Move the x86 combine from D58974 into the DAGCombine VSELECT code and update the SELECT version to use the isBooleanFlip helper as well.
Requested by @spatel on D59006
llvm-svn: 355533
This patch enables combining integer bitcasts of integer build vectors when the new scalar type is legal. I've avoided floating point because the implementation bitcasts float to int along the way and we would need to check the intermediate types for legality
Differential Revision: https://reviews.llvm.org/D58884
llvm-svn: 355324
Support undef shuffle mask indices in the shuffle(concat_vectors, concat_vectors) -> concat_vectors fold
Differential Revision: https://reviews.llvm.org/D58585
llvm-svn: 354793
This fold can occur during legalization, so it can fight with promotion
to the larger type. It apparently takes a special sequence and subtarget
to avoid more basic simplifications that would hide the problem.
But there's a bigger question raised here: why does distributeTruncateThroughAnd()
even exist? It duplicates functionality from a more minimal pattern that we
already have. But getting rid of this function requires some preliminary steps.
https://bugs.llvm.org/show_bug.cgi?id=40793
llvm-svn: 354594
Summary:
A store to an object whose lifetime is about to end can be removed.
See PR40550 for motivation.
Reviewers: niravd
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D57541
llvm-svn: 354244
If we're comparing some value for equality against 2 constants
and those constants have an absolute difference of just 1 bit,
then we can offset and mask off that 1 bit and reduce to a single
compare against zero:
and/or (setcc X, C0, ne), (setcc X, C1, ne/eq) -->
setcc ((add X, -C1), ~(C0 - C1)), 0, ne/eq
https://rise4fun.com/Alive/XslKj
This transform is disabled by default using a TLI hook
("convertSetCCLogicToBitwiseLogic()").
That should be overridden for AArch64, MIPS, Sparc and possibly
others based on the asm shown in:
https://bugs.llvm.org/show_bug.cgi?id=40611
llvm-svn: 353859
Now that we have SimplifyDemandedBits support for funnel shifts (rL353539), we need to simplify funnel shifts back to bitshifts in cases where either argument has been folded to undef/zero.
Differential Revision: https://reviews.llvm.org/D58009
llvm-svn: 353645
The sqrt case is faster and we already do this for the case where
the exponent is 0.25. This adds the 0.75 case which is also not
sensitive to signed zeros.
Patch by Whitney Tsang (Whitney)
Differential revision: https://reviews.llvm.org/D57434
llvm-svn: 353557
Move the (add (umax X, C), -C) --> (usubsat X, C) X86 combine into generic DAGCombiner
First of a number of saturated arithmetic folds that can be moved out of X86-specific code for PR40111.
Differential Revision: https://reviews.llvm.org/D57754
llvm-svn: 353457
I noticed that we are missing this canonicalization in IR:
rL352515
...and then realized that we don't get this right in SDAG either,
so this has to be fixed first regardless of what we choose to do in IR.
The existing fold was limited to scalars and using the wrong predicate
to guard the transform. We have a boolean contents TLI query that can
be used to decide which direction to fold.
This may eventually lead back to the problems/question in:
https://bugs.llvm.org/show_bug.cgi?id=40486
...but it makes no difference to that yet.
Differential Revision: https://reviews.llvm.org/D57401
llvm-svn: 353433
Summary:
If the index isn't constant, this transform inserts a multiply and an add on the index to calculating the base pointer for a scalar load. But we still create a memory operand with an offset of 0 and the size of the scalar access. But the access is really to an unknown offset within the original access size.
This can cause the machine scheduler to incorrectly calculate dependencies between this load and other accesses. In the case we saw, there was a 32 byte vector store that was split into two 16 byte stores, one with offset 0 and one with offset 16. The size of the memory operand for both was 16. The scheduler correctly detected the alias with the offset 0 store, but not the offset 16 store.
This patch discards the pointer info so we don't incorrectly detect aliasing. I wasn't sure if we could keep using the original offset and size without risking some other transform on the load changing the size.
I tried to reduce a test case, but there's still a lot of memory operations needed to get the scheduler to do the bad reordering. So it looked pretty fragile to maintain.
Reviewers: efriedma
Reviewed By: efriedma
Subscribers: arphaman, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D57616
llvm-svn: 353124
Noticed while investigating PR40483, and fixes the basic test case from the bug - but not a more general case.
We're pretty weak at dealing with ADD/SUB combines compared to the SimplifyAssociativeOrCommutative/SimplifyUsingDistributiveLaws abilities that InstCombine can manage.
llvm-svn: 353044
We already have the getConstantOperandVal helper which returns a uint64_t, but along comes the fuzzer and inserts a i128 -1 constant or something and the whole thing asserts.......
I've updated a few obvious cases, and tried to make use of the const reference where possible, but there's more to do. A number of existing oss-fuzz tickets should be fixed if we start using APInt and perform value clamping where necessary.
llvm-svn: 352961
This patch fixes pr39098.
For the attached test case, CombineZExtLogicopShiftLoad can optimize it to
t25: i64 = Constant<1099511627775>
t35: i64 = Constant<0>
t0: ch = EntryToken
t57: i64,ch = load<(load 4 from `i40* undef`, align 8), zext from i32> t0, undef:i64, undef:i64
t58: i64 = srl t57, Constant:i8<1>
t60: i64 = and t58, Constant:i64<524287>
t29: ch = store<(store 5 into `i40* undef`, align 8), trunc to i40> t57:1, t60, undef:i64, undef:i64
But later visitANDLike transforms it to
t25: i64 = Constant<1099511627775>
t35: i64 = Constant<0>
t0: ch = EntryToken
t57: i64,ch = load<(load 4 from `i40* undef`, align 8), zext from i32> t0, undef:i64, undef:i64
t61: i32 = truncate t57
t63: i32 = srl t61, Constant:i8<1>
t64: i32 = and t63, Constant:i32<524287>
t65: i64 = zero_extend t64
t58: i64 = srl t57, Constant:i8<1>
t60: i64 = and t58, Constant:i64<524287>
t29: ch = store<(store 5 into `i40* undef`, align 8), trunc to i40> t57:1, t60, undef:i64, undef:i64
And it triggers CombineZExtLogicopShiftLoad again, causes a dead loop.
Both forms should generate same instructions, CombineZExtLogicopShiftLoad generated IR looks cleaner. But it looks more difficult to prevent visitANDLike to do the transform, so I prevent CombineZExtLogicopShiftLoad to do the transform if the ZExt is free.
Differential Revision: https://reviews.llvm.org/D57491
llvm-svn: 352792
While dangling nodes will eventually be pruned when they are
considered, leaving them disables combines requiring single-use.
Reviewers: Carrot, spatel, craig.topper, RKSimon, efriedma
Subscribers: hiraditya, llvm-commits
Differential Revision: https://reviews.llvm.org/D57520
llvm-svn: 352784
This extends the existing transform for:
add X, 0/1 --> sub X, 0/-1
...to allow the sibling subtraction fold.
This pattern could regress with the proposed change in D57401.
llvm-svn: 352680
This is the sibling fold for insert-of-insert that was added with D56604.
Now that we have x86 shuffle narrowing (D57156), this change shows improvements for
lots of AVX512 reduction code (not sure that we would ever expect extract-of-extract otherwise).
There's a small regression in some of the partial-permute tests (extracting followed by splat).
That is tracked by PR40500:
https://bugs.llvm.org/show_bug.cgi?id=40500
Differential Revision: https://reviews.llvm.org/D57336
llvm-svn: 352528
The current check in CombineToPreIndexedLoadStore is too
conversative, preventing a pre-indexed store when the base pointer
is a predecessor of the value being stored. Instead, we should check
the pointer operand of the store.
Differential Revision: https://reviews.llvm.org/D56719
llvm-svn: 351933
vecbo (insertsubv undef, X, Z), (insertsubv undef, Y, Z) --> insertsubv VecC, (vecbo X, Y), Z
This is another step in generic vector narrowing. It's also a step towards more horizontal op
formation specifically for x86 (although we still failed to match those in the affected tests).
The scalarization cases are also not optimal (we should be scalarizing those), but it's still
an improvement to use a narrower vector op when we know part of the result must be constant
because both inputs are undef in some vector lanes.
I think a similar match but checking for a constant operand might help some of the cases in
D51553.
Differential Revision: https://reviews.llvm.org/D56875
llvm-svn: 351825
The regression test is reduced from the example shown in D56281.
This does raise a question as noted in the test file: do we want
to handle this pattern? I don't have a motivating example for
that on x86 yet, but it seems like we could have that pattern
there too, so we could avoid the back-and-forth using a shuffle.
llvm-svn: 351753
to reflect the new license.
We understand that people may be surprised that we're moving the header
entirely to discuss the new license. We checked this carefully with the
Foundation's lawyer and we believe this is the correct approach.
Essentially, all code in the project is now made available by the LLVM
project under our new license, so you will see that the license headers
include that license only. Some of our contributors have contributed
code under our old license, and accordingly, we have retained a copy of
our old license notice in the top-level files in each project and
repository.
llvm-svn: 351636
Similar to D55073. Without this change, the DAG combiner crashes on code
with more than 64k of stores in a single basic block that form parallelizable
chains.
No test case, as it would be very IR file.
Reviewed By: RKSimon
Differential Revision: https://reviews.llvm.org/D56740
llvm-svn: 351571
ReduceLoadWidth can trigger using a shifted mask is used and this
requires that the function return a shl node to correct for the
offset. However, the way that this was implemented meant that the
returned result could be an existing node, which would be incorrect.
This fixes the method of inserting the new node and replacing uses.
Differential Revision: https://reviews.llvm.org/D50432
llvm-svn: 351310
The motivating case for this is shown in the first regression test. We are
transferring to scalar and back rather than just zero-extending with 'vpmovzxdq'.
That's a special-case for a more general pattern as shown here. In all tests,
we're avoiding the vector-scalar-vector moves in favor of vector ops.
We aren't producing optimal shuffle code in some cases though, so the patch is
limited to reduce regressions.
Differential Revision: https://reviews.llvm.org/D56281
llvm-svn: 351198
This pattern:
t33: v8i32 = insert_subvector undef:v8i32, t35, Constant:i64<0>
t21: v16i32 = insert_subvector undef:v16i32, t33, Constant:i64<0>
...shows up in PR33758:
https://bugs.llvm.org/show_bug.cgi?id=33758
...although this patch doesn't make any difference to the final result on that yet.
In the affected tests here, it looks like it just makes RA wiggle. But we might
as well squash this to prevent it interfering with other pattern-matching.
Differential Revision:
https://reviews.llvm.org/D56604
llvm-svn: 351008
As noted in PR39973 and D55558:
https://bugs.llvm.org/show_bug.cgi?id=39973
...this is a partial implementation of a fold that we do as an IR canonicalization in instcombine:
// extelt (binop X, Y), Index --> binop (extelt X, Index), (extelt Y, Index)
We want to have this in the DAG too because as we can see in some of the test diffs (reductions),
the pattern may not be visible in IR.
Given that this is already an IR canonicalization, any backend that would prefer a vector op over
a scalar op is expected to already have the reverse transform in DAG lowering (not sure if that's
a realistic expectation though). The transform is limited with a TLI hook because there's an
existing transform in CodeGenPrepare that tries to do the opposite transform.
Differential Revision: https://reviews.llvm.org/D55722
llvm-svn: 350354
Currently we expand the two nodes separately. This gives DAG combiner an opportunity to optimize the expanded sequence taking into account only one set of users. When we expand the other node we'll create the expansion again, but might not be able to optimize it the same way. So the nodes won't CSE and we'll have two similarish sequences in the same basic block. By expanding both nodes at the same time we'll avoid prematurely optimizing the expansion until both the division and remainder have been replaced.
Improves the test case from PR38217. There may be additional opportunities after this.
Differential Revision: https://reviews.llvm.org/D56145
llvm-svn: 350239
If x has multiple sign bits than it doesn't matter which one we extend from so we can sext from x's msb instead.
The X86 setcc-combine.ll changes are a little weird. It appears we ended up with a (sext_inreg (aext (trunc (extractelt)))) after type legalization. The sext_inreg+aext now gets optimized by this combine to leave (sext (trunc (extractelt))). Then we visit the trunc before we visit the sext. This ends up changing the truncate to an extractvectorelt from a bitcasted vector. I have a follow up patch to fix this.
Differential Revision: https://reviews.llvm.org/D56156
llvm-svn: 350235
It's dangerous to knowingly create an illegal vector type
no matter what stage of combining we're in.
This prevents the missed folding/scalarization seen in:
https://bugs.llvm.org/show_bug.cgi?id=40146
llvm-svn: 350034
trunc (add X, C ) --> add (trunc X), C'
If we're throwing away the top bits of an 'add' instruction, do it in the narrow destination type.
This makes the truncate-able opcode list identical to the sibling transform done in IR (in instcombine).
This change used to show regressions for x86, but those are gone after D55494.
This gets us closer to deleting the x86 custom function (combineTruncatedArithmetic)
that does almost the same thing.
Differential Revision: https://reviews.llvm.org/D55866
llvm-svn: 350006
This saves materializing the immediate. The additional forms are less
common (they don't usually show up for bitfield insert/extract), but
they're still relevant.
I had to add a new target hook to prevent DAGCombine from reversing the
transform. That isn't the only possible way to solve the conflict, but
it seems straightforward enough.
Differential Revision: https://reviews.llvm.org/D55630
llvm-svn: 349857
Now that SimplifyDemandedBits/SimplifyDemandedVectorElts is simplifying vector elements, we're seeing more constant BUILD_VECTOR containing undefs.
This patch provides opt-in support for UNDEF elements in matchBinaryPredicate, passing NULL instead of the result ConstantSDNode* argument.
I've updated the (or (and X, c1), c2) -> (and (or X, c2), c1|c2) fold to demonstrate its use, which I believe is safe for undef cases.
Differential Revision: https://reviews.llvm.org/D55822
llvm-svn: 349629
As described on PR40091, we have several places where zext (and zext_vector_inreg) fold an undef input into an undef output. For zero extensions this is incorrect as the output should guarantee to least have the new upper bits set to zero.
SimplifyDemandedVectorElts is the worst offender (and its the most likely to cause new undefs to appear) but DAGCombiner's tryToFoldExtendOfConstant has a similar issue.
Thanks to @dmgreen for catching this.
Differential Revision: https://reviews.llvm.org/D55883
llvm-svn: 349625
The transform performs a bitwise logic op in a wider type followed by
truncate when both inputs are truncated from the same source type:
logic_op (truncate x), (truncate y) --> truncate (logic_op x, y)
There are a bunch of other checks that should prevent doing this when
it might be harmful.
We already do this transform for scalars in this spot. The vector
limitation was shared with a check for the case when the operands are
extended. I'm not sure if that limit is needed either, but that would
be a separate patch.
Differential Revision: https://reviews.llvm.org/D55448
llvm-svn: 349303
Also exposes an issue in DAGCombiner::visitFunnelShift where we were assuming the shift amount had the result type (after legalization it'll have the targets shift amount type).
llvm-svn: 349298
Summary:
If the setcc already has the target desired type we can reach the getSetCC/getSExtOrTrunc after the MatchingVecType check with the exact same types as the nodes we started with. This causes those causes VsetCC to be CSEd to N0 and the getSExtOrTrunc will CSE to N. When we return N, the caller will think that meant we called CombineTo and did our own worklist management. But that's not what happened. This prevents target hooks from being called for the node.
To fix this, I've now returned SDValue if the setcc is already the desired type. But to avoid some regressions in X86 I've had to disable one of the target combines that wasn't being reached before in the case of a (sext (setcc)). If we get vector widening legalization enabled that entire function will be deleted anyway so hopefully this is only for the short term.
Reviewers: RKSimon, spatel
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D55459
llvm-svn: 349137
This isn't quite NFC, but I don't know how to expose
any outward diffs from these changes. Mostly, this
was confusing because it used 'VT' to refer to the
operand type rather the usual type of the input node.
There's also a large block at the end that is dedicated
solely to matching loads, but that wasn't obvious. This
could probably be split up into separate functions to
make it easier to see.
It's still not clear to me when we make certain transforms
because the legality and constant conditions are
intertwined in a way that might be improved.
llvm-svn: 349095
This is a retry of rL349051 (reverted at rL349056). I changed the check for dead-ness from
number of uses to an opcode test for DELETED_NODE based on existing similar code.
Differential Revision: https://reviews.llvm.org/D55655
llvm-svn: 349058
As discussed on D55511, this caused an issue if the inner node deletes a node that the outer node depends upon. As it doesn't affect any lit-tests and I've only been able to expose this with the D55511 change I'm committing this now.
llvm-svn: 348781
This triggers an assert when combining concat_vectors of a bitcast of
merge_values.
With asserts disabled, it fails to select:
fatal error: error in backend: Cannot select: 0x7ff19d000e90: i32 = any_extend 0x7ff19d000ae8
0x7ff19d000ae8: f64,ch = CopyFromReg 0x7ff19d000c20:1, Register:f64 %1
0x7ff19d000b50: f64 = Register %1
In function: d
Differential Revision: https://reviews.llvm.org/D55507
llvm-svn: 348759
This is effectively re-committing the changes from:
rL347917 (D54640)
rL348195 (D55126)
...which were effectively reverted here:
rL348604
...because the code had a bug that could induce infinite looping
or eventual out-of-memory compilation.
The bug was that this code did not guard against transforming
opaque constants. More details are in the post-commit mailing
list thread for r347917. A reduced test for that is included
in the x86 bool-math.ll file. (I wasn't able to reduce a PPC
backend test for this, but it was almost the same pattern.)
Original commit message for r347917:
The motivating case for this is shown in:
https://bugs.llvm.org/show_bug.cgi?id=32023
and the corresponding rot16.ll regression tests.
Because x86 scalar shift amounts are i8 values, we can end up with trunc-binop-trunc
sequences that don't get folded in IR.
As the TODO comments suggest, there will be regressions if we extend this (for x86,
we mostly seem to be missing LEA opportunities, but there are likely vector folds
missing too). I think those should be considered existing bugs because this is the
same transform that we do as an IR canonicalization in instcombine. We just need
more tests to make those visible independent of this patch.
llvm-svn: 348706
As discussed in the post-commit thread of r347917, this
transform is fighting with an existing transform causing
an infinite loop or out-of-memory, so this is effectively
reverting r347917 and its follow-up r348195 while we
investigate the bug.
llvm-svn: 348604
If this is not a valid way to assign an SDLoc, then we get this
wrong all over SDAG.
I don't know enough about the SDAG to explain this. IIUC, theoretically,
debug info is not supposed to affect codegen. But here it has clearly
affected 3 different targets, and the x86 change is an actual improvement.
llvm-svn: 348552
We shouldn't care about the debug location for a node that
we're creating, but attaching the root of the pattern should
be the best effort. (If this is not true, then we are doing
it wrong all over the SDAG).
This is no-functional-change-intended, and there are no
regression test diffs...and that's what I expected. But
there's a similar line above this diff, where those
assumptions apparently do not hold.
llvm-svn: 348550
This was probably organized as it was because bswap is a unary op.
But that's where the similarity to the other opcodes ends. We should
not limit this transform to scalars, and we should not try it if
either input has other uses. This is another step towards trying to
clean this whole function up to prevent it from causing infinite loops
and memory explosions.
Earlier commits in this series:
rL348501
rL348508
rL348518
llvm-svn: 348534
Unlike some of the folds in hoistLogicOpWithSameOpcodeHands()
above this shuffle transform, this has the expected hasOneUse()
checks in place.
llvm-svn: 348523
This patch introduces a new DAGCombiner rule to simplify concat_vectors nodes:
concat_vectors( bitcast (scalar_to_vector %A), UNDEF)
--> bitcast (scalar_to_vector %A)
This patch only partially addresses PR39257. In particular, it is enough to fix
one of the two problematic cases mentioned in PR39257. However, it is not enough
to fix the original test case posted by Craig; that particular case would
probably require a more complicated approach (and knowledge about used bits).
Before this patch, we used to generate the following code for function PR39257
(-mtriple=x86_64 , -mattr=+avx):
vmovsd (%rdi), %xmm0 # xmm0 = mem[0],zero
vxorps %xmm1, %xmm1, %xmm1
vblendps $3, %xmm0, %xmm1, %xmm0 # xmm0 = xmm0[0,1],xmm1[2,3]
vmovaps %ymm0, (%rsi)
vzeroupper
retq
Now we generate this:
vmovsd (%rdi), %xmm0 # xmm0 = mem[0],zero
vmovaps %ymm0, (%rsi)
vzeroupper
retq
As a side note: that VZEROUPPER is completely redundant...
I guess the vzeroupper insertion pass doesn't realize that the definition of
%xmm0 from vmovsd is already zeroing the upper half of %ymm0. Note that on
%-mcpu=btver2, we don't get that vzeroupper because pass vzeroupper insertion
%pass is disabled.
Differential Revision: https://reviews.llvm.org/D55274
llvm-svn: 348522
The PPC test with 2 extra uses seems clearly better by avoiding this transform.
With 1 extra use, we also prevent an extra register move (although that might
be an RA problem). The general rule should be to only make a change here if
it is always profitable. The x86 diffs are all neutral.
llvm-svn: 348518
The AVX512 diffs are neutral, but the bswap test shows a clear overreach in
hoistLogicOpWithSameOpcodeHands(). If we don't check for other uses, we can
increase the instruction count.
This could also fight with transforms trying to go in the opposite direction
and possibly blow up/infinite loop. This might be enough to solve the bug
noted here:
http://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20181203/608593.html
I did not add the hasOneUse() checks to all opcodes because I see a perf
regression for at least one opcode. We may decide that's irrelevant in the
face of potential compiler crashing, but I'll see if I can salvage that first.
llvm-svn: 348508
Because we're potentially peeking through a bitcast in this transform,
we need to use overall bitwidths rather than number of elements to
determine when it's safe to proceed.
Should fix:
https://bugs.llvm.org/show_bug.cgi?id=39893
llvm-svn: 348383
This is an initial patch to add a minimum level of support for funnel shifts to the SelectionDAG and to begin wiring it up to the X86 SHLD/SHRD instructions.
Some partial legalization code has been added to handle the case for 'SlowSHLD' where we want to expand instead and I've added a few DAG combines so we don't get regressions from the existing DAG builder expansion code.
Differential Revision: https://reviews.llvm.org/D54698
llvm-svn: 348353
Add support for ISD::*_EXTEND and ISD::*_EXTEND_VECTOR_INREG opcodes.
The extra broadcast in trunc-subvector.ll will be fixed in an upcoming patch.
llvm-svn: 348246
This is the smallest vector enhancement I could find to D54640.
Here, we're allowing narrowing to only legal vector ops because we'll see
regressions without that. All of the test diffs are wins from what I can tell.
With AVX/AVX512, we can shrink ymm/zmm ops to xmm.
x86 vector multiplies are the problem case that we're avoiding due to the
patchwork ISA, and it's not clear to me if we can dance around those
regressions using TLI hooks or if we need preliminary patches to plug those
holes.
Differential Revision: https://reviews.llvm.org/D55126
llvm-svn: 348195
Summary:
Under -x86-experimental-vector-widening-legalization, fp_to_uint/fp_to_sint with a smaller than 128 bit vector type results are custom type legalized by promoting the result to a 128 bit vector by promoting the elements, inserting an assertzext/assertsext, then truncating back to original type. The truncate will be further legalizdd to a pack shuffle. In the case of a v8i8 result type, we'll end up with a v8i16 fp_to_sint. This will need to be further legalized during vector op legalization by promoting to v8i32 and then truncating again. Under avx2 this produces good code with two pack instructions, but Under avx512 this will result in a truncate instruction and a packuswb instruction. But we should be able to get away with a single truncate instruction.
The other option is to promote all the way to vXi32 result type during the first type legalization. But in some experimentation that seemed to require more work to produce good code for other configurations.
Reviewers: RKSimon, spatel
Reviewed By: RKSimon
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D54836
llvm-svn: 348158
This change prevents the crash noted in the post-commit comments
for rL347478 :
http://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20181119/605166.html
We can't guarantee that an oversized shift amount is folded away,
so we have to check for it.
Note that I committed an incomplete fix for that crash with:
rL347502
But as discussed here:
http://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20181126/605679.html
...we have to try harder.
So I'm not sure how to expose the bug now (and apparently no fuzzers have found
a way yet either).
On the plus side, we have discovered that we're missing real optimizations by
not simplifying nodes sooner, so the earlier fix still has value, and there's
likely more value in extending that so we can simplify more opcodes and simplify
when doing RAUW and/or putting nodes on the combiner worklist.
Differential Revision: https://reviews.llvm.org/D54954
llvm-svn: 348089
The motivating case for this is shown in:
https://bugs.llvm.org/show_bug.cgi?id=32023
and the corresponding rot16.ll regression tests.
Because x86 scalar shift amounts are i8 values, we can end up with trunc-binop-trunc
sequences that don't get folded in IR.
As the TODO comments suggest, there will be regressions if we extend this (for x86,
we mostly seem to be missing LEA opportunities, but there are likely vector folds
missing too). I think those should be considered existing bugs because this is the
same transform that we do as an IR canonicalization in instcombine. We just need
more tests to make those visible independent of this patch.
Differential Revision: https://reviews.llvm.org/D54640
llvm-svn: 347917
This should likely be adjusted to limit this transform
further, but these diffs should be clear wins.
If we have blendv/conditional move, then we should assume
those are cheap ops. The loads become independent of the
compare, so those can be speculated before we need to use
the values in the blend/mov.
llvm-svn: 347526
rL347502 moved the null sibling, so we should group all of these
together. I'm not sure why these aren't methods of the SDValue
class itself, but that's another patch if that's possible.
llvm-svn: 347523
...and use them to avoid creating obviously undef values as
discussed in the post-commit thread for r347478.
The diffs in vector div/rem show that we were missing real
optimizations by creating bogus shift nodes.
llvm-svn: 347502
We fail to canonicalize IR this way (prefer 'not' ops to arbitrary 'xor'),
but that would not matter without this patch because DAGCombiner was
reversing that transform. I think we need this transform in the backend
regardless of what happens in IR to catch cases where the shift-xor
is formed late from GEP or other ops.
https://rise4fun.com/Alive/NC1
Name: shl
Pre: (-1 << C2) == C1
%shl = shl i8 %x, C2
%r = xor i8 %shl, C1
=>
%not = xor i8 %x, -1
%r = shl i8 %not, C2
Name: shr
Pre: (-1 u>> C2) == C1
%sh = lshr i8 %x, C2
%r = xor i8 %sh, C1
=>
%not = xor i8 %x, -1
%r = lshr i8 %not, C2
https://bugs.llvm.org/show_bug.cgi?id=39657
llvm-svn: 347478
This transform needs to be limited.
We are converting to a constant pool load very early, and we
are turning loads that are independent of the select condition
(and therefore speculatable) into a dependent non-speculatable
load.
We may also be transferring a condition code from an FP register
to integer to create that dependent load.
llvm-svn: 347424
This is another step in vector narrowing - a follow-up to D53784
(and hoping to eventually squash potential regressions seen in
D51553).
The x86 test diffs are wins, but the AArch64 diff is probably not.
That problem already exists independent of this patch (see PR39722), but it
went unnoticed in the previous patch because there were no regression tests
that showed the possibility.
The x86 diff in i64-mem-copy.ll is close. Given the frequency throttling
concerns with using wider vector ops, an extra extract to reduce vector
width is the right trade-off at this level of codegen.
Differential Revision: https://reviews.llvm.org/D54392
llvm-svn: 347356
This uncovered an off-by-one typo in SimplifyDemandedVectorElts's INSERT_SUBVECTOR handling as its bounds check was bailing on safe indices.
llvm-svn: 347313
Consistently use (!LegalOperations || isOperationLegalOrCustom) for all node pairs.
Differential Revision: https://reviews.llvm.org/D53478
llvm-svn: 347255
Sadly, this duplicates (twice) the logic from InstSimplify. There
might be some way to at least share the DAG versions of the code,
but copying the folds seems to be the standard method to ensure
that we don't miss these folds.
Unlike in IR, we don't run DAGCombiner to fixpoint, so there's no
way to ensure that we do these kinds of simplifications unless the
code is repeated at node creation time and during combines.
There were other tests that would become worthless with this
improvement that I changed as pre-commits:
rL347161
rL347164
rL347165
rL347166
rL347167
I'm not sure how to salvage the remaining tests (diffs in this patch).
So the x86 tests verify that the new code is working as intended.
The AMDGPU test is actually similar to my motivating case: we have
some undef value that has survived to machine IR in an x86 test, and
then it gets folded in some weird way, or we crash if we don't transfer
the undef flag. But we would have been better off never getting to that
point by doing these simplifications.
This will lead back to PR32023 someday...
https://bugs.llvm.org/show_bug.cgi?id=32023
llvm-svn: 347170
It should be ok to create a new build_vector after legal operations so long as it doesn't cause an infinite loop in DAG combiner.
Unfortunately, X86's custom constant folding in combineVSZext is hiding any test changes from this. But I'm trying to get to a point where that X86 specific code isn't necessary at all.
Differential Revision: https://reviews.llvm.org/D54285
llvm-svn: 346728
Summary:
Handle extra output from index loads in cases where we wish to
forward a load value directly from a preceeding store.
Fixes PR39571.
Reviewers: peter.smith, rengolin
Subscribers: javed.absar, hiraditya, arphaman, llvm-commits
Differential Revision: https://reviews.llvm.org/D54265
llvm-svn: 346654
This is a long-awaited follow-up suggested in D33578. Since then, we've picked up even more
opportunities for vector narrowing from changes like D53784, so there are a lot of test diffs.
Apart from 2-3 strange cases, these are all wins.
I've structured this to be no-functional-change-intended for any target except for x86
because I couldn't tell if AArch64, ARM, and AMDGPU would improve or not. All of those
targets have existing regression tests (4, 4, 10 files respectively) that would be
affected. Also, Hexagon overrides the shouldReduceLoadWidth() hook, but doesn't show
any regression test diffs. The trade-off is deciding if an extra vector load is better
than a single wide load + extract_subvector.
For x86, this is almost always better (on paper at least) because we often can fold
loads into subsequent ops and not increase the official instruction count. There's also
some unknown -- but potentially large -- benefit from using narrower vector ops if wide
ops are implemented with multiple uops and/or frequency throttling is avoided.
Differential Revision: https://reviews.llvm.org/D54073
llvm-svn: 346595
It's possible for vector op legalization to generate a shuffle. If that happens we should give a chance for DAG combine to combine that with a build_vector input.
I also fixed a bug in combineShuffleOfScalars that was considering the number of uses on a undef input to a shuffle. We don't care how many times undef is used.
Differential Revision: https://reviews.llvm.org/D54283
llvm-svn: 346530
The DAGCombiner tries to SimplifySelectCC as follows:
select_cc(x, y, 16, 0, cc) -> shl(zext(set_cc(x, y, cc)), 4)
It can't cope with the situation of reordered operands:
select_cc(x, y, 0, 16, cc)
In that case we just need to swap the operands and invert the Condition Code:
select_cc(x, y, 16, 0, ~cc)
Differential Revision: https://reviews.llvm.org/D53236
llvm-svn: 346484
FindBetterNeighborChains simulateanously improves the chain
dependencies of a chain of related stores avoiding the generation of
extra token factors. For chains longer than the GatherAllAliasDepths,
stores further down in the chain will necessarily fail, a potentially
significant waste and preventing otherwise trivial parallelization.
This patch directly parallelize the chains of stores before improving
each store. This generally improves DAG-level parallelism.
Reviewers: courbet, spatel, RKSimon, bogner, efriedma, craig.topper, rnk
Subscribers: sdardis, javed.absar, hiraditya, jrtc27, atanasyan, llvm-commits
Differential Revision: https://reviews.llvm.org/D53552
llvm-svn: 346432
The original code avoided creating a zero vector after type legalization, but if we're after type legalization the type we have is legal. The real hazard we need to avoid is creating a build vector after op legalization. tryFoldToZero takes care of checking for this.
llvm-svn: 346119
These methods were just wrappers around getNode with additional asserts (identical and repeated 3 times). But getNode already has a switch that can be used to hold these asserts that allows them to be shared for all 3 opcodes. This also enables checking on the places that create these nodes without using the wrappers.
The rest of the patch is just changing all callers to use getNode directly.
llvm-svn: 346087
We already have custom lowering for the AVX case in LegalizeVectorOps. So its better to keep the regular extend op around as long as possible.
I had to qualify one place in DAG combine that created illegal vector extending load operations. This change by itself had no effect on any tests which is why its included here.
I've made a few cleanups to the custom lowering. The sign extend code no longer creates an identity shuffle with undef elements. The zero extend code now emits a zero_extend_vector_inreg instead of an unpckl with a zero vector.
For the high half of the custom lowering of zero_extend/any_extend, we're now using an unpckh with a zero vector or undef. Previously we used used a pshufd to move the upper 64-bits to the lower 64-bits and then used a zero_extend_vector_inreg. I think the zero vector should require less execution resources and be smaller code size.
Differential Revision: https://reviews.llvm.org/D54024
llvm-svn: 346043
reduceBuildVecConvertToConvertBuildVec vectorizes int2float in the DAGCombiner, which means that even if the LV/SLP has decided to keep scalar code using the cost models, this will override this.
While there are cases where vectorization is necessary in the DAG (mainly due to legalization artefacts), I don't think this is the case here, we should assume that the vectorizers know what they are doing.
Differential Revision: https://reviews.llvm.org/D53712
llvm-svn: 345964
I'm having trouble creating a test case for the ISD::TRUNCATE part of this that shows any codegen differences. But I was able to test the setcc path which is what the test changes here cover.
llvm-svn: 345908
The test causes a crash because we were trying to extract v4f32 to v3f32, and the
narrowing factor was then 4/3 = 1 producing a bogus narrow type.
This should fix:
https://bugs.llvm.org/show_bug.cgi?id=39511
llvm-svn: 345842
Summary:
Normalize the offset for endianess before checking
if the store cover the load in ForwardStoreValueToDirectLoad.
Without this we missed out on some optimizations for big
endian targets. If for example having a 4 bytes store followed
by a 1 byte load, loading the least significant byte from the
store, the STCoversLD check would fail (see @test4 in
test/CodeGen/AArch64/load-store-forwarding.ll).
This patch also fixes a problem seen in an out-of-tree target.
The target has i40 as a legal type, it is big endian,
and the StoreSize for i40 is 48 bits. So when normalizing
the offset for endianess we need to take the StoreSize into
account (assuming that padding added when storing into
a larger StoreSize always is added at the most significant
end).
Reviewers: niravd
Reviewed By: niravd
Subscribers: javed.absar, kristof.beyls, llvm-commits, uabelho
Differential Revision: https://reviews.llvm.org/D53776
llvm-svn: 345636
Narrowing vector binops came up in the demanded bits discussion in D52912.
I don't think we're going to be able to do this transform in IR as a canonicalization
because of the risk of creating unsupported widths for vector ops, but we already have
a DAG TLI hook to allow what I was hoping for: isExtractSubvectorCheap(). This is
currently enabled for x86, ARM, and AArch64 (although only x86 has existing regression
test diffs).
This is artificially limited to not look through bitcasts because there are so many
test diffs already, but that's marked with a TODO and is a small follow-up.
Differential Revision: https://reviews.llvm.org/D53784
llvm-svn: 345602
Summary:
Changes all uses of minnan/maxnan to minimum/maximum
globally. These names emphasize that the semantic difference between
these operations is more than just NaN-propagation.
Reviewers: arsenm, aheejin, dschuff, javed.absar
Subscribers: jholewinski, sdardis, wdng, sbc100, jgravelle-google, jrtc27, atanasyan, llvm-commits
Differential Revision: https://reviews.llvm.org/D53112
llvm-svn: 345218
Until now, we've only checked whether merging stores would cause a cycle via
the value argument, but the address and indexed offset arguments are also
capable of creating cycles in some situations.
The addresses are all base+offset with notionally the same base, but the base
SDNode may still be different (e.g. via an indexed load in one case, and an
ISD::ADD elsewhere). This allows cycles to creep in if one of these sources
depends on another.
The indexed offset is usually undef (representing a non-indexed store), but on
some architectures (e.g. 32-bit ARM-mode ARM) it can be an arbitrary value,
again allowing dependency cycles to creep in.
llvm-svn: 345200
When implementing memset's today we often see this pattern:
$x0 = MOV 0xXYXYXYXYXYXYXYXY
store $x0, ...
$w1 = MOV 0xXYXYXYXY
store $w1, ...
We first create a 64bit constant in a 64bit register with all bytes the
same and then create a 32bit constant with all bytes the same in a 32bit
register. In many targets we could just access the lower byte of the
64bit register instead.
- Ideally this would be handled by the ConstantHoist pass but it runs
too early when memset isn't expanded yet.
- The memset expansion code already had this optimization implemented,
however SelectionDAG constantfolding would constantfold the
"trunc(bigconstnat)" pattern to "smallconstant".
- This patch makes the memset expansion mark the constant as Opaque and
stop DAGCombiner from constant folding in this situation. (Similar to
how ConstantHoisting marks things as Opaque to avoid folding
ADD/SUB/etc.)
Differential Revision: https://reviews.llvm.org/D53181
llvm-svn: 345102
I've included a fix to DAGCombiner::ForwardStoreValueToDirectLoad that I believe will prevent the previous miscompile.
Original commit message:
Theoretically this was done to simplify the amount of isel patterns that were needed. But it also meant a substantial number of our isel patterns have to match an explicit bitcast. By making the vXi32/vXi16/vXi8 types legal for loads, DAG combiner should be able to change the load type to rem
I had to add some additional plain load instruction patterns and a few other special cases, but overall the isel table has reduced in size by ~12000 bytes. So it looks like this promotion was hurting us more than helping.
I still have one crash in vector-trunc.ll that I'm hoping @RKSimon can help with. It seems to relate to using getTargetConstantFromNode on a load that was shrunk due to an extract_subvector combine after the constant pool entry was created. So we end up decoding more mask elements than the lo
I'm hoping this patch will simplify the number of patterns needed to remove the and/or/xor promotion.
Reviewers: RKSimon, spatel
Reviewed By: RKSimon
Subscribers: llvm-commits, RKSimon
Differential Revision: https://reviews.llvm.org/D53306
llvm-svn: 344965
Introduce new versions that follow the IEEE semantics
to help with legalization that may need quieted inputs.
There are some regressions from inserting unnecessary
canonicalizes when these are matched from fast math
fcmp + select which should be fixed in a future commit.
llvm-svn: 344914
This is a late backend subset of the IR transform added with:
D52439
We can confirm that the conversion to a 'trunc' is correct by running:
$ opt -instcombine -data-layout="e"
(assuming the IR transforms are correct; change "e" to "E" for big-endian)
As discussed in PR39016:
https://bugs.llvm.org/show_bug.cgi?id=39016
...the pattern may emerge during legalization, so that's we are waiting for an
insertelement to become a scalar_to_vector in the pattern matching here.
The DAG allows for fun variations that are not possible in IR. Result types for
extracts and scalar_to_vector don't necessarily match input types, so that means
we have to be a bit more careful in the transform (see code comments).
The tests show that we don't handle cases that require a shift (as we did in the
IR version). I've left that as a potential follow-up because I'm not sure if
that's a real concern at this late stage.
Differential Revision: https://reviews.llvm.org/D53201
llvm-svn: 344872
I want to add another pattern here that includes scalar_to_vector,
so this makes that patch smaller. I was hoping to remove the
hasOneUse() check because it shouldn't be necessary for common
codegen, but an AMDGPU test has a comment suggesting that the
extra check makes things better on one of those targets.
llvm-svn: 344320
Summary:
Extend analysis forwarding loads from preceeding stores to work with
extended loads and truncated stores to the same address so long as the
load is fully subsumed by the store.
Hexagon's swp-epilog-phis.ll and swp-memrefs-epilog1.ll test are
deleted as they've no longer seem to be relevant.
Reviewers: RKSimon, rnk, kparzysz, javed.absar
Subscribers: sdardis, nemanjai, hiraditya, atanasyan, llvm-commits
Differential Revision: https://reviews.llvm.org/D49200
llvm-svn: 344142
We already do the following combines:
(bitcast int (and (bitcast fp X to int), 0x7fff...) to fp) -> fabs X
(bitcast int (xor (bitcast fp X to int), 0x8000...) to fp) -> fneg X
When the target has "bit preserving fp logic". This patch just extends it
to also combine:
(bitcast int (or (bitcast fp X to int), 0x8000...) to fp) -> fneg (fabs X)
As some targets have fnabs and even those that don't can efficiently lower
both the fabs and the fneg.
Differential revision: https://reviews.llvm.org/D44548
llvm-svn: 344093
This change is proposed as a part of D44548, but we
need this independently to avoid regressions from improved
undef propagation in SimplifyDemandedVectorElts().
llvm-svn: 343940
And use that to transform fsub with zero constant operands.
The integer part isn't used yet, but it is proposed for use in
D44548, so adding both enhancements here makes that
patch simpler.
llvm-svn: 343865
This fixes a case of bad index calculation when merging mismatching
vector types. This changes the existing code to just use the existing
extract_{subvector|element} and a bitcast (instead of bitcast first and
then newly created extract_xxx) so we don't need to adjust any indices
in the first place.
rdar://44584718
Differential Revision: https://reviews.llvm.org/D52681
llvm-svn: 343493
The SINT_TO_FP<->UINT_TO_FP combines for non-negative integers should only occur for legal ops once LegalOperations = true
No test case to hand, noticed when investigating PR38226 + PR38970
llvm-svn: 343405
DAGCombine will try to fold two loads that feed a SELECT or SELECT_CC
after the select, resulting in a select of an address and a single
load after.
If either of the loads depend on the other, this is not legal as it
could introduce cycles. However, it only checked this if the opcode
was a SELECT, and not for a SELECT_CC.
Unfortunately, the only reproducer I have for this is for our
downstream target. I've tried getting it to trigger on an upstream one
but haven't been successful.
Patch thanks to Bevin Hansson.
llvm-svn: 342980
This is a preliminary step towards solving PR14613:
https://bugs.llvm.org/show_bug.cgi?id=14613
If we have an 'add' instruction that sets flags, we can use that to eliminate an
explicit compare instruction or some other instruction (cmn) that sets flags for
use in the later select.
As shown in the unchanged tests that use 'icmp ugt %x, %a', we're effectively
reversing an IR icmp canonicalization that replaces a variable operand with a
constant:
https://rise4fun.com/Alive/V1Q
But we're not using 'uaddo' in those cases via DAG transforms. This happens in
CGP after D8889 without checking target lowering to see if the op is supported.
So AArch already shows 'uaddo' codegen for the i8/i16/i32/i64 test variants with
"using_cmp_sum" in the title. That's the pattern that CGP matches as an unsigned
saturated add and converts to uaddo without checking target capabilities.
This patch is gated by isOperationLegalOrCustom(ISD::UADDO, VT), so we see only
see AArch diffs for i32/i64 in the tests with "using_cmp_notval" in the title
(unlike x86 which sees improvements for all sizes because all sizes are 'custom').
But the AArch code (like x86) looks better when translated to 'uaddo' in all cases.
So someone that is involved with AArch may want to set i8/i16 to 'custom' for UADDO,
so this patch will fire on those tests.
Another possibility given the existing behavior: we could remove the legal-or-custom
check altogether because we're assuming that a UADDO sequence is canonical/optimal
before we ever reach here. But that seems like a bug to me. If the target doesn't
have an add-with-flags op, then it's not likely that we'll get optimal DAG combining
using a UADDO node. This is similar justification for why we don't canonicalize IR to
the overflow math intrinsic sibling (llvm.uadd.with.overflow) for UADDO in the first
place.
Differential Revision: https://reviews.llvm.org/D51929
llvm-svn: 342886
This code handled SCALAR_TO_VECTOR being returned by the recursion, but the code that used to return SCALAR_TO_VECTOR was removed in 2015.
llvm-svn: 342856
This comment was misleading about why we were restricting to before legalize types. The reason given would only apply to before legalize ops. But there is a before legalize types reason that should also be listed.
llvm-svn: 342851
This is an alternative to https://reviews.llvm.org/D37896. We can't decompose
multiplies generically without a target hook to tell us when it's profitable.
ARM and AArch64 may be able to remove some existing code that overlaps with
this transform.
This extends D52195 and may resolve PR34474:
https://bugs.llvm.org/show_bug.cgi?id=34474
(still an open question about transforming legal vector multiplies, but we
could open another bug report for those)
llvm-svn: 342844
x86 had 2 versions of peekThroughBitcast. DAGCombiner had 1. Plus, it had a 1-off implementation for the one-use variant.
Move the x86 versions of the code to SelectionDAG, so we don't have different copies of the code.
No functional change intended.
I'm putting this next to isBitwiseNot() because I am planning to use it in there. Another option is next to the
helpers in the ISD namespace (eg, ISD::isConstantSplatVector()). But if there's no good reason for those to be
there, I'd prefer to pull other helpers over to SelectionDAG in follow-up steps.
Differential Revision: https://reviews.llvm.org/D52285
llvm-svn: 342669
The test diff in not-and-simplify.ll is from a use in SimplifyDemandedBits,
and the test diff in add.ll is from a DAGCombiner transform.
llvm-svn: 342594
This is an alternative to D37896. I don't see a way to decompose multiplies
generically without a target hook to tell us when it's profitable.
ARM and AArch64 may be able to remove some duplicate code that overlaps with
this transform.
As a first step, we're only getting the most clear wins on the vector examples
requested in PR34474:
https://bugs.llvm.org/show_bug.cgi?id=34474
As noted in the code comment, it's likely that the x86 constraints are tighter
than necessary, but it may not always be a win to replace a pmullw/pmulld.
Differential Revision: https://reviews.llvm.org/D52195
llvm-svn: 342554
This is a follow-up suggested in D51630 and originally proposed as an IR transform in D49040.
Copying the motivational statement by @evandro from that patch:
"This transformation helps some benchmarks in SPEC CPU2000 and CPU2006, such as 188.ammp,
447.dealII, 453.povray, and especially 300.twolf, as well as some proprietary benchmarks.
Otherwise, no regressions on x86-64 or A64."
I'm proposing to add only the minimum support for a DAG node here. Since we don't have an
LLVM IR intrinsic for cbrt, and there are no other DAG ways to create a FCBRT node yet, I
don't think we need to worry about DAG builder, legalization, a strict variant, etc. We
should be able to expand as needed when adding more functionality/transforms. For reference,
these are transform suggestions currently listed in SimplifyLibCalls.cpp:
// * cbrt(expN(X)) -> expN(x/3)
// * cbrt(sqrt(x)) -> pow(x,1/6)
// * cbrt(cbrt(x)) -> pow(x,1/9)
Also, given that we bail out on long double for now, there should not be any logical
differences between platforms (unless there's some platform out there that has pow()
but not cbrt()).
Differential Revision: https://reviews.llvm.org/D51753
llvm-svn: 342348
Add support for bitcasts from float type to an integer type of the same element bitwidth.
There maybe cases where we need to support different widths (e.g. as SSE __m128i is treated as v2i64) - but I haven't seen cases of this in the wild yet.
llvm-svn: 341652
This was proposed as an IR transform in D49306, but it was not clearly justifiable as a canonicalization.
Here, we only do the transform when the target tells us that sqrt can be lowered with inline code.
This is the basic case. Some potential enhancements are in the TODO comments:
1. Generalize the transform for other exponents (allow more than 2 sqrt calcs if that's really cheaper).
2. If we have less fast-math-flags, generate code to avoid -0.0 and/or INF.
3. Allow the transform when optimizing/minimizing size (might require a target hook to get that right).
Note that by default, x86 converts single-precision sqrt calcs into sqrt reciprocal estimate with
refinement. That codegen is controlled by CPU attributes and can be manually overridden. We have plenty
of test coverage for that already, so I didn't bother to include extra testing for that here. AArch uses
its full-precision ops in all cases (not sure if that's the intended behavior or not, but that should
also be covered by existing tests).
Differential Revision: https://reviews.llvm.org/D51630
llvm-svn: 341481
Summary:
I'm not sure if this patch is correct or if it needs more qualifying somehow. Bitcast shouldn't change the size of the load so it should be ok? We already do something similar for stores. We'll change the type of a volatile store if the resulting store is Legal or Custom. I'm not sure we should be allowing Custom there...
I was playing around with converting X86 atomic loads/stores(except seq_cst) into regular volatile loads and stores during lowering. This would allow some special RMW isel patterns in X86InstrCompiler.td to be removed. But there's some floating point patterns in there that didn't work because we don't fold (f64 (bitconvert (i64 volatile load))) or (f32 (bitconvert (i32 volatile load))).
Reviewers: efriedma, atanasyan, arsenm
Reviewed By: efriedma
Subscribers: jvesely, arsenm, sdardis, kzhuravl, wdng, yaxunl, dstuttard, tpr, t-tye, arichardson, jrtc27, atanasyan, jfb, llvm-commits
Differential Revision: https://reviews.llvm.org/D50491
llvm-svn: 340797
I noticed this along with the patterns in D51125, but when the index is variable,
we don't convert insertelement into a build_vector.
For x86, that means these get expanded at legalization time into the loading/spilling
code that we see in the tests. I think it's always better to avoid going to memory on
these, and we get the optimal 'broadcast' if it's available.
I suspect other targets may want to look at enabling the hook. AArch64 and AMDGPU have
regression tests that would be affected (although I did not check what would happen in
those cases). In the most basic cases shown here, AArch64 would probably do much
better with a splat.
Differential Revision: https://reviews.llvm.org/D51186
llvm-svn: 340705
Previously we allowed the store to be Custom. But without knowing for sure that the Custom handling won't split the store, we shouldn't convert a volatile store. We also probably shouldn't be creating a store the requires custom handling after LegalizeOps. This could lead to an infinite loop if the custom handling was to insert a bitcast. Though I guess isStoreBitCastBeneficial could be used to block such a loop.
The test changes here are due to the volatile part of this. The stores in the test are all volatile and i32 stores are marked custom, So we are no longer converting them
This is related to D50491 where I was trying to allow some bitcasting of volatile loads
Differential Revision: https://reviews.llvm.org/D50578
llvm-svn: 340626
During combining, ReduceLoadWdith is used to combine AND nodes that
mask loads into narrow loads. This patch allows the mask to be a
shifted constant. This results in a narrow load which is then left
shifted to compensate for the new offset.
Differential Revision: https://reviews.llvm.org/D50432
llvm-svn: 340261
Summary:
I believe this restores the behavior we had before r339147.
Fixes PR38622.
Reviewers: RKSimon, chandlerc, spatel
Reviewed By: chandlerc
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D50936
llvm-svn: 340120
Add support for cases where only some c1+c2 results exceed the max bitshift, clamping accordingly.
Differential Revision: https://reviews.llvm.org/D35722
llvm-svn: 340010
When nodes are reassociated the vector-reduction flag gets lost.
The test case is here is what would happen if you had a sum of absolute differences loop that started with a non-zero but contant sum and that loop was unrolled. The vectorizer will generate a constant vector for the initial value. And DAGCombiner reassociate tries to move it down the addition tree erasing the vector-reduction flag. Interestingly this moves constants the opposite direction of the reassociate IR pass.
I've chosen to just punt on the reassociate, but I suppose we could maybe preserve the flag if both nodes have it set.
Differential Revision: https://reviews.llvm.org/D50827
llvm-svn: 339946
Intentionally excluding nodes from the DAGCombine worklist is likely to
lead to weird optimizations and infinite loops, so it's generally a bad
idea.
To avoid the infinite loops, fix DAGCombine to use the
isDesirableToCommuteWithShift target hook before performing the
transforms in question, and implement the target hook in the ARM backend
disable the transforms in question.
Fixes https://bugs.llvm.org/show_bug.cgi?id=38530 . (I don't have a
reduced testcase for that bug. But we should have sufficient test
coverage for PerformSHLSimplify given that we're not playing weird
tricks with the worklist. I can try to bugpoint it if necessary,
though.)
Differential Revision: https://reviews.llvm.org/D50667
llvm-svn: 339734
Summary: This change provides a common optimization path for both Unsafe and FMF driven optimization for this fsub fold adding reassociation, as it the flag that most closely represents the translation
Reviewers: spatel, wristow, arsenm
Reviewed By: spatel
Subscribers: wdng
Differential Revision: https://reviews.llvm.org/D50195
llvm-svn: 339357
isNegatibleForFree() should not matter here (as the test diffs show)
because it's always a win to replace an fsub+fadd with fneg. The
problem in D50195 persists because either (1) we are doing these
folds in the wrong order or (2) we're missing another fold for fadd.
llvm-svn: 339299
I don't know if it's possible to expose this diff in a test,
but we should always try simplifications (no new nodes created)
before more complicated transforms for efficiency (similar to
what we do in IR).
llvm-svn: 339298
The isConstOrConstSplat result is only used in a ISD::matchUnaryPredicate call which can perform the equivalent iteration just as quickly.
llvm-svn: 339262
Scatter could have multiple identical indices. We need to maintain sequential order. We get this right in LegalizeVectorTypes, but not in this code.
Differential Revision: https://reviews.llvm.org/D50374
llvm-svn: 339157
This was missed in D50185.
NFC until we add actual non-uniform support to BuildSDIV (similar BuildUDIV support in D49248) - for now it just early outs.
llvm-svn: 339147
This patch refactors the existing TargetLowering::BuildUDIV base implementation to support non-uniform constant vector denominators.
It also includes a fold for MULHU by pow2 constants to SRL which can now more readily occur from BuildUDIV.
Differential Revision: https://reviews.llvm.org/D49248
llvm-svn: 339121
Src0 doesn't really convey any meaning to what the operand is. Passthru matches what's used in the documentation for the intrinsic this comes from.
llvm-svn: 339101
First step towards a BuildSDIV equivalent to D49248 for non-uniform vector support - this just pushes the splat detection down into TargetLowering::BuildSDIV where its still used.
Differential Revision: https://reviews.llvm.org/D50185
llvm-svn: 338838
The vector contains the SDNodes that these functions create. The number of nodes is always a small number so we should use SmallVector to avoid a heap allocation.
llvm-svn: 338329
This is exchanging a sub-of-1 with add-of-minus-1:
https://rise4fun.com/Alive/plKAH
This is another step towards improving select-of-constants codegen (see D48970).
x86 is the motivating target, and those diffs all appear to be wins. PPC and AArch64 look neutral.
I've limited this to early combining (!LegalOperations) in case a target wants to reverse it, but
I think canonicalizing to 'add' is more likely to produce further transforms because we have more
folds for 'add'.
Differential Revision: https://reviews.llvm.org/D49924
llvm-svn: 338317
Thinking about it more it might be possible for the later nodes to be folded in getNode in such a way that the other created nodes are left dead. This can cause use counts to be incorrect on nodes that aren't dead.
So its probably safer to leave this alone.
llvm-svn: 338298
Summary:
Attempt to extract a shrl from a udiv or a shl from a mul if this allows a rotate to be formed. This targets cases where the input to a rotate pattern was a mul or udiv by a constant and InstCombine merged one of the shifts with the op.
Patch by: sameconrad (Sam Conrad)
Reviewers: RKSimon, craig.topper, spatel, lebedev.ri, javed.absar
Reviewed By: lebedev.ri
Subscribers: efriedma, kparzysz, llvm-commits
Differential Revision: https://reviews.llvm.org/D47681
llvm-svn: 338270
The DAGCombiner has a mechanism for ensuring all nodes have been visited at least once. Every time a node is visited, it makes sure its operands have been in the worklist at least once. This ensures that when multiple nodes are created by a combine, only the last node needs to be returned. The earlier nodes can all be found Through this operand check. These means we don't need to explicitly add nodes to the worklist when a combine creates multiple nodes.
I've removed the most obvious cases here. There are probably more than can be removed.
llvm-svn: 338222
This removes the need for an assert to ensure the pointer isn't null.
Years ago we had ifs the checked the pointer was non-null before very access to the vector. These checks were removed and replaced with a single assert. But a reference seems more suitable here.
llvm-svn: 338205
This is a follow-up suggested in D48970.
Alive proofs:
https://rise4fun.com/Alive/sII
We can eliminate an instruction in the usual select-of-constants
to bit hack transform by adjusting the add/sub with constant.
This is always a win.
There are more transforms that are likely wins, but they may need
target hooks in case some targets do not benefit.
This is another step towards making up for canonicalizing to
select-of-constants in rL331486.
llvm-svn: 338132
The DAGCombiner has a system for ensuring all nodes are visited. It doesn't require an AddToWorkList for every node that is created by a combine.
llvm-svn: 338079
When merging through a TokenFactor we need to check that the
load may be ordered such that no other aliasing memory operations may
happen. It is not sufficient to just check that the load is a member
of the chain token factor as it there may be a indirect chain. Require
the load's chain has only one use.
This fixes PR37826.
Reviewers: spatel, davide, efriedma, craig.topper, RKSimon
Subscribers: hiraditya, llvm-commits
Differential Revision: https://reviews.llvm.org/D49388
llvm-svn: 337560
If we are only extracting vector elements via EXTRACT_VECTOR_ELT(s) we may be able to use SimplifyDemandedVectorElts to avoid unnecessary vector ops.
Differential Revision: https://reviews.llvm.org/D49262
llvm-svn: 337258
This is almost the same as an existing IR canonicalization in instcombine,
so I'm assuming this is a good early generic DAG combine too.
The motivation comes from reduced bit-hacking for select-of-constants in IR
after rL331486. We want to restore that functionality in the DAG as noted in
the commit comments for that change and the llvm-dev discussion here:
http://lists.llvm.org/pipermail/llvm-dev/2018-July/124433.html
The PPC and AArch tests show that those targets are already doing something
similar. x86 will be neutral in the minimal case and generally better when
this pattern is extended with other ops as shown in the signbit-shift.ll tests.
Note the asymmetry: we don't include the (extend (ifneg X)) transform because
it already exists in SimplifySelectCC(), and that is verified in the later
unchanged tests in the signbit-shift.ll files. Without the 'not' op, the
general transform to use a shift is always a win because that's a single
instruction.
Alive proofs:
https://rise4fun.com/Alive/ysli
Name: if pos, get -1
%c = icmp sgt i16 %x, -1
%r = sext i1 %c to i16
=>
%n = xor i16 %x, -1
%r = ashr i16 %n, 15
Name: if pos, get 1
%c = icmp sgt i16 %x, -1
%r = zext i1 %c to i16
=>
%n = xor i16 %x, -1
%r = lshr i16 %n, 15
Differential Revision: https://reviews.llvm.org/D48970
llvm-svn: 337130
Reuse this function as to test correctness and profitability of
reducing width of either load or store operations.
Reviewsers: samparker
Differential Revision: https://reviews.llvm.org/D48624
llvm-svn: 336800
This allows us to use SelectionDAG::isKnownNeverZero in DAGCombiner::visitREM (visitSDIVLike/visitUDIVLike handle the checking for constants).
llvm-svn: 336779
First stage in PR38057 - support non-uniform constant vectors in the combine to reuse the division-by-constant logic.
We can definitely do better for srem pow2 remainders (and avoid that extra multiply....) but this at least helps keep everything on the vector unit.
Differential Revision: https://reviews.llvm.org/D48975
llvm-svn: 336774
As suggested by @efriedma on D48975, this patch separates the BuildDiv/Pow2 style optimizations from the rest of the visitSDIV/visitUDIV to make it easier to reuse the combines and will allow us to avoid some rather nasty node recursive combining in visitREM.
llvm-svn: 336656
Summary:
This adds a reverse transform for the instcombine canonicalizations
that were added in D47980, D47981.
As discussed later, that was worse at least for the code size,
and potentially for the performance, too.
https://rise4fun.com/Alive/Zmpl
Reviewers: craig.topper, RKSimon, spatel
Reviewed By: spatel
Subscribers: reames, llvm-commits
Differential Revision: https://reviews.llvm.org/D48768
llvm-svn: 336585
As discussed on PR37989, this patch adds EXTRACT_SUBVECTOR handling to TargetLowering::SimplifyDemandedVectorElts and calls it from DAGCombiner::visitEXTRACT_SUBVECTOR.
Differential Revision: https://reviews.llvm.org/D48825
llvm-svn: 336490
D48278
Allow to reduce redundant shift masks.
For example:
x1 = x & 0xAB00
x2 = (x >> 8) & 0xAB
can be reduced to:
x1 = x & 0xAB00
x2 = x1 >> 8
It only allows folding when the masks and shift values are constants.
llvm-svn: 336426
The combine added in commit 329525 overlooked the case where one, but not all, of the divisor elements is -1, -1 is the only power of two value for which the sdiv expansion recipe breaks.
Thanks to @zvi for the original patch.
Differential Revision: https://reviews.llvm.org/D45806
llvm-svn: 336048
We could get away with it for constant folded cases, but not for rL335719.
Thanks to Krzysztof Parzyszek for noticing.
Reapply original commit rL335821 which was reverted at rL335871 due to a WebAssembly bug that was fixed at rL335884.
llvm-svn: 335886
As noted in the D44909 review, the transform from (fptosi+sitofp) to ftrunc
can produce -0.0 where the original code does not:
#include <stdio.h>
int main(int argc) {
float x;
x = -0.8 * argc;
printf("%f\n", (float)((int)x));
return 0;
}
$ clang -O0 -mavx fp.c ; ./a.out
0.000000
$ clang -O1 -mavx fp.c ; ./a.out
-0.000000
Ideally, we'd use IR/node flags to predicate the transform, but the IR parser
doesn't currently allow fast-math-flags on the cast instructions. So for now,
just use the function attribute that corresponds to clang's "-fno-signed-zeros"
option.
Differential Revision: https://reviews.llvm.org/D48085
llvm-svn: 335761
This patch has the same motivating example as D48466:
define void @foo(i64 %x, i32 %c.0282.in, i32 %d.0280, i32* %ptr0, i32* %ptr1) {
%c.0282 = and i32 %c.0282.in, 268435455
%a16 = lshr i64 32508, %x
%a17 = and i64 %a16, 1
%tobool = icmp eq i64 %a17, 0
%. = select i1 %tobool, i32 1, i32 2
%.286 = select i1 %tobool, i32 27, i32 26
%shr97 = lshr i32 %c.0282, %.
%shl98 = shl i32 %c.0282.in, %.286
%or99 = or i32 %shr97, %shl98
%shr100 = lshr i32 %d.0280, %.
%shl101 = shl i32 %d.0280, %.286
%or102 = or i32 %shr100, %shl101
store i32 %or99, i32* %ptr0
store i32 %or102, i32* %ptr1
ret void
}
...but I'm trying to kill the setcc bool math sooner rather than later.
By matching a larger pattern that includes both the low-bit mask and the trailing add/sub,
we can create a universally good fold because we always eliminate the condition code
intermediate value.
Here are Alive proofs for these (currently instcombine folds the 'add' variants, but
misses the 'sub' patterns):
https://rise4fun.com/Alive/Gsyp
Name: sub of zext cmp mask
%a = and i8 %x, 1
%c = icmp eq i8 %a, 0
%z = zext i1 %c to i32
%r = sub i32 C1, %z
=>
%optional_cast = zext i8 %a to i32
%r = add i32 %optional_cast, C1-1
Name: add of zext cmp mask
%a = and i32 %x, 1
%c = icmp eq i32 %a, 0
%z = zext i1 %c to i8
%r = add i8 %z, C1
=>
%optional_cast = trunc i32 %a to i8
%r = sub i8 C1+1, %optional_cast
All of the tests look like improvements or neutral to me. But it is possible that x86
test+set+bitop is better than what we now show here. I suspect we could do better by
adding another fold for the 'sub' variants.
We start with select-of-constant in IR in the larger motivating test, so that's why I
included tests with selects. Proofs for those variants:
https://rise4fun.com/Alive/Bx1
Name: true const is bigger
Pre: C2 == (C1 + 1)
%a = and i8 %x, 1
%c = icmp eq i8 %a, 0
%r = select i1 %c, i64 C2, i64 C1
=>
%z = zext i8 %a to i64
%r = sub i64 C2, %z
Name: false const is bigger
Pre: C2 == (C1 + 1)
%a = and i8 %x, 1
%c = icmp eq i8 %a, 0
%r = select i1 %c, i64 C1, i64 C2
=>
%z = zext i8 %a to i64
%r = add i64 C1, %z
Differential Revision: https://reviews.llvm.org/D48466
llvm-svn: 335433
Allowed folding for "and/or" binops with non-constant operand if
arguments of select are 0/-1 values.
Normally this code with "and" opcode does not get to a DAG combiner
and simplified yet in the InstCombine. However AMDGPU produces it
during lowering and InstCombine has no chance to optimize it out.
In turn the same pattern with "or" opcode can reach DAG.
Differential Revision: https://reviews.llvm.org/D48301
llvm-svn: 335250
The alignment parameter to getExtLoad is treated as a base alignment,
not the alignment of the load (base + offset). When we infer a better
alignment for a Ptr we need to ensure that it applies to the base to
prevent the alignment on the load from being wrong.
This fixes a bug where the alignment could then be used to incorrectly
prove noalias between a load and a store, leading to a miscompile.
Differential Revision: https://reviews.llvm.org/D48029
llvm-svn: 335210
Previously this folding was done only if select is a first operand.
However, for non-commutative operations constant may go before
select.
Differential Revision: https://reviews.llvm.org/D48223
llvm-svn: 335167
Summary: This patch originated from D46562 and is a proper subset, with some issues addressed.
Reviewers: spatel, hfinkel, wristow, arsenm, javed.absar
Reviewed By: spatel
Subscribers: wdng, nhaehnle
Differential Revision: https://reviews.llvm.org/D47909
llvm-svn: 334996
Summary: Refactoring for all constant cases which require AllowNewConst and some staging for future fmf usage.
Reviewers: spatel, hfinkel, wristow
Reviewed By: spatel
Subscribers: nhaehnle
Differential Revision: https://reviews.llvm.org/D48289
llvm-svn: 334984
Summary: This patch originated from D47388 and is a proper subset of the originating changes, containing only the fmf optimization guard extensions.
Reviewers: spatel, hfinkel, wristow, arsenm, javed.absar, rampitec, nhaehnle, nemanjai
Reviewed By: rampitec, nhaehnle
Subscribers: tpr, nemanjai, wdng
Differential Revision: https://reviews.llvm.org/D47918
llvm-svn: 334876
Summary: This patch originated from D46562 and is a proper subset, with some issues addressed.
Reviewers: spatel, hfinkel, wristow, arsenm
Reviewed By: spatel
Subscribers: wdng, nhaehnle
Differential Revision: https://reviews.llvm.org/D47954
llvm-svn: 334862
Summary:
Here we relax the old constraint which utilized unsafe with the TargetOption flag HonorSignDependentRoundingFPMathOption, with the assertion that unsafe is no longer needed or never was required for correctness on FDIV/FMUL.
Reviewers: spatel, hfinkel, wristow, arsenm, javed.absar
Reviewed By: spatel
Subscribers: efriedma, wdng, tpr
Differential Revision: https://reviews.llvm.org/D48057
llvm-svn: 334769
Summary: A FMF constraint is added to FADD with unsafe still available as the fallback
Reviewers: spatel, wristow, arsenm, hfinkel
Reviewed By: spatel
Subscribers: wdng
Differential Revision: https://reviews.llvm.org/D48180
llvm-svn: 334753
We're constant folding here, so we shouldn't check uses. This matches
the IR optimizer behavior.
The x86 test shows the expected win. The AArch64 test shows something
else. This only seems to happen if the "generic" AArch64 CPU model is
used by MachineCombiner, so I'll file a bug report to follow-up.
llvm-svn: 334608
Summary: This patch originated from D46562 and is a proper subset, with some issues addressed for fmul.
Reviewers: spatel, hfinkel, wristow, arsenm
Reviewed By: spatel
Subscribers: nhaehnle, wdng
Differential Revision: https://reviews.llvm.org/D47911
llvm-svn: 334514
Implement default legalization of rotates: either in terms of the rotation
in the opposite direction (if legal), or in terms of shifts and ors.
Implement generating of rotate instructions for Hexagon. Hexagon only
supports rotates by an immediate value, so implement custom lowering of
ROTL/ROTR on Hexagon. If a rotate is not legal, use the default expansion.
Differential Revision: https://reviews.llvm.org/D47725
llvm-svn: 334497
This would fail before because 1x vectors aren't legal,
so instead just use the scalar type.
Avoids regressions in a future AMDGPU commit to add
v4i16/v4f16 as legal types.
Test update is just the one test that this triggers
on in tree now. It wasn't checking anything before.
The result is completely changed since the selects
are eliminated. Not sure if it's considered better
or not.
llvm-svn: 334440
This patch started off much more general and ambitious, but it's been a nightmare
seeing all the ways x86 vector codegen can go wrong.
So the code is still structured to allow extending easily, but it's currently
limited in several ways:
1. Only handle cases with an extending load.
2. Only handle cases with a zero constant compare.
3. Ignore setcc with vector bitmask (SetCCWidth != 1) - so AVX512 should be unaffected.
The motivating case from PR37427:
https://bugs.llvm.org/show_bug.cgi?id=37427
...is the 1st test, and that shows the expected win - we eliminated the unnecessary
intermediate cast.
There's a clear regression in the last test (sgt_zero_fp_select) because we longer
recognize a 'SHRUNKBLEND' opportunity. I think that general problem is also present
in sgt_zero, so I'll try to fix that in a follow-up. We need to match a sign-bit
setcc from a sign-extended operand and remove it.
Differential Revision: https://reviews.llvm.org/D47330
llvm-svn: 334378
SmallSet forwards to SmallPtrSet for pointer types. SmallPtrSet supports iteration, but a normal SmallSet doesn't. So if it wasn't for the forwarding, this wouldn't work.
These places were found by hiding the begin/end methods in the SmallSet forwarding
llvm-svn: 334343
Summary: This patch originated from D46562 and is a proper subset, with some issues addressed for fsub.
Reviewers: spatel, hfinkel, wristow, arsenm
Reviewed By: spatel
Subscribers: wdng
Differential Revision: https://reviews.llvm.org/D47910
llvm-svn: 334306
While trying to propagate AND masks back to loads, we currently allow
one non-load node to be included as a leaf in chain. This fix now
limits that node to produce only a single data value.
Differential Revision: https://reviews.llvm.org/D47878
llvm-svn: 334268
Summary: This change uses fmf subflags to guard fma optimizations as well as unsafe. These changes originated from D46483 and have been simplified via getNode.
Reviewers: spatel, arsenm, hfinkel, javed.absar
Reviewed By: spatel
Subscribers: nemanjai, wdng
Differential Revision: https://reviews.llvm.org/D47388
llvm-svn: 334242
This avoids regressions in a future AMDGPU change
to make v4i16/v4f16 legal. For these types, build_vector
is implemented as bitcasted operations on v2i32. This
combine was creating v4i16s out of what would have been
already been a v2i32 build_vector, creating a mess
of nodes that never get cleaned up.
I'm not sure this is the right condition to check.
I initially tried just checking for the legality of the
new build_vector. This works for my case, but breaks dozens
of x86 tests. A Mips test seems to show some improvement
or at least a neutral change. I don't want to think
about how long it would take to analyze the set of
different x86 vector operations impacted.
Test included in future commit.
llvm-svn: 334218
Summary:
This change uses fmf subflags to guard optimizations as well as unsafe. These changes originated from D46483.
It contains only context for fsqrt.
Reviewers: spatel, hfinkel, arsenm
Reviewed By: spatel
Subscribers: hfinkel, wdng, andrew.w.kaylor, wristow, efriedma, nemanjai
Differential Revision: https://reviews.llvm.org/D47749
llvm-svn: 334113
Summary: This include variant for add, uaddo and addcarry. usubo and subcarry require the carry to be flipped to preserve semantic, but we chose to do the transform anyway in that case as to push the transform down the carry chain.
Reviewers: efriedma, spatel, RKSimon, zvi, bkramer
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D46505
llvm-svn: 333943
Summary: It has been deprecated in favor of SETCCCARRY for a year now and isn't used by any in tree backend.
Reviewers: efriedma, craig.topper, dblaikie, bkramer
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D47685
llvm-svn: 333939
Summary:
As pointed out in D46528, we errneously transform cases like `xor X, -1`,
even though we use said function.
It's because the `-1` is actually a bitcast there.
So i think we can just look through it in the function.
Differential Revision: https://reviews.llvm.org/D47156
llvm-svn: 332905
Summary:
This **appears** to be the last missing piece for the masked merge pattern handling in the backend.
This is [[ https://bugs.llvm.org/show_bug.cgi?id=37104 | PR37104 ]].
[[ https://bugs.llvm.org/show_bug.cgi?id=6773 | PR6773 ]] will introduce an IR canonicalization that is likely bad for the end assembly.
Previously, `andps`+`andnps` / `bsl` would be generated. (see `@out`)
Now, they would no longer be generated (see `@in`), and we need to make sure that they are generated.
Differential Revision: https://reviews.llvm.org/D46528
llvm-svn: 332904
SimplifyDemandedBits can remove bits from the masks for the shift amounts we need to see to detect rotates.
This patch uses zeroes from computeKnownBits to fill in some of these mask bits to make the match work.
As currently written this calls computeKnownBits even when the mask hasn't been simplified because it made the code simpler. If we're worried about compile time performance we can improve this.
I know we're talking about making a rotate intrinsic, but hopefully we can go ahead and do this change and just make sure the rotate intrinsic also handles it.
Differential Revision: https://reviews.llvm.org/D47116
llvm-svn: 332895
As part of merging stores we check that fusing the nodes does not
cause a cycle due to one candidate store being indirectly dependent on
another store (this may happen via chained memory copies). This is
done by searching if a store is a predecessor to another store's
value.
Prune the search at the candidate search's root node which is a
predecessor to all candidate stores. This reduces the
size of the subgraph searched in large basic blocks.
Reviewers: jyknight
Subscribers: llvm-commits, hiraditya
Differential Revision: https://reviews.llvm.org/D46955
llvm-svn: 332490
search. NFCI.
Migrate single-use and non-volatility, non-indexed requirements on
stores of immediate store values to candidate collection pass from
later stage.
llvm-svn: 332392
The DEBUG() macro is very generic so it might clash with other projects.
The renaming was done as follows:
- git grep -l 'DEBUG' | xargs sed -i 's/\bDEBUG\s\?(/LLVM_DEBUG(/g'
- git diff -U0 master | ../clang/tools/clang-format/clang-format-diff.py -i -p1 -style LLVM
- Manual change to APInt
- Manually chage DOCS as regex doesn't match it.
In the transition period the DEBUG() macro is still present and aliased
to the LLVM_DEBUG() one.
Differential Revision: https://reviews.llvm.org/D43624
llvm-svn: 332240
ExtendSetCCUses updates SETCC nodes which use a load (OriginalLoad) to
reflect a simplification to the load (ExtLoad).
Based on my reading, ExtendSetCCUses may create new nodes to extend a
constant attached to a SETCC. It also creates fresh SETCC nodes which
refer to any updated operands.
ISTM that the location applied to the new constant and SETCC nodes
should be the same as the location of the ExtLoad.
This was suggested by Adrian in https://reviews.llvm.org/D45995.
Part of: llvm.org/PR37262
Differential Revision: https://reviews.llvm.org/D46216
llvm-svn: 332119
This teaches tryToFoldExtOfLoad to set the right location on a
newly-created extload. With that in place, the logic for performing a
certain ([s|z]ext (load ...)) combine becomes identical for sexts and
zexts, and we can get rid of one copy of the logic.
The test case churn is due to dependencies on IROrders inherited from
the wrong SDLoc.
Part of: llvm.org/PR37262
Differential Revision: https://reviews.llvm.org/D46158
llvm-svn: 332118
Part of the logic for combining (zext (load ...)) and (sext (load ...))
is duplicated. This creates problems because bugs in one version have to
be fixed again in the other version.
To address this, as a first step, I've extracted the duplicate logic
into a helper. I'll fix the debug location bug in the helper and
eliminate the copy of its logic in a followup.
Part of: llvm.org/PR37262
Differential Revision: https://reviews.llvm.org/D46157
llvm-svn: 332117
Summary:
The combine in rebuildSetCC may be combined to another
node leaving our references stale. Keep a handle on
it to avoid stale references.
Fixes PR36602.
Reviewers: dbabokin, RKSimon, eli.friedman, davide
Subscribers: hiraditya, uabelho, JesperAntonsson, qcolombet, llvm-commits
Differential Revision: https://reviews.llvm.org/D46404
llvm-svn: 331985
Previously if !LegalOperations we would blindly call getBitcast and hope that getNode would constant fold it. But if the conversion is between a vector and a scalar, getNode has no simplification.
This means we would just get back the original N. We would then return that N which would make the caller of visitBITCAST think that we used CombineTo and did our own worklist management. This prevents target specific optimizations from being called for vector/scalar bitcasts until after legal operations.
llvm-svn: 331896
The previous value of 8192 resulted in severe compile time hits in
some pathological cases.
rdar://39781410
Differential Revision: https://reviews.llvm.org/D46581
llvm-svn: 331888
Summary:
Split off from D46031.
The previous patch, D46493, completely disabled unfolding in case of immediates.
But we can do better:
{F6120274} {F6120277}
https://rise4fun.com/Alive/xJS
Reviewers: spatel, craig.topper
Reviewed By: spatel
Subscribers: andreadb, llvm-commits
Differential Revision: https://reviews.llvm.org/D46494
llvm-svn: 331685
Summary:
Split off from D46031.
In masked merge case, this degrades IPC by decreasing instruction count.
{F6108777}
The next patch should be able to recover and improve this.
This also affects the transform @spatel have added in D27489 / rL289738,
and the test coverage for X86 was missing.
But after i have added it, and looked at the changes in MCA, i'm somewhat confused.
{F6093591} {F6093592} {F6093593}
I'd say this regression is an improvement, since `IPC` increased in that case?
Reviewers: spatel, craig.topper
Reviewed By: spatel
Subscribers: andreadb, llvm-commits, spatel
Differential Revision: https://reviews.llvm.org/D46493
llvm-svn: 331684
Summary:
Split off form D46031.
It seems we don't want to transform the pattern if the `xor`'s are actually `not`'s.
In vector case, this breaks `andnpd` / `vandnps` patterns.
That being said, we may want to re-visit this `not` handling, maybe in D46073.
Reviewers: spatel, craig.topper, javed.absar
Reviewed By: spatel
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D46492
llvm-svn: 331595
Inspired by r331508, I did a grep and found these.
Mostly just change from dyn_cast to cast. Some cases also showed a dyn_cast result being converted to bool, so those I changed to isa.
llvm-svn: 331577
Summary: Adding support for Fast flags in the SDNode to leverage fast math sub flag usage.
Reviewers: spatel, arsenm, jbhateja, hfinkel, escha, qcolombet, echristo, wristow, javed.absar
Reviewed By: spatel
Subscribers: llvm-commits, rampitec, nhaehnle, tstellar, FarhanaAleen, nemanjai, javed.absar, jbhateja, hfinkel, wdng
Differential Revision: https://reviews.llvm.org/D45710
llvm-svn: 331547
The logic for this combine is almost identical to the logic for a
(sext (sextload x)) combine.
This commit factors out the logic so it can be shared by both combines,
and corrects the SDLoc assigned in the zext version of the combine.
Prior to this patch, for the given test case, we would apply the
location associated with the udiv instruction to instructions which
perform the load.
Part of: llvm.org/PR37262
llvm-svn: 331303
Prior to this patch, for the given test case, we would apply the
location associated with the sdiv instruction to instructions which
perform the load.
Part of: llvm.org/PR37262.
Differential Revision: https://reviews.llvm.org/D46222
llvm-svn: 331302
In DAGCombiner, we try to simplify this pattern:
([s|z]ext (load ...))
Conceptually, a new extload which is created while splitting the load
should have the same debug location as the load.
Making this change affects the IROrder of the new load, causing some
test case churn.
In practice, the new location is never different from the location of
the [s|z]ext, at least not during check-llvm or a stage2 build.
Part of: llvm.org/PR37262
Differential Revision: https://reviews.llvm.org/D46156
llvm-svn: 331301
Setting the right SDLoc on a newly-created zextload fixes a line table
bug which resulted in non-linear stepping behavior.
Several backend tests contained CHECK lines which relied on the IROrder
inherited from the wrong SDLoc. This patch breaks that dependence where
feasbile and regenerates test cases where not.
In some cases, changing a node's IROrder may alter register allocation
and spill behavior. This can affect performance. I have chosen not to
prevent this by applying a "known good" IROrder to SDLocs, as this may
hide a more general bug in the scheduler, or cause regressions on other
test inputs.
rdar://33755881, Part of: llvm.org/PR37262
Differential Revision: https://reviews.llvm.org/D45995
llvm-svn: 331300
We've been running doxygen with the autobrief option for a couple of
years now. This makes the \brief markers into our comments
redundant. Since they are a visual distraction and we don't want to
encourage more \brief markers in new code either, this patch removes
them all.
Patch produced by
for i in $(git grep -l '\\brief'); do perl -pi -e 's/\\brief //g' $i & done
Differential Revision: https://reviews.llvm.org/D46290
llvm-svn: 331272
Summary:
D42479 (rL329525) enabled SDIV combine for pow2 non-splat vector
dividers. But when there is a 1 in a vector, the instruction sequence to
be generated involves shifting a value by the number of its bit widths,
which is undefined
(c64f4dbfe3/lib/CodeGen/SelectionDAG/DAGCombiner.cpp (L6000-L6006)).
Especially, in architectures that do not support vector instructions,
each of element in a vector will be computed separately using scalar
operations, and then the resulting value will be undef for '1' values
in a vector.
(All 1's vector is fine; only vectors mixed with 1 and others will be
affected.)
Reviewers: RKSimon, jgravelle-google
Subscribers: jfb, dschuff, sbc100, jgravelle-google, llvm-commits
Differential Revision: https://reviews.llvm.org/D46161
llvm-svn: 331092
As noted, the attribute name is subject to change once we have
the clang side implemented, but it's clear that we need some
kind of attribute-based predication here based on the discussion
for:
rL330437
llvm-svn: 330951
As discussed in the post-review comments for rL330437,
we need to guard this fold to allow existing code to
keep working with the undefined behavior that they've
come to rely on.
That would mean duplicating more code than we already
have, so let's fix that first.
llvm-svn: 330947
We were previously prefering ZEXTLOAD over EXTLOAD if it is legal. This triggers during X86's promotion of i16->i32. Not sure about other targets.
Using ZEXTLOAD can prevent folding it to SEXTLOAD later if we were to promote a sign extended operand like we would need for SRA. However, X86 doesn't currently promote i16 SRA. I was looking into doing that which is how I found this issue.
This is also blocking our ability to fold 4 byte aligned EXTLOADs with "loadi32". This is what caused most of the test changes here.
Differential Revision: https://reviews.llvm.org/D45585#inline-402825
llvm-svn: 330781
Summary:
This is [[ https://bugs.llvm.org/show_bug.cgi?id=37104 | PR37104 ]].
[[ https://bugs.llvm.org/show_bug.cgi?id=6773 | PR6773 ]] will introduce an IR canonicalization that is likely bad for the end assembly.
Previously, `andl`+`andn`/`andps`+`andnps` / `bic`/`bsl` would be generated. (see `@out`)
Now, they would no longer be generated (see `@in`).
So we need to make sure that they are still generated.
If the mask is constant, we do nothing. InstCombine should have unfolded it.
Else, i use `hasAndNot()` TLI hook.
For now, only handle scalars.
https://rise4fun.com/Alive/bO6
----
I *really* don't like the code i wrote in `DAGCombiner::unfoldMaskedMerge()`.
It is super fragile. Is there something like IR Pattern Matchers for this?
Reviewers: spatel, craig.topper, RKSimon, javed.absar
Reviewed By: spatel
Subscribers: andreadb, courbet, kristof.beyls, javed.absar, rengolin, nemanjai, llvm-commits
Differential Revision: https://reviews.llvm.org/D45733
llvm-svn: 330646
This was originally committed at rL328921 and reverted at rL329920 to
investigate failures in Chrome. This time I've added to the ReleaseNotes
to warn users of the potential of exposing UB and let me repeat that
here for more exposure:
Optimization of floating-point casts is improved. This may cause surprising
results for code that is relying on undefined behavior. Code sanitizers can
be used to detect affected patterns such as this:
int main() {
float x = 4294967296.0f;
x = (float)((int)x);
printf("junk in the ftrunc: %f\n", x);
return 0;
}
$ clang -O1 ftrunc.c -fsanitize=undefined ; ./a.out
ftrunc.c:5:15: runtime error: 4.29497e+09 is outside the range of
representable values of type 'int'
junk in the ftrunc: 0.000000
Original commit message:
fptosi / fptoui round towards zero, and that's the same behavior as ISD::FTRUNC,
so replace a pair of casts with the equivalent node. We don't have to account for
special cases (NaN, INF) because out-of-range casts are undefined.
Differential Revision: https://reviews.llvm.org/D44909
llvm-svn: 330437
This is a transform that I limited in instcombine in rL329821 because it was
creating more instructions in IR when the cast has multiple uses.
But if the cast is free, then we can do the transform regardless of other
uses because it improves the potential throughput of the calculation by
removing a dependency on the fneg.
Differential Revision: https://reviews.llvm.org/D45598
llvm-svn: 330098
This change is exposing UB in source code - as was warned/predicted. :)
See D44909 for discussion. Reverting while we figure out how to fix things.
llvm-svn: 329920
Recommitting r329283, third time lucky...
If the SRL node is only used by an AND, we may be able to set the
ExtVT to the width of the mask, making the AND redundant. To support
this, another check has been added in isLegalNarrowLoad which queries
whether the load is valid.
Differential Revision: https://reviews.llvm.org/D41350
llvm-svn: 329551
In our real world application, we found the following optimization is missed in DAGCombiner
(zext (and/or/xor (shl/shr (load x), cst), cst)) -> (and/or/xor (shl/shr (zextload x), (zext cst)), (zext cst))
If the user of original zext is an add, it may enable further lea optimization on x86.
This patch add a new function CombineZExtLogicopShiftLoad to do this optimization.
Differential Revision: https://reviews.llvm.org/D44402
llvm-svn: 329516
Summary:
r327219 added wrappers to std::sort which randomly shuffle the container before sorting.
This will help in uncovering non-determinism caused due to undefined sorting
order of objects having the same key.
To make use of that infrastructure we need to invoke llvm::sort instead of std::sort.
Note: This patch is one of a series of patches to replace *all* std::sort to llvm::sort.
Refer the comments section in D44363 for a list of all the required patches.
Reviewers: bogner, rnk, MatzeB, RKSimon
Reviewed By: rnk
Subscribers: JDevlieghere, javed.absar, llvm-commits
Differential Revision: https://reviews.llvm.org/D45133
llvm-svn: 329435
Recommitting rL321259. Previosuly this caused an issue with PPCBE but
I didn't receieve a reproducer and didn't have the time to follow up.
If the issue appears again, please provide a reproducer so I can fix
it.
Original commit message:
If the SRL node is only used by an AND, we may be able to set the
ExtVT to the width of the mask, making the AND redundant. To support
this, another check has been added in isLegalNarrowLoad which queries
whether the load is valid.
Differential Revision: https://reviews.llvm.org/D41350
llvm-svn: 329160
fptosi / fptoui round towards zero, and that's the same behavior as ISD::FTRUNC,
so replace a pair of casts with the equivalent node. We don't have to account for
special cases (NaN, INF) because out-of-range casts are undefined.
Differential Revision: https://reviews.llvm.org/D44909
llvm-svn: 328921
The code has bugs dealing with -0.0.
Since D44550 introduced FABS pattern folding in InstCombine,
this patch removes the now-redundant code that causes
https://bugs.llvm.org/show_bug.cgi?id=36600.
Patch by Mikhail Dvoretckii!
Differential Revision: https://reviews.llvm.org/D44683
llvm-svn: 328872
Currently EVT is in the IR layer only because of Function.cpp needing a very small piece of the functionality of EVT::getEVTString(). The rest of EVT is used in codegen making CodeGen a better place for it.
The previous code converted a Type* to EVT and then called getEVTString. This was only expected to handle the primitive types from Type*. Since there only a few primitive types, we can just print them as strings directly.
Differential Revision: https://reviews.llvm.org/D45017
llvm-svn: 328806
This is used by llvm tblgen as well as by LLVM Targets, so the only
common place is Support for now. (maybe we need another target for these
sorts of things - but for now I'm at least making them correct & we can
make them better if/when people have strong feelings)
llvm-svn: 328395
In our real world application, we found the following optimization is missed in DAGCombiner
(zext (and/or/xor (shl/shr (load x), cst), cst)) -> (and/or/xor (shl/shr (zextload x), (zext cst)), (zext cst))
If the user of original zext is an add, it may enable further lea optimization on x86.
This patch add a new function CombineZExtLogicopShiftLoad to do this optimization.
Differential Revision: https://reviews.llvm.org/D44402
llvm-svn: 328252
