Extend OpDSL with a `defines` method that can set the `hasCanonicalizer` flag for an OpDSL operation. If the flag is set via `defines(Canonicalizer)` the operation needs to implement the `getCanonicalizationPatterns` method. The revision specifies the flag for linalg.fill_tensor and adds an empty `FillTensorOp::getCanonicalizationPatterns` implementation.
This revision is a preparation step to replace linalg.fill by its OpDSL counterpart linalg.fill_tensor. The two are only functionally equivalent if both specify the same canonicalization patterns. The revision is thus a prerequisite for the linalg.fill replacement.
Depends On D120725
Reviewed By: nicolasvasilache
Differential Revision: https://reviews.llvm.org/D120726
Enhance `LinalgTileAndFuseTensorOpsPattern` with an additional rewrite signature that returns the result of the rewrite.
Reviewed By: nicolasvasilache
Differential Revision: https://reviews.llvm.org/D121212
Add a FillOpInterface similar to the contraction and convolution op interfaces. The FillOpInterface is a preparation step to replace linalg.fill by its OpDSL version linalg.fill_tensor. The interface implements the `value()`, `output()`, and `result()` methods that by default are not available on linalg.fill_tensor.
Reviewed By: nicolasvasilache
Differential Revision: https://reviews.llvm.org/D120725
As a fly-by fix, enable the complexity-changing assertions in __debug_less
only when the full debug mode is enabled, since debugging level 0 is usually
understood to only contain basic assertions that do not change the complexity
of algorithms.
Differential Revision: https://reviews.llvm.org/D121129
We were not creating an evaluation context for the TU scope, so we
never popped an evaluation context for it. Popping the evaluation
context triggers a number of diagnostics, including warnings about
immediate invocations that we were previously missing.
Note: I think we have an additional issue that we should solve, but not
as part of this patch. I don't think Clang is properly modeling static
initialization as happening before constant expression evaluation. I
think structure members members are zero initialized per
http://eel.is/c++draft/basic.start.static#1,
https://eel.is/c++draft/basic.start.static#2.sentence-2, and
http://eel.is/c++draft/dcl.init#general-6.2 and the new test case
actually should be accepted. However, it's also worth noting that other
compilers behave the way this patch makes Clang behave:
https://godbolt.org/z/T7noqhdPr
The existing module symbol mangling scheme turns out to be
undemangleable. It is also desirable to switch to the
strong-ownership model as the hoped-for C++17 compatibility turns out
to be fragile, and we also now have a better way of controlling that.
The issue is captured on the ABI list at:
https://github.com/itanium-cxx-abi/cxx-abi/issues/134
A document describing the issues and new mangling is at:
https://drive.google.com/file/d/1qQjqptzOFT_lfXH8L6-iD9nCRi34wjft/view
This patch is the code-generation part. I have a demangler too, but
that patch is based on some to-be-landed refactoring of the demangler.
The old mangling is unceremoniously dropped. No backwards
compatibility, no deprectated old-mangling flag. It was always
labelled experimental. (Old and new manglings cannot be confused.)
Reviewed By: ChuanqiXu
Differential Revision: https://reviews.llvm.org/D118352
ld64 breaks down `__objc_classrefs` on a per-word level and deduplicates
them. This greatly reduces the number of bind entries emitted (and
therefore the amount of work `dyld` has to do at runtime). For
chromium_framework, this change to LLD cuts the number of (non-lazy)
binds from 912 to 190, getting us to parity with ld64 in this aspect.
Reviewed By: #lld-macho, thakis
Differential Revision: https://reviews.llvm.org/D121053
`__cfstring` has embedded addends that foil ICF's hashing / equality
checks. (We can ignore embedded addends when doing ICF because the same
information gets recorded in our Reloc structs.) Therefore, in order to
properly dedup CFStrings, we create a mutable copy of the CFString and
zero out the embedded addends before performing any hashing / equality
checks.
(We did in fact have a partial implementation of CFString deduplication
already. However, it only worked when the cstrings they point to are at
identical offsets in their object files.)
I anticipate this approach can be extended to other similar
statically-allocated struct sections in the future.
In addition, we previously treated all references with differing addends
as unequal. This is not true when the references are to literals:
different addends may point to the same literal in the output binary. In
particular, `__cfstring` has such references to `__cstring`. I've
adjusted ICF's `equalsConstant` logic accordingly, and I've added a few
more tests to make sure the addend-comparison code path is adequately
covered.
Fixes https://github.com/llvm/llvm-project/issues/51281.
Reviewed By: #lld-macho, Roger
Differential Revision: https://reviews.llvm.org/D120137
These are some checks that make sense in libc++ IMO. The checks after `#TODO: investigate these checks` are candidates, but they can't be enabled without some cleanup.
Reviewed By: ldionne, #libc
Spies: aheejin, libcxx-commits
Differential Revision: https://reviews.llvm.org/D120925
This patch adds tail call support to the 32-bit Sparc backend.
Two new instructions are defined, TAIL_CALL and TAIL_CALLri. They are
encoded the same as CALL and BINDri, but are marked with isReturn so
that the epilogue gets emitted. In contrast to CALL, TAIL_CALL is not
marked with isCall. This makes it possible to use the leaf function
optimization when the only call a function makes is a tail call.
TAIL_CALL modifies the return address in %o7, so for leaf functions
the value in %o7 needs to be restored after the call. For normal
functions which uses the restore instruction this is not necessary.
Reviewed By: koakuma
Differential Revision: https://reviews.llvm.org/D51206
Make a further improvement to decrease verbosity of the API: ASTContext
provides SourceManager access.
Reviewed By: sammccall
Differential Revision: https://reviews.llvm.org/D119842
Before, the code:
```
int Value { get; } = 0;
int Value { init; } = 0;
```
was formatted incoherently:
```
int Value { get; } = 0;
int Value { init; }
= 0;
```
because `init` was not recognised as an accessor specifier.
Reviewed By: MyDeveloperDay, HazardyKnusperkeks
Differential Revision: https://reviews.llvm.org/D121132
As noticed in D119654, by adding the masked intrinsics results together
we can end up with the selects being canonicalized away from the
intrinsic - this isn't what we want to test here so replace with a
insertvalue chain into a aggregate instead to retain all the results.
Differential Revision: https://reviews.llvm.org/D121196
FdClose is a subjet to the same atomicity problem as MemoryRangeFreed
(memory state is not "monotoic" wrt race detection).
So we need to lock the thread slot in FdClose the same way we do
in MemoryRangeFreed.
This fixes the modified stress.cpp test.
Reviewed By: vitalybuka, melver
Differential Revision: https://reviews.llvm.org/D121143
This patch disables TestIOHandlerProcessSTDIO.py for Arm/AArch64 Linux
to silence random test failures on buildbots. IO handler tests are known
to randomly fail on arm/aarch64 linux buildbots due to pexpect timeouts.
Currently, the transfer mask is materialized by generating the vector
comparison: [offset + 0, .., offset + length - 1] < [dim, .., dim]
A better alternative is to materialize the transfer mask by using the
operation: `vector.create_mask (dim - offset)`, which will generate
simpler code and compose better with scalable vectors.
Differential Revision: https://reviews.llvm.org/D120487
Otherwise we will fail to generate the definition of a defaulted destructor,
if the only reference was in a templated temporary.
rdar://89366678
Differential Revision: https://reviews.llvm.org/D120426
As noticed in D119654, by adding the masked intrinsics results together
we can end up with the selects being canonicalized away from the
intrinsic - this isn't what we want to test here so replace with a
insertvalue chain into a aggregate instead to retain all the results.
Differential Revision: https://reviews.llvm.org/D121188
Introduce a new attribute "function-inline-cost-multiplier" which
multiplies the inline cost of a call site (or all calls to a callee) by
the multiplier.
When processing the list of calls created by inlining, check each call
to see if the new call's callee is in the same SCC as the original
callee. If so, set the "function-inline-cost-multiplier" attribute of
the new call site to double the original call site's attribute value.
This does not happen when the original call site is intra-SCC.
This is an alternative to D120584, which marks the call sites as
noinline.
Hopefully fixes PR45253.
Reviewed By: davidxl
Differential Revision: https://reviews.llvm.org/D121084
A load via pointer cast to constant will return true from
pointsToConstantMemory which is not necessarily so.
Fixes: SWDEV-326463
Differential Revision: https://reviews.llvm.org/D121172
This is useful when building a complete toolchain to ensure that CRT
is built after builtins but before the rest of the compiler-rt.
Differential Revision: https://reviews.llvm.org/D120682
Add support of parsing .csky_attribute directive and emit related target attributes in .csky.attribute section.
It does not emit attribute directive in assembly code, so only emit target attributes in ELF streamer.
In ELF streamer, it handles the header EFlag and the csky_attribute section which contains some attribute items.
The EFlag and attribute items are calculated from feature bits based on Subtarget.
This relands commit 731347431976509823e38329a96fcbc69fe98cd2.
It failed on Windows/Mac because `-fjmc` is only checked for ELF targets.
Check the flag unconditionally instead and issue a warning for non-ELF targets.
If the user disables de-globalization we did not seed the AAHeapToShared
and AAHeapToStack but we still could end up with them through in-flight
lookups. With this patch we disable AAHeapToShared completely if the
user disabled de-globalization. Heap-2-stack is still run though.
Differential Revision: https://reviews.llvm.org/D121059
An event pool, similar to the stream pool, needs to be kept per device.
For one, events are associated with cuda contexts which means we cannot
destroy the former after the latter. Also, CUDA documentation states
streams and events need to be associated with the same context, which
we did not ensure at all.
Differential Revision: https://reviews.llvm.org/D120142
There are two problems this patch tries to address:
1) We currently free resources in a random order wrt. plugin and
libomptarget destruction. This patch should ensure the CUDA plugin
is less fragile if something during the deinitialization goes wrong.
2) We need to support (hard) pause runtime calls eventually. This patch
allows us to free all associated resources, though we cannot
reinitialize the device yet.
Follow up patch will associate one event pool per device/context.
Differential Revision: https://reviews.llvm.org/D120089
Rewrite isInnermostAffineForOp utility to make it more direct/efficient.
Drop unnecessary check. NFC.
Differential Revision: https://reviews.llvm.org/D121170
This patch removes the unintended resolution of locally scoped absolute symbols
(which was causing unexpected definition errors).
It stops using the JITSymbolFlags::Absolute flag (it isn't set or used elsewhere,
and causes mismatch-flags asserts), and adds JITSymbolFlags::Exported to default
scoped absolute symbols.
Finally, we now set the scope of absolute symbols correctly in
MachOLinkGraphBuilder.
Use TII::getRegClass to return a valid regclass or a nullptr
if the RC is unknown for a given OpIdx. This fixes a potential
crash occurred while getting the RC from a variadic instruction.
Reviewed By: arsenm
Differential Revision: https://reviews.llvm.org/D120813
