Matt Arsenault 168a5152ad AMDGPU: Make SIInsertWaits about a factor of 4 faster ... This was the slowest target custom pass and was spending 80% of the time in getMinimalPhysRegClass which was called for every register operand. Try to use the statically known register class when possible from the instruction's MCOperandInfo. There are a few pseudo instructions which are not well behaved with unknown register classes which still require the expensive physical register class search. There are a few other possibilities for making this even faster, such as not inspecting implicit operands. For now those are checked because it is technically possible to have a scalar load into exec or vcc which can be implicitly used. llvm-svn: 249079		2015-10-01 21:43:15 +00:00
..
Analysis	Refactor computeKnownBits alignment handling code	2015-09-30 11:55:45 +00:00
AsmParser	HHVM calling conventions.	2015-09-29 22:09:16 +00:00
Bitcode	[Bitcode][Asm] Teach LLVM to read and write operand bundles.	2015-09-24 23:34:52 +00:00
CodeGen	[WinEH] Emit __C_specific_handler tables for the new IR	2015-10-01 21:38:24 +00:00
DebugInfo	Use makeArrayRef or None to avoid unnecessarily mentioning the ArrayRef type extra times. NFC	2015-09-21 05:32:41 +00:00
ExecutionEngine	Fix performance problem in long-running SectionMemoryManagers	2015-10-01 02:45:07 +00:00
Fuzzer	[libFuzzer] Marking exported symbols as visible. Patch by Mike Aizatsky	2015-09-30 22:22:37 +00:00
IR	Fix debug info with SafeStack.	2015-09-30 19:55:43 +00:00
IRReader
LibDriver	There is only one saver of strings.	2015-08-13 01:07:02 +00:00
LineEditor
Linker	[opaque pointer types] Switch a few cases of getElementType over, since I had them lying around anyway	2015-09-14 20:29:26 +00:00
LTO	Reapply "LTO: Disable extra verify runs in release builds"	2015-09-15 23:05:59 +00:00
MC	MCAsmInfo: Allow targets to specify when the .section directive should be omitted	2015-09-25 21:41:14 +00:00
Object	Prune trailing whitespaces.	2015-09-22 11:19:03 +00:00
Option	Add an ArgList::AddAllArgs that accepts a vector of OptSpecifier.	2015-07-29 17:34:41 +00:00
Passes	[PM] Port SROA to the new pass manager.	2015-09-12 09:09:14 +00:00
ProfileData	InstrProf: Don't call std::unique twice here	2015-09-30 02:02:08 +00:00
Support	[BranchProbability] Manually round the floating point output.	2015-09-26 10:09:36 +00:00
TableGen	TableGen: Support folding casts from bits to int	2015-07-31 01:12:06 +00:00
Target	AMDGPU: Make SIInsertWaits about a factor of 4 faster	2015-10-01 21:43:15 +00:00
Transforms	[InstCombine] Remove trivially empty lifetime start/end ranges.	2015-10-01 14:54:31 +00:00
CMakeLists.txt
LLVMBuild.txt
Makefile