Device clause when it occurs with **target enter data** and **target exit data** must be declared with some non negative value. So some changes were made to evaluate the device clause argument to non negative value and throw the expected error when it takes negative value as argument.
Reviewed By: clementval
Differential Revision: https://reviews.llvm.org/D119141
When, for example, using "ascend" operator, I was wondering whether
the bit order of the dag argument will be reversed in ascend order
as well.
This patch clarifies it.
Reviewed By: myhsu
Differential Revision: https://reviews.llvm.org/D119653
This patch fixes the issue that the for loop in `applyToShadowMapEntries`
is infinite because `Itr` is not incremented in `CB`. Fixes#53727.
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D119471
D108887 fixed alignment mismatch by changing the caller's alignment in
ABI. However, we found some cases that still assume the alignment is
vector size. This patch fixes them to avoid the runtime crash.
Reviewed By: rnk
Differential Revision: https://reviews.llvm.org/D114536
`<filesystem>` header has been around for a while now, so we can safely remove
`<experimental/filesystem>` header. `_LIBCPP_DEPRECATED_EXPERIMENTAL_FILESYSTEM`
suggests we were going to remove `<experimental/filesystem>` in llvm 11 release,
but we never did. So, remove the experimental header now, its associated tests,
and the `_LIBCPP_DEPRECATED_EXPERIMENTAL_FILESYSTEM` macro.
Differential Revision: https://reviews.llvm.org/D119603
While matching widening multiply, if we matched an extend from i8->i32,
i16->i64 or i8->i64, we need to reintroduce a narrower extend. If we're
matching a vwmulsu we need to use a sext for op0 and a zext for op1.
This bug exists in LLVM 14 and will need to be backported.
Differential Revision: https://reviews.llvm.org/D119618
Now that the old device runtime has been deleted there is only a single
target that differs by the triple and the architecture. Simplify the
scheme for identifying the library but directly using the triple.
Reviewed By: JonChesterfield
Differential Revision: https://reviews.llvm.org/D119638
This reverts commit 5ebdb07e7e.
Enabling shrink wrap by default can cause assertions or crashes, and
these should first be investigated and fixed. For now, reverting the
change so it can be cherry-picked into 14.0.0 is the safest choice.
I have not found a way to expose a difference for this patch in a test
because it only triggers for a one-use load, but this is the code that
was adapted into D118376 and caused miscompiles. The new code pattern
is the same as what we do in narrowExtractedVectorLoad() (reduces load
width for a subvector extract).
This removes seemingly unnecessary manual worklist management and fixes
the chain updating via "SelectionDAG::makeEquivalentMemoryOrdering()".
Differential Revision: https://reviews.llvm.org/D119549
Extend the existing fold to use SimplifyMultipleUseDemandedBits as well as SimplifyDemandedVectorElts/SimplifyDemandedBits when attempting to simplify based off known zero vector elements.
To find uniform shift/rotation amounts, we currently use SelectionDAG::getSplatValue which creates a node that extracts the scalar value from the source vector, this makes it more difficult for later combines to remove the extraction and stay on the SIMD unit, and can be a problem when the scalar type is illegal (i.e. i64 vs v2i64 on 32-bit targets).
This patch begins to use SelectionDAG::getSplatSourceVector (which SelectionDAG::getSplatValue uses internally) and adds a new variant of getTargetVShiftNode that takes the source vector and the splat index, and adjusts the vector in place to create the zero-extended value suitable for the SSE PSLL/PSRL/PSRA uniform instructions.
I'm still addressing a number of regressions when used for normal vector shifts, so I've just handled the funnelshift/rotation lowering for this first patch. I can then focus on the yak shaving (SimplifyDemandedBits/Elts in particular) necessary to always use SelectionDAG::getSplatSourceVector.
Differential Revision: https://reviews.llvm.org/D119090
Apparently modern versions of ounit2 can only be found as "ounit2"
rather than "oUnit" version 2. Update the CMake check to support both
variants. This makes the OCaml tests run again with ounit2-2.2.4.
Differential Revision: https://reviews.llvm.org/D119079
A significant number of our tests in C accidentally use functions
without prototypes. This patch converts the function signatures to have
a prototype for the situations where the test is not specific to K&R C
declarations. e.g.,
void func();
becomes
void func(void);
This is the eighth batch of tests being updated (there are a
significant number of other tests left to be updated).
The pointer is referenced immediately, so assert the cast is correct instead of returning nullptr
It's only later iterations of the loop where the getParent() call might return nullptr
