[llvm-bugs] [Bug 47295] New: Invalid error on long long literal -9223372036854775808LL as a long long array value
https://bugs.llvm.org/show_bug.cgi?id=47295
Bug ID: 47295
Summary: Invalid error on long long literal
-9223372036854775808LL as a long long array value
Product: clang
Version: 6.0
Hardware: PC
OS: Windows NT
Status: NEW
Severity: normal
Priority: P
Component: C++
Assignee: unassignedclangb...@nondot.org
Reporter: francis.andre.kampb...@orange.fr
CC: blitzrak...@gmail.com, dgre...@apple.com,
erik.pilking...@gmail.com, llvm-bugs@lists.llvm.org,
richard-l...@metafoo.co.uk
Hello
Compiling the sample below;
static const long long jjlonglong = -9223372036854775808LL;
static const long long jjlonglong1 = 0x;
static const long long jjCharData[][41] = {
{2LL, 1LL, -9223372036854775808LL}
};
int main() {
return 0;
}
Clang is producing those messages
x.cpp:2:38: warning: integer literal is too large to be represented in a signed
integer type, interpreting as unsigned [-Wimplicitly-unsigned-literal]
static const long long jjlonglong = -9223372036854775808LL;
^
x.cpp:6:14: warning: integer literal is too large to be represented in a signed
integer type, interpreting as unsigned [-Wimplicitly-unsigned-literal]
{2LL, 1LL, -9223372036854775808LL}
^
x.cpp:6:13: error: constant expression evaluates to 9223372036854775808 which
cannot be narrowed to type 'long long' [-Wc++11-narrowing]
{2LL, 1LL, -9223372036854775808LL}
^~
x.cpp:6:13: note: insert an explicit cast to silence this issue
{2LL, 1LL, -9223372036854775808LL}
^~
static_cast( )
The same long long literal as -9223372036854775808LL should produce the same
warning, be used as initializer of a simple variable or a long long array
member.
--
You are receiving this mail because:
You are on the CC list for the bug.___
llvm-bugs mailing list
llvm-bugs@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 47296] New: [BDCE] does not salvage/invalidate debuginfo when rewriting sext as zext
https://bugs.llvm.org/show_bug.cgi?id=47296
Bug ID: 47296
Summary: [BDCE] does not salvage/invalidate debuginfo when
rewriting sext as zext
Product: libraries
Version: trunk
Hardware: PC
OS: Linux
Status: NEW
Severity: enhancement
Priority: P
Component: Scalar Optimizations
Assignee: unassignedb...@nondot.org
Reporter: bjorn.a.petters...@ericsson.com
CC: llvm-bugs@lists.llvm.org
The patch from https://reviews.llvm.org/D60413 (commit 7bf168390fd05460b1c0df)
teached bit-tracking dead code elimination to rewrite:
%6 = sext i32 %5 to i64, !dbg !26
call void @llvm.dbg.value(metadata i64 %6, metadata !18, metadata
!DIExpression()), !dbg !27
as
%6 = zext i32 %5 to i64, !dbg !26
call void @llvm.dbg.value(metadata i64 %6, metadata !18, metadata
!DIExpression()), !dbg !27
when no uses of %6 depended on the most significant bits.
However, since the metadata references in the dbg.value intrinsics aren't dealt
with in any special way, a debugger would print the wrong value for variable
!18 (showing a zero extended value instead of a sign extended value).
I figure that it would be more correct to invalidate the debug-uses (presenting
an undef value), or somehow magically "salvage" the correct value by some fancy
DIExpression based on %5.
Here is an example with full IR (compile with "opt -bdce"):
;-
; ModuleID = './example.ll'
source_filename = "./example.ll"
; Function Attrs: nounwind uwtable
define i32 @foo(i32 %0, i32 %1, i32* %2, i32* %3) #0 !dbg !7 {
%5 = add nsw i32 %0, 2, !dbg !25
%6 = sext i32 %5 to i64, !dbg !26
call void @llvm.dbg.value(metadata i64 %6, metadata !18, metadata
!DIExpression()), !dbg !27
%7 = icmp sgt i32 %1, 0, !dbg !28
br i1 %7, label %8, label %11, !dbg !29
8:; preds = %4
%9 = or i64 %6, 5, !dbg !30
call void @llvm.dbg.value(metadata i64 %9, metadata !20, metadata
!DIExpression()), !dbg !31
%10 = trunc i64 %9 to i32, !dbg !32
store i32 %10, i32* %2, align 4, !dbg !33, !tbaa !34
br label %14, !dbg !38
11: ; preds = %4
%12 = add nsw i64 %6, 3, !dbg !39
call void @llvm.dbg.value(metadata i64 %12, metadata !23, metadata
!DIExpression()), !dbg !40
%13 = trunc i64 %12 to i32, !dbg !41
store i32 %13, i32* %3, align 4, !dbg !42, !tbaa !34
br label %14
14: ; preds = %11, %8
ret i32 undef, !dbg !43
}
; Function Attrs: nounwind readnone speculatable willreturn
declare void @llvm.dbg.value(metadata, metadata, metadata) #1
attributes #0 = { nounwind uwtable }
attributes #1 = { nounwind readnone speculatable willreturn }
!llvm.dbg.cu = !{!0}
!llvm.module.flags = !{!3, !4, !5}
!llvm.ident = !{!6}
!0 = distinct !DICompileUnit(language: DW_LANG_C99, file: !1, producer: "clang
version 12.0.0 (https://github.com/llvm/llvm-project.git
2b37174b9a5db235e493cb72e4454cc08a1b1791)", isOptimized: true, runtimeVersion:
0, emissionKind: FullDebug, enums: !2, splitDebugInlining: false,
nameTableKind: None)
!1 = !DIFile(filename: "example.c", directory: "/home/ce")
!2 = !{}
!3 = !{i32 7, !"Dwarf Version", i32 4}
!4 = !{i32 2, !"Debug Info Version", i32 3}
!5 = !{i32 1, !"wchar_size", i32 4}
!6 = !{!"clang version 12.0.0 (https://github.com/llvm/llvm-project.git
2b37174b9a5db235e493cb72e4454cc08a1b1791)"}
!7 = distinct !DISubprogram(name: "foo", scope: !8, file: !8, line: 2, type:
!9, scopeLine: 2, flags: DIFlagPrototyped | DIFlagAllCallsDescribed, spFlags:
DISPFlagDefinition | DISPFlagOptimized, unit: !0, retainedNodes: !13)
!8 = !DIFile(filename: "", directory: "/home/ce")
!9 = !DISubroutineType(types: !10)
!10 = !{!11, !11, !11, !12, !12}
!11 = !DIBasicType(name: "int", size: 32, encoding: DW_ATE_signed)
!12 = !DIDerivedType(tag: DW_TAG_pointer_type, baseType: !11, size: 64)
!13 = !{!14, !15, !16, !17, !18, !20, !23}
!14 = !DILocalVariable(name: "X", arg: 1, scope: !7, file: !8, line: 2, type:
!11)
!15 = !DILocalVariable(name: "Y", arg: 2, scope: !7, file: !8, line: 2, type:
!11)
!16 = !DILocalVariable(name: "Z", arg: 3, scope: !7, file: !8, line: 2, type:
!12)
!17 = !DILocalVariable(name: "W", arg: 4, scope: !7, file: !8, line: 2, type:
!12)
!18 = !DILocalVariable(name: "LX", scope: !7, file: !8, line: 3, type: !19)
!19 = !DIBasicType(name: "long int", size: 64, encoding: DW_ATE_signed)
!20 = !DILocalVariable(name: "T", scope: !21, file: !8, line: 5, type: !19)
!21 = distinct !DILexicalBlock(scope: !22, file: !8, line: 4, column: 16)
!22 = distinct !DILexicalBlock(scope: !7, file: !8, line: 4, column: 9)
!23 = !DILocalVariable(name: "T", scope: !24, file: !8, line: 8, type: !19)
!24 = distinct !DILexicalBlock(scope: !22, file: !8, line: 7, column: 12)
!25 = !DILocation(line: 3, column
[llvm-bugs] [Bug 46268] Comparison against fixed-point type leads to crash
https://bugs.llvm.org/show_bug.cgi?id=46268 Bevin Hansson changed: What|Removed |Added Resolution|--- |FIXED Status|CONFIRMED |RESOLVED CC||bevin.hans...@ericsson.com --- Comment #10 from Bevin Hansson --- Fixed in d4408fe17f33. -- You are receiving this mail because: You are on the CC list for the bug.___ llvm-bugs mailing list llvm-bugs@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] Issue 25158 in oss-fuzz: llvm:clang-fuzzer: Stack-overflow in llvm::detail::IEEEFloat::IEEEFloat
Updates: Labels: ClusterFuzz-Verified Status: Verified Comment #1 on issue 25158 by ClusterFuzz-External: llvm:clang-fuzzer: Stack-overflow in llvm::detail::IEEEFloat::IEEEFloat https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=25158#c1 ClusterFuzz testcase 4867275684052992 is verified as fixed in https://oss-fuzz.com/revisions?job=libfuzzer_asan_llvm&range=202008230608:202008240603 If this is incorrect, please file a bug on https://github.com/google/oss-fuzz/issues/new -- You received this message because: 1. You were specifically CC'd on the issue You may adjust your notification preferences at: https://bugs.chromium.org/hosting/settings Reply to this email to add a comment.___ llvm-bugs mailing list llvm-bugs@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 47297] New: Assertion `false && "Pass modifies its input and doesn't report it."' for -jump-threading pass
https://bugs.llvm.org/show_bug.cgi?id=47297 Bug ID: 47297 Summary: Assertion `false && "Pass modifies its input and doesn't report it."' for -jump-threading pass Product: new-bugs Version: trunk Hardware: PC OS: Windows NT Status: NEW Severity: enhancement Priority: P Component: new bugs Assignee: unassignedb...@nondot.org Reporter: david.stenb...@ericsson.com CC: htmldevelo...@gmail.com, llvm-bugs@lists.llvm.org Created attachment 23887 --> https://bugs.llvm.org/attachment.cgi?id=23887&action=edit IR reproducer. LLVM commit: ada2e8ea67393aa8c44fe8e9d46be62df6d1c702 Requires an expensive checks build to reproduce. $ opt -S -jump-threading jump-threading.ll Pass modifies its input and doesn't report it: Jump Threading opt: ../lib/IR/LegacyPassManager.cpp:1593: bool llvm::FPPassManager::runOnFunction(llvm::Function &): Assertion `false && "Pass modifies its input and doesn't report it."' failed. -- You are receiving this mail because: You are on the CC list for the bug.___ llvm-bugs mailing list llvm-bugs@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 47276] Crash when running clang-tidy
https://bugs.llvm.org/show_bug.cgi?id=47276 Eugene Zelenko changed: What|Removed |Added Assignee|unassignedclangbugs@nondot. |dcough...@apple.com |org | CC||dcough...@apple.com, ||eugene.zele...@gmail.com, ||llvm-bugs@lists.llvm.org Product|clang-tools-extra |clang Component|clang-tidy |Static Analyzer -- You are receiving this mail because: You are on the CC list for the bug.___ llvm-bugs mailing list llvm-bugs@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 47123] lld fails to build on targets that need -latomic
https://bugs.llvm.org/show_bug.cgi?id=47123 Hans Wennborg changed: What|Removed |Added CC||h...@chromium.org Resolution|--- |FIXED Status|NEW |RESOLVED --- Comment #5 from Hans Wennborg --- Cherry-picked to 11.x as 1708358fbbf58d4a870be015ef21eeea9adae87b. Thanks! -- You are receiving this mail because: You are on the CC list for the bug.___ llvm-bugs mailing list llvm-bugs@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 46725] [meta] 11.0.0 Release Blockers
https://bugs.llvm.org/show_bug.cgi?id=46725 Bug 46725 depends on bug 47123, which changed state. Bug 47123 Summary: lld fails to build on targets that need -latomic https://bugs.llvm.org/show_bug.cgi?id=47123 What|Removed |Added Status|NEW |RESOLVED Resolution|--- |FIXED -- You are receiving this mail because: You are on the CC list for the bug.___ llvm-bugs mailing list llvm-bugs@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 47222] Merge ca77ab49 into 11.0 branch: Disable use of _ExtInt with '__atomic' builtins
https://bugs.llvm.org/show_bug.cgi?id=47222 Hans Wennborg changed: What|Removed |Added Status|NEW |RESOLVED Resolution|--- |FIXED --- Comment #1 from Hans Wennborg --- Pushed to 11.x as 82e48a579024d0ffbc352702ec0c52b47a6fe691. Thanks! -- You are receiving this mail because: You are on the CC list for the bug.___ llvm-bugs mailing list llvm-bugs@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 47259] PowerPC64: Incorrect codegen for mtsprg[0-3] on powerpc64 breaks FreeBSD powerpc64 kernel
https://bugs.llvm.org/show_bug.cgi?id=47259 Hans Wennborg changed: What|Removed |Added Status|REOPENED|RESOLVED CC||h...@chromium.org Resolution|--- |FIXED --- Comment #3 from Hans Wennborg --- Cherry-picked to 11.x as dcdf2aff02a154f9f862d6f01feff3de1389cc47. Shouldn't this be covered by tests? -- You are receiving this mail because: You are on the CC list for the bug.___ llvm-bugs mailing list llvm-bugs@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 46725] [meta] 11.0.0 Release Blockers
https://bugs.llvm.org/show_bug.cgi?id=46725 Bug 46725 depends on bug 47259, which changed state. Bug 47259 Summary: PowerPC64: Incorrect codegen for mtsprg[0-3] on powerpc64 breaks FreeBSD powerpc64 kernel https://bugs.llvm.org/show_bug.cgi?id=47259 What|Removed |Added Status|REOPENED|RESOLVED Resolution|--- |FIXED -- You are receiving this mail because: You are on the CC list for the bug.___ llvm-bugs mailing list llvm-bugs@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 46725] [meta] 11.0.0 Release Blockers
https://bugs.llvm.org/show_bug.cgi?id=46725 Bug 46725 depends on bug 47284, which changed state. Bug 47284 Summary: [lldb] Merge 1ce07cd614be into 11.0 https://bugs.llvm.org/show_bug.cgi?id=47284 What|Removed |Added Status|NEW |RESOLVED Resolution|--- |FIXED -- You are receiving this mail because: You are on the CC list for the bug.___ llvm-bugs mailing list llvm-bugs@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 47284] [lldb] Merge 1ce07cd614be into 11.0
https://bugs.llvm.org/show_bug.cgi?id=47284 Hans Wennborg changed: What|Removed |Added Status|NEW |RESOLVED Resolution|--- |FIXED CC||h...@chromium.org --- Comment #1 from Hans Wennborg --- Pushed to 11.x as 6406b6fa5ac8192b0861c343509d98368b555d12. -- You are receiving this mail because: You are on the CC list for the bug.___ llvm-bugs mailing list llvm-bugs@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 47293] Merge 2b37174b9a5db235e493cb72e4454cc08a1b179 into 11.0
https://bugs.llvm.org/show_bug.cgi?id=47293 Hans Wennborg changed: What|Removed |Added Resolution|--- |FIXED CC||h...@chromium.org Status|NEW |RESOLVED --- Comment #1 from Hans Wennborg --- Pushed to 11.x as 0c001a171c7d671f0129d69c6a47b159544cdca6. -- You are receiving this mail because: You are on the CC list for the bug.___ llvm-bugs mailing list llvm-bugs@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 46725] [meta] 11.0.0 Release Blockers
https://bugs.llvm.org/show_bug.cgi?id=46725 Bug 46725 depends on bug 47293, which changed state. Bug 47293 Summary: Merge 2b37174b9a5db235e493cb72e4454cc08a1b179 into 11.0 https://bugs.llvm.org/show_bug.cgi?id=47293 What|Removed |Added Status|NEW |RESOLVED Resolution|--- |FIXED -- You are receiving this mail because: You are on the CC list for the bug.___ llvm-bugs mailing list llvm-bugs@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 47294] Merge bf3577ef64c300ba7841a90a4e09e1e305271976 into 11.0
https://bugs.llvm.org/show_bug.cgi?id=47294 Hans Wennborg changed: What|Removed |Added Status|NEW |RESOLVED CC||h...@chromium.org Resolution|--- |FIXED --- Comment #1 from Hans Wennborg --- Pushed to 11.x as e4f4d48665526c6d8001e8101dc539a7f7653aec. -- You are receiving this mail because: You are on the CC list for the bug.___ llvm-bugs mailing list llvm-bugs@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 46725] [meta] 11.0.0 Release Blockers
https://bugs.llvm.org/show_bug.cgi?id=46725 Bug 46725 depends on bug 47294, which changed state. Bug 47294 Summary: Merge bf3577ef64c300ba7841a90a4e09e1e305271976 into 11.0 https://bugs.llvm.org/show_bug.cgi?id=47294 What|Removed |Added Status|NEW |RESOLVED Resolution|--- |FIXED -- You are receiving this mail because: You are on the CC list for the bug.___ llvm-bugs mailing list llvm-bugs@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 47298] New: Failure to optimize strlen of string with known length
https://bugs.llvm.org/show_bug.cgi?id=47298
Bug ID: 47298
Summary: Failure to optimize strlen of string with known length
Product: libraries
Version: trunk
Hardware: PC
OS: Linux
Status: NEW
Severity: enhancement
Priority: P
Component: Support Libraries
Assignee: unassignedb...@nondot.org
Reporter: gabrav...@gmail.com
CC: llvm-bugs@lists.llvm.org
bool f()
{
char a[4];
strcpy(a, "123");
return strlen(a) == 3;
}
This can be optimized to `return true;`. This transformation is done by GCC,
but not by LLVM.
--
You are receiving this mail because:
You are on the CC list for the bug.___
llvm-bugs mailing list
llvm-bugs@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 47299] New: llvm.get.active.lane.mask crashes for vl=64
https://bugs.llvm.org/show_bug.cgi?id=47299
Bug ID: 47299
Summary: llvm.get.active.lane.mask crashes for vl=64
Product: new-bugs
Version: unspecified
Hardware: PC
OS: Linux
Status: NEW
Severity: enhancement
Priority: P
Component: new bugs
Assignee: unassignedb...@nondot.org
Reporter: ajc...@google.com
CC: htmldevelo...@gmail.com, llvm-bugs@lists.llvm.org
Given
declare <16 x i1> @llvm.get.active.lane.mask.v16i1.i64(i64, i64)
define <16 x i1> @create_mask64(i64 %0, i64 %1) {
%3 = call <16 x i1> @llvm.get.active.lane.mask.v16i1.i64(i64 %0, i64 %1)
ret <16 x i1> %3
}
and compiling this for avx512 yields the following code without issues
vpbroadcastqzmm0, rsi
vpbroadcastqzmm1, rdi
vpaddq zmm2, zmm1, zmmword ptr [rip + .LCPI0_0]
vpcmpleuq k0, zmm2, zmm0
vpaddq zmm3, zmm1, zmmword ptr [rip + .LCPI0_1]
vpcmpleuq k1, zmm3, zmm0
kunpckbwk0, k1, k0
vpcmpltuq k1, zmm2, zmm1
vpcmpltuq k2, zmm3, zmm1
kunpckbwk1, k2, k1
kandnw k0, k1, k0
vpmovm2bxmm0, k0
vzeroupper
ret
However, using a longer vector length, like 64, crashes the backend
declare <64 x i1> @llvm.get.active.lane.mask.v64i1.i64(i64, i64)
define <64 x i1> @create_mask64(i64 %0, i64 %1) {
%3 = call <64 x i1> @llvm.get.active.lane.mask.v64i1.i64(i64 %0, i64 %1)
ret <64 x i1> %3
}
PLEASE submit a bug report to https://bugs.llvm.org/ and include the crash
backtrace.
Stack dump:
0. Program arguments: PATH/llvm/llvm-project/build/bin/llc -O3
--x86-asm-syntax=intel -mtriple=x86_64-grtev4-linux-gnu -march=x86-64
-mcpu=skylake-avx512 -mattr=fma,avx512f
1. Running pass 'Function Pass Manager' on module ''.
2. Running pass 'X86 DAG->DAG Instruction Selection' on function
'@create_mask64'
#0 0x5581034306a6 llvm::sys::PrintStackTrace(llvm::raw_ostream&)
PATH/llvm/llvm-project/llvm/lib/Support/Unix/Signals.inc:563:22
#1 0x558103430739 PrintStackTraceSignalHandler(void*)
PATH/llvm/llvm-project/llvm/lib/Support/Unix/Signals.inc:624:1
#2 0x55810342e806 llvm::sys::RunSignalHandlers()
PATH/llvm/llvm-project/llvm/lib/Support/Signals.cpp:70:20
#3 0x5581034300f0 SignalHandler(int)
PATH/llvm/llvm-project/llvm/lib/Support/Unix/Signals.inc:405:1
#4 0x7f956d292110 __restore_rt
(/lib/x86_64-linux-gnu/libpthread.so.0+0x14110)
#5 0x558100bf3d9e llvm::Type::getTypeID() const
PATH/llvm/llvm-project/llvm/include/llvm/IR/Type.h:135:37
#6 0x558100bfa344 llvm::Type::isVectorTy() const
PATH/llvm/llvm-project/llvm/include/llvm/IR/Type.h:232:46
#7 0x55810292fb34 llvm::EVT::isExtendedVector() const
PATH/llvm/llvm-project/llvm/lib/CodeGen/ValueTypes.cpp:75:1
#8 0x55810292fe16 llvm::EVT::isExtendedScalableVector() const
PATH/llvm/llvm-project/llvm/lib/CodeGen/ValueTypes.cpp:114:29
#9 0x558100cc0710 llvm::EVT::isScalableVector() const
PATH/llvm/llvm-project/llvm/include/llvm/CodeGen/ValueTypes.h:163:74
#10 0x5581031faae0 FoldBUILD_VECTOR(llvm::SDLoc const&, llvm::EVT,
llvm::ArrayRef, llvm::SelectionDAG&)
PATH/llvm/llvm-project/llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp:4296:3
#11 0x55810321010b llvm::SelectionDAG::getNode(unsigned int, llvm::SDLoc
const&, llvm::EVT, llvm::ArrayRef, llvm::SDNodeFlags)
PATH/llvm/llvm-project/llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp:7512:56
#12 0x558100d32f32 llvm::SelectionDAG::getBuildVector(llvm::EVT,
llvm::SDLoc const&, llvm::ArrayRef)
PATH/llvm/llvm-project/llvm/include/llvm/CodeGen/SelectionDAG.h:775:50
#13 0x5581031b6d2f
llvm::SelectionDAGBuilder::visitIntrinsicCall(llvm::CallInst const&, unsigned
int)
PATH/llvm/llvm-project/llvm/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp:6911:65
--
You are receiving this mail because:
You are on the CC list for the bug.___
llvm-bugs mailing list
llvm-bugs@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 47300] New: [Loop Idiom Recognize] Degrade performance when arrays are replaced by memset/memcpy
https://bugs.llvm.org/show_bug.cgi?id=47300
Bug ID: 47300
Summary: [Loop Idiom Recognize] Degrade performance when arrays
are replaced by memset/memcpy
Product: libraries
Version: trunk
Hardware: Other
OS: Linux
Status: NEW
Severity: enhancement
Priority: P
Component: Loop Optimizer
Assignee: unassignedb...@nondot.org
Reporter: anhtu...@ca.ibm.com
CC: llvm-bugs@lists.llvm.org
Created attachment 23888
--> https://bugs.llvm.org/attachment.cgi?id=23888&action=edit
The IR file of work.c when LIRP is not run.
If the trip count is "small", LIPR should not replace legal stores with
memset/memcpy. The following testcase can be used to demonstrate how Loop Idiom
Recognize Pass (LIRP) degrades the performance.
1) Compiler work.c to an IR file called work.ll.
2) Call opt with -loop-idiom to produce an IR file containing memset
instruction
opt -basic-aa -loop-idiom -S work.ll -o work.yes.ll
3) Inspect to make sure LIRP did replace store with memset in work.yes.ll, but
not in work.ll
5) Compile the test.c, and link it with the IR from (1) and then the IR from
(2).
clang++ -c test.c
clang++ test.o work.ll -o no
clang++ test.o work.yes.ll -o yes
6) Run both the executables on a quiet machine. On my performance machine,
times spent are:
With memset: **Time elapsed: 1.4215**
Without memset: **Time elapsed: 1.3611**
test.c
```
#include
#include
int work(int A[], int sizeI, int sizeL);
int main() {
int A[3] = {1, 2, 3};
int res = 1;
clock_t begin = clock();
res = work(A, , 3);
clock_t end = clock();
double time_spent = (double)(end - begin) / CLOCKS_PER_SEC;
printf("Time elapsed: %4.4f\n", time_spent);
return res;
}
```
work.c
```
int work(int A[], int sizeI, int arraySize) {
for (int i = 0; i < sizeI; ++i)
for (int j = 0; j < sizeI; ++j)
for (int k = 0; k < arraySize; ++k)
A[k] = 0;
return A[arraySize - 1];
}
```
IR before calling opt with **-loop-idiom**
```
; ModuleID = './work.ll'
source_filename = "work.c"
target datalayout = "e-m:e-i64:64-n32:64"
target triple = "powerpc64le-unknown-linux-gnu"
; Function Attrs: noinline nounwind
define dso_local signext i32 @_Z4workPiii(i32* %A, i32 signext %sizeI, i32
signext %arraySize) #0 {
entry:
%cmp6 = icmp slt i32 0, %sizeI
br i1 %cmp6, label %for.body.preheader, label %for.end12
for.body.preheader: ; preds = %entry
br label %for.body
for.body: ; preds =
%for.body.preheader, %for.inc10
%i.07 = phi i32 [ %inc11, %for.inc10 ], [ 0, %for.body.preheader ]
%cmp23 = icmp slt i32 0, %sizeI
br i1 %cmp23, label %for.body3.preheader, label %for.inc10
for.body3.preheader: ; preds = %for.body
br label %for.body3
for.body3:; preds =
%for.body3.preheader, %for.inc7
%j.04 = phi i32 [ %inc8, %for.inc7 ], [ 0, %for.body3.preheader ]
%cmp51 = icmp slt i32 0, %arraySize
br i1 %cmp51, label %for.body6.preheader, label %for.inc7
for.body6.preheader: ; preds = %for.body3
br label %for.body6
for.body6:; preds =
%for.body6.preheader, %for.body6
%k.02 = phi i32 [ %inc, %for.body6 ], [ 0, %for.body6.preheader ]
%idxprom = sext i32 %k.02 to i64
%arrayidx = getelementptr inbounds i32, i32* %A, i64 %idxprom
store i32 0, i32* %arrayidx, align 4
%inc = add nsw i32 %k.02, 1
%cmp5 = icmp slt i32 %inc, %arraySize
br i1 %cmp5, label %for.body6, label %for.inc7.loopexit
for.inc7.loopexit:; preds = %for.body6
br label %for.inc7
for.inc7: ; preds = %for.inc7.loopexit,
%for.body3
%inc8 = add nsw i32 %j.04, 1
%cmp2 = icmp slt i32 %inc8, %sizeI
br i1 %cmp2, label %for.body3, label %for.inc10.loopexit
for.inc10.loopexit: ; preds = %for.inc7
br label %for.inc10
for.inc10:; preds =
%for.inc10.loopexit, %for.body
%inc11 = add nsw i32 %i.07, 1
%cmp = icmp slt i32 %inc11, %sizeI
br i1 %cmp, label %for.body, label %for.end12.loopexit
for.end12.loopexit: ; preds = %for.inc10
br label %for.end12
for.end12:; preds =
%for.end12.loopexit, %entry
%sub = sub nsw i32 %arraySize, 1
%idxprom13 = sext i32 %sub to i64
%arrayidx14 = getelementptr inbounds i32, i32* %A, i64 %idxprom13
%0 = load i32, i32* %arrayidx14, align 4
ret i32 %0
}
attributes #0 = { noinline nounwind
"correctly-rounded-divide-sqrt-fp-math"="false" "disable-tail-calls"="false"
"frame-pointer"="all" "less-precise-fpmad"="false" "min-legal-vector-width"="0"
"no-infs-fp-math"="false" "no-jump-tables"="false" "no-nans-fp-math"="false"
"no-
[llvm-bugs] [Bug 47301] New: clang coverage shows incorrect number of macro expansions
https://bugs.llvm.org/show_bug.cgi?id=47301
Bug ID: 47301
Summary: clang coverage shows incorrect number of macro
expansions
Product: clang
Version: trunk
Hardware: PC
OS: All
Status: NEW
Severity: enhancement
Priority: P
Component: C++
Assignee: unassignedclangb...@nondot.org
Reporter: zequa...@google.com
CC: blitzrak...@gmail.com, dgre...@apple.com,
erik.pilking...@gmail.com, llvm-bugs@lists.llvm.org,
richard-l...@metafoo.co.uk
In the following code example, it shows the macro executed 4 times, but it
should be 1. It also generates 4 identical regions, File 1, 4:15 -> 4:24 = #0 ,
which I believe is the cause for that.
$ clang++ -fcoverage-mapping -fprofile-instr-generate b.cpp -Xclang
-dump-coverage-mapping && ./a.out && llvm-profdata merge -sparse
default.profraw -o a.profdata && llvm-cov show -color ./a.out
-instr-profile=a.profdata -debug-only=coverage-mapping
main:
File 0, 3:12 -> 7:2 = #0
Expansion,File 0, 5:5 -> 5:6 = #0 (Expanded file = 1)
File 1, 4:15 -> 4:24 = #0
File 1, 4:15 -> 4:24 = #0
File 1, 4:15 -> 4:24 = #0
File 1, 4:15 -> 4:24 = #0
Counter in file 0 3:12 -> 7:2, #0
Counter in file 0 5:5 -> 5:6, Expands to file 1
Counter in file 1 4:15 -> 4:24, #0
Counter in file 1 4:15 -> 4:24, #0
Counter in file 1 4:15 -> 4:24, #0
Counter in file 1 4:15 -> 4:24, #0
Emitting segments for file:
/usr/local/google/home/zequanwu/llvm-project/build/master/b.cpp
Combined regions:
3:12 -> 7:2 (count=1)
4:15 -> 4:24 (count=4)
5:5 -> 5:6 (count=1)
Segment at 3:12 (count = 1), RegionEntry
Segment at 4:15 (count = 4), RegionEntry
Segment at 4:24 (count = 1)
Segment at 5:5 (count = 1), RegionEntry
Segment at 5:6 (count = 1)
Segment at 7:2 (count = 0), Skipped
1| |#include
2| |
3| 1|int main() {
4| 4|#define O std::cout
5| 1|O << "" << "" << "";
6| 1|return 0;
7| 1|}
--
You are receiving this mail because:
You are on the CC list for the bug.___
llvm-bugs mailing list
llvm-bugs@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] Issue 25199 in oss-fuzz: llvm:clang-fuzzer: Stack-overflow in ComplexExprEvaluator::VisitBinaryOperator
Status: New Owner: CC: k...@google.com, masc...@google.com, jdevlieg...@apple.com, igm...@gmail.com, d...@google.com, mit...@google.com, bigchees...@gmail.com, eney...@google.com, llvm-b...@lists.llvm.org, j...@chromium.org, v...@apple.com, mitchphi...@outlook.com, xpl...@gmail.com, akils...@apple.com Labels: ClusterFuzz Stability-Memory-AddressSanitizer Reproducible Engine-libfuzzer OS-Linux Proj-llvm Reported-2020-08-25 Type: Bug New issue 25199 by ClusterFuzz-External: llvm:clang-fuzzer: Stack-overflow in ComplexExprEvaluator::VisitBinaryOperator https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=25199 Detailed Report: https://oss-fuzz.com/testcase?key=5661313114832896 Project: llvm Fuzzing Engine: libFuzzer Fuzz Target: clang-fuzzer Job Type: libfuzzer_asan_llvm Platform Id: linux Crash Type: Stack-overflow Crash Address: 0x7ffc0706d800 Crash State: ComplexExprEvaluator::VisitBinaryOperator clang::StmtVisitorBase::Visit Evaluate Sanitizer: address (ASAN) Regressed: https://oss-fuzz.com/revisions?job=libfuzzer_asan_llvm&range=201912130427:201912200437 Reproducer Testcase: https://oss-fuzz.com/download?testcase_id=5661313114832896 Issue filed automatically. See https://google.github.io/oss-fuzz/advanced-topics/reproducing for instructions to reproduce this bug locally. When you fix this bug, please * mention the fix revision(s). * state whether the bug was a short-lived regression or an old bug in any stable releases. * add any other useful information. This information can help downstream consumers. If you need to contact the OSS-Fuzz team with a question, concern, or any other feedback, please file an issue at https://github.com/google/oss-fuzz/issues. Comments on individual Monorail issues are not monitored. -- You received this message because: 1. You were specifically CC'd on the issue You may adjust your notification preferences at: https://bugs.chromium.org/hosting/settings Reply to this email to add a comment.___ llvm-bugs mailing list llvm-bugs@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 47302] New: mingw-w64: std::wcout after _setmode(_fileno(stdout), _O_WTEXT) doesn't work
https://bugs.llvm.org/show_bug.cgi?id=47302
Bug ID: 47302
Summary: mingw-w64: std::wcout after _setmode(_fileno(stdout),
_O_WTEXT) doesn't work
Product: libc++
Version: 10.0
Hardware: PC
OS: Windows NT
Status: NEW
Severity: normal
Priority: P
Component: All Bugs
Assignee: unassignedclangb...@nondot.org
Reporter: nguyen.long.908...@gmail.com
CC: llvm-bugs@lists.llvm.org, mclow.li...@gmail.com
Take this code snippet:
#include
#include
#include
int wmain(int argc, const wchar_t* argv[])
{
_setmode(_fileno(stdout), _O_WTEXT);
std::wcout << L"Thử nghiệm\n";
return 0;
}
On MS STL, this correctly outputs the string, but on libc++ it outputs nothing.
Additional notes:
Add -municode when compiling with clang, and add /utf-8 when compiling with
MSVC
--
You are receiving this mail because:
You are on the CC list for the bug.___
llvm-bugs mailing list
llvm-bugs@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
