We use a bump pointer to implement malloc for the hermetic tests.
Currently, we bump the pointer up by any amount. This means that calling
`malloc(1)` will misalign the buffer so any following `malloc(8)`
accesses will not be aligned. This causes problems in architectures
which require alignment.

Reviewed By: sivachandra

Differential Revision: https://reviews.llvm.org/D149863

Matches the CMake build, otherwise we're rebuilding tblgen (and everything else) when there's a change almost anywhere in LLVM.

This commit splits a function that both parses MD nodes from YAML into
DI{Expr,Loc,Variable} objects AND adds an entry to the MF variable table, so
that each of those jobs is done separately.

It will enable subsequent patches to reuse the MD node parsing code.

Differential Revision: https://reviews.llvm.org/D149870

This patch lowers the data clause on the OpenACC compute construct
to their corresponding acc data operand operation.
The decomposition is the same as in D149673.

Note that `private` and `firstprivate` are not lowered to data operand operation as they do not have one and will likely have dedicated design/process.

Depends on D149673

Reviewed By: razvanlupusoru, jeanPerier

Differential Revision: https://reviews.llvm.org/D149785

Update the HCFG builder to directly connect the created CFG to the
existing Plan's entry. This allows removing `setEntry`, which can cause
leaks when the existing entry is replaced.

Should fix
https://lab.llvm.org/buildbot/#/builders/5/builds/33455/steps/13/logs/stdio

When tooling out there produces a reproducer that is archived, the first thing
a user is likely to expect is to process this as they do with any MLIR file.
However https://reviews.llvm.org/D126447 changed the behavior of mlir-opt to
eliminate the `--run-reproducer` option and instead automatically run it when
present in the input file. This creates a discrepancy in how mlir-opt behaves
when fed with an input file, and is surprising for users.
The explicit passing of `--run-reproducer` to express user-intent seems more
in line with what is expected from `mlir-opt`.

Reviewed By: rriddle, jpienaar

Differential Revision: https://reviews.llvm.org/D149820

Prior to P2448R2 we were more aggressive in diagnosing ill-formed
constexpr functions. Many of these restrictions were relaxed and now it
is not required for defaulted comparison operators to call constexpr
functions.

This behavior is extended to before C++23 and diagnostic for it's use
can be enabled w/ -pedantic or -Wc++2b-default-comp-relaxed-constexpr

This fixes: https://github.com/llvm/llvm-project/issues/61238

Differential Revision: https://reviews.llvm.org/D146090

Last use removed in f7420453e80b9294273009efcebaceac2383269e.

D149553 changes the name of the LangOptions member. This change allows
libc++ to work with either name.

Reviewed By: philnik, #libc

Differential Revision: https://reviews.llvm.org/D149554

This patch lowers the data clause on the OpenACC data construct
to their corresponding acc data operand operation.
The copy clause is decomposed into acc.copyin before and acc.copyout after
the acc.data operation.
The copyout close is decomposed into acc.create before and acc.copyout after
the acc.data operation.
The attach clause is decomposed into acc.attach before and acc.detach after
the acc.data operation.

Depends on D149601

Reviewed By: jeanPerier

Differential Revision: https://reviews.llvm.org/D149673

Will be necessary for correct locking on volta.

API-only change to help with rebasing the rest of the stack on this.

Reviewed By: jhuber6

Differential Revision: https://reviews.llvm.org/D149871

Dumping jump table and tail call fdata is covered by subsequent iteration over
successors.

Reviewed By: #bolt, maksfb

Differential Revision: https://reviews.llvm.org/D149799

Reviewed By: aartbik, vzakhari

Differential Revision: https://reviews.llvm.org/D149874

During the ISO C++ Committee meeting plenary session the C++23 Standard
has been voted as technical complete.

This updates the reference to c++2b to c++23 and updates the __cplusplus
macro.

Drive-by fixes c++1z -> c++17 and c++2a -> c++20 when seen.

Reviewed By: aaron.ballman

Differential Revision: https://reviews.llvm.org/D149553

Sema.h is huge. This makes a small reduction to it by moving
EnterExpressionEvaluationContext into a new header, since it is an
independent component.

Differential Revision: https://reviews.llvm.org/D149796

This addresses the issue found by: https://lab.llvm.org/buildbot#builders/178/builds/4571

insts.

If the vectorizable GEP node is built, which should not be scheduled,
and at least one node is a non-gep instruction, need to insert the
vectorized instructions before the last instruction in the list, not
before the first one, otherwise the instructions may be emitted in the
wrong order.

The AOK_SizeDirective part from 5b37c181
(2014-08) seems unneeded nowadays (the root cause has likely been fixed
elsewhere). The part abuses that `call dword ptr foo` assembles the same way as `call
foo` in Intel syntax, which is going to be fixed (changed) by D149579.

The generated object files for
CodeGen/ms-inline-asm{,-functions,-variables,-static-variable}.c and
CodeGenCXX/ms-inline-asm-fields.cpp are unchanged
(-mno-incremental-linker-compatible) with just this patch. When D149579 is
subsequently applied, the FIXME part of `kptr` in
CodeGen/ms-inline-asm-functions.c will be fixed.

Differential Revision: https://reviews.llvm.org/D149695

This reverts commit bfe72059, and follow
on fix e3e6bc69, due to some remaining
instability exposed by the bot enabling expensive checks:
https://lab.llvm.org/buildbot/#/builders/42/builds/9842

Add support to lower data exit operations and use it for the
copyout, detach and delete clause on the `!$acc exit data` construct.

This patch refactor the data entry op creation so most part of the code
can be shared with the data exit op.

Reviewed By: jeanPerier

Differential Revision: https://reviews.llvm.org/D149601

All the used OpenMP pragmas are available with OpenMP 4.0, so they can all share a single conditional block.

Reviewed By: ldionne, #libc

Spies: yaxunl, sstefan1, guansong, jplehr, libcxx-commits, sunshaoce

Differential Revision: https://reviews.llvm.org/D149603

The asm in a naked function may reasonably expect the argument registers and the
return address register (if present) to be live.

When using -pg and -finstrument-functions, functions are instrumented by adding
a function call to `_mcount/__cyg_profile_func_enter/__cyg_profile_func_enter_bare`/etc,
which will clobber these registers. If the return address register is clobbered,
the function will be unable to return to the caller, possibly causing an
infinite loop.

```
__attribute__((naked)) void g() {
#if defined(__arm__)
  __asm__("bx lr");
#else
  __asm__("ret");
#endif
}

int main() { g(); }
```

It seems that the only one reasonable way to handle the combination is to
disable instrumenting for naked functions.

GCC PR: https://gcc.gnu.org/PR109707
Close https://github.com/llvm/llvm-project/issues/62504

Reviewed By: hans

Differential Revision: https://reviews.llvm.org/D149721

hlfir.assign, in general, ends up calling the Assign runtime that asserts
that the types of LHS and RHS match. In case of implicit logical<->integer
conversions (allowed as an extension) the operands of hlfir.assign
have non-matching types. This change makes sure that the lowering
produces explicit type cast (either as a scalar fir.convert or
as a hlfir.elemental producing array expression).

Reviewed By: jeanPerier

Differential Revision: https://reviews.llvm.org/D149765

This addressed some unresolved comments in https://reviews.llvm.org/D142930

Reviewed By: aartbik, wrengr

Differential Revision: https://reviews.llvm.org/D148565

Reviewed By: ldionne, #libc

Spies: sstefan1, jplehr, libcxx-commits, miyuki

Differential Revision: https://reviews.llvm.org/D149787

This patch migrates the MapCombinedInfoTy from Clang codegen to OpenMPIRBuilder.

Differential Revision: https://reviews.llvm.org/D149666

Fix a leak introduced in b85a402d and flagged by LSan
https://lab.llvm.org/buildbot#builders/5/builds/33452

Now that a large amount of GPU tests can be run in parallel, we should
document spurious failures. It is a well-known issue that launching many
GPU applications in parallel can lead to various problems, the worst of
which being an indefinite hang.

Differential Revision: https://reviews.llvm.org/D149857

Partially `fix` https://github.com/llvm/llvm-project/issues/62441.

Extend isImpliedCondOperands() to handle ugt/uge and sgt/sge predicates.

alive2 proof: https://alive2.llvm.org/ce/z/jLFDAv and
https://alive2.llvm.org/ce/z/Z8idUd

Differential Revision: https://reviews.llvm.org/D149510

ugt/uge/sgt/sge test for the extension to isImpliedCondOperands
case from issue 62441 and 61393:
(X >> Z) <=(u) Y ==> X <=(to u) Y and (X > Y +_{nuw} 1) ==> X != Y

Differential Revision: https://reviews.llvm.org/D149512

Depends on D149002

Differential Revision: https://reviews.llvm.org/D149346

Add a callback from interception to allow asan on Windows to produce
better error messages. If an unrecoverable error occured when
intercepting functions, print a message before terminating.

Additionally, when encountering unknown instructions, a more helpful
message containing the address and the bytes of the unknown instruction
is now printed to help identify the issue and make it easier to propose
a fix.

Depends on D149549

Differential Revision: https://reviews.llvm.org/D149002

Do not treat unknown instructions as a fatal error. In most cases,
failure to intercept a function is reported by the caller, though
requires setting verbosity to 1 or higher to be visible.

Better error message reporting for asan will be added in a separate
patch.

Differential Revision: https://reviews.llvm.org/D149549

The i686-w64-windows-gnu target does not use SEH (which MSVC uses),
but DWARF-2 exception handling or possibly sjlj depending on the
toolchain build options. On this target we have to actually intercept
functions in libc++ and libunwind which handles throwing exceptions.
This fixes the `TestCases/intercept-rethrow-exception.cpp` test.

The x86_64-w64-windows-gnu target already works because it uses SEH
which is handled by intercepting RaiseException, so this change does not
affect x86_64.

Depends on https://reviews.llvm.org/D148990

Differential Revision: https://reviews.llvm.org/D148991

These assembly patterns are needed to intercept some libc++ and
libunwind functions built by Clang for i686-w64-windows-gnu target.

Differential Revision: https://reviews.llvm.org/D148990

Differential Revision: https://reviews.llvm.org/D149696

Follow up to bfe72059 to require asserts
which is needed for the use of -stats. This showed up in the following
bot failure: https://lab.llvm.org/buildbot/#/builders/91/builds/16760

The patch makes visitFSUBForFMACombine serve vp.fsub too. It helps DAGCombiner
to fuse vp.fsub and vp.fmul patterns to vp.fma.

Reviewed By: luke

Differential Revision: https://reviews.llvm.org/D149821

This also fixes a missing close li tag as a drive-by