Simon Pilgrim 89ca3e72ca

[CostModel][X86] Reduce worst case v8i16/v16i8 SSE2 shuffle costs (#124789) ...

These were based off instruction count, not throughput - we can probably improve these further, but these throughput numbers match the worse expanded shuffles we see in the vector-shuffle-128-v* codegen tests.

2025-01-29 10:23:09 +00:00

..

abs-codesize.ll

…

abs-latency.ll

…

abs-sizelatency.ll

[CostModel][X86] Adjust ABS scalar SizeLatency cost to 3uops

2024-06-11 10:29:18 +01:00

abs.ll

…

aggregates.ll

…

alternate-shuffle-cost.ll

[CostModel][X86] Attempt to match cheap v4f32 shuffles that map to SHUFPS instruction (#121778)

2025-01-06 17:54:36 +00:00

arith-fix.ll

[CostModel][X86] Tweak SSE2 v2i64 multiply costs based off D46276 script

2023-06-14 11:06:15 +01:00

arith-fma.ll

…

arith-fp-codesize.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

arith-fp-latency.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

arith-fp-sizelatency.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

arith-fp.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

arith-int-codesize.ll

[X86] Lower vXi8 multiplies using PMADDUBSW on SSSE3+ targets (#95690)

2024-06-25 12:25:56 +01:00

arith-int-latency.ll

[X86] Lower vXi8 multiplies using PMADDUBSW on SSSE3+ targets (#95690)

2024-06-25 12:25:56 +01:00

arith-int-sizelatency.ll

[X86] Lower vXi8 multiplies using PMADDUBSW on SSSE3+ targets (#95690)

2024-06-25 12:25:56 +01:00

arith-int.ll

[X86] Lower vXi8 multiplies using PMADDUBSW on SSSE3+ targets (#95690)

2024-06-25 12:25:56 +01:00

arith-overflow.ll

[CostModel][X86] Add missing AVX512 vector mul overflow intrinsic costs

2024-07-29 15:53:42 +01:00

arith-sminmax-codesize.ll

…

arith-sminmax-latency.ll

…

arith-sminmax-sizelatency.ll

…

arith-sminmax.ll

…

arith-ssat-codesize.ll

[CostModel][X86] Add add/sat sat intrinsic costs

2024-07-25 14:35:39 +01:00

arith-ssat-latency.ll

[CostModel][X86] Add add/sat sat intrinsic costs

2024-07-25 14:35:39 +01:00

arith-ssat-sizelatency.ll

[CostModel][X86] Add add/sat sat intrinsic costs

2024-07-25 14:35:39 +01:00

arith-ssat.ll

[CostModel][X86] Add add/sat sat intrinsic costs

2024-07-25 14:35:39 +01:00

arith-uminmax-codesize.ll

…

arith-uminmax-latency.ll

…

arith-uminmax-sizelatency.ll

…

arith-uminmax.ll

…

arith-usat-codesize.ll

[CostModel][X86] Add add/sat sat intrinsic costs

2024-07-25 14:35:39 +01:00

arith-usat-latency.ll

[CostModel][X86] Add add/sat sat intrinsic costs

2024-07-25 14:35:39 +01:00

arith-usat-sizelatency.ll

[CostModel][X86] Add add/sat sat intrinsic costs

2024-07-25 14:35:39 +01:00

arith-usat.ll

[CostModel][X86] Add add/sat sat intrinsic costs

2024-07-25 14:35:39 +01:00

bitreverse-codesize.ll

[CostModel][X86] Update BITREVERSE costs for GFNI targets

2024-04-17 15:46:38 +01:00

bitreverse-latency.ll

[CostModel][X86] Update BITREVERSE costs for GFNI targets

2024-04-17 15:46:38 +01:00

bitreverse-sizelatency.ll

[CostModel][X86] Update BITREVERSE costs for GFNI targets

2024-04-17 15:46:38 +01:00

bitreverse.ll

[CostModel][X86] Update BITREVERSE costs for GFNI targets

2024-04-17 15:46:38 +01:00

bswap-codesize.ll

[CostModel][X86] Add explicit Silvermont test coverage for bswap costs

2024-01-02 11:21:32 +00:00

bswap-latency.ll

[CostModel][X86] Update SSSE3/AVX1 BSWAP costs

2024-01-02 12:06:48 +00:00

bswap-sizelatency.ll

[CostModel][X86] Add explicit Silvermont test coverage for bswap costs

2024-01-02 11:21:32 +00:00

bswap-store.ll

[CostModel] Convert some tests to opaque pointers (NFC)

2022-12-15 09:50:34 +01:00

bswap.ll

[CostModel][X86] Update SSSE3/AVX1 BSWAP costs

2024-01-02 12:06:48 +00:00

cast.ll

[CostModel][X86] Update AVX1 sext v4i1 -> v4i64 cost based off worst case llvm-mca numbers

2024-04-04 17:17:55 +01:00

costmodel.ll

[CostModel][X86] Improve add/sub/mul overflow intrinsic costs

2024-07-25 16:01:05 +01:00

ctlz-codesize.ll

[CostModel][X86] Update baseline CTTZ/CTLZ costs for x86_64 (#124312)

2025-01-26 14:43:51 +00:00

ctlz-latency.ll

[CostModel][X86] Add missing costkinds for scalar CTLZ/CTTZ instructions

2024-08-20 15:26:04 +01:00

ctlz-sizelatency.ll

[CostModel][X86] Update baseline CTTZ/CTLZ costs for x86_64 (#124312)

2025-01-26 14:43:51 +00:00

ctlz.ll

[CostModel][X86] Update baseline CTTZ/CTLZ costs for x86_64 (#124312)

2025-01-26 14:43:51 +00:00

ctpop-codesize.ll

[CostModel][X86] Fix expanded CTPOP i8 costs

2024-02-21 14:54:50 +00:00

ctpop-latency.ll

…

ctpop-sizelatency.ll

[CostModel][X86] Fix expanded CTPOP i8 costs

2024-02-21 14:54:50 +00:00

ctpop.ll

…

cttz-codesize.ll

[CostModel][X86] Update baseline CTTZ/CTLZ costs for x86_64 (#124312)

2025-01-26 14:43:51 +00:00

cttz-latency.ll

[CostModel][X86] Add missing costkinds for scalar CTLZ/CTTZ instructions

2024-08-20 15:26:04 +01:00

cttz-sizelatency.ll

[CostModel][X86] Update baseline CTTZ/CTLZ costs for x86_64 (#124312)

2025-01-26 14:43:51 +00:00

cttz.ll

[CostModel][X86] Update baseline CTTZ/CTLZ costs for x86_64 (#124312)

2025-01-26 14:43:51 +00:00

div-codesize.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

div-latency.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

div-sizelatency.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

div.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

extend.ll

[CostModel][X86] Update AVX1 sext v4i1 -> v4i64 cost based off worst case llvm-mca numbers

2024-04-04 17:17:55 +01:00

fcmp-codesize.ll

…

fcmp-latency.ll

…

fcmp-sizelatency.ll

…

fcmp.ll

…

fmaxnum-codesize.ll

[CostModel][X86] Add latency/code-size/size-latency target costs for minnum/maxnum intrinsics

2023-04-13 18:07:03 +01:00

fmaxnum-latency.ll

[CostModel][X86] Add latency/code-size/size-latency target costs for minnum/maxnum intrinsics

2023-04-13 18:07:03 +01:00

fmaxnum-size-latency.ll

[CostModel][X86] Add latency/code-size/size-latency target costs for minnum/maxnum intrinsics

2023-04-13 18:07:03 +01:00

fmaxnum.ll

[CostModel][X86] Add latency/code-size/size-latency target costs for minnum/maxnum intrinsics

2023-04-13 18:07:03 +01:00

fminnum-codesize.ll

[CostModel][X86] Add latency/code-size/size-latency target costs for minnum/maxnum intrinsics

2023-04-13 18:07:03 +01:00

fminnum-latency.ll

[CostModel][X86] Add latency/code-size/size-latency target costs for minnum/maxnum intrinsics

2023-04-13 18:07:03 +01:00

fminnum-size-latency.ll

[CostModel][X86] Add latency/code-size/size-latency target costs for minnum/maxnum intrinsics

2023-04-13 18:07:03 +01:00

fminnum.ll

[CostModel][X86] Add latency/code-size/size-latency target costs for minnum/maxnum intrinsics

2023-04-13 18:07:03 +01:00

fptoi_sat.ll

[X86][FP16][BF16] Improve vectorization of fcmp (#116153)

2024-11-26 08:17:48 +08:00

fptosi.ll

[X86] Rewrite getScalarizationOverhead()

2022-11-15 21:07:12 +03:00

fptoui.ll

[X86] Rewrite getScalarizationOverhead()

2022-11-15 21:07:12 +03:00

free-intrinsics.ll

[DebugInfo][RemoveDIs] Final final test-maintenence patch (#80988)

2024-02-07 14:31:52 +00:00

fround.ll

[CostModel][X86] fround.ll - add <2 x float> test coverage

2023-09-27 17:28:50 +01:00

fshl-codesize.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

fshl-latency.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

fshl-sizelatency.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

fshl.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

fshr-codesize.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

fshr-latency.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

fshr-sizelatency.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

fshr.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

gep.ll

[CostModel] Convert some tests to opaque pointers (NFC)

2022-12-15 09:50:34 +01:00

i32.ll

…

icmp0.ll

[CostModel][X86] Reduce the extra costs for ICMP complex predicates when an operand is constant

2024-02-21 16:19:39 +00:00

icmp-codesize.ll

[TTI] Add cost model support for [u|s]cmp (#106824)

2024-08-31 21:25:04 +08:00

icmp-latency.ll

[TTI] Add cost model support for [u|s]cmp (#106824)

2024-08-31 21:25:04 +08:00

icmp-sizelatency.ll

[TTI] Add cost model support for [u|s]cmp (#106824)

2024-08-31 21:25:04 +08:00

icmp.ll

[TTI] Add cost model support for [u|s]cmp (#106824)

2024-08-31 21:25:04 +08:00

insert-extract-at-zero-inseltpoison.ll

…

insert-extract-at-zero.ll

…

intrinsic-cost-kinds.ll

[CostModel][X86] Update baseline CTTZ/CTLZ costs for x86_64 (#124312)

2025-01-26 14:43:51 +00:00

intrinsic-cost.ll

[CostModel] Convert some tests to opaque pointers (NFC)

2022-12-15 09:50:34 +01:00

lit.local.cfg

[NFC][Py Reformat] Reformat lit.local.cfg python files in llvm

2023-05-17 17:03:15 +02:00

load_store.ll

[CostModel][X86] getMemoryOpCost - increase cost of sub-32-bit vector load/stores

2023-04-23 21:48:25 +01:00

load-broadcast.ll

[CostModel][X86] Broadcast shuffles can be free if they are from a one-use load

2024-04-23 11:11:15 +01:00

load-bswap.ll

[CostModel] Convert some tests to opaque pointers (NFC)

2022-12-15 09:50:34 +01:00

load-to-trunc.ll

[opt] Infer DataLayout from triple if not specified

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

logicalop.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

loop_v2-inseltpoison.ll

[SLP]Fix cost of the broadcast buildvector/gather.

2023-01-06 09:25:05 -08:00

loop_v2.ll

[SLP]Fix cost of the broadcast buildvector/gather.

2023-01-06 09:25:05 -08:00

masked-intrinsic-codesize.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

masked-intrinsic-cost-inseltpoison.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

masked-intrinsic-cost.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

masked-intrinsic-latency.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

masked-intrinsic-sizelatency.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

min-legal-vector-width.ll

[opt] Infer DataLayout from triple if not specified

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

mul32.ll

…

mul64.ll

[CostModel][X86] Tweak SSE2 v2i64 multiply costs based off D46276 script

2023-06-14 11:06:15 +01:00

mul-codesize.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

mul-latency.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

mul-sizelatency.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

mul.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

powi.ll

[X86] Rewrite getScalarizationOverhead()

2022-11-15 21:07:12 +03:00

reduce-add.ll

…

reduce-and.ll

…

reduce-fadd.ll

…

reduce-fmax.ll

[CostModel][X86] Add latency/code-size/size-latency target costs for minnum/maxnum intrinsics

2023-04-13 18:07:03 +01:00

reduce-fmin.ll

[CostModel][X86] Add latency/code-size/size-latency target costs for minnum/maxnum intrinsics

2023-04-13 18:07:03 +01:00

reduce-fmul.ll

…

reduce-mul.ll

[CostModel][X86] Tweak SSE2 v2i64 multiply costs based off D46276 script

2023-06-14 11:06:15 +01:00

reduce-or.ll

…

reduce-smax.ll

[TTI][X86] getMinMaxCost - use existing integer min/max intrinsic cost values instead of maintaining a duplicate cost table

2023-04-12 15:33:12 +01:00

reduce-smin.ll

[TTI][X86] getMinMaxCost - use existing integer min/max intrinsic cost values instead of maintaining a duplicate cost table

2023-04-12 15:33:12 +01:00

reduce-umax.ll

[TTI][X86] getMinMaxCost - use existing integer min/max intrinsic cost values instead of maintaining a duplicate cost table

2023-04-12 15:33:12 +01:00

reduce-umin.ll

[TTI][X86] getMinMaxCost - use existing integer min/max intrinsic cost values instead of maintaining a duplicate cost table

2023-04-12 15:33:12 +01:00

reduce-xor.ll

[CostModel][X86] Reduce worst case v8i16/v16i8 SSE2 shuffle costs (#124789)

2025-01-29 10:23:09 +00:00

reduction.ll

[CostModel][X86] Reduce worst case v8i16/v16i8 SSE2 shuffle costs (#124789)

2025-01-29 10:23:09 +00:00

rem-codesize.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

rem-latency.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

rem-sizelatency.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

rem.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

scalarize.ll

[test] Change llc -march= to -mtriple=

2024-12-15 13:08:02 -08:00

select-codesize.ll

…

select-latency.ll

…

select-sizelatency.ll

…

select.ll

…

shuffle-broadcast-codesize.ll

[CostModel][X86] getShuffleCost - add fallback (to half vector) for bfloat vector shuffle costs

2023-10-05 11:12:40 +01:00

shuffle-broadcast-fp16-codesize.ll

[CostModel][X86] Add CostKinds test coverage for shufflevector instruction

2022-11-23 10:30:48 +08:00

shuffle-broadcast-fp16-latency.ll

[CostModel][X86] Add CostKinds test coverage for shufflevector instruction

2022-11-23 10:30:48 +08:00

shuffle-broadcast-fp16-sizelatency.ll

[CostModel][X86] Add CostKinds test coverage for shufflevector instruction

2022-11-23 10:30:48 +08:00

shuffle-broadcast-fp16.ll

…

shuffle-broadcast-latency.ll

[CostModel][X86] getShuffleCost - add fallback (to half vector) for bfloat vector shuffle costs

2023-10-05 11:12:40 +01:00

shuffle-broadcast-sizelatency.ll

[CostModel][X86] getShuffleCost - add fallback (to half vector) for bfloat vector shuffle costs

2023-10-05 11:12:40 +01:00

shuffle-broadcast.ll

[CostModel][X86] getShuffleCost - add fallback (to half vector) for bfloat vector shuffle costs

2023-10-05 11:12:40 +01:00

shuffle-concat_subvector-codesize.ll

[CostModel][X86] getShuffleCost - use processShuffleMasks to split SK_PermuteTwoSrc shuffles to legal types (#120599)

2024-12-20 10:39:45 +00:00

shuffle-concat_subvector-latency.ll

[CostModel][X86] getShuffleCost - use processShuffleMasks to split SK_PermuteTwoSrc shuffles to legal types (#120599)

2024-12-20 10:39:45 +00:00

shuffle-concat_subvector-sizelatency.ll

[CostModel][X86] getShuffleCost - use processShuffleMasks to split SK_PermuteTwoSrc shuffles to legal types (#120599)

2024-12-20 10:39:45 +00:00

shuffle-concat_subvector.ll

[CostModel][X86] getShuffleCost - use processShuffleMasks to split SK_PermuteTwoSrc shuffles to legal types (#120599)

2024-12-20 10:39:45 +00:00

shuffle-extract_subvector-codesize.ll

[CostModel][X86] Add initial costs for non-lane-crossing one/two input shuffles (#114680)

2024-11-04 10:19:02 +00:00

shuffle-extract_subvector-latency.ll

[CostModel][X86] Add initial costs for non-lane-crossing one/two input shuffles (#114680)

2024-11-04 10:19:02 +00:00

shuffle-extract_subvector-sizelatency.ll

[CostModel][X86] Add initial costs for non-lane-crossing one/two input shuffles (#114680)

2024-11-04 10:19:02 +00:00

shuffle-extract_subvector.ll

[CostModel][X86] Reduce worst case v8i16/v16i8 SSE2 shuffle costs (#124789)

2025-01-29 10:23:09 +00:00

shuffle-insert_subvector-codesize.ll

[CostModel][X86] Attempt to match cheap v4f32 shuffles that map to SHUFPS instruction (#121778)

2025-01-06 17:54:36 +00:00

shuffle-insert_subvector-latency.ll

[CostModel][X86] Attempt to match cheap v4f32 shuffles that map to SHUFPS instruction (#121778)

2025-01-06 17:54:36 +00:00

shuffle-insert_subvector-sizelatency.ll

[CostModel][X86] Attempt to match cheap v4f32 shuffles that map to SHUFPS instruction (#121778)

2025-01-06 17:54:36 +00:00

shuffle-insert_subvector.ll

[CostModel][X86] Reduce worst case v8i16/v16i8 SSE2 shuffle costs (#124789)

2025-01-29 10:23:09 +00:00

shuffle-load-codesize.ll

[CostModel][X86] Add CostKinds test coverage for shufflevector instruction

2022-11-23 10:30:48 +08:00

shuffle-load-latency.ll

[CostModel][X86] Add CostKinds test coverage for shufflevector instruction

2022-11-23 10:30:48 +08:00

shuffle-load-sizelatency.ll

[CostModel][X86] Add CostKinds test coverage for shufflevector instruction

2022-11-23 10:30:48 +08:00

shuffle-load.ll

[CostModel][X86] getMemoryOpCost - increase cost of sub-32-bit vector load/stores

2023-04-23 21:48:25 +01:00

shuffle-non-pow-2-codesize.ll

[CostModel[X86] Update shuffle non-pow-2 tests to not analyse shuffle(undef,undef)

2024-12-16 18:32:13 +00:00

shuffle-non-pow-2-latency.ll

[CostModel[X86] Update shuffle non-pow-2 tests to not analyse shuffle(undef,undef)

2024-12-16 18:32:13 +00:00

shuffle-non-pow-2-sizelatency.ll

[CostModel[X86] Update shuffle non-pow-2 tests to not analyse shuffle(undef,undef)

2024-12-16 18:32:13 +00:00

shuffle-non-pow-2.ll

[CostModel[X86] Update shuffle non-pow-2 tests to not analyse shuffle(undef,undef)

2024-12-16 18:32:13 +00:00

shuffle-replication-i1-codesize.ll

[CostModel][X86] getCastInstrCost - improve CostKind adjustment when splitting src/dst types

2024-05-03 12:17:18 +01:00

shuffle-replication-i1-latency.ll

[CostModel][X86] getCastInstrCost - improve CostKind adjustment when splitting src/dst types

2024-05-03 12:17:18 +01:00

shuffle-replication-i1-sizelatency.ll

[CostModel][X86] getCastInstrCost - improve CostKind adjustment when splitting src/dst types

2024-05-03 12:17:18 +01:00

shuffle-replication-i1.ll

[NFC][Costmodel][X86] Replication shuffle: AVX512F can promote i1 to i32.

2022-12-13 21:21:07 +03:00

shuffle-replication-i8-codesize.ll

[CostModel][X86] getCastInstrCost - improve CostKind adjustment when splitting src/dst types

2024-05-03 12:17:18 +01:00

shuffle-replication-i8-latency.ll

[CostModel][X86] getCastInstrCost - improve CostKind adjustment when splitting src/dst types

2024-05-03 12:17:18 +01:00

shuffle-replication-i8-sizelatency.ll

[CostModel][X86] getCastInstrCost - improve CostKind adjustment when splitting src/dst types

2024-05-03 12:17:18 +01:00

shuffle-replication-i8.ll

[X86] Rewrite getScalarizationOverhead()

2022-11-15 21:07:12 +03:00

shuffle-replication-i16-codesize.ll

[CostModel][X86] getCastInstrCost - improve CostKind adjustment when splitting src/dst types

2024-05-03 12:17:18 +01:00

shuffle-replication-i16-latency.ll

[CostModel][X86] getCastInstrCost - improve CostKind adjustment when splitting src/dst types

2024-05-03 12:17:18 +01:00

shuffle-replication-i16-sizelatency.ll

[CostModel][X86] getCastInstrCost - improve CostKind adjustment when splitting src/dst types

2024-05-03 12:17:18 +01:00

shuffle-replication-i16.ll

[X86] Rewrite getScalarizationOverhead()

2022-11-15 21:07:12 +03:00

shuffle-replication-i32-codesize.ll

[CostModel][X86] Add CostKinds test coverage for shufflevector instruction

2022-11-23 10:30:48 +08:00

shuffle-replication-i32-latency.ll

[CostModel][X86] Fix permute latency cost

2022-11-23 19:17:16 +08:00

shuffle-replication-i32-sizelatency.ll

[CostModel][X86] Add CostKinds test coverage for shufflevector instruction

2022-11-23 10:30:48 +08:00

shuffle-replication-i32.ll

[X86] Rewrite getScalarizationOverhead()

2022-11-15 21:07:12 +03:00

shuffle-replication-i64-codesize.ll

[CostModel][X86] Add CostKinds test coverage for shufflevector instruction

2022-11-23 10:30:48 +08:00

shuffle-replication-i64-latency.ll

[CostModel][X86] Fix permute latency cost

2022-11-23 19:17:16 +08:00

shuffle-replication-i64-sizelatency.ll

[CostModel][X86] Add CostKinds test coverage for shufflevector instruction

2022-11-23 10:30:48 +08:00

shuffle-replication-i64.ll

[X86] Rewrite getScalarizationOverhead()

2022-11-15 21:07:12 +03:00

shuffle-reverse-codesize.ll

[CostModel][X86] Add CostKinds test coverage for shufflevector instruction

2022-11-23 10:30:48 +08:00

shuffle-reverse-fp16-codesize.ll

[CostModel][X86] Add CostKinds test coverage for shufflevector instruction

2022-11-23 10:30:48 +08:00

shuffle-reverse-fp16-latency.ll

[CostModel][X86] Add CostKinds test coverage for shufflevector instruction

2022-11-23 10:30:48 +08:00

shuffle-reverse-fp16-sizelatency.ll

[CostModel][X86] Add CostKinds test coverage for shufflevector instruction

2022-11-23 10:30:48 +08:00

shuffle-reverse-fp16.ll

…

shuffle-reverse-latency.ll

[CostModel][X86] Fix permute latency cost

2022-11-23 19:17:16 +08:00

shuffle-reverse-sizelatency.ll

[CostModel][X86] Add CostKinds test coverage for shufflevector instruction

2022-11-23 10:30:48 +08:00

shuffle-reverse.ll

…

shuffle-select-codesize.ll

[CostModel][X86] getShuffleCost - use processShuffleMasks for all shuffle kinds to legal types (#120599) (#121760)

2025-01-06 13:32:55 +00:00

shuffle-select-latency.ll

[CostModel][X86] getShuffleCost - use processShuffleMasks for all shuffle kinds to legal types (#120599) (#121760)

2025-01-06 13:32:55 +00:00

shuffle-select-sizelatency.ll

[CostModel][X86] getShuffleCost - use processShuffleMasks for all shuffle kinds to legal types (#120599) (#121760)

2025-01-06 13:32:55 +00:00

shuffle-select.ll

[CostModel][X86] getShuffleCost - use processShuffleMasks for all shuffle kinds to legal types (#120599) (#121760)

2025-01-06 13:32:55 +00:00

shuffle-single-src-codesize.ll

[IR] Change shufflevector undef mask to poison

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

shuffle-single-src-fp16-codesize.ll

[CostModel][X86] Add CostKinds test coverage for shufflevector instruction

2022-11-23 10:30:48 +08:00

shuffle-single-src-fp16-latency.ll

[CostModel][X86] Add CostKinds test coverage for shufflevector instruction

2022-11-23 10:30:48 +08:00

shuffle-single-src-fp16-sizelatency.ll

[CostModel][X86] Add CostKinds test coverage for shufflevector instruction

2022-11-23 10:30:48 +08:00

shuffle-single-src-fp16.ll

…

shuffle-single-src-latency.ll

[SLP]Add cost estimation for gather node reshuffling

2024-12-24 15:35:29 -05:00

shuffle-single-src-sizelatency.ll

[IR] Change shufflevector undef mask to poison

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

shuffle-single-src.ll

[CostModel][X86] Reduce worst case v8i16/v16i8 SSE2 shuffle costs (#124789)

2025-01-29 10:23:09 +00:00

shuffle-splat-codesize.ll

[SLP]Add cost estimation for gather node reshuffling

2024-12-24 15:35:29 -05:00

shuffle-splat-latency.ll

[SLP]Add cost estimation for gather node reshuffling

2024-12-24 15:35:29 -05:00

shuffle-splat-sizelatency.ll

[SLP]Add cost estimation for gather node reshuffling

2024-12-24 15:35:29 -05:00

shuffle-splat.ll

[SLP]Add cost estimation for gather node reshuffling

2024-12-24 15:35:29 -05:00

shuffle-splice-codesize.ll

[CostModel][X86] Attempt to match cheap v4f32 shuffles that map to SHUFPS instruction (#121778)

2025-01-06 17:54:36 +00:00

shuffle-splice-latency.ll

[CostModel][X86] Attempt to match cheap v4f32 shuffles that map to SHUFPS instruction (#121778)

2025-01-06 17:54:36 +00:00

shuffle-splice-sizelatency.ll

[CostModel][X86] Attempt to match cheap v4f32 shuffles that map to SHUFPS instruction (#121778)

2025-01-06 17:54:36 +00:00

shuffle-splice.ll

[CostModel][X86] Attempt to match cheap v4f32 shuffles that map to SHUFPS instruction (#121778)

2025-01-06 17:54:36 +00:00

shuffle-transpose-codesize.ll

[TTI][X86] getShuffleCosts - for SK_PermuteTwoSrc, if the masks are known to be "inlane" no need to scale the costs by worst-case legalization (#117999)

2024-12-01 12:01:47 +00:00

shuffle-transpose-latency.ll

[TTI][X86] getShuffleCosts - for SK_PermuteTwoSrc, if the masks are known to be "inlane" no need to scale the costs by worst-case legalization (#117999)

2024-12-01 12:01:47 +00:00

shuffle-transpose-sizelatency.ll

[TTI][X86] getShuffleCosts - for SK_PermuteTwoSrc, if the masks are known to be "inlane" no need to scale the costs by worst-case legalization (#117999)

2024-12-01 12:01:47 +00:00

shuffle-transpose.ll

[CostModel][X86] Reduce worst case v8i16/v16i8 SSE2 shuffle costs (#124789)

2025-01-29 10:23:09 +00:00

shuffle-two-src-codesize.ll

[CostModel][X86] Attempt to match v4f32 shuffles that map to MOVSS/INSERTPS instruction

2025-01-07 11:31:44 +00:00

shuffle-two-src-fp16-codesize.ll

[CostModel][X86] getShuffleCost - use processShuffleMasks to split SK_PermuteTwoSrc shuffles to legal types (#120599)

2024-12-20 10:39:45 +00:00

shuffle-two-src-fp16-latency.ll

[CostModel][X86] getShuffleCost - use processShuffleMasks to split SK_PermuteTwoSrc shuffles to legal types (#120599)

2024-12-20 10:39:45 +00:00

shuffle-two-src-fp16-sizelatency.ll

[CostModel][X86] getShuffleCost - use processShuffleMasks to split SK_PermuteTwoSrc shuffles to legal types (#120599)

2024-12-20 10:39:45 +00:00

shuffle-two-src-fp16.ll

[CostModel][X86] getShuffleCost - use processShuffleMasks to split SK_PermuteTwoSrc shuffles to legal types (#120599)

2024-12-20 10:39:45 +00:00

shuffle-two-src-latency.ll

[CostModel][X86] Attempt to match v4f32 shuffles that map to MOVSS/INSERTPS instruction

2025-01-07 11:31:44 +00:00

shuffle-two-src-sizelatency.ll

[CostModel][X86] Attempt to match v4f32 shuffles that map to MOVSS/INSERTPS instruction

2025-01-07 11:31:44 +00:00

shuffle-two-src.ll

[CostModel][X86] Reduce worst case v8i16/v16i8 SSE2 shuffle costs (#124789)

2025-01-29 10:23:09 +00:00

sitofp.ll

[X86] Rewrite getScalarizationOverhead()

2022-11-15 21:07:12 +03:00

size-cost.ll

[opt] Infer DataLayout from triple if not specified

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

slm-arith-costs.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

sse-itoi.ll

[CostModel] Convert some tests to opaque pointers (NFC)

2022-12-15 09:50:34 +01:00

tiny.ll

…

trunc-codesize.ll

[CostModel][X86] getCastInstrCost - improve CostKind adjustment when splitting src/dst types

2024-05-03 12:17:18 +01:00

trunc-latency.ll

[CostModel][X86] getCastInstrCost - improve CostKind adjustment when splitting src/dst types

2024-05-03 12:17:18 +01:00

trunc-sizelatency.ll

[CostModel][X86] getCastInstrCost - improve CostKind adjustment when splitting src/dst types

2024-05-03 12:17:18 +01:00

trunc.ll

[opt] Infer DataLayout from triple if not specified

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

uitofp.ll

…

uniformshift-inseltpoison.ll

…

uniformshift.ll

…

vdiv-cost.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

vector_gep-inseltpoison.ll

[CostModel] Convert some tests to opaque pointers (NFC)

2022-12-15 09:50:34 +01:00

vector_gep.ll

[CostModel] Convert some tests to opaque pointers (NFC)

2022-12-15 09:50:34 +01:00

vector-extract.ll

[CostModel][X86] getMemoryOpCost - increase cost of sub-32-bit vector load/stores

2023-04-23 21:48:25 +01:00

vector-insert-inseltpoison.ll

[CostModel][X86] Reduce worst case v8i16/v16i8 SSE2 shuffle costs (#124789)

2025-01-29 10:23:09 +00:00

vector-insert-value.ll

[CostModel][X86] Reduce worst case v8i16/v16i8 SSE2 shuffle costs (#124789)

2025-01-29 10:23:09 +00:00

vector-insert.ll

[CostModel][X86] Reduce worst case v8i16/v16i8 SSE2 shuffle costs (#124789)

2025-01-29 10:23:09 +00:00

vectorized-loop.ll

[CostModel] Convert some tests to opaque pointers (NFC)

2022-12-15 09:57:59 +01:00

vshift-ashr-codesize.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

vshift-ashr-cost-inseltpoison.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

vshift-ashr-cost.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

vshift-ashr-latency.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

vshift-ashr-sizelatency.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

vshift-lshr-codesize.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

vshift-lshr-cost-inseltpoison.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

vshift-lshr-cost.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

vshift-lshr-latency.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

vshift-lshr-sizelatency.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

vshift-shl-codesize.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

vshift-shl-cost-inseltpoison.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

vshift-shl-cost.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

vshift-shl-latency.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00

vshift-shl-sizelatency.ll

[LLVM][IR] Use splat syntax when printing Constant[Data]Vector. (#112548)

2024-11-06 11:53:33 +00:00