Implemented as many of Michael's suggestions as were possible:
* clang-format the added code while it is still fresh.
* tried to change Value* to Instruction* in many places in computeMinimumValueSizes - unfortunately there are several places where Constants need to be handled so this wasn't possible.
* Reduce the pass list on loop-vectorization-factors.ll.
* Fix a bug where we were querying MinBWs for I->getOperand(0) but using MinBWs[I].
llvm-svn: 252469
Summary:
LAA currently generates a set of SCEV predicates that must be checked by users.
In the case of Loop Distribute/Loop Load Elimination, no such predicates could have
been emitted, since we don't allow stride versioning. However, in the future there
could be SCEV predicates that will need to be checked.
This change adds support for SCEV predicate versioning in the Loop Distribute, Loop
Load Eliminate and the loop versioning infrastructure.
Reviewers: anemet
Subscribers: mssimpso, sanjoy, llvm-commits
Differential Revision: http://reviews.llvm.org/D14240
llvm-svn: 252467
Summary:
Teach the FunctionAttrs to do the right thing for IR with operand
bundles.
Reviewers: reames, chandlerc
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D14408
llvm-svn: 252387
Summary:
This change fixes an iterator wraparound bug in
`determinePointerReadAttrs`.
Ideally, ++'ing off the `end()` of an iplist should result in a failed
assert, but currently iplist seems to silently wrap to the head of the
list on `end()++`. This is why the bad behavior is difficult to
demonstrate.
Reviewers: chandlerc, reames
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D14350
llvm-svn: 252386
FoldPHIArgZextsIntoPHI cannot insert an instruction after the PHI if
there is an EHPad in the BB. Doing so would result in an instruction
inserted after a terminator.
llvm-svn: 252377
Some implicit ilist iterator conversions have crept back into Analysis,
Transforms, Hexagon, and llvm-stress. This removes them.
I'll commit a patch immediately after this to disallow them (in a
separate patch so that it's easy to revert if necessary).
llvm-svn: 252371
We tried to insert a cast of a phi in a block whose terminator is an
EHPad. This is invalid. Do not attempt the transform in these
circumstances.
llvm-svn: 252370
This marker prevents optimization passes from adding 'tail' or
'musttail' markers to a call. Is is used to prevent tail call
optimization from being performed on the call.
rdar://problem/22667622
Differential Revision: http://reviews.llvm.org/D12923
llvm-svn: 252368
The SLPVectorizer had a very crude way of trying to benefit
from associativity: it tried to optimize for splat/broadcast
or in order to have the same operator on the same side.
This is benefitial to the cost model and allows more vectorization
to occur.
This patch improve the logic and make the detection optimal (locally,
we don't look at the full tree but only at the immediate children).
Should fix https://llvm.org/bugs/show_bug.cgi?id=25247
Reviewers: mzolotukhin
Differential Revision: http://reviews.llvm.org/D13996
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 252337
Summary:
This change makes the `isImpliedCondition` interface similar to the rest
of the functions in ValueTracking (in that it takes a DataLayout,
AssumptionCache etc.). This is an NFC, intended to make a later diff
less noisy.
Depends on D14369
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D14391
llvm-svn: 252333
Previously, subprograms contained a metadata reference to the function they
described. Because most clients need to get or set a subprogram for a given
function rather than the other way around, this created unneeded inefficiency.
For example, many passes needed to call the function llvm::makeSubprogramMap()
to build a mapping from functions to subprograms, and the IR linker needed to
fix up function references in a way that caused quadratic complexity in the IR
linking phase of LTO.
This change reverses the direction of the edge by storing the subprogram as
function-level metadata and removing DISubprogram's function field.
Since this is an IR change, a bitcode upgrade has been provided.
Fixes PR23367. An upgrade script for textual IR for out-of-tree clients is
attached to the PR.
Differential Revision: http://reviews.llvm.org/D14265
llvm-svn: 252219
inalloca variables were not treated as static allocas, therefore didn't
participate in regular stack instrumentation. We don't want them to
participate in dynamic alloca instrumentation as well.
llvm-svn: 252213
We were correctly skipping dbginfo intrinsics and terminators, but the initial bailout wasn't, causing it to bail out on almost any block.
llvm-svn: 252152
Summary:
Remove the loop over the uses of the CallSite in ArgumentUsesTracker.
Since we have the `Use *` for actual argument operand, we can just use
pointer subtraction.
The time complexity remains the same though (except for a vararg
argument) -- `std::advance` is O(UseIndex) for the ArgumentList
iterator.
The real motivation is to make a later change adding support for operand
bundles simpler.
Reviewers: reames, chandlerc, nlewycky
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D14363
llvm-svn: 252141
With this change, instrumentation code and reader/write
code related to profile data structs are kept strictly
in-sync. THis will be extended to cfe and compile-rt
references as well.
Differential Revision: http://reviews.llvm.org/D13843
llvm-svn: 252113
We can often end up with conditional stores that cannot be speculated. They can come from fairly simple, idiomatic code:
if (c & flag1)
*a = x;
if (c & flag2)
*a = y;
...
There is no dominating or post-dominating store to a, so it is not legal to move the store unconditionally to the end of the sequence and cache the intermediate result in a register, as we would like to.
It is, however, legal to merge the stores together and do the store once:
tmp = undef;
if (c & flag1)
tmp = x;
if (c & flag2)
tmp = y;
if (c & flag1 || c & flag2)
*a = tmp;
The real power in this optimization is that it allows arbitrary length ladders such as these to be completely and trivially if-converted. The typical code I'd expect this to trigger on often uses binary-AND with constants as the condition (as in the above example), which means the ending condition can simply be truncated into a single binary-AND too: 'if (c & (flag1|flag2))'. As in the general case there are bitwise operators here, the ladder can often be optimized further too.
This optimization involves potentially increasing register pressure. Even in the simplest case, the lifetime of the first predicate is extended. This can be elided in some cases such as using binary-AND on constants, but not in the general case. Threading 'tmp' through all branches can also increase register pressure.
The optimization as in this patch is enabled by default but kept in a very conservative mode. It will only optimize if it thinks the resultant code should be if-convertable, and additionally if it can thread 'tmp' through at least one existing PHI, so it will only ever in the worst case create one more PHI and extend the lifetime of a predicate.
This doesn't trigger much in LNT, unfortunately, but it does trigger in a big way in a third party test suite.
llvm-svn: 252051
In my previous change to CVP (251606), I made CVP much more aggressive about trying to constant fold comparisons. This patch is a reversal in direction. Rather than being agressive about every compare, we restore the non-block local restriction for most, and then try hard for compares feeding returns.
The motivation for this is two fold:
* The more I thought about it, the less comfortable I got with the possible compile time impact of the other approach. There have been no reported issues, but after talking to a couple of folks, I've come to the conclusion the time probably isn't justified.
* It turns out we need to know the context to leverage the full power of LVI. In particular, asking about something at the end of it's block (the use of a compare in a return) will frequently get more precise results than something in the middle of a block. This is an implementation detail, but it's also hard to get around since mid-block queries have to reason about possible throwing instructions and don't get to use most of LVI's block focused infrastructure. This will become particular important when combined with http://reviews.llvm.org/D14263.
Differential Revision: http://reviews.llvm.org/D14271
llvm-svn: 252032
Summary:
The goal of this pass is to perform store-to-load forwarding across the
backedge of a loop. E.g.:
for (i)
A[i + 1] = A[i] + B[i]
=>
T = A[0]
for (i)
T = T + B[i]
A[i + 1] = T
The pass relies on loop dependence analysis via LoopAccessAnalisys to
find opportunities of loop-carried dependences with a distance of one
between a store and a load. Since it's using LoopAccessAnalysis, it was
easy to also add support for versioning away may-aliasing intervening
stores that would otherwise prevent this transformation.
This optimization is also performed by Load-PRE in GVN without the
option of multi-versioning. As was discussed with Daniel Berlin in
http://reviews.llvm.org/D9548, this is inferior to a more loop-aware
solution applied here. Hopefully, we will be able to remove some
complexity from GVN/MemorySSA as a consequence.
In the long run, we may want to extend this pass (or create a new one if
there is little overlap) to also eliminate loop-indepedent redundant
loads and store that *require* versioning due to may-aliasing
intervening stores/loads. I have some motivating cases for store
elimination. My plan right now is to wait for MemorySSA to come online
first rather than using memdep for this.
The main motiviation for this pass is the 456.hmmer loop in SPECint2006
where after distributing the original loop and vectorizing the top part,
we are left with the critical path exposed in the bottom loop. Being
able to promote the memory dependence into a register depedence (even
though the HW does perform store-to-load fowarding as well) results in a
major gain (~20%). This gain also transfers over to x86: it's
around 8-10%.
Right now the pass is off by default and can be enabled
with -enable-loop-load-elim. On the LNT testsuite, there are two
performance changes (negative number -> improvement):
1. -28% in Polybench/linear-algebra/solvers/dynprog: the length of the
critical paths is reduced
2. +2% in Polybench/stencils/adi: Unfortunately, I couldn't reproduce this
outside of LNT
The pass is scheduled after the loop vectorizer (which is after loop
distribution). The rational is to try to reuse LAA state, rather than
recomputing it. The order between LV and LLE is not critical because
normally LV does not touch scalar st->ld forwarding cases where
vectorizing would inhibit the CPU's st->ld forwarding to kick in.
LoopLoadElimination requires LAA to provide the full set of dependences
(including forward dependences). LAA is known to omit loop-independent
dependences in certain situations. The big comment before
removeDependencesFromMultipleStores explains why this should not occur
for the cases that we're interested in.
Reviewers: dberlin, hfinkel
Subscribers: junbuml, dberlin, mssimpso, rengolin, sanjoy, llvm-commits
Differential Revision: http://reviews.llvm.org/D13259
llvm-svn: 252017
Summary:
We now collect all types of dependences including lexically forward
deps not just "interesting" ones.
Reviewers: hfinkel
Subscribers: rengolin, llvm-commits
Differential Revision: http://reviews.llvm.org/D13256
llvm-svn: 251985
Commit 251839 triggers miscompiles on some bots:
http://lab.llvm.org:8011/builders/perf-x86_64-penryn-O3-polly-fast/builds/13723
(The commit is listed in 13722, but due to an existing failure introduced in
13721 and reverted in 13723 the failure is only visible in 13723)
To verify r251839 is indeed the only change that triggered the buildbot failures
and to ensure the buildbots remain green while investigating I temporarily
revert this commit. At the current state it is unclear if this commit introduced
some miscompile or if it only exposed code to Polly that is subsequently
miscompiled by Polly.
llvm-svn: 251901
To be able to maximize the bandwidth during vectorization, this patch provides a new flag vectorizer-maximize-bandwidth. When it is turned on, the vectorizer will determine the vectorization factor (VF) using the smallest instead of widest type in the loop. To avoid increasing register pressure too much, estimates of the register usage for different VFs are calculated so that we only choose a VF when its register usage doesn't exceed the number of available registers.
This is the second attempt to submit this patch. The first attempt got a test failure on ARM. This patch is updated to try to fix the failure (more specifically, by handling the case when VF=1).
Differential revision: http://reviews.llvm.org/D8943
llvm-svn: 251850
This reverts commit r251837, due to a number of bot failures of the form:
/home/grosser/buildslave/perf-x86_64-penryn-O3-polly-fast/llvm.obj/tools/llvm-link/Release+Asserts/llvm-link.o:llvm-link.cpp:function
loadIndex(llvm::LLVMContext&, llvm::Module const*): error: undefined
reference to
'llvm::object::FunctionIndexObjectFile::create(llvm::MemoryBufferRef,
llvm::LLVMContext&, llvm::Module const*, bool)'
/home/grosser/buildslave/perf-x86_64-penryn-O3-polly-fast/llvm.obj/tools/llvm-link/Release+Asserts/llvm-link.o:llvm-link.cpp:function
loadIndex(llvm::LLVMContext&, llvm::Module const*): error: undefined
reference to 'llvm::object::FunctionIndexObjectFile::takeIndex()'
I'm not sure why these are happening - I added Object to the requred
libraries in tools/llvm-link/LLVMBuild.txt and the LLVM_LINK_COMPONENTS
in tools/llvm-link/CMakeLists.txt. Confirmed for my build that these
symbols come out of libLLVMObject.a. What am I missing?
llvm-svn: 251841
Summary:
This patch adds support to check if a loop has loop invariant conditions which lead to loop exits. If so, we know that if the exit path is taken, it is at the first loop iteration. If there is an induction variable used in that exit path whose value has not been updated, it will keep its initial value passing from loop preheader. We can therefore rewrite the exit value with
its initial value. This will help remove phis created by LCSSA and enable other optimizations like loop unswitch.
Reviewers: sanjoy
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D13974
llvm-svn: 251839
Summary:
Support for necessary linkage changes and symbol renaming during
ThinLTO function importing.
Also includes llvm-link support for manually importing functions
and associated llvm-link based tests.
Note that this does not include support for intelligently importing
metadata, which is currently imported duplicate times. That support will
be in the follow-on patch, and currently is ignored by the tests.
Reviewers: dexonsmith, joker.eph, davidxl
Subscribers: tobiasvk, tejohnson, llvm-commits
Differential Revision: http://reviews.llvm.org/D13515
llvm-svn: 251837
Summary:
SCEV Predicates represent conditions that typically cannot be derived from
static analysis, but can be used to reduce SCEV expressions to forms which are
usable for different optimizers.
ScalarEvolution now has the rewriteUsingPredicate method which can simplify a
SCEV expression using a SCEVPredicateSet. The normal workflow of a pass using
SCEVPredicates would be to hold a SCEVPredicateSet and every time assumptions
need to be made a new SCEV Predicate would be created and added to the set.
Each time after calling getSCEV, the user will call the rewriteUsingPredicate
method.
We add two types of predicates
SCEVPredicateSet - implements a set of predicates
SCEVEqualPredicate - tests for equality between two SCEV expressions
We use the SCEVEqualPredicate to re-implement stride versioning. Every time we
version a stride, we will add a SCEVEqualPredicate to the context.
Instead of adding specific stride checks, LoopVectorize now adds a more
generic SCEV check.
We only need to add support for this in the LoopVectorizer since this is the
only pass that will do stride versioning.
Reviewers: mzolotukhin, anemet, hfinkel, sanjoy
Subscribers: sanjoy, hfinkel, rengolin, jmolloy, llvm-commits
Differential Revision: http://reviews.llvm.org/D13595
llvm-svn: 251800
The initial coverage checking code for sample records failed to count
records inside inlined profiles. This change fixes the oversight.
llvm-svn: 251752
from its pass harness by providing a lambda to query for AA results.
This allows the legacy pass to easily provide a lambda that uses the
special helpers to construct function AA results from a legacy CGSCC
pass. With the new pass manager (the next patch) the lambda just
directly wraps the intuitive query API.
llvm-svn: 251715
Update the discriminator assignment algorithm
* If a scope has already been assigned a discriminator, do not reassign a nested discriminator for it.
* If the file and line both match, even if the column does not match, we should assign a new discriminator for the stmt.
original code:
; #1 int foo(int i) {
; #2 if (i == 3 || i == 5) return 100; else return 99;
; #3 }
; i == 3: discriminator 0
; i == 5: discriminator 2
; return 100: discriminator 1
; return 99: discriminator 3
llvm-svn: 251689
Update the discriminator assignment algorithm
* If a scope has already been assigned a discriminator, do not reassign a nested discriminator for it.
* If the file and line both match, even if the column does not match, we should assign a new discriminator for the stmt.
original code:
; #1 int foo(int i) {
; #2 if (i == 3 || i == 5) return 100; else return 99;
; #3 }
; i == 3: discriminator 0
; i == 5: discriminator 2
; return 100: discriminator 1
; return 99: discriminator 3
llvm-svn: 251685
* If a scope has already been assigned a discriminator, do not reassign a nested discriminator for it.
* If the file and line both match, even if the column does not match, we should assign a new discriminator for the stmt.
original code:
; #1 int foo(int i) {
; #2 if (i == 3 || i == 5) return 100; else return 99;
; #3 }
; i == 3: discriminator 0
; i == 5: discriminator 2
; return 100: discriminator 1
; return 99: discriminator 3
llvm-svn: 251680
transformations in FunctionAttrs rather than building a new one each
time.
This isn't trivial because there are different heuristics from different
passes for exactly what set they want. The primary difference is whether
an *overridable* function completely disables the synthesis of
attributes. I've modeled this by directly testing for overridable, and
using the common set that excludes external and opt-none functions.
This does cause some changes by disabling more optimizations in the face
of opt-none. Specifically, we were still optimizing *calls* to opt-none
functions based on their attributes, just not the bodies. It seems
better to be conservative on both fronts given the intended semanticas
here (best effort to not assume or disturb anything). I've not tried to
test this change as it seems complex, brittle, and not important to the
implicit contract of opt-none. Instead, it seems more like a choice that
should be dictated by the simplified implementation and the change to be
acceptable differences within the space of opt-none.
A big benefit here is that these transformations no longer rely on the
legacy pass manager's SCC types, they just work on generic sets of
function pointers. This will make it easy to re-use their logic in the
new pass manager.
I've also made the transforms static functions instead of members where
trivial while I was touching the signatures.
llvm-svn: 251640
This patch unify the 39-bit and 42-bit mapping for aarch64 to use only
one instrumentation algorithm. This removes compiler flag
SANITIZER_AARCH64_VMA requirement for MSAN on aarch64.
The mapping to use now is for 39 and 42-bits:
0x00000000000ULL-0x01000000000ULL MappingDesc::INVALID
0x01000000000ULL-0x02000000000ULL MappingDesc::SHADOW
0x02000000000ULL-0x03000000000ULL MappingDesc::ORIGIN
0x03000000000ULL-0x04000000000ULL MappingDesc::SHADOW
0x04000000000ULL-0x05000000000ULL MappingDesc::ORIGIN
0x05000000000ULL-0x06000000000ULL MappingDesc::APP
0x06000000000ULL-0x07000000000ULL MappingDesc::INVALID
0x07000000000ULL-0x08000000000ULL MappingDesc::APP
And only for 42-bits:
0x08000000000ULL-0x09000000000ULL MappingDesc::INVALID
0x09000000000ULL-0x0A000000000ULL MappingDesc::SHADOW
0x0A000000000ULL-0x0B000000000ULL MappingDesc::ORIGIN
0x0B000000000ULL-0x0F000000000ULL MappingDesc::INVALID
0x0F000000000ULL-0x10000000000ULL MappingDesc::APP
0x10000000000ULL-0x11000000000ULL MappingDesc::INVALID
0x11000000000ULL-0x12000000000ULL MappingDesc::APP
0x12000000000ULL-0x17000000000ULL MappingDesc::INVALID
0x17000000000ULL-0x18000000000ULL MappingDesc::SHADOW
0x18000000000ULL-0x19000000000ULL MappingDesc::ORIGIN
0x19000000000ULL-0x20000000000ULL MappingDesc::INVALID
0x20000000000ULL-0x21000000000ULL MappingDesc::APP
0x21000000000ULL-0x26000000000ULL MappingDesc::INVALID
0x26000000000ULL-0x27000000000ULL MappingDesc::SHADOW
0x27000000000ULL-0x28000000000ULL MappingDesc::ORIGIN
0x28000000000ULL-0x29000000000ULL MappingDesc::SHADOW
0x29000000000ULL-0x2A000000000ULL MappingDesc::ORIGIN
0x2A000000000ULL-0x2B000000000ULL MappingDesc::APP
0x2B000000000ULL-0x2C000000000ULL MappingDesc::INVALID
0x2C000000000ULL-0x2D000000000ULL MappingDesc::SHADOW
0x2D000000000ULL-0x2E000000000ULL MappingDesc::ORIGIN
0x2E000000000ULL-0x2F000000000ULL MappingDesc::APP
0x2F000000000ULL-0x39000000000ULL MappingDesc::INVALID
0x39000000000ULL-0x3A000000000ULL MappingDesc::SHADOW
0x3A000000000ULL-0x3B000000000ULL MappingDesc::ORIGIN
0x3B000000000ULL-0x3C000000000ULL MappingDesc::APP
0x3C000000000ULL-0x3D000000000ULL MappingDesc::INVALID
0x3D000000000ULL-0x3E000000000ULL MappingDesc::SHADOW
0x3E000000000ULL-0x3F000000000ULL MappingDesc::ORIGIN
0x3F000000000ULL-0x40000000000ULL MappingDesc::APP
And although complex it provides a better memory utilization that
previous one.
llvm-svn: 251624
Clang driver now injects -u<hook_var> flag in the linker
command line, in which case user function is not needed
any more.
Differential Revision: http://reviews.llvm.org/D14033
llvm-svn: 251612
Somewhat shockingly for an analysis pass which is computing constant ranges, LVI did not understand the ranges provided by range metadata.
As part of this change, I included a change to CVP primarily because doing so made it much easier to write small self contained test cases. CVP was previously only handling the non-local operand case, but given that LVI can sometimes figure out information about instructions standalone, I don't see any reason to restrict this. There could possibly be a compile time impact from this, but I suspect it should be minimal. If anyone has an example which substaintially regresses, please let me know. I could restrict the block local handling to ICmps feeding Terminator instructions if needed.
Note that this patch continues a somewhat bad practice in LVI. In many cases, we know facts about values, and separate context sensitive facts about values. LVI makes no effort to distinguish and will frequently cache the same value fact repeatedly for different contexts. I would like to change this, but that's a large enough change that I want it to go in separately with clear documentation of what's changing. Other examples of this include the non-null handling, and arguments.
As a meta comment: the entire motivation of this change was being able to write smaller (aka reasonable sized) test cases for a future patch teaching LVI about select instructions.
Differential Revision: http://reviews.llvm.org/D13543
llvm-svn: 251606
The most common use case is when eliminating redundant range checks in an example like the following:
c = a[i+1] + a[i];
Note that all the smarts of the transform (the implication engine) is already in ValueTracking and is tested directly through InstructionSimplify.
Differential Revision: http://reviews.llvm.org/D13040
llvm-svn: 251596
To be able to maximize the bandwidth during vectorization, this patch provides a new flag vectorizer-maximize-bandwidth. When it is turned on, the vectorizer will determine the vectorization factor (VF) using the smallest instead of widest type in the loop. To avoid increasing register pressure too much, estimates of the register usage for different VFs are calculated so that we only choose a VF when its register usage doesn't exceed the number of available registers.
llvm-svn: 251592
This adds the flag -mllvm -sample-profile-check-coverage=N to the
SampleProfile pass. N is the percent of input sample records that the
user expects to apply. If the pass does not use N% (or more) of the
sample records in the input, it emits a warning.
This is useful to detect some forms of stale profiles. If the code has
drifted enough from the original profile, there will be records that do
not match the IR anymore.
This will not detect cases where a sample profile record for line L is
referring to some other instructions that also used to be at line L.
llvm-svn: 251568
Summary:
If P branches to Q conditional on C and Q branches to R conditional on
C' and C => C' then the branch conditional on C' can be folded to an
unconditional branch.
Reviewers: reames
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D13972
llvm-svn: 251557
The pass was keeping around a lot of per-function data (visited blocks,
edges, dominance, etc) that is just taking up memory for no reason. In
fact, from function to function it could potentially confuse the
propagator since some maps are indexed by line offsets which can be
common between functions.
llvm-svn: 251531
I think these were affected by a change way back when to stop printing newlines in Value::dump() by default. This change simply allows the debug output to be readable.
NFC.
llvm-svn: 251517
Summary:
This patch adds support to check if a loop has loop invariant conditions which lead to loop exits. If so, we know that if the exit path is taken, it is at the first loop iteration. If there is an induction variable used in that exit path whose value has not been updated, it will keep its initial value passing from loop preheader. We can therefore rewrite the exit value with
its initial value. This will help remove phis created by LCSSA and enable other optimizations like loop unswitch.
Reviewers: sanjoy
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D13974
llvm-svn: 251492
CatchReturnInst has side-effects: it runs a destructor. This destructor
could conceivably run forever/call exit/etc. and should not be removed.
llvm-svn: 251461
Summary:
This change could be way off-piste, I'm looking for any feedback on whether it's an acceptable approach.
It never seems to be a problem to gobble up as many reduction values as can be found, and then to attempt to reduce the resulting tree. Some of the workloads I'm looking at have been aggressively unrolled by hand, and by selecting reduction widths that are not constrained by a vector register size, it becomes possible to profitably vectorize. My test case shows such an unrolling which SLP was not vectorizing (on neither ARM nor X86) before this patch, but with it does vectorize.
I measure no significant compile time impact of this change when combined with D13949 and D14063. There are also no significant performance regressions on ARM/AArch64 in SPEC or LNT.
The more principled approach I thought of was to generate several candidate tree's and use the cost model to pick the cheapest one. That seemed like quite a big design change (the algorithms seem very much one-shot), and would likely be a costly thing for compile time. This seemed to do the job at very little cost, but I'm worried I've misunderstood something!
Reviewers: nadav, jmolloy
Subscribers: mssimpso, llvm-commits, aemerson
Differential Revision: http://reviews.llvm.org/D14116
llvm-svn: 251428
Summary:
Currently, when the SLP vectorizer considers whether a phi is part of a reduction, it dismisses phi's whose incoming blocks are not the same as the block containing the phi. For the patterns I'm looking at, extending this rule to allow phis whose incoming block is a containing loop latch allows me to vectorize certain workloads.
There is no significant compile-time impact, and combined with D13949, no performance improvement measured in ARM/AArch64 in any of SPEC2000, SPEC2006 or LNT.
Reviewers: jmolloy, mcrosier, nadav
Subscribers: mssimpso, nadav, aemerson, llvm-commits
Differential Revision: http://reviews.llvm.org/D14063
llvm-svn: 251425
Summary:
Certain workloads, in particular sum-of-absdiff loops, can be vectorized using SLP if it can treat select instructions as reduction values.
The test case is a bit awkward. The AArch64 cost model needs some tuning to not be so pessimistic about selects. I've had to tweak the SLP threshold here.
Reviewers: jmolloy, mzolotukhin, spatel, nadav
Subscribers: nadav, mssimpso, aemerson, llvm-commits
Differential Revision: http://reviews.llvm.org/D13949
llvm-svn: 251424
When emitting a remark for a conditional branch annotation, the remark
uses the line location information of the conditional branch in the
message. In some cases, that information is unavailable and the
optimization would segfaul. I'm still not sure whether this is a bug or
WAI, but the optimizer should not die because of this.
llvm-svn: 251420
No functionality changed here, but the indentation is substantially
reduced and IMO the code is much easier to read. I've also added some
helpful comments.
This is just a clean-up I wrote while studying the code, and that has
been in my backlog for a while.
llvm-svn: 251381
We should remove noalias along with dereference and dereference_or_null attributes
because statepoint could potentially touch the entire heap including noalias objects.
Differential Revision: http://reviews.llvm.org/D14032
llvm-svn: 251333
This adds a couple of optimization remarks to the SamplePGO
transformation. When it decides to inline a hot function (to mimic the
inline stack and repeat useful inline decisions in the original build).
It will also report branch destinations. For instance, given the code
fragment:
6 if (i < 1000)
7 sum -= i;
8 else
9 sum += -i * rand();
If the 'else' branch is taken most of the time, building this code with
-Rpass=sample-profile will produce:
a.cc:9:14: remark: most popular destination for conditional branches at small.cc:6:9 [-Rpass=sample-profile]
sum += -i * rand();
^
llvm-svn: 251330
Android libc provides a fixed TLS slot for the unsafe stack pointer,
and this change implements direct access to that slot on AArch64 via
__builtin_thread_pointer() + offset.
This change also moves more code into TargetLowering and its
target-specific subclasses to get rid of target-specific codegen
in SafeStackPass.
This change does not touch the ARM backend because ARM lowers
builting_thread_pointer as aeabi_read_tp, which is not available
on Android.
The previous iteration of this change was reverted in r250461. This
version leaves the generic, compiler-rt based implementation in
SafeStack.cpp instead of moving it to TargetLoweringBase in order to
allow testing without a TargetMachine.
llvm-svn: 251324
Vectorization of memory instruction (Load/Store) is possible when the pointer is coming from GEP. The GEP analysis allows to estimate the profit.
In some cases we have a "bitcast" between GEP and memory instruction.
I added code that skips the "bitcast".
http://reviews.llvm.org/D13886
llvm-svn: 251291
Summary:
InstCombine tries to transform GEP(PHI(GEP1, GEP2, ..)) into GEP(GEP(PHI(...))
when possible. However, this may leave the old PHI node around. Even if we
do end up folding the GEPs, having an extra PHI node might not be beneficial.
This change makes the transformation more conservative. We now only do this if
the PHI has only one use, and can therefore be removed after the transformation.
Reviewers: jmolloy, majnemer
Subscribers: mcrosier, mssimpso, llvm-commits
Differential Revision: http://reviews.llvm.org/D13887
llvm-svn: 251281
First, the motivation: LLVM currently does not realize that:
((2072 >> (L == 0)) >> 7) & 1 == 0
where L is some arbitrary value. Whether you right-shift 2072 by 7 or by 8, the
lowest-order bit is always zero. There are obviously several ways to go about
fixing this, but the generic solution pursued in this patch is to teach
computeKnownBits something about shifts by a non-constant amount. Previously,
we would give up completely on these. Instead, in cases where we know something
about the low-order bits of the shift-amount operand, we can combine (and
together) the associated restrictions for all shift amounts consistent with
that knowledge. As a further generalization, I refactored all of the logic for
all three kinds of shifts to have this capability. This works well in the above
case, for example, because the dynamic shift amount can only be 0 or 1, and
thus we can say a lot about the known bits of the result.
This brings us to the second part of this change: Even when we know all of the
bits of a value via computeKnownBits, nothing used to constant-fold the result.
This introduces the necessary code into InstCombine and InstSimplify. I've
added it into both because:
1. InstCombine won't automatically pick up the associated logic in
InstSimplify (InstCombine uses InstSimplify, but not via the API that
passes in the original instruction).
2. Putting the logic in InstCombine allows the resulting simplifications to become
part of the iterative worklist
3. Putting the logic in InstSimplify allows the resulting simplifications to be
used by everywhere else that calls SimplifyInstruction (inlining, unrolling,
and many others).
And this requires a small change to our definition of an ephemeral value so
that we don't break the rest case from r246696 (where the icmp feeding the
@llvm.assume, is also feeding a br). Under the old definition, the icmp would
not be considered ephemeral (because it is used by the br), but this causes the
assume to remove itself (in addition to simplifying the branch structure), and
it seems more-useful to prevent that from happening.
llvm-svn: 251146
After some look-ahead PRE was added for GEPs, an instruction could end
up in the table of candidates before it was actually inspected. When
this happened the pass might decide it was the best candidate to
replace itself. This didn't go well.
Should fix PR25291
llvm-svn: 251145
Summary:
This change teaches the LLVM inliner to not inline through callsites
with unknown operand bundles. Currently all operand bundles are
"unknown" operand bundles but in the near future we will add support for
inlining through some select kinds of operand bundles.
Reviewers: reames, chandlerc, majnemer
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D14001
llvm-svn: 251141
This patch converts the remaining references to literal
strings for names of profile runtime entites (such as
profile runtime hook, runtime hook use function, profile
init method, register function etc).
Also added documentation for all the new interfaces.
llvm-svn: 251093
The insertLoop() API is only used to add new loops, and has confusing
ownership semantics. Simplify it by replacing it with addLoop().
llvm-svn: 251064
Summary: Currently SimplifyResume can convert an invoke instruction to a call instruction if its landing pad is trivial. In practice we could have several invoke instructions with trivial landing pads and share a common rethrow block, and in the common rethrow block, all the landing pads join to a phi node. The patch extends SimplifyResume to check the phi of landing pad and their incoming blocks. If any of them is trivial, remove it from the phi node and convert the invoke instruction to a call instruction.
Reviewers: hfinkel, reames
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D13718
llvm-svn: 251061
This is a clean up patch that defines instr prof section and variable
name prefixes in a common header with access helper functions.
clang FE change will be done as a follow up once this patch is in.
Differential Revision: http://reviews.llvm.org/D13919
llvm-svn: 251058
As an invariant, BasicBlocks cannot be empty when passed to a transform.
This is not the case for MachineBasicBlocks and the Sink pass was ported
from the MachineSink pass which would explain the check's existence.
llvm-svn: 251057
* Don't instrument promotable dynamic allocas:
We already have a test that checks that promotable dynamic allocas are
ignored, as well as static promotable allocas. Make sure this test will
still pass if/when we enable dynamic alloca instrumentation by default.
* Handle lifetime intrinsics before handling dynamic allocas:
lifetime intrinsics may refer to dynamic allocas, so we need to emit
instrumentation before these dynamic allocas would be replaced.
Differential Revision: http://reviews.llvm.org/D12704
llvm-svn: 251045
SimplifyTerminatorOnSelect didn't consider the possibility that the
condition might be related to one of PHI nodes.
This fixes PR25267.
llvm-svn: 250922
In some cases (as illustrated in the unittest), lineno can be less than the heade_lineno because the function body are included from some other files. In this case, offset will be negative. This patch makes clang still able to match the profile to IR in this situation.
http://reviews.llvm.org/D13914
llvm-svn: 250873
It is now possible to infer intrinsic modref behaviour purely from intrinsic attributes.
This change will allow to completely remove GET_INTRINSIC_MODREF_BEHAVIOR table.
Differential Revision: http://reviews.llvm.org/D13907
llvm-svn: 250860
Summary: In r231241, TargetLibraryInfoWrapperPass was added to
`getAnalysisUsage` for `AddressSanitizer`, but the corresponding
`INITIALIZE_PASS_DEPENDENCY` was not added.
Reviewers: dvyukov, chandlerc, kcc
Subscribers: kcc, llvm-commits
Differential Revision: http://reviews.llvm.org/D13629
llvm-svn: 250813
`normalizeForInvokeSafepoint` in RewriteStatepointsForGC.cpp, as it is
written today, deals with `gc.relocate` and `gc.result` uses of a
statepoint equally well. This change documents this fact and adds a
test case.
There is no functional change here -- only documentation of existing
functionality.
llvm-svn: 250784
Allow LLVM to optimize the sequence like the following:
%inc = add nsw i32 %i, 1
%cmp = icmp slt %n, %inc
into:
%cmp = icmp sle i32 %n, %i
The case is not handled previously due to the complexity of compuation of %n.
Hence, LLVM cannot swap operands of icmp accordingly.
llvm-svn: 250746
Besides the usual, I finally added an overload to
`BasicBlock::splitBasicBlock()` that accepts an `Instruction*` instead
of `BasicBlock::iterator`. Someone can go back and remove this overload
later (after updating the callers I'm going to skip going forward), but
the most common call seems to be
`BB->splitBasicBlock(BB->getTerminator(), ...)` and I'm not sure it's
better to add `->getIterator()` to every one than have the overload.
It's pretty hard to get the usage wrong.
llvm-svn: 250745
Originally I planned to use the same interface for masked gather/scatter and set isConsecutive to "false" in this case.
Now I'm implementing masked gather/scatter and see that the interface is inconvenient. I want to add interfaces isLegalMaskedGather() / isLegalMaskedScatter() instead of using the "Consecutive" parameter in the existing interfaces.
Differential Revision: http://reviews.llvm.org/D13850
llvm-svn: 250686
This patch improves support for combining the SSE4A EXTRQ(I) and INSERTQ(I) intrinsics:
1 - Converts INSERTQ/EXTRQ calls to INSERTQI/EXTRQI if the 'bit index' and 'length' operands are constant
2 - Converts INSERTQI/EXTRQI calls to shufflevector if the bit index/length are both byte aligned (we can already lower shuffles to INSERTQI/EXTRQI if its useful)
3 - Constant folding support
4 - Add zeroinitializer handling
Differential Revision: http://reviews.llvm.org/D13348
llvm-svn: 250609
The `"statepoint-id"` and `"statepoint-num-patch-bytes"` attributes are
used solely to determine properties of the `gc.statepoint` being
created. Once the `gc.statepoint` is in place, these should be removed.
llvm-svn: 250491
Summary:
This is a step towards using operand bundles to carry deopt state till
RewriteStatepointsForGC. The change adds a flag to
RewriteStatepointsForGC that teaches it to pick up deopt state from a
`"deopt"` operand bundle attached to the `call` or `invoke` it is
wrapping.
The command line flag added, `-rs4gc-use-deopt-bundles`, will only exist
for a short while. Once we are able to pipe deopt bundle state through
the full optimization pipeline without problems, we will "constant fold"
`-rs4gc-use-deopt-bundles` to `true`.
Reviewers: swaroop.sridhar, reames
Subscribers: llvm-commits, sanjoy
Differential Revision: http://reviews.llvm.org/D13372
llvm-svn: 250489
Summary:
`cloneArithmeticIVUser` currently trips over expression like `add %iv,
-1` when `%iv` is being zero extended -- it tries to construct the
widened use as `add %iv.zext, zext(-1)` and (correctly) fails to prove
equivalence to `zext(add %iv, -1)` (here the SCEV for `%iv` is
`{1,+,1}`).
This change teaches `IndVars` to try sign extending the non-IV operand
if that makes the newly constructed IV use equivalent to the widened
narrow IV use.
Reviewers: atrick, hfinkel, reames
Subscribers: sanjoy, llvm-commits
Differential Revision: http://reviews.llvm.org/D13717
llvm-svn: 250483
Summary:
This NFC splitting is intended to make a later diff easier to follow.
It just tail duplicates `cloneIVUser` into `cloneArithmeticIVUser` and
`cloneBitwiseIVUser`.
Reviewers: atrick, hfinkel, reames
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D13716
llvm-svn: 250481
Android libc provides a fixed TLS slot for the unsafe stack pointer,
and this change implements direct access to that slot on AArch64 via
__builtin_thread_pointer() + offset.
This change also moves more code into TargetLowering and its
target-specific subclasses to get rid of target-specific codegen
in SafeStackPass.
This change does not touch the ARM backend because ARM lowers
builting_thread_pointer as aeabi_read_tp, which is not available
on Android.
llvm-svn: 250456
Turns out this approach is buggy. In discussion about follow on work, Sanjoy pointed out that we could be subject to circular logic problems.
Consider:
if (i u< L) leave()
if ((i + 1) u< L) leave()
print(a[i] + a[i+1])
If we know that L is less than UINT_MAX, we could possible prove (in a control dependent way) that i + 1 does not overflow. This gives us:
if (i u< L) leave()
if ((i +nuw 1) u< L) leave()
print(a[i] + a[i+1])
If we now do the transform this patch proposed, we end up with:
if ((i +nuw 1) u< L) leave_appropriately()
print(a[i] + a[i+1])
That would be a miscompile when i==-1. The problem here is that the control dependent nuw bits got used to prove something about the first condition. That's obviously invalid.
This won't happen today, but since I plan to enhance LVI/CVP with exactly that transform at some point in the not too distant future...
llvm-svn: 250430
This adjusts all integers in the reader/writer to reflect the types
stored on profile files. They should all be unsigned 32-bit or 64-bit
values. Changed all associated internal types to be uint32_t or
uint64_t.
The only place that needed some adjustments is in the sample profile
transformation. Altough the weight read from the profile are 64-bit
values, the internal API for branch weights only accepts 32-bit values.
The pass now saturates weights that overflow uint32_t.
llvm-svn: 250427
With r250345 and r250343, we start to observe the following failure
when bootstrap clang with lto and pgo:
PHI node entries do not match predecessors!
%.sroa.029.3.i = phi %"class.llvm::SDNode.13298"* [ null, %30953 ], [ null, %31017 ], [ null, %30998 ], [ null, %_ZN4llvm8dyn_castINS_14ConstantSDNodeENS_7SDValueEEENS_10cast_rettyIT_T0_E8ret_typeERS5_.exit.i.1804 ], [ null, %30975 ], [ null, %30991 ], [ null, %_ZNK4llvm3EVT13getScalarTypeEv.exit.i.1812 ], [ %..sroa.029.0.i, %_ZN4llvm11SmallVectorIiLj8EED1Ev.exit.i.1826 ], !dbg !451895
label %30998
label %_ZNK4llvm3EVTeqES0_.exit19.thread.i
LLVM ERROR: Broken function found, compilation aborted!
I will re-commit this if the bot does not recover.
llvm-svn: 250366
Currently in JumpThreading pass, the branch weight metadata is not updated after CFG modification. Consider the jump threading on PredBB, BB, and SuccBB. After jump threading, the weight on BB->SuccBB should be adjusted as some of it is contributed by the edge PredBB->BB, which doesn't exist anymore. This patch tries to update the edge weight in metadata on BB->SuccBB by scaling it by 1 - Freq(PredBB->BB) / Freq(BB->SuccBB).
This is the third attempt to submit this patch, while the first two led to failures in some FDO tests. After investigation, it is the edge weight normalization that caused those failures. In this patch the edge weight normalization is fixed so that there is no zero weight in the output and the sum of all weights can fit in 32-bit integer. Several unit tests are added.
Differential revision: http://reviews.llvm.org/D10979
llvm-svn: 250345
If we have a series of branches which are all unlikely to fail, we can possibly combine them into a single check on the fastpath combined with a bit of dispatch logic on the slowpath. We don't want to do this unconditionally since it requires speculating instructions past a branch, but if the profiling metadata on the branch indicates profitability, this can reduce the number of checks needed along the fast path.
The canonical example this is trying to handle is removing the second bounds check implied by the Java code: a[i] + a[i+1]. Note that it can currently only do so for really simple conditions and the values of a[i] can't be used anywhere except in the addition. (i.e. the load has to have been sunk already and not prevent speculation.) I plan on extending this transform over the next few days to handle alternate sequences.
Differential Revision: http://reviews.llvm.org/D13070
llvm-svn: 250343
We forgot to append the terminatepad's arguments which resulted in us
treating the old terminatepad as an argument to the new terminatepad
causing us to crash immediately. Instead, add the old terminatepad's
arguments to the new terminatepad.
This fixes PR25155.
llvm-svn: 250234
Remove remaining `ilist_iterator` implicit conversions from
LLVMScalarOpts.
This change exposed some scary behaviour in
lib/Transforms/Scalar/SCCP.cpp around line 1770. This patch changes a
call from `Function::begin()` to `&Function::front()`, since the return
was immediately being passed into another function that takes a
`Function*`. `Function::front()` started to assert, since the function
was empty. Note that `Function::end()` does not point at a legal
`Function*` -- it points at an `ilist_half_node` -- so the other
function was getting garbage before. (I added the missing check for
`Function::isDeclaration()`.)
Otherwise, no functionality change intended.
llvm-svn: 250211
Currently in JumpThreading pass, the branch weight metadata is not updated after CFG modification. Consider the jump threading on PredBB, BB, and SuccBB. After jump threading, the weight on BB->SuccBB should be adjusted as some of it is contributed by the edge PredBB->BB, which doesn't exist anymore. This patch tries to update the edge weight in metadata on BB->SuccBB by scaling it by 1 - Freq(PredBB->BB) / Freq(BB->SuccBB).
Differential revision: http://reviews.llvm.org/D10979
llvm-svn: 250204
On Linux, the profile runtime can use __start_SECTNAME and __stop_SECTNAME
symbols defined by the linker to locate the start and end location of
a named section (with C name). This eliminates the need for instrumented
binary to call __llvm_profile_register_function during start-up time.
llvm-svn: 250199
As discussed in D13348 - the INSERTQI range combining code is wrong in that it confuses the insertion bit index with an extraction bit index.
The remaining legal combines are very unlikely (especially once we've converted to shuffles in D13348) so I'm removing the optimization.
llvm-svn: 250160
Now that all the known faults with GlobalsAA have been fixed, flip the big switch on -enable-non-lto-gmr again.
Feel free to pester me with any more bugs found, and don't hesitate to flip the switch back off.
llvm-svn: 250157
Continuing the work from last week to remove implicit ilist iterator
conversions. First related commit was probably r249767, with some more
motivation in r249925. This edition gets LLVMTransformUtils compiling
without the implicit conversions.
No functional change intended.
llvm-svn: 250142
In JumpThreading pass, the branch weight metadata is not updated after CFG modification. Consider the jump threading on PredBB, BB, and SuccBB. After jump threading, the weight on BB->SuccBB should be adjusted as some of it is contributed by the edge PredBB->BB, which doesn't exist anymore. This patch tries to update the edge weight in metadata on BB->SuccBB by scaling it by 1 - Freq(PredBB->BB) / Freq(BB->SuccBB).
Differential revision: http://reviews.llvm.org/D10979
llvm-svn: 250089
GlobalOpt currently merges stores into the initialisers of internal,
externally_initialized globals, but should not do so as the value of the global
may change between the initialiser and any code in the module being run.
llvm-svn: 250035
C semantics force sub-int-sized values (e.g. i8, i16) to be promoted to int
type (e.g. i32) whenever arithmetic is performed on them.
For targets with native i8 or i16 operations, usually InstCombine can shrink
the arithmetic type down again. However InstCombine refuses to create illegal
types, so for targets without i8 or i16 registers, the lengthening and
shrinking remains.
Most SIMD ISAs (e.g. NEON) however support vectors of i8 or i16 even when
their scalar equivalents do not, so during vectorization it is important to
remove these lengthens and truncates when deciding the profitability of
vectorization.
The algorithm this uses starts at truncs and icmps, trawling their use-def
chains until they terminate or instructions outside the loop are found (or
unsafe instructions like inttoptr casts are found). If the use-def chains
starting from different root instructions (truncs/icmps) meet, they are
unioned. The demanded bits of each node in the graph are ORed together to form
an overall mask of the demanded bits in the entire graph. The minimum bitwidth
that graph can be truncated to is the bitwidth minus the number of leading
zeroes in the overall mask.
The intention is that this algorithm should "first do no harm", so it will
never insert extra cast instructions. This is why the use-def graphs are
unioned, so that subgraphs with different minimum bitwidths do not need casts
inserted between them.
This algorithm works hard to reduce compile time impact. DemandedBits are only
queried if there are extends of illegal types and if a truncate to an illegal
type is seen. In the general case, this results in a simple linear scan of the
instructions in the loop.
No non-noise compile time impact was seen on a clang bootstrap build.
llvm-svn: 250032
Doing so could cause the post-unswitching convergent ops to be
control-dependent on the unswitch condition where they were not before.
This check could be refined to allow unswitching where the convergent
operation was already control-dependent on the unswitch condition.
llvm-svn: 249874
This covers the common case of operations that cannot be sunk.
Operations that cannot be hoisted should already be handled properly via
the safe-to-speculate rules and mechanisms.
llvm-svn: 249865
http://reviews.llvm.org/D13576
As we are using hierarchical profile, there is no need to keep HeaderLineno a member variable. This is because each level of the inline stack will have its own header lineno. One should use the head lineno of its own inline stack level instead of the actual symbol.
llvm-svn: 249848
Pass MemCpyOpt doesn't check if a store instruction is nontemporal.
As a consequence, adjacent nontemporal stores are always merged into a
memset call.
Example:
;;;
define void @foo(<4 x float>* nocapture %p) {
entry:
store <4 x float> zeroinitializer, <4 x float>* %p, align 16, !nontemporal !0
%p1 = getelementptr inbounds <4 x float>, <4 x float>* %dst, i64 1
store <4 x float> zeroinitializer, <4 x float>* %p1, align 16, !nontemporal !0
ret void
}
!0 = !{i32 1}
;;;
In this example, the two nontemporal stores are combined to a memset of zero
which does not preserve the nontemporal hint. Later on the backend (tested on a
x86-64 corei7) expands that memset call into a sequence of two normal 16-byte
aligned vector stores.
opt -memcpyopt example.ll -S -o - | llc -mcpu=corei7 -o -
Before:
xorps %xmm0, %xmm0
movaps %xmm0, 16(%rdi)
movaps %xmm0, (%rdi)
With this patch, we no longer merge nontemporal stores into calls to memset.
In this example, llc correctly expands the two stores into two movntps:
xorps %xmm0, %xmm0
movntps %xmm0, 16(%rdi)
movntps %xmm0, (%rdi)
In theory, we could extend the usage of !nontemporal metadata to memcpy/memset
calls. However a change like that would only have the effect of forcing the
backend to expand !nontemporal memsets back to sequences of store instructions.
A memset library call would not have exactly the same semantic of a builtin
!nontemporal memset call. So, SelectionDAG will have to conservatively expand
it back to a sequence of !nontemporal stores (effectively undoing the merging).
Differential Revision: http://reviews.llvm.org/D13519
llvm-svn: 249820
Summary:
These non-semantic changes will help make a later change adding
support for deopt operand bundles more streamlined.
Reviewers: reames, swaroop.sridhar
Subscribers: sanjoy, llvm-commits
Differential Revision: http://reviews.llvm.org/D13491
llvm-svn: 249779
Summary:
This will be used in a later change to RewriteStatepointsForGC.
Reviewers: reames, swaroop.sridhar
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D13490
llvm-svn: 249777
Summary: Use `const auto &` instead of `auto` in `makeStatepointExplicit`.
Reviewers: reames, swaroop.sridhar
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D13454
llvm-svn: 249776
This is an implementation of
https://github.com/google/sanitizers/issues/579
It has a number of advantages over the current mapping:
* Works for non-PIE executables.
* Does not require ASLR; as a consequence, debugging MSan programs in
gdb no longer requires "set disable-randomization off".
* Supports linux kernels >=4.1.2.
* The code is marginally faster and smaller.
This is an ABI break. We never really promised ABI stability, but
this patch includes a courtesy escape hatch: a compile-time macro
that reverts back to the old mapping layout.
llvm-svn: 249753
This is a partial fix for PR24886:
https://llvm.org/bugs/show_bug.cgi?id=24886
Without this IR transform, the backend (x86 at least) was producing inefficient code.
This patch is making 2 assumptions:
1. The canonical form of a fabs() operation is, in fact, the LLVM fabs() intrinsic.
2. The high bit of an FP value is always the sign bit; as noted in the bug report, this isn't specified by the LangRef.
Differential Revision: http://reviews.llvm.org/D13076
llvm-svn: 249702
Summary:
After r249211, SCEV can see through some LCSSA phis. Add a
`replacementPreservesLCSSAForm` check before replacing uses of these phi
nodes with a simplified use of the induction variable to avoid breaking
LCSSA.
Fixes 25047.
Depends on D13460.
Reviewers: atrick, hfinkel
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D13461
llvm-svn: 249575
Summary:
Some target intrinsics can access multiple elements, using the pointer as a
base address (e.g. AArch64 ld4). When trying to CSE such instructions,
it must be checked the available value comes from a compatible instruction
because the pointer is not enough to discriminate whether the value is
correct.
Reviewers: ssijaric
Subscribers: mcrosier, llvm-commits, aemerson
Differential Revision: http://reviews.llvm.org/D13475
llvm-svn: 249523
I don't think this assert adds much value, and removing it and related
variables avoids an "unused variable" warning in release builds.
llvm-svn: 249511
Summary:
A series of cosmetic cleanup changes to RewriteStatepointsForGC:
- Rename variables to LLVM style
- Remove some redundant asserts
- Remove an unsued `Pass *` parameter
- Remove unnecessary variables
- Use C++11 idioms where applicable
- Pass CallSite by value, not reference
Reviewers: reames, swaroop.sridhar
Subscribers: llvm-commits, sanjoy
Differential Revision: http://reviews.llvm.org/D13370
llvm-svn: 249508
This will allow us to optimize code such as:
int f(int *p) {
int x;
return p == &x;
}
as well as:
int *allocate(void);
int f() {
int x;
int *p = allocate();
return p == &x;
}
The folding can only be done under certain circumstances. Even though p and &x
cannot alias, the comparison must still return true if the pointer
representations are equal. If a user successfully generates a p that's a
correct guess for &x, comparison should return true even though p is an invalid
pointer.
This patch argues that if the address of the alloca isn't observable outside the
function, the function can act as-if the address is impossible to guess from the
outside. The tricky part is keeping the act consistent: if we fold p == &x to
false in one place, we must make sure to fold any other comparisons based on
those pointers similarly. To ensure that, we only fold when &x is involved
exactly once in comparison instructions.
Differential Revision: http://reviews.llvm.org/D13358
llvm-svn: 249490
Summary:
After r249211, `getSCEV(X) == getSCEV(Y)` does not guarantee that X and
Y are related in the dominator tree, even if X is an operand to Y (I've
included a toy example in comments, and a real example as a test case).
This commit changes `SimplifyIndVar` to require a `DominatorTree`. I
don't think this is a problem because `ScalarEvolution` requires it
anyway.
Fixes PR25051.
Depends on D13459.
Reviewers: atrick, hfinkel
Subscribers: joker.eph, llvm-commits, sanjoy
Differential Revision: http://reviews.llvm.org/D13460
llvm-svn: 249471
Summary:
- Add CoreCLR to if/else ladders and switches as appropriate.
- Rename isMSVCEHPersonality to isFuncletEHPersonality to better
reflect what it captures.
Reviewers: majnemer, andrew.w.kaylor, rnk
Subscribers: pgavlin, AndyAyers, llvm-commits
Differential Revision: http://reviews.llvm.org/D13449
llvm-svn: 249455
If the mask of a select instruction is a ConstantVector, method
SimplifyDemandedVectorElts iterates over the mask elements to identify which
values are selected from the select inputs.
Before this patch, method SimplifyDemandedVectorElts always used method
Constant::isNullValue() to check if a value in the mask was zero. Unfortunately
that method always returns false when called on a ConstantExpr.
This patch fixes the problem in SimplifyDemandedVectorElts by adding an explicit
check for ConstantExpr values. Now, if a value in the mask is a ConstantExpr, we
avoid calling isNullValue() on it.
Fixes PR24922.
Differential Revision: http://reviews.llvm.org/D13219
llvm-svn: 249390
Otherwise, the map will observe changes as long as MergeFunctions is alive. This
is bad because follow-up passes could replace-all-uses-with on the key of an
entry in the map. The value handle callback of ValueMap however asserts that the
key type matches.
rdar://22971893
llvm-svn: 249327
The most important part required to make clang
devirtualization works ( ͡°͜ʖ ͡°).
The code is able to find non local dependencies, but unfortunatelly
because the caller can only handle local dependencies, I had to add
some restrictions to look for dependencies only in the same BB.
http://reviews.llvm.org/D12992
llvm-svn: 249196
When trying to optimize fortified library functions use the right
location to insert new instructions in order to preserve correct
def-use order.
This fixes an issue where a misplaced instruction definition would
happen to be *after* one of its use after a RAUW, forming invalid IR.
This behavior was introduced by r227250.
Differential Revision: http://reviews.llvm.org/D13301
rdar://problem/22802369
llvm-svn: 249092
Summary:
Some passes may open up opportunities for optimizations, leaving empty
lifetime start/end ranges. For example, with the following code:
void foo(char *, char *);
void bar(int Size, bool flag) {
for (int i = 0; i < Size; ++i) {
char text[1];
char buff[1];
if (flag)
foo(text, buff); // BBFoo
}
}
the loop unswitch pass will create 2 versions of the loop, one with
flag==true, and the other one with flag==false, but always leaving
the BBFoo basic block, with lifetime ranges covering the scope of the for
loop. Simplify CFG will then remove BBFoo in the case where flag==false,
but will leave the lifetime markers.
This patch teaches InstCombine to remove trivially empty lifetime marker
ranges, that is ranges ending right after they were started (ignoring
debug info or other lifetime markers in the range).
This fixes PR24598: excessive compile time after r234581.
Reviewers: reames, chandlerc
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D13305
llvm-svn: 249018
Summary:
The instructions SeenExprs records may be deleted during rewriting.
FindClosestMatchingDominator should ignore these deleted instructions.
Fixes PR24301.
Reviewers: grosser
Subscribers: grosser, llvm-commits
Differential Revision: http://reviews.llvm.org/D13315
llvm-svn: 248983
Summary:
Given an array of i2 elements, 4 consecutive scalar loads will be lowered to
i8-sized loads and thus will access 4 consecutive bytes in memory. If we
vectorize these loads into a single <4 x i2> load, it'll access only 1 byte in
memory. Hence, we should prohibit vectorization in such cases.
PS: Initial patch was proposed by Arnold.
Reviewers: aschwaighofer, nadav, hfinkel
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D13277
llvm-svn: 248943
Same strategy as simplifyInstructionsInBlock. ~1/3 less time
on my test suite. This pass doesn't have many in-tree users,
but getting rid of an O(N^2) worst case and making it cleaner
should at least make it a viable alternative to ADCE, since
it's now consistently somewhat faster.
llvm-svn: 248927
Usually large blocks are not a problem. But if a large block (> 10k instructions)
contains many (potential) chains of vector instructions, and those chains are
spread over a wide range of instructions, then scheduling becomes a compile time problem.
This change introduces a limit for the accumulate scheduling region size of a block.
For real-world functions this limit will never be exceeded (it's about 10x larger than
the maximum value seen in the test-suite and external test suite).
llvm-svn: 248917
This patch teaches InstCombiner how to convert a SSSE3/AVX2 byte shuffle to a
builtin shuffle if the mask is constant.
Converting byte shuffle intrinsic calls to builtin shuffles can help finding
more opportunities for combining shuffles later on in selection dag.
We may end up with byte shuffles with constant masks as the result of inlining.
Differential Revision: http://reviews.llvm.org/D13252
llvm-svn: 248913
Currently SimplifyDemandedVectorElts can only peek through bitcasts if the vectors have the same number of elements.
This patch fixes and enables some existing (disabled) code to support bitcasting to vectors with more/fewer elements. It currently only accepts cases when vectors alias cleanly (i.e. number of elements are an exact multiple of the other vector).
This was added to improve the demanded vector elements support for SSE vector shifts which require the __m128i (<2 x i64>) argument type to be bitcast to the vector type for the builtin shift. I've added extra tests for various additional bitcasts.
Differential Revision: http://reviews.llvm.org/D12935
llvm-svn: 248784
Summary: This patch adds block frequency analysis to LoopUnswitch pass to recognize hot/cold regions. For cold regions the pass only performs trivial unswitches since they do not increase code size, and for hot regions everything works as before. This helps to minimize code growth in cold regions and be more aggressive in hot regions. Currently the default cold regions are blocks with frequencies below 20% of function entry frequency, and it can be adjusted via -loop-unswitch-cold-block-frequency flag. The entire feature is controlled via -loop-unswitch-with-block-frequency flag and it is off by default.
Reviewers: broune, silvas, dnovillo, reames
Subscribers: davidxl, llvm-commits
Differential Revision: http://reviews.llvm.org/D11605
llvm-svn: 248777
Place new and update dbg.declare calls immediately after the
corresponding alloca.
Current code in replaceDbgDeclareForAlloca puts the new dbg.declare
at the end of the basic block. LLVM codegen has problems emitting
debug info in a situation when dbg.declare appears after all uses of
the variable. This usually kinda works for inlining and ASan (two
users of this function) but not for SafeStack (see the pending change
in http://reviews.llvm.org/D13178).
llvm-svn: 248769
Patch by Jake VanAdrighem!
Summary:
Fix the way we sort the llvm.used and llvm.compiler.used members.
This bug seems to have been introduced in rL183756 through a set of improper casts to GlobalValue*. In subsequent patches this problem was missed and transformed into a getName call on a ConstantExpr.
Reviewers: silvas
Subscribers: silvas, llvm-commits
Differential Revision: http://reviews.llvm.org/D12851
llvm-svn: 248728
1. Use a worklist, not a recursive approach, to avoid needless
revisitation and being repeatedly forced to jump back to the
start of the BB if a handle is invalidated.
2. Only insert operands to the worklist if they become unused
after a dead instruction is removed, so we don’t have to
visit them again in most cases.
3. Use a SmallSetVector to track the worklist.
4. Instead of pre-initting the SmallSetVector like in
DeadCodeEliminationPass, only put things into the worklist
if they have to be revisited after the first run-through.
This minimizes how much the actual SmallSetVector gets used,
which saves a lot of time.
llvm-svn: 248727
Originally, debug intrinsics and annotation intrinsics may prevent
the loop to be rerolled, now they are ignored.
Differential Revision: http://reviews.llvm.org/D13150
llvm-svn: 248718
This is one step towards solving PR24766:
https://llvm.org/bugs/show_bug.cgi?id=24766
We were not producing the same IR for these two C functions because the store
to the temp bool causes extra zexts:
#include <stdbool.h>
bool switchy(char x1, char x2, char condition) {
bool conditionMet = false;
switch (condition) {
case 0: conditionMet = (x1 == x2); break;
case 1: conditionMet = (x1 <= x2); break;
}
return conditionMet;
}
bool switchy2(char x1, char x2, char condition) {
switch (condition) {
case 0: return (x1 == x2);
case 1: return (x1 <= x2);
}
return false;
}
As noted in the code comments, this test case manages to avoid the more general existing
phi optimizations where there are only 2 phi inputs or where there are no constant phi
args mixed in with the casts ops. It seems like a corner case, but if we don't catch it,
then I don't think we can get SimplifyCFG to further optimize towards the canonical form
for this function shown in the bug report.
Differential Revision: http://reviews.llvm.org/D12866
llvm-svn: 248689
Summary:
Factor the code that rewrites invokes to calls and rewrites WinEH
terminators to their "unwind to caller" equivalents into a helper in
Utils/Local, and use it in the three places I'm aware of that need to do
this.
Reviewers: andrew.w.kaylor, majnemer, rnk
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D13152
llvm-svn: 248677
This is a fix for PR22723:
https://llvm.org/bugs/show_bug.cgi?id=22723
My first attempt at this was to change what I thought was the root problem:
xor (zext i1 X to i32), 1 --> zext (xor i1 X, true) to i32
...but we create the opposite pattern in InstCombiner::visitZExt(), so infinite loop!
My next idea was to fix the matchIfNot() implementation in PatternMatch, but that would
mean potentially returning a different size for the match than what was input. I think
this would require all users of m_Not to check the size of the returned match, so I
abandoned that idea.
I settled on just fixing the exact case presented in the PR. This patch does allow the
2 functions in PR22723 to compile identically (x86):
bool test(bool x, bool y) { return !x | !y; }
bool test(bool x, bool y) { return !x || !y; }
...
andb %sil, %dil
xorb $1, %dil
movb %dil, %al
retq
Differential Revision: http://reviews.llvm.org/D12705
llvm-svn: 248634
Loop unswitching produces conditional branches with constant condition,
and it's beneficial for later passes to clean this up with simplify-cfg.
We do this after the second invocation of loop-unswitch, but not after
the first one. Not doing so might cause problem for passes like
LoopUnroll, whose estimate of loop body size would be less accurate.
Reviewers: hfinkel
Differential Revision: http://reviews.llvm.org/D13064
llvm-svn: 248460
Nothing is expected to change, except we do less redundant work in
clean-up.
Reviewers: hfinkel
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D12951
llvm-svn: 248444
In -fprofile-instr-generate compilation, to remove the redundant profile
variables for the COMDAT functions, these variables are placed in the same
COMDAT group as its associated function. This way when the COMDAT function
is not picked by the linker, those profile variables will also not be
output in the final binary. This may cause warning when mix link objects
built w and wo -fprofile-instr-generate.
This patch puts the profile variables for COMDAT functions to its own COMDAT
group to avoid the problem.
Patch by xur.
Differential Revision: http://reviews.llvm.org/D12248
llvm-svn: 248440
This patch changes the order of GEPs generated by Splitting GEPs
pass, specially when one of the GEPs has constant and the base is
loop invariant, then we will generate the GEP with constant first
when beneficial, to expose more cases for LICM.
If originally Splitting GEP generate the following:
do.body.i:
%idxprom.i = sext i32 %shr.i to i64
%2 = bitcast %typeD* %s to i8*
%3 = shl i64 %idxprom.i, 2
%uglygep = getelementptr i8, i8* %2, i64 %3
%uglygep7 = getelementptr i8, i8* %uglygep, i64 1032
...
Now it genereates:
do.body.i:
%idxprom.i = sext i32 %shr.i to i64
%2 = bitcast %typeD* %s to i8*
%3 = shl i64 %idxprom.i, 2
%uglygep = getelementptr i8, i8* %2, i64 1032
%uglygep7 = getelementptr i8, i8* %uglygep, i64 %3
...
For no-loop cases, the original way of generating GEPs seems to
expose more CSE cases, so we don't change the logic for no-loop
cases, and only limit our change to the specific case we are
interested in.
llvm-svn: 248420
Add two new ways of accessing the unsafe stack pointer:
* At a fixed offset from the thread TLS base. This is very similar to
StackProtector cookies, but we plan to extend it to other backends
(ARM in particular) soon. Bionic-side implementation here:
https://android-review.googlesource.com/170988.
* Via a function call, as a fallback for platforms that provide
neither a fixed TLS slot, nor a reasonable TLS implementation (i.e.
not emutls).
This is a re-commit of a change in r248357 that was reverted in
r248358.
llvm-svn: 248405
This changes the behavior of AddAligntmentAssumptions to match its
comment. I.e, prove the asserted alignment in the context of the caller,
not the callee.
Thanks to Mehdi Amini for seeing the issue here! Also to Artur Pilipenko
who also saw a fix for the issue.
rdar://22521387
Differential Revision: http://reviews.llvm.org/D12997
llvm-svn: 248390
Invoking a function which returns an aggregate can sometimes be
transformed to return a scalar value. However, this means that we need
to create an insertvalue instruction(s) to recreate the correct
aggregate type. We achieved this by inserting an insertvalue
instruction at the invoke's normal successor. However, this is not
feasible if the normal successor uses the invoke's return value inside a
PHI node.
Instead, split the edge between the invoke and the unwind successor and
create the insertvalue instruction in the new basic block. The new
basic block's successor will be the old invoke successor which leaves
us with IR which is well behaved.
This fixes PR24906.
llvm-svn: 248387
This change allows dead store elimination to remove zero and null stores into memory freshly allocated with calloc-like function.
Differential Revision: http://reviews.llvm.org/D13021
llvm-svn: 248374
This patches removes the x86.sse41.pmovsx* intrinsics, provides a suitable upgrade path and updates relevant tests to sign extend a subvector instead.
LLVM counterpart to D12835
Differential Revision: http://reviews.llvm.org/D13002
llvm-svn: 248368
Summary:
It is fairly common to call SE->getConstant(Ty, 0) or
SE->getConstant(Ty, 1); this change makes such uses a little bit
briefer.
I've refactored the call sites I could find easily to use getZero /
getOne.
Reviewers: hfinkel, majnemer, reames
Subscribers: sanjoy, llvm-commits
Differential Revision: http://reviews.llvm.org/D12947
llvm-svn: 248362
Add two new ways of accessing the unsafe stack pointer:
* At a fixed offset from the thread TLS base. This is very similar to
StackProtector cookies, but we plan to extend it to other backends
(ARM in particular) soon. Bionic-side implementation here:
https://android-review.googlesource.com/170988.
* Via a function call, as a fallback for platforms that provide
neither a fixed TLS slot, nor a reasonable TLS implementation (i.e.
not emutls).
llvm-svn: 248357
Apart from checking that GlobalVariable is a constant, we should check
that it's not a weak constant, in which case we can't propagate its
value.
llvm-svn: 248327
Summary:
We should either require the DT info to be available, or check if it's
available in every place we use DT (and we already miss such check in
one place, which causes failures in some cases). As other loop passes
preserve DT and it's usually available, it makes sense to just require
it here.
There is no regression test, because the bug only shows up if pass
manager decides to clean DT info right before LoopUnswitch. If
loop-unswitch is run separately, DT is available, so bug isn't exposed.
Reviewers: chandlerc, hfinkel
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D13036
llvm-svn: 248230
We know that an argmemonly function can only access memory pointed to by it's pointer arguments. Rather than needing to consider all possible stores as aliasing (as we do for a readonly function), we can only consider the aliasing of the pointer arguments.
Note that this change only addresses hoisting. I'm thinking about how to address speculation safety as well, but that will be a different change.
FYI, argmemonly disallows accessing memory through non-pointer typed arguments.
Differential Revision: http://reviews.llvm.org/D12771
llvm-svn: 248220
We're currently losing any fast-math flags when synthesizing fcmps for
min/max reductions. In LV, make sure we copy over the scalar inst's
flags. In LoopUtils, we know we only ever match patterns with
hasUnsafeAlgebra, so apply that to any synthesized ops.
llvm-svn: 248201
evaluate whether 'readonly' or 'readnone' apply to a given function.
This both reduces indentation and will make it easy to share the logic
with a new pass manager implementation.
llvm-svn: 248181
