After landing the globalization optimizations, the precense of globalization on
the device that was not put in shared or stack memory is a failed optimization
with performance consequences so it should indicate a missed remark.
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D104735
They are already provided by Sema, deserializing from preamble if need
be. Moreover category names are meaningless outside interface/implementation
context, hence they were only causing noise.
Differential Revision: https://reviews.llvm.org/D104540
Slowly we are moving toward full support of sparse tensor *outputs*. First
step was support for all-dense annotated "sparse" tensors. This step adds
support for truly sparse tensors, but only for operations in which the values
of a tensor change, but not the nonzero structure (this was refered to as
"simply dynamic" in the [Bik96] thesis).
Some background text was posted on discourse:
https://llvm.discourse.group/t/sparse-tensors-in-mlir/3389/25
Reviewed By: gussmith23
Differential Revision: https://reviews.llvm.org/D104577
This change adds an option which, in addition to dumping the record
layout as is done by -fdump-record-layouts, causes us to compute the
layout for all complete record types (rather than the as-needed basis
which is usually done by clang), so that we will dump them as well.
This is useful if we are looking for layout differences across large
code bases without needing to instantiate every type we are interested in.
Reviewed By: dexonsmith
Differential Revision: https://reviews.llvm.org/D104484
The OpenMP 5.1 standard defines the environment variable
`OMP_TEAMS_THREAD_LIMIT` to limit the number of threads that will be run in a
single block. This patch adds support for this into the AMDGPU and CUDA
plugins.
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D103923
This used to be important for reducing lock contention when accessing identifiers, but
the cost of the cache can be quite large if parsing in a multi-threaded context. After
D104167, the win of keeping a cache is not worth the cost.
Differential Revision: https://reviews.llvm.org/D104737
Operations currently rely on the string name of attributes during attribute lookup/removal/replacement, in build methods, and more. This unfortunately means that some of the most used APIs in MLIR require string comparisons, additional hashing(+mutex locking) to construct Identifiers, and more. This revision remedies this by caching identifiers for all of the attributes of the operation in its corresponding AbstractOperation. Just updating the autogenerated usages brings up to a 15% reduction in compile time, greatly reducing the cost of interacting with the attributes of an operation. This number can grow even higher as we use these methods in handwritten C++ code.
Methods for accessing these cached identifiers are exposed via `<attr-name>AttrName` methods on the derived operation class. Moving forward, users should generally use these methods over raw strings when an attribute name is necessary.
Differential Revision: https://reviews.llvm.org/D104167
Currently the lambda body indents relative to where the lambda signature is located. This instead lets the user
choose to align the lambda body relative to the parent scope that contains the lambda declaration. Thus:
someFunction([] {
lambdaBody();
});
will always have the same indentation of the body even when the lambda signature goes on a new line:
someFunction(
[] {
lambdaBody();
});
whereas before lambdaBody would be indented 6 spaces.
Differential Revision: https://reviews.llvm.org/D102706
Global initializers may be ConstantArrays. They need to be checked
explicitly, because different-yet-still-equivalent type names may be
used for each, and/or a GEP instruction may appear in one.
The only wrinkle is that we can't process the "blockaddress" arguments
of the callbr until the blocks have been equated. So we force them to be
"unified" before checking.
This was left out when the callbr instruction was added.
Differential Revision: https://reviews.llvm.org/D104606
This reverts commit ed7086ad46.
This reverts commit b9792638b0.
This breaks cmake with message:
CMake Error at llvm-project/compiler-rt/CMakeLists.txt:449:
Parse error. Expected "(", got newline with text "
When the load type is changed to ptr, we need the load pointer type
to also be ptr, because it's not allowed to create a pointer to an
opaque pointer. This is achieved by adjusting the getPointerTo() API
to return an opaque pointer for an opaque pointer base type.
Differential Revision: https://reviews.llvm.org/D104718
Summary:
The changes introduced in D97680 turns this command line option into a no-op so
it can be removed entirely.
Reviewed By: tianshilei1992
Differential Revision: https://reviews.llvm.org/D102940
* indicates whether work has been started or completed
* consolidates content that was split for dependency reasons (iff
everything has been merged)
* makes things a lot more fine-grained
* turns sub-CSVs into lists
* puts links into description section and removes patch column
* adds links to c++draft on occasion
These changes heavily prioritise the the reader of the generated HTML
file, not the source.
Differential Revision: https://reviews.llvm.org/D103295
We want to disable the use of undefined symbols on Fuchsia, but there
are cases where it might be desirable so may it configurable.
Differential Revision: https://reviews.llvm.org/D104728
Right now the Attributor defaults to 32 fixed point iterations unless it is set
explicitly by a command line flag. This patch allows this to be configured when
the attributor instance is created. The maximum is then increased in OpenMPOpt
if the target is a kernel. This is because the globalization analysis can result
in larger iteration counts due to many dependent instances running at once.
Depends on D102444
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D104416
This reverts commit e1adf90826.
This appears to affect the way that C++ mangled symbols appear in the
import library when using a .def file that names a C++ free function
with no name decoration. I will follow up with a reduced test case
shortly.
Summary:
This patch adds support for the Attributor to emit remarks on behalf of some
other pass. The attributor can now optionally take a callback function that
returns an OptimizationRemarkEmitter object when given a Function pointer. If
this is availible then a remark will be emitted for the corresponding pass
name.
Depends on D102197
Reviewed By: sstefan1 thegameg
Differential Revision: https://reviews.llvm.org/D102444
These used to consistently be zeroed pre-gfx9, but gfx9 made the
situation complicated since now some still do and some don't. This
also manages to pick up a few cases that the pattern fails to optimize
away.
We handle some cases with instruction patterns, but some get
through. In particular this improves the integer cases.
C++03 didn't support `explicit` conversion operators;
but Clang's C++03 mode does, as an extension, so we can use it.
This lets us make the conversion explicit in `std::function` (even in '03),
and remove some silly metaprogramming in `std::basic_ios`.
Drive-by improvements to the tests for these operators, in addition
to making sure all these tests also run in `c++03` mode.
Differential Revision: https://reviews.llvm.org/D104682
Summary:
The changes to globalization introduced in D97680 introduce a large amount of overhead by default. The old globalization method would always ignore globalization code if executing in SPMD mode. This wasn't strictly correct as data sharing is still possible in SPMD mode. The new interface is correct but introduces globalization code even when unnecessary. This optimization will use the existing HeapToStack transformation in the attributor to allow for unneeded globalization to be replaced with thread-private stack memory. This is done using the newly introduced library instances for the RTL functions added in D102087.
Depends on D97818
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D102197
Summary:
The changes to globalization introduced in D97680 created two new functions to
push / pop shareably memory on the GPU, __kmpc_alloc_shared and
__kmpc_free_shared. This patch adds these new runtime functions to the
library info so they can be used by the HeapToStack attributor interface. This
optimization replaces malloc / free pairs with stack memory if legal.
Reviewed By: tianshilei1992
Differential Revision: https://reviews.llvm.org/D102087
0 latency instructions now get processed and retired properly within the in-order pipeline. Had to fix a bug within TimelineView.cpp as well that would show up when a 0 latency instruction was the first instruction in the source.
Differential Revision: https://reviews.llvm.org/D104675
We can do this optimization in the majority of cases, but we currently
don't have a way to do it. We do not track/model which instructions
have which behavior, the control bit to change the high bit behavior,
or making use of preserved bits at all. This is a bit fuzzy since we
don't know precisely how the source instruction will be lowered, but
that only really matters in one case (for fma_mixlo).
We do need to fixup some of these cases after selection, but the
pattern helps eliminate many of these zexts.
Summary:
generate eh_info when vector registers are saved according to the traceback table.
struct eh_info_t {
unsigned version; /* EH info version 0 */
#if defined(64BIT)
char _pad[4]; /* padding */
#endif
unsigned long lsda; /* Pointer to Language Specific Data Area */
unsigned long personality; /* Pointer to the personality routine */
};
the value of lsda and personality is zero when the number of vector registers saved is large zero and there is not personality of the function
Reviewers: Jason Liu
Differential Revision: https://reviews.llvm.org/D103651