[llvm] [IA][RISCV] Support VP loads/stores in InterleavedAccessPass (PR #120490)

[Luke Lau via llvm-commits]

================
@@ -0,0 +1,816 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 2
+; RUN: llc < %s -mtriple=riscv32 -mattr=+v,m -O2 | FileCheck -check-prefixes=CHECK,RV32 %s
+; RUN: llc < %s -mtriple=riscv64 -mattr=+v,m -O2 | FileCheck -check-prefixes=CHECK,RV64 %s
+
+define {<vscale x 2 x i32>, <vscale x 2 x i32>} @load_factor2_v2(ptr %ptr, i32 %evl) {
+; RV32-LABEL: load_factor2_v2:
+; RV32:       # %bb.0:
+; RV32-NEXT:    slli a1, a1, 1
+; RV32-NEXT:    srli a1, a1, 1
+; RV32-NEXT:    vsetvli zero, a1, e32, m1, ta, ma
+; RV32-NEXT:    vlseg2e32.v v8, (a0)
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: load_factor2_v2:
+; RV64:       # %bb.0:
+; RV64-NEXT:    slli a1, a1, 33
+; RV64-NEXT:    srli a1, a1, 33
+; RV64-NEXT:    vsetvli zero, a1, e32, m1, ta, ma
+; RV64-NEXT:    vlseg2e32.v v8, (a0)
+; RV64-NEXT:    ret
+  %rvl = mul i32 %evl, 2
+  %wide.masked.load = call <vscale x 4 x i32> @llvm.vp.load.nxv4i32.p0(ptr %ptr, <vscale x 4 x i1> splat (i1 true), i32 %rvl)
+  %deinterleaved.results = call { <vscale x 2 x i32>, <vscale x 2 x i32> } @llvm.vector.deinterleave2.nxv4i32(<vscale x 4 x i32> %wide.masked.load)
+  %t0 = extractvalue { <vscale x 2 x i32>, <vscale x 2 x i32> } %deinterleaved.results, 0
+  %t1 = extractvalue { <vscale x 2 x i32>, <vscale x 2 x i32> } %deinterleaved.results, 1
+  %res0 = insertvalue { <vscale x 2 x i32>, <vscale x 2 x i32> } undef, <vscale x 2 x i32> %t0, 0
+  %res1 = insertvalue { <vscale x 2 x i32>, <vscale x 2 x i32> } %res0, <vscale x 2 x i32> %t1, 1
+  ret { <vscale x 2 x i32>, <vscale x 2 x i32> } %res1
+}
+
+define {<vscale x 2 x i32>, <vscale x 2 x i32>, <vscale x 2 x i32>, <vscale x 2 x i32>} @load_factor4_v2(ptr %ptr, i32 %evl) {
+; RV32-LABEL: load_factor4_v2:
+; RV32:       # %bb.0:
+; RV32-NEXT:    slli a1, a1, 2
+; RV32-NEXT:    srli a1, a1, 2
+; RV32-NEXT:    vsetvli zero, a1, e32, m1, ta, ma
+; RV32-NEXT:    vlseg4e32.v v8, (a0)
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: load_factor4_v2:
+; RV64:       # %bb.0:
+; RV64-NEXT:    slli a1, a1, 34
+; RV64-NEXT:    srli a1, a1, 34
+; RV64-NEXT:    vsetvli zero, a1, e32, m1, ta, ma
+; RV64-NEXT:    vlseg4e32.v v8, (a0)
+; RV64-NEXT:    ret
+  %rvl = mul i32 %evl, 4
+  %wide.masked.load = call <vscale x 8 x i32> @llvm.vp.load.nxv8i32.p0(ptr %ptr, <vscale x 8 x i1> splat (i1 true), i32 %rvl)
+  %d0 = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> %wide.masked.load)
+  %d0.0 = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } %d0, 0
+  %d0.1 = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } %d0, 1
+  %d1 = call { <vscale x 2 x i32>, <vscale x 2 x i32> } @llvm.vector.deinterleave2.nxv4i32(<vscale x 4 x i32> %d0.0)
+  %t0 = extractvalue { <vscale x 2 x i32>, <vscale x 2 x i32> } %d1, 0
+  %t2 = extractvalue { <vscale x 2 x i32>, <vscale x 2 x i32> } %d1, 1
+  %d2 = call { <vscale x 2 x i32>, <vscale x 2 x i32> } @llvm.vector.deinterleave2.nxv4i32(<vscale x 4 x i32> %d0.1)
+  %t1 = extractvalue { <vscale x 2 x i32>, <vscale x 2 x i32> } %d2, 0
+  %t3 = extractvalue { <vscale x 2 x i32>, <vscale x 2 x i32> } %d2, 1
+
+  %res0 = insertvalue { <vscale x 2 x i32>, <vscale x 2 x i32>, <vscale x 2 x i32>, <vscale x 2 x i32> } undef, <vscale x 2 x i32> %t0, 0
----------------
lukel97 wrote:

Should probably replace undef with poison in this file

https://github.com/llvm/llvm-project/pull/120490