Summary:
In case of non-alloca pointers, we check for whether it is a pointer
from malloc-like calls and it is not captured. In such case, we can
promote the pointer, as the caller will have no way to access this pointer
even if there is unwinding in middle of the loop.
Reviewers: hfinkel, sanjoy, reames, eli.friedman
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D28834
llvm-svn: 292510
Type identifiers are exported by:
- Adding coarse-grained information about how to test the type
identifier to the summary.
- Creating symbols in the object file (aliases and absolute symbols)
containing fine-grained information about the type identifier.
Differential Revision: https://reviews.llvm.org/D28424
llvm-svn: 292462
This changes the vectorizer to explicitly use the loopsimplify and lcssa utils,
instead of "requiring" the transformations as if they were analyses.
This is not NFC, since it changes the LCSSA behavior - we no longer run LCSSA
for all loops, but rather only for the loops we expect to modify.
Differential Revision: https://reviews.llvm.org/D28868
llvm-svn: 292456
We currently check whether a reduction has a single outside user. We don't
really need to require that - we just need to make sure a single value is
used externally. The number of external users of that value shouldn't actually
matter.
Differential Revision: https://reviews.llvm.org/D28830
llvm-svn: 292424
claims to test.
LoopSimplify was unifying the multiple exits in this test case, making
it never even test the multiple exit handling of LoopDeletion. Doh.
Now it works (thanks to a great idea from mkuper) and will fail if we
ever change something to make it stop working.
llvm-svn: 292331
Summary: Partial unrolling should have separate threshold with full unrolling.
Reviewers: efriedma, mzolotukhin
Reviewed By: efriedma, mzolotukhin
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D28831
llvm-svn: 292293
Summary: Add a test case for LICM when promoting locals that may be read after the throw within the loop.
Reviewers: eli.friedman, hfinkel, sanjoy
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D28822
llvm-svn: 292261
If a memory instruction will be vectorized, but it's pointer operand is
non-consecutive-like, the instruction is a gather or scatter operation. Its
pointer operand will be non-uniform. This should fix PR31671.
Reference: https://llvm.org/bugs/show_bug.cgi?id=31671
Differential Revision: https://reviews.llvm.org/D28819
llvm-svn: 292254
runnig LCSSA over them prior to running the loop pipeline.
This also teaches the loop PM to verify that LCSSA form is preserved
throughout the pipeline's run across the loop nest.
Most of the test updates just leverage this new functionality. One has to be
relaxed with the new PM as IVUsers is less powerful when it sees LCSSA input.
Differential Revision: https://reviews.llvm.org/D28743
llvm-svn: 292241
Also, add the corresponding match to the AssumptionCache's 'Affected Values' list.
Differential Revision: https://reviews.llvm.org/D28485
llvm-svn: 292239
Add missing fabs(fpext) optimzation that worked with the call,
and also fixes it creating a second fpext when there were multiple
uses.
llvm-svn: 292172
Simplify a pshufb shuffle mask based on the elements of the mask that are actually demanded.
Differential Revision: https://reviews.llvm.org/D28745
llvm-svn: 292101
First, I've moved a test of IVUsers from the LSR tree to a dedicated
IVUsers test directory. I've also simplified its RUN line now that the
new pass manager's loop PM is providing analyses on their own.
No functionality changed, but it makes subsequent changes cleaner.
llvm-svn: 292060
cover domtree and alias analysis. These are the pretty clear analyses
that we would always want to survive this pass.
To make these survive, we also need to preserve the assumption cache.
Added a test that verifies the important bits of this preservation.
llvm-svn: 292037
Allows LLVM to optimize sequences like the following:
%add = add nuw i32 %x, 1
%cmp = icmp ugt i32 %add, %y
Into:
%cmp = icmp uge i32 %x, %y
Previously, only signed comparisons were being handled.
Decrements could also be handled, but 'sub nuw %x, 1' is currently canonicalized to
'add %x, -1' in InstCombineAddSub, losing the nuw flag. Removing that canonicalization
seems like it might have far-reaching ramifications so I kept this simple for now.
Patch by Matti Niemenmaa!
Differential Revision: https://reviews.llvm.org/D24700
llvm-svn: 291975
Summary:
This is a testcase where phi node cycling happens, and because we do
not order the leaders by domination or anything similar, the leader
keeps changing.
Using std::set for the members is too expensive, and we actually don't
need them sorted all the time, only at leader changes.
We could keep both a set and a vector, and keep them mostly sorted and
resort as necessary, or use a set and a fibheap, but all of this seems
premature.
After running some statistics, we are able to avoid the vast majority
of sorting by keeping a "next leader" field. Most congruence classes only have
leader changes once or twice during GVN.
Reviewers: davide
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D28594
llvm-svn: 291968
Summary:
Memory Dependence Analysis was limited to return only local dependencies
for invariant.group handling. Now it returns NonLocal when it finds it
and then by asking getNonLocalPointerDependency we get found dep.
Thanks to this we are able to devirtualize loops!
void indirect(A &a, int n) {
for (int i = 0 ; i < n; i++)
a.foo();
}
void test(int n) {
A a;
indirect(a);
}
After inlining a.foo() will be changed to direct call, even if foo and A::A()
is external (but only if vtable definition is be available).
Reviewers: nlewycky, dberlin, chandlerc, rsmith
Subscribers: mehdi_amini, davide, llvm-commits
Differential Revision: https://reviews.llvm.org/D28137
llvm-svn: 291762
This test seems to have largely been relying on asserts being tripped.
It had a very specific and somewhat uninteresting grep of the output,
but it never really did anything to cause SCEV to be preserved across
loop simplify, certainly not explicitly. And a later addition to it
actually added CHECK lines despite the test never running FileCheck.
Now we actually print SCEV before and after loop simplify to make sure
it is *changing* and being *updated*. Which seems to be much more likely
the point of the test.
llvm-svn: 291740
This means that we can use a shorter instruction sequence in the case where
the size is a power of two and on the boundary between two representations.
Differential Revision: https://reviews.llvm.org/D28421
llvm-svn: 291706
classes, and updating checking to allow for equivalence through
reachability.
(Sadly, the checking here is not perfect, and can't be made perfect,
so we'll have to disable it after we are satisfied with correctness.
Right now it is just "very unlikely" to happen.)
llvm-svn: 291698
The removed assert seems bogus - it's perfectly legal for the roots of the
vectorized subtrees to be equal even if the original scalar values aren't,
if the original scalars happen to be equivalent.
This fixes PR31599.
Differential Revision: https://reviews.llvm.org/D28539
llvm-svn: 291692
Summary:
Revert LowerTypeTests: Split the pass in two: a resolution phase and a lowering phase.
This change separates how type identifiers are resolved from how intrinsic
calls are lowered. All information required to lower an intrinsic call
is stored in a new TypeIdLowering data structure. The idea is that this
data structure can either be initialized using the module itself during
regular LTO, or using the module summary in ThinLTO backends.
Original URL: https://reviews.llvm.org/D28341
Reviewers: pcc
Subscribers: mehdi_amini, llvm-commits
Differential Revision: https://reviews.llvm.org/D28532
llvm-svn: 291684
updated instructions:
pmulld, pmullw, pmulhw, mulsd, mulps, mulpd, divss, divps, divsd, divpd, addpd and subpd.
special optimization case which replaces pmulld with pmullw\pmulhw\pshuf seq.
In case if the real operands bitwidth <= 16.
Differential Revision: https://reviews.llvm.org/D28104
llvm-svn: 291657
These are interesting again because the user may not be aware that this
is a common reason preventing LICM.
A const is removed from an instruction pointer declaration in order to
pass it to ORE.
Differential Revision: https://reviews.llvm.org/D27940
llvm-svn: 291649
This patch reverts r291588: [PGO] Turn off comdat renaming in IR PGO by default,
as we are seeing some hash mismatches in our internal tests.
llvm-svn: 291621
Bail out instead of asserting when we encounter this situation,
which can actually happen.
The reason the test uses the new PM is that the "bad" phi, incidentally, gets
cleaned up by LoopSimplify. But LICM can create this kind of phi and preserve
loop simplify form, so the cleanup has no chance to run.
This fixes PR31190.
We may want to solve this in a less conservative manner, since this phi is
actually uniform within the inner loop (or we may want LICM to output a cleaner
promotion to begin with).
Differential Revision: https://reviews.llvm.org/D28490
llvm-svn: 291589
Summary:
In IR PGO we append the function hash to comdat functions to avoid the
potential hash mismatch. This turns out not legal in some cases: if the comdat
function is address-taken and used in comparison. Renaming changes the semantic.
This patch turns off comdat renaming by default.
To alleviate the hash mismatch issue, we now rename the profile variable
for comdat functions. Profile allows co-existing multiple versions of profiles
with different hash value. The inlined copy will always has the correct profile
counter. The out-of-line copy might not have the correct count. But we will
not have the bogus mismatch warning.
Reviewers: davidxl
Subscribers: llvm-commits, xur
Differential Revision: https://reviews.llvm.org/D28416
llvm-svn: 291588
In some cases StructurizeCfg updates root node, but dominator info
remains unchanges, it causes crash when expensive checks are enabled.
To cope with this problem a new method was added to DominatorTreeBase
that allows adding new root nodes, it is called in StructurizeCfg to
put dominator tree in sync.
This change fixes PR27488.
Differential Revision: https://reviews.llvm.org/D28114
llvm-svn: 291530
This patch delays the fix-up step for external induction variable users until
after the dominator tree has been properly updated. This should fix PR30742.
The SCEVExpander in InductionDescriptor::transform can generate code in the
wrong location if the dominator tree is not up-to-date. We should work towards
keeping the dominator tree up-to-date throughout the transformation.
Reference: https://llvm.org/bugs/show_bug.cgi?id=30742
Differential Revision: https://reviews.llvm.org/D28168
llvm-svn: 291462
Summary:
By using stripPointerCasts we can get to the root
value and then walk down the bitcast graph
Reviewers: reames
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D28181
llvm-svn: 291405
fabs(x * x) is not generally safe to assume x is positive if x is a NaN.
This is also less general than it could be, so this will be replaced
with a transformation on the intrinsic.
llvm-svn: 291359
Summary: LLVM's non-standard notion of phi nodes means we can't both try to substitute for undef in phi nodes *and* use phi nodes as leaders all the time. This changes NewGVN to use the same semantics as SimplifyPHINode to decide which phi nodes are equivalent.
Reviewers: davide
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D28312
llvm-svn: 291308
This is fixing a bug where Loop Vectorization is widening a load but
with a lower alignment. Hoisting the load without propagating the alignment
will allow inst-combine to later deduce a higher alignment that what the pointer
actually is.
Differential Revision: https://reviews.llvm.org/D28408
llvm-svn: 291281
This change separates how type identifiers are resolved from how intrinsic
calls are lowered. All information required to lower an intrinsic call
is stored in a new TypeIdLowering data structure. The idea is that this
data structure can either be initialized using the module itself during
regular LTO, or using the module summary in ThinLTO backends.
Differential Revision: https://reviews.llvm.org/D28341
llvm-svn: 291205
Promotion is always legal when a store within the loop is guaranteed to execute.
However, this is not a necessary condition - for promotion to be memory model
semantics-preserving, it is enough to have a store that dominates every exit
block. This is because if the store dominates every exit block, the fact the
exit block was executed implies the original store was executed as well.
Differential Revision: https://reviews.llvm.org/D28147
llvm-svn: 291171
This code seems to be target dependent which may not be the same for all targets.
Passed the decision whether the given stride is complex or not to the target by sending stride information via SCEV to getAddressComputationCost instead of 'IsComplex'.
Specifically at X86 targets we dont see any significant address computation cost in case of the strided access in general.
Differential Revision: https://reviews.llvm.org/D27518
llvm-svn: 291106
Set up basic YAML I/O support for module summaries, plumb the summary into
the pass and add a few command line flags to test YAML I/O support. Bitcode
support to come separately, as will the code in LowerTypeTests that actually
uses the summary. Also add a couple of tests that pass by virtue of the pass
doing nothing with the summary (which happens to be the correct thing to do
for those tests).
Differential Revision: https://reviews.llvm.org/D28041
llvm-svn: 291069
performing partial redundancy elimination (PRE). Not doing so can cause jumpy line
tables and confusing (though correct) source attributions.
Differential Revision: https://reviews.llvm.org/D27857
llvm-svn: 291037
We can perform the following:
(add (zext (add nuw X, C1)), C2) -> (zext (add nuw X, C1+C2))
This is only possible if C2 is negative and C2 is greater than or equal to negative C1.
llvm-svn: 290927
Summary:
Regardless how the loop body weight is distributed, we should preserve
total loop body weight. i.e. we should have same weight reaching the body of the loop
or its duplicates in peeled and unpeeled case.
Reviewers: mkuper, davidxl, anemet
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D28179
llvm-svn: 290833
Summary:
gep 0, 0 is equivalent to bitcast. LLVM canonicalizes it
to getelementptr because it make SROA can then handle it.
Simple case like
void g(A &a) {
z(a);
if (glob)
a.foo();
}
void testG() {
A a;
g(a);
}
was not devirtualized with -fstrict-vtable-pointers because luck of
handling for gep 0 in Memory Dependence Analysis
Reviewers: dberlin, nlewycky, chandlerc
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D28126
llvm-svn: 290763
This is similar to the allocfn case - if an alloca is not captured, then it's
necessarily thread-local.
Differential Revision: https://reviews.llvm.org/D28170
llvm-svn: 290738
Summary:
The current loop complete unroll algorithm checks if unrolling complete will reduce the runtime by a certain percentage. If yes, it will apply a fixed boosting factor to the threshold (by discounting cost). The problem for this approach is that the threshold abruptly. This patch makes the boosting factor a function of runtime reduction percentage, capped by a fixed threshold. In this way, the threshold changes continuously.
The patch also simplified the code by reducing one parameter in UP.
The patch only affects code-gen of two speccpu2006 benchmark:
445.gobmk binary size decreases 0.08%, no performance change.
464.h264ref binary size increases 0.24%, no performance change.
Reviewers: mzolotukhin, chandlerc
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D26989
llvm-svn: 290737
The accidentally had trivially dead code. Also needed to adjust the rounding mode to not CUR_DIRECTION so the intrinsics don't get converted to native operations before going through SimplifyDemandedVectorElts.
llvm-svn: 290702
This is an orthogonal and separated layer instead of being embedded
inside the pass manager. While it adds a small amount of complexity, it
is fairly minimal and the composability and control seems worth the
cost.
The logic for this ends up being nicely isolated and targeted. It should
be easy to experiment with different iteration strategies wrapped around
the CGSCC bottom-up walk using this kind of facility.
The mechanism used to track devirtualization is the simplest one I came
up with. I think it handles most of the cases the existing iteration
machinery handles, but I haven't done a *very* in depth analysis. It
does however match the basic intended semantics, and we can tweak or
tune its exact behavior incrementally as necessary. One thing that we
may want to revisit is freshly building the value handle set on each
iteration. While I don't think this will be a significant cost (it is
strictly fewer value handles but more churn of value handes than the old
call graph), it is conceivable that we'll want a somewhat more clever
tracking mechanism. My hope is to layer that on as a follow up patch
with data supporting any implementation complexity it adds.
This code also provides for a basic count heuristic: if the number of
indirect calls decreases and the number of direct calls increases for
a given function in the SCC, we assume devirtualization is responsible.
This matches the heuristics currently used in the legacy pass manager.
Differential Revision: https://reviews.llvm.org/D23114
llvm-svn: 290665
analyses when we're about to break apart an SCC.
We can't wait until after breaking apart the SCC to invalidate things:
1) Which SCC do we then invalidate? All of them?
2) Even if we invalidate all of them, a newly created SCC may not have
a proxy that will convey the invalidation to functions!
Previously we only invalidated one of the SCCs and too late. This led to
stale analyses remaining in the cache. And because the caching strategy
actually works, they would get used and chaos would ensue.
Doing invalidation early is somewhat pessimizing though if we *know*
that the SCC structure won't change. So it turns out that the design to
make the mutation API force the caller to know the *kind* of mutation in
advance was indeed 100% correct and we didn't do enough of it. So this
change also splits two cases of switching a call edge to a ref edge into
two separate APIs so that callers can clearly test for this and take the
easy path without invalidating when appropriate. This is particularly
important in this case as we expect most inlines to be between functions
in separate SCCs and so the common case is that we don't have to so
aggressively invalidate analyses.
The LCG API change in turn needed some basic cleanups and better testing
in its unittest. No interesting functionality changed there other than
more coverage of the returned sequence of SCCs.
While this seems like an obvious improvement over the current state, I'd
like to revisit the core concept of invalidating within the CG-update
layer at all. I'm wondering if we would be better served forcing the
callers to handle the invalidation beforehand in the cases that they
can handle it. An interesting example is when we want to teach the
inliner to *update and preserve* analyses. But we can cross that bridge
when we get there.
With this patch, the new pass manager an build all of the LLVM test
suite at -O3 and everything passes. =D I haven't bootstrapped yet and
I'm sure there are still plenty of bugs, but this gives a nice baseline
so I'm going to increasingly focus on fleshing out the missing
functionality, especially the bits that are just turned off right now in
order to let us establish this baseline.
llvm-svn: 290664
when they are call edges at the leaf but may (transitively) be reached
via ref edges.
It turns out there is a simple rule: insert everything as a ref edge
which is a safe conservative default. Then we let the existing update
logic handle promoting some of those to call edges.
Note that it would be fairly cheap to make these call edges right away
if that is desirable by testing whether there is some existing call path
from the source to the target. It just seemed like slightly more
complexity in this code path that isn't strictly necessary. If anyone
feels strongly about handling this differently I'm happy to change it.
llvm-svn: 290649
This adds a combine that canonicalizes a chain of inserts which broadcasts
a value into a single insert + a splat shufflevector.
This fixes PR31286.
Differential Revision: https://reviews.llvm.org/D27992
llvm-svn: 290641
most of the inliner test cases.
The inliner involves a bunch of interesting code and tends to be where
most of the issues I've seen experimenting with the new PM lie. All of
these test cases pass, but I'd like to keep some more thorough coverage
here so doing a fairly blanket enabling.
There are a handful of interesting tests I've not enabled yet because
they're focused on the always inliner, or on functionality that doesn't
(yet) exist in the inliner.
llvm-svn: 290592
skipping indirectly recursive inline chains.
To do this, we implicitly build an inline stack for each callsite and
check prior to inlining that doing so would not form a cycle. This uses
the exact same technique and even shares some code with the legacy PM
inliner.
This solution remains deeply unsatisfying to me because it means we
cannot actually iterate the inliner externally. Doing so would not be
able to easily detect and avoid such cycles. Some day I would very much
like to have a solution that works without this internal state to detect
cycles, but this is not that day.
llvm-svn: 290590
Nothing really interesting here, but I had to improve the test to use
variables rather than hard coding value names as we happen to end up
with different value names in the new PM.
llvm-svn: 290589
We currently ignore the `allocsize` attribute on functions calls with
the `nobuiltin` attribute when trying to lower `@llvm.objectsize`. We
shouldn't care about `nobuiltin` here: `allocsize` is explicitly added
by the user, not inferred based on a function's symbol.
llvm-svn: 290588
PMULDQ/PMULUDQ vXi64 instructions only use the even numbered v2Xi32 input elements which SimplifyDemandedVectorElts should try and use.
This builds on r290554 which added supported for 128 and 256-bit.
llvm-svn: 290582
This mostly involved converting from grep to FileCheck and tidying up
the IR used.
In one case (invoke_test-3.ll) the test had become completely pointless
as we use 'resume' rather than 'unwind' now, and even then it did not
occur at the end of the line.
llvm-svn: 290570
An earlier commit added support for unmasked scalar operations. At that time isel wouldn't generate an optimal sequence for masked operations, but that has now been fixed.
llvm-svn: 290566
inside of `InlineFunction`. Prior to this, call instructions are
specifically being rewritten and replaced within the inlined region,
invalidating some of the call sites.
Several of these regions are using the same technique to walk the
inlined region so this seems clearly safe up to this point.
I've also added a short circuit to the scan for call sites based on what
other code is doing.
With this, the most common crash I've found in the new inliner code is
fixed. I've turned it on for another test case that covers this
scenario.
I'll make my way through most of the other inliner test cases
just to get some easy coverage next.
llvm-svn: 290562
removing fully-dead comdats without removing dead entries in comdats
with live members.
This factors the core logic out of the current inliner's internals to
a reusable utility and leverages that in both places. The factored out
code should also be (minorly) more efficient in cases where we have very
few dead functions or dead comdats to consider.
I've added a test case to cover this behavior of the always inliner.
This is the last significant bug in the new PM's always inliner I've
found (so far).
llvm-svn: 290557
PMULDQ/PMULUDQ vXi64 instructions only use the even numbered v2Xi32 input elements which SimplifyDemandedVectorElts should try and use.
Differential Revision: https://reviews.llvm.org/D28119
llvm-svn: 290554
The current GVN algorithm folds unconditional branches to, it claims,
expose more PRE oportunities. The folding, if really needed,
(which is not sure, as it's not really proved it improves analysis)
can be done by an earlier cleanup pass instead of GVN itself.
Ack'ed/SGTM'd by Daniel Berlin.
Differential Revision: https://reviews.llvm.org/D28117
llvm-svn: 290546
systematically and document in the test what all is going on.
This replaces the PR-named test that was the only coverage for GlobalDCE
and comdats previously. I wrote this because I wasn't certain how
comdat DCE was supposed to work and wanted to step through what
GlobalDCE did to fully understand it. After talking to folks and reading
the code and really staring at things it all makes sense but it seemed
good to help write down some of this in a more explicit and fully
covering test case.
For example, it seemed like a bug that GlobalDCE didn't consider comdat
participation of ifuncs. Specifically it seemed like an accident because
testing didn't really cover that case. But in fact, ifuncs specifically
cannot participate in a comdat despite having that API. The new test
case covers this and explicitly documents that DCE gets to fire here
even though there are comdats involved.
Also, we didn't have any positive tests for the challenging cases such
as usage cycles between comdat participants that might make them seem
alive except that there is no external edge into the cycle.
llvm-svn: 290537
Summary:
I only do this for unmasked cases for now because isel is failing to fold the mask. I'll try to fix that soon.
I'll do the same thing for packed add/sub/mul/div in a future patch.
Reviewers: delena, RKSimon, zvi, craig.topper
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D27879
llvm-svn: 290535
Summary:
This patch adds support for converting the masked vpermv intrinsics into shufflevector instructions if the indices are constants.
We also need to wrap a select instruction around the shuffle to take care of the masking part. InstCombine will take care of optimizing the select if the mask is constant so I didn't bother checking for that.
Reviewers: zvi, delena, spatel, RKSimon
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D27825
llvm-svn: 290530
whether functions are removed, and fix the new PM's always inliner to
actually pass this test.
Without this, the new PM's always inliner leaves all the functions
kicking around which won't work out very well given the semantics of
always inline.
Doing this really highlights how frustrating the current alwaysinline
semantic contract is though -- why can we put it on *external*
functions, etc?
Also I've added a number of tricky and interesting test cases for
removing functions with the always inliner. There is one remaining case
not handled -- fully removing comdats -- and I've left a FIXME about
this.
llvm-svn: 290457
The pass creates some state which expects to be cleaned up by
a later instance of the same pass. opt-bisect happens to expose
this not ideal design because calling skipLoop() will result in
this state not being cleaned up at times and an assertion firing
in `doFinalization()`. Chandler tells me the new pass manager will
give us options to avoid these design traps, but until it's not ready,
we need a workaround for the current pass infrastructure. Fix provided
by Andy Kaylor, see the review for a complete discussion.
Differential Revision: https://reviews.llvm.org/D25848
llvm-svn: 290427
Use a dummy private function with inline asm calls instead of module
level asm blocks for CFI jumptables.
The main advantage is that now jumptable codegen can be affected by
the function attributes (like target_cpu on ARM). Module level asm
gets the default subtarget based on the target triple, which is often
not good enough.
This change also uses asm constraints/arguments to reference
jumptable targets and aliases directly. We no longer do asm name
mangling in an IR pass.
Differential Revision: https://reviews.llvm.org/D28012
llvm-svn: 290384
The code have been developed by Daniel Berlin over the years, and
the new implementation goal is that of addressing shortcomings of
the current GVN infrastructure, i.e. long compile time for large
testcases, lack of phi predication, no load/store value numbering
etc...
The current code just implements the "core" GVN algorithm, although
other pieces (load coercion, phi handling, predicate system) are
already implemented in a branch out of tree. Once the core is stable,
we'll start adding pieces on top of the base framework.
The test currently living in test/Transform/NewGVN are a copy
of the ones in GVN, with proper `XFAIL` (missing features in NewGVN).
A flag will be added in a future commit to enable NewGVN, so that
interested parties can exercise this code easily.
Differential Revision: https://reviews.llvm.org/D26224
llvm-svn: 290346
This patch renumbers the metadata nodes in debug info testcases after
https://reviews.llvm.org/D26769. This is a separate patch because it
causes so much churn. This was implemented with a python script that
pipes the testcases through llvm-as - | llvm-dis - and then goes
through the original and new output side-by side to insert all
comments at a close-enough location.
Differential Revision: https://reviews.llvm.org/D27765
llvm-svn: 290292
In r267672, where the loop distribution pragma was introduced, I tried
it hard to keep the old behavior for opt: when opt is invoked
with -loop-distribute, it should distribute the loop (it's off by
default when ran via the optimization pipeline).
As MichaelZ has discovered this has the unintended consequence of
breaking a very common developer work-flow to reproduce compilations
using opt: First you print the pass pipeline of clang
with -debug-pass=Arguments and then invoking opt with the returned
arguments.
clang -debug-pass will include -loop-distribute but the pass is invoked
with default=off so nothing happens unless the loop carries the pragma.
While through opt (default=on) we will try to distribute all loops.
This changes opt's default to off as well to match clang. The tests are
modified to explicitly enable the transformation.
llvm-svn: 290235
We're currently doing nearly the same thing for @llvm.objectsize in
three different places: two of them are missing checks for overflow,
and one of them could subtly break if InstCombine gets much smarter
about removing alloc sites. Seems like a good idea to not do that.
llvm-svn: 290214
This doesn't implement *every* feature of the existing inliner, but
tries to implement the most important ones for building a functional
optimization pipeline and beginning to sort out bugs, regressions, and
other problems.
Notable, but intentional omissions:
- No alloca merging support. Why? Because it isn't clear we want to do
this at all. Active discussion and investigation is going on to remove
it, so for simplicity I omitted it.
- No support for trying to iterate on "internally" devirtualized calls.
Why? Because it adds what I suspect is inappropriate coupling for
little or no benefit. We will have an outer iteration system that
tracks devirtualization including that from function passes and
iterates already. We should improve that rather than approximate it
here.
- Optimization remarks. Why? Purely to make the patch smaller, no other
reason at all.
The last one I'll probably work on almost immediately. But I wanted to
skip it in the initial patch to try to focus the change as much as
possible as there is already a lot of code moving around and both of
these *could* be skipped without really disrupting the core logic.
A summary of the different things happening here:
1) Adding the usual new PM class and rigging.
2) Fixing minor underlying assumptions in the inline cost analysis or
inline logic that don't generally hold in the new PM world.
3) Adding the core pass logic which is in essence a loop over the calls
in the nodes in the call graph. This is a bit duplicated from the old
inliner, but only a handful of lines could realistically be shared.
(I tried at first, and it really didn't help anything.) All told,
this is only about 100 lines of code, and most of that is the
mechanics of wiring up analyses from the new PM world.
4) Updating the LazyCallGraph (in the new PM) based on the *newly
inlined* calls and references. This is very minimal because we cannot
form cycles.
5) When inlining removes the last use of a function, eagerly nuking the
body of the function so that any "one use remaining" inline cost
heuristics are immediately refined, and queuing these functions to be
completely deleted once inlining is complete and the call graph
updated to reflect that they have become dead.
6) After all the inlining for a particular function, updating the
LazyCallGraph and the CGSCC pass manager to reflect the
function-local simplifications that are done immediately and
internally by the inline utilties. These are the exact same
fundamental set of CG updates done by arbitrary function passes.
7) Adding a bunch of test cases to specifically target CGSCC and other
subtle aspects in the new PM world.
Many thanks to the careful review from Easwaran and Sanjoy and others!
Differential Revision: https://reviews.llvm.org/D24226
llvm-svn: 290161
This patch implements PR31013 by introducing a
DIGlobalVariableExpression that holds a pair of DIGlobalVariable and
DIExpression.
Currently, DIGlobalVariables holds a DIExpression. This is not the
best way to model this:
(1) The DIGlobalVariable should describe the source level variable,
not how to get to its location.
(2) It makes it unsafe/hard to update the expressions when we call
replaceExpression on the DIGLobalVariable.
(3) It makes it impossible to represent a global variable that is in
more than one location (e.g., a variable with multiple
DW_OP_LLVM_fragment-s). We also moved away from attaching the
DIExpression to DILocalVariable for the same reasons.
This reapplies r289902 with additional testcase upgrades and a change
to the Bitcode record for DIGlobalVariable, that makes upgrading the
old format unambiguous also for variables without DIExpressions.
<rdar://problem/29250149>
https://llvm.org/bugs/show_bug.cgi?id=31013
Differential Revision: https://reviews.llvm.org/D26769
llvm-svn: 290153
Background/motivation - I was circling back around to:
https://llvm.org/bugs/show_bug.cgi?id=28296
I made a simple patch for that and noticed some regressions, so added test cases for
those with rL281055, and this is hopefully the minimal fix for just those cases.
But as you can see from the surrounding untouched folds, we are missing commuted patterns
all over the place, and of course there are no regression tests to cover any of those cases.
We could sprinkle "m_c_" dust all over this file and catch most of the missing folds, but
then we still wouldn't have test coverage, and we'd still miss some fraction of commuted
patterns because they require adjustments to the match order.
I'm aware of the concern about the potential compile-time performance impact of adding
matches like this (currently being discussed on llvm-dev), but I don't think there's any
evidence yet to suggest that handling commutative pattern matching more thoroughly is not
a worthwhile goal of InstCombine.
Differential Revision: https://reviews.llvm.org/D24419
llvm-svn: 290067
This is recommit of r287553 after fixing the invalid loop info after eliminating an empty block and unit test failures in AVR and WebAssembly :
Summary: Merging an empty case block into the header block of switch could cause ISel to add COPY instructions in the header of switch, instead of the case block, if the case block is used as an incoming block of a PHI. This could potentially increase dynamic instructions, especially when the switch is in a loop. I added a test case which was reduced from the benchmark I was targetting.
Reviewers: t.p.northover, mcrosier, manmanren, wmi, joerg, davidxl
Subscribers: joerg, qcolombet, danielcdh, hfinkel, mcrosier, llvm-commits
Differential Revision: https://reviews.llvm.org/D22696
llvm-svn: 289988
This reverts commit 289920 (again).
I forgot to implement a Bitcode upgrade for the case where a DIGlobalVariable
has not DIExpression. Unfortunately it is not possible to safely upgrade
these variables without adding a flag to the bitcode record indicating which
version they are.
My plan of record is to roll the planned follow-up patch that adds a
unit: field to DIGlobalVariable into this patch before recomitting.
This way we only need one Bitcode upgrade for both changes (with a
version flag in the bitcode record to safely distinguish the record
formats).
Sorry for the churn!
llvm-svn: 289982
This patch reapplies r289863. The original patch was reverted because it
exposed a bug causing the loop vectorizer to crash in the Python runtime on
PPC. The underlying issue was fixed with r289958.
llvm-svn: 289975
`dropUnknownNonDebugMetadata` takes a list of "known" metadata IDs. The
only reason it worked at all is that `getMetadataID` returns something
unrelated -- it returns the subclass ID of the receiver (which is used
in `dyn_cast` etc.). That does not numerically match
`LLVMContext::MD_invariant_group` and ends up dropping `invariant_group`
along with every other metadata that does not numerically match
`LLVMContext::MD_invariant_group`.
llvm-svn: 289973
After r288909, instructions feeding predicated instructions may be scalarized
if profitable. Since these instructions will remain scalar, we shouldn't
attempt to type-shrink them. We should only truncate vector types to their
minimal bit widths. This bug was exposed by enabling the vectorization of loops
containing conditional stores by default.
llvm-svn: 289958
This is recommit of r287553 after fixing the invalid loop info after eliminating an empty block:
Summary: Merging an empty case block into the header block of switch could cause ISel to add COPY instructions in the header of switch, instead of the case block, if the case block is used as an incoming block of a PHI. This could potentially increase dynamic instructions, especially when the switch is in a loop. I added a test case which was reduced from the benchmark I was targetting.
Reviewers: t.p.northover, mcrosier, manmanren, wmi, joerg, davidxl
Subscribers: joerg, qcolombet, danielcdh, hfinkel, mcrosier, llvm-commits
Differential Revision: https://reviews.llvm.org/D22696
llvm-svn: 289951
stores by default
This uncovers a crasher in the loop vectorizer on PPC when building the
Python runtime. I'll send the testcase to the review thread for the
original commit.
llvm-svn: 289934
This patch implements PR31013 by introducing a
DIGlobalVariableExpression that holds a pair of DIGlobalVariable and
DIExpression.
Currently, DIGlobalVariables holds a DIExpression. This is not the
best way to model this:
(1) The DIGlobalVariable should describe the source level variable,
not how to get to its location.
(2) It makes it unsafe/hard to update the expressions when we call
replaceExpression on the DIGLobalVariable.
(3) It makes it impossible to represent a global variable that is in
more than one location (e.g., a variable with multiple
DW_OP_LLVM_fragment-s). We also moved away from attaching the
DIExpression to DILocalVariable for the same reasons.
This reapplies r289902 with additional testcase upgrades.
<rdar://problem/29250149>
https://llvm.org/bugs/show_bug.cgi?id=31013
Differential Revision: https://reviews.llvm.org/D26769
llvm-svn: 289920
This patch implements PR31013 by introducing a
DIGlobalVariableExpression that holds a pair of DIGlobalVariable and
DIExpression.
Currently, DIGlobalVariables holds a DIExpression. This is not the
best way to model this:
(1) The DIGlobalVariable should describe the source level variable,
not how to get to its location.
(2) It makes it unsafe/hard to update the expressions when we call
replaceExpression on the DIGLobalVariable.
(3) It makes it impossible to represent a global variable that is in
more than one location (e.g., a variable with multiple
DW_OP_LLVM_fragment-s). We also moved away from attaching the
DIExpression to DILocalVariable for the same reasons.
<rdar://problem/29250149>
https://llvm.org/bugs/show_bug.cgi?id=31013
Differential Revision: https://reviews.llvm.org/D26769
llvm-svn: 289902
This pass prepares a module containing type metadata for ThinLTO by splitting
it into regular and thin LTO parts if possible, and writing both parts to
a multi-module bitcode file. Modules that do not contain type metadata are
written unmodified as a single module.
All globals with type metadata are added to the regular LTO module, and
the rest are added to the thin LTO module.
Differential Revision: https://reviews.llvm.org/D27324
llvm-svn: 289899
This patch sets the default value of the "-enable-cond-stores-vec" command line
option to "true".
Differential Revision: https://reviews.llvm.org/D27814
llvm-svn: 289863
Min/max canonicalization (r287585) exposes the fact that we're missing combines for min/max patterns.
This patch won't solve the example that was attached to that thread, so something else still needs fixing.
The line between InstCombine and InstSimplify gets blurry here because sometimes the icmp instruction that
we want to fold to already exists, but sometimes it's the swapped form of what we want.
Corresponding changes for smax/umin/umax to follow.
Differential Revision: https://reviews.llvm.org/D27531
llvm-svn: 289855
This is split out from D27696, since it turned out to be a bug fix and
not part of the NFC efficiency change.
Keep the same adjusted (possibly decayed) threshold in both the worklist
and the ImportList. Otherwise if we encountered it first along a cold
path, the callee would be added to the worklist with a lower decayed
threshold than when it is later encountered along a hot path. But the
logic uses the threshold recorded in the ImportList entry to check if
we should re-add it, and without this patch the threshold recorded there
is the same along both paths so we don't re-add it. Using the
same possibly decayed threshold in the ImportList ensures we re-add it
later with the higher non-decayed hot path threshold.
llvm-svn: 289843
A number of new patterns for simplifying and/xor of icmp:
(icmp ne %x, 0) ^ (icmp ne %y, 0) => icmp ne %x, %y if the following is true:
1- (%x = and %a, %mask) and (%y = and %b, %mask)
2- %mask is a power of 2.
(icmp eq %x, 0) & (icmp ne %y, 0) => icmp ult %x, %y if the following is true:
1- (%x = and %a, %mask1) and (%y = and %b, %mask2)
2- Let %t be the smallest power of 2 where %mask1 & %t != 0. Then for any
%s that is a power of 2 and %s & %mask2 != 0, we must have %s <= %t.
For example if %mask1 = 24 and %mask2 = 16, setting %s = 16 and %t = 8
violates condition (2) above. So this optimization cannot be applied.
llvm-svn: 289813
After r289755, the AssumptionCache is no longer needed. Variables affected by
assumptions are now found by using the new operand-bundle-based scheme. This
new scheme is more computationally efficient, and also we need much less
code...
llvm-svn: 289756
There was an efficiency problem with how we processed @llvm.assume in
ValueTracking (and other places). The AssumptionCache tracked all of the
assumptions in a given function. In order to find assumptions relevant to
computing known bits, etc. we searched every assumption in the function. For
ValueTracking, that means that we did O(#assumes * #values) work in InstCombine
and other passes (with a constant factor that can be quite large because we'd
repeat this search at every level of recursion of the analysis).
Several of us discussed this situation at the last developers' meeting, and
this implements the discussed solution: Make the values that an assume might
affect operands of the assume itself. To avoid exposing this detail to
frontends and passes that need not worry about it, I've used the new
operand-bundle feature to add these extra call "operands" in a way that does
not affect the intrinsic's signature. I think this solution is relatively
clean. InstCombine adds these extra operands based on what ValueTracking, LVI,
etc. will need and then those passes need only search the users of the values
under consideration. This should fix the computational-complexity problem.
At this point, no passes depend on the AssumptionCache, and so I'll remove
that as a follow-up change.
Differential Revision: https://reviews.llvm.org/D27259
llvm-svn: 289755
Summary: SampleProfileLoader pass may be invoked twice by LTO. The 2nd pass should not append more summary info as it is already preset by the 1st pass.
Reviewers: eraman, davidxl
Subscribers: mehdi_amini, llvm-commits
Differential Revision: https://reviews.llvm.org/D27733
llvm-svn: 289725
Summary:
Move GVNHoist to later in the optimization pipeline, specifically, to
the function simplification part of the pipeline. The new pipeline
location allows GVNHoist to run on a function after its callees have
been inlined but before the function has been considered for inlining
into its callers, exposing more opportunities for hoisting.
Performance results on AArch64 kryo:
Improvements:
Benchmarks/CoyoteBench/fftbench -24.952%
spec2006/bzip2 -4.071%
internal bmark -3.177%
Benchmarks/PAQ8p/paq8p -1.754%
spec2000/perlbmk -1.328%
spec2006/h264ref -1.140%
Regressions:
internal bmark +1.818%
Benchmarks/mafft/pairlocalalign +1.084%
Reviewers: sebpop, dberlin, hiraditya
Subscribers: aemerson, mehdi_amini, mcrosier, llvm-commits
Differential Revision: https://reviews.llvm.org/D27722
llvm-svn: 289696
Summary:
This patch will add loop metadata on the pre and post loops generated by IRCE.
Currently, we have metadata for disabling optimizations such as vectorization,
unrolling, loop distribution and LICM versioning (and confirmed that these
optimizations check for the metadata before proceeding with the transformation).
The pre and post loops generated by IRCE need not go through loop opts (since
these are slow paths).
Added two test cases as well.
Reviewers: sanjoy, reames
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D26806
llvm-svn: 289588
We currently check if the exact trip count is known and is smaller than the
"tiny loop" bound. We should be checking the maximum bound on the trip count
instead.
Differential Revision: https://reviews.llvm.org/D27690
llvm-svn: 289583
Summary:
This is last in of a series of patches to evolve ADCE.cpp to support
removing of unnecessary control flow.
This patch adds the code to update the control and data flow graphs
to remove the dead control flow.
Also update unit tests to test the capability to remove dead,
may-be-infinite loop which is enabled by the switch
-adce-remove-loops.
Previous patches:
D23824 [ADCE] Add handling of PHI nodes when removing control flow
D23559 [ADCE] Add control dependence computation
D23225 [ADCE] Modify data structures to support removing control flow
D23065 [ADCE] Refactor anticipating new functionality (NFC)
D23102 [ADCE] Refactoring for new functionality (NFC)
Reviewers: dberlin, majnemer, nadav, mehdi_amini
Subscribers: llvm-commits, david2050, freik, twoh
Differential Revision: https://reviews.llvm.org/D24918
llvm-svn: 289548
Only the lower bits of the input element are used. And only the lower element can be undef since the upper bits are zeroed.
Have InstCombineCalls call SimplifyDemandedVectorElts for these intrinsics to reuse this support.
llvm-svn: 289523
Summary:
Since we don't break BBs for function calls. We might get some insane counts
(wrap of unsigned) in the presence of noreturn calls.
This patch sets these counts to zero instead of the wrapped number.
Reviewers: davidxl
Subscribers: xur, eraman, llvm-commits
Differential Revision: https://reviews.llvm.org/D27602
llvm-svn: 289521
This patch ensures the correct minimum bit width during type-shrinking.
Previously when type-shrinking, we always sign-extended values back to their
original width. However, if we are going to sign-extend, and the sign bit is
unknown, we have to increase the minimum bit width by one bit so the
sign-extend will fill the upper bits correctly. If the sign bit is known to be
zero, we can perform a zero-extend instead. This should fix PR31243.
Reference: https://llvm.org/bugs/show_bug.cgi?id=31243
Differential Revision: https://reviews.llvm.org/D27466
llvm-svn: 289470
Reverts r289412. It caused an OOB PHI operand access in instcombine when
ASan is enabled. Reduction in progress.
Also reverts "[SCEVExpander] Add a test case related to r289412"
llvm-svn: 289453
We could truncate the condition and then try to fold the add into the
original condition value causing wrong case constants to be used.
Move the offset transform ahead of the truncate transform and return
after each transform, so there's no chance of getting confused values.
Fix for:
https://llvm.org/bugs/show_bug.cgi?id=31260
llvm-svn: 289442
SCEVExpand computes the insertion point for the components of a SCEV to be code
generated. When it comes to generating code for a division, SCEVexpand would
not be able to check (at compilation time) all the conditions necessary to avoid
a division by zero. The patch disables hoisting of expressions containing
divisions by anything other than non-zero constants in order to avoid hoisting
these expressions past conditions that should hold before doing the division.
The patch passes check-all on x86_64-linux.
Differential Revision: https://reviews.llvm.org/D27216
llvm-svn: 289412
Summary:
This change adds some verification in the IR verifier around struct path
TBAA metadata.
Other than some basic sanity checks (e.g. we get constant integers where
we expect constant integers), this checks:
- That by the time an struct access tuple `(base-type, offset)` is
"reduced" to a scalar base type, the offset is `0`. For instance, in
C++ you can't start from, say `("struct-a", 16)`, and end up with
`("int", 4)` -- by the time the base type is `"int"`, the offset
better be zero. In particular, a variant of this invariant is needed
for `llvm::getMostGenericTBAA` to be correct.
- That there are no cycles in a struct path.
- That struct type nodes have their offsets listed in an ascending
order.
- That when generating the struct access path, you eventually reach the
access type listed in the tbaa tag node.
Reviewers: dexonsmith, chandlerc, reames, mehdi_amini, manmanren
Subscribers: mcrosier, llvm-commits
Differential Revision: https://reviews.llvm.org/D26438
llvm-svn: 289402
This teaches SimplifyDemandedElts that the FMA can be removed if the lower element isn't used. It also teaches it that if upper elements of the first operand aren't used then we can simplify them.
llvm-svn: 289377
The motivating example is:
extern int patatino;
int goo() {
int x = 0;
for (int i = 0; i < 1000000; ++i) {
x *= patatino;
}
return x;
}
Currently SCCP will not realize that this function returns always zero,
therefore will try to unroll and vectorize the loop at -O3 producing an
awful lot of (useless) code. With this change, it will just produce:
0000000000000000 <g>:
xor %eax,%eax
retq
llvm-svn: 289175
Summary:
Attaching !absolute_symbol to a global variable does two things:
1) Marks it as an absolute symbol reference.
2) Specifies the value range of that symbol's address.
Teach the X86 backend to allow absolute symbols to appear in place of
immediates by extending the relocImm and mov64imm32 matchers. Start using
relocImm in more places where it is legal.
As previously proposed on llvm-dev:
http://lists.llvm.org/pipermail/llvm-dev/2016-October/105800.html
Differential Revision: https://reviews.llvm.org/D25878
llvm-svn: 289087
When trying to vectorize trees that start at insertelement instructions
function tryToVectorizeList() uses vectorization factor calculated as
MinVecRegSize/ScalarTypeSize. But sometimes it does not work as tree
cost for this fixed vectorization factor is too high.
Patch tries to improve the situation. It tries different vectorization
factors from max(PowerOf2Floor(NumberOfVectorizedValues),
MinVecRegSize/ScalarTypeSize) to MinVecRegSize/ScalarTypeSize and tries
to choose the best one.
Differential Revision: https://reviews.llvm.org/D27215
llvm-svn: 289043
Replace @progbits in the section directive with %progbits, because "@" starts a comment on arm/thumb.
Use b.w branch instruction.
Use .thumb_function and .thumb_set for proper arm/thumb interwork. This way jumptable entry addresses on thumb have bit 0 set (correctly). This does not affect CFI check math, because the address of the jumptable start also has that bit set.
This does not work on thumbv5, because it does not support b.w, and the linker would not insert a veneer (trampoline?) to extend the range of b.n. We may need to do full-range plt-style jumptables on thumbv54, which are 12 bytes per entry. Another option is "push lr; bl; pop pc" (4 bytes) but that needs unwinding instructions, etc.
Differential Revision: https://reviews.llvm.org/D27499
llvm-svn: 289008
The fix committed in r288851 doesn't cover all the cases.
In particular, if we have an instruction with side effects
which has a no non-dbg use not depending on the bits, we still
perform RAUW destroying the dbg.value's first argument.
Prevent metadata from being replaced here to avoid the issue.
Differential Revision: https://reviews.llvm.org/D27534
llvm-svn: 288987
The tests that already work are folded in InstSimplify, so those
tests should be redundant and we can remove them if they don't
seem worthwhile for completeness.
llvm-svn: 288957
This patch attempts to scalarize the operand expressions of predicated
instructions if they were conditionally executed in the original loop. After
scalarization, the expressions will be sunk inside the blocks created for the
predicated instructions. The transformation essentially performs
un-if-conversion on the operands.
The cost model has been updated to determine if scalarization is profitable. It
compares the cost of a vectorized instruction, assuming it will be
if-converted, to the cost of the scalarized instruction, assuming that the
instructions corresponding to each vector lane will be sunk inside a predicated
block, possibly avoiding execution. If it's more profitable to scalarize the
entire expression tree feeding the predicated instruction, the expression will
be scalarized; otherwise, it will be vectorized. We only consider the cost of
the entire expression to accurately estimate the cost of the required
insertelement and extractelement instructions.
Differential Revision: https://reviews.llvm.org/D26083
llvm-svn: 288909
In the case of a fully redundant load LI dominated by an equivalent load V, GVN
should always preserve the original debug location of V. Otherwise, we risk to
introduce an incorrect stepping.
If V has debug info, then clearly it should not be modified. If V has a null
debugloc, then it is still potentially incorrect to propagate LI's debugloc
because LI may not post-dominate V.
Differential Revision: https://reviews.llvm.org/D27468
llvm-svn: 288903
As Eli noted in the post-commit thread for r288833, the use of
swapOperands() may not be allowed in InstSimplify, so I'm
removing those calls here pending further review.
The swap mutates the icmp, and there doesn't appear to be precedent
for instruction mutation in InstSimplify.
I didn't actually have any tests for those cases, so I'm adding
a few here.
llvm-svn: 288855
BDCE has two phases:
1. It asks SimplifyDemandedBits if all the bits of an instruction are dead, and if so,
replaces all its uses with the constant zero.
2. Then, it asks SimplifyDemandedBits again if the instruction is really dead
(no side effects etc..) and if so, eliminates it.
Now, in 1) if all the bits of an instruction are dead, we may end up replacing a dbg use:
%call = tail call i32 (...) @g() #4, !dbg !15
tail call void @llvm.dbg.value(metadata i32 %call, i64 0, metadata !8, metadata !16), !dbg !17
->
%call = tail call i32 (...) @g() #4, !dbg !15
tail call void @llvm.dbg.value(metadata i32 0, i64 0, metadata !8, metadata !16), !dbg !17
but not eliminating the call because it may have arbitrary side effects.
In other words, we lose some debug informations.
This patch fixes the problem making sure that BDCE does nothing with the instruction if
it has side effects and no non-dbg uses.
Differential Revision: https://reviews.llvm.org/D27471
llvm-svn: 288851
All of these (and a few more) are already handled by InstCombine,
but we shouldn't have to wait until then to simplify these because
they're cheap to deal with here in InstSimplify.
This is the 'and' sibling of the earlier 'or' patch:
https://reviews.llvm.org/rL288833
llvm-svn: 288841
All of these (and a few more) are already handled by InstCombine,
but we shouldn't have to wait until then to simplify these because
they're cheap to deal with here in InstSimplify.
llvm-svn: 288833
Summary:
If LAA expands a bound that is loop invariant, but not hoisted out
of the loop body, it used to use that value anyway, causing a
non-domination error, because the memcheck block is of course not
dominated by the scalar loop body. Detect this situation and expand
the SCEV expression instead.
Fixes PR31251
Reviewers: anemet
Subscribers: mzolotukhin, llvm-commits
Differential Revision: https://reviews.llvm.org/D27397
llvm-svn: 288705
so we can stop using DW_OP_bit_piece with the wrong semantics.
The entire back story can be found here:
http://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20161114/405934.html
The gist is that in LLVM we've been misinterpreting DW_OP_bit_piece's
offset field to mean the offset into the source variable rather than
the offset into the location at the top the DWARF expression stack. In
order to be able to fix this in a subsequent patch, this patch
introduces a dedicated DW_OP_LLVM_fragment operation with the
semantics that we used to apply to DW_OP_bit_piece, which is what we
actually need while inside of LLVM. This patch is complete with a
bitcode upgrade for expressions using the old format. It does not yet
fix the DWARF backend to use DW_OP_bit_piece correctly.
Implementation note: We discussed several options for implementing
this, including reserving a dedicated field in DIExpression for the
fragment size and offset, but using an custom operator at the end of
the expression works just fine and is more efficient because we then
only pay for it when we need it.
Differential Revision: https://reviews.llvm.org/D27361
rdar://problem/29335809
llvm-svn: 288683
This solves a secondary problem seen in PR6137:
https://llvm.org/bugs/show_bug.cgi?id=6137#c6
This is similar to the bitwise logic op fold added with:
https://reviews.llvm.org/rL287707
And like that patch, I'm artificially restricting the
transform from vector <-> scalar types until we're sure
that the backend can handle that.
llvm-svn: 288584
VSX has instructions lxsiwax/lxsdx that can load 32/64 bit value into VSX register cheaply. That patch makes it known to memory cost model, so the vectorization of the test case in pr30990 is beneficial.
Differential Revision: https://reviews.llvm.org/D26713
llvm-svn: 288560
For -O0 there might be unreachable BBs, which breaks the assumption that all the
BBs have an auxiliary data structure. In this patch, we add another interface
called findBBInfo() so that a nullptr can be returned for the unreachable BBs
(and the callers can ignore those BBs).
This fixes the bug reported
https://llvm.org/bugs/show_bug.cgi?id=31209
Differential Revision: https://reviews.llvm.org/D27280
llvm-svn: 288528
This reverts commit r288497, as it broke the AArch64 build of Compiler-RT's
builtins (twice: once in r288412 and once in r288497). We should investigate
this offline.
llvm-svn: 288508
When trying to vectorize trees that start at insertelement instructions
function tryToVectorizeList() uses vectorization factor calculated as
MinVecRegSize/ScalarTypeSize. But sometimes it does not work as tree
cost for this fixed vectorization factor is too high.
Patch tries to improve the situation. It tries different vectorization
factors from max(PowerOf2Floor(NumberOfVectorizedValues),
MinVecRegSize/ScalarTypeSize) to MinVecRegSize/ScalarTypeSize and tries
to choose the best one.
Differential Revision: https://reviews.llvm.org/D27215
llvm-svn: 288497
The instcombine code which folds loads and stores into their use types can trip up if the use is a bitcast to a type which we can't directly load or store in the IR. In principle, such types shouldn't exist, but in practice they do today. This is a workaround to avoid a bug while we work towards the long term goal.
Differential Revision: https://reviews.llvm.org/D24365
llvm-svn: 288415
When trying to vectorize trees that start at insertelement instructions
function tryToVectorizeList() uses vectorization factor calculated as
MinVecRegSize/ScalarTypeSize. But sometimes it does not work as tree
cost for this fixed vectorization factor is too high.
Patch tries to improve the situation. It tries different vectorization
factors from max(PowerOf2Floor(NumberOfVectorizedValues),
MinVecRegSize/ScalarTypeSize) to MinVecRegSize/ScalarTypeSize and tries
to choose the best one.
Differential Revision: https://reviews.llvm.org/D27215
llvm-svn: 288412
Currently when cost of scalar operations is evaluated the vector type is
used for scalar operations. Patch fixes this issue and fixes evaluation
of the vector operations cost.
Several test showed that vector cost model is too optimistic. It
allowed vectorization of 8 or less add/fadd operations, though scalar
code is faster. Actually, only for 16 or more operations vector code
provides better performance.
Differential Revision: https://reviews.llvm.org/D26277
llvm-svn: 288398
[recommitting after the fix in r288307]
This requires some changes to the opt-diag API. Hal and I have
discussed this at the Dev Meeting and came up with a streaming delimiter
(setExtraArgs) to solve this.
Arguments after this delimiter are only included in the optimization
records and not in the remarks printed in the compiler output. (Note,
how in the test the content of the YAML file changes but the remarks on
the compiler output don't.)
This implements the green GVN message with a bug fix at line
http://lab.llvm.org:8080/artifacts/opt-view_test-suite/build/SingleSource/Benchmarks/Dhrystone/CMakeFiles/dry.dir/html/_org_test-suite_SingleSource_Benchmarks_Dhrystone_dry.c.html#L446
The fix is that now we properly include the constant value in the
message: "load of type i32 eliminated in favor of 7"
Differential Revision: https://reviews.llvm.org/D26489
llvm-svn: 288380
If LoopInfo is available during GVN, BasicAA will use it. However
MergeBlockIntoPredecessor does not update LI as it merges blocks.
This didn't use to cause problems because LI was freed before
GVN/BasicAA. Now with OptimizationRemarkEmitter, the lifetime of LI is
extended so LI needs to be kept up-to-date during GVN.
Differential Revision: https://reviews.llvm.org/D27288
llvm-svn: 288307
This implements PGO-driven loop peeling.
The basic idea is that when the average dynamic trip-count of a loop is known,
based on PGO, to be low, we can expect a performance win by peeling off the
first several iterations of that loop.
Unlike unrolling based on a known trip count, or a trip count multiple, this
doesn't save us the conditional check and branch on each iteration. However,
it does allow us to simplify the straight-line code we get (constant-folding,
etc.). This is important given that we know that we will usually only hit this
code, and not the actual loop.
This is currently disabled by default.
Differential Revision: https://reviews.llvm.org/D25963
llvm-svn: 288274
Michel Dänzer reported that r288051, "[StructurizeCFG] Use range-based
for loops", introduced a bug into rebuildSSA, wherein we were iterating
over an instruction's use list while modifying it, without taking care
to do this correctly.
llvm-svn: 288200
Currently SLP vectorizer tries to vectorize a binary operation and dies
immediately after unsuccessful the first unsuccessfull attempt. Patch
tries to improve the situation, trying to vectorize all binary
operations of all children nodes in the binop tree.
Differential Revision: https://reviews.llvm.org/D25517
llvm-svn: 288115
Preserving lifetime markers isn't as important as allowing promotion,
so just drop the lifetime markers if necessary.
This also fixes an assertion failure where other parts of SROA assumed
that lifetime markers never block promotion.
Fixes https://llvm.org/bugs/show_bug.cgi?id=29139.
Differential Revision: https://reviews.llvm.org/D24854
llvm-svn: 288074
In r286814, the algorithm for calculating inline costs changed. This
caused more inlining to take place which is especially apparent
in optsize and minsize modes.
As the cost calculation removed a skewed behaviour (we were inconsistent
about the cost of calls) it isn't possible to update the thresholds to
get exactly the same behaviour as before. However, this threshold change
accounts for the very common case where an inline candidate has no
calls within it. In this case, r286814 would inline around 5-6 more (IR)
instructions.
The changes to -Oz have been heavily benchmarked. The "obvious" value
for the inline threshold at -Oz is zero, but due to inaccuracies in the
inline heuristics this can actually cause code size increases due to
not inlining key thunk functions (that then disappear). Experimentally,
5 was the sweet spot for code size over the test-suite.
For -Os, this change removes the outlier results shown up by green dragon
(http://104.154.54.203/db_default/v4/nts/13248).
Fixes D26848.
llvm-svn: 288024
Summary:
The iterative algorithm for Loop Unswitching may render some of the branches unreachable in the unswitched loops.
Given the exponential nature of the algorithm, this is quite an overhead.
This patch fixes this problem by selectively unswitching only those branches within a loop that are reachable from the loop header.
Reviewers: Michael Zolothukin, Anna Thomas, Weiming Zhao.
Subscribers: llvm-commits.
Differential Revision: http://reviews.llvm.org/D26299
llvm-svn: 287925
Summary:
The "getVectorizablePrefix" method would give up if it found an aliasing load for a store chain.
In practice, the aliasing load can be treated as a memory barrier and all stores that precede it
are a valid vectorizable prefix.
Issue found by volkan in D26962. Testcase is a pruned version of the one in the original patch.
Reviewers: jlebar, arsenm, tstellarAMD
Subscribers: mzolotukhin, wdng, nhaehnle, anna, volkan, llvm-commits
Differential Revision: https://reviews.llvm.org/D27008
llvm-svn: 287781
Without this test, you can just remove the code fixing the
switch to the first constant in ResolvedUndefs in and everything
pass. This test, instead, fails with an assertion if the code
is removed. Found while refactoring SCCP to integrate undef in
the solver.
llvm-svn: 287731
We visit and/or, we try to derive a lattice value for the
instruction even if one of the operands is overdefined.
If the non-overdefined value is still 'unknown' just return and wait
for ResolvedUndefsIn to "plug in" the correct value. This simplifies
the logic a bit. While I'm here add tests for missing cases.
llvm-svn: 287709
In PR27925:
https://llvm.org/bugs/show_bug.cgi?id=27925
...we proposed adding this fold to eliminate a bitcast. In D20774, there was
some concern about changing the type of a bitwise op as well as creating
bitcasts that might not be free for a target. However, if we're strictly
eliminating an instruction (by limiting this to one-use ops), then we should
be able to do this in InstCombine.
But we're cautiously restricting the transform for now to vector types to
avoid possible backend problems. A transform to make sure the logic op is
legal for the target should be added to reverse this transform and improve
codegen.
Differential Revision: https://reviews.llvm.org/D26641
llvm-svn: 287707
Summary:
Previously, CGP would unconditionally sink addrspacecast instructions,
even going so far as to sink them into a loop.
Now we check that the cast is "cheap", as defined by TLI.
We introduce a new "is-cheap" function to TLI rather than using
isNopAddrSpaceCast because some GPU platforms want the ability to ask
for non-nop casts to be sunk.
Reviewers: arsenm, tra
Subscribers: jholewinski, wdng, llvm-commits
Differential Revision: https://reviews.llvm.org/D26923
llvm-svn: 287591
Allow using an instruction other than a mul or phi as the base for
root-finding. For example, the included testcase includes a loop
which requires using a getelementptr as the base for root-finding.
Differential Revision: https://reviews.llvm.org/D26529
llvm-svn: 287588
This is a first step towards canonicalization and improved folding/codegen
for integer min/max as discussed here:
http://lists.llvm.org/pipermail/llvm-dev/2016-November/106868.html
Here, we're just matching the simplest min/max patterns and adjusting the
icmp predicate while swapping the select operands.
I've included FIXME tests in test/Transforms/InstCombine/select_meta.ll
so it's easier to see how this might be extended (corresponds to the TODO
comment in the code). That's also why I'm using matchSelectPattern()
rather than a simpler check; once the backend is patched, we can just
remove some of the restrictions to allow the obfuscated min/max patterns
in the FIXME tests to be matched.
Differential Revision: https://reviews.llvm.org/D26525
llvm-svn: 287585
Summary:
D26704 fixed the non-determinism in codegen by sorting basic blocks before
iteration so as to have a defined iteration order. As a result we need to fix
the names (numbers) of the temporaries in the following unit tests:
test/Transforms/Util/MemorySSA/multi-edges.ll
test/Transforms/Util/MemorySSA/multiple-backedges-hal.ll
Reviewers: dberlin, david2050, mgrang
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D26926
llvm-svn: 287575
This patch fixes the non-determinism caused due to iterating SmallPtrSet's
which was uncovered due to the experimental "reverse iteration order " patch:
https://reviews.llvm.org/D26718
The following unit tests failed because of the undefined order of iteration.
LLVM :: Transforms/Util/MemorySSA/cyclicphi.ll
LLVM :: Transforms/Util/MemorySSA/many-dom-backedge.ll
LLVM :: Transforms/Util/MemorySSA/many-doms.ll
LLVM :: Transforms/Util/MemorySSA/phi-translation.ll
Reviewers: dberlin, mgrang
Subscribers: dberlin, llvm-commits, david2050
Differential Revision: https://reviews.llvm.org/D26704
llvm-svn: 287563
Summary: Merging an empty case block into the header block of switch could cause
ISel to add COPY instructions in the header of switch, instead of the case
block, if the case block is used as an incoming block of a PHI. This could
potentially increase dynamic instructions, especially when the switch is in a
loop. I added a test case which was reduced from the benchmark I was targetting.
Reviewers: t.p.northover, mcrosier, manmanren, wmi, davidxl
Subscribers: qcolombet, danielcdh, hfinkel, mcrosier, llvm-commits
Differential Revision: https://reviews.llvm.org/D22696
llvm-svn: 287553
Currently LLVM assumes that a pointer addrspacecasted to a different addr space is equivalent to trunc or zext bitwise, which is not true. For example, in amdgcn target, when a null pointer is addrspacecasted from addr space 4 to 0, its value is changed from i64 0 to i32 -1.
This patch teaches LLVM not to assume known bits of addrspacecast instruction to its operand.
Differential Revision: https://reviews.llvm.org/D26803
llvm-svn: 287545
This is a prerequisite patch for D26556:
https://reviews.llvm.org/D26556
...because there was no direct coverage for these folds (which in some cases are adding instructions).
llvm-svn: 287400
insertUniqueBackedgeBlock in lib/Transforms/Utils/LoopSimplify.cpp now
propagates existing llvm.loop metadata to newly the added backedge.
llvm::TryToSimplifyUncondBranchFromEmptyBlock in lib/Transforms/Utils/Local.cpp
now propagates existing llvm.loop metadata to the branch instructions in the
predecessor blocks of the empty block that is removed.
Differential Revision: https://reviews.llvm.org/D26495
llvm-svn: 287341
This is a straightforward extension of the existing support for 32/64-bit element types. Just needed to add the additional instrinsics to the switches.
llvm-svn: 287316
Summary:
This extends FCOPYSIGN support to 512-bit vectors.
I've also added tests to show what the 128-bit and 256-bit cases look like with broadcast loads.
Reviewers: delena, zvi, RKSimon, spatel
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D26791
llvm-svn: 287298
Summary:
For flat loop, even if it is hot, it is not a good idea to unroll in runtime, thus we set a lower partial unroll threshold.
For hot loop, we set a higher unroll threshold and allows expensive tripcount computation to allow more aggressive unrolling.
Reviewers: davidxl, mzolotukhin
Subscribers: sanjoy, mehdi_amini, llvm-commits
Differential Revision: https://reviews.llvm.org/D26527
llvm-svn: 287186
This pass splits globals into elements using inrange annotations on
getelementptr indices.
Differential Revision: https://reviews.llvm.org/D22295
llvm-svn: 287178
Summary: These intrinsics have been unused for clang for a while. This patch removes them. We auto upgrade them to extractelements, a scalar operation and then an insertelement. This matches the sequence used by clangs intrinsic file.
Reviewers: zvi, delena, RKSimon
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D26660
llvm-svn: 287083
Summary:
We don't do BypassSlowDivision when the denominator is a constant, but
we do do it when the numerator is a constant.
This patch makes two related changes to BypassSlowDivision when the
numerator is a constant:
* If the numerator is too large to fit into the bypass width, don't
bypass slow division (because we'll never run the smaller-width
code).
* If we bypass slow division where the numerator is a constant, don't
OR together the numerator and denominator when determining whether
both operands fit within the bypass width. We need to check only the
denominator.
Reviewers: tra
Subscribers: llvm-commits, jholewinski
Differential Revision: https://reviews.llvm.org/D26699
llvm-svn: 287062
In RateRegister of existing LSR, if a formula contains a Reg which is a SCEVAddRecExpr,
and this SCEVAddRecExpr's loop is an outerloop, the formula will be marked as Loser
and dropped.
Suppose we have an IR that %for.body is outerloop and %for.body2 is innerloop. LSR only
handle inner loop now so only %for.body2 will be handled.
Using the logic above, formula like
reg(%array) + reg({1,+, %size}<%for.body>) + 1*reg({0,+,1}<%for.body2>) will be dropped
no matter what because reg({1,+, %size}<%for.body>) is a SCEVAddRecExpr type reg related
with outerloop. Only formula like
reg(%array) + 1*reg({{1,+, %size}<%for.body>,+,1}<nuw><nsw><%for.body2>) will be kept
because the SCEVAddRecExpr related with outerloop is folded into the initial value of the
SCEVAddRecExpr related with current loop.
But in some cases, we do need to share the basic induction variable
reg{0 ,+, 1}<%for.body2> among LSR Uses to reduce the final total number of induction
variables used by LSR, so we don't want to drop the formula like
reg(%array) + reg({1,+, %size}<%for.body>) + 1*reg({0,+,1}<%for.body2>) unconditionally.
From the existing comment, it tries to avoid considering multiple level loops at the same time.
However, existing LSR only handles innermost loop, so for any SCEVAddRecExpr with a loop other
than current loop, it is an invariant and will be simple to handle, and the formula doesn't have
to be dropped.
Differential Revision: https://reviews.llvm.org/D26429
llvm-svn: 286999
When both WidenIV::getWideRecurrence and WidenIV::getExtendedOperandRecurrence
return non-null but different WideAddRec, if getWideRecurrence is called
before getExtendedOperandRecurrence, we won't bother to call
getExtendedOperandRecurrence again. But As we know it is possible that after
SCEV folding, we cannot prove the legality using the SCEVAddRecExpr returned
by getWideRecurrence. Meanwhile if getExtendedOperandRecurrence returns non-null
WideAddRec, we know for sure that it is legal to do widening for current instruction.
So it is better to put getExtendedOperandRecurrence before getWideRecurrence, which
will increase the chance of successful widening.
Differential Revision: https://reviews.llvm.org/D26059
llvm-svn: 286987
The register usage algorithm incorrectly treats instructions whose value is
not used within the loop (e.g. those that do not produce a value).
The algorithm first calculates the usages within the loop. It iterates over
the instructions in order, and records at which instruction index each use
ends (in fact, they're actually recorded against the next index, as this is
when we want to delete them from the open intervals).
The algorithm then iterates over the instructions again, adding each
instruction in turn to a list of open intervals. Instructions are then
removed from the list of open intervals when they occur in the list of uses
ended at the current index.
The problem is, instructions which are not used in the loop are skipped.
However, although they aren't used, the last use of a value may have been
recorded against that instruction index. In this case, the use is not deleted
from the open intervals, which may then bump up the estimated register usage.
This patch fixes the issue by simply moving the "is used" check after the loop
which erases the uses at the current index.
Differential Revision: https://reviews.llvm.org/D26554
llvm-svn: 286969
Add explicit v16i16/v32i8 ADD/SUB costs, matching the costs of v4i64/v8i32 - they were missing for some reason.
This has side effects on the LV max bandwidth tests (AVX1 now prefers 128-bit vectors vs AVX2 which still prefers 256-bit)
llvm-svn: 286832
When calculating the cost of a call instruction we were applying a heuristic penalty as well as the cost of the instruction itself.
However, when calculating the benefit from inlining we weren't discounting the equivalent penalty for the call instruction that would be removed! This caused skew in the calculation and meant we wouldn't inline in the following, trivial case:
int g() {
h();
}
int f() {
g();
}
llvm-svn: 286814
This is PR28376.
Unfortunately given the current structure of optimization diagnostics we
lack the capability to tell whether the user has
passed -Rpass-analysis=loop-vectorize since this is local to the
front-end (BackendConsumer::OptimizationRemarkHandler).
So rather than printing this even if the user has already
passed -Rpass-analysis, this patch just punts and stops recommending
this option. I don't think that getting this right is worth the
complexity.
Differential Revision: https://reviews.llvm.org/D26563
llvm-svn: 286662
When a function pointer is replaced with a jumptable pointer, special
case is needed to preserve the semantics of extern_weak functions.
Since a jumptable entry can not be extern_weak, we emulate that
behaviour by replacing all references to F (the extern_weak function)
with the following expression: F != nullptr ? JumpTablePtr : nullptr.
Extra special care is needed for global initializers, since most (or
probably all) backends can not lower an initializer that includes
this kind of constant expression. Initializers like that are replaced
with a global constructor (i.e. a runtime initializer).
llvm-svn: 286636
The current implementation is emitting a global constant that happens
to evaluate to the same bytes + relocation as a jump instruction on
X86. This does not work for PIE executables and shared libraries
though, because we end up with a wrong relocation type. And it has no
chance of working on ARM/AArch64 which use different relocation types
for jump instructions (R_ARM_JUMP24) that is never generated for
data.
This change replaces the constant with module-level inline assembly
followed by a hidden declaration of the jump table. Works fine for
ARM/AArch64, but has some drawbacks.
* Extra symbols are added to the static symbol table, which inflate
the size of the unstripped binary a little. Stripped binaries are not
affected. This happens because jump table declarations must be
external (because their body is in the inline asm).
* Original functions that were anonymous are now named
<original name>.cfi, and it affects symbolization sometimes. This is
necessary because the only user of these functions is the (inline
asm) jump table, so they had to be added to @llvm.used, which does
not allow unnamed functions.
llvm-svn: 286611
The r283656 did this in the remark arguments. We also need to do this
in the main function attribute as that is written to YAML as well.
llvm-svn: 286482
Note that the existing metadata checking was re-added by hand because the
script doesn't currently know how to generate checks for lines outside of
functions.
llvm-svn: 286460
Removing the limitation in visitInsertElementInst() causes several regressions
because we're not prepared to fold sequences of shuffles or inserts and extracts
separated by shuffles. Fixing that appears to be a difficult mission because we
are purposely trying to avoid creating shuffles with arbitrary shuffle masks
because some targets may choke on those.
https://llvm.org/bugs/show_bug.cgi?id=30923
llvm-svn: 286423
Summary:
The change will test the change in r286159.
The idea behind the change: Make the dbg location different between loop header and preheader/exit. Originally, dbg location 21 exists in 3 BBs: preheader, header, critical edge (exit). Update the debug location of inside the loop header from !21 to !22 so that it will reflect the correct location.
Reviewers: probinson
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D26428
llvm-svn: 286403
Scalar Evolution asserts when not all the operands of an Add Recurrence
Expression are loop invariants. Loop Strength Reduction should only
create affine Add Recurrences, so that both the start and the step of
the expression are loop invariants.
Differential Revision: https://reviews.llvm.org/D26185
llvm-svn: 286347
Summary: For functions with profile data, we are confident that loop sink will be optimal in sinking code.
Reviewers: davidxl, hfinkel
Subscribers: mehdi_amini, mzolotukhin, llvm-commits
Differential Revision: https://reviews.llvm.org/D26155
llvm-svn: 286325
As the test change shows, we can increase the critical path by adding
a 'not' instruction, so make sure that we're actually removing an
instruction if we do this transform.
This transform could also cause us to miss folds of min/max pairs.
llvm-svn: 286315
For example, it invalidates the domtree, causing assertions
in later passes which need dominator infos. Make it preserve
GlobalsAA, as suggested by Eli.
Differential Revision: https://reviews.llvm.org/D26381
llvm-svn: 286271
Summary:
These are good candidates for jump threading. This enables later opts
(such as InstCombine) to combine instructions from the selects with
instructions out of the selects. SimplifyCFG will fold the select
again if unfolding wasn't worth it.
Patch by James Molloy and Pablo Barrio.
Reviewers: rengolin, haicheng, sebpop
Subscribers: jojo, jmolloy, llvm-commits
Differential Revision: https://reviews.llvm.org/D26391
llvm-svn: 286236
This patch avoids scalarization of CTLZ by instead expanding to use CTPOP (ref: "Hacker's Delight") when the necessary operations are available.
This also adds the necessary cost models for X86 SSE2 targets (the main beneficiary) to ensure vectorization only happens when its useful.
Differential Revision: https://reviews.llvm.org/D25910
llvm-svn: 286233
With this we get a new field in the YAML record if the value being
streamed out has a debug location. For examples, please see the changes
to the tests.
This is then used in opt-viewer to display a link for the callee
function in the inlining remarks.
Differential Revision: https://reviews.llvm.org/D26366
llvm-svn: 286169
Summary:
In some specific scenarios with well understood operand bundle types
(like `"deopt"`) it may be possible to go ahead and convert recursion to
iteration, but TailRecursionElimination does not have that logic today
so avoid doing the right thing for now.
I need some input on whether `"funclet"` operand bundles should also
block tail recursion elimination. If not, I'll allow TRE across calls
with `"funclet"` operand bundles and add a test case.
Reviewers: rnk, majnemer, nlewycky, ahatanak
Subscribers: mcrosier, llvm-commits
Differential Revision: https://reviews.llvm.org/D26270
llvm-svn: 286147
Argument evaluation order is one of the edge cases where Clang differs
from GCC, yielding different IR depending on which compiler LLVM was
built with. Make the order deterministic and tune the test to actually
verify the order instead of trying to hide it.
llvm-svn: 286126
This was reverted at r285866 because there was a crash handling a scalar
select of vectors. I added a check for that pattern and a test case based
on the example provided in the post-commit thread for r285732.
llvm-svn: 286113
This reverts commit r285732.
This change introduced a new assertion failure in the following
testcase at -O2:
typedef short __v8hi __attribute__((__vector_size__(16)));
__v8hi foo(__v8hi &V1, __v8hi &V2, unsigned mask) {
__v8hi Result = V1;
if (mask & 0x80)
Result[0] = V2[0];
return Result;
}
llvm-svn: 285866
Summary:
It was detected that the reassociate pass could enter an inifite
loop when analysing dead code. Simply skipping to analyse basic
blocks that are dead avoids such problems (and as a side effect
we avoid spending time on optimising dead code).
The solution is using the same Reverse Post Order ordering of the
basic blocks when doing the optimisations, as when building the
precalculated rank map. A nice side-effect of this solution is
that we now know that we only try to do optimisations for blocks
with ranked instructions.
Fixes https://llvm.org/bugs/show_bug.cgi?id=30818
Reviewers: llvm-commits, davide, eli.friedman, mehdi_amini
Subscribers: dberlin
Differential Revision: https://reviews.llvm.org/D26154
llvm-svn: 285793
I think the former 'test50' had a typo making it functionally equivalent
to the former 'test49'; changed the predicate to provide more coverage.
llvm-svn: 285706
This patch introduces the combine:
(C1 shift (A add C2)) -> ((C1 shift C2) shift A)
iff A and C2 are both positive
If both A and C2 are know to be positive then we can safely split into 2 shifts, permitting the folding of the Inner shift.
Fix for the spec benchmark case mentioned by @nadav on PR15141 (assuming we can prove that the inputs as positive).
Differential Revision: https://reviews.llvm.org/D26000
llvm-svn: 285696
possible pointer-wrap-around concerns, in some cases.
Before this patch, collectConstStridedAccesses (part of interleaved-accesses
analysis) called getPtrStride with [Assume=false, ShouldCheckWrap=true] when
examining all candidate pointers. This is too conservative. Instead, this
patch makes collectConstStridedAccesses use an optimistic approach, calling
getPtrStride with [Assume=true, ShouldCheckWrap=false], and then, once the
candidate interleave groups have been formed, revisits the pointer-wrapping
analysis but only where it matters: namely, in groups that have gaps, and where
the gaps are not at the very end of the group (in which case the loop is
peeled). This second time getPtrStride is called with [Assume=false,
ShouldCheckWrap=true], but this could further be improved to using Assume=true,
once we also add the logic to track that we are not going to meet the scev
runtime checks threshold.
Differential Revision: https://reviews.llvm.org/D25276
llvm-svn: 285517
Try harder to detect obfuscated min/max patterns: the initial pattern was added with D9352 / rL236202.
There was a bug fix for PR27137 at rL264996, but I think we can do better by folding the corresponding
smax pattern and commuted variants.
The codegen tests demonstrate the effect of ValueTracking on the backend via SelectionDAGBuilder. We
can't expose these differences minimally in IR because we don't have smin/smax intrinsics for IR.
Differential Revision: https://reviews.llvm.org/D26091
llvm-svn: 285499
Summary:
This "pass" eagerly creates div and rem instructions even when only one
is needed -- it relies on a later pass (machine DCE?) to clean them up.
This is problematic not just from a cleanliness perspective (this pass
is running during CodeGenPrepare, so should leave the IR in a better
state), but it also creates a problem for instruction selection. If we
always have a div+rem, isel will always select a divrem instruction (if
possible), even when a single div or rem would do.
Specifically, in NVPTX, we want to compute rem from the output of div,
if available. But if a div is not available, we want to leave the rem
alone. This transformation is overeager if div is always available.
Because this code runs as part of CodeGenPrepare, it's nontrivial to
write a test for this change. But this will effectively be tested by
a later patch which adds the aforementioned change to NVPTX isel.
Reviewers: tra
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D26088
llvm-svn: 285460
Summary:
In BypassSlowDivision's short-dividend path, we would create e.g.
udiv exact i32 %a, %b
"exact" here means that we are asserting that %a is a multiple of %b.
But we have no reason to believe this must be true -- this is just a
bug, as far as I can tell.
Reviewers: tra
Subscribers: jholewinski, llvm-commits
Differential Revision: https://reviews.llvm.org/D26097
llvm-svn: 285459
Fixes PR 30784. Discussed with Justin, who pointed out that
in the new PassManager infrastructure we can have more fine-grained
control on which analyses we want to preserve, but this is the
best we can do with the current infrastructure.
llvm-svn: 285380
Summary: LICM may hoist instructions to preheader speculatively. Before code generation, we need to sink down the hoisted instructions inside to loop if it's beneficial. This pass is a reverse of LICM: looking at instructions in preheader and sinks the instruction to basic blocks inside the loop body if basic block frequency is smaller than the preheader frequency.
Reviewers: hfinkel, davidxl, chandlerc
Subscribers: anna, modocache, mgorny, beanz, reames, dberlin, chandlerc, mcrosier, junbuml, sanjoy, mzolotukhin, llvm-commits
Differential Revision: https://reviews.llvm.org/D22778
llvm-svn: 285308
After successfull horizontal reduction vectorization attempt for PHI node
vectorizer tries to update root binary op by combining vectorized tree
and the ReductionPHI node. But during vectorization this ReductionPHI
can be vectorized itself and replaced by the `undef` value, while the
instruction itself is marked for deletion. This 'marked for deletion'
PHI node then can be used in new binary operation, causing "Use still
stuck around after Def is destroyed" crash upon PHI node deletion.
Also the test is fixed to make it perform actual testing.
Differential Revision: https://reviews.llvm.org/D25671
llvm-svn: 285286
Summary:
Extends InstSimplify to handle both `x >=u x >> y` and `x >=u x udiv y`.
This is a folloup of rL258422 and
https://github.com/rust-lang/rust/pull/30917 where llvm failed to
optimize away the bounds checking in a binary search.
Patch by Arthur Silva!
Reviewers: sanjoy
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D25941
llvm-svn: 285228
Summary: This patch introduces updateDiscriminator to DILocation so that it can be directly called by AddDiscriminator. It also makes it easier to update the discriminator later.
Reviewers: dnovillo, dblaikie, aprantl, echristo
Subscribers: mehdi_amini, llvm-commits
Differential Revision: https://reviews.llvm.org/D25959
llvm-svn: 285207
Summary:
Select instruction annotation in IR PGO uses the edge count to infer the
branch count. It's currently placed in setInstrumentedCounts() where
no all the BB counts have been computed. This leads to wrong branch weights.
Move the annotation after all BB counts are populated.
Reviewers: davidxl
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D25961
llvm-svn: 285128
The original patch of the A->B->A BitCast optimization was reverted by r274094 because it may cause infinite loop inside compiler https://llvm.org/bugs/show_bug.cgi?id=27996.
The problem is with following code
xB = load (type B);
xA = load (type A);
+yA = (A)xB; B -> A
+zAn = PHI[yA, xA]; PHI
+zBn = (B)zAn; // A -> B
store zAn;
store zBn;
optimizeBitCastFromPhi generates
+zBn = (B)zAn; // A -> B
and expects it will be combined with the following store instruction to another
store zAn
Unfortunately before combineStoreToValueType is called on the store instruction, optimizeBitCastFromPhi is called on the new BitCast again, and this pattern repeats indefinitely.
optimizeBitCastFromPhi only generates BitCast for load/store instructions, only the BitCast before store can cause the reexecution of optimizeBitCastFromPhi, and BitCast before store can easily be handled by InstCombineLoadStoreAlloca.cpp. So the solution to the problem is if all users of a CI are store instructions, we should not do optimizeBitCastFromPhi on it. Then optimizeBitCastFromPhi will not be called on the new BitCast instructions.
Differential Revision: https://reviews.llvm.org/D23896
llvm-svn: 285116
When we predicate an instruction (div, rem, store) we place the instruction in
its own basic block within the vectorized loop. If a predicated instruction has
scalar operands, it's possible to recursively sink these scalar expressions
into the predicated block so that they might avoid execution. This patch sinks
as much scalar computation as possible into predicated blocks. We previously
were able to sink such operands only if they were extractelement instructions.
Differential Revision: https://reviews.llvm.org/D25632
llvm-svn: 285097
This adds a new function to DebugInfo.cpp that takes an llvm::Module
as input and removes all debug info metadata that is not directly
needed for line tables, thus effectively stripping all type and
variable information from the module.
The primary motivation for this feature was the bitcode work flow
(cf. http://lists.llvm.org/pipermail/llvm-dev/2016-June/100643.html
for more background). This is not wired up yet, but will be in
subsequent patches. For testing, the new functionality is exposed to
opt with a -strip-nonlinetable-debuginfo option.
The secondary use-case (and one that works right now!) is as a
reduction pass in bugpoint. I added two new bugpoint options
(-disable-strip-debuginfo and -disable-strip-debug-types) to control
the new features. By default it will first attempt to remove all debug
information, then only the type info, and then proceed to hack at any
remaining MDNodes.
Thanks to Adrian Prantl for stewarding this patch!
llvm-svn: 285094
Now that MemorySSA keeps track of whether MemoryUses are optimized, use
getClobberingMemoryAccess() to check MemoryUse memory dependencies since
it should no longer be so expensive.
This is a follow-up change to https://reviews.llvm.org/D25881
llvm-svn: 285080
This fixes a bug in the handling of lexical scopes, when more than one
scope is defined on the same line or functions are inlined into call
sites that are on the same line as the function definition. This
situation can easily happen in macro expansions.
The problem is solved by introducing a SmallDenseMap<DIScope *,
DILexicalBlockFile *, 1> that keeps track of all the different lexical
scopes that share a line/file location.
Fixes PR30681.
llvm-svn: 284998
Summary:
When using MemorySSA, re-optimize MemoryPhis when removing a store since
this may create MemoryPhis with all identical arguments.
Also, when using MemorySSA to check if two MemoryUses are reading from
the same version of the heap, use the defining access instead of calling
getClobberingAccess, since the latter can currently result in many more
AA calls. Once the MemorySSA use optimization tracking changes are
done, we can remove this limitation, which should result in more loads
being CSE'd.
Reviewers: dberlin
Subscribers: mcrosier, llvm-commits
Differential Revision: https://reviews.llvm.org/D25881
llvm-svn: 284984
Summary:
These are good candidates for jump threading. This enables later opts
(such as InstCombine) to combine instructions from the selects with
instructions out of the selects. SimplifyCFG will fold the select
again if unfolding wasn't worth it.
Patch by James Molloy and Pablo Barrio.
Reviewers: reames, bkramer, mcrosier, gberry, haicheng, jmolloy, sebpop
Subscribers: jojo, rengolin, llvm-commits
Differential Revision: https://reviews.llvm.org/D25477
llvm-svn: 284971
Summary:
Utility pass to remove gc.relocates created by rewrite statepoints for GC.
With respect to safepoint verification, the IR generated would be incorrect, and cannot run
as such.
This would be a single transformation on the final optimized IR.
The benefit of the pass is for easy analysis when the IRs are 'polluted' by too
many gc.relocates.
Added tests.
test run: All RS4GC tests with -verify option. Local downstream tests on large
IR files. This also works when the pointer being gc.relocated is another
gc.relocate.
Reviewers: sanjoy, reames
Subscribers: beanz, mgorny, llvm-commits
Differential Revision: https://reviews.llvm.org/D25096
llvm-svn: 284855
When we have a loop with a known upper bound on the number of iterations, and
furthermore know that either the number of iterations will be either exactly
that upper bound or zero, then we can fully unroll up to that upper bound
keeping only the first loop test to check for the zero iteration case.
Most of the work here is in plumbing this 'max-or-zero' information from the
part of scalar evolution where it's detected through to loop unrolling. I've
also gone for the safe default of 'false' everywhere but howManyLessThans which
could probably be improved.
Differential Revision: https://reviews.llvm.org/D25682
llvm-svn: 284818
There's no agreement about this patch. I personally find the
PRE machinery of the current GVN hard enough to reason about
that I'm not sure I'll try to land this again, instead of working
on the rewrite).
llvm-svn: 284796
0 - X --> X, if X is 0 or the minimum signed value
0 - X --> 0, if X is 0 or the minimum signed value and the sub is NSW
I noticed this pattern might be created in the backend after the change from D25485,
so we'll want to add a similar fold for the DAG.
The use of computeKnownBits in InstSimplify may be something to investigate if the
compile time of InstSimplify is noticeable. We could replace computeKnownBits with
specific pattern matchers or limit the recursion.
Differential Revision: https://reviews.llvm.org/D25785
llvm-svn: 284649
Some instructions from the original loop, when vectorized, can become trivially
dead. This happens because of the way we structure the new loop. For example,
we create new induction variables and induction variable "steps" in the new
loop. Thus, when we go to vectorize the original induction variable update, it
may no longer be needed due to the instructions we've already created. This
patch prevents us from creating these redundant instructions. This reduces code
size before simplification and allows greater flexibility in code generation
since we have fewer unnecessary instruction uses.
Differential Revision: https://reviews.llvm.org/D25631
llvm-svn: 284631
This change is motivated by the case when IndVarSimplify doesn't widen a comparison of IV increment because it can't prove IV increment being non-negative. We end up with a redundant trunc of the widened increment on this example.
for.body:
%i = phi i32 [ %start, %for.body.lr.ph ], [ %i.inc, %for.inc ]
%within_limits = icmp ult i32 %i, 64
br i1 %within_limits, label %continue, label %for.end
continue:
%i.i64 = zext i32 %i to i64
%arrayidx = getelementptr inbounds i32, i32* %base, i64 %i.i64
%val = load i32, i32* %arrayidx, align 4
br label %for.inc
for.inc:
%i.inc = add nsw nuw i32 %i, 1
%cmp = icmp slt i32 %i.inc, %limit
br i1 %cmp, label %for.body, label %for.end
There is a range check inside of the loop which guarantees the IV to be non-negative. NSW on the increment guarantees that the increment is also non-negative. Teach IndVarSimplify to use the range check to prove non-negativity of loop increments.
Reviewed By: sanjoy
Differential Revision: https://reviews.llvm.org/D25738
llvm-svn: 284629
Summary:
This pass shrink-wraps a condition to some library calls where the call
result is not used. For example:
sqrt(val);
is transformed to
if (val < 0)
sqrt(val);
Even if the result of library call is not being used, the compiler cannot
safely delete the call because the function can set errno on error
conditions.
Note in many functions, the error condition solely depends on the incoming
parameter. In this optimization, we can generate the condition can lead to
the errno to shrink-wrap the call. Since the chances of hitting the error
condition is low, the runtime call is effectively eliminated.
These partially dead calls are usually results of C++ abstraction penalty
exposed by inlining. This optimization hits 108 times in 19 C/C++ programs
in SPEC2006.
Reviewers: hfinkel, mehdi_amini, davidxl
Subscribers: modocache, mgorny, mehdi_amini, xur, llvm-commits, beanz
Differential Revision: https://reviews.llvm.org/D24414
llvm-svn: 284542
Summary:
The original implementation is in r261607, which was reverted in r269726 to accomendate the ProfileSummaryInfo analysis pass. The new implementation:
1. add a new metadata for function section prefix
2. query against ProfileSummaryInfo in CGP to set the correct section prefix for each function
3. output the section prefix set by CGP
Reviewers: davidxl, eraman
Subscribers: vsk, llvm-commits
Differential Revision: https://reviews.llvm.org/D24989
llvm-svn: 284533
As discussed on PR28461 we currently miss the chance to lower "fptosi <2 x double> %arg to <2 x i32>" to cvttpd2dq due to its use of illegal types.
This patch adds support for fptosi to 2i32 from both 2f64 and 2f32.
It also recognises that cvttpd2dq zeroes the upper 64-bits of the xmm result (similar to D23797) - we still don't do this for the cvttpd2dq/cvttps2dq intrinsics - this can be done in a future patch.
Differential Revision: https://reviews.llvm.org/D23808
llvm-svn: 284459
If -coverage is passed, but -g is not, clang populates the PassManager
pipeline with StripSymbols(debugOnly = true).
The stripSymbol pass therefore scans the list of named metadata,
drops !llvm.dbg.cu, but leaves !llvm.gcov and !0 (the compileUnit MD)
around. The verifier runs, and finds out that there's a CU not listed
in !llvm.dbg.cu (as it was previously dropped) -> crash.
When we strip debug info, so, check if there's coverage data,
and strip it as well, in order to avoid pending metadata left around.
Differential Revision: https://reviews.llvm.org/D25689
llvm-svn: 284418
Summary: Debug info should *not* affect code generation. This patch properly handles debug info to make sure the generated code are the same with or without debug info.
Reviewers: davidxl, mzolotukhin, jmolloy
Subscribers: aprantl, llvm-commits
Differential Revision: https://reviews.llvm.org/D25286
llvm-svn: 284415
Not all ConstantExprs can be represented by a global variable, for example most
pointer arithmetic other than addition of a constant, so we can't convert these
values from switch statements to lookup tables.
Differential Revision: https://reviews.llvm.org/D25550
llvm-svn: 284379
In theory this could be generalized to move anything where
we prove the operands are available, but that would require
rewriting PRE. As NewGVN will hopefully come soon, and we're
trying to rewrite PRE in terms of NewGVN+MemorySSA, it's probably
not worth spending too much time on it. Fix provided by
Daniel Berlin!
llvm-svn: 284311
X86. The pass optimizes as a unit the entire wide load + shuffles pattern
produced by interleaved vectorization. This initial patch optimizes one pattern
(64-bit elements interleaved by a factor of 4). Future patches will generalize
to additional patterns.
Patch by Farhana Aleen
Differential revision: http://reviews.llvm.org/D24681
llvm-svn: 284260
This test was apparently checking for 2 independent folds, but we have
plenty of tests for those individual folds already. We are lacking
vector tests, however, because we don't have the shift folds for vectors.
llvm-svn: 284243
Prefer add/zext because they are better supported in terms of value-tracking.
Note that the backend should be prepared for this IR canonicalization
(including vector types) after:
https://reviews.llvm.org/rL284015
Differential Revision: https://reviews.llvm.org/D25135
llvm-svn: 284241
This patch modifies the cost calculation of predicated instructions (div and
rem) to avoid the accumulation of rounding errors due to multiple truncating
integer divisions. The calculation for predicated stores will be addressed in a
follow-on patch since we currently don't scale the cost of predicated stores by
block probability.
Differential Revision: https://reviews.llvm.org/D25333
llvm-svn: 284123
This is with an extra change to avoid calling MemoryLocation::get() on a call instruction.
Differential Revision: https://reviews.llvm.org/D25542
llvm-svn: 284098
This CL didn't actually address the test case in PR30499, and clang
still crashes.
Also revert dependent change "Memory-SSA cleanup of clobbers interface, NFC"
Reverts r283965 and r283967.
llvm-svn: 284093
Reappy r284044 after revert in r284051. Krzysztof fixed the error in r284049.
The original summary:
This patch tries to fully unroll loops having break statement like this
for (int i = 0; i < 8; i++) {
if (a[i] == value) {
found = true;
break;
}
}
GCC can fully unroll such loops, but currently LLVM cannot because LLVM only
supports loops having exact constant trip counts.
The upper bound of the trip count can be obtained from calling
ScalarEvolution::getMaxBackedgeTakenCount(). Part of the patch is the
refactoring work in SCEV to prevent duplicating code.
The feature of using the upper bound is enabled under the same circumstance
when runtime unrolling is enabled since both are used to unroll loops without
knowing the exact constant trip count.
llvm-svn: 284053
This patch tries to fully unroll loops having break statement like this
for (int i = 0; i < 8; i++) {
if (a[i] == value) {
found = true;
break;
}
}
GCC can fully unroll such loops, but currently LLVM cannot because LLVM only
supports loops having exact constant trip counts.
The upper bound of the trip count can be obtained from calling
ScalarEvolution::getMaxBackedgeTakenCount(). Part of the patch is the
refactoring work in SCEV to prevent duplicating code.
The feature of using the upper bound is enabled under the same circumstance
when runtime unrolling is enabled since both are used to unroll loops without
knowing the exact constant trip count.
Differential Revision: https://reviews.llvm.org/D24790
llvm-svn: 284044
Branch folder removes implicit defs if they are the only non-branching
instructions in a block, and the branches do not use the defined registers.
The problem is that in some cases these implicit defs are required for
the liveness information to be correct.
Differential Revision: https://reviews.llvm.org/D25478
llvm-svn: 284036
Summary:
Constant bundle operands may need to retain their constant-ness for
correctness. I'll admit that this is slightly odd, but it looks like
SimplifyCFG already does this for things like @llvm.frameaddress and
@llvm.stackmap, so I suppose adding one more case is not a big deal.
It is possible to add a mechanism to denote bundle operands that need to
remain constants, but that's probably too complicated for the time
being.
Reviewers: jmolloy
Subscribers: mcrosier, llvm-commits
Differential Revision: https://reviews.llvm.org/D25502
llvm-svn: 284028
Since this change is known to cause performance degradations in some cases it's commited under a temporary flag which is turned off by default.
Patch by Li Huang
Differential Revision: https://reviews.llvm.org/D18777
llvm-svn: 284022
An arithmetic shift can be safely changed to a logical shift if the first
operand is known positive. This allows ComputeKnownBits (and similar analysis)
to determine the sign bit of the shifted value in some cases. In turn, this
allows InstCombine to canonicalize a signed comparison (a > 0) into an equality
check (a != 0).
PR30577
Differential Revision: https://reviews.llvm.org/D25119
llvm-svn: 284013
As discussed by Andrea on PR30486, we have an unsafe cast to an Instruction type in the select combine which doesn't take into account that it could be a ConstantExpr instead.
Differential Revision: https://reviews.llvm.org/D25466
llvm-svn: 284000
This is a refreshed version of a patch that was reverted: it fixes
the problems reported in both PR30216 and PR30499, and
contains all the test-cases from both bugs.
To hoist stores past loads, we used to search for potential
conflicting loads on the hoisting path by following a MemorySSA
def-def link from the store to be hoisted to the previous
defining memory access, and from there we followed the def-use
chains to all the uses that occur on the hoisting path. The
problem is that the def-def link may point to a store that does
not alias with the store to be hoisted, and so the loads that are
walked may not alias with the store to be hoisted, and even as in
the testcase of PR30216, the loads that may alias with the store
to be hoisted are not visited.
The current patch visits all loads on the path from the store to
be hoisted to the hoisting position and uses the alias analysis
to ask whether the store may alias the load. I was not able to
use the MemorySSA functionality to ask for whether load and
store are clobbered: I'm not sure which function to call, so I
used a call to AA->isNoAlias().
Store past store is still working as before using a MemorySSA
query: I added an extra test to pr30216.ll to make sure store
past store does not regress.
Tested on x86_64-linux with check and a test-suite run.
Differential Revision: https://reviews.llvm.org/D25476
llvm-svn: 283965
When combining an integer load with !range metadata that does not include 0 to a pointer load, make sure emit !nonnull metadata on the newly-created pointer load. This prevents the !nonnull metadata from being dropped during a ptrtoint/inttoptr pair.
This fixes PR30597.
Patch by Ariel Ben-Yehuda!
Differential Revision: https://reviews.llvm.org/D25215
llvm-svn: 283836
Fixed copy+paste vector alignment to correct for per-element scalar loads
Increased to 512-bit data sizes in preparation of avx512 tests
llvm-svn: 283748
Value names may be prefixed with a binary '1' to indicate that the
backend should not modify the symbols due to any platform naming
convention.
This should not show up in the YAML opt record file because it breaks
the YAML parser.
llvm-svn: 283656
Summary:
If heap allocation of a coroutine is elided, we need to make sure that we will update an address stored in the coroutine frame from f.destroy to f.cleanup.
Before this change, CoroSplit synthesized these stores after coro.begin:
```
store void (%f.Frame*)* @f.resume, void (%f.Frame*)** %resume.addr
store void (%f.Frame*)* @f.destroy, void (%f.Frame*)** %destroy.addr
```
In those cases where we did heap elision, but were not able to devirtualize all indirect calls, destroy call will attempt to "free" the coroutine frame stored on the stack. Oops.
Now we use select to put an appropriate coroutine subfunction in the destroy slot. As bellow:
```
store void (%f.Frame*)* @f.resume, void (%f.Frame*)** %resume.addr
%0 = select i1 %need.alloc, void (%f.Frame*)* @f.destroy, void (%f.Frame*)* @f.cleanup
store void (%f.Frame*)* %0, void (%f.Frame*)** %destroy.addr
```
Reviewers: majnemer
Subscribers: mehdi_amini, llvm-commits
Differential Revision: https://reviews.llvm.org/D25377
llvm-svn: 283625
Summary: Add tests for cases where we have zero coverage in RS4GC.
Reviewers: sanjoy, reames
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D25341
llvm-svn: 283591
If we're going to canonicalize IR towards select of constants, try harder to create those.
Also, don't lose the metadata.
This is actually 4 related transforms in one patch:
// select X, (sext X), C --> select X, -1, C
// select X, (zext X), C --> select X, 1, C
// select X, C, (sext X) --> select X, C, 0
// select X, C, (zext X) --> select X, C, 0
Differential Revision: https://reviews.llvm.org/D25126
llvm-svn: 283575
Previously, we marked the branch conditions of latch blocks uniform after
vectorization if they were instructions contained in the loop. However, if a
condition instruction has users other than the branch, it may not remain
uniform. This patch ensures the conditions we mark uniform are only used by the
branch. This should fix PR30627.
Reference: https://llvm.org/bugs/show_bug.cgi?id=30627
llvm-svn: 283563
Summary:
While walking defs of pointer operands we were assuming that the pointer
size would remain constant. This is not true, because addresspacecast
instructions may cast the pointer to an address space with a different
pointer width.
This partial reverts r282612, which was a more conservative solution
to this problem.
Reviewers: reames, sanjoy, apilipenko
Subscribers: wdng, llvm-commits
Differential Revision: https://reviews.llvm.org/D24772
llvm-svn: 283557
unrolling.
The next code is not vectorized by the SLPVectorizer:
```
int test(unsigned int *p) {
int sum = 0;
for (int i = 0; i < 8; i++)
sum += p[i];
return sum;
}
```
During optimization this loop is fully unrolled and SLPVectorizer is
unable to vectorize it. Patch tries to fix this problem.
Differential Revision: https://reviews.llvm.org/D24796
llvm-svn: 283535
With the ROPI and RWPI relocation models we can't always have pointers
to global data or functions in constant data, so don't try to convert switches
into lookup tables if any value in the lookup table would require a relocation.
We can still safely emit lookup tables of other values, such as simple
constants.
Differential Revision: https://reviews.llvm.org/D24462
llvm-svn: 283530
GetCaseResults assumed that a terminator with one successor was an
unconditional branch. This is not necessarily the case, it could be a
cleanupret.
Strengthen the check by querying whether or not the terminator is
exceptional.
llvm-svn: 283517
Vectorizer tests in the target-independent directory should not have a target
triple. If a test really needs to query a specific backend, it belongs in the
right target subdirectory (which "REQUIRES" the right backend). Otherwise, it
should not specify a triple.
llvm-svn: 283512
Add a weak alias to the renamed Comdat function in IR level instrumentation,
using it's original name. This ensures the same behavior w/ and w/o IR
instrumentation, even for non standard conforming code.
Differential Revision: http://reviews.llvm.org/D25339
llvm-svn: 283490
This adds a new function to DebugInfo.cpp that takes an llvm::Module
as input and removes all debug info metadata that is not directly
needed for line tables, thus effectively stripping all type and
variable information from the module.
The primary motivation for this feature was the bitcode work flow
(cf. http://lists.llvm.org/pipermail/llvm-dev/2016-June/100643.html
for more background). This is not wired up yet, but will be in
subsequent patches. For testing, the new functionality is exposed to
opt with a -strip-nonlinetable-debuginfo option.
The secondary use-case (and one that works right now!) is as a
reduction pass in bugpoint. I added two new bugpoint options
(-disable-strip-debuginfo and -disable-strip-debug-types) to control
the new features. By default it will first attempt to remove all debug
information, then only the type info, and then proceed to hack at any
remaining MDNodes.
llvm-svn: 283473
The purpose of the YAML diagnostic output file is to collect information on
optimizations performed, or not performed, for later processing by tools that
help users (and compiler developers) understand how code was optimized. As
such, the diagnostics that appear in the file should not be coupled to what a
user might want to see summarized for them as the compiler runs, and in fact,
because the user likely does not know what optimization diagnostics their tools
might want to use, the user cannot provide a useful filter regardless. As such,
we shouldn't filter the diagnostics going to the output file.
Differential Revision: https://reviews.llvm.org/D25224
llvm-svn: 283236
Splitting the edge is nontrivial because of the landing pad, and we would
currently assert trying to do it.
Differential Revision: https://reviews.llvm.org/D24680
llvm-svn: 283129
This should fix:
https://llvm.org/bugs/show_bug.cgi?id=30433
There are a couple of open questions about the codegen:
1. Should we let scalar ops be scalars and avoid vector constant loads/splats?
2. Should we have a pass to combine constants such as the inverted pair that we have here?
Differential Revision: https://reviews.llvm.org/D25165
llvm-svn: 283119
Summary:
In the case below, %Result.i19 is defined between coro.save and coro.suspend and used after coro.suspend. We need to correctly place such a value into the coroutine frame.
```
%save = call token @llvm.coro.save(i8* null)
%Result.i19 = getelementptr inbounds %"struct.lean_future<int>::Awaiter", %"struct.lean_future<int>::Awaiter"* %ref.tmp7, i64 0, i32 0
%suspend = call i8 @llvm.coro.suspend(token %save, i1 false)
switch i8 %suspend, label %exit [
i8 0, label %await.ready
i8 1, label %exit
]
await.ready:
%val = load i32, i32* %Result.i19
```
Reviewers: majnemer
Subscribers: llvm-commits, mehdi_amini
Differential Revision: https://reviews.llvm.org/D24418
llvm-svn: 282902
Summary:
Without the fix, if there was a function inlined into the coroutine with debug information, CloneFunctionInto(NewF, &F, VMap, /*ModuleLevelChanges=*/true, Returns); would duplicate all of the debug information including the DICompileUnit.
We know use VMap to indicate that debug metadata for a File, Unit and FunctionType should not be duplicated when we creating clones that will become f.resume, f.destroy and f.cleanup.
Reviewers: majnemer
Subscribers: mehdi_amini, llvm-commits
Differential Revision: https://reviews.llvm.org/D24417
llvm-svn: 282899
Summary: Not all coro.subfn.addr intrinsics can be eliminated in CoroElide through devirtualization. Those that remain need to be lowered in CoroCleanup.
Reviewers: majnemer
Subscribers: llvm-commits, mehdi_amini
Differential Revision: https://reviews.llvm.org/D24412
llvm-svn: 282897
Summary: Debug info should *not* affect optimization decisions. This patch updates loop unroller cost model to make it not affected by debug info.
Reviewers: davidxl, mzolotukhin
Subscribers: haicheng, llvm-commits, mzolotukhin
Differential Revision: https://reviews.llvm.org/D25098
llvm-svn: 282894
When building the steps for scalar induction variables, we previously attempted
to determine if all the scalar users of the induction variable were uniform. If
they were, we would only emit the step corresponding to vector lane zero. This
optimization was too aggressive. We generally don't know the entire set of
induction variable users that will be scalar. We have
isScalarAfterVectorization, but this is only a conservative estimate of the
instructions that will be scalarized. Thus, an induction variable may have
scalar users that aren't already known to be scalar. To avoid emitting unused
steps, we can only check that the induction variable is uniform. This should
fix PR30542.
Reference: https://llvm.org/bugs/show_bug.cgi?id=30542
llvm-svn: 282863
Summary:
We don't want to decay hot callsites to import chains of hot
callsites. The same mechanism is used in LIPO.
Reviewers: tejohnson, eraman, mehdi_amini
Subscribers: llvm-commits, mehdi_amini
Differential Revision: https://reviews.llvm.org/D24976
llvm-svn: 282833
Summary:
Not tunned up heuristic, but with this small heuristic there is about
+0.10% improvement on SPEC 2006
Reviewers: tejohnson, mehdi_amini, eraman
Subscribers: mehdi_amini, llvm-commits
Differential Revision: https://reviews.llvm.org/D24940
llvm-svn: 282733
Summary:
The patch fixes regression caused by two earlier patches D18777 and D18867.
Reviewers: reames, sanjoy
Differential Revision: http://reviews.llvm.org/D24280
From: Li Huang
llvm-svn: 282650
Also, remove unnecessary function attributes, parameters, and comments.
It looks like at least some of these tests are not minimal though...
llvm-svn: 282620
Pointers in different addrspaces can have different sizes, so it's not valid to look through addrspace cast calculating base and offset for a value.
This is similar to D13008.
Reviewed By: reames
Differential Revision: https://reviews.llvm.org/D24729
llvm-svn: 282612
There is really no reason for these to be separate.
The vectorizer started this pretty bad tradition that the text of the
missed remarks is pretty meaningless, i.e. vectorization failed. There,
you have to query analysis to get the full picture.
I think we should just explain the reason for missing the optimization
in the missed remark when possible. Analysis remarks should provide
information that the pass gathers regardless whether the optimization is
passing or not.
llvm-svn: 282542
(Re-committed after moving the template specialization under the yaml
namespace. GCC was complaining about this.)
This allows various presentation of this data using an external tool.
This was first recommended here[1].
As an example, consider this module:
1 int foo();
2 int bar();
3
4 int baz() {
5 return foo() + bar();
6 }
The inliner generates these missed-optimization remarks today (the
hotness information is pulled from PGO):
remark: /tmp/s.c:5:10: foo will not be inlined into baz (hotness: 30)
remark: /tmp/s.c:5:18: bar will not be inlined into baz (hotness: 30)
Now with -pass-remarks-output=<yaml-file>, we generate this YAML file:
--- !Missed
Pass: inline
Name: NotInlined
DebugLoc: { File: /tmp/s.c, Line: 5, Column: 10 }
Function: baz
Hotness: 30
Args:
- Callee: foo
- String: will not be inlined into
- Caller: baz
...
--- !Missed
Pass: inline
Name: NotInlined
DebugLoc: { File: /tmp/s.c, Line: 5, Column: 18 }
Function: baz
Hotness: 30
Args:
- Callee: bar
- String: will not be inlined into
- Caller: baz
...
This is a summary of the high-level decisions:
* There is a new streaming interface to emit optimization remarks.
E.g. for the inliner remark above:
ORE.emit(DiagnosticInfoOptimizationRemarkMissed(
DEBUG_TYPE, "NotInlined", &I)
<< NV("Callee", Callee) << " will not be inlined into "
<< NV("Caller", CS.getCaller()) << setIsVerbose());
NV stands for named value and allows the YAML client to process a remark
using its name (NotInlined) and the named arguments (Callee and Caller)
without parsing the text of the message.
Subsequent patches will update ORE users to use the new streaming API.
* I am using YAML I/O for writing the YAML file. YAML I/O requires you
to specify reading and writing at once but reading is highly non-trivial
for some of the more complex LLVM types. Since it's not clear that we
(ever) want to use LLVM to parse this YAML file, the code supports and
asserts that we're writing only.
On the other hand, I did experiment that the class hierarchy starting at
DiagnosticInfoOptimizationBase can be mapped back from YAML generated
here (see D24479).
* The YAML stream is stored in the LLVM context.
* In the example, we can probably further specify the IR value used,
i.e. print "Function" rather than "Value".
* As before hotness is computed in the analysis pass instead of
DiganosticInfo. This avoids the layering problem since BFI is in
Analysis while DiagnosticInfo is in IR.
[1] https://reviews.llvm.org/D19678#419445
Differential Revision: https://reviews.llvm.org/D24587
llvm-svn: 282539
This allows various presentation of this data using an external tool.
This was first recommended here[1].
As an example, consider this module:
1 int foo();
2 int bar();
3
4 int baz() {
5 return foo() + bar();
6 }
The inliner generates these missed-optimization remarks today (the
hotness information is pulled from PGO):
remark: /tmp/s.c:5:10: foo will not be inlined into baz (hotness: 30)
remark: /tmp/s.c:5:18: bar will not be inlined into baz (hotness: 30)
Now with -pass-remarks-output=<yaml-file>, we generate this YAML file:
--- !Missed
Pass: inline
Name: NotInlined
DebugLoc: { File: /tmp/s.c, Line: 5, Column: 10 }
Function: baz
Hotness: 30
Args:
- Callee: foo
- String: will not be inlined into
- Caller: baz
...
--- !Missed
Pass: inline
Name: NotInlined
DebugLoc: { File: /tmp/s.c, Line: 5, Column: 18 }
Function: baz
Hotness: 30
Args:
- Callee: bar
- String: will not be inlined into
- Caller: baz
...
This is a summary of the high-level decisions:
* There is a new streaming interface to emit optimization remarks.
E.g. for the inliner remark above:
ORE.emit(DiagnosticInfoOptimizationRemarkMissed(
DEBUG_TYPE, "NotInlined", &I)
<< NV("Callee", Callee) << " will not be inlined into "
<< NV("Caller", CS.getCaller()) << setIsVerbose());
NV stands for named value and allows the YAML client to process a remark
using its name (NotInlined) and the named arguments (Callee and Caller)
without parsing the text of the message.
Subsequent patches will update ORE users to use the new streaming API.
* I am using YAML I/O for writing the YAML file. YAML I/O requires you
to specify reading and writing at once but reading is highly non-trivial
for some of the more complex LLVM types. Since it's not clear that we
(ever) want to use LLVM to parse this YAML file, the code supports and
asserts that we're writing only.
On the other hand, I did experiment that the class hierarchy starting at
DiagnosticInfoOptimizationBase can be mapped back from YAML generated
here (see D24479).
* The YAML stream is stored in the LLVM context.
* In the example, we can probably further specify the IR value used,
i.e. print "Function" rather than "Value".
* As before hotness is computed in the analysis pass instead of
DiganosticInfo. This avoids the layering problem since BFI is in
Analysis while DiagnosticInfo is in IR.
[1] https://reviews.llvm.org/D19678#419445
Differential Revision: https://reviews.llvm.org/D24587
llvm-svn: 282499
Summary:
We don't currently need this facility for CFI. Disabling individual hot methods proved
to be a better strategy in Chrome.
Also, the design of the feature is suboptimal, as pointed out by Peter Collingbourne.
Reviewers: pcc
Subscribers: kcc
Differential Revision: https://reviews.llvm.org/D24948
llvm-svn: 282461
Summary:
This patch improves thinlto importer
by importing 3x larger functions that are called from hot block.
I compared performance with the trunk on spec, and there
were about 2% on povray and 3.33% on milc. These results seems
to be consistant and match the results Teresa got with her simple
heuristic. Some benchmarks got slower but I think they are just
noisy (mcf, xalancbmki, omnetpp)- running the benchmarks again with
more iterations to confirm. Geomean of all benchmarks including the noisy ones
were about +0.02%.
I see much better improvement on google branch with Easwaran patch
for pgo callsite inlining (the inliner actually inline those big functions)
Over all I see +0.5% improvement, and I get +8.65% on povray.
So I guess we will see much bigger change when Easwaran patch will land
(it depends on new pass manager), but it is still worth putting this to trunk
before it.
Implementation details changes:
- Removed CallsiteCount.
- ProfileCount got replaced by Hotness
- hot-import-multiplier is set to 3.0 for now,
didn't have time to tune it up, but I see that we get most of the interesting
functions with 3, so there is no much performance difference with higher, and
binary size doesn't grow as much as with 10.0.
Reviewers: eraman, mehdi_amini, tejohnson
Subscribers: mehdi_amini, llvm-commits
Differential Revision: https://reviews.llvm.org/D24638
llvm-svn: 282437
This patch ensures that we actually scalarize instructions marked scalar after
vectorization. Previously, such instructions may have been vectorized instead.
Differential Revision: https://reviews.llvm.org/D23889
llvm-svn: 282418
Summary:
If coroutine has no suspend points, remove heap allocation and turn a coroutine into a normal function.
Also, if a pattern is detected that coroutine resumes or destroys itself prior to coro.suspend call, turn the suspend point into a simple jump to resume or cleanup label. This pattern occurs when coroutines are used to propagate errors in functions that return expected<T>.
Reviewers: majnemer
Subscribers: mehdi_amini, llvm-commits
Differential Revision: https://reviews.llvm.org/D24408
llvm-svn: 282414
The index of the new insertelement instruction was evaluated in the
wrong way, it was considered as the index of the inserted value instead
of index of the position, where the value should be inserted.
llvm-svn: 282401
This patch fixes PR30366.
Function foldUDivShl() worked under the assumption that one of the values
in input to the function was always an instance of llvm::Instruction.
However, function visitUDivOperand() (the only user of foldUDivShl) was
clearly violating that precondition; internally, visitUDivOperand() uses pattern
matches to check the operands of a udiv. Pattern matchers for binary operators
know how to handle both Instruction and ConstantExpr values.
This patch fixes the problem in foldUDivShl(). Now we use pattern matchers
instead of explicit casts to Instruction. The reduced test case from PR30366
has been added to test file InstCombine/udiv-simplify.ll.
Differential Revision: https://reviews.llvm.org/D24565
llvm-svn: 282398
If inserting more than one constant into a vector:
define <4 x float> @foo(<4 x float> %x) {
%ins1 = insertelement <4 x float> %x, float 1.0, i32 1
%ins2 = insertelement <4 x float> %ins1, float 2.0, i32 2
ret <4 x float> %ins2
}
InstCombine could reduce that to a shufflevector:
define <4 x float> @goo(<4 x float> %x) {
%shuf = shufflevector <4 x float> %x, <4 x float> <float undef, float 1.0, float 2.0, float undef>, <4 x i32><i32 0, i32 5, i32 6, i32 3>
ret <4 x float> %shuf
}
Also, InstCombine tries to convert shuffle instruction to single insertelement, if one of the vectors is a constant vector and only a single element from this constant should be used in shuffle, i.e.
shufflevector <4 x float> %v, <4 x float> <float undef, float 1.0, float
undef, float undef>, <4 x i32> <i32 0, i32 5, i32 undef, i32 undef> ->
insertelement <4 x float> %v, float 1.0, 1
Differential Revision: https://reviews.llvm.org/D24182
llvm-svn: 282237
We already have the udiv variant of this transform, so I think this is ok for
InstCombine too even though there is an increase in IR instructions. As the
tests and TODO comments show, the transform can lead to follow-on combines.
This should fix: https://llvm.org/bugs/show_bug.cgi?id=28672
Differential Revision: https://reviews.llvm.org/D24527
llvm-svn: 282209
and also the dependent r282175 "GVN-hoist: do not dereference null pointers"
It's causing compiler crashes building Harfbuzz (PR30499).
llvm-svn: 282199
To hoist stores past loads, we used to search for potential
conflicting loads on the hoisting path by following a MemorySSA
def-def link from the store to be hoisted to the previous
defining memory access, and from there we followed the def-use
chains to all the uses that occur on the hoisting path. The
problem is that the def-def link may point to a store that does
not alias with the store to be hoisted, and so the loads that are
walked may not alias with the store to be hoisted, and even as in
the testcase of PR30216, the loads that may alias with the store
to be hoisted are not visited.
The current patch visits all loads on the path from the store to
be hoisted to the hoisting position and uses the alias analysis
to ask whether the store may alias the load. I was not able to
use the MemorySSA functionality to ask for whether load and
store are clobbered: I'm not sure which function to call, so I
used a call to AA->isNoAlias().
Store past store is still working as before using a MemorySSA
query: I added an extra test to pr30216.ll to make sure store
past store does not regress.
Differential Revision: https://reviews.llvm.org/D24517
llvm-svn: 282168
Without this patch, GVN-hoist would think that a branch instruction is a scalar instruction
and would try to value number it. The patch filters out all such kind of irrelevant instructions.
A bit frustrating is that there is no easy way to discard all those very infrequent instructions,
a bit like isa<TerminatorInst> that stands for a large family of instructions. I'm thinking that
checking for those very infrequent other instructions would cost us more in compilation time
than just letting those instructions getting numbered, so I'm still thinking that a simpler check:
if (isa<TerminatorInst>(I))
return false;
is better than listing all the other less frequent instructions.
Differential Revision: https://reviews.llvm.org/D23929
llvm-svn: 282160
Currently, we give up on loop interchange if we encounter a flow dependency
anywhere in the loop list. Worse yet, we don't even track output dependencies.
This patch updates the dependency matrix computation to track flow and output
dependencies in the same way we track anti dependencies.
This improves an internal workload by 2.2x.
Note the loop interchange pass is off by default and it can be enabled with
'-mllvm -enable-loopinterchange'
Differential Revision: https://reviews.llvm.org/D24564
llvm-svn: 282101
If we identify an instruction as uniform after vectorization, we know that we
should only use the value corresponding to the first vector lane of each unroll
iteration. However, when scalarizing such instructions, we still produce values
for the other vector lanes. This patch prevents us from generating the unused
scalars.
Differential Revision: https://reviews.llvm.org/D24275
llvm-svn: 282087
Summary: Now that we have more precise debug info, we should change back to use maximum to get basic block weight.
Reviewers: dnovillo
Subscribers: andreadb, llvm-commits
Differential Revision: https://reviews.llvm.org/D24788
llvm-svn: 282084
SROA doesn't preserve the llvm.mem.parallel_loop_access metadata when it
transforms loads/stores. This patch fixes a couple occurences of this
issue.
(Partially addresses PR28981).
Differential Revision: https://reviews.llvm.org/D23549
llvm-svn: 281960
Summary: Callsites in the same basic block should share the same hotness. This patch checks for the hottest callsite in the same basic block, and use the hotness for all callsites in that basic block for early inline decisions. It also fixes the test to add "-S" so theat the "CHECK-NOT" is actually checking the content.
Reviewers: dnovillo
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D24734
llvm-svn: 281927
Summary: It does not make sense to set equal weights for all unkown branches as we have static branch prediction available.
Reviewers: dnovillo
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D24732
llvm-svn: 281912
Summary: The call target count profile is directly derived from LBR branch->target data. This is more reliable than instruction frequency profiles that could be moved across basic block boundaries. This patches uses call target count profile to annotate call instructions.
Reviewers: davidxl, dnovillo
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D24410
llvm-svn: 281911
When phi nodes are created in the -mem2reg phase, the @llvm.dbg.declare
entries are converted to @llvm.dbg.value entries at the place where the
store instructions existed. However no entry is created to describe
the resulting value of the phi node.
The effect of this is especially noticeable in for loops which have a
constant for the intial value; the loop control variable's location
would be described as the intial constant value in the loop body once
the -mem2reg optimization phase was run.
This change adds the creation of the @llvm.dbg.value entries to describe
variables whose location is the result of a phi node created in -mem2reg.
Also when the phi node is finally lowered to a machine instruction it
is important that the lowered "load" instruction is placed before the
associated DEBUG_VALUE entry describing the value loaded.
Differential Revision: https://reviews.llvm.org/D23715
llvm-svn: 281895
We were updating metadata but not IR flags. Because we pick an arbitrary instruction to be the CSE candidate, it comes down to luck (50% or less chance) if this results in broken codegen or not, which is why PR30373 which is actually not the fault of the commit it was bisected down to.
Fixes PR30373.
llvm-svn: 281889
Summary: Previously we reline on inst-combine to remove inlinable invoke instructions. This causes trouble because a few extra optimizations are schedule early that could introduce too much CFG change (e.g. simplifycfg removes too much control flow). This patch handles invoke instruction in-place during sample profile annotation, so that we do not rely on instcombine to remove those invoke instructions.
Reviewers: davidxl, dnovillo
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D24409
llvm-svn: 281870
Summary:
This fixes an issue when files are compiled with -flto=thin
at default -O0. We need to rename anonymous globals before attempting
to write the module summary because all values need names for
the summary. This was happening at -O1 and above, but not before
the early exit when constructing the pipeline for -O0.
Also add an internal -prepare-for-thinlto option to enable this
to be tested via opt.
Fixes PR30419.
Reviewers: mehdi_amini
Subscribers: probinson, llvm-commits, mehdi_amini
Differential Revision: https://reviews.llvm.org/D24701
llvm-svn: 281840
This is a fix for PR30318.
Clang may generate IR where an alloca is already live when entering a
BB with lifetime.start. In this case, conservatively extend the
alloca lifetime all the way back to the block entry.
llvm-svn: 281784
computeKnownBits() already works for integer vectors, so allow vector types when calling that from InstCombine.
I don't think the change to use m_APInt in computeKnownBits is strictly necessary because we do check for
ConstantVector later, but it's more efficient to handle the splat case without needing to loop on vector elements.
This should work with InstSimplify, but doesn't yet, so I made that a FIXME comment on the test for PR24942:
https://llvm.org/bugs/show_bug.cgi?id=24942
Differential Revision: https://reviews.llvm.org/D24677
llvm-svn: 281777
A follow-up patch will rename this pass and the source file accordingly,
but I figured the non-NFC change will be easier to spot in isolation.
Differential Revision: https://reviews.llvm.org/D24641
llvm-svn: 281744
These 2 helper functions were already using APInt internally, so just
change the API and caller to allow folds for splats. The scalar
regression tests look quite thorough, so I just added a couple of
tests to prove that vectors are handled too.
These folds should be grouped with the other cmp+shift folds though.
That can be an NFC follow-up.
llvm-svn: 281663
GlobalOpt is already dead-code-eliminating global definitions. With
this change it also takes care of declarations.
Hopefully this should make it now a strict superset of GlobalDCE.
This is important for LTO/ThinLTO as we don't want the linker to see
"undefined reference" when it processes the input files: it could
prevent proper internalization (or even load an extra file from a
static archive, changing the behavior of the program!).
llvm-svn: 281653
The patch is to partially fix PR10584. Correlated Value Propagation queries LVI
to check non-null for pointer params of each callsite. If we know the def of
param is an alloca instruction, we know it is non-null and can return early from
LVI. Similarly, CVP queries LVI to check whether pointer for each mem access is
constant. If the def of the pointer is an alloca instruction, we know it is not
a constant pointer. These shortcuts can reduce the cost of CVP significantly.
Differential Revision: https://reviews.llvm.org/D18066
llvm-svn: 281586
This patch moves the processing of pointer induction variables in
collectLoopUniforms from the consecutive pointer phase of the analysis to the
phi node phase. Previously, if a pointer induction variable was used by both a
scalarized non-memory instruction as well as a vectorized memory instruction,
we would incorrectly identify the pointer as uniform. Pointer induction
variables should be treated the same as other phi nodes. That is, they are
uniform if all users of the induction variable and induction variable update
are uniform.
Differential Revision: https://reviews.llvm.org/D24511
llvm-svn: 281485
ObjC library call with call return.
ARC contraction tries to replace uses of an argument passed to an
objective-c library call with the call return value. For example, in the
following IR, it replaces uses of argument %9 and uses of the values
discovered traversing the chain upwards (%7 and %8) with the call return
%10, if they are dominated by the call to @objc_autoreleaseReturnValue.
This transformation enables code-gen to tail-call the call to
@objc_autoreleaseReturnValue, which is necessary to enable auto release
return value optimization.
%7 = tail call i8* @objc_loadWeakRetained(i8** %6)
%8 = bitcast i8* %7 to %0*
%9 = bitcast %0* %8 to i8*
%10 = tail call i8* @objc_autoreleaseReturnValue(i8* %9)
ret %0* %8
Since r276727, llvm started removing redundant bitcasts and as a result
started feeding the following IR to ARC contraction:
%7 = tail call i8* @objc_loadWeakRetained(i8** %6)
%8 = bitcast i8* %7 to %0*
%9 = tail call i8* @objc_autoreleaseReturnValue(i8* %7)
ret %0* %8
ARC contraction no longer does the optimization described above since it
only traverses the chain upwards and fails to recognize that the
function return can be replaced by the call return. This commit changes
ARC contraction to traverse the chain downwards too and replace uses of
bitcasts with the call return.
rdar://problem/28011339
Differential Revision: https://reviews.llvm.org/D24523
llvm-svn: 281419
The constant folder didn't know how to always fold bitcasts of constant integer
vectors. In particular, it was unable to handle the case where a constant vector
had some undef elements, and the resulting (i.e. bitcasted) vector type had more
elements than the original vector type.
Example:
%cast = bitcast <2 x i64><i64 undef, i64 2> to <4 x i32>
On a little endian target, %cast could have been folded to:
<4 x i32><i32 undef, i32 undef, i32 2, i32 0>
This patch improves the folding logic by teaching how to correctly propagate
undef elements in the folded vector.
Differential Revision: https://reviews.llvm.org/D24301
llvm-svn: 281343
InstSimplify doesn't always know how to fold a bitcast of a constant vector.
In particular, the logic in InstSimplify doesn't know how to handle the case
where the constant vector in input contains some undef elements, and the
number of elements is smaller than the number of elements of the bitcast
vector type.
llvm-svn: 281332
Teach SimplifyLibcalls that in can treat functions annotated with
apcs, aapcs or aapcs_vfp like normal C functions if they only take
and return integer or pointer values, and the target is not iOS.
Differential Revision: https://reviews.llvm.org/D24453
llvm-svn: 281322
This patch reverses the edge from DIGlobalVariable to GlobalVariable.
This will allow us to more easily preserve debug info metadata when
manipulating global variables.
Fixes PR30362. A program for upgrading test cases is attached to that
bug.
Differential Revision: http://reviews.llvm.org/D20147
llvm-svn: 281284
Trying to infer the 'returned' attribute if an argument is already
'returned' can lead to verification failure: inference might determine
that a different argument is passed through which would result in two
different arguments marked as 'returned'.
This fixes PR30350.
llvm-svn: 281221
This should *actually* fix PR30244. This cranks up the workaround for PR30188 so that we never sink loads or stores of allocas.
The idea is that these should be removed by SROA/Mem2Reg, and any movement of them may well confuse SROA or just cause unwanted code churn. It's not ideal that the midend should be crippled like this, but that unwanted churn can really cause significant regressions in important workloads (tsan).
llvm-svn: 281162
Exposed by PR30244, we will split a block currently if we think we can sink at least one instruction. However this isn't right - the reason we split predecessors is so that we can sink instructions that otherwise couldn't be sunk because it isn't safe to do so - stores, for example.
So, change the heuristic to only split if it thinks it can sink at least one non-speculatable instruction.
Should fix PR30244.
llvm-svn: 281160
Summary:
This will let e.g. the load/store vectorizer propagate this metadata
appropriately.
Reviewers: arsenm
Subscribers: tra, jholewinski, hfinkel, mzolotukhin
Differential Revision: https://reviews.llvm.org/D23479
llvm-svn: 281153
This would create a bitcast use which fails the verifier: swifterror values may
only be used by loads, stores, and as function arguments.
rdar://28233244
llvm-svn: 281114
I was looking to fix a bug in getComplexity(), and these cases showed up as
obvious failures. I'm not sure how to find these in general though.
llvm-svn: 281055
Summary:
If one of the uses of the value is a single edge PHINode, handle it.
Original:
%val = something
<suspend>
%p = PHINode [%val]
After Spill + Part13:
%val = something
%slot = gep val.spill.slot
store %val, %slot
<suspend>
%p = load %slot
Plus tiny fixes/changes:
* use correct index for coro.free in CoroCleanup
* fixup id parameter in coro.free to allow authoring coroutine in plain C with __builtins
Reviewers: majnemer
Subscribers: mehdi_amini, llvm-commits
Differential Revision: https://reviews.llvm.org/D24242
llvm-svn: 281020
Summary: The hoisted instruction is executed speculatively. It could affect the debugging experience as user would see gdb go into code that may not be expected to execute. It will also affect sample profile accuracy by assigning incorrect frequency to source within then/else branch.
Reviewers: davidxl, dblaikie, chandlerc, kcc, echristo
Subscribers: mehdi_amini, probinson, eric_niebler, andreadb, llvm-commits
Differential Revision: https://reviews.llvm.org/D24164
llvm-svn: 280995
The test case included in r280979 wasn't checking what it was supposed to be
checking for the predicated store case. Fixing the test revealed that the
multi-use case (when a pointer is used by both vectorized and scalarized memory
accesses) wasn't being handled properly. We can't skip over
non-consecutive-like pointers since they may have looked consecutive-like with
a different memory access.
llvm-svn: 280992
Previously, all consecutive pointers were marked uniform after vectorization.
However, if a consecutive pointer is used by a memory access that is eventually
scalarized, the pointer won't remain uniform after all. An example is
predicated stores. Even though a predicated store may be consecutive, it will
still be scalarized, making it's pointer operand non-uniform.
This patch updates the logic in collectLoopUniforms to consider the cases where
a memory access may be scalarized. If a memory access may be scalarized, its
pointer operand is not marked uniform. The determination of whether a given
memory instruction will be scalarized or not has been moved into a common
function that is used by the vectorizer, cost model, and legality analysis.
Differential Revision: https://reviews.llvm.org/D24271
llvm-svn: 280979
Summary:
When cloning blocks for prologue/epilogue we need to replicate the loop
structure from the original loop. It wasn't a problem for the innermost
loops, but it led to an incorrect loop info when we unrolled a loop with
a child loop - in this case created prologue-loop had a child loop, but
loop info didn't reflect that.
This fixes PR28888.
Reviewers: chandlerc, sanjoy, hfinkel
Subscribers: llvm-commits, silvas
Differential Revision: https://reviews.llvm.org/D24203
llvm-svn: 280901
We can't create metadata-valued PHIs; don't try to do so when sinking.
I created a test case for this using the @llvm.type.test intrinsic, because it
takes a metadata parameter and does not have severe side effects (thus
SimplifyCFG is willing to otherwise sink it).
Previously, running the test case would crash with:
Invalid use of metadata!
%.sink = select i1 %flag, metadata <...>, metadata <0x4e45dc0>
LLVM ERROR: Broken function found, compilation aborted!
llvm-svn: 280866
This is a revert of r280676 which was a revert of r280637;
ie, this is r280637 again. It was speculatively reverted to
help debug buildbot failures.
llvm-svn: 280861
Summary:
LSV replaces multiple adjacent loads with one vectorized load and a
bunch of extractelement instructions. This patch makes the
extractelement instructions' names match those of the original loads,
for (hopefully) improved readability.
Reviewers: asbirlea, tstellarAMD
Subscribers: arsenm, mzolotukhin
Differential Revision: https://reviews.llvm.org/D23748
llvm-svn: 280818
This fixes a similar issue to the one already fixed by r280804
(revieved in D24256). Revision 280804 fixed the problem with unsafe dyn_casts
in the extrq/extrqi combining logic. However, it turns out that even the
insertq/insertqi logic was affected by the same problem.
llvm-svn: 280807
This patch fixes an assertion failure caused by unsafe dynamic casts on the
constant operands of sse4a intrinsic calls to extrq/extrqi
The combine logic that simplifies sse4a extrq/extrqi intrinsic calls currently
checks if the input operands are constants. Internally, that logic relies on
dyn_casts of values returned by calls to method Constant::getAggregateElement.
However, method getAggregateElemet may return nullptr if the constant element
cannot be retrieved. So, all the dyn_casts can potentially fail. This is what
happens for example if a constexpr value is passed in input to an extrq/extrqi
intrinsic call.
This patch fixes the problem by using a dyn_cast_or_null (instead of a simple
dyn_cast) on the result of each call to Constant::getAggregateElement.
Added reproducible test cases to x86-sse4a.ll.
Differential Revision: https://reviews.llvm.org/D24256
llvm-svn: 280804
I should have realised this the first time around, but if we're avoiding sinking stores where the operands come from allocas so they don't create selects, we also have to do the same for loads because SROA will be just as defective looking at loads of selected addresses as stores.
Fixes PR30188 (again).
llvm-svn: 280792
PR30292 showed a case where our PHI checking wasn't correct. We were checking that all values were used by the same PHI before deciding to sink, but we weren't checking that the incoming values for that PHI were what we expected. As a result, we had to bail out after block splitting which caused us to never reach a steady state in SimplifyCFG.
Fixes PR30292.
llvm-svn: 280790
Currently the pass updates branch weights in the IR if the function has
any PGO info (entry frequency is set). However we could still have
regions of the CFG that does not have branch weights collected (e.g. a
cold region). In this case we'd use static estimates. Since static
estimates for branches are determined independently, they are
inconsistent. Updating them can "randomly" inflate block frequencies.
I've run into this in a completely cold loop of h264ref from
SPEC. -Rpass-with-hotness showed the loop to be completely cold during
inlining (before JT) but completely hot during vectorization (after JT).
The new testcase demonstrate the problem. We check array elements
against 1, 2 and 3 in a loop. The check against 3 is the loop-exiting
check. The block names should be self-explanatory.
In this example, jump threading incorrectly updates the weight of the
loop-exiting branch to 0, drastically inflating the frequency of the
loop (in the range of billions).
There is no run-time profile info for edges inside the loop, so branch
probabilities are estimated. These are the resulting branch and block
frequencies for the loop body:
check_1 (16)
(8) / |
eq_1 | (8)
\ |
check_2 (16)
(8) / |
eq_2 | (8)
\ |
check_3 (16)
(1) / |
(loop exit) | (15)
|
(back edge)
First we thread eq_1 -> check_2 to check_3. Frequencies are updated to
remove the frequency of eq_1 from check_2 and then from the false edge
leaving check_2. Changed frequencies are highlighted with * *:
check_1 (16)
(8) / |
eq_1~ | (8)
/ |
/ check_2 (*8*)
/ (8) / |
\ eq_2 | (*0*)
\ \ |
` --- check_3 (16)
(1) / |
(loop exit) | (15)
|
(back edge)
Next we thread eq_1 -> check_3 and eq_2 -> check_3 to check_1 as new
back edges. Frequencies are updated to remove the frequency of eq_1 and
eq_3 from check_3 and then the false edge leaving check_3 (changed
frequencies are highlighted with * *):
check_1 (16)
(8) / |
eq_1~ | (8)
/ |
/ check_2 (*8*)
/ (8) / |
/-- eq_2~ | (*0*)
(back edge) |
check_3 (*0*)
(*0*) / |
(loop exit) | (*0*)
|
(back edge)
As a result, the loop exit edge ends up with 0 frequency which in turn makes
the loop header to have maximum frequency.
There are a few potential problems here:
1. The profile data seems odd. There is a single profile sample of the
loop being entered. On the other hand, there are no weights inside the
loop.
2. Based on static estimation we shouldn't set edges to "extreme"
values, i.e. extremely likely or unlikely.
3. We shouldn't create profile metadata that is calculated from static
estimation. I am not sure what policy is but it seems to make sense to
treat profile metadata as something that is known to originate from
profiling. Estimated probabilities should only be reflected in BPI/BFI.
Any one of these would probably fix the immediate problem. I went for 3
because I think it's a good policy to have and added a FIXME about 2.
Differential Revision: https://reviews.llvm.org/D24118
llvm-svn: 280713
Summary:
Move early uses of spilled variables after CoroBegin.
For example, if a parameter had address taken, we may end up with the code
like:
define @f(i32 %n) {
%n.addr = alloca i32
store %n, %n.addr
...
call @coro.begin
This patch fixes the problem by moving uses of spilled variables after CoroBegin.
Reviewers: majnemer
Subscribers: mehdi_amini, llvm-commits
Differential Revision: https://reviews.llvm.org/D24234
llvm-svn: 280678
This test code previously caused a failure in the module verifier,
because SimplifyCFG created this invalid instruction, which tries to
take the address of inline asm:
%.sink = select i1 %1, i64 ()* asm "mov $0, #1", "=r", i64 ()* asm %"mov $0, #2", "=r"
This has been fixed recently, presumably by James Molloy's patches that
re-wrote and changed parts of SimplifyCFG, so this patch just adds a
regression test for it.
Differential Revision: https://reviews.llvm.org/D24231
llvm-svn: 280660
Summary:
A frontend may designate a particular suspend to be final, by setting the second argument of the coro.suspend intrinsic to true. Such a suspend point has two properties:
* it is possible to check whether a suspended coroutine is at the final suspend point via coro.done intrinsic;
* a resumption of a coroutine stopped at the final suspend point leads to undefined behavior. The only possible action for a coroutine at a final suspend point is destroying it via coro.destroy intrinsic.
This patch adds final suspend handling logic to CoroEarly and CoroSplit passes.
Now, the final suspend point example from docs\Coroutines.rst compiles and produces expected result (see test/Transform/Coroutines/ex5.ll).
Reviewers: majnemer
Subscribers: mehdi_amini, llvm-commits
Differential Revision: https://reviews.llvm.org/D24068
llvm-svn: 280646
memcpy with ld/st.
When InstCombine replaces a memcpy with loads+stores it does not copy over the
llvm.mem.parallel_loop_access from the memcpy instruction. This patch fixes
that.
Differential Revision: https://reviews.llvm.org/D23499
llvm-svn: 280617
Summary:
The inliner may need to determine where a given funclet unwinds to,
and this determination may depend on other funclets throughout the
funclet tree. The code that performs this walk in getUnwindDestToken
memoizes results to avoid redundant computations. In the case that
a funclet's unwind destination is derived from its ancestor, there's
code to walk back down the tree from the ancestor updating the memo
map of its descendants to record the unwind destination. This change
fixes that code to account for the case that some descendant has a
different unwind destination, which can happen if that unwind dest
is a descendant of the EHPad being queried and thus didn't determine
its unwind destination.
Also update test inline-funclets.ll, which is supposed to cover such
scenarios, to include a case that fails an assertion without this fix
but passes with it.
Fixes PR29151.
Reviewers: majnemer
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D24117
llvm-svn: 280610
For the store of a wide value merged from a pair of values, especially int-fp pair,
sometimes it is more efficent to split it into separate narrow stores, which can
remove the bitwise instructions or sink them to colder places.
Now the feature is only enabled on x86 target, and only store of int-fp pair is
splitted. It is possible that the application scope gets extended with perf evidence
support in the future.
Differential Revision: https://reviews.llvm.org/D22840
llvm-svn: 280505
The motivating case occurs with SSE/AVX scalar intrinsics, so this is a first step towards
shrinking that to a single shufflevector.
Note that the transform is intentionally limited to shuffles that are equivalent to vector
selects to avoid creating arbitrary shuffle masks that may not lower well.
This should solve PR29126:
https://llvm.org/bugs/show_bug.cgi?id=29126
Differential Revision: https://reviews.llvm.org/D23886
llvm-svn: 280504
For uniform instructions, we're only required to generate a scalar value for
the first vector lane of each unroll iteration. Thus, if we have a reverse
interleaved group, computing the member index off the scalar GEP corresponding
to the last vector lane of its pointer operand technically makes the GEP
non-uniform. We should compute the member index off the first scalar GEP
instead.
I've added the updated member index computation to the existing reverse
interleaved group test.
llvm-svn: 280497
This patch fixes a crash caused by an incorrect folding of an ordered comparison
between a packed floating point vector and a splat vector of NaN.
An ordered comparison between a vector and a constant vector of NaN, should
always be folded into a constant vector where each element is i1 false.
Since revision 266175, SimplifyFCmpInst folds the ordered fcmp into a scalar
'false'. Later on, this would cause an assertion failure, since the value type
of the folded value doesn't match the expected value type of the uses of the
original instruction: "Assertion failed: New->getType() == getType() &&
"replaceAllUses of value with new value of different type!".
This patch fixes the issue and adds a test case to the already existing test
InstSimplify/floating-point-compares.ll.
Differential Revision: https://reviews.llvm.org/D24143
llvm-svn: 280488
We're sinking stores, which is a good thing, but in the process creating selects for the store address operand, which SROA/Mem2Reg can't look through, which caused serious regressions.
The real fix is in SROA, which I'll be looking into.
llvm-svn: 280470
While removing a scalar shackle from an icmp fold, I noticed that I couldn't find any tests to trigger
this code path.
The 'and' shrinking transform should be handled by InstCombiner::foldCastedBitwiseLogic()
or eliminated with InstSimplify. The icmp narrowing is part of InstCombiner::foldICmpWithCastAndCast().
Differential Revision: https://reviews.llvm.org/D24031
llvm-svn: 280370
This was a real restriction in the original version of SinkIfThenCodeToEnd. Now it's been rewritten, the restriction can be lifted.
As part of this, we handle a very common and useful case where one of the incoming branches is actually conditional. Consider:
if (a)
x(1);
else if (b)
x(2);
This produces the following CFG:
[if]
/ \
[x(1)] [if]
| | \
| | \
| [x(2)] |
\ | /
[ end ]
[end] has two unconditional predecessor arcs and one conditional. The conditional refers to the implicit empty 'else' arc. This same pattern can also be caused by an empty default block in a switch.
We can't sink the call to x() down to end because no call to x() happens on the third incoming arc (assume that x() has sideeffects for the sake of argument; if something is safe to speculate we could indeed sink nevertheless but this cannot happen in the general case and causes many extra selects).
We are now able to detect this case and split off the unconditional arcs to a common successor:
[if]
/ \
[x(1)] [if]
| | \
| | \
| [x(2)] |
\ / |
[sink.split] |
\ /
[ end ]
Now we can sink the call to x() into %sink.split. This can cause significant code simplification in many testcases.
llvm-svn: 280364
r279460 rewrote this function to be able to handle more than two incoming edges and took pains to ensure this didn't regress anything.
This time we change the logic for determining if an instruction should be sunk. Previously we used a single pass greedy algorithm - sink instructions until one requires more than one PHI node or we run out of instructions to sink.
This had the problem that sinking instructions that had non-identical but trivially the same operands needed extra logic so we sunk them aggressively. For example:
%a = load i32* %b %d = load i32* %b
%c = gep i32* %a, i32 0 %e = gep i32* %d, i32 1
Sinking %c and %e would naively require two PHI merges as %a != %d. But the loads are obviously equivalent (and maybe can't be hoisted because there is no common predecessor).
This is why we implemented the fairly complex function areValuesTriviallySame(), to look through trivial differences like this. However it's just not clever enough.
Instead, throw areValuesTriviallySame away, use pointer equality to check equivalence of operands and switch to a two-stage algorithm.
In the "scan" stage, we look at every sinkable instruction in isolation from end of block to front. If it's sinkable, we keep track of all operands that required PHI merging.
In the "sink" stage, we iteratively sink the last non-terminator in the source blocks. But when calculating how many PHIs are actually required to be inserted (to work out if we should stop or not) we remove any values that have already been sunk from the set of PHI-merges required, which allows us to be more aggressive.
This turns an algorithm with potentially recursive lookahead (looking through GEPs, casts, loads and any other instruction potentially not CSE'd) to two linear scans.
llvm-svn: 280351
As discussed in https://reviews.llvm.org/D22666, our current mechanism to
support -pg profiling, where we insert calls to mcount(), or some similar
function, is fundamentally broken. We insert these calls in the frontend, which
means they get duplicated when inlining, and so the accumulated execution
counts for the inlined-into functions are wrong.
Because we don't want the presence of these functions to affect optimizaton,
they should be inserted in the backend. Here's a pass which would do just that.
The knowledge of the name of the counting function lives in the frontend, so
we're passing it here as a function attribute. Clang will be updated to use
this mechanism.
Differential Revision: https://reviews.llvm.org/D22825
llvm-svn: 280347
-fprofile-dir=path allows the user to specify where .gcda files should be
emitted when the program is run. In particular, this is the first flag that
causes the .gcno and .o files to have different paths, LLVM is extended to
support this. -fprofile-dir= does not change the file name in the .gcno (and
thus where lcov looks for the source) but it does change the name in the .gcda
(and thus where the runtime library writes the .gcda file). It's different from
a GCOV_PREFIX because a user can observe that the GCOV_PREFIX_STRIP will strip
paths off of -fprofile-dir= but not off of a supplied GCOV_PREFIX.
To implement this we split -coverage-file into -coverage-data-file and
-coverage-notes-file to specify the two different names. The !llvm.gcov
metadata node grows from a 2-element form {string coverage-file, node dbg.cu}
to 3-elements, {string coverage-notes-file, string coverage-data-file, node
dbg.cu}. In the 3-element form, the file name is already "mangled" with
.gcno/.gcda suffixes, while the 2-element form left that to the middle end
pass.
llvm-svn: 280306
Summary:
Use MemorySSA, if requested, to do less conservative memory dependency
checking.
This change doesn't enable the MemorySSA enhanced EarlyCSE in the
default pipelines, so should be NFC.
Reviewers: dberlin, sanjoy, reames, majnemer
Subscribers: mcrosier, llvm-commits
Differential Revision: http://reviews.llvm.org/D19821
llvm-svn: 280279
This is a first step towards supporting deopt value lowering and reporting entirely with the register allocator. I hope to build on this in the near future to support live-on-return semantics, but I have a use case which allows me to test and investigate code quality with just the live-in semantics so I've chosen to start there. For those curious, my use cases is our implementation of the "__llvm_deoptimize" function we bind to @llvm.deoptimize. I'm choosing not to hard code that fact in the patch and instead make it configurable via function attributes.
The basic approach here is modelled on what is done for the "Live In" values on stackmaps and patchpoints. (A secondary goal here is to remove one of the last barriers to merging the pseudo instructions.) We start by adding the operands directly to the STATEPOINT SDNode. Once we've lowered to MI, we extend the remat logic used by the register allocator to fold virtual register uses into StackMap::Indirect entries as needed. This does rely on the fact that the register allocator rematerializes. If it didn't along some code path, we could end up with more vregs than physical registers and fail to allocate.
Today, we *only* fold in the register allocator. This can create some weird effects when combined with arguments passed on the stack because we don't fold them appropriately. I have an idea how to fix that, but it needs this patch in place to work on that effectively. (There's some weird interaction with the scheduler as well, more investigation needed.)
My near term plan is to land this patch off-by-default, experiment in my local tree to identify any correctness issues and then start fixing codegen problems one by one as I find them. Once I have the live-in lowering fully working (both correctness and code quality), I'm hoping to move on to the live-on-return semantics. Note: I don't have any *known* miscompiles with this patch enabled, but I'm pretty sure I'll find at least a couple. Thus, the "experimental" tag and the fact it's off by default.
Differential Revision: https://reviews.llvm.org/D24000
llvm-svn: 280250
We check that a sinking candidate is used by only one PHI node during our legality checks. However for instructions that are used by other sinking candidates our heuristic is less conservative. This can result in a candidate actually being illegal when we come to sink it because of how we sunk a predecessor. Do the used-by-only-one-PHI checks again during sinking to ensure we don't crash.
llvm-svn: 280228
We're sinking stores, which is a good thing, but in the process creating selects for the store address operand, which SROA/Mem2Reg can't look through, which caused serious regressions.
The real fix is in SROA, which I'll be looking into.
llvm-svn: 280219
This was a real restriction in the original version of SinkIfThenCodeToEnd. Now it's been rewritten, the restriction can be lifted.
As part of this, we handle a very common and useful case where one of the incoming branches is actually conditional. Consider:
if (a)
x(1);
else if (b)
x(2);
This produces the following CFG:
[if]
/ \
[x(1)] [if]
| | \
| | \
| [x(2)] |
\ | /
[ end ]
[end] has two unconditional predecessor arcs and one conditional. The conditional refers to the implicit empty 'else' arc. This same pattern can also be caused by an empty default block in a switch.
We can't sink the call to x() down to end because no call to x() happens on the third incoming arc (assume that x() has sideeffects for the sake of argument; if something is safe to speculate we could indeed sink nevertheless but this cannot happen in the general case and causes many extra selects).
We are now able to detect this case and split off the unconditional arcs to a common successor:
[if]
/ \
[x(1)] [if]
| | \
| | \
| [x(2)] |
\ / |
[sink.split] |
\ /
[ end ]
Now we can sink the call to x() into %sink.split. This can cause significant code simplification in many testcases.
llvm-svn: 280217
r279460 rewrote this function to be able to handle more than two incoming edges and took pains to ensure this didn't regress anything.
This time we change the logic for determining if an instruction should be sunk. Previously we used a single pass greedy algorithm - sink instructions until one requires more than one PHI node or we run out of instructions to sink.
This had the problem that sinking instructions that had non-identical but trivially the same operands needed extra logic so we sunk them aggressively. For example:
%a = load i32* %b %d = load i32* %b
%c = gep i32* %a, i32 0 %e = gep i32* %d, i32 1
Sinking %c and %e would naively require two PHI merges as %a != %d. But the loads are obviously equivalent (and maybe can't be hoisted because there is no common predecessor).
This is why we implemented the fairly complex function areValuesTriviallySame(), to look through trivial differences like this. However it's just not clever enough.
Instead, throw areValuesTriviallySame away, use pointer equality to check equivalence of operands and switch to a two-stage algorithm.
In the "scan" stage, we look at every sinkable instruction in isolation from end of block to front. If it's sinkable, we keep track of all operands that required PHI merging.
In the "sink" stage, we iteratively sink the last non-terminator in the source blocks. But when calculating how many PHIs are actually required to be inserted (to work out if we should stop or not) we remove any values that have already been sunk from the set of PHI-merges required, which allows us to be more aggressive.
This turns an algorithm with potentially recursive lookahead (looking through GEPs, casts, loads and any other instruction potentially not CSE'd) to two linear scans.
llvm-svn: 280216
This was deliberately disabled during my rewrite of SinkIfThenToEnd to keep behaviour
at least vaguely consistent with the previous version and keep it as close to NFC as
I could.
There's no real reason not to merge sideeffect calls though, so let's do it! Small fixup
along the way to ensure we don't create indirect calls.
Should fix PR28964.
llvm-svn: 280215
Summary:
1) CoroEarly now lowers llvm.coro.promise intrinsic that allows to obtain
a coroutine promise pointer from a coroutine frame and vice versa.
2) CoroFrame now interprets Promise argument of llvm.coro.begin to
place CoroutinPromise alloca at a deterministic offset from the coroutine frame.
Now, the coroutine promise example from docs\Coroutines.rst compiles and produces expected result (see test/Transform/Coroutines/ex4.ll).
Reviewers: majnemer
Subscribers: llvm-commits, mehdi_amini
Differential Revision: https://reviews.llvm.org/D23993
llvm-svn: 280184
Summary:
LSV was using two vector sets (heads and tails) to track pairs of adjiacent position to vectorize.
A recent optimization is trying to obtain the longest chain to vectorize and assumes the positions
in heads(H) and tails(T) match, which is not the case is there are multiple tails for the same head.
e.g.:
i1: store a[0]
i2: store a[1]
i3: store a[1]
Leads to:
H: i1
T: i2 i3
Instead of:
H: i1 i1
T: i2 i3
So the positions for instructions that follow i3 will have different indexes in H/T.
This patch resolves PR29148.
This issue also surfaced the fact that if the chain is too long, and TLI
returns a "not-fast" answer, the whole chain will be abandoned for
vectorization, even though a smaller one would be beneficial.
Added a testcase and FIXME for this.
Reviewers: tstellarAMD, arsenm, jlebar
Subscribers: mzolotukhin, wdng, llvm-commits
Differential Revision: https://reviews.llvm.org/D24057
llvm-svn: 280179
We don't need to limit predication to blocks that have a single incoming
edge, we just need to use the right mask.
This fixes PR30172.
Differential Revision: https://reviews.llvm.org/D24009
llvm-svn: 280148
Summary:
Fix a couple issues limiting the application of indirect call promotion
in ThinLTO mode:
- Invoke indirect call promotion before globalopt, since it may
eliminate imported functions which appear unreferenced.
- Invoke indirect call promotion with InLTO=true so that the PGOFuncName
metadata is used to get the name for locals which would have been
renamed during promotion.
Reviewers: davidxl, mehdi_amini
Subscribers: Prazek, llvm-commits, mehdi_amini
Differential Revision: https://reviews.llvm.org/D24004
llvm-svn: 280024
After r279649 when getting a vector value from VectorLoopValueMap, we create an
insertelement sequence on-demand if the value has been scalarized instead of
vectorized. We previously inserted this insertelement sequence before the
value's first vector user. However, this insert location is problematic if that
user is the phi node of a first-order recurrence. With this patch, we move the
insertelement sequence after the last scalar instruction we created when
scalarizing the value. Thus, the value's vector definition in the new loop will
immediately follow its scalar definitions. This should fix PR30183.
Reference: https://llvm.org/bugs/show_bug.cgi?id=30183
llvm-svn: 280001
Summary:
While walking the use chain for identifying rematerializable values in RS4GC,
add the case where the current value and base value are the same PHI nodes.
This will aid rematerialization of geps and casts instead of relocating.
Reviewers: sanjoy, reames, igor
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D23920
llvm-svn: 279975
Assuming the default FP env, we should not treat fdiv and frem any differently in terms of
trapping behavior than any other FP op. Ie, FP ops do not trap with the default FP env.
This matches how we treat the fdiv/frem in IR with isSafeToSpeculativelyExecute() and in
the backend after:
https://reviews.llvm.org/rL279970
llvm-svn: 279973
Summary:
[Coroutines] Part 9: Add cleanup subfunction.
This patch completes coroutine heap allocation elision. Now, the heap elision example from docs\Coroutines.rst compiles and produces expected result (see test/Transform/Coroutines/ex3.ll)
Intrinsic Changes:
* coro.free gets a token parameter tying it to coro.id to allow reliably discovering all coro.frees associated with a particular coroutine.
* coro.id gets an extra parameter that points back to a coroutine function. This allows to check whether a coro.id describes the enclosing function or it belongs to a different function that was later inlined.
CoroSplit now creates three subfunctions:
# f$resume - resume logic
# f$destroy - cleanup logic, followed by a deallocation code
# f$cleanup - just the cleanup code
CoroElide pass during devirtualization replaces coro.destroy with either f$destroy or f$cleanup depending whether heap elision is performed or not.
Other fixes, improvements:
* Fixed buglet in Shape::buildFrame that was not creating coro.save properly if coroutine has more than one suspend point.
* Switched to using variable width suspend index field (no longer limited to 32 bit index field can be as little as i1 or as large as i<whatever-size_t-is>)
Reviewers: majnemer
Subscribers: llvm-commits, mehdi_amini
Differential Revision: https://reviews.llvm.org/D23844
llvm-svn: 279971
Fixed a bug in run-time checks for possible memory conflicts inside loop.
The bug is in Low <-> High boundaries calculation. The High boundary should be calculated as "last memory access pointer + element size".
Differential revision: https://reviews.llvm.org/D23176
llvm-svn: 279930
Summary:
This is obviously an interesting case because it may motivate code
restructuring or LTO.
Reporting this requires instantiation of ORE in the loop where the call
sites are first gathered. I've checked compile-time
overhead *with* -Rpass-with-hotness and the worst slow-down was 6% in
mcf and quickly tailing off. As before without -Rpass-with-hotness
there is no overhead.
Because this could be a pretty noisy diagnostics, it is currently
qualified as 'verbose'. As of this patch, 'verbose' diagnostics are
only emitted with -Rpass-with-hotness, i.e. when the output is expected
to be filtered.
Reviewers: eraman, chandlerc, davidxl, hfinkel
Subscribers: tejohnson, Prazek, davide, llvm-commits
Differential Revision: https://reviews.llvm.org/D23415
llvm-svn: 279860
Summary:
This fixes pr29105. The reason is that lifetime marks creates new
aliasing pointers the original ones, but before this patch aliases
were not checked in performMemCpyToMemSetOptzn.
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D23846
llvm-svn: 279769
It is invalid to hoist stores or loads if they are not executed on all paths
from the hoisting point to the exit of the function. In the testcase, there are
paths in the loop that do not execute the stores or the loads, and so hoisting
them within the loop is unsafe.
The problem is that the current implementation of hoistingFromAllPaths is
incomplete: it walks all blocks dominated by the hoisting point, and does not
return false when the loop contains a path on which the hoisted ld/st is
not executed.
Differential Revision: https://reviews.llvm.org/D23843
llvm-svn: 279732
This patch unifies the data structures we use for mapping instructions from the
original loop to their corresponding instructions in the new loop. Previously,
we maintained two distinct maps for this purpose: WidenMap and ScalarIVMap.
WidenMap maintained the vector values each instruction from the old loop was
represented with, and ScalarIVMap maintained the scalar values each scalarized
induction variable was represented with. With this patch, all values created
for the new loop are maintained in VectorLoopValueMap.
The change allows for several simplifications. Previously, when an instruction
was scalarized, we had to insert the scalar values into vectors in order to
maintain the mapping in WidenMap. Then, if a user of the scalarized value was
also scalar, we had to extract the scalar values from the temporary vector we
created. We now aovid these unnecessary scalar-to-vector-to-scalar conversions.
If a scalarized value is used by a scalar instruction, the scalar value is used
directly. However, if the scalarized value is needed by a vector instruction,
we generate the needed insertelement instructions on-demand.
A common idiom in several locations in the code (including the scalarization
code), is to first get the vector values an instruction from the original loop
maps to, and then extract a particular scalar value. This patch adds
getScalarValue for this purpose along side getVectorValue as an interface into
VectorLoopValueMap. These functions work together to return the requested
values if they're available or to produce them if they're not.
The mapping has also be made less permissive. Entries can be added to
VectorLoopValue map with the new initVector and initScalar functions.
getVectorValue has been modified to return a constant reference to the mapped
entries.
There's no real functional change with this patch; however, in some cases we
will generate slightly different code. For example, instead of an insertelement
sequence following the definition of an instruction, it will now precede the
first use of that instruction. This can be seen in the test case changes.
Differential Revision: https://reviews.llvm.org/D23169
llvm-svn: 279649
I'm not sure if the `!isa<CallInst>(Inst) &&
!isa<TerminatorInst>(Inst))` bit is correct either, but this fixes the
case we know is broken.
llvm-svn: 279647
div/rem instructions in basic blocks that require predication currently prevent
vectorization. This patch extends the existing mechanism for predicating stores
to handle other instructions and leverages it to predicate divs and rems.
Differential Revision: https://reviews.llvm.org/D22918
llvm-svn: 279620
Summary:
This patch adds coroutine frame building algorithm. Now, simple coroutines such as ex0.ll and ex1.ll (first examples from docs\Coroutines.rst can be compiled).
Documentation and overview is here: http://llvm.org/docs/Coroutines.html.
Upstreaming sequence (rough plan)
1.Add documentation. (https://reviews.llvm.org/D22603)
2.Add coroutine intrinsics. (https://reviews.llvm.org/D22659)
...
7. Split coroutine into subfunctions. (https://reviews.llvm.org/D23461)
8. Coroutine Frame Building algorithm <= we are here
9. Add f.cleanup subfunction.
10+. The rest of the logic
Reviewers: majnemer
Subscribers: mehdi_amini, llvm-commits
Differential Revision: https://reviews.llvm.org/D23586
llvm-svn: 279609
The test case included with r279125 exposed an existing signed integer
overflow. Since getTreeCost can return INT_MAX, we can't sum this cost together
with other costs, such as getReductionCost.
This patch removes the possibility of assigning a cost of INT_MAX. Since we
were previously using INT_MAX as an indicator for "should not vectorize", we
now explicitly check this condition with "isTreeTinyAndNotFullyVectorizable"
before computing a cost.
This patch adds a run-line to the test case used for r279125 that ensures we
don't vectorize. Previously, this line would vectorize the test case by chance
due to undefined behavior in the cost calculation.
Differential Revision: https://reviews.llvm.org/D23723
llvm-svn: 279562
...because like the corresponding code, this is just too big to keep adding to.
And the next step is to add a vector version of each of these tests to show
missed folds.
Also, auto-generate CHECK lines and add comments for the tests that correspond to
the source code.
llvm-svn: 279530
[Recommitting now an unrelated assertion in SROA is sorted out]
The new version has several advantages:
1) IMSHO it's more readable and neater
2) It handles loads and stores properly
3) It can handle any number of incoming blocks rather than just two. I'll be taking advantage of this in a followup patch.
With this change we can now finally sink load-modify-store idioms such as:
if (a)
return *b += 3;
else
return *b += 4;
=>
%z = load i32, i32* %y
%.sink = select i1 %a, i32 5, i32 7
%b = add i32 %z, %.sink
store i32 %b, i32* %y
ret i32 %b
When this works for switches it'll be even more powerful.
Round 4. This time we should handle all instructions correctly, and not replace any operands that need to be constant with variables.
This was really hard to determine safely, so the helper function should be put into the Instruction API. I'll do that as a followup.
llvm-svn: 279460
Summary: We can allow sinking if the single user block has only one unique predecessor, regardless of the number of edges. Note that a switch statement with multiple cases can have the same destination.
Reviewers: mcrosier, majnemer, spatel, reames
Subscribers: reames, mcrosier, llvm-commits
Differential Revision: https://reviews.llvm.org/D23722
llvm-svn: 279448
The new version has several advantages:
1) IMSHO it's more readable and neater
2) It handles loads and stores properly
3) It can handle any number of incoming blocks rather than just two. I'll be taking advantage of this in a followup patch.
With this change we can now finally sink load-modify-store idioms such as:
if (a)
return *b += 3;
else
return *b += 4;
=>
%z = load i32, i32* %y
%.sink = select i1 %a, i32 5, i32 7
%b = add i32 %z, %.sink
store i32 %b, i32* %y
ret i32 %b
When this works for switches it'll be even more powerful.
Round 4. This time we should handle all instructions correctly, and not replace any operands that need to be constant with variables.
This was really hard to determine safely, so the helper function should be put into the Instruction API. I'll do that as a followup.
llvm-svn: 279443
This change cause performance regression on MultiSource/Benchmarks/TSVC/Symbolics-flt/Symbolics-flt from LNT and some other bechmarks.
See https://reviews.llvm.org/D18777 for details.
llvm-svn: 279433
This change needs to be reverted in order to revert -r278267 which cause performance regression on MultiSource/Benchmarks/TSVC/Symbolics-flt/Symbolics-flt from LNT and some other bechmarks.
See comments on https://reviews.llvm.org/D18777 for details.
llvm-svn: 279432
The test case included in r279125 exposed existing undefined behavior in the
SLP vectorizer that it did not introduce. This patch reapplies the original
patch, but modifies the test case to avoid hitting the undefined behavior. This
allows us to close PR28330 while keeping the UBSan bot happy. The undefined
behavior the original test uncovered will be addressed in a follow-on patch.
Reference: https://llvm.org/bugs/show_bug.cgi?id=28330
llvm-svn: 279370
This is a partial enablement (move the ConstantInt guard down) because there are many
different folds here and one of the later ones will require reworking 'isSignBitCheck'.
llvm-svn: 279339
The intended transform is:
// Simplify icmp eq (or (ptrtoint P), (ptrtoint Q)), 0
// -> and (icmp eq P, null), (icmp eq Q, null).
P and Q are both pointer types, but may have different types. We need
two calls to getNullValue() to make the icmps.
llvm-svn: 279271
CGSCC use a WeakVH to track call sites. RAUW a call within a function
can result in that WeakVH getting confused about whether or not the call
site is still around.
llvm-svn: 279268
Of course, we really need to refactor and fix all of the cmp predicates,
but this one is interesting because without it, we later perform an
information-losing transform of icmp (shl 1, Y), C, and we can't recover
the better fold.
llvm-svn: 279263
The new version has several advantages:
1) IMSHO it's more readable and neater
2) It handles loads and stores properly
3) It can handle any number of incoming blocks rather than just two. I'll be taking advantage of this in a followup patch.
With this change we can now finally sink load-modify-store idioms such as:
if (a)
return *b += 3;
else
return *b += 4;
=>
%z = load i32, i32* %y
%.sink = select i1 %a, i32 5, i32 7
%b = add i32 %z, %.sink
store i32 %b, i32* %y
ret i32 %b
When this works for switches it'll be even more powerful.
llvm-svn: 279229
We abort building vectorizable trees in some cases (e.g., if the maximum
recursion depth is reached, if the region size is too large, etc.). If this
happens for a reduction, we can be left with a root entry that needs to be
gathered. For these cases, we need make sure we actually set VectorizedValue to
the resulting vector.
This patch ensures we properly set VectorizedValue, and it also ensures the
insertelement sequence generated for the gathers is inserted at the correct
location.
Reference: https://llvm.org/bugs/show_bug.cgi?id=28330
Differential Revison: https://reviews.llvm.org/D23410
llvm-svn: 279125
It causes a regression on our internal benchmark. Introduce cvp-dont-process flag and set it off by default while investigating the regression.
llvm-svn: 279082
Also, add a scalar test to demonstrate one of the intermediate folds that
is necessary to accomplish the existing, multi-step test. And simplify
the vector tests to only check the final piece of that multi-step transform.
llvm-svn: 278995
minimal and boring form than the old pass manager's version.
This pass does the very minimal amount of work necessary to inline
functions declared as always-inline. It doesn't support a wide array of
things that the legacy pass manager did support, but is alse ... about
20 lines of code. So it has that going for it. Notably things this
doesn't support:
- Array alloca merging
- To support the above, bottom-up inlining with careful history
tracking and call graph updates
- DCE of the functions that become dead after this inlining.
- Inlining through call instructions with the always_inline attribute.
Instead, it focuses on inlining functions with that attribute.
The first I've omitted because I'm hoping to just turn it off for the
primary pass manager. If that doesn't pan out, I can add it here but it
will be reasonably expensive to do so.
The second should really be handled by running global-dce after the
inliner. I don't want to re-implement the non-trivial logic necessary to
do comdat-correct DCE of functions. This means the -O0 pipeline will
have to be at least 'always-inline,global-dce', but that seems
reasonable to me. If others are seriously worried about this I'd like to
hear about it and understand why. Again, this is all solveable by
factoring that logic into a utility and calling it here, but I'd like to
wait to do that until there is a clear reason why the existing
pass-based factoring won't work.
The final point is a serious one. I can fairly easily add support for
this, but it seems both costly and a confusing construct for the use
case of the always inliner running at -O0. This attribute can of course
still impact the normal inliner easily (although I find that
a questionable re-use of the same attribute). I've started a discussion
to sort out what semantics we want here and based on that can figure out
if it makes sense ta have this complexity at O0 or not.
One other advantage of this design is that it should be quite a bit
faster due to checking for whether the function is a viable candidate
for inlining exactly once per function instead of doing it for each call
site.
Anyways, hopefully a reasonable starting point for this pass.
Differential Revision: https://reviews.llvm.org/D23299
llvm-svn: 278896
It is pretty easy to get it down to O(nlogn + mlogm). This
implementation has the added benefit of automatically deduplicating
entries between the two sets.
llvm-svn: 278837
I have audited all the callers of concatenate and none require duplicate
entries to service concatenation.
These duplicates serve no purpose but to needlessly embiggen the IR.
N.B. Layering getMostGenericAliasScope on top of concatenate makes it
O(nlogn + mlogm) instead of O(n*m).
llvm-svn: 278836
Summary:
This patch adds simple coroutine splitting logic to CoroSplit pass.
Documentation and overview is here: http://llvm.org/docs/Coroutines.html.
Upstreaming sequence (rough plan)
1.Add documentation. (https://reviews.llvm.org/D22603)
2.Add coroutine intrinsics. (https://reviews.llvm.org/D22659)
...
7. Split coroutine into subfunctions <= we are here
8. Coroutine Frame Building algorithm
9. Handle coroutine with unwinds
10+. The rest of the logic
Reviewers: majnemer
Subscribers: llvm-commits, mehdi_amini
Differential Revision: https://reviews.llvm.org/D23461
llvm-svn: 278830
This reverts commit r278660.
It causes downstream assertion failure in InstCombine on shuffle
instructions. Comes up in __mm_swizzle_epi32.
llvm-svn: 278672
The new version has several advantages:
1) IMSHO it's more readable and neater
2) It handles loads and stores properly
3) It can handle any number of incoming blocks rather than just two. I'll be taking advantage of this in a followup patch.
With this change we can now finally sink load-modify-store idioms such as:
if (a)
return *b += 3;
else
return *b += 4;
=>
%z = load i32, i32* %y
%.sink = select i1 %a, i32 5, i32 7
%b = add i32 %z, %.sink
store i32 %b, i32* %y
ret i32 %b
When this works for switches it'll be even more powerful.
llvm-svn: 278660
If a loop is not rotated (for example when optimizing for size), the latch is not the backedge. If we promote an expression to post-inc form, we not only increase register pressure and add a COPY for that IV expression but for all IVs!
Motivating testcase:
void f(float *a, float *b, float *c, int n) {
while (n-- > 0)
*c++ = *a++ + *b++;
}
It's imperative that the pointer increments be located in the latch block and not the header block; if not, we cannot use post-increment loads and stores and we have to keep both the post-inc and pre-inc values around until the end of the latch which bloats register usage.
llvm-svn: 278658
IRCE has the ability to further version pre-loops and post-loops that it
created, but this isn't useful at all. This change teaches IRCE to
leave behind some metadata in the loops it creates (by cloning the main
loop) so that these new loops are not re-processed by IRCE.
Today this bug is hidden by another bug -- IRCE does not update LoopInfo
properly so the loop pass manager does not re-invoke IRCE on the loops
it split out. However, once the latter is fixed the bug addressed in
this change causes IRCE to infinite-loop in some cases (e.g. it splits
out a pre-loop, a pre-pre-loop from that, a pre-pre-pre-loop from that
and so on).
llvm-svn: 278617
The (negative) test case is supposed to check that IRCE does not muck
with range checks it cannot handle, not that it does the right thing in
the absence of profiling information.
llvm-svn: 278612
Loops containing `indirectbr` may not be in simplified form, even after
running LoopSimplify. Reject then gracefully, instead of tripping an
assert.
llvm-svn: 278611
Summary:
Refactor the existing support into a LoopDataPrefetch implementation
class and a LoopDataPrefetchLegacyPass class that invokes it.
Add a new LoopDataPrefetchPass for the new pass manager that utilizes
the LoopDataPrefetch implementation class.
Reviewers: mehdi_amini
Subscribers: sanjoy, mzolotukhin, nemanjai, llvm-commits
Differential Revision: https://reviews.llvm.org/D23483
llvm-svn: 278591
`IVVisitor::visitCast` used to have the invariant that if the
instruction it was passed was a sext or zext instruction, the result of
the instruction would be wider than the induction variable. This is no
longer true after rL275037, so this change teaches `IndVarSimplify` s
implementation of `IVVisitor::visitCast` to work with the relaxed
invariant.
A corresponding change to SimplifyIndVar to preserve the said invariant
after rL275037 would also work, but given how `IVVisitor::visitCast` is
spelled (no indication of said invariant), I figured the current fix is
cleaner.
Fixes PR28935.
llvm-svn: 278584
InnerLoopVectorizer shouldn't handle a loop with cycles inside the loop
body, even if that cycle isn't a natural loop.
Fixes PR28541.
Differential Revision: https://reviews.llvm.org/D22952
llvm-svn: 278573
They aren't static, and moving them to the entry block across something
else will only result in tears.
Root cause of http://crbug.com/636558.
llvm-svn: 278571
Summary: The refined propagation algorithm is more accurate and robust.
Reviewers: davidxl, dnovillo
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D23224
llvm-svn: 278522
Rewrite Visited[Cond] = getValueFromConditionImpl(..., Visited) statement which can lead to a memory corruption since getValueFromConditionImpl changes Visited map and invalidates the iterators.
llvm-svn: 278514
Take range metadata into account for conditions like this:
%length = load i32, i32* %length_ptr, !range !{i32 0, i32 2147483647}
%cmp = icmp ult i32 %a, %length
This is a common pattern for range checks where the length of the array is dynamically loaded.
Reviewed By: sanjoy
Differential Revision: https://reviews.llvm.org/D23267
llvm-svn: 278496
Currently LVI can only gather value constraints from comparisons like:
* icmp <pred> Val, ...
* icmp ult (add Val, Offset), ...
In fact we can handle any predicate in latter comparisons.
Reviewed By: sanjoy
Differential Revision: https://reviews.llvm.org/D23357
llvm-svn: 278493
Summary:
1. Make coroutine representation more robust against optimization that may duplicate instruction by introducing coro.id intrinsics that returns a token that will get fed into coro.alloc and coro.begin. Due to coro.id returning a token, it won't get duplicated and can be used as reliable indicator of coroutine identify when a particular coroutine call gets inlined.
2. Move last three arguments of coro.begin into coro.id as they will be shared if coro.begin will get duplicated.
3. doc + test + code updated to support the new intrinsic.
Reviewers: mehdi_amini, majnemer
Subscribers: mehdi_amini, llvm-commits
Differential Revision: https://reviews.llvm.org/D23412
llvm-svn: 278481
Summary:
This patch adds IsVariadicFunction bit to summary in order
to not import variadic functions. Inliner doesn't inline
variadic functions because it is hard to reason about it.
This one small fix improves Importer by about 16%
(going from 86% to 100% of imported functions that are
inlined anywhere)
on some spec benchmarks like 'int' and others.
Reviewers: eraman, mehdi_amini, tejohnson
Subscribers: mehdi_amini, llvm-commits
Differential Revision: https://reviews.llvm.org/D23339
llvm-svn: 278432
When legal, extending trip count in the loop control logic generates better code compared to truncating IV. This is because
(1) extending trip count is a loop invariant operation (see genLoopLimit where we prove trip count is loop invariant).
(2) Scalar Evolution seems to have problems understanding trunc when computing loop trip count. So removing them allows better analysis performed in Scalar Evolution. (In particular this fixes PR 28363 which is the motivation for this change).
I am not going to perform any performance test. Any degradation caused by this should be an indication of a bug elsewhere.
To prove legality, we rely on SCEV to prove zext(trunc(IV)) == IV (or similarly for sext). If this holds, we can prove equivalence of trunc(IV)==ExitCnt (1) and IV == zext(ExitCnt). Simply take zext of boths sides of (1) and apply the proven equivalence.
This commit contains changes in a newly added testcase which was not included in the previous commit (which was reverted later on).
https://reviews.llvm.org/D23075
llvm-svn: 278421
Summary:
This is an extension of the fix in r271424. That fix dealt with builder
insert points being moved by SCEV expansion, but only for the lifetime
of the expand call. This change modifies the interface so that LSR can
safely call expand multiple times at the same insert point and do the
right thing if one of the expansions decides to move the original insert
point.
This is a fix for PR28719.
Reviewers: sanjoy
Subscribers: llvm-commits, mcrosier, mzolotukhin
Differential Revision: https://reviews.llvm.org/D23342
llvm-svn: 278413
Summary:
This fixes PR 28933 by making sure GVNHoist does not try to recreate memory
accesses when it has not actually moved them.
Reviewers: sebpop
Subscribers: llvm-commits, george.burgess.iv
Differential Revision: https://reviews.llvm.org/D23411
llvm-svn: 278401
Summary:
Keep track of all methods for which we have devirtualized at least
one call and then print them sorted alphabetically. That allows to
avoid duplicates and also makes the order deterministic.
Add optimization names into the remarks, so that it's easier to
understand how has each method been devirtualized.
Fix a bug when wrong methods could have been reported for
tryVirtualConstProp.
Reviewers: kcc, mehdi_amini
Differential Revision: https://reviews.llvm.org/D23297
llvm-svn: 278389
When legal, extending trip count in the loop control logic generates better code compared to truncating IV. This is because
(1) extending trip count is a loop invariant operation (see genLoopLimit where we prove trip count is loop invariant).
(2) Scalar Evolution seems to have problems understanding trunc when computing loop trip count. So removing them allows better analysis performed in Scalar Evolution. (In particular this fixes PR 28363 which is the motivation for this change).
I am not going to perform any performance test. Any degradation caused by this should be an indication of a bug elsewhere.
To prove legality, we rely on SCEV to prove zext(trunc(IV)) == IV (or similarly for sext). If this holds, we can prove equivalence of trunc(IV)==ExitCnt (1) and IV == zext(ExitCnt). Simply take zext of boths sides of (1) and apply the proven equivalence.
https://reviews.llvm.org/D23075
llvm-svn: 278334
Change --no-pgo-warn-missing to -pgo-warn-missing-function
and negate the default. /NFC
Add more test to make sure the warning is off by default
llvm-svn: 278314
Summary:
A particular coroutine usage pattern, where a coroutine is created, manipulated and
destroyed by the same calling function, is common for coroutines implementing
RAII idiom and is suitable for allocation elision optimization which avoid
dynamic allocation by storing the coroutine frame as a static `alloca` in its
caller.
coro.free and coro.alloc intrinsics are used to indicate which code needs to be suppressed
when dynamic allocation elision happens:
```
entry:
%elide = call i8* @llvm.coro.alloc()
%need.dyn.alloc = icmp ne i8* %elide, null
br i1 %need.dyn.alloc, label %coro.begin, label %dyn.alloc
dyn.alloc:
%alloc = call i8* @CustomAlloc(i32 4)
br label %coro.begin
coro.begin:
%phi = phi i8* [ %elide, %entry ], [ %alloc, %dyn.alloc ]
%hdl = call i8* @llvm.coro.begin(i8* %phi, i32 0, i8* null,
i8* bitcast ([2 x void (%f.frame*)*]* @f.resumers to i8*))
```
and
```
%mem = call i8* @llvm.coro.free(i8* %hdl)
%need.dyn.free = icmp ne i8* %mem, null
br i1 %need.dyn.free, label %dyn.free, label %if.end
dyn.free:
call void @CustomFree(i8* %mem)
br label %if.end
if.end:
...
```
If heap allocation elision is performed, we replace coro.alloc with a static alloca on the caller frame and coro.free with null constant.
Also, we need to make sure that if there are any tail calls referencing the coroutine frame, we need to remote tail call attribute, since now coroutine frame lives on the stack.
Documentation and overview is here: http://llvm.org/docs/Coroutines.html.
Upstreaming sequence (rough plan)
1.Add documentation. (https://reviews.llvm.org/D22603)
2.Add coroutine intrinsics. (https://reviews.llvm.org/D22659)
3.Add empty coroutine passes. (https://reviews.llvm.org/D22847)
4.Add coroutine devirtualization + tests.
ab) Lower coro.resume and coro.destroy (https://reviews.llvm.org/D22998)
c) Do devirtualization (https://reviews.llvm.org/D23229)
5.Add CGSCC restart trigger + tests. (https://reviews.llvm.org/D23234)
6.Add coroutine heap elision + tests. <= we are here
7.Add the rest of the logic (split into more patches)
Reviewers: mehdi_amini, majnemer
Subscribers: mehdi_amini, llvm-commits
Differential Revision: https://reviews.llvm.org/D23245
llvm-svn: 278242
Teach LVI how to gather information from conditions in the form of (cond1 && cond2). Our out-of-tree front-end emits range checks in this form.
Reviewed By: sanjoy
Differential Revision: http://reviews.llvm.org/D23200
llvm-svn: 278231
This is a resubmission of previously reverted r277592. It was hitting overly strong assertion in getConstantRange which was relaxed in r278217.
Use LVI to prove that adds do not wrap. The change is motivated by https://llvm.org/bugs/show_bug.cgi?id=28620 bug and it's the first step to fix that problem.
Reviewed By: sanjoy
Differential Revision: http://reviews.llvm.org/D23059
llvm-svn: 278220
Hal pointed out that the semantic of our intrinsic and the libc
call are slightly different. Add a comment while I'm here to
explain why we can't emit an intrinsic. Thanks Hal!
llvm-svn: 278200
Summary:
The inliner not being a function pass requires the work-around of
generating the OptimizationRemarkEmitter and in turn BFI on demand.
This will go away after the new PM is ready.
BFI is only computed inside ORE if the user has requested hotness
information for optimization diagnostitics (-pass-remark-with-hotness at
the 'opt' level). Thus there is no additional overhead without the
flag.
Reviewers: hfinkel, davidxl, eraman
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D22694
llvm-svn: 278185
Summary:
This hopefully fixes PR28825. The problem now was that a value from the
original loop was used in a subloop, which became a sibling after separation.
While a subloop doesn't need an lcssa phi node, a sibling does, and that's
where we broke LCSSA. The most natural way to fix this now is to simply call
formLCSSA on the original loop: it'll do what we've been doing before plus
it'll cover situations described above.
I think we don't need to run formLCSSARecursively here, and we have an assert
to verify this (I've tried testing it on LLVM testsuite + SPECs). I'd be happy
to be corrected here though.
I also changed a run line in the test from '-lcssa -loop-unroll' to
'-lcssa -loop-simplify -indvars', because it exercises LCSSA
preservation to the same extent, but also makes less unrelated
transformation on the CFG, which makes it easier to verify.
Reviewers: chandlerc, sanjoy, silvas
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D23288
llvm-svn: 278173
Summary:
We teach alias analysis that invariant.start is readonly.
This helps with GVN and memcopy optimizations that currently treat.
invariant.start as a clobber.
We need to treat this as readonly, so that DSE does not incorrectly
remove stores prior to the invariant.start
Reviewers: sanjoy, reames, majnemer, dberlin
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D23214
llvm-svn: 278138
no prof data for func warning is turned off by default
due to its high verbosity and minimal usefulness.
Differential Revision: http://reviews.llvm.org/D23295
llvm-svn: 278127
Summary:
In the use optimizer, we need to keep of whether the lower bound still
dominates us or else we may decide a lower bound is still valid when it
is not due to intervening pushes/pops. Fixes PR28880 (and probably a
bunch of other things).
Reviewers: george.burgess.iv
Subscribers: MatzeB, llvm-commits, sebpop
Differential Revision: https://reviews.llvm.org/D23237
llvm-svn: 277978
Summary:
The correctness fix here is that when we CSE a load with another load,
we need to combine the metadata on the two loads. This matches the
behavior of other passes, like instcombine and GVN.
There's also a minor optimization improvement here: for load PRE, the
aliasing metadata on the inserted load should be the same as the
metadata on the original load. Not sure why the old code was throwing
it away.
Issue found by inspection.
Differential Revision: http://reviews.llvm.org/D21460
llvm-svn: 277977
Summary:
CoroSplit pass processes the coroutine twice. First, it lets it go through
complete IPO optimization pipeline as a single function. It forces restart
of the pipeline by inserting an indirect call to an empty function "coro.devirt.trigger"
which is devirtualized by CoroElide pass that triggers a restart of the pipeline by CGPassManager.
(In later patches, when CoroSplit pass sees the same coroutine the second time, it splits it up,
adds coroutine subfunctions to the SCC to be processed by IPO pipeline.)
Documentation and overview is here: http://llvm.org/docs/Coroutines.html.
Upstreaming sequence (rough plan)
1.Add documentation. (https://reviews.llvm.org/D22603)
2.Add coroutine intrinsics. (https://reviews.llvm.org/D22659)
3.Add empty coroutine passes. (https://reviews.llvm.org/D22847)
4.Add coroutine devirtualization + tests.
ab) Lower coro.resume and coro.destroy (https://reviews.llvm.org/D22998)
c) Do devirtualization (https://reviews.llvm.org/D23229)
5.Add CGSCC restart trigger + tests. <= we are here
6.Add coroutine heap elision + tests.
7.Add the rest of the logic (split into more patches)
Reviewers: mehdi_amini, majnemer
Subscribers: llvm-commits, mehdi_amini
Differential Revision: https://reviews.llvm.org/D23234
llvm-svn: 277936
Summary:
This is the 4c patch of the coroutine series. CoroElide pass now checks if PostSplit coro.begin
is referenced by coro.subfn.addr intrinsics. If so replace coro.subfn.addrs with an appropriate coroutine
subfunction associated with that coro.begin.
Documentation and overview is here: http://llvm.org/docs/Coroutines.html.
Upstreaming sequence (rough plan)
1.Add documentation. (https://reviews.llvm.org/D22603)
2.Add coroutine intrinsics. (https://reviews.llvm.org/D22659)
3.Add empty coroutine passes. (https://reviews.llvm.org/D22847)
4.Add coroutine devirtualization + tests.
ab) Lower coro.resume and coro.destroy (https://reviews.llvm.org/D22998)
c) Do devirtualization <= we are here
5.Add CGSCC restart trigger + tests.
6.Add coroutine heap elision + tests.
7.Add the rest of the logic (split into more patches)
Reviewers: majnemer
Subscribers: mehdi_amini, llvm-commits
Differential Revision: https://reviews.llvm.org/D23229
llvm-svn: 277908
Fixes PR28764. Right now there is no way to test this, but (as
mentioned on the PR) with Michael Zolotukhin's yet to be checked in
LoopSimplify verfier, 8 of the llvm-lit tests for IRCE crash.
llvm-svn: 277891
This fixes PR28825. The problem was that we only checked if a value from
a created inner loop is used in the outer loop, and fixed LCSSA for
them. But we missed to fixup LCSSA for values used in exits of the outer
loop.
llvm-svn: 277877
Summary: Hot callsites should have higher threshold than inline hints. This patch uses separate threshold parameter for hot callsites.
Reviewers: davidxl, eraman
Subscribers: Prazek, llvm-commits
Differential Revision: https://reviews.llvm.org/D22368
llvm-svn: 277860
Summary:
Chrome on Linux uses WholeProgramDevirt for speed ups, and it's
important to detect regressions on both sides: the toolchain,
if fewer methods get devirtualized after an update, and Chrome,
if an innocently looking change caused many hot methods become
virtual again.
The need to track devirtualized methods is not Chrome-specific,
but it's probably the only user of the pass at this time.
Reviewers: kcc
Differential Revision: https://reviews.llvm.org/D23219
llvm-svn: 277856
Summary: We do not care about intrinsic calls when assigning discriminators.
Reviewers: davidxl, dnovillo
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D23212
llvm-svn: 277843
Summary:
Having -O0 in opt allows testing that -O0 optimization
pipeline is built correctly.
Reviewers: majnemer
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D23208
llvm-svn: 277829
This generated IR based on the order of evaluation, which is different
between GCC and Clang. With that in mind you get bootstrap miscompares
if you compare a Clang built with GCC-built Clang vs. Clang built with
Clang-built Clang. Diagnosing that made my head hurt.
This also reverts commit r277337, which "fixed" the test case.
llvm-svn: 277820
Summary:
Turn (select C, (sext A), B) into (sext (select C, A, B')) when A is i1 and
B is a compatible constant, also for zext instead of sext. This will then be
further folded into logical operations.
The transformation would be valid for non-i1 types as well, but other parts of
InstCombine prefer to have sext from non-i1 as an operand of select.
Motivated by the shader compiler frontend in Mesa for AMDGPU, which emits i32
for boolean operations. With this change, the boolean logic is fully
recovered.
Reviewers: majnemer, spatel, tstellarAMD
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D22747
llvm-svn: 277801
The patch splits a complex && if condition into easier to read and understand
logic. That wrong early exit condition was letting some instructions with not
all operands available pass through when HoistingGeps was true.
Differential Revision: https://reviews.llvm.org/D23174
llvm-svn: 277785
Shifts with a uniform but non-constant count were considered very expensive to
vectorize, because the splat of the uniform count and the shift would tend to
appear in different blocks. That made the splat invisible to ISel, and we'd
scalarize the shift at codegen time.
Since r201655, CodeGenPrepare sinks those splats to be next to their use, and we
are able to select the appropriate vector shifts. This updates the cost model to
to take this into account by making shifts by a uniform cheap again.
Differential Revision: https://reviews.llvm.org/D23049
llvm-svn: 277782
PR28848 had a very nice reduction of the underlying cause of the bug.
Our ValueMap had, in an entry for an Instruction, a ConstantInt.
This is not at all unexpected but should be handled properly.
llvm-svn: 277773
This is the forth patch in the coroutine series. CoroEaly pass now lowers coro.resume
and coro.destroy intrinsics by replacing them with an indirect call to an address
returned by coro.subfn.addr intrinsic. This is done so that CGPassManager recognizes
devirtualization when CoroElide replaces a call to coro.subfn.addr with an appropriate
function address.
Patch by Gor Nishanov!
Differential Revision: https://reviews.llvm.org/D22998
llvm-svn: 277765
I'm removing a misplaced pair of more specific folds from InstCombine in this patch as well,
so we know where those folds are happening in InstSimplify.
llvm-svn: 277738
Summary:
TargetBaseAlign is no longer required since LSV checks if target allows misaligned accesses.
A constant defining a base alignment is still needed for stack accesses where alignment can be adjusted.
Previous patch (D22936) was reverted because tests were failing. This patch also fixes the cause of those failures:
- x86 failing tests either did not have the right target, or the right alignment.
- NVPTX failing tests did not have the right alignment.
- AMDGPU failing test (merge-stores) should allow vectorization with the given alignment but the target info
considers <3xi32> a non-standard type and gives up early. This patch removes the condition and only checks
for a maximum size allowed and relies on the next condition checking for %4 for correctness.
This should be revisited to include 3xi32 as a MVT type (on arsenm's non-immediate todo list).
Note that checking the sizeInBits for a MVT is undefined (leads to an assertion failure),
so we need to create an EVT, hence the interface change in allowsMisaligned to include the Context.
Reviewers: arsenm, jlebar, tstellarAMD
Subscribers: jholewinski, arsenm, mzolotukhin, llvm-commits
Differential Revision: https://reviews.llvm.org/D23068
llvm-svn: 277735
With this patch we compute the MemorySSA once and update it in the code generator.
Differential Revision: https://reviews.llvm.org/D22966
llvm-svn: 277649
Some of these tests need to be cleaned up further to make it obvious
what they're testing, but as a first step remove all instances of
"grep".
llvm-svn: 277648
This reverts commit r277611 and the followup r277614.
Bootstrap builds and chromium builds are crashing during inlining after
this change.
llvm-svn: 277642
This is a follow-up to r277637. It teaches MemorySSA that invariant
loads (and loads of provably constant memory) are always liveOnEntry.
llvm-svn: 277640
This patch makes MemorySSA recognize atomic/volatile loads, and makes
MSSA treat said loads specially. This allows us to be a bit more
aggressive in some cases.
Administrative note: Revision was LGTM'ed by reames in person.
Additionally, this doesn't include the `invariant.load` recognition in
the differential revision, because I feel it's better to commit that
separately. Will commit soon.
Differential Revision: https://reviews.llvm.org/D16875
llvm-svn: 277637
Summary:
InstCombine unfolds expressions of the form `zext(or(icmp, icmp))` to `or(zext(icmp), zext(icmp))` such that in a later iteration of InstCombine the exposed `zext(icmp)` instructions can be optimized. We now combine this unfolding and the subsequent `zext(icmp)` optimization to be performed together. Since the unfolding doesn't happen separately anymore, we also again enable the folding of `logic(cast(icmp), cast(icmp))` expressions to `cast(logic(icmp, icmp))` which had been disabled due to its interference with the unfolding transformation.
Tested via `make check` and `lnt`.
Background
==========
For a better understanding on how it came to this change we subsequently summarize its history. In commit r275989 we've already tried to enable the folding of `logic(cast(icmp), cast(icmp))` to `cast(logic(icmp, icmp))` which had to be reverted in r276106 because it could lead to an endless loop in InstCombine (also see http://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20160718/374347.html). The root of this problem is that in `visitZExt()` in InstCombineCasts.cpp there also exists a reverse of the above folding transformation, that unfolds `zext(or(icmp, icmp))` to `or(zext(icmp), zext(icmp))` in order to expose `zext(icmp)` operations which would then possibly be eliminated by subsequent iterations of InstCombine. However, before these `zext(icmp)` would be eliminated the folding from r275989 could kick in and cause InstCombine to endlessly switch back and forth between the folding and the unfolding transformation. This is the reason why we now combine the `zext`-unfolding and the elimination of the exposed `zext(icmp)` to happen at one go because this enables us to still allow the cast-folding in `logic(cast(icmp), cast(icmp))` without entering an endless loop again.
Details on the submitted changes
================================
- In `visitZExt()` we combine the unfolding and optimization of `zext` instructions.
- In `transformZExtICmp()` we have to use `Builder->CreateIntCast()` instead of `CastInst::CreateIntegerCast()` to make sure that the new `CastInst` is inserted in a `BasicBlock`. The new calls to `transformZExtICmp()` that we introduce in `visitZExt()` would otherwise cause according assertions to be triggered (in our case this happend, for example, with lnt for the MultiSource/Applications/sqlite3 and SingleSource/Regression/C++/EH/recursive-throw tests). The subsequent usage of `replaceInstUsesWith()` is necessary to ensure that the new `CastInst` replaces the `ZExtInst` accordingly.
- In InstCombineAndOrXor.cpp we again allow the folding of casts on `icmp` instructions.
- The instruction order in the optimized IR for the zext-or-icmp.ll test case is different with the introduced changes.
- The test cases in zext.ll have been adopted from the reverted commits r275989 and r276105.
Reviewers: grosser, majnemer, spatel
Subscribers: eli.friedman, majnemer, llvm-commits
Differential Revision: https://reviews.llvm.org/D22864
Contributed-by: Matthias Reisinger <d412vv1n@gmail.com>
llvm-svn: 277635
This removes the restriction for the icmp constant, but as noted by the FIXME comments,
we still need to change individual checks for binop operand constants.
llvm-svn: 277629
This is a fix for PR28697.
An MDNode can indirectly refer to a GlobalValue, through a
ConstantAsMetadata. When the GlobalValue is deleted, the MDNode operand
is reset to `nullptr`. If the node is uniqued, this can lead to a
hard-to-detect cache invalidation in a Metadata map that's shared across
an LLVMContext.
Consider:
1. A map from Metadata* to `T` called RemappedMDs.
2. A node that references a global variable, `!{i1* @GV}`.
3. Insert `!{i1* @GV} -> SomeT` in the map.
4. Delete `@GV`, leaving behind `!{null} -> SomeT`.
Looking up the generic and uninteresting `!{null}` gives you `SomeT`,
which is likely related to `@GV`. Worse, `SomeT`'s lifetime may be tied
to the deleted `@GV`.
This occurs in practice in the shared ValueMap used since r266579 in the
IRMover. Other code that handles more than one Module (with different
lifetimes) in the same LLVMContext could hit it too.
The fix here is a partial revert of r225223: in the rare case that an
MDNode operand is a ConstantAsMetadata (i.e., wrapping a node from the
Value hierarchy), drop uniquing if it gets replaced with `nullptr`.
This changes step #4 above to leave behind `distinct !{null} -> SomeT`,
which can't be confused with the generic `!{null}`.
In theory, this can cause some churn in the LLVMContext's MDNode
uniquing map when Values are being deleted. However:
- The number of GlobalValues referenced from uniqued MDNodes is
expected to be quite small. E.g., the debug info metadata schema
only references GlobalValues from distinct nodes.
- Other Constants have the lifetime of the LLVMContext, whose teardown
is careful to drop references before deleting the constants.
As a result, I don't expect a compile time regression from this change.
llvm-svn: 277625
We were able to figure out that the result of a call is some constant.
While propagating that fact, we added the constant to the value map.
This is problematic because it results in us losing the call site when
processing the value map.
This fixes PR28802.
llvm-svn: 277611
reason about and less error prone.
The core idea is to fully parse the text without trying to identify
passes or structure. This is done with a single state machine. There
were various bugs in the logic around this previously that were repeated
and scattered across the code. Having a single routine makes it much
easier to fix and get correct. For example, this routine doesn't suffer
from PR28577.
Then the actual pass construction is handled using *much* easier to read
code and simple loops, with particular pass manager construction sunk to
live with other pass construction. This is especially nice as the pass
managers *are* in fact passes.
Finally, the "implicit" pass manager synthesis is done much more simply
by forming "pre-parsed" structures rather than having to duplicate tons
of logic.
One of the bugs fixed by this was evident in the tests where we accepted
a pipeline that wasn't really well formed. Another bug is PR28577 for
which I have added a test case.
The code is less efficient than the previous code but I'm really hoping
that's not a priority. ;]
Thanks to Sean for the review!
Differential Revision: https://reviews.llvm.org/D22724
llvm-svn: 277561
Summary:
Sometimes, bitsets could get really large (>300k entries) and
we might want to drop a check, as it would have a too much cost.
Adding a flag to control how much penalty are we willing to pay
for bitsets.
Reviewers: kcc
Differential Revision: https://reviews.llvm.org/D23088
llvm-svn: 277556
As agreed in post-commit review of r265388, I'm switching the flag to
its original value until the 90% runtime performance regression on
SingleSource/Benchmarks/Stanford/Bubblesort is addressed.
llvm-svn: 277524
Update comment for isOutOfScope and add a testcase for uniform value being used
out of scope.
Differential Revision: https://reviews.llvm.org/D23073
llvm-svn: 277515
This patch enables the vectorizer to generate both scalar and vector versions
of an integer induction variable for a given loop. Previously, we only
generated a scalar induction variable if we knew all its users were going to be
scalar. Otherwise, we generated a vector induction variable. In the case of a
loop with both scalar and vector users of the induction variable, we would
generate the vector induction variable and extract scalar values from it for
the scalar users. With this patch, we now generate both versions of the
induction variable when there are both scalar and vector users and select which
version to use based on whether the user is scalar or vector.
Differential Revision: https://reviews.llvm.org/D22869
llvm-svn: 277474
This patch refactors the logic in collectLoopUniforms and
collectValuesToIgnore, untangling the concepts of "uniform" and "scalar". It
adds isScalarAfterVectorization along side isUniformAfterVectorization to
distinguish the two. Known scalar values include those that are uniform,
getelementptr instructions that won't be vectorized, and induction variables
and induction variable update instructions whose users are all known to be
scalar.
This patch includes the following functional changes:
- In collectLoopUniforms, we mark uniform the pointer operands of interleaved
accesses. Although non-consecutive, these pointers are treated like
consecutive pointers during vectorization.
- In collectValuesToIgnore, we insert a value into VecValuesToIgnore if it
isScalarAfterVectorization rather than isUniformAfterVectorization. This
differs from the previous functionaly in that we now add getelementptr
instructions that will not be vectorized into VecValuesToIgnore.
This patch also removes the ValuesNotWidened set used for induction variable
scalarization since, after the above changes, it is now equivalent to
isScalarAfterVectorization.
Differential Revision: https://reviews.llvm.org/D22867
llvm-svn: 277460
Added ability to estimate the entry count of the extracted function and
the branch probabilities of the exit branches.
Patch by River Riddle!
Differential Revision: https://reviews.llvm.org/D22744
llvm-svn: 277411
Summary: This patch implements CFI for WebAssembly. It modifies the
LowerTypeTest pass to pre-assign table indexes to functions that are
called indirectly, and lowers type checks to test against the
appropriate table indexes. It also modifies the WebAssembly backend to
support a special ".indidx" assembly directive that propagates the table
index assignments out to the linker.
Patch by Dominic Chen
Differential Revision: https://reviews.llvm.org/D21768
llvm-svn: 277398
Using RAUW was wrong here; if we have a switch transform such as:
18 -> 6 then
6 -> 0
If we use RAUW, while performing the second transform the *transformed* 6
from the first will be also replaced, so we end up with:
18 -> 0
6 -> 0
Found by clang stage2 bootstrap; testcase added.
llvm-svn: 277332
It looks like the two independent parts of the rotate operation (a lshr and shl) are being reordered on some bots. Add CHECK-DAGs to account for this.
llvm-svn: 277329
If a switch is sparse and all the cases (once sorted) are in arithmetic progression, we can extract the common factor out of the switch and create a dense switch. For example:
switch (i) {
case 5: ...
case 9: ...
case 13: ...
case 17: ...
}
can become:
if ( (i - 5) % 4 ) goto default;
switch ((i - 5) / 4) {
case 0: ...
case 1: ...
case 2: ...
case 3: ...
}
or even better:
switch ( ROTR(i - 5, 2) {
case 0: ...
case 1: ...
case 2: ...
case 3: ...
}
The division and remainder operations could be costly so we only do this if the factor is a power of two, and emit a right-rotate instead of a divide/remainder sequence. Dense switches can be lowered significantly better than sparse switches and can even be transformed into lookup tables.
llvm-svn: 277325
When extracting a set of blocks make sure to inherit all of the target
dependent attributes to make sure that the function will be valid for
lowering. One example is the "target-features" attribute for x86, if the
extracted region has functionality that relies on a specific feature it
will fail to be lowered.
This also allows for extracted functions to be valid for inlining, at
least back into the parent function, as the target attributes are tested
when inlining for compatibility.
Patch by River Riddle!
Differential Revision: https://reviews.llvm.org/D22713
llvm-svn: 277315
LoopUnroll is a loop pass, so the analysis of OptimizationRemarkEmitter
is added to the common function analysis passes that loop passes
depend on.
The BFI and indirectly BPI used in this pass is computed lazily so no
overhead should be observed unless -pass-remarks-with-hotness is used.
This is how the patch affects the O3 pipeline:
Dominator Tree Construction
Natural Loop Information
Canonicalize natural loops
Loop-Closed SSA Form Pass
Basic Alias Analysis (stateless AA impl)
Function Alias Analysis Results
Scalar Evolution Analysis
+ Lazy Branch Probability Analysis
+ Lazy Block Frequency Analysis
+ Optimization Remark Emitter
Loop Pass Manager
Rotate Loops
Loop Invariant Code Motion
Unswitch loops
Simplify the CFG
Dominator Tree Construction
Basic Alias Analysis (stateless AA impl)
Function Alias Analysis Results
Combine redundant instructions
Natural Loop Information
Canonicalize natural loops
Loop-Closed SSA Form Pass
Scalar Evolution Analysis
+ Lazy Branch Probability Analysis
+ Lazy Block Frequency Analysis
+ Optimization Remark Emitter
Loop Pass Manager
Induction Variable Simplification
Recognize loop idioms
Delete dead loops
Unroll loops
...
llvm-svn: 277203
Patch by Sunita Marathe
Third try, now following fixes to MSan to handle mempcy in such a way that this commit won't break the MSan buildbots. (Thanks, Evegenii!)
llvm-svn: 277189
Some instructions may have their uses replaced with a symbolic constant.
However, the instruction may still have side effects which percludes it
from being removed from the function. EarlyCSE treated such an
instruction as if it were removed, resulting in PR28763.
llvm-svn: 277114
An undef vector element can be treated as if it had any value. Folding
such a vector element to 0 in a bitcast can open up further folding
opportunities.
llvm-svn: 277104
ConstantExpr::getWithOperands does much of the hard work that
ConstantFoldInstOperandsImpl tries to do but more completely.
This lets us fold ExtractValue/InsertValue expressions.
llvm-svn: 277100
A ConstantVector can have ConstantExpr operands and vice versa.
However, the folder had no ability to fold ConstantVectors which, in
some cases, was an optimization barrier.
Instead, rephrase the folder in terms of Constants instead of
ConstantExprs and teach callers how to deal with failure.
llvm-svn: 277099
Summary:
copypasta doc of ImportedFunctionsInliningStatistics class
\brief Calculate and dump ThinLTO specific inliner stats.
The main statistics are:
(1) Number of inlined imported functions,
(2) Number of imported functions inlined into importing module (indirect),
(3) Number of non imported functions inlined into importing module
(indirect).
The difference between first and the second is that first stat counts
all performed inlines on imported functions, but the second one only the
functions that have been eventually inlined to a function in the importing
module (by a chain of inlines). Because llvm uses bottom-up inliner, it is
possible to e.g. import function `A`, `B` and then inline `B` to `A`,
and after this `A` might be too big to be inlined into some other function
that calls it. It calculates this statistic by building graph, where
the nodes are functions, and edges are performed inlines and then by marking
the edges starting from not imported function.
If `Verbose` is set to true, then it also dumps statistics
per each inlined function, sorted by the greatest inlines count like
- number of performed inlines
- number of performed inlines to importing module
Reviewers: eraman, tejohnson, mehdi_amini
Subscribers: mehdi_amini, llvm-commits
Differential Revision: https://reviews.llvm.org/D22491
llvm-svn: 277089
Summary:
The motivation is the same as in D22141: In order to add the hotness
attribute to optimization remarks we need BFI to be available in all
passes that emit optimization remarks. BFI depends on BPI so unless we
make this lazy as well we would still compute BPI unconditionally.
The solution is to use the new LazyBPI pass in LazyBFI and only compute
BPI when computation of BFI is requested by the client.
I extended the laziness test using a LoopDistribute test to also cover
BPI.
Reviewers: hfinkel, davidxl
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D22835
llvm-svn: 277083
Summary:
Asan stack-use-after-scope check should poison alloca even if there is
no access between start and end.
This is possible for code like this:
for (int i = 0; i < 3; i++) {
int x;
p = &x;
}
"Loop Invariant Code Motion" will move "p = &x;" out of the loop, making
start/end range empty.
PR27453
Reviewers: eugenis
Differential Revision: https://reviews.llvm.org/D22842
llvm-svn: 277072
Summary:
Asan stack-use-after-scope check should poison alloca even if there is
no access between start and end.
This is possible for code like this:
for (int i = 0; i < 3; i++) {
int x;
p = &x;
}
"Loop Invariant Code Motion" will move "p = &x;" out of the loop, making
start/end range empty.
PR27453
Reviewers: eugenis
Differential Revision: https://reviews.llvm.org/D22842
llvm-svn: 277068
This adds boilerplate code for all coroutine passes,
the passes are no-ops for now.
Also, a small test has been added to verify that passes execute in
the expected order or not at all if coroutine support is disabled.
Patch by Gor Nishanov!
Differential Revision: https://reviews.llvm.org/D22847
llvm-svn: 277033
When folding an expression, we run ConstantFoldConstantExpression on
each operand of that expression.
However, ConstantFoldConstantExpression can fail and retur nullptr.
Previously, we would bail on further refining the expression.
Instead, use the original operand and see if we can refine a later
operand.
llvm-svn: 276959
Summary:
When we ask the builder to create a bitcast on a constant, we get back a
constant, not an instruction.
Reviewers: asbirlea
Subscribers: jholewinski, mzolotukhin, llvm-commits, arsenm
Differential Revision: https://reviews.llvm.org/D22878
llvm-svn: 276922
When loading or storing in a field of a struct like "a.b.c", GVN is able to
detect the equivalent expressions, and GVN-hoist would fail in the code
generation. This is because the GEPs are not hoisted as scalar operations to
avoid moving the GEPs too far from their ld/st instruction when the ld/st is not
movable. So we end up having to generate code for the GEP of a ld/st when we
move the ld/st. In the case of a GEP referring to another GEP as in "a.b.c" we
need to code generate all the GEPs necessary to make all the operands available
at the new location for the ld/st. With this patch we recursively walk through
the GEP operands checking whether all operands are available, and in the case of
a GEP operand, it recursively makes all its operands available. Code generation
happens from the inner GEPs out until reaching the GEP that appears as an
operand of the ld/st.
Differential Revision: https://reviews.llvm.org/D22599
llvm-svn: 276841
Instead of DFS numbering basic blocks we now DFS number instructions that avoids
the costly operation of which instruction comes first in a basic block.
Patch mostly written by Daniel Berlin.
Differential Revision: https://reviews.llvm.org/D22777
llvm-svn: 276714
Pre-instrumentation inline (pre-inliner) greatly improves the IR
instrumentation code performance, among other benefits. One issue of the
pre-inliner is it can introduce CFG-mismatch for COMDAT functions. This
is due to the fact that the same COMDAT function may have different early
inline decisions across different modules -- that means different copies
of COMDAT functions will have different CFG checksum.
In this patch, we propose a partially renaming the COMDAT group and its
member function/variable so we have different profile counter for each
version. We will post-fix the COMDAT function and the group name with its
FunctionHash.
Differential Revision: http://reviews.llvm.org/D22600
llvm-svn: 276673
The public InlineFunction utility assumes that the passed in
InlineFunctionInfo has a valid AssumptionCacheTracker.
Patch by River Riddle!
Differential Revision: https://reviews.llvm.org/D22706
llvm-svn: 276609
If we two loads of two different alignments, we must use the minimum of
the two alignments when hoisting. Same deal for stores.
For allocas, use the maximum of the two allocas.
llvm-svn: 276601
Allowed loop vectorization with secondary FP IVs. Like this:
float *A;
float x = init;
for (int i=0; i < N; ++i) {
A[i] = x;
x -= fp_inc;
}
The auto-vectorization is possible when the induction binary operator is "fast" or the function has "unsafe" attribute.
Differential Revision: https://reviews.llvm.org/D21330
llvm-svn: 276554
When vectorizing a tree rooted at a store bundle, we currently try to sort the
stores before building the tree, so that the stores can be vectorized. For other
trees, the order of the root bundle - which determines the order of all other
bundles - is arbitrary. That is bad, since if a leaf bundle of consecutive loads
happens to appear in the wrong order, we will not vectorize it.
This is partially mitigated when the root is a binary operator, by trying to
build a "reversed" tree when that's considered profitable. This patch extends the
workaround we have for binops to trees rooted in a horizontal reduction.
This fixes PR28474.
Differential Revision: https://reviews.llvm.org/D22554
llvm-svn: 276477
Recommiting r275571 after fixing crash reported in PR28270.
Now we erase elements of IOL in deleteDeadInstruction().
Original Summary:
This change use the overlap interval map built from partial overwrite tracking to perform shortening MemIntrinsics.
Add test cases which was missing opportunities before.
llvm-svn: 276452
Summary:
The llvm.invariant.start and llvm.invariant.end intrinsics currently
support specifying invariant memory objects only in the default address
space.
With this change, these intrinsics are overloaded for any adddress space
for memory objects
and we can use these llvm invariant intrinsics in non-default address
spaces.
Example: llvm.invariant.start.p1i8(i64 4, i8 addrspace(1)* %ptr)
This overloaded intrinsic is needed for representing final or invariant
memory in managed languages.
Reviewers: apilipenko, reames
Subscribers: llvm-commits
llvm-svn: 276447
Just because we can constant fold the result of an instruction does not
imply that we can delete the instruction. It may have side effects.
This fixes PR28655.
llvm-svn: 276389
If `-irce-skip-profitability-checks` is passed in, IRCE will kick in in
all cases where it is legal for it to kick in. This flag is intended to
help diagnose and analyse performance issues.
llvm-svn: 276372
Do not clone stored values unless they are GEPs that are special cased to avoid
hoisting them without hoisting their associated ld/st.
Differential revision: https://reviews.llvm.org/D22652
llvm-svn: 276358
rL245171 exposed a hole in InstSimplify that manifested in a strange way in PR28466:
https://llvm.org/bugs/show_bug.cgi?id=28466
It's possible to use trunc + icmp sgt/slt in place of an and + icmp eq/ne, so we need to
recognize that pattern to eliminate selects that are choosing between some value and some
bitmasked version of that value.
Note that there is significant room for improvement (refactoring) and enhancement (more
patterns, possibly in InstCombine rather than here).
Differential Revision: https://reviews.llvm.org/D22537
llvm-svn: 276341
This patch moves the update instruction for vectorized integer induction phi
nodes to the end of the latch block. This ensures consistent placement of all
induction updates across all the kinds of int inductions we create (scalar,
splat vector, or vector phi).
Differential Revision: https://reviews.llvm.org/D22416
llvm-svn: 276339
Summary:
The llvm.invariant.start and llvm.invariant.end intrinsics currently
support specifying invariant memory objects only in the default address space.
With this change, these intrinsics are overloaded for any adddress space for memory objects
and we can use these llvm invariant intrinsics in non-default address spaces.
Example: llvm.invariant.start.p1i8(i64 4, i8 addrspace(1)* %ptr)
This overloaded intrinsic is needed for representing final or invariant memory in managed languages.
Reviewers: tstellarAMD, reames, apilipenko
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D22519
llvm-svn: 276316
The earlier change added hotness attribute to missed-optimization
remarks. This follows up with the analysis remarks (the ones explaining
the reason for the missed optimization).
llvm-svn: 276192
We can replace the return values with undef if we replaced all
the call uses with a constant/undef.
Differential Revision: https://reviews.llvm.org/D22336
llvm-svn: 276174
Summary:
Previously we wouldn't move loads/stores across instructions that had
side-effects, where that was defined as may-write or may-throw. But
this is not sufficiently restrictive: Stores can't safely be moved
across instructions that may load.
This patch also adds a DEBUG check that all instructions in our chain
are either loads or stores.
Reviewers: asbirlea
Subscribers: llvm-commits, jholewinski, arsenm, mzolotukhin
Differential Revision: https://reviews.llvm.org/D22547
llvm-svn: 276171
Summary:
Previously if we had a chain that contained a side-effecting
instruction, we wouldn't vectorize it at all. Now we'll vectorize
everything that comes before the side-effecting instruction.
Reviewers: asbirlea
Subscribers: arsenm, jholewinski, llvm-commits, mzolotukhin
Differential Revision: https://reviews.llvm.org/D22536
llvm-svn: 276170
Summary: In r275989 we enabled the folding of `logic(cast(icmp), cast(icmp))` to `cast(logic(icmp, icmp))`. Here we add more test cases to assure this folding works for all logical operations `and`/`or`/`xor`.
Reviewers: grosser
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D22561
Contributed-by: Matthias Reisinger
llvm-svn: 276105
This patch adds costs for the vectorized implementations of CTPOP, the default values were seriously underestimating the cost of these and was encouraging vectorization on targets where serialized use of POPCNT would be much better.
Differential Revision: https://reviews.llvm.org/D22456
llvm-svn: 276104
Summary:
getVectorizablePrefix previously didn't work properly in the face of
aliasing loads/stores. It unwittingly assumed that the loads/stores
appeared in the BB in address order. If they didn't, it would do the
wrong thing.
Reviewers: asbirlea, tstellarAMD
Subscribers: arsenm, llvm-commits, mzolotukhin
Differential Revision: https://reviews.llvm.org/D22535
llvm-svn: 276072
Revert "[LoopSimplify] Update LCSSA after separating nested loops."
This reverts commit r275891.
Revert "[LCSSA] Post-process PHI-nodes created by SSAUpdate when constructing LCSSA form."
This reverts commit r275883.
llvm-svn: 276064
We just set PreserveLCSSA to always true since we don't have an
analogous method `mustPreserveAnalysisID(LCSSA)`.
Also port LoopInfo verifier pass to test LoopUnrollPass.
llvm-svn: 276063
Summary:
Previously, the insertion point for stores was the last instruction in
Chain *before calling getVectorizablePrefixEndIdx*. Thus if
getVectorizablePrefixEndIdx didn't return Chain.size(), we still would
insert at the last instruction in Chain.
This patch changes our internal API a bit in an attempt to make it less
prone to this sort of error. As a result, we end up recalculating the
Chain's boundary instructions, but I think worrying about the speed hit
of this is a premature optimization right now.
Reviewers: asbirlea, tstellarAMD
Subscribers: mzolotukhin, arsenm, llvm-commits
Differential Revision: https://reviews.llvm.org/D22534
llvm-svn: 276056
Summary:
This helps keep us honest -- there were a number of ways we could screw
up and still have passed this test.
Reviewers: asbirlea
Subscribers: llvm-commits, arsenm
Differential Revision: https://reviews.llvm.org/D22531
llvm-svn: 276053
The pattern may look more obviously like a sext if written as:
define i32 @g(i16 %x) {
%zext = zext i16 %x to i32
%xor = xor i32 %zext, 32768
%add = add i32 %xor, -32768
ret i32 %add
}
We already have that fold in visitAdd().
Differential Revision: https://reviews.llvm.org/D22477
llvm-svn: 276035
We skipped over ReturnInsts which didn't return an argument which would
lead us to incorrectly conclude that an argument returned by another
ReturnInst was 'returned'.
This reverts commit r275756.
This fixes PR28610.
llvm-svn: 276008
Summary:
Currently, InstCombine is already able to fold expressions of the form `logic(cast(A), cast(B))` to the simpler form `cast(logic(A, B))`, where logic designates one of `and`/`or`/`xor`. This transformation is implemented in `foldCastedBitwiseLogic()` in InstCombineAndOrXor.cpp. However, this optimization will not be performed if both `A` and `B` are `icmp` instructions. The decision to preclude casts of `icmp` instructions originates in r48715 in combination with r261707, and can be best understood by the title of the former one:
> Transform (zext (or (icmp), (icmp))) to (or (zext (cimp), (zext icmp))) if at least one of the (zext icmp) can be transformed to eliminate an icmp.
Apparently, it introduced a transformation that is a reverse of the transformation that is done in `foldCastedBitwiseLogic()`. Its purpose is to expose pairs of `zext icmp` that would subsequently be optimized by `transformZExtICmp()` in InstCombineCasts.cpp. Therefore, in order to avoid an endless loop of switching back and forth between these two transformations, the one in `foldCastedBitwiseLogic()` has been restricted to exclude `icmp` instructions which is mirrored in the responsible check:
`if ((!isa<ICmpInst>(Cast0Src) || !isa<ICmpInst>(Cast1Src)) && ...`
This check seems to sort out more cases than necessary because:
- the reverse transformation is obviously done for `or` instructions only
- and also not every `zext icmp` pair is necessarily the result of this reverse transformation
Therefore we now remove this check and replace it by a more finegrained one in `shouldOptimizeCast()` that now rejects only those `logic(zext(icmp), zext(icmp))` that would be able to be optimized by `transformZExtICmp()`, which also avoids the mentioned endless loop. That means we are now able to also simplify expressions of the form `logic(cast(icmp), cast(icmp))` to `cast(logic(icmp, icmp))` (`cast` being an arbitrary `CastInst`).
As an example, consider the following IR snippet
```
%1 = icmp sgt i64 %a, %b
%2 = zext i1 %1 to i8
%3 = icmp slt i64 %a, %c
%4 = zext i1 %3 to i8
%5 = and i8 %2, %4
```
which would now be transformed to
```
%1 = icmp sgt i64 %a, %b
%2 = icmp slt i64 %a, %c
%3 = and i1 %1, %2
%4 = zext i1 %3 to i8
```
This issue became apparent when experimenting with the programming language Julia, which makes use of LLVM. Currently, Julia lowers its `Bool` datatype to LLVM's `i8` (also see https://github.com/JuliaLang/julia/pull/17225). In fact, the above IR example is the lowered form of the Julia snippet `(a > b) & (a < c)`. Like shown above, this may introduce `zext` operations, casting between `i1` and `i8`, which could for example hinder ScalarEvolution and Polly on certain code.
Reviewers: grosser, vtjnash, majnemer
Subscribers: majnemer, llvm-commits
Differential Revision: https://reviews.llvm.org/D22511
Contributed-by: Matthias Reisinger
llvm-svn: 275989
D20859 and D20860 attempted to replace the SSE (V)CVTTPS2DQ and VCVTTPD2DQ truncating conversions with generic IR instead.
It turns out that the behaviour of these intrinsics is different enough from generic IR that this will cause problems, INF/NAN/out of range values are guaranteed to result in a 0x80000000 value - which plays havoc with constant folding which converts them to either zero or UNDEF. This is also an issue with the scalar implementations (which were already generic IR and what I was trying to match).
This patch changes both scalar and packed versions back to using x86-specific builtins.
It also deals with the other scalar conversion cases that are runtime rounding mode dependent and can have similar issues with constant folding.
A companion clang patch is at D22105
Differential Revision: https://reviews.llvm.org/D22106
llvm-svn: 275981
This patch updates MemorySSA's use-optimizing walker to be more
accurate and, in some cases, faster.
Essentially, this changed our core walking algorithm from a
cache-as-you-go DFS to an iteratively expanded DFS, with all of the
caching happening at the end. Said expansion happens when we hit a Phi,
P; we'll try to do the smallest amount of work possible to see if
optimizing above that Phi is legal in the first place. If so, we'll
expand the search to see if we can optimize to the next phi, etc.
An iteratively expanded DFS lets us potentially quit earlier (because we
don't assume that we can optimize above all phis) than our old walker.
Additionally, because we don't cache as we go, we can now optimize above
loops.
As an added bonus, this patch adds a ton of verification (if
EXPENSIVE_CHECKS are enabled), so finding bugs is easier.
Differential Revision: https://reviews.llvm.org/D21777
llvm-svn: 275940
For instructions in uniform set, they will not have vector versions so
add them to VecValuesToIgnore.
For induction vars, those only used in uniform instructions or consecutive
ptrs instructions have already been added to VecValuesToIgnore above. For
those induction vars which are only used in uniform instructions or
non-consecutive/non-gather scatter ptr instructions, the related phi and
update will also be added into VecValuesToIgnore set.
The change will make the vector RegUsages estimation less conservative.
Differential Revision: https://reviews.llvm.org/D20474
The recommit fixed the testcase global_alias.ll.
llvm-svn: 275936
For instructions in uniform set, they will not have vector versions so
add them to VecValuesToIgnore.
For induction vars, those only used in uniform instructions or consecutive
ptrs instructions have already been added to VecValuesToIgnore above. For
those induction vars which are only used in uniform instructions or
non-consecutive/non-gather scatter ptr instructions, the related phi and
update will also be added into VecValuesToIgnore set.
The change will make the vector RegUsages estimation less conservative.
Differential Revision: https://reviews.llvm.org/D20474
llvm-svn: 275912
Summary:
Usually LCSSA survives this transformation, but in some cases (see
attached test) it doesn't: values from the original loop after
separating might be used from the outer loop. Before the transformation
it was the same loop, so LCSSA phis were not required.
This fixes PR28272.
Reviewers: sanjoy, hfinkel, chandlerc
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D21665
llvm-svn: 275891
Summary:
The direct motivation for the port is to ensure that the OptRemarkEmitter
tests work with the new PM.
This remains a function pass because we not only create multiple loops
but could also version the original loop.
In the test I need to invoke opt
with -passes='require<aa>,loop-distribute'. LoopDistribute does not
directly depend on AA however LAA does. LAA uses getCachedResult so
I *think* we need manually pull in 'aa'.
Reviewers: davidxl, silvas
Subscribers: sanjoy, llvm-commits, mzolotukhin
Differential Revision: https://reviews.llvm.org/D22437
llvm-svn: 275811
While debugging GVNHoist, I found it confusing that the entries in a
VNtoInsns were not always value numbers. They _usually_ were except for
StoreInst in which case they were a hash of two different value numbers.
This leads to two observations:
- It is more difficult to debug things when the semantic contents of
VNtoInsns changes over time.
- Using a single value number is not much cheaper, the value of
VNtoInsns is a SmallVector.
- It is not immediately clear what the algorithm would do if there were
hash collisions in the StoreInst case.
Using a DenseMap of std::pair sidesteps all of this.
N.B. The changes in the test were due their sensitivity to the
iteration order of VNtoInsns which has changed.
llvm-svn: 275761
This reverts also r275029, "Update Clang tests after adding inference for the returned argument attribute"
It broke LTO build. Seems miscompilation.
llvm-svn: 275756
Summary:
To enable profile-guided indirect call promotion in ThinLTO mode, we
simply add call graph edges for each profitable target from the profile
to the summaries, then the summary-guided importing will consider the
callee for importing as usual.
Also we need to enable the indirect call promotion pass creation in the
PassManagerBuilder when PerformThinLTO=true (we are in the ThinLTO
backend), so that the newly imported functions are considered for
promotion in the backends.
The IC promotion profiles refer to callees by GUID, which required
adding GUIDs to the per-module VST in bitcode (and assigning them
valueIds similar to how they are assigned valueIds in the combined
index).
Reviewers: mehdi_amini, xur
Subscribers: mehdi_amini, davidxl, llvm-commits
Differential Revision: http://reviews.llvm.org/D21932
llvm-svn: 275707
This is a partial implementation of a general fold for associative+commutative operators:
(op (cast (op X, C2)), C1) --> (cast (op X, op (C1, C2)))
(op (cast (op X, C2)), C1) --> (op (cast X), op (C1, C2))
There are 7 associative operators and 13 cast types, so this could potentially go a lot further.
Differential Revision: https://reviews.llvm.org/D22421
llvm-svn: 275684
This reverts commit r275042; the initial commit triggered self-hosting failures
on ARM/AArch64. James Molloy identified the problematic backend code, which has
been disabled in r275677. Trying again...
Original commit message:
Let FuncAttrs infer the 'returned' argument attribute
A function can have one argument with the 'returned' attribute, indicating that
the associated argument is always the return value of the function. Add
FuncAttrs inference logic.
llvm-svn: 275678
With r275532 fixing miscompilation of GVN, "inbounds" on certain GEPs in these
tests cannot be preserved any more. Left a TODO in the tests for future
reference.
llvm-svn: 275596
This patch adds a selected set of cleanup passes including a pre-inline pass
before LLVM IR PGO instrumentation. The inline is only intended to apply those
obvious/trivial ones before instrumentation so that much less instrumentation
is needed to get better profiling information. This will drastically improve
the instrumented code performance for large C++ applications. Another benefit
is the context sensitive counts that can potentially improve the PGO
optimization.
Differential Revision: http://reviews.llvm.org/D21405
llvm-svn: 275588
Summary:
This is the first set of changes implementing the RFC from
http://thread.gmane.org/gmane.comp.compilers.llvm.devel/98334
This is a cross-sectional patch; rather than implementing the hotness
attribute for all optimization remarks and all passes in a patch set, it
implements it for the 'missed-optimization' remark for Loop
Distribution. My goal is to shake out the design issues before scaling
it up to other types and passes.
Hotness is computed as an integer as the multiplication of the block
frequency with the function entry count. It's only printed in opt
currently since clang prints the diagnostic fields directly. E.g.:
remark: /tmp/t.c:3:3: loop not distributed: use -Rpass-analysis=loop-distribute for more info (hotness: 300)
A new API added is similar to emitOptimizationRemarkMissed. The
difference is that it additionally takes a code region that the
diagnostic corresponds to. From this, hotness is computed using BFI.
The new API is exposed via an analysis pass so that it can be made
dependent on LazyBFI. (Thanks to Hal for the analysis pass idea.)
This feature can all be enabled by setDiagnosticHotnessRequested in the
LLVM context. If this is off, LazyBFI is not calculated (D22141) so
there should be no overhead.
A new command-line option is added to turn this on in opt.
My plan is to switch all user of emitOptimizationRemark* to use this
module instead.
Reviewers: hfinkel
Subscribers: rcox2, mzolotukhin, llvm-commits
Differential Revision: http://reviews.llvm.org/D21771
llvm-svn: 275583
Summary:
This change use the overlap interval map built from partial overwrite tracking to perform shortening MemIntrinsics.
Add test cases which was missing opportunities before.
Reviewers: hfinkel, eeckstein, mcrosier
Subscribers: mcrosier, llvm-commits
Differential Revision: https://reviews.llvm.org/D21909
llvm-svn: 275571
This pass hoists duplicated computations in the program. The primary goal of
gvn-hoist is to reduce the size of functions before inline heuristics to reduce
the total cost of function inlining.
Pass written by Sebastian Pop, Aditya Kumar, Xiaoyu Hu, and Brian Rzycki.
Important algorithmic contributions by Daniel Berlin under the form of reviews.
Differential Revision: http://reviews.llvm.org/D19338
llvm-svn: 275561
Fix for PR 28418.
opt never finishes compiling a test when -gvn option is passed.
The problem is caused by the fact that GVN fails to fold a constant expression.
Differential Revision: https://reviews.llvm.org/D22185
llvm-svn: 275483
This patch allows the formation of interleaved access groups in loops
containing predicated blocks. However, the predicated accesses are prevented
from forming groups.
Differential Revision: https://reviews.llvm.org/D19694
llvm-svn: 275471
Summary: Port Dead Loop Deletion Pass to the new pass manager.
Reviewers: silvas, davide
Subscribers: llvm-commits, sanjoy, mcrosier
Differential Revision: https://reviews.llvm.org/D21483
llvm-svn: 275453
This patch prevents increases in the number of instructions, pre-instcombine,
due to induction variable scalarization. An increase in instructions can lead
to an increase in the compile-time required to simplify the induction
variables. We now maintain a new map for scalarized induction variables to
prevent us from converting between the scalar and vector forms.
This patch should resolve compile-time regressions seen after r274627.
llvm-svn: 275419
enables the code size optimisation to fold a rem and div into a single
aeabi_uidivmod call. This was not happening before because sdiv was converted
but srem not, and instructions with different signedness are not combined.
Differential Revision: http://reviews.llvm.org/D22214
llvm-svn: 275403
This pass hoists duplicated computations in the program. The primary goal of
gvn-hoist is to reduce the size of functions before inline heuristics to reduce
the total cost of function inlining.
Pass written by Sebastian Pop, Aditya Kumar, Xiaoyu Hu, and Brian Rzycki.
Important algorithmic contributions by Daniel Berlin under the form of reviews.
Differential Revision: http://reviews.llvm.org/D19338
llvm-svn: 275401
constant hoisting. It not only takes into account the number of uses and the
cost of expressions in which constants appear, but now also the resulting
integer range of the offsets. Thus, the algorithm maximizes the number of uses
within an integer range that will enable more efficient code generation. On
ARM, for example, this will enable code size optimisations because less
negative offsets will be created. Negative offsets/immediates are not supported
by Thumb1 thus preventing more compact instruction encoding.
Differential Revision: http://reviews.llvm.org/D21183
llvm-svn: 275382
We were able to fold masked loads with an all-ones mask to a normal
load. However, we couldn't turn a masked load with a mask with mixed
ones and undefs into a normal load.
llvm-svn: 275380
This now should also work with the interprocedural variant of the pass.
Slightly easier now that the yak is shaved.
Differential Revision: http://reviews.llvm.org/D22329
llvm-svn: 275363
Summary:
In Scalarizer::gather we see if we already have a scattered form of Op,
and in that case use the new form.
In the particular case of PR28108, the found ValueVector SV has size 2,
where the first Value is nullptr, and the second is indeed a proper Value.
The nullptr then caused an assert to blow when we tried to do
cast<Instruction>(SV[I]).
With this patch we check SV[I] before doing the cast, and if it's nullptr
we just skip over it.
I don't know the Scalarizer well enough to know if this is the best fix
or if something should be done else where to prevent the nullptr from
being in the ValueVector at all, but at least this avoids the crash
and looking at the test case output it looks reasonable.
Reviewers: hfinkel, frasercrmck, wala, mehdi_amini
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D21518
llvm-svn: 275359
Treat loads which clip before the start of a global initializer the same
way we treat clipping beyond the end of the initializer: use zeros.
llvm-svn: 275345
Summary:
LSV used to abort vectorizing a chain for interleaved load/store accesses that alias.
Allow a valid prefix of the chain to be vectorized, mark just the prefix and retry vectorizing the remaining chain.
Reviewers: llvm-commits, jlebar, arsenm
Subscribers: mzolotukhin
Differential Revision: http://reviews.llvm.org/D22119
llvm-svn: 275317
In D21740, we discussed trying to make this a more general matcher. However, I didn't see a clean
way to handle the regular m_Not cases and these non-splat vector patterns, so I've opted for the
direct approach here. If there are other potential uses of areInverseVectorBitmasks(), we could
move that helper function to a higher level.
There is an open question as to which is of these forms should be considered the canonical IR:
%sel = select <4 x i1> <i1 true, i1 false, i1 false, i1 true>, <4 x i32> %a, <4 x i32> %b
%shuf = shufflevector <4 x i32> %a, <4 x i32> %b, <4 x i32> <i32 0, i32 5, i32 6, i32 3>
Differential Revision: http://reviews.llvm.org/D22114
llvm-svn: 275289
The LCSSA pass itself will not generate several redundant PHI nodes in a single
exit block. However, such redundant PHI nodes don't violate LCSSA form, and may
be introduced by passes that preserve LCSSA, and/or preserved by the LCSSA pass
itself. So, assuming a single PHI node per exit block is not safe.
llvm-svn: 275217
Summary:
It's useful to have some visibility about which call sites are devirtualized,
especially for debug purposes. Another use case is a regression test on the
application side (like, Chromium).
Reviewers: pcc
Differential Revision: http://reviews.llvm.org/D22252
llvm-svn: 275145
Summary:
Aiming to correct the ordering of loads/stores. This patch changes the
insert point for loads to the position of the first load.
It updates the ordering method for loads to insert before, rather than after.
Before this patch the following sequence:
"load a[1], store a[1], store a[0], load a[2]"
Would incorrectly vectorize to "store a[0,1], load a[1,2]".
The correctness check was assuming the insertion point for loads is at
the position of the first load, when in practice it was at the last
load. An alternative fix would have been to invert the correctness check.
The current fix changes insert position but also requires reordering of
instructions before the vectorized load.
Updated testcases to reflect the changes.
Reviewers: tstellarAMD, llvm-commits, jlebar, arsenm
Subscribers: mzolotukhin
Differential Revision: http://reviews.llvm.org/D22071
llvm-svn: 275117
Make some AVX and AVX512 cast costs more precise.
Based on part of a patch by Elena Demikhovsky (D15604).
Differential Revision: http://reviews.llvm.org/D22064
llvm-svn: 275106
There's a little bit of churn in this patch because the initialization
mechanism is now shared between the old and the new PM. Other than
that, it's just a pretty mechanical translation.
llvm-svn: 275082
Summary:
For sample-based PGO, using BFI to calculate callsite count is sometime not accurate. This is because with sampling based approach, if a callsite resides in a hot loop deeply nested in a bunch of cold branches, the callsite's BFI frequency would be inaccurately calculated due to lack of samples in the cold branch.
E.g.
if (A1 && A2 && A3 && ..... && A10) {
for (i=0; i < 100000000; i++) {
callsite();
}
}
Assume that A1 to A100 are all 100% taken, and callsite has 1000 samples and thus is considerred hot. Because the loop's trip count is huge, it's normal that all branches outside the loop has no sample at all. As a result, we can only use static branch probability to derive the the frequency of the loop header. Assuming that static heuristic thinks each branch is 50% taken, then the count calculated from BFI will be 1/(2^10) of the actual value.
In order to get more accurate callsite count, we directly annotate the weight on the call instruction, and directly use it when checking callsite hotness.
Note that this mechanism can also be shared by instrumentation based callsite hotness analysis. The side benefit is that it breaks the dependency from Inliner to BFI as call count is embedded in the IR.
Reviewers: davidxl, eraman, dnovillo
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D22118
llvm-svn: 275073
Summary: Handle the case when there is only one incoming/outgoing edge for a visited basic block: use the block weight to adjust edge weight even when the edge has been visited before. This can help reduce inaccuracies introduced by incorrect basic block profile, as shown in the updated unittest.
Reviewers: davidxl, dnovillo
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D22180
llvm-svn: 275072
For functions which are known to return a specific argument, pointer-comparison
folding can look through the function calls as part of its analysis.
Differential Revision: http://reviews.llvm.org/D9387
llvm-svn: 275039
If a function is known to return one of its arguments, we can use that in order
to compute known bits of the return value.
Differential Revision: http://reviews.llvm.org/D9397
llvm-svn: 275036
A function can have one argument with the 'returned' attribute, indicating that
the associated argument is always the return value of the function. Add
FuncAttrs inference logic.
Differential Revision: http://reviews.llvm.org/D22202
llvm-svn: 275027
ConstantInt::getSExtValue may fail on >64-bit integers. Add checks to call
getSExtValue only on narrow integers.
As a minor aside, simplify slsr-gep.ll to remove unnecessary load instructions.
llvm-svn: 274982
While here move simplifyLoop() function to the new header, as
suggested by Chandler in the review.
Differential Revision: http://reviews.llvm.org/D21404
llvm-svn: 274959
Summary:
This way the metadata will be only generated when asserts enabled,
or when -enable-import-metadata specified
FIXED missing colon on requires.
Reviewers: tejohnson, eraman, mehdi_amini
Subscribers: mehdi_amini, llvm-commits
Differential Revision: http://reviews.llvm.org/D22167
llvm-svn: 274947
Xin Tong