linux-kernel - Re: [kvm-unit-tests patch v3 3/8] x86/pmu: Fix incorrect masking of fixed counters

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: <aR-WAGVNZwNh7Xo8@google.com>
Date: Thu, 20 Nov 2025 14:28:16 -0800
From: Sean Christopherson <seanjc@...gle.com>
To: Dapeng Mi <dapeng1.mi@...ux.intel.com>
Cc: Paolo Bonzini <pbonzini@...hat.com>, kvm@...r.kernel.org, linux-kernel@...r.kernel.org, 
	Jim Mattson <jmattson@...gle.com>, Mingwei Zhang <mizhang@...gle.com>, 
	Zide Chen <zide.chen@...el.com>, Das Sandipan <Sandipan.Das@....com>, 
	Shukla Manali <Manali.Shukla@....com>, Xiaoyao Li <xiaoyao.li@...el.com>, 
	Dapeng Mi <dapeng1.mi@...el.com>, dongsheng <dongsheng.x.zhang@...el.com>, 
	Yi Lai <yi1.lai@...el.com>
Subject: Re: [kvm-unit-tests patch v3 3/8] x86/pmu: Fix incorrect masking of
 fixed counters

On Wed, Sep 03, 2025, Dapeng Mi wrote:
> From: dongsheng <dongsheng.x.zhang@...el.com>
> 
> The current implementation mistakenly limits the width of fixed counters
> to the width of GP counters. Corrects the logic to ensure fixed counters
> are properly masked according to their own width.
> 
> Opportunistically refine the GP counter bitwidth processing code.
> 
> Signed-off-by: dongsheng <dongsheng.x.zhang@...el.com>
> Co-developed-by: Dapeng Mi <dapeng1.mi@...ux.intel.com>
> Signed-off-by: Dapeng Mi <dapeng1.mi@...ux.intel.com>
> Tested-by: Yi Lai <yi1.lai@...el.com>
> ---
>  x86/pmu.c | 8 +++-----
>  1 file changed, 3 insertions(+), 5 deletions(-)
> 
> diff --git a/x86/pmu.c b/x86/pmu.c
> index 04946d10..44c728a5 100644
> --- a/x86/pmu.c
> +++ b/x86/pmu.c
> @@ -556,18 +556,16 @@ static void check_counter_overflow(void)
>  		int idx;
>  
>  		cnt.count = overflow_preset;
> -		if (pmu_use_full_writes())
> -			cnt.count &= (1ull << pmu.gp_counter_width) - 1;
> -
>  		if (i == pmu.nr_gp_counters) {
>  			if (!pmu.is_intel)
>  				break;
>  
>  			cnt.ctr = fixed_events[0].unit_sel;
> -			cnt.count = measure_for_overflow(&cnt);

Per commit 7ec3b67a ("x86/pmu: Reset the expected count of the fixed counter 0
when i386"), re-measuring for the fixed counter is necessary when running a 32-bit
guests.  I didn't see failures (spotted this by inspection), but I don't see any
point in making this change without good reason.

> -			cnt.count &= (1ull << pmu.gp_counter_width) - 1;
> +			cnt.count &= (1ull << pmu.fixed_counter_width) - 1;
>  		} else {
>  			cnt.ctr = MSR_GP_COUNTERx(i);
> +			if (pmu_use_full_writes())
> +				cnt.count &= (1ull << pmu.gp_counter_width) - 1;
>  		}
>  
>  		if (i % 2)
> -- 
> 2.34.1
>