📄 copy_sse_core2.s

📁 Optimized GotoBLAS libraries
💻 S
字号:
/*********************************************************************//*                                                                   *//*             Optimized BLAS libraries                              *//*                     By Kazushige Goto <kgoto@tacc.utexas.edu>     *//*                                                                   *//* Copyright (c) The University of Texas, 2005. All rights reserved. *//* UNIVERSITY EXPRESSLY DISCLAIMS ANY AND ALL WARRANTIES CONCERNING  *//* THIS SOFTWARE AND DOCUMENTATION, INCLUDING ANY WARRANTIES OF      *//* MERCHANTABILITY, FITNESS FOR ANY PARTICULAR PURPOSE,              *//* NON-INFRINGEMENT AND WARRANTIES OF PERFORMANCE, AND ANY WARRANTY  *//* THAT MIGHT OTHERWISE ARISE FROM COURSE OF DEALING OR USAGE OF     *//* TRADE. NO WARRANTY IS EITHER EXPRESS OR IMPLIED WITH RESPECT TO   *//* THE USE OF THE SOFTWARE OR DOCUMENTATION.                         *//* Under no circumstances shall University be liable for incidental, *//* special, indirect, direct or consequential damages or loss of     *//* profits, interruption of business, or related expenses which may  *//* arise from use of Software or Documentation, including but not    *//* limited to those resulting from defects in Software and/or        *//* Documentation, or loss or inaccuracy of data of any kind.         *//*********************************************************************/#define ASSEMBLER#include "common.h"#define M	ARG1	/* rdi */#define X	ARG2	/* rsi */#define INCX	ARG3	/* rdx */#define Y	ARG4	/* rcx */#ifndef WINDOWS_ABI#define INCY	ARG5	/* r8  */#define FLAG	ARG6#else#define INCY	%r10#define FLAG	%r11#endif	PROLOGUE	PROFCODE#ifdef WINDOWS_ABI	movq	40(%rsp), INCY#endif	SAVEREGISTERS	leaq	(, INCX, SIZE), INCX	leaq	(, INCY, SIZE), INCY	cmpq	$SIZE, INCX	jne	.L50	cmpq	$SIZE, INCY	jne	.L50	subq	$-32 * SIZE, X	subq	$-32 * SIZE, Y	testq	$SIZE, Y	je	.L05	movss	-32 * SIZE(X), %xmm0	movss	%xmm0, -32 * SIZE(Y)	addq	$1 * SIZE, X	addq	$1 * SIZE, Y	decq	M	jle	.L19	ALIGN_3.L05:	testq	$2 * SIZE, Y	je	.L10	cmpq	$1, M	je	.L17	movsd	-32 * SIZE(X), %xmm0	movsd	%xmm0, -32 * SIZE(Y)	addq	$2 * SIZE, X	addq	$2 * SIZE, Y	subq	$2, M	jle	.L19	ALIGN_4.L10:	testq	$3 * SIZE, X	jne	.L20	movq	M,  %rax	sarq	$5, %rax	jle	.L13	movaps	-32 * SIZE(X), %xmm0	movaps	-28 * SIZE(X), %xmm1	movaps	-24 * SIZE(X), %xmm2	movaps	-20 * SIZE(X), %xmm3	movaps	-16 * SIZE(X), %xmm4	movaps	-12 * SIZE(X), %xmm5	movaps	 -8 * SIZE(X), %xmm6	movaps	 -4 * SIZE(X), %xmm7	decq	%rax	jle	.L12	ALIGN_3.L11:	movaps	%xmm0, -32 * SIZE(Y)	movaps	 0 * SIZE(X), %xmm0	movaps	%xmm1, -28 * SIZE(Y)	movaps	 4 * SIZE(X), %xmm1	movaps	%xmm2, -24 * SIZE(Y)	movaps	 8 * SIZE(X), %xmm2	movaps	%xmm3, -20 * SIZE(Y)	movaps	12 * SIZE(X), %xmm3	movaps	%xmm4, -16 * SIZE(Y)	movaps	16 * SIZE(X), %xmm4	movaps	%xmm5, -12 * SIZE(Y)	movaps	20 * SIZE(X), %xmm5	movaps	%xmm6,  -8 * SIZE(Y)	movaps	24 * SIZE(X), %xmm6	movaps	%xmm7,  -4 * SIZE(Y)	subq	$-32 * SIZE, Y	movaps	28 * SIZE(X), %xmm7	subq	$-32 * SIZE, X	subq	$1, %rax	jg,pt	.L11	ALIGN_3.L12:	movaps	%xmm0, -32 * SIZE(Y)	movaps	%xmm1, -28 * SIZE(Y)	movaps	%xmm2, -24 * SIZE(Y)	movaps	%xmm3, -20 * SIZE(Y)	movaps	%xmm4, -16 * SIZE(Y)	movaps	%xmm5, -12 * SIZE(Y)	movaps	%xmm6,  -8 * SIZE(Y)	movaps	%xmm7,  -4 * SIZE(Y)	subq	$-32 * SIZE, X	subq	$-32 * SIZE, Y	ALIGN_3.L13:	testq	$16, M	jle	.L14	ALIGN_3	movaps	-32 * SIZE(X), %xmm0	movaps	-28 * SIZE(X), %xmm1	movaps	-24 * SIZE(X), %xmm2	movaps	-20 * SIZE(X), %xmm3	movaps	%xmm0, -32 * SIZE(Y)	movaps	%xmm1, -28 * SIZE(Y)	movaps	%xmm2, -24 * SIZE(Y)	movaps	%xmm3, -20 * SIZE(Y)	addq	$16 * SIZE, X	addq	$16 * SIZE, Y	ALIGN_3.L14:	testq	$8, M	jle	.L15	ALIGN_3	movaps	-32 * SIZE(X), %xmm0	movaps	-28 * SIZE(X), %xmm1	movaps	%xmm0, -32 * SIZE(Y)	movaps	%xmm1, -28 * SIZE(Y)	addq	$8 * SIZE, X	addq	$8 * SIZE, Y	ALIGN_3.L15:	testq	$4, M	jle	.L16	ALIGN_3	movaps	-32 * SIZE(X), %xmm0	movaps	%xmm0, -32 * SIZE(Y)	addq	$4 * SIZE, X	addq	$4 * SIZE, Y	ALIGN_3.L16:	testq	$2, M	jle	.L17	ALIGN_3	movsd	-32 * SIZE(X), %xmm0	movsd	%xmm0, 	-32 * SIZE(Y)	addq	$2 * SIZE, X	addq	$2 * SIZE, Y	ALIGN_3.L17:	testq	$1, M	jle	.L19	ALIGN_3	movss	-32 * SIZE(X), %xmm0	movss	%xmm0, -32 * SIZE(Y)	ALIGN_3.L19:	xorq	%rax,%rax	RESTOREREGISTERS	ret	ALIGN_3.L20:	testq	$2 * SIZE, X	jne	.L30	movaps	-33 * SIZE(X), %xmm0	movq	M,  %rax	sarq	$5, %rax	jle	.L23	movaps	-29 * SIZE(X), %xmm1	movaps	-25 * SIZE(X), %xmm2	movaps	-21 * SIZE(X), %xmm3	movaps	-17 * SIZE(X), %xmm4	movaps	-13 * SIZE(X), %xmm5	movaps	 -9 * SIZE(X), %xmm6	movaps	 -5 * SIZE(X), %xmm7	movaps	 -1 * SIZE(X), %xmm8	decq	%rax	jle .L22	ALIGN_3.L21:	movaps	%xmm1, %xmm9	palignr	$4, %xmm0, %xmm1	movaps	%xmm1, -32 * SIZE(Y)	movaps	  3 * SIZE(X), %xmm1	movaps	%xmm2, %xmm10	palignr	$4, %xmm9, %xmm2	movaps	%xmm2, -28 * SIZE(Y)	movaps	  7 * SIZE(X), %xmm2	movaps	%xmm3, %xmm11	palignr	$4, %xmm10, %xmm3	movaps	%xmm3, -24 * SIZE(Y)	movaps	 11 * SIZE(X), %xmm3	movaps	%xmm4, %xmm12	palignr	$4, %xmm11, %xmm4	movaps	%xmm4, -20 * SIZE(Y)	movaps	 15 * SIZE(X), %xmm4	movaps	%xmm5, %xmm13	palignr	$4, %xmm12, %xmm5	movaps	%xmm5, -16 * SIZE(Y)	movaps	 19 * SIZE(X), %xmm5	movaps	%xmm6, %xmm14	palignr	$4, %xmm13, %xmm6	movaps	%xmm6, -12 * SIZE(Y)	movaps	 23 * SIZE(X), %xmm6	movaps	%xmm7, %xmm15	palignr	$4, %xmm14, %xmm7	movaps	%xmm7,  -8 * SIZE(Y)	movaps	 27 * SIZE(X), %xmm7	movaps	%xmm8, %xmm0	palignr	$4, %xmm15, %xmm8	movaps	%xmm8,  -4 * SIZE(Y)	subq	$-32 * SIZE, Y	movaps	 31 * SIZE(X), %xmm8	subq	$-32 * SIZE, X	subq	$1, %rax	jg,pt	.L21	ALIGN_3.L22:	movaps	%xmm1, %xmm9	palignr	$4, %xmm0, %xmm1	movaps	%xmm1, -32 * SIZE(Y)	movaps	%xmm2, %xmm10	palignr	$4, %xmm9, %xmm2	movaps	%xmm2, -28 * SIZE(Y)	movaps	%xmm3, %xmm11	palignr	$4, %xmm10, %xmm3	movaps	%xmm3, -24 * SIZE(Y)	movaps	%xmm4, %xmm12	palignr	$4, %xmm11, %xmm4	movaps	%xmm4, -20 * SIZE(Y)	movaps	%xmm5, %xmm13	palignr	$4, %xmm12, %xmm5	movaps	%xmm5, -16 * SIZE(Y)	movaps	%xmm6, %xmm14	palignr	$4, %xmm13, %xmm6	movaps	%xmm6, -12 * SIZE(Y)	movaps	%xmm7, %xmm15	palignr	$4, %xmm14, %xmm7	movaps	%xmm7,  -8 * SIZE(Y)	movaps	%xmm8, %xmm0	palignr	$4, %xmm15, %xmm8	movaps	%xmm8,  -4 * SIZE(Y)	subq	$-32 * SIZE, X	subq	$-32 * SIZE, Y	ALIGN_3.L23:	testq	$16, M	jle	.L24	ALIGN_3	movaps	-29 * SIZE(X), %xmm1	movaps	-25 * SIZE(X), %xmm2	movaps	-21 * SIZE(X), %xmm3	movaps	-17 * SIZE(X), %xmm4	movaps	%xmm1, %xmm5	movaps	%xmm2, %xmm6	palignr	$4, %xmm0, %xmm1	palignr	$4, %xmm5, %xmm2	movaps	%xmm3, %xmm7	movaps	%xmm4, %xmm0	palignr	$4, %xmm6, %xmm3	palignr	$4, %xmm7, %xmm4	movaps	%xmm1, -32 * SIZE(Y)	movaps	%xmm2, -28 * SIZE(Y)	movaps	%xmm3, -24 * SIZE(Y)	movaps	%xmm4, -20 * SIZE(Y)	addq	$16 * SIZE, X	addq	$16 * SIZE, Y	ALIGN_3.L24:	testq	$8, M	jle	.L25	ALIGN_3	movaps	-29 * SIZE(X), %xmm1	movaps	-25 * SIZE(X), %xmm2	movaps	%xmm1, %xmm3	movaps	%xmm2, %xmm4	palignr	$4, %xmm0, %xmm1	palignr	$4, %xmm3, %xmm2	movaps	%xmm4, %xmm0	movaps	%xmm1, -32 * SIZE(Y)	movaps	%xmm2, -28 * SIZE(Y)	addq	$8 * SIZE, X	addq	$8 * SIZE, Y	ALIGN_3.L25:	testq	$4, M	jle	.L26	ALIGN_3	movaps	-29 * SIZE(X), %xmm1	palignr	$4, %xmm0, %xmm1	movaps	%xmm1, -32 * SIZE(Y)	addq	$4 * SIZE, X	addq	$4 * SIZE, Y	ALIGN_3.L26:	testq	$2, M	jle	.L27	ALIGN_3	movsd	-32 * SIZE(X), %xmm0	movsd	%xmm0, 	-32 * SIZE(Y)	addq	$2 * SIZE, X	addq	$2 * SIZE, Y	ALIGN_3.L27:	testq	$1, M	jle	.L29	ALIGN_3	movss	-32 * SIZE(X), %xmm0	movss	%xmm0, 	-32 * SIZE(Y)	ALIGN_3.L29:	xorq	%rax,%rax	RESTOREREGISTERS	ret	ALIGN_3.L30:	testq	$1 * SIZE, X	jne	.L40	movaps	 -34 * SIZE(X), %xmm0	movq	M,  %rax	sarq	$5, %rax	jle	.L33	movaps	 -30 * SIZE(X), %xmm1	movaps	 -26 * SIZE(X), %xmm2	movaps	 -22 * SIZE(X), %xmm3	movaps	 -18 * SIZE(X), %xmm4	movaps	 -14 * SIZE(X), %xmm5	movaps	 -10 * SIZE(X), %xmm6	movaps	  -6 * SIZE(X), %xmm7	decq	%rax	jle .L32	ALIGN_4.L31:	SHUFPD_1 %xmm1, %xmm0	movaps	%xmm0, -32 * SIZE(Y)	movaps	 -2 * SIZE(X), %xmm0	SHUFPD_1 %xmm2, %xmm1	movaps	%xmm1, -28 * SIZE(Y)	movaps	  2 * SIZE(X), %xmm1	SHUFPD_1 %xmm3, %xmm2	movaps	%xmm2, -24 * SIZE(Y)	movaps	  6 * SIZE(X), %xmm2	SHUFPD_1 %xmm4, %xmm3	movaps	%xmm3, -20 * SIZE(Y)	movaps	 10 * SIZE(X), %xmm3	SHUFPD_1 %xmm5, %xmm4	movaps	%xmm4, -16 * SIZE(Y)	movaps	 14 * SIZE(X), %xmm4	SHUFPD_1 %xmm6, %xmm5	movaps	%xmm5, -12 * SIZE(Y)	movaps	 18 * SIZE(X), %xmm5	SHUFPD_1 %xmm7, %xmm6	movaps	%xmm6,  -8 * SIZE(Y)	movaps	 22 * SIZE(X), %xmm6	SHUFPD_1 %xmm0, %xmm7	movaps	%xmm7,  -4 * SIZE(Y)	subq	$-32 * SIZE, Y	movaps	 26 * SIZE(X), %xmm7	subq	$-32 * SIZE, X	subq	$1, %rax	jg,pt	.L31	ALIGN_3.L32:	SHUFPD_1 %xmm1, %xmm0	movaps	%xmm0, -32 * SIZE(Y)	movaps	 -2 * SIZE(X), %xmm0	SHUFPD_1 %xmm2, %xmm1	movaps	%xmm1, -28 * SIZE(Y)	SHUFPD_1 %xmm3, %xmm2	movaps	%xmm2, -24 * SIZE(Y)	SHUFPD_1 %xmm4, %xmm3	movaps	%xmm3, -20 * SIZE(Y)	SHUFPD_1 %xmm5, %xmm4	movaps	%xmm4, -16 * SIZE(Y)	SHUFPD_1 %xmm6, %xmm5	movaps	%xmm5, -12 * SIZE(Y)	SHUFPD_1 %xmm7, %xmm6	movaps	%xmm6,  -8 * SIZE(Y)	subq	$-32 * SIZE, X	SHUFPD_1 %xmm0, %xmm7	movaps	%xmm7,  -4 * SIZE(Y)	subq	$-32 * SIZE, Y	ALIGN_3.L33:	testq	$16, M	jle	.L34	ALIGN_3	movaps	-30 * SIZE(X), %xmm1	movaps	-26 * SIZE(X), %xmm2	movaps	-22 * SIZE(X), %xmm3	movaps	-18 * SIZE(X), %xmm8	SHUFPD_1 %xmm1, %xmm0	movaps	%xmm0, -32 * SIZE(Y)	movaps	%xmm8, %xmm0	SHUFPD_1 %xmm2, %xmm1	movaps	%xmm1, -28 * SIZE(Y)	SHUFPD_1 %xmm3, %xmm2	movaps	%xmm2, -24 * SIZE(Y)	addq	$16 * SIZE, X	SHUFPD_1 %xmm8, %xmm3	movaps	%xmm3, -20 * SIZE(Y)	addq	$16 * SIZE, Y	ALIGN_3.L34:	testq	$8, M	jle	.L35	ALIGN_3 	movaps	-30 * SIZE(X), %xmm1	SHUFPD_1 %xmm1, %xmm0	movaps	-26 * SIZE(X), %xmm2	SHUFPD_1 %xmm2, %xmm1	movaps	%xmm0, -32 * SIZE(Y)	movaps	%xmm1, -28 * SIZE(Y)	movaps	%xmm2, %xmm0	addq	$8 * SIZE, X	addq	$8 * SIZE, Y	ALIGN_3.L35:	testq	$4, M	jle	.L36	ALIGN_3	movaps	-30 * SIZE(X), %xmm1	SHUFPD_1 %xmm1, %xmm0	movaps	%xmm0, -32 * SIZE(Y)	movaps	%xmm1, %xmm0	addq	$4 * SIZE, X	addq	$4 * SIZE, Y	ALIGN_3.L36:	testq	$2, M	jle	.L37	ALIGN_3	movhps	%xmm0, -32 * SIZE(Y)	addq	$2 * SIZE, X	addq	$2 * SIZE, Y	ALIGN_3.L37:	testq	$1, M	jle	.L39	ALIGN_3	movss	-32 * SIZE(X), %xmm0	movss	%xmm0, 	-32 * SIZE(Y)	ALIGN_3.L39:	xorq	%rax,%rax	RESTOREREGISTERS	ret	ALIGN_3.L40:	movaps	-35 * SIZE(X), %xmm0	movq	M,  %rax	sarq	$5, %rax	jle	.L43	movaps	-31 * SIZE(X), %xmm1	movaps	-27 * SIZE(X), %xmm2	movaps	-23 * SIZE(X), %xmm3	movaps	-19 * SIZE(X), %xmm4	movaps	-15 * SIZE(X), %xmm5	movaps	-11 * SIZE(X), %xmm6	movaps	 -7 * SIZE(X), %xmm7	movaps	 -3 * SIZE(X), %xmm8	decq	%rax	jle .L42	ALIGN_3.L41:	movaps	%xmm1, %xmm9	palignr	$12, %xmm0, %xmm1	movaps	%xmm1, -32 * SIZE(Y)	movaps	  1 * SIZE(X), %xmm1	movaps	%xmm2, %xmm10	palignr	$12, %xmm9, %xmm2	movaps	%xmm2, -28 * SIZE(Y)	movaps	  5 * SIZE(X), %xmm2	movaps	%xmm3, %xmm11	palignr	$12, %xmm10, %xmm3	movaps	%xmm3, -24 * SIZE(Y)	movaps	  9 * SIZE(X), %xmm3	movaps	%xmm4, %xmm12	palignr	$12, %xmm11, %xmm4	movaps	%xmm4, -20 * SIZE(Y)	movaps	 13 * SIZE(X), %xmm4	movaps	%xmm5, %xmm13	palignr	$12, %xmm12, %xmm5	movaps	%xmm5, -16 * SIZE(Y)	movaps	 17 * SIZE(X), %xmm5	movaps	%xmm6, %xmm14	palignr	$12, %xmm13, %xmm6	movaps	%xmm6, -12 * SIZE(Y)	movaps	 21 * SIZE(X), %xmm6	movaps	%xmm7, %xmm15	palignr	$12, %xmm14, %xmm7	movaps	%xmm7,  -8 * SIZE(Y)	movaps	 25 * SIZE(X), %xmm7	movaps	%xmm8, %xmm0	palignr	$12, %xmm15, %xmm8	movaps	%xmm8,  -4 * SIZE(Y)	subq	$-32 * SIZE, Y	movaps	 29 * SIZE(X), %xmm8	subq	$-32 * SIZE, X	subq	$1, %rax	jg,pt	.L41	ALIGN_3.L42:	movaps	%xmm1, %xmm9	palignr	$12, %xmm0, %xmm1	movaps	%xmm1, -32 * SIZE(Y)	movaps	%xmm2, %xmm10	palignr	$12, %xmm9, %xmm2	movaps	%xmm2, -28 * SIZE(Y)	movaps	%xmm3, %xmm11	palignr	$12, %xmm10, %xmm3	movaps	%xmm3, -24 * SIZE(Y)	movaps	%xmm4, %xmm12	palignr	$12, %xmm11, %xmm4	movaps	%xmm4, -20 * SIZE(Y)	movaps	%xmm5, %xmm13	palignr	$12, %xmm12, %xmm5	movaps	%xmm5, -16 * SIZE(Y)	movaps	%xmm6, %xmm14	palignr	$12, %xmm13, %xmm6	movaps	%xmm6, -12 * SIZE(Y)	movaps	%xmm7, %xmm15	palignr	$12, %xmm14, %xmm7	movaps	%xmm7,  -8 * SIZE(Y)	movaps	%xmm8, %xmm0	palignr	$12, %xmm15, %xmm8	movaps	%xmm8,  -4 * SIZE(Y)	subq	$-32 * SIZE, X	subq	$-32 * SIZE, Y	ALIGN_3.L43:	testq	$16, M	jle	.L44	ALIGN_3	movaps	-31 * SIZE(X), %xmm1	movaps	-27 * SIZE(X), %xmm2	movaps	-23 * SIZE(X), %xmm3	movaps	-19 * SIZE(X), %xmm4	movaps	%xmm1, %xmm5	movaps	%xmm2, %xmm6	palignr	$12, %xmm0, %xmm1	palignr	$12, %xmm5, %xmm2	movaps	%xmm3, %xmm7	movaps	%xmm4, %xmm0	palignr	$12, %xmm6, %xmm3	palignr	$12, %xmm7, %xmm4	movaps	%xmm1, -32 * SIZE(Y)	movaps	%xmm2, -28 * SIZE(Y)	movaps	%xmm3, -24 * SIZE(Y)	movaps	%xmm4, -20 * SIZE(Y)	addq	$16 * SIZE, X	addq	$16 * SIZE, Y	ALIGN_3.L44:	testq	$8, M	jle	.L45	ALIGN_3	movaps	-31 * SIZE(X), %xmm1	movaps	-27 * SIZE(X), %xmm2	movaps	%xmm1, %xmm3	movaps	%xmm2, %xmm4	palignr	$12, %xmm0, %xmm1	palignr	$12, %xmm3, %xmm2	movaps	%xmm4, %xmm0	movaps	%xmm1, -32 * SIZE(Y)	movaps	%xmm2, -28 * SIZE(Y)	addq	$8 * SIZE, X	addq	$8 * SIZE, Y	ALIGN_3.L45:	testq	$4, M	jle	.L46	ALIGN_3	movaps	-31 * SIZE(X), %xmm1	palignr	$12, %xmm0, %xmm1	movaps	%xmm1, -32 * SIZE(Y)	addq	$4 * SIZE, X	addq	$4 * SIZE, Y	ALIGN_3.L46:	testq	$2, M	jle	.L47	ALIGN_3	movsd	-32 * SIZE(X), %xmm0	movsd	%xmm0, 	-32 * SIZE(Y)	addq	$2 * SIZE, X	addq	$2 * SIZE, Y	ALIGN_3.L47:	testq	$1, M	jle	.L49	ALIGN_3	movss	-32 * SIZE(X), %xmm0	movss	%xmm0, 	-32 * SIZE(Y)	ALIGN_3.L49:	xorq	%rax,%rax	RESTOREREGISTERS	ret	ALIGN_3.L50:	movq	M,  %rax	sarq	$3, %rax	jle	.L55	ALIGN_3.L51:	movss	0 * SIZE(X), %xmm0	addq	INCX, X	movss	0 * SIZE(X), %xmm1	addq	INCX, X	movss	0 * SIZE(X), %xmm2	addq	INCX, X	movss	0 * SIZE(X), %xmm3	addq	INCX, X	movss	0 * SIZE(X), %xmm4	addq	INCX, X	movss	0 * SIZE(X), %xmm5	addq	INCX, X	movss	0 * SIZE(X), %xmm6	addq	INCX, X	movss	0 * SIZE(X), %xmm7	addq	INCX, X	movss	%xmm0, 0 * SIZE(Y)	addq	INCY, Y	movss	%xmm1, 0 * SIZE(Y)	addq	INCY, Y	movss	%xmm2, 0 * SIZE(Y)	addq	INCY, Y	movss	%xmm3, 0 * SIZE(Y)	addq	INCY, Y	movss	%xmm4, 0 * SIZE(Y)	addq	INCY, Y	movss	%xmm5, 0 * SIZE(Y)	addq	INCY, Y	movss	%xmm6, 0 * SIZE(Y)	addq	INCY, Y	movss	%xmm7, 0 * SIZE(Y)	addq	INCY, Y	decq	%rax	jg	.L51	ALIGN_3.L55:	movq	M,  %rax	andq	$7, %rax	jle	.L57	ALIGN_3.L56:	movss	(X), %xmm0	addq	INCX, X	movss	%xmm0, (Y)	addq	INCY, Y	decq	%rax	jg	.L56	ALIGN_3.L57:	xorq	%rax, %rax	RESTOREREGISTERS	ret	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 + -