On Mon, Jun 15, 2020 at 05:20PM +0200, Peter Zijlstra wrote: > On Mon, Jun 15, 2020 at 05:03:27PM +0200, Peter Zijlstra wrote: > > > Yes, I think so. x86_64 needs lib/memcpy_64.S in .noinstr.text then. For > > i386 it's an __always_inline inline-asm thing. > > Bah, I tried writing it without memcpy, but clang inserts memcpy anyway > :/
Hmm, __builtin_memcpy() won't help either. Turns out, Clang 11 got __builtin_memcpy_inline(): https://reviews.llvm.org/D73543 The below works, no more crash on either KASAN or KCSAN with Clang. We can test if we have it with __has_feature(__builtin_memcpy_inline) (although that's currently not working as expected, trying to fix :-/). Would a memcpy_inline() be generally useful? It's not just Clang but also GCC that isn't entirely upfront about which memcpy is inlined and which isn't. If the compiler has __builtin_memcpy_inline(), we can use it, otherwise the arch likely has to provide the implementation. Thoughts? Thanks, -- Marco ------ >8 ------ diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c index af75109485c2..3e07beae2a75 100644 --- a/arch/x86/kernel/traps.c +++ b/arch/x86/kernel/traps.c @@ -690,13 +690,13 @@ struct bad_iret_stack *fixup_bad_iret(struct bad_iret_stack *s) (struct bad_iret_stack *)__this_cpu_read(cpu_tss_rw.x86_tss.sp0) - 1; /* Copy the IRET target to the temporary storage. */ - memcpy(&tmp.regs.ip, (void *)s->regs.sp, 5*8); + __builtin_memcpy_inline(&tmp.regs.ip, (void *)s->regs.sp, 5*8); /* Copy the remainder of the stack from the current stack. */ - memcpy(&tmp, s, offsetof(struct bad_iret_stack, regs.ip)); + __builtin_memcpy_inline(&tmp, s, offsetof(struct bad_iret_stack, regs.ip)); /* Update the entry stack */ - memcpy(new_stack, &tmp, sizeof(tmp)); + __builtin_memcpy_inline(new_stack, &tmp, sizeof(tmp)); BUG_ON(!user_mode(&new_stack->regs)); return new_stack;
