📄 daxpy.s
字号:
;; { .mmi (p13) LDFPD f49, f52 = [X3] (p13) LDFPD f50, f51 = [Y1], 2 * SIZE tbit.nz p14, p0 = N, 1 } { .mmi (p13) LDFD f48 = [X1], 3 * SIZE (p13) adds X3 = 4 * SIZE, X3 nop __LINE__ } ;; { .mmi (p13) LDFD f53 = [X1], 1 * SIZE (p13) LDFPD f54, f55 = [Y1], 2 * SIZE tbit.nz p15, p0 = N, 0 } ;; { .mmi (p14) LDFD f56 = [X1], 2 * SIZE (p14) LDFPD f58, f59 = [Y1], 2 * SIZE nop __LINE__ } { .mmi (p14) LDFD f57 = [X3] nop __LINE__ nop __LINE__ } ;; { .mmi (p15) LDFD f60 = [X1] (p15) LDFD f61 = [Y1] nop __LINE__ } ;; (p12) FMA f6 = ALPHA, f32, f34 (p12) FMA f7 = ALPHA, f36, f38 (p12) FMA f10 = ALPHA, f33, f35 (p12) FMA f11 = ALPHA, f37, f39 (p12) FMA f12 = ALPHA, f40, f42 (p12) FMA f13 = ALPHA, f44, f46 (p12) FMA f14 = ALPHA, f41, f43 (p12) FMA f15 = ALPHA, f45, f47 ;; { .mmf (p12) STFD [YY1] = f6, 1 * SIZE (p12) STFD [YY2] = f7, 1 * SIZE (p13) FMA f6 = ALPHA, f48, f50 } { .mmf nop __LINE__ nop __LINE__ (p13) FMA f7 = ALPHA, f52, f54 } ;; { .mmf (p12) STFD [YY1] = f10, 3 * SIZE (p12) STFD [YY2] = f11, 3 * SIZE (p13) FMA f10 = ALPHA, f49, f51 } { .mmf nop __LINE__ nop __LINE__ (p13) FMA f11 = ALPHA, f53, f55 } ;; { .mmf (p12) STFD [YY1] = f12, 1 * SIZE (p12) STFD [YY2] = f13, 1 * SIZE (p14) FMA f12 = ALPHA, f56, f58 } ;; { .mmf (p12) STFD [YY1] = f14, 3 * SIZE (p12) STFD [YY2] = f15, 3 * SIZE (p14) FMA f13 = ALPHA, f57, f59 } ;; { .mmf (p13) STFD [YY1] = f6, 1 * SIZE (p13) STFD [YY2] = f7, 1 * SIZE (p15) FMA f14 = ALPHA, f60, f61 } ;; { .mmi (p13) STFD [YY1] = f10, 3 * SIZE (p13) STFD [YY2] = f11, 3 * SIZE } ;; { .mmi (p14) STFD [YY1] = f12, 1 * SIZE ;; (p14) STFD [YY1] = f13, 1 * SIZE nop __LINE__ } ;; { .mmb (p15) STFD [YY1] = f14 nop __LINE__ br.ret.sptk.many b0 } ;; .align 32.L40: { .mmi (p10) STFD [YY] = f9 cmp.eq p7 ,p0 = -1, I mov ar.lc = I } { .mib adds PREX = (PREFETCHSIZE + 2) * SIZE, X1 adds PREY = (PREFETCHSIZE + 8) * SIZE, Y1 (p7) br.cond.dpnt .L45 } ;; .align 32.L42: { .mmf (p18) STFD [YY1] = f6, 1 * SIZE (p18) STFD [YY2] = f7, 1 * SIZE (p18) FMA f6 = ALPHA, f58, f106 } { .mmf (p17) LDFPD f54, f57 = [X1], 2 * SIZE (p16) LDFPD f80, f83 = [Y1], 2 * SIZE (p18) FMA f7 = ALPHA, f64, f112 } ;; { .mmf (p18) STFD [YY1] = f10, 3 * SIZE (p18) STFD [YY2] = f11, 3 * SIZE (p18) FMA f64 = ALPHA, f61, f109 } { .mmf (p17) LDFPD f60, f63 = [X1], 2 * SIZE (p16) LDFPD f86, f89 = [Y1], 2 * SIZE (p18) FMA f11 = ALPHA, f67, f115 } ;; { .mmf (p18) STFD [YY1] = f12, 1 * SIZE (p18) STFD [YY2] = f13, 1 * SIZE (p18) FMA f67 = ALPHA, f70, f118 } { .mmf (p17) LDFPD f66, f69 = [X1], 2 * SIZE (p16) LDFPD f92, f95 = [Y1], 2 * SIZE (p18) FMA f13 = ALPHA, f76, f124 } ;; { .mmf (p18) STFD [YY1] = f14, 3 * SIZE (p18) STFD [YY2] = f15, 3 * SIZE (p18) FMA f9 = ALPHA, f73, f121 } { .mmf (p17) LDFPD f72, f75 = [X1], 2 * SIZE (p16) LDFPD f98, f101 = [Y1], 2 * SIZE (p18) FMA f15 = ALPHA, f79, f127 } ;; { .mmi (p18) STFD [YY1] = f6, 1 * SIZE (p18) STFD [YY2] = f7, 1 * SIZE (p17) adds X3 = 2 * SIZE, X1 } { .mmf (p16) LDFPD f104, f107 = [Y1], 2 * SIZE (p17) LDFD f78 = [X1], 1 * SIZE (p17) FMA f6 = ALPHA, f33, f81 } ;; { .mmf (p16) LDFPD f110, f113 = [Y1], 2 * SIZE (p16) lfetch.fault.nt1 [PREX], 16 * SIZE (p17) FMA f7 = ALPHA, f39, f87 } { .mmf (p16) LDFD f32 = [X1], 1 * SIZE (p17) FMA f10 = ALPHA, f36, f84 } ;; { .mmf (p18) STFD [YY1] = f64, 3 * SIZE (p18) STFD [YY2] = f11, 3 * SIZE (p17) FMA f11 = ALPHA, f42, f90 } { .mmf (p16) LDFPD f35, f38 = [X1], 2 * SIZE (p16) LDFPD f116, f119 = [Y1], 2 * SIZE (p17) FMA f12 = ALPHA, f45, f93 } ;; { .mmf (p18) STFD [YY1] = f67, 1 * SIZE (p18) STFD [YY2] = f13, 1 * SIZE (p17) FMA f13 = ALPHA, f51, f99 } { .mmf (p16) LDFPD f41, f44 = [X1], 2 * SIZE (p16) LDFPD f122, f125 = [Y1], 2 * SIZE (p17) FMA f14 = ALPHA, f48, f96 } ;; { .mmf (p18) STFD [YY1] = f9, 3 * SIZE (p18) STFD [YY2] = f15, 3 * SIZE (p17) FMA f15 = ALPHA, f54, f102 } { .mmb (p16) lfetch.fault.excl.nt1 [PREY], 16 * SIZE (p16) LDFPD f47, f50 = [X1], 2 * SIZE br.ctop.sptk.few .L42 } ;; .align 32.L45: { .mmi (p12) LDFPD f33, f36 = [X3] (p12) LDFPD f34, f35 = [Y1], 2 * SIZE mov pr = PR, -65474 } { .mmi (p12) LDFD f32 = [X1], 3 * SIZE (p12) adds X3 = 8 * SIZE, X3 nop __LINE__ } ;; { .mmi (p12) LDFPD f37, f40 = [X1], 2 * SIZE (p12) LDFPD f38, f39 = [Y1], 2 * SIZE mov ar.lc = ARLC } ;; { .mmb (p12) LDFPD f41, f44 = [X1], 2 * SIZE (p12) LDFPD f42, f43 = [Y1], 2 * SIZE (p11) br.ret.dpnt.many b0 } ;; { .mmi (p12) LDFD f45 = [X1], 1 * SIZE (p12) LDFPD f46, f47 = [Y1], 2 * SIZE tbit.nz p13, p0 = N, 2 } ;; { .mmi (p13) LDFPD f49, f52 = [X3] (p13) LDFPD f50, f51 = [Y1], 2 * SIZE tbit.nz p14, p0 = N, 1 } { .mmi (p13) LDFD f48 = [X1], 3 * SIZE (p13) adds X3 = 4 * SIZE, X3 nop __LINE__ } ;; { .mmi (p13) LDFD f53 = [X1], 1 * SIZE (p13) LDFPD f54, f55 = [Y1], 2 * SIZE tbit.nz p15, p0 = N, 0 } ;; { .mmi (p14) LDFD f56 = [X1], 2 * SIZE (p14) LDFPD f58, f59 = [Y1], 2 * SIZE nop __LINE__ } { .mmi (p14) LDFD f57 = [X3] nop __LINE__ nop __LINE__ } ;; { .mmi (p15) LDFD f60 = [X1] (p15) LDFD f61 = [Y1] nop __LINE__ } ;; (p12) FMA f6 = ALPHA, f32, f34 (p12) FMA f7 = ALPHA, f36, f38 (p12) FMA f10 = ALPHA, f33, f35 (p12) FMA f11 = ALPHA, f37, f39 (p12) FMA f12 = ALPHA, f40, f42 (p12) FMA f13 = ALPHA, f44, f46 (p12) FMA f14 = ALPHA, f41, f43 (p12) FMA f15 = ALPHA, f45, f47 ;; { .mmf (p12) STFD [YY1] = f6, 1 * SIZE (p12) STFD [YY2] = f7, 1 * SIZE (p13) FMA f6 = ALPHA, f48, f50 } { .mmf nop __LINE__ nop __LINE__ (p13) FMA f7 = ALPHA, f52, f54 } ;; { .mmf (p12) STFD [YY1] = f10, 3 * SIZE (p12) STFD [YY2] = f11, 3 * SIZE (p13) FMA f10 = ALPHA, f49, f51 } { .mmf nop __LINE__ nop __LINE__ (p13) FMA f11 = ALPHA, f53, f55 } ;; { .mmf (p12) STFD [YY1] = f12, 1 * SIZE (p12) STFD [YY2] = f13, 1 * SIZE (p14) FMA f12 = ALPHA, f56, f58 } ;; { .mmf (p12) STFD [YY1] = f14, 3 * SIZE (p12) STFD [YY2] = f15, 3 * SIZE (p14) FMA f13 = ALPHA, f57, f59 } ;; { .mmf (p13) STFD [YY1] = f6, 1 * SIZE (p13) STFD [YY2] = f7, 1 * SIZE (p15) FMA f14 = ALPHA, f60, f61 } ;; { .mmi (p13) STFD [YY1] = f10, 3 * SIZE (p13) STFD [YY2] = f11, 3 * SIZE } ;; { .mmi (p14) STFD [YY1] = f12, 1 * SIZE ;; (p14) STFD [YY1] = f13, 1 * SIZE nop __LINE__ } ;; { .mmb (p15) STFD [YY1] = f14 nop __LINE__ br.ret.sptk.many b0 } ;; .align 32.L100: { .mfi cmp.eq p16, p0 = r0, r0 (p10) FMA f9 = ALPHA, f32, f33 shr I = N, 4 } ;; { .mmi cmp.eq p11, p0 = r0, J adds I = -1, I mov ar.ec = 3 } { .mmi shladd INCX16 = INCX, 4, r0 shladd INCY16 = INCY, 4, r0 tbit.nz p12, p0 = N, 3 } ;; { .mmi (p10) STFD [YY] = f9 cmp.eq p7 ,p0 = -1, I mov ar.lc = I } { .mib adds PREX = (PREFETCHSIZE + 2) * SIZE, X1 adds PREY = (PREFETCHSIZE + 2) * SIZE, Y1 (p7) br.cond.dpnt .L115 } ;; .align 32.L112: { .mmf (p18) STFD [YY1] = f6 (p16) lfetch.fault.nt1 [PREX], INCX16 (p18) FMA f12 = ALPHA, f46, f94 } { .mmi (p16) LDFD f32 = [X1], INCX (p16) LDFD f80 = [Y1], INCY (p18) add YY1 = YY1, INCY } ;; { .mmf (p18) STFD [YY1] = f7 (p18) add YY1 = YY1, INCY (p18) FMA f13 = ALPHA, f49, f97 } { .mmi (p16) LDFD f35 = [X1], INCX (p16) LDFD f83 = [Y1], INCY nop __LINE__ } ;; { .mmf (p18) STFD [YY1] = f10 (p18) add YY1 = YY1, INCY (p18) FMA f14 = ALPHA, f52, f100 } { .mmi (p16) LDFD f38 = [X1], INCX (p16) LDFD f86 = [Y1], INCY nop __LINE__ } ;; { .mmf (p18) STFD [YY1] = f11 (p18) add YY1 = YY1, INCY (p18) FMA f15 = ALPHA, f55, f103 } { .mmi (p16) LDFD f41 = [X1], INCX (p16) LDFD f89 = [Y1], INCY nop __LINE__ } ;; { .mmf (p18) STFD [YY1] = f12 (p18) add YY1 = YY1, INCY (p18) FMA f6 = ALPHA, f58, f106 } { .mmi (p16) LDFD f44 = [X1], INCX (p16) LDFD f92 = [Y1], INCY nop __LINE__ } ;; { .mmf (p18) STFD [YY1] = f13 (p18) add YY1 = YY1, INCY (p18) FMA f7 = ALPHA, f61, f109 } { .mmi (p16) LDFD f47 = [X1], INCX (p16) LDFD f95 = [Y1], INCY nop __LINE__ } ;; { .mmf (p18) STFD [YY1] = f14 (p18) add YY1 = YY1, INCY (p18) FMA f10 = ALPHA, f64, f112 } { .mmi (p16) LDFD f50 = [X1], INCX (p16) LDFD f98 = [Y1], INCY nop __LINE__ } ;; { .mmf (p18) STFD [YY1] = f15 (p18) add YY1 = YY1, INCY (p18) FMA f11 = ALPHA, f67, f115 } { .mmi (p16) LDFD f53 = [X1], INCX (p16) LDFD f101 = [Y1], INCY nop __LINE__ } ;; { .mmf (p18) STFD [YY1] = f6 (p16) lfetch.fault.excl.nt1 [PREY], INCY16 (p18) FMA f12 = ALPHA, f70, f118 } { .mmi (p16) LDFD f56 = [X1], INCX (p16) LDFD f104 = [Y1], INCY (p18) add YY1 = YY1, INCY } ;; { .mmf (p18) STFD [YY1] = f7 (p18) add YY1 = YY1, INCY (p18) FMA f13 = ALPHA, f73, f121 } { .mmi (p16) LDFD f59 = [X1], INCX (p16) LDFD f107 = [Y1], INCY nop __LINE__ } ;; { .mmf (p18) STFD [YY1] = f10 (p18) add YY1 = YY1, INCY (p18) FMA f14 = ALPHA, f76, f124 } { .mmi (p16) LDFD f62 = [X1], INCX (p16) LDFD f110 = [Y1], INCY nop __LINE__ } ;; { .mmf (p18) STFD [YY1] = f11 (p18) add YY1 = YY1, INCY (p18) FMA f15 = ALPHA, f79, f127 } { .mmi (p16) LDFD f65 = [X1], INCX (p16) LDFD f113 = [Y1], INCY nop __LINE__ } ;; { .mmf (p18) STFD [YY1] = f12 (p18) add YY1 = YY1, INCY (p17) FMA f6 = ALPHA, f33, f81 } { .mmi (p16) LDFD f68 = [X1], INCX (p16) LDFD f116 = [Y1], INCY nop __LINE__ } ;; { .mmf (p18) STFD [YY1] = f13 (p18) add YY1 = YY1, INCY (p17) FMA f7 = ALPHA, f36, f84 } { .mmi (p16) LDFD f71 = [X1], INCX (p16) LDFD f119 = [Y1], INCY nop __LINE__ } ;; { .mmf (p18) STFD [YY1] = f14 (p18) add YY1 = YY1, INCY (p17) FMA f10 = ALPHA, f39, f87 } { .mmi (p16) LDFD f74 = [X1], INCX (p16) LDFD f122 = [Y1], INCY nop __LINE__ } ;; { .mmf (p18) STFD [YY1] = f15 (p18) add YY1 = YY1, INCY (p17) FMA f11 = ALPHA, f42, f90 } { .mmb (p16) LDFD f77 = [X1], INCX (p16) LDFD f125 = [Y1], INCY br.ctop.sptk.few .L112 } ;; .align 32.L115: { .mmi (p12) LDFD f32 = [X1], INCX (p12) LDFD f34 = [Y1], INCY mov pr = PR, -65474 } ;; { .mmi (p12) LDFD f33 = [X1], INCX (p12) LDFD f35 = [Y1], INCY mov ar.lc = ARLC } ;; { .mmb (p12) LDFD f36 = [X1], INCX (p12) LDFD f38 = [Y1], INCY (p11) br.ret.dpnt.many b0 } ;; { .mmi (p12) LDFD f37 = [X1], INCX (p12) LDFD f39 = [Y1], INCY tbit.nz p13, p0 = N, 2 } ;; { .mmi (p12) LDFD f40 = [X1], INCX (p12) LDFD f42 = [Y1], INCY tbit.nz p14, p0 = N, 1 } ;; { .mmi (p12) LDFD f41 = [X1], INCX (p12) LDFD f43 = [Y1], INCY tbit.nz p15, p0 = N, 0 } ;; { .mmf (p12) LDFD f44 = [X1], INCX (p12) LDFD f46 = [Y1], INCY (p12) FMA f6 = ALPHA, f32, f34 } ;; { .mmf (p12) LDFD f45 = [X1], INCX (p12) LDFD f47 = [Y1], INCY (p12) FMA f7 = ALPHA, f33, f35 } ;; { .mmf (p13) LDFD f48 = [X1], INCX (p13) LDFD f50 = [Y1], INCY (p12) FMA f10 = ALPHA, f36, f38 } ;; { .mmf (p13) LDFD f49 = [X1], INCX (p13) LDFD f51 = [Y1], INCY (p12) FMA f11 = ALPHA, f37, f39 } ;; { .mmf (p12) STFD [YY1] = f6 (p12) add YY1 = YY1, INCY (p12) FMA f12 = ALPHA, f40, f42 } { .mmi (p13) LDFD f52 = [X1], INCX (p13) LDFD f54 = [Y1], INCY nop __LINE__ } ;; { .mmf (p12) STFD [YY1] = f7 (p12) add YY1 = YY1, INCY (p12) FMA f13 = ALPHA, f41, f43 } { .mmi (p13) LDFD f53 = [X1], INCX (p13) LDFD f55 = [Y1], INCY nop __LINE__ } ;; { .mmf (p12) STFD [YY1] = f10 (p12) add YY1 = YY1, INCY (p12) FMA f14 = ALPHA, f44, f46 } { .mmi (p14) LDFD f56 = [X1], INCX (p14) LDFD f58 = [Y1], INCY nop __LINE__ } ;; { .mmf (p12) STFD [YY1] = f11 (p12) add YY1 = YY1, INCY (p12) FMA f15 = ALPHA, f45, f47 } { .mmi (p14) LDFD f57 = [X1], INCX (p14) LDFD f59 = [Y1], INCY nop __LINE__ } ;; { .mmf (p12) STFD [YY1] = f12 (p12) add YY1 = YY1, INCY (p13) FMA f6 = ALPHA, f48, f50 } { .mmi (p15) LDFD f60 = [X1], INCX (p15) LDFD f61 = [Y1], INCY nop __LINE__ } ;; { .mmf (p12) STFD [YY1] = f13 (p12) add YY1 = YY1, INCY (p13) FMA f7 = ALPHA, f49, f51 } ;; { .mmf (p12) STFD [YY1] = f14 (p12) add YY1 = YY1, INCY (p13) FMA f10 = ALPHA, f52, f54 } ;; { .mmf (p12) STFD [YY1] = f15 (p12) add YY1 = YY1, INCY (p13) FMA f11 = ALPHA, f53, f55 } ;; ;; { .mmf (p13) STFD [YY1] = f6 (p13) add YY1 = YY1, INCY (p14) FMA f12 = ALPHA, f56, f58 } ;; { .mmf (p13) STFD [YY1] = f7 (p13) add YY1 = YY1, INCY (p14) FMA f13 = ALPHA, f57, f59 } ;; { .mmf (p13) STFD [YY1] = f10 (p13) add YY1 = YY1, INCY (p15) FMA f14 = ALPHA, f60, f61 } ;; { .mmi (p13) STFD [YY1] = f11 (p13) add YY1 = YY1, INCY nop __LINE__ } ;; { .mmi (p14) STFD [YY1] = f12 (p14) add YY1 = YY1, INCY nop __LINE__ } ;; { .mmi (p14) STFD [YY1] = f13 (p14) add YY1 = YY1, INCY nop __LINE__ } ;; { .mmb (p15) STFD [YY1] = f14 nop __LINE__ br.ret.sptk.many b0 } ;; EPILOGUE
⌨️ 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 + -