Chandler Carruth 7f583a4201 [x86] Dramatically improve v8i16 shuffle lowering by not using its ... terribly complex partial blend logic. This code path was one of the more complex and bug prone when it first went in and it hasn't faired much better. Ultimately, with the simpler basis for unpack lowering and support bit-math blending, this is completely obsolete. In the worst case without this we generate different but equivalent instructions. However, in many cases we generate much better code. This is especially true when blends or pshufb is available. This does expose one (minor) weakness of the unpack lowering that I'll try to address. In case you were wondering, this is actually a big part of what I've been trying to pull off in the recent string of commits. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@229853 91177308-0d34-0410-b5e6-96231b3b80d8		2015-02-19 14:08:24 +00:00
..
Analysis	Reformat.	2015-02-18 08:36:14 +00:00
AsmParser	IR: Drop scope from MDTemplateParameter	2015-02-19 00:37:21 +00:00
Bitcode	IR: Drop scope from MDTemplateParameter	2015-02-19 00:37:21 +00:00
CodeGen	[x86,sdag] Two interrelated changes to the x86 and sdag code.	2015-02-19 10:36:19 +00:00
DebugInfo	llvm-pdbdump: Add flags controlling the type of values to dump.	2015-02-15 20:27:53 +00:00
ExecutionEngine	Don't deference the section_end() iterator.	2015-02-17 20:07:28 +00:00
Fuzzer
IR	Implement invoke statepoint verification.	2015-02-19 11:28:47 +00:00
IRReader
LineEditor
Linker	Prefer SmallVector::append/insert over push_back loops.	2015-02-17 15:29:18 +00:00
LTO
MC	Reverting r229831 due to multiple ARM/PPC/MIPS build-bot failures.	2015-02-19 11:38:11 +00:00
Object	Introduce Target::createNullTargetStreamer and use it from IRObjectFile.	2015-02-19 00:45:02 +00:00
Option	Prefer SmallVector::append/insert over push_back loops.	2015-02-17 15:29:18 +00:00
ProfileData	InstrProf: Don't combine expansion regions with code regions	2015-02-18 19:01:06 +00:00
Support	[Support/Timer] Make GetMallocUsage() aware of jemalloc.	2015-02-19 07:27:14 +00:00
TableGen	MSVC 2013 does not ICE on this code in the same fashion that MSVC 2012 did; NFC.	2015-02-16 19:33:36 +00:00
Target	[x86] Dramatically improve v8i16 shuffle lowering by not using its	2015-02-19 14:08:24 +00:00
Transforms	[objc-arc] Introduce the concept of RCIdentity and rename all relevant functions to use that name. NFC.	2015-02-19 00:42:38 +00:00
CMakeLists.txt
LLVMBuild.txt
Makefile