Broadwell/Haswell were completely overriding the class defs - we can remove those overrides entirely by just choosing better class defs (plus a fix for missing mmx folded load).
The `target_link_libraries()` call operates on `gtest_main` rather than
`llvm_gtest_main`. I think it was missed while prefixing all the tools
in 38151a08c21e0cdacd52af03e4bdff258f0bac6f.
Differential Revision: https://reviews.llvm.org/D137080
Unswitching adjusts the CFG in ways that may invalidate cached loop
dispositions. Clear all cached block and loop dispositions during
trivial unswitching. The same is already done for non-trivial
unswitching.
Fixes#58751.
The WriteCvtSD2SS/WriteCvtPD2PS* classes were mostly unused as the models were needlessly overriding all instructions - in some cases the folded pattern overrides were entirely missing (but I've confirmed they just have an additional Port23 use)
There were a couple of typos (confirmed with Agner/uops.info) - Skylake/Icelake uses Port5+Port01 for XMM/YMM, Skylake uses Port5+Port05 for ZMM but Icelake uses Port5+Port0
This mirrors what we have done in the classic algorithms
Reviewed By: ldionne, #libc
Spies: libcxx-commits
Differential Revision: https://reviews.llvm.org/D137186
Represent the select type statement + type guard statement
the same way the select case statement and case statement are represented.
controlSuccessor was not correctly attributed to the next type guard stmt.
Reviewed By: PeteSteinfeld, vdonaldson
Differential Revision: https://reviews.llvm.org/D137460
Replaces hard coded uses of WriteVPMOV256 for all the vector truncations instructions
We still need to work out how to fix folded stores (see Issue #36236)
This was never added in D52886, probably because the defs were already missing the old ReadAfterLd attribute, but they do exhibit the same behaviour of reading the dst reg after the load.
Just use the default WriteCvtSS2I/WriteCvtSD2I folded variants (already used by the VCVTSS2SI/VCVTSD2SI AVX variants).
Confirmed with agner and uops.info - there should be no difference between the SSE/AVX1 variants in folded load latency
Broadwell, Haswell and SkylakeClient were completely overriding the WriteCvtI2PD defs - we can remove those overrides entirely by replacing the unused WriteCvtI2PD values
There's plenty more of these in the scheduler models - I'm looking at improving warnings in llvm-tblgen to catch them all
This patch replaces the backing file path key to "file_path" to keep it
consistent.
rdar://101652618
Signed-off-by: Med Ismail Bennani <medismail.bennani@gmail.com>
It can happen that the originator of a crash report doesn't have access
to certain images. When that's the case, ReportCrash won't show the
source info in the crash report stack frames, but only the stack address
and image name.
This patch fixes a bug in the crashlog stackframe parser regular
expression to optionally match the source info group.
rdar://101934135
Differential Revision: https://reviews.llvm.org/D137466
Signed-off-by: Med Ismail Bennani <medismail.bennani@gmail.com>
The KeyTy of attribute/type storage classes provide enough information for
automatically implementing the necessary sub element interface methods. This
removes the need for derived classes to do it themselves, which is both much
nicer and easier to handle certain invariants (e.g. null handling). In cases where
explicitly handling for parameter types is necessary, they can provide an implementation
of `AttrTypeSubElementHandler` to opt-in to support.
This tickles a few things alias wise, which annoyingly messes with tests that hard
code specific affine map numbers.
Differential Revision: https://reviews.llvm.org/D137374
The first result of the extract_strided_metadata operation is a MemRef,
not a naked pointer.
This patch fixes the lowering of this operation in MemRefToLLVM so that
we properly materialize the full MemRef structure and not just the base,
naked, pointer.
Differential Revision: https://reviews.llvm.org/D137364
But I can not reproduce the problem on my local machine. My local machine run:
222 0x5a6780
222 0x7fffbef9400e
222 0x5a677e 0x5a6780 0x7fffbef936c8
222 0x376f8e 0x376f90 0x7fffbef94008
222 0x281f20
222 0x7fffbef9400e
PASSED
Previously, we only checked for duplicate zero entries when merging a
MemOPSize table (see D92074), but a user recently provided a reproducer
demonstrating that other entries can also be duplicated. As demonstrated
by the test in this patch, PGOMemOPSizeOpt can potentially generate
invalid IR for non-zero, non-consecutive duplicate entries. This seems
to be a rare case, since the duplicate entry is often below the
threshold, but possible. This patch extends the existing warning to
check for any duplicate values in the table, both in the optimization
and in llvm-profdata.
Differential Revision: https://reviews.llvm.org/D136211
In `MLInlineAdvisor::getAdviceImpl`, we call `getCachedFPI` twice, once
for the caller, once for the callee, so the second may invalidate the
reference obtained by the first because the underlying implementation of
the cache is a `DenseMap`. `std::map` doesn't have that problem.
Sometimes libc++'s stddef.h wrapper gets included, which defines
::nullptr_t. This test is compiled with -Wshadow -Werror, so shadowing
::nullptr_t with a nullptr_t in main is an error. Include cstddef,
which is guaranteed to define std::nullptr_t in C++11 and forward.
Reviewed By: ldionne, #libc_abi
Differential Revision: https://reviews.llvm.org/D137127
It is caused by regenerate captured var value when processing the
has_device_addr, the captured var value has been generated in
GenerateOpenMPCapturedVars and passed as Arg in generateInfoForCapture.
The fix just use Arg instead regenerated just same as is_device_ptr
This was trying to add segments beyond the new and use,
so skip additional segments.
This would hit (S < E && "Cannot create empty or backwards segment").
This is for case when native target like X86 is not in LLVM_TARGETS_TO_BUILD.
Right now LLVM_DEFAULT_TARGET_TRIPLE is set to LLVM_HOST_TRIPLE even when native target is not enabled,
As a result, many lit tests will fail because default_triple is set for lit test but not enabled when build LLVM.
Reviewed By: smeenai
Differential Revision: https://reviews.llvm.org/D134972
