qemu-stable
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [PULL 4/9] target/i386: Adjust feature level according to FEAT_7_1_E


From: Michael Tokarev
Subject: Re: [PULL 4/9] target/i386: Adjust feature level according to FEAT_7_1_EDX
Date: Mon, 17 Jul 2023 14:19:59 +0300
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101 Thunderbird/102.13.0

07.07.2023 17:04, Paolo Bonzini wrote:
From: Tao Su <tao1.su@linux.intel.com>

If FEAT_7_1_EAX is 0 and FEAT_7_1_EDX is non-zero, as is the case
with a Granite Rapids host and
'-cpu host,-avx-vnni,-avx512-bf16,-fzrm,-fsrs,-fsrc,-amx-fp16', we can't
get CPUID_7_1 leaf even though CPUID_7_1_EDX has non-zero value.

Update cpuid_level_func7 according to CPUID_7_1_EDX, otherwise
guest may report wrong maximum number sub-leaves in leaf 07H.

Fixes: eaaa197d5b11 ("target/i386: Add support for AVX-VNNI-INT8 in CPUID 
enumeration")
Cc: qemu-stable@nongnu.org

This patch is not applicable for any stable qemu releases, since the
commit it fixes (eaaa197d5b11, which also introduced FEAT_7_1_EDX) is
not in any released version, it is v8.0.0-290-geaaa197d5b.

This is just for the information (I'm not picking it up for stable),
there's nothing wrong with it tagged for stable initially.

Thanks!

/mjt



reply via email to

[Prev in Thread] Current Thread [Next in Thread]