Skip to content

[AMDGPU] Patterns for <2 x bfloat> fneg (fabs) #142911

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
Jun 5, 2025
Merged

[AMDGPU] Patterns for <2 x bfloat> fneg (fabs) #142911

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
11 changes: 5 additions & 6 deletions llvm/lib/Target/AMDGPU/SIInstructions.td
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1840,22 +1840,21 @@ def : GCNPat <
(UniformUnaryFrag<fabs> (v2fp16vt SReg_32:$src)),
(S_AND_B32 SReg_32:$src, (S_MOV_B32 (i32 0x7fff7fff)))
>;
}

// This is really (fneg (fabs v2f16:$src))
//
// fabs is not reported as free because there is modifier for it in
// VOP3P instructions, so it is turned into the bit op.
def : GCNPat <
(UniformUnaryFrag<fneg> (v2f16 (bitconvert (and_oneuse (i32 SReg_32:$src), 0x7fff7fff)))),
(UniformUnaryFrag<fneg> (v2fp16vt (bitconvert (and_oneuse (i32 SReg_32:$src), 0x7fff7fff)))),
(S_OR_B32 SReg_32:$src, (S_MOV_B32 (i32 0x80008000))) // Set sign bit
>;

def : GCNPat <
(UniformUnaryFrag<fneg> (v2f16 (fabs SReg_32:$src))),
(UniformUnaryFrag<fneg> (v2fp16vt (fabs SReg_32:$src))),
(S_OR_B32 SReg_32:$src, (S_MOV_B32 (i32 0x80008000))) // Set sign bit
>;

}

// COPY_TO_REGCLASS is needed to avoid using SCC from S_XOR_B32 instead
// of the real value.
Expand Down Expand Up @@ -1986,12 +1985,12 @@ def : GCNPat <
(fabs (v2fp16vt VGPR_32:$src)),
(V_AND_B32_e64 (S_MOV_B32 (i32 0x7fff7fff)), VGPR_32:$src)
>;
}

def : GCNPat <
(fneg (v2f16 (fabs VGPR_32:$src))),
(fneg (v2fp16vt (fabs VGPR_32:$src))),
(V_OR_B32_e64 (S_MOV_B32 (i32 0x80008000)), VGPR_32:$src)
>;
}

def : GCNPat <
(fabs (f64 VReg_64:$src)),
Expand Down
38 changes: 16 additions & 22 deletions llvm/test/CodeGen/AMDGPU/fneg-fabs.bf16.ll
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -523,8 +523,7 @@ define amdgpu_kernel void @s_fneg_fabs_v2bf16_non_bc_src(ptr addrspace(1) %out,
; VI-NEXT: v_cndmask_b32_e32 v1, v2, v3, vcc
; VI-NEXT: v_lshrrev_b32_e32 v1, 16, v1
; VI-NEXT: v_alignbit_b32 v0, v1, v0, 16
; VI-NEXT: v_and_b32_e32 v0, 0x7fff7fff, v0
; VI-NEXT: v_xor_b32_e32 v2, 0x80008000, v0
; VI-NEXT: v_or_b32_e32 v2, 0x80008000, v0
; VI-NEXT: v_mov_b32_e32 v0, s0
; VI-NEXT: v_mov_b32_e32 v1, s1
; VI-NEXT: flat_store_dword v[0:1], v2
Expand Down Expand Up @@ -556,8 +555,7 @@ define amdgpu_kernel void @s_fneg_fabs_v2bf16_non_bc_src(ptr addrspace(1) %out,
; GFX9-NEXT: v_lshrrev_b32_e32 v1, 16, v1
; GFX9-NEXT: v_and_b32_sdwa v2, v3, v2 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:WORD_1
; GFX9-NEXT: v_lshl_or_b32 v1, v1, 16, v2
; GFX9-NEXT: v_and_b32_e32 v1, 0x7fff7fff, v1
; GFX9-NEXT: v_xor_b32_e32 v1, 0x80008000, v1
; GFX9-NEXT: v_or_b32_e32 v1, 0x80008000, v1
; GFX9-NEXT: global_store_dword v0, v1, s[0:1]
; GFX9-NEXT: s_endpgm
;
Expand Down Expand Up @@ -590,9 +588,9 @@ define amdgpu_kernel void @s_fneg_fabs_v2bf16_non_bc_src(ptr addrspace(1) %out,
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
; GFX11-NEXT: v_lshrrev_b32_e32 v1, 16, v1
; GFX11-NEXT: v_lshl_or_b32 v0, v1, 16, v0
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
; GFX11-NEXT: v_dual_mov_b32 v1, 0 :: v_dual_and_b32 v0, 0x7fff7fff, v0
; GFX11-NEXT: v_xor_b32_e32 v0, 0x80008000, v0
; GFX11-NEXT: v_mov_b32_e32 v1, 0
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_2)
; GFX11-NEXT: v_or_b32_e32 v0, 0x80008000, v0
; GFX11-NEXT: s_waitcnt lgkmcnt(0)
; GFX11-NEXT: global_store_b32 v1, v0, s[0:1]
; GFX11-NEXT: s_endpgm
Expand Down Expand Up @@ -634,8 +632,7 @@ define amdgpu_kernel void @s_fneg_fabs_v2bf16_bc_src(ptr addrspace(1) %out, <2 x
; VI-NEXT: s_mov_b32 flat_scratch_lo, s13
; VI-NEXT: s_lshr_b32 flat_scratch_hi, s12, 8
; VI-NEXT: s_waitcnt lgkmcnt(0)
; VI-NEXT: s_and_b32 s2, s2, 0x7fff7fff
; VI-NEXT: s_xor_b32 s2, s2, 0x80008000
; VI-NEXT: s_or_b32 s2, s2, 0x80008000
; VI-NEXT: v_mov_b32_e32 v0, s0
; VI-NEXT: v_mov_b32_e32 v1, s1
; VI-NEXT: v_mov_b32_e32 v2, s2
Expand All @@ -648,8 +645,7 @@ define amdgpu_kernel void @s_fneg_fabs_v2bf16_bc_src(ptr addrspace(1) %out, <2 x
; GFX9-NEXT: s_load_dwordx2 s[0:1], s[8:9], 0x0
; GFX9-NEXT: v_mov_b32_e32 v0, 0
; GFX9-NEXT: s_waitcnt lgkmcnt(0)
; GFX9-NEXT: s_and_b32 s2, s2, 0x7fff7fff
; GFX9-NEXT: s_xor_b32 s2, s2, 0x80008000
; GFX9-NEXT: s_or_b32 s2, s2, 0x80008000
; GFX9-NEXT: v_mov_b32_e32 v1, s2
; GFX9-NEXT: global_store_dword v0, v1, s[0:1]
; GFX9-NEXT: s_endpgm
Expand All @@ -660,9 +656,8 @@ define amdgpu_kernel void @s_fneg_fabs_v2bf16_bc_src(ptr addrspace(1) %out, <2 x
; GFX11-NEXT: s_load_b32 s2, s[4:5], 0x8
; GFX11-NEXT: s_load_b64 s[0:1], s[4:5], 0x0
; GFX11-NEXT: s_waitcnt lgkmcnt(0)
; GFX11-NEXT: s_and_b32 s2, s2, 0x7fff7fff
; GFX11-NEXT: s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(SALU_CYCLE_1)
; GFX11-NEXT: s_xor_b32 s2, s2, 0x80008000
; GFX11-NEXT: s_or_b32 s2, s2, 0x80008000
; GFX11-NEXT: s_delay_alu instid0(SALU_CYCLE_1)
; GFX11-NEXT: v_dual_mov_b32 v0, 0 :: v_dual_mov_b32 v1, s2
; GFX11-NEXT: global_store_b32 v0, v1, s[0:1]
; GFX11-NEXT: s_endpgm
Expand Down Expand Up @@ -977,7 +972,7 @@ define amdgpu_kernel void @s_fneg_multi_use_fabs_v2bf16(ptr addrspace(1) %out0,
; VI-NEXT: v_mov_b32_e32 v0, s0
; VI-NEXT: s_and_b32 s0, s4, 0x7fff7fff
; VI-NEXT: v_mov_b32_e32 v1, s1
; VI-NEXT: s_xor_b32 s1, s0, 0x80008000
; VI-NEXT: s_or_b32 s1, s4, 0x80008000
; VI-NEXT: v_mov_b32_e32 v4, s0
; VI-NEXT: v_mov_b32_e32 v2, s2
; VI-NEXT: v_mov_b32_e32 v3, s3
Expand All @@ -992,12 +987,12 @@ define amdgpu_kernel void @s_fneg_multi_use_fabs_v2bf16(ptr addrspace(1) %out0,
; GFX9-NEXT: s_load_dwordx4 s[0:3], s[8:9], 0x0
; GFX9-NEXT: v_mov_b32_e32 v0, 0
; GFX9-NEXT: s_waitcnt lgkmcnt(0)
; GFX9-NEXT: s_and_b32 s4, s4, 0x7fff7fff
; GFX9-NEXT: s_xor_b32 s5, s4, 0x80008000
; GFX9-NEXT: v_mov_b32_e32 v1, s4
; GFX9-NEXT: global_store_dword v0, v1, s[0:1]
; GFX9-NEXT: s_and_b32 s5, s4, 0x7fff7fff
; GFX9-NEXT: s_or_b32 s4, s4, 0x80008000
; GFX9-NEXT: v_mov_b32_e32 v1, s5
; GFX9-NEXT: global_store_dword v0, v1, s[2:3]
; GFX9-NEXT: v_mov_b32_e32 v2, s4
; GFX9-NEXT: global_store_dword v0, v1, s[0:1]
; GFX9-NEXT: global_store_dword v0, v2, s[2:3]
; GFX9-NEXT: s_endpgm
;
; GFX11-LABEL: s_fneg_multi_use_fabs_v2bf16:
Expand All @@ -1007,9 +1002,8 @@ define amdgpu_kernel void @s_fneg_multi_use_fabs_v2bf16(ptr addrspace(1) %out0,
; GFX11-NEXT: s_load_b128 s[0:3], s[4:5], 0x0
; GFX11-NEXT: s_waitcnt lgkmcnt(0)
; GFX11-NEXT: s_and_b32 s4, s6, 0x7fff7fff
; GFX11-NEXT: s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_1) | instid1(SALU_CYCLE_1)
; GFX11-NEXT: s_or_b32 s5, s6, 0x80008000
; GFX11-NEXT: v_dual_mov_b32 v0, 0 :: v_dual_mov_b32 v1, s4
; GFX11-NEXT: s_xor_b32 s5, s4, 0x80008000
; GFX11-NEXT: v_mov_b32_e32 v2, s5
; GFX11-NEXT: s_clause 0x1
; GFX11-NEXT: global_store_b32 v0, v1, s[0:1]
Expand Down