In addition to target-dependent attributes, we can also preserve a
white-listed subset of target independent function attributes. The white-list
excludes problematic attributes, most prominently:
* attributes related to memory accesses, as alloca instructions
could be moved in/out of the extracted block
* control-flow dependent attributes, like no_return or thunk, as the
relerelevant instructions might or might not get extracted.
Thanks @efriedma and @aemerson for providing a set of attributes that cannot be
propagated.
Reviewers: efriedma, davidxl, davide, silvas
Reviewed By: efriedma
Differential Revision: https://reviews.llvm.org/D41334
llvm-svn: 321961
If the varargs are not accessed by a function, we can inline the
function.
Reviewers: dblaikie, chandlerc, davide, efriedma, rnk, hfinkel
Reviewed By: efriedma
Differential Revision: https://reviews.llvm.org/D41335
llvm-svn: 321940
In the minimal case, this won't remove instructions, but it still improves
uses of existing values.
In the motivating example from PR35834, it does remove instructions, and
sets that case up to be optimized by something like D41603:
https://reviews.llvm.org/D41603
llvm-svn: 321936
This is the last step needed to fix PR33325:
https://bugs.llvm.org/show_bug.cgi?id=33325
We're trading branch and compares for loads and logic ops.
This makes the code smaller and hopefully faster in most cases.
The 24-byte test shows an interesting construct: we load the trailing scalar
elements into vector registers and generate the same pcmpeq+movmsk code that
we expected for a pair of full vector elements (see the 32- and 64-byte tests).
Differential Revision: https://reviews.llvm.org/D41714
llvm-svn: 321934
Besides the bug of omitting the inverse transform of max(~a, ~b) --> ~min(a, b),
the use checking and operand creation were off. We were potentially creating
repeated identical instructions of existing values. This led to infinite
looping after I added the extra folds.
By using the simpler m_Not matcher and not creating new 'not' ops for a and b,
we avoid that problem. It's possible that not using IsFreeToInvert() here is
more limiting than the simpler matcher, but there are no tests for anything
more exotic. It's also possible that we should relax the use checking further
to handle a case like PR35834:
https://bugs.llvm.org/show_bug.cgi?id=35834
...but we can make that a follow-up if it is needed.
llvm-svn: 321882
Summary:
See D37528 for a previous (non-deferred) version of this
patch and its description.
Preserves dominance in a deferred manner using a new class
DeferredDominance. This reduces the performance impact of
updating the DominatorTree at every edge insertion and
deletion. A user may call DDT->flush() within JumpThreading
for an up-to-date DT. This patch currently has one flush()
at the end of runImpl() to ensure DT is preserved across
the pass.
LVI is also preserved to help subsequent passes such as
CorrelatedValuePropagation. LVI is simpler to maintain and
is done immediately (not deferred). The code to perfom the
preversation was minimally altered and was simply marked
as preserved for the PassManager to be informed.
This extends the analysis available to JumpThreading for
future enhancements. One example is loop boundary threading.
Reviewers: dberlin, kuhar, sebpop
Reviewed By: kuhar, sebpop
Subscribers: hiraditya, llvm-commits
Differential Revision: https://reviews.llvm.org/D40146
llvm-svn: 321825
Summary:
I basically copied this patch from here:
https://reviews.llvm.org/D1251
But I skipped some of the refactoring to make the patch more clean.
The new outer3/inner3 test case in ptr-diff.ll triggers the
following assert without this patch:
lib/IR/Constants.cpp:1834: static llvm::Constant *llvm::ConstantExpr::getCompare(unsigned short, llvm::Constant *, llvm::Constant *, bool): Assertion `C1->getType() == C2->getType() && "Op types should be identical!"' failed.
The other new test cases makes sure that there is code coverage
for all modifications in InlineCost.cpp (getting different values
due to not fetching sizes for address space zero). I only guarantee
code coverage for those tests. The tests are not written in a way
that they would break if not having the corrections in
InlineCost.cpp. I found it quite hard to fine tune the tests into
getting different results based on the pointer sizes (except for
the test case where we hit an assert if not teaching InlineCost
about address spaces).
Reviewers: chandlerc, arsenm, haicheng
Reviewed By: arsenm
Subscribers: wdng, eraman, llvm-commits, haicheng
Differential Revision: https://reviews.llvm.org/D40455
llvm-svn: 321809
These were inspired by a very old review I'm about to abandon (https://reviews.llvm.org/D7061). Several of the test cases from that worked without modification and expanding test coverage of such cases is always worthwhile.
llvm-svn: 321764
The work order was changed in r228186 from SCC order
to RPO with an arbitrary sorting function. The sorting
function attempted to move inner loop nodes earlier. This
was was apparently relying on an assumption that every block
in a given loop / the same loop depth would be seen before
visiting another loop. In the broken testcase, a block
outside of the loop was encountered before moving onto
another block in the same loop. The testcase would then
structurize such that one blocks unconditional successor
could never be reached.
Revert to plain RPO for the analysis phase. This fixes
detecting edges as backedges that aren't really.
The processing phase does use another visited set, and
I'm unclear on whether the order there is as important.
An arbitrary order doesn't work, and triggers some infinite
loops. The reversed RPO list seems to work and is closer
to the order that was used before, minus the arbitary
custom sorting.
A few of the changed tests now produce smaller code,
and a few are slightly worse looking.
llvm-svn: 321751
Summary:
We are incorrectly updating the LI when loop-simplify generates
dedicated exit blocks for a loop. The issue is that there's an implicit
assumption that the Preds passed into UpdateAnalysisInformation are
reachable. However, this is not true and breaks LI by incorrectly
updating the header of a loop.
One such case is when we generate dedicated exits when the exit block is
a landing pad (through SplitLandingPadPredecessors). There maybe other
cases as well, since we do not guarantee that Preds passed in are
reachable basic blocks.
The added test case shows how loop-simplify breaks LI for the outer loop (and DT in turn)
after we try to generate the LoopSimplifyForm.
Reviewers: davide, chandlerc, sanjoy
Reviewed By: davide
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D41519
llvm-svn: 321653
In one case, we were handling out of bounds, but not undef indices. In the other, we were handling undef (with the comment making the analogy to out of bounds), but not out of bounds. Be consistent and treat both undef and constant out of bounds indices as producing undefined results.
As a side effect, this also protects instcombine from having to handle large constant indices as we always simplify first.
llvm-svn: 321575
Went to reduce another fuzzer failure to find it's already been fixed, but the test case is slightly different so it's worth adding anyways.
Reduced from oss-fuzz #4768 test case
llvm-svn: 321573
Summary:
New pass manager driver passes DebugPM (-debug-pass-manager) flag into
individual PassManager constructors in order to enable debug logging.
FunctionToLoopPassAdaptor has its own internal LoopCanonicalizationPM
which never gets its debug logging enabled and that means canonicalization
passes like LoopSimplify are never present in -debug-pass-manager output.
Extending FunctionToLoopPassAdaptor's constructor and
createFunctionToLoopPassAdaptor wrapper with an optional
boolean DebugLogging argument.
Passing debug-logging flags there as appropriate.
Reviewers: chandlerc, davide
Reviewed By: davide
Subscribers: mehdi_amini, eraman, llvm-commits, JDevlieghere
Differential Revision: https://reviews.llvm.org/D41586
llvm-svn: 321548
`RewriteStatepointsForGC` iterates over function blocks and their predecessors
in order of declaration. One of outcomes of this is that callsites are placed in
arbitrary order which has nothing to do with travelsar order.
On the other hand, function `recomputeLiveInValues` asserts that bases are
added to `Info.PointerToBase` before their deried pointers are updated. But
if call sites are processed in order different from RPOT, this is not necessarily
true. We cannot guarantee that the base was placed there before every
pointer derived from it. All we can guarantee is that this base was marked as
known base by this point.
This patch replaces the fact that we assert from checking that the base was
added to the map with assert that the base was marked as known base.
Differential Revision: https://reviews.llvm.org/D41593
llvm-svn: 321517
This reverts r321138. It seems there are still underlying issues with
memdep. PR35519 seems to still be present if debug info is enabled. We
end up losing a memcpy. Somehow during store to memset merging, we
insert the memset after the memcpy or fail to update the memdep analysis
to account for the newly inserted memset of a pair.
Reduced test case:
#include <assert.h>
#include <stdio.h>
#include <string>
#include <utility>
#include <vector>
void do_push_back(
std::vector<std::pair<std::string, std::vector<std::string>>>* crls) {
crls->push_back(std::make_pair(std::string(), std::vector<std::string>()));
}
int __attribute__((optnone)) main() {
// Put some data in the vector and then remove it so we take the push_back
// fast path.
std::vector<std::pair<std::string, std::vector<std::string>>> crl_set;
crl_set.push_back({"asdf", {}});
crl_set.pop_back();
printf("first word in vector storage: %p\n", *(void**)crl_set.data());
// Do the push_back which may fail to initialize the data.
do_push_back(&crl_set);
auto* first = &crl_set.back().first;
printf("first word in vector storage (should be zero): %p\n",
*(void**)crl_set.data());
assert(first->empty());
puts("ok");
}
Compile with libc++, enable optimizations, and enable debug info:
$ clang++ -stdlib=libc++ -g -O2 t.cpp -o t.exe -Wl,-rpath=llvm/build/lib
This program will assert with this change.
llvm-svn: 321510
This is a preliminary step for the patch discussed in D41136 (and denoted here with the FIXME comment).
When we match an FP min/max that is cast to integer, any intermediate difference between +0.0 or -0.0
should be muted in the result by the conversion (either fptosi or fptoui) of the result. Thus, we can
enable 'nsz' for the purpose of matching fmin/fmax.
Note that there's probably room to generalize this more, possibly by fixing the current calls to the
weak version of isKnownNonZero() in matchSelectPattern() to the more powerful recursive version.
Differential Revision: https://reviews.llvm.org/D41333
llvm-svn: 321456
By following the single predecessors of the predecessors of the call
site, we do not need to restrict the control flow.
Reviewed By: junbuml, davide
Differential Revision: https://reviews.llvm.org/D40729
llvm-svn: 321413
If after if-conversion, most of the instructions in this new BB construct a long and slow dependence chain, it may be slower than cmp/branch, even if the branch has a high miss rate, because the control dependence is transformed into data dependence, and control dependence can be speculated, and thus, the second part can execute in parallel with the first part on modern OOO processor.
This patch checks for the long dependence chain, and give up if-conversion if find one.
Differential Revision: https://reviews.llvm.org/D39352
llvm-svn: 321377
Currently, inline cost model considers a binary operator as free only if both
its operands are constants. Some simple cases are missing such as a + 0, a - a,
etc. This patch modifies visitBinaryOperator() to call SimplifyBinOp() without
going through simplifyInstruction() to get rid of the constant restriction.
Thus, visitAnd() and visitOr() are not needed.
Differential Revision: https://reviews.llvm.org/D41494
llvm-svn: 321366
The penalty is currently getting applied in a bunch of places where it
doesn't make sense, like bitcasts (which are free) and calls (which
were getting the call penalty applied twice). Instead, just apply the
penalty to binary operators and floating-point casts.
While I'm here, also fix getFPOpCost() to do the right thing in more
cases, so we don't have to dig into function attributes.
Differential Revision: https://reviews.llvm.org/D41522
llvm-svn: 321332
This patch modifies the indirect call promotion utilities by exposing and using
an unconditional call promotion interface. The unconditional promotion
interface (i.e., call promotion without creating an if-then-else) can be used
if it's known that an indirect call has only one possible callee. The existing
conditional promotion interface uses this unconditional interface to promote an
indirect call after it has been versioned and placed within the "then" block.
A consequence of unconditional promotion is that the fix-up operations for phi
nodes in the normal destination of invoke instructions are changed. This is
necessary because the existing implementation assumed that an invoke had been
versioned, creating a "merge" block where a return value bitcast could be
placed. In the new implementation, the edge between a promoted invoke's parent
block and its normal destination is split if needed to add a bitcast for the
return value. If the invoke is also versioned, the phi node merging the return
value of the promoted and original invoke instructions is placed in the "merge"
block.
Differential Revision: https://reviews.llvm.org/D40751
llvm-svn: 321210
Summary:
The function section prefix for PGO based layout (e.g. hot/unlikely)
should look at the hotness of all blocks not just the entry BB.
A function with a cold entry but a very hot loop should be placed in the
hot section, for example, so that it is located close to other hot
functions it may call. For SamplePGO it was already looking at the
branch weights on calls, and I made that code conditional on whether
this is SamplePGO since it was essentially a noop for instrumentation
PGO anyway.
Reviewers: davidxl
Subscribers: eraman, llvm-commits
Differential Revision: https://reviews.llvm.org/D41395
llvm-svn: 321197
canVectorize is only checking if the loop has a normalized pre-header if DoExtraAnalysis is true.
This doesn't make sense to me because reporting analysis information shouldn't alter legality
checks. This is probably the result of a last minute minor change before committing (?).
Patch by Diego Caballero.
Reviewed By: fhahn
Differential Revision: https://reviews.llvm.org/D40973
llvm-svn: 321172
This teaches memcpyopt to make a non-local memdep query when a local query
indicates that the dependency is non-local. This notably allows it to
eliminate many more llvm.memcpy calls in common Rust code, often by 20-30%.
This is r319482 and r319483, along with fixes for PR35519: fix the
optimization that merges stores into memsets to preserve cached memdep
info, and fix memdep's non-local caching strategy to not assume that larger
queries are always more conservative than smaller ones.
Fixes PR28958 and PR35519.
Differential Revision: https://reviews.llvm.org/D40802
llvm-svn: 321138
PRE in JumpThreading should not be able to hoist copy of non-speculable loads across
instructions that don't always transfer execution to their successors, otherwise they may
introduce an unsafe load which otherwise would not be executed.
The same problem for GVN was fixed as rL316975.
Differential Revision: https://reviews.llvm.org/D40347
llvm-svn: 321063
There are cases when two tags with different base types denote
accesses to the same direct or indirect member of a structure
type. Currently, merging of such tags results in a tag that
represents an access to an object that has the type of that
member. This patch changes this so that if one of the accesses
encloses the other, then the generic tag is the one of the
enclosed access.
Differential Revision: https://reviews.llvm.org/D39557
llvm-svn: 321019
Summary:
In r277849, getEntryCount was changed to return None when the entry
count was 0, specifically for SamplePGO where it means no samples were
recorded. However, for instrumentation PGO a 0 entry count should be
returned directly, since it does mean that the function was completely
cold. Otherwise we end up treating these functions conservatively
in isFunctionEntryCold() and isColdBB().
Instead, for SamplePGO use -1 when there are no samples, and change
getEntryCount to return None when the value is -1.
Reviewers: danielcdh, davidxl
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D41307
llvm-svn: 321018
This patch introduce a switch to control splitting of non-whole-alloca slices with default off.
The switch will be default on again after fixing an issue reported in PR35657.
llvm-svn: 320958
When we put the value in select placeholder we must pass
the value through simplification tracker due to the value might
be already simplified and erased.
This is a fix for PR35658.
Reviewers: john.brawn, uabelho
Reviewed By: john.brawn
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D41251
llvm-svn: 320956
We want to do this for 2 reasons:
1. Value tracking does not recognize the ashr variant, so it would fail to match for cases like D39766.
2. DAGCombiner does better at producing optimal codegen when we have the cmp+sel pattern.
More detail about what happens in the backend:
1. DAGCombiner has a generic transform for all targets to convert the scalar cmp+sel variant of abs
into the shift variant. That is the opposite of this IR canonicalization.
2. DAGCombiner has a generic transform for all targets to convert the vector cmp+sel variant of abs
into either an ABS node or the shift variant. That is again the opposite of this IR canonicalization.
3. DAGCombiner has a generic transform for all targets to convert the exact shift variants produced by #1 or #2
into an ISD::ABS node. Note: It would be an efficiency improvement if we had #1 go directly to an ABS node
when that's legal/custom.
4. The pattern matching above is incomplete, so it is possible to escape the intended/optimal codegen in a
variety of ways.
a. For #2, the vector path is missing the case for setlt with a '1' constant.
b. For #3, we are missing a match for commuted versions of the shift variants.
5. Therefore, this IR canonicalization can only help get us to the optimal codegen. The version of cmp+sel
produced by this patch will be recognized in the DAG and converted to an ABS node when possible or the
shift sequence when not.
6. In the following examples with this patch applied, we may get conditional moves rather than the shift
produced by the generic DAGCombiner transforms. The conditional move is created using a target-specific
decision for any given target. Whether it is optimal or not for a particular subtarget may be up for debate.
define i32 @abs_shifty(i32 %x) {
%signbit = ashr i32 %x, 31
%add = add i32 %signbit, %x
%abs = xor i32 %signbit, %add
ret i32 %abs
}
define i32 @abs_cmpsubsel(i32 %x) {
%cmp = icmp slt i32 %x, zeroinitializer
%sub = sub i32 zeroinitializer, %x
%abs = select i1 %cmp, i32 %sub, i32 %x
ret i32 %abs
}
define <4 x i32> @abs_shifty_vec(<4 x i32> %x) {
%signbit = ashr <4 x i32> %x, <i32 31, i32 31, i32 31, i32 31>
%add = add <4 x i32> %signbit, %x
%abs = xor <4 x i32> %signbit, %add
ret <4 x i32> %abs
}
define <4 x i32> @abs_cmpsubsel_vec(<4 x i32> %x) {
%cmp = icmp slt <4 x i32> %x, zeroinitializer
%sub = sub <4 x i32> zeroinitializer, %x
%abs = select <4 x i1> %cmp, <4 x i32> %sub, <4 x i32> %x
ret <4 x i32> %abs
}
> $ ./opt -instcombine shiftyabs.ll -S | ./llc -o - -mtriple=x86_64 -mattr=avx
> abs_shifty:
> movl %edi, %eax
> negl %eax
> cmovll %edi, %eax
> retq
>
> abs_cmpsubsel:
> movl %edi, %eax
> negl %eax
> cmovll %edi, %eax
> retq
>
> abs_shifty_vec:
> vpabsd %xmm0, %xmm0
> retq
>
> abs_cmpsubsel_vec:
> vpabsd %xmm0, %xmm0
> retq
>
> $ ./opt -instcombine shiftyabs.ll -S | ./llc -o - -mtriple=aarch64
> abs_shifty:
> cmp w0, #0 // =0
> cneg w0, w0, mi
> ret
>
> abs_cmpsubsel:
> cmp w0, #0 // =0
> cneg w0, w0, mi
> ret
>
> abs_shifty_vec:
> abs v0.4s, v0.4s
> ret
>
> abs_cmpsubsel_vec:
> abs v0.4s, v0.4s
> ret
>
> $ ./opt -instcombine shiftyabs.ll -S | ./llc -o - -mtriple=powerpc64le
> abs_shifty:
> srawi 4, 3, 31
> add 3, 3, 4
> xor 3, 3, 4
> blr
>
> abs_cmpsubsel:
> srawi 4, 3, 31
> add 3, 3, 4
> xor 3, 3, 4
> blr
>
> abs_shifty_vec:
> vspltisw 3, -16
> vspltisw 4, 15
> vsubuwm 3, 4, 3
> vsraw 3, 2, 3
> vadduwm 2, 2, 3
> xxlxor 34, 34, 35
> blr
>
> abs_cmpsubsel_vec:
> vspltisw 3, -16
> vspltisw 4, 15
> vsubuwm 3, 4, 3
> vsraw 3, 2, 3
> vadduwm 2, 2, 3
> xxlxor 34, 34, 35
> blr
>
Differential Revision: https://reviews.llvm.org/D40984
llvm-svn: 320921
Changes to the original scalar loop during LV code gen cause the return value
of Legal->isConsecutivePtr() to be inconsistent with the return value during
legal/cost phases (further analysis and information of the bug is in D39346).
This patch is an alternative fix to PR34965 following the CM_Widen approach
proposed by Ayal and Gil in D39346. It extends InstWidening enum with
CM_Widen_Reverse to properly record the widening decision for consecutive
reverse memory accesses and, consequently, get rid of the
Legal->isConsetuviePtr() call in LV code gen. I think this is a simpler/cleaner
solution to PR34965 than the one in D39346.
Fixes PR34965.
Patch by Diego Caballero, thanks!
Differential Revision: https://reviews.llvm.org/D40742
llvm-svn: 320913
Summary:
1. Use stream 0 only for combined module. Previously if combined module was not
processes ThinLTO used the stream for own output. However small changes in input,
could trigger combined module and shuffle outputs making life of llvm::LTO harder.
2. Always process combined module and write output to stream 0. Processing empty
combined module is cheap and allows llvm::LTO users to avoid implementing processing
which is already done in llvm::LTO.
Subscribers: mehdi_amini, inglorion, eraman, hiraditya
Differential Revision: https://reviews.llvm.org/D41267
llvm-svn: 320905
When unsafe algerbra is allowed calls to cabs(r) can be replaced by:
sqrt(creal(r)*creal(r) + cimag(r)*cimag(r))
Patch by Paul Walker, thanks!
Differential Revision: https://reviews.llvm.org/D40069
llvm-svn: 320901
Summary:
This implements a missing feature to allow importing of aliases, which
was previously disabled because alias cannot be available_externally.
We instead import an alias as a copy of its aliasee.
Some additional work was required in the IndexBitcodeWriter for the
distributed build case, to ensure that the aliasee has a value id
in the distributed index file (i.e. even when it is not being
imported directly).
This is a performance win in codes that have many aliases, e.g. C++
applications that have many constructor and destructor aliases.
Reviewers: pcc
Subscribers: mehdi_amini, inglorion, eraman, llvm-commits
Differential Revision: https://reviews.llvm.org/D40747
llvm-svn: 320895
This recommits r320823 reverted due to the test failure in sink-foldable.ll and
an unused variable. Added "REQUIRES: aarch64-registered-target" in the test
and removed unused variable.
Original commit message:
Continue trying to sink an instruction if its users in the loop is foldable.
This will allow the instruction to be folded in the loop by decoupling it from
the user outside of the loop.
Reviewers: hfinkel, majnemer, davidxl, efriedma, danielcdh, bmakam, mcrosier
Reviewed By: hfinkel
Subscribers: javed.absar, bmakam, mcrosier, llvm-commits
Differential Revision: https://reviews.llvm.org/D37076
llvm-svn: 320858
This recommit r320823 after fixing a test failure.
Original commit message:
Continue trying to sink an instruction if its users in the loop is foldable.
This will allow the instruction to be folded in the loop by decoupling it from
the user outside of the loop.
Reviewers: hfinkel, majnemer, davidxl, efriedma, danielcdh, bmakam, mcrosier
Reviewed By: hfinkel
Subscribers: javed.absar, bmakam, mcrosier, llvm-commits
Differential Revision: https://reviews.llvm.org/D37076
llvm-svn: 320833
Summary:
Continue trying to sink an instruction if its users in the loop is foldable.
This will allow the instruction to be folded in the loop by decoupling it from
the user outside of the loop.
Reviewers: hfinkel, majnemer, davidxl, efriedma, danielcdh, bmakam, mcrosier
Reviewed By: hfinkel
Subscribers: javed.absar, bmakam, mcrosier, llvm-commits
Differential Revision: https://reviews.llvm.org/D37076
llvm-svn: 320823
SROA analysis of InlineCost can figure out that some stores can be removed
after inlining and then the repeated loads clobbered by these stores are also
free. This patch finds these clobbered loads and adjust the inline cost
accordingly.
Differential Revision: https://reviews.llvm.org/D33946
llvm-svn: 320814
Summary:
The port is nearly straightforward.
The only complication is related to the analyses handling,
since one of the analyses used in this module pass is domtree,
which is a function analysis. That requires asking for the results
of each function and disallows a single interface for run-on-module
pass action.
Decided to copy-paste the main body of this pass.
Most of its code is requesting analyses anyway, so not that much
of a copy-paste.
The rest of the code movement is to transform all the implementation
helper functions like stripNonValidData into non-member statics.
Extended all the related LLVM tests with new-pass-manager use.
No failures.
Reviewers: sanjoy, anna, reames
Reviewed By: anna
Subscribers: skatkov, llvm-commits
Differential Revision: https://reviews.llvm.org/D41162
llvm-svn: 320796
We cannot move the insertion point to header if SCEV contains div/rem
operations due to they may go over check for zero denominator.
Reviewers: sanjoy, mkazantsev, sebpop
Reviewed By: sebpop
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D41229
llvm-svn: 320789
This should solve:
https://bugs.llvm.org/show_bug.cgi?id=34603
...by preventing SimplifyCFG from altering redundant instructions before early-cse has a chance to run.
It changes the default (canonical-forming) behavior of SimplifyCFG, so we're only doing the
sinking transform later in the optimization pipeline.
Differential Revision: https://reviews.llvm.org/D38566
llvm-svn: 320749
In SLPVectorizer, the vector build instructions (insertvalue for aggregate type) is passed to BoUpSLP.buildTree, it is treated as UserIgnoreList, so later in cost estimation, the cost of these instructions are not counted.
For aggregate value, later usage are more likely to be done in scalar registers, either used as individual scalars or used as a whole for function call or return value. Ignore scalar extraction instructions may cause too aggressive vectorization for aggregate values, and slow down performance. So for vectorization of aggregate value, the scalar extraction instructions are required in cost estimation.
Differential Revision: https://reviews.llvm.org/D41139
llvm-svn: 320736
Summary:
The function is meant to recurse until it comes upon the
phi it's looking for. However, with the current condition,
it will recurse until it finds anything _but_ the phi.
The function will even fail for simple cases like:
%i = phi i32 [ %inc, %loop ], ...
...
%inc = add i32 %i, 1
because the base condition will not happen when the phi
is recursed to, and the recursion will end with a 'false'
result since the previous instruction is a phi.
Reviewers: sanjoy, atrick
Reviewed By: sanjoy
Subscribers: Ka-Ka, bjope, llvm-commits
Committing on behalf of: Bevin Hansson (bevinh)
Differential Revision: https://reviews.llvm.org/D40946
llvm-svn: 320700
This patch fix this FIXME in visitPHI()
FIXME: We should potentially be tracking values through phi nodes,
especially when they collapse to a single value due to deleted CFG edges
during inlining.
Differential Revision: https://reviews.llvm.org/D38594
llvm-svn: 320699
Summary:
Passing AliasAnalysis results instead of nullptr appears to work just fine.
A couple new-pass-manager tests updated to align with new order of analyses.
Reviewers: chandlerc, spatel, craig.topper
Reviewed By: chandlerc
Subscribers: mehdi_amini, eraman, llvm-commits
Differential Revision: https://reviews.llvm.org/D41203
llvm-svn: 320687
D30041 extended SCEVPredicateRewriter to improve handling of Phi nodes whose
update chain involves casts; PSCEV can now build an AddRecurrence for some
forms of such phi nodes, under the proper runtime overflow test. This means
that we can identify such phi nodes as an induction, and the loop-vectorizer
can now vectorize such inductions, however inefficiently. The vectorizer
doesn't know that it can ignore the casts, and so it vectorizes them.
This patch records the casts in the InductionDescriptor, so that they could
be marked to be ignored for cost calculation (we use VecValuesToIgnore for
that) and ignored for vectorization/widening/scalarization (i.e. treated as
TriviallyDead).
In addition to marking all these casts to be ignored, we also need to make
sure that each cast is mapped to the right vector value in the vector loop body
(be it a widened, vectorized, or scalarized induction). So whenever an
induction phi is mapped to a vector value (during vectorization/widening/
scalarization), we also map the respective cast instruction (if exists) to that
vector value. (If the phi-update sequence of an induction involves more than one
cast, then the above mapping to vector value is relevant only for the last cast
of the sequence as we allow only the "last cast" to be used outside the
induction update chain itself).
This is the last step in addressing PR30654.
llvm-svn: 320672
Summary:
See D37528 for a previous (non-deferred) version of this
patch and its description.
Preserves dominance in a deferred manner using a new class
DeferredDominance. This reduces the performance impact of
updating the DominatorTree at every edge insertion and
deletion. A user may call DDT->flush() within JumpThreading
for an up-to-date DT. This patch currently has one flush()
at the end of runImpl() to ensure DT is preserved across
the pass.
LVI is also preserved to help subsequent passes such as
CorrelatedValuePropagation. LVI is simpler to maintain and
is done immediately (not deferred). The code to perfom the
preversation was minimally altered and was simply marked
as preserved for the PassManager to be informed.
This extends the analysis available to JumpThreading for
future enhancements. One example is loop boundary threading.
Reviewers: dberlin, kuhar, sebpop
Reviewed By: kuhar, sebpop
Subscribers: hiraditya, llvm-commits
Differential Revision: https://reviews.llvm.org/D40146
llvm-svn: 320612
w.r.t. the paper
"A Practical Improvement to the Partial Redundancy Elimination in SSA Form"
(https://sites.google.com/site/jongsoopark/home/ssapre.pdf)
Proper dominance check was missing here, so having a loopinfo should not be required.
Committing this diff as this fixes the bug, if there are
further concerns, I'll be happy to work on them.
Differential Revision: https://reviews.llvm.org/D39781
llvm-svn: 320607
Summary:
This patch tries to vectorize loads of consecutive memory accesses, accessed
in non-consecutive or jumbled way. An earlier attempt was made with patch D26905
which was reverted back due to some basic issue with representing the 'use mask' of
jumbled accesses.
This patch fixes the mask representation by recording the 'use mask' in the usertree entry.
Change-Id: I9fe7f5045f065d84c126fa307ef6ebe0787296df
Reviewers: mkuper, loladiro, Ayal, zvi, danielcdh
Reviewed By: Ayal
Subscribers: mgrang, dcaballe, hans, mzolotukhin
Differential Revision: https://reviews.llvm.org/D36130
llvm-svn: 320548
Summary:
This change makes the call site creation more general if any of the
arguments is predicated on a condition in the call site's predecessors.
If we find a callsite, that potentially can be split, we collect the set
of conditions for the call site's predecessors (currently only 2
predecessors are allowed). To do that, we traverse each predecessor's
predecessors as long as it only has single predecessors and record the
condition, if it is relevant to the call site. For each condition, we
also check if the condition is taken or not. In case it is not taken,
we record the inverse predicate.
We use the recorded conditions to create the new call sites and split
the basic block.
This has 2 benefits: (1) it is slightly easier to see what is going on
(IMO) and (2) we can easily extend it to handle more complex control
flow.
Reviewers: davidxl, junbuml
Reviewed By: junbuml
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D40728
llvm-svn: 320547
Summary:
If we have pattern `store (load(bitcast(select (cmp(V1, V2), &V1,
&V2)))), bitcast)`, but the load is used in other instructions, it leads
to looping in InstCombiner. Patch adds additional check that all users
of the load instructions are stores and then replaces all uses of load
instruction by the new one with new type.
Reviewers: RKSimon, spatel, majnemer
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D41072
llvm-svn: 320525
This algorithm (explained more in the source code) takes into account
global redundancies by building a "pair map" to find common subexprs.
The primary motivation of this is to handle situations like
foo = (a * b) * c
bar = (a * d) * c
where we currently don't identify that "a * c" is redundant.
Accordingly, it prioritizes the emission of a * c so that CSE
can remove the redundant calculation later.
Does not change the actual reassociation algorithm -- only the
order in which the reassociated operand chain is reconstructed.
Gives ~1.5% floating point math instruction count reduction on
a large offline suite of graphics shaders.
llvm-svn: 320515
Summary:
The PGO gen/use passes currently fail with an assert failure if there's a
critical edge whose source is an IndirectBr instruction and that edge
needs to be instrumented.
To avoid this in certain cases, split IndirectBr critical edges in the PGO
gen/use passes. This works for blocks with single indirectbr predecessors,
but not for those with multiple indirectbr predecessors (splitting an
IndirectBr critical edge isn't always possible.)
Reviewers: davidxl, xur
Reviewed By: davidxl
Subscribers: efriedma, llvm-commits, mehdi_amini
Differential Revision: https://reviews.llvm.org/D40699
llvm-svn: 320511
Summary:
If we have pattern `store (load(bitcast(select (cmp(V1, V2), &V1,
&V2)))), bitcast)`, but the load is used in other instructions, it leads
to looping in InstCombiner. Patch adds additional check that all users
of the load instructions are stores and then replaces all uses of load
instruction by the new one with new type.
Reviewers: RKSimon, spatel, majnemer
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D41072
llvm-svn: 320510
Summary:
If we have pattern `store (load(bitcast(select (cmp(V1, V2), &V1,
&V2)))), bitcast)`, but the load is used in other instructions, it leads
to looping in InstCombiner. Patch adds additional check that all users
of the load instructions are stores and then replaces all uses of load
instruction by the new one with new type.
Reviewers: RKSimon, spatel, majnemer
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D41072
llvm-svn: 320499
Summary:
If we have pattern `store (load(bitcast(select (cmp(V1, V2), &V1,
&V2)))), bitcast)`, but the load is used in other instructions, it leads
to looping in InstCombiner. Patch adds additional check that all users
of the load instructions are stores and then replaces all uses of load
instruction by the new one with new type.
Reviewers: RKSimon, spatel, majnemer
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D41072
llvm-svn: 320496
Summary:
If we have pattern `store (load(bitcast(select (cmp(V1, V2), &V1,
&V2)))), bitcast)`, but the load is used in other instructions, it leads
to looping in InstCombiner. Patch adds additional check that all users
of the load instructions are stores and then replaces all uses of load
instruction by the new one with new type.
Reviewers: RKSimon, spatel, majnemer
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D41072
llvm-svn: 320488
If we have pattern `store (load(bitcast(select (cmp(V1, V2), &V1,
&V2)))), bitcast)`, but the load is used in other instructions, it leads
to looping in InstCombiner. Patch adds additional check that all users
of the load instructions are stores and then replaces all uses of load
instruction by the new one with new type.
Reviewers: RKSimon, spatel, majnemer
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D41072
llvm-svn: 320483
Summary:
Currently, in InstCombineLoadStoreAlloca, we have simplification
rules for the following cases:
1. load off a null
2. load off a GEP with null base
3. store to a null
This patch adds support for the fourth case which is store into a
GEP with null base. Since this is UB as well (and directly analogous to
the load off a GEP with null base), we can substitute the stored val
with undef in instcombine, so that SimplifyCFG can optimize this code
into unreachable code.
Note: Right now, simplifyCFG hasn't been taught about optimizing
this to unreachable and adding an llvm.trap (this is already done for
the above 3 cases).
Reviewers: majnemer, hfinkel, sanjoy, davide
Reviewed by: sanjoy, davide
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D41026
llvm-svn: 320480
VecValuesToIgnore holds values that will not appear in the vectorized loop.
We should therefore ignore their cost when VF > 1.
Differential Revision: https://reviews.llvm.org/D40883
llvm-svn: 320463
Summary:
This solves PR35616.
We don't want the compiler to generate different code when we compile
with/without -g, so we now ignore debug intrinsics when determining if
the optimization can trigger or not.
Reviewers: junbuml
Subscribers: davide, JDevlieghere, llvm-commits
Differential Revision: https://reviews.llvm.org/D41068
llvm-svn: 320460
The tests fail (opt asserts) on Windows.
> Summary:
> If we have pattern `store (load(bitcast(select (cmp(V1, V2), &V1,
> &V2)))), bitcast)`, but the load is used in other instructions, it leads
> to looping in InstCombiner. Patch adds additional check that all users
> of the load instructions are stores and then replaces all uses of load
> instruction by the new one with new type.
>
> Reviewers: RKSimon, spatel, majnemer
>
> Subscribers: llvm-commits
>
> Differential Revision: https://reviews.llvm.org/D41072
llvm-svn: 320421
Summary:
If we have pattern `store (load(bitcast(select (cmp(V1, V2), &V1,
&V2)))), bitcast)`, but the load is used in other instructions, it leads
to looping in InstCombiner. Patch adds additional check that all users
of the load instructions are stores and then replaces all uses of load
instruction by the new one with new type.
Reviewers: RKSimon, spatel, majnemer
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D41072
llvm-svn: 320407
CreateAddRecFromPHIWithCastsImpl() adds an IncrementNUSW overflow predicate
which allows the PSCEV rewriter to rewrite this scev expression:
(zext i8 {0, + , (trunc i32 step to i8)} to i32)
into
{0, +, (sext i8 (trunc i32 step to i8) to i32)}
But then it adds the wrong Equal predicate:
%step == (zext i8 (trunc i32 %step to i8) to i32).
instead of:
%step == (sext i8 (trunc i32 %step to i8) to i32)
This is fixed here.
Differential Revision: https://reviews.llvm.org/D40641
llvm-svn: 320298
Summary:
This is LLVM instrumentation for the new HWASan tool. It is basically
a stripped down copy of ASan at this point, w/o stack or global
support. Instrumenation adds a global constructor + runtime callbacks
for every load and store.
HWASan comes with its own IR attribute.
A brief design document can be found in
clang/docs/HardwareAssistedAddressSanitizerDesign.rst (submitted earlier).
Reviewers: kcc, pcc, alekseyshl
Subscribers: srhines, mehdi_amini, mgorny, javed.absar, eraman, llvm-commits, hiraditya
Differential Revision: https://reviews.llvm.org/D40932
llvm-svn: 320217
The Debugify pass synthesizes debug info for IR. It's paired with a
CheckDebugify pass which determines how much of the original debug info
is preserved. These passes make it easier to create targeted tests for
debug info preservation.
Here is the Debugify algorithm:
NextLine = 1
for (Instruction &I : M)
attach DebugLoc(NextLine++) to I
NextVar = 1
for (Instruction &I : M)
if (canAttachDebugValue(I))
attach dbg.value(NextVar++) to I
The CheckDebugify pass expects contiguous ranges of DILocations and
DILocalVariables. If it fails to find all of the expected debug info, it
prints a specific error to stderr which can be FileChecked.
This was discussed on llvm-dev in the thread:
"Passes to add/validate synthetic debug info"
Differential Revision: https://reviews.llvm.org/D40512
llvm-svn: 320202
Summary:
If a partially inlined function has debug info, we have to add debug
locations to the call instruction calling the outlined function.
We use the debug location of the first instruction in the outlined
function, as the introduced call transfers control to this statement and
there is no other equivalent line in the source code.
We also use the same debug location for the branch instruction added
to jump from artificial entry block for the outlined function, which just
jumps to the first actual basic block of the outlined function.
Reviewers: davide, aprantl, rriddle, dblaikie, danielcdh, wmi
Reviewed By: aprantl, rriddle, danielcdh
Subscribers: eraman, JDevlieghere, llvm-commits
Differential Revision: https://reviews.llvm.org/D40413
llvm-svn: 320199
Causes unexpected memory issue with New PM this time.
The new PM invalidates BPI but not BFI, leaving the
reference to BPI from BFI invalid.
Abandon this patch. There is a more general solution
which also handles runtime infinite loop (but not statically).
llvm-svn: 320180
Summary:
If we have the code like this:
```
float a, b;
a = std::max(a ,b);
```
it is converted into something like this:
```
%call = call dereferenceable(4) float* @_ZSt3maxIfERKT_S2_S2_(float* nonnull dereferenceable(4) %a.addr, float* nonnull dereferenceable(4) %b.addr)
%1 = bitcast float* %call to i32*
%2 = load i32, i32* %1, align 4
%3 = bitcast float* %a.addr to i32*
store i32 %2, i32* %3, align 4
```
After inlinning this code is converted to the next:
```
%1 = load float, float* %a.addr
%2 = load float, float* %b.addr
%cmp.i = fcmp fast olt float %1, %2
%__b.__a.i = select i1 %cmp.i, float* %a.addr, float* %b.addr
%3 = bitcast float* %__b.__a.i to i32*
%4 = load i32, i32* %3, align 4
%5 = bitcast float* %arrayidx to i32*
store i32 %4, i32* %5, align 4
```
This pattern is not recognized as minmax pattern.
Patch solves this problem by converting sequence
```
store (bitcast, (load bitcast (select ((cmp V1, V2), &V1, &V2))))
```
to a sequence
```
store (,load (select((cmp V1, V2), &V1, &V2)))
```
After this the code is recognized as minmax pattern.
Reviewers: RKSimon, spatel
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D40304
llvm-svn: 320157
As a new access is generated spanning across multiple fields, we need to
propagate alias info from all the fields to form the most generic alias info.
rdar://35602528
Differential Revision: https://reviews.llvm.org/D40617
llvm-svn: 319979
Summary:
An undef extract index can be arbitrarily chosen to be an
out-of-range index value, which would result in the instruction being undef.
This change closes a gap identified while working on lowering vector permute intrinsics
with variable index vectors to pure LLVM IR.
Reviewers: arsenm, spatel, majnemer
Reviewed By: arsenm, spatel
Subscribers: fhahn, nhaehnle, wdng, llvm-commits
Differential Revision: https://reviews.llvm.org/D40231
llvm-svn: 319910
This caused PR35519.
> [memcpyopt] Teach memcpyopt to optimize across basic blocks
>
> This teaches memcpyopt to make a non-local memdep query when a local query
> indicates that the dependency is non-local. This notably allows it to
> eliminate many more llvm.memcpy calls in common Rust code, often by 20-30%.
>
> Fixes PR28958.
>
> Differential Revision: https://reviews.llvm.org/D38374
>
> [memcpyopt] Commit file missed in r319482.
>
> This change was meant to be included with r319482 but was accidentally
> omitted.
llvm-svn: 319873
This uses ConstantRange::makeGuaranteedNoWrapRegion's newly-added handling for subtraction to allow CVP to remove some subtraction overflow checks.
Differential Revision: https://reviews.llvm.org/D40039
llvm-svn: 319807
Summary:
A true or false result is expected from a comparison, but it seems the possibility of undef was overlooked, which could lead to a failed assert. This is fixed by this patch by bailing out if we encounter undef.
The bug is old and the assert has been there since the end of 2014, so it seems this is unusual enough to forego optimization.
Patch by JesperAntonsson.
Reviewers: spatel, eeckstein, hans
Reviewed By: hans
Subscribers: uabelho, llvm-commits
Differential Revision: https://reviews.llvm.org/D40639
llvm-svn: 319768
Follow-up of r316824. This patch supports the vector type for both current and
previous index when factoring out the current one into the previous one.
Differential Revision: https://reviews.llvm.org/D39556
llvm-svn: 319683
(This reapplies r314253. r314253 was reverted on r314482 because of a
correctness regression on P100, but that regression was identified to be
something else.)
Summary:
Don't bail out on constant divisors for divisions that can be narrowed without
introducing control flow . This gives us a 32 bit multiply instead of an
emulated 64 bit multiply in the generated PTX assembly.
Reviewers: jlebar
Subscribers: jholewinski, mcrosier, llvm-commits
Differential Revision: https://reviews.llvm.org/D38265
llvm-svn: 319677
Summary:
Currently, we only support predication for forward loops with step
of 1. This patch enables loop predication for reverse or
countdownLoops, which satisfy the following conditions:
1. The step of the IV is -1.
2. The loop has a singe latch as B(X) = X <pred>
latchLimit with pred as s> or u>
3. The IV of the guard is the decrement
IV of the latch condition (Guard is: G(X) = X-1 u< guardLimit).
This patch was downstream for a while and is the last series of patches
that's from our LP implementation downstream.
Reviewers: apilipenko, mkazantsev, sanjoy
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D40353
llvm-svn: 319659
Turns out we can have comparisons which are indirect users of the induction variable that we can make invariant. In this case, there is no loop invariant value contributing and we'd fail an assert.
The test case was found by a java fuzzer and reduced. It's a real cornercase. You have to have a static loop which we've already proven only executes once, but haven't broken the backedge on, and an inner phi whose result can be constant folded by SCEV using exit count reasoning but not proven by isKnownPredicate. To my knowledge, only the fuzzer has hit this case.
llvm-svn: 319583
These are blocks that haven't not been executed during training. For large
projects this could make a significant difference. For the project, I was
looking at, I got an order of magnitude decrease in the size of the total YAML
files with this and r319235.
Differential Revision: https://reviews.llvm.org/D40678
Re-commit after fixing the failing testcase in rL319576, rL319577 and
rL319578.
llvm-svn: 319581
These are blocks that haven't not been executed during training. For large
projects this could make a significant difference. For the project, I was
looking at, I got an order of magnitude decrease in the size of the total YAML
files with this and r319235.
Differential Revision: https://reviews.llvm.org/D40678
llvm-svn: 319556
It causes builds to fail with "Instruction does not dominate all uses" (PR35497).
> Patch tries to improve vectorization of the following code:
>
> void add1(int * __restrict dst, const int * __restrict src) {
> *dst++ = *src++;
> *dst++ = *src++ + 1;
> *dst++ = *src++ + 2;
> *dst++ = *src++ + 3;
> }
> Allows to vectorize even if the very first operation is not a binary add, but just a load.
>
> Fixed issues related to previous commit.
>
> Reviewers: spatel, mzolotukhin, mkuper, hfinkel, RKSimon, filcab, ABataev
>
> Reviewed By: ABataev, RKSimon
>
> Subscribers: llvm-commits, RKSimon
>
> Differential Revision: https://reviews.llvm.org/D28907
llvm-svn: 319550
Summary:
1/ Operand folding during complex pattern matching for LEAs has been extended, such that it promotes Scale to
accommodate similar operand appearing in the DAG e.g.
T1 = A + B
T2 = T1 + 10
T3 = T2 + A
For above DAG rooted at T3, X86AddressMode will now look like
Base = B , Index = A , Scale = 2 , Disp = 10
2/ During OptimizeLEAPass down the pipeline factorization is now performed over LEAs so that if there is an opportunity
then complex LEAs (having 3 operands) could be factored out e.g.
leal 1(%rax,%rcx,1), %rdx
leal 1(%rax,%rcx,2), %rcx
will be factored as following
leal 1(%rax,%rcx,1), %rdx
leal (%rdx,%rcx) , %edx
3/ Aggressive operand folding for AM based selection for LEAs is sensitive to loops, thus avoiding creation of any complex LEAs within a loop.
4/ Simplify LEA converts (lea (BASE,1,INDEX,0) --> add (BASE, INDEX) which offers better through put.
PR32755 will be taken care of by this pathc.
Previous patch revisions : r313343 , r314886
Reviewers: lsaba, RKSimon, craig.topper, qcolombet, jmolloy, jbhateja
Reviewed By: lsaba, RKSimon, jbhateja
Subscribers: jmolloy, spatel, igorb, llvm-commits
Differential Revision: https://reviews.llvm.org/D35014
llvm-svn: 319543
Summary:
A true or false result is expected from a comparison, but it seems the possibility of undef was overlooked, which could lead to a failed assert. This is fixed by this patch by bailing out if we encounter undef.
The bug is old and the assert has been there since the end of 2014, so it seems this is unusual enough to forego optimization.
Patch by: JesperAntonsson
Reviewers: spatel, eeckstein, hans
Reviewed By: hans
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D40639
llvm-svn: 319537
Patch tries to improve vectorization of the following code:
void add1(int * __restrict dst, const int * __restrict src) {
*dst++ = *src++;
*dst++ = *src++ + 1;
*dst++ = *src++ + 2;
*dst++ = *src++ + 3;
}
Allows to vectorize even if the very first operation is not a binary add, but just a load.
Fixed issues related to previous commit.
Reviewers: spatel, mzolotukhin, mkuper, hfinkel, RKSimon, filcab, ABataev
Reviewed By: ABataev, RKSimon
Subscribers: llvm-commits, RKSimon
Differential Revision: https://reviews.llvm.org/D28907
llvm-svn: 319531
If the thin module has no references to an internal global in the
merged module, we need to make sure to preserve that property if the
global is a member of a comdat group, as otherwise promotion can end
up adding global symbols to the comdat, which is not allowed.
This situation can arise if the external global in the thin module
has dead constant users, which would cause use_empty() to return
false and would cause us to try to promote it. To prevent this from
happening, discard the dead constant users before asking whether a
global is empty.
Differential Revision: https://reviews.llvm.org/D40593
llvm-svn: 319494
This teaches memcpyopt to make a non-local memdep query when a local query
indicates that the dependency is non-local. This notably allows it to
eliminate many more llvm.memcpy calls in common Rust code, often by 20-30%.
Fixes PR28958.
Differential Revision: https://reviews.llvm.org/D38374
llvm-svn: 319482
Currently, SROA splits loads and stores only when they are accessing the whole alloca.
This patch relaxes this limitation to allow splitting a load/store if all other loads and stores to the alloca are disjoint to or fully included in the current load/store. If there is no other load or store that crosses the boundary of the current load/store, the current splitting implementation works as is.
The whole-alloca loads and stores meet this new condition and so they are still splittable.
Here is a simplified motivating example.
struct record {
long long a;
int b;
int c;
};
int func(struct record r) {
for (int i = 0; i < r.c; i++)
r.b++;
return r.b;
}
When updating r.b (or r.c as well), LLVM generates redundant instructions on some platforms (such as x86_64, ppc64); here, r.b and r.c are packed into one 64-bit GPR when the struct is passed as a method argument.
With this patch, the above example is compiled into only few instructions without loop.
Without the patch, unnecessary loop-carried dependency is introduced by SROA and the loop cannot be eliminated by the later optimizers.
Differential Revision: https://reviews.llvm.org/D32998
llvm-svn: 319407
If common type is different we should bail out due to we will not be
able to create a select or Phi of these values.
Basically it is done in ExtAddrMode::compare however it does not work
if we handle the null first and then two values of different types.
so add a check in initializeMap as well. The check in ExtAddrMode::compare
is used as earlier bail out.
Reviewers: reames, john.brawn
Reviewed By: john.brawn
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D40479
llvm-svn: 319292
The core idea is to (re-)introduce some redundancies where their cost is
hidden by the cost of materializing immediates for constant operands of
PHI nodes. When the cost of the redundancies is covered by this,
avoiding materializing the immediate has numerous benefits:
1) Less register pressure
2) Potential for further folding / combining
3) Potential for more efficient instructions due to immediate operand
As a motivating example, consider the remarkably different cost on x86
of a SHL instruction with an immediate operand versus a register
operand.
This pattern turns up surprisingly frequently, but is somewhat rarely
obvious as a significant performance problem.
The pass is entirely target independent, but it does rely on the target
cost model in TTI to decide when to speculate things around the PHI
node. I've included x86-focused tests, but any target that sets up its
immediate cost model should benefit from this pass.
There is probably more that can be done in this space, but the pass
as-is is enough to get some important performance on our internal
benchmarks, and should be generally performance neutral, but help with
more extensive benchmarking is always welcome.
One awkward part is that this pass has to be scheduled after
*everything* that can eliminate these kinds of redundancies. This
includes SimplifyCFG, GVN, etc. I'm open to suggestions about better
places to put this. We could in theory make it part of the codegen pass
pipeline, but there doesn't really seem to be a good reason for that --
it isn't "lowering" in any sense and only relies on pretty standard cost
model based TTI queries, so it seems to fit well with the "optimization"
pipeline model. Still, further thoughts on the pipeline position are
welcome.
I've also only implemented this in the new pass manager. If folks are
very interested, I can try to add it to the old PM as well, but I didn't
really see much point (my use case is already switched over to the new
PM).
I've tested this pretty heavily without issue. A wide range of
benchmarks internally show no change outside the noise, and I don't see
any significant changes in SPEC either. However, the size class
computation in tcmalloc is substantially improved by this, which turns
into a 2% to 4% win on the hottest path through tcmalloc for us, so
there are definitely important cases where this is going to make
a substantial difference.
Differential revision: https://reviews.llvm.org/D37467
llvm-svn: 319164
Summary:
I think we do not need to analyze debug intrinsics here, as they should
not impact codegen. This has 2 benefits: 1) slightly less work to do and
2) avoiding generating optimization remarks for converting calls to
debug intrinsics to tail calls, which are not really helpful for users.
Based on work by Sander de Smalen.
Reviewers: davide, trentxintong, aprantl
Reviewed By: aprantl
Subscribers: llvm-commits, JDevlieghere
Tags: #debug-info
Differential Revision: https://reviews.llvm.org/D40440
llvm-svn: 319158
This is to address a problem similar to those in D37460 for Scalar PRE. We should not
PRE across an instruction that may not pass execution to its successor unless it is safe
to speculatively execute it.
Differential Revision: https://reviews.llvm.org/D38619
llvm-svn: 319147
Revert "[SROA] Propagate !range metadata when moving loads."
Revert "[Mem2Reg] Clang-format unformatted parts of this file. NFCI."
Davide says they broke a bot.
llvm-svn: 319131
This tries to propagate !range metadata to a pre-existing load
when a load is optimized out. This is done instead of adding an
assume because converting loads to and from assumes creates a
lot of IR.
Patch by Ariel Ben-Yehuda.
Differential Revision: https://reviews.llvm.org/D37216
llvm-svn: 319096
enum TailCallKind { TCK_None = 0, TCK_Tail = 1, TCK_MustTail = 2,
TCK_NoTail = 3 };
TCK_NoTail is greater than TCK_Tail so taking the min does not do the
correct thing.
rdar://35639547
llvm-svn: 319075
The current way that trivial addressing modes are detected incorrectly thinks
that null pointers are non-trivial, leading to an infinite loop where we keep
duplicating the same select. Fix this by aware of null when deciding if an
addressing mode is trivial.
Differential Revision: https://reviews.llvm.org/D40447
llvm-svn: 319019
CodeGenPrepare sinks address computations from one basic block to another
and attempts to reuse address computations that have already been sunk. If
the same address computation appears twice with the first instance as an
operand of a load whose result is an operand to a simplifable select,
CodeGenPrepare simplifies the select and recursively erases the now dead
instructions. CodeGenPrepare then attempts to use the erased address
computation for the second load.
Fix this by erasing the cached address value if it has zero uses before
looking for the address value in the sunken address map.
This partially resolves PR35209.
Thanks to Alexander Richardson for reporting the issue!
This fixed version relands r318032 which was reverted in r318049 due to
sanitizer buildbot failures.
Reviewers: john.brawn
Differential Revision: https://reviews.llvm.org/D39841
llvm-svn: 318956
This patch extends the recent work in optimizeMemoryInst to make it able to
combine more ExtAddrMode fields than just the BaseReg.
This fixes some benchmark regressions introduced by r309397, where GVN PRE is
hoisting a getelementptr such that it can no longer be combined into the
addressing mode of the load or store that uses it.
Differential Revision: https://reviews.llvm.org/D38133
llvm-svn: 318949
After the dataflow algorithm proves that an argument is constant,
it replaces it value with the integer constant and drops the lattice
value associated to the DEF.
e.g. in the example we have @f() that's called twice:
call @f(undef, ...)
call @f(2, ...)
`undef` MEET 2 = 2 so we replace the argument and all its uses with
the constant 2.
Shortly after, tryToReplaceWithConstantRange() tries to get the lattice
value for the argument we just replaced, causing an assertion.
This function is a little peculiar as it runs when we're doing replacement
and not as part of the solver but still queries the solver.
The fix is that of checking whether we replaced the value already and
get a temporary lattice value for the constant.
Thanks to Zhendong Su for the report!
Fixes PR35357.
llvm-svn: 318817
properlyDominates() shouldn't be used as sort key. It causes different output between stdlibc++ and libc++.
Instead, I introduced RPOT. In most cases, it works for CSE.
llvm-svn: 318743
Summary:
Add the following heuristics for irreducible loop metadata:
- When an irreducible loop header is missing the loop header weight metadata,
give it the minimum weight seen among other headers.
- Annotate indirectbr targets with the loop header weight metadata (as they are
likely to become irreducible loop headers after indirectbr tail duplication.)
These greatly improve the accuracy of the block frequency info of the Python
interpreter loop (eg. from ~3-16x off down to ~40-55% off) and the Python
performance (eg. unpack_sequence from ~50% slower to ~8% faster than GCC) due to
better register allocation under PGO.
Reviewers: davidxl
Reviewed By: davidxl
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D39980
llvm-svn: 318693
Summary:
SROA can fail in rewriting alloca but still rewrite a phi resulting
in dead instruction elimination. The Changed flag was not being set
correctly, resulting in downstream passes using stale analyses.
The included test case will assert during the second BDCE pass as a
result.
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D39921
llvm-svn: 318677
This patch adds a new abstraction layer to VPlan and leverages it to model the planned
instructions that manipulate masks (AND, OR, NOT), introduced during predication.
The new VPValue and VPUser classes model how data flows into, through and out
of a VPlan, forming the vertices of a planned Def-Use graph. The new
VPInstruction class is a generic single-instruction Recipe that models a
planned instruction along with its opcode, operands and users. See
VectorizationPlan.rst for more details.
Differential Revision: https://reviews.llvm.org/D38676
llvm-svn: 318645
In rL316552, we ban intersection of unsigned latch range with signed range check and vice
versa, unless the entire range check iteration space is known positive. It was a correct
functional fix that saved us from dealing with ambiguous values, but it also appeared
to be a very restrictive limitation. In particular, in the following case:
loop:
%iv = phi i32 [ 0, %preheader ], [ %iv.next, %latch]
%iv.offset = add i32 %iv, 10
%rc = icmp slt i32 %iv.offset, %len
br i1 %rc, label %latch, label %deopt
latch:
%iv.next = add i32 %iv, 11
%cond = icmp i32 ult %iv.next, 100
br it %cond, label %loop, label %exit
Here, the unsigned iteration range is `[0, 100)`, and the safe range for range
check is `[-10, %len - 10)`. For unsigned iteration spaces, we use unsigned
min/max functions for range intersection. Given this, we wanted to avoid dealing
with `-10` because it is interpreted as a very big unsigned value. Semantically, range
check's safe range goes through unsigned border, so in fact it is two disjoint
ranges in IV's iteration space. Intersection of such ranges is not trivial, so we prohibited
this case saying that we are not allowed to intersect such ranges.
What semantics of this safe range actually means is that we can start from `-10` and go
up increasing the `%iv` by one until we reach `%len - 10` (for simplicity let's assume that
`%len - 10` is a reasonably big positive value).
In particular, this safe iteration space includes `0, 1, 2, ..., %len - 11`. So if we were able to return
safe iteration space `[0, %len - 10)`, we could safely intersect it with IV's iteration space. All
values in this range are non-negative, so using signed/unsigned min/max for them is unambiguous.
In this patch, we alter the algorithm of safe range calculation so that it returnes a subset of the
original safe space which is represented by one continuous range that does not go through wrap.
In order to reach this, we use modified SCEV substraction function. It can be imagined as a function
that substracts by `1` (or `-1`) as long as the further substraction does not cause a wrap in IV iteration
space. This allows us to perform IRCE in many situations when we deal with IV space and range check
of different types (in terms of signed/unsigned).
We apply this approach for both matching and not matching types of IV iteration space and the
range check. One implication of this is that now IRCE became smarter in detection of empty safe
ranges. For example, in this case:
loop:
%iv = phi i32 [ %begin, %preheader ], [ %iv.next, %latch]
%iv.offset = sub i32 %iv, 10
%rc = icmp ult i32 %iv.offset, %len
br i1 %rc, label %latch, label %deopt
latch:
%iv.next = add i32 %iv, 11
%cond = icmp i32 ult %iv.next, 100
br it %cond, label %loop, label %exit
If `%len` was less than 10 but SCEV failed to trivially prove that `%begin - 10 >u %len- 10`,
we could end up executing entire loop in safe preloop while the main loop was still generated,
but never executed. Now, cutting the ranges so that if both `begin - 10` and `%len - 10` overflow,
we have a trivially empty range of `[0, 0)`. This in some cases prevents us from meaningless optimization.
Differential Revision: https://reviews.llvm.org/D39954
llvm-svn: 318639
We must collect all AddModes even if they are the same.
This is due to Original value is different but we need all original
values collected as they are used as anchors in common phi finding.
Reviewers: john.brawn, reames
Reviewed By: john.brawn
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D40166
llvm-svn: 318638
As the first test shows, we could transform an llvm intrinsic which never sets errno
into a libcall which could set errno (even though it's marked readnone?), so that's
not ideal.
It's possible that we can also transform a libcall which could set errno to an
intrinsic given the fast-math-flags constraint, but that's deferred to determine
exactly which set of FMF are needed.
Differential Revision: https://reviews.llvm.org/D40150
llvm-svn: 318628
The 'ord' and 'uno' predicates have a logic operation for NAN built into their definitions:
FCMP_ORD = 7, ///< 0 1 1 1 True if ordered (no nans)
FCMP_UNO = 8, ///< 1 0 0 0 True if unordered: isnan(X) | isnan(Y)
So we can simplify patterns like this:
(fcmp ord (known NNAN), X) && (fcmp ord X, Y) --> fcmp ord X, Y
(fcmp uno (known NNAN), X) || (fcmp uno X, Y) --> fcmp uno X, Y
It might be better to split this into (X uno 0) | (Y uno 0) as a canonicalization, but that
would be another patch.
Differential Revision: https://reviews.llvm.org/D40130
llvm-svn: 318627
Summary:
With this patch I tried to reduce the complexity of the code sightly, by
removing some indirection. Please let me know what you think.
Reviewers: junbuml, mcrosier, davidxl
Reviewed By: junbuml
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D40037
llvm-svn: 318593
Summary: This change fix PR35342 by replacing only the current use with undef in unreachable blocks.
Reviewers: efriedma, mcrosier, igor-laevsky
Reviewed By: efriedma
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D40184
llvm-svn: 318551
Florian Hahn