Sanjoy Das 2d9ad10711 Reland "Relax constraints for reduction vectorization" ...

Change from original commit: move test (that uses an X86 triple) into the X86
subdirectory.

Original description:
Gating vectorizing reductions on *all* fastmath flags seems unnecessary;
`reassoc` should be sufficient.

Reviewers: tvvikram, mkuper, kristof.beyls, sdesmalen, Ayal

Reviewed By: sdesmalen

Subscribers: dcaballe, huntergr, jmolloy, mcrosier, jlebar, bixia, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D57728

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@355889 91177308-0d34-0410-b5e6-96231b3b80d8

2019-03-12 01:31:44 +00:00

..

already-vectorized.ll

[Loop Vectorize] Added a separate metadata

2017-08-20 10:32:41 +00:00

assume.ll

…

avx1.ll

[x86] fix cost model inaccuracy for vector memory ops

2016-03-09 22:23:33 +00:00

avx512.ll

[X86] Add support for passing 'prefer-vector-width' function attribute into X86Subtarget and exposing via X86's getRegisterWidth TTI interface.

2018-01-20 00:26:08 +00:00

consecutive-ptr-cg-bug.ll

[SCEV] Smart range calculation for SCEVUnknown Phis

2018-03-01 06:56:48 +00:00

consecutive-ptr-uniforms.ll

Support arbitrary address space pointers in masked gather/scatter intrinsics.

2017-05-03 12:28:54 +00:00

constant-fold.ll

[LV] Do not create SCEVs on broken IR in emitTransformedIndex. PR39160

2018-10-08 05:46:29 +00:00

constant-vector-operand.ll

…

conversion-cost.ll

…

cost-model.ll

[X86] Enable interleaved memory access by default

2016-10-20 21:04:31 +00:00

float-induction-x86.ll

[SimplifyCFG] use pass options and remove the latesimplifycfg pass

2017-10-28 18:43:07 +00:00

force-ifcvt.ll

Introduce llvm.loop.parallel_accesses and llvm.access.group metadata.

2018-12-20 04:58:07 +00:00

fp32_to_uint32-cost-model.ll

…

fp64_to_uint32-cost-model.ll

Revert "r306473 - re-commit r306336: Enable vectorizer-maximize-bandwidth by default."

2017-07-01 03:24:09 +00:00

fp_to_sint8-cost-model.ll

…

funclet.ll

[LoopVectorize] Add operand bundles to vectorized functions

2016-04-29 07:09:48 +00:00

gather_scatter.ll

[ValueTracking] Look through casts when determining non-nullness

2019-01-26 23:40:35 +00:00

gather-cost.ll

Second attempt at r285517.

2016-10-31 13:17:31 +00:00

gather-vs-interleave.ll

[X86][LoopVectorize] Replace -mcpu=skylake-avx512 with -mattr=avx512f in some tests that failed when experimenting with defaulting to -mprefer-vector-width=256 for skylake-avx512.

2018-12-01 01:38:44 +00:00

gcc-examples.ll

Revert "r306473 - re-commit r306336: Enable vectorizer-maximize-bandwidth by default."

2017-07-01 03:24:09 +00:00

illegal-parallel-loop-uniform-write.ll

[LoopVectorize] auto-generate complete checks; NFC

2018-12-18 22:23:04 +00:00

imprecise-through-phis.ll

[CostModel][X86] Add explicit vector select costs

2019-01-20 13:55:01 +00:00

int128_no_gather.ll

[X86] Remove PCOMMIT instruction support since Intel has deprecated this instruction with no plans to release products with it.

2017-02-08 05:45:39 +00:00

interleaved-accesses-large-gap.ll

[InterleavedAccessAnalysis] Fix integer overflow in insertMember.

2019-03-07 17:50:16 +00:00

interleaving.ll

[X86] enable memory interleaving for X86\SLM arch.

2017-01-25 09:14:48 +00:00

invariant-load-gather.ll

[X86][LoopVectorize] Replace -mcpu=skylake-avx512 with -mattr=avx512f in some tests that failed when experimenting with defaulting to -mprefer-vector-width=256 for skylake-avx512.

2018-12-01 01:38:44 +00:00

invariant-store-vectorization.ll

[X86][LoopVectorize] Replace -mcpu=skylake-avx512 with -mattr=avx512f in some tests that failed when experimenting with defaulting to -mprefer-vector-width=256 for skylake-avx512.

2018-12-01 01:38:44 +00:00

lit.local.cfg

…

masked_load_store.ll

[ValueTracking] More accurate unsigned sub overflow detection

2019-02-28 18:04:20 +00:00

max-mstore.ll

Masked Store in Loop Vectorizer - bugfix

2016-04-26 20:18:04 +00:00

metadata-enable.ll

[LV] Preserve inbounds on created GEPs

2018-05-01 15:35:08 +00:00

min-trip-count-switch.ll

…

mul_slm_16bit.ll

[X86] fixing failed test in commit: r291657

2017-01-11 09:03:11 +00:00

no_fpmath_with_hotness.ll

[DebugInfo] Add DILabel metadata and intrinsic llvm.dbg.label.

2018-05-09 02:40:45 +00:00

no_fpmath.ll

[DebugInfo] Add DILabel metadata and intrinsic llvm.dbg.label.

2018-05-09 02:40:45 +00:00

no-vector.ll

…

optsize.ll

[LV] Avoid vectorizing loops under opt for size that involve SCEV checks

2018-11-02 09:16:12 +00:00

parallel-loops-after-reg2mem.ll

Introduce llvm.loop.parallel_accesses and llvm.access.group metadata.

2018-12-20 04:58:07 +00:00

parallel-loops.ll

Introduce llvm.loop.parallel_accesses and llvm.access.group metadata.

2018-12-20 04:58:07 +00:00

powof2div.ll

…

pr23997.ll

[X86][CET] Changing -fcf-protection behavior to comply with gcc (LLVM part)

2018-05-18 11:58:25 +00:00

pr34438.ll

Introduce llvm.loop.parallel_accesses and llvm.access.group metadata.

2018-12-20 04:58:07 +00:00

pr35432.ll

Revert "[SCEV][NFC] Check NoWrap flags before lexicographical comparison of SCEVs"

2018-08-27 21:41:37 +00:00

pr36524.ll

Revert r325687 (workaround for PR36032).

2018-03-22 22:04:39 +00:00

pr39160.ll

[LV] Move test for r343954 into x86 subdirectory

2018-10-09 22:40:04 +00:00

propagate-metadata.ll

SLPVectorizer: Move propagateMetadata to VectorUtils

2016-06-30 21:17:59 +00:00

ptr-indvar-crash.ll

…

rauw-bug.ll

…

reduction-crash.ll

Revert "r306473 - re-commit r306336: Enable vectorizer-maximize-bandwidth by default."

2017-07-01 03:24:09 +00:00

reduction-fastmath.ll

Reland "Relax constraints for reduction vectorization"

2019-03-12 01:31:44 +00:00

reduction-small-size.ll

[LV] Ignore the cost of values that will not appear in the vectorized loop

2017-12-12 08:57:43 +00:00

redundant-vf2-cost.ll

Move redundant-vf2-cost.ll test to X86 directory

2018-06-15 18:46:03 +00:00

reg-usage-debug.ll

[DebugInfo] Add DILabel metadata and intrinsic llvm.dbg.label.

2018-05-09 02:40:45 +00:00

reg-usage.ll

[LV] Avoid computing the register usage for default VF. NFC

2017-09-13 19:35:45 +00:00

register-assumption.ll

[LoopVectorize] Register cloned assumptions

2016-04-08 16:37:10 +00:00

scatter_crash.ll

Support arbitrary address space pointers in masked gather/scatter intrinsics.

2017-05-03 12:28:54 +00:00

slm-no-vectorize.ll

llvm/test/Transforms/LoopVectorize/X86/slm-no-vectorize.ll: -debug is available in +Asserts.

2017-07-02 14:25:27 +00:00

small-size.ll

[LV] Fold tail by masking to vectorize loops of arbitrary trip count under opt for size

2018-10-18 15:03:15 +00:00

strided_load_cost.ll

Currently isLikelyComplexAddressComputation tries to figure out if the given stride seems to be 'complex' and need some extra cost for address computation handling.

2017-01-05 14:03:41 +00:00

struct-store.ll

…

svml-calls-finite.ll

[TLI] Add mapping for various '__<func>_finite' forms of the math routines to SVML routines

2017-05-12 22:11:26 +00:00

svml-calls.ll

[TargetLibraryInfo] add mappings from LLVM sin/cos intrinsics to SVML calls

2018-06-07 18:21:24 +00:00

tripcount.ll

…

uint64_to_fp64-cost-model.ll

[CostModel][X86] Add realistic vXi64 uitofp vXf64 costs

2018-10-25 13:06:20 +00:00

uniform_load.ll

Loop vectorization with uniform load

2016-04-10 16:53:19 +00:00

uniform-phi.ll

[LV] First order recurrence phis should not be treated as uniform

2018-09-04 22:12:23 +00:00

uniformshift.ll

[LV, X86] Be more optimistic about vectorizing shifts.

2016-08-04 22:48:03 +00:00

unroll_selection.ll

…

unroll-pm.ll

…

unroll-small-loops.ll

…

veclib-calls.ll

[SLPVectorizer] Vectorizing the libm sqrt to llvm's sqrt intrinsic requires nnan

2016-04-06 07:04:53 +00:00

vect.omp.force.ll

Introduce llvm.loop.parallel_accesses and llvm.access.group metadata.

2018-12-20 04:58:07 +00:00

vect.omp.force.small-tc.ll

Refactor setAlreadyUnrolled() and setAlreadyVectorized().

2019-02-11 19:45:44 +00:00

vector_max_bandwidth.ll

Introduce llvm.loop.parallel_accesses and llvm.access.group metadata.

2018-12-20 04:58:07 +00:00

vector_ptr_load_store.ll

Add a flag vectorizer-maximize-bandwidth in loop vectorizer to enable using larger vectorization factor.

2015-11-02 22:53:48 +00:00

vector-scalar-select-cost.ll

[CostModel][X86] Add explicit vector select costs

2019-01-20 13:55:01 +00:00

vectorization-remarks-loopid-dbg.ll

[DebugInfo] Add DILabel metadata and intrinsic llvm.dbg.label.

2018-05-09 02:40:45 +00:00

vectorization-remarks-missed.ll

[Unroll/UnrollAndJam/Vectorizer/Distribute] Add followup loop attributes.

2018-12-12 17:32:52 +00:00

vectorization-remarks-profitable.ll

[DebugInfo] Add DILabel metadata and intrinsic llvm.dbg.label.

2018-05-09 02:40:45 +00:00

vectorization-remarks.ll

[DebugInfo] Add DILabel metadata and intrinsic llvm.dbg.label.

2018-05-09 02:40:45 +00:00

vectorize-only-for-real.ll

[LoopVectorize] Don't vectorize loops when everything will be scalarized

2016-03-30 19:37:08 +00:00

x86_fp80-vector-store.ll

…

x86-interleaved-accesses-masked-group.ll

[LV] Support vectorization of interleave-groups that require an epilog under

2018-10-31 09:57:56 +00:00

x86-pr39099.ll

[IAI,LV] Avoid creating interleave-groups for predicated accesse

2018-10-07 06:57:25 +00:00

x86-predication.ll

[LV] Introduce VPBlendRecipe, VPWidenMemoryInstructionRecipe

2017-11-14 12:09:30 +00:00