History

x86/kvm: Provide optimized version of vcpu_is_preempted() for x86-64

It was found when running fio sequential write test with a XFS ramdisk on a KVM guest running on a 2-socket x86-64 system, the %CPU times as reported by perf were as follows: 69.75% 0.59% fio [k] down_write 69.15% 0.01% fio [k] call_rwsem_down_write_failed 67.12% 1.12% fio [k] rwsem_down_write_failed 63.48% 52.77% fio [k] osq_lock 9.46% 7.88% fio [k] __raw_callee_save___kvm_vcpu_is_preempt 3.93% 3.93% fio [k] __kvm_vcpu_is_preempted Making vcpu_is_preempted() a callee-save function has a relatively high cost on x86-64 primarily due to at least one more cacheline of data access from the saving and restoring of registers (8 of them) to and from stack as well as one more level of function call. To reduce this performance overhead, an optimized assembly version of the the __raw_callee_save___kvm_vcpu_is_preempt() function is provided for x86-64. With this patch applied on a KVM guest on a 2-socket 16-core 32-thread system with 16 parallel jobs (8 on each socket), the aggregrate bandwidth of the fio test on an XFS ramdisk were as follows: I/O Type w/o patch with patch -------- --------- ---------- random read 8141.2 MB/s 8497.1 MB/s seq read 8229.4 MB/s 8304.2 MB/s random write 1675.5 MB/s 1701.5 MB/s seq write 1681.3 MB/s 1699.9 MB/s There are some increases in the aggregated bandwidth because of the patch. The perf data now became: 70.78% 0.58% fio [k] down_write 70.20% 0.01% fio [k] call_rwsem_down_write_failed 69.70% 1.17% fio [k] rwsem_down_write_failed 59.91% 55.42% fio [k] osq_lock 10.14% 10.14% fio [k] __kvm_vcpu_is_preempted The assembly code was verified by using a test kernel module to compare the output of C __kvm_vcpu_is_preempted() and that of assembly __raw_callee_save___kvm_vcpu_is_preempt() to verify that they matched.

Waiman Long committed 9 years ago

Commit dd0fd8b1 parent 6c629852 files · +33 −0Browse files

Unified · Split

2 files
  1. arch/x86/kernel/asm-offsets_64.c +9 −0
  2. arch/x86/kernel/kvm.c +24 −0
arch/x86/kernel/asm-offsets_64.c+9−0
@@ -13,6 +13,10 @@ static char syscalls_ia32[] = { · 12 unmodified lines
1313 #include <asm/syscalls_32.h>
1414 };
1515
16+#if defined(CONFIG_KVM_GUEST) && defined(CONFIG_PARAVIRT_SPINLOCKS)
17+#include <asm/kvm_para.h>
18+#endif
19+
1620 int main(void)
1721 {
1822 #ifdef CONFIG_PARAVIRT
@@ -22,6 +26,11 @@ int main(void) · 3 unmodified lines
2226 BLANK();
2327 #endif
2428
29+#if defined(CONFIG_KVM_GUEST) && defined(CONFIG_PARAVIRT_SPINLOCKS)
30+ OFFSET(KVM_STEAL_TIME_preempted, kvm_steal_time, preempted);
31+ BLANK();
32+#endif
33+
2534 #define ENTRY(entry) OFFSET(pt_regs_ ## entry, pt_regs, entry)
2635 ENTRY(bx);
2736 ENTRY(cx);
arch/x86/kernel/kvm.c+24−0
@@ -589,6 +589,7 @@ out: · 588 unmodified lines
589589 local_irq_restore(flags);
590590 }
591591
592+#ifdef CONFIG_X86_32
592593 __visible bool __kvm_vcpu_is_preempted(long cpu)
593594 {
594595 struct kvm_steal_time *src = &per_cpu(steal_time, cpu);
@@ -597,6 +598,29 @@ __visible bool __kvm_vcpu_is_preempted(long cpu) · 2 unmodified lines
597598 }
598599 PV_CALLEE_SAVE_REGS_THUNK(__kvm_vcpu_is_preempted);
599600
601+#else
602+
603+#include <asm/asm-offsets.h>
604+
605+extern bool __raw_callee_save___kvm_vcpu_is_preempted(long);
606+
607+/*
608+ * Hand-optimize version for x86-64 to avoid 8 64-bit register saving and
609+ * restoring to/from the stack.
610+ */
611+asm(
612+".pushsection .text;"
613+".global __raw_callee_save___kvm_vcpu_is_preempted;"
614+".type __raw_callee_save___kvm_vcpu_is_preempted, @function;"
615+"__raw_callee_save___kvm_vcpu_is_preempted:"
616+"movq __per_cpu_offset(,%rdi,8), %rax;"
617+"cmpb $0, " __stringify(KVM_STEAL_TIME_preempted) "+steal_time(%rax);"
618+"setne %al;"
619+"ret;"
620+".popsection");
621+
622+#endif
623+
600624 /*
601625 * Setup pv_lock_ops to exploit KVM_FEATURE_PV_UNHALT if present.
602626 */

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