Jingyue Wu 5733100450 Divergence analysis for GPU programs ... Summary: Some optimizations such as jump threading and loop unswitching can negatively affect performance when applied to divergent branches. The divergence analysis added in this patch conservatively estimates which branches in a GPU program can diverge. This information can then help LLVM to run certain optimizations selectively. Test Plan: test/Analysis/DivergenceAnalysis/NVPTX/diverge.ll Reviewers: resistor, hfinkel, eliben, meheff, jholewinski Subscribers: broune, bjarke.roune, madhur13490, tstellarAMD, dberlin, echristo, jholewinski, llvm-commits Differential Revision: http://reviews.llvm.org/D8576 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@234567 91177308-0d34-0410-b5e6-96231b3b80d8		2015-04-10 05:03:50 +00:00
..
Analysis	Divergence analysis for GPU programs	2015-04-10 05:03:50 +00:00
AsmParser	DebugInfo: Move DIFlag-related API from DIDescriptor to DebugNode	2015-04-07 01:21:40 +00:00
Bitcode	IR: Remove MDTupleTypedArrayWrapper::operator MDTuple*()	2015-04-07 16:50:39 +00:00
CodeGen	[WinEHPrepare] Don't rely on the order of IR	2015-04-10 04:56:17 +00:00
DebugInfo
ExecutionEngine	[jitlistener] Remove unused code	2015-04-08 13:17:48 +00:00
Fuzzer	Removing a spurious space; NFC.	2015-04-06 16:09:13 +00:00
IR	IR: Preserve use-list order by default in bitcode	2015-04-09 17:41:20 +00:00
IRReader
LineEditor
Linker	IR: Stop using DIDescriptor::is*() and auto-casting	2015-04-06 23:18:49 +00:00
LTO	Simplify use of formatted_raw_ostream.	2015-04-09 21:06:08 +00:00
MC	Simplify use of formatted_raw_ostream.	2015-04-09 21:06:08 +00:00
Object
Option
Passes
ProfileData
Support	Misc cleanup. NFC.	2015-04-09 16:59:07 +00:00
TableGen
Target	Divergence analysis for GPU programs	2015-04-10 05:03:50 +00:00
Transforms	Eliminate O(n^2) worst-case behavior in SSA construction	2015-04-08 18:26:20 +00:00
CMakeLists.txt
LLVMBuild.txt
Makefile