[llvm-bugs] [Bug 26203] New: Assertion `QualifierLoc && "should be created for dependent qualifiers"' failed.
https://llvm.org/bugs/show_bug.cgi?id=26203
Bug ID: 26203
Summary: Assertion `QualifierLoc && "should be created for
dependent qualifiers"' failed.
Product: clang
Version: unspecified
Hardware: PC
OS: All
Status: NEW
Severity: normal
Priority: P
Component: C++11
Assignee: unassignedclangb...@nondot.org
Reporter: david.majne...@gmail.com
CC: dgre...@apple.com, llvm-bugs@lists.llvm.org
Classification: Unclassified
consider:
template
inline void f(T) {
struct S {
inline operator T() {}
static void X() {
&operator T;
}
};
}
int main() {
f(3);
}
run with:
clang -cc1 -S -std=c++11 -x c++ -emit-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
http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 26204] New: Feature request: Align consecutive #define macros
https://llvm.org/bugs/show_bug.cgi?id=26204 Bug ID: 26204 Summary: Feature request: Align consecutive #define macros Product: clang Version: unspecified Hardware: All OS: All Status: NEW Severity: normal Priority: P Component: Formatter Assignee: unassignedclangb...@nondot.org Reporter: eyco...@airspan.com CC: djas...@google.com, kli...@google.com, llvm-bugs@lists.llvm.org Classification: Unclassified Hello. Just like there's AlignConsecutiveAssignments, I'd like to do the same for #defines. Example: #define LWR_A 1 #define LWR_AAA 2 #define LWR_ 4 #define LWR_BB 8 #define LWR_BBAA 16 Should become: #define LWR_A1 #define LWR_AAA 2 #define LWR_ 4 #define LWR_BB 8 #define LWR_BBAA 16 Thank you. -- 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 26206] New: TestExitDuringStep sometimes hangs on android
https://llvm.org/bugs/show_bug.cgi?id=26206 Bug ID: 26206 Summary: TestExitDuringStep sometimes hangs on android Product: lldb Version: unspecified Hardware: PC OS: Linux Status: NEW Severity: normal Priority: P Component: All Bugs Assignee: lldb-...@lists.llvm.org Reporter: lab...@google.com CC: llvm-bugs@lists.llvm.org Classification: Unclassified Test sometimes hangs on the linux->android buildbot. The logs seem to indicate that the process is simply not hitting the breakpoint after the initial "run" command. It's hard to diagnose further without getting logs from the server (which we are not collecting at the moment). -- 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 19396] ARM code emitter misoptimization with 16-bit multiplication
https://llvm.org/bugs/show_bug.cgi?id=19396 Renato Golin changed: What|Removed |Added Status|NEW |RESOLVED CC||renato.go...@linaro.org Resolution|--- |FIXED --- Comment #1 from Renato Golin --- Trunk is 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 17136] ARMLoadStoreOptimizer: MemOps selected incorrectly
https://llvm.org/bugs/show_bug.cgi?id=17136 Renato Golin changed: What|Removed |Added Status|NEW |RESOLVED CC||renato.go...@linaro.org Resolution|--- |WONTFIX --- Comment #5 from Renato Golin --- We need a bug in OSS trunk, otherwise will be hard to track and keep fixed. Sorry. -- 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 26207] New: Assertion failure in ScalarEvolution::getBackedgeTakenInfo
https://llvm.org/bugs/show_bug.cgi?id=26207
Bug ID: 26207
Summary: Assertion failure in
ScalarEvolution::getBackedgeTakenInfo
Product: libraries
Version: trunk
Hardware: PC
OS: Linux
Status: NEW
Severity: normal
Priority: P
Component: Scalar Optimizations
Assignee: unassignedb...@nondot.org
Reporter: mattias.v.eriks...@ericsson.com
CC: llvm-bugs@lists.llvm.org
Classification: Unclassified
Created attachment 15661
--> https://llvm.org/bugs/attachment.cgi?id=15661&action=edit
reduced input file
I am getting a crash in "indvars":
opt -S -indvars bugpoint-reduced-simplified.ll
Attempt to use a SCEVCouldNotCompute object!
UNREACHABLE executed at ../lib/Analysis/ScalarEvolution.cpp:9287!
0 opt 0x0190453e
llvm::sys::PrintStackTrace(llvm::raw_ostream&) + 46
1 opt 0x01904949
2 opt 0x01903253 llvm::sys::RunSignalHandlers() + 131
3 opt 0x01904d8f
4 libpthread.so.0 0x7effc7f64340
5 libc.so.6 0x7effc718ccc9 gsignal + 57
6 libc.so.6 0x7effc71900d8 abort + 328
7 opt 0x01895e16
8 opt 0x00c3678d
llvm::ScalarEvolution::computeLoopDisposition(llvm::SCEV const*, llvm::Loop
const*) + 1085
9 opt 0x00c3622f
llvm::ScalarEvolution::getLoopDisposition(llvm::SCEV const*, llvm::Loop const*)
+ 239
10 opt 0x00c178b5
llvm::ScalarEvolution::isLoopInvariant(llvm::SCEV const*, llvm::Loop const*) +
37
11 opt 0x00c24dd9
llvm::ScalarEvolution::getBackedgeTakenInfo(llvm::Loop const*) + 377
12 opt 0x00c2523f
llvm::ScalarEvolution::getBackedgeTakenCount(llvm::Loop const*) + 47
13 opt 0x016c6534
14 opt 0x00ba7122
llvm::LPPassManager::runOnFunction(llvm::Function&) + 1138
15 opt 0x0134d260
llvm::FPPassManager::runOnFunction(llvm::Function&) + 416
16 opt 0x0134d595
llvm::FPPassManager::runOnModule(llvm::Module&) + 117
17 opt 0x0134dd81
18 opt 0x0134d87b
llvm::legacy::PassManagerImpl::run(llvm::Module&) + 347
19 opt 0x0134e2c1
llvm::legacy::PassManager::run(llvm::Module&) + 33
20 opt 0x007de450 main + 8576
21 libc.so.6 0x7effc7177ec5 __libc_start_main + 245
22 opt 0x007b8a48
Stack dump:
0. Program arguments: build/bin/opt -S -indvars
bugpoint-reduced-simplified.ll
1. Running pass 'Function Pass Manager' on module
'bugpoint-reduced-simplified.ll'.
2. Running pass 'Loop Pass Manager' on function '@main'
3. Running pass 'Induction Variable Simplification' on basic block
'%bb1.i'
Aborted
input file:
; ModuleID = 'bugpoint-reduced-simplified.bc'
target triple = "x86_64-unknown-linux-gnu"
define void @main() {
br label %bb2
bb2: ; preds = %bb1.i, %bb2, %0
%_tmp44.i = icmp slt i16 undef, 2
br i1 %_tmp44.i, label %bb1.i, label %bb2
bb1.i:; preds = %bb1.i, %bb2
%_tmp25.i = phi i16 [ undef, %bb2 ], [ %_tmp6.i, %bb1.i ]
%_tmp6.i = add nsw i16 %_tmp25.i, 1
%_tmp10.i = icmp sge i16 %_tmp6.i, 2
%exitcond.i = icmp eq i16 %_tmp6.i, 2
%or.cond = and i1 %_tmp10.i, %exitcond.i
br i1 %or.cond, label %bb2, label %bb1.i
}
This check in ScalarEvolution.cpp looks suspicious:
if (Result.getExact(this) != getCouldNotCompute()) {
assert(isLoopInvariant(Result.getExact(this), L) &&
isLoopInvariant(Result.getMax(this), L) &&
"Computed backedge-taken count isn't loop invariant for loop!");
++NumTripCountsComputed;
}
The assertion is guarded for the getExact() case, but not for the getMax()
case. It is Result.getMax() that returns the SCEVCouldNotCompute here.
--
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 24549] ARM backend crash when compiling half-precision FP op without hardware FP
https://llvm.org/bugs/show_bug.cgi?id=24549
Renato Golin changed:
What|Removed |Added
Status|NEW |RESOLVED
CC||renato.go...@linaro.org
Resolution|--- |FIXED
--- Comment #5 from Renato Golin ---
The test now produces:
push{r4, r5, r6, lr}
ldrhr0, [r0]
movr4, r2
movr5, r1
bl__aeabi_h2f
movr6, r0
ldrhr0, [r5]
bl__aeabi_h2f
movr1, r0
movr0, r6
bl__aeabi_fadd
bl__aeabi_f2h
strhr0, [r4]
pop{r4, r5, r6, pc}
Which, though a bit ugly, seems correct.
--
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 21313] [aarch64 disassembler] Some instructions do not have correct condition flag set
https://llvm.org/bugs/show_bug.cgi?id=21313 Renato Golin changed: What|Removed |Added Status|NEW |RESOLVED CC||renato.go...@linaro.org Resolution|--- |INVALID --- Comment #2 from Renato Golin --- A lot has changed since 3.3. Without a test case, command line options and trunk behaviour, there isn't much we can do. -- 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 8690] Block that inherits 'self' from enclosing block can get wrong value (ARM)
https://llvm.org/bugs/show_bug.cgi?id=8690 Renato Golin changed: What|Removed |Added Status|NEW |RESOLVED CC||renato.go...@linaro.org Resolution|--- |INVALID --- Comment #1 from Renato Golin --- Apple specific bugs should go into Apple's own bug tracker. -- 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 10833] LLVM generates wrong reloc type for global_ctors (section .init_array) for ARM target
https://llvm.org/bugs/show_bug.cgi?id=10833 Renato Golin changed: What|Removed |Added Status|ASSIGNED|RESOLVED CC||renato.go...@linaro.org Resolution|--- |INVALID --- Comment #4 from Renato Golin --- ARM is now using LLVM technology for their tools, so this feature is likely already fixed. If not, contacting ARM customer support is probably the best way to go from here. -- 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 12636] LLVM crashes with -O2 on ARM
https://llvm.org/bugs/show_bug.cgi?id=12636 Renato Golin changed: What|Removed |Added Status|NEW |RESOLVED CC||renato.go...@linaro.org Resolution|--- |FIXED --- Comment #9 from Renato Golin --- Working on trunk. -- 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 12988] ARM NEON intrinsic vextq_u16() produces very bad code
https://llvm.org/bugs/show_bug.cgi?id=12988
Renato Golin changed:
What|Removed |Added
Status|NEW |RESOLVED
CC||renato.go...@linaro.org
Resolution|--- |FIXED
--- Comment #1 from Renato Golin ---
Fixed on trunk:
.LBB0_1:@ %for.body
@ =>This Inner Loop Header: Depth=1
vld1.16{d16, d17}, [r1], r2
subsr0, r0, #1
vmlal.u16q1, d17, d4[0]
vmlal.u16q0, d16, d4[0]
vext.16q2, q2, q2, #1
bne.LBB0_1
@ BB#2: @ %for.end
bxlr
--
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 26191] Info from .cfi directives should by default go into .debug_frame, not .eh_frame section
https://llvm.org/bugs/show_bug.cgi?id=26191 Bart Jacobs changed: What|Removed |Added Status|NEW |RESOLVED Resolution|--- |INVALID --- Comment #7 from Bart Jacobs --- I think I figured out why people say the x86_64 ABI requires .eh_frame. The main reason is: being able to get a correct backtrace, even in the absence of debug symbols, is considered important for diagnosis and monitoring. Putting the info in the .debug_frame section means it is only there in debug builds, and it gets stripped when stripping debug info. That is considered undesirable. I can appreciate that. Still, I wish the use case of generating a backtrace were separated from the use case of exception handling. In particular, I wish the Objective-C runtime (through libunwind) did not interpret the presence of .eh_frame in my OCaml binary as an invitation to walk into my OCaml stack frames (and crash on the incorrect unwind info). I wish it were possible to mark frames as "exception-safe" or not and to have the Objective-C runtime not walk into frames not marked as exception-safe. But that's a separate request, I guess. So I'm marking this as resolved. -- 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 26208] New: Handling of .space is really slow
https://llvm.org/bugs/show_bug.cgi?id=26208 Bug ID: 26208 Summary: Handling of .space is really slow Product: libraries Version: trunk Hardware: PC OS: Linux Status: NEW Severity: normal Priority: P Component: Loop Optimizer Assignee: unassignedb...@nondot.org Reporter: rafael.espind...@gmail.com CC: llvm-bugs@lists.llvm.org Classification: Unclassified A file with just .space 0x1 takes 14 seconds to assemble with llvm-mc and 4 seconds with gas. -- 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 13203] vqshrn NEON instruction does not support #0 immediate shift
https://llvm.org/bugs/show_bug.cgi?id=13203 Renato Golin changed: What|Removed |Added Status|NEW |RESOLVED CC||renato.go...@linaro.org Resolution|--- |INVALID --- Comment #1 from Renato Golin --- ARM ARM states (A8.8.381): " The immediate value, in the range 1 to /2. See the description of for how is encoded." If the shift value is zero, you can use VQMOVN. Changing that immediate to non-zero value works as expected. -- 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 13275] Problem when passing small vector as parameter for ARM + NEON target
https://llvm.org/bugs/show_bug.cgi?id=13275
Renato Golin changed:
What|Removed |Added
Status|NEW |RESOLVED
CC||renato.go...@linaro.org
Resolution|--- |INVALID
--- Comment #2 from Renato Golin ---
Funny enough, GAS is now perfectly ok with the syntax, and throws errors if we
use 64:
$ arm-none-eabi-as -march=armv7-a -mcpu=cortex-a9 -mfpu=neon -mfloat-abi=hard
bar.s
bar.s: Assembler messages:
bar.s:30: Error: unsupported alignment for instruction -- `vld1.16
{d16[0]},[r1:64]'
bar.s:31: Error: unsupported alignment for instruction -- `vld1.16
{d17[0]},[r0:64]'
While LLVM-MC seems to be on the same page:
$ echo "vld1.16{d16[0]}, [r1:64]" | llvm-mc -triple=armv7-a-linux-gnueabihf
-mcpu=cortex-a9 -show-encoding
.text
:1:22: error: alignment must be 16 or omitted
vld1.16{d16[0]}, [r1:64]
^
Even though the ARM ARM (A8.8.320) seems to imply it should be in bits, both
tools want in bytes, and identical to the element size (which makes some
sense).
--
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 13556] ARM AsmPrinter doesn't support %H modifier in inline ASM
https://llvm.org/bugs/show_bug.cgi?id=13556 Renato Golin changed: What|Removed |Added Status|NEW |RESOLVED CC||renato.go...@linaro.org Resolution|--- |FIXED --- Comment #3 from Renato Golin --- Seems to work on trunk. -- 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 13743] Lacks an option to set AllowFPOpFusion on ARM Target
https://llvm.org/bugs/show_bug.cgi?id=13743 Renato Golin changed: What|Removed |Added Status|NEW |RESOLVED CC||renato.go...@linaro.org Resolution|--- |INVALID -- 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 12769] unreachable executed in MCAssembler on ARM with -mthumb and -mcpu=cortex-a9
https://llvm.org/bugs/show_bug.cgi?id=12769 Renato Golin changed: What|Removed |Added Status|NEW |RESOLVED CC||renato.go...@linaro.org Resolution|--- |FIXED --- Comment #8 from Renato Golin --- works on trunk -- 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 1642] SjLj (setjmp/longjmp)-based exception handling in clang seems to be unsupported for anything non arm/darwin
https://llvm.org/bugs/show_bug.cgi?id=1642 Renato Golin changed: What|Removed |Added Status|REOPENED|RESOLVED CC||renato.go...@linaro.org Resolution|--- |FIXED --- Comment #9 from Renato Golin --- llvm-gcc is dead. Clang supports both SJLJ and EHABI exception handling. -- 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 12637] LLVM crashes with -O2 on ARM during SelectionDAG
https://llvm.org/bugs/show_bug.cgi?id=12637 Renato Golin changed: What|Removed |Added Status|NEW |RESOLVED CC||renato.go...@linaro.org Resolution|--- |FIXED --- Comment #6 from Renato Golin --- works on trunk -- 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 25576] clang crashes at -O1 and above on x86_64-linux-gnu
https://llvm.org/bugs/show_bug.cgi?id=25576 Sanjoy Das changed: What|Removed |Added Status|NEW |RESOLVED Resolution|--- |FIXED --- Comment #4 from Sanjoy Das --- Fixed in 258016 -- 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 15057] sign-extension folding for ARM (and other backends) depends too much on block-placement
https://llvm.org/bugs/show_bug.cgi?id=15057 Renato Golin changed: What|Removed |Added Status|ASSIGNED|RESOLVED CC||renato.go...@linaro.org Resolution|--- |FIXED --- Comment #2 from Renato Golin --- Option doesn't exist anymore, and the code generated by -O3 is much better than any example here. -- 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 16002] RuntimeDyld generates incorrect Stub Functions on Arm
https://llvm.org/bugs/show_bug.cgi?id=16002 Renato Golin changed: What|Removed |Added Status|NEW |RESOLVED CC||renato.go...@linaro.org Resolution|--- |INVALID --- Comment #1 from Renato Golin --- Patches should be submitted to for review on http://reviews.llvm.org/ or one of the commit mailing lists. -- 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 16741] ARM isel hits assert with simple program
https://llvm.org/bugs/show_bug.cgi?id=16741 Renato Golin changed: What|Removed |Added Status|NEW |RESOLVED CC||renato.go...@linaro.org Resolution|--- |FIXED --- Comment #1 from Renato Golin --- works on trunk -- 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 12946] "Wrong MachineOperand accessor" assertion hit when compiling NEON intrinsics code
https://llvm.org/bugs/show_bug.cgi?id=12946 Renato Golin changed: What|Removed |Added Status|NEW |RESOLVED CC||renato.go...@linaro.org Resolution|--- |FIXED --- Comment #3 from Renato Golin --- works on trunk -- 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 17031] llc on ARM doesn't emit exception catch
https://llvm.org/bugs/show_bug.cgi?id=17031 Renato Golin changed: What|Removed |Added Status|NEW |RESOLVED CC||renato.go...@linaro.org Resolution|--- |FIXED --- Comment #5 from Renato Golin --- behaviour enabled by default -- 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 17817] ARMv7: Kaleidoscope "2" returns "1.000000".
https://llvm.org/bugs/show_bug.cgi?id=17817 Renato Golin changed: What|Removed |Added Status|NEW |RESOLVED CC||renato.go...@linaro.org Resolution|--- |INVALID -- 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 17881] srem/urem i64 for ARM aeabi generates __moddi3 instead of __aeabi_ldivmod
https://llvm.org/bugs/show_bug.cgi?id=17881 Renato Golin changed: What|Removed |Added Status|NEW |RESOLVED Resolution|--- |FIXED --- Comment #4 from Renato Golin --- Fixed on trunk -- 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 18112] arm call generates incorrect code
https://llvm.org/bugs/show_bug.cgi?id=18112 Renato Golin changed: What|Removed |Added Status|NEW |RESOLVED CC||renato.go...@linaro.org Resolution|--- |FIXED --- Comment #1 from Renato Golin --- works on trunk -- 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 18368] [thumbv8] assertions fails when check for eligibility of IT block
https://llvm.org/bugs/show_bug.cgi?id=18368 Renato Golin changed: What|Removed |Added Status|NEW |RESOLVED CC||renato.go...@linaro.org Resolution|--- |FIXED --- Comment #1 from Renato Golin --- works on trunk -- 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 18370] [thumbv8] insts that sets CPSR are bundled into IT block mistakenly
https://llvm.org/bugs/show_bug.cgi?id=18370 Renato Golin changed: What|Removed |Added Status|NEW |RESOLVED CC||renato.go...@linaro.org Resolution|--- |FIXED --- Comment #1 from Renato Golin --- works on trunk -- 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 14071] UNREACHABLE executed in rewriteARMFrameIndex while handling __builtin_prefetch of local variable on ARM
https://llvm.org/bugs/show_bug.cgi?id=14071 Renato Golin changed: What|Removed |Added Status|NEW |RESOLVED CC||renato.go...@linaro.org Resolution|--- |FIXED --- Comment #3 from Renato Golin --- works on trunk -- 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 20056] [AArch64] [Assembler] LDR x0, = not recognized
https://llvm.org/bugs/show_bug.cgi?id=20056 Renato Golin changed: What|Removed |Added Status|NEW |RESOLVED CC||renato.go...@linaro.org Resolution|--- |FIXED --- Comment #3 from Renato Golin --- works on trunk -- 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 21187] trunk/llvm/tools/clang/utils/TableGen/NeonEmitter.cpp:1941: possible missing call to erase ?
https://llvm.org/bugs/show_bug.cgi?id=21187 Renato Golin changed: What|Removed |Added Status|NEW |RESOLVED CC||renato.go...@linaro.org 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 26059] [meta] 3.8.0 Release Blockers
https://llvm.org/bugs/show_bug.cgi?id=26059 Bug 26059 depends on bug 26145, which changed state. Bug 26145 Summary: OpenMP runtime fails to build with -Wdate-time in release builds https://llvm.org/bugs/show_bug.cgi?id=26145 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 26145] OpenMP runtime fails to build with -Wdate-time in release builds
https://llvm.org/bugs/show_bug.cgi?id=26145 Hans Wennborg changed: What|Removed |Added Status|NEW |RESOLVED Resolution|--- |FIXED --- Comment #1 from Hans Wennborg --- r257833 -- 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 21077] R_AARCH64_ABS64 used with TLS symbol
https://llvm.org/bugs/show_bug.cgi?id=21077 Renato Golin changed: What|Removed |Added Status|NEW |RESOLVED CC||renato.go...@linaro.org Resolution|--- |FIXED --- Comment #13 from Renato Golin --- works on trunk with BFD 2.25 -- 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 16855] llvm generates invalid mov command for arm7tdmi target
https://llvm.org/bugs/show_bug.cgi?id=16855 Renato Golin changed: What|Removed |Added Status|NEW |RESOLVED CC||renato.go...@linaro.org Resolution|--- |FIXED --- Comment #3 from Renato Golin --- fixed on trunk -- 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 26128] libcxx fails to compile with newlib for cmath functions
https://llvm.org/bugs/show_bug.cgi?id=26128 Rian Quinn changed: What|Removed |Added Status|NEW |RESOLVED Resolution|--- |INVALID --- Comment #15 from Rian Quinn --- I was able to verify that when using cmake, this does _not_ seem to be an issue. We compiled using: cmake ../libcxx/ \ -DLLVM_PATH=~/clang/llvm \ -DLIBCXX_CXX_ABI=libcxxabi \ -DLIBCXX_CXX_ABI_INCLUDE_PATHS=../libcxxabi/include \ -DCMAKE_INSTALL_PREFIX=$HOME/opt/cross/x86_64-elf/ \ -DLIBCXX_SYSROOT=$HOME/opt/cross/x86_64-elf/ \ -DCMAKE_C_COMPILER=$HOME/opt/cross/bin/x86_64-bareflank-gcc \ -DCMAKE_CXX_COMPILER=$HOME/opt/cross/bin/x86_64-bareflank-g++ \ -DLIBCXX_ENABLE_THREADS=OFF \ -DLIBCXX_ENABLE_MONOTONIC_CLOCK=OFF We had to wrap gcc in a bash script so that we could properly handle compiling vs linking as cmake insists on using g++ for linking instead of ld, but it seems to be working great. I will note that when you compile with GCC, you get a pretty alarming number of warnings. And it doesn't seem to compile with "-fno-exceptions" as it complains about "-Werror=return-type". I'll mark this as resolved, as it seems there is some magic in cmake that is making this error go away. -- 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 26208] Handling of .space is really slow
https://llvm.org/bugs/show_bug.cgi?id=26208 Rafael Ávila de Espíndola changed: What|Removed |Added Status|ASSIGNED|RESOLVED Resolution|--- |FIXED --- Comment #2 from Rafael Ávila de Espíndola --- Fixed in r258149. -- 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 23591] [AArch64] llc crashes with assertion "Illegal shifted immedate value!" on AArch64.
https://llvm.org/bugs/show_bug.cgi?id=23591 Renato Golin changed: What|Removed |Added Status|NEW |RESOLVED CC||renato.go...@linaro.org Resolution|--- |FIXED --- Comment #1 from Renato Golin --- works on trunk -- 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 19136] [ARM] crash while folding SP-update into vpush/vpop
https://llvm.org/bugs/show_bug.cgi?id=19136 Renato Golin changed: What|Removed |Added Status|NEW |RESOLVED CC||renato.go...@linaro.org Resolution|--- |FIXED --- Comment #1 from Renato Golin --- works on trunk -- 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 21988] [MC][Aarch64] -no-exec-stack causes crash
https://llvm.org/bugs/show_bug.cgi?id=21988 Renato Golin changed: What|Removed |Added Status|NEW |RESOLVED CC||renato.go...@linaro.org Resolution|--- |FIXED --- Comment #1 from Renato Golin --- works on trunk -- 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 19972] Bad code generation in post-ra ARMLoadStoreOptimizer pass in thumb1 mode
https://llvm.org/bugs/show_bug.cgi?id=19972
Renato Golin changed:
What|Removed |Added
Status|REOPENED|RESOLVED
CC||renato.go...@linaro.org
Resolution|--- |FIXED
--- Comment #7 from Renato Golin ---
Trunk generates:
push{r7, lr}
addr7, sp, #0
ldmr0!, {r1, r2}
subsr0, #8
blbar
pop{r7, pc}
which, while inefficient (the sub is unnecessary with no write-back), looks
correct.
If people feel strongly about reducing code size for that sequence in Thumb1,
please open a new enhancement bug.
--
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 24931] arm: Assertion failed: (!VRM->hasPhys(SplitVirtReg->reg) && "Register already assigned"), function allocatePhysRegs
https://llvm.org/bugs/show_bug.cgi?id=24931 Renato Golin changed: What|Removed |Added Status|NEW |RESOLVED CC||renato.go...@linaro.org Resolution|--- |FIXED --- Comment #1 from Renato Golin --- works on trunk -- 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 16857] llvm generates pop pc on ARMv4 in thumb mode (broken interworking)
https://llvm.org/bugs/show_bug.cgi?id=16857 Renato Golin changed: What|Removed |Added Status|NEW |RESOLVED CC||renato.go...@linaro.org Resolution|--- |FIXED --- Comment #7 from Renato Golin --- fixed on trunk -- 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 26210] New: clang-cl and msvc-cl disagree about static initialization
https://llvm.org/bugs/show_bug.cgi?id=26210
Bug ID: 26210
Summary: clang-cl and msvc-cl disagree about static
initialization
Product: clang
Version: unspecified
Hardware: PC
OS: All
Status: NEW
Severity: normal
Priority: P
Component: LLVM Codegen
Assignee: unassignedclangb...@nondot.org
Reporter: jmuizel...@mozilla.com
CC: llvm-bugs@lists.llvm.org
Classification: Unclassified
When compiling a function like this:
inline const T *Ops() {
static int k;
static const T sOps = true ? 0 : &k;
return &sOps;
}
msvc puts sOps in a read-write section, while clang puts sOps in a read-only
section. If the code for Ops() is not included in the object file from clang,
the linker can pick the read-only sOps from clang and Ops() implementation from
msvc. When the msvc Ops() runs it will crash trying to write to sOps.
I've put up a repo that reproduces this issue here:
https://github.com/jrmuizel/read-only-symbol-crash/
We're currently trying to bisect some clang related compilation issues and this
is preventing us from linking object files from clang and cl together. I
suspect using the clang front end on top of cl2 backend will also suffer from
this interoperability issue.
--
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 24931] arm: Assertion failed: (!VRM->hasPhys(SplitVirtReg->reg) && "Register already assigned"), function allocatePhysRegs
https://llvm.org/bugs/show_bug.cgi?id=24931 Renato Golin changed: What|Removed |Added Status|RESOLVED|REOPENED Component|new bugs|Backend: ARM Resolution|FIXED |--- Product|new-bugs|libraries --- Comment #3 from Renato Golin --- Ok, I'm reopening the bug. 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 http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 26207] Assertion failure in ScalarEvolution::getBackedgeTakenInfo
https://llvm.org/bugs/show_bug.cgi?id=26207 Sanjoy Das changed: What|Removed |Added Status|NEW |RESOLVED Resolution|--- |FIXED --- Comment #2 from Sanjoy Das --- Should be fixed in r258184 -- 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 26059] [meta] 3.8.0 Release Blockers
https://llvm.org/bugs/show_bug.cgi?id=26059 Bug 26059 depends on bug 26164, which changed state. Bug 26164 Summary: check-libomp fails on darwin https://llvm.org/bugs/show_bug.cgi?id=26164 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 26164] check-libomp fails on darwin
https://llvm.org/bugs/show_bug.cgi?id=26164 Hans Wennborg changed: What|Removed |Added Status|NEW |RESOLVED Resolution|--- |FIXED --- Comment #1 from Hans Wennborg --- The tests pass on my Mac now after r258169 and the fixes for Issue 26136 got merged. -- 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 26211] New: clang-3.9: Assert when re-defining sqrt() function
https://llvm.org/bugs/show_bug.cgi?id=26211
Bug ID: 26211
Summary: clang-3.9: Assert when re-defining sqrt() function
Product: clang
Version: trunk
Hardware: PC
OS: Linux
Status: NEW
Severity: normal
Priority: P
Component: LLVM Codegen
Assignee: unassignedclangb...@nondot.org
Reporter: brzy...@gmail.com
CC: llvm-bugs@lists.llvm.org
Classification: Unclassified
There is a compiler assert when overwriting sqrt(). The output from clang is
the following:
$
/sarc-c/compiler_tmp/tools/cross/llvm-3.9/2016-01-19-18-47-02-3a5bf2b-master/bin/aarch64-sarc-linux-gnu-clang
-O0 main-7c7b0f.c
main-7c7b0f.c:1:5: warning: incompatible redeclaration of library function
'sqrt' [-Wincompatible-library-redeclaration]
int sqrt(int n) {
^
main-7c7b0f.c:1:5: note: 'sqrt' is a builtin with type 'double (double)'
1 warning generated.
brzycki@spa103 ~/x $
/sarc-c/compiler_tmp/tools/cross/llvm-3.9/2016-01-19-18-47-02-3a5bf2b-master/bin/aarch64-sarc-linux-gnu-clang
-Wno-incompatible-library-redeclaration -O0 main-7c7b0f.c
brzycki@spa103 ~/x $
/sarc-c/compiler_tmp/tools/cross/llvm-3.9/2016-01-19-18-47-02-3a5bf2b-master/bin/aarch64-sarc-linux-gnu-clang
-Wno-incompatible-library-redeclaration -O1 main-7c7b0f.c
clang-3.9: /tmp/tmp.pwSURnmdYf.spa189/llvm/lib/IR/Instruction.cpp:152: bool
llvm::Instruction::hasUnsafeAlgebra() const: Assertion
`isa(this) && "getting fast-math flag on invalid op"' failed.
0 clang-3.9 0x02ec6ccb
llvm::sys::PrintStackTrace(llvm::raw_ostream&) + 44
1 clang-3.9 0x02ec7048
2 clang-3.9 0x02ec56d9 llvm::sys::RunSignalHandlers() + 138
3 clang-3.9 0x02ec6718
4 libpthread.so.0 0x003bd240f710
5 libc.so.6 0x003bd1c32925 gsignal + 53
6 libc.so.6 0x003bd1c34105 abort + 373
7 libc.so.6 0x003bd1c2ba4e
8 libc.so.6 0x003bd1c2bb10 __assert_perror_fail + 0
9 clang-3.9 0x029e2068 llvm::Instruction::hasUnsafeAlgebra()
const + 76
10 clang-3.9 0x02f6a1c9
llvm::LibCallSimplifier::optimizeSqrt(llvm::CallInst*, llvm::IRBuilder >&) + 405
11 clang-3.9 0x02f6e6e0
llvm::LibCallSimplifier::optimizeCall(llvm::CallInst*) + 1434
12 clang-3.9 0x02b593a2
13 clang-3.9 0x02b59f70
14 clang-3.9 0x02b53bf0
15 clang-3.9 0x02b2a1c6
16 clang-3.9 0x02b2923f
17 clang-3.9 0x02b27741
18 clang-3.9 0x02b23b30
19 clang-3.9 0x02b24b7d
20 clang-3.9 0x02b24e74
21 clang-3.9 0x02a1a24b
llvm::FPPassManager::runOnFunction(llvm::Function&) + 315
22 clang-3.9 0x02a1a3c2
llvm::FPPassManager::runOnModule(llvm::Module&) + 118
23 clang-3.9 0x02a1a70f
24 clang-3.9 0x02a1addd
llvm::legacy::PassManagerImpl::run(llvm::Module&) + 271
25 clang-3.9 0x02a1afe9
llvm::legacy::PassManager::run(llvm::Module&) + 39
26 clang-3.9 0x030af67a
27 clang-3.9 0x030af7ab
clang::EmitBackendOutput(clang::DiagnosticsEngine&, clang::CodeGenOptions
const&, clang::TargetOptions const&, clang::LangOptions const&,
llvm::StringRef, llvm::Module*, clang::BackendAction, llvm::raw_pwrite_stream*)
+ 127
28 clang-3.9 0x03994c3c
29 clang-3.9 0x041e8d51 clang::ParseAST(clang::Sema&, bool, bool)
+ 827
30 clang-3.9 0x0354380f clang::ASTFrontendAction::ExecuteAction()
+ 323
31 clang-3.9 0x03997832 clang::CodeGenAction::ExecuteAction() +
1722
32 clang-3.9 0x035432a5 clang::FrontendAction::Execute() + 135
33 clang-3.9 0x034fc58f
clang::CompilerInstance::ExecuteAction(clang::FrontendAction&) + 1109
34 clang-3.9 0x0363af4e
clang::ExecuteCompilerInvocation(clang::CompilerInstance*) + 1160
35 clang-3.9 0x016b88d6 cc1_main(llvm::ArrayRef,
char const*, void*) + 1019
36 clang-3.9 0x016af6c5
37 clang-3.9 0x016b014d main + 1965
38 libc.so.6 0x003bd1c1ed1d __libc_start_main + 253
39 clang-3.9 0x016ab9c9
Stack dump:
0. Program arguments:
/sarc-c/compiler_tmp/tools/cross/llvm-3.9/2016-01-19-18-47-02-3a5bf2b-master/bin/clang-3.9
-cc1 -triple aarch64-sarc-linux-gnu -emit-obj -disable-free -main-file-name
main-7c7b0f.c -mrelocation-model static -mthread-model posix -mdisable-fp-elim
-fmath-errno -masm-verbose -mconstructor-aliases -fuse-init-array -target-cpu
generic -target-feature +neon -target-abi aapcs -dwarf-column-info
-debugger-tuning=gdb -resource-dir
/sarc-c/compiler_tmp/tools/cross/llvm-3.9/2016-01-19-18-47-02-3a5bf2b-master/bin/../lib/clang/3.9.0
-isysroot
/sarc-c/compiler_tmp/tools/cross/aarch64/sysroot-llvm/aarch64-sarc-linux-gnu/sysroot
-internal-isystem
/sarc-c/compiler_tmp/tools/cross/aarch64/sysroot-llvm/aarch64-sarc-lin
[llvm-bugs] [Bug 25980] Some files in libc++/test are missing license headers
https://llvm.org/bugs/show_bug.cgi?id=25980 Eric Fiselier changed: What|Removed |Added Status|ASSIGNED|RESOLVED Resolution|--- |FIXED --- Comment #5 from Eric Fiselier --- Added in r258196. -- 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 26212] New: LLVM Windows 64 bit installer overrides values from the 32 bit installer
https://llvm.org/bugs/show_bug.cgi?id=26212 Bug ID: 26212 Summary: LLVM Windows 64 bit installer overrides values from the 32 bit installer Product: Packaging Version: unspecified Hardware: PC OS: Windows NT Status: NEW Severity: normal Priority: P Component: Windows Installer Assignee: unassignedb...@nondot.org Reporter: cristian.a...@gmail.com CC: llvm-bugs@lists.llvm.org Classification: Unclassified I have downloaded the official LLVM 3.7.1 32 and 64 bit windows installers. Environment: Windows 10 64 bit. First I have installed the 32 bit LLVM 3.7.1. The following registry key HKEY_LOCAL_MACHINE\SOFTWARE\WOW6432Node\LLVM\LLVM had the default value set to: "C:\Program Files (x86)\LLVM" Then I have installed the 64 bit LLVM 3.7.1. The first thing that popped up was this dialog: --- LLVM Setup --- LLVM is already installed. Do you want to uninstall the old version before installing the new one? --- Yes No Cancel --- I have clicked on "No". The registry key HKEY_LOCAL_MACHINE\SOFTWARE\WOW6432Node\LLVM\LLVM defaul t value was changed to "C:\Program Files\LLVM". Please make sure that the 64 bit installer checks and writes to the 64 bit registry "HKEY_LOCAL_MACHINE\SOFTWARE\LLVM\LLVM". These paths are needed by tools to query the existence of LLVM on Windows. For more information see: http://nsis.sourceforge.net/Reference/SetRegView and the x64.nsh NSIS include http://forums.winamp.com/showthread.php?threadid=301724 -- 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 25976] A bunch of new files in libcxx and libcxxabi are missing license headers
https://llvm.org/bugs/show_bug.cgi?id=25976 Eric Fiselier changed: What|Removed |Added Status|NEW |RESOLVED Resolution|--- |FIXED --- Comment #3 from Eric Fiselier --- Added libc++abi headers in r258201. -- 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 26136] segfault during Register Coalescing
https://llvm.org/bugs/show_bug.cgi?id=26136 Hans Wennborg 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 26212] LLVM Windows 64 bit installer overrides values from the 32 bit installer
https://llvm.org/bugs/show_bug.cgi?id=26212 Hans Wennborg changed: What|Removed |Added Status|NEW |RESOLVED CC||h...@chromium.org Resolution|--- |WONTFIX --- Comment #3 from Hans Wennborg --- This behaviour is intentional. The installer and Visual Studio integration are only set up to handle one installed LLVM version at the time. -- 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 26213] New: Assertion at construction of local struct object using inherited constructor from template struct base
https://llvm.org/bugs/show_bug.cgi?id=26213
Bug ID: 26213
Summary: Assertion at construction of local struct object using
inherited constructor from template struct base
Product: new-bugs
Version: 3.7
Hardware: Macintosh
OS: Linux
Status: NEW
Severity: normal
Priority: P
Component: new bugs
Assignee: unassignedb...@nondot.org
Reporter: kyle.mark...@intel.com
CC: llvm-bugs@lists.llvm.org
Classification: Unclassified
Created attachment 15669
--> https://llvm.org/bugs/attachment.cgi?id=15669&action=edit
Preprocessed source
Assertion: clang/lib/Sema/TreeTransform.h:4726: bool
clang::TreeTransform::TransformExceptionSpec(clang::SourceLocation,
clang::FunctionProtoType::ExceptionSpecInfo&,
llvm::SmallVectorImpl&, bool&) [with Derived =
{anonymous}::TemplateInstantiator]: Assertion `ESI.Type != EST_Uninstantiated
&& ESI.Type != EST_Unevaluated' failed.
clang version 3.7.0 (tags/RELEASE_370/final)
Target: x86_64-unknown-linux-gnu
Thread model: posix
clang: note: diagnostic msg: PLEASE submit a bug report to
http://llvm.org/bugs/ and include the crash backtrace, preprocessed source, and
associated run script.
Operating system is:
SUSE Linux Enterprise Server 11 (x86_64)
VERSION = 11
PATCHLEVEL = 2
This is the offending minimized source (bad.cpp):
template
struct A {
T value;
template A(U u) : value(u) { }
};
void foo() {
struct B : A {
using A::A;
};
B(0);
}
Preprocessed source is attached. I cannot attach the run script because it
contains path names that are potentially sensitive, but the command was:
clang++ -std=c++11 bad.cpp -c
Backtrace:
0 libLLVM-3.7.so 0x765a6315
llvm::sys::PrintStackTrace(llvm::raw_ostream&) + 37
1 libLLVM-3.7.so 0x765a5e83
2 libpthread.so.0 0x74916810
3 libc.so.6 0x737db755 gsignal + 53
4 libc.so.6 0x737dcd31 abort + 385
5 libc.so.6 0x737d4610 __assert_fail + 240
6 clang 0x00ecdbc6
7 clang 0x00ecdd3a
clang::Sema::SubstExceptionSpec(clang::FunctionDecl*, clang::FunctionProtoType
const*, clang::MultiLevelTemplateArgumentList const&) + 298
8 clang 0x00ee76af
clang::TemplateDeclInstantiator::InitFunctionInstantiation(clang::FunctionDecl*,
clang::FunctionDecl*) + 895
9 clang 0x00ee79a1
clang::TemplateDeclInstantiator::InitMethodInstantiation(clang::CXXMethodDecl*,
clang::CXXMethodDecl*) + 17
10 clang 0x00ef42ac
clang::TemplateDeclInstantiator::VisitCXXMethodDecl(clang::CXXMethodDecl*,
clang::TemplateParameterList*, bool) + 2652
11 clang 0x00eead3c clang::Sema::SubstDecl(clang::Decl*,
clang::DeclContext*, clang::MultiLevelTemplateArgumentList const&) + 140
12 clang 0x00e8b3cd
clang::Sema::FinishTemplateArgumentDeduction(clang::FunctionTemplateDecl*,
llvm::SmallVectorImpl&, unsigned int,
clang::FunctionDecl*&, clang::sema::TemplateDeductionInfo&,
llvm::SmallVectorImpl const*, bool) + 2397
13 clang 0x00e9c8d3
clang::Sema::DeduceTemplateArguments(clang::FunctionTemplateDecl*,
clang::TemplateArgumentListInfo*, llvm::ArrayRef,
clang::FunctionDecl*&, clang::sema::TemplateDeductionInfo&, bool) + 915
14 clang 0x00dd3510
clang::Sema::AddTemplateOverloadCandidate(clang::FunctionTemplateDecl*,
clang::DeclAccessPair, clang::TemplateArgumentListInfo*,
llvm::ArrayRef, clang::OverloadCandidateSet&, bool, bool) + 400
15 clang 0x00d2f0dd
16 clang 0x00d37636
17 clang 0x00d408d6
clang::InitializationSequence::InitializeFrom(clang::Sema&,
clang::InitializedEntity const&, clang::InitializationKind const&,
llvm::MutableArrayRef, bool) + 2742
18 clang 0x00af8850
19 clang 0x00afb70a
20 clang 0x00afdb1b
21 clang 0x00afdda9
clang::Sema::BuildCXXFunctionalCastExpr(clang::TypeSourceInfo*,
clang::SourceLocation, clang::Expr*, clang::SourceLocation) + 185
22 clang 0x00ccceed
clang::Sema::BuildCXXTypeConstructExpr(clang::TypeSourceInfo*,
clang::SourceLocation, llvm::MutableArrayRef,
clang::SourceLocation) + 893
23 clang 0x00ccd304
clang::Sema::ActOnCXXTypeConstructExpr(clang::OpaquePtr,
clang::SourceLocation, llvm::MutableArrayRef,
clang::SourceLocation) + 84
24 clang 0x00a6685e
clang::Parser::ParseCXXTypeConstructExpression(clang::DeclSpec const&) + 1726
25 clang 0x00a5a18f clang::Parser::ParseCastExpression(bool,
bool, bool&, clang::Parser::TypeCastState) + 3535
26 clang 0x00a5b84d clang::Parser::ParseCastExpression(bool,
bool, clang::Parser::TypeCastState) + 29
27 clang 0x00a5cacf
clang::Parser::ParseAssignmentExpression(clang::Parser::TypeCastState) + 3
[llvm-bugs] [Bug 26107] Shrink wrapping breaks llgo on X86-64
https://llvm.org/bugs/show_bug.cgi?id=26107 Quentin Colombet changed: What|Removed |Added Status|NEW |RESOLVED Resolution|--- |FIXED --- Comment #9 from Quentin Colombet --- I have disabled shrink-wrapping for functions that have the split-stack attribute or the HiPE calling convention in revision 258221. The targeting of both callbacks is not ready to handle a prologue block that is not the entry block of the function. Fixing that is probably easy (we need to insert the new blocks right in front of the prologue instead of the begin of the function and patch the predecessors of the prologue block to adjust the jump to the new block) but I don’t have time to do it right now. -- 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 26214] New: IR verifier slows down linking Chrome with LTO by 2.5x
https://llvm.org/bugs/show_bug.cgi?id=26214
Bug ID: 26214
Summary: IR verifier slows down linking Chrome with LTO by 2.5x
Product: new-bugs
Version: unspecified
Hardware: PC
OS: Linux
Status: NEW
Severity: normal
Priority: P
Component: new bugs
Assignee: unassignedb...@nondot.org
Reporter: kra...@google.com
CC: llvm-bugs@lists.llvm.org
Classification: Unclassified
IR verifier added in http://reviews.llvm.org/rL257825 and
http://reviews.llvm.org/rL257823 significantly slows down LTO build. The larger
the target, the higher ratio of the slowdown. Linking Chrome with LTO is 27
mins before the changes, and 88 minutes after them.
Here is the profile:
+ 40.50%ld.gold LLVMgold.so [.] bool llvm::function_ref::callback_fn<(anonymous
namespace)::Verifier::visitGlobalValue(llvm::GlobalValue
const&)::{lambda(llvm::Valu
+ 19.61%ld.gold LLVMgold.so [.] llvm::Use::getImpliedUser()
const
+ 7.20%ld.gold LLVMgold.so [.]
llvm::IRMover::move(llvm::Module&, llvm::ArrayRef,
std::function)>, llvm
+ 2.63%ld.gold LLVMgold.so [.]
llvm::SmallPtrSetImplBase::FindBucketFor(void const*) const
+ 1.92%ld.gold [kernel.kallsyms][k] 0x81180fa0
+ 1.60%ld.gold LLVMgold.so [.] forEachUser(llvm::Value
const*, llvm::SmallPtrSet&, llvm::function_ref) [clone .constprop.1086]
+ 1.58%ld.gold libc-2.19.so [.] _int_malloc
+ 1.18%ld.gold LLVMgold.so [.] llvm::Use::getUser() const
+ 0.82%ld.gold LLVMgold.so [.]
llvm::SmallPtrSetImplBase::insert_imp(void const*)
+ 0.76%ld.gold LLVMgold.so [.]
llvm::NamedMDNode::getOperand(unsigned int) const
This regression is severe for our use case (deploying Control Flow Integrity in
Chrome), but I don't think it's limited just to us. Anyone who uses LTO is
affected.
--
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 23937] AVX512: scalar FMA 132/231 instructions are not supported
https://llvm.org/bugs/show_bug.cgi?id=23937 Hal Finkel changed: What|Removed |Added Status|NEW |RESOLVED CC||hfin...@anl.gov Resolution|--- |FIXED --- Comment #1 from Hal Finkel --- This works on trunk: $ echo '0x62 0xf2 0xf5 0x08 0xa9 0xc2' | llvm-mc -triple x86_64-unknown-unknown -mcpu=knl -disassemble .text vfmadd213sd%xmm2, %xmm1, %xmm0 $ echo '0x62 0xf2 0xf5 0x08 0xb9 0xc2' | llvm-mc -triple x86_64-unknown-unknown -mcpu=knl -disassemble .text vfmadd231sd%xmm2, %xmm1, %xmm0 $ echo '0x62 0xf2 0xf5 0x08 0x99 0xc2' | llvm-mc -triple x86_64-unknown-unknown -mcpu=knl -disassemble .text vfmadd132sd%xmm2, %xmm1, %xmm0 -- 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 26215] New: clang-format should insert braces for conditionals with a multi-line body
https://llvm.org/bugs/show_bug.cgi?id=26215
Bug ID: 26215
Summary: clang-format should insert braces for conditionals
with a multi-line body
Product: clang
Version: unspecified
Hardware: PC
OS: Linux
Status: NEW
Severity: normal
Priority: P
Component: Formatter
Assignee: unassignedclangb...@nondot.org
Reporter: dch...@google.com
CC: djas...@google.com, kli...@google.com,
llvm-bugs@lists.llvm.org
Classification: Unclassified
>From http://google.github.io/styleguide/cppguide.html#Conditionals:
"In general, curly braces are not required for single-line statements, but they
are allowed if you like them; conditional or loop statements with complex
conditions or statements may be more readable with curly braces. Some projects
require that an if must always always have an accompanying brace."
The implication here seems to be that any conditional with a multi-line body
should use braces.
Right now, clang-format won't make any formatting changes to this snippet:
if (some_condition)
some_very_long_variable_name =
SomeVeryLongFunctionName(some_very_long_parameter_name);
But I believe that the style guide implies that properly formatted code should
look like this:
if (some_condition) {
some_very_long_variable_name =
SomeVeryLongFunctionName(some_very_long_parameter_name);
}
--
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 26216] New: compiling w -fno-exceptions generates errors on "throw" in templates that are never instantiated
https://llvm.org/bugs/show_bug.cgi?id=26216
Bug ID: 26216
Summary: compiling w -fno-exceptions generates errors on
"throw" in templates that are never instantiated
Product: clang
Version: trunk
Hardware: Macintosh
OS: All
Status: NEW
Severity: normal
Priority: P
Component: C++
Assignee: unassignedclangb...@nondot.org
Reporter: mclow.li...@gmail.com
CC: dgre...@apple.com, llvm-bugs@lists.llvm.org
Classification: Unclassified
I ran into this in the libc++ test suite.
If -fno-exceptions is defined, the following code compiles w/o error using GCC,
but gets an error with clang.
My expectation matches gcc's behavior; no error should be generated
unless/until the template is instantiated.
Reproducer:
#include
#include
template
struct wrap {
wrap() : t_() {}
wrap(const T& t) : t_(t) {}
T & get() { return t_; }
void toss () { throw std::runtime_error("Toss"); }
T t_;
};
int main() {
std::cout << "Hello World" << std::endl;
}
--
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 26147] Verify error: Unwind edges out of a funclet pad must have the same unwind dest
https://llvm.org/bugs/show_bug.cgi?id=26147 Joseph Tremoulet changed: What|Removed |Added Status|ASSIGNED|RESOLVED Resolution|--- |FIXED --- Comment #9 from Joseph Tremoulet --- Fix committed as r258273. Hans: If I'm understanding http://llvm.org/docs/HowToReleaseLLVM.html#release-timeline correctly, it should just get picked up for 3.8? Please let me know if there's something else I need to do for this w.r.t. the release. -- 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 26059] [meta] 3.8.0 Release Blockers
https://llvm.org/bugs/show_bug.cgi?id=26059 Bug 26059 depends on bug 26147, which changed state. Bug 26147 Summary: Verify error: Unwind edges out of a funclet pad must have the same unwind dest https://llvm.org/bugs/show_bug.cgi?id=26147 What|Removed |Added Status|ASSIGNED|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 20211] libc++ is not compatible with musl libc library
https://llvm.org/bugs/show_bug.cgi?id=20211 Eric Fiselier changed: What|Removed |Added Status|ASSIGNED|RESOLVED Resolution|--- |FIXED Assignee|mclow.li...@gmail.com |e...@efcs.ca --- Comment #7 from Eric Fiselier --- libc++ can now build and install for MUSL libc, but it requires that libc++ is built with the CMake option `-DLIBCXX_HAS_MUSL_LIBC`. This option ensures that a modified "__config" header is installed that defines `_LIBCPP_HAS_MUSL_LIBC`. See http://libcxx.llvm.org/docs/DesignDocs/CapturingConfigInfo.html for more information. -- 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 15685] OpenCL 'char' is not signed
https://llvm.org/bugs/show_bug.cgi?id=15685 Igor Chesnokov changed: What|Removed |Added Status|NEW |RESOLVED Resolution|--- |WORKSFORME --- Comment #1 from Igor Chesnokov --- This is already fixed in Clang 3.9.0. "@foo = constant i32 -1, align 4" output appears for platforms x86, ppc32, ppc64, ppc64le, x86, x86-64 targets. In 3.9.0 the sample requires to define foo as "constant in foo". -- 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
