This is a pretty rare case, when CxtI and assume are
in the same basic block, with assume being located later.
We were already checking that assumption was guaranteed to be
executed, but we omitted CxtI itself from consideration,
and as the test (miscompile) shows, that is incorrect.
As noted in D71660 review by @nikic.
@escape() may throw here, we don't know that assumption, which is located
afterwards in the same block, is executed, therefore %load arg of
call to @escape() can not be marked as non-null.
As noted in D71660 review by @nikic.
A function marked `noreturn` may contain unreachable terminators: these
should not be considered cold, as the function may be a trampoline.
rdar://58068594
Summary:
Ignore looking at blocks that are unreachable from entry when
collecting candidates for hosting.
Normally the consthoist pass is executed in the llc pipeline,
just after unreachableblockelim. So it is abnormal to have code
that is unreachable from the entry block. But when running the
pass as part of opt, for example as part of fuzzy testing, we
might trigger various kinds of asserts when collecting candidates
if we include unreachable blocks in that analysis.
It seems like a waste of time to hoist constants in unreachble
blocks, so the solution is to simply ignore such blocks when
collecting the hoisting candidates.
The two added test cases used to end up in two different asserts,
and the intention with the checks is just to verify that we no
longer fail.
Fixes: PR43903
Reviewers: spatel
Reviewed By: spatel
Subscribers: hiraditya, uabelho, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D71678
In certain situations after inlining and simplification we end up with
code that is _almost_ a min/max pattern, but contains constants that
have been demand-bit optimised to the wrong values, ending up with code
like:
%1 = icmp slt i32 %shr, -128
%2 = select i1 %1, i32 128, i32 %shr
%.inv = icmp sgt i32 %shr, 127
%spec.select.i = select i1 %.inv, i32 127, i32 %2
%conv7 = trunc i32 %spec.select.i to i8
This should be turned into a min/max pattern, but the -128 in the first
select was instead transformed into 128, as only the bottom byte was
ever demanded.
To fix this, I've put in further canonicalisation for the immediates of
selects, preferring to use the same value as the icmp if available.
Differential Revision: https://reviews.llvm.org/D71516
Summary:
This patch adds instructions to the InstCombine worklist after they are properly inserted. This way we don't get `<badref>`s printed when logging added instructions.
It also adds a check in `Worklist::Add` that ensures that all added instructions have parents.
Simple test case that illustrates the difference when run with `--debug-only=instcombine`:
```
define i32 @test35(i32 %a, i32 %b) {
%1 = or i32 %a, 1135
%2 = or i32 %1, %b
ret i32 %2
}
```
Before this patch:
```
INSTCOMBINE ITERATION #1 on test35
IC: ADDING: 3 instrs to worklist
IC: Visiting: %1 = or i32 %a, 1135
IC: Visiting: %2 = or i32 %1, %b
IC: ADD: %2 = or i32 %a, %b
IC: Old = %3 = or i32 %1, %b
New = <badref> = or i32 %2, 1135
IC: ADD: <badref> = or i32 %2, 1135
...
```
With this patch:
```
INSTCOMBINE ITERATION #1 on test35
IC: ADDING: 3 instrs to worklist
IC: Visiting: %1 = or i32 %a, 1135
IC: Visiting: %2 = or i32 %1, %b
IC: ADD: %2 = or i32 %a, %b
IC: Old = %3 = or i32 %1, %b
New = <badref> = or i32 %2, 1135
IC: ADD: %3 = or i32 %2, 1135
...
```
Reviewers: fhahn, davide, spatel, foad, grosser, nikic
Reviewed By: nikic
Subscribers: nikic, lebedev.ri, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D71093
Summary:
This patch teaches InstCombine to accept a new parameter: maximum number of iterations over functions.
InstCombine tries to simplify instructions by iterating over the whole function until the function stops changing. As a consequence, the last iteration before reaching a fixpoint visits all instructions in the worklist and never performs any rewrites.
Bounding the number of iterations can have 2 benefits:
* In case the users of the pass can make a good guess about the number of required iterations, we can save the time normally spent on the last iteration that doesn't change anything.
* When the wants to use InstCombine as a cleanup pass, it may be enough to run just a few iterations and stop even before reaching a fixpoint. This can be also useful for implementing a lightweight pass pipeline (think `-O1`).
This patch does not change the behavior of opt or Clang -- limiting the number of iterations is entirely opt-in.
Reviewers: fhahn, davide, spatel, foad, nlopes, grosser, lebedev.ri, nikic, xbolva00
Reviewed By: spatel
Subscribers: craig.topper, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D71145
This reverts commit 1f3dd83cc1, reapplying
commit bb1b0bc4e5.
The original commit failed on some builds seemingly due to the use of a
bracketed constructor with an std::array, i.e. `std::array<> arr({...})`.
Previously, LLVM had no functional way of performing casts inside of a
DIExpression(), which made salvaging cast instructions other than Noop
casts impossible. This patch enables the salvaging of casts by using the
DW_OP_LLVM_convert operator for SExt and Trunc instructions.
There is another issue which is exposed by this fix, in which fragment
DIExpressions (which are preserved more readily by this patch) for
values that must be split across registers in ISel trigger an assertion,
as the 'split' fragments extend beyond the bounds of the fragment
DIExpression causing an error. This patch also fixes this issue by
checking the fragment status of DIExpressions which are to be split, and
dropping fragments that are invalid.
We somehow missed doing this when we were working on Power9 exploitation.
This just adds the missing legalization and cost for producing the vector
intrinsics.
Differential revision: https://reviews.llvm.org/D70436
Summary:This PR move instructions from FC0.Latch bottom up to the
beginning of FC1.Latch as long as they are proven safe.
To illustrate why this is beneficial, let's consider the following
example:
Before Fusion:
header1:
br header2
header2:
br header2, latch1
latch1:
br header1, preheader3
preheader3:
br header3
header3:
br header4
header4:
br header4, latch3
latch3:
br header3, exit3
After Fusion (before this PR):
header1:
br header2
header2:
br header2, latch1
latch1:
br header3
header3:
br header4
header4:
br header4, latch3
latch3:
br header1, exit3
Note that preheader3 is removed during fusion before this PR.
Notice that we cannot fuse loop2 with loop4 as there exists block latch1
in between.
This PR move instructions from latch1 to beginning of latch3, and remove
block latch1. LoopFusion is now able to fuse loop nest recursively.
After Fusion (after this PR):
header1:
br header2
header2:
br header3
header3:
br header4
header4:
br header2, latch3
latch3:
br header1, exit3
Reviewer: kbarton, jdoerfert, Meinersbur, dmgreen, fhahn, hfinkel,
bmahjour, etiotto
Reviewed By: kbarton, Meinersbur
Subscribers: hiraditya, llvm-commits
Tag: LLVM
Differential Revision: https://reviews.llvm.org/D71165
Summary:
Add trimming of unused components of s_buffer_load.
Extend trimming of *buffer_load to also include
unused components at the beginning of vectors and update offset.
Subscribers: kzhuravl, jvesely, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D70315
Summary:
This patch restricts loop fusion to only consider rotated loops as valid candidates.
This simplifies the analysis and transformation and aligns with other loop optimizations.
Reviewers: jdoerfert, Meinersbur, dmgreen, etiotto, Whitney, fhahn, hfinkel
Reviewed By: Meinersbur
Subscribers: ormris, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D71025
Summary:
In commit d60f34c20a (llvm-svn 317128,
PR35113) MergeBlockIntoPredecessor was changed into
discarding some dbg.value intrinsics referring to
PHI values, post-splice due to loop rotation.
That elimination of dbg.value intrinsics did not
consider which dbg.value to keep depending on the
context (e.g. if the variable is changing its value
several times inside the basic block).
In the past that hasn't been such a big problem since
CodeGenPrepare::placeDbgValues has moved the dbg.value
to be next to the PHI node anyway. But after commit
00e238896c CodeGenPrepare isn't doing that
any longer, so we need to be more careful when avoiding
duplicate dbg.value intrinsics in MergeBlockIntoPredecessor.
This patch replaces the code that tried to avoid duplicate
dbg.values by using the RemoveRedundantDbgInstrs helper.
Reviewers: aprantl, jmorse, vsk
Reviewed By: aprantl, vsk
Subscribers: jholewinski, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D71480
Summary:
In commit d60f34c20a (llvm-svn 317128,
PR35113) MergeBlockIntoPredecessor was changed into
discarding some dbg.value intrinsics referring to
PHI values, post-splice due to loop rotation.
That elimination of dbg.value intrinsics does not
consider which dbg.value to keep based on the context.
Such as always keeping the one that comes first textually,
or the need to keep several of them in case the variable
is changing it's value several times inside the basic block.
In the past that hasn't been such a big problem since
CodeGenPrepare::placeDbgValues has moved the dbg.value
to be next to the PHI node anyway. But after commit
00e238896c CodeGenPrepare isn't doing that
any longer, so we need to be more careful when avoiding
duplicate dbg.value intrinsics in MergeBlockIntoPredecessor.
This patch is just a pre commit of the test case.
Subscribers: llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D71479
Summary:
Add a RemoveRedundantDbgInstrs to BasicBlockUtils with the
goal to remove redundant dbg intrinsics from a basic block.
This can be useful after various transforms, as it might
be simpler to do a filtering of dbg intrinsics after the
transform than during the transform.
One primary use case would be to replace a too aggressive
removal done by MergeBlockIntoPredecessor, seen at loop
rotate (not done in this patch).
The elimination algorithm currently focuses on dbg.value
intrinsics and is doing two iterations over the BB.
First we iterate backward starting at the last instruction
in the BB. Whenever a consecutive sequence of dbg.value
instructions are found we keep the last dbg.value for
each variable found (variable fragments are identified
using the {DILocalVariable, FragmentInfo, inlinedAt}
triple as given by the DebugVariable helper class).
Next we iterate forward starting at the first instruction
in the BB. Whenever we find a dbg.value describing a
DebugVariable (identified by {DILocalVariable, inlinedAt})
we save the {DIValue, DIExpression} that describes that
variables value. But if the variable already was mapped
to the same {DIValue, DIExpression} pair we instead drop
the second dbg.value.
To ease the process of making lit tests for this utility a
new pass is introduced called RedundantDbgInstElimination.
It can be executed by opt using -redundant-dbg-inst-elim.
Reviewers: aprantl, jmorse, vsk
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D71478
shuf (inselt ?, C, IndexC), undef, <IndexC, IndexC...> --> <C, C...>
This is another missing shuffle fold pattern uncovered by the
shuffle correctness fix from D70246.
The problem was visible in the post-commit thread example, but
we managed to overcome the limitation for that particular case
with D71220.
This is something like the inverse of the previous fix - there
we didn't demand the inserted scalar, and here we are only
demanding an inserted scalar.
Differential Revision: https://reviews.llvm.org/D71488
Summary:
In preparation of D65531 as well as the reuse of these tests for the
Attributor, we modernize them and use the update_test_checks to simplify
updates.
This was done with the update_test_checks after D68819 and D68850.
Reviewers: hfinkel, vsk, dblaikie, davidxl, tejohnson, tstellar, echristo, chandlerc, efriedma, lebedev.ri
Subscribers: bollu, arphaman, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D68766
The Attributor can, to some degree, do what IPConstantProp does. We can
consequently use the corner cases already collected and tested for in
the IPConstantProp tests to improve Attributor test coverage.
This exposed various bugs fixed in previous Attributor patches.
Not all functionality of IPConstantProp is available in AAValueSimplify
and AAIsDead so some tests show that we cannot perform the expected
constant propagation.
Reviewers: fhahn, efriedma, mssimpso, davide
Subscribers: bollu, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D69748
GEP index size can be specified in the DataLayout, introduced in D42123. However, there were still places
in which getIndexSizeInBits was used interchangeably with getPointerSizeInBits. This notably caused issues
with Instcombine's visitPtrToInt; but the unit tests was incorrect, so this remained undiscovered.
This fixes the buildbot failures.
Differential Revision: https://reviews.llvm.org/D68328
Patch by Joseph Faulls!
This is a rebase of the change over D70376, which fixes an LVI cache
invalidation issue that also affected this patch.
-----
Related to D69686. As noted there, LVI currently behaves differently
for integer and pointer values: For integers, the block value is always
valid inside the basic block, while for pointers it is only valid at
the end of the basic block. I believe the integer behavior is the
correct one, and CVP relies on it via its getConstantRange() uses.
The reason for the special pointer behavior is that LVI checks whether
a pointer is dereferenced in a given basic block and marks it as
non-null in that case. Of course, this information is valid only after
the dereferencing instruction, or in conservative approximation,
at the end of the block.
This patch changes the treatment of dereferencability: Instead of
including it inside the block value, we instead treat it as something
similar to an assume (it essentially is a non-nullness assume) and
incorporate this information in intersectAssumeOrGuardBlockValueConstantRange()
if the context instruction is the terminator of the basic block.
This happens either when determining an edge-value internally in LVI,
or when a terminator was explicitly passed to getValueAt(). The latter
case makes this change not fully NFC, because we can now fold
terminator icmps based on the dereferencability information in the
same block. This is the reason why I changed one JumpThreading test
(it would optimize the condition away without the change).
Of course, we do not want to recompute dereferencability on each
intersectAssume call, so we need a new cache for this. The
dereferencability analysis requires walking the entire basic block
and computing underlying objects of all memory operands. This was
previously done separately for each queried pointer value. In the
new implementation (both because this makes the caching simpler,
and because it is faster), I instead only walk the full BB once and
cache all the dereferenced pointers. So the traversal is now performed
only once per BB, instead of once per queried pointer value.
I think the overall model now makes more sense than before, and there
will be no more pitfalls due to differing integer/pointer behavior.
Differential Revision: https://reviews.llvm.org/D69914
When we reason about the pointer argument that is byval we actually
reason about a local copy of the value passed at the call site. This was
not the case before and we wrongly introduced attributes based on the
surrounding function.
AAMemoryBehaviorArgument, AAMemoryBehaviorCallSiteArgument and
AANoCaptureCallSiteArgument are made aware of byval now. The code
to skip "subsuming positions" for reasoning follows a common pattern and
we should refactor it. A TODO was added.
Discovered by @efriedma as part of D69748.
This is the first patch adding an initial set of matrix intrinsics and a
corresponding lowering pass. This has been discussed on llvm-dev:
http://lists.llvm.org/pipermail/llvm-dev/2019-October/136240.html
The first patch introduces four new intrinsics (transpose, multiply,
columnwise load and store) and a LowerMatrixIntrinsics pass, that
lowers those intrinsics to vector operations.
Matrixes are embedded in a 'flat' vector (e.g. a 4 x 4 float matrix
embedded in a <16 x float> vector) and the intrinsics take the dimension
information as parameters. Those parameters need to be ConstantInt.
For the memory layout, we initially assume column-major, but in the RFC
we also described how to extend the intrinsics to support row-major as
well.
For the initial lowering, we split the input of the intrinsics into a
set of column vectors, transform those column vectors and concatenate
the result columns to a flat result vector.
This allows us to lower the intrinsics without any shape propagation, as
mentioned in the RFC. In follow-up patches, we plan to submit the
following improvements:
* Shape propagation to eliminate the embedding/splitting for each
intrinsic.
* Fused & tiled lowering of multiply and other operations.
* Optimization remarks highlighting matrix expressions and costs.
* Generate loops for operations on large matrixes.
* More general block processing for operation on large vectors,
exploiting shape information.
We would like to add dedicated transpose, columnwise load and store
intrinsics, even though they are not strictly necessary. For example, we
could instead emit a large shufflevector instruction instead of the
transpose. But we expect that to
(1) become unwieldy for larger matrixes (even for 16x16 matrixes,
the resulting shufflevector masks would be huge),
(2) risk instcombine making small changes, causing us to fail to
detect the transpose, preventing better lowerings
For the load/store, we are additionally planning on exploiting the
intrinsics for better alias analysis.
Reviewers: anemet, Gerolf, reames, hfinkel, andrew.w.kaylor, efriedma, rengolin
Reviewed By: anemet
Differential Revision: https://reviews.llvm.org/D70456
GEP index size can be specified in the DataLayout, introduced in D42123. However, there were still places
in which getIndexSizeInBits was used interchangeably with getPointerSizeInBits. This notably caused issues
with Instcombine's visitPtrToInt; but the unit tests was incorrect, so this remained undiscovered.
Differential Revision: https://reviews.llvm.org/D68328
Patch by Joseph Faulls!
Summary: AutoFDO compilation has two places that do inlining - the sample profile loader that does inlining with context sensitive profile, and the regular inliner as CGSCC pass. Ideally we want most inlining to come from sample profile loader as that is driven by context sensitive profile and also retains context sensitivity after inlining. However the reality is most of the inlining actually happens during regular inliner. To track the number of inline instances from sample profile loader and help move more inlining to sample profile loader, I'm adding statistics and optimization remarks for sample profile loader's inlining.
Reviewers: wmi, davidxl
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D70584
Fix for https://bugs.llvm.org/show_bug.cgi?id=40846.
This adds a combine for cases where a (a + b) < a style overflow
check is performed, but with a + b being the result of
uadd.with.overflow, so the overflow result is also already available
and we can just use it. Subsequently GVN/CSE will deduplicate the extracts.
We can run into this situation if you have both a uadd.with.overflow
and a manual add + overflow check in the same function (on the same
operands), in which case GVN will rewrite the add to the with.overflow
result and leave you with this pattern.
The implementation is a bit ugly because I'm handling the various
canonicalization edge cases.
This does not yet handle the negated version of this pattern.
Differential Revision: https://reviews.llvm.org/D58644
If the pointer was loaded/stored before the null check, the check
is redundant and can be removed. For now the optimizers do not
remove the nullptr check, see https://gcc.godbolt.org/z/H2r5GG.
The patch allows to use more nonnull constraints. Also, it found
one more optimization in some PowerPC test. This is my first llvm
review, I am free to any comments.
Differential Revision: https://reviews.llvm.org/D71177
Fix for https://bugs.llvm.org/show_bug.cgi?id=44236. This code was
originally introduced in rG36512330041201e10f5429361bbd79b1afac1ea1.
However, the attribute copying was done in the wrong place (in general
call replacement, not thunk generation) and a proper fix was
implemented in D12581.
Previously this code was just unnecessary but harmless (because
FunctionComparator ensured that the attributes of the two functions
are exactly the same), but since byval was changed to accept a type
this copying is actively wrong and may result in malformed IR.
Differential Revision: https://reviews.llvm.org/D71173
This pattern is noted as a regression from:
D70246
...where we removed an over-aggressive shuffle simplification.
SimplifyDemandedVectorElts fails to catch this case when the insert has multiple uses,
so I'm proposing to pattern match the minimal sequence directly. This fold does not
conflict with any of our current shuffle undef/poison semantics.
Differential Revision: https://reviews.llvm.org/D71220
basic blocks
Originally applied in 72ce759928.
Fixed a build failure caused by incorrect use of cast instead of
dyn_cast.
This reverts commit 8b0780f795.
Summary:
This change is in preparation to reuse these test for the Attributor.
It mainly is to remove UB, make it clear what is tested, and use
"modern" run lines.
Reviewers: fhahn, efriedma, mssimpso, davide
Subscribers: bollu, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D69747
Summary:
Same as D60846 and D69571 but with a fix for the problem encountered
after them. Both times it was a missing context adjustment in the
handling of PHI nodes.
The reproducers created from the bugs that caused the old commits to be
reverted are included.
Reviewers: nikic, nlopes, mkazantsev, spatel, dlrobertson, uabelho, hakzsam, hans
Subscribers: hiraditya, bollu, asbirlea, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D71181
With the extra optimisations we have done, these should now be fine to
enable by default. Which is what this patch does.
Differential Revision: https://reviews.llvm.org/D70968
This attempts to teach the cost model in Arm that code such as:
%s = shl i32 %a, 3
%a = and i32 %s, %b
Can under Arm or Thumb2 become:
and r0, r1, r2, lsl #3
So the cost of the shift can essentially be free. To do this without
trying to artificially adjust the cost of the "and" instruction, it
needs to get the users of the shl and check if they are a type of
instruction that the shift can be folded into. And so it needs to have
access to the actual instruction in getArithmeticInstrCost, which if
available is added as an extra parameter much like getCastInstrCost.
We otherwise limit it to shifts with a single user, which should
hopefully handle most of the cases. The list of instruction that the
shift can be folded into include ADC, ADD, AND, BIC, CMP, EOR, MVN, ORR,
ORN, RSB, SBC and SUB. This translates to Add, Sub, And, Or, Xor and
ICmp.
Differential Revision: https://reviews.llvm.org/D70966
This adds some extra cost model tests for shifts, and does some minor
adjustments to some Neon code to make it clear as to what it applies to.
Both NFC.
This caused "Too many bits for uint64_t" asserts when building Chromium. See
https://crbug.com/1031978#c2 for a reproducer. I'll follow up on the
llvm-commits thread with a creduced version.
> ARMCodeGenPrepare has already been generalized and renamed to
> TypePromotion. We've had it enabled and tested downstream for a
> while, so enable it by default.
>
> Differential Revision: https://reviews.llvm.org/D70998
This is another transform suggested in PR44153:
https://bugs.llvm.org/show_bug.cgi?id=44153
The backend for some targets already manages to get
this if it converts copysign to bitwise logic.
Alas, using half the available vector registers in a single instruction
is just too much for the register allocator to handle. The mve-vldst4.ll
test here fails when these instructions are enabled at present. This
patch disables the generation of VLD4 and VST4 by adding a
mve-max-interleave-factor option, which we currently default to 2.
Differential Revision: https://reviews.llvm.org/D71109
Currently we fail to pick the right insertion point when
PreviousLastPart of a first-order-recurrence is a PHI node not in the
LoopVectorBody. This can happen when PreviousLastPart is produce in a
predicated block. In that case, we should pick the insertion point in
the BB the PHI is in.
Fixes PR44020.
Reviewers: hsaito, fhahn, Ayal, dorit
Reviewed By: Ayal
Differential Revision: https://reviews.llvm.org/D71071
Summary:
Sample profile loader of AutoFDO tries to replay previous inlining using context sensitive profile. The replay only repeats inlining if the call site block is hot. As a result it punts inlining of small functions, some of which can be beneficial for size, and will still be inlined by CSGCC inliner later. The oscillation between sample profile loader's inlining and regular CGSSC inlining cause unnecessary loss of context-sensitive profile. It doesn't have much impact for inline decision itself, but it negatively affects post-inline profile quality as CGSCC inliner have to scale counts which is not as accurate as the original context sensitive profile, and bad post-inline profile can misguide code layout.
This change added regular Inline Cost calculation for sample profile loader, so we can inline small functions upfront under switch -sample-profile-inline-size. In addition -sample-profile-cold-inline-threshold is added so we can tune the separate size threshold - currently the default is chosen to be the same as regular inliner's cold call-site threshold.
Reviewers: wmi, davidxl
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D70750
In general ValueHandleBase::ValueIsRAUWd shouldn't be called when not
all uses of the value were actually replaced, though, currently
formLCSSAForInstructions calls it when it inserts LCSSA-phis.
Calls of ValueHandleBase::ValueIsRAUWd were added to LCSSA specifically
to update/invalidate SCEV. In the best case these calls duplicate some
of the work already done by SE->forgetValue, though in case when SCEV of
the value is SCEVUnknown, SCEV replaces the underlying value of
SCEVUnknown with the new value (i.e. acts like LCSSA-phi actually fully
replaces the value it is created for), which leads to SCEV being
corrupted because LCSSA-phi rarely dominates all uses of its inputs.
Fixes bug https://bugs.llvm.org/show_bug.cgi?id=44058.
Reviewers: fhahn, efriedma, reames, sanjoy.google
Reviewed By: fhahn
Subscribers: hiraditya, javed.absar, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D70593
Summary:
AutoFDO's sample profile loader processes function in arbitrary source code order, so if I change the order of two functions in source code, the inline decision can change. This also prevented the use of context-sensitive profile to do specialization while inlining. This commit enforces SCC top-down order for sample profile loader. With this change, we can now do specialization, as illustrated by the added test case:
Say if we have A->B->C and D->B->C call path, we want to inline C into B when root inliner is B, but not when root inliner is A or D, this is not possible without enforcing top-down order. E.g. Once C is inlined into B, A and D can only choose to inline (B->C) as a whole or nothing, but what we want is only inline B into A and D, not its recursive callee C. If we process functions in top-down order, this is no longer a problem, which is what this commit is doing.
This change is guarded with a new switch "-sample-profile-top-down-load" for tuning, and it depends on D70653. Eventually, top-down can be the default order for sample profile loader.
Reviewers: wmi, davidxl
Subscribers: hiraditya, llvm-commits, tejohnson
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D70655
Summary:
When sample profile loader decides not to inline a previously inlined call-site, we adjust the profile of outlined function simply by scaling up its profile counts by call-site count. This means the context-sensitive profile of that inlined instance will be thrown away. This commit try to keep context-sensitive profile for such cases:
- Instead of scaling outlined function's profile, we now properly merge the FunctionSamples of inlined instance into outlined function, including all recursively inlined profile.
- Instead of adjusting the profile for negative inline decision at the end of the sample profile loader pass, we do the profile merge right after processing each function. This change paired with top-down ordering of annotation/inline-replay (a separate diff) will make sure we recursively merge profile back before the profile is used for annotation and inline replay.
A new switch -sample-profile-merge-inlinee is added to enable the new profile merge for tuning. It should be the default behavior eventually.
Reviewers: wmi, davidxl
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D70653
Summary:
Emit a value debug intrinsic (with OP_deref) when an alloca address is
passed to a function call after going through a bitcast.
This generates an FP or SP-relative location for the local variable in
the following case:
int x;
use((void *)&x;
Reviewers: aprantl, vsk, pcc
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D70752
Summary:
This reverts commit c3b06d0c39.
Reason for revert: Caused miscompiles when inserting assume for undef.
Also adds a test to prevent similar breakage in future.
Fixes PR44154.
Reviewers: rnk, jdoerfert, efriedma, xbolva00
Reviewed By: rnk
Subscribers: thakis, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D70933
Summary:
D68408 proposes to greatly improve our negation sinking abilities.
But in current canonicalization, we produce `sub A, zext(B)`,
which we will consider non-canonical and try to sink that negation,
undoing the existing canonicalization.
So unless we explicitly stop producing previous canonicalization,
we will have two conflicting folds, and will end up endlessly looping.
This inverts canonicalization, and adds back the obvious fold
that we'd miss:
* `sub [nsw] Op0, sext/zext (bool Y) -> add [nsw] Op0, zext/sext (bool Y)`
https://rise4fun.com/Alive/xx4
* `sext(bool) + C -> bool ? C - 1 : C`
https://rise4fun.com/Alive/fBl
It is obvious that `@ossfuzz_9880()` / `@lshr_out_of_range()`/`@ashr_out_of_range()`
(oss-fuzz 4871) are no longer folded as much, though those aren't really worrying.
Reviewers: spatel, efriedma, t.p.northover, hfinkel
Reviewed By: spatel
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D71064
The patch makes sure that the LastThrowing pointer does not point to any instruction deleted by call to DeleteDeadInstruction.
While iterating through the instructions the pass maintains a pointer to the lastThrowing Instruction. A call to deleteDeadInstruction deletes a dead store and other instructions feeding the original dead instruction which also become dead. The instruction pointed by the lastThrowing pointer could also be deleted by the call to DeleteDeadInstruction and thus it becomes a dangling pointer. Because of this, we see an error in the next iteration.
In the patch, we maintain a list of throwing instructions encountered previously and use the last non deleted throwing instruction from the container.
Patch by Ankit <quic_aankit@quicinc.com>
Reviewers: fhahn, bcahoon, efriedma
Reviewed By: fhahn
Differential Revision: https://reviews.llvm.org/D65326
ARMCodeGenPrepare has already been generalized and renamed to
TypePromotion. We've had it enabled and tested downstream for a
while, so enable it by default.
Differential Revision: https://reviews.llvm.org/D70998
The PHI node checks for inner loop exits are too permissive currently.
As indicated by an existing comment, we should only allow LCSSA PHI
nodes that are part of reductions or are only used outside of the loop
nest. We ensure this by checking the users of the LCSSA PHIs.
Specifically, it is not safe to use an exiting value from the inner loop in the latch of the outer
loop.
It also moves the inner loop exit check before the outer loop exit
check.
Fixes PR43473.
Reviewers: efriedma, mcrosier
Reviewed By: efriedma
Differential Revision: https://reviews.llvm.org/D68144
When basic blocks are killed, either due to being empty or to being an if.then
or if.else block whose complement contains identical instructions, some of the
debug intrinsics in that block are lost. This patch sinks those intrinsics
into the single successor block, setting them Undef if necessary to
prevent debug info from falling out-of-date.
Differential Revision: https://reviews.llvm.org/D70318
SCEV caches the exiting blocks when computing exit counts. In
SimpleLoopUnswitch, we split the exit block of the loop to unswitch.
Currently we only invalidate the loop containing that exit block, but if
that block is the exiting block for a parent loop, we have stale cache
entries. We have to invalidate the top-most loop that contains the exit
block as exiting block. We might also be able to skip invalidating the
loop containing the exit block, if the exit block is not an exiting
block of that loop.
There are also 2 more places in SimpleLoopUnswitch, that use a similar
problematic approach to get the loop to invalidate. If the patch makes
sense, I will also update those places to a similar approach (they deal
with multiple exit blocks, so we cannot directly re-use
getTopMostExitingLoop).
Fixes PR43972.
Reviewers: skatkov, reames, asbirlea, chandlerc
Reviewed By: asbirlea
Differential Revision: https://reviews.llvm.org/D70786
Currently, getIntImmCost returns TCC_Free for almost all intrinsics.
For most AArch64 specific intrinsics however, it looks like integer
constants cannot be folded into most of them (at least the ones I checked).
Unless we know that we can fold integer operands with the intrinsic, we
handle more cases correctly by returning the cost to materialize the
immediate than return TCC_Free.
Reviewers: SjoerdMeijer, dmgreen, t.p.northover, ributzka
Reviewed By: SjoerdMeijer
Differential Revision: https://reviews.llvm.org/D70669
This reverts these two commits
[InstCombine] Turn (extractelement <1 x i64/double> (bitcast (x86_mmx))) into a single bitcast from x86_mmx to i64/double.
[InstCombine] Don't transform bitcasts between x86_mmx and v1i64 into insertelement/extractelement
We're seeing at least one internal test failure related to a
bitcast that was previously before an inline assembly block
containing emms being placed after it. This leads to the mmx
state ending up not empty after the emms. IR has no way to
make any specific guarantees about this. Reverting these patches
to get back to previous behavior which at least worked for this
test.
Fix PR40816: avoid considering scalar-with-predication instructions as also
uniform-after-vectorization.
Instructions identified as "scalar with predication" will be "vectorized" using
a replicating region. If such instructions are also optimized as "uniform after
vectorization", namely when only the first of VF lanes is used, such a
replicating region becomes erroneous - only the first instance of the region can
and should be formed. Fix such cases by not considering such instructions as
"uniform after vectorization".
Differential Revision: https://reviews.llvm.org/D70298
Summary:
Make SLPVectorize to recognize homogeneous aggregates like
`{<2 x float>, <2 x float>}`, `{{float, float}, {float, float}}`,
`[2 x {float, float}]` and so on.
It's a follow-up of https://reviews.llvm.org/D70068.
Merged `findBuildVector()` and `findBuildAggregate()` to
one `findBuildAggregate()` function making it recursive
to recognize multidimensional aggregates. Aggregates required
to be homogeneous.
Reviewers: RKSimon, ABataev, dtemirbulatov, spatel, vporpo
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D70587
Convert ARMCodeGenPrepare into a generic type promotion pass by:
- Removing the insertion of arm specific intrinsics to handle narrow
types as we weren't using this.
- Removing ARMSubtarget references.
- Now query a generic TLI object to know which types should be
promoted and what they should be promoted to.
- Move all codegen tests into Transforms folder and testing using opt
and not llc, which is how they should have been written in the
first place...
The pass searches up from icmp operands in an attempt to safely
promote types so we can avoid generating unnecessary unsigned extends
during DAG ISel.
Differential Revision: https://reviews.llvm.org/D69556
Summary:
This cropped up in the Linux kernel where cold code was placed in an
incompatible section.
Reviewers: compnerd, vsk, tejohnson
Reviewed By: vsk
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D70925
As described here:
https://bugs.llvm.org/show_bug.cgi?id=44186
The match() code safely allows undef values, but we can't safely
propagate a vector constant that contains an undef to the new
compare instruction.
Summary:
If a user writing C code using the ACLE MVE intrinsics generates a
predicate and then complements it, then the resulting IR will use the
`pred_v2i` IR intrinsic to turn some `<n x i1>` vector into a 16-bit
integer; complement that integer; and convert back. This will generate
machine code that moves the predicate out of the `P0` register,
complements it in an integer GPR, and moves it back in again.
This InstCombine rule replaces `i2v(~v2i(x))` with a direct complement
of the original predicate vector, which we can already instruction-
select as the VPNOT instruction which complements P0 in place.
Reviewers: ostannard, MarkMurrayARM, dmgreen
Reviewed By: dmgreen
Subscribers: kristof.beyls, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D70484
Summary:
This patch moves the test cases related to Attributor to `Transforms/Attributor` folder.
We have used `Transforms/FunctionAttrs` as the primary folder for Attributor test but we need to change testing way now.
For the test cases which I think functionattrs doesn't infer anything something like (willreturn, nosync, value-simplify, h2s ..etc), I moved them with the command `git mv`.
For the test cases in which functoinattrs and attributor are tested, I copied the test to the folder and remove the check only used by functoinattrs.
Reviewers: jdoerfert, sstefan1
Reviewed By: jdoerfert
Subscribers: jfb, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D70843
rL341831 moved one-use check higher up, restricting a few folds
that produced a single instruction from two instructions to the case
where the inner instruction would go away.
Original commit message:
> InstCombine: move hasOneUse check to the top of foldICmpAddConstant
>
> 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.
From the commit message alone, there doesn't seem to be a
deeper motivation, deeper problem that was trying to solve,
other than 'fixing the wrong one-use check'.
As i have briefly discusses in IRC with Tim, the original motivation
can no longer be recovered, too much time has passed.
However i believe that the original fold was doing the right thing,
we should be performing such a transformation even if the inner `add`
will not go away - that will still unchain the comparison from `add`,
it will no longer need to wait for `add` to compute.
Doing so doesn't seem to break any particular idioms,
as least as far as i can see.
References https://bugs.llvm.org/show_bug.cgi?id=44100
If the sign of the sign argument is known (this could be extended to use ValueTracking),
then we can use fneg+fabs to clear/set the sign bit of the magnitude argument.
http://llvm.org/docs/LangRef.html#llvm-copysign-intrinsic
This transform is already done in DAGCombiner, but we can do it sooner in IR as
suggested in PR44153:
https://bugs.llvm.org/show_bug.cgi?id=44153
We have effectively no analysis for copysign in IR, so we are taking the unusual step
of increasing the number of IR instructions for the negative constant case.
Differential Revision: https://reviews.llvm.org/D70792
Summary:
optimizeVectorResize is rewriting patterns like:
%1 = bitcast vector %src to integer
%2 = trunc/zext %1
%dst = bitcast %2 to vector
Since bitcasting between integer an vector types gives
different integer values depending on endianness, we need
to take endianness into account. As it happens the old
implementation only produced the correct result for little
endian targets.
Fixes: https://bugs.llvm.org/show_bug.cgi?id=44178
Reviewers: spatel, lattner, lebedev.ri
Reviewed By: spatel, lebedev.ri
Subscribers: lebedev.ri, hiraditya, uabelho, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D70844
The constants come through as add %x, -C, not a sub as would be
expected. They need some extra matchers to canonicalise them towards
usub_sat.
Differential Revision: https://reviews.llvm.org/D69514
This adjusts the one use checks in the the usub_sat fold code to not
increase instruction count, but otherwise do the fold. Reviewed as a
part of D69514.
Some tests in test/Transforms/InstCombine/cast.ll depend on
endianness. Added a second run line to run the tests with both
big and little endian. In the past we only compiled for big
endian, and then it was hard to see if any big endian bugfixes
would impact the little endian result etc.
Summary:
This patch introduces the deduction based on load/store instructions whose pointer operand is a non-inbounds GEP instruction.
For example if we have,
```
void f(int *u){
u[0] = 0;
u[1] = 1;
u[2] = 2;
}
```
then u must be dereferenceable(12).
This patch is inspired by D64258
Reviewers: jdoerfert, spatel, hfinkel, RKSimon, sstefan1, xbolva00, dtemirbulatov
Reviewed By: jdoerfert
Subscribers: jfb, lebedev.ri, xbolva00, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D70714
Currently every time we encounter an indirect call of a known function,
we try to evaluate the inline cost of that function. In case of a
recursion, that evaluation never stops.
The solution I propose is to evaluate only the indirect call of the
function, while any further indirect calls (of a known function) will be
treated just as direct function calls, which, actually, never tries to
evaluate the call.
Fixes PR35469.
Differential Revision: https://reviews.llvm.org/D69349
I'm not sure what the effect of this change will be on all of the affected
tests or a larger benchmark, but it fixes the horizontal add/sub problems
noted here:
https://reviews.llvm.org/D59710?vs=227972&id=228095&whitespace=ignore-most#toc
The costs are based on reciprocal throughput numbers in Agner's tables for
PEXTR*; these appear to be very slow ops on Silvermont.
This is a small step towards the larger motivation discussed in PR43605:
https://bugs.llvm.org/show_bug.cgi?id=43605
Also, it seems likely that insert/extract is the source of perf regressions on
other CPUs (up to 30%) that were cited as part of the reason to revert D59710,
so maybe we'll extend the table-based approach to other subtargets.
Differential Revision: https://reviews.llvm.org/D70607
This reapplies: 8ff85ed905
Original commit message:
As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there.
This change doesn't include any change to move from selection dag to fast isel
and that will come with other numbers that should help inform that decision.
There also haven't been any real debuggability studies with this pipeline yet,
this is just the initial start done so that people could see it and we could start
tweaking after.
Test updates: Outside of the newpm tests most of the updates are coming from either
optimization passes not run anymore (and without a compelling argument at the moment)
that were largely used for canonicalization in clang.
Original post:
http://lists.llvm.org/pipermail/llvm-dev/2019-April/131494.html
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D65410
This reverts commit c9ddb02659.
This is correct for any value including NaN/inf.
We don't have this fold directly in the backend either,
but x86 manages to get it after converting things to bitops.
Summary:
This patch enables us to track GEP instruction in align deduction.
If a pointer `B` is defined as `A+Offset` and known to have alignment `C`, there exists some integer Q such that
```
A + Offset = C * Q = B
```
So we can say that the maximum power of two which is a divisor of gcd(Offset, C) is an alignment.
Reviewers: jdoerfert, sstefan1
Reviewed By: jdoerfert
Subscribers: lebedev.ri, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D70392
This change doesn't include any change to move from selection dag to fast isel
and that will come with other numbers that should help inform that decision.
There also haven't been any real debuggability studies with this pipeline yet,
this is just the initial start done so that people could see it and we could start
tweaking after.
Test updates: Outside of the newpm tests most of the updates are coming from either
optimization passes not run anymore (and without a compelling argument at the moment)
that were largely used for canonicalization in clang.
Original post:
http://lists.llvm.org/pipermail/llvm-dev/2019-April/131494.html
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D65410
This version contains 2 fixes for reported issues:
1. Make sure we do not try to sink terminator instructions.
2. Make sure we bail out, if we try to sink an instruction that needs to
stay in place for another recurrence.
Original message:
If the recurrence PHI node has a single user, we can sink any
instruction without side effects, given that all users are dominated by
the instruction computing the incoming value of the next iteration
('Previous'). We can sink instructions that may cause traps, because
that only causes the trap to occur later, but not on any new paths.
With the relaxed check, we also have to make sure that we do not have a
direct cycle (meaning PHI user == 'Previous), which indicates a
reduction relation, which potentially gets missed by
ReductionDescriptor.
As follow-ups, we can also sink stores, iff they do not alias with
other instructions we move them across and we could also support sinking
chains of instructions and multiple users of the PHI.
Fixes PR43398.
Reviewers: hsaito, dcaballe, Ayal, rengolin
Reviewed By: Ayal
Differential Revision: https://reviews.llvm.org/D69228
Currently the assertion in updateSuccessor is overly strict in some
cases and overly relaxed in other cases. For branches to the inner and
outer loop preheader it is too strict, because they can either be
unconditional branches or conditional branches with duplicate targets.
Both cases are fine and we can allow updating multiple successors.
On the other hand, we have to at least update one successor. This patch
adds such an assertion.
And simultaneously enhance SimplifyDemandedVectorElts() to rcognize that
pattern. That preserves some of the old optimizations in IR.
Given a shuffle that includes undef elements in an otherwise identity mask like:
define <4 x float> @shuffle(<4 x float> %arg) {
%shuf = shufflevector <4 x float> %arg, <4 x float> undef, <4 x i32> <i32 undef, i32 1, i32 2, i32 3>
ret <4 x float> %shuf
}
We were simplifying that to the input operand.
But as discussed in PR43958:
https://bugs.llvm.org/show_bug.cgi?id=43958
...that means that per-vector-element poison that would be stopped by the shuffle can now
leak to the result.
Also note that we still have (and there are tests for) the same transform with no undef
elements in the mask (a fully-defined identity mask). I don't think there's any
controversy about that case - it's a valid transform under any interpretation of
shufflevector/undef/poison.
Looking at a few of the diffs into codegen, I don't see any difference in final asm. So
depending on your perspective, that's good (no real loss of optimization power) or bad
(poison exists in the DAG, so we only partially fixed the bug).
Differential Revision: https://reviews.llvm.org/D70246
Currently every time we encounter an indirect call of a known function,
we try to evaluate the inline cost of that function. In case of a
recursion, that evaluation never stops.
The solution presented is to evaluate only the indirect call of the
function, while any further indirect calls (of a known function) will be
treated just as direct function calls, which, actually, never tries to
evaluate the call.
Fixes PR35469.
Differential Revision: https://reviews.llvm.org/D69349
Summary:
Vector aggregate is homogeneous aggregate of vectors like `{ <2 x float>, <2 x float> }`.
This patch allows `findBuildAggregate()` to consider vector aggregates as
well as scalar ones. For instance, `{ <2 x float>, <2 x float> }` maps to `<4 x float>`.
Fixes vector part of llvm.org/PR42022
Reviewers: RKSimon
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D70068
Summary:
Without this patch, the jump threading pass ignores profiling data
whenever we invoke the pass with the new pass manager.
Specifically, JumpThreadingPass::run calls runImpl with class variable
HasProfileData always set to false. In turn, runImpl sets
HasProfileData to false again:
HasProfileData = HasProfileData_;
In the end, we don't use profiling data at all with the new pass
manager.
This patch fixes the problem by passing F.hasProfileData() to runImpl.
The bug appears to have been introduced at:
https://reviews.llvm.org/D41461
which removed local variable HasProfileData in JumpThreadingPass::run
even though there was one more use left in the same function. As a
result, the remaining use ended referring to the class variable
instead.
Note that F.hasProfileData is an extremely lightweight function, so I
don't see the need to cache its result. Once this patch is approved,
I'm planning to stop caching the result of F.hasProfileData in
runOnFunction.
Reviewers: wmi, eli.friedman
Subscribers: hiraditya, jfb, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D70509
We may end up with a case where we have a widenable branch above the loop, but not all widenable branches within the loop have been removed. Since a widenable branch inhibit SCEVs ability to reason about exit counts (by design), we have a tradeoff between effectiveness of this optimization and allowing future widening of the branches within the loop. LoopPred is thought to be one of the most important optimizations for range check elimination, so let's pay the cost.
Bit-Tracking Dead Code Elimination (bdce) do not mark dbg.value as undef after
deleting instruction. which shows invalid state of variable in debugger. This
patches fixes this by marking the dbg.value as undef which depends on dead
instruction.
This fixes https://bugs.llvm.org/show_bug.cgi?id=41925
Patch by kamlesh kumar!
Differential Revision: https://reviews.llvm.org/D70040
As a reminder, a "widenable branch" is the pattern "br i1 (and i1 X, WC()), label %taken, label %untaken" where "WC" is the widenable condition intrinsics. The semantics of such a branch (derived from the semantics of WC) is that a new condition can be added into the condition arbitrarily without violating legality.
Broaden the definition in two ways:
Allow swapped operands to the br (and X, WC()) form
Allow widenable branch w/trivial condition (i.e. true) which takes form of br i1 WC()
The former is just general robustness (e.g. for X = non-instruction this is what instcombine produces). The later is specifically important as partial unswitching of a widenable range check produces exactly this form above the loop.
Differential Revision: https://reviews.llvm.org/D70502
Follow-up of cb47b8783: don't query TTI->preferPredicateOverEpilogue when
option -prefer-predicate-over-epilog is set to false, i.e. when we prefer not
to predicate the loop.
Differential Revision: https://reviews.llvm.org/D70382
After speaking with Sanjay - seeing a number of miscompiles and working
on tracking down a testcase. None of the follow on patches seem to
have helped so far.
This reverts commit 8a0aa5310b.
After speaking with Sanjay - seeing a number of miscompiles and working
on tracking down a testcase. None of the follow on patches seem to
have helped so far.
This reverts commit 7ff57705ba.
This code has never been enabled. While it is tested, it's complicating some refactoring. If we decide to re-implement this, doing it in SimplifyCFG would probably make more sense anyways.
Unswitch (and other loop transforms) like to generate loop exit blocks with unconditional successors, and phi nodes (LCSSA, or simple multiple exiting blocks sharing an exit). Generalize the "likely very rare exit" check slightly to handle this form.
Pair up inlined AutoreleaseRV calls with their matching RetainRV or
ClaimRV.
- RetainRV cancels out AutoreleaseRV. Delete both instructions.
- ClaimRV is a peephole for RetainRV+Release. Delete AutoreleaseRV and
replace ClaimRV with Release.
This avoids problems where more aggressive inlining triggers memory
regressions.
This patch is happy to skip over non-callable instructions and non-ARC
intrinsics looking for the pair. It is likely sound to also skip over
opaque function calls, but that's harder to reason about, and it's not
relevant to the goal here: if there's an opaque function call splitting
up a pair, it's very unlikely that a handshake would have happened
dynamically without inlining.
Note that this patch also subsumes the previous logic that looked
backwards from ReleaseRV.
https://reviews.llvm.org/D70370
rdar://problem/46509586
The 1st attempt was reverted because it revealed an existing
bug where we could produce invalid IR (use of value before
definition). That should be fixed with:
rG39de82ecc9c2
The bug manifests as replacing a reduction operand with an undef
value.
The problem appears to be limited to cases where a min/max reduction
has extra uses of the compare operand to the select.
In the general case, we are tracking "ExternallyUsedValues" and
an "IgnoreList" of the reduction operations, but those may not apply
to the final compare+select in a min/max reduction.
For that, we use replaceAllUsesWith (RAUW) to ensure that the new
vectorized reduction values are transferred to all subsequent users.
Differential Revision: https://reviews.llvm.org/D70148
Now that we have the intrinsics, we can add VLD2/4 and VST2/4 lowering
for MVE. This works the same way as Neon, recognising the load/shuffles
combination and converting them into intrinsics in a pre-isel pass,
which just calls getMaxSupportedInterleaveFactor, lowerInterleavedLoad
and lowerInterleavedStore.
The main difference to Neon is that we do not have a VLD3 instruction.
Otherwise most of the code works very similarly, with just some minor
differences in the form of the intrinsics to work around. VLD3 is
disabled by making isLegalInterleavedAccessType return false for those
cases.
We may need some other future adjustments, such as VLD4 take up half the
available registers so should maybe cost more. This patch should get the
basics in though.
Differential Revision: https://reviews.llvm.org/D69392
Roman Lebedev