[PATCH] [X86] Allow atomic operations using immediates to avoid using a register

[Nadav Rotem]

Robin, 

I am sorry for the delay in review.  The changes LGTM. 

Thanks,
Nadav

> On Sep 2, 2014, at 1:29 PM, Robin Morisset <morisset at google.com> wrote:
> 
> ping.
> 
> 
> On Wed, Aug 27, 2014 at 2:16 PM, Robin Morisset <morisset at google.com <mailto:morisset at google.com>> wrote:
> ping, still waiting on someone with knowledge of the X86 backend.
> 
> 
> On Thu, Aug 7, 2014 at 4:12 PM, Robin Morisset <morisset at google.com <mailto:morisset at google.com>> wrote:
>   Add back the modification to X86CodeEmitter as the Jit resurrected
> 
> + an extra comment on the implicitness of the lock prefix before xchg.
> 
> http://reviews.llvm.org/D4796 <http://reviews.llvm.org/D4796>
> 
> Files:
>   lib/Target/X86/X86CodeEmitter.cpp
>   lib/Target/X86/X86InstrCompiler.td
>   lib/Target/X86/X86MCInstLower.cpp
>   test/CodeGen/X86/atomic_mi.ll
> 
> 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20140902/867e5b7b/attachment.html>