Sanjay Patel 5fb1d32848 [x86-64] allow mfence even with -mno-sse (PR23203) ...

As shown in:
https://llvm.org/bugs/show_bug.cgi?id=23203
...we currently die because lowering believes that mfence is allowed without SSE2 on x86-64,
but the instruction def doesn't know that.

I don't know if allowing mfence without SSE is right, but if not, at least now it's consistently wrong. :)

Differential Revision: http://reviews.llvm.org/D17219



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@260828 91177308-0d34-0410-b5e6-96231b3b80d8

2016-02-13 17:26:29 +00:00

..

Analysis

[ConstantFolding] Reduce APInt and APFloat copying.

2016-02-13 16:54:14 +00:00

AsmParser

…

Bitcode

Restore "[ThinLTO] Use MD5 hash in function index." with fix

2016-02-10 21:55:02 +00:00

CodeGen

[AggressiveAntiDepBreaker] Skip some unnecessary BitVector copies.

2016-02-13 16:39:39 +00:00

DebugInfo

…

ExecutionEngine

[Orc] Add lazy-JITting support for i386.

2016-02-10 01:02:33 +00:00

Fuzzer

[libFuzzer] don't require seed in fuzzer::Mutate, instead use the global Fuzzer object for fuzzer::Mutate. This makes custom mutators fast

2016-02-13 06:24:18 +00:00

IR

[ConstantFolding] Reduce APInt and APFloat copying.

2016-02-13 16:54:14 +00:00

IRReader

…

LibDriver

…

LineEditor

…

Linker

Delete the deprecated LLVMLinkModules.

2016-02-12 15:28:45 +00:00

LTO

…

MC

[MC] Fixed parsing of macro arguments where expressions with spaces are present.

2016-02-11 13:48:49 +00:00

Object

Use copy initialization.

2016-02-11 19:54:18 +00:00

Option

…

Passes

…

ProfileData

[PGO] Add another interface for annotateValueSite

2016-02-12 21:36:17 +00:00

Support

Add AMDGPU related triple vendors/OSes

2016-02-13 01:56:21 +00:00

TableGen

…

Target

[x86-64] allow mfence even with -mno-sse (PR23203)

2016-02-13 17:26:29 +00:00

Transforms

Use ArrayRef to hide SmallVector details, kill a useless vector copy along the way.

2016-02-13 16:01:12 +00:00

CMakeLists.txt

…

LLVMBuild.txt

…