diff options
Diffstat (limited to 'crypto/bn/asm/x86_64-gcc.c')
| -rw-r--r-- | crypto/bn/asm/x86_64-gcc.c | 34 |
1 files changed, 23 insertions, 11 deletions
diff --git a/crypto/bn/asm/x86_64-gcc.c b/crypto/bn/asm/x86_64-gcc.c index d13ec5a468..a0c49ee978 100644 --- a/crypto/bn/asm/x86_64-gcc.c +++ b/crypto/bn/asm/x86_64-gcc.c @@ -1,3 +1,7 @@ +#include "../bn_lcl.h" +#if !(defined(__GNUC__) && __GNUC__>=2) +# include "../bn_asm.c" /* kind of dirty hack for Sun Studio */ +#else /* * x86_64 BIGNUM accelerator version 0.1, December 2002. * @@ -51,7 +55,14 @@ * machine. */ +#ifdef _WIN64 +#define BN_ULONG unsigned long long +#else #define BN_ULONG unsigned long +#endif + +#undef mul +#undef mul_add /* * "m"(a), "+m"(r) is the way to favor DirectPath µ-code; @@ -94,7 +105,7 @@ : "a"(a) \ : "cc"); -BN_ULONG bn_mul_add_words(BN_ULONG *rp, BN_ULONG *ap, int num, BN_ULONG w) +BN_ULONG bn_mul_add_words(BN_ULONG *rp, const BN_ULONG *ap, int num, BN_ULONG w) { BN_ULONG c1=0; @@ -118,7 +129,7 @@ BN_ULONG bn_mul_add_words(BN_ULONG *rp, BN_ULONG *ap, int num, BN_ULONG w) return(c1); } -BN_ULONG bn_mul_words(BN_ULONG *rp, BN_ULONG *ap, int num, BN_ULONG w) +BN_ULONG bn_mul_words(BN_ULONG *rp, const BN_ULONG *ap, int num, BN_ULONG w) { BN_ULONG c1=0; @@ -141,7 +152,7 @@ BN_ULONG bn_mul_words(BN_ULONG *rp, BN_ULONG *ap, int num, BN_ULONG w) return(c1); } -void bn_sqr_words(BN_ULONG *r, BN_ULONG *a, int n) +void bn_sqr_words(BN_ULONG *r, const BN_ULONG *a, int n) { if (n <= 0) return; @@ -172,21 +183,21 @@ BN_ULONG bn_div_words(BN_ULONG h, BN_ULONG l, BN_ULONG d) return ret; } -BN_ULONG bn_add_words (BN_ULONG *rp, BN_ULONG *ap, BN_ULONG *bp,int n) +BN_ULONG bn_add_words (BN_ULONG *rp, const BN_ULONG *ap, const BN_ULONG *bp,int n) { BN_ULONG ret=0,i=0; if (n <= 0) return 0; asm ( " subq %2,%2 \n" - ".align 16 \n" + ".p2align 4 \n" "1: movq (%4,%2,8),%0 \n" " adcq (%5,%2,8),%0 \n" " movq %0,(%3,%2,8) \n" " leaq 1(%2),%2 \n" " loop 1b \n" " sbbq %0,%0 \n" - : "+a"(ret),"+c"(n),"+r"(i) + : "=&a"(ret),"+c"(n),"=&r"(i) : "r"(rp),"r"(ap),"r"(bp) : "cc" ); @@ -195,21 +206,21 @@ BN_ULONG bn_add_words (BN_ULONG *rp, BN_ULONG *ap, BN_ULONG *bp,int n) } #ifndef SIMICS -BN_ULONG bn_sub_words (BN_ULONG *rp, BN_ULONG *ap, BN_ULONG *bp,int n) +BN_ULONG bn_sub_words (BN_ULONG *rp, const BN_ULONG *ap, const BN_ULONG *bp,int n) { BN_ULONG ret=0,i=0; if (n <= 0) return 0; asm ( " subq %2,%2 \n" - ".align 16 \n" + ".p2align 4 \n" "1: movq (%4,%2,8),%0 \n" " sbbq (%5,%2,8),%0 \n" " movq %0,(%3,%2,8) \n" " leaq 1(%2),%2 \n" " loop 1b \n" " sbbq %0,%0 \n" - : "+a"(ret),"+c"(n),"+r"(i) + : "=&a"(ret),"+c"(n),"=&r"(i) : "r"(rp),"r"(ap),"r"(bp) : "cc" ); @@ -482,7 +493,7 @@ void bn_mul_comba4(BN_ULONG *r, BN_ULONG *a, BN_ULONG *b) r[7]=c2; } -void bn_sqr_comba8(BN_ULONG *r, BN_ULONG *a) +void bn_sqr_comba8(BN_ULONG *r, const BN_ULONG *a) { BN_ULONG t1,t2; BN_ULONG c1,c2,c3; @@ -558,7 +569,7 @@ void bn_sqr_comba8(BN_ULONG *r, BN_ULONG *a) r[15]=c1; } -void bn_sqr_comba4(BN_ULONG *r, BN_ULONG *a) +void bn_sqr_comba4(BN_ULONG *r, const BN_ULONG *a) { BN_ULONG t1,t2; BN_ULONG c1,c2,c3; @@ -591,3 +602,4 @@ void bn_sqr_comba4(BN_ULONG *r, BN_ULONG *a) r[6]=c1; r[7]=c2; } +#endif |