Sanjay Patel bb9333c350 [InstCombine] fold cast of right-shift if high bits are not demanded (2nd try) ...

The 1st try at this was reverted because it caused an infinite loop in instcombine.
That should be fixed after:
1cd6b44f267b

(masked) trunc (lshr X, C) --> (masked) lshr (trunc X), C

Narrowing the shift should be better for analysis and can lead
to follow-on transforms as shown.

Attempt at a general proof in Alive2:
https://alive2.llvm.org/ce/z/tRnnSF

Here are a couple of the specific tests:
https://alive2.llvm.org/ce/z/bCnTp-
https://alive2.llvm.org/ce/z/TfaHnb

Differential Revision: https://reviews.llvm.org/D110170

2021-09-23 09:41:37 -04:00

..

CMakeLists.txt

llvmbuildectomy - replace llvm-build by plain cmake

2020-11-13 10:35:24 +01:00

InstCombineAddSub.cpp

[InstCombine] update code/test comments; NFC

2021-09-11 10:53:53 -04:00

InstCombineAndOrXor.cpp

[InstCombine] allow splat vectors for narrowing masked fold

2021-09-17 11:24:16 -04:00

InstCombineAtomicRMW.cpp

Add missing namespace closure comments. NFCI.

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

InstCombineCalls.cpp

[InstCombine] Move InstCombineWorklist to Utils to allow reuse (NFC).

2021-09-22 08:47:21 +01:00

InstCombineCasts.cpp

[InstCombine] Update InstCombine to use poison instead of undef for shufflevector's placeholder (1/3)

2021-09-22 23:18:51 +09:00

InstCombineCompares.cpp

[InstCombine] Update InstCombine to use poison instead of undef for shufflevector's placeholder (2/3)

2021-09-23 00:14:50 +09:00

InstCombineInternal.h

[InstCombine] Move InstCombineWorklist to Utils to allow reuse (NFC).

2021-09-22 08:47:21 +01:00

InstCombineLoadStoreAlloca.cpp

[IR] Return AAMDNodes from Instruction::getMetadata() (NFC)

2021-09-16 21:06:57 +02:00

InstCombineMulDivRem.cpp

[InstCombine] Move InstCombineWorklist to Utils to allow reuse (NFC).

2021-09-22 08:47:21 +01:00

InstCombineNegator.cpp

[Transforms] Use range-based for loops (NFC)

2021-02-08 22:33:53 -08:00

InstCombinePHI.cpp

[InstCombine] Fold phi ( inttoptr/ptrtoint x ) to phi (x)

2021-08-03 17:52:59 +05:30

InstCombineSelect.cpp

Don't fold (select C, (gep Ptr, Idx), Ptr) if C is vector but Idx is scalar

2021-09-22 18:11:33 +08:00

InstCombineShifts.cpp

[InstCombine] add one-use check to shift-shift transform

2021-09-22 16:31:12 -04:00

InstCombineSimplifyDemanded.cpp

[InstCombine] fold cast of right-shift if high bits are not demanded (2nd try)

2021-09-23 09:41:37 -04:00

InstCombineVectorOps.cpp

[NFC][InstCombine] Fix inconsistent comments

2021-09-23 09:31:39 +09:00

InstructionCombining.cpp

[InstSimplify] Don't lose inbounds when simplifying a GEP

2021-09-23 09:25:06 +01:00