diff options
| author | Linus Torvalds <torvalds@linux-foundation.org> | 2018-01-21 21:48:35 +0300 |
|---|---|---|
| committer | Linus Torvalds <torvalds@linux-foundation.org> | 2018-01-21 21:48:35 +0300 |
| commit | 5515114211c5096987b8e622c2ca57929cebabaa (patch) | |
| tree | 73924d90c4a9435b9eb0b29093b7b8a121c78864 /arch/x86/include | |
| parent | 319f1e04c7e7016233f998a1a48cd3d17da2108f (diff) | |
| parent | 3f7d875566d8e79c5e0b2c9a413e91b2c29e0854 (diff) | |
| download | linux-5515114211c5096987b8e622c2ca57929cebabaa.tar.xz | |
Merge branch 'x86-pti-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 pti fixes from Thomas Gleixner:
"A small set of fixes for the meltdown/spectre mitigations:
- Make kprobes aware of retpolines to prevent probes in the retpoline
thunks.
- Make the machine check exception speculation protected. MCE used to
issue an indirect call directly from the ASM entry code. Convert
that to a direct call into a C-function and issue the indirect call
from there so the compiler can add the retpoline protection,
- Make the vmexit_fill_RSB() assembly less stupid
- Fix a typo in the PTI documentation"
* 'x86-pti-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/retpoline: Optimize inline assembler for vmexit_fill_RSB
x86/pti: Document fix wrong index
kprobes/x86: Disable optimizing on the function jumps to indirect thunk
kprobes/x86: Blacklist indirect thunk functions for kprobes
retpoline: Introduce start/end markers of indirect thunk
x86/mce: Make machine check speculation protected
Diffstat (limited to 'arch/x86/include')
| -rw-r--r-- | arch/x86/include/asm/nospec-branch.h | 10 | ||||
| -rw-r--r-- | arch/x86/include/asm/traps.h | 1 |
2 files changed, 8 insertions, 3 deletions
diff --git a/arch/x86/include/asm/nospec-branch.h b/arch/x86/include/asm/nospec-branch.h index 7b45d8424150..4ad41087ce0e 100644 --- a/arch/x86/include/asm/nospec-branch.h +++ b/arch/x86/include/asm/nospec-branch.h @@ -194,6 +194,9 @@ enum spectre_v2_mitigation { SPECTRE_V2_IBRS, }; +extern char __indirect_thunk_start[]; +extern char __indirect_thunk_end[]; + /* * On VMEXIT we must ensure that no RSB predictions learned in the guest * can be followed in the host, by overwriting the RSB completely. Both @@ -203,16 +206,17 @@ enum spectre_v2_mitigation { static inline void vmexit_fill_RSB(void) { #ifdef CONFIG_RETPOLINE - unsigned long loops = RSB_CLEAR_LOOPS / 2; + unsigned long loops; asm volatile (ANNOTATE_NOSPEC_ALTERNATIVE ALTERNATIVE("jmp 910f", __stringify(__FILL_RETURN_BUFFER(%0, RSB_CLEAR_LOOPS, %1)), X86_FEATURE_RETPOLINE) "910:" - : "=&r" (loops), ASM_CALL_CONSTRAINT - : "r" (loops) : "memory" ); + : "=r" (loops), ASM_CALL_CONSTRAINT + : : "memory" ); #endif } + #endif /* __ASSEMBLY__ */ #endif /* __NOSPEC_BRANCH_H__ */ diff --git a/arch/x86/include/asm/traps.h b/arch/x86/include/asm/traps.h index 31051f35cbb7..3de69330e6c5 100644 --- a/arch/x86/include/asm/traps.h +++ b/arch/x86/include/asm/traps.h @@ -88,6 +88,7 @@ dotraplinkage void do_simd_coprocessor_error(struct pt_regs *, long); #ifdef CONFIG_X86_32 dotraplinkage void do_iret_error(struct pt_regs *, long); #endif +dotraplinkage void do_mce(struct pt_regs *, long); static inline int get_si_code(unsigned long condition) { |