⭐ 欢迎来到虫虫下载站! | 📦 资源下载 📁 资源专辑 ℹ️ 关于我们
⭐ 虫虫下载站

📄 trsm_kernel_lt.s

📁 Optimized GotoBLAS libraries
💻 S
📖 第 1 页 / 共 5 页
字号:
	adds	C9  = 4 * SIZE, C1	}	;;	{ .mfi	STFD	[AOFFSET] = f123	adds	AOFFSET = - 59 * SIZE, AOFFSET	}	{ .mfi	STFD	[AOFFSET2] = f127	adds	AOFFSET2 = - 59 * SIZE, AOFFSET2	}	;;#endif	{ .mmf	STFD	[C1 ] = f64, SIZE	STFD	[C9 ] = f68, SIZE	mov	f64  = f0	}	;;	{ .mmi	STFD	[C1 ] = f65, SIZE	STFD	[C9 ] = f69, SIZE	adds	C10 = 4 * SIZE, C2	}	;;	{ .mmi	STFD	[C1 ] = f66, SIZE	STFD	[C9 ] = f70, SIZE	nop	__LINE__	}	;;	{ .mmi	STFD	[C1 ] = f67, 5 * SIZE	STFD	[C9 ] = f71	adds	C11 = 4 * SIZE, C3	}	;;	{ .mmf	STFD	[C2 ] = f72, SIZE	STFD	[C10] = f76, SIZE	mov	f72  = f0	}	;;	{ .mmi	STFD	[C2 ] = f73, SIZE	STFD	[C10] = f77, SIZE	nop	__LINE__	}	;;	{ .mmi	STFD	[C2 ] = f74, SIZE	STFD	[C10] = f78, SIZE	adds	C12 = 4 * SIZE, C4	}	;;	{ .mmi	STFD	[C2 ] = f75, 5 * SIZE	STFD	[C10] = f79	nop	__LINE__	}	;;	{ .mmf	STFD	[C3 ] = f80, SIZE	STFD	[C11] = f84, SIZE	mov	f80  = f0	}	;;	{ .mmi	STFD	[C3 ] = f81, SIZE	STFD	[C11] = f85, SIZE	adds	C13 = 4 * SIZE, C5	}	;;	{ .mmi	STFD	[C3 ] = f82, SIZE	STFD	[C11] = f86, SIZE	nop	__LINE__	}	;;	{ .mmi	STFD	[C3 ] = f83, 5 * SIZE	STFD	[C11] = f87	adds	C14 = 4 * SIZE, C6	}	;;	{ .mmf	STFD	[C4 ] = f88, SIZE	STFD	[C12] = f92, SIZE	mov	f88  = f0	}	;;	{ .mmi	STFD	[C4 ] = f89, SIZE	STFD	[C12] = f93, SIZE	nop	__LINE__	}	;;	{ .mmi	STFD	[C4 ] = f90, SIZE	STFD	[C12] = f94, SIZE	adds	C16 = 4 * SIZE, C8	}	;;	{ .mmi	STFD	[C4 ] = f91, 5 * SIZE	STFD	[C12] = f95	cmp.ne	p6, p0 = 1, I	}	;;	{ .mmf	STFD	[C5 ] = f96,  SIZE	STFD	[C13] = f100, SIZE	mov	f96  = f0	}	;;	{ .mmi	STFD	[C5 ] = f97,  SIZE	STFD	[C13] = f101, SIZE	adds	I = -1, I	}	;;	{ .mmi	STFD	[C5 ] = f98,  SIZE	STFD	[C13] = f102, SIZE	nop	__LINE__	}	;;	{ .mmi	STFD	[C5 ] = f99,  5 * SIZE	STFD	[C13] = f103	adds	C15 = 4 * SIZE, C7	}	;;	{ .mmf	STFD	[C6 ] = f104, SIZE	STFD	[C14] = f108, SIZE	mov	f104 = f0	}	;;	{ .mmi	STFD	[C6 ] = f105, SIZE	STFD	[C14] = f109, SIZE	nop	__LINE__	}	;;	{ .mmi	STFD	[C6 ] = f106, SIZE	STFD	[C14] = f110, SIZE	sub	L = K, KK	}	;;	{ .mmi	STFD	[C6 ] = f107, 5 * SIZE	STFD	[C14] = f111	nop	__LINE__	}	;;	{ .mmf	STFD	[C7 ] = f112, SIZE	STFD	[C15] = f116, SIZE	mov	f112 = f0	}	;;	{ .mmi	STFD	[C7 ] = f113, SIZE	STFD	[C15] = f117, SIZE	shladd	L = L, BASE_SHIFT, r0	}	;;	{ .mmi	STFD	[C7 ] = f114, SIZE	STFD	[C15] = f118, SIZE	shladd	AOFFSET = L, 3, AOFFSET	}	;;	{ .mmi	STFD	[C7 ] = f115, 5 * SIZE	STFD	[C15] = f119	shladd	BOFFSET = L, 3, BOFFSET	}	;;	{ .mmf	STFD	[C8 ] = f120, SIZE	STFD	[C16] = f124, SIZE	mov	f120 = f0	}	;;	{ .mmi	STFD	[C8 ] = f121, SIZE	STFD	[C16] = f125, SIZE#ifdef LT	adds	KK =  8, KK#else	nop	__LINE__#endif	}	;;	{ .mmi	STFD	[C8 ] = f122, SIZE	STFD	[C16] = f126, SIZE	mov	L = KK	}	;;	{ .mmb	STFD	[C8 ] = f123, 5 * SIZE	STFD	[C16] = f127	(p6)	br.cond.dptk .L011	}	;;.L020:	{ .mib	mov	L = KK	tbit.z	p6, p0 = M, 2	(p6)	br.cond.dptk .L030	}	;;	{ .mmi	cmp.ne	p7, p0 = r0, L	adds	BOFFSET = 0 * SIZE, B	nop	__LINE__	}	;;	{ .mmf	(p7) LDFPD	f48, f49 = [BOFFSET], 2 * SIZE	setf.d	f73  = r0	mov	f65  = f0	}	;;	{ .mfi	setf.d	f105 = r0	mov	f81  = f0	adds	L =  1, L	}	{ .mfi	adds	PREA = (PREFETCHSIZE + 8) * SIZE, AOFFSET	mov	f89  = f0	cmp.eq	p3, p0 = r0, r0	}	;;	{ .mfi	(p7) LDFPD	f50, f51 = [BOFFSET], 2 * SIZE	mov	f113 = f0	tbit.z	p12, p0 = L, 0	}	{ .mfi	setf.d	f97  = r0	mov	f121 = f0	shr	L = L, 1	}	;;	{ .mmf	(p7) LDFPD	f52, f53 = [BOFFSET], 2 * SIZE	setf.d	f66  = r0	mov	f67  = f0	}	{ .mfi	setf.d	f74  = r0	mov	f75  = f0	adds	L =  -1, L	}	;;	{ .mmf	(p7) LDFPD	f54, f55 = [BOFFSET], 2 * SIZE	setf.d	f82  = r0	mov	f83  = f0	}	{ .mfi	setf.d	f90  = r0	mov	f91  = f0	cmp.eq  p6, p0 = -1, L	}	;;	{ .mmf	(p7) LDFPD	f32, f33 = [AOFFSET], 2 * SIZE	setf.d	f98  = r0	mov	f99  = f0	}	{ .mfi	setf.d	f106 = r0	mov	f107 = f0	mov	ar.lc = L	}	;;	{ .mmf	(p7) LDFPD	f34, f35  = [AOFFSET], 2 * SIZE	setf.d	f114 = r0	mov	f115 = f0	}	{ .mfb	setf.d	f122 = r0	mov	f123 = f0	(p6) br.cond.dpnt   .L028	}	;;.L022:	{ .mfi	lfetch.nt1	[PREA],  8 * SIZE	FMA	f64   = f32, f48, f64	// A1 * B1	adds	PREB = (PREFETCHSIZE + 0) * SIZE, BOFFSET	}	{ .mfi	nop	__LINE__	FMA	f72   = f32, f49, f72	// A1 * B2	(p12) cmp.ne p3, p0 =  0, L	}	;;	{ .mfi	lfetch.nt1	[PREB],  16 * SIZE	FMA	f80   = f32, f50, f80	// A1 * B3	cmp.ne	p4, p5 =  0, L	}	{ .mfb	nop	__LINE__	FMA	f88   = f32, f51, f88	// A1 * B4	nop	__LINE__	}	;;	{ .mfi	(p3) LDFPD	f40, f41 = [AOFFSET], 2 * SIZE	FMA	f96   = f32, f52, f96	// A1 * B5	(p5) adds	C9  = 2 * SIZE, C1	}	{ .mfi	nop	__LINE__	FMA	f104  = f32, f53, f104	// A1 * B6	(p5) adds	C10 = 2 * SIZE, C2	}	;;	{ .mfi	(p3) LDFPD	f56, f57 = [BOFFSET],   2 * SIZE	FMA	f112  = f32, f54, f112	// A1 * B7	(p5) adds	C11 = 2 * SIZE, C3	}	{ .mfi	nop	__LINE__	FMA	f120  = f32, f55, f120	// A1 * B8	(p5) adds	C12 = 2 * SIZE, C4	}	;;	{ .mfi	(p3) LDFPD	f58, f59 = [BOFFSET],  2 * SIZE	FMA	f65   = f33, f48, f65	// A2 * B1	(p5) adds	C13 = 2 * SIZE, C5	}	{ .mfi	nop	__LINE__	FMA	f73   = f33, f49, f73	// A2 * B2	(p5) adds	C14 = 2 * SIZE, C6	}	;;	{ .mfi	(p3) LDFPD	f60, f61 = [BOFFSET], 2 * SIZE	FMA	f81   = f33, f50, f81	// A2 * B3	(p5) adds	C15 = 2 * SIZE, C7	}	{ .mfi	nop	__LINE__	FMA	f89   = f33, f51, f89	// A2 * B4	(p5) adds	C16 = 2 * SIZE, C8	}	;;	{ .mfb	(p3) LDFPD	f62, f63 = [BOFFSET], 2 * SIZE	FMA	f97   = f33, f52, f97	// A2 * B5	nop	__LINE__	}	{ .mfb	nop	__LINE__	FMA	f105  = f33, f53, f105	// A2 * B6	nop	__LINE__	}	;;	{ .mfb	(p3) LDFPD	f42, f43 = [AOFFSET], 2 * SIZE	FMA	f113  = f33, f54, f113	// A2 * B7	nop	__LINE__	}	{ .mfb	nop	__LINE__	FMA	f121  = f33, f55, f121	// A2 * B8	nop	__LINE__	}	;;	{ .mfb	nop	__LINE__	FMA	f66   = f34, f48, f66	// A3 * B1	nop	__LINE__	}	{ .mfb	nop	__LINE__	FMA	f74   = f34, f49, f74	// A3 * B2	nop	__LINE__	}	;;	{ .mfb	nop	__LINE__	FMA	f82   = f34, f50, f82	// A3 * B3	nop	__LINE__	}	{ .mfb	nop	__LINE__	FMA	f90   = f34, f51, f90	// A3 * B4	nop	__LINE__	}	;;	{ .mfb	nop	__LINE__	FMA	f98   = f34, f52, f98	// A3 * B5	nop	__LINE__	}	{ .mfb	nop	__LINE__	FMA	f106  = f34, f53, f106	// A3 * B6	nop	__LINE__	}	{ .mfb	nop	__LINE__	FMA	f114  = f34, f54, f114	// A3 * B7	nop	__LINE__	}	{ .mfb	nop	__LINE__	FMA	f122  = f34, f55, f122	// A3 * B8	nop	__LINE__	}	{ .mfb	nop	__LINE__	FMA	f67   = f35, f48, f67	// A4 * B1	nop	__LINE__	}	{ .mfb	nop	__LINE__	FMA	f75   = f35, f49, f75	// A4 * B2	nop	__LINE__	}	{ .mfb	nop	__LINE__	FMA	f83   = f35, f50, f83	// A4 * B3	nop	__LINE__	}	{ .mfb	nop	__LINE__	FMA	f91   = f35, f51, f91	// A4 * B4	nop	__LINE__	}	{ .mfb	(p4) LDFPD	f32, f33 = [AOFFSET],   2 * SIZE	FMA	f99   = f35, f52, f99	// A4 * B5	nop	__LINE__	}	{ .mfb	nop	__LINE__	FMA	f107  = f35, f53, f107	// A4 * B6	nop	__LINE__	}	{ .mfb	(p4) LDFPD	f48, f49 = [BOFFSET],   2 * SIZE	FMA	f115  = f35, f54, f115	// A4 * B7	nop	__LINE__	}	{ .mfb	nop	__LINE__	FMA	f123  = f35, f55, f123	// A4 * B8	nop	__LINE__	}	;;	{ .mfb	(p4) LDFPD	f50, f51 = [BOFFSET],  2 * SIZE	(p3) FMA	f64   = f40, f56, f64	// A1 * B1	nop	__LINE__	}	{ .mfb	nop	__LINE__	(p3) FMA	f72   = f40, f57, f72	// A1 * B2	nop	__LINE__	}	;;	{ .mfb	(p4) LDFPD	f52, f53 = [BOFFSET], 2 * SIZE	(p3) FMA	f80   = f40, f58, f80	// A1 * B3	nop	__LINE__	}	{ .mfb	nop	__LINE__	(p3) FMA	f88   = f40, f59, f88	// A1 * B4	nop	__LINE__	}	;;	{ .mfb	nop	__LINE__	(p3) FMA	f96   = f40, f60, f96	// A1 * B5	nop	__LINE__	}	{ .mfb	nop	__LINE__	(p3) FMA	f104  = f40, f61, f104	// A1 * B6	nop	__LINE__	}	;;	{ .mfb	nop	__LINE__	(p3) FMA	f112  = f40, f62, f112	// A1 * B7	nop	__LINE__	}	{ .mfb	nop	__LINE__	(p3) FMA	f120  = f40, f63, f120	// A1 * B8	nop	__LINE__	}	;;	{ .mfb	(p4) LDFPD	f54, f55 = [BOFFSET], 2 * SIZE	(p3) FMA	f65   = f41, f56, f65	// A2 * B1	nop	__LINE__	}	{ .mfb	(p3) FMA	f73   = f41, f57, f73	// A2 * B2	nop	__LINE__	}	{ .mfb	(p4) LDFPD	f34, f35 = [AOFFSET], 2 * SIZE	(p3) FMA	f81   = f41, f58, f81	// A2 * B3	nop	__LINE__	}	{ .mfb	(p3) FMA	f89   = f41, f59, f89	// A2 * B4	nop	__LINE__	}	;;	{ .mfb	nop	__LINE__	(p3) FMA	f97   = f41, f60, f97	// A2 * B5	nop	__LINE__	}	{ .mfb	nop	__LINE__	(p3) FMA	f105  = f41, f61, f105	// A2 * B6	nop	__LINE__	}	;;	{ .mfb	nop	__LINE__	(p3) FMA	f113  = f41, f62, f113	// A2 * B7	nop	__LINE__	}	{ .mfb	nop	__LINE__	(p3) FMA	f121  = f41, f63, f121	// A2 * B8	nop	__LINE__	}	;;	{ .mfb	nop	__LINE__	(p3) FMA	f66   = f42, f56, f66	// A3 * B1	nop	__LINE__	}	{ .mfb	nop	__LINE__	(p3) FMA	f74   = f42, f57, f74	// A3 * B2	nop	__LINE__	}	;;	{ .mfb	nop	__LINE__	(p3) FMA	f82   = f42, f58, f82	// A3 * B3	nop	__LINE__	}	{ .mfb	nop	__LINE__	(p3) FMA	f90   = f42, f59, f90	// A3 * B4	nop	__LINE__	}	;;	{ .mfb	nop	__LINE__	(p3) FMA	f98   = f42, f60, f98	// A3 * B5	nop	__LINE__	}	{ .mfb	nop	__LINE__	(p3) FMA	f106  = f42, f61, f106	// A3 * B6	nop	__LINE__	}	;;	{ .mfb	nop	__LINE__	(p3) FMA	f114  = f42, f62, f114	// A3 * B7	nop	__LINE__	}	{ .mfb	nop	__LINE__	(p3) FMA	f122  = f42, f63, f122	// A3 * B8	nop	__LINE__	}	;;	{ .mfb	nop	__LINE__	(p3) FMA	f67   = f43, f56, f67	// A4 * B1	nop	__LINE__	}	{ .mfb	nop	__LINE__	(p3) FMA	f75   = f43, f57, f75	// A4 * B2	nop	__LINE__	}	;;	{ .mfb	nop	__LINE__	(p3) FMA	f83   = f43, f58, f83	// A4 * B3	nop	__LINE__	}	{ .mfb	nop	__LINE__	(p3) FMA	f91   = f43, f59, f91	// A4 * B4	nop	__LINE__	}	;;	{ .mfb	nop	__LINE__	(p3) FMA	f99   = f43, f60, f99	// A4 * B5	nop	__LINE__	}	{ .mfb	nop	__LINE__	(p3) FMA	f107  = f43, f61, f107	// A4 * B6	nop	__LINE__	}	;;	{ .mfi	nop	__LINE__	(p3) FMA	f115  = f43, f62, f115	// A4 * B7	adds	L = -1, L	}	{ .mfb	nop	__LINE__	(p3) FMA	f123  = f43, f63, f123	// A4 * B8	br.cloop.sptk.few .L022	}	;;.L028:#ifdef LT	LDFPD	f32, f33 = [BOFFSET], 2 * SIZE	;;	LDFPD	f34, f35 = [BOFFSET], 2 * SIZE	;;	LDFPD	f36, f37 = [BOFFSET], 2 * SIZE	;;	LDFPD	f38, f39 = [BOFFSET], 2 * SIZE	;;	LDFPD	f40, f41 = [BOFFSET], 2 * SIZE	;;	LDFPD	f42, f43 = [BOFFSET], 2 * SIZE	;;	LDFPD	f44, f45 = [BOFFSET], 2 * SIZE	;;	LDFPD	f46, f47 = [BOFFSET], 2 * SIZE	;;	{ .mfi	LDFPD	f48, f49 = [BOFFSET], 2 * SIZE	FSUB	f64  = f32, f64	nop	__LINE__	}	{ .mfi	nop	__LINE__	FSUB	f72  = f33, f72	nop	__LINE__	}	;;	{ .mfi	LDFPD	f50, f51 = [BOFFSET], 2 * SIZE	FSUB	f80  = f34, f80	nop	__LINE__	}	{ .mfi	nop	__LINE__	FSUB	f88  = f35, f88	nop	__LINE__	}	;;	{ .mfi	LDFPD	f52, f53 = [BOFFSET], 2 * SIZE	FSUB	f96  = f36, f96	nop	__LINE__	}	{ .mfi	nop	__LINE__	FSUB	f104 = f37, f104	nop	__LINE__	}	;;	{ .mfi	LDFPD	f54, f55 = [BOFFSET], 2 * SIZE	FSUB	f112 = f38, f112	nop	__LINE__	}	{ .mfi	nop	__LINE__	FSUB	f120 = f39, f120	nop	__LINE__	}	;;	{ .mfi	LDFPD	f56, f57 = [BOFFSET], 2 * SIZE	FSUB	f65  = f40, f65	nop	__LINE__	}	{ .mfi	nop	__LINE__	FSUB	f73  = f41, f73	nop	__LINE__	}	;;	{ .mfi	LDFPD	f58, f59 = [BOFFSET], 2 * SIZE	FSUB	f81  = f42, f81	nop	__LINE__	}	{ .mfi	nop	__LINE__	FSUB	f89  = f43, f89	nop	__LINE__	}	;;	{ .mfi	LDFPD	f60, f61 = [BOFFSET], 2 * SIZE	FSUB	f97  = f44, f97	nop	__LINE__	}	{ .mfi	nop	__LINE__	FSUB	f105 = f45, f105	nop	__LINE__	}	;;	{ .mfi	LDFPD	f62, f63 = [BOFFSET]	FSUB	f113 = f46, f113	adds	BOFFSET = -30 * SIZE, BOFFSET	}	{ .mfi	nop	__LINE__	FSUB	f121 = f47, f121	nop	__LINE__	}	;;	FSUB	f66  = f48, f66	FSUB	f74  = f49, f74	FSUB	f82  = f50, f82	FSUB	f90  = f51, f90	FSUB	f98  = f52, f98	FSUB	f106 = f53, f106	FSUB	f114 = f54, f114	FSUB	f122 = f55, f122	;;	FSUB	f67  = f56, f67	FSUB	f75  = f57, f75	FSUB	f83  = f58, f83	FSUB	f91  = f59, f91	FSUB	f99  = f60, f99	FSUB	f107 = f61, f107	FSUB	f115 = f62, f115	FSUB	f123 = f63, f123	;;#else	LDFPD	f32, f33 = [AOFFSET], 2 * SIZE	;;	LDFPD	f34, f35 = [AOFFSET], 2 * SIZE	;;	LDFPD	f36, f37 = [AOFFSET], 2 * SIZE	;;	LDFPD	f38, f39 = [AOFFSET], 2 * SIZE	;;	LDFPD	f40, f41 = [AOFFSET], 2 * SIZE	;;	LDFPD	f42, f43 = [AOFFSET], 2 * SIZE	;;	LDFPD	f44, f45 = [AOFFSET], 2 * SIZE	;;	LDFPD	f46, f47 = [AOFFSET], 2 * SIZE	;;	LDFPD	f48, f49 = [AOFFSET], 2 * SIZE	;;	LDFPD	f50, f51 = [AOFFSET], 2 * SIZE	;;	LDFPD	f52, f53 = [AOFFSET], 2 * SIZE	;;	LDFPD	f54, f55 = [AOFFSET], 2 * SIZE	;;	LDFPD	f56, f57 = [AOFFSET], 2 * SIZE	;;	LDFPD	f58, f59 = [AOFFSET], 2 * SIZE	;;	LDFPD	f60, f61 = [AOFFSET], 2 * SIZE	;;	LDFPD	f62, f63 = [AOFFSET]	adds	AOFFSET = -30

⌨️ 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 + -