As pointed out by @Zingam the paper was implemented in libc++ as an
extension. This patch does the bookkeeping. The inital release version
is based on historical release dates.

Completes:
- Add a conditional noexcept specification to std::apply

This reverts commit 51bf0dff.

There are test failures on Windows.

A couple of clauses are allowed multiple times when they are separated
by a device_type clause. This patch updates the ACC.td file to move
these clauses to the `allowedClause` list and the
`CheckAllowedOncePerGroup` function is used to make sure they appear
only once on the directive or for each device_type.

These tests were initially pushed together with
https://github.com/llvm/llvm-project/pull/75864 but they were triggering
some buildbot failure (sanitizers). They now make use of the
`OwningOpRef` so all the resources are correctly destroyed at the end of
each tests.
They will be extended to includes all the extra getter functions added
with device_type support.

Certain compilers do not seem to like the static assert with a string, causing
a implicit conversion. It can be removed as it should not be reachable and the
mlir::failure should handle it correctly in case it is.

An empty string attribute value (e.g. a parent-umbrella: "") is
equivalent to omitting it. Theres no reason to write it out.

This adds a ReductionElementalConversion transform to
OptimizedBufferizationPass, taking hlfir::count(hlfir::elemental) and
generating the inline loop to perform the count of true elements. This
lets us generate a single loop instead of ending up as two plus a
temporary.

Any and All should be able to share the same code with a different
function/initial value.

Replaced certain AST matchers in ExprMutationAnalyzer with a more direct
utilization of AST classes. The primary bottleneck was identified in the
canResolveToExpr AST matcher. Since this matcher was employed multiple
times and used recursively, each invocation led to the constant creation
and destruction of other matchers within it. Additionally, the continual
comparison of DynTypedNode resulted in significant performance
degradation.

The optimization was tested on the TargetLowering.cpp file. Originally,
the check took 156 seconds on that file, but after implementing this
enhancement, it now takes approximately 40 seconds, making it nearly
four times faster.

Despite this improvement, there are still numerous issues in this file.
To further reduce the computational cost of this class, it is advisable
to consider removing the remaining matchers and exploring alternatives
such as leveraging RecursiveASTVisitor and increasi...

-Rename to GPRPair.
-Rename registers to be named like X10_X11 instead of X10_PD. Except X0
 which is now X0_Pair since it is not paired with X1.
-Use unknown size and offset for the subreg indices. This might
 be a functional change, but does not affect any lit tests.

Prior to this, attempts to bind a bit-field to an NTTP of reference type
produced an error because references to subobjects in NTTPs are
disallowed. But C++20 allows references to subobjects in NTTPs generally
(see
[P1907R1](https://www.open-std.org/jtc1/sc22/wg21/docs/papers/2019/p1907r1.html)).
Without this change, implementing P1907R1 would cause a bug allowing
bit-fields to be bound to reference template arguments.

Extracted from https://reviews.llvm.org/D140996

This is in preparation for moving the run of AMDGPUAttributor earlier.
Currently it infers the lack of the corresponding intrinsic calls,
so if we introduce new ones we need to remove the attribute from any
possible transitive callers. This is more conservative than necessary,
we could try to identify specific subgraphs where LDS globals are not
used.

Other options include teaching the attributor to avoid adding it in
cases
where the lowering may choose the table, but this seems more complex.
Alternatively could add a second run which doesn't seem worth it.

Depends #71349

Since vector loads and stores from scalar memrefs translate to
llvm.load/store, add the ability to tag said loads and stores as
nontemporal. This mirrors functionality available in memref.load/store.

Add the vop3p op_sel fields in getInsVOP3P instead of getInsVOP3Base.
Also start using defvar for some of the intermediate fields. let
overrides of all the visible fields are really difficult to follow.

libclc does not have a half implementation for erf/erfc
Add one based on the float implementation by extending the input and
truncating the output.

Fixes #77147.

This adds a subset of the C++20 calendar data formatters:
- day,
- month,
- year,
- month_day,
- month_day_last, and
- year_month_day.

A followup patch will add the missing calendar data formatters:
- weekday,
- weekday_indexed,
- weekday_last,
- month_weekday,
- month_weekday_last,
- year_month,
- year_month_day_last
- year_month_weekday, and
- year_month_weekday_last.

This is required for users of `TypeQuery` that limit the set of
languages of the query using APIs such as
`GetSupportedLanguagesForTypes` or
`GetSupportedLanguagesForExpressions`.

Example usage: https://github.com/apple/llvm-project/pull/7885

Phab no longer knows about new revisions.

Fixes: #77411
When substituting deduced type, noexcept expr in method should be
instantiated and evaluated.
ThisScrope should be switched to method context instead of origin sema
context

This patch adds a configuration of the libc++ test suite that enables
optimizations when building the tests. It also adds a new CI
configuration to exercise this on a regular basis. This is added in the
context of [1], which requires building with optimizations in order to
hit the bug.

[1]: https://github.com/llvm/llvm-project/issues/68552

This PR exposes four PGO functions 

- `__llvm_profile_set_filename`
- `__llvm_profile_reset_counters`, 
- `__llvm_profile_dump` 
- `__llvm_orderfile_dump` 

to user programs through the new header `instr_prof_interface.h` under
`compiler-rt/include/profile`. This way, the user can include the header
`profile/instr_prof_interface.h` to introduce these four names to their
programs.

Additionally, this PR defines macro `__LLVM_INSTR_PROFILE_GENERATE` when
the program is compiled with profile generation, and defines macro
`__LLVM_INSTR_PROFILE_USE` when the program is compiled with profile
use. `__LLVM_INSTR_PROFILE_GENERATE` together with
`instr_prof_interface.h` define the PGO functions only when the program
is compiled with profile generation. When profile generation is off,
these PGO functions are defined away and leave no trace in the user's
program.

Background:
https://discourse.llvm.org/t/pgo-are-the-llvm-profile-functions-stable-c-apis-across-llvm-releases/75832

This change disables it on z/OS and AIX since it fails on both platforms with:
fatal error: error in backend: Objective-C support is unimplemented for object file format

This document captures the design philosophy of the acc dialect. It also
shares the rationale behind the design and implementation of various
operations - and ties that back to the dialect design goals.

Co-authored-by: Valentin Clement <clementval@gmail.com>
Co-authored-by: Slava Zakharin <szakharin@nvidia.com>

Previously, registers created for temporary defs in apply patterns were
rendered as uses, resulting in machine verifier errors.

LiveDebugValues requires CFG only has one entry. BranchFolding and
MachineBlockPlacement may remove all predecessors of landing pad which
leaves it to be another entry.

This is the logical equivalent for #76710 for APInt and uses the same
naming scheme.

Converted existing users through:
`git grep -l "cast<ConstantSDNode>\(.*\).*getAPIntValueValue" | xargs
sed -E -i
's/cast<ConstantSDNode>\((.*)\)->getAPIntValue/\1->getAsAPIntVal/'`

…9c849f42cf

This patch is based on a previous PR https://reviews.llvm.org/D144657
that added alloca address space handling to MLIR's DataLayout and DLTI
interface. This patch aims to add identical features to import and
access the global and program memory space through MLIR's
DataLayout/DLTI system.

[Flang][Driver] Enable gpulibc/nogpulibc options for Flang, which allows linking of GPU LIBC for the fortran and OpenMP runtime (#77135)

This patch seeks to add the -gpulibc and -nogpulibc for Flang, which
allows the linking of the GPU libc library, this allows the use of
memcpy and other useful library functions for GPU.

In particular, this allows the Fortran runtime (written in C++) to be
compiled for offload and then linked against the GPU LIBC library via
this option to resolve memcpy and other C library functions that the
fortran runtime depends on for AMD GPU devices (and likely other GPU
devices).

This is the current method I've tested and found to be able to utilise
the Fortran runtime when compiled for AMD GPU, albeit it requires
compiling libc for GPU and then the Fortran runtime for GPU, so not
particularly straight forward or user friendly yet.

Activating this option will allow the subset of C functions to also be
utilised for GPU in other C/C++ based Fortran libraries if any are made
when linking against GPU libc.