linux-kernel - Re: [PATCH] cpufreq: intel_pstate: Check IDA feature only during MSR 0x199 write

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 for Android: free password hash cracker in your pocket

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

Message-ID: <CAJZ5v0g4FWk3w3Kby5p0ZgMfoTjt0XA=6g=x1UKQhaAUGashMg@mail.gmail.com>
Date: Wed, 12 Nov 2025 18:24:38 +0100
From: "Rafael J. Wysocki" <rafael@...nel.org>
To: Srinivas Pandruvada <srinivas.pandruvada@...ux.intel.com>
Cc: rafael@...nel.org, linux-pm@...r.kernel.org, linux-kernel@...r.kernel.org, 
	trenn@...e.de, Aaron Rainbolt <arainbolt@...cus.org>
Subject: Re: [PATCH] cpufreq: intel_pstate: Check IDA feature only during MSR
 0x199 write

On Tue, Nov 11, 2025 at 2:09 AM Srinivas Pandruvada
<srinivas.pandruvada@...ux.intel.com> wrote:
>
> Commit ac4e04d9e378 ("cpufreq: intel_pstate: Unchecked MSR aceess in
> legacy mode") introduced a check for feature X86_FEATURE_IDA to verify
> turbo mode support. Although this is the correct way to check for turbo
> mode, it causes issues on some platforms that disable turbo during OS
> boot but enable it later. Without this feature check, users were able to
> write 0 to /sys/devices/system/cpu/intel_pstate/no_turbo post-boot to
> get turbo mode frequencies.
>
> To restore the old behavior while still addressing the unchecked MSR
> issue on some Skylake-X systems, limit the X86_FEATURE_IDA check to only
> when setting MSR 0x199 Turbo Engage Bit (bit 32).
>
> Fixes: ac4e04d9e378 ("cpufreq: intel_pstate: Unchecked MSR aceess in legacy mode")
> Reported-by: Aaron Rainbolt <arainbolt@...cus.org>
> Closes: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/2122531
> Tested-by: Aaron Rainbolt <arainbolt@...cus.org>
> Signed-off-by: Srinivas Pandruvada <srinivas.pandruvada@...ux.intel.com>
> ---
>  drivers/cpufreq/intel_pstate.c | 9 ++++-----
>  1 file changed, 4 insertions(+), 5 deletions(-)
>
> diff --git a/drivers/cpufreq/intel_pstate.c b/drivers/cpufreq/intel_pstate.c
> index 38897bb14a2c..492a10f1bdbf 100644
> --- a/drivers/cpufreq/intel_pstate.c
> +++ b/drivers/cpufreq/intel_pstate.c
> @@ -603,9 +603,6 @@ static bool turbo_is_disabled(void)
>  {
>         u64 misc_en;
>
> -       if (!cpu_feature_enabled(X86_FEATURE_IDA))
> -               return true;
> -
>         rdmsrq(MSR_IA32_MISC_ENABLE, misc_en);
>
>         return !!(misc_en & MSR_IA32_MISC_ENABLE_TURBO_DISABLE);
> @@ -2106,7 +2103,8 @@ static u64 atom_get_val(struct cpudata *cpudata, int pstate)
>         u32 vid;
>
>         val = (u64)pstate << 8;
> -       if (READ_ONCE(global.no_turbo) && !READ_ONCE(global.turbo_disabled))
> +       if (READ_ONCE(global.no_turbo) && !READ_ONCE(global.turbo_disabled) &&
> +           cpu_feature_enabled(X86_FEATURE_IDA))
>                 val |= (u64)1 << 32;
>
>         vid_fp = cpudata->vid.min + mul_fp(
> @@ -2271,7 +2269,8 @@ static u64 core_get_val(struct cpudata *cpudata, int pstate)
>         u64 val;
>
>         val = (u64)pstate << 8;
> -       if (READ_ONCE(global.no_turbo) && !READ_ONCE(global.turbo_disabled))
> +       if (READ_ONCE(global.no_turbo) && !READ_ONCE(global.turbo_disabled) &&
> +           cpu_feature_enabled(X86_FEATURE_IDA))
>                 val |= (u64)1 << 32;
>
>         return val;
> --

Applied as 6.18-rc material, with some adjustments in the subject and changelog.

Thanks!