[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <4737f149-c5b7-8a51-7cc5-8bda6e98308b@ispras.ru>
Date: Tue, 28 Feb 2023 21:47:22 +0300 (MSK)
From: Alexander Monakov <amonakov@...ras.ru>
To: Borislav Petkov <bp@...en8.de>
cc: Andrew Cooper <andrew.cooper3@...rix.com>,
Tavis Ormandy <taviso@...il.com>, linux-kernel@...r.kernel.org,
x86@...nel.org
Subject: Re: x86: AMD Zen2 ymm registers rolling back
On Wed, 22 Feb 2023, Borislav Petkov wrote:
> On Wed, Feb 22, 2023 at 09:38:09AM +0000, Andrew Cooper wrote:
> > This sounds suspiciously like an errata which was fixed with a ucode
> > update last year.
>
> Yes, it looks like it.
>
> Alternatively, you can try booting with "clearcpuid=xsaves" - that
> should take care of your observation too but yeah, you should rather
> update your microcode.
Hi folks,
I can reproduce this bug on AMD Renoir SoC:
vendor_id : AuthenticAMD
cpu family : 23
model : 96
model name : AMD Ryzen 5 4600G with Radeon Graphics
stepping : 1
microcode : 0x8600104
for which there's no microcode update, the microcode_amd_fam17h.bin file
in the linux-firmware.git repo carries only the following patches:
$ ./amd_ucode_info.py microcode_amd_fam17h.bin
Microcode patches in microcode_amd_fam17h.bin:
Family=0x17 Model=0x08 Stepping=0x02: Patch=0x0800820d Length=3200 bytes
Family=0x17 Model=0x01 Stepping=0x02: Patch=0x0800126e Length=3200 bytes
Family=0x17 Model=0x31 Stepping=0x00: Patch=0x08301055 Length=3200 bytes
I've seen microcode version increase after a BIOS update, so it seems like
internally microcode patches exist for Renoir too, but it's up to hardware
vendors to pick them up as part of a BIOS update. Is there any chance of
a conventional release like for the above three CPU models?
Thanks.
Alexander
Powered by blists - more mailing lists