On 21.02.2023 22:26, Sergey Dyasli wrote: > On Tue, Feb 21, 2023 at 2:03 PM Jan Beulich <jbeul...@suse.com> wrote: >> >> On 15.02.2023 16:38, Sergey Dyasli wrote: >>> --- a/xen/arch/x86/cpu/microcode/core.c >>> +++ b/xen/arch/x86/cpu/microcode/core.c >>> @@ -398,10 +398,16 @@ static int cf_check microcode_nmi_callback( >>> (!ucode_in_nmi && cpu == primary) ) >>> return 0; >>> >>> - if ( cpu == primary ) >>> + if ( boot_cpu_data.x86_vendor == X86_VENDOR_AMD ) >> >> Given their origin, I'm pretty certain Hygon wants treating the same here >> and below. > > Hygon? ucode_ops is currently initialised only for Amd and Intel.
Hmm, you're right, we still haven't sorted that aspect. I'm inclined to say though that adding Hygon in your changes right away reduces the burden later on. And it'll do no harm as long as early_microcode_init() isn't properly dealing with Hygon. > Speaking of which, I'm thinking about adding a new function > is_cpu_primary() there. This would make the core code much cleaner. > I'll see if I can make it work. Thanks - I was actually meaning to suggest something like that, realizing the potential improvement only after sending the earlier reply. Even just a static helper (without new hook) may already improve things. Jan
