## Mercurial > dropbear

### view libtommath/bn_fast_s_mp_sqr.c @ 350:01e4180895ba

Find changesets by keywords (author, files, the commit message), revision
number or hash, or revset expression.

merge of '9a4e042fd565f46141e81e0c1ab90260303348fe'
and 'bea3887a5875cf3ab8a1331e15e698b37b61fe37'

author | Matt Johnston <matt@ucc.asn.au> |
---|---|

date | Mon, 07 Aug 2006 13:41:23 +0000 |

parents | eed26cff980b |

children | 5ff8218bcee9 |

line wrap: on

line source

#include <tommath.h> #ifdef BN_FAST_S_MP_SQR_C /* LibTomMath, multiple-precision integer library -- Tom St Denis * * LibTomMath is a library that provides multiple-precision * integer arithmetic as well as number theoretic functionality. * * The library was designed directly after the MPI library by * Michael Fromberger but has been written from scratch with * additional optimizations in place. * * The library is free for all purposes without any express * guarantee it works. * * Tom St Denis, [email protected], http://math.libtomcrypt.org */ /* the jist of squaring... * you do like mult except the offset of the tmpx [one that * starts closer to zero] can't equal the offset of tmpy. * So basically you set up iy like before then you min it with * (ty-tx) so that it never happens. You double all those * you add in the inner loop After that loop you do the squares and add them in. */ int fast_s_mp_sqr (mp_int * a, mp_int * b) { int olduse, res, pa, ix, iz; mp_digit W[MP_WARRAY], *tmpx; mp_word W1; /* grow the destination as required */ pa = a->used + a->used; if (b->alloc < pa) { if ((res = mp_grow (b, pa)) != MP_OKAY) { return res; } } /* number of output digits to produce */ W1 = 0; for (ix = 0; ix < pa; ix++) { int tx, ty, iy; mp_word _W; mp_digit *tmpy; /* clear counter */ _W = 0; /* get offsets into the two bignums */ ty = MIN(a->used-1, ix); tx = ix - ty; /* setup temp aliases */ tmpx = a->dp + tx; tmpy = a->dp + ty; /* this is the number of times the loop will iterrate, essentially while (tx++ < a->used && ty-- >= 0) { ... } */ iy = MIN(a->used-tx, ty+1); /* now for squaring tx can never equal ty * we halve the distance since they approach at a rate of 2x * and we have to round because odd cases need to be executed */ iy = MIN(iy, (ty-tx+1)>>1); /* execute loop */ for (iz = 0; iz < iy; iz++) { _W += ((mp_word)*tmpx++)*((mp_word)*tmpy--); } /* double the inner product and add carry */ _W = _W + _W + W1; /* even columns have the square term in them */ if ((ix&1) == 0) { _W += ((mp_word)a->dp[ix>>1])*((mp_word)a->dp[ix>>1]); } /* store it */ W[ix] = (mp_digit)(_W & MP_MASK); /* make next carry */ W1 = _W >> ((mp_word)DIGIT_BIT); } /* setup dest */ olduse = b->used; b->used = a->used+a->used; { mp_digit *tmpb; tmpb = b->dp; for (ix = 0; ix < pa; ix++) { *tmpb++ = W[ix] & MP_MASK; } /* clear unused digits [that existed in the old copy of c] */ for (; ix < olduse; ix++) { *tmpb++ = 0; } } mp_clamp (b); return MP_OKAY; } #endif