This is my first LLVM patch, so please tell me if there are any process issues.
The main observation for this patch is that we can lower UMIN/UMAX with v8i16 by using unsigned saturated subtractions in a clever way. Previously this operation was lowered by turning the signbit of both inputs and the output which turns the unsigned minimum/maximum into a signed one.
We could use this trick in reverse for lowering SMIN/SMAX with v16i8 instead. In terms of latency/throughput this is the needs one large move instruction. It's just that the sign bit turning has an increased chance of being optimized further. This is particularly apparent in the "reduce" test cases. However due to the slight regression in the single use case, this patch no longer proposes this.
Unfortunately this argument also applies in reverse to the new lowering of UMIN/UMAX with v8i16 which regresses the "horizontal-reduce-umax", "horizontal-reduce-umin", "vector-reduce-umin" and "vector-reduce-umax" test cases a bit with this patch. Maybe some extra casework would be possible to avoid this. However independent of that I believe that the benefits in the common case of just 1 to 3 chained min/max instructions outweighs the downsides in that specific case.
Patch By: @TomHender (Tom Hender) ActuallyaDeviloper
Differential Revision: https://reviews.llvm.org/D87236
Changes TTI function getIntImmCostInst to take an additional Instruction parameter,
which enables us to be able to check it is part of a min(max())/max(min()) pattern that will match SSAT.
We can then mark the constant used as free to prevent it being hoisted so SSAT can still be generated.
Required minor changes in some non-ARM backends to allow for the optional parameter to be included.
Differential Revision: https://reviews.llvm.org/D87457
Other types can be handled in future patches but their uniform / non-uniform costs are more similar and don't appear to cause many vectorization issues.
Currently, getCastInstrCost has limited information about the cast it's
rating, often just the opcode and types. Sometimes there is a context
instruction as well, but it isn't trustworthy: for instance, when the
vectorizer is rating a plan, it calls getCastInstrCost with the old
instructions when, in fact, it's trying to evaluate the cost of the
instruction post-vectorization. Thus, the current system can get the
cost of certain casts incorrect as the correct cost can vary greatly
based on the context in which it's used.
For example, if the vectorizer queries getCastInstrCost to evaluate the
cost of a sext(load) with tail predication enabled, getCastInstrCost
will think it's free most of the time, but it's not always free. On ARM
MVE, a VLD2 group cannot be extended like a normal VLDR can. Similar
situations can come up with how masked loads can be extended when being
split.
To fix that, this path adds a new parameter to getCastInstrCost to give
it a hint about the context of the cast. It adds a CastContextHint enum
which contains the type of the load/store being created by the
vectorizer - one for each of the types it can produce.
Original patch by Pierre van Houtryve
Differential Revision: https://reviews.llvm.org/D79162
These cost methods don't make much sense in X86Subtarget. Make
them methods in X86's TTI and move the feature checks from the
X86Subtarget constructor into these methods.
Reviewed By: RKSimon
Differential Revision: https://reviews.llvm.org/D84594
The main interface has been migrated to Align already but a few backends where broadening the type from Align to MaybeAlign.
This patch makes sure all implementations conform to the public API.
Differential Revision: https://reviews.llvm.org/D82465
Summary:
Get back `const` partially lost in one of recent changes.
Additionally specify explicit qualifiers in few places.
Reviewers: samparker
Reviewed By: samparker
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D82383
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
Add the remaining arithmetic opcodes into the generic implementation
of getUserCost and then call this from getInstructionThroughput. Most
of the backends have been modified to return the base implementation
for cost kinds other RecipThroughput. The outlier here is AMDGPU
which already uses getArithmeticInstrCost for all the cost kinds.
This change means that most of the opcodes can be removed from that
backends implementation of getUserCost.
Differential Revision: https://reviews.llvm.org/D80992
Add cases for icmp, fcmp and select into the switch statement of the
generic getUserCost implementation with getInstructionThroughput then
calling into it. The BasicTTI and backend implementations have be set
to return a default value (1) when a cost other than throughput is
being queried.
Differential Revision: https://reviews.llvm.org/D80550
Use getMemoryOpCost from the generic implementation of getUserCost
and have getInstructionThroughput return the result of that for loads
and stores.
This also means that the X86 implementation of getUserCost can be
removed with the functionality folded into its getMemoryOpCost.
Differential Revision: https://reviews.llvm.org/D80984
Add the remaining cast instruction opcodes to the base implementation
of getUserCost and directly return the result. This allows
getInstructionThroughput to return getUserCost for the casts. This
has required changes to PPC and SystemZ because they implement
getUserCost and/or getCastInstrCost with adjustments for vector
operations. Adjusts have also been made in the remaining backends
that implement the method so that they still produce a cost of zero
or one for cost kinds other than throughput.
Differential Revision: https://reviews.llvm.org/D79848
Recommitting most of the remaining changes from
259eb619ff, but excluding the call to
getUserCost from getInstructionThroughput. Though there's still no
test changes, I doubt that this is an NFC...
With the two getIntrinsicInstrCosts folded into one, now fold in the
scalar/code-size orientated getIntrinsicCost. The remaining scalar
intrinsics were memcpy, cttz and ctlz which now have special handling
in the BasicTTI implementation.
This had required a change in the AMDGPU backend for fabs as it
should always be 'free'. I've also changed the X86 backend to return
the BaseT implementation when the CostKind isn't RecipThroughput.
Differential Revision: https://reviews.llvm.org/D80012
With the two getIntrinsicInstrCosts folded into one, now fold in the
scalar/code-size orientated getIntrinsicCost. This involved sinking
cost of the TTIImpl into the base implementation, as it performs no
target checks. The opcodes remaining were memcpy, cttz and ctlz which
now have special handling in the BasicTTI implementation.
getInstructionThroughput can now directly return the result of
getUserCost.
This had required a change in the AMDGPU backend for fabs and its
always 'free'. I've also changed the X86 backend to return '1' for
any intrinsic when the CostKind isn't RecipThroughput.
Though this intended to be a non-functional change, there are many
paths being combined here so I would be very surprised if this didn't
have an effect.
Differential Revision: https://reviews.llvm.org/D80012
Combine the two API calls into one by introducing a structure to hold
the relevant data. This has the added benefit of moving the boiler
plate code for arguments and flags, into the constructors. This is
intended to be a non-functional change, but the complicated web of
logic involved here makes it very hard to guarantee.
Differential Revision: https://reviews.llvm.org/D79941
getScalarizationOverhead is only ever called with vectors (and we already had a load of cast<VectorType> calls immediately inside the functions).
Followup to D78357
Reviewed By: @samparker
Differential Revision: https://reviews.llvm.org/D79341
Make the kind of cost explicit throughout the cost model which,
apart from making the cost clear, will allow the generic parts to
calculate better costs. It will also allow some backends to
approximate and correlate the different costs if they wish. Another
benefit is that it will also help simplify the cost model around
immediate and intrinsic costs, where we currently have multiple APIs.
RFC thread:
http://lists.llvm.org/pipermail/llvm-dev/2020-April/141263.html
Differential Revision: https://reviews.llvm.org/D79002
Also fix some cost tables for vXi1 types to match the costs entries for the types they will be promoted to.
Differential Revision: https://reviews.llvm.org/D79045
vpermw is 2 uops. vpermt2b/vpermt2w are two shuffle uops and a port 015 uop. Weirdly vpermb is a single uop.
This patch bumps the cost to 2 for these operations. Maybe should go to 3 for the vpermt2*, but I've started conservative.
I've also removed a few entries that were now the same as earlier subtargets or that I didn't think we really did. Like I don't think we extend v32i8 to v32i16, shuffle, and then truncate.
Differential Revision: https://reviews.llvm.org/D79148
We generate much better code these days than we used to. And we use the same sequence for AVX1 and AVX2 for these
For v4i64->v4i32 we generate:
vextractf128 xmm1, ymm0, 1
vshufps xmm0, xmm0, xmm1, 136 # xmm0 = xmm0[0,2],xmm1[0,2]
And for v8i64->v8i32 we generate:
vperm2f128 ymm2, ymm0, ymm1, 49 # ymm2 = ymm0[2,3],ymm1[2,3]
vinsertf128 ymm0, ymm0, xmm1, 1
vshufps ymm0, ymm0, ymm2, 136 # ymm0 = ymm0[0,2],ymm2[0,2],ymm0[4,6],ymm2[4,6]
Differential Revision: https://reviews.llvm.org/D79109
The improvements to the x86 vector insert/extract element costs in D74976 resulted in the estimated costs for vector initialization and scalarization increasing higher than should be expected. This is particularly noticeable on pre-SSE4 targets where the available of legal INSERT_VECTOR_ELT ops is more limited.
This patch does 2 things:
1 - it implements X86TTIImpl::getScalarizationOverhead to more accurately represent the typical costs of a ISD::BUILD_VECTOR pattern.
2 - it adds a DemandedElts mask to getScalarizationOverhead to permit the SLP's BoUpSLP::getGatherCost to be rewritten to use it directly instead of accumulating raw vector insertion costs.
This fixes PR45418 where a v4i8 (zext'd to v4i32) was no longer vectorizing.
A future patch should extend X86TTIImpl::getScalarizationOverhead to tweak the EXTRACT_VECTOR_ELT scalarization costs as well.
Reviewed By: @craig.topper
Differential Revision: https://reviews.llvm.org/D78216
I've modified isTruncateFree to get an accurate cost for types that need to be split. I'm planning to look into fixing it for all vectors, but need more cost cleanups first.
Differential Revision: https://reviews.llvm.org/D78973
There are several different types of cost that TTI tries to provide
explicit information for: throughput, latency, code size along with
a vague 'intersection of code-size cost and execution cost'.
The vectorizer is a keen user of RecipThroughput and there's at least
'getInstructionThroughput' and 'getArithmeticInstrCost' designed to
help with this cost. The latency cost has a single use and a single
implementation. The intersection cost appears to cover most of the
rest of the API.
getUserCost is explicitly called from within TTI when the user has
been explicit in wanting the code size (also only one use) as well
as a few passes which are concerned with a mixture of size and/or
a relative cost. In many cases these costs are closely related, such
as when multiple instructions are required, but one evident diverging
cost in this function is for div/rem.
This patch adds an argument so that the cost required is explicit,
so that we can make the important distinction when necessary.
Differential Revision: https://reviews.llvm.org/D78635
All avx512 truncate instructions except vXi64->vXi32 are 2 uops
on port 5. So raise their costs to 2. Except when we have an
earlier faster sequence like pshufb for 128 bit input vectors.
Add a lower cost of 3 v16i16->v16i8 with avx512f where we can
extend to v16i32 then truncate. And a cost of 2 for avx512bw with
and without avx512vl. There we can use vpmovwb with either a ymm
or zmm input. Both of these beat masking, splitting, and using
packuswb which is our avx/avx2 codegen.
We're currently getting this from the default implementation. But
I don't like how the cost model came to this answer and I might
be making some changes there.
If one of caller/callee has disabled ZMM registers due to
prefer-vector-width=256, we were previously
disabling argument promotion as the ABI might be incompatible since
one side will split 512-bit vectors in this case.
But if we can see that the types are all scalar this shouldn't be
a problem.
This patch assumes that pointer element type reflects the type that
the argument will be promoted to.
Differential Revision: https://reviews.llvm.org/D78770
The API for shuffles and reductions uses generic Type parameters,
instead of VectorType, and so assertions and casts are used a lot.
This patch makes those types explicit, which means that the clients
can't be lazy, but results in less ambiguity, and that can only be a
good thing.
Bugzilla: https://bugs.llvm.org/show_bug.cgi?id=45562
Differential Revision: https://reviews.llvm.org/D78357
Summary:
Remove usages of asserting vector getters in Type in preparation for the
VectorType refactor. The existence of these functions complicates the
refactor while adding little value.
Reviewers: craig.topper, sdesmalen, efriedma, RKSimon
Reviewed By: efriedma
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D77264
This moves v32i16/v64i8 to a model consistent with how we
treat integer types with avx1.
This does change the ABI for types vXi16/vXi8 vectors larger than
512 bits to pass in multiple zmms instead of multiple ymms. We'd
already hacked some code to make v64i8/v32i16 pass in zmm.
Cost model is still a bit of a mess. In some place I tried to
match existing behavior. But really we need to account for
splitting and concating costs. Cost model for shuffles is
especially pessimistic.
Differential Revision: https://reviews.llvm.org/D76212
If we're inserting into v2i8/v4i8/v8i8/v2i16/v4i16 style sub-128bit vectors ensure we don't use the SK_PermuteTwoSrc cost of the legalized value type - this is a followup to rG12c629ec6c59 which added equivalent sub-128bit shuffle costs
This is similar to what I recently did for getArithmeticReductionCost.
I'm trying to account for the narrowing from 512->256->128 as we go.
I've also added a new helper method getMinMaxCost that tries to
handle the cases where we have native min/max instructions and
fall back to cmp+select when we don't.
Differential Revision: https://reviews.llvm.org/D76634
v2i8/v4i8/v8i8 + v2i16/v4i16 all show up in vectorizer code and by just using the legalized types (v16i8/v8i16) we're highly exaggerating the actual cost of the shuffle.
Summary:
This allows doing `memcmp(p, q, 7)` with 2 loads instead of a call to
memcmp.
This fixes part of PR45147.
Reviewers: spatel
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D76133
This patch attempts to more accurately model the reduction of
power of 2 vectors of types we natively support. This takes into
account the narrowing of vectors that occur as we go from 512
bits to 256 bits, to 128 bits. It also takes into account the use
of wider elements in the shuffles for the first 2 steps of a
reduction from 128 bits. And uses a v8i16 shift for the final step
of vXi8 reduction.
The default implementation uses the legalized type for the arithmetic
for all levels. And uses the single source permute cost of the
legalized type for all levels. This penalizes things like
lack of v16i8 pshufb on pre-sse3 targets and the splitting and
joining that needs to be done for integer types on AVX1. We never
need v16i8 shuffle for a reduction and we only need split AVX1 ops
when type the type wide and needs to be split. I think we're still
over costing splits and joins for AVX1, but we're closer now.
I've also removed all pairwise special casing because I don't
think we ever want to generate that on X86. I've also adjusted
the add handling to more accurately account for any type splitting
that occurs before we reach a legal type.
Differential Revision: https://reviews.llvm.org/D76478
Previously we multiplied the cost for the table entries by the number of splits needed. But that implies that each split goes through a reduction to scalar independently. I think what really happens is that the we AND/OR the split pieces until we're down to a single value with a legal type and then do special reduction sequence on that.
So to model that this patch takes the number of splits minus one multiplied by the cost of a AND/OR at the legal element count and adds that on top of the table lookup.
Differential Revision: https://reviews.llvm.org/D76400
Refines the gather/scatter cost model, but also changes the TTI
function getIntrinsicInstrCost to accept an additional parameter
which is needed for the gather/scatter cost evaluation.
This did require trivial changes in some non-ARM backends to
adopt the new parameter.
Extending gathers and truncating scatters are now priced cheaper.
Differential Revision: https://reviews.llvm.org/D75525
This tries to improve the accuracy of extract/insert element costs by accounting for subvector extraction/insertion for >128-bit vectors and the shuffling of elements to/from the 0'th index.
It also adds INSERTPS for f32 types and PINSR/PEXTR costs for integer types (at the moment we assume the same cost as MOVD/MOVQ - which isn't always true).
Differential Revision: https://reviews.llvm.org/D74976
D74976 will handle larger vector types, but since SLM doesn't support AVX+ then we will always be extracting from 128-bit vectors so don't need to scale the cost.
We seem to be inheriting the cost from sse4.1. But if we have 256-bit registers we should be able to do this with just one extract to split the 16i16 and two v8i16->v8i32 operations so our cost should be 3 not 4.
Differential Revision: https://reviews.llvm.org/D73646
Summary:
Previously we did this with isel patterns that used garbage in
the widened part of the source. But that's not valid for strictfp.
So now we custom widen and use zeroes for the widened elemens for
strictfp.
This replaces D71864.
Reviewers: RKSimon, spatel, andrew.w.kaylor, pengfei, LiuChen3
Reviewed By: pengfei
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D71879
Add an extra parameter so alignment can be taken under
consideration in gather/scatter legalization.
Differential Revision: https://reviews.llvm.org/D71610
Soon Intrinsic::ID will be a plain integer, so this overload will not be
possible.
Rename both overloads to ensure that downstream targets observe this as
a build failure instead of a runtime failure.
Split off from D71320
Reviewers: efriedma
Differential Revision: https://reviews.llvm.org/D71381
This attempts to teach the cost model in Arm that code such as:
%s = shl i32 %a, 3
%a = and i32 %s, %b
Can under Arm or Thumb2 become:
and r0, r1, r2, lsl #3
So the cost of the shift can essentially be free. To do this without
trying to artificially adjust the cost of the "and" instruction, it
needs to get the users of the shl and check if they are a type of
instruction that the shift can be folded into. And so it needs to have
access to the actual instruction in getArithmeticInstrCost, which if
available is added as an extra parameter much like getCastInstrCost.
We otherwise limit it to shifts with a single user, which should
hopefully handle most of the cases. The list of instruction that the
shift can be folded into include ADC, ADD, AND, BIC, CMP, EOR, MVN, ORR,
ORN, RSB, SBC and SUB. This translates to Add, Sub, And, Or, Xor and
ICmp.
Differential Revision: https://reviews.llvm.org/D70966
This is a follow-up to D70607 where we made any
extract element on SLM more costly than default. But that is
pessimistic for extract from element 0 because that corresponds
to x86 movd/movq instructions. These generally have >1 cycle
latency, but they are probably implemented as single uop
instructions.
Note that no vectorization tests are affected by this change.
Also, no targets besides SLM are affected because those are
falling through to the default cost of 1 anyway. But this will
become visible/important if we add more specializations via cost
tables.
Differential Revision: https://reviews.llvm.org/D71023
I'm not sure what the effect of this change will be on all of the affected
tests or a larger benchmark, but it fixes the horizontal add/sub problems
noted here:
https://reviews.llvm.org/D59710?vs=227972&id=228095&whitespace=ignore-most#toc
The costs are based on reciprocal throughput numbers in Agner's tables for
PEXTR*; these appear to be very slow ops on Silvermont.
This is a small step towards the larger motivation discussed in PR43605:
https://bugs.llvm.org/show_bug.cgi?id=43605
Also, it seems likely that insert/extract is the source of perf regressions on
other CPUs (up to 30%) that were cited as part of the reason to revert D59710,
so maybe we'll extend the table-based approach to other subtargets.
Differential Revision: https://reviews.llvm.org/D70607
This is no longer needed after widening legalization as we
custom legalize v8i8 ourselves.
Added entries to the cost model, but bumped the cost slightly
to account for the truncate shuffle that wasn't costed before.
This better represents the kshift+binop we'd get for each stage
before the final extract. Its likely we'll do even better by
doing a kmov and a cmp with a GPR, but this is a good start.
The default handling was costing a worst case single source
permute shuffle of the vector before the binop. This worst
case assumes the shuffle might have to be emulated with
extracts and inserts. But since we know we're doing a reduction
we can assume we'll get kshift lowering.
There's still some room for improvement here, but this is
much better than it was.
Update TargetTransformInfo to allow AVX1 to use YMM registers for memcmp.
This is a follow up to D68632 which enabled XOR compares which made this possible.
This also updates the memcmp-optsize.ll test unlike the first patch.
https://reviews.llvm.org/D69658
Update TargetTransformInfo to allow AVX1 to use YMM registers for memcmp.
This is a follow up to D68632 which enabled XOR compares which made this possible.
https://reviews.llvm.org/D69658
Summary:
We don't pattern match pairwise shuffles in SelectionDAG. So we
should only return the optimized costs if its not a pairwise
shuffle.
I think SLP vectorizer gives priority to non pairwise shuffle if
the cost is the same. And the look up for reduction intrinsics
passes false for the pairwise flag. So this probably has no real
effect today.
Reviewers: RKSimon
Reviewed By: RKSimon
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D69083
Add specific scalar costs for CTLZ instructions, we can't discriminate between CTLZ and CTLZ_ZERO_UNDEF so we have to assume the worst. Given how BSR is often a microcoded nightmare on some older targets we might still be underestimating it.
For targets supporting LZCNT (Intel Haswell+ or AMD Fam10+), we provide overrides that assume 1cy costs.
llvm-svn: 374786
Add specific scalar costs for ctpop instructions, these are based on the llvm-mca's SLM throughput numbers (the oldest model we have).
For targets supporting POPCNT, we provide overrides that assume 1cy costs.
llvm-svn: 374775
Add an extra parameter so the backend can take the alignment into
consideration.
Differential Revision: https://reviews.llvm.org/D68400
llvm-svn: 374763
I can't see any notable differences in costs between SSE2 and SSE42 arches for FADD/ADD reduction, so I've lowered the target to just SSE2.
I've also added vXi8 sum reduction costs in line with the PSADBW codegen and discussions on PR42674.
llvm-svn: 374655
In loop-vectorize, interleave count and vector factor depend on target register number. Currently, it does not
estimate different register pressure for different register class separately(especially for scalar type,
float type should not be on the same position with int type), so it's not accurate. Specifically,
it causes too many times interleaving/unrolling, result in too many register spills in loop body and hurting performance.
So we need classify the register classes in IR level, and importantly these are abstract register classes,
and are not the target register class of backend provided in td file. It's used to establish the mapping between
the types of IR values and the number of simultaneous live ranges to which we'd like to limit for some set of those types.
For example, POWER target, register num is special when VSX is enabled. When VSX is enabled, the number of int scalar register is 32(GPR),
float is 64(VSR), but for int and float vector register both are 64(VSR). So there should be 2 kinds of register class when vsx is enabled,
and 3 kinds of register class when VSX is NOT enabled.
It runs on POWER target, it makes big(+~30%) performance improvement in one specific bmk(503.bwaves_r) of spec2017 and no other obvious degressions.
Differential revision: https://reviews.llvm.org/D67148
llvm-svn: 374634
Also Revert "[LoopVectorize] Fix non-debug builds after rL374017"
This reverts commit 9f41deccc0.
This reverts commit 18b6fe07bc.
The patch is breaking PowerPC internal build, checked with author, reverting
on behalf of him for now due to timezone.
llvm-svn: 374091
In loop-vectorize, interleave count and vector factor depend on target register number. Currently, it does not
estimate different register pressure for different register class separately(especially for scalar type,
float type should not be on the same position with int type), so it's not accurate. Specifically,
it causes too many times interleaving/unrolling, result in too many register spills in loop body and hurting performance.
So we need classify the register classes in IR level, and importantly these are abstract register classes,
and are not the target register class of backend provided in td file. It's used to establish the mapping between
the types of IR values and the number of simultaneous live ranges to which we'd like to limit for some set of those types.
For example, POWER target, register num is special when VSX is enabled. When VSX is enabled, the number of int scalar register is 32(GPR),
float is 64(VSR), but for int and float vector register both are 64(VSR). So there should be 2 kinds of register class when vsx is enabled,
and 3 kinds of register class when VSX is NOT enabled.
It runs on POWER target, it makes big(+~30%) performance improvement in one specific bmk(503.bwaves_r) of spec2017 and no other obvious degressions.
Differential revision: https://reviews.llvm.org/D67148
llvm-svn: 374017
This was added back to allow some performance regressions to be
investigated. The main perf issue was fixed shortly after adding
this back and no other major issues have been reported. So I
think its safe to remove this again.
llvm-svn: 373174
SLM is 2 x slower for <2 x i64> comparison ops than other vector types, we should account for this like we do for SLM <2 x i64> add/sub/mul costs.
This should remove some of the SLM codegen diffs in D43582
llvm-svn: 372954
We are missing costs for a lot of truncation cases, I'm hoping to address all the 'zero cost' cases in trunc.ll
I thought this was a vector widening side effect, but even before this we had some interesting LV decisions (notably over indvars) being made due to these zero costs.
llvm-svn: 372498
This merges the 32-bit and 64-bit mode code to just use Custom
for both i32 and i64. We already had most of the handling in
the custom handling due to the AVX512 having legal fp_to_uint.
Just needed to add the i32->i64 promotion handling. Refactor
the fp_to_uint code in the custom handler to simplify the
number of times we check things.
Tweak cost model tables to match the default handling we were
getting due to Expand before.
llvm-svn: 370700
I don't really understand the costs we're using for fp_to_sint,
but prior to widening legalization we used 20 as the cost for this
via the v2i64->v2f64 entry. That number seems better than the 40
we got with widening legalization. So now we need either a
v2i32->v2f64 entry or a v4i32->v2f64 entry depending on whether
AVX is enabled or not since we skip the first SSE2 table look up
under AVX.
llvm-svn: 369628
Google is reporting performance issues with the new default behavior
and have asked for a way to switch back to the old behavior while we
investigate and make fixes.
I've restored all of the code that had since been removed and added
additional checks of the command flag onto code paths that are
not otherwise guarded by a check of getTypeAction.
I've also modified the cost model tables to hopefully get us back
to the previous costs.
Hopefully we won't need to support this for very long since we
have no test coverage of the old behavior so we can very easily
break it.
llvm-svn: 369332
Now that we're using widening legalization. We need to improve our extract_subvector cost model for these types. This patch begins by modeling these as a subvector extract followed by a permute. I've left FIXMEs in the code for future improvements.
Differential Revision: https://reviews.llvm.org/D65892
llvm-svn: 369022
Now that we legalize by widening, the element types here won't change. Previously these were modeled as the elements being widened and then the instruction might become an AND or SHL/ASHR pair. But now they'll become something like a ZERO_EXTEND_VECTOR_INREG/SIGN_EXTEND_VECTOR_INREG.
For AVX2, when the destination type is legal its clear the cost should be 1 since we have extend instructions that can produce 256 bit vectors from less than 128 bit vectors. I'm a little less sure about AVX1 costs, but I think the ones I changed were definitely too high, but they might still be too high.
Differential Revision: https://reviews.llvm.org/D66169
llvm-svn: 368858
The assert that caused this to be reverted should be fixed now.
Original commit message:
This patch changes our defualt legalization behavior for 16, 32, and
64 bit vectors with i8/i16/i32/i64 scalar types from promotion to
widening. For example, v8i8 will now be widened to v16i8 instead of
promoted to v8i16. This keeps the elements widths the same and pads
with undef elements. We believe this is a better legalization strategy.
But it carries some issues due to the fragmented vector ISA. For
example, i8 shifts and multiplies get widened and then later have
to be promoted/split into vXi16 vectors.
This has the potential to cause regressions so we wanted to get
it in early in the 10.0 cycle so we have plenty of time to
address them.
Next steps will be to merge tests that explicitly test the command
line option. And then we can remove the option and its associated
code.
llvm-svn: 368183
This patch changes our defualt legalization behavior for 16, 32, and
64 bit vectors with i8/i16/i32/i64 scalar types from promotion to
widening. For example, v8i8 will now be widened to v16i8 instead of
promoted to v8i16. This keeps the elements widths the same and pads
with undef elements. We believe this is a better legalization strategy.
But it carries some issues due to the fragmented vector ISA. For
example, i8 shifts and multiplies get widened and then later have
to be promoted/split into vXi16 vectors.
This has the potential to cause regressions so we wanted to get
it in early in the 10.0 cycle so we have plenty of time to
address them.
Next steps will be to merge tests that explicitly test the command
line option. And then we can remove the option and its associated
code.
llvm-svn: 367901
When considering a loop containing nontemporal stores or loads for
vectorization, suppress the vectorization if the corresponding
vectorized store or load with the aligment of the original scaler
memory op is not supported with the nontemporal hint on the target.
This adds two new functions:
bool isLegalNTStore(Type *DataType, unsigned Alignment) const;
bool isLegalNTLoad(Type *DataType, unsigned Alignment) const;
to TTI, leaving the target independent default implementation as
returning true, but with overriding implementations for X86 that
check the legality based on available Subtarget features.
This fixes https://llvm.org/PR40759
Differential Revision: https://reviews.llvm.org/D61764
llvm-svn: 363581
The original costs stopped at SSE42, I've added conservative estimates for everything down to SSE1/SSE2 and moved some of the SSE42 costs to SSE41 (really only the addition of PCMPGT makes any difference).
I've also added missing vXi8 costs (we use PHMINPOSUW for i8/i16 for scarily quick results) and 256-bit vector costs for AVX1.
llvm-svn: 360528
On pre-AVX512 targets we can use MOVMSK to extract reduced boolean results. This is properly optimized, annoyingly AVX512 isn't and produces code that is almost as bad as the (unchanged) costs suggest......
Differential Revision: https://reviews.llvm.org/D60403
llvm-svn: 358574
Summary:
This adds a BranchFusion feature to replace the usage of the MacroFusion
for AMD CPUs.
See D59688 for context.
Reviewers: andreadb, lebedev.ri
Subscribers: hiraditya, jdoerfert, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D59872
llvm-svn: 357171
This adds support for scalarizing these intrinsics as well the X86TargetTransformInfo support to avoid scalarizing them in the cases X86 can handle.
I've omitted handling special cases for constant masks for this first pass. Though CodeGenPrepare can constant fold the branch conditions and remove some of the control flow anyway.
Fixes PR40994 and is covers most of PR3666. Might want to implement constant masks to close that.
Differential Revision: https://reviews.llvm.org/D59180
llvm-svn: 356687
We were just checking pointer size and type primitive size. But this caused unintended things like vectors of half being accepted by masked load/store.
For FP we now explicitly check for only double and float.
For pointers we now let any pointer through. Trusting that only 32 and 64 would be used to generate assembly.
We only check bitwidth after checking that the type is an integer.
llvm-svn: 355667
As this has broken the lto bootstrap build for 3 days and is
showing a significant regression on the Dither_benchmark results (from
the LLVM benchmark suite) -- specifically, on the
BENCHMARK_FLOYD_DITHER_128, BENCHMARK_FLOYD_DITHER_256, and
BENCHMARK_FLOYD_DITHER_512; the others are unchanged. These have
regressed by about 28% on Skylake, 34% on Haswell, and over 40% on
Sandybridge.
This reverts commit r353923.
llvm-svn: 354434
The use of the -mprefer-vector-width=256 command line option mixed with functions
using vector intrinsics can create situations where one function thinks 512 vectors
are legal, but another fucntion does not.
If a 512 bit vector is passed between them via a pointer, its possible ArgumentPromotion
might try to pass by value instead. This will result in type legalization for the two
functions handling the 512 bit vector differently leading to runtime failures.
Had the 512 bit vector been passed by value from clang codegen, both functions would
have been tagged with a min-legal-vector-width=512 function attribute. That would
make them be legalized the same way.
I observed this issue in 32-bit mode where a union containing a 512 bit vector was
being passed by a function that used intrinsics to one that did not. The caller
ended up passing in zmm0 and the callee tried to read it from ymm0 and ymm1.
The fix implemented here is just to consider it a mismatch if two functions
would handle 512 bit differently without looking at the types that are being
considered. This is the easist and safest fix, but it can be improved in the future.
Differential Revision: https://reviews.llvm.org/D58390
llvm-svn: 354376
Tuning flags don't have any effect on the available instructions so aren't a good reason to prevent inlining.
There are also some ISA flags that don't have any intrinsics our ABI requirements that we can exclude. I've put only the most basic ones like cmpxchg16b and lahfsahf. These are interesting because they aren't present in all 64-bit CPUs, but we have codegen workarounds when they aren't present.
Loosening these checks can help with scenarios where a caller has a more specific CPU than a callee. The default tuning flags on our generic 'x86-64' CPU can currently make it inline compatible with other CPUs. I've also added an example test for 'nocona' and 'prescott' where 'nocona' is just a 64-bit capable version of 'prescott' but in 32-bit mode they should be completely compatible.
I've based the implementation here of the similar code in AMDGPU.
Differential Revision: https://reviews.llvm.org/D58371
llvm-svn: 354355
Try to use 64-bit SLP vectorization. In addition to horizontal instrs
this change triggers optimizations for partial vector operations (for instance,
using low halfs of 128-bit registers xmm0 and xmm1 to multiply <2 x float> by
<2 x float>).
Fixes llvm.org/PR32433
llvm-svn: 353923
Followup to D56636, this time handling the UADDSAT case by expanding
uadd.sat(a, b) to umin(a, ~b) + b.
Differential Revision: https://reviews.llvm.org/D56869
llvm-svn: 352409
First step towards PR40376, this patch adds support for getCmpSelInstrCost to use the (optional) Instruction CmpInst predicate to indicate the type of integer comparison we're performing and alter the costs accordingly.
Differential Revision: https://reviews.llvm.org/D57013
llvm-svn: 351810
Prior to SSE41 (and sometimes on AVX1), vector select has to be performed as a ((X & C)|(Y & ~C)) bit select.
Exposes a couple of issues with the min/max reduction costs (which only go down to SSE42 for some reason).
The increase pre-SSE41 selection costs also prevent a couple of tests from firing any longer, so I've either tweaked the target or added AVX tests as well to the existing SSE2 tests.
llvm-svn: 351685
to reflect the new license.
We understand that people may be surprised that we're moving the header
entirely to discuss the new license. We checked this carefully with the
Foundation's lawyer and we believe this is the correct approach.
Essentially, all code in the project is now made available by the LLVM
project under our new license, so you will see that the license headers
include that license only. Some of our contributors have contributed
code under our old license, and accordingly, we have retained a copy of
our old license notice in the top-level files in each project and
repository.
llvm-svn: 351636
There are no test changes here in the existing cost model
regression tests because integer add/sub have a default
legal cost of 1 already. This would break, however, if
we custom lower those ops because the default cost model
assumes that custom-lowered ops are more expensive.
This is similar to the change in rL350403. See discussion
in D56011 for more details. When we enhance that patch to
handle integer ops, we need this cost model change to avoid
unintended diffs here from the custom lowering.
llvm-svn: 350496
Noticed in D56011 - handle the case that scalar fp ops are quicker on P3 than P4
Add the other costs so that we're not relying on the default "is legal/custom" cost logic.
llvm-svn: 350403
Summary:
This allows expanding {7,11,13,14,15,21,22,23,25,26,27,28,29,30,31}-byte memcmp
in just two loads on X86. These were previously calling memcmp.
Reviewers: spatel, gchatelet
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D55263
llvm-svn: 349731
This is an initial patch to add a minimum level of support for funnel shifts to the SelectionDAG and to begin wiring it up to the X86 SHLD/SHRD instructions.
Some partial legalization code has been added to handle the case for 'SlowSHLD' where we want to expand instead and I've added a few DAG combines so we don't get regressions from the existing DAG builder expansion code.
Differential Revision: https://reviews.llvm.org/D54698
llvm-svn: 348353
Unlike most cost model functions this code makes a lot of table lookups without using the results from getTypeLegalizationCost. This means 512-bit vectors can be looked up even when the type isn't legal.
This patch adds a check around the two tables that contain 512-bit types to make sure that neither of the types would be split by type legalization. Meaning 512 bit types are illegal. I wanted to write this in a somewhat generic way that uses type legalization query hooks. But if prefered, I can switch to just using is512BitVector and the subtarget feature.
Differential Revision: https://reviews.llvm.org/D54984
llvm-svn: 347786
This fixes some of scalarization costs reported for sext/zext using avx512bw. This does not fix all scalarization costs being reported. Just the worst.
I've restricted this only to combinations of types that are legal with avx512bw like v32i1/v64i1/v32i16/v64i8 and conversions between vXi1 and vXi8/vXi16 with legal vXi8/vXi16 result types.
Differential Revision: https://reviews.llvm.org/D54979
llvm-svn: 347785
We're seeing some issues internally where we sent some intrinsics into the cost model that the getTypeLegalizationCost call fails on, but X86 specific tables don't care about. Our base class implementation takes care of them. We'd just like X86 backend to ignore them.
This patch makes sure the switch returned something X86 cares about and skips the table lookups and type legalization call if not. Probably more efficient too since we don't go scanning the tables for every intrinsic we could possibly see.
Differential Revision: https://reviews.llvm.org/D54711
llvm-svn: 347248
Add support for the expansion of funnelshift/rotates to getIntrinsicInstrCost.
This also required us to move the X86 fshl/fshr costs to the same place as the rotates to avoid expansion and get correct scalarization vs vectorization costs.
llvm-svn: 346854
When we repeat the 2 shifting operands then this is a bit rotation - annoyingly this has to be done in the other getIntrinsicInstrCost than most intrinsics as we need to check the operands are the same.
llvm-svn: 346688
optsize using masked wide loads
Under Opt for Size, the vectorizer does not vectorize interleave-groups that
have gaps at the end of the group (such as a loop that reads only the even
elements: a[2*i]) because that implies that we'll require a scalar epilogue
(which is not allowed under Opt for Size). This patch extends the support for
masked-interleave-groups (introduced by D53011 for conditional accesses) to
also cover the case of gaps in a group of loads; Targets that enable the
masked-interleave-group feature don't have to invalidate interleave-groups of
loads with gaps; they could now use masked wide-loads and shuffles (if that's
what the cost model selects).
Reviewers: Ayal, hsaito, dcaballe, fhahn
Reviewed By: Ayal
Differential Revision: https://reviews.llvm.org/D53668
llvm-svn: 345705
Simon Pilgrim