[llvm-commits] [llvm] r167967 - in /llvm/trunk: lib/Target/X86/X86ISelLowering.cpp test/CodeGen/X86/2008-10-27-StackRealignment.ll test/CodeGen/X86/memcpy-2.ll test/CodeGen/X86/memset-sse-stack-realignment.ll test/CodeGen/X86/memset64-on-x86-32.ll

[Eli Friedman]

On Wed, Nov 14, 2012 at 12:08 PM, Benjamin Kramer
<benny.kra at googlemail.com> wrote:
> Author: d0k
> Date: Wed Nov 14 14:08:40 2012
> New Revision: 167967
>
> URL: http://llvm.org/viewvc/llvm-project?rev=167967&view=rev
> Log:
> X86: Enable SSE memory intrinsics even when stack alignment is less than 16 bytes.
>
> The stack realignment code was fixed to work when there is stack realignment and
> a dynamic alloca is present so this shouldn't cause correctness issues anymore.

No alignment issues, true, but there is a performance cost to
realigning the stack, especially since we don't optimize the alignment
out when we don't actually spill an SSE/AVX register.  Not sure how
much of an issue this is in practice.

-Eli