函数源码

Linux Kernel

v5.5.9

Brick Technologies Co., Ltd

Source File:lib\crypto\curve25519-fiat32.c Create Date:2022-07-27 07:43:49
首页 Copyright©Brick

755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
void curve25519_generic(u8 out[CURVE25519_KEY_SIZE],
            const u8 scalar[CURVE25519_KEY_SIZE],
            const u8 point[CURVE25519_KEY_SIZE])
{
    fe x1, x2, z2, x3, z3;
    fe_loose x2l, z2l, x3l;
    unsigned swap = 0;
    int pos;
    u8 e[32];
 
    memcpy(e, scalar, 32);
    curve25519_clamp_secret(e);
 
    /* The following implementation was transcribed to Coq and proven to
     * correspond to unary scalar multiplication in affine coordinates given
     * that x1 != 0 is the x coordinate of some point on the curve. It was
     * also checked in Coq that doing a ladderstep with x1 = x3 = 0 gives
     * z2' = z3' = 0, and z2 = z3 = 0 gives z2' = z3' = 0. The statement was
     * quantified over the underlying field, so it applies to Curve25519
     * itself and the quadratic twist of Curve25519. It was not proven in
     * Coq that prime-field arithmetic correctly simulates extension-field
     * arithmetic on prime-field values. The decoding of the byte array
     * representation of e was not considered.
     *
     * Specification of Montgomery curves in affine coordinates:
     *
     * Proof that these form a group that is isomorphic to a Weierstrass
     * curve:
     *
     * Coq transcription and correctness proof of the loop
     * (where scalarbits=255):
     * preconditions: 0 <= e < 2^255 (not necessarily e < order),
     * fe_invert(0) = 0
     */
    fe_frombytes(&x1, point);
    fe_1(&x2);
    fe_0(&z2);
    fe_copy(&x3, &x1);
    fe_1(&z3);
 
    for (pos = 254; pos >= 0; --pos) {
        fe tmp0, tmp1;
        fe_loose tmp0l, tmp1l;
        /* loop invariant as of right before the test, for the case
         * where x1 != 0:
         *   pos >= -1; if z2 = 0 then x2 is nonzero; if z3 = 0 then x3
         *   is nonzero
         *   let r := e >> (pos+1) in the following equalities of
         *   projective points:
         *   to_xz (r*P)     === if swap then (x3, z3) else (x2, z2)
         *   to_xz ((r+1)*P) === if swap then (x2, z2) else (x3, z3)
         *   x1 is the nonzero x coordinate of the nonzero
         *   point (r*P-(r+1)*P)
         */
        unsigned b = 1 & (e[pos / 8] >> (pos & 7));
        swap ^= b;
        fe_cswap(&x2, &x3, swap);
        fe_cswap(&z2, &z3, swap);
        swap = b;
        /* Coq transcription of ladderstep formula (called from
         * transcribed loop):
         */
        fe_sub(&tmp0l, &x3, &z3);
        fe_sub(&tmp1l, &x2, &z2);
        fe_add(&x2l, &x2, &z2);
        fe_add(&z2l, &x3, &z3);
        fe_mul_tll(&z3, &tmp0l, &x2l);
        fe_mul_tll(&z2, &z2l, &tmp1l);
        fe_sq_tl(&tmp0, &tmp1l);
        fe_sq_tl(&tmp1, &x2l);
        fe_add(&x3l, &z3, &z2);
        fe_sub(&z2l, &z3, &z2);
        fe_mul_ttt(&x2, &tmp1, &tmp0);
        fe_sub(&tmp1l, &tmp1, &tmp0);
        fe_sq_tl(&z2, &z2l);
        fe_mul121666(&z3, &tmp1l);
        fe_sq_tl(&x3, &x3l);
        fe_add(&tmp0l, &tmp0, &z3);
        fe_mul_ttt(&z3, &x1, &z2);
        fe_mul_tll(&z2, &tmp1l, &tmp0l);
    }
    /* here pos=-1, so r=e, so to_xz (e*P) === if swap then (x3, z3)
     * else (x2, z2)
     */
    fe_cswap(&x2, &x3, swap);
    fe_cswap(&z2, &z3, swap);
 
    fe_invert(&z2, &z2);
    fe_mul_ttt(&x2, &x2, &z2);
    fe_tobytes(out, &x2);
 
    memzero_explicit(&x1, sizeof(x1));
    memzero_explicit(&x2, sizeof(x2));
    memzero_explicit(&z2, sizeof(z2));
    memzero_explicit(&x3, sizeof(x3));
    memzero_explicit(&z3, sizeof(z3));
    memzero_explicit(&x2l, sizeof(x2l));
    memzero_explicit(&z2l, sizeof(z2l));
    memzero_explicit(&x3l, sizeof(x3l));
    memzero_explicit(&e, sizeof(e));
}