This is meant to be used with clang's __builtin_dynamic_object_size.
When 'true' is passed to this parameter, the intrinsic has the
potential to be folded into instructions that will be evaluated
at run time. When 'false', the objectsize intrinsic behaviour is
unchanged.
rdar://32212419
Differential revision: https://reviews.llvm.org/D56761
llvm-svn: 352664
The point is that this simplifies integration of new intrinsics into SimplifiedDemandedVectorElts, and ensures we don't miss any existing ones.
This is intended to be NFC-ish, but as seen from the diffs, can produce slightly different output. This is due to order of transforms w/in instcombine resulting in two slightly different fixed points. That's something we should fix, but isn't a problem w/this patch per se.
Differential Revision: https://reviews.llvm.org/D57398
llvm-svn: 352653
I'm circling back around to a loose end from D51929.
The backend (either CGP or DAG) doesn't recognize this pattern, so we end up with different asm for these IR variants.
Regardless of any future changes to canonicalize to saturation/overflow intrinsics, we want to get raw IR variations
into the minimal number of raw IR forms. If/when we can canonicalize to intrinsics, that will make that step easier.
Pre: C2 == ~C1
%a = add i32 %x, C1
%c = icmp ugt i32 %x, C2
%r = select i1 %c, i32 -1, i32 %a
=>
%a = add i32 %x, C1
%c2 = icmp ult i32 %x, C2
%r = select i1 %c2, i32 %a, i32 -1
https://rise4fun.com/Alive/pkH
Differential Revision: https://reviews.llvm.org/D57352
llvm-svn: 352536
GEPs can produce either scalar or vector results. If we're extracting only a subset of the vector lanes, simplifying the operands is helpful in eliminating redundant computation, and (eventually) allowing further optimizations
Differential Revision: https://reviews.llvm.org/D57177
llvm-svn: 352440
This causes a couple of changes in the upgrade tests as signed/unsigned eq/ne are equivalent and we constant fold true/false codes, these changes are the same as what we already do for avx512 cmp/ucmp.
Noticed while cleaning up vector integer comparison costs for PR40376.
llvm-svn: 351697
Followup to D55745, this time handling comparisons with ugt and ult
predicates (which are the canonical forms for non-equality predicates).
For ctlz we can convert into a simple icmp, for cttz we can convert
into a mask check.
Differential Revision: https://reviews.llvm.org/D56355
llvm-svn: 351645
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
Summary:
InstCombine's sinking algorithm only thinks about memory. It doesn't
think about non-memory constraints like stack object lifetime. It can
sink dynamic allocas across a stacksave call, which may be used with
stackrestore, which can incorrectly reduce the lifetime of the dynamic
alloca.
Fixes PR40365
Reviewers: hfinkel, efriedma
Subscribers: hiraditya, llvm-commits
Differential Revision: https://reviews.llvm.org/D56872
llvm-svn: 351475
InstCombine is able to transform mem transfer instrinsic to alone store or store/load pair.
It might result in generation of unaligned atomic load/store which later in backend
will be transformed to libcall. It is not an evident gain and it is better to keep intrinsic as is
and handle it at backend.
Reviewers: reames, anna, apilipenko, mkazantsev
Reviewed By: reames
Subscribers: t.p.northover, jfb, llvm-commits
Differential Revision: https://reviews.llvm.org/D56582
llvm-svn: 351295
Summary:
This allows moving the condition from the intrinsic to the standard ICmp
opcode, so that LLVM can do simplifications on it. The icmp.i1 intrinsic
is an identity for retrieving the SGPR mask.
And we can also get the mask from and i1, or i1, xor i1.
Reviewers: arsenm, nhaehnle
Subscribers: kzhuravl, jvesely, wdng, yaxunl, dstuttard, tpr, t-tye, llvm-commits
Differential Revision: https://reviews.llvm.org/D52060
llvm-svn: 351150
TFE and LWE support requires extra result registers that are written in the
event of a failure in order to detect that failure case.
The specific use-case that initiated these changes is sparse texture support.
This means that if image intrinsics are used with either option turned on, the
programmer must ensure that the return type can contain all of the expected
results. This can result in redundant registers since the vector size must be a
power-of-2.
This change takes roughly 6 parts:
1. Modify the instruction defs in tablegen to add new instruction variants that
can accomodate the extra return values.
2. Updates to lowerImage in SIISelLowering.cpp to accomodate setting TFE or LWE
(where the bulk of the work for these instruction types is now done)
3. Extra verification code to catch cases where intrinsics have been used but
insufficient return registers are used.
4. Modification to the adjustWritemask optimisation to account for TFE/LWE being
enabled (requires extra registers to be maintained for error return value).
5. An extra pass to zero initialize the error value return - this is because if
the error does not occur, the register is not written and thus must be zeroed
before use. Also added a new (on by default) option to ensure ALL return values
are zero-initialized that is required for sparse texture support.
6. Disable the inst_combine optimization in the presence of tfe/lwe (later TODO
for this to re-enable and handle correctly).
There's an additional fix now to avoid a dmask=0
For an image intrinsic with tfe where all result channels except tfe
were unused, I was getting an image instruction with dmask=0 and only a
single vgpr result for tfe. That is incorrect because the hardware
assumes there is at least one vgpr result, plus the one for tfe.
Fixed by forcing dmask to 1, which gives the desired two vgpr result
with tfe in the second one.
The TFE or LWE result is returned from the intrinsics using an aggregate
type. Look in the test code provided to see how this works, but in essence IR
code to invoke the intrinsic looks as follows:
%v = call {<4 x float>,i32} @llvm.amdgcn.image.load.1d.v4f32i32.i32(i32 15,
i32 %s, <8 x i32> %rsrc, i32 1, i32 0)
%v.vec = extractvalue {<4 x float>, i32} %v, 0
%v.err = extractvalue {<4 x float>, i32} %v, 1
This re-submit of the change also includes a slight modification in
SIISelLowering.cpp to work-around a compiler bug for the powerpc_le
platform that caused a buildbot failure on a previous submission.
Differential revision: https://reviews.llvm.org/D48826
Change-Id: If222bc03642e76cf98059a6bef5d5bffeda38dda
Work around for ppcle compiler bug
Change-Id: Ie284cf24b2271215be1b9dc95b485fd15000e32b
llvm-svn: 351054
This is matching the equivalent of the DAG expansion,
so it should never end up with worse perf than the
original code even if the target doesn't have a rotate
instruction.
llvm-svn: 350672
The cttz/ctlz intrinsics have a parameter specifying whether the
result is undefined for zero. cttz(x, false) can be relaxed to
cttz(x, true) if x is known non-zero, and in fact such an optimization
is already performed. However, this currently doesn't work if x is
non-zero as a result of a select rather than an explicit branch.
This patch adds handling for this case, thus allowing
x != 0 ? cttz(x, false) : y to simplify to x != 0 ? cttz(x, true) : y.
Differential Revision: https://reviews.llvm.org/D55786
llvm-svn: 350463
Similar to rL350199 - there are no known analysis/codegen holes for
funnel shift intrinsics now, so we can canonicalize the 6+ regular
instructions to funnel shift to improve vectorization, inlining,
unrolling, etc.
llvm-svn: 350419
The final piece of IR-level analysis to allow this was committed with:
rL350188
Using the intrinsics should improve transforms based on cost models
like vectorization and inlining.
The backend should be prepared too, so we can now canonicalize more
sequences of shift/logic to the intrinsics and know that the end
result should be equal or better to the original code even if the
target does not have an actual rotate instruction.
llvm-svn: 350199
Instruction::isLifetimeStartOrEnd() checks whether an Instruction is an
llvm.lifetime.start or an llvm.lifetime.end intrinsic.
This was suggested as a cleanup in D55967.
Differential Revision: https://reviews.llvm.org/D56019
llvm-svn: 349964
The current llvm.mem.parallel_loop_access metadata has a problem in that
it uses LoopIDs. LoopID unfortunately is not loop identifier. It is
neither unique (there's even a regression test assigning the some LoopID
to multiple loops; can otherwise happen if passes such as LoopVersioning
make copies of entire loops) nor persistent (every time a property is
removed/added from a LoopID's MDNode, it will also receive a new LoopID;
this happens e.g. when calling Loop::setLoopAlreadyUnrolled()).
Since most loop transformation passes change the loop attributes (even
if it just to mark that a loop should not be processed again as
llvm.loop.isvectorized does, for the versioned and unversioned loop),
the parallel access information is lost for any subsequent pass.
This patch unlinks LoopIDs and parallel accesses.
llvm.mem.parallel_loop_access metadata on instruction is replaced by
llvm.access.group metadata. llvm.access.group points to a distinct
MDNode with no operands (avoiding the problem to ever need to add/remove
operands), called "access group". Alternatively, it can point to a list
of access groups. The LoopID then has an attribute
llvm.loop.parallel_accesses with all the access groups that are parallel
(no dependencies carries by this loop).
This intentionally avoid any kind of "ID". Loops that are clones/have
their attributes modifies retain the llvm.loop.parallel_accesses
attribute. Access instructions that a cloned point to the same access
group. It is not necessary for each access to have it's own "ID" MDNode,
but those memory access instructions with the same behavior can be
grouped together.
The behavior of llvm.mem.parallel_loop_access is not changed by this
patch, but should be considered deprecated.
Differential Revision: https://reviews.llvm.org/D52116
llvm-svn: 349725
Checking whether a number has a certain number of trailing / leading
zeros means checking whether it is of the form XXXX1000 / 0001XXXX,
which can be done with an and+icmp.
Related to https://bugs.llvm.org/show_bug.cgi?id=28668. As a next
step, this can be extended to non-equality predicates.
Differential Revision: https://reviews.llvm.org/D55745
llvm-svn: 349530
As the FIXME indicates, this has the potential to go
overboard. So I'm not sure if it's even worth keeping
this vs. iteratively doing simple matches, but we might
as well clean it up.
llvm-svn: 349523
The problem is shown specifically for a case with vector multiply here:
https://bugs.llvm.org/show_bug.cgi?id=40032
...and this might mask the original backend bug for ARM shown in:
https://bugs.llvm.org/show_bug.cgi?id=39967
As the test diffs here show, we were (and probably still aren't) doing
these kinds of transforms in a principled way. We are producing more or
equal wide instructions than we started with in some cases, so we still
need to restrict/correct other transforms from overstepping.
If there are perf regressions from this change, we can either carve out
exceptions to the general IR rules, or improve the backend to do these
transforms when we know the transform is profitable. That's probably
similar to a change like D55448.
Differential Revision: https://reviews.llvm.org/D55744
llvm-svn: 349389
This fixes https://bugs.llvm.org/show_bug.cgi?id=39908.
The evaluateGEPOffsetExpression() function simplifies GEP offsets for
use in comparisons against zero, basically by converting X*Scale+Offset==0
to X+Offset/Scale==0 if Scale divides Offset. However, before this is done,
Offset is masked down to the pointer size. This results in incorrect
results for negative Offsets, because we basically end up dividing the
32-bit offset *zero* extended to 64-bit bits (rather than sign extended).
Fix this by explicitly sign extending the truncated value.
Differential Revision: https://reviews.llvm.org/D55449
llvm-svn: 348987
call iM movmsk(sext <N x i1> X) --> zext (bitcast <N x i1> X to iN) to iM
This has the potential to create less-than-8-bit scalar types as shown in
some of the test diffs, but it looks like the backend knows how to deal
with that in these patterns. This is the simple part of the fix suggested in:
https://bugs.llvm.org/show_bug.cgi?id=39927
Differential Revision: https://reviews.llvm.org/D55529
llvm-svn: 348862
I was finally able to quantify what i thought was missing in the fix,
it was vector constants. If we have a scalar (and %x, -1),
it will be instsimplified before we reach this code,
but if it is a vector, we may still have a -1 element.
Thus, we want to avoid the fold if *at least one* element is -1.
Or in other words, ignoring the undef elements, no sign bits
should be set. Thus, m_NonNegative().
A follow-up for rL348181
https://bugs.llvm.org/show_bug.cgi?id=39861
llvm-svn: 348462
Extracting from a splat constant is always handled by InstSimplify.
Move the test for this from InstCombine to InstSimplify to make
sure that stays true.
llvm-svn: 348423
The tests here are based on the motivating cases from D54827.
More background:
1. We don't get these cases in general with SimplifyCFG because the root
of the pattern match is an icmp, not a branch. I'm not sure how often
we encounter this pattern vs. the seemingly more likely case with
branches, but I don't see evidence to leave the minimal pattern
unoptimized.
2. This has a chance of increasing compile-time because we're using a
ValueTracking call to handle the match. The motivating cases could be
handled with a simpler pair of calls to isImpliedTrueByMatchingCmp/
isImpliedFalseByMatchingCmp, but I saw that we have a more
comprehensive wrapper around those, so we might as well use it here
unless there's evidence that it's significantly slower.
3. Ideally, we'd handle the fold to constants in InstSimplify, but as
with the existing code here, we could extend this to handle cases
where the result is not a constant, but a new combined predicate.
That would mean splitting the logic across the 2 passes and possibly
duplicating the pattern-matching cost.
4. As mentioned in D54827, this seems like the kind of thing that should
be handled in Correlated Value Propagation, but that pass is currently
limited to dealing with instructions with constant operands, so extending
this bit of InstCombine is the smallest/easiest way to get these patterns
optimized.
llvm-svn: 348367
Move it out from under the constant check, reorder
predicates, add comments. This makes it easier to
extend to handle the non-constant case.
llvm-svn: 348284
There's a potential small enhancement to this code that could
solve the cases currently under proposal in D54827 via SimplifyCFG.
Whether instcombine should be doing this kind of semi-non-local
analysis in the first place is an open question, but separating
the logic out can only help if/when we decide to move it to a
different pass.
AFAICT, any proposal to do this in SimplifyCFG could also be seen
as an overreach + it would be incomplete to start the fold from a
branch rather than an icmp.
There's another question here about the code for processUGT_ADDCST_ADD().
That part may be completely dead after rL234638 ?
llvm-svn: 348273
When we have a shuffle that extends a source vector with undefs
and then do some binop on that, we must make sure that the extra
elements remain undef with that binop if we reverse the order of
the binop and shuffle.
'or' is probably the easiest example to show the bug because
'or C, undef --> -1' (not undef). But there are other
opcode/constant combinations where this is true as shown by
the 'shl' test.
llvm-svn: 348191
There are potential improvements to the structure of this API
raised by D54994, but remove some cosmetic blemishes before
making any functional changes.
llvm-svn: 348149
Extend ssub.sat(X, C) -> sadd.sat(X, -C) canonicalization to also
support non-splat vector constants. This is done by generalizing
the implementation of the isNotMinSignedValue() helper to return
true for constants that are non-splat, but don't contain any
signed min elements.
Differential Revision: https://reviews.llvm.org/D55011
llvm-svn: 348072
Also revert fix r347876
One of the buildbots was reporting a failure in some relevant tests that I can't
repro or explain at present, so reverting until I can isolate.
llvm-svn: 347911
This is an almost direct move of the functionality from InstCombine to
InstSimplify. There's no reason not to do this in InstSimplify because
we never create a new value with this transform.
(There's a question of whether any dominance-based transform belongs in
either of these passes, but that's a separate issue.)
I've changed 1 of the conditions for the fold (1 of the blocks for the
branch must be the block we started with) into an assert because I'm not
sure how that could ever be false.
We need 1 extra check to make sure that the instruction itself is in a
basic block because passes other than InstCombine may be using InstSimplify
as an analysis on values that are not wired up yet.
The 3-way compare changes show that InstCombine has some kind of
phase-ordering hole. Otherwise, we would have already gotten the intended
final result that we now show here.
llvm-svn: 347896
TFE and LWE support requires extra result registers that are written in the
event of a failure in order to detect that failure case.
The specific use-case that initiated these changes is sparse texture support.
This means that if image intrinsics are used with either option turned on, the
programmer must ensure that the return type can contain all of the expected
results. This can result in redundant registers since the vector size must be a
power-of-2.
This change takes roughly 6 parts:
1. Modify the instruction defs in tablegen to add new instruction variants that
can accomodate the extra return values.
2. Updates to lowerImage in SIISelLowering.cpp to accomodate setting TFE or LWE
(where the bulk of the work for these instruction types is now done)
3. Extra verification code to catch cases where intrinsics have been used but
insufficient return registers are used.
4. Modification to the adjustWritemask optimisation to account for TFE/LWE being
enabled (requires extra registers to be maintained for error return value).
5. An extra pass to zero initialize the error value return - this is because if
the error does not occur, the register is not written and thus must be zeroed
before use. Also added a new (on by default) option to ensure ALL return values
are zero-initialized that is required for sparse texture support.
6. Disable the inst_combine optimization in the presence of tfe/lwe (later TODO
for this to re-enable and handle correctly).
There's an additional fix now to avoid a dmask=0
For an image intrinsic with tfe where all result channels except tfe
were unused, I was getting an image instruction with dmask=0 and only a
single vgpr result for tfe. That is incorrect because the hardware
assumes there is at least one vgpr result, plus the one for tfe.
Fixed by forcing dmask to 1, which gives the desired two vgpr result
with tfe in the second one.
The TFE or LWE result is returned from the intrinsics using an aggregate
type. Look in the test code provided to see how this works, but in essence IR
code to invoke the intrinsic looks as follows:
%v = call {<4 x float>,i32} @llvm.amdgcn.image.load.1d.v4f32i32.i32(i32 15,
i32 %s, <8 x i32> %rsrc, i32 1, i32 0)
%v.vec = extractvalue {<4 x float>, i32} %v, 0
%v.err = extractvalue {<4 x float>, i32} %v, 1
Differential revision: https://reviews.llvm.org/D48826
Change-Id: If222bc03642e76cf98059a6bef5d5bffeda38dda
llvm-svn: 347871
Combine
sat(sat(X + C1) + C2) -> sat(X + (C1+C2))
and
sat(sat(X - C1) - C2) -> sat(X - (C1+C2))
if the sign of C1 and C2 matches.
In the unsigned case we can compute C1+C2 with saturating arithmetic,
and InstSimplify will reduce this just to the saturation value. For
the signed case, we cannot perform the simplification if the result
of the addition overflows.
This change is part of https://reviews.llvm.org/D54534.
llvm-svn: 347773
Canonicalize ssub.sat(X, C) to ssub.sat(X, -C) if C is constant and
not signed minimum. This will help further optimizations to apply.
This change is part of https://reviews.llvm.org/D54534.
llvm-svn: 347772
If ValueTracking can determine that the add/sub can newer overflow,
replace it with the corresponding nuw/nsw add/sub.
Additionally, for the unsigned case, if ValueTracking determines
that the add/sub always overflows, replace the result with the
saturation value.
This change is part of https://reviews.llvm.org/D54534.
llvm-svn: 347770
If a saturating add intrinsic has one constant argument, make sure
it is on the RHS. This will simplify further transformations.
This change is part of https://reviews.llvm.org/D54534.
llvm-svn: 347769
I tried to change this, not quite realising the logic behind what we
were doing. Hopefully this comment will help the next person to come
along.
llvm-svn: 347653
Support funnel shifts in InstCombine demanded bits simplification.
If the shift amount is constant, we can determine both the demanded
bits of the operands, as well as the known bits of the result.
If one of the operands has no demanded bits, it will be replaced
by undef and the funnel shift will be simplified into a simple shift
due to the simplifications added in D54778.
Differential Revision: https://reviews.llvm.org/D54869
llvm-svn: 347515
The following simplifications are implemented:
* `fshl(X, 0, C) -> shl X, C%BW`
* `fshl(X, undef, C) -> shl X, C%BW` (assuming undef = 0)
* `fshl(0, X, C) -> lshr X, BW-C%BW`
* `fshl(undef, X, C) -> lshr X, BW-C%BW` (assuming undef = 0)
* `fshr(X, 0, C) -> shl X, (BW-C%BW)`
* `fshr(X, undef, C) -> shl X, BW-C%BW` (assuming undef = 0)
* `fshr(0, X, C) -> lshr X, C%BW`
* `fshr(undef, X, C) -> lshr, X, C%BW` (assuming undef = 0)
The simplification is only performed if the shift amount C is constant,
because we can explicitly compute C%BW and BW-C%BW in this case.
Differential Revision: https://reviews.llvm.org/D54778
llvm-svn: 347505
Add methods to BasicBlock which make it easier to efficiently check
whether a block has N (or more) predecessors.
This can be more efficient than using pred_size(), which is a linear
time operation.
We might consider adding similar methods for successors. I haven't done
so in this patch because succ_size() is already O(1).
With this patch applied, I measured a 0.065% compile-time reduction in
user time for running `opt -O3` on the sqlite3 amalgamation (30 trials).
The change in mergeStoreIntoSuccessor alone saves 45 million linked list
iterations in a stage2 Release build of llc.
See llvm.org/PR39702 for a harder but more general way of achieving
similar results.
Differential Revision: https://reviews.llvm.org/D54686
llvm-svn: 347256
Summary:
These asserts are based on the assumption that the order of true/false operands in a select and those in the compare would always be the same.
This fixes PR39595.
Reviewers: craig.topper, spatel, dmgreen
Reviewed By: craig.topper
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D54359
llvm-svn: 346874
The shift amount of a funnel shift is modulo the scalar bitwidth:
http://llvm.org/docs/LangRef.html#llvm-fshl-intrinsic
...so we can use demanded bits analysis on that operand to simplify it
when we have a power-of-2 bitwidth.
This is another step towards canonicalizing {shift/shift/or} to the
intrinsics in IR.
Differential Revision: https://reviews.llvm.org/D54478
llvm-svn: 346814
The cmp+branch variant of this pattern is shown in:
https://bugs.llvm.org/show_bug.cgi?id=34924
...and as discussed there, we probably can't transform
that without a rotate intrinsic. We do have that now
via funnel shift, but we're not quite ready to
canonicalize IR to that form yet. The case with 'select'
should already be transformed though, so that's this patch.
The sequence with negation followed by masking is what we
use in the backend and partly in clang (though that part
should be updated).
https://rise4fun.com/Alive/TplC
%cmp = icmp eq i32 %shamt, 0
%sub = sub i32 32, %shamt
%shr = lshr i32 %x, %shamt
%shl = shl i32 %x, %sub
%or = or i32 %shr, %shl
%r = select i1 %cmp, i32 %x, i32 %or
=>
%neg = sub i32 0, %shamt
%masked = and i32 %shamt, 31
%maskedneg = and i32 %neg, 31
%shl2 = lshr i32 %x, %masked
%shr2 = shl i32 %x, %maskedneg
%r = or i32 %shl2, %shr2
llvm-svn: 346807
This is a longer variant for the pattern handled in
rL346713
This one includes zexts.
Eventually, we should canonicalize all rotate patterns
to the funnel shift intrinsics, but we need a bit more
infrastructure to make sure the vectorizers handle those
intrinsics as well as the shift+logic ops.
https://rise4fun.com/Alive/FMn
Name: narrow rotateright
%neg = sub i8 0, %shamt
%rshamt = and i8 %shamt, 7
%rshamtconv = zext i8 %rshamt to i32
%lshamt = and i8 %neg, 7
%lshamtconv = zext i8 %lshamt to i32
%conv = zext i8 %x to i32
%shr = lshr i32 %conv, %rshamtconv
%shl = shl i32 %conv, %lshamtconv
%or = or i32 %shl, %shr
%r = trunc i32 %or to i8
=>
%maskedShAmt2 = and i8 %shamt, 7
%negShAmt2 = sub i8 0, %shamt
%maskedNegShAmt2 = and i8 %negShAmt2, 7
%shl2 = lshr i8 %x, %maskedShAmt2
%shr2 = shl i8 %x, %maskedNegShAmt2
%r = or i8 %shl2, %shr2
llvm-svn: 346716
The sub-pattern for the shift amount in a rotate can take on
several different forms, and there's apparently no way to
canonicalize those without seeing the entire rotate sequence.
This is the form noted in:
https://bugs.llvm.org/show_bug.cgi?id=39624https://rise4fun.com/Alive/qnT
%zx = zext i8 %x to i32
%maskedShAmt = and i32 %shAmt, 7
%shl = shl i32 %zx, %maskedShAmt
%negShAmt = sub i32 0, %shAmt
%maskedNegShAmt = and i32 %negShAmt, 7
%shr = lshr i32 %zx, %maskedNegShAmt
%rot = or i32 %shl, %shr
%r = trunc i32 %rot to i8
=>
%truncShAmt = trunc i32 %shAmt to i8
%maskedShAmt2 = and i8 %truncShAmt, 7
%shl2 = shl i8 %x, %maskedShAmt2
%negShAmt2 = sub i8 0, %truncShAmt
%maskedNegShAmt2 = and i8 %negShAmt2, 7
%shr2 = lshr i8 %x, %maskedNegShAmt2
%r = or i8 %shl2, %shr2
llvm-svn: 346713
Noticed via inspection. Appears to be largely innocious in practice, but slight code change could have resulted in either visit order dependent missed optimizations or infinite loops. May be a minor compile time problem today.
llvm-svn: 346698
Summary:
When the 3rd argument to these intrinsics is zero, lowering them
to shift instructions produces poison values, since we end up with
shift amounts equal to the number of bits in the shifted value. This
means we can only lower these intrinsics if we can prove that the
3rd argument is not zero.
Reviewers: arsenm
Reviewed By: arsenm
Subscribers: bnieuwenhuizen, jvesely, wdng, nhaehnle, llvm-commits
Differential Revision: https://reviews.llvm.org/D53739
llvm-svn: 346422
By morphing the instruction rather than deleting and creating a new one,
we retain fast-math-flags and potentially other metadata (profile info?).
llvm-svn: 346331
The sibling fold for 'oge' --> 'ord' was already here,
but this half was missing.
The result of fabs() must be positive or nan, so asking
if the result is negative or nan is the same as asking
if the result is nan.
This is another step towards fixing:
https://bugs.llvm.org/show_bug.cgi?id=39475
llvm-svn: 346321
As shown, this is used to eliminate redundant code in InstCombine,
and there are more cases where we should be using this pattern, but
we're currently unintentionally dropping flags.
llvm-svn: 346282
This is another part of solving PR39475:
https://bugs.llvm.org/show_bug.cgi?id=39475
This might be enough to fix that particular issue, but as noted
with the FIXME, we're still dropping FMF on other folds around here.
llvm-svn: 346234
This is NFCI for InstCombine because it calls InstSimplify,
so I left the tests for this transform there. As noted in
the code comment, we can allow this fold more often by using
FMF and/or value tracking.
llvm-svn: 346169
As stated in IEEE-754 and discussed in:
https://bugs.llvm.org/show_bug.cgi?id=38086
...the sign of zero does not affect any FP compare predicate.
Known regressions were fixed with:
rL346097 (D54001)
rL346143
The transform will help reduce pattern-matching complexity to solve:
https://bugs.llvm.org/show_bug.cgi?id=39475
...as well as improve CSE and codegen (a zero constant is almost always
easier to produce than 0x80..00).
llvm-svn: 346147
It looks like we correctly removed edge cases with 0.0 from D50714,
but we were a bit conservative because getBinOpIdentity() doesn't
distinguish between +0.0 and -0.0 and 'nsz' is effectively always
true for fcmp (see discussion in:
https://bugs.llvm.org/show_bug.cgi?id=38086
Without this change, we would get regressions by canonicalizing
to +0.0 in all fcmp, and that's a step towards solving:
https://bugs.llvm.org/show_bug.cgi?id=39475
llvm-svn: 346143
The 'OLT' case was updated at rL266175, so I assume it was just an
oversight that 'UGE' was not included because that patch handled
both predicates in InstSimplify.
llvm-svn: 345727
InstCombine features an optimization that essentially replaces:
if (a)
free(a)
into:
free(a)
Right now, this optimization is gated by the minsize attribute and therefore
we only perform it if we can prove that we are going to be able to eliminate
the branch and the destination block.
However when casts are involved the optimization would fail to apply, because
the optimization was not smart enough to realize that it is possible to also
move the casts away from the destination block and that is harmless to the
performance since they are just noops.
E.g.,
foo(int *a)
if (a)
free((char*)a)
Wouldn't be optimized by instcombine, because
- We would refuse to hoist the `bitcast i32* %a to i8` in the source block
- We would fail to see that `bitcast i32* %a to i8` and %a are the same value.
This patch fixes both these problems:
- It teaches the pattern matching of the comparison how to look
through casts.
- It checks that whether the additional instruction in the destination block
can be hoisted and are harmless performance-wise.
- It hoists all the code of the destination block in the source block.
Differential Revision: D53356
llvm-svn: 345644
shuffle (insert ?, Scalar, IndexC), V1, Mask --> insert V1, Scalar, IndexC'
The motivating case is at least a couple of steps away: I noticed that
SLPVectorizer does not analyze shuffles as well as sequences of
insert/extract in PR34724:
https://bugs.llvm.org/show_bug.cgi?id=34724
...so SLP may fail to vectorize when source code has shuffles to start
with or instcombine has converted insert/extract to shuffles.
Independent of that, an insertelement is always a simpler op for IR
analysis vs. a shuffle, so we should transform to insert when possible.
I don't think there's any codegen concern here - if a target can't insert
a scalar directly to some fixed element in a vector (x86?), then this
should get expanded to the insert+shuffle that we started with.
Differential Revision: https://reviews.llvm.org/D53507
llvm-svn: 345607
Replacing BinaryOperator::isFNeg(...) to avoid regressions when we
separate FNeg from the FSub IR instruction.
Differential Revision: https://reviews.llvm.org/D53650
llvm-svn: 345295
The original patch was committed here:
rL344609
...and reverted:
rL344612
...because it did not properly check/test data types before calling
ComputeNumSignBits().
The tests that caused bot failures for the previous commit are
over-reaching front-end tests that run the entire -O optimizer
pipeline:
Clang :: CodeGen/builtins-systemz-zvector.c
Clang :: CodeGen/builtins-systemz-zvector2.c
I've added a negative test here to ensure coverage for that case.
The new early exit check also tests the type of the 'B' parameter,
so we don't waste time on matching if either value is unsuitable.
Original commit message:
This is part of solving PR37549:
https://bugs.llvm.org/show_bug.cgi?id=37549
The patterns shown here are a special case of something
that we already convert to select. Using ComputeNumSignBits()
catches that case (but not the more complicated motivating
patterns yet).
The backend has hooks/logic to convert back to logic ops
if that's better for the target.
llvm-svn: 345149
There's probably some vector-with-undef-element pattern
that shows an improvement, so this is probably not quite
'NFC'.
This is the last step towards removing the fake binop
queries for not/neg. Ie, there are no more uses of those
functions in trunk. Fneg should follow.
llvm-svn: 345050
I couldn't tell from svn history when these checks were added,
but it pre-dates the split of instcombine into its own directory
at rL92459.
The motivation for changing the check is partly shown by the
code in PR34724:
https://bugs.llvm.org/show_bug.cgi?id=34724
There are also existing regression tests for SLPVectorizer with
sequences of extract+insert that are likely assumed to become
shuffles by the vectorizer cost models.
llvm-svn: 344854
Summary:
In several places in the code we use the following pattern:
if (hasUnaryFloatFn(&TLI, Ty, LibFunc_tan, LibFunc_tanf, LibFunc_tanl)) {
[...]
Value *Res = emitUnaryFloatFnCall(X, TLI.getName(LibFunc_tan), B, Attrs);
[...]
}
In short, we check if there is a lib-function for a certain type, and then
we _always_ fetch the name of the "double" version of the lib function and
construct a call to the appropriate function, that we just checked exists,
using that "double" name as a basis.
This is of course a problem in cases where the target doesn't support the
"double" version, but e.g. only the "float" version.
In that case TLI.getName(LibFunc_tan) returns "", and
emitUnaryFloatFnCall happily appends an "f" to "", and we erroneously end
up with a call to a function called "f".
To solve this, the above pattern is changed to
if (hasUnaryFloatFn(&TLI, Ty, LibFunc_tan, LibFunc_tanf, LibFunc_tanl)) {
[...]
Value *Res = emitUnaryFloatFnCall(X, &TLI, LibFunc_tan, LibFunc_tanf,
LibFunc_tanl, B, Attrs);
[...]
}
I.e instead of first fetching the name of the "double" version and then
letting emitUnaryFloatFnCall() add the final "f" or "l", we let
emitUnaryFloatFnCall() fetch the right name from TLI.
Reviewers: eli.friedman, efriedma
Reviewed By: efriedma
Subscribers: efriedma, bjope, llvm-commits
Differential Revision: https://reviews.llvm.org/D53370
llvm-svn: 344725
I noticed a missing check and added it at rL344610, but there actually
are codegen tests that will fail without that, so I'll edit those and
submit a fixed patch with more tests.
llvm-svn: 344612
This is part of solving PR37549:
https://bugs.llvm.org/show_bug.cgi?id=37549
The patterns shown here are a special case of something
that we already convert to select. Using ComputeNumSignBits()
catches that case (but not the more complicated motivating
patterns yet).
The backend has hooks/logic to convert back to logic ops
if that's better for the target.
llvm-svn: 344609
by `getTerminator()` calls instead be declared as `Instruction`.
This is the biggest remaining chunk of the usage of `getTerminator()`
that insists on the narrow type and so is an easy batch of updates.
Several files saw more extensive updates where this would cascade to
requiring API updates within the file to use `Instruction` instead of
`TerminatorInst`. All of these were trivial in nature (pervasively using
`Instruction` instead just worked).
llvm-svn: 344502
This is part of the missing IR-level folding noted in D52912.
This should be ok as a canonicalization because the new shuffle mask can't
be any more complicated than the existing shuffle mask. If there's some
target where the shorter vector shuffle is not legal, it should just end up
expanding to something like the pair of shuffles that we're starting with here.
Differential Revision: https://reviews.llvm.org/D53037
llvm-svn: 344476
InstCombine keeps a worklist and assumes that optimizations don't
eraseFromParent() the instruction, which SimplifyLibCalls violates. This change
adds a new callback to SimplifyLibCalls to let clients specify their own hander
for erasing actions.
Differential Revision: https://reviews.llvm.org/D52729
llvm-svn: 344251
This is the umin alternative to the umax code from rL344237. We use
DeMorgans law on the umax case to bring us to the same thing on umin,
but using countLeadingOnes, not countLeadingZeros.
Differential Revision: https://reviews.llvm.org/D53036
llvm-svn: 344239
Use the demanded bits of umax(A,C) to prove we can just use A so long as the
lowest non-zero bit of DemandMask is higher than the highest non-zero bit of C
Differential Revision: https://reviews.llvm.org/D53033
llvm-svn: 344237
The IRBuilder CreateIntrinsic method wouldn't allow you to specify the
types that you wanted the intrinsic to be mangled with. To fix this
I've:
- Added an ArrayRef<Type *> member to both CreateIntrinsic overloads.
- Used that array to pass into the Intrinsic::getDeclaration call.
- Added a CreateUnaryIntrinsic to replace the most common use of
CreateIntrinsic where the type was auto-deduced from operand 0.
- Added a bunch more unit tests to test Create*Intrinsic calls that
weren't being tested (including the FMF flag that wasn't checked).
This was suggested as part of the AMDGPU specific atomic optimizer
review (https://reviews.llvm.org/D51969).
Differential Revision: https://reviews.llvm.org/D52087
llvm-svn: 343962
Currently running the @insertelem_after_gep function below through the InstCombine pass with opt produces invalid IR.
Input:
```
define void @insertelem_after_gep(<16 x i32>* %t0) {
%t1 = bitcast <16 x i32>* %t0 to [16 x i32]*
%t2 = addrspacecast [16 x i32]* %t1 to [16 x i32] addrspace(3)*
%t3 = getelementptr inbounds [16 x i32], [16 x i32] addrspace(3)* %t2, i64 0, i64 0
%t4 = insertelement <16 x i32 addrspace(3)*> undef, i32 addrspace(3)* %t3, i32 0
call void @extern_vec_pointers_func(<16 x i32 addrspace(3)*> %t4)
ret void
}
```
Output:
```
define void @insertelem_after_gep(<16 x i32>* %t0) {
%t3 = getelementptr inbounds <16 x i32>, <16 x i32>* %t0, i64 0, i64 0
%t4 = insertelement <16 x i32 addrspace(3)*> undef, i32 addrspace(3)* %t3, i32 0
call void @my_extern_func(<16 x i32 addrspace(3)*> %t4)
ret void
}
```
Which although causes no complaints when produced, isn't valid IR as the insertelement use of the %t3 GEP expects an address space.
```
opt: /tmp/bad.ll:52:73: error: '%t3' defined with type 'i32*' but expected 'i32 addrspace(3)*'
%t4 = insertelement <16 x i32 addrspace(3)*> undef, i32 addrspace(3)* %t3, i32 0
```
I've fixed this by adding an addrspacecast after the GEP in the InstCombine pass, and including a check for this type mismatch to the verifier.
Reviewers: spatel, lebedev.ri
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D52294
llvm-svn: 343956
We established the (unfortunately complicated) rules for UB/poison
propagation with vector ops in:
D48893
D48987
D49047
It's clear from the affected tests that we are potentially creating
poison where none existed before the transforms. For add/sub/mul,
the answer is simple: just drop the flags because the extra undef
vector lanes are generally more valuable for analysis and codegen.
llvm-svn: 343819
This is a follow-up to rL343482 / D52439.
This was a pattern that initially caused the commit to be reverted because
the transform requires a bitcast as shown here.
llvm-svn: 343794
We're a long way from D50992 and D51553, but this is where we have to start.
We weren't back-propagating undefs into binop constant values for anything but
add/sub/mul/and/or/xor.
This is likely because we have to be careful about not introducing UB/poison
with div/rem/shift. But I suspect we already are getting the poison part wrong
for add/sub/mul (although it may not be possible to expose the bug currently
because we use SimplifyDemandedVectorElts from a limited set of opcodes).
See the discussion/implementation from D48987 and D49047.
This patch just enables functionality for FP ops because those do not have
UB/poison potential.
llvm-svn: 343727
1. Fix include ordering.
2. Improve variable name (width is bitwidth not number-of-elements).
3. Add local Opcode variable to reduce code duplication.
llvm-svn: 343694
This is an attempt to get out of a local-minimum that instcombine currently
gets stuck in. We essentially combine two optimisations at once, ~a - ~b = b-a
and min(~a, ~b) = ~max(a, b), only doing the transform if the result is at
least neutral. This involves using IsFreeToInvert, which has been expanded a
little to include selects that can be easily inverted.
This is trying to fix PR35875, using the ideas from Sanjay. It is a large
improvement to one of our rgb to cmy kernels.
Differential Revision: https://reviews.llvm.org/D52177
llvm-svn: 343569
Summary:
This is a continuation of the fix for PR34627 "InstCombine assertion at vector gep/icmp folding". (I just realized bugpoint had fuzzed the original test for me, so I had fixed another trigger of the same assert in adjacent code in InstCombine.)
This patch avoids optimizing an icmp (to look only at the base pointers) when the resulting icmp would have a different type.
The patch adds a testcase and also cleans up and shrinks the pre-existing test for the adjacent assert trigger.
Reviewers: lebedev.ri, majnemer, spatel
Reviewed By: lebedev.ri
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D52494
llvm-svn: 343486
This was originally committed at rL343407, but reverted at
rL343458 because it crashed trying to handle a case where
the destination type is FP. This version of the patch adds
a check for that possibility. Tests added at rL343480.
Original commit message:
This transform is requested for the backend in:
https://bugs.llvm.org/show_bug.cgi?id=39016
...but I figured it was worth doing in IR too, and it's probably
easier to implement here, so that's this patch.
In the simplest case, we are just truncating a scalar value. If the
extract index doesn't correspond to the LSBs of the scalar, then we
have to shift-right before the truncate. Endian-ness makes this tricky,
but hopefully the ASCII-art helps visualize the transform.
Differential Revision: https://reviews.llvm.org/D52439
llvm-svn: 343482
This caused Chromium builds to fail with "Illegal Trunc" assertion.
See https://crbug.com/890723 for repro.
> This transform is requested for the backend in:
> https://bugs.llvm.org/show_bug.cgi?id=39016
> ...but I figured it was worth doing in IR too, and it's probably
> easier to implement here, so that's this patch.
>
> In the simplest case, we are just truncating a scalar value. If the
> extract index doesn't correspond to the LSBs of the scalar, then we
> have to shift-right before the truncate. Endian-ness makes this tricky,
> but hopefully the ASCII-art helps visualize the transform.
>
> Differential Revision: https://reviews.llvm.org/D52439
llvm-svn: 343458
This transform is requested for the backend in:
https://bugs.llvm.org/show_bug.cgi?id=39016
...but I figured it was worth doing in IR too, and it's probably
easier to implement here, so that's this patch.
In the simplest case, we are just truncating a scalar value. If the
extract index doesn't correspond to the LSBs of the scalar, then we
have to shift-right before the truncate. Endian-ness makes this tricky,
but hopefully the ASCII-art helps visualize the transform.
Differential Revision: https://reviews.llvm.org/D52439
llvm-svn: 343407
As noted in post-commit comments for D52548, the limitation on
increasing vector length can be applied by opcode.
As a first step, this patch only allows insertelement to be
widened because that has no logical downsides for IR and has
little risk of pessimizing codegen.
This may cause PR39132 to go into hiding during a full compile,
but that bug is not fixed.
llvm-svn: 343406
InstCombine would propagate shufflevector insts that had wider output vectors onto
predecessors, which would sometimes push undef's onto the divisor of a div/rem and
result in bad codegen.
I've fixed this by just banning propagating shufflevector back if the result of
the shufflevector is wider than the input vectors.
Patch by: @sheredom (Neil Henning)
Differential Revision: https://reviews.llvm.org/D52548
llvm-svn: 343329
When C is not zero and infinites are not allowed (C / X) > 0 is a sign
test. Depending on the sign of C, the predicate must be swapped.
E.g.:
foo(double X) {
if ((-2.0 / X) <= 0) ...
}
=>
foo(double X) {
if (X >= 0) ...
}
Patch by: @marels (Martin Elshuber)
Differential Revision: https://reviews.llvm.org/D51942
llvm-svn: 343228
The motivating case from:
https://bugs.llvm.org/show_bug.cgi?id=33026
...has no shuffles now. This kind of pattern may occur during
vectorization when targets have lumpy ISAs like SSE/AVX.
llvm-svn: 342988
We can handle patterns where the elements have different
sizes, so refactoring ahead of trying to add another blob
within these clauses.
llvm-svn: 342918
'width' of a vector usually refers to the bit-width.
https://bugs.llvm.org/show_bug.cgi?id=39016
shows a case where we could extend this fold to handle
a case where the number of elements in the bitcasted
vector is not equal to the resulting value.
llvm-svn: 342902
Follow-up to rL342324 (D52059):
Missing optimizations with blendv are shown in:
https://bugs.llvm.org/show_bug.cgi?id=38814
This is an easier and more powerful solution than adding pattern matching for a few
special cases in the backend. The potential danger with this transform in IR is that
the condition value can get separated from the select, and the backend might not be
able to make a blendv out of it again.
llvm-svn: 342806
Summary: This restores the combine that was reverted in r341883. The infinite loop from the failing test no longer occurs due to changes from r342163.
Reviewers: spatel, dmgreen
Reviewed By: spatel
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D52070
llvm-svn: 342797
Summary:
Same as to D52146.
`((1 << y)+(-1))` is simply non-canoniacal version of `~(-1 << y)`: https://rise4fun.com/Alive/0vl
We can not canonicalize it due to the extra uses. But we can handle it here.
Reviewers: spatel, craig.topper, RKSimon
Reviewed By: spatel
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D52147
llvm-svn: 342547
Summary:
Two folds are happening here:
1. https://rise4fun.com/Alive/oaFX
2. And then `foldICmpWithHighBitMask()` (D52001): https://rise4fun.com/Alive/wsP4
This change doesn't just add the handling for eq/ne predicates,
it actually builds upon the previous `foldICmpWithLowBitMaskedVal()` work,
so **all** the 16 fold variants* are immediately supported.
I'm indeed only testing these two predicates.
I do not feel like re-proving all 16 folds*, because they were already proven
for the general case of constant with all-ones in low bits. So as long as
the mask produces all-ones in low bits, i'm pretty sure the fold is valid.
But required, i can re-prove, let me know.
* eq/ne are commutative - 4 folds; ult/ule/ugt/uge - are not commutative (the commuted variant is InstSimplified), 4 folds; slt/sle/sgt/sge are not commutative - 4 folds. 12 folds in total.
https://bugs.llvm.org/show_bug.cgi?id=38123https://bugs.llvm.org/show_bug.cgi?id=38708
Reviewers: spatel, craig.topper, RKSimon
Reviewed By: spatel
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D52146
llvm-svn: 342546
Summary:
If the sub doesn't overflow in the original type we can move it above the sext/zext.
This is similar to what we do for add. The overflow checking for sub is currently weaker than add, so the test cases are constructed for what is supported.
Reviewers: spatel
Reviewed By: spatel
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D52075
llvm-svn: 342335
Missing optimizations with blendv are shown in:
https://bugs.llvm.org/show_bug.cgi?id=38814
If this works, it's an easier and more powerful solution than adding pattern matching
for a few special cases in the backend. The potential danger with this transform in IR
is that the condition value can get separated from the select, and the backend might
not be able to make a blendv out of it again. I don't think that's too likely, but
I've kept this patch minimal with a 'TODO', so we can test that theory in the wild
before expanding the transform.
Differential Revision: https://reviews.llvm.org/D52059
llvm-svn: 342324
Similar to rL342278:
The test diffs are all cosmetic due to the change in
value naming, but I'm including that to show that the
new code does perform these folds rather than something
else in instcombine.
D52075 should be able to use this code too rather than
duplicating all of the logic.
llvm-svn: 342292
The test diffs are all cosmetic due to the change in
value naming, but I'm including that to show that the
new code does perform these folds rather than something
else in instcombine.
llvm-svn: 342278
Summary:
It is sometimes important to check that some newly-computed value
is non-negative and only n bits wide (where n is a variable.)
There are many ways to check that:
https://godbolt.org/z/o4RB8D
The last variant seems best?
(I'm sure there are some other variations i haven't thought of..)
More complicated, canonical pattern:
https://rise4fun.com/Alive/uhA
We do need to have two `switch()`'es like this,
to not mismatch the swappable predicates.
https://bugs.llvm.org/show_bug.cgi?id=38708
Reviewers: spatel, craig.topper, RKSimon
Reviewed By: spatel
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D52001
llvm-svn: 342173
This allows the xor to be removed completely.
This might help with recomitting r341674, but seems good regardless.
Coincidentally fixes PR38915.
Differential Revision: https://reviews.llvm.org/D51964
llvm-svn: 342163
I accidentally committed this diff with rL342147 because
I had applied D51964. We probably do need those checks,
but D51964 has tests and more discussion/motivation,
so they should be re-added with that patch.
llvm-svn: 342149
I don't have a test case for this, but it's motivated by
the discussion in D51964, and I've added TODO comments for
the better fix - move simplifications into instsimplify
because that's more efficient and reduces risk of infinite
loops in instcombine caused by transforms trying to do the
opposite folds.
In this case, we know that the transform that tries to move
'not' through min/max can be fooled by the multiple uses
of a value in another min/max, so try to squash the
foldSPFofSPF() patterns first.
llvm-svn: 342147
Summary:
It is sometimes important to check that some newly-computed value
is non-negative and only `n` bits wide (where `n` is a variable.)
There are **many** ways to check that:
https://godbolt.org/z/o4RB8D
The last variant seems best?
(I'm sure there are some other variations i haven't thought of..)
Let's handle the second variant first, since it is much simpler.
https://rise4fun.com/Alive/LYjYhttps://bugs.llvm.org/show_bug.cgi?id=38708
Reviewers: spatel, craig.topper, RKSimon
Reviewed By: spatel
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D51985
llvm-svn: 342067
Name: op_ugt_sum
%a = add i8 %x, %y
%r = icmp ugt i8 %x, %a
=>
%notx = xor i8 %x, -1
%r = icmp ugt i8 %y, %notx
Name: sum_ult_op
%a = add i8 %x, %y
%r = icmp ult i8 %a, %x
=>
%notx = xor i8 %x, -1
%r = icmp ugt i8 %y, %notx
https://rise4fun.com/Alive/ZRxI
AFAICT, this doesn't interfere with any add-saturation patterns
because those have >1 use for the 'add'. But this should be
better for IR analysis and codegen in the basic cases.
This is another fold inspired by PR14613:
https://bugs.llvm.org/show_bug.cgi?id=14613
llvm-svn: 342004
These are the folds in Alive;
Name: xor_ult
Pre: isPowerOf2(-C1)
%xor = xor i8 %x, C1
%r = icmp ult i8 %xor, C1
=>
%r = icmp ugt i8 %x, ~C1
Name: xor_ugt
Pre: isPowerOf2(C1+1)
%xor = xor i8 %x, C1
%r = icmp ugt i8 %xor, C1
=>
%r = icmp ugt i8 %x, C1
https://rise4fun.com/Alive/Vty
The ugt case in its simplest form was already handled by DemandedBits,
but that's not ideal as shown in the multi-use test.
I'm not sure if these are all of the symmetrical folds, but I adjusted
the existing code for one of the folds to try to show the similarities.
There's no obvious connection, but this is another preliminary step
for PR14613...
https://bugs.llvm.org/show_bug.cgi?id=14613
llvm-svn: 341997
I noticed that we were not back-propagating undef lanes to shuffle masks when we have a
shuffle that reduces the vector width. This is part of investigating/solving PR38691:
https://bugs.llvm.org/show_bug.cgi?id=38691
The DAG equivalent was proposed with:
D51696
Differential Revision: https://reviews.llvm.org/D51433
llvm-svn: 341981
For vectors, getPrimitiveSizeInBits returns the full vector width. This code should using the element size for vectors. This could be fixed by calling getScalarSizeInBits, but its even easier to just get it from the APInt we're checking.
Differential Revision: https://reviews.llvm.org/D51938
llvm-svn: 341971
Summary:
Revert min/max changes in rL341674 dues to high compile times causing timeouts (PR38897).
Checking in to unblock failing builds. Patch available for post-commit review and re-revert once resolved.
Working on a smaller reproducer for PR38897.
Reviewers: craig.topper, spatel
Subscribers: sanjoy, jlebar, llvm-commits
Differential Revision: https://reviews.llvm.org/D51897
llvm-svn: 341883
There were two combines not covered by the check before now, neither of which
actually differed from normal in the benefit analysis.
The most recent seems to be because it was just added at the top of the
function (naturally). The older is from way back in 2008 (r46687) when we just
didn't put those checks in so routinely, and has been diligently maintained
since.
llvm-svn: 341831
shuf (sel (shuf NarrowCond, undef, WideMask), X, Y), undef, NarrowMask) -->
sel NarrowCond, (shuf X, undef, NarrowMask), (shuf Y, undef, NarrowMask)
The motivating case from:
https://bugs.llvm.org/show_bug.cgi?id=38691
...is the last regression test. In that case, we're just left with the narrow select.
Note that if we do create new shuffles, they use the existing extraction identity mask,
so there's no danger that this transform creates arbitrary shuffles.
Differential Revision: https://reviews.llvm.org/D51496
llvm-svn: 341708
If the ~X wasn't able to simplify above the max/min, we might be able to simplify it by moving it below the max/min.
I had to modify the ~(min/max ~X, Y) transform to prevent getting stuck in a loop when we saw the new ~(max/min X, ~Y) before the ~Y had been folded away to remove the new not.
Differential Revision: https://reviews.llvm.org/D51398
llvm-svn: 341674
If OtherOpT or OtherOpF have scalar types and the condition is a vector,
we would create an invalid select.
Reviewers: spatel, john.brawn, mssimpso, craig.topper
Reviewed By: spatel
Differential Revision: https://reviews.llvm.org/D51781
llvm-svn: 341666
This fold is needed to avoid a regression when we try
to recommit rL300977.
We can't see the most basic win currently because
demanded bits changes the patterns:
https://rise4fun.com/Alive/plpp
llvm-svn: 341559
I'm preparing to add the same functionality both here and to the DAG
version of this code in D51696 / D51433, so try to make those cases
as similar as possible to avoid bugs.
llvm-svn: 341545
I'm probably missing some way to use m_Deferred to remove the code
duplication, but that can be a follow-up.
The improvement in demand_shrink_nsw.ll is an example of missing
the fold because the pattern matching was deficient. I didn't try
to follow the bits in that test, but Alive says it's correct:
https://rise4fun.com/Alive/ugc
llvm-svn: 341426
This is just a cleanup step. The TODO comments show
what is wrong with the 'and' version of the fold.
Fixing this should be part of recommitting:
rL300977
llvm-svn: 341405
The fold was implemented for the general case but use-limitation,
but the later constant version which didn't check uses was only
matching splat constants.
llvm-svn: 341292
This is no-outwardly-visible-change intended, so no test.
But the code is smaller and more efficient. The check for
a 'not' op is intended to avoid the expensive value tracking
call when it should not be necessary, and it might prevent
infinite looping when we resurrect:
rL300977
llvm-svn: 341280
This is a follow-up to rL339604 which did the same transform
for a sin libcall. The handling of intrinsics vs. libcalls
is unfortunately scattered, so I'm just adding this next to
the existing transform for llvm.cos for now.
This should resolve PR38458:
https://bugs.llvm.org/show_bug.cgi?id=38458
If the call was already negated, the negates will cancel
each other out.
llvm-svn: 340952
We were calling getNumUses to check for 1 or 2 uses. But getNumUses is linear in the number of uses. We can instead use !hasNUsesOrMore(3) which will stop the linear scan as soon as it determines there are at least 3 uses even if there are more.
llvm-svn: 340939
This lines up with the behavior of an existing transform where if both
operands of the binop are shuffled, we allow moving the binop before the
shuffle regardless of whether the shuffle changes the size of the vector.
llvm-svn: 340787
This is a bit awkward in a handful of places where we didn't even have
an instruction and now we have to see if we can build one. But on the
whole, this seems like a win and at worst a reasonable cost for removing
`TerminatorInst`.
All of this is part of the removal of `TerminatorInst` from the
`Instruction` type hierarchy.
llvm-svn: 340701
The core get and set routines move to the `Instruction` class. These
routines are only valid to call on instructions which are terminators.
The iterator and *generic* range based access move to `CFG.h` where all
the other generic successor and predecessor access lives. While moving
the iterator here, simplify it using the iterator utilities LLVM
provides and updates coding style as much as reasonable. The APIs remain
pointer-heavy when they could better use references, and retain the odd
behavior of `operator*` and `operator->` that is common in LLVM
iterators. Adjusting this API, if desired, should be a follow-up step.
Non-generic range iteration is added for the two instructions where
there is an especially easy mechanism and where there was code
attempting to use the range accessor from a specific subclass:
`indirectbr` and `br`. In both cases, the successors are contiguous
operands and can be easily iterated via the operand list.
This is the first major patch in removing the `TerminatorInst` type from
the IR's instruction type hierarchy. This change was discussed in an RFC
here and was pretty clearly positive:
http://lists.llvm.org/pipermail/llvm-dev/2018-May/123407.html
There will be a series of much more mechanical changes following this
one to complete this move.
Differential Revision: https://reviews.llvm.org/D47467
llvm-svn: 340698
This patch makes the DoesKMove argument non-optional, to force people
to think about it. Most cases where it is false are either code hoisting
or code sinking, where we pick one instruction from a set of
equal instructions among different code paths.
Reviewers: dberlin, nlopes, efriedma, davide
Reviewed By: efriedma
Differential Revision: https://reviews.llvm.org/D47475
llvm-svn: 340606
I'm assuming its easier to make sure the RHS of an XOR is all ones than it is to check for the many select patterns we have. So lets check that first. Same with the one use check.
llvm-svn: 340321
This commit fixes a (gcc 7.3.0) [-Wunused-function] warning caused by the
presence of unused method FaddCombine::createFDiv().
The last use of that method was removed at r339519.
llvm-svn: 340014