Constant hoisting may hide the constant value behind bitcast for And's
operand. Track down the constant to make the BFI result consistent
regardless of hoisting.
Differential Revision: https://reviews.llvm.org/D91450
aliasGEP() currently implements some special handling for the case
where all variable offsets are positive, in which case the constant
offset can be taken as the minimal offset. However, it does not
perform the same handling for the all-negative case. This means that
the alias-analysis result between two GEPs is asymmetric:
If GEP1 - GEP2 is all-positive, then GEP2 - GEP1 is all-negative,
and the first will result in NoAlias, while the second will result
in MayAlias.
Apart from producing sub-optimal results for one order, this also
violates our caching assumption. In particular, if BatchAA is used,
the cached result depends on the order of the GEPs in the first query.
This results in an inconsistency in BatchAA and AA results, which
is how I noticed this issue in the first place.
Differential Revision: https://reviews.llvm.org/D91383
This patch fixes the function isWideningInstruction for scalable vectors.
Now the cost model can check the widening pattern for SVE.
Differential Revision: https://reviews.llvm.org/D91260
In an effort to make code around flag determination more readable, and (possibly) prepare for a follow up change, factor out some of the flag detection logic. In the process, reduce the number of locations we mutate wrap flags by a couple.
Note that this isn't NFC. The old code tried for NSW xor (NUW || NW). This is, two different paths computed different sets of wrap flags. The new code will try for all three. The result is that some expressions end up with a few extra flags set.
This was changed recently with D90554 / f7eac51b9b
...because we had a regression testing blindspot for intrinsics
that are expected to be lowered to libcalls.
In general, we want the *size* cost for a scalar call to be cheap
even if the other costs are expensive - we expect it to just be
a branch with some optional stack manipulation.
It is likely that we will want to carve out some
exceptions/overrides to this rule as follow-up patches for
calls that have some general and/or target-specific difference
to the expected lowering.
This was noticed as a regression in unrolling, so we have a test
for that now along with a couple of direct cost model tests.
If the assumed scalarization costs for the oversized vector
calls are not realistic, that would be another follow-up
refinement of the cost models.
The SCEV code for constructing GEP expressions currently assumes
that the addition of the base and all the offsets is nsw if the GEP
is inbounds. While the addition of the offsets is indeed nsw, the
addition to the base address is not, as the base address is
interpreted as an unsigned value.
Fix the GEP expression code to not assume nsw for the base+offset
calculation. However, do assume nuw if we know that the offset is
non-negative. With this, we use the same behavior as the
construction of GEP addrecs does. (Modulo the fact that we
disregard SCEV unification, as the pre-existing FIXME points out).
Differential Revision: https://reviews.llvm.org/D90648
The GEP aliasing code currently checks for the GEP decomposition
limit being reached (i.e., we did not reach the "final" underlying
object). As far as I can see, these checks are not necessary. It is
perfectly fine to work with a GEP whose base can still be further
decomposed.
Looking back through the commit history, these checks were originally
introduced in 1a444489e9. However, I
believe that the problem this was intended to address was later
properly fixed with 1726fc698c, and
the checks are no longer necessary since then (and were not the
right fix in the first place).
Differential Revision: https://reviews.llvm.org/D91010
Summary:
Expand the print-memoryssa and print<memoryssa> passes with a new hidden
option -cfg-dot-mssa that names a file. When set, a dot-cfg style file
will be generated into the named file with the memoryssa comments retained
and those blocks containing them shown in light pink. The option does
nothing in isolation.
Author: Jamie Schmeiser <schmeise@ca.ibm.com>
Reviewed By: asbirlea (Alina Sbirlea), dblaikie (David Blaikie)
Differential Revision: https://reviews.llvm.org/D90638
Summary:
Expand the print-memoryssa and print<memoryssa> passes with a new hidden
option -cfg-dot-mssa that names a file. When set, a dot-cfg style file
will be generated into the named file with the memoryssa comments retained
and those blocks containing them shown in light pink. The option does
nothing in isolation.
Author: Jamie Schmeiser <schmeise@ca.ibm.com>
Reviewed By: asbirlea (Alina Sbirlea), dblaikie (David Blaikie)
Differential Revision: https://reviews.llvm.org/D90638
This is the last step in removing cost-kind as a consideration in the basic class model for intrinsics.
See D89461 for the start of that.
Subsequent commits dealt with each of the special-case intrinsics that had customization here in the
basic class. This should remove a barrier to retrying
D87188 (canonicalization to the abs intrinsic).
The ARM and x86 cost diffs seen here may be wrong because the target-specific overrides have their own
bugs, but we hope this is less wrong - if something has a significant throughput cost, then it should
have a significant size / blended cost too by default.
The only behavioral diff in current regression tests is shown in the x86 scatter-gather test (which is
misplaced or broken because it runs the entire -O3 pipeline) - we unrolled less, and we assume that is
a improvement.
Differential Revision: https://reviews.llvm.org/D90554
Our range computation methods benefit from no-wrap flags. But if the ranges
were first computed before the flags were set, the cached range will be too
pessimistic.
We need to drop cached ranges whenever we sharpen AddRec's no wrap flags.
Differential Revision: https://reviews.llvm.org/D89847
Reviewed By: fhahn
This is the cmp/sel sibling to D90692.
Again, the reasoning is: the throughput cost is number of instructions/uops,
so size/blended costs are identical except in special cases (for example,
fdiv or other known-expensive machine instructions or things like MVE that
may require cracking into >1 uops).
We need to check for a valid (non-null) condition type parameter because
SimplifyCFG may pass nullptr for that (and so we will crash multiple
regression tests without that check). I'm not sure if passing nullptr makes
sense, but other code in the cost model does appear to check if that param
is set or not.
Differential Revision: https://reviews.llvm.org/D90781
This is based on the same idea that I am using for the basic model implementation
and what I have partly already done for x86: throughput cost is number of
instructions/uops, so size/blended costs are identical except in special cases
(for example, fdiv or other known-expensive machine instructions or things like
MVE that may require cracking into >1 uop)).
Differential Revision: https://reviews.llvm.org/D90692
As noted in D90554, there's an opcode typo in using an easily
misused cost model API: getCmpSelInstrCost(). Beyond that, the
assumed sequence of ops is questionable, but that would be
another patch.
My guess is that the x86 test diffs show that we are probably
wrong both before and after this change, so there will be no
practical difference.
As an example, I tried this test which shows a cost of '7'
either way:
define <4 x i32> @sadd(<4 x i32> %va, <4 x i32> %vb) {
%V4I32 = call {<4 x i32>, <4 x i1>} @llvm.sadd.with.overflow.v4i32(<4 x i32> %va, <4 x i32> %vb)
%ov = extractvalue {<4 x i32>, <4 x i1>} %V4I32, 1
%r = extractvalue {<4 x i32>, <4 x i1>} %V4I32, 0
%z = select <4 x i1> %ov, <4 x i32> <i32 42, i32 42, i32 42, i32 42>, <4 x i32> %r
ret <4 x i32> %z
}
$ llc -o - sadd.ll -mattr=avx
vpaddd %xmm1, %xmm0, %xmm2
vpcmpgtd %xmm2, %xmm0, %xmm0
vpxor %xmm0, %xmm1, %xmm0
vblendvps %xmm0, LCPI0_0(%rip), %xmm2, %xmm0a
Differential Revision: https://reviews.llvm.org/D90681
This patch changes the intrinsics cost model to assume that by default
target intrinsics are cheap. This didn't seem to be the case for all
intrinsics, and is potentially an MVE problem due to our scalarization
overheads. Cheap seems to be a good default in general though.
Differential Revision: https://reviews.llvm.org/D90597
```
// The legacy PM CGPassManager discovers SCCs this way:
for function in the source order
tarjanSCC(function)
// While the new PM CGSCCPassManager does:
for function in the reversed source order [1]
discover a reference graph SCC
build call graph SCCs inside the reference graph SCC
```
In the common cases, reference graph ~= call graph, the new PM order is
undesired because for `a | b | c` (3 independent functions), the new PM will
process them in the reversed order: c, b, a. If `a <-> b <-> c`, we can see
that `-print-after-all` will report the sole SCC as `scc: (c, b, a)`.
This patch corrects the iteration order. The discovered SCC order will match
the legacy PM in the common cases.
For some tests (`Transforms/Inline/cgscc-*.ll` and
`unittests/Analysis/CGSCCPassManagerTest.cpp`), the behaviors are dependent on
the SCC discovery order and there are too many check lines for the particular
order. This patch simply reverses the function order to avoid changing too many
check lines.
Differential Revision: https://reviews.llvm.org/D90566
As noticed in D90554 ,
the AVX2 costs for 256-bit vectors did not include FMAXNUM entries,
so we fell back to AVX1 which assumes those ops will be split into
128-bit halves or something close to that.
Differential Revision: https://reviews.llvm.org/D90613
This reverts the revert commit 408c4408fa.
This version of the patch includes a fix for a crash caused by
treating ICmp/FCmp constant expressions as instructions.
Original message:
On some targets, like AArch64, vector selects can be efficiently lowered
if the vector condition is a compare with a supported predicate.
This patch adds a new argument to getCmpSelInstrCost, to indicate the
predicate of the feeding select condition. Note that it is not
sufficient to use the context instruction when querying the cost of a
vector select starting from a scalar one, because the condition of the
vector select could be composed of compares with different predicates.
This change greatly improves modeling the costs of certain
compare/select patterns on AArch64.
I am also planning on putting up patches to make use of the new argument in
SLPVectorizer & LV.
If the elt size is unknown due to it being a pointer, a comparison
against 0 will cause an assert. Make sure the elt size is large enough
before comparing and for the moment just return the scalar cost.
CallInst::updateProfWeight() creates branch_weights with i64 instead of i32.
To be more consistent everywhere and remove lots of casts from uint64_t
to uint32_t, use i64 for branch_weights.
Reviewed By: davidxl
Differential Revision: https://reviews.llvm.org/D88609
I'm assuming the standard size integer instructions for this end up as something like:
mulq %rsi
seto %al
And the 'mul' generally has reciprocal throughput of 1 on typical implementations
(higher latency, but that's not handled here).
The default costs may end up much higher than that, and that's what we see in the test diffs.
Vector types are left as a 'TODO'.
Differential Revision: https://reviews.llvm.org/D90431
On some targets, like AArch64, vector selects can be efficiently lowered
if the vector condition is a compare with a supported predicate.
This patch adds a new argument to getCmpSelInstrCost, to indicate the
predicate of the feeding select condition. Note that it is not
sufficient to use the context instruction when querying the cost of a
vector select starting from a scalar one, because the condition of the
vector select could be composed of compares with different predicates.
This change greatly improves modeling the costs of certain
compare/select patterns on AArch64.
I am also planning on putting up patches to make use of the new argument in
SLPVectorizer & LV.
Reviewed By: dmgreen, RKSimon
Differential Revision: https://reviews.llvm.org/D90070
If we've got an SCEVPtrToIntExpr(op), where op is not an SCEVUnknown,
we want to sink the SCEVPtrToIntExpr into an operand,
so that the operation is performed on integers,
and eventually we end up with just an `SCEVPtrToIntExpr(SCEVUnknown)`.
Reviewed By: mkazantsev
Differential Revision: https://reviews.llvm.org/D89692
And use it to model LLVM IR's `ptrtoint` cast.
This is essentially an alternative to D88806, but with no chance for
all the problems it caused due to having the cast as implicit there.
(see rG7ee6c402474a2f5fd21c403e7529f97f6362fdb3)
As we've established by now, there are at least two reasons why we want this:
* It will allow SCEV to actually model the `ptrtoint` casts
and their operands, instead of treating them as `SCEVUnknown`
* It should help with initial problem of PR46786 - this should eventually allow us
to not loose pointer-ness of an expression in more cases
As discussed in [[ https://bugs.llvm.org/show_bug.cgi?id=46786 | PR46786 ]], in principle,
we could just extend `SCEVUnknown` with a `is ptrtoint` cast, because `ScalarEvolution::getPtrToIntExpr()`
should sink the cast as far down into the expression as possible,
so in the end we should always end up with `SCEVPtrToIntExpr` of `SCEVUnknown`.
But i think that it isn't the best solution, because it doesn't really matter
from memory consumption side - there probably won't be *that* many `SCEVPtrToIntExpr`s
for it to matter, and it allows for much better discoverability.
Reviewed By: mkazantsev
Differential Revision: https://reviews.llvm.org/D89456
Completing the series of FIXME removals for special-case intrinsics:
50dfa19cc7f2c25c7079c963bde01501ea93d85d
This one looks quite different than the others. The size/blended
cost is still potentially very far off from the throughput cost,
but this is hopefully not worse on the whole. It looks like the
underlying costs for the expanded shift/logic have their own
cost-kind limitations. Also, we are not asking the target if
it has a legal funnel shift op, so we just assume that the
intrinsic gets expanded.
When we need to prove implication of expressions of different type width,
the default strategy is to widen everything to wider type and prove in this
type. This does not interact well with AddRecs with negative steps and
unsigned predicates: such AddRec will likely not have a `nuw` flag, and its
`zext` to wider type will not be an AddRec. In contraty, `trunc` of an AddRec
in some cases can easily be proved to be an `AddRec` too.
This patch introduces an alternative way to handling implications of different
type widths. If we can prove that wider type values actually fit in the narrow type,
we truncate them and prove the implication in narrow type.
The return was due to revert of underlying patch that this one depends on.
Unit test temporarily disabled because the required logic in SCEV is switched
off due to compile time reasons.
Differential Revision: https://reviews.llvm.org/D89548
vnot (xor -1) should be equivalent to the AArch64 specific AArch64ISD::NOT
node, but allow more folding thanks to all the target independent
optimizations. Specifically this allows select(icmp ne, x, y) to
become "cmeq; bsl y, x" as opposed to needing to convert the predicate
with "cmeq; mvn; bsl x, y"
Unfortunately there is a regression in a cmtst test, but the code it
selected from was already non-canonical, with instcombine preferring to
use an eq predicate instead. Plus the more common case of icmp ne is
improved.
Differential Revision: https://reviews.llvm.org/D90126
We can sharpen the range of a AddRec if we know that it does not
self-wrap and know the symbolic iteration count in the loop. If we can
evaluate the value of AddRec on the last iteration and prove that at least
one its intermediate value lies between start and end, then no-wrap flag
allows us to conclude that all of them also lie between start and end. So
the estimate of range can be improved to union of ranges of start and end.
Switched off by default, can be turned on by flag.
Differential Revision: https://reviews.llvm.org/D89381
Reviewed By: lebedev.ri, nikic
This was originally part of:
f2c25c7079
but that was reverted because there was an underlying bug in
processing the vector type of these intrinsics. That was
fixed with:
74ffc823ed
This is similar in spirit to 01ea93d85d (memcpy) except that
here the underlying caller assumptions were created for vectorizer
use (throughput) rather than other passes.
That meant targets could have an enormous throughput cost with no
corresponding size, latency, or blended cost increase.
Paraphrasing from the previous commits:
This may not make sense for some callers, but at least now the
costs will be consistently wrong instead of mysteriously wrong.
Targets should provide better overrides if the current modeling
is not accurate.
This patch is to add the support of the value tracking of the alignment assume bundle.
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D88669
CallInst::updateProfWeight() creates branch_weights with i64 instead of i32.
To be more consistent everywhere and remove lots of casts from uint64_t
to uint32_t, use i64 for branch_weights.
Reviewed By: davidxl
Differential Revision: https://reviews.llvm.org/D88609
In each 128-lane, if there is at least one index is demanded and not all
indices are demanded and this 128-lane is not the first 128-lane of the
legalized-vector, then this 128-lane needs a extracti128;
If in each 128-lane, there is at least one index is demanded, this 128-lane
needs a inserti128.
The following cases will help you build a better understanding:
Assume we insert several elements into a v8i32 vector in avx2,
Case#1: inserting into 1th index needs vpinsrd + inserti128
Case#2: inserting into 5th index needs extracti128 + vpinsrd +
inserti128
Case#3: inserting into 4,5,6,7 index needs 4*vpinsrd + inserti128.
Reviewed By: pengfei, RKSimon
Differential Revision: https://reviews.llvm.org/D89767
We do not need to use the implicit cast here. We can instead can rely on
a comparison between two TypeSize objects instead. This algorithm will
work fine with scalable vectors.
Reviewed By: DavidTruby
Differential Revision: https://reviews.llvm.org/D90146
The warning would fire when calling getGEPCost for analyzing the cost of
a GEP instruction. This would result in the use of the now deprecated
implicit cast of TypeSize to uint64_t through the overloaded operator.
This patch fixes the issue by using getKnownMinSize instead of the
implicit cast. This is possible because the code is already
scalable-vector aware. The semantic behaviour of the code is unchanged
by this patch.
Reviewed By: sdesmalen, fpetrogalli
Differential Revision: https://reviews.llvm.org/D89872
This allows using annotation in a much more contexts than it currently has.
especially when annotation with template or constexpr.
Reviewed By: aaron.ballman
Differential Revision: https://reviews.llvm.org/D88645
This is a modified 2nd try of 22d10b8ab4
(reverted by 1c8371692d because it managed
to expose an existing crashing bug that should be fixed by
74ffc823 ).
Original commit message:
This is similar in spirit to 01ea93d85d (memcpy) except that
here the underlying caller assumptions were created for vectorizer
use (throughput) rather than other passes.
That meant targets could have an enormous throughput cost with no
corresponding size, latency, or blended cost increase.
The ARM costs show a small difference between throughput and
size because there's an underlying difference in cmp/sel
costs that is also predicated on cost-kind.
Paraphrasing from the previous commits:
This may not make sense for some callers, but at least now the
costs will be consistently wrong instead of mysteriously wrong.
Targets should provide better overrides if the current modeling
is not accurate.
I'm not sure if/how this ever worked, but it must not be tested
currently because the basic tests added here were crashing as
noted in the post-review comments for 1c83716 (which reverted
another cost-model fix in 22d10b8ab4).
Same change as 0dda633317, but for
mul expressions. We want to first fold any constant operans and
then strengthen the nowrap flags, as we can compute more precise
flags at that point.
Establish parity with the handling of add expressions, by always
constant folding mul expression operands before checking the depth
limit (this is a non-recursive simplification). The code was already
unconditionally constant folding the case where all operands were
constants, but was not folding multiple constant operands together
if there were also non-constant operands.
This requires picking out a different demonstration for depth-based
folding differences in the limit-depth.ll test.
We should first try to constant fold the add expression and only
strengthen nowrap flags afterwards. This allows us to determine
stronger flags if e.g. only two operands are left after constant
folding (and thus "guaranteed no wrap region" code applies) or the
resulting operands are non-negative and thus nsw->nuw strengthening
applies.
This is similar in spirit to 01ea93d85d (memcpy) except that
here the underlying caller assumptions were created for vectorizer
use (throughput) rather than other passes.
That meant targets could have an enormous throughput cost with no
corresponding size, latency, or blended cost increase.
The ARM costs show a small difference between throughput and
size because there's an underlying difference in cmp/sel
costs that is also predicated on cost-kind.
Paraphrasing from the previous commits:
This may not make sense for some callers, but at least now the
costs will be consistently wrong instead of mysteriously wrong.
Targets should provide better overrides if the current modeling
is not accurate.
1. Throughput and codesize costs estimations was separated and updated.
2. Updated fdiv cost estimation for different cases.
3. Added scalarization processing for types that are treated as !isSimple() to
improve codesize estimation in getArithmeticInstrCost() and
getArithmeticInstrCost(). The code was borrowed from TCK_RecipThroughput path
of base implementation.
Next step is unify scalarization part in base class that is currently works for
TCK_RecipThroughput path only.
Reviewed By: rampitec
Differential Revision: https://reviews.llvm.org/D89973
This patch adds a specialized implementation of getIntrinsicInstrCost
and add initial cost-modeling for min/max vector intrinsics.
AArch64 NEON support umin/smin/umax/smax for vectors
<8 x i8>, <16 x i8>, <4 x i16>, <8 x i16>, <2 x i32> and <4 x i32>.
Notably, it does not support vectors with i64 elements.
This change by itself should have very little impact on codegen, but in
follow-up patches I plan to teach the vectorizers to consider using
those intrinsics on platforms where it is profitable, e.g. because there
is no general 'select'-like instruction.
The current cost returned should be better for throughput, latency and size.
Reviewed By: dmgreen
Differential Revision: https://reviews.llvm.org/D89953
Visited phi blocks only need to be added for the duration of the
recursive alias queries, they should not leak into following code.
Once again, while this also improves analysis precision, this is
mainly intended to clarify the applicability scope of VisitedPhiBBs.
We only need the VisitedPhiBBs to disambiguate comparisons of
values from two different loop iterations. If we're comparing
two phis from the same basic block in lock-step, the compared
values will always be on the same iteration.
While this also increases precision, this is mainly intended
to clarify the scope of VisitedPhiBBs.
This is similar in spirit to 01ea93d85d (memcpy) except that
here the underlying caller assumptions were created for vectorizer
use (throughput) rather than other passes.
That meant ARM could have an enormous throughput cost with no
corresponding size, latency, or blended cost increase. X86 has
the same throughput restriction as the basic implementation, so
it is still unchanged.
Paraphrasing from the previous commit:
This may not make sense for some callers, but at least now the
costs will be consistently wrong instead of mysteriously wrong.
Targets should provide better overrides if the current modeling
is not accurate.
The default implementation base returns TCC_Expensive (currently
set to '4'), so that explains the test diff. This probably does
not make sense for most callers, but at least now the costs will
be consistently wrong instead of mysteriously wrong.
The ARM target has an override that tries to model codegen expansion,
and that should likely be adapted for general usage.
This probably does not affect anything because the vectorizers are
the primary users of the throughput cost, but memcpy is not listed
as a trivially vectorizable intrinsic.
When we need to prove implication of expressions of different type width,
the default strategy is to widen everything to wider type and prove in this
type. This does not interact well with AddRecs with negative steps and
unsigned predicates: such AddRec will likely not have a `nuw` flag, and its
`zext` to wider type will not be an AddRec. In contraty, `trunc` of an AddRec
in some cases can easily be proved to be an `AddRec` too.
This patch introduces an alternative way to handling implications of different
type widths. If we can prove that wider type values actually fit in the narrow type,
we truncate them and prove the implication in narrow type.
Differential Revision: https://reviews.llvm.org/D89548
Reviewed By: fhahn
This reverts commit a10a64e7e3.
It broke polly/test/ScopInfo/NonAffine/non-affine-loop-condition-dependent-access_3.ll
The difference suggests that this may be a serious issue.
D70365 allows us to make attributes default. This is a follow up to
actually make nosync, nofree and willreturn default. The approach we
chose, for now, is to opt-in to default attributes to avoid introducing
problems to target specific intrinsics. Intrinsics with default
attributes can be created using `DefaultAttrsIntrinsic` class.
Fixed wrapping range case & proof methods reduced to constant range
checks to save compile time.
Differential Revision: https://reviews.llvm.org/D89381
It's not pretty, but probably better than modelling it
as an opaque SCEVUnknown, i guess.
It is relevant e.g. for the loop that was brought up in
https://bugs.llvm.org/show_bug.cgi?id=46786#c26
as an example of what we'd be able to better analyze
once SCEV handles `ptrtoint` (D89456).
But as it is evident, even if we deal with `ptrtoint` there,
we also fail to model such an `ashr`.
Also, modeling of mul-of-exact-shr/div could use improvement.
As per alive2:
https://alive2.llvm.org/ce/z/tnfZKd
```
define i8 @src(i8 %0) {
%2 = ashr exact i8 %0, 4
ret i8 %2
}
declare i8 @llvm.abs(i8, i1)
declare i8 @llvm.smin(i8, i8)
declare i8 @llvm.smax(i8, i8)
define i8 @tgt(i8 %x) {
%abs_x = call i8 @llvm.abs(i8 %x, i1 false)
%div = udiv exact i8 %abs_x, 16
%t0 = call i8 @llvm.smax(i8 %x, i8 -1)
%t1 = call i8 @llvm.smin(i8 %t0, i8 1)
%r = mul nsw i8 %div, %t1
ret i8 %r
}
```
Transformation seems to be correct!
This adds some basic costs for MVE reductions - currently just costing
the simple legal add vectors as a single MVE instruction. More complex
costing can be added in the future when the framework more readily
allows it.
Differential Revision: https://reviews.llvm.org/D88980
This adds a very basic cost for active_lane_mask under MVE - making the
assumption that they will be free and then apologizing for that in a
comment.
In reality they may either be free (by being nicely folded into a tail
predicated loop), cost the same as a VCTP or be expanded into vdup's,
adds and cmp's. It is difficult to detect the difference from a single
getIntrinsicInstrCost call, so makes the assumption that the vectorizer
is adding them, and only added them where it makes sense.
We may need to change this in the future to better model predicate costs
in the vectorizer, especially at -Os or non-tail predicated loops. The
vectorizer currently does not query the cost of these instructions but
that will change in the future and a zero cost there probably makes the
most sense at the moment.
Differential Revision: https://reviews.llvm.org/D88989
It was reverted because of negative compile time impact. In this version,
less powerful proof methods are used (non-recursive reasoning only), and
scope limited to constant End values to avoid explision of complex proofs.
Differential Revision: https://reviews.llvm.org/D89381
After investigation by @asbirlea, the issue that caused the
revert appears to be an issue in the original source, rather
than a problem with the compiler.
This patch enables MemorySSA DSE again.
This reverts commit 915310bf14.
We can sharpen the range of a AddRec if we know that it does not
self-wrap and know the symbolic iteration count in the loop. If we can
evaluate the value of AddRec on the last iteration and prove that at least
one its intermediate value lies between start and end, then no-wrap flag
allows us to conclude that all of them also lie between start and end. So
the estimate of range can be improved to union of ranges of start and end.
Differential Revision: https://reviews.llvm.org/D89381
Reviewed By: efriedma
The cost modeling for intrinsics is a patchwork based on different
expectations from the callers, so it's a mess. I'm hoping to untangle
this to allow canonicalization to the new min/max intrinsics in IR.
The general goal is to remove the cost-kind restriction here in the
basic implementation class. Ie, if some intrinsic has throughput cost
of 104, assume that it has the same size, latency, and blended costs.
Effectively, an intrinsic with cost N is composed of N simple
instructions. If that's not correct, the target should provide a more
accurate override.
The x86-64 SSE2 subtarget cost diffs require explanation:
1. The scalar ctlz/cttz are assuming "BSR+XOR+CMOV" or
"TEST+BSF+CMOV/BRANCH", so not cheap.
2. The 128-bit SSE vector width versions assume cost of 18 or 26
(no explanation provided in the tables, but this corresponds to a
bunch of shift/logic/compare).
3. The 512-bit vectors in the test file are scaled up by a factor of
4 from the legal vector width costs.
4. The plain latency cost-kind is not affected in this patch because
that calc is diverted before we get to getIntrinsicInstrCost().
Differential Revision: https://reviews.llvm.org/D89461
While we haven't encountered an earth-shattering problem with this yet,
by now it is pretty evident that trying to model the ptr->int cast
implicitly leads to having to update every single place that assumed
no such cast could be needed. That is of course the wrong approach.
Let's back this out, and re-attempt with some another approach,
possibly one originally suggested by Eli Friedman in
https://bugs.llvm.org/show_bug.cgi?id=46786#c20
which should hopefully spare us this pain and more.
This reverts commits 1fb6104293,
7324616660,
aaafe350bb,
e92a8e0c74.
I've kept&improved the tests though.
This is bigger/uglier than before, but it should allow fixing
all of the broken paths more easily. Test coverage added with
rGfab028b and other commits.
This is not NFC - the scalable vector test would crash
without this patch.
Testing for the various cost model "TargetCostKind" is limited,
and testing for scalable vectors is limited. The motivating
example of an intrinsic is not included here yet because that
just crashes.
Much similar to the ZExt/Trunc handling.
Thanks goes to Alexander Richardson for nudging towards noticing this one proactively.
The appropriate (currently crashing) test coverage added.
This relands commit 1c021c64ca which was
reverted in commit 17cec6a11a because
an assertion was being triggered, since `BuildConstantFromSCEV()`
wasn't updated to handle the case where the constant we want to truncate
is actually a pointer. I was unsuccessful in coming up with a test case
where we'd end there with constant zext/sext of a pointer,
so i didn't handle those cases there until there is a test case.
Original commit message:
While we indeed can't treat them as no-ops, i believe we can/should
do better than just modelling them as `unknown`. `inttoptr` story
is complicated, but for `ptrtoint`, it seems straight-forward
to model it just as a zext-or-trunc of unknown.
This may be important now that we track towards
making inttoptr/ptrtoint casts not no-op,
and towards preventing folding them into loads/etc
(see D88979/D88789/D88788)
Reviewed By: mkazantsev
Differential Revision: https://reviews.llvm.org/D88806
> While we indeed can't treat them as no-ops, i believe we can/should
> do better than just modelling them as `unknown`. `inttoptr` story
> is complicated, but for `ptrtoint`, it seems straight-forward
> to model it just as a zext-or-trunc of unknown.
>
> This may be important now that we track towards
> making inttoptr/ptrtoint casts not no-op,
> and towards preventing folding them into loads/etc
> (see D88979/D88789/D88788)
>
> Reviewed By: mkazantsev
>
> Differential Revision: https://reviews.llvm.org/D88806
It caused the following assert during Chromium builds:
llvm/lib/IR/Constants.cpp:1868:
static llvm::Constant *llvm::ConstantExpr::getTrunc(llvm::Constant *, llvm::Type *, bool):
Assertion `C->getType()->isIntOrIntVectorTy() && "Trunc operand must be integer"' failed.
See code review for a link to a reproducer.
This reverts commit 1c021c64ca.
While we indeed can't treat them as no-ops, i believe we can/should
do better than just modelling them as `unknown`. `inttoptr` story
is complicated, but for `ptrtoint`, it seems straight-forward
to model it just as a zext-or-trunc of unknown.
This may be important now that we track towards
making inttoptr/ptrtoint casts not no-op,
and towards preventing folding them into loads/etc
(see D88979/D88789/D88788)
Reviewed By: mkazantsev
Differential Revision: https://reviews.llvm.org/D88806
This is my first LLVM patch, so please tell me if there are any process issues.
The main observation for this patch is that we can lower UMIN/UMAX with v8i16 by using unsigned saturated subtractions in a clever way. Previously this operation was lowered by turning the signbit of both inputs and the output which turns the unsigned minimum/maximum into a signed one.
We could use this trick in reverse for lowering SMIN/SMAX with v16i8 instead. In terms of latency/throughput this is the needs one large move instruction. It's just that the sign bit turning has an increased chance of being optimized further. This is particularly apparent in the "reduce" test cases. However due to the slight regression in the single use case, this patch no longer proposes this.
Unfortunately this argument also applies in reverse to the new lowering of UMIN/UMAX with v8i16 which regresses the "horizontal-reduce-umax", "horizontal-reduce-umin", "vector-reduce-umin" and "vector-reduce-umax" test cases a bit with this patch. Maybe some extra casework would be possible to avoid this. However independent of that I believe that the benefits in the common case of just 1 to 3 chained min/max instructions outweighs the downsides in that specific case.
Patch By: @TomHender (Tom Hender) ActuallyaDeviloper
Differential Revision: https://reviews.llvm.org/D87236
This one is weird...
globals-aa needs to be already computed at licm, or else a function pass
can't run a module analysis and won't have access to globals-aa.
But the globals-aa result is impacted by instcombine in a way that
affects what the test is expecting. If globals-aa is computed before
instcombine, it is cached and globals-aa used in licm won't contain the
necessary info provided by instcombine.
Another catch is that if we don't invalidate AAManager, it will use the
cached AAManager that instcombine requested, which may not contain
globals-aa. So we have to invalidate<aa> so that licm can recompute
an AAManager with the globals-aa created by the require<globals-aa>.
This is essentially the problem described in https://reviews.llvm.org/D84259.
Reviewed By: asbirlea
Differential Revision: https://reviews.llvm.org/D88118
This patch achieves two things:
1. It breaks up the `join_blocks` interface between the SDA to the DA to
return two separate sets for divergent loops exits and divergent,
disjoint path joins.
2. It updates the SDA algorithm to run in O(n) time and improves the
precision on divergent loop exits.
This fixes `https://bugs.llvm.org/show_bug.cgi?id=46372` (by virtue of
the improved `join_blocks` interface) and revealed an imprecise expected
result in the `Analysis/DivergenceAnalysis/AMDGPU/hidden_loopdiverge.ll`
test.
Reviewed By: sameerds
Differential Revision: https://reviews.llvm.org/D84413
Similar to collecting information from branches guarding a loop, we can
also collect information from assumes dominating the loop header.
Fixes PR47247.
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D87854
There appears to be a mis-compile with MemorySSA-backed DSE in
combination with llvm.lifetime.end. It currently appears like
DSE is doing the right thing and the llvm.lifetime.end markers
are incorrect. The reverted patch uncovers the mis-compile.
This patch temporarily switches back to the legacy DSE
implementation, while we investigate.
This reverts commit 9d172c8e9c.
This introduces an analysis pass that wraps IRSimilarityIdentifier,
and adds a printer pass to examine in what function similarities are
being found.
Test for what the printer pass can find are in
test/Analysis/IRSimilarityIdentifier.
Reviewed by: paquette, jroelofs
Differential Revision: https://reviews.llvm.org/D86973
For some expressions, we can use information from loop guards when
we are looking for a maximum. This patch applies information from
loop guards to the expression used to compute the maximum backedge
taken count in howFarToZero. It currently replaces an unknown
expression X with UMin(X, Y), if the loop is guarded by
X ult Y.
This patch is minimal in what conditions it applies, and there
are a few TODOs to generalize.
This partly addresses PR40961. We will also need an update to
LV to address it completely.
Reviewed By: reames
Differential Revision: https://reviews.llvm.org/D67178
Better use isZero() and isIntN() in SystemZTargetTransformInfo rather than
calling getZExtValue() since the immediate operand may be wider than 64 bits,
which is not allowed with getZExtValue().
Fixes https://bugs.llvm.org/show_bug.cgi?id=47600
Review: Simon Pilgrim
-debug-pass is a legacy PM only option.
Some tests checks that the pass returned that it made a change,
which is not relevant to the NPM, since passes return PreservedAnalyses.
Some tests check that passes are freed at the proper time, which is also
not relevant to the NPM.
Reviewed By: asbirlea
Differential Revision: https://reviews.llvm.org/D87945
NPM passes just use the normal versions of these analyses instead.
Also pin any tests with -analyze to legacy PM.
Reviewed By: asbirlea
Differential Revision: https://reviews.llvm.org/D87857
This commit was originally because it was suspected to cause a crash,
but a reproducer did not surface.
A crash that was exposed by this change was fixed in 1d8f2e5292.
This reverts the revert commit 0581c0b0ee.
LSR claims to MemorySSA, but we also have to make sure it is preserved
when splitting critical edges. This can be done by passing MSSAU to
SplitCriticalEdge.
Fixes PR47557.
This switches to using DSE + MemorySSA by default again, after
fixing the issues reported after the first commit.
Notable fixes fc82006331, a0017c2bc2.
This reverts commit 3a59628f3c.
I think we need to be even more conservative when traversing memory
phis, to make sure we catch any loop carried dependences.
This approach updates fillInCurrentPair to use unknown sizes for
locations when we walk over a phi, unless the location is guaranteed to
be loop-invariant for any possible loop. Using an unknown size for
locations should ensure we catch all memory accesses to locations after
the given memory location, which includes loop-carried dependences.
Reviewed By: asbirlea
Differential Revision: https://reviews.llvm.org/D87778
This adds test cases for PR40961 and PR47247. They illustrate cases in
which the max backedge-taken count can be improved by information from
the loop guards.
As @efriedma pointed out in D86301, this "not equal to 0 check" of
get.active.lane.mask's second operand needs to live here in Lint and not the
Verifier.
Differential Revision: https://reviews.llvm.org/D87228
Really it should be named print<alias-sets>, but for the sake of
changing fewer tests, added a TODO to rename after NPM switch and test
cleanup.
Reviewed By: ychen
Differential Revision: https://reviews.llvm.org/D87713
'require<globals-aa>' is needed to make globals-aa work in NPM, since
globals-aa is a module analysis but function passes cannot run module
analyses on demand.
So don't skip translating alias analyses to 'require<>'.
Reviewed By: asbirlea
Differential Revision: https://reviews.llvm.org/D87743
Was missing MODULE_ALIAS_ANALYSIS, previously only FUNCTION_ALIAS_ANALYSIS was taken into account.
Reviewed By: asbirlea
Differential Revision: https://reviews.llvm.org/D87664
Pin RUN lines with -analyze to legacy PM, add corresponding NPM RUN line if missing.
Reviewed By: asbirlea
Differential Revision: https://reviews.llvm.org/D87658
1ce82015f6 added a fix to restrict phi optimizations after phi
translations. But the current use of performedPhiTranslation only
checked whether phi translation happened for the first iterator and
missed cases where phi translations happens at subsequent
iterators/upwards defs.
This patch changes upward_defs_iteartor to take a pointer to a bool, so
we can easily ensure the final value includes all visited defs, while
still being able to conveniently use it with make_range & co.
The tests have been updated and I plan to move them from the MSSA
directory up.
Some end-to-end tests needed small adjustments. One difference to the
legacy DSE is that legacy DSE also deletes trivially dead instructions
that are unrelated to memory operations. Because MemorySSA-backed DSE
just walks the MemorySSA, we only visit/check memory instructions. But
removing unrelated dead instructions is not really DSE's job and other
passes will clean up.
One noteworthy change is in llvm/test/Transforms/Coroutines/ArgAddr.ll,
but I think this comes down to legacy DSE not handling instructions that
may throw correctly in that case. To cover this with MemorySSA-backed
DSE, we need an update to llvm.coro.begin to treat it's return value to
belong to the same underlying object as the passed pointer.
There are some minor cases MemorySSA-backed DSE currently misses, e.g. related
to atomic operations, but I think those can be implemented after the switch.
This has been discussed on llvm-dev:
http://lists.llvm.org/pipermail/llvm-dev/2020-August/144417.html
For the MultiSource/SPEC2000/SPEC2006 the number of eliminated stores
goes from ~17500 (legayc DSE) to ~26300 (MemorySSA-backed). More numbers
and details in the thread on llvm-dev.
Impact on CTMark:
```
Legacy Pass Manager
exec instrs size-text
O3 + 0.60% - 0.27%
ReleaseThinLTO + 1.00% - 0.42%
ReleaseLTO-g. + 0.77% - 0.33%
RelThinLTO (link only) + 0.87% - 0.42%
RelLO-g (link only) + 0.78% - 0.33%
```
http://llvm-compile-time-tracker.com/compare.php?from=3f22e96d95c71ded906c67067d75278efb0a2525&to=ae8be4642533ff03803967ee9d7017c0d73b0ee0&stat=instructions
```
New Pass Manager
exec instrs. size-text
O3 + 0.95% - 0.25%
ReleaseThinLTO + 1.34% - 0.41%
ReleaseLTO-g. + 1.71% - 0.35%
RelThinLTO (link only) + 0.96% - 0.41%
RelLO-g (link only) + 2.21% - 0.35%
```
http://195.201.131.214:8000/compare.php?from=3f22e96d95c71ded906c67067d75278efb0a2525&to=ae8be4642533ff03803967ee9d7017c0d73b0ee0&stat=instructions
Reviewed By: asbirlea, xbolva00, nikic
Differential Revision: https://reviews.llvm.org/D87163
Other types can be handled in future patches but their uniform / non-uniform costs are more similar and don't appear to cause many vectorization issues.
When optimising for size, make the cost of i1 logical operations
relatively expensive so that optimisations don't try to combine
predicates.
Differential Revision: https://reviews.llvm.org/D86525
This adjusts the description of `llvm.memcpy` to also allow operands
to be equal. This is in line with what Clang currently expects.
This change is intended to be temporary and followed by re-introduce
a variant with the non-overlapping guarantee for cases where we can
actually ensure that property in the front-end.
See the links below for more details:
http://lists.llvm.org/pipermail/cfe-dev/2020-August/066614.html
and PR11763.
Reviewed By: nikic
Differential Revision: https://reviews.llvm.org/D86815
Recognize umin/umax/smin/smax intrinsics and convert them to the
already existing SCEV nodes of the same name.
In the future we'll want SCEVExpander to also produce the intrinsics,
but we're not ready for that yet.
Differential Revision: https://reviews.llvm.org/D87160
As discussed in D86843, -earlycse-debug-hash should be used in more regression
tests to catch inconsistency between the hashing and the equivalence check.
Differential Revision: https://reviews.llvm.org/D86863
Summary:
Analyses are preserved in MemCpyOptimizer.
Get analyses before running the pass and store the pointers, instead of
using lambdas and getting them every time on demand.
Reviewers: lenary, deadalnix, mehdi_amini, nikic, efriedma
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D74494
MemoryPhis with a single value are correct, but can lead to errors when
updating. Clean up single entry Phis newly added when cloning blocks.
Resolves PR46574.
This patch adds support for memcmp in MemoryLocation::getForArgument.
memcmp reads from the first 2 arguments up to the number of bytes of the
third argument.
Reviewed By: efriedma
Differential Revision: https://reviews.llvm.org/D86725
As discussed in
http://lists.llvm.org/pipermail/llvm-dev/2020-July/143801.html.
Currently no users outside of unit tests.
Replace all instances in tests of -constprop with -instsimplify.
Notable changes in tests:
* vscale.ll - @llvm.sadd.sat.nxv16i8 is evaluated by instsimplify, use a fake intrinsic instead
* InsertElement.ll - insertelement undef is removed by instsimplify in @insertelement_undef
llvm/test/Transforms/ConstProp moved to llvm/test/Transforms/InstSimplify/ConstProp
Reviewed By: lattner, nikic
Differential Revision: https://reviews.llvm.org/D85159
InstSimplify should do all transformations that ConstProp does, but
one thing that ConstProp does that InstSimplify wouldn't is inline
vector instructions that are constants, e.g. into a ret.
Previously vector instructions wouldn't be inlined in InstSimplify
because llvm::Simplify*Instruction() would return nullptr for specific
instructions, such as vector instructions that were actually constants,
if it couldn't simplify them.
This changes SimplifyInsertElementInst, SimplifyExtractElementInst, and
SimplifyShuffleVectorInst to return a vector constant when possible.
Reviewed By: efriedma
Differential Revision: https://reviews.llvm.org/D85946
MVE Gather scatter codegeneration is looking a lot better than it used
to, but still has some issues. The instructions we currently model as 1
cycle per element, which is a bit low for some cases. Increasing the
cost by the MVECostFactor brings them in-line with our other instruction
costs. This will have the effect of only generating then when the extra
benefit is more likely to overcome some of the issues. Notably in
running out of registers and vectorizing loops that could otherwise be
SLP vectorized.
In the short-term whilst we look at other ways of dealing with those
more directly, we can increase the costs of gathers to make them more
likely to be beneficial when created.
Differential Revision: https://reviews.llvm.org/D86444
Summary: The LCSSA pass (required for all loop passes) sometimes adds
additional blocks containing LCSSA variables, and checkLoopsStructure
may return false even when the loops are perfectly nested in this case.
This is because the successor of the exit block of the inner loop now
points to the LCSSA block instead of the latch block of the outer loop.
Examples are shown in the test nests-with-lcssa.ll.
To fix the issue, the successor of the exit block of the inner loop can
now point to a block in which all instructions are LCSSA phi node
(except the terminator), and the sole successor of that block should
point to the latch block of the outer loop.
Reviewed By: Whitney, etiotto
Differential Revision: https://reviews.llvm.org/D86133
In getCastInstrCost when the instruction is a truncate we were relying
upon the implicit TypeSize -> uint64_t cast when asking if a given type
has the same size as a legal integer. I've changed the code to only
ask the question if the type is fixed length.
I have also changed InstCombinerImpl::SimplifyDemandedUseBits to bail
out for now if the type is a scalable vector.
I've added the following new tests:
Analysis/CostModel/AArch64/sve-trunc.ll
Transforms/InstCombine/AArch64/sve-trunc.ll
for both of these fixes.
Differential revision: https://reviews.llvm.org/D86432
test referenced a relative path to a file, but the path was not correct
relative to the project the test is in
Differential Revision: https://reviews.llvm.org/D86368
Modify the ARM getCmpSelInstrCost implementation for the code size
costs of selects. Now consider the legalization cost and increase
the cost of i1 because those values wouldn't live in a general purpose
register. We also make selects +1 more expensive to account for the IT
instruction.
Differential Revision: https://reviews.llvm.org/D82091
StackLifetime class collects lifetime marker of an `alloca` by collect
the user of `BitCast` who is the user of the `alloca`. However, either
the `alloca` itself could be used with the lifetime marker or the `BitCast`
of the `alloca` could be transformed to other instructions. (e.g.,
it may be transformed to all zero reps in `InstCombine` pass).
This patch tries to fix this process in `collectMarkers` functions.
Reviewed By: vitalybuka
Differential Revision: https://reviews.llvm.org/D85399
The current demand propagator for addition will mark all input bits at and right of the alive output bit as alive. But carry won't propagate beyond a bit for which both operands are zero (or one/zero in the case of subtraction) so a more accurate answer is possible given known bits.
I derived a propagator by working through truth tables and using a bit-reversed addition to make demand ripple to the right, but I'm not sure how to make a convincing argument for its correctness in the comments yet. Nevertheless, here's a minimal implementation and test to get feedback.
This would help in a situation where, for example, four bytes (<128) packed into an int are added with four others SIMD-style but only one of the four results is actually read.
Known A: 0_______0_______0_______0_______
Known B: 0_______0_______0_______0_______
AOut: 00000000001000000000000000000000
AB, current: 00000000001111111111111111111111
AB, patch: 00000000001111111000000000000000
Committed on behalf of: @rrika (Erika)
Differential Revision: https://reviews.llvm.org/D72423
If we can't identify alloca used in lifetime marker we
need to assume to worst case scenario.
Reviewed By: eugenis
Differential Revision: https://reviews.llvm.org/D84630
Similarly as for pointers, even for integers a == b is usually false.
GCC also uses this heuristic.
Reviewed By: ebrevnov
Differential Revision: https://reviews.llvm.org/D85781
Similarly as for pointers, even for integers a == b is usually false.
GCC also uses this heuristic.
Reviewed By: ebrevnov
Differential Revision: https://reviews.llvm.org/D85781
Similarly as for pointers, even for integers a == b is usually false.
GCC also uses this heuristic.
Reviewed By: ebrevnov
Differential Revision: https://reviews.llvm.org/D85781
This reverts commit e441b7a7a0.
This patch causes a compile error in tensorflow opensource project. The stack trace looks like:
Point of crash:
llvm/include/llvm/Analysis/LoopInfoImpl.h : line 35
(gdb) ptype *this
type = const class llvm::LoopBase<llvm::BasicBlock, llvm::Loop> [with BlockT = llvm::BasicBlock, LoopT = llvm::Loop]
(gdb) p *this
$1 = {ParentLoop = 0x0, SubLoops = std::vector of length 0, capacity 0, Blocks = std::vector of length 0, capacity 1,
DenseBlockSet = {<llvm::SmallPtrSetImpl<llvm::BasicBlock const*>> = {<llvm::SmallPtrSetImplBase> = {<llvm::DebugEpochBase> = {Epoch = 3}, SmallArray = 0x1b2bf6c8, CurArray = 0x1b2bf6c8,
CurArraySize = 8, NumNonEmpty = 0, NumTombstones = 0}, <No data fields>}, SmallStorage = {0xfffffffffffffffe, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0}}, IsInvalid = true}
(gdb) p *this->DenseBlockSet->CurArray
$2 = (const void *) 0xfffffffffffffffe
I will try to get a case from tensorflow or use creduce to get a small case.
Now that SCEVExpander can preserve LCSSA form,
we do not have to worry about LCSSA form when
trying to look through PHIs. SCEVExpander will take
care of inserting LCSSA PHI nodes as required.
This increases precision of the analysis in some cases.
Reviewed By: mkazantsev, bmahjour
Differential Revision: https://reviews.llvm.org/D71539
This is the max version of D85046.
This change causes binary changes in 44 out of 237 benchmarks (out of
MultiSource/SPEC2000/SPEC2006)
Reviewed By: lebedev.ri
Differential Revision: https://reviews.llvm.org/D85189
Constant fold both the trapping and saturating versions of the
WebAssembly truncation intrinsics. The tests are adapted from the
WebAssembly spec tests for the corresponding instructions.
Requested in PR46982.
Differential Revision: https://reviews.llvm.org/D85392
If we can't identify alloca used in lifetime marker we
need to assume to worst case scenario.
Reviewed By: eugenis
Differential Revision: https://reviews.llvm.org/D84630
Add cases of fused fmul+fadd/fsub with f16 and f64 operands to cost model.
Also added operations with contract attribute.
Fixed line endings in test.
Reviewed By: rampitec
Differential Revision: https://reviews.llvm.org/D84995
As mentioned in
http://lists.llvm.org/pipermail/llvm-dev/2020-July/143395.html,
loop-unswitch has not been ported to the NPM. Instead people are using
simple-loop-unswitch.
Pin all tests in Transforms/LoopUnswitch to legacy PM and replace all
other uses of loop-unswitch with simple-loop-unswitch.
One test that didn't fit into the above was
2014-06-21-congruent-constant.ll which seems to only pass with
loop-unswitch. That is also pinned to legacy PM.
Now all tests containing "-loop-unswitch" anywhere in the test succeed with
NPM turned on by default.
Reviewed By: ychen
Differential Revision: https://reviews.llvm.org/D85360
-print-memoryssa in legacy PM is print<memoryssa> in NPM.
Pin tests with -print-memoryssa to legacy PM.
Add corresponding tests for NPM where missing.
This fixes "unknown pass name 'print-memoryssa'".
Some tests still fail in Analysis/MemorySSA due to other passes that
haven't been ported.
pr43427.ll and pr43438.ll required adding -aa-pipeline=basic-aa,
-loop-simplify (since it doesn't run on legacy PM by default), and
decrementing some of the MemoryPhi numbers.
Reviewed By: asbirlea
Differential Revision: https://reviews.llvm.org/D85333
As with other targets, set the throughput cost of control-flow
instructions to free so that we don't miss out of vectorization
opportunities.
Differential Revision: https://reviews.llvm.org/D85283
This option was added a while back, to help improve AA around pointer
phi loops. It looks for phi(gep(phi, const), x) loops, checking if x can
then prove more precise aliasing info.
Differential Revision: https://reviews.llvm.org/D82998
Wei Wang