Summary:
If we attempt to unfold an SUnit in ScheduleDAG that results in
finding an already scheduled load, we must should abort the
unfold as it will not improve scheduling.
This fixes PR32610.
Reviewers: jmolloy, sunfish, bogner, spatel
Subscribers: llvm-commits, MatzeB
Differential Revision: https://reviews.llvm.org/D32911
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@304321 91177308-0d34-0410-b5e6-96231b3b80d8
Correct references to alignment of store which may be deleted in a
previous iteration of merge. Instead use first store that would be
merged.
Corrects pr33172's use-after-poison caught by ASan.
Reviewers: spatel, hfinkel, RKSimon
Reviewed By: RKSimon
Subscribers: thegameg, javed.absar, llvm-commits
Differential Revision: https://reviews.llvm.org/D33686
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@304299 91177308-0d34-0410-b5e6-96231b3b80d8
This code was compensating for FPOWI defaulting to Legal and many targets not changing it to Expand. This was fixed in r304215 to default to Expand so this special handling should no longer be necessary.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@304221 91177308-0d34-0410-b5e6-96231b3b80d8
Summary:
Currently FPOWI defaults to Legal and LegalizeDAG.cpp turns Legal into Expand for this opcode because Legal is a "lie".
This patch changes the default for this opcode to Expand and removes the hack from LegalizeDAG.cpp. It also removes all the code in the targets that set this opcode to Expand themselves since they can just rely on the default.
Reviewers: spatel, RKSimon, efriedma
Reviewed By: RKSimon
Subscribers: jfb, dschuff, sbc100, jgravelle-google, nemanjai, javed.absar, andrew.w.kaylor, llvm-commits
Differential Revision: https://reviews.llvm.org/D33530
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@304215 91177308-0d34-0410-b5e6-96231b3b80d8
If we have (extract_subvector(load wide vector)) with no other users,
that can just be (load narrow vector). This is intentionally conservative.
Follow-ups may loosen the one-use constraint to account for the extract cost
or just remove the one-use check.
The memop chain updating is based on code that already exists multiple times
in x86 lowering, so that should be pulled into a helper function as a follow-up.
Background: this is a potential improvement noticed via regressions caused by
making x86's peekThroughBitcasts() not loop on consecutive bitcasts (see
comments in D33137).
Differential Revision: https://reviews.llvm.org/D33578
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@304072 91177308-0d34-0410-b5e6-96231b3b80d8
In the best case:
extract (binop (concat X1, X2), (concat Y1, Y2)), N --> binop XN, YN
...we kill all of the extract/concat and just have narrow binops remaining.
If only one of the binop operands is amenable, this transform is still
worthwhile because we kill some of the extract/concat.
Optional bitcasting makes the code more complicated, but there doesn't
seem to be a way to avoid that.
The TODO about extending to more than bitwise logic is there because we really
will regress several x86 tests including madd, psad, and even a plain
integer-multiply-by-2 or shift-left-by-1. I don't think there's anything
fundamentally wrong with this patch that would cause those regressions; those
folds are just missing or brittle.
If we extend to more binops, I found that this patch will fire on at least one
non-x86 regression test. There's an ARM NEON test in
test/CodeGen/ARM/coalesce-subregs.ll with a pattern like:
t5: v2f32 = vector_shuffle<0,3> t2, t4
t6: v1i64 = bitcast t5
t8: v1i64 = BUILD_VECTOR Constant:i64<0>
t9: v2i64 = concat_vectors t6, t8
t10: v4f32 = bitcast t9
t12: v4f32 = fmul t11, t10
t13: v2i64 = bitcast t12
t16: v1i64 = extract_subvector t13, Constant:i32<0>
There was no functional change in the codegen from this transform from what I
could see though.
For the x86 test changes:
1. PR32790() is the closest call. We don't reduce the AVX1 instruction count in that case,
but we improve throughput. Also, on a core like Jaguar that double-pumps 256-bit ops,
there's an unseen win because two 128-bit ops have the same cost as the wider 256-bit op.
SSE/AVX2/AXV512 are not affected which is expected because only AVX1 has the extract/concat
ops to match the pattern.
2. do_not_use_256bit_op() is the best case. Everyone wins by avoiding the concat/extract.
Related bug for IR filed as: https://bugs.llvm.org/show_bug.cgi?id=33026
3. The SSE diffs in vector-trunc-math.ll are just scheduling/RA, so nothing real AFAICT.
4. The AVX1 diffs in vector-tzcnt-256.ll are all the same pattern: we reduced the instruction
count by one in each case by eliminating two insert/extract while adding one narrower logic op.
https://bugs.llvm.org/show_bug.cgi?id=32790
Differential Revision: https://reviews.llvm.org/D33137
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@303997 91177308-0d34-0410-b5e6-96231b3b80d8
Currently getOptimalMemOpType returns i32 for large enough sizes without
checking for alignment, leading to poor code generation when misaligned accesses
aren't permitted as we generate a word store then later split it up into byte
stores. This means we inadvertantly go over the MaxStoresPerMemcpy limit and for
memset we splat the memset value into a word then immediately split it up
again.
Fix this by leaving it up to FindOptimalMemOpLowering to figure out which type
to use, but also fix a bug there where it wasn't correctly checking if
misaligned memory accesses are allowed.
Differential Revision: https://reviews.llvm.org/D33442
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@303990 91177308-0d34-0410-b5e6-96231b3b80d8
C++14 added user-defined literal support for complex numbers so that you can
write something like "complex<double> val = 2i". However, there is an existing
GNU extension supporting this syntax and interpreting the result as a _Complex
type.
This changes parsing so that such literals are interpreted in terms of C++14's
operators if an overload is present but otherwise falls back to the original
GNU extension.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@303694 91177308-0d34-0410-b5e6-96231b3b80d8
Refactor the strlen optimization code to work for both strlen and wcslen.
This especially helps with programs in the wild where people pass
L"string"s to const std::wstring& function parameters and the wstring
constructor gets inlined.
This also fixes a lingerind API problem/bug in getConstantStringInfo()
where zeroinitializers would always give you an empty string (without a
length) back regardless of the actual length of the initializer which
did not work well in the TrimAtNul==false causing the PR mentioned
below.
Note that the fixed getConstantStringInfo() needed fixes to SelectionDAG
memcpy lowering and may lead to some cases for out-of-bounds
zeroinitializer accesses not getting optimized anymore. So some code
with UB may produce out of bound memory reads now instead of just
producing zeros.
The refactoring "accidentally" fixes http://llvm.org/PR32124
Differential Revision: https://reviews.llvm.org/D32839
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@303461 91177308-0d34-0410-b5e6-96231b3b80d8
Summary:
While this makes some case better and some case worse - so it's unclear if it is a worthy combine just by itself - this is a useful canonicalisation.
As per discussion in D32756 .
Reviewers: jyknight, nemanjai, mkuper, spatel, RKSimon, zvi, bkramer
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D32916
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@303441 91177308-0d34-0410-b5e6-96231b3b80d8
Summary:
There are several places in the codebase that try to calculate a maximum value in a Statistic object. We currently do this in one of two ways:
MaxNumFoo = std::max(MaxNumFoo, NumFoo);
or
MaxNumFoo = (MaxNumFoo > NumFoo) ? MaxNumFoo : NumFoo;
The first version reads from MaxNumFoo one time and uncontionally rwrites to it. The second version possibly reads it twice depending on the result of the first compare. But we have no way of knowing if the value was changed by another thread between the reads and the writes.
This patch adds a method to the Statistic object that can ensure that we only store if our value is the max and the previous max didn't change after we read it. If it changed we'll recheck if our value should still be the max or not and try again.
This spawned from an audit I'm trying to do of all places we uses the implicit conversion to unsigned on the Statistics objects. See my previous thread on llvm-dev https://groups.google.com/forum/#!topic/llvm-dev/yfvxiorKrDQ
Reviewers: dberlin, chandlerc, hfinkel, dblaikie
Reviewed By: chandlerc
Subscribers: llvm-commits, sanjoy
Differential Revision: https://reviews.llvm.org/D33301
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@303318 91177308-0d34-0410-b5e6-96231b3b80d8
Summary:
In SelectionDAG, when a store is immediately chained to another store
to the same address, elide the first store as it has no observable
effects. This is causes small improvements dealing with intrinsics
lowered to stores.
Test notes:
* Many testcases overwrite store addresses multiple times and needed
minor changes, mainly making stores volatile to prevent the
optimization from optimizing the test away.
* Many X86 test cases optimized out instructions associated with
associated with va_start.
* Note that test_splat in CodeGen/AArch64/misched-stp.ll no longer has
dependencies to check and can probably be removed and potentially
replaced with another test.
Reviewers: rnk, john.brawn
Subscribers: aemerson, rengolin, qcolombet, jyknight, nemanjai, nhaehnle, javed.absar, llvm-commits
Differential Revision: https://reviews.llvm.org/D33206
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@303198 91177308-0d34-0410-b5e6-96231b3b80d8
This function gives the wrong answer on some non-ELF platforms in some
cases. The function that does the right thing lives in Mangler.h. To try to
discourage people from using this function, give it a different name.
Differential Revision: https://reviews.llvm.org/D33162
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@303134 91177308-0d34-0410-b5e6-96231b3b80d8
Summary:
Eli pointed out that it's unsafe to combine the shifts to ISD::SHL etc.,
because those are not defined for b > sizeof(a) * 8, even after some of
the combiners run.
However, PPCISD::SHL defines that behavior (as the instructions themselves).
Move the combination to the backend.
The tests in shift_mask.ll still pass.
Reviewers: echristo, hfinkel, efriedma, iteratee
Subscribers: nemanjai, llvm-commits
Differential Revision: https://reviews.llvm.org/D33076
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@302937 91177308-0d34-0410-b5e6-96231b3b80d8
This patch adds min/max population count, leading/trailing zero/one bit counting methods.
The min methods return answers based on bits that are known without considering unknown bits. The max methods give answers taking into account the largest count that unknown bits could give.
Differential Revision: https://reviews.llvm.org/D32931
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@302925 91177308-0d34-0410-b5e6-96231b3b80d8
Updates the MSP430 target to generate EABI-compatible libcall names.
As a byproduct, adjusts the hardware multiplier options available in
the MSP430 target, adds support for promotion of the ISD::MUL operation
for 8-bit integers, and correctly marks R11 as used by call instructions.
Patch by Andrew Wygle.
Differential Revision: https://reviews.llvm.org/D32676
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@302820 91177308-0d34-0410-b5e6-96231b3b80d8
Summary:
Allow consecutive stores whose values come from consecutive loads to
merged in the presense of other uses of the loads. Previously this was
disallowed as in general the merged load cannot be shared with the
other uses. Merging N stores into 1 may cause as many as N redundant
loads. However in the context of caching this should have neglible
affect on memory pressure and reduce instruction count making it
almost always a win.
Fixes PR32086.
Reviewers: spatel, jyknight, andreadb, hfinkel, efriedma
Reviewed By: efriedma
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D30471
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@302712 91177308-0d34-0410-b5e6-96231b3b80d8
Before r247167, the pass manager builder controlled which AA
implementations were used, exporting them all in the AliasAnalysis
analysis group.
Now, AAResultsWrapperPass always uses BasicAA, but still uses other AA
implementations if made available in the pass pipeline.
But regardless, SDAGISel is required at O0, and really doesn't need to
be doing fancy optimizations based on useful AA results.
Don't require AA at CodeGenOpt::None, and only use it otherwise.
This does have a functional impact (and one testcase is pessimized
because we can't reuse a load). But I think that's desirable no matter
what.
Note that this alone doesn't result in less DT computations: TwoAddress
was previously able to reuse the DT we computed for SDAG. That will be
fixed separately.
Differential Revision: https://reviews.llvm.org/D32766
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@302611 91177308-0d34-0410-b5e6-96231b3b80d8
Using arguments with attribute inalloca creates problems for verification
of machine representation. This attribute instructs the backend that the
argument is prepared in stack prior to CALLSEQ_START..CALLSEQ_END
sequence (see http://llvm.org/docs/InAlloca.htm for details). Frame size
stored in CALLSEQ_START in this case does not count the size of this
argument. However CALLSEQ_END still keeps total frame size, as caller can
be responsible for cleanup of entire frame. So CALLSEQ_START and
CALLSEQ_END keep different frame size and the difference is treated by
MachineVerifier as stack error. Currently there is no way to distinguish
this case from actual errors.
This patch adds additional argument to CALLSEQ_START and its
target-specific counterparts to keep size of stack that is set up prior to
the call frame sequence. This argument allows MachineVerifier to calculate
actual frame size associated with frame setup instruction and correctly
process the case of inalloca arguments.
The changes made by the patch are:
- Frame setup instructions get the second mandatory argument. It
affects all targets that use frame pseudo instructions and touched many
files although the changes are uniform.
- Access to frame properties are implemented using special instructions
rather than calls getOperand(N).getImm(). For X86 and ARM such
replacement was made previously.
- Changes that reflect appearance of additional argument of frame setup
instruction. These involve proper instruction initialization and
methods that access instruction arguments.
- MachineVerifier retrieves frame size using method, which reports sum of
frame parts initialized inside frame instruction pair and outside it.
The patch implements approach proposed by Quentin Colombet in
https://bugs.llvm.org/show_bug.cgi?id=27481#c1.
It fixes 9 tests failed with machine verifier enabled and listed
in PR27481.
Differential Revision: https://reviews.llvm.org/D32394
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@302527 91177308-0d34-0410-b5e6-96231b3b80d8
- This change allows targets to opt-in to using them instead of the log2
shufflevector algorithm.
- The SLP and Loop vectorizers have the common code to do shuffle reductions
factored out into LoopUtils, and now have a unified interface for generating
reductions regardless of the preference of the target. LoopUtils now uses TTI
to determine what kind of reductions the target wants to handle.
- For CodeGen, basic legalization support is added.
Differential Revision: https://reviews.llvm.org/D30086
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@302514 91177308-0d34-0410-b5e6-96231b3b80d8
This reverts commit r302461.
It appears to be causing failures compiling gtest with debug info on the
Linux sanitizer bot. I was unable to reproduce the failure locally,
however.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@302504 91177308-0d34-0410-b5e6-96231b3b80d8
Summary:
For inalloca functions, this is a very common code pattern:
%argpack = type <{ i32, i32, i32 }>
define void @f(%argpack* inalloca %args) {
entry:
%a = getelementptr inbounds %argpack, %argpack* %args, i32 0, i32 0
%b = getelementptr inbounds %argpack, %argpack* %args, i32 0, i32 1
%c = getelementptr inbounds %argpack, %argpack* %args, i32 0, i32 2
tail call void @llvm.dbg.declare(metadata i32* %a, ... "a")
tail call void @llvm.dbg.declare(metadata i32* %c, ... "b")
tail call void @llvm.dbg.declare(metadata i32* %b, ... "c")
Even though these GEPs can be simplified to a constant offset from EBP
or RSP, we don't do that at -O0, and each GEP is computed into a
register. Registers used to compute argument addresses are typically
spilled and clobbered very quickly after the initial computation, so
live debug variable tracking loses information very quickly if we use
DBG_VALUE instructions.
This change moves processing of dbg.declare between argument lowering
and basic block isel, so that we can ask if an argument has a frame
index or not. If the argument lives in a register as is the case for
byval arguments on some targets, then we don't put it in the side table
and during ISel we emit DBG_VALUE instructions.
Reviewers: aprantl
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D32980
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@302483 91177308-0d34-0410-b5e6-96231b3b80d8
