With the STL containers, I didn't enable move operations in C++03 mode
because that would change the overload resolution for things that today
are copy operations. With iostreams, though, the copy operations aren't
present at all, and so I see no problem with enabling move operations
even in (Clang's greatly extended) C++03 mode.

Clang's C++03 mode does not support delegating constructors.

Differential Revision: https://reviews.llvm.org/D104310

This reverts commit 7d6e589f.

Windows build was unbroken.

- In the caller of the overridden `parseStatement` function (i.e. the `AsmParser::Run()`) in the case of an error **and** if we're not at the start of the statement, we "eat" up until the end of the current statement, so we don't have to process it again.
- However, in the HLASMAsmParser class what's happening is that, if an error occurs at the very start of the statement (for example, you invoke the HLASMAsmParser to parse a gnu directive), we will error out, but we never really progress in terms of the next token in the statement to parse. We simply keep looping processing the same error over and over again (partly because we're at the start of the statement)
- To remedy this, when the `parseAsHLASMLabel` function fails, before returning, we "eat" until the end of the statement function, so we don't process it anymore.

Reviewed By: uweigand

Differential Revision: https://reviews.llvm.org/D104869

This reverts commit 51e434fc because of a
build bot failure in test-suite::GCC-C-execute-pr60960.test that I need to
investigate.

This reverts commit 4f1238e4.
Buildbot fails on predecessor patch

Revert "[clang][PATCH][nfc] Refactor TargetInfo::adjust to pass DiagnosticsEngine to allow diagnostics on target-unsupported options"

This reverts commit 2c02b0c3.
buildbot fails

There is a constraint that coro.suspend instructions need to be in their
own blocks. The coro split pass initially creates IR that obeys this constraint
(which is later checked). Sinking rematerializable instructions into these
blocks breaks that constraint.

Instead rematerialize in the predecessor block to the suspend's single
predecessor block.

Differential Revision: https://reviews.llvm.org/D104051

This patch adds a new clang builtin, __arithmetic_fence. The purpose of the
builtin is to provide the user fine control, at the expression level, over
floating point optimization when -ffast-math (-ffp-model=fast) is enabled.
The builtin prevents the optimizer from rearranging floating point expression
evaluation. The new option fprotect-parens has the same effect on
parenthesized expressions, forcing the optimizer to respect the parentheses.

Reviewed By: aaron.ballman, kpn

Differential Revision: https://reviews.llvm.org/D100118

[clang][PATCH][nfc] Refactor TargetInfo::adjust to pass DiagnosticsEngine to allow diagnostics on target-unsupported options

Reviewed By: aaron.ballman

Differential Revision: https://reviews.llvm.org/D104729

[llvm][clang][fpenv] Create new intrinsic llvm.arith.fence to control FP optimization at expression level

This intrinsic blocks floating point transformations by the optimizer.

Author: Pengfei

Reviewed By: LuoYuanke, Andy Kaylor, Craig Topper, kpn

Differential Revision: https://reviews.llvm.org/D99675

Differential Revision: https://reviews.llvm.org/D104669

MemRefDataFlow performs mem2reg style operations for affine load/stores. Unfortunately, it is not presently correct in the presence of external operations such as memref.cast, or function calls. This diff extends the functionality of the pass to remain correct in the presence of such ops.

Differential Revision: https://reviews.llvm.org/D104053

Gate the include of alloca.h behind _WIN32 guard.

Differential Revision: https://reviews.llvm.org/D105036

Added the option `-altivec-src-compat=[mixed,gcc,xl]`. The default at this time is `mixed`.

The default behavior for clang is for all vector compares to return a scalar unless the vectors being
compared are vector bool or vector pixel. In that case the compare returns a
vector. With the gcc case all vector compares return vectors and in the xl case
all vector compares return scalars.

This patch does not change the default behavior of clang.

This option will be used in future patches to implement behaviour compatibility for the vector bool/pixel types.

Reviewed By: bmahjour

Differential Revision: https://reviews.llvm.org/D103615

A canonicalization accidentally will remove a memref allocation if it is only stored into. However, this is incorrect if the allocation is the value being stored, not the allocation being stored into.

Differential Revision: https://reviews.llvm.org/D104947

This is just for consistency with all other instruction multiclasses
that pass around pseudo names as arguments.

This reverts commit 6f3b775c.
Test fails flakily, see comments on https://reviews.llvm.org/D103967

Also revert follow-up "[Analyzer] Attempt to fix windows bots test
failure b/c of new-line"
This reverts commit fe0e861a.

This reverts commit adace796.
Still breaks things, see comment on https://reviews.llvm.org/D91722

Previously xscale was known to everything apart
from the ELF streamer so we would crash as soon
as you tried to output an object file.

Reviewed By: nickdesaulniers

Differential Revision: https://reviews.llvm.org/D104776

Precommit test case for a change to GetIndexDifference in D99424.

Updated Fortran LLVM Test-suite docs to reflect latest changes.

Reviewed By: Meinersbur, xgupta

Differential Revision: https://reviews.llvm.org/D104961

Given a select that returns the logical negation of the condition, replace it with a not of the condition.

Differential Revision: https://reviews.llvm.org/D104966

The underlying TargetProcessControl API changes with D104694. Once it landed I can patch and re-enable the example.

This reverts commit e9396449.

Breaks Windows build.

On AIX the alignment implementation has the storage aligned to the
preferred alignment instead of the alignment of a type. Macro guard
these tests for AIX and have them pass when the "reference alignment" is
less than or equal to the alignment observed. In other words, the
alignment applied is at least as strict as the required alignment.

Reviewed By: hubert.reinterpretcast

Differential Revision: https://reviews.llvm.org/D104786

This patch relands https://reviews.llvm.org/D104454, but fixes some failing
builds on Mac OS which apparently has a different definition for size_t,
that caused 'ambiguous operator overload' for the implicit conversion
of TypeSize to a scalar value.

This reverts commit b732e6c9.

Peephole optimizer should not be introducing sub-reg definitions
as they are illegal in machine SSA phase. This patch modifies
the optimizer to not emit sub-register definitions.

Reviewed By: arsenm

Differential Revision: https://reviews.llvm.org/D103408

This helps remove extra comparisons when generating masks for fixed
length masked operations.

Differential Revision: https://reviews.llvm.org/D104910

In typeinfo there is a reinterpret_cast between a uintptr_t and size_t. These are two integer types and therefore a reinterpret_cast is not right for this situation. It looks like it may have been copied and pasted from above in the file. An implicit cast works in it's place.

Reviewed By: ldionne, #libc

Differential Revision: https://reviews.llvm.org/D104814

Compiler crashes at an assertion while casting operands to PtrToIntInst at some cases when
ptrtoint is present as an explicit operand to inttoptr. Explicit instruction operator as
operand can not be casted to an Instruction.

This patch replaces cast from PtrToInst to Operator which are later checked for constant
expressions.

Differential Revision: https://reviews.llvm.org/D105002

Reviewed By: dylanmckay

Differential Revision: https://reviews.llvm.org/D99239

Adds legalizer, register bank select, and instruction
select support for G_SBFX and G_UBFX. These opcodes generate
scalar or vector ALU bitfield extract instructions for
AMDGPU. The instructions allow both constant or register
values for the offset and width operands.

The 32-bit scalar version is expanded to a sequence that
combines the offset and width into a single register.

There are no 64-bit vgpr bitfield extract instructions, so the
operations are expanded to a sequence of instructions that
implement the operation. If the width is a constant,
then the 32-bit bitfield extract instructions are used.

Moved the AArch64 specific code for creating G_SBFX to
CombinerHelper.cpp so that it can be used by other targets.
Only bitfield extracts with constant offset and width values
are handled currently.

Differential Revision: https://reviews.llvm.org/D100149

Partially implement P0315R4.

This patch allow lambda in unevaluated context.
It does not implement temp.deduct/9.

This lowering uses a library call to implement copying in the general case, i.e.,
supporting arbitrary rank and strided layouts.