Justin Holewinski a1535e3b9b [NVPTX] Honor alignment on vector loads/stores ...

We were not considering the stated alignment on vector loads/stores,
leading us to generate vector instructions even when we do not have
sufficient alignment.

Now, for IR like:

  %1 = load <4 x float>, <4 x float>* %ptr, align 4

we will generate correct, conservative PTX like:

  ld.f32 ... [%ptr]
  ld.f32 ... [%ptr+4]
  ld.f32 ... [%ptr+8]
  ld.f32 ... [%ptr+12]

Or if we have an alignment of 8 (for example), we can
generate code like:

  ld.v2.f32 ... [%ptr]
  ld.v2.f32 ... [%ptr+8]

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@213186 91177308-0d34-0410-b5e6-96231b3b80d8

2014-07-16 19:45:35 +00:00

..

InstPrinter

[NVPTX] Rename registers %fl -> %fd and %rl -> %rd

2014-07-16 16:26:58 +00:00

MCTargetDesc

…

TargetInfo

…

cl_common_defines.h

…

CMakeLists.txt

…

LLVMBuild.txt

NVPTX/LLVMBuild.txt: Add "Scalar" to required_libraries. It is really referenced.

2014-07-14 02:52:19 +00:00

Makefile

…

ManagedStringPool.h

…

NVPTX.h

…

NVPTX.td

[NVPTX] Add target options for PTX 3.2/4.0 and SM 5.0 (Maxwell)

2014-06-27 18:35:18 +00:00

NVPTXAllocaHoisting.cpp

…

NVPTXAllocaHoisting.h

…

NVPTXAsmPrinter.cpp

[NVPTX] Rename registers %fl -> %fd and %rl -> %rd

2014-07-16 16:26:58 +00:00

NVPTXAsmPrinter.h

…

NVPTXAssignValidGlobalNames.cpp

…

NVPTXFavorNonGenericAddrSpaces.cpp

…

NVPTXFrameLowering.cpp

Move the constructor for NVPTXFrameLowering into the implementation

2014-06-27 02:05:24 +00:00

NVPTXFrameLowering.h

Move the constructor for NVPTXFrameLowering into the implementation

2014-06-27 02:05:24 +00:00

NVPTXGenericToNVVM.cpp

[NVPTX] Do not process samplers in GenericToNVVM

2014-06-27 18:36:02 +00:00

NVPTXImageOptimizer.cpp

…

NVPTXInstrFormats.td

…

NVPTXInstrInfo.cpp

Remove caching of the target machine in NVPTXInstrInfo and

2014-06-27 01:27:08 +00:00

NVPTXInstrInfo.h

Remove caching of the target machine in NVPTXInstrInfo and

2014-06-27 01:27:08 +00:00

NVPTXInstrInfo.td

[NVPTX] Add support for [SHL,SRA,SRL]_PARTS

2014-06-27 18:35:40 +00:00

NVPTXIntrinsics.td

[NVPTX] Fix handling of ldg/ldu intrinsics.

2014-06-27 18:35:51 +00:00

NVPTXISelDAGToDAG.cpp

[NVPTX] Fix handling of ldg/ldu intrinsics.

2014-06-27 18:35:51 +00:00

NVPTXISelDAGToDAG.h

[NVPTX] Fix handling of ldg/ldu intrinsics.

2014-06-27 18:35:51 +00:00

NVPTXISelLowering.cpp

[NVPTX] Honor alignment on vector loads/stores

2014-07-16 19:45:35 +00:00

NVPTXISelLowering.h

[codegen,aarch64] Add a target hook to the code generator to control

2014-07-03 00:23:43 +00:00

NVPTXLowerAggrCopies.cpp

…

NVPTXLowerAggrCopies.h

…

NVPTXMachineFunctionInfo.h

…

NVPTXMCExpr.cpp

…

NVPTXMCExpr.h

…

NVPTXPrologEpilogPass.cpp

…

NVPTXRegisterInfo.cpp

[NVPTX] Rename registers %fl -> %fd and %rl -> %rd

2014-07-16 16:26:58 +00:00

NVPTXRegisterInfo.h

…

NVPTXRegisterInfo.td

[NVPTX] Rename registers %fl -> %fd and %rl -> %rd

2014-07-16 16:26:58 +00:00

NVPTXReplaceImageHandles.cpp

…

NVPTXSection.h

…

NVPTXSubtarget.cpp

[NVPTX] Add target options for PTX 3.2/4.0 and SM 5.0 (Maxwell)

2014-06-27 18:35:18 +00:00

NVPTXSubtarget.h

[NVPTX] Update sub-target feature detection

2014-06-27 18:35:16 +00:00

NVPTXTargetMachine.cpp

[NVPTX] Directly control the Machine SSA passes that are invoked for NVPTX.

2014-06-27 18:35:14 +00:00

NVPTXTargetMachine.h

Move NVPTX subtarget dependent variables from the target machine

2014-06-27 04:33:14 +00:00

NVPTXTargetObjectFile.h

CodeGen: Stick constant pool entries in COMDAT sections for WinCOFF

2014-07-14 22:57:27 +00:00

NVPTXutil.cpp

…

NVPTXutil.h

…

NVPTXUtilities.cpp

…

NVPTXUtilities.h

…

NVPTXVector.td

…

NVVMReflect.cpp

[NVPTX] Add reflect intrinsic (better than matching by function name)

2014-06-27 18:36:11 +00:00