Re: [tip: x86/urgent] x86/CPU/AMD: Add RDSEED fix for Zen5

[Peter Jung]

On 10/28/25 13:01, tip-bot2 for Gregory Price wrote:
> The following commit has been merged into the x86/urgent branch of tip:
>
> Commit-ID:     607b9fb2ce248cc5b633c5949e0153838992c152
> Gitweb:https://git.kernel.org/tip/607b9fb2ce248cc5b633c5949e0153838992c152
> Author:        Gregory Price<gourry@xxxxxxxxxx>
> AuthorDate:    Mon, 20 Oct 2025 11:13:55 +02:00
> Committer:     Borislav Petkov (AMD)<bp@xxxxxxxxx>
> CommitterDate: Tue, 28 Oct 2025 12:37:49 +01:00
>
> x86/CPU/AMD: Add RDSEED fix for Zen5
>
> There's an issue with RDSEED's 16-bit and 32-bit register output
> variants on Zen5 which return a random value of 0 "at a rate inconsistent
> with randomness while incorrectly signaling success (CF=1)". Search the
> web for AMD-SB-7055 for more detail.
>
> Add a fix glue which checks microcode revisions.
>
>    [ bp: Add microcode revisions checking, rewrite. ]
>
> Cc:stable@xxxxxxxxxxxxxxx
> Signed-off-by: Gregory Price<gourry@xxxxxxxxxx>
> Signed-off-by: Borislav Petkov (AMD)<bp@xxxxxxxxx>
> Link:https://lore.kernel.org/r/20251018024010.4112396-1-gourry@xxxxxxxxxx
> ---
>   arch/x86/kernel/cpu/amd.c | 10 ++++++++++
>   1 file changed, 10 insertions(+)
>
> diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
> index ccaa51c..bc29be6 100644
> --- a/arch/x86/kernel/cpu/amd.c
> +++ b/arch/x86/kernel/cpu/amd.c
> @@ -1035,8 +1035,18 @@ static void init_amd_zen4(struct cpuinfo_x86 *c)
>   	}
>   }
>   
> +static const struct x86_cpu_id zen5_rdseed_microcode[] = {
> +	ZEN_MODEL_STEP_UCODE(0x1a, 0x02, 0x1, 0x0b00215a),
> +	ZEN_MODEL_STEP_UCODE(0x1a, 0x11, 0x0, 0x0b101054),
> +};
> +
>   static void init_amd_zen5(struct cpuinfo_x86 *c)
>   {
> +	if (!x86_match_min_microcode_rev(zen5_rdseed_microcode)) {
> +		clear_cpu_cap(c, X86_FEATURE_RDSEED);
> +		msr_clear_bit(MSR_AMD64_CPUID_FN_7, 18);
> +		pr_emerg_once("RDSEED32 is broken. Disabling the corresponding CPUID bit.\n");
> +	}
>   }
>   
>   static void init_amd(struct cpuinfo_x86 *c)

Hi all,

This fix seems to break quite a bunch of users in CachyOS. There has 
been now several users reporting that there system can not get properly 
into the graphical interface.

CachyOS is compiling the packages with -march=znver5 and the GCC 
compiler currently does pass RDSEED.

This patch results into that also Client CPUs (Strix Point, Granite 
Ridge), can not execute this. There has been a microcode fix deployed in 
linux-firmware for Turin, but no other microcode changes seen yet.

I think it would be possible to exclude clients or providing a fix for this.


Example log:

Nov 03 13:37:33 hells drkonqi-coredump-processor[1073]: Incompatible 
processor. This Qt build requires the following features:
Nov 03 13:37:33 hells drkonqi-coredump-processor[1073]:     rdseed
Nov 03 13:37:33 hells systemd-coredump[1077]: Process 1073 
(drkonqi-coredum) of user 0 terminated abnormally with signal 6/ABRT, 
processing...
Nov 03 13:37:33 hells systemd[1]: Started Process Core Dump (PID 
1077/UID 0).
Nov 03 13:37:33 hells systemd[1]: Started Pass systemd-coredump journal 
entries to relevant user for potential DrKonqi handling.
Nov 03 13:37:33 hells drkonqi-coredump-processor[1079]: Incompatible 
processor. This Qt build requires the following features:
Nov 03 13:37:33 hells drkonqi-coredump-processor[1079]:     rdseed
Nov 03 13:37:33 hells systemd-coredump[1082]: Process 1079 
(drkonqi-coredum) of user 0 terminated abnormally with signal 6/ABRT, 
processing...
Nov 03 13:37:33 hells systemd-coredump[1071]: Process 1049 (sddm) of 
user 0 dumped core.


Best regards,

Peter