Skip to content

[AMDGPU][GFX11] Add more test coverage for FMA instructions. #65935

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Sep 12, 2023
Merged

[AMDGPU][GFX11] Add more test coverage for FMA instructions. #65935

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
343 changes: 235 additions & 108 deletions llvm/test/CodeGen/AMDGPU/dagcombine-fma-fmad.ll
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,71 +1,140 @@
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -march=amdgcn -mcpu=gfx1010 -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefix=GCN %s
; RUN: llc -march=amdgcn -mcpu=gfx1010 -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,GFX10 %s
; RUN: llc -march=amdgcn -mcpu=gfx1100 -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,GFX11 %s

define amdgpu_ps float @_amdgpu_ps_main() #0 {
; GCN-LABEL: _amdgpu_ps_main:
; GCN: ; %bb.0: ; %.entry
; GCN-NEXT: image_sample v[0:1], v[0:1], s[0:7], s[0:3] dmask:0x3 dim:SQ_RSRC_IMG_2D
; GCN-NEXT: v_mov_b32_e32 v4, 0
; GCN-NEXT: s_waitcnt vmcnt(0)
; GCN-NEXT: s_clause 0x1
; GCN-NEXT: image_sample v2, v[0:1], s[0:7], s[0:3] dmask:0x4 dim:SQ_RSRC_IMG_2D
; GCN-NEXT: image_sample v3, v[0:1], s[0:7], s[0:3] dmask:0x1 dim:SQ_RSRC_IMG_2D
; GCN-NEXT: s_waitcnt vmcnt(0)
; GCN-NEXT: image_load_mip v4, v[2:4], s[0:7] dmask:0x4 dim:SQ_RSRC_IMG_2D unorm
; GCN-NEXT: s_clause 0x3
; GCN-NEXT: s_buffer_load_dword s24, s[0:3], 0x5c
; GCN-NEXT: s_buffer_load_dword s28, s[0:3], 0x7c
; GCN-NEXT: s_buffer_load_dword s29, s[0:3], 0xc0
; GCN-NEXT: s_waitcnt_depctr 0xffe3
; GCN-NEXT: s_nop 0
; GCN-NEXT: s_buffer_load_dwordx4 s[0:3], s[0:3], 0x40
; GCN-NEXT: s_waitcnt lgkmcnt(0)
; GCN-NEXT: s_clause 0x1
; GCN-NEXT: s_buffer_load_dwordx4 s[4:7], s[0:3], 0x50
; GCN-NEXT: s_nop 0
; GCN-NEXT: s_buffer_load_dword s0, s[0:3], 0x2c
; GCN-NEXT: v_sub_f32_e64 v5, s24, s28
; GCN-NEXT: s_waitcnt lgkmcnt(0)
; GCN-NEXT: s_clause 0x4
; GCN-NEXT: s_buffer_load_dwordx4 s[8:11], s[0:3], 0x60
; GCN-NEXT: s_buffer_load_dwordx4 s[12:15], s[0:3], 0x20
; GCN-NEXT: s_buffer_load_dwordx4 s[16:19], s[0:3], 0x0
; GCN-NEXT: s_buffer_load_dwordx4 s[20:23], s[0:3], 0x70
; GCN-NEXT: s_buffer_load_dwordx4 s[24:27], s[0:3], 0x10
; GCN-NEXT: v_fma_f32 v1, v1, v5, s28
; GCN-NEXT: v_max_f32_e64 v6, s0, s0 clamp
; GCN-NEXT: v_add_f32_e64 v5, s29, -1.0
; GCN-NEXT: v_sub_f32_e32 v8, s0, v1
; GCN-NEXT: v_fma_f32 v7, -s2, v6, s6
; GCN-NEXT: v_fma_f32 v5, v6, v5, 1.0
; GCN-NEXT: v_mad_f32 v10, s2, v6, v2
; GCN-NEXT: s_mov_b32 s0, 0x3c23d70a
; GCN-NEXT: v_fmac_f32_e32 v1, v6, v8
; GCN-NEXT: v_fmac_f32_e32 v10, v7, v6
; GCN-NEXT: s_waitcnt lgkmcnt(0)
; GCN-NEXT: v_mul_f32_e32 v9, s10, v0
; GCN-NEXT: v_fma_f32 v0, -v0, s10, s14
; GCN-NEXT: v_mul_f32_e32 v8, s18, v2
; GCN-NEXT: v_mul_f32_e32 v3, s22, v3
; GCN-NEXT: v_fmac_f32_e32 v9, v0, v6
; GCN-NEXT: v_sub_f32_e32 v0, v1, v5
; GCN-NEXT: v_mul_f32_e32 v1, v8, v6
; GCN-NEXT: v_mul_f32_e32 v7, v6, v3
; GCN-NEXT: v_fma_f32 v3, -v6, v3, v9
; GCN-NEXT: v_fmac_f32_e32 v5, v0, v6
; GCN-NEXT: v_fma_f32 v0, v2, s26, -v1
; GCN-NEXT: v_fmac_f32_e32 v7, v3, v6
; GCN-NEXT: v_fmac_f32_e32 v1, v0, v6
; GCN-NEXT: v_mul_f32_e32 v0, v2, v6
; GCN-NEXT: s_waitcnt vmcnt(0)
; GCN-NEXT: v_add_f32_e32 v4, v4, v10
; GCN-NEXT: v_mul_f32_e32 v3, v4, v6
; GCN-NEXT: v_fmaak_f32 v4, s0, v5, 0x3ca3d70a
; GCN-NEXT: v_mul_f32_e32 v1, v3, v1
; GCN-NEXT: v_mul_f32_e32 v2, v7, v4
; GCN-NEXT: v_fmac_f32_e32 v1, v2, v0
; GCN-NEXT: v_max_f32_e32 v0, 0, v1
; GCN-NEXT: ; return to shader part epilog
; GFX10-LABEL: _amdgpu_ps_main:
; GFX10: ; %bb.0: ; %.entry
; GFX10-NEXT: image_sample v[0:1], v[0:1], s[0:7], s[0:3] dmask:0x3 dim:SQ_RSRC_IMG_2D
; GFX10-NEXT: v_mov_b32_e32 v4, 0
; GFX10-NEXT: s_waitcnt vmcnt(0)
; GFX10-NEXT: s_clause 0x1
; GFX10-NEXT: image_sample v2, v[0:1], s[0:7], s[0:3] dmask:0x4 dim:SQ_RSRC_IMG_2D
; GFX10-NEXT: image_sample v3, v[0:1], s[0:7], s[0:3] dmask:0x1 dim:SQ_RSRC_IMG_2D
; GFX10-NEXT: s_waitcnt vmcnt(0)
; GFX10-NEXT: image_load_mip v4, v[2:4], s[0:7] dmask:0x4 dim:SQ_RSRC_IMG_2D unorm
; GFX10-NEXT: s_clause 0x3
; GFX10-NEXT: s_buffer_load_dword s24, s[0:3], 0x5c
; GFX10-NEXT: s_buffer_load_dword s28, s[0:3], 0x7c
; GFX10-NEXT: s_buffer_load_dword s29, s[0:3], 0xc0
; GFX10-NEXT: s_waitcnt_depctr 0xffe3
; GFX10-NEXT: s_nop 0
; GFX10-NEXT: s_buffer_load_dwordx4 s[0:3], s[0:3], 0x40
; GFX10-NEXT: s_waitcnt lgkmcnt(0)
; GFX10-NEXT: s_clause 0x1
; GFX10-NEXT: s_buffer_load_dwordx4 s[4:7], s[0:3], 0x50
; GFX10-NEXT: s_nop 0
; GFX10-NEXT: s_buffer_load_dword s0, s[0:3], 0x2c
; GFX10-NEXT: v_sub_f32_e64 v5, s24, s28
; GFX10-NEXT: s_waitcnt lgkmcnt(0)
; GFX10-NEXT: s_clause 0x4
; GFX10-NEXT: s_buffer_load_dwordx4 s[8:11], s[0:3], 0x60
; GFX10-NEXT: s_buffer_load_dwordx4 s[12:15], s[0:3], 0x20
; GFX10-NEXT: s_buffer_load_dwordx4 s[16:19], s[0:3], 0x0
; GFX10-NEXT: s_buffer_load_dwordx4 s[20:23], s[0:3], 0x70
; GFX10-NEXT: s_buffer_load_dwordx4 s[24:27], s[0:3], 0x10
; GFX10-NEXT: v_fma_f32 v1, v1, v5, s28
; GFX10-NEXT: v_max_f32_e64 v6, s0, s0 clamp
; GFX10-NEXT: v_add_f32_e64 v5, s29, -1.0
; GFX10-NEXT: v_sub_f32_e32 v8, s0, v1
; GFX10-NEXT: v_fma_f32 v7, -s2, v6, s6
; GFX10-NEXT: v_fma_f32 v5, v6, v5, 1.0
; GFX10-NEXT: v_mad_f32 v10, s2, v6, v2
; GFX10-NEXT: s_mov_b32 s0, 0x3c23d70a
; GFX10-NEXT: v_fmac_f32_e32 v1, v6, v8
; GFX10-NEXT: v_fmac_f32_e32 v10, v7, v6
; GFX10-NEXT: s_waitcnt lgkmcnt(0)
; GFX10-NEXT: v_mul_f32_e32 v9, s10, v0
; GFX10-NEXT: v_fma_f32 v0, -v0, s10, s14
; GFX10-NEXT: v_mul_f32_e32 v8, s18, v2
; GFX10-NEXT: v_mul_f32_e32 v3, s22, v3
; GFX10-NEXT: v_fmac_f32_e32 v9, v0, v6
; GFX10-NEXT: v_sub_f32_e32 v0, v1, v5
; GFX10-NEXT: v_mul_f32_e32 v1, v8, v6
; GFX10-NEXT: v_mul_f32_e32 v7, v6, v3
; GFX10-NEXT: v_fma_f32 v3, -v6, v3, v9
; GFX10-NEXT: v_fmac_f32_e32 v5, v0, v6
; GFX10-NEXT: v_fma_f32 v0, v2, s26, -v1
; GFX10-NEXT: v_fmac_f32_e32 v7, v3, v6
; GFX10-NEXT: v_fmac_f32_e32 v1, v0, v6
; GFX10-NEXT: v_mul_f32_e32 v0, v2, v6
; GFX10-NEXT: s_waitcnt vmcnt(0)
; GFX10-NEXT: v_add_f32_e32 v4, v4, v10
; GFX10-NEXT: v_mul_f32_e32 v3, v4, v6
; GFX10-NEXT: v_fmaak_f32 v4, s0, v5, 0x3ca3d70a
; GFX10-NEXT: v_mul_f32_e32 v1, v3, v1
; GFX10-NEXT: v_mul_f32_e32 v2, v7, v4
; GFX10-NEXT: v_fmac_f32_e32 v1, v2, v0
; GFX10-NEXT: v_max_f32_e32 v0, 0, v1
; GFX10-NEXT: ; return to shader part epilog
;
; GFX11-LABEL: _amdgpu_ps_main:
; GFX11: ; %bb.0: ; %.entry
; GFX11-NEXT: image_sample v[0:1], v[0:1], s[0:7], s[0:3] dmask:0x3 dim:SQ_RSRC_IMG_2D
; GFX11-NEXT: v_mov_b32_e32 v4, 0
; GFX11-NEXT: s_waitcnt vmcnt(0)
; GFX11-NEXT: s_clause 0x1
; GFX11-NEXT: image_sample v2, v[0:1], s[0:7], s[0:3] dmask:0x4 dim:SQ_RSRC_IMG_2D
; GFX11-NEXT: image_sample v3, v[0:1], s[0:7], s[0:3] dmask:0x1 dim:SQ_RSRC_IMG_2D
; GFX11-NEXT: s_waitcnt vmcnt(0)
; GFX11-NEXT: image_load_mip v4, v[2:4], s[0:7] dmask:0x4 dim:SQ_RSRC_IMG_2D unorm
; GFX11-NEXT: s_clause 0x3
; GFX11-NEXT: s_buffer_load_b32 s24, s[0:3], 0x5c
; GFX11-NEXT: s_buffer_load_b32 s28, s[0:3], 0x7c
; GFX11-NEXT: s_buffer_load_b32 s29, s[0:3], 0xc0
; GFX11-NEXT: s_buffer_load_b128 s[0:3], s[0:3], 0x40
; GFX11-NEXT: s_waitcnt lgkmcnt(0)
; GFX11-NEXT: s_clause 0x1
; GFX11-NEXT: s_buffer_load_b128 s[4:7], s[0:3], 0x50
; GFX11-NEXT: s_buffer_load_b32 s0, s[0:3], 0x2c
; GFX11-NEXT: v_sub_f32_e64 v5, s24, s28
; GFX11-NEXT: s_waitcnt lgkmcnt(0)
; GFX11-NEXT: s_clause 0x3
; GFX11-NEXT: s_buffer_load_b128 s[8:11], s[0:3], 0x60
; GFX11-NEXT: s_buffer_load_b128 s[12:15], s[0:3], 0x20
; GFX11-NEXT: s_buffer_load_b128 s[16:19], s[0:3], 0x0
; GFX11-NEXT: s_buffer_load_b128 s[20:23], s[0:3], 0x70
; GFX11-NEXT: v_fma_f32 v1, v1, v5, s28
; GFX11-NEXT: v_max_f32_e64 v6, s0, s0 clamp
; GFX11-NEXT: s_buffer_load_b128 s[24:27], s[0:3], 0x10
; GFX11-NEXT: v_add_f32_e64 v5, s29, -1.0
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_3)
; GFX11-NEXT: v_sub_f32_e32 v8, s0, v1
; GFX11-NEXT: v_fma_f32 v7, -s2, v6, s6
; GFX11-NEXT: v_fma_f32 v10, s2, v6, v2
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_4)
; GFX11-NEXT: v_fma_f32 v5, v6, v5, 1.0
; GFX11-NEXT: s_mov_b32 s0, 0x3c23d70a
; GFX11-NEXT: s_waitcnt lgkmcnt(0)
; GFX11-NEXT: v_mul_f32_e32 v9, s10, v0
; GFX11-NEXT: v_fma_f32 v0, -v0, s10, s14
; GFX11-NEXT: v_mul_f32_e32 v3, s22, v3
; GFX11-NEXT: v_dual_fmac_f32 v1, v6, v8 :: v_dual_mul_f32 v8, s18, v2
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_3)
; GFX11-NEXT: v_fmac_f32_e32 v9, v0, v6
; GFX11-NEXT: v_dual_fmac_f32 v10, v7, v6 :: v_dual_mul_f32 v7, v6, v3
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_3)
; GFX11-NEXT: v_sub_f32_e32 v0, v1, v5
; GFX11-NEXT: v_fma_f32 v3, -v6, v3, v9
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_3)
; GFX11-NEXT: v_fmac_f32_e32 v7, v3, v6
; GFX11-NEXT: v_fmac_f32_e32 v5, v0, v6
; GFX11-NEXT: v_mul_f32_e32 v1, v8, v6
; GFX11-NEXT: s_waitcnt vmcnt(0)
; GFX11-NEXT: v_add_f32_e32 v4, v4, v10
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_3)
; GFX11-NEXT: v_dual_mul_f32 v3, v4, v6 :: v_dual_fmaak_f32 v4, s0, v5, 0x3ca3d70a
; GFX11-NEXT: v_fma_f32 v0, v2, s26, -v1
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_4)
; GFX11-NEXT: v_fmac_f32_e32 v1, v0, v6
; GFX11-NEXT: v_mul_f32_e32 v0, v2, v6
; GFX11-NEXT: v_mul_f32_e32 v2, v7, v4
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_1)
; GFX11-NEXT: v_mul_f32_e32 v1, v3, v1
; GFX11-NEXT: v_fmac_f32_e32 v1, v2, v0
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1)
; GFX11-NEXT: v_max_f32_e32 v0, 0, v1
; GFX11-NEXT: ; return to shader part epilog
.entry:
%0 = call <3 x float> @llvm.amdgcn.image.sample.2d.v3f32.f32(i32 7, float undef, float undef, <8 x i32> undef, <4 x i32> undef, i1 false, i32 0, i32 0)
%.i2243 = extractelement <3 x float> %0, i32 2
Expand Down Expand Up @@ -168,13 +237,22 @@ define amdgpu_ps float @_amdgpu_ps_main() #0 {
}

define float @fmac_sequence_simple(float %a, float %b, float %c, float %d, float %e) #0 {
; GCN-LABEL: fmac_sequence_simple:
; GCN: ; %bb.0:
; GCN-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; GCN-NEXT: v_fma_f32 v2, v2, v3, v4
; GCN-NEXT: v_fmac_f32_e32 v2, v0, v1
; GCN-NEXT: v_mov_b32_e32 v0, v2
; GCN-NEXT: s_setpc_b64 s[30:31]
; GFX10-LABEL: fmac_sequence_simple:
; GFX10: ; %bb.0:
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; GFX10-NEXT: v_fma_f32 v2, v2, v3, v4
; GFX10-NEXT: v_fmac_f32_e32 v2, v0, v1
; GFX10-NEXT: v_mov_b32_e32 v0, v2
; GFX10-NEXT: s_setpc_b64 s[30:31]
;
; GFX11-LABEL: fmac_sequence_simple:
; GFX11: ; %bb.0:
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; GFX11-NEXT: v_fma_f32 v2, v2, v3, v4
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
; GFX11-NEXT: v_fmac_f32_e32 v2, v0, v1
; GFX11-NEXT: v_mov_b32_e32 v0, v2
; GFX11-NEXT: s_setpc_b64 s[30:31]
%t0 = fmul fast float %a, %b
%t1 = fmul fast float %c, %d
%t2 = fadd fast float %t0, %t1
Expand All @@ -183,14 +261,25 @@ define float @fmac_sequence_simple(float %a, float %b, float %c, float %d, float
}

define float @fmac_sequence_innermost_fmul(float %a, float %b, float %c, float %d, float %e, float %f, float %g) #0 {
; GCN-LABEL: fmac_sequence_innermost_fmul:
; GCN: ; %bb.0:
; GCN-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; GCN-NEXT: v_mad_f32 v2, v2, v3, v6
; GCN-NEXT: v_fmac_f32_e32 v2, v0, v1
; GCN-NEXT: v_fmac_f32_e32 v2, v4, v5
; GCN-NEXT: v_mov_b32_e32 v0, v2
; GCN-NEXT: s_setpc_b64 s[30:31]
; GFX10-LABEL: fmac_sequence_innermost_fmul:
; GFX10: ; %bb.0:
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; GFX10-NEXT: v_mad_f32 v2, v2, v3, v6
; GFX10-NEXT: v_fmac_f32_e32 v2, v0, v1
; GFX10-NEXT: v_fmac_f32_e32 v2, v4, v5
; GFX10-NEXT: v_mov_b32_e32 v0, v2
; GFX10-NEXT: s_setpc_b64 s[30:31]
;
; GFX11-LABEL: fmac_sequence_innermost_fmul:
; GFX11: ; %bb.0:
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; GFX11-NEXT: v_fma_f32 v2, v2, v3, v6
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
; GFX11-NEXT: v_fmac_f32_e32 v2, v0, v1
; GFX11-NEXT: v_fmac_f32_e32 v2, v4, v5
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1)
; GFX11-NEXT: v_mov_b32_e32 v0, v2
; GFX11-NEXT: s_setpc_b64 s[30:31]
%t0 = fmul fast float %a, %b
%t1 = fmul fast float %c, %d
%t2 = fadd fast float %t0, %t1
Expand All @@ -201,14 +290,25 @@ define float @fmac_sequence_innermost_fmul(float %a, float %b, float %c, float %
}

define float @fmac_sequence_innermost_fmul_swapped_operands(float %a, float %b, float %c, float %d, float %e, float %f, float %g) #0 {
; GCN-LABEL: fmac_sequence_innermost_fmul_swapped_operands:
; GCN: ; %bb.0:
; GCN-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; GCN-NEXT: v_mad_f32 v2, v2, v3, v6
; GCN-NEXT: v_fmac_f32_e32 v2, v0, v1
; GCN-NEXT: v_fmac_f32_e32 v2, v4, v5
; GCN-NEXT: v_mov_b32_e32 v0, v2
; GCN-NEXT: s_setpc_b64 s[30:31]
; GFX10-LABEL: fmac_sequence_innermost_fmul_swapped_operands:
; GFX10: ; %bb.0:
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; GFX10-NEXT: v_mad_f32 v2, v2, v3, v6
; GFX10-NEXT: v_fmac_f32_e32 v2, v0, v1
; GFX10-NEXT: v_fmac_f32_e32 v2, v4, v5
; GFX10-NEXT: v_mov_b32_e32 v0, v2
; GFX10-NEXT: s_setpc_b64 s[30:31]
;
; GFX11-LABEL: fmac_sequence_innermost_fmul_swapped_operands:
; GFX11: ; %bb.0:
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; GFX11-NEXT: v_fma_f32 v2, v2, v3, v6
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
; GFX11-NEXT: v_fmac_f32_e32 v2, v0, v1
; GFX11-NEXT: v_fmac_f32_e32 v2, v4, v5
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1)
; GFX11-NEXT: v_mov_b32_e32 v0, v2
; GFX11-NEXT: s_setpc_b64 s[30:31]
%t0 = fmul fast float %a, %b
%t1 = fmul fast float %c, %d
%t2 = fadd fast float %t0, %t1
Expand All @@ -219,12 +319,20 @@ define float @fmac_sequence_innermost_fmul_swapped_operands(float %a, float %b,
}

define amdgpu_ps float @fmac_sequence_innermost_fmul_sgpr(float inreg %a, float inreg %b, float inreg %c, float inreg %d, float inreg %e, float inreg %f, float %g) #0 {
; GCN-LABEL: fmac_sequence_innermost_fmul_sgpr:
; GCN: ; %bb.0:
; GCN-NEXT: v_mac_f32_e64 v0, s2, s3
; GCN-NEXT: v_fmac_f32_e64 v0, s0, s1
; GCN-NEXT: v_fmac_f32_e64 v0, s4, s5
; GCN-NEXT: ; return to shader part epilog
; GFX10-LABEL: fmac_sequence_innermost_fmul_sgpr:
; GFX10: ; %bb.0:
; GFX10-NEXT: v_mac_f32_e64 v0, s2, s3
; GFX10-NEXT: v_fmac_f32_e64 v0, s0, s1
; GFX10-NEXT: v_fmac_f32_e64 v0, s4, s5
; GFX10-NEXT: ; return to shader part epilog
;
; GFX11-LABEL: fmac_sequence_innermost_fmul_sgpr:
; GFX11: ; %bb.0:
; GFX11-NEXT: v_fmac_f32_e64 v0, s2, s3
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
; GFX11-NEXT: v_fmac_f32_e64 v0, s0, s1
; GFX11-NEXT: v_fmac_f32_e64 v0, s4, s5
; GFX11-NEXT: ; return to shader part epilog
%t0 = fmul fast float %a, %b
%t1 = fmul fast float %c, %d
%t2 = fadd fast float %t0, %t1
Expand All @@ -235,14 +343,25 @@ define amdgpu_ps float @fmac_sequence_innermost_fmul_sgpr(float inreg %a, float
}

define amdgpu_ps float @fmac_sequence_innermost_fmul_multiple_use(float inreg %a, float inreg %b, float inreg %c, float inreg %d, float inreg %e, float inreg %f, float %g) #0 {
; GCN-LABEL: fmac_sequence_innermost_fmul_multiple_use:
; GCN: ; %bb.0:
; GCN-NEXT: v_mul_f32_e64 v1, s2, s3
; GCN-NEXT: v_fmac_f32_e64 v1, s0, s1
; GCN-NEXT: v_fma_f32 v2, s5, s4, v1
; GCN-NEXT: v_fmac_f32_e32 v1, s5, v2
; GCN-NEXT: v_add_f32_e32 v0, v1, v0
; GCN-NEXT: ; return to shader part epilog
; GFX10-LABEL: fmac_sequence_innermost_fmul_multiple_use:
; GFX10: ; %bb.0:
; GFX10-NEXT: v_mul_f32_e64 v1, s2, s3
; GFX10-NEXT: v_fmac_f32_e64 v1, s0, s1
; GFX10-NEXT: v_fma_f32 v2, s5, s4, v1
; GFX10-NEXT: v_fmac_f32_e32 v1, s5, v2
; GFX10-NEXT: v_add_f32_e32 v0, v1, v0
; GFX10-NEXT: ; return to shader part epilog
;
; GFX11-LABEL: fmac_sequence_innermost_fmul_multiple_use:
; GFX11: ; %bb.0:
; GFX11-NEXT: v_mul_f32_e64 v1, s2, s3
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
; GFX11-NEXT: v_fmac_f32_e64 v1, s0, s1
; GFX11-NEXT: v_fma_f32 v2, s5, s4, v1
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
; GFX11-NEXT: v_fmac_f32_e32 v1, s5, v2
; GFX11-NEXT: v_add_f32_e32 v0, v1, v0
; GFX11-NEXT: ; return to shader part epilog
%t0 = fmul fast float %a, %b
%t1 = fmul fast float %c, %d
%t2 = fadd fast float %t0, %t1
Expand All @@ -258,12 +377,20 @@ define amdgpu_ps float @fmac_sequence_innermost_fmul_multiple_use(float inreg %a
; selecting it as a multiply. In some cases the multiply is better because
; SIFoldOperands can fold it into a previous instruction as an output modifier.
define amdgpu_ps float @fma_vs_output_modifier(float %x, i32 %n) #0 {
; GCN-LABEL: fma_vs_output_modifier:
; GCN: ; %bb.0:
; GCN-NEXT: v_cvt_f32_i32_e64 v1, v1 mul:2
; GCN-NEXT: v_mul_f32_e32 v0, v0, v0
; GCN-NEXT: v_mul_f32_e32 v0, v0, v1
; GCN-NEXT: ; return to shader part epilog
; GFX10-LABEL: fma_vs_output_modifier:
; GFX10: ; %bb.0:
; GFX10-NEXT: v_cvt_f32_i32_e64 v1, v1 mul:2
; GFX10-NEXT: v_mul_f32_e32 v0, v0, v0
; GFX10-NEXT: v_mul_f32_e32 v0, v0, v1
; GFX10-NEXT: ; return to shader part epilog
;
; GFX11-LABEL: fma_vs_output_modifier:
; GFX11: ; %bb.0:
; GFX11-NEXT: v_cvt_f32_i32_e64 v1, v1 mul:2
; GFX11-NEXT: v_mul_f32_e32 v0, v0, v0
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1)
; GFX11-NEXT: v_mul_f32_e32 v0, v0, v1
; GFX11-NEXT: ; return to shader part epilog
%s = sitofp i32 %n to float
%m = fmul contract float %x, %x
%a = fmul contract float %m, 2.0
Expand Down
Loading