📄 zgemm3m_kernel.s
字号:
(p3) FMA f105 = f41, f61, f105 // A2 * B6 nop __LINE__ } ;;/* 40 */ { .mfi (p5) LDFD f6 = [C1 ], SIZE (p3) FMA f113 = f41, f62, f113 // A2 * B7 nop __LINE__ } { .mfi (p5) LDFD f7 = [C9 ], SIZE (p3) FMA f121 = f41, f63, f121 // A2 * B8 nop __LINE__ } ;; /* 41 */ { .mfi (p5) LDFD f10 = [C1 ], SIZE (p3) FMA f66 = f42, f56, f66 // A3 * B1 nop __LINE__ } { .mfi (p5) LDFD f11 = [C9 ], SIZE (p3) FMA f74 = f42, f57, f74 // A3 * B2 nop __LINE__ } ;;/* 42 */ { .mfi (p5) LDFD f12 = [C1 ], SIZE (p3) FMA f82 = f42, f58, f82 // A3 * B3 nop __LINE__ } { .mfi (p5) LDFD f13 = [C9 ], SIZE (p3) FMA f90 = f42, f59, f90 // A3 * B4 nop __LINE__ } ;;/* 43 */ { .mfi (p5) LDFD f14 = [C1 ], 5 * SIZE (p3) FMA f98 = f42, f60, f98 // A3 * B5 nop __LINE__ } { .mfi (p5) LDFD f15 = [C9 ], 5 * SIZE (p3) FMA f106 = f42, f61, f106 // A3 * B6 nop __LINE__ } ;;/* 44 */ { .mfi (p5) LDFD f16 = [C1 ], SIZE (p3) FMA f114 = f42, f62, f114 // A3 * B7 nop __LINE__ } { .mfi (p5) LDFD f17 = [C9 ], SIZE (p3) FMA f122 = f42, f63, f122 // A3 * B8 nop __LINE__ } ;;/* 45 */ { .mfi (p5) LDFD f18 = [C1 ], SIZE (p3) FMA f67 = f43, f56, f67 // A4 * B1 nop __LINE__ } { .mfi (p5) LDFD f19 = [C9 ], SIZE (p3) FMA f75 = f43, f57, f75 // A4 * B2 nop __LINE__ } ;;/* 46 */ { .mfi (p5) LDFD f20 = [C1 ], SIZE (p3) FMA f83 = f43, f58, f83 // A4 * B3 nop __LINE__ } { .mfi (p5) LDFD f21 = [C9 ], SIZE (p3) FMA f91 = f43, f59, f91 // A4 * B4 nop __LINE__ } ;;/* 47 */ { .mfi (p5) LDFD f22 = [C1 ], - 11 * SIZE (p3) FMA f99 = f43, f60, f99 // A4 * B5 nop __LINE__ } { .mfi (p5) LDFD f23 = [C9 ], - 11 * SIZE (p3) FMA f107 = f43, f61, f107 // A4 * B6 nop __LINE__ } ;;/* 48 */ { .mfi (p5) LDFD f24 = [C2 ], SIZE (p3) FMA f115 = f43, f62, f115 // A4 * B7 nop __LINE__ } { .mfi (p5) LDFD f25 = [C10], SIZE (p3) FMA f123 = f43, f63, f123 // A4 * B8 nop __LINE__ } ;;/* 49 */ { .mfi (p5) LDFD f26 = [C2 ], SIZE (p3) FMA f68 = f44, f56, f68 // A5 * B1 nop __LINE__ } { .mfi (p5) LDFD f27 = [C10], SIZE (p3) FMA f76 = f44, f57, f76 // A5 * B2 nop __LINE__ } ;;/* 50 */ { .mfi (p5) LDFD f28 = [C2 ], SIZE (p3) FMA f84 = f44, f58, f84 // A5 * B3 nop __LINE__ } { .mfi (p5) LDFD f29 = [C10], SIZE (p3) FMA f92 = f44, f59, f92 // A5 * B4 nop __LINE__ } ;;/* 51 */ { .mfi (p5) LDFD f30 = [C2 ], 5 * SIZE (p3) FMA f100 = f44, f60, f100 // A5 * B5 nop __LINE__ } { .mfi (p5) LDFD f31 = [C10], 5 * SIZE (p3) FMA f108 = f44, f61, f108 // A5 * B6 nop __LINE__ } ;;/* 52 */ { .mfi (p5) LDFD f32 = [C2 ], SIZE (p3) FMA f116 = f44, f62, f116 // A5 * B7 nop __LINE__ } { .mfi (p5) LDFD f33 = [C10], SIZE (p3) FMA f124 = f44, f63, f124 // A5 * B8 nop __LINE__ } ;;/* 53 */ { .mfi (p5) LDFD f34 = [C2 ], SIZE (p3) FMA f69 = f45, f56, f69 // A6 * B1 nop __LINE__ } { .mfi (p5) LDFD f35 = [C10], SIZE (p3) FMA f77 = f45, f57, f77 // A6 * B2 nop __LINE__ } ;;/* 54 */ { .mfi (p5) LDFD f36 = [C2 ], SIZE (p3) FMA f85 = f45, f58, f85 // A6 * B3 nop __LINE__ } { .mfi (p5) LDFD f37 = [C10], SIZE (p3) FMA f93 = f45, f59, f93 // A6 * B4 nop __LINE__ } ;;/* 55 */ { .mfi (p5) LDFD f38 = [C2 ], - 11 * SIZE (p3) FMA f101 = f45, f60, f101 // A6 * B5 nop __LINE__ } { .mfi (p5) LDFD f39 = [C10], - 11 * SIZE (p3) FMA f109 = f45, f61, f109 // A6 * B6 nop __LINE__ } ;;/* 56 */ { .mfi (p5) LDFD f48 = [C3 ], SIZE (p3) FMA f117 = f45, f62, f117 // A6 * B7 nop __LINE__ } { .mfi (p5) LDFD f49 = [C11], SIZE (p3) FMA f125 = f45, f63, f125 // A6 * B8 nop __LINE__ } ;;/* 57 */ { .mfi (p5) LDFD f50 = [C3 ], SIZE (p3) FMA f70 = f46, f56, f70 // A7 * B1 nop __LINE__ } { .mfi (p5) LDFD f51 = [C11], SIZE (p3) FMA f78 = f46, f57, f78 // A7 * B2 nop __LINE__ } ;;/* 58 */ { .mfi (p5) LDFD f52 = [C3 ], SIZE (p3) FMA f86 = f46, f58, f86 // A7 * B3 nop __LINE__ } { .mfi (p5) LDFD f53 = [C11], SIZE (p3) FMA f94 = f46, f59, f94 // A7 * B4 nop __LINE__ } ;;/* 59 */ { .mfi (p5) LDFD f54 = [C3 ], 5 * SIZE (p3) FMA f102 = f46, f60, f102 // A7 * B5 nop __LINE__ } { .mfi (p5) LDFD f55 = [C11], 5 * SIZE (p3) FMA f110 = f46, f61, f110 // A7 * B6 nop __LINE__ } ;;/* 60 */ { .mfi (p5) LDFD f40 = [C3 ], SIZE (p3) FMA f118 = f46, f62, f118 // A7 * B7 nop __LINE__ } { .mfi (p5) LDFD f41 = [C11], SIZE (p3) FMA f126 = f46, f63, f126 // A7 * B8 nop __LINE__ } ;;/* 61 */ { .mfi (p5) LDFD f42 = [C3 ], SIZE (p3) FMA f71 = f47, f56, f71 // A8 * B1 nop __LINE__ } { .mfi (p5) LDFD f43 = [C11], SIZE (p3) FMA f79 = f47, f57, f79 // A8 * B2 nop __LINE__ } ;;/* 62 */ { .mfi (p5) LDFD f44 = [C3 ], SIZE (p3) FMA f87 = f47, f58, f87 // A8 * B3 nop __LINE__ } { .mfi (p5) LDFD f45 = [C11], SIZE (p3) FMA f95 = f47, f59, f95 // A8 * B4 nop __LINE__ } ;;/* 63 */ { .mfi (p5) LDFD f46 = [C3 ], - 11 * SIZE (p3) FMA f103 = f47, f60, f103 // A8 * B5 nop __LINE__ } { .mfi (p5) LDFD f56 = [C11], - 11 * SIZE (p3) FMA f111 = f47, f61, f111 // A8 * B6 nop __LINE__ } ;;/* 64 */ { .mfi (p5) LDFD f57 = [C4 ], SIZE (p3) FMA f119 = f47, f62, f119 // A8 * B7 adds L = -1, L } { .mfb (p5) LDFD f58 = [C12], SIZE (p3) FMA f127 = f47, f63, f127 // A8 * B8 br.cloop.sptk.few .L012 } ;;.L013: { .mmf (p5) LDFD f59 = [C4 ], SIZE (p5) LDFD f60 = [C12], SIZE FMA f6 = ALPHA_R, f64, f6 } { .mmf cmp.ne p6, p0 = 1, I nop __LINE__ FMA f7 = ALPHA_R, f66, f7 } ;; { .mmf (p5) LDFD f61 = [C4 ], SIZE (p5) LDFD f62 = [C12], SIZE FMA f10 = ALPHA_I, f64, f10 } { .mmf nop __LINE__ nop __LINE__ FMA f11 = ALPHA_I, f66, f11 } ;; { .mmf (p5) LDFD f63 = [C4 ], 5 * SIZE (p5) LDFD f47 = [C12], 5 * SIZE FMA f12 = ALPHA_R, f65, f12 } { .mmf nop __LINE__ nop __LINE__ FMA f13 = ALPHA_R, f67, f13 } ;; { .mfi (p5) LDFD f64 = [C4 ], SIZE FMA f14 = ALPHA_I, f65, f14 nop __LINE__ } { .mfi (p5) LDFD f65 = [C12], SIZE FMA f15 = ALPHA_I, f67, f15 nop __LINE__ } ;; { .mmf STFD [C1 ] = f6, SIZE STFD [C9 ] = f7, SIZE FMA f16 = ALPHA_R, f68, f16 } { .mmf (p5) LDFD f6 = [C4 ], SIZE (p5) LDFD f7 = [C12], SIZE FMA f17 = ALPHA_R, f70, f17 } ;; { .mmf STFD [C1 ] = f10, SIZE STFD [C9 ] = f11, SIZE FMA f18 = ALPHA_I, f68, f18 } { .mmf (p5) LDFD f10 = [C4 ], SIZE (p5) LDFD f11 = [C12], SIZE FMA f19 = ALPHA_I, f70, f19 } ;; { .mmf STFD [C1 ] = f12, SIZE STFD [C9 ] = f13, SIZE FMA f20 = ALPHA_R, f69, f20 } { .mmf (p5) LDFD f12 = [C4 ], - 11 * SIZE (p5) LDFD f13 = [C12], - 11 * SIZE FMA f21 = ALPHA_R, f71, f21 } ;; { .mmf STFD [C1 ] = f14, 5 * SIZE STFD [C9 ] = f15, 5 * SIZE FMA f22 = ALPHA_I, f69, f22 } { .mmf (p5) LDFD f14 = [C5 ], SIZE (p5) LDFD f15 = [C13], SIZE FMA f23 = ALPHA_I, f71, f23 } ;; { .mmf STFD [C1 ] = f16, SIZE STFD [C9 ] = f17, SIZE FMA f24 = ALPHA_R, f72, f24 } { .mmf (p5) LDFD f16 = [C5 ], SIZE (p5) LDFD f17 = [C13], SIZE FMA f25 = ALPHA_R, f74, f25 } ;; { .mmf STFD [C1 ] = f18, SIZE STFD [C9 ] = f19, SIZE FMA f26 = ALPHA_I, f72, f26 } { .mmf (p5) LDFD f18 = [C5 ], SIZE (p5) LDFD f19 = [C13], SIZE FMA f27 = ALPHA_I, f74, f27 } ;; { .mmf STFD [C1 ] = f20, SIZE STFD [C9 ] = f21, SIZE FMA f28 = ALPHA_R, f73, f28 } { .mmf (p5) LDFD f20 = [C5 ], 5 * SIZE (p5) LDFD f21 = [C13], 5 * SIZE FMA f29 = ALPHA_R, f75, f29 } ;; { .mmf STFD [C1 ] = f22, 5 * SIZE STFD [C9 ] = f23, 5 * SIZE FMA f30 = ALPHA_I, f73, f30 } { .mmf (p5) LDFD f22 = [C5 ], SIZE (p5) LDFD f23 = [C13], SIZE FMA f31 = ALPHA_I, f75, f31 } ;; { .mmf STFD [C2 ] = f24, SIZE STFD [C10] = f25, SIZE FMA f32 = ALPHA_R, f76, f32 } { .mmf (p5) LDFD f24 = [C5 ], SIZE (p5) LDFD f25 = [C13], SIZE FMA f33 = ALPHA_R, f78, f33 } ;; { .mmf STFD [C2 ] = f26, SIZE STFD [C10] = f27, SIZE FMA f34 = ALPHA_I, f76, f34 } { .mmf (p5) LDFD f26 = [C5 ], SIZE (p5) LDFD f27 = [C13], SIZE FMA f35 = ALPHA_I, f78, f35 } ;; { .mmf STFD [C2 ] = f28, SIZE STFD [C10] = f29, SIZE FMA f36 = ALPHA_R, f77, f36 } { .mmf (p5) LDFD f28 = [C5 ], - 11 * SIZE (p5) LDFD f29 = [C13], - 11 * SIZE FMA f37 = ALPHA_R, f79, f37 } ;; { .mmf STFD [C2 ] = f30, 5 * SIZE STFD [C10] = f31, 5 * SIZE FMA f38 = ALPHA_I, f77, f38 } { .mmf (p5) LDFD f30 = [C6 ], SIZE (p5) LDFD f31 = [C14], SIZE FMA f39 = ALPHA_I, f79, f39 } ;; { .mmf STFD [C2 ] = f32, SIZE STFD [C10] = f33, SIZE FMA f48 = ALPHA_R, f80, f48 } { .mmf (p5) LDFD f32 = [C6 ], SIZE (p5) LDFD f33 = [C14], SIZE FMA f49 = ALPHA_R, f82, f49 } ;; { .mmf STFD [C2 ] = f34, SIZE STFD [C10] = f35, SIZE FMA f50 = ALPHA_I, f80, f50 } { .mmf (p5) LDFD f34 = [C6 ], SIZE (p5) LDFD f35 = [C14], SIZE FMA f51 = ALPHA_I, f82, f51 } ;; { .mmf STFD [C2 ] = f36, SIZE STFD [C10] = f37, SIZE FMA f52 = ALPHA_R, f81, f52 } { .mmf (p5) LDFD f36 = [C6 ], 5 * SIZE (p5) LDFD f37 = [C14], 5 * SIZE FMA f53 = ALPHA_R, f83, f53 } ;; { .mmf STFD [C2 ] = f38, 5 * SIZE STFD [C10] = f39, 5 * SIZE FMA f54 = ALPHA_I, f81, f54 } { .mmf (p5) LDFD f38 = [C6 ], SIZE (p5) LDFD f39 = [C14], SIZE FMA f55 = ALPHA_I, f83, f55 } ;; { .mmf STFD [C3 ] = f48, SIZE STFD [C11] = f49, SIZE FMA f40 = ALPHA_R, f84, f40 } { .mmf (p5) LDFD f48 = [C6 ], SIZE (p5) LDFD f49 = [C14], SIZE FMA f41 = ALPHA_R, f86, f41 } ;; { .mmf STFD [C3 ] = f50, SIZE STFD [C11] = f51, SIZE FMA f42 = ALPHA_I, f84, f42 } { .mmf (p5) LDFD f50 = [C6 ], SIZE (p5) LDFD f51 = [C14], SIZE FMA f43 = ALPHA_I, f86, f43 } ;; { .mmf STFD [C3 ] = f52, SIZE STFD [C11] = f53, SIZE FMA f44 = ALPHA_R, f85, f44 } { .mmf (p5) LDFD f52 = [C6 ], - 11 * SIZE (p5) LDFD f53 = [C14], - 11 * SIZE FMA f45 = ALPHA_R, f87, f45 } ;; { .mmf STFD [C3 ] = f54, 5 * SIZE STFD [C11] = f55, 5 * SIZE FMA f46 = ALPHA_I, f85, f46 } { .mmf (p5) LDFD f54 = [C7 ], SIZE (p5) LDFD f55 = [C15], SIZE FMA f56 = ALPHA_I, f87, f56 } ;; { .mmf STFD [C3 ] = f40, SIZE STFD [C11] = f41, SIZE FMA f57 = ALPHA_R, f88, f57 } { .mmf (p5) LDFD f40 = [C7 ], SIZE (p5) LDFD f41 = [C15], SIZE FMA f58 = ALPHA_R, f90, f58 } ;; { .mmf STFD [C3 ] = f42, SIZE STFD [C11] = f43, SIZE FMA f59 = ALPHA_I, f88, f59 } { .mmf (p5) LDFD f42 = [C7 ], SIZE (p5) LDFD f43 = [C15], SIZE FMA f60 = ALPHA_I, f90, f60 } ;; { .mmf STFD [C3 ] = f44, SIZE STFD [C11] = f45, SIZE FMA f61 = ALPHA_R, f89, f61 } { .mmf (p5) LDFD f44 = [C7 ], 5 * SIZE (p5) LDFD f45 = [C15], 5 * SIZE FMA f62 = ALPHA_R, f91, f62 } ;; { .mmf STFD [C3 ] = f46, 5 * SIZE STFD [C11] = f56, 5 * SIZE FMA f63 = ALPHA_I, f89, f63 } { .mmf (p5) LDFD f46 = [C7 ], SIZE (p5) LDFD f56 = [C15], SIZE FMA f47 = ALPHA_I, f91, f47 } ;; { .mmf STFD [C4 ] = f57, SIZE STFD [C12] = f58, SIZE FMA f64 = ALPHA_R, f92, f64 } { .mmf (p5) LDFD f57 = [C7 ], SIZE (p5) LDFD f58 = [C15], SIZE FMA f65 = ALPHA_R, f94, f65 } ;; { .mmf STFD [C4 ] = f59, SIZE STFD [C12] = f60, SIZE FMA f6 = ALPHA_I, f92, f6 } { .mmf (p5) LDFD f59 = [C7 ], SIZE (p5) LDFD f60 = [C15], SIZE FMA f7 = ALPHA_I, f94, f7 } ;; { .mmf STFD [C4 ] = f61, SIZE STFD [C12] = f62, SIZE FMA f10 = ALPHA_R, f93, f10 } { .mmf (p5) LDFD f61 = [C7 ], - 11 * SIZE (p5) LDFD f62 = [C15], - 11 * SIZE FMA f11 = ALPHA_R, f95, f11 } ;; { .mmf STFD [C4 ] = f63, 5 * SIZE STFD [C12] = f47, 5 * SIZE FMA f12 = ALPHA_I, f93, f12 } { .mmf (p5) LDFD f63 = [C8 ], SIZE (p5) LDFD f47 = [C16], SIZE FMA f13 = ALPHA_I, f95, f13 } ;; { .mmf STFD [C4 ] = f64, SIZE STFD [C12] = f65, SIZE FMA f14 = ALPHA_R, f96, f14 } { .mmf (p5) LDFD f64 = [C8 ], SIZE (p5) LDFD f65 = [C16], SIZE FMA f15 = ALPHA_R, f98, f15 } ;; { .mmf STFD [C4 ] = f6, SIZE STFD [C12] = f7, SIZE FMA f16 = ALPHA_I, f96, f16 } { .mmf (p5) LDFD f6 = [C8 ], SIZE (p5) LDFD f7 = [C16], SIZE FMA f17 = ALPHA_I, f98, f17 } ;; { .mmf STFD [C4 ] = f10, SIZE STFD [C12] = f11, SIZE FMA f18 = ALPHA_R, f97, f18 } { .mmf (p5) LDFD f10 = [C8 ], 5 * SIZE (p5) LDFD f11 = [C16], 5 * SIZE FMA f19 = ALPHA_R, f99, f19 } ;; { .mmf STFD [C4 ] = f12, 5 * SIZE STFD [C12] = f13, 5 * SIZE FMA f20 = ALPHA_I, f97, f20 } { .mmf (p5) LDFD f12 = [C8 ], SIZE (p5) LDFD f13 = [C16], SIZE FMA f21 = ALPHA_I, f99, f21 } ;; { .mmf STFD [C5 ] = f14, SIZE STFD [C13] = f15, SIZE FMA f22 = ALPHA_R, f100, f22 } { .mmf (p5) LDFD f14 = [C8 ], SIZE (p5) LDFD f15 = [C16], SIZE FMA f23 = ALPHA_R, f102, f23 } ;; { .mmf STFD [C5 ] = f16, SIZE STFD [C13] = f17, SIZE FMA f24 = ALPHA_I, f100, f24 } { .mmf (p5) LDFD f16 = [C8 ], SIZE (p5) LDFD f17 = [C16], SIZE FMA f25 = ALPHA_I, f102, f25 } ;; { .mmf STFD [C5 ] = f18, SIZE STFD [C13] = f19, SIZE FMA f26 = ALPHA_R, f101, f26 } { .mmf (p5) LDFD f18 = [C8 ], - 11 * SIZE (p5) LDFD f19 = [C16], - 11 * SIZE FMA f27 = ALPHA_R, f103, f27 } ;; { .mmf STFD [C5 ] = f20, 5 * SIZE STFD [C13] = f21, 5 * SIZE FMA f28 = ALPHA_I, f101, f28 } { .mmf nop __LINE__ nop __LINE__ FMA f29 = ALPHA_I, f103, f29 } ;; { .mmf STFD [C5 ] = f22, SIZE STFD [C13] = f23, SIZE FMA f30 = ALPHA_R, f104, f30 } { .mmf nop __LINE__ nop __LINE__ FMA f31 = ALPHA_R, f106, f31 } ;; { .mmf STFD [C5 ] = f24, SIZE STFD [C13] = f25, SIZE FMA f32 = ALPHA_I, f104, f32 } { .mmf nop __LINE__ nop __LINE__ FMA f33 = ALPHA_I, f106, f33 } ;; { .mmf STFD [C5 ] = f26, SIZE STFD [C13] = f27, SIZE FMA f34 = ALPHA_R, f105, f34 } { .mmf nop __LINE__ nop __LINE__ FMA f35 = ALPHA_R, f107, f35 } ;; { .mmf STFD [C5 ] = f28, 5 * SIZE STFD [C13] = f29, 5 * SIZE FMA f36 = ALPHA_I, f105, f36 } { .mmf nop __LINE__ nop __LINE__ FMA f37 = ALPHA_I, f107, f37 } ;; { .mmf STFD [C6 ] = f30, SIZE STFD [C14] = f31, SIZE FMA f38 = ALPHA_R, f108, f38 } { .mmf nop __LINE__ nop __LINE__ FMA f39 = ALPHA_R, f110, f39 } ;; { .mmf STFD [C6 ] = f32, SIZE STFD [C14] = f33, SIZE FMA f48 = ALPHA_I, f108, f48 }
⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -