1 /* mpn_toom_interpolate_8pts -- Interpolate for toom54, 63, 72.
3 Contributed to the GNU project by Marco Bodrato.
5 THE FUNCTION IN THIS FILE IS INTERNAL WITH A MUTABLE INTERFACE. IT IS ONLY
6 SAFE TO REACH IT THROUGH DOCUMENTED INTERFACES. IN FACT, IT IS ALMOST
7 GUARANTEED THAT IT WILL CHANGE OR DISAPPEAR IN A FUTURE GNU MP RELEASE.
9 Copyright 2009 Free Software Foundation, Inc.
11 This file is part of the GNU MP Library.
13 The GNU MP Library is free software; you can redistribute it and/or modify
14 it under the terms of the GNU Lesser General Public License as published by
15 the Free Software Foundation; either version 3 of the License, or (at your
16 option) any later version.
18 The GNU MP Library is distributed in the hope that it will be useful, but
19 WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
20 or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public
21 License for more details.
23 You should have received a copy of the GNU Lesser General Public License
24 along with the GNU MP Library. If not, see http://www.gnu.org/licenses/. */
29 #define BINVERT_3 MODLIMB_INVERSE_3
32 ((((GMP_NUMB_MAX >> (GMP_NUMB_BITS % 4)) / 15) * 14 * 16 & GMP_NUMB_MAX) + 15)
34 #define BINVERT_45 ((BINVERT_15 * BINVERT_3) & GMP_NUMB_MASK)
36 #ifndef mpn_divexact_by3
37 #if HAVE_NATIVE_mpn_pi1_bdiv_q_1
38 #define mpn_divexact_by3(dst,src,size) mpn_pi1_bdiv_q_1(dst,src,size,3,BINVERT_3,0)
40 #define mpn_divexact_by3(dst,src,size) mpn_divexact_1(dst,src,size,3)
44 #ifndef mpn_divexact_by45
45 #if GMP_NUMB_BITS % 12 == 0
46 #define mpn_divexact_by45(dst,src,size) \
47 (63 & 19 * mpn_bdiv_dbm1 (dst, src, size, __GMP_CAST (mp_limb_t, GMP_NUMB_MASK / 45)))
49 #if HAVE_NATIVE_mpn_pi1_bdiv_q_1
50 #define mpn_divexact_by45(dst,src,size) mpn_pi1_bdiv_q_1(dst,src,size,45,BINVERT_45,0)
52 #define mpn_divexact_by45(dst,src,size) mpn_divexact_1(dst,src,size,45)
57 #if HAVE_NATIVE_mpn_sublsh_n
58 #define DO_mpn_sublsh_n(dst,src,n,s,ws) mpn_sublsh_n (dst,src,n,s)
61 DO_mpn_sublsh_n (mp_ptr dst, mp_srcptr src, mp_size_t n, unsigned int s, mp_ptr ws)
64 return mpn_submul_1(dst,src,n,CNST_LIMB(1) <<(s));
67 __cy = mpn_lshift (ws,src,n,s);
68 return __cy + mpn_sub_n (dst,dst,ws,n);
74 #if HAVE_NATIVE_mpn_subrsh
75 #define DO_mpn_subrsh(dst,nd,src,ns,s,ws) mpn_subrsh (dst,nd,src,ns,s)
77 /* This is not a correct definition, it assumes no carry */
78 #define DO_mpn_subrsh(dst,nd,src,ns,s,ws) \
81 MPN_DECR_U (dst, nd, src[0] >> s); \
82 __cy = DO_mpn_sublsh_n (dst, src + 1, ns - 1, GMP_NUMB_BITS - s, ws); \
83 MPN_DECR_U (dst + ns - 1, nd - ns + 1, __cy); \
87 /* Interpolation for Toom-4.5 (or Toom-4), using the evaluation
88 points: infinity(4.5 only), 4, -4, 2, -2, 1, -1, 0. More precisely,
89 we want to compute f(2^(GMP_NUMB_BITS * n)) for a polynomial f of
90 degree 7 (or 6), given the 8 (rsp. 7) values:
92 r1 = limit at infinity of f(x) / x^7,
101 All couples of the form f(n),f(-n) must be already mixed with
102 toom_couple_handling(f(n),...,f(-n),...)
104 The result is stored in {pp, spt + 7*n (or 6*n)}.
105 At entry, r8 is stored at {pp, 2n},
106 r5 is stored at {pp + 3n, 3n + 1}.
108 The other values are 2n+... limbs each (with most significant limbs small).
110 All intermediate results are positive.
111 Inputs are destroyed.
115 mpn_toom_interpolate_8pts (mp_ptr pp, mp_size_t n,
116 mp_ptr r3, mp_ptr r7,
117 mp_size_t spt, mp_ptr ws)
121 r5 = (pp + 3 * n); /* 3n+1 */
122 r1 = (pp + 7 * n); /* spt */
124 /******************************* interpolation *****************************/
126 DO_mpn_subrsh(r3+n, 2 * n + 1, pp, 2 * n, 4, ws);
127 cy = DO_mpn_sublsh_n (r3, r1, spt, 12, ws);
128 MPN_DECR_U (r3 + spt, 3 * n + 1 - spt, cy);
130 DO_mpn_subrsh(r5+n, 2 * n + 1, pp, 2 * n, 2, ws);
131 cy = DO_mpn_sublsh_n (r5, r1, spt, 6, ws);
132 MPN_DECR_U (r5 + spt, 3 * n + 1 - spt, cy);
134 r7[3*n] -= mpn_sub_n (r7+n, r7+n, pp, 2 * n);
135 cy = mpn_sub_n (r7, r7, r1, spt);
136 MPN_DECR_U (r7 + spt, 3 * n + 1 - spt, cy);
138 ASSERT_NOCARRY(mpn_sub_n (r3, r3, r5, 3 * n + 1));
139 ASSERT_NOCARRY(mpn_rshift(r3, r3, 3 * n + 1, 2));
141 ASSERT_NOCARRY(mpn_sub_n (r5, r5, r7, 3 * n + 1));
143 ASSERT_NOCARRY(mpn_sub_n (r3, r3, r5, 3 * n + 1));
145 mpn_divexact_by45 (r3, r3, 3 * n + 1);
147 ASSERT_NOCARRY(mpn_divexact_by3 (r5, r5, 3 * n + 1));
149 ASSERT_NOCARRY(DO_mpn_sublsh_n (r5, r3, 3 * n + 1, 2, ws));
151 /* last interpolation steps... */
152 /* ... are mixed with recomposition */
154 /***************************** recomposition *******************************/
156 pp[] prior to operations:
157 |_H r1|_L r1|____||_H r5|_M_r5|_L r5|_____|_H r8|_L r8|pp
159 summation scheme for remaining operations:
160 |____8|n___7|n___6|n___5|n___4|n___3|n___2|n____|n____|pp
161 |_H r1|_L r1|____||_H*r5|_M r5|_L r5|_____|_H_r8|_L r8|pp
168 cy = mpn_add_n (pp + n, pp + n, r7, n); /* Hr8+Lr7-Lr5 */
169 cy-= mpn_sub_n (pp + n, pp + n, r5, n);
171 MPN_DECR_U (r7 + n, 2*n + 1, 1);
173 MPN_INCR_U (r7 + n, 2*n + 1, cy);
175 cy = mpn_sub_n (pp + 2*n, r7 + n, r5 + n, n); /* Mr7-Mr5 */
176 MPN_DECR_U (r7 + 2*n, n + 1, cy);
178 cy = mpn_add_n (pp + 3*n, r5, r7+ 2*n, n+1); /* Hr7+Lr5 */
179 r5[3*n]+= mpn_add_n (r5 + 2*n, r5 + 2*n, r3, n); /* Hr5+Lr3 */
180 cy-= mpn_sub_n (pp + 3*n, pp + 3*n, r5 + 2*n, n+1); /* Hr7-Hr5+Lr5-Lr3 */
181 if (UNLIKELY(0 > cy))
182 MPN_DECR_U (r5 + n + 1, 2*n, 1);
184 MPN_INCR_U (r5 + n + 1, 2*n, cy);
186 ASSERT_NOCARRY(mpn_sub_n(pp + 4*n, r5 + n, r3 + n, 2*n +1)); /* Mr5-Mr3,Hr5-Hr3 */
188 cy = mpn_add_1 (pp + 6*n, r3 + n, n, pp[6*n]);
189 MPN_INCR_U (r3 + 2*n, n + 1, cy);
190 cy = r3[3*n] + mpn_add_n (pp + 7*n, pp + 7*n, r3 + 2*n, n);
191 if (LIKELY(spt != n))
192 MPN_INCR_U (pp + 8*n, spt - n, cy);