Simon Pilgrim 212b2bbcd1 [VectorCombine][X86] foldShuffleOfCastops - fold shuffle(cast(x),cast(y)) -> cast(shuffle(x,y)) iff cost efficient (#87510) ...

Based off the existing foldShuffleOfBinops fold

Fixes #67803

2024-04-04 11:22:37 +01:00

..

extract-binop-inseltpoison.ll

…

extract-binop.ll

[IR] Change shufflevector undef mask to poison

2023-04-27 14:41:10 +01:00

extract-cmp-binop.ll

[VectorCombine] Use poison insteaf of undef as placeholder [NFC]

2023-07-19 10:29:08 +01:00

extract-cmp.ll

[IR] Change shufflevector undef mask to poison

2023-04-27 14:41:10 +01:00

extract-fneg-insert.ll

…

insert-binop-inseltpoison.ll

…

insert-binop-with-constant-inseltpoison.ll

[opt] Infer DataLayout from triple if not specified

2023-10-26 12:07:37 -07:00

insert-binop-with-constant.ll

[opt] Infer DataLayout from triple if not specified

2023-10-26 12:07:37 -07:00

insert-binop.ll

[NewPM][test] Use -passes syntax in VectorCombine lit tests

2021-10-20 15:16:17 +02:00

lit.local.cfg

…

load-inseltpoison.ll

…

load-widening.ll

[Transforms] Convert tests to opaque pointers (NFC)

2024-02-05 11:57:34 +01:00

load.ll

[IR] Change shufflevector undef mask to poison

2023-04-27 14:41:10 +01:00

no-sse-inseltpoison.ll

…

no-sse.ll

…

reduction-of-truncations.ll

[VectorCombine] Fold reduce(trunc(x)) -> trunc(reduce(x)) iff cost effective (#81852)

2024-02-19 11:32:23 +00:00

reduction-two-vecs-combine.ll

[VectorCombine] Add special handling for truncating shuffles (#70013)

2023-10-24 15:03:43 +01:00

scalarize-cmp-inseltpoison.ll

[opt] Infer DataLayout from triple if not specified

2023-10-26 12:07:37 -07:00

scalarize-cmp.ll

[opt] Infer DataLayout from triple if not specified

2023-10-26 12:07:37 -07:00

scalarize-vector-gep.ll

…

select-shuffle.ll

[IR] Change shufflevector undef mask to poison

2023-04-27 14:41:10 +01:00

shuffle-inseltpoison.ll

[VectorCombine] foldBitcastShuffle - peek through any residual bitcasts before creating a new bitcast on top (#86119)

2024-04-02 10:58:45 +01:00

shuffle-of-casts.ll

[VectorCombine][X86] foldShuffleOfCastops - fold shuffle(cast(x),cast(y)) -> cast(shuffle(x,y)) iff cost efficient (#87510)

2024-04-04 11:22:37 +01:00

shuffle.ll

[VectorCombine] foldBitcastShuffle - peek through any residual bitcasts before creating a new bitcast on top (#86119)

2024-04-02 10:58:45 +01:00