Owen Anderson 9917700af4 Add a target-specific DAG combine on X86 to fold the common pattern of ...

fence-atomic-fence down to just the atomic op.  This is possible thanks to
X86's relatively strong memory model, which guarantees that locked instructions
(which are used to implement atomics) are implicit fences.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@74435 91177308-0d34-0410-b5e6-96231b3b80d8

2009-06-29 18:04:45 +00:00

..

Analysis

Missed one.

2009-06-29 00:53:49 +00:00

Archive

Avoid leaking memory in an error path. Noticed

2009-06-11 08:09:49 +00:00

AsmParser

Address review comments: add 3 ARM calling conventions.

2009-06-16 18:50:49 +00:00

Bitcode

Fix PR4336: Iterating over use-def chains doesn't seem to be deterministic.

2009-06-16 05:15:21 +00:00

CodeGen

Eliminate a layer of indirection in LoopInfo and MachineLoopInfo.

2009-06-27 21:22:48 +00:00

CompilerDriver

Make dynamic LLVMC plugins work on Windows (finally!).

2009-06-29 03:09:15 +00:00

Debugger

Oops...I committed too much.

2009-03-13 04:39:26 +00:00

ExecutionEngine

Get rid of unnecessary global variables.

2009-06-26 16:46:15 +00:00

Linker

Add a new "available_externally" linkage type. This is intended

2009-04-13 05:44:34 +00:00

MC

MC: Truncate values when printing, to keep 'as' happy.

2009-06-25 21:03:18 +00:00

Support

Add triple for OpenBSD.

2009-06-29 13:36:13 +00:00

System

Fix compilation without pthreads.

2009-06-26 08:48:03 +00:00

Target

Add a target-specific DAG combine on X86 to fold the common pattern of

2009-06-29 18:04:45 +00:00

Transforms

Don't try to split a loop when the controlling icmp instruction

2009-06-27 22:58:27 +00:00

VMCore

Add more vector ValueTypes for AVX and other extended vector instruction

2009-06-29 16:47:10 +00:00

Makefile

Start flushing out MCContext.

2009-06-23 22:01:43 +00:00