All of lore.kernel.org
 help / color / mirror / Atom feed
From: Sean Christopherson <seanjc@google.com>
To: Yang Weijiang <weijiang.yang@intel.com>
Cc: pbonzini@redhat.com, kvm@vger.kernel.org
Subject: Re: [kvm-unit-tests PATCH v4 2/2] x86: Check platform vPMU capabilities before run lbr tests
Date: Thu, 7 Jul 2022 20:28:36 +0000	[thread overview]
Message-ID: <YsdB9Is3oHgSFg8S@google.com> (raw)
In-Reply-To: <20220628093203.73160-2-weijiang.yang@intel.com>

On Tue, Jun 28, 2022, Yang Weijiang wrote:
> Use new helper to check whether pmu is available and Perfmon/Debug
> capbilities are supported before read MSR_IA32_PERF_CAPABILITIES to
> avoid test failure. The issue can be captured when enable_pmu=0.
> 
> Suggested-by: Sean Christopherson <seanjc@google.com>
> Signed-off-by: Yang Weijiang <weijiang.yang@intel.com>
> 
> v4:
> - Put the X86_FEATURE_PDCM to the right place. [Sean]
> ---

<version info goes here>

>  lib/x86/processor.h |  1 +
>  x86/pmu_lbr.c       | 32 +++++++++++++-------------------
>  2 files changed, 14 insertions(+), 19 deletions(-)
> 
> diff --git a/lib/x86/processor.h b/lib/x86/processor.h
> index 7b6ee92..7a35c7f 100644
> --- a/lib/x86/processor.h
> +++ b/lib/x86/processor.h
> @@ -146,6 +146,7 @@ static inline bool is_intel(void)
>   */
>  #define	X86_FEATURE_MWAIT		(CPUID(0x1, 0, ECX, 3))
>  #define	X86_FEATURE_VMX			(CPUID(0x1, 0, ECX, 5))
> +#define	X86_FEATURE_PDCM		(CPUID(0x1, 0, ECX, 15))
>  #define	X86_FEATURE_PCID		(CPUID(0x1, 0, ECX, 17))
>  #define	X86_FEATURE_MOVBE		(CPUID(0x1, 0, ECX, 22))
>  #define	X86_FEATURE_TSC_DEADLINE_TIMER	(CPUID(0x1, 0, ECX, 24))
> diff --git a/x86/pmu_lbr.c b/x86/pmu_lbr.c
> index 688634d..497df1e 100644
> --- a/x86/pmu_lbr.c
> +++ b/x86/pmu_lbr.c
> @@ -15,6 +15,7 @@
>  #define MSR_LBR_SELECT		0x000001c8
>  
>  volatile int count;
> +u32 lbr_from, lbr_to;
>  
>  static noinline int compute_flag(int i)
>  {
> @@ -38,18 +39,6 @@ static noinline int lbr_test(void)
>  	return 0;
>  }
>  
> -union cpuid10_eax {
> -	struct {
> -		unsigned int version_id:8;
> -		unsigned int num_counters:8;
> -		unsigned int bit_width:8;
> -		unsigned int mask_length:8;
> -	} split;
> -	unsigned int full;
> -} eax;
> -
> -u32 lbr_from, lbr_to;
> -
>  static void init_lbr(void *index)
>  {
>  	wrmsr(lbr_from + *(int *) index, 0);
> @@ -63,7 +52,7 @@ static bool test_init_lbr_from_exception(u64 index)
>  
>  int main(int ac, char **av)
>  {
> -	struct cpuid id = cpuid(10);
> +	u8 version = pmu_version();
>  	u64 perf_cap;
>  	int max, i;
>  
> @@ -74,19 +63,24 @@ int main(int ac, char **av)
>  		return 0;
>  	}
>  
> -	perf_cap = rdmsr(MSR_IA32_PERF_CAPABILITIES);
> -	eax.full = id.a;
> -
> -	if (!eax.split.version_id) {
> +	if (!version) {

If the previous patch exposes cpu_has_pmu(), then this open coded check goes away
in favor of the more obvious:

	if (!cpu_has_pmu()) {

>  		printf("No pmu is detected!\n");
>  		return report_summary();
>  	}
> +
> +	if (!this_cpu_has(X86_FEATURE_PDCM)) {
> +		printf("Perfmon/Debug Capabilities MSR isn't supported\n");
> +		return report_summary();
> +	}
> +
> +	perf_cap = rdmsr(MSR_IA32_PERF_CAPABILITIES);
> +
>  	if (!(perf_cap & PMU_CAP_LBR_FMT)) {
> -		printf("No LBR is detected!\n");
> +		printf("(Architectural) LBR is not supported.\n");
>  		return report_summary();
>  	}
>  
> -	printf("PMU version:		 %d\n", eax.split.version_id);
> +	printf("PMU version:		 %d\n", version);

And with the open coded check gone, this can be:
	
	printf("PMU version:		 %d\n", pmu_version());

>  	printf("LBR version:		 %ld\n", perf_cap & PMU_CAP_LBR_FMT);
>  
>  	/* Look for LBR from and to MSRs */
> -- 
> 2.27.0
> 

  reply	other threads:[~2022-07-07 20:28 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-06-28  9:32 [kvm-unit-tests PATCH v4 1/2] x86: Skip perf related tests when platform cannot support Yang Weijiang
2022-06-28  9:32 ` [kvm-unit-tests PATCH v4 2/2] x86: Check platform vPMU capabilities before run lbr tests Yang Weijiang
2022-07-07 20:28   ` Sean Christopherson [this message]
2022-07-08  0:57     ` Yang, Weijiang
2022-07-07 20:26 ` [kvm-unit-tests PATCH v4 1/2] x86: Skip perf related tests when platform cannot support Sean Christopherson
2022-07-08  0:48   ` Yang, Weijiang

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=YsdB9Is3oHgSFg8S@google.com \
    --to=seanjc@google.com \
    --cc=kvm@vger.kernel.org \
    --cc=pbonzini@redhat.com \
    --cc=weijiang.yang@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.