As discussed on PR24888, until SSE42 we don't have access to PCMPGTQ for v2i64 comparisons, but the cost models don't reflect this, resulting in over-optimistic vectorizaton.
This patch adds SSE2 'base level' costs that match what a typical target is capable of and only reduces the v2i64 costs at SSE42.
Technically SSE41 provides a PCMPEQQ v2i64 equality test, but as getCmpSelInstrCost doesn't give us a way to discriminate between comparison test types we can't easily make use of this, otherwise we could split the cost of integer equality and greater-than tests to give better costings of each.
Differential Revision: http://reviews.llvm.org/D20057
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@268972 91177308-0d34-0410-b5e6-96231b3b80d8
Summary:
rL256194 transforms truncations between vectors of integers into PACKUS/PACKSS
operations during DAG combine. This generates better code for truncate, so cost
of truncate needs to be changed but looks like it got changed only in SSE2 table
Whereas this change is also applicable for SSE4.1, so the cost of truncate needs
to be changed for that as well. Cost of “TRUNCATE v16i32 to v16i8” & “TRUNCATE
v16i16 to v16i8” should be same in SSE4.1 & SSE2 table. Removing their cost from
SSE4.1, so it will fall back to SSE2.
Reviewers: Simon Pilgrim
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@267123 91177308-0d34-0410-b5e6-96231b3b80d8
This is a resubmittion of 263158 change.
This patch fixes the problem which occurs when loop-vectorize tries to use @llvm.masked.load/store intrinsic for a non-default addrspace pointer. It fails with "Calling a function with a bad signature!" assertion in CallInst constructor because it tries to pass a non-default addrspace pointer to the pointer argument which has default addrspace.
The fix is to add pointer type as another overloaded type to @llvm.masked.load/store intrinsics.
Reviewed By: reames
Differential Revision: http://reviews.llvm.org/D17270
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@266086 91177308-0d34-0410-b5e6-96231b3b80d8
PPC has a vector popcount, this lets the vectorizer use the correct cost
for it. Tweak X86 test to use an intrinsic that's actually scalarized (we
have a somewhat efficient lowering for vector popcount using SSE, the
cost model finds that now).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@265005 91177308-0d34-0410-b5e6-96231b3b80d8
This patch fixes the problem which occurs when loop-vectorize tries to use @llvm.masked.load/store intrinsic for a non-default addrspace pointer. It fails with "Calling a function with a bad signature!" assertion in CallInst constructor because it tries to pass a non-default addrspace pointer to the pointer argument which has default addrspace.
The fix is to add pointer type as another overloaded type to @llvm.masked.load/store intrinsics.
Reviewed By: reames
Differential Revision: http://reviews.llvm.org/D17270
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@263158 91177308-0d34-0410-b5e6-96231b3b80d8
This patch transforms truncation between vectors of integers into
X86ISD::PACKUS/PACKSS operations during DAG combine. We don't do it in
lowering phase because after type legalization, the original truncation
will be turned into a BUILD_VECTOR with each element that is extracted
from a vector and then truncated, and from them it is difficult to do
this optimization. This greatly improves the performance of truncations
on some specific types.
Cost table is updated accordingly.
Differential revision: http://reviews.llvm.org/D14588
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@256194 91177308-0d34-0410-b5e6-96231b3b80d8
Previously in the conversion cost table there are no entries for integer-integer
conversions on SSE2. This will result in imprecise costs for certain vectorized
operations. This patch adds those entries for SSE2 and SSE4.1. The cost numbers
are counted from the result of running llc on the new test case in this patch.
Differential revision: http://reviews.llvm.org/D15132
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@255315 91177308-0d34-0410-b5e6-96231b3b80d8
Currently in LLVM's cost model, a vectorized arithmetic instruction will have
high cost if its type is split into multiple registers. However, this
punishment is too heavy and unnecessary. The overhead of the split should not
be on arithmetic instructions but instructions that implement the split. Note
that during vectorization we have calculated the register pressure, and we
only choose proper interleaving factor (and also vectorization factor) so
that we don't use more registers than the maximum number.
Here is a very simple example: if a vadd has the cost 1, and if we double VF
so that we need two registers to perform it, then its cost will become 4 with
the current implementation, which will prevent us to use larger VF.
Differential revision: http://reviews.llvm.org/D15159
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@254671 91177308-0d34-0410-b5e6-96231b3b80d8
I checked and updated the cost of AVX-512 conversion operations. Added cost of conversion operations in DQ mode.
Conversion of illegal types that requires vector split is not calculated right now (like for other X86 targets).
Differential Revision: http://reviews.llvm.org/D15074
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@254494 91177308-0d34-0410-b5e6-96231b3b80d8
The cost for scalarized operations is computed as N * (scalar operation
cost + 1 extractelement + 1 insertelement). This partially fixes
inflating the cost of scalarized operations since every operation is
scalarized and free. I don't think we want any cost asociated with
scalarization, but for now insertelement is still counted. I'm not sure
if we should pretend that insertelement is also free, or add a way
to compute a custom scalarization cost.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@254438 91177308-0d34-0410-b5e6-96231b3b80d8
The underlying issues surrounding codegen for 32-bit vselects have been resolved. The pessimistic costs for 64-bit vselects remain due to the bad
scalarization that is still happening there.
I tested this on A57 in T32, A32 and A64 modes. I saw no regressions, and some improvements.
From my benchmarks, I saw these improvements in A57 (T32)
spec.cpu2000.ref.177_mesa 5.95%
lnt.SingleSource/Benchmarks/Shootout/strcat 12.93%
lnt.MultiSource/Benchmarks/MiBench/telecomm-CRC32/telecomm-CRC32 11.89%
I also measured A57 A32, A53 T32 and A9 T32 and found no performance regressions. I see much bigger wins in third-party benchmarks with this change
Differential Revision: http://reviews.llvm.org/D14743
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@253349 91177308-0d34-0410-b5e6-96231b3b80d8
There are several dodgy costings due to AVX1 legalizing 256-bit integer vectors that need fixing.
As discussed in D8690.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@249983 91177308-0d34-0410-b5e6-96231b3b80d8
There are several dodgy costings due to AVX1 legalizing 256-bit integer vectors that need fixing.
As discussed in D8690.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@249981 91177308-0d34-0410-b5e6-96231b3b80d8
The XOP shifts just have logical/arithmetic versions and the left/right shifts are controlled by whether the value is positive/negative. Because of this I've added new X86ISD nodes instead of trying to force them to use the existing shift nodes.
Additionally Excavator cores (bdver4) support XOP and AVX2 - meaning that it should use the AVX2 shifts when it can and fall back to XOP in other cases.
Differential Revision: http://reviews.llvm.org/D8690
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@248878 91177308-0d34-0410-b5e6-96231b3b80d8
Summary:
We are not scalarizing the wide selects in codegen for i16 and i32 and
therefore we can remove the amortization factor. We still have issues
with i64 vectors in codegen though.
Reviewers: mcrosier
Subscribers: mcrosier, aemerson, llvm-commits, rengolin
Differential Revision: http://reviews.llvm.org/D12724
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@247156 91177308-0d34-0410-b5e6-96231b3b80d8
Pre-P8, when we generate code for unaligned vector loads (for Altivec and QPX
types), even when accounting for the combining that takes place for multiple
consecutive such loads, there is at least one load instructions and one
permutation for each load. Make sure the cost reported reflects the cost of the
permutes as well.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@246807 91177308-0d34-0410-b5e6-96231b3b80d8
I'm adding a regression test to better cover code generation for unaligned
vector loads and stores, but there's no functional change to the code
generation here. There is an improvement to the cost model for unaligned vector
loads and stores, mostly for QPX (for which we were not previously accounting
for the permutation-based loads), and the cost model implementation is cleaner.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@246712 91177308-0d34-0410-b5e6-96231b3b80d8
Summary:
This change limits the minimum cost of an insert/extract
element operation to 2 in cases where this would result
in mixing of NEON and VFP code.
Reviewers: rengolin
Subscribers: mssimpso, aemerson, llvm-commits, rengolin
Differential Revision: http://reviews.llvm.org/D12030
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@245225 91177308-0d34-0410-b5e6-96231b3b80d8
This patch vectorizes the v2i64/v4i64 ASHR shift operations - the last remaining integer vector shifts that are still being transferred to/from the scalar unit to be completed.
Differential Revision: http://reviews.llvm.org/D11439
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@243569 91177308-0d34-0410-b5e6-96231b3b80d8
r243250 appeared to break clang/test/Analysis/dead-store.c on one of the build
slaves, but I couldn't reproduce this failure locally. Probably a false
positive as I saw this test was broken by r243246 or r243247 too but passed
later without people fixing anything.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@243253 91177308-0d34-0410-b5e6-96231b3b80d8
Summary:
This patch updates TargetTransformInfoImplCRTPBase::getGEPCost to consider
addressing modes. It now returns TCC_Free when the GEP can be completely folded
to an addresing mode.
I started this patch as I refactored SLSR. Function isGEPFoldable looks common
and is indeed used by some WIP of mine. So I extracted that logic to getGEPCost.
Furthermore, I noticed getGEPCost wasn't directly tested anywhere. The best
testing bed seems CostModel, but its getInstructionCost method invokes
getAddressComputationCost for GEPs which provides very coarse estimation. So
this patch also makes getInstructionCost call the updated getGEPCost for GEPs.
This change inevitably breaks some tests because the cost model changes, but
nothing looks seriously wrong -- if we believe the new cost model is the right
way to go, these tests should be updated.
This patch is not perfect yet -- the comments in some tests need to be updated.
I want to know whether this is a right approach before fixing those details.
Reviewers: chandlerc, hfinkel
Subscribers: aschwaighofer, llvm-commits, aemerson
Differential Revision: http://reviews.llvm.org/D9819
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@243250 91177308-0d34-0410-b5e6-96231b3b80d8
While the v4i32 shl operation is already vectorized using a cvttps2dq/pmulld pattern, the lshr/ashr opeations are still scalarized.
This patch adds vectorization support for non-uniform v4i32 shift operations - it splats constant shift amounts to allow them to use the immediate sse shift instructions, or extracts/zero-extends non-constant shift amounts. The individual results are then blended together.
Differential Revision: http://reviews.llvm.org/D11063
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@241989 91177308-0d34-0410-b5e6-96231b3b80d8
Merged separate (but equivalent) SSE2/AVX512F tests.
Removed codegen tests since these are already done better in test/CodeGen/X86.
The actual cost values still need to be updated to match recent codegen improvements.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@240219 91177308-0d34-0410-b5e6-96231b3b80d8
This patch ensures that SHL/SRL/SRA shifts for i8 and i16 vectors avoid scalarization. It builds on the existing i8 SHL vectorized implementation of moving the shift bits up to the sign bit position and separating the 4, 2 & 1 bit shifts with several improvements:
1 - SSE41 targets can use (v)pblendvb directly with the sign bit instead of performing a comparison to feed into a VSELECT node.
2 - pre-SSE41 targets were masking + comparing with an 0x80 constant - we avoid this by using the fact that a set sign bit means a negative integer which can be compared against zero to then feed into VSELECT, avoiding the need for a constant mask (zero generation is much cheaper).
3 - SRA i8 needs to be unpacked to the upper byte of a i16 so that the i16 psraw instruction can be correctly used for sign extension - we have to do more work than for SHL/SRL but perf tests indicate that this is still beneficial.
The i16 implementation is similar but simpler than for i8 - we have to do 8, 4, 2 & 1 bit shifts but less shift masking is involved. SSE41 use of (v)pblendvb requires that the i16 shift amount is splatted to both bytes however.
Tested on SSE2, SSE41 and AVX machines.
Differential Revision: http://reviews.llvm.org/D9474
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239509 91177308-0d34-0410-b5e6-96231b3b80d8
