diff options
| author | Linus Torvalds | 2024-03-12 09:45:34 -0700 |
|---|---|---|
| committer | Linus Torvalds | 2024-03-12 09:45:34 -0700 |
| commit | e66c58f743513119f703f3a47f0f93a8e82c0028 (patch) | |
| tree | 0b8b40835b16002511bf18a127fd47079220dd42 /arch/x86 | |
| parent | 0e33cf955f07e3991e45109cb3e29fbc9ca51d06 (diff) | |
| parent | 532a0c57d7ff75e8f07d4e25cba4184989e2a241 (diff) | |
Merge tag 'x86-apic-2024-03-12' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 APIC fixup from Dave Hansen:
"Revert VERW fixed addressing patch.
The reverted commit is not x86/apic material and was cruft left over
from a merge.
I believe the sequence of events went something like this:
- The commit in question was added to x86/urgent
- x86/urgent was merged into x86/apic to resolve a conflict
- The commit was zapped from x86/urgent, but *not* from x86/apic
- x86/apic got pullled (yesterday)
I think we need to be a bit more vigilant when zapping things to make
sure none of the other branches are depending on the zapped material"
* tag 'x86-apic-2024-03-12' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
Revert "x86/bugs: Use fixed addressing for VERW operand"
Diffstat (limited to 'arch/x86')
| -rw-r--r-- | arch/x86/include/asm/nospec-branch.h | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/arch/x86/include/asm/nospec-branch.h b/arch/x86/include/asm/nospec-branch.h index d0b8bb762838..fc3a8a3c7ffe 100644 --- a/arch/x86/include/asm/nospec-branch.h +++ b/arch/x86/include/asm/nospec-branch.h @@ -314,7 +314,7 @@ * Note: Only the memory operand variant of VERW clears the CPU buffers. */ .macro CLEAR_CPU_BUFFERS - ALTERNATIVE "", __stringify(verw mds_verw_sel), X86_FEATURE_CLEAR_CPU_BUF + ALTERNATIVE "", __stringify(verw _ASM_RIP(mds_verw_sel)), X86_FEATURE_CLEAR_CPU_BUF .endm #else /* __ASSEMBLY__ */ |