diff options
Diffstat (limited to 'crypto/openssl/crypto/bn')
-rwxr-xr-x | crypto/openssl/crypto/bn/asm/armv4-gf2m.pl | 24 | ||||
-rw-r--r-- | crypto/openssl/crypto/bn/asm/ia64.S | 4 | ||||
-rwxr-xr-x | crypto/openssl/crypto/bn/asm/s390x-gf2m.pl | 6 | ||||
-rwxr-xr-x | crypto/openssl/crypto/bn/asm/x86-gf2m.pl | 16 | ||||
-rw-r--r-- | crypto/openssl/crypto/bn/asm/x86_64-gcc.c | 2 | ||||
-rwxr-xr-x | crypto/openssl/crypto/bn/asm/x86_64-gf2m.pl | 16 | ||||
-rw-r--r-- | crypto/openssl/crypto/bn/bn_exp.c | 7 | ||||
-rw-r--r-- | crypto/openssl/crypto/bn/bn_gcd.c | 2 | ||||
-rw-r--r-- | crypto/openssl/crypto/bn/bn_gf2m.c | 11 | ||||
-rw-r--r-- | crypto/openssl/crypto/bn/bn_mont.c | 9 | ||||
-rw-r--r-- | crypto/openssl/crypto/bn/bn_recp.c | 4 | ||||
-rw-r--r-- | crypto/openssl/crypto/bn/bn_x931p.c | 7 | ||||
-rw-r--r-- | crypto/openssl/crypto/bn/bntest.c | 56 |
13 files changed, 115 insertions, 49 deletions
diff --git a/crypto/openssl/crypto/bn/asm/armv4-gf2m.pl b/crypto/openssl/crypto/bn/asm/armv4-gf2m.pl index c52e0b7..22ad1f8 100755 --- a/crypto/openssl/crypto/bn/asm/armv4-gf2m.pl +++ b/crypto/openssl/crypto/bn/asm/armv4-gf2m.pl @@ -41,13 +41,13 @@ $code=<<___; .align 5 mul_1x1_neon: vshl.u64 `&Dlo("q1")`,d16,#8 @ q1-q3 are slided $a - vmull.p8 `&Q("d0")`,d16,d17 @ a·bb + vmull.p8 `&Q("d0")`,d16,d17 @ a·bb vshl.u64 `&Dlo("q2")`,d16,#16 - vmull.p8 q1,`&Dlo("q1")`,d17 @ a<<8·bb + vmull.p8 q1,`&Dlo("q1")`,d17 @ a<<8·bb vshl.u64 `&Dlo("q3")`,d16,#24 - vmull.p8 q2,`&Dlo("q2")`,d17 @ a<<16·bb + vmull.p8 q2,`&Dlo("q2")`,d17 @ a<<16·bb vshr.u64 `&Dlo("q1")`,#8 - vmull.p8 q3,`&Dlo("q3")`,d17 @ a<<24·bb + vmull.p8 q3,`&Dlo("q3")`,d17 @ a<<24·bb vshl.u64 `&Dhi("q1")`,#24 veor d0,`&Dlo("q1")` vshr.u64 `&Dlo("q2")`,#16 @@ -158,7 +158,7 @@ ___ ################ # void bn_GF2m_mul_2x2(BN_ULONG *r, # BN_ULONG a1,BN_ULONG a0, -# BN_ULONG b1,BN_ULONG b0); # r[3..0]=a1a0·b1b0 +# BN_ULONG b1,BN_ULONG b0); # r[3..0]=a1a0·b1b0 ($A1,$B1,$A0,$B0,$A1B1,$A0B0)=map("d$_",(18..23)); @@ -184,20 +184,20 @@ bn_GF2m_mul_2x2: vmov d16,$A1 vmov d17,$B1 - bl mul_1x1_neon @ a1·b1 + bl mul_1x1_neon @ a1·b1 vmov $A1B1,d0 vmov d16,$A0 vmov d17,$B0 - bl mul_1x1_neon @ a0·b0 + bl mul_1x1_neon @ a0·b0 vmov $A0B0,d0 veor d16,$A0,$A1 veor d17,$B0,$B1 veor $A0,$A0B0,$A1B1 - bl mul_1x1_neon @ (a0+a1)·(b0+b1) + bl mul_1x1_neon @ (a0+a1)·(b0+b1) - veor d0,$A0 @ (a0+a1)·(b0+b1)-a0·b0-a1·b1 + veor d0,$A0 @ (a0+a1)·(b0+b1)-a0·b0-a1·b1 vshl.u64 d1,d0,#32 vshr.u64 d0,d0,#32 veor $A0B0,d1 @@ -220,7 +220,7 @@ $code.=<<___; mov $mask,#7<<2 sub sp,sp,#32 @ allocate tab[8] - bl mul_1x1_ialu @ a1·b1 + bl mul_1x1_ialu @ a1·b1 str $lo,[$ret,#8] str $hi,[$ret,#12] @@ -230,13 +230,13 @@ $code.=<<___; eor r2,r2,$a eor $b,$b,r3 eor $a,$a,r2 - bl mul_1x1_ialu @ a0·b0 + bl mul_1x1_ialu @ a0·b0 str $lo,[$ret] str $hi,[$ret,#4] eor $a,$a,r2 eor $b,$b,r3 - bl mul_1x1_ialu @ (a1+a0)·(b1+b0) + bl mul_1x1_ialu @ (a1+a0)·(b1+b0) ___ @r=map("r$_",(6..9)); $code.=<<___; diff --git a/crypto/openssl/crypto/bn/asm/ia64.S b/crypto/openssl/crypto/bn/asm/ia64.S index 951abc5..a9a42ab 100644 --- a/crypto/openssl/crypto/bn/asm/ia64.S +++ b/crypto/openssl/crypto/bn/asm/ia64.S @@ -422,7 +422,7 @@ bn_mul_add_words: // This loop spins in 3*(n+10) ticks on Itanium and in 2*(n+10) on // Itanium 2. Yes, unlike previous versions it scales:-) Previous -// version was peforming *all* additions in IALU and was starving +// version was performing *all* additions in IALU and was starving // for those even on Itanium 2. In this version one addition is // moved to FPU and is folded with multiplication. This is at cost // of propogating the result from previous call to this subroutine @@ -568,7 +568,7 @@ bn_sqr_comba8: // I've estimated this routine to run in ~120 ticks, but in reality // (i.e. according to ar.itc) it takes ~160 ticks. Are those extra // cycles consumed for instructions fetch? Or did I misinterpret some -// clause in Itanium µ-architecture manual? Comments are welcomed and +// clause in Itanium µ-architecture manual? Comments are welcomed and // highly appreciated. // // On Itanium 2 it takes ~190 ticks. This is because of stalls on diff --git a/crypto/openssl/crypto/bn/asm/s390x-gf2m.pl b/crypto/openssl/crypto/bn/asm/s390x-gf2m.pl index cd9f13e..9d18d40 100755 --- a/crypto/openssl/crypto/bn/asm/s390x-gf2m.pl +++ b/crypto/openssl/crypto/bn/asm/s390x-gf2m.pl @@ -172,19 +172,19 @@ ___ if ($SIZE_T==8) { my @r=map("%r$_",(6..9)); $code.=<<___; - bras $ra,_mul_1x1 # a1·b1 + bras $ra,_mul_1x1 # a1·b1 stmg $lo,$hi,16($rp) lg $a,`$stdframe+128+4*$SIZE_T`($sp) lg $b,`$stdframe+128+6*$SIZE_T`($sp) - bras $ra,_mul_1x1 # a0·b0 + bras $ra,_mul_1x1 # a0·b0 stmg $lo,$hi,0($rp) lg $a,`$stdframe+128+3*$SIZE_T`($sp) lg $b,`$stdframe+128+5*$SIZE_T`($sp) xg $a,`$stdframe+128+4*$SIZE_T`($sp) xg $b,`$stdframe+128+6*$SIZE_T`($sp) - bras $ra,_mul_1x1 # (a0+a1)·(b0+b1) + bras $ra,_mul_1x1 # (a0+a1)·(b0+b1) lmg @r[0],@r[3],0($rp) xgr $lo,$hi diff --git a/crypto/openssl/crypto/bn/asm/x86-gf2m.pl b/crypto/openssl/crypto/bn/asm/x86-gf2m.pl index 808a1e5..b579530 100755 --- a/crypto/openssl/crypto/bn/asm/x86-gf2m.pl +++ b/crypto/openssl/crypto/bn/asm/x86-gf2m.pl @@ -14,7 +14,7 @@ # the time being... Except that it has three code paths: pure integer # code suitable for any x86 CPU, MMX code suitable for PIII and later # and PCLMULQDQ suitable for Westmere and later. Improvement varies -# from one benchmark and µ-arch to another. Below are interval values +# from one benchmark and µ-arch to another. Below are interval values # for 163- and 571-bit ECDH benchmarks relative to compiler-generated # code: # @@ -226,22 +226,22 @@ if ($sse2) { &push ("edi"); &mov ($a,&wparam(1)); &mov ($b,&wparam(3)); - &call ("_mul_1x1_mmx"); # a1·b1 + &call ("_mul_1x1_mmx"); # a1·b1 &movq ("mm7",$R); &mov ($a,&wparam(2)); &mov ($b,&wparam(4)); - &call ("_mul_1x1_mmx"); # a0·b0 + &call ("_mul_1x1_mmx"); # a0·b0 &movq ("mm6",$R); &mov ($a,&wparam(1)); &mov ($b,&wparam(3)); &xor ($a,&wparam(2)); &xor ($b,&wparam(4)); - &call ("_mul_1x1_mmx"); # (a0+a1)·(b0+b1) + &call ("_mul_1x1_mmx"); # (a0+a1)·(b0+b1) &pxor ($R,"mm7"); &mov ($a,&wparam(0)); - &pxor ($R,"mm6"); # (a0+a1)·(b0+b1)-a1·b1-a0·b0 + &pxor ($R,"mm6"); # (a0+a1)·(b0+b1)-a1·b1-a0·b0 &movq ($A,$R); &psllq ($R,32); @@ -266,13 +266,13 @@ if ($sse2) { &mov ($a,&wparam(1)); &mov ($b,&wparam(3)); - &call ("_mul_1x1_ialu"); # a1·b1 + &call ("_mul_1x1_ialu"); # a1·b1 &mov (&DWP(8,"esp"),$lo); &mov (&DWP(12,"esp"),$hi); &mov ($a,&wparam(2)); &mov ($b,&wparam(4)); - &call ("_mul_1x1_ialu"); # a0·b0 + &call ("_mul_1x1_ialu"); # a0·b0 &mov (&DWP(0,"esp"),$lo); &mov (&DWP(4,"esp"),$hi); @@ -280,7 +280,7 @@ if ($sse2) { &mov ($b,&wparam(3)); &xor ($a,&wparam(2)); &xor ($b,&wparam(4)); - &call ("_mul_1x1_ialu"); # (a0+a1)·(b0+b1) + &call ("_mul_1x1_ialu"); # (a0+a1)·(b0+b1) &mov ("ebp",&wparam(0)); @r=("ebx","ecx","edi","esi"); diff --git a/crypto/openssl/crypto/bn/asm/x86_64-gcc.c b/crypto/openssl/crypto/bn/asm/x86_64-gcc.c index 9c5074b..0a5bb28 100644 --- a/crypto/openssl/crypto/bn/asm/x86_64-gcc.c +++ b/crypto/openssl/crypto/bn/asm/x86_64-gcc.c @@ -66,7 +66,7 @@ # undef sqr /*- - * "m"(a), "+m"(r) is the way to favor DirectPath µ-code; + * "m"(a), "+m"(r) is the way to favor DirectPath µ-code; * "g"(0) let the compiler to decide where does it * want to keep the value of zero; */ diff --git a/crypto/openssl/crypto/bn/asm/x86_64-gf2m.pl b/crypto/openssl/crypto/bn/asm/x86_64-gf2m.pl index 226c66c..42bbec2 100755 --- a/crypto/openssl/crypto/bn/asm/x86_64-gf2m.pl +++ b/crypto/openssl/crypto/bn/asm/x86_64-gf2m.pl @@ -13,7 +13,7 @@ # in bn_gf2m.c. It's kind of low-hanging mechanical port from C for # the time being... Except that it has two code paths: code suitable # for any x86_64 CPU and PCLMULQDQ one suitable for Westmere and -# later. Improvement varies from one benchmark and µ-arch to another. +# later. Improvement varies from one benchmark and µ-arch to another. # Vanilla code path is at most 20% faster than compiler-generated code # [not very impressive], while PCLMULQDQ - whole 85%-160% better on # 163- and 571-bit ECDH benchmarks on Intel CPUs. Keep in mind that @@ -184,13 +184,13 @@ ___ $code.=<<___; movdqa %xmm0,%xmm4 movdqa %xmm1,%xmm5 - pclmulqdq \$0,%xmm1,%xmm0 # a1·b1 + pclmulqdq \$0,%xmm1,%xmm0 # a1·b1 pxor %xmm2,%xmm4 pxor %xmm3,%xmm5 - pclmulqdq \$0,%xmm3,%xmm2 # a0·b0 - pclmulqdq \$0,%xmm5,%xmm4 # (a0+a1)·(b0+b1) + pclmulqdq \$0,%xmm3,%xmm2 # a0·b0 + pclmulqdq \$0,%xmm5,%xmm4 # (a0+a1)·(b0+b1) xorps %xmm0,%xmm4 - xorps %xmm2,%xmm4 # (a0+a1)·(b0+b1)-a0·b0-a1·b1 + xorps %xmm2,%xmm4 # (a0+a1)·(b0+b1)-a0·b0-a1·b1 movdqa %xmm4,%xmm5 pslldq \$8,%xmm4 psrldq \$8,%xmm5 @@ -225,13 +225,13 @@ $code.=<<___; mov \$0xf,$mask mov $a1,$a mov $b1,$b - call _mul_1x1 # a1·b1 + call _mul_1x1 # a1·b1 mov $lo,16(%rsp) mov $hi,24(%rsp) mov 48(%rsp),$a mov 64(%rsp),$b - call _mul_1x1 # a0·b0 + call _mul_1x1 # a0·b0 mov $lo,0(%rsp) mov $hi,8(%rsp) @@ -239,7 +239,7 @@ $code.=<<___; mov 56(%rsp),$b xor 48(%rsp),$a xor 64(%rsp),$b - call _mul_1x1 # (a0+a1)·(b0+b1) + call _mul_1x1 # (a0+a1)·(b0+b1) ___ @r=("%rbx","%rcx","%rdi","%rsi"); $code.=<<___; diff --git a/crypto/openssl/crypto/bn/bn_exp.c b/crypto/openssl/crypto/bn/bn_exp.c index 27146c8..7e33ba9 100644 --- a/crypto/openssl/crypto/bn/bn_exp.c +++ b/crypto/openssl/crypto/bn/bn_exp.c @@ -599,12 +599,13 @@ int BN_mod_exp_mont_consttime(BIGNUM *rr, const BIGNUM *a, const BIGNUM *p, bn_check_top(p); bn_check_top(m); - top = m->top; - - if (!(m->d[0] & 1)) { + if (!BN_is_odd(m)) { BNerr(BN_F_BN_MOD_EXP_MONT_CONSTTIME, BN_R_CALLED_WITH_EVEN_MODULUS); return (0); } + + top = m->top; + bits = BN_num_bits(p); if (bits == 0) { ret = BN_one(rr); diff --git a/crypto/openssl/crypto/bn/bn_gcd.c b/crypto/openssl/crypto/bn/bn_gcd.c index 97c55ab..ce59fe7 100644 --- a/crypto/openssl/crypto/bn/bn_gcd.c +++ b/crypto/openssl/crypto/bn/bn_gcd.c @@ -583,6 +583,7 @@ static BIGNUM *BN_mod_inverse_no_branch(BIGNUM *in, * BN_div_no_branch will be called eventually. */ pB = &local_B; + local_B.flags = 0; BN_with_flags(pB, B, BN_FLG_CONSTTIME); if (!BN_nnmod(B, pB, A, ctx)) goto err; @@ -610,6 +611,7 @@ static BIGNUM *BN_mod_inverse_no_branch(BIGNUM *in, * BN_div_no_branch will be called eventually. */ pA = &local_A; + local_A.flags = 0; BN_with_flags(pA, A, BN_FLG_CONSTTIME); /* (D, M) := (A/B, A%B) ... */ diff --git a/crypto/openssl/crypto/bn/bn_gf2m.c b/crypto/openssl/crypto/bn/bn_gf2m.c index a0ba8de..8ad44b4 100644 --- a/crypto/openssl/crypto/bn/bn_gf2m.c +++ b/crypto/openssl/crypto/bn/bn_gf2m.c @@ -576,7 +576,7 @@ int BN_GF2m_mod_sqr_arr(BIGNUM *r, const BIGNUM *a, const int p[], bn_check_top(a); BN_CTX_start(ctx); if ((s = BN_CTX_get(ctx)) == NULL) - return 0; + goto err; if (!bn_wexpand(s, 2 * a->top)) goto err; @@ -700,18 +700,21 @@ int BN_GF2m_mod_inv(BIGNUM *r, const BIGNUM *a, const BIGNUM *p, BN_CTX *ctx) int top = p->top; BN_ULONG *udp, *bdp, *vdp, *cdp; - bn_wexpand(u, top); + if (!bn_wexpand(u, top)) + goto err; udp = u->d; for (i = u->top; i < top; i++) udp[i] = 0; u->top = top; - bn_wexpand(b, top); + if (!bn_wexpand(b, top)) + goto err; bdp = b->d; bdp[0] = 1; for (i = 1; i < top; i++) bdp[i] = 0; b->top = top; - bn_wexpand(c, top); + if (!bn_wexpand(c, top)) + goto err; cdp = c->d; for (i = 0; i < top; i++) cdp[i] = 0; diff --git a/crypto/openssl/crypto/bn/bn_mont.c b/crypto/openssl/crypto/bn/bn_mont.c index aadd5db..be95bd55 100644 --- a/crypto/openssl/crypto/bn/bn_mont.c +++ b/crypto/openssl/crypto/bn/bn_mont.c @@ -361,9 +361,9 @@ void BN_MONT_CTX_free(BN_MONT_CTX *mont) if (mont == NULL) return; - BN_free(&(mont->RR)); - BN_free(&(mont->N)); - BN_free(&(mont->Ni)); + BN_clear_free(&(mont->RR)); + BN_clear_free(&(mont->N)); + BN_clear_free(&(mont->Ni)); if (mont->flags & BN_FLG_MALLOCED) OPENSSL_free(mont); } @@ -373,6 +373,9 @@ int BN_MONT_CTX_set(BN_MONT_CTX *mont, const BIGNUM *mod, BN_CTX *ctx) int ret = 0; BIGNUM *Ri, *R; + if (BN_is_zero(mod)) + return 0; + BN_CTX_start(ctx); if ((Ri = BN_CTX_get(ctx)) == NULL) goto err; diff --git a/crypto/openssl/crypto/bn/bn_recp.c b/crypto/openssl/crypto/bn/bn_recp.c index 6826f93..7497ac6 100644 --- a/crypto/openssl/crypto/bn/bn_recp.c +++ b/crypto/openssl/crypto/bn/bn_recp.c @@ -152,8 +152,10 @@ int BN_div_recp(BIGNUM *dv, BIGNUM *rem, const BIGNUM *m, if (BN_ucmp(m, &(recp->N)) < 0) { BN_zero(d); - if (!BN_copy(r, m)) + if (!BN_copy(r, m)) { + BN_CTX_end(ctx); return 0; + } BN_CTX_end(ctx); return (1); } diff --git a/crypto/openssl/crypto/bn/bn_x931p.c b/crypto/openssl/crypto/bn/bn_x931p.c index 6d76b12..efa48bd 100644 --- a/crypto/openssl/crypto/bn/bn_x931p.c +++ b/crypto/openssl/crypto/bn/bn_x931p.c @@ -213,14 +213,14 @@ int BN_X931_generate_Xpq(BIGNUM *Xp, BIGNUM *Xq, int nbits, BN_CTX *ctx) * exceeded. */ if (!BN_rand(Xp, nbits, 1, 0)) - return 0; + goto err; BN_CTX_start(ctx); t = BN_CTX_get(ctx); for (i = 0; i < 1000; i++) { if (!BN_rand(Xq, nbits, 1, 0)) - return 0; + goto err; /* Check that |Xp - Xq| > 2^(nbits - 100) */ BN_sub(t, Xp, Xq); if (BN_num_bits(t) > (nbits - 100)) @@ -234,6 +234,9 @@ int BN_X931_generate_Xpq(BIGNUM *Xp, BIGNUM *Xq, int nbits, BN_CTX *ctx) return 0; + err: + BN_CTX_end(ctx); + return 0; } /* diff --git a/crypto/openssl/crypto/bn/bntest.c b/crypto/openssl/crypto/bn/bntest.c index 06662c5..6d55049 100644 --- a/crypto/openssl/crypto/bn/bntest.c +++ b/crypto/openssl/crypto/bn/bntest.c @@ -441,6 +441,14 @@ int test_div(BIO *bp, BN_CTX *ctx) BN_init(&d); BN_init(&e); + BN_one(&a); + BN_zero(&b); + + if (BN_div(&d, &c, &a, &b, ctx)) { + fprintf(stderr, "Division by zero succeeded!\n"); + return 0; + } + for (i = 0; i < num0 + num1; i++) { if (i < num1) { BN_bntest_rand(&a, 400, 0, 0); @@ -516,9 +524,9 @@ int test_div_word(BIO *bp) do { BN_bntest_rand(&a, 512, -1, 0); BN_bntest_rand(&b, BN_BITS2, -1, 0); - s = b.d[0]; - } while (!s); + } while (BN_is_zero(&b)); + s = b.d[0]; BN_copy(&b, &a); r = BN_div_word(&b, s); @@ -781,6 +789,18 @@ int test_mont(BIO *bp, BN_CTX *ctx) if (mont == NULL) return 0; + BN_zero(&n); + if (BN_MONT_CTX_set(mont, &n, ctx)) { + fprintf(stderr, "BN_MONT_CTX_set succeeded for zero modulus!\n"); + return 0; + } + + BN_set_word(&n, 16); + if (BN_MONT_CTX_set(mont, &n, ctx)) { + fprintf(stderr, "BN_MONT_CTX_set succeeded for even modulus!\n"); + return 0; + } + BN_bntest_rand(&a, 100, 0, 0); BN_bntest_rand(&b, 100, 0, 0); for (i = 0; i < num2; i++) { @@ -887,6 +907,14 @@ int test_mod_mul(BIO *bp, BN_CTX *ctx) d = BN_new(); e = BN_new(); + BN_one(a); + BN_one(b); + BN_zero(c); + if (BN_mod_mul(e, a, b, c, ctx)) { + fprintf(stderr, "BN_mod_mul with zero modulus succeeded!\n"); + return 0; + } + for (j = 0; j < 3; j++) { BN_bntest_rand(c, 1024, 0, 0); for (i = 0; i < num0; i++) { @@ -952,6 +980,14 @@ int test_mod_exp(BIO *bp, BN_CTX *ctx) d = BN_new(); e = BN_new(); + BN_one(a); + BN_one(b); + BN_zero(c); + if (BN_mod_exp(d, a, b, c, ctx)) { + fprintf(stderr, "BN_mod_exp with zero modulus succeeded!\n"); + return 0; + } + BN_bntest_rand(c, 30, 0, 1); /* must be odd for montgomery */ for (i = 0; i < num2; i++) { BN_bntest_rand(a, 20 + i * 5, 0, 0); @@ -999,6 +1035,22 @@ int test_mod_exp_mont_consttime(BIO *bp, BN_CTX *ctx) d = BN_new(); e = BN_new(); + BN_one(a); + BN_one(b); + BN_zero(c); + if (BN_mod_exp_mont_consttime(d, a, b, c, ctx, NULL)) { + fprintf(stderr, "BN_mod_exp_mont_consttime with zero modulus " + "succeeded\n"); + return 0; + } + + BN_set_word(c, 16); + if (BN_mod_exp_mont_consttime(d, a, b, c, ctx, NULL)) { + fprintf(stderr, "BN_mod_exp_mont_consttime with even modulus " + "succeeded\n"); + return 0; + } + BN_bntest_rand(c, 30, 0, 1); /* must be odd for montgomery */ for (i = 0; i < num2; i++) { BN_bntest_rand(a, 20 + i * 5, 0, 0); |