Differential Revision: https://reviews.llvm.org/D149690

Removes the redundant Ack/Data bit manipulation.

Represents the inbox/outbox state with one bit instead of two. This will
be useful if we change to a packed representation and otherwise cuts the
runtime state space from 16 to 4.

Further simplification is possible, this patch is intentionally minimal.
- can_{send,recv}_data are now in == out
- {client,server}::try_open can be factored into Process:try_open

This implements the state machine of D148191, modulo differences in atomic
ordering and fences.

Reviewed By: jhuber6

Differential Revision: https://reviews.llvm.org/D149788

The wasm64 versions of the v128.storeX_lane instructions was incorrectly defined
as returning a v128 value, which resulted in spurious drop instructions being
emitted and causing validation to fail. This was not caught earlier because
wasm64 has been experimental and not well tested. Update the relevant test file
to test both wasm32 and wasm64.

Fixes #62443.

Differential Revision: https://reviews.llvm.org/D149780

Reduction clause on loop or compute constructs
is not currently lowered but no TODO message was issued.
Add a proper TODO so the user is informed.

Reviewed By: razvanlupusoru

Differential Revision: https://reviews.llvm.org/D149789

This changes the test in GlobalISel, which makes it match the test
elsewhere.

Differential Revision: https://reviews.llvm.org/D149795

We use a map of Argument->AllocaInst when mapping Arguments to the
AllocaInst created for them.

Said map is declared from "Value" and called a "DbgPtrCache". This
commit:
* replaces Value to the more specialized Argument class, to reflect the
intent better (i.e. we are _always_ mapping Arguments).
* replaces the name "DbgPtrCache" with the more explicit "ArgToAllocaMap",
as it is not clear reading the code what a "DbgPtr" is.

Differential Revision: https://reviews.llvm.org/D149748

Reviewed By: ldionne, Mordante, #libc

Spies: grandinj, libcxx-commits, arphaman

Differential Revision: https://reviews.llvm.org/D148478

Prevent operation reordering with fence instead of a comment.

The mailboxes are in shared memory and the locks structure in device memory.
If the mailboxes are read and then the lock taken, the lock says nothing
about the current or future state of those mail boxes. The relaxed atomic
fetch_or can be reordered before the relaxed atomic loads of unrelated
variables unless there is a fence preventing this.

Patches both Client::try_open and Server::try_open, one of which is missing
an optimisation and the other is missing the comment, but which otherwise
could be Process::try_open followed by buffer->opcode = opcode in Client.

Reviewed By: jhuber6

Differential Revision: https://reviews.llvm.org/D149790

Required after D148767 for flang+debug+slibs build.

Reviewed By: chapuni, clementval

Differential Revision: https://reviews.llvm.org/D149764

Add support for argument attributes on workgroup and private
attributions for GPU functions. These arguments are outside the range
of getNumArguments() and get printed separately, so the default
mechanism for function argument attributes can't be used on them.

Having done this, check for the `llvm.align` attribute on workgroup or
private attributions in a `gpu.func` and pass it through to the
relevant allocation op (creating a global or alloca). This allows
people creating kernels that use multiple workgroup buffers to set an
alignment.

(This could, in the future, be a GPU dialect `alignment` attribute,
but I've taken the simpler route of using the LLVM version instead for
simplicity and because I don't know how this might impact backends
like Vulkan)

Reviewed By: nirvedhmeshram

Differential Revision: https://reviews.llvm.org/D148965

Reviewed By: ldionne, #libc

Spies: libcxx-commits, miyuki

Differential Revision: https://reviews.llvm.org/D149599

A  DBG_VALUE or DBG_VALUE_LIST must always return a non-null
DILocalVariable, the ARMLoadStoreOptimizer code that move’s DBG_VALUE
and DBG_VALUE_LIST instructions if their corresponding loads have been
moved, currently just continues if it finds a DBG_VALUE or
DBG_VALUE_LIST with a null DILocalVariable, change that to an assert.

Differential revision: https://reviews.llvm.org/D149762

Not all AMDGPU targets support all atomic operations. For example,
there are not atomic floating-point adds on the gfx10 series. Add a
pass to emulate these operations using a compare-and-swap loop, by
analogy to the generic atomicrmw rewrite in MemrefToLLVM.

This pass is named generally, as in the future we may have a
memref-to-amdgpu that translates constructs like atomicrmw fmax (which
doesn't generally exist in LLVM) to the relevant intrinsics, which may
themselves require emulation.

Since the AMDGPU dialect now has a pass that operates on it, the
dialect's directory structure is reorganized to match other similarly
complex dialects.

The pass should be run before amdgpu-to-rocdl if desired.

This commit also adds f64 support to atomic_fmax.

Depends on D148722

Reviewed By: nirvedhmeshram

Differential Revision: https://reviews.llvm.org/D148724

This commit adds the buffer cmpswap intrinsic to the ROCDL dialect and
its corresponding AMDGPU dialect wrappers.

Reviewed By: nirvedhmeshram

Differential Revision: https://reviews.llvm.org/D148722

This reverts commit bf6ff4fd.

There is a bot failure where we are getting the correct remarks output
but in a different order. I'll need to investigate to see where we are
having nondeterministic behavior.

This diff switches the approach to comparison of constraint expressions
to the new one based on template args substitution.
It continues the effort to fix our handling of out-of-line definitions
of constrained templates.
This is a recommit of e3b1083e.

Differential revision: https://reviews.llvm.org/D146178

This ensures we get the correct FileCharacteristic during scanning. In a
yet-to-be-upstreamed branch this fixes observable failures, but it's
also good to handle this on principle: the FileCharacteristic is a
property of the file that is observable in the scanner, so there is
nothing preventing us from depending on it.

rdar://108627403

Differential Revision: https://reviews.llvm.org/D149777

Applies cloning decisions to the IR, cloning functions and updating
calls. For Regular LTO, the IR is updated directly during function
assignment, whereas for ThinLTO it is recorded in the summary index
(a subsequent patch will apply to the IR via the index during the
ThinLTO backend.

The function assignment and cloning proceeds greedily, and we create new
clones as needed when we find an incompatible assignment of function
clones to callsite clones (i.e. when different callers need to invoke
different combinations of callsite clones).

Depends on D140949.

Differential Revision: https://reviews.llvm.org/D141077

Simplifying poison mask elements yields poison values.

Differential Revision: https://reviews.llvm.org/D149544

This is a cheap pass so there's no need to limit to -O3.
This removes some differences between various pipelines.

Code size regressions should be addressed with
https://reviews.llvm.org/D149768.

Reviewed By: nikic

Differential Revision: https://reviews.llvm.org/D148269

In code we use `#include "llvm/Lib/Header.h"` which is located in
"llvm/include/llvm/Lib/Header.h", so we use "llvm/include/" as a header
search path. We should put modulemaps in the same directory and
shouldn't rely on clang to search in immediate subdirectories.

rdar://106677321

Differential Revision: https://reviews.llvm.org/D148776

Yet another amendment to 9bb28a18, this
addresses issues found in:
https://lab.llvm.org/buildbot/#/builders/188/builds/29204

This includes a couple of changes:
1. Moves the code that changes the root node out of the `TryToReduce` lambda and out of the traversal loop.
2. Since that code moved, there isn't much left in `TryToReduce` so the code was inlined.
3. The phi node variable `P` was also being used as a flag that turns on/off the exploration of operands as new seeds. This patch uses a new variable `TryOperandsAsNewSeeds` for this.
4. Simplifies the code executed when vectorization fails.

The logic of the code should be identical to the original, but I may be missing something not caught by tests.

Differential Revision: https://reviews.llvm.org/D149627

Re-land D145441 with data layout upgrade code fixed to not break OpenMP.

This reverts commit 3f2fbe92.

Differential Revision: https://reviews.llvm.org/D149776

Added basic implementation of ShuffleCostBuilder class in
ShuffleCostEstimator and generalized BaseShuffleAnalysis::createShuffle
function to support emission of Value */InstructionCost for the
vectorization/cost estimation.

Differential Revision: https://reviews.llvm.org/D149171

This addresses the issue found in:
https://lab.llvm.org/buildbot/#/builders/245/builds/7882

This time, the issue was that not every platform has the same alignment
values, so those were removed from the test case.

If the value count is > 1, we use a splat and thus only materialize the constant once, not once per element.

This addresses the issue found by:
https://lab.llvm.org/buildbot/#/builders/197/builds/4783
https://lab.llvm.org/buildbot/#/builders/188/builds/29201

We are relying on deleting the working directory to trigger a later
error in getcwd, but some platforms (e.g. Solaris) disallow deleting the
working directory of the process. Skip the test in that case.

Differential Revision: https://reviews.llvm.org/D149760

We have no use for debug info for the scanner modules, and writing raw
ast files speeds up scanning ~15% in some cases. Note that the compile
commands produced by the scanner will still build the obj format (if
requested), and the scanner can *read* obj format pcms, e.g. from a PCH.

rdar://108807592

Differential Revision: https://reviews.llvm.org/D149693

No call sites interpreted this value meaningfully. Simplify this
interface.

Reviewed By: MaskRay

Differential Revision: https://reviews.llvm.org/D149707

Factor out utility to simplify D147964 as sugested.

We filed some CD ballot comments which WG14 considered during the
ballot comment resolution meetings in Jan and Feb 2023, and this
updates our implementation based on the decisions reached. Those
decisions were (paraphrased for brevity):

US 9-034 (REJECTED)
  allow (void *)nullptr to be a null pointer constant
US 10-035 (ACCEPTED)
  accept the following code, as in C++:
  void func(nullptr_t); func(0);
US 22-058 (REJECTED)
  accept the following code, as in C++:
  nullptr_t val; (void)(1 ? val : 0); (void)(1 ? nullptr : 0);
US 23-062 (REJECTED)
  reject the following code, as in C++:
  nullptr_t val; bool b1 = val; bool b2 = nullptr;
US 24-061 (ACCEPTED)
  accept the following code, as in C++:
  nullptr_t val; val = 0;
US 21-068 (ACCEPTED)
  accept the following code, as in C++:
  (nullptr_t)nullptr;
GB-071 (ACCEPTED)
  accept the following code, as in C++:
  nullptr_t val; (void)(val == nullptr);

This patch updates the implementation as appropriate, but is primarily
focused around US 10-035, US 24-061, and US 23-062 in terms of
functional changes.

Differential Revision: https://reviews.llvm.org/D148800

Keeping this false could end up with extra iterations on a lot of loops
that aren't real ones (e.g. they could be a do-while-false for macros),
and makes the analyses very slow.

This patch changes the default for
CFG::BuildOptions.PruneTriviallyFalseEdges to true to avoid it.

Reviewed By: ymandel, xazax.hun, gribozavr2

Differential Revision: https://reviews.llvm.org/D149640

Argument promotion mostly works on functions with more than one caller (otherwise the function would be inlined or is dead), so there's a good chance that performing this increases code size since we introduce loads at every call site. If any caller is marked minsize, bail.

We could compare the number of loads/stores removed from the function with the number of loads introduced in callers, but that's TODO.

Reviewed By: nikic

Differential Revision: https://reviews.llvm.org/D149768

Internal testing showed that the change made in commit 62fcfc5a needed
more test coverage. Specifically, the imported function shouldn't be
externally visibile and the whole test needed to be run in regular LTO
mode.

Differential Revision: https://reviews.llvm.org/D148950

Decimal long doubles are not commonly used, and aren't currently
supported by the algorithm used for decimal float conversions. To avoid
giving incorrect answers, this patch adds a temporary exception to print
long doubles in hexadecimal even when decimal is requested.

Reviewed By: sivachandra

Differential Revision: https://reviews.llvm.org/D148179

This patch adds the function "socket" from the header "sys/socket". It's
a simple syscall wrapper, and I plan on adding the related functions in
a followup patch.

Reviewed By: sivachandra

Differential Revision: https://reviews.llvm.org/D149622