[llvm-bugs] [Bug 34858] New: clang-cl crash with constexpr ctor
https://bugs.llvm.org/show_bug.cgi?id=34858
Bug ID: 34858
Summary: clang-cl crash with constexpr ctor
Product: clang
Version: trunk
Hardware: PC
OS: Windows NT
Status: NEW
Severity: enhancement
Priority: P
Component: C++14
Assignee: unassignedclangb...@nondot.org
Reporter: powerch...@web.de
CC: llvm-bugs@lists.llvm.org
struct S
{
constexpr S(int const & = 0) {}
};
S a[2]; // crash, see log below and attached files
S b[2] {}; // crash
// In contrast:
S c[2] {0}; // OK
S d[1]; // OK
S e[1] {}; // OK
--
1>Assertion failed: Result.isUninit() && "temporary created multiple times",
file C:\src\llvm_package_314411\llvm\tools\clang\lib\AST\ExprConstant.cpp, line
1107
1>Wrote crash dump file "C:\Users\x\AppData\Local\Temp\CL.exe-308f23.dmp"
1>0x01E82CA8 (0x0016 0x04A7D360 0x768A8350 0x0003)
1>0x768453E2 (0x04A7E2AE 0x04A7D360 0x001803B8 0x058CCBBC), abort() + 0x32
bytes(s)
1>0x76846CE7 (0x0453 0x0ADD3A00 0x058CD7A0 0x058CCBCC), _get_wpgmptr() +
0x1777 bytes(s)
1>0x7684614B (0x0453 0x036AE040 0x001803A0 0x036AE040), _get_wpgmptr() +
0xBDB bytes(s)
1>0x76846DE6 (0x04A7E2AE 0x04A7D360 0x0453 0x058CCC44), _wassert() + 0x16
bytes(s)
1>0x036AE040 (0x 0x0ADD3B30 0x058CD7A0 0x0ADD3A00)
1>0x036A5F3C (0x774137EE 0x774E6E38 0x0050 0x000A)
1>0x036A21BA (0x058CCDC4 0x767CE7EB 0x000A 0x)
1>0x036A3B7D (0x058CD788 0x058CCEC0 0x00102800 0x268D563B)
1>0x03694C57 (0x058CD914 0x03634A6E 0x00129FF0 0x0ADD3B30)
1>0x02BF19E8 (0x00129FF0 0x0ADD3B30 0x0ADCBDD0 0x0001)
1>0x03634A6E (0x058CD930 0x058CD93C 0x058CD9B0 0x)
1>0x02BB0EE7 (0x 0x268D563B 0x058CD97C 0x058CD9A8)
1>clang-cl.exe : error : clang frontend command failed due to signal (use -v to
see invocation)
1>clang version 6.0.0 (trunk)
1>Target: i686-pc-windows-msvc
1>Thread model: posix
1>InstalledDir: C:\Program Files (x86)\LLVM\msbuild-bin
1>clang-cl.exe: note: diagnostic msg: PLEASE submit a bug report to
http://llvm.org/bugs/ and include the crash backtrace, preprocessed source, and
associated run script.
1>clang-cl.exe: note: diagnostic msg:
1>
1>
1>PLEASE ATTACH THE FOLLOWING FILES TO THE BUG REPORT:
1>Preprocessed source(s) and associated run script(s) are located at:
1>clang-cl.exe: note: diagnostic msg:
C:\Users\x\AppData\Local\Temp\State1-fca48a.cpp
1>clang-cl.exe: note: diagnostic msg:
C:\Users\x\AppData\Local\Temp\State1-fca48a.sh
1>clang-cl.exe: note: diagnostic msg:
1>
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
http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 34859] New: Isel crash with large constant (csmith)
https://bugs.llvm.org/show_bug.cgi?id=34859
Bug ID: 34859
Summary: Isel crash with large constant (csmith)
Product: libraries
Version: trunk
Hardware: PC
OS: Linux
Status: NEW
Severity: enhancement
Priority: P
Component: Backend: SystemZ
Assignee: unassignedb...@nondot.org
Reporter: pauls...@linux.vnet.ibm.com
CC: llvm-bugs@lists.llvm.org
This (reduced) program:
@g_272 = external global i64, align 8
@g_276 = external global i16, align 2
; Function Attrs: noreturn nounwind
define void @main(i32 signext) local_unnamed_addr #0 {
store i16 -1, i16* @g_276, align 2
%2 = load i16, i16* @g_276, align 2
%3 = icmp ne i16 %2, 0
%4 = zext i1 %3 to i64
%5 = or i64 %4, 3944173009226982604
store i64 %5, i64* @g_272, align 8
tail call void @transparent_crc(i64 3944173009226982604, i32 signext undef)
unreachable
}
declare void @transparent_crc(i64, i32 signext) local_unnamed_addr #1
+
bin/llc -mtriple=s390x-linux-gnu -mcpu=z13
crashes with
LLVM ERROR: Cannot select: t47: i64 = Constant<3944173007420784641>
DAG:
Optimized legalized selection DAG: BB#0 'main:'
SelectionDAG has 25 nodes:
t0: ch = EntryToken
t45: i64 = SystemZISD::PCREL_WRAPPER TargetGlobalAddress:i64 0
t41: ch = store t0, Constant:i32<65535>,
t45, undef:i64
t18: i64 = or Constant:i64<1>,
OpaqueConstant:i64<3944173009226982604>
t43: i64 = SystemZISD::PCREL_WRAPPER TargetGlobalAddress:i64 0
t34: ch = store t0, t18, t43, undef:i64
t35: ch = TokenFactor t41, t34
t24: ch,glue = callseq_start t35, TargetConstant:i64<0>,
TargetConstant:i64<0>
t28: ch,glue = CopyToReg t24, Register:i64 %R2D,
OpaqueConstant:i64<3944173009226982604>
t30: ch,glue = CopyToReg t28, Register:i64 %R3D, Constant:i64<0>, t28:1
t26: i64 = SystemZISD::PCREL_WRAPPER TargetGlobalAddress:i64 0
t32: ch,glue = SystemZISD::CALL t30, t26, Register:i64 %R2D, Register:i64
%R3D, RegisterMask:Untyped, t30:1
t33: ch,glue = callseq_end t32, TargetConstant:i64<0>, TargetConstant:i64<0>,
t32:1
--
You are receiving this mail because:
You are on the CC list for the bug.___
llvm-bugs mailing list
llvm-bugs@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 34712] --hash-style=both by default (like recent ld.bfd and ld.gold)
https://bugs.llvm.org/show_bug.cgi?id=34712 George Rimar changed: What|Removed |Added Status|ASSIGNED|RESOLVED Resolution|--- |FIXED --- Comment #2 from George Rimar --- r315051 + r315052 -- You are receiving this mail because: You are on the CC list for the bug.___ llvm-bugs mailing list llvm-bugs@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 34705] Crash when a symbol references external version definition
https://bugs.llvm.org/show_bug.cgi?id=34705 George Rimar changed: What|Removed |Added Status|NEW |RESOLVED Resolution|--- |FIXED --- Comment #8 from George Rimar --- r315053 + r315054. -- You are receiving this mail because: You are on the CC list for the bug.___ llvm-bugs mailing list llvm-bugs@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 34860] New: Csmith program gives bad checksum with -disable-basicaa
https://bugs.llvm.org/show_bug.cgi?id=34860 Bug ID: 34860 Summary: Csmith program gives bad checksum with -disable-basicaa Product: libraries Version: trunk Hardware: PC OS: Linux Status: NEW Severity: enhancement Priority: P Component: Common Code Generator Code Assignee: unassignedb...@nondot.org Reporter: pauls...@linux.vnet.ibm.com CC: llvm-bugs@lists.llvm.org Created attachment 19247 --> https://bugs.llvm.org/attachment.cgi?id=19247&action=edit bugpoint reduced testcase with same checksums / options Preprocessed orinal c-program: wrong10.i 49B9B5C5: /root/llvm/build/llvm-dev/bin/clang -O3 -march=z13 wrong10.c -I/root/csmith/runtime -o a.out -w 5FFE0748: /root/llvm/build/llvm-dev/bin/clang -O3 -march=z13 wrong10.c -I/root/csmith/runtime -o a.out -w -mllvm -disable-basicaa (Out of 15 builds with randomized options did all but one gave checksum 49B9B5C5). reduced program (also prints a number of other lines): wrong10.reduced.bc /root/llvm/build/llvm-dev/bin/clang -O3 -march=z13 ./wrong10.reduced.bc -o a.out -w; ./a.out |& grep checksum checksum = 49B9B5C5 /root/llvm/build/llvm-dev/bin/clang -O3 -march=z13 ./wrong10.reduced.bc -o a.out -w -mllvm -disable-basicaa; ./a.out |& grep checksum checksum = 5FFE0748 Note : the commits were: LLVM : e5e1c85 / trunk@314416 clang: 0e1f5c3 / cfe/trunk@314391 -- You are receiving this mail because: You are on the CC list for the bug.___ llvm-bugs mailing list llvm-bugs@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 10114] EltsFromConsecutiveLoads does not preserve chain successor dependencies from the original loads.
https://bugs.llvm.org/show_bug.cgi?id=10114 Artur Pilipenko changed: What|Removed |Added Status|NEW |RESOLVED Resolution|--- |FIXED --- Comment #7 from Artur Pilipenko --- Yes, fixed by https://reviews.llvm.org/rL314988. -- You are receiving this mail because: You are on the CC list for the bug.___ llvm-bugs mailing list llvm-bugs@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 34861] New: backport SVN r313398 to 5.1
https://bugs.llvm.org/show_bug.cgi?id=34861 Bug ID: 34861 Summary: backport SVN r313398 to 5.1 Product: libraries Version: trunk Hardware: PC OS: All Status: NEW Severity: enhancement Priority: P Component: Bitcode Reader Assignee: unassignedb...@nondot.org Reporter: compn...@compnerd.org CC: llvm-bugs@lists.llvm.org We should back port SVN r313398 to the 5.1 release as that can break LTO compatibility across releases. -- You are receiving this mail because: You are on the CC list for the bug.___ llvm-bugs mailing list llvm-bugs@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 34862] New: clang crashes compiling "volatile long double resld = __builtin_canonicalizel(LD)"
https://bugs.llvm.org/show_bug.cgi?id=34862
Bug ID: 34862
Summary: clang crashes compiling "volatile long double resld =
__builtin_canonicalizel(LD)"
Product: libraries
Version: trunk
Hardware: PC
OS: All
Status: NEW
Severity: normal
Priority: P
Component: Backend: X86
Assignee: unassignedb...@nondot.org
Reporter: ahata...@gmail.com
CC: llvm-bugs@lists.llvm.org
clang crashes when compiling the following code:
$ cat canon-crash.c
void test_float_builtin_ops(long double LD) {
volatile long double resld = __builtin_canonicalizel(LD);
}
$ clang canon-crash.c -Os -o - -S
fatal error: error in backend: Cannot select: t7: f80 = fcanonicalize t3
t3: f80,ch = load t0, FrameIndex:i64<-1>,
undef:i64
t1: i64 = FrameIndex<-1>
t2: i64 = undef
In function: test_float_builtin_ops
--
You are receiving this mail because:
You are on the CC list for the bug.___
llvm-bugs mailing list
llvm-bugs@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 34863] New: Call frame optimization fails for immediate arguments of 0 or -1
https://bugs.llvm.org/show_bug.cgi?id=34863
Bug ID: 34863
Summary: Call frame optimization fails for immediate arguments
of 0 or -1
Product: libraries
Version: trunk
Hardware: PC
OS: Windows NT
Status: NEW
Severity: enhancement
Priority: P
Component: Backend: X86
Assignee: unassignedb...@nondot.org
Reporter: david.l.kreit...@intel.com
CC: llvm-bugs@lists.llvm.org
The X86CallFrameOptimization fails to recognize outgoing parameter stores of 0,
-1 that are implemented using AND, OR. So we are missing out on some
opportunities to use PUSH instead of MOV.
bash-4.2$ cat t.ll
declare void @f2(i32, i32, i32, i32, i32, i32, i32, i32)
define void @f1(i32 %x) minsize nounwind {
entry:
tail call void @f2(i32 %x, i32 %x, i32 %x, i32 %x, i32 %x, i32 %x, i32 0, i32
-1) #2
ret void
}
bash-4.2$ llc -mtriple=i686-unknown-unknown t.ll -o -
.text
.file "t.ll"
.globl f1 # -- Begin function f1
.p2align4, 0x90
.type f1,@function
f1: # @f1
# BB#0: # %entry
subl$32, %esp
movl36(%esp), %eax
movl%eax, 20(%esp)
movl%eax, 16(%esp)
movl%eax, 12(%esp)
movl%eax, 8(%esp)
movl%eax, 4(%esp)
movl%eax, (%esp)
orl $-1, 28(%esp)
andl$0, 24(%esp)
calll f2
addl$32, %esp
retl
.Lfunc_end0:
.size f1, .Lfunc_end0-f1
# -- End function
.section".note.GNU-stack","",@progbits
bash-4.2$ llc -mtriple=x86_64-unknown-unknown t.ll -o -
.text
.file "t.ll"
.globl f1 # -- Begin function f1
.p2align4, 0x90
.type f1,@function
f1: # @f1
# BB#0: # %entry
subq$24, %rsp
orl $-1, 8(%rsp)
andl$0, (%rsp)
movl%edi, %esi
movl%edi, %edx
movl%edi, %ecx
movl%edi, %r8d
movl%edi, %r9d
callq f2
addq$24, %rsp
retq
.Lfunc_end0:
.size f1, .Lfunc_end0-f1
# -- End function
.section".note.GNU-stack","",@progbits
--
You are receiving this mail because:
You are on the CC list for the bug.___
llvm-bugs mailing list
llvm-bugs@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 34864] New: Deadlock inside llvm::parallel::TaskList::spawn stops LLD compiled with MinGW-w64
https://bugs.llvm.org/show_bug.cgi?id=34864 Bug ID: 34864 Summary: Deadlock inside llvm::parallel::TaskList::spawn stops LLD compiled with MinGW-w64 Product: libraries Version: trunk Hardware: Other OS: Linux Status: NEW Severity: enhancement Priority: P Component: Support Libraries Assignee: unassignedb...@nondot.org Reporter: vasek.ge...@gmail.com CC: llvm-bugs@lists.llvm.org I managed to build the latest trunk with MinGW-w64 GCC 7.1.0 (posix-seh) and it works fine producing cross-platform code. During the tests I noticed that linking an ELF target on Windows host platform with LLD results in a deadlock in llvm::parallel::TaskList::spawn (basically, here: https://llvm.org/doxygen/Parallel_8cpp_source.html#l0098 ). The correct ELF file is produced only if -Wlm--no-threads switch is added to clang command line. I tried to change the add() function like described here: https://stackoverflow.com/questions/15912322/condition-variable-deadlock but the deadlock moved into ~Latch destructor. -- You are receiving this mail because: You are on the CC list for the bug.___ llvm-bugs mailing list llvm-bugs@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 33430] [ubsan] false negative: pointer overflow check misses subtractions with unsigned offsets
https://bugs.llvm.org/show_bug.cgi?id=33430 Vedant Kumar changed: What|Removed |Added Status|NEW |RESOLVED Resolution|--- |FIXED --- Comment #1 from Vedant Kumar --- This was fixed in cfe/r307955 -- You are receiving this mail because: You are on the CC list for the bug.___ llvm-bugs mailing list llvm-bugs@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 34865] New: Clang doesn't require references to be listed in lambda capture
https://bugs.llvm.org/show_bug.cgi?id=34865
Bug ID: 34865
Summary: Clang doesn't require references to be listed in
lambda capture
Product: clang
Version: 5.0
Hardware: PC
OS: Linux
Status: NEW
Severity: normal
Priority: P
Component: C++11
Assignee: unassignedclangb...@nondot.org
Reporter: redbeard0...@gmail.com
CC: dgre...@apple.com, llvm-bugs@lists.llvm.org
Clang will compile this faulty code without an error:
int var = 0;
int main() {
int& varRef = var;
return []{return varRef;}();
}
This also leads to an incorrect warning if varRef is correctly captured (which
is how we noticed this issue in the first place):
> clang++ test2.cpp --std=c++17 -Wall
test2.cpp:4:14: warning: lambda capture 'varRef' is not required to be captured
for this use [-Wunused-lambda-capture]
return [&varRef]{return varRef;}();
^
This is a reduced example from this real-world code:
https://github.com/mongodb/mongo/blob/r3.5.13/src/mongo/db/repl/replication_coordinator_external_state_impl.cpp#L383-L402
(clang++-5.0 -Wall warns that kRsOplogNamespace is unused even though it is
used on line 402)
--
You are receiving this mail because:
You are on the CC list for the bug.___
llvm-bugs mailing list
llvm-bugs@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 34243] Clang should default to -fms-compatibility-version=19.11
https://bugs.llvm.org/show_bug.cgi?id=34243 Reid Kleckner changed: 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 http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 34849] clang-cl fails to dllexport specialization of class template member function when the class template is dllexport
https://bugs.llvm.org/show_bug.cgi?id=34849 Reid Kleckner changed: What|Removed |Added Resolution|FIXED |--- CC||r...@google.com Status|RESOLVED|REOPENED --- Comment #2 from Reid Kleckner --- Had to revert due to base.dll link failures in Chromium. -- You are receiving this mail because: You are on the CC list for the bug.___ llvm-bugs mailing list llvm-bugs@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 34866] New: [Regression] Infinite loop in SoftenFloatOperand
https://bugs.llvm.org/show_bug.cgi?id=34866
Bug ID: 34866
Summary: [Regression] Infinite loop in SoftenFloatOperand
Product: libraries
Version: 5.0
Hardware: PC
OS: Windows NT
Status: NEW
Severity: enhancement
Priority: P
Component: Common Code Generator Code
Assignee: unassignedb...@nondot.org
Reporter: bma...@codeaurora.org
CC: c...@google.com, florian.h...@arm.com,
llvm-bugs@lists.llvm.org, mcros...@codeaurora.org
Blocks: 34492
The following code no longer compiles:
$ cat test.c
typedef unsigned __int128 u128_t;
typedef __float128 f128_t;
typedef union
{
f128_t f;
u128_t i;
} union128_t;
int test(volatile f128_t *start_X_ptr, volatile f128_t *start_Y_ptr)
{
for (unsigned iter_count = 1; iter_count <= 2; iter_count++)
{
union128_t read_valX = { .f = *start_X_ptr };
union128_t read_valY = { .f = *start_Y_ptr };
u128_t expected_val = ((iter_count - 1) * read_valY.i);
if (read_valX.i != expected_val)
{
return -1;
}
union128_t write_valX = { .i = (iter_count * (read_valX.i + 1)) };
union128_t write_valY = { .i = (read_valX.i + 1) };
*start_X_ptr = write_valX.f;
*start_Y_ptr = write_valY.f;
}
return 0;
}
because of an infinite loop in SoftenFloatOperand.
$ clang -O3 -c test.c -mllvm -debug-only=legalize-types
Soften float operand 0: t33: v16i8 = bitcast t20
Soften float operand 0: t33: v16i8 = bitcast t20
Soften float operand 0: t33: v16i8 = bitcast t20
Soften float operand 0: t33: v16i8 = bitcast t20
Soften float operand 0: t33: v16i8 = bitcast t20
Soften float operand 0: t33: v16i8 = bitcast t20
Soften float operand 0: t33: v16i8 = bitcast t20
Soften float operand 0: t33: v16i8 = bitcast t20
Works with LLVM4.0.1: https://godbolt.org/g/W8cULV
I suspect this was introduced in r254653 and got exposed sometime between
03/14/2017 and 03/17/2017.
Referenced Bugs:
https://bugs.llvm.org/show_bug.cgi?id=34492
[Bug 34492] [meta] 5.0.1 Release Blockers
--
You are receiving this mail because:
You are on the CC list for the bug.___
llvm-bugs mailing list
llvm-bugs@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 34867] New: Merge r315086 to 5.0.1
https://bugs.llvm.org/show_bug.cgi?id=34867 Bug ID: 34867 Summary: Merge r315086 to 5.0.1 Product: libraries Version: trunk Hardware: PC OS: All Status: NEW Severity: enhancement Priority: P Component: Bitcode Reader Assignee: unassignedb...@nondot.org Reporter: compn...@compnerd.org CC: llvm-bugs@lists.llvm.org -- You are receiving this mail because: You are on the CC list for the bug.___ llvm-bugs mailing list llvm-bugs@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
