forked from OSchip/llvm-project
2858 lines
112 KiB
LLVM
2858 lines
112 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
|
|
; RUN: llc -mtriple=amdgcn -mcpu=gfx700 -verify-machineinstrs < %s | FileCheck --check-prefixes=GFX7 %s
|
|
; RUN: llc -mtriple=amdgcn -mcpu=gfx803 -verify-machineinstrs < %s | FileCheck --check-prefixes=GFX8 %s
|
|
; RUN: llc -mtriple=amdgcn -mcpu=gfx900 -verify-machineinstrs < %s | FileCheck --check-prefixes=GFX9-NODL %s
|
|
; RUN: llc -mtriple=amdgcn -mcpu=gfx906 -verify-machineinstrs < %s | FileCheck --check-prefixes=GFX9-DL %s
|
|
; RUN: llc -mtriple=amdgcn -mcpu=gfx1011 -verify-machineinstrs < %s | FileCheck --check-prefixes=GFX10-DL %s
|
|
; RUN: llc -mtriple=amdgcn -mcpu=gfx1012 -verify-machineinstrs < %s | FileCheck --check-prefixes=GFX10-DL %s
|
|
|
|
; add(mul(S0.x, S1.y),
|
|
; add (mul (S0.y, S1.y), S3)) -> v_dot2_{I|U}32_{I|U}16(S1, S2, S3)
|
|
|
|
define amdgpu_kernel void @udot2(<2 x i16> addrspace(1)* %src1,
|
|
; GFX7-LABEL: udot2:
|
|
; GFX7: ; %bb.0: ; %entry
|
|
; GFX7-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x9
|
|
; GFX7-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0xd
|
|
; GFX7-NEXT: s_mov_b32 s8, 0xffff
|
|
; GFX7-NEXT: s_mov_b32 s3, 0xf000
|
|
; GFX7-NEXT: s_mov_b32 s2, -1
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_load_dword s4, s[4:5], 0x0
|
|
; GFX7-NEXT: s_load_dword s5, s[6:7], 0x0
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_lshr_b32 s6, s4, 16
|
|
; GFX7-NEXT: s_lshr_b32 s7, s5, 16
|
|
; GFX7-NEXT: s_and_b32 s4, s4, s8
|
|
; GFX7-NEXT: s_and_b32 s5, s5, s8
|
|
; GFX7-NEXT: s_load_dword s8, s[0:1], 0x0
|
|
; GFX7-NEXT: v_mov_b32_e32 v0, s6
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: v_mov_b32_e32 v1, s8
|
|
; GFX7-NEXT: v_mad_u32_u24 v0, s7, v0, v1
|
|
; GFX7-NEXT: v_mov_b32_e32 v1, s4
|
|
; GFX7-NEXT: v_mad_u32_u24 v0, s5, v1, v0
|
|
; GFX7-NEXT: buffer_store_dword v0, off, s[0:3], 0
|
|
; GFX7-NEXT: s_endpgm
|
|
;
|
|
; GFX8-LABEL: udot2:
|
|
; GFX8: ; %bb.0: ; %entry
|
|
; GFX8-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX8-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX8-NEXT: s_mov_b32 s2, 0xffff
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_load_dword s3, s[4:5], 0x0
|
|
; GFX8-NEXT: s_load_dword s4, s[6:7], 0x0
|
|
; GFX8-NEXT: s_load_dword s5, s[0:1], 0x0
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_and_b32 s6, s3, s2
|
|
; GFX8-NEXT: s_lshr_b32 s3, s3, 16
|
|
; GFX8-NEXT: s_and_b32 s2, s4, s2
|
|
; GFX8-NEXT: s_lshr_b32 s4, s4, 16
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s5
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s3
|
|
; GFX8-NEXT: v_mad_u32_u24 v0, s4, v1, v0
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s6
|
|
; GFX8-NEXT: v_mad_u32_u24 v2, s2, v1, v0
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX8-NEXT: flat_store_dword v[0:1], v2
|
|
; GFX8-NEXT: s_endpgm
|
|
;
|
|
; GFX9-NODL-LABEL: udot2:
|
|
; GFX9-NODL: ; %bb.0: ; %entry
|
|
; GFX9-NODL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-NODL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-NODL-NEXT: s_mov_b32 s2, 0xffff
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_load_dword s3, s[4:5], 0x0
|
|
; GFX9-NODL-NEXT: s_load_dword s4, s[6:7], 0x0
|
|
; GFX9-NODL-NEXT: s_load_dword s5, s[0:1], 0x0
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_and_b32 s6, s3, s2
|
|
; GFX9-NODL-NEXT: s_lshr_b32 s3, s3, 16
|
|
; GFX9-NODL-NEXT: s_and_b32 s2, s4, s2
|
|
; GFX9-NODL-NEXT: s_lshr_b32 s4, s4, 16
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s5
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s3
|
|
; GFX9-NODL-NEXT: v_mad_u32_u24 v0, s4, v1, v0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s6
|
|
; GFX9-NODL-NEXT: v_mad_u32_u24 v2, s2, v1, v0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-NODL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-NODL-NEXT: s_endpgm
|
|
;
|
|
; GFX9-DL-LABEL: udot2:
|
|
; GFX9-DL: ; %bb.0: ; %entry
|
|
; GFX9-DL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-DL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: s_load_dword s2, s[4:5], 0x0
|
|
; GFX9-DL-NEXT: s_load_dword s3, s[0:1], 0x0
|
|
; GFX9-DL-NEXT: s_load_dword s4, s[6:7], 0x0
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s2
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s3
|
|
; GFX9-DL-NEXT: v_dot2_u32_u16 v2, s4, v0, v1
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-DL-NEXT: s_endpgm
|
|
;
|
|
; GFX10-DL-LABEL: udot2:
|
|
; GFX10-DL: ; %bb.0: ; %entry
|
|
; GFX10-DL-NEXT: s_clause 0x1
|
|
; GFX10-DL-NEXT: s_load_dwordx2 s[4:5], s[0:1], 0x34
|
|
; GFX10-DL-NEXT: s_load_dwordx4 s[0:3], s[0:1], 0x24
|
|
; GFX10-DL-NEXT: ; implicit-def: $vcc_hi
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: s_load_dword s6, s[4:5], 0x0
|
|
; GFX10-DL-NEXT: s_load_dword s0, s[0:1], 0x0
|
|
; GFX10-DL-NEXT: s_load_dword s1, s[2:3], 0x0
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s6
|
|
; GFX10-DL-NEXT: v_dot2_u32_u16 v2, s1, s0, v0
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v1, s5
|
|
; GFX10-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX10-DL-NEXT: s_endpgm
|
|
<2 x i16> addrspace(1)* %src2,
|
|
i32 addrspace(1)* nocapture %dst) {
|
|
entry:
|
|
%vec1 = load <2 x i16>, <2 x i16> addrspace(1)* %src1
|
|
%vec2 = load <2 x i16>, <2 x i16> addrspace(1)* %src2
|
|
|
|
%s1.elt1 = extractelement <2 x i16> %vec1, i64 0
|
|
%conv = zext i16 %s1.elt1 to i32
|
|
%s2.elt1 = extractelement <2 x i16> %vec2, i64 0
|
|
%conv2 = zext i16 %s2.elt1 to i32
|
|
%mul1 = mul nuw i32 %conv2, %conv
|
|
|
|
%s1.elt2 = extractelement <2 x i16> %vec1, i64 1
|
|
%conv3 = zext i16 %s1.elt2 to i32
|
|
%s2.elt2 = extractelement <2 x i16> %vec2, i64 1
|
|
%conv4 = zext i16 %s2.elt2 to i32
|
|
%mul2 = mul nuw i32 %conv4, %conv3
|
|
|
|
%s3 = load i32, i32 addrspace(1)* %dst, align 4
|
|
%add = add i32 %mul2, %s3
|
|
%add6 = add i32 %add, %mul1
|
|
store i32 %add6, i32 addrspace(1)* %dst, align 4
|
|
ret void
|
|
}
|
|
|
|
; TODO: Support this pattern
|
|
; add(S3,
|
|
; add (mul (S0.y, S1.y), mul (S0.y, S1.y))) -> v_dot2_{I|U}32_{I|U}16(S1, S2, S3)
|
|
define amdgpu_kernel void @udot2_MulMul(<2 x i16> addrspace(1)* %src1,
|
|
; GFX7-LABEL: udot2_MulMul:
|
|
; GFX7: ; %bb.0: ; %entry
|
|
; GFX7-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x9
|
|
; GFX7-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0xd
|
|
; GFX7-NEXT: s_mov_b32 s8, 0xffff
|
|
; GFX7-NEXT: s_mov_b32 s3, 0xf000
|
|
; GFX7-NEXT: s_mov_b32 s2, -1
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_load_dword s4, s[4:5], 0x0
|
|
; GFX7-NEXT: s_load_dword s5, s[6:7], 0x0
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_lshr_b32 s6, s4, 16
|
|
; GFX7-NEXT: s_and_b32 s4, s4, s8
|
|
; GFX7-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX7-NEXT: s_load_dword s4, s[0:1], 0x0
|
|
; GFX7-NEXT: s_lshr_b32 s7, s5, 16
|
|
; GFX7-NEXT: s_and_b32 s5, s5, s8
|
|
; GFX7-NEXT: v_mul_u32_u24_e32 v0, s5, v0
|
|
; GFX7-NEXT: v_mov_b32_e32 v1, s6
|
|
; GFX7-NEXT: v_mad_u32_u24 v0, s7, v1, v0
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: v_add_i32_e32 v0, vcc, s4, v0
|
|
; GFX7-NEXT: buffer_store_dword v0, off, s[0:3], 0
|
|
; GFX7-NEXT: s_endpgm
|
|
;
|
|
; GFX8-LABEL: udot2_MulMul:
|
|
; GFX8: ; %bb.0: ; %entry
|
|
; GFX8-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX8-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX8-NEXT: s_mov_b32 s2, 0xffff
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_load_dword s3, s[4:5], 0x0
|
|
; GFX8-NEXT: s_load_dword s4, s[6:7], 0x0
|
|
; GFX8-NEXT: s_load_dword s5, s[0:1], 0x0
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_and_b32 s6, s3, s2
|
|
; GFX8-NEXT: s_and_b32 s2, s4, s2
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s6
|
|
; GFX8-NEXT: s_lshr_b32 s3, s3, 16
|
|
; GFX8-NEXT: s_lshr_b32 s4, s4, 16
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s3
|
|
; GFX8-NEXT: v_mul_u32_u24_e32 v0, s2, v0
|
|
; GFX8-NEXT: v_mad_u32_u24 v0, s4, v1, v0
|
|
; GFX8-NEXT: v_add_u32_e32 v2, vcc, s5, v0
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX8-NEXT: flat_store_dword v[0:1], v2
|
|
; GFX8-NEXT: s_endpgm
|
|
;
|
|
; GFX9-NODL-LABEL: udot2_MulMul:
|
|
; GFX9-NODL: ; %bb.0: ; %entry
|
|
; GFX9-NODL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-NODL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-NODL-NEXT: s_mov_b32 s2, 0xffff
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_load_dword s3, s[4:5], 0x0
|
|
; GFX9-NODL-NEXT: s_load_dword s4, s[6:7], 0x0
|
|
; GFX9-NODL-NEXT: s_load_dword s5, s[0:1], 0x0
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_and_b32 s6, s3, s2
|
|
; GFX9-NODL-NEXT: s_and_b32 s2, s4, s2
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s6
|
|
; GFX9-NODL-NEXT: s_lshr_b32 s3, s3, 16
|
|
; GFX9-NODL-NEXT: s_lshr_b32 s4, s4, 16
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s3
|
|
; GFX9-NODL-NEXT: v_mul_u32_u24_e32 v0, s2, v0
|
|
; GFX9-NODL-NEXT: v_mad_u32_u24 v0, s4, v1, v0
|
|
; GFX9-NODL-NEXT: v_add_u32_e32 v2, s5, v0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-NODL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-NODL-NEXT: s_endpgm
|
|
;
|
|
; GFX9-DL-LABEL: udot2_MulMul:
|
|
; GFX9-DL: ; %bb.0: ; %entry
|
|
; GFX9-DL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-DL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-DL-NEXT: s_mov_b32 s2, 0xffff
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: s_load_dword s3, s[4:5], 0x0
|
|
; GFX9-DL-NEXT: s_load_dword s4, s[6:7], 0x0
|
|
; GFX9-DL-NEXT: s_load_dword s5, s[0:1], 0x0
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: s_and_b32 s6, s3, s2
|
|
; GFX9-DL-NEXT: s_and_b32 s2, s4, s2
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s6
|
|
; GFX9-DL-NEXT: s_lshr_b32 s3, s3, 16
|
|
; GFX9-DL-NEXT: s_lshr_b32 s4, s4, 16
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s3
|
|
; GFX9-DL-NEXT: v_mul_u32_u24_e32 v0, s2, v0
|
|
; GFX9-DL-NEXT: v_mad_u32_u24 v0, s4, v1, v0
|
|
; GFX9-DL-NEXT: v_add_u32_e32 v2, s5, v0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-DL-NEXT: s_endpgm
|
|
;
|
|
; GFX10-DL-LABEL: udot2_MulMul:
|
|
; GFX10-DL: ; %bb.0: ; %entry
|
|
; GFX10-DL-NEXT: s_clause 0x1
|
|
; GFX10-DL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX10-DL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX10-DL-NEXT: ; implicit-def: $vcc_hi
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: s_load_dword s2, s[4:5], 0x0
|
|
; GFX10-DL-NEXT: s_load_dword s3, s[6:7], 0x0
|
|
; GFX10-DL-NEXT: s_load_dword s4, s[0:1], 0x0
|
|
; GFX10-DL-NEXT: s_mov_b32 s5, 0xffff
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: s_and_b32 s6, s2, s5
|
|
; GFX10-DL-NEXT: s_and_b32 s5, s3, s5
|
|
; GFX10-DL-NEXT: s_lshr_b32 s2, s2, 16
|
|
; GFX10-DL-NEXT: v_mul_u32_u24_e64 v0, s5, s6
|
|
; GFX10-DL-NEXT: s_lshr_b32 s3, s3, 16
|
|
; GFX10-DL-NEXT: v_mad_u32_u24 v0, s3, s2, v0
|
|
; GFX10-DL-NEXT: v_add_nc_u32_e32 v2, s4, v0
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX10-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX10-DL-NEXT: s_endpgm
|
|
<2 x i16> addrspace(1)* %src2,
|
|
i32 addrspace(1)* nocapture %dst) {
|
|
entry:
|
|
%vec1 = load <2 x i16>, <2 x i16> addrspace(1)* %src1
|
|
%vec2 = load <2 x i16>, <2 x i16> addrspace(1)* %src2
|
|
|
|
%s1.elt1 = extractelement <2 x i16> %vec1, i64 0
|
|
%conv = zext i16 %s1.elt1 to i32
|
|
%s2.elt1 = extractelement <2 x i16> %vec2, i64 0
|
|
%conv2 = zext i16 %s2.elt1 to i32
|
|
%mul1 = mul nuw i32 %conv2, %conv
|
|
|
|
%s1.elt2 = extractelement <2 x i16> %vec1, i64 1
|
|
%conv3 = zext i16 %s1.elt2 to i32
|
|
%s2.elt2 = extractelement <2 x i16> %vec2, i64 1
|
|
%conv4 = zext i16 %s2.elt2 to i32
|
|
%mul2 = mul nuw i32 %conv4, %conv3
|
|
%s3 = load i32, i32 addrspace(1)* %dst, align 4
|
|
%add = add i32 %mul2, %mul1
|
|
%add6 = add i32 %add, %s3
|
|
store i32 %add6, i32 addrspace(1)* %dst, align 4
|
|
ret void
|
|
}
|
|
|
|
define amdgpu_kernel void @idot2(<2 x i16> addrspace(1)* %src1,
|
|
; GFX7-LABEL: idot2:
|
|
; GFX7: ; %bb.0: ; %entry
|
|
; GFX7-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x9
|
|
; GFX7-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0xd
|
|
; GFX7-NEXT: s_mov_b32 s3, 0xf000
|
|
; GFX7-NEXT: s_mov_b32 s2, -1
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_load_dword s4, s[4:5], 0x0
|
|
; GFX7-NEXT: s_load_dword s5, s[6:7], 0x0
|
|
; GFX7-NEXT: s_load_dword s6, s[0:1], 0x0
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_sext_i32_i16 s7, s4
|
|
; GFX7-NEXT: s_ashr_i32 s4, s4, 16
|
|
; GFX7-NEXT: s_sext_i32_i16 s8, s5
|
|
; GFX7-NEXT: s_ashr_i32 s5, s5, 16
|
|
; GFX7-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX7-NEXT: v_mov_b32_e32 v1, s6
|
|
; GFX7-NEXT: v_mad_i32_i24 v0, s5, v0, v1
|
|
; GFX7-NEXT: v_mov_b32_e32 v1, s7
|
|
; GFX7-NEXT: v_mad_i32_i24 v0, s8, v1, v0
|
|
; GFX7-NEXT: buffer_store_dword v0, off, s[0:3], 0
|
|
; GFX7-NEXT: s_endpgm
|
|
;
|
|
; GFX8-LABEL: idot2:
|
|
; GFX8: ; %bb.0: ; %entry
|
|
; GFX8-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX8-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_load_dword s2, s[4:5], 0x0
|
|
; GFX8-NEXT: s_load_dword s3, s[6:7], 0x0
|
|
; GFX8-NEXT: s_load_dword s4, s[0:1], 0x0
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_sext_i32_i16 s5, s2
|
|
; GFX8-NEXT: s_ashr_i32 s2, s2, 16
|
|
; GFX8-NEXT: s_sext_i32_i16 s6, s3
|
|
; GFX8-NEXT: s_ashr_i32 s3, s3, 16
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s2
|
|
; GFX8-NEXT: v_mov_b32_e32 v2, s5
|
|
; GFX8-NEXT: v_mad_i32_i24 v0, s3, v1, v0
|
|
; GFX8-NEXT: v_mad_i32_i24 v2, s6, v2, v0
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX8-NEXT: flat_store_dword v[0:1], v2
|
|
; GFX8-NEXT: s_endpgm
|
|
;
|
|
; GFX9-NODL-LABEL: idot2:
|
|
; GFX9-NODL: ; %bb.0: ; %entry
|
|
; GFX9-NODL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-NODL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_load_dword s2, s[4:5], 0x0
|
|
; GFX9-NODL-NEXT: s_load_dword s3, s[6:7], 0x0
|
|
; GFX9-NODL-NEXT: s_load_dword s4, s[0:1], 0x0
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_sext_i32_i16 s5, s2
|
|
; GFX9-NODL-NEXT: s_ashr_i32 s2, s2, 16
|
|
; GFX9-NODL-NEXT: s_sext_i32_i16 s6, s3
|
|
; GFX9-NODL-NEXT: s_ashr_i32 s3, s3, 16
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s2
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v2, s5
|
|
; GFX9-NODL-NEXT: v_mad_i32_i24 v0, s3, v1, v0
|
|
; GFX9-NODL-NEXT: v_mad_i32_i24 v2, s6, v2, v0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-NODL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-NODL-NEXT: s_endpgm
|
|
;
|
|
; GFX9-DL-LABEL: idot2:
|
|
; GFX9-DL: ; %bb.0: ; %entry
|
|
; GFX9-DL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-DL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: s_load_dword s2, s[4:5], 0x0
|
|
; GFX9-DL-NEXT: s_load_dword s3, s[0:1], 0x0
|
|
; GFX9-DL-NEXT: s_load_dword s4, s[6:7], 0x0
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s2
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s3
|
|
; GFX9-DL-NEXT: v_dot2_i32_i16 v2, s4, v0, v1
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-DL-NEXT: s_endpgm
|
|
;
|
|
; GFX10-DL-LABEL: idot2:
|
|
; GFX10-DL: ; %bb.0: ; %entry
|
|
; GFX10-DL-NEXT: s_clause 0x1
|
|
; GFX10-DL-NEXT: s_load_dwordx2 s[4:5], s[0:1], 0x34
|
|
; GFX10-DL-NEXT: s_load_dwordx4 s[0:3], s[0:1], 0x24
|
|
; GFX10-DL-NEXT: ; implicit-def: $vcc_hi
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: s_load_dword s6, s[4:5], 0x0
|
|
; GFX10-DL-NEXT: s_load_dword s0, s[0:1], 0x0
|
|
; GFX10-DL-NEXT: s_load_dword s1, s[2:3], 0x0
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s6
|
|
; GFX10-DL-NEXT: v_dot2_i32_i16 v2, s1, s0, v0
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v1, s5
|
|
; GFX10-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX10-DL-NEXT: s_endpgm
|
|
<2 x i16> addrspace(1)* %src2,
|
|
i32 addrspace(1)* nocapture %dst) {
|
|
entry:
|
|
%vec1 = load <2 x i16>, <2 x i16> addrspace(1)* %src1
|
|
%vec2 = load <2 x i16>, <2 x i16> addrspace(1)* %src2
|
|
|
|
%s1.elt1 = extractelement <2 x i16> %vec1, i64 0
|
|
%conv = sext i16 %s1.elt1 to i32
|
|
%s2.elt1 = extractelement <2 x i16> %vec2, i64 0
|
|
%conv2 = sext i16 %s2.elt1 to i32
|
|
%mul1 = mul nuw i32 %conv2, %conv
|
|
|
|
%s1.elt2 = extractelement <2 x i16> %vec1, i64 1
|
|
%conv3 = sext i16 %s1.elt2 to i32
|
|
%s2.elt2 = extractelement <2 x i16> %vec2, i64 1
|
|
%conv4 = sext i16 %s2.elt2 to i32
|
|
%mul2 = mul nuw i32 %conv4, %conv3
|
|
|
|
%s3 = load i32, i32 addrspace(1)* %dst, align 4
|
|
%add = add i32 %mul2, %s3
|
|
%add6 = add i32 %add, %mul1
|
|
store i32 %add6, i32 addrspace(1)* %dst, align 4
|
|
ret void
|
|
}
|
|
|
|
define amdgpu_kernel void @idot2_MixedTypedMul(<2 x i16> addrspace(1)* %src1,
|
|
; GFX7-LABEL: idot2_MixedTypedMul:
|
|
; GFX7: ; %bb.0: ; %entry
|
|
; GFX7-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x9
|
|
; GFX7-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0xd
|
|
; GFX7-NEXT: s_mov_b32 s3, 0xf000
|
|
; GFX7-NEXT: s_mov_b32 s2, -1
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_load_dword s4, s[4:5], 0x0
|
|
; GFX7-NEXT: s_load_dword s5, s[6:7], 0x0
|
|
; GFX7-NEXT: s_load_dword s6, s[0:1], 0x0
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_lshr_b32 s7, s4, 16
|
|
; GFX7-NEXT: s_lshr_b32 s8, s5, 16
|
|
; GFX7-NEXT: s_sext_i32_i16 s4, s4
|
|
; GFX7-NEXT: v_mov_b32_e32 v0, s7
|
|
; GFX7-NEXT: v_mov_b32_e32 v1, s6
|
|
; GFX7-NEXT: v_mad_u32_u24 v0, s8, v0, v1
|
|
; GFX7-NEXT: s_sext_i32_i16 s5, s5
|
|
; GFX7-NEXT: v_mov_b32_e32 v1, s4
|
|
; GFX7-NEXT: v_mad_i32_i24 v0, s5, v1, v0
|
|
; GFX7-NEXT: buffer_store_dword v0, off, s[0:3], 0
|
|
; GFX7-NEXT: s_endpgm
|
|
;
|
|
; GFX8-LABEL: idot2_MixedTypedMul:
|
|
; GFX8: ; %bb.0: ; %entry
|
|
; GFX8-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX8-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_load_dword s2, s[4:5], 0x0
|
|
; GFX8-NEXT: s_load_dword s3, s[6:7], 0x0
|
|
; GFX8-NEXT: s_load_dword s4, s[0:1], 0x0
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_sext_i32_i16 s5, s2
|
|
; GFX8-NEXT: s_lshr_b32 s2, s2, 16
|
|
; GFX8-NEXT: s_sext_i32_i16 s6, s3
|
|
; GFX8-NEXT: s_lshr_b32 s3, s3, 16
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s2
|
|
; GFX8-NEXT: v_mov_b32_e32 v2, s5
|
|
; GFX8-NEXT: v_mad_u32_u24 v0, s3, v1, v0
|
|
; GFX8-NEXT: v_mad_i32_i24 v2, s6, v2, v0
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX8-NEXT: flat_store_dword v[0:1], v2
|
|
; GFX8-NEXT: s_endpgm
|
|
;
|
|
; GFX9-NODL-LABEL: idot2_MixedTypedMul:
|
|
; GFX9-NODL: ; %bb.0: ; %entry
|
|
; GFX9-NODL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-NODL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_load_dword s2, s[4:5], 0x0
|
|
; GFX9-NODL-NEXT: s_load_dword s3, s[6:7], 0x0
|
|
; GFX9-NODL-NEXT: s_load_dword s4, s[0:1], 0x0
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_sext_i32_i16 s5, s2
|
|
; GFX9-NODL-NEXT: s_lshr_b32 s2, s2, 16
|
|
; GFX9-NODL-NEXT: s_sext_i32_i16 s6, s3
|
|
; GFX9-NODL-NEXT: s_lshr_b32 s3, s3, 16
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s2
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v2, s5
|
|
; GFX9-NODL-NEXT: v_mad_u32_u24 v0, s3, v1, v0
|
|
; GFX9-NODL-NEXT: v_mad_i32_i24 v2, s6, v2, v0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-NODL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-NODL-NEXT: s_endpgm
|
|
;
|
|
; GFX9-DL-LABEL: idot2_MixedTypedMul:
|
|
; GFX9-DL: ; %bb.0: ; %entry
|
|
; GFX9-DL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-DL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: s_load_dword s2, s[4:5], 0x0
|
|
; GFX9-DL-NEXT: s_load_dword s3, s[6:7], 0x0
|
|
; GFX9-DL-NEXT: s_load_dword s4, s[0:1], 0x0
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: s_sext_i32_i16 s5, s2
|
|
; GFX9-DL-NEXT: s_lshr_b32 s2, s2, 16
|
|
; GFX9-DL-NEXT: s_sext_i32_i16 s6, s3
|
|
; GFX9-DL-NEXT: s_lshr_b32 s3, s3, 16
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s2
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v2, s5
|
|
; GFX9-DL-NEXT: v_mad_u32_u24 v0, s3, v1, v0
|
|
; GFX9-DL-NEXT: v_mad_i32_i24 v2, s6, v2, v0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-DL-NEXT: s_endpgm
|
|
;
|
|
; GFX10-DL-LABEL: idot2_MixedTypedMul:
|
|
; GFX10-DL: ; %bb.0: ; %entry
|
|
; GFX10-DL-NEXT: s_clause 0x1
|
|
; GFX10-DL-NEXT: s_load_dwordx2 s[4:5], s[0:1], 0x34
|
|
; GFX10-DL-NEXT: s_load_dwordx4 s[0:3], s[0:1], 0x24
|
|
; GFX10-DL-NEXT: ; implicit-def: $vcc_hi
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: s_load_dword s6, s[4:5], 0x0
|
|
; GFX10-DL-NEXT: s_load_dword s0, s[0:1], 0x0
|
|
; GFX10-DL-NEXT: s_load_dword s1, s[2:3], 0x0
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s6
|
|
; GFX10-DL-NEXT: s_lshr_b32 s2, s0, 16
|
|
; GFX10-DL-NEXT: s_lshr_b32 s3, s1, 16
|
|
; GFX10-DL-NEXT: s_sext_i32_i16 s0, s0
|
|
; GFX10-DL-NEXT: s_sext_i32_i16 s1, s1
|
|
; GFX10-DL-NEXT: v_mad_u32_u24 v0, s3, s2, v0
|
|
; GFX10-DL-NEXT: v_mad_i32_i24 v2, s1, s0, v0
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v1, s5
|
|
; GFX10-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX10-DL-NEXT: s_endpgm
|
|
<2 x i16> addrspace(1)* %src2,
|
|
i32 addrspace(1)* nocapture %dst) {
|
|
entry:
|
|
%vec1 = load <2 x i16>, <2 x i16> addrspace(1)* %src1
|
|
%vec2 = load <2 x i16>, <2 x i16> addrspace(1)* %src2
|
|
|
|
%s1.elt1 = extractelement <2 x i16> %vec1, i64 0
|
|
%conv = sext i16 %s1.elt1 to i32
|
|
%s2.elt1 = extractelement <2 x i16> %vec2, i64 0
|
|
%conv2 = sext i16 %s2.elt1 to i32
|
|
%mul1 = mul nuw i32 %conv2, %conv
|
|
|
|
%s1.elt2 = extractelement <2 x i16> %vec1, i64 1
|
|
%conv3 = zext i16 %s1.elt2 to i32
|
|
%s2.elt2 = extractelement <2 x i16> %vec2, i64 1
|
|
%conv4 = zext i16 %s2.elt2 to i32
|
|
%mul2 = mul nuw i32 %conv4, %conv3
|
|
|
|
%s3 = load i32, i32 addrspace(1)* %dst, align 4
|
|
%add = add i32 %mul2, %s3
|
|
%add6 = add i32 %add, %mul1
|
|
store i32 %add6, i32 addrspace(1)* %dst, align 4
|
|
ret void
|
|
}
|
|
|
|
define amdgpu_kernel void @udot2_alt_AddOperands(<2 x i16> addrspace(1)* %src1,
|
|
; GFX7-LABEL: udot2_alt_AddOperands:
|
|
; GFX7: ; %bb.0: ; %entry
|
|
; GFX7-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x9
|
|
; GFX7-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0xd
|
|
; GFX7-NEXT: s_mov_b32 s8, 0xffff
|
|
; GFX7-NEXT: s_mov_b32 s3, 0xf000
|
|
; GFX7-NEXT: s_mov_b32 s2, -1
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_load_dword s4, s[4:5], 0x0
|
|
; GFX7-NEXT: s_load_dword s5, s[6:7], 0x0
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_lshr_b32 s6, s4, 16
|
|
; GFX7-NEXT: s_lshr_b32 s7, s5, 16
|
|
; GFX7-NEXT: s_and_b32 s4, s4, s8
|
|
; GFX7-NEXT: s_and_b32 s5, s5, s8
|
|
; GFX7-NEXT: s_load_dword s8, s[0:1], 0x0
|
|
; GFX7-NEXT: v_mov_b32_e32 v0, s6
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: v_mov_b32_e32 v1, s8
|
|
; GFX7-NEXT: v_mad_u32_u24 v0, s7, v0, v1
|
|
; GFX7-NEXT: v_mov_b32_e32 v1, s4
|
|
; GFX7-NEXT: v_mad_u32_u24 v0, s5, v1, v0
|
|
; GFX7-NEXT: buffer_store_dword v0, off, s[0:3], 0
|
|
; GFX7-NEXT: s_endpgm
|
|
;
|
|
; GFX8-LABEL: udot2_alt_AddOperands:
|
|
; GFX8: ; %bb.0: ; %entry
|
|
; GFX8-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX8-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX8-NEXT: s_mov_b32 s2, 0xffff
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_load_dword s3, s[4:5], 0x0
|
|
; GFX8-NEXT: s_load_dword s4, s[6:7], 0x0
|
|
; GFX8-NEXT: s_load_dword s5, s[0:1], 0x0
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_and_b32 s6, s3, s2
|
|
; GFX8-NEXT: s_lshr_b32 s3, s3, 16
|
|
; GFX8-NEXT: s_and_b32 s2, s4, s2
|
|
; GFX8-NEXT: s_lshr_b32 s4, s4, 16
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s5
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s3
|
|
; GFX8-NEXT: v_mad_u32_u24 v0, s4, v1, v0
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s6
|
|
; GFX8-NEXT: v_mad_u32_u24 v2, s2, v1, v0
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX8-NEXT: flat_store_dword v[0:1], v2
|
|
; GFX8-NEXT: s_endpgm
|
|
;
|
|
; GFX9-NODL-LABEL: udot2_alt_AddOperands:
|
|
; GFX9-NODL: ; %bb.0: ; %entry
|
|
; GFX9-NODL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-NODL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-NODL-NEXT: s_mov_b32 s2, 0xffff
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_load_dword s3, s[4:5], 0x0
|
|
; GFX9-NODL-NEXT: s_load_dword s4, s[6:7], 0x0
|
|
; GFX9-NODL-NEXT: s_load_dword s5, s[0:1], 0x0
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_and_b32 s6, s3, s2
|
|
; GFX9-NODL-NEXT: s_lshr_b32 s3, s3, 16
|
|
; GFX9-NODL-NEXT: s_and_b32 s2, s4, s2
|
|
; GFX9-NODL-NEXT: s_lshr_b32 s4, s4, 16
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s5
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s3
|
|
; GFX9-NODL-NEXT: v_mad_u32_u24 v0, s4, v1, v0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s6
|
|
; GFX9-NODL-NEXT: v_mad_u32_u24 v2, s2, v1, v0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-NODL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-NODL-NEXT: s_endpgm
|
|
;
|
|
; GFX9-DL-LABEL: udot2_alt_AddOperands:
|
|
; GFX9-DL: ; %bb.0: ; %entry
|
|
; GFX9-DL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-DL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: s_load_dword s2, s[4:5], 0x0
|
|
; GFX9-DL-NEXT: s_load_dword s3, s[0:1], 0x0
|
|
; GFX9-DL-NEXT: s_load_dword s4, s[6:7], 0x0
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s2
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s3
|
|
; GFX9-DL-NEXT: v_dot2_u32_u16 v2, s4, v0, v1
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-DL-NEXT: s_endpgm
|
|
;
|
|
; GFX10-DL-LABEL: udot2_alt_AddOperands:
|
|
; GFX10-DL: ; %bb.0: ; %entry
|
|
; GFX10-DL-NEXT: s_clause 0x1
|
|
; GFX10-DL-NEXT: s_load_dwordx2 s[4:5], s[0:1], 0x34
|
|
; GFX10-DL-NEXT: s_load_dwordx4 s[0:3], s[0:1], 0x24
|
|
; GFX10-DL-NEXT: ; implicit-def: $vcc_hi
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: s_load_dword s6, s[4:5], 0x0
|
|
; GFX10-DL-NEXT: s_load_dword s0, s[0:1], 0x0
|
|
; GFX10-DL-NEXT: s_load_dword s1, s[2:3], 0x0
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s6
|
|
; GFX10-DL-NEXT: v_dot2_u32_u16 v2, s1, s0, v0
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v1, s5
|
|
; GFX10-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX10-DL-NEXT: s_endpgm
|
|
<2 x i16> addrspace(1)* %src2,
|
|
i32 addrspace(1)* nocapture %dst) {
|
|
entry:
|
|
%vec1 = load <2 x i16>, <2 x i16> addrspace(1)* %src1
|
|
%vec2 = load <2 x i16>, <2 x i16> addrspace(1)* %src2
|
|
|
|
%s1.elt1 = extractelement <2 x i16> %vec1, i64 0
|
|
%conv = zext i16 %s1.elt1 to i32
|
|
%s2.elt1 = extractelement <2 x i16> %vec2, i64 0
|
|
%conv2 = zext i16 %s2.elt1 to i32
|
|
%mul1 = mul nuw i32 %conv2, %conv
|
|
|
|
%s1.elt2 = extractelement <2 x i16> %vec1, i64 1
|
|
%conv3 = zext i16 %s1.elt2 to i32
|
|
%s2.elt2 = extractelement <2 x i16> %vec2, i64 1
|
|
%conv4 = zext i16 %s2.elt2 to i32
|
|
%mul2 = mul nuw i32 %conv4, %conv3
|
|
|
|
%s3 = load i32, i32 addrspace(1)* %dst, align 4
|
|
%add = add i32 %s3, %mul2
|
|
%add6 = add i32 %mul1, %add
|
|
store i32 %add6, i32 addrspace(1)* %dst, align 4
|
|
ret void
|
|
}
|
|
|
|
define amdgpu_kernel void @idot2_MixedExt(<2 x i16> addrspace(1)* %src1,
|
|
; GFX7-LABEL: idot2_MixedExt:
|
|
; GFX7: ; %bb.0: ; %entry
|
|
; GFX7-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x9
|
|
; GFX7-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0xd
|
|
; GFX7-NEXT: s_mov_b32 s3, 0xf000
|
|
; GFX7-NEXT: s_mov_b32 s2, -1
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_load_dword s4, s[4:5], 0x0
|
|
; GFX7-NEXT: s_load_dword s5, s[6:7], 0x0
|
|
; GFX7-NEXT: s_load_dword s6, s[0:1], 0x0
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_sext_i32_i16 s7, s4
|
|
; GFX7-NEXT: s_ashr_i32 s4, s4, 16
|
|
; GFX7-NEXT: s_and_b32 s8, s5, 0xffff
|
|
; GFX7-NEXT: s_ashr_i32 s5, s5, 16
|
|
; GFX7-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX7-NEXT: v_mov_b32_e32 v1, s6
|
|
; GFX7-NEXT: v_mad_i32_i24 v0, s5, v0, v1
|
|
; GFX7-NEXT: v_mov_b32_e32 v1, s7
|
|
; GFX7-NEXT: v_mad_i32_i24 v0, s8, v1, v0
|
|
; GFX7-NEXT: buffer_store_dword v0, off, s[0:3], 0
|
|
; GFX7-NEXT: s_endpgm
|
|
;
|
|
; GFX8-LABEL: idot2_MixedExt:
|
|
; GFX8: ; %bb.0: ; %entry
|
|
; GFX8-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX8-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_load_dword s2, s[4:5], 0x0
|
|
; GFX8-NEXT: s_load_dword s3, s[6:7], 0x0
|
|
; GFX8-NEXT: s_load_dword s4, s[0:1], 0x0
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_sext_i32_i16 s5, s2
|
|
; GFX8-NEXT: s_ashr_i32 s2, s2, 16
|
|
; GFX8-NEXT: s_and_b32 s6, s3, 0xffff
|
|
; GFX8-NEXT: s_ashr_i32 s3, s3, 16
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s2
|
|
; GFX8-NEXT: v_mov_b32_e32 v2, s5
|
|
; GFX8-NEXT: v_mad_i32_i24 v0, s3, v1, v0
|
|
; GFX8-NEXT: v_mad_i32_i24 v2, s6, v2, v0
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX8-NEXT: flat_store_dword v[0:1], v2
|
|
; GFX8-NEXT: s_endpgm
|
|
;
|
|
; GFX9-NODL-LABEL: idot2_MixedExt:
|
|
; GFX9-NODL: ; %bb.0: ; %entry
|
|
; GFX9-NODL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-NODL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_load_dword s2, s[4:5], 0x0
|
|
; GFX9-NODL-NEXT: s_load_dword s3, s[6:7], 0x0
|
|
; GFX9-NODL-NEXT: s_load_dword s4, s[0:1], 0x0
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_sext_i32_i16 s5, s2
|
|
; GFX9-NODL-NEXT: s_ashr_i32 s2, s2, 16
|
|
; GFX9-NODL-NEXT: s_and_b32 s6, s3, 0xffff
|
|
; GFX9-NODL-NEXT: s_ashr_i32 s3, s3, 16
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s2
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v2, s5
|
|
; GFX9-NODL-NEXT: v_mad_i32_i24 v0, s3, v1, v0
|
|
; GFX9-NODL-NEXT: v_mad_i32_i24 v2, s6, v2, v0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-NODL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-NODL-NEXT: s_endpgm
|
|
;
|
|
; GFX9-DL-LABEL: idot2_MixedExt:
|
|
; GFX9-DL: ; %bb.0: ; %entry
|
|
; GFX9-DL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-DL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: s_load_dword s2, s[4:5], 0x0
|
|
; GFX9-DL-NEXT: s_load_dword s3, s[6:7], 0x0
|
|
; GFX9-DL-NEXT: s_load_dword s4, s[0:1], 0x0
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: s_sext_i32_i16 s5, s2
|
|
; GFX9-DL-NEXT: s_ashr_i32 s2, s2, 16
|
|
; GFX9-DL-NEXT: s_and_b32 s6, s3, 0xffff
|
|
; GFX9-DL-NEXT: s_ashr_i32 s3, s3, 16
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s2
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v2, s5
|
|
; GFX9-DL-NEXT: v_mad_i32_i24 v0, s3, v1, v0
|
|
; GFX9-DL-NEXT: v_mad_i32_i24 v2, s6, v2, v0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-DL-NEXT: s_endpgm
|
|
;
|
|
; GFX10-DL-LABEL: idot2_MixedExt:
|
|
; GFX10-DL: ; %bb.0: ; %entry
|
|
; GFX10-DL-NEXT: s_clause 0x1
|
|
; GFX10-DL-NEXT: s_load_dwordx2 s[4:5], s[0:1], 0x34
|
|
; GFX10-DL-NEXT: s_load_dwordx4 s[0:3], s[0:1], 0x24
|
|
; GFX10-DL-NEXT: ; implicit-def: $vcc_hi
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: s_load_dword s6, s[4:5], 0x0
|
|
; GFX10-DL-NEXT: s_load_dword s0, s[0:1], 0x0
|
|
; GFX10-DL-NEXT: s_load_dword s1, s[2:3], 0x0
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s6
|
|
; GFX10-DL-NEXT: s_ashr_i32 s2, s0, 16
|
|
; GFX10-DL-NEXT: s_ashr_i32 s3, s1, 16
|
|
; GFX10-DL-NEXT: s_sext_i32_i16 s0, s0
|
|
; GFX10-DL-NEXT: s_and_b32 s1, s1, 0xffff
|
|
; GFX10-DL-NEXT: v_mad_i32_i24 v0, s3, s2, v0
|
|
; GFX10-DL-NEXT: v_mad_i32_i24 v2, s1, s0, v0
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v1, s5
|
|
; GFX10-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX10-DL-NEXT: s_endpgm
|
|
<2 x i16> addrspace(1)* %src2,
|
|
i32 addrspace(1)* nocapture %dst) {
|
|
entry:
|
|
%vec1 = load <2 x i16>, <2 x i16> addrspace(1)* %src1
|
|
%vec2 = load <2 x i16>, <2 x i16> addrspace(1)* %src2
|
|
|
|
%s1.elt1 = extractelement <2 x i16> %vec1, i64 0
|
|
%conv = sext i16 %s1.elt1 to i32
|
|
%s2.elt1 = extractelement <2 x i16> %vec2, i64 0
|
|
%conv2 = zext i16 %s2.elt1 to i32
|
|
%mul1 = mul nuw i32 %conv2, %conv
|
|
|
|
%s1.elt2 = extractelement <2 x i16> %vec1, i64 1
|
|
%conv3 = sext i16 %s1.elt2 to i32
|
|
%s2.elt2 = extractelement <2 x i16> %vec2, i64 1
|
|
%conv4 = sext i16 %s2.elt2 to i32
|
|
%mul2 = mul nuw i32 %conv4, %conv3
|
|
|
|
%s3 = load i32, i32 addrspace(1)* %dst, align 4
|
|
%add = add i32 %mul2, %s3
|
|
%add6 = add i32 %add, %mul1
|
|
store i32 %add6, i32 addrspace(1)* %dst, align 4
|
|
ret void
|
|
}
|
|
|
|
define amdgpu_kernel void @notudot2_SameVec(<2 x i16> addrspace(1)* %src1,
|
|
; GFX7-LABEL: notudot2_SameVec:
|
|
; GFX7: ; %bb.0: ; %entry
|
|
; GFX7-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x9
|
|
; GFX7-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0xd
|
|
; GFX7-NEXT: s_mov_b32 s3, 0xf000
|
|
; GFX7-NEXT: s_mov_b32 s2, -1
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_load_dword s6, s[6:7], 0x0
|
|
; GFX7-NEXT: s_load_dword s7, s[0:1], 0x0
|
|
; GFX7-NEXT: s_load_dword s4, s[4:5], 0x0
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_lshr_b32 s5, s6, 16
|
|
; GFX7-NEXT: v_mov_b32_e32 v0, s7
|
|
; GFX7-NEXT: s_and_b32 s4, s4, 0xffff
|
|
; GFX7-NEXT: v_mad_u32_u24 v0, s5, s5, v0
|
|
; GFX7-NEXT: v_mad_u32_u24 v0, s4, s4, v0
|
|
; GFX7-NEXT: buffer_store_dword v0, off, s[0:3], 0
|
|
; GFX7-NEXT: s_endpgm
|
|
;
|
|
; GFX8-LABEL: notudot2_SameVec:
|
|
; GFX8: ; %bb.0: ; %entry
|
|
; GFX8-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX8-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_load_dword s2, s[6:7], 0x0
|
|
; GFX8-NEXT: s_load_dword s3, s[0:1], 0x0
|
|
; GFX8-NEXT: s_load_dword s4, s[4:5], 0x0
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_lshr_b32 s2, s2, 16
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s3
|
|
; GFX8-NEXT: s_and_b32 s4, s4, 0xffff
|
|
; GFX8-NEXT: v_mad_u32_u24 v0, s2, s2, v0
|
|
; GFX8-NEXT: v_mad_u32_u24 v2, s4, s4, v0
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX8-NEXT: flat_store_dword v[0:1], v2
|
|
; GFX8-NEXT: s_endpgm
|
|
;
|
|
; GFX9-NODL-LABEL: notudot2_SameVec:
|
|
; GFX9-NODL: ; %bb.0: ; %entry
|
|
; GFX9-NODL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-NODL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_load_dword s2, s[6:7], 0x0
|
|
; GFX9-NODL-NEXT: s_load_dword s3, s[0:1], 0x0
|
|
; GFX9-NODL-NEXT: s_load_dword s4, s[4:5], 0x0
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_lshr_b32 s2, s2, 16
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s3
|
|
; GFX9-NODL-NEXT: s_and_b32 s4, s4, 0xffff
|
|
; GFX9-NODL-NEXT: v_mad_u32_u24 v0, s2, s2, v0
|
|
; GFX9-NODL-NEXT: v_mad_u32_u24 v2, s4, s4, v0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-NODL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-NODL-NEXT: s_endpgm
|
|
;
|
|
; GFX9-DL-LABEL: notudot2_SameVec:
|
|
; GFX9-DL: ; %bb.0: ; %entry
|
|
; GFX9-DL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-DL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: s_load_dword s2, s[6:7], 0x0
|
|
; GFX9-DL-NEXT: s_load_dword s3, s[0:1], 0x0
|
|
; GFX9-DL-NEXT: s_load_dword s4, s[4:5], 0x0
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: s_lshr_b32 s2, s2, 16
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s3
|
|
; GFX9-DL-NEXT: s_and_b32 s4, s4, 0xffff
|
|
; GFX9-DL-NEXT: v_mad_u32_u24 v0, s2, s2, v0
|
|
; GFX9-DL-NEXT: v_mad_u32_u24 v2, s4, s4, v0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-DL-NEXT: s_endpgm
|
|
;
|
|
; GFX10-DL-LABEL: notudot2_SameVec:
|
|
; GFX10-DL: ; %bb.0: ; %entry
|
|
; GFX10-DL-NEXT: s_clause 0x1
|
|
; GFX10-DL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX10-DL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX10-DL-NEXT: ; implicit-def: $vcc_hi
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: s_load_dword s2, s[6:7], 0x0
|
|
; GFX10-DL-NEXT: s_load_dword s3, s[0:1], 0x0
|
|
; GFX10-DL-NEXT: s_load_dword s4, s[4:5], 0x0
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: s_lshr_b32 s2, s2, 16
|
|
; GFX10-DL-NEXT: v_mad_u32_u24 v0, s2, s2, s3
|
|
; GFX10-DL-NEXT: s_and_b32 s2, s4, 0xffff
|
|
; GFX10-DL-NEXT: v_mad_u32_u24 v2, s2, s2, v0
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX10-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX10-DL-NEXT: s_endpgm
|
|
<2 x i16> addrspace(1)* %src2,
|
|
i32 addrspace(1)* nocapture %dst) {
|
|
entry:
|
|
%vec1 = load <2 x i16>, <2 x i16> addrspace(1)* %src1
|
|
%vec2 = load <2 x i16>, <2 x i16> addrspace(1)* %src2
|
|
|
|
%s1.elt1 = extractelement <2 x i16> %vec1, i64 0
|
|
%conv = zext i16 %s1.elt1 to i32
|
|
%s2.elt1 = extractelement <2 x i16> %vec1, i64 0
|
|
%conv2 = zext i16 %s2.elt1 to i32
|
|
%mul1 = mul i32 %conv2, %conv
|
|
|
|
%s1.elt2 = extractelement <2 x i16> %vec2, i64 1
|
|
%conv3 = zext i16 %s1.elt2 to i32
|
|
%s2.elt2 = extractelement <2 x i16> %vec2, i64 1
|
|
%conv4 = zext i16 %s2.elt2 to i32
|
|
%mul2 = mul i32 %conv4, %conv3
|
|
|
|
%s3 = load i32, i32 addrspace(1)* %dst, align 4
|
|
%add = add i32 %mul2, %s3
|
|
%add6 = add i32 %add, %mul1
|
|
store i32 %add6, i32 addrspace(1)* %dst, align 4
|
|
ret void
|
|
}
|
|
|
|
define amdgpu_kernel void @udot2_v4i16(<4 x i16> addrspace(1)* %src1,
|
|
; GFX7-LABEL: udot2_v4i16:
|
|
; GFX7: ; %bb.0: ; %entry
|
|
; GFX7-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x9
|
|
; GFX7-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0xd
|
|
; GFX7-NEXT: s_mov_b32 s8, 0xffff
|
|
; GFX7-NEXT: s_mov_b32 s3, 0xf000
|
|
; GFX7-NEXT: s_mov_b32 s2, -1
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_load_dword s4, s[4:5], 0x0
|
|
; GFX7-NEXT: s_load_dword s5, s[6:7], 0x0
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_and_b32 s6, s4, s8
|
|
; GFX7-NEXT: s_and_b32 s7, s5, s8
|
|
; GFX7-NEXT: s_load_dword s8, s[0:1], 0x0
|
|
; GFX7-NEXT: s_lshr_b32 s4, s4, 16
|
|
; GFX7-NEXT: s_lshr_b32 s5, s5, 16
|
|
; GFX7-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: v_mov_b32_e32 v1, s8
|
|
; GFX7-NEXT: v_mad_u32_u24 v0, s5, v0, v1
|
|
; GFX7-NEXT: v_mov_b32_e32 v1, s6
|
|
; GFX7-NEXT: v_mad_u32_u24 v0, s7, v1, v0
|
|
; GFX7-NEXT: buffer_store_dword v0, off, s[0:3], 0
|
|
; GFX7-NEXT: s_endpgm
|
|
;
|
|
; GFX8-LABEL: udot2_v4i16:
|
|
; GFX8: ; %bb.0: ; %entry
|
|
; GFX8-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX8-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX8-NEXT: s_mov_b32 s2, 0xffff
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_load_dword s3, s[4:5], 0x0
|
|
; GFX8-NEXT: s_load_dword s4, s[6:7], 0x0
|
|
; GFX8-NEXT: s_load_dword s5, s[0:1], 0x0
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_and_b32 s6, s3, s2
|
|
; GFX8-NEXT: s_lshr_b32 s3, s3, 16
|
|
; GFX8-NEXT: s_and_b32 s2, s4, s2
|
|
; GFX8-NEXT: s_lshr_b32 s4, s4, 16
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s5
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s3
|
|
; GFX8-NEXT: v_mad_u32_u24 v0, s4, v1, v0
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s6
|
|
; GFX8-NEXT: v_mad_u32_u24 v2, s2, v1, v0
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX8-NEXT: flat_store_dword v[0:1], v2
|
|
; GFX8-NEXT: s_endpgm
|
|
;
|
|
; GFX9-NODL-LABEL: udot2_v4i16:
|
|
; GFX9-NODL: ; %bb.0: ; %entry
|
|
; GFX9-NODL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-NODL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-NODL-NEXT: s_mov_b32 s2, 0xffff
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_load_dword s3, s[4:5], 0x0
|
|
; GFX9-NODL-NEXT: s_load_dword s4, s[6:7], 0x0
|
|
; GFX9-NODL-NEXT: s_load_dword s5, s[0:1], 0x0
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_and_b32 s6, s3, s2
|
|
; GFX9-NODL-NEXT: s_lshr_b32 s3, s3, 16
|
|
; GFX9-NODL-NEXT: s_and_b32 s2, s4, s2
|
|
; GFX9-NODL-NEXT: s_lshr_b32 s4, s4, 16
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s5
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s3
|
|
; GFX9-NODL-NEXT: v_mad_u32_u24 v0, s4, v1, v0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s6
|
|
; GFX9-NODL-NEXT: v_mad_u32_u24 v2, s2, v1, v0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-NODL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-NODL-NEXT: s_endpgm
|
|
;
|
|
; GFX9-DL-LABEL: udot2_v4i16:
|
|
; GFX9-DL: ; %bb.0: ; %entry
|
|
; GFX9-DL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-DL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: s_load_dword s2, s[4:5], 0x0
|
|
; GFX9-DL-NEXT: s_load_dword s3, s[0:1], 0x0
|
|
; GFX9-DL-NEXT: s_load_dword s4, s[6:7], 0x0
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s2
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s3
|
|
; GFX9-DL-NEXT: v_dot2_u32_u16 v2, s4, v0, v1
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-DL-NEXT: s_endpgm
|
|
;
|
|
; GFX10-DL-LABEL: udot2_v4i16:
|
|
; GFX10-DL: ; %bb.0: ; %entry
|
|
; GFX10-DL-NEXT: s_clause 0x1
|
|
; GFX10-DL-NEXT: s_load_dwordx2 s[4:5], s[0:1], 0x34
|
|
; GFX10-DL-NEXT: s_load_dwordx4 s[0:3], s[0:1], 0x24
|
|
; GFX10-DL-NEXT: ; implicit-def: $vcc_hi
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: s_load_dword s6, s[4:5], 0x0
|
|
; GFX10-DL-NEXT: s_load_dword s0, s[0:1], 0x0
|
|
; GFX10-DL-NEXT: s_load_dword s1, s[2:3], 0x0
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s6
|
|
; GFX10-DL-NEXT: v_dot2_u32_u16 v2, s1, s0, v0
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v1, s5
|
|
; GFX10-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX10-DL-NEXT: s_endpgm
|
|
<4 x i16> addrspace(1)* %src2,
|
|
i32 addrspace(1)* nocapture %dst) {
|
|
entry:
|
|
%vec1 = load <4 x i16>, <4 x i16> addrspace(1)* %src1
|
|
%vec2 = load <4 x i16>, <4 x i16> addrspace(1)* %src2
|
|
|
|
%s1.elt1 = extractelement <4 x i16> %vec1, i64 0
|
|
%conv = zext i16 %s1.elt1 to i32
|
|
%s2.elt1 = extractelement <4 x i16> %vec2, i64 0
|
|
%conv2 = zext i16 %s2.elt1 to i32
|
|
%mul1 = mul i32 %conv2, %conv
|
|
|
|
%s1.elt2 = extractelement <4 x i16> %vec1, i64 1
|
|
%conv3 = zext i16 %s1.elt2 to i32
|
|
%s2.elt2 = extractelement <4 x i16> %vec2, i64 1
|
|
%conv4 = zext i16 %s2.elt2 to i32
|
|
%mul2 = mul i32 %conv4, %conv3
|
|
|
|
%s3 = load i32, i32 addrspace(1)* %dst, align 4
|
|
%add = add i32 %mul2, %s3
|
|
%add6 = add i32 %add, %mul1
|
|
store i32 %add6, i32 addrspace(1)* %dst, align 4
|
|
ret void
|
|
}
|
|
|
|
define amdgpu_kernel void @udot2_v4i16_Hi(<4 x i16> addrspace(1)* %src1,
|
|
; GFX7-LABEL: udot2_v4i16_Hi:
|
|
; GFX7: ; %bb.0: ; %entry
|
|
; GFX7-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x9
|
|
; GFX7-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0xd
|
|
; GFX7-NEXT: s_mov_b32 s8, 0xffff
|
|
; GFX7-NEXT: s_mov_b32 s3, 0xf000
|
|
; GFX7-NEXT: s_mov_b32 s2, -1
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_load_dword s4, s[4:5], 0x1
|
|
; GFX7-NEXT: s_load_dword s5, s[6:7], 0x1
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_and_b32 s6, s4, s8
|
|
; GFX7-NEXT: s_and_b32 s7, s5, s8
|
|
; GFX7-NEXT: s_load_dword s8, s[0:1], 0x0
|
|
; GFX7-NEXT: s_lshr_b32 s4, s4, 16
|
|
; GFX7-NEXT: s_lshr_b32 s5, s5, 16
|
|
; GFX7-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: v_mov_b32_e32 v1, s8
|
|
; GFX7-NEXT: v_mad_u32_u24 v0, s5, v0, v1
|
|
; GFX7-NEXT: v_mov_b32_e32 v1, s6
|
|
; GFX7-NEXT: v_mad_u32_u24 v0, s7, v1, v0
|
|
; GFX7-NEXT: buffer_store_dword v0, off, s[0:3], 0
|
|
; GFX7-NEXT: s_endpgm
|
|
;
|
|
; GFX8-LABEL: udot2_v4i16_Hi:
|
|
; GFX8: ; %bb.0: ; %entry
|
|
; GFX8-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX8-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX8-NEXT: s_mov_b32 s2, 0xffff
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_load_dword s3, s[4:5], 0x4
|
|
; GFX8-NEXT: s_load_dword s4, s[6:7], 0x4
|
|
; GFX8-NEXT: s_load_dword s5, s[0:1], 0x0
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_and_b32 s6, s3, s2
|
|
; GFX8-NEXT: s_lshr_b32 s3, s3, 16
|
|
; GFX8-NEXT: s_and_b32 s2, s4, s2
|
|
; GFX8-NEXT: s_lshr_b32 s4, s4, 16
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s5
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s3
|
|
; GFX8-NEXT: v_mad_u32_u24 v0, s4, v1, v0
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s6
|
|
; GFX8-NEXT: v_mad_u32_u24 v2, s2, v1, v0
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX8-NEXT: flat_store_dword v[0:1], v2
|
|
; GFX8-NEXT: s_endpgm
|
|
;
|
|
; GFX9-NODL-LABEL: udot2_v4i16_Hi:
|
|
; GFX9-NODL: ; %bb.0: ; %entry
|
|
; GFX9-NODL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-NODL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-NODL-NEXT: s_mov_b32 s2, 0xffff
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_load_dword s3, s[4:5], 0x4
|
|
; GFX9-NODL-NEXT: s_load_dword s4, s[6:7], 0x4
|
|
; GFX9-NODL-NEXT: s_load_dword s5, s[0:1], 0x0
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_and_b32 s6, s3, s2
|
|
; GFX9-NODL-NEXT: s_lshr_b32 s3, s3, 16
|
|
; GFX9-NODL-NEXT: s_and_b32 s2, s4, s2
|
|
; GFX9-NODL-NEXT: s_lshr_b32 s4, s4, 16
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s5
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s3
|
|
; GFX9-NODL-NEXT: v_mad_u32_u24 v0, s4, v1, v0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s6
|
|
; GFX9-NODL-NEXT: v_mad_u32_u24 v2, s2, v1, v0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-NODL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-NODL-NEXT: s_endpgm
|
|
;
|
|
; GFX9-DL-LABEL: udot2_v4i16_Hi:
|
|
; GFX9-DL: ; %bb.0: ; %entry
|
|
; GFX9-DL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-DL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: s_load_dword s2, s[4:5], 0x4
|
|
; GFX9-DL-NEXT: s_load_dword s3, s[0:1], 0x0
|
|
; GFX9-DL-NEXT: s_load_dword s4, s[6:7], 0x4
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s2
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s3
|
|
; GFX9-DL-NEXT: v_dot2_u32_u16 v2, s4, v0, v1
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-DL-NEXT: s_endpgm
|
|
;
|
|
; GFX10-DL-LABEL: udot2_v4i16_Hi:
|
|
; GFX10-DL: ; %bb.0: ; %entry
|
|
; GFX10-DL-NEXT: s_clause 0x1
|
|
; GFX10-DL-NEXT: s_load_dwordx2 s[4:5], s[0:1], 0x34
|
|
; GFX10-DL-NEXT: s_load_dwordx4 s[0:3], s[0:1], 0x24
|
|
; GFX10-DL-NEXT: ; implicit-def: $vcc_hi
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: s_load_dword s6, s[4:5], 0x0
|
|
; GFX10-DL-NEXT: s_load_dword s0, s[0:1], 0x4
|
|
; GFX10-DL-NEXT: s_load_dword s1, s[2:3], 0x4
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s6
|
|
; GFX10-DL-NEXT: v_dot2_u32_u16 v2, s1, s0, v0
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v1, s5
|
|
; GFX10-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX10-DL-NEXT: s_endpgm
|
|
<4 x i16> addrspace(1)* %src2,
|
|
i32 addrspace(1)* nocapture %dst) {
|
|
entry:
|
|
%vec1 = load <4 x i16>, <4 x i16> addrspace(1)* %src1
|
|
%vec2 = load <4 x i16>, <4 x i16> addrspace(1)* %src2
|
|
|
|
%s1.elt1 = extractelement <4 x i16> %vec1, i64 2
|
|
%conv = zext i16 %s1.elt1 to i32
|
|
%s2.elt1 = extractelement <4 x i16> %vec2, i64 2
|
|
%conv2 = zext i16 %s2.elt1 to i32
|
|
%mul1 = mul i32 %conv2, %conv
|
|
|
|
%s1.elt2 = extractelement <4 x i16> %vec1, i64 3
|
|
%conv3 = zext i16 %s1.elt2 to i32
|
|
%s2.elt2 = extractelement <4 x i16> %vec2, i64 3
|
|
%conv4 = zext i16 %s2.elt2 to i32
|
|
%mul2 = mul i32 %conv4, %conv3
|
|
|
|
%s3 = load i32, i32 addrspace(1)* %dst, align 4
|
|
%add = add i32 %mul2, %s3
|
|
%add6 = add i32 %add, %mul1
|
|
store i32 %add6, i32 addrspace(1)* %dst, align 4
|
|
ret void
|
|
}
|
|
|
|
define amdgpu_kernel void @notudot2_v4i16_Even(<4 x i16> addrspace(1)* %src1,
|
|
; GFX7-LABEL: notudot2_v4i16_Even:
|
|
; GFX7: ; %bb.0: ; %entry
|
|
; GFX7-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x9
|
|
; GFX7-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0xd
|
|
; GFX7-NEXT: s_mov_b32 s8, 0xffff
|
|
; GFX7-NEXT: s_mov_b32 s3, 0xf000
|
|
; GFX7-NEXT: s_mov_b32 s2, -1
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_load_dwordx2 s[4:5], s[4:5], 0x0
|
|
; GFX7-NEXT: s_load_dwordx2 s[6:7], s[6:7], 0x0
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_and_b32 s5, s5, s8
|
|
; GFX7-NEXT: s_and_b32 s4, s4, s8
|
|
; GFX7-NEXT: s_and_b32 s6, s6, s8
|
|
; GFX7-NEXT: s_and_b32 s7, s7, s8
|
|
; GFX7-NEXT: s_load_dword s8, s[0:1], 0x0
|
|
; GFX7-NEXT: v_mov_b32_e32 v0, s5
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: v_mov_b32_e32 v1, s8
|
|
; GFX7-NEXT: v_mad_u32_u24 v0, s7, v0, v1
|
|
; GFX7-NEXT: v_mov_b32_e32 v1, s4
|
|
; GFX7-NEXT: v_mad_u32_u24 v0, s6, v1, v0
|
|
; GFX7-NEXT: buffer_store_dword v0, off, s[0:3], 0
|
|
; GFX7-NEXT: s_endpgm
|
|
;
|
|
; GFX8-LABEL: notudot2_v4i16_Even:
|
|
; GFX8: ; %bb.0: ; %entry
|
|
; GFX8-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX8-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX8-NEXT: s_mov_b32 s8, 0xffff
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_load_dwordx2 s[2:3], s[4:5], 0x0
|
|
; GFX8-NEXT: s_load_dwordx2 s[4:5], s[6:7], 0x0
|
|
; GFX8-NEXT: s_load_dword s6, s[0:1], 0x0
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_and_b32 s3, s3, s8
|
|
; GFX8-NEXT: s_and_b32 s2, s2, s8
|
|
; GFX8-NEXT: s_and_b32 s5, s5, s8
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s6
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s3
|
|
; GFX8-NEXT: v_mad_u32_u24 v0, s5, v1, v0
|
|
; GFX8-NEXT: s_and_b32 s4, s4, s8
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s2
|
|
; GFX8-NEXT: v_mad_u32_u24 v2, s4, v1, v0
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX8-NEXT: flat_store_dword v[0:1], v2
|
|
; GFX8-NEXT: s_endpgm
|
|
;
|
|
; GFX9-NODL-LABEL: notudot2_v4i16_Even:
|
|
; GFX9-NODL: ; %bb.0: ; %entry
|
|
; GFX9-NODL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-NODL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-NODL-NEXT: s_mov_b32 s8, 0xffff
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_load_dwordx2 s[2:3], s[4:5], 0x0
|
|
; GFX9-NODL-NEXT: s_load_dwordx2 s[4:5], s[6:7], 0x0
|
|
; GFX9-NODL-NEXT: s_load_dword s6, s[0:1], 0x0
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_and_b32 s3, s3, s8
|
|
; GFX9-NODL-NEXT: s_and_b32 s2, s2, s8
|
|
; GFX9-NODL-NEXT: s_and_b32 s5, s5, s8
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s6
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s3
|
|
; GFX9-NODL-NEXT: v_mad_u32_u24 v0, s5, v1, v0
|
|
; GFX9-NODL-NEXT: s_and_b32 s4, s4, s8
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s2
|
|
; GFX9-NODL-NEXT: v_mad_u32_u24 v2, s4, v1, v0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-NODL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-NODL-NEXT: s_endpgm
|
|
;
|
|
; GFX9-DL-LABEL: notudot2_v4i16_Even:
|
|
; GFX9-DL: ; %bb.0: ; %entry
|
|
; GFX9-DL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-DL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-DL-NEXT: s_mov_b32 s8, 0xffff
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: s_load_dwordx2 s[2:3], s[4:5], 0x0
|
|
; GFX9-DL-NEXT: s_load_dwordx2 s[4:5], s[6:7], 0x0
|
|
; GFX9-DL-NEXT: s_load_dword s6, s[0:1], 0x0
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: s_and_b32 s3, s3, s8
|
|
; GFX9-DL-NEXT: s_and_b32 s2, s2, s8
|
|
; GFX9-DL-NEXT: s_and_b32 s5, s5, s8
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s6
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s3
|
|
; GFX9-DL-NEXT: v_mad_u32_u24 v0, s5, v1, v0
|
|
; GFX9-DL-NEXT: s_and_b32 s4, s4, s8
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s2
|
|
; GFX9-DL-NEXT: v_mad_u32_u24 v2, s4, v1, v0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-DL-NEXT: s_endpgm
|
|
;
|
|
; GFX10-DL-LABEL: notudot2_v4i16_Even:
|
|
; GFX10-DL: ; %bb.0: ; %entry
|
|
; GFX10-DL-NEXT: s_clause 0x1
|
|
; GFX10-DL-NEXT: s_load_dwordx2 s[4:5], s[0:1], 0x34
|
|
; GFX10-DL-NEXT: s_load_dwordx4 s[0:3], s[0:1], 0x24
|
|
; GFX10-DL-NEXT: s_mov_b32 s7, 0xffff
|
|
; GFX10-DL-NEXT: ; implicit-def: $vcc_hi
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: s_load_dword s6, s[4:5], 0x0
|
|
; GFX10-DL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x0
|
|
; GFX10-DL-NEXT: s_load_dwordx2 s[2:3], s[2:3], 0x0
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s6
|
|
; GFX10-DL-NEXT: s_and_b32 s1, s1, s7
|
|
; GFX10-DL-NEXT: s_and_b32 s3, s3, s7
|
|
; GFX10-DL-NEXT: s_and_b32 s0, s0, s7
|
|
; GFX10-DL-NEXT: v_mad_u32_u24 v0, s3, s1, v0
|
|
; GFX10-DL-NEXT: s_and_b32 s1, s2, s7
|
|
; GFX10-DL-NEXT: v_mad_u32_u24 v2, s1, s0, v0
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v1, s5
|
|
; GFX10-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX10-DL-NEXT: s_endpgm
|
|
<4 x i16> addrspace(1)* %src2,
|
|
i32 addrspace(1)* nocapture %dst) {
|
|
entry:
|
|
%vec1 = load <4 x i16>, <4 x i16> addrspace(1)* %src1
|
|
%vec2 = load <4 x i16>, <4 x i16> addrspace(1)* %src2
|
|
|
|
%s1.elt1 = extractelement <4 x i16> %vec1, i64 0
|
|
%conv = zext i16 %s1.elt1 to i32
|
|
%s2.elt1 = extractelement <4 x i16> %vec2, i64 0
|
|
%conv2 = zext i16 %s2.elt1 to i32
|
|
%mul1 = mul i32 %conv2, %conv
|
|
|
|
%s1.elt2 = extractelement <4 x i16> %vec1, i64 2
|
|
%conv3 = zext i16 %s1.elt2 to i32
|
|
%s2.elt2 = extractelement <4 x i16> %vec2, i64 2
|
|
%conv4 = zext i16 %s2.elt2 to i32
|
|
%mul2 = mul i32 %conv4, %conv3
|
|
|
|
%s3 = load i32, i32 addrspace(1)* %dst, align 4
|
|
%add = add i32 %mul2, %s3
|
|
%add6 = add i32 %add, %mul1
|
|
store i32 %add6, i32 addrspace(1)* %dst, align 4
|
|
ret void
|
|
}
|
|
|
|
define amdgpu_kernel void @notudot2_v4i16_Middle(<4 x i16> addrspace(1)* %src1,
|
|
; GFX7-LABEL: notudot2_v4i16_Middle:
|
|
; GFX7: ; %bb.0: ; %entry
|
|
; GFX7-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x9
|
|
; GFX7-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0xd
|
|
; GFX7-NEXT: s_mov_b32 s8, 0xffff
|
|
; GFX7-NEXT: s_mov_b32 s3, 0xf000
|
|
; GFX7-NEXT: s_mov_b32 s2, -1
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_load_dwordx2 s[4:5], s[4:5], 0x0
|
|
; GFX7-NEXT: s_load_dwordx2 s[6:7], s[6:7], 0x0
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_and_b32 s5, s5, s8
|
|
; GFX7-NEXT: s_and_b32 s7, s7, s8
|
|
; GFX7-NEXT: s_load_dword s8, s[0:1], 0x0
|
|
; GFX7-NEXT: s_lshr_b32 s4, s4, 16
|
|
; GFX7-NEXT: v_mov_b32_e32 v0, s5
|
|
; GFX7-NEXT: s_lshr_b32 s6, s6, 16
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: v_mov_b32_e32 v1, s8
|
|
; GFX7-NEXT: v_mad_u32_u24 v0, s7, v0, v1
|
|
; GFX7-NEXT: v_mov_b32_e32 v1, s4
|
|
; GFX7-NEXT: v_mad_u32_u24 v0, s6, v1, v0
|
|
; GFX7-NEXT: buffer_store_dword v0, off, s[0:3], 0
|
|
; GFX7-NEXT: s_endpgm
|
|
;
|
|
; GFX8-LABEL: notudot2_v4i16_Middle:
|
|
; GFX8: ; %bb.0: ; %entry
|
|
; GFX8-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX8-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX8-NEXT: s_mov_b32 s8, 0xffff
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_load_dwordx2 s[2:3], s[4:5], 0x0
|
|
; GFX8-NEXT: s_load_dwordx2 s[4:5], s[6:7], 0x0
|
|
; GFX8-NEXT: s_load_dword s6, s[0:1], 0x0
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_and_b32 s3, s3, s8
|
|
; GFX8-NEXT: s_lshr_b32 s2, s2, 16
|
|
; GFX8-NEXT: s_and_b32 s5, s5, s8
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s6
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s3
|
|
; GFX8-NEXT: v_mad_u32_u24 v0, s5, v1, v0
|
|
; GFX8-NEXT: s_lshr_b32 s4, s4, 16
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s2
|
|
; GFX8-NEXT: v_mad_u32_u24 v2, s4, v1, v0
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX8-NEXT: flat_store_dword v[0:1], v2
|
|
; GFX8-NEXT: s_endpgm
|
|
;
|
|
; GFX9-NODL-LABEL: notudot2_v4i16_Middle:
|
|
; GFX9-NODL: ; %bb.0: ; %entry
|
|
; GFX9-NODL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-NODL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-NODL-NEXT: s_mov_b32 s8, 0xffff
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_load_dwordx2 s[2:3], s[4:5], 0x0
|
|
; GFX9-NODL-NEXT: s_load_dwordx2 s[4:5], s[6:7], 0x0
|
|
; GFX9-NODL-NEXT: s_load_dword s6, s[0:1], 0x0
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_and_b32 s3, s3, s8
|
|
; GFX9-NODL-NEXT: s_lshr_b32 s2, s2, 16
|
|
; GFX9-NODL-NEXT: s_and_b32 s5, s5, s8
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s6
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s3
|
|
; GFX9-NODL-NEXT: v_mad_u32_u24 v0, s5, v1, v0
|
|
; GFX9-NODL-NEXT: s_lshr_b32 s4, s4, 16
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s2
|
|
; GFX9-NODL-NEXT: v_mad_u32_u24 v2, s4, v1, v0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-NODL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-NODL-NEXT: s_endpgm
|
|
;
|
|
; GFX9-DL-LABEL: notudot2_v4i16_Middle:
|
|
; GFX9-DL: ; %bb.0: ; %entry
|
|
; GFX9-DL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-DL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-DL-NEXT: s_mov_b32 s8, 0xffff
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: s_load_dwordx2 s[2:3], s[4:5], 0x0
|
|
; GFX9-DL-NEXT: s_load_dwordx2 s[4:5], s[6:7], 0x0
|
|
; GFX9-DL-NEXT: s_load_dword s6, s[0:1], 0x0
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: s_and_b32 s3, s3, s8
|
|
; GFX9-DL-NEXT: s_lshr_b32 s2, s2, 16
|
|
; GFX9-DL-NEXT: s_and_b32 s5, s5, s8
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s6
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s3
|
|
; GFX9-DL-NEXT: v_mad_u32_u24 v0, s5, v1, v0
|
|
; GFX9-DL-NEXT: s_lshr_b32 s4, s4, 16
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s2
|
|
; GFX9-DL-NEXT: v_mad_u32_u24 v2, s4, v1, v0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-DL-NEXT: s_endpgm
|
|
;
|
|
; GFX10-DL-LABEL: notudot2_v4i16_Middle:
|
|
; GFX10-DL: ; %bb.0: ; %entry
|
|
; GFX10-DL-NEXT: s_clause 0x1
|
|
; GFX10-DL-NEXT: s_load_dwordx2 s[4:5], s[0:1], 0x34
|
|
; GFX10-DL-NEXT: s_load_dwordx4 s[0:3], s[0:1], 0x24
|
|
; GFX10-DL-NEXT: s_mov_b32 s7, 0xffff
|
|
; GFX10-DL-NEXT: ; implicit-def: $vcc_hi
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: s_load_dword s6, s[4:5], 0x0
|
|
; GFX10-DL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x0
|
|
; GFX10-DL-NEXT: s_load_dwordx2 s[2:3], s[2:3], 0x0
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s6
|
|
; GFX10-DL-NEXT: s_and_b32 s1, s1, s7
|
|
; GFX10-DL-NEXT: s_and_b32 s3, s3, s7
|
|
; GFX10-DL-NEXT: s_lshr_b32 s0, s0, 16
|
|
; GFX10-DL-NEXT: v_mad_u32_u24 v0, s3, s1, v0
|
|
; GFX10-DL-NEXT: s_lshr_b32 s1, s2, 16
|
|
; GFX10-DL-NEXT: v_mad_u32_u24 v2, s1, s0, v0
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v1, s5
|
|
; GFX10-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX10-DL-NEXT: s_endpgm
|
|
<4 x i16> addrspace(1)* %src2,
|
|
i32 addrspace(1)* nocapture %dst) {
|
|
entry:
|
|
%vec1 = load <4 x i16>, <4 x i16> addrspace(1)* %src1
|
|
%vec2 = load <4 x i16>, <4 x i16> addrspace(1)* %src2
|
|
|
|
%s1.elt1 = extractelement <4 x i16> %vec1, i64 1
|
|
%conv = zext i16 %s1.elt1 to i32
|
|
%s2.elt1 = extractelement <4 x i16> %vec2, i64 1
|
|
%conv2 = zext i16 %s2.elt1 to i32
|
|
%mul1 = mul i32 %conv2, %conv
|
|
|
|
%s1.elt2 = extractelement <4 x i16> %vec1, i64 2
|
|
%conv3 = zext i16 %s1.elt2 to i32
|
|
%s2.elt2 = extractelement <4 x i16> %vec2, i64 2
|
|
%conv4 = zext i16 %s2.elt2 to i32
|
|
%mul2 = mul i32 %conv4, %conv3
|
|
|
|
%s3 = load i32, i32 addrspace(1)* %dst, align 4
|
|
%add = add i32 %mul2, %s3
|
|
%add6 = add i32 %add, %mul1
|
|
store i32 %add6, i32 addrspace(1)* %dst, align 4
|
|
ret void
|
|
}
|
|
|
|
define amdgpu_kernel void @notudot2_DiffIndex(<2 x i16> addrspace(1)* %src1,
|
|
; GFX7-LABEL: notudot2_DiffIndex:
|
|
; GFX7: ; %bb.0: ; %entry
|
|
; GFX7-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x9
|
|
; GFX7-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0xd
|
|
; GFX7-NEXT: s_mov_b32 s8, 0xffff
|
|
; GFX7-NEXT: s_mov_b32 s3, 0xf000
|
|
; GFX7-NEXT: s_mov_b32 s2, -1
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_load_dword s4, s[4:5], 0x0
|
|
; GFX7-NEXT: s_load_dword s5, s[6:7], 0x0
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_lshr_b32 s6, s4, 16
|
|
; GFX7-NEXT: s_lshr_b32 s7, s5, 16
|
|
; GFX7-NEXT: s_and_b32 s4, s4, s8
|
|
; GFX7-NEXT: s_and_b32 s5, s5, s8
|
|
; GFX7-NEXT: s_load_dword s8, s[0:1], 0x0
|
|
; GFX7-NEXT: v_mov_b32_e32 v0, s6
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: v_mov_b32_e32 v1, s8
|
|
; GFX7-NEXT: v_mad_u32_u24 v0, s5, v0, v1
|
|
; GFX7-NEXT: v_mov_b32_e32 v1, s4
|
|
; GFX7-NEXT: v_mad_u32_u24 v0, s7, v1, v0
|
|
; GFX7-NEXT: buffer_store_dword v0, off, s[0:3], 0
|
|
; GFX7-NEXT: s_endpgm
|
|
;
|
|
; GFX8-LABEL: notudot2_DiffIndex:
|
|
; GFX8: ; %bb.0: ; %entry
|
|
; GFX8-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX8-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX8-NEXT: s_mov_b32 s2, 0xffff
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_load_dword s3, s[4:5], 0x0
|
|
; GFX8-NEXT: s_load_dword s4, s[6:7], 0x0
|
|
; GFX8-NEXT: s_load_dword s5, s[0:1], 0x0
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_and_b32 s6, s3, s2
|
|
; GFX8-NEXT: s_lshr_b32 s3, s3, 16
|
|
; GFX8-NEXT: s_and_b32 s2, s4, s2
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s5
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s3
|
|
; GFX8-NEXT: v_mad_u32_u24 v0, s2, v1, v0
|
|
; GFX8-NEXT: s_lshr_b32 s7, s4, 16
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s6
|
|
; GFX8-NEXT: v_mad_u32_u24 v2, s7, v1, v0
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX8-NEXT: flat_store_dword v[0:1], v2
|
|
; GFX8-NEXT: s_endpgm
|
|
;
|
|
; GFX9-NODL-LABEL: notudot2_DiffIndex:
|
|
; GFX9-NODL: ; %bb.0: ; %entry
|
|
; GFX9-NODL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-NODL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-NODL-NEXT: s_mov_b32 s2, 0xffff
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_load_dword s3, s[4:5], 0x0
|
|
; GFX9-NODL-NEXT: s_load_dword s4, s[6:7], 0x0
|
|
; GFX9-NODL-NEXT: s_load_dword s5, s[0:1], 0x0
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_and_b32 s6, s3, s2
|
|
; GFX9-NODL-NEXT: s_lshr_b32 s3, s3, 16
|
|
; GFX9-NODL-NEXT: s_and_b32 s2, s4, s2
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s5
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s3
|
|
; GFX9-NODL-NEXT: v_mad_u32_u24 v0, s2, v1, v0
|
|
; GFX9-NODL-NEXT: s_lshr_b32 s7, s4, 16
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s6
|
|
; GFX9-NODL-NEXT: v_mad_u32_u24 v2, s7, v1, v0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-NODL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-NODL-NEXT: s_endpgm
|
|
;
|
|
; GFX9-DL-LABEL: notudot2_DiffIndex:
|
|
; GFX9-DL: ; %bb.0: ; %entry
|
|
; GFX9-DL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-DL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-DL-NEXT: s_mov_b32 s2, 0xffff
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: s_load_dword s3, s[4:5], 0x0
|
|
; GFX9-DL-NEXT: s_load_dword s4, s[6:7], 0x0
|
|
; GFX9-DL-NEXT: s_load_dword s5, s[0:1], 0x0
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: s_and_b32 s6, s3, s2
|
|
; GFX9-DL-NEXT: s_lshr_b32 s3, s3, 16
|
|
; GFX9-DL-NEXT: s_and_b32 s2, s4, s2
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s5
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s3
|
|
; GFX9-DL-NEXT: v_mad_u32_u24 v0, s2, v1, v0
|
|
; GFX9-DL-NEXT: s_lshr_b32 s7, s4, 16
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s6
|
|
; GFX9-DL-NEXT: v_mad_u32_u24 v2, s7, v1, v0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-DL-NEXT: s_endpgm
|
|
;
|
|
; GFX10-DL-LABEL: notudot2_DiffIndex:
|
|
; GFX10-DL: ; %bb.0: ; %entry
|
|
; GFX10-DL-NEXT: s_clause 0x1
|
|
; GFX10-DL-NEXT: s_load_dwordx2 s[4:5], s[0:1], 0x34
|
|
; GFX10-DL-NEXT: s_load_dwordx4 s[0:3], s[0:1], 0x24
|
|
; GFX10-DL-NEXT: ; implicit-def: $vcc_hi
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: s_load_dword s6, s[4:5], 0x0
|
|
; GFX10-DL-NEXT: s_load_dword s0, s[0:1], 0x0
|
|
; GFX10-DL-NEXT: s_load_dword s1, s[2:3], 0x0
|
|
; GFX10-DL-NEXT: s_mov_b32 s2, 0xffff
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s6
|
|
; GFX10-DL-NEXT: s_lshr_b32 s3, s0, 16
|
|
; GFX10-DL-NEXT: s_and_b32 s6, s1, s2
|
|
; GFX10-DL-NEXT: s_and_b32 s0, s0, s2
|
|
; GFX10-DL-NEXT: s_lshr_b32 s1, s1, 16
|
|
; GFX10-DL-NEXT: v_mad_u32_u24 v0, s6, s3, v0
|
|
; GFX10-DL-NEXT: v_mad_u32_u24 v2, s1, s0, v0
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v1, s5
|
|
; GFX10-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX10-DL-NEXT: s_endpgm
|
|
<2 x i16> addrspace(1)* %src2,
|
|
i32 addrspace(1)* nocapture %dst) {
|
|
entry:
|
|
%vec1 = load <2 x i16>, <2 x i16> addrspace(1)* %src1
|
|
%vec2 = load <2 x i16>, <2 x i16> addrspace(1)* %src2
|
|
|
|
%s1.elt1 = extractelement <2 x i16> %vec1, i64 0
|
|
%conv = zext i16 %s1.elt1 to i32
|
|
%s2.elt1 = extractelement <2 x i16> %vec2, i64 1
|
|
%conv2 = zext i16 %s2.elt1 to i32
|
|
%mul1 = mul i32 %conv2, %conv
|
|
|
|
%s1.elt2 = extractelement <2 x i16> %vec1, i64 1
|
|
%conv3 = zext i16 %s1.elt2 to i32
|
|
%s2.elt2 = extractelement <2 x i16> %vec2, i64 0
|
|
%conv4 = zext i16 %s2.elt2 to i32
|
|
%mul2 = mul i32 %conv4, %conv3
|
|
|
|
%s3 = load i32, i32 addrspace(1)* %dst, align 4
|
|
%add = add i32 %mul2, %s3
|
|
%add6 = add i32 %add, %mul1
|
|
store i32 %add6, i32 addrspace(1)* %dst, align 4
|
|
ret void
|
|
}
|
|
|
|
define amdgpu_kernel void @udot2_MultipleUses_add1(<2 x i16> addrspace(1)* %src1,
|
|
; GFX7-LABEL: udot2_MultipleUses_add1:
|
|
; GFX7: ; %bb.0: ; %entry
|
|
; GFX7-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x9
|
|
; GFX7-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0xd
|
|
; GFX7-NEXT: s_mov_b32 s8, 0xffff
|
|
; GFX7-NEXT: s_mov_b32 s3, 0xf000
|
|
; GFX7-NEXT: s_mov_b32 s2, -1
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_load_dword s4, s[4:5], 0x0
|
|
; GFX7-NEXT: s_load_dword s5, s[6:7], 0x0
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_lshr_b32 s6, s4, 16
|
|
; GFX7-NEXT: s_lshr_b32 s7, s5, 16
|
|
; GFX7-NEXT: s_and_b32 s4, s4, s8
|
|
; GFX7-NEXT: s_and_b32 s5, s5, s8
|
|
; GFX7-NEXT: s_load_dword s8, s[0:1], 0x0
|
|
; GFX7-NEXT: v_mov_b32_e32 v0, s6
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: v_mov_b32_e32 v1, s8
|
|
; GFX7-NEXT: v_mad_u32_u24 v0, s7, v0, v1
|
|
; GFX7-NEXT: v_mov_b32_e32 v1, s4
|
|
; GFX7-NEXT: v_mad_u32_u24 v1, s5, v1, v0
|
|
; GFX7-NEXT: v_add_i32_e32 v0, vcc, v0, v1
|
|
; GFX7-NEXT: buffer_store_dword v0, off, s[0:3], 0
|
|
; GFX7-NEXT: s_endpgm
|
|
;
|
|
; GFX8-LABEL: udot2_MultipleUses_add1:
|
|
; GFX8: ; %bb.0: ; %entry
|
|
; GFX8-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX8-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX8-NEXT: s_mov_b32 s2, 0xffff
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_load_dword s3, s[4:5], 0x0
|
|
; GFX8-NEXT: s_load_dword s4, s[6:7], 0x0
|
|
; GFX8-NEXT: s_load_dword s5, s[0:1], 0x0
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_and_b32 s6, s3, s2
|
|
; GFX8-NEXT: s_lshr_b32 s3, s3, 16
|
|
; GFX8-NEXT: s_and_b32 s2, s4, s2
|
|
; GFX8-NEXT: s_lshr_b32 s4, s4, 16
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s5
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s3
|
|
; GFX8-NEXT: v_mad_u32_u24 v0, s4, v1, v0
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s6
|
|
; GFX8-NEXT: v_mad_u32_u24 v1, s2, v1, v0
|
|
; GFX8-NEXT: v_add_u32_e32 v2, vcc, v0, v1
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX8-NEXT: flat_store_dword v[0:1], v2
|
|
; GFX8-NEXT: s_endpgm
|
|
;
|
|
; GFX9-NODL-LABEL: udot2_MultipleUses_add1:
|
|
; GFX9-NODL: ; %bb.0: ; %entry
|
|
; GFX9-NODL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-NODL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-NODL-NEXT: s_mov_b32 s2, 0xffff
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_load_dword s3, s[4:5], 0x0
|
|
; GFX9-NODL-NEXT: s_load_dword s4, s[6:7], 0x0
|
|
; GFX9-NODL-NEXT: s_load_dword s5, s[0:1], 0x0
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_and_b32 s6, s3, s2
|
|
; GFX9-NODL-NEXT: s_lshr_b32 s3, s3, 16
|
|
; GFX9-NODL-NEXT: s_and_b32 s2, s4, s2
|
|
; GFX9-NODL-NEXT: s_lshr_b32 s4, s4, 16
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s5
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s3
|
|
; GFX9-NODL-NEXT: v_mad_u32_u24 v0, s4, v1, v0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s6
|
|
; GFX9-NODL-NEXT: v_mad_u32_u24 v1, s2, v1, v0
|
|
; GFX9-NODL-NEXT: v_add_u32_e32 v2, v1, v0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-NODL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-NODL-NEXT: s_endpgm
|
|
;
|
|
; GFX9-DL-LABEL: udot2_MultipleUses_add1:
|
|
; GFX9-DL: ; %bb.0: ; %entry
|
|
; GFX9-DL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-DL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-DL-NEXT: s_mov_b32 s2, 0xffff
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: s_load_dword s3, s[4:5], 0x0
|
|
; GFX9-DL-NEXT: s_load_dword s4, s[6:7], 0x0
|
|
; GFX9-DL-NEXT: s_load_dword s5, s[0:1], 0x0
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: s_and_b32 s6, s3, s2
|
|
; GFX9-DL-NEXT: s_lshr_b32 s3, s3, 16
|
|
; GFX9-DL-NEXT: s_and_b32 s2, s4, s2
|
|
; GFX9-DL-NEXT: s_lshr_b32 s4, s4, 16
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s5
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s3
|
|
; GFX9-DL-NEXT: v_mad_u32_u24 v0, s4, v1, v0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s6
|
|
; GFX9-DL-NEXT: v_mad_u32_u24 v1, s2, v1, v0
|
|
; GFX9-DL-NEXT: v_add_u32_e32 v2, v1, v0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-DL-NEXT: s_endpgm
|
|
;
|
|
; GFX10-DL-LABEL: udot2_MultipleUses_add1:
|
|
; GFX10-DL: ; %bb.0: ; %entry
|
|
; GFX10-DL-NEXT: s_clause 0x1
|
|
; GFX10-DL-NEXT: s_load_dwordx2 s[4:5], s[0:1], 0x34
|
|
; GFX10-DL-NEXT: s_load_dwordx4 s[0:3], s[0:1], 0x24
|
|
; GFX10-DL-NEXT: ; implicit-def: $vcc_hi
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: s_load_dword s6, s[4:5], 0x0
|
|
; GFX10-DL-NEXT: s_load_dword s0, s[0:1], 0x0
|
|
; GFX10-DL-NEXT: s_load_dword s1, s[2:3], 0x0
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s6
|
|
; GFX10-DL-NEXT: s_lshr_b32 s2, s0, 16
|
|
; GFX10-DL-NEXT: s_lshr_b32 s3, s1, 16
|
|
; GFX10-DL-NEXT: s_mov_b32 s6, 0xffff
|
|
; GFX10-DL-NEXT: v_mad_u32_u24 v0, s3, s2, v0
|
|
; GFX10-DL-NEXT: s_and_b32 s0, s0, s6
|
|
; GFX10-DL-NEXT: s_and_b32 s1, s1, s6
|
|
; GFX10-DL-NEXT: v_mad_u32_u24 v1, s1, s0, v0
|
|
; GFX10-DL-NEXT: v_add_nc_u32_e32 v2, v1, v0
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v1, s5
|
|
; GFX10-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX10-DL-NEXT: s_endpgm
|
|
<2 x i16> addrspace(1)* %src2,
|
|
i32 addrspace(1)* nocapture %dst) {
|
|
entry:
|
|
%vec1 = load <2 x i16>, <2 x i16> addrspace(1)* %src1
|
|
%vec2 = load <2 x i16>, <2 x i16> addrspace(1)* %src2
|
|
|
|
%s1.elt1 = extractelement <2 x i16> %vec1, i64 0
|
|
%conv = zext i16 %s1.elt1 to i32
|
|
%s2.elt1 = extractelement <2 x i16> %vec2, i64 0
|
|
%conv2 = zext i16 %s2.elt1 to i32
|
|
%mul1 = mul i32 %conv2, %conv
|
|
|
|
%s1.elt2 = extractelement <2 x i16> %vec1, i64 1
|
|
%conv3 = zext i16 %s1.elt2 to i32
|
|
%s2.elt2 = extractelement <2 x i16> %vec2, i64 1
|
|
%conv4 = zext i16 %s2.elt2 to i32
|
|
%mul2 = mul i32 %conv4, %conv3
|
|
|
|
%s3 = load i32, i32 addrspace(1)* %dst, align 4
|
|
%add1 = add i32 %mul2, %s3
|
|
%add2 = add i32 %add1, %mul1
|
|
|
|
%res = add i32 %add2, %add1
|
|
store i32 %res, i32 addrspace(1)* %dst, align 4
|
|
ret void
|
|
}
|
|
|
|
define amdgpu_kernel void @idot2_MultipleUses_add1(<2 x i16> addrspace(1)* %src1,
|
|
; GFX7-LABEL: idot2_MultipleUses_add1:
|
|
; GFX7: ; %bb.0: ; %entry
|
|
; GFX7-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x9
|
|
; GFX7-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0xd
|
|
; GFX7-NEXT: s_mov_b32 s3, 0xf000
|
|
; GFX7-NEXT: s_mov_b32 s2, -1
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_load_dword s4, s[4:5], 0x0
|
|
; GFX7-NEXT: s_load_dword s5, s[6:7], 0x0
|
|
; GFX7-NEXT: s_load_dword s6, s[0:1], 0x0
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_sext_i32_i16 s7, s4
|
|
; GFX7-NEXT: s_ashr_i32 s4, s4, 16
|
|
; GFX7-NEXT: s_sext_i32_i16 s8, s5
|
|
; GFX7-NEXT: s_ashr_i32 s5, s5, 16
|
|
; GFX7-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX7-NEXT: v_mov_b32_e32 v1, s6
|
|
; GFX7-NEXT: v_mad_i32_i24 v0, s5, v0, v1
|
|
; GFX7-NEXT: v_mov_b32_e32 v1, s7
|
|
; GFX7-NEXT: v_mad_i32_i24 v1, s8, v1, v0
|
|
; GFX7-NEXT: v_add_i32_e32 v0, vcc, v0, v1
|
|
; GFX7-NEXT: buffer_store_dword v0, off, s[0:3], 0
|
|
; GFX7-NEXT: s_endpgm
|
|
;
|
|
; GFX8-LABEL: idot2_MultipleUses_add1:
|
|
; GFX8: ; %bb.0: ; %entry
|
|
; GFX8-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX8-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_load_dword s2, s[4:5], 0x0
|
|
; GFX8-NEXT: s_load_dword s3, s[6:7], 0x0
|
|
; GFX8-NEXT: s_load_dword s4, s[0:1], 0x0
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_sext_i32_i16 s5, s2
|
|
; GFX8-NEXT: s_ashr_i32 s2, s2, 16
|
|
; GFX8-NEXT: s_sext_i32_i16 s6, s3
|
|
; GFX8-NEXT: s_ashr_i32 s3, s3, 16
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s2
|
|
; GFX8-NEXT: v_mov_b32_e32 v2, s5
|
|
; GFX8-NEXT: v_mad_i32_i24 v0, s3, v1, v0
|
|
; GFX8-NEXT: v_mad_i32_i24 v1, s6, v2, v0
|
|
; GFX8-NEXT: v_add_u32_e32 v2, vcc, v0, v1
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX8-NEXT: flat_store_dword v[0:1], v2
|
|
; GFX8-NEXT: s_endpgm
|
|
;
|
|
; GFX9-NODL-LABEL: idot2_MultipleUses_add1:
|
|
; GFX9-NODL: ; %bb.0: ; %entry
|
|
; GFX9-NODL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-NODL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_load_dword s2, s[4:5], 0x0
|
|
; GFX9-NODL-NEXT: s_load_dword s3, s[6:7], 0x0
|
|
; GFX9-NODL-NEXT: s_load_dword s4, s[0:1], 0x0
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_sext_i32_i16 s5, s2
|
|
; GFX9-NODL-NEXT: s_ashr_i32 s2, s2, 16
|
|
; GFX9-NODL-NEXT: s_sext_i32_i16 s6, s3
|
|
; GFX9-NODL-NEXT: s_ashr_i32 s3, s3, 16
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s2
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v2, s5
|
|
; GFX9-NODL-NEXT: v_mad_i32_i24 v0, s3, v1, v0
|
|
; GFX9-NODL-NEXT: v_mad_i32_i24 v1, s6, v2, v0
|
|
; GFX9-NODL-NEXT: v_add_u32_e32 v2, v1, v0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-NODL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-NODL-NEXT: s_endpgm
|
|
;
|
|
; GFX9-DL-LABEL: idot2_MultipleUses_add1:
|
|
; GFX9-DL: ; %bb.0: ; %entry
|
|
; GFX9-DL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-DL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: s_load_dword s2, s[4:5], 0x0
|
|
; GFX9-DL-NEXT: s_load_dword s3, s[6:7], 0x0
|
|
; GFX9-DL-NEXT: s_load_dword s4, s[0:1], 0x0
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: s_sext_i32_i16 s5, s2
|
|
; GFX9-DL-NEXT: s_ashr_i32 s2, s2, 16
|
|
; GFX9-DL-NEXT: s_sext_i32_i16 s6, s3
|
|
; GFX9-DL-NEXT: s_ashr_i32 s3, s3, 16
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s2
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v2, s5
|
|
; GFX9-DL-NEXT: v_mad_i32_i24 v0, s3, v1, v0
|
|
; GFX9-DL-NEXT: v_mad_i32_i24 v1, s6, v2, v0
|
|
; GFX9-DL-NEXT: v_add_u32_e32 v2, v1, v0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-DL-NEXT: s_endpgm
|
|
;
|
|
; GFX10-DL-LABEL: idot2_MultipleUses_add1:
|
|
; GFX10-DL: ; %bb.0: ; %entry
|
|
; GFX10-DL-NEXT: s_clause 0x1
|
|
; GFX10-DL-NEXT: s_load_dwordx2 s[4:5], s[0:1], 0x34
|
|
; GFX10-DL-NEXT: s_load_dwordx4 s[0:3], s[0:1], 0x24
|
|
; GFX10-DL-NEXT: ; implicit-def: $vcc_hi
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: s_load_dword s6, s[4:5], 0x0
|
|
; GFX10-DL-NEXT: s_load_dword s0, s[0:1], 0x0
|
|
; GFX10-DL-NEXT: s_load_dword s1, s[2:3], 0x0
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s6
|
|
; GFX10-DL-NEXT: s_ashr_i32 s2, s0, 16
|
|
; GFX10-DL-NEXT: s_ashr_i32 s3, s1, 16
|
|
; GFX10-DL-NEXT: s_sext_i32_i16 s0, s0
|
|
; GFX10-DL-NEXT: s_sext_i32_i16 s1, s1
|
|
; GFX10-DL-NEXT: v_mad_i32_i24 v0, s3, s2, v0
|
|
; GFX10-DL-NEXT: v_mad_i32_i24 v1, s1, s0, v0
|
|
; GFX10-DL-NEXT: v_add_nc_u32_e32 v2, v1, v0
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v1, s5
|
|
; GFX10-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX10-DL-NEXT: s_endpgm
|
|
<2 x i16> addrspace(1)* %src2,
|
|
i32 addrspace(1)* nocapture %dst) {
|
|
entry:
|
|
%vec1 = load <2 x i16>, <2 x i16> addrspace(1)* %src1
|
|
%vec2 = load <2 x i16>, <2 x i16> addrspace(1)* %src2
|
|
|
|
%s1.elt1 = extractelement <2 x i16> %vec1, i64 0
|
|
%conv = sext i16 %s1.elt1 to i32
|
|
%s2.elt1 = extractelement <2 x i16> %vec2, i64 0
|
|
%conv2 = sext i16 %s2.elt1 to i32
|
|
%mul1 = mul i32 %conv2, %conv
|
|
|
|
%s1.elt2 = extractelement <2 x i16> %vec1, i64 1
|
|
%conv3 = sext i16 %s1.elt2 to i32
|
|
%s2.elt2 = extractelement <2 x i16> %vec2, i64 1
|
|
%conv4 = sext i16 %s2.elt2 to i32
|
|
%mul2 = mul i32 %conv4, %conv3
|
|
|
|
%s3 = load i32, i32 addrspace(1)* %dst, align 4
|
|
%add1 = add i32 %mul2, %s3
|
|
%add2 = add i32 %add1, %mul1
|
|
|
|
%res = add i32 %add2, %add1
|
|
store i32 %res, i32 addrspace(1)* %dst, align 4
|
|
ret void
|
|
}
|
|
|
|
define amdgpu_kernel void @udot2_MultipleUses_mul1(<2 x i16> addrspace(1)* %src1,
|
|
; GFX7-LABEL: udot2_MultipleUses_mul1:
|
|
; GFX7: ; %bb.0: ; %entry
|
|
; GFX7-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x9
|
|
; GFX7-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0xd
|
|
; GFX7-NEXT: s_mov_b32 s8, 0xffff
|
|
; GFX7-NEXT: s_mov_b32 s3, 0xf000
|
|
; GFX7-NEXT: s_mov_b32 s2, -1
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_load_dword s4, s[4:5], 0x0
|
|
; GFX7-NEXT: s_load_dword s5, s[6:7], 0x0
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_lshr_b32 s6, s4, 16
|
|
; GFX7-NEXT: s_lshr_b32 s7, s5, 16
|
|
; GFX7-NEXT: s_and_b32 s4, s4, s8
|
|
; GFX7-NEXT: s_and_b32 s5, s5, s8
|
|
; GFX7-NEXT: s_load_dword s8, s[0:1], 0x0
|
|
; GFX7-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX7-NEXT: v_mov_b32_e32 v2, s6
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: v_mov_b32_e32 v1, s8
|
|
; GFX7-NEXT: v_mad_u32_u24 v1, s5, v0, v1
|
|
; GFX7-NEXT: v_mad_u32_u24 v1, s7, v2, v1
|
|
; GFX7-NEXT: v_mad_u32_u24 v0, s5, v0, v1
|
|
; GFX7-NEXT: buffer_store_dword v0, off, s[0:3], 0
|
|
; GFX7-NEXT: s_endpgm
|
|
;
|
|
; GFX8-LABEL: udot2_MultipleUses_mul1:
|
|
; GFX8: ; %bb.0: ; %entry
|
|
; GFX8-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX8-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX8-NEXT: s_mov_b32 s2, 0xffff
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_load_dword s3, s[4:5], 0x0
|
|
; GFX8-NEXT: s_load_dword s4, s[6:7], 0x0
|
|
; GFX8-NEXT: s_load_dword s5, s[0:1], 0x0
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_and_b32 s6, s3, s2
|
|
; GFX8-NEXT: s_and_b32 s2, s4, s2
|
|
; GFX8-NEXT: s_lshr_b32 s3, s3, 16
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s5
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s6
|
|
; GFX8-NEXT: s_lshr_b32 s4, s4, 16
|
|
; GFX8-NEXT: v_mad_u32_u24 v0, s2, v1, v0
|
|
; GFX8-NEXT: v_mov_b32_e32 v2, s3
|
|
; GFX8-NEXT: v_mad_u32_u24 v0, s4, v2, v0
|
|
; GFX8-NEXT: v_mad_u32_u24 v2, s2, v1, v0
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX8-NEXT: flat_store_dword v[0:1], v2
|
|
; GFX8-NEXT: s_endpgm
|
|
;
|
|
; GFX9-NODL-LABEL: udot2_MultipleUses_mul1:
|
|
; GFX9-NODL: ; %bb.0: ; %entry
|
|
; GFX9-NODL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-NODL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-NODL-NEXT: s_mov_b32 s2, 0xffff
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_load_dword s3, s[4:5], 0x0
|
|
; GFX9-NODL-NEXT: s_load_dword s4, s[6:7], 0x0
|
|
; GFX9-NODL-NEXT: s_load_dword s5, s[0:1], 0x0
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_and_b32 s6, s3, s2
|
|
; GFX9-NODL-NEXT: s_and_b32 s2, s4, s2
|
|
; GFX9-NODL-NEXT: s_lshr_b32 s3, s3, 16
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s5
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s6
|
|
; GFX9-NODL-NEXT: s_lshr_b32 s4, s4, 16
|
|
; GFX9-NODL-NEXT: v_mad_u32_u24 v0, s2, v1, v0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v2, s3
|
|
; GFX9-NODL-NEXT: v_mad_u32_u24 v0, s4, v2, v0
|
|
; GFX9-NODL-NEXT: v_mad_u32_u24 v2, s2, v1, v0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-NODL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-NODL-NEXT: s_endpgm
|
|
;
|
|
; GFX9-DL-LABEL: udot2_MultipleUses_mul1:
|
|
; GFX9-DL: ; %bb.0: ; %entry
|
|
; GFX9-DL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-DL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-DL-NEXT: s_mov_b32 s2, 0xffff
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: s_load_dword s3, s[4:5], 0x0
|
|
; GFX9-DL-NEXT: s_load_dword s4, s[6:7], 0x0
|
|
; GFX9-DL-NEXT: s_load_dword s5, s[0:1], 0x0
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: s_and_b32 s6, s3, s2
|
|
; GFX9-DL-NEXT: s_and_b32 s2, s4, s2
|
|
; GFX9-DL-NEXT: s_lshr_b32 s3, s3, 16
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s5
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s6
|
|
; GFX9-DL-NEXT: s_lshr_b32 s4, s4, 16
|
|
; GFX9-DL-NEXT: v_mad_u32_u24 v0, s2, v1, v0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v2, s3
|
|
; GFX9-DL-NEXT: v_mad_u32_u24 v0, s4, v2, v0
|
|
; GFX9-DL-NEXT: v_mad_u32_u24 v2, s2, v1, v0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-DL-NEXT: s_endpgm
|
|
;
|
|
; GFX10-DL-LABEL: udot2_MultipleUses_mul1:
|
|
; GFX10-DL: ; %bb.0: ; %entry
|
|
; GFX10-DL-NEXT: s_clause 0x1
|
|
; GFX10-DL-NEXT: s_load_dwordx2 s[4:5], s[0:1], 0x34
|
|
; GFX10-DL-NEXT: s_load_dwordx4 s[0:3], s[0:1], 0x24
|
|
; GFX10-DL-NEXT: ; implicit-def: $vcc_hi
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: s_load_dword s6, s[4:5], 0x0
|
|
; GFX10-DL-NEXT: s_load_dword s0, s[0:1], 0x0
|
|
; GFX10-DL-NEXT: s_load_dword s1, s[2:3], 0x0
|
|
; GFX10-DL-NEXT: s_mov_b32 s2, 0xffff
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s6
|
|
; GFX10-DL-NEXT: s_and_b32 s3, s0, s2
|
|
; GFX10-DL-NEXT: s_and_b32 s2, s1, s2
|
|
; GFX10-DL-NEXT: s_lshr_b32 s0, s0, 16
|
|
; GFX10-DL-NEXT: s_lshr_b32 s1, s1, 16
|
|
; GFX10-DL-NEXT: v_mad_u32_u24 v0, s2, s3, v0
|
|
; GFX10-DL-NEXT: v_mad_u32_u24 v0, s1, s0, v0
|
|
; GFX10-DL-NEXT: v_mad_u32_u24 v2, s2, s3, v0
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v1, s5
|
|
; GFX10-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX10-DL-NEXT: s_endpgm
|
|
<2 x i16> addrspace(1)* %src2,
|
|
i32 addrspace(1)* nocapture %dst) {
|
|
entry:
|
|
%vec1 = load <2 x i16>, <2 x i16> addrspace(1)* %src1
|
|
%vec2 = load <2 x i16>, <2 x i16> addrspace(1)* %src2
|
|
|
|
%s1.elt1 = extractelement <2 x i16> %vec1, i64 0
|
|
%conv = zext i16 %s1.elt1 to i32
|
|
%s2.elt1 = extractelement <2 x i16> %vec2, i64 0
|
|
%conv2 = zext i16 %s2.elt1 to i32
|
|
%mul1 = mul i32 %conv2, %conv
|
|
|
|
%s1.elt2 = extractelement <2 x i16> %vec1, i64 1
|
|
%conv3 = zext i16 %s1.elt2 to i32
|
|
%s2.elt2 = extractelement <2 x i16> %vec2, i64 1
|
|
%conv4 = zext i16 %s2.elt2 to i32
|
|
%mul2 = mul i32 %conv4, %conv3
|
|
|
|
%s3 = load i32, i32 addrspace(1)* %dst, align 4
|
|
%add0 = add i32 %mul1, %s3
|
|
|
|
%add1 = add i32 %mul2, %add0
|
|
%add2 = add i32 %add1, %mul1
|
|
|
|
store i32 %add2, i32 addrspace(1)* %dst, align 4
|
|
ret void
|
|
}
|
|
|
|
define amdgpu_kernel void @idot2_MultipleUses_mul1(<2 x i16> addrspace(1)* %src1,
|
|
; GFX7-LABEL: idot2_MultipleUses_mul1:
|
|
; GFX7: ; %bb.0: ; %entry
|
|
; GFX7-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x9
|
|
; GFX7-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0xd
|
|
; GFX7-NEXT: s_mov_b32 s3, 0xf000
|
|
; GFX7-NEXT: s_mov_b32 s2, -1
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_load_dword s4, s[4:5], 0x0
|
|
; GFX7-NEXT: s_load_dword s5, s[6:7], 0x0
|
|
; GFX7-NEXT: s_load_dword s6, s[0:1], 0x0
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_sext_i32_i16 s7, s4
|
|
; GFX7-NEXT: s_sext_i32_i16 s8, s5
|
|
; GFX7-NEXT: s_ashr_i32 s4, s4, 16
|
|
; GFX7-NEXT: v_mov_b32_e32 v0, s7
|
|
; GFX7-NEXT: v_mov_b32_e32 v1, s6
|
|
; GFX7-NEXT: s_ashr_i32 s5, s5, 16
|
|
; GFX7-NEXT: v_mad_i32_i24 v1, s8, v0, v1
|
|
; GFX7-NEXT: v_mov_b32_e32 v2, s4
|
|
; GFX7-NEXT: v_mad_i32_i24 v1, s5, v2, v1
|
|
; GFX7-NEXT: v_mad_i32_i24 v0, s8, v0, v1
|
|
; GFX7-NEXT: buffer_store_dword v0, off, s[0:3], 0
|
|
; GFX7-NEXT: s_endpgm
|
|
;
|
|
; GFX8-LABEL: idot2_MultipleUses_mul1:
|
|
; GFX8: ; %bb.0: ; %entry
|
|
; GFX8-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX8-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_load_dword s2, s[4:5], 0x0
|
|
; GFX8-NEXT: s_load_dword s3, s[6:7], 0x0
|
|
; GFX8-NEXT: s_load_dword s4, s[0:1], 0x0
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_sext_i32_i16 s5, s2
|
|
; GFX8-NEXT: s_sext_i32_i16 s6, s3
|
|
; GFX8-NEXT: s_ashr_i32 s2, s2, 16
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s5
|
|
; GFX8-NEXT: s_ashr_i32 s3, s3, 16
|
|
; GFX8-NEXT: v_mov_b32_e32 v2, s2
|
|
; GFX8-NEXT: v_mad_i32_i24 v0, s6, v1, v0
|
|
; GFX8-NEXT: v_mad_i32_i24 v0, s3, v2, v0
|
|
; GFX8-NEXT: v_mad_i32_i24 v2, s6, v1, v0
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX8-NEXT: flat_store_dword v[0:1], v2
|
|
; GFX8-NEXT: s_endpgm
|
|
;
|
|
; GFX9-NODL-LABEL: idot2_MultipleUses_mul1:
|
|
; GFX9-NODL: ; %bb.0: ; %entry
|
|
; GFX9-NODL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-NODL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_load_dword s2, s[4:5], 0x0
|
|
; GFX9-NODL-NEXT: s_load_dword s3, s[6:7], 0x0
|
|
; GFX9-NODL-NEXT: s_load_dword s4, s[0:1], 0x0
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_sext_i32_i16 s5, s2
|
|
; GFX9-NODL-NEXT: s_sext_i32_i16 s6, s3
|
|
; GFX9-NODL-NEXT: s_ashr_i32 s2, s2, 16
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s5
|
|
; GFX9-NODL-NEXT: s_ashr_i32 s3, s3, 16
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v2, s2
|
|
; GFX9-NODL-NEXT: v_mad_i32_i24 v0, s6, v1, v0
|
|
; GFX9-NODL-NEXT: v_mad_i32_i24 v0, s3, v2, v0
|
|
; GFX9-NODL-NEXT: v_mad_i32_i24 v2, s6, v1, v0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-NODL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-NODL-NEXT: s_endpgm
|
|
;
|
|
; GFX9-DL-LABEL: idot2_MultipleUses_mul1:
|
|
; GFX9-DL: ; %bb.0: ; %entry
|
|
; GFX9-DL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-DL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: s_load_dword s2, s[4:5], 0x0
|
|
; GFX9-DL-NEXT: s_load_dword s3, s[6:7], 0x0
|
|
; GFX9-DL-NEXT: s_load_dword s4, s[0:1], 0x0
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: s_sext_i32_i16 s5, s2
|
|
; GFX9-DL-NEXT: s_sext_i32_i16 s6, s3
|
|
; GFX9-DL-NEXT: s_ashr_i32 s2, s2, 16
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s5
|
|
; GFX9-DL-NEXT: s_ashr_i32 s3, s3, 16
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v2, s2
|
|
; GFX9-DL-NEXT: v_mad_i32_i24 v0, s6, v1, v0
|
|
; GFX9-DL-NEXT: v_mad_i32_i24 v0, s3, v2, v0
|
|
; GFX9-DL-NEXT: v_mad_i32_i24 v2, s6, v1, v0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-DL-NEXT: s_endpgm
|
|
;
|
|
; GFX10-DL-LABEL: idot2_MultipleUses_mul1:
|
|
; GFX10-DL: ; %bb.0: ; %entry
|
|
; GFX10-DL-NEXT: s_clause 0x1
|
|
; GFX10-DL-NEXT: s_load_dwordx2 s[4:5], s[0:1], 0x34
|
|
; GFX10-DL-NEXT: s_load_dwordx4 s[0:3], s[0:1], 0x24
|
|
; GFX10-DL-NEXT: ; implicit-def: $vcc_hi
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: s_load_dword s6, s[4:5], 0x0
|
|
; GFX10-DL-NEXT: s_load_dword s0, s[0:1], 0x0
|
|
; GFX10-DL-NEXT: s_load_dword s1, s[2:3], 0x0
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s6
|
|
; GFX10-DL-NEXT: s_sext_i32_i16 s2, s0
|
|
; GFX10-DL-NEXT: s_sext_i32_i16 s3, s1
|
|
; GFX10-DL-NEXT: s_ashr_i32 s0, s0, 16
|
|
; GFX10-DL-NEXT: s_ashr_i32 s1, s1, 16
|
|
; GFX10-DL-NEXT: v_mad_i32_i24 v0, s3, s2, v0
|
|
; GFX10-DL-NEXT: v_mad_i32_i24 v0, s1, s0, v0
|
|
; GFX10-DL-NEXT: v_mad_i32_i24 v2, s3, s2, v0
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v1, s5
|
|
; GFX10-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX10-DL-NEXT: s_endpgm
|
|
<2 x i16> addrspace(1)* %src2,
|
|
i32 addrspace(1)* nocapture %dst) {
|
|
entry:
|
|
%vec1 = load <2 x i16>, <2 x i16> addrspace(1)* %src1
|
|
%vec2 = load <2 x i16>, <2 x i16> addrspace(1)* %src2
|
|
|
|
%s1.elt1 = extractelement <2 x i16> %vec1, i64 0
|
|
%conv = sext i16 %s1.elt1 to i32
|
|
%s2.elt1 = extractelement <2 x i16> %vec2, i64 0
|
|
%conv2 = sext i16 %s2.elt1 to i32
|
|
%mul1 = mul i32 %conv2, %conv
|
|
|
|
%s1.elt2 = extractelement <2 x i16> %vec1, i64 1
|
|
%conv3 = sext i16 %s1.elt2 to i32
|
|
%s2.elt2 = extractelement <2 x i16> %vec2, i64 1
|
|
%conv4 = sext i16 %s2.elt2 to i32
|
|
%mul2 = mul i32 %conv4, %conv3
|
|
|
|
%s3 = load i32, i32 addrspace(1)* %dst, align 4
|
|
%add0 = add i32 %mul1, %s3
|
|
|
|
%add1 = add i32 %mul2, %add0
|
|
%add2 = add i32 %add1, %mul1
|
|
|
|
store i32 %add2, i32 addrspace(1)* %dst, align 4
|
|
ret void
|
|
}
|
|
|
|
define amdgpu_kernel void @udot2_MultipleUses_mul2(<2 x i16> addrspace(1)* %src1,
|
|
; GFX7-LABEL: udot2_MultipleUses_mul2:
|
|
; GFX7: ; %bb.0: ; %entry
|
|
; GFX7-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x9
|
|
; GFX7-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0xd
|
|
; GFX7-NEXT: s_mov_b32 s8, 0xffff
|
|
; GFX7-NEXT: s_mov_b32 s3, 0xf000
|
|
; GFX7-NEXT: s_mov_b32 s2, -1
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_load_dword s4, s[4:5], 0x0
|
|
; GFX7-NEXT: s_load_dword s5, s[6:7], 0x0
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_lshr_b32 s6, s4, 16
|
|
; GFX7-NEXT: s_lshr_b32 s7, s5, 16
|
|
; GFX7-NEXT: s_and_b32 s4, s4, s8
|
|
; GFX7-NEXT: s_and_b32 s5, s5, s8
|
|
; GFX7-NEXT: s_load_dword s8, s[0:1], 0x0
|
|
; GFX7-NEXT: v_mov_b32_e32 v0, s6
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: v_mov_b32_e32 v1, s8
|
|
; GFX7-NEXT: v_mad_u32_u24 v1, s7, v0, v1
|
|
; GFX7-NEXT: v_mad_u32_u24 v0, s7, v0, v1
|
|
; GFX7-NEXT: v_mov_b32_e32 v1, s4
|
|
; GFX7-NEXT: v_mad_u32_u24 v0, s5, v1, v0
|
|
; GFX7-NEXT: buffer_store_dword v0, off, s[0:3], 0
|
|
; GFX7-NEXT: s_endpgm
|
|
;
|
|
; GFX8-LABEL: udot2_MultipleUses_mul2:
|
|
; GFX8: ; %bb.0: ; %entry
|
|
; GFX8-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX8-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX8-NEXT: s_mov_b32 s2, 0xffff
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_load_dword s3, s[4:5], 0x0
|
|
; GFX8-NEXT: s_load_dword s4, s[6:7], 0x0
|
|
; GFX8-NEXT: s_load_dword s5, s[0:1], 0x0
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_and_b32 s6, s3, s2
|
|
; GFX8-NEXT: s_lshr_b32 s3, s3, 16
|
|
; GFX8-NEXT: s_and_b32 s2, s4, s2
|
|
; GFX8-NEXT: s_lshr_b32 s4, s4, 16
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s5
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s3
|
|
; GFX8-NEXT: v_mad_u32_u24 v0, s4, v1, v0
|
|
; GFX8-NEXT: v_mad_u32_u24 v0, s4, v1, v0
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s6
|
|
; GFX8-NEXT: v_mad_u32_u24 v2, s2, v1, v0
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX8-NEXT: flat_store_dword v[0:1], v2
|
|
; GFX8-NEXT: s_endpgm
|
|
;
|
|
; GFX9-NODL-LABEL: udot2_MultipleUses_mul2:
|
|
; GFX9-NODL: ; %bb.0: ; %entry
|
|
; GFX9-NODL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-NODL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-NODL-NEXT: s_mov_b32 s2, 0xffff
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_load_dword s3, s[4:5], 0x0
|
|
; GFX9-NODL-NEXT: s_load_dword s4, s[6:7], 0x0
|
|
; GFX9-NODL-NEXT: s_load_dword s5, s[0:1], 0x0
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_and_b32 s6, s3, s2
|
|
; GFX9-NODL-NEXT: s_lshr_b32 s3, s3, 16
|
|
; GFX9-NODL-NEXT: s_and_b32 s2, s4, s2
|
|
; GFX9-NODL-NEXT: s_lshr_b32 s4, s4, 16
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s5
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s3
|
|
; GFX9-NODL-NEXT: v_mad_u32_u24 v0, s4, v1, v0
|
|
; GFX9-NODL-NEXT: v_mad_u32_u24 v0, s4, v1, v0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s6
|
|
; GFX9-NODL-NEXT: v_mad_u32_u24 v2, s2, v1, v0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-NODL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-NODL-NEXT: s_endpgm
|
|
;
|
|
; GFX9-DL-LABEL: udot2_MultipleUses_mul2:
|
|
; GFX9-DL: ; %bb.0: ; %entry
|
|
; GFX9-DL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-DL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-DL-NEXT: s_mov_b32 s2, 0xffff
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: s_load_dword s3, s[4:5], 0x0
|
|
; GFX9-DL-NEXT: s_load_dword s4, s[6:7], 0x0
|
|
; GFX9-DL-NEXT: s_load_dword s5, s[0:1], 0x0
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: s_and_b32 s6, s3, s2
|
|
; GFX9-DL-NEXT: s_lshr_b32 s3, s3, 16
|
|
; GFX9-DL-NEXT: s_and_b32 s2, s4, s2
|
|
; GFX9-DL-NEXT: s_lshr_b32 s4, s4, 16
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s5
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s3
|
|
; GFX9-DL-NEXT: v_mad_u32_u24 v0, s4, v1, v0
|
|
; GFX9-DL-NEXT: v_mad_u32_u24 v0, s4, v1, v0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s6
|
|
; GFX9-DL-NEXT: v_mad_u32_u24 v2, s2, v1, v0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-DL-NEXT: s_endpgm
|
|
;
|
|
; GFX10-DL-LABEL: udot2_MultipleUses_mul2:
|
|
; GFX10-DL: ; %bb.0: ; %entry
|
|
; GFX10-DL-NEXT: s_clause 0x1
|
|
; GFX10-DL-NEXT: s_load_dwordx2 s[4:5], s[0:1], 0x34
|
|
; GFX10-DL-NEXT: s_load_dwordx4 s[0:3], s[0:1], 0x24
|
|
; GFX10-DL-NEXT: ; implicit-def: $vcc_hi
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: s_load_dword s6, s[4:5], 0x0
|
|
; GFX10-DL-NEXT: s_load_dword s0, s[0:1], 0x0
|
|
; GFX10-DL-NEXT: s_load_dword s1, s[2:3], 0x0
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s6
|
|
; GFX10-DL-NEXT: s_lshr_b32 s2, s0, 16
|
|
; GFX10-DL-NEXT: s_lshr_b32 s3, s1, 16
|
|
; GFX10-DL-NEXT: s_mov_b32 s6, 0xffff
|
|
; GFX10-DL-NEXT: v_mad_u32_u24 v0, s3, s2, v0
|
|
; GFX10-DL-NEXT: s_and_b32 s0, s0, s6
|
|
; GFX10-DL-NEXT: s_and_b32 s1, s1, s6
|
|
; GFX10-DL-NEXT: v_mad_u32_u24 v0, s3, s2, v0
|
|
; GFX10-DL-NEXT: v_mad_u32_u24 v2, s1, s0, v0
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v1, s5
|
|
; GFX10-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX10-DL-NEXT: s_endpgm
|
|
<2 x i16> addrspace(1)* %src2,
|
|
i32 addrspace(1)* nocapture %dst) {
|
|
entry:
|
|
%vec1 = load <2 x i16>, <2 x i16> addrspace(1)* %src1
|
|
%vec2 = load <2 x i16>, <2 x i16> addrspace(1)* %src2
|
|
|
|
%s1.elt1 = extractelement <2 x i16> %vec1, i64 0
|
|
%conv = zext i16 %s1.elt1 to i32
|
|
%s2.elt1 = extractelement <2 x i16> %vec2, i64 0
|
|
%conv2 = zext i16 %s2.elt1 to i32
|
|
%mul1 = mul i32 %conv2, %conv
|
|
|
|
%s1.elt2 = extractelement <2 x i16> %vec1, i64 1
|
|
%conv3 = zext i16 %s1.elt2 to i32
|
|
%s2.elt2 = extractelement <2 x i16> %vec2, i64 1
|
|
%conv4 = zext i16 %s2.elt2 to i32
|
|
%mul2 = mul i32 %conv4, %conv3
|
|
|
|
%s3 = load i32, i32 addrspace(1)* %dst, align 4
|
|
%add0 = add i32 %mul2, %s3
|
|
|
|
%add1 = add i32 %mul2, %add0
|
|
%add2 = add i32 %add1, %mul1
|
|
|
|
store i32 %add2, i32 addrspace(1)* %dst, align 4
|
|
ret void
|
|
}
|
|
|
|
define amdgpu_kernel void @idot2_MultipleUses_mul2(<2 x i16> addrspace(1)* %src1,
|
|
; GFX7-LABEL: idot2_MultipleUses_mul2:
|
|
; GFX7: ; %bb.0: ; %entry
|
|
; GFX7-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x9
|
|
; GFX7-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0xd
|
|
; GFX7-NEXT: s_mov_b32 s3, 0xf000
|
|
; GFX7-NEXT: s_mov_b32 s2, -1
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_load_dword s4, s[4:5], 0x0
|
|
; GFX7-NEXT: s_load_dword s5, s[6:7], 0x0
|
|
; GFX7-NEXT: s_load_dword s6, s[0:1], 0x0
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_sext_i32_i16 s7, s4
|
|
; GFX7-NEXT: s_ashr_i32 s4, s4, 16
|
|
; GFX7-NEXT: s_sext_i32_i16 s8, s5
|
|
; GFX7-NEXT: s_ashr_i32 s5, s5, 16
|
|
; GFX7-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX7-NEXT: v_mov_b32_e32 v1, s6
|
|
; GFX7-NEXT: v_mad_i32_i24 v1, s5, v0, v1
|
|
; GFX7-NEXT: v_mad_i32_i24 v0, s5, v0, v1
|
|
; GFX7-NEXT: v_mov_b32_e32 v1, s7
|
|
; GFX7-NEXT: v_mad_i32_i24 v0, s8, v1, v0
|
|
; GFX7-NEXT: buffer_store_dword v0, off, s[0:3], 0
|
|
; GFX7-NEXT: s_endpgm
|
|
;
|
|
; GFX8-LABEL: idot2_MultipleUses_mul2:
|
|
; GFX8: ; %bb.0: ; %entry
|
|
; GFX8-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX8-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_load_dword s2, s[4:5], 0x0
|
|
; GFX8-NEXT: s_load_dword s3, s[6:7], 0x0
|
|
; GFX8-NEXT: s_load_dword s4, s[0:1], 0x0
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_sext_i32_i16 s5, s2
|
|
; GFX8-NEXT: s_ashr_i32 s2, s2, 16
|
|
; GFX8-NEXT: s_sext_i32_i16 s6, s3
|
|
; GFX8-NEXT: s_ashr_i32 s3, s3, 16
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s2
|
|
; GFX8-NEXT: v_mad_i32_i24 v0, s3, v1, v0
|
|
; GFX8-NEXT: v_mov_b32_e32 v2, s5
|
|
; GFX8-NEXT: v_mad_i32_i24 v0, s3, v1, v0
|
|
; GFX8-NEXT: v_mad_i32_i24 v2, s6, v2, v0
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX8-NEXT: flat_store_dword v[0:1], v2
|
|
; GFX8-NEXT: s_endpgm
|
|
;
|
|
; GFX9-NODL-LABEL: idot2_MultipleUses_mul2:
|
|
; GFX9-NODL: ; %bb.0: ; %entry
|
|
; GFX9-NODL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-NODL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_load_dword s2, s[4:5], 0x0
|
|
; GFX9-NODL-NEXT: s_load_dword s3, s[6:7], 0x0
|
|
; GFX9-NODL-NEXT: s_load_dword s4, s[0:1], 0x0
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_sext_i32_i16 s5, s2
|
|
; GFX9-NODL-NEXT: s_ashr_i32 s2, s2, 16
|
|
; GFX9-NODL-NEXT: s_sext_i32_i16 s6, s3
|
|
; GFX9-NODL-NEXT: s_ashr_i32 s3, s3, 16
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s2
|
|
; GFX9-NODL-NEXT: v_mad_i32_i24 v0, s3, v1, v0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v2, s5
|
|
; GFX9-NODL-NEXT: v_mad_i32_i24 v0, s3, v1, v0
|
|
; GFX9-NODL-NEXT: v_mad_i32_i24 v2, s6, v2, v0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-NODL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-NODL-NEXT: s_endpgm
|
|
;
|
|
; GFX9-DL-LABEL: idot2_MultipleUses_mul2:
|
|
; GFX9-DL: ; %bb.0: ; %entry
|
|
; GFX9-DL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-DL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: s_load_dword s2, s[4:5], 0x0
|
|
; GFX9-DL-NEXT: s_load_dword s3, s[6:7], 0x0
|
|
; GFX9-DL-NEXT: s_load_dword s4, s[0:1], 0x0
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: s_sext_i32_i16 s5, s2
|
|
; GFX9-DL-NEXT: s_ashr_i32 s2, s2, 16
|
|
; GFX9-DL-NEXT: s_sext_i32_i16 s6, s3
|
|
; GFX9-DL-NEXT: s_ashr_i32 s3, s3, 16
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s2
|
|
; GFX9-DL-NEXT: v_mad_i32_i24 v0, s3, v1, v0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v2, s5
|
|
; GFX9-DL-NEXT: v_mad_i32_i24 v0, s3, v1, v0
|
|
; GFX9-DL-NEXT: v_mad_i32_i24 v2, s6, v2, v0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-DL-NEXT: s_endpgm
|
|
;
|
|
; GFX10-DL-LABEL: idot2_MultipleUses_mul2:
|
|
; GFX10-DL: ; %bb.0: ; %entry
|
|
; GFX10-DL-NEXT: s_clause 0x1
|
|
; GFX10-DL-NEXT: s_load_dwordx2 s[4:5], s[0:1], 0x34
|
|
; GFX10-DL-NEXT: s_load_dwordx4 s[0:3], s[0:1], 0x24
|
|
; GFX10-DL-NEXT: ; implicit-def: $vcc_hi
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: s_load_dword s6, s[4:5], 0x0
|
|
; GFX10-DL-NEXT: s_load_dword s0, s[0:1], 0x0
|
|
; GFX10-DL-NEXT: s_load_dword s1, s[2:3], 0x0
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s6
|
|
; GFX10-DL-NEXT: s_ashr_i32 s2, s0, 16
|
|
; GFX10-DL-NEXT: s_ashr_i32 s3, s1, 16
|
|
; GFX10-DL-NEXT: s_sext_i32_i16 s0, s0
|
|
; GFX10-DL-NEXT: s_sext_i32_i16 s1, s1
|
|
; GFX10-DL-NEXT: v_mad_i32_i24 v0, s3, s2, v0
|
|
; GFX10-DL-NEXT: v_mad_i32_i24 v0, s3, s2, v0
|
|
; GFX10-DL-NEXT: v_mad_i32_i24 v2, s1, s0, v0
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v1, s5
|
|
; GFX10-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX10-DL-NEXT: s_endpgm
|
|
<2 x i16> addrspace(1)* %src2,
|
|
i32 addrspace(1)* nocapture %dst) {
|
|
entry:
|
|
%vec1 = load <2 x i16>, <2 x i16> addrspace(1)* %src1
|
|
%vec2 = load <2 x i16>, <2 x i16> addrspace(1)* %src2
|
|
|
|
%s1.elt1 = extractelement <2 x i16> %vec1, i64 0
|
|
%conv = sext i16 %s1.elt1 to i32
|
|
%s2.elt1 = extractelement <2 x i16> %vec2, i64 0
|
|
%conv2 = sext i16 %s2.elt1 to i32
|
|
%mul1 = mul i32 %conv2, %conv
|
|
|
|
%s1.elt2 = extractelement <2 x i16> %vec1, i64 1
|
|
%conv3 = sext i16 %s1.elt2 to i32
|
|
%s2.elt2 = extractelement <2 x i16> %vec2, i64 1
|
|
%conv4 = sext i16 %s2.elt2 to i32
|
|
%mul2 = mul i32 %conv4, %conv3
|
|
|
|
%s3 = load i32, i32 addrspace(1)* %dst, align 4
|
|
%add0 = add i32 %mul2, %s3
|
|
|
|
%add1 = add i32 %mul2, %add0
|
|
%add2 = add i32 %add1, %mul1
|
|
|
|
store i32 %add2, i32 addrspace(1)* %dst, align 4
|
|
ret void
|
|
}
|
|
|
|
define amdgpu_kernel void @udot2_acc16(<2 x i16> addrspace(1)* %src1,
|
|
; GFX7-LABEL: udot2_acc16:
|
|
; GFX7: ; %bb.0: ; %entry
|
|
; GFX7-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x9
|
|
; GFX7-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0xd
|
|
; GFX7-NEXT: s_mov_b32 s3, 0xf000
|
|
; GFX7-NEXT: s_mov_b32 s2, -1
|
|
; GFX7-NEXT: s_mov_b32 s8, 0xffff
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: buffer_load_ushort v0, off, s[0:3], 0
|
|
; GFX7-NEXT: s_load_dword s4, s[4:5], 0x0
|
|
; GFX7-NEXT: s_load_dword s5, s[6:7], 0x0
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_lshr_b32 s6, s4, 16
|
|
; GFX7-NEXT: s_lshr_b32 s7, s5, 16
|
|
; GFX7-NEXT: v_mov_b32_e32 v1, s7
|
|
; GFX7-NEXT: s_and_b32 s5, s5, s8
|
|
; GFX7-NEXT: s_and_b32 s4, s4, s8
|
|
; GFX7-NEXT: s_waitcnt vmcnt(0)
|
|
; GFX7-NEXT: v_mad_u32_u24 v0, s6, v1, v0
|
|
; GFX7-NEXT: v_mov_b32_e32 v1, s5
|
|
; GFX7-NEXT: v_mad_u32_u24 v0, s4, v1, v0
|
|
; GFX7-NEXT: buffer_store_short v0, off, s[0:3], 0
|
|
; GFX7-NEXT: s_endpgm
|
|
;
|
|
; GFX8-LABEL: udot2_acc16:
|
|
; GFX8: ; %bb.0: ; %entry
|
|
; GFX8-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX8-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX8-NEXT: flat_load_ushort v2, v[0:1]
|
|
; GFX8-NEXT: s_load_dword s1, s[4:5], 0x0
|
|
; GFX8-NEXT: s_load_dword s2, s[6:7], 0x0
|
|
; GFX8-NEXT: s_mov_b32 s0, 0xffff
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: s_and_b32 s3, s2, s0
|
|
; GFX8-NEXT: s_lshr_b32 s2, s2, 16
|
|
; GFX8-NEXT: s_and_b32 s0, s1, s0
|
|
; GFX8-NEXT: s_lshr_b32 s1, s1, 16
|
|
; GFX8-NEXT: v_mov_b32_e32 v3, s2
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0)
|
|
; GFX8-NEXT: v_mad_u32_u24 v2, s1, v3, v2
|
|
; GFX8-NEXT: v_mov_b32_e32 v3, s3
|
|
; GFX8-NEXT: v_mad_u32_u24 v2, s0, v3, v2
|
|
; GFX8-NEXT: flat_store_short v[0:1], v2
|
|
; GFX8-NEXT: s_endpgm
|
|
;
|
|
; GFX9-NODL-LABEL: udot2_acc16:
|
|
; GFX9-NODL: ; %bb.0: ; %entry
|
|
; GFX9-NODL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-NODL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-NODL-NEXT: global_load_ushort v2, v[0:1], off
|
|
; GFX9-NODL-NEXT: s_load_dword s1, s[4:5], 0x0
|
|
; GFX9-NODL-NEXT: s_load_dword s2, s[6:7], 0x0
|
|
; GFX9-NODL-NEXT: s_mov_b32 s0, 0xffff
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: s_and_b32 s3, s2, s0
|
|
; GFX9-NODL-NEXT: s_lshr_b32 s2, s2, 16
|
|
; GFX9-NODL-NEXT: s_and_b32 s0, s1, s0
|
|
; GFX9-NODL-NEXT: s_lshr_b32 s1, s1, 16
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v3, s2
|
|
; GFX9-NODL-NEXT: s_waitcnt vmcnt(0)
|
|
; GFX9-NODL-NEXT: v_mad_u32_u24 v2, s1, v3, v2
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v3, s3
|
|
; GFX9-NODL-NEXT: v_mad_u32_u24 v2, s0, v3, v2
|
|
; GFX9-NODL-NEXT: global_store_short v[0:1], v2, off
|
|
; GFX9-NODL-NEXT: s_endpgm
|
|
;
|
|
; GFX9-DL-LABEL: udot2_acc16:
|
|
; GFX9-DL: ; %bb.0: ; %entry
|
|
; GFX9-DL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-DL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: s_load_dword s2, s[4:5], 0x0
|
|
; GFX9-DL-NEXT: s_load_dword s3, s[6:7], 0x0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-DL-NEXT: global_load_ushort v2, v[0:1], off
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v3, s3
|
|
; GFX9-DL-NEXT: s_waitcnt vmcnt(0)
|
|
; GFX9-DL-NEXT: v_dot2_u32_u16 v2, s2, v3, v2
|
|
; GFX9-DL-NEXT: global_store_short v[0:1], v2, off
|
|
; GFX9-DL-NEXT: s_endpgm
|
|
;
|
|
; GFX10-DL-LABEL: udot2_acc16:
|
|
; GFX10-DL: ; %bb.0: ; %entry
|
|
; GFX10-DL-NEXT: s_load_dwordx2 s[2:3], s[0:1], 0x34
|
|
; GFX10-DL-NEXT: ; implicit-def: $vcc_hi
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s2
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v1, s3
|
|
; GFX10-DL-NEXT: s_load_dwordx4 s[0:3], s[0:1], 0x24
|
|
; GFX10-DL-NEXT: global_load_ushort v2, v[0:1], off
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: s_load_dword s0, s[0:1], 0x0
|
|
; GFX10-DL-NEXT: s_load_dword s1, s[2:3], 0x0
|
|
; GFX10-DL-NEXT: s_waitcnt vmcnt(0) lgkmcnt(0)
|
|
; GFX10-DL-NEXT: v_dot2_u32_u16 v2, s0, s1, v2
|
|
; GFX10-DL-NEXT: global_store_short v[0:1], v2, off
|
|
; GFX10-DL-NEXT: s_endpgm
|
|
<2 x i16> addrspace(1)* %src2,
|
|
i16 addrspace(1)* nocapture %dst) {
|
|
entry:
|
|
%v1 = load <2 x i16>, <2 x i16> addrspace(1)* %src1
|
|
%v2 = load <2 x i16>, <2 x i16> addrspace(1)* %src2
|
|
|
|
%v1e1 = extractelement <2 x i16> %v1, i64 0
|
|
%v2e1 = extractelement <2 x i16> %v2, i64 0
|
|
%mul1 = mul i16 %v1e1, %v2e1
|
|
|
|
%v1e2 = extractelement <2 x i16> %v1, i64 1
|
|
%v2e2 = extractelement <2 x i16> %v2, i64 1
|
|
%mul2 = mul i16 %v1e2, %v2e2
|
|
|
|
%s2 = load i16, i16 addrspace(1)* %dst, align 2
|
|
%add1 = add i16 %mul2, %s2
|
|
%add2 = add i16 %add1, %mul1
|
|
store i16 %add2, i16 addrspace(1)* %dst, align 2
|
|
ret void
|
|
}
|
|
|
|
define amdgpu_kernel void @notsdot2_sext8(<2 x i8> addrspace(1)* %src1,
|
|
; GFX7-LABEL: notsdot2_sext8:
|
|
; GFX7: ; %bb.0: ; %entry
|
|
; GFX7-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x9
|
|
; GFX7-NEXT: s_load_dwordx2 s[8:9], s[0:1], 0xd
|
|
; GFX7-NEXT: s_mov_b32 s3, 0xf000
|
|
; GFX7-NEXT: s_mov_b32 s2, -1
|
|
; GFX7-NEXT: s_mov_b32 s10, s2
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: s_mov_b32 s0, s4
|
|
; GFX7-NEXT: s_mov_b32 s1, s5
|
|
; GFX7-NEXT: s_mov_b32 s4, s6
|
|
; GFX7-NEXT: s_mov_b32 s5, s7
|
|
; GFX7-NEXT: s_mov_b32 s6, s2
|
|
; GFX7-NEXT: s_mov_b32 s7, s3
|
|
; GFX7-NEXT: buffer_load_ushort v0, off, s[0:3], 0
|
|
; GFX7-NEXT: buffer_load_ushort v1, off, s[4:7], 0
|
|
; GFX7-NEXT: s_load_dword s0, s[8:9], 0x0
|
|
; GFX7-NEXT: s_mov_b32 s11, s3
|
|
; GFX7-NEXT: s_waitcnt vmcnt(1)
|
|
; GFX7-NEXT: v_bfe_i32 v2, v0, 0, 8
|
|
; GFX7-NEXT: s_waitcnt vmcnt(0)
|
|
; GFX7-NEXT: v_bfe_i32 v3, v1, 0, 8
|
|
; GFX7-NEXT: v_bfe_i32 v0, v0, 8, 8
|
|
; GFX7-NEXT: v_bfe_i32 v1, v1, 8, 8
|
|
; GFX7-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX7-NEXT: v_mad_i32_i24 v0, v1, v0, s0
|
|
; GFX7-NEXT: v_mad_i32_i24 v0, v3, v2, v0
|
|
; GFX7-NEXT: buffer_store_dword v0, off, s[8:11], 0
|
|
; GFX7-NEXT: s_endpgm
|
|
;
|
|
; GFX8-LABEL: notsdot2_sext8:
|
|
; GFX8: ; %bb.0: ; %entry
|
|
; GFX8-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX8-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX8-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s5
|
|
; GFX8-NEXT: v_mov_b32_e32 v2, s6
|
|
; GFX8-NEXT: v_mov_b32_e32 v3, s7
|
|
; GFX8-NEXT: flat_load_ushort v0, v[0:1]
|
|
; GFX8-NEXT: flat_load_ushort v1, v[2:3]
|
|
; GFX8-NEXT: s_load_dword s2, s[0:1], 0x0
|
|
; GFX8-NEXT: s_waitcnt vmcnt(1) lgkmcnt(0)
|
|
; GFX8-NEXT: v_bfe_i32 v2, v0, 0, 8
|
|
; GFX8-NEXT: v_lshrrev_b16_e32 v0, 8, v0
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0)
|
|
; GFX8-NEXT: v_bfe_i32 v3, v1, 0, 8
|
|
; GFX8-NEXT: v_lshrrev_b16_e32 v1, 8, v1
|
|
; GFX8-NEXT: v_bfe_i32 v0, v0, 0, 8
|
|
; GFX8-NEXT: v_bfe_i32 v1, v1, 0, 8
|
|
; GFX8-NEXT: v_mad_i32_i24 v0, v1, v0, s2
|
|
; GFX8-NEXT: v_mad_i32_i24 v2, v3, v2, v0
|
|
; GFX8-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX8-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX8-NEXT: flat_store_dword v[0:1], v2
|
|
; GFX8-NEXT: s_endpgm
|
|
;
|
|
; GFX9-NODL-LABEL: notsdot2_sext8:
|
|
; GFX9-NODL: ; %bb.0: ; %entry
|
|
; GFX9-NODL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-NODL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s5
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v2, s6
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v3, s7
|
|
; GFX9-NODL-NEXT: global_load_ushort v0, v[0:1], off
|
|
; GFX9-NODL-NEXT: global_load_ushort v1, v[2:3], off
|
|
; GFX9-NODL-NEXT: s_load_dword s2, s[0:1], 0x0
|
|
; GFX9-NODL-NEXT: s_waitcnt vmcnt(1)
|
|
; GFX9-NODL-NEXT: v_bfe_i32 v2, v0, 0, 8
|
|
; GFX9-NODL-NEXT: v_lshrrev_b16_e32 v0, 8, v0
|
|
; GFX9-NODL-NEXT: s_waitcnt vmcnt(0)
|
|
; GFX9-NODL-NEXT: v_bfe_i32 v3, v1, 0, 8
|
|
; GFX9-NODL-NEXT: v_lshrrev_b16_e32 v1, 8, v1
|
|
; GFX9-NODL-NEXT: v_bfe_i32 v0, v0, 0, 8
|
|
; GFX9-NODL-NEXT: v_bfe_i32 v1, v1, 0, 8
|
|
; GFX9-NODL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-NODL-NEXT: v_mad_i32_i24 v0, v1, v0, s2
|
|
; GFX9-NODL-NEXT: v_mad_i32_i24 v2, v3, v2, v0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-NODL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-NODL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-NODL-NEXT: s_endpgm
|
|
;
|
|
; GFX9-DL-LABEL: notsdot2_sext8:
|
|
; GFX9-DL: ; %bb.0: ; %entry
|
|
; GFX9-DL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX9-DL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s5
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v2, s6
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v3, s7
|
|
; GFX9-DL-NEXT: global_load_ushort v0, v[0:1], off
|
|
; GFX9-DL-NEXT: global_load_ushort v1, v[2:3], off
|
|
; GFX9-DL-NEXT: s_load_dword s2, s[0:1], 0x0
|
|
; GFX9-DL-NEXT: s_waitcnt vmcnt(1)
|
|
; GFX9-DL-NEXT: v_bfe_i32 v2, v0, 0, 8
|
|
; GFX9-DL-NEXT: v_lshrrev_b16_e32 v0, 8, v0
|
|
; GFX9-DL-NEXT: s_waitcnt vmcnt(0)
|
|
; GFX9-DL-NEXT: v_bfe_i32 v3, v1, 0, 8
|
|
; GFX9-DL-NEXT: v_lshrrev_b16_e32 v1, 8, v1
|
|
; GFX9-DL-NEXT: v_bfe_i32 v0, v0, 0, 8
|
|
; GFX9-DL-NEXT: v_bfe_i32 v1, v1, 0, 8
|
|
; GFX9-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX9-DL-NEXT: v_mad_i32_i24 v0, v1, v0, s2
|
|
; GFX9-DL-NEXT: v_mad_i32_i24 v2, v3, v2, v0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX9-DL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX9-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX9-DL-NEXT: s_endpgm
|
|
;
|
|
; GFX10-DL-LABEL: notsdot2_sext8:
|
|
; GFX10-DL: ; %bb.0: ; %entry
|
|
; GFX10-DL-NEXT: s_clause 0x1
|
|
; GFX10-DL-NEXT: s_load_dwordx4 s[4:7], s[0:1], 0x24
|
|
; GFX10-DL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x34
|
|
; GFX10-DL-NEXT: ; implicit-def: $vcc_hi
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s4
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v1, s5
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v2, s6
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v3, s7
|
|
; GFX10-DL-NEXT: global_load_ushort v0, v[0:1], off
|
|
; GFX10-DL-NEXT: global_load_ushort v1, v[2:3], off
|
|
; GFX10-DL-NEXT: s_load_dword s2, s[0:1], 0x0
|
|
; GFX10-DL-NEXT: s_waitcnt vmcnt(1)
|
|
; GFX10-DL-NEXT: v_lshrrev_b16_e64 v2, 8, v0
|
|
; GFX10-DL-NEXT: s_waitcnt vmcnt(0)
|
|
; GFX10-DL-NEXT: v_lshrrev_b16_e64 v3, 8, v1
|
|
; GFX10-DL-NEXT: v_bfe_i32 v0, v0, 0, 8
|
|
; GFX10-DL-NEXT: v_bfe_i32 v1, v1, 0, 8
|
|
; GFX10-DL-NEXT: v_bfe_i32 v2, v2, 0, 8
|
|
; GFX10-DL-NEXT: v_bfe_i32 v3, v3, 0, 8
|
|
; GFX10-DL-NEXT: s_waitcnt lgkmcnt(0)
|
|
; GFX10-DL-NEXT: v_mad_i32_i24 v2, v3, v2, s2
|
|
; GFX10-DL-NEXT: v_mad_i32_i24 v2, v1, v0, v2
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v0, s0
|
|
; GFX10-DL-NEXT: v_mov_b32_e32 v1, s1
|
|
; GFX10-DL-NEXT: global_store_dword v[0:1], v2, off
|
|
; GFX10-DL-NEXT: s_endpgm
|
|
<2 x i8> addrspace(1)* %src2,
|
|
i32 addrspace(1)* nocapture %dst) {
|
|
entry:
|
|
%vec1 = load <2 x i8>, <2 x i8> addrspace(1)* %src1
|
|
%vec2 = load <2 x i8>, <2 x i8> addrspace(1)* %src2
|
|
|
|
%s1.elt1 = extractelement <2 x i8> %vec1, i64 0
|
|
%conv = sext i8 %s1.elt1 to i32
|
|
%s2.elt1 = extractelement <2 x i8> %vec2, i64 0
|
|
%conv2 = sext i8 %s2.elt1 to i32
|
|
%mul1 = mul nuw i32 %conv2, %conv
|
|
|
|
%s1.elt2 = extractelement <2 x i8> %vec1, i64 1
|
|
%conv3 = sext i8 %s1.elt2 to i32
|
|
%s2.elt2 = extractelement <2 x i8> %vec2, i64 1
|
|
%conv4 = sext i8 %s2.elt2 to i32
|
|
%mul2 = mul nuw i32 %conv4, %conv3
|
|
|
|
%s3 = load i32, i32 addrspace(1)* %dst, align 4
|
|
%add = add i32 %mul2, %s3
|
|
%add6 = add i32 %add, %mul1
|
|
store i32 %add6, i32 addrspace(1)* %dst, align 4
|
|
ret void
|
|
}
|