[X86][Costmodel] Load/store i8 Stride=2 VF=4 interleaving costs

[lebedev.ri]

Summary

The only sched models that for cpu's that support avx2
but not avx512 are: haswell, broadwell, skylake, zen1-3

Identical to VF=2.

For load we have:
https://godbolt.org/z/sGE41GYo7 - for intels Block RThroughput: =2.0; for ryzens, Block RThroughput: <=1.0
So pick cost of 2.

For store we have:
https://godbolt.org/z/ba5r3s9xa - for intels Block RThroughput: =1.0; for ryzens, Block RThroughput: <=0.5
So pick cost of 1.

I'm directly using the shuffling asm the llc produced,
without any manual fixups that may be needed
to ensure sequential execution.

[RKSimon]

Before we go any further - do we gain anything by updating BaseT::getInterleavedMemoryOpCost to use getScalarizationOverhead?

[lebedev.ri]

In D110704#3030093, @RKSimon wrote:

Before we go any further - do we gain anything by updating BaseT::getInterleavedMemoryOpCost to use getScalarizationOverhead?

It depends on your definition of "anything". It basically doesn't help,
because the only improvement that gives us not overestimating XMM subreg insertion/extraction from YMM,
The costs still bogusly high.

[lebedev.ri]

(I've posted proof of concept: D110713)

In D110704#3030176, @lebedev.ri wrote:

In D110704#3030093, @RKSimon wrote:

Before we go any further - do we gain anything by updating BaseT::getInterleavedMemoryOpCost to use getScalarizationOverhead?

It depends on your definition of "anything". It basically doesn't help,
because the only improvement that gives us not overestimating XMM subreg insertion/extraction from YMM,
The costs still bogusly high.

FWIW i agree that this is not really great, but what we need is D100486,
and afterwards we should materialize shuffle masks and query their costs.
I was planning on doing that, as i have noted previously in some review.
But even that will overestimate the cost.

[RKSimon]

LGTM

llvm/lib/Target/X86/X86TargetTransformInfo.cpp

InstructionCost X86TTIImpl::getInterleavedMemoryOpCostAVX2(
   // TODO: Complete for other data-types and strides.
   // Each combination of Stride, element bit width and VF results in a different
   // sequence; The cost tables are therefore accessed with:
   // Factor (stride) and VectorType=VFxiN.
   // The Cost accounts only for the shuffle sequence;
   // The cost of the loads/stores is accounted for separately.
   //
   static const CostTblEntry AVX2InterleavedLoadTbl[] = {
       {2, MVT::v2i8, 2},  // (load 4i8 and) deinterleave into 2 x 2i8
+      {2, MVT::v4i8, 2},  // (load 8i8 and) deinterleave into 2 x 4i8
 
       {2, MVT::v2i16, 2}, // (load 4i16 and) deinterleave into 2 x 2i16
       {2, MVT::v4i16, 2}, // (load 8i16 and) deinterleave into 2 x 4i16
       {2, MVT::v8i16, 6}, // (load 16i16 and) deinterleave into 2 x 8i16
       {2, MVT::v16i16, 9}, // (load 32i16 and) deinterleave into 2 x 16i16
       {2, MVT::v32i16, 18}, // (load 64i16 and) deinterleave into 2 x 32i16
 
       {2, MVT::v4i64, 6}, // (load 8i64 and) deinterleave into 2 x 4i64
 
       {3, MVT::v2i8, 10},  // (load 6i8 and) deinterleave into 3 x 2i8
       {3, MVT::v4i8, 4},   // (load 12i8 and) deinterleave into 3 x 4i8
       {3, MVT::v8i8, 9},   // (load 24i8 and) deinterleave into 3 x 8i8
       {3, MVT::v16i8, 11}, // (load 48i8 and) deinterleave into 3 x 16i8
       {3, MVT::v32i8, 13}, // (load 96i8 and) deinterleave into 3 x 32i8
 
       {3, MVT::v8i32, 17}, // (load 24i32 and) deinterleave into 3 x 8i32
 
       {4, MVT::v2i8, 12},  // (load 8i8 and) deinterleave into 4 x 2i8
       {4, MVT::v4i8, 4},   // (load 16i8 and) deinterleave into 4 x 4i8
       {4, MVT::v8i8, 20},  // (load 32i8 and) deinterleave into 4 x 8i8
       {4, MVT::v16i8, 39}, // (load 64i8 and) deinterleave into 4 x 16i8
       {4, MVT::v32i8, 80}, // (load 128i8 and) deinterleave into 4 x 32i8
 
       {4, MVT::v2i16, 6}, // (load 8i16 and) deinterleave into 4 x 2i16
       {4, MVT::v4i16, 17}, // (load 16i16 and) deinterleave into 4 x 4i16
       {4, MVT::v8i16, 33}, // (load 32i16 and) deinterleave into 4 x 8i16
       {4, MVT::v16i16, 75}, // (load 64i16 and) deinterleave into 4 x 16i16
       {4, MVT::v32i16, 150}, // (load 128i16 and) deinterleave into 4 x 32i16
 
       {6, MVT::v2i16, 13}, // (load 12i16 and) deinterleave into 6 x 2i16
       {6, MVT::v4i16, 9}, // (load 24i16 and) deinterleave into 6 x 4i16
       {6, MVT::v8i16, 39}, // (load 48i16 and) deinterleave into 6 x 8i16
       {6, MVT::v16i16, 106}, // (load 96i16 and) deinterleave into 6 x 16i16
 
       {8, MVT::v8i32, 40} // (load 64i32 and) deinterleave into 8 x 8i32
   };
 
   static const CostTblEntry AVX2InterleavedStoreTbl[] = {
       {2, MVT::v2i8, 1}, // interleave 2 x 2i8 into 4i8 (and store)
+      {2, MVT::v4i8, 1}, // interleave 2 x 4i8 into 8i8 (and store)
 
       {2, MVT::v2i16, 1}, // interleave 2 x 2i16 into 4i16 (and store)
       {2, MVT::v4i16, 1}, // interleave 2 x 4i16 into 8i16 (and store)
       {2, MVT::v8i16, 3}, // interleave 2 x 8i16 into 16i16 (and store)
       {2, MVT::v16i16, 4}, // interleave 2 x 16i16 into 32i16 (and store)
       {2, MVT::v32i16, 8}, // interleave 2 x 32i16 into 64i16 (and store)
 
       {2, MVT::v4i64, 6}, // interleave 2 x 4i64 into 8i64 (and store)
 
       {3, MVT::v2i8, 7},   // interleave 3 x 2i8 into 6i8 (and store)
       {3, MVT::v4i8, 8},   // interleave 3 x 4i8 into 12i8 (and store)
       {3, MVT::v8i8, 11},  // interleave 3 x 8i8 into 24i8 (and store)
       {3, MVT::v16i8, 11}, // interleave 3 x 16i8 into 48i8 (and store)
       {3, MVT::v32i8, 13}, // interleave 3 x 32i8 into 96i8 (and store)
 
       {4, MVT::v2i8, 12},  // interleave 4 x 2i8 into 8i8 (and store)
       {4, MVT::v4i8, 9},   // interleave 4 x 4i8 into 16i8 (and store)
       {4, MVT::v8i8, 10},  // interleave 4 x 8i8 into 32i8 (and store)
       {4, MVT::v16i8, 10}, // interleave 4 x 16i8 into 64i8 (and store)
       {4, MVT::v32i8, 12}, // interleave 4 x 32i8 into 128i8 (and store)
 
       {4, MVT::v2i16, 2},  // interleave 4 x 2i16 into 8i16 (and store)
       {4, MVT::v4i16, 6},  // interleave 4 x 4i16 into 16i16 (and store)
       {4, MVT::v8i16, 10},  // interleave 4 x 8i16 into 32i16 (and store)
       {4, MVT::v16i16, 32},  // interleave 4 x 16i16 into 64i16 (and store)
       {4, MVT::v32i16, 64},  // interleave 4 x 32i16 into 128i16 (and store)
 
       {6, MVT::v2i16, 10},  // interleave 6 x 2i16 into 12i16 (and store)
       {6, MVT::v4i16, 15},  // interleave 6 x 4i16 into 24i16 (and store)
       {6, MVT::v8i16, 21},  // interleave 6 x 8i16 into 48i16 (and store)
       {6, MVT::v16i16, 58},  // interleave 6 x 16i16 into 96i16 (and store)
   };
 
   if (Opcode == Instruction::Load) {
     if (const auto *Entry =
             CostTableLookup(AVX2InterleavedLoadTbl, Factor, ETy.getSimpleVT()))
       return MemOpCosts + Entry->Cost;
   } else {
     assert(Opcode == Instruction::Store &&

llvm/test/Analysis/CostModel/X86/interleaved-load-i8-stride-2.ll

 ; AVX1: LV: Found an estimated cost of 9 for VF 2 For instruction:   %v0 = load i8, i8* %in0, align 1
 ; AVX1: LV: Found an estimated cost of 17 for VF 4 For instruction:   %v0 = load i8, i8* %in0, align 1
 ; AVX1: LV: Found an estimated cost of 33 for VF 8 For instruction:   %v0 = load i8, i8* %in0, align 1
 ; AVX1: LV: Found an estimated cost of 81 for VF 16 For instruction:   %v0 = load i8, i8* %in0, align 1
 ; AVX1: LV: Found an estimated cost of 226 for VF 32 For instruction:   %v0 = load i8, i8* %in0, align 1
 ;
 ; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction:   %v0 = load i8, i8* %in0, align 1
 ; AVX2: LV: Found an estimated cost of 3 for VF 2 For instruction:   %v0 = load i8, i8* %in0, align 1
-; AVX2: LV: Found an estimated cost of 17 for VF 4 For instruction:   %v0 = load i8, i8* %in0, align 1
+; AVX2: LV: Found an estimated cost of 3 for VF 4 For instruction:   %v0 = load i8, i8* %in0, align 1
 ; AVX2: LV: Found an estimated cost of 33 for VF 8 For instruction:   %v0 = load i8, i8* %in0, align 1
 ; AVX2: LV: Found an estimated cost of 81 for VF 16 For instruction:   %v0 = load i8, i8* %in0, align 1
 ; AVX2: LV: Found an estimated cost of 226 for VF 32 For instruction:   %v0 = load i8, i8* %in0, align 1
 ;
 ; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction:   %v0 = load i8, i8* %in0, align 1
 ; AVX512: LV: Found an estimated cost of 3 for VF 2 For instruction:   %v0 = load i8, i8* %in0, align 1
 ; AVX512: LV: Found an estimated cost of 3 for VF 4 For instruction:   %v0 = load i8, i8* %in0, align 1
 ; AVX512: LV: Found an estimated cost of 3 for VF 8 For instruction:   %v0 = load i8, i8* %in0, align 1

llvm/test/Analysis/CostModel/X86/interleaved-store-i8-stride-2.ll

 ; AVX1: LV: Found an estimated cost of 9 for VF 2 For instruction:   store i8 %v1, i8* %out1, align 1
 ; AVX1: LV: Found an estimated cost of 17 for VF 4 For instruction:   store i8 %v1, i8* %out1, align 1
 ; AVX1: LV: Found an estimated cost of 33 for VF 8 For instruction:   store i8 %v1, i8* %out1, align 1
 ; AVX1: LV: Found an estimated cost of 97 for VF 16 For instruction:   store i8 %v1, i8* %out1, align 1
 ; AVX1: LV: Found an estimated cost of 226 for VF 32 For instruction:   store i8 %v1, i8* %out1, align 1
 
 ; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction:   store i8 %v1, i8* %out1, align 1
 ; AVX2: LV: Found an estimated cost of 2 for VF 2 For instruction:   store i8 %v1, i8* %out1, align 1
-; AVX2: LV: Found an estimated cost of 17 for VF 4 For instruction:   store i8 %v1, i8* %out1, align 1
+; AVX2: LV: Found an estimated cost of 2 for VF 4 For instruction:   store i8 %v1, i8* %out1, align 1
 ; AVX2: LV: Found an estimated cost of 33 for VF 8 For instruction:   store i8 %v1, i8* %out1, align 1
 ; AVX2: LV: Found an estimated cost of 97 for VF 16 For instruction:   store i8 %v1, i8* %out1, align 1
 ; AVX2: LV: Found an estimated cost of 226 for VF 32 For instruction:   store i8 %v1, i8* %out1, align 1
 ;
 ; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction:   store i8 %v1, i8* %out1, align 1
 ; AVX512: LV: Found an estimated cost of 4 for VF 2 For instruction:   store i8 %v1, i8* %out1, align 1
 ; AVX512: LV: Found an estimated cost of 4 for VF 4 For instruction:   store i8 %v1, i8* %out1, align 1
 ; AVX512: LV: Found an estimated cost of 4 for VF 8 For instruction:   store i8 %v1, i8* %out1, align 1