The last template parameter of llvm::support::endian::{read,write}
defaults to unaligned, so we can drop that at call sites.

Identified with misc-unused-using-decls.

Fix issue https://github.com/llvm/llvm-project/issues/66990

For LLVM_DIAGNOSTIC_DIR we want to create a subdirectory while for
DSYMUTIL_REPRODUCER_PATH we want to use the directory specified. The
latter makes writing a test a whole lot easier. Keep both to support
both scenarios.

This reverts commit b5ff71e2.  As described in
https://github.com/llvm/llvm-project/issues/68730, this appears to have exposed
an existing liveness issue.  Revert to green until we can figure out how to
address the root cause.

Note: This was not a clean revert.  I ended up doing it by hand.

This reverts commit 606f89ab while investigating bot failures.

The `po` alias now matches the behavior of the `expression` command when
the it can apply a Fix-It to an expression.
Modifications

- Add has `m_fixed_expression` to the `CommandObjectDWIMPrint` class a
`protected` member that stores the post Fix-It expression, just like the
`CommandObjectExpression` class.
- Converted messages to present tense.
- Add test cases that confirms a Fix-It for a C++ expression for both
`po` and `expressions`

rdar://115317419



Co-authored-by: Pete Lawrence <plawrence@apple.com>

A recent change to ubsan

(https://github.com/llvm/llvm-project/commit/792674400f6f04a074a3827349ed0e2ac10067f6)
exposed an unaligned load in MachOReader (see
https://lab.llvm.org/buildbot/#/builders/85/builds/19482 for an
example).

This patch fixes it by dropping the alignment.

In multi threaded application, it is possible for one thread to
terminate the program while another is flushing profile information. We
setup a signal mask to delay SIGKILL so that we can safely flush the
profile.

This patch setups the mask earlier: it reduces the window during which a
SIGKILL can end the computation prematurely.

This fixes https://bugzilla.mozilla.org/show_bug.cgi?id=1850940

, where a
profiled firefox was encountering several:

LLVM Profile Error: Failed to write file "default_*.profraw": Broken
pipe

Co-authored-by: serge-sans-paille <sguelton@mozilla.com>

Update documentation.

This is the 1st PR of [Mesh sharding
RFC](https://discourse.llvm.org/t/open-mlir-meeting-9-28-2023-rfc-sharding-framework-design-for-device-mesh/73695),
includes

Includes:

- mesh.cluster op
- mesh.shard op (the mesh.annotate op in the RFC slides, the name is
modified a bit from @stellaraccident 's advice, which I think might be a
bit more concise)
- MeshSharding attribute

This commit causes the following issue with sanitizers:

`include/c++/v1/__debug_utils/strict_weak_ordering_check.h:52: assertion
!__comp(*(__first + __b), *(__first + __a)) failed: Your comparator is
not a valid strict-weak ordering`

probably due to an invalid sort().

Revert "[MLIR][Transforms] Fix Mem2Reg removal order to respect
dominance (#68687)"

This reverts commit be81f42b.

Previously, libcxx forced all strings created during constant evaluation
to point to allocated memory. That was done due to implementation
difficultites, but it turns out not to be necessary. This patch permits
the use of SSO strings during constant evaluation, and also simplifies
the implementation.

This does have a downside in terms of enabling users to accidentally
write non-portable code, however, which I've documented in
UsingLibcxx.rst.

In particular, whether `constinit std::string x = "...";` will
successfully compile now depends on whether the string is smaller than
the SSO capacity -- in libc++, up to 22 bytes on 64-bit platforms, and
up to 10 bytes on 32-bit platforms. By comparison, libstdc++ and MSVC
have an SSO capacity of 15 bytes, except that in libstdc++,
constant-initialized strings cannot be used as function-locals because
the object contains a pointer to itself.

Closes #68434

We lower G_PTR_ADD to G_ADD before selection. We choose to do this here
and not post-legalization because G_PTR_ADD can still be combined with
other instructions during selection.

We use tablegen patterns to select for G_ICMP. There are existing
SelectionDAG patterns for selecting G_LOAD, G_STORE and G_ZEXTLOAD with
XLenVT type data. We introduce GIAddrRegImm to successfully import those
patterns. For pointer and i32 loads and stores, we introduce new patterns, since these are not legal types in SelectionDAG.

Calls to variadic functions do not seem to require any special handling.

To allow for a smoother transition, keep the safe mode working as is in
the LLVM 18 release (the first release that aims to make hardening
available), then deprecate it in LLVM 19.

We observe intermittent failures of that test and need some time to
investigate. Hence, for now, we disable it.

We do not define anything beyond the C Library's limits.h, so it
shouldn't be needed. This removes the need for a GCC-specific
workaround, see [1].

[1]: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107795

Differential Revision: https://reviews.llvm.org/D138384

Summary:
The implementation of `assert` has an if statement so that only the
first thread in the warp prints the assertion. On modern NVPTX
architecture, this can be printed out of order with the abort call. This
would lead to only a portion of the message being printed and then
exiting the program. By adding a mandatory warp sync we force the full
string to be printed before we continue to the abort.

There are no 64-bit variants of these ALU / SETHI instructions in V9.
Remove these instruction definitions and add patterns to match DAG nodes
to the generic instructions defined in SparcInstrInfo.td.

This is not strictly NFC because of the changes in
`2011-01-11-FrameAddr.ll` test. The reason is that Sparc delay slot
filler pass handled ADDrr but not ADDXrr, which are now the same
instruction.

Rename DSYMUTIL_REPRODUCER_PATH to LLVM_DIAGNOSTIC_DIR as the latter is
more generic and already set in our downstream build.

Add support for assumed shape arrays in lowering of the copy region of
the firstprivate recipe. Information is passed in block arguments as it
is done for the reduction recipe.

Extend the logic in addInfoForInductions to support decrementing
inductions with a step of -1.

Fixes #64881.

In some cases clobbering store can be safely skipped if it can only must
or no alias with memory location and it writes the same value. This
patch supports simple case when the value from memory location was
loaded in the same basic block before the store and there are no
modifications between them.

CDSort function reordering outperforms the existing default heuristic (
hfsort/C^3) in terms of the performance of generated binaries while
being (almost) as fast. Thus, the suggestion is to change the default.
The speedup is up to 1.5% perf for large front-end binaries, and can be
moderate/neutral for "small" benchmarks.

High-level **perf impact** on two selected binaries:
clang-10 binary (built with LTO+AutoFDO/CSSPGO): wins on top of C^3 in
[0.3%..0.8%]
rocksDB-8 binary (built with LTO+CSSPGO): wins on top of C^3 in
[0.8%..1.5%]

More detailed measurements on the clang binary is at
[here](https://reviews.llvm.org/D152834#4445042)

Refer to Zcmp rather than Zca. Also, this pass will only modify POP
instructions (replacing with POPRET if possible).

None of the Zc* extensions are experimental any more, so the comment was
out of date.

The file no longer contains any extensions that are still experimental.

Integrating MTE globals on Android revealed a lot of cases where
libraries are built as both archives and DSOs, and they're linked into
fully static and dynamic executables respectively.

MTE globals doesn't work for fully static executables. They need a
dynamic loader to process the special R_AARCH64_RELATIVE relocation
semantics with the encoded offset. Fully static executables that had
out-of-bounds derived symbols (like 'int* foo_end = foo[16]') crash
under MTE globals w/ static executables. So, LLD in its current form
simply errors out when you try and compile a fully static executable
that has a single MTE global variable in it.

It seems like a much better idea to simply have LLD not do the special
work for MTE globals in fully static contexts, and to drop any
unnecessary metadata. This means that you can build archives with MTE
globals and link them into both fully-static and dynamic executables.

Legalize bigger types for i8 and i16 vector types for G_VECREDUCE_ADD

-lowering tosa reduce operations to linalg.reduce
op which indicates explicit reduce operation.

-the linalg reduce op doesn't have iterator
types or indexing maps instead it has a reduction
dimension which would provide the necessary
information.

Per discussion on https://github.com/llvm/llvm-project/pull/68453 , the
make_buffer_rsrc intrinsic was incorrectly marked noclobber. This commit
fixes the issue.

Right now DWARFContext for DWO/DWP that is created is not thread safe. 
Changed it so that thread safety is inherited from the main
binary DWARFContext.

The `release` flag is misleading and its semantics are not well defined.
Originally this was meant to allow for different `LIBC_NAMESPACE`
depending on whether the code was considered stabled and released or
unstable. It appears that we may have a canary environment that is
neither released or dev. As a consequence we move the `LIBC_NAMESPACE`
definition to its own file and each environment can override this file
with whatever makes sense.