AMDGPU: Remove llvm.AMDGPU.kill

This is the last of the old AMDGPU intrinsics.

llvm-svn: 348615

5 files changed, 56 insertions, 82 deletions

diff --git a/llvm/test/CodeGen/AMDGPU/default-fp-mode.ll b/llvm/test/CodeGen/AMDGPU/default-fp-mode.ll
index 6867afee170..94d518e63ce 100644
--- a/llvm/test/CodeGen/AMDGPU/default-fp-mode.ll
+++ b/llvm/test/CodeGen/AMDGPU/default-fp-mode.ll
@@ -86,12 +86,10 @@ define amdgpu_kernel void @test_f32_f16_f64_denormals(half addrspace(1)* %out0,
 ; GCN: IeeeMode: 0
 define amdgpu_gs void @kill_gs_const() {
 main_body:
-  %0 = icmp ule i32 0, 3
-  %1 = select i1 %0, float 1.000000e+00, float -1.000000e+00
-  call void @llvm.AMDGPU.kill(float %1)
-  %2 = icmp ule i32 3, 0
-  %3 = select i1 %2, float 1.000000e+00, float -1.000000e+00
-  call void @llvm.AMDGPU.kill(float %3)
+  %cmp0 = icmp ule i32 0, 3
+  call void @llvm.amdgcn.kill(i1 %cmp0)
+  %cmp1 = icmp ule i32 3, 0
+  call void @llvm.amdgcn.kill(i1 %cmp1)
   ret void
 }
 
@@ -100,12 +98,12 @@ main_body:
 define amdgpu_ps float @kill_vcc_implicit_def([6 x <16 x i8>] addrspace(4)* byval, [17 x <16 x i8>] addrspace(4)* byval, [17 x <4 x i32>] addrspace(4)* byval, [34 x <8 x i32>] addrspace(4)* byval, float inreg, i32 inreg, <2 x i32>, <2 x i32>, <2 x i32>, <3 x i32>, <2 x i32>, <2 x i32>, <2 x i32>, float, float, float, float, float, float, i32, float, float) {
 entry:
   %tmp0 = fcmp olt float %13, 0.0
-  call void @llvm.AMDGPU.kill(float %14)
+  call void @llvm.amdgcn.kill(i1 %tmp0)
   %tmp1 = select i1 %tmp0, float 1.0, float 0.0
   ret float %tmp1
 }
 
-declare void @llvm.AMDGPU.kill(float)
+declare void @llvm.amdgcn.kill(i1)
 
 attributes #0 = { nounwind "target-cpu"="tahiti" }
 attributes #1 = { nounwind "target-cpu"="fiji" }
diff --git a/llvm/test/CodeGen/AMDGPU/llvm.AMDGPU.kill.ll b/llvm/test/CodeGen/AMDGPU/llvm.AMDGPU.kill.ll
deleted file mode 100644
index 6374c9e02ed..00000000000
--- a/llvm/test/CodeGen/AMDGPU/llvm.AMDGPU.kill.ll
+++ /dev/null
@@ -1,35 +0,0 @@
-; RUN: llc -march=amdgcn -mcpu=verde -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s
-; RUN: llc -march=amdgcn -mcpu=tonga -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s
-
-; SI-LABEL: {{^}}kill_gs_const:
-; SI-NOT: v_cmpx_le_f32
-; SI: s_mov_b64 exec, 0
-define amdgpu_gs void @kill_gs_const() {
-main_body:
-  %tmp = icmp ule i32 0, 3
-  %tmp1 = select i1 %tmp, float 1.000000e+00, float -1.000000e+00
-  call void @llvm.AMDGPU.kill(float %tmp1)
-  %tmp2 = icmp ule i32 3, 0
-  %tmp3 = select i1 %tmp2, float 1.000000e+00, float -1.000000e+00
-  call void @llvm.AMDGPU.kill(float %tmp3)
-  ret void
-}
-
-; SI-LABEL: {{^}}kill_vcc_implicit_def:
-; SI-NOT: v_cmp_gt_f32_e32 vcc,
-; SI: v_cmp_gt_f32_e64 [[CMP:s\[[0-9]+:[0-9]+\]]], 0, v{{[0-9]+}}
-; SI: v_cmpx_le_f32_e32 vcc, 0, v{{[0-9]+}}
-; SI: v_cndmask_b32_e64 v{{[0-9]+}}, 0, 1.0, [[CMP]]
-define amdgpu_ps void @kill_vcc_implicit_def([6 x <16 x i8>] addrspace(4)* byval %arg, [17 x <16 x i8>] addrspace(4)* byval %arg1, [17 x <4 x i32>] addrspace(4)* byval %arg2, [34 x <8 x i32>] addrspace(4)* byval %arg3, float inreg %arg4, i32 inreg %arg5, <2 x i32> %arg6, <2 x i32> %arg7, <2 x i32> %arg8, <3 x i32> %arg9, <2 x i32> %arg10, <2 x i32> %arg11, <2 x i32> %arg12, float %arg13, float %arg14, float %arg15, float %arg16, float %arg17, float %arg18, i32 %arg19, float %arg20, float %arg21) {
-entry:
-  %tmp0 = fcmp olt float %arg13, 0.000000e+00
-  call void @llvm.AMDGPU.kill(float %arg14)
-  %tmp1 = select i1 %tmp0, float 1.000000e+00, float 0.000000e+00
-  call void @llvm.amdgcn.exp.f32(i32 1, i32 15, float %tmp1, float %tmp1, float %tmp1, float %tmp1, i1 true, i1 true) #0
-  ret void
-}
-
-declare void @llvm.AMDGPU.kill(float) #0
-declare void @llvm.amdgcn.exp.f32(i32, i32, float, float, float, float, i1, i1) #0
-
-attributes #0 = { nounwind }
diff --git a/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.wqm.vote.ll b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.wqm.vote.ll
index 1946e6a3618..57af85b67bf 100644
--- a/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.wqm.vote.ll
+++ b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.wqm.vote.ll
@@ -33,20 +33,17 @@ main_body:
 ;CHECK-LABEL: {{^}}kill:
 ;CHECK: v_cmp_eq_u32_e32 [[CMP:[^,]+]], v0, v1
 ;CHECK: s_wqm_b64 [[WQM:[^,]+]], [[CMP]]
-;FIXME: This could just be: s_and_b64 exec, exec, [[WQM]]
-;CHECK: v_cndmask_b32_e64 [[KILL:[^,]+]], -1.0, 1.0, [[WQM]]
-;CHECK: v_cmpx_le_f32_e32 {{[^,]+}}, 0, [[KILL]]
+;CHECK: s_and_b64 exec, exec, [[WQM]]
 ;CHECK: s_endpgm
 define amdgpu_ps void @kill(i32 %v0, i32 %v1) #1 {
 main_body:
   %c = icmp eq i32 %v0, %v1
   %w = call i1 @llvm.amdgcn.wqm.vote(i1 %c)
-  %r = select i1 %w, float 1.0, float -1.0
-  call void @llvm.AMDGPU.kill(float %r)
+  call void @llvm.amdgcn.kill(i1 %w)
   ret void
 }
 
-declare void @llvm.AMDGPU.kill(float) #1
+declare void @llvm.amdgcn.kill(i1) #1
 declare i1 @llvm.amdgcn.wqm.vote(i1)
 
 attributes #1 = { nounwind }
diff --git a/llvm/test/CodeGen/AMDGPU/skip-if-dead.ll b/llvm/test/CodeGen/AMDGPU/skip-if-dead.ll
index 42a28b95273..a02e6f8349c 100644
--- a/llvm/test/CodeGen/AMDGPU/skip-if-dead.ll
+++ b/llvm/test/CodeGen/AMDGPU/skip-if-dead.ll
@@ -2,9 +2,10 @@
 
 ; CHECK-LABEL: {{^}}test_kill_depth_0_imm_pos:
 ; CHECK-NEXT: ; %bb.0:
+; CHECK-NEXT: ; %bb.1:
 ; CHECK-NEXT: s_endpgm
 define amdgpu_ps void @test_kill_depth_0_imm_pos() #0 {
-  call void @llvm.AMDGPU.kill(float 0.0)
+  call void @llvm.amdgcn.kill(i1 true)
   ret void
 }
 
@@ -14,7 +15,7 @@ define amdgpu_ps void @test_kill_depth_0_imm_pos() #0 {
 ; CHECK-NEXT: ; %bb.1:
 ; CHECK-NEXT: s_endpgm
 define amdgpu_ps void @test_kill_depth_0_imm_neg() #0 {
-  call void @llvm.AMDGPU.kill(float -0.0)
+  call void @llvm.amdgcn.kill(i1 false)
   ret void
 }
 
@@ -27,58 +28,62 @@ define amdgpu_ps void @test_kill_depth_0_imm_neg() #0 {
 ; CHECK-NEXT: ; %bb.2:
 ; CHECK-NEXT: s_endpgm
 define amdgpu_ps void @test_kill_depth_0_imm_neg_x2() #0 {
-  call void @llvm.AMDGPU.kill(float -0.0)
-  call void @llvm.AMDGPU.kill(float -1.0)
+  call void @llvm.amdgcn.kill(i1 false)
+  call void @llvm.amdgcn.kill(i1 false)
   ret void
 }
 
 ; CHECK-LABEL: {{^}}test_kill_depth_var:
 ; CHECK-NEXT: ; %bb.0:
-; CHECK-NEXT: v_cmpx_le_f32_e32 vcc, 0, v0
+; CHECK-NEXT: v_cmpx_gt_f32_e32 vcc, 0, v0
 ; CHECK-NEXT: ; %bb.1:
 ; CHECK-NEXT: s_endpgm
 define amdgpu_ps void @test_kill_depth_var(float %x) #0 {
-  call void @llvm.AMDGPU.kill(float %x)
+  %cmp = fcmp olt float %x, 0.0
+  call void @llvm.amdgcn.kill(i1 %cmp)
   ret void
 }
 
 ; FIXME: Ideally only one would be emitted
 ; CHECK-LABEL: {{^}}test_kill_depth_var_x2_same:
 ; CHECK-NEXT: ; %bb.0:
-; CHECK-NEXT: v_cmpx_le_f32_e32 vcc, 0, v0
+; CHECK-NEXT: v_cmpx_gt_f32_e32 vcc, 0, v0
 ; CHECK-NEXT: ; %bb.1:
-; CHECK-NEXT: v_cmpx_le_f32_e32 vcc, 0, v0
+; CHECK-NEXT: v_cmpx_gt_f32_e32 vcc, 0, v0
 ; CHECK-NEXT: ; %bb.2:
 ; CHECK-NEXT: s_endpgm
 define amdgpu_ps void @test_kill_depth_var_x2_same(float %x) #0 {
-  call void @llvm.AMDGPU.kill(float %x)
-  call void @llvm.AMDGPU.kill(float %x)
+  %cmp = fcmp olt float %x, 0.0
+  call void @llvm.amdgcn.kill(i1 %cmp)
+  call void @llvm.amdgcn.kill(i1 %cmp)
   ret void
 }
 
 ; CHECK-LABEL: {{^}}test_kill_depth_var_x2:
 ; CHECK-NEXT: ; %bb.0:
-; CHECK-NEXT: v_cmpx_le_f32_e32 vcc, 0, v0
+; CHECK-NEXT: v_cmpx_gt_f32_e32 vcc, 0, v0
 ; CHECK-NEXT: ; %bb.1:
-; CHECK-NEXT: v_cmpx_le_f32_e32 vcc, 0, v1
+; CHECK-NEXT: v_cmpx_gt_f32_e32 vcc, 0, v1
 ; CHECK-NEXT: ; %bb.2:
 ; CHECK-NEXT: s_endpgm
 define amdgpu_ps void @test_kill_depth_var_x2(float %x, float %y) #0 {
-  call void @llvm.AMDGPU.kill(float %x)
-  call void @llvm.AMDGPU.kill(float %y)
+  %cmp.x = fcmp olt float %x, 0.0
+  call void @llvm.amdgcn.kill(i1 %cmp.x)
+  %cmp.y = fcmp olt float %y, 0.0
+  call void @llvm.amdgcn.kill(i1 %cmp.y)
   ret void
 }
 
 ; CHECK-LABEL: {{^}}test_kill_depth_var_x2_instructions:
 ; CHECK-NEXT: ; %bb.0:
-; CHECK-NEXT: v_cmpx_le_f32_e32 vcc, 0, v0
+; CHECK-NEXT: v_cmpx_gt_f32_e32 vcc, 0, v0
 ; CHECK-NEXT: s_cbranch_execnz BB6_2
 ; CHECK-NEXT: ; %bb.1:
 ; CHECK-NEXT: exp
 ; CHECK-NEXT: s_endpgm
 ; CHECK-NEXT: BB6_2:
 ; CHECK: v_mov_b32_e64 v7, -1
-; CHECK: v_cmpx_le_f32_e32 vcc, 0, v7
+; CHECK: v_cmpx_gt_f32_e32 vcc, 0, v7
 ; CHECK-NEXT: s_cbranch_execnz BB6_4
 ; CHECK-NEXT: ; %bb.3:
 ; CHECK-NEXT: exp
@@ -86,9 +91,11 @@ define amdgpu_ps void @test_kill_depth_var_x2(float %x, float %y) #0 {
 ; CHECK-NEXT: BB6_4:
 ; CHECK-NEXT: s_endpgm
 define amdgpu_ps void @test_kill_depth_var_x2_instructions(float %x) #0 {
-  call void @llvm.AMDGPU.kill(float %x)
+  %cmp.x = fcmp olt float %x, 0.0
+  call void @llvm.amdgcn.kill(i1 %cmp.x)
   %y = call float asm sideeffect "v_mov_b32_e64 v7, -1", "={v7}"()
-  call void @llvm.AMDGPU.kill(float %y)
+  %cmp.y = fcmp olt float %y, 0.0
+  call void @llvm.amdgcn.kill(i1 %cmp.y)
   ret void
 }
 
@@ -111,7 +118,7 @@ define amdgpu_ps void @test_kill_depth_var_x2_instructions(float %x) #0 {
 ; CHECK: v_nop_e64
 ; CHECK: v_nop_e64
 
-; CHECK: v_cmpx_le_f32_e32 vcc, 0, v7
+; CHECK: v_cmpx_gt_f32_e32 vcc, 0, v7
 ; CHECK-NEXT: s_cbranch_execnz [[SPLIT_BB:BB[0-9]+_[0-9]+]]
 ; CHECK-NEXT: ; %bb.2:
 ; CHECK-NEXT: exp null off, off, off, off done vm
@@ -137,7 +144,8 @@ bb:
     v_nop_e64
     v_nop_e64
     v_nop_e64", "={v7}"()
-  call void @llvm.AMDGPU.kill(float %var)
+  %cmp.var = fcmp olt float %var, 0.0
+  call void @llvm.amdgcn.kill(i1 %cmp.var)
   br label %exit
 
 exit:
@@ -162,7 +170,7 @@ exit:
 ; CHECK: ;;#ASMEND
 ; CHECK: v_mov_b32_e64 v8, -1
 ; CHECK: ;;#ASMEND
-; CHECK: v_cmpx_le_f32_e32 vcc, 0, v7
+; CHECK: v_cmpx_gt_f32_e32 vcc, 0, v7
 ; CHECK-NEXT: s_cbranch_execnz [[SPLIT_BB:BB[0-9]+_[0-9]+]]
 
 ; CHECK-NEXT: ; %bb.2:
@@ -196,7 +204,8 @@ bb:
     v_nop_e64
     v_nop_e64", "={v7}"()
   %live.across = call float asm sideeffect "v_mov_b32_e64 v8, -1", "={v8}"()
-  call void @llvm.AMDGPU.kill(float %var)
+  %cmp.var = fcmp olt float %var, 0.0
+  call void @llvm.amdgcn.kill(i1 %cmp.var)
   store volatile float %live.across, float addrspace(1)* undef
   %live.out = call float asm sideeffect "v_mov_b32_e64 v9, -2", "={v9}"()
   br label %exit
@@ -221,7 +230,7 @@ exit:
 
 ; CHECK: v_mov_b32_e64 v7, -1
 ; CHECK: v_nop_e64
-; CHECK: v_cmpx_le_f32_e32 vcc, 0, v7
+; CHECK: v_cmpx_gt_f32_e32 vcc, 0, v7
 
 ; CHECK-NEXT: ; %bb.3:
 ; CHECK: buffer_load_dword [[LOAD:v[0-9]+]]
@@ -251,7 +260,8 @@ bb:
     v_nop_e64
     v_nop_e64
     v_nop_e64", "={v7}"()
-  call void @llvm.AMDGPU.kill(float %var)
+  %cmp.var = fcmp olt float %var, 0.0
+  call void @llvm.amdgcn.kill(i1 %cmp.var)
   %vgpr = load volatile i32, i32 addrspace(1)* undef
   %loop.cond = icmp eq i32 %vgpr, 0
   br i1 %loop.cond, label %bb, label %exit
@@ -264,7 +274,7 @@ exit:
 ; bug 28550
 ; CHECK-LABEL: {{^}}phi_use_def_before_kill:
 ; CHECK: v_cndmask_b32_e64 [[PHIREG:v[0-9]+]], 0, -1.0,
-; CHECK: v_cmpx_le_f32_e32 vcc, 0,
+; CHECK: v_cmpx_lt_f32_e32 vcc, 0,
 ; CHECK-NEXT: s_cbranch_execnz [[BB4:BB[0-9]+_[0-9]+]]
 
 ; CHECK: exp
@@ -288,7 +298,8 @@ bb:
   %tmp = fadd float %x, 1.000000e+00
   %tmp1 = fcmp olt float 0.000000e+00, %tmp
   %tmp2 = select i1 %tmp1, float -1.000000e+00, float 0.000000e+00
-  call void @llvm.AMDGPU.kill(float %tmp2)
+  %cmp.tmp2 = fcmp olt float %tmp2, 0.0
+  call void @llvm.amdgcn.kill(i1 %cmp.tmp2)
   br i1 undef, label %phibb, label %bb8
 
 phibb:
@@ -335,7 +346,7 @@ bb5:                                              ; preds = %bb4, %bb3
   unreachable
 
 bb6:                                              ; preds = %bb
-  call void @llvm.AMDGPU.kill(float -1.000000e+00)
+  call void @llvm.amdgcn.kill(i1 false)
   unreachable
 
 bb7:                                              ; preds = %bb4
@@ -348,7 +359,7 @@ bb7:                                              ; preds = %bb4
 ; CHECK: s_xor_b64
 ; CHECK-NEXT: mask branch [[BB4:BB[0-9]+_[0-9]+]]
 
-; CHECK: v_cmpx_le_f32_e32 vcc, 0,
+; CHECK: v_cmpx_gt_f32_e32 vcc, 0,
 ; CHECK: [[BB4]]:
 ; CHECK: s_or_b64 exec, exec
 ; CHECK: image_sample_c
@@ -369,7 +380,8 @@ bb:
   br i1 %tmp, label %bb3, label %bb4
 
 bb3:                                              ; preds = %bb
-  call void @llvm.AMDGPU.kill(float %arg)
+  %cmp.arg = fcmp olt float %arg, 0.0
+  call void @llvm.amdgcn.kill(i1 %cmp.arg)
   br label %bb4
 
 bb4:                                              ; preds = %bb3, %bb
@@ -387,7 +399,7 @@ bb9:                                              ; preds = %bb4
 }
 
 declare <4 x float> @llvm.amdgcn.image.sample.c.1d.v4f32.f32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare void @llvm.AMDGPU.kill(float) #0
+declare void @llvm.amdgcn.kill(i1) #0
 
 attributes #0 = { nounwind }
 attributes #1 = { nounwind readonly }
diff --git a/llvm/test/CodeGen/AMDGPU/wqm.ll b/llvm/test/CodeGen/AMDGPU/wqm.ll
index 81232df43d1..3d181836848 100644
--- a/llvm/test/CodeGen/AMDGPU/wqm.ll
+++ b/llvm/test/CodeGen/AMDGPU/wqm.ll
@@ -586,7 +586,8 @@ main_body:
   %data.0 = extractelement <2 x float> %data, i32 0
   call void @llvm.amdgcn.buffer.store.f32(float %data.0, <4 x i32> undef, i32 %idx.0, i32 0, i1 0, i1 0)
 
-  call void @llvm.AMDGPU.kill(float %z)
+  %z.cmp = fcmp olt float %z, 0.0
+  call void @llvm.amdgcn.kill(i1 %z.cmp)
 
   %idx.1 = extractelement <2 x i32> %idx, i32 1
   %data.1 = extractelement <2 x float> %data, i32 1
@@ -619,7 +620,8 @@ main_body:
 
   call void @llvm.amdgcn.buffer.store.f32(float %data, <4 x i32> undef, i32 0, i32 0, i1 0, i1 0)
 
-  call void @llvm.AMDGPU.kill(float %z)
+  %z.cmp = fcmp olt float %z, 0.0
+  call void @llvm.amdgcn.kill(i1 %z.cmp)
 
   ret <4 x float> %dtex
 }
@@ -826,7 +828,7 @@ declare <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i32(i32, i32, <8 x i32>, i3
 declare float @llvm.amdgcn.buffer.load.f32(<4 x i32>, i32, i32, i1, i1) #3
 declare <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #3
 declare <4 x float> @llvm.amdgcn.image.sample.2d.v4f32.f32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #3
-declare void @llvm.AMDGPU.kill(float) #1
+declare void @llvm.amdgcn.kill(i1) #1
 declare float @llvm.amdgcn.wqm.f32(float) #3
 declare i32 @llvm.amdgcn.wqm.i32(i32) #3
 declare float @llvm.amdgcn.wwm.f32(float) #3