shylie/llvm-project
1
0
Fork 0
Code Issues Pull Requests Actions 6 Packages Projects Releases Wiki Activity
llvm-project/llvm/test/Analysis/CostModel/X86
History
Nikita Popov d3b86b79fa Reapply "[CodeGen][X86] Expand USUBSAT to UMAX+SUB, also for vectors" 
Related to https://bugs.llvm.org/show_bug.cgi?id=40123.

Rather than scalarizing, expand a vector USUBSAT into UMAX+SUB,
which produces much better code for X86.

Reapplying with updated SLPVectorizer tests.

Differential Revision: https://reviews.llvm.org/D56636

llvm-svn: 351219
2019-01-15 18:43:41 +00:00
..
alternate-shuffle-cost.ll
[CostModel] Replace ShuffleKind::SK_Alternate with ShuffleKind::SK_Select (PR33744)
2018-06-12 16:12:29 +00:00
arith-fma.ll
…
arith-fp.ll
[CostModel][X86] Fix SSE1 FADD/FSUB costs
2019-01-04 16:55:57 +00:00
arith-ssat.ll
[X86] Add ADD/SUB SSAT/USAT vector costs (PR40123)
2019-01-03 11:38:42 +00:00
arith-usat.ll
Reapply "[CodeGen][X86] Expand USUBSAT to UMAX+SUB, also for vectors"
2019-01-15 18:43:41 +00:00
arith.ll
…
bitreverse.ll
…
bswap.ll
…
cast-widen.ll
[X86] Add cost model tests for fp_to_int/int_to_fp with -x86-experimental-vector-widening-legalization
2018-11-27 19:44:34 +00:00
cast.ll
…
cmp.ll
…
costmodel.ll
…
ctlz.ll
…
ctpop.ll
…
cttz.ll
…
div.ll
[CostModel][X86] Add vXi8 vector division by constants costs.
2018-10-24 18:44:12 +00:00
extend.ll
[X86] Add some cost model entries for sext/zext for avx512bw
2018-11-28 18:11:39 +00:00
fptosi-widen.ll
[X86] Add cost model tests for fp_to_int/int_to_fp with -x86-experimental-vector-widening-legalization
2018-11-27 19:44:34 +00:00
fptosi.ll
…
fptoui-widen.ll
[X86] Add cost model tests for fp_to_int/int_to_fp with -x86-experimental-vector-widening-legalization
2018-11-27 19:44:34 +00:00
fptoui.ll
[X86] Correct the cost of (v4i32 (fptoui (v4f64))) under AVX512F.
2018-08-26 18:47:44 +00:00
fround.ll
[CostModel][X86] Add cost tests for fp rounding intrinsics
2018-07-02 17:07:01 +00:00
fshl.ll
[TTI] getOperandInfo - a broadcast shuffle means the result is OK_UniformValue
2018-11-14 15:04:08 +00:00
fshr.ll
[TTI] getOperandInfo - a broadcast shuffle means the result is OK_UniformValue
2018-11-14 15:04:08 +00:00
gep.ll
…
i32.ll
…
insert-extract-at-zero.ll
…
interleave-load-i32.ll
[X86] Replace '-mcpu=skx' with -mattr=avx512f or -mattr=avx512bw in interleave/strided load/store cost model tests.
2018-12-01 00:21:49 +00:00
interleave-store-i32.ll
[X86] Replace '-mcpu=skx' with -mattr=avx512f or -mattr=avx512bw in interleave/strided load/store cost model tests.
2018-12-01 00:21:49 +00:00
interleaved-load-float.ll
…
interleaved-load-i8.ll
…
interleaved-load-store-double.ll
…
interleaved-load-store-i64.ll
…
interleaved-store-i8.ll
…
intrinsic-cost.ll
…
lit.local.cfg
…
load_store.ll
…
loop_v2.ll
…
masked-intrinsic-cost-widen.ll
[X86] Add cost model test for masked load an store with -x86-experimental-vector-widening-legalization
2018-11-27 19:44:36 +00:00
masked-intrinsic-cost.ll
[CostModel] Add more realistic SK_InsertSubvector generic costs.
2018-11-12 15:20:24 +00:00
min-legal-vector-width.ll
[X86] Make X86TTIImpl::getCastInstrCost properly handle the case where AVX512 is enabled, but 512-bit vectors aren't legal.
2018-11-28 18:11:42 +00:00
reduce-add-widen.ll
[CostModel][X86] Fix overcounting arithmetic cost in illegal types in getArithmeticReductionCost/getMinMaxReductionCost
2018-12-07 18:20:56 +00:00
reduce-add.ll
[CostModel][X86] Fix overcounting arithmetic cost in illegal types in getArithmeticReductionCost/getMinMaxReductionCost
2018-12-07 18:20:56 +00:00
reduce-and-widen.ll
[CostModel][X86] Fix overcounting arithmetic cost in illegal types in getArithmeticReductionCost/getMinMaxReductionCost
2018-12-07 18:20:56 +00:00
reduce-and.ll
[CostModel][X86] Fix overcounting arithmetic cost in illegal types in getArithmeticReductionCost/getMinMaxReductionCost
2018-12-07 18:20:56 +00:00
reduce-mul-widen.ll
[CostModel][X86] Fix overcounting arithmetic cost in illegal types in getArithmeticReductionCost/getMinMaxReductionCost
2018-12-07 18:20:56 +00:00
reduce-mul.ll
[CostModel][X86] Fix overcounting arithmetic cost in illegal types in getArithmeticReductionCost/getMinMaxReductionCost
2018-12-07 18:20:56 +00:00
reduce-or-widen.ll
[CostModel][X86] Fix overcounting arithmetic cost in illegal types in getArithmeticReductionCost/getMinMaxReductionCost
2018-12-07 18:20:56 +00:00
reduce-or.ll
[CostModel][X86] Fix overcounting arithmetic cost in illegal types in getArithmeticReductionCost/getMinMaxReductionCost
2018-12-07 18:20:56 +00:00
reduce-smax-widen.ll
[CostModel][X86][AArch64] Adjust cost of the scalarization part of min/max reduction.
2018-12-10 06:58:58 +00:00
reduce-smax.ll
[CostModel][X86][AArch64] Adjust cost of the scalarization part of min/max reduction.
2018-12-10 06:58:58 +00:00
reduce-smin-widen.ll
[CostModel][X86][AArch64] Adjust cost of the scalarization part of min/max reduction.
2018-12-10 06:58:58 +00:00
reduce-smin.ll
[CostModel][X86][AArch64] Adjust cost of the scalarization part of min/max reduction.
2018-12-10 06:58:58 +00:00
reduce-umax-widen.ll
[CostModel][X86][AArch64] Adjust cost of the scalarization part of min/max reduction.
2018-12-10 06:58:58 +00:00
reduce-umax.ll
[CostModel][X86][AArch64] Adjust cost of the scalarization part of min/max reduction.
2018-12-10 06:58:58 +00:00
reduce-umin-widen.ll
[CostModel][X86][AArch64] Adjust cost of the scalarization part of min/max reduction.
2018-12-10 06:58:58 +00:00
reduce-umin.ll
[CostModel][X86][AArch64] Adjust cost of the scalarization part of min/max reduction.
2018-12-10 06:58:58 +00:00
reduce-xor-widen.ll
[CostModel][X86] Fix overcounting arithmetic cost in illegal types in getArithmeticReductionCost/getMinMaxReductionCost
2018-12-07 18:20:56 +00:00
reduce-xor.ll
[CostModel][X86] Fix overcounting arithmetic cost in illegal types in getArithmeticReductionCost/getMinMaxReductionCost
2018-12-07 18:20:56 +00:00
reduction.ll
[CostModel][X86] Don't count 2 shuffles on the last level of a pairwise arithmetic or min/max reduction
2018-12-13 19:08:10 +00:00
rem.ll
[CostModel][X86] Add vXi8 vector division by constants costs.
2018-10-24 18:44:12 +00:00
scalarize.ll
…
shuffle-broadcast.ll
[CostModel] Recognise BROADCAST shuffle mask if the elements come from the second src
2018-06-13 16:52:02 +00:00
shuffle-extract_subvector.ll
[CostModel][X86] SK_ExtractSubvector is cheap if the (legal) subvector is aligned within the source vector
2018-11-12 15:48:06 +00:00
shuffle-insert_subvector.ll
[CostModel][X86] Add some initial extract/insert subvector shuffle cost tests
2018-10-20 17:38:33 +00:00
shuffle-reverse.ll
[CostModel] Recognise REVERSE shuffle mask if the elements come from the second src
2018-06-14 09:35:00 +00:00
shuffle-single-src.ll
[X86][AVX] Reduce v4f64/v4i64 shuffle costs (PR37882)
2018-06-21 11:37:13 +00:00
shuffle-transpose.ll
[TTI][X86] Treat SK_Transpose shuffles as SK_PermuteTwoSrc - there's no difference in lowering.
2018-10-23 16:45:26 +00:00
shuffle-two-src.ll
[X86][AVX] Reduce v4f64/v4i64 shuffle costs (PR37882)
2018-06-21 11:37:13 +00:00
sitofp-widen.ll
[X86] Add cost model tests for fp_to_int/int_to_fp with -x86-experimental-vector-widening-legalization
2018-11-27 19:44:34 +00:00
sitofp.ll
…
slm-arith-costs.ll
…
sse-itoi.ll
…
strided-load-i8.ll
[X86] Replace '-mcpu=skx' with -mattr=avx512f or -mattr=avx512bw in interleave/strided load/store cost model tests.
2018-12-01 00:21:49 +00:00
strided-load-i16.ll
[X86] Replace '-mcpu=skx' with -mattr=avx512f or -mattr=avx512bw in interleave/strided load/store cost model tests.
2018-12-01 00:21:49 +00:00
strided-load-i32.ll
[X86] Replace '-mcpu=skx' with -mattr=avx512f or -mattr=avx512bw in interleave/strided load/store cost model tests.
2018-12-01 00:21:49 +00:00
strided-load-i64.ll
[X86] Replace '-mcpu=skx' with -mattr=avx512f or -mattr=avx512bw in interleave/strided load/store cost model tests.
2018-12-01 00:21:49 +00:00
testshiftashr-widen.ll
[X86] Add cost model tests for shifts with -x86-experimental-vector-widening-legalization.
2018-11-27 19:44:30 +00:00
testshiftashr.ll
[X86] Add a combine for back to back VSRAI instructions
2018-11-28 18:03:38 +00:00
testshiftlshr-widen.ll
[X86] Add cost model tests for shifts with -x86-experimental-vector-widening-legalization.
2018-11-27 19:44:30 +00:00
testshiftlshr.ll
[X86] Add -LABEL to some FileCheck checks. NFC
2018-10-26 17:21:19 +00:00
testshiftshl-widen.ll
[X86] Add cost model tests for shifts with -x86-experimental-vector-widening-legalization.
2018-11-27 19:44:30 +00:00
testshiftshl.ll
[X86] Add -LABEL to some FileCheck checks. NFC
2018-10-26 17:21:19 +00:00
tiny.ll
…
trunc.ll
[CostModel][X86] Add truncate cost tests to cover all legal destination types
2019-01-03 14:49:39 +00:00
uitofp-widen.ll
[X86] Add cost model tests for fp_to_int/int_to_fp with -x86-experimental-vector-widening-legalization
2018-11-27 19:44:34 +00:00
uitofp.ll
[CostModel][X86] Add realistic vXi64 uitofp vXf64 costs
2018-10-25 13:06:20 +00:00
uniformshift.ll
…
vdiv-cost.ll
[CostModel][X86] Add vXi8 vector division by constants costs.
2018-10-24 18:44:12 +00:00
vector_gep.ll
…
vector-extract.ll
…
vector-insert.ll
…
vectorized-loop.ll
…
vselect-cost.ll
…
vshift-ashr-cost.ll
[CostModel][X86] Fix constant vector XOP rights shifts
2018-11-13 16:40:10 +00:00
vshift-lshr-cost.ll
[CostModel][X86] Fix constant vector XOP rights shifts
2018-11-13 16:40:10 +00:00
vshift-shl-cost.ll
…