[clang] 6a097e2 - [10/11][POC][Clang][RISCV] Define vget for tuple type

[via cfe-commits]

Author: eopXD
Date: 2023-05-22T04:29:28-07:00
New Revision: 6a097e279c7831d12e9ed56ca5cc021e68315c12

URL: https://github.com/llvm/llvm-project/commit/6a097e279c7831d12e9ed56ca5cc021e68315c12
DIFF: https://github.com/llvm/llvm-project/commit/6a097e279c7831d12e9ed56ca5cc021e68315c12.diff

LOG: [10/11][POC][Clang][RISCV] Define vget for tuple type

For the cover letter of this patch-set, please checkout D146872.

Depends on D147915.

This is the 10th patch of the patch-set.

This patch is a proof-of-concept and will be extended to full coverage
in the future. Only vget for tuple type of NF=2, EEW=32, LMUL=1 is
defined now.

Reviewed By: craig.topper

Differential Revision: https://reviews.llvm.org/D147916

Added: 
    clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/vget_tuple.c

Modified: 
    clang/include/clang/Basic/riscv_vector.td
    clang/lib/Sema/SemaChecking.cpp
    clang/lib/Support/RISCVVIntrinsicUtils.cpp
    clang/test/CodeGen/RISCV/rvv-intrinsics-handcrafted/vget-index-out-of-range.c

Removed: 
    


################################################################################
diff  --git a/clang/include/clang/Basic/riscv_vector.td b/clang/include/clang/Basic/riscv_vector.td
index c7e6a1b76dab9..3b070109b2e3d 100644
--- a/clang/include/clang/Basic/riscv_vector.td
+++ b/clang/include/clang/Basic/riscv_vector.td
@@ -2536,6 +2536,11 @@ let HasMasked = false, HasVL = false, IRName = "" in {
   let Name = "vget_v", MaskedPolicyScheme = NonePolicy,
       ManualCodegen = [{
       {
+        if (isa<StructType>(Ops[0]->getType())) // For tuple type
+          // Extract value from index (operand 1) of vtuple (operand 0)
+          return Builder.CreateExtractValue(
+            Ops[0],
+            {(unsigned)cast<ConstantInt>(Ops[1])->getZExtValue()});
         auto *VecTy = cast<ScalableVectorType>(ResultType);
         auto *OpVecTy = cast<ScalableVectorType>(Ops[0]->getType());
         // Mask to only valid indices.
@@ -2553,6 +2558,12 @@ let HasMasked = false, HasVL = false, IRName = "" in {
       def : RVVBuiltin<"v" # dst_lmul # "v", dst_lmul # "vvKz", "csilxfd", dst_lmul # "v">;
       def : RVVBuiltin<"Uv" # dst_lmul # "Uv", dst_lmul # "UvUvKz", "csil", dst_lmul # "Uv">;
     }
+    foreach nf = [2] in {
+      let Log2LMUL = [0] in {
+        defvar T = "(Tuple:" # nf # ")";
+        def : RVVBuiltin<T # "vv", "v" # T # "vKz", "i", "v">;
+      }
+    }
   }
 
   let Name = "vset_v", Log2LMUL = [0, 1, 2], MaskedPolicyScheme = NonePolicy,

diff  --git a/clang/lib/Sema/SemaChecking.cpp b/clang/lib/Sema/SemaChecking.cpp
index 06739beba4f91..7aceade260ad3 100644
--- a/clang/lib/Sema/SemaChecking.cpp
+++ b/clang/lib/Sema/SemaChecking.cpp
@@ -4522,9 +4522,12 @@ bool Sema::CheckRISCVBuiltinFunctionCall(const TargetInfo &TI,
     ASTContext::BuiltinVectorTypeInfo VecInfo =
         Context.getBuiltinVectorTypeInfo(cast<BuiltinType>(
             TheCall->getArg(0)->getType().getCanonicalType().getTypePtr()));
-    unsigned MaxIndex =
-        (VecInfo.EC.getKnownMinValue() * VecInfo.NumVectors) /
-        (ResVecInfo.EC.getKnownMinValue() * ResVecInfo.NumVectors);
+    unsigned MaxIndex;
+    if (VecInfo.NumVectors != 1) // vget for tuple type
+      MaxIndex = VecInfo.NumVectors;
+    else // vget for non-tuple type
+      MaxIndex = (VecInfo.EC.getKnownMinValue() * VecInfo.NumVectors) /
+                 (ResVecInfo.EC.getKnownMinValue() * ResVecInfo.NumVectors);
     return SemaBuiltinConstantArgRange(TheCall, 1, 0, MaxIndex - 1);
   }
   case RISCVVector::BI__builtin_rvv_vset_v: {

diff  --git a/clang/lib/Support/RISCVVIntrinsicUtils.cpp b/clang/lib/Support/RISCVVIntrinsicUtils.cpp
index 353e0cb4422f2..74849d8137484 100644
--- a/clang/lib/Support/RISCVVIntrinsicUtils.cpp
+++ b/clang/lib/Support/RISCVVIntrinsicUtils.cpp
@@ -332,6 +332,8 @@ void RVVType::initShortStr() {
   }
   if (isVector())
     ShortStr += LMUL.str();
+  if (isTuple())
+    ShortStr += "x" + utostr(NF);
 }
 
 void RVVType::applyBasicType() {

diff  --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/vget_tuple.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/vget_tuple.c
new file mode 100644
index 0000000000000..48a797ad3ab70
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/vget_tuple.c
@@ -0,0 +1,20 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 2
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v -target-feature +zfh \
+// RUN:   -target-feature +experimental-zvfh -disable-O0-optnone  \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x i32> @test_vget_v_i32m1x2_i32m1
+// CHECK-RV64-SAME: (<vscale x 2 x i32> [[SRC_COERCE0:%.*]], <vscale x 2 x i32> [[SRC_COERCE1:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 2 x i32>, <vscale x 2 x i32> } poison, <vscale x 2 x i32> [[SRC_COERCE0]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = insertvalue { <vscale x 2 x i32>, <vscale x 2 x i32> } [[TMP0]], <vscale x 2 x i32> [[SRC_COERCE1]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x i32>, <vscale x 2 x i32> } [[TMP1]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP2]]
+//
+vint32m1_t test_vget_v_i32m1x2_i32m1(vint32m1x2_t src) {
+  return __riscv_vget_v_i32m1x2_i32m1(src, 0);
+}

diff  --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-handcrafted/vget-index-out-of-range.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-handcrafted/vget-index-out-of-range.c
index bd8d38b791e93..8ce6ffed7b4d5 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-handcrafted/vget-index-out-of-range.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-handcrafted/vget-index-out-of-range.c
@@ -339,3 +339,8 @@ vfloat16m4_t test_vget_v_f16m8_f16m4(vfloat16m8_t src) {
   // expected-error at +1 {{argument value 2 is outside the valid range [0, 1]}}
   return __riscv_vget_v_f16m8_f16m4(src, 2);
 }
+
+vint32m1_t test_vget_v_i32m1x2_i32m1(vint32m1x2_t src) {
+  // expected-error at +1 {{argument value 2 is outside the valid range [0, 1]}}
+  return __riscv_vget_v_i32m1x2_i32m1(src, 2);
+}