github.com/aergoio/aergo@v1.3.1/libtool/src/gmp-6.1.2/mpn/x86/pentium4/copyd.asm (about) 1 dnl Pentium-4 mpn_copyd -- copy limb vector, decrementing. 2 3 dnl Copyright 1999-2001 Free Software Foundation, Inc. 4 5 dnl This file is part of the GNU MP Library. 6 dnl 7 dnl The GNU MP Library is free software; you can redistribute it and/or modify 8 dnl it under the terms of either: 9 dnl 10 dnl * the GNU Lesser General Public License as published by the Free 11 dnl Software Foundation; either version 3 of the License, or (at your 12 dnl option) any later version. 13 dnl 14 dnl or 15 dnl 16 dnl * the GNU General Public License as published by the Free Software 17 dnl Foundation; either version 2 of the License, or (at your option) any 18 dnl later version. 19 dnl 20 dnl or both in parallel, as here. 21 dnl 22 dnl The GNU MP Library is distributed in the hope that it will be useful, but 23 dnl WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY 24 dnl or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License 25 dnl for more details. 26 dnl 27 dnl You should have received copies of the GNU General Public License and the 28 dnl GNU Lesser General Public License along with the GNU MP Library. If not, 29 dnl see https://www.gnu.org/licenses/. 30 31 32 dnl The std/rep/movsl/cld is very slow for small blocks on pentium4. Its 33 dnl startup time seems to be about 165 cycles. It then needs 2.6 c/l. 34 dnl We therefore use an open-coded 2 c/l copying loop. 35 36 dnl Ultimately, we may want to use 64-bit movq or 128-bit movdqu in some 37 dnl nifty unrolled arrangement. Clearly, that could reach much higher 38 dnl speeds, at least for large blocks. 39 40 include(`../config.m4') 41 42 43 defframe(PARAM_SIZE, 12) 44 defframe(PARAM_SRC, 8) 45 defframe(PARAM_DST, 4) 46 47 TEXT 48 ALIGN(8) 49 50 PROLOGUE(mpn_copyd) 51 deflit(`FRAME',0) 52 53 movl PARAM_SIZE, %ecx 54 55 movl PARAM_SRC, %eax 56 movl PARAM_DST, %edx 57 movl %ebx, PARAM_SIZE 58 addl $-1, %ecx 59 js L(end) 60 61 L(loop): 62 movl (%eax,%ecx,4), %ebx 63 movl %ebx, (%edx,%ecx,4) 64 addl $-1, %ecx 65 66 jns L(loop) 67 L(end): 68 movl PARAM_SIZE, %ebx 69 ret 70 71 EPILOGUE()