| ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py |
| ; RUN: llc -verify-machineinstrs < %s | FileCheck %s |
| |
| target triple = "aarch64-linux-gnu" |
| |
| define <vscale x 8 x half> @test_fclamp_f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c) #0 { |
| ; CHECK-LABEL: test_fclamp_f16: |
| ; CHECK: // %bb.0: |
| ; CHECK-NEXT: fclamp z0.h, z1.h, z2.h |
| ; CHECK-NEXT: ret |
| %res = call <vscale x 8 x half> @llvm.aarch64.sve.fclamp.nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c) |
| ret <vscale x 8 x half> %res |
| } |
| |
| define <vscale x 4 x float> @test_fclamp_f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b, <vscale x 4 x float> %c) #0 { |
| ; CHECK-LABEL: test_fclamp_f32: |
| ; CHECK: // %bb.0: |
| ; CHECK-NEXT: fclamp z0.s, z1.s, z2.s |
| ; CHECK-NEXT: ret |
| %res = call <vscale x 4 x float> @llvm.aarch64.sve.fclamp.nxv4f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b, <vscale x 4 x float> %c) |
| ret <vscale x 4 x float> %res |
| } |
| |
| define <vscale x 2 x double> @test_fclamp_f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b, <vscale x 2 x double> %c) #0 { |
| ; CHECK-LABEL: test_fclamp_f64: |
| ; CHECK: // %bb.0: |
| ; CHECK-NEXT: fclamp z0.d, z1.d, z2.d |
| ; CHECK-NEXT: ret |
| %res = call <vscale x 2 x double> @llvm.aarch64.sve.fclamp.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b, <vscale x 2 x double> %c) |
| ret <vscale x 2 x double> %res |
| } |
| |
| define { <vscale x 8 x half>, <vscale x 8 x half> } @test_fclamp_single_x2_f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c, <vscale x 8 x half> %d) #1 { |
| ; CHECK-LABEL: test_fclamp_single_x2_f16: |
| ; CHECK: // %bb.0: |
| ; CHECK-NEXT: // kill: def $z1 killed $z1 killed $z0_z1 def $z0_z1 |
| ; CHECK-NEXT: // kill: def $z0 killed $z0 killed $z0_z1 def $z0_z1 |
| ; CHECK-NEXT: fclamp { z0.h, z1.h }, z2.h, z3.h |
| ; CHECK-NEXT: ret |
| %res = call { <vscale x 8 x half>, <vscale x 8 x half> } @llvm.aarch64.sve.fclamp.single.x2.nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c, <vscale x 8 x half> %d) |
| ret { <vscale x 8 x half>, <vscale x 8 x half> } %res |
| } |
| |
| define { <vscale x 4 x float>, <vscale x 4 x float> } @test_fclamp_single_x2_f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b, <vscale x 4 x float> %c, <vscale x 4 x float> %d) #1 { |
| ; CHECK-LABEL: test_fclamp_single_x2_f32: |
| ; CHECK: // %bb.0: |
| ; CHECK-NEXT: // kill: def $z1 killed $z1 killed $z0_z1 def $z0_z1 |
| ; CHECK-NEXT: // kill: def $z0 killed $z0 killed $z0_z1 def $z0_z1 |
| ; CHECK-NEXT: fclamp { z0.s, z1.s }, z2.s, z3.s |
| ; CHECK-NEXT: ret |
| %res = call { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.aarch64.sve.fclamp.single.x2.nxv4f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b, <vscale x 4 x float> %c, <vscale x 4 x float> %d) |
| ret { <vscale x 4 x float>, <vscale x 4 x float> } %res |
| } |
| |
| define { <vscale x 2 x double>, <vscale x 2 x double> } @test_fclamp_single_x2_f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b, <vscale x 2 x double> %c, <vscale x 2 x double> %d) #1 { |
| ; CHECK-LABEL: test_fclamp_single_x2_f64: |
| ; CHECK: // %bb.0: |
| ; CHECK-NEXT: // kill: def $z1 killed $z1 killed $z0_z1 def $z0_z1 |
| ; CHECK-NEXT: // kill: def $z0 killed $z0 killed $z0_z1 def $z0_z1 |
| ; CHECK-NEXT: fclamp { z0.d, z1.d }, z2.d, z3.d |
| ; CHECK-NEXT: ret |
| %res = call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.aarch64.sve.fclamp.single.x2.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b, <vscale x 2 x double> %c, <vscale x 2 x double> %d) |
| ret { <vscale x 2 x double>, <vscale x 2 x double> } %res |
| } |
| |
| |
| define { <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half> } @test_fclamp_single_x4_f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c, <vscale x 8 x half> %d, <vscale x 8 x half> %e, <vscale x 8 x half> %f) #1 { |
| ; CHECK-LABEL: test_fclamp_single_x4_f16: |
| ; CHECK: // %bb.0: |
| ; CHECK-NEXT: // kill: def $z3 killed $z3 killed $z0_z1_z2_z3 def $z0_z1_z2_z3 |
| ; CHECK-NEXT: // kill: def $z2 killed $z2 killed $z0_z1_z2_z3 def $z0_z1_z2_z3 |
| ; CHECK-NEXT: // kill: def $z1 killed $z1 killed $z0_z1_z2_z3 def $z0_z1_z2_z3 |
| ; CHECK-NEXT: // kill: def $z0 killed $z0 killed $z0_z1_z2_z3 def $z0_z1_z2_z3 |
| ; CHECK-NEXT: fclamp { z0.h - z3.h }, z4.h, z5.h |
| ; CHECK-NEXT: ret |
| %res = call { <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half> } @llvm.aarch64.sve.fclamp.single.x4.nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c, <vscale x 8 x half> %d, <vscale x 8 x half> %e, <vscale x 8 x half> %f) |
| ret { <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half> } %res |
| } |
| |
| define { <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float> } @test_fclamp_single_x4_f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b, <vscale x 4 x float> %c, <vscale x 4 x float> %d, <vscale x 4 x float> %e, <vscale x 4 x float> %f) #1 { |
| ; CHECK-LABEL: test_fclamp_single_x4_f32: |
| ; CHECK: // %bb.0: |
| ; CHECK-NEXT: // kill: def $z3 killed $z3 killed $z0_z1_z2_z3 def $z0_z1_z2_z3 |
| ; CHECK-NEXT: // kill: def $z2 killed $z2 killed $z0_z1_z2_z3 def $z0_z1_z2_z3 |
| ; CHECK-NEXT: // kill: def $z1 killed $z1 killed $z0_z1_z2_z3 def $z0_z1_z2_z3 |
| ; CHECK-NEXT: // kill: def $z0 killed $z0 killed $z0_z1_z2_z3 def $z0_z1_z2_z3 |
| ; CHECK-NEXT: fclamp { z0.s - z3.s }, z4.s, z5.s |
| ; CHECK-NEXT: ret |
| %res = call { <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float> } @llvm.aarch64.sve.fclamp.single.x4.nxv4f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b, <vscale x 4 x float> %c, <vscale x 4 x float> %d, <vscale x 4 x float> %e, <vscale x 4 x float> %f) |
| ret { <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float> } %res |
| } |
| |
| define { <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double> } @test_fclamp_single_x4_f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b, <vscale x 2 x double> %c, <vscale x 2 x double> %d, <vscale x 2 x double> %e, <vscale x 2 x double> %f) #1 { |
| ; CHECK-LABEL: test_fclamp_single_x4_f64: |
| ; CHECK: // %bb.0: |
| ; CHECK-NEXT: // kill: def $z3 killed $z3 killed $z0_z1_z2_z3 def $z0_z1_z2_z3 |
| ; CHECK-NEXT: // kill: def $z2 killed $z2 killed $z0_z1_z2_z3 def $z0_z1_z2_z3 |
| ; CHECK-NEXT: // kill: def $z1 killed $z1 killed $z0_z1_z2_z3 def $z0_z1_z2_z3 |
| ; CHECK-NEXT: // kill: def $z0 killed $z0 killed $z0_z1_z2_z3 def $z0_z1_z2_z3 |
| ; CHECK-NEXT: fclamp { z0.d - z3.d }, z4.d, z5.d |
| ; CHECK-NEXT: ret |
| %res = call { <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double> } @llvm.aarch64.sve.fclamp.single.x4.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b, <vscale x 2 x double> %c, <vscale x 2 x double> %d, <vscale x 2 x double> %e, <vscale x 2 x double> %f) |
| ret { <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double> } %res |
| } |
| |
| |
| attributes #0 = { "target-features"="+sve2p1" } |
| attributes #1 = { "target-features"="+sme2" } |
| |
| declare <vscale x 8 x half> @llvm.aarch64.sve.fclamp.nxv8f16(<vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half>) |
| declare <vscale x 4 x float> @llvm.aarch64.sve.fclamp.nxv4f32(<vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float>) |
| declare <vscale x 2 x double> @llvm.aarch64.sve.fclamp.nxv2f64(<vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double>) |
| |
| declare { <vscale x 8 x half>, <vscale x 8 x half> } @llvm.aarch64.sve.fclamp.single.x2.nxv8f16(<vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half>) |
| declare { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.aarch64.sve.fclamp.single.x2.nxv4f32(<vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float>) |
| declare { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.aarch64.sve.fclamp.single.x2.nxv2f64(<vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double>) |
| |
| declare { <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half> } @llvm.aarch64.sve.fclamp.single.x4.nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c, <vscale x 8 x half> %d, <vscale x 8 x half> %e, <vscale x 8 x half> %f) |
| declare { <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float> } @llvm.aarch64.sve.fclamp.single.x4.nxv4f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b, <vscale x 4 x float> %c, <vscale x 4 x float> %d, <vscale x 4 x float> %e, <vscale x 4 x float> %f) |
| declare { <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double> } @llvm.aarch64.sve.fclamp.single.x4.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b, <vscale x 2 x double> %c, <vscale x 2 x double> %d, <vscale x 2 x double> %e, <vscale x 2 x double> %f) |