diff options
| author | Vipin Sharma <vipinsh@google.com> | 2022-11-03 12:17:19 -0700 |
|---|---|---|
| committer | Sean Christopherson <seanjc@google.com> | 2022-11-16 10:03:24 -0800 |
| commit | d886724ea81c6a4dc5e37d4ee09287a31ab8335e (patch) | |
| tree | 9a8efec469058b33158eaef207065b9f1429c382 /tools/testing/selftests/kvm/lib/perf_test_util.c | |
| parent | 0001725d0f9b5d749540021befb67c117d566416 (diff) | |
| download | linux-d886724ea81c6a4dc5e37d4ee09287a31ab8335e.tar.gz linux-d886724ea81c6a4dc5e37d4ee09287a31ab8335e.tar.bz2 linux-d886724ea81c6a4dc5e37d4ee09287a31ab8335e.zip | |
KVM: selftests: Allowing running dirty_log_perf_test on specific CPUs
Add a command line option, -c, to pin vCPUs to physical CPUs (pCPUs),
i.e. to force vCPUs to run on specific pCPUs.
Requirement to implement this feature came in discussion on the patch
"Make page tables for eager page splitting NUMA aware"
https://lore.kernel.org/lkml/YuhPT2drgqL+osLl@google.com/
This feature is useful as it provides a way to analyze performance based
on the vCPUs and dirty log worker locations, like on the different NUMA
nodes or on the same NUMA nodes.
To keep things simple, implementation is intentionally very limited,
either all of the vCPUs will be pinned followed by an optional main
thread or nothing will be pinned.
Signed-off-by: Vipin Sharma <vipinsh@google.com>
Suggested-by: David Matlack <dmatlack@google.com>
Reviewed-by: Sean Christopherson <seanjc@google.com>
Link: https://lore.kernel.org/r/20221103191719.1559407-8-vipinsh@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Diffstat (limited to 'tools/testing/selftests/kvm/lib/perf_test_util.c')
| -rw-r--r-- | tools/testing/selftests/kvm/lib/perf_test_util.c | 8 |
1 files changed, 7 insertions, 1 deletions
diff --git a/tools/testing/selftests/kvm/lib/perf_test_util.c b/tools/testing/selftests/kvm/lib/perf_test_util.c index 9618b37c66f7..3a1d0a44419b 100644 --- a/tools/testing/selftests/kvm/lib/perf_test_util.c +++ b/tools/testing/selftests/kvm/lib/perf_test_util.c @@ -2,6 +2,8 @@ /* * Copyright (C) 2020, Google LLC. */ +#define _GNU_SOURCE + #include <inttypes.h> #include "kvm_util.h" @@ -243,6 +245,10 @@ void __weak perf_test_setup_nested(struct kvm_vm *vm, int nr_vcpus, struct kvm_v static void *vcpu_thread_main(void *data) { struct vcpu_thread *vcpu = data; + int vcpu_idx = vcpu->vcpu_idx; + + if (perf_test_args.pin_vcpus) + kvm_pin_this_task_to_pcpu(perf_test_args.vcpu_to_pcpu[vcpu_idx]); WRITE_ONCE(vcpu->running, true); @@ -255,7 +261,7 @@ static void *vcpu_thread_main(void *data) while (!READ_ONCE(all_vcpu_threads_running)) ; - vcpu_thread_fn(&perf_test_args.vcpu_args[vcpu->vcpu_idx]); + vcpu_thread_fn(&perf_test_args.vcpu_args[vcpu_idx]); return NULL; } |