linux-kernel - Re: [PATCH v1 2/8] KVM: x86: PEBS output to Intel PT MSRs emulation

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

Message-ID: <20191029150243.GM4097@hirez.programming.kicks-ass.net>
Date:   Tue, 29 Oct 2019 16:02:43 +0100
From:   Peter Zijlstra <peterz@...radead.org>
To:     Luwei Kang <luwei.kang@...el.com>
Cc:     kvm@...r.kernel.org, linux-kernel@...r.kernel.org,
        pbonzini@...hat.com, rkrcmar@...hat.com,
        sean.j.christopherson@...el.com, vkuznets@...hat.com,
        wanpengli@...cent.com, jmattson@...gle.com, joro@...tes.org,
        tglx@...utronix.de, mingo@...hat.com, bp@...en8.de, hpa@...or.com,
        x86@...nel.org, ak@...ux.intel.com, thomas.lendacky@....com,
        acme@...nel.org, mark.rutland@....com,
        alexander.shishkin@...ux.intel.com, jolsa@...hat.com,
        namhyung@...nel.org
Subject: Re: [PATCH v1 2/8] KVM: x86: PEBS output to Intel PT MSRs emulation

On Sun, Oct 27, 2019 at 07:11:11PM -0400, Luwei Kang wrote:
> Intel new hardware introduces a mechanism to direct PEBS records
> output into the Intel PT buffer that can be used for enabling PEBS
> in KVM guest. This patch implements the registers read and write
> emulation when PEBS is supported in KVM guest.
> 
> KMM needs to reprogram the counters when the value of these MSRs
> be changed that to make sure it can take effect in hardware.
> 
> Signed-off-by: Luwei Kang <luwei.kang@...el.com>
> ---
>  arch/x86/include/asm/kvm_host.h  |  4 +++
>  arch/x86/include/asm/msr-index.h |  6 ++++
>  arch/x86/kvm/vmx/capabilities.h  | 15 ++++++++++
>  arch/x86/kvm/vmx/pmu_intel.c     | 63 ++++++++++++++++++++++++++++++++++++++--
>  4 files changed, 86 insertions(+), 2 deletions(-)
> 
> diff --git a/arch/x86/include/asm/msr-index.h b/arch/x86/include/asm/msr-index.h
> index 20ce682..d22f8d9 100644
> --- a/arch/x86/include/asm/msr-index.h
> +++ b/arch/x86/include/asm/msr-index.h
> @@ -131,9 +131,13 @@
>  #define LBR_INFO_ABORT			BIT_ULL(61)
>  #define LBR_INFO_CYCLES			0xffff
>  
> +#define MSR_IA32_PEBS_PMI_AFTER_REC	BIT_ULL(60)
> +#define MSR_IA32_PEBS_OUTPUT_PT		BIT_ULL(61)
> +#define MSR_IA32_PEBS_OUTPUT_MASK	(3ULL << 61)
>  #define MSR_IA32_PEBS_ENABLE		0x000003f1
>  #define MSR_PEBS_DATA_CFG		0x000003f2
>  #define MSR_IA32_DS_AREA		0x00000600
> +#define MSR_IA32_PERF_CAP_PEBS_OUTPUT_PT	BIT_ULL(16)
>  #define MSR_IA32_PERF_CAPABILITIES	0x00000345
>  #define MSR_PEBS_LD_LAT_THRESHOLD	0x000003f6
>  
> @@ -665,6 +669,8 @@
>  #define MSR_IA32_MISC_ENABLE_FERR			(1ULL << MSR_IA32_MISC_ENABLE_FERR_BIT)
>  #define MSR_IA32_MISC_ENABLE_FERR_MULTIPLEX_BIT		10
>  #define MSR_IA32_MISC_ENABLE_FERR_MULTIPLEX		(1ULL << MSR_IA32_MISC_ENABLE_FERR_MULTIPLEX_BIT)
> +#define MSR_IA32_MISC_ENABLE_PEBS_BIT			12
> +#define MSR_IA32_MISC_ENABLE_PEBS			(1ULL << MSR_IA32_MISC_ENABLE_PEBS_BIT)
>  #define MSR_IA32_MISC_ENABLE_TM2_BIT			13
>  #define MSR_IA32_MISC_ENABLE_TM2			(1ULL << MSR_IA32_MISC_ENABLE_TM2_BIT)
>  #define MSR_IA32_MISC_ENABLE_ADJ_PREF_DISABLE_BIT	19

Some of these already exist but are local to perf. Don't blindly
introduce more without unifying.