linux-kernel - Re: [PATCH v2 0/2] Inject a PMI for KVM Guest when ToPA buffer is filled

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <9c0efe99-0878-a136-b4db-68d09c486232@redhat.com>
Date:   Fri, 22 Feb 2019 18:24:55 +0100
From:   Paolo Bonzini <pbonzini@...hat.com>
To:     Luwei Kang <luwei.kang@...el.com>, linux-kernel@...r.kernel.org,
        kvm@...r.kernel.org
Cc:     tglx@...utronix.de, mingo@...hat.com, bp@...en8.de, hpa@...or.com,
        x86@...nel.org, rkrcmar@...hat.com, peterz@...radead.org,
        acme@...nel.org, alexander.shishkin@...ux.intel.com,
        jolsa@...hat.com, namhyung@...nel.org, kan.liang@...ux.intel.com,
        ak@...ux.intel.com, konrad.wilk@...cle.com,
        Janakarajan.Natarajan@....com, mattst88@...il.com,
        tim.c.chen@...ux.intel.com
Subject: Re: [PATCH v2 0/2] Inject a PMI for KVM Guest when ToPA buffer is
 filled

On 19/02/19 01:26, Luwei Kang wrote:
> Each intel processor trace table of physical addresses (ToPA) entry
> has an INT bit. If this bit is set, the processor will signal a
> performance-monitoring interrupt (PMI) when the corresponding trace
> output region is filled. This patch set will inject a PMI for Intel
> Processor Trace when ToPA buffer is filled.
> 
>>>From v1:
>  - Exported a global function pointers may not a good chioce.
>    Add a new member in kvm_guest_cbs to send Intel PT PMI for KVM guest.
> 
> Luwei Kang (2):
>   KVM: x86: Inject PMI for KVM guest
>   KVM: x86: Add support of clear Trace_ToPA_PMI status
> 
>  arch/x86/events/intel/core.c     |  6 +++++-
>  arch/x86/include/asm/kvm_host.h  |  1 +
>  arch/x86/include/asm/msr-index.h |  8 ++++++++
>  arch/x86/kvm/vmx/pmu_intel.c     |  8 +++++++-
>  arch/x86/kvm/x86.c               | 10 ++++++++++
>  include/linux/perf_event.h       |  1 +
>  6 files changed, 32 insertions(+), 2 deletions(-)
> 

Queued, with just a little change in patch 2:

diff --git a/arch/x86/kvm/vmx/pmu_intel.c b/arch/x86/kvm/vmx/pmu_intel.c
index 6dee7cf24442..f8502c376b37 100644
--- a/arch/x86/kvm/vmx/pmu_intel.c
+++ b/arch/x86/kvm/vmx/pmu_intel.c
@@ -297,9 +297,9 @@ static void intel_pmu_refresh(struct kvm_vcpu *vcpu)
 	pmu->global_ctrl = ((1ull << pmu->nr_arch_gp_counters) - 1) |
 		(((1ull << pmu->nr_arch_fixed_counters) - 1) << INTEL_PMC_IDX_FIXED);
 	pmu->global_ctrl_mask = ~pmu->global_ctrl;
-	pmu->global_ovf_ctrl_mask = ~(pmu->global_ctrl |
-				MSR_CORE_PERF_GLOBAL_OVF_CTRL_OVF_BUF |
-				MSR_CORE_PERF_GLOBAL_OVF_CTRL_COND_CHGD);
+	pmu->global_ovf_ctrl_mask = pmu->global_ctrl_mask
+			& ~(MSR_CORE_PERF_GLOBAL_OVF_CTRL_OVF_BUF |
+			    MSR_CORE_PERF_GLOBAL_OVF_CTRL_COND_CHGD);
 	if (kvm_x86_ops->pt_supported())
 		pmu->global_ovf_ctrl_mask &=
 				~MSR_CORE_PERF_GLOBAL_OVF_CTRL_TRACE_TOPA_PMI;

this makes the review easier, and also makes the assignment look
the same as the one inside the "if (kvm_x86_ops->pt_supported())".

Paolo