[PATCH] D128784: [RISCV] Add cost model for vector.reverse mask operation

[WangLian via Phabricator via llvm-commits]

Jimerlife created this revision.
Jimerlife added reviewers: craig.topper, reames, frasercrmck, benshi001.
Jimerlife added a project: LLVM.
Herald added subscribers: sunshaoce, VincentWu, luke957, StephenFan, vkmr, evandro, luismarques, apazos, sameer.abuasal, s.egerton, Jim, benna, psnobl, jocewei, PkmX, the_o, brucehoult, MartinMosbeck, rogfer01, edward-jones, zzheng, jrtc27, shiva0217, kito-cheng, niosHD, sabuasal, simoncook, johnrusso, rbar, asb, hiraditya, arichardson.
Herald added a project: All.
Jimerlife requested review of this revision.
Herald added subscribers: llvm-commits, pcwang-thead, eopXD, jacquesguan, MaskRay.

For vector.reverse mask operation need additional extend and truncate. So, I try to adjust this cost model.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D128784

Files:
  llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
  llvm/test/Analysis/CostModel/RISCV/rvv-shuffle.ll


Index: llvm/test/Analysis/CostModel/RISCV/rvv-shuffle.ll
===================================================================
--- llvm/test/Analysis/CostModel/RISCV/rvv-shuffle.ll
+++ llvm/test/Analysis/CostModel/RISCV/rvv-shuffle.ll
@@ -65,10 +65,10 @@
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %reverse_nxv8i32 = call <vscale x 8 x i32> @llvm.experimental.vector.reverse.nxv8i32(<vscale x 8 x i32> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %reverse_nxv2i64 = call <vscale x 2 x i64> @llvm.experimental.vector.reverse.nxv2i64(<vscale x 2 x i64> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %reverse_nxv4i64 = call <vscale x 4 x i64> @llvm.experimental.vector.reverse.nxv4i64(<vscale x 4 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %reverse_nxv16i1 = call <vscale x 16 x i1> @llvm.experimental.vector.reverse.nxv16i1(<vscale x 16 x i1> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %reverse_nxv8i1 = call <vscale x 8 x i1> @llvm.experimental.vector.reverse.nxv8i1(<vscale x 8 x i1> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %reverse_nxv4i1 = call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %reverse_nxv2i1 = call <vscale x 2 x i1> @llvm.experimental.vector.reverse.nxv2i1(<vscale x 2 x i1> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %reverse_nxv16i1 = call <vscale x 16 x i1> @llvm.experimental.vector.reverse.nxv16i1(<vscale x 16 x i1> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %reverse_nxv8i1 = call <vscale x 8 x i1> @llvm.experimental.vector.reverse.nxv8i1(<vscale x 8 x i1> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %reverse_nxv4i1 = call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %reverse_nxv2i1 = call <vscale x 2 x i1> @llvm.experimental.vector.reverse.nxv2i1(<vscale x 2 x i1> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret void
 ;
   %reverse_nxv16i8 = call <vscale x 16 x i8> @llvm.experimental.vector.reverse.nxv16i8(<vscale x 16 x i8> undef)
Index: llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
===================================================================
--- llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
+++ llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
@@ -198,6 +198,9 @@
       //   vid.v v9
       //   vrsub.vx v10, v9, a0
       //   vrgather.vv v9, v8, v10
+      if (Tp->getElementType()->isIntegerTy(1))
+        // Mask operation additionally required extend and truncate
+        return LT.first * 8;
       return LT.first * 6;
     }
   }


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D128784.440864.patch
Type: text/x-patch
Size: 2998 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20220629/15157ec9/attachment.bin>