derek/TheRock
1
0 Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity
Files
1a182e2dac58879f215c0fad0f5f78c76dcafe56
TheRock/compiler/amd-llvm/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-trunc-vp.ll
T
derek.holloway 7502bf64a4
CodeQL / Analyze (actions) (push) Canceled after 0s
Details
CodeQL / Analyze (python) (push) Canceled after 0s
Details
pre-commit / pre-commit (push) Canceled after 0s
Details
Multi-Arch CI / setup (push) Canceled after 0s
Details
Publish build_manylinux_x86_64 images / publish_build_manylinux_x86_64 (push) Canceled after 0s
Details
Setup (Multi-Arch) / setup (push) Canceled after 0s
Details
Multi-Arch CI / Linux::${{ fromJSON(needs.setup.outputs.linux_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Details
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Details
Publish no_rocm_image_ubi10 image / publish_no_rocm_image_ubi10 (push) Canceled after 0s
Details
Multi-Arch CI / Windows::${{ fromJSON(needs.setup.outputs.windows_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Details
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Details
Multi-Arch CI / Manifest Diff (push) Canceled after 0s
Details
TheRock Manifest Diff Report / Generate Manifest Diff Report (push) Canceled after 0s
Details
Multi-Arch CI / CI Summary (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04 (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04_ocl_rt images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Details
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Details
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04_rocgdb images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Details
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Details
Unit Tests / Unit Tests :: ubuntu-24.04 (push) Canceled after 0s
Details
Unit Tests / Unit Tests :: windows-2022 (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_media (push) Canceled after 0s
Details
Unit Tests / Unit Tests Summary (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_openmpi (push) Canceled after 0s
Details
Flatten subprojects into a single base monorepo
2026-09-02 16:34:47 -07:00

323 lines
12 KiB
LLVM
Raw Blame History

; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -mtriple=riscv32 -mattr=+v,+m -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,RV32
; RUN: llc -mtriple=riscv64 -mattr=+v,+m -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,RV64
define <2 x i7> @vtrunc_v2i7_v2i16(<2 x i16> %a, <2 x i1> %m, i32 zeroext %vl) {
; CHECK-LABEL: vtrunc_v2i7_v2i16:
; CHECK: # %bb.0:
; CHECK-NEXT: vsetivli zero, 2, e8, mf8, ta, ma
; CHECK-NEXT: vnsrl.wi v8, v8, 0
; CHECK-NEXT: ret
%v = call <2 x i7> @llvm.vp.trunc.v2i7.v2i16(<2 x i16> %a, <2 x i1> %m, i32 %vl)
ret <2 x i7> %v
}
define <2 x i8> @vtrunc_v2i8_v2i15(<2 x i15> %a, <2 x i1> %m, i32 zeroext %vl) {
; CHECK-LABEL: vtrunc_v2i8_v2i15:
; CHECK: # %bb.0:
; CHECK-NEXT: vsetivli zero, 2, e8, mf8, ta, ma
; CHECK-NEXT: vnsrl.wi v8, v8, 0
; CHECK-NEXT: ret
%v = call <2 x i8> @llvm.vp.trunc.v2i8.v2i15(<2 x i15> %a, <2 x i1> %m, i32 %vl)
ret <2 x i8> %v
}
define <2 x i8> @vtrunc_v2i8_v2i16(<2 x i16> %a, <2 x i1> %m, i32 zeroext %vl) {
; CHECK-LABEL: vtrunc_v2i8_v2i16:
; CHECK: # %bb.0:
; CHECK-NEXT: vsetivli zero, 2, e8, mf8, ta, ma
; CHECK-NEXT: vnsrl.wi v8, v8, 0
; CHECK-NEXT: ret
%v = call <2 x i8> @llvm.vp.trunc.v2i8.v2i16(<2 x i16> %a, <2 x i1> %m, i32 %vl)
ret <2 x i8> %v
}
define <2 x i8> @vtrunc_v2i8_v2i16_unmasked(<2 x i16> %a, i32 zeroext %vl) {
; CHECK-LABEL: vtrunc_v2i8_v2i16_unmasked:
; CHECK: # %bb.0:
; CHECK-NEXT: vsetivli zero, 2, e8, mf8, ta, ma
; CHECK-NEXT: vnsrl.wi v8, v8, 0
; CHECK-NEXT: ret
%v = call <2 x i8> @llvm.vp.trunc.v2i8.v2i16(<2 x i16> %a, <2 x i1> <i1 true, i1 true>, i32 %vl)
ret <2 x i8> %v
}
define <128 x i7> @vtrunc_v128i7_v128i16(<128 x i16> %a, <128 x i1> %m, i32 zeroext %vl) {
; CHECK-LABEL: vtrunc_v128i7_v128i16:
; CHECK: # %bb.0:
; CHECK-NEXT: li a0, 64
; CHECK-NEXT: vsetvli zero, a0, e8, m4, ta, ma
; CHECK-NEXT: vnsrl.wi v24, v16, 0
; CHECK-NEXT: vnsrl.wi v16, v8, 0
; CHECK-NEXT: li a1, 128
; CHECK-NEXT: vsetvli zero, a1, e8, m8, ta, ma
; CHECK-NEXT: vslideup.vx v16, v24, a0
; CHECK-NEXT: vmv.v.v v8, v16
; CHECK-NEXT: ret
%v = call <128 x i7> @llvm.vp.trunc.v128i7.v128i16(<128 x i16> %a, <128 x i1> %m, i32 %vl)
ret <128 x i7> %v
}
define <2 x i8> @vtrunc_v2i8_v2i32(<2 x i32> %a, <2 x i1> %m, i32 zeroext %vl) {
; CHECK-LABEL: vtrunc_v2i8_v2i32:
; CHECK: # %bb.0:
; CHECK-NEXT: vsetivli zero, 2, e16, mf4, ta, ma
; CHECK-NEXT: vnsrl.wi v8, v8, 0
; CHECK-NEXT: vsetvli zero, zero, e8, mf8, ta, ma
; CHECK-NEXT: vnsrl.wi v8, v8, 0
; CHECK-NEXT: ret
%v = call <2 x i8> @llvm.vp.trunc.v2i8.v2i32(<2 x i32> %a, <2 x i1> %m, i32 %vl)
ret <2 x i8> %v
}
define <2 x i8> @vtrunc_v2i8_v2i32_unmasked(<2 x i32> %a, i32 zeroext %vl) {
; CHECK-LABEL: vtrunc_v2i8_v2i32_unmasked:
; CHECK: # %bb.0:
; CHECK-NEXT: vsetivli zero, 2, e16, mf4, ta, ma
; CHECK-NEXT: vnsrl.wi v8, v8, 0
; CHECK-NEXT: vsetvli zero, zero, e8, mf8, ta, ma
; CHECK-NEXT: vnsrl.wi v8, v8, 0
; CHECK-NEXT: ret
%v = call <2 x i8> @llvm.vp.trunc.v2i8.v2i32(<2 x i32> %a, <2 x i1> <i1 true, i1 true>, i32 %vl)
ret <2 x i8> %v
}
define <2 x i8> @vtrunc_v2i8_v2i64(<2 x i64> %a, <2 x i1> %m, i32 zeroext %vl) {
; CHECK-LABEL: vtrunc_v2i8_v2i64:
; CHECK: # %bb.0:
; CHECK-NEXT: vsetivli zero, 2, e32, mf2, ta, ma
; CHECK-NEXT: vnsrl.wi v8, v8, 0
; CHECK-NEXT: vsetvli zero, zero, e16, mf4, ta, ma
; CHECK-NEXT: vnsrl.wi v8, v8, 0
; CHECK-NEXT: vsetvli zero, zero, e8, mf8, ta, ma
; CHECK-NEXT: vnsrl.wi v8, v8, 0
; CHECK-NEXT: ret
%v = call <2 x i8> @llvm.vp.trunc.v2i8.v2i64(<2 x i64> %a, <2 x i1> %m, i32 %vl)
ret <2 x i8> %v
}
define <2 x i8> @vtrunc_v2i8_v2i64_unmasked(<2 x i64> %a, i32 zeroext %vl) {
; CHECK-LABEL: vtrunc_v2i8_v2i64_unmasked:
; CHECK: # %bb.0:
; CHECK-NEXT: vsetivli zero, 2, e32, mf2, ta, ma
; CHECK-NEXT: vnsrl.wi v8, v8, 0
; CHECK-NEXT: vsetvli zero, zero, e16, mf4, ta, ma
; CHECK-NEXT: vnsrl.wi v8, v8, 0
; CHECK-NEXT: vsetvli zero, zero, e8, mf8, ta, ma
; CHECK-NEXT: vnsrl.wi v8, v8, 0
; CHECK-NEXT: ret
%v = call <2 x i8> @llvm.vp.trunc.v2i8.v2i64(<2 x i64> %a, <2 x i1> <i1 true, i1 true>, i32 %vl)
ret <2 x i8> %v
}
define <2 x i16> @vtrunc_v2i16_v2i32(<2 x i32> %a, <2 x i1> %m, i32 zeroext %vl) {
; CHECK-LABEL: vtrunc_v2i16_v2i32:
; CHECK: # %bb.0:
; CHECK-NEXT: vsetivli zero, 2, e16, mf4, ta, ma
; CHECK-NEXT: vnsrl.wi v8, v8, 0
; CHECK-NEXT: ret
%v = call <2 x i16> @llvm.vp.trunc.v2i16.v2i32(<2 x i32> %a, <2 x i1> %m, i32 %vl)
ret <2 x i16> %v
}
define <2 x i16> @vtrunc_v2i16_v2i32_unmasked(<2 x i32> %a, i32 zeroext %vl) {
; CHECK-LABEL: vtrunc_v2i16_v2i32_unmasked:
; CHECK: # %bb.0:
; CHECK-NEXT: vsetivli zero, 2, e16, mf4, ta, ma
; CHECK-NEXT: vnsrl.wi v8, v8, 0
; CHECK-NEXT: ret
%v = call <2 x i16> @llvm.vp.trunc.v2i16.v2i32(<2 x i32> %a, <2 x i1> <i1 true, i1 true>, i32 %vl)
ret <2 x i16> %v
}
define <2 x i16> @vtrunc_v2i16_v2i64(<2 x i64> %a, <2 x i1> %m, i32 zeroext %vl) {
; CHECK-LABEL: vtrunc_v2i16_v2i64:
; CHECK: # %bb.0:
; CHECK-NEXT: vsetivli zero, 2, e32, mf2, ta, ma
; CHECK-NEXT: vnsrl.wi v8, v8, 0
; CHECK-NEXT: vsetvli zero, zero, e16, mf4, ta, ma
; CHECK-NEXT: vnsrl.wi v8, v8, 0
; CHECK-NEXT: ret
%v = call <2 x i16> @llvm.vp.trunc.v2i16.v2i64(<2 x i64> %a, <2 x i1> %m, i32 %vl)
ret <2 x i16> %v
}
define <2 x i16> @vtrunc_v2i16_v2i64_unmasked(<2 x i64> %a, i32 zeroext %vl) {
; CHECK-LABEL: vtrunc_v2i16_v2i64_unmasked:
; CHECK: # %bb.0:
; CHECK-NEXT: vsetivli zero, 2, e32, mf2, ta, ma
; CHECK-NEXT: vnsrl.wi v8, v8, 0
; CHECK-NEXT: vsetvli zero, zero, e16, mf4, ta, ma
; CHECK-NEXT: vnsrl.wi v8, v8, 0
; CHECK-NEXT: ret
%v = call <2 x i16> @llvm.vp.trunc.v2i16.v2i64(<2 x i64> %a, <2 x i1> <i1 true, i1 true>, i32 %vl)
ret <2 x i16> %v
}
define <15 x i16> @vtrunc_v15i16_v15i64(<15 x i64> %a, <15 x i1> %m, i32 zeroext %vl) {
; CHECK-LABEL: vtrunc_v15i16_v15i64:
; CHECK: # %bb.0:
; CHECK-NEXT: vsetivli zero, 16, e32, m4, ta, ma
; CHECK-NEXT: vnsrl.wi v16, v8, 0
; CHECK-NEXT: vsetvli zero, zero, e16, m2, ta, ma
; CHECK-NEXT: vnsrl.wi v8, v16, 0
; CHECK-NEXT: ret
%v = call <15 x i16> @llvm.vp.trunc.v15i16.v15i64(<15 x i64> %a, <15 x i1> %m, i32 %vl)
ret <15 x i16> %v
}
define <2 x i32> @vtrunc_v2i32_v2i64(<2 x i64> %a, <2 x i1> %m, i32 zeroext %vl) {
; CHECK-LABEL: vtrunc_v2i32_v2i64:
; CHECK: # %bb.0:
; CHECK-NEXT: vsetivli zero, 2, e32, mf2, ta, ma
; CHECK-NEXT: vnsrl.wi v8, v8, 0
; CHECK-NEXT: ret
%v = call <2 x i32> @llvm.vp.trunc.v2i32.v2i64(<2 x i64> %a, <2 x i1> %m, i32 %vl)
ret <2 x i32> %v
}
define <2 x i32> @vtrunc_v2i32_v2i64_unmasked(<2 x i64> %a, i32 zeroext %vl) {
; CHECK-LABEL: vtrunc_v2i32_v2i64_unmasked:
; CHECK: # %bb.0:
; CHECK-NEXT: vsetivli zero, 2, e32, mf2, ta, ma
; CHECK-NEXT: vnsrl.wi v8, v8, 0
; CHECK-NEXT: ret
%v = call <2 x i32> @llvm.vp.trunc.v2i32.v2i64(<2 x i64> %a, <2 x i1> <i1 true, i1 true>, i32 %vl)
ret <2 x i32> %v
}
define <128 x i32> @vtrunc_v128i32_v128i64(<128 x i64> %a, <128 x i1> %m, i32 zeroext %vl) {
; CHECK-LABEL: vtrunc_v128i32_v128i64:
; CHECK: # %bb.0:
; CHECK-NEXT: addi sp, sp, -16
; CHECK-NEXT: .cfi_def_cfa_offset 16
; CHECK-NEXT: csrr a2, vlenb
; CHECK-NEXT: li a3, 40
; CHECK-NEXT: mul a2, a2, a3
; CHECK-NEXT: sub sp, sp, a2
; CHECK-NEXT: .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0x28, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 40 * vlenb
; CHECK-NEXT: addi a2, a1, 512
; CHECK-NEXT: addi a3, a1, 640
; CHECK-NEXT: addi a4, a1, 256
; CHECK-NEXT: vsetivli zero, 16, e32, m4, ta, ma
; CHECK-NEXT: vle64.v v24, (a2)
; CHECK-NEXT: csrr a2, vlenb
; CHECK-NEXT: slli a2, a2, 5
; CHECK-NEXT: add a2, sp, a2
; CHECK-NEXT: addi a2, a2, 16
; CHECK-NEXT: vs8r.v v24, (a2) # vscale x 64-byte Folded Spill
; CHECK-NEXT: addi a5, a1, 384
; CHECK-NEXT: vle64.v v24, (a3)
; CHECK-NEXT: csrr a2, vlenb
; CHECK-NEXT: li a3, 24
; CHECK-NEXT: mul a2, a2, a3
; CHECK-NEXT: add a2, sp, a2
; CHECK-NEXT: addi a2, a2, 16
; CHECK-NEXT: vs8r.v v24, (a2) # vscale x 64-byte Folded Spill
; CHECK-NEXT: li a2, 32
; CHECK-NEXT: vnsrl.wi v24, v16, 0
; CHECK-NEXT: vnsrl.wi v16, v8, 0
; CHECK-NEXT: vsetvli zero, a2, e32, m8, ta, ma
; CHECK-NEXT: vslideup.vi v16, v24, 16
; CHECK-NEXT: vsetivli zero, 16, e64, m8, ta, ma
; CHECK-NEXT: vle64.v v0, (a1)
; CHECK-NEXT: addi a1, a1, 128
; CHECK-NEXT: vle64.v v8, (a4)
; CHECK-NEXT: csrr a3, vlenb
; CHECK-NEXT: slli a3, a3, 3
; CHECK-NEXT: add a3, sp, a3
; CHECK-NEXT: addi a3, a3, 16
; CHECK-NEXT: vs8r.v v8, (a3) # vscale x 64-byte Folded Spill
; CHECK-NEXT: vle64.v v24, (a1)
; CHECK-NEXT: vle64.v v8, (a5)
; CHECK-NEXT: addi a1, sp, 16
; CHECK-NEXT: vs8r.v v8, (a1) # vscale x 64-byte Folded Spill
; CHECK-NEXT: vsetvli zero, a2, e32, m8, ta, ma
; CHECK-NEXT: vse32.v v16, (a0)
; CHECK-NEXT: vsetivli zero, 16, e32, m4, ta, ma
; CHECK-NEXT: vnsrl.wi v8, v0, 0
; CHECK-NEXT: vnsrl.wi v16, v24, 0
; CHECK-NEXT: vsetvli zero, a2, e32, m8, ta, ma
; CHECK-NEXT: vslideup.vi v8, v16, 16
; CHECK-NEXT: csrr a1, vlenb
; CHECK-NEXT: slli a1, a1, 4
; CHECK-NEXT: add a1, sp, a1
; CHECK-NEXT: addi a1, a1, 16
; CHECK-NEXT: vs8r.v v8, (a1) # vscale x 64-byte Folded Spill
; CHECK-NEXT: addi a1, sp, 16
; CHECK-NEXT: vl8r.v v8, (a1) # vscale x 64-byte Folded Reload
; CHECK-NEXT: vsetivli zero, 16, e32, m4, ta, ma
; CHECK-NEXT: vnsrl.wi v24, v8, 0
; CHECK-NEXT: csrr a1, vlenb
; CHECK-NEXT: slli a1, a1, 3
; CHECK-NEXT: add a1, sp, a1
; CHECK-NEXT: addi a1, a1, 16
; CHECK-NEXT: vl8r.v v8, (a1) # vscale x 64-byte Folded Reload
; CHECK-NEXT: vnsrl.wi v16, v8, 0
; CHECK-NEXT: vsetvli zero, a2, e32, m8, ta, ma
; CHECK-NEXT: vslideup.vi v16, v24, 16
; CHECK-NEXT: csrr a1, vlenb
; CHECK-NEXT: li a3, 24
; CHECK-NEXT: mul a1, a1, a3
; CHECK-NEXT: add a1, sp, a1
; CHECK-NEXT: addi a1, a1, 16
; CHECK-NEXT: vl8r.v v8, (a1) # vscale x 64-byte Folded Reload
; CHECK-NEXT: vsetivli zero, 16, e32, m4, ta, ma
; CHECK-NEXT: vnsrl.wi v24, v8, 0
; CHECK-NEXT: csrr a1, vlenb
; CHECK-NEXT: slli a1, a1, 5
; CHECK-NEXT: add a1, sp, a1
; CHECK-NEXT: addi a1, a1, 16
; CHECK-NEXT: vl8r.v v8, (a1) # vscale x 64-byte Folded Reload
; CHECK-NEXT: vnsrl.wi v0, v8, 0
; CHECK-NEXT: vsetvli zero, a2, e32, m8, ta, ma
; CHECK-NEXT: vslideup.vi v0, v24, 16
; CHECK-NEXT: addi a1, a0, 384
; CHECK-NEXT: vse32.v v0, (a1)
; CHECK-NEXT: addi a1, a0, 256
; CHECK-NEXT: vse32.v v16, (a1)
; CHECK-NEXT: addi a0, a0, 128
; CHECK-NEXT: csrr a1, vlenb
; CHECK-NEXT: slli a1, a1, 4
; CHECK-NEXT: add a1, sp, a1
; CHECK-NEXT: addi a1, a1, 16
; CHECK-NEXT: vl8r.v v8, (a1) # vscale x 64-byte Folded Reload
; CHECK-NEXT: vse32.v v8, (a0)
; CHECK-NEXT: csrr a0, vlenb
; CHECK-NEXT: li a1, 40
; CHECK-NEXT: mul a0, a0, a1
; CHECK-NEXT: add sp, sp, a0
; CHECK-NEXT: .cfi_def_cfa sp, 16
; CHECK-NEXT: addi sp, sp, 16
; CHECK-NEXT: .cfi_def_cfa_offset 0
; CHECK-NEXT: ret
%v = call <128 x i32> @llvm.vp.trunc.v128i32.v128i64(<128 x i64> %a, <128 x i1> %m, i32 %vl)
ret <128 x i32> %v
}
define <32 x i32> @vtrunc_v32i32_v32i64(<32 x i64> %a, <32 x i1> %m, i32 zeroext %vl) {
; CHECK-LABEL: vtrunc_v32i32_v32i64:
; CHECK: # %bb.0:
; CHECK-NEXT: vsetivli zero, 16, e32, m4, ta, ma
; CHECK-NEXT: vnsrl.wi v24, v16, 0
; CHECK-NEXT: vnsrl.wi v16, v8, 0
; CHECK-NEXT: li a0, 32
; CHECK-NEXT: vsetvli zero, a0, e32, m8, ta, ma
; CHECK-NEXT: vslideup.vi v16, v24, 16
; CHECK-NEXT: vmv.v.v v8, v16
; CHECK-NEXT: ret
%v = call <32 x i32> @llvm.vp.trunc.v32i32.v32i64(<32 x i64> %a, <32 x i1> %m, i32 %vl)
ret <32 x i32> %v
}
define <2 x i7> @vtrunc_v2i7_v2i8(<2 x i8> %a, <2 x i1> %m, i32 zeroext %vl) {
; CHECK-LABEL: vtrunc_v2i7_v2i8:
; CHECK: # %bb.0:
; CHECK-NEXT: ret
%v = call <2 x i7> @llvm.vp.trunc.v2i7.v2i8(<2 x i8> %a, <2 x i1> %m, i32 %vl)
ret <2 x i7> %v
}
;; NOTE: These prefixes are unused and the list is autogenerated. Do not add tests below this line:
; RV32: {{.*}}
; RV64: {{.*}}
Reference in new issue View Git Blame Copy Permalink