maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/llvm/test/CodeGen/AMDGPU/fmad-formation-fmul-distrib...

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

261 lines
11 KiB
LLVM
Raw Normal View History

DAG: Fix wrong legality check for ISD::FMAD Since 1725f2884175ca618d29b06e35f5c6ebd618053d, this should check isFMADLegalForFAddFSub rather than the the plain isOperationLegal. This would assert in a subset of cases due to an oddity in how FMAD is selected. We will allow FMA formation pre-legalize, but not FMAD even in cases where it would be valid. The current hook requires passing in the root fadd/fsub. However, in this distributed case, this would be far more complicated to pass in the relevant operand. AMDGPU doesn't get any value from the node, and only needs the type and is the only implementor, so I'm not sure why we have this complexity. Just rename and expand the assert to avoid the more complicated checks spread through the distribution logic.
2020-04-12 21:47:03 +08:00
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
[NFC] Removed unused prefixes in test/CodeGen/AMDGPU More patches to follow. This covers the pertinent tests starting with e, f, and g. Differential Revision: https://reviews.llvm.org/D94124
2021-01-06 06:37:29 +08:00
; RUN: llc -march=amdgcn -mcpu=tahiti -denormal-fp-math-f32=ieee < %s | FileCheck --check-prefix=FMA %s
; RUN: llc -march=amdgcn -mcpu=verde -denormal-fp-math-f32=ieee < %s | FileCheck --check-prefix=NOFUSE %s
; RUN: llc -march=amdgcn -mcpu=fiji -denormal-fp-math-f32=ieee < %s | FileCheck --check-prefix=NOFUSE %s
; RUN: llc -march=amdgcn -mcpu=tonga -denormal-fp-math-f32=ieee < %s | FileCheck --check-prefix=NOFUSE %s
; RUN: llc -march=amdgcn -mcpu=gfx900 -denormal-fp-math-f32=ieee < %s | FileCheck --check-prefix=FMA %s
[AMDGPU] Extend gfx10 test coverage. NFC. Differential Revision: https://reviews.llvm.org/D99267
2021-03-29 17:12:46 +08:00
; RUN: llc -march=amdgcn -mcpu=gfx1010 -denormal-fp-math-f32=ieee < %s | FileCheck --check-prefix=FMAGFX10 %s
DAG: Fix wrong legality check for ISD::FMAD Since 1725f2884175ca618d29b06e35f5c6ebd618053d, this should check isFMADLegalForFAddFSub rather than the the plain isOperationLegal. This would assert in a subset of cases due to an oddity in how FMAD is selected. We will allow FMA formation pre-legalize, but not FMAD even in cases where it would be valid. The current hook requires passing in the root fadd/fsub. However, in this distributed case, this would be far more complicated to pass in the relevant operand. AMDGPU doesn't get any value from the node, and only needs the type and is the only implementor, so I'm not sure why we have this complexity. Just rename and expand the assert to avoid the more complicated checks spread through the distribution logic.
2020-04-12 21:47:03 +08:00
[NFC] Removed unused prefixes in test/CodeGen/AMDGPU More patches to follow. This covers the pertinent tests starting with e, f, and g. Differential Revision: https://reviews.llvm.org/D94124
2021-01-06 06:37:29 +08:00
; RUN: llc -march=amdgcn -mcpu=tahiti -denormal-fp-math-f32=preserve-sign < %s | FileCheck --check-prefix=FMAD %s
; RUN: llc -march=amdgcn -mcpu=verde -denormal-fp-math-f32=preserve-sign < %s | FileCheck --check-prefix=FMAD %s
; RUN: llc -march=amdgcn -mcpu=fiji -denormal-fp-math-f32=preserve-sign < %s | FileCheck --check-prefix=FMAD %s
; RUN: llc -march=amdgcn -mcpu=tonga -denormal-fp-math-f32=preserve-sign < %s | FileCheck --check-prefix=FMAD %s
; RUN: llc -march=amdgcn -mcpu=gfx900 -denormal-fp-math-f32=preserve-sign < %s | FileCheck --check-prefix=FMAD %s
[AMDGPU] Extend gfx10 test coverage. NFC. Differential Revision: https://reviews.llvm.org/D99267
2021-03-29 17:12:46 +08:00
; RUN: llc -march=amdgcn -mcpu=gfx1010 -denormal-fp-math-f32=preserve-sign < %s | FileCheck --check-prefix=FMADGFX10 %s
DAG: Fix wrong legality check for ISD::FMAD Since 1725f2884175ca618d29b06e35f5c6ebd618053d, this should check isFMADLegalForFAddFSub rather than the the plain isOperationLegal. This would assert in a subset of cases due to an oddity in how FMAD is selected. We will allow FMA formation pre-legalize, but not FMAD even in cases where it would be valid. The current hook requires passing in the root fadd/fsub. However, in this distributed case, this would be far more complicated to pass in the relevant operand. AMDGPU doesn't get any value from the node, and only needs the type and is the only implementor, so I'm not sure why we have this complexity. Just rename and expand the assert to avoid the more complicated checks spread through the distribution logic.
2020-04-12 21:47:03 +08:00
; Check for incorrect fmad formation when distributing
define float @unsafe_fmul_fadd_distribute_fast_f32(float %arg0, float %arg1) #0 {
; FMA-LABEL: unsafe_fmul_fadd_distribute_fast_f32:
; FMA: ; %bb.0:
; FMA-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; FMA-NEXT: v_fma_f32 v0, v1, v0, v0
; FMA-NEXT: s_setpc_b64 s[30:31]
;
; NOFUSE-LABEL: unsafe_fmul_fadd_distribute_fast_f32:
; NOFUSE: ; %bb.0:
; NOFUSE-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; NOFUSE-NEXT: v_add_f32_e32 v1, 1.0, v1
; NOFUSE-NEXT: v_mul_f32_e32 v0, v0, v1
; NOFUSE-NEXT: s_setpc_b64 s[30:31]
;
[AMDGPU] Extend gfx10 test coverage. NFC. Differential Revision: https://reviews.llvm.org/D99267
2021-03-29 17:12:46 +08:00
; FMAGFX10-LABEL: unsafe_fmul_fadd_distribute_fast_f32:
; FMAGFX10: ; %bb.0:
; FMAGFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; FMAGFX10-NEXT: s_waitcnt_vscnt null, 0x0
; FMAGFX10-NEXT: v_fmac_f32_e32 v0, v1, v0
; FMAGFX10-NEXT: s_setpc_b64 s[30:31]
;
DAG: Fix wrong legality check for ISD::FMAD Since 1725f2884175ca618d29b06e35f5c6ebd618053d, this should check isFMADLegalForFAddFSub rather than the the plain isOperationLegal. This would assert in a subset of cases due to an oddity in how FMAD is selected. We will allow FMA formation pre-legalize, but not FMAD even in cases where it would be valid. The current hook requires passing in the root fadd/fsub. However, in this distributed case, this would be far more complicated to pass in the relevant operand. AMDGPU doesn't get any value from the node, and only needs the type and is the only implementor, so I'm not sure why we have this complexity. Just rename and expand the assert to avoid the more complicated checks spread through the distribution logic.
2020-04-12 21:47:03 +08:00
; FMAD-LABEL: unsafe_fmul_fadd_distribute_fast_f32:
; FMAD: ; %bb.0:
; FMAD-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; FMAD-NEXT: v_mac_f32_e32 v0, v1, v0
; FMAD-NEXT: s_setpc_b64 s[30:31]
[AMDGPU] Extend gfx10 test coverage. NFC. Differential Revision: https://reviews.llvm.org/D99267
2021-03-29 17:12:46 +08:00
;
; FMADGFX10-LABEL: unsafe_fmul_fadd_distribute_fast_f32:
; FMADGFX10: ; %bb.0:
; FMADGFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; FMADGFX10-NEXT: s_waitcnt_vscnt null, 0x0
; FMADGFX10-NEXT: v_fmac_f32_e32 v0, v1, v0
; FMADGFX10-NEXT: s_setpc_b64 s[30:31]
DAG: Fix wrong legality check for ISD::FMAD Since 1725f2884175ca618d29b06e35f5c6ebd618053d, this should check isFMADLegalForFAddFSub rather than the the plain isOperationLegal. This would assert in a subset of cases due to an oddity in how FMAD is selected. We will allow FMA formation pre-legalize, but not FMAD even in cases where it would be valid. The current hook requires passing in the root fadd/fsub. However, in this distributed case, this would be far more complicated to pass in the relevant operand. AMDGPU doesn't get any value from the node, and only needs the type and is the only implementor, so I'm not sure why we have this complexity. Just rename and expand the assert to avoid the more complicated checks spread through the distribution logic.
2020-04-12 21:47:03 +08:00
%add = fadd fast float %arg1, 1.0
%tmp1 = fmul fast float %arg0, %add
ret float %tmp1
}
define float @unsafe_fmul_fsub_distribute_fast_f32(float %arg0, float %arg1) #0 {
; FMA-LABEL: unsafe_fmul_fsub_distribute_fast_f32:
; FMA: ; %bb.0:
; FMA-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; FMA-NEXT: v_fma_f32 v0, -v1, v0, v0
; FMA-NEXT: s_setpc_b64 s[30:31]
;
; NOFUSE-LABEL: unsafe_fmul_fsub_distribute_fast_f32:
; NOFUSE: ; %bb.0:
; NOFUSE-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; NOFUSE-NEXT: v_sub_f32_e32 v1, 1.0, v1
; NOFUSE-NEXT: v_mul_f32_e32 v0, v0, v1
; NOFUSE-NEXT: s_setpc_b64 s[30:31]
;
[AMDGPU] Extend gfx10 test coverage. NFC. Differential Revision: https://reviews.llvm.org/D99267
2021-03-29 17:12:46 +08:00
; FMAGFX10-LABEL: unsafe_fmul_fsub_distribute_fast_f32:
; FMAGFX10: ; %bb.0:
; FMAGFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; FMAGFX10-NEXT: s_waitcnt_vscnt null, 0x0
; FMAGFX10-NEXT: v_fmac_f32_e64 v0, -v1, v0
; FMAGFX10-NEXT: s_setpc_b64 s[30:31]
;
DAG: Fix wrong legality check for ISD::FMAD Since 1725f2884175ca618d29b06e35f5c6ebd618053d, this should check isFMADLegalForFAddFSub rather than the the plain isOperationLegal. This would assert in a subset of cases due to an oddity in how FMAD is selected. We will allow FMA formation pre-legalize, but not FMAD even in cases where it would be valid. The current hook requires passing in the root fadd/fsub. However, in this distributed case, this would be far more complicated to pass in the relevant operand. AMDGPU doesn't get any value from the node, and only needs the type and is the only implementor, so I'm not sure why we have this complexity. Just rename and expand the assert to avoid the more complicated checks spread through the distribution logic.
2020-04-12 21:47:03 +08:00
; FMAD-LABEL: unsafe_fmul_fsub_distribute_fast_f32:
; FMAD: ; %bb.0:
; FMAD-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; FMAD-NEXT: v_mad_f32 v0, -v1, v0, v0
; FMAD-NEXT: s_setpc_b64 s[30:31]
[AMDGPU] Extend gfx10 test coverage. NFC. Differential Revision: https://reviews.llvm.org/D99267
2021-03-29 17:12:46 +08:00
;
; FMADGFX10-LABEL: unsafe_fmul_fsub_distribute_fast_f32:
; FMADGFX10: ; %bb.0:
; FMADGFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; FMADGFX10-NEXT: s_waitcnt_vscnt null, 0x0
; FMADGFX10-NEXT: v_fmac_f32_e64 v0, -v1, v0
; FMADGFX10-NEXT: s_setpc_b64 s[30:31]
DAG: Fix wrong legality check for ISD::FMAD Since 1725f2884175ca618d29b06e35f5c6ebd618053d, this should check isFMADLegalForFAddFSub rather than the the plain isOperationLegal. This would assert in a subset of cases due to an oddity in how FMAD is selected. We will allow FMA formation pre-legalize, but not FMAD even in cases where it would be valid. The current hook requires passing in the root fadd/fsub. However, in this distributed case, this would be far more complicated to pass in the relevant operand. AMDGPU doesn't get any value from the node, and only needs the type and is the only implementor, so I'm not sure why we have this complexity. Just rename and expand the assert to avoid the more complicated checks spread through the distribution logic.
2020-04-12 21:47:03 +08:00
%add = fsub fast float 1.0, %arg1
%tmp1 = fmul fast float %arg0, %add
ret float %tmp1
}
define <2 x float> @unsafe_fmul_fadd_distribute_fast_v2f32(<2 x float> %arg0, <2 x float> %arg1) #0 {
; FMA-LABEL: unsafe_fmul_fadd_distribute_fast_v2f32:
; FMA: ; %bb.0:
; FMA-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; FMA-NEXT: v_fma_f32 v0, v2, v0, v0
; FMA-NEXT: v_fma_f32 v1, v3, v1, v1
; FMA-NEXT: s_setpc_b64 s[30:31]
;
; NOFUSE-LABEL: unsafe_fmul_fadd_distribute_fast_v2f32:
; NOFUSE: ; %bb.0:
; NOFUSE-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; NOFUSE-NEXT: v_add_f32_e32 v3, 1.0, v3
; NOFUSE-NEXT: v_add_f32_e32 v2, 1.0, v2
; NOFUSE-NEXT: v_mul_f32_e32 v0, v0, v2
; NOFUSE-NEXT: v_mul_f32_e32 v1, v1, v3
; NOFUSE-NEXT: s_setpc_b64 s[30:31]
;
[AMDGPU] Extend gfx10 test coverage. NFC. Differential Revision: https://reviews.llvm.org/D99267
2021-03-29 17:12:46 +08:00
; FMAGFX10-LABEL: unsafe_fmul_fadd_distribute_fast_v2f32:
; FMAGFX10: ; %bb.0:
; FMAGFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; FMAGFX10-NEXT: s_waitcnt_vscnt null, 0x0
; FMAGFX10-NEXT: v_fmac_f32_e32 v0, v2, v0
; FMAGFX10-NEXT: v_fmac_f32_e32 v1, v3, v1
; FMAGFX10-NEXT: s_setpc_b64 s[30:31]
;
DAG: Fix wrong legality check for ISD::FMAD Since 1725f2884175ca618d29b06e35f5c6ebd618053d, this should check isFMADLegalForFAddFSub rather than the the plain isOperationLegal. This would assert in a subset of cases due to an oddity in how FMAD is selected. We will allow FMA formation pre-legalize, but not FMAD even in cases where it would be valid. The current hook requires passing in the root fadd/fsub. However, in this distributed case, this would be far more complicated to pass in the relevant operand. AMDGPU doesn't get any value from the node, and only needs the type and is the only implementor, so I'm not sure why we have this complexity. Just rename and expand the assert to avoid the more complicated checks spread through the distribution logic.
2020-04-12 21:47:03 +08:00
; FMAD-LABEL: unsafe_fmul_fadd_distribute_fast_v2f32:
; FMAD: ; %bb.0:
; FMAD-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; FMAD-NEXT: v_mac_f32_e32 v0, v2, v0
; FMAD-NEXT: v_mac_f32_e32 v1, v3, v1
; FMAD-NEXT: s_setpc_b64 s[30:31]
[AMDGPU] Extend gfx10 test coverage. NFC. Differential Revision: https://reviews.llvm.org/D99267
2021-03-29 17:12:46 +08:00
;
; FMADGFX10-LABEL: unsafe_fmul_fadd_distribute_fast_v2f32:
; FMADGFX10: ; %bb.0:
; FMADGFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; FMADGFX10-NEXT: s_waitcnt_vscnt null, 0x0
; FMADGFX10-NEXT: v_fmac_f32_e32 v0, v2, v0
; FMADGFX10-NEXT: v_fmac_f32_e32 v1, v3, v1
; FMADGFX10-NEXT: s_setpc_b64 s[30:31]
DAG: Fix wrong legality check for ISD::FMAD Since 1725f2884175ca618d29b06e35f5c6ebd618053d, this should check isFMADLegalForFAddFSub rather than the the plain isOperationLegal. This would assert in a subset of cases due to an oddity in how FMAD is selected. We will allow FMA formation pre-legalize, but not FMAD even in cases where it would be valid. The current hook requires passing in the root fadd/fsub. However, in this distributed case, this would be far more complicated to pass in the relevant operand. AMDGPU doesn't get any value from the node, and only needs the type and is the only implementor, so I'm not sure why we have this complexity. Just rename and expand the assert to avoid the more complicated checks spread through the distribution logic.
2020-04-12 21:47:03 +08:00
%add = fadd fast <2 x float> %arg1, <float 1.0, float 1.0>
%tmp1 = fmul fast <2 x float> %arg0, %add
ret <2 x float> %tmp1
}
define <2 x float> @unsafe_fmul_fsub_distribute_fast_v2f32(<2 x float> %arg0, <2 x float> %arg1) #0 {
; FMA-LABEL: unsafe_fmul_fsub_distribute_fast_v2f32:
; FMA: ; %bb.0:
; FMA-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; FMA-NEXT: v_fma_f32 v0, -v2, v0, v0
; FMA-NEXT: v_fma_f32 v1, -v3, v1, v1
; FMA-NEXT: s_setpc_b64 s[30:31]
;
; NOFUSE-LABEL: unsafe_fmul_fsub_distribute_fast_v2f32:
; NOFUSE: ; %bb.0:
; NOFUSE-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; NOFUSE-NEXT: v_sub_f32_e32 v3, 1.0, v3
; NOFUSE-NEXT: v_sub_f32_e32 v2, 1.0, v2
; NOFUSE-NEXT: v_mul_f32_e32 v0, v0, v2
; NOFUSE-NEXT: v_mul_f32_e32 v1, v1, v3
; NOFUSE-NEXT: s_setpc_b64 s[30:31]
;
[AMDGPU] Extend gfx10 test coverage. NFC. Differential Revision: https://reviews.llvm.org/D99267
2021-03-29 17:12:46 +08:00
; FMAGFX10-LABEL: unsafe_fmul_fsub_distribute_fast_v2f32:
; FMAGFX10: ; %bb.0:
; FMAGFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; FMAGFX10-NEXT: s_waitcnt_vscnt null, 0x0
; FMAGFX10-NEXT: v_fmac_f32_e64 v0, -v2, v0
; FMAGFX10-NEXT: v_fmac_f32_e64 v1, -v3, v1
; FMAGFX10-NEXT: s_setpc_b64 s[30:31]
;
DAG: Fix wrong legality check for ISD::FMAD Since 1725f2884175ca618d29b06e35f5c6ebd618053d, this should check isFMADLegalForFAddFSub rather than the the plain isOperationLegal. This would assert in a subset of cases due to an oddity in how FMAD is selected. We will allow FMA formation pre-legalize, but not FMAD even in cases where it would be valid. The current hook requires passing in the root fadd/fsub. However, in this distributed case, this would be far more complicated to pass in the relevant operand. AMDGPU doesn't get any value from the node, and only needs the type and is the only implementor, so I'm not sure why we have this complexity. Just rename and expand the assert to avoid the more complicated checks spread through the distribution logic.
2020-04-12 21:47:03 +08:00
; FMAD-LABEL: unsafe_fmul_fsub_distribute_fast_v2f32:
; FMAD: ; %bb.0:
; FMAD-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; FMAD-NEXT: v_mad_f32 v0, -v2, v0, v0
; FMAD-NEXT: v_mad_f32 v1, -v3, v1, v1
; FMAD-NEXT: s_setpc_b64 s[30:31]
[AMDGPU] Extend gfx10 test coverage. NFC. Differential Revision: https://reviews.llvm.org/D99267
2021-03-29 17:12:46 +08:00
;
; FMADGFX10-LABEL: unsafe_fmul_fsub_distribute_fast_v2f32:
; FMADGFX10: ; %bb.0:
; FMADGFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; FMADGFX10-NEXT: s_waitcnt_vscnt null, 0x0
; FMADGFX10-NEXT: v_fmac_f32_e64 v0, -v2, v0
; FMADGFX10-NEXT: v_fmac_f32_e64 v1, -v3, v1
; FMADGFX10-NEXT: s_setpc_b64 s[30:31]
DAG: Fix wrong legality check for ISD::FMAD Since 1725f2884175ca618d29b06e35f5c6ebd618053d, this should check isFMADLegalForFAddFSub rather than the the plain isOperationLegal. This would assert in a subset of cases due to an oddity in how FMAD is selected. We will allow FMA formation pre-legalize, but not FMAD even in cases where it would be valid. The current hook requires passing in the root fadd/fsub. However, in this distributed case, this would be far more complicated to pass in the relevant operand. AMDGPU doesn't get any value from the node, and only needs the type and is the only implementor, so I'm not sure why we have this complexity. Just rename and expand the assert to avoid the more complicated checks spread through the distribution logic.
2020-04-12 21:47:03 +08:00
%add = fsub fast <2 x float> <float 1.0, float 1.0>, %arg1
%tmp1 = fmul fast <2 x float> %arg0, %add
ret <2 x float> %tmp1
}
define <2 x float> @unsafe_fast_fmul_fadd_distribute_post_legalize_f32(float %arg0, <2 x float> %arg1) #0 {
; FMA-LABEL: unsafe_fast_fmul_fadd_distribute_post_legalize_f32:
; FMA: ; %bb.0:
; FMA-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; FMA-NEXT: v_fma_f32 v0, v0, v1, v1
; FMA-NEXT: s_setpc_b64 s[30:31]
;
; NOFUSE-LABEL: unsafe_fast_fmul_fadd_distribute_post_legalize_f32:
; NOFUSE: ; %bb.0:
; NOFUSE-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; NOFUSE-NEXT: v_add_f32_e32 v0, 1.0, v0
; NOFUSE-NEXT: v_mul_f32_e32 v0, v1, v0
; NOFUSE-NEXT: s_setpc_b64 s[30:31]
;
[AMDGPU] Extend gfx10 test coverage. NFC. Differential Revision: https://reviews.llvm.org/D99267
2021-03-29 17:12:46 +08:00
; FMAGFX10-LABEL: unsafe_fast_fmul_fadd_distribute_post_legalize_f32:
; FMAGFX10: ; %bb.0:
; FMAGFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; FMAGFX10-NEXT: s_waitcnt_vscnt null, 0x0
; FMAGFX10-NEXT: v_fma_f32 v0, v0, v1, v1
; FMAGFX10-NEXT: s_setpc_b64 s[30:31]
;
DAG: Fix wrong legality check for ISD::FMAD Since 1725f2884175ca618d29b06e35f5c6ebd618053d, this should check isFMADLegalForFAddFSub rather than the the plain isOperationLegal. This would assert in a subset of cases due to an oddity in how FMAD is selected. We will allow FMA formation pre-legalize, but not FMAD even in cases where it would be valid. The current hook requires passing in the root fadd/fsub. However, in this distributed case, this would be far more complicated to pass in the relevant operand. AMDGPU doesn't get any value from the node, and only needs the type and is the only implementor, so I'm not sure why we have this complexity. Just rename and expand the assert to avoid the more complicated checks spread through the distribution logic.
2020-04-12 21:47:03 +08:00
; FMAD-LABEL: unsafe_fast_fmul_fadd_distribute_post_legalize_f32:
; FMAD: ; %bb.0:
; FMAD-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; FMAD-NEXT: v_mad_f32 v0, v0, v1, v1
; FMAD-NEXT: s_setpc_b64 s[30:31]
[AMDGPU] Extend gfx10 test coverage. NFC. Differential Revision: https://reviews.llvm.org/D99267
2021-03-29 17:12:46 +08:00
;
; FMADGFX10-LABEL: unsafe_fast_fmul_fadd_distribute_post_legalize_f32:
; FMADGFX10: ; %bb.0:
; FMADGFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; FMADGFX10-NEXT: s_waitcnt_vscnt null, 0x0
; FMADGFX10-NEXT: v_mad_f32 v0, v0, v1, v1
; FMADGFX10-NEXT: s_setpc_b64 s[30:31]
DAG: Fix wrong legality check for ISD::FMAD Since 1725f2884175ca618d29b06e35f5c6ebd618053d, this should check isFMADLegalForFAddFSub rather than the the plain isOperationLegal. This would assert in a subset of cases due to an oddity in how FMAD is selected. We will allow FMA formation pre-legalize, but not FMAD even in cases where it would be valid. The current hook requires passing in the root fadd/fsub. However, in this distributed case, this would be far more complicated to pass in the relevant operand. AMDGPU doesn't get any value from the node, and only needs the type and is the only implementor, so I'm not sure why we have this complexity. Just rename and expand the assert to avoid the more complicated checks spread through the distribution logic.
2020-04-12 21:47:03 +08:00
%add = fadd fast float %arg0, 1.0
%splat = insertelement <2 x float> undef, float %add, i32 0
%tmp1 = fmul fast <2 x float> %arg1, %splat
ret <2 x float> %tmp1
}
define <2 x float> @unsafe_fast_fmul_fsub_ditribute_post_legalize(float %arg0, <2 x float> %arg1) #0 {
; FMA-LABEL: unsafe_fast_fmul_fsub_ditribute_post_legalize:
; FMA: ; %bb.0:
; FMA-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; FMA-NEXT: v_fma_f32 v0, -v0, v1, v1
; FMA-NEXT: s_setpc_b64 s[30:31]
;
; NOFUSE-LABEL: unsafe_fast_fmul_fsub_ditribute_post_legalize:
; NOFUSE: ; %bb.0:
; NOFUSE-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; NOFUSE-NEXT: v_sub_f32_e32 v0, 1.0, v0
; NOFUSE-NEXT: v_mul_f32_e32 v0, v1, v0
; NOFUSE-NEXT: s_setpc_b64 s[30:31]
;
[AMDGPU] Extend gfx10 test coverage. NFC. Differential Revision: https://reviews.llvm.org/D99267
2021-03-29 17:12:46 +08:00
; FMAGFX10-LABEL: unsafe_fast_fmul_fsub_ditribute_post_legalize:
; FMAGFX10: ; %bb.0:
; FMAGFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; FMAGFX10-NEXT: s_waitcnt_vscnt null, 0x0
; FMAGFX10-NEXT: v_fma_f32 v0, v1, -v0, v1
; FMAGFX10-NEXT: s_setpc_b64 s[30:31]
;
DAG: Fix wrong legality check for ISD::FMAD Since 1725f2884175ca618d29b06e35f5c6ebd618053d, this should check isFMADLegalForFAddFSub rather than the the plain isOperationLegal. This would assert in a subset of cases due to an oddity in how FMAD is selected. We will allow FMA formation pre-legalize, but not FMAD even in cases where it would be valid. The current hook requires passing in the root fadd/fsub. However, in this distributed case, this would be far more complicated to pass in the relevant operand. AMDGPU doesn't get any value from the node, and only needs the type and is the only implementor, so I'm not sure why we have this complexity. Just rename and expand the assert to avoid the more complicated checks spread through the distribution logic.
2020-04-12 21:47:03 +08:00
; FMAD-LABEL: unsafe_fast_fmul_fsub_ditribute_post_legalize:
; FMAD: ; %bb.0:
; FMAD-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; FMAD-NEXT: v_mad_f32 v0, -v0, v1, v1
; FMAD-NEXT: s_setpc_b64 s[30:31]
[AMDGPU] Extend gfx10 test coverage. NFC. Differential Revision: https://reviews.llvm.org/D99267
2021-03-29 17:12:46 +08:00
;
; FMADGFX10-LABEL: unsafe_fast_fmul_fsub_ditribute_post_legalize:
; FMADGFX10: ; %bb.0:
; FMADGFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; FMADGFX10-NEXT: s_waitcnt_vscnt null, 0x0
; FMADGFX10-NEXT: v_mad_f32 v0, -v0, v1, v1
; FMADGFX10-NEXT: s_setpc_b64 s[30:31]
DAG: Fix wrong legality check for ISD::FMAD Since 1725f2884175ca618d29b06e35f5c6ebd618053d, this should check isFMADLegalForFAddFSub rather than the the plain isOperationLegal. This would assert in a subset of cases due to an oddity in how FMAD is selected. We will allow FMA formation pre-legalize, but not FMAD even in cases where it would be valid. The current hook requires passing in the root fadd/fsub. However, in this distributed case, this would be far more complicated to pass in the relevant operand. AMDGPU doesn't get any value from the node, and only needs the type and is the only implementor, so I'm not sure why we have this complexity. Just rename and expand the assert to avoid the more complicated checks spread through the distribution logic.
2020-04-12 21:47:03 +08:00
%sub = fsub fast float 1.0, %arg0
%splat = insertelement <2 x float> undef, float %sub, i32 0
%tmp1 = fmul fast <2 x float> %arg1, %splat
ret <2 x float> %tmp1
}
attributes #0 = { "no-infs-fp-math"="true" "unsafe-fp-math"="true" }