[PATCH] D124960: [RISCV] Add a special case to treat riscv-v-vector-bits-min=-1 as meaning use Zvl*b value.

[Craig Topper via Phabricator via llvm-commits]

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

riscv-v-vector-bits-min is primarily used to opt-in to the
autovectorizer. The vector width can be determined from Zvl*b.

This patch adds support treating -1 as meaning use Zvl*b so we can
still opt-in to autovectorization without needing to repeat a
vector width already given by Zvl*b or -mcpu.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D124960

Files:
  llvm/lib/Target/RISCV/RISCVSubtarget.cpp
  llvm/test/CodeGen/RISCV/rvv/fixed-vectors-int.ll


Index: llvm/test/CodeGen/RISCV/rvv/fixed-vectors-int.ll
===================================================================
--- llvm/test/CodeGen/RISCV/rvv/fixed-vectors-int.ll
+++ llvm/test/CodeGen/RISCV/rvv/fixed-vectors-int.ll
@@ -3,6 +3,8 @@
 ; RUN: llc -mtriple=riscv64 -mattr=+v -riscv-v-vector-bits-min=128 -riscv-v-fixed-length-vector-lmul-max=2 -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,RV64,LMULMAX2,LMULMAX2-RV64
 ; RUN: llc -mtriple=riscv32 -mattr=+v -riscv-v-vector-bits-min=128 -riscv-v-fixed-length-vector-lmul-max=1 -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,RV32,LMULMAX1,LMULMAX1-RV32
 ; RUN: llc -mtriple=riscv64 -mattr=+v -riscv-v-vector-bits-min=128 -riscv-v-fixed-length-vector-lmul-max=1 -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,RV64,LMULMAX1,LMULMAX1-RV64
+; RUN: llc -mtriple=riscv32 -mattr=+v,+zvl128b -riscv-v-vector-bits-min=-1 -riscv-v-fixed-length-vector-lmul-max=1 -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,RV32,LMULMAX1,LMULMAX1-RV32
+; RUN: llc -mtriple=riscv64 -mattr=+v,+zvl128b -riscv-v-vector-bits-min=-1 -riscv-v-fixed-length-vector-lmul-max=1 -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,RV64,LMULMAX1,LMULMAX1-RV64
 
 define void @add_v16i8(<16 x i8>* %x, <16 x i8>* %y) {
 ; CHECK-LABEL: add_v16i8:
Index: llvm/lib/Target/RISCV/RISCVSubtarget.cpp
===================================================================
--- llvm/lib/Target/RISCV/RISCVSubtarget.cpp
+++ llvm/lib/Target/RISCV/RISCVSubtarget.cpp
@@ -28,16 +28,18 @@
 #define GET_SUBTARGETINFO_CTOR
 #include "RISCVGenSubtargetInfo.inc"
 
-static cl::opt<unsigned> RVVVectorBitsMax(
+static cl::opt<int> RVVVectorBitsMax(
     "riscv-v-vector-bits-max",
     cl::desc("Assume V extension vector registers are at most this big, "
              "with zero meaning no maximum size is assumed."),
     cl::init(0), cl::Hidden);
 
-static cl::opt<unsigned> RVVVectorBitsMin(
+static cl::opt<int> RVVVectorBitsMin(
     "riscv-v-vector-bits-min",
     cl::desc("Assume V extension vector registers are at least this big, "
-             "with zero meaning no minimum size is assumed."),
+             "with zero meaning no minimum size is assumed. A value of -1 "
+             "means use Zvl*b extension. This is primarily used to enable "
+             "autovectorization with fixed width vectors."),
     cl::init(0), cl::Hidden);
 
 static cl::opt<unsigned> RVVVectorLMULMax(
@@ -136,7 +138,7 @@
 
   // ZvlLen specifies the minimum required vlen. The upper bound provided by
   // riscv-v-vector-bits-max should be no less than it.
-  if (RVVVectorBitsMax < ZvlLen)
+  if (RVVVectorBitsMax < (int)ZvlLen)
     report_fatal_error("riscv-v-vector-bits-max specified is lower "
                        "than the Zvl*b limitation");
 
@@ -154,14 +156,18 @@
 }
 
 unsigned RISCVSubtarget::getMinRVVVectorSizeInBits() const {
+  assert(hasVInstructions() &&
+         "Tried to get vector length without Zve or V extension support!");
+
+  if (RVVVectorBitsMin == -1)
+    return ZvlLen;
+
   // ZvlLen specifies the minimum required vlen. The lower bound provided by
   // riscv-v-vector-bits-min should be no less than it.
-  if (RVVVectorBitsMin != 0 && RVVVectorBitsMin < ZvlLen)
+  if (RVVVectorBitsMin != 0 && RVVVectorBitsMin < (int)ZvlLen)
     report_fatal_error("riscv-v-vector-bits-min specified is lower "
                        "than the Zvl*b limitation");
 
-  assert(hasVInstructions() &&
-         "Tried to get vector length without Zve or V extension support!");
   // FIXME: Change to >= 32 when VLEN = 32 is supported
   assert(
       (RVVVectorBitsMin == 0 ||


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D124960.427123.patch
Type: text/x-patch
Size: 3688 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20220504/3c82d3a5/attachment.bin>