maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/llvm/test/Transforms/LoopVectorize/runtime-check.ll

86 lines
3.5 KiB
LLVM
Raw Normal View History

Rename getMaximumUnrollFactor -> getMaxInterleaveFactor; also rename option names controlling this variable. "Unroll" is not the appropriate name for this variable. Clang already uses the term "interleave" in pragmas and metadata for this. Differential Revision: http://reviews.llvm.org/D5066 llvm-svn: 217528
2014-09-11 01:58:16 +08:00
; RUN: opt < %s -loop-vectorize -force-vector-interleave=1 -force-vector-width=4 -dce -instcombine -S | FileCheck %s
Add support for memory runtime check. When we can, we calculate array bounds. If the arrays are found to be disjoint then we run the vectorized version of the loop. If they are not, we run the scalar code. llvm-svn: 167608
2012-11-09 15:09:44 +08:00
target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128"
; Make sure we vectorize this loop:
; int foo(float *a, float *b, int n) {
; for (int i=0; i<n; ++i)
; a[i] = b[i] * 3;
; }
[LoopVectorize] Use ReplaceInstWithInst() helper where appropriate. This is mostly an NFC, which increases code readability (instead of saving old terminator, generating new one in front of old, and deleting old, we just call a function). However, it would additionaly copy the debug location from old instruction to replacement, which would help PR23837. llvm-svn: 241197
2015-07-02 06:18:30 +08:00
;CHECK-LABEL: define i32 @foo
LoopVectorizer: Emit memory checks into their own basic block. This separates the check for "too few elements to run the vector loop" from the "memory overlap" check, giving a lot nicer code and allowing to skip the memory checks when we're not going to execute the vector code anyways. We still leave the decision of whether to emit the memory checks as branches or setccs, but it seems to be doing a good job. If ugly code pops up we may want to emit them as separate blocks too. Small speedup on MultiSource/Benchmarks/MallocBench/espresso. Most of this is legwork to allow multiple bypass blocks while updating PHIs, dominators and loop info. llvm-svn: 172902
2013-01-19 21:57:58 +08:00
;CHECK: for.body.preheader:
[LV] Don't bail to MiddleBlock if a runtime check fails, bail to ScalarPH instead We were bailing to two places if our runtime checks failed. If the initial overflow check failed, we'd go to ScalarPH. If any other check failed, we'd go to MiddleBlock. This caused us to have to have an extra PHI per induction and reduction as the vector loop's exit block was not dominated by its latch. There's no need to have this behavior - if we just always go to ScalarPH we can get rid of a bunch of complexity. llvm-svn: 246637
2015-09-02 18:15:39 +08:00
;CHECK: br i1 %cmp.zero, label %scalar.ph, label %vector.memcheck, !dbg [[BODY_LOC:![0-9]+]]
LoopVectorizer: Emit memory checks into their own basic block. This separates the check for "too few elements to run the vector loop" from the "memory overlap" check, giving a lot nicer code and allowing to skip the memory checks when we're not going to execute the vector code anyways. We still leave the decision of whether to emit the memory checks as branches or setccs, but it seems to be doing a good job. If ugly code pops up we may want to emit them as separate blocks too. Small speedup on MultiSource/Benchmarks/MallocBench/espresso. Most of this is legwork to allow multiple bypass blocks while updating PHIs, dominators and loop info. llvm-svn: 172902
2013-01-19 21:57:58 +08:00
;CHECK: vector.memcheck:
[LV] Don't bail to MiddleBlock if a runtime check fails, bail to ScalarPH instead We were bailing to two places if our runtime checks failed. If the initial overflow check failed, we'd go to ScalarPH. If any other check failed, we'd go to MiddleBlock. This caused us to have to have an extra PHI per induction and reduction as the vector loop's exit block was not dominated by its latch. There's no need to have this behavior - if we just always go to ScalarPH we can get rid of a bunch of complexity. llvm-svn: 246637
2015-09-02 18:15:39 +08:00
;CHECK: br i1 %memcheck.conflict, label %scalar.ph, label %vector.ph, !dbg [[BODY_LOC]]
Add support for memory runtime check. When we can, we calculate array bounds. If the arrays are found to be disjoint then we run the vectorized version of the loop. If they are not, we run the scalar code. llvm-svn: 167608
2012-11-09 15:09:44 +08:00
;CHECK: load <4 x float>
define i32 @foo(float* nocapture %a, float* nocapture %b, i32 %n) nounwind uwtable ssp {
entry:
[LoopVectorize] Use ReplaceInstWithInst() helper where appropriate. This is mostly an NFC, which increases code readability (instead of saving old terminator, generating new one in front of old, and deleting old, we just call a function). However, it would additionaly copy the debug location from old instruction to replacement, which would help PR23837. llvm-svn: 241197
2015-07-02 06:18:30 +08:00
%cmp6 = icmp sgt i32 %n, 0, !dbg !6
br i1 %cmp6, label %for.body, label %for.end, !dbg !6
Add support for memory runtime check. When we can, we calculate array bounds. If the arrays are found to be disjoint then we run the vectorized version of the loop. If they are not, we run the scalar code. llvm-svn: 167608
2012-11-09 15:09:44 +08:00
for.body: ; preds = %entry, %for.body
[LoopVectorize] Use ReplaceInstWithInst() helper where appropriate. This is mostly an NFC, which increases code readability (instead of saving old terminator, generating new one in front of old, and deleting old, we just call a function). However, it would additionaly copy the debug location from old instruction to replacement, which would help PR23837. llvm-svn: 241197
2015-07-02 06:18:30 +08:00
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ], !dbg !7
%arrayidx = getelementptr inbounds float, float* %b, i64 %indvars.iv, !dbg !7
%0 = load float, float* %arrayidx, align 4, !dbg !7
%mul = fmul float %0, 3.000000e+00, !dbg !7
%arrayidx2 = getelementptr inbounds float, float* %a, i64 %indvars.iv, !dbg !7
store float %mul, float* %arrayidx2, align 4, !dbg !7
%indvars.iv.next = add i64 %indvars.iv, 1, !dbg !7
%lftr.wideiv = trunc i64 %indvars.iv.next to i32, !dbg !7
%exitcond = icmp eq i32 %lftr.wideiv, %n, !dbg !7
br i1 %exitcond, label %for.end, label %for.body, !dbg !7
Add support for memory runtime check. When we can, we calculate array bounds. If the arrays are found to be disjoint then we run the vectorized version of the loop. If they are not, we run the scalar code. llvm-svn: 167608
2012-11-09 15:09:44 +08:00
for.end: ; preds = %for.body, %entry
[LoopVectorize] Use ReplaceInstWithInst() helper where appropriate. This is mostly an NFC, which increases code readability (instead of saving old terminator, generating new one in front of old, and deleting old, we just call a function). However, it would additionaly copy the debug location from old instruction to replacement, which would help PR23837. llvm-svn: 241197
2015-07-02 06:18:30 +08:00
ret i32 undef, !dbg !8
Add support for memory runtime check. When we can, we calculate array bounds. If the arrays are found to be disjoint then we run the vectorized version of the loop. If they are not, we run the scalar code. llvm-svn: 167608
2012-11-09 15:09:44 +08:00
}
LoopVectorizer: If dependency checks fail try runtime checks When a dependence check fails we can still try to vectorize loops with runtime array bounds checks. This helps linpack to vectorize a loop in dgefa. And we are back to 2x of the scalar performance on a corei7-avx. radar://15339680 llvm-svn: 193853
2013-11-01 11:05:07 +08:00
; Make sure that we try to vectorize loops with a runtime check if the
; dependency check fails.
; CHECK-LABEL: test_runtime_check
; CHECK: <4 x float>
define void @test_runtime_check(float* %a, float %b, i64 %offset, i64 %offset2, i64 %n) {
entry:
br label %for.body
for.body:
%iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]
%ind.sum = add i64 %iv, %offset
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction One of several parallel first steps to remove the target type of pointers, replacing them with a single opaque pointer type. This adds an explicit type parameter to the gep instruction so that when the first parameter becomes an opaque pointer type, the type to gep through is still available to the instructions. * This doesn't modify gep operators, only instructions (operators will be handled separately) * Textual IR changes only. Bitcode (including upgrade) and changing the in-memory representation will be in separate changes. * geps of vectors are transformed as: getelementptr <4 x float*> %x, ... ->getelementptr float, <4 x float*> %x, ... Then, once the opaque pointer type is introduced, this will ultimately look like: getelementptr float, <4 x ptr> %x with the unambiguous interpretation that it is a vector of pointers to float. * address spaces remain on the pointer, not the type: getelementptr float addrspace(1)* %x ->getelementptr float, float addrspace(1)* %x Then, eventually: getelementptr float, ptr addrspace(1) %x Importantly, the massive amount of test case churn has been automated by same crappy python code. I had to manually update a few test cases that wouldn't fit the script's model (r228970,r229196,r229197,r229198). The python script just massages stdin and writes the result to stdout, I then wrapped that in a shell script to handle replacing files, then using the usual find+xargs to migrate all the files. update.py: import fileinput import sys import re ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))") normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))") def conv(match, line): if not match: return line line = match.groups()[0] if len(match.groups()[5]) == 0: line += match.groups()[2] line += match.groups()[3] line += ", " line += match.groups()[1] line += "\n" return line for line in sys.stdin: if line.find("getelementptr ") == line.find("getelementptr inbounds"): if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("): line = conv(re.match(ibrep, line), line) elif line.find("getelementptr ") != line.find("getelementptr ("): line = conv(re.match(normrep, line), line) sys.stdout.write(line) apply.sh: for name in "$@" do python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name" rm -f "$name.tmp" done The actual commands: From llvm/src: find test/ -name *.ll | xargs ./apply.sh From llvm/src/tools/clang: find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}" From llvm/src/tools/polly: find test/ -name *.ll | xargs ./apply.sh After that, check-all (with llvm, clang, clang-tools-extra, lld, compiler-rt, and polly all checked out). The extra 'rm' in the apply.sh script is due to a few files in clang's test suite using interesting unicode stuff that my python script was throwing exceptions on. None of those files needed to be migrated, so it seemed sufficient to ignore those cases. Reviewers: rafael, dexonsmith, grosser Differential Revision: http://reviews.llvm.org/D7636 llvm-svn: 230786
2015-02-28 03:29:02 +08:00
%arr.idx = getelementptr inbounds float, float* %a, i64 %ind.sum
[opaque pointer type] Add textual IR support for explicit type parameter to load instruction Essentially the same as the GEP change in r230786. A similar migration script can be used to update test cases, though a few more test case improvements/changes were required this time around: (r229269-r229278) import fileinput import sys import re pat = re.compile(r"((?:=|:|^)\s*load (?:atomic )?(?:volatile )?(.*?))(| addrspace\(\d+\) *)\*($| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$)") for line in sys.stdin: sys.stdout.write(re.sub(pat, r"\1, \2\3*\4", line)) Reviewers: rafael, dexonsmith, grosser Differential Revision: http://reviews.llvm.org/D7649 llvm-svn: 230794
2015-02-28 05:17:42 +08:00
%l1 = load float, float* %arr.idx, align 4
LoopVectorizer: If dependency checks fail try runtime checks When a dependence check fails we can still try to vectorize loops with runtime array bounds checks. This helps linpack to vectorize a loop in dgefa. And we are back to 2x of the scalar performance on a corei7-avx. radar://15339680 llvm-svn: 193853
2013-11-01 11:05:07 +08:00
%ind.sum2 = add i64 %iv, %offset2
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction One of several parallel first steps to remove the target type of pointers, replacing them with a single opaque pointer type. This adds an explicit type parameter to the gep instruction so that when the first parameter becomes an opaque pointer type, the type to gep through is still available to the instructions. * This doesn't modify gep operators, only instructions (operators will be handled separately) * Textual IR changes only. Bitcode (including upgrade) and changing the in-memory representation will be in separate changes. * geps of vectors are transformed as: getelementptr <4 x float*> %x, ... ->getelementptr float, <4 x float*> %x, ... Then, once the opaque pointer type is introduced, this will ultimately look like: getelementptr float, <4 x ptr> %x with the unambiguous interpretation that it is a vector of pointers to float. * address spaces remain on the pointer, not the type: getelementptr float addrspace(1)* %x ->getelementptr float, float addrspace(1)* %x Then, eventually: getelementptr float, ptr addrspace(1) %x Importantly, the massive amount of test case churn has been automated by same crappy python code. I had to manually update a few test cases that wouldn't fit the script's model (r228970,r229196,r229197,r229198). The python script just massages stdin and writes the result to stdout, I then wrapped that in a shell script to handle replacing files, then using the usual find+xargs to migrate all the files. update.py: import fileinput import sys import re ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))") normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))") def conv(match, line): if not match: return line line = match.groups()[0] if len(match.groups()[5]) == 0: line += match.groups()[2] line += match.groups()[3] line += ", " line += match.groups()[1] line += "\n" return line for line in sys.stdin: if line.find("getelementptr ") == line.find("getelementptr inbounds"): if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("): line = conv(re.match(ibrep, line), line) elif line.find("getelementptr ") != line.find("getelementptr ("): line = conv(re.match(normrep, line), line) sys.stdout.write(line) apply.sh: for name in "$@" do python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name" rm -f "$name.tmp" done The actual commands: From llvm/src: find test/ -name *.ll | xargs ./apply.sh From llvm/src/tools/clang: find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}" From llvm/src/tools/polly: find test/ -name *.ll | xargs ./apply.sh After that, check-all (with llvm, clang, clang-tools-extra, lld, compiler-rt, and polly all checked out). The extra 'rm' in the apply.sh script is due to a few files in clang's test suite using interesting unicode stuff that my python script was throwing exceptions on. None of those files needed to be migrated, so it seemed sufficient to ignore those cases. Reviewers: rafael, dexonsmith, grosser Differential Revision: http://reviews.llvm.org/D7636 llvm-svn: 230786
2015-02-28 03:29:02 +08:00
%arr.idx2 = getelementptr inbounds float, float* %a, i64 %ind.sum2
[opaque pointer type] Add textual IR support for explicit type parameter to load instruction Essentially the same as the GEP change in r230786. A similar migration script can be used to update test cases, though a few more test case improvements/changes were required this time around: (r229269-r229278) import fileinput import sys import re pat = re.compile(r"((?:=|:|^)\s*load (?:atomic )?(?:volatile )?(.*?))(| addrspace\(\d+\) *)\*($| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$)") for line in sys.stdin: sys.stdout.write(re.sub(pat, r"\1, \2\3*\4", line)) Reviewers: rafael, dexonsmith, grosser Differential Revision: http://reviews.llvm.org/D7649 llvm-svn: 230794
2015-02-28 05:17:42 +08:00
%l2 = load float, float* %arr.idx2, align 4
LoopVectorizer: If dependency checks fail try runtime checks When a dependence check fails we can still try to vectorize loops with runtime array bounds checks. This helps linpack to vectorize a loop in dgefa. And we are back to 2x of the scalar performance on a corei7-avx. radar://15339680 llvm-svn: 193853
2013-11-01 11:05:07 +08:00
%m = fmul fast float %b, %l2
%ad = fadd fast float %l1, %m
store float %ad, float* %arr.idx, align 4
%iv.next = add nuw nsw i64 %iv, 1
%exitcond = icmp eq i64 %iv.next, %n
br i1 %exitcond, label %loopexit, label %for.body
loopexit:
ret void
}
[LoopVectorize] Use ReplaceInstWithInst() helper where appropriate. This is mostly an NFC, which increases code readability (instead of saving old terminator, generating new one in front of old, and deleting old, we just call a function). However, it would additionaly copy the debug location from old instruction to replacement, which would help PR23837. llvm-svn: 241197
2015-07-02 06:18:30 +08:00
; CHECK: [[BODY_LOC]] = !DILocation(line: 101, column: 1, scope: !{{.*}})
!llvm.module.flags = !{!0, !1}
[DebugInfo] Fix tests so that each subprogram belongs to a CU. llvm-svn: 265490
2016-04-06 07:37:08 +08:00
!llvm.dbg.cu = !{!9}
[LoopVectorize] Use ReplaceInstWithInst() helper where appropriate. This is mostly an NFC, which increases code readability (instead of saving old terminator, generating new one in front of old, and deleting old, we just call a function). However, it would additionaly copy the debug location from old instruction to replacement, which would help PR23837. llvm-svn: 241197
2015-07-02 06:18:30 +08:00
!0 = !{i32 2, !"Dwarf Version", i32 4}
!1 = !{i32 2, !"Debug Info Version", i32 3}
!2 = !{}
!3 = !DISubroutineType(types: !2)
!4 = !DIFile(filename: "test.cpp", directory: "/tmp")
[PR27284] Reverse the ownership between DICompileUnit and DISubprogram. Currently each Function points to a DISubprogram and DISubprogram has a scope field. For member functions the scope is a DICompositeType. DIScopes point to the DICompileUnit to facilitate type uniquing. Distinct DISubprograms (with isDefinition: true) are not part of the type hierarchy and cannot be uniqued. This change removes the subprograms list from DICompileUnit and instead adds a pointer to the owning compile unit to distinct DISubprograms. This would make it easy for ThinLTO to strip unneeded DISubprograms and their transitively referenced debug info. Motivation ---------- Materializing DISubprograms is currently the most expensive operation when doing a ThinLTO build of clang. We want the DISubprogram to be stored in a separate Bitcode block (or the same block as the function body) so we can avoid having to expensively deserialize all DISubprograms together with the global metadata. If a function has been inlined into another subprogram we need to store a reference the block containing the inlined subprogram. Attached to https://llvm.org/bugs/show_bug.cgi?id=27284 is a python script that updates LLVM IR testcases to the new format. http://reviews.llvm.org/D19034 <rdar://problem/25256815> llvm-svn: 266446
2016-04-15 23:57:41 +08:00
!5 = distinct !DISubprogram(name: "foo", scope: !4, file: !4, line: 99, type: !3, isLocal: false, isDefinition: true, scopeLine: 100, flags: DIFlagPrototyped, isOptimized: false, unit: !9, variables: !2)
[LoopVectorize] Use ReplaceInstWithInst() helper where appropriate. This is mostly an NFC, which increases code readability (instead of saving old terminator, generating new one in front of old, and deleting old, we just call a function). However, it would additionaly copy the debug location from old instruction to replacement, which would help PR23837. llvm-svn: 241197
2015-07-02 06:18:30 +08:00
!6 = !DILocation(line: 100, column: 1, scope: !5)
!7 = !DILocation(line: 101, column: 1, scope: !5)
!8 = !DILocation(line: 102, column: 1, scope: !5)
[DebugInfo] Fix tests so that each subprogram belongs to a CU. llvm-svn: 265490
2016-04-06 07:37:08 +08:00
!9 = distinct !DICompileUnit(language: DW_LANG_C99, producer: "clang",
file: !10,
isOptimized: true, flags: "-O2",
[PR27284] Reverse the ownership between DICompileUnit and DISubprogram. Currently each Function points to a DISubprogram and DISubprogram has a scope field. For member functions the scope is a DICompositeType. DIScopes point to the DICompileUnit to facilitate type uniquing. Distinct DISubprograms (with isDefinition: true) are not part of the type hierarchy and cannot be uniqued. This change removes the subprograms list from DICompileUnit and instead adds a pointer to the owning compile unit to distinct DISubprograms. This would make it easy for ThinLTO to strip unneeded DISubprograms and their transitively referenced debug info. Motivation ---------- Materializing DISubprograms is currently the most expensive operation when doing a ThinLTO build of clang. We want the DISubprogram to be stored in a separate Bitcode block (or the same block as the function body) so we can avoid having to expensively deserialize all DISubprograms together with the global metadata. If a function has been inlined into another subprogram we need to store a reference the block containing the inlined subprogram. Attached to https://llvm.org/bugs/show_bug.cgi?id=27284 is a python script that updates LLVM IR testcases to the new format. http://reviews.llvm.org/D19034 <rdar://problem/25256815> llvm-svn: 266446
2016-04-15 23:57:41 +08:00
splitDebugFilename: "abc.debug", emissionKind: 2)
[DebugInfo] Fix tests so that each subprogram belongs to a CU. llvm-svn: 265490
2016-04-06 07:37:08 +08:00
!10 = !DIFile(filename: "path/to/file", directory: "/path/to/dir")
!11 = !{i32 2, !"Debug Info Version", i32 3}