KVM: selftests: fix hyperv_clock test
PATCH 1/1 KVM: selftests: fix hyperv_clock test
The test was mistakenly using addr_gpa2hva on a gva and that happened to work accidentally.
commit 106a2e766eae ("KVM: selftests:
Lower the min virtual address for misc page allocations") revealed this bug.
--- tools/testing/selftests/kvm/x86_64/hyperv_clock.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
tools/testing/selftests/kvm/x86_64/hyperv_clock.c +1 -1
@@ -215,7 +215,7 @@ int main(void)vcpu_set_hv_cpuid (vm,VCPU_ID ); tsc_page_gva =vm_vaddr_alloc_page (vm); -memset (addr_gpa2hva (vm, tsc_page_gva),0x0 ,getpagesize ()); +memset (addr_gva2hva (vm, tsc_page_gva),0x0 ,getpagesize ());TEST_ASSERT ((addr_gva2gpa (vm, tsc_page_gva) & (getpagesize () -1 )) ==0 ,"TSC page has to be page aligned\n" );vcpu_args_set (vm,VCPU_ID ,2 , tsc_page_gva,addr_gva2gpa (vm, tsc_page_gva));
Signature · 1 line
2.26.3
Paolo Bonzini
maintainerQuoted 8 lines · Maxim Levitsky
The test was mistakenly using
addr_gpa2hvaon a gva and that happened to work accidentally.commit
106a2e766eae("KVM: selftests:
Lower the min virtual address for misc page allocations") revealed this bug.Fixes 106a2e766eae ("KVM: selftests: Lower the min virtual address for misc page allocations")
The right "Fixes" is 2c7f76b4c42b ("selftests: kvm: Add basic Hyper-V clocksources tests", 2021-03-18).
Paolo
Quoted 19 lines
Signed-off-by Maxim Levitsky <[email protected]>--- tools/testing/selftests/kvm/x86_64/hyperv_clock.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)tools/testing/selftests/kvm/x86_64/hyperv_clock.c +1
-1@@ -215,7 +215,7 @@ int main(void)vcpu_set_hv_cpuid (vm,VCPU_ID ); tsc_page_gva =vm_vaddr_alloc_page (vm); -memset (addr_gpa2hva (vm, tsc_page_gva),0x0 ,getpagesize ()); +memset (addr_gva2hva (vm, tsc_page_gva),0x0 ,getpagesize ());TEST_ASSERT ((addr_gva2gpa (vm, tsc_page_gva) & (getpagesize () -1 )) ==0 ,"TSC page has to be page aligned\n" );vcpu_args_set (vm,VCPU_ID ,2 , tsc_page_gva,addr_gva2gpa (vm, tsc_page_gva));
Quoted 4 lines · Paolo Bonzini
Quoted 3 lines · Maxim Levitsky
The test was mistakenly using
addr_gpa2hvaon a gva and that happened to work accidentally.
Quoted 8 lines
Quoted 5 lines
commit
106a2e766eae("KVM: selftests:
Lower the min virtual address for misc page allocations") revealed this bug.Fixes 106a2e766eae ("KVM: selftests: Lower the min virtual address for misc page allocations")The right "Fixes" is
2c7f76b4c42b("selftests: kvm: Add basic Hyper-V clocksources tests", 2021-03-18).
Makes sense. I wasn't sure about which commit to blame here. I'll resend the patch.
Best regards,
Maxim Levitsky
Quoted 22 lines
Paolo
Quoted 19 lines
Signed-off-by Maxim Levitsky <[email protected]>--- tools/testing/selftests/kvm/x86_64/hyperv_clock.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)tools/testing/selftests/kvm/x86_64/hyperv_clock.c +1
-1@@ -215,7 +215,7 @@ int main(void)vcpu_set_hv_cpuid (vm,VCPU_ID ); tsc_page_gva =vm_vaddr_alloc_page (vm); -memset (addr_gpa2hva (vm, tsc_page_gva),0x0 ,getpagesize ()); +memset (addr_gva2hva (vm, tsc_page_gva),0x0 ,getpagesize ());TEST_ASSERT ((addr_gva2gpa (vm, tsc_page_gva) & (getpagesize () -1 )) ==0 ,"TSC page has to be page aligned\n" );vcpu_args_set (vm,VCPU_ID ,2 , tsc_page_gva,addr_gva2gpa (vm, tsc_page_gva));