Ahmed Bougacha bc47ceef43 [ARM] Don't break alignment when combining base updates into load/stores. ... r223862/r224203 tried to also combine base-updating load/stores. There was a mistake there: the alignment was added as is as an operand to the ARMISD::VLD/VST node. However, the VLD/VST selection logic doesn't care about less-than-standard alignment attributes. For example, no matter the alignment of a v2i64 load (say 1), SelectVLD picks VLD1q64 (because of the memory type). But VLD1q64 ("vld1.64 {dXX, dYY}") is 8-aligned, per ARMARMv7a 3.2.1. For the 1-aligned load, what we really want is VLD1q8. This commit introduces bitcasts if necessary, and changes the vld/vst type to one whose standard alignment matches the original load/store alignment. Differential Revision: http://reviews.llvm.org/D6759 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@224754 91177308-0d34-0410-b5e6-96231b3b80d8		2014-12-23 06:07:31 +00:00
..
Analysis	InstSimplify: Don't bother if getScalarSizeInBits returns zero	2014-12-20 04:45:33 +00:00
AsmParser
Bitcode
CodeGen	Make musttail more robust for vector types on x86	2014-12-22 23:58:37 +00:00
DebugInfo	[DebugInfo] Move all DWARF headers to the public include directory.	2014-12-19 18:26:33 +00:00
ExecutionEngine	[C API] Expose LLVMGetGlobalValueAddress and LLVMGetFunctionAddress.	2014-12-22 18:53:11 +00:00
IR	The leak detector is dead, long live asan and valgrind.	2014-12-22 13:00:36 +00:00
IRReader
LineEditor
Linker	Rename MapValue(Metadata*) to MapMetadata()	2014-12-19 06:06:18 +00:00
LTO
MC	Fix Windows unwind info for functions in sections other than .text	2014-12-22 22:10:08 +00:00
Object	Add printing the LC_ROUTINES load commands with llvm-objdump’s -private-headers.	2014-12-19 22:25:22 +00:00
Option
ProfileData
Support
TableGen
Target	[ARM] Don't break alignment when combining base updates into load/stores.	2014-12-23 06:07:31 +00:00
Transforms	Revert r224739: Debug info: Teach SROA how to update debug info for	2014-12-23 02:58:14 +00:00
CMakeLists.txt
LLVMBuild.txt
Makefile