This reverts commit 579cc9ad.
This breaks on Windows.

Ensure that libSupport does not carry any static global initializer.
libSupport can be embedded in use cases where we don't want to load all
cl::opt unless we want to parse the command line.
ManagedStatic can be used to enable lazy-initialization of globals.

Summary: yaml2obj shouldn't create the string table that isn't needed
         - doing so wastes time and disk space.

Reviewed By: jhenderson

Differential Revision: https://reviews.llvm.org/D106420

This adds the optional wrappers around things, however this isn't sufficient yet for CL 3.0 without generic address space, I've got one more additional patch to add all those APIs, but this is an easier to review precursor.

Reviewed By: Anastasia

Differential Revision: https://reviews.llvm.org/D106111

This reverts commit 5eb2e9aa.
This broke MacOS builds, needs to have a safer check guarding the flag
addition.

Ensure that libSupport does not carry any static global initializer.
libSupport can be embedded in use cases where we don't want to load all
cl::opt unless we want to parse the command line.
ManagedStatic can be used to enable lazy-initialization of globals.

The motivation for this caching wasn't clear, remove it in an effort to
simplify the code and make libSupport free of global dynamic constructor.

Reviewed By: dexonsmith

Differential Revision: https://reviews.llvm.org/D106206

This is consistent with the two other usages of prof md in this pass.

Same as D106650, but for `FoldTwoEntryPHINode()`

Reviewed By: spatel

Differential Revision: https://reviews.llvm.org/D106717

If the branch isn't `unpredictable`, and it is predicted to *not* branch
to the block we are considering speculatively executing,
then it seems counter-productive to execute the code that is predicted not to be executed.

Reviewed By: spatel

Differential Revision: https://reviews.llvm.org/D106650

glibc/elf/tst-absolute-zero-lib.lds uses `"absolute" = 0;`

These symbols are somewhat interesting in that they create non-existing
segments, which as far as I know is the only way to create segments
that don't contain any sections.

Final part of part of PR50760. Like D106629, but for segments instead
of sections. I'm not aware of anything that needs this in practice.

Differential Revision: https://reviews.llvm.org/D106767

Fixes the output segment name if both -rename_section and
-rename_segment are used and the post-section-rename segment
name is the same as the pre-segment-rename segment name to
match ld64's behavior.

The motivation is that segment$start$ can create section-less segments,
and this makes a corner case in the interaction between segment$start and
-rename_segment in the upcoming segment$start patch.

Differential Revision: https://reviews.llvm.org/D106766

No behavior change.

Differential Revision: https://reviews.llvm.org/D106765

If hsa_init fails, subsequent calls into hsa are not safe. Except for
hsa_init, but we don't retry on failure.

This patch:
- deletes a print that called into hsa to ask why it can't call into hsa
- drops a merge conflict block next to that print
- reliably initializes number of devices to zero
- skips the plugin destructor contents if the constructor failed to init hsa

Tested by making hsa_init return error, and by forcing the dynamic library
use which was then deleted from disk. Before this patch, both segv. After it,
friendly message about offloading being unavailable.

Reviewed By: jdoerfert

Differential Revision: https://reviews.llvm.org/D106774

This reverts commit a6eb3462.
The test fails, I screwed something up.

No behavior change.

Differential Revision: https://reviews.llvm.org/D106765

Makes clang crash: https://reviews.llvm.org/D105008#2903350
This reverts commit d2a73fb4.

Also revert a minor formatting follow-up:
This reverts commit 82834a67.

Inaccurate error handling around hsa_init

This reverts commit e30b3b23.

so that IMAGE_COMDAT_SELECT_LARGEST refers to the correct example.

Begin replacing individual getMemIntrinsicNode calls and setup (for X86ISD::VBROADCAST_LOAD + X86ISD::SUBV_BROADCAST_LOAD opcodes) with this getBROADCAST_LOAD helper.

Default to building the amdgpu plugin to use dlopen when hsa is
not found instead of disabling it.

Reviewed By: jdoerfert

Differential Revision: https://reviews.llvm.org/D106600

This patch introduces a new RAII struct that will temporarily make an OpenMP
RTL function have external linkage. This is done before the attributor is
invoked to prevent it from incorrectly removing some function definitions that
we will use later. For example, if we determine all calls to one function are
dead, because it has internal linkage it can safely be removed. Later when we
try to get an instance to that function to modify the source using
`getOrCreateRuntimeFunction` we will then get an empty declaration for that
function that won't be defined anywhere. This patch prevents this from
occurring.

Reviewed By: jdoerfert

Differential Revision: https://reviews.llvm.org/D106707

The stack adjustment for local deallocation was incorrectly ported.

Reviewed By: MaskRay

Differential Revision: https://reviews.llvm.org/D106760

gcc 11 introduced support for depend clause, but the gomp interface of libomp
does not yet handle the information.
Also remove -fopenmp-version=50, which is no longer needed for clang, but not
supported by gcc.

If the rotation amount is a known splat, perform the modulo on the splat source, and then perform the splat. That way the amount-extension performed later by LowerScalarVariableShift can fold the splats away without any multiple-use issues.

Fixes one of the concerns raised on D104156

Rather than adding methods for dropping these attributes in
various places, add a function that returns an AttrBuilder with
these attributes, which can then be used with existing methods
for dropping attributes. This is with an eye on D104641, which
also needs to drop them from returns, not just parameters.

Also be more explicit about the semantics of the method in the
documentation. Refer to UB rather than Undef, which is what this
is actually about.

From LangRef:

> if the parameter or return pointer is null, poison value is
> returned or passed instead. The nonnull attribute should be
> combined with the noundef attribute to ensure a pointer is not
> null or otherwise the behavior is undefined.

Dropping noundef is sufficient to prevent UB. Including nonnull
in this method just muddies the semantics.

This still breaks buildbots

Bug Fix for PR: https://llvm.org/PR47960

This patch makes sure that the fast math flag used in the 'select'
instruction is the same as the 'fabs' instruction after the transformation.

Differential Revision: https://reviews.llvm.org/D101727

We build `deviceRTLs` with `-O1` by default, which also triggers OpenMPOpt. When
the info cache is created, some attributes are removed. As a result, although we
mark a few functions `noinline`, they are still inlined when the bitcode library
is generated. This can cause an issue in middle end optimization.

Reviewed By: jdoerfert

Differential Revision: https://reviews.llvm.org/D106710

[NFC][Codegen][X86] Improve test coverage for repeated insertions of the same scalar into different elements

To simplify diff in future patch

To simplify diffs in future patch

This is a minimum extension of D106607 to allow folding for
2 non-zero constantsi that can be materialized as immediates..

In the reduced test examples, we save 1 instruction by rolling
the constants into LEA/ADD. In the motivating test from the bullet
benchmark, we absorb both of the constant moves into add ops via
LEA magic, so we reduce by 2 instructions.

Differential Revision: https://reviews.llvm.org/D106684