Searched refs:predication (Results 1 – 25 of 114) sorted by relevance
12345
/external/llvm-project/llvm/docs/Proposals/ |
D | VectorPredication.rst | 12 This proposal defines a roadmap towards native vector predication in LLVM, 20 implementation to achieve native vector predication in LLVM. The VP prototype 76 - InstCombine/InstSimplify expect predication in regular Instructions (Stage (3) 79 Result: Native vector predication in IR. 87 .. [VPRFC] RFC: Prototype & Roadmap for vector predication in LLVM,
|
/external/llvm-project/llvm/test/Transforms/LoopVectorize/ARM/ |
D | tail-folding-reductions-allowed.ll | 1 ; RUN: opt -mtriple=thumbv8.1m.main-arm-eabihf -mattr=+mve.fp -loop-vectorize -tail-predication=ena… 4 ; RUN: opt -mtriple=thumbv8.1m.main-arm-eabihf -mattr=+mve.fp -loop-vectorize -tail-predication=ena… 7 ; RUN: opt -mtriple=thumbv8.1m.main-arm-eabihf -mattr=+mve.fp -loop-vectorize -tail-predication=for…
|
D | prefer-tail-loop-folding.ll | 3 ; RUN: -tail-predication=enabled -loop-vectorize -S < %s | \ 7 ; RUN: -tail-predication=enabled -loop-vectorize \ 13 ; RUN: -tail-predication=enabled -loop-vectorize \ 19 ; RUN: -tail-predication=disabled -loop-vectorize \ 28 ; RUN: -tail-predication=enabled -loop-vectorize \ 34 ; RUN: -tail-predication=enabled -loop-vectorize \ 41 ; RUN: -tail-predication=enabled -loop-vectorize \ 48 ; RUN: -tail-predication=enabled -loop-vectorize \
|
D | tail-folding-loop-hint.ll | 1 ; RUN: opt -mtriple=thumbv8.1m.main-arm-eabihf -mattr=+mve.fp -loop-vectorize -tail-predication=ena… 5 ; this test case, the TTI hook rejects tail-predication:
|
D | tail-folding-scalar-epilogue-fallback.ll | 2 ; RUN: opt -S -loop-vectorize -mattr=+armv8.1-m.main,+mve.fp -tail-predication=disabled< %s | FileC… 3 ; RUN: opt -S -loop-vectorize -mattr=+armv8.1-m.main,+mve.fp -tail-predication=enabled < %s | FileC…
|
D | tail-folding-prefer-flag.ll | 1 ; RUN: opt -mtriple=thumbv8.1m.main-arm-eabihf -mattr=+mve.fp -loop-vectorize -tail-predication=ena… 4 ; RUN: opt -mtriple=thumbv8.1m.main-arm-eabihf -mattr=+mve.fp -loop-vectorize -tail-predication=ena…
|
D | tail-folding-counting-down.ll | 2 ; RUN: opt < %s -loop-vectorize -tail-predication=enabled -prefer-predicate-over-epilogue=predicate… 3 ; RUN: opt < %s -loop-vectorize -tail-predication=enabled -prefer-predicate-over-epilogue=predicate… 4 ; RUN: opt < %s -loop-vectorize -tail-predication=enabled -S | FileCheck %s --check-prefixes=COMMON… 304 ; TODO: if tail-predication is requested, tail-folding isn't triggered because 344 ; With -disable-mve-tail-predication=false, the target hook returns
|
/external/llvm-project/llvm/test/Transforms/LoopPredication/ |
D | unswitch-exit-loop.ll | 1 ; RUN: opt < %s -loop-predication -S | FileCheck %s 3 ;; This is a simplified copy of @unswitch_exit_form test that should trigger loop-predication
|
D | profitability.ll | 2 ; RUN: opt -S -loop-predication -loop-predication-skip-profitability-checks=false < %s 2>&1 | FileC… 3 …N: opt -S -loop-predication-skip-profitability-checks=false -passes='require<scalar-evolution>,req…
|
D | reverse.ll | 2 ; RUN: opt -S -loop-predication -loop-predication-enable-count-down-loop=true < %s 2>&1 | FileCheck… 3 ; RUN: opt -S -passes='require<scalar-evolution>,loop(loop-predication)' -loop-predication-enable-c…
|
D | visited.ll | 2 ; RUN: opt -S -loop-predication < %s 2>&1 | FileCheck %s 3 ; RUN: opt -S -passes='require<scalar-evolution>,loop(loop-predication)' < %s 2>&1 | FileCheck %s
|
/external/llvm-project/llvm/test/CodeGen/Thumb2/LowOverheadLoops/ |
D | tail-pred-narrow.ll | 1 ; RUN: opt -mtriple=thumbv8.1m.main -mve-tail-predication -tail-predication=enabled -mattr=+mve,+lo…
|
D | varying-outer-2d-reduction.ll | 3 ; RUN: llc -mtriple=thumbv8.1m.main -mattr=+mve -tail-predication=enabled %s -o - | \ 6 ; Forcing tail-predication should not be necessary here, so we check the same 8 ; RUN: llc -mtriple=thumbv8.1m.main -mattr=+mve -tail-predication=force-enabled %s -o - | \ 11 ; RUN: llc -mtriple=thumbv8.1m.main -mattr=+mve -tail-predication=enabled-no-reductions %s -o - | \ 14 ; RUN: llc -mtriple=thumbv8.1m.main -mattr=+mve -tail-predication=force-enabled-no-reductions %s -o…
|
D | vector-reduce-mve-tail.ll | 2 ; RUN: opt -mtriple=thumbv8.1m.main -mve-tail-predication -tail-predication=enabled -mattr=+mve %s …
|
D | tail-pred-disabled-in-loloops.ll | 3 ; RUN: llc -mtriple=thumbv8.1m.main -mattr=+mve -tail-predication=enabled \ 6 ; RUN: llc -mtriple=thumbv8.1m.main -mattr=+mve -tail-predication=enabled \
|
D | predicated-liveout-unknown-lanes.ll | 2 ; RUN: llc -mtriple=thumbv8.1m.main-none-none-eabi -mattr=+mve.fp -O3 -tail-predication=force-enabl…
|
D | tail-pred-intrinsic-fabs.ll | 2 …mbv8.1m.main-none-none-eabi -mattr=+mve.fp -verify-machineinstrs -tail-predication=enabled -o - %s…
|
D | lsr-profitable-chain.ll | 2 ; RUN: llc -O3 -tail-predication=enabled -mtriple=thumbv8.1m.main -mattr=+mve,+mve.fp %s -o - | Fil…
|
D | vector-unroll.ll | 1 ; RUN: opt -mtriple=thumbv8.1m.main -mve-tail-predication -tail-predication=enabled -mattr=+mve,+lo…
|
/external/llvm-project/llvm/test/CodeGen/Hexagon/ |
D | expand-condsets-def-undef.mir | 6 # during predication. In the code below, %2.isub_hi is invalidated
|
/external/mesa3d/docs/relnotes/ |
D | 19.2.6.rst | 65 - v3d: Fix predication with atomic image operations
|
D | 20.1.9.rst | 65 - intel/fs: Disable sample mask predication for scratch stores
|
/external/llvm-project/llvm/test/Transforms/LoopVectorize/ |
D | tail-folding-counting-down.ll | 4 ; is vectorized with preferred predication.
|
/external/llvm-project/llvm/test/Transforms/LoopVectorize/X86/ |
D | consecutive-ptr-uniforms.ll | 73 ; Check that scalar with predication instructions are not considered uniform 75 ; having a single instance (out of VF). The predication stems from a tiny count
|
/external/llvm-project/llvm/test/CodeGen/Thumb2/ |
D | mve-gather-tailpred.ll | 2 …bv8.1m.main-none-none-eabi -mattr=+mve.fp -enable-arm-maskedldst -tail-predication=force-enabled %…
|
12345