Re: [PATCH] x86/ucode/AMD: apply the patch early on every logical thread

2023-01-09 Thread Sergey Dyasli
On Thu, Jan 5, 2023 at 10:56 PM Andrew Cooper wrote: > > diff --git a/xen/arch/x86/cpu/microcode/private.h > > b/xen/arch/x86/cpu/microcode/private.h > > index 73b095d5bf..c4c6729f56 100644 > > --- a/xen/arch/x86/cpu/microcode/private.h > > +++ b/xen/arch/x86/cpu/microcode/private.h > > @@ -7,6

Re: [PATCH] x86/ucode/AMD: apply the patch early on every logical thread

2023-01-05 Thread Andrew Cooper
On 05/01/2023 1:20 pm, Sergey Dyasli wrote: > The original issue has been reported on AMD Bulldozer-based CPUs where > ucode loading loses the LWP feature bit in order to gain the IBPB bit. > LWP disabling is per-SMT core modification and needs to happen on each > sibling SMT thread despite the

[PATCH] x86/ucode/AMD: apply the patch early on every logical thread

2023-01-05 Thread Sergey Dyasli
The original issue has been reported on AMD Bulldozer-based CPUs where ucode loading loses the LWP feature bit in order to gain the IBPB bit. LWP disabling is per-SMT core modification and needs to happen on each sibling SMT thread despite the shared microcode engine. Otherwise, logical CPUs will