| ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py |
| ; 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 |
| ; 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 |
| |
| define void @buildvec_vid_v16i8(ptr %x) { |
| ; CHECK-LABEL: buildvec_vid_v16i8: |
| ; CHECK: # %bb.0: |
| ; CHECK-NEXT: vsetivli zero, 16, e8, m1, ta, ma |
| ; CHECK-NEXT: vid.v v8 |
| ; CHECK-NEXT: vse8.v v8, (a0) |
| ; CHECK-NEXT: ret |
| store <16 x i8> <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 8, i8 9, i8 10, i8 11, i8 12, i8 13, i8 14, i8 15>, ptr %x |
| ret void |
| } |
| |
| define void @buildvec_vid_undefelts_v16i8(ptr %x) { |
| ; CHECK-LABEL: buildvec_vid_undefelts_v16i8: |
| ; CHECK: # %bb.0: |
| ; CHECK-NEXT: vsetivli zero, 16, e8, m1, ta, ma |
| ; CHECK-NEXT: vid.v v8 |
| ; CHECK-NEXT: vse8.v v8, (a0) |
| ; CHECK-NEXT: ret |
| store <16 x i8> <i8 0, i8 1, i8 2, i8 undef, i8 4, i8 undef, i8 6, i8 7, i8 8, i8 9, i8 10, i8 11, i8 12, i8 13, i8 14, i8 15>, ptr %x |
| ret void |
| } |
| |
| ; TODO: Could do VID then insertelement on missing elements |
| define void @buildvec_notquite_vid_v16i8(ptr %x) { |
| ; CHECK-LABEL: buildvec_notquite_vid_v16i8: |
| ; CHECK: # %bb.0: |
| ; CHECK-NEXT: lui a1, %hi(.LCPI2_0) |
| ; CHECK-NEXT: addi a1, a1, %lo(.LCPI2_0) |
| ; CHECK-NEXT: vsetivli zero, 16, e8, m1, ta, ma |
| ; CHECK-NEXT: vle8.v v8, (a1) |
| ; CHECK-NEXT: vse8.v v8, (a0) |
| ; CHECK-NEXT: ret |
| store <16 x i8> <i8 0, i8 1, i8 3, i8 3, i8 4, i8 5, i8 6, i8 7, i8 8, i8 9, i8 10, i8 11, i8 12, i8 13, i8 14, i8 15>, ptr %x |
| ret void |
| } |
| |
| define void @buildvec_vid_plus_imm_v16i8(ptr %x) { |
| ; CHECK-LABEL: buildvec_vid_plus_imm_v16i8: |
| ; CHECK: # %bb.0: |
| ; CHECK-NEXT: vsetivli zero, 16, e8, m1, ta, ma |
| ; CHECK-NEXT: vid.v v8 |
| ; CHECK-NEXT: vadd.vi v8, v8, 2 |
| ; CHECK-NEXT: vse8.v v8, (a0) |
| ; CHECK-NEXT: ret |
| store <16 x i8> <i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 8, i8 9, i8 10, i8 11, i8 12, i8 13, i8 14, i8 15, i8 16, i8 17>, ptr %x |
| ret void |
| } |
| |
| define void @buildvec_vid_mpy_imm_v16i8(ptr %x) { |
| ; CHECK-LABEL: buildvec_vid_mpy_imm_v16i8: |
| ; CHECK: # %bb.0: |
| ; CHECK-NEXT: vsetivli zero, 16, e8, m1, ta, ma |
| ; CHECK-NEXT: vid.v v8 |
| ; CHECK-NEXT: li a1, 3 |
| ; CHECK-NEXT: vmul.vx v8, v8, a1 |
| ; CHECK-NEXT: vse8.v v8, (a0) |
| ; CHECK-NEXT: ret |
| store <16 x i8> <i8 0, i8 3, i8 6, i8 9, i8 12, i8 15, i8 18, i8 21, i8 24, i8 27, i8 30, i8 33, i8 36, i8 39, i8 42, i8 45>, ptr %x |
| ret void |
| } |
| |
| ; Some tests return this struct because the stores end up being scalarized. |
| %x4v4i8 = type {<4 x i8>, <4 x i8>, <4 x i8>, <4 x i8>} |
| |
| define %x4v4i8 @buildvec_vid_step2_add0_v4i8() { |
| ; CHECK-LABEL: buildvec_vid_step2_add0_v4i8: |
| ; CHECK: # %bb.0: |
| ; CHECK-NEXT: vsetivli zero, 4, e8, mf4, ta, ma |
| ; CHECK-NEXT: vid.v v8 |
| ; CHECK-NEXT: vadd.vv v8, v8, v8 |
| ; CHECK-NEXT: vmv1r.v v9, v8 |
| ; CHECK-NEXT: vmv1r.v v10, v8 |
| ; CHECK-NEXT: vmv1r.v v11, v8 |
| ; CHECK-NEXT: ret |
| %1 = insertvalue %x4v4i8 poison, <4 x i8> <i8 0, i8 2, i8 4, i8 6>, 0 |
| %2 = insertvalue %x4v4i8 %1, <4 x i8> <i8 undef, i8 2, i8 4, i8 6>, 1 |
| %3 = insertvalue %x4v4i8 %2, <4 x i8> <i8 undef, i8 undef, i8 4, i8 6>, 2 |
| %4 = insertvalue %x4v4i8 %3, <4 x i8> <i8 0, i8 undef, i8 undef, i8 6>, 3 |
| ret %x4v4i8 %4 |
| } |
| |
| define %x4v4i8 @buildvec_vid_step2_add1_v4i8() { |
| ; CHECK-LABEL: buildvec_vid_step2_add1_v4i8: |
| ; CHECK: # %bb.0: |
| ; CHECK-NEXT: vsetivli zero, 4, e8, mf4, ta, ma |
| ; CHECK-NEXT: vid.v v8 |
| ; CHECK-NEXT: vadd.vv v8, v8, v8 |
| ; CHECK-NEXT: vadd.vi v8, v8, 1 |
| ; CHECK-NEXT: vmv1r.v v9, v8 |
| ; CHECK-NEXT: vmv1r.v v10, v8 |
| ; CHECK-NEXT: vmv1r.v v11, v8 |
| ; CHECK-NEXT: ret |
| %1 = insertvalue %x4v4i8 poison, <4 x i8> <i8 1, i8 3, i8 5, i8 7>, 0 |
| %2 = insertvalue %x4v4i8 %1, <4 x i8> <i8 undef, i8 3, i8 5, i8 7>, 1 |
| %3 = insertvalue %x4v4i8 %2, <4 x i8> <i8 undef, i8 undef, i8 5, i8 7>, 2 |
| %4 = insertvalue %x4v4i8 %3, <4 x i8> <i8 1, i8 undef, i8 undef, i8 7>, 3 |
| ret %x4v4i8 %4 |
| } |
| |
| ; FIXME: This could generate vrsub.vi but the (ISD::MUL X, -1) we generate |
| ; while lowering ISD::BUILD_VECTOR is custom-lowered to RISCVISD::MUL_VL before |
| ; being combined. |
| define %x4v4i8 @buildvec_vid_stepn1_add0_v4i8() { |
| ; CHECK-LABEL: buildvec_vid_stepn1_add0_v4i8: |
| ; CHECK: # %bb.0: |
| ; CHECK-NEXT: vsetivli zero, 4, e8, mf4, ta, ma |
| ; CHECK-NEXT: vid.v v8 |
| ; CHECK-NEXT: vrsub.vi v8, v8, 0 |
| ; CHECK-NEXT: vmv1r.v v9, v8 |
| ; CHECK-NEXT: vmv1r.v v10, v8 |
| ; CHECK-NEXT: vmv1r.v v11, v8 |
| ; CHECK-NEXT: ret |
| %1 = insertvalue %x4v4i8 poison, <4 x i8> <i8 0, i8 -1, i8 -2, i8 -3>, 0 |
| %2 = insertvalue %x4v4i8 %1, <4 x i8> <i8 undef, i8 -1, i8 -2, i8 -3>, 1 |
| %3 = insertvalue %x4v4i8 %2, <4 x i8> <i8 undef, i8 undef, i8 -2, i8 -3>, 2 |
| %4 = insertvalue %x4v4i8 %3, <4 x i8> <i8 0, i8 undef, i8 undef, i8 -3>, 3 |
| ret %x4v4i8 %4 |
| } |
| |
| define %x4v4i8 @buildvec_vid_stepn2_add0_v4i8() { |
| ; CHECK-LABEL: buildvec_vid_stepn2_add0_v4i8: |
| ; CHECK: # %bb.0: |
| ; CHECK-NEXT: vsetivli zero, 4, e8, mf4, ta, ma |
| ; CHECK-NEXT: vid.v v8 |
| ; CHECK-NEXT: vadd.vv v8, v8, v8 |
| ; CHECK-NEXT: vrsub.vi v8, v8, 0 |
| ; CHECK-NEXT: vmv.v.i v11, -6 |
| ; CHECK-NEXT: vmv1r.v v9, v8 |
| ; CHECK-NEXT: vmv1r.v v10, v8 |
| ; CHECK-NEXT: ret |
| %1 = insertvalue %x4v4i8 poison, <4 x i8> <i8 0, i8 -2, i8 -4, i8 -6>, 0 |
| %2 = insertvalue %x4v4i8 %1, <4 x i8> <i8 undef, i8 -2, i8 -4, i8 -6>, 1 |
| %3 = insertvalue %x4v4i8 %2, <4 x i8> <i8 undef, i8 undef, i8 -4, i8 -6>, 2 |
| %4 = insertvalue %x4v4i8 %3, <4 x i8> <i8 undef, i8 undef, i8 undef, i8 -6>, 3 |
| ret %x4v4i8 %4 |
| } |
| |
| define <4 x i8> @buildvec_vid_stepn2_add3_v4i8() { |
| ; CHECK-LABEL: buildvec_vid_stepn2_add3_v4i8: |
| ; CHECK: # %bb.0: |
| ; CHECK-NEXT: vsetivli zero, 4, e8, mf4, ta, ma |
| ; CHECK-NEXT: vid.v v8 |
| ; CHECK-NEXT: vadd.vv v8, v8, v8 |
| ; CHECK-NEXT: vrsub.vi v8, v8, 3 |
| ; CHECK-NEXT: ret |
| ret <4 x i8> <i8 3, i8 1, i8 -1, i8 -3> |
| } |
| |
| define <4 x i8> @buildvec_vid_stepn3_add3_v4i8() { |
| ; CHECK-LABEL: buildvec_vid_stepn3_add3_v4i8: |
| ; CHECK: # %bb.0: |
| ; CHECK-NEXT: vsetivli zero, 4, e8, mf4, ta, ma |
| ; CHECK-NEXT: vmv.v.i v9, 3 |
| ; CHECK-NEXT: vid.v v8 |
| ; CHECK-NEXT: li a0, -3 |
| ; CHECK-NEXT: vmadd.vx v8, a0, v9 |
| ; CHECK-NEXT: ret |
| ret <4 x i8> <i8 3, i8 0, i8 -3, i8 -6> |
| } |
| |
| define void @buildvec_vid_stepn3_addn3_v4i32(ptr %z0, ptr %z1, ptr %z2, ptr %z3) { |
| ; CHECK-LABEL: buildvec_vid_stepn3_addn3_v4i32: |
| ; CHECK: # %bb.0: |
| ; CHECK-NEXT: vsetivli zero, 4, e32, m1, ta, ma |
| ; CHECK-NEXT: vmv.v.i v8, -3 |
| ; CHECK-NEXT: vid.v v9 |
| ; CHECK-NEXT: li a4, -3 |
| ; CHECK-NEXT: vmadd.vx v9, a4, v8 |
| ; CHECK-NEXT: vse32.v v9, (a0) |
| ; CHECK-NEXT: vse32.v v9, (a1) |
| ; CHECK-NEXT: vse32.v v9, (a2) |
| ; CHECK-NEXT: vse32.v v9, (a3) |
| ; CHECK-NEXT: ret |
| store <4 x i32> <i32 -3, i32 -6, i32 -9, i32 -12>, ptr %z0 |
| store <4 x i32> <i32 undef, i32 -6, i32 -9, i32 -12>, ptr %z1 |
| store <4 x i32> <i32 undef, i32 undef, i32 -9, i32 -12>, ptr %z2 |
| store <4 x i32> <i32 -3, i32 undef, i32 undef, i32 -12>, ptr %z3 |
| ret void |
| } |
| |
| ; FIXME: RV32 doesn't catch this pattern due to BUILD_VECTOR legalization. |
| define <4 x i64> @buildvec_vid_step1_add0_v4i64() { |
| ; RV32-LABEL: buildvec_vid_step1_add0_v4i64: |
| ; RV32: # %bb.0: |
| ; RV32-NEXT: vsetivli zero, 4, e32, m1, ta, ma |
| ; RV32-NEXT: vmv.v.i v9, 1 |
| ; RV32-NEXT: vmv.v.i v8, 0 |
| ; RV32-NEXT: vsetivli zero, 3, e32, m1, tu, ma |
| ; RV32-NEXT: vslideup.vi v8, v9, 2 |
| ; RV32-NEXT: lui a0, %hi(.LCPI12_0) |
| ; RV32-NEXT: addi a0, a0, %lo(.LCPI12_0) |
| ; RV32-NEXT: vsetivli zero, 4, e32, m1, ta, ma |
| ; RV32-NEXT: vle32.v v9, (a0) |
| ; RV32-NEXT: ret |
| ; |
| ; RV64-LABEL: buildvec_vid_step1_add0_v4i64: |
| ; RV64: # %bb.0: |
| ; RV64-NEXT: vsetivli zero, 2, e64, m1, ta, ma |
| ; RV64-NEXT: vid.v v8 |
| ; RV64-NEXT: vadd.vi v9, v8, 2 |
| ; RV64-NEXT: ret |
| ret <4 x i64> <i64 0, i64 1, i64 2, i64 3> |
| } |
| |
| define <4 x i64> @buildvec_vid_step2_add0_v4i64() { |
| ; RV32-LABEL: buildvec_vid_step2_add0_v4i64: |
| ; RV32: # %bb.0: |
| ; RV32-NEXT: vsetivli zero, 4, e32, m1, ta, ma |
| ; RV32-NEXT: vmv.v.i v9, 2 |
| ; RV32-NEXT: vmv.v.i v8, 0 |
| ; RV32-NEXT: vsetivli zero, 3, e32, m1, tu, ma |
| ; RV32-NEXT: vslideup.vi v8, v9, 2 |
| ; RV32-NEXT: lui a0, %hi(.LCPI13_0) |
| ; RV32-NEXT: addi a0, a0, %lo(.LCPI13_0) |
| ; RV32-NEXT: vsetivli zero, 4, e32, m1, ta, ma |
| ; RV32-NEXT: vle32.v v9, (a0) |
| ; RV32-NEXT: ret |
| ; |
| ; RV64-LABEL: buildvec_vid_step2_add0_v4i64: |
| ; RV64: # %bb.0: |
| ; RV64-NEXT: vsetivli zero, 2, e64, m1, ta, ma |
| ; RV64-NEXT: vid.v v8 |
| ; RV64-NEXT: vadd.vv v8, v8, v8 |
| ; RV64-NEXT: vadd.vi v9, v8, 4 |
| ; RV64-NEXT: ret |
| ret <4 x i64> <i64 0, i64 2, i64 4, i64 6> |
| } |
| |
| %x6v4i8 = type {<4 x i8>, <4 x i8>, <4 x i8>, <4 x i8>, <4 x i8>, <4 x i8>} |
| |
| define %x6v4i8 @buildvec_no_vid_v4i8() { |
| ; CHECK-LABEL: buildvec_no_vid_v4i8: |
| ; CHECK: # %bb.0: |
| ; CHECK-NEXT: lui a0, %hi(.LCPI14_0) |
| ; CHECK-NEXT: addi a0, a0, %lo(.LCPI14_0) |
| ; CHECK-NEXT: vsetivli zero, 4, e8, mf4, ta, ma |
| ; CHECK-NEXT: vle8.v v8, (a0) |
| ; CHECK-NEXT: lui a0, %hi(.LCPI14_1) |
| ; CHECK-NEXT: addi a0, a0, %lo(.LCPI14_1) |
| ; CHECK-NEXT: vle8.v v9, (a0) |
| ; CHECK-NEXT: li a0, 1 |
| ; CHECK-NEXT: slli a0, a0, 11 |
| ; CHECK-NEXT: vsetivli zero, 2, e16, mf4, ta, ma |
| ; CHECK-NEXT: vmv.v.x v10, a0 |
| ; CHECK-NEXT: li a0, 2047 |
| ; CHECK-NEXT: vmv.v.x v11, a0 |
| ; CHECK-NEXT: vsetivli zero, 4, e8, mf4, ta, ma |
| ; CHECK-NEXT: lui a0, %hi(.LCPI14_2) |
| ; CHECK-NEXT: addi a0, a0, %lo(.LCPI14_2) |
| ; CHECK-NEXT: vle8.v v13, (a0) |
| ; CHECK-NEXT: vmv.v.i v12, -2 |
| ; CHECK-NEXT: ret |
| %1 = insertvalue %x6v4i8 poison, <4 x i8> <i8 1, i8 3, i8 6, i8 7>, 0 |
| %2 = insertvalue %x6v4i8 %1, <4 x i8> <i8 undef, i8 2, i8 5, i8 7>, 1 |
| %3 = insertvalue %x6v4i8 %2, <4 x i8> <i8 0, i8 undef, i8 undef, i8 8>, 2 |
| %4 = insertvalue %x6v4i8 %3, <4 x i8> <i8 -1, i8 undef, i8 undef, i8 7>, 3 |
| %5 = insertvalue %x6v4i8 %4, <4 x i8> <i8 -2, i8 undef, i8 undef, i8 undef>, 4 |
| %6 = insertvalue %x6v4i8 %5, <4 x i8> <i8 -1, i8 -2, i8 -4, i8 -5>, 5 |
| ret %x6v4i8 %6 |
| } |
| |
| define void @buildvec_dominant0_v8i16(ptr %x) { |
| ; CHECK-LABEL: buildvec_dominant0_v8i16: |
| ; CHECK: # %bb.0: |
| ; CHECK-NEXT: vsetivli zero, 8, e16, m1, ta, ma |
| ; CHECK-NEXT: vmv.s.x v8, zero |
| ; CHECK-NEXT: vmv.v.i v9, 8 |
| ; CHECK-NEXT: vsetivli zero, 4, e16, m1, tu, ma |
| ; CHECK-NEXT: vslideup.vi v9, v8, 3 |
| ; CHECK-NEXT: vsetivli zero, 8, e16, m1, ta, ma |
| ; CHECK-NEXT: vse16.v v9, (a0) |
| ; CHECK-NEXT: ret |
| store <8 x i16> <i16 8, i16 8, i16 undef, i16 0, i16 8, i16 undef, i16 8, i16 8>, ptr %x |
| ret void |
| } |
| |
| define void @buildvec_dominant1_v8i16(ptr %x) { |
| ; CHECK-LABEL: buildvec_dominant1_v8i16: |
| ; CHECK: # %bb.0: |
| ; CHECK-NEXT: vsetivli zero, 8, e16, m1, ta, ma |
| ; CHECK-NEXT: vmv.v.i v8, 8 |
| ; CHECK-NEXT: vse16.v v8, (a0) |
| ; CHECK-NEXT: ret |
| store <8 x i16> <i16 undef, i16 8, i16 undef, i16 undef, i16 undef, i16 undef, i16 undef, i16 undef>, ptr %x |
| ret void |
| } |
| |
| define <2 x i8> @buildvec_dominant0_v2i8() { |
| ; CHECK-LABEL: buildvec_dominant0_v2i8: |
| ; CHECK: # %bb.0: |
| ; CHECK-NEXT: ret |
| ret <2 x i8> <i8 undef, i8 undef> |
| } |
| |
| define <2 x i8> @buildvec_dominant1_v2i8() { |
| ; CHECK-LABEL: buildvec_dominant1_v2i8: |
| ; CHECK: # %bb.0: |
| ; CHECK-NEXT: vsetivli zero, 2, e8, mf8, ta, ma |
| ; CHECK-NEXT: vmv.v.i v8, -1 |
| ; CHECK-NEXT: ret |
| ret <2 x i8> <i8 undef, i8 -1> |
| } |
| |
| define <2 x i8> @buildvec_dominant2_v2i8() { |
| ; CHECK-LABEL: buildvec_dominant2_v2i8: |
| ; CHECK: # %bb.0: |
| ; CHECK-NEXT: vsetivli zero, 2, e8, mf8, ta, ma |
| ; CHECK-NEXT: vid.v v8 |
| ; CHECK-NEXT: vrsub.vi v8, v8, 0 |
| ; CHECK-NEXT: ret |
| ret <2 x i8> <i8 0, i8 -1> |
| } |
| |
| define void @buildvec_dominant0_v2i32(ptr %x) { |
| ; RV32-LABEL: buildvec_dominant0_v2i32: |
| ; RV32: # %bb.0: |
| ; RV32-NEXT: lui a1, %hi(.LCPI20_0) |
| ; RV32-NEXT: addi a1, a1, %lo(.LCPI20_0) |
| ; RV32-NEXT: vsetivli zero, 4, e32, m1, ta, ma |
| ; RV32-NEXT: vle32.v v8, (a1) |
| ; RV32-NEXT: vse32.v v8, (a0) |
| ; RV32-NEXT: ret |
| ; |
| ; RV64-LABEL: buildvec_dominant0_v2i32: |
| ; RV64: # %bb.0: |
| ; RV64-NEXT: lui a1, %hi(.LCPI20_0) |
| ; RV64-NEXT: ld a1, %lo(.LCPI20_0)(a1) |
| ; RV64-NEXT: vsetivli zero, 2, e64, m1, ta, ma |
| ; RV64-NEXT: vmv.v.i v8, -1 |
| ; RV64-NEXT: vsetvli zero, zero, e64, m1, tu, ma |
| ; RV64-NEXT: vmv.s.x v8, a1 |
| ; RV64-NEXT: vse64.v v8, (a0) |
| ; RV64-NEXT: ret |
| store <2 x i64> <i64 2049638230412172402, i64 -1>, ptr %x |
| ret void |
| } |
| |
| define void @buildvec_dominant1_optsize_v2i32(ptr %x) optsize { |
| ; RV32-LABEL: buildvec_dominant1_optsize_v2i32: |
| ; RV32: # %bb.0: |
| ; RV32-NEXT: lui a1, %hi(.LCPI21_0) |
| ; RV32-NEXT: addi a1, a1, %lo(.LCPI21_0) |
| ; RV32-NEXT: vsetivli zero, 4, e32, m1, ta, ma |
| ; RV32-NEXT: vle32.v v8, (a1) |
| ; RV32-NEXT: vse32.v v8, (a0) |
| ; RV32-NEXT: ret |
| ; |
| ; RV64-LABEL: buildvec_dominant1_optsize_v2i32: |
| ; RV64: # %bb.0: |
| ; RV64-NEXT: lui a1, %hi(.LCPI21_0) |
| ; RV64-NEXT: addi a1, a1, %lo(.LCPI21_0) |
| ; RV64-NEXT: vsetivli zero, 2, e64, m1, ta, ma |
| ; RV64-NEXT: vle64.v v8, (a1) |
| ; RV64-NEXT: vse64.v v8, (a0) |
| ; RV64-NEXT: ret |
| store <2 x i64> <i64 2049638230412172402, i64 -1>, ptr %x |
| ret void |
| } |
| |
| define void @buildvec_seq_v8i8_v4i16(ptr %x) { |
| ; CHECK-LABEL: buildvec_seq_v8i8_v4i16: |
| ; CHECK: # %bb.0: |
| ; CHECK-NEXT: li a1, 513 |
| ; CHECK-NEXT: vsetivli zero, 4, e16, mf2, ta, ma |
| ; CHECK-NEXT: vmv.v.x v8, a1 |
| ; CHECK-NEXT: vsetivli zero, 8, e8, mf2, ta, ma |
| ; CHECK-NEXT: vse8.v v8, (a0) |
| ; CHECK-NEXT: ret |
| store <8 x i8> <i8 1, i8 2, i8 1, i8 2, i8 1, i8 2, i8 undef, i8 2>, ptr %x |
| ret void |
| } |
| |
| define void @buildvec_seq_v8i8_v2i32(ptr %x) { |
| ; RV32-LABEL: buildvec_seq_v8i8_v2i32: |
| ; RV32: # %bb.0: |
| ; RV32-NEXT: lui a1, 48 |
| ; RV32-NEXT: addi a1, a1, 513 |
| ; RV32-NEXT: vsetivli zero, 2, e32, mf2, ta, ma |
| ; RV32-NEXT: vmv.v.x v8, a1 |
| ; RV32-NEXT: vsetivli zero, 8, e8, mf2, ta, ma |
| ; RV32-NEXT: vse8.v v8, (a0) |
| ; RV32-NEXT: ret |
| ; |
| ; RV64-LABEL: buildvec_seq_v8i8_v2i32: |
| ; RV64: # %bb.0: |
| ; RV64-NEXT: lui a1, 48 |
| ; RV64-NEXT: addiw a1, a1, 513 |
| ; RV64-NEXT: vsetivli zero, 2, e32, mf2, ta, ma |
| ; RV64-NEXT: vmv.v.x v8, a1 |
| ; RV64-NEXT: vsetivli zero, 8, e8, mf2, ta, ma |
| ; RV64-NEXT: vse8.v v8, (a0) |
| ; RV64-NEXT: ret |
| store <8 x i8> <i8 1, i8 2, i8 3, i8 undef, i8 1, i8 2, i8 3, i8 undef>, ptr %x |
| ret void |
| } |
| |
| define void @buildvec_seq_v16i8_v2i64(ptr %x) { |
| ; RV32-LABEL: buildvec_seq_v16i8_v2i64: |
| ; RV32: # %bb.0: |
| ; RV32-NEXT: lui a1, %hi(.LCPI24_0) |
| ; RV32-NEXT: addi a1, a1, %lo(.LCPI24_0) |
| ; RV32-NEXT: vsetivli zero, 16, e8, m1, ta, ma |
| ; RV32-NEXT: vle8.v v8, (a1) |
| ; RV32-NEXT: vse8.v v8, (a0) |
| ; RV32-NEXT: ret |
| ; |
| ; RV64-LABEL: buildvec_seq_v16i8_v2i64: |
| ; RV64: # %bb.0: |
| ; RV64-NEXT: lui a1, %hi(.LCPI24_0) |
| ; RV64-NEXT: addi a1, a1, %lo(.LCPI24_0) |
| ; RV64-NEXT: vsetivli zero, 2, e64, m1, ta, ma |
| ; RV64-NEXT: vlse64.v v8, (a1), zero |
| ; RV64-NEXT: vsetivli zero, 16, e8, m1, ta, ma |
| ; RV64-NEXT: vse8.v v8, (a0) |
| ; RV64-NEXT: ret |
| store <16 x i8> <i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 8, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 8>, ptr %x |
| ret void |
| } |
| |
| define void @buildvec_seq2_v16i8_v2i64(ptr %x) { |
| ; RV32-LABEL: buildvec_seq2_v16i8_v2i64: |
| ; RV32: # %bb.0: |
| ; RV32-NEXT: lui a1, 528432 |
| ; RV32-NEXT: addi a1, a1, 513 |
| ; RV32-NEXT: vsetivli zero, 2, e64, m1, ta, ma |
| ; RV32-NEXT: vmv.v.x v8, a1 |
| ; RV32-NEXT: vsetivli zero, 16, e8, m1, ta, ma |
| ; RV32-NEXT: vse8.v v8, (a0) |
| ; RV32-NEXT: ret |
| ; |
| ; RV64-LABEL: buildvec_seq2_v16i8_v2i64: |
| ; RV64: # %bb.0: |
| ; RV64-NEXT: lui a1, 528432 |
| ; RV64-NEXT: addiw a1, a1, 513 |
| ; RV64-NEXT: vsetivli zero, 2, e64, m1, ta, ma |
| ; RV64-NEXT: vmv.v.x v8, a1 |
| ; RV64-NEXT: vsetivli zero, 16, e8, m1, ta, ma |
| ; RV64-NEXT: vse8.v v8, (a0) |
| ; RV64-NEXT: ret |
| store <16 x i8> <i8 1, i8 2, i8 3, i8 129, i8 -1, i8 -1, i8 -1, i8 -1, i8 1, i8 2, i8 3, i8 129, i8 -1, i8 -1, i8 -1, i8 -1>, ptr %x |
| ret void |
| } |
| |
| define void @buildvec_seq_v9i8(ptr %x) { |
| ; RV32-LABEL: buildvec_seq_v9i8: |
| ; RV32: # %bb.0: |
| ; RV32-NEXT: li a1, 3 |
| ; RV32-NEXT: sb a1, 8(a0) |
| ; RV32-NEXT: li a1, 73 |
| ; RV32-NEXT: vsetivli zero, 8, e8, mf2, ta, ma |
| ; RV32-NEXT: vmv.s.x v0, a1 |
| ; RV32-NEXT: vmv.v.i v9, 2 |
| ; RV32-NEXT: li a1, 36 |
| ; RV32-NEXT: vmv.s.x v8, a1 |
| ; RV32-NEXT: vmerge.vim v9, v9, 1, v0 |
| ; RV32-NEXT: vmv1r.v v0, v8 |
| ; RV32-NEXT: vmerge.vim v8, v9, 3, v0 |
| ; RV32-NEXT: vse8.v v8, (a0) |
| ; RV32-NEXT: ret |
| ; |
| ; RV64-LABEL: buildvec_seq_v9i8: |
| ; RV64: # %bb.0: |
| ; RV64-NEXT: lui a1, %hi(.LCPI26_0) |
| ; RV64-NEXT: ld a1, %lo(.LCPI26_0)(a1) |
| ; RV64-NEXT: li a2, 3 |
| ; RV64-NEXT: sb a2, 8(a0) |
| ; RV64-NEXT: sd a1, 0(a0) |
| ; RV64-NEXT: ret |
| store <9 x i8> <i8 1, i8 2, i8 3, i8 1, i8 2, i8 3, i8 1, i8 2, i8 3>, ptr %x |
| ret void |
| } |
| |
| define void @buildvec_seq_v4i16_v2i32(ptr %x) { |
| ; CHECK-LABEL: buildvec_seq_v4i16_v2i32: |
| ; CHECK: # %bb.0: |
| ; CHECK-NEXT: li a1, -127 |
| ; CHECK-NEXT: vsetivli zero, 2, e32, mf2, ta, ma |
| ; CHECK-NEXT: vmv.v.x v8, a1 |
| ; CHECK-NEXT: vsetivli zero, 4, e16, mf2, ta, ma |
| ; CHECK-NEXT: vse16.v v8, (a0) |
| ; CHECK-NEXT: ret |
| store <4 x i16> <i16 -127, i16 -1, i16 -127, i16 -1>, ptr %x |
| ret void |
| } |
| |
| define void @buildvec_vid_step1o2_v4i32(ptr %z0, ptr %z1, ptr %z2, ptr %z3, ptr %z4, ptr %z5, ptr %z6) { |
| ; CHECK-LABEL: buildvec_vid_step1o2_v4i32: |
| ; CHECK: # %bb.0: |
| ; CHECK-NEXT: vsetivli zero, 4, e32, m1, ta, ma |
| ; CHECK-NEXT: vid.v v8 |
| ; CHECK-NEXT: vsrl.vi v8, v8, 1 |
| ; CHECK-NEXT: vse32.v v8, (a0) |
| ; CHECK-NEXT: vse32.v v8, (a1) |
| ; CHECK-NEXT: vse32.v v8, (a2) |
| ; CHECK-NEXT: vse32.v v8, (a3) |
| ; CHECK-NEXT: vse32.v v8, (a4) |
| ; CHECK-NEXT: vmv.s.x v8, zero |
| ; CHECK-NEXT: vmv.v.i v9, 1 |
| ; CHECK-NEXT: vsetivli zero, 2, e32, m1, tu, ma |
| ; CHECK-NEXT: vmv1r.v v10, v9 |
| ; CHECK-NEXT: vslideup.vi v10, v8, 1 |
| ; CHECK-NEXT: vsetivli zero, 4, e32, m1, ta, ma |
| ; CHECK-NEXT: vse32.v v10, (a5) |
| ; CHECK-NEXT: vmv.v.i v8, 0 |
| ; CHECK-NEXT: vslideup.vi v8, v9, 3 |
| ; CHECK-NEXT: vse32.v v8, (a6) |
| ; CHECK-NEXT: ret |
| store <4 x i32> <i32 0, i32 0, i32 1, i32 1>, ptr %z0 |
| store <4 x i32> <i32 0, i32 0, i32 1, i32 undef>, ptr %z1 |
| store <4 x i32> <i32 0, i32 undef, i32 1, i32 1>, ptr %z2 |
| store <4 x i32> <i32 undef, i32 0, i32 undef, i32 1>, ptr %z3 |
| store <4 x i32> <i32 0, i32 undef, i32 1, i32 undef>, ptr %z4 |
| ; We don't catch this one |
| store <4 x i32> <i32 undef, i32 0, i32 1, i32 1>, ptr %z5 |
| ; We catch this one but as VID/3 rather than VID/2 |
| store <4 x i32> <i32 0, i32 0, i32 undef, i32 1>, ptr %z6 |
| ret void |
| } |
| |
| define void @buildvec_vid_step1o2_add3_v4i16(ptr %z0, ptr %z1, ptr %z2, ptr %z3, ptr %z4, ptr %z5, ptr %z6) { |
| ; CHECK-LABEL: buildvec_vid_step1o2_add3_v4i16: |
| ; CHECK: # %bb.0: |
| ; CHECK-NEXT: vsetivli zero, 4, e16, mf2, ta, ma |
| ; CHECK-NEXT: vid.v v8 |
| ; CHECK-NEXT: vsrl.vi v8, v8, 1 |
| ; CHECK-NEXT: vadd.vi v8, v8, 3 |
| ; CHECK-NEXT: vse16.v v8, (a0) |
| ; CHECK-NEXT: vse16.v v8, (a1) |
| ; CHECK-NEXT: vse16.v v8, (a2) |
| ; CHECK-NEXT: vse16.v v8, (a3) |
| ; CHECK-NEXT: vse16.v v8, (a4) |
| ; CHECK-NEXT: vmv.v.i v8, 3 |
| ; CHECK-NEXT: vmv.v.i v9, 4 |
| ; CHECK-NEXT: vsetivli zero, 2, e16, mf2, tu, ma |
| ; CHECK-NEXT: vmv1r.v v10, v9 |
| ; CHECK-NEXT: vslideup.vi v10, v8, 1 |
| ; CHECK-NEXT: vsetivli zero, 4, e16, mf2, ta, ma |
| ; CHECK-NEXT: vse16.v v10, (a5) |
| ; CHECK-NEXT: vslideup.vi v8, v9, 3 |
| ; CHECK-NEXT: vse16.v v8, (a6) |
| ; CHECK-NEXT: ret |
| store <4 x i16> <i16 3, i16 3, i16 4, i16 4>, ptr %z0 |
| store <4 x i16> <i16 3, i16 3, i16 4, i16 undef>, ptr %z1 |
| store <4 x i16> <i16 3, i16 undef, i16 4, i16 4>, ptr %z2 |
| store <4 x i16> <i16 undef, i16 3, i16 undef, i16 4>, ptr %z3 |
| store <4 x i16> <i16 3, i16 undef, i16 4, i16 undef>, ptr %z4 |
| ; We don't catch this one |
| store <4 x i16> <i16 undef, i16 3, i16 4, i16 4>, ptr %z5 |
| ; We catch this one but as VID/3 rather than VID/2 |
| store <4 x i16> <i16 3, i16 3, i16 undef, i16 4>, ptr %z6 |
| ret void |
| } |
| |
| define void @buildvec_vid_stepn1o4_addn5_v8i8(ptr %z0) { |
| ; CHECK-LABEL: buildvec_vid_stepn1o4_addn5_v8i8: |
| ; CHECK: # %bb.0: |
| ; CHECK-NEXT: vsetivli zero, 8, e8, mf2, ta, ma |
| ; CHECK-NEXT: vid.v v8 |
| ; CHECK-NEXT: vsrl.vi v8, v8, 2 |
| ; CHECK-NEXT: vrsub.vi v8, v8, -5 |
| ; CHECK-NEXT: vse8.v v8, (a0) |
| ; CHECK-NEXT: ret |
| store <8 x i8> <i8 -5, i8 -5, i8 -5, i8 -5, i8 -6, i8 -6, i8 -6, i8 -6>, ptr %z0 |
| ret void |
| } |
| |
| define void @buildvec_vid_mpy_imm_v8i16(ptr %x) { |
| ; CHECK-LABEL: buildvec_vid_mpy_imm_v8i16: |
| ; CHECK: # %bb.0: |
| ; CHECK-NEXT: vsetivli zero, 8, e16, m1, ta, ma |
| ; CHECK-NEXT: vid.v v8 |
| ; CHECK-NEXT: li a1, 17 |
| ; CHECK-NEXT: vmul.vx v8, v8, a1 |
| ; CHECK-NEXT: vse16.v v8, (a0) |
| ; CHECK-NEXT: ret |
| store <8 x i16> <i16 0, i16 17, i16 34, i16 51, i16 68, i16 85, i16 102, i16 119>, ptr %x |
| ret void |
| } |
| |
| define void @buildvec_vid_shl_imm_v8i16(ptr %x) { |
| ; CHECK-LABEL: buildvec_vid_shl_imm_v8i16: |
| ; CHECK: # %bb.0: |
| ; CHECK-NEXT: vsetivli zero, 8, e16, m1, ta, ma |
| ; CHECK-NEXT: vid.v v8 |
| ; CHECK-NEXT: vsll.vi v8, v8, 9 |
| ; CHECK-NEXT: vse16.v v8, (a0) |
| ; CHECK-NEXT: ret |
| store <8 x i16> <i16 0, i16 512, i16 1024, i16 1536, i16 2048, i16 2560, i16 3072, i16 3584>, ptr %x |
| ret void |
| } |
| |
| define <4 x i32> @splat_c3_v4i32(<4 x i32> %v) { |
| ; CHECK-LABEL: splat_c3_v4i32: |
| ; CHECK: # %bb.0: |
| ; CHECK-NEXT: vsetivli zero, 4, e32, m1, ta, ma |
| ; CHECK-NEXT: vrgather.vi v9, v8, 3 |
| ; CHECK-NEXT: vmv.v.v v8, v9 |
| ; CHECK-NEXT: ret |
| %x = extractelement <4 x i32> %v, i32 3 |
| %ins = insertelement <4 x i32> poison, i32 %x, i32 0 |
| %splat = shufflevector <4 x i32> %ins, <4 x i32> poison, <4 x i32> zeroinitializer |
| ret <4 x i32> %splat |
| } |
| |
| define <4 x i32> @splat_idx_v4i32(<4 x i32> %v, i64 %idx) { |
| ; CHECK-LABEL: splat_idx_v4i32: |
| ; CHECK: # %bb.0: |
| ; CHECK-NEXT: vsetivli zero, 4, e32, m1, ta, ma |
| ; CHECK-NEXT: vrgather.vx v9, v8, a0 |
| ; CHECK-NEXT: vmv.v.v v8, v9 |
| ; CHECK-NEXT: ret |
| %x = extractelement <4 x i32> %v, i64 %idx |
| %ins = insertelement <4 x i32> poison, i32 %x, i32 0 |
| %splat = shufflevector <4 x i32> %ins, <4 x i32> poison, <4 x i32> zeroinitializer |
| ret <4 x i32> %splat |
| } |
| |
| define <8 x i16> @splat_c4_v8i16(<8 x i16> %v) { |
| ; CHECK-LABEL: splat_c4_v8i16: |
| ; CHECK: # %bb.0: |
| ; CHECK-NEXT: vsetivli zero, 8, e16, m1, ta, ma |
| ; CHECK-NEXT: vrgather.vi v9, v8, 4 |
| ; CHECK-NEXT: vmv.v.v v8, v9 |
| ; CHECK-NEXT: ret |
| %x = extractelement <8 x i16> %v, i32 4 |
| %ins = insertelement <8 x i16> poison, i16 %x, i32 0 |
| %splat = shufflevector <8 x i16> %ins, <8 x i16> poison, <8 x i32> zeroinitializer |
| ret <8 x i16> %splat |
| } |
| |
| define <8 x i16> @splat_idx_v8i16(<8 x i16> %v, i64 %idx) { |
| ; CHECK-LABEL: splat_idx_v8i16: |
| ; CHECK: # %bb.0: |
| ; CHECK-NEXT: vsetivli zero, 8, e16, m1, ta, ma |
| ; CHECK-NEXT: vrgather.vx v9, v8, a0 |
| ; CHECK-NEXT: vmv.v.v v8, v9 |
| ; CHECK-NEXT: ret |
| %x = extractelement <8 x i16> %v, i64 %idx |
| %ins = insertelement <8 x i16> poison, i16 %x, i32 0 |
| %splat = shufflevector <8 x i16> %ins, <8 x i16> poison, <8 x i32> zeroinitializer |
| ret <8 x i16> %splat |
| } |
| |
| define <4 x i8> @buildvec_not_vid_v4i8_1() { |
| ; CHECK-LABEL: buildvec_not_vid_v4i8_1: |
| ; CHECK: # %bb.0: |
| ; CHECK-NEXT: lui a0, %hi(.LCPI37_0) |
| ; CHECK-NEXT: addi a0, a0, %lo(.LCPI37_0) |
| ; CHECK-NEXT: vsetivli zero, 4, e8, mf4, ta, ma |
| ; CHECK-NEXT: vle8.v v8, (a0) |
| ; CHECK-NEXT: ret |
| ret <4 x i8> <i8 0, i8 0, i8 2, i8 3> |
| } |
| |
| define <4 x i8> @buildvec_not_vid_v4i8_2() { |
| ; CHECK-LABEL: buildvec_not_vid_v4i8_2: |
| ; CHECK: # %bb.0: |
| ; CHECK-NEXT: lui a0, %hi(.LCPI38_0) |
| ; CHECK-NEXT: addi a0, a0, %lo(.LCPI38_0) |
| ; CHECK-NEXT: vsetivli zero, 4, e8, mf4, ta, ma |
| ; CHECK-NEXT: vle8.v v8, (a0) |
| ; CHECK-NEXT: ret |
| ret <4 x i8> <i8 3, i8 3, i8 1, i8 0> |
| } |
| |
| ; We match this as a VID sequence (-3 / 8) + 5 but choose not to introduce |
| ; division to compute it. |
| define <16 x i8> @buildvec_not_vid_v16i8() { |
| ; CHECK-LABEL: buildvec_not_vid_v16i8: |
| ; CHECK: # %bb.0: |
| ; CHECK-NEXT: vsetivli zero, 16, e8, m1, ta, ma |
| ; CHECK-NEXT: vmv.v.i v9, 3 |
| ; CHECK-NEXT: vmv.v.i v8, 0 |
| ; CHECK-NEXT: vsetivli zero, 7, e8, m1, tu, ma |
| ; CHECK-NEXT: vslideup.vi v8, v9, 6 |
| ; CHECK-NEXT: ret |
| ret <16 x i8> <i8 undef, i8 undef, i8 undef, i8 undef, i8 undef, i8 undef, i8 3, i8 undef, i8 undef, i8 undef, i8 undef, i8 undef, i8 undef, i8 undef, i8 0, i8 0> |
| } |