Currently the cost model under-estimates the cost of certain
FP16 conversions.
This patch updates getCastInstrCost to return more accurate costs for
the cases improved in c2ed9fd054.
Reviewed By: dmgreen
Differential Revision: https://reviews.llvm.org/D113700
The costs of vector shifts was 2 as opposed to 1, as the nodes are
marked custom. Fix this like the others and mark the nodes as cheap.
Differential Revision: https://reviews.llvm.org/D120773
In Clang we can attach TBAA metadata based on the load/store intrinsics
based on the operation's element type.
This also contains changes to InstCombine where the AArch64-specific
intrinsics are transformed into generic LLVM load/store operations,
to ensure that all metadata is transferred to the new instruction.
There will be some further work after this patch to also emit TBAA
metadata for SVE's gather/scatter- and struct load/store intrinsics.
Reviewed By: paulwalker-arm
Differential Revision: https://reviews.llvm.org/D119319
The current cost-model overestimates the cost of vector compares &
selects for ordered floating point compares. This patch fixes that by
extending the existing logic for integer predicates.
Reviewed By: dmgreen
Differential Revision: https://reviews.llvm.org/D118256
We have some bitcasts which we know will be simplified,
so their cost is zero.
Reviewed By: david-arm, sdesmalen
Differential Revision: https://reviews.llvm.org/D118019
Generated code resulted in redundant aarch64.sve.convert.to.svbool
calls for AArch64 Binary Operations. Narrow the more precise operands
instead of widening the less precise operands
Differential Revision: https://reviews.llvm.org/D116730
A lot of neon intrinsics work lane-wise, meaning that non-demanded
elements in and not demanded out. This teaches that to
AArch64TTIImpl::simplifyDemandedVectorEltsIntrinsic for some simple
single-input truncate intrinsics, which can help remove unnecessary
instructions.
Differential Revision: https://reviews.llvm.org/D117097
If we are inserting into or extracting from a scalable vector we do
not know the number of elements at runtime, so we can only let the
index wrap for fixed-length vectors.
Tests added here:
Analysis/CostModel/AArch64/sve-insert-extract.ll
Differential Revision: https://reviews.llvm.org/D117099
Similar to D116732, this adds basic scalar sadd_with_overflow,
uadd_with_overflow, ssub_with_overflow and usub_with_overflow costs for
aarch64, which are usually quite efficiently lowered.
Differential Revision: https://reviews.llvm.org/D116734
This adds some AArch64 specific smul_with_overflow and umul_with_overflow
costs, overriding the default costs. The code generation for these mul
with overflow intrinsics is usually better than the default expansion on
AArch64. The costs come from https://godbolt.org/z/zEzYhMWqo with various
types, or llvm/test/CodeGen/AArch64/arm64-xaluo.ll.
Differential Revision: https://reviews.llvm.org/D116732
This patch adds on an overhead cost for gathers and scatters, which
is a rough estimate based on performance investigations I have
performed on SVE hardware for various micro-benchmarks.
Differential Revision: https://reviews.llvm.org/D115143
This patch extends the "is all active predicate" check to cover
cases where the predicate is casted but in a way that doesn't
change its "all active" status.
Differential Revision: https://reviews.llvm.org/D115047
We can not swap multiplicand and multiplier because the sve intrinsics
are predicated. Imagine lanes in vectors having the following values:
pg = 0
multiplicand = 1 (from dup)
multiplier = 2
The resulting value should be 1, but if we swap multiplicand and multiplier it will become 2,
which is incorrect.
Differential Revision: https://reviews.llvm.org/D114577
InstCombine AArch64 LD1/ST1 to llvm.masked.load/llvm.masked.store
and LD1/ST1 to load/store when a ptrue all predicate pattern operand
is present.
This allows existing IR optimizations such as dead-load removal to
occur.
Differential Revision: https://reviews.llvm.org/D113489
This adds support for SVE structured loads/stores to the relevant target
hooks, such that we can support these instructions in the InterleavedAccess
pass.
Depends on D112078
Differential Revision: https://reviews.llvm.org/D112303
Combine FADD and FMUL intrinsics into FMA when the result of the FMUL is an FADD operand
with one only use and both use the same predicate.
Differential Revision: https://reviews.llvm.org/D111638
This patch introduces a new function:
AArch64Subtarget::getVScaleForTuning
that returns a value for vscale that can be used for tuning the cost
model when using scalable vectors. The VScaleForTuning option in
AArch64Subtarget is initialised according to the following rules:
1. If the user has specified the CPU to tune for we use that, else
2. If the target CPU was specified we use that, else
3. The tuning is set to "generic".
For CPUs of type "generic" I have assumed that vscale=2.
New tests added here:
Analysis/CostModel/AArch64/sve-gather.ll
Analysis/CostModel/AArch64/sve-scatter.ll
Transforms/LoopVectorize/AArch64/sve-strict-fadd-cost.ll
Differential Revision: https://reviews.llvm.org/D110259
This patch adds further support for vectorisation of loops that involve
selecting an integer value based on a previous comparison. Consider the
following C++ loop:
int r = a;
for (int i = 0; i < n; i++) {
if (src[i] > 3) {
r = b;
}
src[i] += 2;
}
We should be able to vectorise this loop because all we are doing is
selecting between two states - 'a' and 'b' - both of which are loop
invariant. This just involves building a vector of values that contain
either 'a' or 'b', where the final reduced value will be 'b' if any lane
contains 'b'.
The IR generated by clang typically looks like this:
%phi = phi i32 [ %a, %entry ], [ %phi.update, %for.body ]
...
%pred = icmp ugt i32 %val, i32 3
%phi.update = select i1 %pred, i32 %b, i32 %phi
We already detect min/max patterns, which also involve a select + cmp.
However, with the min/max patterns we are selecting loaded values (and
hence loop variant) in the loop. In addition we only support certain
cmp predicates. This patch adds a new pattern matching function
(isSelectCmpPattern) and new RecurKind enums - SelectICmp & SelectFCmp.
We only support selecting values that are integer and loop invariant,
however we can support any kind of compare - integer or float.
Tests have been added here:
Transforms/LoopVectorize/AArch64/sve-select-cmp.ll
Transforms/LoopVectorize/select-cmp-predicated.ll
Transforms/LoopVectorize/select-cmp.ll
Differential Revision: https://reviews.llvm.org/D108136
This patch adds further support for vectorisation of loops that involve
selecting an integer value based on a previous comparison. Consider the
following C++ loop:
int r = a;
for (int i = 0; i < n; i++) {
if (src[i] > 3) {
r = b;
}
src[i] += 2;
}
We should be able to vectorise this loop because all we are doing is
selecting between two states - 'a' and 'b' - both of which are loop
invariant. This just involves building a vector of values that contain
either 'a' or 'b', where the final reduced value will be 'b' if any lane
contains 'b'.
The IR generated by clang typically looks like this:
%phi = phi i32 [ %a, %entry ], [ %phi.update, %for.body ]
...
%pred = icmp ugt i32 %val, i32 3
%phi.update = select i1 %pred, i32 %b, i32 %phi
We already detect min/max patterns, which also involve a select + cmp.
However, with the min/max patterns we are selecting loaded values (and
hence loop variant) in the loop. In addition we only support certain
cmp predicates. This patch adds a new pattern matching function
(isSelectCmpPattern) and new RecurKind enums - SelectICmp & SelectFCmp.
We only support selecting values that are integer and loop invariant,
however we can support any kind of compare - integer or float.
Tests have been added here:
Transforms/LoopVectorize/AArch64/sve-select-cmp.ll
Transforms/LoopVectorize/select-cmp-predicated.ll
Transforms/LoopVectorize/select-cmp.ll
Differential Revision: https://reviews.llvm.org/D108136
Fix static analysis warning that we check for null Entry after dereferencing it.
I don't think this can actually happen as i8/i16 should legalize to use the i32 path which should return a cost - but I'd rather play it safe that rely on an implicit type legalization.
Several FP instructions (fadd, fsub, etc.) were incorrectly assigned
a higher cost for SVE because they have custom lowering, however we
know they are legal. This patch explicitly assigns a cost of 2 to
these opcodes.
Tests added here:
Analysis/CostModel/AArch64/arith-fp-sve.ll
Differential Revision: https://reviews.llvm.org/D108993
This patch solely moves convert operation between SVE predicate pattern
and element number into two small functions. It's pre-commit patch for optimize
pture with known sve register width.
Differential Revision: https://reviews.llvm.org/D108705
Tell the cost model to use the scalable calculation for non-neon fixed vector.
This results in a cheaper cost for fixed-length SVE masked gathers/scatters
allowing the vectorizor to emit them more frequently.
For tight loops like this:
float r = 0;
for (int i = 0; i < n; i++) {
r += a[i];
}
it's better not to vectorise at -O3 using fixed-width ordered reductions
on AArch64 targets. Although the resulting number of instructions in the
generated code ends up being comparable to not vectorising at all, there
may be additional costs on some CPUs, for example perhaps the scheduling
is worse. It makes sense to deter vectorisation in tight loops.
Differential Revision: https://reviews.llvm.org/D108292
Removed AArch64 usage of the getMaxVScale interface, replacing it with
the vscale_range(min, max) IR Attribute.
Reviewed By: paulwalker-arm
Differential Revision: https://reviews.llvm.org/D106277
Replace vector unpack operation with a scalar extend operation.
unpack(splat(X)) --> splat(extend(X))
If we have both, unpkhi and unpklo, for the same vector then we may
save a register in some cases, e.g:
Hi = unpkhi (splat(X))
Lo = unpklo(splat(X))
--> Hi = Lo = splat(extend(X))
Differential Revision: https://reviews.llvm.org/D106929
Change-Id: I77c5c201131e3a50de1cdccbdcf84420f5b2244b
Move the last{a,b} operation to the vector operand of the binary instruction if
the binop's operand is a splat value. This essentially converts the binop
to a scalar operation.
Example:
// If x and/or y is a splat value:
lastX (binop (x, y)) --> binop(lastX(x), lastX(y))
Differential Revision: https://reviews.llvm.org/D106932
Change-Id: I93ff5302f9a7972405ee0d3854cf115f072e99c0
This takes the existing SVE costing for the various min/max reduction
intrinsics and expands it to NEON, where I believe it applies equally
well.
In the process it changes the lowering to use min/max cost, as opposed
to summing up the cost of ICmp+Select.
Differential Revision: https://reviews.llvm.org/D106239
I'm not sure this is the best way to approach this,
but the situation is rather not very detectable unless we explicitly call it out when refusing to advise to unroll.
Reviewed By: efriedma
Differential Revision: https://reviews.llvm.org/D107271
I have added a new FastMathFlags parameter to getArithmeticReductionCost
to indicate what type of reduction we are performing:
1. Tree-wise. This is the typical fast-math reduction that involves
continually splitting a vector up into halves and adding each
half together until we get a scalar result. This is the default
behaviour for integers, whereas for floating point we only do this
if reassociation is allowed.
2. Ordered. This now allows us to estimate the cost of performing
a strict vector reduction by treating it as a series of scalar
operations in lane order. This is the case when FP reassociation
is not permitted. For scalable vectors this is more difficult
because at compile time we do not know how many lanes there are,
and so we use the worst case maximum vscale value.
I have also fixed getTypeBasedIntrinsicInstrCost to pass in the
FastMathFlags, which meant fixing up some X86 tests where we always
assumed the vector.reduce.fadd/mul intrinsics were 'fast'.
New tests have been added here:
Analysis/CostModel/AArch64/reduce-fadd.ll
Analysis/CostModel/AArch64/sve-intrinsics.ll
Transforms/LoopVectorize/AArch64/strict-fadd-cost.ll
Transforms/LoopVectorize/AArch64/sve-strict-fadd-cost.ll
Differential Revision: https://reviews.llvm.org/D105432
This adds some missing single source shuffle costs for AArch64, of i16
and i8 vectors. v4i16 are the same as v4i32 with a worse case cost of 3
coming from the perfect shuffle tables. The larger vector sizes expand
into a constant pool, plus a load (and adrp) and a tbl. I arbitrarily
chose 8 for the cost to be expensive but not too expensive.
Differential Revision: https://reviews.llvm.org/D106241
This changes the cost to (LT.first-1) * cost(add) + 2, where the cost of
an add is assumed to be 1. This brings it inline with the other
reductions.
Differential Revision: https://reviews.llvm.org/D106240
At the moment, <vscale x 1 x eltty> are not yet fully handled by the
code-generator, so to avoid vectorizing loops with that VF, we mark the
cost for these types as invalid.
The reason for not adding a new "TTI::getMinimumScalableVF" is because
the type is supposed to be a type that can be legalized. It partially is,
although the support for these types need some more work.
Reviewed By: paulwalker-arm, dmgreen
Differential Revision: https://reviews.llvm.org/D103882
This patch removes the IsPairwiseForm flag from the Reduction Cost TTI
hooks, along with some accompanying code for pattern matching reductions
from trees starting at extract elements. IsPairWise is now assumed to be
false, which was the predominant way that the value was used from both
the Loop and SLP vectorizers. Since the adjustments such as D93860, the
SLP vectorizer has not relied upon this distinction between paiwise and
non-pairwise reductions.
This also removes some code that was detecting reductions trees starting
from extract elements inside the costmodel. This case was
double-counting costs though, adding the individual costs on the
individual instruction _and_ the total cost of the reduction. Removing
it changes the costs in llvm/test/Analysis/CostModel/X86/reduction.ll to
not double count. The cost of reduction intrinsics is still tested
through the various tests in
llvm/test/Analysis/CostModel/X86/reduce-xyz.ll.
Differential Revision: https://reviews.llvm.org/D105484
This patch adds a TTI function, isElementTypeLegalForScalableVector, to query
whether it is possible to vectorize a given element type. This is called by
isLegalToVectorizeInstTypesForScalable to reject scalable vectorization if
any of the instruction types in the loop are unsupported, e.g:
int foo(__int128_t* ptr, int N)
#pragma clang loop vectorize_width(4, scalable)
for (int i=0; i<N; ++i)
ptr[i] = ptr[i] + 42;
This example currently crashes if we attempt to vectorize since i128 is not a
supported type for scalable vectorization.
Reviewed By: sdesmalen, david-arm
Differential Revision: https://reviews.llvm.org/D102253
This patch adds a new ShuffleKind SK_Splice and then handle the cost in
getShuffleCost, as in experimental.vector.reverse.
Differential Revision: https://reviews.llvm.org/D104630
Loads of <4 x i8> vectors were modeled as extremely expensive. And while we
don't have a load instruction that supports this, it isn't that expensive to
create a vector of i8 elements. The codegen for this was fixed/optimised in
D105110. This now tweaks the cost model and enables SLP vectorisation of my
motivating case loadi8.ll.
Differential Revision: https://reviews.llvm.org/D103629
OR, XOR and AND entries are added to the cost table. An extra cost
is added when vector splitting occurs.
This is done to address the issue of a missed SLP vectorization
opportunity due to unreasonably high costs being attributed to the vector
Or reduction (see: https://bugs.llvm.org/show_bug.cgi?id=44593).
Differential Revision: https://reviews.llvm.org/D104538
Added a case for CTPOP to AArch64TTIImpl::getIntrinsicInstrCost so that
the cost estimate matches the codegen in
test/CodeGen/AArch64/arm64-vpopcnt.ll
Differential Revision: https://reviews.llvm.org/D103952
We were passing the RecurrenceDescriptor by value to most of the reduction analysis methods, despite it being rather bulky with TrackingVH members (that can be costly to copy). In all these cases we're only using the RecurrenceDescriptor for rather basic purposes (access to types/kinds etc.).
Differential Revision: https://reviews.llvm.org/D104029
Fixes getTypeConversion to return `TypeScalarizeScalableVector` when a scalable vector
type cannot be legalized by widening/splitting. When this is the method of legalization
found, getTypeLegalizationCost will return an Invalid cost.
The getMemoryOpCost, getMaskedMemoryOpCost & getGatherScatterOpCost functions already call
getTypeLegalizationCost and will now also return an Invalid cost for unsupported types.
Reviewed By: sdesmalen, david-arm
Differential Revision: https://reviews.llvm.org/D102515
Use llvm.experimental.vector.insert instead of storing into an alloca
when generating code for these intrinsics. This defers the codegen of
the generated vector to instruction selection, allowing existing
shufflevector style optimizations to apply.
Additionally, introduce a new target transform that can recognise fixed
predicate patterns in the svbool variants of these intrinsics.
Differential Revision: https://reviews.llvm.org/D103082
Due to the dependency on runtime unrolling, UnJ is only
enabled by default on in-order scheduling models,
and if a cpu is specified through -mcpu.
Differential Revision: https://reviews.llvm.org/D103604
Follow up to D101357 / 3fa6510f6.
Supersedes D102330.
Goal: Use flags setting rdffrs instead of rdffr + ptest.
Problem: RDFFR_P doesn't have have a flags setting equivalent.
Solution: in instcombine, canonicalize to RDFFR_PP at the IR level, and
rely on RDFFR_PP+PTEST => RDFFRS_PP optimization in
AArch64InstrInfo::optimizePTestInstr.
While here:
* Test that rdffr.z+ptest generates a rdffrs.
* Use update_{test,llc}_checks.py on the tests.
* Use sve attribute on functions.
Differential Revision: https://reviews.llvm.org/D102623
Added an extra analysis for better choosing of shuffle kind in
getShuffleCost functions for better cost estimation if mask was
provided.
Differential Revision: https://reviews.llvm.org/D100865
Added an extra analysis for better choosing of shuffle kind in
getShuffleCost functions for better cost estimation if mask was
provided.
Differential Revision: https://reviews.llvm.org/D100865
By converting the SVE intrinsic to a normal LLVM insertelement we give
the code generator a better chance to remove transitions between GPRs
and VPRs
Co-authored-by: Paul Walker <paul.walker@arm.com>
Depends on D101302
Differential Revision: https://reviews.llvm.org/D101167
As part of this the ptrue coalescing done in SVEIntrinsicOpts has been
modified to not introduce redundant converts, since the convert removal
will no longer run after that optimisation to clean up.
Differential Revision: https://reviews.llvm.org/D101302
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
We were missing some instruction costs when converting vectors of
floating point half types into integers, so I've added those here.
I also manually generated assembly code for each FP->int case and
looked at the number of instructions generated, which meant
adjusting some of the existing costs too.
I've updated an existing test to reflect the new costs:
Analysis/CostModel/AArch64/sve-fptoi.ll
Differential Revision: https://reviews.llvm.org/D99935
Introduced the cost of thre reverse shuffles for AArch64, currently just
copied the costs for PermuteSingleSrc.
Differential Revision: https://reviews.llvm.org/D100871
when the predicate used by last{a,b} specifies a known vector length.
For example:
aarch64_sve_lasta(VL1, D) -> extractelement(D, #1)
aarch64_sve_lastb(VL1, D) -> extractelement(D, #0)
Co-authored-by: Paul Walker <paul.walker@arm.com>
Differential Revision: https://reviews.llvm.org/D100476
At the moment, getMemoryOpCost returns 1 for all inputs if CostKind is
CodeSize or SizeAndLatency. This fools LoopUnroll into thinking memory
operations on large vectors have a cost of one, even if they will get
expanded to a large number of memory operations in the backend.
This patch updates getMemoryOpCost to return the cost for the type
legalization for both CodeSize and SizeAndLatency. This should more
accurately reflect the number of memory operations required.
I am not sure how latency should properly be included in SizeAndLatency
from the description, but returning the size cost should be clearly more
accurate.
This does not cause any binary changes when building
MultiSource/SPEC2000/SPEC2006 with -O3 -flto for AArch64, likely because
large vector memops are not really formed by code emitted from Clang.
But using the C/C++ matrix extension can easily result in code with very
large vector operations directly from Clang, e.g.
https://clang.godbolt.org/z/6xzxcTGvb
Reviewed By: samparker
Differential Revision: https://reviews.llvm.org/D100291
David Green