--- /dev/null
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 2
+; RUN: llc -mtriple=riscv32 -mattr=+v -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,CHECK-RV32
+; RUN: llc -mtriple=riscv64 -mattr=+v -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,CHECK-RV64
+; RUN: llc -mtriple=riscv32 -mattr=+v,+experimental-zvbb -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK-ZVBB,CHECK-ZVBB32
+; RUN: llc -mtriple=riscv64 -mattr=+v,+experimental-zvbb -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK-ZVBB,CHECK-ZVBB64
+
+declare <vscale x 1 x i8> @llvm.fshl.nxv1i8(<vscale x 1 x i8>, <vscale x 1 x i8>, <vscale x 1 x i8>)
+
+define <vscale x 1 x i8> @vrol_vv_nxv1i8(<vscale x 1 x i8> %a, <vscale x 1 x i8> %b) {
+; CHECK-LABEL: vrol_vv_nxv1i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e8, mf8, ta, ma
+; CHECK-NEXT: vand.vi v10, v9, 7
+; CHECK-NEXT: vsll.vv v10, v8, v10
+; CHECK-NEXT: vrsub.vi v9, v9, 0
+; CHECK-NEXT: vand.vi v9, v9, 7
+; CHECK-NEXT: vsrl.vv v8, v8, v9
+; CHECK-NEXT: vor.vv v8, v10, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vv_nxv1i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e8, mf8, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vv v8, v8, v9
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 1 x i8> @llvm.fshl.nxv1i8(<vscale x 1 x i8> %a, <vscale x 1 x i8> %a, <vscale x 1 x i8> %b)
+ ret <vscale x 1 x i8> %x
+}
+
+define <vscale x 1 x i8> @vrol_vx_nxv1i8(<vscale x 1 x i8> %a, i8 %b) {
+; CHECK-LABEL: vrol_vx_nxv1i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a1, zero, e8, mf8, ta, ma
+; CHECK-NEXT: vmv.v.x v9, a0
+; CHECK-NEXT: vand.vi v10, v9, 7
+; CHECK-NEXT: vsll.vv v10, v8, v10
+; CHECK-NEXT: vrsub.vi v9, v9, 0
+; CHECK-NEXT: vand.vi v9, v9, 7
+; CHECK-NEXT: vsrl.vv v8, v8, v9
+; CHECK-NEXT: vor.vv v8, v10, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vx_nxv1i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e8, mf8, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 1 x i8> poison, i8 %b, i32 0
+ %b.splat = shufflevector <vscale x 1 x i8> %b.head, <vscale x 1 x i8> poison, <vscale x 1 x i32> zeroinitializer
+ %x = call <vscale x 1 x i8> @llvm.fshl.nxv1i8(<vscale x 1 x i8> %a, <vscale x 1 x i8> %a, <vscale x 1 x i8> %b.splat)
+ ret <vscale x 1 x i8> %x
+}
+
+declare <vscale x 2 x i8> @llvm.fshl.nxv2i8(<vscale x 2 x i8>, <vscale x 2 x i8>, <vscale x 2 x i8>)
+
+define <vscale x 2 x i8> @vrol_vv_nxv2i8(<vscale x 2 x i8> %a, <vscale x 2 x i8> %b) {
+; CHECK-LABEL: vrol_vv_nxv2i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e8, mf4, ta, ma
+; CHECK-NEXT: vand.vi v10, v9, 7
+; CHECK-NEXT: vsll.vv v10, v8, v10
+; CHECK-NEXT: vrsub.vi v9, v9, 0
+; CHECK-NEXT: vand.vi v9, v9, 7
+; CHECK-NEXT: vsrl.vv v8, v8, v9
+; CHECK-NEXT: vor.vv v8, v10, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vv_nxv2i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e8, mf4, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vv v8, v8, v9
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 2 x i8> @llvm.fshl.nxv2i8(<vscale x 2 x i8> %a, <vscale x 2 x i8> %a, <vscale x 2 x i8> %b)
+ ret <vscale x 2 x i8> %x
+}
+
+define <vscale x 2 x i8> @vrol_vx_nxv2i8(<vscale x 2 x i8> %a, i8 %b) {
+; CHECK-LABEL: vrol_vx_nxv2i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a1, zero, e8, mf4, ta, ma
+; CHECK-NEXT: vmv.v.x v9, a0
+; CHECK-NEXT: vand.vi v10, v9, 7
+; CHECK-NEXT: vsll.vv v10, v8, v10
+; CHECK-NEXT: vrsub.vi v9, v9, 0
+; CHECK-NEXT: vand.vi v9, v9, 7
+; CHECK-NEXT: vsrl.vv v8, v8, v9
+; CHECK-NEXT: vor.vv v8, v10, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vx_nxv2i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e8, mf4, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 2 x i8> poison, i8 %b, i32 0
+ %b.splat = shufflevector <vscale x 2 x i8> %b.head, <vscale x 2 x i8> poison, <vscale x 2 x i32> zeroinitializer
+ %x = call <vscale x 2 x i8> @llvm.fshl.nxv2i8(<vscale x 2 x i8> %a, <vscale x 2 x i8> %a, <vscale x 2 x i8> %b.splat)
+ ret <vscale x 2 x i8> %x
+}
+
+declare <vscale x 4 x i8> @llvm.fshl.nxv4i8(<vscale x 4 x i8>, <vscale x 4 x i8>, <vscale x 4 x i8>)
+
+define <vscale x 4 x i8> @vrol_vv_nxv4i8(<vscale x 4 x i8> %a, <vscale x 4 x i8> %b) {
+; CHECK-LABEL: vrol_vv_nxv4i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e8, mf2, ta, ma
+; CHECK-NEXT: vand.vi v10, v9, 7
+; CHECK-NEXT: vsll.vv v10, v8, v10
+; CHECK-NEXT: vrsub.vi v9, v9, 0
+; CHECK-NEXT: vand.vi v9, v9, 7
+; CHECK-NEXT: vsrl.vv v8, v8, v9
+; CHECK-NEXT: vor.vv v8, v10, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vv_nxv4i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e8, mf2, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vv v8, v8, v9
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 4 x i8> @llvm.fshl.nxv4i8(<vscale x 4 x i8> %a, <vscale x 4 x i8> %a, <vscale x 4 x i8> %b)
+ ret <vscale x 4 x i8> %x
+}
+
+define <vscale x 4 x i8> @vrol_vx_nxv4i8(<vscale x 4 x i8> %a, i8 %b) {
+; CHECK-LABEL: vrol_vx_nxv4i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a1, zero, e8, mf2, ta, ma
+; CHECK-NEXT: vmv.v.x v9, a0
+; CHECK-NEXT: vand.vi v10, v9, 7
+; CHECK-NEXT: vsll.vv v10, v8, v10
+; CHECK-NEXT: vrsub.vi v9, v9, 0
+; CHECK-NEXT: vand.vi v9, v9, 7
+; CHECK-NEXT: vsrl.vv v8, v8, v9
+; CHECK-NEXT: vor.vv v8, v10, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vx_nxv4i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e8, mf2, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 4 x i8> poison, i8 %b, i32 0
+ %b.splat = shufflevector <vscale x 4 x i8> %b.head, <vscale x 4 x i8> poison, <vscale x 4 x i32> zeroinitializer
+ %x = call <vscale x 4 x i8> @llvm.fshl.nxv4i8(<vscale x 4 x i8> %a, <vscale x 4 x i8> %a, <vscale x 4 x i8> %b.splat)
+ ret <vscale x 4 x i8> %x
+}
+
+declare <vscale x 8 x i8> @llvm.fshl.nxv8i8(<vscale x 8 x i8>, <vscale x 8 x i8>, <vscale x 8 x i8>)
+
+define <vscale x 8 x i8> @vrol_vv_nxv8i8(<vscale x 8 x i8> %a, <vscale x 8 x i8> %b) {
+; CHECK-LABEL: vrol_vv_nxv8i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e8, m1, ta, ma
+; CHECK-NEXT: vand.vi v10, v9, 7
+; CHECK-NEXT: vsll.vv v10, v8, v10
+; CHECK-NEXT: vrsub.vi v9, v9, 0
+; CHECK-NEXT: vand.vi v9, v9, 7
+; CHECK-NEXT: vsrl.vv v8, v8, v9
+; CHECK-NEXT: vor.vv v8, v10, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vv_nxv8i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e8, m1, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vv v8, v8, v9
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 8 x i8> @llvm.fshl.nxv8i8(<vscale x 8 x i8> %a, <vscale x 8 x i8> %a, <vscale x 8 x i8> %b)
+ ret <vscale x 8 x i8> %x
+}
+
+define <vscale x 8 x i8> @vrol_vx_nxv8i8(<vscale x 8 x i8> %a, i8 %b) {
+; CHECK-LABEL: vrol_vx_nxv8i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a1, zero, e8, m1, ta, ma
+; CHECK-NEXT: vmv.v.x v9, a0
+; CHECK-NEXT: vand.vi v10, v9, 7
+; CHECK-NEXT: vsll.vv v10, v8, v10
+; CHECK-NEXT: vrsub.vi v9, v9, 0
+; CHECK-NEXT: vand.vi v9, v9, 7
+; CHECK-NEXT: vsrl.vv v8, v8, v9
+; CHECK-NEXT: vor.vv v8, v10, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vx_nxv8i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e8, m1, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 8 x i8> poison, i8 %b, i32 0
+ %b.splat = shufflevector <vscale x 8 x i8> %b.head, <vscale x 8 x i8> poison, <vscale x 8 x i32> zeroinitializer
+ %x = call <vscale x 8 x i8> @llvm.fshl.nxv8i8(<vscale x 8 x i8> %a, <vscale x 8 x i8> %a, <vscale x 8 x i8> %b.splat)
+ ret <vscale x 8 x i8> %x
+}
+
+declare <vscale x 16 x i8> @llvm.fshl.nxv16i8(<vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>)
+
+define <vscale x 16 x i8> @vrol_vv_nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %b) {
+; CHECK-LABEL: vrol_vv_nxv16i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e8, m2, ta, ma
+; CHECK-NEXT: vand.vi v12, v10, 7
+; CHECK-NEXT: vsll.vv v12, v8, v12
+; CHECK-NEXT: vrsub.vi v10, v10, 0
+; CHECK-NEXT: vand.vi v10, v10, 7
+; CHECK-NEXT: vsrl.vv v8, v8, v10
+; CHECK-NEXT: vor.vv v8, v12, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vv_nxv16i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e8, m2, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vv v8, v8, v10
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 16 x i8> @llvm.fshl.nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %a, <vscale x 16 x i8> %b)
+ ret <vscale x 16 x i8> %x
+}
+
+define <vscale x 16 x i8> @vrol_vx_nxv16i8(<vscale x 16 x i8> %a, i8 %b) {
+; CHECK-LABEL: vrol_vx_nxv16i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a1, zero, e8, m2, ta, ma
+; CHECK-NEXT: vmv.v.x v10, a0
+; CHECK-NEXT: vand.vi v12, v10, 7
+; CHECK-NEXT: vsll.vv v12, v8, v12
+; CHECK-NEXT: vrsub.vi v10, v10, 0
+; CHECK-NEXT: vand.vi v10, v10, 7
+; CHECK-NEXT: vsrl.vv v8, v8, v10
+; CHECK-NEXT: vor.vv v8, v12, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vx_nxv16i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e8, m2, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 16 x i8> poison, i8 %b, i32 0
+ %b.splat = shufflevector <vscale x 16 x i8> %b.head, <vscale x 16 x i8> poison, <vscale x 16 x i32> zeroinitializer
+ %x = call <vscale x 16 x i8> @llvm.fshl.nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %a, <vscale x 16 x i8> %b.splat)
+ ret <vscale x 16 x i8> %x
+}
+
+declare <vscale x 32 x i8> @llvm.fshl.nxv32i8(<vscale x 32 x i8>, <vscale x 32 x i8>, <vscale x 32 x i8>)
+
+define <vscale x 32 x i8> @vrol_vv_nxv32i8(<vscale x 32 x i8> %a, <vscale x 32 x i8> %b) {
+; CHECK-LABEL: vrol_vv_nxv32i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e8, m4, ta, ma
+; CHECK-NEXT: vand.vi v16, v12, 7
+; CHECK-NEXT: vsll.vv v16, v8, v16
+; CHECK-NEXT: vrsub.vi v12, v12, 0
+; CHECK-NEXT: vand.vi v12, v12, 7
+; CHECK-NEXT: vsrl.vv v8, v8, v12
+; CHECK-NEXT: vor.vv v8, v16, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vv_nxv32i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e8, m4, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vv v8, v8, v12
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 32 x i8> @llvm.fshl.nxv32i8(<vscale x 32 x i8> %a, <vscale x 32 x i8> %a, <vscale x 32 x i8> %b)
+ ret <vscale x 32 x i8> %x
+}
+
+define <vscale x 32 x i8> @vrol_vx_nxv32i8(<vscale x 32 x i8> %a, i8 %b) {
+; CHECK-LABEL: vrol_vx_nxv32i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a1, zero, e8, m4, ta, ma
+; CHECK-NEXT: vmv.v.x v12, a0
+; CHECK-NEXT: vand.vi v16, v12, 7
+; CHECK-NEXT: vsll.vv v16, v8, v16
+; CHECK-NEXT: vrsub.vi v12, v12, 0
+; CHECK-NEXT: vand.vi v12, v12, 7
+; CHECK-NEXT: vsrl.vv v8, v8, v12
+; CHECK-NEXT: vor.vv v8, v16, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vx_nxv32i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e8, m4, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 32 x i8> poison, i8 %b, i32 0
+ %b.splat = shufflevector <vscale x 32 x i8> %b.head, <vscale x 32 x i8> poison, <vscale x 32 x i32> zeroinitializer
+ %x = call <vscale x 32 x i8> @llvm.fshl.nxv32i8(<vscale x 32 x i8> %a, <vscale x 32 x i8> %a, <vscale x 32 x i8> %b.splat)
+ ret <vscale x 32 x i8> %x
+}
+
+declare <vscale x 64 x i8> @llvm.fshl.nxv64i8(<vscale x 64 x i8>, <vscale x 64 x i8>, <vscale x 64 x i8>)
+
+define <vscale x 64 x i8> @vrol_vv_nxv64i8(<vscale x 64 x i8> %a, <vscale x 64 x i8> %b) {
+; CHECK-LABEL: vrol_vv_nxv64i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e8, m8, ta, ma
+; CHECK-NEXT: vand.vi v24, v16, 7
+; CHECK-NEXT: vsll.vv v24, v8, v24
+; CHECK-NEXT: vrsub.vi v16, v16, 0
+; CHECK-NEXT: vand.vi v16, v16, 7
+; CHECK-NEXT: vsrl.vv v8, v8, v16
+; CHECK-NEXT: vor.vv v8, v24, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vv_nxv64i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e8, m8, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vv v8, v8, v16
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 64 x i8> @llvm.fshl.nxv64i8(<vscale x 64 x i8> %a, <vscale x 64 x i8> %a, <vscale x 64 x i8> %b)
+ ret <vscale x 64 x i8> %x
+}
+
+define <vscale x 64 x i8> @vrol_vx_nxv64i8(<vscale x 64 x i8> %a, i8 %b) {
+; CHECK-LABEL: vrol_vx_nxv64i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a1, zero, e8, m8, ta, ma
+; CHECK-NEXT: vmv.v.x v16, a0
+; CHECK-NEXT: vand.vi v24, v16, 7
+; CHECK-NEXT: vsll.vv v24, v8, v24
+; CHECK-NEXT: vrsub.vi v16, v16, 0
+; CHECK-NEXT: vand.vi v16, v16, 7
+; CHECK-NEXT: vsrl.vv v8, v8, v16
+; CHECK-NEXT: vor.vv v8, v24, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vx_nxv64i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e8, m8, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 64 x i8> poison, i8 %b, i32 0
+ %b.splat = shufflevector <vscale x 64 x i8> %b.head, <vscale x 64 x i8> poison, <vscale x 64 x i32> zeroinitializer
+ %x = call <vscale x 64 x i8> @llvm.fshl.nxv64i8(<vscale x 64 x i8> %a, <vscale x 64 x i8> %a, <vscale x 64 x i8> %b.splat)
+ ret <vscale x 64 x i8> %x
+}
+
+declare <vscale x 1 x i16> @llvm.fshl.nxv1i16(<vscale x 1 x i16>, <vscale x 1 x i16>, <vscale x 1 x i16>)
+
+define <vscale x 1 x i16> @vrol_vv_nxv1i16(<vscale x 1 x i16> %a, <vscale x 1 x i16> %b) {
+; CHECK-LABEL: vrol_vv_nxv1i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e16, mf4, ta, ma
+; CHECK-NEXT: vand.vi v10, v9, 15
+; CHECK-NEXT: vsll.vv v10, v8, v10
+; CHECK-NEXT: vrsub.vi v9, v9, 0
+; CHECK-NEXT: vand.vi v9, v9, 15
+; CHECK-NEXT: vsrl.vv v8, v8, v9
+; CHECK-NEXT: vor.vv v8, v10, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vv_nxv1i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e16, mf4, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vv v8, v8, v9
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 1 x i16> @llvm.fshl.nxv1i16(<vscale x 1 x i16> %a, <vscale x 1 x i16> %a, <vscale x 1 x i16> %b)
+ ret <vscale x 1 x i16> %x
+}
+
+define <vscale x 1 x i16> @vrol_vx_nxv1i16(<vscale x 1 x i16> %a, i16 %b) {
+; CHECK-LABEL: vrol_vx_nxv1i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a1, zero, e16, mf4, ta, ma
+; CHECK-NEXT: vmv.v.x v9, a0
+; CHECK-NEXT: vand.vi v10, v9, 15
+; CHECK-NEXT: vsll.vv v10, v8, v10
+; CHECK-NEXT: vrsub.vi v9, v9, 0
+; CHECK-NEXT: vand.vi v9, v9, 15
+; CHECK-NEXT: vsrl.vv v8, v8, v9
+; CHECK-NEXT: vor.vv v8, v10, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vx_nxv1i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e16, mf4, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 1 x i16> poison, i16 %b, i32 0
+ %b.splat = shufflevector <vscale x 1 x i16> %b.head, <vscale x 1 x i16> poison, <vscale x 1 x i32> zeroinitializer
+ %x = call <vscale x 1 x i16> @llvm.fshl.nxv1i16(<vscale x 1 x i16> %a, <vscale x 1 x i16> %a, <vscale x 1 x i16> %b.splat)
+ ret <vscale x 1 x i16> %x
+}
+
+declare <vscale x 2 x i16> @llvm.fshl.nxv2i16(<vscale x 2 x i16>, <vscale x 2 x i16>, <vscale x 2 x i16>)
+
+define <vscale x 2 x i16> @vrol_vv_nxv2i16(<vscale x 2 x i16> %a, <vscale x 2 x i16> %b) {
+; CHECK-LABEL: vrol_vv_nxv2i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e16, mf2, ta, ma
+; CHECK-NEXT: vand.vi v10, v9, 15
+; CHECK-NEXT: vsll.vv v10, v8, v10
+; CHECK-NEXT: vrsub.vi v9, v9, 0
+; CHECK-NEXT: vand.vi v9, v9, 15
+; CHECK-NEXT: vsrl.vv v8, v8, v9
+; CHECK-NEXT: vor.vv v8, v10, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vv_nxv2i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e16, mf2, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vv v8, v8, v9
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 2 x i16> @llvm.fshl.nxv2i16(<vscale x 2 x i16> %a, <vscale x 2 x i16> %a, <vscale x 2 x i16> %b)
+ ret <vscale x 2 x i16> %x
+}
+
+define <vscale x 2 x i16> @vrol_vx_nxv2i16(<vscale x 2 x i16> %a, i16 %b) {
+; CHECK-LABEL: vrol_vx_nxv2i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a1, zero, e16, mf2, ta, ma
+; CHECK-NEXT: vmv.v.x v9, a0
+; CHECK-NEXT: vand.vi v10, v9, 15
+; CHECK-NEXT: vsll.vv v10, v8, v10
+; CHECK-NEXT: vrsub.vi v9, v9, 0
+; CHECK-NEXT: vand.vi v9, v9, 15
+; CHECK-NEXT: vsrl.vv v8, v8, v9
+; CHECK-NEXT: vor.vv v8, v10, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vx_nxv2i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e16, mf2, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 2 x i16> poison, i16 %b, i32 0
+ %b.splat = shufflevector <vscale x 2 x i16> %b.head, <vscale x 2 x i16> poison, <vscale x 2 x i32> zeroinitializer
+ %x = call <vscale x 2 x i16> @llvm.fshl.nxv2i16(<vscale x 2 x i16> %a, <vscale x 2 x i16> %a, <vscale x 2 x i16> %b.splat)
+ ret <vscale x 2 x i16> %x
+}
+
+declare <vscale x 4 x i16> @llvm.fshl.nxv4i16(<vscale x 4 x i16>, <vscale x 4 x i16>, <vscale x 4 x i16>)
+
+define <vscale x 4 x i16> @vrol_vv_nxv4i16(<vscale x 4 x i16> %a, <vscale x 4 x i16> %b) {
+; CHECK-LABEL: vrol_vv_nxv4i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e16, m1, ta, ma
+; CHECK-NEXT: vand.vi v10, v9, 15
+; CHECK-NEXT: vsll.vv v10, v8, v10
+; CHECK-NEXT: vrsub.vi v9, v9, 0
+; CHECK-NEXT: vand.vi v9, v9, 15
+; CHECK-NEXT: vsrl.vv v8, v8, v9
+; CHECK-NEXT: vor.vv v8, v10, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vv_nxv4i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e16, m1, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vv v8, v8, v9
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 4 x i16> @llvm.fshl.nxv4i16(<vscale x 4 x i16> %a, <vscale x 4 x i16> %a, <vscale x 4 x i16> %b)
+ ret <vscale x 4 x i16> %x
+}
+
+define <vscale x 4 x i16> @vrol_vx_nxv4i16(<vscale x 4 x i16> %a, i16 %b) {
+; CHECK-LABEL: vrol_vx_nxv4i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a1, zero, e16, m1, ta, ma
+; CHECK-NEXT: vmv.v.x v9, a0
+; CHECK-NEXT: vand.vi v10, v9, 15
+; CHECK-NEXT: vsll.vv v10, v8, v10
+; CHECK-NEXT: vrsub.vi v9, v9, 0
+; CHECK-NEXT: vand.vi v9, v9, 15
+; CHECK-NEXT: vsrl.vv v8, v8, v9
+; CHECK-NEXT: vor.vv v8, v10, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vx_nxv4i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e16, m1, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 4 x i16> poison, i16 %b, i32 0
+ %b.splat = shufflevector <vscale x 4 x i16> %b.head, <vscale x 4 x i16> poison, <vscale x 4 x i32> zeroinitializer
+ %x = call <vscale x 4 x i16> @llvm.fshl.nxv4i16(<vscale x 4 x i16> %a, <vscale x 4 x i16> %a, <vscale x 4 x i16> %b.splat)
+ ret <vscale x 4 x i16> %x
+}
+
+declare <vscale x 8 x i16> @llvm.fshl.nxv8i16(<vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>)
+
+define <vscale x 8 x i16> @vrol_vv_nxv8i16(<vscale x 8 x i16> %a, <vscale x 8 x i16> %b) {
+; CHECK-LABEL: vrol_vv_nxv8i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e16, m2, ta, ma
+; CHECK-NEXT: vand.vi v12, v10, 15
+; CHECK-NEXT: vsll.vv v12, v8, v12
+; CHECK-NEXT: vrsub.vi v10, v10, 0
+; CHECK-NEXT: vand.vi v10, v10, 15
+; CHECK-NEXT: vsrl.vv v8, v8, v10
+; CHECK-NEXT: vor.vv v8, v12, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vv_nxv8i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e16, m2, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vv v8, v8, v10
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 8 x i16> @llvm.fshl.nxv8i16(<vscale x 8 x i16> %a, <vscale x 8 x i16> %a, <vscale x 8 x i16> %b)
+ ret <vscale x 8 x i16> %x
+}
+
+define <vscale x 8 x i16> @vrol_vx_nxv8i16(<vscale x 8 x i16> %a, i16 %b) {
+; CHECK-LABEL: vrol_vx_nxv8i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a1, zero, e16, m2, ta, ma
+; CHECK-NEXT: vmv.v.x v10, a0
+; CHECK-NEXT: vand.vi v12, v10, 15
+; CHECK-NEXT: vsll.vv v12, v8, v12
+; CHECK-NEXT: vrsub.vi v10, v10, 0
+; CHECK-NEXT: vand.vi v10, v10, 15
+; CHECK-NEXT: vsrl.vv v8, v8, v10
+; CHECK-NEXT: vor.vv v8, v12, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vx_nxv8i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e16, m2, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 8 x i16> poison, i16 %b, i32 0
+ %b.splat = shufflevector <vscale x 8 x i16> %b.head, <vscale x 8 x i16> poison, <vscale x 8 x i32> zeroinitializer
+ %x = call <vscale x 8 x i16> @llvm.fshl.nxv8i16(<vscale x 8 x i16> %a, <vscale x 8 x i16> %a, <vscale x 8 x i16> %b.splat)
+ ret <vscale x 8 x i16> %x
+}
+
+declare <vscale x 16 x i16> @llvm.fshl.nxv16i16(<vscale x 16 x i16>, <vscale x 16 x i16>, <vscale x 16 x i16>)
+
+define <vscale x 16 x i16> @vrol_vv_nxv16i16(<vscale x 16 x i16> %a, <vscale x 16 x i16> %b) {
+; CHECK-LABEL: vrol_vv_nxv16i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e16, m4, ta, ma
+; CHECK-NEXT: vand.vi v16, v12, 15
+; CHECK-NEXT: vsll.vv v16, v8, v16
+; CHECK-NEXT: vrsub.vi v12, v12, 0
+; CHECK-NEXT: vand.vi v12, v12, 15
+; CHECK-NEXT: vsrl.vv v8, v8, v12
+; CHECK-NEXT: vor.vv v8, v16, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vv_nxv16i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e16, m4, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vv v8, v8, v12
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 16 x i16> @llvm.fshl.nxv16i16(<vscale x 16 x i16> %a, <vscale x 16 x i16> %a, <vscale x 16 x i16> %b)
+ ret <vscale x 16 x i16> %x
+}
+
+define <vscale x 16 x i16> @vrol_vx_nxv16i16(<vscale x 16 x i16> %a, i16 %b) {
+; CHECK-LABEL: vrol_vx_nxv16i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a1, zero, e16, m4, ta, ma
+; CHECK-NEXT: vmv.v.x v12, a0
+; CHECK-NEXT: vand.vi v16, v12, 15
+; CHECK-NEXT: vsll.vv v16, v8, v16
+; CHECK-NEXT: vrsub.vi v12, v12, 0
+; CHECK-NEXT: vand.vi v12, v12, 15
+; CHECK-NEXT: vsrl.vv v8, v8, v12
+; CHECK-NEXT: vor.vv v8, v16, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vx_nxv16i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e16, m4, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 16 x i16> poison, i16 %b, i32 0
+ %b.splat = shufflevector <vscale x 16 x i16> %b.head, <vscale x 16 x i16> poison, <vscale x 16 x i32> zeroinitializer
+ %x = call <vscale x 16 x i16> @llvm.fshl.nxv16i16(<vscale x 16 x i16> %a, <vscale x 16 x i16> %a, <vscale x 16 x i16> %b.splat)
+ ret <vscale x 16 x i16> %x
+}
+
+declare <vscale x 32 x i16> @llvm.fshl.nxv32i16(<vscale x 32 x i16>, <vscale x 32 x i16>, <vscale x 32 x i16>)
+
+define <vscale x 32 x i16> @vrol_vv_nxv32i16(<vscale x 32 x i16> %a, <vscale x 32 x i16> %b) {
+; CHECK-LABEL: vrol_vv_nxv32i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e16, m8, ta, ma
+; CHECK-NEXT: vand.vi v24, v16, 15
+; CHECK-NEXT: vsll.vv v24, v8, v24
+; CHECK-NEXT: vrsub.vi v16, v16, 0
+; CHECK-NEXT: vand.vi v16, v16, 15
+; CHECK-NEXT: vsrl.vv v8, v8, v16
+; CHECK-NEXT: vor.vv v8, v24, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vv_nxv32i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e16, m8, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vv v8, v8, v16
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 32 x i16> @llvm.fshl.nxv32i16(<vscale x 32 x i16> %a, <vscale x 32 x i16> %a, <vscale x 32 x i16> %b)
+ ret <vscale x 32 x i16> %x
+}
+
+define <vscale x 32 x i16> @vrol_vx_nxv32i16(<vscale x 32 x i16> %a, i16 %b) {
+; CHECK-LABEL: vrol_vx_nxv32i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a1, zero, e16, m8, ta, ma
+; CHECK-NEXT: vmv.v.x v16, a0
+; CHECK-NEXT: vand.vi v24, v16, 15
+; CHECK-NEXT: vsll.vv v24, v8, v24
+; CHECK-NEXT: vrsub.vi v16, v16, 0
+; CHECK-NEXT: vand.vi v16, v16, 15
+; CHECK-NEXT: vsrl.vv v8, v8, v16
+; CHECK-NEXT: vor.vv v8, v24, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vx_nxv32i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e16, m8, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 32 x i16> poison, i16 %b, i32 0
+ %b.splat = shufflevector <vscale x 32 x i16> %b.head, <vscale x 32 x i16> poison, <vscale x 32 x i32> zeroinitializer
+ %x = call <vscale x 32 x i16> @llvm.fshl.nxv32i16(<vscale x 32 x i16> %a, <vscale x 32 x i16> %a, <vscale x 32 x i16> %b.splat)
+ ret <vscale x 32 x i16> %x
+}
+
+declare <vscale x 1 x i32> @llvm.fshl.nxv1i32(<vscale x 1 x i32>, <vscale x 1 x i32>, <vscale x 1 x i32>)
+
+define <vscale x 1 x i32> @vrol_vv_nxv1i32(<vscale x 1 x i32> %a, <vscale x 1 x i32> %b) {
+; CHECK-LABEL: vrol_vv_nxv1i32:
+; CHECK: # %bb.0:
+; CHECK-NEXT: li a0, 31
+; CHECK-NEXT: vsetvli a1, zero, e32, mf2, ta, ma
+; CHECK-NEXT: vand.vx v10, v9, a0
+; CHECK-NEXT: vsll.vv v10, v8, v10
+; CHECK-NEXT: vrsub.vi v9, v9, 0
+; CHECK-NEXT: vand.vx v9, v9, a0
+; CHECK-NEXT: vsrl.vv v8, v8, v9
+; CHECK-NEXT: vor.vv v8, v10, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vv_nxv1i32:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e32, mf2, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vv v8, v8, v9
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 1 x i32> @llvm.fshl.nxv1i32(<vscale x 1 x i32> %a, <vscale x 1 x i32> %a, <vscale x 1 x i32> %b)
+ ret <vscale x 1 x i32> %x
+}
+
+define <vscale x 1 x i32> @vrol_vx_nxv1i32(<vscale x 1 x i32> %a, i32 %b) {
+; CHECK-RV32-LABEL: vrol_vx_nxv1i32:
+; CHECK-RV32: # %bb.0:
+; CHECK-RV32-NEXT: andi a1, a0, 31
+; CHECK-RV32-NEXT: vsetvli a2, zero, e32, mf2, ta, ma
+; CHECK-RV32-NEXT: vsll.vx v9, v8, a1
+; CHECK-RV32-NEXT: neg a0, a0
+; CHECK-RV32-NEXT: andi a0, a0, 31
+; CHECK-RV32-NEXT: vsrl.vx v8, v8, a0
+; CHECK-RV32-NEXT: vor.vv v8, v9, v8
+; CHECK-RV32-NEXT: ret
+;
+; CHECK-RV64-LABEL: vrol_vx_nxv1i32:
+; CHECK-RV64: # %bb.0:
+; CHECK-RV64-NEXT: vsetvli a1, zero, e32, mf2, ta, ma
+; CHECK-RV64-NEXT: vmv.v.x v9, a0
+; CHECK-RV64-NEXT: li a0, 31
+; CHECK-RV64-NEXT: vand.vx v10, v9, a0
+; CHECK-RV64-NEXT: vsll.vv v10, v8, v10
+; CHECK-RV64-NEXT: vrsub.vi v9, v9, 0
+; CHECK-RV64-NEXT: vand.vx v9, v9, a0
+; CHECK-RV64-NEXT: vsrl.vv v8, v8, v9
+; CHECK-RV64-NEXT: vor.vv v8, v10, v8
+; CHECK-RV64-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vx_nxv1i32:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e32, mf2, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 1 x i32> poison, i32 %b, i32 0
+ %b.splat = shufflevector <vscale x 1 x i32> %b.head, <vscale x 1 x i32> poison, <vscale x 1 x i32> zeroinitializer
+ %x = call <vscale x 1 x i32> @llvm.fshl.nxv1i32(<vscale x 1 x i32> %a, <vscale x 1 x i32> %a, <vscale x 1 x i32> %b.splat)
+ ret <vscale x 1 x i32> %x
+}
+
+declare <vscale x 2 x i32> @llvm.fshl.nxv2i32(<vscale x 2 x i32>, <vscale x 2 x i32>, <vscale x 2 x i32>)
+
+define <vscale x 2 x i32> @vrol_vv_nxv2i32(<vscale x 2 x i32> %a, <vscale x 2 x i32> %b) {
+; CHECK-LABEL: vrol_vv_nxv2i32:
+; CHECK: # %bb.0:
+; CHECK-NEXT: li a0, 31
+; CHECK-NEXT: vsetvli a1, zero, e32, m1, ta, ma
+; CHECK-NEXT: vand.vx v10, v9, a0
+; CHECK-NEXT: vsll.vv v10, v8, v10
+; CHECK-NEXT: vrsub.vi v9, v9, 0
+; CHECK-NEXT: vand.vx v9, v9, a0
+; CHECK-NEXT: vsrl.vv v8, v8, v9
+; CHECK-NEXT: vor.vv v8, v10, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vv_nxv2i32:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e32, m1, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vv v8, v8, v9
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 2 x i32> @llvm.fshl.nxv2i32(<vscale x 2 x i32> %a, <vscale x 2 x i32> %a, <vscale x 2 x i32> %b)
+ ret <vscale x 2 x i32> %x
+}
+
+define <vscale x 2 x i32> @vrol_vx_nxv2i32(<vscale x 2 x i32> %a, i32 %b) {
+; CHECK-RV32-LABEL: vrol_vx_nxv2i32:
+; CHECK-RV32: # %bb.0:
+; CHECK-RV32-NEXT: andi a1, a0, 31
+; CHECK-RV32-NEXT: vsetvli a2, zero, e32, m1, ta, ma
+; CHECK-RV32-NEXT: vsll.vx v9, v8, a1
+; CHECK-RV32-NEXT: neg a0, a0
+; CHECK-RV32-NEXT: andi a0, a0, 31
+; CHECK-RV32-NEXT: vsrl.vx v8, v8, a0
+; CHECK-RV32-NEXT: vor.vv v8, v9, v8
+; CHECK-RV32-NEXT: ret
+;
+; CHECK-RV64-LABEL: vrol_vx_nxv2i32:
+; CHECK-RV64: # %bb.0:
+; CHECK-RV64-NEXT: vsetvli a1, zero, e32, m1, ta, ma
+; CHECK-RV64-NEXT: vmv.v.x v9, a0
+; CHECK-RV64-NEXT: li a0, 31
+; CHECK-RV64-NEXT: vand.vx v10, v9, a0
+; CHECK-RV64-NEXT: vsll.vv v10, v8, v10
+; CHECK-RV64-NEXT: vrsub.vi v9, v9, 0
+; CHECK-RV64-NEXT: vand.vx v9, v9, a0
+; CHECK-RV64-NEXT: vsrl.vv v8, v8, v9
+; CHECK-RV64-NEXT: vor.vv v8, v10, v8
+; CHECK-RV64-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vx_nxv2i32:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e32, m1, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 2 x i32> poison, i32 %b, i32 0
+ %b.splat = shufflevector <vscale x 2 x i32> %b.head, <vscale x 2 x i32> poison, <vscale x 2 x i32> zeroinitializer
+ %x = call <vscale x 2 x i32> @llvm.fshl.nxv2i32(<vscale x 2 x i32> %a, <vscale x 2 x i32> %a, <vscale x 2 x i32> %b.splat)
+ ret <vscale x 2 x i32> %x
+}
+
+declare <vscale x 4 x i32> @llvm.fshl.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>)
+
+define <vscale x 4 x i32> @vrol_vv_nxv4i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b) {
+; CHECK-LABEL: vrol_vv_nxv4i32:
+; CHECK: # %bb.0:
+; CHECK-NEXT: li a0, 31
+; CHECK-NEXT: vsetvli a1, zero, e32, m2, ta, ma
+; CHECK-NEXT: vand.vx v12, v10, a0
+; CHECK-NEXT: vsll.vv v12, v8, v12
+; CHECK-NEXT: vrsub.vi v10, v10, 0
+; CHECK-NEXT: vand.vx v10, v10, a0
+; CHECK-NEXT: vsrl.vv v8, v8, v10
+; CHECK-NEXT: vor.vv v8, v12, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vv_nxv4i32:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e32, m2, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vv v8, v8, v10
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 4 x i32> @llvm.fshl.nxv4i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %a, <vscale x 4 x i32> %b)
+ ret <vscale x 4 x i32> %x
+}
+
+define <vscale x 4 x i32> @vrol_vx_nxv4i32(<vscale x 4 x i32> %a, i32 %b) {
+; CHECK-RV32-LABEL: vrol_vx_nxv4i32:
+; CHECK-RV32: # %bb.0:
+; CHECK-RV32-NEXT: andi a1, a0, 31
+; CHECK-RV32-NEXT: vsetvli a2, zero, e32, m2, ta, ma
+; CHECK-RV32-NEXT: vsll.vx v10, v8, a1
+; CHECK-RV32-NEXT: neg a0, a0
+; CHECK-RV32-NEXT: andi a0, a0, 31
+; CHECK-RV32-NEXT: vsrl.vx v8, v8, a0
+; CHECK-RV32-NEXT: vor.vv v8, v10, v8
+; CHECK-RV32-NEXT: ret
+;
+; CHECK-RV64-LABEL: vrol_vx_nxv4i32:
+; CHECK-RV64: # %bb.0:
+; CHECK-RV64-NEXT: vsetvli a1, zero, e32, m2, ta, ma
+; CHECK-RV64-NEXT: vmv.v.x v10, a0
+; CHECK-RV64-NEXT: li a0, 31
+; CHECK-RV64-NEXT: vand.vx v12, v10, a0
+; CHECK-RV64-NEXT: vsll.vv v12, v8, v12
+; CHECK-RV64-NEXT: vrsub.vi v10, v10, 0
+; CHECK-RV64-NEXT: vand.vx v10, v10, a0
+; CHECK-RV64-NEXT: vsrl.vv v8, v8, v10
+; CHECK-RV64-NEXT: vor.vv v8, v12, v8
+; CHECK-RV64-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vx_nxv4i32:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e32, m2, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 4 x i32> poison, i32 %b, i32 0
+ %b.splat = shufflevector <vscale x 4 x i32> %b.head, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
+ %x = call <vscale x 4 x i32> @llvm.fshl.nxv4i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %a, <vscale x 4 x i32> %b.splat)
+ ret <vscale x 4 x i32> %x
+}
+
+declare <vscale x 8 x i32> @llvm.fshl.nxv8i32(<vscale x 8 x i32>, <vscale x 8 x i32>, <vscale x 8 x i32>)
+
+define <vscale x 8 x i32> @vrol_vv_nxv8i32(<vscale x 8 x i32> %a, <vscale x 8 x i32> %b) {
+; CHECK-LABEL: vrol_vv_nxv8i32:
+; CHECK: # %bb.0:
+; CHECK-NEXT: li a0, 31
+; CHECK-NEXT: vsetvli a1, zero, e32, m4, ta, ma
+; CHECK-NEXT: vand.vx v16, v12, a0
+; CHECK-NEXT: vsll.vv v16, v8, v16
+; CHECK-NEXT: vrsub.vi v12, v12, 0
+; CHECK-NEXT: vand.vx v12, v12, a0
+; CHECK-NEXT: vsrl.vv v8, v8, v12
+; CHECK-NEXT: vor.vv v8, v16, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vv_nxv8i32:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e32, m4, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vv v8, v8, v12
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 8 x i32> @llvm.fshl.nxv8i32(<vscale x 8 x i32> %a, <vscale x 8 x i32> %a, <vscale x 8 x i32> %b)
+ ret <vscale x 8 x i32> %x
+}
+
+define <vscale x 8 x i32> @vrol_vx_nxv8i32(<vscale x 8 x i32> %a, i32 %b) {
+; CHECK-RV32-LABEL: vrol_vx_nxv8i32:
+; CHECK-RV32: # %bb.0:
+; CHECK-RV32-NEXT: andi a1, a0, 31
+; CHECK-RV32-NEXT: vsetvli a2, zero, e32, m4, ta, ma
+; CHECK-RV32-NEXT: vsll.vx v12, v8, a1
+; CHECK-RV32-NEXT: neg a0, a0
+; CHECK-RV32-NEXT: andi a0, a0, 31
+; CHECK-RV32-NEXT: vsrl.vx v8, v8, a0
+; CHECK-RV32-NEXT: vor.vv v8, v12, v8
+; CHECK-RV32-NEXT: ret
+;
+; CHECK-RV64-LABEL: vrol_vx_nxv8i32:
+; CHECK-RV64: # %bb.0:
+; CHECK-RV64-NEXT: vsetvli a1, zero, e32, m4, ta, ma
+; CHECK-RV64-NEXT: vmv.v.x v12, a0
+; CHECK-RV64-NEXT: li a0, 31
+; CHECK-RV64-NEXT: vand.vx v16, v12, a0
+; CHECK-RV64-NEXT: vsll.vv v16, v8, v16
+; CHECK-RV64-NEXT: vrsub.vi v12, v12, 0
+; CHECK-RV64-NEXT: vand.vx v12, v12, a0
+; CHECK-RV64-NEXT: vsrl.vv v8, v8, v12
+; CHECK-RV64-NEXT: vor.vv v8, v16, v8
+; CHECK-RV64-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vx_nxv8i32:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e32, m4, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 8 x i32> poison, i32 %b, i32 0
+ %b.splat = shufflevector <vscale x 8 x i32> %b.head, <vscale x 8 x i32> poison, <vscale x 8 x i32> zeroinitializer
+ %x = call <vscale x 8 x i32> @llvm.fshl.nxv8i32(<vscale x 8 x i32> %a, <vscale x 8 x i32> %a, <vscale x 8 x i32> %b.splat)
+ ret <vscale x 8 x i32> %x
+}
+
+declare <vscale x 16 x i32> @llvm.fshl.nxv16i32(<vscale x 16 x i32>, <vscale x 16 x i32>, <vscale x 16 x i32>)
+
+define <vscale x 16 x i32> @vrol_vv_nxv16i32(<vscale x 16 x i32> %a, <vscale x 16 x i32> %b) {
+; CHECK-LABEL: vrol_vv_nxv16i32:
+; CHECK: # %bb.0:
+; CHECK-NEXT: li a0, 31
+; CHECK-NEXT: vsetvli a1, zero, e32, m8, ta, ma
+; CHECK-NEXT: vand.vx v24, v16, a0
+; CHECK-NEXT: vsll.vv v24, v8, v24
+; CHECK-NEXT: vrsub.vi v16, v16, 0
+; CHECK-NEXT: vand.vx v16, v16, a0
+; CHECK-NEXT: vsrl.vv v8, v8, v16
+; CHECK-NEXT: vor.vv v8, v24, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vv_nxv16i32:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e32, m8, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vv v8, v8, v16
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 16 x i32> @llvm.fshl.nxv16i32(<vscale x 16 x i32> %a, <vscale x 16 x i32> %a, <vscale x 16 x i32> %b)
+ ret <vscale x 16 x i32> %x
+}
+
+define <vscale x 16 x i32> @vrol_vx_nxv16i32(<vscale x 16 x i32> %a, i32 %b) {
+; CHECK-RV32-LABEL: vrol_vx_nxv16i32:
+; CHECK-RV32: # %bb.0:
+; CHECK-RV32-NEXT: andi a1, a0, 31
+; CHECK-RV32-NEXT: vsetvli a2, zero, e32, m8, ta, ma
+; CHECK-RV32-NEXT: vsll.vx v16, v8, a1
+; CHECK-RV32-NEXT: neg a0, a0
+; CHECK-RV32-NEXT: andi a0, a0, 31
+; CHECK-RV32-NEXT: vsrl.vx v8, v8, a0
+; CHECK-RV32-NEXT: vor.vv v8, v16, v8
+; CHECK-RV32-NEXT: ret
+;
+; CHECK-RV64-LABEL: vrol_vx_nxv16i32:
+; CHECK-RV64: # %bb.0:
+; CHECK-RV64-NEXT: vsetvli a1, zero, e32, m8, ta, ma
+; CHECK-RV64-NEXT: vmv.v.x v16, a0
+; CHECK-RV64-NEXT: li a0, 31
+; CHECK-RV64-NEXT: vand.vx v24, v16, a0
+; CHECK-RV64-NEXT: vsll.vv v24, v8, v24
+; CHECK-RV64-NEXT: vrsub.vi v16, v16, 0
+; CHECK-RV64-NEXT: vand.vx v16, v16, a0
+; CHECK-RV64-NEXT: vsrl.vv v8, v8, v16
+; CHECK-RV64-NEXT: vor.vv v8, v24, v8
+; CHECK-RV64-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vx_nxv16i32:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e32, m8, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 16 x i32> poison, i32 %b, i32 0
+ %b.splat = shufflevector <vscale x 16 x i32> %b.head, <vscale x 16 x i32> poison, <vscale x 16 x i32> zeroinitializer
+ %x = call <vscale x 16 x i32> @llvm.fshl.nxv16i32(<vscale x 16 x i32> %a, <vscale x 16 x i32> %a, <vscale x 16 x i32> %b.splat)
+ ret <vscale x 16 x i32> %x
+}
+
+declare <vscale x 1 x i64> @llvm.fshl.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>)
+
+define <vscale x 1 x i64> @vrol_vv_nxv1i64(<vscale x 1 x i64> %a, <vscale x 1 x i64> %b) {
+; CHECK-LABEL: vrol_vv_nxv1i64:
+; CHECK: # %bb.0:
+; CHECK-NEXT: li a0, 63
+; CHECK-NEXT: vsetvli a1, zero, e64, m1, ta, ma
+; CHECK-NEXT: vand.vx v10, v9, a0
+; CHECK-NEXT: vsll.vv v10, v8, v10
+; CHECK-NEXT: vrsub.vi v9, v9, 0
+; CHECK-NEXT: vand.vx v9, v9, a0
+; CHECK-NEXT: vsrl.vv v8, v8, v9
+; CHECK-NEXT: vor.vv v8, v10, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vv_nxv1i64:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vv v8, v8, v9
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 1 x i64> @llvm.fshl.nxv1i64(<vscale x 1 x i64> %a, <vscale x 1 x i64> %a, <vscale x 1 x i64> %b)
+ ret <vscale x 1 x i64> %x
+}
+
+define <vscale x 1 x i64> @vrol_vx_nxv1i64(<vscale x 1 x i64> %a, i64 %b) {
+; CHECK-RV32-LABEL: vrol_vx_nxv1i64:
+; CHECK-RV32: # %bb.0:
+; CHECK-RV32-NEXT: addi sp, sp, -16
+; CHECK-RV32-NEXT: .cfi_def_cfa_offset 16
+; CHECK-RV32-NEXT: sw a1, 12(sp)
+; CHECK-RV32-NEXT: sw a0, 8(sp)
+; CHECK-RV32-NEXT: addi a0, sp, 8
+; CHECK-RV32-NEXT: vsetvli a1, zero, e64, m1, ta, ma
+; CHECK-RV32-NEXT: vlse64.v v9, (a0), zero
+; CHECK-RV32-NEXT: li a0, 63
+; CHECK-RV32-NEXT: vand.vx v10, v9, a0
+; CHECK-RV32-NEXT: vsll.vv v10, v8, v10
+; CHECK-RV32-NEXT: vrsub.vi v9, v9, 0
+; CHECK-RV32-NEXT: vand.vx v9, v9, a0
+; CHECK-RV32-NEXT: vsrl.vv v8, v8, v9
+; CHECK-RV32-NEXT: vor.vv v8, v10, v8
+; CHECK-RV32-NEXT: addi sp, sp, 16
+; CHECK-RV32-NEXT: ret
+;
+; CHECK-RV64-LABEL: vrol_vx_nxv1i64:
+; CHECK-RV64: # %bb.0:
+; CHECK-RV64-NEXT: andi a1, a0, 63
+; CHECK-RV64-NEXT: vsetvli a2, zero, e64, m1, ta, ma
+; CHECK-RV64-NEXT: vsll.vx v9, v8, a1
+; CHECK-RV64-NEXT: negw a0, a0
+; CHECK-RV64-NEXT: andi a0, a0, 63
+; CHECK-RV64-NEXT: vsrl.vx v8, v8, a0
+; CHECK-RV64-NEXT: vor.vv v8, v9, v8
+; CHECK-RV64-NEXT: ret
+;
+; CHECK-ZVBB32-LABEL: vrol_vx_nxv1i64:
+; CHECK-ZVBB32: # %bb.0:
+; CHECK-ZVBB32-NEXT: addi sp, sp, -16
+; CHECK-ZVBB32-NEXT: .cfi_def_cfa_offset 16
+; CHECK-ZVBB32-NEXT: sw a1, 12(sp)
+; CHECK-ZVBB32-NEXT: sw a0, 8(sp)
+; CHECK-ZVBB32-NEXT: addi a0, sp, 8
+; CHECK-ZVBB32-NEXT: vsetvli a1, zero, e64, m1, ta, ma
+; CHECK-ZVBB32-NEXT: vlse64.v v9, (a0), zero
+; CHECK-ZVBB32-NEXT: vrol.vv v8, v8, v9
+; CHECK-ZVBB32-NEXT: addi sp, sp, 16
+; CHECK-ZVBB32-NEXT: ret
+;
+; CHECK-ZVBB64-LABEL: vrol_vx_nxv1i64:
+; CHECK-ZVBB64: # %bb.0:
+; CHECK-ZVBB64-NEXT: vsetvli a1, zero, e64, m1, ta, ma
+; CHECK-ZVBB64-NEXT: vrol.vx v8, v8, a0
+; CHECK-ZVBB64-NEXT: ret
+ %b.head = insertelement <vscale x 1 x i64> poison, i64 %b, i32 0
+ %b.splat = shufflevector <vscale x 1 x i64> %b.head, <vscale x 1 x i64> poison, <vscale x 1 x i32> zeroinitializer
+ %x = call <vscale x 1 x i64> @llvm.fshl.nxv1i64(<vscale x 1 x i64> %a, <vscale x 1 x i64> %a, <vscale x 1 x i64> %b.splat)
+ ret <vscale x 1 x i64> %x
+}
+
+declare <vscale x 2 x i64> @llvm.fshl.nxv2i64(<vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>)
+
+define <vscale x 2 x i64> @vrol_vv_nxv2i64(<vscale x 2 x i64> %a, <vscale x 2 x i64> %b) {
+; CHECK-LABEL: vrol_vv_nxv2i64:
+; CHECK: # %bb.0:
+; CHECK-NEXT: li a0, 63
+; CHECK-NEXT: vsetvli a1, zero, e64, m2, ta, ma
+; CHECK-NEXT: vand.vx v12, v10, a0
+; CHECK-NEXT: vsll.vv v12, v8, v12
+; CHECK-NEXT: vrsub.vi v10, v10, 0
+; CHECK-NEXT: vand.vx v10, v10, a0
+; CHECK-NEXT: vsrl.vv v8, v8, v10
+; CHECK-NEXT: vor.vv v8, v12, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vv_nxv2i64:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e64, m2, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vv v8, v8, v10
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 2 x i64> @llvm.fshl.nxv2i64(<vscale x 2 x i64> %a, <vscale x 2 x i64> %a, <vscale x 2 x i64> %b)
+ ret <vscale x 2 x i64> %x
+}
+
+define <vscale x 2 x i64> @vrol_vx_nxv2i64(<vscale x 2 x i64> %a, i64 %b) {
+; CHECK-RV32-LABEL: vrol_vx_nxv2i64:
+; CHECK-RV32: # %bb.0:
+; CHECK-RV32-NEXT: addi sp, sp, -16
+; CHECK-RV32-NEXT: .cfi_def_cfa_offset 16
+; CHECK-RV32-NEXT: sw a1, 12(sp)
+; CHECK-RV32-NEXT: sw a0, 8(sp)
+; CHECK-RV32-NEXT: addi a0, sp, 8
+; CHECK-RV32-NEXT: vsetvli a1, zero, e64, m2, ta, ma
+; CHECK-RV32-NEXT: vlse64.v v10, (a0), zero
+; CHECK-RV32-NEXT: li a0, 63
+; CHECK-RV32-NEXT: vand.vx v12, v10, a0
+; CHECK-RV32-NEXT: vsll.vv v12, v8, v12
+; CHECK-RV32-NEXT: vrsub.vi v10, v10, 0
+; CHECK-RV32-NEXT: vand.vx v10, v10, a0
+; CHECK-RV32-NEXT: vsrl.vv v8, v8, v10
+; CHECK-RV32-NEXT: vor.vv v8, v12, v8
+; CHECK-RV32-NEXT: addi sp, sp, 16
+; CHECK-RV32-NEXT: ret
+;
+; CHECK-RV64-LABEL: vrol_vx_nxv2i64:
+; CHECK-RV64: # %bb.0:
+; CHECK-RV64-NEXT: andi a1, a0, 63
+; CHECK-RV64-NEXT: vsetvli a2, zero, e64, m2, ta, ma
+; CHECK-RV64-NEXT: vsll.vx v10, v8, a1
+; CHECK-RV64-NEXT: negw a0, a0
+; CHECK-RV64-NEXT: andi a0, a0, 63
+; CHECK-RV64-NEXT: vsrl.vx v8, v8, a0
+; CHECK-RV64-NEXT: vor.vv v8, v10, v8
+; CHECK-RV64-NEXT: ret
+;
+; CHECK-ZVBB32-LABEL: vrol_vx_nxv2i64:
+; CHECK-ZVBB32: # %bb.0:
+; CHECK-ZVBB32-NEXT: addi sp, sp, -16
+; CHECK-ZVBB32-NEXT: .cfi_def_cfa_offset 16
+; CHECK-ZVBB32-NEXT: sw a1, 12(sp)
+; CHECK-ZVBB32-NEXT: sw a0, 8(sp)
+; CHECK-ZVBB32-NEXT: addi a0, sp, 8
+; CHECK-ZVBB32-NEXT: vsetvli a1, zero, e64, m2, ta, ma
+; CHECK-ZVBB32-NEXT: vlse64.v v10, (a0), zero
+; CHECK-ZVBB32-NEXT: vrol.vv v8, v8, v10
+; CHECK-ZVBB32-NEXT: addi sp, sp, 16
+; CHECK-ZVBB32-NEXT: ret
+;
+; CHECK-ZVBB64-LABEL: vrol_vx_nxv2i64:
+; CHECK-ZVBB64: # %bb.0:
+; CHECK-ZVBB64-NEXT: vsetvli a1, zero, e64, m2, ta, ma
+; CHECK-ZVBB64-NEXT: vrol.vx v8, v8, a0
+; CHECK-ZVBB64-NEXT: ret
+ %b.head = insertelement <vscale x 2 x i64> poison, i64 %b, i32 0
+ %b.splat = shufflevector <vscale x 2 x i64> %b.head, <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer
+ %x = call <vscale x 2 x i64> @llvm.fshl.nxv2i64(<vscale x 2 x i64> %a, <vscale x 2 x i64> %a, <vscale x 2 x i64> %b.splat)
+ ret <vscale x 2 x i64> %x
+}
+
+declare <vscale x 4 x i64> @llvm.fshl.nxv4i64(<vscale x 4 x i64>, <vscale x 4 x i64>, <vscale x 4 x i64>)
+
+define <vscale x 4 x i64> @vrol_vv_nxv4i64(<vscale x 4 x i64> %a, <vscale x 4 x i64> %b) {
+; CHECK-LABEL: vrol_vv_nxv4i64:
+; CHECK: # %bb.0:
+; CHECK-NEXT: li a0, 63
+; CHECK-NEXT: vsetvli a1, zero, e64, m4, ta, ma
+; CHECK-NEXT: vand.vx v16, v12, a0
+; CHECK-NEXT: vsll.vv v16, v8, v16
+; CHECK-NEXT: vrsub.vi v12, v12, 0
+; CHECK-NEXT: vand.vx v12, v12, a0
+; CHECK-NEXT: vsrl.vv v8, v8, v12
+; CHECK-NEXT: vor.vv v8, v16, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vv_nxv4i64:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e64, m4, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vv v8, v8, v12
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 4 x i64> @llvm.fshl.nxv4i64(<vscale x 4 x i64> %a, <vscale x 4 x i64> %a, <vscale x 4 x i64> %b)
+ ret <vscale x 4 x i64> %x
+}
+
+define <vscale x 4 x i64> @vrol_vx_nxv4i64(<vscale x 4 x i64> %a, i64 %b) {
+; CHECK-RV32-LABEL: vrol_vx_nxv4i64:
+; CHECK-RV32: # %bb.0:
+; CHECK-RV32-NEXT: addi sp, sp, -16
+; CHECK-RV32-NEXT: .cfi_def_cfa_offset 16
+; CHECK-RV32-NEXT: sw a1, 12(sp)
+; CHECK-RV32-NEXT: sw a0, 8(sp)
+; CHECK-RV32-NEXT: addi a0, sp, 8
+; CHECK-RV32-NEXT: vsetvli a1, zero, e64, m4, ta, ma
+; CHECK-RV32-NEXT: vlse64.v v12, (a0), zero
+; CHECK-RV32-NEXT: li a0, 63
+; CHECK-RV32-NEXT: vand.vx v16, v12, a0
+; CHECK-RV32-NEXT: vsll.vv v16, v8, v16
+; CHECK-RV32-NEXT: vrsub.vi v12, v12, 0
+; CHECK-RV32-NEXT: vand.vx v12, v12, a0
+; CHECK-RV32-NEXT: vsrl.vv v8, v8, v12
+; CHECK-RV32-NEXT: vor.vv v8, v16, v8
+; CHECK-RV32-NEXT: addi sp, sp, 16
+; CHECK-RV32-NEXT: ret
+;
+; CHECK-RV64-LABEL: vrol_vx_nxv4i64:
+; CHECK-RV64: # %bb.0:
+; CHECK-RV64-NEXT: andi a1, a0, 63
+; CHECK-RV64-NEXT: vsetvli a2, zero, e64, m4, ta, ma
+; CHECK-RV64-NEXT: vsll.vx v12, v8, a1
+; CHECK-RV64-NEXT: negw a0, a0
+; CHECK-RV64-NEXT: andi a0, a0, 63
+; CHECK-RV64-NEXT: vsrl.vx v8, v8, a0
+; CHECK-RV64-NEXT: vor.vv v8, v12, v8
+; CHECK-RV64-NEXT: ret
+;
+; CHECK-ZVBB32-LABEL: vrol_vx_nxv4i64:
+; CHECK-ZVBB32: # %bb.0:
+; CHECK-ZVBB32-NEXT: addi sp, sp, -16
+; CHECK-ZVBB32-NEXT: .cfi_def_cfa_offset 16
+; CHECK-ZVBB32-NEXT: sw a1, 12(sp)
+; CHECK-ZVBB32-NEXT: sw a0, 8(sp)
+; CHECK-ZVBB32-NEXT: addi a0, sp, 8
+; CHECK-ZVBB32-NEXT: vsetvli a1, zero, e64, m4, ta, ma
+; CHECK-ZVBB32-NEXT: vlse64.v v12, (a0), zero
+; CHECK-ZVBB32-NEXT: vrol.vv v8, v8, v12
+; CHECK-ZVBB32-NEXT: addi sp, sp, 16
+; CHECK-ZVBB32-NEXT: ret
+;
+; CHECK-ZVBB64-LABEL: vrol_vx_nxv4i64:
+; CHECK-ZVBB64: # %bb.0:
+; CHECK-ZVBB64-NEXT: vsetvli a1, zero, e64, m4, ta, ma
+; CHECK-ZVBB64-NEXT: vrol.vx v8, v8, a0
+; CHECK-ZVBB64-NEXT: ret
+ %b.head = insertelement <vscale x 4 x i64> poison, i64 %b, i32 0
+ %b.splat = shufflevector <vscale x 4 x i64> %b.head, <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer
+ %x = call <vscale x 4 x i64> @llvm.fshl.nxv4i64(<vscale x 4 x i64> %a, <vscale x 4 x i64> %a, <vscale x 4 x i64> %b.splat)
+ ret <vscale x 4 x i64> %x
+}
+
+declare <vscale x 8 x i64> @llvm.fshl.nxv8i64(<vscale x 8 x i64>, <vscale x 8 x i64>, <vscale x 8 x i64>)
+
+define <vscale x 8 x i64> @vrol_vv_nxv8i64(<vscale x 8 x i64> %a, <vscale x 8 x i64> %b) {
+; CHECK-LABEL: vrol_vv_nxv8i64:
+; CHECK: # %bb.0:
+; CHECK-NEXT: li a0, 63
+; CHECK-NEXT: vsetvli a1, zero, e64, m8, ta, ma
+; CHECK-NEXT: vand.vx v24, v16, a0
+; CHECK-NEXT: vsll.vv v24, v8, v24
+; CHECK-NEXT: vrsub.vi v16, v16, 0
+; CHECK-NEXT: vand.vx v16, v16, a0
+; CHECK-NEXT: vsrl.vv v8, v8, v16
+; CHECK-NEXT: vor.vv v8, v24, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vrol_vv_nxv8i64:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e64, m8, ta, ma
+; CHECK-ZVBB-NEXT: vrol.vv v8, v8, v16
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 8 x i64> @llvm.fshl.nxv8i64(<vscale x 8 x i64> %a, <vscale x 8 x i64> %a, <vscale x 8 x i64> %b)
+ ret <vscale x 8 x i64> %x
+}
+
+define <vscale x 8 x i64> @vrol_vx_nxv8i64(<vscale x 8 x i64> %a, i64 %b) {
+; CHECK-RV32-LABEL: vrol_vx_nxv8i64:
+; CHECK-RV32: # %bb.0:
+; CHECK-RV32-NEXT: addi sp, sp, -16
+; CHECK-RV32-NEXT: .cfi_def_cfa_offset 16
+; CHECK-RV32-NEXT: sw a1, 12(sp)
+; CHECK-RV32-NEXT: sw a0, 8(sp)
+; CHECK-RV32-NEXT: addi a0, sp, 8
+; CHECK-RV32-NEXT: vsetvli a1, zero, e64, m8, ta, ma
+; CHECK-RV32-NEXT: vlse64.v v16, (a0), zero
+; CHECK-RV32-NEXT: li a0, 63
+; CHECK-RV32-NEXT: vand.vx v24, v16, a0
+; CHECK-RV32-NEXT: vsll.vv v24, v8, v24
+; CHECK-RV32-NEXT: vrsub.vi v16, v16, 0
+; CHECK-RV32-NEXT: vand.vx v16, v16, a0
+; CHECK-RV32-NEXT: vsrl.vv v8, v8, v16
+; CHECK-RV32-NEXT: vor.vv v8, v24, v8
+; CHECK-RV32-NEXT: addi sp, sp, 16
+; CHECK-RV32-NEXT: ret
+;
+; CHECK-RV64-LABEL: vrol_vx_nxv8i64:
+; CHECK-RV64: # %bb.0:
+; CHECK-RV64-NEXT: andi a1, a0, 63
+; CHECK-RV64-NEXT: vsetvli a2, zero, e64, m8, ta, ma
+; CHECK-RV64-NEXT: vsll.vx v16, v8, a1
+; CHECK-RV64-NEXT: negw a0, a0
+; CHECK-RV64-NEXT: andi a0, a0, 63
+; CHECK-RV64-NEXT: vsrl.vx v8, v8, a0
+; CHECK-RV64-NEXT: vor.vv v8, v16, v8
+; CHECK-RV64-NEXT: ret
+;
+; CHECK-ZVBB32-LABEL: vrol_vx_nxv8i64:
+; CHECK-ZVBB32: # %bb.0:
+; CHECK-ZVBB32-NEXT: addi sp, sp, -16
+; CHECK-ZVBB32-NEXT: .cfi_def_cfa_offset 16
+; CHECK-ZVBB32-NEXT: sw a1, 12(sp)
+; CHECK-ZVBB32-NEXT: sw a0, 8(sp)
+; CHECK-ZVBB32-NEXT: addi a0, sp, 8
+; CHECK-ZVBB32-NEXT: vsetvli a1, zero, e64, m8, ta, ma
+; CHECK-ZVBB32-NEXT: vlse64.v v16, (a0), zero
+; CHECK-ZVBB32-NEXT: vrol.vv v8, v8, v16
+; CHECK-ZVBB32-NEXT: addi sp, sp, 16
+; CHECK-ZVBB32-NEXT: ret
+;
+; CHECK-ZVBB64-LABEL: vrol_vx_nxv8i64:
+; CHECK-ZVBB64: # %bb.0:
+; CHECK-ZVBB64-NEXT: vsetvli a1, zero, e64, m8, ta, ma
+; CHECK-ZVBB64-NEXT: vrol.vx v8, v8, a0
+; CHECK-ZVBB64-NEXT: ret
+ %b.head = insertelement <vscale x 8 x i64> poison, i64 %b, i32 0
+ %b.splat = shufflevector <vscale x 8 x i64> %b.head, <vscale x 8 x i64> poison, <vscale x 8 x i32> zeroinitializer
+ %x = call <vscale x 8 x i64> @llvm.fshl.nxv8i64(<vscale x 8 x i64> %a, <vscale x 8 x i64> %a, <vscale x 8 x i64> %b.splat)
+ ret <vscale x 8 x i64> %x
+}
+
--- /dev/null
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 2
+; RUN: llc -mtriple=riscv32 -mattr=+v -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,CHECK-RV32
+; RUN: llc -mtriple=riscv64 -mattr=+v -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,CHECK-RV64
+; RUN: llc -mtriple=riscv32 -mattr=+v,+experimental-zvbb -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK-ZVBB,CHECK-ZVBB32
+; RUN: llc -mtriple=riscv64 -mattr=+v,+experimental-zvbb -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK-ZVBB,CHECK-ZVBB64
+
+declare <vscale x 1 x i8> @llvm.fshr.nxv1i8(<vscale x 1 x i8>, <vscale x 1 x i8>, <vscale x 1 x i8>)
+declare <vscale x 1 x i8> @llvm.fshl.nxv1i8(<vscale x 1 x i8>, <vscale x 1 x i8>, <vscale x 1 x i8>)
+
+define <vscale x 1 x i8> @vror_vv_nxv1i8(<vscale x 1 x i8> %a, <vscale x 1 x i8> %b) {
+; CHECK-LABEL: vror_vv_nxv1i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e8, mf8, ta, ma
+; CHECK-NEXT: vand.vi v10, v9, 7
+; CHECK-NEXT: vsrl.vv v10, v8, v10
+; CHECK-NEXT: vrsub.vi v9, v9, 0
+; CHECK-NEXT: vand.vi v9, v9, 7
+; CHECK-NEXT: vsll.vv v8, v8, v9
+; CHECK-NEXT: vor.vv v8, v10, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vv_nxv1i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e8, mf8, ta, ma
+; CHECK-ZVBB-NEXT: vror.vv v8, v8, v9
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 1 x i8> @llvm.fshr.nxv1i8(<vscale x 1 x i8> %a, <vscale x 1 x i8> %a, <vscale x 1 x i8> %b)
+ ret <vscale x 1 x i8> %x
+}
+
+define <vscale x 1 x i8> @vror_vx_nxv1i8(<vscale x 1 x i8> %a, i8 %b) {
+; CHECK-LABEL: vror_vx_nxv1i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a1, zero, e8, mf8, ta, ma
+; CHECK-NEXT: vmv.v.x v9, a0
+; CHECK-NEXT: vand.vi v10, v9, 7
+; CHECK-NEXT: vsrl.vv v10, v8, v10
+; CHECK-NEXT: vrsub.vi v9, v9, 0
+; CHECK-NEXT: vand.vi v9, v9, 7
+; CHECK-NEXT: vsll.vv v8, v8, v9
+; CHECK-NEXT: vor.vv v8, v10, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vx_nxv1i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e8, mf8, ta, ma
+; CHECK-ZVBB-NEXT: vror.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 1 x i8> poison, i8 %b, i32 0
+ %b.splat = shufflevector <vscale x 1 x i8> %b.head, <vscale x 1 x i8> poison, <vscale x 1 x i32> zeroinitializer
+ %x = call <vscale x 1 x i8> @llvm.fshr.nxv1i8(<vscale x 1 x i8> %a, <vscale x 1 x i8> %a, <vscale x 1 x i8> %b.splat)
+ ret <vscale x 1 x i8> %x
+}
+
+define <vscale x 1 x i8> @vror_vi_nxv1i8(<vscale x 1 x i8> %a) {
+; CHECK-LABEL: vror_vi_nxv1i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e8, mf8, ta, ma
+; CHECK-NEXT: vsll.vi v9, v8, 7
+; CHECK-NEXT: vsrl.vi v8, v8, 1
+; CHECK-NEXT: vor.vv v8, v8, v9
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_nxv1i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e8, mf8, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 1
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 1 x i8> @llvm.fshr.nxv1i8(<vscale x 1 x i8> %a, <vscale x 1 x i8> %a, <vscale x 1 x i8> shufflevector(<vscale x 1 x i8> insertelement(<vscale x 1 x i8> poison, i8 1, i32 0), <vscale x 1 x i8> poison, <vscale x 1 x i32> zeroinitializer))
+ ret <vscale x 1 x i8> %x
+}
+
+define <vscale x 1 x i8> @vror_vi_rotl_nxv1i8(<vscale x 1 x i8> %a) {
+; CHECK-LABEL: vror_vi_rotl_nxv1i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e8, mf8, ta, ma
+; CHECK-NEXT: vsrl.vi v9, v8, 7
+; CHECK-NEXT: vadd.vv v8, v8, v8
+; CHECK-NEXT: vor.vv v8, v8, v9
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_rotl_nxv1i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e8, mf8, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 63
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 1 x i8> @llvm.fshl.nxv1i8(<vscale x 1 x i8> %a, <vscale x 1 x i8> %a, <vscale x 1 x i8> shufflevector(<vscale x 1 x i8> insertelement(<vscale x 1 x i8> poison, i8 1, i32 0), <vscale x 1 x i8> poison, <vscale x 1 x i32> zeroinitializer))
+ ret <vscale x 1 x i8> %x
+}
+
+declare <vscale x 2 x i8> @llvm.fshr.nxv2i8(<vscale x 2 x i8>, <vscale x 2 x i8>, <vscale x 2 x i8>)
+declare <vscale x 2 x i8> @llvm.fshl.nxv2i8(<vscale x 2 x i8>, <vscale x 2 x i8>, <vscale x 2 x i8>)
+
+define <vscale x 2 x i8> @vror_vv_nxv2i8(<vscale x 2 x i8> %a, <vscale x 2 x i8> %b) {
+; CHECK-LABEL: vror_vv_nxv2i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e8, mf4, ta, ma
+; CHECK-NEXT: vand.vi v10, v9, 7
+; CHECK-NEXT: vsrl.vv v10, v8, v10
+; CHECK-NEXT: vrsub.vi v9, v9, 0
+; CHECK-NEXT: vand.vi v9, v9, 7
+; CHECK-NEXT: vsll.vv v8, v8, v9
+; CHECK-NEXT: vor.vv v8, v10, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vv_nxv2i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e8, mf4, ta, ma
+; CHECK-ZVBB-NEXT: vror.vv v8, v8, v9
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 2 x i8> @llvm.fshr.nxv2i8(<vscale x 2 x i8> %a, <vscale x 2 x i8> %a, <vscale x 2 x i8> %b)
+ ret <vscale x 2 x i8> %x
+}
+
+define <vscale x 2 x i8> @vror_vx_nxv2i8(<vscale x 2 x i8> %a, i8 %b) {
+; CHECK-LABEL: vror_vx_nxv2i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a1, zero, e8, mf4, ta, ma
+; CHECK-NEXT: vmv.v.x v9, a0
+; CHECK-NEXT: vand.vi v10, v9, 7
+; CHECK-NEXT: vsrl.vv v10, v8, v10
+; CHECK-NEXT: vrsub.vi v9, v9, 0
+; CHECK-NEXT: vand.vi v9, v9, 7
+; CHECK-NEXT: vsll.vv v8, v8, v9
+; CHECK-NEXT: vor.vv v8, v10, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vx_nxv2i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e8, mf4, ta, ma
+; CHECK-ZVBB-NEXT: vror.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 2 x i8> poison, i8 %b, i32 0
+ %b.splat = shufflevector <vscale x 2 x i8> %b.head, <vscale x 2 x i8> poison, <vscale x 2 x i32> zeroinitializer
+ %x = call <vscale x 2 x i8> @llvm.fshr.nxv2i8(<vscale x 2 x i8> %a, <vscale x 2 x i8> %a, <vscale x 2 x i8> %b.splat)
+ ret <vscale x 2 x i8> %x
+}
+
+define <vscale x 2 x i8> @vror_vi_nxv2i8(<vscale x 2 x i8> %a) {
+; CHECK-LABEL: vror_vi_nxv2i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e8, mf4, ta, ma
+; CHECK-NEXT: vsll.vi v9, v8, 7
+; CHECK-NEXT: vsrl.vi v8, v8, 1
+; CHECK-NEXT: vor.vv v8, v8, v9
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_nxv2i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e8, mf4, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 1
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 2 x i8> @llvm.fshr.nxv2i8(<vscale x 2 x i8> %a, <vscale x 2 x i8> %a, <vscale x 2 x i8> shufflevector(<vscale x 2 x i8> insertelement(<vscale x 2 x i8> poison, i8 1, i32 0), <vscale x 2 x i8> poison, <vscale x 2 x i32> zeroinitializer))
+ ret <vscale x 2 x i8> %x
+}
+
+define <vscale x 2 x i8> @vror_vi_rotl_nxv2i8(<vscale x 2 x i8> %a) {
+; CHECK-LABEL: vror_vi_rotl_nxv2i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e8, mf4, ta, ma
+; CHECK-NEXT: vsrl.vi v9, v8, 7
+; CHECK-NEXT: vadd.vv v8, v8, v8
+; CHECK-NEXT: vor.vv v8, v8, v9
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_rotl_nxv2i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e8, mf4, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 63
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 2 x i8> @llvm.fshl.nxv2i8(<vscale x 2 x i8> %a, <vscale x 2 x i8> %a, <vscale x 2 x i8> shufflevector(<vscale x 2 x i8> insertelement(<vscale x 2 x i8> poison, i8 1, i32 0), <vscale x 2 x i8> poison, <vscale x 2 x i32> zeroinitializer))
+ ret <vscale x 2 x i8> %x
+}
+
+declare <vscale x 4 x i8> @llvm.fshr.nxv4i8(<vscale x 4 x i8>, <vscale x 4 x i8>, <vscale x 4 x i8>)
+declare <vscale x 4 x i8> @llvm.fshl.nxv4i8(<vscale x 4 x i8>, <vscale x 4 x i8>, <vscale x 4 x i8>)
+
+define <vscale x 4 x i8> @vror_vv_nxv4i8(<vscale x 4 x i8> %a, <vscale x 4 x i8> %b) {
+; CHECK-LABEL: vror_vv_nxv4i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e8, mf2, ta, ma
+; CHECK-NEXT: vand.vi v10, v9, 7
+; CHECK-NEXT: vsrl.vv v10, v8, v10
+; CHECK-NEXT: vrsub.vi v9, v9, 0
+; CHECK-NEXT: vand.vi v9, v9, 7
+; CHECK-NEXT: vsll.vv v8, v8, v9
+; CHECK-NEXT: vor.vv v8, v10, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vv_nxv4i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e8, mf2, ta, ma
+; CHECK-ZVBB-NEXT: vror.vv v8, v8, v9
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 4 x i8> @llvm.fshr.nxv4i8(<vscale x 4 x i8> %a, <vscale x 4 x i8> %a, <vscale x 4 x i8> %b)
+ ret <vscale x 4 x i8> %x
+}
+
+define <vscale x 4 x i8> @vror_vx_nxv4i8(<vscale x 4 x i8> %a, i8 %b) {
+; CHECK-LABEL: vror_vx_nxv4i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a1, zero, e8, mf2, ta, ma
+; CHECK-NEXT: vmv.v.x v9, a0
+; CHECK-NEXT: vand.vi v10, v9, 7
+; CHECK-NEXT: vsrl.vv v10, v8, v10
+; CHECK-NEXT: vrsub.vi v9, v9, 0
+; CHECK-NEXT: vand.vi v9, v9, 7
+; CHECK-NEXT: vsll.vv v8, v8, v9
+; CHECK-NEXT: vor.vv v8, v10, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vx_nxv4i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e8, mf2, ta, ma
+; CHECK-ZVBB-NEXT: vror.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 4 x i8> poison, i8 %b, i32 0
+ %b.splat = shufflevector <vscale x 4 x i8> %b.head, <vscale x 4 x i8> poison, <vscale x 4 x i32> zeroinitializer
+ %x = call <vscale x 4 x i8> @llvm.fshr.nxv4i8(<vscale x 4 x i8> %a, <vscale x 4 x i8> %a, <vscale x 4 x i8> %b.splat)
+ ret <vscale x 4 x i8> %x
+}
+
+define <vscale x 4 x i8> @vror_vi_nxv4i8(<vscale x 4 x i8> %a) {
+; CHECK-LABEL: vror_vi_nxv4i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e8, mf2, ta, ma
+; CHECK-NEXT: vsll.vi v9, v8, 7
+; CHECK-NEXT: vsrl.vi v8, v8, 1
+; CHECK-NEXT: vor.vv v8, v8, v9
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_nxv4i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e8, mf2, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 1
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 4 x i8> @llvm.fshr.nxv4i8(<vscale x 4 x i8> %a, <vscale x 4 x i8> %a, <vscale x 4 x i8> shufflevector(<vscale x 4 x i8> insertelement(<vscale x 4 x i8> poison, i8 1, i32 0), <vscale x 4 x i8> poison, <vscale x 4 x i32> zeroinitializer))
+ ret <vscale x 4 x i8> %x
+}
+
+define <vscale x 4 x i8> @vror_vi_rotl_nxv4i8(<vscale x 4 x i8> %a) {
+; CHECK-LABEL: vror_vi_rotl_nxv4i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e8, mf2, ta, ma
+; CHECK-NEXT: vsrl.vi v9, v8, 7
+; CHECK-NEXT: vadd.vv v8, v8, v8
+; CHECK-NEXT: vor.vv v8, v8, v9
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_rotl_nxv4i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e8, mf2, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 63
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 4 x i8> @llvm.fshl.nxv4i8(<vscale x 4 x i8> %a, <vscale x 4 x i8> %a, <vscale x 4 x i8> shufflevector(<vscale x 4 x i8> insertelement(<vscale x 4 x i8> poison, i8 1, i32 0), <vscale x 4 x i8> poison, <vscale x 4 x i32> zeroinitializer))
+ ret <vscale x 4 x i8> %x
+}
+
+declare <vscale x 8 x i8> @llvm.fshr.nxv8i8(<vscale x 8 x i8>, <vscale x 8 x i8>, <vscale x 8 x i8>)
+declare <vscale x 8 x i8> @llvm.fshl.nxv8i8(<vscale x 8 x i8>, <vscale x 8 x i8>, <vscale x 8 x i8>)
+
+define <vscale x 8 x i8> @vror_vv_nxv8i8(<vscale x 8 x i8> %a, <vscale x 8 x i8> %b) {
+; CHECK-LABEL: vror_vv_nxv8i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e8, m1, ta, ma
+; CHECK-NEXT: vand.vi v10, v9, 7
+; CHECK-NEXT: vsrl.vv v10, v8, v10
+; CHECK-NEXT: vrsub.vi v9, v9, 0
+; CHECK-NEXT: vand.vi v9, v9, 7
+; CHECK-NEXT: vsll.vv v8, v8, v9
+; CHECK-NEXT: vor.vv v8, v10, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vv_nxv8i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e8, m1, ta, ma
+; CHECK-ZVBB-NEXT: vror.vv v8, v8, v9
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 8 x i8> @llvm.fshr.nxv8i8(<vscale x 8 x i8> %a, <vscale x 8 x i8> %a, <vscale x 8 x i8> %b)
+ ret <vscale x 8 x i8> %x
+}
+
+define <vscale x 8 x i8> @vror_vx_nxv8i8(<vscale x 8 x i8> %a, i8 %b) {
+; CHECK-LABEL: vror_vx_nxv8i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a1, zero, e8, m1, ta, ma
+; CHECK-NEXT: vmv.v.x v9, a0
+; CHECK-NEXT: vand.vi v10, v9, 7
+; CHECK-NEXT: vsrl.vv v10, v8, v10
+; CHECK-NEXT: vrsub.vi v9, v9, 0
+; CHECK-NEXT: vand.vi v9, v9, 7
+; CHECK-NEXT: vsll.vv v8, v8, v9
+; CHECK-NEXT: vor.vv v8, v10, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vx_nxv8i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e8, m1, ta, ma
+; CHECK-ZVBB-NEXT: vror.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 8 x i8> poison, i8 %b, i32 0
+ %b.splat = shufflevector <vscale x 8 x i8> %b.head, <vscale x 8 x i8> poison, <vscale x 8 x i32> zeroinitializer
+ %x = call <vscale x 8 x i8> @llvm.fshr.nxv8i8(<vscale x 8 x i8> %a, <vscale x 8 x i8> %a, <vscale x 8 x i8> %b.splat)
+ ret <vscale x 8 x i8> %x
+}
+
+define <vscale x 8 x i8> @vror_vi_nxv8i8(<vscale x 8 x i8> %a) {
+; CHECK-LABEL: vror_vi_nxv8i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e8, m1, ta, ma
+; CHECK-NEXT: vsll.vi v9, v8, 7
+; CHECK-NEXT: vsrl.vi v8, v8, 1
+; CHECK-NEXT: vor.vv v8, v8, v9
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_nxv8i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e8, m1, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 1
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 8 x i8> @llvm.fshr.nxv8i8(<vscale x 8 x i8> %a, <vscale x 8 x i8> %a, <vscale x 8 x i8> shufflevector(<vscale x 8 x i8> insertelement(<vscale x 8 x i8> poison, i8 1, i32 0), <vscale x 8 x i8> poison, <vscale x 8 x i32> zeroinitializer))
+ ret <vscale x 8 x i8> %x
+}
+
+define <vscale x 8 x i8> @vror_vi_rotl_nxv8i8(<vscale x 8 x i8> %a) {
+; CHECK-LABEL: vror_vi_rotl_nxv8i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e8, m1, ta, ma
+; CHECK-NEXT: vsrl.vi v9, v8, 7
+; CHECK-NEXT: vadd.vv v8, v8, v8
+; CHECK-NEXT: vor.vv v8, v8, v9
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_rotl_nxv8i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e8, m1, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 63
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 8 x i8> @llvm.fshl.nxv8i8(<vscale x 8 x i8> %a, <vscale x 8 x i8> %a, <vscale x 8 x i8> shufflevector(<vscale x 8 x i8> insertelement(<vscale x 8 x i8> poison, i8 1, i32 0), <vscale x 8 x i8> poison, <vscale x 8 x i32> zeroinitializer))
+ ret <vscale x 8 x i8> %x
+}
+
+declare <vscale x 16 x i8> @llvm.fshr.nxv16i8(<vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>)
+declare <vscale x 16 x i8> @llvm.fshl.nxv16i8(<vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>)
+
+define <vscale x 16 x i8> @vror_vv_nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %b) {
+; CHECK-LABEL: vror_vv_nxv16i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e8, m2, ta, ma
+; CHECK-NEXT: vand.vi v12, v10, 7
+; CHECK-NEXT: vsrl.vv v12, v8, v12
+; CHECK-NEXT: vrsub.vi v10, v10, 0
+; CHECK-NEXT: vand.vi v10, v10, 7
+; CHECK-NEXT: vsll.vv v8, v8, v10
+; CHECK-NEXT: vor.vv v8, v12, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vv_nxv16i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e8, m2, ta, ma
+; CHECK-ZVBB-NEXT: vror.vv v8, v8, v10
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 16 x i8> @llvm.fshr.nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %a, <vscale x 16 x i8> %b)
+ ret <vscale x 16 x i8> %x
+}
+
+define <vscale x 16 x i8> @vror_vx_nxv16i8(<vscale x 16 x i8> %a, i8 %b) {
+; CHECK-LABEL: vror_vx_nxv16i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a1, zero, e8, m2, ta, ma
+; CHECK-NEXT: vmv.v.x v10, a0
+; CHECK-NEXT: vand.vi v12, v10, 7
+; CHECK-NEXT: vsrl.vv v12, v8, v12
+; CHECK-NEXT: vrsub.vi v10, v10, 0
+; CHECK-NEXT: vand.vi v10, v10, 7
+; CHECK-NEXT: vsll.vv v8, v8, v10
+; CHECK-NEXT: vor.vv v8, v12, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vx_nxv16i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e8, m2, ta, ma
+; CHECK-ZVBB-NEXT: vror.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 16 x i8> poison, i8 %b, i32 0
+ %b.splat = shufflevector <vscale x 16 x i8> %b.head, <vscale x 16 x i8> poison, <vscale x 16 x i32> zeroinitializer
+ %x = call <vscale x 16 x i8> @llvm.fshr.nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %a, <vscale x 16 x i8> %b.splat)
+ ret <vscale x 16 x i8> %x
+}
+
+define <vscale x 16 x i8> @vror_vi_nxv16i8(<vscale x 16 x i8> %a) {
+; CHECK-LABEL: vror_vi_nxv16i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e8, m2, ta, ma
+; CHECK-NEXT: vsll.vi v10, v8, 7
+; CHECK-NEXT: vsrl.vi v8, v8, 1
+; CHECK-NEXT: vor.vv v8, v8, v10
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_nxv16i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e8, m2, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 1
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 16 x i8> @llvm.fshr.nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %a, <vscale x 16 x i8> shufflevector(<vscale x 16 x i8> insertelement(<vscale x 16 x i8> poison, i8 1, i32 0), <vscale x 16 x i8> poison, <vscale x 16 x i32> zeroinitializer))
+ ret <vscale x 16 x i8> %x
+}
+
+define <vscale x 16 x i8> @vror_vi_rotl_nxv16i8(<vscale x 16 x i8> %a) {
+; CHECK-LABEL: vror_vi_rotl_nxv16i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e8, m2, ta, ma
+; CHECK-NEXT: vsrl.vi v10, v8, 7
+; CHECK-NEXT: vadd.vv v8, v8, v8
+; CHECK-NEXT: vor.vv v8, v8, v10
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_rotl_nxv16i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e8, m2, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 63
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 16 x i8> @llvm.fshl.nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %a, <vscale x 16 x i8> shufflevector(<vscale x 16 x i8> insertelement(<vscale x 16 x i8> poison, i8 1, i32 0), <vscale x 16 x i8> poison, <vscale x 16 x i32> zeroinitializer))
+ ret <vscale x 16 x i8> %x
+}
+
+declare <vscale x 32 x i8> @llvm.fshr.nxv32i8(<vscale x 32 x i8>, <vscale x 32 x i8>, <vscale x 32 x i8>)
+declare <vscale x 32 x i8> @llvm.fshl.nxv32i8(<vscale x 32 x i8>, <vscale x 32 x i8>, <vscale x 32 x i8>)
+
+define <vscale x 32 x i8> @vror_vv_nxv32i8(<vscale x 32 x i8> %a, <vscale x 32 x i8> %b) {
+; CHECK-LABEL: vror_vv_nxv32i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e8, m4, ta, ma
+; CHECK-NEXT: vand.vi v16, v12, 7
+; CHECK-NEXT: vsrl.vv v16, v8, v16
+; CHECK-NEXT: vrsub.vi v12, v12, 0
+; CHECK-NEXT: vand.vi v12, v12, 7
+; CHECK-NEXT: vsll.vv v8, v8, v12
+; CHECK-NEXT: vor.vv v8, v16, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vv_nxv32i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e8, m4, ta, ma
+; CHECK-ZVBB-NEXT: vror.vv v8, v8, v12
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 32 x i8> @llvm.fshr.nxv32i8(<vscale x 32 x i8> %a, <vscale x 32 x i8> %a, <vscale x 32 x i8> %b)
+ ret <vscale x 32 x i8> %x
+}
+
+define <vscale x 32 x i8> @vror_vx_nxv32i8(<vscale x 32 x i8> %a, i8 %b) {
+; CHECK-LABEL: vror_vx_nxv32i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a1, zero, e8, m4, ta, ma
+; CHECK-NEXT: vmv.v.x v12, a0
+; CHECK-NEXT: vand.vi v16, v12, 7
+; CHECK-NEXT: vsrl.vv v16, v8, v16
+; CHECK-NEXT: vrsub.vi v12, v12, 0
+; CHECK-NEXT: vand.vi v12, v12, 7
+; CHECK-NEXT: vsll.vv v8, v8, v12
+; CHECK-NEXT: vor.vv v8, v16, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vx_nxv32i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e8, m4, ta, ma
+; CHECK-ZVBB-NEXT: vror.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 32 x i8> poison, i8 %b, i32 0
+ %b.splat = shufflevector <vscale x 32 x i8> %b.head, <vscale x 32 x i8> poison, <vscale x 32 x i32> zeroinitializer
+ %x = call <vscale x 32 x i8> @llvm.fshr.nxv32i8(<vscale x 32 x i8> %a, <vscale x 32 x i8> %a, <vscale x 32 x i8> %b.splat)
+ ret <vscale x 32 x i8> %x
+}
+
+define <vscale x 32 x i8> @vror_vi_nxv32i8(<vscale x 32 x i8> %a) {
+; CHECK-LABEL: vror_vi_nxv32i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e8, m4, ta, ma
+; CHECK-NEXT: vsll.vi v12, v8, 7
+; CHECK-NEXT: vsrl.vi v8, v8, 1
+; CHECK-NEXT: vor.vv v8, v8, v12
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_nxv32i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e8, m4, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 1
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 32 x i8> @llvm.fshr.nxv32i8(<vscale x 32 x i8> %a, <vscale x 32 x i8> %a, <vscale x 32 x i8> shufflevector(<vscale x 32 x i8> insertelement(<vscale x 32 x i8> poison, i8 1, i32 0), <vscale x 32 x i8> poison, <vscale x 32 x i32> zeroinitializer))
+ ret <vscale x 32 x i8> %x
+}
+
+define <vscale x 32 x i8> @vror_vi_rotl_nxv32i8(<vscale x 32 x i8> %a) {
+; CHECK-LABEL: vror_vi_rotl_nxv32i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e8, m4, ta, ma
+; CHECK-NEXT: vsrl.vi v12, v8, 7
+; CHECK-NEXT: vadd.vv v8, v8, v8
+; CHECK-NEXT: vor.vv v8, v8, v12
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_rotl_nxv32i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e8, m4, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 63
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 32 x i8> @llvm.fshl.nxv32i8(<vscale x 32 x i8> %a, <vscale x 32 x i8> %a, <vscale x 32 x i8> shufflevector(<vscale x 32 x i8> insertelement(<vscale x 32 x i8> poison, i8 1, i32 0), <vscale x 32 x i8> poison, <vscale x 32 x i32> zeroinitializer))
+ ret <vscale x 32 x i8> %x
+}
+
+declare <vscale x 64 x i8> @llvm.fshr.nxv64i8(<vscale x 64 x i8>, <vscale x 64 x i8>, <vscale x 64 x i8>)
+declare <vscale x 64 x i8> @llvm.fshl.nxv64i8(<vscale x 64 x i8>, <vscale x 64 x i8>, <vscale x 64 x i8>)
+
+define <vscale x 64 x i8> @vror_vv_nxv64i8(<vscale x 64 x i8> %a, <vscale x 64 x i8> %b) {
+; CHECK-LABEL: vror_vv_nxv64i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e8, m8, ta, ma
+; CHECK-NEXT: vand.vi v24, v16, 7
+; CHECK-NEXT: vsrl.vv v24, v8, v24
+; CHECK-NEXT: vrsub.vi v16, v16, 0
+; CHECK-NEXT: vand.vi v16, v16, 7
+; CHECK-NEXT: vsll.vv v8, v8, v16
+; CHECK-NEXT: vor.vv v8, v24, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vv_nxv64i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e8, m8, ta, ma
+; CHECK-ZVBB-NEXT: vror.vv v8, v8, v16
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 64 x i8> @llvm.fshr.nxv64i8(<vscale x 64 x i8> %a, <vscale x 64 x i8> %a, <vscale x 64 x i8> %b)
+ ret <vscale x 64 x i8> %x
+}
+
+define <vscale x 64 x i8> @vror_vx_nxv64i8(<vscale x 64 x i8> %a, i8 %b) {
+; CHECK-LABEL: vror_vx_nxv64i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a1, zero, e8, m8, ta, ma
+; CHECK-NEXT: vmv.v.x v16, a0
+; CHECK-NEXT: vand.vi v24, v16, 7
+; CHECK-NEXT: vsrl.vv v24, v8, v24
+; CHECK-NEXT: vrsub.vi v16, v16, 0
+; CHECK-NEXT: vand.vi v16, v16, 7
+; CHECK-NEXT: vsll.vv v8, v8, v16
+; CHECK-NEXT: vor.vv v8, v24, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vx_nxv64i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e8, m8, ta, ma
+; CHECK-ZVBB-NEXT: vror.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 64 x i8> poison, i8 %b, i32 0
+ %b.splat = shufflevector <vscale x 64 x i8> %b.head, <vscale x 64 x i8> poison, <vscale x 64 x i32> zeroinitializer
+ %x = call <vscale x 64 x i8> @llvm.fshr.nxv64i8(<vscale x 64 x i8> %a, <vscale x 64 x i8> %a, <vscale x 64 x i8> %b.splat)
+ ret <vscale x 64 x i8> %x
+}
+
+define <vscale x 64 x i8> @vror_vi_nxv64i8(<vscale x 64 x i8> %a) {
+; CHECK-LABEL: vror_vi_nxv64i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e8, m8, ta, ma
+; CHECK-NEXT: vsll.vi v16, v8, 7
+; CHECK-NEXT: vsrl.vi v8, v8, 1
+; CHECK-NEXT: vor.vv v8, v8, v16
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_nxv64i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e8, m8, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 1
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 64 x i8> @llvm.fshr.nxv64i8(<vscale x 64 x i8> %a, <vscale x 64 x i8> %a, <vscale x 64 x i8> shufflevector(<vscale x 64 x i8> insertelement(<vscale x 64 x i8> poison, i8 1, i32 0), <vscale x 64 x i8> poison, <vscale x 64 x i32> zeroinitializer))
+ ret <vscale x 64 x i8> %x
+}
+
+define <vscale x 64 x i8> @vror_vi_rotl_nxv64i8(<vscale x 64 x i8> %a) {
+; CHECK-LABEL: vror_vi_rotl_nxv64i8:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e8, m8, ta, ma
+; CHECK-NEXT: vsrl.vi v16, v8, 7
+; CHECK-NEXT: vadd.vv v8, v8, v8
+; CHECK-NEXT: vor.vv v8, v8, v16
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_rotl_nxv64i8:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e8, m8, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 63
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 64 x i8> @llvm.fshl.nxv64i8(<vscale x 64 x i8> %a, <vscale x 64 x i8> %a, <vscale x 64 x i8> shufflevector(<vscale x 64 x i8> insertelement(<vscale x 64 x i8> poison, i8 1, i32 0), <vscale x 64 x i8> poison, <vscale x 64 x i32> zeroinitializer))
+ ret <vscale x 64 x i8> %x
+}
+
+declare <vscale x 1 x i16> @llvm.fshr.nxv1i16(<vscale x 1 x i16>, <vscale x 1 x i16>, <vscale x 1 x i16>)
+declare <vscale x 1 x i16> @llvm.fshl.nxv1i16(<vscale x 1 x i16>, <vscale x 1 x i16>, <vscale x 1 x i16>)
+
+define <vscale x 1 x i16> @vror_vv_nxv1i16(<vscale x 1 x i16> %a, <vscale x 1 x i16> %b) {
+; CHECK-LABEL: vror_vv_nxv1i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e16, mf4, ta, ma
+; CHECK-NEXT: vand.vi v10, v9, 15
+; CHECK-NEXT: vsrl.vv v10, v8, v10
+; CHECK-NEXT: vrsub.vi v9, v9, 0
+; CHECK-NEXT: vand.vi v9, v9, 15
+; CHECK-NEXT: vsll.vv v8, v8, v9
+; CHECK-NEXT: vor.vv v8, v10, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vv_nxv1i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e16, mf4, ta, ma
+; CHECK-ZVBB-NEXT: vror.vv v8, v8, v9
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 1 x i16> @llvm.fshr.nxv1i16(<vscale x 1 x i16> %a, <vscale x 1 x i16> %a, <vscale x 1 x i16> %b)
+ ret <vscale x 1 x i16> %x
+}
+
+define <vscale x 1 x i16> @vror_vx_nxv1i16(<vscale x 1 x i16> %a, i16 %b) {
+; CHECK-LABEL: vror_vx_nxv1i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a1, zero, e16, mf4, ta, ma
+; CHECK-NEXT: vmv.v.x v9, a0
+; CHECK-NEXT: vand.vi v10, v9, 15
+; CHECK-NEXT: vsrl.vv v10, v8, v10
+; CHECK-NEXT: vrsub.vi v9, v9, 0
+; CHECK-NEXT: vand.vi v9, v9, 15
+; CHECK-NEXT: vsll.vv v8, v8, v9
+; CHECK-NEXT: vor.vv v8, v10, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vx_nxv1i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e16, mf4, ta, ma
+; CHECK-ZVBB-NEXT: vror.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 1 x i16> poison, i16 %b, i32 0
+ %b.splat = shufflevector <vscale x 1 x i16> %b.head, <vscale x 1 x i16> poison, <vscale x 1 x i32> zeroinitializer
+ %x = call <vscale x 1 x i16> @llvm.fshr.nxv1i16(<vscale x 1 x i16> %a, <vscale x 1 x i16> %a, <vscale x 1 x i16> %b.splat)
+ ret <vscale x 1 x i16> %x
+}
+
+define <vscale x 1 x i16> @vror_vi_nxv1i16(<vscale x 1 x i16> %a) {
+; CHECK-LABEL: vror_vi_nxv1i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e16, mf4, ta, ma
+; CHECK-NEXT: vsll.vi v9, v8, 15
+; CHECK-NEXT: vsrl.vi v8, v8, 1
+; CHECK-NEXT: vor.vv v8, v8, v9
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_nxv1i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e16, mf4, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 1
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 1 x i16> @llvm.fshr.nxv1i16(<vscale x 1 x i16> %a, <vscale x 1 x i16> %a, <vscale x 1 x i16> shufflevector(<vscale x 1 x i16> insertelement(<vscale x 1 x i16> poison, i16 1, i32 0), <vscale x 1 x i16> poison, <vscale x 1 x i32> zeroinitializer))
+ ret <vscale x 1 x i16> %x
+}
+
+define <vscale x 1 x i16> @vror_vi_rotl_nxv1i16(<vscale x 1 x i16> %a) {
+; CHECK-LABEL: vror_vi_rotl_nxv1i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e16, mf4, ta, ma
+; CHECK-NEXT: vsrl.vi v9, v8, 15
+; CHECK-NEXT: vadd.vv v8, v8, v8
+; CHECK-NEXT: vor.vv v8, v8, v9
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_rotl_nxv1i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e16, mf4, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 63
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 1 x i16> @llvm.fshl.nxv1i16(<vscale x 1 x i16> %a, <vscale x 1 x i16> %a, <vscale x 1 x i16> shufflevector(<vscale x 1 x i16> insertelement(<vscale x 1 x i16> poison, i16 1, i32 0), <vscale x 1 x i16> poison, <vscale x 1 x i32> zeroinitializer))
+ ret <vscale x 1 x i16> %x
+}
+
+declare <vscale x 2 x i16> @llvm.fshr.nxv2i16(<vscale x 2 x i16>, <vscale x 2 x i16>, <vscale x 2 x i16>)
+declare <vscale x 2 x i16> @llvm.fshl.nxv2i16(<vscale x 2 x i16>, <vscale x 2 x i16>, <vscale x 2 x i16>)
+
+define <vscale x 2 x i16> @vror_vv_nxv2i16(<vscale x 2 x i16> %a, <vscale x 2 x i16> %b) {
+; CHECK-LABEL: vror_vv_nxv2i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e16, mf2, ta, ma
+; CHECK-NEXT: vand.vi v10, v9, 15
+; CHECK-NEXT: vsrl.vv v10, v8, v10
+; CHECK-NEXT: vrsub.vi v9, v9, 0
+; CHECK-NEXT: vand.vi v9, v9, 15
+; CHECK-NEXT: vsll.vv v8, v8, v9
+; CHECK-NEXT: vor.vv v8, v10, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vv_nxv2i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e16, mf2, ta, ma
+; CHECK-ZVBB-NEXT: vror.vv v8, v8, v9
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 2 x i16> @llvm.fshr.nxv2i16(<vscale x 2 x i16> %a, <vscale x 2 x i16> %a, <vscale x 2 x i16> %b)
+ ret <vscale x 2 x i16> %x
+}
+
+define <vscale x 2 x i16> @vror_vx_nxv2i16(<vscale x 2 x i16> %a, i16 %b) {
+; CHECK-LABEL: vror_vx_nxv2i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a1, zero, e16, mf2, ta, ma
+; CHECK-NEXT: vmv.v.x v9, a0
+; CHECK-NEXT: vand.vi v10, v9, 15
+; CHECK-NEXT: vsrl.vv v10, v8, v10
+; CHECK-NEXT: vrsub.vi v9, v9, 0
+; CHECK-NEXT: vand.vi v9, v9, 15
+; CHECK-NEXT: vsll.vv v8, v8, v9
+; CHECK-NEXT: vor.vv v8, v10, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vx_nxv2i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e16, mf2, ta, ma
+; CHECK-ZVBB-NEXT: vror.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 2 x i16> poison, i16 %b, i32 0
+ %b.splat = shufflevector <vscale x 2 x i16> %b.head, <vscale x 2 x i16> poison, <vscale x 2 x i32> zeroinitializer
+ %x = call <vscale x 2 x i16> @llvm.fshr.nxv2i16(<vscale x 2 x i16> %a, <vscale x 2 x i16> %a, <vscale x 2 x i16> %b.splat)
+ ret <vscale x 2 x i16> %x
+}
+
+define <vscale x 2 x i16> @vror_vi_nxv2i16(<vscale x 2 x i16> %a) {
+; CHECK-LABEL: vror_vi_nxv2i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e16, mf2, ta, ma
+; CHECK-NEXT: vsll.vi v9, v8, 15
+; CHECK-NEXT: vsrl.vi v8, v8, 1
+; CHECK-NEXT: vor.vv v8, v8, v9
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_nxv2i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e16, mf2, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 1
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 2 x i16> @llvm.fshr.nxv2i16(<vscale x 2 x i16> %a, <vscale x 2 x i16> %a, <vscale x 2 x i16> shufflevector(<vscale x 2 x i16> insertelement(<vscale x 2 x i16> poison, i16 1, i32 0), <vscale x 2 x i16> poison, <vscale x 2 x i32> zeroinitializer))
+ ret <vscale x 2 x i16> %x
+}
+
+define <vscale x 2 x i16> @vror_vi_rotl_nxv2i16(<vscale x 2 x i16> %a) {
+; CHECK-LABEL: vror_vi_rotl_nxv2i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e16, mf2, ta, ma
+; CHECK-NEXT: vsrl.vi v9, v8, 15
+; CHECK-NEXT: vadd.vv v8, v8, v8
+; CHECK-NEXT: vor.vv v8, v8, v9
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_rotl_nxv2i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e16, mf2, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 63
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 2 x i16> @llvm.fshl.nxv2i16(<vscale x 2 x i16> %a, <vscale x 2 x i16> %a, <vscale x 2 x i16> shufflevector(<vscale x 2 x i16> insertelement(<vscale x 2 x i16> poison, i16 1, i32 0), <vscale x 2 x i16> poison, <vscale x 2 x i32> zeroinitializer))
+ ret <vscale x 2 x i16> %x
+}
+
+declare <vscale x 4 x i16> @llvm.fshr.nxv4i16(<vscale x 4 x i16>, <vscale x 4 x i16>, <vscale x 4 x i16>)
+declare <vscale x 4 x i16> @llvm.fshl.nxv4i16(<vscale x 4 x i16>, <vscale x 4 x i16>, <vscale x 4 x i16>)
+
+define <vscale x 4 x i16> @vror_vv_nxv4i16(<vscale x 4 x i16> %a, <vscale x 4 x i16> %b) {
+; CHECK-LABEL: vror_vv_nxv4i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e16, m1, ta, ma
+; CHECK-NEXT: vand.vi v10, v9, 15
+; CHECK-NEXT: vsrl.vv v10, v8, v10
+; CHECK-NEXT: vrsub.vi v9, v9, 0
+; CHECK-NEXT: vand.vi v9, v9, 15
+; CHECK-NEXT: vsll.vv v8, v8, v9
+; CHECK-NEXT: vor.vv v8, v10, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vv_nxv4i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e16, m1, ta, ma
+; CHECK-ZVBB-NEXT: vror.vv v8, v8, v9
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 4 x i16> @llvm.fshr.nxv4i16(<vscale x 4 x i16> %a, <vscale x 4 x i16> %a, <vscale x 4 x i16> %b)
+ ret <vscale x 4 x i16> %x
+}
+
+define <vscale x 4 x i16> @vror_vx_nxv4i16(<vscale x 4 x i16> %a, i16 %b) {
+; CHECK-LABEL: vror_vx_nxv4i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a1, zero, e16, m1, ta, ma
+; CHECK-NEXT: vmv.v.x v9, a0
+; CHECK-NEXT: vand.vi v10, v9, 15
+; CHECK-NEXT: vsrl.vv v10, v8, v10
+; CHECK-NEXT: vrsub.vi v9, v9, 0
+; CHECK-NEXT: vand.vi v9, v9, 15
+; CHECK-NEXT: vsll.vv v8, v8, v9
+; CHECK-NEXT: vor.vv v8, v10, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vx_nxv4i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e16, m1, ta, ma
+; CHECK-ZVBB-NEXT: vror.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 4 x i16> poison, i16 %b, i32 0
+ %b.splat = shufflevector <vscale x 4 x i16> %b.head, <vscale x 4 x i16> poison, <vscale x 4 x i32> zeroinitializer
+ %x = call <vscale x 4 x i16> @llvm.fshr.nxv4i16(<vscale x 4 x i16> %a, <vscale x 4 x i16> %a, <vscale x 4 x i16> %b.splat)
+ ret <vscale x 4 x i16> %x
+}
+
+define <vscale x 4 x i16> @vror_vi_nxv4i16(<vscale x 4 x i16> %a) {
+; CHECK-LABEL: vror_vi_nxv4i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e16, m1, ta, ma
+; CHECK-NEXT: vsll.vi v9, v8, 15
+; CHECK-NEXT: vsrl.vi v8, v8, 1
+; CHECK-NEXT: vor.vv v8, v8, v9
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_nxv4i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e16, m1, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 1
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 4 x i16> @llvm.fshr.nxv4i16(<vscale x 4 x i16> %a, <vscale x 4 x i16> %a, <vscale x 4 x i16> shufflevector(<vscale x 4 x i16> insertelement(<vscale x 4 x i16> poison, i16 1, i32 0), <vscale x 4 x i16> poison, <vscale x 4 x i32> zeroinitializer))
+ ret <vscale x 4 x i16> %x
+}
+
+define <vscale x 4 x i16> @vror_vi_rotl_nxv4i16(<vscale x 4 x i16> %a) {
+; CHECK-LABEL: vror_vi_rotl_nxv4i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e16, m1, ta, ma
+; CHECK-NEXT: vsrl.vi v9, v8, 15
+; CHECK-NEXT: vadd.vv v8, v8, v8
+; CHECK-NEXT: vor.vv v8, v8, v9
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_rotl_nxv4i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e16, m1, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 63
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 4 x i16> @llvm.fshl.nxv4i16(<vscale x 4 x i16> %a, <vscale x 4 x i16> %a, <vscale x 4 x i16> shufflevector(<vscale x 4 x i16> insertelement(<vscale x 4 x i16> poison, i16 1, i32 0), <vscale x 4 x i16> poison, <vscale x 4 x i32> zeroinitializer))
+ ret <vscale x 4 x i16> %x
+}
+
+declare <vscale x 8 x i16> @llvm.fshr.nxv8i16(<vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>)
+declare <vscale x 8 x i16> @llvm.fshl.nxv8i16(<vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>)
+
+define <vscale x 8 x i16> @vror_vv_nxv8i16(<vscale x 8 x i16> %a, <vscale x 8 x i16> %b) {
+; CHECK-LABEL: vror_vv_nxv8i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e16, m2, ta, ma
+; CHECK-NEXT: vand.vi v12, v10, 15
+; CHECK-NEXT: vsrl.vv v12, v8, v12
+; CHECK-NEXT: vrsub.vi v10, v10, 0
+; CHECK-NEXT: vand.vi v10, v10, 15
+; CHECK-NEXT: vsll.vv v8, v8, v10
+; CHECK-NEXT: vor.vv v8, v12, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vv_nxv8i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e16, m2, ta, ma
+; CHECK-ZVBB-NEXT: vror.vv v8, v8, v10
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 8 x i16> @llvm.fshr.nxv8i16(<vscale x 8 x i16> %a, <vscale x 8 x i16> %a, <vscale x 8 x i16> %b)
+ ret <vscale x 8 x i16> %x
+}
+
+define <vscale x 8 x i16> @vror_vx_nxv8i16(<vscale x 8 x i16> %a, i16 %b) {
+; CHECK-LABEL: vror_vx_nxv8i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a1, zero, e16, m2, ta, ma
+; CHECK-NEXT: vmv.v.x v10, a0
+; CHECK-NEXT: vand.vi v12, v10, 15
+; CHECK-NEXT: vsrl.vv v12, v8, v12
+; CHECK-NEXT: vrsub.vi v10, v10, 0
+; CHECK-NEXT: vand.vi v10, v10, 15
+; CHECK-NEXT: vsll.vv v8, v8, v10
+; CHECK-NEXT: vor.vv v8, v12, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vx_nxv8i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e16, m2, ta, ma
+; CHECK-ZVBB-NEXT: vror.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 8 x i16> poison, i16 %b, i32 0
+ %b.splat = shufflevector <vscale x 8 x i16> %b.head, <vscale x 8 x i16> poison, <vscale x 8 x i32> zeroinitializer
+ %x = call <vscale x 8 x i16> @llvm.fshr.nxv8i16(<vscale x 8 x i16> %a, <vscale x 8 x i16> %a, <vscale x 8 x i16> %b.splat)
+ ret <vscale x 8 x i16> %x
+}
+
+define <vscale x 8 x i16> @vror_vi_nxv8i16(<vscale x 8 x i16> %a) {
+; CHECK-LABEL: vror_vi_nxv8i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e16, m2, ta, ma
+; CHECK-NEXT: vsll.vi v10, v8, 15
+; CHECK-NEXT: vsrl.vi v8, v8, 1
+; CHECK-NEXT: vor.vv v8, v8, v10
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_nxv8i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e16, m2, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 1
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 8 x i16> @llvm.fshr.nxv8i16(<vscale x 8 x i16> %a, <vscale x 8 x i16> %a, <vscale x 8 x i16> shufflevector(<vscale x 8 x i16> insertelement(<vscale x 8 x i16> poison, i16 1, i32 0), <vscale x 8 x i16> poison, <vscale x 8 x i32> zeroinitializer))
+ ret <vscale x 8 x i16> %x
+}
+
+define <vscale x 8 x i16> @vror_vi_rotl_nxv8i16(<vscale x 8 x i16> %a) {
+; CHECK-LABEL: vror_vi_rotl_nxv8i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e16, m2, ta, ma
+; CHECK-NEXT: vsrl.vi v10, v8, 15
+; CHECK-NEXT: vadd.vv v8, v8, v8
+; CHECK-NEXT: vor.vv v8, v8, v10
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_rotl_nxv8i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e16, m2, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 63
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 8 x i16> @llvm.fshl.nxv8i16(<vscale x 8 x i16> %a, <vscale x 8 x i16> %a, <vscale x 8 x i16> shufflevector(<vscale x 8 x i16> insertelement(<vscale x 8 x i16> poison, i16 1, i32 0), <vscale x 8 x i16> poison, <vscale x 8 x i32> zeroinitializer))
+ ret <vscale x 8 x i16> %x
+}
+
+declare <vscale x 16 x i16> @llvm.fshr.nxv16i16(<vscale x 16 x i16>, <vscale x 16 x i16>, <vscale x 16 x i16>)
+declare <vscale x 16 x i16> @llvm.fshl.nxv16i16(<vscale x 16 x i16>, <vscale x 16 x i16>, <vscale x 16 x i16>)
+
+define <vscale x 16 x i16> @vror_vv_nxv16i16(<vscale x 16 x i16> %a, <vscale x 16 x i16> %b) {
+; CHECK-LABEL: vror_vv_nxv16i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e16, m4, ta, ma
+; CHECK-NEXT: vand.vi v16, v12, 15
+; CHECK-NEXT: vsrl.vv v16, v8, v16
+; CHECK-NEXT: vrsub.vi v12, v12, 0
+; CHECK-NEXT: vand.vi v12, v12, 15
+; CHECK-NEXT: vsll.vv v8, v8, v12
+; CHECK-NEXT: vor.vv v8, v16, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vv_nxv16i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e16, m4, ta, ma
+; CHECK-ZVBB-NEXT: vror.vv v8, v8, v12
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 16 x i16> @llvm.fshr.nxv16i16(<vscale x 16 x i16> %a, <vscale x 16 x i16> %a, <vscale x 16 x i16> %b)
+ ret <vscale x 16 x i16> %x
+}
+
+define <vscale x 16 x i16> @vror_vx_nxv16i16(<vscale x 16 x i16> %a, i16 %b) {
+; CHECK-LABEL: vror_vx_nxv16i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a1, zero, e16, m4, ta, ma
+; CHECK-NEXT: vmv.v.x v12, a0
+; CHECK-NEXT: vand.vi v16, v12, 15
+; CHECK-NEXT: vsrl.vv v16, v8, v16
+; CHECK-NEXT: vrsub.vi v12, v12, 0
+; CHECK-NEXT: vand.vi v12, v12, 15
+; CHECK-NEXT: vsll.vv v8, v8, v12
+; CHECK-NEXT: vor.vv v8, v16, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vx_nxv16i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e16, m4, ta, ma
+; CHECK-ZVBB-NEXT: vror.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 16 x i16> poison, i16 %b, i32 0
+ %b.splat = shufflevector <vscale x 16 x i16> %b.head, <vscale x 16 x i16> poison, <vscale x 16 x i32> zeroinitializer
+ %x = call <vscale x 16 x i16> @llvm.fshr.nxv16i16(<vscale x 16 x i16> %a, <vscale x 16 x i16> %a, <vscale x 16 x i16> %b.splat)
+ ret <vscale x 16 x i16> %x
+}
+
+define <vscale x 16 x i16> @vror_vi_nxv16i16(<vscale x 16 x i16> %a) {
+; CHECK-LABEL: vror_vi_nxv16i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e16, m4, ta, ma
+; CHECK-NEXT: vsll.vi v12, v8, 15
+; CHECK-NEXT: vsrl.vi v8, v8, 1
+; CHECK-NEXT: vor.vv v8, v8, v12
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_nxv16i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e16, m4, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 1
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 16 x i16> @llvm.fshr.nxv16i16(<vscale x 16 x i16> %a, <vscale x 16 x i16> %a, <vscale x 16 x i16> shufflevector(<vscale x 16 x i16> insertelement(<vscale x 16 x i16> poison, i16 1, i32 0), <vscale x 16 x i16> poison, <vscale x 16 x i32> zeroinitializer))
+ ret <vscale x 16 x i16> %x
+}
+
+define <vscale x 16 x i16> @vror_vi_rotl_nxv16i16(<vscale x 16 x i16> %a) {
+; CHECK-LABEL: vror_vi_rotl_nxv16i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e16, m4, ta, ma
+; CHECK-NEXT: vsrl.vi v12, v8, 15
+; CHECK-NEXT: vadd.vv v8, v8, v8
+; CHECK-NEXT: vor.vv v8, v8, v12
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_rotl_nxv16i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e16, m4, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 63
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 16 x i16> @llvm.fshl.nxv16i16(<vscale x 16 x i16> %a, <vscale x 16 x i16> %a, <vscale x 16 x i16> shufflevector(<vscale x 16 x i16> insertelement(<vscale x 16 x i16> poison, i16 1, i32 0), <vscale x 16 x i16> poison, <vscale x 16 x i32> zeroinitializer))
+ ret <vscale x 16 x i16> %x
+}
+
+declare <vscale x 32 x i16> @llvm.fshr.nxv32i16(<vscale x 32 x i16>, <vscale x 32 x i16>, <vscale x 32 x i16>)
+declare <vscale x 32 x i16> @llvm.fshl.nxv32i16(<vscale x 32 x i16>, <vscale x 32 x i16>, <vscale x 32 x i16>)
+
+define <vscale x 32 x i16> @vror_vv_nxv32i16(<vscale x 32 x i16> %a, <vscale x 32 x i16> %b) {
+; CHECK-LABEL: vror_vv_nxv32i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e16, m8, ta, ma
+; CHECK-NEXT: vand.vi v24, v16, 15
+; CHECK-NEXT: vsrl.vv v24, v8, v24
+; CHECK-NEXT: vrsub.vi v16, v16, 0
+; CHECK-NEXT: vand.vi v16, v16, 15
+; CHECK-NEXT: vsll.vv v8, v8, v16
+; CHECK-NEXT: vor.vv v8, v24, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vv_nxv32i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e16, m8, ta, ma
+; CHECK-ZVBB-NEXT: vror.vv v8, v8, v16
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 32 x i16> @llvm.fshr.nxv32i16(<vscale x 32 x i16> %a, <vscale x 32 x i16> %a, <vscale x 32 x i16> %b)
+ ret <vscale x 32 x i16> %x
+}
+
+define <vscale x 32 x i16> @vror_vx_nxv32i16(<vscale x 32 x i16> %a, i16 %b) {
+; CHECK-LABEL: vror_vx_nxv32i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a1, zero, e16, m8, ta, ma
+; CHECK-NEXT: vmv.v.x v16, a0
+; CHECK-NEXT: vand.vi v24, v16, 15
+; CHECK-NEXT: vsrl.vv v24, v8, v24
+; CHECK-NEXT: vrsub.vi v16, v16, 0
+; CHECK-NEXT: vand.vi v16, v16, 15
+; CHECK-NEXT: vsll.vv v8, v8, v16
+; CHECK-NEXT: vor.vv v8, v24, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vx_nxv32i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e16, m8, ta, ma
+; CHECK-ZVBB-NEXT: vror.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 32 x i16> poison, i16 %b, i32 0
+ %b.splat = shufflevector <vscale x 32 x i16> %b.head, <vscale x 32 x i16> poison, <vscale x 32 x i32> zeroinitializer
+ %x = call <vscale x 32 x i16> @llvm.fshr.nxv32i16(<vscale x 32 x i16> %a, <vscale x 32 x i16> %a, <vscale x 32 x i16> %b.splat)
+ ret <vscale x 32 x i16> %x
+}
+
+define <vscale x 32 x i16> @vror_vi_nxv32i16(<vscale x 32 x i16> %a) {
+; CHECK-LABEL: vror_vi_nxv32i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e16, m8, ta, ma
+; CHECK-NEXT: vsll.vi v16, v8, 15
+; CHECK-NEXT: vsrl.vi v8, v8, 1
+; CHECK-NEXT: vor.vv v8, v8, v16
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_nxv32i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e16, m8, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 1
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 32 x i16> @llvm.fshr.nxv32i16(<vscale x 32 x i16> %a, <vscale x 32 x i16> %a, <vscale x 32 x i16> shufflevector(<vscale x 32 x i16> insertelement(<vscale x 32 x i16> poison, i16 1, i32 0), <vscale x 32 x i16> poison, <vscale x 32 x i32> zeroinitializer))
+ ret <vscale x 32 x i16> %x
+}
+
+define <vscale x 32 x i16> @vror_vi_rotl_nxv32i16(<vscale x 32 x i16> %a) {
+; CHECK-LABEL: vror_vi_rotl_nxv32i16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e16, m8, ta, ma
+; CHECK-NEXT: vsrl.vi v16, v8, 15
+; CHECK-NEXT: vadd.vv v8, v8, v8
+; CHECK-NEXT: vor.vv v8, v8, v16
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_rotl_nxv32i16:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e16, m8, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 63
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 32 x i16> @llvm.fshl.nxv32i16(<vscale x 32 x i16> %a, <vscale x 32 x i16> %a, <vscale x 32 x i16> shufflevector(<vscale x 32 x i16> insertelement(<vscale x 32 x i16> poison, i16 1, i32 0), <vscale x 32 x i16> poison, <vscale x 32 x i32> zeroinitializer))
+ ret <vscale x 32 x i16> %x
+}
+
+declare <vscale x 1 x i32> @llvm.fshr.nxv1i32(<vscale x 1 x i32>, <vscale x 1 x i32>, <vscale x 1 x i32>)
+declare <vscale x 1 x i32> @llvm.fshl.nxv1i32(<vscale x 1 x i32>, <vscale x 1 x i32>, <vscale x 1 x i32>)
+
+define <vscale x 1 x i32> @vror_vv_nxv1i32(<vscale x 1 x i32> %a, <vscale x 1 x i32> %b) {
+; CHECK-LABEL: vror_vv_nxv1i32:
+; CHECK: # %bb.0:
+; CHECK-NEXT: li a0, 31
+; CHECK-NEXT: vsetvli a1, zero, e32, mf2, ta, ma
+; CHECK-NEXT: vand.vx v10, v9, a0
+; CHECK-NEXT: vsrl.vv v10, v8, v10
+; CHECK-NEXT: vrsub.vi v9, v9, 0
+; CHECK-NEXT: vand.vx v9, v9, a0
+; CHECK-NEXT: vsll.vv v8, v8, v9
+; CHECK-NEXT: vor.vv v8, v10, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vv_nxv1i32:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e32, mf2, ta, ma
+; CHECK-ZVBB-NEXT: vror.vv v8, v8, v9
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 1 x i32> @llvm.fshr.nxv1i32(<vscale x 1 x i32> %a, <vscale x 1 x i32> %a, <vscale x 1 x i32> %b)
+ ret <vscale x 1 x i32> %x
+}
+
+define <vscale x 1 x i32> @vror_vx_nxv1i32(<vscale x 1 x i32> %a, i32 %b) {
+; CHECK-RV32-LABEL: vror_vx_nxv1i32:
+; CHECK-RV32: # %bb.0:
+; CHECK-RV32-NEXT: andi a1, a0, 31
+; CHECK-RV32-NEXT: vsetvli a2, zero, e32, mf2, ta, ma
+; CHECK-RV32-NEXT: vsrl.vx v9, v8, a1
+; CHECK-RV32-NEXT: neg a0, a0
+; CHECK-RV32-NEXT: andi a0, a0, 31
+; CHECK-RV32-NEXT: vsll.vx v8, v8, a0
+; CHECK-RV32-NEXT: vor.vv v8, v9, v8
+; CHECK-RV32-NEXT: ret
+;
+; CHECK-RV64-LABEL: vror_vx_nxv1i32:
+; CHECK-RV64: # %bb.0:
+; CHECK-RV64-NEXT: vsetvli a1, zero, e32, mf2, ta, ma
+; CHECK-RV64-NEXT: vmv.v.x v9, a0
+; CHECK-RV64-NEXT: li a0, 31
+; CHECK-RV64-NEXT: vand.vx v10, v9, a0
+; CHECK-RV64-NEXT: vsrl.vv v10, v8, v10
+; CHECK-RV64-NEXT: vrsub.vi v9, v9, 0
+; CHECK-RV64-NEXT: vand.vx v9, v9, a0
+; CHECK-RV64-NEXT: vsll.vv v8, v8, v9
+; CHECK-RV64-NEXT: vor.vv v8, v10, v8
+; CHECK-RV64-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vx_nxv1i32:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e32, mf2, ta, ma
+; CHECK-ZVBB-NEXT: vror.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 1 x i32> poison, i32 %b, i32 0
+ %b.splat = shufflevector <vscale x 1 x i32> %b.head, <vscale x 1 x i32> poison, <vscale x 1 x i32> zeroinitializer
+ %x = call <vscale x 1 x i32> @llvm.fshr.nxv1i32(<vscale x 1 x i32> %a, <vscale x 1 x i32> %a, <vscale x 1 x i32> %b.splat)
+ ret <vscale x 1 x i32> %x
+}
+
+define <vscale x 1 x i32> @vror_vi_nxv1i32(<vscale x 1 x i32> %a) {
+; CHECK-LABEL: vror_vi_nxv1i32:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e32, mf2, ta, ma
+; CHECK-NEXT: vsll.vi v9, v8, 31
+; CHECK-NEXT: vsrl.vi v8, v8, 1
+; CHECK-NEXT: vor.vv v8, v8, v9
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_nxv1i32:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e32, mf2, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 1
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 1 x i32> @llvm.fshr.nxv1i32(<vscale x 1 x i32> %a, <vscale x 1 x i32> %a, <vscale x 1 x i32> shufflevector(<vscale x 1 x i32> insertelement(<vscale x 1 x i32> poison, i32 1, i32 0), <vscale x 1 x i32> poison, <vscale x 1 x i32> zeroinitializer))
+ ret <vscale x 1 x i32> %x
+}
+
+define <vscale x 1 x i32> @vror_vi_rotl_nxv1i32(<vscale x 1 x i32> %a) {
+; CHECK-LABEL: vror_vi_rotl_nxv1i32:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e32, mf2, ta, ma
+; CHECK-NEXT: vsrl.vi v9, v8, 31
+; CHECK-NEXT: vadd.vv v8, v8, v8
+; CHECK-NEXT: vor.vv v8, v8, v9
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_rotl_nxv1i32:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e32, mf2, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 63
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 1 x i32> @llvm.fshl.nxv1i32(<vscale x 1 x i32> %a, <vscale x 1 x i32> %a, <vscale x 1 x i32> shufflevector(<vscale x 1 x i32> insertelement(<vscale x 1 x i32> poison, i32 1, i32 0), <vscale x 1 x i32> poison, <vscale x 1 x i32> zeroinitializer))
+ ret <vscale x 1 x i32> %x
+}
+
+declare <vscale x 2 x i32> @llvm.fshr.nxv2i32(<vscale x 2 x i32>, <vscale x 2 x i32>, <vscale x 2 x i32>)
+declare <vscale x 2 x i32> @llvm.fshl.nxv2i32(<vscale x 2 x i32>, <vscale x 2 x i32>, <vscale x 2 x i32>)
+
+define <vscale x 2 x i32> @vror_vv_nxv2i32(<vscale x 2 x i32> %a, <vscale x 2 x i32> %b) {
+; CHECK-LABEL: vror_vv_nxv2i32:
+; CHECK: # %bb.0:
+; CHECK-NEXT: li a0, 31
+; CHECK-NEXT: vsetvli a1, zero, e32, m1, ta, ma
+; CHECK-NEXT: vand.vx v10, v9, a0
+; CHECK-NEXT: vsrl.vv v10, v8, v10
+; CHECK-NEXT: vrsub.vi v9, v9, 0
+; CHECK-NEXT: vand.vx v9, v9, a0
+; CHECK-NEXT: vsll.vv v8, v8, v9
+; CHECK-NEXT: vor.vv v8, v10, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vv_nxv2i32:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e32, m1, ta, ma
+; CHECK-ZVBB-NEXT: vror.vv v8, v8, v9
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 2 x i32> @llvm.fshr.nxv2i32(<vscale x 2 x i32> %a, <vscale x 2 x i32> %a, <vscale x 2 x i32> %b)
+ ret <vscale x 2 x i32> %x
+}
+
+define <vscale x 2 x i32> @vror_vx_nxv2i32(<vscale x 2 x i32> %a, i32 %b) {
+; CHECK-RV32-LABEL: vror_vx_nxv2i32:
+; CHECK-RV32: # %bb.0:
+; CHECK-RV32-NEXT: andi a1, a0, 31
+; CHECK-RV32-NEXT: vsetvli a2, zero, e32, m1, ta, ma
+; CHECK-RV32-NEXT: vsrl.vx v9, v8, a1
+; CHECK-RV32-NEXT: neg a0, a0
+; CHECK-RV32-NEXT: andi a0, a0, 31
+; CHECK-RV32-NEXT: vsll.vx v8, v8, a0
+; CHECK-RV32-NEXT: vor.vv v8, v9, v8
+; CHECK-RV32-NEXT: ret
+;
+; CHECK-RV64-LABEL: vror_vx_nxv2i32:
+; CHECK-RV64: # %bb.0:
+; CHECK-RV64-NEXT: vsetvli a1, zero, e32, m1, ta, ma
+; CHECK-RV64-NEXT: vmv.v.x v9, a0
+; CHECK-RV64-NEXT: li a0, 31
+; CHECK-RV64-NEXT: vand.vx v10, v9, a0
+; CHECK-RV64-NEXT: vsrl.vv v10, v8, v10
+; CHECK-RV64-NEXT: vrsub.vi v9, v9, 0
+; CHECK-RV64-NEXT: vand.vx v9, v9, a0
+; CHECK-RV64-NEXT: vsll.vv v8, v8, v9
+; CHECK-RV64-NEXT: vor.vv v8, v10, v8
+; CHECK-RV64-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vx_nxv2i32:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e32, m1, ta, ma
+; CHECK-ZVBB-NEXT: vror.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 2 x i32> poison, i32 %b, i32 0
+ %b.splat = shufflevector <vscale x 2 x i32> %b.head, <vscale x 2 x i32> poison, <vscale x 2 x i32> zeroinitializer
+ %x = call <vscale x 2 x i32> @llvm.fshr.nxv2i32(<vscale x 2 x i32> %a, <vscale x 2 x i32> %a, <vscale x 2 x i32> %b.splat)
+ ret <vscale x 2 x i32> %x
+}
+
+define <vscale x 2 x i32> @vror_vi_nxv2i32(<vscale x 2 x i32> %a) {
+; CHECK-LABEL: vror_vi_nxv2i32:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e32, m1, ta, ma
+; CHECK-NEXT: vsll.vi v9, v8, 31
+; CHECK-NEXT: vsrl.vi v8, v8, 1
+; CHECK-NEXT: vor.vv v8, v8, v9
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_nxv2i32:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e32, m1, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 1
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 2 x i32> @llvm.fshr.nxv2i32(<vscale x 2 x i32> %a, <vscale x 2 x i32> %a, <vscale x 2 x i32> shufflevector(<vscale x 2 x i32> insertelement(<vscale x 2 x i32> poison, i32 1, i32 0), <vscale x 2 x i32> poison, <vscale x 2 x i32> zeroinitializer))
+ ret <vscale x 2 x i32> %x
+}
+
+define <vscale x 2 x i32> @vror_vi_rotl_nxv2i32(<vscale x 2 x i32> %a) {
+; CHECK-LABEL: vror_vi_rotl_nxv2i32:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e32, m1, ta, ma
+; CHECK-NEXT: vsrl.vi v9, v8, 31
+; CHECK-NEXT: vadd.vv v8, v8, v8
+; CHECK-NEXT: vor.vv v8, v8, v9
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_rotl_nxv2i32:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e32, m1, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 63
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 2 x i32> @llvm.fshl.nxv2i32(<vscale x 2 x i32> %a, <vscale x 2 x i32> %a, <vscale x 2 x i32> shufflevector(<vscale x 2 x i32> insertelement(<vscale x 2 x i32> poison, i32 1, i32 0), <vscale x 2 x i32> poison, <vscale x 2 x i32> zeroinitializer))
+ ret <vscale x 2 x i32> %x
+}
+
+declare <vscale x 4 x i32> @llvm.fshr.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>)
+declare <vscale x 4 x i32> @llvm.fshl.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>)
+
+define <vscale x 4 x i32> @vror_vv_nxv4i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b) {
+; CHECK-LABEL: vror_vv_nxv4i32:
+; CHECK: # %bb.0:
+; CHECK-NEXT: li a0, 31
+; CHECK-NEXT: vsetvli a1, zero, e32, m2, ta, ma
+; CHECK-NEXT: vand.vx v12, v10, a0
+; CHECK-NEXT: vsrl.vv v12, v8, v12
+; CHECK-NEXT: vrsub.vi v10, v10, 0
+; CHECK-NEXT: vand.vx v10, v10, a0
+; CHECK-NEXT: vsll.vv v8, v8, v10
+; CHECK-NEXT: vor.vv v8, v12, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vv_nxv4i32:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e32, m2, ta, ma
+; CHECK-ZVBB-NEXT: vror.vv v8, v8, v10
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 4 x i32> @llvm.fshr.nxv4i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %a, <vscale x 4 x i32> %b)
+ ret <vscale x 4 x i32> %x
+}
+
+define <vscale x 4 x i32> @vror_vx_nxv4i32(<vscale x 4 x i32> %a, i32 %b) {
+; CHECK-RV32-LABEL: vror_vx_nxv4i32:
+; CHECK-RV32: # %bb.0:
+; CHECK-RV32-NEXT: andi a1, a0, 31
+; CHECK-RV32-NEXT: vsetvli a2, zero, e32, m2, ta, ma
+; CHECK-RV32-NEXT: vsrl.vx v10, v8, a1
+; CHECK-RV32-NEXT: neg a0, a0
+; CHECK-RV32-NEXT: andi a0, a0, 31
+; CHECK-RV32-NEXT: vsll.vx v8, v8, a0
+; CHECK-RV32-NEXT: vor.vv v8, v10, v8
+; CHECK-RV32-NEXT: ret
+;
+; CHECK-RV64-LABEL: vror_vx_nxv4i32:
+; CHECK-RV64: # %bb.0:
+; CHECK-RV64-NEXT: vsetvli a1, zero, e32, m2, ta, ma
+; CHECK-RV64-NEXT: vmv.v.x v10, a0
+; CHECK-RV64-NEXT: li a0, 31
+; CHECK-RV64-NEXT: vand.vx v12, v10, a0
+; CHECK-RV64-NEXT: vsrl.vv v12, v8, v12
+; CHECK-RV64-NEXT: vrsub.vi v10, v10, 0
+; CHECK-RV64-NEXT: vand.vx v10, v10, a0
+; CHECK-RV64-NEXT: vsll.vv v8, v8, v10
+; CHECK-RV64-NEXT: vor.vv v8, v12, v8
+; CHECK-RV64-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vx_nxv4i32:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e32, m2, ta, ma
+; CHECK-ZVBB-NEXT: vror.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 4 x i32> poison, i32 %b, i32 0
+ %b.splat = shufflevector <vscale x 4 x i32> %b.head, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
+ %x = call <vscale x 4 x i32> @llvm.fshr.nxv4i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %a, <vscale x 4 x i32> %b.splat)
+ ret <vscale x 4 x i32> %x
+}
+
+define <vscale x 4 x i32> @vror_vi_nxv4i32(<vscale x 4 x i32> %a) {
+; CHECK-LABEL: vror_vi_nxv4i32:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e32, m2, ta, ma
+; CHECK-NEXT: vsll.vi v10, v8, 31
+; CHECK-NEXT: vsrl.vi v8, v8, 1
+; CHECK-NEXT: vor.vv v8, v8, v10
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_nxv4i32:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e32, m2, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 1
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 4 x i32> @llvm.fshr.nxv4i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %a, <vscale x 4 x i32> shufflevector(<vscale x 4 x i32> insertelement(<vscale x 4 x i32> poison, i32 1, i32 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer))
+ ret <vscale x 4 x i32> %x
+}
+
+define <vscale x 4 x i32> @vror_vi_rotl_nxv4i32(<vscale x 4 x i32> %a) {
+; CHECK-LABEL: vror_vi_rotl_nxv4i32:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e32, m2, ta, ma
+; CHECK-NEXT: vsrl.vi v10, v8, 31
+; CHECK-NEXT: vadd.vv v8, v8, v8
+; CHECK-NEXT: vor.vv v8, v8, v10
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_rotl_nxv4i32:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e32, m2, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 63
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 4 x i32> @llvm.fshl.nxv4i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %a, <vscale x 4 x i32> shufflevector(<vscale x 4 x i32> insertelement(<vscale x 4 x i32> poison, i32 1, i32 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer))
+ ret <vscale x 4 x i32> %x
+}
+
+declare <vscale x 8 x i32> @llvm.fshr.nxv8i32(<vscale x 8 x i32>, <vscale x 8 x i32>, <vscale x 8 x i32>)
+declare <vscale x 8 x i32> @llvm.fshl.nxv8i32(<vscale x 8 x i32>, <vscale x 8 x i32>, <vscale x 8 x i32>)
+
+define <vscale x 8 x i32> @vror_vv_nxv8i32(<vscale x 8 x i32> %a, <vscale x 8 x i32> %b) {
+; CHECK-LABEL: vror_vv_nxv8i32:
+; CHECK: # %bb.0:
+; CHECK-NEXT: li a0, 31
+; CHECK-NEXT: vsetvli a1, zero, e32, m4, ta, ma
+; CHECK-NEXT: vand.vx v16, v12, a0
+; CHECK-NEXT: vsrl.vv v16, v8, v16
+; CHECK-NEXT: vrsub.vi v12, v12, 0
+; CHECK-NEXT: vand.vx v12, v12, a0
+; CHECK-NEXT: vsll.vv v8, v8, v12
+; CHECK-NEXT: vor.vv v8, v16, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vv_nxv8i32:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e32, m4, ta, ma
+; CHECK-ZVBB-NEXT: vror.vv v8, v8, v12
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 8 x i32> @llvm.fshr.nxv8i32(<vscale x 8 x i32> %a, <vscale x 8 x i32> %a, <vscale x 8 x i32> %b)
+ ret <vscale x 8 x i32> %x
+}
+
+define <vscale x 8 x i32> @vror_vx_nxv8i32(<vscale x 8 x i32> %a, i32 %b) {
+; CHECK-RV32-LABEL: vror_vx_nxv8i32:
+; CHECK-RV32: # %bb.0:
+; CHECK-RV32-NEXT: andi a1, a0, 31
+; CHECK-RV32-NEXT: vsetvli a2, zero, e32, m4, ta, ma
+; CHECK-RV32-NEXT: vsrl.vx v12, v8, a1
+; CHECK-RV32-NEXT: neg a0, a0
+; CHECK-RV32-NEXT: andi a0, a0, 31
+; CHECK-RV32-NEXT: vsll.vx v8, v8, a0
+; CHECK-RV32-NEXT: vor.vv v8, v12, v8
+; CHECK-RV32-NEXT: ret
+;
+; CHECK-RV64-LABEL: vror_vx_nxv8i32:
+; CHECK-RV64: # %bb.0:
+; CHECK-RV64-NEXT: vsetvli a1, zero, e32, m4, ta, ma
+; CHECK-RV64-NEXT: vmv.v.x v12, a0
+; CHECK-RV64-NEXT: li a0, 31
+; CHECK-RV64-NEXT: vand.vx v16, v12, a0
+; CHECK-RV64-NEXT: vsrl.vv v16, v8, v16
+; CHECK-RV64-NEXT: vrsub.vi v12, v12, 0
+; CHECK-RV64-NEXT: vand.vx v12, v12, a0
+; CHECK-RV64-NEXT: vsll.vv v8, v8, v12
+; CHECK-RV64-NEXT: vor.vv v8, v16, v8
+; CHECK-RV64-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vx_nxv8i32:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e32, m4, ta, ma
+; CHECK-ZVBB-NEXT: vror.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 8 x i32> poison, i32 %b, i32 0
+ %b.splat = shufflevector <vscale x 8 x i32> %b.head, <vscale x 8 x i32> poison, <vscale x 8 x i32> zeroinitializer
+ %x = call <vscale x 8 x i32> @llvm.fshr.nxv8i32(<vscale x 8 x i32> %a, <vscale x 8 x i32> %a, <vscale x 8 x i32> %b.splat)
+ ret <vscale x 8 x i32> %x
+}
+
+define <vscale x 8 x i32> @vror_vi_nxv8i32(<vscale x 8 x i32> %a) {
+; CHECK-LABEL: vror_vi_nxv8i32:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e32, m4, ta, ma
+; CHECK-NEXT: vsll.vi v12, v8, 31
+; CHECK-NEXT: vsrl.vi v8, v8, 1
+; CHECK-NEXT: vor.vv v8, v8, v12
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_nxv8i32:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e32, m4, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 1
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 8 x i32> @llvm.fshr.nxv8i32(<vscale x 8 x i32> %a, <vscale x 8 x i32> %a, <vscale x 8 x i32> shufflevector(<vscale x 8 x i32> insertelement(<vscale x 8 x i32> poison, i32 1, i32 0), <vscale x 8 x i32> poison, <vscale x 8 x i32> zeroinitializer))
+ ret <vscale x 8 x i32> %x
+}
+
+define <vscale x 8 x i32> @vror_vi_rotl_nxv8i32(<vscale x 8 x i32> %a) {
+; CHECK-LABEL: vror_vi_rotl_nxv8i32:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e32, m4, ta, ma
+; CHECK-NEXT: vsrl.vi v12, v8, 31
+; CHECK-NEXT: vadd.vv v8, v8, v8
+; CHECK-NEXT: vor.vv v8, v8, v12
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_rotl_nxv8i32:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e32, m4, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 63
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 8 x i32> @llvm.fshl.nxv8i32(<vscale x 8 x i32> %a, <vscale x 8 x i32> %a, <vscale x 8 x i32> shufflevector(<vscale x 8 x i32> insertelement(<vscale x 8 x i32> poison, i32 1, i32 0), <vscale x 8 x i32> poison, <vscale x 8 x i32> zeroinitializer))
+ ret <vscale x 8 x i32> %x
+}
+
+declare <vscale x 16 x i32> @llvm.fshr.nxv16i32(<vscale x 16 x i32>, <vscale x 16 x i32>, <vscale x 16 x i32>)
+declare <vscale x 16 x i32> @llvm.fshl.nxv16i32(<vscale x 16 x i32>, <vscale x 16 x i32>, <vscale x 16 x i32>)
+
+define <vscale x 16 x i32> @vror_vv_nxv16i32(<vscale x 16 x i32> %a, <vscale x 16 x i32> %b) {
+; CHECK-LABEL: vror_vv_nxv16i32:
+; CHECK: # %bb.0:
+; CHECK-NEXT: li a0, 31
+; CHECK-NEXT: vsetvli a1, zero, e32, m8, ta, ma
+; CHECK-NEXT: vand.vx v24, v16, a0
+; CHECK-NEXT: vsrl.vv v24, v8, v24
+; CHECK-NEXT: vrsub.vi v16, v16, 0
+; CHECK-NEXT: vand.vx v16, v16, a0
+; CHECK-NEXT: vsll.vv v8, v8, v16
+; CHECK-NEXT: vor.vv v8, v24, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vv_nxv16i32:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e32, m8, ta, ma
+; CHECK-ZVBB-NEXT: vror.vv v8, v8, v16
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 16 x i32> @llvm.fshr.nxv16i32(<vscale x 16 x i32> %a, <vscale x 16 x i32> %a, <vscale x 16 x i32> %b)
+ ret <vscale x 16 x i32> %x
+}
+
+define <vscale x 16 x i32> @vror_vx_nxv16i32(<vscale x 16 x i32> %a, i32 %b) {
+; CHECK-RV32-LABEL: vror_vx_nxv16i32:
+; CHECK-RV32: # %bb.0:
+; CHECK-RV32-NEXT: andi a1, a0, 31
+; CHECK-RV32-NEXT: vsetvli a2, zero, e32, m8, ta, ma
+; CHECK-RV32-NEXT: vsrl.vx v16, v8, a1
+; CHECK-RV32-NEXT: neg a0, a0
+; CHECK-RV32-NEXT: andi a0, a0, 31
+; CHECK-RV32-NEXT: vsll.vx v8, v8, a0
+; CHECK-RV32-NEXT: vor.vv v8, v16, v8
+; CHECK-RV32-NEXT: ret
+;
+; CHECK-RV64-LABEL: vror_vx_nxv16i32:
+; CHECK-RV64: # %bb.0:
+; CHECK-RV64-NEXT: vsetvli a1, zero, e32, m8, ta, ma
+; CHECK-RV64-NEXT: vmv.v.x v16, a0
+; CHECK-RV64-NEXT: li a0, 31
+; CHECK-RV64-NEXT: vand.vx v24, v16, a0
+; CHECK-RV64-NEXT: vsrl.vv v24, v8, v24
+; CHECK-RV64-NEXT: vrsub.vi v16, v16, 0
+; CHECK-RV64-NEXT: vand.vx v16, v16, a0
+; CHECK-RV64-NEXT: vsll.vv v8, v8, v16
+; CHECK-RV64-NEXT: vor.vv v8, v24, v8
+; CHECK-RV64-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vx_nxv16i32:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a1, zero, e32, m8, ta, ma
+; CHECK-ZVBB-NEXT: vror.vx v8, v8, a0
+; CHECK-ZVBB-NEXT: ret
+ %b.head = insertelement <vscale x 16 x i32> poison, i32 %b, i32 0
+ %b.splat = shufflevector <vscale x 16 x i32> %b.head, <vscale x 16 x i32> poison, <vscale x 16 x i32> zeroinitializer
+ %x = call <vscale x 16 x i32> @llvm.fshr.nxv16i32(<vscale x 16 x i32> %a, <vscale x 16 x i32> %a, <vscale x 16 x i32> %b.splat)
+ ret <vscale x 16 x i32> %x
+}
+
+define <vscale x 16 x i32> @vror_vi_nxv16i32(<vscale x 16 x i32> %a) {
+; CHECK-LABEL: vror_vi_nxv16i32:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e32, m8, ta, ma
+; CHECK-NEXT: vsll.vi v16, v8, 31
+; CHECK-NEXT: vsrl.vi v8, v8, 1
+; CHECK-NEXT: vor.vv v8, v8, v16
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_nxv16i32:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e32, m8, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 1
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 16 x i32> @llvm.fshr.nxv16i32(<vscale x 16 x i32> %a, <vscale x 16 x i32> %a, <vscale x 16 x i32> shufflevector(<vscale x 16 x i32> insertelement(<vscale x 16 x i32> poison, i32 1, i32 0), <vscale x 16 x i32> poison, <vscale x 16 x i32> zeroinitializer))
+ ret <vscale x 16 x i32> %x
+}
+
+define <vscale x 16 x i32> @vror_vi_rotl_nxv16i32(<vscale x 16 x i32> %a) {
+; CHECK-LABEL: vror_vi_rotl_nxv16i32:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e32, m8, ta, ma
+; CHECK-NEXT: vsrl.vi v16, v8, 31
+; CHECK-NEXT: vadd.vv v8, v8, v8
+; CHECK-NEXT: vor.vv v8, v8, v16
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_rotl_nxv16i32:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e32, m8, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 63
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 16 x i32> @llvm.fshl.nxv16i32(<vscale x 16 x i32> %a, <vscale x 16 x i32> %a, <vscale x 16 x i32> shufflevector(<vscale x 16 x i32> insertelement(<vscale x 16 x i32> poison, i32 1, i32 0), <vscale x 16 x i32> poison, <vscale x 16 x i32> zeroinitializer))
+ ret <vscale x 16 x i32> %x
+}
+
+declare <vscale x 1 x i64> @llvm.fshr.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>)
+declare <vscale x 1 x i64> @llvm.fshl.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>)
+
+define <vscale x 1 x i64> @vror_vv_nxv1i64(<vscale x 1 x i64> %a, <vscale x 1 x i64> %b) {
+; CHECK-LABEL: vror_vv_nxv1i64:
+; CHECK: # %bb.0:
+; CHECK-NEXT: li a0, 63
+; CHECK-NEXT: vsetvli a1, zero, e64, m1, ta, ma
+; CHECK-NEXT: vand.vx v10, v9, a0
+; CHECK-NEXT: vsrl.vv v10, v8, v10
+; CHECK-NEXT: vrsub.vi v9, v9, 0
+; CHECK-NEXT: vand.vx v9, v9, a0
+; CHECK-NEXT: vsll.vv v8, v8, v9
+; CHECK-NEXT: vor.vv v8, v10, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vv_nxv1i64:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-ZVBB-NEXT: vror.vv v8, v8, v9
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 1 x i64> @llvm.fshr.nxv1i64(<vscale x 1 x i64> %a, <vscale x 1 x i64> %a, <vscale x 1 x i64> %b)
+ ret <vscale x 1 x i64> %x
+}
+
+define <vscale x 1 x i64> @vror_vx_nxv1i64(<vscale x 1 x i64> %a, i64 %b) {
+; CHECK-RV32-LABEL: vror_vx_nxv1i64:
+; CHECK-RV32: # %bb.0:
+; CHECK-RV32-NEXT: addi sp, sp, -16
+; CHECK-RV32-NEXT: .cfi_def_cfa_offset 16
+; CHECK-RV32-NEXT: sw a1, 12(sp)
+; CHECK-RV32-NEXT: sw a0, 8(sp)
+; CHECK-RV32-NEXT: addi a0, sp, 8
+; CHECK-RV32-NEXT: vsetvli a1, zero, e64, m1, ta, ma
+; CHECK-RV32-NEXT: vlse64.v v9, (a0), zero
+; CHECK-RV32-NEXT: li a0, 63
+; CHECK-RV32-NEXT: vand.vx v10, v9, a0
+; CHECK-RV32-NEXT: vsrl.vv v10, v8, v10
+; CHECK-RV32-NEXT: vrsub.vi v9, v9, 0
+; CHECK-RV32-NEXT: vand.vx v9, v9, a0
+; CHECK-RV32-NEXT: vsll.vv v8, v8, v9
+; CHECK-RV32-NEXT: vor.vv v8, v10, v8
+; CHECK-RV32-NEXT: addi sp, sp, 16
+; CHECK-RV32-NEXT: ret
+;
+; CHECK-RV64-LABEL: vror_vx_nxv1i64:
+; CHECK-RV64: # %bb.0:
+; CHECK-RV64-NEXT: andi a1, a0, 63
+; CHECK-RV64-NEXT: vsetvli a2, zero, e64, m1, ta, ma
+; CHECK-RV64-NEXT: vsrl.vx v9, v8, a1
+; CHECK-RV64-NEXT: negw a0, a0
+; CHECK-RV64-NEXT: andi a0, a0, 63
+; CHECK-RV64-NEXT: vsll.vx v8, v8, a0
+; CHECK-RV64-NEXT: vor.vv v8, v9, v8
+; CHECK-RV64-NEXT: ret
+;
+; CHECK-ZVBB32-LABEL: vror_vx_nxv1i64:
+; CHECK-ZVBB32: # %bb.0:
+; CHECK-ZVBB32-NEXT: addi sp, sp, -16
+; CHECK-ZVBB32-NEXT: .cfi_def_cfa_offset 16
+; CHECK-ZVBB32-NEXT: sw a1, 12(sp)
+; CHECK-ZVBB32-NEXT: sw a0, 8(sp)
+; CHECK-ZVBB32-NEXT: addi a0, sp, 8
+; CHECK-ZVBB32-NEXT: vsetvli a1, zero, e64, m1, ta, ma
+; CHECK-ZVBB32-NEXT: vlse64.v v9, (a0), zero
+; CHECK-ZVBB32-NEXT: vror.vv v8, v8, v9
+; CHECK-ZVBB32-NEXT: addi sp, sp, 16
+; CHECK-ZVBB32-NEXT: ret
+;
+; CHECK-ZVBB64-LABEL: vror_vx_nxv1i64:
+; CHECK-ZVBB64: # %bb.0:
+; CHECK-ZVBB64-NEXT: vsetvli a1, zero, e64, m1, ta, ma
+; CHECK-ZVBB64-NEXT: vror.vx v8, v8, a0
+; CHECK-ZVBB64-NEXT: ret
+ %b.head = insertelement <vscale x 1 x i64> poison, i64 %b, i32 0
+ %b.splat = shufflevector <vscale x 1 x i64> %b.head, <vscale x 1 x i64> poison, <vscale x 1 x i32> zeroinitializer
+ %x = call <vscale x 1 x i64> @llvm.fshr.nxv1i64(<vscale x 1 x i64> %a, <vscale x 1 x i64> %a, <vscale x 1 x i64> %b.splat)
+ ret <vscale x 1 x i64> %x
+}
+
+define <vscale x 1 x i64> @vror_vi_nxv1i64(<vscale x 1 x i64> %a) {
+; CHECK-RV32-LABEL: vror_vi_nxv1i64:
+; CHECK-RV32: # %bb.0:
+; CHECK-RV32-NEXT: li a0, 63
+; CHECK-RV32-NEXT: vsetvli a1, zero, e64, m1, ta, ma
+; CHECK-RV32-NEXT: vmv.v.x v9, a0
+; CHECK-RV32-NEXT: vand.vi v9, v9, 1
+; CHECK-RV32-NEXT: vsrl.vv v9, v8, v9
+; CHECK-RV32-NEXT: vmv.v.i v10, 1
+; CHECK-RV32-NEXT: vrsub.vi v10, v10, 0
+; CHECK-RV32-NEXT: vand.vx v10, v10, a0
+; CHECK-RV32-NEXT: vsll.vv v8, v8, v10
+; CHECK-RV32-NEXT: vor.vv v8, v9, v8
+; CHECK-RV32-NEXT: ret
+;
+; CHECK-RV64-LABEL: vror_vi_nxv1i64:
+; CHECK-RV64: # %bb.0:
+; CHECK-RV64-NEXT: li a0, 63
+; CHECK-RV64-NEXT: vsetvli a1, zero, e64, m1, ta, ma
+; CHECK-RV64-NEXT: vsll.vx v9, v8, a0
+; CHECK-RV64-NEXT: vsrl.vi v8, v8, 1
+; CHECK-RV64-NEXT: vor.vv v8, v8, v9
+; CHECK-RV64-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_nxv1i64:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 1
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 1 x i64> @llvm.fshr.nxv1i64(<vscale x 1 x i64> %a, <vscale x 1 x i64> %a, <vscale x 1 x i64> shufflevector(<vscale x 1 x i64> insertelement(<vscale x 1 x i64> poison, i64 1, i32 0), <vscale x 1 x i64> poison, <vscale x 1 x i32> zeroinitializer))
+ ret <vscale x 1 x i64> %x
+}
+
+define <vscale x 1 x i64> @vror_vi_rotl_nxv1i64(<vscale x 1 x i64> %a) {
+; CHECK-RV32-LABEL: vror_vi_rotl_nxv1i64:
+; CHECK-RV32: # %bb.0:
+; CHECK-RV32-NEXT: li a0, 63
+; CHECK-RV32-NEXT: vsetvli a1, zero, e64, m1, ta, ma
+; CHECK-RV32-NEXT: vmv.v.x v9, a0
+; CHECK-RV32-NEXT: vand.vi v9, v9, 1
+; CHECK-RV32-NEXT: vsll.vv v9, v8, v9
+; CHECK-RV32-NEXT: vmv.v.i v10, 1
+; CHECK-RV32-NEXT: vrsub.vi v10, v10, 0
+; CHECK-RV32-NEXT: vand.vx v10, v10, a0
+; CHECK-RV32-NEXT: vsrl.vv v8, v8, v10
+; CHECK-RV32-NEXT: vor.vv v8, v9, v8
+; CHECK-RV32-NEXT: ret
+;
+; CHECK-RV64-LABEL: vror_vi_rotl_nxv1i64:
+; CHECK-RV64: # %bb.0:
+; CHECK-RV64-NEXT: li a0, 63
+; CHECK-RV64-NEXT: vsetvli a1, zero, e64, m1, ta, ma
+; CHECK-RV64-NEXT: vsrl.vx v9, v8, a0
+; CHECK-RV64-NEXT: vadd.vv v8, v8, v8
+; CHECK-RV64-NEXT: vor.vv v8, v8, v9
+; CHECK-RV64-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_rotl_nxv1i64:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 63
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 1 x i64> @llvm.fshl.nxv1i64(<vscale x 1 x i64> %a, <vscale x 1 x i64> %a, <vscale x 1 x i64> shufflevector(<vscale x 1 x i64> insertelement(<vscale x 1 x i64> poison, i64 1, i32 0), <vscale x 1 x i64> poison, <vscale x 1 x i32> zeroinitializer))
+ ret <vscale x 1 x i64> %x
+}
+
+declare <vscale x 2 x i64> @llvm.fshr.nxv2i64(<vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>)
+declare <vscale x 2 x i64> @llvm.fshl.nxv2i64(<vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>)
+
+define <vscale x 2 x i64> @vror_vv_nxv2i64(<vscale x 2 x i64> %a, <vscale x 2 x i64> %b) {
+; CHECK-LABEL: vror_vv_nxv2i64:
+; CHECK: # %bb.0:
+; CHECK-NEXT: li a0, 63
+; CHECK-NEXT: vsetvli a1, zero, e64, m2, ta, ma
+; CHECK-NEXT: vand.vx v12, v10, a0
+; CHECK-NEXT: vsrl.vv v12, v8, v12
+; CHECK-NEXT: vrsub.vi v10, v10, 0
+; CHECK-NEXT: vand.vx v10, v10, a0
+; CHECK-NEXT: vsll.vv v8, v8, v10
+; CHECK-NEXT: vor.vv v8, v12, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vv_nxv2i64:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e64, m2, ta, ma
+; CHECK-ZVBB-NEXT: vror.vv v8, v8, v10
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 2 x i64> @llvm.fshr.nxv2i64(<vscale x 2 x i64> %a, <vscale x 2 x i64> %a, <vscale x 2 x i64> %b)
+ ret <vscale x 2 x i64> %x
+}
+
+define <vscale x 2 x i64> @vror_vx_nxv2i64(<vscale x 2 x i64> %a, i64 %b) {
+; CHECK-RV32-LABEL: vror_vx_nxv2i64:
+; CHECK-RV32: # %bb.0:
+; CHECK-RV32-NEXT: addi sp, sp, -16
+; CHECK-RV32-NEXT: .cfi_def_cfa_offset 16
+; CHECK-RV32-NEXT: sw a1, 12(sp)
+; CHECK-RV32-NEXT: sw a0, 8(sp)
+; CHECK-RV32-NEXT: addi a0, sp, 8
+; CHECK-RV32-NEXT: vsetvli a1, zero, e64, m2, ta, ma
+; CHECK-RV32-NEXT: vlse64.v v10, (a0), zero
+; CHECK-RV32-NEXT: li a0, 63
+; CHECK-RV32-NEXT: vand.vx v12, v10, a0
+; CHECK-RV32-NEXT: vsrl.vv v12, v8, v12
+; CHECK-RV32-NEXT: vrsub.vi v10, v10, 0
+; CHECK-RV32-NEXT: vand.vx v10, v10, a0
+; CHECK-RV32-NEXT: vsll.vv v8, v8, v10
+; CHECK-RV32-NEXT: vor.vv v8, v12, v8
+; CHECK-RV32-NEXT: addi sp, sp, 16
+; CHECK-RV32-NEXT: ret
+;
+; CHECK-RV64-LABEL: vror_vx_nxv2i64:
+; CHECK-RV64: # %bb.0:
+; CHECK-RV64-NEXT: andi a1, a0, 63
+; CHECK-RV64-NEXT: vsetvli a2, zero, e64, m2, ta, ma
+; CHECK-RV64-NEXT: vsrl.vx v10, v8, a1
+; CHECK-RV64-NEXT: negw a0, a0
+; CHECK-RV64-NEXT: andi a0, a0, 63
+; CHECK-RV64-NEXT: vsll.vx v8, v8, a0
+; CHECK-RV64-NEXT: vor.vv v8, v10, v8
+; CHECK-RV64-NEXT: ret
+;
+; CHECK-ZVBB32-LABEL: vror_vx_nxv2i64:
+; CHECK-ZVBB32: # %bb.0:
+; CHECK-ZVBB32-NEXT: addi sp, sp, -16
+; CHECK-ZVBB32-NEXT: .cfi_def_cfa_offset 16
+; CHECK-ZVBB32-NEXT: sw a1, 12(sp)
+; CHECK-ZVBB32-NEXT: sw a0, 8(sp)
+; CHECK-ZVBB32-NEXT: addi a0, sp, 8
+; CHECK-ZVBB32-NEXT: vsetvli a1, zero, e64, m2, ta, ma
+; CHECK-ZVBB32-NEXT: vlse64.v v10, (a0), zero
+; CHECK-ZVBB32-NEXT: vror.vv v8, v8, v10
+; CHECK-ZVBB32-NEXT: addi sp, sp, 16
+; CHECK-ZVBB32-NEXT: ret
+;
+; CHECK-ZVBB64-LABEL: vror_vx_nxv2i64:
+; CHECK-ZVBB64: # %bb.0:
+; CHECK-ZVBB64-NEXT: vsetvli a1, zero, e64, m2, ta, ma
+; CHECK-ZVBB64-NEXT: vror.vx v8, v8, a0
+; CHECK-ZVBB64-NEXT: ret
+ %b.head = insertelement <vscale x 2 x i64> poison, i64 %b, i32 0
+ %b.splat = shufflevector <vscale x 2 x i64> %b.head, <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer
+ %x = call <vscale x 2 x i64> @llvm.fshr.nxv2i64(<vscale x 2 x i64> %a, <vscale x 2 x i64> %a, <vscale x 2 x i64> %b.splat)
+ ret <vscale x 2 x i64> %x
+}
+
+define <vscale x 2 x i64> @vror_vi_nxv2i64(<vscale x 2 x i64> %a) {
+; CHECK-RV32-LABEL: vror_vi_nxv2i64:
+; CHECK-RV32: # %bb.0:
+; CHECK-RV32-NEXT: li a0, 63
+; CHECK-RV32-NEXT: vsetvli a1, zero, e64, m2, ta, ma
+; CHECK-RV32-NEXT: vmv.v.x v10, a0
+; CHECK-RV32-NEXT: vand.vi v10, v10, 1
+; CHECK-RV32-NEXT: vsrl.vv v10, v8, v10
+; CHECK-RV32-NEXT: vmv.v.i v12, 1
+; CHECK-RV32-NEXT: vrsub.vi v12, v12, 0
+; CHECK-RV32-NEXT: vand.vx v12, v12, a0
+; CHECK-RV32-NEXT: vsll.vv v8, v8, v12
+; CHECK-RV32-NEXT: vor.vv v8, v10, v8
+; CHECK-RV32-NEXT: ret
+;
+; CHECK-RV64-LABEL: vror_vi_nxv2i64:
+; CHECK-RV64: # %bb.0:
+; CHECK-RV64-NEXT: li a0, 63
+; CHECK-RV64-NEXT: vsetvli a1, zero, e64, m2, ta, ma
+; CHECK-RV64-NEXT: vsll.vx v10, v8, a0
+; CHECK-RV64-NEXT: vsrl.vi v8, v8, 1
+; CHECK-RV64-NEXT: vor.vv v8, v8, v10
+; CHECK-RV64-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_nxv2i64:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e64, m2, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 1
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 2 x i64> @llvm.fshr.nxv2i64(<vscale x 2 x i64> %a, <vscale x 2 x i64> %a, <vscale x 2 x i64> shufflevector(<vscale x 2 x i64> insertelement(<vscale x 2 x i64> poison, i64 1, i32 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer))
+ ret <vscale x 2 x i64> %x
+}
+
+define <vscale x 2 x i64> @vror_vi_rotl_nxv2i64(<vscale x 2 x i64> %a) {
+; CHECK-RV32-LABEL: vror_vi_rotl_nxv2i64:
+; CHECK-RV32: # %bb.0:
+; CHECK-RV32-NEXT: li a0, 63
+; CHECK-RV32-NEXT: vsetvli a1, zero, e64, m2, ta, ma
+; CHECK-RV32-NEXT: vmv.v.x v10, a0
+; CHECK-RV32-NEXT: vand.vi v10, v10, 1
+; CHECK-RV32-NEXT: vsll.vv v10, v8, v10
+; CHECK-RV32-NEXT: vmv.v.i v12, 1
+; CHECK-RV32-NEXT: vrsub.vi v12, v12, 0
+; CHECK-RV32-NEXT: vand.vx v12, v12, a0
+; CHECK-RV32-NEXT: vsrl.vv v8, v8, v12
+; CHECK-RV32-NEXT: vor.vv v8, v10, v8
+; CHECK-RV32-NEXT: ret
+;
+; CHECK-RV64-LABEL: vror_vi_rotl_nxv2i64:
+; CHECK-RV64: # %bb.0:
+; CHECK-RV64-NEXT: li a0, 63
+; CHECK-RV64-NEXT: vsetvli a1, zero, e64, m2, ta, ma
+; CHECK-RV64-NEXT: vsrl.vx v10, v8, a0
+; CHECK-RV64-NEXT: vadd.vv v8, v8, v8
+; CHECK-RV64-NEXT: vor.vv v8, v8, v10
+; CHECK-RV64-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_rotl_nxv2i64:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e64, m2, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 63
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 2 x i64> @llvm.fshl.nxv2i64(<vscale x 2 x i64> %a, <vscale x 2 x i64> %a, <vscale x 2 x i64> shufflevector(<vscale x 2 x i64> insertelement(<vscale x 2 x i64> poison, i64 1, i32 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer))
+ ret <vscale x 2 x i64> %x
+}
+
+declare <vscale x 4 x i64> @llvm.fshr.nxv4i64(<vscale x 4 x i64>, <vscale x 4 x i64>, <vscale x 4 x i64>)
+declare <vscale x 4 x i64> @llvm.fshl.nxv4i64(<vscale x 4 x i64>, <vscale x 4 x i64>, <vscale x 4 x i64>)
+
+define <vscale x 4 x i64> @vror_vv_nxv4i64(<vscale x 4 x i64> %a, <vscale x 4 x i64> %b) {
+; CHECK-LABEL: vror_vv_nxv4i64:
+; CHECK: # %bb.0:
+; CHECK-NEXT: li a0, 63
+; CHECK-NEXT: vsetvli a1, zero, e64, m4, ta, ma
+; CHECK-NEXT: vand.vx v16, v12, a0
+; CHECK-NEXT: vsrl.vv v16, v8, v16
+; CHECK-NEXT: vrsub.vi v12, v12, 0
+; CHECK-NEXT: vand.vx v12, v12, a0
+; CHECK-NEXT: vsll.vv v8, v8, v12
+; CHECK-NEXT: vor.vv v8, v16, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vv_nxv4i64:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e64, m4, ta, ma
+; CHECK-ZVBB-NEXT: vror.vv v8, v8, v12
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 4 x i64> @llvm.fshr.nxv4i64(<vscale x 4 x i64> %a, <vscale x 4 x i64> %a, <vscale x 4 x i64> %b)
+ ret <vscale x 4 x i64> %x
+}
+
+define <vscale x 4 x i64> @vror_vx_nxv4i64(<vscale x 4 x i64> %a, i64 %b) {
+; CHECK-RV32-LABEL: vror_vx_nxv4i64:
+; CHECK-RV32: # %bb.0:
+; CHECK-RV32-NEXT: addi sp, sp, -16
+; CHECK-RV32-NEXT: .cfi_def_cfa_offset 16
+; CHECK-RV32-NEXT: sw a1, 12(sp)
+; CHECK-RV32-NEXT: sw a0, 8(sp)
+; CHECK-RV32-NEXT: addi a0, sp, 8
+; CHECK-RV32-NEXT: vsetvli a1, zero, e64, m4, ta, ma
+; CHECK-RV32-NEXT: vlse64.v v12, (a0), zero
+; CHECK-RV32-NEXT: li a0, 63
+; CHECK-RV32-NEXT: vand.vx v16, v12, a0
+; CHECK-RV32-NEXT: vsrl.vv v16, v8, v16
+; CHECK-RV32-NEXT: vrsub.vi v12, v12, 0
+; CHECK-RV32-NEXT: vand.vx v12, v12, a0
+; CHECK-RV32-NEXT: vsll.vv v8, v8, v12
+; CHECK-RV32-NEXT: vor.vv v8, v16, v8
+; CHECK-RV32-NEXT: addi sp, sp, 16
+; CHECK-RV32-NEXT: ret
+;
+; CHECK-RV64-LABEL: vror_vx_nxv4i64:
+; CHECK-RV64: # %bb.0:
+; CHECK-RV64-NEXT: andi a1, a0, 63
+; CHECK-RV64-NEXT: vsetvli a2, zero, e64, m4, ta, ma
+; CHECK-RV64-NEXT: vsrl.vx v12, v8, a1
+; CHECK-RV64-NEXT: negw a0, a0
+; CHECK-RV64-NEXT: andi a0, a0, 63
+; CHECK-RV64-NEXT: vsll.vx v8, v8, a0
+; CHECK-RV64-NEXT: vor.vv v8, v12, v8
+; CHECK-RV64-NEXT: ret
+;
+; CHECK-ZVBB32-LABEL: vror_vx_nxv4i64:
+; CHECK-ZVBB32: # %bb.0:
+; CHECK-ZVBB32-NEXT: addi sp, sp, -16
+; CHECK-ZVBB32-NEXT: .cfi_def_cfa_offset 16
+; CHECK-ZVBB32-NEXT: sw a1, 12(sp)
+; CHECK-ZVBB32-NEXT: sw a0, 8(sp)
+; CHECK-ZVBB32-NEXT: addi a0, sp, 8
+; CHECK-ZVBB32-NEXT: vsetvli a1, zero, e64, m4, ta, ma
+; CHECK-ZVBB32-NEXT: vlse64.v v12, (a0), zero
+; CHECK-ZVBB32-NEXT: vror.vv v8, v8, v12
+; CHECK-ZVBB32-NEXT: addi sp, sp, 16
+; CHECK-ZVBB32-NEXT: ret
+;
+; CHECK-ZVBB64-LABEL: vror_vx_nxv4i64:
+; CHECK-ZVBB64: # %bb.0:
+; CHECK-ZVBB64-NEXT: vsetvli a1, zero, e64, m4, ta, ma
+; CHECK-ZVBB64-NEXT: vror.vx v8, v8, a0
+; CHECK-ZVBB64-NEXT: ret
+ %b.head = insertelement <vscale x 4 x i64> poison, i64 %b, i32 0
+ %b.splat = shufflevector <vscale x 4 x i64> %b.head, <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer
+ %x = call <vscale x 4 x i64> @llvm.fshr.nxv4i64(<vscale x 4 x i64> %a, <vscale x 4 x i64> %a, <vscale x 4 x i64> %b.splat)
+ ret <vscale x 4 x i64> %x
+}
+
+define <vscale x 4 x i64> @vror_vi_nxv4i64(<vscale x 4 x i64> %a) {
+; CHECK-RV32-LABEL: vror_vi_nxv4i64:
+; CHECK-RV32: # %bb.0:
+; CHECK-RV32-NEXT: li a0, 63
+; CHECK-RV32-NEXT: vsetvli a1, zero, e64, m4, ta, ma
+; CHECK-RV32-NEXT: vmv.v.x v12, a0
+; CHECK-RV32-NEXT: vand.vi v12, v12, 1
+; CHECK-RV32-NEXT: vsrl.vv v12, v8, v12
+; CHECK-RV32-NEXT: vmv.v.i v16, 1
+; CHECK-RV32-NEXT: vrsub.vi v16, v16, 0
+; CHECK-RV32-NEXT: vand.vx v16, v16, a0
+; CHECK-RV32-NEXT: vsll.vv v8, v8, v16
+; CHECK-RV32-NEXT: vor.vv v8, v12, v8
+; CHECK-RV32-NEXT: ret
+;
+; CHECK-RV64-LABEL: vror_vi_nxv4i64:
+; CHECK-RV64: # %bb.0:
+; CHECK-RV64-NEXT: li a0, 63
+; CHECK-RV64-NEXT: vsetvli a1, zero, e64, m4, ta, ma
+; CHECK-RV64-NEXT: vsll.vx v12, v8, a0
+; CHECK-RV64-NEXT: vsrl.vi v8, v8, 1
+; CHECK-RV64-NEXT: vor.vv v8, v8, v12
+; CHECK-RV64-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_nxv4i64:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e64, m4, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 1
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 4 x i64> @llvm.fshr.nxv4i64(<vscale x 4 x i64> %a, <vscale x 4 x i64> %a, <vscale x 4 x i64> shufflevector(<vscale x 4 x i64> insertelement(<vscale x 4 x i64> poison, i64 1, i32 0), <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer))
+ ret <vscale x 4 x i64> %x
+}
+
+define <vscale x 4 x i64> @vror_vi_rotl_nxv4i64(<vscale x 4 x i64> %a) {
+; CHECK-RV32-LABEL: vror_vi_rotl_nxv4i64:
+; CHECK-RV32: # %bb.0:
+; CHECK-RV32-NEXT: li a0, 63
+; CHECK-RV32-NEXT: vsetvli a1, zero, e64, m4, ta, ma
+; CHECK-RV32-NEXT: vmv.v.x v12, a0
+; CHECK-RV32-NEXT: vand.vi v12, v12, 1
+; CHECK-RV32-NEXT: vsll.vv v12, v8, v12
+; CHECK-RV32-NEXT: vmv.v.i v16, 1
+; CHECK-RV32-NEXT: vrsub.vi v16, v16, 0
+; CHECK-RV32-NEXT: vand.vx v16, v16, a0
+; CHECK-RV32-NEXT: vsrl.vv v8, v8, v16
+; CHECK-RV32-NEXT: vor.vv v8, v12, v8
+; CHECK-RV32-NEXT: ret
+;
+; CHECK-RV64-LABEL: vror_vi_rotl_nxv4i64:
+; CHECK-RV64: # %bb.0:
+; CHECK-RV64-NEXT: li a0, 63
+; CHECK-RV64-NEXT: vsetvli a1, zero, e64, m4, ta, ma
+; CHECK-RV64-NEXT: vsrl.vx v12, v8, a0
+; CHECK-RV64-NEXT: vadd.vv v8, v8, v8
+; CHECK-RV64-NEXT: vor.vv v8, v8, v12
+; CHECK-RV64-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_rotl_nxv4i64:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e64, m4, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 63
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 4 x i64> @llvm.fshl.nxv4i64(<vscale x 4 x i64> %a, <vscale x 4 x i64> %a, <vscale x 4 x i64> shufflevector(<vscale x 4 x i64> insertelement(<vscale x 4 x i64> poison, i64 1, i32 0), <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer))
+ ret <vscale x 4 x i64> %x
+}
+
+declare <vscale x 8 x i64> @llvm.fshr.nxv8i64(<vscale x 8 x i64>, <vscale x 8 x i64>, <vscale x 8 x i64>)
+declare <vscale x 8 x i64> @llvm.fshl.nxv8i64(<vscale x 8 x i64>, <vscale x 8 x i64>, <vscale x 8 x i64>)
+
+define <vscale x 8 x i64> @vror_vv_nxv8i64(<vscale x 8 x i64> %a, <vscale x 8 x i64> %b) {
+; CHECK-LABEL: vror_vv_nxv8i64:
+; CHECK: # %bb.0:
+; CHECK-NEXT: li a0, 63
+; CHECK-NEXT: vsetvli a1, zero, e64, m8, ta, ma
+; CHECK-NEXT: vand.vx v24, v16, a0
+; CHECK-NEXT: vsrl.vv v24, v8, v24
+; CHECK-NEXT: vrsub.vi v16, v16, 0
+; CHECK-NEXT: vand.vx v16, v16, a0
+; CHECK-NEXT: vsll.vv v8, v8, v16
+; CHECK-NEXT: vor.vv v8, v24, v8
+; CHECK-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vv_nxv8i64:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e64, m8, ta, ma
+; CHECK-ZVBB-NEXT: vror.vv v8, v8, v16
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 8 x i64> @llvm.fshr.nxv8i64(<vscale x 8 x i64> %a, <vscale x 8 x i64> %a, <vscale x 8 x i64> %b)
+ ret <vscale x 8 x i64> %x
+}
+
+define <vscale x 8 x i64> @vror_vx_nxv8i64(<vscale x 8 x i64> %a, i64 %b) {
+; CHECK-RV32-LABEL: vror_vx_nxv8i64:
+; CHECK-RV32: # %bb.0:
+; CHECK-RV32-NEXT: addi sp, sp, -16
+; CHECK-RV32-NEXT: .cfi_def_cfa_offset 16
+; CHECK-RV32-NEXT: sw a1, 12(sp)
+; CHECK-RV32-NEXT: sw a0, 8(sp)
+; CHECK-RV32-NEXT: addi a0, sp, 8
+; CHECK-RV32-NEXT: vsetvli a1, zero, e64, m8, ta, ma
+; CHECK-RV32-NEXT: vlse64.v v16, (a0), zero
+; CHECK-RV32-NEXT: li a0, 63
+; CHECK-RV32-NEXT: vand.vx v24, v16, a0
+; CHECK-RV32-NEXT: vsrl.vv v24, v8, v24
+; CHECK-RV32-NEXT: vrsub.vi v16, v16, 0
+; CHECK-RV32-NEXT: vand.vx v16, v16, a0
+; CHECK-RV32-NEXT: vsll.vv v8, v8, v16
+; CHECK-RV32-NEXT: vor.vv v8, v24, v8
+; CHECK-RV32-NEXT: addi sp, sp, 16
+; CHECK-RV32-NEXT: ret
+;
+; CHECK-RV64-LABEL: vror_vx_nxv8i64:
+; CHECK-RV64: # %bb.0:
+; CHECK-RV64-NEXT: andi a1, a0, 63
+; CHECK-RV64-NEXT: vsetvli a2, zero, e64, m8, ta, ma
+; CHECK-RV64-NEXT: vsrl.vx v16, v8, a1
+; CHECK-RV64-NEXT: negw a0, a0
+; CHECK-RV64-NEXT: andi a0, a0, 63
+; CHECK-RV64-NEXT: vsll.vx v8, v8, a0
+; CHECK-RV64-NEXT: vor.vv v8, v16, v8
+; CHECK-RV64-NEXT: ret
+;
+; CHECK-ZVBB32-LABEL: vror_vx_nxv8i64:
+; CHECK-ZVBB32: # %bb.0:
+; CHECK-ZVBB32-NEXT: addi sp, sp, -16
+; CHECK-ZVBB32-NEXT: .cfi_def_cfa_offset 16
+; CHECK-ZVBB32-NEXT: sw a1, 12(sp)
+; CHECK-ZVBB32-NEXT: sw a0, 8(sp)
+; CHECK-ZVBB32-NEXT: addi a0, sp, 8
+; CHECK-ZVBB32-NEXT: vsetvli a1, zero, e64, m8, ta, ma
+; CHECK-ZVBB32-NEXT: vlse64.v v16, (a0), zero
+; CHECK-ZVBB32-NEXT: vror.vv v8, v8, v16
+; CHECK-ZVBB32-NEXT: addi sp, sp, 16
+; CHECK-ZVBB32-NEXT: ret
+;
+; CHECK-ZVBB64-LABEL: vror_vx_nxv8i64:
+; CHECK-ZVBB64: # %bb.0:
+; CHECK-ZVBB64-NEXT: vsetvli a1, zero, e64, m8, ta, ma
+; CHECK-ZVBB64-NEXT: vror.vx v8, v8, a0
+; CHECK-ZVBB64-NEXT: ret
+ %b.head = insertelement <vscale x 8 x i64> poison, i64 %b, i32 0
+ %b.splat = shufflevector <vscale x 8 x i64> %b.head, <vscale x 8 x i64> poison, <vscale x 8 x i32> zeroinitializer
+ %x = call <vscale x 8 x i64> @llvm.fshr.nxv8i64(<vscale x 8 x i64> %a, <vscale x 8 x i64> %a, <vscale x 8 x i64> %b.splat)
+ ret <vscale x 8 x i64> %x
+}
+
+define <vscale x 8 x i64> @vror_vi_nxv8i64(<vscale x 8 x i64> %a) {
+; CHECK-RV32-LABEL: vror_vi_nxv8i64:
+; CHECK-RV32: # %bb.0:
+; CHECK-RV32-NEXT: li a0, 63
+; CHECK-RV32-NEXT: vsetvli a1, zero, e64, m8, ta, ma
+; CHECK-RV32-NEXT: vmv.v.x v16, a0
+; CHECK-RV32-NEXT: vand.vi v16, v16, 1
+; CHECK-RV32-NEXT: vsrl.vv v16, v8, v16
+; CHECK-RV32-NEXT: vmv.v.i v24, 1
+; CHECK-RV32-NEXT: vrsub.vi v24, v24, 0
+; CHECK-RV32-NEXT: vand.vx v24, v24, a0
+; CHECK-RV32-NEXT: vsll.vv v8, v8, v24
+; CHECK-RV32-NEXT: vor.vv v8, v16, v8
+; CHECK-RV32-NEXT: ret
+;
+; CHECK-RV64-LABEL: vror_vi_nxv8i64:
+; CHECK-RV64: # %bb.0:
+; CHECK-RV64-NEXT: li a0, 63
+; CHECK-RV64-NEXT: vsetvli a1, zero, e64, m8, ta, ma
+; CHECK-RV64-NEXT: vsll.vx v16, v8, a0
+; CHECK-RV64-NEXT: vsrl.vi v8, v8, 1
+; CHECK-RV64-NEXT: vor.vv v8, v8, v16
+; CHECK-RV64-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_nxv8i64:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e64, m8, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 1
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 8 x i64> @llvm.fshr.nxv8i64(<vscale x 8 x i64> %a, <vscale x 8 x i64> %a, <vscale x 8 x i64> shufflevector(<vscale x 8 x i64> insertelement(<vscale x 8 x i64> poison, i64 1, i32 0), <vscale x 8 x i64> poison, <vscale x 8 x i32> zeroinitializer))
+ ret <vscale x 8 x i64> %x
+}
+
+define <vscale x 8 x i64> @vror_vi_rotl_nxv8i64(<vscale x 8 x i64> %a) {
+; CHECK-RV32-LABEL: vror_vi_rotl_nxv8i64:
+; CHECK-RV32: # %bb.0:
+; CHECK-RV32-NEXT: li a0, 63
+; CHECK-RV32-NEXT: vsetvli a1, zero, e64, m8, ta, ma
+; CHECK-RV32-NEXT: vmv.v.x v16, a0
+; CHECK-RV32-NEXT: vand.vi v16, v16, 1
+; CHECK-RV32-NEXT: vsll.vv v16, v8, v16
+; CHECK-RV32-NEXT: vmv.v.i v24, 1
+; CHECK-RV32-NEXT: vrsub.vi v24, v24, 0
+; CHECK-RV32-NEXT: vand.vx v24, v24, a0
+; CHECK-RV32-NEXT: vsrl.vv v8, v8, v24
+; CHECK-RV32-NEXT: vor.vv v8, v16, v8
+; CHECK-RV32-NEXT: ret
+;
+; CHECK-RV64-LABEL: vror_vi_rotl_nxv8i64:
+; CHECK-RV64: # %bb.0:
+; CHECK-RV64-NEXT: li a0, 63
+; CHECK-RV64-NEXT: vsetvli a1, zero, e64, m8, ta, ma
+; CHECK-RV64-NEXT: vsrl.vx v16, v8, a0
+; CHECK-RV64-NEXT: vadd.vv v8, v8, v8
+; CHECK-RV64-NEXT: vor.vv v8, v8, v16
+; CHECK-RV64-NEXT: ret
+;
+; CHECK-ZVBB-LABEL: vror_vi_rotl_nxv8i64:
+; CHECK-ZVBB: # %bb.0:
+; CHECK-ZVBB-NEXT: vsetvli a0, zero, e64, m8, ta, ma
+; CHECK-ZVBB-NEXT: vror.vi v8, v8, 63
+; CHECK-ZVBB-NEXT: ret
+ %x = call <vscale x 8 x i64> @llvm.fshl.nxv8i64(<vscale x 8 x i64> %a, <vscale x 8 x i64> %a, <vscale x 8 x i64> shufflevector(<vscale x 8 x i64> insertelement(<vscale x 8 x i64> poison, i64 1, i32 0), <vscale x 8 x i64> poison, <vscale x 8 x i32> zeroinitializer))
+ ret <vscale x 8 x i64> %x
+}
+