On 15.03.2023 13:11, Greg Kroah-Hartman wrote:
From: Andrew Cooper <andrew.cooper3@xxxxxxxxxx> commit b0563468eeac88ebc70559d52a0b66efc37e4e9d upstream. AMD Erratum 1386 is summarised as: XSAVES Instruction May Fail to Save XMM Registers to the Provided State Save Area This piece of accidental chronomancy causes the %xmm registers to occasionally reset back to an older value. Ignore the XSAVES feature on all AMD Zen1/2 hardware. The XSAVEC instruction (which works fine) is equivalent on affected parts. [ bp: Typos, move it into the F17h-specific function. ] Reported-by: Tavis Ormandy <taviso@xxxxxxxxx> Signed-off-by: Andrew Cooper <andrew.cooper3@xxxxxxxxxx> Signed-off-by: Borislav Petkov (AMD) <bp@xxxxxxxxx> Cc: <stable@xxxxxxxxxx> Link: https://lore.kernel.org/r/20230307174643.1240184-1-andrew.cooper3@xxxxxxxxxx Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx> --- arch/x86/kernel/cpu/amd.c | 9 +++++++++ 1 file changed, 9 insertions(+) --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -205,6 +205,15 @@ static void init_amd_k6(struct cpuinfo_x return; } #endif + /* + * Work around Erratum 1386. The XSAVES instruction malfunctions in + * certain circumstances on Zen1/2 uarch, and not all parts have had + * updated microcode at the time of writing (March 2023). + * + * Affected parts all have no supervisor XSAVE states, meaning that + * the XSAVEC instruction (which works fine) is equivalent. + */ + clear_cpu_cap(c, X86_FEATURE_XSAVES); }
This is essentially a well-intended NOP since K6 well predates XSAVES, and init_amd_k6() is *not* called for Zen CPUs. This workaround should have been added to init_amd_zn() function instead. 4.19 and 4.14 backports of this patch have the same problem. Thanks, Maciej
