Suyog Sarda 1dea0dc279 This patch recognizes (+ (+ v0, v1) (+ v2, v3)), reorders them for bundling into vector of loads, ... and vectorizes it. Test case : float hadd(float* a) { return (a[0] + a[1]) + (a[2] + a[3]); } AArch64 assembly before patch : ldp s0, s1, [x0] ldp s2, s3, [x0, #8] fadd s0, s0, s1 fadd s1, s2, s3 fadd s0, s0, s1 ret AArch64 assembly after patch : ldp d0, d1, [x0] fadd v0.2s, v0.2s, v1.2s faddp s0, v0.2s ret Reviewed Link : http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20141208/248531.html git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@224119 91177308-0d34-0410-b5e6-96231b3b80d8		2014-12-12 12:53:44 +00:00
..
Analysis	Fix PR21694. r219517 added a use of SCEV divide in HowFarToZero computation. This divide can produce incorrect results as we are using an unsigned divide for what should be a modular divide. This change reverts back to a more conservative computation using trailing zeros.	2014-12-10 22:53:52 +00:00
AsmParser	AsmParser: Don't crash on an ill-formed MDNodeVector	2014-12-11 20:51:54 +00:00
Bitcode	Bitcode: Add missing "Remove in 4.0" comments	2014-12-12 02:11:31 +00:00
CodeGen	[MachineScheduler] Fix for PR21807: minor code difference building with/without -g.	2014-12-12 12:41:22 +00:00
DebugInfo	Make DWARFAcceleratorTable::dump() const.	2014-11-20 16:21:11 +00:00
ExecutionEngine	Move three methods only used by MCJIT to MCJIT.	2014-12-10 20:46:55 +00:00
IR	Document that PassManager::add() may delete the pass right away.	2014-12-12 01:27:01 +00:00
IRReader	Remove unused variable. NFC.	2014-11-06 23:16:57 +00:00
LineEditor
Linker	IR: Split Metadata from Value	2014-12-09 18:38:53 +00:00
LTO	IR: Split Metadata from Value	2014-12-09 18:38:53 +00:00
MC	clang-formatted ranged loops and assignment, NFC.	2014-12-04 08:30:39 +00:00
Object	Move three methods only used by MCJIT to MCJIT.	2014-12-10 20:46:55 +00:00
Option
ProfileData	llvm-cov: Sink some reporting logic into CoverageMapping	2014-11-14 01:50:32 +00:00
Support	Remove a convoluted way of calling close by moving the call to the only caller.	2014-12-11 20:12:55 +00:00
TableGen	Use range-based for loops. NFC	2014-12-11 07:04:54 +00:00
Target	Emit Tag_ABI_FP_16bit_format build attribute.	2014-12-12 11:59:18 +00:00
Transforms	This patch recognizes (+ (+ v0, v1) (+ v2, v3)), reorders them for bundling into vector of loads,	2014-12-12 12:53:44 +00:00
CMakeLists.txt
LLVMBuild.txt
Makefile