Sanjay Patel 0527c8749b [InstCombine] ease alignment restriction for converting masked load to normal load ...

I think we initially made this fold conservative to be safer, but we do not
need the alignment attribute/metadata limitation because the masked load
intrinsic itself specifies the alignment. A normal vector load is better for
IR transforms and should be no worse in codegen than the masked alternative.
If it is worse for some target, the backend can reverse this transform.

Differential Revision: https://reviews.llvm.org/D88505

2020-09-29 15:26:22 -04:00

..

CMakeLists.txt

[InstCombine] Move target-specific inst combining

2020-07-22 15:59:49 +02:00

InstCombineAddSub.cpp

[InstCombine] Add parentheses in assert to silence GCC warning. NFC.

2020-09-23 09:03:01 +03:00

InstCombineAndOrXor.cpp

[InstCombine] matchRotate - force splat of uniform constant rotation amounts (PR46895)

2020-09-28 15:12:41 +01:00

InstCombineAtomicRMW.cpp

Add missing namespace closure comments. NFCI.

2020-09-23 16:19:25 +01:00

InstCombineCalls.cpp

[InstCombine] ease alignment restriction for converting masked load to normal load

2020-09-29 15:26:22 -04:00

InstCombineCasts.cpp

[InstCombine] visitTrunc - trunc (*shr (trunc A), C) --> trunc(*shr A, C)

2020-09-29 18:27:42 +01:00

InstCombineCompares.cpp

[InstCombine][SVE] Skip scalable type for InstCombiner::getFlippedStrictnessPredicateAndConstant.

2020-09-18 11:26:36 -07:00

InstCombineInternal.h

[InstCombine] erase instructions leading up to unreachable

2020-09-07 10:44:08 -04:00

InstCombineLoadStoreAlloca.cpp

[SVE] Remove calls to VectorType::getNumElements from InstCombine

2020-08-31 12:59:10 -07:00

InstCombineMulDivRem.cpp

[InstCombine] Fold mul of abs intrinsic

2020-09-05 12:37:45 +02:00

InstCombineNegator.cpp

[InstCombine] Negator: freeze is freely negatible if it's operand is negatible

2020-08-23 23:28:19 +03:00

InstCombinePHI.cpp

Add missing namespace closure comments. NFCI.

2020-09-23 16:19:25 +01:00

InstCombineSelect.cpp

Revert "[InstCombine] Canonicalize SPF_ABS to abs intrinc"

2020-09-18 09:38:26 +02:00

InstCombineShifts.cpp

[InstCombine] canonicalize 'not' ops before logical shifts

2020-08-22 09:38:13 -04:00

InstCombineSimplifyDemanded.cpp

[SVE] Remove calls to VectorType::getNumElements from InstCombine

2020-08-31 12:59:10 -07:00

InstCombineVectorOps.cpp

[InstCombine] Simplify code. NFCI.

2020-09-27 19:11:07 +02:00

InstructionCombining.cpp

[InstCombine] use unary shuffle creator to reduce code duplication; NFC

2020-09-21 15:34:24 -04:00

LLVMBuild.txt

…