[llvm-bugs] [Bug 46247] New: [Simple Register Coalescing] Assertion `VNI && "No live value at use."' failed.
https://bugs.llvm.org/show_bug.cgi?id=46247 Bug ID: 46247 Summary: [Simple Register Coalescing] Assertion `VNI && "No live value at use."' failed. Product: libraries Version: trunk Hardware: PC OS: Linux Status: NEW Severity: normal Priority: P Component: Register Allocator Assignee: unassignedb...@nondot.org Reporter: pauls...@linux.vnet.ibm.com CC: llvm-bugs@lists.llvm.org, quentin.colom...@gmail.com Created attachment 23592 --> https://bugs.llvm.org/attachment.cgi?id=23592&action=edit reduced testcase I found a test case which which triggered this assert with machine DCE disabled. In the reduced test case, it seems that the whole block is in fact dead and it would normally become an empty block after Machine DCE. However, if run with llc -mtriple=s390x-linux-gnu -mcpu=z15 -O3 -o out.s tc_nolivevalatuse.ll -disable-machine-dce -misched=shuffle , the RegisterCoalescer deletes a lot of dead defs, but not all of them. Three instructions remain, and the mischeduler will then be run and trigger the assert. It seems that the RegisterCoalescer is not updating data structures correctly while deleting instructions. llvm-project/llvm/lib/CodeGen/MachineScheduler.cpp:1144: void llvm::ScheduleDAGMILive::updatePressureDiffs(llvm::ArrayRef): Assertion `VNI && "No live value at use."' failed. ... #8 0x02aa24927b36 llvm::ScheduleDAGMILive::updatePressureDiffs ... -- 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 46248] New: `-fmodules-local-submodule-visibility` does not work correctly with Objective-C
https://bugs.llvm.org/show_bug.cgi?id=46248
Bug ID: 46248
Summary: `-fmodules-local-submodule-visibility` does not work
correctly with Objective-C
Product: clang
Version: trunk
Hardware: All
OS: All
Status: NEW
Severity: enhancement
Priority: P
Component: Frontend
Assignee: unassignedclangb...@nondot.org
Reporter: mboe...@google.com
CC: llvm-bugs@lists.llvm.org, neeil...@live.com,
richard-l...@metafoo.co.uk
Multiple places in AST/DeclObjC.{h,cpp} and Sema/SemaDeclObjC.cpp use
Decl::isHidden() instead of Sema::isVisible().
These two methods are equivalent only if -fmodules-local-submodule-visibility
is not turned on.
If -fmodules-local-submodule-visibility is turned on, Decl::isHidden() can
return true for declarations that have been made visible by importing the
(sub)module that owns them. (See also the FIXME in the comment for
Decl::isHidden().) As a result, the Objective-C code using Decl::isHidden()
will falsely assume that such declarations are hidden when in fact they are
visible.
For examples of how this causes wrong behavior, witness how adding the
-fmodules-local-submodule-visibility option to tests objc-categories.m,
objc-category-2.m and objc-category-3.m in test/Modules causes these tests to
fail.
Fixing this in Sema/SemaDeclObjC.cpp is easy, as Decl::isHidden() can simply be
replaced by Sema::isVisible().
Fixing this in AST/DeclObjC.{h,cpp} is more difficult as AST is not allowed to
depend on Sema. As a result, the functions in question and all of their callers
need to be moved to Sema. From what I have seen, this is a non-trivial
undertaking.
For more context, see also description on this change:
https://github.com/llvm/llvm-project/commit/90dc5254725e32c0fab65213c554b689a2a0bbbf
Also note this in-review change, which changes the name of Decl::isHidden() to
more accurately reflect what it does:
https://reviews.llvm.org/D81392
--
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 46249] New: [X86] VPSHUFD and VPERMQ with immediates combined into VPERMD with variable mask
https://bugs.llvm.org/show_bug.cgi?id=46249 Bug ID: 46249 Summary: [X86] VPSHUFD and VPERMQ with immediates combined into VPERMD with variable mask Product: libraries Version: trunk Hardware: PC OS: All Status: NEW Severity: enhancement Priority: P Component: Backend: X86 Assignee: unassignedb...@nondot.org Reporter: craig.top...@gmail.com CC: craig.top...@gmail.com, llvm-bugs@lists.llvm.org, llvm-...@redking.me.uk, spatel+l...@rotateright.com As noted here https://twitter.com/damageboy/status/1270266352406802432 We combined t45: v16i32 = vector_shuffle<3,2,1,0,7,6,5,4,11,10,9,8,15,14,13,12> t44, undef:v16i32 t46: v16i32 = vector_shuffle<8,9,10,11,12,13,14,15,0,1,2,3,4,5,6,7> t45, undef:v16i32 into t258: v16i32 = vector_shuffle<11,10,9,8,15,14,13,12,3,2,1,0,7,6,5,4> Which then required vpermd. This happened before lowering and lowering doesn't know to recreate the two shuffles. Simon, would hasFastVariableShuffle() also have created a VPERMD if lowering had managed to make 2 immediate shuffles? -- 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 46250] New: Random `parser end of file`
https://bugs.llvm.org/show_bug.cgi?id=46250 Bug ID: 46250 Summary: Random `parser end of file` Product: clang Version: 10.0 Hardware: PC OS: other Status: NEW Severity: normal Priority: P Component: -New Bugs Assignee: unassignedclangb...@nondot.org Reporter: ste...@profanter.me CC: htmldevelo...@gmail.com, llvm-bugs@lists.llvm.org, neeil...@live.com, richard-l...@metafoo.co.uk -- 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 46250] Random `parser end of file`
https://bugs.llvm.org/show_bug.cgi?id=46250 Stefan Profanter changed: What|Removed |Added Resolution|--- |INVALID Status|NEW |RESOLVED --- Comment #1 from Stefan Profanter --- Sorry, accidentally clicked on submit. Please ignore this! -- 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 42986] Load/store combine fails to merge complex address math
https://bugs.llvm.org/show_bug.cgi?id=42986 Simon Pilgrim changed: What|Removed |Added Resolution|--- |WONTFIX Status|NEW |RESOLVED --- Comment #4 from Simon Pilgrim --- Changing the offsets to size_t avoids the problem with overflow semantics. -- 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 46251] New: [Matrix] Implement literal initializers for matrix types.
https://bugs.llvm.org/show_bug.cgi?id=46251 Bug ID: 46251 Summary: [Matrix] Implement literal initializers for matrix types. Product: clang Version: unspecified Hardware: PC OS: All Status: NEW Severity: enhancement Priority: P Component: C++ Assignee: unassignedclangb...@nondot.org Reporter: florian_h...@apple.com CC: blitzrak...@gmail.com, dgre...@apple.com, erik.pilking...@gmail.com, llvm-bugs@lists.llvm.org, richard-l...@metafoo.co.uk Define and implement initialization syntax for matrix types (as mentioned in the TODO section: https://clang.llvm.org/docs/MatrixTypes.html#todos) Once this is implemented, the commented out test in https://github.com/llvm/llvm-project/blob/3323a628ec821b8b75d3b60bf1510931f97d3883/clang/test/CodeGenCXX/matrix-type-builtins.cpp#L78 should be enabled. -- 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 45886] Wrong variable value change during debugging at Og
https://bugs.llvm.org/show_bug.cgi?id=45886 Jan Kratochvil changed: What|Removed |Added Fixed By Commit(s)||4515d35f5c9e2bdcdbcb9fe9028 ||d4a84871958c5 Status|CONFIRMED |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 46252] New: New CodeGen/AMDGPU/inline-constraints.ll tests fail on x86 Linux
https://bugs.llvm.org/show_bug.cgi?id=46252
Bug ID: 46252
Summary: New CodeGen/AMDGPU/inline-constraints.ll tests fail on
x86 Linux
Product: new-bugs
Version: trunk
Hardware: PC
OS: Linux
Status: NEW
Severity: normal
Priority: P
Component: new bugs
Assignee: unassignedb...@nondot.org
Reporter: lukebe...@hotmail.com
CC: htmldevelo...@gmail.com, llvm-bugs@lists.llvm.org
After
https://github.com/llvm/llvm-project/commit/b087b91c917087bc53d47282a16ee4af78bfe286
[AMDGPU] Added 'A' constraint for inline assembler
the inline-constraints.ll lit test is failing on x86 Linux. I discovered this
with Arch 32, gcc 9.3.0.
Here is the build log:
llvm-project/llvm/test/CodeGen/AMDGPU/inline-constraints.ll:178:8: error: GCN:
expected string not found in input
; GCN: v_mov_b32 {{v[0-9]+}}, 0xbf00
^
:690:25: note: scanning from here
inline_A_constraint_F2: ; @inline_A_constraint_F2
^
:694:2: note: possible intended match here
v_mov_b32 v0, 0x5f24720bf00
^
/llvm-project/llvm/test/CodeGen/AMDGPU/inline-constraints.ll:185:8: error: GCN:
expected string not found in input
; GCN: v_mov_b32 {{v[0-9]+}}, 0x4000
^
:722:25: note: scanning from here
inline_A_constraint_F3: ; @inline_A_constraint_F3
^
:726:2: note: possible intended match here
v_mov_b32 v0, 0x5f247204000
^
/llvm-project/llvm/test/CodeGen/AMDGPU/inline-constraints.ll:192:8: error: GCN:
expected string not found in input
; GCN: v_mov_b32 {{v[0-9]+}}, 0xc080
^
:754:25: note: scanning from here
inline_A_constraint_F4: ; @inline_A_constraint_F4
^
:758:2: note: possible intended match here
v_mov_b32 v0, 0x5f24720c080
^
/llvm-project/llvm/test/CodeGen/AMDGPU/inline-constraints.ll:207:8: error: GCN:
expected string not found in input
; GCN: v_mov_b32 {{v[0-9]+}}, 0x3f00
^
:815:25: note: scanning from here
inline_A_constraint_F6: ; @inline_A_constraint_F6
^
:819:2: note: possible intended match here
v_mov_b32 v0, 0x5f247203f00
^
--
Testing: 0.. 10.. 20.. 30.. 40.. 50.. 60.. 70.. 80.. 90..
Failed Tests (1):
LLVM :: CodeGen/AMDGPU/inline-constraints.ll
--
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 22790 in oss-fuzz: llvm:clang-fuzzer: Stack-overflow in ComplexExprEvaluator::VisitBinaryOperator
Updates: Labels: ClusterFuzz-Verified Status: Verified Comment #1 on issue 22790 by ClusterFuzz-External: llvm:clang-fuzzer: Stack-overflow in ComplexExprEvaluator::VisitBinaryOperator https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=22790#c1 ClusterFuzz testcase 4809915205156864 is verified as fixed in https://oss-fuzz.com/revisions?job=libfuzzer_asan_llvm&range=202006080300:202006090257 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 46253] New: malloc analyzer check crashes
https://bugs.llvm.org/show_bug.cgi?id=46253
Bug ID: 46253
Summary: malloc analyzer check crashes
Product: clang
Version: trunk
Hardware: PC
OS: Linux
Status: NEW
Severity: enhancement
Priority: P
Component: Static Analyzer
Assignee: dcough...@apple.com
Reporter: hok...@google.com
CC: dcough...@apple.com, llvm-bugs@lists.llvm.org
./clang-tidy -checks="-*,clang-analyzer*" /tmp/crash.cpp
crash.cpp:
class a {
void realloc(int, bool = false) { realloc(1); }
};
assert.h assertion failed at
llvm-project/clang/lib/StaticAnalyzer/Core/SValBuilder.cpp:427 in
clang::ento::SVal
clang::ento::SValBuilder::evalBinOp(clang::ento::ProgramStateRef,
BinaryOperator::Opcode, clang::ento::SVal, clang::ento::SVal, clang::QualType):
op == BO_Add
*** Check failure stack trace: ***
@ 0x564aecc437c6 __assert_fail
@ 0x564aeb336cbf clang::ento::SValBuilder::evalBinOp()
@ 0x564aeb336efc clang::ento::SValBuilder::evalEQ()
@ 0x564aeb13e777 (anonymous namespace)::MallocChecker::ReallocMemAux()
@ 0x564aeb135ba2 (anonymous namespace)::MallocChecker::checkRealloc()
@ 0x564aeb1454bd clang::ento::check::PostCall::_checkCall<>()
@ 0x564aeb29b3f2
clang::ento::CheckerManager::runCheckersForCallEvent()
@ 0x564aeb2e067e clang::ento::ExprEngine::evalCall()
@ 0x564aeb2e0333 clang::ento::ExprEngine::VisitCallExpr()
@ 0x564aeb2bfe8d clang::ento::ExprEngine::Visit()
@ 0x564aeb2bb96c clang::ento::ExprEngine::ProcessStmt()
@ 0x564aeb2bb648 clang::ento::ExprEngine::processCFGElement()
--
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 25078] [Analyzer] Assertion "System is over constrained" after truncating 64 bits integers to 32 bits.
https://bugs.llvm.org/show_bug.cgi?id=25078 pierre gousseau changed: What|Removed |Added Resolution|--- |FIXED Status|NEW |RESOLVED --- Comment #3 from pierre gousseau --- Thanks Denys for checking, yes closing as fixed 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 https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
[llvm-bugs] [Bug 46254] New: [clang-format] BreakBeforeBraces: Whitesmiths ; In a ‘switch’ block, the ‘default’ case is always missing and indentation level.
https://bugs.llvm.org/show_bug.cgi?id=46254
Bug ID: 46254
Summary: [clang-format] BreakBeforeBraces: Whitesmiths ; In a
‘switch’ block, the ‘default’ case is always missing
and indentation level.
Product: clang
Version: trunk
Hardware: PC
OS: Windows NT
Status: NEW
Severity: release blocker
Priority: P
Component: Formatter
Assignee: unassignedclangb...@nondot.org
Reporter: smass...@matrox.com
CC: djas...@google.com, kli...@google.com,
llvm-bugs@lists.llvm.org
Bug found in upcoming llvm 11.0.
https://prereleases.llvm.org/win-snapshots/ClangFormat-2663a25f.vsix
In ‘clang-format’, when using "BreakBeforeBraces: Whitesmiths".
In a ‘switch’ block, the ‘default’ case is always missing and indentation
level. See below.
(It seems related to: https://reviews.llvm.org/D67627)
switch(x)
{
case 0:
{
foo(x + 1);
}
break;
case 1:
{
goo(x % 42);
}
break;
default:
{
printf("the default case");
}
break;
}
--
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 46255] New: Anonymous structs/unions can violate "Last member element" constraint on flexible arrays
https://bugs.llvm.org/show_bug.cgi?id=46255
Bug ID: 46255
Summary: Anonymous structs/unions can violate "Last member
element" constraint on flexible arrays
Product: clang
Version: trunk
Hardware: PC
OS: Linux
Status: NEW
Severity: normal
Priority: P
Component: C
Assignee: unassignedclangb...@nondot.org
Reporter: vaib...@linux.ibm.com
CC: blitzrak...@gmail.com, dgre...@apple.com,
erik.pilking...@gmail.com, llvm-bugs@lists.llvm.org,
richard-l...@metafoo.co.uk
Description:
In 'C' mode defining a struct containing 2 or more structs having flexible
arrays should result in 'gnu-variable-sized-type-not-at-end' warning getting
triggered as flexible arrays should always be at the end of the containing
struct.
However if the member struct elements are wrapped in an anonymous union or a
struct then clang erroneously ignores this violation.
Example:
Consider following struct definitions ( https://godbolt.org/z/7nnBGz ):
struct A { int a; int b[]; };
struct B { int a; int b[]; };
struct C {
/* warning: field 'a' with variable sized type 'struct A' not at
* the end of a struct or class is a GNU extension
* [-Werror,-Wgnu-variable-sized-type-not-at-end]
*/
struct A a;
struct B b;
};
struct C_U {
/* No warning reported */
struct A a;
union {
struct B b;
};
};
struct C_S {
/* No warning reported */
struct A a;
struct {
struct B b;
};
};
struct C_SU {
/* No warning reported */
struct A a;
struct {
struct B b;
};
union {
struct A aa;
};
};
Exception:
This issue is not observed in 'C++' mode ( https://godbolt.org/z/3nNzap )
--
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 46078] out-of-tree build can't find runtime/config.h
https://bugs.llvm.org/show_bug.cgi?id=46078 Tim Keith changed: What|Removed |Added Assignee|unassignedb...@nondot.org |isu...@gmail.com Status|CONFIRMED |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 46256] New: [coroutines] coroutines with potentially-throwing 'co_await promise.final_suspend()' expressions should be ill-formed
https://bugs.llvm.org/show_bug.cgi?id=46256 Bug ID: 46256 Summary: [coroutines] coroutines with potentially-throwing 'co_await promise.final_suspend()' expressions should be ill-formed Product: clang Version: trunk Hardware: All OS: All Status: NEW Severity: normal Priority: P Component: C++2a Assignee: unassignedclangb...@nondot.org Reporter: lewissba...@gmail.com CC: blitzrak...@gmail.com, erik.pilking...@gmail.com, llvm-bugs@lists.llvm.org, richard-l...@metafoo.co.uk The wording of [dcl.fct.def.coroutine]/15 states: > The expression co_await promise.final_suspend() shall not be > potentially-throwing ([except.spec]). See http://eel.is/c++draft/dcl.fct.def.coroutine#15 and http://eel.is/c++draft/except.spec#6 I believe this means that the compiler should reject with a diagnostic any coroutine with a promise_type where `co_await promise.final_suspend()` is not noexcept. ie. all of the following must be declared noexcept (if they form part of the await-expression): - promise_type::final_suspend() - finalSuspendObj.operator co_await() - finalSuspendAwaiter.await_ready() - finalSuspendAwaiter.await_suspend() - finalSuspendAwaiter.await_resume() - finalSuspedObj destructor - finalSuspendAwaiter destructor Clang (and GCC) currently accepts code that that does not declare final_suspend() as noexcept. e.g. see https://godbolt.org/z/Et_5eo -- 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 45594] lld resolves reloc to sym in discarded section to 0 if sym_name < originating_sym_name
https://bugs.llvm.org/show_bug.cgi?id=45594 Fangrui Song changed: What|Removed |Added Status|CONFIRMED |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 40971] [Coverage] Assertion when using _Pragma("unroll()")
https://bugs.llvm.org/show_bug.cgi?id=40971 Yuanfang Chen changed: What|Removed |Added Resolution|--- |FIXED Status|CONFIRMED |RESOLVED --- Comment #2 from Yuanfang Chen --- Fixed with 9c2e770034d04a0966b84cac14a9e3fb3b0d4ab7 -- 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 46257] New: data race in libomptarget
https://bugs.llvm.org/show_bug.cgi?id=46257 Bug ID: 46257 Summary: data race in libomptarget Product: OpenMP Version: unspecified Hardware: PC OS: Linux Status: NEW Severity: normal Priority: P Component: Runtime Library Assignee: unassignedb...@nondot.org Reporter: xw111lu...@gmail.com CC: llvm-bugs@lists.llvm.org My code offloads from multiple host threads simultaneously. I'm running on x86-64 + NVIDIA GPU. git clone https://github.com/ye-luo/miniqmc.git cd miniqmc/build cmake -DCMAKE_CXX_COMPILER=clang++ -DENABLE_OFFLOAD=1 -DUSE_OBJECT_TARGET=ON .. make -j15 miniqmc_sync_move OMP_NUM_THREADS=8 ./bin/miniqmc_sync_move -w 80 -c 2 -n 1 Libomptarget fatal error 1: failure of target construct while offloading is mandatory Libomptarget fatal error 1: failure of target construct while offloading is mandatory Segmentation fault (core dumped) OMP_NUM_THREADS=1 runs fine. The same code runs well on Power9+V100 and XL compiler. -- 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 46237] Size regression -Os constant propagation difference results in larger code
https://bugs.llvm.org/show_bug.cgi?id=46237 Sanjay Patel changed: What|Removed |Added Status|NEW |RESOLVED Resolution|--- |FIXED --- Comment #3 from Sanjay Patel --- https://github.com/llvm/llvm-project/commit/6f6d2d238360883039cd17986c9ef598d04995a3 -- 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 46255] Anonymous structs/unions can violate "Last member element" constraint on flexible arrays
https://bugs.llvm.org/show_bug.cgi?id=46255 Erich Keane changed: What|Removed |Added Status|CONFIRMED |RESOLVED Fixed By Commit(s)||113b0d7d0bd637743efb050ad61 ||9dd0c6d306e96/90ee8cf63613d ||f525c4dab5e715fbae5f7310a78 Resolution|--- |FIXED --- Comment #1 from Erich Keane --- 113b0d7d0bd637743efb050ad619dd0c6d306e96 Fixed in that commit, but that had a mistake in it, so THAT was fixed in: 90ee8cf63613df525c4dab5e715fbae5f7310a78 -- 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 46258] New: [AArch64] Some functions compiled without BTI with -fsanitize=cfi
https://bugs.llvm.org/show_bug.cgi?id=46258
Bug ID: 46258
Summary: [AArch64] Some functions compiled without BTI with
-fsanitize=cfi
Product: new-bugs
Version: trunk
Hardware: PC
OS: Linux
Status: NEW
Severity: normal
Priority: P
Component: new bugs
Assignee: pe...@pcc.me.uk
Reporter: samitolva...@google.com
CC: htmldevelo...@gmail.com, keesc...@chromium.org,
llvm-bugs@lists.llvm.org
When compiling a binary with both -mbranch-protection=bti and -fsanitize=cfi,
LLVM warns about functions being compiled without BTI:
$ echo "void a() {}" > test.c
$ clang -target aarch64-linux-gnu- -mbranch-protection=bti \
-flto=thin -fvisibility=default -fsanitize=cfi -fsanitize-cfi-cross-dso \
-c test.c
$ ld.lld -r -o test.lto.o test.o
warning: some functions compiled with BTI and some compiled without BTI
warning: not setting BTI in feature flags
Looking at the disassembly of test.o, it looks like the compiler generated
__cfi_check and __cfi_check_fail functions don't have the
"branch-target-enforcement" attribute.
--
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 46259] New: Clang pointer to VLA conversions disallowed in C++ mode.
https://bugs.llvm.org/show_bug.cgi?id=46259
Bug ID: 46259
Summary: Clang pointer to VLA conversions disallowed in C++
mode.
Product: clang
Version: trunk
Hardware: PC
OS: Windows NT
Status: NEW
Severity: enhancement
Priority: P
Component: C++
Assignee: unassignedclangb...@nondot.org
Reporter: erich.ke...@intel.com
CC: blitzrak...@gmail.com, dgre...@apple.com,
erik.pilking...@gmail.com, llvm-bugs@lists.llvm.org,
richard-l...@metafoo.co.uk
See https://godbolt.org/z/dBvJLK
void foo(void *a, int x) {
int (*VLA) [x] = (int(*)[x]) a;
VLA = (int(*)[x]) a;
}
:4:11: error: cannot initialize a variable of type 'int (*)[x]' with an
rvalue of type 'int (*)[x]'
int (*VLA) [x] = (int(*)[x]) a;
^ ~
:5:11: error: incompatible pointer types assigning to 'int (*)[x]' from
'int (*)[x]'
VLA = (int(*)[x]) a;
^
2 errors generated.
Compiler returned: 1
The two types listed in both error message (despite syntactically and textually
identical) are actually pointers to different types, thus conversions aren't
allowed. In C mode the IsPointerConversion goes through typesAreConvertible
and mergeTypes and decides they are the compatible. However, the rules in C++
disallow these compatible conversions.
GCC allows these, and it seems to me that these two types should be allowed as
they are effectively the 'same' type despite being represented as different
types in our AST.
SO, to fix this I think we either need to
a: better canonicalize VLA types based on their represented types/bounds,
though this wouldn't fix the situation where the value of 'x' above could
change between them, so it would accept some incompatible types (where hte
value of "x" changed between the two), and disallow others.
b: Simply treat these as 'same-type' as a condition in isPointerConversion. I
think we can just do a minimal reproduction of mergeTypes to recursively check
ONLY VLA and element types. C mode ALSO allows fixed-size arrays on either
side of the conversion, so we may consider making this work for all array
types?
--
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 42783] Add a way to avoid calls to noreturn functions from being merged
https://bugs.llvm.org/show_bug.cgi?id=42783 Zequan Wu changed: What|Removed |Added Resolution|--- |FIXED Status|CONFIRMED |RESOLVED --- Comment #5 from Zequan Wu --- A new attribute called "nomerge" has been introduced to prevent callsites merging. You can read more about it on https://clang.llvm.org/docs/AttributeReference.html#nomerge -- 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 46260] New: Using builtin_assume to optimize pointer "ptr + size / sizeof(Struct)" does not work
https://bugs.llvm.org/show_bug.cgi?id=46260
Bug ID: 46260
Summary: Using builtin_assume to optimize pointer "ptr + size /
sizeof(Struct)" does not work
Product: clang
Version: trunk
Hardware: All
OS: All
Status: NEW
Severity: normal
Priority: P
Component: C++
Assignee: unassignedclangb...@nondot.org
Reporter: mv...@google.com
CC: blitzrak...@gmail.com, dgre...@apple.com,
erik.pilking...@gmail.com, llvm-bugs@lists.llvm.org,
richard-l...@metafoo.co.uk
Background: for some vector logic, I attempted to minimize the code for keeping
the distance for some 'begin' vs 'end' span.
i.e.:
void SomeFunction() {
size_type size = byte_size(begin_, end_);
begin_ = grow_or_something(begin_, );
end_ = add_byte_size(begin_, size);
}
auto byte_size(S* begin, S* end) {
return (end - begin) * sizeof(S);
}
auto add_byte_size(S* begin, size_t n) {
__builtin_assume((n % sizeof(S)) == 0);
return begin + n / sizeof(S);
}
The 'add_byte_size()' method refuses to do the raw add, despite my 'assume this
is a multiple of sizeof(S)'. The only think that works (reliably) is cast it
through intptr, which I'd prefer not to. (especially as the real code works
with pointer traits / pointer types making it even more verbose / boilerplate.
See examples / attempts / what works at https://gcc.godbolt.org/z/94Upor
--
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 46261] New: [SLP] SLP vectorizer change causes performance loss
https://bugs.llvm.org/show_bug.cgi?id=46261 Bug ID: 46261 Summary: [SLP] SLP vectorizer change causes performance loss Product: libraries Version: trunk Hardware: PC OS: Windows NT Status: NEW Severity: normal Priority: P Component: Scalar Optimizations Assignee: unassignedb...@nondot.org Reporter: wolfgang_p...@playstation.sony.com CC: llvm-bugs@lists.llvm.org Created attachment 23593 --> https://bugs.llvm.org/attachment.cgi?id=23593&action=edit Test case After the commit https://github.com/llvm/llvm-project/commit/6a18a9548761b266b28a49f705a568677c24b59b the attached test experiences a roughly 50% slowdown. Compile with clang -O2 -march=btver2. The IR dump after SLP shows that, before the commit, a substantial number of shuffle instructions are present, e.g. %shuffle = shufflevector <2 x float> %14, <2 x float> undef, <4 x i32> whereas after the commit, the IR dump does not show any. -- 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 45624] clang fails to detect using an object before it is initialized
https://bugs.llvm.org/show_bug.cgi?id=45624 Zequan Wu changed: What|Removed |Added Status|CONFIRMED |RESOLVED Resolution|--- |FIXED --- Comment #11 from Zequan Wu --- A new flag -Wuninitialized-const-reference has been introduced to diagnose a warning when passing uninitialized variables as const reference parameters to a function. https://reviews.llvm.org/rG170b6869b563dd3393d99f3e03d389b9058d5f24 -- 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 46005] Name mangling problem when initializing inline std::function with lambda
https://bugs.llvm.org/show_bug.cgi?id=46005 Zequan Wu changed: What|Removed |Added Resolution|--- |DUPLICATE Status|NEW |RESOLVED CC||zequa...@google.com --- Comment #1 from Zequan Wu --- *** This bug has been marked as a duplicate of bug 45213 *** -- 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 46262] New: Miscompilation leads to system out to memory
https://bugs.llvm.org/show_bug.cgi?id=46262
Bug ID: 46262
Summary: Miscompilation leads to system out to memory
Product: clang
Version: trunk
Hardware: PC
OS: Windows NT
Status: NEW
Severity: release blocker
Priority: P
Component: -New Bugs
Assignee: unassignedclangb...@nondot.org
Reporter: changoch...@gmail.com
CC: htmldevelo...@gmail.com, llvm-bugs@lists.llvm.org,
neeil...@live.com, richard-l...@metafoo.co.uk
We found a bug that can exhaust the memory of the os.
POC:
---
int a;
int b() {
int c;
short d, *e = b();
switch (c)
do {
case 3:
if (a)
e = &d;
case 6:
e++;
} while (e);
return 1;
}
---
Cmdline:
---
clang-11 -cc1 -emit-obj -O3 -o /dev/null -x c poc.c
---
Clang-version:
---
clang version 11.0.0 (https://github.com/llvm/llvm-project.git
641d5ac4d1965990fcf981f153369b038816cd16)
Target: x86_64-unknown-linux-gnu
Thread model: posix
InstalledDir: /home//llvm-project/build_clean/bin
---
The compilation exhausts the 256G on my system. Tested on the latest
update-to-date master branch of clang.
--
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 46263] New: The lld cannot identify LMA overlapping when the VMA is not specified.
https://bugs.llvm.org/show_bug.cgi?id=46263
Bug ID: 46263
Summary: The lld cannot identify LMA overlapping when the VMA
is not specified.
Product: lld
Version: unspecified
Hardware: PC
OS: Windows NT
Status: NEW
Severity: enhancement
Priority: P
Component: ELF
Assignee: unassignedb...@nondot.org
Reporter: 2077213...@qq.com
CC: llvm-bugs@lists.llvm.org, smithp...@googlemail.com
I used the latest lld and one of the versions of lld and found that the latest
version had a problem.
MEMORY {
DDR : o = 0xfb0, l = 185M
TCM : o = 0x1F40, l = 128K
}
SECTIONS {
.text 0xfb0 : {
* (.text);
_etext = .;
} > DDR
.mdata : AT(LOADADDR(.text) + SIZEOF(.text)) {
_data = .;
*(.data);
_edata = .;
} > TCM
.bss : { // This is not reasonable because the LMA of .bss overlaps that of
.mdata.
_bstart = .;
*(.bss) * (COMMON);
_bend = .;
} > DDR
.ARM.exidx : {
*(.ARM.exidx);
} > DDR
}
// The latest linker. from https://github.com/llvm/llvm-project Source code
compilation
ld.lld -T my.ld test.o
objdump -hw a.out
## the .bss LMA is a Strange numbers.
Idx Name Size VMA LMA File off Algn Flags
0 .text 0008 0fb0 0fb0 1000 2**2 CONTENTS,
ALLOC, LOAD, READONLY, CODE
1 .mdata0004 1f40 0fb8 2000 2**2 CONTENTS,
ALLOC, LOAD, DATA
2 .bss 0008 0fb8 00200010 2004 2**2 ALLOC
3 .ARM.exidx0010 0fb00010 00200018 2010 2**2 CONTENTS,
ALLOC, LOAD, READONLY, DATA
4 .comment 00bc 2020 2**0 CONTENTS,
READONLY
5 .ARM.attributes 003e 20dc 2**0 CONTENTS,
READONLY
// one of version lld. The overlap can be identified.
//./ld.lld -v
//LLD 11.0.0 (https://github.com/llvm/llvm-project.git
07d68c24aa19483e44db4336b0935b00a5d69949) (compatible with GNU linkers)
ld.lld -T gcc.ld test.o
ld.lld: warning: lld uses blx instruction, no object with architecture
supporting feature detected
ld.lld: error: section .mdata load address range overlaps with .bss
>>> .mdata range is [0xFB8, 0xFBB]
>>> .bss range is [0xFB8, 0xFBF]
--
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
