hamdist.asm revision 1.1.1.2
1dnl Alpha ev67 mpn_hamdist -- mpn hamming distance. 2 3dnl Copyright 2003, 2005 Free Software Foundation, Inc. 4 5dnl This file is part of the GNU MP Library. 6dnl 7dnl The GNU MP Library is free software; you can redistribute it and/or modify 8dnl it under the terms of either: 9dnl 10dnl * the GNU Lesser General Public License as published by the Free 11dnl Software Foundation; either version 3 of the License, or (at your 12dnl option) any later version. 13dnl 14dnl or 15dnl 16dnl * the GNU General Public License as published by the Free Software 17dnl Foundation; either version 2 of the License, or (at your option) any 18dnl later version. 19dnl 20dnl or both in parallel, as here. 21dnl 22dnl The GNU MP Library is distributed in the hope that it will be useful, but 23dnl WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY 24dnl or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License 25dnl for more details. 26dnl 27dnl You should have received copies of the GNU General Public License and the 28dnl GNU Lesser General Public License along with the GNU MP Library. If not, 29dnl see https://www.gnu.org/licenses/. 30 31include(`../config.m4') 32 33 34C ev67: 2.5 cycles/limb 35 36 37C unsigned long mpn_hamdist (mp_srcptr xp, mp_srcptr yp, mp_size_t size); 38C 39C The hope was for 2.0 c/l here, but that isn't achieved. We're limited by 40C renaming register shortage. Since we need 5 instructions per limb, further 41C unrolling could approach 1.5 c/l. 42C 43C The main loop processes two limbs from each operand on each iteration. An 44C odd size is handled by processing xp[0]^yp[0] at the start. If the size 45C is even that result is discarded, and is repeated by the main loop. 46C 47 48ASM_START() 49PROLOGUE(mpn_hamdist) 50 51 C r16 xp 52 C r17 yp 53 C r18 size 54 55 ldq r1, 0(r16) C L0 xp[0] 56 ldq r2, 0(r17) C L1 yp[0] 57 and r18, 1, r8 C U1 1 if size odd 58 srl r18, 1, r18 C U0 size, limb pairs 59 60 clr r0 C L0 initial total 61 s8addq r8, r17, r17 C U1 yp++ if size odd 62 s8addq r8, r16, r16 C L1 xp++ if size odd 63 clr r6 C U0 dummy initial xor 1 64 65 xor r1, r2, r5 C L initial xor 0 66 beq r18, L(one) C U if size==1 67 68 cmoveq r8, r31, r5 C L discard first limb if size even 69 unop C U 70 71 72 ALIGN(16) 73L(top): 74 C r0 total accumulating 75 C r7 xor 0 76 C r8 xor 1 77 C r16 xp, incrementing 78 C r17 yp, incrementing 79 C r18 size, limb pairs, decrementing 80 81 ldq r1, 0(r16) C L 82 ldq r2, 0(r17) C L 83 ctpop r5, r7 C U0 84 lda r16, 16(r16) C U 85 86 ldq r3, -8(r16) C L 87 ldq r4, 8(r17) C L 88 ctpop r6, r8 C U0 89 lda r17, 16(r17) C U 90 91 ldl r31, 256(r16) C L prefetch 92 ldl r31, 256(r17) C L prefetch 93 xor r1, r2, r5 C U 94 lda r18, -1(r18) C U 95 96 xor r3, r4, r6 C U 97 addq r0, r7, r0 C L 98 addq r0, r8, r0 C L 99 bne r18, L(top) C U 100 101 102 ctpop r6, r8 C U0 103 addq r0, r8, r0 C L 104L(one): 105 ctpop r5, r7 C U0 106 addq r0, r7, r0 C L 107 108 ret r31, (r26), 1 C L0 109 110EPILOGUE() 111ASM_END() 112