Simon Pilgrim 5c91aa6603 [InstCombine] Fold or(zext(bswap(x)),shl(zext(bswap(y)),bw/2)) -> bswap(or(zext(x),shl(zext(y), bw/2)) ...

This adds a general combine that can be used to fold:

  or(zext(OP(x)), shl(zext(OP(y)),bw/2))
-->
  OP(or(zext(x), shl(zext(y),bw/2)))

Allowing us to widen 'concat-able' style or+zext patterns - I've just set this up for BSWAP but we could use this for other similar ops (BITREVERSE for instance).

We already do something similar for bitop(bswap(x),bswap(y)) --> bswap(bitop(x,y))

Fixes PR45715

Reviewed By: @lebedev.ri

Differential Revision: https://reviews.llvm.org/D79041

2020-05-05 12:30:10 +01:00

..

CMakeLists.txt

[InstCombine] Negator - sink sinkable negations

2020-04-21 22:00:23 +03:00

InstCombineAddSub.cpp

[InstCombine] Negator - sink sinkable negations

2020-04-21 22:00:23 +03:00

InstCombineAndOrXor.cpp

[InstCombine] Fold or(zext(bswap(x)),shl(zext(bswap(y)),bw/2)) -> bswap(or(zext(x),shl(zext(y), bw/2))

2020-05-05 12:30:10 +01:00

InstCombineAtomicRMW.cpp

[NFC] Modernize misc. uses of Align/MaybeAlign APIs.

2020-04-06 17:53:04 -07:00

InstCombineCalls.cpp

[NFC] Rename *ByValOrInalloca* to *PassPointeeByValue*

2020-04-30 09:42:13 -07:00

InstCombineCasts.cpp

Upgrade users of 'new ShuffleVectorInst' to pass indices as an int array

2020-04-15 14:29:43 +02:00

InstCombineCompares.cpp

[InstCombine] generalize canonicalization of masked equality comparisons

2020-04-25 11:31:57 -04:00

InstCombineInternal.h

[InstCombine] Negator: 'or' with no common bits set is just 'add'

2020-04-28 19:16:32 +03:00

InstCombineLoadStoreAlloca.cpp

[SVE] Remove calls to isScalable from Transforms

2020-04-23 13:50:07 -07:00

InstCombineMulDivRem.cpp

[InstCombine] Fold (mul(abs(x),abs(x))) -> (mul(x,x)) (PR39476)

2020-05-04 15:21:52 +01:00

InstCombineNegator.cpp

[InstCombine] Negator: 'or' with no common bits set is just 'add'

2020-04-28 19:16:32 +03:00

InstCombinePHI.cpp

[InstCombine] Use replaceOperand() in non zero phi transform

2020-03-31 21:54:21 +02:00

InstCombineSelect.cpp

[InstCombine] use select-of-constants with set/clear bit mask patterns

2020-05-03 09:44:43 -04:00

InstCombineShifts.cpp

[SVE] Update API ConstantVector::getSplat() to use ElementCount.

2020-03-12 13:22:41 -07:00

InstCombineSimplifyDemanded.cpp

Upgrade calls to CreateShuffleVector to use the preferred form of passing an array of ints

2020-04-15 12:51:38 +02:00

InstCombineTables.td

…

InstCombineVectorOps.cpp

Upgrade users of 'new ShuffleVectorInst' to pass indices as an int array

2020-04-15 14:29:43 +02:00

InstructionCombining.cpp

[SVE] Fix invalid usage of VectorType::getNumElements() in InstCombine

2020-05-01 10:56:29 -07:00

LLVMBuild.txt

…