History

lavc/h264idct: fix RISC-V group multiplier

After the branch, the expected SEW/LMUL ratio is 1 byte/vector. So we have to set the same ratio before branching (QEMU does not care, but real hardware does).

Rémi Denis-Courmont committed 1 year ago

Commit bbb0fde1 parent 8a5b74f1 file · +4 −4Browse files

Unified · Split

libavcodec/riscv/h264idct_rvv.S+4−4
@@ -644,7 +644,7 @@ func ff_h264_idct4_add\type\()_\depth\()_rvv, zve32x · 643 unmodified lines
644 sd a4, 24(sp)644 sd a4, 24(sp)
645 ld a0, 0(a0) # dest[0]645 ld a0, 0(a0) # dest[0]
646 addi a1, a1, 16 * 4 # &block_offset[16]646 addi a1, a1, 16 * 4 # &block_offset[16]
647 vsetivli zero, 4, e8, mf4, ta, ma647 vsetivli zero, 4, e8, m1, ta, ma
648 jal .Lidct4_add4_\depth\()_rvv648 jal .Lidct4_add4_\depth\()_rvv
649 649
650 ld a4, 24(sp) # nnzc650 ld a4, 24(sp) # nnzc
@@ -655,7 +655,7 @@ func ff_h264_idct4_add\type\()_\depth\()_rvv, zve32x · 4 unmodified lines
655 ld a0, 8(a0) # dest[1]655 ld a0, 8(a0) # dest[1]
656 lla t0, ff_h264_scan8 + 32656 lla t0, ff_h264_scan8 + 32
657.ifc \type, 8_422657.ifc \type, 8_422
658 vsetivli zero, 4, e8, mf4, ta, ma658 vsetivli zero, 4, e8, m1, ta, ma
659 jal .Lidct4_add4_\depth\()_rvv659 jal .Lidct4_add4_\depth\()_rvv
660 660
661 ld a4, 24(sp) # nnzc661 ld a4, 24(sp) # nnzc
@@ -665,7 +665,7 @@ func ff_h264_idct4_add\type\()_\depth\()_rvv, zve32x · 3 unmodified lines
665 addi a1, t5, (-8 - 4) * 4 # &block_offset[24]665 addi a1, t5, (-8 - 4) * 4 # &block_offset[24]
666 ld a0, 0(a0) # dest[0]666 ld a0, 0(a0) # dest[0]
667 lla t0, ff_h264_scan8 + 24667 lla t0, ff_h264_scan8 + 24
668 vsetivli zero, 4, e8, mf4, ta, ma668 vsetivli zero, 4, e8, m1, ta, ma
669 jal .Lidct4_add4_\depth\()_rvv669 jal .Lidct4_add4_\depth\()_rvv
670 670
671 ld a4, 24(sp) # nnzc671 ld a4, 24(sp) # nnzc
@@ -679,7 +679,7 @@ func ff_h264_idct4_add\type\()_\depth\()_rvv, zve32x · 7 unmodified lines
679 ld ra, 8(sp)679 ld ra, 8(sp)
680 ld s0, 0(sp)680 ld s0, 0(sp)
681 addi sp, sp, 32681 addi sp, sp, 32
682 vsetivli zero, 4, e8, mf4, ta, ma682 vsetivli zero, 4, e8, m1, ta, ma
683 j .Lidct4_add4_\depth\()_rvv683 j .Lidct4_add4_\depth\()_rvv
684endfunc684endfunc
685.endm685.endm

Keyboard shortcuts

General

Show keyboard shortcuts
?
Focus the current search
/

Issues and pull requests

Open filters
f

Issues, pull requests and Inbox

Select next item
j↓
Select previous item
k↑
Open selected item
Enter
Dismiss filters or clear selection
Esc

Repository code

Go to file
t