lr.[wd]: mayLoad = 1, mayStore = 0
sc.[wd]: mayLoad = 0, mayStore = 1

all other AMOs: mayLoad = 1, mayStore = 1

This allows us to support larger stack offsets for FrameLowering.

Fixes #88365.

Use the std `if () continue;` idiom before falling into the
processing.

FormatManager::GetCategoryForLanguage and
FormatManager::GetCategory(can_create = true) can be called concurrently
and they both take the TypeCategory::m_map_mutex and the
FormatManager::m_language_categories_mutex but in reverse order.

On one thread, GetCategoryForLanguage takes m_language_categories_mutex
and then ends calling TypeCategoryMap::Get which takes m_map_mutex

On another thread GetCategory calls TypeCategoryMap::Add which takes
m_map_mutex and then calls FormatManager::Changed() which takes
m_language_categories_mutex

If both threads are running concurrently, we have a dead lock.

The patch releases the m_map_mutex before calling Changed which avoids
the dead lock.

---------

Co-authored-by: Vincent Belliard <v-bulle@github.com>

ROCm installation path is used for finding and automatically linking
required bitcode libraries for OpenMP AMDGPU offload.

Reported issue: https://github.com/llvm/llvm-project/issues/82553

This PR fix a AST dump issue since
https://github.com/llvm/llvm-project/pull/80001



When Clang dumps `CXXDefaultArgExpr`/`CXXDefaultInitExpr`, there has no
recursively dump the complete `CXXDefaultArgExpr`/`CXXDefaultInitExpr`.

Since this PR, Clang will recursively dump a
`CXXDefaultArgExpr`/`CXXDefaultInitExpr` node, even if the node has no
rewritten init.

*Consider*:
```
struct A {
  int arr[1];
};

struct B {
  const A &a = A{{0}};
};

void test() {
  B b{};
}
```

*Before*:
```
`-FunctionDecl <line:9:1, line:11:1> line:9:6 test 'void ()'
  `-CompoundStmt <col:13, line:11:1>
    `-DeclStmt <line:10:3, col:8>
      `-VarDecl <col:3, col:7> col:5 b 'B' listinit
        `-InitListExpr <col:6, col:7> 'B'
          `-CXXDefaultInitExpr <col:7> 'const A' lvalue has rewritten init
            `-ExprWithCleanups <line:6:16, col:21> 'const A' lvalue
```

*After*:
```
`-FunctionDecl 0x15a9455a8 <line:9:1, line:11:1> line:9:6 test 'void ()'
  `-CompoundStmt 0x15a945850 <col:13, line:11:1>
    `-DeclStmt 0x15a945838 <line:10:3, col:8>
      `-VarDecl 0x15a945708 <col:3, col:7> col:5 b 'B' listinit
        `-InitListExpr 0x15a9457b0 <col:6, col:7> 'B'
          `-CXXDefaultInitExpr 0x15a9457f8 <col:7> 'const A' lvalue has rewritten init
            `-ExprWithCleanups 0x15a945568 <line:6:16, col:21> 'const A' lvalue
              `-MaterializeTemporaryExpr 0x15a945500 <col:16, col:21> 'const A' lvalue extended by Field 0x15a945160 'a' 'const A &'
                `-ImplicitCastExpr 0x15a9454e8 <col:16, col:21> 'const A' <NoOp>
                  `-CXXFunctionalCastExpr 0x15a9454c0 <col:16, col:21> 'A' functional cast to A <NoOp>
                    `-InitListExpr 0x15a9452c0 <col:17, col:21> 'A'
                      `-InitListExpr 0x15a945308 <col:18, col:20> 'int[1]'
                        `-IntegerLiteral 0x15a945210 <col:19> 'int' 0
```

---------

Signed-off-by: yronglin <yronglin777@gmail.com>

This includes the lldb-dap executable in the MacOS and Linux
distributions of Swift. Currently there is a commit in the Apple repo to
do this for just MacOS https://github.com/apple/llvm-project/pull/8176.
This PR extends this to both Linux and MacOS and brings the change
upstream.

@JDevlieghere @adrian-prantl

The load narrowing part of TargetLowering::SimplifySetCC is updated
according to this:

1) The offset calculation (for big endian) did not work properly for
   non byte-sized types. This is basically solved by an early exit
   if the memory type isn't byte-sized. But the code is also corrected
   to use the store size when calculating the offset.
2) To still allow some optimizations for non-byte-sized types the
   TargetLowering::isPaddedAtMostSignificantBitsWhenStored hook is
   added. By default it assumes that scalar integer types are padded
   starting at the most significant bits, if the type needs padding
   when being stored to memory.
3) Allow optimizing when isPaddedAtMostSignificantBitsWhenStored is
   true, as that hook makes it possible for TargetLowering to know
   how the non byte-sized value is aligned in memory.
4) Update the algorithm to always search for a narrowed load with
   a power-of-2 byte-sized type. In the past the algorithm started
   with the the width of the original load, and then divided it by
   two for each iteration. But for a type such as i48 that would
   just end up trying to narrow the load into a i24 or i12 load,
   and then we would fail sooner or later due to not finding a
   newVT that fulfilled newVT.isRound().
   With this new approach we can narrow the i48 load into either
   an i8, i16 or i32 load. By checking if such a load is allowed
(e.g. alignment wise) for any "multiple of 8 offset", then we can find
   more opportunities for the optimization to trigger. So even for a
   byte-sized type such as i32 we may now end up narrowing the load
   into loading the 16 bits starting at offset 8 (if that is allowed
   by the target). The old algorithm did not even consider that case.
5) Also start using getObjectPtrOffset instead of getMemBasePlusOffset
   when creating the new ptr. This way we get "nsw" on the add.

These test cases show some miscomplies for big-endian when dealing
with non byte-sized loads. One part of the problem is that LLVM IR
isn't really telling where the padding goes for non byte-sized
loads/stores. So currently TargetLowering::SimplifySetCC can't assume
anything about it. But the implementation also do not consider that
the TypeStoreSize could be larger than the TypeSize, resulting in
the offset calculation being wrong for big-endian.

Pre-commit for https://github.com/llvm/llvm-project/pull/87646

[lldb][test] Add tests for evaluating local variables whose name clashes with Objective-C types (#87807)

Depends on https://github.com/llvm/llvm-project/pull/87767

This pull request implements LWG3736: move_iterator missing
disable_sized_sentinel_for specialization.

Co-authored-by: Wang Pengcheng <wangpengcheng.pp@bytedance.com>

Co-authored-by: Wang Pengcheng <wangpengcheng.pp@bytedance.com>

Co-authored-by: Wang Pengcheng <wangpengcheng.pp@bytedance.com>

Co-authored-by: Wang Pengcheng <wangpengcheng.pp@bytedance.com>

Co-authored-by: Wang Pengcheng <wangpengcheng.pp@bytedance.com>

Co-authored-by: Wang Pengcheng <wangpengcheng.pp@bytedance.com>

This pull request implements LWG3643: Missing constexpr in
std::counted_iterator. Specifically, one overload of
std::counted_operator::operator++ was not marked as constexpr,
despite being eligible for it after the introduction of try-block
support in constexpr functions in C++20.

When the alloca is too big for vectorization, the function could have
already
been modified in previous iteration of the `for` loop.

This reverts commit 29e8bfc1.
This patch didn't handle vector return type correctly.

[VectorCombine] foldShuffleOfCastops - ensure we can scale shuffle masks between bitcasted vector types

Don't just assert that the src/dst vector element counts are multiples of one another - in general IR this can actually happen.

Reported by @mikaelholmen

Reverts llvm/llvm-project#88395

This broke the powerpc buildbot. That build doesn't support using
`std::filesystem` in flang unit tests.

This is a GNU extension:
https://gcc.gnu.org/onlinedocs/gfortran/ACCESS.html

Used in SALMON:
https://salmon-tddft.jp/download.html

Unfortunately the intrinsic takes a file path to operate on so there
isn't an easy way to make the test robust. The unit test expects to be
able to create, set read write and execute permissions, and delete files
called
  `std::filesystem::temp_directory_path() / <test_name>.<pid>`

The test will fail if a file already exists with that name.

I have not implemented the intrinsic on Windows because this is wrapping
a POSIX system call and Windows doesn't support all of the permission
bits tested by the intrinsic. I don't have a Windows machine easily
available to check if Gfortran implements this intrinsic on Windows.

This originally dealt with tbss, but now handles any bss-like section.
So the comment is inaccurate. Also, the `{}` on the messaging seem
unnecessary.

Resolve #84905

This patch covers
[CWG393](https://cplusplus.github.io/CWG/issues/393.html) "Pointer to
array of unknown bound in template argument list in parameter",
[CWG528](https://cplusplus.github.io/CWG/issues/528.html) "Why are
incomplete class types not allowed with `typeid`?",
[CWG550](https://cplusplus.github.io/CWG/issues/550.html) "Pointer to
array of unknown bound in parameter declarations",
[CWG553](https://cplusplus.github.io/CWG/issues/553.html) "Problems with
friend allocation and deallocation functions",
[CWG555](https://cplusplus.github.io/CWG/issues/555.html)
"Pseudo-destructor name lookup",
[CWG560](https://cplusplus.github.io/CWG/issues/560.html) "Use of the
`typename` keyword in return types".

CWG393 is on this list, because CWG550 is marked as a duplicate of
CWG393.

Test for CWG553 has been already written, but it was missing a status
comment. As a drive-by fix, I'm adding missing status comments to
CWG1584 and CWG1903 as well.

CWG555 used CWG466 test, and also a variation of that test to test
references. CWG466 is now testing non-reference non-pointer case.

CWG560 showcases again that converting warnings to errors in DR tests
via `-pedantic-errors` doesn't make things more clear. By default that
test is accepted with an extension warning since we implemented
[P0634R3](https://wg21.link/p0634r3) "Down with `typename`!" in Clang
16.

This diagnostic used to talk about complex integer types but is issued
for use of the increment or decrement operators on any complex type,
not just integral ones.

The diagnostic also had zero test coverage, so new coverage is added
along with the rewording.

This patch updates Flang lowering to use the new set of OpenMP clause
operand structures and their groupings into directive-specific sets of
clause operands.

It simplifies the passing of information from the clause processor and
the creation of operations.

The `DataSharingProcessor` is slightly modified to not hold delayed
privatization state. Instead, optional arguments are added to
`processStep1` which are only passed when delayed privatization is used.
This enables using the clause operand structure for `private` and
removes the need for the ad-hoc `DelayedPrivatizationInfo` structure.

The processing of the `schedule` clause is updated to process the
`chunk` modifier rather than requiring two separate calls to the
`ClauseProcessor`.

Lowering of a block-associated `ordered` construct is updated to emit a
TODO error if the `simd` clause is specified, since it is not currently
supported by the `ClauseProcessor` or later compilation stages.

Removed processing of `schedule` from `omp.simdloop`, as it doesn't
apply to `simd` constructs.

If the extract_subvector is cheap, attempt to extract directly from an inserted subvector

Refactor `CUDAFunctionTarget` into a scoped enum at namespace scope, so that it can be forward declared. This is done in preparation for `SemaCUDA`.

By default the scheduling info of instructions into a BUNDLE are given a
latency of 0 as they operate on the implicit register of the bundle.
This modifies that for AArch64 so that the latency is adjusted to use
the latency from the instruction in the bundle instead. This essentially
assumes that the bundled instructions are executed in a single cycle,
which for AArch64 is probably OK considering they are mostly used for
MOVPFX bundles, where this can help create slightly better scheduling
especially for in-order cores.

This patch introduces an operation intended to hold loop information
associated to the `omp.distribute`, `omp.simdloop`, `omp.taskloop` and
`omp.wsloop` operations. This is a stopgap solution to unblock work on
transitioning these operations to becoming wrappers, as discussed in
[this
RFC](https://discourse.llvm.org/t/rfc-representing-combined-composite-constructs-in-the-openmp-dialect/76986).

Long-term, this operation will likely be replaced by
`omp.canonical_loop`, which is being designed to address missing support
for loop transformations, etc.

In preparation for `SemaCUDA`, which requires this enum to be forward-declarable.

This diff causes the `tblgen`-erated printProperties() function to skip
printing a `DefaultValuedAttr` property when the value is equal to the
default.

Co-authored-by: Biao Wang <biaow@nvidia.com>

Fortran mandates "CHARACTER(1), VALUE" be passed as a C "char" in calls
to BIND(C) procedures (F'2023 18.3.7 (4)). Lowering passed them by
memory instead. Update call interface lowering code to pass them by
register. Fix related test and update it to use HLFIR.