Compiler-rt started emitting the macho_embedded libraries in
`<resource_dir>/lib/darwin/macho_embedded` after
https://reviews.llvm.org/D105765 / 1e03c37b,
so update the clang's driver to reflect that.

Differential Revision: https://reviews.llvm.org/D115403

Place include elf.h under !apple condition

Differential Revision: https://reviews.llvm.org/D119038

Many CI runs are very similar in nature. Let's put them into groups for a better overview

Reviewed By: ldionne, #libc

Spies: libcxx-commits, arichardson, mstorsjo

Differential Revision: https://reviews.llvm.org/D119059

Enable peephole optimizations for aarch64.
Also small code refactoring - add PeepholeOpts under Peepholes class.

Vladislav Khmelevsky,
Advanced Software Technology Lab, Huawei

Differential Revision: https://reviews.llvm.org/D118732

We use splat_vector for FP nodes without VL, not SplatPat which handles
splat_vector and integer VMV_V_X_VL.

Reduces isel table size by a few hundred bytes.

Differential Revision: https://reviews.llvm.org/D119185

Some members are public but should be private. Nothing requires they are public
right now, so make them private.

Differential Revision: https://reviews.llvm.org/D119097

This reverts commit d7303364.

Back in https://reviews.llvm.org/D109459, we stopped using the C++03
emulation for std::nullptr_t by default, which was an ABI break. We
still left a knob for users to turn it back on if they were broken by
the change, with a note that we would remove that knob after one release.

The time has now come to remove the knob and clean up the std::nullptr_t
emulation.

Differential Revision: https://reviews.llvm.org/D114786

There are a few different test passes that check elementwise fusion in
Linalg. Consolidate them to a single pass controlled by different pass
options (in keeping with how `TestLinalgTransforms` exists).

Update `__init__.py` generation to implement `__lldb_init_module`, which calls
`__lldb_init_module` on submodules that define it.

This allows the use case where a user runs `command script import lldb.macosx`.
With this change, the `__lldb_init_module` function in `crashlog.py` and
`heap.py` will be run, which is where command registration is occurring.

Differential Revision: https://reviews.llvm.org/D119179

libtool can currently produce 2 warnings:

1. No symbols were in the object file
2. An object file with the same basename was specified multiple times

The first warning here is often harmless and may just mean you have some
translation units with no symbols for the target you're building for.
The second warning can lead to real issues like those mentioned in
https://reviews.llvm.org/D113130 where ODR violations can slip in.

This introduces a new -warnings_as_errors flag that can be used by build
systems that want to verify they never hit these warnings. For example
with bazel the libtool caller first uniques names to make sure the
duplicate base name case is not possible, but if that doesn't work as
expected, having it fail would be preferred.

It's also worth noting that llvm-libtool-darwin works around an issue
that cctools libtool experiences related to debug info and duplicate
basenames, the workaround is described here:
https://github.com/llvm/llvm-project/blob/30baa5d2a450d5e302d8cba3fc7a26a59d4b7ae1/llvm/lib/Object/ArchiveWriter.cpp#L424-L465
And it avoids this bug:
https://github.com/keith/radars/tree/f0cbbb1c37126ec6528c132510b29e08566377a7/DuplicateBasenameIssue

Differential Revision: https://reviews.llvm.org/D118931

This is a translation of the transform added to InstCombine with:
D118539

containers

Add test cases for iteration over the ordered associative container from
end to begin using operator--

Reviewed By: Quuxplusone, rarutyun

Differential Revision: https://reviews.llvm.org/D118114

This patch replaces the ValueRAII pointer with a default 'nullptr'
value. Previously this was initialized as a reference to an existing
variable. The use of this variable caused overhead as the compiler could
not look through the uses and determine that it was unused if 'Active'
was not set. Because of this accesses to the variable would be left in
the runtime once compiled.

Fixes #53641

Reviewed By: jdoerfert

Differential Revision: https://reviews.llvm.org/D119187

Fix the verification function of spirv::ConstantOp to allow nesting
array attributes.

Reviewed By: antiagainst

Differential Revision: https://reviews.llvm.org/D118939

[Bitstream] Temporarily disable UBSan for invalid bitcode tests. This fixes failing sanitizer buildbots.

Tests added by
  https://github.com/llvm/llvm-project/commit/3c86642edd28f1ce970882edaba8dce468ec7401
  https://github.com/llvm/llvm-project/commit/f4fca0fbb052e55935b483c8955c440b59511ce8
cause existing code to exhibit UB:
  https://github.com/llvm/llvm-project/blob/b4c6d1bb379192cb5b712fda9f60cd105f21194f/llvm/include/llvm/Bitstream/BitstreamReader.h#L244

llvm/include/llvm/Bitstream/BitstreamReader.h:244:51:
runtime error: shift exponent 35 is too large for 32-bit type 'unsigned int'

Hitting the UBSan error before the expected error causes the test to fail under UBSan:

Reviewed By: vitalybuka

Differential Revision: https://reviews.llvm.org/D119182

Fix mixed-sign comparison warnings and an out-of-bound access in high_precision_decimal.h

Reviewed By: michaelrj

Differential Revision: https://reviews.llvm.org/D119156

The unused vararibles in the tests are intentional.

Reviewed By: sivachandra

Differential Revision: https://reviews.llvm.org/D119183

Implement log1pf correctly rounded to all rounding modes relying on logf implementation for exponent > 2^(-8).

Reviewed By: sivachandra, zimmermann6

Differential Revision: https://reviews.llvm.org/D118962

Aligned new does not require size to be a multiple of alignment, so
memalign is the correct choice instead of aligned_alloc.

Fixes false reports for unaligned sizes.

Reviewed By: eugenis

Differential Revision: https://reviews.llvm.org/D119161

This patch fixes a bug introduced in commit 4eaf5846. Commit
4eaf5846 sets address space of function type as program
address space unconditionally. This breaks types which have
address space qualifiers. E.g. __ptr32.

This patch fixes the bug by using address space qualifiers if
present.

Differential Revision: https://reviews.llvm.org/D119045

This is translated from recent changes to the IR version of this function:
D119060
D119139

This is adapted from existing tests for instcombine.
We want to keep the backend logic synchronized with
that as much as possible.

See D119139 / D119060 / D118539

Previously, importing `crashlog` resulted in a message being printed. The
message was about other commands (those in heap.py), not `crashlog`. The
changes in D117237 made it so that the heap.py messages were printed only when
importing `lldb.macosx.heap`, not when importing `lldb.macosx.crashlog`. Some
users may see no output and think `crashlog` wasn't successfully loaded. This
ensures users see that `crashlog` is loaded.

rdar://88283132

Differential Revision: https://reviews.llvm.org/D119155

A significant number of our tests in C accidentally use functions
without prototypes. This patch converts the function signatures to have
a prototype for the situations where the test is not specific to K&R C
declarations. e.g.,

  void func();

becomes

  void func(void);

This is the fourth batch of tests being updated (there are a significant
number of other tests left to be updated).

Major user-facing changes:
llvm/Bitcode/BitcodeReader.h no longer includes llvm/IR/ModuleSummaryIndex.h

Some statistics:
clang++ -E  -Iinclude -I../llvm/include ../llvm/lib/Bitcode/Reader/*.cpp -std=c++14 -fno-rtti -fno-exceptions | wc -l
after: 493335
before: 539640

Discourse thread on the topic: https://discourse.llvm.org/t/include-what-you-use-include-cleanup/

Differential Revision: https://reviews.llvm.org/D119091

Summary:
This patch changes the ClangLinkerWrapper to use the executable path
when searching for the lld binary. Previously we relied on the program
name. Also not finding 'llvm-strip' is not considered an error anymore
because it is an optional optimization.

As Pavel pointed out, on Apple Silicon "b main" stops at a point after
the variable has already been initialized. This patch updates the test
case to avoids that. I've also split the test into separate files so its
easier to reproduce the individual scenarios without having to build any
shared state.

This patch makes the uncontrovertial changes to the pipeline.

Reviewed By: ldionne, #libc

Spies: libcxx-commits, arichardson

Differential Revision: https://reviews.llvm.org/D119041

The vectorizer will choose at times to "vectorize" loops with a scalar
factor (VF=1) with interleaving (IC > 1). This can occasionally produce
better code than the unroller (notable for reductions where it can
produce independent reduction chains that are combined after the loop).
At times this is not very beneficial though, for example when runtime
checks are needed or when the scalar code requires predication.

This addresses the second point, preventing the vectorizer from
interleaving when the scalar loop will require predication. This
prevents it from making a bit of a mess, that is worse than the original
and better left for the unroller to unroll if beneficial. It helps
reverse some of the regressions from D118090.

Differential Revision: https://reviews.llvm.org/D118566