Home
last modified time | relevance | path

Searched refs:SLP (Results 1 – 25 of 54) sorted by relevance

123

/llvm-project-15.0.7/llvm/test/Transforms/SLPVectorizer/X86/
H A Dopt.ll2 …t < %s -O3 -S -mtriple=x86_64-apple-macosx10.8.0 -mcpu=corei7-avx | FileCheck %s --check-prefix=SLP
11 ; SLP-LABEL: @test1(
12 ; SLP-NEXT: entry:
13 ; SLP-NEXT: [[TMP0:%.*]] = bitcast double* [[A:%.*]] to <2 x double>*
14 ; SLP-NEXT: [[TMP1:%.*]] = load <2 x double>, <2 x double>* [[TMP0]], align 8
15 ; SLP-NEXT: [[TMP2:%.*]] = bitcast double* [[B:%.*]] to <2 x double>*
16 ; SLP-NEXT: [[TMP3:%.*]] = load <2 x double>, <2 x double>* [[TMP2]], align 8
17 ; SLP-NEXT: [[TMP4:%.*]] = fmul <2 x double> [[TMP1]], [[TMP3]]
18 ; SLP-NEXT: [[TMP5:%.*]] = bitcast double* [[C:%.*]] to <2 x double>*
19 ; SLP-NEXT: store <2 x double> [[TMP4]], <2 x double>* [[TMP5]], align 8
[all …]
H A Dremark_unsupported.ll5 ; This type is not supported by SLP
44 ; YAML-NEXT: - String: 'Cannot SLP vectorize list: type '
H A Dextractelement-multiple-uses.ll10 ; YAML: - String: 'SLP vectorized with cost '
H A Dvector_gep.ll7 ; This test checks that SLP vectorizer does not fail on vector GEP.
H A Dremark_alternate.ll10 ; YAML-NEXT: - String: 'SLP vectorized with cost '
H A Dbroadcast_long.ll10 ; YAML-NEXT: - String: 'Stores SLP vectorized with cost '
H A Dremark_extract_broadcast.ll22 ; YAML-NEXT: - String: 'Stores SLP vectorized with cost '
H A Dgep.ll7 ; Test if SLP can handle GEP expressions.
H A Dcatchswitch.ll4 ; This used to crash in SLP vectorization when attempting to set the
H A Dsaxpy.ll7 ; SLP vectorization example from http://cs.stanford.edu/people/eschkufz/research/asplos291-schkufza…
H A Drevectorized_rdx_crash.ll6 ; SLP crashed when tried to delete instruction with uses.
/llvm-project-15.0.7/llvm/test/tools/llvm-lto2/X86/
H A Dslp-vectorize-pm.ll3 ; Test SLP and Loop Vectorization are enabled by default at O2 and O3.
5 ; RUN: -save-temps 2>&1 | FileCheck %s --check-prefix=CHECK-O0-SLP
9 ; RUN: -save-temps 2>&1 | FileCheck %s --check-prefix=CHECK-O1-SLP
13 ; RUN: -save-temps 2>&1 | FileCheck %s --check-prefix=CHECK-O2-SLP
17 ; RUN: -save-temps 2>&1 | FileCheck %s --check-prefix=CHECK-O3-SLP
20 ; CHECK-O0-SLP-NOT: Running pass: SLPVectorizerPass
21 ; CHECK-O1-SLP-NOT: Running pass: SLPVectorizerPass
22 ; CHECK-O2-SLP: Running pass: SLPVectorizerPass
23 ; CHECK-O3-SLP: Running pass: SLPVectorizerPass
/llvm-project-15.0.7/lld/test/ELF/lto/
H A Dslp-vectorize-pm.ll4 ; Test SLP and Loop Vectorization are enabled by default at O2 and O3.
5 …pt=O0 --plugin-opt=save-temps -shared -o %t1.o %t.o 2>&1 | FileCheck %s --check-prefix=CHECK-O0-SLP
8 …pt=O1 --plugin-opt=save-temps -shared -o %t2.o %t.o 2>&1 | FileCheck %s --check-prefix=CHECK-O1-SLP
11 …pt=O2 --plugin-opt=save-temps -shared -o %t3.o %t.o 2>&1 | FileCheck %s --check-prefix=CHECK-O2-SLP
14 …pt=O3 --plugin-opt=save-temps -shared -o %t4.o %t.o 2>&1 | FileCheck %s --check-prefix=CHECK-O3-SLP
17 ; CHECK-O0-SLP-NOT: Running pass: SLPVectorizerPass
18 ; CHECK-O1-SLP-NOT: Running pass: SLPVectorizerPass
19 ; CHECK-O2-SLP: Running pass: SLPVectorizerPass
20 ; CHECK-O3-SLP: Running pass: SLPVectorizerPass
/llvm-project-15.0.7/llvm/test/tools/gold/X86/
H A Dslp-vectorize-pm.ll3 ; Test SLP and Loop Vectorization are enabled by default at O2 and O3.
12 ; RUN: -o %t2.o %t.o 2>&1 | FileCheck %s --check-prefix=CHECK-O0-SLP
23 ; RUN: -o %t3.o %t.o 2>&1 | FileCheck %s --check-prefix=CHECK-O1-SLP
34 ; RUN: -o %t4.o %t.o 2>&1 | FileCheck %s --check-prefix=CHECK-O2-SLP
45 ; RUN: -o %t5.o %t.o 2>&1 | FileCheck %s --check-prefix=CHECK-O3-SLP
48 ; CHECK-O0-SLP-NOT: Running pass: SLPVectorizerPass
49 ; CHECK-O1-SLP-NOT: Running pass: SLPVectorizerPass
50 ; CHECK-O2-SLP: Running pass: SLPVectorizerPass
51 ; CHECK-O3-SLP: Running pass: SLPVectorizerPass
/llvm-project-15.0.7/llvm/test/Transforms/SLPVectorizer/SystemZ/
H A DSLP-cmp-cost-query.ll2 ; RUN: opt -mtriple=systemz-unknown -mcpu=z13 -slp-vectorizer -debug-only=SLP \
5 ; Check that SLP vectorizer gets the right cost difference for a compare
34 ; CHECK: SLP: Adding cost -1 for bundle that starts with %4 = icmp ult i32 %2, %1.
/llvm-project-15.0.7/llvm/test/Transforms/SLPVectorizer/
H A Dslp-umax-rdx-matcher-crash.ll4 ; Given LLVM IR caused associative reduction matching routine crash in SLP.
42 ; Given LLVM IR caused crash in SLP.
H A Dint_sideeffect.ll6 ; SLP vectorization across a @llvm.sideeffect.
/llvm-project-15.0.7/llvm/docs/
H A DVectorizers.rst9 which operates on Loops, and the :ref:`SLP Vectorizer
12 The SLP vectorizer merges multiple scalars that are found in the code into
16 Both the Loop Vectorizer and the SLP Vectorizer are enabled by default.
422 The SLP Vectorizer
428 The goal of SLP vectorization (a.k.a. superword-level parallelism) is
446 The SLP-vectorizer processes the code bottom-up, across basic blocks, in search of scalars to combi…
451 The SLP Vectorizer is enabled by default, but it can be disabled
/llvm-project-15.0.7/llvm/test/Transforms/SLPVectorizer/AArch64/
H A Dmul.ll14 ; If we SLP vectorise this then we end up with something like this because we
30 ; If we don't SLP vectorise but scalarize this we get this instead:
H A Dremarks.ll12 ; CHECK: remark: /tmp/s.c:5:10: Stores SLP vectorized with cost -4 and with tree size 3
/llvm-project-15.0.7/llvm/test/CodeGen/X86/
H A Dchain_order.ll4 ; A test from pifft (after SLP-vectorization) that fails when we drop the chain on newly merged loa…
/llvm-project-15.0.7/llvm/docs/Proposals/
H A DVectorizationPlan.rst64 b. SLP vectorization.
67 VF and UF), mixed vectorization: vectorizing a loop with SLP patterns
211 1. SLP Vectorizer: one can compare the VPlan model with LLVM's existing SLP
/llvm-project-15.0.7/llvm/test/Transforms/SLPVectorizer/WebAssembly/
H A Dno-vectorize-rotate.ll4 ; Regression test for a bug in the SLP vectorizer that was causing
/llvm-project-15.0.7/llvm/test/Other/
H A Dopt-pipeline-vector-passes.ll14 ; SLP does not run at -O1. Loop vectorization runs, but it only
/llvm-project-15.0.7/llvm/test/Transforms/SLPVectorizer/VE/
H A Ddisable_slp.ll4 ; Make sure SLP does not trigger for VE on an appealing set of combinable loads

123