| Message ID | 20240306230153.786365-1-seanjc@google.com (mailing list archive) |
|---|---|
| Headers | show |
| Series | x86/pmu: PEBS fixes and new testcases | expand |
On 3/7/2024 7:01 AM, Sean Christopherson wrote: > One bug fix where pmu_pebs attempts to enable PEBS for fixed counter on > CPUs without Extended PEBS, and two new testcases to verify adaptive > PEBS functionality. > > The new testcases are intended both to demonstrate that adaptive PEBS > virtualization is currently broken, and to serve as a gatekeeper for > re-enabling adapative PEBS in the future. > > https://lore.kernel.org/all/ZeepGjHCeSfadANM@google.com > > Sean Christopherson (4): > x86/pmu: Enable PEBS on fixed counters iff baseline PEBS is support > x86/pmu: Iterate over adaptive PEBS flag combinations > x86/pmu: Test adaptive PEBS without any adaptive counters > x86/pmu: Add a PEBS test to verify the host LBRs aren't leaked to the > guest > > lib/x86/pmu.h | 6 ++- > x86/pmu_pebs.c | 109 +++++++++++++++++++++++++++---------------------- > 2 files changed, 66 insertions(+), 49 deletions(-) Tested-by: Dapeng Mi <dapeng1.mi@linux.intel.com> > > base-commit: 55dd01b4f066577b49f03fb7146723c4a65822c4
On Wed, 06 Mar 2024 15:01:49 -0800, Sean Christopherson wrote: > One bug fix where pmu_pebs attempts to enable PEBS for fixed counter on > CPUs without Extended PEBS, and two new testcases to verify adaptive > PEBS functionality. > > The new testcases are intended both to demonstrate that adaptive PEBS > virtualization is currently broken, and to serve as a gatekeeper for > re-enabling adapative PEBS in the future. > > [...] Applied to kvm-x86 next. Dapeng, I didn't address your feedback about adding finer grained message prefixes. I'm not opposed to doing so, I'm just extremely short on cycles and want to get the fixes landed. [1/4] x86/pmu: Enable PEBS on fixed counters iff baseline PEBS is support https://github.com/kvm-x86/kvm-unit-tests/commit/79aa106cd427 [2/4] x86/pmu: Iterate over adaptive PEBS flag combinations https://github.com/kvm-x86/kvm-unit-tests/commit/fc17d5276b38 [3/4] x86/pmu: Test adaptive PEBS without any adaptive counters https://github.com/kvm-x86/kvm-unit-tests/commit/2cb2af7f53db [4/4] x86/pmu: Add a PEBS test to verify the host LBRs aren't leaked to the guest https://github.com/kvm-x86/kvm-unit-tests/commit/8d0f574f4e4d -- https://github.com/kvm-x86/kvm-unit-tests/tree/next
> -----Original Message----- > From: Sean Christopherson <seanjc@google.com> > Sent: Thursday, June 6, 2024 7:21 AM > To: Sean Christopherson <seanjc@google.com>; Paolo Bonzini > <pbonzini@redhat.com> > Cc: kvm@vger.kernel.org; Like Xu <like.xu.linux@gmail.com>; Mingwei Zhang > <mizhang@google.com>; Zhenyu Wang <zhenyuw@linux.intel.com>; Zhang, > Xiong Y <xiong.y.zhang@intel.com>; Lv, Zhiyuan <zhiyuan.lv@intel.com>; Mi, > Dapeng1 <dapeng1.mi@intel.com> > Subject: Re: [kvm-unit-tests PATCH 0/4] x86/pmu: PEBS fixes and new testcases > > On Wed, 06 Mar 2024 15:01:49 -0800, Sean Christopherson wrote: > > One bug fix where pmu_pebs attempts to enable PEBS for fixed counter > > on CPUs without Extended PEBS, and two new testcases to verify > > adaptive PEBS functionality. > > > > The new testcases are intended both to demonstrate that adaptive PEBS > > virtualization is currently broken, and to serve as a gatekeeper for > > re-enabling adapative PEBS in the future. > > > > [...] > > Applied to kvm-x86 next. Dapeng, I didn't address your feedback about adding > finer grained message prefixes. I'm not opposed to doing so, I'm just extremely > short on cycles and want to get the fixes landed. It's fine. I think I can help to add it.
One bug fix where pmu_pebs attempts to enable PEBS for fixed counter on CPUs without Extended PEBS, and two new testcases to verify adaptive PEBS functionality. The new testcases are intended both to demonstrate that adaptive PEBS virtualization is currently broken, and to serve as a gatekeeper for re-enabling adapative PEBS in the future. https://lore.kernel.org/all/ZeepGjHCeSfadANM@google.com Sean Christopherson (4): x86/pmu: Enable PEBS on fixed counters iff baseline PEBS is support x86/pmu: Iterate over adaptive PEBS flag combinations x86/pmu: Test adaptive PEBS without any adaptive counters x86/pmu: Add a PEBS test to verify the host LBRs aren't leaked to the guest lib/x86/pmu.h | 6 ++- x86/pmu_pebs.c | 109 +++++++++++++++++++++++++++---------------------- 2 files changed, 66 insertions(+), 49 deletions(-) base-commit: 55dd01b4f066577b49f03fb7146723c4a65822c4