The loop on IVOperand's incoming values assumes IVOperand to be an
induction variable on the loop over which `S Pred X` is invariant;
otherwise loop invariant incoming values to IVOperand are not guaranteed
to dominate the comparision.
This fixes PR26973.
llvm-svn: 263827
In the <DisplayString> of PointerIntPair , I cast the pointer to the actual type, so VS can leverage it while visualizing, not unlike the recent change to PointerUnion visualization.
In the expansion, the current code is casting to the incorrect type (wrong number of stars), so I fixed that as well.
llvm-svn: 263821
This patch adds unscaled loads and sign-extend loads to the TII
getMemOpBaseRegImmOfs API, which is used to control clustering in the MI
scheduler. This is done to create more opportunities for load pairing. I've
also added the scaled LDRSWui instruction, which was missing from the scaled
instructions. Finally, I've added support in shouldClusterLoads for clustering
adjacent sext and zext loads that too can be paired by the load/store optimizer.
Differential Revision: http://reviews.llvm.org/D18048
llvm-svn: 263819
Summary:
These dependencies would be used in the future to reduce the number
of instrumented blocks(http://reviews.llvm.org/rL262103)
This is submitted as a separate CL because of previous problems with
ARM.
Subscribers: aemerson
Differential Revision: http://reviews.llvm.org/D18227
llvm-svn: 263797
Summary:
Allow the selection of BUFFER_LOAD_FORMAT_x and _XY. Do this now before
the frontend patches land in Mesa. Eventually, we may want to automatically
reduce the size of loads at the LLVM IR level, which requires such overloads,
and in some cases Mesa can generate them directly.
Reviewers: tstellarAMD, arsenm
Subscribers: arsenm, llvm-commits
Differential Revision: http://reviews.llvm.org/D18255
llvm-svn: 263792
Summary:
These intrinsics expose the BUFFER_ATOMIC_* instructions and will be used
by Mesa to implement atomics with buffer semantics. The intrinsic interface
matches that of buffer.load.format and buffer.store.format, except that the
GLC bit is not exposed (it is automatically deduced based on whether the
return value is used).
The change of hasSideEffects is required for TableGen to accept the pattern
that matches the intrinsic.
Reviewers: tstellarAMD, arsenm
Subscribers: arsenm, rivanvx, llvm-commits
Differential Revision: http://reviews.llvm.org/D18151
llvm-svn: 263791
Summary:
We cannot easily deduce that an offset is in an SGPR, but the Mesa frontend
cannot easily make use of an explicit soffset parameter either. Furthermore,
it is likely that in the future, LLVM will be in a better position than the
frontend to choose an SGPR offset if possible.
Since there aren't any frontend uses of these intrinsics in upstream
repositories yet, I would like to take this opportunity to change the
intrinsic signatures to a single offset parameter, which is then selected
to immediate offsets or voffsets using a ComplexPattern.
Reviewers: arsenm, tstellarAMD, mareko
Subscribers: arsenm, llvm-commits
Differential Revision: http://reviews.llvm.org/D18218
llvm-svn: 263790
Now that the resolved path cache stores the StringRef's, its
best to just always cache the results, even when realpath isn't
used. This way we'll still avoid the StringMap hashing and lookup.
This also conveniently reorganises this code in a way I need for
a future patch.
llvm-svn: 263777
ResolvedPaths was storing std::string's as a cache. We would then take those strings and look them up in the internString pool to get a unique StringRef for each path.
This patch changes ResolvedPaths to store the StringRef pointing in to the internString pool itself. This way, when getResolvedPath returns a string, we know we have the StringRef we would find in the pool anyway. We can avoid the duplicate memory of the std::string's, and also the time from the lookup.
Unfortunately my profiles show no runtime change here, but it should still save memory allocations which is nice.
Reviewed by Frederic Riss.
Differential Revision: http://reviews.llvm.org/D18259
llvm-svn: 263774
Summary:
It can hurt performance to prefetch ahead too much. Be conservative for
now and don't prefetch ahead more than 3 iterations on Cyclone.
Reviewers: hfinkel
Subscribers: llvm-commits, mzolotukhin
Differential Revision: http://reviews.llvm.org/D17949
llvm-svn: 263772
Summary:
And use this TTI for Cyclone. As it was explained in the original RFC
(http://thread.gmane.org/gmane.comp.compilers.llvm.devel/92758), the HW
prefetcher work up to 2KB strides.
I am also adding tests for this and the previous change (D17943):
* Cyclone prefetching accesses with a large stride
* Cyclone not prefetching accesses with a small stride
* Generic Aarch64 subtarget not prefetching either
Reviewers: hfinkel
Subscribers: aemerson, rengolin, llvm-commits, mzolotukhin
Differential Revision: http://reviews.llvm.org/D17945
llvm-svn: 263771
Summary:
This wires up the pass for Cyclone but keeps it off for now because we
need a few more TTIs.
The getPrefetchMinStride value is not very well tuned right now but it
works well with CFP2006/433.milc which motivated this.
Tests will be added as part of the upcoming large-stride prefetching
patch.
Reviewers: t.p.northover
Subscribers: llvm-commits, aemerson, hfinkel, rengolin
Differential Revision: http://reviews.llvm.org/D17943
llvm-svn: 263770
A virtual index of -1u indicates that the subprogram's virtual index is
unrepresentable (for example, when using the relative vtable ABI), so do
not emit a DW_AT_vtable_elem_location attribute for it.
Differential Revision: http://reviews.llvm.org/D18236
llvm-svn: 263765
MSVC as usual:
C:\Buildbot\Slave\llvm-clang-lld-x86_64-scei-ps4-windows10pro-fast\llvm.src\include\llvm/ADT/STLExtras.h(120):
error C2100: illegal indirection
C:\Buildbot\Slave\llvm-clang-lld-x86_64-scei-ps4-windows10pro-fast\llvm.src\include\llvm/IR/Instructions.h(3966):
note: see reference to class template instantiation
'llvm::mapped_iterator<llvm::User::op_iterator,llvm::CatchSwitchInst::DerefFnTy>'
being compiled
This reverts commit e091dd63f1f34e043748e28ad160d3bc17731168.
llvm-svn: 263760
For fcmp, major concern about the following 6 cases is NaN result. The
comparison result consists of 4 bits, indicating lt, eq, gt and un (unordered),
only one of which will be set. The result is generated by fcmpu
instruction. However, bc instruction only inspects one of the first 3
bits, so when un is set, bc instruction may jump to to an undesired
place.
More specifically, if we expect an unordered comparison and un is set, we
expect to always go to true branch; in such case UEQ, UGT and ULT still
give false, which are undesired; but UNE, UGE, ULE happen to give true,
since they are tested by inspecting !eq, !lt, !gt, respectively.
Similarly, for ordered comparison, when un is set, we always expect the
result to be false. In such case OGT, OLT and OEQ is good, since they are
actually testing GT, LT, and EQ respectively, which are false. OGE, OLE
and ONE are tested through !lt, !gt and !eq, and these are true.
llvm-svn: 263753
idiom.
Most LLVM tool code exits immediately when an error is encountered and prints an
error message to stderr. The ExitOnError class supports this by providing two
call operators - one for Errors, and one for Expected<T>s. Calls to code that
can return Errors (or Expected<T>s) can use these calls to bail out on error,
and otherwise continue as if the operation had succeeded. E.g.
Error foo();
Expected<int> bar();
int main(int argc, char *argv[]) {
ExitOnError ExitOnErr;
ExitOnErr.setBanner(std::string("Error in ") + argv[0] + ":");
// Exit if foo returns an error. No need to manually check error return.
ExitOnErr(foo());
// Exit if bar returns an error, otherwise unwrap the contained int and
// continue.
int X = ExitOnErr(bar());
// ...
return 0;
}
llvm-svn: 263749
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:
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
