Summary:
As discussed on llvm-dev[1].
This change adds the basic boilerplate code around having this intrinsic
in LLVM:
- Changes in Intrinsics.td, and the IR Verifier
- A lowering pass to lower @llvm.experimental.guard to normal
control flow
- Inliner support
[1]: http://lists.llvm.org/pipermail/llvm-dev/2016-February/095523.html
Reviewers: reames, atrick, chandlerc, rnk, JosephTremoulet, echristo
Subscribers: mcrosier, llvm-commits
Differential Revision: http://reviews.llvm.org/D18527
llvm-svn: 264976
Commit r260791 contained an error in that it would introduce a cross-module
reference in the old module. It also introduced O(N^2) complexity in the
module cloner by requiring the entire module to be visited for each function.
Fix both of these problems by avoiding use of the CloneDebugInfoMetadata
function (which is only designed to do intra-module cloning) and cloning
function-attached metadata in the same way that we clone all other metadata.
Differential Revision: http://reviews.llvm.org/D18583
llvm-svn: 264935
Widening a PHI requires us to insert a trunc.
The logical place for this trunc is in the same BB as the PHI.
This is not possible if the BB is terminated by a catchswitch.
This fixes PR27133.
llvm-svn: 264926
Summary:
This gives callers flexibility to pass lambdas with captures, which lets
callers avoid the C-style void*-ptr closure style. (Currently, callers
in clang store state in the PassManagerBuilderBase arg.)
No functional change, and the new API is backwards-compatible.
Reviewers: chandlerc
Subscribers: joker.eph, cfe-commits
Differential Revision: http://reviews.llvm.org/D18613
llvm-svn: 264918
This change prevents the loop vectorizer from vectorizing when all of the vector
types it generates will be scalarized. I've run into this problem on the PPC's QPX
vector ISA, which only holds floating-point vector types. The loop vectorizer
will, however, happily vectorize loops with purely integer computation. Here's
an example:
LV: The Smallest and Widest types: 32 / 32 bits.
LV: The Widest register is: 256 bits.
LV: Found an estimated cost of 0 for VF 1 For instruction: %indvars.iv25 = phi i64 [ 0, %entry ], [ %indvars.iv.next26, %for.body ]
LV: Found an estimated cost of 0 for VF 1 For instruction: %arrayidx = getelementptr inbounds [1600 x i32], [1600 x i32]* %a, i64 0, i64 %indvars.iv25
LV: Found an estimated cost of 0 for VF 1 For instruction: %2 = trunc i64 %indvars.iv25 to i32
LV: Found an estimated cost of 1 for VF 1 For instruction: store i32 %2, i32* %arrayidx, align 4
LV: Found an estimated cost of 1 for VF 1 For instruction: %indvars.iv.next26 = add nuw nsw i64 %indvars.iv25, 1
LV: Found an estimated cost of 1 for VF 1 For instruction: %exitcond27 = icmp eq i64 %indvars.iv.next26, 1600
LV: Found an estimated cost of 0 for VF 1 For instruction: br i1 %exitcond27, label %for.cond.cleanup, label %for.body
LV: Scalar loop costs: 3.
LV: Found an estimated cost of 0 for VF 2 For instruction: %indvars.iv25 = phi i64 [ 0, %entry ], [ %indvars.iv.next26, %for.body ]
LV: Found an estimated cost of 0 for VF 2 For instruction: %arrayidx = getelementptr inbounds [1600 x i32], [1600 x i32]* %a, i64 0, i64 %indvars.iv25
LV: Found an estimated cost of 0 for VF 2 For instruction: %2 = trunc i64 %indvars.iv25 to i32
LV: Found an estimated cost of 2 for VF 2 For instruction: store i32 %2, i32* %arrayidx, align 4
LV: Found an estimated cost of 1 for VF 2 For instruction: %indvars.iv.next26 = add nuw nsw i64 %indvars.iv25, 1
LV: Found an estimated cost of 1 for VF 2 For instruction: %exitcond27 = icmp eq i64 %indvars.iv.next26, 1600
LV: Found an estimated cost of 0 for VF 2 For instruction: br i1 %exitcond27, label %for.cond.cleanup, label %for.body
LV: Vector loop of width 2 costs: 2.
LV: Found an estimated cost of 0 for VF 4 For instruction: %indvars.iv25 = phi i64 [ 0, %entry ], [ %indvars.iv.next26, %for.body ]
LV: Found an estimated cost of 0 for VF 4 For instruction: %arrayidx = getelementptr inbounds [1600 x i32], [1600 x i32]* %a, i64 0, i64 %indvars.iv25
LV: Found an estimated cost of 0 for VF 4 For instruction: %2 = trunc i64 %indvars.iv25 to i32
LV: Found an estimated cost of 4 for VF 4 For instruction: store i32 %2, i32* %arrayidx, align 4
LV: Found an estimated cost of 1 for VF 4 For instruction: %indvars.iv.next26 = add nuw nsw i64 %indvars.iv25, 1
LV: Found an estimated cost of 1 for VF 4 For instruction: %exitcond27 = icmp eq i64 %indvars.iv.next26, 1600
LV: Found an estimated cost of 0 for VF 4 For instruction: br i1 %exitcond27, label %for.cond.cleanup, label %for.body
LV: Vector loop of width 4 costs: 1.
...
LV: Selecting VF: 8.
LV: The target has 32 registers
LV(REG): Calculating max register usage:
LV(REG): At #0 Interval # 0
LV(REG): At #1 Interval # 1
LV(REG): At #2 Interval # 2
LV(REG): At #4 Interval # 1
LV(REG): At #5 Interval # 1
LV(REG): VF = 8
The problem is that the cost model here is not wrong, exactly. Since all of
these operations are scalarized, their cost (aside from the uniform ones) are
indeed VF*(scalar cost), just as the model suggests. In fact, the larger the VF
picked, the lower the relative overhead from the loop itself (and the
induction-variable update and check), and so in a sense, picking the largest VF
here is the right thing to do.
The problem is that vectorizing like this, where all of the vectors will be
scalarized in the backend, isn't really vectorizing, but rather interleaving.
By itself, this would be okay, but then the vectorizer itself also interleaves,
and that's where the problem manifests itself. There's aren't actually enough
scalar registers to support the normal interleave factor multiplied by a factor
of VF (8 in this example). In other words, the problem with this is that our
register-pressure heuristic does not account for scalarization.
While we might want to improve our register-pressure heuristic, I don't think
this is the right motivating case for that work. Here we have a more-basic
problem: The job of the vectorizer is to vectorize things (interleaving aside),
and if the IR it generates won't generate any actual vector code, then
something is wrong. Thus, if every type looks like it will be scalarized (i.e.
will be split into VF or more parts), then don't consider that VF.
This is not a problem specific to PPC/QPX, however. The problem comes up under
SSE on x86 too, and as such, this change fixes PR26837 too. I've added Sanjay's
reduced test case from PR26837 to this commit.
Differential Revision: http://reviews.llvm.org/D18537
llvm-svn: 264904
PGOFuncNames are used as the key to retrieve the Function definition from the
MD5 stored in the profile. For internal linkage function, we prefix the source
file name to the PGOFuncNames. LTO's internalization privatizes many global linkage
symbols. This happens after value profile annotation, but those internal
linkage functions should not have a source prefix. To differentiate compiler
generated internal symbols from original ones, PGOFuncName meta data are
created and attached to the original internal symbols in the value profile
annotation step. If a symbol does not have the meta data, its original linkage
must be non-internal.
Also add a new map that maps PGOFuncName's MD5 value to the function definition.
Differential Revision: http://reviews.llvm.org/D17895
llvm-svn: 264902
These checks are redundant and can be removed
Reviewers: hans
Subscribers: llvm-commits, mzolotukhin
Differential Revision: http://reviews.llvm.org/D18564
llvm-svn: 264872
Prior to this patch, the MemorySSA caching visitor would cache all
calls that it visited. When paired with phi optimization, this can be
problematic. Consider:
define void @foo() {
; 1 = MemoryDef(liveOnEntry)
call void @clobberFunction()
br i1 undef, label %if.end, label %if.then
if.then:
; MemoryUse(??)
call void @readOnlyFunction()
; 2 = MemoryDef(1)
call void @clobberFunction()
br label %if.end
if.end:
; 3 = MemoryPhi(...)
; MemoryUse(?)
call void @readOnlyFunction()
ret void
}
When optimizing MemoryUse(?), we visit defs 1 and 2, so we note to
cache them later. We ultimately end up not being able to optimize
passed the Phi, so we set MemoryUse(?) to point to the Phi. We then
cache the clobbering call for def 1 to be the Phi.
This commit changes this behavior so that we wipe out any calls
added to VisistedCalls while visiting the defs of a phi we couldn't
optimize.
Aside: With this patch, we now can bootstrap clang/LLVM without a
single MemorySSA verifier failure. Woohoo. :)
llvm-svn: 264820
This patch teaches the caching MemorySSA walker a few things:
1. Not to walk Phis we've walked before. It seems that we tried to do
this before, but it didn't work so well in cases like:
define void @foo() {
%1 = alloca i8
%2 = alloca i8
br label %begin
begin:
; 3 = MemoryPhi({%0,liveOnEntry},{%end,2})
; 1 = MemoryDef(3)
store i8 0, i8* %2
br label %end
end:
; MemoryUse(?)
load i8, i8* %1
; 2 = MemoryDef(1)
store i8 0, i8* %2
br label %begin
}
Because we wouldn't put Phis in Q.Visited until we tried to visit them.
So, when trying to optimize MemoryUse(?):
- We would visit 3 above
- ...Which would make us put {%0,liveOnEntry} in Q.Visited
- ...Which would make us visit {%0,liveOnEntry}
- ...Which would make us put {%end,2} in Q.Visited
- ...Which would make us visit {%end,2}
- ...Which would make us visit 3
- ...Which would realize we've already visited everything in 3
- ...Which would make us conservatively return 3.
In the added test-case, (@looped_visitedonlyonce) this behavior would
cause us to give incorrect results. Specifically, we'd visit 4 twice
in the same query, but on the second visit, we'd skip while.cond because
it had been visited, visit if.then/if.then2, and cache "1" as the
clobbering def on the way back.
2. If we try to walk the defs of a {Phi,MemLoc} and see it has been
visited before, just hand back the Phi we're trying to optimize.
I promise this isn't as terrible as it seems. :)
We now insert {Phi,MemLoc} pairs just before walking the Phi's upward
defs. So, we check the cache for the {Phi,MemLoc} pair before checking
if we've already walked the Phi.
The {Phi,MemLoc} pair is (almost?) always guaranteed to have a cache
entry if we've already fully walked it, because we cache as we go.
So, if the {Phi,MemLoc} pair isn't in cache, either:
(a) we must be in the process of visiting it (in which case, we can't
give a better answer in a cache-as-we-go DFS walker)
(b) we visited it, but didn't cache it on the way back (...which seems
to require `ModifyingAccess` to not dominate `StartingAccess`,
so I'm 99% sure that would be an error. If it's not an error, I
haven't been able to get it to happen locally, so I suspect it's
rare.)
- - - - -
As a consequence of this change, we no longer skip upward defs of phis,
so we can kill the `VisitedOnlyOne` check. This gives us better accuracy
than we had before, at the cost of potentially doing a bit more work
when we have a loop.
llvm-svn: 264814
This is effectively NFC, minus the renaming of the options
(-cyclone-prefetch-distance -> -prefetch-distance).
The change was requested by Tim in D17943.
llvm-svn: 264806
We have known races on profile counters, which can be reproduced by enabling
-fsanitize=thread and -fprofile-instr-generate simultaneously on a
multi-threaded program. This patch avoids reporting those races by not
instrumenting the reads and writes coming from the instruction profiler.
llvm-svn: 264805
During ADCE, track which debug info scopes still have live references
from the code, and delete debug info intrinsics for the dead ones.
These intrinsics describe the locations of variables (in registers or
stack slots). If there's no code left corresponding to a variable's
scope, then there's no way to reference the variable in the debugger and
it doesn't matter what its value is.
I add a DEBUG printout when the described location in an SSA register,
in case it helps some trying to track down why locations get lost.
However, we still delete these; the scope itself isn't attached to any
real code, so the ship has already sailed.
llvm-svn: 264800
Since we have moved to a model where functions are imported in bulk from
each source module after making summary-based importing decisions, there
is no longer a need to link metadata as a postpass, and all users have
been removed.
This essentially reverts r255909 and follow-on fixes.
llvm-svn: 264763
When eliminating or merging almost empty basic blocks, the existence of non-trivial PHI nodes
is currently used to recognize potential loops of which the block is the header and keep the block.
However, the current algorithm fails if the loops' exit condition is evaluated only with volatile
values hence no PHI nodes in the header. Especially when such a loop is an outer loop of a nested
loop, the loop is collapsed into a single loop which prevent later optimizations from being
applied (e.g., transforming nested loops into simplified forms and loop vectorization).
The patch augments the existing PHI node-based check by adding a pre-test if the BB actually
belongs to a set of loop headers and not eliminating it if yes.
llvm-svn: 264697
Personality is copied as part of copyFunctionAttributes, but it is
invalid on a declaration. Remove the personality attribute it the
function body is not cloned.
Also add a verifier run over output modules in the llvm-split tool.
llvm-svn: 264667
On OS X El Capitan and iOS 9, the linker supports a new section
attribute, live_support, which allows dead stripping to remove dead
globals along with the ASAN metadata about them.
With this change __asan_global structures are emitted in a new
__DATA,__asan_globals section on Darwin.
Additionally, there is a __DATA,__asan_liveness section with the
live_support attribute. Each entry in this section is simply a tuple
that binds together the liveness of a global variable and its ASAN
metadata structure. Thus the metadata structure will be alive if and
only if the global it references is also alive.
Review: http://reviews.llvm.org/D16737
llvm-svn: 264645
When eliminating or merging almost empty basic blocks, the existence of non-trivial PHI nodes
is currently used to recognize potential loops of which the block is the header and keep the block.
However, the current algorithm fails if the loops' exit condition is evaluated only with volatile
values hence no PHI nodes in the header. Especially when such a loop is an outer loop of a nested
loop, the loop is collapsed into a single loop which prevent later optimizations from being
applied (e.g., transforming nested loops into simplified forms and loop vectorization).
The patch augments the existing PHI node-based check by adding a pre-test if the BB actually
belongs to a set of loop headers and not eliminating it if yes.
llvm-svn: 264596
Don't set the function hotness attribute on the fly. This changes the CFG
branch probability of the caller function, which leads to inconsistent BB
ordering. This patch moves the attribute setting to a separated loop after
the counts in all functions are populated.
Fixes PR27024 - PGO instrumentation profile data is not reflected in correct
basic blocks.
Differential Revision: http://reviews.llvm.org/D18491
llvm-svn: 264594
Summary:
Add a statistic to count the number of imported functions. Also, add a
new -print-imports option to emit a trace of imported functions, that
works even for an NDEBUG build.
Note that emitOptimizationRemark does not work for the above printing as
it expects a Function object and DebugLoc, neither of which we have
with summary-based importing.
This is part 2 of D18487, the first part was committed separately as
r264536.
Reviewers: joker.eph
Subscribers: llvm-commits, joker.eph
Differential Revision: http://reviews.llvm.org/D18487
llvm-svn: 264537
With r264503, aliases are now being added to the GlobalsToImport set
even when their aliasees can't be imported due to their linkage type.
While the importing worked correctly (the aliases imported as
declarations) due to the logic in doImportAsDefinition, there is no
point to adding them to the GlobalsToImport set.
Additionally, with D18487 it was resulting in incorrectly printing a
message indicating that the alias was imported.
To avoid this, delay adding aliases to the GlobalsToImport set until
after the linkage type of the aliasee is checked.
This patch is part of D18487.
llvm-svn: 264536
Summary:
Now that the summary contains the full reference/call graph, we can
replace the existing function importer that loads and inspect the IR
to iteratively walk the call graph by a traversal based purely on the
summary information. Decouple the actual importing decision from any
IR manipulation.
Reviewers: tejohnson
Subscribers: llvm-commits, joker.eph
Differential Revision: http://reviews.llvm.org/D18343
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 264503
This changes RS4GC to lower calls to ``@llvm.experimental.deoptimize``
to gc.statepoints wrapping ``__llvm_deoptimize``, and changes
``callsGCLeafFunction`` to recognize ``@llvm.experimental.deoptimize``
as a non GC leaf function.
I've had to hard code the ``"__llvm_deoptimize"`` name in
RewriteStatepointsForGC; since ``TargetLibraryInfo`` is available only
during codegen. This isn't without precedent in the codebase, so I'm
not overtly concerned.
llvm-svn: 264456
We try to hoist the insertion point as high as possible to encourage
sharing. However, we must be careful not to hoist into a catchswitch as
it is both an EHPad and a terminator.
llvm-svn: 264344
isDependenceDistanceOfOne asserts that the store and the load access
through the same type. This function is also used by
removeDependencesFromMultipleStores so we need to make sure we filter
out mismatching types before reaching this point.
Now we do this when the initial candidates are gathered.
This is a refinement of the fix made in r262267.
Fixes PR27048.
llvm-svn: 264313
There are a few bugs in the walker that this patch addresses.
Primarily:
- Caching can break when we have multiple BBs without phis
- We weren't optimizing some phis properly
- Because of how the DFS iterator works, there were times where we
wouldn't cache any results of our DFS
I left the test cases with FIXMEs in, because I'm not sure how much
effort it will take to get those to work (read: We'll probably
ultimately have to end up redoing the walker, or we'll have to come up
with some creative caching tricks), and more test coverage = better.
Differential Revision: http://reviews.llvm.org/D18065
llvm-svn: 264180
When you have multiple LCSSA (single-operand) PHIs that are converted
into two-operand PHIs due to versioning, only assert that the PHI
currently being converted has a single operand. I.e. we don't want to
check PHIs that were converted earlier in the loop.
Fixes PR27023.
Thanks to Karl-Johan Karlsson for the minimized testcase!
llvm-svn: 264081
It's a bug fix.
For rerolled loops SE trip count remains unchanged. It leads to incorrect work of the next passes.
My patch just resets SE info for rerolled loop forcing SE to re-evaluate it next time it requested.
I also added a verifier call in the exisitng test to be sure no invalid SE data remain. Without my fix this test would fail with -verify-scev.
Differential Revision: http://reviews.llvm.org/D18316
llvm-svn: 264051
Summary:
Without tree pruning clang has 2,667,552 points.
Wiht only dominators pruning: 1,515,586.
With both dominators & predominators pruning: 1,340,534.
Resubmit of r262103.
Differential Revision: http://reviews.llvm.org/D18341
llvm-svn: 264003
If we have a BB with only MemoryDefs, live-in calculations will ignore
it. This means we get results like this:
define void @foo(i8* %p) {
; 1 = MemoryDef(liveOnEntry)
store i8 0, i8* %p
br i1 undef, label %if.then, label %if.end
if.then:
; 2 = MemoryDef(1)
store i8 1, i8* %p
br label %if.end
if.end:
; 3 = MemoryDef(1)
store i8 2, i8* %p
ret void
}
...When there should be a MemoryPhi in the `if.end` BB.
This patch fixes that behavior.
llvm-svn: 263991
The sinpi/cospi can be replaced with sincospi to remove unnecessary
computations. However, we need to make sure that the calls are within
the same function!
This fixes PR26993.
llvm-svn: 263875
Summary:
ThinLTO is relying on linkInModule to import selected function.
However a lot of "magic" was hidden in linkInModule and the IRMover,
who would rename and promote global variables on the fly.
This is moving to an approach where the steps are decoupled and the
client is reponsible to specify the list of globals to import.
As a consequence some test are changed because they were relying on
the previous behavior which was importing the definition of *every*
single global without control on the client side.
Now the burden is on the client to decide if a global has to be imported
or not.
Reviewers: tejohnson
Subscribers: joker.eph, llvm-commits
Differential Revision: http://reviews.llvm.org/D18122
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 263863
The loop on IVOperand's incoming values assumes IVOperand to be an
induction variable on the loop over which `S Pred X` is invariant;
otherwise loop invariant incoming values to IVOperand are not guaranteed
to dominate the comparision.
This fixes PR26973.
llvm-svn: 263827
Summary:
These dependencies would be used in the future to reduce the number
of instrumented blocks(http://reviews.llvm.org/rL262103)
This is submitted as a separate CL because of previous problems with
ARM.
Subscribers: aemerson
Differential Revision: http://reviews.llvm.org/D18227
llvm-svn: 263797
Summary:
It can hurt performance to prefetch ahead too much. Be conservative for
now and don't prefetch ahead more than 3 iterations on Cyclone.
Reviewers: hfinkel
Subscribers: llvm-commits, mzolotukhin
Differential Revision: http://reviews.llvm.org/D17949
llvm-svn: 263772
Summary:
And use this TTI for Cyclone. As it was explained in the original RFC
(http://thread.gmane.org/gmane.comp.compilers.llvm.devel/92758), the HW
prefetcher work up to 2KB strides.
I am also adding tests for this and the previous change (D17943):
* Cyclone prefetching accesses with a large stride
* Cyclone not prefetching accesses with a small stride
* Generic Aarch64 subtarget not prefetching either
Reviewers: hfinkel
Subscribers: aemerson, rengolin, llvm-commits, mzolotukhin
Differential Revision: http://reviews.llvm.org/D17945
llvm-svn: 263771
Summary:
Use the new LoopVersioning facility (D16712) to add noalias metadata in
the vector loop if we versioned with memchecks. This can enable some
optimization opportunities further down the pipeline (see the included
test or the benchmark improvement quoted in D16712).
The test also covers the bug I had in the initial version in D16712.
The vectorizer did not previously use LoopVersioning. The reason is
that the vectorizer performs its transformations in single shot. It
creates an empty single-block vector loop that it then populates with
the widened, if-converted instructions. Thus creating an intermediate
versioned scalar loop seems wasteful.
So this patch (rather than bringing in LoopVersioning fully) adds a
special interface to LoopVersioning to allow the vectorizer to add
no-alias annotation while still performing its own versioning.
As the vectorizer propagates metadata from the instructions in the
original loop to the vector instructions we also check the pointer in
the original instruction and see if LoopVersioning can add no-alias
metadata based on the issued memchecks.
Reviewers: hfinkel, nadav, mzolotukhin
Subscribers: mzolotukhin, llvm-commits
Differential Revision: http://reviews.llvm.org/D17191
llvm-svn: 263744
Summary:
If we decide to version a loop to benefit a transformation, it makes
sense to record the now non-aliasing accesses in the newly versioned
loop. This allows non-aliasing information to be used by subsequent
passes.
One example is 456.hmmer in SPECint2006 where after loop distribution,
we vectorize one of the newly distributed loops. To vectorize we
version this loop to fully disambiguate may-aliasing accesses. If we
add the noalias markers, we can use the same information in a later DSE
pass to eliminate some dead stores which amounts to ~25% of the
instructions of this hot memory-pipeline-bound loop. The overall
performance improves by 18% on our ARM64.
The scoped noalias annotation is added in LoopVersioning. The patch
then enables this for loop distribution. A follow-on patch will enable
it for the vectorizer. Eventually this should be run by default when
versioning the loop but first I'd like to get some feedback whether my
understanding and application of scoped noalias metadata is correct.
Essentially my approach was to have a separate alias domain for each
versioning of the loop. For example, if we first version in loop
distribution and then in vectorization of the distributed loops, we have
a different set of memchecks for each versioning. By keeping the scopes
in different domains they can conveniently be defined independently
since different alias domains don't affect each other.
As written, I also have a separate domain for each loop. This is not
necessary and we could save some metadata here by using the same domain
across the different loops. I don't think it's a big deal either way.
Probably the best is to review the tests first to see if I mapped this
problem correctly to scoped noalias markers. I have plenty of comments
in the tests.
Note that the interface is prepared for the vectorizer which needs the
annotateInstWithNoAlias API. The vectorizer does not use LoopVersioning
so we need a way to pass in the versioned instructions. This is also
why the maps have to become part of the object state.
Also currently, we only have an AA-aware DSE after the vectorizer if we
also run the LTO pipeline. Depending how widely this triggers we may
want to schedule a DSE toward the end of the regular pass pipeline.
Reviewers: hfinkel, nadav, ashutosh.nema
Subscribers: mssimpso, aemerson, llvm-commits, mcrosier
Differential Revision: http://reviews.llvm.org/D16712
llvm-svn: 263743
This is similar to D18133 where we allowed profile weights on select instructions.
This extends that change to also allow the 'unpredictable' attribute of branches to apply to selects.
A test to check that 'unpredictable' metadata is preserved when cloning instructions was checked in at:
http://reviews.llvm.org/rL263648
Differential Revision: http://reviews.llvm.org/D18220
llvm-svn: 263716
This splits out the logic that maps the `"statepoint-id"` attribute into
the actual statepoint ID, and the `"statepoint-num-patch-bytes"`
attribute into the number of patchable bytes the statpeoint is lowered
into. The new home of this logic is in IR/Statepoint.cpp, and this
refactoring will support similar functionality when lowering calls with
deopt operand bundles in the future.
llvm-svn: 263685
Summary:
Fix LSRInstance::HoistInsertPosition() to check the original insert
position block first for a canonical insertion point that is dominated
by all inputs. This leads to SCEV being able to reuse more instructions
since it currently tracks the instructions it creates for reuse by
keeping a table of <Value, insert point> pairs.
Reviewers: atrick
Subscribers: mcrosier, mzolotukhin, llvm-commits
Differential Revision: http://reviews.llvm.org/D18001
llvm-svn: 263644
The latent bug that LLE exposed in the LoopVectorizer was resolved
(PR26952).
The pass can be disabled with -mllvm -enable-loop-load-elim=0
llvm-svn: 263595
There is something strange going on with debug info (.eh_frame_hdr)
disappearing when msan.module_ctor are placed in comdat sections.
Moving this functionality under flag, disabled by default.
llvm-svn: 263579
This was a latent bug that got exposed by the change to add LoopSimplify
as a dependence to LoopLoadElimination. Since LoopInfo was corrupted
after LV, LoopSimplify mis-compiled nbench in the test-suite (more
details in the PR).
The problem was that when we create the blocks for predicated stores we
didn't add those to any loops.
The original testcase for store predication provides coverage for this
assuming we verify LI on the way out of LV.
Fixes PR26952.
llvm-svn: 263565
Since the static getGlobalIdentifier and getGUID methods are now called
for global values other than functions, reflect that by moving these
methods to the GlobalValue class.
llvm-svn: 263524
(Resubmitting after fixing missing file issue)
With the changes in r263275, there are now more than just functions in
the summary. Completed the renaming of data structures (started in
r263275) to reflect the wider scope. In particular, changed the
FunctionIndex* data structures to ModuleIndex*, and renamed related
variables and comments. Also renamed the files to reflect the changes.
A companion clang patch will immediately succeed this patch to reflect
this renaming.
llvm-svn: 263513
Summary:
Specifically, when we perform runtime loop unrolling of a loop that
contains a convergent op, we can only unroll k times, where k divides
the loop trip multiple.
Without this change, we'll happily unroll e.g. the following loop
for (int i = 0; i < N; ++i) {
if (i == 0) convergent_op();
foo();
}
into
int i = 0;
if (N % 2 == 1) {
convergent_op();
foo();
++i;
}
for (; i < N - 1; i += 2) {
if (i == 0) convergent_op();
foo();
foo();
}.
This is unsafe, because we've just added a control-flow dependency to
the convergent op in the prelude.
In general, runtime unrolling loops that contain convergent ops is safe
only if we don't have emit a prelude, which occurs when the unroll count
divides the trip multiple.
Reviewers: resistor
Subscribers: llvm-commits, mzolotukhin
Differential Revision: http://reviews.llvm.org/D17526
llvm-svn: 263509
Summary: This now try to reorder instructions in order to help create the optimizable pattern.
Reviewers: craig.topper, spatel, dexonsmith, Prazek, chandlerc, joker.eph, majnemer
Differential Revision: http://reviews.llvm.org/D16523
llvm-svn: 263503
With the changes in r263275, there are now more than just functions in
the summary. Completed the renaming of data structures (started in
r263275) to reflect the wider scope. In particular, changed the
FunctionIndex* data structures to ModuleIndex*, and renamed related
variables and comments. Also renamed the files to reflect the changes.
A companion clang patch will immediately succeed this patch to reflect
this renaming.
llvm-svn: 263490
As noted in:
https://llvm.org/bugs/show_bug.cgi?id=26636
This doesn't accomplish anything on its own. It's the first step towards preserving
and using branch weights with selects.
The next step would be to make sure we're propagating the info in all of the other
places where we create selects (SimplifyCFG, InstCombine, etc). I don't think there's
an easy fix to make this happen; we have to look at each transform individually to
determine how to correctly propagate the weights.
Along with that step, we need to then use the weights when making subsequent transform
decisions such as discussed in http://reviews.llvm.org/D16836.
The inliner test is independent but closely related. It verifies that metadata is
preserved when both branches and selects are cloned.
Differential Revision: http://reviews.llvm.org/D18133
llvm-svn: 263482
Summary:
Previously we had a notion of convergent functions but not of convergent
calls. This is insufficient to correctly analyze calls where the target
is unknown, e.g. indirect calls.
Now a call is convergent if it targets a known-convergent function, or
if it's explicitly marked as convergent. As usual, we can remove
convergent where we can prove that no convergent operations are
performed in the call.
Originally landed as r261544, then reverted in r261544 for (incidental)
build breakage. Re-landed here with no changes.
Reviewers: chandlerc, jingyue
Subscribers: llvm-commits, tra, jhen, hfinkel
Differential Revision: http://reviews.llvm.org/D17739
llvm-svn: 263481
The motivating example is this
for (j = n; j > 1; j = i) {
i = j / 2;
}
The signed division is safely to be changed to an unsigned division (j is known
to be larger than 1 from the loop guard) and later turned into a single shift
without considering the sign bit.
llvm-svn: 263406
This follows up on the related AVX instruction transforms, but this
one is too strange to do anything more with. Intel's behavioral
description of this instruction in its Software Developer's Manual
is tragi-comic.
llvm-svn: 263340
commit ae14bf6488e8441f0f6d74f00455555f6f3943ac
Author: Mehdi Amini <mehdi.amini@apple.com>
Date: Fri Mar 11 17:15:50 2016 +0000
Remove PreserveNames template parameter from IRBuilder
Summary:
Following r263086, we are now relying on a flag on the Context to
discard Value names in release builds.
Reviewers: chandlerc
Subscribers: mzolotukhin, llvm-commits
Differential Revision: http://reviews.llvm.org/D18023
From: Mehdi Amini <mehdi.amini@apple.com>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@263258
91177308-0d34-0410-b5e6-96231b3b80d8
until we can figure out what to do about clang and Release build testing.
This reverts commit 263258.
llvm-svn: 263321
Summary:
This intrinsic, together with deoptimization operand bundles, allow
frontends to express transfer of control and frame-local state from
one (typically more specialized, hence faster) version of a function
into another (typically more generic, hence slower) version.
In languages with a fully integrated managed runtime this intrinsic can
be used to implement "uncommon trap" like functionality. In unmanaged
languages like C and C++, this intrinsic can be used to represent the
slow paths of specialized functions.
Note: this change does not address how `@llvm.experimental_deoptimize`
is lowered. That will be done in a later change.
Reviewers: chandlerc, rnk, atrick, reames
Subscribers: llvm-commits, kmod, mjacob, maksfb, mcrosier, JosephTremoulet
Differential Revision: http://reviews.llvm.org/D17732
llvm-svn: 263281
Value profile instrumentation treats inline asm calls like they are
indirect calls. This causes problems when the 'Callee' is passed to a
ptrtoint cast -- the verifier rightly claims that this is bogus and
crashes opt.
llvm-svn: 263278
Summary:
This patch adds support for including a full reference graph including
call graph edges and other GV references in the summary.
The reference graph edges can be used to make importing decisions
without materializing any source modules, can be used in the plugin
to make file staging decisions for distributed build systems, and is
expected to have other uses.
The call graph edges are recorded in each function summary in the
bitcode via a list of <CalleeValueIds, StaticCount> tuples when no PGO
data exists, or <CalleeValueId, StaticCount, ProfileCount> pairs when
there is PGO, where the ValueId can be mapped to the function GUID via
the ValueSymbolTable. In the function index in memory, the call graph
edges reference the target via the CalleeGUID instead of the
CalleeValueId.
The reference graph edges are recorded in each summary record with a
list of referenced value IDs, which can be mapped to value GUID via the
ValueSymbolTable.
Addtionally, a new summary record type is added to record references
from global variable initializers. A number of bitcode records and data
structures have been renamed to reflect the newly expanded scope of the
summary beyond functions. More cleanup will follow.
Reviewers: joker.eph, davidxl
Subscribers: joker.eph, llvm-commits
Differential Revision: http://reviews.llvm.org/D17212
llvm-svn: 263275
Summary:
Following r263086, we are now relying on a flag on the Context to
discard Value names in release builds.
Reviewers: chandlerc
Subscribers: mzolotukhin, llvm-commits
Differential Revision: http://reviews.llvm.org/D18023
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 263258
Summary:
Following r263086, we are replacing this by a runtime check.
More cleanup will follow on the IRBuilder itself, but I submitted
this patch separately as SROA has a fancy "prefixInserter" class
that needs extra-love.
Reviewers: chandlerc
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D18022
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 263256
member type.
Because of how this type is used by the ValueTable, it cannot actually
have hidden visibility. GCC actually nicely warns about this but Clang
just silently ... I don't even know. =/ We should do a better job either
way though.
This should resolve a bunch of the GCC warnings about visibility that
the port of GVN triggered and make the visibility story a bit more
correct.
llvm-svn: 263250
much to my horror, so use variables to fix it in place.
This terrifies me. Both basic-aa and memdep will provide more precise
information when the domtree and/or the loop info is available. Because
of this, if your pass (like GVN) requires domtree, and then queries
memdep or basic-aa, it will get more precise results. If it does this in
the other order, it gets less precise results.
All of the ideas I have for fixing this are, essentially, terrible. Here
I've just caused us to stop having unspecified behavior as different
implementations evaluate the order of these arguments differently. I'm
actually rather glad that they do, or the fragility of memdep and
basic-aa would have gone on unnoticed. I've left comments so we don't
immediately break this again. This should fix bots whose host compilers
evaluate the order of arguments differently from Clang.
llvm-svn: 263231
This was originally a pointer to support pass managers which didn't use
AnalysisManagers. However, that doesn't realistically come up much and
the complexity of supporting it doesn't really make sense.
In fact, *many* parts of the pass manager were just assuming the pointer
was never null already. This at least makes it much more explicit and
clear.
llvm-svn: 263219
Since the names are used in a loop this does more work in debug builds. In
release builds value names are generally discarded so we don't have to do
the concatenation at all. It's also simpler code, no functional change
intended.
llvm-svn: 263215
tests to run GVN in both modes.
This is mostly the boring refactoring just like SROA and other complex
transformation passes. There is some trickiness in that GVN's
ValueNumber class requires hand holding to get to compile cleanly. I'm
open to suggestions about a better pattern there, but I tried several
before settling on this. I was trying to balance my desire to sink as
much implementation detail into the source file as possible without
introducing overly many layers of abstraction.
Much like with SROA, the design of this system is made somewhat more
cumbersome by the need to support both pass managers without duplicating
the significant state and logic of the pass. The same compromise is
struck here.
I've also left a FIXME in a doxygen comment as the GVN pass seems to
have pretty woeful documentation within it. I'd like to submit this with
the FIXME and let those more deeply familiar backfill the information
here now that we have a nice place in an interface to put that kind of
documentaiton.
Differential Revision: http://reviews.llvm.org/D18019
llvm-svn: 263208
llvm::getDISubprogram walks the instructions in a function, looking for one in the scope of the current function, so that it can find the !dbg entry for the subprogram itself.
Now that !dbg is attached to functions, this should not be necessary. This patch changes all uses to just query the subprogram directly on the function.
Ideally this should be NFC, but in reality its possible that a function:
has no !dbg (in which case there's likely a bug somewhere in an opt pass), or
that none of the instructions had a scope referencing the function, so we used to not find the !dbg on the function but now we will
Reviewed by Duncan Exon Smith.
Differential Revision: http://reviews.llvm.org/D18074
llvm-svn: 263184
The code assumed that we always had a preheader without making the pass
dependent on LoopSimplify.
Thanks to Mattias Eriksson V for reporting this.
llvm-svn: 263173
of, and I misdiagnosed for months and months.
Andrea has had a patch for this forever, but I just couldn't see how
it was fixing the root cause of the problem. It didn't make sense to me,
even though the patch was perfectly good and the analysis of the actual
failure event was *fantastic*.
Well, I came back to it today because the patch has sat for *far* too
long and needs attention and decided I wouldn't let it go until I really
understood what was going on. After quite some time in the debugger,
I finally realized that in fact I had just missed an important case with
my previous attempt to fix PR22093 in r225149. Not only do we need to
handle loads that won't be split, but stores-of-loads that we won't
split. We *do* actually have enough logic in the presplitting to form
new slices for split stores.... *unless* we decided not to split them!
I'm so sorry that it took me this long to come to the realization that
this is the issue. It seems so obvious in hind sight (of course).
Anyways, the fix becomes *much* smaller and more focused. The fact that
we're left doing integer smashing is related to the FIXME in my original
commit: fundamentally, we're not aggressive about pre-splitting for
loads and stores to the same alloca. If we want to get aggressive about
this, it'll need both what Andrea had put into the proposed fix, but
also a *lot* more logic to essentially iteratively pre-split the alloca
until we can't do any more. As I said in that commit log, its really
unclear that this is the right call. Instead, the integer blending and
letting targets lower this to narrower stores seems slightly better. But
we definitely shouldn't really go down that path just to fix this bug.
Again, tons of thanks are owed to Andrea and others at Sony for working
on this bug. I really should have seen what was going on here and
re-directed them sooner. =////
llvm-svn: 263121
We already have the instruction extracted into 'I', just cast that to
a store the way we do for loads. Also, we don't enter the if unless SI
is non-null, so don't test it again for null.
I'm pretty sure the entire test there can be nuked, but this is just the
trivial cleanup.
llvm-svn: 263112
MinVecRegSize is currently hardcoded to 128; this patch adds a cl::opt
to allow changing it. I tried not to change any existing behavior for the default
case.
Differential revision: http://reviews.llvm.org/D13278
llvm-svn: 263089
need to be changed for porting to the new pass manager.
Also sink the comment on the ValueTable class back to that class instead
of it dangling on an anonymous namespace.
No functionality changed.
llvm-svn: 263084
This is a fairly straightforward port to the new pass manager with one
exception. It removes a very questionable use of releaseMemory() in
the old pass to invalidate its caches between runs on a function.
I don't think this is really guaranteed to be safe. I've just used the
more direct port to the new PM to address this by nuking the results
object each time the pass runs. While this could cause some minor malloc
traffic increase, I don't expect the compile time performance hit to be
noticable, and it makes the correctness and other aspects of the pass
much easier to reason about. In some cases, it may make things faster by
making the sets and maps smaller with better locality. Indeed, the
measurements collected by Bruno (thanks!!!) show mostly compile time
improvements.
There is sadly very limited testing at this point as there are only two
tests of memdep, and both rely on GVN. I'll be porting GVN next and that
will exercise this heavily though.
Differential Revision: http://reviews.llvm.org/D17962
llvm-svn: 263082
This patch teaches LICM's implementation of store promotion to exploit the fact that the memory location being accessed might be provable thread local. The fact it's thread local weakens the requirements for where we can insert stores since no other thread can observe the write. This allows us perform store promotion even in cases where the store is not guaranteed to execute in the loop.
Two key assumption worth drawing out is that this assumes a) no-capture is strong enough to imply no-escape, and b) standard allocation functions like malloc, calloc, and operator new return values which can be assumed not to have previously escaped.
In future work, it would be nice to generalize this so that it works without directly seeing the allocation site. I believe that the nocapture return attribute should be suitable for this purpose, but haven't investigated carefully. It's also likely that we could support unescaped allocas with similar reasoning, but since SROA and Mem2Reg should destroy those, they're less interesting than they first might seem.
Differential Revision: http://reviews.llvm.org/D16783
llvm-svn: 263072
When checking whether an smin is positive, we can move the comparison to one of the inputs if the other is known positive. If the known positive one is the min, then the other can't be negative. If the other is the min, then we compute the min.
Differential Revision: http://reviews.llvm.org/D17873
llvm-svn: 263059
I somehow missed this. The case in GCC (global_alloc) was similar to
the new testcase except it had an array of structs rather than a two
dimensional array.
Fixes RP26885.
llvm-svn: 263058
As part of r251146 InstCombine was extended to call computeKnownBits on
every value in the function to determine whether it happens to be
constant. This increases typical compiletime by 1-3% (5% in irgen+opt
time) in my measurements. On the other hand this case did not trigger
once in the whole llvm-testsuite.
This patch introduces the notion of ExpensiveCombines which are only
enabled for OptLevel > 2. I removed the check in InstructionSimplify as
that is called from various places where the OptLevel is not known but
given the rarity of the situation I think a check in InstCombine is
enough.
Differential Revision: http://reviews.llvm.org/D16835
llvm-svn: 263047
Original commit message:
calculate builtin_object_size if argument is a removable pointer
This patch fixes calculating correct value for builtin_object_size function
when pointer is used only in builtin_object_size function call and never
after that.
Patch by Strahinja Petrovic.
Differential Revision: http://reviews.llvm.org/D17337
Reland the original change with a small modification (first do a null check
and then do the cast) to satisfy ubsan.
llvm-svn: 263011
TSan instrumentation functions for atomic stores, loads, and cmpxchg work on
integer value types. This patch adds casts before calling TSan instrumentation
functions in cases where the value is a pointer.
Differential Revision: http://reviews.llvm.org/D17833
llvm-svn: 262876
This lets select sub-targets enable this pass. The patch implements the
idea from the recent llvm-dev thread:
http://thread.gmane.org/gmane.comp.compilers.llvm.devel/94925
The goal is to enable the LoopDataPrefetch pass for the Cyclone
sub-target only within Aarch64.
Positive and negative tests will be included in an upcoming patch that
enables selective prefetching of large-strided accesses on Cyclone.
llvm-svn: 262844
This reverts commit r262250.
It causes SPEC2006/gcc to generate wrong result (166.s) in AArch64 when
running with *ref* data set. The error happens with
"-Ofast -flto -fuse-ld=gold" or "-O3 -fno-strict-aliasing".
llvm-svn: 262839
This code has been successfully used to bootstrap libc++ in a no-asserts
mode for a very long time, so the code that follows cannot be completely
incorrect. I've added a test that shows the current behavior for this
kind of code with DFSan. If it is desirable for DFSan to do something
special when processing an invoke of a variadic function, it can be
added, but we shouldn't keep an assert that we've been ignoring due to
release builds anyways.
llvm-svn: 262829
Given that we're not actually reducing the instruction count in the included
regression tests, I think we would call this a canonicalization step.
The motivation comes from the example in PR26702:
https://llvm.org/bugs/show_bug.cgi?id=26702
If we hoist the bitwise logic ahead of the bitcast, the previously unoptimizable
example of:
define <4 x i32> @is_negative(<4 x i32> %x) {
%lobit = ashr <4 x i32> %x, <i32 31, i32 31, i32 31, i32 31>
%not = xor <4 x i32> %lobit, <i32 -1, i32 -1, i32 -1, i32 -1>
%bc = bitcast <4 x i32> %not to <2 x i64>
%notnot = xor <2 x i64> %bc, <i64 -1, i64 -1>
%bc2 = bitcast <2 x i64> %notnot to <4 x i32>
ret <4 x i32> %bc2
}
Simplifies to the expected:
define <4 x i32> @is_negative(<4 x i32> %x) {
%lobit = ashr <4 x i32> %x, <i32 31, i32 31, i32 31, i32 31>
ret <4 x i32> %lobit
}
Differential Revision: http://reviews.llvm.org/D17583
llvm-svn: 262645
This patch provides the following infrastructure for PGO enhancements in inliner:
Enable the use of block level profile information in inliner
Incremental update of block frequency information during inlining
Update the function entry counts of callees when they get inlined into callers.
Differential Revision: http://reviews.llvm.org/D16381
llvm-svn: 262636
Summary: With discriminator, LineLocation can uniquely identify a callsite without the need to specifying callee name. Remove Callee function name from the key, and put it in the value (FunctionSamples).
Reviewers: davidxl, dnovillo
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D17827
llvm-svn: 262634
The vectorization of first-order recurrences (r261346) caused PR26734. When
detecting these recurrences, we need to ensure that the previous value is
actually defined inside the loop. This patch includes the fix and test case.
llvm-svn: 262624
Summary: This is the last step toward supporting aggregate memory access in instcombine. This explodes stores of arrays into a serie of stores for each element, allowing them to be optimized.
Reviewers: joker.eph, reames, hfinkel, majnemer, mgrang
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D17828
llvm-svn: 262530
Summary: This is another step toward improving fca support. This unpack load of array in a series of load to array's elements.
Reviewers: chandlerc, joker.eph, majnemer, reames, hfinkel
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D15890
llvm-svn: 262521
parts of the AA interface out of the base class of every single AA
result object.
Because this logic reformulates the query in terms of some other aspect
of the API, it would easily cause O(n^2) query patterns in alias
analysis. These could in turn be magnified further based on the number
of call arguments, and then further based on the number of AA queries
made for a particular call. This ended up causing problems for Rust that
were actually noticable enough to get a bug (PR26564) and probably other
places as well.
When originally re-working the AA infrastructure, the desire was to
regularize the pattern of refinement without losing any generality.
While I think it was successful, that is clearly proving to be too
costly. And the cost is needless: we gain no actual improvement for this
generality of making a direct query to tbaa actually be able to
re-use some other alias analysis's refinement logic for one of the other
APIs, or some such. In short, this is entirely wasted work.
To the extent possible, delegation to other API surfaces should be done
at the aggregation layer so that we can avoid re-walking the
aggregation. In fact, this significantly simplifies the logic as we no
longer need to smuggle the aggregation layer into each alias analysis
(or the TargetLibraryInfo into each alias analysis just so we can form
argument memory locations!).
However, we also have some delegation logic inside of BasicAA and some
of it even makes sense. When the delegation logic is baking in specific
knowledge of aliasing properties of the LLVM IR, as opposed to simply
reformulating the query to utilize a different alias analysis interface
entry point, it makes a lot of sense to restrict that logic to
a different layer such as BasicAA. So one aspect of the delegation that
was in every AA base class is that when we don't have operand bundles,
we re-use function AA results as a fallback for callsite alias results.
This relies on the IR properties of calls and functions w.r.t. aliasing,
and so seems a better fit to BasicAA. I've lifted the logic up to that
point where it seems to be a natural fit. This still does a bit of
redundant work (we query function attributes twice, once via the
callsite and once via the function AA query) but it is *exactly* twice
here, no more.
The end result is that all of the delegation logic is hoisted out of the
base class and into either the aggregation layer when it is a pure
retargeting to a different API surface, or into BasicAA when it relies
on the IR's aliasing properties. This should fix the quadratic query
pattern reported in PR26564, although I don't have a stand-alone test
case to reproduce it.
It also seems general goodness. Now the numerous AAs that don't need
target library info don't carry it around and depend on it. I think
I can even rip out the general access to the aggregation layer and only
expose that in BasicAA as it is the only place where we re-query in that
manner.
However, this is a non-trivial change to the AA infrastructure so I want
to get some additional eyes on this before it lands. Sadly, it can't
wait long because we should really cherry pick this into 3.8 if we're
going to go this route.
Differential Revision: http://reviews.llvm.org/D17329
llvm-svn: 262490
As noted in the code comment, I don't think we can do the same transform that we do for
*scalar* integers comparisons to *vector* integers comparisons because it might pessimize
the general case.
Exhibit A for an incomplete integer comparison ISA remains x86 SSE/AVX: it only has EQ and GT
for integer vectors.
But we should now recognize all the variants of this construct and produce the optimal code
for the cases shown in:
https://llvm.org/bugs/show_bug.cgi?id=26701
llvm-svn: 262424
Summary: SampleProfile pass needs to be performed after InstructionCombiningPass, which helps eliminate un-inlinable function calls.
Reviewers: davidxl, dnovillo
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D17742
llvm-svn: 262419
Summary:
This adds the beginning of an update API to preserve MemorySSA. In particular,
this patch adds a way to remove memory SSA accesses when instructions are
deleted.
It also adds relevant unit testing infrastructure for MemorySSA's API.
(There is an actual user of this API, i will make that diff dependent on this one. In practice, a ton of opt passes remove memory instructions, so it's hopefully an obviously useful API :P)
Reviewers: hfinkel, reames, george.burgess.iv
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D17157
llvm-svn: 262362
This patch fixes calculating correct value for builtin_object_size function
when pointer is used only in builtin_object_size function call and never
after that.
Patch by Strahinja Petrovic.
Differential Revision: http://reviews.llvm.org/D17337
llvm-svn: 262337
The intended effect of this patch in conjunction with:
http://reviews.llvm.org/rL259392http://reviews.llvm.org/rL260145
is that customers using the AVX intrinsics in C will benefit from combines when
the load mask is constant:
__m128 mload_zeros(float *f) {
return _mm_maskload_ps(f, _mm_set1_epi32(0));
}
__m128 mload_fakeones(float *f) {
return _mm_maskload_ps(f, _mm_set1_epi32(1));
}
__m128 mload_ones(float *f) {
return _mm_maskload_ps(f, _mm_set1_epi32(0x80000000));
}
__m128 mload_oneset(float *f) {
return _mm_maskload_ps(f, _mm_set_epi32(0x80000000, 0, 0, 0));
}
...so none of the above will actually generate a masked load for optimized code.
This is the masked load counterpart to:
http://reviews.llvm.org/rL262064
llvm-svn: 262269
Summary:
I re-benchmarked this and results are similar to original results in
D13259:
On ARM64:
SingleSource/Benchmarks/Polybench/linear-algebra/solvers/dynprog -59.27%
SingleSource/Benchmarks/Polybench/stencils/adi -19.78%
On x86:
SingleSource/Benchmarks/Polybench/linear-algebra/solvers/dynprog -27.14%
And of course the original ~20% gain on SPECint_2006/456.hmmer with Loop
Distribution.
In terms of compile time, there is ~5% increase on both
SingleSource/Benchmarks/Misc/oourafft and
SingleSource/Benchmarks/Linkpack/linkpack-pc. These are both very tiny
loop-intensive programs where SCEV computations dominates compile time.
The reason that time spent in SCEV increases has to do with the design
of the old pass manager. If a transform pass does not preserve an
analysis we *invalidate* the analysis even if there was *no*
modification made by the transform pass.
This means that currently we don't take advantage of LLE and LV sharing
the same analysis (LAA) and unfortunately we recompute LAA *and* SCEV
for LLE.
(There should be a way to work around this limitation in the case of
SCEV and LAA since both compute things on demand and internally cache
their result. Thus we could pretend that transform passes preserve
these analyses and manually invalidate them upon actual modification.
On the other hand the new pass manager is supposed to solve so I am not
sure if this is worthwhile.)
Reviewers: hfinkel, dberlin
Subscribers: dberlin, reames, mssimpso, aemerson, joker.eph, llvm-commits
Differential Revision: http://reviews.llvm.org/D16300
llvm-svn: 262250
Summary:
The PS4 linker seems to handle this fine.
Hi David, it seems that indeed most ELF linkers support
__{start,stop}_SECNAME, as our proprietary linker does as well.
This follows the pattern of r250679 w.r.t. the testing.
Maggie, Phillip, Paul: I've tested this with the PS4 SDK 3.5 toolchain
prerelease and it seems to work fine.
Reviewers: davidxl
Subscribers: probinson, phillip.power, MaggieYi
Differential Revision: http://reviews.llvm.org/D17672
llvm-svn: 262112
merged into a loop that was subsequently unrolled (or otherwise nuked).
In this case it can't merge in the ASTs for any remaining nested loops,
it needs to re-add their instructions dircetly.
The fix is very isolated, but I've pulled the code for merging blocks
into the AST into a single place in the process. The only behavior
change is in the case which would have crashed before.
This fixes a crash reported by Mikael Holmen on the list after r261316
restored much of the loop pass pipelining and allowed us to actually do
this kind of nested transformation sequenc. I've taken that test case
and further reduced it into the somewhat twisty maze of loops in the
included test case. This does in fact trigger the bug even in this
reduced form.
llvm-svn: 262108
Summary:
Without tree pruning clang has 2,667,552 points.
Wiht only dominators pruning: 1,515,586.
With both dominators & predominators pruning: 1,340,534.
Differential Revision: http://reviews.llvm.org/D17671
llvm-svn: 262103
The intended effect of this patch in conjunction with:
http://reviews.llvm.org/rL259392http://reviews.llvm.org/rL260145
is that customers using the AVX intrinsics in C will benefit from combines when
the store mask is constant:
void mstore_zero_mask(float *f, __m128 v) {
_mm_maskstore_ps(f, _mm_set1_epi32(0), v);
}
void mstore_fake_ones_mask(float *f, __m128 v) {
_mm_maskstore_ps(f, _mm_set1_epi32(1), v);
}
void mstore_ones_mask(float *f, __m128 v) {
_mm_maskstore_ps(f, _mm_set1_epi32(0x80000000), v);
}
void mstore_one_set_elt_mask(float *f, __m128 v) {
_mm_maskstore_ps(f, _mm_set_epi32(0x80000000, 0, 0, 0), v);
}
...so none of the above will actually generate a masked store for optimized code.
Differential Revision: http://reviews.llvm.org/D17485
llvm-svn: 262064
Summary: Check that we're using SCEV for the same loop we're simulating. Otherwise, we might try to use the iteration number of the current loop in SCEV expressions for inner/outer loops IVs, which is clearly incorrect.
Reviewers: chandlerc, hfinkel
Subscribers: sanjoy, llvm-commits, mzolotukhin
Differential Revision: http://reviews.llvm.org/D17632
llvm-svn: 261958
Summary:
This is the first simple attempt to reduce number of coverage-
instrumented blocks.
If a basic block dominates all its successors, then its coverage
information is useless to us. Ingore such blocks if
santizer-coverage-prune-tree option is set.
Differential Revision: http://reviews.llvm.org/D17626
llvm-svn: 261949
The cleanupret instruction has an invariant that it's 'from' operand be
a cleanuppad. This invariant was violated when we removed a dead block
which removed a cleanuppad leaving behind a cleanupret with an undef
'from' operand.
This was solved in r261731 by staving off the removal of the dead block
to a later pass.
However, it occured to me that we do not need to do this.
Instead, we can simply avoid processing the cleanupret if it has an
undef 'from' operand because we know that it will be removed soon.
llvm-svn: 261754
This is part of the payoff for the refactoring in:
http://reviews.llvm.org/rL261649http://reviews.llvm.org/rL261707
In addition to removing a pile of duplicated code, the xor case was
missing the optimization for vector types because it checked
"SrcTy->isIntegerTy()" rather than "SrcTy->isIntOrIntVectorTy()"
like 'and' and 'or' were already doing.
This solves part of:
https://llvm.org/bugs/show_bug.cgi?id=26702
llvm-svn: 261750
This is a part of the refactoring to unify isSafeToLoadUnconditionally and isDereferenceablePointer functions. In subsequent change I'm going to eliminate isDerferenceableAndAlignedPointer from Loads API, leaving isSafeToLoadSpecualtively the only function to check is load instruction can be speculated.
Reviewed By: hfinkel
Differential Revision: http://reviews.llvm.org/D16180
llvm-svn: 261736
DeleteDeadBlock was called indiscriminately, leading to cleanuprets with
undef cleanuppad references.
Instead, try to drain the BB of most of it's instructions if it is
unreachable. We can then remove the BB if it solely consists of a
terminator (and maybe some phis).
llvm-svn: 261731
Note: The 'and' case in foldCastedBitwiseLogic() is inheriting one extra
check from the nearly identical 'or' case:
if ((!isa<ICmpInst>(Cast0Src) || !isa<ICmpInst>(Cast1Src))
But I'm not sure how to expose that difference in a regression test.
Without that check, the 'or' path will infinite loop on:
test/Transforms/InstCombine/zext-or-icmp.ll
because the zext-or-icmp fold is attempting a reverse transform.
The refactoring should extend to the 'xor' case next to solve part of
PR26702.
llvm-svn: 261707
It is problematic if the inlinee has a cleanupret which unwinds to
caller and we inline it into a call site which doesn't unwind.
If the funclet unwinds anywhere other than to the caller,
then we will give the funclet two unwind destinations.
This will result in a verifier failure.
Seeing as how the caller wasn't an invoke (which would locally unwind)
and that the funclet cannot unwind to caller, we must conclude that an
'unwind to caller' cleanupret is dynamically unreachable.
This fixes PR26698.
Differential Revision: http://reviews.llvm.org/D17536
llvm-svn: 261656
This is a straight cut and paste of the existing code and is intended to
be the first step in solving part of PR26702:
https://llvm.org/bugs/show_bug.cgi?id=26702
We should be able to reuse most of this and delete the nearly identical
existing code in visitOr(). Then, we can enhance visitXor() to use the
same code too.
llvm-svn: 261649
Summary:
When we completely unroll a loop, it's pretty easy to update DT in-place and
thus avoid rebuilding it. DT recalculation is one of the most time-consuming
tasks in loop-unroll, so avoiding it at least in case of full unroll should be
beneficial.
On some extreme (but still real-world) tests this patch improves compile time by
~2x.
Reviewers: escha, jmolloy, hfinkel, sanjoy, chandlerc
Subscribers: joker.eph, sanjoy, llvm-commits
Differential Revision: http://reviews.llvm.org/D17473
llvm-svn: 261595
Summary:
Since this is an IR pass it's nice to be able to write tests without
llc. This is the counterpart of the llc test under
CodeGen/PowerPC/loop-data-prefetch.ll.
Reviewers: hfinkel
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D17464
llvm-svn: 261578
The issue was that we only required LCSSA rebuilding if the immediate
parent-loop had values used outside of it. The fix is to enaable the
same logic for all outer loops, not only immediate parent.
llvm-svn: 261575
This flag was part of a migration to a new means of handling vectors-of-points which was described in the llvm-dev thread "FYI: Relocating vector of pointers". The old code path has been off by default for a while without complaints, so time to cleanup.
llvm-svn: 261569
This change reverts "246133 [RewriteStatepointsForGC] Reduce the number of new instructions for base pointers" and a follow on bugfix 12575.
As pointed out in pr25846, this code suffers from a memory corruption bug. Since I'm (empirically) not going to get back to this any time soon, simply reverting the problematic change is the right answer.
llvm-svn: 261565
Summary:
Previously we had a notion of convergent functions but not of convergent
calls. This is insufficient to correctly analyze calls where the target
is unknown, e.g. indirect calls.
Now a call is convergent if it targets a known-convergent function, or
if it's explicitly marked as convergent. As usual, we can remove
convergent where we can prove that no convergent operations are
performed in the call.
Reviewers: chandlerc, jingyue
Subscribers: hfinkel, jhen, tra, llvm-commits
Differential Revision: http://reviews.llvm.org/D17317
llvm-svn: 261544
I missed == and != when I removed implicit conversions between iterators
and pointers in r252380 since they were defined outside ilist_iterator.
Since they depend on getNodePtrUnchecked(), they indirectly rely on UB.
This commit removes all uses of these operators. (I'll delete the
operators themselves in a separate commit so that it can be easily
reverted if necessary.)
There should be NFC here.
llvm-svn: 261498
Stop relying on `getNodePtrUnchecked()` being useful on invalid
iterators. This function is documented to be for internal use only, and
the pointer type will eventually have to change to remove UB from
ilist_iterator. Instead, check the iterator before it has been
invalidated.
llvm-svn: 261497
Cleanuppads may be merged together if one is the only predecessor of the
other in which case a simple transform can be performed: replace the
a cleanupret with a branch and remove an unnecessary cleanuppad.
Differential Revision: http://reviews.llvm.org/D17459
llvm-svn: 261390
This patch enables the vectorization of first-order recurrences. A first-order
recurrence is a non-reduction recurrence relation in which the value of the
recurrence in the current loop iteration equals a value defined in the previous
iteration. The load PRE of the GVN pass often creates these recurrences by
hoisting loads from within loops.
In this patch, we add a new recurrence kind for first-order phi nodes and
attempt to vectorize them if possible. Vectorization is performed by shuffling
the values for the current and previous iterations. The vectorization cost
estimate is updated to account for the added shuffle instruction.
Contributed-by: Matthew Simpson and Chad Rosier <mcrosier@codeaurora.org>
Differential Revision: http://reviews.llvm.org/D16197
llvm-svn: 261346
Summary:
If we don't have the first and last access of an interleaved load group,
the first and last wide load in the loop can do an out of bounds
access. Even though we discard results from speculative loads,
this can cause problems, since it can technically generate page faults
(or worse).
We now discard interleaved load groups that don't have the first and
load in the group.
Reviewers: hfinkel, rengolin
Subscribers: rengolin, llvm-commits, mzolotukhin, anemet
Differential Revision: http://reviews.llvm.org/D17332
llvm-svn: 261331
routine.
We were getting this wrong in small ways and generally being very
inconsistent about it across loop passes. Instead, let's have a common
place where we do this. One minor downside is that this will require
some analyses like SCEV in more places than they are strictly needed.
However, this seems benign as these analyses are complete no-ops, and
without this consistency we can in many cases end up with the legacy
pass manager scheduling deciding to split up a loop pass pipeline in
order to run the function analysis half-way through. It is very, very
annoying to fix these without just being very pedantic across the board.
The only loop passes I've not updated here are ones that use
AU.setPreservesAll() such as IVUsers (an analysis) and the pass printer.
They seemed less relevant.
With this patch, almost all of the problems in PR24804 around loop pass
pipelines are fixed. The one remaining issue is that we run simplify-cfg
and instcombine in the middle of the loop pass pipeline. We've recently
added some loop variants of these passes that would seem substantially
cleaner to use, but this at least gets us much closer to the previous
state. Notably, the seven loop pass managers is down to three.
I've not updated the loop passes using LoopAccessAnalysis because that
analysis hasn't been fully wired into LoopSimplify/LCSSA, and it isn't
clear that those transforms want to support those forms anyways. They
all run late anyways, so this is harmless. Similarly, LSR is left alone
because it already carefully manages its forms and doesn't need to get
fused into a single loop pass manager with a bunch of other loop passes.
LoopReroll didn't use loop simplified form previously, and I've updated
the test case to match the trivially different output.
Finally, I've also factored all the pass initialization for the passes
that use this technique as well, so that should be done regularly and
reliably.
Thanks to James for the help reviewing and thinking about this stuff,
and Ben for help thinking about it as well!
Differential Revision: http://reviews.llvm.org/D17435
llvm-svn: 261316
more places to prevent gratuitous re-"runs" of these passes.
The passes themselves don't do any work when run, but we keep spending
time scheduling and running these needlessly when we really don't need
to do so.
This is the first patch towards fixing the really horrible loop pass
pipeline fragmentation pointed out by Sanjoy in PR24804.
llvm-svn: 261302
This patch is part of the work to make PPCLoopDataPrefetch
target-independent
(http://thread.gmane.org/gmane.comp.compilers.llvm.devel/92758).
Obviously the pass still only used from PPC at this point. Subsequent
patches will start driving this from ARM64 as well.
Due to the previous patch most lines should show up as moved lines.
llvm-svn: 261265
Commit r259357 was reverted because it caused PR26629. We were assuming all
roots of a vectorizable tree could be truncated to the same width, which is not
the case in general. This commit reapplies the patch along with a fix and a new
test case to ensure we don't regress because of this issue again. This should
fix PR26629.
llvm-svn: 261212
convert one test to use this.
This is a particularly significant milestone because it required
a working per-function AA framework which can be queried over each
function from within a CGSCC transform pass (and additionally a module
analysis to be accessible). This is essentially *the* point of the
entire pass manager rewrite. A CGSCC transform is able to query for
multiple different function's analysis results. It works. The whole
thing appears to actually work and accomplish the original goal. While
we were able to hack function attrs and basic-aa to "work" in the old
pass manager, this port doesn't use any of that, it directly leverages
the new fundamental functionality.
For this to work, the CGSCC framework also has to support SCC-based
behavior analysis, etc. The only part of the CGSCC pass infrastructure
not sorted out at this point are the updates in the face of inlining and
running function passes that mutate the call graph.
The changes are pretty boring and boiler-plate. Most of the work was
factored into more focused preperatory patches. But this is what wires
it all together.
llvm-svn: 261203
This function is used to check whether a dbg.value intrinsic has already
been inserted, but without comparing the DIExpression, it would erroneously
fire on split aggregates and only the first scalar would survive.
Found via http://reviews.llvm.org/D16867.
<rdar://problem/24456528>
llvm-svn: 261145
Loop vectorizer now knows to vectorize GEP and create masked gather and scatter intrinsics for random memory access.
The feature is enabled on AVX-512 target.
Differential Revision: http://reviews.llvm.org/D15690
llvm-svn: 261140
Summary: Store and loads unpacked by instcombine do not always have the right alignement. This explicitely compute the alignement and set it.
Reviewers: dblaikie, majnemer, reames, hfinkel, joker.eph
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D17326
llvm-svn: 261139
When support for objc_unsafeClaimAutoreleasedReturnValue has been added to the
ARC optimizer in r258970, one case was missed which would lead the optimizer
to execute an llvm_unreachable. In this case, just handle ClaimRV in the same
way we handle RetainRV.
llvm-svn: 261134
Summary:
On the contrary to Full LTO, ThinLTO can afford to shift compile time
from the frontend to the linker: both phases are parallel (even if
it is not totally "free": projects like clang are reusing product
from the "compile phase" for multiple link, think about
libLLVMSupport reused for opt, llc, etc.).
This pipeline is based on the proposal in D13443 for full LTO. We
didn't move forward on this proposal because the LTO link was far too
long after that. We believe that we can afford it with ThinLTO.
The ThinLTO pipeline integrates in the regular O2/O3 flow:
- The compile phase perform the inliner with a somehow lighter
function simplification. (TODO: tune the inliner thresholds here)
This is intendend to simplify the IR and get rid of obvious things
like linkonce_odr that will be inlined.
- The link phase will run the pipeline from the start, extended with
some specific passes that leverage the augmented knowledge we have
during LTO. Especially after the inliner is done, a sequence of
globalDCE/globalOpt is performed, followed by another run of the
"function simplification" passes. It is not clear if this part
of the pipeline will stay as is, as the split model of ThinLTO
does not allow the same benefit as FullLTO without added tricks.
The measurements on the public test suite as well as on our internal
suite show an overall net improvement. The binary size for the clang
executable is reduced by 5%. We're still tuning it with the bringup
of ThinLTO and it will evolve, but this should provide a good starting
point.
Reviewers: tejohnson
Differential Revision: http://reviews.llvm.org/D17115
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 261029
It is intended to contains the passes run over a function after the
inliner is done with a function and before it moves to its callers.
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 261028
Summary:
Extending findExistingExpansion can use existing value in ExprValueMap.
This patch gives 0.3~0.5% performance improvements on
benchmarks(test-suite, spec2000, spec2006, commercial benchmark)
Reviewers: mzolotukhin, sanjoy, zzheng
Differential Revision: http://reviews.llvm.org/D15559
llvm-svn: 260938
Summary:
While shrinking types according to the required bits, we can
encounter insert/extract element instructions. This will cause us to
reach an llvm_unreachable statement.
This change adds support for truncating insert/extract element
operations, and adds a regression test.
Reviewers: jmolloy
Subscribers: mzolotukhin, llvm-commits
Differential Revision: http://reviews.llvm.org/D17078
llvm-svn: 260893
LICM starts with an *empty* AST, and then merges in each sub-loop. While the
add code is appropriate for sub-loop 2 and up, it's utterly unnecessary for
sub-loop 1. If the AST starts off empty, we can just clone/move the contents
of the subloop into the containing AST.
Reviewed-by: Philip Reames <listmail@philipreames.com>
Differential Revision: http://reviews.llvm.org/D16753
llvm-svn: 260892
than the SCC object, and have it scan the instruction stream directly
rather than relying on call records.
This makes the behavior of this routine consistent between libc routines
and LLVM intrinsics for libc routines. We can go and start teaching it
about those being norecurse, but we should behave the same for the
intrinsic and the libc routine rather than differently. I chatted with
James Molloy and the inconsistency doesn't seem intentional and likely
is due to intrinsic calls not being modelled in the call graph analyses.
This also fixes a bug where we would deduce norecurse on optnone
functions, when generally we try to handle optnone functions as-if they
were replaceable and thus unanalyzable.
llvm-svn: 260813
Summary:
Export the CloneDebugInfoMetadata utility, which clones all debug info
associated with a function into the first module. Also use this function
in CloneModule on each function we clone (the CloneFunction entrypoint
already does this).
Without this, cloning a module will lead to DI quality regressions,
especially since r252219 reversed the Function <-> DISubprogram edge
(before we could get lucky and have this edge preserved if the
DISubprogram itself was, e.g. due to location metadata).
This was verified to fix missing debug information in julia and
a unittest to verify the new behavior is included.
Patch by Yichao Yu! Thanks!
Reviewers: loladiro, pcc
Differential Revision: http://reviews.llvm.org/D17165
llvm-svn: 260791
Summary:
Performing this optimization duplicates the call to the convergent
function and adds new control-flow dependencies, which is a no-no.
Reviewers: jingyue
Subscribers: broune, hfinkel, tra, resistor, joker.eph, arsenm, llvm-commits, mzolotukhin
Differential Revision: http://reviews.llvm.org/D17128
llvm-svn: 260730
Summary:
Calls to convergent functions can be duplicated, but only if the
duplicates are not control-flow dependent on any additional values.
Loop rotation doesn't meet the bar.
Reviewers: jingyue
Subscribers: mzolotukhin, llvm-commits, arsenm, joker.eph, resistor, tra, hfinkel, broune
Differential Revision: http://reviews.llvm.org/D17127
llvm-svn: 260729
The attached patch removes all of the block local code for performing X-load forwarding by reusing the code used in the non-local case.
The motivation here is to remove duplication and in the process increase our test coverage of some fairly tricky code. I have some upcoming changes I'll be proposing in this area and wanted to have the code cleaned up a bit first.
Note: The review for this mostly happened in email which didn't make it to phabricator on the 258882 commit thread.
Differential Revision: http://reviews.llvm.org/D16608
llvm-svn: 260711
In short, before r252926 we were comparing an unsigned (StoreSize) against an a
APInt (Stride), which is fine and well. After we were zero extending the Stride
and then converting to an unsigned, which is not the same thing. Obviously,
Stides can also be negative. This commit just restores the original behavior.
AFAICT, it's not possible to write a test case to expose the issue because
the code already has checks to make sure the StoreSize can't overflow an
unsigned (which prevents the Stride from overflowing an unsigned as well).
llvm-svn: 260706
For some cases, InstCombine replaces the sequence of xor/sub instruction
followed by cmp instruction into a single cmp instruction.
However, this replacement may result suboptimal result especially when
the xor/sub has more than one use, as discussed in
bug 26465 (https://llvm.org/bugs/show_bug.cgi?id=26465).
This patch make the replacement happen only when xor/sub has only one
use.
Differential Revision: http://reviews.llvm.org/D16915
Patch by Taewook Oh!
llvm-svn: 260695
node set rather than walking the SCC directly.
This directly exposes the functions and has already had null entries
filtered out. We also don't need need to handle optnone as it has
already been handled in the caller -- we never try to remove convergent
when there are optnone functions in the SCC.
With this change, the code for removing convergent should work with the
new pass manager and a different SCC analysis.
llvm-svn: 260668
with the test for a non-convergent intrinsic call.
While it is possible to use the call records to search for function
calls, we're going to do an instruction scan anyways to find the
intrinsics, we can handle both cases while scanning instructions. This
will also make the logic more amenable to the new pass manager which
doesn't use the same call graph structure.
My next patch will remove use of CallGraphNode entirely and allow this
code to work with both the old and new pass manager. Fortunately, it
should also get strictly simpler without changing functionality.
llvm-svn: 260666
MSan adds a constructor to each translation unit that calls
__msan_init, and does nothing else. The idea is to run __msan_init
before any instrumented code. This results in multiple constructors
and multiple .init_array entries in the final binary, one per
translation unit. This is absolutely unnecessary; one would be
enough.
This change moves the constructors to a comdat group in order to drop
the extra ones.
llvm-svn: 260632
Original commit message:
[InstCombine] Fold IntToPtr and PtrToInt into preceding loads.
Currently we only fold a BitCast into a Load when the BitCast is its
only user.
Do the same for any no-op cast.
Patch by Philip Pfaffe!
Differential Revision: http://reviews.llvm.org/D9152
llvm-svn: 260612
Summary:
On the contrary to Full LTO, ThinLTO can afford to shift compile time
from the frontend to the linker: both phases are parallel.
This pipeline is based on the proposal in D13443 for full LTO. We ]
didn't move forward on this proposal because the link was far too long
after that.
This patch refactor the "function simplification" passes that are part
of the inliner loop in a helper function (this part is NFC and can be
commited separately to simplify the diff). The ThinLTO pipeline
integrates in the regular O2/O3 flow:
- The compile phase perform the inliner with a somehow lighter
function simplification. (TODO: tune the inliner thresholds here)
This is intendend to simplify the IR and get rid of obvious things
like linkonce_odr that will be inlined.
- The link phase will run the pipeline from the start, extended with
some specific passes that leverage the augmented knowledge we have
during LTO. Especially after the inliner is done, a sequence of
globalDCE/globalOpt is performed, followed by another run of the
"function simplification" passes.
The measurements on the public test suite as well as on our internal
suite show an overall net improvement. The binary size for the clang
executable is reduced by 5%. We're still tuning it with the bringup
of ThinLTO but this should provide a good starting point.
Reviewers: tejohnson
Subscribers: joker.eph, llvm-commits, dexonsmith
Differential Revision: http://reviews.llvm.org/D17115
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 260604
It is intended to contains the passes run over a function after the
inliner is done with a function and before it moves to its callers.
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 260603
When optimizing a extractvalue(load), we generate a load from the
aggregate type. This load didn't have alignment set and so would
get the alignment of the type. This breaks when the type is packed
and so the alignment should be lower.
For example, loading { int, int } would give us alignment of 4, but
the original load from this type may have an alignment of 1 if packed.
Reviewed by David Majnemer
Differential revision: http://reviews.llvm.org/D17158
llvm-svn: 260587
Summary:
When a PHI is used only to be compared with zero, it is possible to replace an
incoming value with any non-zero constant if the incoming value can be proved as
a known nonzero value. For example, in below code, we can replace the incoming value %v with
any non-zero constant based on the fact that the PHI is only used to be compared with zero
and %v is a known non-zero value:
%v = select %cond, 1, 2
%p = phi [%v, BB] ...
%c = icmp eq, %p, 0
Reviewers: mcrosier, jmolloy, sanjoy
Subscribers: hfinkel, mcrosier, majnemer, llvm-commits, haicheng, bmakam, mssimpso, gberry
Differential Revision: http://reviews.llvm.org/D16240
llvm-svn: 260530
Make sure we split ":" from the end of the global function id (which
is <path>:<function> for local functions) instead of the beginning to
avoid splitting at the wrong place for Windows file paths that contain
a ":".
llvm-svn: 260469
The current function importer will walk the callgraph, importing
transitively any callee that is below the threshold. This can
lead to import very deep which is costly in compile time and not
necessarily beneficial as most of the inline would happen in
imported function and not necessarilly in user code.
The actual factor has been carefully chosen by flipping a coin ;)
Some tuning need to be done (just at the existing limiting threshold).
Reviewers: tejohnson
Differential Revision: http://reviews.llvm.org/D17082
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 260466
There is not reason to pass an array of "char *" to rebuild a set if
the client already has one.
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 260462
It looks like clang has a couple of test cases which caught the fact LVI was not slightly more precise after 260439. When looking at the failures, it struck me as wasteful to be querying nullness of a constant via LVI, so instead of tweaking the clang tests, let's just stop querying constants from this source.
llvm-svn: 260451
This restores commit r260408, along with a fix for a bot failure.
The bot failure was caused by dereferencing a unique_ptr in the same
call instruction parameter list where it was passed via std::move.
Apparently due to luck this was not exposed when I built the compiler
with clang, only with gcc.
llvm-svn: 260442
Summary:
This patch uses the lower 64-bits of the MD5 hash of a function name as
a GUID in the function index, instead of storing function names. Any
local functions are first given a global name by prepending the original
source file name. This is the same naming scheme and GUID used by PGO in
the indexed profile format.
This change has a couple of benefits. The primary benefit is size
reduction in the combined index file, for example 483.xalancbmk's
combined index file was reduced by around 70%. It should also result in
memory savings for the index file in memory, as the in-memory map is
also indexed by the hash instead of the string.
Second, this enables integration with indirect call promotion, since the
indirect call profile targets are recorded using the same global naming
convention and hash. This will enable the function importer to easily
locate function summaries for indirect call profile targets to enable
their import and subsequent promotion.
The original source file name is recorded in the bitcode in a new
module-level record for use in the ThinLTO backend pipeline.
Reviewers: davidxl, joker.eph
Subscribers: llvm-commits, joker.eph
Differential Revision: http://reviews.llvm.org/D17028
llvm-svn: 260408
This patch reads the indirect-call value records in the profile and makes the
annotation in the indirect-call instruction. This is for IR level profile
instrumentation.
Differential Revision: http://reviews.llvm.org/D16935
llvm-svn: 260400
Summary:
As discussed on IRC, move the ThinLTOGlobalProcessing code out of
the linker, and into TransformUtils. The name of the class is changed
to FunctionImportGlobalProcessing.
Reviewers: joker.eph, rafael
Subscribers: joker.eph, llvm-commits
Differential Revision: http://reviews.llvm.org/D17081
llvm-svn: 260395
This patch uses one bit in profile version to differentiate Clang
instrumentation and IR level instrumentation profiles.
PGOInstrumenation generates a COMDAT variable __llvm_profile_raw_version so
that the compiler runtime can set the right profile kind.
For Maco-O platform, we generate the variable as linkonce_odr linkage as
COMDAT is not supported.
PGOInstrumenation now checks this bit to make sure it's an IR level
instrumentation profile.
The patch was submitted as r260164 but reverted due to a Darwin test breakage.
Original Differential Revision: http://reviews.llvm.org/D15540
Differential Revision: http://reviews.llvm.org/D17020
llvm-svn: 260385
Summary:
Tests for this will be added once the AMDGPU backend enables this
option.
Reviewers: arsenm
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D16602
llvm-svn: 260336
Summary:
Remove the convergent attribute on any functions which provably do not
contain or invoke any convergent functions.
After this change, we'll be able to modify clang to conservatively add
'convergent' to all functions when compiling CUDA.
Reviewers: jingyue, joker.eph
Subscribers: llvm-commits, tra, jhen, hfinkel, resistor, chandlerc, arsenm
Differential Revision: http://reviews.llvm.org/D17013
llvm-svn: 260319
This pass implements whole program optimization of virtual calls in cases
where we know (via bitset information) that the list of callees is fixed. This
includes the following:
- Single implementation devirtualization: if a virtual call has a single
possible callee, replace all calls with a direct call to that callee.
- Virtual constant propagation: if the virtual function's return type is an
integer <=64 bits and all possible callees are readnone, for each class and
each list of constant arguments: evaluate the function, store the return
value alongside the virtual table, and rewrite each virtual call as a load
from the virtual table.
- Uniform return value optimization: if the conditions for virtual constant
propagation hold and each function returns the same constant value, replace
each virtual call with that constant.
- Unique return value optimization for i1 return values: if the conditions
for virtual constant propagation hold and a single vtable's function
returns 0, or a single vtable's function returns 1, replace each virtual
call with a comparison of the vptr against that vtable's address.
Differential Revision: http://reviews.llvm.org/D16795
llvm-svn: 260312
We introduced gc.relocates of vector-of-pointer types a couple of weeks back. Somehow, I missed updating the InstCombine rule to account for this. If we hit this code path with a vector-of-pointers gc.relocate, we'd crash on a cast<PointerType>.
I also took the chance to do a bit of code style cleanup.
llvm-svn: 260279
FunctionAttrs does an "optimistic" analysis of SCCs as a unit, which
means normally it is able to disregard calls from an SCC into itself.
However, calls and invokes with operand bundles are allowed to have
memory effects not fully described by the memory effects on the call
target, so we can't be optimistic around operand-bundled calls from an
SCC into itself.
llvm-svn: 260244
Summary:
Passes that call `getAnalysisIfAvailable<T>` also need to call
`addUsedIfAvailable<T>` in `getAnalysisUsage` to indicate to the
legacy pass manager that it uses `T`. This contract was being
violated by passes that used `createLegacyPMAAResults`. This change
fixes this by exposing a helper in AliasAnalysis.h,
`addUsedAAAnalyses`, that is complementary to createLegacyPMAAResults
and does the right thing when called from `getAnalysisUsage`.
Reviewers: chandlerc
Subscribers: mcrosier, llvm-commits
Differential Revision: http://reviews.llvm.org/D17010
llvm-svn: 260183
Summary:
Unrolling Analyzer is already pretty complicated, and it becomes harder and harder to exercise it with usual IR tests, as with them we can only check the final decision: whether the loop is unrolled or not. This change factors this framework out from LoopUnrollPass to analyses, which allows to use unit tests.
The change itself is supposed to be NFC, except adding a couple of tests.
I plan to add more tests as I add new functionality and find/fix bugs.
Reviewers: chandlerc, hfinkel, sanjoy
Subscribers: zzheng, sanjoy, llvm-commits
Differential Revision: http://reviews.llvm.org/D16623
llvm-svn: 260169
This patch uses one bit in profile version to differentiate Clang
instrumentation and IR level instrumentation profiles.
PGOInstrumenation generates a COMDAT variable __llvm_profile_raw_version so
that the compiler runtime can set the right profile kind.
PGOInstrumenation now checks this bit to make sure it's an IR level
instrumentation profile.
Differential Revision: http://reviews.llvm.org/D15540
llvm-svn: 260146
This reduces sizes of instrumented object files, final binaries,
process images, and raw profile data.
The format of the indexed profile data remain the same.
Differential Revision: http://reviews.llvm.org/D16388
llvm-svn: 260117
sanitizer issue. The PredicatedScalarEvolution's copy constructor
wasn't copying the Generation value, and was leaving it un-initialized.
Original commit message:
[SCEV][LAA] Add no wrap SCEV predicates and use use them to improve strided pointer detection
Summary:
This change adds no wrap SCEV predicates with:
- support for runtime checking
- support for expression rewriting:
(sext ({x,+,y}) -> {sext(x),+,sext(y)}
(zext ({x,+,y}) -> {zext(x),+,sext(y)}
Note that we are sign extending the increment of the SCEV, even for
the zext case. This is needed to cover the fairly common case where y would
be a (small) negative integer. In order to do this, this change adds two new
flags: nusw and nssw that are applicable to AddRecExprs and permit the
transformations above.
We also change isStridedPtr in LAA to be able to make use of
these predicates. With this feature we should now always be able to
work around overflow issues in the dependence analysis.
Reviewers: mzolotukhin, sanjoy, anemet
Subscribers: mzolotukhin, sanjoy, llvm-commits, rengolin, jmolloy, hfinkel
Differential Revision: http://reviews.llvm.org/D15412
llvm-svn: 260112
Change a return statement of ComputeValueKnownInPredecessors() to be the same as
the rest return statements of the function. Otherwise, it might return true with
an empty Result when the current basic block has no predecessors and trigger the
first assert of JumpThreading::ProcessThreadableEdges().
llvm-svn: 260110
We shouldn't assert when there are no memchecks, since we
can have SCEV checks. There is already an assert covering
the case where there are no SCEV checks or memchecks.
This also changes the LAA pointer wrapping versioning test
to use the loop versioning pass (this was how I managed to
trigger the assert in the loop versioning pass).
llvm-svn: 260086
Summary:
This change adds no wrap SCEV predicates with:
- support for runtime checking
- support for expression rewriting:
(sext ({x,+,y}) -> {sext(x),+,sext(y)}
(zext ({x,+,y}) -> {zext(x),+,sext(y)}
Note that we are sign extending the increment of the SCEV, even for
the zext case. This is needed to cover the fairly common case where y would
be a (small) negative integer. In order to do this, this change adds two new
flags: nusw and nssw that are applicable to AddRecExprs and permit the
transformations above.
We also change isStridedPtr in LAA to be able to make use of
these predicates. With this feature we should now always be able to
work around overflow issues in the dependence analysis.
Reviewers: mzolotukhin, sanjoy, anemet
Subscribers: mzolotukhin, sanjoy, llvm-commits, rengolin, jmolloy, hfinkel
Differential Revision: http://reviews.llvm.org/D15412
llvm-svn: 260085
As discussed in https://github.com/google/sanitizers/issues/398, with current
implementation of poisoning globals we can have some CHECK failures or false
positives in case of mixing instrumented and non-instrumented code due to ASan
poisons innocent globals from non-sanitized binary/library. We can use private
aliases to avoid such errors. In addition, to preserve ODR violation detection,
we introduce new __odr_asan_gen_XXX symbol for each instrumented global that
indicates if this global was already registered. To detect ODR violation in
runtime, we should only check the value of indicator and report an error if it
isn't equal to zero.
Differential Revision: http://reviews.llvm.org/D15642
llvm-svn: 260075
Summary:
When alias analysis is uncertain about the aliasing between any two accesses,
it will return MayAlias. This uncertainty from alias analysis restricts LICM
from proceeding further. In cases where alias analysis is uncertain we might
use loop versioning as an alternative.
Loop Versioning will create a version of the loop with aggressive aliasing
assumptions in addition to the original with conservative (default) aliasing
assumptions. The version of the loop making aggressive aliasing assumptions
will have all the memory accesses marked as no-alias. These two versions of
loop will be preceded by a memory runtime check. This runtime check consists
of bound checks for all unique memory accessed in loop, and it ensures the
lack of memory aliasing. The result of the runtime check determines which of
the loop versions is executed: If the runtime check detects any memory
aliasing, then the original loop is executed. Otherwise, the version with
aggressive aliasing assumptions is used.
The pass is off by default and can be enabled with command line option
-enable-loop-versioning-licm.
Reviewers: hfinkel, anemet, chatur01, reames
Subscribers: MatzeB, grosser, joker.eph, sanjoy, javed.absar, sbaranga,
llvm-commits
Differential Revision: http://reviews.llvm.org/D9151
llvm-svn: 259986
In r255133 (reapplied r253126) we started to avoid redundant
recomputation of LCSSA after loop-unrolling. This patch moves one step
further in this direction - now we can avoid it for much wider range of
loops, as we start to look at IR and try to figure out if the
transformation actually breaks LCSSA phis or makes it necessary to
insert new ones.
Differential Revision: http://reviews.llvm.org/D16838
llvm-svn: 259869
Summary:
Passing the rematerialized values map to insertRematerializationStores by
value looks to be a simple oversight; update it to pass by reference.
Reviewers: reames, sanjoy
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D16911
llvm-svn: 259867
Current SCEV expansion will expand SCEV as a sequence of operations
and doesn't utilize the value already existed. This will introduce
redundent computation which may not be cleaned up throughly by
following optimizations.
This patch introduces an ExprValueMap which is a map from SCEV to the
set of equal values with the same SCEV. When a SCEV is expanded, the
set of values is checked and reused whenever possible before generating
a sequence of operations.
The original commit triggered regressions in Polly tests. The regressions
exposed two problems which have been fixed in current version.
1. Polly will generate a new function based on the old one. To generate an
instruction for the new function, it builds SCEV for the old instruction,
applies some tranformation on the SCEV generated, then expands the transformed
SCEV and insert the expanded value into new function. Because SCEV expansion
may reuse value cached in ExprValueMap, the value in old function may be
inserted into new function, which is wrong.
In SCEVExpander::expand, there is a logic to check the cached value to
be used should dominate the insertion point. However, for the above
case, the check always passes. That is because the insertion point is
in a new function, which is unreachable from the old function. However
for unreachable node, DominatorTreeBase::dominates thinks it will be
dominated by any other node.
The fix is to simply add a check that the cached value to be used in
expansion should be in the same function as the insertion point instruction.
2. When the SCEV is of scConstant type, expanding it directly is cheaper than
reusing a normal value cached. Although in the cached value set in ExprValueMap,
there is a Constant type value, but it is not easy to find it out -- the cached
Value set is not sorted according to the potential cost. Existing reuse logic
in SCEVExpander::expand simply chooses the first legal element from the cached
value set.
The fix is that when the SCEV is of scConstant type, don't try the reuse
logic. simply expand it.
Differential Revision: http://reviews.llvm.org/D12090
llvm-svn: 259736
D16251)
Summary:
This is a simpler fix to the problem than the dominator approach in
http://reviews.llvm.org/D16251. It adds only values into the gather() while loop
that have been seen before.
The actual endless loop is in the constant compare gather() routine in
Utils/SimplifyCFG.cpp. The same value ret.0.off0.i is pushed back into the
queue:
%.ret.0.off0.i = or i1 %.ret.0.off0.i, %cmp10.i
Here is what happens at the IR level:
for.cond.i: ; preds = %if.end6.i,
%if.end.i54
%ix.0.i = phi i32 [ 0, %if.end.i54 ], [ %inc.i55, %if.end6.i ]
%ret.0.off0.i = phi i1 [false, %if.end.i54], [%.ret.0.off0.i, %if.end6.i] <<<
%cmp2.i = icmp ult i32 %ix.0.i, %11
br i1 %cmp2.i, label %for.body.i, label %LBJ_TmpSimpleNeedExt.exit
if.end6.i: ; preds = %for.body.i
%cmp10.i = icmp ugt i32 %conv.i, %add9.i
%.ret.0.off0.i = or i1 %ret.0.off0.i, %cmp10.i <<<
When if.end.i54 gets eliminated which removes the definition of ret.0.off0.i.
The result is the expression %.ret.0.off0.i = or i1 %.ret.0.off0.i, %cmp10.i
(Note the first ‘or’ operand is now %.ret.0.off0.i, and *NOT* %ret.0.off0.i).
And
now there is use of .ret.0.off0.i before a definition which triggers the
“endless” loop in gather():
while(!DFT.empty()) {
V = DFT.pop_back_val(); // V is .ret.0.off0.i
if (Instruction *I = dyn_cast<Instruction>(V)) {
// If it is a || (or && depending on isEQ), process the operands.
if (I->getOpcode() == (isEQ ? Instruction::Or : Instruction::And)) {
DFT.push_back(I->getOperand(1)); // This is now .ret.0.off0.i also
DFT.push_back(I->getOperand(0));
continue; // “endless loop” for .ret.0.off0.i
}
Reviewers: reames, ahatanak
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D16839
llvm-svn: 259730
Bail out if we have a PHI on an EHPad that gets a value from a
CatchSwitchInst. Because the CatchSwitchInst cannot be split, there is
no good place to stick any instructions.
This fixes PR26373.
llvm-svn: 259702
According to git bisect, this is the root cause of a miscompile for Regex in
libLLVMSupport. I am still working on reducing a test case.
The actual bug may be elsewhere and this commit just exposed it.
Anyway, at the moment, to reproduce, follow these steps:
1. Build clang and libLTO in release mode.
2. Create a new build directory <stage2> and cd into it.
3. Use clang and libLTO from #1 to build llvm-extract in Release mode + asserts
using -O2 -flto
4. Run llvm-extract -ralias '.*bar' -S test/Other/extract-alias.ll
Result:
program doesn't contain global named '.*bar'!
Expected result:
@a0a0bar = alias void ()* @bar
@a0bar = alias void ()* @bar
declare void @bar()
Note: In step #3, if you don't use lto or asserts, the miscompile disappears.
llvm-svn: 259674
Current SCEV expansion will expand SCEV as a sequence of operations
and doesn't utilize the value already existed. This will introduce
redundent computation which may not be cleaned up throughly by
following optimizations.
This patch introduces an ExprValueMap which is a map from SCEV to the
set of equal values with the same SCEV. When a SCEV is expanded, the
set of values is checked and reused whenever possible before generating
a sequence of operations.
Differential Revision: http://reviews.llvm.org/D12090
llvm-svn: 259662
Summary:
LoopVersioning is a transform utility that transform passes can use to
run-time disambiguate may-aliasing accesses. I'd like to also expose as
pass to allow it to be unit-tested.
I am planning to add support for non-aliasing annotation in
LoopVersioning and I'd like to be able to write tests directly using
this pass.
(After that feature is done, the pass could also be used to look for
optimization opportunities that are hidden behind incomplete alias
information at compile time.)
The pass drives LoopVersioning in its default way which is to fully
disambiguate may-aliasing accesses no matter how many checks are
required.
Reviewers: hfinkel, ashutosh.nema, sbaranga
Subscribers: zzheng, mssimpso, llvm-commits, sanjoy
Differential Revision: http://reviews.llvm.org/D16612
llvm-svn: 259610
Please see include/llvm/Transforms/Utils/MemorySSA.h for a description
of MemorySSA, and what it does.
Differential Revision: http://reviews.llvm.org/D7864
llvm-svn: 259595
A masked store with a zero mask means there's no store.
A masked store with an allOnes mask means it's a normal vector store.
This is a continuation of:
http://reviews.llvm.org/rL259369
llvm-svn: 259392
This miscompile came about because we tried to use a transform which was
only appropriate for xor operators when addition was present.
This fixes PR26407.
llvm-svn: 259375
A masked load with a zero mask means there's no load.
A masked load with an allOnes mask means it's a normal vector load.
Differential Revision: http://reviews.llvm.org/D16691
llvm-svn: 259369
In the future, we will vectorize recurrences other than reductions. This patch
renames a few variables and updates their associated comments to enable them to
be reused for non-reduction PHI nodes.
This change was requested in the review for D16197.
llvm-svn: 259364
Loop transformations can sometimes fail because the loop, while in
valid rotated LCSSA form, is not in a canonical CFG form. This is
an extremely simple pass that just merges obviously redundant
blocks, which can be used to fix some known failure cases. In the
future, it may be enhanced with more cases (and have code shared with
SimplifyCFG).
This allows us to run LoopSimplifyCFG -> LoopRotate -> LoopUnroll,
so that SimplifyCFG cleans up the loop before Rotate tries to run.
Not currently used in the pass manager, since this pass doesn't do
anything unless you can hook it up in an LPM with other loop passes.
It'll be added once Chandler cleans up things to allow this.
Tested in a custom pipeline out of tree to confirm it works in
practice (in addition to the included trivial test).
llvm-svn: 259256
We would infinite loop because we created a shufflevector that was wider than
needed and then failed to combine that with the insertelement. When subsequently
visiting the extractelement from that shuffle, we see that it's unnecessary,
delete it, and trigger another visit to the insertelement.
llvm-svn: 259236
- Locally declare struct, and call it BaseDerivedPair
- Use a lambda to compare, instead of a singleton with uninitialized
fields
- Add a constructor to BaseDerivedPair and use SmallVector::emplace_back
llvm-svn: 259208
The full diff for the test directory may be hard to read because of the
filename clash; so here's all that happened as far as the tests are
concerned:
```
cd test/Transforms/RewriteStatepointsForGC
git rm *ll
git mv deopt-bundles/* ./
rmdir deopt-bundles
find . -name '*.ll' | xargs gsed -i 's/-rs4gc-use-deopt-bundles //g'
```
llvm-svn: 259129
These changes are aimed at bringing PlaceSafepoints up to code with the
LLVM coding guidelines:
- Fix variable naming
- Use DenseSet instead of std::set
- Remove dead code
- Minor local code simplifications
llvm-svn: 259112
Sanjoy Das