Summary:
Use the new LoopVersioning facility (D16712) to add noalias metadata in
the vector loop if we versioned with memchecks. This can enable some
optimization opportunities further down the pipeline (see the included
test or the benchmark improvement quoted in D16712).
The test also covers the bug I had in the initial version in D16712.
The vectorizer did not previously use LoopVersioning. The reason is
that the vectorizer performs its transformations in single shot. It
creates an empty single-block vector loop that it then populates with
the widened, if-converted instructions. Thus creating an intermediate
versioned scalar loop seems wasteful.
So this patch (rather than bringing in LoopVersioning fully) adds a
special interface to LoopVersioning to allow the vectorizer to add
no-alias annotation while still performing its own versioning.
As the vectorizer propagates metadata from the instructions in the
original loop to the vector instructions we also check the pointer in
the original instruction and see if LoopVersioning can add no-alias
metadata based on the issued memchecks.
Reviewers: hfinkel, nadav, mzolotukhin
Subscribers: mzolotukhin, llvm-commits
Differential Revision: http://reviews.llvm.org/D17191
llvm-svn: 263744
Summary:
If we decide to version a loop to benefit a transformation, it makes
sense to record the now non-aliasing accesses in the newly versioned
loop. This allows non-aliasing information to be used by subsequent
passes.
One example is 456.hmmer in SPECint2006 where after loop distribution,
we vectorize one of the newly distributed loops. To vectorize we
version this loop to fully disambiguate may-aliasing accesses. If we
add the noalias markers, we can use the same information in a later DSE
pass to eliminate some dead stores which amounts to ~25% of the
instructions of this hot memory-pipeline-bound loop. The overall
performance improves by 18% on our ARM64.
The scoped noalias annotation is added in LoopVersioning. The patch
then enables this for loop distribution. A follow-on patch will enable
it for the vectorizer. Eventually this should be run by default when
versioning the loop but first I'd like to get some feedback whether my
understanding and application of scoped noalias metadata is correct.
Essentially my approach was to have a separate alias domain for each
versioning of the loop. For example, if we first version in loop
distribution and then in vectorization of the distributed loops, we have
a different set of memchecks for each versioning. By keeping the scopes
in different domains they can conveniently be defined independently
since different alias domains don't affect each other.
As written, I also have a separate domain for each loop. This is not
necessary and we could save some metadata here by using the same domain
across the different loops. I don't think it's a big deal either way.
Probably the best is to review the tests first to see if I mapped this
problem correctly to scoped noalias markers. I have plenty of comments
in the tests.
Note that the interface is prepared for the vectorizer which needs the
annotateInstWithNoAlias API. The vectorizer does not use LoopVersioning
so we need a way to pass in the versioned instructions. This is also
why the maps have to become part of the object state.
Also currently, we only have an AA-aware DSE after the vectorizer if we
also run the LTO pipeline. Depending how widely this triggers we may
want to schedule a DSE toward the end of the regular pass pipeline.
Reviewers: hfinkel, nadav, ashutosh.nema
Subscribers: mssimpso, aemerson, llvm-commits, mcrosier
Differential Revision: http://reviews.llvm.org/D16712
llvm-svn: 263743
I hit a crash in the bitcode reader on some corrupt input where an
MDString had somehow been attached to an instruction instead of an
MDNode. This input is pretty bogus, but we shouldn't be crashing on bad
input here.
This change adds error handling in all of the places where we
currently have unchecked casts from Metadata to MDNode, which means
we'll error out instead of crashing for that sort of input.
Unfortunately, I don't have tests. Hitting this requires flipping bits
in the input bitcode, and committing corrupt binary files to catch
these cases is a bit too opaque and unmaintainable.
llvm-svn: 263742
Summary: ...as that is apparently what MSVC does
Reviewers: rnk
Patch by Stephan Bergmann
Differential Revision: http://reviews.llvm.org/D15267
llvm-svn: 263738
Summary:
The gdb-remote async thread cannot modify thread state while the main thread
holds a lock on the state. Don't use locking thread iteration for bt all.
Specifically, the deadlock manifests when lldb attempts to JIT code to
symbolicate objective c while backtracing. As part of this code path,
SetPrivateState() is called on an async thread. This async thread will
block waiting for the thread_list lock held by the main thread in
CommandObjectIterateOverThreads. The main thread will also block on the
async thread during DoResume (although with a timeout), leading to a
deadlock. Due to the timeout, the deadlock is not immediately apparent,
but the inferior will be left in an invalid state after the bt all completes,
and objective-c symbols will not be successfully resolved in the backtrace.
Reviewers: andrew.w.kaylor, jingham, clayborg
Subscribers: sas, lldb-commits
Differential Revision: http://reviews.llvm.org/D18075
Change by Francis Ricci <fjricci@fb.com>
llvm-svn: 263735
Summary:
The multiprocessing.Queue.put() call can hang if we try queueing all the
tests before starting to take them out of the queue.
The current implementation hangs if tests exceed 2^^15, on Mac OS X.
This might happen with a ninja check-all if one has a bunch of llvm
projects.
Reviewers: delcypher, bkramer
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D17609
llvm-svn: 263731
Autogenerated from the corresponding assembler tests with a few FIXME added (will fix soon).
Differential Revision: http://reviews.llvm.org/D18249
llvm-svn: 263729
This patch prevents CTR loops optimization when using soft float operations
inside loop body. Soft float operations use function calls, but function
calls are not allowed inside CTR optimized loops.
Patch by Aleksandar Beserminji.
Differential Revision: http://reviews.llvm.org/D17600
llvm-svn: 263727
Summary:
MRI::eliminateFrameIndex can emit several instructions to do address
calculations; these can usually be stackified. Because instructions with
FI operands can have subsequent operands which may be expression trees,
find the top of the leftmost tree and insert the code before it, to keep
the LIFO property.
Also use stackified registers when writing back the SP value to memory
in the epilog; it's unnecessary because SP will not be used after the
epilog, and it results in better code.
Differential Revision: http://reviews.llvm.org/D18234
llvm-svn: 263725
Some COFF tests used INT_MIN for the alignment of the directive section.
This is invalid; replace the alignment with something more sensible: 1.
llvm-svn: 263723
Symmary:
ds_permute/ds_bpermute do not read memory so s_waitcnt is not needed.
Reviewers
arsenm, tstellarAMD
Subscribers
llvm-commits, arsenm
Differential Revision:
http://reviews.llvm.org/D18197
llvm-svn: 263720
Summary:
As explained by the comment, threads will typically see different values
returned by atomic instructions even if the arguments are equal.
Reviewers: arsenm, tstellarAMD
Subscribers: arsenm, llvm-commits
Differential Revision: http://reviews.llvm.org/D18156
llvm-svn: 263719
The System/ mock is large and too complex for this test. It can cause
the tests to fail in mysterious ways as it depends on the resource dir
being present, which is not really supported for driver tests (using
%clang instead of %clang_cc1). Copy the tree and trim out all the
%unnecessary fat.
llvm-svn: 263718
We were being too aggressive in trying to combine a shuffle into a blend-with-zero pattern, often resulting in a endless loop of contrasting combines
This patch stops the combine if we already have a blend in place (means we miss some domain corrections)
llvm-svn: 263717
This is similar to D18133 where we allowed profile weights on select instructions.
This extends that change to also allow the 'unpredictable' attribute of branches to apply to selects.
A test to check that 'unpredictable' metadata is preserved when cloning instructions was checked in at:
http://reviews.llvm.org/rL263648
Differential Revision: http://reviews.llvm.org/D18220
llvm-svn: 263716
The two changes together weakened the test and caused a regression with division
handling in MSVC mode. They were applied to avoid an assertion being triggered
in the block frequency analysis. However, the underlying problem was simply
being masked rather than solved properly. Address the actual underlying problem
and revert the changes. Rather than analyze the cause of the assertion, the
division failure was assumed to be an overflow.
The underlying issue was a subtle bug in the BB construction in the emission of
the div-by-zero check (WIN__DBZCHK). We did not construct the proper successor
information in the basic blocks, nor did we update the PHIs associated with the
basic block when we split them. This would result in assertions being triggered
in the block frequency analysis pass.
Although the original tests are being removed, the tests themselves performed
very little in terms of validation but merely tested that we did not assert when
generating code. Update this with new tests that actually ensure that we do not
regress on the code generation.
llvm-svn: 263714
If a call takes a single argument, using AlwaysBreak can lead to lots
of wasted lines and additional indentation without improving the
readability in a significant way.
Before:
caaaaaaaaaaaall(
caaaaaaaaaaaall(
caaaaaaaaaaaall(
caaaaaaaaaaaaaaaaaaaaaaall(aaaaaaaaaaaaaa, aaaaaaaaa))));
After:
caaaaaaaaaaaall(caaaaaaaaaaaall(caaaaaaaaaaaall(
caaaaaaaaaaaaaaaaaaaaaaall(aaaaaaaaaaaaaa, aaaaaaaaa))));
llvm-svn: 263709
OpenMP 4.5 allows privatization of non-static data members in non-static
member functions. This patch adds support of private data members in
'copyprivate' clauses.
llvm-svn: 263706
It might be hard to recognize a hexadecimal number without '0x' prefix.
Besides that '0x' prefix corresponds to GNU objdump behaviour.
Differential Revision: http://reviews.llvm.org/D18207
llvm-svn: 263705
That allows, for example, to print hex-formatted immediates using
llvm-objdump --print-imm-hex command line option.
Differential Revision: http://reviews.llvm.org/D18195
llvm-svn: 263704
Summary:
This should eliminate all occurrences of this within LLVMMipsAsmParser.
This patch is in response to http://reviews.llvm.org/D17983. I was unable
to reproduce the warnings on my machine so please advise if this fixes the
warnings.
Reviewers: ariccio, vkalintiris, dsanders
Subscribers: dblaikie, dsanders, llvm-commits
Differential Revision: http://reviews.llvm.org/D18087
llvm-svn: 263703
OpenMP 4.0 allows to define custom reduction operations using '#pragma
omp declare reduction' construct. Patch allows to use this custom
defined reduction operations in 'reduction' clauses.
llvm-svn: 263701
The feature check is already in place when building the library but wasn't
honored for the tests.
Differential Revision: http://reviews.llvm.org/D18205
llvm-svn: 263699
On OS X 10.11+, we have "automatic interceptors", so we don't need to use DYLD_INSERT_LIBRARIES when launching instrumented programs. However, non-instrumented programs that load TSan late (e.g. via dlopen) are currently broken, as TSan will still try to initialize, but the program will crash/hang at random places (because the interceptors don't work). This patch adds an explicit check that interceptors are working, and if not, it aborts and prints out an error message suggesting to explicitly use DYLD_INSERT_LIBRARIES.
TSan unit tests run with a statically linked runtime, where interceptors don't work. To avoid aborting the process in this case, the patch replaces `DisableReexec()` with a weak `ReexecDisabled()` function which is defined to return true in unit tests.
Differential Revision: http://reviews.llvm.org/D18212
llvm-svn: 263695
-pie
--pic-executable
Create a position independent executable. This is currently only
supported on ELF platforms. Position independent executables are
similar to shared libraries in that they are relocated by the
dynamic linker to the virtual address the OS chooses for them
(which can vary between invocations). Like normal dynamically
linked executables they can be executed and symbols defined in the
executable cannot be overridden by shared libraries.
Differential revision: http://reviews.llvm.org/D18183
llvm-svn: 263693
Adam Nemet