Merge tag 'noinstr-lds-2020-05-19' into core/rcu

Get the noinstr section and annotation markers to base the RCU parts on.

Thomas Gleixner committed 6 years ago

Commit 1ed09482 parents 68f0f26 6553896886 files · +9233 −4559Browse files

Unified · Split

886 files

.clang-format +14 −4
Documentation/admin-guide/device-mapper/dm-integrity.rst +9 −6
Documentation/admin-guide/kernel-parameters.txt +1 −2
Documentation/arm64/amu.rst +26 −21
Documentation/devicetree/bindings/Makefile +12 −9
Documentation/devicetree/bindings/display/panel/leadtek,ltk500hd1829.yaml +0 −1
Documentation/devicetree/bindings/display/panel/lvds.yaml +9 −1
Documentation/devicetree/bindings/display/panel/xinpeng,xpp055c272.yaml +0 −1
Documentation/devicetree/bindings/dma/socionext,uniphier-xdmac.yaml +3 −4
Documentation/devicetree/bindings/iio/adc/st,stm32-adc.yaml +1 −1
Documentation/devicetree/bindings/mfd/st,stpmic1.yaml +0 −2
Documentation/devicetree/bindings/phy/qcom,qusb2-phy.yaml +3 −3
Documentation/devicetree/bindings/regulator/mps,mp5416.yaml +0 −1
Documentation/devicetree/bindings/regulator/mps,mpq7920.yaml +2 −1
Documentation/devicetree/bindings/regulator/rohm,bd71828-regulator.yaml +5 −1
Documentation/devicetree/bindings/regulator/rohm,bd71837-regulator.yaml +5 −1
Documentation/devicetree/bindings/regulator/rohm,bd71847-regulator.yaml +5 −1
Documentation/devicetree/bindings/sound/rockchip-i2s.yaml +3 −0
Documentation/devicetree/bindings/sound/rockchip-spdif.txt +0 −45
Documentation/devicetree/bindings/sound/rockchip-spdif.yaml +101 −0
Documentation/devicetree/bindings/usb/renesas,usb3-peri.yaml +1 −0
Documentation/devicetree/bindings/usb/renesas,usbhs.yaml +1 −0
Documentation/devicetree/bindings/usb/usb-xhci.txt +2 −1
Documentation/filesystems/debugfs.rst +2 −2
Documentation/kbuild/makefiles.rst +2 −1
Documentation/networking/devlink/ice.rst +2 −2
Documentation/networking/ip-sysctl.txt +7 −0
Documentation/virt/kvm/index.rst +2 −0
Documentation/virt/kvm/running-nested-guests.rst +276 −0
MAINTAINERS +42 −46
Makefile +12 −5
arch/arc/include/asm/module.h +0 −5
arch/arc/include/asm/vermagic.h +8 −0
arch/arm/boot/dts/bcm2835-rpi.dtsi +3 −0
arch/arm/boot/dts/bcm283x.dtsi +1 −0
arch/arm/boot/dts/omap3-n950-n9.dtsi +5 −0
arch/arm/crypto/chacha-glue.c +11 −3
arch/arm/crypto/nhpoly1305-neon-glue.c +1 −1
arch/arm/crypto/poly1305-glue.c +11 −4
arch/arm/include/asm/futex.h +7 −2
arch/arm/include/asm/module.h +0 −24
arch/arm/include/asm/vermagic.h +31 −0
arch/arm/mach-imx/Makefile +2 −0
arch/arm64/crypto/chacha-neon-glue.c +11 −3
arch/arm64/crypto/nhpoly1305-neon-glue.c +1 −1
arch/arm64/crypto/poly1305-glue.c +11 −4
arch/arm64/include/asm/module.h +0 −2
arch/arm64/include/asm/pointer_auth.h +6 −3
arch/arm64/include/asm/vermagic.h +10 −0
arch/arm64/kernel/vdso/Makefile +1 −1
arch/arm64/kvm/guest.c +7 −0
arch/arm64/kvm/hyp/entry.S +23 −0
arch/arm64/kvm/hyp/hyp-entry.S +0 −1
arch/arm64/kvm/hyp/sysreg-sr.c +3 −14
arch/arm64/mm/hugetlbpage.c +2 −0
arch/h8300/kernel/.gitignore +2 −0
arch/hexagon/include/asm/vermagic.h +4 −4
arch/ia64/include/asm/module.h +0 −4
arch/ia64/include/asm/vermagic.h +15 −0
arch/mips/include/asm/module.h +0 −61
arch/mips/include/asm/vermagic.h +66 −0
arch/nds32/include/asm/vermagic.h +3 −5
arch/powerpc/include/asm/module.h +0 −18
arch/powerpc/include/asm/vermagic.h +20 −0
arch/powerpc/kernel/entry_32.S +1 −1
arch/powerpc/kernel/setup_64.c +2 −0
arch/powerpc/kernel/vmlinux.lds.S +1 −0
arch/powerpc/kvm/book3s_64_mmu_hv.c +5 −4
arch/powerpc/kvm/book3s_64_mmu_radix.c +5 −4
arch/powerpc/kvm/powerpc.c +1 −0
arch/powerpc/mm/nohash/8xx.c +3 −0
arch/powerpc/platforms/Kconfig.cputype +1 −1
arch/riscv/Kconfig +1 −1
arch/riscv/include/asm/csr.h +0 −3
arch/riscv/include/asm/hwcap.h +22 −0
arch/riscv/include/asm/module.h +0 −2
arch/riscv/include/asm/set_memory.h +0 −8
arch/riscv/include/asm/vermagic.h +9 −0
arch/riscv/kernel/cpu_ops.c +2 −2
arch/riscv/kernel/cpufeature.c +80 −3
arch/riscv/kernel/sbi.c +10 −7
arch/riscv/kernel/smp.c +2 −0
arch/riscv/kernel/stacktrace.c +2 −2
arch/riscv/kernel/vdso/Makefile +4 −4
arch/riscv/kernel/vdso/note.S +12 −0
arch/riscv/mm/init.c +2 −17
arch/s390/boot/uv.c +0 −2
arch/s390/kernel/diag.c +1 −1
arch/s390/kernel/smp.c +2 −2
arch/s390/kernel/trace.c +1 −1
arch/s390/kernel/uv.c +2 −1
arch/s390/kvm/interrupt.c +1 −1
arch/s390/kvm/kvm-s390.c +4 −0
arch/s390/kvm/priv.c +3 −1
arch/s390/lib/uaccess.c +4 −0
arch/s390/mm/pgalloc.c +14 −2
arch/s390/pci/pci_irq.c +3 −2
arch/sh/include/asm/module.h +0 −28
arch/sh/include/asm/vermagic.h +34 −0
arch/sh/mm/init.c +1 −1
arch/um/Makefile +1 −0
arch/x86/Kconfig +1 −1
arch/x86/crypto/blake2s-glue.c +4 −6
arch/x86/crypto/chacha_glue.c +11 −3
arch/x86/crypto/nhpoly1305-avx2-glue.c +1 −1
arch/x86/crypto/nhpoly1305-sse2-glue.c +1 −1
arch/x86/crypto/poly1305_glue.c +6 −7
arch/x86/entry/calling.h +21 −19
arch/x86/entry/entry_64.S +7 −7
arch/x86/events/intel/cstate.c +1 −0
arch/x86/hyperv/hv_init.c +10 −2
arch/x86/include/asm/ftrace.h +3 −2
arch/x86/include/asm/kvm_host.h +2 −2
arch/x86/include/asm/module.h +0 −60
arch/x86/include/asm/mshyperv.h +2 −0
arch/x86/include/asm/nospec-branch.h +0 −21
arch/x86/include/asm/unwind.h +1 −1
arch/x86/include/asm/vermagic.h +68 −0
arch/x86/kernel/apic/apic.c +14 −13
arch/x86/kernel/dumpstack_64.c +2 −1
arch/x86/kernel/smpboot.c +33 −14
arch/x86/kernel/unwind_frame.c +3 −0
arch/x86/kernel/unwind_orc.c +74 −39
arch/x86/kvm/Makefile +4 −0
arch/x86/kvm/ioapic.c +5 −5
arch/x86/kvm/svm/sev.c +6 −2
arch/x86/kvm/svm/svm.c +3 −9
arch/x86/kvm/svm/vmenter.S +9 −1
arch/x86/kvm/vmx/nested.c +20 −3
arch/x86/kvm/vmx/vmenter.S +3 −0
arch/x86/kvm/vmx/vmx.c +1 −1
arch/x86/kvm/x86.c +25 −17
arch/x86/mm/pat/set_memory.c +8 −4
arch/x86/net/bpf_jit_comp.c +15 −3
arch/x86/net/bpf_jit_comp32.c +22 −6
arch/xtensa/include/asm/vermagic.h +6 −9
block/bfq-iosched.c +4 −2
block/blk-cgroup.c +1 −1
block/blk-iocost.c +73 −48
block/partitions/core.c +1 −1
crypto/lrw.c +3 −3
crypto/xts.c +3 −3
drivers/acpi/button.c +0 −12
drivers/acpi/device_pm.c +2 −2
drivers/acpi/pci_link.c +2 −2
drivers/amba/bus.c +1 −0
drivers/ata/sata_inic162x.c +1 −1
drivers/base/component.c +5 −3
drivers/base/core.c +6 −1
drivers/base/dd.c +8 −12
drivers/base/firmware_loader/fallback_table.c +0 −1
drivers/base/platform.c +2 −0
drivers/base/power/main.c +1 −1
drivers/block/null_blk.h +19 −10
drivers/block/null_blk_main.c +31 −31
drivers/block/null_blk_zoned.c +33 −12
drivers/block/virtio_blk.c +79 −8
drivers/bus/mhi/core/init.c +3 −4
drivers/bus/mhi/core/internal.h +0 −3
drivers/bus/mhi/core/main.c +5 −13
drivers/bus/mhi/core/pm.c +5 −1
drivers/char/hw_random/virtio-rng.c +1 −0
drivers/char/tpm/tpm-interface.c +1 −1
drivers/char/tpm/tpm2-cmd.c +1 −0
drivers/char/tpm/tpm_ibmvtpm.c +73 −63
drivers/char/tpm/tpm_tis_core.c +7 −1
drivers/counter/104-quad-8.c +160 −34
drivers/cpufreq/intel_pstate.c +1 −1
drivers/crypto/caam/caamalg.c +7 −3
drivers/crypto/caam/caamhash.c +6 −2
drivers/crypto/caam/caampkc.c +6 −2
drivers/crypto/chelsio/chcr_ktls.c +48 −41
drivers/dma-buf/dma-buf.c +4 −3
drivers/dma/Kconfig +2 −1
drivers/dma/dmaengine.c +26 −34
drivers/dma/dmatest.c +3 −3
drivers/dma/mmp_tdma.c +4 −1
drivers/dma/pch_dma.c +1 −1
drivers/dma/tegra20-apb-dma.c +9 −0
drivers/dma/ti/k3-psil.c +1 −0
drivers/dma/xilinx/xilinx_dma.c +10 −10
drivers/firmware/efi/tpm.c +1 −1
drivers/firmware/imx/Kconfig +2 −2
drivers/firmware/xilinx/zynqmp-debug.c +1 −1
drivers/fpga/dfl-pci.c +4 −2
drivers/fpga/zynq-fpga.c +2 −1
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +2 −5
drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +2 −1
drivers/gpu/drm/amd/amdgpu/navi10_sdma_pkt_open.h +16 −0
drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c +13 −1
drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +52 −17
drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c +9 −6
drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h +2 −1
drivers/gpu/drm/amd/display/dc/core/dc.c +2 −3
drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c +27 −0
drivers/gpu/drm/amd/display/dc/core/dc_stream.c +4 −36
drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c +1 −0
drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c +10 −0
drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h +1 −0
drivers/gpu/drm/amd/display/dc/dcn10/dcn10_init.c +1 −0
drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.c +15 −0
drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.h +14 −6
drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c +12 −2
drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c +2 −1
drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c +1 −0
drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.c +1 −0
drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.h +2 −1
drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c +27 −8
drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c +1 −0
drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c +33 −44
drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.c +4 −4
drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h +16 −0
drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h +1 −0
drivers/gpu/drm/amd/display/dc/os_types.h +1 −1
drivers/gpu/drm/amd/powerplay/amd_powerplay.c +5 −4
drivers/gpu/drm/amd/powerplay/hwmgr/processpptables.c +26 −0
drivers/gpu/drm/amd/powerplay/renoir_ppt.c +6 −1
drivers/gpu/drm/bridge/analogix/analogix-anx6345.c +3 −0
drivers/gpu/drm/drm_dp_mst_topology.c +7 −2
drivers/gpu/drm/drm_edid.c +1 −1
drivers/gpu/drm/drm_hdcp.c +7 −1
drivers/gpu/drm/i915/display/intel_ddi.c +11 −3
drivers/gpu/drm/i915/display/intel_display_power.c +6 −6
drivers/gpu/drm/i915/display/intel_dp.c +2 −3
drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c +1 −0
drivers/gpu/drm/i915/display/intel_hdmi.c +2 −1
drivers/gpu/drm/i915/display/intel_sprite.c +12 −5
drivers/gpu/drm/i915/gem/i915_gem_tiling.c +18 −2
drivers/gpu/drm/i915/gem/selftests/huge_pages.c +8 −4
drivers/gpu/drm/i915/gt/intel_rps.c +6 −3
drivers/gpu/drm/i915/gt/intel_timeline.c +2 −0
drivers/gpu/drm/i915/i915_drv.h +2 −0
drivers/gpu/drm/i915/i915_irq.c +2 −4
drivers/gpu/drm/i915/i915_reg.h +2 −2
drivers/gpu/drm/i915/i915_vma.c +6 −4
drivers/gpu/drm/ingenic/ingenic-drm.c +1 −0
drivers/gpu/drm/meson/meson_dw_hdmi.c +1 −3
drivers/gpu/drm/qxl/qxl_cmd.c +5 −5
drivers/gpu/drm/qxl/qxl_display.c +3 −3
drivers/gpu/drm/qxl/qxl_draw.c +4 −3
drivers/gpu/drm/qxl/qxl_image.c +2 −1
drivers/gpu/drm/qxl/qxl_ioctl.c +1 −4
drivers/gpu/drm/scheduler/sched_main.c +1 −1
drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c +1 −1
drivers/gpu/drm/tidss/tidss_crtc.c +13 −3
drivers/gpu/drm/tidss/tidss_encoder.c +11 −3
drivers/gpu/drm/tidss/tidss_plane.c +18 −6
drivers/gpu/drm/virtio/virtgpu_drv.h +1 −0
drivers/gpu/drm/virtio/virtgpu_gem.c +3 −0
drivers/gpu/drm/virtio/virtgpu_ioctl.c +2 −2
drivers/gpu/drm/virtio/virtgpu_kms.c +7 −11
drivers/hid/Kconfig +1 −0
drivers/hid/hid-alps.c +1 −0
drivers/hid/hid-ids.h +7 −1
drivers/hid/hid-lg-g15.c +4 −0
drivers/hid/hid-multitouch.c +3 −0
drivers/hid/hid-quirks.c +1 −0
drivers/hid/i2c-hid/i2c-hid-core.c +2 −0
drivers/hid/usbhid/hid-core.c +29 −8
drivers/hid/usbhid/usbhid.h +1 −0
drivers/hid/wacom_sys.c +3 −1
drivers/hid/wacom_wac.c +22 −66
drivers/hv/hv.c +1 −5
drivers/hv/hv_trace.h +2 −2
drivers/hv/vmbus_drv.c +34 −9
drivers/i2c/busses/i2c-amd-mp2-pci.c +1 −1
drivers/i2c/busses/i2c-aspeed.c +4 −1
drivers/i2c/busses/i2c-bcm-iproc.c +3 −0
drivers/i2c/busses/i2c-tegra.c +12 −24
drivers/iio/adc/ad7192.c +47 −16
drivers/iio/adc/ad7793.c +1 −1
drivers/iio/adc/stm32-adc.c +28 −3
drivers/iio/adc/ti-ads8344.c +3 −3
drivers/iio/adc/xilinx-xadc-core.c +74 −21
drivers/iio/common/st_sensors/st_sensors_core.c +1 −1
drivers/iio/dac/ad5770r.c +1 −1
drivers/iio/imu/inv_mpu6050/inv_mpu_core.c +10 −1
drivers/iio/imu/st_lsm6dsx/st_lsm6dsx.h +3 −0
drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_buffer.c +16 −7
drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c +23 −1
drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_shub.c +22 −7
drivers/iio/industrialio-core.c +2 −5
drivers/infiniband/core/cm.c +14 −12
drivers/infiniband/core/rdma_core.c +4 −5
drivers/infiniband/core/uverbs_main.c +4 −0
drivers/infiniband/hw/i40iw/i40iw_ctrl.c +1 −1
drivers/infiniband/hw/mlx4/main.c +2 −1
drivers/infiniband/hw/mlx5/qp.c +3 −1
drivers/infiniband/sw/rdmavt/cq.c +2 −2
drivers/infiniband/sw/rdmavt/mmap.c +2 −2
drivers/infiniband/sw/rdmavt/qp.c +2 −2
drivers/infiniband/sw/rdmavt/srq.c +2 −2
drivers/infiniband/sw/siw/siw_qp_tx.c +11 −4
drivers/interconnect/qcom/bcm-voter.c +3 −2
drivers/interconnect/qcom/osm-l3.c +2 −2
drivers/interconnect/qcom/sdm845.c +8 −8
drivers/iommu/Kconfig +2 −2
drivers/iommu/amd_iommu.c +154 −44
drivers/iommu/amd_iommu_init.c +1 −1
drivers/iommu/amd_iommu_types.h +7 −2
drivers/iommu/intel-iommu.c +2 −2
drivers/iommu/iommu.c +2 −1
drivers/iommu/qcom_iommu.c +4 −1
drivers/iommu/virtio-iommu.c +1 −1
drivers/md/dm-mpath.c +4 −2
drivers/md/dm-verity-fec.c +1 −1
drivers/md/dm-writecache.c +37 −15
drivers/misc/mei/hw-me.c +8 −0
drivers/misc/mei/hw-me.h +4 −0
drivers/misc/mei/pci-me.c +3 −2
drivers/misc/mic/Kconfig +1 −1
drivers/mmc/core/mmc_ops.c +1 −1
drivers/mmc/host/cqhci.c +10 −11
drivers/mmc/host/meson-mx-sdio.c +1 −10
drivers/mmc/host/sdhci-msm.c +2 −0
drivers/mmc/host/sdhci-pci-core.c +3 −0
drivers/mmc/host/sdhci-xenon.c +10 −0
drivers/most/core.c +1 −1
drivers/net/caif/Kconfig +1 −1
drivers/net/dsa/b53/b53_common.c +33 −5
drivers/net/dsa/b53/b53_regs.h +6 −2
drivers/net/dsa/mv88e6xxx/Kconfig +1 −1
drivers/net/dsa/mv88e6xxx/chip.c +0 −4
drivers/net/dsa/ocelot/felix.c +1 −0
drivers/net/dsa/ocelot/felix.h +1 −0
drivers/net/dsa/ocelot/felix_vsc9959.c +1 −0
drivers/net/dsa/sja1105/Kconfig +1 −0
drivers/net/dsa/sja1105/sja1105_ptp.c +18 −8
drivers/net/ethernet/amazon/ena/ena_netdev.h +1 −1
drivers/net/ethernet/aquantia/atlantic/aq_pci_func.c +1 −1
drivers/net/ethernet/broadcom/bcmsysport.c +2 −2
drivers/net/ethernet/broadcom/bgmac-platform.c +2 −7
drivers/net/ethernet/broadcom/bnxt/bnxt.c +13 −7
drivers/net/ethernet/broadcom/bnxt/bnxt.h +0 −1
drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.h +1 −1
drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c +2 −8
drivers/net/ethernet/broadcom/genet/bcmgenet.c +5 −1
drivers/net/ethernet/cadence/Kconfig +1 −1
drivers/net/ethernet/cadence/macb_main.c +12 −12
drivers/net/ethernet/cavium/Kconfig +1 −1
drivers/net/ethernet/chelsio/cxgb4/cudbg_lib.c +20 −7
drivers/net/ethernet/chelsio/cxgb4/cxgb4_ptp.c +6 −21
drivers/net/ethernet/chelsio/cxgb4/sge.c +37 −3
drivers/net/ethernet/chelsio/cxgb4/t4_regs.h +3 −0
drivers/net/ethernet/freescale/enetc/enetc_pci_mdio.c +1 −1
drivers/net/ethernet/ibm/ibmvnic.c +2 −1
drivers/net/ethernet/intel/e1000/e1000_main.c +1 −1
drivers/net/ethernet/intel/i40e/i40e_main.c +2 −2
drivers/net/ethernet/marvell/mvpp2/mvpp2_cls.c +3 −0
drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c +2 −0
drivers/net/ethernet/mellanox/mlx4/en_tx.c +13 −1
drivers/net/ethernet/mellanox/mlx4/main.c +3 −1
drivers/net/ethernet/mellanox/mlx5/core/Kconfig +4 −4
drivers/net/ethernet/mellanox/mlx5/core/cmd.c +5 −1
drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c +3 −3
drivers/net/ethernet/mellanox/mlx5/core/en.h +2 −1
drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c +12 −11
drivers/net/ethernet/mellanox/mlx5/core/en/xsk/tx.c +3 −0
drivers/net/ethernet/mellanox/mlx5/core/en_main.c +6 −1
drivers/net/ethernet/mellanox/mlx5/core/en_rep.c +2 −7
drivers/net/ethernet/mellanox/mlx5/core/en_rx.c +5 −3
drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c +5 −1
drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +9 −9
drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c +13 −1
drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_actions.c +2 −2
drivers/net/ethernet/mellanox/mlxsw/spectrum2_acl_tcam.c +2 −2
drivers/net/ethernet/mellanox/mlxsw/spectrum_acl.c +1 −1
drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c +10 −2
drivers/net/ethernet/mellanox/mlxsw/spectrum_flower.c +2 −1
drivers/net/ethernet/mellanox/mlxsw/spectrum_mr_tcam.c +2 −2
drivers/net/ethernet/moxa/moxart_ether.c +1 −1
drivers/net/ethernet/mscc/ocelot.c +11 −6
drivers/net/ethernet/mscc/ocelot_regs.c +1 −0
drivers/net/ethernet/natsemi/jazzsonic.c +4 −2
drivers/net/ethernet/netronome/nfp/abm/main.c +1 −0
drivers/net/ethernet/pensando/ionic/ionic_debugfs.c +1 −2
drivers/net/ethernet/pensando/ionic/ionic_lif.c +2 −2
drivers/net/ethernet/sgi/meth.h +0 −16
drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c +189 −0
drivers/net/ethernet/stmicro/stmmac/dwmac-intel.h +23 −0
drivers/net/ethernet/stmicro/stmmac/dwmac-meson8b.c +1 −0
drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c +6 −3
drivers/net/ethernet/stmicro/stmmac/dwmac5.c +1 −1
drivers/net/ethernet/stmicro/stmmac/stmmac_hwtstamp.c +8 −4
drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +26 −6
drivers/net/ethernet/ti/Kconfig +1 −2
drivers/net/ethernet/ti/am65-cpsw-nuss.c +3 −2
drivers/net/ethernet/toshiba/tc35815.c +1 −1
drivers/net/ethernet/xscale/ixp4xx_eth.c +2 −0
drivers/net/geneve.c +1 −1
drivers/net/gtp.c +5 −4
drivers/net/hyperv/netvsc_drv.c +2 −1
drivers/net/ipa/gsi.c +9 −2
drivers/net/ipa/gsi_reg.h +2 −0
drivers/net/ipa/ipa_endpoint.c +2 −5
drivers/net/macsec.c +12 −6
drivers/net/macvlan.c +1 −1
drivers/net/phy/bcm84881.c +3 −3
drivers/net/phy/dp83640.c +1 −1
drivers/net/phy/dp83822.c +14 −16
drivers/net/phy/dp83tc811.c +12 −9
drivers/net/phy/marvell10g.c +28 −2
drivers/net/phy/microchip_t1.c +171 −0
drivers/net/team/team.c +4 −0
drivers/net/usb/qmi_wwan.c +1 −0
drivers/net/vrf.c +6 −4
drivers/net/vxlan.c +3 −3
drivers/net/wimax/i2400m/usb-fw.c +1 −0
drivers/net/wireguard/queueing.c +3 −1
drivers/net/wireguard/receive.c +10 −11
drivers/net/wireguard/selftest/ratelimiter.c +2 −2
drivers/net/wireguard/send.c +10 −10
drivers/net/wireguard/socket.c +0 −12
drivers/net/wireless/intel/iwlegacy/3945-rs.c +1 −1
drivers/net/wireless/intel/iwlegacy/4965-rs.c +1 −1
drivers/net/wireless/intel/iwlwifi/dvm/rs.c +1 −1
drivers/net/wireless/intel/iwlwifi/fw/acpi.c +7 −2
drivers/net/wireless/intel/iwlwifi/fw/api/txq.h +3 −3
drivers/net/wireless/intel/iwlwifi/iwl-drv.c +1 −1
drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c +2 −4
drivers/net/wireless/intel/iwlwifi/mvm/fw.c +11 −14
drivers/net/wireless/intel/iwlwifi/mvm/rs.c +1 −1
drivers/net/wireless/intel/iwlwifi/mvm/rx.c +11 −2
drivers/net/wireless/intel/iwlwifi/mvm/sta.c +7 −2
drivers/net/wireless/intel/iwlwifi/pcie/ctxt-info-gen3.c +14 −4
drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c +3 −0
drivers/net/wireless/mac80211_hwsim.c +1 −1
drivers/net/wireless/realtek/rtlwifi/rc.c +1 −1
drivers/nvme/host/core.c +3 −1
drivers/nvme/host/pci.c +5 −1
drivers/pci/quirks.c +7 −0
drivers/phy/qualcomm/phy-qcom-qusb2.c +7 −0
drivers/phy/qualcomm/phy-qcom-usb-hs-28nm.c +21 −11
drivers/phy/tegra/Kconfig +2 −1
drivers/platform/chrome/cros_ec_sensorhub.c +46 −34
drivers/platform/chrome/cros_ec_sensorhub_ring.c +46 −27
drivers/platform/x86/asus-nb-wmi.c +24 −0
drivers/platform/x86/intel-uncore-frequency.c +1 −1
drivers/platform/x86/intel_pmc_core.c +5 −19
drivers/platform/x86/intel_pmc_core.h +0 −2
drivers/platform/x86/surface3_power.c +2 −2
drivers/platform/x86/thinkpad_acpi.c +1 −1
drivers/platform/x86/xiaomi-wmi.c +2 −2
drivers/ptp/ptp_ines.c +2 −2
drivers/regulator/core.c +11 −14
drivers/remoteproc/mtk_common.h +1 −1
drivers/remoteproc/mtk_scp.c +3 −3
drivers/remoteproc/qcom_q6v5_mss.c +20 −13
drivers/remoteproc/remoteproc_sysfs.c +1 −0
drivers/remoteproc/stm32_rproc.c +1 −0
drivers/rpmsg/mtk_rpmsg.c +1 −0
drivers/s390/net/qeth_core_main.c +5 −5
drivers/scsi/Kconfig +1 −1
drivers/scsi/ibmvscsi/ibmvfc.c +5 −0
drivers/scsi/ibmvscsi/ibmvscsi.c +0 −4
drivers/scsi/qla2xxx/qla_attr.c +1 −1
drivers/scsi/qla2xxx/qla_mbx.c +1 −1
drivers/scsi/qla2xxx/qla_os.c +17 −18
drivers/scsi/scsi_lib.c +1 −0
drivers/scsi/sg.c +3 −1
drivers/soc/fsl/dpio/dpio-service.c +14 −5
drivers/soc/fsl/dpio/qbman-portal.c +2 −4
drivers/soc/imx/Kconfig +1 −0
drivers/soc/xilinx/Kconfig +2 −2
drivers/staging/comedi/comedi_fops.c +3 −1
drivers/staging/comedi/drivers/dt2815.c +3 −0
drivers/staging/gasket/apex_driver.c +0 −7
drivers/staging/gasket/gasket_core.c +4 −0
drivers/staging/gasket/gasket_sysfs.c +1 −2
drivers/staging/gasket/gasket_sysfs.h +0 −4
drivers/staging/ks7010/TODO +0 −1
drivers/staging/vt6656/key.c +3 −11
drivers/staging/vt6656/main_usb.c +17 −14
drivers/staging/vt6656/usbpipe.c +2 −1
drivers/target/target_core_iblock.c +1 −1
drivers/thunderbolt/usb4.c +3 −0
drivers/tty/hvc/Kconfig +1 −1
drivers/tty/hvc/hvc_console.c +14 −9
drivers/tty/rocket.c +14 −11
drivers/tty/serial/Kconfig +1 −1
drivers/tty/serial/owl-uart.c +7 −0
drivers/tty/serial/sh-sci.c +10 −3
drivers/tty/serial/sunhv.c +3 −0
drivers/tty/serial/xilinx_uartps.c +50 −162
drivers/tty/sysrq.c +2 −0
drivers/tty/vt/vt.c +10 −4
drivers/usb/chipidea/ci_hdrc_msm.c +1 −1
drivers/usb/class/cdc-acm.c +31 −5
drivers/usb/class/cdc-acm.h +4 −1
drivers/usb/core/devio.c +2 −3
drivers/usb/core/hub.c +15 −3
drivers/usb/core/message.c +10 −3
drivers/usb/core/quirks.c +4 −0
drivers/usb/dwc3/core.h +6 −2
drivers/usb/dwc3/gadget.c +47 −29
drivers/usb/early/xhci-dbc.c +4 −4
drivers/usb/early/xhci-dbc.h +16 −2
drivers/usb/gadget/function/f_fs.c +4 −0
drivers/usb/gadget/legacy/raw_gadget.c +42 −29
drivers/usb/gadget/udc/atmel_usba_udc.c +2 −2
drivers/usb/gadget/udc/bdc/bdc_ep.c +1 −1
drivers/usb/host/xhci-hub.c +9 −0
drivers/usb/host/xhci-ring.c +40 −6
drivers/usb/host/xhci.c +7 −7
drivers/usb/host/xhci.h +3 −2
drivers/usb/misc/sisusbvga/sisusb.c +10 −10
drivers/usb/misc/sisusbvga/sisusb_init.h +7 −7
drivers/usb/serial/garmin_gps.c +2 −2
drivers/usb/serial/qcserial.c +1 −0
drivers/usb/storage/uas.c +43 −3
drivers/usb/storage/unusual_devs.h +7 −0
drivers/usb/storage/unusual_uas.h +7 −0
drivers/usb/typec/bus.c +4 −1
drivers/usb/typec/mux/intel_pmc_mux.c +6 −2
drivers/usb/typec/mux/pi3usb30532.c +2 −2
drivers/usb/typec/tcpm/tcpm.c +26 −0
drivers/vdpa/Kconfig +6 −12
drivers/vdpa/ifcvf/ifcvf_base.c +0 −2
drivers/vdpa/ifcvf/ifcvf_main.c +1 −3
drivers/vdpa/vdpa.c +1 −1
drivers/vdpa/vdpa_sim/vdpa_sim.c +2 −2
drivers/vfio/vfio_iommu_type1.c +5 −5
drivers/vhost/Kconfig +16 −5
drivers/vhost/net.c +15 −13
drivers/vhost/scsi.c +7 −7
drivers/vhost/test.c +7 −7
drivers/vhost/vdpa.c +0 −5
drivers/vhost/vhost.h +27 −0
drivers/vhost/vringh.c +5 −0
drivers/vhost/vsock.c +23 −12
drivers/virtio/Kconfig +1 −1
drivers/virtio/virtio_balloon.c +2 −2
drivers/virtio/virtio_input.c +1 −0
fs/afs/cmservice.c +1 −1
fs/afs/fs_probe.c +1 −4
fs/afs/internal.h +1 −3
fs/afs/rotate.c +3 −3
fs/afs/server.c +2 −5
fs/afs/vl_rotate.c +2 −2
fs/afs/volume.c +5 −3
fs/block_dev.c +10 −1
fs/btrfs/backref.c +1 −1
fs/btrfs/block-group.c +14 −6
fs/btrfs/discard.h +1 −1
fs/btrfs/disk-io.c +32 −4
fs/btrfs/relocation.c +1 −0
fs/btrfs/transaction.c +11 −2
fs/btrfs/tree-log.c +40 −3
fs/buffer.c +1 −1
fs/ceph/caps.c +2 −1
fs/ceph/debugfs.c +1 −1
fs/ceph/mds_client.c +3 −5
fs/ceph/quota.c +2 −2
fs/cifs/cifsglob.h +2 −1
fs/cifs/connect.c +6 −0
fs/cifs/misc.c +65 −17
fs/cifs/smb2ops.c +5 −0
fs/configfs/dir.c +1 −0
fs/coredump.c +10 −0
fs/debugfs/file.c +3 −12
fs/eventpoll.c +33 −28
fs/exfat/balloc.c +0 −3
fs/exfat/exfat_fs.h +1 −0
fs/exfat/file.c +2 −0
fs/exfat/misc.c +13 −1
fs/exfat/namei.c +7 −0
fs/exfat/super.c +22 −31
fs/io_uring.c +58 −75
fs/ioctl.c +8 −0
fs/iomap/fiemap.c +1 −4
fs/nfs/nfs3acl.c +15 −7
fs/nfs/nfs4proc.c +9 −2
fs/nfs/pnfs.c +5 −6
fs/nfs/super.c +1 −1
fs/nfsd/nfs4callback.c +3 −1
fs/nfsd/nfs4state.c +2 −0
fs/ocfs2/dlmfs/dlmfs.c +12 −15
fs/pnode.c +4 −5
fs/proc/base.c +0 −1
fs/proc/vmcore.c +3 −2
fs/splice.c +18 −27
fs/super.c +1 −1
fs/vboxsf/super.c +1 −1
include/asm-generic/Kbuild +1 −0
include/asm-generic/sections.h +3 −0
include/asm-generic/vermagic.h +7 −0
include/asm-generic/vmlinux.lds.h +10 −0
include/linux/amba/bus.h +1 −0
include/linux/backing-dev-defs.h +1 −1
include/linux/backing-dev.h +1 −8
include/linux/compiler.h +53 −0
include/linux/compiler_types.h +4 −0
include/linux/debugfs.h +4 −8
include/linux/dma-buf.h +1 −2
include/linux/dmaengine.h +6 −6
include/linux/fs.h +1 −1
include/linux/iio/iio.h +1 −1
include/linux/kvm_host.h +1 −1
include/linux/lsm_hook_defs.h +1 −1
include/linux/mhi.h +10 −6
include/linux/nfs_xdr.h +2 −0
include/linux/platform_data/cros_ec_sensorhub.h +1 −0
include/linux/platform_device.h +1 −0
include/linux/pnp.h +9 −20
include/linux/stmmac.h +2 −0
include/linux/sunrpc/clnt.h +12 −1
include/linux/sunrpc/svc_rdma.h +1 −0
include/linux/tcp.h +0 −51
include/linux/tracepoint.h +1 −21
include/linux/tty.h +1 −1
include/linux/vdpa.h +1 −1
include/linux/vermagic.h +5 −3
include/linux/virtio.h +0 −1
include/linux/virtio_net.h +24 −2
include/linux/virtio_vsock.h +1 −0
include/linux/vmalloc.h +1 −1
include/linux/vringh.h +6 −0
include/net/flow_offload.h +8 −1
include/net/inet_ecn.h +55 −2
include/net/ip6_fib.h +4 −0
include/net/mac80211.h +3 −1
include/net/mptcp.h +0 −3
include/net/ndisc.h +1 −1
include/net/net_namespace.h +7 −0
include/net/sch_generic.h +1 −0
include/net/tcp.h +1 −1
include/soc/mscc/ocelot.h +1 −0
include/soc/tegra/pmc.h +9 −6
include/sound/soc-dai.h +0 −1
include/sound/soc.h +3 −0
include/trace/events/gpu_mem.h +1 −1
include/trace/events/iocost.h +3 −3
include/trace/events/rpcrdma.h +40 −22
include/trace/events/wbt.h +4 −4
include/trace/events/writeback.h +0 −1
include/uapi/drm/amdgpu_drm.h +4 −0
include/uapi/linux/bpf.h +2 −2
include/uapi/linux/dlm_device.h +2 −2
include/uapi/linux/dma-buf.h +6 −0
include/uapi/linux/fiemap.h +1 −1
include/uapi/linux/hyperv.h +2 −2
include/uapi/linux/if_arcnet.h +3 −3
include/uapi/linux/mmc/ioctl.h +1 −1
include/uapi/linux/net_dropmon.h +2 −2
include/uapi/linux/netfilter_bridge/ebt_among.h +1 −1
include/uapi/linux/virtio_balloon.h +9 −2
include/uapi/scsi/scsi_bsg_fc.h +1 −1
include/vdso/datapage.h +2 −2
init/Kconfig +0 −18
init/initramfs.c +1 −1
init/main.c +52 −17
ipc/mqueue.c +26 −8
kernel/audit.c +3 −0
kernel/bpf/cpumap.c +1 −1
kernel/bpf/syscall.c +10 −3
kernel/bpf/verifier.c +28 −10
kernel/events/core.c +10 −3
kernel/exit.c +1 −0
kernel/kcov.c +2 −2
kernel/power/hibernate.c +7 −0
kernel/sched/core.c +2 −7
kernel/signal.c +5 −1
kernel/trace/Kconfig +0 −1
kernel/trace/ftrace.c +1 −0
kernel/trace/preemptirq_delay_test.c +24 −6
kernel/trace/trace.c +15 −1
kernel/trace/trace_boot.c +8 −12
kernel/trace/trace_events_hist.c +7 −0
kernel/trace/trace_kprobe.c +7 −1
kernel/trace/tracing_map.c +3 −3
kernel/umh.c +5 −0
lib/Kconfig.ubsan +6 −9
lib/kunit/test.c +1 −1
lib/mpi/longlong.h +17 −17
mm/backing-dev.c +11 −2
mm/gup.c +1 −1
mm/hugetlb.c +8 −6
mm/ksm.c +10 −2
mm/madvise.c +18 −0
mm/memcontrol.c +9 −6
mm/page_alloc.c +9 −0
mm/percpu.c +10 −4
mm/shmem.c +8 −5
mm/slub.c +40 −17
mm/vmalloc.c +13 −3
mm/vmscan.c +0 −1
net/atm/common.c +11 −11
net/atm/lec.c +6 −0
net/batman-adv/bat_v_ogm.c +1 −1
net/batman-adv/network-coding.c +1 −8
net/batman-adv/sysfs.c +2 −1
net/bridge/br_netlink.c +1 −0
net/core/devlink.c +10 −2
net/core/drop_monitor.c +7 −4
net/core/neighbour.c +3 −3
net/core/netclassid_cgroup.c +1 −3
net/core/sock.c +0 −1
net/dsa/dsa2.c +1 −1
net/dsa/master.c +2 −1
net/dsa/slave.c +6 −10
net/hsr/hsr_slave.c +1 −1
net/ipv4/fib_semantics.c +3 −3
net/ipv4/tcp_input.c +0 −7
net/ipv4/xfrm4_output.c +0 −2
net/ipv6/ipv6_sockglue.c +6 −7
net/ipv6/route.c +25 −0
net/ipv6/rpl.c +4 −3
net/ipv6/seg6.c +8 −2
net/ipv6/xfrm6_output.c +0 −2
net/mac80211/main.c +3 −2
net/mac80211/rate.c +4 −11
net/mac80211/rate.h +23 −0
net/mac80211/rc80211_minstrel_ht.c +13 −6
net/mac80211/sta_info.c +2 −1
net/mptcp/options.c +41 −55
net/mptcp/pm_netlink.c +7 −5
net/mptcp/protocol.c +20 −10
net/mptcp/protocol.h +44 −5
net/mptcp/subflow.c +119 −52
net/netfilter/nf_flow_table_core.c +4 −2
net/netfilter/nf_nat_proto.c +3 −5
net/netfilter/nfnetlink_osf.c +7 −5
net/netlabel/Kconfig +1 −1
net/netrom/nr_route.c +1 −0
net/openvswitch/conntrack.c +2 −1
net/openvswitch/datapath.c +3 −1
net/sched/cls_api.c +16 −6
net/sched/sch_choke.c +2 −1
net/sched/sch_etf.c +4 −3
net/sched/sch_fq_codel.c +1 −1
net/sched/sch_sfq.c +9 −0
net/sched/sch_skbprio.c +3 −0
net/sctp/sm_make_chunk.c +5 −1
net/sctp/sm_statefuns.c +3 −3
net/sunrpc/cache.c +3 −2
net/sunrpc/clnt.c +18 −6
net/sunrpc/svc_xprt.c +2 −3
net/sunrpc/svcsock.c +4 −0
net/sunrpc/xprtrdma/rpc_rdma.c +11 −4
net/sunrpc/xprtrdma/svc_rdma_backchannel.c +2 −0
net/sunrpc/xprtrdma/svc_rdma_recvfrom.c +22 −0
net/sunrpc/xprtrdma/svc_rdma_rw.c +1 −2
net/sunrpc/xprtrdma/svc_rdma_sendto.c +12 −17
net/sunrpc/xprtrdma/svc_rdma_transport.c +0 −5
net/sunrpc/xprtrdma/verbs.c +2 −1
net/sunrpc/xprtsock.c +1 −0
net/tipc/crypto.c +1 −0
net/tipc/node.c +3 −1
net/tipc/topsrv.c +3 −2
net/tls/tls_sw.c +5 −2
net/vmw_vsock/virtio_transport_common.c +4 −0
net/x25/x25_dev.c +3 −1
net/x25/x25_subr.c +6 −0
samples/trace_events/trace-events-sample.h +1 −1
samples/vfio-mdev/mdpy.c +1 −1
scripts/Makefile.lib +1 −1
scripts/checkpatch.pl +1 −1
scripts/config +4 −1
scripts/decodecode +1 −1
scripts/gcc-plugins/Makefile +1 −0
scripts/gcc-plugins/gcc-common.h +4 −0
scripts/gcc-plugins/stackleak_plugin.c +2 −3
scripts/gdb/linux/rbtree.py +2 −2
scripts/kallsyms.c +1 −1
scripts/mod/modpost.c +1 −1
security/selinux/hooks.c +45 −25
security/selinux/ss/conditional.c +1 −1
sound/core/oss/pcm_plugin.c +6 −4
sound/isa/opti9xx/miro.c +6 −3
sound/isa/opti9xx/opti92x-ad1848.c +6 −3
sound/pci/hda/hda_codec.c +17 −11
sound/pci/hda/hda_intel.c +7 −20
sound/pci/hda/patch_hdmi.c +14 −1
sound/pci/hda/patch_realtek.c +9 −3
sound/soc/amd/acp3x-rt5682-max9836.c +3 −3
sound/soc/codecs/Kconfig +3 −0
sound/soc/codecs/hdac_hdmi.c +3 −3
sound/soc/codecs/madera.c +0 −4
sound/soc/codecs/sgtl5000.c +34 −0
sound/soc/codecs/sgtl5000.h +1 −0
sound/soc/codecs/tas571x.c +15 −5
sound/soc/codecs/wm8960.c +1 −2
sound/soc/codecs/wsa881x.c +4 −0
sound/soc/intel/common/soc-acpi-intel-cml-match.c +0 −8
sound/soc/intel/common/soc-acpi-intel-icl-match.c +0 −8
sound/soc/meson/axg-card.c +3 −1
sound/soc/meson/gx-card.c +3 −1
sound/soc/qcom/apq8096.c +1 −3
sound/soc/qcom/qdsp6/q6afe-dai.c +16 −0
sound/soc/qcom/sdm845.c +1 −3
sound/soc/samsung/s3c-i2s-v2.c +0 −57
sound/soc/samsung/s3c2412-i2s.c +56 −0
sound/soc/sh/rcar/ssi.c +10 −1
sound/soc/sh/rcar/ssiu.c +1 −1
sound/soc/soc-dai.c +2 −9
sound/soc/soc-dapm.c +75 −72
sound/soc/soc-pcm.c +11 −2
sound/soc/soc-topology.c +89 −26
sound/soc/sof/intel/bdw.c +16 −0
sound/soc/sof/intel/byt.c +48 −0
sound/soc/stm/stm32_sai_sub.c +6 −8
sound/usb/format.c +51 −0
sound/usb/line6/podhd.c +5 −17
sound/usb/mixer.c +34 −3
sound/usb/mixer.h +10 −0
sound/usb/mixer_maps.c +34 −3
sound/usb/mixer_quirks.c +8 −4
sound/usb/quirks-table.h +14 −84
sound/usb/quirks.c +15 −1
sound/usb/usx2y/usbusx2yaudio.c +2 −0
tools/bootconfig/main.c +6 −3
tools/bpf/bpftool/struct_ops.c +7 −1
tools/bpf/runqslower/Makefile +1 −1
tools/build/feature/test-sync-compare-and-swap.c +1 −1
tools/cgroup/iocost_monitor.py +6 −1
tools/include/uapi/linux/bpf.h +1 −1
tools/lib/bpf/netlink.c +2 −0
tools/objtool/check.c +14 −3
tools/objtool/elf.c +1 −1
tools/objtool/elf.h +5 −4
tools/power/pm-graph/Makefile +4 −0
tools/power/pm-graph/README +114 −10
tools/power/pm-graph/bootgraph.py +1 −1
tools/power/pm-graph/sleepgraph.8 +49 −16
tools/power/pm-graph/sleepgraph.py +451 −344
tools/testing/selftests/bpf/prog_tests/fexit_bpf2bpf.c +24 −6
tools/testing/selftests/bpf/progs/connect4_prog.c +16 −12
tools/testing/selftests/bpf/progs/freplace_connect4.c +18 −0
tools/testing/selftests/bpf/progs/test_btf_haskv.c +5 −13
tools/testing/selftests/bpf/progs/test_btf_newkv.c +5 −13
tools/testing/selftests/bpf/progs/test_btf_nokv.c +5 −13
tools/testing/selftests/bpf/test_btf.c +1 −1
tools/testing/selftests/bpf/verifier/stack_ptr.c +40 −0
tools/testing/selftests/bpf/verifier/value_illegal_alu.c +1 −0
tools/testing/selftests/filesystems/epoll/epoll_wakeup_test.c +146 −0
tools/testing/selftests/ftrace/config +1 −0
tools/testing/selftests/ftrace/ftracetest +30 −2
tools/testing/selftests/ftrace/test.d/ftrace/fgraph-filter-stack.tc +1 −4
tools/testing/selftests/ftrace/test.d/ftrace/fgraph-filter.tc +2 −0
tools/testing/selftests/ftrace/test.d/ftrace/func-filter-glob.tc +2 −0
tools/testing/selftests/ftrace/test.d/ftrace/func-filter-notrace-pid.tc +1 −4
tools/testing/selftests/ftrace/test.d/ftrace/func-filter-pid.tc +1 −4
tools/testing/selftests/ftrace/test.d/ftrace/func-filter-stacktrace.tc +1 −1
tools/testing/selftests/ftrace/test.d/ftrace/func_event_triggers.tc +1 −4
tools/testing/selftests/ftrace/test.d/ftrace/func_mod_trace.tc +1 −1
tools/testing/selftests/ftrace/test.d/ftrace/func_profiler.tc +1 −4
tools/testing/selftests/ftrace/test.d/ftrace/func_set_ftrace_file.tc +1 −4
tools/testing/selftests/ftrace/test.d/ftrace/func_stack_tracer.tc +2 −0
tools/testing/selftests/ftrace/test.d/ftrace/func_traceonoff_triggers.tc +1 −4
tools/testing/selftests/ftrace/test.d/functions +6 −0
tools/testing/selftests/ftrace/test.d/kprobe/kprobe_args_type.tc +1 −1
tools/testing/selftests/ftrace/test.d/kprobe/kprobe_ftrace.tc +2 −0
tools/testing/selftests/gpio/Makefile +6 −6
tools/testing/selftests/intel_pstate/Makefile +1 −1
tools/testing/selftests/ipc/msgque.c +1 −1
tools/testing/selftests/kselftest/runner.sh +1 −1
tools/testing/selftests/kselftest_deps.sh +272 −0
tools/testing/selftests/kselftest_harness.h +3 −3
tools/testing/selftests/kvm/Makefile +28 −1
tools/testing/selftests/kvm/include/evmcs.h +2 −2
tools/testing/selftests/kvm/lib/x86_64/vmx.c +3 −0
tools/testing/selftests/memfd/Makefile +14 −3
tools/testing/selftests/net/fib_nexthops.sh +23 −0
tools/testing/selftests/net/fib_tests.sh +8 −2
tools/testing/selftests/net/tcp_mmap.c +5 −2
tools/testing/selftests/seccomp/seccomp_bpf.c +8 −6
tools/testing/selftests/tpm2/test_smoke.sh +2 −11
tools/testing/selftests/tpm2/test_space.sh +1 −8
tools/testing/selftests/vm/Makefile +2 −2
tools/testing/selftests/vm/run_vmtests +1 −1
tools/testing/selftests/wireguard/netns.sh +51 −3
tools/testing/selftests/wireguard/qemu/arch/powerpc64le.config +1 −0
tools/testing/selftests/wireguard/qemu/debug.config +0 −1
tools/virtio/Makefile +3 −2
tools/virtio/asm/barrier.h +1 −0
tools/virtio/generated/autoconf.h +0 −0
tools/virtio/linux/compiler.h +1 −0
tools/vm/Makefile +2 −0
virt/kvm/arm/hyp/aarch32.c +6 −2
virt/kvm/arm/psci.c +40 −0
virt/kvm/arm/vgic/vgic-init.c +16 −3
virt/kvm/arm/vgic/vgic-its.c +9 −2
virt/kvm/arm/vgic/vgic-mmio-v2.c +10 −6
virt/kvm/arm/vgic/vgic-mmio-v3.c +18 −13
virt/kvm/arm/vgic/vgic-mmio.c +170 −58
virt/kvm/arm/vgic/vgic-mmio.h +19 −0

@@ -142,10 +142,13 @@ ForEachMacros: · 141 unmodified lines
142	142	- 'for_each_card_auxs'
143	143	- 'for_each_card_auxs_safe'
144	144	- 'for_each_card_components'
	145	+ - 'for_each_card_dapms'
145	146	- 'for_each_card_pre_auxs'
146	147	- 'for_each_card_prelinks'
147	148	- 'for_each_card_rtds'
148	149	- 'for_each_card_rtds_safe'
	150	+ - 'for_each_card_widgets'
	151	+ - 'for_each_card_widgets_safe'
149	152	- 'for_each_cgroup_storage_type'
150	153	- 'for_each_child_of_node'
151	154	- 'for_each_clear_bit'
@@ -160,6 +163,7 @@ ForEachMacros: · 8 unmodified lines
160	163	- 'for_each_cpu_and'
161	164	- 'for_each_cpu_not'
162	165	- 'for_each_cpu_wrap'
	166	+ - 'for_each_dapm_widgets'
163	167	- 'for_each_dev_addr'
164	168	- 'for_each_dev_scope'
165	169	- 'for_each_displayid_db'
@@ -170,7 +174,6 @@ ForEachMacros: · 4 unmodified lines
170	174	- 'for_each_dpcm_fe'
171	175	- 'for_each_drhd_unit'
172	176	- 'for_each_dss_dev'
173		− - 'for_each_efi_handle'
174	177	- 'for_each_efi_memory_desc'
175	178	- 'for_each_efi_memory_desc_in_map'
176	179	- 'for_each_element'
@@ -191,6 +194,7 @@ ForEachMacros: · 14 unmodified lines
191	194	- 'for_each_ip_tunnel_rcu'
192	195	- 'for_each_irq_nr'
193	196	- 'for_each_link_codecs'
	197	+ - 'for_each_link_cpus'
194	198	- 'for_each_link_platforms'
195	199	- 'for_each_lru'
196	200	- 'for_each_matching_node'
@@ -250,6 +254,7 @@ ForEachMacros: · 53 unmodified lines
250	254	- 'for_each_pci_bridge'
251	255	- 'for_each_pci_dev'
252	256	- 'for_each_pci_msi_entry'
	257	+ - 'for_each_pcm_streams'
253	258	- 'for_each_populated_zone'
254	259	- 'for_each_possible_cpu'
255	260	- 'for_each_present_cpu'
@@ -260,9 +265,12 @@ ForEachMacros: · 4 unmodified lines
260	265	- 'for_each_property_of_node'
261	266	- 'for_each_registered_fb'
262	267	- 'for_each_reserved_mem_region'
263		− - 'for_each_rtd_codec_dai'
264		− - 'for_each_rtd_codec_dai_rollback'
	268	+ - 'for_each_rtd_codec_dais'
	269	+ - 'for_each_rtd_codec_dais_rollback'
265	270	- 'for_each_rtd_components'
	271	+ - 'for_each_rtd_cpu_dais'
	272	+ - 'for_each_rtd_cpu_dais_rollback'
	273	+ - 'for_each_rtd_dais'
266	274	- 'for_each_set_bit'
267	275	- 'for_each_set_bit_from'
268	276	- 'for_each_set_clump8'
@@ -334,6 +342,7 @@ ForEachMacros: · 65 unmodified lines
334	342	- 'klp_for_each_object'
335	343	- 'klp_for_each_object_safe'
336	344	- 'klp_for_each_object_static'
	345	+ - 'kunit_suite_for_each_test_case'
337	346	- 'kvm_for_each_memslot'
338	347	- 'kvm_for_each_vcpu'
339	348	- 'list_for_each'
@@ -387,6 +396,7 @@ ForEachMacros: · 47 unmodified lines
387	396	- 'of_property_for_each_string'
388	397	- 'of_property_for_each_u32'
389	398	- 'pci_bus_for_each_resource'
	399	+ - 'pcm_for_each_format'
390	400	- 'ping_portaddr_for_each_entry'
391	401	- 'plist_for_each'
392	402	- 'plist_for_each_continue'
@@ -482,7 +492,7 @@ KeepEmptyLinesAtTheStartOfBlocks: false · 89 unmodified lines
482	492	MacroBlockBegin: ''
483	493	MacroBlockEnd: ''
484	494	MaxEmptyLinesToKeep: 1
485		−NamespaceIndentation: Inner
	495	+NamespaceIndentation: None
486	496	#ObjCBinPackProtocolList: Auto # Unknown to clang-format-5.0
487	497	ObjCBlockIndentWidth: 8
488	498	ObjCSpaceAfterProperty: true

Documentation/admin-guide/device-mapper/dm-integrity.rst+9−6

@@ -182,12 +182,15 @@ fix_padding · 181 unmodified lines
182	182	space-efficient. If this option is not present, large padding is
183	183	used - that is for compatibility with older kernels.
184	184
185		−
186		−The journal mode (D/J), buffer_sectors, journal_watermark, commit_time can
187		−be changed when reloading the target (load an inactive table and swap the
188		−tables with suspend and resume). The other arguments should not be changed
189		−when reloading the target because the layout of disk data depend on them
190		−and the reloaded target would be non-functional.
	185	+allow_discards
	186	+ Allow block discard requests (a.k.a. TRIM) for the integrity device.
	187	+ Discards are only allowed to devices using internal hash.
	188	+
	189	+The journal mode (D/J), buffer_sectors, journal_watermark, commit_time and
	190	+allow_discards can be changed when reloading the target (load an inactive
	191	+table and swap the tables with suspend and resume). The other arguments
	192	+should not be changed when reloading the target because the layout of disk
	193	+data depend on them and the reloaded target would be non-functional.
191	194
192	195
193	196	The layout of the formatted block device:

Documentation/admin-guide/kernel-parameters.txt+1−2

@@ -5206,8 +5206,7 @@ · 5205 unmodified lines
5206	5206
5207	5207	usbcore.old_scheme_first=
5208	5208	[USB] Start with the old device initialization
5209		− scheme, applies only to low and full-speed devices
5210		− (default 0 = off).
	5209	+ scheme (default 0 = off).
5211	5210
5212	5211	usbcore.usbfs_memory_mb=
5213	5212	[USB] Memory limit (in MB) for buffers allocated by

Documentation/arm64/amu.rst+26−21

@@ -23,13 +23,14 @@ optional external memory-mapped interface. · 22 unmodified lines
23	23
24	24	Version 1 of the Activity Monitors architecture implements a counter group
25	25	of four fixed and architecturally defined 64-bit event counters.
26		− - CPU cycle counter: increments at the frequency of the CPU.
27		− - Constant counter: increments at the fixed frequency of the system
28		− clock.
29		− - Instructions retired: increments with every architecturally executed
30		− instruction.
31		− - Memory stall cycles: counts instruction dispatch stall cycles caused by
32		− misses in the last level cache within the clock domain.
	26	+
	27	+- CPU cycle counter: increments at the frequency of the CPU.
	28	+- Constant counter: increments at the fixed frequency of the system
	29	+ clock.
	30	+- Instructions retired: increments with every architecturally executed
	31	+ instruction.
	32	+- Memory stall cycles: counts instruction dispatch stall cycles caused by
	33	+ misses in the last level cache within the clock domain.
33	34
34	35	When in WFI or WFE these counters do not increment.
35	36
@@ -57,11 +58,12 @@ counters, only the presence of the extension. · 21 unmodified lines
57	58
58	59	Firmware (code running at higher exception levels, e.g. arm-tf) support is
59	60	needed to:
60		− - Enable access for lower exception levels (EL2 and EL1) to the AMU
61		− registers.
62		− - Enable the counters. If not enabled these will read as 0.
63		− - Save/restore the counters before/after the CPU is being put/brought up
64		− from the 'off' power state.
	61	+
	62	+- Enable access for lower exception levels (EL2 and EL1) to the AMU
	63	+ registers.
	64	+- Enable the counters. If not enabled these will read as 0.
	65	+- Save/restore the counters before/after the CPU is being put/brought up
	66	+ from the 'off' power state.
65	67
66	68	When using kernels that have this feature enabled but boot with broken
67	69	firmware the user may experience panics or lockups when accessing the
@@ -78,10 +80,11 @@ are not trapped in EL2/EL3. · 10 unmodified lines
78	80
79	81	The fixed counters of AMUv1 are accessible though the following system
80	82	register definitions:
81		− - SYS_AMEVCNTR0_CORE_EL0
82		− - SYS_AMEVCNTR0_CONST_EL0
83		− - SYS_AMEVCNTR0_INST_RET_EL0
84		− - SYS_AMEVCNTR0_MEM_STALL_EL0
	83	+
	84	+- SYS_AMEVCNTR0_CORE_EL0
	85	+- SYS_AMEVCNTR0_CONST_EL0
	86	+- SYS_AMEVCNTR0_INST_RET_EL0
	87	+- SYS_AMEVCNTR0_MEM_STALL_EL0
85	88
86	89	Auxiliary platform specific counters can be accessed using
87	90	SYS_AMEVCNTR1_EL0(n), where n is a value between 0 and 15.
@@ -93,9 +96,10 @@ Userspace access · 5 unmodified lines
93	96	----------------
94	97
95	98	Currently, access from userspace to the AMU registers is disabled due to:
96		− - Security reasons: they might expose information about code executed in
97		− secure mode.
98		− - Purpose: AMU counters are intended for system management use.
	99	+
	100	+- Security reasons: they might expose information about code executed in
	101	+ secure mode.
	102	+- Purpose: AMU counters are intended for system management use.
99	103
100	104	Also, the presence of the feature is not visible to userspace.
101	105
@@ -105,8 +109,9 @@ Virtualization · 3 unmodified lines
105	109
106	110	Currently, access from userspace (EL0) and kernelspace (EL1) on the KVM
107	111	guest side is disabled due to:
108		− - Security reasons: they might expose information about code executed
109		− by other guests or the host.
	112	+
	113	+- Security reasons: they might expose information about code executed
	114	+ by other guests or the host.
110	115
111	116	Any attempt to access the AMU registers will result in an UNDEFINED
112	117	exception being injected into the guest.

Documentation/devicetree/bindings/Makefile+12−9

@@ -2,6 +2,7 @@ · 1 unmodified line
2	2	DT_DOC_CHECKER ?= dt-doc-validate
3	3	DT_EXTRACT_EX ?= dt-extract-example
4	4	DT_MK_SCHEMA ?= dt-mk-schema
	5	+DT_MK_SCHEMA_USERONLY_FLAG := $(if $(DT_SCHEMA_FILES), -u)
5	6
6	7	quiet_cmd_chk_binding = CHKDT $(patsubst $(srctree)/%,%,$<)
7	8	cmd_chk_binding = $(DT_DOC_CHECKER) -u $(srctree)/$(src) $< ; \
@@ -13,16 +14,18 @@ $(obj)/%.example.dts: $(src)/%.yaml FORCE · 5 unmodified lines
13	14	# Use full schemas when checking %.example.dts
14	15	DT_TMP_SCHEMA := $(obj)/processed-schema-examples.yaml
15	16
	17	+find_cmd = find $(srctree)/$(src) \( -name '*.yaml' ! \
	18	+ -name 'processed-schema*' ! \
	19	+ -name '*.example.dt.yaml' \)
	20	+
16	21	quiet_cmd_mk_schema = SCHEMA $@
17		− cmd_mk_schema = $(DT_MK_SCHEMA) $(DT_MK_SCHEMA_FLAGS) -o $@ $(real-prereqs)
	22	+ cmd_mk_schema = rm -f $@ ; \
	23	+ $(if $(DT_MK_SCHEMA_FLAGS), \
	24	+ echo $(real-prereqs), \
	25	+ $(find_cmd)) \| \
	26	+ xargs $(DT_MK_SCHEMA) $(DT_MK_SCHEMA_FLAGS) >> $@
18	27
19		−DT_DOCS = $(addprefix $(src)/, \
20		− $(shell \
21		− cd $(srctree)/$(src) && \
22		− find * \( -name '*.yaml' ! \
23		− -name 'processed-schema*' ! \
24		− -name '*.example.dt.yaml' \) \
25		− ))
	28	+DT_DOCS = $(shell $(find_cmd) \| sed -e 's\|^$(srctree)/\|\|')
26	29
27	30	DT_SCHEMA_FILES ?= $(DT_DOCS)
28	31
@@ -37,7 +40,7 @@ override DTC_FLAGS := \ · 8 unmodified lines
37	40	$(obj)/processed-schema-examples.yaml: $(DT_DOCS) FORCE
38	41	$(call if_changed,mk_schema)
39	42
40		−$(obj)/processed-schema.yaml: DT_MK_SCHEMA_FLAGS := -u
	43	+$(obj)/processed-schema.yaml: DT_MK_SCHEMA_FLAGS := $(DT_MK_SCHEMA_USERONLY_FLAG)
41	44	$(obj)/processed-schema.yaml: $(DT_SCHEMA_FILES) FORCE
42	45	$(call if_changed,mk_schema)
43	46

Documentation/devicetree/bindings/display/panel/leadtek,ltk500hd1829.yaml+0−1

@@ -37,7 +37,6 @@ examples: · 36 unmodified lines
37	37	dsi {
38	38	#address-cells = <1>;
39	39	#size-cells = <0>;
40		− reg = <0xff450000 0x1000>;
41	40
42	41	panel@0 {
43	42	compatible = "leadtek,ltk500hd1829";

Documentation/devicetree/bindings/display/panel/lvds.yaml+9−1

@@ -96,12 +96,20 @@ properties: · 95 unmodified lines
96	96	If set, reverse the bit order described in the data mappings below on all
97	97	data lanes, transmitting bits for slots 6 to 0 instead of 0 to 6.
98	98
	99	+ port: true
	100	+ ports: true
	101	+
99	102	required:
100	103	- compatible
101	104	- data-mapping
102	105	- width-mm
103	106	- height-mm
104	107	- panel-timing
105		− - port
	108	+
	109	+oneOf:
	110	+ - required:
	111	+ - port
	112	+ - required:
	113	+ - ports
106	114
107	115	...

Documentation/devicetree/bindings/display/panel/xinpeng,xpp055c272.yaml+0−1

@@ -37,7 +37,6 @@ examples: · 36 unmodified lines
37	37	dsi {
38	38	#address-cells = <1>;
39	39	#size-cells = <0>;
40		− reg = <0xff450000 0x1000>;
41	40
42	41	panel@0 {
43	42	compatible = "xinpeng,xpp055c272";

Documentation/devicetree/bindings/dma/socionext,uniphier-xdmac.yaml+3−4

@@ -22,9 +22,7 @@ properties: · 21 unmodified lines
22	22	const: socionext,uniphier-xdmac
23	23
24	24	reg:
25		− items:
26		− - description: XDMAC base register region (offset and length)
27		− - description: XDMAC extension register region (offset and length)
	25	+ maxItems: 1
28	26
29	27	interrupts:
30	28	maxItems: 1
@@ -49,12 +47,13 @@ required: · 18 unmodified lines
49	47	- reg
50	48	- interrupts
51	49	- "#dma-cells"
	50	+ - dma-channels
52	51
53	52	examples:
54	53	- \|
55	54	xdmac: dma-controller@5fc10000 {
56	55	compatible = "socionext,uniphier-xdmac";
57		− reg = <0x5fc10000 0x1000>, <0x5fc20000 0x800>;
	56	+ reg = <0x5fc10000 0x5300>;
58	57	interrupts = <0 188 4>;
59	58	#dma-cells = <2>;
60	59	dma-channels = <16>;

Documentation/devicetree/bindings/iio/adc/st,stm32-adc.yaml+1−1

@@ -1,7 +1,7 @@
1	1	# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause)
2	2	%YAML 1.2
3	3	---
4		−$id: "http://devicetree.org/schemas/bindings/iio/adc/st,stm32-adc.yaml#"
	4	+$id: "http://devicetree.org/schemas/iio/adc/st,stm32-adc.yaml#"
5	5	$schema: "http://devicetree.org/meta-schemas/core.yaml#"
6	6
7	7	title: STMicroelectronics STM32 ADC bindings

Documentation/devicetree/bindings/mfd/st,stpmic1.yaml+0−2

@@ -259,8 +259,6 @@ properties: · 258 unmodified lines
259	259
260	260	additionalProperties: false
261	261
262		− additionalProperties: false
263		−
264	262	additionalProperties: false
265	263
266	264	required:

Documentation/devicetree/bindings/phy/qcom,qusb2-phy.yaml+3−3

@@ -97,7 +97,7 @@ then: · 96 unmodified lines
97	97	- $ref: /schemas/types.yaml#/definitions/uint32
98	98	- minimum: 0
99	99	maximum: 63
100		− default: 0
	100	+ default: 32
101	101
102	102	qcom,charge-ctrl-value:
103	103	description:
@@ -130,7 +130,7 @@ then: · 26 unmodified lines
130	130	- $ref: /schemas/types.yaml#/definitions/uint32
131	131	- minimum: 0
132	132	maximum: 3
133		− default: 2
	133	+ default: 0
134	134
135	135	qcom,preemphasis-width:
136	136	description:
@@ -152,7 +152,7 @@ then: · 15 unmodified lines
152	152	- $ref: /schemas/types.yaml#/definitions/uint32
153	153	- minimum: 0
154	154	maximum: 3
155		− default: 0
	155	+ default: 1
156	156
157	157	required:
158	158	- compatible

Documentation/devicetree/bindings/regulator/mps,mp5416.yaml+0−1

@@ -37,7 +37,6 @@ properties: · 36 unmodified lines
37	37	type: object
38	38
39	39	additionalProperties: false
40		− additionalProperties: false
41	40
42	41	required:
43	42	- compatible

Documentation/devicetree/bindings/regulator/mps,mpq7920.yaml+2−1

@@ -75,7 +75,8 @@ properties: · 74 unmodified lines
75	75	description: \|
76	76	disables over voltage protection of this buck
77	77
78		− additionalProperties: false
	78	+ unevaluatedProperties: false
	79	+
79	80	additionalProperties: false
80	81
81	82	required:

Documentation/devicetree/bindings/regulator/rohm,bd71828-regulator.yaml+5−1

@@ -35,6 +35,8 @@ patternProperties: · 34 unmodified lines
35	35	description:
36	36	should be "ldo1", ..., "ldo7"
37	37
	38	+ unevaluatedProperties: false
	39	+
38	40	"^BUCK[1-7]$":
39	41	type: object
40	42	allOf:
@@ -103,5 +105,7 @@ patternProperties: · 62 unmodified lines
103	105
104	106	required:
105	107	- regulator-name
106		− additionalProperties: false
	108	+
	109	+ unevaluatedProperties: false
	110	+
107	111	additionalProperties: false

Documentation/devicetree/bindings/regulator/rohm,bd71837-regulator.yaml+5−1

@@ -41,6 +41,8 @@ patternProperties: · 40 unmodified lines
41	41	description:
42	42	should be "ldo1", ..., "ldo7"
43	43
	44	+ unevaluatedProperties: false
	45	+
44	46	"^BUCK[1-8]$":
45	47	type: object
46	48	allOf:
@@ -99,5 +101,7 @@ patternProperties: · 52 unmodified lines
99	101
100	102	required:
101	103	- regulator-name
102		− additionalProperties: false
	104	+
	105	+ unevaluatedProperties: false
	106	+
103	107	additionalProperties: false

Documentation/devicetree/bindings/regulator/rohm,bd71847-regulator.yaml+5−1

@@ -40,6 +40,8 @@ patternProperties: · 39 unmodified lines
40	40	description:
41	41	should be "ldo1", ..., "ldo6"
42	42
	43	+ unevaluatedProperties: false
	44	+
43	45	"^BUCK[1-6]$":
44	46	type: object
45	47	allOf:
@@ -93,5 +95,7 @@ patternProperties: · 47 unmodified lines
93	95
94	96	required:
95	97	- regulator-name
96		− additionalProperties: false
	98	+
	99	+ unevaluatedProperties: false
	100	+
97	101	additionalProperties: false

Documentation/devicetree/bindings/sound/rockchip-i2s.yaml+3−0

@@ -56,6 +56,9 @@ properties: · 55 unmodified lines
56	56	- const: tx
57	57	- const: rx
58	58
	59	+ power-domains:
	60	+ maxItems: 1
	61	+
59	62	rockchip,capture-channels:
60	63	allOf:
61	64	- $ref: /schemas/types.yaml#/definitions/uint32

Documentation/devicetree/bindings/sound/rockchip-spdif.txt+0−45

@@ -1,45 +0,0 @@
1		−* Rockchip SPDIF transceiver
2		−
3		−The S/PDIF audio block is a stereo transceiver that allows the
4		−processor to receive and transmit digital audio via an coaxial cable or
5		−a fibre cable.
6		−
7		−Required properties:
8		−
9		−- compatible: should be one of the following:
10		− - "rockchip,rk3066-spdif"
11		− - "rockchip,rk3188-spdif"
12		− - "rockchip,rk3228-spdif"
13		− - "rockchip,rk3288-spdif"
14		− - "rockchip,rk3328-spdif"
15		− - "rockchip,rk3366-spdif"
16		− - "rockchip,rk3368-spdif"
17		− - "rockchip,rk3399-spdif"
18		−- reg: physical base address of the controller and length of memory mapped
19		− region.
20		−- interrupts: should contain the SPDIF interrupt.
21		−- dmas: DMA specifiers for tx dma. See the DMA client binding,
22		− Documentation/devicetree/bindings/dma/dma.txt
23		−- dma-names: should be "tx"
24		−- clocks: a list of phandle + clock-specifier pairs, one for each entry
25		− in clock-names.
26		−- clock-names: should contain following:
27		− - "hclk": clock for SPDIF controller
28		− - "mclk" : clock for SPDIF bus
29		−
30		−Required properties on RK3288:
31		− - rockchip,grf: the phandle of the syscon node for the general register
32		− file (GRF)
33		−
34		−Example for the rk3188 SPDIF controller:
35		−
36		−spdif: spdif@1011e000 {
37		− compatible = "rockchip,rk3188-spdif", "rockchip,rk3066-spdif";
38		− reg = <0x1011e000 0x2000>;
39		− interrupts = <GIC_SPI 32 IRQ_TYPE_LEVEL_HIGH>;
40		− dmas = <&dmac1_s 8>;
41		− dma-names = "tx";
42		− clock-names = "hclk", "mclk";
43		− clocks = <&cru HCLK_SPDIF>, <&cru SCLK_SPDIF>;
44		− #sound-dai-cells = <0>;
45		−};

Documentation/devicetree/bindings/sound/rockchip-spdif.yaml+101−0

@@ -0,0 +1,101 @@
	1	+# SPDX-License-Identifier: GPL-2.0
	2	+%YAML 1.2
	3	+---
	4	+$id: http://devicetree.org/schemas/sound/rockchip-spdif.yaml#
	5	+$schema: http://devicetree.org/meta-schemas/core.yaml#
	6	+
	7	+title: Rockchip SPDIF transceiver
	8	+
	9	+description:
	10	+ The S/PDIF audio block is a stereo transceiver that allows the
	11	+ processor to receive and transmit digital audio via a coaxial or
	12	+ fibre cable.
	13	+
	14	+maintainers:
	15	+ - Heiko Stuebner <heiko@sntech.de>
	16	+
	17	+properties:
	18	+ compatible:
	19	+ oneOf:
	20	+ - const: rockchip,rk3066-spdif
	21	+ - const: rockchip,rk3228-spdif
	22	+ - const: rockchip,rk3328-spdif
	23	+ - const: rockchip,rk3366-spdif
	24	+ - const: rockchip,rk3368-spdif
	25	+ - const: rockchip,rk3399-spdif
	26	+ - items:
	27	+ - enum:
	28	+ - rockchip,rk3188-spdif
	29	+ - rockchip,rk3288-spdif
	30	+ - const: rockchip,rk3066-spdif
	31	+
	32	+ reg:
	33	+ maxItems: 1
	34	+
	35	+ interrupts:
	36	+ maxItems: 1
	37	+
	38	+ clocks:
	39	+ items:
	40	+ - description: clock for SPDIF bus
	41	+ - description: clock for SPDIF controller
	42	+
	43	+ clock-names:
	44	+ items:
	45	+ - const: mclk
	46	+ - const: hclk
	47	+
	48	+ dmas:
	49	+ maxItems: 1
	50	+
	51	+ dma-names:
	52	+ const: tx
	53	+
	54	+ power-domains:
	55	+ maxItems: 1
	56	+
	57	+ rockchip,grf:
	58	+ $ref: /schemas/types.yaml#/definitions/phandle
	59	+ description:
	60	+ The phandle of the syscon node for the GRF register.
	61	+ Required property on RK3288.
	62	+
	63	+ "#sound-dai-cells":
	64	+ const: 0
	65	+
	66	+required:
	67	+ - compatible
	68	+ - reg
	69	+ - interrupts
	70	+ - clocks
	71	+ - clock-names
	72	+ - dmas
	73	+ - dma-names
	74	+ - "#sound-dai-cells"
	75	+
	76	+if:
	77	+ properties:
	78	+ compatible:
	79	+ contains:
	80	+ const: rockchip,rk3288-spdif
	81	+
	82	+then:
	83	+ required:
	84	+ - rockchip,grf
	85	+
	86	+additionalProperties: false
	87	+
	88	+examples:
	89	+ - \|
	90	+ #include <dt-bindings/clock/rk3188-cru.h>
	91	+ #include <dt-bindings/interrupt-controller/arm-gic.h>
	92	+ spdif: spdif@1011e000 {
	93	+ compatible = "rockchip,rk3188-spdif", "rockchip,rk3066-spdif";
	94	+ reg = <0x1011e000 0x2000>;
	95	+ interrupts = <GIC_SPI 32 IRQ_TYPE_LEVEL_HIGH>;
	96	+ clocks = <&cru SCLK_SPDIF>, <&cru HCLK_SPDIF>;
	97	+ clock-names = "mclk", "hclk";
	98	+ dmas = <&dmac1_s 8>;
	99	+ dma-names = "tx";
	100	+ #sound-dai-cells = <0>;
	101	+ };

Documentation/devicetree/bindings/usb/renesas,usb3-peri.yaml+1−0

@@ -18,6 +18,7 @@ properties: · 17 unmodified lines
18	18	- renesas,r8a774c0-usb3-peri # RZ/G2E
19	19	- renesas,r8a7795-usb3-peri # R-Car H3
20	20	- renesas,r8a7796-usb3-peri # R-Car M3-W
	21	+ - renesas,r8a77961-usb3-peri # R-Car M3-W+
21	22	- renesas,r8a77965-usb3-peri # R-Car M3-N
22	23	- renesas,r8a77990-usb3-peri # R-Car E3
23	24	- const: renesas,rcar-gen3-usb3-peri

Documentation/devicetree/bindings/usb/renesas,usbhs.yaml+1−0

@@ -40,6 +40,7 @@ properties: · 39 unmodified lines
40	40	- renesas,usbhs-r8a774c0 # RZ/G2E
41	41	- renesas,usbhs-r8a7795 # R-Car H3
42	42	- renesas,usbhs-r8a7796 # R-Car M3-W
	43	+ - renesas,usbhs-r8a77961 # R-Car M3-W+
43	44	- renesas,usbhs-r8a77965 # R-Car M3-N
44	45	- renesas,usbhs-r8a77990 # R-Car E3
45	46	- renesas,usbhs-r8a77995 # R-Car D3

Documentation/devicetree/bindings/usb/usb-xhci.txt+2−1

@@ -16,7 +16,8 @@ Required properties: · 15 unmodified lines
16	16	- "renesas,xhci-r8a7791" for r8a7791 SoC
17	17	- "renesas,xhci-r8a7793" for r8a7793 SoC
18	18	- "renesas,xhci-r8a7795" for r8a7795 SoC
19		− - "renesas,xhci-r8a7796" for r8a7796 SoC
	19	+ - "renesas,xhci-r8a7796" for r8a77960 SoC
	20	+ - "renesas,xhci-r8a77961" for r8a77961 SoC
20	21	- "renesas,xhci-r8a77965" for r8a77965 SoC
21	22	- "renesas,xhci-r8a77990" for r8a77990 SoC
22	23	- "renesas,rcar-gen2-xhci" for a generic R-Car Gen2 or RZ/G1 compatible

Documentation/filesystems/debugfs.rst+2−2

@@ -79,8 +79,8 @@ created with any of:: · 78 unmodified lines
79	79	struct dentry parent, u8 value);
80	80	void debugfs_create_u16(const char *name, umode_t mode,
81	81	struct dentry parent, u16 value);
82		− struct dentry debugfs_create_u32(const char name, umode_t mode,
83		− struct dentry parent, u32 value);
	82	+ void debugfs_create_u32(const char *name, umode_t mode,
	83	+ struct dentry parent, u32 value);
84	84	void debugfs_create_u64(const char *name, umode_t mode,
85	85	struct dentry parent, u64 value);
86	86

Documentation/kbuild/makefiles.rst+2−1

@@ -1241,7 +1241,8 @@ When kbuild executes, the following steps are followed (roughly): · 1240 unmodified lines
1241	1241	will be displayed with "make KBUILD_VERBOSE=0".
1242	1242
1243	1243
1244		−--- 6.9 Preprocessing linker scripts
	1244	+6.9 Preprocessing linker scripts
	1245	+--------------------------------
1245	1246
1246	1247	When the vmlinux image is built, the linker script
1247	1248	arch/$(ARCH)/kernel/vmlinux.lds is used.

Documentation/networking/devlink/ice.rst+2−2

@@ -61,8 +61,8 @@ The ``ice`` driver reports the following versions · 60 unmodified lines
61	61	- running
62	62	- ICE OS Default Package
63	63	- The name of the DDP package that is active in the device. The DDP
64		− package is loaded by the driver during initialization. Each varation
65		− of DDP package shall have a unique name.
	64	+ package is loaded by the driver during initialization. Each
	65	+ variation of the DDP package has a unique name.
66	66	* - ``fw.app``
67	67	- running
68	68	- 1.3.1.0

Documentation/networking/ip-sysctl.txt+7−0

@@ -983,6 +983,13 @@ ip_early_demux - BOOLEAN · 982 unmodified lines
983	983	reduces overall throughput, in such case you should disable it.
984	984	Default: 1
985	985
	986	+ping_group_range - 2 INTEGERS
	987	+ Restrict ICMP_PROTO datagram sockets to users in the group range.
	988	+ The default is "1 0", meaning, that nobody (not even root) may
	989	+ create ping sockets. Setting it to "100 100" would grant permissions
	990	+ to the single group. "0 4294967295" would enable it for the world, "100
	991	+ 4294967295" would enable it for the users, but not daemons.
	992	+
986	993	tcp_early_demux - BOOLEAN
987	994	Enable early demux for established TCP sockets.
988	995	Default: 1

Documentation/virt/kvm/index.rst+2−0

@@ -28,3 +28,5 @@ KVM · 27 unmodified lines
28	28	arm/index
29	29
30	30	devices/index
	31	+
	32	+ running-nested-guests

Documentation/virt/kvm/running-nested-guests.rst+276−0

@@ -0,0 +1,276 @@
	1	+==============================
	2	+Running nested guests with KVM
	3	+==============================
	4	+
	5	+A nested guest is the ability to run a guest inside another guest (it
	6	+can be KVM-based or a different hypervisor). The straightforward
	7	+example is a KVM guest that in turn runs on a KVM guest (the rest of
	8	+this document is built on this example)::
	9	+
	10	+ .----------------. .----------------.
	11	+ \| \| \| \|
	12	+ \| L2 \| \| L2 \|
	13	+ \| (Nested Guest) \| \| (Nested Guest) \|
	14	+ \| \| \| \|
	15	+ \|----------------'--'----------------\|
	16	+ \| \|
	17	+ \| L1 (Guest Hypervisor) \|
	18	+ \| KVM (/dev/kvm) \|
	19	+ \| \|
	20	+ .------------------------------------------------------.
	21	+ \| L0 (Host Hypervisor) \|
	22	+ \| KVM (/dev/kvm) \|
	23	+ \|------------------------------------------------------\|
	24	+ \| Hardware (with virtualization extensions) \|
	25	+ '------------------------------------------------------'
	26	+
	27	+Terminology:
	28	+
	29	+- L0 – level-0; the bare metal host, running KVM
	30	+
	31	+- L1 – level-1 guest; a VM running on L0; also called the "guest
	32	+ hypervisor", as it itself is capable of running KVM.
	33	+
	34	+- L2 – level-2 guest; a VM running on L1, this is the "nested guest"
	35	+
	36	+.. note:: The above diagram is modelled after the x86 architecture;
	37	+ s390x, ppc64 and other architectures are likely to have
	38	+ a different design for nesting.
	39	+
	40	+ For example, s390x always has an LPAR (LogicalPARtition)
	41	+ hypervisor running on bare metal, adding another layer and
	42	+ resulting in at least four levels in a nested setup — L0 (bare
	43	+ metal, running the LPAR hypervisor), L1 (host hypervisor), L2
	44	+ (guest hypervisor), L3 (nested guest).
	45	+
	46	+ This document will stick with the three-level terminology (L0,
	47	+ L1, and L2) for all architectures; and will largely focus on
	48	+ x86.
	49	+
	50	+
	51	+Use Cases
	52	+---------
	53	+
	54	+There are several scenarios where nested KVM can be useful, to name a
	55	+few:
	56	+
	57	+- As a developer, you want to test your software on different operating
	58	+ systems (OSes). Instead of renting multiple VMs from a Cloud
	59	+ Provider, using nested KVM lets you rent a large enough "guest
	60	+ hypervisor" (level-1 guest). This in turn allows you to create
	61	+ multiple nested guests (level-2 guests), running different OSes, on
	62	+ which you can develop and test your software.
	63	+
	64	+- Live migration of "guest hypervisors" and their nested guests, for
	65	+ load balancing, disaster recovery, etc.
	66	+
	67	+- VM image creation tools (e.g. ``virt-install``, etc) often run
	68	+ their own VM, and users expect these to work inside a VM.
	69	+
	70	+- Some OSes use virtualization internally for security (e.g. to let
	71	+ applications run safely in isolation).
	72	+
	73	+
	74	+Enabling "nested" (x86)
	75	+-----------------------
	76	+
	77	+From Linux kernel v4.19 onwards, the ``nested`` KVM parameter is enabled
	78	+by default for Intel and AMD. (Though your Linux distribution might
	79	+override this default.)
	80	+
	81	+In case you are running a Linux kernel older than v4.19, to enable
	82	+nesting, set the ``nested`` KVM module parameter to ``Y`` or ``1``. To
	83	+persist this setting across reboots, you can add it in a config file, as
	84	+shown below:
	85	+
	86	+1. On the bare metal host (L0), list the kernel modules and ensure that
	87	+ the KVM modules::
	88	+
	89	+ $ lsmod \| grep -i kvm
	90	+ kvm_intel 133627 0
	91	+ kvm 435079 1 kvm_intel
	92	+
	93	+2. Show information for ``kvm_intel`` module::
	94	+
	95	+ $ modinfo kvm_intel \| grep -i nested
	96	+ parm: nested:bool
	97	+
	98	+3. For the nested KVM configuration to persist across reboots, place the
	99	+ below in ``/etc/modprobed/kvm_intel.conf`` (create the file if it
	100	+ doesn't exist)::
	101	+
	102	+ $ cat /etc/modprobe.d/kvm_intel.conf
	103	+ options kvm-intel nested=y
	104	+
	105	+4. Unload and re-load the KVM Intel module::
	106	+
	107	+ $ sudo rmmod kvm-intel
	108	+ $ sudo modprobe kvm-intel
	109	+
	110	+5. Verify if the ``nested`` parameter for KVM is enabled::
	111	+
	112	+ $ cat /sys/module/kvm_intel/parameters/nested
	113	+ Y
	114	+
	115	+For AMD hosts, the process is the same as above, except that the module
	116	+name is ``kvm-amd``.
	117	+
	118	+
	119	+Additional nested-related kernel parameters (x86)
	120	+-------------------------------------------------
	121	+
	122	+If your hardware is sufficiently advanced (Intel Haswell processor or
	123	+higher, which has newer hardware virt extensions), the following
	124	+additional features will also be enabled by default: "Shadow VMCS
	125	+(Virtual Machine Control Structure)", APIC Virtualization on your bare
	126	+metal host (L0). Parameters for Intel hosts::
	127	+
	128	+ $ cat /sys/module/kvm_intel/parameters/enable_shadow_vmcs
	129	+ Y
	130	+
	131	+ $ cat /sys/module/kvm_intel/parameters/enable_apicv
	132	+ Y
	133	+
	134	+ $ cat /sys/module/kvm_intel/parameters/ept
	135	+ Y
	136	+
	137	+.. note:: If you suspect your L2 (i.e. nested guest) is running slower,
	138	+ ensure the above are enabled (particularly
	139	+ ``enable_shadow_vmcs`` and ``ept``).
	140	+
	141	+
	142	+Starting a nested guest (x86)
	143	+-----------------------------
	144	+
	145	+Once your bare metal host (L0) is configured for nesting, you should be
	146	+able to start an L1 guest with::
	147	+
	148	+ $ qemu-kvm -cpu host [...]
	149	+
	150	+The above will pass through the host CPU's capabilities as-is to the
	151	+gues); or for better live migration compatibility, use a named CPU
	152	+model supported by QEMU. e.g.::
	153	+
	154	+ $ qemu-kvm -cpu Haswell-noTSX-IBRS,vmx=on
	155	+
	156	+then the guest hypervisor will subsequently be capable of running a
	157	+nested guest with accelerated KVM.
	158	+
	159	+
	160	+Enabling "nested" (s390x)
	161	+-------------------------
	162	+
	163	+1. On the host hypervisor (L0), enable the ``nested`` parameter on
	164	+ s390x::
	165	+
	166	+ $ rmmod kvm
	167	+ $ modprobe kvm nested=1
	168	+
	169	+.. note:: On s390x, the kernel parameter ``hpage`` is mutually exclusive
	170	+ with the ``nested`` paramter — i.e. to be able to enable
	171	+ ``nested``, the ``hpage`` parameter must be disabled.
	172	+
	173	+2. The guest hypervisor (L1) must be provided with the ``sie`` CPU
	174	+ feature — with QEMU, this can be done by using "host passthrough"
	175	+ (via the command-line ``-cpu host``).
	176	+
	177	+3. Now the KVM module can be loaded in the L1 (guest hypervisor)::
	178	+
	179	+ $ modprobe kvm
	180	+
	181	+
	182	+Live migration with nested KVM
	183	+------------------------------
	184	+
	185	+Migrating an L1 guest, with a live nested guest in it, to another
	186	+bare metal host, works as of Linux kernel 5.3 and QEMU 4.2.0 for
	187	+Intel x86 systems, and even on older versions for s390x.
	188	+
	189	+On AMD systems, once an L1 guest has started an L2 guest, the L1 guest
	190	+should no longer be migrated or saved (refer to QEMU documentation on
	191	+"savevm"/"loadvm") until the L2 guest shuts down. Attempting to migrate
	192	+or save-and-load an L1 guest while an L2 guest is running will result in
	193	+undefined behavior. You might see a ``kernel BUG!`` entry in ``dmesg``, a
	194	+kernel 'oops', or an outright kernel panic. Such a migrated or loaded L1
	195	+guest can no longer be considered stable or secure, and must be restarted.
	196	+Migrating an L1 guest merely configured to support nesting, while not
	197	+actually running L2 guests, is expected to function normally even on AMD
	198	+systems but may fail once guests are started.
	199	+
	200	+Migrating an L2 guest is always expected to succeed, so all the following
	201	+scenarios should work even on AMD systems:
	202	+
	203	+- Migrating a nested guest (L2) to another L1 guest on the same bare
	204	+ metal host.
	205	+
	206	+- Migrating a nested guest (L2) to another L1 guest on a different
	207	+ bare metal host.
	208	+
	209	+- Migrating a nested guest (L2) to a bare metal host.
	210	+
	211	+Reporting bugs from nested setups
	212	+-----------------------------------
	213	+
	214	+Debugging "nested" problems can involve sifting through log files across
	215	+L0, L1 and L2; this can result in tedious back-n-forth between the bug
	216	+reporter and the bug fixer.
	217	+
	218	+- Mention that you are in a "nested" setup. If you are running any kind
	219	+ of "nesting" at all, say so. Unfortunately, this needs to be called
	220	+ out because when reporting bugs, people tend to forget to even
	221	+ mention that they're using nested virtualization.
	222	+
	223	+- Ensure you are actually running KVM on KVM. Sometimes people do not
	224	+ have KVM enabled for their guest hypervisor (L1), which results in
	225	+ them running with pure emulation or what QEMU calls it as "TCG", but
	226	+ they think they're running nested KVM. Thus confusing "nested Virt"
	227	+ (which could also mean, QEMU on KVM) with "nested KVM" (KVM on KVM).
	228	+
	229	+Information to collect (generic)
	230	+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
	231	+
	232	+The following is not an exhaustive list, but a very good starting point:
	233	+
	234	+ - Kernel, libvirt, and QEMU version from L0
	235	+
	236	+ - Kernel, libvirt and QEMU version from L1
	237	+
	238	+ - QEMU command-line of L1 -- when using libvirt, you'll find it here:
	239	+ ``/var/log/libvirt/qemu/instance.log``
	240	+
	241	+ - QEMU command-line of L2 -- as above, when using libvirt, get the
	242	+ complete libvirt-generated QEMU command-line
	243	+
	244	+ - ``cat /sys/cpuinfo`` from L0
	245	+
	246	+ - ``cat /sys/cpuinfo`` from L1
	247	+
	248	+ - ``lscpu`` from L0
	249	+
	250	+ - ``lscpu`` from L1
	251	+
	252	+ - Full ``dmesg`` output from L0
	253	+
	254	+ - Full ``dmesg`` output from L1
	255	+
	256	+x86-specific info to collect
	257	+~~~~~~~~~~~~~~~~~~~~~~~~~~~~
	258	+
	259	+Both the below commands, ``x86info`` and ``dmidecode``, should be
	260	+available on most Linux distributions with the same name:
	261	+
	262	+ - Output of: ``x86info -a`` from L0
	263	+
	264	+ - Output of: ``x86info -a`` from L1
	265	+
	266	+ - Output of: ``dmidecode`` from L0
	267	+
	268	+ - Output of: ``dmidecode`` from L1
	269	+
	270	+s390x-specific info to collect
	271	+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
	272	+
	273	+Along with the earlier mentioned generic details, the below is
	274	+also recommended:
	275	+
	276	+ - ``/proc/sysinfo`` from L1; this will also include the info from L0

MAINTAINERS+42−46

@@ -189,7 +189,7 @@ F: drivers/net/hamradio/6pack.c · 188 unmodified lines
189	189	M: Johannes Berg <johannes@sipsolutions.net>
190	190	L: linux-wireless@vger.kernel.org
191	191	S: Maintained
192		−W: http://wireless.kernel.org/
	192	+W: https://wireless.wiki.kernel.org/
193	193	T: git git://git.kernel.org/pub/scm/linux/kernel/git/jberg/mac80211.git
194	194	T: git git://git.kernel.org/pub/scm/linux/kernel/git/jberg/mac80211-next.git
195	195	F: Documentation/driver-api/80211/cfg80211.rst
@@ -505,7 +505,7 @@ F: drivers/hwmon/adm1029.c · 309 unmodified lines
505	505	ADM8211 WIRELESS DRIVER
506	506	L: linux-wireless@vger.kernel.org
507	507	S: Orphan
508		−W: http://wireless.kernel.org/
	508	+W: https://wireless.wiki.kernel.org/
509	509	F: drivers/net/wireless/admtek/adm8211.*
510	510
511	511	ADP1653 FLASH CONTROLLER DRIVER
@@ -570,7 +570,7 @@ F: Documentation/devicetree/bindings/iio/accel/adi,adxl345.yaml · 58 unmodified lines
570	570	F: drivers/input/misc/adxl34x.c
571	571
572	572	ADXL372 THREE-AXIS DIGITAL ACCELEROMETER DRIVER
573		−M: Stefan Popa <stefan.popa@analog.com>
	573	+M: Michael Hennerich <michael.hennerich@analog.com>
574	574	S: Supported
575	575	W: http://ez.analog.com/community/linux-device-drivers
576	576	F: Documentation/devicetree/bindings/iio/accel/adi,adxl372.yaml
@@ -922,7 +922,7 @@ F: arch/arm64/boot/dts/amd/amd-seattle-xgbe*.dtsi · 345 unmodified lines
922	922	F: drivers/net/ethernet/amd/xgbe/
923	923
924	924	ANALOG DEVICES INC AD5686 DRIVER
925		−M: Stefan Popa <stefan.popa@analog.com>
	925	+M: Michael Hennerich <Michael.Hennerich@analog.com>
926	926	L: linux-pm@vger.kernel.org
927	927	S: Supported
928	928	W: http://ez.analog.com/community/linux-device-drivers
@@ -930,7 +930,7 @@ F: drivers/iio/dac/ad5686* · 1 unmodified line
930	930	F: drivers/iio/dac/ad5696*
931	931
932	932	ANALOG DEVICES INC AD5758 DRIVER
933		−M: Stefan Popa <stefan.popa@analog.com>
	933	+M: Michael Hennerich <Michael.Hennerich@analog.com>
934	934	L: linux-iio@vger.kernel.org
935	935	S: Supported
936	936	W: http://ez.analog.com/community/linux-device-drivers
@@ -946,7 +946,7 @@ F: Documentation/devicetree/bindings/iio/adc/adi,ad7091r5.yaml · 9 unmodified lines
946	946	F: drivers/iio/adc/ad7091r5.c
947	947
948	948	ANALOG DEVICES INC AD7124 DRIVER
949		−M: Stefan Popa <stefan.popa@analog.com>
	949	+M: Michael Hennerich <Michael.Hennerich@analog.com>
950	950	L: linux-iio@vger.kernel.org
951	951	S: Supported
952	952	W: http://ez.analog.com/community/linux-device-drivers
@@ -970,7 +970,7 @@ F: Documentation/devicetree/bindings/iio/adc/adi,ad7292.yaml · 17 unmodified lines
970	970	F: drivers/iio/adc/ad7292.c
971	971
972	972	ANALOG DEVICES INC AD7606 DRIVER
973		−M: Stefan Popa <stefan.popa@analog.com>
	973	+M: Michael Hennerich <Michael.Hennerich@analog.com>
974	974	M: Beniamin Bia <beniamin.bia@analog.com>
975	975	L: linux-iio@vger.kernel.org
976	976	S: Supported
@@ -979,7 +979,7 @@ F: Documentation/devicetree/bindings/iio/adc/adi,ad7606.yaml · 2 unmodified lines
979	979	F: drivers/iio/adc/ad7606.c
980	980
981	981	ANALOG DEVICES INC AD7768-1 DRIVER
982		−M: Stefan Popa <stefan.popa@analog.com>
	982	+M: Michael Hennerich <Michael.Hennerich@analog.com>
983	983	L: linux-iio@vger.kernel.org
984	984	S: Supported
985	985	W: http://ez.analog.com/community/linux-device-drivers
@@ -1040,7 +1040,7 @@ F: Documentation/devicetree/bindings/hwmon/adi,adm1177.yaml · 54 unmodified lines
1040	1040	F: drivers/hwmon/adm1177.c
1041	1041
1042	1042	ANALOG DEVICES INC ADP5061 DRIVER
1043		−M: Stefan Popa <stefan.popa@analog.com>
	1043	+M: Michael Hennerich <Michael.Hennerich@analog.com>
1044	1044	L: linux-pm@vger.kernel.org
1045	1045	S: Supported
1046	1046	W: http://ez.analog.com/community/linux-device-drivers
@@ -1109,7 +1109,6 @@ F: drivers/iio/amplifiers/hmc425a.c · 62 unmodified lines
1109	1109	ANALOG DEVICES INC IIO DRIVERS
1110	1110	M: Lars-Peter Clausen <lars@metafoo.de>
1111	1111	M: Michael Hennerich <Michael.Hennerich@analog.com>
1112		−M: Stefan Popa <stefan.popa@analog.com>
1113	1112	S: Supported
1114	1113	W: http://wiki.analog.com/
1115	1114	W: http://ez.analog.com/community/linux-device-drivers
@@ -2850,14 +2849,14 @@ M: Nick Kossifidis <mickflemm@gmail.com> · 1734 unmodified lines
2850	2849	M: Luis Chamberlain <mcgrof@kernel.org>
2851	2850	L: linux-wireless@vger.kernel.org
2852	2851	S: Maintained
2853		−W: http://wireless.kernel.org/en/users/Drivers/ath5k
	2852	+W: https://wireless.wiki.kernel.org/en/users/Drivers/ath5k
2854	2853	F: drivers/net/wireless/ath/ath5k/
2855	2854
2856	2855	ATHEROS ATH6KL WIRELESS DRIVER
2857	2856	M: Kalle Valo <kvalo@codeaurora.org>
2858	2857	L: linux-wireless@vger.kernel.org
2859	2858	S: Supported
2860		−W: http://wireless.kernel.org/en/users/Drivers/ath6kl
	2859	+W: https://wireless.wiki.kernel.org/en/users/Drivers/ath6kl
2861	2860	T: git git://git.kernel.org/pub/scm/linux/kernel/git/kvalo/ath.git
2862	2861	F: drivers/net/wireless/ath/ath6kl/
2863	2862
@@ -3020,7 +3019,7 @@ B43 WIRELESS DRIVER · 156 unmodified lines
3020	3019	L: linux-wireless@vger.kernel.org
3021	3020	L: b43-dev@lists.infradead.org
3022	3021	S: Odd Fixes
3023		−W: http://wireless.kernel.org/en/users/Drivers/b43
	3022	+W: https://wireless.wiki.kernel.org/en/users/Drivers/b43
3024	3023	F: drivers/net/wireless/broadcom/b43/
3025	3024
3026	3025	B43LEGACY WIRELESS DRIVER
@@ -3028,7 +3027,7 @@ M: Larry Finger <Larry.Finger@lwfinger.net> · 1 unmodified line
3028	3027	L: linux-wireless@vger.kernel.org
3029	3028	L: b43-dev@lists.infradead.org
3030	3029	S: Maintained
3031		−W: http://wireless.kernel.org/en/users/Drivers/b43
	3030	+W: https://wireless.wiki.kernel.org/en/users/Drivers/b43
3032	3031	F: drivers/net/wireless/broadcom/b43legacy/
3033	3032
3034	3033	BACKLIGHT CLASS/SUBSYSTEM
@@ -3658,7 +3657,7 @@ L: linux-btrfs@vger.kernel.org · 623 unmodified lines
3658	3657	S: Maintained
3659	3658	W: http://btrfs.wiki.kernel.org/
3660	3659	Q: http://patchwork.kernel.org/project/linux-btrfs/list/
3661		−T: git git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs.git
	3660	+T: git git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux.git
3662	3661	F: Documentation/filesystems/btrfs.rst
3663	3662	F: fs/btrfs/
3664	3663	F: include/linux/btrfs*
@@ -3843,7 +3842,7 @@ CARL9170 LINUX COMMUNITY WIRELESS DRIVER · 178 unmodified lines
3843	3842	M: Christian Lamparter <chunkeey@googlemail.com>
3844	3843	L: linux-wireless@vger.kernel.org
3845	3844	S: Maintained
3846		−W: http://wireless.kernel.org/en/users/Drivers/carl9170
	3845	+W: https://wireless.wiki.kernel.org/en/users/Drivers/carl9170
3847	3846	F: drivers/net/wireless/ath/carl9170/
3848	3847
3849	3848	CAVIUM I2C DRIVER
@@ -3937,11 +3936,9 @@ F: arch/powerpc/platforms/cell/ · 87 unmodified lines
3937	3936	CEPH COMMON CODE (LIBCEPH)
3938	3937	M: Ilya Dryomov <idryomov@gmail.com>
3939	3938	M: Jeff Layton <jlayton@kernel.org>
3940		−M: Sage Weil <sage@redhat.com>
3941	3939	L: ceph-devel@vger.kernel.org
3942	3940	S: Supported
3943	3941	W: http://ceph.com/
3944		−T: git git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client.git
3945	3942	T: git git://github.com/ceph/ceph-client.git
3946	3943	F: include/linux/ceph/
3947	3944	F: include/linux/crush/
@@ -3949,12 +3946,10 @@ F: net/ceph/ · 1 unmodified line
3949	3946
3950	3947	CEPH DISTRIBUTED FILE SYSTEM CLIENT (CEPH)
3951	3948	M: Jeff Layton <jlayton@kernel.org>
3952		−M: Sage Weil <sage@redhat.com>
3953	3949	M: Ilya Dryomov <idryomov@gmail.com>
3954	3950	L: ceph-devel@vger.kernel.org
3955	3951	S: Supported
3956	3952	W: http://ceph.com/
3957		−T: git git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client.git
3958	3953	T: git git://github.com/ceph/ceph-client.git
3959	3954	F: Documentation/filesystems/ceph.rst
3960	3955	F: fs/ceph/
@@ -5176,6 +5171,7 @@ S: Maintained · 1215 unmodified lines
5176	5171	F: drivers/soc/fsl/dpio
5177	5172
5178	5173	DPAA2 ETHERNET DRIVER
	5174	+M: Ioana Ciornei <ioana.ciornei@nxp.com>
5179	5175	M: Ioana Radulescu <ruxandra.radulescu@nxp.com>
5180	5176	L: netdev@vger.kernel.org
5181	5177	S: Maintained
@@ -5935,9 +5931,9 @@ F: lib/dynamic_debug.c · 753 unmodified lines
5935	5931	DYNAMIC INTERRUPT MODERATION
5936	5932	M: Tal Gilboa <talgi@mellanox.com>
5937	5933	S: Maintained
	5934	+F: Documentation/networking/net_dim.rst
5938	5935	F: include/linux/dim.h
5939	5936	F: lib/dim/
5940		−F: Documentation/networking/net_dim.rst
5941	5937
5942	5938	DZ DECSTATION DZ11 SERIAL DRIVER
5943	5939	M: "Maciej W. Rozycki" <macro@linux-mips.org>
@@ -7119,9 +7115,10 @@ F: include/uapi/asm-generic/ · 1175 unmodified lines
7119	7115
7120	7116	GENERIC PHY FRAMEWORK
7121	7117	M: Kishon Vijay Abraham I <kishon@ti.com>
	7118	+M: Vinod Koul <vkoul@kernel.org>
7122	7119	L: linux-kernel@vger.kernel.org
7123	7120	S: Supported
7124		−T: git git://git.kernel.org/pub/scm/linux/kernel/git/kishon/linux-phy.git
	7121	+T: git git://git.kernel.org/pub/scm/linux/kernel/git/phy/linux-phy.git
7125	7122	F: Documentation/devicetree/bindings/phy/
7126	7123	F: drivers/phy/
7127	7124	F: include/linux/phy/
@@ -7746,11 +7743,6 @@ L: platform-driver-x86@vger.kernel.org · 618 unmodified lines
7746	7743	S: Orphan
7747	7744	F: drivers/platform/x86/tc1100-wmi.c
7748	7745
7749		−HP100: Driver for HP 10/100 Mbit/s Voice Grade Network Adapter Series
7750		−M: Jaroslav Kysela <perex@perex.cz>
7751		−S: Obsolete
7752		−F: drivers/staging/hp/hp100.*
7753		−
7754	7746	HPET: High Precision Event Timers driver
7755	7747	M: Clemens Ladisch <clemens@ladisch.de>
7756	7748	S: Maintained
@@ -9329,6 +9321,7 @@ M: Christian Borntraeger <borntraeger@de.ibm.com> · 1572 unmodified lines
9329	9321	M: Janosch Frank <frankja@linux.ibm.com>
9330	9322	R: David Hildenbrand <david@redhat.com>
9331	9323	R: Cornelia Huck <cohuck@redhat.com>
	9324	+R: Claudio Imbrenda <imbrenda@linux.ibm.com>
9332	9325	L: kvm@vger.kernel.org
9333	9326	S: Supported
9334	9327	W: http://www.ibm.com/developerworks/linux/linux390/
@@ -9416,6 +9409,13 @@ F: include/linux/keyctl.h · 81 unmodified lines
9416	9409	F: include/uapi/linux/keyctl.h
9417	9410	F: security/keys/
9418	9411
	9412	+KFIFO
	9413	+M: Stefani Seibold <stefani@seibold.net>
	9414	+S: Maintained
	9415	+F: include/linux/kfifo.h
	9416	+F: lib/kfifo.c
	9417	+F: samples/kfifo/
	9418	+
9419	9419	KGDB / KDB /debug_core
9420	9420	M: Jason Wessel <jason.wessel@windriver.com>
9421	9421	M: Daniel Thompson <daniel.thompson@linaro.org>
@@ -10067,7 +10067,7 @@ MAC80211 · 645 unmodified lines
10067	10067	M: Johannes Berg <johannes@sipsolutions.net>
10068	10068	L: linux-wireless@vger.kernel.org
10069	10069	S: Maintained
10070		−W: http://wireless.kernel.org/
	10070	+W: https://wireless.wiki.kernel.org/
10071	10071	T: git git://git.kernel.org/pub/scm/linux/kernel/git/jberg/mac80211.git
10072	10072	T: git git://git.kernel.org/pub/scm/linux/kernel/git/jberg/mac80211-next.git
10073	10073	F: Documentation/networking/mac80211-injection.txt
@@ -10697,7 +10697,6 @@ MEDIATEK MT76 WIRELESS LAN DRIVER · 623 unmodified lines
10697	10697	M: Felix Fietkau <nbd@nbd.name>
10698	10698	M: Lorenzo Bianconi <lorenzo.bianconi83@gmail.com>
10699	10699	R: Ryder Lee <ryder.lee@mediatek.com>
10700		−R: Roy Luo <royluo@google.com>
10701	10700	L: linux-wireless@vger.kernel.org
10702	10701	S: Maintained
10703	10702	F: drivers/net/wireless/mediatek/mt76/
@@ -12648,7 +12647,7 @@ F: fs/orangefs/ · 1944 unmodified lines
12648	12647	ORINOCO DRIVER
12649	12648	L: linux-wireless@vger.kernel.org
12650	12649	S: Orphan
12651		−W: http://wireless.kernel.org/en/users/Drivers/orinoco
	12650	+W: https://wireless.wiki.kernel.org/en/users/Drivers/orinoco
12652	12651	W: http://www.nongnu.org/orinoco/
12653	12652	F: drivers/net/wireless/intersil/orinoco/
12654	12653
@@ -12674,7 +12673,7 @@ P54 WIRELESS DRIVER · 19 unmodified lines
12674	12673	M: Christian Lamparter <chunkeey@googlemail.com>
12675	12674	L: linux-wireless@vger.kernel.org
12676	12675	S: Maintained
12677		−W: http://wireless.kernel.org/en/users/Drivers/p54
	12676	+W: https://wireless.wiki.kernel.org/en/users/Drivers/p54
12678	12677	F: drivers/net/wireless/intersil/p54/
12679	12678
12680	12679	PACKING
@@ -13042,7 +13041,7 @@ F: drivers/pci/controller/pci-xgene-msi.c · 361 unmodified lines
13042	13041
13043	13042	PCI NATIVE HOST BRIDGE AND ENDPOINT DRIVERS
13044	13043	M: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
13045		−R: Andrew Murray <amurray@thegoodpenguin.co.uk>
	13044	+R: Rob Herring <robh@kernel.org>
13046	13045	L: linux-pci@vger.kernel.org
13047	13046	S: Supported
13048	13047	Q: http://patchwork.ozlabs.org/project/linux-pci/list/
@@ -13595,7 +13594,7 @@ PRISM54 WIRELESS DRIVER · 546 unmodified lines
13595	13594	M: Luis Chamberlain <mcgrof@kernel.org>
13596	13595	L: linux-wireless@vger.kernel.org
13597	13596	S: Obsolete
13598		−W: http://wireless.kernel.org/en/users/Drivers/p54
	13597	+W: https://wireless.wiki.kernel.org/en/users/Drivers/p54
13599	13598	F: drivers/net/wireless/intersil/prism54/
13600	13599
13601	13600	PROC FILESYSTEM
@@ -13936,7 +13935,7 @@ QUALCOMM ATHEROS ATH10K WIRELESS DRIVER · 334 unmodified lines
13936	13935	M: Kalle Valo <kvalo@codeaurora.org>
13937	13936	L: ath10k@lists.infradead.org
13938	13937	S: Supported
13939		−W: http://wireless.kernel.org/en/users/Drivers/ath10k
	13938	+W: https://wireless.wiki.kernel.org/en/users/Drivers/ath10k
13940	13939	T: git git://git.kernel.org/pub/scm/linux/kernel/git/kvalo/ath.git
13941	13940	F: drivers/net/wireless/ath/ath10k/
13942	13941
@@ -13951,7 +13950,7 @@ QUALCOMM ATHEROS ATH9K WIRELESS DRIVER · 8 unmodified lines
13951	13950	M: QCA ath9k Development <ath9k-devel@qca.qualcomm.com>
13952	13951	L: linux-wireless@vger.kernel.org
13953	13952	S: Supported
13954		−W: http://wireless.kernel.org/en/users/Drivers/ath9k
	13953	+W: https://wireless.wiki.kernel.org/en/users/Drivers/ath9k
13955	13954	F: drivers/net/wireless/ath/ath9k/
13956	13955
13957	13956	QUALCOMM CAMERA SUBSYSTEM DRIVER
@@ -14048,13 +14047,12 @@ QUALCOMM WCN36XX WIRELESS DRIVER · 90 unmodified lines
14048	14047	M: Kalle Valo <kvalo@codeaurora.org>
14049	14048	L: wcn36xx@lists.infradead.org
14050	14049	S: Supported
14051		−W: http://wireless.kernel.org/en/users/Drivers/wcn36xx
	14050	+W: https://wireless.wiki.kernel.org/en/users/Drivers/wcn36xx
14052	14051	T: git git://github.com/KrasnikovEugene/wcn36xx.git
14053	14052	F: drivers/net/wireless/ath/wcn36xx/
14054	14053
14055	14054	QUANTENNA QTNFMAC WIRELESS DRIVER
14056	14055	M: Igor Mitsyanko <imitsyanko@quantenna.com>
14057		−M: Avinash Patil <avinashp@quantenna.com>
14058	14056	M: Sergey Matyukevich <smatyukevich@quantenna.com>
14059	14057	L: linux-wireless@vger.kernel.org
14060	14058	S: Maintained
@@ -14096,12 +14094,10 @@ F: drivers/media/radio/radio-tea5777.c · 35 unmodified lines
14096	14094
14097	14095	RADOS BLOCK DEVICE (RBD)
14098	14096	M: Ilya Dryomov <idryomov@gmail.com>
14099		−M: Sage Weil <sage@redhat.com>
14100	14097	R: Dongsheng Yang <dongsheng.yang@easystack.cn>
14101	14098	L: ceph-devel@vger.kernel.org
14102	14099	S: Supported
14103	14100	W: http://ceph.com/
14104		−T: git git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client.git
14105	14101	T: git git://github.com/ceph/ceph-client.git
14106	14102	F: Documentation/ABI/testing/sysfs-bus-rbd
14107	14103	F: drivers/block/rbd.c
@@ -14276,7 +14272,7 @@ REALTEK WIRELESS DRIVER (rtlwifi family) · 168 unmodified lines
14276	14272	M: Ping-Ke Shih <pkshih@realtek.com>
14277	14273	L: linux-wireless@vger.kernel.org
14278	14274	S: Maintained
14279		−W: http://wireless.kernel.org/
	14275	+W: https://wireless.wiki.kernel.org/
14280	14276	T: git git://git.kernel.org/pub/scm/linux/kernel/git/linville/wireless-testing.git
14281	14277	F: drivers/net/wireless/realtek/rtlwifi/
14282	14278
@@ -14411,7 +14407,7 @@ RFKILL · 128 unmodified lines
14411	14407	M: Johannes Berg <johannes@sipsolutions.net>
14412	14408	L: linux-wireless@vger.kernel.org
14413	14409	S: Maintained
14414		−W: http://wireless.kernel.org/
	14410	+W: https://wireless.wiki.kernel.org/
14415	14411	T: git git://git.kernel.org/pub/scm/linux/kernel/git/jberg/mac80211.git
14416	14412	T: git git://git.kernel.org/pub/scm/linux/kernel/git/jberg/mac80211-next.git
14417	14413	F: Documentation/ABI/stable/sysfs-class-rfkill
@@ -14560,7 +14556,7 @@ F: drivers/media/dvb-frontends/rtl2832_sdr* · 142 unmodified lines
14560	14556	RTL8180 WIRELESS DRIVER
14561	14557	L: linux-wireless@vger.kernel.org
14562	14558	S: Orphan
14563		−W: http://wireless.kernel.org/
	14559	+W: https://wireless.wiki.kernel.org/
14564	14560	T: git git://git.kernel.org/pub/scm/linux/kernel/git/linville/wireless-testing.git
14565	14561	F: drivers/net/wireless/realtek/rtl818x/rtl8180/
14566	14562
@@ -14570,7 +14566,7 @@ M: Hin-Tak Leung <htl10@users.sourceforge.net> · 3 unmodified lines
14570	14566	M: Larry Finger <Larry.Finger@lwfinger.net>
14571	14567	L: linux-wireless@vger.kernel.org
14572	14568	S: Maintained
14573		−W: http://wireless.kernel.org/
	14569	+W: https://wireless.wiki.kernel.org/
14574	14570	T: git git://git.kernel.org/pub/scm/linux/kernel/git/linville/wireless-testing.git
14575	14571	F: drivers/net/wireless/realtek/rtl818x/rtl8187/
14576	14572
@@ -16925,8 +16921,8 @@ F: drivers/media/platform/ti-vpe/ · 2348 unmodified lines
16925	16921	TI WILINK WIRELESS DRIVERS
16926	16922	L: linux-wireless@vger.kernel.org
16927	16923	S: Orphan
16928		−W: http://wireless.kernel.org/en/users/Drivers/wl12xx
16929		−W: http://wireless.kernel.org/en/users/Drivers/wl1251
	16924	+W: https://wireless.wiki.kernel.org/en/users/Drivers/wl12xx
	16925	+W: https://wireless.wiki.kernel.org/en/users/Drivers/wl1251
16930	16926	T: git git://git.kernel.org/pub/scm/linux/kernel/git/luca/wl12xx.git
16931	16927	F: drivers/net/wireless/ti/
16932	16928	F: include/linux/wl12xx.h
@@ -18208,7 +18204,7 @@ M: Maya Erez <merez@codeaurora.org> · 1275 unmodified lines
18208	18204	L: linux-wireless@vger.kernel.org
18209	18205	L: wil6210@qti.qualcomm.com
18210	18206	S: Supported
18211		−W: http://wireless.kernel.org/en/users/Drivers/wil6210
	18207	+W: https://wireless.wiki.kernel.org/en/users/Drivers/wil6210
18212	18208	F: drivers/net/wireless/ath/wil6210/
18213	18209
18214	18210	WIMAX STACK

Makefile+12−5

@@ -2,7 +2,7 @@ · 1 unmodified line
2	2	VERSION = 5
3	3	PATCHLEVEL = 7
4	4	SUBLEVEL = 0
5		−EXTRAVERSION = -rc2
	5	+EXTRAVERSION = -rc5
6	6	NAME = Kleptomaniac Octopus
7	7
8	8	# DOCUMENTATION
@@ -729,10 +729,6 @@ else ifdef CONFIG_CC_OPTIMIZE_FOR_SIZE · 720 unmodified lines
729	729	KBUILD_CFLAGS += -Os
730	730	endif
731	731
732		−ifdef CONFIG_CC_DISABLE_WARN_MAYBE_UNINITIALIZED
733		−KBUILD_CFLAGS += -Wno-maybe-uninitialized
734		−endif
735		−
736	732	# Tell gcc to never replace conditional load with a non-conditional one
737	733	KBUILD_CFLAGS += $(call cc-option,--param=allow-store-data-races=0)
738	734	KBUILD_CFLAGS += $(call cc-option,-fno-allow-store-data-races)
@@ -881,6 +877,17 @@ KBUILD_CFLAGS += -Wno-pointer-sign · 142 unmodified lines
881	877	# disable stringop warnings in gcc 8+
882	878	KBUILD_CFLAGS += $(call cc-disable-warning, stringop-truncation)
883	879
	880	+# We'll want to enable this eventually, but it's not going away for 5.7 at least
	881	+KBUILD_CFLAGS += $(call cc-disable-warning, zero-length-bounds)
	882	+KBUILD_CFLAGS += $(call cc-disable-warning, array-bounds)
	883	+KBUILD_CFLAGS += $(call cc-disable-warning, stringop-overflow)
	884	+
	885	+# Another good warning that we'll want to enable eventually
	886	+KBUILD_CFLAGS += $(call cc-disable-warning, restrict)
	887	+
	888	+# Enabled with W=2, disabled by default as noisy
	889	+KBUILD_CFLAGS += $(call cc-disable-warning, maybe-uninitialized)
	890	+
884	891	# disable invalid "can't wrap" optimizations for signed / pointers
885	892	KBUILD_CFLAGS += $(call cc-option,-fno-strict-overflow)
886	893

arch/arc/include/asm/module.h+0−5

@@ -3,7 +3,6 @@ · 2 unmodified lines
3	3	* Copyright (C) 2004, 2007-2010, 2011-2012 Synopsys, Inc. (www.synopsys.com)
4	4	*
5	5	* Amit Bhor, Sameer Dhavale: Codito Technologies 2004
6		−
7	6	*/
8	7
9	8	#ifndef _ASM_ARC_MODULE_H
@@ -19,8 +18,4 @@ struct mod_arch_specific { · 9 unmodified lines
19	18	const char *secstr;
20	19	};
21	20
22		−#define MODULE_PROC_FAMILY "ARC700"
23		−
24		−#define MODULE_ARCH_VERMAGIC MODULE_PROC_FAMILY
25		−
26	21	#endif /* _ASM_ARC_MODULE_H */

arch/arc/include/asm/vermagic.h+8−0

@@ -0,0 +1,8 @@
	1	+/* SPDX-License-Identifier: GPL-2.0 */
	2	+
	3	+#ifndef _ASM_VERMAGIC_H
	4	+#define _ASM_VERMAGIC_H
	5	+
	6	+#define MODULE_ARCH_VERMAGIC "ARC700"
	7	+
	8	+#endif /* _ASM_VERMAGIC_H */

arch/arm/boot/dts/bcm2835-rpi.dtsi+3−0

@@ -14,6 +14,9 @@ · 13 unmodified lines
14	14	soc {
15	15	firmware: firmware {
16	16	compatible = "raspberrypi,bcm2835-firmware", "simple-bus";
	17	+ #address-cells = <1>;
	18	+ #size-cells = <1>;
	19	+
17	20	mboxes = <&mailbox>;
18	21	dma-ranges;
19	22	};

arch/arm/boot/dts/bcm283x.dtsi+1−0

@@ -372,6 +372,7 @@ · 371 unmodified lines
372	372	"dsi0_ddr2",
373	373	"dsi0_ddr";
374	374
	375	+ status = "disabled";
375	376	};
376	377
377	378	aux: aux@7e215000 {

arch/arm/boot/dts/omap3-n950-n9.dtsi+5−0

@@ -341,6 +341,11 @@ · 340 unmodified lines
341	341	status = "disabled";
342	342	};
343	343
	344	+/* RNG not directly accessible on N950/N9. */
	345	+&rng_target {
	346	+ status = "disabled";
	347	+};
	348	+
344	349	&usb_otg_hs {
345	350	interface-type = <0>;
346	351	usb-phy = <&usb2_phy>;

arch/arm/crypto/chacha-glue.c+11−3

@@ -91,9 +91,17 @@ void chacha_crypt_arch(u32 state, u8 dst, const u8 *src, unsigned int bytes, · 90 unmodified lines
91	91	return;
92	92	}
93	93
94		− kernel_neon_begin();
95		− chacha_doneon(state, dst, src, bytes, nrounds);
96		− kernel_neon_end();
	94	+ do {
	95	+ unsigned int todo = min_t(unsigned int, bytes, SZ_4K);
	96	+
	97	+ kernel_neon_begin();
	98	+ chacha_doneon(state, dst, src, todo, nrounds);
	99	+ kernel_neon_end();
	100	+
	101	+ bytes -= todo;
	102	+ src += todo;
	103	+ dst += todo;
	104	+ } while (bytes);
97	105	}
98	106	EXPORT_SYMBOL(chacha_crypt_arch);
99	107

arch/arm/crypto/nhpoly1305-neon-glue.c+1−1

@@ -30,7 +30,7 @@ static int nhpoly1305_neon_update(struct shash_desc *desc, · 29 unmodified lines
30	30	return crypto_nhpoly1305_update(desc, src, srclen);
31	31
32	32	do {
33		− unsigned int n = min_t(unsigned int, srclen, PAGE_SIZE);
	33	+ unsigned int n = min_t(unsigned int, srclen, SZ_4K);
34	34
35	35	kernel_neon_begin();
36	36	crypto_nhpoly1305_update_helper(desc, src, n, _nh_neon);

arch/arm/crypto/poly1305-glue.c+11−4

@@ -160,13 +160,20 @@ void poly1305_update_arch(struct poly1305_desc_ctx dctx, const u8 src, · 159 unmodified lines
160	160	unsigned int len = round_down(nbytes, POLY1305_BLOCK_SIZE);
161	161
162	162	if (static_branch_likely(&have_neon) && do_neon) {
163		− kernel_neon_begin();
164		− poly1305_blocks_neon(&dctx->h, src, len, 1);
165		− kernel_neon_end();
	163	+ do {
	164	+ unsigned int todo = min_t(unsigned int, len, SZ_4K);
	165	+
	166	+ kernel_neon_begin();
	167	+ poly1305_blocks_neon(&dctx->h, src, todo, 1);
	168	+ kernel_neon_end();
	169	+
	170	+ len -= todo;
	171	+ src += todo;
	172	+ } while (len);
166	173	} else {
167	174	poly1305_blocks_arm(&dctx->h, src, len, 1);
	175	+ src += len;
168	176	}
169		− src += len;
170	177	nbytes %= POLY1305_BLOCK_SIZE;
171	178	}
172	179

arch/arm/include/asm/futex.h+7−2

@@ -165,8 +165,13 @@ arch_futex_atomic_op_inuser(int op, int oparg, int oval, u32 __user uaddr) · 164 unmodified lines
165	165	preempt_enable();
166	166	#endif
167	167
168		− if (!ret)
169		− *oval = oldval;
	168	+ /*
	169	+ * Store unconditionally. If ret != 0 the extra store is the least
	170	+ * of the worries but GCC cannot figure out that __futex_atomic_op()
	171	+ * is either setting ret to -EFAULT or storing the old value in
	172	+ * oldval which results in a uninitialized warning at the call site.
	173	+ */
	174	+ *oval = oldval;
170	175
171	176	return ret;
172	177	}

arch/arm/include/asm/module.h+0−24

@@ -37,30 +37,6 @@ struct mod_arch_specific { · 36 unmodified lines
37	37	struct module;
38	38	u32 get_module_plt(struct module *mod, unsigned long loc, Elf32_Addr val);
39	39
40		−/*
41		− * Add the ARM architecture version to the version magic string
42		− */
43		−#define MODULE_ARCH_VERMAGIC_ARMVSN "ARMv" __stringify(__LINUX_ARM_ARCH__) " "
44		−
45		−/* Add __virt_to_phys patching state as well */
46		−#ifdef CONFIG_ARM_PATCH_PHYS_VIRT
47		−#define MODULE_ARCH_VERMAGIC_P2V "p2v8 "
48		−#else
49		−#define MODULE_ARCH_VERMAGIC_P2V ""
50		−#endif
51		−
52		−/* Add instruction set architecture tag to distinguish ARM/Thumb kernels */
53		−#ifdef CONFIG_THUMB2_KERNEL
54		−#define MODULE_ARCH_VERMAGIC_ARMTHUMB "thumb2 "
55		−#else
56		−#define MODULE_ARCH_VERMAGIC_ARMTHUMB ""
57		−#endif
58		−
59		−#define MODULE_ARCH_VERMAGIC \
60		− MODULE_ARCH_VERMAGIC_ARMVSN \
61		− MODULE_ARCH_VERMAGIC_ARMTHUMB \
62		− MODULE_ARCH_VERMAGIC_P2V
63		−
64	40	#ifdef CONFIG_THUMB2_KERNEL
65	41	#define HAVE_ARCH_KALLSYMS_SYMBOL_VALUE
66	42	static inline unsigned long kallsyms_symbol_value(const Elf_Sym *sym)

arch/arm/include/asm/vermagic.h+31−0

@@ -0,0 +1,31 @@
	1	+/* SPDX-License-Identifier: GPL-2.0 */
	2	+#ifndef _ASM_VERMAGIC_H
	3	+#define _ASM_VERMAGIC_H
	4	+
	5	+#include <linux/stringify.h>
	6	+
	7	+/*
	8	+ * Add the ARM architecture version to the version magic string
	9	+ */
	10	+#define MODULE_ARCH_VERMAGIC_ARMVSN "ARMv" __stringify(__LINUX_ARM_ARCH__) " "
	11	+
	12	+/* Add __virt_to_phys patching state as well */
	13	+#ifdef CONFIG_ARM_PATCH_PHYS_VIRT
	14	+#define MODULE_ARCH_VERMAGIC_P2V "p2v8 "
	15	+#else
	16	+#define MODULE_ARCH_VERMAGIC_P2V ""
	17	+#endif
	18	+
	19	+/* Add instruction set architecture tag to distinguish ARM/Thumb kernels */
	20	+#ifdef CONFIG_THUMB2_KERNEL
	21	+#define MODULE_ARCH_VERMAGIC_ARMTHUMB "thumb2 "
	22	+#else
	23	+#define MODULE_ARCH_VERMAGIC_ARMTHUMB ""
	24	+#endif
	25	+
	26	+#define MODULE_ARCH_VERMAGIC \
	27	+ MODULE_ARCH_VERMAGIC_ARMVSN \
	28	+ MODULE_ARCH_VERMAGIC_ARMTHUMB \
	29	+ MODULE_ARCH_VERMAGIC_P2V
	30	+
	31	+#endif /* _ASM_VERMAGIC_H */

arch/arm/mach-imx/Makefile+2−0

@@ -91,8 +91,10 @@ AFLAGS_suspend-imx6.o :=-Wa,-march=armv7-a · 90 unmodified lines
91	91	obj-$(CONFIG_SOC_IMX6) += suspend-imx6.o
92	92	obj-$(CONFIG_SOC_IMX53) += suspend-imx53.o
93	93	endif
	94	+ifeq ($(CONFIG_ARM_CPU_SUSPEND),y)
94	95	AFLAGS_resume-imx6.o :=-Wa,-march=armv7-a
95	96	obj-$(CONFIG_SOC_IMX6) += resume-imx6.o
	97	+endif
96	98	obj-$(CONFIG_SOC_IMX6) += pm-imx6.o
97	99
98	100	obj-$(CONFIG_SOC_IMX1) += mach-imx1.o

arch/arm64/crypto/chacha-neon-glue.c+11−3

@@ -87,9 +87,17 @@ void chacha_crypt_arch(u32 state, u8 dst, const u8 *src, unsigned int bytes, · 86 unmodified lines
87	87	!crypto_simd_usable())
88	88	return chacha_crypt_generic(state, dst, src, bytes, nrounds);
89	89
90		− kernel_neon_begin();
91		− chacha_doneon(state, dst, src, bytes, nrounds);
92		− kernel_neon_end();
	90	+ do {
	91	+ unsigned int todo = min_t(unsigned int, bytes, SZ_4K);
	92	+
	93	+ kernel_neon_begin();
	94	+ chacha_doneon(state, dst, src, todo, nrounds);
	95	+ kernel_neon_end();
	96	+
	97	+ bytes -= todo;
	98	+ src += todo;
	99	+ dst += todo;
	100	+ } while (bytes);
93	101	}
94	102	EXPORT_SYMBOL(chacha_crypt_arch);
95	103

arch/arm64/crypto/nhpoly1305-neon-glue.c+1−1

@@ -30,7 +30,7 @@ static int nhpoly1305_neon_update(struct shash_desc *desc, · 29 unmodified lines
30	30	return crypto_nhpoly1305_update(desc, src, srclen);
31	31
32	32	do {
33		− unsigned int n = min_t(unsigned int, srclen, PAGE_SIZE);
	33	+ unsigned int n = min_t(unsigned int, srclen, SZ_4K);
34	34
35	35	kernel_neon_begin();
36	36	crypto_nhpoly1305_update_helper(desc, src, n, _nh_neon);

arch/arm64/crypto/poly1305-glue.c+11−4

@@ -143,13 +143,20 @@ void poly1305_update_arch(struct poly1305_desc_ctx dctx, const u8 src, · 142 unmodified lines
143	143	unsigned int len = round_down(nbytes, POLY1305_BLOCK_SIZE);
144	144
145	145	if (static_branch_likely(&have_neon) && crypto_simd_usable()) {
146		− kernel_neon_begin();
147		− poly1305_blocks_neon(&dctx->h, src, len, 1);
148		− kernel_neon_end();
	146	+ do {
	147	+ unsigned int todo = min_t(unsigned int, len, SZ_4K);
	148	+
	149	+ kernel_neon_begin();
	150	+ poly1305_blocks_neon(&dctx->h, src, todo, 1);
	151	+ kernel_neon_end();
	152	+
	153	+ len -= todo;
	154	+ src += todo;
	155	+ } while (len);
149	156	} else {
150	157	poly1305_blocks(&dctx->h, src, len, 1);
	158	+ src += len;
151	159	}
152		− src += len;
153	160	nbytes %= POLY1305_BLOCK_SIZE;
154	161	}
155	162

arch/arm64/include/asm/module.h+0−2

@@ -7,8 +7,6 @@ · 6 unmodified lines
7	7
8	8	#include <asm-generic/module.h>
9	9
10		−#define MODULE_ARCH_VERMAGIC "aarch64"
11		−
12	10	#ifdef CONFIG_ARM64_MODULE_PLTS
13	11	struct mod_plt_sec {
14	12	int plt_shndx;

arch/arm64/include/asm/pointer_auth.h+6−3

@@ -47,7 +47,7 @@ static inline void ptrauth_keys_init_user(struct ptrauth_keys_user *keys) · 46 unmodified lines
47	47	get_random_bytes(&keys->apga, sizeof(keys->apga));
48	48	}
49	49
50		−#define __ptrauth_key_install(k, v) \
	50	+#define __ptrauth_key_install_nosync(k, v) \
51	51	do { \
52	52	struct ptrauth_key __pki_v = (v); \
53	53	write_sysreg_s(__pki_v.lo, SYS_ ## k ## KEYLO_EL1); \
@@ -62,8 +62,11 @@ static __always_inline void ptrauth_keys_init_kernel(struct ptrauth_keys_kernel · 8 unmodified lines
62	62
63	63	static __always_inline void ptrauth_keys_switch_kernel(struct ptrauth_keys_kernel *keys)
64	64	{
65		− if (system_supports_address_auth())
66		− __ptrauth_key_install(APIA, keys->apia);
	65	+ if (!system_supports_address_auth())
	66	+ return;
	67	+
	68	+ __ptrauth_key_install_nosync(APIA, keys->apia);
	69	+ isb();
67	70	}
68	71
69	72	extern int ptrauth_prctl_reset_keys(struct task_struct *tsk, unsigned long arg);

arch/arm64/include/asm/vermagic.h+10−0

@@ -0,0 +1,10 @@
	1	+/* SPDX-License-Identifier: GPL-2.0-only */
	2	+/*
	3	+ * Copyright (C) 2012 ARM Ltd.
	4	+ */
	5	+#ifndef _ASM_VERMAGIC_H
	6	+#define _ASM_VERMAGIC_H
	7	+
	8	+#define MODULE_ARCH_VERMAGIC "aarch64"
	9	+
	10	+#endif /* _ASM_VERMAGIC_H */

arch/arm64/kernel/vdso/Makefile+1−1

@@ -32,7 +32,7 @@ UBSAN_SANITIZE := n · 31 unmodified lines
32	32	OBJECT_FILES_NON_STANDARD := y
33	33	KCOV_INSTRUMENT := n
34	34
35		−CFLAGS_vgettimeofday.o = -O2 -mcmodel=tiny
	35	+CFLAGS_vgettimeofday.o = -O2 -mcmodel=tiny -fasynchronous-unwind-tables
36	36
37	37	ifneq ($(c-gettimeofday-y),)
38	38	CFLAGS_vgettimeofday.o += -include $(c-gettimeofday-y)

arch/arm64/kvm/guest.c+7−0

@@ -200,6 +200,13 @@ static int set_core_reg(struct kvm_vcpu vcpu, const struct kvm_one_reg reg) · 199 unmodified lines
200	200	}
201	201
202	202	memcpy((u32 *)regs + off, valp, KVM_REG_SIZE(reg->id));
	203	+
	204	+ if (*vcpu_cpsr(vcpu) & PSR_MODE32_BIT) {
	205	+ int i;
	206	+
	207	+ for (i = 0; i < 16; i++)
	208	+ vcpu_reg32(vcpu, i) = (u32)vcpu_reg32(vcpu, i);
	209	+ }
203	210	out:
204	211	return err;
205	212	}

arch/arm64/kvm/hyp/entry.S+23−0

@@ -18,6 +18,7 @@ · 17 unmodified lines
18	18
19	19	#define CPU_GP_REG_OFFSET(x) (CPU_GP_REGS + x)
20	20	#define CPU_XREG_OFFSET(x) CPU_GP_REG_OFFSET(CPU_USER_PT_REGS + 8*x)
	21	+#define CPU_SP_EL0_OFFSET (CPU_XREG_OFFSET(30) + 8)
21	22
22	23	.text
23	24	.pushsection .hyp.text, "ax"
@@ -47,6 +48,16 @@ · 23 unmodified lines
47	48	ldp x29, lr, [\ctxt, #CPU_XREG_OFFSET(29)]
48	49	.endm
49	50
	51	+.macro save_sp_el0 ctxt, tmp
	52	+ mrs \tmp, sp_el0
	53	+ str \tmp, [\ctxt, #CPU_SP_EL0_OFFSET]
	54	+.endm
	55	+
	56	+.macro restore_sp_el0 ctxt, tmp
	57	+ ldr \tmp, [\ctxt, #CPU_SP_EL0_OFFSET]
	58	+ msr sp_el0, \tmp
	59	+.endm
	60	+
50	61	/*
51	62	* u64 __guest_enter(struct kvm_vcpu *vcpu,
52	63	* struct kvm_cpu_context *host_ctxt);
@@ -60,6 +71,9 @@ SYM_FUNC_START(__guest_enter) · 7 unmodified lines
60	71	// Store the host regs
61	72	save_callee_saved_regs x1
62	73
	74	+ // Save the host's sp_el0
	75	+ save_sp_el0 x1, x2
	76	+
63	77	// Now the host state is stored if we have a pending RAS SError it must
64	78	// affect the host. If any asynchronous exception is pending we defer
65	79	// the guest entry. The DSB isn't necessary before v8.2 as any SError
@@ -83,6 +97,9 @@ alternative_else_nop_endif · 17 unmodified lines
83	97	// when this feature is enabled for kernel code.
84	98	ptrauth_switch_to_guest x29, x0, x1, x2
85	99
	100	+ // Restore the guest's sp_el0
	101	+ restore_sp_el0 x29, x0
	102	+
86	103	// Restore guest regs x0-x17
87	104	ldp x0, x1, [x29, #CPU_XREG_OFFSET(0)]
88	105	ldp x2, x3, [x29, #CPU_XREG_OFFSET(2)]
@@ -130,6 +147,9 @@ SYM_INNER_LABEL(__guest_exit, SYM_L_GLOBAL) · 41 unmodified lines
130	147	// Store the guest regs x18-x29, lr
131	148	save_callee_saved_regs x1
132	149
	150	+ // Store the guest's sp_el0
	151	+ save_sp_el0 x1, x2
	152	+
133	153	get_host_ctxt x2, x3
134	154
135	155	// Macro ptrauth_switch_to_guest format:
@@ -139,6 +159,9 @@ SYM_INNER_LABEL(__guest_exit, SYM_L_GLOBAL) · 3 unmodified lines
139	159	// when this feature is enabled for kernel code.
140	160	ptrauth_switch_to_host x1, x2, x3, x4, x5
141	161
	162	+ // Restore the hosts's sp_el0
	163	+ restore_sp_el0 x2, x3
	164	+
142	165	// Now restore the host regs
143	166	restore_callee_saved_regs x2
144	167

arch/arm64/kvm/hyp/hyp-entry.S+0−1

@@ -198,7 +198,6 @@ SYM_CODE_END(__hyp_panic) · 197 unmodified lines
198	198	.macro invalid_vector label, target = __hyp_panic
199	199	.align 2
200	200	SYM_CODE_START(\label)
201		−\label:
202	201	b \target
203	202	SYM_CODE_END(\label)
204	203	.endm

arch/arm64/kvm/hyp/sysreg-sr.c+3−14

@@ -15,8 +15,9 @@ · 14 unmodified lines
15	15	/*
16	16	* Non-VHE: Both host and guest must save everything.
17	17	*
18		− * VHE: Host and guest must save mdscr_el1 and sp_el0 (and the PC and pstate,
19		− * which are handled as part of the el2 return state) on every switch.
	18	+ * VHE: Host and guest must save mdscr_el1 and sp_el0 (and the PC and
	19	+ * pstate, which are handled as part of the el2 return state) on every
	20	+ * switch (sp_el0 is being dealt with in the assembly code).
20	21	* tpidr_el0 and tpidrro_el0 only need to be switched when going
21	22	* to host userspace or a different VCPU. EL1 registers only need to be
22	23	* switched when potentially going to run a different VCPU. The latter two
@@ -26,12 +27,6 @@ · 3 unmodified lines
26	27	static void __hyp_text __sysreg_save_common_state(struct kvm_cpu_context *ctxt)
27	28	{
28	29	ctxt->sys_regs[MDSCR_EL1] = read_sysreg(mdscr_el1);
29		−
30		− /*
31		− * The host arm64 Linux uses sp_el0 to point to 'current' and it must
32		− * therefore be saved/restored on every entry/exit to/from the guest.
33		− */
34		− ctxt->gp_regs.regs.sp = read_sysreg(sp_el0);
35	30	}
36	31
37	32	static void __hyp_text __sysreg_save_user_state(struct kvm_cpu_context *ctxt)
@@ -99,12 +94,6 @@ NOKPROBE_SYMBOL(sysreg_save_guest_state_vhe); · 61 unmodified lines
99	94	static void __hyp_text __sysreg_restore_common_state(struct kvm_cpu_context *ctxt)
100	95	{
101	96	write_sysreg(ctxt->sys_regs[MDSCR_EL1], mdscr_el1);
102		−
103		− /*
104		− * The host arm64 Linux uses sp_el0 to point to 'current' and it must
105		− * therefore be saved/restored on every entry/exit to/from the guest.
106		− */
107		− write_sysreg(ctxt->gp_regs.regs.sp, sp_el0);
108	97	}
109	98
110	99	static void __hyp_text __sysreg_restore_user_state(struct kvm_cpu_context *ctxt)

arch/arm64/mm/hugetlbpage.c+2−0

@@ -230,6 +230,8 @@ pte_t huge_pte_alloc(struct mm_struct mm, · 229 unmodified lines
230	230	ptep = (pte_t *)pudp;
231	231	} else if (sz == (CONT_PTE_SIZE)) {
232	232	pmdp = pmd_alloc(mm, pudp, addr);
	233	+ if (!pmdp)
	234	+ return NULL;
233	235
234	236	WARN_ON(addr & (sz - 1));
235	237	/*

arch/h8300/kernel/.gitignore+2−0

@@ -0,0 +1,2 @@
	1	+# SPDX-License-Identifier: GPL-2.0-only
	2	+vmlinux.lds

arch/hexagon/include/asm/vermagic.hrenamed from arch/hexagon/include/asm/module.h+4−4

@@ -3,11 +3,11 @@ · 2 unmodified lines
3	3	* Copyright (c) 2010-2011, The Linux Foundation. All rights reserved.
4	4	*/
5	5
6		−#ifndef _ASM_MODULE_H
7		−#define _ASM_MODULE_H
	6	+#ifndef _ASM_VERMAGIC_H
	7	+#define _ASM_VERMAGIC_H
8	8
9		−#include <asm-generic/module.h>
	9	+#include <linux/stringify.h>
10	10
11	11	#define MODULE_ARCH_VERMAGIC __stringify(PROCESSOR_MODEL_NAME) " "
12	12
13		−#endif
	13	+#endif /* _ASM_VERMAGIC_H */

arch/ia64/include/asm/module.h+0−4

@@ -26,10 +26,6 @@ struct mod_arch_specific { · 25 unmodified lines
26	26	unsigned int next_got_entry; /* index of next available got entry */
27	27	};
28	28
29		−#define MODULE_PROC_FAMILY "ia64"
30		−#define MODULE_ARCH_VERMAGIC MODULE_PROC_FAMILY \
31		− "gcc-" __stringify(__GNUC__) "." __stringify(__GNUC_MINOR__)
32		−
33	29	#define ARCH_SHF_SMALL SHF_IA_64_SHORT
34	30
35	31	#endif /* _ASM_IA64_MODULE_H */

arch/ia64/include/asm/vermagic.h+15−0

@@ -0,0 +1,15 @@
	1	+/* SPDX-License-Identifier: GPL-2.0 */
	2	+/*
	3	+ * Copyright (C) 2003 Hewlett-Packard Co
	4	+ * David Mosberger-Tang <davidm@hpl.hp.com>
	5	+ */
	6	+
	7	+#ifndef _ASM_VERMAGIC_H
	8	+#define _ASM_VERMAGIC_H
	9	+
	10	+#include <linux/stringify.h>
	11	+
	12	+#define MODULE_ARCH_VERMAGIC "ia64" \
	13	+ "gcc-" __stringify(__GNUC__) "." __stringify(__GNUC_MINOR__)
	14	+
	15	+#endif /* _ASM_VERMAGIC_H */

arch/mips/include/asm/module.h+0−61

@@ -83,65 +83,4 @@ search_module_dbetables(unsigned long addr) · 82 unmodified lines
83	83	}
84	84	#endif
85	85
86		−#ifdef CONFIG_CPU_BMIPS
87		−#define MODULE_PROC_FAMILY "BMIPS "
88		−#elif defined CONFIG_CPU_MIPS32_R1
89		−#define MODULE_PROC_FAMILY "MIPS32_R1 "
90		−#elif defined CONFIG_CPU_MIPS32_R2
91		−#define MODULE_PROC_FAMILY "MIPS32_R2 "
92		−#elif defined CONFIG_CPU_MIPS32_R6
93		−#define MODULE_PROC_FAMILY "MIPS32_R6 "
94		−#elif defined CONFIG_CPU_MIPS64_R1
95		−#define MODULE_PROC_FAMILY "MIPS64_R1 "
96		−#elif defined CONFIG_CPU_MIPS64_R2
97		−#define MODULE_PROC_FAMILY "MIPS64_R2 "
98		−#elif defined CONFIG_CPU_MIPS64_R6
99		−#define MODULE_PROC_FAMILY "MIPS64_R6 "
100		−#elif defined CONFIG_CPU_R3000
101		−#define MODULE_PROC_FAMILY "R3000 "
102		−#elif defined CONFIG_CPU_TX39XX
103		−#define MODULE_PROC_FAMILY "TX39XX "
104		−#elif defined CONFIG_CPU_VR41XX
105		−#define MODULE_PROC_FAMILY "VR41XX "
106		−#elif defined CONFIG_CPU_R4X00
107		−#define MODULE_PROC_FAMILY "R4X00 "
108		−#elif defined CONFIG_CPU_TX49XX
109		−#define MODULE_PROC_FAMILY "TX49XX "
110		−#elif defined CONFIG_CPU_R5000
111		−#define MODULE_PROC_FAMILY "R5000 "
112		−#elif defined CONFIG_CPU_R5500
113		−#define MODULE_PROC_FAMILY "R5500 "
114		−#elif defined CONFIG_CPU_NEVADA
115		−#define MODULE_PROC_FAMILY "NEVADA "
116		−#elif defined CONFIG_CPU_R10000
117		−#define MODULE_PROC_FAMILY "R10000 "
118		−#elif defined CONFIG_CPU_RM7000
119		−#define MODULE_PROC_FAMILY "RM7000 "
120		−#elif defined CONFIG_CPU_SB1
121		−#define MODULE_PROC_FAMILY "SB1 "
122		−#elif defined CONFIG_CPU_LOONGSON32
123		−#define MODULE_PROC_FAMILY "LOONGSON32 "
124		−#elif defined CONFIG_CPU_LOONGSON2EF
125		−#define MODULE_PROC_FAMILY "LOONGSON2EF "
126		−#elif defined CONFIG_CPU_LOONGSON64
127		−#define MODULE_PROC_FAMILY "LOONGSON64 "
128		−#elif defined CONFIG_CPU_CAVIUM_OCTEON
129		−#define MODULE_PROC_FAMILY "OCTEON "
130		−#elif defined CONFIG_CPU_XLR
131		−#define MODULE_PROC_FAMILY "XLR "
132		−#elif defined CONFIG_CPU_XLP
133		−#define MODULE_PROC_FAMILY "XLP "
134		−#else
135		−#error MODULE_PROC_FAMILY undefined for your processor configuration
136		−#endif
137		−
138		−#ifdef CONFIG_32BIT
139		−#define MODULE_KERNEL_TYPE "32BIT "
140		−#elif defined CONFIG_64BIT
141		−#define MODULE_KERNEL_TYPE "64BIT "
142		−#endif
143		−
144		−#define MODULE_ARCH_VERMAGIC \
145		− MODULE_PROC_FAMILY MODULE_KERNEL_TYPE
146		−
147	86	#endif /* _ASM_MODULE_H */

arch/mips/include/asm/vermagic.h+66−0

@@ -0,0 +1,66 @@
	1	+/* SPDX-License-Identifier: GPL-2.0 */
	2	+#ifndef _ASM_VERMAGIC_H
	3	+#define _ASM_VERMAGIC_H
	4	+
	5	+#ifdef CONFIG_CPU_BMIPS
	6	+#define MODULE_PROC_FAMILY "BMIPS "
	7	+#elif defined CONFIG_CPU_MIPS32_R1
	8	+#define MODULE_PROC_FAMILY "MIPS32_R1 "
	9	+#elif defined CONFIG_CPU_MIPS32_R2
	10	+#define MODULE_PROC_FAMILY "MIPS32_R2 "
	11	+#elif defined CONFIG_CPU_MIPS32_R6
	12	+#define MODULE_PROC_FAMILY "MIPS32_R6 "
	13	+#elif defined CONFIG_CPU_MIPS64_R1
	14	+#define MODULE_PROC_FAMILY "MIPS64_R1 "
	15	+#elif defined CONFIG_CPU_MIPS64_R2
	16	+#define MODULE_PROC_FAMILY "MIPS64_R2 "
	17	+#elif defined CONFIG_CPU_MIPS64_R6
	18	+#define MODULE_PROC_FAMILY "MIPS64_R6 "
	19	+#elif defined CONFIG_CPU_R3000
	20	+#define MODULE_PROC_FAMILY "R3000 "
	21	+#elif defined CONFIG_CPU_TX39XX
	22	+#define MODULE_PROC_FAMILY "TX39XX "
	23	+#elif defined CONFIG_CPU_VR41XX
	24	+#define MODULE_PROC_FAMILY "VR41XX "
	25	+#elif defined CONFIG_CPU_R4X00
	26	+#define MODULE_PROC_FAMILY "R4X00 "
	27	+#elif defined CONFIG_CPU_TX49XX
	28	+#define MODULE_PROC_FAMILY "TX49XX "
	29	+#elif defined CONFIG_CPU_R5000
	30	+#define MODULE_PROC_FAMILY "R5000 "
	31	+#elif defined CONFIG_CPU_R5500
	32	+#define MODULE_PROC_FAMILY "R5500 "
	33	+#elif defined CONFIG_CPU_NEVADA
	34	+#define MODULE_PROC_FAMILY "NEVADA "
	35	+#elif defined CONFIG_CPU_R10000
	36	+#define MODULE_PROC_FAMILY "R10000 "
	37	+#elif defined CONFIG_CPU_RM7000
	38	+#define MODULE_PROC_FAMILY "RM7000 "
	39	+#elif defined CONFIG_CPU_SB1
	40	+#define MODULE_PROC_FAMILY "SB1 "
	41	+#elif defined CONFIG_CPU_LOONGSON32
	42	+#define MODULE_PROC_FAMILY "LOONGSON32 "
	43	+#elif defined CONFIG_CPU_LOONGSON2EF
	44	+#define MODULE_PROC_FAMILY "LOONGSON2EF "
	45	+#elif defined CONFIG_CPU_LOONGSON64
	46	+#define MODULE_PROC_FAMILY "LOONGSON64 "
	47	+#elif defined CONFIG_CPU_CAVIUM_OCTEON
	48	+#define MODULE_PROC_FAMILY "OCTEON "
	49	+#elif defined CONFIG_CPU_XLR
	50	+#define MODULE_PROC_FAMILY "XLR "
	51	+#elif defined CONFIG_CPU_XLP
	52	+#define MODULE_PROC_FAMILY "XLP "
	53	+#else
	54	+#error MODULE_PROC_FAMILY undefined for your processor configuration
	55	+#endif
	56	+
	57	+#ifdef CONFIG_32BIT
	58	+#define MODULE_KERNEL_TYPE "32BIT "
	59	+#elif defined CONFIG_64BIT
	60	+#define MODULE_KERNEL_TYPE "64BIT "
	61	+#endif
	62	+
	63	+#define MODULE_ARCH_VERMAGIC \
	64	+ MODULE_PROC_FAMILY MODULE_KERNEL_TYPE
	65	+
	66	+#endif /* _ASM_VERMAGIC_H */

arch/nds32/include/asm/vermagic.hrenamed from arch/nds32/include/asm/module.h+3−5

@@ -1,11 +1,9 @@
1	1	/* SPDX-License-Identifier: GPL-2.0 */
2	2	// Copyright (C) 2005-2017 Andes Technology Corporation
3	3
4		−#ifndef _ASM_NDS32_MODULE_H
5		−#define _ASM_NDS32_MODULE_H
6		−
7		−#include <asm-generic/module.h>
	4	+#ifndef _ASM_VERMAGIC_H
	5	+#define _ASM_VERMAGIC_H
8	6
9	7	#define MODULE_ARCH_VERMAGIC "NDS32v3"
10	8
11		−#endif /* _ASM_NDS32_MODULE_H */
	9	+#endif /* _ASM_VERMAGIC_H */

arch/powerpc/include/asm/module.h+0−18

@@ -3,28 +3,10 @@ · 2 unmodified lines
3	3	#define _ASM_POWERPC_MODULE_H
4	4	#ifdef __KERNEL__
5	5
6		−/*
7		− */
8		−
9	6	#include <linux/list.h>
10	7	#include <asm/bug.h>
11	8	#include <asm-generic/module.h>
12	9
13		−
14		−#ifdef CONFIG_MPROFILE_KERNEL
15		−#define MODULE_ARCH_VERMAGIC_FTRACE "mprofile-kernel "
16		−#else
17		−#define MODULE_ARCH_VERMAGIC_FTRACE ""
18		−#endif
19		−
20		−#ifdef CONFIG_RELOCATABLE
21		−#define MODULE_ARCH_VERMAGIC_RELOCATABLE "relocatable "
22		−#else
23		−#define MODULE_ARCH_VERMAGIC_RELOCATABLE ""
24		−#endif
25		−
26		−#define MODULE_ARCH_VERMAGIC MODULE_ARCH_VERMAGIC_FTRACE MODULE_ARCH_VERMAGIC_RELOCATABLE
27		−
28	10	#ifndef __powerpc64__
29	11	/*
30	12	* Thanks to Paul M for explaining this.

arch/powerpc/include/asm/vermagic.h+20−0

@@ -0,0 +1,20 @@
	1	+/* SPDX-License-Identifier: GPL-2.0 */
	2	+#ifndef _ASM_VERMAGIC_H
	3	+#define _ASM_VERMAGIC_H
	4	+
	5	+#ifdef CONFIG_MPROFILE_KERNEL
	6	+#define MODULE_ARCH_VERMAGIC_FTRACE "mprofile-kernel "
	7	+#else
	8	+#define MODULE_ARCH_VERMAGIC_FTRACE ""
	9	+#endif
	10	+
	11	+#ifdef CONFIG_RELOCATABLE
	12	+#define MODULE_ARCH_VERMAGIC_RELOCATABLE "relocatable "
	13	+#else
	14	+#define MODULE_ARCH_VERMAGIC_RELOCATABLE ""
	15	+#endif
	16	+
	17	+#define MODULE_ARCH_VERMAGIC \
	18	+ MODULE_ARCH_VERMAGIC_FTRACE MODULE_ARCH_VERMAGIC_RELOCATABLE
	19	+
	20	+#endif /* _ASM_VERMAGIC_H */

arch/powerpc/kernel/entry_32.S+1−1

@@ -732,7 +732,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_SPE) · 731 unmodified lines
732	732	stw r10,_CCR(r1)
733	733	stw r1,KSP(r3) /* Set old stack pointer */
734	734
735		− kuap_check r2, r4
	735	+ kuap_check r2, r0
736	736	#ifdef CONFIG_SMP
737	737	/* We need a sync somewhere here to make sure that if the
738	738	* previous task gets rescheduled on another CPU, it sees all

arch/powerpc/kernel/setup_64.c+2−0

@@ -534,6 +534,8 @@ static bool __init parse_cache_info(struct device_node *np, · 533 unmodified lines
534	534	lsizep = of_get_property(np, propnames[3], NULL);
535	535	if (bsizep == NULL)
536	536	bsizep = lsizep;
	537	+ if (lsizep == NULL)
	538	+ lsizep = bsizep;
537	539	if (lsizep != NULL)
538	540	lsize = be32_to_cpu(*lsizep);
539	541	if (bsizep != NULL)

arch/powerpc/kernel/vmlinux.lds.S+1−0

@@ -90,6 +90,7 @@ SECTIONS · 89 unmodified lines
90	90	#ifdef CONFIG_PPC64
91	91	*(.tramp.ftrace.text);
92	92	#endif
	93	+ NOINSTR_TEXT
93	94	SCHED_TEXT
94	95	CPUIDLE_TEXT
95	96	LOCK_TEXT

arch/powerpc/kvm/book3s_64_mmu_hv.c+5−4

@@ -604,18 +604,19 @@ int kvmppc_book3s_hv_page_fault(struct kvm_run run, struct kvm_vcpu vcpu, · 603 unmodified lines
604	604	*/
605	605	local_irq_disable();
606	606	ptep = __find_linux_pte(vcpu->arch.pgdir, hva, NULL, &shift);
	607	+ pte = __pte(0);
	608	+ if (ptep)
	609	+ pte = *ptep;
	610	+ local_irq_enable();
607	611	/*
608	612	* If the PTE disappeared temporarily due to a THP
609	613	* collapse, just return and let the guest try again.
610	614	*/
611		− if (!ptep) {
612		− local_irq_enable();
	615	+ if (!pte_present(pte)) {
613	616	if (page)
614	617	put_page(page);
615	618	return RESUME_GUEST;
616	619	}
617		− pte = *ptep;
618		− local_irq_enable();
619	620	hpa = pte_pfn(pte) << PAGE_SHIFT;
620	621	pte_size = PAGE_SIZE;
621	622	if (shift)

arch/powerpc/kvm/book3s_64_mmu_radix.c+5−4

@@ -815,18 +815,19 @@ int kvmppc_book3s_instantiate_page(struct kvm_vcpu *vcpu, · 814 unmodified lines
815	815	*/
816	816	local_irq_disable();
817	817	ptep = __find_linux_pte(vcpu->arch.pgdir, hva, NULL, &shift);
	818	+ pte = __pte(0);
	819	+ if (ptep)
	820	+ pte = *ptep;
	821	+ local_irq_enable();
818	822	/*
819	823	* If the PTE disappeared temporarily due to a THP
820	824	* collapse, just return and let the guest try again.
821	825	*/
822		− if (!ptep) {
823		− local_irq_enable();
	826	+ if (!pte_present(pte)) {
824	827	if (page)
825	828	put_page(page);
826	829	return RESUME_GUEST;
827	830	}
828		− pte = *ptep;
829		− local_irq_enable();
830	831
831	832	/* If we're logging dirty pages, always map single pages */
832	833	large_enable = !(memslot->flags & KVM_MEM_LOG_DIRTY_PAGES);

arch/powerpc/kvm/powerpc.c+1−0

@@ -521,6 +521,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) · 520 unmodified lines
521	521	case KVM_CAP_IOEVENTFD:
522	522	case KVM_CAP_DEVICE_CTRL:
523	523	case KVM_CAP_IMMEDIATE_EXIT:
	524	+ case KVM_CAP_SET_GUEST_DEBUG:
524	525	r = 1;
525	526	break;
526	527	case KVM_CAP_PPC_GUEST_DEBUG_SSTEP:

arch/powerpc/mm/nohash/8xx.c+3−0

@@ -185,6 +185,7 @@ void mmu_mark_initmem_nx(void) · 184 unmodified lines
185	185	mmu_mapin_ram_chunk(etext8, einittext8, PAGE_KERNEL);
186	186	}
187	187	}
	188	+ _tlbil_all();
188	189	}
189	190
190	191	#ifdef CONFIG_STRICT_KERNEL_RWX
@@ -199,6 +200,8 @@ void mmu_mark_rodata_ro(void) · 8 unmodified lines
199	200	~(LARGE_PAGE_SIZE_8M - 1)));
200	201	mmu_patch_addis(&patch__dtlbmiss_romem_top, -__pa(_sinittext));
201	202
	203	+ _tlbil_all();
	204	+
202	205	/* Update page tables for PTDUMP and BDI */
203	206	mmu_mapin_ram_chunk(0, sinittext, __pgprot(0));
204	207	mmu_mapin_ram_chunk(0, etext, PAGE_KERNEL_ROX);

arch/powerpc/platforms/Kconfig.cputype+1−1

@@ -397,7 +397,7 @@ config PPC_KUAP · 396 unmodified lines
397	397
398	398	config PPC_KUAP_DEBUG
399	399	bool "Extra debugging for Kernel Userspace Access Protection"
400		− depends on PPC_KUAP && (PPC_RADIX_MMU \|\| PPC_32)
	400	+ depends on PPC_KUAP && (PPC_RADIX_MMU \|\| PPC32)
401	401	help
402	402	Add extra debugging for Kernel Userspace Access Protection (KUAP)
403	403	If you're unsure, say N.

arch/riscv/Kconfig+1−1

@@ -60,7 +60,7 @@ config RISCV · 59 unmodified lines
60	60	select ARCH_HAS_GIGANTIC_PAGE
61	61	select ARCH_HAS_SET_DIRECT_MAP
62	62	select ARCH_HAS_SET_MEMORY
63		− select ARCH_HAS_STRICT_KERNEL_RWX
	63	+ select ARCH_HAS_STRICT_KERNEL_RWX if MMU
64	64	select ARCH_WANT_HUGE_PMD_SHARE if 64BIT
65	65	select SPARSEMEM_STATIC if 32BIT
66	66	select ARCH_WANT_DEFAULT_TOPDOWN_MMAP_LAYOUT if MMU

arch/riscv/include/asm/csr.h+0−3

@@ -51,13 +51,10 @@ · 50 unmodified lines
51	51	#define CAUSE_IRQ_FLAG (_AC(1, UL) << (__riscv_xlen - 1))
52	52
53	53	/* Interrupt causes (minus the high bit) */
54		−#define IRQ_U_SOFT 0
55	54	#define IRQ_S_SOFT 1
56	55	#define IRQ_M_SOFT 3
57		−#define IRQ_U_TIMER 4
58	56	#define IRQ_S_TIMER 5
59	57	#define IRQ_M_TIMER 7
60		−#define IRQ_U_EXT 8
61	58	#define IRQ_S_EXT 9
62	59	#define IRQ_M_EXT 11
63	60

arch/riscv/include/asm/hwcap.h+22−0

@@ -8,6 +8,7 @@ · 7 unmodified lines
8	8	#ifndef _ASM_RISCV_HWCAP_H
9	9	#define _ASM_RISCV_HWCAP_H
10	10
	11	+#include <linux/bits.h>
11	12	#include <uapi/asm/hwcap.h>
12	13
13	14	#ifndef __ASSEMBLY__
@@ -22,6 +23,27 @@ enum { · 8 unmodified lines
22	23	};
23	24
24	25	extern unsigned long elf_hwcap;
	26	+
	27	+#define RISCV_ISA_EXT_a ('a' - 'a')
	28	+#define RISCV_ISA_EXT_c ('c' - 'a')
	29	+#define RISCV_ISA_EXT_d ('d' - 'a')
	30	+#define RISCV_ISA_EXT_f ('f' - 'a')
	31	+#define RISCV_ISA_EXT_h ('h' - 'a')
	32	+#define RISCV_ISA_EXT_i ('i' - 'a')
	33	+#define RISCV_ISA_EXT_m ('m' - 'a')
	34	+#define RISCV_ISA_EXT_s ('s' - 'a')
	35	+#define RISCV_ISA_EXT_u ('u' - 'a')
	36	+
	37	+#define RISCV_ISA_EXT_MAX 64
	38	+
	39	+unsigned long riscv_isa_extension_base(const unsigned long *isa_bitmap);
	40	+
	41	+#define riscv_isa_extension_mask(ext) BIT_MASK(RISCV_ISA_EXT_##ext)
	42	+
	43	+bool __riscv_isa_extension_available(const unsigned long *isa_bitmap, int bit);
	44	+#define riscv_isa_extension_available(isa_bitmap, ext) \
	45	+ __riscv_isa_extension_available(isa_bitmap, RISCV_ISA_EXT_##ext)
	46	+
25	47	#endif
26	48
27	49	#endif /* _ASM_RISCV_HWCAP_H */

arch/riscv/include/asm/module.h+0−2

@@ -6,8 +6,6 @@ · 5 unmodified lines
6	6
7	7	#include <asm-generic/module.h>
8	8
9		−#define MODULE_ARCH_VERMAGIC "riscv"
10		−
11	9	struct module;
12	10	unsigned long module_emit_got_entry(struct module *mod, unsigned long val);
13	11	unsigned long module_emit_plt_entry(struct module *mod, unsigned long val);

arch/riscv/include/asm/set_memory.h+0−8

@@ -22,14 +22,6 @@ static inline int set_memory_x(unsigned long addr, int numpages) { return 0; } · 21 unmodified lines
22	22	static inline int set_memory_nx(unsigned long addr, int numpages) { return 0; }
23	23	#endif
24	24
25		−#ifdef CONFIG_STRICT_KERNEL_RWX
26		−void set_kernel_text_ro(void);
27		−void set_kernel_text_rw(void);
28		−#else
29		−static inline void set_kernel_text_ro(void) { }
30		−static inline void set_kernel_text_rw(void) { }
31		−#endif
32		−
33	25	int set_direct_map_invalid_noflush(struct page *page);
34	26	int set_direct_map_default_noflush(struct page *page);
35	27

arch/riscv/include/asm/vermagic.h+9−0

@@ -0,0 +1,9 @@
	1	+/* SPDX-License-Identifier: GPL-2.0 */
	2	+/* Copyright (C) 2017 Andes Technology Corporation */
	3	+
	4	+#ifndef _ASM_VERMAGIC_H
	5	+#define _ASM_VERMAGIC_H
	6	+
	7	+#define MODULE_ARCH_VERMAGIC "riscv"
	8	+
	9	+#endif /* _ASM_VERMAGIC_H */

arch/riscv/kernel/cpu_ops.c+2−2

@@ -15,8 +15,8 @@ · 14 unmodified lines
15	15
16	16	const struct cpu_operations *cpu_ops[NR_CPUS] __ro_after_init;
17	17
18		−void *__cpu_up_stack_pointer[NR_CPUS];
19		−void *__cpu_up_task_pointer[NR_CPUS];
	18	+void *__cpu_up_stack_pointer[NR_CPUS] __section(.data);
	19	+void *__cpu_up_task_pointer[NR_CPUS] __section(.data);
20	20
21	21	extern const struct cpu_operations cpu_ops_sbi;
22	22	extern const struct cpu_operations cpu_ops_spinwait;

arch/riscv/kernel/cpufeature.c+80−3

@@ -6,6 +6,7 @@ · 5 unmodified lines
6	6	* Copyright (C) 2017 SiFive
7	7	*/
8	8
	9	+#include <linux/bitmap.h>
9	10	#include <linux/of.h>
10	11	#include <asm/processor.h>
11	12	#include <asm/hwcap.h>
@@ -13,15 +14,57 @@ · 1 unmodified line
13	14	#include <asm/switch_to.h>
14	15
15	16	unsigned long elf_hwcap __read_mostly;
	17	+
	18	+/* Host ISA bitmap */
	19	+static DECLARE_BITMAP(riscv_isa, RISCV_ISA_EXT_MAX) __read_mostly;
	20	+
16	21	#ifdef CONFIG_FPU
17	22	bool has_fpu __read_mostly;
18	23	#endif
19	24
	25	+/**
	26	+ * riscv_isa_extension_base() - Get base extension word
	27	+ *
	28	+ * @isa_bitmap: ISA bitmap to use
	29	+ * Return: base extension word as unsigned long value
	30	+ *
	31	+ * NOTE: If isa_bitmap is NULL then Host ISA bitmap will be used.
	32	+ */
	33	+unsigned long riscv_isa_extension_base(const unsigned long *isa_bitmap)
	34	+{
	35	+ if (!isa_bitmap)
	36	+ return riscv_isa[0];
	37	+ return isa_bitmap[0];
	38	+}
	39	+EXPORT_SYMBOL_GPL(riscv_isa_extension_base);
	40	+
	41	+/**
	42	+ * __riscv_isa_extension_available() - Check whether given extension
	43	+ * is available or not
	44	+ *
	45	+ * @isa_bitmap: ISA bitmap to use
	46	+ * @bit: bit position of the desired extension
	47	+ * Return: true or false
	48	+ *
	49	+ * NOTE: If isa_bitmap is NULL then Host ISA bitmap will be used.
	50	+ */
	51	+bool __riscv_isa_extension_available(const unsigned long *isa_bitmap, int bit)
	52	+{
	53	+ const unsigned long *bmap = (isa_bitmap) ? isa_bitmap : riscv_isa;
	54	+
	55	+ if (bit >= RISCV_ISA_EXT_MAX)
	56	+ return false;
	57	+
	58	+ return test_bit(bit, bmap) ? true : false;
	59	+}
	60	+EXPORT_SYMBOL_GPL(__riscv_isa_extension_available);
	61	+
20	62	void riscv_fill_hwcap(void)
21	63	{
22	64	struct device_node *node;
23	65	const char *isa;
24		− size_t i;
	66	+ char print_str[BITS_PER_LONG + 1];
	67	+ size_t i, j, isa_len;
25	68	static unsigned long isa2hwcap[256] = {0};
26	69
27	70	isa2hwcap['i'] = isa2hwcap['I'] = COMPAT_HWCAP_ISA_I;
@@ -33,8 +76,11 @@ void riscv_fill_hwcap(void) · 5 unmodified lines
33	76
34	77	elf_hwcap = 0;
35	78
	79	+ bitmap_zero(riscv_isa, RISCV_ISA_EXT_MAX);
	80	+
36	81	for_each_of_cpu_node(node) {
37	82	unsigned long this_hwcap = 0;
	83	+ unsigned long this_isa = 0;
38	84
39	85	if (riscv_of_processor_hartid(node) < 0)
40	86	continue;
@@ -44,8 +90,24 @@ void riscv_fill_hwcap(void) · 3 unmodified lines
44	90	continue;
45	91	}
46	92
47		− for (i = 0; i < strlen(isa); ++i)
	93	+ i = 0;
	94	+ isa_len = strlen(isa);
	95	+#if IS_ENABLED(CONFIG_32BIT)
	96	+ if (!strncmp(isa, "rv32", 4))
	97	+ i += 4;
	98	+#elif IS_ENABLED(CONFIG_64BIT)
	99	+ if (!strncmp(isa, "rv64", 4))
	100	+ i += 4;
	101	+#endif
	102	+ for (; i < isa_len; ++i) {
48	103	this_hwcap \|= isa2hwcap[(unsigned char)(isa[i])];
	104	+ /*
	105	+ * TODO: X, Y and Z extension parsing for Host ISA
	106	+ * bitmap will be added in-future.
	107	+ */
	108	+ if ('a' <= isa[i] && isa[i] < 'x')
	109	+ this_isa \|= (1UL << (isa[i] - 'a'));
	110	+ }
49	111
50	112	/*
51	113	* All "okay" hart should have same isa. Set HWCAP based on
@@ -56,6 +118,11 @@ void riscv_fill_hwcap(void) · 4 unmodified lines
56	118	elf_hwcap &= this_hwcap;
57	119	else
58	120	elf_hwcap = this_hwcap;
	121	+
	122	+ if (riscv_isa[0])
	123	+ riscv_isa[0] &= this_isa;
	124	+ else
	125	+ riscv_isa[0] = this_isa;
59	126	}
60	127
61	128	/* We don't support systems with F but without D, so mask those out
@@ -65,7 +132,17 @@ void riscv_fill_hwcap(void) · 3 unmodified lines
65	132	elf_hwcap &= ~COMPAT_HWCAP_ISA_F;
66	133	}
67	134
68		− pr_info("elf_hwcap is 0x%lx\n", elf_hwcap);
	135	+ memset(print_str, 0, sizeof(print_str));
	136	+ for (i = 0, j = 0; i < BITS_PER_LONG; i++)
	137	+ if (riscv_isa[0] & BIT_MASK(i))
	138	+ print_str[j++] = (char)('a' + i);
	139	+ pr_info("riscv: ISA extensions %s\n", print_str);
	140	+
	141	+ memset(print_str, 0, sizeof(print_str));
	142	+ for (i = 0, j = 0; i < BITS_PER_LONG; i++)
	143	+ if (elf_hwcap & BIT_MASK(i))
	144	+ print_str[j++] = (char)('a' + i);
	145	+ pr_info("riscv: ELF capabilities %s\n", print_str);
69	146
70	147	#ifdef CONFIG_FPU
71	148	if (elf_hwcap & (COMPAT_HWCAP_ISA_F \| COMPAT_HWCAP_ISA_D))

arch/riscv/kernel/sbi.c+10−7

@@ -102,7 +102,7 @@ void sbi_shutdown(void) · 101 unmodified lines
102	102	{
103	103	sbi_ecall(SBI_EXT_0_1_SHUTDOWN, 0, 0, 0, 0, 0, 0, 0);
104	104	}
105		−EXPORT_SYMBOL(sbi_set_timer);
	105	+EXPORT_SYMBOL(sbi_shutdown);
106	106
107	107	/**
108	108	* sbi_clear_ipi() - Clear any pending IPIs for the calling hart.
@@ -113,7 +113,7 @@ void sbi_clear_ipi(void) · 4 unmodified lines
113	113	{
114	114	sbi_ecall(SBI_EXT_0_1_CLEAR_IPI, 0, 0, 0, 0, 0, 0, 0);
115	115	}
116		−EXPORT_SYMBOL(sbi_shutdown);
	116	+EXPORT_SYMBOL(sbi_clear_ipi);
117	117
118	118	/**
119	119	* sbi_set_timer_v01() - Program the timer for next timer event.
@@ -167,6 +167,11 @@ static int __sbi_rfence_v01(int fid, const unsigned long *hart_mask, · 47 unmodified lines
167	167
168	168	return result;
169	169	}
	170	+
	171	+static void sbi_set_power_off(void)
	172	+{
	173	+ pm_power_off = sbi_shutdown;
	174	+}
170	175	#else
171	176	static void __sbi_set_timer_v01(uint64_t stime_value)
172	177	{
@@ -191,6 +196,8 @@ static int __sbi_rfence_v01(int fid, const unsigned long *hart_mask, · 18 unmodified lines
191	196
192	197	return 0;
193	198	}
	199	+
	200	+static void sbi_set_power_off(void) {}
194	201	#endif /* CONFIG_RISCV_SBI_V01 */
195	202
196	203	static void __sbi_set_timer_v02(uint64_t stime_value)
@@ -540,16 +547,12 @@ static inline long sbi_get_firmware_version(void) · 343 unmodified lines
540	547	return __sbi_base_ecall(SBI_EXT_BASE_GET_IMP_VERSION);
541	548	}
542	549
543		−static void sbi_power_off(void)
544		−{
545		− sbi_shutdown();
546		−}
547	550
548	551	int __init sbi_init(void)
549	552	{
550	553	int ret;
551	554
552		− pm_power_off = sbi_power_off;
	555	+ sbi_set_power_off();
553	556	ret = sbi_get_spec_version();
554	557	if (ret > 0)
555	558	sbi_spec_version = ret;

arch/riscv/kernel/smp.c+2−0

@@ -10,6 +10,7 @@ · 9 unmodified lines
10	10
11	11	#include <linux/cpu.h>
12	12	#include <linux/interrupt.h>
	13	+#include <linux/module.h>
13	14	#include <linux/profile.h>
14	15	#include <linux/smp.h>
15	16	#include <linux/sched.h>
@@ -63,6 +64,7 @@ void riscv_cpuid_to_hartid_mask(const struct cpumask in, struct cpumask out) · 47 unmodified lines
63	64	for_each_cpu(cpu, in)
64	65	cpumask_set_cpu(cpuid_to_hartid_map(cpu), out);
65	66	}
	67	+EXPORT_SYMBOL_GPL(riscv_cpuid_to_hartid_mask);
66	68
67	69	bool arch_match_cpu_phys_id(int cpu, u64 phys_id)
68	70	{

arch/riscv/kernel/stacktrace.c+2−2

@@ -12,6 +12,8 @@ · 11 unmodified lines
12	12	#include <linux/stacktrace.h>
13	13	#include <linux/ftrace.h>
14	14
	15	+register unsigned long sp_in_global __asm__("sp");
	16	+
15	17	#ifdef CONFIG_FRAME_POINTER
16	18
17	19	struct stackframe {
@@ -19,8 +21,6 @@ struct stackframe { · 1 unmodified line
19	21	unsigned long ra;
20	22	};
21	23
22		−register unsigned long sp_in_global __asm__("sp");
23		−
24	24	void notrace walk_stackframe(struct task_struct task, struct pt_regs regs,
25	25	bool (fn)(unsigned long, void ), void *arg)
26	26	{

arch/riscv/kernel/vdso/Makefile+4−4

@@ -12,7 +12,7 @@ vdso-syms += getcpu · 11 unmodified lines
12	12	vdso-syms += flush_icache
13	13
14	14	# Files to link into the vdso
15		−obj-vdso = $(patsubst %, %.o, $(vdso-syms))
	15	+obj-vdso = $(patsubst %, %.o, $(vdso-syms)) note.o
16	16
17	17	# Build rules
18	18	targets := $(obj-vdso) vdso.so vdso.so.dbg vdso.lds vdso-dummy.o
@@ -33,15 +33,15 @@ $(obj)/vdso.so.dbg: $(src)/vdso.lds $(obj-vdso) FORCE · 14 unmodified lines
33	33	$(call if_changed,vdsold)
34	34
35	35	# We also create a special relocatable object that should mirror the symbol
36		−# table and layout of the linked DSO. With ld -R we can then refer to
37		−# these symbols in the kernel code rather than hand-coded addresses.
	36	+# table and layout of the linked DSO. With ld --just-symbols we can then
	37	+# refer to these symbols in the kernel code rather than hand-coded addresses.
38	38
39	39	SYSCFLAGS_vdso.so.dbg = -shared -s -Wl,-soname=linux-vdso.so.1 \
40	40	-Wl,--build-id -Wl,--hash-style=both
41	41	$(obj)/vdso-dummy.o: $(src)/vdso.lds $(obj)/rt_sigreturn.o FORCE
42	42	$(call if_changed,vdsold)
43	43
44		−LDFLAGS_vdso-syms.o := -r -R
	44	+LDFLAGS_vdso-syms.o := -r --just-symbols
45	45	$(obj)/vdso-syms.o: $(obj)/vdso-dummy.o FORCE
46	46	$(call if_changed,ld)
47	47

arch/riscv/kernel/vdso/note.S+12−0

@@ -0,0 +1,12 @@
	1	+/* SPDX-License-Identifier: GPL-2.0-or-later */
	2	+/*
	3	+ * This supplies .note.* sections to go into the PT_NOTE inside the vDSO text.
	4	+ * Here we can supply some information useful to userland.
	5	+ */
	6	+
	7	+#include <linux/elfnote.h>
	8	+#include <linux/version.h>
	9	+
	10	+ELFNOTE_START(Linux, 0, "a")
	11	+ .long LINUX_VERSION_CODE
	12	+ELFNOTE_END

arch/riscv/mm/init.c+2−17

@@ -150,7 +150,8 @@ void __init setup_bootmem(void) · 149 unmodified lines
150	150	memblock_reserve(vmlinux_start, vmlinux_end - vmlinux_start);
151	151
152	152	set_max_mapnr(PFN_DOWN(mem_size));
153		− max_low_pfn = PFN_DOWN(memblock_end_of_DRAM());
	153	+ max_pfn = PFN_DOWN(memblock_end_of_DRAM());
	154	+ max_low_pfn = max_pfn;
154	155
155	156	#ifdef CONFIG_BLK_DEV_INITRD
156	157	setup_initrd();
@@ -501,22 +502,6 @@ static inline void setup_vm_final(void) · 344 unmodified lines
501	502	#endif /* CONFIG_MMU */
502	503
503	504	#ifdef CONFIG_STRICT_KERNEL_RWX
504		−void set_kernel_text_rw(void)
505		−{
506		− unsigned long text_start = (unsigned long)_text;
507		− unsigned long text_end = (unsigned long)_etext;
508		−
509		− set_memory_rw(text_start, (text_end - text_start) >> PAGE_SHIFT);
510		−}
511		−
512		−void set_kernel_text_ro(void)
513		−{
514		− unsigned long text_start = (unsigned long)_text;
515		− unsigned long text_end = (unsigned long)_etext;
516		−
517		− set_memory_ro(text_start, (text_end - text_start) >> PAGE_SHIFT);
518		−}
519		−
520	505	void mark_rodata_ro(void)
521	506	{
522	507	unsigned long text_start = (unsigned long)_text;

arch/s390/boot/uv.c+0−2

@@ -7,9 +7,7 @@ · 6 unmodified lines
7	7	#ifdef CONFIG_PROTECTED_VIRTUALIZATION_GUEST
8	8	int __bootdata_preserved(prot_virt_guest);
9	9	#endif
10		−#if IS_ENABLED(CONFIG_KVM)
11	10	struct uv_info __bootdata_preserved(uv_info);
12		−#endif
13	11
14	12	void uv_query_info(void)
15	13	{

arch/s390/kernel/diag.c+1−1

@@ -133,7 +133,7 @@ void diag_stat_inc(enum diag_stat_enum nr) · 132 unmodified lines
133	133	}
134	134	EXPORT_SYMBOL(diag_stat_inc);
135	135
136		−void diag_stat_inc_norecursion(enum diag_stat_enum nr)
	136	+void notrace diag_stat_inc_norecursion(enum diag_stat_enum nr)
137	137	{
138	138	this_cpu_inc(diag_stat.counter[nr]);
139	139	trace_s390_diagnose_norecursion(diag_map[nr].code);

arch/s390/kernel/smp.c+2−2

@@ -403,7 +403,7 @@ int smp_find_processor_id(u16 address) · 402 unmodified lines
403	403	return -1;
404	404	}
405	405
406		−bool arch_vcpu_is_preempted(int cpu)
	406	+bool notrace arch_vcpu_is_preempted(int cpu)
407	407	{
408	408	if (test_cpu_flag_of(CIF_ENABLED_WAIT, cpu))
409	409	return false;
@@ -413,7 +413,7 @@ bool arch_vcpu_is_preempted(int cpu) · 3 unmodified lines
413	413	}
414	414	EXPORT_SYMBOL(arch_vcpu_is_preempted);
415	415
416		−void smp_yield_cpu(int cpu)
	416	+void notrace smp_yield_cpu(int cpu)
417	417	{
418	418	if (!MACHINE_HAS_DIAG9C)
419	419	return;

arch/s390/kernel/trace.c+1−1

@@ -14,7 +14,7 @@ EXPORT_TRACEPOINT_SYMBOL(s390_diagnose); · 13 unmodified lines
14	14
15	15	static DEFINE_PER_CPU(unsigned int, diagnose_trace_depth);
16	16
17		−void trace_s390_diagnose_norecursion(int diag_nr)
	17	+void notrace trace_s390_diagnose_norecursion(int diag_nr)
18	18	{
19	19	unsigned long flags;
20	20	unsigned int *depth;

arch/s390/kernel/uv.c+2−1

@@ -23,10 +23,11 @@ · 22 unmodified lines
23	23	int __bootdata_preserved(prot_virt_guest);
24	24	#endif
25	25
	26	+struct uv_info __bootdata_preserved(uv_info);
	27	+
26	28	#if IS_ENABLED(CONFIG_KVM)
27	29	int prot_virt_host;
28	30	EXPORT_SYMBOL(prot_virt_host);
29		−struct uv_info __bootdata_preserved(uv_info);
30	31	EXPORT_SYMBOL(uv_info);
31	32
32	33	static int __init prot_virt_setup(char *val)

arch/s390/kvm/interrupt.c+1−1

@@ -393,7 +393,7 @@ static unsigned long deliverable_irqs(struct kvm_vcpu *vcpu) · 392 unmodified lines
393	393	if (psw_mchk_disabled(vcpu))
394	394	active_mask &= ~IRQ_PEND_MCHK_MASK;
395	395	/* PV guest cpus can have a single interruption injected at a time. */
396		− if (kvm_s390_pv_cpu_is_protected(vcpu) &&
	396	+ if (kvm_s390_pv_cpu_get_handle(vcpu) &&
397	397	vcpu->arch.sie_block->iictl != IICTL_CODE_NONE)
398	398	active_mask &= ~(IRQ_PEND_EXT_II_MASK \|
399	399	IRQ_PEND_IO_MASK \|

arch/s390/kvm/kvm-s390.c+4−0

@@ -545,6 +545,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) · 544 unmodified lines
545	545	case KVM_CAP_S390_AIS:
546	546	case KVM_CAP_S390_AIS_MIGRATION:
547	547	case KVM_CAP_S390_VCPU_RESETS:
	548	+ case KVM_CAP_SET_GUEST_DEBUG:
548	549	r = 1;
549	550	break;
550	551	case KVM_CAP_S390_HPAGE_1M:
@@ -1939,6 +1940,9 @@ static int gfn_to_memslot_approx(struct kvm_memslots *slots, gfn_t gfn) · 1388 unmodified lines
1939	1940	start = slot + 1;
1940	1941	}
1941	1942
	1943	+ if (start >= slots->used_slots)
	1944	+ return slots->used_slots - 1;
	1945	+
1942	1946	if (gfn >= memslots[start].base_gfn &&
1943	1947	gfn < memslots[start].base_gfn + memslots[start].npages) {
1944	1948	atomic_set(&slots->lru_slot, start);

arch/s390/kvm/priv.c+3−1

@@ -626,10 +626,12 @@ static int handle_pqap(struct kvm_vcpu *vcpu) · 625 unmodified lines
626	626	* available for the guest are AQIC and TAPQ with the t bit set
627	627	* since we do not set IC.3 (FIII) we currently will only intercept
628	628	* the AQIC function code.
	629	+ * Note: running nested under z/VM can result in intercepts for other
	630	+ * function codes, e.g. PQAP(QCI). We do not support this and bail out.
629	631	*/
630	632	reg0 = vcpu->run->s.regs.gprs[0];
631	633	fc = (reg0 >> 24) & 0xff;
632		− if (WARN_ON_ONCE(fc != 0x03))
	634	+ if (fc != 0x03)
633	635	return -EOPNOTSUPP;
634	636
635	637	/* PQAP instruction is allowed for guest kernel only */

arch/s390/lib/uaccess.c+4−0

@@ -64,10 +64,13 @@ mm_segment_t enable_sacf_uaccess(void) · 63 unmodified lines
64	64	{
65	65	mm_segment_t old_fs;
66	66	unsigned long asce, cr;
	67	+ unsigned long flags;
67	68
68	69	old_fs = current->thread.mm_segment;
69	70	if (old_fs & 1)
70	71	return old_fs;
	72	+ /* protect against a concurrent page table upgrade */
	73	+ local_irq_save(flags);
71	74	current->thread.mm_segment \|= 1;
72	75	asce = S390_lowcore.kernel_asce;
73	76	if (likely(old_fs == USER_DS)) {
@@ -83,6 +86,7 @@ mm_segment_t enable_sacf_uaccess(void) · 9 unmodified lines
83	86	__ctl_load(asce, 7, 7);
84	87	set_cpu_flag(CIF_ASCE_SECONDARY);
85	88	}
	89	+ local_irq_restore(flags);
86	90	return old_fs;
87	91	}
88	92	EXPORT_SYMBOL(enable_sacf_uaccess);

arch/s390/mm/pgalloc.c+14−2

@@ -70,8 +70,20 @@ static void __crst_table_upgrade(void *arg) · 69 unmodified lines
70	70	{
71	71	struct mm_struct *mm = arg;
72	72
73		− if (current->active_mm == mm)
74		− set_user_asce(mm);
	73	+ /* we must change all active ASCEs to avoid the creation of new TLBs */
	74	+ if (current->active_mm == mm) {
	75	+ S390_lowcore.user_asce = mm->context.asce;
	76	+ if (current->thread.mm_segment == USER_DS) {
	77	+ __ctl_load(S390_lowcore.user_asce, 1, 1);
	78	+ /* Mark user-ASCE present in CR1 */
	79	+ clear_cpu_flag(CIF_ASCE_PRIMARY);
	80	+ }
	81	+ if (current->thread.mm_segment == USER_DS_SACF) {
	82	+ __ctl_load(S390_lowcore.user_asce, 7, 7);
	83	+ /* enable_sacf_uaccess does all or nothing */
	84	+ WARN_ON(!test_cpu_flag(CIF_ASCE_SECONDARY));
	85	+ }
	86	+ }
75	87	__tlb_flush_local();
76	88	}
77	89

arch/s390/pci/pci_irq.c+3−2

@@ -115,7 +115,6 @@ static struct irq_chip zpci_irq_chip = { · 114 unmodified lines
115	115	.name = "PCI-MSI",
116	116	.irq_unmask = pci_msi_unmask_irq,
117	117	.irq_mask = pci_msi_mask_irq,
118		− .irq_set_affinity = zpci_set_irq_affinity,
119	118	};
120	119
121	120	static void zpci_handle_cpu_local_irq(bool rescan)
@@ -276,7 +275,9 @@ int arch_setup_msi_irqs(struct pci_dev *pdev, int nvec, int type) · 154 unmodified lines
276	275	rc = -EIO;
277	276	if (hwirq - bit >= msi_vecs)
278	277	break;
279		− irq = __irq_alloc_descs(-1, 0, 1, 0, THIS_MODULE, msi->affinity);
	278	+ irq = __irq_alloc_descs(-1, 0, 1, 0, THIS_MODULE,
	279	+ (irq_delivery == DIRECTED) ?
	280	+ msi->affinity : NULL);
280	281	if (irq < 0)
281	282	return -ENOMEM;
282	283	rc = irq_set_msi_desc(irq, msi);

arch/sh/include/asm/module.h+0−28

@@ -11,32 +11,4 @@ struct mod_arch_specific { · 10 unmodified lines
11	11	};
12	12	#endif
13	13
14		−#ifdef CONFIG_CPU_LITTLE_ENDIAN
15		−# ifdef CONFIG_CPU_SH2
16		−# define MODULE_PROC_FAMILY "SH2LE "
17		−# elif defined CONFIG_CPU_SH3
18		−# define MODULE_PROC_FAMILY "SH3LE "
19		−# elif defined CONFIG_CPU_SH4
20		−# define MODULE_PROC_FAMILY "SH4LE "
21		−# elif defined CONFIG_CPU_SH5
22		−# define MODULE_PROC_FAMILY "SH5LE "
23		−# else
24		−# error unknown processor family
25		−# endif
26		−#else
27		−# ifdef CONFIG_CPU_SH2
28		−# define MODULE_PROC_FAMILY "SH2BE "
29		−# elif defined CONFIG_CPU_SH3
30		−# define MODULE_PROC_FAMILY "SH3BE "
31		−# elif defined CONFIG_CPU_SH4
32		−# define MODULE_PROC_FAMILY "SH4BE "
33		−# elif defined CONFIG_CPU_SH5
34		−# define MODULE_PROC_FAMILY "SH5BE "
35		−# else
36		−# error unknown processor family
37		−# endif
38		−#endif
39		−
40		−#define MODULE_ARCH_VERMAGIC MODULE_PROC_FAMILY
41		−
42	14	#endif /* _ASM_SH_MODULE_H */

arch/sh/include/asm/vermagic.h+34−0

@@ -0,0 +1,34 @@
	1	+/* SPDX-License-Identifier: GPL-2.0 */
	2	+
	3	+#ifndef _ASM_VERMAGIC_H
	4	+#define _ASM_VERMAGIC_H
	5	+
	6	+#ifdef CONFIG_CPU_LITTLE_ENDIAN
	7	+# ifdef CONFIG_CPU_SH2
	8	+# define MODULE_PROC_FAMILY "SH2LE "
	9	+# elif defined CONFIG_CPU_SH3
	10	+# define MODULE_PROC_FAMILY "SH3LE "
	11	+# elif defined CONFIG_CPU_SH4
	12	+# define MODULE_PROC_FAMILY "SH4LE "
	13	+# elif defined CONFIG_CPU_SH5
	14	+# define MODULE_PROC_FAMILY "SH5LE "
	15	+# else
	16	+# error unknown processor family
	17	+# endif
	18	+#else
	19	+# ifdef CONFIG_CPU_SH2
	20	+# define MODULE_PROC_FAMILY "SH2BE "
	21	+# elif defined CONFIG_CPU_SH3
	22	+# define MODULE_PROC_FAMILY "SH3BE "
	23	+# elif defined CONFIG_CPU_SH4
	24	+# define MODULE_PROC_FAMILY "SH4BE "
	25	+# elif defined CONFIG_CPU_SH5
	26	+# define MODULE_PROC_FAMILY "SH5BE "
	27	+# else
	28	+# error unknown processor family
	29	+# endif
	30	+#endif
	31	+
	32	+#define MODULE_ARCH_VERMAGIC MODULE_PROC_FAMILY
	33	+
	34	+#endif /* _ASM_VERMAGIC_H */

arch/sh/mm/init.c+1−1

@@ -412,7 +412,7 @@ int arch_add_memory(int nid, u64 start, u64 size, · 411 unmodified lines
412	412	unsigned long nr_pages = size >> PAGE_SHIFT;
413	413	int ret;
414	414
415		− if (WARN_ON_ONCE(params->pgprot.pgprot != PAGE_KERNEL.pgprot)
	415	+ if (WARN_ON_ONCE(params->pgprot.pgprot != PAGE_KERNEL.pgprot))
416	416	return -EINVAL;
417	417
418	418	/* We only have ZONE_NORMAL, so this is easy.. */

arch/um/Makefile+1−0

@@ -140,6 +140,7 @@ export CFLAGS_vmlinux := $(LINK-y) $(LINK_WRAPS) $(LD_FLAGS_CMDLINE) · 139 unmodified lines
140	140	# When cleaning we don't include .config, so we don't include
141	141	# TT or skas makefiles and don't clean skas_ptregs.h.
142	142	CLEAN_FILES += linux x.i gmon.out
	143	+MRPROPER_DIRS += arch/$(SUBARCH)/include/generated
143	144
144	145	archclean:
145	146	@find . \( -name '.bb' -o -name '.bbg' -o -name '*.da' \

arch/x86/Kconfig+1−1

@@ -149,7 +149,7 @@ config X86 · 148 unmodified lines
149	149	select HAVE_ARCH_TRACEHOOK
150	150	select HAVE_ARCH_TRANSPARENT_HUGEPAGE
151	151	select HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD if X86_64
152		− select HAVE_ARCH_USERFAULTFD_WP if USERFAULTFD
	152	+ select HAVE_ARCH_USERFAULTFD_WP if X86_64 && USERFAULTFD
153	153	select HAVE_ARCH_VMAP_STACK if X86_64
154	154	select HAVE_ARCH_WITHIN_STACK_FRAMES
155	155	select HAVE_ASM_MODVERSIONS

arch/x86/crypto/blake2s-glue.c+4−6

@@ -32,16 +32,16 @@ void blake2s_compress_arch(struct blake2s_state *state, · 31 unmodified lines
32	32	const u32 inc)
33	33	{
34	34	/* SIMD disables preemption, so relax after processing each page. */
35		− BUILD_BUG_ON(PAGE_SIZE / BLAKE2S_BLOCK_SIZE < 8);
	35	+ BUILD_BUG_ON(SZ_4K / BLAKE2S_BLOCK_SIZE < 8);
36	36
37	37	if (!static_branch_likely(&blake2s_use_ssse3) \|\| !crypto_simd_usable()) {
38	38	blake2s_compress_generic(state, block, nblocks, inc);
39	39	return;
40	40	}
41	41
42		− for (;;) {
	42	+ do {
43	43	const size_t blocks = min_t(size_t, nblocks,
44		− PAGE_SIZE / BLAKE2S_BLOCK_SIZE);
	44	+ SZ_4K / BLAKE2S_BLOCK_SIZE);
45	45
46	46	kernel_fpu_begin();
47	47	if (IS_ENABLED(CONFIG_AS_AVX512) &&
@@ -52,10 +52,8 @@ void blake2s_compress_arch(struct blake2s_state *state, · 4 unmodified lines
52	52	kernel_fpu_end();
53	53
54	54	nblocks -= blocks;
55		− if (!nblocks)
56		− break;
57	55	block += blocks * BLAKE2S_BLOCK_SIZE;
58		− }
	56	+ } while (nblocks);
59	57	}
60	58	EXPORT_SYMBOL(blake2s_compress_arch);
61	59

arch/x86/crypto/chacha_glue.c+11−3

@@ -153,9 +153,17 @@ void chacha_crypt_arch(u32 state, u8 dst, const u8 *src, unsigned int bytes, · 152 unmodified lines
153	153	bytes <= CHACHA_BLOCK_SIZE)
154	154	return chacha_crypt_generic(state, dst, src, bytes, nrounds);
155	155
156		− kernel_fpu_begin();
157		− chacha_dosimd(state, dst, src, bytes, nrounds);
158		− kernel_fpu_end();
	156	+ do {
	157	+ unsigned int todo = min_t(unsigned int, bytes, SZ_4K);
	158	+
	159	+ kernel_fpu_begin();
	160	+ chacha_dosimd(state, dst, src, todo, nrounds);
	161	+ kernel_fpu_end();
	162	+
	163	+ bytes -= todo;
	164	+ src += todo;
	165	+ dst += todo;
	166	+ } while (bytes);
159	167	}
160	168	EXPORT_SYMBOL(chacha_crypt_arch);
161	169

arch/x86/crypto/nhpoly1305-avx2-glue.c+1−1

@@ -29,7 +29,7 @@ static int nhpoly1305_avx2_update(struct shash_desc *desc, · 28 unmodified lines
29	29	return crypto_nhpoly1305_update(desc, src, srclen);
30	30
31	31	do {
32		− unsigned int n = min_t(unsigned int, srclen, PAGE_SIZE);
	32	+ unsigned int n = min_t(unsigned int, srclen, SZ_4K);
33	33
34	34	kernel_fpu_begin();
35	35	crypto_nhpoly1305_update_helper(desc, src, n, _nh_avx2);

arch/x86/crypto/nhpoly1305-sse2-glue.c+1−1

@@ -29,7 +29,7 @@ static int nhpoly1305_sse2_update(struct shash_desc *desc, · 28 unmodified lines
29	29	return crypto_nhpoly1305_update(desc, src, srclen);
30	30
31	31	do {
32		− unsigned int n = min_t(unsigned int, srclen, PAGE_SIZE);
	32	+ unsigned int n = min_t(unsigned int, srclen, SZ_4K);
33	33
34	34	kernel_fpu_begin();
35	35	crypto_nhpoly1305_update_helper(desc, src, n, _nh_sse2);

arch/x86/crypto/poly1305_glue.c+6−7

@@ -91,8 +91,8 @@ static void poly1305_simd_blocks(void ctx, const u8 inp, size_t len, · 90 unmodified lines
91	91	struct poly1305_arch_internal *state = ctx;
92	92
93	93	/* SIMD disables preemption, so relax after processing each page. */
94		− BUILD_BUG_ON(PAGE_SIZE < POLY1305_BLOCK_SIZE \|\|
95		− PAGE_SIZE % POLY1305_BLOCK_SIZE);
	94	+ BUILD_BUG_ON(SZ_4K < POLY1305_BLOCK_SIZE \|\|
	95	+ SZ_4K % POLY1305_BLOCK_SIZE);
96	96
97	97	if (!static_branch_likely(&poly1305_use_avx) \|\|
98	98	(len < (POLY1305_BLOCK_SIZE * 18) && !state->is_base2_26) \|\|
@@ -102,8 +102,8 @@ static void poly1305_simd_blocks(void ctx, const u8 inp, size_t len, · 3 unmodified lines
102	102	return;
103	103	}
104	104
105		− for (;;) {
106		− const size_t bytes = min_t(size_t, len, PAGE_SIZE);
	105	+ do {
	106	+ const size_t bytes = min_t(size_t, len, SZ_4K);
107	107
108	108	kernel_fpu_begin();
109	109	if (IS_ENABLED(CONFIG_AS_AVX512) && static_branch_likely(&poly1305_use_avx512))
@@ -113,11 +113,10 @@ static void poly1305_simd_blocks(void ctx, const u8 inp, size_t len, · 3 unmodified lines
113	113	else
114	114	poly1305_blocks_avx(ctx, inp, bytes, padbit);
115	115	kernel_fpu_end();
	116	+
116	117	len -= bytes;
117		− if (!len)
118		− break;
119	118	inp += bytes;
120		− }
	119	+ } while (len);
121	120	}
122	121
123	122	static void poly1305_simd_emit(void *ctx, u8 mac[POLY1305_DIGEST_SIZE],

arch/x86/entry/calling.h+21−19

@@ -98,13 +98,6 @@ For 32-bit we have the following conventions - kernel is built with · 97 unmodified lines
98	98	#define SIZEOF_PTREGS 21*8
99	99
100	100	.macro PUSH_AND_CLEAR_REGS rdx=%rdx rax=%rax save_ret=0
101		− /*
102		− * Push registers and sanitize registers of values that a
103		− * speculation attack might otherwise want to exploit. The
104		− * lower registers are likely clobbered well before they
105		− * could be put to use in a speculative execution gadget.
106		− * Interleave XOR with PUSH for better uop scheduling:
107		− */
108	101	.if \save_ret
109	102	pushq %rsi /* pt_regs->si */
110	103	movq 8(%rsp), %rsi /* temporarily store the return address in %rsi */
@@ -114,34 +107,43 @@ For 32-bit we have the following conventions - kernel is built with · 3 unmodified lines
114	107	pushq %rsi /* pt_regs->si */
115	108	.endif
116	109	pushq \rdx /* pt_regs->dx */
117		− xorl %edx, %edx /* nospec dx */
118	110	pushq %rcx /* pt_regs->cx */
119		− xorl %ecx, %ecx /* nospec cx */
120	111	pushq \rax /* pt_regs->ax */
121	112	pushq %r8 /* pt_regs->r8 */
122		− xorl %r8d, %r8d /* nospec r8 */
123	113	pushq %r9 /* pt_regs->r9 */
124		− xorl %r9d, %r9d /* nospec r9 */
125	114	pushq %r10 /* pt_regs->r10 */
126		− xorl %r10d, %r10d /* nospec r10 */
127	115	pushq %r11 /* pt_regs->r11 */
128		− xorl %r11d, %r11d /* nospec r11*/
129	116	pushq %rbx /* pt_regs->rbx */
130		− xorl %ebx, %ebx /* nospec rbx*/
131	117	pushq %rbp /* pt_regs->rbp */
132		− xorl %ebp, %ebp /* nospec rbp*/
133	118	pushq %r12 /* pt_regs->r12 */
134		− xorl %r12d, %r12d /* nospec r12*/
135	119	pushq %r13 /* pt_regs->r13 */
136		− xorl %r13d, %r13d /* nospec r13*/
137	120	pushq %r14 /* pt_regs->r14 */
138		− xorl %r14d, %r14d /* nospec r14*/
139	121	pushq %r15 /* pt_regs->r15 */
140		− xorl %r15d, %r15d /* nospec r15*/
141	122	UNWIND_HINT_REGS
	123	+
142	124	.if \save_ret
143	125	pushq %rsi /* return address on top of stack */
144	126	.endif
	127	+
	128	+ /*
	129	+ * Sanitize registers of values that a speculation attack might
	130	+ * otherwise want to exploit. The lower registers are likely clobbered
	131	+ * well before they could be put to use in a speculative execution
	132	+ * gadget.
	133	+ */
	134	+ xorl %edx, %edx /* nospec dx */
	135	+ xorl %ecx, %ecx /* nospec cx */
	136	+ xorl %r8d, %r8d /* nospec r8 */
	137	+ xorl %r9d, %r9d /* nospec r9 */
	138	+ xorl %r10d, %r10d /* nospec r10 */
	139	+ xorl %r11d, %r11d /* nospec r11 */
	140	+ xorl %ebx, %ebx /* nospec rbx */
	141	+ xorl %ebp, %ebp /* nospec rbp */
	142	+ xorl %r12d, %r12d /* nospec r12 */
	143	+ xorl %r13d, %r13d /* nospec r13 */
	144	+ xorl %r14d, %r14d /* nospec r14 */
	145	+ xorl %r15d, %r15d /* nospec r15 */
	146	+
145	147	.endm
146	148
147	149	.macro POP_REGS pop_rdi=1 skip_r11rcx=0

arch/x86/entry/entry_64.S+7−7

@@ -249,7 +249,6 @@ SYM_INNER_LABEL(entry_SYSCALL_64_after_hwframe, SYM_L_GLOBAL) · 248 unmodified lines
249	249	*/
250	250	syscall_return_via_sysret:
251	251	/* rcx and r11 are already restored (see code above) */
252		− UNWIND_HINT_EMPTY
253	252	POP_REGS pop_rdi=0 skip_r11rcx=1
254	253
255	254	/*
@@ -258,6 +257,7 @@ syscall_return_via_sysret: · 2 unmodified lines
258	257	*/
259	258	movq %rsp, %rdi
260	259	movq PER_CPU_VAR(cpu_tss_rw + TSS_sp0), %rsp
	260	+ UNWIND_HINT_EMPTY
261	261
262	262	pushq RSP-RDI(%rdi) /* RSP */
263	263	pushq (%rdi) /* RDI */
@@ -279,8 +279,7 @@ SYM_CODE_END(entry_SYSCALL_64) · 15 unmodified lines
279	279	* %rdi: prev task
280	280	* %rsi: next task
281	281	*/
282		−SYM_CODE_START(__switch_to_asm)
283		− UNWIND_HINT_FUNC
	282	+SYM_FUNC_START(__switch_to_asm)
284	283	/*
285	284	* Save callee-saved registers
286	285	* This must match the order in inactive_task_frame
@@ -321,7 +320,7 @@ SYM_CODE_START(__switch_to_asm) · 34 unmodified lines
321	320	popq %rbp
322	321
323	322	jmp __switch_to
324		−SYM_CODE_END(__switch_to_asm)
	323	+SYM_FUNC_END(__switch_to_asm)
325	324
326	325	/*
327	326	* A newly forked process directly context switches into this address.
@@ -512,7 +511,7 @@ SYM_CODE_END(spurious_entries_start) · 184 unmodified lines
512	511	* +----------------------------------------------------+
513	512	*/
514	513	SYM_CODE_START(interrupt_entry)
515		− UNWIND_HINT_FUNC
	514	+ UNWIND_HINT_IRET_REGS offset=16
516	515	ASM_CLAC
517	516	cld
518	517
@@ -544,9 +543,9 @@ SYM_CODE_START(interrupt_entry) · 25 unmodified lines
544	543	pushq 58(%rdi) / regs->eflags */
545	544	pushq 48(%rdi) / regs->cs */
546	545	pushq 38(%rdi) / regs->ip */
	546	+ UNWIND_HINT_IRET_REGS
547	547	pushq 28(%rdi) / regs->orig_ax */
548	548	pushq 8(%rdi) /* return address */
549		− UNWIND_HINT_FUNC
550	549
551	550	movq (%rdi), %rdi
552	551	jmp 2f
@@ -637,6 +636,7 @@ SYM_INNER_LABEL(swapgs_restore_regs_and_return_to_usermode, SYM_L_GLOBAL) · 84 unmodified lines
637	636	*/
638	637	movq %rsp, %rdi
639	638	movq PER_CPU_VAR(cpu_tss_rw + TSS_sp0), %rsp
	639	+ UNWIND_HINT_EMPTY
640	640
641	641	/* Copy the IRET frame to the trampoline stack. */
642	642	pushq 68(%rdi) / SS */
@@ -1739,7 +1739,7 @@ SYM_CODE_START(rewind_stack_do_exit) · 1096 unmodified lines
1739	1739
1740	1740	movq PER_CPU_VAR(cpu_current_top_of_stack), %rax
1741	1741	leaq -PTREGS_SIZE(%rax), %rsp
1742		− UNWIND_HINT_FUNC sp_offset=PTREGS_SIZE
	1742	+ UNWIND_HINT_REGS
1743	1743
1744	1744	call do_exit
1745	1745	SYM_CODE_END(rewind_stack_do_exit)

arch/x86/events/intel/cstate.c+1−0

@@ -643,6 +643,7 @@ static const struct x86_cpu_id intel_cstates_match[] __initconst = { · 642 unmodified lines
643	643	X86_MATCH_INTEL_FAM6_MODEL(ATOM_GOLDMONT_PLUS, &glm_cstates),
644	644	X86_MATCH_INTEL_FAM6_MODEL(ATOM_TREMONT_D, &glm_cstates),
645	645	X86_MATCH_INTEL_FAM6_MODEL(ATOM_TREMONT, &glm_cstates),
	646	+ X86_MATCH_INTEL_FAM6_MODEL(ATOM_TREMONT_L, &glm_cstates),
646	647
647	648	X86_MATCH_INTEL_FAM6_MODEL(ICELAKE_L, &icl_cstates),
648	649	X86_MATCH_INTEL_FAM6_MODEL(ICELAKE, &icl_cstates),

arch/x86/hyperv/hv_init.c+10−2

@@ -73,7 +73,8 @@ static int hv_cpu_init(unsigned int cpu) · 72 unmodified lines
73	73	struct page *pg;
74	74
75	75	input_arg = (void **)this_cpu_ptr(hyperv_pcpu_input_arg);
76		− pg = alloc_page(GFP_KERNEL);
	76	+ /* hv_cpu_init() can be called with IRQs disabled from hv_resume() */
	77	+ pg = alloc_page(irqs_disabled() ? GFP_ATOMIC : GFP_KERNEL);
77	78	if (unlikely(!pg))
78	79	return -ENOMEM;
79	80	*input_arg = page_address(pg);
@@ -254,6 +255,7 @@ static int __init hv_pci_init(void) · 174 unmodified lines
254	255	static int hv_suspend(void)
255	256	{
256	257	union hv_x64_msr_hypercall_contents hypercall_msr;
	258	+ int ret;
257	259
258	260	/*
259	261	* Reset the hypercall page as it is going to be invalidated
@@ -270,12 +272,17 @@ static int hv_suspend(void) · 10 unmodified lines
270	272	hypercall_msr.enable = 0;
271	273	wrmsrl(HV_X64_MSR_HYPERCALL, hypercall_msr.as_uint64);
272	274
273		− return 0;
	275	+ ret = hv_cpu_die(0);
	276	+ return ret;
274	277	}
275	278
276	279	static void hv_resume(void)
277	280	{
278	281	union hv_x64_msr_hypercall_contents hypercall_msr;
	282	+ int ret;
	283	+
	284	+ ret = hv_cpu_init(0);
	285	+ WARN_ON(ret);
279	286
280	287	/* Re-enable the hypercall page */
281	288	rdmsrl(HV_X64_MSR_HYPERCALL, hypercall_msr.as_uint64);
@@ -288,6 +295,7 @@ static void hv_resume(void) · 6 unmodified lines
288	295	hv_hypercall_pg_saved = NULL;
289	296	}
290	297
	298	+/* Note: when the ops are called, only CPU0 is online and IRQs are disabled. */
291	299	static struct syscore_ops hv_syscore_ops = {
292	300	.suspend = hv_suspend,
293	301	.resume = hv_resume,

arch/x86/include/asm/ftrace.h+3−2

@@ -61,11 +61,12 @@ static inline bool arch_syscall_match_sym_name(const char sym, const char name · 60 unmodified lines
61	61	{
62	62	/*
63	63	* Compare the symbol name with the system call name. Skip the
64		− * "__x64_sys", "__ia32_sys" or simple "sys" prefix.
	64	+ * "__x64_sys", "__ia32_sys", "__do_sys" or simple "sys" prefix.
65	65	*/
66	66	return !strcmp(sym + 3, name + 3) \|\|
67	67	(!strncmp(sym, "__x64_", 6) && !strcmp(sym + 9, name + 3)) \|\|
68		− (!strncmp(sym, "__ia32_", 7) && !strcmp(sym + 10, name + 3));
	68	+ (!strncmp(sym, "__ia32_", 7) && !strcmp(sym + 10, name + 3)) \|\|
	69	+ (!strncmp(sym, "__do_sys", 8) && !strcmp(sym + 8, name + 3));
69	70	}
70	71
71	72	#ifndef COMPILE_OFFSETS

arch/x86/include/asm/kvm_host.h+2−2

@@ -1663,8 +1663,8 @@ void kvm_set_msi_irq(struct kvm kvm, struct kvm_kernel_irq_routing_entry e, · 1662 unmodified lines
1663	1663	static inline bool kvm_irq_is_postable(struct kvm_lapic_irq *irq)
1664	1664	{
1665	1665	/* We can only post Fixed and LowPrio IRQs */
1666		− return (irq->delivery_mode == dest_Fixed \|\|
1667		− irq->delivery_mode == dest_LowestPrio);
	1666	+ return (irq->delivery_mode == APIC_DM_FIXED \|\|
	1667	+ irq->delivery_mode == APIC_DM_LOWEST);
1668	1668	}
1669	1669
1670	1670	static inline void kvm_arch_vcpu_blocking(struct kvm_vcpu *vcpu)

arch/x86/include/asm/module.h+0−60

@@ -13,64 +13,4 @@ struct mod_arch_specific { · 12 unmodified lines
13	13	#endif
14	14	};
15	15
16		−#ifdef CONFIG_X86_64
17		−/* X86_64 does not define MODULE_PROC_FAMILY */
18		−#elif defined CONFIG_M486SX
19		−#define MODULE_PROC_FAMILY "486SX "
20		−#elif defined CONFIG_M486
21		−#define MODULE_PROC_FAMILY "486 "
22		−#elif defined CONFIG_M586
23		−#define MODULE_PROC_FAMILY "586 "
24		−#elif defined CONFIG_M586TSC
25		−#define MODULE_PROC_FAMILY "586TSC "
26		−#elif defined CONFIG_M586MMX
27		−#define MODULE_PROC_FAMILY "586MMX "
28		−#elif defined CONFIG_MCORE2
29		−#define MODULE_PROC_FAMILY "CORE2 "
30		−#elif defined CONFIG_MATOM
31		−#define MODULE_PROC_FAMILY "ATOM "
32		−#elif defined CONFIG_M686
33		−#define MODULE_PROC_FAMILY "686 "
34		−#elif defined CONFIG_MPENTIUMII
35		−#define MODULE_PROC_FAMILY "PENTIUMII "
36		−#elif defined CONFIG_MPENTIUMIII
37		−#define MODULE_PROC_FAMILY "PENTIUMIII "
38		−#elif defined CONFIG_MPENTIUMM
39		−#define MODULE_PROC_FAMILY "PENTIUMM "
40		−#elif defined CONFIG_MPENTIUM4
41		−#define MODULE_PROC_FAMILY "PENTIUM4 "
42		−#elif defined CONFIG_MK6
43		−#define MODULE_PROC_FAMILY "K6 "
44		−#elif defined CONFIG_MK7
45		−#define MODULE_PROC_FAMILY "K7 "
46		−#elif defined CONFIG_MK8
47		−#define MODULE_PROC_FAMILY "K8 "
48		−#elif defined CONFIG_MELAN
49		−#define MODULE_PROC_FAMILY "ELAN "
50		−#elif defined CONFIG_MCRUSOE
51		−#define MODULE_PROC_FAMILY "CRUSOE "
52		−#elif defined CONFIG_MEFFICEON
53		−#define MODULE_PROC_FAMILY "EFFICEON "
54		−#elif defined CONFIG_MWINCHIPC6
55		−#define MODULE_PROC_FAMILY "WINCHIPC6 "
56		−#elif defined CONFIG_MWINCHIP3D
57		−#define MODULE_PROC_FAMILY "WINCHIP3D "
58		−#elif defined CONFIG_MCYRIXIII
59		−#define MODULE_PROC_FAMILY "CYRIXIII "
60		−#elif defined CONFIG_MVIAC3_2
61		−#define MODULE_PROC_FAMILY "VIAC3-2 "
62		−#elif defined CONFIG_MVIAC7
63		−#define MODULE_PROC_FAMILY "VIAC7 "
64		−#elif defined CONFIG_MGEODEGX1
65		−#define MODULE_PROC_FAMILY "GEODEGX1 "
66		−#elif defined CONFIG_MGEODE_LX
67		−#define MODULE_PROC_FAMILY "GEODE "
68		−#else
69		−#error unknown processor family
70		−#endif
71		−
72		−#ifdef CONFIG_X86_32
73		−# define MODULE_ARCH_VERMAGIC MODULE_PROC_FAMILY
74		−#endif
75		−
76	16	#endif /* _ASM_X86_MODULE_H */

arch/x86/include/asm/mshyperv.h+2−0

@@ -35,6 +35,8 @@ typedef int (*hyperv_fill_flush_list_func)( · 34 unmodified lines
35	35	rdmsrl(HV_X64_MSR_SINT0 + int_num, val)
36	36	#define hv_set_synint_state(int_num, val) \
37	37	wrmsrl(HV_X64_MSR_SINT0 + int_num, val)
	38	+#define hv_recommend_using_aeoi() \
	39	+ (!(ms_hyperv.hints & HV_DEPRECATING_AEOI_RECOMMENDED))
38	40
39	41	#define hv_get_crash_ctl(val) \
40	42	rdmsrl(HV_X64_MSR_CRASH_CTL, val)

arch/x86/include/asm/nospec-branch.h+0−21

@@ -237,27 +237,6 @@ enum ssb_mitigation { · 236 unmodified lines
237	237	extern char __indirect_thunk_start[];
238	238	extern char __indirect_thunk_end[];
239	239
240		−/*
241		− * On VMEXIT we must ensure that no RSB predictions learned in the guest
242		− * can be followed in the host, by overwriting the RSB completely. Both
243		− * retpoline and IBRS mitigations for Spectre v2 need this; only on future
244		− * CPUs with IBRS_ALL might it be avoided.
245		− */
246		−static inline void vmexit_fill_RSB(void)
247		−{
248		−#ifdef CONFIG_RETPOLINE
249		− unsigned long loops;
250		−
251		− asm volatile (ANNOTATE_NOSPEC_ALTERNATIVE
252		− ALTERNATIVE("jmp 910f",
253		− __stringify(__FILL_RETURN_BUFFER(%0, RSB_CLEAR_LOOPS, %1)),
254		− X86_FEATURE_RETPOLINE)
255		− "910:"
256		− : "=r" (loops), ASM_CALL_CONSTRAINT
257		− : : "memory" );
258		−#endif
259		−}
260		−
261	240	static __always_inline
262	241	void alternative_msr_write(unsigned int msr, u64 val, unsigned int feature)
263	242	{

arch/x86/include/asm/unwind.h+1−1

@@ -19,7 +19,7 @@ struct unwind_state { · 18 unmodified lines
19	19	#if defined(CONFIG_UNWINDER_ORC)
20	20	bool signal, full_regs;
21	21	unsigned long sp, bp, ip;
22		− struct pt_regs *regs;
	22	+ struct pt_regs regs, prev_regs;
23	23	#elif defined(CONFIG_UNWINDER_FRAME_POINTER)
24	24	bool got_irq;
25	25	unsigned long bp, orig_sp, ip;

arch/x86/include/asm/vermagic.h+68−0

@@ -0,0 +1,68 @@
	1	+/* SPDX-License-Identifier: GPL-2.0 */
	2	+
	3	+#ifndef _ASM_VERMAGIC_H
	4	+#define _ASM_VERMAGIC_H
	5	+
	6	+#ifdef CONFIG_X86_64
	7	+/* X86_64 does not define MODULE_PROC_FAMILY */
	8	+#elif defined CONFIG_M486SX
	9	+#define MODULE_PROC_FAMILY "486SX "
	10	+#elif defined CONFIG_M486
	11	+#define MODULE_PROC_FAMILY "486 "
	12	+#elif defined CONFIG_M586
	13	+#define MODULE_PROC_FAMILY "586 "
	14	+#elif defined CONFIG_M586TSC
	15	+#define MODULE_PROC_FAMILY "586TSC "
	16	+#elif defined CONFIG_M586MMX
	17	+#define MODULE_PROC_FAMILY "586MMX "
	18	+#elif defined CONFIG_MCORE2
	19	+#define MODULE_PROC_FAMILY "CORE2 "
	20	+#elif defined CONFIG_MATOM
	21	+#define MODULE_PROC_FAMILY "ATOM "
	22	+#elif defined CONFIG_M686
	23	+#define MODULE_PROC_FAMILY "686 "
	24	+#elif defined CONFIG_MPENTIUMII
	25	+#define MODULE_PROC_FAMILY "PENTIUMII "
	26	+#elif defined CONFIG_MPENTIUMIII
	27	+#define MODULE_PROC_FAMILY "PENTIUMIII "
	28	+#elif defined CONFIG_MPENTIUMM
	29	+#define MODULE_PROC_FAMILY "PENTIUMM "
	30	+#elif defined CONFIG_MPENTIUM4
	31	+#define MODULE_PROC_FAMILY "PENTIUM4 "
	32	+#elif defined CONFIG_MK6
	33	+#define MODULE_PROC_FAMILY "K6 "
	34	+#elif defined CONFIG_MK7
	35	+#define MODULE_PROC_FAMILY "K7 "
	36	+#elif defined CONFIG_MK8
	37	+#define MODULE_PROC_FAMILY "K8 "
	38	+#elif defined CONFIG_MELAN
	39	+#define MODULE_PROC_FAMILY "ELAN "
	40	+#elif defined CONFIG_MCRUSOE
	41	+#define MODULE_PROC_FAMILY "CRUSOE "
	42	+#elif defined CONFIG_MEFFICEON
	43	+#define MODULE_PROC_FAMILY "EFFICEON "
	44	+#elif defined CONFIG_MWINCHIPC6
	45	+#define MODULE_PROC_FAMILY "WINCHIPC6 "
	46	+#elif defined CONFIG_MWINCHIP3D
	47	+#define MODULE_PROC_FAMILY "WINCHIP3D "
	48	+#elif defined CONFIG_MCYRIXIII
	49	+#define MODULE_PROC_FAMILY "CYRIXIII "
	50	+#elif defined CONFIG_MVIAC3_2
	51	+#define MODULE_PROC_FAMILY "VIAC3-2 "
	52	+#elif defined CONFIG_MVIAC7
	53	+#define MODULE_PROC_FAMILY "VIAC7 "
	54	+#elif defined CONFIG_MGEODEGX1
	55	+#define MODULE_PROC_FAMILY "GEODEGX1 "
	56	+#elif defined CONFIG_MGEODE_LX
	57	+#define MODULE_PROC_FAMILY "GEODE "
	58	+#else
	59	+#error unknown processor family
	60	+#endif
	61	+
	62	+#ifdef CONFIG_X86_32
	63	+# define MODULE_ARCH_VERMAGIC MODULE_PROC_FAMILY
	64	+#else
	65	+# define MODULE_ARCH_VERMAGIC ""
	66	+#endif
	67	+
	68	+#endif /* _ASM_VERMAGIC_H */

arch/x86/kernel/apic/apic.c+14−13

@@ -352,8 +352,6 @@ static void __setup_APIC_LVTT(unsigned int clocks, int oneshot, int irqen) · 351 unmodified lines
352	352	* According to Intel, MFENCE can do the serialization here.
353	353	*/
354	354	asm volatile("mfence" : : : "memory");
355		−
356		− printk_once(KERN_DEBUG "TSC deadline timer enabled\n");
357	355	return;
358	356	}
359	357
@@ -546,7 +544,7 @@ static struct clock_event_device lapic_clockevent = { · 186 unmodified lines
546	544	};
547	545	static DEFINE_PER_CPU(struct clock_event_device, lapic_events);
548	546
549		−static u32 hsx_deadline_rev(void)
	547	+static __init u32 hsx_deadline_rev(void)
550	548	{
551	549	switch (boot_cpu_data.x86_stepping) {
552	550	case 0x02: return 0x3a; /* EP */
@@ -556,7 +554,7 @@ static u32 hsx_deadline_rev(void) · 3 unmodified lines
556	554	return ~0U;
557	555	}
558	556
559		−static u32 bdx_deadline_rev(void)
	557	+static __init u32 bdx_deadline_rev(void)
560	558	{
561	559	switch (boot_cpu_data.x86_stepping) {
562	560	case 0x02: return 0x00000011;
@@ -568,7 +566,7 @@ static u32 bdx_deadline_rev(void) · 5 unmodified lines
568	566	return ~0U;
569	567	}
570	568
571		−static u32 skx_deadline_rev(void)
	569	+static __init u32 skx_deadline_rev(void)
572	570	{
573	571	switch (boot_cpu_data.x86_stepping) {
574	572	case 0x03: return 0x01000136;
@@ -581,7 +579,7 @@ static u32 skx_deadline_rev(void) · 6 unmodified lines
581	579	return ~0U;
582	580	}
583	581
584		−static const struct x86_cpu_id deadline_match[] = {
	582	+static const struct x86_cpu_id deadline_match[] __initconst = {
585	583	X86_MATCH_INTEL_FAM6_MODEL( HASWELL_X, &hsx_deadline_rev),
586	584	X86_MATCH_INTEL_FAM6_MODEL( BROADWELL_X, 0x0b000020),
587	585	X86_MATCH_INTEL_FAM6_MODEL( BROADWELL_D, &bdx_deadline_rev),
@@ -603,18 +601,19 @@ static const struct x86_cpu_id deadline_match[] = { · 15 unmodified lines
603	601	{},
604	602	};
605	603
606		−static void apic_check_deadline_errata(void)
	604	+static __init bool apic_validate_deadline_timer(void)
607	605	{
608	606	const struct x86_cpu_id *m;
609	607	u32 rev;
610	608
611		− if (!boot_cpu_has(X86_FEATURE_TSC_DEADLINE_TIMER) \|\|
612		− boot_cpu_has(X86_FEATURE_HYPERVISOR))
613		− return;
	609	+ if (!boot_cpu_has(X86_FEATURE_TSC_DEADLINE_TIMER))
	610	+ return false;
	611	+ if (boot_cpu_has(X86_FEATURE_HYPERVISOR))
	612	+ return true;
614	613
615	614	m = x86_match_cpu(deadline_match);
616	615	if (!m)
617		− return;
	616	+ return true;
618	617
619	618	/*
620	619	* Function pointers will have the MSB set due to address layout,
@@ -626,11 +625,12 @@ static void apic_check_deadline_errata(void) · 5 unmodified lines
626	625	rev = (u32)m->driver_data;
627	626
628	627	if (boot_cpu_data.microcode >= rev)
629		− return;
	628	+ return true;
630	629
631	630	setup_clear_cpu_cap(X86_FEATURE_TSC_DEADLINE_TIMER);
632	631	pr_err(FW_BUG "TSC_DEADLINE disabled due to Errata; "
633	632	"please update microcode to version: 0x%x (or later)\n", rev);
	633	+ return false;
634	634	}
635	635
636	636	/*
@@ -2092,7 +2092,8 @@ void __init init_apic_mappings(void) · 1455 unmodified lines
2092	2092	{
2093	2093	unsigned int new_apicid;
2094	2094
2095		− apic_check_deadline_errata();
	2095	+ if (apic_validate_deadline_timer())
	2096	+ pr_debug("TSC deadline timer available\n");
2096	2097
2097	2098	if (x2apic_mode) {
2098	2099	boot_cpu_physical_apicid = read_apic_id();

arch/x86/kernel/dumpstack_64.c+2−1

@@ -183,7 +183,8 @@ recursion_check: · 182 unmodified lines
183	183	*/
184	184	if (visit_mask) {
185	185	if (*visit_mask & (1UL << info->type)) {
186		− printk_deferred_once(KERN_WARNING "WARNING: stack recursion on stack type %d\n", info->type);
	186	+ if (task == current)
	187	+ printk_deferred_once(KERN_WARNING "WARNING: stack recursion on stack type %d\n", info->type);
187	188	goto unknown;
188	189	}
189	190	*visit_mask \|= 1UL << info->type;

arch/x86/kernel/smpboot.c+33−14

@@ -147,7 +147,7 @@ static inline void smpboot_restore_warm_reset_vector(void) · 146 unmodified lines
147	147	((volatile u32 )phys_to_virt(TRAMPOLINE_PHYS_LOW)) = 0;
148	148	}
149	149
150		−static void init_freq_invariance(void);
	150	+static void init_freq_invariance(bool secondary);
151	151
152	152	/*
153	153	* Report back to the Boot Processor during boot time or to the caller processor
@@ -185,7 +185,7 @@ static void smp_callin(void) · 31 unmodified lines
185	185	*/
186	186	set_cpu_sibling_map(raw_smp_processor_id());
187	187
188		− init_freq_invariance();
	188	+ init_freq_invariance(true);
189	189
190	190	/*
191	191	* Get our bogomips.
@@ -1341,7 +1341,7 @@ void __init native_smp_prepare_cpus(unsigned int max_cpus) · 1149 unmodified lines
1341	1341	set_sched_topology(x86_topology);
1342	1342
1343	1343	set_cpu_sibling_map(0);
1344		− init_freq_invariance();
	1344	+ init_freq_invariance(false);
1345	1345	smp_sanity_check();
1346	1346
1347	1347	switch (apic_intr_mode) {
@@ -1877,9 +1877,6 @@ static bool knl_set_max_freq_ratio(u64 base_freq, u64 turbo_freq, · 529 unmodified lines
1877	1877	int err, i;
1878	1878	u64 msr;
1879	1879
1880		− if (!x86_match_cpu(has_knl_turbo_ratio_limits))
1881		− return false;
1882		−
1883	1880	err = rdmsrl_safe(MSR_PLATFORM_INFO, base_freq);
1884	1881	if (err)
1885	1882	return false;
@@ -1945,18 +1942,23 @@ static bool skx_set_max_freq_ratio(u64 base_freq, u64 turbo_freq, int size) · 59 unmodified lines
1945	1942
1946	1943	static bool core_set_max_freq_ratio(u64 base_freq, u64 turbo_freq)
1947	1944	{
	1945	+ u64 msr;
1948	1946	int err;
1949	1947
1950	1948	err = rdmsrl_safe(MSR_PLATFORM_INFO, base_freq);
1951	1949	if (err)
1952	1950	return false;
1953	1951
1954		− err = rdmsrl_safe(MSR_TURBO_RATIO_LIMIT, turbo_freq);
	1952	+ err = rdmsrl_safe(MSR_TURBO_RATIO_LIMIT, &msr);
1955	1953	if (err)
1956	1954	return false;
1957	1955
1958		− base_freq = (base_freq >> 8) & 0xFF; /* max P state */
1959		− turbo_freq = (turbo_freq >> 24) & 0xFF; /* 4C turbo */
	1956	+ base_freq = (base_freq >> 8) & 0xFF; /* max P state */
	1957	+ turbo_freq = (msr >> 24) & 0xFF; / 4C turbo */
	1958	+
	1959	+ /* The CPU may have less than 4 cores */
	1960	+ if (!*turbo_freq)
	1961	+ turbo_freq = msr & 0xFF; / 1C turbo */
1960	1962
1961	1963	return true;
1962	1964	}
@@ -1972,7 +1974,8 @@ static bool intel_set_max_freq_ratio(void) · 9 unmodified lines
1972	1974	skx_set_max_freq_ratio(&base_freq, &turbo_freq, 1))
1973	1975	goto out;
1974	1976
1975		− if (knl_set_max_freq_ratio(&base_freq, &turbo_freq, 1))
	1977	+ if (x86_match_cpu(has_knl_turbo_ratio_limits) &&
	1978	+ knl_set_max_freq_ratio(&base_freq, &turbo_freq, 1))
1976	1979	goto out;
1977	1980
1978	1981	if (x86_match_cpu(has_skx_turbo_ratio_limits) &&
@@ -1985,13 +1988,22 @@ static bool intel_set_max_freq_ratio(void) · 6 unmodified lines
1985	1988	return false;
1986	1989
1987	1990	out:
	1991	+ /*
	1992	+ * Some hypervisors advertise X86_FEATURE_APERFMPERF
	1993	+ * but then fill all MSR's with zeroes.
	1994	+ */
	1995	+ if (!base_freq) {
	1996	+ pr_debug("Couldn't determine cpu base frequency, necessary for scale-invariant accounting.\n");
	1997	+ return false;
	1998	+ }
	1999	+
1988	2000	arch_turbo_freq_ratio = div_u64(turbo_freq * SCHED_CAPACITY_SCALE,
1989	2001	base_freq);
1990	2002	arch_set_max_freq_ratio(turbo_disabled());
1991	2003	return true;
1992	2004	}
1993	2005
1994		−static void init_counter_refs(void *arg)
	2006	+static void init_counter_refs(void)
1995	2007	{
1996	2008	u64 aperf, mperf;
1997	2009
@@ -2002,18 +2014,25 @@ static void init_counter_refs(void *arg) · 4 unmodified lines
2002	2014	this_cpu_write(arch_prev_mperf, mperf);
2003	2015	}
2004	2016
2005		−static void init_freq_invariance(void)
	2017	+static void init_freq_invariance(bool secondary)
2006	2018	{
2007	2019	bool ret = false;
2008	2020
2009		− if (smp_processor_id() != 0 \|\| !boot_cpu_has(X86_FEATURE_APERFMPERF))
	2021	+ if (!boot_cpu_has(X86_FEATURE_APERFMPERF))
2010	2022	return;
2011	2023
	2024	+ if (secondary) {
	2025	+ if (static_branch_likely(&arch_scale_freq_key)) {
	2026	+ init_counter_refs();
	2027	+ }
	2028	+ return;
	2029	+ }
	2030	+
2012	2031	if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL)
2013	2032	ret = intel_set_max_freq_ratio();
2014	2033
2015	2034	if (ret) {
2016		− on_each_cpu(init_counter_refs, NULL, 1);
	2035	+ init_counter_refs();
2017	2036	static_branch_enable(&arch_scale_freq_key);
2018	2037	} else {
2019	2038	pr_debug("Couldn't determine max cpu frequency, necessary for scale-invariant accounting.\n");

arch/x86/kernel/unwind_frame.c+3−0

@@ -344,6 +344,9 @@ bad_address: · 343 unmodified lines
344	344	if (IS_ENABLED(CONFIG_X86_32))
345	345	goto the_end;
346	346
	347	+ if (state->task != current)
	348	+ goto the_end;
	349	+
347	350	if (state->regs) {
348	351	printk_deferred_once(KERN_WARNING
349	352	"WARNING: kernel stack regs at %p in %s:%d has bad 'bp' value %p\n",

arch/x86/kernel/unwind_orc.c+74−39

@@ -8,19 +8,21 @@ · 7 unmodified lines
8	8	#include <asm/orc_lookup.h>
9	9
10	10	#define orc_warn(fmt, ...) \
11		− printk_deferred_once(KERN_WARNING pr_fmt("WARNING: " fmt), ##__VA_ARGS__)
	11	+ printk_deferred_once(KERN_WARNING "WARNING: " fmt, ##__VA_ARGS__)
	12	+
	13	+#define orc_warn_current(args...) \
	14	+({ \
	15	+ if (state->task == current) \
	16	+ orc_warn(args); \
	17	+})
12	18
13	19	extern int __start_orc_unwind_ip[];
14	20	extern int __stop_orc_unwind_ip[];
15	21	extern struct orc_entry __start_orc_unwind[];
16	22	extern struct orc_entry __stop_orc_unwind[];
17	23
18		−static DEFINE_MUTEX(sort_mutex);
19		−int *cur_orc_ip_table = __start_orc_unwind_ip;
20		−struct orc_entry *cur_orc_table = __start_orc_unwind;
21		−
22		−unsigned int lookup_num_blocks;
23		−bool orc_init;
	24	+static bool orc_init __ro_after_init;
	25	+static unsigned int lookup_num_blocks __ro_after_init;
24	26
25	27	static inline unsigned long orc_ip(const int *ip)
26	28	{
@@ -142,9 +144,6 @@ static struct orc_entry *orc_find(unsigned long ip) · 115 unmodified lines
142	144	{
143	145	static struct orc_entry *orc;
144	146
145		− if (!orc_init)
146		− return NULL;
147		−
148	147	if (ip == 0)
149	148	return &null_orc_entry;
150	149
@@ -189,6 +188,10 @@ static struct orc_entry *orc_find(unsigned long ip) · 38 unmodified lines
189	188
190	189	#ifdef CONFIG_MODULES
191	190
	191	+static DEFINE_MUTEX(sort_mutex);
	192	+static int *cur_orc_ip_table = __start_orc_unwind_ip;
	193	+static struct orc_entry *cur_orc_table = __start_orc_unwind;
	194	+
192	195	static void orc_sort_swap(void _a, void _b, int size)
193	196	{
194	197	struct orc_entry orc_a, orc_b;
@@ -381,9 +384,38 @@ static bool deref_stack_iret_regs(struct unwind_state *state, unsigned long addr · 186 unmodified lines
381	384	return true;
382	385	}
383	386
	387	+/*
	388	+ * If state->regs is non-NULL, and points to a full pt_regs, just get the reg
	389	+ * value from state->regs.
	390	+ *
	391	+ * Otherwise, if state->regs just points to IRET regs, and the previous frame
	392	+ * had full regs, it's safe to get the value from the previous regs. This can
	393	+ * happen when early/late IRQ entry code gets interrupted by an NMI.
	394	+ */
	395	+static bool get_reg(struct unwind_state *state, unsigned int reg_off,
	396	+ unsigned long *val)
	397	+{
	398	+ unsigned int reg = reg_off/8;
	399	+
	400	+ if (!state->regs)
	401	+ return false;
	402	+
	403	+ if (state->full_regs) {
	404	+ val = ((unsigned long )state->regs)[reg];
	405	+ return true;
	406	+ }
	407	+
	408	+ if (state->prev_regs) {
	409	+ val = ((unsigned long )state->prev_regs)[reg];
	410	+ return true;
	411	+ }
	412	+
	413	+ return false;
	414	+}
	415	+
384	416	bool unwind_next_frame(struct unwind_state *state)
385	417	{
386		− unsigned long ip_p, sp, orig_ip = state->ip, prev_sp = state->sp;
	418	+ unsigned long ip_p, sp, tmp, orig_ip = state->ip, prev_sp = state->sp;
387	419	enum stack_type prev_type = state->stack_info.type;
388	420	struct orc_entry *orc;
389	421	bool indirect = false;
@@ -445,43 +477,39 @@ bool unwind_next_frame(struct unwind_state *state) · 55 unmodified lines
445	477	break;
446	478
447	479	case ORC_REG_R10:
448		− if (!state->regs \|\| !state->full_regs) {
449		− orc_warn("missing regs for base reg R10 at ip %pB\n",
450		− (void *)state->ip);
	480	+ if (!get_reg(state, offsetof(struct pt_regs, r10), &sp)) {
	481	+ orc_warn_current("missing R10 value at %pB\n",
	482	+ (void *)state->ip);
451	483	goto err;
452	484	}
453		− sp = state->regs->r10;
454	485	break;
455	486
456	487	case ORC_REG_R13:
457		− if (!state->regs \|\| !state->full_regs) {
458		− orc_warn("missing regs for base reg R13 at ip %pB\n",
459		− (void *)state->ip);
	488	+ if (!get_reg(state, offsetof(struct pt_regs, r13), &sp)) {
	489	+ orc_warn_current("missing R13 value at %pB\n",
	490	+ (void *)state->ip);
460	491	goto err;
461	492	}
462		− sp = state->regs->r13;
463	493	break;
464	494
465	495	case ORC_REG_DI:
466		− if (!state->regs \|\| !state->full_regs) {
467		− orc_warn("missing regs for base reg DI at ip %pB\n",
468		− (void *)state->ip);
	496	+ if (!get_reg(state, offsetof(struct pt_regs, di), &sp)) {
	497	+ orc_warn_current("missing RDI value at %pB\n",
	498	+ (void *)state->ip);
469	499	goto err;
470	500	}
471		− sp = state->regs->di;
472	501	break;
473	502
474	503	case ORC_REG_DX:
475		− if (!state->regs \|\| !state->full_regs) {
476		− orc_warn("missing regs for base reg DX at ip %pB\n",
477		− (void *)state->ip);
	504	+ if (!get_reg(state, offsetof(struct pt_regs, dx), &sp)) {
	505	+ orc_warn_current("missing DX value at %pB\n",
	506	+ (void *)state->ip);
478	507	goto err;
479	508	}
480		− sp = state->regs->dx;
481	509	break;
482	510
483	511	default:
484		− orc_warn("unknown SP base reg %d for ip %pB\n",
	512	+ orc_warn("unknown SP base reg %d at %pB\n",
485	513	orc->sp_reg, (void *)state->ip);
486	514	goto err;
487	515	}
@@ -504,44 +532,48 @@ bool unwind_next_frame(struct unwind_state *state) · 16 unmodified lines
504	532
505	533	state->sp = sp;
506	534	state->regs = NULL;
	535	+ state->prev_regs = NULL;
507	536	state->signal = false;
508	537	break;
509	538
510	539	case ORC_TYPE_REGS:
511	540	if (!deref_stack_regs(state, sp, &state->ip, &state->sp)) {
512		− orc_warn("can't dereference registers at %p for ip %pB\n",
513		− (void )sp, (void )orig_ip);
	541	+ orc_warn_current("can't access registers at %pB\n",
	542	+ (void *)orig_ip);
514	543	goto err;
515	544	}
516	545
517	546	state->regs = (struct pt_regs *)sp;
	547	+ state->prev_regs = NULL;
518	548	state->full_regs = true;
519	549	state->signal = true;
520	550	break;
521	551
522	552	case ORC_TYPE_REGS_IRET:
523	553	if (!deref_stack_iret_regs(state, sp, &state->ip, &state->sp)) {
524		− orc_warn("can't dereference iret registers at %p for ip %pB\n",
525		− (void )sp, (void )orig_ip);
	554	+ orc_warn_current("can't access iret registers at %pB\n",
	555	+ (void *)orig_ip);
526	556	goto err;
527	557	}
528	558
	559	+ if (state->full_regs)
	560	+ state->prev_regs = state->regs;
529	561	state->regs = (void *)sp - IRET_FRAME_OFFSET;
530	562	state->full_regs = false;
531	563	state->signal = true;
532	564	break;
533	565
534	566	default:
535		− orc_warn("unknown .orc_unwind entry type %d for ip %pB\n",
	567	+ orc_warn("unknown .orc_unwind entry type %d at %pB\n",
536	568	orc->type, (void *)orig_ip);
537		− break;
	569	+ goto err;
538	570	}
539	571
540	572	/* Find BP: */
541	573	switch (orc->bp_reg) {
542	574	case ORC_REG_UNDEFINED:
543		− if (state->regs && state->full_regs)
544		− state->bp = state->regs->bp;
	575	+ if (get_reg(state, offsetof(struct pt_regs, bp), &tmp))
	576	+ state->bp = tmp;
545	577	break;
546	578
547	579	case ORC_REG_PREV_SP:
@@ -564,8 +596,8 @@ bool unwind_next_frame(struct unwind_state *state) · 16 unmodified lines
564	596	if (state->stack_info.type == prev_type &&
565	597	on_stack(&state->stack_info, (void *)state->sp, sizeof(long)) &&
566	598	state->sp <= prev_sp) {
567		− orc_warn("stack going in the wrong direction? ip=%pB\n",
568		− (void *)orig_ip);
	599	+ orc_warn_current("stack going in the wrong direction? at %pB\n",
	600	+ (void *)orig_ip);
569	601	goto err;
570	602	}
571	603
@@ -585,6 +617,9 @@ EXPORT_SYMBOL_GPL(unwind_next_frame); · 13 unmodified lines
585	617	void __unwind_start(struct unwind_state state, struct task_struct task,
586	618	struct pt_regs regs, unsigned long first_frame)
587	619	{
	620	+ if (!orc_init)
	621	+ goto done;
	622	+
588	623	memset(state, 0, sizeof(*state));
589	624	state->task = task;
590	625
@@ -651,7 +686,7 @@ void __unwind_start(struct unwind_state state, struct task_struct task, · 60 unmodified lines
651	686	/* Otherwise, skip ahead to the user-specified starting frame: */
652	687	while (!unwind_done(state) &&
653	688	(!on_stack(&state->stack_info, first_frame, sizeof(long)) \|\|
654		− state->sp <= (unsigned long)first_frame))
	689	+ state->sp < (unsigned long)first_frame))
655	690	unwind_next_frame(state);
656	691
657	692	return;

arch/x86/kvm/Makefile+4−0

@@ -3,6 +3,10 @@ · 2 unmodified lines
3	3	ccflags-y += -Iarch/x86/kvm
4	4	ccflags-$(CONFIG_KVM_WERROR) += -Werror
5	5
	6	+ifeq ($(CONFIG_FRAME_POINTER),y)
	7	+OBJECT_FILES_NON_STANDARD_vmenter.o := y
	8	+endif
	9	+
6	10	KVM := ../../../virt/kvm
7	11
8	12	kvm-y += $(KVM)/kvm_main.o $(KVM)/coalesced_mmio.o \

arch/x86/kvm/ioapic.c+5−5

@@ -225,12 +225,12 @@ static int ioapic_set_irq(struct kvm_ioapic *ioapic, unsigned int irq, · 224 unmodified lines
225	225	}
226	226
227	227	/*
228		− * AMD SVM AVIC accelerate EOI write and do not trap,
229		− * in-kernel IOAPIC will not be able to receive the EOI.
230		− * In this case, we do lazy update of the pending EOI when
231		− * trying to set IOAPIC irq.
	228	+ * AMD SVM AVIC accelerate EOI write iff the interrupt is edge
	229	+ * triggered, in which case the in-kernel IOAPIC will not be able
	230	+ * to receive the EOI. In this case, we do a lazy update of the
	231	+ * pending EOI when trying to set IOAPIC irq.
232	232	*/
233		− if (kvm_apicv_activated(ioapic->kvm))
	233	+ if (edge && kvm_apicv_activated(ioapic->kvm))
234	234	ioapic_lazy_update_eoi(ioapic, irq);
235	235
236	236	/*

arch/x86/kvm/svm/sev.c+6−2

@@ -12,6 +12,7 @@ · 11 unmodified lines
12	12	#include <linux/kernel.h>
13	13	#include <linux/highmem.h>
14	14	#include <linux/psp-sev.h>
	15	+#include <linux/pagemap.h>
15	16	#include <linux/swap.h>
16	17
17	18	#include "x86.h"
@@ -344,7 +345,7 @@ static struct page *sev_pin_memory(struct kvm kvm, unsigned long uaddr, · 326 unmodified lines
344	345	return NULL;
345	346
346	347	/* Pin the user virtual address. */
347		− npinned = get_user_pages_fast(uaddr, npages, FOLL_WRITE, pages);
	348	+ npinned = get_user_pages_fast(uaddr, npages, write ? FOLL_WRITE : 0, pages);
348	349	if (npinned != npages) {
349	350	pr_err("SEV: Failure locking %lu pages.\n", npages);
350	351	goto err;
@@ -1117,7 +1118,7 @@ int __init sev_hardware_setup(void) · 766 unmodified lines
1117	1118	/* Maximum number of encrypted guests supported simultaneously */
1118	1119	max_sev_asid = cpuid_ecx(0x8000001F);
1119	1120
1120		− if (!max_sev_asid)
	1121	+ if (!svm_sev_enabled())
1121	1122	return 1;
1122	1123
1123	1124	/* Minimum ASID value that should be used for SEV guest */
@@ -1156,6 +1157,9 @@ err: · 32 unmodified lines
1156	1157
1157	1158	void sev_hardware_teardown(void)
1158	1159	{
	1160	+ if (!svm_sev_enabled())
	1161	+ return;
	1162	+
1159	1163	bitmap_free(sev_asid_bitmap);
1160	1164	bitmap_free(sev_reclaim_asid_bitmap);
1161	1165

arch/x86/kvm/svm/svm.c+3−9

@@ -1752,6 +1752,8 @@ static int db_interception(struct vcpu_svm *svm) · 1751 unmodified lines
1752	1752	if (svm->vcpu.guest_debug &
1753	1753	(KVM_GUESTDBG_SINGLESTEP \| KVM_GUESTDBG_USE_HW_BP)) {
1754	1754	kvm_run->exit_reason = KVM_EXIT_DEBUG;
	1755	+ kvm_run->debug.arch.dr6 = svm->vmcb->save.dr6;
	1756	+ kvm_run->debug.arch.dr7 = svm->vmcb->save.dr7;
1755	1757	kvm_run->debug.arch.pc =
1756	1758	svm->vmcb->save.cs.base + svm->vmcb->save.rip;
1757	1759	kvm_run->debug.arch.exception = DB_VECTOR;
@@ -3276,7 +3278,7 @@ static void svm_cancel_injection(struct kvm_vcpu *vcpu) · 1518 unmodified lines
3276	3278	svm_complete_interrupts(svm);
3277	3279	}
3278	3280
3279		−bool __svm_vcpu_run(unsigned long vmcb_pa, unsigned long *regs);
	3281	+void __svm_vcpu_run(unsigned long vmcb_pa, unsigned long *regs);
3280	3282
3281	3283	static void svm_vcpu_run(struct kvm_vcpu *vcpu)
3282	3284	{
@@ -3330,13 +3332,8 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu) · 47 unmodified lines
3330	3332	*/
3331	3333	x86_spec_ctrl_set_guest(svm->spec_ctrl, svm->virt_spec_ctrl);
3332	3334
3333		− local_irq_enable();
3334		−
3335	3335	__svm_vcpu_run(svm->vmcb_pa, (unsigned long *)&svm->vcpu.arch.regs);
3336	3336
3337		− /* Eliminate branch target predictions from guest mode */
3338		− vmexit_fill_RSB();
3339		−
3340	3337	#ifdef CONFIG_X86_64
3341	3338	wrmsrl(MSR_GS_BASE, svm->host.gs_base);
3342	3339	#else
@@ -3366,8 +3363,6 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu) · 23 unmodified lines
3366	3363
3367	3364	reload_tss(vcpu);
3368	3365
3369		− local_irq_disable();
3370		−
3371	3366	x86_spec_ctrl_restore_host(svm->spec_ctrl, svm->virt_spec_ctrl);
3372	3367
3373	3368	vcpu->arch.cr2 = svm->vmcb->save.cr2;
@@ -3411,7 +3406,6 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu) · 37 unmodified lines
3411	3406
3412	3407	mark_all_clean(svm->vmcb);
3413	3408	}
3414		−STACK_FRAME_NON_STANDARD(svm_vcpu_run);
3415	3409
3416	3410	static void svm_load_mmu_pgd(struct kvm_vcpu *vcpu, unsigned long root)
3417	3411	{

arch/x86/kvm/svm/vmenter.S+9−1

@@ -3,6 +3,7 @@ · 2 unmodified lines
3	3	#include <asm/asm.h>
4	4	#include <asm/bitsperlong.h>
5	5	#include <asm/kvm_vcpu_regs.h>
	6	+#include <asm/nospec-branch.h>
6	7
7	8	#define WORD_SIZE (BITS_PER_LONG / 8)
8	9
@@ -35,7 +36,6 @@ · 26 unmodified lines
35	36	*/
36	37	SYM_FUNC_START(__svm_vcpu_run)
37	38	push %_ASM_BP
38		− mov %_ASM_SP, %_ASM_BP
39	39	#ifdef CONFIG_X86_64
40	40	push %r15
41	41	push %r14
@@ -78,6 +78,7 @@ SYM_FUNC_START(__svm_vcpu_run) · 36 unmodified lines
78	78	pop %_ASM_AX
79	79
80	80	/* Enter guest mode */
	81	+ sti
81	82	1: vmload %_ASM_AX
82	83	jmp 3f
83	84	2: cmpb $0, kvm_rebooting
@@ -99,6 +100,13 @@ SYM_FUNC_START(__svm_vcpu_run) · 15 unmodified lines
99	100	ud2
100	101	_ASM_EXTABLE(5b, 6b)
101	102	7:
	103	+ cli
	104	+
	105	+#ifdef CONFIG_RETPOLINE
	106	+ /* IMPORTANT: Stuff the RSB immediately after VM-Exit, before RET! */
	107	+ FILL_RETURN_BUFFER %_ASM_AX, RSB_CLEAR_LOOPS, X86_FEATURE_RETPOLINE
	108	+#endif
	109	+
102	110	/* "POP" @regs to RAX. */
103	111	pop %_ASM_AX
104	112

arch/x86/kvm/vmx/nested.c+20−3

@@ -5165,7 +5165,7 @@ static int handle_invept(struct kvm_vcpu *vcpu) · 5164 unmodified lines
5165	5165	*/
5166	5166	break;
5167	5167	default:
5168		− BUG_ON(1);
	5168	+ BUG();
5169	5169	break;
5170	5170	}
5171	5171
@@ -5533,8 +5533,25 @@ static bool nested_vmx_exit_handled_vmcs_access(struct kvm_vcpu *vcpu, · 361 unmodified lines
5533	5533	return 1 & (b >> (field & 7));
5534	5534	}
5535	5535
	5536	+static bool nested_vmx_exit_handled_mtf(struct vmcs12 *vmcs12)
	5537	+{
	5538	+ u32 entry_intr_info = vmcs12->vm_entry_intr_info_field;
	5539	+
	5540	+ if (nested_cpu_has_mtf(vmcs12))
	5541	+ return true;
	5542	+
	5543	+ /*
	5544	+ * An MTF VM-exit may be injected into the guest by setting the
	5545	+ * interruption-type to 7 (other event) and the vector field to 0. Such
	5546	+ * is the case regardless of the 'monitor trap flag' VM-execution
	5547	+ * control.
	5548	+ */
	5549	+ return entry_intr_info == (INTR_INFO_VALID_MASK
	5550	+ \| INTR_TYPE_OTHER_EVENT);
	5551	+}
	5552	+
5536	5553	/*
5537		− * Return 1 if we should exit from L2 to L1 to handle an exit, or 0 if we
	5554	+ * Return true if we should exit from L2 to L1 to handle an exit, or false if we
5538	5555	* should handle it ourselves in L0 (and then continue L2). Only call this
5539	5556	* when in is_guest_mode (L2).
5540	5557	*/
@@ -5633,7 +5650,7 @@ bool nested_vmx_exit_reflected(struct kvm_vcpu *vcpu, u32 exit_reason) · 92 unmodified lines
5633	5650	case EXIT_REASON_MWAIT_INSTRUCTION:
5634	5651	return nested_cpu_has(vmcs12, CPU_BASED_MWAIT_EXITING);
5635	5652	case EXIT_REASON_MONITOR_TRAP_FLAG:
5636		− return nested_cpu_has_mtf(vmcs12);
	5653	+ return nested_vmx_exit_handled_mtf(vmcs12);
5637	5654	case EXIT_REASON_MONITOR_INSTRUCTION:
5638	5655	return nested_cpu_has(vmcs12, CPU_BASED_MONITOR_EXITING);
5639	5656	case EXIT_REASON_PAUSE_INSTRUCTION:

arch/x86/kvm/vmx/vmenter.S+3−0

@@ -82,6 +82,9 @@ SYM_FUNC_START(vmx_vmexit) · 81 unmodified lines
82	82	/* IMPORTANT: Stuff the RSB immediately after VM-Exit, before RET! */
83	83	FILL_RETURN_BUFFER %_ASM_AX, RSB_CLEAR_LOOPS, X86_FEATURE_RETPOLINE
84	84
	85	+ /* Clear RFLAGS.CF and RFLAGS.ZF to preserve VM-Exit, i.e. !VM-Fail. */
	86	+ or $1, %_ASM_AX
	87	+
85	88	pop %_ASM_AX
86	89	.Lvmexit_skip_rsb:
87	90	#endif

arch/x86/kvm/vmx/vmx.c+1−1

@@ -4572,7 +4572,7 @@ static int handle_rmode_exception(struct kvm_vcpu *vcpu, · 4571 unmodified lines
4572	4572	*/
4573	4573	static void kvm_machine_check(void)
4574	4574	{
4575		−#if defined(CONFIG_X86_MCE) && defined(CONFIG_X86_64)
	4575	+#if defined(CONFIG_X86_MCE)
4576	4576	struct pt_regs regs = {
4577	4577	.cs = 3, /* Fake ring 3 no matter what the guest ran on */
4578	4578	.flags = X86_EFLAGS_IF,

arch/x86/kvm/x86.c+25−17

@@ -926,19 +926,6 @@ EXPORT_SYMBOL_GPL(kvm_set_xcr); · 925 unmodified lines
926	926	__reserved_bits; \
927	927	})
928	928
929		−static u64 kvm_host_cr4_reserved_bits(struct cpuinfo_x86 *c)
930		−{
931		− u64 reserved_bits = __cr4_reserved_bits(cpu_has, c);
932		−
933		− if (kvm_cpu_cap_has(X86_FEATURE_LA57))
934		− reserved_bits &= ~X86_CR4_LA57;
935		−
936		− if (kvm_cpu_cap_has(X86_FEATURE_UMIP))
937		− reserved_bits &= ~X86_CR4_UMIP;
938		−
939		− return reserved_bits;
940		−}
941		−
942	929	static int kvm_valid_cr4(struct kvm_vcpu *vcpu, unsigned long cr4)
943	930	{
944	931	if (cr4 & cr4_reserved_bits)
@@ -3060,6 +3047,17 @@ int kvm_get_msr_common(struct kvm_vcpu vcpu, struct msr_data msr_info) · 2115 unmodified lines
3060	3047	case MSR_IA32_PERF_CTL:
3061	3048	case MSR_AMD64_DC_CFG:
3062	3049	case MSR_F15H_EX_CFG:
	3050	+ /*
	3051	+ * Intel Sandy Bridge CPUs must support the RAPL (running average power
	3052	+ * limit) MSRs. Just return 0, as we do not want to expose the host
	3053	+ * data here. Do not conditionalize this on CPUID, as KVM does not do
	3054	+ * so for existing CPU-specific MSRs.
	3055	+ */
	3056	+ case MSR_RAPL_POWER_UNIT:
	3057	+ case MSR_PP0_ENERGY_STATUS: /* Power plane 0 (core) */
	3058	+ case MSR_PP1_ENERGY_STATUS: /* Power plane 1 (graphics uncore) */
	3059	+ case MSR_PKG_ENERGY_STATUS: /* Total package */
	3060	+ case MSR_DRAM_ENERGY_STATUS: /* DRAM controller */
3063	3061	msr_info->data = 0;
3064	3062	break;
3065	3063	case MSR_F15H_PERF_CTL0 ... MSR_F15H_PERF_CTR5:
@@ -3374,6 +3372,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) · 308 unmodified lines
3374	3372	case KVM_CAP_GET_MSR_FEATURES:
3375	3373	case KVM_CAP_MSR_PLATFORM_INFO:
3376	3374	case KVM_CAP_EXCEPTION_PAYLOAD:
	3375	+ case KVM_CAP_SET_GUEST_DEBUG:
3377	3376	r = 1;
3378	3377	break;
3379	3378	case KVM_CAP_SYNC_REGS:
@@ -5049,10 +5048,13 @@ set_identity_unlock: · 1669 unmodified lines
5049	5048	r = -EFAULT;
5050	5049	if (copy_from_user(&u.ps, argp, sizeof(u.ps)))
5051	5050	goto out;
	5051	+ mutex_lock(&kvm->lock);
5052	5052	r = -ENXIO;
5053	5053	if (!kvm->arch.vpit)
5054		− goto out;
	5054	+ goto set_pit_out;
5055	5055	r = kvm_vm_ioctl_set_pit(kvm, &u.ps);
	5056	+set_pit_out:
	5057	+ mutex_unlock(&kvm->lock);
5056	5058	break;
5057	5059	}
5058	5060	case KVM_GET_PIT2: {
@@ -5072,10 +5074,13 @@ set_identity_unlock: · 13 unmodified lines
5072	5074	r = -EFAULT;
5073	5075	if (copy_from_user(&u.ps2, argp, sizeof(u.ps2)))
5074	5076	goto out;
	5077	+ mutex_lock(&kvm->lock);
5075	5078	r = -ENXIO;
5076	5079	if (!kvm->arch.vpit)
5077		− goto out;
	5080	+ goto set_pit2_out;
5078	5081	r = kvm_vm_ioctl_set_pit2(kvm, &u.ps2);
	5082	+set_pit2_out:
	5083	+ mutex_unlock(&kvm->lock);
5079	5084	break;
5080	5085	}
5081	5086	case KVM_REINJECT_CONTROL: {
@@ -9658,7 +9663,9 @@ int kvm_arch_hardware_setup(void *opaque) · 4576 unmodified lines
9658	9663	if (!kvm_cpu_cap_has(X86_FEATURE_XSAVES))
9659	9664	supported_xss = 0;
9660	9665
9661		− cr4_reserved_bits = kvm_host_cr4_reserved_bits(&boot_cpu_data);
	9666	+#define __kvm_cpu_cap_has(UNUSED_, f) kvm_cpu_cap_has(f)
	9667	+ cr4_reserved_bits = __cr4_reserved_bits(__kvm_cpu_cap_has, UNUSED_);
	9668	+#undef __kvm_cpu_cap_has
9662	9669
9663	9670	if (kvm_has_tsc_control) {
9664	9671	/*
@@ -9690,7 +9697,8 @@ int kvm_arch_check_processor_compat(void *opaque) · 25 unmodified lines
9690	9697
9691	9698	WARN_ON(!irqs_disabled());
9692	9699
9693		− if (kvm_host_cr4_reserved_bits(c) != cr4_reserved_bits)
	9700	+ if (__cr4_reserved_bits(cpu_has, c) !=
	9701	+ __cr4_reserved_bits(cpu_has, &boot_cpu_data))
9694	9702	return -EIO;
9695	9703
9696	9704	return ops->check_processor_compatibility();

arch/x86/mm/pat/set_memory.c+8−4

@@ -43,7 +43,8 @@ struct cpa_data { · 42 unmodified lines
43	43	unsigned long pfn;
44	44	unsigned int flags;
45	45	unsigned int force_split : 1,
46		− force_static_prot : 1;
	46	+ force_static_prot : 1,
	47	+ force_flush_all : 1;
47	48	struct page **pages;
48	49	};
49	50
@@ -355,10 +356,10 @@ static void cpa_flush(struct cpa_data *data, int cache) · 305 unmodified lines
355	356	return;
356	357	}
357	358
358		− if (cpa->numpages <= tlb_single_page_flush_ceiling)
359		− on_each_cpu(__cpa_flush_tlb, cpa, 1);
360		− else
	359	+ if (cpa->force_flush_all \|\| cpa->numpages > tlb_single_page_flush_ceiling)
361	360	flush_tlb_all();
	361	+ else
	362	+ on_each_cpu(__cpa_flush_tlb, cpa, 1);
362	363
363	364	if (!cache)
364	365	return;
@@ -1598,6 +1599,8 @@ static int cpa_process_alias(struct cpa_data *cpa) · 1233 unmodified lines
1598	1599	alias_cpa.flags &= ~(CPA_PAGES_ARRAY \| CPA_ARRAY);
1599	1600	alias_cpa.curpage = 0;
1600	1601
	1602	+ cpa->force_flush_all = 1;
	1603	+
1601	1604	ret = __change_page_attr_set_clr(&alias_cpa, 0);
1602	1605	if (ret)
1603	1606	return ret;
@@ -1618,6 +1621,7 @@ static int cpa_process_alias(struct cpa_data *cpa) · 14 unmodified lines
1618	1621	alias_cpa.flags &= ~(CPA_PAGES_ARRAY \| CPA_ARRAY);
1619	1622	alias_cpa.curpage = 0;
1620	1623
	1624	+ cpa->force_flush_all = 1;
1621	1625	/*
1622	1626	* The high mapping range is imprecise, so ignore the
1623	1627	* return value.

arch/x86/net/bpf_jit_comp.c+15−3

@@ -158,6 +158,19 @@ static bool is_ereg(u32 reg) · 157 unmodified lines
158	158	BIT(BPF_REG_AX));
159	159	}
160	160
	161	+/*
	162	+ * is_ereg_8l() == true if BPF register 'reg' is mapped to access x86-64
	163	+ * lower 8-bit registers dil,sil,bpl,spl,r8b..r15b, which need extra byte
	164	+ * of encoding. al,cl,dl,bl have simpler encoding.
	165	+ */
	166	+static bool is_ereg_8l(u32 reg)
	167	+{
	168	+ return is_ereg(reg) \|\|
	169	+ (1 << reg) & (BIT(BPF_REG_1) \|
	170	+ BIT(BPF_REG_2) \|
	171	+ BIT(BPF_REG_FP));
	172	+}
	173	+
161	174	static bool is_axreg(u32 reg)
162	175	{
163	176	return reg == BPF_REG_0;
@@ -598,9 +611,8 @@ static void emit_stx(u8 **pprog, u32 size, u32 dst_reg, u32 src_reg, int off) · 434 unmodified lines
598	611	switch (size) {
599	612	case BPF_B:
600	613	/* Emit 'mov byte ptr [rax + off], al' */
601		− if (is_ereg(dst_reg) \|\| is_ereg(src_reg) \|\|
602		− /* We have to add extra byte for x86 SIL, DIL regs */
603		− src_reg == BPF_REG_1 \|\| src_reg == BPF_REG_2)
	614	+ if (is_ereg(dst_reg) \|\| is_ereg_8l(src_reg))
	615	+ /* Add extra byte for eregs or SIL,DIL,BPL in src_reg */
604	616	EMIT2(add_2mod(0x40, dst_reg, src_reg), 0x88);
605	617	else
606	618	EMIT1(0x88);

arch/x86/net/bpf_jit_comp32.c+22−6

@@ -1847,14 +1847,16 @@ static int do_jit(struct bpf_prog bpf_prog, int addrs, u8 *image, · 1846 unmodified lines
1847	1847	case BPF_B:
1848	1848	case BPF_H:
1849	1849	case BPF_W:
1850		− if (!bpf_prog->aux->verifier_zext)
	1850	+ if (bpf_prog->aux->verifier_zext)
1851	1851	break;
1852	1852	if (dstk) {
1853	1853	EMIT3(0xC7, add_1reg(0x40, IA32_EBP),
1854	1854	STACK_VAR(dst_hi));
1855	1855	EMIT(0x0, 4);
1856	1856	} else {
1857		− EMIT3(0xC7, add_1reg(0xC0, dst_hi), 0);
	1857	+ /* xor dst_hi,dst_hi */
	1858	+ EMIT2(0x33,
	1859	+ add_2reg(0xC0, dst_hi, dst_hi));
1858	1860	}
1859	1861	break;
1860	1862	case BPF_DW:
@@ -2013,8 +2015,8 @@ static int do_jit(struct bpf_prog bpf_prog, int addrs, u8 *image, · 152 unmodified lines
2013	2015	case BPF_JMP \| BPF_JSET \| BPF_X:
2014	2016	case BPF_JMP32 \| BPF_JSET \| BPF_X: {
2015	2017	bool is_jmp64 = BPF_CLASS(insn->code) == BPF_JMP;
2016		− u8 dreg_lo = dstk ? IA32_EAX : dst_lo;
2017		− u8 dreg_hi = dstk ? IA32_EDX : dst_hi;
	2018	+ u8 dreg_lo = IA32_EAX;
	2019	+ u8 dreg_hi = IA32_EDX;
2018	2020	u8 sreg_lo = sstk ? IA32_ECX : src_lo;
2019	2021	u8 sreg_hi = sstk ? IA32_EBX : src_hi;
2020	2022
@@ -2026,6 +2028,13 @@ static int do_jit(struct bpf_prog bpf_prog, int addrs, u8 *image, · 5 unmodified lines
2026	2028	add_2reg(0x40, IA32_EBP,
2027	2029	IA32_EDX),
2028	2030	STACK_VAR(dst_hi));
	2031	+ } else {
	2032	+ /* mov dreg_lo,dst_lo */
	2033	+ EMIT2(0x89, add_2reg(0xC0, dreg_lo, dst_lo));
	2034	+ if (is_jmp64)
	2035	+ /* mov dreg_hi,dst_hi */
	2036	+ EMIT2(0x89,
	2037	+ add_2reg(0xC0, dreg_hi, dst_hi));
2029	2038	}
2030	2039
2031	2040	if (sstk) {
@@ -2050,8 +2059,8 @@ static int do_jit(struct bpf_prog bpf_prog, int addrs, u8 *image, · 18 unmodified lines
2050	2059	case BPF_JMP \| BPF_JSET \| BPF_K:
2051	2060	case BPF_JMP32 \| BPF_JSET \| BPF_K: {
2052	2061	bool is_jmp64 = BPF_CLASS(insn->code) == BPF_JMP;
2053		− u8 dreg_lo = dstk ? IA32_EAX : dst_lo;
2054		− u8 dreg_hi = dstk ? IA32_EDX : dst_hi;
	2062	+ u8 dreg_lo = IA32_EAX;
	2063	+ u8 dreg_hi = IA32_EDX;
2055	2064	u8 sreg_lo = IA32_ECX;
2056	2065	u8 sreg_hi = IA32_EBX;
2057	2066	u32 hi;
@@ -2064,6 +2073,13 @@ static int do_jit(struct bpf_prog bpf_prog, int addrs, u8 *image, · 6 unmodified lines
2064	2073	add_2reg(0x40, IA32_EBP,
2065	2074	IA32_EDX),
2066	2075	STACK_VAR(dst_hi));
	2076	+ } else {
	2077	+ /* mov dreg_lo,dst_lo */
	2078	+ EMIT2(0x89, add_2reg(0xC0, dreg_lo, dst_lo));
	2079	+ if (is_jmp64)
	2080	+ /* mov dreg_hi,dst_hi */
	2081	+ EMIT2(0x89,
	2082	+ add_2reg(0xC0, dreg_hi, dst_hi));
2067	2083	}
2068	2084
2069	2085	/* mov ecx,imm32 */

arch/xtensa/include/asm/vermagic.hrenamed from arch/xtensa/include/asm/module.h+6−9

@@ -1,8 +1,4 @@
1	1	/*
2		− * include/asm-xtensa/module.h
3		− *
4		− * This file contains the module code specific to the Xtensa architecture.
5		− *
6	2	* This file is subject to the terms and conditions of the GNU General Public
7	3	* License. See the file "COPYING" in the main directory of this archive
8	4	* for more details.
@@ -10,11 +6,12 @@ · 1 unmodified line
10	6	* Copyright (C) 2001 - 2005 Tensilica Inc.
11	7	*/
12	8
13		−#ifndef _XTENSA_MODULE_H
14		−#define _XTENSA_MODULE_H
	9	+#ifndef _ASM_VERMAGIC_H
	10	+#define _ASM_VERMAGIC_H
15	11
16		−#define MODULE_ARCH_VERMAGIC "xtensa-" __stringify(XCHAL_CORE_ID) " "
	12	+#include <linux/stringify.h>
	13	+#include <variant/core.h>
17	14
18		−#include <asm-generic/module.h>
	15	+#define MODULE_ARCH_VERMAGIC "xtensa-" __stringify(XCHAL_CORE_ID) " "
19	16
20		−#endif /* _XTENSA_MODULE_H */
	17	+#endif /* _ASM_VERMAGIC_H */

block/bfq-iosched.c+4−2

@@ -123,6 +123,7 @@ · 122 unmodified lines
123	123	#include <linux/ioprio.h>
124	124	#include <linux/sbitmap.h>
125	125	#include <linux/delay.h>
	126	+#include <linux/backing-dev.h>
126	127
127	128	#include "blk.h"
128	129	#include "blk-mq.h"
@@ -4976,8 +4977,9 @@ bfq_set_next_ioprio_data(struct bfq_queue bfqq, struct bfq_io_cq bic) · 4847 unmodified lines
4976	4977	ioprio_class = IOPRIO_PRIO_CLASS(bic->ioprio);
4977	4978	switch (ioprio_class) {
4978	4979	default:
4979		− dev_err(bfqq->bfqd->queue->backing_dev_info->dev,
4980		− "bfq: bad prio class %d\n", ioprio_class);
	4980	+ pr_err("bdi %s: bfq: bad prio class %d\n",
	4981	+ bdi_dev_name(bfqq->bfqd->queue->backing_dev_info),
	4982	+ ioprio_class);
4981	4983	/* fall through */
4982	4984	case IOPRIO_CLASS_NONE:
4983	4985	/*

block/blk-cgroup.c+1−1

@@ -496,7 +496,7 @@ const char blkg_dev_name(struct blkcg_gq blkg) · 495 unmodified lines
496	496	{
497	497	/* some drivers (floppy) instantiate a queue w/o disk registered */
498	498	if (blkg->q->backing_dev_info->dev)
499		− return dev_name(blkg->q->backing_dev_info->dev);
	499	+ return bdi_dev_name(blkg->q->backing_dev_info);
500	500	return NULL;
501	501	}
502	502

block/blk-iocost.c+73−48

@@ -466,7 +466,7 @@ struct ioc_gq { · 465 unmodified lines
466	466	*/
467	467	atomic64_t vtime;
468	468	atomic64_t done_vtime;
469		− atomic64_t abs_vdebt;
	469	+ u64 abs_vdebt;
470	470	u64 last_vtime;
471	471
472	472	/*
@@ -1142,7 +1142,7 @@ static void iocg_kick_waitq(struct ioc_gq iocg, struct ioc_now now) · 669 unmodified lines
1142	1142	struct iocg_wake_ctx ctx = { .iocg = iocg };
1143	1143	u64 margin_ns = (u64)(ioc->period_us *
1144	1144	WAITQ_TIMER_MARGIN_PCT / 100) * NSEC_PER_USEC;
1145		− u64 abs_vdebt, vdebt, vshortage, expires, oexpires;
	1145	+ u64 vdebt, vshortage, expires, oexpires;
1146	1146	s64 vbudget;
1147	1147	u32 hw_inuse;
1148	1148
@@ -1152,18 +1152,15 @@ static void iocg_kick_waitq(struct ioc_gq iocg, struct ioc_now now) · 3 unmodified lines
1152	1152	vbudget = now->vnow - atomic64_read(&iocg->vtime);
1153	1153
1154	1154	/* pay off debt */
1155		− abs_vdebt = atomic64_read(&iocg->abs_vdebt);
1156		− vdebt = abs_cost_to_cost(abs_vdebt, hw_inuse);
	1155	+ vdebt = abs_cost_to_cost(iocg->abs_vdebt, hw_inuse);
1157	1156	if (vdebt && vbudget > 0) {
1158	1157	u64 delta = min_t(u64, vbudget, vdebt);
1159	1158	u64 abs_delta = min(cost_to_abs_cost(delta, hw_inuse),
1160		− abs_vdebt);
	1159	+ iocg->abs_vdebt);
1161	1160
1162	1161	atomic64_add(delta, &iocg->vtime);
1163	1162	atomic64_add(delta, &iocg->done_vtime);
1164		− atomic64_sub(abs_delta, &iocg->abs_vdebt);
1165		− if (WARN_ON_ONCE(atomic64_read(&iocg->abs_vdebt) < 0))
1166		− atomic64_set(&iocg->abs_vdebt, 0);
	1163	+ iocg->abs_vdebt -= abs_delta;
1167	1164	}
1168	1165
1169	1166	/*
@@ -1219,12 +1216,18 @@ static bool iocg_kick_delay(struct ioc_gq iocg, struct ioc_now now, u64 cost) · 49 unmodified lines
1219	1216	u64 expires, oexpires;
1220	1217	u32 hw_inuse;
1221	1218
	1219	+ lockdep_assert_held(&iocg->waitq.lock);
	1220	+
1222	1221	/* debt-adjust vtime */
1223	1222	current_hweight(iocg, NULL, &hw_inuse);
1224		− vtime += abs_cost_to_cost(atomic64_read(&iocg->abs_vdebt), hw_inuse);
	1223	+ vtime += abs_cost_to_cost(iocg->abs_vdebt, hw_inuse);
1225	1224
1226		− /* clear or maintain depending on the overage */
1227		− if (time_before_eq64(vtime, now->vnow)) {
	1225	+ /*
	1226	+ * Clear or maintain depending on the overage. Non-zero vdebt is what
	1227	+ * guarantees that @iocg is online and future iocg_kick_delay() will
	1228	+ * clear use_delay. Don't leave it on when there's no vdebt.
	1229	+ */
	1230	+ if (!iocg->abs_vdebt \|\| time_before_eq64(vtime, now->vnow)) {
1228	1231	blkcg_clear_delay(blkg);
1229	1232	return false;
1230	1233	}
@@ -1258,9 +1261,12 @@ static enum hrtimer_restart iocg_delay_timer_fn(struct hrtimer *timer) · 27 unmodified lines
1258	1261	{
1259	1262	struct ioc_gq *iocg = container_of(timer, struct ioc_gq, delay_timer);
1260	1263	struct ioc_now now;
	1264	+ unsigned long flags;
1261	1265
	1266	+ spin_lock_irqsave(&iocg->waitq.lock, flags);
1262	1267	ioc_now(iocg->ioc, &now);
1263	1268	iocg_kick_delay(iocg, &now, 0);
	1269	+ spin_unlock_irqrestore(&iocg->waitq.lock, flags);
1264	1270
1265	1271	return HRTIMER_NORESTART;
1266	1272	}
@@ -1368,14 +1374,13 @@ static void ioc_timer_fn(struct timer_list *timer) · 101 unmodified lines
1368	1374	* should have woken up in the last period and expire idle iocgs.
1369	1375	*/
1370	1376	list_for_each_entry_safe(iocg, tiocg, &ioc->active_iocgs, active_list) {
1371		− if (!waitqueue_active(&iocg->waitq) &&
1372		− !atomic64_read(&iocg->abs_vdebt) && !iocg_is_idle(iocg))
	1377	+ if (!waitqueue_active(&iocg->waitq) && iocg->abs_vdebt &&
	1378	+ !iocg_is_idle(iocg))
1373	1379	continue;
1374	1380
1375	1381	spin_lock(&iocg->waitq.lock);
1376	1382
1377		− if (waitqueue_active(&iocg->waitq) \|\|
1378		− atomic64_read(&iocg->abs_vdebt)) {
	1383	+ if (waitqueue_active(&iocg->waitq) \|\| iocg->abs_vdebt) {
1379	1384	/* might be oversleeping vtime / hweight changes, kick */
1380	1385	iocg_kick_waitq(iocg, &now);
1381	1386	iocg_kick_delay(iocg, &now, 0);
@@ -1591,7 +1596,7 @@ skip_surplus_transfers: · 209 unmodified lines
1591	1596	vrate_min, vrate_max);
1592	1597	}
1593	1598
1594		− trace_iocost_ioc_vrate_adj(ioc, vrate, &missed_ppm, rq_wait_pct,
	1599	+ trace_iocost_ioc_vrate_adj(ioc, vrate, missed_ppm, rq_wait_pct,
1595	1600	nr_lagging, nr_shortages,
1596	1601	nr_surpluses);
1597	1602
@@ -1600,7 +1605,7 @@ skip_surplus_transfers: · 2 unmodified lines
1600	1605	ioc->period_us * vrate * INUSE_MARGIN_PCT, 100);
1601	1606	} else if (ioc->busy_level != prev_busy_level \|\| nr_lagging) {
1602	1607	trace_iocost_ioc_vrate_adj(ioc, atomic64_read(&ioc->vtime_rate),
1603		− &missed_ppm, rq_wait_pct, nr_lagging,
	1608	+ missed_ppm, rq_wait_pct, nr_lagging,
1604	1609	nr_shortages, nr_surpluses);
1605	1610	}
1606	1611
@@ -1718,28 +1723,49 @@ static void ioc_rqos_throttle(struct rq_qos rqos, struct bio bio) · 111 unmodified lines
1718	1723	* tests are racy but the races aren't systemic - we only miss once
1719	1724	* in a while which is fine.
1720	1725	*/
1721		− if (!waitqueue_active(&iocg->waitq) &&
1722		− !atomic64_read(&iocg->abs_vdebt) &&
	1726	+ if (!waitqueue_active(&iocg->waitq) && !iocg->abs_vdebt &&
1723	1727	time_before_eq64(vtime + cost, now.vnow)) {
1724	1728	iocg_commit_bio(iocg, bio, cost);
1725	1729	return;
1726	1730	}
1727	1731
1728	1732	/*
1729		− * We're over budget. If @bio has to be issued regardless,
1730		− * remember the abs_cost instead of advancing vtime.
1731		− * iocg_kick_waitq() will pay off the debt before waking more IOs.
	1733	+ * We activated above but w/o any synchronization. Deactivation is
	1734	+ * synchronized with waitq.lock and we won't get deactivated as long
	1735	+ * as we're waiting or has debt, so we're good if we're activated
	1736	+ * here. In the unlikely case that we aren't, just issue the IO.
	1737	+ */
	1738	+ spin_lock_irq(&iocg->waitq.lock);
	1739	+
	1740	+ if (unlikely(list_empty(&iocg->active_list))) {
	1741	+ spin_unlock_irq(&iocg->waitq.lock);
	1742	+ iocg_commit_bio(iocg, bio, cost);
	1743	+ return;
	1744	+ }
	1745	+
	1746	+ /*
	1747	+ * We're over budget. If @bio has to be issued regardless, remember
	1748	+ * the abs_cost instead of advancing vtime. iocg_kick_waitq() will pay
	1749	+ * off the debt before waking more IOs.
	1750	+ *
1732	1751	* This way, the debt is continuously paid off each period with the
1733		− * actual budget available to the cgroup. If we just wound vtime,
1734		− * we would incorrectly use the current hw_inuse for the entire
1735		− * amount which, for example, can lead to the cgroup staying
1736		− * blocked for a long time even with substantially raised hw_inuse.
	1752	+ * actual budget available to the cgroup. If we just wound vtime, we
	1753	+ * would incorrectly use the current hw_inuse for the entire amount
	1754	+ * which, for example, can lead to the cgroup staying blocked for a
	1755	+ * long time even with substantially raised hw_inuse.
	1756	+ *
	1757	+ * An iocg with vdebt should stay online so that the timer can keep
	1758	+ * deducting its vdebt and [de]activate use_delay mechanism
	1759	+ * accordingly. We don't want to race against the timer trying to
	1760	+ * clear them and leave @iocg inactive w/ dangling use_delay heavily
	1761	+ * penalizing the cgroup and its descendants.
1737	1762	*/
1738	1763	if (bio_issue_as_root_blkg(bio) \|\| fatal_signal_pending(current)) {
1739		− atomic64_add(abs_cost, &iocg->abs_vdebt);
	1764	+ iocg->abs_vdebt += abs_cost;
1740	1765	if (iocg_kick_delay(iocg, &now, cost))
1741	1766	blkcg_schedule_throttle(rqos->q,
1742	1767	(bio->bi_opf & REQ_SWAP) == REQ_SWAP);
	1768	+ spin_unlock_irq(&iocg->waitq.lock);
1743	1769	return;
1744	1770	}
1745	1771
@@ -1756,20 +1782,6 @@ static void ioc_rqos_throttle(struct rq_qos rqos, struct bio bio) · 10 unmodified lines
1756	1782	* All waiters are on iocg->waitq and the wait states are
1757	1783	* synchronized using waitq.lock.
1758	1784	*/
1759		− spin_lock_irq(&iocg->waitq.lock);
1760		−
1761		− /*
1762		− * We activated above but w/o any synchronization. Deactivation is
1763		− * synchronized with waitq.lock and we won't get deactivated as
1764		− * long as we're waiting, so we're good if we're activated here.
1765		− * In the unlikely case that we are deactivated, just issue the IO.
1766		− */
1767		− if (unlikely(list_empty(&iocg->active_list))) {
1768		− spin_unlock_irq(&iocg->waitq.lock);
1769		− iocg_commit_bio(iocg, bio, cost);
1770		− return;
1771		− }
1772		−
1773	1785	init_waitqueue_func_entry(&wait.wait, iocg_wake_fn);
1774	1786	wait.wait.private = current;
1775	1787	wait.bio = bio;
@@ -1801,6 +1813,7 @@ static void ioc_rqos_merge(struct rq_qos rqos, struct request rq, · 25 unmodified lines
1801	1813	struct ioc_now now;
1802	1814	u32 hw_inuse;
1803	1815	u64 abs_cost, cost;
	1816	+ unsigned long flags;
1804	1817
1805	1818	/* bypass if disabled or for root cgroup */
1806	1819	if (!ioc->enabled \|\| !iocg->level)
@@ -1820,15 +1833,28 @@ static void ioc_rqos_merge(struct rq_qos rqos, struct request rq, · 13 unmodified lines
1820	1833	iocg->cursor = bio_end;
1821	1834
1822	1835	/*
1823		− * Charge if there's enough vtime budget and the existing request
1824		− * has cost assigned. Otherwise, account it as debt. See debt
1825		− * handling in ioc_rqos_throttle() for details.
	1836	+ * Charge if there's enough vtime budget and the existing request has
	1837	+ * cost assigned.
1826	1838	*/
1827	1839	if (rq->bio && rq->bio->bi_iocost_cost &&
1828		− time_before_eq64(atomic64_read(&iocg->vtime) + cost, now.vnow))
	1840	+ time_before_eq64(atomic64_read(&iocg->vtime) + cost, now.vnow)) {
1829	1841	iocg_commit_bio(iocg, bio, cost);
1830		− else
1831		− atomic64_add(abs_cost, &iocg->abs_vdebt);
	1842	+ return;
	1843	+ }
	1844	+
	1845	+ /*
	1846	+ * Otherwise, account it as debt if @iocg is online, which it should
	1847	+ * be for the vast majority of cases. See debt handling in
	1848	+ * ioc_rqos_throttle() for details.
	1849	+ */
	1850	+ spin_lock_irqsave(&iocg->waitq.lock, flags);
	1851	+ if (likely(!list_empty(&iocg->active_list))) {
	1852	+ iocg->abs_vdebt += abs_cost;
	1853	+ iocg_kick_delay(iocg, &now, cost);
	1854	+ } else {
	1855	+ iocg_commit_bio(iocg, bio, cost);
	1856	+ }
	1857	+ spin_unlock_irqrestore(&iocg->waitq.lock, flags);
1832	1858	}
1833	1859
1834	1860	static void ioc_rqos_done_bio(struct rq_qos rqos, struct bio bio)
@@ -1998,7 +2024,6 @@ static void ioc_pd_init(struct blkg_policy_data *pd) · 163 unmodified lines
1998	2024	iocg->ioc = ioc;
1999	2025	atomic64_set(&iocg->vtime, now.vnow);
2000	2026	atomic64_set(&iocg->done_vtime, now.vnow);
2001		− atomic64_set(&iocg->abs_vdebt, 0);
2002	2027	atomic64_set(&iocg->active_period, atomic64_read(&ioc->cur_period));
2003	2028	INIT_LIST_HEAD(&iocg->active_list);
2004	2029	iocg->hweight_active = HWEIGHT_WHOLE;

block/partitions/core.c+1−1

@@ -496,7 +496,7 @@ int blk_drop_partitions(struct gendisk disk, struct block_device bdev) · 495 unmodified lines
496	496
497	497	if (!disk_part_scan_enabled(disk))
498	498	return 0;
499		− if (bdev->bd_part_count \|\| bdev->bd_openers > 1)
	499	+ if (bdev->bd_part_count)
500	500	return -EBUSY;
501	501	res = invalidate_partition(disk, 0);
502	502	if (res)

crypto/lrw.c+3−3

@@ -287,7 +287,7 @@ static void exit_tfm(struct crypto_skcipher *tfm) · 286 unmodified lines
287	287	crypto_free_skcipher(ctx->child);
288	288	}
289	289
290		−static void free(struct skcipher_instance *inst)
	290	+static void free_inst(struct skcipher_instance *inst)
291	291	{
292	292	crypto_drop_skcipher(skcipher_instance_ctx(inst));
293	293	kfree(inst);
@@ -400,12 +400,12 @@ static int create(struct crypto_template tmpl, struct rtattr *tb) · 106 unmodified lines
400	400	inst->alg.encrypt = encrypt;
401	401	inst->alg.decrypt = decrypt;
402	402
403		− inst->free = free;
	403	+ inst->free = free_inst;
404	404
405	405	err = skcipher_register_instance(tmpl, inst);
406	406	if (err) {
407	407	err_free_inst:
408		− free(inst);
	408	+ free_inst(inst);
409	409	}
410	410	return err;
411	411	}

crypto/xts.c+3−3

@@ -322,7 +322,7 @@ static void exit_tfm(struct crypto_skcipher *tfm) · 321 unmodified lines
322	322	crypto_free_cipher(ctx->tweak);
323	323	}
324	324
325		−static void free(struct skcipher_instance *inst)
	325	+static void free_inst(struct skcipher_instance *inst)
326	326	{
327	327	crypto_drop_skcipher(skcipher_instance_ctx(inst));
328	328	kfree(inst);
@@ -434,12 +434,12 @@ static int create(struct crypto_template tmpl, struct rtattr *tb) · 105 unmodified lines
434	434	inst->alg.encrypt = encrypt;
435	435	inst->alg.decrypt = decrypt;
436	436
437		− inst->free = free;
	437	+ inst->free = free_inst;
438	438
439	439	err = skcipher_register_instance(tmpl, inst);
440	440	if (err) {
441	441	err_free_inst:
442		− free(inst);
	442	+ free_inst(inst);
443	443	}
444	444	return err;
445	445	}

drivers/acpi/button.c+0−12

@@ -87,18 +87,6 @@ static const struct dmi_system_id dmi_lid_quirks[] = { · 86 unmodified lines
87	87	},
88	88	.driver_data = (void *)(long)ACPI_BUTTON_LID_INIT_DISABLED,
89	89	},
90		− {
91		− /*
92		− * Asus T200TA, _LID keeps reporting closed after every second
93		− * openening of the lid. Causing immediate re-suspend after
94		− * opening every other open. Using LID_INIT_OPEN fixes this.
95		− */
96		− .matches = {
97		− DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
98		− DMI_MATCH(DMI_PRODUCT_NAME, "T200TA"),
99		− },
100		− .driver_data = (void *)(long)ACPI_BUTTON_LID_INIT_OPEN,
101		− },
102	90	{
103	91	/* GP-electronic T701, _LID method points to a floating GPIO */
104	92	.matches = {

drivers/acpi/device_pm.c+2−2

@@ -273,13 +273,13 @@ int acpi_device_set_power(struct acpi_device *device, int state) · 272 unmodified lines
273	273	end:
274	274	if (result) {
275	275	dev_warn(&device->dev, "Failed to change power state to %s\n",
276		− acpi_power_state_string(state));
	276	+ acpi_power_state_string(target_state));
277	277	} else {
278	278	device->power.state = target_state;
279	279	ACPI_DEBUG_PRINT((ACPI_DB_INFO,
280	280	"Device [%s] transitioned to %s\n",
281	281	device->pnp.bus_id,
282		− acpi_power_state_string(state)));
	282	+ acpi_power_state_string(target_state)));
283	283	}
284	284
285	285	return result;

drivers/acpi/pci_link.c+2−2

@@ -322,10 +322,10 @@ static int acpi_pci_link_set(struct acpi_pci_link *link, int irq) · 321 unmodified lines
322	322	resource->res.data.extended_irq.polarity =
323	323	link->irq.polarity;
324	324	if (link->irq.triggering == ACPI_EDGE_SENSITIVE)
325		− resource->res.data.irq.shareable =
	325	+ resource->res.data.extended_irq.shareable =
326	326	ACPI_EXCLUSIVE;
327	327	else
328		− resource->res.data.irq.shareable = ACPI_SHARED;
	328	+ resource->res.data.extended_irq.shareable = ACPI_SHARED;
329	329	resource->res.data.extended_irq.interrupt_count = 1;
330	330	resource->res.data.extended_irq.interrupts[0] = irq;
331	331	/* ignore resource_source, it's optional */

drivers/amba/bus.c+1−0

@@ -645,6 +645,7 @@ static void amba_device_initialize(struct amba_device dev, const char name) · 644 unmodified lines
645	645	dev->dev.release = amba_device_release;
646	646	dev->dev.bus = &amba_bustype;
647	647	dev->dev.dma_mask = &dev->dev.coherent_dma_mask;
	648	+ dev->dev.dma_parms = &dev->dma_parms;
648	649	dev->res.name = dev_name(&dev->dev);
649	650	}
650	651

drivers/ata/sata_inic162x.c+1−1

@@ -145,7 +145,7 @@ enum { · 144 unmodified lines
145	145
146	146	/* PORT_IDMA_CTL bits */
147	147	IDMA_CTL_RST_ATA = (1 << 2), /* hardreset ATA bus */
148		− IDMA_CTL_RST_IDMA = (1 << 5), /* reset IDMA machinary */
	148	+ IDMA_CTL_RST_IDMA = (1 << 5), /* reset IDMA machinery */
149	149	IDMA_CTL_GO = (1 << 7), /* IDMA mode go */
150	150	IDMA_CTL_ATA_NIEN = (1 << 8), /* ATA IRQ disable */
151	151

drivers/base/component.c+5−3

@@ -256,7 +256,8 @@ static int try_to_bring_up_master(struct master *master, · 255 unmodified lines
256	256	ret = master->ops->bind(master->dev);
257	257	if (ret < 0) {
258	258	devres_release_group(master->dev, NULL);
259		− dev_info(master->dev, "master bind failed: %d\n", ret);
	259	+ if (ret != -EPROBE_DEFER)
	260	+ dev_info(master->dev, "master bind failed: %d\n", ret);
260	261	return ret;
261	262	}
262	263
@@ -611,8 +612,9 @@ static int component_bind(struct component component, struct master master, · 348 unmodified lines
611	612	devres_release_group(component->dev, NULL);
612	613	devres_release_group(master->dev, NULL);
613	614
614		− dev_err(master->dev, "failed to bind %s (ops %ps): %d\n",
615		− dev_name(component->dev), component->ops, ret);
	615	+ if (ret != -EPROBE_DEFER)
	616	+ dev_err(master->dev, "failed to bind %s (ops %ps): %d\n",
	617	+ dev_name(component->dev), component->ops, ret);
616	618	}
617	619
618	620	return ret;

drivers/base/core.c+6−1

@@ -2370,6 +2370,11 @@ u32 fw_devlink_get_flags(void) · 2369 unmodified lines
2370	2370	return fw_devlink_flags;
2371	2371	}
2372	2372
	2373	+static bool fw_devlink_is_permissive(void)
	2374	+{
	2375	+ return fw_devlink_flags == DL_FLAG_SYNC_STATE_ONLY;
	2376	+}
	2377	+
2373	2378	/**
2374	2379	* device_add - add device to device hierarchy.
2375	2380	* @dev: device.
@@ -2524,7 +2529,7 @@ int device_add(struct device *dev) · 148 unmodified lines
2524	2529	if (fw_devlink_flags && is_fwnode_dev &&
2525	2530	fwnode_has_op(dev->fwnode, add_links)) {
2526	2531	fw_ret = fwnode_call_int_op(dev->fwnode, add_links, dev);
2527		− if (fw_ret == -ENODEV)
	2532	+ if (fw_ret == -ENODEV && !fw_devlink_is_permissive())
2528	2533	device_link_wait_for_mandatory_supplier(dev);
2529	2534	else if (fw_ret)
2530	2535	device_link_wait_for_optional_supplier(dev);

drivers/base/dd.c+8−12

@@ -224,17 +224,9 @@ static int deferred_devs_show(struct seq_file s, void data) · 223 unmodified lines
224	224	}
225	225	DEFINE_SHOW_ATTRIBUTE(deferred_devs);
226	226
227		−#ifdef CONFIG_MODULES
228		−/*
229		− * In the case of modules, set the default probe timeout to
230		− * 30 seconds to give userland some time to load needed modules
231		− */
232		−int driver_deferred_probe_timeout = 30;
233		−#else
234		−/* In the case of !modules, no probe timeout needed */
235		−int driver_deferred_probe_timeout = -1;
236		−#endif
	227	+int driver_deferred_probe_timeout;
237	228	EXPORT_SYMBOL_GPL(driver_deferred_probe_timeout);
	229	+static DECLARE_WAIT_QUEUE_HEAD(probe_timeout_waitqueue);
238	230
239	231	static int __init deferred_probe_timeout_setup(char *str)
240	232	{
@@ -266,8 +258,8 @@ int driver_deferred_probe_check_state(struct device *dev) · 25 unmodified lines
266	258	return -ENODEV;
267	259	}
268	260
269		− if (!driver_deferred_probe_timeout) {
270		− dev_WARN(dev, "deferred probe timeout, ignoring dependency");
	261	+ if (!driver_deferred_probe_timeout && initcalls_done) {
	262	+ dev_warn(dev, "deferred probe timeout, ignoring dependency");
271	263	return -ETIMEDOUT;
272	264	}
273	265
@@ -284,6 +276,7 @@ static void deferred_probe_timeout_work_func(struct work_struct *work) · 10 unmodified lines
284	276
285	277	list_for_each_entry_safe(private, p, &deferred_probe_pending_list, deferred_probe)
286	278	dev_info(private->device, "deferred probe pending");
	279	+ wake_up(&probe_timeout_waitqueue);
287	280	}
288	281	static DECLARE_DELAYED_WORK(deferred_probe_timeout_work, deferred_probe_timeout_work_func);
289	282
@@ -658,6 +651,9 @@ int driver_probe_done(void) · 368 unmodified lines
658	651	*/
659	652	void wait_for_device_probe(void)
660	653	{
	654	+ /* wait for probe timeout */
	655	+ wait_event(probe_timeout_waitqueue, !driver_deferred_probe_timeout);
	656	+
661	657	/* wait for the deferred probe workqueue to finish */
662	658	flush_work(&deferred_probe_work);
663	659

drivers/base/firmware_loader/fallback_table.c+0−1

@@ -45,5 +45,4 @@ struct ctl_table firmware_config_table[] = { · 44 unmodified lines
45	45	},
46	46	{ }
47	47	};
48		−EXPORT_SYMBOL_GPL(firmware_config_table);
49	48	#endif

drivers/base/platform.c+2−0

@@ -380,6 +380,8 @@ struct platform_object { · 379 unmodified lines
380	380	*/
381	381	static void setup_pdev_dma_masks(struct platform_device *pdev)
382	382	{
	383	+ pdev->dev.dma_parms = &pdev->dma_parms;
	384	+
383	385	if (!pdev->dev.coherent_dma_mask)
384	386	pdev->dev.coherent_dma_mask = DMA_BIT_MASK(32);
385	387	if (!pdev->dev.dma_mask) {

drivers/base/power/main.c+1−1

@@ -730,7 +730,7 @@ static bool dpm_async_fn(struct device *dev, async_func_t func) · 729 unmodified lines
730	730
731	731	if (is_async(dev)) {
732	732	get_device(dev);
733		− async_schedule(func, dev);
	733	+ async_schedule_dev(func, dev);
734	734	return true;
735	735	}
736	736

drivers/block/null_blk.h+19−10

@@ -85,26 +85,35 @@ struct nullb { · 84 unmodified lines
85	85	char disk_name[DISK_NAME_LEN];
86	86	};
87	87
	88	+blk_status_t null_process_cmd(struct nullb_cmd *cmd,
	89	+ enum req_opf op, sector_t sector,
	90	+ unsigned int nr_sectors);
	91	+
88	92	#ifdef CONFIG_BLK_DEV_ZONED
89		−int null_zone_init(struct nullb_device *dev);
90		−void null_zone_exit(struct nullb_device *dev);
	93	+int null_init_zoned_dev(struct nullb_device dev, struct request_queue q);
	94	+int null_register_zoned_dev(struct nullb *nullb);
	95	+void null_free_zoned_dev(struct nullb_device *dev);
91	96	int null_report_zones(struct gendisk *disk, sector_t sector,
92	97	unsigned int nr_zones, report_zones_cb cb, void *data);
93		−blk_status_t null_handle_zoned(struct nullb_cmd *cmd,
94		− enum req_opf op, sector_t sector,
95		− sector_t nr_sectors);
	98	+blk_status_t null_process_zoned_cmd(struct nullb_cmd *cmd,
	99	+ enum req_opf op, sector_t sector,
	100	+ sector_t nr_sectors);
96	101	size_t null_zone_valid_read_len(struct nullb *nullb,
97	102	sector_t sector, unsigned int len);
98	103	#else
99		−static inline int null_zone_init(struct nullb_device *dev)
	104	+static inline int null_init_zoned_dev(struct nullb_device *dev,
	105	+ struct request_queue *q)
100	106	{
101	107	pr_err("CONFIG_BLK_DEV_ZONED not enabled\n");
102	108	return -EINVAL;
103	109	}
104		−static inline void null_zone_exit(struct nullb_device *dev) {}
105		−static inline blk_status_t null_handle_zoned(struct nullb_cmd *cmd,
106		− enum req_opf op, sector_t sector,
107		− sector_t nr_sectors)
	110	+static inline int null_register_zoned_dev(struct nullb *nullb)
	111	+{
	112	+ return -ENODEV;
	113	+}
	114	+static inline void null_free_zoned_dev(struct nullb_device *dev) {}
	115	+static inline blk_status_t null_process_zoned_cmd(struct nullb_cmd *cmd,
	116	+ enum req_opf op, sector_t sector, sector_t nr_sectors)
108	117	{
109	118	return BLK_STS_NOTSUPP;
110	119	}

drivers/block/null_blk_main.c+31−31

@@ -580,7 +580,7 @@ static void null_free_dev(struct nullb_device *dev) · 579 unmodified lines
580	580	if (!dev)
581	581	return;
582	582
583		− null_zone_exit(dev);
	583	+ null_free_zoned_dev(dev);
584	584	badblocks_exit(&dev->badblocks);
585	585	kfree(dev);
586	586	}
@@ -1276,6 +1276,25 @@ static inline void nullb_complete_cmd(struct nullb_cmd *cmd) · 689 unmodified lines
1276	1276	}
1277	1277	}
1278	1278
	1279	+blk_status_t null_process_cmd(struct nullb_cmd *cmd,
	1280	+ enum req_opf op, sector_t sector,
	1281	+ unsigned int nr_sectors)
	1282	+{
	1283	+ struct nullb_device *dev = cmd->nq->dev;
	1284	+ blk_status_t ret;
	1285	+
	1286	+ if (dev->badblocks.shift != -1) {
	1287	+ ret = null_handle_badblocks(cmd, sector, nr_sectors);
	1288	+ if (ret != BLK_STS_OK)
	1289	+ return ret;
	1290	+ }
	1291	+
	1292	+ if (dev->memory_backed)
	1293	+ return null_handle_memory_backed(cmd, op);
	1294	+
	1295	+ return BLK_STS_OK;
	1296	+}
	1297	+
1279	1298	static blk_status_t null_handle_cmd(struct nullb_cmd *cmd, sector_t sector,
1280	1299	sector_t nr_sectors, enum req_opf op)
1281	1300	{
@@ -1294,17 +1313,11 @@ static blk_status_t null_handle_cmd(struct nullb_cmd *cmd, sector_t sector, · 12 unmodified lines
1294	1313	goto out;
1295	1314	}
1296	1315
1297		− if (nullb->dev->badblocks.shift != -1) {
1298		− cmd->error = null_handle_badblocks(cmd, sector, nr_sectors);
1299		− if (cmd->error != BLK_STS_OK)
1300		− goto out;
1301		− }
1302		−
1303		− if (dev->memory_backed)
1304		− cmd->error = null_handle_memory_backed(cmd, op);
1305		−
1306		− if (!cmd->error && dev->zoned)
1307		− cmd->error = null_handle_zoned(cmd, op, sector, nr_sectors);
	1316	+ if (dev->zoned)
	1317	+ cmd->error = null_process_zoned_cmd(cmd, op,
	1318	+ sector, nr_sectors);
	1319	+ else
	1320	+ cmd->error = null_process_cmd(cmd, op, sector, nr_sectors);
1308	1321
1309	1322	out:
1310	1323	nullb_complete_cmd(cmd);
@@ -1605,19 +1618,12 @@ static int null_gendisk_register(struct nullb *nullb) · 294 unmodified lines
1605	1618	disk->queue = nullb->q;
1606	1619	strncpy(disk->disk_name, nullb->disk_name, DISK_NAME_LEN);
1607	1620
1608		−#ifdef CONFIG_BLK_DEV_ZONED
1609	1621	if (nullb->dev->zoned) {
1610		− if (queue_is_mq(nullb->q)) {
1611		− int ret = blk_revalidate_disk_zones(disk);
1612		− if (ret)
1613		− return ret;
1614		− } else {
1615		− blk_queue_chunk_sectors(nullb->q,
1616		− nullb->dev->zone_size_sects);
1617		− nullb->q->nr_zones = blkdev_nr_zones(disk);
1618		− }
	1622	+ int ret = null_register_zoned_dev(nullb);
	1623	+
	1624	+ if (ret)
	1625	+ return ret;
1619	1626	}
1620		−#endif
1621	1627
1622	1628	add_disk(disk);
1623	1629	return 0;
@@ -1773,14 +1779,9 @@ static int null_add_dev(struct nullb_device *dev) · 149 unmodified lines
1773	1779	}
1774	1780
1775	1781	if (dev->zoned) {
1776		− rv = null_zone_init(dev);
	1782	+ rv = null_init_zoned_dev(dev, nullb->q);
1777	1783	if (rv)
1778	1784	goto out_cleanup_blk_queue;
1779		−
1780		− nullb->q->limits.zoned = BLK_ZONED_HM;
1781		− blk_queue_flag_set(QUEUE_FLAG_ZONE_RESETALL, nullb->q);
1782		− blk_queue_required_elevator_features(nullb->q,
1783		− ELEVATOR_F_ZBD_SEQ_WRITE);
1784	1785	}
1785	1786
1786	1787	nullb->q->queuedata = nullb;
@@ -1809,8 +1810,7 @@ static int null_add_dev(struct nullb_device *dev) · 22 unmodified lines
1809	1810
1810	1811	return 0;
1811	1812	out_cleanup_zone:
1812		− if (dev->zoned)
1813		− null_zone_exit(dev);
	1813	+ null_free_zoned_dev(dev);
1814	1814	out_cleanup_blk_queue:
1815	1815	blk_cleanup_queue(nullb->q);
1816	1816	out_cleanup_tags:

drivers/block/null_blk_zoned.c+33−12

@@ -13,7 +13,7 @@ static inline unsigned int null_zone_no(struct nullb_device *dev, sector_t sect) · 12 unmodified lines
13	13	return sect >> ilog2(dev->zone_size_sects);
14	14	}
15	15
16		−int null_zone_init(struct nullb_device *dev)
	16	+int null_init_zoned_dev(struct nullb_device dev, struct request_queue q)
17	17	{
18	18	sector_t dev_size = (sector_t)dev->size * 1024 * 1024;
19	19	sector_t sector = 0;
@@ -61,10 +61,27 @@ int null_zone_init(struct nullb_device *dev) · 41 unmodified lines
61	61	sector += dev->zone_size_sects;
62	62	}
63	63
	64	+ q->limits.zoned = BLK_ZONED_HM;
	65	+ blk_queue_flag_set(QUEUE_FLAG_ZONE_RESETALL, q);
	66	+ blk_queue_required_elevator_features(q, ELEVATOR_F_ZBD_SEQ_WRITE);
	67	+
	68	+ return 0;
	69	+}
	70	+
	71	+int null_register_zoned_dev(struct nullb *nullb)
	72	+{
	73	+ struct request_queue *q = nullb->q;
	74	+
	75	+ if (queue_is_mq(q))
	76	+ return blk_revalidate_disk_zones(nullb->disk);
	77	+
	78	+ blk_queue_chunk_sectors(q, nullb->dev->zone_size_sects);
	79	+ q->nr_zones = blkdev_nr_zones(nullb->disk);
	80	+
64	81	return 0;
65	82	}
66	83
67		−void null_zone_exit(struct nullb_device *dev)
	84	+void null_free_zoned_dev(struct nullb_device *dev)
68	85	{
69	86	kvfree(dev->zones);
70	87	}
@@ -126,11 +143,16 @@ static blk_status_t null_zone_write(struct nullb_cmd *cmd, sector_t sector, · 55 unmodified lines
126	143	struct nullb_device *dev = cmd->nq->dev;
127	144	unsigned int zno = null_zone_no(dev, sector);
128	145	struct blk_zone *zone = &dev->zones[zno];
	146	+ blk_status_t ret;
	147	+
	148	+ trace_nullb_zone_op(cmd, zno, zone->cond);
	149	+
	150	+ if (zone->type == BLK_ZONE_TYPE_CONVENTIONAL)
	151	+ return null_process_cmd(cmd, REQ_OP_WRITE, sector, nr_sectors);
129	152
130	153	switch (zone->cond) {
131	154	case BLK_ZONE_COND_FULL:
132	155	/* Cannot write to a full zone */
133		− cmd->error = BLK_STS_IOERR;
134	156	return BLK_STS_IOERR;
135	157	case BLK_ZONE_COND_EMPTY:
136	158	case BLK_ZONE_COND_IMP_OPEN:
@@ -143,19 +165,18 @@ static blk_status_t null_zone_write(struct nullb_cmd *cmd, sector_t sector, · 6 unmodified lines
143	165	if (zone->cond != BLK_ZONE_COND_EXP_OPEN)
144	166	zone->cond = BLK_ZONE_COND_IMP_OPEN;
145	167
	168	+ ret = null_process_cmd(cmd, REQ_OP_WRITE, sector, nr_sectors);
	169	+ if (ret != BLK_STS_OK)
	170	+ return ret;
	171	+
146	172	zone->wp += nr_sectors;
147	173	if (zone->wp == zone->start + zone->len)
148	174	zone->cond = BLK_ZONE_COND_FULL;
149		− break;
150		− case BLK_ZONE_COND_NOT_WP:
151		− break;
	175	+ return BLK_STS_OK;
152	176	default:
153	177	/* Invalid zone condition */
154	178	return BLK_STS_IOERR;
155	179	}
156		−
157		− trace_nullb_zone_op(cmd, zno, zone->cond);
158		− return BLK_STS_OK;
159	180	}
160	181
161	182	static blk_status_t null_zone_mgmt(struct nullb_cmd *cmd, enum req_opf op,
@@ -216,8 +237,8 @@ static blk_status_t null_zone_mgmt(struct nullb_cmd *cmd, enum req_opf op, · 54 unmodified lines
216	237	return BLK_STS_OK;
217	238	}
218	239
219		−blk_status_t null_handle_zoned(struct nullb_cmd *cmd, enum req_opf op,
220		− sector_t sector, sector_t nr_sectors)
	240	+blk_status_t null_process_zoned_cmd(struct nullb_cmd *cmd, enum req_opf op,
	241	+ sector_t sector, sector_t nr_sectors)
221	242	{
222	243	switch (op) {
223	244	case REQ_OP_WRITE:
@@ -229,6 +250,6 @@ blk_status_t null_handle_zoned(struct nullb_cmd *cmd, enum req_opf op, · 5 unmodified lines
229	250	case REQ_OP_ZONE_FINISH:
230	251	return null_zone_mgmt(cmd, op, sector);
231	252	default:
232		− return BLK_STS_OK;
	253	+ return null_process_cmd(cmd, op, sector, nr_sectors);
233	254	}
234	255	}

drivers/block/virtio_blk.c+79−8

@@ -15,6 +15,7 @@ · 14 unmodified lines
15	15	#include <linux/blk-mq.h>
16	16	#include <linux/blk-mq-virtio.h>
17	17	#include <linux/numa.h>
	18	+#include <uapi/linux/virtio_ring.h>
18	19
19	20	#define PART_BITS 4
20	21	#define VQ_NAME_LEN 16
@@ -32,6 +33,15 @@ struct virtio_blk_vq { · 11 unmodified lines
32	33	} ____cacheline_aligned_in_smp;
33	34
34	35	struct virtio_blk {
	36	+ /*
	37	+ * This mutex must be held by anything that may run after
	38	+ * virtblk_remove() sets vblk->vdev to NULL.
	39	+ *
	40	+ * blk-mq, virtqueue processing, and sysfs attribute code paths are
	41	+ * shut down before vblk->vdev is set to NULL and therefore do not need
	42	+ * to hold this mutex.
	43	+ */
	44	+ struct mutex vdev_mutex;
35	45	struct virtio_device *vdev;
36	46
37	47	/* The disk structure for the kernel. */
@@ -43,6 +53,13 @@ struct virtio_blk { · 5 unmodified lines
43	53	/* Process context for config space updates */
44	54	struct work_struct config_work;
45	55
	56	+ /*
	57	+ * Tracks references from block_device_operations open/release and
	58	+ * virtio_driver probe/remove so this object can be freed once no
	59	+ * longer in use.
	60	+ */
	61	+ refcount_t refs;
	62	+
46	63	/* What host tells us, plus 2 for header & tailer. */
47	64	unsigned int sg_elems;
48	65
@@ -294,10 +311,55 @@ out: · 245 unmodified lines
294	311	return err;
295	312	}
296	313
	314	+static void virtblk_get(struct virtio_blk *vblk)
	315	+{
	316	+ refcount_inc(&vblk->refs);
	317	+}
	318	+
	319	+static void virtblk_put(struct virtio_blk *vblk)
	320	+{
	321	+ if (refcount_dec_and_test(&vblk->refs)) {
	322	+ ida_simple_remove(&vd_index_ida, vblk->index);
	323	+ mutex_destroy(&vblk->vdev_mutex);
	324	+ kfree(vblk);
	325	+ }
	326	+}
	327	+
	328	+static int virtblk_open(struct block_device *bd, fmode_t mode)
	329	+{
	330	+ struct virtio_blk *vblk = bd->bd_disk->private_data;
	331	+ int ret = 0;
	332	+
	333	+ mutex_lock(&vblk->vdev_mutex);
	334	+
	335	+ if (vblk->vdev)
	336	+ virtblk_get(vblk);
	337	+ else
	338	+ ret = -ENXIO;
	339	+
	340	+ mutex_unlock(&vblk->vdev_mutex);
	341	+ return ret;
	342	+}
	343	+
	344	+static void virtblk_release(struct gendisk *disk, fmode_t mode)
	345	+{
	346	+ struct virtio_blk *vblk = disk->private_data;
	347	+
	348	+ virtblk_put(vblk);
	349	+}
	350	+
297	351	/* We provide getgeo only to please some old bootloader/partitioning tools */
298	352	static int virtblk_getgeo(struct block_device bd, struct hd_geometry geo)
299	353	{
300	354	struct virtio_blk *vblk = bd->bd_disk->private_data;
	355	+ int ret = 0;
	356	+
	357	+ mutex_lock(&vblk->vdev_mutex);
	358	+
	359	+ if (!vblk->vdev) {
	360	+ ret = -ENXIO;
	361	+ goto out;
	362	+ }
301	363
302	364	/* see if the host passed in geometry config */
303	365	if (virtio_has_feature(vblk->vdev, VIRTIO_BLK_F_GEOMETRY)) {
@@ -313,11 +375,15 @@ static int virtblk_getgeo(struct block_device bd, struct hd_geometry geo) · 9 unmodified lines
313	375	geo->sectors = 1 << 5;
314	376	geo->cylinders = get_capacity(bd->bd_disk) >> 11;
315	377	}
316		− return 0;
	378	+out:
	379	+ mutex_unlock(&vblk->vdev_mutex);
	380	+ return ret;
317	381	}
318	382
319	383	static const struct block_device_operations virtblk_fops = {
320	384	.owner = THIS_MODULE,
	385	+ .open = virtblk_open,
	386	+ .release = virtblk_release,
321	387	.getgeo = virtblk_getgeo,
322	388	};
323	389
@@ -654,6 +720,10 @@ static int virtblk_probe(struct virtio_device *vdev) · 330 unmodified lines
654	720	goto out_free_index;
655	721	}
656	722
	723	+ /* This reference is dropped in virtblk_remove(). */
	724	+ refcount_set(&vblk->refs, 1);
	725	+ mutex_init(&vblk->vdev_mutex);
	726	+
657	727	vblk->vdev = vdev;
658	728	vblk->sg_elems = sg_elems;
659	729
@@ -819,8 +889,6 @@ out: · 159 unmodified lines
819	889	static void virtblk_remove(struct virtio_device *vdev)
820	890	{
821	891	struct virtio_blk *vblk = vdev->priv;
822		− int index = vblk->index;
823		− int refc;
824	892
825	893	/* Make sure no work handler is accessing the device. */
826	894	flush_work(&vblk->config_work);
@@ -830,18 +898,21 @@ static void virtblk_remove(struct virtio_device *vdev) · 3 unmodified lines
830	898
831	899	blk_mq_free_tag_set(&vblk->tag_set);
832	900
	901	+ mutex_lock(&vblk->vdev_mutex);
	902	+
833	903	/* Stop all the virtqueues. */
834	904	vdev->config->reset(vdev);
835	905
836		− refc = kref_read(&disk_to_dev(vblk->disk)->kobj.kref);
	906	+ /* Virtqueues are stopped, nothing can use vblk->vdev anymore. */
	907	+ vblk->vdev = NULL;
	908	+
837	909	put_disk(vblk->disk);
838	910	vdev->config->del_vqs(vdev);
839	911	kfree(vblk->vqs);
840		− kfree(vblk);
841	912
842		− /* Only free device id if we don't have any users */
843		− if (refc == 1)
844		− ida_simple_remove(&vd_index_ida, index);
	913	+ mutex_unlock(&vblk->vdev_mutex);
	914	+
	915	+ virtblk_put(vblk);
845	916	}
846	917
847	918	#ifdef CONFIG_PM_SLEEP

drivers/bus/mhi/core/init.c+3−4

@@ -812,10 +812,9 @@ int mhi_register_controller(struct mhi_controller *mhi_cntrl, · 811 unmodified lines
812	812	if (!mhi_cntrl)
813	813	return -EINVAL;
814	814
815		− if (!mhi_cntrl->runtime_get \|\| !mhi_cntrl->runtime_put)
816		− return -EINVAL;
817		−
818		− if (!mhi_cntrl->status_cb \|\| !mhi_cntrl->link_status)
	815	+ if (!mhi_cntrl->runtime_get \|\| !mhi_cntrl->runtime_put \|\|
	816	+ !mhi_cntrl->status_cb \|\| !mhi_cntrl->read_reg \|\|
	817	+ !mhi_cntrl->write_reg)
819	818	return -EINVAL;
820	819
821	820	ret = parse_config(mhi_cntrl, config);

drivers/bus/mhi/core/internal.h+0−3

@@ -11,9 +11,6 @@ · 10 unmodified lines
11	11
12	12	extern struct bus_type mhi_bus_type;
13	13
14		−/* MHI MMIO register mapping */
15		−#define PCI_INVALID_READ(val) (val == U32_MAX)
16		−
17	14	#define MHIREGLEN (0x0)
18	15	#define MHIREGLEN_MHIREGLEN_MASK (0xFFFFFFFF)
19	16	#define MHIREGLEN_MHIREGLEN_SHIFT (0)

drivers/bus/mhi/core/main.c+5−13

@@ -18,16 +18,7 @@ · 17 unmodified lines
18	18	int __must_check mhi_read_reg(struct mhi_controller *mhi_cntrl,
19	19	void __iomem base, u32 offset, u32 out)
20	20	{
21		− u32 tmp = readl(base + offset);
22		−
23		− /* If there is any unexpected value, query the link status */
24		− if (PCI_INVALID_READ(tmp) &&
25		− mhi_cntrl->link_status(mhi_cntrl))
26		− return -EIO;
27		−
28		− *out = tmp;
29		−
30		− return 0;
	21	+ return mhi_cntrl->read_reg(mhi_cntrl, base + offset, out);
31	22	}
32	23
33	24	int __must_check mhi_read_reg_field(struct mhi_controller *mhi_cntrl,
@@ -49,7 +40,7 @@ int __must_check mhi_read_reg_field(struct mhi_controller *mhi_cntrl, · 15 unmodified lines
49	40	void mhi_write_reg(struct mhi_controller mhi_cntrl, void __iomem base,
50	41	u32 offset, u32 val)
51	42	{
52		− writel(val, base + offset);
	43	+ mhi_cntrl->write_reg(mhi_cntrl, base + offset, val);
53	44	}
54	45
55	46	void mhi_write_reg_field(struct mhi_controller mhi_cntrl, void __iomem base,
@@ -294,7 +285,7 @@ void mhi_create_devices(struct mhi_controller *mhi_cntrl) · 238 unmodified lines
294	285	!(mhi_chan->ee_mask & BIT(mhi_cntrl->ee)))
295	286	continue;
296	287	mhi_dev = mhi_alloc_device(mhi_cntrl);
297		− if (!mhi_dev)
	288	+ if (IS_ERR(mhi_dev))
298	289	return;
299	290
300	291	mhi_dev->dev_type = MHI_DEVICE_XFER;
@@ -336,7 +327,8 @@ void mhi_create_devices(struct mhi_controller *mhi_cntrl) · 35 unmodified lines
336	327
337	328	/* Channel name is same for both UL and DL */
338	329	mhi_dev->chan_name = mhi_chan->name;
339		− dev_set_name(&mhi_dev->dev, "%04x_%s", mhi_chan->chan,
	330	+ dev_set_name(&mhi_dev->dev, "%s_%s",
	331	+ dev_name(mhi_cntrl->cntrl_dev),
340	332	mhi_dev->chan_name);
341	333
342	334	/* Init wakeup source if available */

drivers/bus/mhi/core/pm.c+5−1

@@ -902,7 +902,11 @@ int mhi_sync_power_up(struct mhi_controller *mhi_cntrl) · 901 unmodified lines
902	902	MHI_PM_IN_ERROR_STATE(mhi_cntrl->pm_state),
903	903	msecs_to_jiffies(mhi_cntrl->timeout_ms));
904	904
905		− return (MHI_IN_MISSION_MODE(mhi_cntrl->ee)) ? 0 : -EIO;
	905	+ ret = (MHI_IN_MISSION_MODE(mhi_cntrl->ee)) ? 0 : -ETIMEDOUT;
	906	+ if (ret)
	907	+ mhi_power_down(mhi_cntrl, false);
	908	+
	909	+ return ret;
906	910	}
907	911	EXPORT_SYMBOL(mhi_sync_power_up);
908	912

drivers/char/hw_random/virtio-rng.c+1−0

@@ -11,6 +11,7 @@ · 10 unmodified lines
11	11	#include <linux/virtio.h>
12	12	#include <linux/virtio_rng.h>
13	13	#include <linux/module.h>
	14	+#include <linux/slab.h>
14	15
15	16	static DEFINE_IDA(rng_index_ida);
16	17

drivers/char/tpm/tpm-interface.c+1−1

@@ -323,7 +323,7 @@ int tpm_pcr_extend(struct tpm_chip *chip, u32 pcr_idx, · 322 unmodified lines
323	323
324	324	for (i = 0; i < chip->nr_allocated_banks; i++) {
325	325	if (digests[i].alg_id != chip->allocated_banks[i].alg_id) {
326		− rc = EINVAL;
	326	+ rc = -EINVAL;
327	327	goto out;
328	328	}
329	329	}

drivers/char/tpm/tpm2-cmd.c+1−0

@@ -681,6 +681,7 @@ out: · 680 unmodified lines
681	681	rc = -ENODEV;
682	682	return rc;
683	683	}
	684	+EXPORT_SYMBOL_GPL(tpm2_get_cc_attrs_tbl);
684	685
685	686	/**
686	687	* tpm2_startup - turn on the TPM

drivers/char/tpm/tpm_ibmvtpm.c+73−63

@@ -1,6 +1,6 @@
1	1	// SPDX-License-Identifier: GPL-2.0-only
2	2	/*
3		− * Copyright (C) 2012 IBM Corporation
	3	+ * Copyright (C) 2012-2020 IBM Corporation
4	4	*
5	5	* Author: Ashley Lai <ashleydlai@gmail.com>
6	6	*
@@ -134,6 +134,64 @@ static int tpm_ibmvtpm_recv(struct tpm_chip chip, u8 buf, size_t count) · 127 unmodified lines
134	134	return len;
135	135	}
136	136
	137	+/**
	138	+ * ibmvtpm_crq_send_init - Send a CRQ initialize message
	139	+ * @ibmvtpm: vtpm device struct
	140	+ *
	141	+ * Return:
	142	+ * 0 on success.
	143	+ * Non-zero on failure.
	144	+ */
	145	+static int ibmvtpm_crq_send_init(struct ibmvtpm_dev *ibmvtpm)
	146	+{
	147	+ int rc;
	148	+
	149	+ rc = ibmvtpm_send_crq_word(ibmvtpm->vdev, INIT_CRQ_CMD);
	150	+ if (rc != H_SUCCESS)
	151	+ dev_err(ibmvtpm->dev,
	152	+ "%s failed rc=%d\n", __func__, rc);
	153	+
	154	+ return rc;
	155	+}
	156	+
	157	+/**
	158	+ * tpm_ibmvtpm_resume - Resume from suspend
	159	+ *
	160	+ * @dev: device struct
	161	+ *
	162	+ * Return: Always 0.
	163	+ */
	164	+static int tpm_ibmvtpm_resume(struct device *dev)
	165	+{
	166	+ struct tpm_chip *chip = dev_get_drvdata(dev);
	167	+ struct ibmvtpm_dev *ibmvtpm = dev_get_drvdata(&chip->dev);
	168	+ int rc = 0;
	169	+
	170	+ do {
	171	+ if (rc)
	172	+ msleep(100);
	173	+ rc = plpar_hcall_norets(H_ENABLE_CRQ,
	174	+ ibmvtpm->vdev->unit_address);
	175	+ } while (rc == H_IN_PROGRESS \|\| rc == H_BUSY \|\| H_IS_LONG_BUSY(rc));
	176	+
	177	+ if (rc) {
	178	+ dev_err(dev, "Error enabling ibmvtpm rc=%d\n", rc);
	179	+ return rc;
	180	+ }
	181	+
	182	+ rc = vio_enable_interrupts(ibmvtpm->vdev);
	183	+ if (rc) {
	184	+ dev_err(dev, "Error vio_enable_interrupts rc=%d\n", rc);
	185	+ return rc;
	186	+ }
	187	+
	188	+ rc = ibmvtpm_crq_send_init(ibmvtpm);
	189	+ if (rc)
	190	+ dev_err(dev, "Error send_init rc=%d\n", rc);
	191	+
	192	+ return rc;
	193	+}
	194	+
137	195	/**
138	196	* tpm_ibmvtpm_send() - Send a TPM command
139	197	* @chip: tpm chip struct
@@ -147,6 +205,7 @@ static int tpm_ibmvtpm_recv(struct tpm_chip chip, u8 buf, size_t count) · 7 unmodified lines
147	205	static int tpm_ibmvtpm_send(struct tpm_chip chip, u8 buf, size_t count)
148	206	{
149	207	struct ibmvtpm_dev *ibmvtpm = dev_get_drvdata(&chip->dev);
	208	+ bool retry = true;
150	209	int rc, sig;
151	210
152	211	if (!ibmvtpm->rtce_buf) {
@@ -180,18 +239,27 @@ static int tpm_ibmvtpm_send(struct tpm_chip chip, u8 buf, size_t count) · 27 unmodified lines
180	239	*/
181	240	ibmvtpm->tpm_processing_cmd = true;
182	241
	242	+again:
183	243	rc = ibmvtpm_send_crq(ibmvtpm->vdev,
184	244	IBMVTPM_VALID_CMD, VTPM_TPM_COMMAND,
185	245	count, ibmvtpm->rtce_dma_handle);
186	246	if (rc != H_SUCCESS) {
	247	+ /*
	248	+ * H_CLOSED can be returned after LPM resume. Call
	249	+ * tpm_ibmvtpm_resume() to re-enable the CRQ then retry
	250	+ * ibmvtpm_send_crq() once before failing.
	251	+ */
	252	+ if (rc == H_CLOSED && retry) {
	253	+ tpm_ibmvtpm_resume(ibmvtpm->dev);
	254	+ retry = false;
	255	+ goto again;
	256	+ }
187	257	dev_err(ibmvtpm->dev, "tpm_ibmvtpm_send failed rc=%d\n", rc);
188		− rc = 0;
189	258	ibmvtpm->tpm_processing_cmd = false;
190		− } else
191		− rc = 0;
	259	+ }
192	260
193	261	spin_unlock(&ibmvtpm->rtce_lock);
194		− return rc;
	262	+ return 0;
195	263	}
196	264
197	265	static void tpm_ibmvtpm_cancel(struct tpm_chip *chip)
@@ -269,26 +337,6 @@ static int ibmvtpm_crq_send_init_complete(struct ibmvtpm_dev *ibmvtpm) · 71 unmodified lines
269	337	return rc;
270	338	}
271	339
272		−/**
273		− * ibmvtpm_crq_send_init - Send a CRQ initialize message
274		− * @ibmvtpm: vtpm device struct
275		− *
276		− * Return:
277		− * 0 on success.
278		− * Non-zero on failure.
279		− */
280		−static int ibmvtpm_crq_send_init(struct ibmvtpm_dev *ibmvtpm)
281		−{
282		− int rc;
283		−
284		− rc = ibmvtpm_send_crq_word(ibmvtpm->vdev, INIT_CRQ_CMD);
285		− if (rc != H_SUCCESS)
286		− dev_err(ibmvtpm->dev,
287		− "ibmvtpm_crq_send_init failed rc=%d\n", rc);
288		−
289		− return rc;
290		−}
291		−
292	340	/**
293	341	* tpm_ibmvtpm_remove - ibm vtpm remove entry point
294	342	* @vdev: vio device struct
@@ -401,44 +449,6 @@ static int ibmvtpm_reset_crq(struct ibmvtpm_dev *ibmvtpm) · 106 unmodified lines
401	449	ibmvtpm->crq_dma_handle, CRQ_RES_BUF_SIZE);
402	450	}
403	451
404		−/**
405		− * tpm_ibmvtpm_resume - Resume from suspend
406		− *
407		− * @dev: device struct
408		− *
409		− * Return: Always 0.
410		− */
411		−static int tpm_ibmvtpm_resume(struct device *dev)
412		−{
413		− struct tpm_chip *chip = dev_get_drvdata(dev);
414		− struct ibmvtpm_dev *ibmvtpm = dev_get_drvdata(&chip->dev);
415		− int rc = 0;
416		−
417		− do {
418		− if (rc)
419		− msleep(100);
420		− rc = plpar_hcall_norets(H_ENABLE_CRQ,
421		− ibmvtpm->vdev->unit_address);
422		− } while (rc == H_IN_PROGRESS \|\| rc == H_BUSY \|\| H_IS_LONG_BUSY(rc));
423		−
424		− if (rc) {
425		− dev_err(dev, "Error enabling ibmvtpm rc=%d\n", rc);
426		− return rc;
427		− }
428		−
429		− rc = vio_enable_interrupts(ibmvtpm->vdev);
430		− if (rc) {
431		− dev_err(dev, "Error vio_enable_interrupts rc=%d\n", rc);
432		− return rc;
433		− }
434		−
435		− rc = ibmvtpm_crq_send_init(ibmvtpm);
436		− if (rc)
437		− dev_err(dev, "Error send_init rc=%d\n", rc);
438		−
439		− return rc;
440		−}
441		−
442	452	static bool tpm_ibmvtpm_req_canceled(struct tpm_chip *chip, u8 status)
443	453	{
444	454	return (status == 0);

drivers/char/tpm/tpm_tis_core.c+7−1

@@ -433,6 +433,9 @@ static void disable_interrupts(struct tpm_chip *chip) · 432 unmodified lines
433	433	u32 intmask;
434	434	int rc;
435	435
	436	+ if (priv->irq == 0)
	437	+ return;
	438	+
436	439	rc = tpm_tis_read32(priv, TPM_INT_ENABLE(priv->locality), &intmask);
437	440	if (rc < 0)
438	441	intmask = 0;
@@ -1062,9 +1065,12 @@ int tpm_tis_core_init(struct device dev, struct tpm_tis_data priv, int irq, · 623 unmodified lines
1062	1065	if (irq) {
1063	1066	tpm_tis_probe_irq_single(chip, intmask, IRQF_SHARED,
1064	1067	irq);
1065		− if (!(chip->flags & TPM_CHIP_FLAG_IRQ))
	1068	+ if (!(chip->flags & TPM_CHIP_FLAG_IRQ)) {
1066	1069	dev_err(&chip->dev, FW_BUG
1067	1070	"TPM interrupt not working, polling instead\n");
	1071	+
	1072	+ disable_interrupts(chip);
	1073	+ }
1068	1074	} else {
1069	1075	tpm_tis_probe_irq(chip, intmask);
1070	1076	}

drivers/counter/104-quad-8.c+160−34

@@ -44,6 +44,7 @@ MODULE_PARM_DESC(base, "ACCES 104-QUAD-8 base addresses"); · 43 unmodified lines
44	44	* @base: base port address of the IIO device
45	45	*/
46	46	struct quad8_iio {
	47	+ struct mutex lock;
47	48	struct counter_device counter;
48	49	unsigned int fck_prescaler[QUAD8_NUM_COUNTERS];
49	50	unsigned int preset[QUAD8_NUM_COUNTERS];
@@ -123,6 +124,8 @@ static int quad8_read_raw(struct iio_dev *indio_dev, · 73 unmodified lines
123	124	/* Borrow XOR Carry effectively doubles count range */
124	125	*val = (borrow ^ carry) << 24;
125	126
	127	+ mutex_lock(&priv->lock);
	128	+
126	129	/* Reset Byte Pointer; transfer Counter to Output Latch */
127	130	outb(QUAD8_CTR_RLD \| QUAD8_RLD_RESET_BP \| QUAD8_RLD_CNTR_OUT,
128	131	base_offset + 1);
@@ -130,6 +133,8 @@ static int quad8_read_raw(struct iio_dev *indio_dev, · 1 unmodified line
130	133	for (i = 0; i < 3; i++)
131	134	val \|= (unsigned int)inb(base_offset) << (8 i);
132	135
	136	+ mutex_unlock(&priv->lock);
	137	+
133	138	return IIO_VAL_INT;
134	139	case IIO_CHAN_INFO_ENABLE:
135	140	*val = priv->ab_enable[chan->channel];
@@ -160,6 +165,8 @@ static int quad8_write_raw(struct iio_dev *indio_dev, · 24 unmodified lines
160	165	if ((unsigned int)val > 0xFFFFFF)
161	166	return -EINVAL;
162	167
	168	+ mutex_lock(&priv->lock);
	169	+
163	170	/* Reset Byte Pointer */
164	171	outb(QUAD8_CTR_RLD \| QUAD8_RLD_RESET_BP, base_offset + 1);
165	172
@@ -183,12 +190,16 @@ static int quad8_write_raw(struct iio_dev *indio_dev, · 17 unmodified lines
183	190	/* Reset Error flag */
184	191	outb(QUAD8_CTR_RLD \| QUAD8_RLD_RESET_E, base_offset + 1);
185	192
	193	+ mutex_unlock(&priv->lock);
	194	+
186	195	return 0;
187	196	case IIO_CHAN_INFO_ENABLE:
188	197	/* only boolean values accepted */
189	198	if (val < 0 \|\| val > 1)
190	199	return -EINVAL;
191	200
	201	+ mutex_lock(&priv->lock);
	202	+
192	203	priv->ab_enable[chan->channel] = val;
193	204
194	205	ior_cfg = val \| priv->preset_enable[chan->channel] << 1;
@@ -196,11 +207,18 @@ static int quad8_write_raw(struct iio_dev *indio_dev, · 1 unmodified line
196	207	/* Load I/O control configuration */
197	208	outb(QUAD8_CTR_IOR \| ior_cfg, base_offset + 1);
198	209
	210	+ mutex_unlock(&priv->lock);
	211	+
199	212	return 0;
200	213	case IIO_CHAN_INFO_SCALE:
	214	+ mutex_lock(&priv->lock);
	215	+
201	216	/* Quadrature scaling only available in quadrature mode */
202		− if (!priv->quadrature_mode[chan->channel] && (val2 \|\| val != 1))
	217	+ if (!priv->quadrature_mode[chan->channel] &&
	218	+ (val2 \|\| val != 1)) {
	219	+ mutex_unlock(&priv->lock);
203	220	return -EINVAL;
	221	+ }
204	222
205	223	/* Only three gain states (1, 0.5, 0.25) */
206	224	if (val == 1 && !val2)
@@ -214,11 +232,15 @@ static int quad8_write_raw(struct iio_dev *indio_dev, · 7 unmodified lines
214	232	priv->quadrature_scale[chan->channel] = 2;
215	233	break;
216	234	default:
	235	+ mutex_unlock(&priv->lock);
217	236	return -EINVAL;
218	237	}
219		− else
	238	+ else {
	239	+ mutex_unlock(&priv->lock);
220	240	return -EINVAL;
	241	+ }
221	242
	243	+ mutex_unlock(&priv->lock);
222	244	return 0;
223	245	}
224	246
@@ -255,6 +277,8 @@ static ssize_t quad8_write_preset(struct iio_dev *indio_dev, uintptr_t private, · 30 unmodified lines
255	277	if (preset > 0xFFFFFF)
256	278	return -EINVAL;
257	279
	280	+ mutex_lock(&priv->lock);
	281	+
258	282	priv->preset[chan->channel] = preset;
259	283
260	284	/* Reset Byte Pointer */
@@ -264,6 +288,8 @@ static ssize_t quad8_write_preset(struct iio_dev *indio_dev, uintptr_t private, · 3 unmodified lines
264	288	for (i = 0; i < 3; i++)
265	289	outb(preset >> (8 * i), base_offset);
266	290
	291	+ mutex_unlock(&priv->lock);
	292	+
267	293	return len;
268	294	}
269	295
@@ -293,6 +319,8 @@ static ssize_t quad8_write_set_to_preset_on_index(struct iio_dev *indio_dev, · 23 unmodified lines
293	319	/* Preset enable is active low in Input/Output Control register */
294	320	preset_enable = !preset_enable;
295	321
	322	+ mutex_lock(&priv->lock);
	323	+
296	324	priv->preset_enable[chan->channel] = preset_enable;
297	325
298	326	ior_cfg = priv->ab_enable[chan->channel] \|
@@ -301,6 +329,8 @@ static ssize_t quad8_write_set_to_preset_on_index(struct iio_dev *indio_dev, · 2 unmodified lines
301	329	/* Load I/O control configuration to Input / Output Control Register */
302	330	outb(QUAD8_CTR_IOR \| ior_cfg, base_offset);
303	331
	332	+ mutex_unlock(&priv->lock);
	333	+
304	334	return len;
305	335	}
306	336
@@ -358,6 +388,8 @@ static int quad8_set_count_mode(struct iio_dev *indio_dev, · 51 unmodified lines
358	388	unsigned int mode_cfg = cnt_mode << 1;
359	389	const int base_offset = priv->base + 2 * chan->channel + 1;
360	390
	391	+ mutex_lock(&priv->lock);
	392	+
361	393	priv->count_mode[chan->channel] = cnt_mode;
362	394
363	395	/* Add quadrature mode configuration */
@@ -367,6 +399,8 @@ static int quad8_set_count_mode(struct iio_dev *indio_dev, · 3 unmodified lines
367	399	/* Load mode configuration to Counter Mode Register */
368	400	outb(QUAD8_CTR_CMR \| mode_cfg, base_offset);
369	401
	402	+ mutex_unlock(&priv->lock);
	403	+
370	404	return 0;
371	405	}
372	406
@@ -394,19 +428,26 @@ static int quad8_set_synchronous_mode(struct iio_dev *indio_dev, · 21 unmodified lines
394	428	const struct iio_chan_spec *chan, unsigned int synchronous_mode)
395	429	{
396	430	struct quad8_iio *const priv = iio_priv(indio_dev);
397		− const unsigned int idr_cfg = synchronous_mode \|
398		− priv->index_polarity[chan->channel] << 1;
399	431	const int base_offset = priv->base + 2 * chan->channel + 1;
	432	+ unsigned int idr_cfg = synchronous_mode;
	433	+
	434	+ mutex_lock(&priv->lock);
	435	+
	436	+ idr_cfg \|= priv->index_polarity[chan->channel] << 1;
400	437
401	438	/* Index function must be non-synchronous in non-quadrature mode */
402		− if (synchronous_mode && !priv->quadrature_mode[chan->channel])
	439	+ if (synchronous_mode && !priv->quadrature_mode[chan->channel]) {
	440	+ mutex_unlock(&priv->lock);
403	441	return -EINVAL;
	442	+ }
404	443
405	444	priv->synchronous_mode[chan->channel] = synchronous_mode;
406	445
407	446	/* Load Index Control configuration to Index Control Register */
408	447	outb(QUAD8_CTR_IDR \| idr_cfg, base_offset);
409	448
	449	+ mutex_unlock(&priv->lock);
	450	+
410	451	return 0;
411	452	}
412	453
@@ -434,8 +475,12 @@ static int quad8_set_quadrature_mode(struct iio_dev *indio_dev, · 21 unmodified lines
434	475	const struct iio_chan_spec *chan, unsigned int quadrature_mode)
435	476	{
436	477	struct quad8_iio *const priv = iio_priv(indio_dev);
437		− unsigned int mode_cfg = priv->count_mode[chan->channel] << 1;
438	478	const int base_offset = priv->base + 2 * chan->channel + 1;
	479	+ unsigned int mode_cfg;
	480	+
	481	+ mutex_lock(&priv->lock);
	482	+
	483	+ mode_cfg = priv->count_mode[chan->channel] << 1;
439	484
440	485	if (quadrature_mode)
441	486	mode_cfg \|= (priv->quadrature_scale[chan->channel] + 1) << 3;
@@ -453,6 +498,8 @@ static int quad8_set_quadrature_mode(struct iio_dev *indio_dev, · 11 unmodified lines
453	498	/* Load mode configuration to Counter Mode Register */
454	499	outb(QUAD8_CTR_CMR \| mode_cfg, base_offset);
455	500
	501	+ mutex_unlock(&priv->lock);
	502	+
456	503	return 0;
457	504	}
458	505
@@ -480,15 +527,20 @@ static int quad8_set_index_polarity(struct iio_dev *indio_dev, · 21 unmodified lines
480	527	const struct iio_chan_spec *chan, unsigned int index_polarity)
481	528	{
482	529	struct quad8_iio *const priv = iio_priv(indio_dev);
483		− const unsigned int idr_cfg = priv->synchronous_mode[chan->channel] \|
484		− index_polarity << 1;
485	530	const int base_offset = priv->base + 2 * chan->channel + 1;
	531	+ unsigned int idr_cfg = index_polarity << 1;
	532	+
	533	+ mutex_lock(&priv->lock);
	534	+
	535	+ idr_cfg \|= priv->synchronous_mode[chan->channel];
486	536
487	537	priv->index_polarity[chan->channel] = index_polarity;
488	538
489	539	/* Load Index Control configuration to Index Control Register */
490	540	outb(QUAD8_CTR_IDR \| idr_cfg, base_offset);
491	541
	542	+ mutex_unlock(&priv->lock);
	543	+
492	544	return 0;
493	545	}
494	546
@@ -589,7 +641,7 @@ static int quad8_signal_read(struct counter_device *counter, · 94 unmodified lines
589	641	static int quad8_count_read(struct counter_device *counter,
590	642	struct counter_count count, unsigned long val)
591	643	{
592		− const struct quad8_iio *const priv = counter->priv;
	644	+ struct quad8_iio *const priv = counter->priv;
593	645	const int base_offset = priv->base + 2 * count->id;
594	646	unsigned int flags;
595	647	unsigned int borrow;
@@ -603,6 +655,8 @@ static int quad8_count_read(struct counter_device *counter, · 7 unmodified lines
603	655	/* Borrow XOR Carry effectively doubles count range */
604	656	*val = (unsigned long)(borrow ^ carry) << 24;
605	657
	658	+ mutex_lock(&priv->lock);
	659	+
606	660	/* Reset Byte Pointer; transfer Counter to Output Latch */
607	661	outb(QUAD8_CTR_RLD \| QUAD8_RLD_RESET_BP \| QUAD8_RLD_CNTR_OUT,
608	662	base_offset + 1);
@@ -610,13 +664,15 @@ static int quad8_count_read(struct counter_device *counter, · 1 unmodified line
610	664	for (i = 0; i < 3; i++)
611	665	val \|= (unsigned long)inb(base_offset) << (8 i);
612	666
	667	+ mutex_unlock(&priv->lock);
	668	+
613	669	return 0;
614	670	}
615	671
616	672	static int quad8_count_write(struct counter_device *counter,
617	673	struct counter_count *count, unsigned long val)
618	674	{
619		− const struct quad8_iio *const priv = counter->priv;
	675	+ struct quad8_iio *const priv = counter->priv;
620	676	const int base_offset = priv->base + 2 * count->id;
621	677	int i;
622	678
@@ -624,6 +680,8 @@ static int quad8_count_write(struct counter_device *counter, · 1 unmodified line
624	680	if (val > 0xFFFFFF)
625	681	return -EINVAL;
626	682
	683	+ mutex_lock(&priv->lock);
	684	+
627	685	/* Reset Byte Pointer */
628	686	outb(QUAD8_CTR_RLD \| QUAD8_RLD_RESET_BP, base_offset + 1);
629	687
@@ -647,6 +705,8 @@ static int quad8_count_write(struct counter_device *counter, · 17 unmodified lines
647	705	/* Reset Error flag */
648	706	outb(QUAD8_CTR_RLD \| QUAD8_RLD_RESET_E, base_offset + 1);
649	707
	708	+ mutex_unlock(&priv->lock);
	709	+
650	710	return 0;
651	711	}
652	712
@@ -667,13 +727,13 @@ static enum counter_count_function quad8_count_functions_list[] = { · 14 unmodified lines
667	727	static int quad8_function_get(struct counter_device *counter,
668	728	struct counter_count count, size_t function)
669	729	{
670		− const struct quad8_iio *const priv = counter->priv;
	730	+ struct quad8_iio *const priv = counter->priv;
671	731	const int id = count->id;
672		− const unsigned int quadrature_mode = priv->quadrature_mode[id];
673		− const unsigned int scale = priv->quadrature_scale[id];
674	732
675		− if (quadrature_mode)
676		− switch (scale) {
	733	+ mutex_lock(&priv->lock);
	734	+
	735	+ if (priv->quadrature_mode[id])
	736	+ switch (priv->quadrature_scale[id]) {
677	737	case 0:
678	738	*function = QUAD8_COUNT_FUNCTION_QUADRATURE_X1;
679	739	break;
@@ -687,6 +747,8 @@ static int quad8_function_get(struct counter_device *counter, · 7 unmodified lines
687	747	else
688	748	*function = QUAD8_COUNT_FUNCTION_PULSE_DIRECTION;
689	749
	750	+ mutex_unlock(&priv->lock);
	751	+
690	752	return 0;
691	753	}
692	754
@@ -697,10 +759,15 @@ static int quad8_function_set(struct counter_device *counter, · 4 unmodified lines
697	759	const int id = count->id;
698	760	unsigned int *const quadrature_mode = priv->quadrature_mode + id;
699	761	unsigned int *const scale = priv->quadrature_scale + id;
700		− unsigned int mode_cfg = priv->count_mode[id] << 1;
701	762	unsigned int *const synchronous_mode = priv->synchronous_mode + id;
702		− const unsigned int idr_cfg = priv->index_polarity[id] << 1;
703	763	const int base_offset = priv->base + 2 * id + 1;
	764	+ unsigned int mode_cfg;
	765	+ unsigned int idr_cfg;
	766	+
	767	+ mutex_lock(&priv->lock);
	768	+
	769	+ mode_cfg = priv->count_mode[id] << 1;
	770	+ idr_cfg = priv->index_polarity[id] << 1;
704	771
705	772	if (function == QUAD8_COUNT_FUNCTION_PULSE_DIRECTION) {
706	773	*quadrature_mode = 0;
@@ -736,6 +803,8 @@ static int quad8_function_set(struct counter_device *counter, · 29 unmodified lines
736	803	/* Load mode configuration to Counter Mode Register */
737	804	outb(QUAD8_CTR_CMR \| mode_cfg, base_offset);
738	805
	806	+ mutex_unlock(&priv->lock);
	807	+
739	808	return 0;
740	809	}
741	810
@@ -852,15 +921,20 @@ static int quad8_index_polarity_set(struct counter_device *counter, · 110 unmodified lines
852	921	{
853	922	struct quad8_iio *const priv = counter->priv;
854	923	const size_t channel_id = signal->id - 16;
855		− const unsigned int idr_cfg = priv->synchronous_mode[channel_id] \|
856		− index_polarity << 1;
857	924	const int base_offset = priv->base + 2 * channel_id + 1;
	925	+ unsigned int idr_cfg = index_polarity << 1;
	926	+
	927	+ mutex_lock(&priv->lock);
	928	+
	929	+ idr_cfg \|= priv->synchronous_mode[channel_id];
858	930
859	931	priv->index_polarity[channel_id] = index_polarity;
860	932
861	933	/* Load Index Control configuration to Index Control Register */
862	934	outb(QUAD8_CTR_IDR \| idr_cfg, base_offset);
863	935
	936	+ mutex_unlock(&priv->lock);
	937	+
864	938	return 0;
865	939	}
866	940
@@ -887,19 +961,26 @@ static int quad8_synchronous_mode_set(struct counter_device *counter, · 20 unmodified lines
887	961	{
888	962	struct quad8_iio *const priv = counter->priv;
889	963	const size_t channel_id = signal->id - 16;
890		− const unsigned int idr_cfg = synchronous_mode \|
891		− priv->index_polarity[channel_id] << 1;
892	964	const int base_offset = priv->base + 2 * channel_id + 1;
	965	+ unsigned int idr_cfg = synchronous_mode;
	966	+
	967	+ mutex_lock(&priv->lock);
	968	+
	969	+ idr_cfg \|= priv->index_polarity[channel_id] << 1;
893	970
894	971	/* Index function must be non-synchronous in non-quadrature mode */
895		− if (synchronous_mode && !priv->quadrature_mode[channel_id])
	972	+ if (synchronous_mode && !priv->quadrature_mode[channel_id]) {
	973	+ mutex_unlock(&priv->lock);
896	974	return -EINVAL;
	975	+ }
897	976
898	977	priv->synchronous_mode[channel_id] = synchronous_mode;
899	978
900	979	/* Load Index Control configuration to Index Control Register */
901	980	outb(QUAD8_CTR_IDR \| idr_cfg, base_offset);
902	981
	982	+ mutex_unlock(&priv->lock);
	983	+
903	984	return 0;
904	985	}
905	986
@@ -964,6 +1045,8 @@ static int quad8_count_mode_set(struct counter_device *counter, · 58 unmodified lines
964	1045	break;
965	1046	}
966	1047
	1048	+ mutex_lock(&priv->lock);
	1049	+
967	1050	priv->count_mode[count->id] = cnt_mode;
968	1051
969	1052	/* Set count mode configuration value */
@@ -976,6 +1059,8 @@ static int quad8_count_mode_set(struct counter_device *counter, · 6 unmodified lines
976	1059	/* Load mode configuration to Counter Mode Register */
977	1060	outb(QUAD8_CTR_CMR \| mode_cfg, base_offset);
978	1061
	1062	+ mutex_unlock(&priv->lock);
	1063	+
979	1064	return 0;
980	1065	}
981	1066
@@ -1017,6 +1102,8 @@ static ssize_t quad8_count_enable_write(struct counter_device *counter, · 35 unmodified lines
1017	1102	if (err)
1018	1103	return err;
1019	1104
	1105	+ mutex_lock(&priv->lock);
	1106	+
1020	1107	priv->ab_enable[count->id] = ab_enable;
1021	1108
1022	1109	ior_cfg = ab_enable \| priv->preset_enable[count->id] << 1;
@@ -1024,6 +1111,8 @@ static ssize_t quad8_count_enable_write(struct counter_device *counter, · 1 unmodified line
1024	1111	/* Load I/O control configuration */
1025	1112	outb(QUAD8_CTR_IOR \| ior_cfg, base_offset + 1);
1026	1113
	1114	+ mutex_unlock(&priv->lock);
	1115	+
1027	1116	return len;
1028	1117	}
1029	1118
@@ -1052,14 +1141,28 @@ static ssize_t quad8_count_preset_read(struct counter_device *counter, · 22 unmodified lines
1052	1141	return sprintf(buf, "%u\n", priv->preset[count->id]);
1053	1142	}
1054	1143
	1144	+static void quad8_preset_register_set(struct quad8_iio *quad8iio, int id,
	1145	+ unsigned int preset)
	1146	+{
	1147	+ const unsigned int base_offset = quad8iio->base + 2 * id;
	1148	+ int i;
	1149	+
	1150	+ quad8iio->preset[id] = preset;
	1151	+
	1152	+ /* Reset Byte Pointer */
	1153	+ outb(QUAD8_CTR_RLD \| QUAD8_RLD_RESET_BP, base_offset + 1);
	1154	+
	1155	+ /* Set Preset Register */
	1156	+ for (i = 0; i < 3; i++)
	1157	+ outb(preset >> (8 * i), base_offset);
	1158	+}
	1159	+
1055	1160	static ssize_t quad8_count_preset_write(struct counter_device *counter,
1056	1161	struct counter_count count, void private, const char *buf, size_t len)
1057	1162	{
1058	1163	struct quad8_iio *const priv = counter->priv;
1059		− const int base_offset = priv->base + 2 * count->id;
1060	1164	unsigned int preset;
1061	1165	int ret;
1062		− int i;
1063	1166
1064	1167	ret = kstrtouint(buf, 0, &preset);
1065	1168	if (ret)
@@ -1069,14 +1172,11 @@ static ssize_t quad8_count_preset_write(struct counter_device *counter, · 3 unmodified lines
1069	1172	if (preset > 0xFFFFFF)
1070	1173	return -EINVAL;
1071	1174
1072		− priv->preset[count->id] = preset;
	1175	+ mutex_lock(&priv->lock);
1073	1176
1074		− /* Reset Byte Pointer */
1075		− outb(QUAD8_CTR_RLD \| QUAD8_RLD_RESET_BP, base_offset + 1);
	1177	+ quad8_preset_register_set(priv, count->id, preset);
1076	1178
1077		− /* Set Preset Register */
1078		− for (i = 0; i < 3; i++)
1079		− outb(preset >> (8 * i), base_offset);
	1179	+ mutex_unlock(&priv->lock);
1080	1180
1081	1181	return len;
1082	1182	}
@@ -1084,15 +1184,20 @@ static ssize_t quad8_count_preset_write(struct counter_device *counter, · 1 unmodified line
1084	1184	static ssize_t quad8_count_ceiling_read(struct counter_device *counter,
1085	1185	struct counter_count count, void private, char *buf)
1086	1186	{
1087		− const struct quad8_iio *const priv = counter->priv;
	1187	+ struct quad8_iio *const priv = counter->priv;
	1188	+
	1189	+ mutex_lock(&priv->lock);
1088	1190
1089	1191	/* Range Limit and Modulo-N count modes use preset value as ceiling */
1090	1192	switch (priv->count_mode[count->id]) {
1091	1193	case 1:
1092	1194	case 3:
1093		− return quad8_count_preset_read(counter, count, private, buf);
	1195	+ mutex_unlock(&priv->lock);
	1196	+ return sprintf(buf, "%u\n", priv->preset[count->id]);
1094	1197	}
1095	1198
	1199	+ mutex_unlock(&priv->lock);
	1200	+
1096	1201	/* By default 0x1FFFFFF (25 bits unsigned) is maximum count */
1097	1202	return sprintf(buf, "33554431\n");
1098	1203	}
@@ -1101,15 +1206,29 @@ static ssize_t quad8_count_ceiling_write(struct counter_device *counter, · 2 unmodified lines
1101	1206	struct counter_count count, void private, const char *buf, size_t len)
1102	1207	{
1103	1208	struct quad8_iio *const priv = counter->priv;
	1209	+ unsigned int ceiling;
	1210	+ int ret;
	1211	+
	1212	+ ret = kstrtouint(buf, 0, &ceiling);
	1213	+ if (ret)
	1214	+ return ret;
	1215	+
	1216	+ /* Only 24-bit values are supported */
	1217	+ if (ceiling > 0xFFFFFF)
	1218	+ return -EINVAL;
	1219	+
	1220	+ mutex_lock(&priv->lock);
1104	1221
1105	1222	/* Range Limit and Modulo-N count modes use preset value as ceiling */
1106	1223	switch (priv->count_mode[count->id]) {
1107	1224	case 1:
1108	1225	case 3:
1109		− return quad8_count_preset_write(counter, count, private, buf,
1110		− len);
	1226	+ quad8_preset_register_set(priv, count->id, ceiling);
	1227	+ break;
1111	1228	}
1112	1229
	1230	+ mutex_unlock(&priv->lock);
	1231	+
1113	1232	return len;
1114	1233	}
1115	1234
@@ -1137,6 +1256,8 @@ static ssize_t quad8_count_preset_enable_write(struct counter_device *counter, · 21 unmodified lines
1137	1256	/* Preset enable is active low in Input/Output Control register */
1138	1257	preset_enable = !preset_enable;
1139	1258
	1259	+ mutex_lock(&priv->lock);
	1260	+
1140	1261	priv->preset_enable[count->id] = preset_enable;
1141	1262
1142	1263	ior_cfg = priv->ab_enable[count->id] \| (unsigned int)preset_enable << 1;
@@ -1144,6 +1265,8 @@ static ssize_t quad8_count_preset_enable_write(struct counter_device *counter, · 1 unmodified line
1144	1265	/* Load I/O control configuration to Input / Output Control Register */
1145	1266	outb(QUAD8_CTR_IOR \| ior_cfg, base_offset);
1146	1267
	1268	+ mutex_unlock(&priv->lock);
	1269	+
1147	1270	return len;
1148	1271	}
1149	1272
@@ -1429,6 +1552,9 @@ static int quad8_probe(struct device *dev, unsigned int id) · 279 unmodified lines
1429	1552	quad8iio->counter.priv = quad8iio;
1430	1553	quad8iio->base = base[id];
1431	1554
	1555	+ /* Initialize mutex */
	1556	+ mutex_init(&quad8iio->lock);
	1557	+
1432	1558	/* Reset all counters and disable interrupt function */
1433	1559	outb(QUAD8_CHAN_OP_RESET_COUNTERS, base[id] + QUAD8_REG_CHAN_OP);
1434	1560	/* Set initial configuration for all counters */

drivers/cpufreq/intel_pstate.c+1−1

@@ -1059,7 +1059,7 @@ static ssize_t store_no_turbo(struct kobject a, struct kobj_attribute b, · 1058 unmodified lines
1059	1059
1060	1060	update_turbo_state();
1061	1061	if (global.turbo_disabled) {
1062		− pr_warn("Turbo disabled by BIOS or unavailable on processor\n");
	1062	+ pr_notice_once("Turbo disabled by BIOS or unavailable on processor\n");
1063	1063	mutex_unlock(&intel_pstate_limits_lock);
1064	1064	mutex_unlock(&intel_pstate_driver_lock);
1065	1065	return -EPERM;

drivers/crypto/caam/caamalg.c+7−3

@@ -963,10 +963,12 @@ static void aead_crypt_done(struct device jrdev, u32 desc, u32 err, · 962 unmodified lines
963	963	struct caam_drv_private_jr *jrp = dev_get_drvdata(jrdev);
964	964	struct aead_edesc *edesc;
965	965	int ecode = 0;
	966	+ bool has_bklog;
966	967
967	968	dev_dbg(jrdev, "%s %d: err 0x%x\n", __func__, __LINE__, err);
968	969
969	970	edesc = rctx->edesc;
	971	+ has_bklog = edesc->bklog;
970	972
971	973	if (err)
972	974	ecode = caam_jr_strstatus(jrdev, err);
@@ -979,7 +981,7 @@ static void aead_crypt_done(struct device jrdev, u32 desc, u32 err, · 6 unmodified lines
979	981	* If no backlog flag, the completion of the request is done
980	982	* by CAAM, not crypto engine.
981	983	*/
982		− if (!edesc->bklog)
	984	+ if (!has_bklog)
983	985	aead_request_complete(req, ecode);
984	986	else
985	987	crypto_finalize_aead_request(jrp->engine, req, ecode);
@@ -995,10 +997,12 @@ static void skcipher_crypt_done(struct device jrdev, u32 desc, u32 err, · 9 unmodified lines
995	997	struct caam_drv_private_jr *jrp = dev_get_drvdata(jrdev);
996	998	int ivsize = crypto_skcipher_ivsize(skcipher);
997	999	int ecode = 0;
	1000	+ bool has_bklog;
998	1001
999	1002	dev_dbg(jrdev, "%s %d: err 0x%x\n", __func__, __LINE__, err);
1000	1003
1001	1004	edesc = rctx->edesc;
	1005	+ has_bklog = edesc->bklog;
1002	1006	if (err)
1003	1007	ecode = caam_jr_strstatus(jrdev, err);
1004	1008
@@ -1028,7 +1032,7 @@ static void skcipher_crypt_done(struct device jrdev, u32 desc, u32 err, · 23 unmodified lines
1028	1032	* If no backlog flag, the completion of the request is done
1029	1033	* by CAAM, not crypto engine.
1030	1034	*/
1031		− if (!edesc->bklog)
	1035	+ if (!has_bklog)
1032	1036	skcipher_request_complete(req, ecode);
1033	1037	else
1034	1038	crypto_finalize_skcipher_request(jrp->engine, req, ecode);
@@ -1711,7 +1715,7 @@ static struct skcipher_edesc skcipher_edesc_alloc(struct skcipher_request req, · 676 unmodified lines
1711	1715
1712	1716	if (ivsize \|\| mapped_dst_nents > 1)
1713	1717	sg_to_sec4_set_last(edesc->sec4_sg + dst_sg_idx +
1714		− mapped_dst_nents);
	1718	+ mapped_dst_nents - 1 + !!ivsize);
1715	1719
1716	1720	if (sec4_sg_bytes) {
1717	1721	edesc->sec4_sg_dma = dma_map_single(jrdev, edesc->sec4_sg,

drivers/crypto/caam/caamhash.c+6−2

@@ -583,10 +583,12 @@ static inline void ahash_done_cpy(struct device jrdev, u32 desc, u32 err, · 582 unmodified lines
583	583	struct caam_hash_state *state = ahash_request_ctx(req);
584	584	struct caam_hash_ctx *ctx = crypto_ahash_ctx(ahash);
585	585	int ecode = 0;
	586	+ bool has_bklog;
586	587
587	588	dev_dbg(jrdev, "%s %d: err 0x%x\n", __func__, __LINE__, err);
588	589
589	590	edesc = state->edesc;
	591	+ has_bklog = edesc->bklog;
590	592
591	593	if (err)
592	594	ecode = caam_jr_strstatus(jrdev, err);
@@ -603,7 +605,7 @@ static inline void ahash_done_cpy(struct device jrdev, u32 desc, u32 err, · 10 unmodified lines
603	605	* If no backlog flag, the completion of the request is done
604	606	* by CAAM, not crypto engine.
605	607	*/
606		− if (!edesc->bklog)
	608	+ if (!has_bklog)
607	609	req->base.complete(&req->base, ecode);
608	610	else
609	611	crypto_finalize_hash_request(jrp->engine, req, ecode);
@@ -632,10 +634,12 @@ static inline void ahash_done_switch(struct device jrdev, u32 desc, u32 err, · 22 unmodified lines
632	634	struct caam_hash_state *state = ahash_request_ctx(req);
633	635	int digestsize = crypto_ahash_digestsize(ahash);
634	636	int ecode = 0;
	637	+ bool has_bklog;
635	638
636	639	dev_dbg(jrdev, "%s %d: err 0x%x\n", __func__, __LINE__, err);
637	640
638	641	edesc = state->edesc;
	642	+ has_bklog = edesc->bklog;
639	643	if (err)
640	644	ecode = caam_jr_strstatus(jrdev, err);
641	645
@@ -663,7 +667,7 @@ static inline void ahash_done_switch(struct device jrdev, u32 desc, u32 err, · 21 unmodified lines
663	667	* If no backlog flag, the completion of the request is done
664	668	* by CAAM, not crypto engine.
665	669	*/
666		− if (!edesc->bklog)
	670	+ if (!has_bklog)
667	671	req->base.complete(&req->base, ecode);
668	672	else
669	673	crypto_finalize_hash_request(jrp->engine, req, ecode);

drivers/crypto/caam/caampkc.c+6−2

@@ -121,11 +121,13 @@ static void rsa_pub_done(struct device dev, u32 desc, u32 err, void *context) · 120 unmodified lines
121	121	struct caam_drv_private_jr *jrp = dev_get_drvdata(dev);
122	122	struct rsa_edesc *edesc;
123	123	int ecode = 0;
	124	+ bool has_bklog;
124	125
125	126	if (err)
126	127	ecode = caam_jr_strstatus(dev, err);
127	128
128	129	edesc = req_ctx->edesc;
	130	+ has_bklog = edesc->bklog;
129	131
130	132	rsa_pub_unmap(dev, edesc, req);
131	133	rsa_io_unmap(dev, edesc, req);
@@ -135,7 +137,7 @@ static void rsa_pub_done(struct device dev, u32 desc, u32 err, void *context) · 3 unmodified lines
135	137	* If no backlog flag, the completion of the request is done
136	138	* by CAAM, not crypto engine.
137	139	*/
138		− if (!edesc->bklog)
	140	+ if (!has_bklog)
139	141	akcipher_request_complete(req, ecode);
140	142	else
141	143	crypto_finalize_akcipher_request(jrp->engine, req, ecode);
@@ -152,11 +154,13 @@ static void rsa_priv_f_done(struct device dev, u32 desc, u32 err, · 10 unmodified lines
152	154	struct caam_rsa_req_ctx *req_ctx = akcipher_request_ctx(req);
153	155	struct rsa_edesc *edesc;
154	156	int ecode = 0;
	157	+ bool has_bklog;
155	158
156	159	if (err)
157	160	ecode = caam_jr_strstatus(dev, err);
158	161
159	162	edesc = req_ctx->edesc;
	163	+ has_bklog = edesc->bklog;
160	164
161	165	switch (key->priv_form) {
162	166	case FORM1:
@@ -176,7 +180,7 @@ static void rsa_priv_f_done(struct device dev, u32 desc, u32 err, · 13 unmodified lines
176	180	* If no backlog flag, the completion of the request is done
177	181	* by CAAM, not crypto engine.
178	182	*/
179		− if (!edesc->bklog)
	183	+ if (!has_bklog)
180	184	akcipher_request_complete(req, ecode);
181	185	else
182	186	crypto_finalize_akcipher_request(jrp->engine, req, ecode);

drivers/crypto/chelsio/chcr_ktls.c+48−41

@@ -120,12 +120,10 @@ out: · 119 unmodified lines
120	120	static int chcr_ktls_update_connection_state(struct chcr_ktls_info *tx_info,
121	121	int new_state)
122	122	{
123		− unsigned long flags;
124		−
125	123	/* This function can be called from both rx (interrupt context) and tx
126	124	* queue contexts.
127	125	*/
128		− spin_lock_irqsave(&tx_info->lock, flags);
	126	+ spin_lock_bh(&tx_info->lock);
129	127	switch (tx_info->connection_state) {
130	128	case KTLS_CONN_CLOSED:
131	129	tx_info->connection_state = new_state;
@@ -169,7 +167,7 @@ static int chcr_ktls_update_connection_state(struct chcr_ktls_info *tx_info, · 37 unmodified lines
169	167	pr_err("unknown KTLS connection state\n");
170	168	break;
171	169	}
172		− spin_unlock_irqrestore(&tx_info->lock, flags);
	170	+ spin_unlock_bh(&tx_info->lock);
173	171
174	172	return tx_info->connection_state;
175	173	}
@@ -675,41 +673,14 @@ int chcr_ktls_cpl_set_tcb_rpl(struct adapter adap, unsigned char input) · 499 unmodified lines
675	673	return 0;
676	674	}
677	675
678		−/*
679		− * chcr_write_cpl_set_tcb_ulp: update tcb values.
680		− * TCB is responsible to create tcp headers, so all the related values
681		− * should be correctly updated.
682		− * @tx_info - driver specific tls info.
683		− * @q - tx queue on which packet is going out.
684		− * @tid - TCB identifier.
685		− * @pos - current index where should we start writing.
686		− * @word - TCB word.
687		− * @mask - TCB word related mask.
688		− * @val - TCB word related value.
689		− * @reply - set 1 if looking for TP response.
690		− * return - next position to write.
691		− */
692		−static void chcr_write_cpl_set_tcb_ulp(struct chcr_ktls_info tx_info,
693		− struct sge_eth_txq *q, u32 tid,
694		− void *pos, u16 word, u64 mask,
	676	+static void __chcr_write_cpl_set_tcb_ulp(struct chcr_ktls_info tx_info,
	677	+ u32 tid, void *pos, u16 word, u64 mask,
695	678	u64 val, u32 reply)
696	679	{
697	680	struct cpl_set_tcb_field_core *cpl;
698	681	struct ulptx_idata *idata;
699	682	struct ulp_txpkt *txpkt;
700		− void *save_pos = NULL;
701		− u8 buf[48] = {0};
702		− int left;
703	683
704		− left = (void *)q->q.stat - pos;
705		− if (unlikely(left < CHCR_SET_TCB_FIELD_LEN)) {
706		− if (!left) {
707		− pos = q->q.desc;
708		− } else {
709		− save_pos = pos;
710		− pos = buf;
711		− }
712		− }
713	684	/* ULP_TXPKT */
714	685	txpkt = pos;
715	686	txpkt->cmd_dest = htonl(ULPTX_CMD_V(ULP_TX_PKT) \| ULP_TXPKT_DEST_V(0));
@@ -734,18 +705,54 @@ static void chcr_write_cpl_set_tcb_ulp(struct chcr_ktls_info tx_info, · 18 unmodified lines
734	705	idata = (struct ulptx_idata *)(cpl + 1);
735	706	idata->cmd_more = htonl(ULPTX_CMD_V(ULP_TX_SC_NOOP));
736	707	idata->len = htonl(0);
	708	+ pos = idata + 1;
737	709
738		− if (save_pos) {
739		− pos = chcr_copy_to_txd(buf, &q->q, save_pos,
740		− CHCR_SET_TCB_FIELD_LEN);
741		− } else {
742		− /* check again if we are at the end of the queue */
743		− if (left == CHCR_SET_TCB_FIELD_LEN)
	710	+ return pos;
	711	+}
	712	+
	713	+
	714	+/*
	715	+ * chcr_write_cpl_set_tcb_ulp: update tcb values.
	716	+ * TCB is responsible to create tcp headers, so all the related values
	717	+ * should be correctly updated.
	718	+ * @tx_info - driver specific tls info.
	719	+ * @q - tx queue on which packet is going out.
	720	+ * @tid - TCB identifier.
	721	+ * @pos - current index where should we start writing.
	722	+ * @word - TCB word.
	723	+ * @mask - TCB word related mask.
	724	+ * @val - TCB word related value.
	725	+ * @reply - set 1 if looking for TP response.
	726	+ * return - next position to write.
	727	+ */
	728	+static void chcr_write_cpl_set_tcb_ulp(struct chcr_ktls_info tx_info,
	729	+ struct sge_eth_txq *q, u32 tid,
	730	+ void *pos, u16 word, u64 mask,
	731	+ u64 val, u32 reply)
	732	+{
	733	+ int left = (void *)q->q.stat - pos;
	734	+
	735	+ if (unlikely(left < CHCR_SET_TCB_FIELD_LEN)) {
	736	+ if (!left) {
744	737	pos = q->q.desc;
745		− else
746		− pos = idata + 1;
	738	+ } else {
	739	+ u8 buf[48] = {0};
	740	+
	741	+ __chcr_write_cpl_set_tcb_ulp(tx_info, tid, buf, word,
	742	+ mask, val, reply);
	743	+
	744	+ return chcr_copy_to_txd(buf, &q->q, pos,
	745	+ CHCR_SET_TCB_FIELD_LEN);
	746	+ }
747	747	}
748	748
	749	+ pos = __chcr_write_cpl_set_tcb_ulp(tx_info, tid, pos, word,
	750	+ mask, val, reply);
	751	+
	752	+ /* check again if we are at the end of the queue */
	753	+ if (left == CHCR_SET_TCB_FIELD_LEN)
	754	+ pos = q->q.desc;
	755	+
749	756	return pos;
750	757	}
751	758

drivers/dma-buf/dma-buf.c+4−3

@@ -388,7 +388,8 @@ static long dma_buf_ioctl(struct file *file, · 387 unmodified lines
388	388
389	389	return ret;
390	390
391		− case DMA_BUF_SET_NAME:
	391	+ case DMA_BUF_SET_NAME_A:
	392	+ case DMA_BUF_SET_NAME_B:
392	393	return dma_buf_set_name(dmabuf, (const char __user *)arg);
393	394
394	395	default:
@@ -655,8 +656,8 @@ EXPORT_SYMBOL_GPL(dma_buf_put); · 260 unmodified lines
655	656	* calls attach() of dma_buf_ops to allow device-specific attach functionality
656	657	* @dmabuf: [in] buffer to attach device to.
657	658	* @dev: [in] device to be attached.
658		− * @importer_ops [in] importer operations for the attachment
659		− * @importer_priv [in] importer private pointer for the attachment
	659	+ * @importer_ops: [in] importer operations for the attachment
	660	+ * @importer_priv: [in] importer private pointer for the attachment
660	661	*
661	662	* Returns struct dma_buf_attachment pointer for this attachment. Attachments
662	663	* must be cleaned up by calling dma_buf_detach().

drivers/dma/Kconfig+2−1

@@ -241,7 +241,8 @@ config FSL_RAID · 240 unmodified lines
241	241
242	242	config HISI_DMA
243	243	tristate "HiSilicon DMA Engine support"
244		− depends on ARM64 \|\| (COMPILE_TEST && PCI_MSI)
	244	+ depends on ARM64 \|\| COMPILE_TEST
	245	+ depends on PCI_MSI
245	246	select DMA_ENGINE
246	247	select DMA_VIRTUAL_CHANNELS
247	248	help

drivers/dma/dmaengine.c+26−34

@@ -232,10 +232,6 @@ static void chan_dev_release(struct device *dev) · 231 unmodified lines
232	232	struct dma_chan_dev *chan_dev;
233	233
234	234	chan_dev = container_of(dev, typeof(*chan_dev), device);
235		− if (atomic_dec_and_test(chan_dev->idr_ref)) {
236		− ida_free(&dma_ida, chan_dev->dev_id);
237		− kfree(chan_dev->idr_ref);
238		− }
239	235	kfree(chan_dev);
240	236	}
241	237
@@ -1043,27 +1039,9 @@ static int get_dma_id(struct dma_device *device) · 801 unmodified lines
1043	1039	}
1044	1040
1045	1041	static int __dma_async_device_channel_register(struct dma_device *device,
1046		− struct dma_chan *chan,
1047		− int chan_id)
	1042	+ struct dma_chan *chan)
1048	1043	{
1049	1044	int rc = 0;
1050		− int chancnt = device->chancnt;
1051		− atomic_t *idr_ref;
1052		− struct dma_chan *tchan;
1053		−
1054		− tchan = list_first_entry_or_null(&device->channels,
1055		− struct dma_chan, device_node);
1056		− if (!tchan)
1057		− return -ENODEV;
1058		−
1059		− if (tchan->dev) {
1060		− idr_ref = tchan->dev->idr_ref;
1061		− } else {
1062		− idr_ref = kmalloc(sizeof(*idr_ref), GFP_KERNEL);
1063		− if (!idr_ref)
1064		− return -ENOMEM;
1065		− atomic_set(idr_ref, 0);
1066		− }
1067	1045
1068	1046	chan->local = alloc_percpu(typeof(*chan->local));
1069	1047	if (!chan->local)
@@ -1079,29 +1057,36 @@ static int __dma_async_device_channel_register(struct dma_device *device, · 9 unmodified lines
1079	1057	* When the chan_id is a negative value, we are dynamically adding
1080	1058	* the channel. Otherwise we are static enumerating.
1081	1059	*/
1082		− chan->chan_id = chan_id < 0 ? chancnt : chan_id;
	1060	+ mutex_lock(&device->chan_mutex);
	1061	+ chan->chan_id = ida_alloc(&device->chan_ida, GFP_KERNEL);
	1062	+ mutex_unlock(&device->chan_mutex);
	1063	+ if (chan->chan_id < 0) {
	1064	+ pr_err("%s: unable to alloc ida for chan: %d\n",
	1065	+ __func__, chan->chan_id);
	1066	+ goto err_out;
	1067	+ }
	1068	+
1083	1069	chan->dev->device.class = &dma_devclass;
1084	1070	chan->dev->device.parent = device->dev;
1085	1071	chan->dev->chan = chan;
1086		− chan->dev->idr_ref = idr_ref;
1087	1072	chan->dev->dev_id = device->dev_id;
1088		− atomic_inc(idr_ref);
1089	1073	dev_set_name(&chan->dev->device, "dma%dchan%d",
1090	1074	device->dev_id, chan->chan_id);
1091		−
1092	1075	rc = device_register(&chan->dev->device);
1093	1076	if (rc)
1094		− goto err_out;
	1077	+ goto err_out_ida;
1095	1078	chan->client_count = 0;
1096		− device->chancnt = chan->chan_id + 1;
	1079	+ device->chancnt++;
1097	1080
1098	1081	return 0;
1099	1082
	1083	+ err_out_ida:
	1084	+ mutex_lock(&device->chan_mutex);
	1085	+ ida_free(&device->chan_ida, chan->chan_id);
	1086	+ mutex_unlock(&device->chan_mutex);
1100	1087	err_out:
1101	1088	free_percpu(chan->local);
1102	1089	kfree(chan->dev);
1103		− if (atomic_dec_return(idr_ref) == 0)
1104		− kfree(idr_ref);
1105	1090	return rc;
1106	1091	}
1107	1092
@@ -1110,7 +1095,7 @@ int dma_async_device_channel_register(struct dma_device *device, · 2 unmodified lines
1110	1095	{
1111	1096	int rc;
1112	1097
1113		− rc = __dma_async_device_channel_register(device, chan, -1);
	1098	+ rc = __dma_async_device_channel_register(device, chan);
1114	1099	if (rc < 0)
1115	1100	return rc;
1116	1101
@@ -1130,6 +1115,9 @@ static void __dma_async_device_channel_unregister(struct dma_device *device, · 13 unmodified lines
1130	1115	device->chancnt--;
1131	1116	chan->dev->chan = NULL;
1132	1117	mutex_unlock(&dma_list_mutex);
	1118	+ mutex_lock(&device->chan_mutex);
	1119	+ ida_free(&device->chan_ida, chan->chan_id);
	1120	+ mutex_unlock(&device->chan_mutex);
1133	1121	device_unregister(&chan->dev->device);
1134	1122	free_percpu(chan->local);
1135	1123	}
@@ -1152,7 +1140,7 @@ EXPORT_SYMBOL_GPL(dma_async_device_channel_unregister); · 16 unmodified lines
1152	1140	*/
1153	1141	int dma_async_device_register(struct dma_device *device)
1154	1142	{
1155		− int rc, i = 0;
	1143	+ int rc;
1156	1144	struct dma_chan* chan;
1157	1145
1158	1146	if (!device)
@@ -1257,9 +1245,12 @@ int dma_async_device_register(struct dma_device *device) · 98 unmodified lines
1257	1245	if (rc != 0)
1258	1246	return rc;
1259	1247
	1248	+ mutex_init(&device->chan_mutex);
	1249	+ ida_init(&device->chan_ida);
	1250	+
1260	1251	/* represent channels in sysfs. Probably want devs too */
1261	1252	list_for_each_entry(chan, &device->channels, device_node) {
1262		− rc = __dma_async_device_channel_register(device, chan, i++);
	1253	+ rc = __dma_async_device_channel_register(device, chan);
1263	1254	if (rc < 0)
1264	1255	goto err_out;
1265	1256	}
@@ -1334,6 +1325,7 @@ void dma_async_device_unregister(struct dma_device *device) · 68 unmodified lines
1334	1325	*/
1335	1326	dma_cap_set(DMA_PRIVATE, device->cap_mask);
1336	1327	dma_channel_rebalance();
	1328	+ ida_free(&dma_ida, device->dev_id);
1337	1329	dma_device_put(device);
1338	1330	mutex_unlock(&dma_list_mutex);
1339	1331	}

drivers/dma/dmatest.c+3−3

@@ -240,7 +240,7 @@ static bool is_threaded_test_run(struct dmatest_info *info) · 239 unmodified lines
240	240	struct dmatest_thread *thread;
241	241
242	242	list_for_each_entry(thread, &dtc->threads, node) {
243		− if (!thread->done)
	243	+ if (!thread->done && !thread->pending)
244	244	return true;
245	245	}
246	246	}
@@ -662,8 +662,8 @@ static int dmatest_func(void *data) · 415 unmodified lines
662	662	flags = DMA_CTRL_ACK \| DMA_PREP_INTERRUPT;
663	663
664	664	ktime = ktime_get();
665		− while (!kthread_should_stop()
666		− && !(params->iterations && total_tests >= params->iterations)) {
	665	+ while (!(kthread_should_stop() \|\|
	666	+ (params->iterations && total_tests >= params->iterations))) {
667	667	struct dma_async_tx_descriptor *tx = NULL;
668	668	struct dmaengine_unmap_data *um;
669	669	dma_addr_t *dsts;

drivers/dma/mmp_tdma.c+4−1

@@ -363,6 +363,8 @@ static void mmp_tdma_free_descriptor(struct mmp_tdma_chan *tdmac) · 362 unmodified lines
363	363	gen_pool_free(gpool, (unsigned long)tdmac->desc_arr,
364	364	size);
365	365	tdmac->desc_arr = NULL;
	366	+ if (tdmac->status == DMA_ERROR)
	367	+ tdmac->status = DMA_COMPLETE;
366	368
367	369	return;
368	370	}
@@ -443,7 +445,8 @@ static struct dma_async_tx_descriptor *mmp_tdma_prep_dma_cyclic( · 74 unmodified lines
443	445	if (!desc)
444	446	goto err_out;
445	447
446		− mmp_tdma_config_write(chan, direction, &tdmac->slave_config);
	448	+ if (mmp_tdma_config_write(chan, direction, &tdmac->slave_config))
	449	+ goto err_out;
447	450
448	451	while (buf < buf_len) {
449	452	desc = &tdmac->desc_arr[i];

drivers/dma/pch_dma.c+1−1

@@ -865,6 +865,7 @@ static int pch_dma_probe(struct pci_dev *pdev, · 864 unmodified lines
865	865	}
866	866
867	867	pci_set_master(pdev);
	868	+ pd->dma.dev = &pdev->dev;
868	869
869	870	err = request_irq(pdev->irq, pd_irq, IRQF_SHARED, DRV_NAME, pd);
870	871	if (err) {
@@ -880,7 +881,6 @@ static int pch_dma_probe(struct pci_dev *pdev, · 9 unmodified lines
880	881	goto err_free_irq;
881	882	}
882	883
883		− pd->dma.dev = &pdev->dev;
884	884
885	885	INIT_LIST_HEAD(&pd->dma.channels);
886	886

drivers/dma/tegra20-apb-dma.c+9−0

@@ -816,6 +816,13 @@ static bool tegra_dma_eoc_interrupt_deasserted(struct tegra_dma_channel *tdc) · 815 unmodified lines
816	816	static void tegra_dma_synchronize(struct dma_chan *dc)
817	817	{
818	818	struct tegra_dma_channel *tdc = to_tegra_dma_chan(dc);
	819	+ int err;
	820	+
	821	+ err = pm_runtime_get_sync(tdc->tdma->dev);
	822	+ if (err < 0) {
	823	+ dev_err(tdc2dev(tdc), "Failed to synchronize DMA: %d\n", err);
	824	+ return;
	825	+ }
819	826
820	827	/*
821	828	* CPU, which handles interrupt, could be busy in
@@ -825,6 +832,8 @@ static void tegra_dma_synchronize(struct dma_chan *dc) · 3 unmodified lines
825	832	wait_event(tdc->wq, tegra_dma_eoc_interrupt_deasserted(tdc));
826	833
827	834	tasklet_kill(&tdc->tasklet);
	835	+
	836	+ pm_runtime_put(tdc->tdma->dev);
828	837	}
829	838
830	839	static unsigned int tegra_dma_sg_bytes_xferred(struct tegra_dma_channel *tdc,

drivers/dma/ti/k3-psil.c+1−0

@@ -27,6 +27,7 @@ struct psil_endpoint_config *psil_get_ep_config(u32 thread_id) · 26 unmodified lines
27	27	soc_ep_map = &j721e_ep_map;
28	28	} else {
29	29	pr_err("PSIL: No compatible machine found for map\n");
	30	+ mutex_unlock(&ep_map_mutex);
30	31	return ERR_PTR(-ENOTSUPP);
31	32	}
32	33	pr_debug("%s: Using map for %s\n", __func__, soc_ep_map->name);

drivers/dma/xilinx/xilinx_dma.c+10−10

@@ -1230,16 +1230,16 @@ static enum dma_status xilinx_dma_tx_status(struct dma_chan *dchan, · 1229 unmodified lines
1230	1230	return ret;
1231	1231
1232	1232	spin_lock_irqsave(&chan->lock, flags);
1233		−
1234		− desc = list_last_entry(&chan->active_list,
1235		− struct xilinx_dma_tx_descriptor, node);
1236		− /*
1237		− * VDMA and simple mode do not support residue reporting, so the
1238		− * residue field will always be 0.
1239		− */
1240		− if (chan->has_sg && chan->xdev->dma_config->dmatype != XDMA_TYPE_VDMA)
1241		− residue = xilinx_dma_get_residue(chan, desc);
1242		−
	1233	+ if (!list_empty(&chan->active_list)) {
	1234	+ desc = list_last_entry(&chan->active_list,
	1235	+ struct xilinx_dma_tx_descriptor, node);
	1236	+ /*
	1237	+ * VDMA and simple mode do not support residue reporting, so the
	1238	+ * residue field will always be 0.
	1239	+ */
	1240	+ if (chan->has_sg && chan->xdev->dma_config->dmatype != XDMA_TYPE_VDMA)
	1241	+ residue = xilinx_dma_get_residue(chan, desc);
	1242	+ }
1243	1243	spin_unlock_irqrestore(&chan->lock, flags);
1244	1244
1245	1245	dma_set_residue(txstate, residue);

drivers/firmware/efi/tpm.c+1−1

@@ -16,7 +16,7 @@ · 15 unmodified lines
16	16	int efi_tpm_final_log_size;
17	17	EXPORT_SYMBOL(efi_tpm_final_log_size);
18	18
19		−static int tpm2_calc_event_log_size(void data, int count, void size_info)
	19	+static int __init tpm2_calc_event_log_size(void data, int count, void size_info)
20	20	{
21	21	struct tcg_pcr_event2_head *header;
22	22	int event_size, size = 0;

drivers/firmware/imx/Kconfig+2−2

@@ -12,7 +12,7 @@ config IMX_DSP · 11 unmodified lines
12	12
13	13	config IMX_SCU
14	14	bool "IMX SCU Protocol driver"
15		− depends on IMX_MBOX \|\| COMPILE_TEST
	15	+ depends on IMX_MBOX
16	16	help
17	17	The System Controller Firmware (SCFW) is a low-level system function
18	18	which runs on a dedicated Cortex-M core to provide power, clock, and
@@ -24,6 +24,6 @@ config IMX_SCU · 5 unmodified lines
24	24
25	25	config IMX_SCU_PD
26	26	bool "IMX SCU Power Domain driver"
27		− depends on IMX_SCU \|\| COMPILE_TEST
	27	+ depends on IMX_SCU
28	28	help
29	29	The System Controller Firmware (SCFW) based power domain driver.

drivers/firmware/xilinx/zynqmp-debug.c+1−1

@@ -35,7 +35,7 @@ static struct pm_api_info pm_api_list[] = { · 34 unmodified lines
35	35	PM_API(PM_QUERY_DATA),
36	36	};
37	37
38		−struct dentry *firmware_debugfs_root;
	38	+static struct dentry *firmware_debugfs_root;
39	39
40	40	/**
41	41	* zynqmp_pm_argument_value() - Extract argument value from a PM-API request

drivers/fpga/dfl-pci.c+4−2

@@ -248,11 +248,13 @@ static int cci_pci_sriov_configure(struct pci_dev *pcidev, int num_vfs) · 247 unmodified lines
248	248	return ret;
249	249
250	250	ret = pci_enable_sriov(pcidev, num_vfs);
251		− if (ret)
	251	+ if (ret) {
252	252	dfl_fpga_cdev_config_ports_pf(cdev);
	253	+ return ret;
	254	+ }
253	255	}
254	256
255		− return ret;
	257	+ return num_vfs;
256	258	}
257	259
258	260	static void cci_pci_remove(struct pci_dev *pcidev)

drivers/fpga/zynq-fpga.c+2−1

@@ -583,7 +583,8 @@ static int zynq_fpga_probe(struct platform_device *pdev) · 582 unmodified lines
583	583
584	584	priv->clk = devm_clk_get(dev, "ref_clk");
585	585	if (IS_ERR(priv->clk)) {
586		− dev_err(dev, "input clock not found\n");
	586	+ if (PTR_ERR(priv->clk) != -EPROBE_DEFER)
	587	+ dev_err(dev, "input clock not found\n");
587	588	return PTR_ERR(priv->clk);
588	589	}
589	590

drivers/gpu/drm/amd/amdgpu/amdgpu_device.c+2−5

@@ -3372,15 +3372,12 @@ int amdgpu_device_suspend(struct drm_device *dev, bool fbcon) · 3371 unmodified lines
3372	3372	}
3373	3373	}
3374	3374
3375		− amdgpu_device_set_pg_state(adev, AMD_PG_STATE_UNGATE);
3376		− amdgpu_device_set_cg_state(adev, AMD_CG_STATE_UNGATE);
3377		−
3378		− amdgpu_amdkfd_suspend(adev, !fbcon);
3379		−
3380	3375	amdgpu_ras_suspend(adev);
3381	3376
3382	3377	r = amdgpu_device_ip_suspend_phase1(adev);
3383	3378
	3379	+ amdgpu_amdkfd_suspend(adev, !fbcon);
	3380	+
3384	3381	/* evict vram memory */
3385	3382	amdgpu_bo_evict_vram(adev);
3386	3383

drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c+2−1

@@ -85,9 +85,10 @@ · 84 unmodified lines
85	85	* - 3.34.0 - Non-DC can flip correctly between buffers with different pitches
86	86	* - 3.35.0 - Add drm_amdgpu_info_device::tcc_disabled_mask
87	87	* - 3.36.0 - Allow reading more status registers on si/cik
	88	+ * - 3.37.0 - L2 is invalidated before SDMA IBs, needed for correctness
88	89	*/
89	90	#define KMS_DRIVER_MAJOR 3
90		−#define KMS_DRIVER_MINOR 36
	91	+#define KMS_DRIVER_MINOR 37
91	92	#define KMS_DRIVER_PATCHLEVEL 0
92	93
93	94	int amdgpu_vram_limit = 0;

drivers/gpu/drm/amd/amdgpu/navi10_sdma_pkt_open.h+16−0

@@ -73,6 +73,22 @@ · 72 unmodified lines
73	73	#define SDMA_OP_AQL_COPY 0
74	74	#define SDMA_OP_AQL_BARRIER_OR 0
75	75
	76	+#define SDMA_GCR_RANGE_IS_PA (1 << 18)
	77	+#define SDMA_GCR_SEQ(x) (((x) & 0x3) << 16)
	78	+#define SDMA_GCR_GL2_WB (1 << 15)
	79	+#define SDMA_GCR_GL2_INV (1 << 14)
	80	+#define SDMA_GCR_GL2_DISCARD (1 << 13)
	81	+#define SDMA_GCR_GL2_RANGE(x) (((x) & 0x3) << 11)
	82	+#define SDMA_GCR_GL2_US (1 << 10)
	83	+#define SDMA_GCR_GL1_INV (1 << 9)
	84	+#define SDMA_GCR_GLV_INV (1 << 8)
	85	+#define SDMA_GCR_GLK_INV (1 << 7)
	86	+#define SDMA_GCR_GLK_WB (1 << 6)
	87	+#define SDMA_GCR_GLM_INV (1 << 5)
	88	+#define SDMA_GCR_GLM_WB (1 << 4)
	89	+#define SDMA_GCR_GL1_RANGE(x) (((x) & 0x3) << 2)
	90	+#define SDMA_GCR_GLI_INV(x) (((x) & 0x3) << 0)
	91	+
76	92	/define for op field/
77	93	#define SDMA_PKT_HEADER_op_offset 0
78	94	#define SDMA_PKT_HEADER_op_mask 0x000000FF

drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c+13−1

@@ -382,6 +382,18 @@ static void sdma_v5_0_ring_emit_ib(struct amdgpu_ring *ring, · 381 unmodified lines
382	382	unsigned vmid = AMDGPU_JOB_GET_VMID(job);
383	383	uint64_t csa_mc_addr = amdgpu_sdma_get_csa_mc_addr(ring, vmid);
384	384
	385	+ /* Invalidate L2, because if we don't do it, we might get stale cache
	386	+ * lines from previous IBs.
	387	+ */
	388	+ amdgpu_ring_write(ring, SDMA_PKT_HEADER_OP(SDMA_OP_GCR_REQ));
	389	+ amdgpu_ring_write(ring, 0);
	390	+ amdgpu_ring_write(ring, (SDMA_GCR_GL2_INV \|
	391	+ SDMA_GCR_GL2_WB \|
	392	+ SDMA_GCR_GLM_INV \|
	393	+ SDMA_GCR_GLM_WB) << 16);
	394	+ amdgpu_ring_write(ring, 0xffffff80);
	395	+ amdgpu_ring_write(ring, 0xffff);
	396	+
385	397	/* An IB packet must end on a 8 DW boundary--the next dword
386	398	* must be on a 8-dword boundary. Our IB packet below is 6
387	399	* dwords long, thus add x number of NOPs, such that, in
@@ -1595,7 +1607,7 @@ static const struct amdgpu_ring_funcs sdma_v5_0_ring_funcs = { · 1207 unmodified lines
1595	1607	SOC15_FLUSH_GPU_TLB_NUM_WREG * 3 +
1596	1608	SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 6 * 2 +
1597	1609	10 + 10 + 10, /* sdma_v5_0_ring_emit_fence x3 for user fence, vm fence */
1598		− .emit_ib_size = 7 + 6, /* sdma_v5_0_ring_emit_ib */
	1610	+ .emit_ib_size = 5 + 7 + 6, /* sdma_v5_0_ring_emit_ib */
1599	1611	.emit_ib = sdma_v5_0_ring_emit_ib,
1600	1612	.emit_fence = sdma_v5_0_ring_emit_fence,
1601	1613	.emit_pipeline_sync = sdma_v5_0_ring_emit_pipeline_sync,

drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c+52−17

@@ -2008,17 +2008,22 @@ void amdgpu_dm_update_connector_after_detect( · 2007 unmodified lines
2008	2008	dc_sink_retain(aconnector->dc_sink);
2009	2009	if (sink->dc_edid.length == 0) {
2010	2010	aconnector->edid = NULL;
2011		− drm_dp_cec_unset_edid(&aconnector->dm_dp_aux.aux);
	2011	+ if (aconnector->dc_link->aux_mode) {
	2012	+ drm_dp_cec_unset_edid(
	2013	+ &aconnector->dm_dp_aux.aux);
	2014	+ }
2012	2015	} else {
2013	2016	aconnector->edid =
2014		− (struct edid *) sink->dc_edid.raw_edid;
2015		−
	2017	+ (struct edid *)sink->dc_edid.raw_edid;
2016	2018
2017	2019	drm_connector_update_edid_property(connector,
2018		− aconnector->edid);
2019		− drm_dp_cec_set_edid(&aconnector->dm_dp_aux.aux,
2020		− aconnector->edid);
	2020	+ aconnector->edid);
	2021	+
	2022	+ if (aconnector->dc_link->aux_mode)
	2023	+ drm_dp_cec_set_edid(&aconnector->dm_dp_aux.aux,
	2024	+ aconnector->edid);
2021	2025	}
	2026	+
2022	2027	amdgpu_dm_update_freesync_caps(connector, aconnector->edid);
2023	2028	update_connector_ext_caps(aconnector);
2024	2029	} else {
@@ -3340,7 +3345,8 @@ fill_plane_dcc_attributes(struct amdgpu_device *adev, · 1315 unmodified lines
3340	3345	const union dc_tiling_info *tiling_info,
3341	3346	const uint64_t info,
3342	3347	struct dc_plane_dcc_param *dcc,
3343		− struct dc_plane_address *address)
	3348	+ struct dc_plane_address *address,
	3349	+ bool force_disable_dcc)
3344	3350	{
3345	3351	struct dc *dc = adev->dm.dc;
3346	3352	struct dc_dcc_surface_param input;
@@ -3352,6 +3358,9 @@ fill_plane_dcc_attributes(struct amdgpu_device *adev, · 5 unmodified lines
3352	3358	memset(&input, 0, sizeof(input));
3353	3359	memset(&output, 0, sizeof(output));
3354	3360
	3361	+ if (force_disable_dcc)
	3362	+ return 0;
	3363	+
3355	3364	if (!offset)
3356	3365	return 0;
3357	3366
@@ -3401,7 +3410,8 @@ fill_plane_buffer_attributes(struct amdgpu_device *adev, · 43 unmodified lines
3401	3410	union dc_tiling_info *tiling_info,
3402	3411	struct plane_size *plane_size,
3403	3412	struct dc_plane_dcc_param *dcc,
3404		− struct dc_plane_address *address)
	3413	+ struct dc_plane_address *address,
	3414	+ bool force_disable_dcc)
3405	3415	{
3406	3416	const struct drm_framebuffer *fb = &afb->base;
3407	3417	int ret;
@@ -3507,7 +3517,8 @@ fill_plane_buffer_attributes(struct amdgpu_device *adev, · 99 unmodified lines
3507	3517
3508	3518	ret = fill_plane_dcc_attributes(adev, afb, format, rotation,
3509	3519	plane_size, tiling_info,
3510		− tiling_flags, dcc, address);
	3520	+ tiling_flags, dcc, address,
	3521	+ force_disable_dcc);
3511	3522	if (ret)
3512	3523	return ret;
3513	3524	}
@@ -3599,7 +3610,8 @@ fill_dc_plane_info_and_addr(struct amdgpu_device *adev, · 85 unmodified lines
3599	3610	const struct drm_plane_state *plane_state,
3600	3611	const uint64_t tiling_flags,
3601	3612	struct dc_plane_info *plane_info,
3602		− struct dc_plane_address *address)
	3613	+ struct dc_plane_address *address,
	3614	+ bool force_disable_dcc)
3603	3615	{
3604	3616	const struct drm_framebuffer *fb = plane_state->fb;
3605	3617	const struct amdgpu_framebuffer *afb =
@@ -3681,7 +3693,8 @@ fill_dc_plane_info_and_addr(struct amdgpu_device *adev, · 75 unmodified lines
3681	3693	plane_info->rotation, tiling_flags,
3682	3694	&plane_info->tiling_info,
3683	3695	&plane_info->plane_size,
3684		− &plane_info->dcc, address);
	3696	+ &plane_info->dcc, address,
	3697	+ force_disable_dcc);
3685	3698	if (ret)
3686	3699	return ret;
3687	3700
@@ -3704,6 +3717,7 @@ static int fill_dc_plane_attributes(struct amdgpu_device *adev, · 16 unmodified lines
3704	3717	struct dc_plane_info plane_info;
3705	3718	uint64_t tiling_flags;
3706	3719	int ret;
	3720	+ bool force_disable_dcc = false;
3707	3721
3708	3722	ret = fill_dc_scaling_info(plane_state, &scaling_info);
3709	3723	if (ret)
@@ -3718,9 +3732,11 @@ static int fill_dc_plane_attributes(struct amdgpu_device *adev, · 8 unmodified lines
3718	3732	if (ret)
3719	3733	return ret;
3720	3734
	3735	+ force_disable_dcc = adev->asic_type == CHIP_RAVEN && adev->in_suspend;
3721	3736	ret = fill_dc_plane_info_and_addr(adev, plane_state, tiling_flags,
3722	3737	&plane_info,
3723		− &dc_plane_state->address);
	3738	+ &dc_plane_state->address,
	3739	+ force_disable_dcc);
3724	3740	if (ret)
3725	3741	return ret;
3726	3742
@@ -4664,6 +4680,7 @@ static void amdgpu_dm_connector_destroy(struct drm_connector *connector) · 937 unmodified lines
4664	4680	i2c_del_adapter(&aconnector->i2c->base);
4665	4681	kfree(aconnector->i2c);
4666	4682	}
	4683	+ kfree(aconnector->dm_dp_aux.aux.name);
4667	4684
4668	4685	kfree(connector);
4669	4686	}
@@ -4723,10 +4740,19 @@ amdgpu_dm_connector_atomic_duplicate_state(struct drm_connector *connector) · 53 unmodified lines
4723	4740	static int
4724	4741	amdgpu_dm_connector_late_register(struct drm_connector *connector)
4725	4742	{
4726		−#if defined(CONFIG_DEBUG_FS)
4727	4743	struct amdgpu_dm_connector *amdgpu_dm_connector =
4728	4744	to_amdgpu_dm_connector(connector);
	4745	+ int r;
4729	4746
	4747	+ if ((connector->connector_type == DRM_MODE_CONNECTOR_DisplayPort) \|\|
	4748	+ (connector->connector_type == DRM_MODE_CONNECTOR_eDP)) {
	4749	+ amdgpu_dm_connector->dm_dp_aux.aux.dev = connector->kdev;
	4750	+ r = drm_dp_aux_register(&amdgpu_dm_connector->dm_dp_aux.aux);
	4751	+ if (r)
	4752	+ return r;
	4753	+ }
	4754	+
	4755	+#if defined(CONFIG_DEBUG_FS)
4730	4756	connector_debugfs_init(amdgpu_dm_connector);
4731	4757	#endif
4732	4758
@@ -5332,6 +5358,7 @@ static int dm_plane_helper_prepare_fb(struct drm_plane *plane, · 599 unmodified lines
5332	5358	uint64_t tiling_flags;
5333	5359	uint32_t domain;
5334	5360	int r;
	5361	+ bool force_disable_dcc = false;
5335	5362
5336	5363	dm_plane_state_old = to_dm_plane_state(plane->state);
5337	5364	dm_plane_state_new = to_dm_plane_state(new_state);
@@ -5390,11 +5417,13 @@ static int dm_plane_helper_prepare_fb(struct drm_plane *plane, · 52 unmodified lines
5390	5417	dm_plane_state_old->dc_state != dm_plane_state_new->dc_state) {
5391	5418	struct dc_plane_state *plane_state = dm_plane_state_new->dc_state;
5392	5419
	5420	+ force_disable_dcc = adev->asic_type == CHIP_RAVEN && adev->in_suspend;
5393	5421	fill_plane_buffer_attributes(
5394	5422	adev, afb, plane_state->format, plane_state->rotation,
5395	5423	tiling_flags, &plane_state->tiling_info,
5396	5424	&plane_state->plane_size, &plane_state->dcc,
5397		− &plane_state->address);
	5425	+ &plane_state->address,
	5426	+ force_disable_dcc);
5398	5427	}
5399	5428
5400	5429	return 0;
@@ -6092,7 +6121,7 @@ static int amdgpu_dm_connector_init(struct amdgpu_display_manager *dm, · 691 unmodified lines
6092	6121
6093	6122	if (connector_type == DRM_MODE_CONNECTOR_DisplayPort
6094	6123	\|\| connector_type == DRM_MODE_CONNECTOR_eDP)
6095		− amdgpu_dm_initialize_dp_connector(dm, aconnector);
	6124	+ amdgpu_dm_initialize_dp_connector(dm, aconnector, link->link_index);
6096	6125
6097	6126	out_free:
6098	6127	if (res) {
@@ -6666,7 +6695,12 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, · 567 unmodified lines
6666	6695	fill_dc_plane_info_and_addr(
6667	6696	dm->adev, new_plane_state, tiling_flags,
6668	6697	&bundle->plane_infos[planes_count],
6669		− &bundle->flip_addrs[planes_count].address);
	6698	+ &bundle->flip_addrs[planes_count].address,
	6699	+ false);
	6700	+
	6701	+ DRM_DEBUG_DRIVER("plane: id=%d dcc_en=%d\n",
	6702	+ new_plane_state->plane->index,
	6703	+ bundle->plane_infos[planes_count].dcc.enable);
6670	6704
6671	6705	bundle->surface_updates[planes_count].plane_info =
6672	6706	&bundle->plane_infos[planes_count];
@@ -8086,7 +8120,8 @@ dm_determine_update_type_for_commit(struct amdgpu_display_manager *dm, · 1413 unmodified lines
8086	8120	ret = fill_dc_plane_info_and_addr(
8087	8121	dm->adev, new_plane_state, tiling_flags,
8088	8122	plane_info,
8089		− &flip_addr->address);
	8123	+ &flip_addr->address,
	8124	+ false);
8090	8125	if (ret)
8091	8126	goto cleanup;
8092	8127

drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c+9−6

@@ -156,16 +156,16 @@ amdgpu_dm_mst_connector_late_register(struct drm_connector *connector) · 155 unmodified lines
156	156	to_amdgpu_dm_connector(connector);
157	157	int r;
158	158
159		− amdgpu_dm_connector->dm_dp_aux.aux.dev = connector->kdev;
160		− r = drm_dp_aux_register(&amdgpu_dm_connector->dm_dp_aux.aux);
161		− if (r)
	159	+ r = drm_dp_mst_connector_late_register(connector,
	160	+ amdgpu_dm_connector->port);
	161	+ if (r < 0)
162	162	return r;
163	163
164	164	#if defined(CONFIG_DEBUG_FS)
165	165	connector_debugfs_init(amdgpu_dm_connector);
166	166	#endif
167	167
168		− return r;
	168	+ return 0;
169	169	}
170	170
171	171	static void
@@ -472,9 +472,12 @@ static const struct drm_dp_mst_topology_cbs dm_mst_cbs = { · 300 unmodified lines
472	472	};
473	473
474	474	void amdgpu_dm_initialize_dp_connector(struct amdgpu_display_manager *dm,
475		− struct amdgpu_dm_connector *aconnector)
	475	+ struct amdgpu_dm_connector *aconnector,
	476	+ int link_index)
476	477	{
477		− aconnector->dm_dp_aux.aux.name = "dmdc";
	478	+ aconnector->dm_dp_aux.aux.name =
	479	+ kasprintf(GFP_KERNEL, "AMDGPU DM aux hw bus %d",
	480	+ link_index);
478	481	aconnector->dm_dp_aux.aux.transfer = dm_dp_aux_transfer;
479	482	aconnector->dm_dp_aux.ddc_service = aconnector->dc_link->ddc;
480	483

drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h+2−1

@@ -32,7 +32,8 @@ struct amdgpu_dm_connector; · 31 unmodified lines
32	32	int dm_mst_get_pbn_divider(struct dc_link *link);
33	33
34	34	void amdgpu_dm_initialize_dp_connector(struct amdgpu_display_manager *dm,
35		− struct amdgpu_dm_connector *aconnector);
	35	+ struct amdgpu_dm_connector *aconnector,
	36	+ int link_index);
36	37
37	38	#if defined(CONFIG_DRM_AMD_DC_DCN)
38	39	bool compute_mst_dsc_configs_for_state(struct drm_atomic_state *state,

drivers/gpu/drm/amd/display/dc/core/dc.c+2−3

@@ -834,11 +834,10 @@ static void disable_dangling_plane(struct dc dc, struct dc_state context) · 833 unmodified lines
834	834	static void wait_for_no_pipes_pending(struct dc dc, struct dc_state context)
835	835	{
836	836	int i;
837		− int count = 0;
838		− struct pipe_ctx *pipe;
839	837	PERF_TRACE();
840	838	for (i = 0; i < MAX_PIPES; i++) {
841		− pipe = &context->res_ctx.pipe_ctx[i];
	839	+ int count = 0;
	840	+ struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i];
842	841
843	842	if (!pipe->plane_state)
844	843	continue;

drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c+27−0

@@ -2908,6 +2908,12 @@ bool dc_link_handle_hpd_rx_irq(struct dc_link link, union hpd_irq_data out_hpd · 2907 unmodified lines
2908	2908	sizeof(hpd_irq_dpcd_data),
2909	2909	"Status: ");
2910	2910
	2911	+ for (i = 0; i < MAX_PIPES; i++) {
	2912	+ pipe_ctx = &link->dc->current_state->res_ctx.pipe_ctx[i];
	2913	+ if (pipe_ctx && pipe_ctx->stream && pipe_ctx->stream->link == link)
	2914	+ link->dc->hwss.blank_stream(pipe_ctx);
	2915	+ }
	2916	+
2911	2917	for (i = 0; i < MAX_PIPES; i++) {
2912	2918	pipe_ctx = &link->dc->current_state->res_ctx.pipe_ctx[i];
2913	2919	if (pipe_ctx && pipe_ctx->stream && pipe_ctx->stream->link == link)
@@ -2927,6 +2933,12 @@ bool dc_link_handle_hpd_rx_irq(struct dc_link link, union hpd_irq_data out_hpd · 13 unmodified lines
2927	2933	if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST)
2928	2934	dc_link_reallocate_mst_payload(link);
2929	2935
	2936	+ for (i = 0; i < MAX_PIPES; i++) {
	2937	+ pipe_ctx = &link->dc->current_state->res_ctx.pipe_ctx[i];
	2938	+ if (pipe_ctx && pipe_ctx->stream && pipe_ctx->stream->link == link)
	2939	+ link->dc->hwss.unblank_stream(pipe_ctx, &previous_link_settings);
	2940	+ }
	2941	+
2930	2942	status = false;
2931	2943	if (out_link_loss)
2932	2944	*out_link_loss = true;
@@ -4227,6 +4239,21 @@ void dp_set_fec_enable(struct dc_link *link, bool enable) · 1294 unmodified lines
4227	4239	void dpcd_set_source_specific_data(struct dc_link *link)
4228	4240	{
4229	4241	const uint32_t post_oui_delay = 30; // 30ms
	4242	+ uint8_t dspc = 0;
	4243	+ enum dc_status ret = DC_ERROR_UNEXPECTED;
	4244	+
	4245	+ ret = core_link_read_dpcd(link, DP_DOWN_STREAM_PORT_COUNT, &dspc,
	4246	+ sizeof(dspc));
	4247	+
	4248	+ if (ret != DC_OK) {
	4249	+ DC_LOG_ERROR("Error in DP aux read transaction,"
	4250	+ " not writing source specific data\n");
	4251	+ return;
	4252	+ }
	4253	+
	4254	+ /* Return if OUI unsupported */
	4255	+ if (!(dspc & DP_OUI_SUPPORT))
	4256	+ return;
4230	4257
4231	4258	if (!link->dc->vendor_signature.is_valid) {
4232	4259	struct dpcd_amd_signature amd_signature;

drivers/gpu/drm/amd/display/dc/core/dc_stream.c+4−36

@@ -231,34 +231,6 @@ struct dc_stream_status *dc_stream_get_status( · 230 unmodified lines
231	231	return dc_stream_get_status_from_state(dc->current_state, stream);
232	232	}
233	233
234		−static void delay_cursor_until_vupdate(struct pipe_ctx pipe_ctx, struct dc dc)
235		−{
236		−#if defined(CONFIG_DRM_AMD_DC_DCN)
237		− unsigned int vupdate_line;
238		− unsigned int lines_to_vupdate, us_to_vupdate, vpos, nvpos;
239		− struct dc_stream_state *stream = pipe_ctx->stream;
240		− unsigned int us_per_line;
241		−
242		− if (stream->ctx->asic_id.chip_family == FAMILY_RV &&
243		− ASICREV_IS_RAVEN(stream->ctx->asic_id.hw_internal_rev)) {
244		−
245		− vupdate_line = dc->hwss.get_vupdate_offset_from_vsync(pipe_ctx);
246		− if (!dc_stream_get_crtc_position(dc, &stream, 1, &vpos, &nvpos))
247		− return;
248		−
249		− if (vpos >= vupdate_line)
250		− return;
251		−
252		− us_per_line = stream->timing.h_total * 10000 / stream->timing.pix_clk_100hz;
253		− lines_to_vupdate = vupdate_line - vpos;
254		− us_to_vupdate = lines_to_vupdate * us_per_line;
255		−
256		− /* 70 us is a conservative estimate of cursor update time*/
257		− if (us_to_vupdate < 70)
258		− udelay(us_to_vupdate);
259		− }
260		−#endif
261		−}
262	234
263	235	/**
264	236	* dc_stream_set_cursor_attributes() - Update cursor attributes and set cursor surface address
@@ -298,9 +270,7 @@ bool dc_stream_set_cursor_attributes( · 33 unmodified lines
298	270
299	271	if (!pipe_to_program) {
300	272	pipe_to_program = pipe_ctx;
301		−
302		− delay_cursor_until_vupdate(pipe_ctx, dc);
303		− dc->hwss.pipe_control_lock(dc, pipe_to_program, true);
	273	+ dc->hwss.cursor_lock(dc, pipe_to_program, true);
304	274	}
305	275
306	276	dc->hwss.set_cursor_attribute(pipe_ctx);
@@ -309,7 +279,7 @@ bool dc_stream_set_cursor_attributes( · 2 unmodified lines
309	279	}
310	280
311	281	if (pipe_to_program)
312		− dc->hwss.pipe_control_lock(dc, pipe_to_program, false);
	282	+ dc->hwss.cursor_lock(dc, pipe_to_program, false);
313	283
314	284	return true;
315	285	}
@@ -349,16 +319,14 @@ bool dc_stream_set_cursor_position( · 33 unmodified lines
349	319
350	320	if (!pipe_to_program) {
351	321	pipe_to_program = pipe_ctx;
352		−
353		− delay_cursor_until_vupdate(pipe_ctx, dc);
354		− dc->hwss.pipe_control_lock(dc, pipe_to_program, true);
	322	+ dc->hwss.cursor_lock(dc, pipe_to_program, true);
355	323	}
356	324
357	325	dc->hwss.set_cursor_position(pipe_ctx);
358	326	}
359	327
360	328	if (pipe_to_program)
361		− dc->hwss.pipe_control_lock(dc, pipe_to_program, false);
	329	+ dc->hwss.cursor_lock(dc, pipe_to_program, false);
362	330
363	331	return true;
364	332	}

drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c+1−0

@@ -2757,6 +2757,7 @@ static const struct hw_sequencer_funcs dce110_funcs = { · 2756 unmodified lines
2757	2757	.disable_plane = dce110_power_down_fe,
2758	2758	.pipe_control_lock = dce_pipe_control_lock,
2759	2759	.interdependent_update_lock = NULL,
	2760	+ .cursor_lock = dce_pipe_control_lock,
2760	2761	.prepare_bandwidth = dce110_prepare_bandwidth,
2761	2762	.optimize_bandwidth = dce110_optimize_bandwidth,
2762	2763	.set_drr = set_drr,

drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c+10−0

@@ -1625,6 +1625,16 @@ void dcn10_pipe_control_lock( · 1624 unmodified lines
1625	1625	hws->funcs.verify_allow_pstate_change_high(dc);
1626	1626	}
1627	1627
	1628	+void dcn10_cursor_lock(struct dc dc, struct pipe_ctx pipe, bool lock)
	1629	+{
	1630	+ /* cursor lock is per MPCC tree, so only need to lock one pipe per stream */
	1631	+ if (!pipe \|\| pipe->top_pipe)
	1632	+ return;
	1633	+
	1634	+ dc->res_pool->mpc->funcs->cursor_lock(dc->res_pool->mpc,
	1635	+ pipe->stream_res.opp->inst, lock);
	1636	+}
	1637	+
1628	1638	static bool wait_for_reset_trigger_to_occur(
1629	1639	struct dc_context *dc_ctx,
1630	1640	struct timing_generator *tg)

drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h+1−0

@@ -49,6 +49,7 @@ void dcn10_pipe_control_lock( · 48 unmodified lines
49	49	struct dc *dc,
50	50	struct pipe_ctx *pipe,
51	51	bool lock);
	52	+void dcn10_cursor_lock(struct dc dc, struct pipe_ctx pipe, bool lock);
52	53	void dcn10_blank_pixel_data(
53	54	struct dc *dc,
54	55	struct pipe_ctx *pipe_ctx,

drivers/gpu/drm/amd/display/dc/dcn10/dcn10_init.c+1−0

@@ -50,6 +50,7 @@ static const struct hw_sequencer_funcs dcn10_funcs = { · 49 unmodified lines
50	50	.disable_audio_stream = dce110_disable_audio_stream,
51	51	.disable_plane = dcn10_disable_plane,
52	52	.pipe_control_lock = dcn10_pipe_control_lock,
	53	+ .cursor_lock = dcn10_cursor_lock,
53	54	.interdependent_update_lock = dcn10_lock_all_pipes,
54	55	.prepare_bandwidth = dcn10_prepare_bandwidth,
55	56	.optimize_bandwidth = dcn10_optimize_bandwidth,

drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.c+15−0

@@ -223,6 +223,9 @@ struct mpcc *mpc1_insert_plane( · 222 unmodified lines
223	223	REG_SET(MPCC_TOP_SEL[mpcc_id], 0, MPCC_TOP_SEL, dpp_id);
224	224	REG_SET(MPCC_OPP_ID[mpcc_id], 0, MPCC_OPP_ID, tree->opp_id);
225	225
	226	+ /* Configure VUPDATE lock set for this MPCC to map to the OPP */
	227	+ REG_SET(MPCC_UPDATE_LOCK_SEL[mpcc_id], 0, MPCC_UPDATE_LOCK_SEL, tree->opp_id);
	228	+
226	229	/* update mpc tree mux setting */
227	230	if (tree->opp_list == insert_above_mpcc) {
228	231	/* insert the toppest mpcc */
@@ -318,6 +321,7 @@ void mpc1_remove_mpcc( · 89 unmodified lines
318	321	REG_SET(MPCC_TOP_SEL[mpcc_id], 0, MPCC_TOP_SEL, 0xf);
319	322	REG_SET(MPCC_BOT_SEL[mpcc_id], 0, MPCC_BOT_SEL, 0xf);
320	323	REG_SET(MPCC_OPP_ID[mpcc_id], 0, MPCC_OPP_ID, 0xf);
	324	+ REG_SET(MPCC_UPDATE_LOCK_SEL[mpcc_id], 0, MPCC_UPDATE_LOCK_SEL, 0xf);
321	325
322	326	/* mark this mpcc as not in use */
323	327	mpc10->mpcc_in_use_mask &= ~(1 << mpcc_id);
@@ -328,6 +332,7 @@ void mpc1_remove_mpcc( · 4 unmodified lines
328	332	REG_SET(MPCC_TOP_SEL[mpcc_id], 0, MPCC_TOP_SEL, 0xf);
329	333	REG_SET(MPCC_BOT_SEL[mpcc_id], 0, MPCC_BOT_SEL, 0xf);
330	334	REG_SET(MPCC_OPP_ID[mpcc_id], 0, MPCC_OPP_ID, 0xf);
	335	+ REG_SET(MPCC_UPDATE_LOCK_SEL[mpcc_id], 0, MPCC_UPDATE_LOCK_SEL, 0xf);
331	336	}
332	337	}
333	338
@@ -361,6 +366,7 @@ void mpc1_mpc_init(struct mpc *mpc) · 27 unmodified lines
361	366	REG_SET(MPCC_TOP_SEL[mpcc_id], 0, MPCC_TOP_SEL, 0xf);
362	367	REG_SET(MPCC_BOT_SEL[mpcc_id], 0, MPCC_BOT_SEL, 0xf);
363	368	REG_SET(MPCC_OPP_ID[mpcc_id], 0, MPCC_OPP_ID, 0xf);
	369	+ REG_SET(MPCC_UPDATE_LOCK_SEL[mpcc_id], 0, MPCC_UPDATE_LOCK_SEL, 0xf);
364	370
365	371	mpc1_init_mpcc(&(mpc->mpcc_array[mpcc_id]), mpcc_id);
366	372	}
@@ -381,6 +387,7 @@ void mpc1_mpc_init_single_inst(struct mpc *mpc, unsigned int mpcc_id) · 14 unmodified lines
381	387	REG_SET(MPCC_TOP_SEL[mpcc_id], 0, MPCC_TOP_SEL, 0xf);
382	388	REG_SET(MPCC_BOT_SEL[mpcc_id], 0, MPCC_BOT_SEL, 0xf);
383	389	REG_SET(MPCC_OPP_ID[mpcc_id], 0, MPCC_OPP_ID, 0xf);
	390	+ REG_SET(MPCC_UPDATE_LOCK_SEL[mpcc_id], 0, MPCC_UPDATE_LOCK_SEL, 0xf);
384	391
385	392	mpc1_init_mpcc(&(mpc->mpcc_array[mpcc_id]), mpcc_id);
386	393
@@ -453,6 +460,13 @@ void mpc1_read_mpcc_state( · 66 unmodified lines
453	460	MPCC_BUSY, &s->busy);
454	461	}
455	462
	463	+void mpc1_cursor_lock(struct mpc *mpc, int opp_id, bool lock)
	464	+{
	465	+ struct dcn10_mpc *mpc10 = TO_DCN10_MPC(mpc);
	466	+
	467	+ REG_SET(CUR[opp_id], 0, CUR_VUPDATE_LOCK_SET, lock ? 1 : 0);
	468	+}
	469	+
456	470	static const struct mpc_funcs dcn10_mpc_funcs = {
457	471	.read_mpcc_state = mpc1_read_mpcc_state,
458	472	.insert_plane = mpc1_insert_plane,
@@ -464,6 +478,7 @@ static const struct mpc_funcs dcn10_mpc_funcs = { · 5 unmodified lines
464	478	.assert_mpcc_idle_before_connect = mpc1_assert_mpcc_idle_before_connect,
465	479	.init_mpcc_list_from_hw = mpc1_init_mpcc_list_from_hw,
466	480	.update_blending = mpc1_update_blending,
	481	+ .cursor_lock = mpc1_cursor_lock,
467	482	.set_denorm = NULL,
468	483	.set_denorm_clamp = NULL,
469	484	.set_output_csc = NULL,

drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.h+14−6

@@ -39,11 +39,12 @@ · 38 unmodified lines
39	39	SRII(MPCC_BG_G_Y, MPCC, inst),\
40	40	SRII(MPCC_BG_R_CR, MPCC, inst),\
41	41	SRII(MPCC_BG_B_CB, MPCC, inst),\
42		− SRII(MPCC_BG_B_CB, MPCC, inst),\
43		− SRII(MPCC_SM_CONTROL, MPCC, inst)
	42	+ SRII(MPCC_SM_CONTROL, MPCC, inst),\
	43	+ SRII(MPCC_UPDATE_LOCK_SEL, MPCC, inst)
44	44
45	45	#define MPC_OUT_MUX_COMMON_REG_LIST_DCN1_0(inst) \
46		− SRII(MUX, MPC_OUT, inst)
	46	+ SRII(MUX, MPC_OUT, inst),\
	47	+ VUPDATE_SRII(CUR, VUPDATE_LOCK_SET, inst)
47	48
48	49	#define MPC_COMMON_REG_VARIABLE_LIST \
49	50	uint32_t MPCC_TOP_SEL[MAX_MPCC]; \
@@ -55,7 +56,9 @@ · 5 unmodified lines
55	56	uint32_t MPCC_BG_R_CR[MAX_MPCC]; \
56	57	uint32_t MPCC_BG_B_CB[MAX_MPCC]; \
57	58	uint32_t MPCC_SM_CONTROL[MAX_MPCC]; \
58		− uint32_t MUX[MAX_OPP];
	59	+ uint32_t MUX[MAX_OPP]; \
	60	+ uint32_t MPCC_UPDATE_LOCK_SEL[MAX_MPCC]; \
	61	+ uint32_t CUR[MAX_OPP];
59	62
60	63	#define MPC_COMMON_MASK_SH_LIST_DCN1_0(mask_sh)\
61	64	SF(MPCC0_MPCC_TOP_SEL, MPCC_TOP_SEL, mask_sh),\
@@ -78,7 +81,8 @@ · 16 unmodified lines
78	81	SF(MPCC0_MPCC_SM_CONTROL, MPCC_SM_FIELD_ALT, mask_sh),\
79	82	SF(MPCC0_MPCC_SM_CONTROL, MPCC_SM_FORCE_NEXT_FRAME_POL, mask_sh),\
80	83	SF(MPCC0_MPCC_SM_CONTROL, MPCC_SM_FORCE_NEXT_TOP_POL, mask_sh),\
81		− SF(MPC_OUT0_MUX, MPC_OUT_MUX, mask_sh)
	84	+ SF(MPC_OUT0_MUX, MPC_OUT_MUX, mask_sh),\
	85	+ SF(MPCC0_MPCC_UPDATE_LOCK_SEL, MPCC_UPDATE_LOCK_SEL, mask_sh)
82	86
83	87	#define MPC_REG_FIELD_LIST(type) \
84	88	type MPCC_TOP_SEL;\
@@ -101,7 +105,9 @@ · 16 unmodified lines
101	105	type MPCC_SM_FIELD_ALT;\
102	106	type MPCC_SM_FORCE_NEXT_FRAME_POL;\
103	107	type MPCC_SM_FORCE_NEXT_TOP_POL;\
104		− type MPC_OUT_MUX;
	108	+ type MPC_OUT_MUX;\
	109	+ type MPCC_UPDATE_LOCK_SEL;\
	110	+ type CUR_VUPDATE_LOCK_SET;
105	111
106	112	struct dcn_mpc_registers {
107	113	MPC_COMMON_REG_VARIABLE_LIST
@@ -192,4 +198,6 @@ void mpc1_read_mpcc_state( · 84 unmodified lines
192	198	int mpcc_inst,
193	199	struct mpcc_state *s);
194	200
	201	+void mpc1_cursor_lock(struct mpc *mpc, int opp_id, bool lock);
	202	+
195	203	#endif

drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c+12−2

@@ -181,6 +181,14 @@ enum dcn10_clk_src_array_id { · 180 unmodified lines
181	181	.reg_name[id] = BASE(mm ## block ## id ## _ ## reg_name ## _BASE_IDX) + \
182	182	mm ## block ## id ## _ ## reg_name
183	183
	184	+#define VUPDATE_SRII(reg_name, block, id)\
	185	+ .reg_name[id] = BASE(mm ## reg_name ## 0 ## _ ## block ## id ## _BASE_IDX) + \
	186	+ mm ## reg_name ## 0 ## _ ## block ## id
	187	+
	188	+/* set field/register/bitfield name */
	189	+#define SFRB(field_name, reg_name, bitfield, post_fix)\
	190	+ .field_name = reg_name ## __ ## bitfield ## post_fix
	191	+
184	192	/* NBIO */
185	193	#define NBIO_BASE_INNER(seg) \
186	194	NBIF_BASE__INST0_SEG ## seg
@@ -419,11 +427,13 @@ static const struct dcn_mpc_registers mpc_regs = { · 232 unmodified lines
419	427	};
420	428
421	429	static const struct dcn_mpc_shift mpc_shift = {
422		− MPC_COMMON_MASK_SH_LIST_DCN1_0(__SHIFT)
	430	+ MPC_COMMON_MASK_SH_LIST_DCN1_0(__SHIFT),\
	431	+ SFRB(CUR_VUPDATE_LOCK_SET, CUR0_VUPDATE_LOCK_SET0, CUR0_VUPDATE_LOCK_SET, __SHIFT)
423	432	};
424	433
425	434	static const struct dcn_mpc_mask mpc_mask = {
426		− MPC_COMMON_MASK_SH_LIST_DCN1_0(_MASK),
	435	+ MPC_COMMON_MASK_SH_LIST_DCN1_0(_MASK),\
	436	+ SFRB(CUR_VUPDATE_LOCK_SET, CUR0_VUPDATE_LOCK_SET0, CUR0_VUPDATE_LOCK_SET, _MASK)
427	437	};
428	438
429	439	#define tg_regs(id)\

drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c+2−1

@@ -2294,7 +2294,8 @@ void dcn20_fpga_init_hw(struct dc *dc) · 2293 unmodified lines
2294	2294
2295	2295	REG_UPDATE(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_REFDIV, 2);
2296	2296	REG_UPDATE(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_ENABLE, 1);
2297		− REG_WRITE(REFCLK_CNTL, 0);
	2297	+ if (REG(REFCLK_CNTL))
	2298	+ REG_WRITE(REFCLK_CNTL, 0);
2298	2299	//
2299	2300
2300	2301

drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c+1−0

@@ -52,6 +52,7 @@ static const struct hw_sequencer_funcs dcn20_funcs = { · 51 unmodified lines
52	52	.disable_plane = dcn20_disable_plane,
53	53	.pipe_control_lock = dcn20_pipe_control_lock,
54	54	.interdependent_update_lock = dcn10_lock_all_pipes,
	55	+ .cursor_lock = dcn10_cursor_lock,
55	56	.prepare_bandwidth = dcn20_prepare_bandwidth,
56	57	.optimize_bandwidth = dcn20_optimize_bandwidth,
57	58	.update_bandwidth = dcn20_update_bandwidth,

drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.c+1−0

@@ -545,6 +545,7 @@ const struct mpc_funcs dcn20_mpc_funcs = { · 544 unmodified lines
545	545	.mpc_init = mpc1_mpc_init,
546	546	.mpc_init_single_inst = mpc1_mpc_init_single_inst,
547	547	.update_blending = mpc2_update_blending,
	548	+ .cursor_lock = mpc1_cursor_lock,
548	549	.get_mpcc_for_dpp = mpc2_get_mpcc_for_dpp,
549	550	.wait_for_idle = mpc2_assert_idle_mpcc,
550	551	.assert_mpcc_idle_before_connect = mpc2_assert_mpcc_idle_before_connect,

drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.h+2−1

@@ -179,7 +179,8 @@ · 178 unmodified lines
179	179	SF(MPC_OUT0_DENORM_CLAMP_G_Y, MPC_OUT_DENORM_CLAMP_MAX_G_Y, mask_sh),\
180	180	SF(MPC_OUT0_DENORM_CLAMP_G_Y, MPC_OUT_DENORM_CLAMP_MIN_G_Y, mask_sh),\
181	181	SF(MPC_OUT0_DENORM_CLAMP_B_CB, MPC_OUT_DENORM_CLAMP_MAX_B_CB, mask_sh),\
182		− SF(MPC_OUT0_DENORM_CLAMP_B_CB, MPC_OUT_DENORM_CLAMP_MIN_B_CB, mask_sh)
	182	+ SF(MPC_OUT0_DENORM_CLAMP_B_CB, MPC_OUT_DENORM_CLAMP_MIN_B_CB, mask_sh),\
	183	+ SF(CUR_VUPDATE_LOCK_SET0, CUR_VUPDATE_LOCK_SET, mask_sh)
183	184
184	185	/*
185	186	* DCN2 MPC_OCSC debug status register:

drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c+27−8

@@ -508,6 +508,10 @@ enum dcn20_clk_src_array_id { · 507 unmodified lines
508	508	.block ## _ ## reg_name[id] = BASE(mm ## block ## id ## _ ## reg_name ## _BASE_IDX) + \
509	509	mm ## block ## id ## _ ## reg_name
510	510
	511	+#define VUPDATE_SRII(reg_name, block, id)\
	512	+ .reg_name[id] = BASE(mm ## reg_name ## _ ## block ## id ## _BASE_IDX) + \
	513	+ mm ## reg_name ## _ ## block ## id
	514	+
511	515	/* NBIO */
512	516	#define NBIO_BASE_INNER(seg) \
513	517	NBIO_BASE__INST0_SEG ## seg
@@ -3064,25 +3068,32 @@ validate_out: · 2550 unmodified lines
3064	3068	return out;
3065	3069	}
3066	3070
3067		−
3068		−bool dcn20_validate_bandwidth(struct dc dc, struct dc_state context,
3069		− bool fast_validate)
	3071	+/*
	3072	+ * This must be noinline to ensure anything that deals with FP registers
	3073	+ * is contained within this call; previously our compiling with hard-float
	3074	+ * would result in fp instructions being emitted outside of the boundaries
	3075	+ * of the DC_FP_START/END macros, which makes sense as the compiler has no
	3076	+ * idea about what is wrapped and what is not
	3077	+ *
	3078	+ * This is largely just a workaround to avoid breakage introduced with 5.6,
	3079	+ * ideally all fp-using code should be moved into its own file, only that
	3080	+ * should be compiled with hard-float, and all code exported from there
	3081	+ * should be strictly wrapped with DC_FP_START/END
	3082	+ */
	3083	+static noinline bool dcn20_validate_bandwidth_fp(struct dc *dc,
	3084	+ struct dc_state *context, bool fast_validate)
3070	3085	{
3071	3086	bool voltage_supported = false;
3072	3087	bool full_pstate_supported = false;
3073	3088	bool dummy_pstate_supported = false;
3074	3089	double p_state_latency_us;
3075	3090
3076		− DC_FP_START();
3077	3091	p_state_latency_us = context->bw_ctx.dml.soc.dram_clock_change_latency_us;
3078	3092	context->bw_ctx.dml.soc.disable_dram_clock_change_vactive_support =
3079	3093	dc->debug.disable_dram_clock_change_vactive_support;
3080	3094
3081	3095	if (fast_validate) {
3082		− voltage_supported = dcn20_validate_bandwidth_internal(dc, context, true);
3083		−
3084		− DC_FP_END();
3085		− return voltage_supported;
	3096	+ return dcn20_validate_bandwidth_internal(dc, context, true);
3086	3097	}
3087	3098
3088	3099	// Best case, we support full UCLK switch latency
@@ -3111,7 +3122,15 @@ bool dcn20_validate_bandwidth(struct dc dc, struct dc_state context, · 22 unmodified lines
3111	3122
3112	3123	restore_dml_state:
3113	3124	context->bw_ctx.dml.soc.dram_clock_change_latency_us = p_state_latency_us;
	3125	+ return voltage_supported;
	3126	+}
3114	3127
	3128	+bool dcn20_validate_bandwidth(struct dc dc, struct dc_state context,
	3129	+ bool fast_validate)
	3130	+{
	3131	+ bool voltage_supported = false;
	3132	+ DC_FP_START();
	3133	+ voltage_supported = dcn20_validate_bandwidth_fp(dc, context, fast_validate);
3115	3134	DC_FP_END();
3116	3135	return voltage_supported;
3117	3136	}

drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c+1−0

@@ -53,6 +53,7 @@ static const struct hw_sequencer_funcs dcn21_funcs = { · 52 unmodified lines
53	53	.disable_plane = dcn20_disable_plane,
54	54	.pipe_control_lock = dcn20_pipe_control_lock,
55	55	.interdependent_update_lock = dcn10_lock_all_pipes,
	56	+ .cursor_lock = dcn10_cursor_lock,
56	57	.prepare_bandwidth = dcn20_prepare_bandwidth,
57	58	.optimize_bandwidth = dcn20_optimize_bandwidth,
58	59	.update_bandwidth = dcn20_update_bandwidth,

drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c+33−44

@@ -284,7 +284,7 @@ struct _vcs_dpi_soc_bounding_box_st dcn2_1_soc = { · 283 unmodified lines
284	284	.dram_channel_width_bytes = 4,
285	285	.fabric_datapath_to_dcn_data_return_bytes = 32,
286	286	.dcn_downspread_percent = 0.5,
287		− .downspread_percent = 0.5,
	287	+ .downspread_percent = 0.38,
288	288	.dram_page_open_time_ns = 50.0,
289	289	.dram_rw_turnaround_time_ns = 17.5,
290	290	.dram_return_buffer_per_channel_bytes = 8192,
@@ -340,6 +340,10 @@ struct _vcs_dpi_soc_bounding_box_st dcn2_1_soc = { · 49 unmodified lines
340	340	.block ## _ ## reg_name[id] = BASE(mm ## block ## id ## _ ## reg_name ## _BASE_IDX) + \
341	341	mm ## block ## id ## _ ## reg_name
342	342
	343	+#define VUPDATE_SRII(reg_name, block, id)\
	344	+ .reg_name[id] = BASE(mm ## reg_name ## _ ## block ## id ## _BASE_IDX) + \
	345	+ mm ## reg_name ## _ ## block ## id
	346	+
343	347	/* NBIO */
344	348	#define NBIO_BASE_INNER(seg) \
345	349	NBIF0_BASE__INST0_SEG ## seg
@@ -1374,64 +1378,49 @@ static void update_bw_bounding_box(struct dc dc, struct clk_bw_params bw_param · 1028 unmodified lines
1374	1378	{
1375	1379	struct dcn21_resource_pool *pool = TO_DCN21_RES_POOL(dc->res_pool);
1376	1380	struct clk_limit_table *clk_table = &bw_params->clk_table;
1377		− unsigned int i, j, k;
1378		− int closest_clk_lvl;
	1381	+ struct _vcs_dpi_voltage_scaling_st clock_limits[DC__VOLTAGE_STATES];
	1382	+ unsigned int i, j, closest_clk_lvl;
1379	1383
1380	1384	// Default clock levels are used for diags, which may lead to overclocking.
1381		− if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) && !IS_DIAG_DC(dc->ctx->dce_environment)) {
	1385	+ if (!IS_DIAG_DC(dc->ctx->dce_environment)) {
1382	1386	dcn2_1_ip.max_num_otg = pool->base.res_cap->num_timing_generator;
1383	1387	dcn2_1_ip.max_num_dpp = pool->base.pipe_count;
1384	1388	dcn2_1_soc.num_chans = bw_params->num_channels;
1385	1389
1386		− /* Vmin: leave lowest DCN clocks, override with dcfclk, fclk, memclk from fuse */
1387		− dcn2_1_soc.clock_limits[0].state = 0;
1388		− dcn2_1_soc.clock_limits[0].dcfclk_mhz = clk_table->entries[0].dcfclk_mhz;
1389		− dcn2_1_soc.clock_limits[0].fabricclk_mhz = clk_table->entries[0].fclk_mhz;
1390		− dcn2_1_soc.clock_limits[0].socclk_mhz = clk_table->entries[0].socclk_mhz;
1391		− dcn2_1_soc.clock_limits[0].dram_speed_mts = clk_table->entries[0].memclk_mhz * 2;
1392		−
1393		− /*
1394		− * Other levels: find closest DCN clocks that fit the given clock limit using dcfclk
1395		− * as indicator
1396		− */
1397		−
1398		− closest_clk_lvl = -1;
1399		− /* index currently being filled */
1400		− k = 1;
1401		− for (i = 1; i < clk_table->num_entries; i++) {
1402		− /* loop backwards, skip duplicate state*/
1403		− for (j = dcn2_1_soc.num_states - 1; j >= k; j--) {
	1390	+ ASSERT(clk_table->num_entries);
	1391	+ for (i = 0; i < clk_table->num_entries; i++) {
	1392	+ /* loop backwards*/
	1393	+ for (closest_clk_lvl = 0, j = dcn2_1_soc.num_states - 1; j >= 0; j--) {
1404	1394	if ((unsigned int) dcn2_1_soc.clock_limits[j].dcfclk_mhz <= clk_table->entries[i].dcfclk_mhz) {
1405	1395	closest_clk_lvl = j;
1406	1396	break;
1407	1397	}
1408	1398	}
1409	1399
1410		− /* if found a lvl that fits, use the DCN clks from it, if not, go to next clk limit*/
1411		− if (closest_clk_lvl != -1) {
1412		− dcn2_1_soc.clock_limits[k].state = i;
1413		− dcn2_1_soc.clock_limits[k].dcfclk_mhz = clk_table->entries[i].dcfclk_mhz;
1414		− dcn2_1_soc.clock_limits[k].fabricclk_mhz = clk_table->entries[i].fclk_mhz;
1415		− dcn2_1_soc.clock_limits[k].socclk_mhz = clk_table->entries[i].socclk_mhz;
1416		− dcn2_1_soc.clock_limits[k].dram_speed_mts = clk_table->entries[i].memclk_mhz * 2;
1417		−
1418		− dcn2_1_soc.clock_limits[k].dispclk_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].dispclk_mhz;
1419		− dcn2_1_soc.clock_limits[k].dppclk_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].dppclk_mhz;
1420		− dcn2_1_soc.clock_limits[k].dram_bw_per_chan_gbps = dcn2_1_soc.clock_limits[closest_clk_lvl].dram_bw_per_chan_gbps;
1421		− dcn2_1_soc.clock_limits[k].dscclk_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].dscclk_mhz;
1422		− dcn2_1_soc.clock_limits[k].dtbclk_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].dtbclk_mhz;
1423		− dcn2_1_soc.clock_limits[k].phyclk_d18_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].phyclk_d18_mhz;
1424		− dcn2_1_soc.clock_limits[k].phyclk_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].phyclk_mhz;
1425		− k++;
1426		− }
	1400	+ clock_limits[i].state = i;
	1401	+ clock_limits[i].dcfclk_mhz = clk_table->entries[i].dcfclk_mhz;
	1402	+ clock_limits[i].fabricclk_mhz = clk_table->entries[i].fclk_mhz;
	1403	+ clock_limits[i].socclk_mhz = clk_table->entries[i].socclk_mhz;
	1404	+ clock_limits[i].dram_speed_mts = clk_table->entries[i].memclk_mhz * 2;
	1405	+
	1406	+ clock_limits[i].dispclk_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].dispclk_mhz;
	1407	+ clock_limits[i].dppclk_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].dppclk_mhz;
	1408	+ clock_limits[i].dram_bw_per_chan_gbps = dcn2_1_soc.clock_limits[closest_clk_lvl].dram_bw_per_chan_gbps;
	1409	+ clock_limits[i].dscclk_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].dscclk_mhz;
	1410	+ clock_limits[i].dtbclk_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].dtbclk_mhz;
	1411	+ clock_limits[i].phyclk_d18_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].phyclk_d18_mhz;
	1412	+ clock_limits[i].phyclk_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].phyclk_mhz;
	1413	+ }
	1414	+ for (i = 0; i < clk_table->num_entries; i++)
	1415	+ dcn2_1_soc.clock_limits[i] = clock_limits[i];
	1416	+ if (clk_table->num_entries) {
	1417	+ dcn2_1_soc.num_states = clk_table->num_entries;
	1418	+ /* duplicate last level */
	1419	+ dcn2_1_soc.clock_limits[dcn2_1_soc.num_states] = dcn2_1_soc.clock_limits[dcn2_1_soc.num_states - 1];
	1420	+ dcn2_1_soc.clock_limits[dcn2_1_soc.num_states].state = dcn2_1_soc.num_states;
1427	1421	}
1428		− dcn2_1_soc.num_states = k;
1429	1422	}
1430	1423
1431		− /* duplicate last level */
1432		− dcn2_1_soc.clock_limits[dcn2_1_soc.num_states] = dcn2_1_soc.clock_limits[dcn2_1_soc.num_states - 1];
1433		− dcn2_1_soc.clock_limits[dcn2_1_soc.num_states].state = dcn2_1_soc.num_states;
1434		−
1435	1424	dml_init_instance(&dc->dml, &dcn2_1_soc, &dcn2_1_ip, DML_PROJECT_DCN21);
1436	1425	}
1437	1426

drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.c+4−4

@@ -1200,7 +1200,7 @@ static void dml_rq_dlg_get_dlg_params( · 1199 unmodified lines
1200	1200	min_hratio_fact_l = 1.0;
1201	1201	min_hratio_fact_c = 1.0;
1202	1202
1203		− if (htaps_l <= 1)
	1203	+ if (hratio_l <= 1)
1204	1204	min_hratio_fact_l = 2.0;
1205	1205	else if (htaps_l <= 6) {
1206	1206	if ((hratio_l * 2.0) > 4.0)
@@ -1216,7 +1216,7 @@ static void dml_rq_dlg_get_dlg_params( · 9 unmodified lines
1216	1216
1217	1217	hscale_pixel_rate_l = min_hratio_fact_l * dppclk_freq_in_mhz;
1218	1218
1219		− if (htaps_c <= 1)
	1219	+ if (hratio_c <= 1)
1220	1220	min_hratio_fact_c = 2.0;
1221	1221	else if (htaps_c <= 6) {
1222	1222	if ((hratio_c * 2.0) > 4.0)
@@ -1522,8 +1522,8 @@ static void dml_rq_dlg_get_dlg_params( · 299 unmodified lines
1522	1522
1523	1523	disp_dlg_regs->refcyc_per_vm_group_vblank = get_refcyc_per_vm_group_vblank(mode_lib, e2e_pipe_param, num_pipes, pipe_idx) * refclk_freq_in_mhz;
1524	1524	disp_dlg_regs->refcyc_per_vm_group_flip = get_refcyc_per_vm_group_flip(mode_lib, e2e_pipe_param, num_pipes, pipe_idx) * refclk_freq_in_mhz;
1525		− disp_dlg_regs->refcyc_per_vm_req_vblank = get_refcyc_per_vm_req_vblank(mode_lib, e2e_pipe_param, num_pipes, pipe_idx) * refclk_freq_in_mhz;
1526		− disp_dlg_regs->refcyc_per_vm_req_flip = get_refcyc_per_vm_req_flip(mode_lib, e2e_pipe_param, num_pipes, pipe_idx) * refclk_freq_in_mhz;
	1525	+ disp_dlg_regs->refcyc_per_vm_req_vblank = get_refcyc_per_vm_req_vblank(mode_lib, e2e_pipe_param, num_pipes, pipe_idx) * refclk_freq_in_mhz * dml_pow(2, 10);
	1526	+ disp_dlg_regs->refcyc_per_vm_req_flip = get_refcyc_per_vm_req_flip(mode_lib, e2e_pipe_param, num_pipes, pipe_idx) * refclk_freq_in_mhz * dml_pow(2, 10);
1527	1527
1528	1528	// Clamp to max for now
1529	1529	if (disp_dlg_regs->refcyc_per_vm_group_vblank >= (unsigned int)dml_pow(2, 23))

drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h+16−0

@@ -210,6 +210,22 @@ struct mpc_funcs { · 209 unmodified lines
210	210	struct mpcc_blnd_cfg *blnd_cfg,
211	211	int mpcc_id);
212	212
	213	+ /*
	214	+ * Lock cursor updates for the specified OPP.
	215	+ * OPP defines the set of MPCC that are locked together for cursor.
	216	+ *
	217	+ * Parameters:
	218	+ * [in] mpc - MPC context.
	219	+ * [in] opp_id - The OPP to lock cursor updates on
	220	+ * [in] lock - lock/unlock the OPP
	221	+ *
	222	+ * Return: void
	223	+ */
	224	+ void (*cursor_lock)(
	225	+ struct mpc *mpc,
	226	+ int opp_id,
	227	+ bool lock);
	228	+
213	229	struct mpcc* (*get_mpcc_for_dpp)(
214	230	struct mpc_tree *tree,
215	231	int dpp_id);

drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h+1−0

@@ -86,6 +86,7 @@ struct hw_sequencer_funcs { · 85 unmodified lines
86	86	struct dc_state *context, bool lock);
87	87	void (set_flip_control_gsl)(struct pipe_ctx pipe_ctx,
88	88	bool flip_immediate);
	89	+ void (cursor_lock)(struct dc dc, struct pipe_ctx *pipe, bool lock);
89	90
90	91	/* Timing Related */
91	92	void (get_position)(struct pipe_ctx *pipe_ctx, int num_pipes,

drivers/gpu/drm/amd/display/dc/os_types.h+1−1

@@ -108,7 +108,7 @@ · 107 unmodified lines
108	108	#define ASSERT(expr) ASSERT_CRITICAL(expr)
109	109
110	110	#else
111		−#define ASSERT(expr) WARN_ON(!(expr))
	111	+#define ASSERT(expr) WARN_ON_ONCE(!(expr))
112	112	#endif
113	113
114	114	#define BREAK_TO_DEBUGGER() ASSERT(0)

drivers/gpu/drm/amd/powerplay/amd_powerplay.c+5−4

@@ -1435,7 +1435,8 @@ static int pp_get_asic_baco_capability(void handle, bool cap) · 1434 unmodified lines
1435	1435	if (!hwmgr)
1436	1436	return -EINVAL;
1437	1437
1438		− if (!hwmgr->pm_en \|\| !hwmgr->hwmgr_func->get_asic_baco_capability)
	1438	+ if (!(hwmgr->not_vf && amdgpu_dpm) \|\|
	1439	+ !hwmgr->hwmgr_func->get_asic_baco_capability)
1439	1440	return 0;
1440	1441
1441	1442	mutex_lock(&hwmgr->smu_lock);
@@ -1452,8 +1453,7 @@ static int pp_get_asic_baco_state(void handle, int state) · 10 unmodified lines
1452	1453	if (!hwmgr)
1453	1454	return -EINVAL;
1454	1455
1455		− if (!(hwmgr->not_vf && amdgpu_dpm) \|\|
1456		− !hwmgr->hwmgr_func->get_asic_baco_state)
	1456	+ if (!hwmgr->pm_en \|\| !hwmgr->hwmgr_func->get_asic_baco_state)
1457	1457	return 0;
1458	1458
1459	1459	mutex_lock(&hwmgr->smu_lock);
@@ -1470,7 +1470,8 @@ static int pp_set_asic_baco_state(void *handle, int state) · 10 unmodified lines
1470	1470	if (!hwmgr)
1471	1471	return -EINVAL;
1472	1472
1473		− if (!hwmgr->pm_en \|\| !hwmgr->hwmgr_func->set_asic_baco_state)
	1473	+ if (!(hwmgr->not_vf && amdgpu_dpm) \|\|
	1474	+ !hwmgr->hwmgr_func->set_asic_baco_state)
1474	1475	return 0;
1475	1476
1476	1477	mutex_lock(&hwmgr->smu_lock);

drivers/gpu/drm/amd/powerplay/hwmgr/processpptables.c+26−0

@@ -984,6 +984,32 @@ static int init_thermal_controller( · 983 unmodified lines
984	984	struct pp_hwmgr *hwmgr,
985	985	const ATOM_PPLIB_POWERPLAYTABLE *powerplay_table)
986	986	{
	987	+ hwmgr->thermal_controller.ucType =
	988	+ powerplay_table->sThermalController.ucType;
	989	+ hwmgr->thermal_controller.ucI2cLine =
	990	+ powerplay_table->sThermalController.ucI2cLine;
	991	+ hwmgr->thermal_controller.ucI2cAddress =
	992	+ powerplay_table->sThermalController.ucI2cAddress;
	993	+
	994	+ hwmgr->thermal_controller.fanInfo.bNoFan =
	995	+ (0 != (powerplay_table->sThermalController.ucFanParameters &
	996	+ ATOM_PP_FANPARAMETERS_NOFAN));
	997	+
	998	+ hwmgr->thermal_controller.fanInfo.ucTachometerPulsesPerRevolution =
	999	+ powerplay_table->sThermalController.ucFanParameters &
	1000	+ ATOM_PP_FANPARAMETERS_TACHOMETER_PULSES_PER_REVOLUTION_MASK;
	1001	+
	1002	+ hwmgr->thermal_controller.fanInfo.ulMinRPM
	1003	+ = powerplay_table->sThermalController.ucFanMinRPM * 100UL;
	1004	+ hwmgr->thermal_controller.fanInfo.ulMaxRPM
	1005	+ = powerplay_table->sThermalController.ucFanMaxRPM * 100UL;
	1006	+
	1007	+ set_hw_cap(hwmgr,
	1008	+ ATOM_PP_THERMALCONTROLLER_NONE != hwmgr->thermal_controller.ucType,
	1009	+ PHM_PlatformCaps_ThermalController);
	1010	+
	1011	+ hwmgr->thermal_controller.use_hw_fan_control = 1;
	1012	+
987	1013	return 0;
988	1014	}
989	1015

drivers/gpu/drm/amd/powerplay/renoir_ppt.c+6−1

@@ -895,12 +895,17 @@ static int renoir_read_sensor(struct smu_context *smu, · 894 unmodified lines
895	895
896	896	static bool renoir_is_dpm_running(struct smu_context *smu)
897	897	{
	898	+ struct amdgpu_device *adev = smu->adev;
	899	+
898	900	/*
899	901	* Util now, the pmfw hasn't exported the interface of SMU
900	902	* feature mask to APU SKU so just force on all the feature
901	903	* at early initial stage.
902	904	*/
903		− return true;
	905	+ if (adev->in_suspend)
	906	+ return false;
	907	+ else
	908	+ return true;
904	909
905	910	}
906	911

drivers/gpu/drm/bridge/analogix/analogix-anx6345.c+3−0

@@ -485,6 +485,9 @@ static int anx6345_get_modes(struct drm_connector *connector) · 484 unmodified lines
485	485
486	486	num_modes += drm_add_edid_modes(connector, anx6345->edid);
487	487
	488	+ /* Driver currently supports only 6bpc */
	489	+ connector->display_info.bpc = 6;
	490	+
488	491	unlock:
489	492	if (power_off)
490	493	anx6345_poweroff(anx6345);

drivers/gpu/drm/drm_dp_mst_topology.c+7−2

@@ -3442,8 +3442,12 @@ static int drm_dp_send_dpcd_write(struct drm_dp_mst_topology_mgr *mgr, · 3441 unmodified lines
3442	3442	drm_dp_queue_down_tx(mgr, txmsg);
3443	3443
3444	3444	ret = drm_dp_mst_wait_tx_reply(mstb, txmsg);
3445		− if (ret > 0 && txmsg->reply.reply_type == DP_SIDEBAND_REPLY_NAK)
3446		− ret = -EIO;
	3445	+ if (ret > 0) {
	3446	+ if (txmsg->reply.reply_type == DP_SIDEBAND_REPLY_NAK)
	3447	+ ret = -EIO;
	3448	+ else
	3449	+ ret = size;
	3450	+ }
3447	3451
3448	3452	kfree(txmsg);
3449	3453	fail_put:
@@ -4295,6 +4299,7 @@ int drm_dp_atomic_release_vcpi_slots(struct drm_atomic_state *state, · 845 unmodified lines
4295	4299	if (pos->vcpi) {
4296	4300	drm_dp_mst_put_port_malloc(port);
4297	4301	pos->vcpi = 0;
	4302	+ pos->pbn = 0;
4298	4303	}
4299	4304
4300	4305	return 0;

drivers/gpu/drm/drm_edid.c+1−1

@@ -5111,7 +5111,7 @@ static struct drm_display_mode drm_mode_displayid_detailed(struct drm_device d · 5110 unmodified lines
5111	5111	struct drm_display_mode *mode;
5112	5112	unsigned pixel_clock = (timings->pixel_clock[0] \|
5113	5113	(timings->pixel_clock[1] << 8) \|
5114		− (timings->pixel_clock[2] << 16));
	5114	+ (timings->pixel_clock[2] << 16)) + 1;
5115	5115	unsigned hactive = (timings->hactive[0] \| timings->hactive[1] << 8) + 1;
5116	5116	unsigned hblank = (timings->hblank[0] \| timings->hblank[1] << 8) + 1;
5117	5117	unsigned hsync = (timings->hsync[0] \| (timings->hsync[1] & 0x7f) << 8) + 1;

drivers/gpu/drm/drm_hdcp.c+7−1

@@ -241,8 +241,12 @@ static int drm_hdcp_request_srm(struct drm_device *drm_dev, · 240 unmodified lines
241	241
242	242	ret = request_firmware_direct(&fw, (const char *)fw_name,
243	243	drm_dev->dev);
244		− if (ret < 0)
	244	+ if (ret < 0) {
	245	+ *revoked_ksv_cnt = 0;
	246	+ *revoked_ksv_list = NULL;
	247	+ ret = 0;
245	248	goto exit;
	249	+ }
246	250
247	251	if (fw->size && fw->data)
248	252	ret = drm_hdcp_srm_update(fw->data, fw->size, revoked_ksv_list,
@@ -287,6 +291,8 @@ int drm_hdcp_check_ksvs_revoked(struct drm_device drm_dev, u8 ksvs, · 38 unmodified lines
287	291
288	292	ret = drm_hdcp_request_srm(drm_dev, &revoked_ksv_list,
289	293	&revoked_ksv_cnt);
	294	+ if (ret)
	295	+ return ret;
290	296
291	297	/* revoked_ksv_cnt will be zero when above function failed */
292	298	for (i = 0; i < revoked_ksv_cnt; i++)

drivers/gpu/drm/i915/display/intel_ddi.c+11−3

@@ -3141,9 +3141,6 @@ static void hsw_ddi_pre_enable_dp(struct intel_encoder *encoder, · 3140 unmodified lines
3141	3141	intel_dp_set_link_params(intel_dp, crtc_state->port_clock,
3142	3142	crtc_state->lane_count, is_mst);
3143	3143
3144		− intel_dp->regs.dp_tp_ctl = DP_TP_CTL(port);
3145		− intel_dp->regs.dp_tp_status = DP_TP_STATUS(port);
3146		−
3147	3144	intel_edp_panel_on(intel_dp);
3148	3145
3149	3146	intel_ddi_clk_select(encoder, crtc_state);
@@ -3848,12 +3845,18 @@ void intel_ddi_get_config(struct intel_encoder *encoder, · 698 unmodified lines
3848	3845	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
3849	3846	struct intel_crtc *intel_crtc = to_intel_crtc(pipe_config->uapi.crtc);
3850	3847	enum transcoder cpu_transcoder = pipe_config->cpu_transcoder;
	3848	+ struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
3851	3849	u32 temp, flags = 0;
3852	3850
3853	3851	/* XXX: DSI transcoder paranoia */
3854	3852	if (drm_WARN_ON(&dev_priv->drm, transcoder_is_dsi(cpu_transcoder)))
3855	3853	return;
3856	3854
	3855	+ if (INTEL_GEN(dev_priv) >= 12) {
	3856	+ intel_dp->regs.dp_tp_ctl = TGL_DP_TP_CTL(cpu_transcoder);
	3857	+ intel_dp->regs.dp_tp_status = TGL_DP_TP_STATUS(cpu_transcoder);
	3858	+ }
	3859	+
3857	3860	intel_dsc_get_config(encoder, pipe_config);
3858	3861
3859	3862	temp = intel_de_read(dev_priv, TRANS_DDI_FUNC_CTL(cpu_transcoder));
@@ -4173,6 +4176,7 @@ static const struct drm_encoder_funcs intel_ddi_funcs = { · 313 unmodified lines
4173	4176	static struct intel_connector *
4174	4177	intel_ddi_init_dp_connector(struct intel_digital_port *intel_dig_port)
4175	4178	{
	4179	+ struct drm_i915_private *dev_priv = to_i915(intel_dig_port->base.base.dev);
4176	4180	struct intel_connector *connector;
4177	4181	enum port port = intel_dig_port->base.port;
4178	4182
@@ -4183,6 +4187,10 @@ intel_ddi_init_dp_connector(struct intel_digital_port *intel_dig_port) · 4 unmodified lines
4183	4187	intel_dig_port->dp.output_reg = DDI_BUF_CTL(port);
4184	4188	intel_dig_port->dp.prepare_link_retrain =
4185	4189	intel_ddi_prepare_link_retrain;
	4190	+ if (INTEL_GEN(dev_priv) < 12) {
	4191	+ intel_dig_port->dp.regs.dp_tp_ctl = DP_TP_CTL(port);
	4192	+ intel_dig_port->dp.regs.dp_tp_status = DP_TP_STATUS(port);
	4193	+ }
4186	4194
4187	4195	if (!intel_dp_init_connector(intel_dig_port, connector)) {
4188	4196	kfree(connector);

drivers/gpu/drm/i915/display/intel_display_power.c+6−6

@@ -4140,7 +4140,7 @@ static const struct i915_power_well_desc tgl_power_wells[] = { · 4139 unmodified lines
4140	4140	{
4141	4141	.name = "AUX D TBT1",
4142	4142	.domains = TGL_AUX_D_TBT1_IO_POWER_DOMAINS,
4143		− .ops = &hsw_power_well_ops,
	4143	+ .ops = &icl_tc_phy_aux_power_well_ops,
4144	4144	.id = DISP_PW_ID_NONE,
4145	4145	{
4146	4146	.hsw.regs = &icl_aux_power_well_regs,
@@ -4151,7 +4151,7 @@ static const struct i915_power_well_desc tgl_power_wells[] = { · 4 unmodified lines
4151	4151	{
4152	4152	.name = "AUX E TBT2",
4153	4153	.domains = TGL_AUX_E_TBT2_IO_POWER_DOMAINS,
4154		− .ops = &hsw_power_well_ops,
	4154	+ .ops = &icl_tc_phy_aux_power_well_ops,
4155	4155	.id = DISP_PW_ID_NONE,
4156	4156	{
4157	4157	.hsw.regs = &icl_aux_power_well_regs,
@@ -4162,7 +4162,7 @@ static const struct i915_power_well_desc tgl_power_wells[] = { · 4 unmodified lines
4162	4162	{
4163	4163	.name = "AUX F TBT3",
4164	4164	.domains = TGL_AUX_F_TBT3_IO_POWER_DOMAINS,
4165		− .ops = &hsw_power_well_ops,
	4165	+ .ops = &icl_tc_phy_aux_power_well_ops,
4166	4166	.id = DISP_PW_ID_NONE,
4167	4167	{
4168	4168	.hsw.regs = &icl_aux_power_well_regs,
@@ -4173,7 +4173,7 @@ static const struct i915_power_well_desc tgl_power_wells[] = { · 4 unmodified lines
4173	4173	{
4174	4174	.name = "AUX G TBT4",
4175	4175	.domains = TGL_AUX_G_TBT4_IO_POWER_DOMAINS,
4176		− .ops = &hsw_power_well_ops,
	4176	+ .ops = &icl_tc_phy_aux_power_well_ops,
4177	4177	.id = DISP_PW_ID_NONE,
4178	4178	{
4179	4179	.hsw.regs = &icl_aux_power_well_regs,
@@ -4184,7 +4184,7 @@ static const struct i915_power_well_desc tgl_power_wells[] = { · 4 unmodified lines
4184	4184	{
4185	4185	.name = "AUX H TBT5",
4186	4186	.domains = TGL_AUX_H_TBT5_IO_POWER_DOMAINS,
4187		− .ops = &hsw_power_well_ops,
	4187	+ .ops = &icl_tc_phy_aux_power_well_ops,
4188	4188	.id = DISP_PW_ID_NONE,
4189	4189	{
4190	4190	.hsw.regs = &icl_aux_power_well_regs,
@@ -4195,7 +4195,7 @@ static const struct i915_power_well_desc tgl_power_wells[] = { · 4 unmodified lines
4195	4195	{
4196	4196	.name = "AUX I TBT6",
4197	4197	.domains = TGL_AUX_I_TBT6_IO_POWER_DOMAINS,
4198		− .ops = &hsw_power_well_ops,
	4198	+ .ops = &icl_tc_phy_aux_power_well_ops,
4199	4199	.id = DISP_PW_ID_NONE,
4200	4200	{
4201	4201	.hsw.regs = &icl_aux_power_well_regs,

drivers/gpu/drm/i915/display/intel_dp.c+2−3

@@ -2517,9 +2517,6 @@ static void intel_dp_prepare(struct intel_encoder *encoder, · 2516 unmodified lines
2517	2517	intel_crtc_has_type(pipe_config,
2518	2518	INTEL_OUTPUT_DP_MST));
2519	2519
2520		− intel_dp->regs.dp_tp_ctl = DP_TP_CTL(port);
2521		− intel_dp->regs.dp_tp_status = DP_TP_STATUS(port);
2522		−
2523	2520	/*
2524	2521	* There are four kinds of DP registers:
2525	2522	*
@@ -7836,6 +7833,8 @@ bool intel_dp_init(struct drm_i915_private *dev_priv, · 5310 unmodified lines
7836	7833
7837	7834	intel_dig_port->dp.output_reg = output_reg;
7838	7835	intel_dig_port->max_lanes = 4;
	7836	+ intel_dig_port->dp.regs.dp_tp_ctl = DP_TP_CTL(port);
	7837	+ intel_dig_port->dp.regs.dp_tp_status = DP_TP_STATUS(port);
7839	7838
7840	7839	intel_encoder->type = INTEL_OUTPUT_DP;
7841	7840	intel_encoder->power_domain = intel_port_to_power_domain(port);

drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c+1−0

@@ -342,6 +342,7 @@ int intel_dp_aux_init_backlight_funcs(struct intel_connector *intel_connector) · 341 unmodified lines
342	342	*/
343	343	if (dev_priv->vbt.backlight.type !=
344	344	INTEL_BACKLIGHT_VESA_EDP_AUX_INTERFACE &&
	345	+ i915_modparams.enable_dpcd_backlight != 1 &&
345	346	!drm_dp_has_quirk(&intel_dp->desc, intel_dp->edid_quirks,
346	347	DP_QUIRK_FORCE_DPCD_BACKLIGHT)) {
347	348	DRM_DEV_INFO(dev->dev,

drivers/gpu/drm/i915/display/intel_hdmi.c+2−1

@@ -1536,7 +1536,8 @@ bool intel_hdmi_hdcp_check_link(struct intel_digital_port *intel_dig_port) · 1535 unmodified lines
1536	1536	intel_de_write(i915, HDCP_RPRIME(i915, cpu_transcoder, port), ri.reg);
1537	1537
1538	1538	/* Wait for Ri prime match */
1539		− if (wait_for(intel_de_read(i915, HDCP_STATUS(i915, cpu_transcoder, port)) &
	1539	+ if (wait_for((intel_de_read(i915, HDCP_STATUS(i915, cpu_transcoder, port)) &
	1540	+ (HDCP_STATUS_RI_MATCH \| HDCP_STATUS_ENC)) ==
1540	1541	(HDCP_STATUS_RI_MATCH \| HDCP_STATUS_ENC), 1)) {
1541	1542	DRM_ERROR("Ri' mismatch detected, link check failed (%x)\n",
1542	1543	intel_de_read(i915, HDCP_STATUS(i915, cpu_transcoder, port)));

drivers/gpu/drm/i915/display/intel_sprite.c+12−5

@@ -2817,19 +2817,25 @@ static bool skl_plane_format_mod_supported(struct drm_plane *_plane, · 2816 unmodified lines
2817	2817	}
2818	2818	}
2819	2819
2820		−static bool gen12_plane_supports_mc_ccs(enum plane_id plane_id)
	2820	+static bool gen12_plane_supports_mc_ccs(struct drm_i915_private *dev_priv,
	2821	+ enum plane_id plane_id)
2821	2822	{
	2823	+ /* Wa_14010477008:tgl[a0..c0] */
	2824	+ if (IS_TGL_REVID(dev_priv, TGL_REVID_A0, TGL_REVID_C0))
	2825	+ return false;
	2826	+
2822	2827	return plane_id < PLANE_SPRITE4;
2823	2828	}
2824	2829
2825	2830	static bool gen12_plane_format_mod_supported(struct drm_plane *_plane,
2826	2831	u32 format, u64 modifier)
2827	2832	{
	2833	+ struct drm_i915_private *dev_priv = to_i915(_plane->dev);
2828	2834	struct intel_plane *plane = to_intel_plane(_plane);
2829	2835
2830	2836	switch (modifier) {
2831	2837	case I915_FORMAT_MOD_Y_TILED_GEN12_MC_CCS:
2832		− if (!gen12_plane_supports_mc_ccs(plane->id))
	2838	+ if (!gen12_plane_supports_mc_ccs(dev_priv, plane->id))
2833	2839	return false;
2834	2840	/* fall through */
2835	2841	case DRM_FORMAT_MOD_LINEAR:
@@ -2998,9 +3004,10 @@ static const u32 icl_get_plane_formats(struct drm_i915_private dev_priv, · 162 unmodified lines
2998	3004	}
2999	3005	}
3000	3006
3001		−static const u64 *gen12_get_plane_modifiers(enum plane_id plane_id)
	3007	+static const u64 gen12_get_plane_modifiers(struct drm_i915_private dev_priv,
	3008	+ enum plane_id plane_id)
3002	3009	{
3003		− if (gen12_plane_supports_mc_ccs(plane_id))
	3010	+ if (gen12_plane_supports_mc_ccs(dev_priv, plane_id))
3004	3011	return gen12_plane_format_modifiers_mc_ccs;
3005	3012	else
3006	3013	return gen12_plane_format_modifiers_rc_ccs;
@@ -3070,7 +3077,7 @@ skl_universal_plane_create(struct drm_i915_private *dev_priv, · 63 unmodified lines
3070	3077
3071	3078	plane->has_ccs = skl_plane_has_ccs(dev_priv, pipe, plane_id);
3072	3079	if (INTEL_GEN(dev_priv) >= 12) {
3073		− modifiers = gen12_get_plane_modifiers(plane_id);
	3080	+ modifiers = gen12_get_plane_modifiers(dev_priv, plane_id);
3074	3081	plane_funcs = &gen12_plane_funcs;
3075	3082	} else {
3076	3083	if (plane->has_ccs)

drivers/gpu/drm/i915/gem/i915_gem_tiling.c+18−2

@@ -182,21 +182,35 @@ i915_gem_object_fence_prepare(struct drm_i915_gem_object *obj, · 181 unmodified lines
182	182	int tiling_mode, unsigned int stride)
183	183	{
184	184	struct i915_ggtt *ggtt = &to_i915(obj->base.dev)->ggtt;
185		− struct i915_vma *vma;
	185	+ struct i915_vma vma, vn;
	186	+ LIST_HEAD(unbind);
186	187	int ret = 0;
187	188
188	189	if (tiling_mode == I915_TILING_NONE)
189	190	return 0;
190	191
191	192	mutex_lock(&ggtt->vm.mutex);
	193	+
	194	+ spin_lock(&obj->vma.lock);
192	195	for_each_ggtt_vma(vma, obj) {
	196	+ GEM_BUG_ON(vma->vm != &ggtt->vm);
	197	+
193	198	if (i915_vma_fence_prepare(vma, tiling_mode, stride))
194	199	continue;
195	200
	201	+ list_move(&vma->vm_link, &unbind);
	202	+ }
	203	+ spin_unlock(&obj->vma.lock);
	204	+
	205	+ list_for_each_entry_safe(vma, vn, &unbind, vm_link) {
196	206	ret = __i915_vma_unbind(vma);
197		− if (ret)
	207	+ if (ret) {
	208	+ /* Restore the remaining vma on an error */
	209	+ list_splice(&unbind, &ggtt->vm.bound_list);
198	210	break;
	211	+ }
199	212	}
	213	+
200	214	mutex_unlock(&ggtt->vm.mutex);
201	215
202	216	return ret;
@@ -268,6 +282,7 @@ i915_gem_object_set_tiling(struct drm_i915_gem_object *obj, · 65 unmodified lines
268	282	}
269	283	mutex_unlock(&obj->mm.lock);
270	284
	285	+ spin_lock(&obj->vma.lock);
271	286	for_each_ggtt_vma(vma, obj) {
272	287	vma->fence_size =
273	288	i915_gem_fence_size(i915, vma->size, tiling, stride);
@@ -278,6 +293,7 @@ i915_gem_object_set_tiling(struct drm_i915_gem_object *obj, · 4 unmodified lines
278	293	if (vma->fence)
279	294	vma->fence->dirty = true;
280	295	}
	296	+ spin_unlock(&obj->vma.lock);
281	297
282	298	obj->tiling_and_stride = tiling \| stride;
283	299	i915_gem_object_unlock(obj);

drivers/gpu/drm/i915/gem/selftests/huge_pages.c+8−4

@@ -1477,8 +1477,10 @@ static int igt_ppgtt_pin_update(void *arg) · 1476 unmodified lines
1477	1477	unsigned int page_size = BIT(first);
1478	1478
1479	1479	obj = i915_gem_object_create_internal(dev_priv, page_size);
1480		− if (IS_ERR(obj))
1481		− return PTR_ERR(obj);
	1480	+ if (IS_ERR(obj)) {
	1481	+ err = PTR_ERR(obj);
	1482	+ goto out_vm;
	1483	+ }
1482	1484
1483	1485	vma = i915_vma_instance(obj, vm, NULL);
1484	1486	if (IS_ERR(vma)) {
@@ -1531,8 +1533,10 @@ static int igt_ppgtt_pin_update(void *arg) · 46 unmodified lines
1531	1533	}
1532	1534
1533	1535	obj = i915_gem_object_create_internal(dev_priv, PAGE_SIZE);
1534		− if (IS_ERR(obj))
1535		− return PTR_ERR(obj);
	1536	+ if (IS_ERR(obj)) {
	1537	+ err = PTR_ERR(obj);
	1538	+ goto out_vm;
	1539	+ }
1536	1540
1537	1541	vma = i915_vma_instance(obj, vm, NULL);
1538	1542	if (IS_ERR(vma)) {

drivers/gpu/drm/i915/gt/intel_rps.c+6−3

@@ -81,13 +81,14 @@ static void rps_enable_interrupts(struct intel_rps *rps) · 80 unmodified lines
81	81	events = (GEN6_PM_RP_UP_THRESHOLD \|
82	82	GEN6_PM_RP_DOWN_THRESHOLD \|
83	83	GEN6_PM_RP_DOWN_TIMEOUT);
84		−
85	84	WRITE_ONCE(rps->pm_events, events);
	85	+
86	86	spin_lock_irq(&gt->irq_lock);
87	87	gen6_gt_pm_enable_irq(gt, rps->pm_events);
88	88	spin_unlock_irq(&gt->irq_lock);
89	89
90		− set(gt->uncore, GEN6_PMINTRMSK, rps_pm_mask(rps, rps->cur_freq));
	90	+ intel_uncore_write(gt->uncore,
	91	+ GEN6_PMINTRMSK, rps_pm_mask(rps, rps->last_freq));
91	92	}
92	93
93	94	static void gen6_rps_reset_interrupts(struct intel_rps *rps)
@@ -120,7 +121,9 @@ static void rps_disable_interrupts(struct intel_rps *rps) · 26 unmodified lines
120	121	struct intel_gt *gt = rps_to_gt(rps);
121	122
122	123	WRITE_ONCE(rps->pm_events, 0);
123		− set(gt->uncore, GEN6_PMINTRMSK, rps_pm_sanitize_mask(rps, ~0u));
	124	+
	125	+ intel_uncore_write(gt->uncore,
	126	+ GEN6_PMINTRMSK, rps_pm_sanitize_mask(rps, ~0u));
124	127
125	128	spin_lock_irq(&gt->irq_lock);
126	129	gen6_gt_pm_disable_irq(gt, GEN6_PM_RPS_EVENTS);

drivers/gpu/drm/i915/gt/intel_timeline.c+2−0

@@ -521,6 +521,8 @@ int intel_timeline_read_hwsp(struct i915_request *from, · 520 unmodified lines
521	521
522	522	rcu_read_lock();
523	523	cl = rcu_dereference(from->hwsp_cacheline);
	524	+ if (i915_request_completed(from)) /* confirm cacheline is valid */
	525	+ goto unlock;
524	526	if (unlikely(!i915_active_acquire_if_busy(&cl->active)))
525	527	goto unlock; /* seqno wrapped and completed! */
526	528	if (unlikely(i915_request_completed(from)))

drivers/gpu/drm/i915/i915_drv.h+2−0

@@ -1507,6 +1507,8 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915, · 1506 unmodified lines
1507	1507	(IS_ICELAKE(p) && IS_REVID(p, since, until))
1508	1508
1509	1509	#define TGL_REVID_A0 0x0
	1510	+#define TGL_REVID_B0 0x1
	1511	+#define TGL_REVID_C0 0x2
1510	1512
1511	1513	#define IS_TGL_REVID(p, since, until) \
1512	1514	(IS_TIGERLAKE(p) && IS_REVID(p, since, until))

drivers/gpu/drm/i915/i915_irq.c+2−4

@@ -3358,7 +3358,8 @@ static void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv) · 3357 unmodified lines
3358	3358	{
3359	3359	struct intel_uncore *uncore = &dev_priv->uncore;
3360	3360
3361		− u32 de_pipe_masked = GEN8_PIPE_CDCLK_CRC_DONE;
	3361	+ u32 de_pipe_masked = gen8_de_pipe_fault_mask(dev_priv) \|
	3362	+ GEN8_PIPE_CDCLK_CRC_DONE;
3362	3363	u32 de_pipe_enables;
3363	3364	u32 de_port_masked = GEN8_AUX_CHANNEL_A;
3364	3365	u32 de_port_enables;
@@ -3369,13 +3370,10 @@ static void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv) · 4 unmodified lines
3369	3370	de_misc_masked \|= GEN8_DE_MISC_GSE;
3370	3371
3371	3372	if (INTEL_GEN(dev_priv) >= 9) {
3372		− de_pipe_masked \|= GEN9_DE_PIPE_IRQ_FAULT_ERRORS;
3373	3373	de_port_masked \|= GEN9_AUX_CHANNEL_B \| GEN9_AUX_CHANNEL_C \|
3374	3374	GEN9_AUX_CHANNEL_D;
3375	3375	if (IS_GEN9_LP(dev_priv))
3376	3376	de_port_masked \|= BXT_DE_PORT_GMBUS;
3377		− } else {
3378		− de_pipe_masked \|= GEN8_DE_PIPE_IRQ_FAULT_ERRORS;
3379	3377	}
3380	3378
3381	3379	if (INTEL_GEN(dev_priv) >= 11)

drivers/gpu/drm/i915/i915_reg.h+2−2

@@ -34,8 +34,8 @@ · 33 unmodified lines
34	34	* Follow the style described here for new macros, and while changing existing
35	35	* macros. Do not mass change existing definitions just to update the style.
36	36	*
37		− * Layout
38		− * ~~~~~~
	37	+ * File Layout
	38	+ * ~~~~~~~~~~~
39	39	*
40	40	* Keep helper macros near the top. For example, _PIPE() and friends.
41	41	*

drivers/gpu/drm/i915/i915_vma.c+6−4

@@ -158,16 +158,18 @@ vma_create(struct drm_i915_gem_object *obj, · 157 unmodified lines
158	158
159	159	GEM_BUG_ON(!IS_ALIGNED(vma->size, I915_GTT_PAGE_SIZE));
160	160
	161	+ spin_lock(&obj->vma.lock);
	162	+
161	163	if (i915_is_ggtt(vm)) {
162	164	if (unlikely(overflows_type(vma->size, u32)))
163		− goto err_vma;
	165	+ goto err_unlock;
164	166
165	167	vma->fence_size = i915_gem_fence_size(vm->i915, vma->size,
166	168	i915_gem_object_get_tiling(obj),
167	169	i915_gem_object_get_stride(obj));
168	170	if (unlikely(vma->fence_size < vma->size \|\| /* overflow */
169	171	vma->fence_size > vm->total))
170		− goto err_vma;
	172	+ goto err_unlock;
171	173
172	174	GEM_BUG_ON(!IS_ALIGNED(vma->fence_size, I915_GTT_MIN_ALIGNMENT));
173	175
@@ -179,8 +181,6 @@ vma_create(struct drm_i915_gem_object *obj, · 5 unmodified lines
179	181	__set_bit(I915_VMA_GGTT_BIT, __i915_vma_flags(vma));
180	182	}
181	183
182		− spin_lock(&obj->vma.lock);
183		−
184	184	rb = NULL;
185	185	p = &obj->vma.tree.rb_node;
186	186	while (*p) {
@@ -225,6 +225,8 @@ vma_create(struct drm_i915_gem_object *obj, · 38 unmodified lines
225	225
226	226	return vma;
227	227
	228	+err_unlock:
	229	+ spin_unlock(&obj->vma.lock);
228	230	err_vma:
229	231	i915_vma_free(vma);
230	232	return ERR_PTR(-E2BIG);

drivers/gpu/drm/ingenic/ingenic-drm.c+1−0

@@ -843,6 +843,7 @@ static const struct of_device_id ingenic_drm_of_match[] = { · 842 unmodified lines
843	843	{ .compatible = "ingenic,jz4770-lcd", .data = &jz4770_soc_info },
844	844	{ /* sentinel */ },
845	845	};
	846	+MODULE_DEVICE_TABLE(of, ingenic_drm_of_match);
846	847
847	848	static struct platform_driver ingenic_drm_driver = {
848	849	.driver = {

drivers/gpu/drm/meson/meson_dw_hdmi.c+1−3

@@ -1034,10 +1034,8 @@ static int meson_dw_hdmi_bind(struct device dev, struct device master, · 1033 unmodified lines
1034	1034	return PTR_ERR(dw_plat_data->regm);
1035	1035
1036	1036	irq = platform_get_irq(pdev, 0);
1037		− if (irq < 0) {
1038		− dev_err(dev, "Failed to get hdmi top irq\n");
	1037	+ if (irq < 0)
1039	1038	return irq;
1040		− }
1041	1039
1042	1040	ret = devm_request_threaded_irq(dev, irq, dw_hdmi_top_irq,
1043	1041	dw_hdmi_top_thread_irq, IRQF_SHARED,

drivers/gpu/drm/qxl/qxl_cmd.c+5−5

@@ -480,9 +480,10 @@ int qxl_hw_surface_alloc(struct qxl_device *qdev, · 479 unmodified lines
480	480	return ret;
481	481
482	482	ret = qxl_release_reserve_list(release, true);
483		− if (ret)
	483	+ if (ret) {
	484	+ qxl_release_free(qdev, release);
484	485	return ret;
485		−
	486	+ }
486	487	cmd = (struct qxl_surface_cmd *)qxl_release_map(qdev, release);
487	488	cmd->type = QXL_SURFACE_CMD_CREATE;
488	489	cmd->flags = QXL_SURF_FLAG_KEEP_DATA;
@@ -499,8 +500,8 @@ int qxl_hw_surface_alloc(struct qxl_device *qdev, · 10 unmodified lines
499	500	/* no need to add a release to the fence for this surface bo,
500	501	since it is only released when we ask to destroy the surface
501	502	and it would never signal otherwise */
502		− qxl_push_command_ring_release(qdev, release, QXL_CMD_SURFACE, false);
503	503	qxl_release_fence_buffer_objects(release);
	504	+ qxl_push_command_ring_release(qdev, release, QXL_CMD_SURFACE, false);
504	505
505	506	surf->hw_surf_alloc = true;
506	507	spin_lock(&qdev->surf_id_idr_lock);
@@ -542,9 +543,8 @@ int qxl_hw_surface_dealloc(struct qxl_device *qdev, · 35 unmodified lines
542	543	cmd->surface_id = id;
543	544	qxl_release_unmap(qdev, release, &cmd->release_info);
544	545
545		− qxl_push_command_ring_release(qdev, release, QXL_CMD_SURFACE, false);
546		−
547	546	qxl_release_fence_buffer_objects(release);
	547	+ qxl_push_command_ring_release(qdev, release, QXL_CMD_SURFACE, false);
548	548
549	549	return 0;
550	550	}

drivers/gpu/drm/qxl/qxl_display.c+3−3

@@ -510,8 +510,8 @@ static int qxl_primary_apply_cursor(struct drm_plane *plane) · 509 unmodified lines
510	510	cmd->u.set.visible = 1;
511	511	qxl_release_unmap(qdev, release, &cmd->release_info);
512	512
513		− qxl_push_cursor_ring_release(qdev, release, QXL_CMD_CURSOR, false);
514	513	qxl_release_fence_buffer_objects(release);
	514	+ qxl_push_cursor_ring_release(qdev, release, QXL_CMD_CURSOR, false);
515	515
516	516	return ret;
517	517
@@ -652,8 +652,8 @@ static void qxl_cursor_atomic_update(struct drm_plane *plane, · 134 unmodified lines
652	652	cmd->u.position.y = plane->state->crtc_y + fb->hot_y;
653	653
654	654	qxl_release_unmap(qdev, release, &cmd->release_info);
655		− qxl_push_cursor_ring_release(qdev, release, QXL_CMD_CURSOR, false);
656	655	qxl_release_fence_buffer_objects(release);
	656	+ qxl_push_cursor_ring_release(qdev, release, QXL_CMD_CURSOR, false);
657	657
658	658	if (old_cursor_bo != NULL)
659	659	qxl_bo_unpin(old_cursor_bo);
@@ -700,8 +700,8 @@ static void qxl_cursor_atomic_disable(struct drm_plane *plane, · 40 unmodified lines
700	700	cmd->type = QXL_CURSOR_HIDE;
701	701	qxl_release_unmap(qdev, release, &cmd->release_info);
702	702
703		− qxl_push_cursor_ring_release(qdev, release, QXL_CMD_CURSOR, false);
704	703	qxl_release_fence_buffer_objects(release);
	704	+ qxl_push_cursor_ring_release(qdev, release, QXL_CMD_CURSOR, false);
705	705	}
706	706
707	707	static void qxl_update_dumb_head(struct qxl_device *qdev,

drivers/gpu/drm/qxl/qxl_draw.c+4−3

@@ -209,9 +209,10 @@ void qxl_draw_dirty_fb(struct qxl_device *qdev, · 208 unmodified lines
209	209	goto out_release_backoff;
210	210
211	211	rects = drawable_set_clipping(qdev, num_clips, clips_bo);
212		− if (!rects)
	212	+ if (!rects) {
	213	+ ret = -EINVAL;
213	214	goto out_release_backoff;
214		−
	215	+ }
215	216	drawable = (struct qxl_drawable *)qxl_release_map(qdev, release);
216	217
217	218	drawable->clip.type = SPICE_CLIP_TYPE_RECTS;
@@ -242,8 +243,8 @@ void qxl_draw_dirty_fb(struct qxl_device *qdev, · 24 unmodified lines
242	243	}
243	244	qxl_bo_kunmap(clips_bo);
244	245
245		− qxl_push_command_ring_release(qdev, release, QXL_CMD_DRAW, false);
246	246	qxl_release_fence_buffer_objects(release);
	247	+ qxl_push_command_ring_release(qdev, release, QXL_CMD_DRAW, false);
247	248
248	249	out_release_backoff:
249	250	if (ret)

drivers/gpu/drm/qxl/qxl_image.c+2−1

@@ -212,7 +212,8 @@ qxl_image_init_helper(struct qxl_device *qdev, · 211 unmodified lines
212	212	break;
213	213	default:
214	214	DRM_ERROR("unsupported image bit depth\n");
215		− return -EINVAL; /* TODO: cleanup */
	215	+ qxl_bo_kunmap_atomic_page(qdev, image_bo, ptr);
	216	+ return -EINVAL;
216	217	}
217	218	image->u.bitmap.flags = QXL_BITMAP_TOP_DOWN;
218	219	image->u.bitmap.x = width;

drivers/gpu/drm/qxl/qxl_ioctl.c+1−4

@@ -261,11 +261,8 @@ static int qxl_process_single_command(struct qxl_device *qdev, · 260 unmodified lines
261	261	apply_surf_reloc(qdev, &reloc_info[i]);
262	262	}
263	263
	264	+ qxl_release_fence_buffer_objects(release);
264	265	ret = qxl_push_command_ring_release(qdev, release, cmd->type, true);
265		− if (ret)
266		− qxl_release_backoff_reserve_list(release);
267		− else
268		− qxl_release_fence_buffer_objects(release);
269	266
270	267	out_free_bos:
271	268	out_free_release:

drivers/gpu/drm/scheduler/sched_main.c+1−1

@@ -676,7 +676,7 @@ drm_sched_get_cleanup_job(struct drm_gpu_scheduler *sched) · 675 unmodified lines
676	676	*/
677	677	if ((sched->timeout != MAX_SCHEDULE_TIMEOUT &&
678	678	!cancel_delayed_work(&sched->work_tdr)) \|\|
679		− __kthread_should_park(sched->thread))
	679	+ kthread_should_park())
680	680	return NULL;
681	681
682	682	spin_lock(&sched->job_list_lock);

drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c+1−1

@@ -717,7 +717,7 @@ static void sun6i_dsi_encoder_enable(struct drm_encoder *encoder) · 716 unmodified lines
717	717	struct drm_display_mode *mode = &encoder->crtc->state->adjusted_mode;
718	718	struct sun6i_dsi *dsi = encoder_to_sun6i_dsi(encoder);
719	719	struct mipi_dsi_device *device = dsi->device;
720		− union phy_configure_opts opts = { 0 };
	720	+ union phy_configure_opts opts = { };
721	721	struct phy_configure_opts_mipi_dphy *cfg = &opts.mipi_dphy;
722	722	u16 delay;
723	723	int err;

drivers/gpu/drm/tidss/tidss_crtc.c+13−3

@@ -379,9 +379,17 @@ static struct drm_crtc_state tidss_crtc_duplicate_state(struct drm_crtc crtc) · 378 unmodified lines
379	379	return &state->base;
380	380	}
381	381
	382	+static void tidss_crtc_destroy(struct drm_crtc *crtc)
	383	+{
	384	+ struct tidss_crtc *tcrtc = to_tidss_crtc(crtc);
	385	+
	386	+ drm_crtc_cleanup(crtc);
	387	+ kfree(tcrtc);
	388	+}
	389	+
382	390	static const struct drm_crtc_funcs tidss_crtc_funcs = {
383	391	.reset = tidss_crtc_reset,
384		− .destroy = drm_crtc_cleanup,
	392	+ .destroy = tidss_crtc_destroy,
385	393	.set_config = drm_atomic_helper_set_config,
386	394	.page_flip = drm_atomic_helper_page_flip,
387	395	.atomic_duplicate_state = tidss_crtc_duplicate_state,
@@ -400,7 +408,7 @@ struct tidss_crtc tidss_crtc_create(struct tidss_device tidss, · 12 unmodified lines
400	408	bool has_ctm = tidss->feat->vp_feat.color.has_ctm;
401	409	int ret;
402	410
403		− tcrtc = devm_kzalloc(tidss->dev, sizeof(*tcrtc), GFP_KERNEL);
	411	+ tcrtc = kzalloc(sizeof(*tcrtc), GFP_KERNEL);
404	412	if (!tcrtc)
405	413	return ERR_PTR(-ENOMEM);
406	414
@@ -411,8 +419,10 @@ struct tidss_crtc tidss_crtc_create(struct tidss_device tidss, · 4 unmodified lines
411	419
412	420	ret = drm_crtc_init_with_planes(&tidss->ddev, crtc, primary,
413	421	NULL, &tidss_crtc_funcs, NULL);
414		− if (ret < 0)
	422	+ if (ret < 0) {
	423	+ kfree(tcrtc);
415	424	return ERR_PTR(ret);
	425	+ }
416	426
417	427	drm_crtc_helper_add(crtc, &tidss_crtc_helper_funcs);
418	428

drivers/gpu/drm/tidss/tidss_encoder.c+11−3

@@ -55,12 +55,18 @@ static int tidss_encoder_atomic_check(struct drm_encoder *encoder, · 54 unmodified lines
55	55	return 0;
56	56	}
57	57
	58	+static void tidss_encoder_destroy(struct drm_encoder *encoder)
	59	+{
	60	+ drm_encoder_cleanup(encoder);
	61	+ kfree(encoder);
	62	+}
	63	+
58	64	static const struct drm_encoder_helper_funcs encoder_helper_funcs = {
59	65	.atomic_check = tidss_encoder_atomic_check,
60	66	};
61	67
62	68	static const struct drm_encoder_funcs encoder_funcs = {
63		− .destroy = drm_encoder_cleanup,
	69	+ .destroy = tidss_encoder_destroy,
64	70	};
65	71
66	72	struct drm_encoder tidss_encoder_create(struct tidss_device tidss,
@@ -69,7 +75,7 @@ struct drm_encoder tidss_encoder_create(struct tidss_device tidss, · 2 unmodified lines
69	75	struct drm_encoder *enc;
70	76	int ret;
71	77
72		− enc = devm_kzalloc(tidss->dev, sizeof(*enc), GFP_KERNEL);
	78	+ enc = kzalloc(sizeof(*enc), GFP_KERNEL);
73	79	if (!enc)
74	80	return ERR_PTR(-ENOMEM);
75	81
@@ -77,8 +83,10 @@ struct drm_encoder tidss_encoder_create(struct tidss_device tidss, · 1 unmodified line
77	83
78	84	ret = drm_encoder_init(&tidss->ddev, enc, &encoder_funcs,
79	85	encoder_type, NULL);
80		− if (ret < 0)
	86	+ if (ret < 0) {
	87	+ kfree(enc);
81	88	return ERR_PTR(ret);
	89	+ }
82	90
83	91	drm_encoder_helper_add(enc, &encoder_helper_funcs);
84	92

drivers/gpu/drm/tidss/tidss_plane.c+18−6

@@ -141,6 +141,14 @@ static void tidss_plane_atomic_disable(struct drm_plane *plane, · 140 unmodified lines
141	141	dispc_plane_enable(tidss->dispc, tplane->hw_plane_id, false);
142	142	}
143	143
	144	+static void drm_plane_destroy(struct drm_plane *plane)
	145	+{
	146	+ struct tidss_plane *tplane = to_tidss_plane(plane);
	147	+
	148	+ drm_plane_cleanup(plane);
	149	+ kfree(tplane);
	150	+}
	151	+
144	152	static const struct drm_plane_helper_funcs tidss_plane_helper_funcs = {
145	153	.atomic_check = tidss_plane_atomic_check,
146	154	.atomic_update = tidss_plane_atomic_update,
@@ -151,7 +159,7 @@ static const struct drm_plane_funcs tidss_plane_funcs = { · 4 unmodified lines
151	159	.update_plane = drm_atomic_helper_update_plane,
152	160	.disable_plane = drm_atomic_helper_disable_plane,
153	161	.reset = drm_atomic_helper_plane_reset,
154		− .destroy = drm_plane_cleanup,
	162	+ .destroy = drm_plane_destroy,
155	163	.atomic_duplicate_state = drm_atomic_helper_plane_duplicate_state,
156	164	.atomic_destroy_state = drm_atomic_helper_plane_destroy_state,
157	165	};
@@ -175,7 +183,7 @@ struct tidss_plane tidss_plane_create(struct tidss_device tidss, · 17 unmodified lines
175	183	BIT(DRM_MODE_BLEND_COVERAGE));
176	184	int ret;
177	185
178		− tplane = devm_kzalloc(tidss->dev, sizeof(*tplane), GFP_KERNEL);
	186	+ tplane = kzalloc(sizeof(*tplane), GFP_KERNEL);
179	187	if (!tplane)
180	188	return ERR_PTR(-ENOMEM);
181	189
@@ -190,7 +198,7 @@ struct tidss_plane tidss_plane_create(struct tidss_device tidss, · 8 unmodified lines
190	198	formats, num_formats,
191	199	NULL, type, NULL);
192	200	if (ret < 0)
193		− return ERR_PTR(ret);
	201	+ goto err;
194	202
195	203	drm_plane_helper_add(&tplane->plane, &tidss_plane_helper_funcs);
196	204
@@ -203,15 +211,19 @@ struct tidss_plane tidss_plane_create(struct tidss_device tidss, · 6 unmodified lines
203	211	default_encoding,
204	212	default_range);
205	213	if (ret)
206		− return ERR_PTR(ret);
	214	+ goto err;
207	215
208	216	ret = drm_plane_create_alpha_property(&tplane->plane);
209	217	if (ret)
210		− return ERR_PTR(ret);
	218	+ goto err;
211	219
212	220	ret = drm_plane_create_blend_mode_property(&tplane->plane, blend_modes);
213	221	if (ret)
214		− return ERR_PTR(ret);
	222	+ goto err;
215	223
216	224	return tplane;
	225	+
	226	+err:
	227	+ kfree(tplane);
	228	+ return ERR_PTR(ret);
217	229	}

drivers/gpu/drm/virtio/virtgpu_drv.h+1−0

@@ -221,6 +221,7 @@ struct virtio_gpu_fpriv { · 220 unmodified lines
221	221	/* virtio_ioctl.c */
222	222	#define DRM_VIRTIO_NUM_IOCTLS 10
223	223	extern struct drm_ioctl_desc virtio_gpu_ioctls[DRM_VIRTIO_NUM_IOCTLS];
	224	+void virtio_gpu_create_context(struct drm_device dev, struct drm_file file);
224	225
225	226	/* virtio_kms.c */
226	227	int virtio_gpu_init(struct drm_device *dev);

drivers/gpu/drm/virtio/virtgpu_gem.c+3−0

@@ -39,6 +39,9 @@ int virtio_gpu_gem_create(struct drm_file *file, · 38 unmodified lines
39	39	int ret;
40	40	u32 handle;
41	41
	42	+ if (vgdev->has_virgl_3d)
	43	+ virtio_gpu_create_context(dev, file);
	44	+
42	45	ret = virtio_gpu_object_create(vgdev, params, &obj, NULL);
43	46	if (ret < 0)
44	47	return ret;

drivers/gpu/drm/virtio/virtgpu_ioctl.c+2−2

@@ -27,14 +27,14 @@ · 26 unmodified lines
27	27
28	28	#include <linux/file.h>
29	29	#include <linux/sync_file.h>
	30	+#include <linux/uaccess.h>
30	31
31	32	#include <drm/drm_file.h>
32	33	#include <drm/virtgpu_drm.h>
33	34
34	35	#include "virtgpu_drv.h"
35	36
36		−static void virtio_gpu_create_context(struct drm_device *dev,
37		− struct drm_file *file)
	37	+void virtio_gpu_create_context(struct drm_device dev, struct drm_file file)
38	38	{
39	39	struct virtio_gpu_device *vgdev = dev->dev_private;
40	40	struct virtio_gpu_fpriv *vfpriv = file->driver_priv;

drivers/gpu/drm/virtio/virtgpu_kms.c+7−11

@@ -25,6 +25,7 @@ · 24 unmodified lines
25	25
26	26	#include <linux/virtio.h>
27	27	#include <linux/virtio_config.h>
	28	+#include <linux/virtio_ring.h>
28	29
29	30	#include <drm/drm_file.h>
30	31
@@ -52,14 +53,6 @@ static void virtio_gpu_config_changed_work_func(struct work_struct *work) · 21 unmodified lines
52	53	events_clear, &events_clear);
53	54	}
54	55
55		−static void virtio_gpu_context_destroy(struct virtio_gpu_device *vgdev,
56		− uint32_t ctx_id)
57		−{
58		− virtio_gpu_cmd_context_destroy(vgdev, ctx_id);
59		− virtio_gpu_notify(vgdev);
60		− ida_free(&vgdev->ctx_id_ida, ctx_id - 1);
61		−}
62		−
63	56	static void virtio_gpu_init_vq(struct virtio_gpu_queue *vgvq,
64	57	void (work_func)(struct work_struct work))
65	58	{
@@ -274,14 +267,17 @@ int virtio_gpu_driver_open(struct drm_device dev, struct drm_file file) · 208 unmodified lines
274	267	void virtio_gpu_driver_postclose(struct drm_device dev, struct drm_file file)
275	268	{
276	269	struct virtio_gpu_device *vgdev = dev->dev_private;
277		− struct virtio_gpu_fpriv *vfpriv;
	270	+ struct virtio_gpu_fpriv *vfpriv = file->driver_priv;
278	271
279	272	if (!vgdev->has_virgl_3d)
280	273	return;
281	274
282		− vfpriv = file->driver_priv;
	275	+ if (vfpriv->context_created) {
	276	+ virtio_gpu_cmd_context_destroy(vgdev, vfpriv->ctx_id);
	277	+ virtio_gpu_notify(vgdev);
	278	+ }
283	279
284		− virtio_gpu_context_destroy(vgdev, vfpriv->ctx_id);
	280	+ ida_free(&vgdev->ctx_id_ida, vfpriv->ctx_id - 1);
285	281	mutex_destroy(&vfpriv->context_lock);
286	282	kfree(vfpriv);
287	283	file->driver_priv = NULL;

drivers/hid/Kconfig+1−0

@@ -1155,6 +1155,7 @@ config HID_ALPS · 1154 unmodified lines
1155	1155	config HID_MCP2221
1156	1156	tristate "Microchip MCP2221 HID USB-to-I2C/SMbus host support"
1157	1157	depends on USB_HID && I2C
	1158	+ depends on GPIOLIB
1158	1159	---help---
1159	1160	Provides I2C and SMBUS host adapter functionality over USB-HID
1160	1161	through MCP2221 device.

drivers/hid/hid-alps.c+1−0

@@ -802,6 +802,7 @@ static int alps_probe(struct hid_device hdev, const struct hid_device_id id) · 801 unmodified lines
802	802	break;
803	803	case HID_DEVICE_ID_ALPS_U1_DUAL:
804	804	case HID_DEVICE_ID_ALPS_U1:
	805	+ case HID_DEVICE_ID_ALPS_U1_UNICORN_LEGACY:
805	806	data->dev_type = U1;
806	807	break;
807	808	default:

drivers/hid/hid-ids.h+7−1

@@ -79,10 +79,10 @@ · 78 unmodified lines
79	79	#define HID_DEVICE_ID_ALPS_U1_DUAL_PTP 0x121F
80	80	#define HID_DEVICE_ID_ALPS_U1_DUAL_3BTN_PTP 0x1220
81	81	#define HID_DEVICE_ID_ALPS_U1 0x1215
	82	+#define HID_DEVICE_ID_ALPS_U1_UNICORN_LEGACY 0x121E
82	83	#define HID_DEVICE_ID_ALPS_T4_BTNLESS 0x120C
83	84	#define HID_DEVICE_ID_ALPS_1222 0x1222
84	85
85		−
86	86	#define USB_VENDOR_ID_AMI 0x046b
87	87	#define USB_DEVICE_ID_AMI_VIRT_KEYBOARD_AND_MOUSE 0xff10
88	88
@@ -385,6 +385,7 @@ · 296 unmodified lines
385	385	#define USB_DEVICE_ID_DWAV_EGALAX_MULTITOUCH_7349 0x7349
386	386	#define USB_DEVICE_ID_DWAV_EGALAX_MULTITOUCH_73F7 0x73f7
387	387	#define USB_DEVICE_ID_DWAV_EGALAX_MULTITOUCH_A001 0xa001
	388	+#define USB_DEVICE_ID_DWAV_EGALAX_MULTITOUCH_C002 0xc002
388	389
389	390	#define USB_VENDOR_ID_ELAN 0x04f3
390	391	#define USB_DEVICE_ID_TOSHIBA_CLICK_L9W 0x0401
@@ -759,6 +760,7 @@ · 368 unmodified lines
759	760	#define USB_DEVICE_ID_LOGITECH_RUMBLEPAD2 0xc218
760	761	#define USB_DEVICE_ID_LOGITECH_RUMBLEPAD2_2 0xc219
761	762	#define USB_DEVICE_ID_LOGITECH_G15_LCD 0xc222
	763	+#define USB_DEVICE_ID_LOGITECH_G11 0xc225
762	764	#define USB_DEVICE_ID_LOGITECH_G15_V2_LCD 0xc227
763	765	#define USB_DEVICE_ID_LOGITECH_G510 0xc22d
764	766	#define USB_DEVICE_ID_LOGITECH_G510_USB_AUDIO 0xc22e
@@ -1097,6 +1099,9 @@ · 332 unmodified lines
1097	1099	#define USB_DEVICE_ID_SYMBOL_SCANNER_2 0x1300
1098	1100	#define USB_DEVICE_ID_SYMBOL_SCANNER_3 0x1200
1099	1101
	1102	+#define I2C_VENDOR_ID_SYNAPTICS 0x06cb
	1103	+#define I2C_PRODUCT_ID_SYNAPTICS_SYNA2393 0x7a13
	1104	+
1100	1105	#define USB_VENDOR_ID_SYNAPTICS 0x06cb
1101	1106	#define USB_DEVICE_ID_SYNAPTICS_TP 0x0001
1102	1107	#define USB_DEVICE_ID_SYNAPTICS_INT_TP 0x0002
@@ -1111,6 +1116,7 @@ · 8 unmodified lines
1111	1116	#define USB_DEVICE_ID_SYNAPTICS_LTS2 0x1d10
1112	1117	#define USB_DEVICE_ID_SYNAPTICS_HD 0x0ac3
1113	1118	#define USB_DEVICE_ID_SYNAPTICS_QUAD_HD 0x1ac3
	1119	+#define USB_DEVICE_ID_SYNAPTICS_DELL_K12A 0x2819
1114	1120	#define USB_DEVICE_ID_SYNAPTICS_ACER_SWITCH5_012 0x2968
1115	1121	#define USB_DEVICE_ID_SYNAPTICS_TP_V103 0x5710
1116	1122	#define USB_DEVICE_ID_SYNAPTICS_ACER_SWITCH5 0x81a7

drivers/hid/hid-lg-g15.c+4−0

@@ -872,6 +872,10 @@ error_hw_stop: · 871 unmodified lines
872	872	}
873	873
874	874	static const struct hid_device_id lg_g15_devices[] = {
	875	+ /* The G11 is a G15 without the LCD, treat it as a G15 */
	876	+ { HID_USB_DEVICE(USB_VENDOR_ID_LOGITECH,
	877	+ USB_DEVICE_ID_LOGITECH_G11),
	878	+ .driver_data = LG_G15 },
875	879	{ HID_USB_DEVICE(USB_VENDOR_ID_LOGITECH,
876	880	USB_DEVICE_ID_LOGITECH_G15_LCD),
877	881	.driver_data = LG_G15 },

drivers/hid/hid-multitouch.c+3−0

@@ -1922,6 +1922,9 @@ static const struct hid_device_id mt_devices[] = { · 1921 unmodified lines
1922	1922	{ .driver_data = MT_CLS_EGALAX_SERIAL,
1923	1923	MT_USB_DEVICE(USB_VENDOR_ID_DWAV,
1924	1924	USB_DEVICE_ID_DWAV_EGALAX_MULTITOUCH_A001) },
	1925	+ { .driver_data = MT_CLS_EGALAX,
	1926	+ MT_USB_DEVICE(USB_VENDOR_ID_DWAV,
	1927	+ USB_DEVICE_ID_DWAV_EGALAX_MULTITOUCH_C002) },
1925	1928
1926	1929	/* Elitegroup panel */
1927	1930	{ .driver_data = MT_CLS_SERIAL,

drivers/hid/hid-quirks.c+1−0

@@ -163,6 +163,7 @@ static const struct hid_device_id hid_quirks[] = { · 162 unmodified lines
163	163	{ HID_USB_DEVICE(USB_VENDOR_ID_SYNAPTICS, USB_DEVICE_ID_SYNAPTICS_LTS2), HID_QUIRK_NO_INIT_REPORTS },
164	164	{ HID_USB_DEVICE(USB_VENDOR_ID_SYNAPTICS, USB_DEVICE_ID_SYNAPTICS_QUAD_HD), HID_QUIRK_NO_INIT_REPORTS },
165	165	{ HID_USB_DEVICE(USB_VENDOR_ID_SYNAPTICS, USB_DEVICE_ID_SYNAPTICS_TP_V103), HID_QUIRK_NO_INIT_REPORTS },
	166	+ { HID_USB_DEVICE(USB_VENDOR_ID_SYNAPTICS, USB_DEVICE_ID_SYNAPTICS_DELL_K12A), HID_QUIRK_NO_INIT_REPORTS },
166	167	{ HID_USB_DEVICE(USB_VENDOR_ID_TOPMAX, USB_DEVICE_ID_TOPMAX_COBRAPAD), HID_QUIRK_BADPAD },
167	168	{ HID_USB_DEVICE(USB_VENDOR_ID_TOUCHPACK, USB_DEVICE_ID_TOUCHPACK_RTS), HID_QUIRK_MULTI_INPUT },
168	169	{ HID_USB_DEVICE(USB_VENDOR_ID_TPV, USB_DEVICE_ID_TPV_OPTICAL_TOUCHSCREEN_8882), HID_QUIRK_NOGET },

drivers/hid/i2c-hid/i2c-hid-core.c+2−0

@@ -177,6 +177,8 @@ static const struct i2c_hid_quirks { · 176 unmodified lines
177	177	I2C_HID_QUIRK_BOGUS_IRQ },
178	178	{ USB_VENDOR_ID_ALPS_JP, HID_ANY_ID,
179	179	I2C_HID_QUIRK_RESET_ON_RESUME },
	180	+ { I2C_VENDOR_ID_SYNAPTICS, I2C_PRODUCT_ID_SYNAPTICS_SYNA2393,
	181	+ I2C_HID_QUIRK_RESET_ON_RESUME },
180	182	{ USB_VENDOR_ID_ITE, I2C_DEVICE_ID_ITE_LENOVO_LEGION_Y720,
181	183	I2C_HID_QUIRK_BAD_INPUT_SIZE },
182	184	{ 0, 0 }

drivers/hid/usbhid/hid-core.c+29−8

@@ -682,16 +682,21 @@ static int usbhid_open(struct hid_device *hid) · 681 unmodified lines
682	682	struct usbhid_device *usbhid = hid->driver_data;
683	683	int res;
684	684
	685	+ mutex_lock(&usbhid->mutex);
	686	+
685	687	set_bit(HID_OPENED, &usbhid->iofl);
686	688
687		− if (hid->quirks & HID_QUIRK_ALWAYS_POLL)
688		− return 0;
	689	+ if (hid->quirks & HID_QUIRK_ALWAYS_POLL) {
	690	+ res = 0;
	691	+ goto Done;
	692	+ }
689	693
690	694	res = usb_autopm_get_interface(usbhid->intf);
691	695	/* the device must be awake to reliably request remote wakeup */
692	696	if (res < 0) {
693	697	clear_bit(HID_OPENED, &usbhid->iofl);
694		− return -EIO;
	698	+ res = -EIO;
	699	+ goto Done;
695	700	}
696	701
697	702	usbhid->intf->needs_remote_wakeup = 1;
@@ -725,6 +730,9 @@ static int usbhid_open(struct hid_device *hid) · 27 unmodified lines
725	730	msleep(50);
726	731
727	732	clear_bit(HID_RESUME_RUNNING, &usbhid->iofl);
	733	+
	734	+ Done:
	735	+ mutex_unlock(&usbhid->mutex);
728	736	return res;
729	737	}
730	738
@@ -732,6 +740,8 @@ static void usbhid_close(struct hid_device *hid) · 1 unmodified line
732	740	{
733	741	struct usbhid_device *usbhid = hid->driver_data;
734	742
	743	+ mutex_lock(&usbhid->mutex);
	744	+
735	745	/*
736	746	* Make sure we don't restart data acquisition due to
737	747	* a resumption we no longer care about by avoiding racing
@@ -743,12 +753,13 @@ static void usbhid_close(struct hid_device *hid) · 5 unmodified lines
743	753	clear_bit(HID_IN_POLLING, &usbhid->iofl);
744	754	spin_unlock_irq(&usbhid->lock);
745	755
746		− if (hid->quirks & HID_QUIRK_ALWAYS_POLL)
747		− return;
	756	+ if (!(hid->quirks & HID_QUIRK_ALWAYS_POLL)) {
	757	+ hid_cancel_delayed_stuff(usbhid);
	758	+ usb_kill_urb(usbhid->urbin);
	759	+ usbhid->intf->needs_remote_wakeup = 0;
	760	+ }
748	761
749		− hid_cancel_delayed_stuff(usbhid);
750		− usb_kill_urb(usbhid->urbin);
751		− usbhid->intf->needs_remote_wakeup = 0;
	762	+ mutex_unlock(&usbhid->mutex);
752	763	}
753	764
754	765	/*
@@ -1057,6 +1068,8 @@ static int usbhid_start(struct hid_device *hid) · 302 unmodified lines
1057	1068	unsigned int n, insize = 0;
1058	1069	int ret;
1059	1070
	1071	+ mutex_lock(&usbhid->mutex);
	1072	+
1060	1073	clear_bit(HID_DISCONNECTED, &usbhid->iofl);
1061	1074
1062	1075	usbhid->bufsize = HID_MIN_BUFFER_SIZE;
@@ -1177,6 +1190,8 @@ static int usbhid_start(struct hid_device *hid) · 114 unmodified lines
1177	1190	usbhid_set_leds(hid);
1178	1191	device_set_wakeup_enable(&dev->dev, 1);
1179	1192	}
	1193	+
	1194	+ mutex_unlock(&usbhid->mutex);
1180	1195	return 0;
1181	1196
1182	1197	fail:
@@ -1187,6 +1202,7 @@ fail: · 4 unmodified lines
1187	1202	usbhid->urbout = NULL;
1188	1203	usbhid->urbctrl = NULL;
1189	1204	hid_free_buffers(dev, hid);
	1205	+ mutex_unlock(&usbhid->mutex);
1190	1206	return ret;
1191	1207	}
1192	1208
@@ -1202,6 +1218,8 @@ static void usbhid_stop(struct hid_device *hid) · 9 unmodified lines
1202	1218	usbhid->intf->needs_remote_wakeup = 0;
1203	1219	}
1204	1220
	1221	+ mutex_lock(&usbhid->mutex);
	1222	+
1205	1223	clear_bit(HID_STARTED, &usbhid->iofl);
1206	1224	spin_lock_irq(&usbhid->lock); /* Sync with error and led handlers */
1207	1225	set_bit(HID_DISCONNECTED, &usbhid->iofl);
@@ -1222,6 +1240,8 @@ static void usbhid_stop(struct hid_device *hid) · 14 unmodified lines
1222	1240	usbhid->urbout = NULL;
1223	1241
1224	1242	hid_free_buffers(hid_to_usb_dev(hid), hid);
	1243	+
	1244	+ mutex_unlock(&usbhid->mutex);
1225	1245	}
1226	1246
1227	1247	static int usbhid_power(struct hid_device *hid, int lvl)
@@ -1382,6 +1402,7 @@ static int usbhid_probe(struct usb_interface intf, const struct usb_device_id · 154 unmodified lines
1382	1402	INIT_WORK(&usbhid->reset_work, hid_reset);
1383	1403	timer_setup(&usbhid->io_retry, hid_retry_timeout, 0);
1384	1404	spin_lock_init(&usbhid->lock);
	1405	+ mutex_init(&usbhid->mutex);
1385	1406
1386	1407	ret = hid_add_device(hid);
1387	1408	if (ret) {

drivers/hid/usbhid/usbhid.h+1−0

@@ -80,6 +80,7 @@ struct usbhid_device { · 79 unmodified lines
80	80	dma_addr_t outbuf_dma; /* Output buffer dma */
81	81	unsigned long last_out; /* record of last output for timeouts */
82	82
	83	+ struct mutex mutex; /* start/stop/open/close */
83	84	spinlock_t lock; /* fifo spinlock */
84	85	unsigned long iofl; /* I/O flags (CTRL_RUNNING, OUT_RUNNING) */
85	86	struct timer_list io_retry; /* Retry timer */

drivers/hid/wacom_sys.c+3−1

@@ -319,9 +319,11 @@ static void wacom_feature_mapping(struct hid_device *hdev, · 318 unmodified lines
319	319	data[0] = field->report->id;
320	320	ret = wacom_get_report(hdev, HID_FEATURE_REPORT,
321	321	data, n, WAC_CMD_RETRIES);
322		− if (ret == n) {
	322	+ if (ret == n && features->type == HID_GENERIC) {
323	323	ret = hid_report_raw_event(hdev,
324	324	HID_FEATURE_REPORT, data, n, 0);
	325	+ } else if (ret == 2 && features->type != HID_GENERIC) {
	326	+ features->touch_max = data[1];
325	327	} else {
326	328	features->touch_max = 16;
327	329	hid_warn(hdev, "wacom_feature_mapping: "

drivers/hid/wacom_wac.c+22−66

@@ -1427,11 +1427,13 @@ static void wacom_intuos_pro2_bt_pad(struct wacom_wac *wacom) · 1426 unmodified lines
1427	1427	{
1428	1428	struct input_dev *pad_input = wacom->pad_input;
1429	1429	unsigned char *data = wacom->data;
	1430	+ int nbuttons = wacom->features.numbered_buttons;
1430	1431
1431		− int buttons = data[282] \| ((data[281] & 0x40) << 2);
	1432	+ int expresskeys = data[282];
	1433	+ int center = (data[281] & 0x40) >> 6;
1432	1434	int ring = data[285] & 0x7F;
1433	1435	bool ringstatus = data[285] & 0x80;
1434		− bool prox = buttons \|\| ringstatus;
	1436	+ bool prox = expresskeys \|\| center \|\| ringstatus;
1435	1437
1436	1438	/* Fix touchring data: userspace expects 0 at left and increasing clockwise */
1437	1439	ring = 71 - ring;
@@ -1439,7 +1441,8 @@ static void wacom_intuos_pro2_bt_pad(struct wacom_wac *wacom) · 1 unmodified line
1439	1441	if (ring > 71)
1440	1442	ring -= 72;
1441	1443
1442		− wacom_report_numbered_buttons(pad_input, 9, buttons);
	1444	+ wacom_report_numbered_buttons(pad_input, nbuttons,
	1445	+ expresskeys \| (center << (nbuttons - 1)));
1443	1446
1444	1447	input_report_abs(pad_input, ABS_WHEEL, ringstatus ? ring : 0);
1445	1448
@@ -2637,9 +2640,25 @@ static void wacom_wac_finger_pre_report(struct hid_device *hdev, · 1191 unmodified lines
2637	2640	case HID_DG_TIPSWITCH:
2638	2641	hid_data->last_slot_field = equivalent_usage;
2639	2642	break;
	2643	+ case HID_DG_CONTACTCOUNT:
	2644	+ hid_data->cc_report = report->id;
	2645	+ hid_data->cc_index = i;
	2646	+ hid_data->cc_value_index = j;
	2647	+ break;
2640	2648	}
2641	2649	}
2642	2650	}
	2651	+
	2652	+ if (hid_data->cc_report != 0 &&
	2653	+ hid_data->cc_index >= 0) {
	2654	+ struct hid_field *field = report->field[hid_data->cc_index];
	2655	+ int value = field->value[hid_data->cc_value_index];
	2656	+ if (value)
	2657	+ hid_data->num_expected = value;
	2658	+ }
	2659	+ else {
	2660	+ hid_data->num_expected = wacom_wac->features.touch_max;
	2661	+ }
2643	2662	}
2644	2663
2645	2664	static void wacom_wac_finger_report(struct hid_device *hdev,
@@ -2649,7 +2668,6 @@ static void wacom_wac_finger_report(struct hid_device *hdev, · 3 unmodified lines
2649	2668	struct wacom_wac *wacom_wac = &wacom->wacom_wac;
2650	2669	struct input_dev *input = wacom_wac->touch_input;
2651	2670	unsigned touch_max = wacom_wac->features.touch_max;
2652		− struct hid_data *hid_data = &wacom_wac->hid_data;
2653	2671
2654	2672	/* If more packets of data are expected, give us a chance to
2655	2673	* process them rather than immediately syncing a partial
@@ -2663,7 +2681,6 @@ static void wacom_wac_finger_report(struct hid_device *hdev, · 7 unmodified lines
2663	2681
2664	2682	input_sync(input);
2665	2683	wacom_wac->hid_data.num_received = 0;
2666		− hid_data->num_expected = 0;
2667	2684
2668	2685	/* keep touch state for pen event */
2669	2686	wacom_wac->shared->touch_down = wacom_wac_finger_count_touches(wacom_wac);
@@ -2738,73 +2755,12 @@ static void wacom_report_events(struct hid_device *hdev, · 68 unmodified lines
2738	2755	}
2739	2756	}
2740	2757
2741		−static void wacom_set_num_expected(struct hid_device *hdev,
2742		− struct hid_report *report,
2743		− int collection_index,
2744		− struct hid_field *field,
2745		− int field_index)
2746		−{
2747		− struct wacom *wacom = hid_get_drvdata(hdev);
2748		− struct wacom_wac *wacom_wac = &wacom->wacom_wac;
2749		− struct hid_data *hid_data = &wacom_wac->hid_data;
2750		− unsigned int original_collection_level =
2751		− hdev->collection[collection_index].level;
2752		− bool end_collection = false;
2753		− int i;
2754		−
2755		− if (hid_data->num_expected)
2756		− return;
2757		−
2758		− // find the contact count value for this segment
2759		− for (i = field_index; i < report->maxfield && !end_collection; i++) {
2760		− struct hid_field *field = report->field[i];
2761		− unsigned int field_level =
2762		− hdev->collection[field->usage[0].collection_index].level;
2763		− unsigned int j;
2764		−
2765		− if (field_level != original_collection_level)
2766		− continue;
2767		−
2768		− for (j = 0; j < field->maxusage; j++) {
2769		− struct hid_usage *usage = &field->usage[j];
2770		−
2771		− if (usage->collection_index != collection_index) {
2772		− end_collection = true;
2773		− break;
2774		− }
2775		− if (wacom_equivalent_usage(usage->hid) == HID_DG_CONTACTCOUNT) {
2776		− hid_data->cc_report = report->id;
2777		− hid_data->cc_index = i;
2778		− hid_data->cc_value_index = j;
2779		−
2780		− if (hid_data->cc_report != 0 &&
2781		− hid_data->cc_index >= 0) {
2782		−
2783		− struct hid_field *field =
2784		− report->field[hid_data->cc_index];
2785		− int value =
2786		− field->value[hid_data->cc_value_index];
2787		−
2788		− if (value)
2789		− hid_data->num_expected = value;
2790		− }
2791		− }
2792		− }
2793		− }
2794		−
2795		− if (hid_data->cc_report == 0 \|\| hid_data->cc_index < 0)
2796		− hid_data->num_expected = wacom_wac->features.touch_max;
2797		−}
2798		−
2799	2758	static int wacom_wac_collection(struct hid_device hdev, struct hid_report report,
2800	2759	int collection_index, struct hid_field *field,
2801	2760	int field_index)
2802	2761	{
2803	2762	struct wacom *wacom = hid_get_drvdata(hdev);
2804	2763
2805		− if (WACOM_FINGER_FIELD(field))
2806		− wacom_set_num_expected(hdev, report, collection_index, field,
2807		− field_index);
2808	2764	wacom_report_events(hdev, report, collection_index, field_index);
2809	2765
2810	2766	/*

drivers/hv/hv.c+1−5

@@ -184,11 +184,7 @@ void hv_synic_enable_regs(unsigned int cpu) · 183 unmodified lines
184	184
185	185	shared_sint.vector = HYPERVISOR_CALLBACK_VECTOR;
186	186	shared_sint.masked = false;
187		− if (ms_hyperv.hints & HV_DEPRECATING_AEOI_RECOMMENDED)
188		− shared_sint.auto_eoi = false;
189		− else
190		− shared_sint.auto_eoi = true;
191		−
	187	+ shared_sint.auto_eoi = hv_recommend_using_aeoi();
192	188	hv_set_synint_state(VMBUS_MESSAGE_SINT, shared_sint.as_uint64);
193	189
194	190	/* Enable the global synic bit */

drivers/hv/hv_trace.h+2−2

@@ -286,8 +286,8 @@ TRACE_EVENT(vmbus_send_tl_connect_request, · 285 unmodified lines
286	286	__field(int, ret)
287	287	),
288	288	TP_fast_assign(
289		− memcpy(__entry->guest_id, &msg->guest_endpoint_id.b, 16);
290		− memcpy(__entry->host_id, &msg->host_service_id.b, 16);
	289	+ export_guid(__entry->guest_id, &msg->guest_endpoint_id);
	290	+ export_guid(__entry->host_id, &msg->host_service_id);
291	291	__entry->ret = ret;
292	292	),
293	293	TP_printk("sending guest_endpoint_id %pUl, host_service_id %pUl, "

drivers/hv/vmbus_drv.c+34−9

@@ -978,6 +978,9 @@ static int vmbus_resume(struct device *child_device) · 977 unmodified lines
978	978
979	979	return drv->resume(dev);
980	980	}
	981	+#else
	982	+#define vmbus_suspend NULL
	983	+#define vmbus_resume NULL
981	984	#endif /* CONFIG_PM_SLEEP */
982	985
983	986	/*
@@ -997,11 +1000,22 @@ static void vmbus_device_release(struct device *device) · 13 unmodified lines
997	1000	}
998	1001
999	1002	/*
1000		− * Note: we must use SET_NOIRQ_SYSTEM_SLEEP_PM_OPS rather than
1001		− * SET_SYSTEM_SLEEP_PM_OPS: see the comment before vmbus_bus_pm.
	1003	+ * Note: we must use the "noirq" ops: see the comment before vmbus_bus_pm.
	1004	+ *
	1005	+ * suspend_noirq/resume_noirq are set to NULL to support Suspend-to-Idle: we
	1006	+ * shouldn't suspend the vmbus devices upon Suspend-to-Idle, otherwise there
	1007	+ * is no way to wake up a Generation-2 VM.
	1008	+ *
	1009	+ * The other 4 ops are for hibernation.
1002	1010	*/
	1011	+
1003	1012	static const struct dev_pm_ops vmbus_pm = {
1004		− SET_NOIRQ_SYSTEM_SLEEP_PM_OPS(vmbus_suspend, vmbus_resume)
	1013	+ .suspend_noirq = NULL,
	1014	+ .resume_noirq = NULL,
	1015	+ .freeze_noirq = vmbus_suspend,
	1016	+ .thaw_noirq = vmbus_resume,
	1017	+ .poweroff_noirq = vmbus_suspend,
	1018	+ .restore_noirq = vmbus_resume,
1005	1019	};
1006	1020
1007	1021	/* The one and only one */
@@ -2281,6 +2295,9 @@ static int vmbus_bus_resume(struct device *dev) · 1273 unmodified lines
2281	2295
2282	2296	return 0;
2283	2297	}
	2298	+#else
	2299	+#define vmbus_bus_suspend NULL
	2300	+#define vmbus_bus_resume NULL
2284	2301	#endif /* CONFIG_PM_SLEEP */
2285	2302
2286	2303	static const struct acpi_device_id vmbus_acpi_device_ids[] = {
@@ -2291,16 +2308,24 @@ static const struct acpi_device_id vmbus_acpi_device_ids[] = { · 4 unmodified lines
2291	2308	MODULE_DEVICE_TABLE(acpi, vmbus_acpi_device_ids);
2292	2309
2293	2310	/*
2294		− * Note: we must use SET_NOIRQ_SYSTEM_SLEEP_PM_OPS rather than
2295		− * SET_SYSTEM_SLEEP_PM_OPS, otherwise NIC SR-IOV can not work, because the
2296		− * "pci_dev_pm_ops" uses the "noirq" callbacks: in the resume path, the
2297		− * pci "noirq" restore callback runs before "non-noirq" callbacks (see
	2311	+ * Note: we must use the "no_irq" ops, otherwise hibernation can not work with
	2312	+ * PCI device assignment, because "pci_dev_pm_ops" uses the "noirq" ops: in
	2313	+ * the resume path, the pci "noirq" restore op runs before "non-noirq" op (see
2298	2314	* resume_target_kernel() -> dpm_resume_start(), and hibernation_restore() ->
2299	2315	* dpm_resume_end()). This means vmbus_bus_resume() and the pci-hyperv's
2300		− * resume callback must also run via the "noirq" callbacks.
	2316	+ * resume callback must also run via the "noirq" ops.
	2317	+ *
	2318	+ * Set suspend_noirq/resume_noirq to NULL for Suspend-to-Idle: see the comment
	2319	+ * earlier in this file before vmbus_pm.
2301	2320	*/
	2321	+
2302	2322	static const struct dev_pm_ops vmbus_bus_pm = {
2303		− SET_NOIRQ_SYSTEM_SLEEP_PM_OPS(vmbus_bus_suspend, vmbus_bus_resume)
	2323	+ .suspend_noirq = NULL,
	2324	+ .resume_noirq = NULL,
	2325	+ .freeze_noirq = vmbus_bus_suspend,
	2326	+ .thaw_noirq = vmbus_bus_resume,
	2327	+ .poweroff_noirq = vmbus_bus_suspend,
	2328	+ .restore_noirq = vmbus_bus_resume
2304	2329	};
2305	2330
2306	2331	static struct acpi_driver vmbus_acpi_driver = {

drivers/i2c/busses/i2c-amd-mp2-pci.c+1−1

@@ -349,12 +349,12 @@ static int amd_mp2_pci_probe(struct pci_dev *pci_dev, · 348 unmodified lines
349	349	if (!privdata)
350	350	return -ENOMEM;
351	351
	352	+ privdata->pci_dev = pci_dev;
352	353	rc = amd_mp2_pci_init(privdata, pci_dev);
353	354	if (rc)
354	355	return rc;
355	356
356	357	mutex_init(&privdata->c2p_lock);
357		− privdata->pci_dev = pci_dev;
358	358
359	359	pm_runtime_set_autosuspend_delay(&pci_dev->dev, 1000);
360	360	pm_runtime_use_autosuspend(&pci_dev->dev);

drivers/i2c/busses/i2c-aspeed.c+4−1

@@ -603,6 +603,7 @@ static irqreturn_t aspeed_i2c_bus_irq(int irq, void *dev_id) · 602 unmodified lines
603	603	/* Ack all interrupts except for Rx done */
604	604	writel(irq_received & ~ASPEED_I2CD_INTR_RX_DONE,
605	605	bus->base + ASPEED_I2C_INTR_STS_REG);
	606	+ readl(bus->base + ASPEED_I2C_INTR_STS_REG);
606	607	irq_remaining = irq_received;
607	608
608	609	#if IS_ENABLED(CONFIG_I2C_SLAVE)
@@ -645,9 +646,11 @@ static irqreturn_t aspeed_i2c_bus_irq(int irq, void *dev_id) · 36 unmodified lines
645	646	irq_received, irq_handled);
646	647
647	648	/* Ack Rx done */
648		− if (irq_received & ASPEED_I2CD_INTR_RX_DONE)
	649	+ if (irq_received & ASPEED_I2CD_INTR_RX_DONE) {
649	650	writel(ASPEED_I2CD_INTR_RX_DONE,
650	651	bus->base + ASPEED_I2C_INTR_STS_REG);
	652	+ readl(bus->base + ASPEED_I2C_INTR_STS_REG);
	653	+ }
651	654	spin_unlock(&bus->lock);
652	655	return irq_remaining ? IRQ_NONE : IRQ_HANDLED;
653	656	}

drivers/i2c/busses/i2c-bcm-iproc.c+3−0

@@ -360,6 +360,9 @@ static bool bcm_iproc_i2c_slave_isr(struct bcm_iproc_i2c_dev *iproc_i2c, · 359 unmodified lines
360	360	value = (u8)((val >> S_RX_DATA_SHIFT) & S_RX_DATA_MASK);
361	361	i2c_slave_event(iproc_i2c->slave,
362	362	I2C_SLAVE_WRITE_RECEIVED, &value);
	363	+ if (rx_status == I2C_SLAVE_RX_END)
	364	+ i2c_slave_event(iproc_i2c->slave,
	365	+ I2C_SLAVE_STOP, &value);
363	366	}
364	367	} else if (status & BIT(IS_S_TX_UNDERRUN_SHIFT)) {
365	368	/* Master read other than start */

drivers/i2c/busses/i2c-tegra.c+12−24

@@ -996,13 +996,14 @@ tegra_i2c_poll_completion_timeout(struct tegra_i2c_dev *i2c_dev, · 995 unmodified lines
996	996	do {
997	997	u32 status = i2c_readl(i2c_dev, I2C_INT_STATUS);
998	998
999		− if (status)
	999	+ if (status) {
1000	1000	tegra_i2c_isr(i2c_dev->irq, i2c_dev);
1001	1001
1002		− if (completion_done(complete)) {
1003		− s64 delta = ktime_ms_delta(ktimeout, ktime);
	1002	+ if (completion_done(complete)) {
	1003	+ s64 delta = ktime_ms_delta(ktimeout, ktime);
1004	1004
1005		− return msecs_to_jiffies(delta) ?: 1;
	1005	+ return msecs_to_jiffies(delta) ?: 1;
	1006	+ }
1006	1007	}
1007	1008
1008	1009	ktime = ktime_get();
@@ -1029,18 +1030,14 @@ tegra_i2c_wait_completion_timeout(struct tegra_i2c_dev *i2c_dev, · 20 unmodified lines
1029	1030	disable_irq(i2c_dev->irq);
1030	1031
1031	1032	/*
1032		− * Under some rare circumstances (like running KASAN +
1033		− * NFS root) CPU, which handles interrupt, may stuck in
1034		− * uninterruptible state for a significant time. In this
1035		− * case we will get timeout if I2C transfer is running on
1036		− * a sibling CPU, despite of IRQ being raised.
1037		− *
1038		− * In order to handle this rare condition, the IRQ status
1039		− * needs to be checked after timeout.
	1033	+ * There is a chance that completion may happen after IRQ
	1034	+ * synchronization, which is done by disable_irq().
1040	1035	*/
1041		− if (ret == 0)
1042		− ret = tegra_i2c_poll_completion_timeout(i2c_dev,
1043		− complete, 0);
	1036	+ if (ret == 0 && completion_done(complete)) {
	1037	+ dev_warn(i2c_dev->dev,
	1038	+ "completion done after timeout\n");
	1039	+ ret = 1;
	1040	+ }
1044	1041	}
1045	1042
1046	1043	return ret;
@@ -1219,15 +1216,6 @@ static int tegra_i2c_xfer_msg(struct tegra_i2c_dev *i2c_dev, · 172 unmodified lines
1219	1216	time_left = tegra_i2c_wait_completion_timeout(
1220	1217	i2c_dev, &i2c_dev->dma_complete, xfer_time);
1221	1218
1222		− /*
1223		− * Synchronize DMA first, since dmaengine_terminate_sync()
1224		− * performs synchronization after the transfer's termination
1225		− * and we want to get a completion if transfer succeeded.
1226		− */
1227		− dmaengine_synchronize(i2c_dev->msg_read ?
1228		− i2c_dev->rx_dma_chan :
1229		− i2c_dev->tx_dma_chan);
1230		−
1231	1219	dmaengine_terminate_sync(i2c_dev->msg_read ?
1232	1220	i2c_dev->rx_dma_chan :
1233	1221	i2c_dev->tx_dma_chan);

drivers/iio/adc/ad7192.c+47−16

@@ -125,10 +125,10 @@ · 124 unmodified lines
125	125	#define AD7193_CH_AINCOM 0x600 /* AINCOM - AINCOM */
126	126
127	127	/* ID Register Bit Designations (AD7192_REG_ID) */
128		−#define ID_AD7190 0x4
129		−#define ID_AD7192 0x0
130		−#define ID_AD7193 0x2
131		−#define ID_AD7195 0x6
	128	+#define CHIPID_AD7190 0x4
	129	+#define CHIPID_AD7192 0x0
	130	+#define CHIPID_AD7193 0x2
	131	+#define CHIPID_AD7195 0x6
132	132	#define AD7192_ID_MASK 0x0F
133	133
134	134	/* GPOCON Register Bit Designations (AD7192_REG_GPOCON) */
@@ -161,7 +161,20 @@ enum { · 26 unmodified lines
161	161	AD7192_SYSCALIB_FULL_SCALE,
162	162	};
163	163
	164	+enum {
	165	+ ID_AD7190,
	166	+ ID_AD7192,
	167	+ ID_AD7193,
	168	+ ID_AD7195,
	169	+};
	170	+
	171	+struct ad7192_chip_info {
	172	+ unsigned int chip_id;
	173	+ const char *name;
	174	+};
	175	+
164	176	struct ad7192_state {
	177	+ const struct ad7192_chip_info *chip_info;
165	178	struct regulator *avdd;
166	179	struct regulator *dvdd;
167	180	struct clk *mclk;
@@ -172,7 +185,6 @@ struct ad7192_state { · 4 unmodified lines
172	185	u32 conf;
173	186	u32 scale_avail[8][2];
174	187	u8 gpocon;
175		− u8 devid;
176	188	u8 clock_sel;
177	189	struct mutex lock; /* protect sensor state */
178	190	u8 syscalib_mode[8];
@@ -348,7 +360,7 @@ static int ad7192_setup(struct ad7192_state st, struct device_node np) · 169 unmodified lines
348	360
349	361	id &= AD7192_ID_MASK;
350	362
351		− if (id != st->devid)
	363	+ if (id != st->chip_info->chip_id)
352	364	dev_warn(&st->sd.spi->dev, "device ID query failed (0x%X)\n",
353	365	id);
354	366
@@ -363,7 +375,7 @@ static int ad7192_setup(struct ad7192_state st, struct device_node np) · 8 unmodified lines
363	375	st->mode \|= AD7192_MODE_REJ60;
364	376
365	377	refin2_en = of_property_read_bool(np, "adi,refin2-pins-enable");
366		− if (refin2_en && st->devid != ID_AD7195)
	378	+ if (refin2_en && st->chip_info->chip_id != CHIPID_AD7195)
367	379	st->conf \|= AD7192_CONF_REFSEL;
368	380
369	381	st->conf &= ~AD7192_CONF_CHOP;
@@ -859,12 +871,31 @@ static const struct iio_chan_spec ad7193_channels[] = { · 489 unmodified lines
859	871	IIO_CHAN_SOFT_TIMESTAMP(14),
860	872	};
861	873
	874	+static const struct ad7192_chip_info ad7192_chip_info_tbl[] = {
	875	+ [ID_AD7190] = {
	876	+ .chip_id = CHIPID_AD7190,
	877	+ .name = "ad7190",
	878	+ },
	879	+ [ID_AD7192] = {
	880	+ .chip_id = CHIPID_AD7192,
	881	+ .name = "ad7192",
	882	+ },
	883	+ [ID_AD7193] = {
	884	+ .chip_id = CHIPID_AD7193,
	885	+ .name = "ad7193",
	886	+ },
	887	+ [ID_AD7195] = {
	888	+ .chip_id = CHIPID_AD7195,
	889	+ .name = "ad7195",
	890	+ },
	891	+};
	892	+
862	893	static int ad7192_channels_config(struct iio_dev *indio_dev)
863	894	{
864	895	struct ad7192_state *st = iio_priv(indio_dev);
865	896
866		− switch (st->devid) {
867		− case ID_AD7193:
	897	+ switch (st->chip_info->chip_id) {
	898	+ case CHIPID_AD7193:
868	899	indio_dev->channels = ad7193_channels;
869	900	indio_dev->num_channels = ARRAY_SIZE(ad7193_channels);
870	901	break;
@@ -878,10 +909,10 @@ static int ad7192_channels_config(struct iio_dev *indio_dev) · 7 unmodified lines
878	909	}
879	910
880	911	static const struct of_device_id ad7192_of_match[] = {
881		− { .compatible = "adi,ad7190", .data = (void *)ID_AD7190 },
882		− { .compatible = "adi,ad7192", .data = (void *)ID_AD7192 },
883		− { .compatible = "adi,ad7193", .data = (void *)ID_AD7193 },
884		− { .compatible = "adi,ad7195", .data = (void *)ID_AD7195 },
	912	+ { .compatible = "adi,ad7190", .data = &ad7192_chip_info_tbl[ID_AD7190] },
	913	+ { .compatible = "adi,ad7192", .data = &ad7192_chip_info_tbl[ID_AD7192] },
	914	+ { .compatible = "adi,ad7193", .data = &ad7192_chip_info_tbl[ID_AD7193] },
	915	+ { .compatible = "adi,ad7195", .data = &ad7192_chip_info_tbl[ID_AD7195] },
885	916	{}
886	917	};
887	918	MODULE_DEVICE_TABLE(of, ad7192_of_match);
@@ -938,16 +969,16 @@ static int ad7192_probe(struct spi_device *spi) · 50 unmodified lines
938	969	}
939	970
940	971	spi_set_drvdata(spi, indio_dev);
941		− st->devid = (unsigned long)of_device_get_match_data(&spi->dev);
	972	+ st->chip_info = of_device_get_match_data(&spi->dev);
942	973	indio_dev->dev.parent = &spi->dev;
943		− indio_dev->name = spi_get_device_id(spi)->name;
	974	+ indio_dev->name = st->chip_info->name;
944	975	indio_dev->modes = INDIO_DIRECT_MODE;
945	976
946	977	ret = ad7192_channels_config(indio_dev);
947	978	if (ret < 0)
948	979	goto error_disable_dvdd;
949	980
950		− if (st->devid == ID_AD7195)
	981	+ if (st->chip_info->chip_id == CHIPID_AD7195)
951	982	indio_dev->info = &ad7195_info;
952	983	else
953	984	indio_dev->info = &ad7192_info;

drivers/iio/adc/ad7793.c+1−1

@@ -542,7 +542,7 @@ static const struct iio_info ad7797_info = { · 541 unmodified lines
542	542	.read_raw = &ad7793_read_raw,
543	543	.write_raw = &ad7793_write_raw,
544	544	.write_raw_get_fmt = &ad7793_write_raw_get_fmt,
545		− .attrs = &ad7793_attribute_group,
	545	+ .attrs = &ad7797_attribute_group,
546	546	.validate_trigger = ad_sd_validate_trigger,
547	547	};
548	548

drivers/iio/adc/stm32-adc.c+28−3

@@ -1418,8 +1418,30 @@ static unsigned int stm32_adc_dma_residue(struct stm32_adc *adc) · 1417 unmodified lines
1418	1418	static void stm32_adc_dma_buffer_done(void *data)
1419	1419	{
1420	1420	struct iio_dev *indio_dev = data;
	1421	+ struct stm32_adc *adc = iio_priv(indio_dev);
	1422	+ int residue = stm32_adc_dma_residue(adc);
	1423	+
	1424	+ /*
	1425	+ * In DMA mode the trigger services of IIO are not used
	1426	+ * (e.g. no call to iio_trigger_poll).
	1427	+ * Calling irq handler associated to the hardware trigger is not
	1428	+ * relevant as the conversions have already been done. Data
	1429	+ * transfers are performed directly in DMA callback instead.
	1430	+ * This implementation avoids to call trigger irq handler that
	1431	+ * may sleep, in an atomic context (DMA irq handler context).
	1432	+ */
	1433	+ dev_dbg(&indio_dev->dev, "%s bufi=%d\n", __func__, adc->bufi);
1421	1434
1422		− iio_trigger_poll_chained(indio_dev->trig);
	1435	+ while (residue >= indio_dev->scan_bytes) {
	1436	+ u16 buffer = (u16 )&adc->rx_buf[adc->bufi];
	1437	+
	1438	+ iio_push_to_buffers(indio_dev, buffer);
	1439	+
	1440	+ residue -= indio_dev->scan_bytes;
	1441	+ adc->bufi += indio_dev->scan_bytes;
	1442	+ if (adc->bufi >= adc->rx_buf_sz)
	1443	+ adc->bufi = 0;
	1444	+ }
1423	1445	}
1424	1446
1425	1447	static int stm32_adc_dma_start(struct iio_dev *indio_dev)
@@ -1845,6 +1867,7 @@ static int stm32_adc_probe(struct platform_device *pdev) · 419 unmodified lines
1845	1867	{
1846	1868	struct iio_dev *indio_dev;
1847	1869	struct device *dev = &pdev->dev;
	1870	+ irqreturn_t (handler)(int irq, void p) = NULL;
1848	1871	struct stm32_adc *adc;
1849	1872	int ret;
1850	1873
@@ -1911,9 +1934,11 @@ static int stm32_adc_probe(struct platform_device *pdev) · 60 unmodified lines
1911	1934	if (ret < 0)
1912	1935	return ret;
1913	1936
	1937	+ if (!adc->dma_chan)
	1938	+ handler = &stm32_adc_trigger_handler;
	1939	+
1914	1940	ret = iio_triggered_buffer_setup(indio_dev,
1915		− &iio_pollfunc_store_time,
1916		− &stm32_adc_trigger_handler,
	1941	+ &iio_pollfunc_store_time, handler,
1917	1942	&stm32_adc_buffer_setup_ops);
1918	1943	if (ret) {
1919	1944	dev_err(&pdev->dev, "buffer setup failed\n");

drivers/iio/adc/ti-ads8344.c+3−3

@@ -29,7 +29,7 @@ struct ads8344 { · 28 unmodified lines
29	29	struct mutex lock;
30	30
31	31	u8 tx_buf ____cacheline_aligned;
32		− u16 rx_buf;
	32	+ u8 rx_buf[3];
33	33	};
34	34
35	35	#define ADS8344_VOLTAGE_CHANNEL(chan, si) \
@@ -89,11 +89,11 @@ static int ads8344_adc_conversion(struct ads8344 *adc, int channel, · 53 unmodified lines
89	89
90	90	udelay(9);
91	91
92		− ret = spi_read(spi, &adc->rx_buf, 2);
	92	+ ret = spi_read(spi, adc->rx_buf, sizeof(adc->rx_buf));
93	93	if (ret)
94	94	return ret;
95	95
96		− return adc->rx_buf;
	96	+ return adc->rx_buf[0] << 9 \| adc->rx_buf[1] << 1 \| adc->rx_buf[2] >> 7;
97	97	}
98	98
99	99	static int ads8344_read_raw(struct iio_dev *iio,

drivers/iio/adc/xilinx-xadc-core.c+74−21

@@ -102,6 +102,16 @@ static const unsigned int XADC_ZYNQ_UNMASK_TIMEOUT = 500; · 101 unmodified lines
102	102
103	103	#define XADC_FLAGS_BUFFERED BIT(0)
104	104
	105	+/*
	106	+ * The XADC hardware supports a samplerate of up to 1MSPS. Unfortunately it does
	107	+ * not have a hardware FIFO. Which means an interrupt is generated for each
	108	+ * conversion sequence. At 1MSPS sample rate the CPU in ZYNQ7000 is completely
	109	+ * overloaded by the interrupts that it soft-lockups. For this reason the driver
	110	+ * limits the maximum samplerate 150kSPS. At this rate the CPU is fairly busy,
	111	+ * but still responsive.
	112	+ */
	113	+#define XADC_MAX_SAMPLERATE 150000
	114	+
105	115	static void xadc_write_reg(struct xadc *xadc, unsigned int reg,
106	116	uint32_t val)
107	117	{
@@ -674,7 +684,7 @@ static int xadc_trigger_set_state(struct iio_trigger *trigger, bool state) · 566 unmodified lines
674	684
675	685	spin_lock_irqsave(&xadc->lock, flags);
676	686	xadc_read_reg(xadc, XADC_AXI_REG_IPIER, &val);
677		− xadc_write_reg(xadc, XADC_AXI_REG_IPISR, val & XADC_AXI_INT_EOS);
	687	+ xadc_write_reg(xadc, XADC_AXI_REG_IPISR, XADC_AXI_INT_EOS);
678	688	if (state)
679	689	val \|= XADC_AXI_INT_EOS;
680	690	else
@@ -722,13 +732,14 @@ static int xadc_power_adc_b(struct xadc *xadc, unsigned int seq_mode) · 41 unmodified lines
722	732	{
723	733	uint16_t val;
724	734
	735	+ /* Powerdown the ADC-B when it is not needed. */
725	736	switch (seq_mode) {
726	737	case XADC_CONF1_SEQ_SIMULTANEOUS:
727	738	case XADC_CONF1_SEQ_INDEPENDENT:
728		− val = XADC_CONF2_PD_ADC_B;
	739	+ val = 0;
729	740	break;
730	741	default:
731		− val = 0;
	742	+ val = XADC_CONF2_PD_ADC_B;
732	743	break;
733	744	}
734	745
@@ -797,6 +808,16 @@ static int xadc_preenable(struct iio_dev *indio_dev) · 62 unmodified lines
797	808	if (ret)
798	809	goto err;
799	810
	811	+ /*
	812	+ * In simultaneous mode the upper and lower aux channels are samples at
	813	+ * the same time. In this mode the upper 8 bits in the sequencer
	814	+ * register are don't care and the lower 8 bits control two channels
	815	+ * each. As such we must set the bit if either the channel in the lower
	816	+ * group or the upper group is enabled.
	817	+ */
	818	+ if (seq_mode == XADC_CONF1_SEQ_SIMULTANEOUS)
	819	+ scan_mask = ((scan_mask >> 8) \| scan_mask) & 0xff0000;
	820	+
800	821	ret = xadc_write_adc_reg(xadc, XADC_REG_SEQ(1), scan_mask >> 16);
801	822	if (ret)
802	823	goto err;
@@ -823,11 +844,27 @@ static const struct iio_buffer_setup_ops xadc_buffer_ops = { · 20 unmodified lines
823	844	.postdisable = &xadc_postdisable,
824	845	};
825	846
	847	+static int xadc_read_samplerate(struct xadc *xadc)
	848	+{
	849	+ unsigned int div;
	850	+ uint16_t val16;
	851	+ int ret;
	852	+
	853	+ ret = xadc_read_adc_reg(xadc, XADC_REG_CONF2, &val16);
	854	+ if (ret)
	855	+ return ret;
	856	+
	857	+ div = (val16 & XADC_CONF2_DIV_MASK) >> XADC_CONF2_DIV_OFFSET;
	858	+ if (div < 2)
	859	+ div = 2;
	860	+
	861	+ return xadc_get_dclk_rate(xadc) / div / 26;
	862	+}
	863	+
826	864	static int xadc_read_raw(struct iio_dev *indio_dev,
827	865	struct iio_chan_spec const chan, int val, int *val2, long info)
828	866	{
829	867	struct xadc *xadc = iio_priv(indio_dev);
830		− unsigned int div;
831	868	uint16_t val16;
832	869	int ret;
833	870
@@ -880,41 +917,31 @@ static int xadc_read_raw(struct iio_dev *indio_dev, · 46 unmodified lines
880	917	*val = -((273150 << 12) / 503975);
881	918	return IIO_VAL_INT;
882	919	case IIO_CHAN_INFO_SAMP_FREQ:
883		− ret = xadc_read_adc_reg(xadc, XADC_REG_CONF2, &val16);
884		− if (ret)
	920	+ ret = xadc_read_samplerate(xadc);
	921	+ if (ret < 0)
885	922	return ret;
886	923
887		− div = (val16 & XADC_CONF2_DIV_MASK) >> XADC_CONF2_DIV_OFFSET;
888		− if (div < 2)
889		− div = 2;
890		−
891		− *val = xadc_get_dclk_rate(xadc) / div / 26;
892		−
	924	+ *val = ret;
893	925	return IIO_VAL_INT;
894	926	default:
895	927	return -EINVAL;
896	928	}
897	929	}
898	930
899		−static int xadc_write_raw(struct iio_dev *indio_dev,
900		− struct iio_chan_spec const *chan, int val, int val2, long info)
	931	+static int xadc_write_samplerate(struct xadc *xadc, int val)
901	932	{
902		− struct xadc *xadc = iio_priv(indio_dev);
903	933	unsigned long clk_rate = xadc_get_dclk_rate(xadc);
904	934	unsigned int div;
905	935
906	936	if (!clk_rate)
907	937	return -EINVAL;
908	938
909		− if (info != IIO_CHAN_INFO_SAMP_FREQ)
910		− return -EINVAL;
911		−
912	939	if (val <= 0)
913	940	return -EINVAL;
914	941
915	942	/* Max. 150 kSPS */
916		− if (val > 150000)
917		− val = 150000;
	943	+ if (val > XADC_MAX_SAMPLERATE)
	944	+ val = XADC_MAX_SAMPLERATE;
918	945
919	946	val *= 26;
920	947
@@ -927,7 +954,7 @@ static int xadc_write_raw(struct iio_dev *indio_dev, · 6 unmodified lines
927	954	* limit.
928	955	*/
929	956	div = clk_rate / val;
930		− if (clk_rate / div / 26 > 150000)
	957	+ if (clk_rate / div / 26 > XADC_MAX_SAMPLERATE)
931	958	div++;
932	959	if (div < 2)
933	960	div = 2;
@@ -938,6 +965,17 @@ static int xadc_write_raw(struct iio_dev *indio_dev, · 4 unmodified lines
938	965	div << XADC_CONF2_DIV_OFFSET);
939	966	}
940	967
	968	+static int xadc_write_raw(struct iio_dev *indio_dev,
	969	+ struct iio_chan_spec const *chan, int val, int val2, long info)
	970	+{
	971	+ struct xadc *xadc = iio_priv(indio_dev);
	972	+
	973	+ if (info != IIO_CHAN_INFO_SAMP_FREQ)
	974	+ return -EINVAL;
	975	+
	976	+ return xadc_write_samplerate(xadc, val);
	977	+}
	978	+
941	979	static const struct iio_event_spec xadc_temp_events[] = {
942	980	{
943	981	.type = IIO_EV_TYPE_THRESH,
@@ -1223,6 +1261,21 @@ static int xadc_probe(struct platform_device *pdev) · 279 unmodified lines
1223	1261	if (ret)
1224	1262	goto err_free_samplerate_trigger;
1225	1263
	1264	+ /*
	1265	+ * Make sure not to exceed the maximum samplerate since otherwise the
	1266	+ * resulting interrupt storm will soft-lock the system.
	1267	+ */
	1268	+ if (xadc->ops->flags & XADC_FLAGS_BUFFERED) {
	1269	+ ret = xadc_read_samplerate(xadc);
	1270	+ if (ret < 0)
	1271	+ goto err_free_samplerate_trigger;
	1272	+ if (ret > XADC_MAX_SAMPLERATE) {
	1273	+ ret = xadc_write_samplerate(xadc, XADC_MAX_SAMPLERATE);
	1274	+ if (ret < 0)
	1275	+ goto err_free_samplerate_trigger;
	1276	+ }
	1277	+ }
	1278	+
1226	1279	ret = request_irq(xadc->irq, xadc->ops->interrupt_handler, 0,
1227	1280	dev_name(&pdev->dev), indio_dev);
1228	1281	if (ret)

drivers/iio/common/st_sensors/st_sensors_core.c+1−1

@@ -79,7 +79,7 @@ int st_sensors_set_odr(struct iio_dev *indio_dev, unsigned int odr) · 78 unmodified lines
79	79	struct st_sensor_odr_avl odr_out = {0, 0};
80	80	struct st_sensor_data *sdata = iio_priv(indio_dev);
81	81
82		− if (!sdata->sensor_settings->odr.addr)
	82	+ if (!sdata->sensor_settings->odr.mask)
83	83	return 0;
84	84
85	85	err = st_sensors_match_odr(sdata->sensor_settings, odr, &odr_out);

drivers/iio/dac/ad5770r.c+1−1

@@ -525,7 +525,7 @@ static int ad5770r_channel_config(struct ad5770r_state *st) · 524 unmodified lines
525	525	ret = fwnode_property_read_u32(child, "num", &num);
526	526	if (ret)
527	527	return ret;
528		− if (num > AD5770R_MAX_CHANNELS)
	528	+ if (num >= AD5770R_MAX_CHANNELS)
529	529	return -EINVAL;
530	530
531	531	ret = fwnode_property_read_u32_array(child,

drivers/iio/imu/inv_mpu6050/inv_mpu_core.c+10−1

@@ -1617,6 +1617,10 @@ static int __maybe_unused inv_mpu_resume(struct device *dev) · 1616 unmodified lines
1617	1617	if (result)
1618	1618	goto out_unlock;
1619	1619
	1620	+ pm_runtime_disable(dev);
	1621	+ pm_runtime_set_active(dev);
	1622	+ pm_runtime_enable(dev);
	1623	+
1620	1624	result = inv_mpu6050_switch_engine(st, true, st->suspended_sensors);
1621	1625	if (result)
1622	1626	goto out_unlock;
@@ -1638,13 +1642,18 @@ static int __maybe_unused inv_mpu_suspend(struct device *dev) · 15 unmodified lines
1638	1642
1639	1643	mutex_lock(&st->lock);
1640	1644
	1645	+ st->suspended_sensors = 0;
	1646	+ if (pm_runtime_suspended(dev)) {
	1647	+ result = 0;
	1648	+ goto out_unlock;
	1649	+ }
	1650	+
1641	1651	if (iio_buffer_enabled(indio_dev)) {
1642	1652	result = inv_mpu6050_prepare_fifo(st, false);
1643	1653	if (result)
1644	1654	goto out_unlock;
1645	1655	}
1646	1656
1647		− st->suspended_sensors = 0;
1648	1657	if (st->chip_config.accl_en)
1649	1658	st->suspended_sensors \|= INV_MPU6050_SENSOR_ACCL;
1650	1659	if (st->chip_config.gyro_en)

drivers/iio/imu/st_lsm6dsx/st_lsm6dsx.h+3−0

@@ -337,6 +337,7 @@ enum st_lsm6dsx_fifo_mode { · 336 unmodified lines
337	337	* @gain: Configured sensor sensitivity.
338	338	* @odr: Output data rate of the sensor [Hz].
339	339	* @watermark: Sensor watermark level.
	340	+ * @decimator: Sensor decimation factor.
340	341	* @sip: Number of samples in a given pattern.
341	342	* @ts_ref: Sensor timestamp reference for hw one.
342	343	* @ext_info: Sensor settings if it is connected to i2c controller
@@ -350,11 +351,13 @@ struct st_lsm6dsx_sensor { · 7 unmodified lines
350	351	u32 odr;
351	352
352	353	u16 watermark;
	354	+ u8 decimator;
353	355	u8 sip;
354	356	s64 ts_ref;
355	357
356	358	struct {
357	359	const struct st_lsm6dsx_ext_dev_settings *settings;
	360	+ u32 slv_odr;
358	361	u8 addr;
359	362	} ext_info;
360	363	};

drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_buffer.c+16−7

@@ -93,6 +93,7 @@ st_lsm6dsx_get_decimator_val(struct st_lsm6dsx_sensor *sensor, u32 max_odr) · 92 unmodified lines
93	93	break;
94	94	}
95	95
	96	+ sensor->decimator = decimator;
96	97	return i == max_size ? 0 : st_lsm6dsx_decimator_table[i].val;
97	98	}
98	99
@@ -337,7 +338,7 @@ static inline int st_lsm6dsx_read_block(struct st_lsm6dsx_hw *hw, u8 addr, · 238 unmodified lines
337	338	int st_lsm6dsx_read_fifo(struct st_lsm6dsx_hw *hw)
338	339	{
339	340	struct st_lsm6dsx_sensor acc_sensor, gyro_sensor, *ext_sensor = NULL;
340		− int err, acc_sip, gyro_sip, ts_sip, ext_sip, read_len, offset;
	341	+ int err, sip, acc_sip, gyro_sip, ts_sip, ext_sip, read_len, offset;
341	342	u16 fifo_len, pattern_len = hw->sip * ST_LSM6DSX_SAMPLE_SIZE;
342	343	u16 fifo_diff_mask = hw->settings->fifo_ops.fifo_diff.mask;
343	344	u8 gyro_buff[ST_LSM6DSX_IIO_BUFF_SIZE];
@@ -399,19 +400,20 @@ int st_lsm6dsx_read_fifo(struct st_lsm6dsx_hw *hw) · 55 unmodified lines
399	400	acc_sip = acc_sensor->sip;
400	401	ts_sip = hw->ts_sip;
401	402	offset = 0;
	403	+ sip = 0;
402	404
403	405	while (acc_sip > 0 \|\| gyro_sip > 0 \|\| ext_sip > 0) {
404		− if (gyro_sip > 0) {
	406	+ if (gyro_sip > 0 && !(sip % gyro_sensor->decimator)) {
405	407	memcpy(gyro_buff, &hw->buff[offset],
406	408	ST_LSM6DSX_SAMPLE_SIZE);
407	409	offset += ST_LSM6DSX_SAMPLE_SIZE;
408	410	}
409		− if (acc_sip > 0) {
	411	+ if (acc_sip > 0 && !(sip % acc_sensor->decimator)) {
410	412	memcpy(acc_buff, &hw->buff[offset],
411	413	ST_LSM6DSX_SAMPLE_SIZE);
412	414	offset += ST_LSM6DSX_SAMPLE_SIZE;
413	415	}
414		− if (ext_sip > 0) {
	416	+ if (ext_sip > 0 && !(sip % ext_sensor->decimator)) {
415	417	memcpy(ext_buff, &hw->buff[offset],
416	418	ST_LSM6DSX_SAMPLE_SIZE);
417	419	offset += ST_LSM6DSX_SAMPLE_SIZE;
@@ -441,18 +443,25 @@ int st_lsm6dsx_read_fifo(struct st_lsm6dsx_hw *hw) · 23 unmodified lines
441	443	offset += ST_LSM6DSX_SAMPLE_SIZE;
442	444	}
443	445
444		− if (gyro_sip-- > 0)
	446	+ if (gyro_sip > 0 && !(sip % gyro_sensor->decimator)) {
445	447	iio_push_to_buffers_with_timestamp(
446	448	hw->iio_devs[ST_LSM6DSX_ID_GYRO],
447	449	gyro_buff, gyro_sensor->ts_ref + ts);
448		− if (acc_sip-- > 0)
	450	+ gyro_sip--;
	451	+ }
	452	+ if (acc_sip > 0 && !(sip % acc_sensor->decimator)) {
449	453	iio_push_to_buffers_with_timestamp(
450	454	hw->iio_devs[ST_LSM6DSX_ID_ACC],
451	455	acc_buff, acc_sensor->ts_ref + ts);
452		− if (ext_sip-- > 0)
	456	+ acc_sip--;
	457	+ }
	458	+ if (ext_sip > 0 && !(sip % ext_sensor->decimator)) {
453	459	iio_push_to_buffers_with_timestamp(
454	460	hw->iio_devs[ST_LSM6DSX_ID_EXT0],
455	461	ext_buff, ext_sensor->ts_ref + ts);
	462	+ ext_sip--;
	463	+ }
	464	+ sip++;
456	465	}
457	466	}
458	467

drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c+23−1

@@ -2036,11 +2036,21 @@ static int st_lsm6dsx_init_hw_timer(struct st_lsm6dsx_hw *hw) · 2035 unmodified lines
2036	2036	return 0;
2037	2037	}
2038	2038
2039		−static int st_lsm6dsx_init_device(struct st_lsm6dsx_hw *hw)
	2039	+static int st_lsm6dsx_reset_device(struct st_lsm6dsx_hw *hw)
2040	2040	{
2041	2041	const struct st_lsm6dsx_reg *reg;
2042	2042	int err;
2043	2043
	2044	+ /*
	2045	+ * flush hw FIFO before device reset in order to avoid
	2046	+ * possible races on interrupt line 1. If the first interrupt
	2047	+ * line is asserted during hw reset the device will work in
	2048	+ * I3C-only mode (if it is supported)
	2049	+ */
	2050	+ err = st_lsm6dsx_flush_fifo(hw);
	2051	+ if (err < 0 && err != -ENOTSUPP)
	2052	+ return err;
	2053	+
2044	2054	/* device sw reset */
2045	2055	reg = &hw->settings->reset;
2046	2056	err = regmap_update_bits(hw->regmap, reg->addr, reg->mask,
@@ -2059,6 +2069,18 @@ static int st_lsm6dsx_init_device(struct st_lsm6dsx_hw *hw) · 12 unmodified lines
2059	2069
2060	2070	msleep(50);
2061	2071
	2072	+ return 0;
	2073	+}
	2074	+
	2075	+static int st_lsm6dsx_init_device(struct st_lsm6dsx_hw *hw)
	2076	+{
	2077	+ const struct st_lsm6dsx_reg *reg;
	2078	+ int err;
	2079	+
	2080	+ err = st_lsm6dsx_reset_device(hw);
	2081	+ if (err < 0)
	2082	+ return err;
	2083	+
2062	2084	/* enable Block Data Update */
2063	2085	reg = &hw->settings->bdu;
2064	2086	err = regmap_update_bits(hw->regmap, reg->addr, reg->mask,

drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_shub.c+22−7

@@ -421,7 +421,8 @@ int st_lsm6dsx_shub_set_enable(struct st_lsm6dsx_sensor *sensor, bool enable) · 420 unmodified lines
421	421
422	422	settings = sensor->ext_info.settings;
423	423	if (enable) {
424		− err = st_lsm6dsx_shub_set_odr(sensor, sensor->odr);
	424	+ err = st_lsm6dsx_shub_set_odr(sensor,
	425	+ sensor->ext_info.slv_odr);
425	426	if (err < 0)
426	427	return err;
427	428	} else {
@@ -459,7 +460,7 @@ st_lsm6dsx_shub_read_oneshot(struct st_lsm6dsx_sensor *sensor, · 31 unmodified lines
459	460	if (err < 0)
460	461	return err;
461	462
462		− delay = 1000000000 / sensor->odr;
	463	+ delay = 1000000000 / sensor->ext_info.slv_odr;
463	464	usleep_range(delay, 2 * delay);
464	465
465	466	len = min_t(int, sizeof(data), ch->scan_type.realbits >> 3);
@@ -500,8 +501,8 @@ st_lsm6dsx_shub_read_raw(struct iio_dev *iio_dev, · 34 unmodified lines
500	501	iio_device_release_direct_mode(iio_dev);
501	502	break;
502	503	case IIO_CHAN_INFO_SAMP_FREQ:
503		− *val = sensor->odr / 1000;
504		− val2 = (sensor->odr % 1000) 1000;
	504	+ *val = sensor->ext_info.slv_odr / 1000;
	505	+ val2 = (sensor->ext_info.slv_odr % 1000) 1000;
505	506	ret = IIO_VAL_INT_PLUS_MICRO;
506	507	break;
507	508	case IIO_CHAN_INFO_SCALE:
@@ -535,8 +536,20 @@ st_lsm6dsx_shub_write_raw(struct iio_dev *iio_dev, · 27 unmodified lines
535	536
536	537	val = val * 1000 + val2 / 1000;
537	538	err = st_lsm6dsx_shub_get_odr_val(sensor, val, &data);
538		− if (!err)
539		− sensor->odr = val;
	539	+ if (!err) {
	540	+ struct st_lsm6dsx_hw *hw = sensor->hw;
	541	+ struct st_lsm6dsx_sensor *ref_sensor;
	542	+ u8 odr_val;
	543	+ int odr;
	544	+
	545	+ ref_sensor = iio_priv(hw->iio_devs[ST_LSM6DSX_ID_ACC]);
	546	+ odr = st_lsm6dsx_check_odr(ref_sensor, val, &odr_val);
	547	+ if (odr < 0)
	548	+ return odr;
	549	+
	550	+ sensor->ext_info.slv_odr = val;
	551	+ sensor->odr = odr;
	552	+ }
540	553	break;
541	554	}
542	555	default:
@@ -613,6 +626,7 @@ st_lsm6dsx_shub_alloc_iiodev(struct st_lsm6dsx_hw *hw, · 70 unmodified lines
613	626	const struct st_lsm6dsx_ext_dev_settings *info,
614	627	u8 i2c_addr, const char *name)
615	628	{
	629	+ enum st_lsm6dsx_sensor_id ref_id = ST_LSM6DSX_ID_ACC;
616	630	struct iio_chan_spec *ext_channels;
617	631	struct st_lsm6dsx_sensor *sensor;
618	632	struct iio_dev *iio_dev;
@@ -628,7 +642,8 @@ st_lsm6dsx_shub_alloc_iiodev(struct st_lsm6dsx_hw *hw, · 9 unmodified lines
628	642	sensor = iio_priv(iio_dev);
629	643	sensor->id = id;
630	644	sensor->hw = hw;
631		− sensor->odr = info->odr_table.odr_avl[0].milli_hz;
	645	+ sensor->odr = hw->settings->odr_table[ref_id].odr_avl[0].milli_hz;
	646	+ sensor->ext_info.slv_odr = info->odr_table.odr_avl[0].milli_hz;
632	647	sensor->gain = info->fs_table.fs_avl[0].gain;
633	648	sensor->ext_info.settings = info;
634	649	sensor->ext_info.addr = i2c_addr;

drivers/iio/industrialio-core.c+2−5

@@ -915,14 +915,11 @@ static ssize_t iio_write_channel_info(struct device *dev, · 914 unmodified lines
915	915	return -EINVAL;
916	916	integer = ch;
917	917	} else {
918		− ret = iio_str_to_fixpoint(buf, fract_mult, &integer, &fract);
	918	+ ret = __iio_str_to_fixpoint(buf, fract_mult, &integer, &fract,
	919	+ scale_db);
919	920	if (ret)
920	921	return ret;
921	922	}
922		− ret = __iio_str_to_fixpoint(buf, fract_mult, &integer, &fract,
923		− scale_db);
924		− if (ret)
925		− return ret;
926	923
927	924	ret = indio_dev->info->write_raw(indio_dev, this_attr->c,
928	925	integer, fract, this_attr->address);

drivers/infiniband/core/cm.c+14−12

@@ -862,7 +862,7 @@ static struct cm_id_private cm_alloc_id_priv(struct ib_device device, · 861 unmodified lines
862	862
863	863	ret = xa_alloc_cyclic_irq(&cm.local_id_table, &id, NULL, xa_limit_32b,
864	864	&cm.local_id_next, GFP_KERNEL);
865		− if (ret)
	865	+ if (ret < 0)
866	866	goto error;
867	867	cm_id_priv->id.local_id = (__force __be32)id ^ cm.random_id_operand;
868	868
@@ -1828,11 +1828,9 @@ static void cm_format_mra(struct cm_mra_msg *mra_msg, · 959 unmodified lines
1828	1828
1829	1829	static void cm_format_rej(struct cm_rej_msg *rej_msg,
1830	1830	struct cm_id_private *cm_id_priv,
1831		− enum ib_cm_rej_reason reason,
1832		− void *ari,
1833		− u8 ari_length,
1834		− const void *private_data,
1835		− u8 private_data_len)
	1831	+ enum ib_cm_rej_reason reason, void *ari,
	1832	+ u8 ari_length, const void *private_data,
	1833	+ u8 private_data_len, enum ib_cm_state state)
1836	1834	{
1837	1835	lockdep_assert_held(&cm_id_priv->lock);
1838	1836
@@ -1840,7 +1838,7 @@ static void cm_format_rej(struct cm_rej_msg *rej_msg, · 1 unmodified line
1840	1838	IBA_SET(CM_REJ_REMOTE_COMM_ID, rej_msg,
1841	1839	be32_to_cpu(cm_id_priv->id.remote_id));
1842	1840
1843		− switch(cm_id_priv->id.state) {
	1841	+ switch (state) {
1844	1842	case IB_CM_REQ_RCVD:
1845	1843	IBA_SET(CM_REJ_LOCAL_COMM_ID, rej_msg, be32_to_cpu(0));
1846	1844	IBA_SET(CM_REJ_MESSAGE_REJECTED, rej_msg, CM_MSG_RESPONSE_REQ);
@@ -1905,8 +1903,9 @@ static void cm_dup_req_handler(struct cm_work *work, · 58 unmodified lines
1905	1903	cm_id_priv->private_data_len);
1906	1904	break;
1907	1905	case IB_CM_TIMEWAIT:
1908		− cm_format_rej((struct cm_rej_msg *) msg->mad, cm_id_priv,
1909		− IB_CM_REJ_STALE_CONN, NULL, 0, NULL, 0);
	1906	+ cm_format_rej((struct cm_rej_msg *)msg->mad, cm_id_priv,
	1907	+ IB_CM_REJ_STALE_CONN, NULL, 0, NULL, 0,
	1908	+ IB_CM_TIMEWAIT);
1910	1909	break;
1911	1910	default:
1912	1911	goto unlock;
@@ -2904,6 +2903,7 @@ static int cm_send_rej_locked(struct cm_id_private *cm_id_priv, · 991 unmodified lines
2904	2903	u8 ari_length, const void *private_data,
2905	2904	u8 private_data_len)
2906	2905	{
	2906	+ enum ib_cm_state state = cm_id_priv->id.state;
2907	2907	struct ib_mad_send_buf *msg;
2908	2908	int ret;
2909	2909
@@ -2913,7 +2913,7 @@ static int cm_send_rej_locked(struct cm_id_private *cm_id_priv, · 3 unmodified lines
2913	2913	(ari && ari_length > IB_CM_REJ_ARI_LENGTH))
2914	2914	return -EINVAL;
2915	2915
2916		− switch (cm_id_priv->id.state) {
	2916	+ switch (state) {
2917	2917	case IB_CM_REQ_SENT:
2918	2918	case IB_CM_MRA_REQ_RCVD:
2919	2919	case IB_CM_REQ_RCVD:
@@ -2925,7 +2925,8 @@ static int cm_send_rej_locked(struct cm_id_private *cm_id_priv, · 5 unmodified lines
2925	2925	if (ret)
2926	2926	return ret;
2927	2927	cm_format_rej((struct cm_rej_msg *)msg->mad, cm_id_priv, reason,
2928		− ari, ari_length, private_data, private_data_len);
	2928	+ ari, ari_length, private_data, private_data_len,
	2929	+ state);
2929	2930	break;
2930	2931	case IB_CM_REP_SENT:
2931	2932	case IB_CM_MRA_REP_RCVD:
@@ -2934,7 +2935,8 @@ static int cm_send_rej_locked(struct cm_id_private *cm_id_priv, · 2 unmodified lines
2934	2935	if (ret)
2935	2936	return ret;
2936	2937	cm_format_rej((struct cm_rej_msg *)msg->mad, cm_id_priv, reason,
2937		− ari, ari_length, private_data, private_data_len);
	2938	+ ari, ari_length, private_data, private_data_len,
	2939	+ state);
2938	2940	break;
2939	2941	default:
2940	2942	pr_debug("%s: local_id %d, cm_id->state: %d\n", __func__,

drivers/infiniband/core/rdma_core.c+4−5

@@ -360,7 +360,7 @@ lookup_get_fd_uobject(const struct uverbs_api_object *obj, · 359 unmodified lines
360	360	* uverbs_uobject_fd_release(), and the caller is expected to ensure
361	361	* that release is never done while a call to lookup is possible.
362	362	*/
363		− if (f->f_op != fd_type->fops) {
	363	+ if (f->f_op != fd_type->fops \|\| uobject->ufile != ufile) {
364	364	fput(f);
365	365	return ERR_PTR(-EBADF);
366	366	}
@@ -474,16 +474,15 @@ alloc_begin_fd_uobject(const struct uverbs_api_object *obj, · 107 unmodified lines
474	474	filp = anon_inode_getfile(fd_type->name, fd_type->fops, NULL,
475	475	fd_type->flags);
476	476	if (IS_ERR(filp)) {
	477	+ uverbs_uobject_put(uobj);
477	478	uobj = ERR_CAST(filp);
478		− goto err_uobj;
	479	+ goto err_fd;
479	480	}
480	481	uobj->object = filp;
481	482
482	483	uobj->id = new_fd;
483	484	return uobj;
484	485
485		−err_uobj:
486		− uverbs_uobject_put(uobj);
487	486	err_fd:
488	487	put_unused_fd(new_fd);
489	488	return uobj;
@@ -679,7 +678,6 @@ void rdma_lookup_put_uobject(struct ib_uobject *uobj, · 189 unmodified lines
679	678	enum rdma_lookup_mode mode)
680	679	{
681	680	assert_uverbs_usecnt(uobj, mode);
682		− uobj->uapi_object->type_class->lookup_put(uobj, mode);
683	681	/*
684	682	* In order to unlock an object, either decrease its usecnt for
685	683	* read access or zero it in case of exclusive access. See
@@ -696,6 +694,7 @@ void rdma_lookup_put_uobject(struct ib_uobject *uobj, · 10 unmodified lines
696	694	break;
697	695	}
698	696
	697	+ uobj->uapi_object->type_class->lookup_put(uobj, mode);
699	698	/* Pairs with the kref obtained by type->lookup_get */
700	699	uverbs_uobject_put(uobj);
701	700	}

drivers/infiniband/core/uverbs_main.c+4−0

@@ -820,6 +820,10 @@ void uverbs_user_mmap_disassociate(struct ib_uverbs_file *ufile) · 819 unmodified lines
820	820	ret = mmget_not_zero(mm);
821	821	if (!ret) {
822	822	list_del_init(&priv->list);
	823	+ if (priv->entry) {
	824	+ rdma_user_mmap_entry_put(priv->entry);
	825	+ priv->entry = NULL;
	826	+ }
823	827	mm = NULL;
824	828	continue;
825	829	}

drivers/infiniband/hw/i40iw/i40iw_ctrl.c+1−1

@@ -1046,7 +1046,7 @@ i40iw_sc_query_rdma_features(struct i40iw_sc_cqp *cqp, · 1045 unmodified lines
1046	1046	u64 header;
1047	1047
1048	1048	wqe = i40iw_sc_cqp_get_next_send_wqe(cqp, scratch);
1049		− if (wqe)
	1049	+ if (!wqe)
1050	1050	return I40IW_ERR_RING_FULL;
1051	1051
1052	1052	set_64bit_val(wqe, 32, feat_mem->pa);

drivers/infiniband/hw/mlx4/main.c+2−1

@@ -1499,8 +1499,9 @@ static int __mlx4_ib_create_default_rules( · 1498 unmodified lines
1499	1499	int i;
1500	1500
1501	1501	for (i = 0; i < ARRAY_SIZE(pdefault_rules->rules_create_list); i++) {
	1502	+ union ib_flow_spec ib_spec = {};
1502	1503	int ret;
1503		− union ib_flow_spec ib_spec;
	1504	+
1504	1505	switch (pdefault_rules->rules_create_list[i]) {
1505	1506	case 0:
1506	1507	/* no rule */

drivers/infiniband/hw/mlx5/qp.c+3−1

@@ -5558,7 +5558,9 @@ static void to_rdma_ah_attr(struct mlx5_ib_dev *ibdev, · 5557 unmodified lines
5558	5558	rdma_ah_set_path_bits(ah_attr, path->grh_mlid & 0x7f);
5559	5559	rdma_ah_set_static_rate(ah_attr,
5560	5560	path->static_rate ? path->static_rate - 5 : 0);
5561		− if (path->grh_mlid & (1 << 7)) {
	5561	+
	5562	+ if (path->grh_mlid & (1 << 7) \|\|
	5563	+ ah_attr->type == RDMA_AH_ATTR_TYPE_ROCE) {
5562	5564	u32 tc_fl = be32_to_cpu(path->tclass_flowlabel);
5563	5565
5564	5566	rdma_ah_set_grh(ah_attr, NULL,

drivers/infiniband/sw/rdmavt/cq.c+2−2

@@ -248,8 +248,8 @@ int rvt_create_cq(struct ib_cq ibcq, const struct ib_cq_init_attr attr, · 247 unmodified lines
248	248	*/
249	249	if (udata && udata->outlen >= sizeof(__u64)) {
250	250	cq->ip = rvt_create_mmap_info(rdi, sz, udata, u_wc);
251		− if (!cq->ip) {
252		− err = -ENOMEM;
	251	+ if (IS_ERR(cq->ip)) {
	252	+ err = PTR_ERR(cq->ip);
253	253	goto bail_wc;
254	254	}
255	255

drivers/infiniband/sw/rdmavt/mmap.c+2−2

@@ -154,7 +154,7 @@ done: · 153 unmodified lines
154	154	* @udata: user data (must be valid!)
155	155	* @obj: opaque pointer to a cq, wq etc
156	156	*
157		− * Return: rvt_mmap struct on success
	157	+ * Return: rvt_mmap struct on success, ERR_PTR on failure
158	158	*/
159	159	struct rvt_mmap_info rvt_create_mmap_info(struct rvt_dev_info rdi, u32 size,
160	160	struct ib_udata udata, void obj)
@@ -166,7 +166,7 @@ struct rvt_mmap_info rvt_create_mmap_info(struct rvt_dev_info rdi, u32 size, · 5 unmodified lines
166	166
167	167	ip = kmalloc_node(sizeof(*ip), GFP_KERNEL, rdi->dparms.node);
168	168	if (!ip)
169		− return ip;
	169	+ return ERR_PTR(-ENOMEM);
170	170
171	171	size = PAGE_ALIGN(size);
172	172

drivers/infiniband/sw/rdmavt/qp.c+2−2

@@ -1244,8 +1244,8 @@ struct ib_qp rvt_create_qp(struct ib_pd ibpd, · 1243 unmodified lines
1244	1244
1245	1245	qp->ip = rvt_create_mmap_info(rdi, s, udata,
1246	1246	qp->r_rq.wq);
1247		− if (!qp->ip) {
1248		− ret = ERR_PTR(-ENOMEM);
	1247	+ if (IS_ERR(qp->ip)) {
	1248	+ ret = ERR_CAST(qp->ip);
1249	1249	goto bail_qpn;
1250	1250	}
1251	1251

drivers/infiniband/sw/rdmavt/srq.c+2−2

@@ -111,8 +111,8 @@ int rvt_create_srq(struct ib_srq ibsrq, struct ib_srq_init_attr srq_init_attr, · 110 unmodified lines
111	111	u32 s = sizeof(struct rvt_rwq) + srq->rq.size * sz;
112	112
113	113	srq->ip = rvt_create_mmap_info(dev, s, udata, srq->rq.wq);
114		− if (!srq->ip) {
115		− ret = -ENOMEM;
	114	+ if (IS_ERR(srq->ip)) {
	115	+ ret = PTR_ERR(srq->ip);
116	116	goto bail_wq;
117	117	}
118	118

drivers/infiniband/sw/siw/siw_qp_tx.c+11−4

@@ -920,20 +920,27 @@ static int siw_fastreg_mr(struct ib_pd pd, struct siw_sqe sqe) · 919 unmodified lines
920	920	{
921	921	struct ib_mr base_mr = (struct ib_mr )(uintptr_t)sqe->base_mr;
922	922	struct siw_device *sdev = to_siw_dev(pd->device);
923		− struct siw_mem *mem = siw_mem_id2obj(sdev, sqe->rkey >> 8);
	923	+ struct siw_mem *mem;
924	924	int rv = 0;
925	925
926	926	siw_dbg_pd(pd, "STag 0x%08x\n", sqe->rkey);
927	927
928		− if (unlikely(!mem \|\| !base_mr)) {
	928	+ if (unlikely(!base_mr)) {
929	929	pr_warn("siw: fastreg: STag 0x%08x unknown\n", sqe->rkey);
930	930	return -EINVAL;
931	931	}
	932	+
932	933	if (unlikely(base_mr->rkey >> 8 != sqe->rkey >> 8)) {
933	934	pr_warn("siw: fastreg: STag 0x%08x: bad MR\n", sqe->rkey);
934		− rv = -EINVAL;
935		− goto out;
	935	+ return -EINVAL;
936	936	}
	937	+
	938	+ mem = siw_mem_id2obj(sdev, sqe->rkey >> 8);
	939	+ if (unlikely(!mem)) {
	940	+ pr_warn("siw: fastreg: STag 0x%08x unknown\n", sqe->rkey);
	941	+ return -EINVAL;
	942	+ }
	943	+
937	944	if (unlikely(mem->pd != pd)) {
938	945	pr_warn("siw: fastreg: PD mismatch\n");
939	946	rv = -EINVAL;

drivers/interconnect/qcom/bcm-voter.c+3−2

@@ -96,6 +96,8 @@ static inline void tcs_cmd_gen(struct tcs_cmd *cmd, u64 vote_x, u64 vote_y, · 95 unmodified lines
96	96	if (!cmd)
97	97	return;
98	98
	99	+ memset(cmd, 0, sizeof(*cmd));
	100	+
99	101	if (vote_x == 0 && vote_y == 0)
100	102	valid = false;
101	103
@@ -112,8 +114,7 @@ static inline void tcs_cmd_gen(struct tcs_cmd *cmd, u64 vote_x, u64 vote_y, · 10 unmodified lines
112	114	* Set the wait for completion flag on command that need to be completed
113	115	* before the next command.
114	116	*/
115		− if (commit)
116		− cmd->wait = true;
	117	+ cmd->wait = commit;
117	118	}
118	119
119	120	static void tcs_list_gen(struct list_head *bcm_list, int bucket,

drivers/interconnect/qcom/osm-l3.c+2−2

@@ -78,7 +78,7 @@ static struct qcom_icc_node *sdm845_osm_l3_nodes[] = { · 77 unmodified lines
78	78	[SLAVE_OSM_L3] = &sdm845_osm_l3,
79	79	};
80	80
81		−const static struct qcom_icc_desc sdm845_icc_osm_l3 = {
	81	+static const struct qcom_icc_desc sdm845_icc_osm_l3 = {
82	82	.nodes = sdm845_osm_l3_nodes,
83	83	.num_nodes = ARRAY_SIZE(sdm845_osm_l3_nodes),
84	84	};
@@ -91,7 +91,7 @@ static struct qcom_icc_node *sc7180_osm_l3_nodes[] = { · 6 unmodified lines
91	91	[SLAVE_OSM_L3] = &sc7180_osm_l3,
92	92	};
93	93
94		−const static struct qcom_icc_desc sc7180_icc_osm_l3 = {
	94	+static const struct qcom_icc_desc sc7180_icc_osm_l3 = {
95	95	.nodes = sc7180_osm_l3_nodes,
96	96	.num_nodes = ARRAY_SIZE(sc7180_osm_l3_nodes),
97	97	};

drivers/interconnect/qcom/sdm845.c+8−8

@@ -192,7 +192,7 @@ static struct qcom_icc_node *aggre1_noc_nodes[] = { · 191 unmodified lines
192	192	[SLAVE_ANOC_PCIE_A1NOC_SNOC] = &qns_pcie_a1noc_snoc,
193	193	};
194	194
195		−const static struct qcom_icc_desc sdm845_aggre1_noc = {
	195	+static const struct qcom_icc_desc sdm845_aggre1_noc = {
196	196	.nodes = aggre1_noc_nodes,
197	197	.num_nodes = ARRAY_SIZE(aggre1_noc_nodes),
198	198	.bcms = aggre1_noc_bcms,
@@ -220,7 +220,7 @@ static struct qcom_icc_node *aggre2_noc_nodes[] = { · 21 unmodified lines
220	220	[SLAVE_SERVICE_A2NOC] = &srvc_aggre2_noc,
221	221	};
222	222
223		−const static struct qcom_icc_desc sdm845_aggre2_noc = {
	223	+static const struct qcom_icc_desc sdm845_aggre2_noc = {
224	224	.nodes = aggre2_noc_nodes,
225	225	.num_nodes = ARRAY_SIZE(aggre2_noc_nodes),
226	226	.bcms = aggre2_noc_bcms,
@@ -281,7 +281,7 @@ static struct qcom_icc_node *config_noc_nodes[] = { · 54 unmodified lines
281	281	[SLAVE_SERVICE_CNOC] = &srvc_cnoc,
282	282	};
283	283
284		−const static struct qcom_icc_desc sdm845_config_noc = {
	284	+static const struct qcom_icc_desc sdm845_config_noc = {
285	285	.nodes = config_noc_nodes,
286	286	.num_nodes = ARRAY_SIZE(config_noc_nodes),
287	287	.bcms = config_noc_bcms,
@@ -297,7 +297,7 @@ static struct qcom_icc_node *dc_noc_nodes[] = { · 9 unmodified lines
297	297	[SLAVE_MEM_NOC_CFG] = &qhs_memnoc,
298	298	};
299	299
300		−const static struct qcom_icc_desc sdm845_dc_noc = {
	300	+static const struct qcom_icc_desc sdm845_dc_noc = {
301	301	.nodes = dc_noc_nodes,
302	302	.num_nodes = ARRAY_SIZE(dc_noc_nodes),
303	303	.bcms = dc_noc_bcms,
@@ -315,7 +315,7 @@ static struct qcom_icc_node *gladiator_noc_nodes[] = { · 11 unmodified lines
315	315	[SLAVE_SERVICE_GNOC] = &srvc_gnoc,
316	316	};
317	317
318		−const static struct qcom_icc_desc sdm845_gladiator_noc = {
	318	+static const struct qcom_icc_desc sdm845_gladiator_noc = {
319	319	.nodes = gladiator_noc_nodes,
320	320	.num_nodes = ARRAY_SIZE(gladiator_noc_nodes),
321	321	.bcms = gladiator_noc_bcms,
@@ -350,7 +350,7 @@ static struct qcom_icc_node *mem_noc_nodes[] = { · 28 unmodified lines
350	350	[SLAVE_EBI1] = &ebi,
351	351	};
352	352
353		−const static struct qcom_icc_desc sdm845_mem_noc = {
	353	+static const struct qcom_icc_desc sdm845_mem_noc = {
354	354	.nodes = mem_noc_nodes,
355	355	.num_nodes = ARRAY_SIZE(mem_noc_nodes),
356	356	.bcms = mem_noc_bcms,
@@ -384,7 +384,7 @@ static struct qcom_icc_node *mmss_noc_nodes[] = { · 27 unmodified lines
384	384	[SLAVE_CAMNOC_UNCOMP] = &qns_camnoc_uncomp,
385	385	};
386	386
387		−const static struct qcom_icc_desc sdm845_mmss_noc = {
	387	+static const struct qcom_icc_desc sdm845_mmss_noc = {
388	388	.nodes = mmss_noc_nodes,
389	389	.num_nodes = ARRAY_SIZE(mmss_noc_nodes),
390	390	.bcms = mmss_noc_bcms,
@@ -430,7 +430,7 @@ static struct qcom_icc_node *system_noc_nodes[] = { · 39 unmodified lines
430	430	[SLAVE_TCU] = &xs_sys_tcu_cfg,
431	431	};
432	432
433		−const static struct qcom_icc_desc sdm845_system_noc = {
	433	+static const struct qcom_icc_desc sdm845_system_noc = {
434	434	.nodes = system_noc_nodes,
435	435	.num_nodes = ARRAY_SIZE(system_noc_nodes),
436	436	.bcms = system_noc_bcms,

drivers/iommu/Kconfig+2−2

@@ -362,7 +362,7 @@ config IPMMU_VMSA · 361 unmodified lines
362	362
363	363	config SPAPR_TCE_IOMMU
364	364	bool "sPAPR TCE IOMMU Support"
365		− depends on PPC_POWERNV \|\| PPC_PSERIES \|\| (PPC && COMPILE_TEST)
	365	+ depends on PPC_POWERNV \|\| PPC_PSERIES
366	366	select IOMMU_API
367	367	help
368	368	Enables bits of IOMMU API required by VFIO. The iommu_ops
@@ -457,7 +457,7 @@ config S390_AP_IOMMU · 88 unmodified lines
457	457
458	458	config MTK_IOMMU
459	459	bool "MTK IOMMU Support"
460		− depends on ARM \|\| ARM64 \|\| COMPILE_TEST
	460	+ depends on HAS_DMA
461	461	depends on ARCH_MEDIATEK \|\| COMPILE_TEST
462	462	select ARM_DMA_USE_IOMMU
463	463	select IOMMU_API

drivers/iommu/amd_iommu.c+154−44

@@ -101,6 +101,8 @@ struct kmem_cache *amd_iommu_irq_cache; · 100 unmodified lines
101	101	static void update_domain(struct protection_domain *domain);
102	102	static int protection_domain_init(struct protection_domain *domain);
103	103	static void detach_device(struct device *dev);
	104	+static void update_and_flush_device_table(struct protection_domain *domain,
	105	+ struct domain_pgtable *pgtable);
104	106
105	107	/****************************************************************************
106	108	*
@@ -151,6 +153,26 @@ static struct protection_domain to_pdomain(struct iommu_domain dom) · 44 unmodified lines
151	153	return container_of(dom, struct protection_domain, domain);
152	154	}
153	155
	156	+static void amd_iommu_domain_get_pgtable(struct protection_domain *domain,
	157	+ struct domain_pgtable *pgtable)
	158	+{
	159	+ u64 pt_root = atomic64_read(&domain->pt_root);
	160	+
	161	+ pgtable->root = (u64 *)(pt_root & PAGE_MASK);
	162	+ pgtable->mode = pt_root & 7; /* lowest 3 bits encode pgtable mode */
	163	+}
	164	+
	165	+static u64 amd_iommu_domain_encode_pgtable(u64 *root, int mode)
	166	+{
	167	+ u64 pt_root;
	168	+
	169	+ /* lowest 3 bits encode pgtable mode */
	170	+ pt_root = mode & 7;
	171	+ pt_root \|= (u64)root;
	172	+
	173	+ return pt_root;
	174	+}
	175	+
154	176	static struct iommu_dev_data *alloc_dev_data(u16 devid)
155	177	{
156	178	struct iommu_dev_data *dev_data;
@@ -1397,13 +1419,18 @@ static struct page *free_sub_pt(unsigned long root, int mode, · 1240 unmodified lines
1397	1419
1398	1420	static void free_pagetable(struct protection_domain *domain)
1399	1421	{
1400		− unsigned long root = (unsigned long)domain->pt_root;
	1422	+ struct domain_pgtable pgtable;
1401	1423	struct page *freelist = NULL;
	1424	+ unsigned long root;
1402	1425
1403		− BUG_ON(domain->mode < PAGE_MODE_NONE \|\|
1404		− domain->mode > PAGE_MODE_6_LEVEL);
	1426	+ amd_iommu_domain_get_pgtable(domain, &pgtable);
	1427	+ atomic64_set(&domain->pt_root, 0);
1405	1428
1406		− freelist = free_sub_pt(root, domain->mode, freelist);
	1429	+ BUG_ON(pgtable.mode < PAGE_MODE_NONE \|\|
	1430	+ pgtable.mode > PAGE_MODE_6_LEVEL);
	1431	+
	1432	+ root = (unsigned long)pgtable.root;
	1433	+ freelist = free_sub_pt(root, pgtable.mode, freelist);
1407	1434
1408	1435	free_page_list(freelist);
1409	1436	}
@@ -1417,24 +1444,39 @@ static bool increase_address_space(struct protection_domain *domain, · 7 unmodified lines
1417	1444	unsigned long address,
1418	1445	gfp_t gfp)
1419	1446	{
	1447	+ struct domain_pgtable pgtable;
1420	1448	unsigned long flags;
1421		− bool ret = false;
1422		− u64 *pte;
	1449	+ bool ret = true;
	1450	+ u64 *pte, root;
1423	1451
1424	1452	spin_lock_irqsave(&domain->lock, flags);
1425	1453
1426		− if (address <= PM_LEVEL_SIZE(domain->mode) \|\|
1427		− WARN_ON_ONCE(domain->mode == PAGE_MODE_6_LEVEL))
	1454	+ amd_iommu_domain_get_pgtable(domain, &pgtable);
	1455	+
	1456	+ if (address <= PM_LEVEL_SIZE(pgtable.mode))
	1457	+ goto out;
	1458	+
	1459	+ ret = false;
	1460	+ if (WARN_ON_ONCE(pgtable.mode == PAGE_MODE_6_LEVEL))
1428	1461	goto out;
1429	1462
1430	1463	pte = (void *)get_zeroed_page(gfp);
1431	1464	if (!pte)
1432	1465	goto out;
1433	1466
1434		− *pte = PM_LEVEL_PDE(domain->mode,
1435		− iommu_virt_to_phys(domain->pt_root));
1436		− domain->pt_root = pte;
1437		− domain->mode += 1;
	1467	+ *pte = PM_LEVEL_PDE(pgtable.mode, iommu_virt_to_phys(pgtable.root));
	1468	+
	1469	+ pgtable.root = pte;
	1470	+ pgtable.mode += 1;
	1471	+ update_and_flush_device_table(domain, &pgtable);
	1472	+ domain_flush_complete(domain);
	1473	+
	1474	+ /*
	1475	+ * Device Table needs to be updated and flushed before the new root can
	1476	+ * be published.
	1477	+ */
	1478	+ root = amd_iommu_domain_encode_pgtable(pte, pgtable.mode);
	1479	+ atomic64_set(&domain->pt_root, root);
1438	1480
1439	1481	ret = true;
1440	1482
@@ -1451,16 +1493,29 @@ static u64 alloc_pte(struct protection_domain domain, · 10 unmodified lines
1451	1493	gfp_t gfp,
1452	1494	bool *updated)
1453	1495	{
	1496	+ struct domain_pgtable pgtable;
1454	1497	int level, end_lvl;
1455	1498	u64 pte, page;
1456	1499
1457	1500	BUG_ON(!is_power_of_2(page_size));
1458	1501
1459		− while (address > PM_LEVEL_SIZE(domain->mode))
1460		− updated = increase_address_space(domain, address, gfp) \|\| updated;
	1502	+ amd_iommu_domain_get_pgtable(domain, &pgtable);
	1503	+
	1504	+ while (address > PM_LEVEL_SIZE(pgtable.mode)) {
	1505	+ /*
	1506	+ * Return an error if there is no memory to update the
	1507	+ * page-table.
	1508	+ */
	1509	+ if (!increase_address_space(domain, address, gfp))
	1510	+ return NULL;
	1511	+
	1512	+ /* Read new values to check if update was successful */
	1513	+ amd_iommu_domain_get_pgtable(domain, &pgtable);
	1514	+ }
	1515	+
1461	1516
1462		− level = domain->mode - 1;
1463		− pte = &domain->pt_root[PM_LEVEL_INDEX(level, address)];
	1517	+ level = pgtable.mode - 1;
	1518	+ pte = &pgtable.root[PM_LEVEL_INDEX(level, address)];
1464	1519	address = PAGE_SIZE_ALIGN(address, page_size);
1465	1520	end_lvl = PAGE_SIZE_LEVEL(page_size);
1466	1521
@@ -1536,16 +1591,19 @@ static u64 fetch_pte(struct protection_domain domain, · 69 unmodified lines
1536	1591	unsigned long address,
1537	1592	unsigned long *page_size)
1538	1593	{
	1594	+ struct domain_pgtable pgtable;
1539	1595	int level;
1540	1596	u64 *pte;
1541	1597
1542	1598	*page_size = 0;
1543	1599
1544		− if (address > PM_LEVEL_SIZE(domain->mode))
	1600	+ amd_iommu_domain_get_pgtable(domain, &pgtable);
	1601	+
	1602	+ if (address > PM_LEVEL_SIZE(pgtable.mode))
1545	1603	return NULL;
1546	1604
1547		− level = domain->mode - 1;
1548		− pte = &domain->pt_root[PM_LEVEL_INDEX(level, address)];
	1605	+ level = pgtable.mode - 1;
	1606	+ pte = &pgtable.root[PM_LEVEL_INDEX(level, address)];
1549	1607	*page_size = PTE_LEVEL_PAGE_SIZE(level);
1550	1608
1551	1609	while (level > 0) {
@@ -1660,7 +1718,13 @@ out: · 108 unmodified lines
1660	1718	unsigned long flags;
1661	1719
1662	1720	spin_lock_irqsave(&dom->lock, flags);
1663		− update_domain(dom);
	1721	+ /*
	1722	+ * Flush domain TLB(s) and wait for completion. Any Device-Table
	1723	+ * Updates and flushing already happened in
	1724	+ * increase_address_space().
	1725	+ */
	1726	+ domain_flush_tlb_pde(dom);
	1727	+ domain_flush_complete(dom);
1664	1728	spin_unlock_irqrestore(&dom->lock, flags);
1665	1729	}
1666	1730
@@ -1806,6 +1870,7 @@ static void dma_ops_domain_free(struct protection_domain *domain) · 139 unmodified lines
1806	1870	static struct protection_domain *dma_ops_domain_alloc(void)
1807	1871	{
1808	1872	struct protection_domain *domain;
	1873	+ u64 *pt_root, root;
1809	1874
1810	1875	domain = kzalloc(sizeof(struct protection_domain), GFP_KERNEL);
1811	1876	if (!domain)
@@ -1814,12 +1879,14 @@ static struct protection_domain *dma_ops_domain_alloc(void) · 2 unmodified lines
1814	1879	if (protection_domain_init(domain))
1815	1880	goto free_domain;
1816	1881
1817		− domain->mode = PAGE_MODE_3_LEVEL;
1818		− domain->pt_root = (void *)get_zeroed_page(GFP_KERNEL);
1819		− domain->flags = PD_DMA_OPS_MASK;
1820		− if (!domain->pt_root)
	1882	+ pt_root = (void *)get_zeroed_page(GFP_KERNEL);
	1883	+ if (!pt_root)
1821	1884	goto free_domain;
1822	1885
	1886	+ root = amd_iommu_domain_encode_pgtable(pt_root, PAGE_MODE_3_LEVEL);
	1887	+ atomic64_set(&domain->pt_root, root);
	1888	+ domain->flags = PD_DMA_OPS_MASK;
	1889	+
1823	1890	if (iommu_get_dma_cookie(&domain->domain) == -ENOMEM)
1824	1891	goto free_domain;
1825	1892
@@ -1841,16 +1908,17 @@ static bool dma_ops_domain(struct protection_domain *domain) · 15 unmodified lines
1841	1908	}
1842	1909
1843	1910	static void set_dte_entry(u16 devid, struct protection_domain *domain,
	1911	+ struct domain_pgtable *pgtable,
1844	1912	bool ats, bool ppr)
1845	1913	{
1846	1914	u64 pte_root = 0;
1847	1915	u64 flags = 0;
1848	1916	u32 old_domid;
1849	1917
1850		− if (domain->mode != PAGE_MODE_NONE)
1851		− pte_root = iommu_virt_to_phys(domain->pt_root);
	1918	+ if (pgtable->mode != PAGE_MODE_NONE)
	1919	+ pte_root = iommu_virt_to_phys(pgtable->root);
1852	1920
1853		− pte_root \|= (domain->mode & DEV_ENTRY_MODE_MASK)
	1921	+ pte_root \|= (pgtable->mode & DEV_ENTRY_MODE_MASK)
1854	1922	<< DEV_ENTRY_MODE_SHIFT;
1855	1923	pte_root \|= DTE_FLAG_IR \| DTE_FLAG_IW \| DTE_FLAG_V \| DTE_FLAG_TV;
1856	1924
@@ -1923,6 +1991,7 @@ static void clear_dte_entry(u16 devid) · 66 unmodified lines
1923	1991	static void do_attach(struct iommu_dev_data *dev_data,
1924	1992	struct protection_domain *domain)
1925	1993	{
	1994	+ struct domain_pgtable pgtable;
1926	1995	struct amd_iommu *iommu;
1927	1996	bool ats;
1928	1997
@@ -1938,7 +2007,9 @@ static void do_attach(struct iommu_dev_data *dev_data, · 9 unmodified lines
1938	2007	domain->dev_cnt += 1;
1939	2008
1940	2009	/* Update device table */
1941		− set_dte_entry(dev_data->devid, domain, ats, dev_data->iommu_v2);
	2010	+ amd_iommu_domain_get_pgtable(domain, &pgtable);
	2011	+ set_dte_entry(dev_data->devid, domain, &pgtable,
	2012	+ ats, dev_data->iommu_v2);
1942	2013	clone_aliases(dev_data->pdev);
1943	2014
1944	2015	device_flush_dte(dev_data);
@@ -2249,23 +2320,36 @@ static int amd_iommu_domain_get_attr(struct iommu_domain *domain, · 304 unmodified lines
2249	2320	*
2250	2321	*****************************************************************************/
2251	2322
2252		−static void update_device_table(struct protection_domain *domain)
	2323	+static void update_device_table(struct protection_domain *domain,
	2324	+ struct domain_pgtable *pgtable)
2253	2325	{
2254	2326	struct iommu_dev_data *dev_data;
2255	2327
2256	2328	list_for_each_entry(dev_data, &domain->dev_list, list) {
2257		− set_dte_entry(dev_data->devid, domain, dev_data->ats.enabled,
2258		− dev_data->iommu_v2);
	2329	+ set_dte_entry(dev_data->devid, domain, pgtable,
	2330	+ dev_data->ats.enabled, dev_data->iommu_v2);
2259	2331	clone_aliases(dev_data->pdev);
2260	2332	}
2261	2333	}
2262	2334
	2335	+static void update_and_flush_device_table(struct protection_domain *domain,
	2336	+ struct domain_pgtable *pgtable)
	2337	+{
	2338	+ update_device_table(domain, pgtable);
	2339	+ domain_flush_devices(domain);
	2340	+}
	2341	+
2263	2342	static void update_domain(struct protection_domain *domain)
2264	2343	{
2265		− update_device_table(domain);
	2344	+ struct domain_pgtable pgtable;
2266	2345
2267		− domain_flush_devices(domain);
	2346	+ /* Update device table */
	2347	+ amd_iommu_domain_get_pgtable(domain, &pgtable);
	2348	+ update_and_flush_device_table(domain, &pgtable);
	2349	+
	2350	+ /* Flush domain TLB(s) and wait for completion */
2268	2351	domain_flush_tlb_pde(domain);
	2352	+ domain_flush_complete(domain);
2269	2353	}
2270	2354
2271	2355	int __init amd_iommu_init_api(void)
@@ -2375,6 +2459,7 @@ out_err: · 103 unmodified lines
2375	2459	static struct iommu_domain *amd_iommu_domain_alloc(unsigned type)
2376	2460	{
2377	2461	struct protection_domain *pdomain;
	2462	+ u64 *pt_root, root;
2378	2463
2379	2464	switch (type) {
2380	2465	case IOMMU_DOMAIN_UNMANAGED:
@@ -2382,13 +2467,15 @@ static struct iommu_domain *amd_iommu_domain_alloc(unsigned type) · 1 unmodified line
2382	2467	if (!pdomain)
2383	2468	return NULL;
2384	2469
2385		− pdomain->mode = PAGE_MODE_3_LEVEL;
2386		− pdomain->pt_root = (void *)get_zeroed_page(GFP_KERNEL);
2387		− if (!pdomain->pt_root) {
	2470	+ pt_root = (void *)get_zeroed_page(GFP_KERNEL);
	2471	+ if (!pt_root) {
2388	2472	protection_domain_free(pdomain);
2389	2473	return NULL;
2390	2474	}
2391	2475
	2476	+ root = amd_iommu_domain_encode_pgtable(pt_root, PAGE_MODE_3_LEVEL);
	2477	+ atomic64_set(&pdomain->pt_root, root);
	2478	+
2392	2479	pdomain->domain.geometry.aperture_start = 0;
2393	2480	pdomain->domain.geometry.aperture_end = ~0ULL;
2394	2481	pdomain->domain.geometry.force_aperture = true;
@@ -2406,7 +2493,7 @@ static struct iommu_domain *amd_iommu_domain_alloc(unsigned type) · 11 unmodified lines
2406	2493	if (!pdomain)
2407	2494	return NULL;
2408	2495
2409		− pdomain->mode = PAGE_MODE_NONE;
	2496	+ atomic64_set(&pdomain->pt_root, PAGE_MODE_NONE);
2410	2497	break;
2411	2498	default:
2412	2499	return NULL;
@@ -2418,6 +2505,7 @@ static struct iommu_domain *amd_iommu_domain_alloc(unsigned type) · 5 unmodified lines
2418	2505	static void amd_iommu_domain_free(struct iommu_domain *dom)
2419	2506	{
2420	2507	struct protection_domain *domain;
	2508	+ struct domain_pgtable pgtable;
2421	2509
2422	2510	domain = to_pdomain(dom);
2423	2511
@@ -2435,7 +2523,9 @@ static void amd_iommu_domain_free(struct iommu_domain *dom) · 11 unmodified lines
2435	2523	dma_ops_domain_free(domain);
2436	2524	break;
2437	2525	default:
2438		− if (domain->mode != PAGE_MODE_NONE)
	2526	+ amd_iommu_domain_get_pgtable(domain, &pgtable);
	2527	+
	2528	+ if (pgtable.mode != PAGE_MODE_NONE)
2439	2529	free_pagetable(domain);
2440	2530
2441	2531	if (domain->flags & PD_IOMMUV2_MASK)
@@ -2518,10 +2608,12 @@ static int amd_iommu_map(struct iommu_domain *dom, unsigned long iova, · 76 unmodified lines
2518	2608	gfp_t gfp)
2519	2609	{
2520	2610	struct protection_domain *domain = to_pdomain(dom);
	2611	+ struct domain_pgtable pgtable;
2521	2612	int prot = 0;
2522	2613	int ret;
2523	2614
2524		− if (domain->mode == PAGE_MODE_NONE)
	2615	+ amd_iommu_domain_get_pgtable(domain, &pgtable);
	2616	+ if (pgtable.mode == PAGE_MODE_NONE)
2525	2617	return -EINVAL;
2526	2618
2527	2619	if (iommu_prot & IOMMU_READ)
@@ -2541,8 +2633,10 @@ static size_t amd_iommu_unmap(struct iommu_domain *dom, unsigned long iova, · 13 unmodified lines
2541	2633	struct iommu_iotlb_gather *gather)
2542	2634	{
2543	2635	struct protection_domain *domain = to_pdomain(dom);
	2636	+ struct domain_pgtable pgtable;
2544	2637
2545		− if (domain->mode == PAGE_MODE_NONE)
	2638	+ amd_iommu_domain_get_pgtable(domain, &pgtable);
	2639	+ if (pgtable.mode == PAGE_MODE_NONE)
2546	2640	return 0;
2547	2641
2548	2642	return iommu_unmap_page(domain, iova, page_size);
@@ -2553,9 +2647,11 @@ static phys_addr_t amd_iommu_iova_to_phys(struct iommu_domain *dom, · 4 unmodified lines
2553	2647	{
2554	2648	struct protection_domain *domain = to_pdomain(dom);
2555	2649	unsigned long offset_mask, pte_pgsize;
	2650	+ struct domain_pgtable pgtable;
2556	2651	u64 *pte, __pte;
2557	2652
2558		− if (domain->mode == PAGE_MODE_NONE)
	2653	+ amd_iommu_domain_get_pgtable(domain, &pgtable);
	2654	+ if (pgtable.mode == PAGE_MODE_NONE)
2559	2655	return iova;
2560	2656
2561	2657	pte = fetch_pte(domain, iova, &pte_pgsize);
@@ -2708,16 +2804,26 @@ EXPORT_SYMBOL(amd_iommu_unregister_ppr_notifier); · 146 unmodified lines
2708	2804	void amd_iommu_domain_direct_map(struct iommu_domain *dom)
2709	2805	{
2710	2806	struct protection_domain *domain = to_pdomain(dom);
	2807	+ struct domain_pgtable pgtable;
2711	2808	unsigned long flags;
	2809	+ u64 pt_root;
2712	2810
2713	2811	spin_lock_irqsave(&domain->lock, flags);
2714	2812
	2813	+ /* First save pgtable configuration*/
	2814	+ amd_iommu_domain_get_pgtable(domain, &pgtable);
	2815	+
2715	2816	/* Update data structure */
2716		− domain->mode = PAGE_MODE_NONE;
	2817	+ pt_root = amd_iommu_domain_encode_pgtable(NULL, PAGE_MODE_NONE);
	2818	+ atomic64_set(&domain->pt_root, pt_root);
2717	2819
2718	2820	/* Make changes visible to IOMMUs */
2719	2821	update_domain(domain);
2720	2822
	2823	+ /* Restore old pgtable in domain->ptroot to free page-table */
	2824	+ pt_root = amd_iommu_domain_encode_pgtable(pgtable.root, pgtable.mode);
	2825	+ atomic64_set(&domain->pt_root, pt_root);
	2826	+
2721	2827	/* Page-table is not visible to IOMMU anymore, so free it */
2722	2828	free_pagetable(domain);
2723	2829
@@ -2908,9 +3014,11 @@ static u64 __get_gcr3_pte(u64 root, int level, int pasid, bool alloc) · 184 unmodified lines
2908	3014	static int __set_gcr3(struct protection_domain *domain, int pasid,
2909	3015	unsigned long cr3)
2910	3016	{
	3017	+ struct domain_pgtable pgtable;
2911	3018	u64 *pte;
2912	3019
2913		− if (domain->mode != PAGE_MODE_NONE)
	3020	+ amd_iommu_domain_get_pgtable(domain, &pgtable);
	3021	+ if (pgtable.mode != PAGE_MODE_NONE)
2914	3022	return -EINVAL;
2915	3023
2916	3024	pte = __get_gcr3_pte(domain->gcr3_tbl, domain->glx, pasid, true);
@@ -2924,9 +3032,11 @@ static int __set_gcr3(struct protection_domain *domain, int pasid, · 7 unmodified lines
2924	3032
2925	3033	static int __clear_gcr3(struct protection_domain *domain, int pasid)
2926	3034	{
	3035	+ struct domain_pgtable pgtable;
2927	3036	u64 *pte;
2928	3037
2929		− if (domain->mode != PAGE_MODE_NONE)
	3038	+ amd_iommu_domain_get_pgtable(domain, &pgtable);
	3039	+ if (pgtable.mode != PAGE_MODE_NONE)
2930	3040	return -EINVAL;
2931	3041
2932	3042	pte = __get_gcr3_pte(domain->gcr3_tbl, domain->glx, pasid, false);

drivers/iommu/amd_iommu_init.c+1−1

@@ -2936,7 +2936,7 @@ static int __init parse_amd_iommu_intr(char *str) · 2935 unmodified lines
2936	2936	{
2937	2937	for (; *str; ++str) {
2938	2938	if (strncmp(str, "legacy", 6) == 0) {
2939		− amd_iommu_guest_ir = AMD_IOMMU_GUEST_IR_LEGACY;
	2939	+ amd_iommu_guest_ir = AMD_IOMMU_GUEST_IR_LEGACY_GA;
2940	2940	break;
2941	2941	}
2942	2942	if (strncmp(str, "vapic", 5) == 0) {

drivers/iommu/amd_iommu_types.h+7−2

@@ -468,8 +468,7 @@ struct protection_domain { · 467 unmodified lines
468	468	iommu core code */
469	469	spinlock_t lock; /* mostly used to lock the page table*/
470	470	u16 id; /* the domain id written to the device table */
471		− int mode; /* paging mode (0-6 levels) */
472		− u64 pt_root; / page table root pointer */
	471	+ atomic64_t pt_root; /* pgtable root and pgtable mode */
473	472	int glx; /* Number of levels for GCR3 table */
474	473	u64 gcr3_tbl; / Guest CR3 table */
475	474	unsigned long flags; /* flags to find out type of domain */
@@ -477,6 +476,12 @@ struct protection_domain { · 1 unmodified line
477	476	unsigned dev_iommu[MAX_IOMMUS]; /* per-IOMMU reference count */
478	477	};
479	478
	479	+/* For decocded pt_root */
	480	+struct domain_pgtable {
	481	+ int mode;
	482	+ u64 *root;
	483	+};
	484	+
480	485	/*
481	486	* Structure where we save information about one hardware AMD IOMMU in the
482	487	* system.

drivers/iommu/intel-iommu.c+2−2

@@ -371,11 +371,11 @@ int dmar_disabled = 0; · 370 unmodified lines
371	371	int dmar_disabled = 1;
372	372	#endif /* CONFIG_INTEL_IOMMU_DEFAULT_ON */
373	373
374		−#ifdef INTEL_IOMMU_SCALABLE_MODE_DEFAULT_ON
	374	+#ifdef CONFIG_INTEL_IOMMU_SCALABLE_MODE_DEFAULT_ON
375	375	int intel_iommu_sm = 1;
376	376	#else
377	377	int intel_iommu_sm;
378		−#endif /* INTEL_IOMMU_SCALABLE_MODE_DEFAULT_ON */
	378	+#endif /* CONFIG_INTEL_IOMMU_SCALABLE_MODE_DEFAULT_ON */
379	379
380	380	int intel_iommu_enabled = 0;
381	381	EXPORT_SYMBOL_GPL(intel_iommu_enabled);

drivers/iommu/iommu.c+2−1

@@ -170,6 +170,7 @@ static struct dev_iommu dev_iommu_get(struct device dev) · 169 unmodified lines
170	170
171	171	static void dev_iommu_free(struct device *dev)
172	172	{
	173	+ iommu_fwspec_free(dev);
173	174	kfree(dev->iommu);
174	175	dev->iommu = NULL;
175	176	}
@@ -1428,7 +1429,7 @@ struct iommu_group iommu_group_get_for_dev(struct device dev) · 1252 unmodified lines
1428	1429
1429	1430	return group;
1430	1431	}
1431		−EXPORT_SYMBOL(iommu_group_get_for_dev);
	1432	+EXPORT_SYMBOL_GPL(iommu_group_get_for_dev);
1432	1433
1433	1434	struct iommu_domain iommu_group_default_domain(struct iommu_group group)
1434	1435	{

drivers/iommu/qcom_iommu.c+4−1

@@ -824,8 +824,11 @@ static int qcom_iommu_device_probe(struct platform_device *pdev) · 823 unmodified lines
824	824	qcom_iommu->dev = dev;
825	825
826	826	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
827		− if (res)
	827	+ if (res) {
828	828	qcom_iommu->local_base = devm_ioremap_resource(dev, res);
	829	+ if (IS_ERR(qcom_iommu->local_base))
	830	+ return PTR_ERR(qcom_iommu->local_base);
	831	+ }
829	832
830	833	qcom_iommu->iface_clk = devm_clk_get(dev, "iface");
831	834	if (IS_ERR(qcom_iommu->iface_clk)) {

drivers/iommu/virtio-iommu.c+1−1

@@ -453,7 +453,7 @@ static int viommu_add_resv_mem(struct viommu_endpoint *vdev, · 452 unmodified lines
453	453	if (!region)
454	454	return -ENOMEM;
455	455
456		− list_add(&vdev->resv_regions, &region->list);
	456	+ list_add(&region->list, &vdev->resv_regions);
457	457	return 0;
458	458	}
459	459

drivers/md/dm-mpath.c+4−2

@@ -585,10 +585,12 @@ static struct pgpath __map_bio(struct multipath m, struct bio *bio) · 584 unmodified lines
585	585
586	586	/* Do we need to select a new pgpath? */
587	587	pgpath = READ_ONCE(m->current_pgpath);
588		− queue_io = test_bit(MPATHF_QUEUE_IO, &m->flags);
589		− if (!pgpath \|\| !queue_io)
	588	+ if (!pgpath \|\| !test_bit(MPATHF_QUEUE_IO, &m->flags))
590	589	pgpath = choose_pgpath(m, bio->bi_iter.bi_size);
591	590
	591	+ /* MPATHF_QUEUE_IO might have been cleared by choose_pgpath. */
	592	+ queue_io = test_bit(MPATHF_QUEUE_IO, &m->flags);
	593	+
592	594	if ((pgpath && queue_io) \|\|
593	595	(!pgpath && test_bit(MPATHF_QUEUE_IF_NO_PATH, &m->flags))) {
594	596	/* Queue for the daemon to resubmit */

drivers/md/dm-verity-fec.c+1−1

@@ -435,7 +435,7 @@ int verity_fec_decode(struct dm_verity v, struct dm_verity_io io, · 434 unmodified lines
435	435	fio->level++;
436	436
437	437	if (type == DM_VERITY_BLOCK_TYPE_METADATA)
438		− block += v->data_blocks;
	438	+ block = block - v->hash_start + v->data_blocks;
439	439
440	440	/*
441	441	* For RS(M, N), the continuous FEC data is divided into blocks of N

drivers/md/dm-writecache.c+37−15

@@ -931,6 +931,24 @@ static int writecache_alloc_entries(struct dm_writecache *wc) · 930 unmodified lines
931	931	return 0;
932	932	}
933	933
	934	+static int writecache_read_metadata(struct dm_writecache *wc, sector_t n_sectors)
	935	+{
	936	+ struct dm_io_region region;
	937	+ struct dm_io_request req;
	938	+
	939	+ region.bdev = wc->ssd_dev->bdev;
	940	+ region.sector = wc->start_sector;
	941	+ region.count = n_sectors;
	942	+ req.bi_op = REQ_OP_READ;
	943	+ req.bi_op_flags = REQ_SYNC;
	944	+ req.mem.type = DM_IO_VMA;
	945	+ req.mem.ptr.vma = (char *)wc->memory_map;
	946	+ req.client = wc->dm_io;
	947	+ req.notify.fn = NULL;
	948	+
	949	+ return dm_io(&req, 1, &region, NULL);
	950	+}
	951	+
934	952	static void writecache_resume(struct dm_target *ti)
935	953	{
936	954	struct dm_writecache *wc = ti->private;
@@ -941,8 +959,18 @@ static void writecache_resume(struct dm_target *ti) · 4 unmodified lines
941	959
942	960	wc_lock(wc);
943	961
944		− if (WC_MODE_PMEM(wc))
	962	+ if (WC_MODE_PMEM(wc)) {
945	963	persistent_memory_invalidate_cache(wc->memory_map, wc->memory_map_size);
	964	+ } else {
	965	+ r = writecache_read_metadata(wc, wc->metadata_sectors);
	966	+ if (r) {
	967	+ size_t sb_entries_offset;
	968	+ writecache_error(wc, r, "unable to read metadata: %d", r);
	969	+ sb_entries_offset = offsetof(struct wc_memory_superblock, entries);
	970	+ memset((char *)wc->memory_map + sb_entries_offset, -1,
	971	+ (wc->metadata_sectors << SECTOR_SHIFT) - sb_entries_offset);
	972	+ }
	973	+ }
946	974
947	975	wc->tree = RB_ROOT;
948	976	INIT_LIST_HEAD(&wc->lru);
@@ -2102,6 +2130,12 @@ static int writecache_ctr(struct dm_target ti, unsigned argc, char *argv) · 1153 unmodified lines
2102	2130	ti->error = "Invalid block size";
2103	2131	goto bad;
2104	2132	}
	2133	+ if (wc->block_size < bdev_logical_block_size(wc->dev->bdev) \|\|
	2134	+ wc->block_size < bdev_logical_block_size(wc->ssd_dev->bdev)) {
	2135	+ r = -EINVAL;
	2136	+ ti->error = "Block size is smaller than device logical block size";
	2137	+ goto bad;
	2138	+ }
2105	2139	wc->block_size_bits = __ffs(wc->block_size);
2106	2140
2107	2141	wc->max_writeback_jobs = MAX_WRITEBACK_JOBS;
@@ -2200,8 +2234,6 @@ invalid_optional: · 92 unmodified lines
2200	2234	goto bad;
2201	2235	}
2202	2236	} else {
2203		− struct dm_io_region region;
2204		− struct dm_io_request req;
2205	2237	size_t n_blocks, n_metadata_blocks;
2206	2238	uint64_t n_bitmap_bits;
2207	2239
@@ -2258,19 +2290,9 @@ invalid_optional: · 50 unmodified lines
2258	2290	goto bad;
2259	2291	}
2260	2292
2261		− region.bdev = wc->ssd_dev->bdev;
2262		− region.sector = wc->start_sector;
2263		− region.count = wc->metadata_sectors;
2264		− req.bi_op = REQ_OP_READ;
2265		− req.bi_op_flags = REQ_SYNC;
2266		− req.mem.type = DM_IO_VMA;
2267		− req.mem.ptr.vma = (char *)wc->memory_map;
2268		− req.client = wc->dm_io;
2269		− req.notify.fn = NULL;
2270		−
2271		− r = dm_io(&req, 1, &region, NULL);
	2293	+ r = writecache_read_metadata(wc, wc->block_size >> SECTOR_SHIFT);
2272	2294	if (r) {
2273		− ti->error = "Unable to read metadata";
	2295	+ ti->error = "Unable to read first block of metadata";
2274	2296	goto bad;
2275	2297	}
2276	2298	}

drivers/misc/mei/hw-me.c+8−0

@@ -1465,6 +1465,13 @@ static const struct mei_cfg mei_me_pch12_cfg = { · 1464 unmodified lines
1465	1465	MEI_CFG_DMA_128,
1466	1466	};
1467	1467
	1468	+/* LBG with quirk for SPS Firmware exclusion */
	1469	+static const struct mei_cfg mei_me_pch12_sps_cfg = {
	1470	+ MEI_CFG_PCH8_HFS,
	1471	+ MEI_CFG_FW_VER_SUPP,
	1472	+ MEI_CFG_FW_SPS,
	1473	+};
	1474	+
1468	1475	/* Tiger Lake and newer devices */
1469	1476	static const struct mei_cfg mei_me_pch15_cfg = {
1470	1477	MEI_CFG_PCH8_HFS,
@@ -1487,6 +1494,7 @@ static const struct mei_cfg *const mei_cfg_list[] = { · 16 unmodified lines
1487	1494	[MEI_ME_PCH8_CFG] = &mei_me_pch8_cfg,
1488	1495	[MEI_ME_PCH8_SPS_CFG] = &mei_me_pch8_sps_cfg,
1489	1496	[MEI_ME_PCH12_CFG] = &mei_me_pch12_cfg,
	1497	+ [MEI_ME_PCH12_SPS_CFG] = &mei_me_pch12_sps_cfg,
1490	1498	[MEI_ME_PCH15_CFG] = &mei_me_pch15_cfg,
1491	1499	};
1492	1500

drivers/misc/mei/hw-me.h+4−0

@@ -80,6 +80,9 @@ struct mei_me_hw { · 79 unmodified lines
80	80	* servers platforms with quirk for
81	81	* SPS firmware exclusion.
82	82	* @MEI_ME_PCH12_CFG: Platform Controller Hub Gen12 and newer
	83	+ * @MEI_ME_PCH12_SPS_CFG: Platform Controller Hub Gen12 and newer
	84	+ * servers platforms with quirk for
	85	+ * SPS firmware exclusion.
83	86	* @MEI_ME_PCH15_CFG: Platform Controller Hub Gen15 and newer
84	87	* @MEI_ME_NUM_CFG: Upper Sentinel.
85	88	*/
@@ -93,6 +96,7 @@ enum mei_cfg_idx { · 7 unmodified lines
93	96	MEI_ME_PCH8_CFG,
94	97	MEI_ME_PCH8_SPS_CFG,
95	98	MEI_ME_PCH12_CFG,
	99	+ MEI_ME_PCH12_SPS_CFG,
96	100	MEI_ME_PCH15_CFG,
97	101	MEI_ME_NUM_CFG,
98	102	};

drivers/misc/mei/pci-me.c+3−2

@@ -70,7 +70,7 @@ static const struct pci_device_id mei_me_pci_tbl[] = { · 69 unmodified lines
70	70	{MEI_PCI_DEVICE(MEI_DEV_ID_SPT_2, MEI_ME_PCH8_CFG)},
71	71	{MEI_PCI_DEVICE(MEI_DEV_ID_SPT_H, MEI_ME_PCH8_SPS_CFG)},
72	72	{MEI_PCI_DEVICE(MEI_DEV_ID_SPT_H_2, MEI_ME_PCH8_SPS_CFG)},
73		− {MEI_PCI_DEVICE(MEI_DEV_ID_LBG, MEI_ME_PCH12_CFG)},
	73	+ {MEI_PCI_DEVICE(MEI_DEV_ID_LBG, MEI_ME_PCH12_SPS_CFG)},
74	74
75	75	{MEI_PCI_DEVICE(MEI_DEV_ID_BXT_M, MEI_ME_PCH8_CFG)},
76	76	{MEI_PCI_DEVICE(MEI_DEV_ID_APL_I, MEI_ME_PCH8_CFG)},
@@ -203,11 +203,12 @@ static int mei_me_probe(struct pci_dev pdev, const struct pci_device_id ent) · 126 unmodified lines
203	203	}
204	204	hw = to_me_hw(dev);
205	205	hw->mem_addr = pcim_iomap_table(pdev)[0];
206		− hw->irq = pdev->irq;
207	206	hw->read_fws = mei_me_read_fws;
208	207
209	208	pci_enable_msi(pdev);
210	209
	210	+ hw->irq = pdev->irq;
	211	+
211	212	/* request and enable interrupt */
212	213	irqflags = pci_dev_msi_enabled(pdev) ? IRQF_ONESHOT : IRQF_SHARED;
213	214

drivers/misc/mic/Kconfig+1−1

@@ -116,7 +116,7 @@ config MIC_COSM · 115 unmodified lines
116	116
117	117	config VOP
118	118	tristate "VOP Driver"
119		− depends on VOP_BUS
	119	+ depends on VOP_BUS && VHOST_DPN
120	120	select VHOST_RING
121	121	select VIRTIO
122	122	help

drivers/mmc/core/mmc_ops.c+1−1

@@ -878,7 +878,7 @@ static int mmc_send_hpi_cmd(struct mmc_card *card) · 877 unmodified lines
878	878	* Issued High Priority Interrupt, and check for card status
879	879	* until out-of prg-state.
880	880	*/
881		−int mmc_interrupt_hpi(struct mmc_card *card)
	881	+static int mmc_interrupt_hpi(struct mmc_card *card)
882	882	{
883	883	int err;
884	884	u32 status;

drivers/mmc/host/cqhci.c+10−11

@@ -5,6 +5,7 @@ · 4 unmodified lines
5	5	#include <linux/delay.h>
6	6	#include <linux/highmem.h>
7	7	#include <linux/io.h>
	8	+#include <linux/iopoll.h>
8	9	#include <linux/module.h>
9	10	#include <linux/dma-mapping.h>
10	11	#include <linux/slab.h>
@@ -349,12 +350,16 @@ static int cqhci_enable(struct mmc_host mmc, struct mmc_card card) · 338 unmodified lines
349	350	/* CQHCI is idle and should halt immediately, so set a small timeout */
350	351	#define CQHCI_OFF_TIMEOUT 100
351	352
	353	+static u32 cqhci_read_ctl(struct cqhci_host *cq_host)
	354	+{
	355	+ return cqhci_readl(cq_host, CQHCI_CTL);
	356	+}
	357	+
352	358	static void cqhci_off(struct mmc_host *mmc)
353	359	{
354	360	struct cqhci_host *cq_host = mmc->cqe_private;
355		− ktime_t timeout;
356		− bool timed_out;
357	361	u32 reg;
	362	+ int err;
358	363
359	364	if (!cq_host->enabled \|\| !mmc->cqe_on \|\| cq_host->recovery_halt)
360	365	return;
@@ -364,15 +369,9 @@ static void cqhci_off(struct mmc_host *mmc) · 3 unmodified lines
364	369
365	370	cqhci_writel(cq_host, CQHCI_HALT, CQHCI_CTL);
366	371
367		− timeout = ktime_add_us(ktime_get(), CQHCI_OFF_TIMEOUT);
368		− while (1) {
369		− timed_out = ktime_compare(ktime_get(), timeout) > 0;
370		− reg = cqhci_readl(cq_host, CQHCI_CTL);
371		− if ((reg & CQHCI_HALT) \|\| timed_out)
372		− break;
373		− }
374		−
375		− if (timed_out)
	372	+ err = readx_poll_timeout(cqhci_read_ctl, cq_host, reg,
	373	+ reg & CQHCI_HALT, 0, CQHCI_OFF_TIMEOUT);
	374	+ if (err < 0)
376	375	pr_err("%s: cqhci: CQE stuck on\n", mmc_hostname(mmc));
377	376	else
378	377	pr_debug("%s: cqhci: CQE off\n", mmc_hostname(mmc));

drivers/mmc/host/meson-mx-sdio.c+1−10

@@ -357,14 +357,6 @@ static void meson_mx_mmc_request(struct mmc_host mmc, struct mmc_request mrq) · 356 unmodified lines
357	357	meson_mx_mmc_start_cmd(mmc, mrq->cmd);
358	358	}
359	359
360		−static int meson_mx_mmc_card_busy(struct mmc_host *mmc)
361		−{
362		− struct meson_mx_mmc_host *host = mmc_priv(mmc);
363		− u32 irqc = readl(host->base + MESON_MX_SDIO_IRQC);
364		−
365		− return !!(irqc & MESON_MX_SDIO_IRQC_FORCE_DATA_DAT_MASK);
366		−}
367		−
368	360	static void meson_mx_mmc_read_response(struct mmc_host *mmc,
369	361	struct mmc_command *cmd)
370	362	{
@@ -506,7 +498,6 @@ static void meson_mx_mmc_timeout(struct timer_list *t) · 135 unmodified lines
506	498	static struct mmc_host_ops meson_mx_mmc_ops = {
507	499	.request = meson_mx_mmc_request,
508	500	.set_ios = meson_mx_mmc_set_ios,
509		− .card_busy = meson_mx_mmc_card_busy,
510	501	.get_cd = mmc_gpio_get_cd,
511	502	.get_ro = mmc_gpio_get_ro,
512	503	};
@@ -570,7 +561,7 @@ static int meson_mx_mmc_add_host(struct meson_mx_mmc_host *host) · 57 unmodified lines
570	561	mmc->f_max = clk_round_rate(host->cfg_div_clk,
571	562	clk_get_rate(host->parent_clk));
572	563
573		− mmc->caps \|= MMC_CAP_ERASE \| MMC_CAP_CMD23;
	564	+ mmc->caps \|= MMC_CAP_ERASE \| MMC_CAP_CMD23 \| MMC_CAP_WAIT_WHILE_BUSY;
574	565	mmc->ops = &meson_mx_mmc_ops;
575	566
576	567	ret = mmc_of_parse(mmc);

drivers/mmc/host/sdhci-msm.c+2−0

@@ -2087,6 +2087,8 @@ static int sdhci_msm_probe(struct platform_device *pdev) · 2086 unmodified lines
2087	2087	goto clk_disable;
2088	2088	}
2089	2089
	2090	+ msm_host->mmc->caps \|= MMC_CAP_WAIT_WHILE_BUSY \| MMC_CAP_NEED_RSP_BUSY;
	2091	+
2090	2092	pm_runtime_get_noresume(&pdev->dev);
2091	2093	pm_runtime_set_active(&pdev->dev);
2092	2094	pm_runtime_enable(&pdev->dev);

drivers/mmc/host/sdhci-pci-core.c+3−0

@@ -601,6 +601,9 @@ static int intel_select_drive_strength(struct mmc_card *card, · 600 unmodified lines
601	601	struct sdhci_pci_slot *slot = sdhci_priv(host);
602	602	struct intel_host *intel_host = sdhci_pci_priv(slot);
603	603
	604	+ if (!(mmc_driver_type_mask(intel_host->drv_strength) & card_drv))
	605	+ return 0;
	606	+
604	607	return intel_host->drv_strength;
605	608	}
606	609

drivers/mmc/host/sdhci-xenon.c+10−0

@@ -235,6 +235,16 @@ static void xenon_voltage_switch(struct sdhci_host *host) · 234 unmodified lines
235	235	{
236	236	/* Wait for 5ms after set 1.8V signal enable bit */
237	237	usleep_range(5000, 5500);
	238	+
	239	+ /*
	240	+ * For some reason the controller's Host Control2 register reports
	241	+ * the bit representing 1.8V signaling as 0 when read after it was
	242	+ * written as 1. Subsequent read reports 1.
	243	+ *
	244	+ * Since this may cause some issues, do an empty read of the Host
	245	+ * Control2 register here to circumvent this.
	246	+ */
	247	+ sdhci_readw(host, SDHCI_HOST_CONTROL2);
238	248	}
239	249
240	250	static const struct sdhci_ops sdhci_xenon_ops = {

drivers/most/core.c+1−1

@@ -1483,7 +1483,7 @@ static void __exit most_exit(void) · 1482 unmodified lines
1483	1483	ida_destroy(&mdev_id);
1484	1484	}
1485	1485
1486		−module_init(most_init);
	1486	+subsys_initcall(most_init);
1487	1487	module_exit(most_exit);
1488	1488	MODULE_LICENSE("GPL");
1489	1489	MODULE_AUTHOR("Christian Gromm <christian.gromm@microchip.com>");

drivers/net/caif/Kconfig+1−1

@@ -50,7 +50,7 @@ config CAIF_HSI · 49 unmodified lines
50	50
51	51	config CAIF_VIRTIO
52	52	tristate "CAIF virtio transport driver"
53		− depends on CAIF && HAS_DMA
	53	+ depends on CAIF && HAS_DMA && VHOST_DPN
54	54	select VHOST_RING
55	55	select VIRTIO
56	56	select GENERIC_ALLOCATOR

drivers/net/dsa/b53/b53_common.c+33−5

@@ -1474,6 +1474,10 @@ static int b53_arl_rw_op(struct b53_device *dev, unsigned int op) · 1473 unmodified lines
1474	1474	reg \|= ARLTBL_RW;
1475	1475	else
1476	1476	reg &= ~ARLTBL_RW;
	1477	+ if (dev->vlan_enabled)
	1478	+ reg &= ~ARLTBL_IVL_SVL_SELECT;
	1479	+ else
	1480	+ reg \|= ARLTBL_IVL_SVL_SELECT;
1477	1481	b53_write8(dev, B53_ARLIO_PAGE, B53_ARLTBL_RW_CTRL, reg);
1478	1482
1479	1483	return b53_arl_op_wait(dev);
@@ -1483,6 +1487,7 @@ static int b53_arl_read(struct b53_device *dev, u64 mac, · 3 unmodified lines
1483	1487	u16 vid, struct b53_arl_entry ent, u8 idx,
1484	1488	bool is_valid)
1485	1489	{
	1490	+ DECLARE_BITMAP(free_bins, B53_ARLTBL_MAX_BIN_ENTRIES);
1486	1491	unsigned int i;
1487	1492	int ret;
1488	1493
@@ -1490,6 +1495,8 @@ static int b53_arl_read(struct b53_device *dev, u64 mac, · 1 unmodified line
1490	1495	if (ret)
1491	1496	return ret;
1492	1497
	1498	+ bitmap_zero(free_bins, dev->num_arl_entries);
	1499	+
1493	1500	/* Read the bins */
1494	1501	for (i = 0; i < dev->num_arl_entries; i++) {
1495	1502	u64 mac_vid;
@@ -1501,13 +1508,24 @@ static int b53_arl_read(struct b53_device *dev, u64 mac, · 5 unmodified lines
1501	1508	B53_ARLTBL_DATA_ENTRY(i), &fwd_entry);
1502	1509	b53_arl_to_entry(ent, mac_vid, fwd_entry);
1503	1510
1504		− if (!(fwd_entry & ARLTBL_VALID))
	1511	+ if (!(fwd_entry & ARLTBL_VALID)) {
	1512	+ set_bit(i, free_bins);
1505	1513	continue;
	1514	+ }
1506	1515	if ((mac_vid & ARLTBL_MAC_MASK) != mac)
1507	1516	continue;
	1517	+ if (dev->vlan_enabled &&
	1518	+ ((mac_vid >> ARLTBL_VID_S) & ARLTBL_VID_MASK) != vid)
	1519	+ continue;
1508	1520	*idx = i;
	1521	+ return 0;
1509	1522	}
1510	1523
	1524	+ if (bitmap_weight(free_bins, dev->num_arl_entries) == 0)
	1525	+ return -ENOSPC;
	1526	+
	1527	+ *idx = find_first_bit(free_bins, dev->num_arl_entries);
	1528	+
1511	1529	return -ENOENT;
1512	1530	}
1513	1531
@@ -1537,10 +1555,21 @@ static int b53_arl_op(struct b53_device *dev, int op, int port, · 23 unmodified lines
1537	1555	if (op)
1538	1556	return ret;
1539	1557
1540		− /* We could not find a matching MAC, so reset to a new entry */
1541		− if (ret) {
	1558	+ switch (ret) {
	1559	+ case -ENOSPC:
	1560	+ dev_dbg(dev->dev, "{%pM,%.4d} no space left in ARL\n",
	1561	+ addr, vid);
	1562	+ return is_valid ? ret : 0;
	1563	+ case -ENOENT:
	1564	+ /* We could not find a matching MAC, so reset to a new entry */
	1565	+ dev_dbg(dev->dev, "{%pM,%.4d} not found, using idx: %d\n",
	1566	+ addr, vid, idx);
1542	1567	fwd_entry = 0;
1543		− idx = 1;
	1568	+ break;
	1569	+ default:
	1570	+ dev_dbg(dev->dev, "{%pM,%.4d} found, using idx: %d\n",
	1571	+ addr, vid, idx);
	1572	+ break;
1544	1573	}
1545	1574
1546	1575	/* For multicast address, the port is a bitmask and the validity
@@ -1558,7 +1587,6 @@ static int b53_arl_op(struct b53_device *dev, int op, int port, · 11 unmodified lines
1558	1587	ent.is_valid = !!(ent.port);
1559	1588	}
1560	1589
1561		− ent.is_valid = is_valid;
1562	1590	ent.vid = vid;
1563	1591	ent.is_static = true;
1564	1592	ent.is_age = false;

drivers/net/dsa/b53/b53_regs.h+6−2

@@ -292,6 +292,7 @@ · 291 unmodified lines
292	292	/* ARL Table Read/Write Register (8 bit) */
293	293	#define B53_ARLTBL_RW_CTRL 0x00
294	294	#define ARLTBL_RW BIT(0)
	295	+#define ARLTBL_IVL_SVL_SELECT BIT(6)
295	296	#define ARLTBL_START_DONE BIT(7)
296	297
297	298	/* MAC Address Index Register (48 bit) */
@@ -304,7 +305,7 @@ · 6 unmodified lines
304	305	*
305	306	* BCM5325 and BCM5365 share most definitions below
306	307	*/
307		−#define B53_ARLTBL_MAC_VID_ENTRY(n) (0x10 * (n))
	308	+#define B53_ARLTBL_MAC_VID_ENTRY(n) ((0x10 * (n)) + 0x10)
308	309	#define ARLTBL_MAC_MASK 0xffffffffffffULL
309	310	#define ARLTBL_VID_S 48
310	311	#define ARLTBL_VID_MASK_25 0xff
@@ -316,13 +317,16 @@ · 5 unmodified lines
316	317	#define ARLTBL_VALID_25 BIT(63)
317	318
318	319	/* ARL Table Data Entry N Registers (32 bit) */
319		−#define B53_ARLTBL_DATA_ENTRY(n) ((0x10 * (n)) + 0x08)
	320	+#define B53_ARLTBL_DATA_ENTRY(n) ((0x10 * (n)) + 0x18)
320	321	#define ARLTBL_DATA_PORT_ID_MASK 0x1ff
321	322	#define ARLTBL_TC(tc) ((3 & tc) << 11)
322	323	#define ARLTBL_AGE BIT(14)
323	324	#define ARLTBL_STATIC BIT(15)
324	325	#define ARLTBL_VALID BIT(16)
325	326
	327	+/* Maximum number of bin entries in the ARL for all switches */
	328	+#define B53_ARLTBL_MAX_BIN_ENTRIES 4
	329	+
326	330	/* ARL Search Control Register (8 bit) */
327	331	#define B53_ARL_SRCH_CTL 0x50
328	332	#define B53_ARL_SRCH_CTL_25 0x20

drivers/net/dsa/mv88e6xxx/Kconfig+1−1

@@ -24,8 +24,8 @@ config NET_DSA_MV88E6XXX_PTP · 23 unmodified lines
24	24	bool "PTP support for Marvell 88E6xxx"
25	25	default n
26	26	depends on NET_DSA_MV88E6XXX_GLOBAL2
	27	+ depends on PTP_1588_CLOCK
27	28	imply NETWORK_PHY_TIMESTAMPING
28		− imply PTP_1588_CLOCK
29	29	help
30	30	Say Y to enable PTP hardware timestamping on Marvell 88E6xxx switch
31	31	chips that support it.

drivers/net/dsa/mv88e6xxx/chip.c+0−4

@@ -3962,7 +3962,6 @@ static const struct mv88e6xxx_ops mv88e6190_ops = { · 3961 unmodified lines
3962	3962	.serdes_get_stats = mv88e6390_serdes_get_stats,
3963	3963	.serdes_get_regs_len = mv88e6390_serdes_get_regs_len,
3964	3964	.serdes_get_regs = mv88e6390_serdes_get_regs,
3965		− .phylink_validate = mv88e6390_phylink_validate,
3966	3965	.gpio_ops = &mv88e6352_gpio_ops,
3967	3966	.phylink_validate = mv88e6390_phylink_validate,
3968	3967	};
@@ -4021,7 +4020,6 @@ static const struct mv88e6xxx_ops mv88e6190x_ops = { · 52 unmodified lines
4021	4020	.serdes_get_stats = mv88e6390_serdes_get_stats,
4022	4021	.serdes_get_regs_len = mv88e6390_serdes_get_regs_len,
4023	4022	.serdes_get_regs = mv88e6390_serdes_get_regs,
4024		− .phylink_validate = mv88e6390_phylink_validate,
4025	4023	.gpio_ops = &mv88e6352_gpio_ops,
4026	4024	.phylink_validate = mv88e6390x_phylink_validate,
4027	4025	};
@@ -4079,7 +4077,6 @@ static const struct mv88e6xxx_ops mv88e6191_ops = { · 51 unmodified lines
4079	4077	.serdes_get_stats = mv88e6390_serdes_get_stats,
4080	4078	.serdes_get_regs_len = mv88e6390_serdes_get_regs_len,
4081	4079	.serdes_get_regs = mv88e6390_serdes_get_regs,
4082		− .phylink_validate = mv88e6390_phylink_validate,
4083	4080	.avb_ops = &mv88e6390_avb_ops,
4084	4081	.ptp_ops = &mv88e6352_ptp_ops,
4085	4082	.phylink_validate = mv88e6390_phylink_validate,
@@ -4235,7 +4232,6 @@ static const struct mv88e6xxx_ops mv88e6290_ops = { · 149 unmodified lines
4235	4232	.serdes_get_stats = mv88e6390_serdes_get_stats,
4236	4233	.serdes_get_regs_len = mv88e6390_serdes_get_regs_len,
4237	4234	.serdes_get_regs = mv88e6390_serdes_get_regs,
4238		− .phylink_validate = mv88e6390_phylink_validate,
4239	4235	.gpio_ops = &mv88e6352_gpio_ops,
4240	4236	.avb_ops = &mv88e6390_avb_ops,
4241	4237	.ptp_ops = &mv88e6352_ptp_ops,

drivers/net/dsa/ocelot/felix.c+1−0

@@ -400,6 +400,7 @@ static int felix_init_structs(struct felix *felix, int num_phys_ports) · 399 unmodified lines
400	400	ocelot->stats_layout = felix->info->stats_layout;
401	401	ocelot->num_stats = felix->info->num_stats;
402	402	ocelot->shared_queue_sz = felix->info->shared_queue_sz;
	403	+ ocelot->num_mact_rows = felix->info->num_mact_rows;
403	404	ocelot->vcap_is2_keys = felix->info->vcap_is2_keys;
404	405	ocelot->vcap_is2_actions= felix->info->vcap_is2_actions;
405	406	ocelot->vcap = felix->info->vcap;

drivers/net/dsa/ocelot/felix.h+1−0

@@ -15,6 +15,7 @@ struct felix_info { · 14 unmodified lines
15	15	const u32 const map;
16	16	const struct ocelot_ops *ops;
17	17	int shared_queue_sz;
	18	+ int num_mact_rows;
18	19	const struct ocelot_stat_layout *stats_layout;
19	20	unsigned int num_stats;
20	21	int num_ports;

drivers/net/dsa/ocelot/felix_vsc9959.c+1−0

@@ -1220,6 +1220,7 @@ struct felix_info felix_info_vsc9959 = { · 1219 unmodified lines
1220	1220	.vcap_is2_actions = vsc9959_vcap_is2_actions,
1221	1221	.vcap = vsc9959_vcap_props,
1222	1222	.shared_queue_sz = 128 * 1024,
	1223	+ .num_mact_rows = 2048,
1223	1224	.num_ports = 6,
1224	1225	.switch_pci_bar = 4,
1225	1226	.imdio_pci_bar = 0,

drivers/net/dsa/sja1105/Kconfig+1−0

@@ -20,6 +20,7 @@ tristate "NXP SJA1105 Ethernet switch family support" · 19 unmodified lines
20	20	config NET_DSA_SJA1105_PTP
21	21	bool "Support for the PTP clock on the NXP SJA1105 Ethernet switch"
22	22	depends on NET_DSA_SJA1105
	23	+ depends on PTP_1588_CLOCK
23	24	help
24	25	This enables support for timestamping and PTP clock manipulations in
25	26	the SJA1105 DSA driver.

drivers/net/dsa/sja1105/sja1105_ptp.c+18−8

@@ -16,14 +16,15 @@ · 15 unmodified lines
16	16
17	17	/* PTPSYNCTS has no interrupt or update mechanism, because the intended
18	18	* hardware use case is for the timestamp to be collected synchronously,
19		− * immediately after the CAS_MASTER SJA1105 switch has triggered a CASSYNC
20		− * pulse on the PTP_CLK pin. When used as a generic extts source, it needs
21		− * polling and a comparison with the old value. The polling interval is just
22		− * the Nyquist rate of a canonical PPS input (e.g. from a GPS module).
23		− * Anything of higher frequency than 1 Hz will be lost, since there is no
24		− * timestamp FIFO.
	19	+ * immediately after the CAS_MASTER SJA1105 switch has performed a CASSYNC
	20	+ * one-shot toggle (no return to level) on the PTP_CLK pin. When used as a
	21	+ * generic extts source, the PTPSYNCTS register needs polling and a comparison
	22	+ * with the old value. The polling interval is configured as the Nyquist rate
	23	+ * of a signal with 50% duty cycle and 1Hz frequency, which is sadly all that
	24	+ * this hardware can do (but may be enough for some setups). Anything of higher
	25	+ * frequency than 1 Hz will be lost, since there is no timestamp FIFO.
25	26	*/
26		−#define SJA1105_EXTTS_INTERVAL (HZ / 2)
	27	+#define SJA1105_EXTTS_INTERVAL (HZ / 4)
27	28
28	29	/* This range is actually +/- SJA1105_MAX_ADJ_PPB
29	30	* divided by 1000 (ppb -> ppm) and with a 16-bit
@@ -754,7 +755,16 @@ static int sja1105_extts_enable(struct sja1105_private *priv, · 724 unmodified lines
754	755	return -EOPNOTSUPP;
755	756
756	757	/* Reject requests with unsupported flags */
757		− if (extts->flags)
	758	+ if (extts->flags & ~(PTP_ENABLE_FEATURE \|
	759	+ PTP_RISING_EDGE \|
	760	+ PTP_FALLING_EDGE \|
	761	+ PTP_STRICT_FLAGS))
	762	+ return -EOPNOTSUPP;
	763	+
	764	+ /* We can only enable time stamping on both edges, sadly. */
	765	+ if ((extts->flags & PTP_STRICT_FLAGS) &&
	766	+ (extts->flags & PTP_ENABLE_FEATURE) &&
	767	+ (extts->flags & PTP_EXTTS_EDGES) != PTP_EXTTS_EDGES)
758	768	return -EOPNOTSUPP;
759	769
760	770	rc = sja1105_change_ptp_clk_pin_func(priv, PTP_PF_EXTTS);

drivers/net/ethernet/amazon/ena/ena_netdev.h+1−1

@@ -69,7 +69,7 @@ · 68 unmodified lines
69	69	* 16kB.
70	70	*/
71	71	#if PAGE_SIZE > SZ_16K
72		−#define ENA_PAGE_SIZE SZ_16K
	72	+#define ENA_PAGE_SIZE (_AC(SZ_16K, UL))
73	73	#else
74	74	#define ENA_PAGE_SIZE PAGE_SIZE
75	75	#endif

drivers/net/ethernet/aquantia/atlantic/aq_pci_func.c+1−1

@@ -57,7 +57,7 @@ static const struct aq_board_revision_s hw_atl_boards[] = { · 56 unmodified lines
57	57	{ AQ_DEVICE_ID_D108, AQ_HWREV_2, &hw_atl_ops_b0, &hw_atl_b0_caps_aqc108, },
58	58	{ AQ_DEVICE_ID_D109, AQ_HWREV_2, &hw_atl_ops_b0, &hw_atl_b0_caps_aqc109, },
59	59
60		− { AQ_DEVICE_ID_AQC100, AQ_HWREV_ANY, &hw_atl_ops_b1, &hw_atl_b0_caps_aqc107, },
	60	+ { AQ_DEVICE_ID_AQC100, AQ_HWREV_ANY, &hw_atl_ops_b1, &hw_atl_b0_caps_aqc100, },
61	61	{ AQ_DEVICE_ID_AQC107, AQ_HWREV_ANY, &hw_atl_ops_b1, &hw_atl_b0_caps_aqc107, },
62	62	{ AQ_DEVICE_ID_AQC108, AQ_HWREV_ANY, &hw_atl_ops_b1, &hw_atl_b0_caps_aqc108, },
63	63	{ AQ_DEVICE_ID_AQC109, AQ_HWREV_ANY, &hw_atl_ops_b1, &hw_atl_b0_caps_aqc109, },

drivers/net/ethernet/broadcom/bcmsysport.c+2−2

@@ -664,7 +664,8 @@ static struct sk_buff bcm_sysport_rx_refill(struct bcm_sysport_priv priv, · 663 unmodified lines
664	664	dma_addr_t mapping;
665	665
666	666	/* Allocate a new SKB for a new packet */
667		− skb = netdev_alloc_skb(priv->netdev, RX_BUF_LENGTH);
	667	+ skb = __netdev_alloc_skb(priv->netdev, RX_BUF_LENGTH,
	668	+ GFP_ATOMIC \| __GFP_NOWARN);
668	669	if (!skb) {
669	670	priv->mib.alloc_rx_buff_failed++;
670	671	netif_err(priv, rx_err, ndev, "SKB alloc failed\n");
@@ -2475,7 +2476,6 @@ static int bcm_sysport_probe(struct platform_device *pdev) · 1804 unmodified lines
2475	2476	priv->wol_irq = platform_get_irq(pdev, 1);
2476	2477	}
2477	2478	if (priv->irq0 <= 0 \|\| (priv->irq1 <= 0 && !priv->is_lite)) {
2478		− dev_err(&pdev->dev, "invalid interrupts\n");
2479	2479	ret = -EINVAL;
2480	2480	goto err_free_netdev;
2481	2481	}

drivers/net/ethernet/broadcom/bgmac-platform.c+2−7

@@ -202,13 +202,8 @@ static int bgmac_probe(struct platform_device *pdev) · 201 unmodified lines
202	202	if (bgmac->irq < 0)
203	203	return bgmac->irq;
204	204
205		− regs = platform_get_resource_byname(pdev, IORESOURCE_MEM, "amac_base");
206		− if (!regs) {
207		− dev_err(&pdev->dev, "Unable to obtain base resource\n");
208		− return -EINVAL;
209		− }
210		−
211		− bgmac->plat.base = devm_ioremap_resource(&pdev->dev, regs);
	205	+ bgmac->plat.base =
	206	+ devm_platform_ioremap_resource_byname(pdev, "amac_base");
212	207	if (IS_ERR(bgmac->plat.base))
213	208	return PTR_ERR(bgmac->plat.base);
214	209

drivers/net/ethernet/broadcom/bnxt/bnxt.c+13−7

@@ -6642,7 +6642,7 @@ static int bnxt_alloc_ctx_pg_tbls(struct bnxt *bp, · 6641 unmodified lines
6642	6642	int rc;
6643	6643
6644	6644	if (!mem_size)
6645		− return 0;
	6645	+ return -EINVAL;
6646	6646
6647	6647	ctx_pg->nr_pages = DIV_ROUND_UP(mem_size, BNXT_PAGE_SIZE);
6648	6648	if (ctx_pg->nr_pages > MAX_CTX_TOTAL_PAGES) {
@@ -9780,6 +9780,7 @@ static netdev_features_t bnxt_fix_features(struct net_device *dev, · 3131 unmodified lines
9780	9780	netdev_features_t features)
9781	9781	{
9782	9782	struct bnxt *bp = netdev_priv(dev);
	9783	+ netdev_features_t vlan_features;
9783	9784
9784	9785	if ((features & NETIF_F_NTUPLE) && !bnxt_rfs_capable(bp))
9785	9786	features &= ~NETIF_F_NTUPLE;
@@ -9796,12 +9797,14 @@ static netdev_features_t bnxt_fix_features(struct net_device *dev, · 10 unmodified lines
9796	9797	/* Both CTAG and STAG VLAN accelaration on the RX side have to be
9797	9798	* turned on or off together.
9798	9799	*/
9799		− if ((features & (NETIF_F_HW_VLAN_CTAG_RX \| NETIF_F_HW_VLAN_STAG_RX)) !=
9800		− (NETIF_F_HW_VLAN_CTAG_RX \| NETIF_F_HW_VLAN_STAG_RX)) {
	9800	+ vlan_features = features & (NETIF_F_HW_VLAN_CTAG_RX \|
	9801	+ NETIF_F_HW_VLAN_STAG_RX);
	9802	+ if (vlan_features != (NETIF_F_HW_VLAN_CTAG_RX \|
	9803	+ NETIF_F_HW_VLAN_STAG_RX)) {
9801	9804	if (dev->features & NETIF_F_HW_VLAN_CTAG_RX)
9802	9805	features &= ~(NETIF_F_HW_VLAN_CTAG_RX \|
9803	9806	NETIF_F_HW_VLAN_STAG_RX);
9804		− else
	9807	+ else if (vlan_features)
9805	9808	features \|= NETIF_F_HW_VLAN_CTAG_RX \|
9806	9809	NETIF_F_HW_VLAN_STAG_RX;
9807	9810	}
@@ -12212,12 +12215,15 @@ static pci_ers_result_t bnxt_io_slot_reset(struct pci_dev *pdev) · 2404 unmodified lines
12212	12215	bnxt_ulp_start(bp, err);
12213	12216	}
12214	12217
12215		− if (result != PCI_ERS_RESULT_RECOVERED && netif_running(netdev))
12216		− dev_close(netdev);
	12218	+ if (result != PCI_ERS_RESULT_RECOVERED) {
	12219	+ if (netif_running(netdev))
	12220	+ dev_close(netdev);
	12221	+ pci_disable_device(pdev);
	12222	+ }
12217	12223
12218	12224	rtnl_unlock();
12219	12225
12220		− return PCI_ERS_RESULT_RECOVERED;
	12226	+ return result;
12221	12227	}
12222	12228
12223	12229	/**

drivers/net/ethernet/broadcom/bnxt/bnxt.h+0−1

@@ -1066,7 +1066,6 @@ struct bnxt_vf_info { · 1065 unmodified lines
1066	1066	#define BNXT_VF_LINK_FORCED 0x4
1067	1067	#define BNXT_VF_LINK_UP 0x8
1068	1068	#define BNXT_VF_TRUST 0x10
1069		− u32 func_flags; /* func cfg flags */
1070	1069	u32 min_tx_rate;
1071	1070	u32 max_tx_rate;
1072	1071	void *hwrm_cmd_req_addr;

drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.h+1−1

@@ -43,7 +43,7 @@ static inline void bnxt_link_bp_to_dl(struct bnxt bp, struct devlink dl) · 42 unmodified lines
43	43	#define BNXT_NVM_CFG_VER_BITS 24
44	44	#define BNXT_NVM_CFG_VER_BYTES 4
45	45
46		−#define BNXT_MSIX_VEC_MAX 1280
	46	+#define BNXT_MSIX_VEC_MAX 512
47	47	#define BNXT_MSIX_VEC_MIN_MAX 128
48	48
49	49	enum bnxt_nvm_dir_type {

drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c+2−8

@@ -85,11 +85,10 @@ int bnxt_set_vf_spoofchk(struct net_device *dev, int vf_id, bool setting) · 84 unmodified lines
85	85	if (old_setting == setting)
86	86	return 0;
87	87
88		− func_flags = vf->func_flags;
89	88	if (setting)
90		− func_flags \|= FUNC_CFG_REQ_FLAGS_SRC_MAC_ADDR_CHECK_ENABLE;
	89	+ func_flags = FUNC_CFG_REQ_FLAGS_SRC_MAC_ADDR_CHECK_ENABLE;
91	90	else
92		− func_flags \|= FUNC_CFG_REQ_FLAGS_SRC_MAC_ADDR_CHECK_DISABLE;
	91	+ func_flags = FUNC_CFG_REQ_FLAGS_SRC_MAC_ADDR_CHECK_DISABLE;
93	92	/*TODO: if the driver supports VLAN filter on guest VLAN,
94	93	* the spoof check should also include vlan anti-spoofing
95	94	*/
@@ -98,7 +97,6 @@ int bnxt_set_vf_spoofchk(struct net_device *dev, int vf_id, bool setting) · 2 unmodified lines
98	97	req.flags = cpu_to_le32(func_flags);
99	98	rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
100	99	if (!rc) {
101		− vf->func_flags = func_flags;
102	100	if (setting)
103	101	vf->flags \|= BNXT_VF_SPOOFCHK;
104	102	else
@@ -228,7 +226,6 @@ int bnxt_set_vf_mac(struct net_device dev, int vf_id, u8 mac) · 123 unmodified lines
228	226	memcpy(vf->mac_addr, mac, ETH_ALEN);
229	227	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FUNC_CFG, -1, -1);
230	228	req.fid = cpu_to_le16(vf->fw_fid);
231		− req.flags = cpu_to_le32(vf->func_flags);
232	229	req.enables = cpu_to_le32(FUNC_CFG_REQ_ENABLES_DFLT_MAC_ADDR);
233	230	memcpy(req.dflt_mac_addr, mac, ETH_ALEN);
234	231	return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
@@ -266,7 +263,6 @@ int bnxt_set_vf_vlan(struct net_device *dev, int vf_id, u16 vlan_id, u8 qos, · 31 unmodified lines
266	263
267	264	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FUNC_CFG, -1, -1);
268	265	req.fid = cpu_to_le16(vf->fw_fid);
269		− req.flags = cpu_to_le32(vf->func_flags);
270	266	req.dflt_vlan = cpu_to_le16(vlan_tag);
271	267	req.enables = cpu_to_le32(FUNC_CFG_REQ_ENABLES_DFLT_VLAN);
272	268	rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
@@ -305,7 +301,6 @@ int bnxt_set_vf_bw(struct net_device *dev, int vf_id, int min_tx_rate, · 32 unmodified lines
305	301	return 0;
306	302	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FUNC_CFG, -1, -1);
307	303	req.fid = cpu_to_le16(vf->fw_fid);
308		− req.flags = cpu_to_le32(vf->func_flags);
309	304	req.enables = cpu_to_le32(FUNC_CFG_REQ_ENABLES_MAX_BW);
310	305	req.max_bw = cpu_to_le32(max_tx_rate);
311	306	req.enables \|= cpu_to_le32(FUNC_CFG_REQ_ENABLES_MIN_BW);
@@ -477,7 +472,6 @@ static void __bnxt_set_vf_params(struct bnxt *bp, int vf_id) · 165 unmodified lines
477	472	vf = &bp->pf.vf[vf_id];
478	473	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FUNC_CFG, -1, -1);
479	474	req.fid = cpu_to_le16(vf->fw_fid);
480		− req.flags = cpu_to_le32(vf->func_flags);
481	475
482	476	if (is_valid_ether_addr(vf->mac_addr)) {
483	477	req.enables \|= cpu_to_le32(FUNC_CFG_REQ_ENABLES_DFLT_MAC_ADDR);

drivers/net/ethernet/broadcom/genet/bcmgenet.c+5−1

@@ -934,6 +934,8 @@ static void bcmgenet_get_ethtool_stats(struct net_device *dev, · 933 unmodified lines
934	934	if (netif_running(dev))
935	935	bcmgenet_update_mib_counters(priv);
936	936
	937	+ dev->netdev_ops->ndo_get_stats(dev);
	938	+
937	939	for (i = 0; i < BCMGENET_STATS_LEN; i++) {
938	940	const struct bcmgenet_stats *s;
939	941	char *p;
@@ -1622,7 +1624,8 @@ static struct sk_buff bcmgenet_rx_refill(struct bcmgenet_priv priv, · 682 unmodified lines
1622	1624	dma_addr_t mapping;
1623	1625
1624	1626	/* Allocate a new Rx skb */
1625		− skb = netdev_alloc_skb(priv->dev, priv->rx_buf_len + SKB_ALIGNMENT);
	1627	+ skb = __netdev_alloc_skb(priv->dev, priv->rx_buf_len + SKB_ALIGNMENT,
	1628	+ GFP_ATOMIC \| __GFP_NOWARN);
1626	1629	if (!skb) {
1627	1630	priv->mib.alloc_rx_buff_failed++;
1628	1631	netif_err(priv, rx_err, priv->dev,
@@ -3156,6 +3159,7 @@ static struct net_device_stats bcmgenet_get_stats(struct net_device dev) · 1527 unmodified lines
3156	3159	dev->stats.rx_packets = rx_packets;
3157	3160	dev->stats.rx_errors = rx_errors;
3158	3161	dev->stats.rx_missed_errors = rx_errors;
	3162	+ dev->stats.rx_dropped = rx_dropped;
3159	3163	return &dev->stats;
3160	3164	}
3161	3165

drivers/net/ethernet/cadence/Kconfig+1−1

@@ -35,8 +35,8 @@ config MACB · 34 unmodified lines
35	35	config MACB_USE_HWSTAMP
36	36	bool "Use IEEE 1588 hwstamp"
37	37	depends on MACB
	38	+ depends on PTP_1588_CLOCK
38	39	default y
39		− imply PTP_1588_CLOCK
40	40	---help---
41	41	Enable IEEE 1588 Precision Time Protocol (PTP) support for MACB.
42	42

drivers/net/ethernet/cadence/macb_main.c+12−12

@@ -334,8 +334,10 @@ static int macb_mdio_read(struct mii_bus *bus, int mii_id, int regnum) · 333 unmodified lines
334	334	int status;
335	335
336	336	status = pm_runtime_get_sync(&bp->pdev->dev);
337		− if (status < 0)
	337	+ if (status < 0) {
	338	+ pm_runtime_put_noidle(&bp->pdev->dev);
338	339	goto mdio_pm_exit;
	340	+ }
339	341
340	342	status = macb_mdio_wait_for_idle(bp);
341	343	if (status < 0)
@@ -386,8 +388,10 @@ static int macb_mdio_write(struct mii_bus *bus, int mii_id, int regnum, · 44 unmodified lines
386	388	int status;
387	389
388	390	status = pm_runtime_get_sync(&bp->pdev->dev);
389		− if (status < 0)
	391	+ if (status < 0) {
	392	+ pm_runtime_put_noidle(&bp->pdev->dev);
390	393	goto mdio_pm_exit;
	394	+ }
391	395
392	396	status = macb_mdio_wait_for_idle(bp);
393	397	if (status < 0)
@@ -3816,8 +3820,10 @@ static int at91ether_open(struct net_device *dev) · 3422 unmodified lines
3816	3820	int ret;
3817	3821
3818	3822	ret = pm_runtime_get_sync(&lp->pdev->dev);
3819		− if (ret < 0)
	3823	+ if (ret < 0) {
	3824	+ pm_runtime_put_noidle(&lp->pdev->dev);
3820	3825	return ret;
	3826	+ }
3821	3827
3822	3828	/* Clear internal statistics */
3823	3829	ctl = macb_readl(lp, NCR);
@@ -4172,15 +4178,9 @@ static int fu540_c000_clk_init(struct platform_device pdev, struct clk *pclk, · 348 unmodified lines
4172	4178
4173	4179	static int fu540_c000_init(struct platform_device *pdev)
4174	4180	{
4175		− struct resource *res;
4176		−
4177		− res = platform_get_resource(pdev, IORESOURCE_MEM, 1);
4178		− if (!res)
4179		− return -ENODEV;
4180		−
4181		− mgmt->reg = ioremap(res->start, resource_size(res));
4182		− if (!mgmt->reg)
4183		− return -ENOMEM;
	4181	+ mgmt->reg = devm_platform_ioremap_resource(pdev, 1);
	4182	+ if (IS_ERR(mgmt->reg))
	4183	+ return PTR_ERR(mgmt->reg);
4184	4184
4185	4185	return macb_init(pdev);
4186	4186	}

drivers/net/ethernet/cavium/Kconfig+1−1

@@ -54,7 +54,7 @@ config THUNDER_NIC_RGX · 53 unmodified lines
54	54	config CAVIUM_PTP
55	55	tristate "Cavium PTP coprocessor as PTP clock"
56	56	depends on 64BIT && PCI
57		− imply PTP_1588_CLOCK
	57	+ depends on PTP_1588_CLOCK
58	58	---help---
59	59	This driver adds support for the Precision Time Protocol Clocks and
60	60	Timestamping coprocessor (PTP) found on Cavium processors.

drivers/net/ethernet/chelsio/cxgb4/cudbg_lib.c+20−7

@@ -1049,9 +1049,9 @@ static void cudbg_t4_fwcache(struct cudbg_init *pdbg_init, · 1048 unmodified lines
1049	1049	}
1050	1050	}
1051	1051
1052		−static unsigned long cudbg_mem_region_size(struct cudbg_init *pdbg_init,
1053		− struct cudbg_error *cudbg_err,
1054		− u8 mem_type)
	1052	+static int cudbg_mem_region_size(struct cudbg_init *pdbg_init,
	1053	+ struct cudbg_error *cudbg_err,
	1054	+ u8 mem_type, unsigned long *region_size)
1055	1055	{
1056	1056	struct adapter *padap = pdbg_init->adap;
1057	1057	struct cudbg_meminfo mem_info;
@@ -1060,15 +1060,23 @@ static unsigned long cudbg_mem_region_size(struct cudbg_init *pdbg_init, · 2 unmodified lines
1060	1060
1061	1061	memset(&mem_info, 0, sizeof(struct cudbg_meminfo));
1062	1062	rc = cudbg_fill_meminfo(padap, &mem_info);
1063		− if (rc)
	1063	+ if (rc) {
	1064	+ cudbg_err->sys_err = rc;
1064	1065	return rc;
	1066	+ }
1065	1067
1066	1068	cudbg_t4_fwcache(pdbg_init, cudbg_err);
1067	1069	rc = cudbg_meminfo_get_mem_index(padap, &mem_info, mem_type, &mc_idx);
1068		− if (rc)
	1070	+ if (rc) {
	1071	+ cudbg_err->sys_err = rc;
1069	1072	return rc;
	1073	+ }
	1074	+
	1075	+ if (region_size)
	1076	+ *region_size = mem_info.avail[mc_idx].limit -
	1077	+ mem_info.avail[mc_idx].base;
1070	1078
1071		− return mem_info.avail[mc_idx].limit - mem_info.avail[mc_idx].base;
	1079	+ return 0;
1072	1080	}
1073	1081
1074	1082	static int cudbg_collect_mem_region(struct cudbg_init *pdbg_init,
@@ -1076,7 +1084,12 @@ static int cudbg_collect_mem_region(struct cudbg_init *pdbg_init, · 1 unmodified line
1076	1084	struct cudbg_error *cudbg_err,
1077	1085	u8 mem_type)
1078	1086	{
1079		− unsigned long size = cudbg_mem_region_size(pdbg_init, cudbg_err, mem_type);
	1087	+ unsigned long size = 0;
	1088	+ int rc;
	1089	+
	1090	+ rc = cudbg_mem_region_size(pdbg_init, cudbg_err, mem_type, &size);
	1091	+ if (rc)
	1092	+ return rc;
1080	1093
1081	1094	return cudbg_read_fw_mem(pdbg_init, dbg_buff, mem_type, size,
1082	1095	cudbg_err);

drivers/net/ethernet/chelsio/cxgb4/cxgb4_ptp.c+6−21

@@ -311,32 +311,17 @@ static int cxgb4_ptp_adjtime(struct ptp_clock_info *ptp, s64 delta) · 310 unmodified lines
311	311	*/
312	312	static int cxgb4_ptp_gettime(struct ptp_clock_info ptp, struct timespec64 ts)
313	313	{
314		− struct adapter adapter = (struct adapter )container_of(ptp,
315		− struct adapter, ptp_clock_info);
316		− struct fw_ptp_cmd c;
	314	+ struct adapter *adapter = container_of(ptp, struct adapter,
	315	+ ptp_clock_info);
317	316	u64 ns;
318		− int err;
319		−
320		− memset(&c, 0, sizeof(c));
321		− c.op_to_portid = cpu_to_be32(FW_CMD_OP_V(FW_PTP_CMD) \|
322		− FW_CMD_REQUEST_F \|
323		− FW_CMD_READ_F \|
324		− FW_PTP_CMD_PORTID_V(0));
325		− c.retval_len16 = cpu_to_be32(FW_CMD_LEN16_V(sizeof(c) / 16));
326		− c.u.ts.sc = FW_PTP_SC_GET_TIME;
327	317
328		− err = t4_wr_mbox(adapter, adapter->mbox, &c, sizeof(c), &c);
329		− if (err < 0) {
330		− dev_err(adapter->pdev_dev,
331		− "PTP: %s error %d\n", __func__, -err);
332		− return err;
333		− }
	318	+ ns = t4_read_reg(adapter, T5_PORT_REG(0, MAC_PORT_PTP_SUM_LO_A));
	319	+ ns \|= (u64)t4_read_reg(adapter,
	320	+ T5_PORT_REG(0, MAC_PORT_PTP_SUM_HI_A)) << 32;
334	321
335	322	/* convert to timespec*/
336		− ns = be64_to_cpu(c.u.ts.tm);
337	323	*ts = ns_to_timespec64(ns);
338		−
339		− return err;
	324	+ return 0;
340	325	}
341	326
342	327	/**

drivers/net/ethernet/chelsio/cxgb4/sge.c+37−3

@@ -2207,6 +2207,9 @@ static void ethofld_hard_xmit(struct net_device *dev, · 2206 unmodified lines
2207	2207	if (unlikely(skip_eotx_wr)) {
2208	2208	start = (u64 *)wr;
2209	2209	eosw_txq->state = next_state;
	2210	+ eosw_txq->cred -= wrlen16;
	2211	+ eosw_txq->ncompl++;
	2212	+ eosw_txq->last_compl = 0;
2210	2213	goto write_wr_headers;
2211	2214	}
2212	2215
@@ -2365,6 +2368,34 @@ netdev_tx_t t4_start_xmit(struct sk_buff skb, struct net_device dev) · 152 unmodified lines
2365	2368	return cxgb4_eth_xmit(skb, dev);
2366	2369	}
2367	2370
	2371	+static void eosw_txq_flush_pending_skbs(struct sge_eosw_txq *eosw_txq)
	2372	+{
	2373	+ int pktcount = eosw_txq->pidx - eosw_txq->last_pidx;
	2374	+ int pidx = eosw_txq->pidx;
	2375	+ struct sk_buff *skb;
	2376	+
	2377	+ if (!pktcount)
	2378	+ return;
	2379	+
	2380	+ if (pktcount < 0)
	2381	+ pktcount += eosw_txq->ndesc;
	2382	+
	2383	+ while (pktcount--) {
	2384	+ pidx--;
	2385	+ if (pidx < 0)
	2386	+ pidx += eosw_txq->ndesc;
	2387	+
	2388	+ skb = eosw_txq->desc[pidx].skb;
	2389	+ if (skb) {
	2390	+ dev_consume_skb_any(skb);
	2391	+ eosw_txq->desc[pidx].skb = NULL;
	2392	+ eosw_txq->inuse--;
	2393	+ }
	2394	+ }
	2395	+
	2396	+ eosw_txq->pidx = eosw_txq->last_pidx + 1;
	2397	+}
	2398	+
2368	2399	/**
2369	2400	* cxgb4_ethofld_send_flowc - Send ETHOFLD flowc request to bind eotid to tc.
2370	2401	* @dev - netdevice
@@ -2440,9 +2471,11 @@ int cxgb4_ethofld_send_flowc(struct net_device *dev, u32 eotid, u32 tc) · 69 unmodified lines
2440	2471	FW_FLOWC_MNEM_EOSTATE_CLOSING :
2441	2472	FW_FLOWC_MNEM_EOSTATE_ESTABLISHED);
2442	2473
2443		− eosw_txq->cred -= len16;
2444		− eosw_txq->ncompl++;
2445		− eosw_txq->last_compl = 0;
	2474	+ /* Free up any pending skbs to ensure there's room for
	2475	+ * termination FLOWC.
	2476	+ */
	2477	+ if (tc == FW_SCHED_CLS_NONE)
	2478	+ eosw_txq_flush_pending_skbs(eosw_txq);
2446	2479
2447	2480	ret = eosw_txq_enqueue(eosw_txq, skb);
2448	2481	if (ret) {
@@ -2695,6 +2728,7 @@ static void ofldtxq_stop(struct sge_uld_txq q, struct fw_wr_hdr wr) · 246 unmodified lines
2695	2728	* is ever running at a time ...
2696	2729	*/
2697	2730	static void service_ofldq(struct sge_uld_txq *q)
	2731	+ __must_hold(&q->sendq.lock)
2698	2732	{
2699	2733	u64 pos, before, *end;
2700	2734	int credits;

drivers/net/ethernet/chelsio/cxgb4/t4_regs.h+3−0

@@ -1906,6 +1906,9 @@ · 1905 unmodified lines
1906	1906
1907	1907	#define MAC_PORT_CFG2_A 0x818
1908	1908
	1909	+#define MAC_PORT_PTP_SUM_LO_A 0x990
	1910	+#define MAC_PORT_PTP_SUM_HI_A 0x994
	1911	+
1909	1912	#define MPS_CMN_CTL_A 0x9000
1910	1913
1911	1914	#define COUNTPAUSEMCRX_S 5

drivers/net/ethernet/freescale/enetc/enetc_pci_mdio.c+1−1

@@ -74,8 +74,8 @@ err_pci_mem_reg: · 73 unmodified lines
74	74	pci_disable_device(pdev);
75	75	err_pci_enable:
76	76	err_mdiobus_alloc:
77		− iounmap(port_regs);
78	77	err_hw_alloc:
	78	+ iounmap(port_regs);
79	79	err_ioremap:
80	80	return err;
81	81	}

drivers/net/ethernet/ibm/ibmvnic.c+2−1

@@ -2189,7 +2189,8 @@ static void __ibmvnic_reset(struct work_struct *work) · 2188 unmodified lines
2189	2189	rc = do_hard_reset(adapter, rwi, reset_state);
2190	2190	rtnl_unlock();
2191	2191	}
2192		− } else {
	2192	+ } else if (!(rwi->reset_reason == VNIC_RESET_FATAL &&
	2193	+ adapter->from_passive_init)) {
2193	2194	rc = do_reset(adapter, rwi, reset_state);
2194	2195	}
2195	2196	kfree(rwi);

drivers/net/ethernet/intel/e1000/e1000_main.c+1−1

@@ -1476,7 +1476,7 @@ static bool e1000_check_64k_bound(struct e1000_adapter adapter, void start, · 1475 unmodified lines
1476	1476	if (hw->mac_type == e1000_82545 \|\|
1477	1477	hw->mac_type == e1000_ce4100 \|\|
1478	1478	hw->mac_type == e1000_82546) {
1479		− return ((begin ^ (end - 1)) >> 16) != 0 ? false : true;
	1479	+ return ((begin ^ (end - 1)) >> 16) == 0;
1480	1480	}
1481	1481
1482	1482	return true;

drivers/net/ethernet/intel/i40e/i40e_main.c+2−2

@@ -1611,7 +1611,7 @@ static int i40e_config_rss_aq(struct i40e_vsi vsi, const u8 seed, · 1610 unmodified lines
1611	1611	}
1612	1612	}
1613	1613	if (lut) {
1614		− bool pf_lut = vsi->type == I40E_VSI_MAIN ? true : false;
	1614	+ bool pf_lut = vsi->type == I40E_VSI_MAIN;
1615	1615
1616	1616	ret = i40e_aq_set_rss_lut(hw, vsi->id, pf_lut, lut, lut_size);
1617	1617	if (ret) {
@@ -11436,7 +11436,7 @@ static int i40e_get_rss_aq(struct i40e_vsi vsi, const u8 seed, · 9818 unmodified lines
11436	11436	}
11437	11437
11438	11438	if (lut) {
11439		− bool pf_lut = vsi->type == I40E_VSI_MAIN ? true : false;
	11439	+ bool pf_lut = vsi->type == I40E_VSI_MAIN;
11440	11440
11441	11441	ret = i40e_aq_get_rss_lut(hw, vsi->id, pf_lut, lut, lut_size);
11442	11442	if (ret) {

drivers/net/ethernet/marvell/mvpp2/mvpp2_cls.c+3−0

@@ -1428,6 +1428,9 @@ int mvpp2_ethtool_cls_rule_del(struct mvpp2_port *port, · 1427 unmodified lines
1428	1428	struct mvpp2_ethtool_fs *efs;
1429	1429	int ret;
1430	1430
	1431	+ if (info->fs.location >= MVPP2_N_RFS_ENTRIES_PER_FLOW)
	1432	+ return -EINVAL;
	1433	+
1431	1434	efs = port->rfs_rules[info->fs.location];
1432	1435	if (!efs)
1433	1436	return -EINVAL;

drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c+2−0

@@ -4329,6 +4329,8 @@ static int mvpp2_ethtool_get_rxfh_context(struct net_device dev, u32 indir, · 4328 unmodified lines
4329	4329
4330	4330	if (!mvpp22_rss_is_supported())
4331	4331	return -EOPNOTSUPP;
	4332	+ if (rss_context >= MVPP22_N_RSS_TABLES)
	4333	+ return -EINVAL;
4332	4334
4333	4335	if (hfunc)
4334	4336	*hfunc = ETH_RSS_HASH_CRC32;

drivers/net/ethernet/mellanox/mlx4/en_tx.c+13−1

@@ -43,6 +43,7 @@ · 42 unmodified lines
43	43	#include <linux/ip.h>
44	44	#include <linux/ipv6.h>
45	45	#include <linux/moduleparam.h>
	46	+#include <linux/indirect_call_wrapper.h>
46	47
47	48	#include "mlx4_en.h"
48	49
@@ -261,6 +262,10 @@ static void mlx4_en_stamp_wqe(struct mlx4_en_priv *priv, · 212 unmodified lines
261	262	}
262	263	}
263	264
	265	+INDIRECT_CALLABLE_DECLARE(u32 mlx4_en_free_tx_desc(struct mlx4_en_priv *priv,
	266	+ struct mlx4_en_tx_ring *ring,
	267	+ int index, u64 timestamp,
	268	+ int napi_mode));
264	269
265	270	u32 mlx4_en_free_tx_desc(struct mlx4_en_priv *priv,
266	271	struct mlx4_en_tx_ring *ring,
@@ -329,6 +334,11 @@ u32 mlx4_en_free_tx_desc(struct mlx4_en_priv *priv, · 62 unmodified lines
329	334	return tx_info->nr_txbb;
330	335	}
331	336
	337	+INDIRECT_CALLABLE_DECLARE(u32 mlx4_en_recycle_tx_desc(struct mlx4_en_priv *priv,
	338	+ struct mlx4_en_tx_ring *ring,
	339	+ int index, u64 timestamp,
	340	+ int napi_mode));
	341	+
332	342	u32 mlx4_en_recycle_tx_desc(struct mlx4_en_priv *priv,
333	343	struct mlx4_en_tx_ring *ring,
334	344	int index, u64 timestamp,
@@ -449,7 +459,9 @@ bool mlx4_en_process_tx_cq(struct net_device *dev, · 114 unmodified lines
449	459	timestamp = mlx4_en_get_cqe_ts(cqe);
450	460
451	461	/* free next descriptor */
452		− last_nr_txbb = ring->free_tx_desc(
	462	+ last_nr_txbb = INDIRECT_CALL_2(ring->free_tx_desc,
	463	+ mlx4_en_free_tx_desc,
	464	+ mlx4_en_recycle_tx_desc,
453	465	priv, ring, ring_index,
454	466	timestamp, napi_budget);
455	467

drivers/net/ethernet/mellanox/mlx4/main.c+3−1

@@ -2550,6 +2550,7 @@ static int mlx4_allocate_default_counters(struct mlx4_dev *dev) · 2549 unmodified lines
2550	2550
2551	2551	if (!err \|\| err == -ENOSPC) {
2552	2552	priv->def_counter[port] = idx;
	2553	+ err = 0;
2553	2554	} else if (err == -ENOENT) {
2554	2555	err = 0;
2555	2556	continue;
@@ -2600,7 +2601,8 @@ int mlx4_counter_alloc(struct mlx4_dev dev, u32 idx, u8 usage) · 44 unmodified lines
2600	2601	MLX4_CMD_TIME_CLASS_A, MLX4_CMD_WRAPPED);
2601	2602	if (!err)
2602	2603	*idx = get_param_l(&out_param);
2603		−
	2604	+ if (WARN_ON(err == -ENOSPC))
	2605	+ err = -EINVAL;
2604	2606	return err;
2605	2607	}
2606	2608	return __mlx4_counter_alloc(dev, idx);

drivers/net/ethernet/mellanox/mlx5/core/Kconfig+4−4

@@ -7,10 +7,10 @@ config MLX5_CORE · 6 unmodified lines
7	7	tristate "Mellanox 5th generation network adapters (ConnectX series) core driver"
8	8	depends on PCI
9	9	select NET_DEVLINK
10		− imply PTP_1588_CLOCK
11		− imply VXLAN
12		− imply MLXFW
13		− imply PCI_HYPERV_INTERFACE
	10	+ depends on VXLAN \|\| !VXLAN
	11	+ depends on MLXFW \|\| !MLXFW
	12	+ depends on PTP_1588_CLOCK \|\| !PTP_1588_CLOCK
	13	+ depends on PCI_HYPERV_INTERFACE \|\| !PCI_HYPERV_INTERFACE
14	14	default n
15	15	---help---
16	16	Core driver for low level functionality of the ConnectX-4 and

drivers/net/ethernet/mellanox/mlx5/core/cmd.c+5−1

@@ -888,7 +888,6 @@ static void cmd_work_handler(struct work_struct *work) · 887 unmodified lines
888	888	}
889	889
890	890	cmd->ent_arr[ent->idx] = ent;
891		− set_bit(MLX5_CMD_ENT_STATE_PENDING_COMP, &ent->state);
892	891	lay = get_inst(cmd, ent->idx);
893	892	ent->lay = lay;
894	893	memset(lay, 0, sizeof(*lay));
@@ -910,6 +909,7 @@ static void cmd_work_handler(struct work_struct *work) · 15 unmodified lines
910	909
911	910	if (ent->callback)
912	911	schedule_delayed_work(&ent->cb_timeout_work, cb_timeout);
	912	+ set_bit(MLX5_CMD_ENT_STATE_PENDING_COMP, &ent->state);
913	913
914	914	/* Skip sending command to fw if internal error */
915	915	if (pci_channel_offline(dev->pdev) \|\|
@@ -922,6 +922,10 @@ static void cmd_work_handler(struct work_struct *work) · 6 unmodified lines
922	922	MLX5_SET(mbox_out, ent->out, syndrome, drv_synd);
923	923
924	924	mlx5_cmd_comp_handler(dev, 1UL << ent->idx, true);
	925	+ /* no doorbell, no need to keep the entry */
	926	+ free_ent(cmd, ent->idx);
	927	+ if (ent->callback)
	928	+ free_cmd(ent);
925	929	return;
926	930	}
927	931

drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c+3−3

@@ -935,7 +935,7 @@ struct mlx5_fw_tracer mlx5_fw_tracer_create(struct mlx5_core_dev dev) · 934 unmodified lines
935	935	return NULL;
936	936	}
937	937
938		− tracer = kzalloc(sizeof(*tracer), GFP_KERNEL);
	938	+ tracer = kvzalloc(sizeof(*tracer), GFP_KERNEL);
939	939	if (!tracer)
940	940	return ERR_PTR(-ENOMEM);
941	941
@@ -982,7 +982,7 @@ destroy_workqueue: · 40 unmodified lines
982	982	tracer->dev = NULL;
983	983	destroy_workqueue(tracer->work_queue);
984	984	free_tracer:
985		− kfree(tracer);
	985	+ kvfree(tracer);
986	986	return ERR_PTR(err);
987	987	}
988	988
@@ -1061,7 +1061,7 @@ void mlx5_fw_tracer_destroy(struct mlx5_fw_tracer *tracer) · 72 unmodified lines
1061	1061	mlx5_fw_tracer_destroy_log_buf(tracer);
1062	1062	flush_workqueue(tracer->work_queue);
1063	1063	destroy_workqueue(tracer->work_queue);
1064		− kfree(tracer);
	1064	+ kvfree(tracer);
1065	1065	}
1066	1066
1067	1067	static int fw_tracer_event(struct notifier_block nb, unsigned long action, void data)

drivers/net/ethernet/mellanox/mlx5/core/en.h+2−1

@@ -367,6 +367,7 @@ enum { · 366 unmodified lines
367	367	MLX5E_SQ_STATE_AM,
368	368	MLX5E_SQ_STATE_TLS,
369	369	MLX5E_SQ_STATE_VLAN_NEED_L2_INLINE,
	370	+ MLX5E_SQ_STATE_PENDING_XSK_TX,
370	371	};
371	372
372	373	struct mlx5e_sq_wqe_info {
@@ -960,7 +961,7 @@ void mlx5e_page_release_dynamic(struct mlx5e_rq *rq, · 587 unmodified lines
960	961	void mlx5e_handle_rx_cqe(struct mlx5e_rq rq, struct mlx5_cqe64 cqe);
961	962	void mlx5e_handle_rx_cqe_mpwrq(struct mlx5e_rq rq, struct mlx5_cqe64 cqe);
962	963	bool mlx5e_post_rx_wqes(struct mlx5e_rq *rq);
963		−void mlx5e_poll_ico_cq(struct mlx5e_cq *cq);
	964	+int mlx5e_poll_ico_cq(struct mlx5e_cq *cq);
964	965	bool mlx5e_post_rx_mpwqes(struct mlx5e_rq *rq);
965	966	void mlx5e_dealloc_rx_wqe(struct mlx5e_rq *rq, u16 ix);
966	967	void mlx5e_dealloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix);

drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c+12−11

@@ -12,6 +12,7 @@ · 11 unmodified lines
12	12	#include <net/flow_offload.h>
13	13	#include <net/netfilter/nf_flow_table.h>
14	14	#include <linux/workqueue.h>
	15	+#include <linux/xarray.h>
15	16
16	17	#include "esw/chains.h"
17	18	#include "en/tc_ct.h"
@@ -35,7 +36,7 @@ struct mlx5_tc_ct_priv { · 17 unmodified lines
35	36	struct mlx5_eswitch *esw;
36	37	const struct net_device *netdev;
37	38	struct idr fte_ids;
38		− struct idr tuple_ids;
	39	+ struct xarray tuple_ids;
39	40	struct rhashtable zone_ht;
40	41	struct mlx5_flow_table *ct;
41	42	struct mlx5_flow_table *ct_nat;
@@ -238,7 +239,7 @@ mlx5_tc_ct_entry_del_rule(struct mlx5_tc_ct_priv *ct_priv, · 196 unmodified lines
238	239
239	240	mlx5_eswitch_del_offloaded_rule(esw, zone_rule->rule, attr);
240	241	mlx5_modify_header_dealloc(esw->dev, attr->modify_hdr);
241		− idr_remove(&ct_priv->tuple_ids, zone_rule->tupleid);
	242	+ xa_erase(&ct_priv->tuple_ids, zone_rule->tupleid);
242	243	}
243	244
244	245	static void
@@ -483,7 +484,7 @@ mlx5_tc_ct_entry_add_rule(struct mlx5_tc_ct_priv *ct_priv, · 238 unmodified lines
483	484	struct mlx5_esw_flow_attr *attr = &zone_rule->attr;
484	485	struct mlx5_eswitch *esw = ct_priv->esw;
485	486	struct mlx5_flow_spec *spec = NULL;
486		− u32 tupleid = 1;
	487	+ u32 tupleid;
487	488	int err;
488	489
489	490	zone_rule->nat = nat;
@@ -493,12 +494,12 @@ mlx5_tc_ct_entry_add_rule(struct mlx5_tc_ct_priv *ct_priv, · 3 unmodified lines
493	494	return -ENOMEM;
494	495
495	496	/* Get tuple unique id */
496		− err = idr_alloc_u32(&ct_priv->tuple_ids, zone_rule, &tupleid,
497		− TUPLE_ID_MAX, GFP_KERNEL);
	497	+ err = xa_alloc(&ct_priv->tuple_ids, &tupleid, zone_rule,
	498	+ XA_LIMIT(1, TUPLE_ID_MAX), GFP_KERNEL);
498	499	if (err) {
499	500	netdev_warn(ct_priv->netdev,
500	501	"Failed to allocate tuple id, err: %d\n", err);
501		− goto err_idr_alloc;
	502	+ goto err_xa_alloc;
502	503	}
503	504	zone_rule->tupleid = tupleid;
504	505
@@ -539,8 +540,8 @@ mlx5_tc_ct_entry_add_rule(struct mlx5_tc_ct_priv *ct_priv, · 34 unmodified lines
539	540	err_rule:
540	541	mlx5_modify_header_dealloc(esw->dev, attr->modify_hdr);
541	542	err_mod_hdr:
542		− idr_remove(&ct_priv->tuple_ids, zone_rule->tupleid);
543		−err_idr_alloc:
	543	+ xa_erase(&ct_priv->tuple_ids, zone_rule->tupleid);
	544	+err_xa_alloc:
544	545	kfree(spec);
545	546	return err;
546	547	}
@@ -1299,7 +1300,7 @@ mlx5_tc_ct_init(struct mlx5_rep_uplink_priv *uplink_priv) · 752 unmodified lines
1299	1300	}
1300	1301
1301	1302	idr_init(&ct_priv->fte_ids);
1302		− idr_init(&ct_priv->tuple_ids);
	1303	+ xa_init_flags(&ct_priv->tuple_ids, XA_FLAGS_ALLOC1);
1303	1304	mutex_init(&ct_priv->control_lock);
1304	1305	rhashtable_init(&ct_priv->zone_ht, &zone_params);
1305	1306
@@ -1334,7 +1335,7 @@ mlx5_tc_ct_clean(struct mlx5_rep_uplink_priv *uplink_priv) · 28 unmodified lines
1334	1335
1335	1336	rhashtable_destroy(&ct_priv->zone_ht);
1336	1337	mutex_destroy(&ct_priv->control_lock);
1337		− idr_destroy(&ct_priv->tuple_ids);
	1338	+ xa_destroy(&ct_priv->tuple_ids);
1338	1339	idr_destroy(&ct_priv->fte_ids);
1339	1340	kfree(ct_priv);
1340	1341
@@ -1352,7 +1353,7 @@ mlx5e_tc_ct_restore_flow(struct mlx5_rep_uplink_priv *uplink_priv, · 11 unmodified lines
1352	1353	if (!ct_priv \|\| !tupleid)
1353	1354	return true;
1354	1355
1355		− zone_rule = idr_find(&ct_priv->tuple_ids, tupleid);
	1356	+ zone_rule = xa_load(&ct_priv->tuple_ids, tupleid);
1356	1357	if (!zone_rule)
1357	1358	return false;
1358	1359

drivers/net/ethernet/mellanox/mlx5/core/en/xsk/tx.c+3−0

@@ -33,6 +33,9 @@ int mlx5e_xsk_wakeup(struct net_device *dev, u32 qid, u32 flags) · 32 unmodified lines
33	33	if (unlikely(!test_bit(MLX5E_SQ_STATE_ENABLED, &c->xskicosq.state)))
34	34	return 0;
35	35
	36	+ if (test_and_set_bit(MLX5E_SQ_STATE_PENDING_XSK_TX, &c->xskicosq.state))
	37	+ return 0;
	38	+
36	39	spin_lock(&c->xskicosq_lock);
37	40	mlx5e_trigger_irq(&c->xskicosq);
38	41	spin_unlock(&c->xskicosq_lock);

drivers/net/ethernet/mellanox/mlx5/core/en_main.c+6−1

@@ -3583,7 +3583,12 @@ mlx5e_get_stats(struct net_device dev, struct rtnl_link_stats64 stats) · 3582 unmodified lines
3583	3583	struct mlx5e_vport_stats *vstats = &priv->stats.vport;
3584	3584	struct mlx5e_pport_stats *pstats = &priv->stats.pport;
3585	3585
3586		− if (!mlx5e_monitor_counter_supported(priv)) {
	3586	+ /* In switchdev mode, monitor counters doesn't monitor
	3587	+ * rx/tx stats of 802_3. The update stats mechanism
	3588	+ * should keep the 802_3 layout counters updated
	3589	+ */
	3590	+ if (!mlx5e_monitor_counter_supported(priv) \|\|
	3591	+ mlx5e_is_uplink_rep(priv)) {
3587	3592	/* update HW stats in background for next time */
3588	3593	mlx5e_queue_update_stats(priv);
3589	3594	}

drivers/net/ethernet/mellanox/mlx5/core/en_rep.c+2−7

@@ -1773,19 +1773,14 @@ static void mlx5e_cleanup_rep_rx(struct mlx5e_priv *priv) · 1772 unmodified lines
1773	1773
1774	1774	static int mlx5e_init_ul_rep_rx(struct mlx5e_priv *priv)
1775	1775	{
1776		− int err = mlx5e_init_rep_rx(priv);
1777		−
1778		− if (err)
1779		− return err;
1780		−
1781	1776	mlx5e_create_q_counters(priv);
1782		− return 0;
	1777	+ return mlx5e_init_rep_rx(priv);
1783	1778	}
1784	1779
1785	1780	static void mlx5e_cleanup_ul_rep_rx(struct mlx5e_priv *priv)
1786	1781	{
1787		− mlx5e_destroy_q_counters(priv);
1788	1782	mlx5e_cleanup_rep_rx(priv);
	1783	+ mlx5e_destroy_q_counters(priv);
1789	1784	}
1790	1785
1791	1786	static int mlx5e_init_uplink_rep_tx(struct mlx5e_rep_priv *rpriv)

drivers/net/ethernet/mellanox/mlx5/core/en_rx.c+5−3

@@ -589,7 +589,7 @@ bool mlx5e_post_rx_wqes(struct mlx5e_rq *rq) · 588 unmodified lines
589	589	return !!err;
590	590	}
591	591
592		−void mlx5e_poll_ico_cq(struct mlx5e_cq *cq)
	592	+int mlx5e_poll_ico_cq(struct mlx5e_cq *cq)
593	593	{
594	594	struct mlx5e_icosq *sq = container_of(cq, struct mlx5e_icosq, cq);
595	595	struct mlx5_cqe64 *cqe;
@@ -597,11 +597,11 @@ void mlx5e_poll_ico_cq(struct mlx5e_cq *cq) · 1 unmodified line
597	597	int i;
598	598
599	599	if (unlikely(!test_bit(MLX5E_SQ_STATE_ENABLED, &sq->state)))
600		− return;
	600	+ return 0;
601	601
602	602	cqe = mlx5_cqwq_get_cqe(&cq->wq);
603	603	if (likely(!cqe))
604		− return;
	604	+ return 0;
605	605
606	606	/* sq->cc must be updated only after mlx5_cqwq_update_db_record(),
607	607	* otherwise a cq overrun may occur
@@ -650,6 +650,8 @@ void mlx5e_poll_ico_cq(struct mlx5e_cq *cq) · 42 unmodified lines
650	650	sq->cc = sqcc;
651	651
652	652	mlx5_cqwq_update_db_record(&cq->wq);
	653	+
	654	+ return i;
653	655	}
654	656
655	657	bool mlx5e_post_rx_mpwqes(struct mlx5e_rq *rq)

drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c+5−1

@@ -152,7 +152,11 @@ int mlx5e_napi_poll(struct napi_struct *napi, int budget) · 151 unmodified lines
152	152	mlx5e_post_rx_wqes,
153	153	rq);
154	154	if (xsk_open) {
155		− mlx5e_poll_ico_cq(&c->xskicosq.cq);
	155	+ if (mlx5e_poll_ico_cq(&c->xskicosq.cq))
	156	+ /* Don't clear the flag if nothing was polled to prevent
	157	+ * queueing more WQEs and overflowing XSKICOSQ.
	158	+ */
	159	+ clear_bit(MLX5E_SQ_STATE_PENDING_XSK_TX, &c->xskicosq.state);
156	160	busy \|= mlx5e_poll_xdpsq_cq(&xsksq->cq);
157	161	busy_xsk \|= mlx5e_napi_xsk_post(xsksq, xskrq);
158	162	}

drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c+9−9

@@ -1550,9 +1550,9 @@ static int esw_create_restore_table(struct mlx5_eswitch *esw) · 1549 unmodified lines
1550	1550	MLX5_FLOW_NAMESPACE_KERNEL, 1,
1551	1551	modact);
1552	1552	if (IS_ERR(mod_hdr)) {
	1553	+ err = PTR_ERR(mod_hdr);
1553	1554	esw_warn(dev, "Failed to create restore mod header, err: %d\n",
1554	1555	err);
1555		− err = PTR_ERR(mod_hdr);
1556	1556	goto err_mod_hdr;
1557	1557	}
1558	1558
@@ -2219,10 +2219,12 @@ static int esw_offloads_steering_init(struct mlx5_eswitch *esw) · 660 unmodified lines
2219	2219	total_vports = num_vfs + MLX5_SPECIAL_VPORTS(esw->dev);
2220	2220
2221	2221	memset(&esw->fdb_table.offloads, 0, sizeof(struct offloads_fdb));
	2222	+ mutex_init(&esw->fdb_table.offloads.vports.lock);
	2223	+ hash_init(esw->fdb_table.offloads.vports.table);
2222	2224
2223	2225	err = esw_create_uplink_offloads_acl_tables(esw);
2224	2226	if (err)
2225		− return err;
	2227	+ goto create_acl_err;
2226	2228
2227	2229	err = esw_create_offloads_table(esw, total_vports);
2228	2230	if (err)
@@ -2240,9 +2242,6 @@ static int esw_offloads_steering_init(struct mlx5_eswitch *esw) · 11 unmodified lines
2240	2242	if (err)
2241	2243	goto create_fg_err;
2242	2244
2243		− mutex_init(&esw->fdb_table.offloads.vports.lock);
2244		− hash_init(esw->fdb_table.offloads.vports.table);
2245		−
2246	2245	return 0;
2247	2246
2248	2247	create_fg_err:
@@ -2253,18 +2252,19 @@ create_restore_err: · 4 unmodified lines
2253	2252	esw_destroy_offloads_table(esw);
2254	2253	create_offloads_err:
2255	2254	esw_destroy_uplink_offloads_acl_tables(esw);
2256		−
	2255	+create_acl_err:
	2256	+ mutex_destroy(&esw->fdb_table.offloads.vports.lock);
2257	2257	return err;
2258	2258	}
2259	2259
2260	2260	static void esw_offloads_steering_cleanup(struct mlx5_eswitch *esw)
2261	2261	{
2262		− mutex_destroy(&esw->fdb_table.offloads.vports.lock);
2263	2262	esw_destroy_vport_rx_group(esw);
2264	2263	esw_destroy_offloads_fdb_tables(esw);
2265	2264	esw_destroy_restore_table(esw);
2266	2265	esw_destroy_offloads_table(esw);
2267	2266	esw_destroy_uplink_offloads_acl_tables(esw);
	2267	+ mutex_destroy(&esw->fdb_table.offloads.vports.lock);
2268	2268	}
2269	2269
2270	2270	static void
@@ -2377,9 +2377,9 @@ int esw_offloads_enable(struct mlx5_eswitch *esw) · 106 unmodified lines
2377	2377	err_vports:
2378	2378	esw_offloads_unload_rep(esw, MLX5_VPORT_UPLINK);
2379	2379	err_uplink:
2380		− esw_set_passing_vport_metadata(esw, false);
2381		−err_steering_init:
2382	2380	esw_offloads_steering_cleanup(esw);
	2381	+err_steering_init:
	2382	+ esw_set_passing_vport_metadata(esw, false);
2383	2383	err_vport_metadata:
2384	2384	mlx5_rdma_disable_roce(esw->dev);
2385	2385	mutex_destroy(&esw->offloads.termtbl_mutex);

drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c+13−1

@@ -695,6 +695,12 @@ static void dr_cq_event(struct mlx5_core_cq *mcq, · 694 unmodified lines
695	695	pr_info("CQ event %u on CQ #%u\n", event, mcq->cqn);
696	696	}
697	697
	698	+static void dr_cq_complete(struct mlx5_core_cq *mcq,
	699	+ struct mlx5_eqe *eqe)
	700	+{
	701	+ pr_err("CQ completion CQ: #%u\n", mcq->cqn);
	702	+}
	703	+
698	704	static struct mlx5dr_cq dr_create_cq(struct mlx5_core_dev mdev,
699	705	struct mlx5_uars_page *uar,
700	706	size_t ncqe)
@@ -756,6 +762,7 @@ static struct mlx5dr_cq dr_create_cq(struct mlx5_core_dev mdev, · 55 unmodified lines
756	762	mlx5_fill_page_frag_array(&cq->wq_ctrl.buf, pas);
757	763
758	764	cq->mcq.event = dr_cq_event;
	765	+ cq->mcq.comp = dr_cq_complete;
759	766
760	767	err = mlx5_core_create_cq(mdev, &cq->mcq, in, inlen, out, sizeof(out));
761	768	kvfree(in);
@@ -767,7 +774,12 @@ static struct mlx5dr_cq dr_create_cq(struct mlx5_core_dev mdev, · 5 unmodified lines
767	774	cq->mcq.set_ci_db = cq->wq_ctrl.db.db;
768	775	cq->mcq.arm_db = cq->wq_ctrl.db.db + 1;
769	776	*cq->mcq.set_ci_db = 0;
770		− *cq->mcq.arm_db = 0;
	777	+
	778	+ /* set no-zero value, in order to avoid the HW to run db-recovery on
	779	+ * CQ that used in polling mode.
	780	+ */
	781	+ *cq->mcq.arm_db = cpu_to_be32(2 << 28);
	782	+
771	783	cq->mcq.vector = 0;
772	784	cq->mcq.irqn = irqn;
773	785	cq->mcq.uar = uar;

drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_actions.c+2−2

@@ -380,7 +380,7 @@ struct mlxsw_afa_block mlxsw_afa_block_create(struct mlxsw_afa mlxsw_afa) · 379 unmodified lines
380	380
381	381	block = kzalloc(sizeof(*block), GFP_KERNEL);
382	382	if (!block)
383		− return NULL;
	383	+ return ERR_PTR(-ENOMEM);
384	384	INIT_LIST_HEAD(&block->resource_list);
385	385	block->afa = mlxsw_afa;
386	386
@@ -408,7 +408,7 @@ err_second_set_create: · 21 unmodified lines
408	408	mlxsw_afa_set_destroy(block->first_set);
409	409	err_first_set_create:
410	410	kfree(block);
411		− return NULL;
	411	+ return ERR_PTR(-ENOMEM);
412	412	}
413	413	EXPORT_SYMBOL(mlxsw_afa_block_create);
414	414

drivers/net/ethernet/mellanox/mlxsw/spectrum2_acl_tcam.c+2−2

@@ -88,8 +88,8 @@ static int mlxsw_sp2_acl_tcam_init(struct mlxsw_sp mlxsw_sp, void priv, · 87 unmodified lines
88	88	* to be written using PEFA register to all indexes for all regions.
89	89	*/
90	90	afa_block = mlxsw_afa_block_create(mlxsw_sp->afa);
91		− if (!afa_block) {
92		− err = -ENOMEM;
	91	+ if (IS_ERR(afa_block)) {
	92	+ err = PTR_ERR(afa_block);
93	93	goto err_afa_block;
94	94	}
95	95	err = mlxsw_afa_block_continue(afa_block);

drivers/net/ethernet/mellanox/mlxsw/spectrum_acl.c+1−1

@@ -464,7 +464,7 @@ mlxsw_sp_acl_rulei_create(struct mlxsw_sp_acl *acl, · 463 unmodified lines
464	464
465	465	rulei = kzalloc(sizeof(*rulei), GFP_KERNEL);
466	466	if (!rulei)
467		− return NULL;
	467	+ return ERR_PTR(-ENOMEM);
468	468
469	469	if (afa_block) {
470	470	rulei->act_block = afa_block;

drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c+10−2

@@ -986,8 +986,9 @@ mlxsw_sp_acl_tcam_vchunk_create(struct mlxsw_sp *mlxsw_sp, · 985 unmodified lines
986	986	unsigned int priority,
987	987	struct mlxsw_afk_element_usage *elusage)
988	988	{
	989	+ struct mlxsw_sp_acl_tcam_vchunk vchunk, vchunk2;
989	990	struct mlxsw_sp_acl_tcam_vregion *vregion;
990		− struct mlxsw_sp_acl_tcam_vchunk *vchunk;
	991	+ struct list_head *pos;
991	992	int err;
992	993
993	994	if (priority == MLXSW_SP_ACL_TCAM_CATCHALL_PRIO)
@@ -1025,7 +1026,14 @@ mlxsw_sp_acl_tcam_vchunk_create(struct mlxsw_sp *mlxsw_sp, · 31 unmodified lines
1025	1026	}
1026	1027
1027	1028	mlxsw_sp_acl_tcam_rehash_ctx_vregion_changed(vregion);
1028		− list_add_tail(&vchunk->list, &vregion->vchunk_list);
	1029	+
	1030	+ /* Position the vchunk inside the list according to priority */
	1031	+ list_for_each(pos, &vregion->vchunk_list) {
	1032	+ vchunk2 = list_entry(pos, typeof(*vchunk2), list);
	1033	+ if (vchunk2->priority > priority)
	1034	+ break;
	1035	+ }
	1036	+ list_add_tail(&vchunk->list, pos);
1029	1037	mutex_unlock(&vregion->lock);
1030	1038
1031	1039	return vchunk;

drivers/net/ethernet/mellanox/mlxsw/spectrum_flower.c+2−1

@@ -36,7 +36,8 @@ static int mlxsw_sp_flower_parse_actions(struct mlxsw_sp *mlxsw_sp, · 35 unmodified lines
36	36	err = mlxsw_sp_acl_rulei_act_count(mlxsw_sp, rulei, extack);
37	37	if (err)
38	38	return err;
39		− } else if (act->hw_stats != FLOW_ACTION_HW_STATS_DISABLED) {
	39	+ } else if (act->hw_stats != FLOW_ACTION_HW_STATS_DISABLED &&
	40	+ act->hw_stats != FLOW_ACTION_HW_STATS_DONT_CARE) {
40	41	NL_SET_ERR_MSG_MOD(extack, "Unsupported action HW stats type");
41	42	return -EOPNOTSUPP;
42	43	}

drivers/net/ethernet/mellanox/mlxsw/spectrum_mr_tcam.c+2−2

@@ -199,8 +199,8 @@ mlxsw_sp_mr_tcam_afa_block_create(struct mlxsw_sp *mlxsw_sp, · 198 unmodified lines
199	199	int err;
200	200
201	201	afa_block = mlxsw_afa_block_create(mlxsw_sp->afa);
202		− if (!afa_block)
203		− return ERR_PTR(-ENOMEM);
	202	+ if (IS_ERR(afa_block))
	203	+ return afa_block;
204	204
205	205	err = mlxsw_afa_block_append_allocated_counter(afa_block,
206	206	counter_index);

drivers/net/ethernet/moxa/moxart_ether.c+1−1

@@ -564,7 +564,7 @@ static int moxart_remove(struct platform_device *pdev) · 563 unmodified lines
564	564	struct net_device *ndev = platform_get_drvdata(pdev);
565	565
566	566	unregister_netdev(ndev);
567		− free_irq(ndev->irq, ndev);
	567	+ devm_free_irq(&pdev->dev, ndev->irq, ndev);
568	568	moxart_mac_free_memory(ndev);
569	569	free_netdev(ndev);
570	570

drivers/net/ethernet/mscc/ocelot.c+11−6

@@ -1031,10 +1031,8 @@ int ocelot_fdb_dump(struct ocelot *ocelot, int port, · 1030 unmodified lines
1031	1031	{
1032	1032	int i, j;
1033	1033
1034		− /* Loop through all the mac tables entries. There are 1024 rows of 4
1035		− * entries.
1036		− */
1037		− for (i = 0; i < 1024; i++) {
	1034	+ /* Loop through all the mac tables entries. */
	1035	+ for (i = 0; i < ocelot->num_mact_rows; i++) {
1038	1036	for (j = 0; j < 4; j++) {
1039	1037	struct ocelot_mact_entry entry;
1040	1038	bool is_static;
@@ -1453,8 +1451,15 @@ static void ocelot_port_attr_stp_state_set(struct ocelot *ocelot, int port, · 412 unmodified lines
1453	1451
1454	1452	void ocelot_set_ageing_time(struct ocelot *ocelot, unsigned int msecs)
1455	1453	{
1456		− ocelot_write(ocelot, ANA_AUTOAGE_AGE_PERIOD(msecs / 2),
1457		− ANA_AUTOAGE);
	1454	+ unsigned int age_period = ANA_AUTOAGE_AGE_PERIOD(msecs / 2000);
	1455	+
	1456	+ /* Setting AGE_PERIOD to zero effectively disables automatic aging,
	1457	+ * which is clearly not what our intention is. So avoid that.
	1458	+ */
	1459	+ if (!age_period)
	1460	+ age_period = 1;
	1461	+
	1462	+ ocelot_rmw(ocelot, age_period, ANA_AUTOAGE_AGE_PERIOD_M, ANA_AUTOAGE);
1458	1463	}
1459	1464	EXPORT_SYMBOL(ocelot_set_ageing_time);
1460	1465

drivers/net/ethernet/mscc/ocelot_regs.c+1−0

@@ -431,6 +431,7 @@ int ocelot_chip_init(struct ocelot ocelot, const struct ocelot_ops ops) · 430 unmodified lines
431	431	ocelot->stats_layout = ocelot_stats_layout;
432	432	ocelot->num_stats = ARRAY_SIZE(ocelot_stats_layout);
433	433	ocelot->shared_queue_sz = 224 * 1024;
	434	+ ocelot->num_mact_rows = 1024;
434	435	ocelot->ops = ops;
435	436
436	437	ret = ocelot_regfields_init(ocelot, ocelot_regfields);

drivers/net/ethernet/natsemi/jazzsonic.c+4−2

@@ -208,11 +208,13 @@ static int jazz_sonic_probe(struct platform_device *pdev) · 207 unmodified lines
208	208
209	209	err = register_netdev(dev);
210	210	if (err)
211		− goto out1;
	211	+ goto undo_probe1;
212	212
213	213	return 0;
214	214
215		−out1:
	215	+undo_probe1:
	216	+ dma_free_coherent(lp->device, SIZEOF_SONIC_DESC * SONIC_BUS_SCALE(lp->dma_bitmode),
	217	+ lp->descriptors, lp->descriptors_laddr);
216	218	release_mem_region(dev->base_addr, SONIC_MEM_SIZE);
217	219	out:
218	220	free_netdev(dev);

drivers/net/ethernet/netronome/nfp/abm/main.c+1−0

@@ -283,6 +283,7 @@ nfp_abm_vnic_set_mac(struct nfp_pf pf, struct nfp_abm abm, struct nfp_net *nn, · 282 unmodified lines
283	283	if (!nfp_nsp_has_hwinfo_lookup(nsp)) {
284	284	nfp_warn(pf->cpp, "NSP doesn't support PF MAC generation\n");
285	285	eth_hw_addr_random(nn->dp.netdev);
	286	+ nfp_nsp_close(nsp);
286	287	return;
287	288	}
288	289

drivers/net/ethernet/pensando/ionic/ionic_debugfs.c+1−2

@@ -170,8 +170,7 @@ void ionic_debugfs_add_qcq(struct ionic_lif lif, struct ionic_qcq qcq) · 169 unmodified lines
170	170	debugfs_create_x64("base_pa", 0400, cq_dentry, &cq->base_pa);
171	171	debugfs_create_u32("num_descs", 0400, cq_dentry, &cq->num_descs);
172	172	debugfs_create_u32("desc_size", 0400, cq_dentry, &cq->desc_size);
173		− debugfs_create_u8("done_color", 0400, cq_dentry,
174		− (u8 *)&cq->done_color);
	173	+ debugfs_create_bool("done_color", 0400, cq_dentry, &cq->done_color);
175	174
176	175	debugfs_create_file("tail", 0400, cq_dentry, cq, &cq_tail_fops);
177	176

drivers/net/ethernet/pensando/ionic/ionic_lif.c+2−2

@@ -2101,6 +2101,7 @@ static void ionic_lif_handle_fw_down(struct ionic_lif *lif) · 2100 unmodified lines
2101	2101	ionic_txrx_free(lif);
2102	2102	}
2103	2103	ionic_lifs_deinit(ionic);
	2104	+ ionic_reset(ionic);
2104	2105	ionic_qcqs_free(lif);
2105	2106
2106	2107	dev_info(ionic->dev, "FW Down: LIFs stopped\n");
@@ -2116,6 +2117,7 @@ static void ionic_lif_handle_fw_up(struct ionic_lif *lif) · 9 unmodified lines
2116	2117
2117	2118	dev_info(ionic->dev, "FW Up: restarting LIFs\n");
2118	2119
	2120	+ ionic_init_devinfo(ionic);
2119	2121	err = ionic_qcqs_alloc(lif);
2120	2122	if (err)
2121	2123	goto err_out;
@@ -2549,8 +2551,6 @@ int ionic_lifs_register(struct ionic *ionic) · 427 unmodified lines
2549	2551	dev_err(ionic->dev, "Cannot register net device, aborting\n");
2550	2552	return err;
2551	2553	}
2552		−
2553		− ionic_link_status_check_request(ionic->master_lif);
2554	2554	ionic->master_lif->registered = true;
2555	2555
2556	2556	return 0;

drivers/net/ethernet/sgi/meth.h+0−16

@@ -1,19 +1,3 @@
1		−
2		−/*
3		− * snull.h -- definitions for the network module
4		− *
5		− * Copyright (C) 2001 Alessandro Rubini and Jonathan Corbet
6		− * Copyright (C) 2001 O'Reilly & Associates
7		− *
8		− * The source code in this file can be freely used, adapted,
9		− * and redistributed in source or binary form, so long as an
10		− * acknowledgment appears in derived source files. The citation
11		− * should list that the code comes from the book "Linux Device
12		− * Drivers" by Alessandro Rubini and Jonathan Corbet, published
13		− * by O'Reilly & Associates. No warranty is attached;
14		− * we cannot take responsibility for errors or fitness for use.
15		− */
16		−
17	1	/* version dependencies have been confined to a separate file */
18	2
19	3	/* Tunable parameters */

drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c+189−0

@@ -5,8 +5,13 @@ · 4 unmodified lines
5	5	#include <linux/clk-provider.h>
6	6	#include <linux/pci.h>
7	7	#include <linux/dmi.h>
	8	+#include "dwmac-intel.h"
8	9	#include "stmmac.h"
9	10
	11	+struct intel_priv_data {
	12	+ int mdio_adhoc_addr; /* mdio address for serdes & etc */
	13	+};
	14	+
10	15	/* This struct is used to associate PCI Function of MAC controller on a board,
11	16	* discovered via DMI, with the address of PHY connected to the MAC. The
12	17	* negative value of the address means that MAC controller is not connected
@@ -49,6 +54,172 @@ static int stmmac_pci_find_phy_addr(struct pci_dev *pdev, · 36 unmodified lines
49	54	return -ENODEV;
50	55	}
51	56
	57	+static int serdes_status_poll(struct stmmac_priv *priv, int phyaddr,
	58	+ int phyreg, u32 mask, u32 val)
	59	+{
	60	+ unsigned int retries = 10;
	61	+ int val_rd;
	62	+
	63	+ do {
	64	+ val_rd = mdiobus_read(priv->mii, phyaddr, phyreg);
	65	+ if ((val_rd & mask) == (val & mask))
	66	+ return 0;
	67	+ udelay(POLL_DELAY_US);
	68	+ } while (--retries);
	69	+
	70	+ return -ETIMEDOUT;
	71	+}
	72	+
	73	+static int intel_serdes_powerup(struct net_device ndev, void priv_data)
	74	+{
	75	+ struct intel_priv_data *intel_priv = priv_data;
	76	+ struct stmmac_priv *priv = netdev_priv(ndev);
	77	+ int serdes_phy_addr = 0;
	78	+ u32 data = 0;
	79	+
	80	+ if (!intel_priv->mdio_adhoc_addr)
	81	+ return 0;
	82	+
	83	+ serdes_phy_addr = intel_priv->mdio_adhoc_addr;
	84	+
	85	+ /* assert clk_req */
	86	+ data = mdiobus_read(priv->mii, serdes_phy_addr,
	87	+ SERDES_GCR0);
	88	+
	89	+ data \|= SERDES_PLL_CLK;
	90	+
	91	+ mdiobus_write(priv->mii, serdes_phy_addr,
	92	+ SERDES_GCR0, data);
	93	+
	94	+ /* check for clk_ack assertion */
	95	+ data = serdes_status_poll(priv, serdes_phy_addr,
	96	+ SERDES_GSR0,
	97	+ SERDES_PLL_CLK,
	98	+ SERDES_PLL_CLK);
	99	+
	100	+ if (data) {
	101	+ dev_err(priv->device, "Serdes PLL clk request timeout\n");
	102	+ return data;
	103	+ }
	104	+
	105	+ /* assert lane reset */
	106	+ data = mdiobus_read(priv->mii, serdes_phy_addr,
	107	+ SERDES_GCR0);
	108	+
	109	+ data \|= SERDES_RST;
	110	+
	111	+ mdiobus_write(priv->mii, serdes_phy_addr,
	112	+ SERDES_GCR0, data);
	113	+
	114	+ /* check for assert lane reset reflection */
	115	+ data = serdes_status_poll(priv, serdes_phy_addr,
	116	+ SERDES_GSR0,
	117	+ SERDES_RST,
	118	+ SERDES_RST);
	119	+
	120	+ if (data) {
	121	+ dev_err(priv->device, "Serdes assert lane reset timeout\n");
	122	+ return data;
	123	+ }
	124	+
	125	+ /* move power state to P0 */
	126	+ data = mdiobus_read(priv->mii, serdes_phy_addr,
	127	+ SERDES_GCR0);
	128	+
	129	+ data &= ~SERDES_PWR_ST_MASK;
	130	+ data \|= SERDES_PWR_ST_P0 << SERDES_PWR_ST_SHIFT;
	131	+
	132	+ mdiobus_write(priv->mii, serdes_phy_addr,
	133	+ SERDES_GCR0, data);
	134	+
	135	+ /* Check for P0 state */
	136	+ data = serdes_status_poll(priv, serdes_phy_addr,
	137	+ SERDES_GSR0,
	138	+ SERDES_PWR_ST_MASK,
	139	+ SERDES_PWR_ST_P0 << SERDES_PWR_ST_SHIFT);
	140	+
	141	+ if (data) {
	142	+ dev_err(priv->device, "Serdes power state P0 timeout.\n");
	143	+ return data;
	144	+ }
	145	+
	146	+ return 0;
	147	+}
	148	+
	149	+static void intel_serdes_powerdown(struct net_device ndev, void intel_data)
	150	+{
	151	+ struct intel_priv_data *intel_priv = intel_data;
	152	+ struct stmmac_priv *priv = netdev_priv(ndev);
	153	+ int serdes_phy_addr = 0;
	154	+ u32 data = 0;
	155	+
	156	+ if (!intel_priv->mdio_adhoc_addr)
	157	+ return;
	158	+
	159	+ serdes_phy_addr = intel_priv->mdio_adhoc_addr;
	160	+
	161	+ /* move power state to P3 */
	162	+ data = mdiobus_read(priv->mii, serdes_phy_addr,
	163	+ SERDES_GCR0);
	164	+
	165	+ data &= ~SERDES_PWR_ST_MASK;
	166	+ data \|= SERDES_PWR_ST_P3 << SERDES_PWR_ST_SHIFT;
	167	+
	168	+ mdiobus_write(priv->mii, serdes_phy_addr,
	169	+ SERDES_GCR0, data);
	170	+
	171	+ /* Check for P3 state */
	172	+ data = serdes_status_poll(priv, serdes_phy_addr,
	173	+ SERDES_GSR0,
	174	+ SERDES_PWR_ST_MASK,
	175	+ SERDES_PWR_ST_P3 << SERDES_PWR_ST_SHIFT);
	176	+
	177	+ if (data) {
	178	+ dev_err(priv->device, "Serdes power state P3 timeout\n");
	179	+ return;
	180	+ }
	181	+
	182	+ /* de-assert clk_req */
	183	+ data = mdiobus_read(priv->mii, serdes_phy_addr,
	184	+ SERDES_GCR0);
	185	+
	186	+ data &= ~SERDES_PLL_CLK;
	187	+
	188	+ mdiobus_write(priv->mii, serdes_phy_addr,
	189	+ SERDES_GCR0, data);
	190	+
	191	+ /* check for clk_ack de-assert */
	192	+ data = serdes_status_poll(priv, serdes_phy_addr,
	193	+ SERDES_GSR0,
	194	+ SERDES_PLL_CLK,
	195	+ (u32)~SERDES_PLL_CLK);
	196	+
	197	+ if (data) {
	198	+ dev_err(priv->device, "Serdes PLL clk de-assert timeout\n");
	199	+ return;
	200	+ }
	201	+
	202	+ /* de-assert lane reset */
	203	+ data = mdiobus_read(priv->mii, serdes_phy_addr,
	204	+ SERDES_GCR0);
	205	+
	206	+ data &= ~SERDES_RST;
	207	+
	208	+ mdiobus_write(priv->mii, serdes_phy_addr,
	209	+ SERDES_GCR0, data);
	210	+
	211	+ /* check for de-assert lane reset reflection */
	212	+ data = serdes_status_poll(priv, serdes_phy_addr,
	213	+ SERDES_GSR0,
	214	+ SERDES_RST,
	215	+ (u32)~SERDES_RST);
	216	+
	217	+ if (data) {
	218	+ dev_err(priv->device, "Serdes de-assert lane reset timeout\n");
	219	+ return;
	220	+ }
	221	+}
	222	+
52	223	static void common_default_data(struct plat_stmmacenet_data *plat)
53	224	{
54	225	plat->clk_csr = 2; /* clk_csr_i = 20-35MHz & MDC = clk_csr_i/16 */
@@ -189,6 +360,9 @@ static int ehl_sgmii_data(struct pci_dev *pdev, · 134 unmodified lines
189	360	plat->phy_addr = 0;
190	361	plat->phy_interface = PHY_INTERFACE_MODE_SGMII;
191	362
	363	+ plat->serdes_powerup = intel_serdes_powerup;
	364	+ plat->serdes_powerdown = intel_serdes_powerdown;
	365	+
192	366	return ehl_common_data(pdev, plat);
193	367	}
194	368
@@ -233,6 +407,8 @@ static int ehl_pse0_sgmii1g_data(struct pci_dev *pdev, · 38 unmodified lines
233	407	struct plat_stmmacenet_data *plat)
234	408	{
235	409	plat->phy_interface = PHY_INTERFACE_MODE_SGMII;
	410	+ plat->serdes_powerup = intel_serdes_powerup;
	411	+ plat->serdes_powerdown = intel_serdes_powerdown;
236	412	return ehl_pse0_common_data(pdev, plat);
237	413	}
238	414
@@ -263,6 +439,8 @@ static int ehl_pse1_sgmii1g_data(struct pci_dev *pdev, · 24 unmodified lines
263	439	struct plat_stmmacenet_data *plat)
264	440	{
265	441	plat->phy_interface = PHY_INTERFACE_MODE_SGMII;
	442	+ plat->serdes_powerup = intel_serdes_powerup;
	443	+ plat->serdes_powerdown = intel_serdes_powerdown;
266	444	return ehl_pse1_common_data(pdev, plat);
267	445	}
268	446
@@ -291,6 +469,8 @@ static int tgl_sgmii_data(struct pci_dev *pdev, · 22 unmodified lines
291	469	plat->bus_id = 1;
292	470	plat->phy_addr = 0;
293	471	plat->phy_interface = PHY_INTERFACE_MODE_SGMII;
	472	+ plat->serdes_powerup = intel_serdes_powerup;
	473	+ plat->serdes_powerdown = intel_serdes_powerdown;
294	474	return tgl_common_data(pdev, plat);
295	475	}
296	476
@@ -417,11 +597,17 @@ static int intel_eth_pci_probe(struct pci_dev *pdev, · 120 unmodified lines
417	597	const struct pci_device_id *id)
418	598	{
419	599	struct stmmac_pci_info info = (struct stmmac_pci_info )id->driver_data;
	600	+ struct intel_priv_data *intel_priv;
420	601	struct plat_stmmacenet_data *plat;
421	602	struct stmmac_resources res;
422	603	int i;
423	604	int ret;
424	605
	606	+ intel_priv = devm_kzalloc(&pdev->dev, sizeof(*intel_priv),
	607	+ GFP_KERNEL);
	608	+ if (!intel_priv)
	609	+ return -ENOMEM;
	610	+
425	611	plat = devm_kzalloc(&pdev->dev, sizeof(*plat), GFP_KERNEL);
426	612	if (!plat)
427	613	return -ENOMEM;
@@ -457,6 +643,9 @@ static int intel_eth_pci_probe(struct pci_dev *pdev, · 29 unmodified lines
457	643
458	644	pci_set_master(pdev);
459	645
	646	+ plat->bsp_priv = intel_priv;
	647	+ intel_priv->mdio_adhoc_addr = 0x15;
	648	+
460	649	ret = info->setup(pdev, plat);
461	650	if (ret)
462	651	return ret;

drivers/net/ethernet/stmicro/stmmac/dwmac-intel.h+23−0

@@ -0,0 +1,23 @@
	1	+/* SPDX-License-Identifier: GPL-2.0 */
	2	+/* Copyright (c) 2020, Intel Corporation
	3	+ * DWMAC Intel header file
	4	+ */
	5	+
	6	+#ifndef __DWMAC_INTEL_H__
	7	+#define __DWMAC_INTEL_H__
	8	+
	9	+#define POLL_DELAY_US 8
	10	+
	11	+/* SERDES Register */
	12	+#define SERDES_GSR0 0x5 /* Global Status Reg0 */
	13	+#define SERDES_GCR0 0xb /* Global Configuration Reg0 */
	14	+
	15	+/* SERDES defines */
	16	+#define SERDES_PLL_CLK BIT(0) /* PLL clk valid signal */
	17	+#define SERDES_RST BIT(2) /* Serdes Reset */
	18	+#define SERDES_PWR_ST_MASK GENMASK(6, 4) /* Serdes Power state*/
	19	+#define SERDES_PWR_ST_SHIFT 4
	20	+#define SERDES_PWR_ST_P0 0x0
	21	+#define SERDES_PWR_ST_P3 0x3
	22	+
	23	+#endif /* __DWMAC_INTEL_H__ */

drivers/net/ethernet/stmicro/stmmac/dwmac-meson8b.c+1−0

@@ -119,6 +119,7 @@ static int meson8b_init_rgmii_tx_clk(struct meson8b_dwmac *dwmac) · 118 unmodified lines
119	119	{ .div = 5, .val = 5, },
120	120	{ .div = 6, .val = 6, },
121	121	{ .div = 7, .val = 7, },
	122	+ { /* end of array */ }
122	123	};
123	124
124	125	clk_configs = devm_kzalloc(dev, sizeof(*clk_configs), GFP_KERNEL);

drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c+6−3

@@ -291,16 +291,19 @@ static int socfpga_gen5_set_phy_mode(struct socfpga_dwmac *dwmac) · 290 unmodified lines
291	291	phymode == PHY_INTERFACE_MODE_MII \|\|
292	292	phymode == PHY_INTERFACE_MODE_GMII \|\|
293	293	phymode == PHY_INTERFACE_MODE_SGMII) {
294		− ctrl \|= SYSMGR_EMACGRP_CTRL_PTP_REF_CLK_MASK << (reg_shift / 2);
295	294	regmap_read(sys_mgr_base_addr, SYSMGR_FPGAGRP_MODULE_REG,
296	295	&module);
297	296	module \|= (SYSMGR_FPGAGRP_MODULE_EMAC << (reg_shift / 2));
298	297	regmap_write(sys_mgr_base_addr, SYSMGR_FPGAGRP_MODULE_REG,
299	298	module);
300		− } else {
301		− ctrl &= ~(SYSMGR_EMACGRP_CTRL_PTP_REF_CLK_MASK << (reg_shift / 2));
302	299	}
303	300
	301	+ if (dwmac->f2h_ptp_ref_clk)
	302	+ ctrl \|= SYSMGR_EMACGRP_CTRL_PTP_REF_CLK_MASK << (reg_shift / 2);
	303	+ else
	304	+ ctrl &= ~(SYSMGR_EMACGRP_CTRL_PTP_REF_CLK_MASK <<
	305	+ (reg_shift / 2));
	306	+
304	307	regmap_write(sys_mgr_base_addr, reg_offset, ctrl);
305	308
306	309	/* Deassert reset for the phy configuration to be sampled by

drivers/net/ethernet/stmicro/stmmac/dwmac5.c+1−1

@@ -624,7 +624,7 @@ int dwmac5_est_configure(void __iomem ioaddr, struct stmmac_est cfg, · 623 unmodified lines
624	624	total_offset += offset;
625	625	}
626	626
627		− total_ctr = cfg->ctr[0] + cfg->ctr[1] * 1000000000;
	627	+ total_ctr = cfg->ctr[0] + cfg->ctr[1] * 1000000000ULL;
628	628	total_ctr += total_offset;
629	629
630	630	ctr_low = do_div(total_ctr, 1000000000);

drivers/net/ethernet/stmicro/stmmac/stmmac_hwtstamp.c+8−4

@@ -27,12 +27,16 @@ static void config_sub_second_increment(void __iomem *ioaddr, · 26 unmodified lines
27	27	unsigned long data;
28	28	u32 reg_value;
29	29
30		− /* For GMAC3.x, 4.x versions, convert the ptp_clock to nano second
31		− * formula = (1/ptp_clock) * 1000000000
32		− * where ptp_clock is 50MHz if fine method is used to update system
	30	+ /* For GMAC3.x, 4.x versions, in "fine adjustement mode" set sub-second
	31	+ * increment to twice the number of nanoseconds of a clock cycle.
	32	+ * The calculation of the default_addend value by the caller will set it
	33	+ * to mid-range = 2^31 when the remainder of this division is zero,
	34	+ * which will make the accumulator overflow once every 2 ptp_clock
	35	+ * cycles, adding twice the number of nanoseconds of a clock cycle :
	36	+ * 2000000000ULL / ptp_clock.
33	37	*/
34	38	if (value & PTP_TCR_TSCFUPDT)
35		− data = (1000000000ULL / 50000000);
	39	+ data = (2000000000ULL / ptp_clock);
36	40	else
37	41	data = (1000000000ULL / ptp_clock);
38	42

drivers/net/ethernet/stmicro/stmmac/stmmac_main.c+26−6

@@ -4060,7 +4060,7 @@ static int stmmac_set_features(struct net_device *netdev, · 4059 unmodified lines
4060	4060	/**
4061	4061	* stmmac_interrupt - main ISR
4062	4062	* @irq: interrupt number.
4063		− * @dev_id: to pass the net device pointer.
	4063	+ * @dev_id: to pass the net device pointer (must be valid).
4064	4064	* Description: this is the main driver interrupt service routine.
4065	4065	* It can call:
4066	4066	* o DMA service routine (to manage incoming frame reception and transmission
@@ -4084,11 +4084,6 @@ static irqreturn_t stmmac_interrupt(int irq, void *dev_id) · 17 unmodified lines
4084	4084	if (priv->irq_wake)
4085	4085	pm_wakeup_event(priv->device, 0);
4086	4086
4087		− if (unlikely(!dev)) {
4088		− netdev_err(priv->dev, "%s: invalid dev pointer\n", __func__);
4089		− return IRQ_NONE;
4090		− }
4091		−
4092	4087	/* Check if adapter is up */
4093	4088	if (test_bit(STMMAC_DOWN, &priv->state))
4094	4089	return IRQ_HANDLED;
@@ -4986,12 +4981,22 @@ int stmmac_dvr_probe(struct device *device, · 891 unmodified lines
4986	4981	goto error_netdev_register;
4987	4982	}
4988	4983
	4984	+ if (priv->plat->serdes_powerup) {
	4985	+ ret = priv->plat->serdes_powerup(ndev,
	4986	+ priv->plat->bsp_priv);
	4987	+
	4988	+ if (ret < 0)
	4989	+ goto error_serdes_powerup;
	4990	+ }
	4991	+
4989	4992	#ifdef CONFIG_DEBUG_FS
4990	4993	stmmac_init_fs(ndev);
4991	4994	#endif
4992	4995
4993	4996	return ret;
4994	4997
	4998	+error_serdes_powerup:
	4999	+ unregister_netdev(ndev);
4995	5000	error_netdev_register:
4996	5001	phylink_destroy(priv->phylink);
4997	5002	error_phy_setup:
@@ -5029,6 +5034,9 @@ int stmmac_dvr_remove(struct device *dev) · 31 unmodified lines
5029	5034
5030	5035	stmmac_stop_all_dma(priv);
5031	5036
	5037	+ if (priv->plat->serdes_powerdown)
	5038	+ priv->plat->serdes_powerdown(ndev, priv->plat->bsp_priv);
	5039	+
5032	5040	stmmac_mac_set(priv, priv->ioaddr, false);
5033	5041	netif_carrier_off(ndev);
5034	5042	unregister_netdev(ndev);
@@ -5081,6 +5089,9 @@ int stmmac_suspend(struct device *dev) · 46 unmodified lines
5081	5089	/* Stop TX/RX DMA */
5082	5090	stmmac_stop_all_dma(priv);
5083	5091
	5092	+ if (priv->plat->serdes_powerdown)
	5093	+ priv->plat->serdes_powerdown(ndev, priv->plat->bsp_priv);
	5094	+
5084	5095	/* Enable Power down mode by programming the PMT regs */
5085	5096	if (device_may_wakeup(priv->device)) {
5086	5097	stmmac_pmt(priv, priv->hw, priv->wolopts);
@@ -5143,6 +5154,7 @@ int stmmac_resume(struct device *dev) · 56 unmodified lines
5143	5154	{
5144	5155	struct net_device *ndev = dev_get_drvdata(dev);
5145	5156	struct stmmac_priv *priv = netdev_priv(ndev);
	5157	+ int ret;
5146	5158
5147	5159	if (!netif_running(ndev))
5148	5160	return 0;
@@ -5170,6 +5182,14 @@ int stmmac_resume(struct device *dev) · 21 unmodified lines
5170	5182	stmmac_mdio_reset(priv->mii);
5171	5183	}
5172	5184
	5185	+ if (priv->plat->serdes_powerup) {
	5186	+ ret = priv->plat->serdes_powerup(ndev,
	5187	+ priv->plat->bsp_priv);
	5188	+
	5189	+ if (ret < 0)
	5190	+ return ret;
	5191	+ }
	5192	+
5173	5193	netif_device_attach(ndev);
5174	5194
5175	5195	mutex_lock(&priv->lock);

drivers/net/ethernet/ti/Kconfig+1−2

@@ -90,9 +90,8 @@ config TI_CPTS · 89 unmodified lines
90	90	config TI_CPTS_MOD
91	91	tristate
92	92	depends on TI_CPTS
	93	+ depends on PTP_1588_CLOCK
93	94	default y if TI_CPSW=y \|\| TI_KEYSTONE_NETCP=y \|\| TI_CPSW_SWITCHDEV=y
94		− select NET_PTP_CLASSIFY
95		− imply PTP_1588_CLOCK
96	95	default m
97	96
98	97	config TI_K3_AM65_CPSW_NUSS

drivers/net/ethernet/ti/am65-cpsw-nuss.c+3−2

@@ -1719,7 +1719,8 @@ static int am65_cpsw_nuss_ndev_add_napi_2g(struct am65_cpsw_common *common) · 1718 unmodified lines
1719	1719
1720	1720	ret = devm_request_irq(dev, tx_chn->irq,
1721	1721	am65_cpsw_nuss_tx_irq,
1722		− 0, tx_chn->tx_chn_name, tx_chn);
	1722	+ IRQF_TRIGGER_HIGH,
	1723	+ tx_chn->tx_chn_name, tx_chn);
1723	1724	if (ret) {
1724	1725	dev_err(dev, "failure requesting tx%u irq %u, %d\n",
1725	1726	tx_chn->id, tx_chn->irq, ret);
@@ -1744,7 +1745,7 @@ static int am65_cpsw_nuss_ndev_reg_2g(struct am65_cpsw_common *common) · 18 unmodified lines
1744	1745
1745	1746	ret = devm_request_irq(dev, common->rx_chns.irq,
1746	1747	am65_cpsw_nuss_rx_irq,
1747		− 0, dev_name(dev), common);
	1748	+ IRQF_TRIGGER_HIGH, dev_name(dev), common);
1748	1749	if (ret) {
1749	1750	dev_err(dev, "failure requesting rx irq %u, %d\n",
1750	1751	common->rx_chns.irq, ret);

drivers/net/ethernet/toshiba/tc35815.c+1−1

@@ -643,7 +643,7 @@ static int tc_mii_probe(struct net_device *dev) · 642 unmodified lines
643	643	linkmode_set_bit(ETHTOOL_LINK_MODE_10baseT_Half_BIT, mask);
644	644	linkmode_set_bit(ETHTOOL_LINK_MODE_100baseT_Half_BIT, mask);
645	645	}
646		− linkmode_and(phydev->supported, phydev->supported, mask);
	646	+ linkmode_andnot(phydev->supported, phydev->supported, mask);
647	647	linkmode_copy(phydev->advertising, phydev->supported);
648	648
649	649	lp->link = 0;

drivers/net/ethernet/xscale/ixp4xx_eth.c+2−0

@@ -1387,6 +1387,8 @@ static int ixp4xx_eth_probe(struct platform_device *pdev) · 1386 unmodified lines
1387	1387	return -ENODEV;
1388	1388	regs_phys = res->start;
1389	1389	port->regs = devm_ioremap_resource(dev, res);
	1390	+ if (IS_ERR(port->regs))
	1391	+ return PTR_ERR(port->regs);
1390	1392
1391	1393	switch (port->id) {
1392	1394	case IXP4XX_ETH_NPEA:

drivers/net/geneve.c+1−1

@@ -1207,7 +1207,7 @@ static int geneve_validate(struct nlattr tb[], struct nlattr data[], · 1206 unmodified lines
1207	1207	enum ifla_geneve_df df = nla_get_u8(data[IFLA_GENEVE_DF]);
1208	1208
1209	1209	if (df < 0 \|\| df > GENEVE_DF_MAX) {
1210		− NL_SET_ERR_MSG_ATTR(extack, tb[IFLA_GENEVE_DF],
	1210	+ NL_SET_ERR_MSG_ATTR(extack, data[IFLA_GENEVE_DF],
1211	1211	"Invalid DF attribute");
1212	1212	return -EINVAL;
1213	1213	}

drivers/net/gtp.c+5−4

@@ -1169,11 +1169,11 @@ out_unlock: · 1168 unmodified lines
1169	1169	static struct genl_family gtp_genl_family;
1170	1170
1171	1171	static int gtp_genl_fill_info(struct sk_buff *skb, u32 snd_portid, u32 snd_seq,
1172		− u32 type, struct pdp_ctx *pctx)
	1172	+ int flags, u32 type, struct pdp_ctx *pctx)
1173	1173	{
1174	1174	void *genlh;
1175	1175
1176		− genlh = genlmsg_put(skb, snd_portid, snd_seq, &gtp_genl_family, 0,
	1176	+ genlh = genlmsg_put(skb, snd_portid, snd_seq, &gtp_genl_family, flags,
1177	1177	type);
1178	1178	if (genlh == NULL)
1179	1179	goto nlmsg_failure;
@@ -1227,8 +1227,8 @@ static int gtp_genl_get_pdp(struct sk_buff skb, struct genl_info info) · 47 unmodified lines
1227	1227	goto err_unlock;
1228	1228	}
1229	1229
1230		− err = gtp_genl_fill_info(skb2, NETLINK_CB(skb).portid,
1231		− info->snd_seq, info->nlhdr->nlmsg_type, pctx);
	1230	+ err = gtp_genl_fill_info(skb2, NETLINK_CB(skb).portid, info->snd_seq,
	1231	+ 0, info->nlhdr->nlmsg_type, pctx);
1232	1232	if (err < 0)
1233	1233	goto err_unlock_free;
1234	1234
@@ -1271,6 +1271,7 @@ static int gtp_genl_dump_pdp(struct sk_buff *skb, · 36 unmodified lines
1271	1271	gtp_genl_fill_info(skb,
1272	1272	NETLINK_CB(cb->skb).portid,
1273	1273	cb->nlh->nlmsg_seq,
	1274	+ NLM_F_MULTI,
1274	1275	cb->nlh->nlmsg_type, pctx)) {
1275	1276	cb->args[0] = i;
1276	1277	cb->args[1] = j;

drivers/net/hyperv/netvsc_drv.c+2−1

@@ -707,7 +707,8 @@ no_memory: · 706 unmodified lines
707	707	goto drop;
708	708	}
709	709
710		−static int netvsc_start_xmit(struct sk_buff skb, struct net_device ndev)
	710	+static netdev_tx_t netvsc_start_xmit(struct sk_buff *skb,
	711	+ struct net_device *ndev)
711	712	{
712	713	return netvsc_xmit(skb, ndev, false);
713	714	}

drivers/net/ipa/gsi.c+9−2

@@ -1041,6 +1041,7 @@ static void gsi_isr_gp_int1(struct gsi *gsi) · 1040 unmodified lines
1041	1041
1042	1042	complete(&gsi->completion);
1043	1043	}
	1044	+
1044	1045	/* Inter-EE interrupt handler */
1045	1046	static void gsi_isr_glob_ee(struct gsi *gsi)
1046	1047	{
@@ -1493,6 +1494,12 @@ static int gsi_generic_command(struct gsi *gsi, u32 channel_id, · 446 unmodified lines
1493	1494	struct completion *completion = &gsi->completion;
1494	1495	u32 val;
1495	1496
	1497	+ /* First zero the result code field */
	1498	+ val = ioread32(gsi->virt + GSI_CNTXT_SCRATCH_0_OFFSET);
	1499	+ val &= ~GENERIC_EE_RESULT_FMASK;
	1500	+ iowrite32(val, gsi->virt + GSI_CNTXT_SCRATCH_0_OFFSET);
	1501	+
	1502	+ /* Now issue the command */
1496	1503	val = u32_encode_bits(opcode, GENERIC_OPCODE_FMASK);
1497	1504	val \|= u32_encode_bits(channel_id, GENERIC_CHID_FMASK);
1498	1505	val \|= u32_encode_bits(GSI_EE_MODEM, GENERIC_EE_FMASK);
@@ -1798,9 +1805,9 @@ static int gsi_channel_init_one(struct gsi *gsi, · 299 unmodified lines
1798	1805
1799	1806	/* Worst case we need an event for every outstanding TRE */
1800	1807	if (data->channel.tre_count > data->channel.event_count) {
1801		− dev_warn(gsi->dev, "channel %u limited to %u TREs\n",
1802		− data->channel_id, data->channel.tre_count);
1803	1808	tre_count = data->channel.event_count;
	1809	+ dev_warn(gsi->dev, "channel %u limited to %u TREs\n",
	1810	+ data->channel_id, tre_count);
1804	1811	} else {
1805	1812	tre_count = data->channel.tre_count;
1806	1813	}

drivers/net/ipa/gsi_reg.h+2−0

@@ -410,6 +410,8 @@ · 409 unmodified lines
410	410	#define INTER_EE_RESULT_FMASK GENMASK(2, 0)
411	411	#define GENERIC_EE_RESULT_FMASK GENMASK(7, 5)
412	412	#define GENERIC_EE_SUCCESS_FVAL 1
	413	+#define GENERIC_EE_INCORRECT_DIRECTION_FVAL 3
	414	+#define GENERIC_EE_INCORRECT_CHANNEL_FVAL 5
413	415	#define GENERIC_EE_NO_RESOURCES_FVAL 7
414	416	#define USB_MAX_PACKET_FMASK GENMASK(15, 15) /* 0: HS; 1: SS */
415	417	#define MHI_BASE_CHANNEL_FMASK GENMASK(31, 24)

drivers/net/ipa/ipa_endpoint.c+2−5

@@ -1283,7 +1283,7 @@ static int ipa_endpoint_stop_rx_dma(struct ipa *ipa) · 1282 unmodified lines
1283	1283	*/
1284	1284	int ipa_endpoint_stop(struct ipa_endpoint *endpoint)
1285	1285	{
1286		− u32 retries = endpoint->toward_ipa ? 0 : IPA_ENDPOINT_STOP_RX_RETRIES;
	1286	+ u32 retries = IPA_ENDPOINT_STOP_RX_RETRIES;
1287	1287	int ret;
1288	1288
1289	1289	do {
@@ -1291,12 +1291,9 @@ int ipa_endpoint_stop(struct ipa_endpoint *endpoint) · 1 unmodified line
1291	1291	struct gsi *gsi = &ipa->gsi;
1292	1292
1293	1293	ret = gsi_channel_stop(gsi, endpoint->channel_id);
1294		− if (ret != -EAGAIN)
	1294	+ if (ret != -EAGAIN \|\| endpoint->toward_ipa)
1295	1295	break;
1296	1296
1297		− if (endpoint->toward_ipa)
1298		− continue;
1299		−
1300	1297	/* For IPA v3.5.1, send a DMA read task and check again */
1301	1298	if (ipa->version == IPA_VERSION_3_5_1) {
1302	1299	ret = ipa_endpoint_stop_rx_dma(ipa);

drivers/net/macsec.c+12−6

@@ -1305,7 +1305,8 @@ static struct crypto_aead macsec_alloc_tfm(char key, int key_len, int icv_len) · 1304 unmodified lines
1305	1305	struct crypto_aead *tfm;
1306	1306	int ret;
1307	1307
1308		− tfm = crypto_alloc_aead("gcm(aes)", 0, 0);
	1308	+ /* Pick a sync gcm(aes) cipher to ensure order is preserved. */
	1309	+ tfm = crypto_alloc_aead("gcm(aes)", 0, CRYPTO_ALG_ASYNC);
1309	1310
1310	1311	if (IS_ERR(tfm))
1311	1312	return tfm;
@@ -2640,11 +2641,12 @@ static int macsec_upd_offload(struct sk_buff skb, struct genl_info info) · 1328 unmodified lines
2640	2641	if (ret)
2641	2642	goto rollback;
2642	2643
2643		− rtnl_unlock();
2644	2644	/* Force features update, since they are different for SW MACSec and
2645	2645	* HW offloading cases.
2646	2646	*/
2647	2647	netdev_update_features(dev);
	2648	+
	2649	+ rtnl_unlock();
2648	2650	return 0;
2649	2651
2650	2652	rollback:
@@ -4002,11 +4004,11 @@ static int macsec_newlink(struct net net, struct net_device dev, · 1351 unmodified lines
4002	4004	struct netlink_ext_ack *extack)
4003	4005	{
4004	4006	struct macsec_dev *macsec = macsec_priv(dev);
	4007	+ rx_handler_func_t *rx_handler;
	4008	+ u8 icv_len = DEFAULT_ICV_LEN;
4005	4009	struct net_device *real_dev;
4006		− int err;
	4010	+ int err, mtu;
4007	4011	sci_t sci;
4008		− u8 icv_len = DEFAULT_ICV_LEN;
4009		− rx_handler_func_t *rx_handler;
4010	4012
4011	4013	if (!tb[IFLA_LINK])
4012	4014	return -EINVAL;
@@ -4033,7 +4035,11 @@ static int macsec_newlink(struct net net, struct net_device dev, · 20 unmodified lines
4033	4035
4034	4036	if (data && data[IFLA_MACSEC_ICV_LEN])
4035	4037	icv_len = nla_get_u8(data[IFLA_MACSEC_ICV_LEN]);
4036		− dev->mtu = real_dev->mtu - icv_len - macsec_extra_len(true);
	4038	+ mtu = real_dev->mtu - icv_len - macsec_extra_len(true);
	4039	+ if (mtu < 0)
	4040	+ dev->mtu = 0;
	4041	+ else
	4042	+ dev->mtu = mtu;
4037	4043
4038	4044	rx_handler = rtnl_dereference(real_dev->rx_handler);
4039	4045	if (rx_handler && rx_handler != macsec_handle_frame)

drivers/net/macvlan.c+1−1

@@ -1704,7 +1704,7 @@ static int macvlan_device_event(struct notifier_block *unused, · 1703 unmodified lines
1704	1704	struct macvlan_dev,
1705	1705	list);
1706	1706
1707		− if (macvlan_sync_address(vlan->dev, dev->dev_addr))
	1707	+ if (vlan && macvlan_sync_address(vlan->dev, dev->dev_addr))
1708	1708	return NOTIFY_BAD;
1709	1709
1710	1710	break;

drivers/net/phy/bcm84881.c+3−3

@@ -155,9 +155,6 @@ static int bcm84881_read_status(struct phy_device *phydev) · 154 unmodified lines
155	155	if (phydev->autoneg == AUTONEG_ENABLE && !phydev->autoneg_complete)
156	156	phydev->link = false;
157	157
158		− if (!phydev->link)
159		− return 0;
160		−
161	158	linkmode_zero(phydev->lp_advertising);
162	159	phydev->speed = SPEED_UNKNOWN;
163	160	phydev->duplex = DUPLEX_UNKNOWN;
@@ -165,6 +162,9 @@ static int bcm84881_read_status(struct phy_device *phydev) · 1 unmodified line
165	162	phydev->asym_pause = 0;
166	163	phydev->mdix = 0;
167	164
	165	+ if (!phydev->link)
	166	+ return 0;
	167	+
168	168	if (phydev->autoneg_complete) {
169	169	val = genphy_c45_read_lpa(phydev);
170	170	if (val < 0)

drivers/net/phy/dp83640.c+1−1

@@ -1120,7 +1120,7 @@ static struct dp83640_clock dp83640_clock_get_bus(struct mii_bus bus) · 1119 unmodified lines
1120	1120	goto out;
1121	1121	}
1122	1122	dp83640_clock_init(clock, bus);
1123		− list_add_tail(&phyter_clocks, &clock->list);
	1123	+ list_add_tail(&clock->list, &phyter_clocks);
1124	1124	out:
1125	1125	mutex_unlock(&phyter_clocks_lock);
1126	1126

drivers/net/phy/dp83822.c+14−16

@@ -137,19 +137,18 @@ static int dp83822_set_wol(struct phy_device *phydev, · 136 unmodified lines
137	137	value &= ~DP83822_WOL_SECURE_ON;
138	138	}
139	139
140		− value \|= (DP83822_WOL_EN \| DP83822_WOL_INDICATION_SEL \|
141		− DP83822_WOL_CLR_INDICATION);
142		− phy_write_mmd(phydev, DP83822_DEVADDR, MII_DP83822_WOL_CFG,
143		− value);
	140	+ /* Clear any pending WoL interrupt */
	141	+ phy_read(phydev, MII_DP83822_MISR2);
	142	+
	143	+ value \|= DP83822_WOL_EN \| DP83822_WOL_INDICATION_SEL \|
	144	+ DP83822_WOL_CLR_INDICATION;
	145	+
	146	+ return phy_write_mmd(phydev, DP83822_DEVADDR,
	147	+ MII_DP83822_WOL_CFG, value);
144	148	} else {
145		− value = phy_read_mmd(phydev, DP83822_DEVADDR,
146		− MII_DP83822_WOL_CFG);
147		− value &= ~DP83822_WOL_EN;
148		− phy_write_mmd(phydev, DP83822_DEVADDR, MII_DP83822_WOL_CFG,
149		− value);
	149	+ return phy_clear_bits_mmd(phydev, DP83822_DEVADDR,
	150	+ MII_DP83822_WOL_CFG, DP83822_WOL_EN);
150	151	}
151		−
152		− return 0;
153	152	}
154	153
155	154	static void dp83822_get_wol(struct phy_device *phydev,
@@ -258,12 +257,11 @@ static int dp83822_config_intr(struct phy_device *phydev) · 102 unmodified lines
258	257
259	258	static int dp83822_config_init(struct phy_device *phydev)
260	259	{
261		− int value;
262		−
263		− value = DP83822_WOL_MAGIC_EN \| DP83822_WOL_SECURE_ON \| DP83822_WOL_EN;
	260	+ int value = DP83822_WOL_EN \| DP83822_WOL_MAGIC_EN \|
	261	+ DP83822_WOL_SECURE_ON;
264	262
265		− return phy_write_mmd(phydev, DP83822_DEVADDR, MII_DP83822_WOL_CFG,
266		− value);
	263	+ return phy_clear_bits_mmd(phydev, DP83822_DEVADDR,
	264	+ MII_DP83822_WOL_CFG, value);
267	265	}
268	266
269	267	static int dp83822_phy_reset(struct phy_device *phydev)

drivers/net/phy/dp83tc811.c+12−9

@@ -139,16 +139,19 @@ static int dp83811_set_wol(struct phy_device *phydev, · 138 unmodified lines
139	139	value &= ~DP83811_WOL_SECURE_ON;
140	140	}
141	141
142		− value \|= (DP83811_WOL_EN \| DP83811_WOL_INDICATION_SEL \|
143		− DP83811_WOL_CLR_INDICATION);
144		− phy_write_mmd(phydev, DP83811_DEVADDR, MII_DP83811_WOL_CFG,
145		− value);
	142	+ /* Clear any pending WoL interrupt */
	143	+ phy_read(phydev, MII_DP83811_INT_STAT1);
	144	+
	145	+ value \|= DP83811_WOL_EN \| DP83811_WOL_INDICATION_SEL \|
	146	+ DP83811_WOL_CLR_INDICATION;
	147	+
	148	+ return phy_write_mmd(phydev, DP83811_DEVADDR,
	149	+ MII_DP83811_WOL_CFG, value);
146	150	} else {
147		− phy_clear_bits_mmd(phydev, DP83811_DEVADDR, MII_DP83811_WOL_CFG,
148		− DP83811_WOL_EN);
	151	+ return phy_clear_bits_mmd(phydev, DP83811_DEVADDR,
	152	+ MII_DP83811_WOL_CFG, DP83811_WOL_EN);
149	153	}
150	154
151		− return 0;
152	155	}
153	156
154	157	static void dp83811_get_wol(struct phy_device *phydev,
@@ -292,8 +295,8 @@ static int dp83811_config_init(struct phy_device *phydev) · 137 unmodified lines
292	295
293	296	value = DP83811_WOL_MAGIC_EN \| DP83811_WOL_SECURE_ON \| DP83811_WOL_EN;
294	297
295		− return phy_write_mmd(phydev, DP83811_DEVADDR, MII_DP83811_WOL_CFG,
296		− value);
	298	+ return phy_clear_bits_mmd(phydev, DP83811_DEVADDR, MII_DP83811_WOL_CFG,
	299	+ value);
297	300	}
298	301
299	302	static int dp83811_phy_reset(struct phy_device *phydev)

drivers/net/phy/marvell10g.c+28−2

@@ -66,6 +66,9 @@ enum { · 65 unmodified lines
66	66	MV_PCS_CSSR1_SPD2_2500 = 0x0004,
67	67	MV_PCS_CSSR1_SPD2_10000 = 0x0000,
68	68
	69	+ /* Temperature read register (88E2110 only) */
	70	+ MV_PCS_TEMP = 0x8042,
	71	+
69	72	/* These registers appear at 0x800X and 0xa00X - the 0xa00X control
70	73	* registers appear to set themselves to the 0x800X when AN is
71	74	* restarted, but status registers appear readable from either.
@@ -77,6 +80,7 @@ enum { · 5 unmodified lines
77	80	MV_V2_PORT_CTRL = 0xf001,
78	81	MV_V2_PORT_CTRL_SWRST = BIT(15),
79	82	MV_V2_PORT_CTRL_PWRDOWN = BIT(11),
	83	+ /* Temperature control/read registers (88X3310 only) */
80	84	MV_V2_TEMP_CTRL = 0xf08a,
81	85	MV_V2_TEMP_CTRL_MASK = 0xc000,
82	86	MV_V2_TEMP_CTRL_SAMPLE = 0x0000,
@@ -104,6 +108,24 @@ static umode_t mv3310_hwmon_is_visible(const void *data, · 21 unmodified lines
104	108	return 0;
105	109	}
106	110
	111	+static int mv3310_hwmon_read_temp_reg(struct phy_device *phydev)
	112	+{
	113	+ return phy_read_mmd(phydev, MDIO_MMD_VEND2, MV_V2_TEMP);
	114	+}
	115	+
	116	+static int mv2110_hwmon_read_temp_reg(struct phy_device *phydev)
	117	+{
	118	+ return phy_read_mmd(phydev, MDIO_MMD_PCS, MV_PCS_TEMP);
	119	+}
	120	+
	121	+static int mv10g_hwmon_read_temp_reg(struct phy_device *phydev)
	122	+{
	123	+ if (phydev->drv->phy_id == MARVELL_PHY_ID_88X3310)
	124	+ return mv3310_hwmon_read_temp_reg(phydev);
	125	+ else /* MARVELL_PHY_ID_88E2110 */
	126	+ return mv2110_hwmon_read_temp_reg(phydev);
	127	+}
	128	+
107	129	static int mv3310_hwmon_read(struct device *dev, enum hwmon_sensor_types type,
108	130	u32 attr, int channel, long *value)
109	131	{
@@ -116,7 +138,7 @@ static int mv3310_hwmon_read(struct device *dev, enum hwmon_sensor_types type, · 6 unmodified lines
116	138	}
117	139
118	140	if (type == hwmon_temp && attr == hwmon_temp_input) {
119		− temp = phy_read_mmd(phydev, MDIO_MMD_VEND2, MV_V2_TEMP);
	141	+ temp = mv10g_hwmon_read_temp_reg(phydev);
120	142	if (temp < 0)
121	143	return temp;
122	144
@@ -169,6 +191,9 @@ static int mv3310_hwmon_config(struct phy_device *phydev, bool enable) · 46 unmodified lines
169	191	u16 val;
170	192	int ret;
171	193
	194	+ if (phydev->drv->phy_id != MARVELL_PHY_ID_88X3310)
	195	+ return 0;
	196	+
172	197	ret = phy_write_mmd(phydev, MDIO_MMD_VEND2, MV_V2_TEMP,
173	198	MV_V2_TEMP_UNKNOWN);
174	199	if (ret < 0)
@@ -246,7 +271,8 @@ static int mv3310_power_up(struct phy_device *phydev) · 71 unmodified lines
246	271	ret = phy_clear_bits_mmd(phydev, MDIO_MMD_VEND2, MV_V2_PORT_CTRL,
247	272	MV_V2_PORT_CTRL_PWRDOWN);
248	273
249		− if (priv->firmware_ver < 0x00030000)
	274	+ if (phydev->drv->phy_id != MARVELL_PHY_ID_88X3310 \|\|
	275	+ priv->firmware_ver < 0x00030000)
250	276	return ret;
251	277
252	278	return phy_set_bits_mmd(phydev, MDIO_MMD_VEND2, MV_V2_PORT_CTRL,

drivers/net/phy/microchip_t1.c+171−0

@@ -3,9 +3,21 @@ · 2 unmodified lines
3	3
4	4	#include <linux/kernel.h>
5	5	#include <linux/module.h>
	6	+#include <linux/delay.h>
6	7	#include <linux/mii.h>
7	8	#include <linux/phy.h>
8	9
	10	+/* External Register Control Register */
	11	+#define LAN87XX_EXT_REG_CTL (0x14)
	12	+#define LAN87XX_EXT_REG_CTL_RD_CTL (0x1000)
	13	+#define LAN87XX_EXT_REG_CTL_WR_CTL (0x0800)
	14	+
	15	+/* External Register Read Data Register */
	16	+#define LAN87XX_EXT_REG_RD_DATA (0x15)
	17	+
	18	+/* External Register Write Data Register */
	19	+#define LAN87XX_EXT_REG_WR_DATA (0x16)
	20	+
9	21	/* Interrupt Source Register */
10	22	#define LAN87XX_INTERRUPT_SOURCE (0x18)
11	23
@@ -14,9 +26,160 @@ · 2 unmodified lines
14	26	#define LAN87XX_MASK_LINK_UP (0x0004)
15	27	#define LAN87XX_MASK_LINK_DOWN (0x0002)
16	28
	29	+/* phyaccess nested types */
	30	+#define PHYACC_ATTR_MODE_READ 0
	31	+#define PHYACC_ATTR_MODE_WRITE 1
	32	+#define PHYACC_ATTR_MODE_MODIFY 2
	33	+
	34	+#define PHYACC_ATTR_BANK_SMI 0
	35	+#define PHYACC_ATTR_BANK_MISC 1
	36	+#define PHYACC_ATTR_BANK_PCS 2
	37	+#define PHYACC_ATTR_BANK_AFE 3
	38	+#define PHYACC_ATTR_BANK_MAX 7
	39	+
17	40	#define DRIVER_AUTHOR "Nisar Sayed <nisar.sayed@microchip.com>"
18	41	#define DRIVER_DESC "Microchip LAN87XX T1 PHY driver"
19	42
	43	+struct access_ereg_val {
	44	+ u8 mode;
	45	+ u8 bank;
	46	+ u8 offset;
	47	+ u16 val;
	48	+ u16 mask;
	49	+};
	50	+
	51	+static int access_ereg(struct phy_device *phydev, u8 mode, u8 bank,
	52	+ u8 offset, u16 val)
	53	+{
	54	+ u16 ereg = 0;
	55	+ int rc = 0;
	56	+
	57	+ if (mode > PHYACC_ATTR_MODE_WRITE \|\| bank > PHYACC_ATTR_BANK_MAX)
	58	+ return -EINVAL;
	59	+
	60	+ if (bank == PHYACC_ATTR_BANK_SMI) {
	61	+ if (mode == PHYACC_ATTR_MODE_WRITE)
	62	+ rc = phy_write(phydev, offset, val);
	63	+ else
	64	+ rc = phy_read(phydev, offset);
	65	+ return rc;
	66	+ }
	67	+
	68	+ if (mode == PHYACC_ATTR_MODE_WRITE) {
	69	+ ereg = LAN87XX_EXT_REG_CTL_WR_CTL;
	70	+ rc = phy_write(phydev, LAN87XX_EXT_REG_WR_DATA, val);
	71	+ if (rc < 0)
	72	+ return rc;
	73	+ } else {
	74	+ ereg = LAN87XX_EXT_REG_CTL_RD_CTL;
	75	+ }
	76	+
	77	+ ereg \|= (bank << 8) \| offset;
	78	+
	79	+ rc = phy_write(phydev, LAN87XX_EXT_REG_CTL, ereg);
	80	+ if (rc < 0)
	81	+ return rc;
	82	+
	83	+ if (mode == PHYACC_ATTR_MODE_READ)
	84	+ rc = phy_read(phydev, LAN87XX_EXT_REG_RD_DATA);
	85	+
	86	+ return rc;
	87	+}
	88	+
	89	+static int access_ereg_modify_changed(struct phy_device *phydev,
	90	+ u8 bank, u8 offset, u16 val, u16 mask)
	91	+{
	92	+ int new = 0, rc = 0;
	93	+
	94	+ if (bank > PHYACC_ATTR_BANK_MAX)
	95	+ return -EINVAL;
	96	+
	97	+ rc = access_ereg(phydev, PHYACC_ATTR_MODE_READ, bank, offset, val);
	98	+ if (rc < 0)
	99	+ return rc;
	100	+
	101	+ new = val \| (rc & (mask ^ 0xFFFF));
	102	+ rc = access_ereg(phydev, PHYACC_ATTR_MODE_WRITE, bank, offset, new);
	103	+
	104	+ return rc;
	105	+}
	106	+
	107	+static int lan87xx_phy_init(struct phy_device *phydev)
	108	+{
	109	+ static const struct access_ereg_val init[] = {
	110	+ /* TX Amplitude = 5 */
	111	+ {PHYACC_ATTR_MODE_MODIFY, PHYACC_ATTR_BANK_AFE, 0x0B,
	112	+ 0x000A, 0x001E},
	113	+ /* Clear SMI interrupts */
	114	+ {PHYACC_ATTR_MODE_READ, PHYACC_ATTR_BANK_SMI, 0x18,
	115	+ 0, 0},
	116	+ /* Clear MISC interrupts */
	117	+ {PHYACC_ATTR_MODE_READ, PHYACC_ATTR_BANK_MISC, 0x08,
	118	+ 0, 0},
	119	+ /* Turn on TC10 Ring Oscillator (ROSC) */
	120	+ {PHYACC_ATTR_MODE_MODIFY, PHYACC_ATTR_BANK_MISC, 0x20,
	121	+ 0x0020, 0x0020},
	122	+ /* WUR Detect Length to 1.2uS, LPC Detect Length to 1.09uS */
	123	+ {PHYACC_ATTR_MODE_WRITE, PHYACC_ATTR_BANK_PCS, 0x20,
	124	+ 0x283C, 0},
	125	+ /* Wake_In Debounce Length to 39uS, Wake_Out Length to 79uS */
	126	+ {PHYACC_ATTR_MODE_WRITE, PHYACC_ATTR_BANK_MISC, 0x21,
	127	+ 0x274F, 0},
	128	+ /* Enable Auto Wake Forward to Wake_Out, ROSC on, Sleep,
	129	+ * and Wake_In to wake PHY
	130	+ */
	131	+ {PHYACC_ATTR_MODE_WRITE, PHYACC_ATTR_BANK_MISC, 0x20,
	132	+ 0x80A7, 0},
	133	+ /* Enable WUP Auto Fwd, Enable Wake on MDI, Wakeup Debouncer
	134	+ * to 128 uS
	135	+ */
	136	+ {PHYACC_ATTR_MODE_WRITE, PHYACC_ATTR_BANK_MISC, 0x24,
	137	+ 0xF110, 0},
	138	+ /* Enable HW Init */
	139	+ {PHYACC_ATTR_MODE_MODIFY, PHYACC_ATTR_BANK_SMI, 0x1A,
	140	+ 0x0100, 0x0100},
	141	+ };
	142	+ int rc, i;
	143	+
	144	+ /* Start manual initialization procedures in Managed Mode */
	145	+ rc = access_ereg_modify_changed(phydev, PHYACC_ATTR_BANK_SMI,
	146	+ 0x1a, 0x0000, 0x0100);
	147	+ if (rc < 0)
	148	+ return rc;
	149	+
	150	+ /* Soft Reset the SMI block */
	151	+ rc = access_ereg_modify_changed(phydev, PHYACC_ATTR_BANK_SMI,
	152	+ 0x00, 0x8000, 0x8000);
	153	+ if (rc < 0)
	154	+ return rc;
	155	+
	156	+ /* Check to see if the self-clearing bit is cleared */
	157	+ usleep_range(1000, 2000);
	158	+ rc = access_ereg(phydev, PHYACC_ATTR_MODE_READ,
	159	+ PHYACC_ATTR_BANK_SMI, 0x00, 0);
	160	+ if (rc < 0)
	161	+ return rc;
	162	+ if ((rc & 0x8000) != 0)
	163	+ return -ETIMEDOUT;
	164	+
	165	+ /* PHY Initialization */
	166	+ for (i = 0; i < ARRAY_SIZE(init); i++) {
	167	+ if (init[i].mode == PHYACC_ATTR_MODE_MODIFY) {
	168	+ rc = access_ereg_modify_changed(phydev, init[i].bank,
	169	+ init[i].offset,
	170	+ init[i].val,
	171	+ init[i].mask);
	172	+ } else {
	173	+ rc = access_ereg(phydev, init[i].mode, init[i].bank,
	174	+ init[i].offset, init[i].val);
	175	+ }
	176	+ if (rc < 0)
	177	+ return rc;
	178	+ }
	179	+
	180	+ return 0;
	181	+}
	182	+
20	183	static int lan87xx_phy_config_intr(struct phy_device *phydev)
21	184	{
22	185	int rc, val = 0;
@@ -40,6 +203,13 @@ static int lan87xx_phy_ack_interrupt(struct phy_device *phydev) · 17 unmodified lines
40	203	return rc < 0 ? rc : 0;
41	204	}
42	205
	206	+static int lan87xx_config_init(struct phy_device *phydev)
	207	+{
	208	+ int rc = lan87xx_phy_init(phydev);
	209	+
	210	+ return rc < 0 ? rc : 0;
	211	+}
	212	+
43	213	static struct phy_driver microchip_t1_phy_driver[] = {
44	214	{
45	215	.phy_id = 0x0007c150,
@@ -48,6 +218,7 @@ static struct phy_driver microchip_t1_phy_driver[] = { · 2 unmodified lines
48	218
49	219	.features = PHY_BASIC_T1_FEATURES,
50	220
	221	+ .config_init = lan87xx_config_init,
51	222	.config_aneg = genphy_config_aneg,
52	223
53	224	.ack_interrupt = lan87xx_phy_ack_interrupt,

drivers/net/team/team.c+4−0

@@ -468,6 +468,9 @@ static const struct team_mode team_mode_get(const char kind) · 467 unmodified lines
468	468	struct team_mode_item *mitem;
469	469	const struct team_mode *mode = NULL;
470	470
	471	+ if (!try_module_get(THIS_MODULE))
	472	+ return NULL;
	473	+
471	474	spin_lock(&mode_list_lock);
472	475	mitem = __find_mode(kind);
473	476	if (!mitem) {
@@ -483,6 +486,7 @@ static const struct team_mode team_mode_get(const char kind) · 9 unmodified lines
483	486	}
484	487
485	488	spin_unlock(&mode_list_lock);
	489	+ module_put(THIS_MODULE);
486	490	return mode;
487	491	}
488	492

drivers/net/usb/qmi_wwan.c+1−0

@@ -1359,6 +1359,7 @@ static const struct usb_device_id products[] = { · 1358 unmodified lines
1359	1359	{QMI_FIXED_INTF(0x413c, 0x81b3, 8)}, /* Dell Wireless 5809e Gobi(TM) 4G LTE Mobile Broadband Card (rev3) */
1360	1360	{QMI_FIXED_INTF(0x413c, 0x81b6, 8)}, /* Dell Wireless 5811e */
1361	1361	{QMI_FIXED_INTF(0x413c, 0x81b6, 10)}, /* Dell Wireless 5811e */
	1362	+ {QMI_FIXED_INTF(0x413c, 0x81cc, 8)}, /* Dell Wireless 5816e */
1362	1363	{QMI_FIXED_INTF(0x413c, 0x81d7, 0)}, /* Dell Wireless 5821e */
1363	1364	{QMI_FIXED_INTF(0x413c, 0x81d7, 1)}, /* Dell Wireless 5821e preproduction config */
1364	1365	{QMI_FIXED_INTF(0x413c, 0x81e0, 0)}, /* Dell Wireless 5821e with eSIM support*/

drivers/net/vrf.c+6−4

@@ -188,8 +188,8 @@ static netdev_tx_t vrf_process_v6_outbound(struct sk_buff *skb, · 187 unmodified lines
188	188	fl6.flowi6_proto = iph->nexthdr;
189	189	fl6.flowi6_flags = FLOWI_FLAG_SKIP_NH_OIF;
190	190
191		− dst = ip6_route_output(net, NULL, &fl6);
192		− if (dst == dst_null)
	191	+ dst = ip6_dst_lookup_flow(net, NULL, &fl6, NULL);
	192	+ if (IS_ERR(dst) \|\| dst == dst_null)
193	193	goto err;
194	194
195	195	skb_dst_drop(skb);
@@ -474,7 +474,8 @@ static struct sk_buff vrf_ip6_out(struct net_device vrf_dev, · 278 unmodified lines
474	474	if (rt6_need_strict(&ipv6_hdr(skb)->daddr))
475	475	return skb;
476	476
477		− if (qdisc_tx_is_default(vrf_dev))
	477	+ if (qdisc_tx_is_default(vrf_dev) \|\|
	478	+ IP6CB(skb)->flags & IP6SKB_XFRM_TRANSFORMED)
478	479	return vrf_ip6_out_direct(vrf_dev, sk, skb);
479	480
480	481	return vrf_ip6_out_redirect(vrf_dev, skb);
@@ -686,7 +687,8 @@ static struct sk_buff vrf_ip_out(struct net_device vrf_dev, · 205 unmodified lines
686	687	ipv4_is_lbcast(ip_hdr(skb)->daddr))
687	688	return skb;
688	689
689		− if (qdisc_tx_is_default(vrf_dev))
	690	+ if (qdisc_tx_is_default(vrf_dev) \|\|
	691	+ IPCB(skb)->flags & IPSKB_XFRM_TRANSFORMED)
690	692	return vrf_ip_out_direct(vrf_dev, sk, skb);
691	693
692	694	return vrf_ip_out_redirect(vrf_dev, skb);

drivers/net/vxlan.c+3−3

@@ -3144,7 +3144,7 @@ static int vxlan_validate(struct nlattr tb[], struct nlattr data[], · 3143 unmodified lines
3144	3144	u32 id = nla_get_u32(data[IFLA_VXLAN_ID]);
3145	3145
3146	3146	if (id >= VXLAN_N_VID) {
3147		− NL_SET_ERR_MSG_ATTR(extack, tb[IFLA_VXLAN_ID],
	3147	+ NL_SET_ERR_MSG_ATTR(extack, data[IFLA_VXLAN_ID],
3148	3148	"VXLAN ID must be lower than 16777216");
3149	3149	return -ERANGE;
3150	3150	}
@@ -3155,7 +3155,7 @@ static int vxlan_validate(struct nlattr tb[], struct nlattr data[], · 4 unmodified lines
3155	3155	= nla_data(data[IFLA_VXLAN_PORT_RANGE]);
3156	3156
3157	3157	if (ntohs(p->high) < ntohs(p->low)) {
3158		− NL_SET_ERR_MSG_ATTR(extack, tb[IFLA_VXLAN_PORT_RANGE],
	3158	+ NL_SET_ERR_MSG_ATTR(extack, data[IFLA_VXLAN_PORT_RANGE],
3159	3159	"Invalid source port range");
3160	3160	return -EINVAL;
3161	3161	}
@@ -3165,7 +3165,7 @@ static int vxlan_validate(struct nlattr tb[], struct nlattr data[], · 3 unmodified lines
3165	3165	enum ifla_vxlan_df df = nla_get_u8(data[IFLA_VXLAN_DF]);
3166	3166
3167	3167	if (df < 0 \|\| df > VXLAN_DF_MAX) {
3168		− NL_SET_ERR_MSG_ATTR(extack, tb[IFLA_VXLAN_DF],
	3168	+ NL_SET_ERR_MSG_ATTR(extack, data[IFLA_VXLAN_DF],
3169	3169	"Invalid DF attribute");
3170	3170	return -EINVAL;
3171	3171	}

drivers/net/wimax/i2400m/usb-fw.c+1−0

@@ -354,6 +354,7 @@ out: · 353 unmodified lines
354	354	usb_autopm_put_interface(i2400mu->usb_iface);
355	355	d_fnend(8, dev, "(i2400m %p ack %p size %zu) = %ld\n",
356	356	i2400m, ack, ack_size, (long) result);
	357	+ usb_put_urb(&notif_urb);
357	358	return result;
358	359
359	360	error_exceeded:

drivers/net/wireguard/queueing.c+3−1

@@ -35,8 +35,10 @@ int wg_packet_queue_init(struct crypt_queue *queue, work_func_t function, · 34 unmodified lines
35	35	if (multicore) {
36	36	queue->worker = wg_packet_percpu_multicore_worker_alloc(
37	37	function, queue);
38		− if (!queue->worker)
	38	+ if (!queue->worker) {
	39	+ ptr_ring_cleanup(&queue->ring, NULL);
39	40	return -ENOMEM;
	41	+ }
40	42	} else {
41	43	INIT_WORK(&queue->work, function);
42	44	}

drivers/net/wireguard/receive.c+10−11

@@ -226,21 +226,20 @@ void wg_packet_handshake_receive_worker(struct work_struct *work) · 225 unmodified lines
226	226	static void keep_key_fresh(struct wg_peer *peer)
227	227	{
228	228	struct noise_keypair *keypair;
229		− bool send = false;
	229	+ bool send;
230	230
231	231	if (peer->sent_lastminute_handshake)
232	232	return;
233	233
234	234	rcu_read_lock_bh();
235	235	keypair = rcu_dereference_bh(peer->keypairs.current_keypair);
236		− if (likely(keypair && READ_ONCE(keypair->sending.is_valid)) &&
237		− keypair->i_am_the_initiator &&
238		− unlikely(wg_birthdate_has_expired(keypair->sending.birthdate,
239		− REJECT_AFTER_TIME - KEEPALIVE_TIMEOUT - REKEY_TIMEOUT)))
240		− send = true;
	236	+ send = keypair && READ_ONCE(keypair->sending.is_valid) &&
	237	+ keypair->i_am_the_initiator &&
	238	+ wg_birthdate_has_expired(keypair->sending.birthdate,
	239	+ REJECT_AFTER_TIME - KEEPALIVE_TIMEOUT - REKEY_TIMEOUT);
241	240	rcu_read_unlock_bh();
242	241
243		− if (send) {
	242	+ if (unlikely(send)) {
244	243	peer->sent_lastminute_handshake = true;
245	244	wg_packet_send_queued_handshake_initiation(peer, false);
246	245	}
@@ -393,13 +392,11 @@ static void wg_packet_consume_data_done(struct wg_peer *peer, · 146 unmodified lines
393	392	len = ntohs(ip_hdr(skb)->tot_len);
394	393	if (unlikely(len < sizeof(struct iphdr)))
395	394	goto dishonest_packet_size;
396		− if (INET_ECN_is_ce(PACKET_CB(skb)->ds))
397		− IP_ECN_set_ce(ip_hdr(skb));
	395	+ INET_ECN_decapsulate(skb, PACKET_CB(skb)->ds, ip_hdr(skb)->tos);
398	396	} else if (skb->protocol == htons(ETH_P_IPV6)) {
399	397	len = ntohs(ipv6_hdr(skb)->payload_len) +
400	398	sizeof(struct ipv6hdr);
401		− if (INET_ECN_is_ce(PACKET_CB(skb)->ds))
402		− IP6_ECN_set_ce(skb, ipv6_hdr(skb));
	399	+ INET_ECN_decapsulate(skb, PACKET_CB(skb)->ds, ipv6_get_dsfield(ipv6_hdr(skb)));
403	400	} else {
404	401	goto dishonest_packet_type;
405	402	}
@@ -518,6 +515,8 @@ void wg_packet_decrypt_worker(struct work_struct *work) · 112 unmodified lines
518	515	&PACKET_CB(skb)->keypair->receiving)) ?
519	516	PACKET_STATE_CRYPTED : PACKET_STATE_DEAD;
520	517	wg_queue_enqueue_per_peer_napi(skb, state);
	518	+ if (need_resched())
	519	+ cond_resched();
521	520	}
522	521	}
523	522

drivers/net/wireguard/selftest/ratelimiter.c+2−2

@@ -120,9 +120,9 @@ bool __init wg_ratelimiter_selftest(void) · 119 unmodified lines
120	120	enum { TRIALS_BEFORE_GIVING_UP = 5000 };
121	121	bool success = false;
122	122	int test = 0, trials;
123		− struct sk_buff skb4, skb6;
	123	+ struct sk_buff skb4, skb6 = NULL;
124	124	struct iphdr *hdr4;
125		− struct ipv6hdr *hdr6;
	125	+ struct ipv6hdr *hdr6 = NULL;
126	126
127	127	if (IS_ENABLED(CONFIG_KASAN) \|\| IS_ENABLED(CONFIG_UBSAN))
128	128	return true;

drivers/net/wireguard/send.c+10−10

@@ -124,20 +124,17 @@ void wg_packet_send_handshake_cookie(struct wg_device *wg, · 123 unmodified lines
124	124	static void keep_key_fresh(struct wg_peer *peer)
125	125	{
126	126	struct noise_keypair *keypair;
127		− bool send = false;
	127	+ bool send;
128	128
129	129	rcu_read_lock_bh();
130	130	keypair = rcu_dereference_bh(peer->keypairs.current_keypair);
131		− if (likely(keypair && READ_ONCE(keypair->sending.is_valid)) &&
132		− (unlikely(atomic64_read(&keypair->sending.counter.counter) >
133		− REKEY_AFTER_MESSAGES) \|\|
134		− (keypair->i_am_the_initiator &&
135		− unlikely(wg_birthdate_has_expired(keypair->sending.birthdate,
136		− REKEY_AFTER_TIME)))))
137		− send = true;
	131	+ send = keypair && READ_ONCE(keypair->sending.is_valid) &&
	132	+ (atomic64_read(&keypair->sending.counter.counter) > REKEY_AFTER_MESSAGES \|\|
	133	+ (keypair->i_am_the_initiator &&
	134	+ wg_birthdate_has_expired(keypair->sending.birthdate, REKEY_AFTER_TIME)));
138	135	rcu_read_unlock_bh();
139	136
140		− if (send)
	137	+ if (unlikely(send))
141	138	wg_packet_send_queued_handshake_initiation(peer, false);
142	139	}
143	140
@@ -281,6 +278,8 @@ void wg_packet_tx_worker(struct work_struct *work) · 137 unmodified lines
281	278
282	279	wg_noise_keypair_put(keypair, false);
283	280	wg_peer_put(peer);
	281	+ if (need_resched())
	282	+ cond_resched();
284	283	}
285	284	}
286	285
@@ -304,7 +303,8 @@ void wg_packet_encrypt_worker(struct work_struct *work) · 17 unmodified lines
304	303	}
305	304	wg_queue_enqueue_per_peer(&PACKET_PEER(first)->tx_queue, first,
306	305	state);
307		−
	306	+ if (need_resched())
	307	+ cond_resched();
308	308	}
309	309	}
310	310

drivers/net/wireguard/socket.c+0−12

@@ -76,12 +76,6 @@ static int send4(struct wg_device wg, struct sk_buff skb, · 75 unmodified lines
76	76	net_dbg_ratelimited("%s: No route to %pISpfsc, error %d\n",
77	77	wg->dev->name, &endpoint->addr, ret);
78	78	goto err;
79		− } else if (unlikely(rt->dst.dev == skb->dev)) {
80		− ip_rt_put(rt);
81		− ret = -ELOOP;
82		− net_dbg_ratelimited("%s: Avoiding routing loop to %pISpfsc\n",
83		− wg->dev->name, &endpoint->addr);
84		− goto err;
85	79	}
86	80	if (cache)
87	81	dst_cache_set_ip4(cache, &rt->dst, fl.saddr);
@@ -149,12 +143,6 @@ static int send6(struct wg_device wg, struct sk_buff skb, · 61 unmodified lines
149	143	net_dbg_ratelimited("%s: No route to %pISpfsc, error %d\n",
150	144	wg->dev->name, &endpoint->addr, ret);
151	145	goto err;
152		− } else if (unlikely(dst->dev == skb->dev)) {
153		− dst_release(dst);
154		− ret = -ELOOP;
155		− net_dbg_ratelimited("%s: Avoiding routing loop to %pISpfsc\n",
156		− wg->dev->name, &endpoint->addr);
157		− goto err;
158	146	}
159	147	if (cache)
160	148	dst_cache_set_ip6(cache, dst, &fl.saddr);

drivers/net/wireless/intel/iwlegacy/3945-rs.c+1−1

@@ -374,7 +374,7 @@ out: · 373 unmodified lines
374	374	}
375	375
376	376	static void *
377		−il3945_rs_alloc(struct ieee80211_hw hw, struct dentry debugfsdir)
	377	+il3945_rs_alloc(struct ieee80211_hw *hw)
378	378	{
379	379	return hw->priv;
380	380	}

drivers/net/wireless/intel/iwlegacy/4965-rs.c+1−1

@@ -2474,7 +2474,7 @@ il4965_rs_fill_link_cmd(struct il_priv il, struct il_lq_sta lq_sta, · 2473 unmodified lines
2474	2474	}
2475	2475
2476	2476	static void *
2477		−il4965_rs_alloc(struct ieee80211_hw hw, struct dentry debugfsdir)
	2477	+il4965_rs_alloc(struct ieee80211_hw *hw)
2478	2478	{
2479	2479	return hw->priv;
2480	2480	}

drivers/net/wireless/intel/iwlwifi/dvm/rs.c+1−1

@@ -3019,7 +3019,7 @@ static void rs_fill_link_cmd(struct iwl_priv *priv, · 3018 unmodified lines
3019	3019	cpu_to_le16(priv->lib->bt_params->agg_time_limit);
3020	3020	}
3021	3021
3022		−static void rs_alloc(struct ieee80211_hw hw, struct dentry *debugfsdir)
	3022	+static void rs_alloc(struct ieee80211_hw hw)
3023	3023	{
3024	3024	return hw->priv;
3025	3025	}

drivers/net/wireless/intel/iwlwifi/fw/acpi.c+7−2

@@ -296,9 +296,14 @@ int iwl_sar_select_profile(struct iwl_fw_runtime *fwrt, · 295 unmodified lines
296	296	if (!prof->enabled) {
297	297	IWL_DEBUG_RADIO(fwrt, "SAR profile %d is disabled.\n",
298	298	profs[i]);
299		− /* if one of the profiles is disabled, we fail all */
300		− return -ENOENT;
	299	+ /*
	300	+ * if one of the profiles is disabled, we
	301	+ * ignore all of them and return 1 to
	302	+ * differentiate disabled from other failures.
	303	+ */
	304	+ return 1;
301	305	}
	306	+
302	307	IWL_DEBUG_INFO(fwrt,
303	308	"SAR EWRD: chain %d profile index %d\n",
304	309	i, profs[i]);

drivers/net/wireless/intel/iwlwifi/fw/api/txq.h+3−3

@@ -8,7 +8,7 @@ · 7 unmodified lines
8	8	* Copyright(c) 2007 - 2014 Intel Corporation. All rights reserved.
9	9	* Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
10	10	* Copyright(c) 2016 - 2017 Intel Deutschland GmbH
11		− * Copyright(c) 2019 Intel Corporation
	11	+ * Copyright(c) 2019 - 2020 Intel Corporation
12	12	*
13	13	* This program is free software; you can redistribute it and/or modify
14	14	* it under the terms of version 2 of the GNU General Public License as
@@ -31,7 +31,7 @@ · 16 unmodified lines
31	31	* Copyright(c) 2005 - 2014 Intel Corporation. All rights reserved.
32	32	* Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
33	33	* Copyright(c) 2016 - 2017 Intel Deutschland GmbH
34		− * Copyright(c) 2019 Intel Corporation
	34	+ * Copyright(c) 2019 - 2020 Intel Corporation
35	35	* All rights reserved.
36	36	*
37	37	* Redistribution and use in source and binary forms, with or without
@@ -99,7 +99,7 @@ enum iwl_mvm_dqa_txq { · 61 unmodified lines
99	99	IWL_MVM_DQA_MAX_MGMT_QUEUE = 8,
100	100	IWL_MVM_DQA_AP_PROBE_RESP_QUEUE = 9,
101	101	IWL_MVM_DQA_MIN_DATA_QUEUE = 10,
102		− IWL_MVM_DQA_MAX_DATA_QUEUE = 31,
	102	+ IWL_MVM_DQA_MAX_DATA_QUEUE = 30,
103	103	};
104	104
105	105	enum iwl_mvm_tx_fifo {

drivers/net/wireless/intel/iwlwifi/iwl-drv.c+1−1

@@ -1467,7 +1467,7 @@ static void iwl_req_fw_callback(const struct firmware ucode_raw, void context) · 1466 unmodified lines
1467	1467	kmemdup(pieces->dbg_conf_tlv[i],
1468	1468	pieces->dbg_conf_tlv_len[i],
1469	1469	GFP_KERNEL);
1470		− if (!pieces->dbg_conf_tlv[i])
	1470	+ if (!drv->fw.dbg.conf_tlv[i])
1471	1471	goto out_free_fw;
1472	1472	}
1473	1473	}

drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c+2−4

@@ -532,8 +532,7 @@ static struct ieee80211_sband_iftype_data iwl_he_capa[] = { · 531 unmodified lines
532	532	IEEE80211_HE_MAC_CAP1_TF_MAC_PAD_DUR_16US \|
533	533	IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_8,
534	534	.mac_cap_info[2] =
535		− IEEE80211_HE_MAC_CAP2_32BIT_BA_BITMAP \|
536		− IEEE80211_HE_MAC_CAP2_ACK_EN,
	535	+ IEEE80211_HE_MAC_CAP2_32BIT_BA_BITMAP,
537	536	.mac_cap_info[3] =
538	537	IEEE80211_HE_MAC_CAP3_OMI_CONTROL \|
539	538	IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_VHT_2,
@@ -617,8 +616,7 @@ static struct ieee80211_sband_iftype_data iwl_he_capa[] = { · 77 unmodified lines
617	616	IEEE80211_HE_MAC_CAP1_TF_MAC_PAD_DUR_16US \|
618	617	IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_8,
619	618	.mac_cap_info[2] =
620		− IEEE80211_HE_MAC_CAP2_BSR \|
621		− IEEE80211_HE_MAC_CAP2_ACK_EN,
	619	+ IEEE80211_HE_MAC_CAP2_BSR,
622	620	.mac_cap_info[3] =
623	621	IEEE80211_HE_MAC_CAP3_OMI_CONTROL \|
624	622	IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_VHT_2,

drivers/net/wireless/intel/iwlwifi/mvm/fw.c+11−14

@@ -727,6 +727,7 @@ int iwl_mvm_sar_select_profile(struct iwl_mvm *mvm, int prof_a, int prof_b) · 726 unmodified lines
727	727	struct iwl_dev_tx_power_cmd_v4 v4;
728	728	} cmd;
729	729
	730	+ int ret;
730	731	u16 len = 0;
731	732
732	733	cmd.v5.v3.set_mode = cpu_to_le32(IWL_TX_POWER_MODE_SET_CHAINS);
@@ -741,9 +742,14 @@ int iwl_mvm_sar_select_profile(struct iwl_mvm *mvm, int prof_a, int prof_b) · 8 unmodified lines
741	742	len = sizeof(cmd.v4.v3);
742	743
743	744
744		− if (iwl_sar_select_profile(&mvm->fwrt, cmd.v5.v3.per_chain_restriction,
745		− prof_a, prof_b))
746		− return -ENOENT;
	745	+ ret = iwl_sar_select_profile(&mvm->fwrt,
	746	+ cmd.v5.v3.per_chain_restriction,
	747	+ prof_a, prof_b);
	748	+
	749	+ /* return on error or if the profile is disabled (positive number) */
	750	+ if (ret)
	751	+ return ret;
	752	+
747	753	IWL_DEBUG_RADIO(mvm, "Sending REDUCE_TX_POWER_CMD per chain\n");
748	754	return iwl_mvm_send_cmd_pdu(mvm, REDUCE_TX_POWER_CMD, 0, len, &cmd);
749	755	}
@@ -1034,16 +1040,7 @@ static int iwl_mvm_sar_init(struct iwl_mvm *mvm) · 284 unmodified lines
1034	1040	"EWRD SAR BIOS table invalid or unavailable. (%d)\n",
1035	1041	ret);
1036	1042
1037		− ret = iwl_mvm_sar_select_profile(mvm, 1, 1);
1038		− /*
1039		− * If we don't have profile 0 from BIOS, just skip it. This
1040		− * means that SAR Geo will not be enabled either, even if we
1041		− * have other valid profiles.
1042		− */
1043		− if (ret == -ENOENT)
1044		− return 1;
1045		−
1046		− return ret;
	1043	+ return iwl_mvm_sar_select_profile(mvm, 1, 1);
1047	1044	}
1048	1045
1049	1046	static int iwl_mvm_load_rt_fw(struct iwl_mvm *mvm)
@@ -1272,7 +1269,7 @@ int iwl_mvm_up(struct iwl_mvm *mvm) · 222 unmodified lines
1272	1269	ret = iwl_mvm_sar_init(mvm);
1273	1270	if (ret == 0) {
1274	1271	ret = iwl_mvm_sar_geo_init(mvm);
1275		− } else if (ret > 0 && !iwl_sar_get_wgds_table(&mvm->fwrt)) {
	1272	+ } else if (ret == -ENOENT && !iwl_sar_get_wgds_table(&mvm->fwrt)) {
1276	1273	/*
1277	1274	* If basic SAR is not available, we check for WGDS,
1278	1275	* which should not be available either. If it is

drivers/net/wireless/intel/iwlwifi/mvm/rs.c+1−1

@@ -3665,7 +3665,7 @@ static void rs_fill_lq_cmd(struct iwl_mvm *mvm, · 3664 unmodified lines
3665	3665	cpu_to_le16(iwl_mvm_coex_agg_time_limit(mvm, sta));
3666	3666	}
3667	3667
3668		−static void rs_alloc(struct ieee80211_hw hw, struct dentry *debugfsdir)
	3668	+static void rs_alloc(struct ieee80211_hw hw)
3669	3669	{
3670	3670	return hw->priv;
3671	3671	}

drivers/net/wireless/intel/iwlwifi/mvm/rx.c+11−2

@@ -8,7 +8,7 @@ · 7 unmodified lines
8	8	* Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved.
9	9	* Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
10	10	* Copyright(c) 2016 - 2017 Intel Deutschland GmbH
11		− * Copyright(c) 2018 - 2019 Intel Corporation
	11	+ * Copyright(c) 2018 - 2020 Intel Corporation
12	12	*
13	13	* This program is free software; you can redistribute it and/or modify
14	14	* it under the terms of version 2 of the GNU General Public License as
@@ -31,7 +31,7 @@ · 16 unmodified lines
31	31	* Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved.
32	32	* Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
33	33	* Copyright(c) 2016 - 2017 Intel Deutschland GmbH
34		− * Copyright(c) 2018 - 2019 Intel Corporation
	34	+ * Copyright(c) 2018 - 2020 Intel Corporation
35	35	* All rights reserved.
36	36	*
37	37	* Redistribution and use in source and binary forms, with or without
@@ -566,6 +566,7 @@ void iwl_mvm_rx_rx_mpdu(struct iwl_mvm mvm, struct napi_struct napi, · 528 unmodified lines
566	566
567	567	struct iwl_mvm_stat_data {
568	568	struct iwl_mvm *mvm;
	569	+ __le32 flags;
569	570	__le32 mac_id;
570	571	u8 beacon_filter_average_energy;
571	572	void *general;
@@ -606,6 +607,13 @@ static void iwl_mvm_stat_iterator(void _data, u8 mac, · 34 unmodified lines
606	607	-general->beacon_average_energy[vif_id];
607	608	}
608	609
	610	+ /* make sure that beacon statistics don't go backwards with TCM
	611	+ * request to clear statistics
	612	+ */
	613	+ if (le32_to_cpu(data->flags) & IWL_STATISTICS_REPLY_FLG_CLEAR)
	614	+ mvmvif->beacon_stats.accu_num_beacons +=
	615	+ mvmvif->beacon_stats.num_beacons;
	616	+
609	617	if (mvmvif->id != id)
610	618	return;
611	619
@@ -763,6 +771,7 @@ void iwl_mvm_handle_rx_statistics(struct iwl_mvm *mvm, · 151 unmodified lines
763	771
764	772	flags = stats->flag;
765	773	}
	774	+ data.flags = flags;
766	775
767	776	iwl_mvm_rx_stats_check_trigger(mvm, pkt);
768	777

drivers/net/wireless/intel/iwlwifi/mvm/sta.c+7−2

@@ -722,6 +722,11 @@ static int iwl_mvm_find_free_queue(struct iwl_mvm *mvm, u8 sta_id, · 721 unmodified lines
722	722
723	723	lockdep_assert_held(&mvm->mutex);
724	724
	725	+ if (WARN(maxq >= mvm->trans->trans_cfg->base_params->num_of_queues,
	726	+ "max queue %d >= num_of_queues (%d)", maxq,
	727	+ mvm->trans->trans_cfg->base_params->num_of_queues))
	728	+ maxq = mvm->trans->trans_cfg->base_params->num_of_queues - 1;
	729	+
725	730	/* This should not be hit with new TX path */
726	731	if (WARN_ON(iwl_mvm_has_new_tx_api(mvm)))
727	732	return -ENOSPC;
@@ -1164,9 +1169,9 @@ static int iwl_mvm_inactivity_check(struct iwl_mvm *mvm, u8 alloc_for_sta) · 436 unmodified lines
1164	1169	inactive_tid_bitmap,
1165	1170	&unshare_queues,
1166	1171	&changetid_queues);
1167		− if (ret >= 0 && free_queue < 0) {
	1172	+ if (ret && free_queue < 0) {
1168	1173	queue_owner = sta;
1169		− free_queue = ret;
	1174	+ free_queue = i;
1170	1175	}
1171	1176	/* only unlock sta lock - we still need the queue info lock */
1172	1177	spin_unlock_bh(&mvmsta->lock);

drivers/net/wireless/intel/iwlwifi/pcie/ctxt-info-gen3.c+14−4

@@ -129,6 +129,18 @@ int iwl_pcie_ctxt_info_gen3_init(struct iwl_trans *trans, · 128 unmodified lines
129	129	int cmdq_size = max_t(u32, IWL_CMD_QUEUE_SIZE,
130	130	trans->cfg->min_txq_size);
131	131
	132	+ switch (trans_pcie->rx_buf_size) {
	133	+ case IWL_AMSDU_DEF:
	134	+ return -EINVAL;
	135	+ case IWL_AMSDU_2K:
	136	+ break;
	137	+ case IWL_AMSDU_4K:
	138	+ case IWL_AMSDU_8K:
	139	+ case IWL_AMSDU_12K:
	140	+ control_flags \|= IWL_PRPH_SCRATCH_RB_SIZE_4K;
	141	+ break;
	142	+ }
	143	+
132	144	/* Allocate prph scratch */
133	145	prph_scratch = dma_alloc_coherent(trans->dev, sizeof(*prph_scratch),
134	146	&trans_pcie->prph_scratch_dma_addr,
@@ -143,10 +155,8 @@ int iwl_pcie_ctxt_info_gen3_init(struct iwl_trans *trans, · 8 unmodified lines
143	155	cpu_to_le16((u16)iwl_read32(trans, CSR_HW_REV));
144	156	prph_sc_ctrl->version.size = cpu_to_le16(sizeof(*prph_scratch) / 4);
145	157
146		− control_flags = IWL_PRPH_SCRATCH_RB_SIZE_4K \|
147		− IWL_PRPH_SCRATCH_MTR_MODE \|
148		− (IWL_PRPH_MTR_FORMAT_256B &
149		− IWL_PRPH_SCRATCH_MTR_FORMAT);
	158	+ control_flags \|= IWL_PRPH_SCRATCH_MTR_MODE;
	159	+ control_flags \|= IWL_PRPH_MTR_FORMAT_256B & IWL_PRPH_SCRATCH_MTR_FORMAT;
150	160
151	161	/* initialize RX default queue */
152	162	prph_sc_ctrl->rbd_cfg.free_rbd_addr =

drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c+3−0

@@ -1418,6 +1418,9 @@ void iwl_trans_pcie_dyn_txq_free(struct iwl_trans *trans, int queue) · 1417 unmodified lines
1418	1418
1419	1419	iwl_pcie_gen2_txq_unmap(trans, queue);
1420	1420
	1421	+ iwl_pcie_gen2_txq_free_memory(trans, trans_pcie->txq[queue]);
	1422	+ trans_pcie->txq[queue] = NULL;
	1423	+
1421	1424	IWL_DEBUG_TX_QUEUES(trans, "Deactivate queue %d\n", queue);
1422	1425	}
1423	1426

drivers/net/wireless/mac80211_hwsim.c+1−1

@@ -4068,7 +4068,7 @@ static void hwsim_virtio_rx_work(struct work_struct *work) · 4067 unmodified lines
4068	4068	}
4069	4069	vq = hwsim_vqs[HWSIM_VQ_RX];
4070	4070	sg_init_one(sg, skb->head, skb_end_offset(skb));
4071		− err = virtqueue_add_inbuf(vq, sg, 1, skb, GFP_KERNEL);
	4071	+ err = virtqueue_add_inbuf(vq, sg, 1, skb, GFP_ATOMIC);
4072	4072	if (WARN(err, "virtqueue_add_inbuf returned %d\n", err))
4073	4073	nlmsg_free(skb);
4074	4074	else

drivers/net/wireless/realtek/rtlwifi/rc.c+1−1

@@ -261,7 +261,7 @@ static void rtl_rate_update(void *ppriv, · 260 unmodified lines
261	261	{
262	262	}
263	263
264		−static void rtl_rate_alloc(struct ieee80211_hw hw, struct dentry *debugfsdir)
	264	+static void rtl_rate_alloc(struct ieee80211_hw hw)
265	265	{
266	266	struct rtl_priv *rtlpriv = rtl_priv(hw);
267	267	return rtlpriv;

drivers/nvme/host/core.c+3−1

@@ -1110,7 +1110,7 @@ static int nvme_identify_ns_descs(struct nvme_ctrl *ctrl, unsigned nsid, · 1109 unmodified lines
1110	1110	* Don't treat an error as fatal, as we potentially already
1111	1111	* have a NGUID or EUI-64.
1112	1112	*/
1113		− if (status > 0)
	1113	+ if (status > 0 && !(status & NVME_SC_DNR))
1114	1114	status = 0;
1115	1115	goto free_data;
1116	1116	}
@@ -3642,6 +3642,8 @@ static void nvme_alloc_ns(struct nvme_ctrl *ctrl, unsigned nsid) · 2525 unmodified lines
3642	3642
3643	3643	return;
3644	3644	out_put_disk:
	3645	+ /* prevent double queue cleanup */
	3646	+ ns->disk->queue = NULL;
3645	3647	put_disk(ns->disk);
3646	3648	out_unlink_ns:
3647	3649	mutex_lock(&ctrl->subsys->lock);

drivers/nvme/host/pci.c+5−1

@@ -973,9 +973,13 @@ static inline void nvme_handle_cqe(struct nvme_queue *nvmeq, u16 idx) · 972 unmodified lines
973	973
974	974	static inline void nvme_update_cq_head(struct nvme_queue *nvmeq)
975	975	{
976		− if (++nvmeq->cq_head == nvmeq->q_depth) {
	976	+ u16 tmp = nvmeq->cq_head + 1;
	977	+
	978	+ if (tmp == nvmeq->q_depth) {
977	979	nvmeq->cq_head = 0;
978	980	nvmeq->cq_phase ^= 1;
	981	+ } else {
	982	+ nvmeq->cq_head = tmp;
979	983	}
980	984	}
981	985

drivers/pci/quirks.c+7−0

@@ -5567,3 +5567,10 @@ static void pci_fixup_no_d0_pme(struct pci_dev *dev) · 5566 unmodified lines
5567	5567	dev->pme_support &= ~(PCI_PM_CAP_PME_D0 >> PCI_PM_CAP_PME_SHIFT);
5568	5568	}
5569	5569	DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_ASMEDIA, 0x2142, pci_fixup_no_d0_pme);
	5570	+
	5571	+static void apex_pci_fixup_class(struct pci_dev *pdev)
	5572	+{
	5573	+ pdev->class = (PCI_CLASS_SYSTEM_OTHER << 8) \| pdev->class;
	5574	+}
	5575	+DECLARE_PCI_FIXUP_CLASS_HEADER(0x1ac1, 0x089a,
	5576	+ PCI_CLASS_NOT_DEFINED, 8, apex_pci_fixup_class);

drivers/phy/qualcomm/phy-qcom-qusb2.c+7−0

@@ -815,6 +815,13 @@ static const struct of_device_id qusb2_phy_of_match_table[] = { · 814 unmodified lines
815	815	}, {
816	816	.compatible = "qcom,msm8998-qusb2-phy",
817	817	.data = &msm8998_phy_cfg,
	818	+ }, {
	819	+ /*
	820	+ * Deprecated. Only here to support legacy device
	821	+ * trees that didn't include "qcom,qusb2-v2-phy"
	822	+ */
	823	+ .compatible = "qcom,sdm845-qusb2-phy",
	824	+ .data = &qusb2_v2_phy_cfg,
818	825	}, {
819	826	.compatible = "qcom,qusb2-v2-phy",
820	827	.data = &qusb2_v2_phy_cfg,

drivers/phy/qualcomm/phy-qcom-usb-hs-28nm.c+21−11

@@ -160,18 +160,11 @@ static int qcom_snps_hsphy_power_on(struct phy *phy) · 159 unmodified lines
160	160	ret = regulator_bulk_enable(VREG_NUM, priv->vregs);
161	161	if (ret)
162	162	return ret;
163		− ret = clk_bulk_prepare_enable(priv->num_clks, priv->clks);
164		− if (ret)
165		− goto err_disable_regulator;
	163	+
166	164	qcom_snps_hsphy_disable_hv_interrupts(priv);
167	165	qcom_snps_hsphy_exit_retention(priv);
168	166
169	167	return 0;
170		−
171		−err_disable_regulator:
172		− regulator_bulk_disable(VREG_NUM, priv->vregs);
173		−
174		− return ret;
175	168	}
176	169
177	170	static int qcom_snps_hsphy_power_off(struct phy *phy)
@@ -180,7 +173,6 @@ static int qcom_snps_hsphy_power_off(struct phy *phy) · 2 unmodified lines
180	173
181	174	qcom_snps_hsphy_enter_retention(priv);
182	175	qcom_snps_hsphy_enable_hv_interrupts(priv);
183		− clk_bulk_disable_unprepare(priv->num_clks, priv->clks);
184	176	regulator_bulk_disable(VREG_NUM, priv->vregs);
185	177
186	178	return 0;
@@ -266,21 +258,39 @@ static int qcom_snps_hsphy_init(struct phy *phy) · 79 unmodified lines
266	258	struct hsphy_priv *priv = phy_get_drvdata(phy);
267	259	int ret;
268	260
269		− ret = qcom_snps_hsphy_reset(priv);
	261	+ ret = clk_bulk_prepare_enable(priv->num_clks, priv->clks);
270	262	if (ret)
271	263	return ret;
272	264
	265	+ ret = qcom_snps_hsphy_reset(priv);
	266	+ if (ret)
	267	+ goto disable_clocks;
	268	+
273	269	qcom_snps_hsphy_init_sequence(priv);
274	270
275	271	ret = qcom_snps_hsphy_por_reset(priv);
276	272	if (ret)
277		− return ret;
	273	+ goto disable_clocks;
	274	+
	275	+ return 0;
	276	+
	277	+disable_clocks:
	278	+ clk_bulk_disable_unprepare(priv->num_clks, priv->clks);
	279	+ return ret;
	280	+}
	281	+
	282	+static int qcom_snps_hsphy_exit(struct phy *phy)
	283	+{
	284	+ struct hsphy_priv *priv = phy_get_drvdata(phy);
	285	+
	286	+ clk_bulk_disable_unprepare(priv->num_clks, priv->clks);
278	287
279	288	return 0;
280	289	}
281	290
282	291	static const struct phy_ops qcom_snps_hsphy_ops = {
283	292	.init = qcom_snps_hsphy_init,
	293	+ .exit = qcom_snps_hsphy_exit,
284	294	.power_on = qcom_snps_hsphy_power_on,
285	295	.power_off = qcom_snps_hsphy_power_off,
286	296	.set_mode = qcom_snps_hsphy_set_mode,

drivers/phy/tegra/Kconfig+2−1

@@ -1,7 +1,8 @@
1	1	# SPDX-License-Identifier: GPL-2.0-only
2	2	config PHY_TEGRA_XUSB
3	3	tristate "NVIDIA Tegra XUSB pad controller driver"
4		− depends on ARCH_TEGRA
	4	+ depends on ARCH_TEGRA && USB_SUPPORT
	5	+ select USB_COMMON
5	6	select USB_CONN_GPIO
6	7	select USB_PHY
7	8	help

drivers/platform/chrome/cros_ec_sensorhub.c+46−34

@@ -52,28 +52,15 @@ static int cros_ec_sensorhub_register(struct device *dev, · 51 unmodified lines
52	52	int sensor_type[MOTIONSENSE_TYPE_MAX] = { 0 };
53	53	struct cros_ec_command *msg = sensorhub->msg;
54	54	struct cros_ec_dev *ec = sensorhub->ec;
55		− int ret, i, sensor_num;
	55	+ int ret, i;
56	56	char *name;
57	57
58		− sensor_num = cros_ec_get_sensor_count(ec);
59		− if (sensor_num < 0) {
60		− dev_err(dev,
61		− "Unable to retrieve sensor information (err:%d)\n",
62		− sensor_num);
63		− return sensor_num;
64		− }
65		−
66		− sensorhub->sensor_num = sensor_num;
67		− if (sensor_num == 0) {
68		− dev_err(dev, "Zero sensors reported.\n");
69		− return -EINVAL;
70		− }
71	58
72	59	msg->version = 1;
73	60	msg->insize = sizeof(struct ec_response_motion_sense);
74	61	msg->outsize = sizeof(struct ec_params_motion_sense);
75	62
76		− for (i = 0; i < sensor_num; i++) {
	63	+ for (i = 0; i < sensorhub->sensor_num; i++) {
77	64	sensorhub->params->cmd = MOTIONSENSE_CMD_INFO;
78	65	sensorhub->params->info.sensor_num = i;
79	66
@@ -140,8 +127,7 @@ static int cros_ec_sensorhub_probe(struct platform_device *pdev) · 60 unmodified lines
140	127	struct cros_ec_dev *ec = dev_get_drvdata(dev->parent);
141	128	struct cros_ec_sensorhub *data;
142	129	struct cros_ec_command *msg;
143		− int ret;
144		− int i;
	130	+ int ret, i, sensor_num;
145	131
146	132	msg = devm_kzalloc(dev, sizeof(struct cros_ec_command) +
147	133	max((u16)sizeof(struct ec_params_motion_sense),
@@ -166,10 +152,52 @@ static int cros_ec_sensorhub_probe(struct platform_device *pdev) · 18 unmodified lines
166	152	dev_set_drvdata(dev, data);
167	153
168	154	/* Check whether this EC is a sensor hub. */
169		− if (cros_ec_check_features(data->ec, EC_FEATURE_MOTION_SENSE)) {
	155	+ if (cros_ec_check_features(ec, EC_FEATURE_MOTION_SENSE)) {
	156	+ sensor_num = cros_ec_get_sensor_count(ec);
	157	+ if (sensor_num < 0) {
	158	+ dev_err(dev,
	159	+ "Unable to retrieve sensor information (err:%d)\n",
	160	+ sensor_num);
	161	+ return sensor_num;
	162	+ }
	163	+ if (sensor_num == 0) {
	164	+ dev_err(dev, "Zero sensors reported.\n");
	165	+ return -EINVAL;
	166	+ }
	167	+ data->sensor_num = sensor_num;
	168	+
	169	+ /*
	170	+ * Prepare the ring handler before enumering the
	171	+ * sensors.
	172	+ */
	173	+ if (cros_ec_check_features(ec, EC_FEATURE_MOTION_SENSE_FIFO)) {
	174	+ ret = cros_ec_sensorhub_ring_allocate(data);
	175	+ if (ret)
	176	+ return ret;
	177	+ }
	178	+
	179	+ /* Enumerate the sensors.*/
170	180	ret = cros_ec_sensorhub_register(dev, data);
171	181	if (ret)
172	182	return ret;
	183	+
	184	+ /*
	185	+ * When the EC does not have a FIFO, the sensors will query
	186	+ * their data themselves via sysfs or a software trigger.
	187	+ */
	188	+ if (cros_ec_check_features(ec, EC_FEATURE_MOTION_SENSE_FIFO)) {
	189	+ ret = cros_ec_sensorhub_ring_add(data);
	190	+ if (ret)
	191	+ return ret;
	192	+ /*
	193	+ * The msg and its data is not under the control of the
	194	+ * ring handler.
	195	+ */
	196	+ return devm_add_action_or_reset(dev,
	197	+ cros_ec_sensorhub_ring_remove,
	198	+ data);
	199	+ }
	200	+
173	201	} else {
174	202	/*
175	203	* If the device has sensors but does not claim to
@@ -184,22 +212,6 @@ static int cros_ec_sensorhub_probe(struct platform_device *pdev) · 8 unmodified lines
184	212	}
185	213	}
186	214
187		− /*
188		− * If the EC does not have a FIFO, the sensors will query their data
189		− * themselves via sysfs or a software trigger.
190		− */
191		− if (cros_ec_check_features(ec, EC_FEATURE_MOTION_SENSE_FIFO)) {
192		− ret = cros_ec_sensorhub_ring_add(data);
193		− if (ret)
194		− return ret;
195		− /*
196		− * The msg and its data is not under the control of the ring
197		− * handler.
198		− */
199		− return devm_add_action_or_reset(dev,
200		− cros_ec_sensorhub_ring_remove,
201		− data);
202		− }
203	215
204	216	return 0;
205	217	}

drivers/platform/chrome/cros_ec_sensorhub_ring.c+46−27

@@ -957,17 +957,15 @@ static int cros_ec_sensorhub_event(struct notifier_block *nb, · 956 unmodified lines
957	957	}
958	958
959	959	/**
960		− * cros_ec_sensorhub_ring_add() - Add the FIFO functionality if the EC
961		− * supports it.
	960	+ * cros_ec_sensorhub_ring_allocate() - Prepare the FIFO functionality if the EC
	961	+ * supports it.
962	962	*
963	963	* @sensorhub : Sensor Hub object.
964	964	*
965	965	* Return: 0 on success.
966	966	*/
967		−int cros_ec_sensorhub_ring_add(struct cros_ec_sensorhub *sensorhub)
	967	+int cros_ec_sensorhub_ring_allocate(struct cros_ec_sensorhub *sensorhub)
968	968	{
969		− struct cros_ec_dev *ec = sensorhub->ec;
970		− int ret;
971	969	int fifo_info_length =
972	970	sizeof(struct ec_response_motion_sense_fifo_info) +
973	971	sizeof(u16) * sensorhub->sensor_num;
@@ -978,6 +976,49 @@ int cros_ec_sensorhub_ring_add(struct cros_ec_sensorhub *sensorhub) · 4 unmodified lines
978	976	if (!sensorhub->fifo_info)
979	977	return -ENOMEM;
980	978
	979	+ /*
	980	+ * Allocate the callback area based on the number of sensors.
	981	+ * Add one for the sensor ring.
	982	+ */
	983	+ sensorhub->push_data = devm_kcalloc(sensorhub->dev,
	984	+ sensorhub->sensor_num,
	985	+ sizeof(*sensorhub->push_data),
	986	+ GFP_KERNEL);
	987	+ if (!sensorhub->push_data)
	988	+ return -ENOMEM;
	989	+
	990	+ sensorhub->tight_timestamps = cros_ec_check_features(
	991	+ sensorhub->ec,
	992	+ EC_FEATURE_MOTION_SENSE_TIGHT_TIMESTAMPS);
	993	+
	994	+ if (sensorhub->tight_timestamps) {
	995	+ sensorhub->batch_state = devm_kcalloc(sensorhub->dev,
	996	+ sensorhub->sensor_num,
	997	+ sizeof(*sensorhub->batch_state),
	998	+ GFP_KERNEL);
	999	+ if (!sensorhub->batch_state)
	1000	+ return -ENOMEM;
	1001	+ }
	1002	+
	1003	+ return 0;
	1004	+}
	1005	+
	1006	+/**
	1007	+ * cros_ec_sensorhub_ring_add() - Add the FIFO functionality if the EC
	1008	+ * supports it.
	1009	+ *
	1010	+ * @sensorhub : Sensor Hub object.
	1011	+ *
	1012	+ * Return: 0 on success.
	1013	+ */
	1014	+int cros_ec_sensorhub_ring_add(struct cros_ec_sensorhub *sensorhub)
	1015	+{
	1016	+ struct cros_ec_dev *ec = sensorhub->ec;
	1017	+ int ret;
	1018	+ int fifo_info_length =
	1019	+ sizeof(struct ec_response_motion_sense_fifo_info) +
	1020	+ sizeof(u16) * sensorhub->sensor_num;
	1021	+
981	1022	/* Retrieve FIFO information */
982	1023	sensorhub->msg->version = 2;
983	1024	sensorhub->params->cmd = MOTIONSENSE_CMD_FIFO_INFO;
@@ -998,31 +1039,9 @@ int cros_ec_sensorhub_ring_add(struct cros_ec_sensorhub *sensorhub) · 14 unmodified lines
998	1039	if (!sensorhub->ring)
999	1040	return -ENOMEM;
1000	1041
1001		− /*
1002		− * Allocate the callback area based on the number of sensors.
1003		− */
1004		− sensorhub->push_data = devm_kcalloc(
1005		− sensorhub->dev, sensorhub->sensor_num,
1006		− sizeof(*sensorhub->push_data),
1007		− GFP_KERNEL);
1008		− if (!sensorhub->push_data)
1009		− return -ENOMEM;
1010		−
1011	1042	sensorhub->fifo_timestamp[CROS_EC_SENSOR_LAST_TS] =
1012	1043	cros_ec_get_time_ns();
1013	1044
1014		− sensorhub->tight_timestamps = cros_ec_check_features(
1015		− ec, EC_FEATURE_MOTION_SENSE_TIGHT_TIMESTAMPS);
1016		−
1017		− if (sensorhub->tight_timestamps) {
1018		− sensorhub->batch_state = devm_kcalloc(sensorhub->dev,
1019		− sensorhub->sensor_num,
1020		− sizeof(*sensorhub->batch_state),
1021		− GFP_KERNEL);
1022		− if (!sensorhub->batch_state)
1023		− return -ENOMEM;
1024		− }
1025		−
1026	1045	/* Register the notifier that will act as a top half interrupt. */
1027	1046	sensorhub->notifier.notifier_call = cros_ec_sensorhub_event;
1028	1047	ret = blocking_notifier_chain_register(&ec->ec_dev->event_notifier,

drivers/platform/x86/asus-nb-wmi.c+24−0

@@ -515,9 +515,33 @@ static struct asus_wmi_driver asus_nb_wmi_driver = { · 514 unmodified lines
515	515	.detect_quirks = asus_nb_wmi_quirks,
516	516	};
517	517
	518	+static const struct dmi_system_id asus_nb_wmi_blacklist[] __initconst = {
	519	+ {
	520	+ /*
	521	+ * asus-nb-wm adds no functionality. The T100TA has a detachable
	522	+ * USB kbd, so no hotkeys and it has no WMI rfkill; and loading
	523	+ * asus-nb-wm causes the camera LED to turn and _stay_ on.
	524	+ */
	525	+ .matches = {
	526	+ DMI_EXACT_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
	527	+ DMI_EXACT_MATCH(DMI_PRODUCT_NAME, "T100TA"),
	528	+ },
	529	+ },
	530	+ {
	531	+ /* The Asus T200TA has the same issue as the T100TA */
	532	+ .matches = {
	533	+ DMI_EXACT_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
	534	+ DMI_EXACT_MATCH(DMI_PRODUCT_NAME, "T200TA"),
	535	+ },
	536	+ },
	537	+ {} /* Terminating entry */
	538	+};
518	539
519	540	static int __init asus_nb_wmi_init(void)
520	541	{
	542	+ if (dmi_check_system(asus_nb_wmi_blacklist))
	543	+ return -ENODEV;
	544	+
521	545	return asus_wmi_register_driver(&asus_nb_wmi_driver);
522	546	}
523	547

drivers/platform/x86/intel-uncore-frequency.c+1−1

@@ -53,7 +53,7 @@ static int uncore_max_entries __read_mostly; · 52 unmodified lines
53	53	/* Storage for uncore data for all instances */
54	54	static struct uncore_data *uncore_instances;
55	55	/* Root of the all uncore sysfs kobjs */
56		−struct kobject *uncore_root_kobj;
	56	+static struct kobject *uncore_root_kobj;
57	57	/* Stores the CPU mask of the target CPUs to use during uncore read/write */
58	58	static cpumask_t uncore_cpu_mask;
59	59	/* CPU online callback register instance */

drivers/platform/x86/intel_pmc_core.c+5−19

@@ -255,7 +255,7 @@ static const struct pmc_bit_map *ext_cnp_pfear_map[] = { · 254 unmodified lines
255	255	};
256	256
257	257	static const struct pmc_bit_map icl_pfear_map[] = {
258		− /* Ice Lake generation onwards only */
	258	+ /* Ice Lake and Jasper Lake generation onwards only */
259	259	{"RES_65", BIT(0)},
260	260	{"RES_66", BIT(1)},
261	261	{"RES_67", BIT(2)},
@@ -274,7 +274,7 @@ static const struct pmc_bit_map *ext_icl_pfear_map[] = { · 12 unmodified lines
274	274	};
275	275
276	276	static const struct pmc_bit_map tgl_pfear_map[] = {
277		− /* Tiger Lake, Elkhart Lake and Jasper Lake generation onwards only */
	277	+ /* Tiger Lake and Elkhart Lake generation onwards only */
278	278	{"PSF9", BIT(0)},
279	279	{"RES_66", BIT(1)},
280	280	{"RES_67", BIT(2)},
@@ -692,7 +692,6 @@ static void pmc_core_lpm_display(struct pmc_dev pmcdev, struct device dev, · 411 unmodified lines
692	692	kfree(lpm_regs);
693	693	}
694	694
695		−#if IS_ENABLED(CONFIG_DEBUG_FS)
696	695	static bool slps0_dbg_latch;
697	696
698	697	static inline u8 pmc_core_reg_read_byte(struct pmc_dev *pmcdev, int offset)
@@ -1133,15 +1132,6 @@ static void pmc_core_dbgfs_register(struct pmc_dev *pmcdev) · 434 unmodified lines
1133	1132	&pmc_core_substate_l_sts_regs_fops);
1134	1133	}
1135	1134	}
1136		−#else
1137		−static inline void pmc_core_dbgfs_register(struct pmc_dev *pmcdev)
1138		−{
1139		−}
1140		−
1141		−static inline void pmc_core_dbgfs_unregister(struct pmc_dev *pmcdev)
1142		−{
1143		−}
1144		−#endif /* CONFIG_DEBUG_FS */
1145	1135
1146	1136	static const struct x86_cpu_id intel_pmc_core_ids[] = {
1147	1137	X86_MATCH_INTEL_FAM6_MODEL(SKYLAKE_L, &spt_reg_map),
@@ -1156,7 +1146,7 @@ static const struct x86_cpu_id intel_pmc_core_ids[] = { · 8 unmodified lines
1156	1146	X86_MATCH_INTEL_FAM6_MODEL(TIGERLAKE_L, &tgl_reg_map),
1157	1147	X86_MATCH_INTEL_FAM6_MODEL(TIGERLAKE, &tgl_reg_map),
1158	1148	X86_MATCH_INTEL_FAM6_MODEL(ATOM_TREMONT, &tgl_reg_map),
1159		− X86_MATCH_INTEL_FAM6_MODEL(ATOM_TREMONT_L, &tgl_reg_map),
	1149	+ X86_MATCH_INTEL_FAM6_MODEL(ATOM_TREMONT_L, &icl_reg_map),
1160	1150	{}
1161	1151	};
1162	1152
@@ -1260,13 +1250,11 @@ static int pmc_core_remove(struct platform_device *pdev) · 97 unmodified lines
1260	1250	return 0;
1261	1251	}
1262	1252
1263		−#ifdef CONFIG_PM_SLEEP
1264		−
1265	1253	static bool warn_on_s0ix_failures;
1266	1254	module_param(warn_on_s0ix_failures, bool, 0644);
1267	1255	MODULE_PARM_DESC(warn_on_s0ix_failures, "Check and warn for S0ix failures");
1268	1256
1269		−static int pmc_core_suspend(struct device *dev)
	1257	+static __maybe_unused int pmc_core_suspend(struct device *dev)
1270	1258	{
1271	1259	struct pmc_dev *pmcdev = dev_get_drvdata(dev);
1272	1260
@@ -1318,7 +1306,7 @@ static inline bool pmc_core_is_s0ix_failed(struct pmc_dev *pmcdev) · 45 unmodified lines
1318	1306	return false;
1319	1307	}
1320	1308
1321		−static int pmc_core_resume(struct device *dev)
	1309	+static __maybe_unused int pmc_core_resume(struct device *dev)
1322	1310	{
1323	1311	struct pmc_dev *pmcdev = dev_get_drvdata(dev);
1324	1312	const struct pmc_bit_map **maps = pmcdev->map->lpm_sts;
@@ -1348,8 +1336,6 @@ static int pmc_core_resume(struct device *dev) · 23 unmodified lines
1348	1336	return 0;
1349	1337	}
1350	1338
1351		−#endif
1352		−
1353	1339	static const struct dev_pm_ops pmc_core_pm_ops = {
1354	1340	SET_LATE_SYSTEM_SLEEP_PM_OPS(pmc_core_suspend, pmc_core_resume)
1355	1341	};

drivers/platform/x86/intel_pmc_core.h+0−2

@@ -282,9 +282,7 @@ struct pmc_dev { · 281 unmodified lines
282	282	u32 base_addr;
283	283	void __iomem *regbase;
284	284	const struct pmc_reg_map *map;
285		−#if IS_ENABLED(CONFIG_DEBUG_FS)
286	285	struct dentry *dbgfs_dir;
287		−#endif /* CONFIG_DEBUG_FS */
288	286	int pmc_xram_read_bit;
289	287	struct mutex lock; /* generic mutex lock for PMC Core */
290	288

drivers/platform/x86/surface3_power.c+2−2

@@ -522,8 +522,8 @@ static int mshw0011_probe(struct i2c_client *client) · 521 unmodified lines
522	522	strlcpy(board_info.type, "MSHW0011-bat0", I2C_NAME_SIZE);
523	523
524	524	bat0 = i2c_acpi_new_device(dev, 1, &board_info);
525		− if (!bat0)
526		− return -ENOMEM;
	525	+ if (IS_ERR(bat0))
	526	+ return PTR_ERR(bat0);
527	527
528	528	data->bat0 = bat0;
529	529	i2c_set_clientdata(bat0, data);

drivers/platform/x86/thinkpad_acpi.c+1−1

@@ -9548,7 +9548,7 @@ static ssize_t tpacpi_battery_store(int what, · 9547 unmodified lines
9548	9548	if (!battery_info.batteries[battery].start_support)
9549	9549	return -ENODEV;
9550	9550	/* valid values are [0, 99] */
9551		− if (value < 0 \|\| value > 99)
	9551	+ if (value > 99)
9552	9552	return -EINVAL;
9553	9553	if (value > battery_info.batteries[battery].charge_stop)
9554	9554	return -EINVAL;

drivers/platform/x86/xiaomi-wmi.c+2−2

@@ -23,7 +23,7 @@ struct xiaomi_wmi { · 22 unmodified lines
23	23	unsigned int key_code;
24	24	};
25	25
26		−int xiaomi_wmi_probe(struct wmi_device wdev, const void context)
	26	+static int xiaomi_wmi_probe(struct wmi_device wdev, const void context)
27	27	{
28	28	struct xiaomi_wmi *data;
29	29
@@ -48,7 +48,7 @@ int xiaomi_wmi_probe(struct wmi_device wdev, const void context) · 18 unmodified lines
48	48	return input_register_device(data->input_dev);
49	49	}
50	50
51		−void xiaomi_wmi_notify(struct wmi_device wdev, union acpi_object dummy)
	51	+static void xiaomi_wmi_notify(struct wmi_device wdev, union acpi_object dummy)
52	52	{
53	53	struct xiaomi_wmi *data;
54	54

drivers/ptp/ptp_ines.c+2−2

@@ -400,8 +400,8 @@ static int ines_hwtstamp(struct mii_timestamper mii_ts, struct ifreq ifr) · 399 unmodified lines
400	400	ines_write32(port, ts_stat_rx, ts_stat_rx);
401	401	ines_write32(port, ts_stat_tx, ts_stat_tx);
402	402
403		− port->rxts_enabled = ts_stat_rx == TS_ENABLE ? true : false;
404		− port->txts_enabled = ts_stat_tx == TS_ENABLE ? true : false;
	403	+ port->rxts_enabled = ts_stat_rx == TS_ENABLE;
	404	+ port->txts_enabled = ts_stat_tx == TS_ENABLE;
405	405
406	406	spin_unlock_irqrestore(&port->lock, flags);
407	407

drivers/regulator/core.c+11−14

@@ -5754,10 +5754,6 @@ static DECLARE_DELAYED_WORK(regulator_init_complete_work, · 5753 unmodified lines
5754	5754
5755	5755	static int __init regulator_init_complete(void)
5756	5756	{
5757		− int delay = driver_deferred_probe_timeout;
5758		−
5759		− if (delay < 0)
5760		− delay = 0;
5761	5757	/*
5762	5758	* Since DT doesn't provide an idiomatic mechanism for
5763	5759	* enabling full constraints and since it's much more natural
@@ -5768,17 +5764,18 @@ static int __init regulator_init_complete(void) · 4 unmodified lines
5768	5764	has_full_constraints = true;
5769	5765
5770	5766	/*
5771		− * If driver_deferred_probe_timeout is set, we punt
5772		− * completion for that many seconds since systems like
5773		− * distros will load many drivers from userspace so consumers
5774		− * might not always be ready yet, this is particularly an
5775		− * issue with laptops where this might bounce the display off
5776		− * then on. Ideally we'd get a notification from userspace
5777		− * when this happens but we don't so just wait a bit and hope
5778		− * we waited long enough. It'd be better if we'd only do
5779		− * this on systems that need it.
	5767	+ * We punt completion for an arbitrary amount of time since
	5768	+ * systems like distros will load many drivers from userspace
	5769	+ * so consumers might not always be ready yet, this is
	5770	+ * particularly an issue with laptops where this might bounce
	5771	+ * the display off then on. Ideally we'd get a notification
	5772	+ * from userspace when this happens but we don't so just wait
	5773	+ * a bit and hope we waited long enough. It'd be better if
	5774	+ * we'd only do this on systems that need it, and a kernel
	5775	+ * command line option might be useful.
5780	5776	*/
5781		− schedule_delayed_work(&regulator_init_complete_work, delay * HZ);
	5777	+ schedule_delayed_work(&regulator_init_complete_work,
	5778	+ msecs_to_jiffies(30000));
5782	5779
5783	5780	return 0;
5784	5781	}

drivers/remoteproc/mtk_common.h+1−1

@@ -68,7 +68,7 @@ struct mtk_scp { · 67 unmodified lines
68	68	wait_queue_head_t ack_wq;
69	69
70	70	void __iomem *cpu_addr;
71		− phys_addr_t phys_addr;
	71	+ dma_addr_t dma_addr;
72	72	size_t dram_size;
73	73
74	74	struct rproc_subdev *rpmsg_subdev;

drivers/remoteproc/mtk_scp.c+3−3

@@ -330,7 +330,7 @@ static void scp_da_to_va(struct rproc rproc, u64 da, size_t len) · 329 unmodified lines
330	330	if (offset >= 0 && (offset + len) < scp->sram_size)
331	331	return (void __force *)scp->sram_base + offset;
332	332	} else {
333		− offset = da - scp->phys_addr;
	333	+ offset = da - scp->dma_addr;
334	334	if (offset >= 0 && (offset + len) < scp->dram_size)
335	335	return (void __force *)scp->cpu_addr + offset;
336	336	}
@@ -451,7 +451,7 @@ static int scp_map_memory_region(struct mtk_scp *scp) · 114 unmodified lines
451	451	/* Reserved SCP code size */
452	452	scp->dram_size = MAX_CODE_SIZE;
453	453	scp->cpu_addr = dma_alloc_coherent(scp->dev, scp->dram_size,
454		− &scp->phys_addr, GFP_KERNEL);
	454	+ &scp->dma_addr, GFP_KERNEL);
455	455	if (!scp->cpu_addr)
456	456	return -ENOMEM;
457	457
@@ -461,7 +461,7 @@ static int scp_map_memory_region(struct mtk_scp *scp) · 3 unmodified lines
461	461	static void scp_unmap_memory_region(struct mtk_scp *scp)
462	462	{
463	463	dma_free_coherent(scp->dev, scp->dram_size, scp->cpu_addr,
464		− scp->phys_addr);
	464	+ scp->dma_addr);
465	465	of_reserved_mem_device_release(scp->dev);
466	466	}
467	467

drivers/remoteproc/qcom_q6v5_mss.c+20−13

@@ -367,7 +367,7 @@ unroll_pd_votes: · 366 unmodified lines
367	367	}
368	368
369	369	return ret;
370		−};
	370	+}
371	371
372	372	static void q6v5_pds_disable(struct q6v5 qproc, struct device *pds,
373	373	size_t pd_count)
@@ -1527,7 +1527,7 @@ unroll_attach: · 1153 unmodified lines
1527	1527	dev_pm_domain_detach(devs[i], false);
1528	1528
1529	1529	return ret;
1530		−};
	1530	+}
1531	1531
1532	1532	static void q6v5_pds_detach(struct q6v5 qproc, struct device *pds,
1533	1533	size_t pd_count)
@@ -1675,7 +1675,7 @@ static int q6v5_probe(struct platform_device *pdev) · 141 unmodified lines
1675	1675	ret = of_property_read_string_index(pdev->dev.of_node, "firmware-name",
1676	1676	1, &qproc->hexagon_mdt_image);
1677	1677	if (ret < 0 && ret != -EINVAL)
1678		− return ret;
	1678	+ goto free_rproc;
1679	1679
1680	1680	platform_set_drvdata(pdev, qproc);
1681	1681
@@ -1766,17 +1766,23 @@ static int q6v5_probe(struct platform_device *pdev) · 84 unmodified lines
1766	1766	qproc->sysmon = qcom_add_sysmon_subdev(rproc, "modem", 0x12);
1767	1767	if (IS_ERR(qproc->sysmon)) {
1768	1768	ret = PTR_ERR(qproc->sysmon);
1769		− goto detach_proxy_pds;
	1769	+ goto remove_subdevs;
1770	1770	}
1771	1771
1772	1772	ret = rproc_add(rproc);
1773	1773	if (ret)
1774		− goto detach_proxy_pds;
	1774	+ goto remove_sysmon_subdev;
1775	1775
1776	1776	return 0;
1777	1777
1778		−detach_proxy_pds:
	1778	+remove_sysmon_subdev:
	1779	+ qcom_remove_sysmon_subdev(qproc->sysmon);
	1780	+remove_subdevs:
1779	1781	qcom_remove_ipa_notify_subdev(qproc->rproc, &qproc->ipa_notify_subdev);
	1782	+ qcom_remove_ssr_subdev(rproc, &qproc->ssr_subdev);
	1783	+ qcom_remove_smd_subdev(rproc, &qproc->smd_subdev);
	1784	+ qcom_remove_glink_subdev(rproc, &qproc->glink_subdev);
	1785	+detach_proxy_pds:
1780	1786	q6v5_pds_detach(qproc, qproc->proxy_pds, qproc->proxy_pd_count);
1781	1787	detach_active_pds:
1782	1788	q6v5_pds_detach(qproc, qproc->active_pds, qproc->active_pd_count);
@@ -1789,19 +1795,20 @@ free_rproc: · 6 unmodified lines
1789	1795	static int q6v5_remove(struct platform_device *pdev)
1790	1796	{
1791	1797	struct q6v5 *qproc = platform_get_drvdata(pdev);
	1798	+ struct rproc *rproc = qproc->rproc;
1792	1799
1793		− rproc_del(qproc->rproc);
	1800	+ rproc_del(rproc);
1794	1801
1795	1802	qcom_remove_sysmon_subdev(qproc->sysmon);
1796		− qcom_remove_ipa_notify_subdev(qproc->rproc, &qproc->ipa_notify_subdev);
1797		− qcom_remove_glink_subdev(qproc->rproc, &qproc->glink_subdev);
1798		− qcom_remove_smd_subdev(qproc->rproc, &qproc->smd_subdev);
1799		− qcom_remove_ssr_subdev(qproc->rproc, &qproc->ssr_subdev);
	1803	+ qcom_remove_ipa_notify_subdev(rproc, &qproc->ipa_notify_subdev);
	1804	+ qcom_remove_ssr_subdev(rproc, &qproc->ssr_subdev);
	1805	+ qcom_remove_smd_subdev(rproc, &qproc->smd_subdev);
	1806	+ qcom_remove_glink_subdev(rproc, &qproc->glink_subdev);
1800	1807
1801		− q6v5_pds_detach(qproc, qproc->active_pds, qproc->active_pd_count);
1802	1808	q6v5_pds_detach(qproc, qproc->proxy_pds, qproc->proxy_pd_count);
	1809	+ q6v5_pds_detach(qproc, qproc->active_pds, qproc->active_pd_count);
1803	1810
1804		− rproc_free(qproc->rproc);
	1811	+ rproc_free(rproc);
1805	1812
1806	1813	return 0;
1807	1814	}

drivers/remoteproc/remoteproc_sysfs.c+1−0

@@ -4,6 +4,7 @@ · 3 unmodified lines
4	4	*/
5	5
6	6	#include <linux/remoteproc.h>
	7	+#include <linux/slab.h>
7	8
8	9	#include "remoteproc_internal.h"
9	10

drivers/remoteproc/stm32_rproc.c+1−0

@@ -19,6 +19,7 @@ · 18 unmodified lines
19	19	#include <linux/regmap.h>
20	20	#include <linux/remoteproc.h>
21	21	#include <linux/reset.h>
	22	+#include <linux/slab.h>
22	23	#include <linux/workqueue.h>
23	24
24	25	#include "remoteproc_internal.h"

drivers/rpmsg/mtk_rpmsg.c+1−0

@@ -8,6 +8,7 @@ · 7 unmodified lines
8	8	#include <linux/platform_device.h>
9	9	#include <linux/remoteproc.h>
10	10	#include <linux/rpmsg/mtk_rpmsg.h>
	11	+#include <linux/slab.h>
11	12	#include <linux/workqueue.h>
12	13
13	14	#include "rpmsg_internal.h"

drivers/s390/net/qeth_core_main.c+5−5

@@ -6717,17 +6717,17 @@ int qeth_stop(struct net_device *dev) · 6716 unmodified lines
6717	6717	unsigned int i;
6718	6718
6719	6719	/* Quiesce the NAPI instances: */
6720		− qeth_for_each_output_queue(card, queue, i) {
	6720	+ qeth_for_each_output_queue(card, queue, i)
6721	6721	napi_disable(&queue->napi);
6722		− del_timer_sync(&queue->timer);
6723		− }
6724	6722
6725	6723	/* Stop .ndo_start_xmit, might still access queue->napi. */
6726	6724	netif_tx_disable(dev);
6727	6725
6728		− /* Queues may get re-allocated, so remove the NAPIs here. */
6729		− qeth_for_each_output_queue(card, queue, i)
	6726	+ qeth_for_each_output_queue(card, queue, i) {
	6727	+ del_timer_sync(&queue->timer);
	6728	+ /* Queues may get re-allocated, so remove the NAPIs. */
6730	6729	netif_napi_del(&queue->napi);
	6730	+ }
6731	6731	} else {
6732	6732	netif_tx_disable(dev);
6733	6733	}

drivers/scsi/Kconfig+1−1

@@ -127,7 +127,7 @@ config CHR_DEV_SG · 126 unmodified lines
127	127
128	128	For scanners, look at SANE (<http://www.sane-project.org/>). For CD
129	129	writer software look at Cdrtools
130		− (<http://cdrecord.berlios.de/private/cdrecord.html>)
	130	+ (<http://cdrtools.sourceforge.net/>)
131	131	and for burning a "disk at once": CDRDAO
132	132	(<http://cdrdao.sourceforge.net/>). Cdparanoia is a high
133	133	quality digital reader of audio CDs (<http://www.xiph.org/paranoia/>).

drivers/scsi/ibmvscsi/ibmvfc.c+5−0

@@ -3640,6 +3640,11 @@ static void ibmvfc_tgt_implicit_logout_and_del(struct ibmvfc_target *tgt) · 3639 unmodified lines
3640	3640	struct ibmvfc_host *vhost = tgt->vhost;
3641	3641	struct ibmvfc_event *evt;
3642	3642
	3643	+ if (!vhost->logged_in) {
	3644	+ ibmvfc_set_tgt_action(tgt, IBMVFC_TGT_ACTION_DEL_RPORT);
	3645	+ return;
	3646	+ }
	3647	+
3643	3648	if (vhost->discovery_threads >= disc_threads)
3644	3649	return;
3645	3650

drivers/scsi/ibmvscsi/ibmvscsi.c+0−4

@@ -2320,16 +2320,12 @@ static int ibmvscsi_probe(struct vio_dev vdev, const struct vio_device_id id) · 2319 unmodified lines
2320	2320	static int ibmvscsi_remove(struct vio_dev *vdev)
2321	2321	{
2322	2322	struct ibmvscsi_host_data *hostdata = dev_get_drvdata(&vdev->dev);
2323		− unsigned long flags;
2324	2323
2325	2324	srp_remove_host(hostdata->host);
2326	2325	scsi_remove_host(hostdata->host);
2327	2326
2328	2327	purge_requests(hostdata, DID_ERROR);
2329		−
2330		− spin_lock_irqsave(hostdata->host->host_lock, flags);
2331	2328	release_event_pool(&hostdata->pool, hostdata);
2332		− spin_unlock_irqrestore(hostdata->host->host_lock, flags);
2333	2329
2334	2330	ibmvscsi_release_crq_queue(&hostdata->queue, hostdata,
2335	2331	max_events);

drivers/scsi/qla2xxx/qla_attr.c+1−1

@@ -3031,11 +3031,11 @@ qla24xx_vport_delete(struct fc_vport *fc_vport) · 3030 unmodified lines
3031	3031	test_bit(FCPORT_UPDATE_NEEDED, &vha->dpc_flags))
3032	3032	msleep(1000);
3033	3033
3034		− qla_nvme_delete(vha);
3035	3034
3036	3035	qla24xx_disable_vp(vha);
3037	3036	qla2x00_wait_for_sess_deletion(vha);
3038	3037
	3038	+ qla_nvme_delete(vha);
3039	3039	vha->flags.delete_progress = 1;
3040	3040
3041	3041	qlt_remove_target(ha, vha);

drivers/scsi/qla2xxx/qla_mbx.c+1−1

@@ -3153,7 +3153,7 @@ qla24xx_abort_command(srb_t *sp) · 3152 unmodified lines
3153	3153	ql_dbg(ql_dbg_mbx + ql_dbg_verbose, vha, 0x108c,
3154	3154	"Entered %s.\n", __func__);
3155	3155
3156		− if (vha->flags.qpairs_available && sp->qpair)
	3156	+ if (sp->qpair)
3157	3157	req = sp->qpair->req;
3158	3158	else
3159	3159	return QLA_FUNCTION_FAILED;

drivers/scsi/qla2xxx/qla_os.c+17−18

@@ -3732,6 +3732,13 @@ qla2x00_remove_one(struct pci_dev *pdev) · 3731 unmodified lines
3732	3732	}
3733	3733	qla2x00_wait_for_hba_ready(base_vha);
3734	3734
	3735	+ /*
	3736	+ * if UNLOADING flag is already set, then continue unload,
	3737	+ * where it was set first.
	3738	+ */
	3739	+ if (test_and_set_bit(UNLOADING, &base_vha->dpc_flags))
	3740	+ return;
	3741	+
3735	3742	if (IS_QLA25XX(ha) \|\| IS_QLA2031(ha) \|\| IS_QLA27XX(ha) \|\|
3736	3743	IS_QLA28XX(ha)) {
3737	3744	if (ha->flags.fw_started)
@@ -3750,15 +3757,6 @@ qla2x00_remove_one(struct pci_dev *pdev) · 12 unmodified lines
3750	3757
3751	3758	qla2x00_wait_for_sess_deletion(base_vha);
3752	3759
3753		− /*
3754		− * if UNLOAD flag is already set, then continue unload,
3755		− * where it was set first.
3756		− */
3757		− if (test_bit(UNLOADING, &base_vha->dpc_flags))
3758		− return;
3759		−
3760		− set_bit(UNLOADING, &base_vha->dpc_flags);
3761		−
3762	3760	qla_nvme_delete(base_vha);
3763	3761
3764	3762	dma_free_coherent(&ha->pdev->dev,
@@ -4864,6 +4862,9 @@ qla2x00_alloc_work(struct scsi_qla_host *vha, enum qla_work_type type) · 1099 unmodified lines
4864	4862	struct qla_work_evt *e;
4865	4863	uint8_t bail;
4866	4864
	4865	+ if (test_bit(UNLOADING, &vha->dpc_flags))
	4866	+ return NULL;
	4867	+
4867	4868	QLA_VHA_MARK_BUSY(vha, bail);
4868	4869	if (bail)
4869	4870	return NULL;
@@ -6628,13 +6629,6 @@ qla2x00_disable_board_on_pci_error(struct work_struct *work) · 1758 unmodified lines
6628	6629	struct pci_dev *pdev = ha->pdev;
6629	6630	scsi_qla_host_t *base_vha = pci_get_drvdata(ha->pdev);
6630	6631
6631		− /*
6632		− * if UNLOAD flag is already set, then continue unload,
6633		− * where it was set first.
6634		− */
6635		− if (test_bit(UNLOADING, &base_vha->dpc_flags))
6636		− return;
6637		−
6638	6632	ql_log(ql_log_warn, base_vha, 0x015b,
6639	6633	"Disabling adapter.\n");
6640	6634
@@ -6645,9 +6639,14 @@ qla2x00_disable_board_on_pci_error(struct work_struct *work) · 4 unmodified lines
6645	6639	return;
6646	6640	}
6647	6641
6648		− qla2x00_wait_for_sess_deletion(base_vha);
	6642	+ /*
	6643	+ * if UNLOADING flag is already set, then continue unload,
	6644	+ * where it was set first.
	6645	+ */
	6646	+ if (test_and_set_bit(UNLOADING, &base_vha->dpc_flags))
	6647	+ return;
6649	6648
6650		− set_bit(UNLOADING, &base_vha->dpc_flags);
	6649	+ qla2x00_wait_for_sess_deletion(base_vha);
6651	6650
6652	6651	qla2x00_delete_all_vps(ha, base_vha);
6653	6652

drivers/scsi/scsi_lib.c+1−0

@@ -2284,6 +2284,7 @@ scsi_device_set_state(struct scsi_device *sdev, enum scsi_device_state state) · 2283 unmodified lines
2284	2284	switch (oldstate) {
2285	2285	case SDEV_RUNNING:
2286	2286	case SDEV_CREATED_BLOCK:
	2287	+ case SDEV_QUIESCE:
2287	2288	case SDEV_OFFLINE:
2288	2289	break;
2289	2290	default:

drivers/scsi/sg.c+3−1

@@ -685,8 +685,10 @@ sg_write(struct file filp, const char __user buf, size_t count, loff_t * ppos) · 684 unmodified lines
685	685	hp->flags = input_size; /* structure abuse ... */
686	686	hp->pack_id = old_hdr.pack_id;
687	687	hp->usr_ptr = NULL;
688		− if (copy_from_user(cmnd, buf, cmd_size))
	688	+ if (copy_from_user(cmnd, buf, cmd_size)) {
	689	+ sg_remove_request(sfp, srp);
689	690	return -EFAULT;
	691	+ }
690	692	/*
691	693	* SG_DXFER_TO_FROM_DEV is functionally equivalent to SG_DXFER_FROM_DEV,
692	694	* but is is possible that the app intended SG_DXFER_TO_DEV, because there

drivers/soc/fsl/dpio/dpio-service.c+14−5

@@ -478,12 +478,18 @@ int dpaa2_io_service_enqueue_multiple_desc_fq(struct dpaa2_io *d, · 477 unmodified lines
478	478	const struct dpaa2_fd *fd,
479	479	int nb)
480	480	{
481		− int i;
482		− struct qbman_eq_desc ed[32];
	481	+ struct qbman_eq_desc *ed;
	482	+ int i, ret;
	483	+
	484	+ ed = kcalloc(sizeof(struct qbman_eq_desc), 32, GFP_KERNEL);
	485	+ if (!ed)
	486	+ return -ENOMEM;
483	487
484	488	d = service_select(d);
485		− if (!d)
486		− return -ENODEV;
	489	+ if (!d) {
	490	+ ret = -ENODEV;
	491	+ goto out;
	492	+ }
487	493
488	494	for (i = 0; i < nb; i++) {
489	495	qbman_eq_desc_clear(&ed[i]);
@@ -491,7 +497,10 @@ int dpaa2_io_service_enqueue_multiple_desc_fq(struct dpaa2_io *d, · 1 unmodified line
491	497	qbman_eq_desc_set_fq(&ed[i], fqid[i]);
492	498	}
493	499
494		− return qbman_swp_enqueue_multiple_desc(d->swp, &ed[0], fd, nb);
	500	+ ret = qbman_swp_enqueue_multiple_desc(d->swp, &ed[0], fd, nb);
	501	+out:
	502	+ kfree(ed);
	503	+ return ret;
495	504	}
496	505	EXPORT_SYMBOL(dpaa2_io_service_enqueue_multiple_desc_fq);
497	506

drivers/soc/fsl/dpio/qbman-portal.c+2−4

@@ -753,7 +753,7 @@ int qbman_swp_enqueue_multiple_mem_back(struct qbman_swp *s, · 752 unmodified lines
753	753	if (!s->eqcr.available) {
754	754	eqcr_ci = s->eqcr.ci;
755	755	p = s->addr_cena + QBMAN_CENA_SWP_EQCR_CI_MEMBACK;
756		− s->eqcr.ci = __raw_readl(p) & full_mask;
	756	+ s->eqcr.ci = *p & full_mask;
757	757	s->eqcr.available = qm_cyc_diff(s->eqcr.pi_ring_size,
758	758	eqcr_ci, s->eqcr.ci);
759	759	if (!s->eqcr.available) {
@@ -823,7 +823,6 @@ int qbman_swp_enqueue_multiple_desc_direct(struct qbman_swp *s, · 63 unmodified lines
823	823	const uint32_t *cl;
824	824	uint32_t eqcr_ci, eqcr_pi, half_mask, full_mask;
825	825	int i, num_enqueued = 0;
826		− uint64_t addr_cena;
827	826
828	827	half_mask = (s->eqcr.pi_ci_mask>>1);
829	828	full_mask = s->eqcr.pi_ci_mask;
@@ -867,7 +866,6 @@ int qbman_swp_enqueue_multiple_desc_direct(struct qbman_swp *s, · 37 unmodified lines
867	866
868	867	/* Flush all the cacheline without load/store in between */
869	868	eqcr_pi = s->eqcr.pi;
870		− addr_cena = (uint64_t)s->addr_cena;
871	869	for (i = 0; i < num_enqueued; i++)
872	870	eqcr_pi++;
873	871	s->eqcr.pi = eqcr_pi & full_mask;
@@ -901,7 +899,7 @@ int qbman_swp_enqueue_multiple_desc_mem_back(struct qbman_swp *s, · 27 unmodified lines
901	899	if (!s->eqcr.available) {
902	900	eqcr_ci = s->eqcr.ci;
903	901	p = s->addr_cena + QBMAN_CENA_SWP_EQCR_CI_MEMBACK;
904		− s->eqcr.ci = __raw_readl(p) & full_mask;
	902	+ s->eqcr.ci = *p & full_mask;
905	903	s->eqcr.available = qm_cyc_diff(s->eqcr.pi_ring_size,
906	904	eqcr_ci, s->eqcr.ci);
907	905	if (!s->eqcr.available)

drivers/soc/imx/Kconfig+1−0

@@ -21,6 +21,7 @@ config SOC_IMX8M · 20 unmodified lines
21	21	bool "i.MX8M SoC family support"
22	22	depends on ARCH_MXC \|\| COMPILE_TEST
23	23	default ARCH_MXC && ARM64
	24	+ select SOC_BUS
24	25	help
25	26	If you say yes here you get support for the NXP i.MX8M family
26	27	support, it will provide the SoC info like SoC family,

drivers/soc/xilinx/Kconfig+2−2

@@ -19,7 +19,7 @@ config XILINX_VCU · 18 unmodified lines
19	19
20	20	config ZYNQMP_POWER
21	21	bool "Enable Xilinx Zynq MPSoC Power Management driver"
22		− depends on PM && ARCH_ZYNQMP
	22	+ depends on PM && ZYNQMP_FIRMWARE
23	23	default y
24	24	select MAILBOX
25	25	select ZYNQMP_IPI_MBOX
@@ -35,7 +35,7 @@ config ZYNQMP_POWER · 9 unmodified lines
35	35	config ZYNQMP_PM_DOMAINS
36	36	bool "Enable Zynq MPSoC generic PM domains"
37	37	default y
38		− depends on PM && ARCH_ZYNQMP && ZYNQMP_FIRMWARE
	38	+ depends on PM && ZYNQMP_FIRMWARE
39	39	select PM_GENERIC_DOMAINS
40	40	help
41	41	Say yes to enable device power management through PM domains

drivers/staging/comedi/comedi_fops.c+3−1

@@ -2725,8 +2725,10 @@ static int comedi_open(struct inode inode, struct file file) · 2724 unmodified lines
2725	2725	}
2726	2726
2727	2727	cfp = kzalloc(sizeof(*cfp), GFP_KERNEL);
2728		− if (!cfp)
	2728	+ if (!cfp) {
	2729	+ comedi_dev_put(dev);
2729	2730	return -ENOMEM;
	2731	+ }
2730	2732
2731	2733	cfp->dev = dev;
2732	2734

drivers/staging/comedi/drivers/dt2815.c+3−0

@@ -92,6 +92,7 @@ static int dt2815_ao_insn(struct comedi_device dev, struct comedi_subdevice s, · 91 unmodified lines
92	92	int ret;
93	93
94	94	for (i = 0; i < insn->n; i++) {
	95	+ /* FIXME: lo bit 0 chooses voltage output or current output */
95	96	lo = ((data[i] & 0x0f) << 4) \| (chan << 1) \| 0x01;
96	97	hi = (data[i] & 0xff0) >> 4;
97	98
@@ -105,6 +106,8 @@ static int dt2815_ao_insn(struct comedi_device dev, struct comedi_subdevice s, · 7 unmodified lines
105	106	if (ret)
106	107	return ret;
107	108
	109	+ outb(hi, dev->iobase + DT2815_DATA);
	110	+
108	111	devpriv->ao_readback[chan] = data[i];
109	112	}
110	113	return i;

drivers/staging/gasket/apex_driver.c+0−7

@@ -570,13 +570,6 @@ static const struct pci_device_id apex_pci_ids[] = { · 569 unmodified lines
570	570	{ PCI_DEVICE(APEX_PCI_VENDOR_ID, APEX_PCI_DEVICE_ID) }, { 0 }
571	571	};
572	572
573		−static void apex_pci_fixup_class(struct pci_dev *pdev)
574		−{
575		− pdev->class = (PCI_CLASS_SYSTEM_OTHER << 8) \| pdev->class;
576		−}
577		−DECLARE_PCI_FIXUP_CLASS_HEADER(APEX_PCI_VENDOR_ID, APEX_PCI_DEVICE_ID,
578		− PCI_CLASS_NOT_DEFINED, 8, apex_pci_fixup_class);
579		−
580	573	static int apex_pci_probe(struct pci_dev *pci_dev,
581	574	const struct pci_device_id *id)
582	575	{

drivers/staging/gasket/gasket_core.c+4−0

@@ -925,6 +925,10 @@ do_map_region(const struct gasket_dev gasket_dev, struct vm_area_struct vma, · 924 unmodified lines
925	925	gasket_get_bar_index(gasket_dev,
926	926	(vma->vm_pgoff << PAGE_SHIFT) +
927	927	driver_desc->legacy_mmap_address_offset);
	928	+
	929	+ if (bar_index < 0)
	930	+ return DO_MAP_REGION_INVALID;
	931	+
928	932	phys_base = gasket_dev->bar_data[bar_index].phys_base + phys_offset;
929	933	while (mapped_bytes < map_length) {
930	934	/*

drivers/staging/gasket/gasket_sysfs.c+1−2

@@ -228,8 +228,7 @@ int gasket_sysfs_create_entries(struct device *device, · 227 unmodified lines
228	228	}
229	229
230	230	mutex_lock(&mapping->mutex);
231		− for (i = 0; strcmp(attrs[i].attr.attr.name, GASKET_ARRAY_END_MARKER);
232		− i++) {
	231	+ for (i = 0; attrs[i].attr.attr.name != NULL; i++) {
233	232	if (mapping->attribute_count == GASKET_SYSFS_MAX_NODES) {
234	233	dev_err(device,
235	234	"Maximum number of sysfs nodes reached for device\n");

drivers/staging/gasket/gasket_sysfs.h+0−4

@@ -30,10 +30,6 @@ · 29 unmodified lines
30	30	*/
31	31	#define GASKET_SYSFS_MAX_NODES 196
32	32
33		−/* End markers for sysfs struct arrays. */
34		−#define GASKET_ARRAY_END_TOKEN GASKET_RESERVED_ARRAY_END
35		−#define GASKET_ARRAY_END_MARKER __stringify(GASKET_ARRAY_END_TOKEN)
36		−
37	33	/*
38	34	* Terminator struct for a gasket_sysfs_attr array. Must be at the end of
39	35	* all gasket_sysfs_attribute arrays.

drivers/staging/ks7010/TODO+0−1

@@ -30,5 +30,4 @@ Now the TODOs: · 29 unmodified lines
30	30
31	31	Please send any patches to:
32	32	Greg Kroah-Hartman <gregkh@linuxfoundation.org>
33		−Wolfram Sang <wsa@the-dreams.de>
34	33	Linux Driver Project Developer List <driverdev-devel@linuxdriverproject.org>

drivers/staging/vt6656/key.c+3−11

@@ -83,9 +83,6 @@ static int vnt_set_keymode(struct ieee80211_hw hw, u8 mac_addr, · 82 unmodified lines
83	83	case VNT_KEY_PAIRWISE:
84	84	key_mode \|= mode;
85	85	key_inx = 4;
86		− /* Don't save entry for pairwise key for station mode */
87		− if (priv->op_mode == NL80211_IFTYPE_STATION)
88		− clear_bit(entry, &priv->key_entry_inuse);
89	86	break;
90	87	default:
91	88	return -EINVAL;
@@ -109,7 +106,6 @@ static int vnt_set_keymode(struct ieee80211_hw hw, u8 mac_addr, · 17 unmodified lines
109	106	int vnt_set_keys(struct ieee80211_hw hw, struct ieee80211_sta sta,
110	107	struct ieee80211_vif vif, struct ieee80211_key_conf key)
111	108	{
112		− struct ieee80211_bss_conf *conf = &vif->bss_conf;
113	109	struct vnt_private *priv = hw->priv;
114	110	u8 *mac_addr = NULL;
115	111	u8 key_dec_mode = 0;
@@ -154,16 +150,12 @@ int vnt_set_keys(struct ieee80211_hw hw, struct ieee80211_sta sta, · 38 unmodified lines
154	150	return -EOPNOTSUPP;
155	151	}
156	152
157		− if (key->flags & IEEE80211_KEY_FLAG_PAIRWISE) {
	153	+ if (key->flags & IEEE80211_KEY_FLAG_PAIRWISE)
158	154	vnt_set_keymode(hw, mac_addr, key, VNT_KEY_PAIRWISE,
159	155	key_dec_mode, true);
160		− } else {
161		− vnt_set_keymode(hw, mac_addr, key, VNT_KEY_DEFAULTKEY,
	156	+ else
	157	+ vnt_set_keymode(hw, mac_addr, key, VNT_KEY_GROUP_ADDRESS,
162	158	key_dec_mode, true);
163	159
164		− vnt_set_keymode(hw, (u8 *)conf->bssid, key,
165		− VNT_KEY_GROUP_ADDRESS, key_dec_mode, true);
166		− }
167		−
168	160	return 0;
169	161	}

drivers/staging/vt6656/main_usb.c+17−14

@@ -625,8 +625,6 @@ static int vnt_add_interface(struct ieee80211_hw hw, struct ieee80211_vif vif) · 624 unmodified lines
625	625
626	626	priv->op_mode = vif->type;
627	627
628		− vnt_set_bss_mode(priv);
629		−
630	628	/* LED blink on TX */
631	629	vnt_mac_set_led(priv, LEDSTS_STS, LEDSTS_INTER);
632	630
@@ -713,7 +711,6 @@ static void vnt_bss_info_changed(struct ieee80211_hw *hw, · 80 unmodified lines
713	711	priv->basic_rates = conf->basic_rates;
714	712
715	713	vnt_update_top_rates(priv);
716		− vnt_set_bss_mode(priv);
717	714
718	715	dev_dbg(&priv->usb->dev, "basic rates %x\n", conf->basic_rates);
719	716	}
@@ -742,11 +739,14 @@ static void vnt_bss_info_changed(struct ieee80211_hw *hw, · 22 unmodified lines
742	739	priv->short_slot_time = false;
743	740
744	741	vnt_set_short_slot_time(priv);
745		− vnt_update_ifs(priv);
746	742	vnt_set_vga_gain_offset(priv, priv->bb_vga[0]);
747	743	vnt_update_pre_ed_threshold(priv, false);
748	744	}
749	745
	746	+ if (changed & (BSS_CHANGED_BASIC_RATES \| BSS_CHANGED_ERP_PREAMBLE \|
	747	+ BSS_CHANGED_ERP_SLOT))
	748	+ vnt_set_bss_mode(priv);
	749	+
750	750	if (changed & BSS_CHANGED_TXPOWER)
751	751	vnt_rf_setpower(priv, priv->current_rate,
752	752	conf->chandef.chan->hw_value);
@@ -770,12 +770,15 @@ static void vnt_bss_info_changed(struct ieee80211_hw *hw, · 17 unmodified lines
770	770	vnt_mac_reg_bits_on(priv, MAC_REG_TFTCTL,
771	771	TFTCTL_TSFCNTREN);
772	772
773		− vnt_adjust_tsf(priv, conf->beacon_rate->hw_value,
774		− conf->sync_tsf, priv->current_tsf);
775		−
776	773	vnt_mac_set_beacon_interval(priv, conf->beacon_int);
777	774
778	775	vnt_reset_next_tbtt(priv, conf->beacon_int);
	776	+
	777	+ vnt_adjust_tsf(priv, conf->beacon_rate->hw_value,
	778	+ conf->sync_tsf, priv->current_tsf);
	779	+
	780	+ vnt_update_next_tbtt(priv,
	781	+ conf->sync_tsf, conf->beacon_int);
779	782	} else {
780	783	vnt_clear_current_tsf(priv);
781	784
@@ -809,15 +812,11 @@ static void vnt_configure(struct ieee80211_hw *hw, · 27 unmodified lines
809	812	{
810	813	struct vnt_private *priv = hw->priv;
811	814	u8 rx_mode = 0;
812		− int rc;
813	815
814	816	*total_flags &= FIF_ALLMULTI \| FIF_OTHER_BSS \| FIF_BCN_PRBRESP_PROMISC;
815	817
816		− rc = vnt_control_in(priv, MESSAGE_TYPE_READ, MAC_REG_RCR,
817		− MESSAGE_REQUEST_MACREG, sizeof(u8), &rx_mode);
818		−
819		− if (!rc)
820		− rx_mode = RCR_MULTICAST \| RCR_BROADCAST;
	818	+ vnt_control_in(priv, MESSAGE_TYPE_READ, MAC_REG_RCR,
	819	+ MESSAGE_REQUEST_MACREG, sizeof(u8), &rx_mode);
821	820
822	821	dev_dbg(&priv->usb->dev, "rx mode in = %x\n", rx_mode);
823	822
@@ -856,8 +855,12 @@ static int vnt_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd, · 32 unmodified lines
856	855	case SET_KEY:
857	856	return vnt_set_keys(hw, sta, vif, key);
858	857	case DISABLE_KEY:
859		− if (test_bit(key->hw_key_idx, &priv->key_entry_inuse))
	858	+ if (test_bit(key->hw_key_idx, &priv->key_entry_inuse)) {
860	859	clear_bit(key->hw_key_idx, &priv->key_entry_inuse);
	860	+
	861	+ vnt_mac_disable_keyentry(priv, key->hw_key_idx);
	862	+ }
	863	+
861	864	default:
862	865	break;
863	866	}

drivers/staging/vt6656/usbpipe.c+2−1

@@ -207,7 +207,8 @@ static void vnt_int_process_data(struct vnt_private *priv) · 206 unmodified lines
207	207	priv->wake_up_count =
208	208	priv->hw->conf.listen_interval;
209	209
210		− --priv->wake_up_count;
	210	+ if (priv->wake_up_count)
	211	+ --priv->wake_up_count;
211	212
212	213	/* Turn on wake up to listen next beacon */
213	214	if (priv->wake_up_count == 1)

drivers/target/target_core_iblock.c+1−1

@@ -432,7 +432,7 @@ iblock_execute_zero_out(struct block_device bdev, struct se_cmd cmd) · 431 unmodified lines
432	432	target_to_linux_sector(dev, cmd->t_task_lba),
433	433	target_to_linux_sector(dev,
434	434	sbc_get_write_same_sectors(cmd)),
435		− GFP_KERNEL, false);
	435	+ GFP_KERNEL, BLKDEV_ZERO_NOUNMAP);
436	436	if (ret)
437	437	return TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
438	438

drivers/thunderbolt/usb4.c+3−0

@@ -182,6 +182,9 @@ static int usb4_switch_op(struct tb_switch sw, u16 opcode, u8 status) · 181 unmodified lines
182	182	return ret;
183	183
184	184	ret = tb_sw_read(sw, &val, TB_CFG_SWITCH, ROUTER_CS_26, 1);
	185	+ if (ret)
	186	+ return ret;
	187	+
185	188	if (val & ROUTER_CS_26_ONS)
186	189	return -EOPNOTSUPP;
187	190

drivers/tty/hvc/Kconfig+1−1

@@ -88,7 +88,7 @@ config HVC_DCC · 87 unmodified lines
88	88
89	89	config HVC_RISCV_SBI
90	90	bool "RISC-V SBI console support"
91		− depends on RISCV_SBI
	91	+ depends on RISCV_SBI_V01
92	92	select HVC_DRIVER
93	93	help
94	94	This enables support for console output via RISC-V SBI calls, which

drivers/tty/hvc/hvc_console.c+14−9

@@ -302,10 +302,6 @@ int hvc_instantiate(uint32_t vtermno, int index, const struct hv_ops *ops) · 301 unmodified lines
302	302	vtermnos[index] = vtermno;
303	303	cons_ops[index] = ops;
304	304
305		− /* reserve all indices up to and including this index */
306		− if (last_hvc < index)
307		− last_hvc = index;
308		−
309	305	/* check if we need to re-register the kernel console */
310	306	hvc_check_console(index);
311	307
@@ -960,13 +956,22 @@ struct hvc_struct *hvc_alloc(uint32_t vtermno, int data, · 648 unmodified lines
960	956	cons_ops[i] == hp->ops)
961	957	break;
962	958
963		− /* no matching slot, just use a counter */
964		− if (i >= MAX_NR_HVC_CONSOLES)
965		− i = ++last_hvc;
	959	+ if (i >= MAX_NR_HVC_CONSOLES) {
	960	+
	961	+ /* find 'empty' slot for console */
	962	+ for (i = 0; i < MAX_NR_HVC_CONSOLES && vtermnos[i] != -1; i++) {
	963	+ }
	964	+
	965	+ /* no matching slot, just use a counter */
	966	+ if (i == MAX_NR_HVC_CONSOLES)
	967	+ i = ++last_hvc + MAX_NR_HVC_CONSOLES;
	968	+ }
966	969
967	970	hp->index = i;
968		− cons_ops[i] = ops;
969		− vtermnos[i] = vtermno;
	971	+ if (i < MAX_NR_HVC_CONSOLES) {
	972	+ cons_ops[i] = ops;
	973	+ vtermnos[i] = vtermno;
	974	+ }
970	975
971	976	list_add_tail(&(hp->next), &hvc_structs);
972	977	mutex_unlock(&hvc_structs_mutex);

drivers/tty/rocket.c+14−11

@@ -632,18 +632,21 @@ init_r_port(int board, int aiop, int chan, struct pci_dev *pci_dev) · 631 unmodified lines
632	632	tty_port_init(&info->port);
633	633	info->port.ops = &rocket_port_ops;
634	634	info->flags &= ~ROCKET_MODE_MASK;
635		− switch (pc104[board][line]) {
636		− case 422:
637		− info->flags \|= ROCKET_MODE_RS422;
638		− break;
639		− case 485:
640		− info->flags \|= ROCKET_MODE_RS485;
641		− break;
642		− case 232:
643		− default:
	635	+ if (board < ARRAY_SIZE(pc104) && line < ARRAY_SIZE(pc104_1))
	636	+ switch (pc104[board][line]) {
	637	+ case 422:
	638	+ info->flags \|= ROCKET_MODE_RS422;
	639	+ break;
	640	+ case 485:
	641	+ info->flags \|= ROCKET_MODE_RS485;
	642	+ break;
	643	+ case 232:
	644	+ default:
	645	+ info->flags \|= ROCKET_MODE_RS232;
	646	+ break;
	647	+ }
	648	+ else
644	649	info->flags \|= ROCKET_MODE_RS232;
645		− break;
646		− }
647	650
648	651	info->intmask = RXF_TRIG \| TXFIFO_MT \| SRC_INT \| DELTA_CD \| DELTA_CTS \| DELTA_DSR;
649	652	if (sInitChan(ctlp, &info->channel, aiop, chan) == 0) {

drivers/tty/serial/Kconfig+1−1

@@ -86,7 +86,7 @@ config SERIAL_EARLYCON_ARM_SEMIHOST · 85 unmodified lines
86	86
87	87	config SERIAL_EARLYCON_RISCV_SBI
88	88	bool "Early console using RISC-V SBI"
89		− depends on RISCV_SBI
	89	+ depends on RISCV_SBI_V01
90	90	select SERIAL_CORE
91	91	select SERIAL_CORE_CONSOLE
92	92	select SERIAL_EARLYCON

drivers/tty/serial/owl-uart.c+7−0

@@ -680,6 +680,12 @@ static int owl_uart_probe(struct platform_device *pdev) · 679 unmodified lines
680	680	return PTR_ERR(owl_port->clk);
681	681	}
682	682
	683	+ ret = clk_prepare_enable(owl_port->clk);
	684	+ if (ret) {
	685	+ dev_err(&pdev->dev, "could not enable clk\n");
	686	+ return ret;
	687	+ }
	688	+
683	689	owl_port->port.dev = &pdev->dev;
684	690	owl_port->port.line = pdev->id;
685	691	owl_port->port.type = PORT_OWL;
@@ -712,6 +718,7 @@ static int owl_uart_remove(struct platform_device *pdev) · 26 unmodified lines
712	718
713	719	uart_remove_one_port(&owl_uart_driver, &owl_port->port);
714	720	owl_uart_ports[pdev->id] = NULL;
	721	+ clk_disable_unprepare(owl_port->clk);
715	722
716	723	return 0;
717	724	}

drivers/tty/serial/sh-sci.c+10−3

@@ -870,9 +870,16 @@ static void sci_receive_chars(struct uart_port *port) · 869 unmodified lines
870	870	tty_insert_flip_char(tport, c, TTY_NORMAL);
871	871	} else {
872	872	for (i = 0; i < count; i++) {
873		− char c = serial_port_in(port, SCxRDR);
874		−
875		− status = serial_port_in(port, SCxSR);
	873	+ char c;
	874	+
	875	+ if (port->type == PORT_SCIF \|\|
	876	+ port->type == PORT_HSCIF) {
	877	+ status = serial_port_in(port, SCxSR);
	878	+ c = serial_port_in(port, SCxRDR);
	879	+ } else {
	880	+ c = serial_port_in(port, SCxRDR);
	881	+ status = serial_port_in(port, SCxSR);
	882	+ }
876	883	if (uart_handle_sysrq_char(port, c)) {
877	884	count--; i--;
878	885	continue;

drivers/tty/serial/sunhv.c+3−0

@@ -567,6 +567,9 @@ static int hv_probe(struct platform_device *op) · 566 unmodified lines
567	567	sunserial_console_match(&sunhv_console, op->dev.of_node,
568	568	&sunhv_reg, port->line, false);
569	569
	570	+ /* We need to initialize lock even for non-registered console */
	571	+ spin_lock_init(&port->lock);
	572	+
570	573	err = uart_add_one_port(&sunhv_reg, port);
571	574	if (err)
572	575	goto out_unregister_driver;

drivers/tty/serial/xilinx_uartps.c+50−162

@@ -26,13 +26,15 @@ · 25 unmodified lines
26	26
27	27	#define CDNS_UART_TTY_NAME "ttyPS"
28	28	#define CDNS_UART_NAME "xuartps"
	29	+#define CDNS_UART_MAJOR 0 /* use dynamic node allocation */
	30	+#define CDNS_UART_MINOR 0 /* works best with devtmpfs */
	31	+#define CDNS_UART_NR_PORTS 16
29	32	#define CDNS_UART_FIFO_SIZE 64 /* FIFO size */
30	33	#define CDNS_UART_REGISTER_SPACE 0x1000
31	34	#define TX_TIMEOUT 500000
32	35
33	36	/* Rx Trigger level */
34	37	static int rx_trigger_level = 56;
35		−static int uartps_major;
36	38	module_param(rx_trigger_level, uint, 0444);
37	39	MODULE_PARM_DESC(rx_trigger_level, "Rx trigger level, 1-63 bytes");
38	40
@@ -188,7 +190,6 @@ MODULE_PARM_DESC(rx_timeout, "Rx timeout, 1-255"); · 149 unmodified lines
188	190	* @pclk: APB clock
189	191	* @cdns_uart_driver: Pointer to UART driver
190	192	* @baud: Current baud rate
191		− * @id: Port ID
192	193	* @clk_rate_change_nb: Notifier block for clock changes
193	194	* @quirks: Flags for RXBS support.
194	195	*/
@@ -198,7 +199,6 @@ struct cdns_uart { · 3 unmodified lines
198	199	struct clk *pclk;
199	200	struct uart_driver *cdns_uart_driver;
200	201	unsigned int baud;
201		− int id;
202	202	struct notifier_block clk_rate_change_nb;
203	203	u32 quirks;
204	204	bool cts_override;
@@ -1133,6 +1133,8 @@ static const struct uart_ops cdns_uart_ops = { · 928 unmodified lines
1133	1133	#endif
1134	1134	};
1135	1135
	1136	+static struct uart_driver cdns_uart_uart_driver;
	1137	+
1136	1138	#ifdef CONFIG_SERIAL_XILINX_PS_UART_CONSOLE
1137	1139	/**
1138	1140	* cdns_uart_console_putchar - write the character to the FIFO buffer
@@ -1272,6 +1274,16 @@ static int cdns_uart_console_setup(struct console co, char options) · 133 unmodified lines
1272	1274
1273	1275	return uart_set_options(port, co, baud, parity, bits, flow);
1274	1276	}
	1277	+
	1278	+static struct console cdns_uart_console = {
	1279	+ .name = CDNS_UART_TTY_NAME,
	1280	+ .write = cdns_uart_console_write,
	1281	+ .device = uart_console_device,
	1282	+ .setup = cdns_uart_console_setup,
	1283	+ .flags = CON_PRINTBUFFER,
	1284	+ .index = -1, /* Specified on the cmdline (e.g. console=ttyPS ) */
	1285	+ .data = &cdns_uart_uart_driver,
	1286	+};
1275	1287	#endif /* CONFIG_SERIAL_XILINX_PS_UART_CONSOLE */
1276	1288
1277	1289	#ifdef CONFIG_PM_SLEEP
@@ -1403,89 +1415,8 @@ static const struct of_device_id cdns_uart_of_match[] = { · 125 unmodified lines
1403	1415	};
1404	1416	MODULE_DEVICE_TABLE(of, cdns_uart_of_match);
1405	1417
1406		−/*
1407		− * Maximum number of instances without alias IDs but if there is alias
1408		− * which target "< MAX_UART_INSTANCES" range this ID can't be used.
1409		− */
1410		−#define MAX_UART_INSTANCES 32
1411		−
1412		−/* Stores static aliases list */
1413		−static DECLARE_BITMAP(alias_bitmap, MAX_UART_INSTANCES);
1414		−static int alias_bitmap_initialized;
1415		−
1416		−/* Stores actual bitmap of allocated IDs with alias IDs together */
1417		−static DECLARE_BITMAP(bitmap, MAX_UART_INSTANCES);
1418		−/* Protect bitmap operations to have unique IDs */
1419		−static DEFINE_MUTEX(bitmap_lock);
1420		−
1421		−static int cdns_get_id(struct platform_device *pdev)
1422		−{
1423		− int id, ret;
1424		−
1425		− mutex_lock(&bitmap_lock);
1426		−
1427		− /* Alias list is stable that's why get alias bitmap only once */
1428		− if (!alias_bitmap_initialized) {
1429		− ret = of_alias_get_alias_list(cdns_uart_of_match, "serial",
1430		− alias_bitmap, MAX_UART_INSTANCES);
1431		− if (ret && ret != -EOVERFLOW) {
1432		− mutex_unlock(&bitmap_lock);
1433		− return ret;
1434		− }
1435		−
1436		− alias_bitmap_initialized++;
1437		− }
1438		−
1439		− /* Make sure that alias ID is not taken by instance without alias */
1440		− bitmap_or(bitmap, bitmap, alias_bitmap, MAX_UART_INSTANCES);
1441		−
1442		− dev_dbg(&pdev->dev, "Alias bitmap: %*pb\n",
1443		− MAX_UART_INSTANCES, bitmap);
1444		−
1445		− /* Look for a serialN alias */
1446		− id = of_alias_get_id(pdev->dev.of_node, "serial");
1447		− if (id < 0) {
1448		− dev_warn(&pdev->dev,
1449		− "No serial alias passed. Using the first free id\n");
1450		−
1451		− /*
1452		− * Start with id 0 and check if there is no serial0 alias
1453		− * which points to device which is compatible with this driver.
1454		− * If alias exists then try next free position.
1455		− */
1456		− id = 0;
1457		−
1458		− for (;;) {
1459		− dev_info(&pdev->dev, "Checking id %d\n", id);
1460		− id = find_next_zero_bit(bitmap, MAX_UART_INSTANCES, id);
1461		−
1462		− /* No free empty instance */
1463		− if (id == MAX_UART_INSTANCES) {
1464		− dev_err(&pdev->dev, "No free ID\n");
1465		− mutex_unlock(&bitmap_lock);
1466		− return -EINVAL;
1467		− }
1468		−
1469		− dev_dbg(&pdev->dev, "The empty id is %d\n", id);
1470		− /* Check if ID is empty */
1471		− if (!test_and_set_bit(id, bitmap)) {
1472		− /* Break the loop if bit is taken */
1473		− dev_dbg(&pdev->dev,
1474		− "Selected ID %d allocation passed\n",
1475		− id);
1476		− break;
1477		− }
1478		− dev_dbg(&pdev->dev,
1479		− "Selected ID %d allocation failed\n", id);
1480		− /* if taking bit fails then try next one */
1481		− id++;
1482		− }
1483		− }
1484		−
1485		− mutex_unlock(&bitmap_lock);
1486		−
1487		− return id;
1488		−}
	1418	+/* Temporary variable for storing number of instances */
	1419	+static int instances;
1489	1420
1490	1421	/**
1491	1422	* cdns_uart_probe - Platform driver probe
@@ -1495,16 +1426,11 @@ static int cdns_get_id(struct platform_device *pdev) · 3 unmodified lines
1495	1426	*/
1496	1427	static int cdns_uart_probe(struct platform_device *pdev)
1497	1428	{
1498		− int rc, irq;
	1429	+ int rc, id, irq;
1499	1430	struct uart_port *port;
1500	1431	struct resource *res;
1501	1432	struct cdns_uart *cdns_uart_data;
1502	1433	const struct of_device_id *match;
1503		− struct uart_driver *cdns_uart_uart_driver;
1504		− char *driver_name;
1505		−#ifdef CONFIG_SERIAL_XILINX_PS_UART_CONSOLE
1506		− struct console *cdns_uart_console;
1507		−#endif
1508	1434
1509	1435	cdns_uart_data = devm_kzalloc(&pdev->dev, sizeof(*cdns_uart_data),
1510	1436	GFP_KERNEL);
@@ -1514,64 +1440,36 @@ static int cdns_uart_probe(struct platform_device *pdev) · 3 unmodified lines
1514	1440	if (!port)
1515	1441	return -ENOMEM;
1516	1442
1517		− cdns_uart_uart_driver = devm_kzalloc(&pdev->dev,
1518		− sizeof(*cdns_uart_uart_driver),
1519		− GFP_KERNEL);
1520		− if (!cdns_uart_uart_driver)
1521		− return -ENOMEM;
1522		−
1523		− cdns_uart_data->id = cdns_get_id(pdev);
1524		− if (cdns_uart_data->id < 0)
1525		− return cdns_uart_data->id;
	1443	+ /* Look for a serialN alias */
	1444	+ id = of_alias_get_id(pdev->dev.of_node, "serial");
	1445	+ if (id < 0)
	1446	+ id = 0;
1526	1447
1527		− /* There is a need to use unique driver name */
1528		− driver_name = devm_kasprintf(&pdev->dev, GFP_KERNEL, "%s%d",
1529		− CDNS_UART_NAME, cdns_uart_data->id);
1530		− if (!driver_name) {
1531		− rc = -ENOMEM;
1532		− goto err_out_id;
	1448	+ if (id >= CDNS_UART_NR_PORTS) {
	1449	+ dev_err(&pdev->dev, "Cannot get uart_port structure\n");
	1450	+ return -ENODEV;
1533	1451	}
1534	1452
1535		− cdns_uart_uart_driver->owner = THIS_MODULE;
1536		− cdns_uart_uart_driver->driver_name = driver_name;
1537		− cdns_uart_uart_driver->dev_name = CDNS_UART_TTY_NAME;
1538		− cdns_uart_uart_driver->major = uartps_major;
1539		− cdns_uart_uart_driver->minor = cdns_uart_data->id;
1540		− cdns_uart_uart_driver->nr = 1;
1541		−
	1453	+ if (!cdns_uart_uart_driver.state) {
	1454	+ cdns_uart_uart_driver.owner = THIS_MODULE;
	1455	+ cdns_uart_uart_driver.driver_name = CDNS_UART_NAME;
	1456	+ cdns_uart_uart_driver.dev_name = CDNS_UART_TTY_NAME;
	1457	+ cdns_uart_uart_driver.major = CDNS_UART_MAJOR;
	1458	+ cdns_uart_uart_driver.minor = CDNS_UART_MINOR;
	1459	+ cdns_uart_uart_driver.nr = CDNS_UART_NR_PORTS;
1542	1460	#ifdef CONFIG_SERIAL_XILINX_PS_UART_CONSOLE
1543		− cdns_uart_console = devm_kzalloc(&pdev->dev, sizeof(*cdns_uart_console),
1544		− GFP_KERNEL);
1545		− if (!cdns_uart_console) {
1546		− rc = -ENOMEM;
1547		− goto err_out_id;
1548		− }
1549		−
1550		− strncpy(cdns_uart_console->name, CDNS_UART_TTY_NAME,
1551		− sizeof(cdns_uart_console->name));
1552		− cdns_uart_console->index = cdns_uart_data->id;
1553		− cdns_uart_console->write = cdns_uart_console_write;
1554		− cdns_uart_console->device = uart_console_device;
1555		− cdns_uart_console->setup = cdns_uart_console_setup;
1556		− cdns_uart_console->flags = CON_PRINTBUFFER;
1557		− cdns_uart_console->data = cdns_uart_uart_driver;
1558		− cdns_uart_uart_driver->cons = cdns_uart_console;
	1461	+ cdns_uart_uart_driver.cons = &cdns_uart_console;
	1462	+ cdns_uart_console.index = id;
1559	1463	#endif
1560	1464
1561		− rc = uart_register_driver(cdns_uart_uart_driver);
1562		− if (rc < 0) {
1563		− dev_err(&pdev->dev, "Failed to register driver\n");
1564		− goto err_out_id;
	1465	+ rc = uart_register_driver(&cdns_uart_uart_driver);
	1466	+ if (rc < 0) {
	1467	+ dev_err(&pdev->dev, "Failed to register driver\n");
	1468	+ return rc;
	1469	+ }
1565	1470	}
1566	1471
1567		− cdns_uart_data->cdns_uart_driver = cdns_uart_uart_driver;
1568		−
1569		− /*
1570		− * Setting up proper name_base needs to be done after uart
1571		− * registration because tty_driver structure is not filled.
1572		− * name_base is 0 by default.
1573		− */
1574		− cdns_uart_uart_driver->tty_driver->name_base = cdns_uart_data->id;
	1472	+ cdns_uart_data->cdns_uart_driver = &cdns_uart_uart_driver;
1575	1473
1576	1474	match = of_match_node(cdns_uart_of_match, pdev->dev.of_node);
1577	1475	if (match && match->data) {
@@ -1649,6 +1547,7 @@ static int cdns_uart_probe(struct platform_device *pdev) · 71 unmodified lines
1649	1547	port->ops = &cdns_uart_ops;
1650	1548	port->fifosize = CDNS_UART_FIFO_SIZE;
1651	1549	port->has_sysrq = IS_ENABLED(CONFIG_SERIAL_XILINX_PS_UART_CONSOLE);
	1550	+ port->line = id;
1652	1551
1653	1552	/*
1654	1553	* Register the port.
@@ -1680,7 +1579,7 @@ static int cdns_uart_probe(struct platform_device *pdev) · 25 unmodified lines
1680	1579	console_port = port;
1681	1580	#endif
1682	1581
1683		− rc = uart_add_one_port(cdns_uart_uart_driver, port);
	1582	+ rc = uart_add_one_port(&cdns_uart_uart_driver, port);
1684	1583	if (rc) {
1685	1584	dev_err(&pdev->dev,
1686	1585	"uart_add_one_port() failed; err=%i\n", rc);
@@ -1690,13 +1589,15 @@ static int cdns_uart_probe(struct platform_device *pdev) · 3 unmodified lines
1690	1589	#ifdef CONFIG_SERIAL_XILINX_PS_UART_CONSOLE
1691	1590	/* This is not port which is used for console that's why clean it up */
1692	1591	if (console_port == port &&
1693		− !(cdns_uart_uart_driver->cons->flags & CON_ENABLED))
	1592	+ !(cdns_uart_uart_driver.cons->flags & CON_ENABLED))
1694	1593	console_port = NULL;
1695	1594	#endif
1696	1595
1697		− uartps_major = cdns_uart_uart_driver->tty_driver->major;
1698	1596	cdns_uart_data->cts_override = of_property_read_bool(pdev->dev.of_node,
1699	1597	"cts-override");
	1598	+
	1599	+ instances++;
	1600	+
1700	1601	return 0;
1701	1602
1702	1603	err_out_pm_disable:
@@ -1712,12 +1613,8 @@ err_out_clk_disable: · 9 unmodified lines
1712	1613	err_out_clk_dis_pclk:
1713	1614	clk_disable_unprepare(cdns_uart_data->pclk);
1714	1615	err_out_unregister_driver:
1715		− uart_unregister_driver(cdns_uart_data->cdns_uart_driver);
1716		−err_out_id:
1717		− mutex_lock(&bitmap_lock);
1718		− if (cdns_uart_data->id < MAX_UART_INSTANCES)
1719		− clear_bit(cdns_uart_data->id, bitmap);
1720		− mutex_unlock(&bitmap_lock);
	1616	+ if (!instances)
	1617	+ uart_unregister_driver(cdns_uart_data->cdns_uart_driver);
1721	1618	return rc;
1722	1619	}
1723	1620
@@ -1740,10 +1637,6 @@ static int cdns_uart_remove(struct platform_device *pdev) · 16 unmodified lines
1740	1637	#endif
1741	1638	rc = uart_remove_one_port(cdns_uart_data->cdns_uart_driver, port);
1742	1639	port->mapbase = 0;
1743		− mutex_lock(&bitmap_lock);
1744		− if (cdns_uart_data->id < MAX_UART_INSTANCES)
1745		− clear_bit(cdns_uart_data->id, bitmap);
1746		− mutex_unlock(&bitmap_lock);
1747	1640	clk_disable_unprepare(cdns_uart_data->uartclk);
1748	1641	clk_disable_unprepare(cdns_uart_data->pclk);
1749	1642	pm_runtime_disable(&pdev->dev);
@@ -1756,13 +1649,8 @@ static int cdns_uart_remove(struct platform_device *pdev) · 6 unmodified lines
1756	1649	console_port = NULL;
1757	1650	#endif
1758	1651
1759		− /* If this is last instance major number should be initialized */
1760		− mutex_lock(&bitmap_lock);
1761		− if (bitmap_empty(bitmap, MAX_UART_INSTANCES))
1762		− uartps_major = 0;
1763		− mutex_unlock(&bitmap_lock);
1764		−
1765		− uart_unregister_driver(cdns_uart_data->cdns_uart_driver);
	1652	+ if (!--instances)
	1653	+ uart_unregister_driver(cdns_uart_data->cdns_uart_driver);
1766	1654	return rc;
1767	1655	}
1768	1656

drivers/tty/sysrq.c+2−0

@@ -74,6 +74,7 @@ int sysrq_mask(void) · 73 unmodified lines
74	74	return 1;
75	75	return sysrq_enabled;
76	76	}
	77	+EXPORT_SYMBOL_GPL(sysrq_mask);
77	78
78	79	/*
79	80	* A value of 1 means 'all', other nonzero values are an op mask:
@@ -1058,6 +1059,7 @@ int sysrq_toggle_support(int enable_mask) · 978 unmodified lines
1058	1059
1059	1060	return 0;
1060	1061	}
	1062	+EXPORT_SYMBOL_GPL(sysrq_toggle_support);
1061	1063
1062	1064	static int __sysrq_swap_key_ops(int key, struct sysrq_key_op *insert_op_p,
1063	1065	struct sysrq_key_op *remove_op_p)

drivers/tty/vt/vt.c+10−4

@@ -81,6 +81,7 @@ · 80 unmodified lines
81	81	#include <linux/errno.h>
82	82	#include <linux/kd.h>
83	83	#include <linux/slab.h>
	84	+#include <linux/vmalloc.h>
84	85	#include <linux/major.h>
85	86	#include <linux/mm.h>
86	87	#include <linux/console.h>
@@ -350,7 +351,7 @@ static struct uni_screen *vc_uniscr_alloc(unsigned int cols, unsigned int rows) · 263 unmodified lines
350	351	/* allocate everything in one go */
351	352	memsize = cols * rows * sizeof(char32_t);
352	353	memsize += rows * sizeof(char32_t *);
353		− p = kmalloc(memsize, GFP_KERNEL);
	354	+ p = vmalloc(memsize);
354	355	if (!p)
355	356	return NULL;
356	357
@@ -364,9 +365,14 @@ static struct uni_screen *vc_uniscr_alloc(unsigned int cols, unsigned int rows) · 7 unmodified lines
364	365	return uniscr;
365	366	}
366	367
	368	+static void vc_uniscr_free(struct uni_screen *uniscr)
	369	+{
	370	+ vfree(uniscr);
	371	+}
	372	+
367	373	static void vc_uniscr_set(struct vc_data vc, struct uni_screen new_uniscr)
368	374	{
369		− kfree(vc->vc_uni_screen);
	375	+ vc_uniscr_free(vc->vc_uni_screen);
370	376	vc->vc_uni_screen = new_uniscr;
371	377	}
372	378
@@ -1206,7 +1212,7 @@ static int vc_do_resize(struct tty_struct tty, struct vc_data vc, · 833 unmodified lines
1206	1212	if (new_cols == vc->vc_cols && new_rows == vc->vc_rows)
1207	1213	return 0;
1208	1214
1209		− if (new_screen_size > (4 << 20))
	1215	+ if (new_screen_size > KMALLOC_MAX_SIZE)
1210	1216	return -EINVAL;
1211	1217	newscreen = kzalloc(new_screen_size, GFP_USER);
1212	1218	if (!newscreen)
@@ -1229,7 +1235,7 @@ static int vc_do_resize(struct tty_struct tty, struct vc_data vc, · 16 unmodified lines
1229	1235	err = resize_screen(vc, new_cols, new_rows, user);
1230	1236	if (err) {
1231	1237	kfree(newscreen);
1232		− kfree(new_uniscr);
	1238	+ vc_uniscr_free(new_uniscr);
1233	1239	return err;
1234	1240	}
1235	1241

drivers/usb/chipidea/ci_hdrc_msm.c+1−1

@@ -114,7 +114,7 @@ static int ci_hdrc_msm_notify_event(struct ci_hdrc *ci, unsigned event) · 113 unmodified lines
114	114	hw_write_id_reg(ci, HS_PHY_GENCONFIG_2,
115	115	HS_PHY_ULPI_TX_PKT_EN_CLR_FIX, 0);
116	116
117		− if (!IS_ERR(ci->platdata->vbus_extcon.edev)) {
	117	+ if (!IS_ERR(ci->platdata->vbus_extcon.edev) \|\| ci->role_switch) {
118	118	hw_write_id_reg(ci, HS_PHY_GENCONFIG_2,
119	119	HS_PHY_SESS_VLD_CTRL_EN,
120	120	HS_PHY_SESS_VLD_CTRL_EN);

drivers/usb/class/cdc-acm.c+31−5

@@ -412,9 +412,12 @@ static void acm_ctrl_irq(struct urb *urb) · 411 unmodified lines
412	412
413	413	exit:
414	414	retval = usb_submit_urb(urb, GFP_ATOMIC);
415		− if (retval && retval != -EPERM)
	415	+ if (retval && retval != -EPERM && retval != -ENODEV)
416	416	dev_err(&acm->control->dev,
417	417	"%s - usb_submit_urb failed: %d\n", __func__, retval);
	418	+ else
	419	+ dev_vdbg(&acm->control->dev,
	420	+ "control resubmission terminated %d\n", retval);
418	421	}
419	422
420	423	static int acm_submit_read_urb(struct acm *acm, int index, gfp_t mem_flags)
@@ -430,6 +433,8 @@ static int acm_submit_read_urb(struct acm *acm, int index, gfp_t mem_flags) · 9 unmodified lines
430	433	dev_err(&acm->data->dev,
431	434	"urb %d failed submission with %d\n",
432	435	index, res);
	436	+ } else {
	437	+ dev_vdbg(&acm->data->dev, "intended failure %d\n", res);
433	438	}
434	439	set_bit(index, &acm->read_urbs_free);
435	440	return res;
@@ -471,6 +476,7 @@ static void acm_read_bulk_callback(struct urb *urb) · 35 unmodified lines
471	476	int status = urb->status;
472	477	bool stopped = false;
473	478	bool stalled = false;
	479	+ bool cooldown = false;
474	480
475	481	dev_vdbg(&acm->data->dev, "got urb %d, len %d, status %d\n",
476	482	rb->index, urb->actual_length, status);
@@ -497,6 +503,14 @@ static void acm_read_bulk_callback(struct urb *urb) · 20 unmodified lines
497	503	__func__, status);
498	504	stopped = true;
499	505	break;
	506	+ case -EOVERFLOW:
	507	+ case -EPROTO:
	508	+ dev_dbg(&acm->data->dev,
	509	+ "%s - cooling babbling device\n", __func__);
	510	+ usb_mark_last_busy(acm->dev);
	511	+ set_bit(rb->index, &acm->urbs_in_error_delay);
	512	+ cooldown = true;
	513	+ break;
500	514	default:
501	515	dev_dbg(&acm->data->dev,
502	516	"%s - nonzero urb status received: %d\n",
@@ -518,9 +532,11 @@ static void acm_read_bulk_callback(struct urb *urb) · 15 unmodified lines
518	532	*/
519	533	smp_mb__after_atomic();
520	534
521		− if (stopped \|\| stalled) {
	535	+ if (stopped \|\| stalled \|\| cooldown) {
522	536	if (stalled)
523	537	schedule_work(&acm->work);
	538	+ else if (cooldown)
	539	+ schedule_delayed_work(&acm->dwork, HZ / 2);
524	540	return;
525	541	}
526	542
@@ -557,14 +573,20 @@ static void acm_softint(struct work_struct *work) · 30 unmodified lines
557	573	struct acm *acm = container_of(work, struct acm, work);
558	574
559	575	if (test_bit(EVENT_RX_STALL, &acm->flags)) {
560		− if (!(usb_autopm_get_interface(acm->data))) {
	576	+ smp_mb(); /* against acm_suspend() */
	577	+ if (!acm->susp_count) {
561	578	for (i = 0; i < acm->rx_buflimit; i++)
562	579	usb_kill_urb(acm->read_urbs[i]);
563	580	usb_clear_halt(acm->dev, acm->in);
564	581	acm_submit_read_urbs(acm, GFP_KERNEL);
565		− usb_autopm_put_interface(acm->data);
	582	+ clear_bit(EVENT_RX_STALL, &acm->flags);
566	583	}
567		− clear_bit(EVENT_RX_STALL, &acm->flags);
	584	+ }
	585	+
	586	+ if (test_and_clear_bit(ACM_ERROR_DELAY, &acm->flags)) {
	587	+ for (i = 0; i < ACM_NR; i++)
	588	+ if (test_and_clear_bit(i, &acm->urbs_in_error_delay))
	589	+ acm_submit_read_urb(acm, i, GFP_NOIO);
568	590	}
569	591
570	592	if (test_and_clear_bit(EVENT_TTY_WAKEUP, &acm->flags))
@@ -1333,6 +1355,7 @@ made_compressed_probe: · 762 unmodified lines
1333	1355	acm->readsize = readsize;
1334	1356	acm->rx_buflimit = num_rx_buf;
1335	1357	INIT_WORK(&acm->work, acm_softint);
	1358	+ INIT_DELAYED_WORK(&acm->dwork, acm_softint);
1336	1359	init_waitqueue_head(&acm->wioctl);
1337	1360	spin_lock_init(&acm->write_lock);
1338	1361	spin_lock_init(&acm->read_lock);
@@ -1542,6 +1565,7 @@ static void acm_disconnect(struct usb_interface *intf) · 203 unmodified lines
1542	1565
1543	1566	acm_kill_urbs(acm);
1544	1567	cancel_work_sync(&acm->work);
	1568	+ cancel_delayed_work_sync(&acm->dwork);
1545	1569
1546	1570	tty_unregister_device(acm_tty_driver, acm->minor);
1547	1571
@@ -1584,6 +1608,8 @@ static int acm_suspend(struct usb_interface *intf, pm_message_t message) · 36 unmodified lines
1584	1608
1585	1609	acm_kill_urbs(acm);
1586	1610	cancel_work_sync(&acm->work);
	1611	+ cancel_delayed_work_sync(&acm->dwork);
	1612	+ acm->urbs_in_error_delay = 0;
1587	1613
1588	1614	return 0;
1589	1615	}

drivers/usb/class/cdc-acm.h+4−1

@@ -109,8 +109,11 @@ struct acm { · 108 unmodified lines
109	109	# define EVENT_TTY_WAKEUP 0
110	110	# define EVENT_RX_STALL 1
111	111	# define ACM_THROTTLED 2
	112	+# define ACM_ERROR_DELAY 3
	113	+ unsigned long urbs_in_error_delay; /* these need to be restarted after a delay */
112	114	struct usb_cdc_line_coding line; /* bits, stop, parity */
113		− struct work_struct work; /* work queue entry for line discipline waking up */
	115	+ struct work_struct work; /* work queue entry for various purposes*/
	116	+ struct delayed_work dwork; /* for cool downs needed in error recovery */
114	117	unsigned int ctrlin; /* input control lines (DCD, DSR, RI, break, overruns) */
115	118	unsigned int ctrlout; /* output control lines (DTR, RTS) */
116	119	struct async_icount iocount; /* counters for control line changes */

drivers/usb/core/devio.c+2−3

@@ -217,6 +217,7 @@ static int usbdev_mmap(struct file file, struct vm_area_struct vma) · 216 unmodified lines
217	217	{
218	218	struct usb_memory *usbm = NULL;
219	219	struct usb_dev_state *ps = file->private_data;
	220	+ struct usb_hcd *hcd = bus_to_hcd(ps->dev->bus);
220	221	size_t size = vma->vm_end - vma->vm_start;
221	222	void *mem;
222	223	unsigned long flags;
@@ -250,9 +251,7 @@ static int usbdev_mmap(struct file file, struct vm_area_struct vma) · 27 unmodified lines
250	251	usbm->vma_use_count = 1;
251	252	INIT_LIST_HEAD(&usbm->memlist);
252	253
253		− if (remap_pfn_range(vma, vma->vm_start,
254		− virt_to_phys(usbm->mem) >> PAGE_SHIFT,
255		− size, vma->vm_page_prot) < 0) {
	254	+ if (dma_mmap_coherent(hcd->self.sysdev, vma, mem, dma_handle, size)) {
256	255	dec_usb_memory_use_count(usbm, &usbm->vma_use_count);
257	256	return -EAGAIN;
258	257	}

drivers/usb/core/hub.c+15−3

@@ -1223,6 +1223,11 @@ static void hub_activate(struct usb_hub *hub, enum hub_activation_type type) · 1222 unmodified lines
1223	1223	#ifdef CONFIG_PM
1224	1224	udev->reset_resume = 1;
1225	1225	#endif
	1226	+ /* Don't set the change_bits when the device
	1227	+ * was powered off.
	1228	+ */
	1229	+ if (test_bit(port1, hub->power_bits))
	1230	+ set_bit(port1, hub->change_bits);
1226	1231
1227	1232	} else {
1228	1233	/* The power session is gone; tell hub_wq */
@@ -2723,13 +2728,11 @@ static bool use_new_scheme(struct usb_device *udev, int retry, · 1494 unmodified lines
2723	2728	{
2724	2729	int old_scheme_first_port =
2725	2730	port_dev->quirks & USB_PORT_QUIRK_OLD_SCHEME;
2726		− int quick_enumeration = (udev->speed == USB_SPEED_HIGH);
2727	2731
2728	2732	if (udev->speed >= USB_SPEED_SUPER)
2729	2733	return false;
2730	2734
2731		− return USE_NEW_SCHEME(retry, old_scheme_first_port \|\| old_scheme_first
2732		− \|\| quick_enumeration);
	2735	+ return USE_NEW_SCHEME(retry, old_scheme_first_port \|\| old_scheme_first);
2733	2736	}
2734	2737
2735	2738	/* Is a USB 3.0 port in the Inactive or Compliance Mode state?
@@ -3088,6 +3091,15 @@ static int check_port_resume_type(struct usb_device *udev, · 352 unmodified lines
3088	3091	if (portchange & USB_PORT_STAT_C_ENABLE)
3089	3092	usb_clear_port_feature(hub->hdev, port1,
3090	3093	USB_PORT_FEAT_C_ENABLE);
	3094	+
	3095	+ /*
	3096	+ * Whatever made this reset-resume necessary may have
	3097	+ * turned on the port1 bit in hub->change_bits. But after
	3098	+ * a successful reset-resume we want the bit to be clear;
	3099	+ * if it was on it would indicate that something happened
	3100	+ * following the reset-resume.
	3101	+ */
	3102	+ clear_bit(port1, hub->change_bits);
3091	3103	}
3092	3104
3093	3105	return status;

drivers/usb/core/message.c+10−3

@@ -589,12 +589,13 @@ void usb_sg_cancel(struct usb_sg_request *io) · 588 unmodified lines
589	589	int i, retval;
590	590
591	591	spin_lock_irqsave(&io->lock, flags);
592		− if (io->status) {
	592	+ if (io->status \|\| io->count == 0) {
593	593	spin_unlock_irqrestore(&io->lock, flags);
594	594	return;
595	595	}
596	596	/* shut everything down */
597	597	io->status = -ECONNRESET;
	598	+ io->count++; /* Keep the request alive until we're done */
598	599	spin_unlock_irqrestore(&io->lock, flags);
599	600
600	601	for (i = io->entries - 1; i >= 0; --i) {
@@ -608,6 +609,12 @@ void usb_sg_cancel(struct usb_sg_request *io) · 7 unmodified lines
608	609	dev_warn(&io->dev->dev, "%s, unlink --> %d\n",
609	610	__func__, retval);
610	611	}
	612	+
	613	+ spin_lock_irqsave(&io->lock, flags);
	614	+ io->count--;
	615	+ if (!io->count)
	616	+ complete(&io->complete);
	617	+ spin_unlock_irqrestore(&io->lock, flags);
611	618	}
612	619	EXPORT_SYMBOL_GPL(usb_sg_cancel);
613	620
@@ -1137,11 +1144,11 @@ void usb_disable_endpoint(struct usb_device *dev, unsigned int epaddr, · 523 unmodified lines
1137	1144
1138	1145	if (usb_endpoint_out(epaddr)) {
1139	1146	ep = dev->ep_out[epnum];
1140		− if (reset_hardware)
	1147	+ if (reset_hardware && epnum != 0)
1141	1148	dev->ep_out[epnum] = NULL;
1142	1149	} else {
1143	1150	ep = dev->ep_in[epnum];
1144		− if (reset_hardware)
	1151	+ if (reset_hardware && epnum != 0)
1145	1152	dev->ep_in[epnum] = NULL;
1146	1153	}
1147	1154	if (ep) {

drivers/usb/core/quirks.c+4−0

@@ -430,6 +430,10 @@ static const struct usb_device_id usb_quirk_list[] = { · 429 unmodified lines
430	430	/* Corsair K70 LUX */
431	431	{ USB_DEVICE(0x1b1c, 0x1b36), .driver_info = USB_QUIRK_DELAY_INIT },
432	432
	433	+ /* Corsair K70 RGB RAPDIFIRE */
	434	+ { USB_DEVICE(0x1b1c, 0x1b38), .driver_info = USB_QUIRK_DELAY_INIT \|
	435	+ USB_QUIRK_DELAY_CTRL_MSG },
	436	+
433	437	/* MIDI keyboard WORLDE MINI */
434	438	{ USB_DEVICE(0x1c75, 0x0204), .driver_info =
435	439	USB_QUIRK_CONFIG_INTF_STRINGS },

drivers/usb/dwc3/core.h+6−2

@@ -307,10 +307,14 @@ · 306 unmodified lines
307	307
308	308	/* Global TX Fifo Size Register */
309	309	#define DWC31_GTXFIFOSIZ_TXFRAMNUM BIT(15) /* DWC_usb31 only */
310		−#define DWC31_GTXFIFOSIZ_TXFDEF(n) ((n) & 0x7fff) /* DWC_usb31 only */
311		−#define DWC3_GTXFIFOSIZ_TXFDEF(n) ((n) & 0xffff)
	310	+#define DWC31_GTXFIFOSIZ_TXFDEP(n) ((n) & 0x7fff) /* DWC_usb31 only */
	311	+#define DWC3_GTXFIFOSIZ_TXFDEP(n) ((n) & 0xffff)
312	312	#define DWC3_GTXFIFOSIZ_TXFSTADDR(n) ((n) & 0xffff0000)
313	313
	314	+/* Global RX Fifo Size Register */
	315	+#define DWC31_GRXFIFOSIZ_RXFDEP(n) ((n) & 0x7fff) /* DWC_usb31 only */
	316	+#define DWC3_GRXFIFOSIZ_RXFDEP(n) ((n) & 0xffff)
	317	+
314	318	/* Global Event Size Registers */
315	319	#define DWC3_GEVNTSIZ_INTMASK BIT(31)
316	320	#define DWC3_GEVNTSIZ_SIZE(n) ((n) & 0xffff)

drivers/usb/dwc3/gadget.c+47−29

@@ -1728,7 +1728,6 @@ static int __dwc3_gadget_wakeup(struct dwc3 *dwc) · 1727 unmodified lines
1728	1728	u32 reg;
1729	1729
1730	1730	u8 link_state;
1731		− u8 speed;
1732	1731
1733	1732	/*
1734	1733	* According to the Databook Remote wakeup request should
@@ -1738,16 +1737,13 @@ static int __dwc3_gadget_wakeup(struct dwc3 *dwc) · 3 unmodified lines
1738	1737	*/
1739	1738	reg = dwc3_readl(dwc->regs, DWC3_DSTS);
1740	1739
1741		− speed = reg & DWC3_DSTS_CONNECTSPD;
1742		− if ((speed == DWC3_DSTS_SUPERSPEED) \|\|
1743		− (speed == DWC3_DSTS_SUPERSPEED_PLUS))
1744		− return 0;
1745		−
1746	1740	link_state = DWC3_DSTS_USBLNKST(reg);
1747	1741
1748	1742	switch (link_state) {
	1743	+ case DWC3_LINK_STATE_RESET:
1749	1744	case DWC3_LINK_STATE_RX_DET: /* in HS, means Early Suspend */
1750	1745	case DWC3_LINK_STATE_U3: /* in HS, means SUSPEND */
	1746	+ case DWC3_LINK_STATE_RESUME:
1751	1747	break;
1752	1748	default:
1753	1749	return -EINVAL;
@@ -2227,7 +2223,6 @@ static int dwc3_gadget_init_in_endpoint(struct dwc3_ep *dep) · 473 unmodified lines
2227	2223	{
2228	2224	struct dwc3 *dwc = dep->dwc;
2229	2225	int mdwidth;
2230		− int kbytes;
2231	2226	int size;
2232	2227
2233	2228	mdwidth = DWC3_MDWIDTH(dwc->hwparams.hwparams0);
@@ -2236,24 +2231,24 @@ static int dwc3_gadget_init_in_endpoint(struct dwc3_ep *dep) · 2 unmodified lines
2236	2231
2237	2232	size = dwc3_readl(dwc->regs, DWC3_GTXFIFOSIZ(dep->number >> 1));
2238	2233	if (dwc3_is_usb31(dwc))
2239		− size = DWC31_GTXFIFOSIZ_TXFDEF(size);
	2234	+ size = DWC31_GTXFIFOSIZ_TXFDEP(size);
2240	2235	else
2241		− size = DWC3_GTXFIFOSIZ_TXFDEF(size);
	2236	+ size = DWC3_GTXFIFOSIZ_TXFDEP(size);
2242	2237
2243	2238	/* FIFO Depth is in MDWDITH bytes. Multiply */
2244	2239	size *= mdwidth;
2245	2240
2246		− kbytes = size / 1024;
2247		− if (kbytes == 0)
2248		− kbytes = 1;
2249		−
2250	2241	/*
2251		− * FIFO sizes account an extra MDWIDTH * (kbytes + 1) bytes for
2252		− * internal overhead. We don't really know how these are used,
2253		− * but documentation say it exists.
	2242	+ * To meet performance requirement, a minimum TxFIFO size of 3x
	2243	+ * MaxPacketSize is recommended for endpoints that support burst and a
	2244	+ * minimum TxFIFO size of 2x MaxPacketSize for endpoints that don't
	2245	+ * support burst. Use those numbers and we can calculate the max packet
	2246	+ * limit as below.
2254	2247	*/
2255		− size -= mdwidth * (kbytes + 1);
2256		− size /= kbytes;
	2248	+ if (dwc->maximum_speed >= USB_SPEED_SUPER)
	2249	+ size /= 3;
	2250	+ else
	2251	+ size /= 2;
2257	2252
2258	2253	usb_ep_set_maxpacket_limit(&dep->endpoint, size);
2259	2254
@@ -2271,8 +2266,39 @@ static int dwc3_gadget_init_in_endpoint(struct dwc3_ep *dep) · 11 unmodified lines
2271	2266	static int dwc3_gadget_init_out_endpoint(struct dwc3_ep *dep)
2272	2267	{
2273	2268	struct dwc3 *dwc = dep->dwc;
	2269	+ int mdwidth;
	2270	+ int size;
	2271	+
	2272	+ mdwidth = DWC3_MDWIDTH(dwc->hwparams.hwparams0);
2274	2273
2275		− usb_ep_set_maxpacket_limit(&dep->endpoint, 1024);
	2274	+ /* MDWIDTH is represented in bits, convert to bytes */
	2275	+ mdwidth /= 8;
	2276	+
	2277	+ /* All OUT endpoints share a single RxFIFO space */
	2278	+ size = dwc3_readl(dwc->regs, DWC3_GRXFIFOSIZ(0));
	2279	+ if (dwc3_is_usb31(dwc))
	2280	+ size = DWC31_GRXFIFOSIZ_RXFDEP(size);
	2281	+ else
	2282	+ size = DWC3_GRXFIFOSIZ_RXFDEP(size);
	2283	+
	2284	+ /* FIFO depth is in MDWDITH bytes */
	2285	+ size *= mdwidth;
	2286	+
	2287	+ /*
	2288	+ * To meet performance requirement, a minimum recommended RxFIFO size
	2289	+ * is defined as follow:
	2290	+ * RxFIFO size >= (3 x MaxPacketSize) +
	2291	+ * (3 x 8 bytes setup packets size) + (16 bytes clock crossing margin)
	2292	+ *
	2293	+ * Then calculate the max packet limit as below.
	2294	+ */
	2295	+ size -= (3 * 8) + 16;
	2296	+ if (size < 0)
	2297	+ size = 0;
	2298	+ else
	2299	+ size /= 3;
	2300	+
	2301	+ usb_ep_set_maxpacket_limit(&dep->endpoint, size);
2276	2302	dep->endpoint.max_streams = 15;
2277	2303	dep->endpoint.ops = &dwc3_gadget_ep_ops;
2278	2304	list_add_tail(&dep->endpoint.ep_list,
@@ -2484,14 +2510,7 @@ static int dwc3_gadget_ep_reclaim_trb_linear(struct dwc3_ep *dep, · 205 unmodified lines
2484	2510
2485	2511	static bool dwc3_gadget_ep_request_completed(struct dwc3_request *req)
2486	2512	{
2487		− /*
2488		− * For OUT direction, host may send less than the setup
2489		− * length. Return true for all OUT requests.
2490		− */
2491		− if (!req->direction)
2492		− return true;
2493		−
2494		− return req->request.actual == req->request.length;
	2513	+ return req->num_pending_sgs == 0;
2495	2514	}
2496	2515
2497	2516	static int dwc3_gadget_ep_cleanup_completed_request(struct dwc3_ep *dep,
@@ -2515,8 +2534,7 @@ static int dwc3_gadget_ep_cleanup_completed_request(struct dwc3_ep *dep, · 17 unmodified lines
2515	2534
2516	2535	req->request.actual = req->request.length - req->remaining;
2517	2536
2518		− if (!dwc3_gadget_ep_request_completed(req) \|\|
2519		− req->num_pending_sgs) {
	2537	+ if (!dwc3_gadget_ep_request_completed(req)) {
2520	2538	__dwc3_gadget_kick_transfer(dep);
2521	2539	goto out;
2522	2540	}

drivers/usb/early/xhci-dbc.c+4−4

@@ -728,19 +728,19 @@ static void xdbc_handle_tx_event(struct xdbc_trb *evt_trb) · 727 unmodified lines
728	728	case COMP_USB_TRANSACTION_ERROR:
729	729	case COMP_STALL_ERROR:
730	730	default:
731		− if (ep_id == XDBC_EPID_OUT)
	731	+ if (ep_id == XDBC_EPID_OUT \|\| ep_id == XDBC_EPID_OUT_INTEL)
732	732	xdbc.flags \|= XDBC_FLAGS_OUT_STALL;
733		− if (ep_id == XDBC_EPID_IN)
	733	+ if (ep_id == XDBC_EPID_IN \|\| ep_id == XDBC_EPID_IN_INTEL)
734	734	xdbc.flags \|= XDBC_FLAGS_IN_STALL;
735	735
736	736	xdbc_trace("endpoint %d stalled\n", ep_id);
737	737	break;
738	738	}
739	739
740		− if (ep_id == XDBC_EPID_IN) {
	740	+ if (ep_id == XDBC_EPID_IN \|\| ep_id == XDBC_EPID_IN_INTEL) {
741	741	xdbc.flags &= ~XDBC_FLAGS_IN_PROCESS;
742	742	xdbc_bulk_transfer(NULL, XDBC_MAX_PACKET, true);
743		− } else if (ep_id == XDBC_EPID_OUT) {
	743	+ } else if (ep_id == XDBC_EPID_OUT \|\| ep_id == XDBC_EPID_OUT_INTEL) {
744	744	xdbc.flags &= ~XDBC_FLAGS_OUT_PROCESS;
745	745	} else {
746	746	xdbc_trace("invalid endpoint id %d\n", ep_id);

drivers/usb/early/xhci-dbc.h+16−2

@@ -120,8 +120,22 @@ struct xdbc_ring { · 119 unmodified lines
120	120	u32 cycle_state;
121	121	};
122	122
123		−#define XDBC_EPID_OUT 2
124		−#define XDBC_EPID_IN 3
	123	+/*
	124	+ * These are the "Endpoint ID" (also known as "Context Index") values for the
	125	+ * OUT Transfer Ring and the IN Transfer Ring of a Debug Capability Context data
	126	+ * structure.
	127	+ * According to the "eXtensible Host Controller Interface for Universal Serial
	128	+ * Bus (xHCI)" specification, section "7.6.3.2 Endpoint Contexts and Transfer
	129	+ * Rings", these should be 0 and 1, and those are the values AMD machines give
	130	+ * you; but Intel machines seem to use the formula from section "4.5.1 Device
	131	+ * Context Index", which is supposed to be used for the Device Context only.
	132	+ * Luckily the values from Intel don't overlap with those from AMD, so we can
	133	+ * just test for both.
	134	+ */
	135	+#define XDBC_EPID_OUT 0
	136	+#define XDBC_EPID_IN 1
	137	+#define XDBC_EPID_OUT_INTEL 2
	138	+#define XDBC_EPID_IN_INTEL 3
125	139
126	140	struct xdbc_state {
127	141	u16 vendor;

drivers/usb/gadget/function/f_fs.c+4−0

@@ -1813,6 +1813,10 @@ static void ffs_data_reset(struct ffs_data *ffs) · 1812 unmodified lines
1813	1813	ffs->state = FFS_READ_DESCRIPTORS;
1814	1814	ffs->setup_state = FFS_NO_SETUP;
1815	1815	ffs->flags = 0;
	1816	+
	1817	+ ffs->ms_os_descs_ext_prop_count = 0;
	1818	+ ffs->ms_os_descs_ext_prop_name_len = 0;
	1819	+ ffs->ms_os_descs_ext_prop_data_len = 0;
1816	1820	}
1817	1821
1818	1822

drivers/usb/gadget/legacy/raw_gadget.c+42−29

@@ -81,6 +81,7 @@ static int raw_event_queue_add(struct raw_event_queue *queue, · 80 unmodified lines
81	81	static struct usb_raw_event *raw_event_queue_fetch(
82	82	struct raw_event_queue *queue)
83	83	{
	84	+ int ret;
84	85	unsigned long flags;
85	86	struct usb_raw_event *event;
86	87
@@ -89,11 +90,18 @@ static struct usb_raw_event *raw_event_queue_fetch( · 2 unmodified lines
89	90	* there's at least one event queued by decrementing the semaphore,
90	91	* and then take the lock to protect queue struct fields.
91	92	*/
92		− if (down_interruptible(&queue->sema))
93		− return NULL;
	93	+ ret = down_interruptible(&queue->sema);
	94	+ if (ret)
	95	+ return ERR_PTR(ret);
94	96	spin_lock_irqsave(&queue->lock, flags);
95		− if (WARN_ON(!queue->size))
96		− return NULL;
	97	+ /*
	98	+ * queue->size must have the same value as queue->sema counter (before
	99	+ * the down_interruptible() call above), so this check is a fail-safe.
	100	+ */
	101	+ if (WARN_ON(!queue->size)) {
	102	+ spin_unlock_irqrestore(&queue->lock, flags);
	103	+ return ERR_PTR(-ENODEV);
	104	+ }
97	105	event = queue->events[0];
98	106	queue->size--;
99	107	memmove(&queue->events[0], &queue->events[1],
@@ -392,9 +400,8 @@ static int raw_ioctl_init(struct raw_dev *dev, unsigned long value) · 292 unmodified lines
392	400	char *udc_device_name;
393	401	unsigned long flags;
394	402
395		− ret = copy_from_user(&arg, (void __user *)value, sizeof(arg));
396		− if (ret)
397		− return ret;
	403	+ if (copy_from_user(&arg, (void __user *)value, sizeof(arg)))
	404	+ return -EFAULT;
398	405
399	406	switch (arg.speed) {
400	407	case USB_SPEED_UNKNOWN:
@@ -501,15 +508,13 @@ out_unlock: · 100 unmodified lines
501	508
502	509	static int raw_ioctl_event_fetch(struct raw_dev *dev, unsigned long value)
503	510	{
504		− int ret = 0;
505	511	struct usb_raw_event arg;
506	512	unsigned long flags;
507	513	struct usb_raw_event *event;
508	514	uint32_t length;
509	515
510		− ret = copy_from_user(&arg, (void __user *)value, sizeof(arg));
511		− if (ret)
512		− return ret;
	516	+ if (copy_from_user(&arg, (void __user *)value, sizeof(arg)))
	517	+ return -EFAULT;
513	518
514	519	spin_lock_irqsave(&dev->lock, flags);
515	520	if (dev->state != STATE_DEV_RUNNING) {
@@ -525,25 +530,31 @@ static int raw_ioctl_event_fetch(struct raw_dev *dev, unsigned long value) · 9 unmodified lines
525	530	spin_unlock_irqrestore(&dev->lock, flags);
526	531
527	532	event = raw_event_queue_fetch(&dev->queue);
528		− if (!event) {
	533	+ if (PTR_ERR(event) == -EINTR) {
529	534	dev_dbg(&dev->gadget->dev, "event fetching interrupted\n");
530	535	return -EINTR;
531	536	}
	537	+ if (IS_ERR(event)) {
	538	+ dev_err(&dev->gadget->dev, "failed to fetch event\n");
	539	+ spin_lock_irqsave(&dev->lock, flags);
	540	+ dev->state = STATE_DEV_FAILED;
	541	+ spin_unlock_irqrestore(&dev->lock, flags);
	542	+ return -ENODEV;
	543	+ }
532	544	length = min(arg.length, event->length);
533		− ret = copy_to_user((void __user *)value, event,
534		− sizeof(*event) + length);
535		− return ret;
	545	+ if (copy_to_user((void __user )value, event, sizeof(event) + length))
	546	+ return -EFAULT;
	547	+
	548	+ return 0;
536	549	}
537	550
538	551	static void raw_alloc_io_data(struct usb_raw_ep_io io, void __user *ptr,
539	552	bool get_from_user)
540	553	{
541		− int ret;
542	554	void *data;
543	555
544		− ret = copy_from_user(io, ptr, sizeof(*io));
545		− if (ret)
546		− return ERR_PTR(ret);
	556	+ if (copy_from_user(io, ptr, sizeof(*io)))
	557	+ return ERR_PTR(-EFAULT);
547	558	if (io->ep >= USB_RAW_MAX_ENDPOINTS)
548	559	return ERR_PTR(-EINVAL);
549	560	if (!usb_raw_io_flags_valid(io->flags))
@@ -658,12 +669,13 @@ static int raw_ioctl_ep0_read(struct raw_dev *dev, unsigned long value) · 108 unmodified lines
658	669	if (IS_ERR(data))
659	670	return PTR_ERR(data);
660	671	ret = raw_process_ep0_io(dev, &io, data, false);
661		− if (ret < 0) {
662		− kfree(data);
663		− return ret;
664		− }
	672	+ if (ret)
	673	+ goto free;
	674	+
665	675	length = min(io.length, (unsigned int)ret);
666		− ret = copy_to_user((void __user *)(value + sizeof(io)), data, length);
	676	+ if (copy_to_user((void __user *)(value + sizeof(io)), data, length))
	677	+ ret = -EFAULT;
	678	+free:
667	679	kfree(data);
668	680	return ret;
669	681	}
@@ -952,12 +964,13 @@ static int raw_ioctl_ep_read(struct raw_dev *dev, unsigned long value) · 282 unmodified lines
952	964	if (IS_ERR(data))
953	965	return PTR_ERR(data);
954	966	ret = raw_process_ep_io(dev, &io, data, false);
955		− if (ret < 0) {
956		− kfree(data);
957		− return ret;
958		− }
	967	+ if (ret)
	968	+ goto free;
	969	+
959	970	length = min(io.length, (unsigned int)ret);
960		− ret = copy_to_user((void __user *)(value + sizeof(io)), data, length);
	971	+ if (copy_to_user((void __user *)(value + sizeof(io)), data, length))
	972	+ ret = -EFAULT;
	973	+free:
961	974	kfree(data);
962	975	return ret;
963	976	}

drivers/usb/gadget/udc/atmel_usba_udc.c+2−2

@@ -1951,10 +1951,10 @@ static irqreturn_t usba_vbus_irq_thread(int irq, void *devid) · 1950 unmodified lines
1951	1951	usba_start(udc);
1952	1952	} else {
1953	1953	udc->suspended = false;
1954		− usba_stop(udc);
1955		−
1956	1954	if (udc->driver->disconnect)
1957	1955	udc->driver->disconnect(&udc->gadget);
	1956	+
	1957	+ usba_stop(udc);
1958	1958	}
1959	1959	udc->vbus_prev = vbus;
1960	1960	}

drivers/usb/gadget/udc/bdc/bdc_ep.c+1−1

@@ -540,7 +540,7 @@ static void bdc_req_complete(struct bdc_ep ep, struct bdc_req req, · 539 unmodified lines
540	540	{
541	541	struct bdc *bdc = ep->bdc;
542	542
543		− if (req == NULL \|\| &req->queue == NULL \|\| &req->usb_req == NULL)
	543	+ if (req == NULL)
544	544	return;
545	545
546	546	dev_dbg(bdc->dev, "%s ep:%s status:%d\n", __func__, ep->name, status);

drivers/usb/host/xhci-hub.c+9−0

@@ -1571,6 +1571,8 @@ int xhci_hub_status_data(struct usb_hcd hcd, char buf) · 1570 unmodified lines
1571	1571	}
1572	1572	if ((temp & PORT_RC))
1573	1573	reset_change = true;
	1574	+ if (temp & PORT_OC)
	1575	+ status = 1;
1574	1576	}
1575	1577	if (!status && !reset_change) {
1576	1578	xhci_dbg(xhci, "%s: stopping port polling.\n", __func__);
@@ -1636,6 +1638,13 @@ retry: · 59 unmodified lines
1636	1638	port_index);
1637	1639	goto retry;
1638	1640	}
	1641	+ /* bail out if port detected a over-current condition */
	1642	+ if (t1 & PORT_OC) {
	1643	+ bus_state->bus_suspended = 0;
	1644	+ spin_unlock_irqrestore(&xhci->lock, flags);
	1645	+ xhci_dbg(xhci, "Bus suspend bailout, port over-current detected\n");
	1646	+ return -EBUSY;
	1647	+ }
1639	1648	/* suspend ports in U0, or bail out for new connect changes */
1640	1649	if ((t1 & PORT_PE) && (t1 & PORT_PLS_MASK) == XDEV_U0) {
1641	1650	if ((t1 & PORT_CSC) && wake_enabled) {

drivers/usb/host/xhci-ring.c+40−6

@@ -547,6 +547,23 @@ void xhci_find_new_dequeue_state(struct xhci_hcd *xhci, · 546 unmodified lines
547	547	stream_id);
548	548	return;
549	549	}
	550	+ /*
	551	+ * A cancelled TD can complete with a stall if HW cached the trb.
	552	+ * In this case driver can't find cur_td, but if the ring is empty we
	553	+ * can move the dequeue pointer to the current enqueue position.
	554	+ */
	555	+ if (!cur_td) {
	556	+ if (list_empty(&ep_ring->td_list)) {
	557	+ state->new_deq_seg = ep_ring->enq_seg;
	558	+ state->new_deq_ptr = ep_ring->enqueue;
	559	+ state->new_cycle_state = ep_ring->cycle_state;
	560	+ goto done;
	561	+ } else {
	562	+ xhci_warn(xhci, "Can't find new dequeue state, missing cur_td\n");
	563	+ return;
	564	+ }
	565	+ }
	566	+
550	567	/* Dig out the cycle state saved by the xHC during the stop ep cmd */
551	568	xhci_dbg_trace(xhci, trace_xhci_dbg_cancel_urb,
552	569	"Finding endpoint context");
@@ -592,6 +609,7 @@ void xhci_find_new_dequeue_state(struct xhci_hcd *xhci, · 39 unmodified lines
592	609	state->new_deq_seg = new_seg;
593	610	state->new_deq_ptr = new_deq;
594	611
	612	+done:
595	613	/* Don't update the ring cycle state for the producer (us). */
596	614	xhci_dbg_trace(xhci, trace_xhci_dbg_cancel_urb,
597	615	"Cycle state = 0x%x", state->new_cycle_state);
@@ -1856,8 +1874,8 @@ static void xhci_cleanup_halted_endpoint(struct xhci_hcd *xhci, · 1258 unmodified lines
1856	1874
1857	1875	if (reset_type == EP_HARD_RESET) {
1858	1876	ep->ep_state \|= EP_HARD_CLEAR_TOGGLE;
1859		− xhci_cleanup_stalled_ring(xhci, ep_index, stream_id, td);
1860		− xhci_clear_hub_tt_buffer(xhci, td, ep);
	1877	+ xhci_cleanup_stalled_ring(xhci, slot_id, ep_index, stream_id,
	1878	+ td);
1861	1879	}
1862	1880	xhci_ring_cmd_db(xhci);
1863	1881	}
@@ -1978,11 +1996,18 @@ static int finish_td(struct xhci_hcd xhci, struct xhci_td td, · 114 unmodified lines
1978	1996	if (trb_comp_code == COMP_STALL_ERROR \|\|
1979	1997	xhci_requires_manual_halt_cleanup(xhci, ep_ctx,
1980	1998	trb_comp_code)) {
1981		− /* Issue a reset endpoint command to clear the host side
1982		− * halt, followed by a set dequeue command to move the
1983		− * dequeue pointer past the TD.
1984		− * The class driver clears the device side halt later.
	1999	+ /*
	2000	+ * xhci internal endpoint state will go to a "halt" state for
	2001	+ * any stall, including default control pipe protocol stall.
	2002	+ * To clear the host side halt we need to issue a reset endpoint
	2003	+ * command, followed by a set dequeue command to move past the
	2004	+ * TD.
	2005	+ * Class drivers clear the device side halt from a functional
	2006	+ * stall later. Hub TT buffer should only be cleared for FS/LS
	2007	+ * devices behind HS hubs for functional stalls.
1985	2008	*/
	2009	+ if ((ep_index != 0) \|\| (trb_comp_code != COMP_STALL_ERROR))
	2010	+ xhci_clear_hub_tt_buffer(xhci, td, ep);
1986	2011	xhci_cleanup_halted_endpoint(xhci, slot_id, ep_index,
1987	2012	ep_ring->stream_id, td, EP_HARD_RESET);
1988	2013	} else {
@@ -2539,6 +2564,15 @@ static int handle_tx_event(struct xhci_hcd *xhci, · 550 unmodified lines
2539	2564	xhci_dbg(xhci, "td_list is empty while skip flag set. Clear skip flag for slot %u ep %u.\n",
2540	2565	slot_id, ep_index);
2541	2566	}
	2567	+ if (trb_comp_code == COMP_STALL_ERROR \|\|
	2568	+ xhci_requires_manual_halt_cleanup(xhci, ep_ctx,
	2569	+ trb_comp_code)) {
	2570	+ xhci_cleanup_halted_endpoint(xhci, slot_id,
	2571	+ ep_index,
	2572	+ ep_ring->stream_id,
	2573	+ NULL,
	2574	+ EP_HARD_RESET);
	2575	+ }
2542	2576	goto cleanup;
2543	2577	}
2544	2578

drivers/usb/host/xhci.c+7−7

@@ -3031,19 +3031,19 @@ static void xhci_setup_input_ctx_for_quirk(struct xhci_hcd *xhci, · 3030 unmodified lines
3031	3031	added_ctxs, added_ctxs);
3032	3032	}
3033	3033
3034		−void xhci_cleanup_stalled_ring(struct xhci_hcd *xhci, unsigned int ep_index,
3035		− unsigned int stream_id, struct xhci_td *td)
	3034	+void xhci_cleanup_stalled_ring(struct xhci_hcd *xhci, unsigned int slot_id,
	3035	+ unsigned int ep_index, unsigned int stream_id,
	3036	+ struct xhci_td *td)
3036	3037	{
3037	3038	struct xhci_dequeue_state deq_state;
3038		− struct usb_device *udev = td->urb->dev;
3039	3039
3040	3040	xhci_dbg_trace(xhci, trace_xhci_dbg_reset_ep,
3041	3041	"Cleaning up stalled endpoint ring");
3042	3042	/* We need to move the HW's dequeue pointer past this TD,
3043	3043	* or it will attempt to resend it on the next doorbell ring.
3044	3044	*/
3045		− xhci_find_new_dequeue_state(xhci, udev->slot_id,
3046		− ep_index, stream_id, td, &deq_state);
	3045	+ xhci_find_new_dequeue_state(xhci, slot_id, ep_index, stream_id, td,
	3046	+ &deq_state);
3047	3047
3048	3048	if (!deq_state.new_deq_ptr \|\| !deq_state.new_deq_seg)
3049	3049	return;
@@ -3054,7 +3054,7 @@ void xhci_cleanup_stalled_ring(struct xhci_hcd *xhci, unsigned int ep_index, · 4 unmodified lines
3054	3054	if (!(xhci->quirks & XHCI_RESET_EP_QUIRK)) {
3055	3055	xhci_dbg_trace(xhci, trace_xhci_dbg_reset_ep,
3056	3056	"Queueing new dequeue state");
3057		− xhci_queue_new_dequeue_state(xhci, udev->slot_id,
	3057	+ xhci_queue_new_dequeue_state(xhci, slot_id,
3058	3058	ep_index, &deq_state);
3059	3059	} else {
3060	3060	/* Better hope no one uses the input context between now and the
@@ -3065,7 +3065,7 @@ void xhci_cleanup_stalled_ring(struct xhci_hcd *xhci, unsigned int ep_index, · 4 unmodified lines
3065	3065	xhci_dbg_trace(xhci, trace_xhci_dbg_quirks,
3066	3066	"Setting up input context for "
3067	3067	"configure endpoint command");
3068		− xhci_setup_input_ctx_for_quirk(xhci, udev->slot_id,
	3068	+ xhci_setup_input_ctx_for_quirk(xhci, slot_id,
3069	3069	ep_index, &deq_state);
3070	3070	}
3071	3071	}

drivers/usb/host/xhci.h+3−2

@@ -2116,8 +2116,9 @@ void xhci_find_new_dequeue_state(struct xhci_hcd *xhci, · 2115 unmodified lines
2116	2116	void xhci_queue_new_dequeue_state(struct xhci_hcd *xhci,
2117	2117	unsigned int slot_id, unsigned int ep_index,
2118	2118	struct xhci_dequeue_state *deq_state);
2119		−void xhci_cleanup_stalled_ring(struct xhci_hcd *xhci, unsigned int ep_index,
2120		− unsigned int stream_id, struct xhci_td *td);
	2119	+void xhci_cleanup_stalled_ring(struct xhci_hcd *xhci, unsigned int slot_id,
	2120	+ unsigned int ep_index, unsigned int stream_id,
	2121	+ struct xhci_td *td);
2121	2122	void xhci_stop_endpoint_command_watchdog(struct timer_list *t);
2122	2123	void xhci_handle_command_timeout(struct work_struct *work);
2123	2124

drivers/usb/misc/sisusbvga/sisusb.c+10−10

@@ -1199,18 +1199,18 @@ static int sisusb_read_mem_bulk(struct sisusb_usb_data *sisusb, u32 addr, · 1198 unmodified lines
1199	1199	/* High level: Gfx (indexed) register access */
1200	1200
1201	1201	#ifdef CONFIG_USB_SISUSBVGA_CON
1202		−int sisusb_setreg(struct sisusb_usb_data *sisusb, int port, u8 data)
	1202	+int sisusb_setreg(struct sisusb_usb_data *sisusb, u32 port, u8 data)
1203	1203	{
1204	1204	return sisusb_write_memio_byte(sisusb, SISUSB_TYPE_IO, port, data);
1205	1205	}
1206	1206
1207		−int sisusb_getreg(struct sisusb_usb_data sisusb, int port, u8 data)
	1207	+int sisusb_getreg(struct sisusb_usb_data sisusb, u32 port, u8 data)
1208	1208	{
1209	1209	return sisusb_read_memio_byte(sisusb, SISUSB_TYPE_IO, port, data);
1210	1210	}
1211	1211	#endif
1212	1212
1213		−int sisusb_setidxreg(struct sisusb_usb_data *sisusb, int port,
	1213	+int sisusb_setidxreg(struct sisusb_usb_data *sisusb, u32 port,
1214	1214	u8 index, u8 data)
1215	1215	{
1216	1216	int ret;
@@ -1220,7 +1220,7 @@ int sisusb_setidxreg(struct sisusb_usb_data *sisusb, int port, · 3 unmodified lines
1220	1220	return ret;
1221	1221	}
1222	1222
1223		−int sisusb_getidxreg(struct sisusb_usb_data *sisusb, int port,
	1223	+int sisusb_getidxreg(struct sisusb_usb_data *sisusb, u32 port,
1224	1224	u8 index, u8 *data)
1225	1225	{
1226	1226	int ret;
@@ -1230,7 +1230,7 @@ int sisusb_getidxreg(struct sisusb_usb_data *sisusb, int port, · 3 unmodified lines
1230	1230	return ret;
1231	1231	}
1232	1232
1233		−int sisusb_setidxregandor(struct sisusb_usb_data *sisusb, int port, u8 idx,
	1233	+int sisusb_setidxregandor(struct sisusb_usb_data *sisusb, u32 port, u8 idx,
1234	1234	u8 myand, u8 myor)
1235	1235	{
1236	1236	int ret;
@@ -1245,7 +1245,7 @@ int sisusb_setidxregandor(struct sisusb_usb_data *sisusb, int port, u8 idx, · 8 unmodified lines
1245	1245	}
1246	1246
1247	1247	static int sisusb_setidxregmask(struct sisusb_usb_data *sisusb,
1248		− int port, u8 idx, u8 data, u8 mask)
	1248	+ u32 port, u8 idx, u8 data, u8 mask)
1249	1249	{
1250	1250	int ret;
1251	1251	u8 tmp;
@@ -1258,13 +1258,13 @@ static int sisusb_setidxregmask(struct sisusb_usb_data *sisusb, · 6 unmodified lines
1258	1258	return ret;
1259	1259	}
1260	1260
1261		−int sisusb_setidxregor(struct sisusb_usb_data *sisusb, int port,
	1261	+int sisusb_setidxregor(struct sisusb_usb_data *sisusb, u32 port,
1262	1262	u8 index, u8 myor)
1263	1263	{
1264	1264	return sisusb_setidxregandor(sisusb, port, index, 0xff, myor);
1265	1265	}
1266	1266
1267		−int sisusb_setidxregand(struct sisusb_usb_data *sisusb, int port,
	1267	+int sisusb_setidxregand(struct sisusb_usb_data *sisusb, u32 port,
1268	1268	u8 idx, u8 myand)
1269	1269	{
1270	1270	return sisusb_setidxregandor(sisusb, port, idx, myand, 0x00);
@@ -2785,8 +2785,8 @@ static loff_t sisusb_lseek(struct file *file, loff_t offset, int orig) · 1514 unmodified lines
2785	2785	static int sisusb_handle_command(struct sisusb_usb_data *sisusb,
2786	2786	struct sisusb_command *y, unsigned long arg)
2787	2787	{
2788		− int retval, port, length;
2789		− u32 address;
	2788	+ int retval, length;
	2789	+ u32 port, address;
2790	2790
2791	2791	/* All our commands require the device
2792	2792	* to be initialized.

drivers/usb/misc/sisusbvga/sisusb_init.h+7−7

@@ -812,17 +812,17 @@ static const struct SiS_VCLKData SiSUSB_VCLKData[] = { · 811 unmodified lines
812	812	int SiSUSBSetMode(struct SiS_Private *SiS_Pr, unsigned short ModeNo);
813	813	int SiSUSBSetVESAMode(struct SiS_Private *SiS_Pr, unsigned short VModeNo);
814	814
815		−extern int sisusb_setreg(struct sisusb_usb_data *sisusb, int port, u8 data);
816		−extern int sisusb_getreg(struct sisusb_usb_data sisusb, int port, u8 data);
817		−extern int sisusb_setidxreg(struct sisusb_usb_data *sisusb, int port,
	815	+extern int sisusb_setreg(struct sisusb_usb_data *sisusb, u32 port, u8 data);
	816	+extern int sisusb_getreg(struct sisusb_usb_data sisusb, u32 port, u8 data);
	817	+extern int sisusb_setidxreg(struct sisusb_usb_data *sisusb, u32 port,
818	818	u8 index, u8 data);
819		−extern int sisusb_getidxreg(struct sisusb_usb_data *sisusb, int port,
	819	+extern int sisusb_getidxreg(struct sisusb_usb_data *sisusb, u32 port,
820	820	u8 index, u8 * data);
821		−extern int sisusb_setidxregandor(struct sisusb_usb_data *sisusb, int port,
	821	+extern int sisusb_setidxregandor(struct sisusb_usb_data *sisusb, u32 port,
822	822	u8 idx, u8 myand, u8 myor);
823		−extern int sisusb_setidxregor(struct sisusb_usb_data *sisusb, int port,
	823	+extern int sisusb_setidxregor(struct sisusb_usb_data *sisusb, u32 port,
824	824	u8 index, u8 myor);
825		−extern int sisusb_setidxregand(struct sisusb_usb_data *sisusb, int port,
	825	+extern int sisusb_setidxregand(struct sisusb_usb_data *sisusb, u32 port,
826	826	u8 idx, u8 myand);
827	827
828	828	void sisusb_delete(struct kref *kref);

drivers/usb/serial/garmin_gps.c+2−2

@@ -1138,8 +1138,8 @@ static void garmin_read_process(struct garmin_data *garmin_data_p, · 1137 unmodified lines
1138	1138	send it directly to the tty port */
1139	1139	if (garmin_data_p->flags & FLAGS_QUEUING) {
1140	1140	pkt_add(garmin_data_p, data, data_length);
1141		− } else if (bulk_data \|\|
1142		− getLayerId(data) == GARMIN_LAYERID_APPL) {
	1141	+ } else if (bulk_data \|\| (data_length >= sizeof(u32) &&
	1142	+ getLayerId(data) == GARMIN_LAYERID_APPL)) {
1143	1143
1144	1144	spin_lock_irqsave(&garmin_data_p->lock, flags);
1145	1145	garmin_data_p->flags \|= APP_RESP_SEEN;

drivers/usb/serial/qcserial.c+1−0

@@ -173,6 +173,7 @@ static const struct usb_device_id id_table[] = { · 172 unmodified lines
173	173	{DEVICE_SWI(0x413c, 0x81b3)}, /* Dell Wireless 5809e Gobi(TM) 4G LTE Mobile Broadband Card (rev3) */
174	174	{DEVICE_SWI(0x413c, 0x81b5)}, /* Dell Wireless 5811e QDL */
175	175	{DEVICE_SWI(0x413c, 0x81b6)}, /* Dell Wireless 5811e QDL */
	176	+ {DEVICE_SWI(0x413c, 0x81cc)}, /* Dell Wireless 5816e */
176	177	{DEVICE_SWI(0x413c, 0x81cf)}, /* Dell Wireless 5819 */
177	178	{DEVICE_SWI(0x413c, 0x81d0)}, /* Dell Wireless 5819 */
178	179	{DEVICE_SWI(0x413c, 0x81d1)}, /* Dell Wireless 5818 */

drivers/usb/storage/uas.c+43−3

@@ -81,6 +81,19 @@ static void uas_free_streams(struct uas_dev_info *devinfo); · 80 unmodified lines
81	81	static void uas_log_cmd_state(struct scsi_cmnd cmnd, const char prefix,
82	82	int status);
83	83
	84	+/*
	85	+ * This driver needs its own workqueue, as we need to control memory allocation.
	86	+ *
	87	+ * In the course of error handling and power management uas_wait_for_pending_cmnds()
	88	+ * needs to flush pending work items. In these contexts we cannot allocate memory
	89	+ * by doing block IO as we would deadlock. For the same reason we cannot wait
	90	+ * for anything allocating memory not heeding these constraints.
	91	+ *
	92	+ * So we have to control all work items that can be on the workqueue we flush.
	93	+ * Hence we cannot share a queue and need our own.
	94	+ */
	95	+static struct workqueue_struct *workqueue;
	96	+
84	97	static void uas_do_work(struct work_struct *work)
85	98	{
86	99	struct uas_dev_info *devinfo =
@@ -109,7 +122,7 @@ static void uas_do_work(struct work_struct *work) · 22 unmodified lines
109	122	if (!err)
110	123	cmdinfo->state &= ~IS_IN_WORK_LIST;
111	124	else
112		− schedule_work(&devinfo->work);
	125	+ queue_work(workqueue, &devinfo->work);
113	126	}
114	127	out:
115	128	spin_unlock_irqrestore(&devinfo->lock, flags);
@@ -134,7 +147,7 @@ static void uas_add_work(struct uas_cmd_info *cmdinfo) · 18 unmodified lines
134	147
135	148	lockdep_assert_held(&devinfo->lock);
136	149	cmdinfo->state \|= IS_IN_WORK_LIST;
137		− schedule_work(&devinfo->work);
	150	+ queue_work(workqueue, &devinfo->work);
138	151	}
139	152
140	153	static void uas_zap_pending(struct uas_dev_info *devinfo, int result)
@@ -190,6 +203,9 @@ static void uas_log_cmd_state(struct scsi_cmnd cmnd, const char prefix, · 49 unmodified lines
190	203	struct uas_cmd_info ci = (void )&cmnd->SCp;
191	204	struct uas_cmd_info cmdinfo = (void )&cmnd->SCp;
192	205
	206	+ if (status == -ENODEV) /* too late */
	207	+ return;
	208	+
193	209	scmd_printk(KERN_INFO, cmnd,
194	210	"%s %d uas-tag %d inflight:%s%s%s%s%s%s%s%s%s%s%s%s ",
195	211	prefix, status, cmdinfo->uas_tag,
@@ -1226,7 +1242,31 @@ static struct usb_driver uas_driver = { · 1030 unmodified lines
1226	1242	.id_table = uas_usb_ids,
1227	1243	};
1228	1244
1229		−module_usb_driver(uas_driver);
	1245	+static int __init uas_init(void)
	1246	+{
	1247	+ int rv;
	1248	+
	1249	+ workqueue = alloc_workqueue("uas", WQ_MEM_RECLAIM, 0);
	1250	+ if (!workqueue)
	1251	+ return -ENOMEM;
	1252	+
	1253	+ rv = usb_register(&uas_driver);
	1254	+ if (rv) {
	1255	+ destroy_workqueue(workqueue);
	1256	+ return -ENOMEM;
	1257	+ }
	1258	+
	1259	+ return 0;
	1260	+}
	1261	+
	1262	+static void __exit uas_exit(void)
	1263	+{
	1264	+ usb_deregister(&uas_driver);
	1265	+ destroy_workqueue(workqueue);
	1266	+}
	1267	+
	1268	+module_init(uas_init);
	1269	+module_exit(uas_exit);
1230	1270
1231	1271	MODULE_LICENSE("GPL");
1232	1272	MODULE_IMPORT_NS(USB_STORAGE);

drivers/usb/storage/unusual_devs.h+7−0

@@ -2323,6 +2323,13 @@ UNUSUAL_DEV( 0x3340, 0xffff, 0x0000, 0x0000, · 2322 unmodified lines
2323	2323	USB_SC_DEVICE,USB_PR_DEVICE,NULL,
2324	2324	US_FL_MAX_SECTORS_64 ),
2325	2325
	2326	+/* Reported by Cyril Roelandt <tipecaml@gmail.com> */
	2327	+UNUSUAL_DEV( 0x357d, 0x7788, 0x0114, 0x0114,
	2328	+ "JMicron",
	2329	+ "USB to ATA/ATAPI Bridge",
	2330	+ USB_SC_DEVICE, USB_PR_DEVICE, NULL,
	2331	+ US_FL_BROKEN_FUA ),
	2332	+
2326	2333	/* Reported by Andrey Rahmatullin <wrar@altlinux.org> */
2327	2334	UNUSUAL_DEV( 0x4102, 0x1020, 0x0100, 0x0100,
2328	2335	"iRiver",

drivers/usb/storage/unusual_uas.h+7−0

@@ -28,6 +28,13 @@ · 27 unmodified lines
28	28	* and don't forget to CC: the USB development list <linux-usb@vger.kernel.org>
29	29	*/
30	30
	31	+/* Reported-by: Julian Groß <julian.g@posteo.de> */
	32	+UNUSUAL_DEV(0x059f, 0x105f, 0x0000, 0x9999,
	33	+ "LaCie",
	34	+ "2Big Quadra USB3",
	35	+ USB_SC_DEVICE, USB_PR_DEVICE, NULL,
	36	+ US_FL_NO_REPORT_OPCODES),
	37	+
31	38	/*
32	39	* Apricorn USB3 dongle sometimes returns "USBSUSBSUSBS" in response to SCSI
33	40	* commands in UAS mode. Observed with the 1.28 firmware; are there others?

drivers/usb/typec/bus.c+4−1

@@ -198,7 +198,10 @@ EXPORT_SYMBOL_GPL(typec_altmode_vdm); · 197 unmodified lines
198	198	const struct typec_altmode *
199	199	typec_altmode_get_partner(struct typec_altmode *adev)
200	200	{
201		− return adev ? &to_altmode(adev)->partner->adev : NULL;
	201	+ if (!adev \|\| !to_altmode(adev)->partner)
	202	+ return NULL;
	203	+
	204	+ return &to_altmode(adev)->partner->adev;
202	205	}
203	206	EXPORT_SYMBOL_GPL(typec_altmode_get_partner);
204	207

drivers/usb/typec/mux/intel_pmc_mux.c+6−2

@@ -157,6 +157,10 @@ pmc_usb_mux_dp(struct pmc_usb_port port, struct typec_mux_state state) · 156 unmodified lines
157	157	req.mode_data \|= (state->mode - TYPEC_STATE_MODAL) <<
158	158	PMC_USB_ALTMODE_DP_MODE_SHIFT;
159	159
	160	+ if (data->status & DP_STATUS_HPD_STATE)
	161	+ req.mode_data \|= PMC_USB_DP_HPD_LVL <<
	162	+ PMC_USB_ALTMODE_DP_MODE_SHIFT;
	163	+
160	164	return pmc_usb_command(port, (void *)&req, sizeof(req));
161	165	}
162	166
@@ -298,11 +302,11 @@ static int pmc_usb_register_port(struct pmc_usb *pmc, int index, · 135 unmodified lines
298	302	struct typec_mux_desc mux_desc = { };
299	303	int ret;
300	304
301		− ret = fwnode_property_read_u8(fwnode, "usb2-port", &port->usb2_port);
	305	+ ret = fwnode_property_read_u8(fwnode, "usb2-port-number", &port->usb2_port);
302	306	if (ret)
303	307	return ret;
304	308
305		− ret = fwnode_property_read_u8(fwnode, "usb3-port", &port->usb3_port);
	309	+ ret = fwnode_property_read_u8(fwnode, "usb3-port-number", &port->usb3_port);
306	310	if (ret)
307	311	return ret;
308	312

drivers/usb/typec/mux/pi3usb30532.c+2−2

@@ -114,8 +114,8 @@ pi3usb30532_mux_set(struct typec_mux mux, struct typec_mux_state state) · 113 unmodified lines
114	114	static int pi3usb30532_probe(struct i2c_client *client)
115	115	{
116	116	struct device *dev = &client->dev;
117		− struct typec_switch_desc sw_desc;
118		− struct typec_mux_desc mux_desc;
	117	+ struct typec_switch_desc sw_desc = { };
	118	+ struct typec_mux_desc mux_desc = { };
119	119	struct pi3usb30532 *pi;
120	120	int ret;
121	121

drivers/usb/typec/tcpm/tcpm.c+26−0

@@ -3794,6 +3794,14 @@ static void _tcpm_cc_change(struct tcpm_port *port, enum typec_cc_status cc1, · 3793 unmodified lines
3794	3794	*/
3795	3795	break;
3796	3796
	3797	+ case PORT_RESET:
	3798	+ case PORT_RESET_WAIT_OFF:
	3799	+ /*
	3800	+ * State set back to default mode once the timer completes.
	3801	+ * Ignore CC changes here.
	3802	+ */
	3803	+ break;
	3804	+
3797	3805	default:
3798	3806	if (tcpm_port_is_disconnected(port))
3799	3807	tcpm_set_state(port, unattached_state(port), 0);
@@ -3855,6 +3863,15 @@ static void _tcpm_pd_vbus_on(struct tcpm_port *port) · 55 unmodified lines
3855	3863	case SRC_TRY_DEBOUNCE:
3856	3864	/* Do nothing, waiting for sink detection */
3857	3865	break;
	3866	+
	3867	+ case PORT_RESET:
	3868	+ case PORT_RESET_WAIT_OFF:
	3869	+ /*
	3870	+ * State set back to default mode once the timer completes.
	3871	+ * Ignore vbus changes here.
	3872	+ */
	3873	+ break;
	3874	+
3858	3875	default:
3859	3876	break;
3860	3877	}
@@ -3908,10 +3925,19 @@ static void _tcpm_pd_vbus_off(struct tcpm_port *port) · 47 unmodified lines
3908	3925	case PORT_RESET_WAIT_OFF:
3909	3926	tcpm_set_state(port, tcpm_default_state(port), 0);
3910	3927	break;
	3928	+
3911	3929	case SRC_TRY_WAIT:
3912	3930	case SRC_TRY_DEBOUNCE:
3913	3931	/* Do nothing, waiting for sink detection */
3914	3932	break;
	3933	+
	3934	+ case PORT_RESET:
	3935	+ /*
	3936	+ * State set back to default mode once the timer completes.
	3937	+ * Ignore vbus changes here.
	3938	+ */
	3939	+ break;
	3940	+
3915	3941	default:
3916	3942	if (port->pwr_role == TYPEC_SINK &&
3917	3943	port->attached)

drivers/vdpa/Kconfig+6−12

@@ -1,21 +1,16 @@
1	1	# SPDX-License-Identifier: GPL-2.0-only
2		−config VDPA
3		− tristate
	2	+menuconfig VDPA
	3	+ tristate "vDPA drivers"
4	4	help
5	5	Enable this module to support vDPA device that uses a
6	6	datapath which complies with virtio specifications with
7	7	vendor specific control path.
8	8
9		−menuconfig VDPA_MENU
10		− bool "VDPA drivers"
11		− default n
12		−
13		−if VDPA_MENU
	9	+if VDPA
14	10
15	11	config VDPA_SIM
16	12	tristate "vDPA device simulator"
17		− depends on RUNTIME_TESTING_MENU
18		− select VDPA
	13	+ depends on RUNTIME_TESTING_MENU && HAS_DMA && VHOST_DPN
19	14	select VHOST_RING
20	15	default n
21	16	help
@@ -24,9 +19,8 @@ config VDPA_SIM · 2 unmodified lines
24	19	development of vDPA.
25	20
26	21	config IFCVF
27		− tristate "Intel IFC VF VDPA driver"
	22	+ tristate "Intel IFC VF vDPA driver"
28	23	depends on PCI_MSI
29		− select VDPA
30	24	default n
31	25	help
32	26	This kernel module can drive Intel IFC VF NIC to offload
@@ -34,4 +28,4 @@ config IFCVF · 1 unmodified line
34	28	To compile this driver as a module, choose M here: the module will
35	29	be called ifcvf.
36	30
37		−endif # VDPA_MENU
	31	+endif # VDPA

drivers/vdpa/ifcvf/ifcvf_base.c+0−2

@@ -301,12 +301,10 @@ int ifcvf_set_vq_state(struct ifcvf_hw *hw, u16 qid, u64 num) · 300 unmodified lines
301	301
302	302	static int ifcvf_hw_enable(struct ifcvf_hw *hw)
303	303	{
304		− struct ifcvf_lm_cfg __iomem *ifcvf_lm;
305	304	struct virtio_pci_common_cfg __iomem *cfg;
306	305	struct ifcvf_adapter *ifcvf;
307	306	u32 i;
308	307
309		− ifcvf_lm = (struct ifcvf_lm_cfg __iomem *)hw->lm_cfg;
310	308	ifcvf = vf_to_adapter(hw);
311	309	cfg = hw->common_cfg;
312	310	ifc_iowrite16(IFCVF_MSI_CONFIG_OFF, &cfg->msix_config);

drivers/vdpa/ifcvf/ifcvf_main.c+1−3

@@ -31,11 +31,9 @@ static irqreturn_t ifcvf_intr_handler(int irq, void *arg) · 30 unmodified lines
31	31	static int ifcvf_start_datapath(void *private)
32	32	{
33	33	struct ifcvf_hw *vf = ifcvf_private_to_vf(private);
34		− struct ifcvf_adapter *ifcvf;
35	34	u8 status;
36	35	int ret;
37	36
38		− ifcvf = vf_to_adapter(vf);
39	37	vf->nr_vring = IFCVF_MAX_QUEUE_PAIRS * 2;
40	38	ret = ifcvf_start_hw(vf);
41	39	if (ret < 0) {
@@ -228,7 +226,7 @@ static u32 ifcvf_vdpa_get_vendor_id(struct vdpa_device *vdpa_dev) · 186 unmodified lines
228	226	return IFCVF_SUBSYS_VENDOR_ID;
229	227	}
230	228
231		−static u16 ifcvf_vdpa_get_vq_align(struct vdpa_device *vdpa_dev)
	229	+static u32 ifcvf_vdpa_get_vq_align(struct vdpa_device *vdpa_dev)
232	230	{
233	231	return IFCVF_QUEUE_ALIGNMENT;
234	232	}

drivers/vdpa/vdpa.c+1−1

@@ -116,7 +116,7 @@ EXPORT_SYMBOL_GPL(__vdpa_alloc_device); · 115 unmodified lines
116	116
117	117	/**
118	118	* vdpa_register_device - register a vDPA device
119		− * Callers must have a succeed call of vdpa_init_device() before.
	119	+ * Callers must have a succeed call of vdpa_alloc_device() before.
120	120	* @vdev: the vdpa device to be registered to vDPA bus
121	121	*
122	122	* Returns an error when fail to add to vDPA bus

drivers/vdpa/vdpa_sim/vdpa_sim.c+2−2

@@ -435,7 +435,7 @@ static u64 vdpasim_get_vq_state(struct vdpa_device *vdpa, u16 idx) · 434 unmodified lines
435	435	return vrh->last_avail_idx;
436	436	}
437	437
438		−static u16 vdpasim_get_vq_align(struct vdpa_device *vdpa)
	438	+static u32 vdpasim_get_vq_align(struct vdpa_device *vdpa)
439	439	{
440	440	return VDPASIM_QUEUE_ALIGN;
441	441	}
@@ -488,7 +488,7 @@ static u8 vdpasim_get_status(struct vdpa_device *vdpa) · 46 unmodified lines
488	488	status = vdpasim->status;
489	489	spin_unlock(&vdpasim->lock);
490	490
491		− return vdpasim->status;
	491	+ return status;
492	492	}
493	493
494	494	static void vdpasim_set_status(struct vdpa_device *vdpa, u8 status)

drivers/vfio/vfio_iommu_type1.c+5−5

@@ -342,8 +342,8 @@ static int vaddr_get_pfn(struct mm_struct *mm, unsigned long vaddr, · 341 unmodified lines
342	342	vma = find_vma_intersection(mm, vaddr, vaddr + 1);
343	343
344	344	if (vma && vma->vm_flags & VM_PFNMAP) {
345		− *pfn = ((vaddr - vma->vm_start) >> PAGE_SHIFT) + vma->vm_pgoff;
346		− if (is_invalid_reserved_pfn(*pfn))
	345	+ if (!follow_pfn(vma, vaddr, pfn) &&
	346	+ is_invalid_reserved_pfn(*pfn))
347	347	ret = 0;
348	348	}
349	349	done:
@@ -555,7 +555,7 @@ static int vfio_iommu_type1_pin_pages(void *iommu_data, · 205 unmodified lines
555	555	continue;
556	556	}
557	557
558		− remote_vaddr = dma->vaddr + iova - dma->iova;
	558	+ remote_vaddr = dma->vaddr + (iova - dma->iova);
559	559	ret = vfio_pin_page_external(dma, remote_vaddr, &phys_pfn[i],
560	560	do_accounting);
561	561	if (ret)
@@ -2345,10 +2345,10 @@ static int vfio_iommu_type1_dma_rw_chunk(struct vfio_iommu *iommu, · 1783 unmodified lines
2345	2345	vaddr = dma->vaddr + offset;
2346	2346
2347	2347	if (write)
2348		− copied = __copy_to_user((void __user )vaddr, data,
	2348	+ copied = copy_to_user((void __user )vaddr, data,
2349	2349	count) ? 0 : count;
2350	2350	else
2351		− copied = __copy_from_user(data, (void __user )vaddr,
	2351	+ copied = copy_from_user(data, (void __user )vaddr,
2352	2352	count) ? 0 : count;
2353	2353	if (kthread)
2354	2354	unuse_mm(mm);

drivers/vhost/Kconfig+16−5

@@ -3,6 +3,8 @@ config VHOST_IOTLB · 2 unmodified lines
3	3	tristate
4	4	help
5	5	Generic IOTLB implementation for vhost and vringh.
	6	+ This option is selected by any driver which needs to support
	7	+ an IOMMU in software.
6	8
7	9	config VHOST_RING
8	10	tristate
@@ -11,6 +13,15 @@ config VHOST_RING · 2 unmodified lines
11	13	This option is selected by any driver which needs to access
12	14	the host side of a virtio ring.
13	15
	16	+config VHOST_DPN
	17	+ bool
	18	+ depends on !ARM \|\| AEABI
	19	+ default y
	20	+ help
	21	+ Anything selecting VHOST or VHOST_RING must depend on VHOST_DPN.
	22	+ This excludes the deprecated ARM ABI since that forces a 4 byte
	23	+ alignment on all structs - incompatible with virtio spec requirements.
	24	+
14	25	config VHOST
15	26	tristate
16	27	select VHOST_IOTLB
@@ -26,7 +37,7 @@ if VHOST_MENU · 9 unmodified lines
26	37
27	38	config VHOST_NET
28	39	tristate "Host kernel accelerator for virtio net"
29		− depends on NET && EVENTFD && (TUN \|\| !TUN) && (TAP \|\| !TAP)
	40	+ depends on NET && EVENTFD && (TUN \|\| !TUN) && (TAP \|\| !TAP) && VHOST_DPN
30	41	select VHOST
31	42	---help---
32	43	This kernel module can be loaded in host kernel to accelerate
@@ -38,7 +49,7 @@ config VHOST_NET · 5 unmodified lines
38	49
39	50	config VHOST_SCSI
40	51	tristate "VHOST_SCSI TCM fabric driver"
41		− depends on TARGET_CORE && EVENTFD
	52	+ depends on TARGET_CORE && EVENTFD && VHOST_DPN
42	53	select VHOST
43	54	default n
44	55	---help---
@@ -47,7 +58,7 @@ config VHOST_SCSI · 2 unmodified lines
47	58
48	59	config VHOST_VSOCK
49	60	tristate "vhost virtio-vsock driver"
50		− depends on VSOCKETS && EVENTFD
	61	+ depends on VSOCKETS && EVENTFD && VHOST_DPN
51	62	select VHOST
52	63	select VIRTIO_VSOCKETS_COMMON
53	64	default n
@@ -61,9 +72,9 @@ config VHOST_VSOCK · 7 unmodified lines
61	72
62	73	config VHOST_VDPA
63	74	tristate "Vhost driver for vDPA-based backend"
64		− depends on EVENTFD
	75	+ depends on EVENTFD && VHOST_DPN
65	76	select VHOST
66		− select VDPA
	77	+ depends on VDPA
67	78	help
68	79	This kernel module can be loaded in host kernel to accelerate
69	80	guest virtio devices with the vDPA-based backends.

drivers/vhost/net.c+15−13

@@ -424,7 +424,7 @@ static void vhost_net_disable_vq(struct vhost_net *n, · 423 unmodified lines
424	424	struct vhost_net_virtqueue *nvq =
425	425	container_of(vq, struct vhost_net_virtqueue, vq);
426	426	struct vhost_poll *poll = n->poll + (nvq - n->vqs);
427		− if (!vq->private_data)
	427	+ if (!vhost_vq_get_backend(vq))
428	428	return;
429	429	vhost_poll_stop(poll);
430	430	}
@@ -437,7 +437,7 @@ static int vhost_net_enable_vq(struct vhost_net *n, · 6 unmodified lines
437	437	struct vhost_poll *poll = n->poll + (nvq - n->vqs);
438	438	struct socket *sock;
439	439
440		− sock = vq->private_data;
	440	+ sock = vhost_vq_get_backend(vq);
441	441	if (!sock)
442	442	return 0;
443	443
@@ -524,7 +524,7 @@ static void vhost_net_busy_poll(struct vhost_net *net, · 80 unmodified lines
524	524	return;
525	525
526	526	vhost_disable_notify(&net->dev, vq);
527		− sock = rvq->private_data;
	527	+ sock = vhost_vq_get_backend(rvq);
528	528
529	529	busyloop_timeout = poll_rx ? rvq->busyloop_timeout:
530	530	tvq->busyloop_timeout;
@@ -570,8 +570,10 @@ static int vhost_net_tx_get_vq_desc(struct vhost_net *net, · 39 unmodified lines
570	570
571	571	if (r == tvq->num && tvq->busyloop_timeout) {
572	572	/* Flush batched packets first */
573		− if (!vhost_sock_zcopy(tvq->private_data))
574		− vhost_tx_batch(net, tnvq, tvq->private_data, msghdr);
	573	+ if (!vhost_sock_zcopy(vhost_vq_get_backend(tvq)))
	574	+ vhost_tx_batch(net, tnvq,
	575	+ vhost_vq_get_backend(tvq),
	576	+ msghdr);
575	577
576	578	vhost_net_busy_poll(net, rvq, tvq, busyloop_intr, false);
577	579
@@ -685,7 +687,7 @@ static int vhost_net_build_xdp(struct vhost_net_virtqueue *nvq, · 107 unmodified lines
685	687	struct vhost_virtqueue *vq = &nvq->vq;
686	688	struct vhost_net *net = container_of(vq->dev, struct vhost_net,
687	689	dev);
688		− struct socket *sock = vq->private_data;
	690	+ struct socket *sock = vhost_vq_get_backend(vq);
689	691	struct page_frag *alloc_frag = &net->page_frag;
690	692	struct virtio_net_hdr *gso;
691	693	struct xdp_buff *xdp = &nvq->xdp[nvq->batched_xdp];
@@ -952,7 +954,7 @@ static void handle_tx(struct vhost_net *net) · 260 unmodified lines
952	954	struct socket *sock;
953	955
954	956	mutex_lock_nested(&vq->mutex, VHOST_NET_VQ_TX);
955		− sock = vq->private_data;
	957	+ sock = vhost_vq_get_backend(vq);
956	958	if (!sock)
957	959	goto out;
958	960
@@ -1121,7 +1123,7 @@ static void handle_rx(struct vhost_net *net) · 162 unmodified lines
1121	1123	int recv_pkts = 0;
1122	1124
1123	1125	mutex_lock_nested(&vq->mutex, VHOST_NET_VQ_RX);
1124		− sock = vq->private_data;
	1126	+ sock = vhost_vq_get_backend(vq);
1125	1127	if (!sock)
1126	1128	goto out;
1127	1129
@@ -1345,9 +1347,9 @@ static struct socket vhost_net_stop_vq(struct vhost_net n, · 217 unmodified lines
1345	1347	container_of(vq, struct vhost_net_virtqueue, vq);
1346	1348
1347	1349	mutex_lock(&vq->mutex);
1348		− sock = vq->private_data;
	1350	+ sock = vhost_vq_get_backend(vq);
1349	1351	vhost_net_disable_vq(n, vq);
1350		− vq->private_data = NULL;
	1352	+ vhost_vq_set_backend(vq, NULL);
1351	1353	vhost_net_buf_unproduce(nvq);
1352	1354	nvq->rx_ring = NULL;
1353	1355	mutex_unlock(&vq->mutex);
@@ -1521,7 +1523,7 @@ static long vhost_net_set_backend(struct vhost_net *n, unsigned index, int fd) · 167 unmodified lines
1521	1523	}
1522	1524
1523	1525	/* start polling new socket */
1524		− oldsock = vq->private_data;
	1526	+ oldsock = vhost_vq_get_backend(vq);
1525	1527	if (sock != oldsock) {
1526	1528	ubufs = vhost_net_ubuf_alloc(vq,
1527	1529	sock && vhost_sock_zcopy(sock));
@@ -1531,7 +1533,7 @@ static long vhost_net_set_backend(struct vhost_net *n, unsigned index, int fd) · 3 unmodified lines
1531	1533	}
1532	1534
1533	1535	vhost_net_disable_vq(n, vq);
1534		− vq->private_data = sock;
	1536	+ vhost_vq_set_backend(vq, sock);
1535	1537	vhost_net_buf_unproduce(nvq);
1536	1538	r = vhost_vq_init_access(vq);
1537	1539	if (r)
@@ -1568,7 +1570,7 @@ static long vhost_net_set_backend(struct vhost_net *n, unsigned index, int fd) · 30 unmodified lines
1568	1570	return 0;
1569	1571
1570	1572	err_used:
1571		− vq->private_data = oldsock;
	1573	+ vhost_vq_set_backend(vq, oldsock);
1572	1574	vhost_net_enable_vq(n, vq);
1573	1575	if (ubufs)
1574	1576	vhost_net_ubuf_put_wait_and_free(ubufs);

drivers/vhost/scsi.c+7−7

@@ -452,7 +452,7 @@ vhost_scsi_do_evt_work(struct vhost_scsi vs, struct vhost_scsi_evt evt) · 451 unmodified lines
452	452	unsigned out, in;
453	453	int head, ret;
454	454
455		− if (!vq->private_data) {
	455	+ if (!vhost_vq_get_backend(vq)) {
456	456	vs->vs_events_missed = true;
457	457	return;
458	458	}
@@ -892,7 +892,7 @@ vhost_scsi_get_req(struct vhost_virtqueue vq, struct vhost_scsi_ctx vc, · 433 unmodified lines
892	892	} else {
893	893	struct vhost_scsi_tpg *vs_tpg, tpg;
894	894
895		− vs_tpg = vq->private_data; /* validated at handler entry */
	895	+ vs_tpg = vhost_vq_get_backend(vq); /* validated at handler entry */
896	896
897	897	tpg = READ_ONCE(vs_tpg[*vc->target]);
898	898	if (unlikely(!tpg)) {
@@ -929,7 +929,7 @@ vhost_scsi_handle_vq(struct vhost_scsi vs, struct vhost_virtqueue vq) · 30 unmodified lines
929	929	* We can handle the vq only after the endpoint is setup by calling the
930	930	* VHOST_SCSI_SET_ENDPOINT ioctl.
931	931	*/
932		− vs_tpg = vq->private_data;
	932	+ vs_tpg = vhost_vq_get_backend(vq);
933	933	if (!vs_tpg)
934	934	goto out;
935	935
@@ -1184,7 +1184,7 @@ vhost_scsi_ctl_handle_vq(struct vhost_scsi vs, struct vhost_virtqueue vq) · 248 unmodified lines
1184	1184	* We can handle the vq only after the endpoint is setup by calling the
1185	1185	* VHOST_SCSI_SET_ENDPOINT ioctl.
1186	1186	*/
1187		− if (!vq->private_data)
	1187	+ if (!vhost_vq_get_backend(vq))
1188	1188	goto out;
1189	1189
1190	1190	memset(&vc, 0, sizeof(vc));
@@ -1322,7 +1322,7 @@ static void vhost_scsi_evt_handle_kick(struct vhost_work *work) · 131 unmodified lines
1322	1322	struct vhost_scsi *vs = container_of(vq->dev, struct vhost_scsi, dev);
1323	1323
1324	1324	mutex_lock(&vq->mutex);
1325		− if (!vq->private_data)
	1325	+ if (!vhost_vq_get_backend(vq))
1326	1326	goto out;
1327	1327
1328	1328	if (vs->vs_events_missed)
@@ -1460,7 +1460,7 @@ vhost_scsi_set_endpoint(struct vhost_scsi *vs, · 131 unmodified lines
1460	1460	for (i = 0; i < VHOST_SCSI_MAX_VQ; i++) {
1461	1461	vq = &vs->vqs[i].vq;
1462	1462	mutex_lock(&vq->mutex);
1463		− vq->private_data = vs_tpg;
	1463	+ vhost_vq_set_backend(vq, vs_tpg);
1464	1464	vhost_vq_init_access(vq);
1465	1465	mutex_unlock(&vq->mutex);
1466	1466	}
@@ -1547,7 +1547,7 @@ vhost_scsi_clear_endpoint(struct vhost_scsi *vs, · 80 unmodified lines
1547	1547	for (i = 0; i < VHOST_SCSI_MAX_VQ; i++) {
1548	1548	vq = &vs->vqs[i].vq;
1549	1549	mutex_lock(&vq->mutex);
1550		− vq->private_data = NULL;
	1550	+ vhost_vq_set_backend(vq, NULL);
1551	1551	mutex_unlock(&vq->mutex);
1552	1552	}
1553	1553	}

drivers/vhost/test.c+7−7

@@ -49,7 +49,7 @@ static void handle_vq(struct vhost_test *n) · 48 unmodified lines
49	49	void *private;
50	50
51	51	mutex_lock(&vq->mutex);
52		− private = vq->private_data;
	52	+ private = vhost_vq_get_backend(vq);
53	53	if (!private) {
54	54	mutex_unlock(&vq->mutex);
55	55	return;
@@ -120,7 +120,7 @@ static int vhost_test_open(struct inode inode, struct file f) · 64 unmodified lines
120	120	vqs[VHOST_TEST_VQ] = &n->vqs[VHOST_TEST_VQ];
121	121	n->vqs[VHOST_TEST_VQ].handle_kick = handle_vq_kick;
122	122	vhost_dev_init(dev, vqs, VHOST_TEST_VQ_MAX, UIO_MAXIOV,
123		− VHOST_TEST_PKT_WEIGHT, VHOST_TEST_WEIGHT);
	123	+ VHOST_TEST_PKT_WEIGHT, VHOST_TEST_WEIGHT, NULL);
124	124
125	125	f->private_data = n;
126	126
@@ -133,8 +133,8 @@ static void vhost_test_stop_vq(struct vhost_test n, · 6 unmodified lines
133	133	void *private;
134	134
135	135	mutex_lock(&vq->mutex);
136		− private = vq->private_data;
137		− vq->private_data = NULL;
	136	+ private = vhost_vq_get_backend(vq);
	137	+ vhost_vq_set_backend(vq, NULL);
138	138	mutex_unlock(&vq->mutex);
139	139	return private;
140	140	}
@@ -198,8 +198,8 @@ static long vhost_test_run(struct vhost_test *n, int test) · 57 unmodified lines
198	198	priv = test ? n : NULL;
199	199
200	200	/* start polling new socket */
201		− oldpriv = vq->private_data;
202		− vq->private_data = priv;
	201	+ oldpriv = vhost_vq_get_backend(vq);
	202	+ vhost_vq_set_backend(vq, priv);
203	203
204	204	r = vhost_vq_init_access(&n->vqs[index]);
205	205
@@ -225,7 +225,7 @@ static long vhost_test_reset_owner(struct vhost_test *n) · 19 unmodified lines
225	225	{
226	226	void *priv = NULL;
227	227	long err;
228		− struct vhost_umem *umem;
	228	+ struct vhost_iotlb *umem;
229	229
230	230	mutex_lock(&n->dev.mutex);
231	231	err = vhost_dev_check_owner(&n->dev);

drivers/vhost/vdpa.c+0−5

@@ -296,7 +296,6 @@ static long vhost_vdpa_vring_ioctl(struct vhost_vdpa *v, unsigned int cmd, · 295 unmodified lines
296	296	struct vdpa_callback cb;
297	297	struct vhost_virtqueue *vq;
298	298	struct vhost_vring_state s;
299		− u8 status;
300	299	u32 idx;
301	300	long r;
302	301
@@ -310,8 +309,6 @@ static long vhost_vdpa_vring_ioctl(struct vhost_vdpa *v, unsigned int cmd, · 7 unmodified lines
310	309	idx = array_index_nospec(idx, v->nvqs);
311	310	vq = &v->vqs[idx];
312	311
313		− status = ops->get_status(vdpa);
314		−
315	312	if (cmd == VHOST_VDPA_SET_VRING_ENABLE) {
316	313	if (copy_from_user(&s, argp, sizeof(s)))
317	314	return -EFAULT;
@@ -678,8 +675,6 @@ static int vhost_vdpa_open(struct inode inode, struct file filep) · 360 unmodified lines
678	675	int nvqs, i, r, opened;
679	676
680	677	v = container_of(inode->i_cdev, struct vhost_vdpa, cdev);
681		− if (!v)
682		− return -ENODEV;
683	678
684	679	opened = atomic_cmpxchg(&v->opened, 0, 1);
685	680	if (opened)

drivers/vhost/vhost.h+27−0

@@ -231,6 +231,33 @@ enum { · 230 unmodified lines
231	231	(1ULL << VIRTIO_F_VERSION_1)
232	232	};
233	233
	234	+/**
	235	+ * vhost_vq_set_backend - Set backend.
	236	+ *
	237	+ * @vq Virtqueue.
	238	+ * @private_data The private data.
	239	+ *
	240	+ * Context: Need to call with vq->mutex acquired.
	241	+ */
	242	+static inline void vhost_vq_set_backend(struct vhost_virtqueue *vq,
	243	+ void *private_data)
	244	+{
	245	+ vq->private_data = private_data;
	246	+}
	247	+
	248	+/**
	249	+ * vhost_vq_get_backend - Get backend.
	250	+ *
	251	+ * @vq Virtqueue.
	252	+ *
	253	+ * Context: Need to call with vq->mutex acquired.
	254	+ * Return: Private data previously set with vhost_vq_set_backend.
	255	+ */
	256	+static inline void vhost_vq_get_backend(struct vhost_virtqueue vq)
	257	+{
	258	+ return vq->private_data;
	259	+}
	260	+
234	261	static inline bool vhost_has_feature(struct vhost_virtqueue *vq, int bit)
235	262	{
236	263	return vq->acked_features & (1ULL << bit);

drivers/vhost/vringh.c+5−0

@@ -13,9 +13,11 @@ · 12 unmodified lines
13	13	#include <linux/uaccess.h>
14	14	#include <linux/slab.h>
15	15	#include <linux/export.h>
	16	+#if IS_REACHABLE(CONFIG_VHOST_IOTLB)
16	17	#include <linux/bvec.h>
17	18	#include <linux/highmem.h>
18	19	#include <linux/vhost_iotlb.h>
	20	+#endif
19	21	#include <uapi/linux/virtio_config.h>
20	22
21	23	static __printf(1,2) __cold void vringh_bad(const char *fmt, ...)
@@ -1059,6 +1061,8 @@ int vringh_need_notify_kern(struct vringh *vrh) · 1037 unmodified lines
1059	1061	}
1060	1062	EXPORT_SYMBOL(vringh_need_notify_kern);
1061	1063
	1064	+#if IS_REACHABLE(CONFIG_VHOST_IOTLB)
	1065	+
1062	1066	static int iotlb_translate(const struct vringh *vrh,
1063	1067	u64 addr, u64 len, struct bio_vec iov[],
1064	1068	int iov_size, u32 perm)
@@ -1416,5 +1420,6 @@ int vringh_need_notify_iotlb(struct vringh *vrh) · 351 unmodified lines
1416	1420	}
1417	1421	EXPORT_SYMBOL(vringh_need_notify_iotlb);
1418	1422
	1423	+#endif
1419	1424
1420	1425	MODULE_LICENSE("GPL");

drivers/vhost/vsock.c+23−12

@@ -91,7 +91,7 @@ vhost_transport_do_send_pkt(struct vhost_vsock *vsock, · 90 unmodified lines
91	91
92	92	mutex_lock(&vq->mutex);
93	93
94		− if (!vq->private_data)
	94	+ if (!vhost_vq_get_backend(vq))
95	95	goto out;
96	96
97	97	/* Avoid further vmexits, we're already processing the virtqueue */
@@ -181,14 +181,14 @@ vhost_transport_do_send_pkt(struct vhost_vsock *vsock, · 83 unmodified lines
181	181	break;
182	182	}
183	183
184		− vhost_add_used(vq, head, sizeof(pkt->hdr) + payload_len);
185		− added = true;
186		−
187		− /* Deliver to monitoring devices all correctly transmitted
188		− * packets.
	184	+ /* Deliver to monitoring devices all packets that we
	185	+ * will transmit.
189	186	*/
190	187	virtio_transport_deliver_tap_pkt(pkt);
191	188
	189	+ vhost_add_used(vq, head, sizeof(pkt->hdr) + payload_len);
	190	+ added = true;
	191	+
192	192	pkt->off += payload_len;
193	193	total_len += payload_len;
194	194
@@ -196,6 +196,12 @@ vhost_transport_do_send_pkt(struct vhost_vsock *vsock, · 1 unmodified line
196	196	* to send it with the next available buffer.
197	197	*/
198	198	if (pkt->off < pkt->len) {
	199	+ /* We are queueing the same virtio_vsock_pkt to handle
	200	+ * the remaining bytes, and we want to deliver it
	201	+ * to monitoring devices in the next iteration.
	202	+ */
	203	+ pkt->tap_delivered = false;
	204	+
199	205	spin_lock_bh(&vsock->send_pkt_list_lock);
200	206	list_add(&pkt->list, &vsock->send_pkt_list);
201	207	spin_unlock_bh(&vsock->send_pkt_list_lock);
@@ -440,7 +446,7 @@ static void vhost_vsock_handle_tx_kick(struct vhost_work *work) · 238 unmodified lines
440	446
441	447	mutex_lock(&vq->mutex);
442	448
443		− if (!vq->private_data)
	449	+ if (!vhost_vq_get_backend(vq))
444	450	goto out;
445	451
446	452	vhost_disable_notify(&vsock->dev, vq);
@@ -533,8 +539,8 @@ static int vhost_vsock_start(struct vhost_vsock *vsock) · 86 unmodified lines
533	539	goto err_vq;
534	540	}
535	541
536		− if (!vq->private_data) {
537		− vq->private_data = vsock;
	542	+ if (!vhost_vq_get_backend(vq)) {
	543	+ vhost_vq_set_backend(vq, vsock);
538	544	ret = vhost_vq_init_access(vq);
539	545	if (ret)
540	546	goto err_vq;
@@ -543,18 +549,23 @@ static int vhost_vsock_start(struct vhost_vsock *vsock) · 2 unmodified lines
543	549	mutex_unlock(&vq->mutex);
544	550	}
545	551
	552	+ /* Some packets may have been queued before the device was started,
	553	+ * let's kick the send worker to send them.
	554	+ */
	555	+ vhost_work_queue(&vsock->dev, &vsock->send_pkt_work);
	556	+
546	557	mutex_unlock(&vsock->dev.mutex);
547	558	return 0;
548	559
549	560	err_vq:
550		− vq->private_data = NULL;
	561	+ vhost_vq_set_backend(vq, NULL);
551	562	mutex_unlock(&vq->mutex);
552	563
553	564	for (i = 0; i < ARRAY_SIZE(vsock->vqs); i++) {
554	565	vq = &vsock->vqs[i];
555	566
556	567	mutex_lock(&vq->mutex);
557		− vq->private_data = NULL;
	568	+ vhost_vq_set_backend(vq, NULL);
558	569	mutex_unlock(&vq->mutex);
559	570	}
560	571	err:
@@ -577,7 +588,7 @@ static int vhost_vsock_stop(struct vhost_vsock *vsock) · 16 unmodified lines
577	588	struct vhost_virtqueue *vq = &vsock->vqs[i];
578	589
579	590	mutex_lock(&vq->mutex);
580		− vq->private_data = NULL;
	591	+ vhost_vq_set_backend(vq, NULL);
581	592	mutex_unlock(&vq->mutex);
582	593	}
583	594

drivers/virtio/Kconfig+1−1

@@ -45,7 +45,7 @@ config VIRTIO_PCI_LEGACY · 44 unmodified lines
45	45
46	46	config VIRTIO_VDPA
47	47	tristate "vDPA driver for virtio devices"
48		− select VDPA
	48	+ depends on VDPA
49	49	select VIRTIO
50	50	help
51	51	This driver provides support for virtio based paravirtual

drivers/virtio/virtio_balloon.c+2−2

@@ -165,7 +165,7 @@ static void tell_host(struct virtio_balloon vb, struct virtqueue vq) · 164 unmodified lines
165	165
166	166	}
167	167
168		−int virtballoon_free_page_report(struct page_reporting_dev_info *pr_dev_info,
	168	+static int virtballoon_free_page_report(struct page_reporting_dev_info *pr_dev_info,
169	169	struct scatterlist *sg, unsigned int nents)
170	170	{
171	171	struct virtio_balloon *vb =
@@ -580,7 +580,7 @@ static u32 virtio_balloon_cmd_id_received(struct virtio_balloon *vb) · 408 unmodified lines
580	580	if (test_and_clear_bit(VIRTIO_BALLOON_CONFIG_READ_CMD_ID,
581	581	&vb->config_read_bitmap))
582	582	virtio_cread(vb->vdev, struct virtio_balloon_config,
583		− free_page_report_cmd_id,
	583	+ free_page_hint_cmd_id,
584	584	&vb->cmd_id_received_cache);
585	585
586	586	return vb->cmd_id_received_cache;

drivers/virtio/virtio_input.c+1−0

@@ -3,6 +3,7 @@ · 2 unmodified lines
3	3	#include <linux/virtio.h>
4	4	#include <linux/virtio_config.h>
5	5	#include <linux/input.h>
	6	+#include <linux/slab.h>
6	7
7	8	#include <uapi/linux/virtio_ids.h>
8	9	#include <uapi/linux/virtio_input.h>

fs/afs/cmservice.c+1−1

@@ -169,7 +169,7 @@ static int afs_record_cm_probe(struct afs_call call, struct afs_server server) · 168 unmodified lines
169	169
170	170	spin_lock(&server->probe_lock);
171	171
172		− if (!test_bit(AFS_SERVER_FL_HAVE_EPOCH, &server->flags)) {
	172	+ if (!test_and_set_bit(AFS_SERVER_FL_HAVE_EPOCH, &server->flags)) {
173	173	server->cm_epoch = call->epoch;
174	174	server->probe.cm_epoch = call->epoch;
175	175	goto out;

fs/afs/fs_probe.c+1−4

@@ -117,11 +117,8 @@ out: · 116 unmodified lines
117	117	(unsigned int)rtt, ret);
118	118
119	119	have_result \|= afs_fs_probe_done(server);
120		− if (have_result) {
121		− server->probe.have_result = true;
122		− wake_up_var(&server->probe.have_result);
	120	+ if (have_result)
123	121	wake_up_all(&server->probe_wq);
124		− }
125	122	}
126	123
127	124	/*

fs/afs/internal.h+1−3

@@ -533,12 +533,10 @@ struct afs_server { · 532 unmodified lines
533	533	u32 abort_code;
534	534	u32 cm_epoch;
535	535	short error;
536		− bool have_result;
537	536	bool responded:1;
538	537	bool is_yfs:1;
539	538	bool not_yfs:1;
540	539	bool local_failure:1;
541		− bool no_epoch:1;
542	540	bool cm_probed:1;
543	541	bool said_rebooted:1;
544	542	bool said_inconsistent:1;
@@ -1335,7 +1333,7 @@ extern struct afs_volume afs_create_volume(struct afs_fs_context ); · 790 unmodified lines
1335	1333	extern void afs_activate_volume(struct afs_volume *);
1336	1334	extern void afs_deactivate_volume(struct afs_volume *);
1337	1335	extern void afs_put_volume(struct afs_cell , struct afs_volume );
1338		−extern int afs_check_volume_status(struct afs_volume , struct key );
	1336	+extern int afs_check_volume_status(struct afs_volume , struct afs_fs_cursor );
1339	1337
1340	1338	/*
1341	1339	* write.c

fs/afs/rotate.c+3−3

@@ -192,7 +192,7 @@ bool afs_select_fileserver(struct afs_fs_cursor *fc) · 191 unmodified lines
192	192	write_unlock(&vnode->volume->servers_lock);
193	193
194	194	set_bit(AFS_VOLUME_NEEDS_UPDATE, &vnode->volume->flags);
195		− error = afs_check_volume_status(vnode->volume, fc->key);
	195	+ error = afs_check_volume_status(vnode->volume, fc);
196	196	if (error < 0)
197	197	goto failed_set_error;
198	198
@@ -281,7 +281,7 @@ bool afs_select_fileserver(struct afs_fs_cursor *fc) · 82 unmodified lines
281	281
282	282	set_bit(AFS_VOLUME_WAIT, &vnode->volume->flags);
283	283	set_bit(AFS_VOLUME_NEEDS_UPDATE, &vnode->volume->flags);
284		− error = afs_check_volume_status(vnode->volume, fc->key);
	284	+ error = afs_check_volume_status(vnode->volume, fc);
285	285	if (error < 0)
286	286	goto failed_set_error;
287	287
@@ -341,7 +341,7 @@ start: · 53 unmodified lines
341	341	/* See if we need to do an update of the volume record. Note that the
342	342	* volume may have moved or even have been deleted.
343	343	*/
344		− error = afs_check_volume_status(vnode->volume, fc->key);
	344	+ error = afs_check_volume_status(vnode->volume, fc);
345	345	if (error < 0)
346	346	goto failed_set_error;
347	347

fs/afs/server.c+2−5

@@ -594,12 +594,9 @@ retry: · 593 unmodified lines
594	594	}
595	595
596	596	ret = wait_on_bit(&server->flags, AFS_SERVER_FL_UPDATING,
597		− TASK_INTERRUPTIBLE);
	597	+ (fc->flags & AFS_FS_CURSOR_INTR) ?
	598	+ TASK_INTERRUPTIBLE : TASK_UNINTERRUPTIBLE);
598	599	if (ret == -ERESTARTSYS) {
599		− if (!(fc->flags & AFS_FS_CURSOR_INTR) && server->addresses) {
600		− _leave(" = t [intr]");
601		− return true;
602		− }
603	600	fc->error = ret;
604	601	_leave(" = f [intr]");
605	602	return false;

fs/afs/vl_rotate.c+2−2

@@ -302,8 +302,8 @@ static void afs_vl_dump_edestaddrreq(const struct afs_vl_cursor *vc) · 301 unmodified lines
302	302	pr_notice("VC: - nr=%u/%u/%u pf=%u\n",
303	303	a->nr_ipv4, a->nr_addrs, a->max_addrs,
304	304	a->preferred);
305		− pr_notice("VC: - pr=%lx R=%lx F=%lx\n",
306		− a->probed, a->responded, a->failed);
	305	+ pr_notice("VC: - R=%lx F=%lx\n",
	306	+ a->responded, a->failed);
307	307	if (a == vc->ac.alist)
308	308	pr_notice("VC: - current\n");
309	309	}

fs/afs/volume.c+5−3

@@ -281,7 +281,7 @@ error: · 280 unmodified lines
281	281	/*
282	282	* Make sure the volume record is up to date.
283	283	*/
284		−int afs_check_volume_status(struct afs_volume volume, struct key key)
	284	+int afs_check_volume_status(struct afs_volume volume, struct afs_fs_cursor fc)
285	285	{
286	286	time64_t now = ktime_get_real_seconds();
287	287	int ret, retries = 0;
@@ -299,7 +299,7 @@ retry: · 11 unmodified lines
299	299	}
300	300
301	301	if (!test_and_set_bit_lock(AFS_VOLUME_UPDATING, &volume->flags)) {
302		− ret = afs_update_volume_status(volume, key);
	302	+ ret = afs_update_volume_status(volume, fc->key);
303	303	clear_bit_unlock(AFS_VOLUME_WAIT, &volume->flags);
304	304	clear_bit_unlock(AFS_VOLUME_UPDATING, &volume->flags);
305	305	wake_up_bit(&volume->flags, AFS_VOLUME_WAIT);
@@ -312,7 +312,9 @@ retry: · 6 unmodified lines
312	312	return 0;
313	313	}
314	314
315		− ret = wait_on_bit(&volume->flags, AFS_VOLUME_WAIT, TASK_INTERRUPTIBLE);
	315	+ ret = wait_on_bit(&volume->flags, AFS_VOLUME_WAIT,
	316	+ (fc->flags & AFS_FS_CURSOR_INTR) ?
	317	+ TASK_INTERRUPTIBLE : TASK_UNINTERRUPTIBLE);
316	318	if (ret == -ERESTARTSYS) {
317	319	_leave(" = %d", ret);
318	320	return ret;

fs/block_dev.c+10−1

@@ -19,7 +19,6 @@ · 18 unmodified lines
19	19	#include <linux/module.h>
20	20	#include <linux/blkpg.h>
21	21	#include <linux/magic.h>
22		−#include <linux/dax.h>
23	22	#include <linux/buffer_head.h>
24	23	#include <linux/swap.h>
25	24	#include <linux/pagevec.h>
@@ -1893,6 +1892,16 @@ static void __blkdev_put(struct block_device *bdev, fmode_t mode, int for_part) · 1867 unmodified lines
1893	1892	struct gendisk *disk = bdev->bd_disk;
1894	1893	struct block_device *victim = NULL;
1895	1894
	1895	+ /*
	1896	+ * Sync early if it looks like we're the last one. If someone else
	1897	+ * opens the block device between now and the decrement of bd_openers
	1898	+ * then we did a sync that we didn't need to, but that's not the end
	1899	+ * of the world and we want to avoid long (could be several minute)
	1900	+ * syncs while holding the mutex.
	1901	+ */
	1902	+ if (bdev->bd_openers == 1)
	1903	+ sync_blockdev(bdev);
	1904	+
1896	1905	mutex_lock_nested(&bdev->bd_mutex, for_part);
1897	1906	if (for_part)
1898	1907	bdev->bd_part_count--;

fs/btrfs/backref.c+1−1

@@ -391,7 +391,7 @@ static int is_shared_data_backref(struct preftrees *preftrees, u64 bytenr) · 390 unmodified lines
391	391	struct rb_node **p = &preftrees->direct.root.rb_root.rb_node;
392	392	struct rb_node *parent = NULL;
393	393	struct prelim_ref *ref = NULL;
394		− struct prelim_ref target = {0};
	394	+ struct prelim_ref target = {};
395	395	int result;
396	396
397	397	target.parent = bytenr;

fs/btrfs/block-group.c+14−6

@@ -916,7 +916,7 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans, · 915 unmodified lines
916	916	path = btrfs_alloc_path();
917	917	if (!path) {
918	918	ret = -ENOMEM;
919		− goto out;
	919	+ goto out_put_group;
920	920	}
921	921
922	922	/*
@@ -954,7 +954,7 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans, · 31 unmodified lines
954	954	ret = btrfs_orphan_add(trans, BTRFS_I(inode));
955	955	if (ret) {
956	956	btrfs_add_delayed_iput(inode);
957		− goto out;
	957	+ goto out_put_group;
958	958	}
959	959	clear_nlink(inode);
960	960	/* One for the block groups ref */
@@ -977,13 +977,13 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans, · 16 unmodified lines
977	977
978	978	ret = btrfs_search_slot(trans, tree_root, &key, path, -1, 1);
979	979	if (ret < 0)
980		− goto out;
	980	+ goto out_put_group;
981	981	if (ret > 0)
982	982	btrfs_release_path(path);
983	983	if (ret == 0) {
984	984	ret = btrfs_del_item(trans, tree_root, path);
985	985	if (ret)
986		− goto out;
	986	+ goto out_put_group;
987	987	btrfs_release_path(path);
988	988	}
989	989
@@ -1102,9 +1102,9 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans, · 112 unmodified lines
1102	1102
1103	1103	ret = remove_block_group_free_space(trans, block_group);
1104	1104	if (ret)
1105		− goto out;
	1105	+ goto out_put_group;
1106	1106
1107		− btrfs_put_block_group(block_group);
	1107	+ /* Once for the block groups rbtree */
1108	1108	btrfs_put_block_group(block_group);
1109	1109
1110	1110	ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
@@ -1127,6 +1127,10 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans, · 16 unmodified lines
1127	1127	/* once for the tree */
1128	1128	free_extent_map(em);
1129	1129	}
	1130	+
	1131	+out_put_group:
	1132	+ /* Once for the lookup reference */
	1133	+ btrfs_put_block_group(block_group);
1130	1134	out:
1131	1135	if (remove_rsv)
1132	1136	btrfs_delayed_refs_rsv_release(fs_info, 1);
@@ -1288,11 +1292,15 @@ static bool clean_pinned_extents(struct btrfs_trans_handle *trans, · 155 unmodified lines
1288	1292	if (ret)
1289	1293	goto err;
1290	1294	mutex_unlock(&fs_info->unused_bg_unpin_mutex);
	1295	+ if (prev_trans)
	1296	+ btrfs_put_transaction(prev_trans);
1291	1297
1292	1298	return true;
1293	1299
1294	1300	err:
1295	1301	mutex_unlock(&fs_info->unused_bg_unpin_mutex);
	1302	+ if (prev_trans)
	1303	+ btrfs_put_transaction(prev_trans);
1296	1304	btrfs_dec_block_group_ro(bg);
1297	1305	return false;
1298	1306	}

fs/btrfs/discard.h+1−1

@@ -1,4 +1,4 @@
1		−// SPDX-License-Identifier: GPL-2.0
	1	+/* SPDX-License-Identifier: GPL-2.0 */
2	2
3	3	#ifndef BTRFS_DISCARD_H
4	4	#define BTRFS_DISCARD_H

fs/btrfs/disk-io.c+32−4

@@ -2036,9 +2036,6 @@ void btrfs_free_fs_roots(struct btrfs_fs_info *fs_info) · 2035 unmodified lines
2036	2036	for (i = 0; i < ret; i++)
2037	2037	btrfs_drop_and_free_fs_root(fs_info, gang[i]);
2038	2038	}
2039		−
2040		− if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state))
2041		− btrfs_free_log_root_tree(NULL, fs_info);
2042	2039	}
2043	2040
2044	2041	static void btrfs_init_scrub(struct btrfs_fs_info *fs_info)
@@ -3888,7 +3885,7 @@ void btrfs_drop_and_free_fs_root(struct btrfs_fs_info *fs_info, · 1843 unmodified lines
3888	3885	spin_unlock(&fs_info->fs_roots_radix_lock);
3889	3886
3890	3887	if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state)) {
3891		− btrfs_free_log(NULL, root);
	3888	+ ASSERT(root->log_root == NULL);
3892	3889	if (root->reloc_root) {
3893	3890	btrfs_put_root(root->reloc_root);
3894	3891	root->reloc_root = NULL;
@@ -4211,6 +4208,36 @@ static void btrfs_error_commit_super(struct btrfs_fs_info *fs_info) · 316 unmodified lines
4211	4208	up_write(&fs_info->cleanup_work_sem);
4212	4209	}
4213	4210
	4211	+static void btrfs_drop_all_logs(struct btrfs_fs_info *fs_info)
	4212	+{
	4213	+ struct btrfs_root *gang[8];
	4214	+ u64 root_objectid = 0;
	4215	+ int ret;
	4216	+
	4217	+ spin_lock(&fs_info->fs_roots_radix_lock);
	4218	+ while ((ret = radix_tree_gang_lookup(&fs_info->fs_roots_radix,
	4219	+ (void **)gang, root_objectid,
	4220	+ ARRAY_SIZE(gang))) != 0) {
	4221	+ int i;
	4222	+
	4223	+ for (i = 0; i < ret; i++)
	4224	+ gang[i] = btrfs_grab_root(gang[i]);
	4225	+ spin_unlock(&fs_info->fs_roots_radix_lock);
	4226	+
	4227	+ for (i = 0; i < ret; i++) {
	4228	+ if (!gang[i])
	4229	+ continue;
	4230	+ root_objectid = gang[i]->root_key.objectid;
	4231	+ btrfs_free_log(NULL, gang[i]);
	4232	+ btrfs_put_root(gang[i]);
	4233	+ }
	4234	+ root_objectid++;
	4235	+ spin_lock(&fs_info->fs_roots_radix_lock);
	4236	+ }
	4237	+ spin_unlock(&fs_info->fs_roots_radix_lock);
	4238	+ btrfs_free_log_root_tree(NULL, fs_info);
	4239	+}
	4240	+
4214	4241	static void btrfs_destroy_ordered_extents(struct btrfs_root *root)
4215	4242	{
4216	4243	struct btrfs_ordered_extent *ordered;
@@ -4603,6 +4630,7 @@ static int btrfs_cleanup_transaction(struct btrfs_fs_info *fs_info) · 386 unmodified lines
4603	4630	btrfs_destroy_delayed_inodes(fs_info);
4604	4631	btrfs_assert_delayed_root_empty(fs_info);
4605	4632	btrfs_destroy_all_delalloc_inodes(fs_info);
	4633	+ btrfs_drop_all_logs(fs_info);
4606	4634	mutex_unlock(&fs_info->transaction_kthread_mutex);
4607	4635
4608	4636	return 0;

fs/btrfs/relocation.c+1−0

@@ -4559,6 +4559,7 @@ int btrfs_recover_relocation(struct btrfs_root *root) · 4558 unmodified lines
4559	4559	if (IS_ERR(fs_root)) {
4560	4560	err = PTR_ERR(fs_root);
4561	4561	list_add_tail(&reloc_root->root_list, &reloc_roots);
	4562	+ btrfs_end_transaction(trans);
4562	4563	goto out_unset;
4563	4564	}
4564	4565

fs/btrfs/transaction.c+11−2

@@ -662,10 +662,19 @@ again: · 661 unmodified lines
662	662	}
663	663
664	664	got_it:
665		− btrfs_record_root_in_trans(h, root);
666		−
667	665	if (!current->journal_info)
668	666	current->journal_info = h;
	667	+
	668	+ /*
	669	+ * btrfs_record_root_in_trans() needs to alloc new extents, and may
	670	+ * call btrfs_join_transaction() while we're also starting a
	671	+ * transaction.
	672	+ *
	673	+ * Thus it need to be called after current->journal_info initialized,
	674	+ * or we can deadlock.
	675	+ */
	676	+ btrfs_record_root_in_trans(h, root);
	677	+
669	678	return h;
670	679
671	680	join_fail:

fs/btrfs/tree-log.c+40−3

@@ -4226,6 +4226,9 @@ static int btrfs_log_prealloc_extents(struct btrfs_trans_handle *trans, · 4225 unmodified lines
4226	4226	const u64 ino = btrfs_ino(inode);
4227	4227	struct btrfs_path *dst_path = NULL;
4228	4228	bool dropped_extents = false;
	4229	+ u64 truncate_offset = i_size;
	4230	+ struct extent_buffer *leaf;
	4231	+ int slot;
4229	4232	int ins_nr = 0;
4230	4233	int start_slot;
4231	4234	int ret;
@@ -4240,9 +4243,43 @@ static int btrfs_log_prealloc_extents(struct btrfs_trans_handle *trans, · 8 unmodified lines
4240	4243	if (ret < 0)
4241	4244	goto out;
4242	4245
	4246	+ /*
	4247	+ * We must check if there is a prealloc extent that starts before the
	4248	+ * i_size and crosses the i_size boundary. This is to ensure later we
	4249	+ * truncate down to the end of that extent and not to the i_size, as
	4250	+ * otherwise we end up losing part of the prealloc extent after a log
	4251	+ * replay and with an implicit hole if there is another prealloc extent
	4252	+ * that starts at an offset beyond i_size.
	4253	+ */
	4254	+ ret = btrfs_previous_item(root, path, ino, BTRFS_EXTENT_DATA_KEY);
	4255	+ if (ret < 0)
	4256	+ goto out;
	4257	+
	4258	+ if (ret == 0) {
	4259	+ struct btrfs_file_extent_item *ei;
	4260	+
	4261	+ leaf = path->nodes[0];
	4262	+ slot = path->slots[0];
	4263	+ ei = btrfs_item_ptr(leaf, slot, struct btrfs_file_extent_item);
	4264	+
	4265	+ if (btrfs_file_extent_type(leaf, ei) ==
	4266	+ BTRFS_FILE_EXTENT_PREALLOC) {
	4267	+ u64 extent_end;
	4268	+
	4269	+ btrfs_item_key_to_cpu(leaf, &key, slot);
	4270	+ extent_end = key.offset +
	4271	+ btrfs_file_extent_num_bytes(leaf, ei);
	4272	+
	4273	+ if (extent_end > i_size)
	4274	+ truncate_offset = extent_end;
	4275	+ }
	4276	+ } else {
	4277	+ ret = 0;
	4278	+ }
	4279	+
4243	4280	while (true) {
4244		− struct extent_buffer *leaf = path->nodes[0];
4245		− int slot = path->slots[0];
	4281	+ leaf = path->nodes[0];
	4282	+ slot = path->slots[0];
4246	4283
4247	4284	if (slot >= btrfs_header_nritems(leaf)) {
4248	4285	if (ins_nr > 0) {
@@ -4280,7 +4317,7 @@ static int btrfs_log_prealloc_extents(struct btrfs_trans_handle *trans, · 31 unmodified lines
4280	4317	ret = btrfs_truncate_inode_items(trans,
4281	4318	root->log_root,
4282	4319	&inode->vfs_inode,
4283		− i_size,
	4320	+ truncate_offset,
4284	4321	BTRFS_EXTENT_DATA_KEY);
4285	4322	} while (ret == -EAGAIN);
4286	4323	if (ret)

fs/buffer.c+1−1

@@ -967,7 +967,7 @@ grow_dev_page(struct block_device *bdev, sector_t block, · 966 unmodified lines
967	967	struct page *page;
968	968	struct buffer_head *bh;
969	969	sector_t end_block;
970		− int ret = 0; /* Will call free_more_memory() */
	970	+ int ret = 0;
971	971	gfp_t gfp_mask;
972	972
973	973	gfp_mask = mapping_gfp_constraint(inode->i_mapping, ~__GFP_FS) \| gfp;

fs/ceph/caps.c+2−1

@@ -2749,7 +2749,7 @@ int ceph_try_get_caps(struct inode *inode, int need, int want, · 2748 unmodified lines
2749	2749
2750	2750	ret = try_get_cap_refs(inode, need, want, 0, flags, got);
2751	2751	/* three special error codes */
2752		− if (ret == -EAGAIN \|\| ret == -EFBIG \|\| ret == -EAGAIN)
	2752	+ if (ret == -EAGAIN \|\| ret == -EFBIG \|\| ret == -ESTALE)
2753	2753	ret = 0;
2754	2754	return ret;
2755	2755	}
@@ -3746,6 +3746,7 @@ retry: · 990 unmodified lines
3746	3746	WARN_ON(1);
3747	3747	tsession = NULL;
3748	3748	target = -1;
	3749	+ mutex_lock(&session->s_mutex);
3749	3750	}
3750	3751	goto retry;
3751	3752

fs/ceph/debugfs.c+1−1

@@ -271,7 +271,7 @@ void ceph_fs_debugfs_init(struct ceph_fs_client *fsc) · 270 unmodified lines
271	271	&congestion_kb_fops);
272	272
273	273	snprintf(name, sizeof(name), "../../bdi/%s",
274		− dev_name(fsc->sb->s_bdi->dev));
	274	+ bdi_dev_name(fsc->sb->s_bdi));
275	275	fsc->debugfs_bdi =
276	276	debugfs_create_symlink("bdi",
277	277	fsc->client->debugfs_dir,

fs/ceph/mds_client.c+3−5

@@ -3251,8 +3251,7 @@ static void handle_session(struct ceph_mds_session *session, · 3250 unmodified lines
3251	3251	void *end = p + msg->front.iov_len;
3252	3252	struct ceph_mds_session_head *h;
3253	3253	u32 op;
3254		− u64 seq;
3255		− unsigned long features = 0;
	3254	+ u64 seq, features = 0;
3256	3255	int wake = 0;
3257	3256	bool blacklisted = false;
3258	3257
@@ -3271,9 +3270,8 @@ static void handle_session(struct ceph_mds_session *session, · 12 unmodified lines
3271	3270	goto bad;
3272	3271	/* version >= 3, feature bits */
3273	3272	ceph_decode_32_safe(&p, end, len, bad);
3274		− ceph_decode_need(&p, end, len, bad);
3275		− memcpy(&features, p, min_t(size_t, len, sizeof(features)));
3276		− p += len;
	3273	+ ceph_decode_64_safe(&p, end, features, bad);
	3274	+ p += len - sizeof(features);
3277	3275	}
3278	3276
3279	3277	mutex_lock(&mdsc->mutex);

fs/ceph/quota.c+2−2

@@ -159,8 +159,8 @@ static struct inode lookup_quotarealm_inode(struct ceph_mds_client mdsc, · 158 unmodified lines
159	159	}
160	160
161	161	if (IS_ERR(in)) {
162		− pr_warn("Can't lookup inode %llx (err: %ld)\n",
163		− realm->ino, PTR_ERR(in));
	162	+ dout("Can't lookup inode %llx (err: %ld)\n",
	163	+ realm->ino, PTR_ERR(in));
164	164	qri->timeout = jiffies + msecs_to_jiffies(60 * 1000); /* XXX */
165	165	} else {
166	166	qri->timeout = 0;

fs/cifs/cifsglob.h+2−1

@@ -1891,7 +1891,8 @@ GLOBAL_EXTERN struct list_head cifs_tcp_ses_list; · 1890 unmodified lines
1891	1891	/*
1892	1892	* This lock protects the cifs_tcp_ses_list, the list of smb sessions per
1893	1893	* tcp session, and the list of tcon's per smb session. It also protects
1894		− * the reference counters for the server, smb session, and tcon. Finally,
	1894	+ * the reference counters for the server, smb session, and tcon. It also
	1895	+ * protects some fields in the TCP_Server_Info struct such as dstaddr. Finally,
1895	1896	* changes to the tcon->tidStatus should be done while holding this lock.
1896	1897	* generally the locks should be taken in order tcp_ses_lock before
1897	1898	* tcon->open_file_lock and that before file->file_info_lock since the

fs/cifs/connect.c+6−0

@@ -375,8 +375,10 @@ static int reconn_set_ipaddr(struct TCP_Server_Info *server) · 374 unmodified lines
375	375	return rc;
376	376	}
377	377
	378	+ spin_lock(&cifs_tcp_ses_lock);
378	379	rc = cifs_convert_address((struct sockaddr *)&server->dstaddr, ipaddr,
379	380	strlen(ipaddr));
	381	+ spin_unlock(&cifs_tcp_ses_lock);
380	382	kfree(ipaddr);
381	383
382	384	return !rc ? -1 : 0;
@@ -3373,6 +3375,10 @@ cifs_find_tcon(struct cifs_ses ses, struct smb_vol volume_info) · 2990 unmodified lines
3373	3375	spin_lock(&cifs_tcp_ses_lock);
3374	3376	list_for_each(tmp, &ses->tcon_list) {
3375	3377	tcon = list_entry(tmp, struct cifs_tcon, tcon_list);
	3378	+#ifdef CONFIG_CIFS_DFS_UPCALL
	3379	+ if (tcon->dfs_path)
	3380	+ continue;
	3381	+#endif
3376	3382	if (!match_tcon(tcon, volume_info))
3377	3383	continue;
3378	3384	++tcon->tc_count;

fs/cifs/misc.c+65−17

@@ -1025,51 +1025,99 @@ int copy_path_name(char dst, const char src) · 1024 unmodified lines
1025	1025	}
1026	1026
1027	1027	struct super_cb_data {
1028		− struct TCP_Server_Info *server;
	1028	+ void *data;
1029	1029	struct super_block *sb;
1030	1030	};
1031	1031
1032		−static void super_cb(struct super_block sb, void arg)
	1032	+static void tcp_super_cb(struct super_block sb, void arg)
1033	1033	{
1034		− struct super_cb_data *d = arg;
	1034	+ struct super_cb_data *sd = arg;
	1035	+ struct TCP_Server_Info *server = sd->data;
1035	1036	struct cifs_sb_info *cifs_sb;
1036	1037	struct cifs_tcon *tcon;
1037	1038
1038		− if (d->sb)
	1039	+ if (sd->sb)
1039	1040	return;
1040	1041
1041	1042	cifs_sb = CIFS_SB(sb);
1042	1043	tcon = cifs_sb_master_tcon(cifs_sb);
1043		− if (tcon->ses->server == d->server)
1044		− d->sb = sb;
	1044	+ if (tcon->ses->server == server)
	1045	+ sd->sb = sb;
1045	1046	}
1046	1047
1047		−struct super_block cifs_get_tcp_super(struct TCP_Server_Info server)
	1048	+static struct super_block __cifs_get_super(void (f)(struct super_block , void ),
	1049	+ void *data)
1048	1050	{
1049		− struct super_cb_data d = {
1050		− .server = server,
	1051	+ struct super_cb_data sd = {
	1052	+ .data = data,
1051	1053	.sb = NULL,
1052	1054	};
1053	1055
1054		− iterate_supers_type(&cifs_fs_type, super_cb, &d);
	1056	+ iterate_supers_type(&cifs_fs_type, f, &sd);
1055	1057
1056		− if (unlikely(!d.sb))
1057		− return ERR_PTR(-ENOENT);
	1058	+ if (!sd.sb)
	1059	+ return ERR_PTR(-EINVAL);
1058	1060	/*
1059	1061	* Grab an active reference in order to prevent automounts (DFS links)
1060	1062	* of expiring and then freeing up our cifs superblock pointer while
1061	1063	* we're doing failover.
1062	1064	*/
1063		− cifs_sb_active(d.sb);
1064		− return d.sb;
	1065	+ cifs_sb_active(sd.sb);
	1066	+ return sd.sb;
1065	1067	}
1066	1068
1067		−void cifs_put_tcp_super(struct super_block *sb)
	1069	+static void __cifs_put_super(struct super_block *sb)
1068	1070	{
1069	1071	if (!IS_ERR_OR_NULL(sb))
1070	1072	cifs_sb_deactive(sb);
1071	1073	}
1072	1074
	1075	+struct super_block cifs_get_tcp_super(struct TCP_Server_Info server)
	1076	+{
	1077	+ return __cifs_get_super(tcp_super_cb, server);
	1078	+}
	1079	+
	1080	+void cifs_put_tcp_super(struct super_block *sb)
	1081	+{
	1082	+ __cifs_put_super(sb);
	1083	+}
	1084	+
	1085	+#ifdef CONFIG_CIFS_DFS_UPCALL
	1086	+static void tcon_super_cb(struct super_block sb, void arg)
	1087	+{
	1088	+ struct super_cb_data *sd = arg;
	1089	+ struct cifs_tcon *tcon = sd->data;
	1090	+ struct cifs_sb_info *cifs_sb;
	1091	+
	1092	+ if (sd->sb)
	1093	+ return;
	1094	+
	1095	+ cifs_sb = CIFS_SB(sb);
	1096	+ if (tcon->dfs_path && cifs_sb->origin_fullpath &&
	1097	+ !strcasecmp(tcon->dfs_path, cifs_sb->origin_fullpath))
	1098	+ sd->sb = sb;
	1099	+}
	1100	+
	1101	+static inline struct super_block cifs_get_tcon_super(struct cifs_tcon tcon)
	1102	+{
	1103	+ return __cifs_get_super(tcon_super_cb, tcon);
	1104	+}
	1105	+
	1106	+static inline void cifs_put_tcon_super(struct super_block *sb)
	1107	+{
	1108	+ __cifs_put_super(sb);
	1109	+}
	1110	+#else
	1111	+static inline struct super_block cifs_get_tcon_super(struct cifs_tcon tcon)
	1112	+{
	1113	+ return ERR_PTR(-EOPNOTSUPP);
	1114	+}
	1115	+
	1116	+static inline void cifs_put_tcon_super(struct super_block *sb)
	1117	+{
	1118	+}
	1119	+#endif
	1120	+
1073	1121	int update_super_prepath(struct cifs_tcon tcon, const char prefix,
1074	1122	size_t prefix_len)
1075	1123	{
@@ -1077,7 +1125,7 @@ int update_super_prepath(struct cifs_tcon tcon, const char prefix, · 1 unmodified line
1077	1125	struct cifs_sb_info *cifs_sb;
1078	1126	int rc = 0;
1079	1127
1080		− sb = cifs_get_tcp_super(tcon->ses->server);
	1128	+ sb = cifs_get_tcon_super(tcon);
1081	1129	if (IS_ERR(sb))
1082	1130	return PTR_ERR(sb);
1083	1131
@@ -1099,6 +1147,6 @@ int update_super_prepath(struct cifs_tcon tcon, const char prefix, · 15 unmodified lines
1099	1147	cifs_sb->mnt_cifs_flags \|= CIFS_MOUNT_USE_PREFIX_PATH;
1100	1148
1101	1149	out:
1102		− cifs_put_tcp_super(sb);
	1150	+ cifs_put_tcon_super(sb);
1103	1151	return rc;
1104	1152	}

fs/cifs/smb2ops.c+5−0

@@ -687,6 +687,11 @@ int open_shroot(unsigned int xid, struct cifs_tcon *tcon, · 686 unmodified lines
687	687	if (smb3_encryption_required(tcon))
688	688	flags \|= CIFS_TRANSFORM_REQ;
689	689
	690	+ if (!server->ops->new_lease_key)
	691	+ return -EIO;
	692	+
	693	+ server->ops->new_lease_key(pfid);
	694	+
690	695	memset(rqst, 0, sizeof(rqst));
691	696	resp_buftype[0] = resp_buftype[1] = CIFS_NO_BUFFER;
692	697	memset(rsp_iov, 0, sizeof(rsp_iov));

fs/configfs/dir.c+1−0

@@ -1519,6 +1519,7 @@ static int configfs_rmdir(struct inode dir, struct dentry dentry) · 1518 unmodified lines
1519	1519	spin_lock(&configfs_dirent_lock);
1520	1520	configfs_detach_rollback(dentry);
1521	1521	spin_unlock(&configfs_dirent_lock);
	1522	+ config_item_put(parent_item);
1522	1523	return -EINTR;
1523	1524	}
1524	1525	frag->frag_dead = true;

fs/coredump.c+10−0

@@ -211,6 +211,8 @@ static int format_corename(struct core_name cn, struct coredump_params cprm, · 210 unmodified lines
211	211	return -ENOMEM;
212	212	(argv)[(argc)++] = 0;
213	213	++pat_ptr;
	214	+ if (!(*pat_ptr))
	215	+ return -ENOMEM;
214	216	}
215	217
216	218	/* Repeat as long as we have more pattern to process and more output
@@ -786,6 +788,14 @@ void do_coredump(const kernel_siginfo_t *siginfo) · 569 unmodified lines
786	788	if (displaced)
787	789	put_files_struct(displaced);
788	790	if (!dump_interrupted()) {
	791	+ /*
	792	+ * umh disabled with CONFIG_STATIC_USERMODEHELPER_PATH="" would
	793	+ * have this set to NULL.
	794	+ */
	795	+ if (!cprm.file) {
	796	+ pr_info("Core dump to \|%s disabled\n", cn.corename);
	797	+ goto close_fail;
	798	+ }
789	799	file_start_write(cprm.file);
790	800	core_dumped = binfmt->core_dump(&cprm);
791	801	file_end_write(cprm.file);

fs/debugfs/file.c+3−12

@@ -506,20 +506,11 @@ DEFINE_DEBUGFS_ATTRIBUTE(fops_u32_wo, NULL, debugfs_u32_set, "%llu\n"); · 505 unmodified lines
506	506	* This function creates a file in debugfs with the given name that
507	507	* contains the value of the variable @value. If the @mode variable is so
508	508	* set, it can be read from, and written to.
509		− *
510		− * This function will return a pointer to a dentry if it succeeds. This
511		− * pointer must be passed to the debugfs_remove() function when the file is
512		− * to be removed (no automatic cleanup happens if your module is unloaded,
513		− * you are responsible here.) If an error occurs, ERR_PTR(-ERROR) will be
514		− * returned.
515		− *
516		− * If debugfs is not enabled in the kernel, the value ERR_PTR(-ENODEV) will
517		− * be returned.
518	509	*/
519		−struct dentry debugfs_create_u32(const char name, umode_t mode,
520		− struct dentry parent, u32 value)
	510	+void debugfs_create_u32(const char name, umode_t mode, struct dentry parent,
	511	+ u32 *value)
521	512	{
522		− return debugfs_create_mode_unsafe(name, mode, parent, value, &fops_u32,
	513	+ debugfs_create_mode_unsafe(name, mode, parent, value, &fops_u32,
523	514	&fops_u32_ro, &fops_u32_wo);
524	515	}
525	516	EXPORT_SYMBOL_GPL(debugfs_create_u32);

fs/eventpoll.c+33−28

@@ -1171,6 +1171,10 @@ static inline bool chain_epi_lockless(struct epitem *epi) · 1170 unmodified lines
1171	1171	{
1172	1172	struct eventpoll *ep = epi->ep;
1173	1173
	1174	+ /* Fast preliminary check */
	1175	+ if (epi->next != EP_UNACTIVE_PTR)
	1176	+ return false;
	1177	+
1174	1178	/* Check that the same epi has not been just chained from another CPU */
1175	1179	if (cmpxchg(&epi->next, EP_UNACTIVE_PTR, NULL) != EP_UNACTIVE_PTR)
1176	1180	return false;
@@ -1237,16 +1241,12 @@ static int ep_poll_callback(wait_queue_entry_t *wait, unsigned mode, int sync, v · 60 unmodified lines
1237	1241	* chained in ep->ovflist and requeued later on.
1238	1242	*/
1239	1243	if (READ_ONCE(ep->ovflist) != EP_UNACTIVE_PTR) {
1240		− if (epi->next == EP_UNACTIVE_PTR &&
1241		− chain_epi_lockless(epi))
	1244	+ if (chain_epi_lockless(epi))
	1245	+ ep_pm_stay_awake_rcu(epi);
	1246	+ } else if (!ep_is_linked(epi)) {
	1247	+ /* In the usual case, add event to ready list. */
	1248	+ if (list_add_tail_lockless(&epi->rdllink, &ep->rdllist))
1242	1249	ep_pm_stay_awake_rcu(epi);
1243		− goto out_unlock;
1244		− }
1245		−
1246		− /* If this file is already in the ready list we exit soon */
1247		− if (!ep_is_linked(epi) &&
1248		− list_add_tail_lockless(&epi->rdllink, &ep->rdllist)) {
1249		− ep_pm_stay_awake_rcu(epi);
1250	1250	}
1251	1251
1252	1252	/*
@@ -1822,7 +1822,6 @@ static int ep_poll(struct eventpoll ep, struct epoll_event __user events, · 569 unmodified lines
1822	1822	{
1823	1823	int res = 0, eavail, timed_out = 0;
1824	1824	u64 slack = 0;
1825		− bool waiter = false;
1826	1825	wait_queue_entry_t wait;
1827	1826	ktime_t expires, *to = NULL;
1828	1827
@@ -1867,21 +1866,23 @@ fetch_events: · 38 unmodified lines
1867	1866	*/
1868	1867	ep_reset_busy_poll_napi_id(ep);
1869	1868
1870		− /*
1871		− * We don't have any available event to return to the caller. We need
1872		− * to sleep here, and we will be woken by ep_poll_callback() when events
1873		− * become available.
1874		− */
1875		− if (!waiter) {
1876		− waiter = true;
1877		− init_waitqueue_entry(&wait, current);
1878		−
	1869	+ do {
	1870	+ /*
	1871	+ * Internally init_wait() uses autoremove_wake_function(),
	1872	+ * thus wait entry is removed from the wait queue on each
	1873	+ * wakeup. Why it is important? In case of several waiters
	1874	+ * each new wakeup will hit the next waiter, giving it the
	1875	+ * chance to harvest new event. Otherwise wakeup can be
	1876	+ * lost. This is also good performance-wise, because on
	1877	+ * normal wakeup path no need to call __remove_wait_queue()
	1878	+ * explicitly, thus ep->lock is not taken, which halts the
	1879	+ * event delivery.
	1880	+ */
	1881	+ init_wait(&wait);
1879	1882	write_lock_irq(&ep->lock);
1880	1883	__add_wait_queue_exclusive(&ep->wq, &wait);
1881	1884	write_unlock_irq(&ep->lock);
1882		− }
1883	1885
1884		− for (;;) {
1885	1886	/*
1886	1887	* We don't want to sleep if the ep_poll_callback() sends us
1887	1888	* a wakeup in between. That's why we set the task state
@@ -1911,10 +1912,20 @@ fetch_events: · 23 unmodified lines
1911	1912	timed_out = 1;
1912	1913	break;
1913	1914	}
1914		− }
	1915	+
	1916	+ /* We were woken up, thus go and try to harvest some events */
	1917	+ eavail = 1;
	1918	+
	1919	+ } while (0);
1915	1920
1916	1921	__set_current_state(TASK_RUNNING);
1917	1922
	1923	+ if (!list_empty_careful(&wait.entry)) {
	1924	+ write_lock_irq(&ep->lock);
	1925	+ __remove_wait_queue(&ep->wq, &wait);
	1926	+ write_unlock_irq(&ep->lock);
	1927	+ }
	1928	+
1918	1929	send_events:
1919	1930	/*
1920	1931	* Try to transfer events to user space. In case we get 0 events and
@@ -1925,12 +1936,6 @@ send_events: · 4 unmodified lines
1925	1936	!(res = ep_send_events(ep, events, maxevents)) && !timed_out)
1926	1937	goto fetch_events;
1927	1938
1928		− if (waiter) {
1929		− write_lock_irq(&ep->lock);
1930		− __remove_wait_queue(&ep->wq, &wait);
1931		− write_unlock_irq(&ep->lock);
1932		− }
1933		−
1934	1939	return res;
1935	1940	}
1936	1941

fs/exfat/balloc.c+0−3

@@ -91,7 +91,6 @@ static int exfat_allocate_bitmap(struct super_block *sb, · 90 unmodified lines
91	91	}
92	92	}
93	93
94		− sbi->pbr_bh = NULL;
95	94	return 0;
96	95	}
97	96
@@ -137,8 +136,6 @@ void exfat_free_bitmap(struct exfat_sb_info *sbi) · 39 unmodified lines
137	136	{
138	137	int i;
139	138
140		− brelse(sbi->pbr_bh);
141		−
142	139	for (i = 0; i < sbi->map_sectors; i++)
143	140	__brelse(sbi->vol_amap[i]);
144	141

fs/exfat/exfat_fs.h+1−0

@@ -507,6 +507,7 @@ void exfat_msg(struct super_block sb, const char lv, const char *fmt, ...) · 506 unmodified lines
507	507	__printf(3, 4) __cold;
508	508	void exfat_get_entry_time(struct exfat_sb_info sbi, struct timespec64 ts,
509	509	u8 tz, __le16 time, __le16 date, u8 time_ms);
	510	+void exfat_truncate_atime(struct timespec64 *ts);
510	511	void exfat_set_entry_time(struct exfat_sb_info sbi, struct timespec64 ts,
511	512	u8 tz, __le16 time, __le16 date, u8 time_ms);
512	513	unsigned short exfat_calc_chksum_2byte(void *data, int len,

fs/exfat/file.c+2−0

@@ -273,6 +273,7 @@ int exfat_getattr(const struct path path, struct kstat stat, · 272 unmodified lines
273	273	struct exfat_inode_info *ei = EXFAT_I(inode);
274	274
275	275	generic_fillattr(inode, stat);
	276	+ exfat_truncate_atime(&stat->atime);
276	277	stat->result_mask \|= STATX_BTIME;
277	278	stat->btime.tv_sec = ei->i_crtime.tv_sec;
278	279	stat->btime.tv_nsec = ei->i_crtime.tv_nsec;
@@ -339,6 +340,7 @@ int exfat_setattr(struct dentry dentry, struct iattr attr) · 60 unmodified lines
339	340	}
340	341
341	342	setattr_copy(inode, attr);
	343	+ exfat_truncate_atime(&inode->i_atime);
342	344	mark_inode_dirty(inode);
343	345
344	346	out:

fs/exfat/misc.c+13−1

@@ -88,7 +88,8 @@ void exfat_get_entry_time(struct exfat_sb_info sbi, struct timespec64 ts, · 87 unmodified lines
88	88	if (time_ms) {
89	89	ts->tv_sec += time_ms / 100;
90	90	ts->tv_nsec = (time_ms % 100) * 10 * NSEC_PER_MSEC;
91		− }
	91	+ } else
	92	+ ts->tv_nsec = 0;
92	93
93	94	if (tz & EXFAT_TZ_VALID)
94	95	/* Adjust timezone to UTC0. */
@@ -124,6 +125,17 @@ void exfat_set_entry_time(struct exfat_sb_info sbi, struct timespec64 ts, · 29 unmodified lines
124	125	*tz = EXFAT_TZ_VALID;
125	126	}
126	127
	128	+/*
	129	+ * The timestamp for access_time has double seconds granularity.
	130	+ * (There is no 10msIncrement field for access_time unlike create/modify_time)
	131	+ * atime also has only a 2-second resolution.
	132	+ */
	133	+void exfat_truncate_atime(struct timespec64 *ts)
	134	+{
	135	+ ts->tv_sec = round_down(ts->tv_sec, 2);
	136	+ ts->tv_nsec = 0;
	137	+}
	138	+
127	139	unsigned short exfat_calc_chksum_2byte(void *data, int len,
128	140	unsigned short chksum, int type)
129	141	{

fs/exfat/namei.c+7−0

@@ -595,6 +595,7 @@ static int exfat_create(struct inode dir, struct dentry dentry, umode_t mode, · 594 unmodified lines
595	595	inode_inc_iversion(inode);
596	596	inode->i_mtime = inode->i_atime = inode->i_ctime =
597	597	EXFAT_I(inode)->i_crtime = current_time(inode);
	598	+ exfat_truncate_atime(&inode->i_atime);
598	599	/* timestamp is already written, so mark_inode_dirty() is unneeded. */
599	600
600	601	d_instantiate(dentry, inode);
@@ -854,6 +855,7 @@ static int exfat_unlink(struct inode dir, struct dentry dentry) · 253 unmodified lines
854	855
855	856	inode_inc_iversion(dir);
856	857	dir->i_mtime = dir->i_atime = current_time(dir);
	858	+ exfat_truncate_atime(&dir->i_atime);
857	859	if (IS_DIRSYNC(dir))
858	860	exfat_sync_inode(dir);
859	861	else
@@ -861,6 +863,7 @@ static int exfat_unlink(struct inode dir, struct dentry dentry) · 1 unmodified line
861	863
862	864	clear_nlink(inode);
863	865	inode->i_mtime = inode->i_atime = current_time(inode);
	866	+ exfat_truncate_atime(&inode->i_atime);
864	867	exfat_unhash_inode(inode);
865	868	exfat_d_version_set(dentry, inode_query_iversion(dir));
866	869	unlock:
@@ -903,6 +906,7 @@ static int exfat_mkdir(struct inode dir, struct dentry dentry, umode_t mode) · 36 unmodified lines
903	906	inode_inc_iversion(inode);
904	907	inode->i_mtime = inode->i_atime = inode->i_ctime =
905	908	EXFAT_I(inode)->i_crtime = current_time(inode);
	909	+ exfat_truncate_atime(&inode->i_atime);
906	910	/* timestamp is already written, so mark_inode_dirty() is unneeded. */
907	911
908	912	d_instantiate(dentry, inode);
@@ -1019,6 +1023,7 @@ static int exfat_rmdir(struct inode dir, struct dentry dentry) · 110 unmodified lines
1019	1023
1020	1024	inode_inc_iversion(dir);
1021	1025	dir->i_mtime = dir->i_atime = current_time(dir);
	1026	+ exfat_truncate_atime(&dir->i_atime);
1022	1027	if (IS_DIRSYNC(dir))
1023	1028	exfat_sync_inode(dir);
1024	1029	else
@@ -1027,6 +1032,7 @@ static int exfat_rmdir(struct inode dir, struct dentry dentry) · 2 unmodified lines
1027	1032
1028	1033	clear_nlink(inode);
1029	1034	inode->i_mtime = inode->i_atime = current_time(inode);
	1035	+ exfat_truncate_atime(&inode->i_atime);
1030	1036	exfat_unhash_inode(inode);
1031	1037	exfat_d_version_set(dentry, inode_query_iversion(dir));
1032	1038	unlock:
@@ -1387,6 +1393,7 @@ static int exfat_rename(struct inode old_dir, struct dentry old_dentry, · 354 unmodified lines
1387	1393	inode_inc_iversion(new_dir);
1388	1394	new_dir->i_ctime = new_dir->i_mtime = new_dir->i_atime =
1389	1395	EXFAT_I(new_dir)->i_crtime = current_time(new_dir);
	1396	+ exfat_truncate_atime(&new_dir->i_atime);
1390	1397	if (IS_DIRSYNC(new_dir))
1391	1398	exfat_sync_inode(new_dir);
1392	1399	else

fs/exfat/super.c+22−31

@@ -49,6 +49,7 @@ static void exfat_put_super(struct super_block *sb) · 48 unmodified lines
49	49	sync_blockdev(sb->s_bdev);
50	50	exfat_set_vol_flags(sb, VOL_CLEAN);
51	51	exfat_free_bitmap(sbi);
	52	+ brelse(sbi->pbr_bh);
52	53	mutex_unlock(&sbi->s_lock);
53	54
54	55	call_rcu(&sbi->rcu, exfat_delayed_free);
@@ -100,7 +101,7 @@ static int exfat_statfs(struct dentry dentry, struct kstatfs buf) · 45 unmodified lines
100	101	int exfat_set_vol_flags(struct super_block *sb, unsigned short new_flag)
101	102	{
102	103	struct exfat_sb_info *sbi = EXFAT_SB(sb);
103		− struct pbr64 *bpb;
	104	+ struct pbr64 bpb = (struct pbr64 )sbi->pbr_bh->b_data;
104	105	bool sync = 0;
105	106
106	107	/* flags are not changed */
@@ -115,15 +116,6 @@ int exfat_set_vol_flags(struct super_block *sb, unsigned short new_flag) · 8 unmodified lines
115	116	if (sb_rdonly(sb))
116	117	return 0;
117	118
118		− if (!sbi->pbr_bh) {
119		− sbi->pbr_bh = sb_bread(sb, 0);
120		− if (!sbi->pbr_bh) {
121		− exfat_msg(sb, KERN_ERR, "failed to read boot sector");
122		− return -ENOMEM;
123		− }
124		− }
125		−
126		− bpb = (struct pbr64 *)sbi->pbr_bh->b_data;
127	119	bpb->bsx.vol_flags = cpu_to_le16(new_flag);
128	120
129	121	if (new_flag == VOL_DIRTY && !buffer_dirty(sbi->pbr_bh))
@@ -159,7 +151,6 @@ static int exfat_show_options(struct seq_file m, struct dentry root) · 29 unmodified lines
159	151	seq_puts(m, ",iocharset=utf8");
160	152	else if (sbi->nls_io)
161	153	seq_printf(m, ",iocharset=%s", sbi->nls_io->charset);
162		− seq_printf(m, ",bps=%ld", sb->s_blocksize);
163	154	if (opts->errors == EXFAT_ERRORS_CONT)
164	155	seq_puts(m, ",errors=continue");
165	156	else if (opts->errors == EXFAT_ERRORS_PANIC)
@@ -351,14 +342,15 @@ static int exfat_read_root(struct inode *inode) · 185 unmodified lines
351	342	exfat_save_attr(inode, ATTR_SUBDIR);
352	343	inode->i_mtime = inode->i_atime = inode->i_ctime = ei->i_crtime =
353	344	current_time(inode);
	345	+ exfat_truncate_atime(&inode->i_atime);
354	346	exfat_cache_init_inode(inode);
355	347	return 0;
356	348	}
357	349
358		−static struct pbr exfat_read_pbr_with_logical_sector(struct super_block sb,
359		− struct buffer_head **prev_bh)
	350	+static struct pbr exfat_read_pbr_with_logical_sector(struct super_block sb)
360	351	{
361		− struct pbr p_pbr = (struct pbr ) (*prev_bh)->b_data;
	352	+ struct exfat_sb_info *sbi = EXFAT_SB(sb);
	353	+ struct pbr p_pbr = (struct pbr ) (sbi->pbr_bh)->b_data;
362	354	unsigned short logical_sect = 0;
363	355
364	356	logical_sect = 1 << p_pbr->bsx.f64.sect_size_bits;
@@ -378,26 +370,23 @@ static struct pbr exfat_read_pbr_with_logical_sector(struct super_block sb, · 13 unmodified lines
378	370	}
379	371
380	372	if (logical_sect > sb->s_blocksize) {
381		− struct buffer_head *bh = NULL;
382		−
383		− __brelse(*prev_bh);
384		− *prev_bh = NULL;
	373	+ brelse(sbi->pbr_bh);
	374	+ sbi->pbr_bh = NULL;
385	375
386	376	if (!sb_set_blocksize(sb, logical_sect)) {
387	377	exfat_msg(sb, KERN_ERR,
388	378	"unable to set blocksize %u", logical_sect);
389	379	return NULL;
390	380	}
391		− bh = sb_bread(sb, 0);
392		− if (!bh) {
	381	+ sbi->pbr_bh = sb_bread(sb, 0);
	382	+ if (!sbi->pbr_bh) {
393	383	exfat_msg(sb, KERN_ERR,
394	384	"unable to read boot sector (logical sector size = %lu)",
395	385	sb->s_blocksize);
396	386	return NULL;
397	387	}
398	388
399		− *prev_bh = bh;
400		− p_pbr = (struct pbr *) bh->b_data;
	389	+ p_pbr = (struct pbr *)sbi->pbr_bh->b_data;
401	390	}
402	391	return p_pbr;
403	392	}
@@ -408,21 +397,20 @@ static int __exfat_fill_super(struct super_block *sb) · 4 unmodified lines
408	397	int ret;
409	398	struct pbr *p_pbr;
410	399	struct pbr64 *p_bpb;
411		− struct buffer_head *bh;
412	400	struct exfat_sb_info *sbi = EXFAT_SB(sb);
413	401
414	402	/* set block size to read super block */
415	403	sb_min_blocksize(sb, 512);
416	404
417	405	/* read boot sector */
418		− bh = sb_bread(sb, 0);
419		− if (!bh) {
	406	+ sbi->pbr_bh = sb_bread(sb, 0);
	407	+ if (!sbi->pbr_bh) {
420	408	exfat_msg(sb, KERN_ERR, "unable to read boot sector");
421	409	return -EIO;
422	410	}
423	411
424	412	/* PRB is read */
425		− p_pbr = (struct pbr *)bh->b_data;
	413	+ p_pbr = (struct pbr *)sbi->pbr_bh->b_data;
426	414
427	415	/* check the validity of PBR */
428	416	if (le16_to_cpu((p_pbr->signature)) != PBR_SIGNATURE) {
@@ -433,7 +421,7 @@ static int __exfat_fill_super(struct super_block *sb) · 4 unmodified lines
433	421
434	422
435	423	/* check logical sector size */
436		− p_pbr = exfat_read_pbr_with_logical_sector(sb, &bh);
	424	+ p_pbr = exfat_read_pbr_with_logical_sector(sb);
437	425	if (!p_pbr) {
438	426	ret = -EIO;
439	427	goto free_bh;
@@ -514,7 +502,7 @@ free_alloc_bitmap: · 74 unmodified lines
514	502	free_upcase_table:
515	503	exfat_free_upcase_table(sbi);
516	504	free_bh:
517		− brelse(bh);
	505	+ brelse(sbi->pbr_bh);
518	506	return ret;
519	507	}
520	508
@@ -531,17 +519,18 @@ static int exfat_fill_super(struct super_block sb, struct fs_context fc) · 10 unmodified lines
531	519	if (opts->discard) {
532	520	struct request_queue *q = bdev_get_queue(sb->s_bdev);
533	521
534		− if (!blk_queue_discard(q))
	522	+ if (!blk_queue_discard(q)) {
535	523	exfat_msg(sb, KERN_WARNING,
536	524	"mounting with \"discard\" option, but the device does not support discard");
537		− opts->discard = 0;
	525	+ opts->discard = 0;
	526	+ }
538	527	}
539	528
540	529	sb->s_flags \|= SB_NODIRATIME;
541	530	sb->s_magic = EXFAT_SUPER_MAGIC;
542	531	sb->s_op = &exfat_sops;
543	532
544		− sb->s_time_gran = 1;
	533	+ sb->s_time_gran = 10 * NSEC_PER_MSEC;
545	534	sb->s_time_min = EXFAT_MIN_TIMESTAMP_SECS;
546	535	sb->s_time_max = EXFAT_MAX_TIMESTAMP_SECS;
547	536
@@ -605,6 +594,7 @@ put_inode: · 57 unmodified lines
605	594	free_table:
606	595	exfat_free_upcase_table(sbi);
607	596	exfat_free_bitmap(sbi);
	597	+ brelse(sbi->pbr_bh);
608	598
609	599	check_nls_io:
610	600	unload_nls(sbi->nls_io);
@@ -717,6 +707,7 @@ static void __exit exit_exfat_fs(void) · 106 unmodified lines
717	707	module_init(init_exfat_fs);
718	708	module_exit(exit_exfat_fs);
719	709
	710	+MODULE_ALIAS_FS("exfat");
720	711	MODULE_LICENSE("GPL");
721	712	MODULE_DESCRIPTION("exFAT filesystem support");
722	713	MODULE_AUTHOR("Samsung Electronics Co., Ltd.");

fs/io_uring.c+58−75

@@ -524,6 +524,7 @@ enum { · 523 unmodified lines
524	524	REQ_F_OVERFLOW_BIT,
525	525	REQ_F_POLLED_BIT,
526	526	REQ_F_BUFFER_SELECTED_BIT,
	527	+ REQ_F_NO_FILE_TABLE_BIT,
527	528
528	529	/* not a real bit, just to check we're not overflowing the space */
529	530	__REQ_F_LAST_BIT,
@@ -577,6 +578,8 @@ enum { · 47 unmodified lines
577	578	REQ_F_POLLED = BIT(REQ_F_POLLED_BIT),
578	579	/* buffer already selected */
579	580	REQ_F_BUFFER_SELECTED = BIT(REQ_F_BUFFER_SELECTED_BIT),
	581	+ /* doesn't need file table for this request */
	582	+ REQ_F_NO_FILE_TABLE = BIT(REQ_F_NO_FILE_TABLE_BIT),
580	583	};
581	584
582	585	struct async_poll {
@@ -677,8 +680,6 @@ struct io_op_def { · 94 unmodified lines
677	680	unsigned needs_mm : 1;
678	681	/* needs req->file assigned */
679	682	unsigned needs_file : 1;
680		− /* needs req->file assigned IFF fd is >= 0 */
681		− unsigned fd_non_neg : 1;
682	683	/* hash wq insertion if file is a regular file */
683	684	unsigned hash_reg_file : 1;
684	685	/* unbound wq insertion if file is a non-regular file */
@@ -781,8 +782,6 @@ static const struct io_op_def io_op_defs[] = { · 96 unmodified lines
781	782	.needs_file = 1,
782	783	},
783	784	[IORING_OP_OPENAT] = {
784		− .needs_file = 1,
785		− .fd_non_neg = 1,
786	785	.file_table = 1,
787	786	.needs_fs = 1,
788	787	},
@@ -796,9 +795,8 @@ static const struct io_op_def io_op_defs[] = { · 7 unmodified lines
796	795	},
797	796	[IORING_OP_STATX] = {
798	797	.needs_mm = 1,
799		− .needs_file = 1,
800		− .fd_non_neg = 1,
801	798	.needs_fs = 1,
	799	+ .file_table = 1,
802	800	},
803	801	[IORING_OP_READ] = {
804	802	.needs_mm = 1,
@@ -833,8 +831,6 @@ static const struct io_op_def io_op_defs[] = { · 28 unmodified lines
833	831	.buffer_select = 1,
834	832	},
835	833	[IORING_OP_OPENAT2] = {
836		− .needs_file = 1,
837		− .fd_non_neg = 1,
838	834	.file_table = 1,
839	835	.needs_fs = 1,
840	836	},
@@ -1291,7 +1287,7 @@ static struct io_kiocb io_get_fallback_req(struct io_ring_ctx ctx) · 450 unmodified lines
1291	1287	struct io_kiocb *req;
1292	1288
1293	1289	req = ctx->fallback_req;
1294		− if (!test_and_set_bit_lock(0, (unsigned long *) ctx->fallback_req))
	1290	+ if (!test_and_set_bit_lock(0, (unsigned long *) &ctx->fallback_req))
1295	1291	return req;
1296	1292
1297	1293	return NULL;
@@ -1378,7 +1374,7 @@ static void __io_free_req(struct io_kiocb *req) · 80 unmodified lines
1378	1374	if (likely(!io_is_fallback_req(req)))
1379	1375	kmem_cache_free(req_cachep, req);
1380	1376	else
1381		− clear_bit_unlock(0, (unsigned long *) req->ctx->fallback_req);
	1377	+ clear_bit_unlock(0, (unsigned long *) &req->ctx->fallback_req);
1382	1378	}
1383	1379
1384	1380	struct req_batch {
@@ -2034,7 +2030,7 @@ static struct file __io_file_get(struct io_submit_state state, int fd) · 649 unmodified lines
2034	2030	* any file. For now, just ensure that anything potentially problematic is done
2035	2031	* inline.
2036	2032	*/
2037		−static bool io_file_supports_async(struct file *file)
	2033	+static bool io_file_supports_async(struct file *file, int rw)
2038	2034	{
2039	2035	umode_t mode = file_inode(file)->i_mode;
2040	2036
@@ -2043,7 +2039,13 @@ static bool io_file_supports_async(struct file *file) · 2 unmodified lines
2043	2039	if (S_ISREG(mode) && file->f_op != &io_uring_fops)
2044	2040	return true;
2045	2041
2046		− return false;
	2042	+ if (!(file->f_mode & FMODE_NOWAIT))
	2043	+ return false;
	2044	+
	2045	+ if (rw == READ)
	2046	+ return file->f_op->read_iter != NULL;
	2047	+
	2048	+ return file->f_op->write_iter != NULL;
2047	2049	}
2048	2050
2049	2051	static int io_prep_rw(struct io_kiocb req, const struct io_uring_sqe sqe,
@@ -2571,7 +2573,7 @@ static int io_read(struct io_kiocb *req, bool force_nonblock) · 521 unmodified lines
2571	2573	* If the file doesn't support async, mark it as REQ_F_MUST_PUNT so
2572	2574	* we know to async punt it even if it was opened O_NONBLOCK
2573	2575	*/
2574		− if (force_nonblock && !io_file_supports_async(req->file))
	2576	+ if (force_nonblock && !io_file_supports_async(req->file, READ))
2575	2577	goto copy_iov;
2576	2578
2577	2579	iov_count = iov_iter_count(&iter);
@@ -2594,7 +2596,8 @@ copy_iov: · 16 unmodified lines
2594	2596	if (ret)
2595	2597	goto out_free;
2596	2598	/* any defer here is final, must blocking retry */
2597		− if (!(req->flags & REQ_F_NOWAIT))
	2599	+ if (!(req->flags & REQ_F_NOWAIT) &&
	2600	+ !file_can_poll(req->file))
2598	2601	req->flags \|= REQ_F_MUST_PUNT;
2599	2602	return -EAGAIN;
2600	2603	}
@@ -2662,7 +2665,7 @@ static int io_write(struct io_kiocb *req, bool force_nonblock) · 61 unmodified lines
2662	2665	* If the file doesn't support async, mark it as REQ_F_MUST_PUNT so
2663	2666	* we know to async punt it even if it was opened O_NONBLOCK
2664	2667	*/
2665		− if (force_nonblock && !io_file_supports_async(req->file))
	2668	+ if (force_nonblock && !io_file_supports_async(req->file, WRITE))
2666	2669	goto copy_iov;
2667	2670
2668	2671	/* file path doesn't support NOWAIT for non-direct_IO */
@@ -2716,7 +2719,8 @@ copy_iov: · 47 unmodified lines
2716	2719	if (ret)
2717	2720	goto out_free;
2718	2721	/* any defer here is final, must blocking retry */
2719		− req->flags \|= REQ_F_MUST_PUNT;
	2722	+ if (!file_can_poll(req->file))
	2723	+ req->flags \|= REQ_F_MUST_PUNT;
2720	2724	return -EAGAIN;
2721	2725	}
2722	2726	}
@@ -2756,15 +2760,6 @@ static int io_splice_prep(struct io_kiocb req, const struct io_uring_sqe sqe) · 33 unmodified lines
2756	2760	return 0;
2757	2761	}
2758	2762
2759		−static bool io_splice_punt(struct file *file)
2760		−{
2761		− if (get_pipe_info(file))
2762		− return false;
2763		− if (!io_file_supports_async(file))
2764		− return true;
2765		− return !(file->f_flags & O_NONBLOCK);
2766		−}
2767		−
2768	2763	static int io_splice(struct io_kiocb *req, bool force_nonblock)
2769	2764	{
2770	2765	struct io_splice *sp = &req->splice;
@@ -2774,11 +2769,8 @@ static int io_splice(struct io_kiocb *req, bool force_nonblock) · 3 unmodified lines
2774	2769	loff_t poff_in, poff_out;
2775	2770	long ret;
2776	2771
2777		− if (force_nonblock) {
2778		− if (io_splice_punt(in) \|\| io_splice_punt(out))
2779		− return -EAGAIN;
2780		− flags \|= SPLICE_F_NONBLOCK;
2781		− }
	2772	+ if (force_nonblock)
	2773	+ return -EAGAIN;
2782	2774
2783	2775	poff_in = (sp->off_in == -1) ? NULL : &sp->off_in;
2784	2776	poff_out = (sp->off_out == -1) ? NULL : &sp->off_out;
@@ -3355,8 +3347,12 @@ static int io_statx(struct io_kiocb *req, bool force_nonblock) · 570 unmodified lines
3355	3347	struct kstat stat;
3356	3348	int ret;
3357	3349
3358		− if (force_nonblock)
	3350	+ if (force_nonblock) {
	3351	+ /* only need file table for an actual valid fd */
	3352	+ if (ctx->dfd == -1 \|\| ctx->dfd == AT_FDCWD)
	3353	+ req->flags \|= REQ_F_NO_FILE_TABLE;
3359	3354	return -EAGAIN;
	3355	+ }
3360	3356
3361	3357	if (vfs_stat_set_lookup_flags(&lookup_flags, ctx->how.flags))
3362	3358	return -EINVAL;
@@ -3502,7 +3498,7 @@ static void io_sync_file_range_finish(struct io_wq_work **workptr) · 139 unmodified lines
3502	3498	if (io_req_cancelled(req))
3503	3499	return;
3504	3500	__io_sync_file_range(req);
3505		− io_put_req(req); /* put submission ref */
	3501	+ io_steal_work(req, workptr);
3506	3502	}
3507	3503
3508	3504	static int io_sync_file_range(struct io_kiocb *req, bool force_nonblock)
@@ -4200,17 +4196,17 @@ static void io_async_task_func(struct callback_head *cb) · 691 unmodified lines
4200	4196
4201	4197	spin_unlock_irq(&ctx->completion_lock);
4202	4198
	4199	+ /* restore ->work in case we need to retry again */
	4200	+ memcpy(&req->work, &apoll->work, sizeof(req->work));
	4201	+
4203	4202	if (canceled) {
4204	4203	kfree(apoll);
4205	4204	io_cqring_ev_posted(ctx);
4206	4205	req_set_fail_links(req);
4207		− io_put_req(req);
	4206	+ io_double_put_req(req);
4208	4207	return;
4209	4208	}
4210	4209
4211		− /* restore ->work in case we need to retry again */
4212		− memcpy(&req->work, &apoll->work, sizeof(req->work));
4213		−
4214	4210	__set_current_state(TASK_RUNNING);
4215	4211	mutex_lock(&ctx->uring_lock);
4216	4212	__io_queue_sqe(req, NULL);
@@ -4369,7 +4365,7 @@ static bool io_poll_remove_one(struct io_kiocb *req) · 152 unmodified lines
4369	4365
4370	4366	hash_del(&req->hash_node);
4371	4367
4372		− if (apoll) {
	4368	+ if (do_complete && apoll) {
4373	4369	/*
4374	4370	* restore ->work because we need to call io_req_work_drop_env.
4375	4371	*/
@@ -5015,7 +5011,7 @@ static int io_req_defer(struct io_kiocb req, const struct io_uring_sqe sqe) · 639 unmodified lines
5015	5011	int ret;
5016	5012
5017	5013	/* Still need defer if there is pending req in defer list. */
5018		− if (!req_need_defer(req) && list_empty(&ctx->defer_list))
	5014	+ if (!req_need_defer(req) && list_empty_careful(&ctx->defer_list))
5019	5015	return 0;
5020	5016
5021	5017	if (!req->io && io_alloc_async_ctx(req))
@@ -5364,15 +5360,6 @@ static void io_wq_submit_work(struct io_wq_work **workptr) · 342 unmodified lines
5364	5360	io_steal_work(req, workptr);
5365	5361	}
5366	5362
5367		−static int io_req_needs_file(struct io_kiocb *req, int fd)
5368		−{
5369		− if (!io_op_defs[req->opcode].needs_file)
5370		− return 0;
5371		− if ((fd == -1 \|\| fd == AT_FDCWD) && io_op_defs[req->opcode].fd_non_neg)
5372		− return 0;
5373		− return 1;
5374		−}
5375		−
5376	5363	static inline struct file io_file_from_index(struct io_ring_ctx ctx,
5377	5364	int index)
5378	5365	{
@@ -5410,14 +5397,11 @@ static int io_file_get(struct io_submit_state state, struct io_kiocb req, · 31 unmodified lines
5410	5397	}
5411	5398
5412	5399	static int io_req_set_file(struct io_submit_state state, struct io_kiocb req,
5413		− int fd, unsigned int flags)
	5400	+ int fd)
5414	5401	{
5415	5402	bool fixed;
5416	5403
5417		− if (!io_req_needs_file(req, fd))
5418		− return 0;
5419		−
5420		− fixed = (flags & IOSQE_FIXED_FILE);
	5404	+ fixed = (req->flags & REQ_F_FIXED_FILE) != 0;
5421	5405	if (unlikely(!fixed && req->needs_fixed_file))
5422	5406	return -EBADF;
5423	5407
@@ -5429,7 +5413,7 @@ static int io_grab_files(struct io_kiocb *req) · 5 unmodified lines
5429	5413	int ret = -EBADF;
5430	5414	struct io_ring_ctx *ctx = req->ctx;
5431	5415
5432		− if (req->work.files)
	5416	+ if (req->work.files \|\| (req->flags & REQ_F_NO_FILE_TABLE))
5433	5417	return 0;
5434	5418	if (!ctx->ring_file)
5435	5419	return -EBADF;
@@ -5794,7 +5778,7 @@ static int io_init_req(struct io_ring_ctx ctx, struct io_kiocb req, · 358 unmodified lines
5794	5778	struct io_submit_state *state, bool async)
5795	5779	{
5796	5780	unsigned int sqe_flags;
5797		− int id, fd;
	5781	+ int id;
5798	5782
5799	5783	/*
5800	5784	* All io need record the previous position, if LINK vs DARIN,
@@ -5846,8 +5830,10 @@ static int io_init_req(struct io_ring_ctx ctx, struct io_kiocb req, · 45 unmodified lines
5846	5830	IOSQE_ASYNC \| IOSQE_FIXED_FILE \|
5847	5831	IOSQE_BUFFER_SELECT \| IOSQE_IO_LINK);
5848	5832
5849		− fd = READ_ONCE(sqe->fd);
5850		− return io_req_set_file(state, req, fd, sqe_flags);
	5833	+ if (!io_op_defs[req->opcode].needs_file)
	5834	+ return 0;
	5835	+
	5836	+ return io_req_set_file(state, req, READ_ONCE(sqe->fd));
5851	5837	}
5852	5838
5853	5839	static int io_submit_sqes(struct io_ring_ctx *ctx, unsigned int nr,
@@ -7327,7 +7313,7 @@ static void io_ring_ctx_wait_and_kill(struct io_ring_ctx *ctx) · 1473 unmodified lines
7327	7313	* it could cause shutdown to hang.
7328	7314	*/
7329	7315	while (ctx->sqo_thread && !wq_has_sleeper(&ctx->sqo_wait))
7330		− cpu_relax();
	7316	+ cond_resched();
7331	7317
7332	7318	io_kill_timeouts(ctx);
7333	7319	io_poll_remove_all(ctx);
@@ -7356,11 +7342,9 @@ static int io_uring_release(struct inode inode, struct file file) · 22 unmodified lines
7356	7342	static void io_uring_cancel_files(struct io_ring_ctx *ctx,
7357	7343	struct files_struct *files)
7358	7344	{
7359		− struct io_kiocb *req;
7360		− DEFINE_WAIT(wait);
7361		−
7362	7345	while (!list_empty_careful(&ctx->inflight_list)) {
7363		− struct io_kiocb *cancel_req = NULL;
	7346	+ struct io_kiocb cancel_req = NULL, req;
	7347	+ DEFINE_WAIT(wait);
7364	7348
7365	7349	spin_lock_irq(&ctx->inflight_lock);
7366	7350	list_for_each_entry(req, &ctx->inflight_list, inflight_entry) {
@@ -7400,6 +7384,7 @@ static void io_uring_cancel_files(struct io_ring_ctx *ctx, · 33 unmodified lines
7400	7384	*/
7401	7385	if (refcount_sub_and_test(2, &cancel_req->refs)) {
7402	7386	io_put_req(cancel_req);
	7387	+ finish_wait(&ctx->inflight_wait, &wait);
7403	7388	continue;
7404	7389	}
7405	7390	}
@@ -7407,8 +7392,8 @@ static void io_uring_cancel_files(struct io_ring_ctx *ctx, · 1 unmodified line
7407	7392	io_wq_cancel_work(ctx->io_wq, &cancel_req->work);
7408	7393	io_put_req(cancel_req);
7409	7394	schedule();
	7395	+ finish_wait(&ctx->inflight_wait, &wait);
7410	7396	}
7411		− finish_wait(&ctx->inflight_wait, &wait);
7412	7397	}
7413	7398
7414	7399	static int io_uring_flush(struct file file, void data)
@@ -7757,7 +7742,8 @@ err: · 342 unmodified lines
7757	7742	return ret;
7758	7743	}
7759	7744
7760		−static int io_uring_create(unsigned entries, struct io_uring_params *p)
	7745	+static int io_uring_create(unsigned entries, struct io_uring_params *p,
	7746	+ struct io_uring_params __user *params)
7761	7747	{
7762	7748	struct user_struct *user = NULL;
7763	7749	struct io_ring_ctx *ctx;
@@ -7849,6 +7835,14 @@ static int io_uring_create(unsigned entries, struct io_uring_params *p) · 85 unmodified lines
7849	7835	p->cq_off.overflow = offsetof(struct io_rings, cq_overflow);
7850	7836	p->cq_off.cqes = offsetof(struct io_rings, cqes);
7851	7837
	7838	+ p->features = IORING_FEAT_SINGLE_MMAP \| IORING_FEAT_NODROP \|
	7839	+ IORING_FEAT_SUBMIT_STABLE \| IORING_FEAT_RW_CUR_POS \|
	7840	+ IORING_FEAT_CUR_PERSONALITY \| IORING_FEAT_FAST_POLL;
	7841	+
	7842	+ if (copy_to_user(params, p, sizeof(*p))) {
	7843	+ ret = -EFAULT;
	7844	+ goto err;
	7845	+ }
7852	7846	/*
7853	7847	* Install ring fd as the very last thing, so we don't risk someone
7854	7848	* having closed it before we finish setup
@@ -7857,9 +7851,6 @@ static int io_uring_create(unsigned entries, struct io_uring_params *p) · 2 unmodified lines
7857	7851	if (ret < 0)
7858	7852	goto err;
7859	7853
7860		− p->features = IORING_FEAT_SINGLE_MMAP \| IORING_FEAT_NODROP \|
7861		− IORING_FEAT_SUBMIT_STABLE \| IORING_FEAT_RW_CUR_POS \|
7862		− IORING_FEAT_CUR_PERSONALITY \| IORING_FEAT_FAST_POLL;
7863	7854	trace_io_uring_create(ret, ctx, p->sq_entries, p->cq_entries, p->flags);
7864	7855	return ret;
7865	7856	err:
@@ -7875,7 +7866,6 @@ err: · 9 unmodified lines
7875	7866	static long io_uring_setup(u32 entries, struct io_uring_params __user *params)
7876	7867	{
7877	7868	struct io_uring_params p;
7878		− long ret;
7879	7869	int i;
7880	7870
7881	7871	if (copy_from_user(&p, params, sizeof(p)))
@@ -7890,14 +7880,7 @@ static long io_uring_setup(u32 entries, struct io_uring_params __user *params) · 8 unmodified lines
7890	7880	IORING_SETUP_CLAMP \| IORING_SETUP_ATTACH_WQ))
7891	7881	return -EINVAL;
7892	7882
7893		− ret = io_uring_create(entries, &p);
7894		− if (ret < 0)
7895		− return ret;
7896		−
7897		− if (copy_to_user(params, &p, sizeof(p)))
7898		− return -EFAULT;
7899		−
7900		− return ret;
	7883	+ return io_uring_create(entries, &p, params);
7901	7884	}
7902	7885
7903	7886	SYSCALL_DEFINE2(io_uring_setup, u32, entries,

fs/ioctl.c+8−0

@@ -55,6 +55,7 @@ EXPORT_SYMBOL(vfs_ioctl); · 54 unmodified lines
55	55	static int ioctl_fibmap(struct file filp, int __user p)
56	56	{
57	57	struct inode *inode = file_inode(filp);
	58	+ struct super_block *sb = inode->i_sb;
58	59	int error, ur_block;
59	60	sector_t block;
60	61
@@ -71,6 +72,13 @@ static int ioctl_fibmap(struct file filp, int __user p) · 10 unmodified lines
71	72	block = ur_block;
72	73	error = bmap(inode, &block);
73	74
	75	+ if (block > INT_MAX) {
	76	+ error = -ERANGE;
	77	+ pr_warn_ratelimited("[%s/%d] FS: %s File: %pD4 would truncate fibmap result\n",
	78	+ current->comm, task_pid_nr(current),
	79	+ sb->s_id, filp);
	80	+ }
	81	+
74	82	if (error)
75	83	ur_block = 0;
76	84	else

fs/iomap/fiemap.c+1−4

@@ -117,10 +117,7 @@ iomap_bmap_actor(struct inode *inode, loff_t pos, loff_t length, · 116 unmodified lines
117	117
118	118	if (iomap->type == IOMAP_MAPPED) {
119	119	addr = (pos - iomap->offset + iomap->addr) >> inode->i_blkbits;
120		− if (addr > INT_MAX)
121		− WARN(1, "would truncate bmap result\n");
122		− else
123		− *bno = addr;
	120	+ *bno = addr;
124	121	}
125	122	return 0;
126	123	}

fs/nfs/nfs3acl.c+15−7

@@ -253,37 +253,45 @@ int nfs3_proc_setacls(struct inode inode, struct posix_acl acl, · 252 unmodified lines
253	253
254	254	int nfs3_set_acl(struct inode inode, struct posix_acl acl, int type)
255	255	{
256		− struct posix_acl alloc = NULL, dfacl = NULL;
	256	+ struct posix_acl orig = acl, dfacl = NULL, *alloc;
257	257	int status;
258	258
259	259	if (S_ISDIR(inode->i_mode)) {
260	260	switch(type) {
261	261	case ACL_TYPE_ACCESS:
262		− alloc = dfacl = get_acl(inode, ACL_TYPE_DEFAULT);
	262	+ alloc = get_acl(inode, ACL_TYPE_DEFAULT);
263	263	if (IS_ERR(alloc))
264	264	goto fail;
	265	+ dfacl = alloc;
265	266	break;
266	267
267	268	case ACL_TYPE_DEFAULT:
268		− dfacl = acl;
269		− alloc = acl = get_acl(inode, ACL_TYPE_ACCESS);
	269	+ alloc = get_acl(inode, ACL_TYPE_ACCESS);
270	270	if (IS_ERR(alloc))
271	271	goto fail;
	272	+ dfacl = acl;
	273	+ acl = alloc;
272	274	break;
273	275	}
274	276	}
275	277
276	278	if (acl == NULL) {
277		− alloc = acl = posix_acl_from_mode(inode->i_mode, GFP_KERNEL);
	279	+ alloc = posix_acl_from_mode(inode->i_mode, GFP_KERNEL);
278	280	if (IS_ERR(alloc))
279	281	goto fail;
	282	+ acl = alloc;
280	283	}
281	284	status = __nfs3_proc_setacls(inode, acl, dfacl);
282		− posix_acl_release(alloc);
	285	+out:
	286	+ if (acl != orig)
	287	+ posix_acl_release(acl);
	288	+ if (dfacl != orig)
	289	+ posix_acl_release(dfacl);
283	290	return status;
284	291
285	292	fail:
286		− return PTR_ERR(alloc);
	293	+ status = PTR_ERR(alloc);
	294	+ goto out;
287	295	}
288	296
289	297	const struct xattr_handler *nfs3_xattr_handlers[] = {

fs/nfs/nfs4proc.c+9−2

@@ -7891,6 +7891,7 @@ static void · 7890 unmodified lines
7891	7891	nfs4_bind_one_conn_to_session_done(struct rpc_task task, void calldata)
7892	7892	{
7893	7893	struct nfs41_bind_conn_to_session_args *args = task->tk_msg.rpc_argp;
	7894	+ struct nfs41_bind_conn_to_session_res *res = task->tk_msg.rpc_resp;
7894	7895	struct nfs_client *clp = args->client;
7895	7896
7896	7897	switch (task->tk_status) {
@@ -7899,6 +7900,12 @@ nfs4_bind_one_conn_to_session_done(struct rpc_task task, void calldata) · 2 unmodified lines
7899	7900	nfs4_schedule_session_recovery(clp->cl_session,
7900	7901	task->tk_status);
7901	7902	}
	7903	+ if (args->dir == NFS4_CDFC4_FORE_OR_BOTH &&
	7904	+ res->dir != NFS4_CDFS4_BOTH) {
	7905	+ rpc_task_close_connection(task);
	7906	+ if (args->retries++ < MAX_BIND_CONN_TO_SESSION_RETRIES)
	7907	+ rpc_restart_call(task);
	7908	+ }
7902	7909	}
7903	7910
7904	7911	static const struct rpc_call_ops nfs4_bind_one_conn_to_session_ops = {
@@ -7921,6 +7928,7 @@ int nfs4_proc_bind_one_conn_to_session(struct rpc_clnt *clnt, · 16 unmodified lines
7921	7928	struct nfs41_bind_conn_to_session_args args = {
7922	7929	.client = clp,
7923	7930	.dir = NFS4_CDFC4_FORE_OR_BOTH,
	7931	+ .retries = 0,
7924	7932	};
7925	7933	struct nfs41_bind_conn_to_session_res res;
7926	7934	struct rpc_message msg = {
@@ -9191,8 +9199,7 @@ nfs4_proc_layoutget(struct nfs4_layoutget lgp, long timeout) · 1264 unmodified lines
9191	9199	nfs4_init_sequence(&lgp->args.seq_args, &lgp->res.seq_res, 0, 0);
9192	9200
9193	9201	task = rpc_run_task(&task_setup_data);
9194		− if (IS_ERR(task))
9195		− return ERR_CAST(task);
	9202	+
9196	9203	status = rpc_wait_for_completion_task(task);
9197	9204	if (status != 0)
9198	9205	goto out;

fs/nfs/pnfs.c+5−6

@@ -1332,13 +1332,15 @@ _pnfs_return_layout(struct inode *ino) · 1331 unmodified lines
1332	1332	!valid_layout) {
1333	1333	spin_unlock(&ino->i_lock);
1334	1334	dprintk("NFS: %s no layout segments to return\n", __func__);
1335		− goto out_put_layout_hdr;
	1335	+ goto out_wait_layoutreturn;
1336	1336	}
1337	1337
1338	1338	send = pnfs_prepare_layoutreturn(lo, &stateid, &cred, NULL);
1339	1339	spin_unlock(&ino->i_lock);
1340	1340	if (send)
1341	1341	status = pnfs_send_layoutreturn(lo, &stateid, &cred, IOMODE_ANY, true);
	1342	+out_wait_layoutreturn:
	1343	+ wait_on_bit(&lo->plh_flags, NFS_LAYOUT_RETURN, TASK_UNINTERRUPTIBLE);
1342	1344	out_put_layout_hdr:
1343	1345	pnfs_free_lseg_list(&tmp_list);
1344	1346	pnfs_put_layout_hdr(lo);
@@ -1456,18 +1458,15 @@ retry: · 111 unmodified lines
1456	1458	/* lo ref dropped in pnfs_roc_release() */
1457	1459	layoutreturn = pnfs_prepare_layoutreturn(lo, &stateid, &lc_cred, &iomode);
1458	1460	/* If the creds don't match, we can't compound the layoutreturn */
1459		− if (!layoutreturn)
	1461	+ if (!layoutreturn \|\| cred_fscmp(cred, lc_cred) != 0)
1460	1462	goto out_noroc;
1461		− if (cred_fscmp(cred, lc_cred) != 0)
1462		− goto out_noroc_put_cred;
1463	1463
1464	1464	roc = layoutreturn;
1465	1465	pnfs_init_layoutreturn_args(args, lo, &stateid, iomode);
1466	1466	res->lrs_present = 0;
1467	1467	layoutreturn = false;
1468		−
1469		−out_noroc_put_cred:
1470	1468	put_cred(lc_cred);
	1469	+
1471	1470	out_noroc:
1472	1471	spin_unlock(&ino->i_lock);
1473	1472	rcu_read_unlock();

fs/nfs/super.c+1−1

@@ -185,7 +185,7 @@ static int __nfs_list_for_each_server(struct list_head *head, · 184 unmodified lines
185	185
186	186	rcu_read_lock();
187	187	list_for_each_entry_rcu(server, head, client_link) {
188		− if (!nfs_sb_active(server->super))
	188	+ if (!(server->super && nfs_sb_active(server->super)))
189	189	continue;
190	190	rcu_read_unlock();
191	191	if (last)

fs/nfsd/nfs4callback.c+3−1

@@ -1312,6 +1312,7 @@ nfsd4_run_cb_work(struct work_struct *work) · 1311 unmodified lines
1312	1312	container_of(work, struct nfsd4_callback, cb_work);
1313	1313	struct nfs4_client *clp = cb->cb_clp;
1314	1314	struct rpc_clnt *clnt;
	1315	+ int flags;
1315	1316
1316	1317	if (cb->cb_need_restart) {
1317	1318	cb->cb_need_restart = false;
@@ -1340,7 +1341,8 @@ nfsd4_run_cb_work(struct work_struct *work) · 22 unmodified lines
1340	1341	}
1341	1342
1342	1343	cb->cb_msg.rpc_cred = clp->cl_cb_cred;
1343		− rpc_call_async(clnt, &cb->cb_msg, RPC_TASK_SOFT \| RPC_TASK_SOFTCONN,
	1344	+ flags = clp->cl_minorversion ? RPC_TASK_NOCONNECT : RPC_TASK_SOFTCONN;
	1345	+ rpc_call_async(clnt, &cb->cb_msg, RPC_TASK_SOFT \| flags,
1344	1346	cb->cb_ops ? &nfsd4_cb_ops : &nfsd4_cb_probe_ops, cb);
1345	1347	}
1346	1348

fs/nfsd/nfs4state.c+2−0

@@ -267,6 +267,8 @@ find_or_allocate_block(struct nfs4_lockowner lo, struct knfsd_fh fh, · 266 unmodified lines
267	267	if (!nbl) {
268	268	nbl= kmalloc(sizeof(*nbl), GFP_KERNEL);
269	269	if (nbl) {
	270	+ INIT_LIST_HEAD(&nbl->nbl_list);
	271	+ INIT_LIST_HEAD(&nbl->nbl_lru);
270	272	fh_copy_shallow(&nbl->nbl_fh, fh);
271	273	locks_init_lock(&nbl->nbl_lock);
272	274	nfsd4_init_cb(&nbl->nbl_cb, lo->lo_owner.so_client,

fs/ocfs2/dlmfs/dlmfs.c+12−15

@@ -275,7 +275,6 @@ static ssize_t dlmfs_file_write(struct file *filp, · 274 unmodified lines
275	275	loff_t *ppos)
276	276	{
277	277	int bytes_left;
278		− ssize_t writelen;
279	278	char *lvb_buf;
280	279	struct inode *inode = file_inode(filp);
281	280
@@ -285,32 +284,30 @@ static ssize_t dlmfs_file_write(struct file *filp, · 3 unmodified lines
285	284	if (*ppos >= i_size_read(inode))
286	285	return -ENOSPC;
287	286
	287	+ /* don't write past the lvb */
	288	+ if (count > i_size_read(inode) - *ppos)
	289	+ count = i_size_read(inode) - *ppos;
	290	+
288	291	if (!count)
289	292	return 0;
290	293
291	294	if (!access_ok(buf, count))
292	295	return -EFAULT;
293	296
294		− /* don't write past the lvb */
295		− if ((count + *ppos) > i_size_read(inode))
296		− writelen = i_size_read(inode) - *ppos;
297		− else
298		− writelen = count - *ppos;
299		−
300		− lvb_buf = kmalloc(writelen, GFP_NOFS);
	297	+ lvb_buf = kmalloc(count, GFP_NOFS);
301	298	if (!lvb_buf)
302	299	return -ENOMEM;
303	300
304		− bytes_left = copy_from_user(lvb_buf, buf, writelen);
305		− writelen -= bytes_left;
306		− if (writelen)
307		− user_dlm_write_lvb(inode, lvb_buf, writelen);
	301	+ bytes_left = copy_from_user(lvb_buf, buf, count);
	302	+ count -= bytes_left;
	303	+ if (count)
	304	+ user_dlm_write_lvb(inode, lvb_buf, count);
308	305
309	306	kfree(lvb_buf);
310	307
311		− ppos = ppos + writelen;
312		− mlog(0, "wrote %zd bytes\n", writelen);
313		− return writelen;
	308	+ ppos = ppos + count;
	309	+ mlog(0, "wrote %zu bytes\n", count);
	310	+ return count;
314	311	}
315	312
316	313	static void dlmfs_init_once(void *foo)

fs/pnode.c+4−5

@@ -261,14 +261,13 @@ static int propagate_one(struct mount *m) · 260 unmodified lines
261	261	child = copy_tree(last_source, last_source->mnt.mnt_root, type);
262	262	if (IS_ERR(child))
263	263	return PTR_ERR(child);
	264	+ read_seqlock_excl(&mount_lock);
264	265	mnt_set_mountpoint(m, mp, child);
	266	+ if (m->mnt_master != dest_master)
	267	+ SET_MNT_MARK(m->mnt_master);
	268	+ read_sequnlock_excl(&mount_lock);
265	269	last_dest = m;
266	270	last_source = child;
267		− if (m->mnt_master != dest_master) {
268		− read_seqlock_excl(&mount_lock);
269		− SET_MNT_MARK(m->mnt_master);
270		− read_sequnlock_excl(&mount_lock);
271		− }
272	271	hlist_add_head(&child->mnt_hash, list);
273	272	return count_mounts(m->mnt_ns, child);
274	273	}

fs/proc/base.c+0−1

@@ -3286,7 +3286,6 @@ static const struct inode_operations proc_tgid_base_inode_operations = { · 3285 unmodified lines
3286	3286	void proc_flush_pid(struct pid *pid)
3287	3287	{
3288	3288	proc_invalidate_siblings_dcache(&pid->inodes, &pid->lock);
3289		− put_pid(pid);
3290	3289	}
3291	3290
3292	3291	static struct dentry proc_pid_instantiate(struct dentry dentry,

fs/proc/vmcore.c+3−2

@@ -266,7 +266,8 @@ static int vmcoredd_mmap_dumps(struct vm_area_struct *vma, unsigned long dst, · 265 unmodified lines
266	266	if (start < offset + dump->size) {
267	267	tsz = min(offset + (u64)dump->size - start, (u64)size);
268	268	buf = dump->buf + start - offset;
269		− if (remap_vmalloc_range_partial(vma, dst, buf, tsz)) {
	269	+ if (remap_vmalloc_range_partial(vma, dst, buf, 0,
	270	+ tsz)) {
270	271	ret = -EFAULT;
271	272	goto out_unlock;
272	273	}
@@ -624,7 +625,7 @@ static int mmap_vmcore(struct file file, struct vm_area_struct vma) · 351 unmodified lines
624	625	tsz = min(elfcorebuf_sz + elfnotes_sz - (size_t)start, size);
625	626	kaddr = elfnotes_buf + start - elfcorebuf_sz - vmcoredd_orig_sz;
626	627	if (remap_vmalloc_range_partial(vma, vma->vm_start + len,
627		− kaddr, tsz))
	628	+ kaddr, 0, tsz))
628	629	goto fail;
629	630
630	631	size -= tsz;

fs/splice.c+18−27

@@ -1118,6 +1118,10 @@ long do_splice(struct file in, loff_t __user off_in, · 1117 unmodified lines
1118	1118	loff_t offset;
1119	1119	long ret;
1120	1120
	1121	+ if (unlikely(!(in->f_mode & FMODE_READ) \|\|
	1122	+ !(out->f_mode & FMODE_WRITE)))
	1123	+ return -EBADF;
	1124	+
1121	1125	ipipe = get_pipe_info(in);
1122	1126	opipe = get_pipe_info(out);
1123	1127
@@ -1125,12 +1129,6 @@ long do_splice(struct file in, loff_t __user off_in, · 1 unmodified line
1125	1129	if (off_in \|\| off_out)
1126	1130	return -ESPIPE;
1127	1131
1128		− if (!(in->f_mode & FMODE_READ))
1129		− return -EBADF;
1130		−
1131		− if (!(out->f_mode & FMODE_WRITE))
1132		− return -EBADF;
1133		−
1134	1132	/* Splicing to self would be fun, but... */
1135	1133	if (ipipe == opipe)
1136	1134	return -EINVAL;
@@ -1153,9 +1151,6 @@ long do_splice(struct file in, loff_t __user off_in, · 16 unmodified lines
1153	1151	offset = out->f_pos;
1154	1152	}
1155	1153
1156		− if (unlikely(!(out->f_mode & FMODE_WRITE)))
1157		− return -EBADF;
1158		−
1159	1154	if (unlikely(out->f_flags & O_APPEND))
1160	1155	return -EINVAL;
1161	1156
@@ -1440,15 +1435,11 @@ SYSCALL_DEFINE6(splice, int, fd_in, loff_t __user *, off_in, · 278 unmodified lines
1440	1435	error = -EBADF;
1441	1436	in = fdget(fd_in);
1442	1437	if (in.file) {
1443		− if (in.file->f_mode & FMODE_READ) {
1444		− out = fdget(fd_out);
1445		− if (out.file) {
1446		− if (out.file->f_mode & FMODE_WRITE)
1447		− error = do_splice(in.file, off_in,
1448		− out.file, off_out,
1449		− len, flags);
1450		− fdput(out);
1451		− }
	1438	+ out = fdget(fd_out);
	1439	+ if (out.file) {
	1440	+ error = do_splice(in.file, off_in, out.file, off_out,
	1441	+ len, flags);
	1442	+ fdput(out);
1452	1443	}
1453	1444	fdput(in);
1454	1445	}
@@ -1770,6 +1761,10 @@ static long do_tee(struct file in, struct file out, size_t len, · 315 unmodified lines
1770	1761	struct pipe_inode_info *opipe = get_pipe_info(out);
1771	1762	int ret = -EINVAL;
1772	1763
	1764	+ if (unlikely(!(in->f_mode & FMODE_READ) \|\|
	1765	+ !(out->f_mode & FMODE_WRITE)))
	1766	+ return -EBADF;
	1767	+
1773	1768	/*
1774	1769	* Duplicate the contents of ipipe to opipe without actually
1775	1770	* copying the data.
@@ -1795,7 +1790,7 @@ static long do_tee(struct file in, struct file out, size_t len, · 19 unmodified lines
1795	1790
1796	1791	SYSCALL_DEFINE4(tee, int, fdin, int, fdout, size_t, len, unsigned int, flags)
1797	1792	{
1798		− struct fd in;
	1793	+ struct fd in, out;
1799	1794	int error;
1800	1795
1801	1796	if (unlikely(flags & ~SPLICE_F_ALL))
@@ -1807,14 +1802,10 @@ SYSCALL_DEFINE4(tee, int, fdin, int, fdout, size_t, len, unsigned int, flags) · 5 unmodified lines
1807	1802	error = -EBADF;
1808	1803	in = fdget(fdin);
1809	1804	if (in.file) {
1810		− if (in.file->f_mode & FMODE_READ) {
1811		− struct fd out = fdget(fdout);
1812		− if (out.file) {
1813		− if (out.file->f_mode & FMODE_WRITE)
1814		− error = do_tee(in.file, out.file,
1815		− len, flags);
1816		− fdput(out);
1817		− }
	1805	+ out = fdget(fdout);
	1806	+ if (out.file) {
	1807	+ error = do_tee(in.file, out.file, len, flags);
	1808	+ fdput(out);
1818	1809	}
1819	1810	fdput(in);
1820	1811	}

fs/super.c+1−1

@@ -1302,8 +1302,8 @@ int get_tree_bdev(struct fs_context *fc, · 1301 unmodified lines
1302	1302	mutex_lock(&bdev->bd_fsfreeze_mutex);
1303	1303	if (bdev->bd_fsfreeze_count > 0) {
1304	1304	mutex_unlock(&bdev->bd_fsfreeze_mutex);
1305		− blkdev_put(bdev, mode);
1306	1305	warnf(fc, "%pg: Can't mount, blockdev is frozen", bdev);
	1306	+ blkdev_put(bdev, mode);
1307	1307	return -EBUSY;
1308	1308	}
1309	1309

fs/vboxsf/super.c+1−1

@@ -164,7 +164,7 @@ static int vboxsf_fill_super(struct super_block sb, struct fs_context fc) · 163 unmodified lines
164	164	goto fail_free;
165	165	}
166	166
167		− err = super_setup_bdi_name(sb, "vboxsf-%s.%d", fc->source, sbi->bdi_id);
	167	+ err = super_setup_bdi_name(sb, "vboxsf-%d", sbi->bdi_id);
168	168	if (err)
169	169	goto fail_free;
170	170

include/asm-generic/Kbuild+1−0

@@ -56,6 +56,7 @@ mandatory-y += topology.h · 55 unmodified lines
56	56	mandatory-y += trace_clock.h
57	57	mandatory-y += uaccess.h
58	58	mandatory-y += unaligned.h
	59	+mandatory-y += vermagic.h
59	60	mandatory-y += vga.h
60	61	mandatory-y += word-at-a-time.h
61	62	mandatory-y += xor.h

include/asm-generic/sections.h+3−0

@@ -53,6 +53,9 @@ extern char __ctors_start[], __ctors_end[]; · 52 unmodified lines
53	53	/* Start and end of .opd section - used for function descriptors. */
54	54	extern char __start_opd[], __end_opd[];
55	55
	56	+/* Start and end of instrumentation protected text section */
	57	+extern char __noinstr_text_start[], __noinstr_text_end[];
	58	+
56	59	extern __visible const void __nosave_begin, __nosave_end;
57	60
58	61	/* Function descriptor handling (if any). Override in asm/sections.h */

include/asm-generic/vermagic.h+7−0

@@ -0,0 +1,7 @@
	1	+/* SPDX-License-Identifier: GPL-2.0-only */
	2	+#ifndef _ASM_GENERIC_VERMAGIC_H
	3	+#define _ASM_GENERIC_VERMAGIC_H
	4	+
	5	+#define MODULE_ARCH_VERMAGIC ""
	6	+
	7	+#endif /* _ASM_GENERIC_VERMAGIC_H */

include/asm-generic/vmlinux.lds.h+10−0

@@ -540,6 +540,15 @@ · 539 unmodified lines
540	540	. = ALIGN((align)); \
541	541	__end_rodata = .;
542	542
	543	+/*
	544	+ * Non-instrumentable text section
	545	+ */
	546	+#define NOINSTR_TEXT \
	547	+ ALIGN_FUNCTION(); \
	548	+ __noinstr_text_start = .; \
	549	+ *(.noinstr.text) \
	550	+ __noinstr_text_end = .;
	551	+
543	552	/*
544	553	* .text section. Map to function alignment to avoid address changes
545	554	* during second ld run in second ld pass when generating System.map
@@ -551,6 +560,7 @@ · 5 unmodified lines
551	560	#define TEXT_TEXT \
552	561	ALIGN_FUNCTION(); \
553	562	*(.text.hot TEXT_MAIN .text.fixup .text.unlikely) \
	563	+ NOINSTR_TEXT \
554	564	*(.text..refcount) \
555	565	*(.ref.text) \
556	566	MEM_KEEP(init.text*) \

include/linux/amba/bus.h+1−0

@@ -65,6 +65,7 @@ struct amba_device { · 64 unmodified lines
65	65	struct device dev;
66	66	struct resource res;
67	67	struct clk *pclk;
	68	+ struct device_dma_parameters dma_parms;
68	69	unsigned int periphid;
69	70	unsigned int cid;
70	71	struct amba_cs_uci_id uci;

include/linux/backing-dev-defs.h+1−1

@@ -54,7 +54,6 @@ enum wb_reason { · 53 unmodified lines
54	54	WB_REASON_SYNC,
55	55	WB_REASON_PERIODIC,
56	56	WB_REASON_LAPTOP_TIMER,
57		− WB_REASON_FREE_MORE_MEM,
58	57	WB_REASON_FS_FREE_SPACE,
59	58	/*
60	59	* There is no bdi forker thread any more and works are done
@@ -220,6 +219,7 @@ struct backing_dev_info { · 159 unmodified lines
220	219	wait_queue_head_t wb_waitq;
221	220
222	221	struct device *dev;
	222	+ char dev_name[64];
223	223	struct device *owner;
224	224
225	225	struct timer_list laptop_mode_wb_timer;

include/linux/backing-dev.h+1−8

@@ -505,13 +505,6 @@ static inline int bdi_rw_congested(struct backing_dev_info *bdi) · 504 unmodified lines
505	505	(1 << WB_async_congested));
506	506	}
507	507
508		−extern const char *bdi_unknown_name;
509		−
510		−static inline const char bdi_dev_name(struct backing_dev_info bdi)
511		−{
512		− if (!bdi \|\| !bdi->dev)
513		− return bdi_unknown_name;
514		− return dev_name(bdi->dev);
515		−}
	508	+const char bdi_dev_name(struct backing_dev_info bdi);
516	509
517	510	#endif /* _LINUX_BACKING_DEV_H */

include/linux/compiler.h+53−0

@@ -120,12 +120,65 @@ void ftrace_likely_update(struct ftrace_likely_data *f, int val, · 119 unmodified lines
120	120	/* Annotate a C jump table to allow objtool to follow the code flow */
121	121	#define __annotate_jump_table __section(.rodata..c_jump_table)
122	122
	123	+#ifdef CONFIG_DEBUG_ENTRY
	124	+/* Begin/end of an instrumentation safe region */
	125	+#define instrumentation_begin() ({ \
	126	+ asm volatile("%c0:\n\t" \
	127	+ ".pushsection .discard.instr_begin\n\t" \
	128	+ ".long %c0b - .\n\t" \
	129	+ ".popsection\n\t" : : "i" (__COUNTER__)); \
	130	+})
	131	+
	132	+/*
	133	+ * Because instrumentation_{begin,end}() can nest, objtool validation considers
	134	+ * _begin() a +1 and _end() a -1 and computes a sum over the instructions.
	135	+ * When the value is greater than 0, we consider instrumentation allowed.
	136	+ *
	137	+ * There is a problem with code like:
	138	+ *
	139	+ * noinstr void foo()
	140	+ * {
	141	+ * instrumentation_begin();
	142	+ * ...
	143	+ * if (cond) {
	144	+ * instrumentation_begin();
	145	+ * ...
	146	+ * instrumentation_end();
	147	+ * }
	148	+ * bar();
	149	+ * instrumentation_end();
	150	+ * }
	151	+ *
	152	+ * If instrumentation_end() would be an empty label, like all the other
	153	+ * annotations, the inner _end(), which is at the end of a conditional block,
	154	+ * would land on the instruction after the block.
	155	+ *
	156	+ * If we then consider the sum of the !cond path, we'll see that the call to
	157	+ * bar() is with a 0-value, even though, we meant it to happen with a positive
	158	+ * value.
	159	+ *
	160	+ * To avoid this, have _end() be a NOP instruction, this ensures it will be
	161	+ * part of the condition block and does not escape.
	162	+ */
	163	+#define instrumentation_end() ({ \
	164	+ asm volatile("%c0: nop\n\t" \
	165	+ ".pushsection .discard.instr_end\n\t" \
	166	+ ".long %c0b - .\n\t" \
	167	+ ".popsection\n\t" : : "i" (__COUNTER__)); \
	168	+})
	169	+#endif /* CONFIG_DEBUG_ENTRY */
	170	+
123	171	#else
124	172	#define annotate_reachable()
125	173	#define annotate_unreachable()
126	174	#define __annotate_jump_table
127	175	#endif
128	176
	177	+#ifndef instrumentation_begin
	178	+#define instrumentation_begin() do { } while(0)
	179	+#define instrumentation_end() do { } while(0)
	180	+#endif
	181	+
129	182	#ifndef ASM_UNREACHABLE
130	183	# define ASM_UNREACHABLE
131	184	#endif

include/linux/compiler_types.h+4−0

@@ -118,6 +118,10 @@ struct ftrace_likely_data { · 117 unmodified lines
118	118	#define notrace __attribute__((__no_instrument_function__))
119	119	#endif
120	120
	121	+/* Section for code which can't be instrumented at all */
	122	+#define noinstr \
	123	+ noinline notrace __attribute((__section__(".noinstr.text")))
	124	+
121	125	/*
122	126	* it doesn't make sense on ARM (currently the only user of __naked)
123	127	* to trace naked functions because then mcount is called without

include/linux/debugfs.h+4−8

@@ -103,8 +103,8 @@ void debugfs_create_u8(const char name, umode_t mode, struct dentry parent, · 102 unmodified lines
103	103	u8 *value);
104	104	void debugfs_create_u16(const char name, umode_t mode, struct dentry parent,
105	105	u16 *value);
106		−struct dentry debugfs_create_u32(const char name, umode_t mode,
107		− struct dentry parent, u32 value);
	106	+void debugfs_create_u32(const char name, umode_t mode, struct dentry parent,
	107	+ u32 *value);
108	108	void debugfs_create_u64(const char name, umode_t mode, struct dentry parent,
109	109	u64 *value);
110	110	struct dentry debugfs_create_ulong(const char name, umode_t mode,
@@ -250,12 +250,8 @@ static inline void debugfs_create_u8(const char *name, umode_t mode, · 139 unmodified lines
250	250	static inline void debugfs_create_u16(const char *name, umode_t mode,
251	251	struct dentry parent, u16 value) { }
252	252
253		−static inline struct dentry debugfs_create_u32(const char name, umode_t mode,
254		− struct dentry *parent,
255		− u32 *value)
256		−{
257		− return ERR_PTR(-ENODEV);
258		−}
	253	+static inline void debugfs_create_u32(const char *name, umode_t mode,
	254	+ struct dentry parent, u32 value) { }
259	255
260	256	static inline void debugfs_create_u64(const char *name, umode_t mode,
261	257	struct dentry parent, u64 value) { }

include/linux/dma-buf.h+1−2

@@ -329,13 +329,12 @@ struct dma_buf { · 328 unmodified lines
329	329
330	330	/**
331	331	* struct dma_buf_attach_ops - importer operations for an attachment
332		− * @move_notify: [optional] notification that the DMA-buf is moving
333	332	*
334	333	* Attachment operations implemented by the importer.
335	334	*/
336	335	struct dma_buf_attach_ops {
337	336	/**
338		− * @move_notify
	337	+ * @move_notify: [optional] notification that the DMA-buf is moving
339	338	*
340	339	* If this callback is provided the framework can avoid pinning the
341	340	* backing store while mappings exists.

include/linux/dmaengine.h+6−6

@@ -83,9 +83,9 @@ enum dma_transfer_direction { · 82 unmodified lines
83	83	/**
84	84	* Interleaved Transfer Request
85	85	* ----------------------------
86		− * A chunk is collection of contiguous bytes to be transfered.
	86	+ * A chunk is collection of contiguous bytes to be transferred.
87	87	* The gap(in bytes) between two chunks is called inter-chunk-gap(ICG).
88		− * ICGs may or maynot change between chunks.
	88	+ * ICGs may or may not change between chunks.
89	89	* A FRAME is the smallest series of contiguous {chunk,icg} pairs,
90	90	* that when repeated an integral number of times, specifies the transfer.
91	91	* A transfer template is specification of a Frame, the number of times
@@ -341,13 +341,11 @@ struct dma_chan { · 249 unmodified lines
341	341	* @chan: driver channel device
342	342	* @device: sysfs device
343	343	* @dev_id: parent dma_device dev_id
344		− * @idr_ref: reference count to gate release of dma_device dev_id
345	344	*/
346	345	struct dma_chan_dev {
347	346	struct dma_chan *chan;
348	347	struct device device;
349	348	int dev_id;
350		− atomic_t *idr_ref;
351	349	};
352	350
353	351	/**
@@ -835,6 +833,8 @@ struct dma_device { · 481 unmodified lines
835	833	int dev_id;
836	834	struct device *dev;
837	835	struct module *owner;
	836	+ struct ida chan_ida;
	837	+ struct mutex chan_mutex; /* to protect chan_ida */
838	838
839	839	u32 src_addr_widths;
840	840	u32 dst_addr_widths;
@@ -1069,7 +1069,7 @@ static inline int dmaengine_terminate_all(struct dma_chan *chan) · 228 unmodified lines
1069	1069	* dmaengine_synchronize() needs to be called before it is safe to free
1070	1070	* any memory that is accessed by previously submitted descriptors or before
1071	1071	* freeing any resources accessed from within the completion callback of any
1072		− * perviously submitted descriptors.
	1072	+ * previously submitted descriptors.
1073	1073	*
1074	1074	* This function can be called from atomic context as well as from within a
1075	1075	* complete callback of a descriptor submitted on the same channel.
@@ -1091,7 +1091,7 @@ static inline int dmaengine_terminate_async(struct dma_chan *chan) · 15 unmodified lines
1091	1091	*
1092	1092	* Synchronizes to the DMA channel termination to the current context. When this
1093	1093	* function returns it is guaranteed that all transfers for previously issued
1094		− * descriptors have stopped and and it is safe to free the memory assoicated
	1094	+ * descriptors have stopped and it is safe to free the memory associated
1095	1095	* with them. Furthermore it is guaranteed that all complete callback functions
1096	1096	* for a previously submitted descriptor have finished running and it is safe to
1097	1097	* free resources accessed from within the complete callbacks.

include/linux/fs.h+1−1

@@ -983,7 +983,7 @@ struct file_handle { · 982 unmodified lines
983	983	__u32 handle_bytes;
984	984	int handle_type;
985	985	/* file identifier */
986		− unsigned char f_handle[0];
	986	+ unsigned char f_handle[];
987	987	};
988	988
989	989	static inline struct file get_file(struct file f)

include/linux/iio/iio.h+1−1

@@ -600,7 +600,7 @@ void iio_device_unregister(struct iio_dev *indio_dev); · 599 unmodified lines
600	600	* 0 on success, negative error number on failure.
601	601	*/
602	602	#define devm_iio_device_register(dev, indio_dev) \
603		− __devm_iio_device_register((dev), (indio_dev), THIS_MODULE);
	603	+ __devm_iio_device_register((dev), (indio_dev), THIS_MODULE)
604	604	int __devm_iio_device_register(struct device dev, struct iio_dev indio_dev,
605	605	struct module *this_mod);
606	606	void devm_iio_device_unregister(struct device dev, struct iio_dev indio_dev);

include/linux/kvm_host.h+1−1

@@ -1048,7 +1048,7 @@ search_memslots(struct kvm_memslots *slots, gfn_t gfn) · 1047 unmodified lines
1048	1048	start = slot + 1;
1049	1049	}
1050	1050
1051		− if (gfn >= memslots[start].base_gfn &&
	1051	+ if (start < slots->used_slots && gfn >= memslots[start].base_gfn &&
1052	1052	gfn < memslots[start].base_gfn + memslots[start].npages) {
1053	1053	atomic_set(&slots->lru_slot, start);
1054	1054	return &memslots[start];

include/linux/lsm_hook_defs.h+1−1

@@ -55,7 +55,7 @@ LSM_HOOK(void, LSM_RET_VOID, bprm_committing_creds, struct linux_binprm *bprm) · 54 unmodified lines
55	55	LSM_HOOK(void, LSM_RET_VOID, bprm_committed_creds, struct linux_binprm *bprm)
56	56	LSM_HOOK(int, 0, fs_context_dup, struct fs_context *fc,
57	57	struct fs_context *src_sc)
58		−LSM_HOOK(int, 0, fs_context_parse_param, struct fs_context *fc,
	58	+LSM_HOOK(int, -ENOPARAM, fs_context_parse_param, struct fs_context *fc,
59	59	struct fs_parameter *param)
60	60	LSM_HOOK(int, 0, sb_alloc_security, struct super_block *sb)
61	61	LSM_HOOK(void, LSM_RET_VOID, sb_free_security, struct super_block *sb)

include/linux/mhi.h+10−6

@@ -53,9 +53,9 @@ enum mhi_callback { · 52 unmodified lines
53	53	* @MHI_CHAIN: Linked transfer
54	54	*/
55	55	enum mhi_flags {
56		− MHI_EOB,
57		− MHI_EOT,
58		− MHI_CHAIN,
	56	+ MHI_EOB = BIT(0),
	57	+ MHI_EOT = BIT(1),
	58	+ MHI_CHAIN = BIT(2),
59	59	};
60	60
61	61	/**
@@ -335,14 +335,15 @@ struct mhi_controller_config { · 273 unmodified lines
335	335	* @syserr_worker: System error worker
336	336	* @state_event: State change event
337	337	* @status_cb: CB function to notify power states of the device (required)
338		− * @link_status: CB function to query link status of the device (required)
339	338	* @wake_get: CB function to assert device wake (optional)
340	339	* @wake_put: CB function to de-assert device wake (optional)
341	340	* @wake_toggle: CB function to assert and de-assert device wake (optional)
342	341	* @runtime_get: CB function to controller runtime resume (required)
343		− * @runtimet_put: CB function to decrement pm usage (required)
	342	+ * @runtime_put: CB function to decrement pm usage (required)
344	343	* @map_single: CB function to create TRE buffer
345	344	* @unmap_single: CB function to destroy TRE buffer
	345	+ * @read_reg: Read a MHI register via the physical link (required)
	346	+ * @write_reg: Write a MHI register via the physical link (required)
346	347	* @buffer_len: Bounce buffer length
347	348	* @bounce_buf: Use of bounce buffer
348	349	* @fbc_download: MHI host needs to do complete image transfer (optional)
@@ -417,7 +418,6 @@ struct mhi_controller { · 68 unmodified lines
417	418
418	419	void (status_cb)(struct mhi_controller mhi_cntrl,
419	420	enum mhi_callback cb);
420		− int (link_status)(struct mhi_controller mhi_cntrl);
421	421	void (wake_get)(struct mhi_controller mhi_cntrl, bool override);
422	422	void (wake_put)(struct mhi_controller mhi_cntrl, bool override);
423	423	void (wake_toggle)(struct mhi_controller mhi_cntrl);
@@ -427,6 +427,10 @@ struct mhi_controller { · 3 unmodified lines
427	427	struct mhi_buf_info *buf);
428	428	void (unmap_single)(struct mhi_controller mhi_cntrl,
429	429	struct mhi_buf_info *buf);
	430	+ int (read_reg)(struct mhi_controller mhi_cntrl, void __iomem *addr,
	431	+ u32 *out);
	432	+ void (write_reg)(struct mhi_controller mhi_cntrl, void __iomem *addr,
	433	+ u32 val);
430	434
431	435	size_t buffer_len;
432	436	bool bounce_buf;

include/linux/nfs_xdr.h+2−0

@@ -1317,11 +1317,13 @@ struct nfs41_impl_id { · 1316 unmodified lines
1317	1317	struct nfstime4 date;
1318	1318	};
1319	1319
	1320	+#define MAX_BIND_CONN_TO_SESSION_RETRIES 3
1320	1321	struct nfs41_bind_conn_to_session_args {
1321	1322	struct nfs_client *client;
1322	1323	struct nfs4_sessionid sessionid;
1323	1324	u32 dir;
1324	1325	bool use_conn_in_rdma_mode;
	1326	+ int retries;
1325	1327	};
1326	1328
1327	1329	struct nfs41_bind_conn_to_session_res {

include/linux/platform_data/cros_ec_sensorhub.h+1−0

@@ -185,6 +185,7 @@ int cros_ec_sensorhub_register_push_data(struct cros_ec_sensorhub *sensorhub, · 184 unmodified lines
185	185	void cros_ec_sensorhub_unregister_push_data(struct cros_ec_sensorhub *sensorhub,
186	186	u8 sensor_num);
187	187
	188	+int cros_ec_sensorhub_ring_allocate(struct cros_ec_sensorhub *sensorhub);
188	189	int cros_ec_sensorhub_ring_add(struct cros_ec_sensorhub *sensorhub);
189	190	void cros_ec_sensorhub_ring_remove(void *arg);
190	191	int cros_ec_sensorhub_ring_fifo_enable(struct cros_ec_sensorhub *sensorhub,

include/linux/platform_device.h+1−0

@@ -25,6 +25,7 @@ struct platform_device { · 24 unmodified lines
25	25	bool id_auto;
26	26	struct device dev;
27	27	u64 platform_dma_mask;
	28	+ struct device_dma_parameters dma_parms;
28	29	u32 num_resources;
29	30	struct resource *resource;
30	31

include/linux/pnp.h+9−20

@@ -220,10 +220,8 @@ struct pnp_card { · 219 unmodified lines
220	220	#define global_to_pnp_card(n) list_entry(n, struct pnp_card, global_list)
221	221	#define protocol_to_pnp_card(n) list_entry(n, struct pnp_card, protocol_list)
222	222	#define to_pnp_card(n) container_of(n, struct pnp_card, dev)
223		−#define pnp_for_each_card(card) \
224		− for((card) = global_to_pnp_card(pnp_cards.next); \
225		− (card) != global_to_pnp_card(&pnp_cards); \
226		− (card) = global_to_pnp_card((card)->global_list.next))
	223	+#define pnp_for_each_card(card) \
	224	+ list_for_each_entry(card, &pnp_cards, global_list)
227	225
228	226	struct pnp_card_link {
229	227	struct pnp_card *card;
@@ -276,14 +274,9 @@ struct pnp_dev { · 46 unmodified lines
276	274	#define card_to_pnp_dev(n) list_entry(n, struct pnp_dev, card_list)
277	275	#define protocol_to_pnp_dev(n) list_entry(n, struct pnp_dev, protocol_list)
278	276	#define to_pnp_dev(n) container_of(n, struct pnp_dev, dev)
279		−#define pnp_for_each_dev(dev) \
280		− for((dev) = global_to_pnp_dev(pnp_global.next); \
281		− (dev) != global_to_pnp_dev(&pnp_global); \
282		− (dev) = global_to_pnp_dev((dev)->global_list.next))
283		−#define card_for_each_dev(card,dev) \
284		− for((dev) = card_to_pnp_dev((card)->devices.next); \
285		− (dev) != card_to_pnp_dev(&(card)->devices); \
286		− (dev) = card_to_pnp_dev((dev)->card_list.next))
	277	+#define pnp_for_each_dev(dev) list_for_each_entry(dev, &pnp_global, global_list)
	278	+#define card_for_each_dev(card, dev) \
	279	+ list_for_each_entry(dev, &(card)->devices, card_list)
287	280	#define pnp_dev_name(dev) (dev)->name
288	281
289	282	static inline void pnp_get_drvdata(struct pnp_dev pdev)
@@ -437,14 +430,10 @@ struct pnp_protocol { · 147 unmodified lines
437	430	};
438	431
439	432	#define to_pnp_protocol(n) list_entry(n, struct pnp_protocol, protocol_list)
440		−#define protocol_for_each_card(protocol,card) \
441		− for((card) = protocol_to_pnp_card((protocol)->cards.next); \
442		− (card) != protocol_to_pnp_card(&(protocol)->cards); \
443		− (card) = protocol_to_pnp_card((card)->protocol_list.next))
444		−#define protocol_for_each_dev(protocol,dev) \
445		− for((dev) = protocol_to_pnp_dev((protocol)->devices.next); \
446		− (dev) != protocol_to_pnp_dev(&(protocol)->devices); \
447		− (dev) = protocol_to_pnp_dev((dev)->protocol_list.next))
	433	+#define protocol_for_each_card(protocol, card) \
	434	+ list_for_each_entry(card, &(protocol)->cards, protocol_list)
	435	+#define protocol_for_each_dev(protocol, dev) \
	436	+ list_for_each_entry(dev, &(protocol)->devices, protocol_list)
448	437
449	438	extern struct bus_type pnp_bus_type;
450	439

include/linux/stmmac.h+2−0

@@ -177,6 +177,8 @@ struct plat_stmmacenet_data { · 176 unmodified lines
177	177	struct stmmac_rxq_cfg rx_queues_cfg[MTL_MAX_RX_QUEUES];
178	178	struct stmmac_txq_cfg tx_queues_cfg[MTL_MAX_TX_QUEUES];
179	179	void (fix_mac_speed)(void priv, unsigned int speed);
	180	+ int (serdes_powerup)(struct net_device ndev, void *priv);
	181	+ void (serdes_powerdown)(struct net_device ndev, void *priv);
180	182	int (init)(struct platform_device pdev, void *priv);
181	183	void (exit)(struct platform_device pdev, void *priv);
182	184	struct mac_device_info (setup)(void *priv);

include/linux/sunrpc/clnt.h+12−1

@@ -71,7 +71,13 @@ struct rpc_clnt { · 70 unmodified lines
71	71	#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
72	72	struct dentry cl_debugfs; / debugfs directory */
73	73	#endif
74		− struct rpc_xprt_iter cl_xpi;
	74	+ /* cl_work is only needed after cl_xpi is no longer used,
	75	+ * and that are of similar size
	76	+ */
	77	+ union {
	78	+ struct rpc_xprt_iter cl_xpi;
	79	+ struct work_struct cl_work;
	80	+ };
75	81	const struct cred *cl_cred;
76	82	};
77	83
@@ -236,4 +242,9 @@ static inline int rpc_reply_expected(struct rpc_task *task) · 158 unmodified lines
236	242	(task->tk_msg.rpc_proc->p_decode != NULL);
237	243	}
238	244
	245	+static inline void rpc_task_close_connection(struct rpc_task *task)
	246	+{
	247	+ if (task->tk_xprt)
	248	+ xprt_force_disconnect(task->tk_xprt);
	249	+}
239	250	#endif /* _LINUX_SUNRPC_CLNT_H */

include/linux/sunrpc/svc_rdma.h+1−0

@@ -170,6 +170,7 @@ extern bool svc_rdma_post_recvs(struct svcxprt_rdma *rdma); · 169 unmodified lines
170	170	extern void svc_rdma_recv_ctxt_put(struct svcxprt_rdma *rdma,
171	171	struct svc_rdma_recv_ctxt *ctxt);
172	172	extern void svc_rdma_flush_recv_queues(struct svcxprt_rdma *rdma);
	173	+extern void svc_rdma_release_rqst(struct svc_rqst *rqstp);
173	174	extern int svc_rdma_recvfrom(struct svc_rqst *);
174	175
175	176	/* svc_rdma_rw.c */

include/linux/tcp.h+0−51

@@ -78,47 +78,6 @@ struct tcp_sack_block { · 77 unmodified lines
78	78	#define TCP_SACK_SEEN (1 << 0) /1 = peer is SACK capable, /
79	79	#define TCP_DSACK_SEEN (1 << 2) /1 = DSACK was received from peer/
80	80
81		−#if IS_ENABLED(CONFIG_MPTCP)
82		−struct mptcp_options_received {
83		− u64 sndr_key;
84		− u64 rcvr_key;
85		− u64 data_ack;
86		− u64 data_seq;
87		− u32 subflow_seq;
88		− u16 data_len;
89		− u16 mp_capable : 1,
90		− mp_join : 1,
91		− dss : 1,
92		− add_addr : 1,
93		− rm_addr : 1,
94		− family : 4,
95		− echo : 1,
96		− backup : 1;
97		− u32 token;
98		− u32 nonce;
99		− u64 thmac;
100		− u8 hmac[20];
101		− u8 join_id;
102		− u8 use_map:1,
103		− dsn64:1,
104		− data_fin:1,
105		− use_ack:1,
106		− ack64:1,
107		− mpc_map:1,
108		− __unused:2;
109		− u8 addr_id;
110		− u8 rm_id;
111		− union {
112		− struct in_addr addr;
113		−#if IS_ENABLED(CONFIG_MPTCP_IPV6)
114		− struct in6_addr addr6;
115		−#endif
116		− };
117		− u64 ahmac;
118		− u16 port;
119		−};
120		−#endif
121		−
122	81	struct tcp_options_received {
123	82	/* PAWS/RTTM data */
124	83	int ts_recent_stamp;/* Time we stored ts_recent (for aging) */
@@ -136,9 +95,6 @@ struct tcp_options_received { · 11 unmodified lines
136	95	u8 num_sacks; /* Number of SACK blocks */
137	96	u16 user_mss; /* mss requested by user in ioctl */
138	97	u16 mss_clamp; /* Maximal mss, negotiated at connection setup */
139		−#if IS_ENABLED(CONFIG_MPTCP)
140		− struct mptcp_options_received mptcp;
141		−#endif
142	98	};
143	99
144	100	static inline void tcp_clear_options(struct tcp_options_received *rx_opt)
@@ -148,13 +104,6 @@ static inline void tcp_clear_options(struct tcp_options_received *rx_opt) · 3 unmodified lines
148	104	#if IS_ENABLED(CONFIG_SMC)
149	105	rx_opt->smc_ok = 0;
150	106	#endif
151		−#if IS_ENABLED(CONFIG_MPTCP)
152		− rx_opt->mptcp.mp_capable = 0;
153		− rx_opt->mptcp.mp_join = 0;
154		− rx_opt->mptcp.add_addr = 0;
155		− rx_opt->mptcp.rm_addr = 0;
156		− rx_opt->mptcp.dss = 0;
157		−#endif
158	107	}
159	108
160	109	/* This is the max number of SACKS that we'll generate and process. It's safe

include/linux/tracepoint.h+1−21

@@ -156,8 +156,7 @@ static inline struct tracepoint tracepoint_ptr_deref(tracepoint_ptr_t p) · 155 unmodified lines
156	156	* Note, the proto and args passed in includes "__data" as the first parameter.
157	157	* The reason for this is to handle the "void" prototype. If a tracepoint
158	158	* has a "void" prototype, then it is invalid to declare a function
159		− * as "(void *, void)". The DECLARE_TRACE_NOARGS() will pass in just
160		− * "void data", where as the DECLARE_TRACE() will pass in "void data, proto".
	159	+ * as "(void *, void)".
161	160	*/
162	161	#define __DO_TRACE(tp, proto, args, cond, rcuidle) \
163	162	do { \
@@ -373,25 +372,6 @@ static inline struct tracepoint tracepoint_ptr_deref(tracepoint_ptr_t p) · 209 unmodified lines
373	372	# define __tracepoint_string
374	373	#endif
375	374
376		−/*
377		− * The need for the DECLARE_TRACE_NOARGS() is to handle the prototype
378		− * (void). "void" is a special value in a function prototype and can
379		− * not be combined with other arguments. Since the DECLARE_TRACE()
380		− * macro adds a data element at the beginning of the prototype,
381		− * we need a way to differentiate "(void *data, proto)" from
382		− * "(void *data, void)". The second prototype is invalid.
383		− *
384		− * DECLARE_TRACE_NOARGS() passes "void" as the tracepoint prototype
385		− * and "void *__data" as the callback prototype.
386		− *
387		− * DECLARE_TRACE() passes "proto" as the tracepoint protoype and
388		− * "void *__data, proto" as the callback prototype.
389		− */
390		−#define DECLARE_TRACE_NOARGS(name) \
391		− __DECLARE_TRACE(name, void, , \
392		− cpu_online(raw_smp_processor_id()), \

5 more lines aren’t shown here.

include/linux/tty.h+1−1

This diff is large: its 9 lines aren’t shown here.

include/linux/vdpa.h+1−1

This diff is large: its 9 lines aren’t shown here.

include/linux/vermagic.h+5−3

This diff is large: its 25 lines aren’t shown here.

include/linux/virtio.h+0−1

This diff is large: its 8 lines aren’t shown here.

include/linux/virtio_net.h+24−2

This diff is large: its 73 lines aren’t shown here.

include/linux/virtio_vsock.h+1−0

This diff is large: its 8 lines aren’t shown here.

include/linux/vmalloc.h+1−1

This diff is large: its 9 lines aren’t shown here.

include/linux/vringh.h+6−0

This diff is large: its 27 lines aren’t shown here.

include/net/flow_offload.h+8−1

This diff is large: its 32 lines aren’t shown here.

include/net/inet_ecn.h+55−2

This diff is large: its 85 lines aren’t shown here.

include/net/ip6_fib.h+4−0

This diff is large: its 18 lines aren’t shown here.

include/net/mac80211.h+3−1

This diff is large: its 11 lines aren’t shown here.

include/net/mptcp.h+0−3

This diff is large: its 12 lines aren’t shown here.

include/net/ndisc.h+1−1

This diff is large: its 9 lines aren’t shown here.

include/net/net_namespace.h+7−0

This diff is large: its 14 lines aren’t shown here.

include/net/sch_generic.h+1−0

This diff is large: its 8 lines aren’t shown here.

include/net/tcp.h+1−1

This diff is large: its 9 lines aren’t shown here.

include/soc/mscc/ocelot.h+1−0

This diff is large: its 8 lines aren’t shown here.

include/soc/tegra/pmc.h+9−6

This diff is large: its 34 lines aren’t shown here.

include/sound/soc-dai.h+0−1

This diff is large: its 8 lines aren’t shown here.

include/sound/soc.h+3−0

This diff is large: its 10 lines aren’t shown here.

include/trace/events/gpu_mem.h+1−1

This diff is large: its 9 lines aren’t shown here.

include/trace/events/iocost.h+3−3

This diff is large: its 20 lines aren’t shown here.

include/trace/events/rpcrdma.h+40−22

This diff is large: its 141 lines aren’t shown here.

include/trace/events/wbt.h+4−4

This diff is large: its 36 lines aren’t shown here.

include/trace/events/writeback.h+0−1

This diff is large: its 8 lines aren’t shown here.

include/uapi/drm/amdgpu_drm.h+4−0

This diff is large: its 11 lines aren’t shown here.

include/uapi/linux/bpf.h+2−2

This diff is large: its 18 lines aren’t shown here.

include/uapi/linux/dlm_device.h+2−2

This diff is large: its 16 lines aren’t shown here.

include/uapi/linux/dma-buf.h+6−0

This diff is large: its 13 lines aren’t shown here.

include/uapi/linux/fiemap.h+1−1

This diff is large: its 9 lines aren’t shown here.

include/uapi/linux/hyperv.h+2−2

This diff is large: its 11 lines aren’t shown here.

include/uapi/linux/if_arcnet.h+3−3

This diff is large: its 27 lines aren’t shown here.

include/uapi/linux/mmc/ioctl.h+1−1

This diff is large: its 9 lines aren’t shown here.

include/uapi/linux/net_dropmon.h+2−2

This diff is large: its 15 lines aren’t shown here.

include/uapi/linux/netfilter_bridge/ebt_among.h+1−1

This diff is large: its 9 lines aren’t shown here.

include/uapi/linux/virtio_balloon.h+9−2

This diff is large: its 18 lines aren’t shown here.

include/uapi/scsi/scsi_bsg_fc.h+1−1

This diff is large: its 9 lines aren’t shown here.

include/vdso/datapage.h+2−2

This diff is large: its 11 lines aren’t shown here.

init/Kconfig+0−18

This diff is large: its 38 lines aren’t shown here.

init/initramfs.c+1−1

This diff is large: its 9 lines aren’t shown here.

init/main.c+52−17

This diff is large: its 106 lines aren’t shown here.

ipc/mqueue.c+26−8

This diff is large: its 69 lines aren’t shown here.

kernel/audit.c+3−0

This diff is large: its 10 lines aren’t shown here.

kernel/bpf/cpumap.c+1−1

This diff is large: its 9 lines aren’t shown here.

kernel/bpf/syscall.c+10−3

This diff is large: its 41 lines aren’t shown here.

kernel/bpf/verifier.c+28−10

This diff is large: its 80 lines aren’t shown here.

kernel/events/core.c+10−3

This diff is large: its 21 lines aren’t shown here.

kernel/exit.c+1−0

This diff is large: its 8 lines aren’t shown here.

kernel/kcov.c+2−2

This diff is large: its 11 lines aren’t shown here.

kernel/power/hibernate.c+7−0

This diff is large: its 14 lines aren’t shown here.

kernel/sched/core.c+2−7

This diff is large: its 16 lines aren’t shown here.

kernel/signal.c+5−1

This diff is large: its 14 lines aren’t shown here.

kernel/trace/Kconfig+0−1

This diff is large: its 8 lines aren’t shown here.

kernel/trace/ftrace.c+1−0

This diff is large: its 8 lines aren’t shown here.

kernel/trace/preemptirq_delay_test.c+24−6

This diff is large: its 59 lines aren’t shown here.

kernel/trace/trace.c+15−1

This diff is large: its 30 lines aren’t shown here.

kernel/trace/trace_boot.c+8−12

This diff is large: its 33 lines aren’t shown here.

kernel/trace/trace_events_hist.c+7−0

This diff is large: its 28 lines aren’t shown here.

kernel/trace/trace_kprobe.c+7−1

This diff is large: its 29 lines aren’t shown here.

kernel/trace/tracing_map.c+3−3

This diff is large: its 27 lines aren’t shown here.

kernel/umh.c+5−0

This diff is large: its 12 lines aren’t shown here.

lib/Kconfig.ubsan+6−9

This diff is large: its 25 lines aren’t shown here.

lib/kunit/test.c+1−1

This diff is large: its 9 lines aren’t shown here.

lib/mpi/longlong.h+17−17

This diff is large: its 85 lines aren’t shown here.

mm/backing-dev.c+11−2

This diff is large: its 34 lines aren’t shown here.

mm/gup.c+1−1

This diff is large: its 9 lines aren’t shown here.

mm/hugetlb.c+8−6

This diff is large: its 35 lines aren’t shown here.

mm/ksm.c+10−2

This diff is large: its 19 lines aren’t shown here.

mm/madvise.c+18−0

This diff is large: its 32 lines aren’t shown here.

mm/memcontrol.c+9−6

This diff is large: its 54 lines aren’t shown here.

mm/page_alloc.c+9−0

This diff is large: its 23 lines aren’t shown here.

mm/percpu.c+10−4

This diff is large: its 35 lines aren’t shown here.

mm/shmem.c+8−5

This diff is large: its 44 lines aren’t shown here.

mm/slub.c+40−17

This diff is large: its 126 lines aren’t shown here.

mm/vmalloc.c+13−3

This diff is large: its 54 lines aren’t shown here.

mm/vmscan.c+0−1

This diff is large: its 8 lines aren’t shown here.

net/atm/common.c+11−11

This diff is large: its 30 lines aren’t shown here.

net/atm/lec.c+6−0

This diff is large: its 13 lines aren’t shown here.

net/batman-adv/bat_v_ogm.c+1−1

This diff is large: its 9 lines aren’t shown here.

net/batman-adv/network-coding.c+1−8

This diff is large: its 17 lines aren’t shown here.

net/batman-adv/sysfs.c+2−1

This diff is large: its 17 lines aren’t shown here.

net/bridge/br_netlink.c+1−0

This diff is large: its 8 lines aren’t shown here.

net/core/devlink.c+10−2

This diff is large: its 35 lines aren’t shown here.

net/core/drop_monitor.c+7−4

This diff is large: its 35 lines aren’t shown here.

net/core/neighbour.c+3−3

This diff is large: its 20 lines aren’t shown here.

net/core/netclassid_cgroup.c+1−3

This diff is large: its 12 lines aren’t shown here.

net/core/sock.c+0−1

This diff is large: its 8 lines aren’t shown here.

net/dsa/dsa2.c+1−1

This diff is large: its 9 lines aren’t shown here.

net/dsa/master.c+2−1

This diff is large: its 10 lines aren’t shown here.

net/dsa/slave.c+6−10

This diff is large: its 39 lines aren’t shown here.

net/hsr/hsr_slave.c+1−1

This diff is large: its 9 lines aren’t shown here.

net/ipv4/fib_semantics.c+3−3

This diff is large: its 20 lines aren’t shown here.

net/ipv4/tcp_input.c+0−7

This diff is large: its 21 lines aren’t shown here.

net/ipv4/xfrm4_output.c+0−2

This diff is large: its 10 lines aren’t shown here.

net/ipv6/ipv6_sockglue.c+6−7

This diff is large: its 22 lines aren’t shown here.

net/ipv6/route.c+25−0

This diff is large: its 49 lines aren’t shown here.

net/ipv6/rpl.c+4−3

This diff is large: its 30 lines aren’t shown here.

net/ipv6/seg6.c+8−2

This diff is large: its 25 lines aren’t shown here.

net/ipv6/xfrm6_output.c+0−2

This diff is large: its 10 lines aren’t shown here.

net/mac80211/main.c+3−2

This diff is large: its 19 lines aren’t shown here.

net/mac80211/rate.c+4−11

This diff is large: its 36 lines aren’t shown here.

net/mac80211/rate.h+23−0

This diff is large: its 30 lines aren’t shown here.

net/mac80211/rc80211_minstrel_ht.c+13−6

This diff is large: its 49 lines aren’t shown here.

net/mac80211/sta_info.c+2−1

This diff is large: its 10 lines aren’t shown here.

net/mptcp/options.c+41−55

This diff is large: its 194 lines aren’t shown here.

net/mptcp/pm_netlink.c+7−5

This diff is large: its 20 lines aren’t shown here.

net/mptcp/protocol.c+20−10

This diff is large: its 81 lines aren’t shown here.

net/mptcp/protocol.h+44−5

This diff is large: its 74 lines aren’t shown here.

net/mptcp/subflow.c+119−52

This diff is large: its 322 lines aren’t shown here.

net/netfilter/nf_flow_table_core.c+4−2

This diff is large: its 15 lines aren’t shown here.

net/netfilter/nf_nat_proto.c+3−5

This diff is large: its 28 lines aren’t shown here.

net/netfilter/nfnetlink_osf.c+7−5

This diff is large: its 42 lines aren’t shown here.

net/netlabel/Kconfig+1−1

This diff is large: its 8 lines aren’t shown here.

net/netrom/nr_route.c+1−0

This diff is large: its 8 lines aren’t shown here.

net/openvswitch/conntrack.c+2−1

This diff is large: its 10 lines aren’t shown here.

net/openvswitch/datapath.c+3−1

This diff is large: its 12 lines aren’t shown here.

net/sched/cls_api.c+16−6

This diff is large: its 66 lines aren’t shown here.

net/sched/sch_choke.c+2−1

This diff is large: its 10 lines aren’t shown here.

net/sched/sch_etf.c+4−3

This diff is large: its 29 lines aren’t shown here.

net/sched/sch_fq_codel.c+1−1

This diff is large: its 9 lines aren’t shown here.

net/sched/sch_sfq.c+9−0

This diff is large: its 16 lines aren’t shown here.

net/sched/sch_skbprio.c+3−0

This diff is large: its 10 lines aren’t shown here.

net/sctp/sm_make_chunk.c+5−1

This diff is large: its 13 lines aren’t shown here.

net/sctp/sm_statefuns.c+3−3

This diff is large: its 27 lines aren’t shown here.

net/sunrpc/cache.c+3−2

This diff is large: its 19 lines aren’t shown here.

net/sunrpc/clnt.c+18−6

This diff is large: its 52 lines aren’t shown here.

net/sunrpc/svc_xprt.c+2−3

This diff is large: its 19 lines aren’t shown here.

net/sunrpc/svcsock.c+4−0

This diff is large: its 18 lines aren’t shown here.

net/sunrpc/xprtrdma/rpc_rdma.c+11−4

This diff is large: its 36 lines aren’t shown here.

net/sunrpc/xprtrdma/svc_rdma_backchannel.c+2−0

This diff is large: its 9 lines aren’t shown here.

net/sunrpc/xprtrdma/svc_rdma_recvfrom.c+22−0

This diff is large: its 36 lines aren’t shown here.

net/sunrpc/xprtrdma/svc_rdma_rw.c+1−2

This diff is large: its 17 lines aren’t shown here.

net/sunrpc/xprtrdma/svc_rdma_sendto.c+12−17

This diff is large: its 58 lines aren’t shown here.

net/sunrpc/xprtrdma/svc_rdma_transport.c+0−5

This diff is large: its 19 lines aren’t shown here.

net/sunrpc/xprtrdma/verbs.c+2−1

This diff is large: its 18 lines aren’t shown here.

net/sunrpc/xprtsock.c+1−0

This diff is large: its 8 lines aren’t shown here.

net/tipc/crypto.c+1−0

This diff is large: its 8 lines aren’t shown here.

net/tipc/node.c+3−1

This diff is large: its 25 lines aren’t shown here.

net/tipc/topsrv.c+3−2

This diff is large: its 14 lines aren’t shown here.

net/tls/tls_sw.c+5−2

This diff is large: its 21 lines aren’t shown here.

net/vmw_vsock/virtio_transport_common.c+4−0

This diff is large: its 12 lines aren’t shown here.

net/x25/x25_dev.c+3−1

This diff is large: its 12 lines aren’t shown here.

net/x25/x25_subr.c+6−0

This diff is large: its 13 lines aren’t shown here.

samples/trace_events/trace-events-sample.h+1−1

This diff is large: its 9 lines aren’t shown here.

samples/vfio-mdev/mdpy.c+1−1

This diff is large: its 9 lines aren’t shown here.

scripts/Makefile.lib+1−1

This diff is large: its 9 lines aren’t shown here.

scripts/checkpatch.pl+1−1

This diff is large: its 9 lines aren’t shown here.

scripts/config+4−1

This diff is large: its 19 lines aren’t shown here.

scripts/decodecode+1−1

This diff is large: its 9 lines aren’t shown here.

scripts/gcc-plugins/Makefile+1−0

This diff is large: its 8 lines aren’t shown here.

scripts/gcc-plugins/gcc-common.h+4−0

This diff is large: its 26 lines aren’t shown here.

scripts/gcc-plugins/stackleak_plugin.c+2−3

This diff is large: its 20 lines aren’t shown here.

scripts/gdb/linux/rbtree.py+2−2

This diff is large: its 18 lines aren’t shown here.

scripts/kallsyms.c+1−1

This diff is large: its 9 lines aren’t shown here.

scripts/mod/modpost.c+1−1

This diff is large: its 9 lines aren’t shown here.

security/selinux/hooks.c+45−25

This diff is large: its 86 lines aren’t shown here.

security/selinux/ss/conditional.c+1−1

This diff is large: its 9 lines aren’t shown here.

sound/core/oss/pcm_plugin.c+6−4

This diff is large: its 35 lines aren’t shown here.

sound/isa/opti9xx/miro.c+6−3

This diff is large: its 17 lines aren’t shown here.

sound/isa/opti9xx/opti92x-ad1848.c+6−3

This diff is large: its 17 lines aren’t shown here.

sound/pci/hda/hda_codec.c+17−11

This diff is large: its 45 lines aren’t shown here.

sound/pci/hda/hda_intel.c+7−20

This diff is large: its 64 lines aren’t shown here.

sound/pci/hda/patch_hdmi.c+14−1

This diff is large: its 45 lines aren’t shown here.

sound/pci/hda/patch_realtek.c+9−3

This diff is large: its 47 lines aren’t shown here.

sound/soc/amd/acp3x-rt5682-max9836.c+3−3

This diff is large: its 13 lines aren’t shown here.

sound/soc/codecs/Kconfig+3−0

This diff is large: its 24 lines aren’t shown here.

sound/soc/codecs/hdac_hdmi.c+3−3

This diff is large: its 18 lines aren’t shown here.

sound/soc/codecs/madera.c+0−4

This diff is large: its 32 lines aren’t shown here.

sound/soc/codecs/sgtl5000.c+34−0

This diff is large: its 41 lines aren’t shown here.

sound/soc/codecs/sgtl5000.h+1−0

This diff is large: its 8 lines aren’t shown here.

sound/soc/codecs/tas571x.c+15−5

This diff is large: its 45 lines aren’t shown here.

sound/soc/codecs/wm8960.c+1−2

This diff is large: its 10 lines aren’t shown here.

sound/soc/codecs/wsa881x.c+4−0

This diff is large: its 32 lines aren’t shown here.

sound/soc/intel/common/soc-acpi-intel-cml-match.c+0−8

This diff is large: its 15 lines aren’t shown here.

sound/soc/intel/common/soc-acpi-intel-icl-match.c+0−8

This diff is large: its 15 lines aren’t shown here.

sound/soc/meson/axg-card.c+3−1

This diff is large: its 12 lines aren’t shown here.

sound/soc/meson/gx-card.c+3−1

This diff is large: its 12 lines aren’t shown here.

sound/soc/qcom/apq8096.c+1−3

This diff is large: its 12 lines aren’t shown here.

sound/soc/qcom/qdsp6/q6afe-dai.c+16−0

This diff is large: its 72 lines aren’t shown here.

sound/soc/qcom/sdm845.c+1−3

This diff is large: its 12 lines aren’t shown here.

sound/soc/samsung/s3c-i2s-v2.c+0−57

This diff is large: its 71 lines aren’t shown here.

sound/soc/samsung/s3c2412-i2s.c+56−0

This diff is large: its 70 lines aren’t shown here.

sound/soc/sh/rcar/ssi.c+10−1

This diff is large: its 28 lines aren’t shown here.

sound/soc/sh/rcar/ssiu.c+1−1

This diff is large: its 9 lines aren’t shown here.

sound/soc/soc-dai.c+2−9

This diff is large: its 27 lines aren’t shown here.

sound/soc/soc-dapm.c+75−72

This diff is large: its 218 lines aren’t shown here.

sound/soc/soc-pcm.c+11−2

This diff is large: its 20 lines aren’t shown here.

sound/soc/soc-topology.c+89−26

This diff is large: its 273 lines aren’t shown here.

sound/soc/sof/intel/bdw.c+16−0

This diff is large: its 26 lines aren’t shown here.

sound/soc/sof/intel/byt.c+48−0

This diff is large: its 70 lines aren’t shown here.

sound/soc/stm/stm32_sai_sub.c+6−8

This diff is large: its 37 lines aren’t shown here.

sound/usb/format.c+51−0

This diff is large: its 65 lines aren’t shown here.

sound/usb/line6/podhd.c+5−17

This diff is large: its 47 lines aren’t shown here.

sound/usb/mixer.c+34−3

This diff is large: its 77 lines aren’t shown here.

sound/usb/mixer.h+10−0

This diff is large: its 24 lines aren’t shown here.

sound/usb/mixer_maps.c+34−3

This diff is large: its 68 lines aren’t shown here.

sound/usb/mixer_quirks.c+8−4

This diff is large: its 20 lines aren’t shown here.

sound/usb/quirks-table.h+14−84

This diff is large: its 110 lines aren’t shown here.

sound/usb/quirks.c+15−1

This diff is large: its 30 lines aren’t shown here.

sound/usb/usx2y/usbusx2yaudio.c+2−0

This diff is large: its 9 lines aren’t shown here.

tools/bootconfig/main.c+6−3

This diff is large: its 38 lines aren’t shown here.

tools/bpf/bpftool/struct_ops.c+7−1

This diff is large: its 22 lines aren’t shown here.

tools/bpf/runqslower/Makefile+1−1

This diff is large: its 9 lines aren’t shown here.

tools/build/feature/test-sync-compare-and-swap.c+1−1

This diff is large: its 9 lines aren’t shown here.

tools/cgroup/iocost_monitor.py+6−1

This diff is large: its 14 lines aren’t shown here.

tools/include/uapi/linux/bpf.h+1−1

This diff is large: its 9 lines aren’t shown here.

tools/lib/bpf/netlink.c+2−0

This diff is large: its 9 lines aren’t shown here.

tools/objtool/check.c+14−3

This diff is large: its 38 lines aren’t shown here.

tools/objtool/elf.c+1−1

This diff is large: its 9 lines aren’t shown here.

tools/objtool/elf.h+5−4

This diff is large: its 23 lines aren’t shown here.

tools/power/pm-graph/Makefile+4−0

This diff is large: its 11 lines aren’t shown here.

tools/power/pm-graph/README+114−10

This diff is large: its 163 lines aren’t shown here.

tools/power/pm-graph/bootgraph.py+1−1

This diff is large: its 6 lines aren’t shown here.

tools/power/pm-graph/sleepgraph.8+49−16

This diff is large: its 112 lines aren’t shown here.

tools/power/pm-graph/sleepgraph.py+451−344

This diff is large: its 1557 lines aren’t shown here.

tools/testing/selftests/bpf/prog_tests/fexit_bpf2bpf.c+24−6

This diff is large: its 84 lines aren’t shown here.

tools/testing/selftests/bpf/progs/connect4_prog.c+16−12

This diff is large: its 46 lines aren’t shown here.

tools/testing/selftests/bpf/progs/freplace_connect4.c+18−0

This diff is large: its 19 lines aren’t shown here.

tools/testing/selftests/bpf/progs/test_btf_haskv.c+5−13

This diff is large: its 42 lines aren’t shown here.

tools/testing/selftests/bpf/progs/test_btf_newkv.c+5−13

This diff is large: its 42 lines aren’t shown here.

tools/testing/selftests/bpf/progs/test_btf_nokv.c+5−13

This diff is large: its 42 lines aren’t shown here.

tools/testing/selftests/bpf/test_btf.c+1−1

This diff is large: its 9 lines aren’t shown here.

tools/testing/selftests/bpf/verifier/stack_ptr.c+40−0

This diff is large: its 44 lines aren’t shown here.

tools/testing/selftests/bpf/verifier/value_illegal_alu.c+1−0

This diff is large: its 8 lines aren’t shown here.

tools/testing/selftests/filesystems/epoll/epoll_wakeup_test.c+146−0

This diff is large: its 158 lines aren’t shown here.

tools/testing/selftests/ftrace/config+1−0

This diff is large: its 7 lines aren’t shown here.

tools/testing/selftests/ftrace/ftracetest+30−2

This diff is large: its 96 lines aren’t shown here.

tools/testing/selftests/ftrace/test.d/ftrace/fgraph-filter-stack.tc+1−4

This diff is large: its 12 lines aren’t shown here.

tools/testing/selftests/ftrace/test.d/ftrace/fgraph-filter.tc+2−0

This diff is large: its 9 lines aren’t shown here.

tools/testing/selftests/ftrace/test.d/ftrace/func-filter-glob.tc+2−0

This diff is large: its 9 lines aren’t shown here.

tools/testing/selftests/ftrace/test.d/ftrace/func-filter-notrace-pid.tc+1−4

This diff is large: its 12 lines aren’t shown here.

tools/testing/selftests/ftrace/test.d/ftrace/func-filter-pid.tc+1−4

This diff is large: its 12 lines aren’t shown here.

tools/testing/selftests/ftrace/test.d/ftrace/func-filter-stacktrace.tc+1−1

This diff is large: its 9 lines aren’t shown here.

tools/testing/selftests/ftrace/test.d/ftrace/func_event_triggers.tc+1−4

This diff is large: its 12 lines aren’t shown here.

tools/testing/selftests/ftrace/test.d/ftrace/func_mod_trace.tc+1−1

This diff is large: its 9 lines aren’t shown here.

tools/testing/selftests/ftrace/test.d/ftrace/func_profiler.tc+1−4

This diff is large: its 12 lines aren’t shown here.

tools/testing/selftests/ftrace/test.d/ftrace/func_set_ftrace_file.tc+1−4

This diff is large: its 12 lines aren’t shown here.

tools/testing/selftests/ftrace/test.d/ftrace/func_stack_tracer.tc+2−0

This diff is large: its 9 lines aren’t shown here.

tools/testing/selftests/ftrace/test.d/ftrace/func_traceonoff_triggers.tc+1−4

This diff is large: its 12 lines aren’t shown here.

tools/testing/selftests/ftrace/test.d/functions+6−0

This diff is large: its 10 lines aren’t shown here.

tools/testing/selftests/ftrace/test.d/kprobe/kprobe_args_type.tc+1−1

This diff is large: its 9 lines aren’t shown here.

tools/testing/selftests/ftrace/test.d/kprobe/kprobe_ftrace.tc+2−0

This diff is large: its 9 lines aren’t shown here.

tools/testing/selftests/gpio/Makefile+6−6

This diff is large: its 20 lines aren’t shown here.

tools/testing/selftests/intel_pstate/Makefile+1−1

This diff is large: its 8 lines aren’t shown here.

tools/testing/selftests/ipc/msgque.c+1−1

This diff is large: its 9 lines aren’t shown here.

tools/testing/selftests/kselftest/runner.sh+1−1

This diff is large: its 9 lines aren’t shown here.

tools/testing/selftests/kselftest_deps.sh+272−0

This diff is large: its 273 lines aren’t shown here.

tools/testing/selftests/kselftest_harness.h+3−3

This diff is large: its 27 lines aren’t shown here.

tools/testing/selftests/kvm/Makefile+28−1

This diff is large: its 52 lines aren’t shown here.

tools/testing/selftests/kvm/include/evmcs.h+2−2

This diff is large: its 11 lines aren’t shown here.

tools/testing/selftests/kvm/lib/x86_64/vmx.c+3−0

This diff is large: its 10 lines aren’t shown here.

tools/testing/selftests/memfd/Makefile+14−3

This diff is large: its 29 lines aren’t shown here.

tools/testing/selftests/net/fib_nexthops.sh+23−0

This diff is large: its 30 lines aren’t shown here.

tools/testing/selftests/net/fib_tests.sh+8−2

This diff is large: its 25 lines aren’t shown here.

tools/testing/selftests/net/tcp_mmap.c+5−2

This diff is large: its 28 lines aren’t shown here.

tools/testing/selftests/seccomp/seccomp_bpf.c+8−6

This diff is large: its 29 lines aren’t shown here.

tools/testing/selftests/tpm2/test_smoke.sh+2−11

This diff is large: its 20 lines aren’t shown here.

tools/testing/selftests/tpm2/test_space.sh+1−8

This diff is large: its 13 lines aren’t shown here.

tools/testing/selftests/vm/Makefile+2−2

This diff is large: its 18 lines aren’t shown here.

tools/testing/selftests/vm/run_vmtests+1−1

This diff is large: its 9 lines aren’t shown here.

tools/testing/selftests/wireguard/netns.sh+51−3

This diff is large: its 96 lines aren’t shown here.

tools/testing/selftests/wireguard/qemu/arch/powerpc64le.config+1−0

This diff is large: its 5 lines aren’t shown here.

tools/testing/selftests/wireguard/qemu/debug.config+0−1

This diff is large: its 8 lines aren’t shown here.

tools/virtio/Makefile+3−2

This diff is large: its 19 lines aren’t shown here.

tools/virtio/asm/barrier.h+1−0

This diff is large: its 6 lines aren’t shown here.

tools/virtio/generated/autoconf.h+0−0

Empty file added.

tools/virtio/linux/compiler.h+1−0

This diff is large: its 6 lines aren’t shown here.

tools/vm/Makefile+2−0

This diff is large: its 9 lines aren’t shown here.

virt/kvm/arm/hyp/aarch32.c+6−2

This diff is large: its 19 lines aren’t shown here.

virt/kvm/arm/psci.c+40−0

This diff is large: its 81 lines aren’t shown here.

virt/kvm/arm/vgic/vgic-init.c+16−3

This diff is large: its 43 lines aren’t shown here.

virt/kvm/arm/vgic/vgic-its.c+9−2

This diff is large: its 24 lines aren’t shown here.

virt/kvm/arm/vgic/vgic-mmio-v2.c+10−6

This diff is large: its 35 lines aren’t shown here.

virt/kvm/arm/vgic/vgic-mmio-v3.c+18−13

This diff is large: its 74 lines aren’t shown here.

virt/kvm/arm/vgic/vgic-mmio.c+170−58

This diff is large: its 326 lines aren’t shown here.

virt/kvm/arm/vgic/vgic-mmio.h+19−0

This diff is large: its 36 lines aren’t shown here.

Merge tag 'noinstr-lds-2020-05-19' into core/rcu

Keyboard shortcuts

General

Issues and pull requests

Issues, pull requests and Inbox

Repository code