This patch simplifies the calculation of certain costs in
getInstructionCost when isScalarAfterVectorization() returns a true value.
There are a few places where we multiply a cost by a number N, i.e.
unsigned N = isScalarAfterVectorization(I, VF) ? VF.getKnownMinValue() : 1;
return N * TTI.getArithmeticInstrCost(...
After some investigation it seems that there are only these cases that occur
in practice:
1. VF is a scalar, in which case N = 1.
2. VF is a vector. We can only get here if: a) the instruction is a
GEP/bitcast/PHI with scalar uses, or b) this is an update to an induction
variable that remains scalar.
I have changed the code so that N is assumed to always be 1. For GEPs
the cost is always 0, since this is calculated later on as part of the
load/store cost. PHI nodes are costed separately and were never previously
multiplied by VF. For all other cases I have added an assert that none of
the users needs scalarising, which didn't fire in any unit tests.
Only one test required fixing and I believe the original cost for the scalar
add instruction to have been wrong, since only one copy remains after
vectorisation.
I have also added a new test for the case when a pointer PHI feeds directly
into a store that will be scalarised as we were previously never testing it.
Differential Revision: https://reviews.llvm.org/D99718
This patch also refactors the way the feasible max VF is calculated,
although this is NFC for fixed-width vectors.
After this change scalable VF hints are no longer truncated/clamped
to a shorter scalable VF, nor does it drop the 'scalable flag' from
the suggested VF to vectorize with a similar VF that is fixed.
Instead, the hint is ignored which means the vectorizer is free
to find a more suitable VF, using the CostModel to determine the
best possible VF.
Reviewed By: c-rhodes, fhahn
Differential Revision: https://reviews.llvm.org/D98509
When using the -enable-strict-reductions flag where UF>1 we generate multiple
Phi nodes, though only one of these is used as an input to the vector.reduce.fadd
intrinsics. The unused Phi nodes are removed later by instcombine.
This patch changes widenPHIInstruction/fixReduction to only generate
one Phi, and adds an additional test for unrolling to strict-fadd.ll
Reviewed By: david-arm
Differential Revision: https://reviews.llvm.org/D100570
This patch simplifies the calculation of certain costs in
getInstructionCost when isScalarAfterVectorization() returns a true value.
There are a few places where we multiply a cost by a number N, i.e.
unsigned N = isScalarAfterVectorization(I, VF) ? VF.getKnownMinValue() : 1;
return N * TTI.getArithmeticInstrCost(...
After some investigation it seems that there are only these cases that occur
in practice:
1. VF is a scalar, in which case N = 1.
2. VF is a vector. We can only get here if: a) the instruction is a
GEP/bitcast/PHI with scalar uses, or b) this is an update to an induction
variable that remains scalar.
I have changed the code so that N is assumed to always be 1. For GEPs
the cost is always 0, since this is calculated later on as part of the
load/store cost. PHI nodes are costed separately and were never previously
multiplied by VF. For all other cases I have added an assert that none of
the users needs scalarising, which didn't fire in any unit tests.
Only one test required fixing and I believe the original cost for the scalar
add instruction to have been wrong, since only one copy remains after
vectorisation.
I have also added a new test for the case when a pointer PHI feeds directly
into a store that will be scalarised as we were previously never testing it.
Differential Revision: https://reviews.llvm.org/D99718
We can already vectorize loops that involve int<>int, fp<>fp, int<>fp
and fp<>int conversions, however we didn't previously have any tests
for them. This patch adds some tests for each conversion type.
Differential Revision: https://reviews.llvm.org/D99951
When vectorising for AArch64 targets if you specify the SVE attribute
we automatically then treat masked loads and stores as legal. Also,
since we have no cost model for masked memory ops we believe it's
cheap to use the masked load/store intrinsics even for fixed width
vectors. This can lead to poor code quality as the intrinsics will
currently be scalarised in the backend. This patch adds a basic
cost model that marks fixed-width masked memory ops as significantly
more expensive than for scalable vectors.
Tests for the cost model are added here:
Transforms/LoopVectorize/AArch64/masked-op-cost.ll
Differential Revision: https://reviews.llvm.org/D100745
This commit fixes a bug where the loop vectoriser fails to predicate
loads/stores when interleaving for targets that support masked
loads and stores.
Code such as:
1 void foo(int *restrict data1, int *restrict data2)
2 {
3 int counter = 1024;
4 while (counter--)
5 if (data1[counter] > data2[counter])
6 data1[counter] = data2[counter];
7 }
... could previously be transformed in such a way that the predicated
store implied by:
if (data1[counter] > data2[counter])
data1[counter] = data2[counter];
... was lost, resulting in miscompiles.
This bug was causing some tests in llvm-test-suite to fail when built
for SVE.
Differential Revision: https://reviews.llvm.org/D99569
Introduced the cost of thre reverse shuffles for AArch64, currently just
copied the costs for PermuteSingleSrc.
Differential Revision: https://reviews.llvm.org/D100871
D98435 added support for in-order reductions and included tests for fixed-width
vectorization with the -enable-strict-reductions flag.
This patch adds similar tests to verify support for scalable vectorization of loops
with in-order reductions.
Reviewed By: david-arm
Differential Revision: https://reviews.llvm.org/D100385
This also fixes a CHECK line in @fadd_strict_unroll which ensures the
changes made to fixReduction() to support in-order reductions with
unrolling are being tested correctly.
There were a few places in widenPHIInstruction where calculations of
offsets were failing to take the runtime calculation of VF into
account for scalable vectors. I've fixed those cases in this patch
as well as adding an assert that we should not be scalarising for
scalable vectors.
Tests are added here:
Transforms/LoopVectorize/AArch64/sve-widen-phi.ll
Differential Revision: https://reviews.llvm.org/D99254
After D98856 these tests will by default break (fatal_error) if any of
the wrong interfaces are used, so there's no longer a need to have a
RUN line that checks for a warning message emitted by the compiler.
Previously we could only vectorize FP reductions if fast math was enabled, as this allows us to
reorder FP operations. However, it may still be beneficial to vectorize the loop by moving
the reduction inside the vectorized loop and making sure that the scalar reduction value
be an input to the horizontal reduction, e.g:
%phi = phi float [ 0.0, %entry ], [ %reduction, %vector_body ]
%load = load <8 x float>
%reduction = call float @llvm.vector.reduce.fadd.v8f32(float %phi, <8 x float> %load)
This patch adds a new flag (IsOrdered) to RecurrenceDescriptor and makes use of the changes added
by D75069 as much as possible, which already teaches the vectorizer about in-loop reductions.
For now in-order reduction support is off by default and controlled with the `-enable-strict-reductions` flag.
Reviewed By: david-arm
Differential Revision: https://reviews.llvm.org/D98435
This patch just adds tests that we can vectorize loop such as these:
for (i = 0; i < n; i++)
dst[i * 7] += 1;
and
for (i = 0; i < n; i++)
if (cond[i])
dst[i * 7] += 1;
using scalable vectors, where we expect to use gathers and scatters in the
vectorized loop. The vector of pointers used for the gather is identical
to those used for the scatter so there should be no memory dependences.
Tests are added here:
Transforms/LoopVectorize/AArch64/sve-large-strides.ll
Differential Revision: https://reviews.llvm.org/D99192
This marks FSIN and other operations to EXPAND for scalable
vectors, so that they are not assumed to be legal by the cost-model.
Depends on D97470
Reviewed By: dmgreen, paulwalker-arm
Differential Revision: https://reviews.llvm.org/D97471
This patch adds support for the vectorization of induction variables when
using scalable vectors, which required the following changes:
1. Removed assert from InnerLoopVectorizer::getStepVector.
2. Modified InnerLoopVectorizer::createVectorIntOrFpInductionPHI to use
a runtime determined value for VF and removed an assert.
3. Modified InnerLoopVectorizer::buildScalarSteps to work for scalable
vectors. I did this by calculating the full vector value for each Part
of the unroll factor (UF) and caching this in the VP state. This means
that we are always able to extract an arbitrary element from the vector
if necessary. In addition to this, I also permitted the caching of the
individual lane values themselves for the known minimum number of elements
in the same way we do for fixed width vectors. This is a further
optimisation that improves the code quality since it avoids unnecessary
extractelement operations when extracting the first lane.
4. Added an assert to InnerLoopVectorizer::widenPHIInstruction, since while
testing some code paths I noticed this is currently broken for scalable
vectors.
Various tests to support different cases have been added here:
Transforms/LoopVectorize/AArch64/sve-inductions.ll
Differential Revision: https://reviews.llvm.org/D98715
This patch simplifies the calculation of certain costs in
getInstructionCost when isScalarAfterVectorization() returns a true value.
There are a few places where we multiply a cost by a number N, i.e.
unsigned N = isScalarAfterVectorization(I, VF) ? VF.getKnownMinValue() : 1;
return N * TTI.getArithmeticInstrCost(...
After some investigation it seems that there are only these cases that occur
in practice:
1. VF is a scalar, in which case N = 1.
2. VF is a vector. We can only get here if: a) the instruction is a
GEP/bitcast with scalar uses, or b) this is an update to an induction variable
that remains scalar.
I have changed the code so that N is assumed to always be 1. For GEPs
the cost is always 0, since this is calculated later on as part of the
load/store cost. For all other cases I have added an assert that none of the
users needs scalarising, which didn't fire in any unit tests.
Only one test required fixing and I believe the original cost for the scalar
add instruction to have been wrong, since only one copy remains after
vectorisation.
Differential Revision: https://reviews.llvm.org/D98512
D95598 added a cost model for broadcast shuffle, which should enable loops
such as the following to vectorize, where the load of b[42] is invariant
and can be done using a scalar load + splat:
for (int i=0; i<n; ++i)
a[i] = b[i] + b[42];
This patch adds tests to verify that we can vectorize such loops.
Reviewed By: joechrisellis
Differential Revision: https://reviews.llvm.org/D98506
This patch adds support for reverse loop vectorization.
It is possible to vectorize the following loop:
```
for (int i = n-1; i >= 0; --i)
a[i] = b[i] + 1.0;
```
with fixed or scalable vector.
The loop-vectorizer will use 'reverse' on the loads/stores to make
sure the lanes themselves are also handled in the right order.
This patch adds support for scalable vector on IRBuilder interface to
create a reverse vector. The IR function
CreateVectorReverse lowers to experimental.vector.reverse for scalable vector
and keedp the original behavior for fixed vector using shuffle reverse.
Differential Revision: https://reviews.llvm.org/D95363
For loops of the form:
void foo(int *a, int *cond, short *inv, long long n) {
for (long long i=0; i<n; ++i) {
if (cond[i])
a[i] = *inv;
}
}
we can vectorise for SVE using masked gather loads where the array
of pointers is simply a vector splat of 'inv' and the mask comes
from the condition 'cond[i] != 0'.
This patch simply adds tests upstream to defend this capability.
Differential Revision: https://reviews.llvm.org/D98043
There are certain loops like this below:
for (int i = 0; i < n; i++) {
a[i] = b[i] + 1;
*inv = a[i];
}
that can only be vectorised if we are able to extract the last lane of the
vectorised form of 'a[i]'. For fixed width vectors this already works since
we know at compile time what the final lane is, however for scalable vectors
this is a different story. This patch adds support for extracting the last
lane from a scalable vector using a runtime determined lane value. I have
added support to VPIteration for runtime-determined lanes that still permit
the caching of values. I did this by introducing a new class called VPLane,
which describes the lane we're dealing with and provides interfaces to get
both the compile-time known lane and the runtime determined value. Whilst
doing this work I couldn't find any explicit tests for extracting the last
lane values of fixed width vectors so I added tests for both scalable and
fixed width vectors.
Differential Revision: https://reviews.llvm.org/D95139
As a followup to D95291, getOperandsScalarizationOverhead was still
using a VF as a vector factor if the arguments were scalar, and would
assert on certain matrix intrinsics with differently sized vector
arguments. This patch removes the VF arg, instead passing the Types
through directly. This should allow it to more accurately compute the
cost without having to guess at which operands will be vectorized,
something difficult with more complex intrinsics.
This adjusts one SVE test as it is now calling the wrong intrinsic vs
veccall. Without invalid InstructCosts the cost of the scalarized
intrinsic is too low. This should get fixed when the cost of
scalarization is accounted for with scalable types.
Differential Revision: https://reviews.llvm.org/D96287
getIntrinsicInstrCost takes a IntrinsicCostAttributes holding various
parameters of the intrinsic being costed. It can either be called with a
scalar intrinsic (RetTy==Scalar, VF==1), with a vector instruction
(RetTy==Vector, VF==1) or from the vectorizer with a scalar type and
vector width (RetTy==Scalar, VF>1). A RetTy==Vector, VF>1 is considered
an error. Both of the vector modes are expected to be treated the same,
but because this is confusing many backends end up getting it wrong.
Instead of trying work with those two values separately this removes the
VF parameter, widening the RetTy/ArgTys by VF used called from the
vectorizer. This keeps things simpler, but does require some other
modifications to keep things consistent.
Most backends look like this will be an improvement (or were not using
getIntrinsicInstrCost). AMDGPU needed the most changes to keep the code
from c230965ccf working. ARM removed the fix in
dfac521da1, webassembly happens to get a fixup for an SLP cost
issue and both X86 and AArch64 seem to now be using better costs from
the vectorizer.
Differential Revision: https://reviews.llvm.org/D95291
This patch extends VPWidenPHIRecipe to manage pairs of incoming
(VPValue, VPBasicBlock) in the VPlan native path. This is made possible
because we now directly manage defined VPValues for recipes.
By keeping both the incoming value and block in the recipe directly,
code-generation in the VPlan native path becomes independent of the
predecessor ordering when fixing up non-induction phis, which currently
can cause crashes in the VPlan native path.
This fixes PR45958.
Reviewed By: sguggill
Differential Revision: https://reviews.llvm.org/D96773
This patch enables scalable vectorization of loops with integer/fast reductions, e.g:
```
unsigned sum = 0;
for (int i = 0; i < n; ++i) {
sum += a[i];
}
```
A new TTI interface, isLegalToVectorizeReduction, has been added to prevent
reductions which are not supported for scalable types from vectorizing.
If the reduction is not supported for a given scalable VF,
computeFeasibleMaxVF will fall back to using fixed-width vectorization.
Reviewed By: david-arm, fhahn, dmgreen
Differential Revision: https://reviews.llvm.org/D95245
Changes `getScalarizationOverhead` to return an invalid cost for scalable VFs
and adds some simple tests for loops containing a function for which
there is a vectorized variant available.
Reviewed By: david-arm
Differential Revision: https://reviews.llvm.org/D96356
The vector reduction intrinsics started life as experimental ops, so backend support
was lacking. As part of promoting them to 1st-class intrinsics, however, codegen
support was added/improved:
D58015
D90247
So I think it is safe to now remove this complication from IR.
Note that we still have an IR-level codegen expansion pass for these as discussed
in D95690. Removing that is another step in simplifying the logic. Also note that
x86 was already unconditionally forming reductions in IR, so there should be no
difference for x86.
I spot checked a couple of the tests here by running them through opt+llc and did
not see any asm diffs.
If we do find functional differences for other targets, it should be possible
to (at least temporarily) restore the shuffle IR with the ExpandReductions IR
pass.
Differential Revision: https://reviews.llvm.org/D96552
This reverts commit 502a67dd7f.
This expose a failure in test-suite build on PowerPC,
revert to unblock buildbot first,
Dave will re-commit in https://reviews.llvm.org/D96287.
Thanks Dave.
getIntrinsicInstrCost takes a IntrinsicCostAttributes holding various
parameters of the intrinsic being costed. It can either be called with a
scalar intrinsic (RetTy==Scalar, VF==1), with a vector instruction
(RetTy==Vector, VF==1) or from the vectorizer with a scalar type and
vector width (RetTy==Scalar, VF>1). A RetTy==Vector, VF>1 is considered
an error. Both of the vector modes are expected to be treated the same,
but because this is confusing many backends end up getting it wrong.
Instead of trying work with those two values separately this removes the
VF parameter, widening the RetTy/ArgTys by VF used called from the
vectorizer. This keeps things simpler, but does require some other
modifications to keep things consistent.
Most backends look like this will be an improvement (or were not using
getIntrinsicInstrCost). AMDGPU needed the most changes to keep the code
from c230965ccf working. ARM removed the fix in
dfac521da1, webassembly happens to get a fixup for an SLP cost
issue and both X86 and AArch64 seem to now be using better costs from
the vectorizer.
Differential Revision: https://reviews.llvm.org/D95291
This patch updates IRBuilder::CreateMaskedGather/Scatter to work
with ScalableVectorType and adds isLegalMaskedGather/Scatter functions
to AArch64TargetTransformInfo. In addition I've fixed up
isLegalMaskedLoad/Store to return true for supported scalar types,
since this is what the vectorizer asks for.
In LoopVectorize.cpp I've changed
LoopVectorizationCostModel::getInterleaveGroupCost to return an invalid
cost for scalable vectors, since currently this relies upon using shuffle
vector for reversing vectors. In addition, in
LoopVectorizationCostModel::setCostBasedWideningDecision I have assumed
that the cost of scalarising memory ops is infinitely expensive.
I have added some simple masked load/store and gather/scatter tests,
including cases where we use gathers and scatters for conditional invariant
loads and stores.
Differential Revision: https://reviews.llvm.org/D95350
I have removed an unnecessary assert in LoopVectorizationCostModel::getInstructionCost
that prevented a cost being calculated for select instructions when using
scalable vectors. In addition, I have changed AArch64TTIImpl::getCmpSelInstrCost
to only do special cost calculations for fixed width vectors and fall
back to the base version for scalable vectors.
I have added a simple cost model test for cmps and selects:
test/Analysis/CostModel/sve-cmpsel.ll
and some simple tests that show we vectorize loops with cmp and select:
test/Transforms/LoopVectorize/AArch64/sve-basic-vec.ll
Differential Revision: https://reviews.llvm.org/D95039
In the following loop:
void foo(int *a, int *b, int N) {
for (int i=0; i<N; ++i)
a[i + 4] = a[i] + b[i];
}
The loop dependence constrains the VF to a maximum of (4, fixed), which
would mean using <4 x i32> as the vector type in vectorization.
Extending this to scalable vectorization, a VF of (4, scalable) implies
a vector type of <vscale x 4 x i32>. To determine if this is legal
vscale must be taken into account. For this example, unless
max(vscale)=1, it's unsafe to vectorize.
For SVE, the number of bits in an SVE register is architecturally
defined to be a multiple of 128 bits with a maximum of 2048 bits, thus
the maximum vscale is 16. In the loop above it is therefore unfeasible
to vectorize with SVE. However, in this loop:
void foo(int *a, int *b, int N) {
#pragma clang loop vectorize_width(X, scalable)
for (int i=0; i<N; ++i)
a[i + 32] = a[i] + b[i];
}
As long as max(vscale) multiplied by the number of lanes 'X' doesn't
exceed the dependence distance, it is safe to vectorize. For SVE a VF of
(2, scalable) is within this constraint, since a vector of <16 x 2 x 32>
will have no dependencies between lanes. For any number of lanes larger
than this it would be unsafe to vectorize.
This patch extends 'computeFeasibleMaxVF' to legalize scalable VFs
specified as loop hints, implementing the following behaviour:
* If the backend does not support scalable vectors, ignore the hint.
* If scalable vectorization is unfeasible given the loop
dependence, like in the first example above for SVE, then use a
fixed VF.
* Accept scalable VFs if it's safe to do so.
* Otherwise, clamp scalable VFs that exceed the maximum safe VF.
Reviewed By: sdesmalen, fhahn, david-arm
Differential Revision: https://reviews.llvm.org/D91718
This patch makes SLP and LV emit operations with initial vectors set to poison constant instead of undef.
This is a part of efforts for using poison vector instead of undef to represent "doesn't care" vector.
The goal is to make nice shufflevector optimizations valid that is currently incorrect due to the tricky interaction between undef and poison (see https://bugs.llvm.org/show_bug.cgi?id=44185 ).
Reviewed By: fhahn
Differential Revision: https://reviews.llvm.org/D94061
This patch updates IRBuilder to create insertelement/shufflevector using poison as a placeholder.
Reviewed By: nikic
Differential Revision: https://reviews.llvm.org/D93793
... so just ensure that we pass DomTreeUpdater it into it.
Fixes DomTree preservation for a large number of tests,
all of which are marked as such so that they do not regress.
In the following loop the dependence distance is 2 and can only be
vectorized if the vector length is no larger than this.
void foo(int *a, int *b, int N) {
#pragma clang loop vectorize(enable) vectorize_width(4)
for (int i=0; i<N; ++i) {
a[i + 2] = a[i] + b[i];
}
}
However, when specifying a VF of 4 via a loop hint this loop is
vectorized. According to [1][2], loop hints are ignored if the
optimization is not safe to apply.
This patch introduces a check to bail of vectorization if the user
specified VF is greater than the maximum feasible VF, unless explicitly
forced with '-force-vector-width=X'.
[1] https://llvm.org/docs/LangRef.html#llvm-loop-vectorize-and-llvm-loop-interleave
[2] https://clang.llvm.org/docs/LanguageExtensions.html#extensions-for-loop-hint-optimizations
Reviewed By: sdesmalen, fhahn, Meinersbur
Differential Revision: https://reviews.llvm.org/D90687
Instruction ExtractValue wasn't handled in
LoopVectorizationCostModel::getInstructionCost(). As a result, it was modeled
as a mul which is not really accurate. Since it is free (most of the times),
this now gets a cost of 0 using getInstructionCost.
This is a follow-up of D92208, that required changing this regression test.
In a follow up I will look at InsertValue which also isn't handled yet.
Differential Revision: https://reviews.llvm.org/D92317
This was modeled to have a cost of 1, but since we do not have a MUL.2d this is
scalarized into vector inserts/extracts and scalar muls.
Motivating precommitted test is test/Transforms/SLPVectorizer/AArch64/mul.ll,
which we don't want to SLP vectorize.
Test Transforms/LoopVectorize/AArch64/extractvalue-no-scalarization-required.ll
unfortunately needed changing, but the reason is documented in
LoopVectorize.cpp:6855:
// The cost of executing VF copies of the scalar instruction. This opcode
// is unknown. Assume that it is the same as 'mul'.
which I will address next as a follow up of this.
Differential Revision: https://reviews.llvm.org/D92208
-Oz normally does not allow loop header duplication so this loop wouldn't be
vectorized. However the vectorization pragma should override this and allow
for loop rotation.
rdar://problem/49281061
Original patch by Adam Nemet.
Reviewed By: Meinersbur
Differential Revision: https://reviews.llvm.org/D59832
The warning would fire when calling isDereferenceableAndAlignedInLoop
with a scalable load. Calling isDereferenceableAndAlignedInLoop with a
scalable load would result in the use of the now deprecated implicit
cast of TypeSize to uint64_t through the overloaded operator.
This patch fixes this issue by:
- no longer considering vector loads as candidates in
canVectorizeWithIfConvert. This doesn't make sense in the context of
identifying scalar loads to vectorize.
- making use of getFixedSize inside isDereferenceableAndAlignedInLoop --
this removes the dependency on the deprecated interface, and will
trigger an assertion error if the function is ever called with a
scalable type.
Reviewed By: sdesmalen
Differential Revision: https://reviews.llvm.org/D89798
For some expressions, we can use information from loop guards when
we are looking for a maximum. This patch applies information from
loop guards to the expression used to compute the maximum backedge
taken count in howFarToZero. It currently replaces an unknown
expression X with UMin(X, Y), if the loop is guarded by
X ult Y.
This patch is minimal in what conditions it applies, and there
are a few TODOs to generalize.
This partly addresses PR40961. We will also need an update to
LV to address it completely.
Reviewed By: reames
Differential Revision: https://reviews.llvm.org/D67178
No widening decisions will be computed for instructions outside the
loop. Do not try to get a widening decision. The load/store will be just
a scalar load, so treating at as normal should be fine I think.
Fixes PR46950.
Reviewed By: dmgreen
Differential Revision: https://reviews.llvm.org/D85087
D79164/2596da31740f changed getCFInstrCost to return 1 per default.
AArch64 did not have its own implementation, hence the throughput cost
of CFI instructions is overestimated. On most cores, most branches should
be predicated and essentially free throughput wise.
This restores a 9% performance regression on a SPEC2006 benchmark on
AArch64 with -O3 LTO & PGO.
This patch effectively restores pre 2596da3174 behavior for AArch64
and undoes the AArch64 test changes of the patch.
Reviewers: samparker, dmgreen, anemet
Reviewed By: samparker
Differential Revision: https://reviews.llvm.org/D82755
Have BasicTTI call the base implementation so that both agree on the
default behaviour, which the default being a cost of '1'. This has
required an X86 specific implementation as it seems to be very
reliant on those instructions being free. Changes are also made to
AMDGPU so that their implementations distinguish between cost kinds,
so that the unrolling isn't affected. PowerPC also has its own
implementation to prevent changes to the reg-usage vectorizer test.
The cost model test changes now reflect that ret instructions are not
generally free.
Differential Revision: https://reviews.llvm.org/D79164
This was reverted because of a miscompilation. At closer inspection, the
problem was actually visible in a changed llvm regression test too. This
one-line follow up fix/recommit will splat the IV, which is what we are trying
to avoid if unnecessary in general, if tail-folding is requested even if all
users are scalar instructions after vectorisation. Because with tail-folding,
the splat IV will be used by the predicate of the masked loads/stores
instructions. The previous version omitted this, which caused the
miscompilation. The original commit message was:
If tail-folding of the scalar remainder loop is applied, the primary induction
variable is splat to a vector and used by the masked load/store vector
instructions, thus the IV does not remain scalar. Because we now mark
that the IV does not remain scalar for these cases, we don't emit the vector IV
if it is not used. Thus, the vectoriser produces less dead code.
Thanks to Ayal Zaks for the direction how to fix this.
If tail-folding of the scalar remainder loop is applied, the primary induction
variable is splat to a vector and used by the masked load/store vector
instructions, thus the IV does not remain scalar. Because we now mark
that the IV does not remain scalar for these cases, we don't emit the vector IV
if it is not used. Thus, the vectoriser produces less dead code.
Thanks to Ayal Zaks for the direction how to fix this.
Differential Revision: https://reviews.llvm.org/D78911
This patch adds a custom implementation of isLegalNTStore to AArch64TTI
that supports vector types that can be directly stored by STNP. Note
that the implementation may not catch all valid cases (e.g. because the
vector is a multiple of 256 and could be broken down to multiple valid 256 bit
stores), but it is good enough for LV to vectorize loops with NT stores,
as LV only passes in a vector with 2 elements to check. LV seems to also
be the only user of isLegalNTStore.
We should also do the same for NT loads, but before that we need to
ensure that we properly lower LDNP of vectors, similar to D72919.
Reviewers: dmgreen, samparker, t.p.northover, ab
Reviewed By: dmgreen
Differential Revision: https://reviews.llvm.org/D73158
Summary:
This commits is a rework of the patch in
https://reviews.llvm.org/D67572.
The rework was requested to prevent out-of-tree performance regression
when vectorizing out-of-tree IR intrinsics. The vectorization of such
intrinsics is enquired via the static function `isTLIScalarize`. For
detail see the discussion in https://reviews.llvm.org/D67572.
Reviewers: uabelho, fhahn, sdesmalen
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D72734
A sequence of additions or multiplications that is known not to wrap, may wrap
if it's order is changed (i.e., reassociated). Therefore when vectorizing
integer sum or product reductions, their no-wrap flags need to be removed.
Fixes PR43828
Patch by Denis Antrushin
Differential Revision: https://reviews.llvm.org/D69563
Loop invariant operands do not need to be scalarized, as we are using
the values outside the loop. We should ignore them when computing the
scalarization overhead.
Fixes PR41294
Reviewers: hsaito, rengolin, dcaballe, Ayal
Reviewed By: Ayal
Differential Revision: https://reviews.llvm.org/D59995
llvm-svn: 366030
This patch uses the mechanism from D62995 to strengthen the
definitions of the reduction intrinsics by letting the scalar
result/accumulator type be overloaded from the vector element type.
For example:
; The LLVM LangRef specifies that the scalar result must equal the
; vector element type, but this is not checked/enforced by LLVM.
declare i32 @llvm.experimental.vector.reduce.or.i32.v4i32(<4 x i32> %a)
This patch changes that into:
declare i32 @llvm.experimental.vector.reduce.or.v4i32(<4 x i32> %a)
Which has the type-constraint more explicit and causes LLVM to check
the result type with the vector element type.
Reviewers: RKSimon, arsenm, rnk, greened, aemerson
Reviewed By: arsenm
Differential Revision: https://reviews.llvm.org/D62996
llvm-svn: 363240
If the given SCEVExpr has no (un)signed flags attached to it, transfer
these to the resulting instruction or use them to find an existing
instruction.
Differential Revision: https://reviews.llvm.org/D61934
llvm-svn: 362687
As it's causing some bot failures (and per request from kbarton).
This reverts commit r358543/ab70da07286e618016e78247e4a24fcb84077fda.
llvm-svn: 358546
1. Use computed VF for stress testing.
2. If the computed VF does not produce vector code (VF smaller than 2), force VF to be 4.
3. Test vectorization of i64 data on AArch64 to make sure we generate VF != 4 (on X86 that was already tested on AVX).
Patch by Francesco Petrogalli <francesco.petrogalli@arm.com>
Differential Revision: https://reviews.llvm.org/D59952
llvm-svn: 358056
With this change, the VPlan native path is triggered with the directive:
#pragma clang loop vectorize(enable)
There is no need to specify the vectorize_width(N) clause.
Patch by Francesco Petrogalli <francesco.petrogalli@arm.com>
Differential Revision: https://reviews.llvm.org/D57598
llvm-svn: 357156
This changeset is modeled after Intel's submission for SVML. It enables
trigonometry functions vectorization via SLEEF: http://sleef.org/.
* A new vectorization library enum is added to TargetLibraryInfo.h: SLEEF.
* A new option is added to TargetLibraryInfoImpl - ClVectorLibrary: SLEEF.
* A comprehensive test case is included in this changeset.
* In a separate changeset (for clang), a new vectorization library argument is
added to -fveclib: -fveclib=SLEEF.
Trigonometry functions that are vectorized by sleef:
acos
asin
atan
atanh
cos
cosh
exp
exp2
exp10
lgamma
log10
log2
log
sin
sinh
sqrt
tan
tanh
tgamma
Patch by Stefan Teleman
Differential Revision: https://reviews.llvm.org/D53927
llvm-svn: 347510
This patch adds a custom trunc store lowering for v4i8 vector types.
Since there is not v.4b register, the v4i8 is promoted to v4i16 (v.4h)
and default action for v4i8 is to extract each element and issue 4
byte stores.
A better strategy would be to extended the promoted v4i16 to v8i16
(with undef elements) and extract and store the word lane which
represents the v4i8 subvectores. The construction:
define void @foo(<4 x i16> %x, i8* nocapture %p) {
%0 = trunc <4 x i16> %x to <4 x i8>
%1 = bitcast i8* %p to <4 x i8>*
store <4 x i8> %0, <4 x i8>* %1, align 4, !tbaa !2
ret void
}
Can be optimized from:
umov w8, v0.h[3]
umov w9, v0.h[2]
umov w10, v0.h[1]
umov w11, v0.h[0]
strb w8, [x0, #3]
strb w9, [x0, #2]
strb w10, [x0, #1]
strb w11, [x0]
ret
To:
xtn v0.8b, v0.8h
str s0, [x0]
ret
The patch also adjust the memory cost for autovectorization, so the C
code:
void foo (const int *src, int width, unsigned char *dst)
{
for (int i = 0; i < width; i++)
*dst++ = *src++;
}
can be vectorized to:
.LBB0_4: // %vector.body
// =>This Inner Loop Header: Depth=1
ldr q0, [x0], #16
subs x12, x12, #4 // =4
xtn v0.4h, v0.4s
xtn v0.8b, v0.8h
st1 { v0.s }[0], [x2], #4
b.ne .LBB0_4
Instead of byte operations.
llvm-svn: 335735
Summary:
This is a fix for PR23997.
The loop vectorizer is not preserving the inbounds property of GEPs that it creates.
This is inhibiting some optimizations. This patch preserves the inbounds property in
the case where a load/store is being fed by an inbounds GEP.
Reviewers: mkuper, javed.absar, hsaito
Reviewed By: hsaito
Subscribers: dcaballe, hsaito, llvm-commits
Differential Revision: https://reviews.llvm.org/D46191
llvm-svn: 331269
Original commit r311077 of D32871 was reverted in r311304 due to failures
reported in PR34248.
This recommit fixes PR34248 by restricting the packing of predicated scalars
into vectors only when vectorizing, avoiding doing so when unrolling w/o
vectorizing. Added a test derived from the reproducer of PR34248.
llvm-svn: 311849
VPlan is an ongoing effort to refactor and extend the Loop Vectorizer. This
patch introduces the VPlan model into LV and uses it to represent the vectorized
code and drive the generation of vectorized IR.
In this patch VPlan models the vectorized loop body: the vectorized control-flow
is represented using VPlan's Hierarchical CFG, with predication refactored from
being a post-vectorization-step into a vectorization planning step modeling
if-then VPRegionBlocks, and generating code inline with non-predicated code. The
vectorized code within each VPBasicBlock is represented as a sequence of
Recipes, each responsible for modelling and generating a sequence of IR
instructions. To keep the size of this commit manageable the Recipes in this
patch are coarse-grained and capture large chunks of LV's code-generation logic.
The constructed VPlans are dumped in dot format under -debug.
This commit retains current vectorizer output, except for minor instruction
reorderings; see associated modifications to lit tests.
For further details on the VPlan model see docs/Proposals/VectorizationPlan.rst
and its references.
Authors: Gil Rapaport and Ayal Zaks
Differential Revision: https://reviews.llvm.org/D32871
llvm-svn: 311077
Summary:
vectorizer-maximize-bandwidth is generally useful in terms of performance. I've tested the impact of changing this to default on speccpu benchmarks on sandybridge machines. The result shows non-negative impact:
spec/2006/fp/C++/444.namd 26.84 -0.31%
spec/2006/fp/C++/447.dealII 46.19 +0.89%
spec/2006/fp/C++/450.soplex 42.92 -0.44%
spec/2006/fp/C++/453.povray 38.57 -2.25%
spec/2006/fp/C/433.milc 24.54 -0.76%
spec/2006/fp/C/470.lbm 41.08 +0.26%
spec/2006/fp/C/482.sphinx3 47.58 -0.99%
spec/2006/int/C++/471.omnetpp 22.06 +1.87%
spec/2006/int/C++/473.astar 22.65 -0.12%
spec/2006/int/C++/483.xalancbmk 33.69 +4.97%
spec/2006/int/C/400.perlbench 33.43 +1.70%
spec/2006/int/C/401.bzip2 23.02 -0.19%
spec/2006/int/C/403.gcc 32.57 -0.43%
spec/2006/int/C/429.mcf 40.35 +0.27%
spec/2006/int/C/445.gobmk 26.96 +0.06%
spec/2006/int/C/456.hmmer 24.4 +0.19%
spec/2006/int/C/458.sjeng 27.91 -0.08%
spec/2006/int/C/462.libquantum 57.47 -0.20%
spec/2006/int/C/464.h264ref 46.52 +1.35%
geometric mean +0.29%
The regression on 453.povray seems real, but is due to secondary effects as all hot functions are bit-identical with and without the flag.
I started this patch to consult upstream opinions on this. It will be greatly appreciated if the community can help test the performance impact of this change on other architectures so that we can decided if this should be target-dependent.
Reviewers: hfinkel, mkuper, davidxl, chandlerc
Reviewed By: chandlerc
Subscribers: rengolin, sanjoy, javed.absar, bjope, dorit, magabari, RKSimon, llvm-commits, mzolotukhin
Differential Revision: https://reviews.llvm.org/D33341
llvm-svn: 306933
Summary:
vectorizer-maximize-bandwidth is generally useful in terms of performance. I've tested the impact of changing this to default on speccpu benchmarks on sandybridge machines. The result shows non-negative impact:
spec/2006/fp/C++/444.namd 26.84 -0.31%
spec/2006/fp/C++/447.dealII 46.19 +0.89%
spec/2006/fp/C++/450.soplex 42.92 -0.44%
spec/2006/fp/C++/453.povray 38.57 -2.25%
spec/2006/fp/C/433.milc 24.54 -0.76%
spec/2006/fp/C/470.lbm 41.08 +0.26%
spec/2006/fp/C/482.sphinx3 47.58 -0.99%
spec/2006/int/C++/471.omnetpp 22.06 +1.87%
spec/2006/int/C++/473.astar 22.65 -0.12%
spec/2006/int/C++/483.xalancbmk 33.69 +4.97%
spec/2006/int/C/400.perlbench 33.43 +1.70%
spec/2006/int/C/401.bzip2 23.02 -0.19%
spec/2006/int/C/403.gcc 32.57 -0.43%
spec/2006/int/C/429.mcf 40.35 +0.27%
spec/2006/int/C/445.gobmk 26.96 +0.06%
spec/2006/int/C/456.hmmer 24.4 +0.19%
spec/2006/int/C/458.sjeng 27.91 -0.08%
spec/2006/int/C/462.libquantum 57.47 -0.20%
spec/2006/int/C/464.h264ref 46.52 +1.35%
geometric mean +0.29%
The regression on 453.povray seems real, but is due to secondary effects as all hot functions are bit-identical with and without the flag.
I started this patch to consult upstream opinions on this. It will be greatly appreciated if the community can help test the performance impact of this change on other architectures so that we can decided if this should be target-dependent.
Reviewers: hfinkel, mkuper, davidxl, chandlerc
Reviewed By: chandlerc
Subscribers: rengolin, sanjoy, javed.absar, bjope, dorit, magabari, RKSimon, llvm-commits, mzolotukhin
Differential Revision: https://reviews.llvm.org/D33341
llvm-svn: 306336
Summary:
vectorizer-maximize-bandwidth is generally useful in terms of performance. I've tested the impact of changing this to default on speccpu benchmarks on sandybridge machines. The result shows non-negative impact:
spec/2006/fp/C++/444.namd 26.84 -0.31%
spec/2006/fp/C++/447.dealII 46.19 +0.89%
spec/2006/fp/C++/450.soplex 42.92 -0.44%
spec/2006/fp/C++/453.povray 38.57 -2.25%
spec/2006/fp/C/433.milc 24.54 -0.76%
spec/2006/fp/C/470.lbm 41.08 +0.26%
spec/2006/fp/C/482.sphinx3 47.58 -0.99%
spec/2006/int/C++/471.omnetpp 22.06 +1.87%
spec/2006/int/C++/473.astar 22.65 -0.12%
spec/2006/int/C++/483.xalancbmk 33.69 +4.97%
spec/2006/int/C/400.perlbench 33.43 +1.70%
spec/2006/int/C/401.bzip2 23.02 -0.19%
spec/2006/int/C/403.gcc 32.57 -0.43%
spec/2006/int/C/429.mcf 40.35 +0.27%
spec/2006/int/C/445.gobmk 26.96 +0.06%
spec/2006/int/C/456.hmmer 24.4 +0.19%
spec/2006/int/C/458.sjeng 27.91 -0.08%
spec/2006/int/C/462.libquantum 57.47 -0.20%
spec/2006/int/C/464.h264ref 46.52 +1.35%
geometric mean +0.29%
The regression on 453.povray seems real, but is due to secondary effects as all hot functions are bit-identical with and without the flag.
I started this patch to consult upstream opinions on this. It will be greatly appreciated if the community can help test the performance impact of this change on other architectures so that we can decided if this should be target-dependent.
Reviewers: hfinkel, mkuper, davidxl, chandlerc
Reviewed By: chandlerc
Subscribers: rengolin, sanjoy, javed.absar, bjope, dorit, magabari, RKSimon, llvm-commits, mzolotukhin
Differential Revision: https://reviews.llvm.org/D33341
llvm-svn: 305960
If we're shrinking a binary operation, it may be the case that the new
operations wraps where the old didn't. If this happens, the behavior
should be well-defined. So, we can't always carry wrapping flags with us
when we shrink operations.
If we do, we get incorrect optimizations in cases like:
void foo(const unsigned char *from, unsigned char *to, int n) {
for (int i = 0; i < n; i++)
to[i] = from[i] - 128;
}
which gets optimized to:
void foo(const unsigned char *from, unsigned char *to, int n) {
for (int i = 0; i < n; i++)
to[i] = from[i] | 128;
}
Because:
- InstCombine turned `sub i32 %from.i, 128` into
`add nuw nsw i32 %from.i, 128`.
- LoopVectorize vectorized the add to be `add nuw nsw <16 x i8>` with a
vector full of `i8 128`s
- InstCombine took advantage of the fact that the newly-shrunken add
"couldn't wrap", and changed the `add` to an `or`.
InstCombine seems happy to figure out whether we can add nuw/nsw on its
own, so I just decided to drop the flags. There are already a number of
places in LoopVectorize where we rely on InstCombine to clean up.
llvm-svn: 305053
r303763 caused build failures in some out-of-tree tests due to an assertion in
TTI. The original patch updated cost estimates for induction variable update
instructions marked for scalarization. However, it didn't consider that the
incoming value of an induction variable phi node could be a cast instruction.
This caused queries for cast instruction costs with a mix of vector and scalar
types. This patch includes a fix for cast instructions and the test case from
PR33193.
The fix was suggested by Jonas Paulsson <paulsson@linux.vnet.ibm.com>.
Reference: https://bugs.llvm.org/show_bug.cgi?id=33193
Original Differential Revision: https://reviews.llvm.org/D33457
llvm-svn: 304235
For non-uniform instructions marked for scalarization, we should update
`VectorTy` when computing instruction costs to reflect the scalar type. In
addition to determining instruction costs, this type is also used to signal
that all instructions in the loop will be scalarized. This currently affects
memory instructions and non-pointer induction variables and their updates. (We
also mark GEPs scalar after vectorization, but their cost is computed together
with memory instructions.) For scalarized induction updates, this patch also
scales the scalar cost by the vectorization factor, corresponding to each
induction step.
llvm-svn: 303763
David Sherwood