The GPU makes use of different address spaces. We generally work with
global memory, thread private memory, and thread shared memory. This
patch simply adds a few preliminary wrappers to map these concepts to
the numerical values the backend uses. Obviously casts between these
will need to be checked by the user.

Reviewed By: arsenm

Differential Revision: https://reviews.llvm.org/D156731

This is a follow-up to D151785, addressing https://github.com/clangd/clangd/issues/1703.

The previous approach of peeling pointer types during a traversal
using getPointeeType might have produced unexpected results; since
the method would implicitly desugar the type if the type being passed
in could not be cast to a Pointer-like Type.

Reviewed By: nridge

Differential Revision: https://reviews.llvm.org/D156300

When dealing with the subunits of a resource group, we should reset
the subunits availability at the first avaiable cycle of the resource
that contains the subunits. Previously, the reset operation was
returning cycle 0, effectively erasing the booking history of the
subunits.

Without this change, when using intervals for models have make use of
subunits, the erasing of resource booking for subunits can raise the
assertion "A resource is being overwritten" in
`ResourceSegments::add`. The test added in the patch is one of such
cases.

Reviewed By: andreadb

Differential Revision: https://reviews.llvm.org/D156530

This reverts commit 93a37067.

The "extern template" declaration of CycleInfo caused problems in a shared build
when CycleInfo was removed from Verifier.cpp. There needs to be an explicit
instantiation corresponding to an extern template in every SO.

This amends 1e06b82b

Adds support for SPV_INTEL_optnone.

Currently still in draft form but I wanted to open this revision
to ask some questions.

Differential Revision: https://reviews.llvm.org/D156297

Differential Revision: https://reviews.llvm.org/D149960

Reviewed By: zixuan-wu

Differential Revision: https://reviews.llvm.org/D156780

Instead of evaluating the entire thing and then pop'ing the value from
the stack, just pass the discard() on.

Set phi inputs to poison whenever we find a dead edge (either
during initial worklist population or the main InstCombine run),
instead of only doing this for successors of dead blocks.

This means that the phi operand is set to poison even if for
critical edges without an intermediate block.

There are quite a few test changes, because the pattern is fairly
common in vectorizer output, for cases where we know the vectorized
loop will be entered.

Differential Revision: https://reviews.llvm.org/D149837

To reduce spurious diffs in future changes.

The GenericConvergenceVerifier stores a pointer *OS, which can be nullptr. This
can crash the verifier.

Avoid the second backedge condition from being optimized away,
that's not the point of the test.

Improve foldOpIntoPhi() for icmp operator to check if incoming PHI value can be replaced with constant based on implied condition.

Depends on D156619.

Differential Revision: https://reviews.llvm.org/D156620

Precommit tests for D156620.

Differential Revision: https://reviews.llvm.org/D156619

Reviewed By: zixuan-wu

Differential Revision: https://reviews.llvm.org/D156543

Reviewed By: zixuan-wu

Differential Revison: https://reviews.llvm.org/D154588

Per C++ standard, programs omitting the definition for initializer_list
is ill-formed, https://eel.is/c++draft/dcl.init.list#2.

Fixes https://github.com/llvm/llvm-project/issues/64198

Differential Revision: https://reviews.llvm.org/D156712

Differential Revision: https://reviews.llvm.org/D149834

This can be nullptr and ubsan found a couple of cases in LLVM's unit
tests.

This patch adds our first integration test for scalable vectorisation in
Linalg. It simply runs `linalg.fill` to fill a scalable vector with a
pre-defined f32 value. The result is printed to stdout.

Note that with scalable architectures, the vector size is not know at
compile time, but it is known at runtime. For this reason, the length of
the output generated by the new test depends on the hardware implementation. For
Arm's SVE we do know that there will be at least 4 f32 elements in every
scalable vector register. CHECK lines were designed accordingly.

In order to see what happens for different implementations of SVE, you
can use the following QEMU settings:
  * qemu-aarch64 -cpu max,sve128=on
  * qemu-aarch64 -cpu max,sve512=on

ATM, this test is only enabled when `MLIR_RUN_ARM_SVE_TESTS` is set.

Differential Revision: https://reviews.llvm.org/D155839

Ignore the expressions and re-do the tests without all the "result
ignored" expected warnings. Those are expected, given the nature of the
tests.

Differential Revision: https://reviews.llvm.org/D149831

The inner 1d vector row can be summed with vector.reduction op. The
earlier mul reduction can't be updated similarly as it currently crashes
in the backend with:

  LLVM ERROR: Expanding reductions for scalable vectors is undefined.

Reviewed By: awarzynski, dcaballe

Differential Revision: https://reviews.llvm.org/D156701

Currently a loop is materialized when lowering ArmSME loads and stores
to intrinsics. This patch introduces two new ops to the ArmSME dialect
that map 1-1 with intrinsics:

  1. arm_sme.load_tile_slice  - Loads a 1D tile slice from
     memory into a 2D SME "virtual tile".
  2. arm_sme.store_tile_slice - Stores a 1D tile slice from a 2D SME
     "virtual tile" into memory.

As well as a new conversion pass '-convert-arm-sme-to-scf' that
materializes loops with these ops. The existing load/store lowering to
intrinsics is updated to use these ops.

Depends on D156517

Discourse thread:
https://discourse.llvm.org/t/loop-materialization-in-armsme/72354

Reviewed By: awarzynski, dcaballe, WanderAway

Differential Revision: https://reviews.llvm.org/D156467

In degenerate cases, it is possible for unreachable code removal
to remove the current instruction. However, we still return the
instruction to report a change, resulting in a use after free.

Instead, perform the change reporting in the same way as
eraseInstFromFunction() does, by directly setting MadeIRChange
and returning nullptr.

Fixes https://github.com/llvm/llvm-project/issues/64235.

The Base expression might cause side-effects, so we still have to
evaluate it.

Differential Revision: https://reviews.llvm.org/D149828

Reviewed By: aaron.ballman

Differential Revision: https://reviews.llvm.org/D156597

The 'get_tile_id' op is currently marked as pure so CSE thinks these ops
are equivalent and replaces with them a single one. This patch removes
pure to prevent this.

Reviewed By: awarzynski

Differential Revision: https://reviews.llvm.org/D156558

Add -canonicalize to simplify Vector to ArmSME test and use CHECK-DAG.

Reviewed By: awarzynski, dcaballe

Differential Revision: https://reviews.llvm.org/D156517

Create a mix-in class with an overloaded constructor that makes the
return type optional.

Reviewed By: ftynse

Differential Revision: https://reviews.llvm.org/D156561

Currently, we consider any instruction that might read from memory to be unsafe for phi-of-ops.
This patch refines that by walking the clobbering memDefs until we either hit a block that
strictly dominates the phi block (safe) or we hit a clobbering memPhi (unsafe).

Differential Revision: https://reviews.llvm.org/D156055

Differential Revision: https://reviews.llvm.org/D154475

The last use was removed by:

  commit 6b31e63b
  Author: Kazu Hirata <kazu@google.com>
  Date:   Sat Jun 24 07:33:50 2023 -0700

It's valuable to have document highlights for labels and be able to find
references to them.

Reviewed By: nridge

Differential Revision: https://reviews.llvm.org/D150124

This reverts commit 3e386b22.

Next to the original fold, this also implements an unnecessary and
inappropriate simplifyICmpWithDominatingAssume() based fold.