X-Git-Url: https://git.librecmc.org/?a=blobdiff_plain;f=crypto%2Fsha%2Fsha512.c;h=aab1f1e8d457931fac527beddd9cdc74c6a7ff6c;hb=32cf5baeae21774db04af2ad2d74567a617c6fec;hp=22d916849755099c0dba1365338f883895178ae3;hpb=63077bd40c67eb1d6be31f0df60470c227b6894e;p=oweals%2Fopenssl.git diff --git a/crypto/sha/sha512.c b/crypto/sha/sha512.c index 22d9168497..aab1f1e8d4 100644 --- a/crypto/sha/sha512.c +++ b/crypto/sha/sha512.c @@ -4,6 +4,8 @@ * according to the OpenSSL license [found in ../../LICENSE]. * ==================================================================== */ +#include +#if !defined(OPENSSL_NO_SHA) && !defined(OPENSSL_NO_SHA512) /* * IMPLEMENTATION NOTES. * @@ -42,14 +44,19 @@ #include #include -#include #include #include #include -const char *SHA512_version="SHA-512" OPENSSL_VERSION_PTEXT; +#include "cryptlib.h" + +__fips_constseg +const char SHA512_version[]="SHA-512" OPENSSL_VERSION_PTEXT; -#if defined(_M_IX86) || defined(_M_AMD64) || defined(__i386) || defined(__x86_64) +#if defined(__i386) || defined(__i386__) || defined(_M_IX86) || \ + defined(__x86_64) || defined(_M_AMD64) || defined(_M_X64) || \ + defined(__s390__) || defined(__s390x__) || \ + defined(SHA512_ASM) #define SHA512_BLOCK_CAN_MANAGE_UNALIGNED_DATA #endif @@ -63,8 +70,9 @@ int SHA384_Init (SHA512_CTX *c) c->h[5]=U64(0x8eb44a8768581511); c->h[6]=U64(0xdb0c2e0d64f98fa7); c->h[7]=U64(0x47b5481dbefa4fa4); + c->Nl=0; c->Nh=0; - c->num=0; + c->num=0; c->md_len=SHA384_DIGEST_LENGTH; return 1; } @@ -78,73 +86,103 @@ int SHA512_Init (SHA512_CTX *c) c->h[5]=U64(0x9b05688c2b3e6c1f); c->h[6]=U64(0x1f83d9abfb41bd6b); c->h[7]=U64(0x5be0cd19137e2179); + c->Nl=0; c->Nh=0; - c->num=0; + c->num=0; c->md_len=SHA512_DIGEST_LENGTH; return 1; } -static void sha512_block (SHA512_CTX *ctx, const void *in, size_t num); +#ifndef SHA512_ASM +static +#endif +void sha512_block_data_order (SHA512_CTX *ctx, const void *in, size_t num); -static int sha512_final (unsigned char *md, SHA512_CTX *c, size_t msz) +int SHA512_Final (unsigned char *md, SHA512_CTX *c) { unsigned char *p=(unsigned char *)c->u.p; size_t n=c->num; - p[n]=0x80; + p[n]=0x80; /* There always is a room for one */ n++; if (n > (sizeof(c->u)-16)) memset (p+n,0,sizeof(c->u)-n), n=0, - sha512_block (c,p,1); + sha512_block_data_order (c,p,1); memset (p+n,0,sizeof(c->u)-16-n); #ifdef B_ENDIAN c->u.d[SHA_LBLOCK-2] = c->Nh; c->u.d[SHA_LBLOCK-1] = c->Nl; #else - p[sizeof(c->u)-1] = (c->Nl)&0xFF; - p[sizeof(c->u)-2] = (c->Nl>>8)&0xFF; - p[sizeof(c->u)-3] = (c->Nl>>16)&0xFF; - p[sizeof(c->u)-4] = (c->Nl>>24)&0xFF; - p[sizeof(c->u)-5] = (c->Nl>>32)&0xFF; - p[sizeof(c->u)-6] = (c->Nl>>40)&0xFF; - p[sizeof(c->u)-7] = (c->Nl>>48)&0xFF; - p[sizeof(c->u)-8] = (c->Nl>>56)&0xFF; - p[sizeof(c->u)-9] = (c->Nh)&0xFF; - p[sizeof(c->u)-10] = (c->Nh>>8)&0xFF; - p[sizeof(c->u)-11] = (c->Nh>>16)&0xFF; - p[sizeof(c->u)-12] = (c->Nh>>24)&0xFF; - p[sizeof(c->u)-13] = (c->Nh>>32)&0xFF; - p[sizeof(c->u)-14] = (c->Nh>>40)&0xFF; - p[sizeof(c->u)-15] = (c->Nh>>48)&0xFF; - p[sizeof(c->u)-16] = (c->Nh>>56)&0xFF; + p[sizeof(c->u)-1] = (unsigned char)(c->Nl); + p[sizeof(c->u)-2] = (unsigned char)(c->Nl>>8); + p[sizeof(c->u)-3] = (unsigned char)(c->Nl>>16); + p[sizeof(c->u)-4] = (unsigned char)(c->Nl>>24); + p[sizeof(c->u)-5] = (unsigned char)(c->Nl>>32); + p[sizeof(c->u)-6] = (unsigned char)(c->Nl>>40); + p[sizeof(c->u)-7] = (unsigned char)(c->Nl>>48); + p[sizeof(c->u)-8] = (unsigned char)(c->Nl>>56); + p[sizeof(c->u)-9] = (unsigned char)(c->Nh); + p[sizeof(c->u)-10] = (unsigned char)(c->Nh>>8); + p[sizeof(c->u)-11] = (unsigned char)(c->Nh>>16); + p[sizeof(c->u)-12] = (unsigned char)(c->Nh>>24); + p[sizeof(c->u)-13] = (unsigned char)(c->Nh>>32); + p[sizeof(c->u)-14] = (unsigned char)(c->Nh>>40); + p[sizeof(c->u)-15] = (unsigned char)(c->Nh>>48); + p[sizeof(c->u)-16] = (unsigned char)(c->Nh>>56); #endif - sha512_block (c,p,1); + sha512_block_data_order (c,p,1); if (md==0) return 0; - for (n=0;msz>0;n++,msz-=8) + switch (c->md_len) { - SHA_LONG64 t = c->h[n]; - - *(md++) = (t>>56)&0xFF; *(md++) = (t>>48)&0xFF; - *(md++) = (t>>40)&0xFF; *(md++) = (t>>32)&0xFF; - *(md++) = (t>>24)&0xFF; *(md++) = (t>>16)&0xFF; - *(md++) = (t>>8)&0xFF; *(md++) = (t)&0xFF; + /* Let compiler decide if it's appropriate to unroll... */ + case SHA384_DIGEST_LENGTH: + for (n=0;nh[n]; + + *(md++) = (unsigned char)(t>>56); + *(md++) = (unsigned char)(t>>48); + *(md++) = (unsigned char)(t>>40); + *(md++) = (unsigned char)(t>>32); + *(md++) = (unsigned char)(t>>24); + *(md++) = (unsigned char)(t>>16); + *(md++) = (unsigned char)(t>>8); + *(md++) = (unsigned char)(t); + } + break; + case SHA512_DIGEST_LENGTH: + for (n=0;nh[n]; + + *(md++) = (unsigned char)(t>>56); + *(md++) = (unsigned char)(t>>48); + *(md++) = (unsigned char)(t>>40); + *(md++) = (unsigned char)(t>>32); + *(md++) = (unsigned char)(t>>24); + *(md++) = (unsigned char)(t>>16); + *(md++) = (unsigned char)(t>>8); + *(md++) = (unsigned char)(t); + } + break; + /* ... as well as make sure md_len is not abused. */ + default: return 0; } return 1; } int SHA384_Final (unsigned char *md,SHA512_CTX *c) -{ return sha512_final (md,c,SHA384_DIGEST_LENGTH); } -int SHA512_Final (unsigned char *md,SHA512_CTX *c) -{ return sha512_final (md,c,SHA512_DIGEST_LENGTH); } +{ return SHA512_Final (md,c); } int SHA512_Update (SHA512_CTX *c, const void *_data, size_t len) { SHA_LONG64 l; - unsigned char *p=c->u.p,*data=(unsigned char *)_data; + unsigned char *p=c->u.p; + const unsigned char *data=(const unsigned char *)_data; if (len==0) return 1; @@ -159,13 +197,13 @@ int SHA512_Update (SHA512_CTX *c, const void *_data, size_t len) if (len < n) { - memcpy (p+c->num,data,len), c->num += len; + memcpy (p+c->num,data,len), c->num += (unsigned int)len; return 1; } else { memcpy (p+c->num,data,n), c->num = 0; len-=n, data+=n; - sha512_block (c,p,1); + sha512_block_data_order (c,p,1); } } @@ -175,12 +213,12 @@ int SHA512_Update (SHA512_CTX *c, const void *_data, size_t len) if ((size_t)data%sizeof(c->u.d[0]) != 0) while (len >= sizeof(c->u)) memcpy (p,data,sizeof(c->u)), - sha512_block (c,p,1), + sha512_block_data_order (c,p,1), len -= sizeof(c->u), data += sizeof(c->u); else #endif - sha512_block (c,data,len/sizeof(c->u)), + sha512_block_data_order (c,data,len/sizeof(c->u)), data += len, len %= sizeof(c->u), data -= len; @@ -195,7 +233,7 @@ int SHA384_Update (SHA512_CTX *c, const void *data, size_t len) { return SHA512_Update (c,data,len); } void SHA512_Transform (SHA512_CTX *c, const unsigned char *data) -{ sha512_block (c,data,1); } +{ sha512_block_data_order (c,data,1); } unsigned char *SHA384(const unsigned char *d, size_t n, unsigned char *md) { @@ -205,7 +243,7 @@ unsigned char *SHA384(const unsigned char *d, size_t n, unsigned char *md) if (md == NULL) md=m; SHA384_Init(&c); SHA512_Update(&c,d,n); - sha512_final(md,&c,sizeof(m)); + SHA512_Final(md,&c); OPENSSL_cleanse(&c,sizeof(c)); return(md); } @@ -218,11 +256,13 @@ unsigned char *SHA512(const unsigned char *d, size_t n, unsigned char *md) if (md == NULL) md=m; SHA512_Init(&c); SHA512_Update(&c,d,n); - sha512_final(md,&c,sizeof(m)); + SHA512_Final(md,&c); OPENSSL_cleanse(&c,sizeof(c)); return(md); } +#ifndef SHA512_ASM +__fips_constseg static const SHA_LONG64 K512[80] = { U64(0x428a2f98d728ae22),U64(0x7137449123ef65cd), U64(0xb5c0fbcfec4d3b2f),U64(0xe9b5dba58189dbbc), @@ -268,40 +308,79 @@ static const SHA_LONG64 K512[80] = { #ifndef PEDANTIC # if defined(__GNUC__) && __GNUC__>=2 && !defined(OPENSSL_NO_ASM) && !defined(OPENSSL_NO_INLINE_ASM) # if defined(__x86_64) || defined(__x86_64__) -# define PULL64(x) ({ SHA_LONG64 ret=*((SHA_LONG64 *)(&(x))); \ - asm ("bswapq %0" \ - : "=r"(ret) \ - : "0"(ret)); ret; }) -# endif -# endif -#endif - -#ifndef PULL64 -#define B(x,j) (((SHA_LONG64)(*(((unsigned char *)(&x))+j)))<<((7-j)*8)) -#define PULL64(x) (B(x,0)|B(x,1)|B(x,2)|B(x,3)|B(x,4)|B(x,5)|B(x,6)|B(x,7)) -#endif - -#ifndef PEDANTIC -# if defined(_MSC_VER) -# if defined(_WIN64) /* applies to both IA-64 and AMD64 */ -# define ROTR(a,n) _rotr64((a),n) -# endif -# elif defined(__GNUC__) && __GNUC__>=2 && !defined(OPENSSL_NO_ASM) && !defined(OPENSSL_NO_INLINE_ASM) -# if defined(__x86_64) || defined(__x86_64__) -# define ROTR(a,n) ({ unsigned long ret; \ +# define ROTR(a,n) ({ SHA_LONG64 ret; \ asm ("rorq %1,%0" \ : "=r"(ret) \ : "J"(n),"0"(a) \ : "cc"); ret; }) -# elif defined(_ARCH_PPC) && defined(__64BIT__) -# define ROTR(a,n) ({ unsigned long ret; \ +# if !defined(B_ENDIAN) +# define PULL64(x) ({ SHA_LONG64 ret=*((const SHA_LONG64 *)(&(x))); \ + asm ("bswapq %0" \ + : "=r"(ret) \ + : "0"(ret)); ret; }) +# endif +# elif (defined(__i386) || defined(__i386__)) && !defined(B_ENDIAN) +# if defined(I386_ONLY) +# define PULL64(x) ({ const unsigned int *p=(const unsigned int *)(&(x));\ + unsigned int hi=p[0],lo=p[1]; \ + asm("xchgb %%ah,%%al;xchgb %%dh,%%dl;"\ + "roll $16,%%eax; roll $16,%%edx; "\ + "xchgb %%ah,%%al;xchgb %%dh,%%dl;" \ + : "=a"(lo),"=d"(hi) \ + : "0"(lo),"1"(hi) : "cc"); \ + ((SHA_LONG64)hi)<<32|lo; }) +# else +# define PULL64(x) ({ const unsigned int *p=(const unsigned int *)(&(x));\ + unsigned int hi=p[0],lo=p[1]; \ + asm ("bswapl %0; bswapl %1;" \ + : "=r"(lo),"=r"(hi) \ + : "0"(lo),"1"(hi)); \ + ((SHA_LONG64)hi)<<32|lo; }) +# endif +# elif (defined(_ARCH_PPC) && defined(__64BIT__)) || defined(_ARCH_PPC64) +# define ROTR(a,n) ({ SHA_LONG64 ret; \ asm ("rotrdi %0,%1,%2" \ : "=r"(ret) \ : "r"(a),"K"(n)); ret; }) # endif +# elif defined(_MSC_VER) +# if defined(_WIN64) /* applies to both IA-64 and AMD64 */ +# pragma intrinsic(_rotr64) +# define ROTR(a,n) _rotr64((a),n) +# endif +# if defined(_M_IX86) && !defined(OPENSSL_NO_ASM) && !defined(OPENSSL_NO_INLINE_ASM) +# if defined(I386_ONLY) + static SHA_LONG64 __fastcall __pull64be(const void *x) + { _asm mov edx, [ecx + 0] + _asm mov eax, [ecx + 4] + _asm xchg dh,dl + _asm xchg ah,al + _asm rol edx,16 + _asm rol eax,16 + _asm xchg dh,dl + _asm xchg ah,al + } +# else + static SHA_LONG64 __fastcall __pull64be(const void *x) + { _asm mov edx, [ecx + 0] + _asm mov eax, [ecx + 4] + _asm bswap edx + _asm bswap eax + } +# endif +# define PULL64(x) __pull64be(&(x)) +# if _MSC_VER<=1200 +# pragma inline_depth(0) +# endif +# endif # endif #endif +#ifndef PULL64 +#define B(x,j) (((SHA_LONG64)(*(((const unsigned char *)(&x))+j)))<<((7-j)*8)) +#define PULL64(x) (B(x,0)|B(x,1)|B(x,2)|B(x,3)|B(x,4)|B(x,5)|B(x,6)|B(x,7)) +#endif + #ifndef ROTR #define ROTR(x,s) (((x)>>s) | (x)<<(64-s)) #endif @@ -314,28 +393,74 @@ static const SHA_LONG64 K512[80] = { #define Ch(x,y,z) (((x) & (y)) ^ ((~(x)) & (z))) #define Maj(x,y,z) (((x) & (y)) ^ ((x) & (z)) ^ ((y) & (z))) -#if defined(OPENSSL_IA32_SSE2) && !defined(OPENSSL_NO_ASM) -#define GO_FOR_SSE2(ctx,in,num) do { \ - extern int OPENSSL_ia32cap; \ - void sha512_block_sse2(void *,const void *,size_t); \ - if (!(OPENSSL_ia32cap & (1<<26))) break; \ - sha512_block_sse2(ctx->h,in,num); return; \ - } while (0) + +#if defined(__i386) || defined(__i386__) || defined(_M_IX86) +/* + * This code should give better results on 32-bit CPU with less than + * ~24 registers, both size and performance wise... + */ +static void sha512_block_data_order (SHA512_CTX *ctx, const void *in, size_t num) + { + const SHA_LONG64 *W=in; + SHA_LONG64 A,E,T; + SHA_LONG64 X[9+80],*F; + int i; + + while (num--) { + + F = X+80; + A = ctx->h[0]; F[1] = ctx->h[1]; + F[2] = ctx->h[2]; F[3] = ctx->h[3]; + E = ctx->h[4]; F[5] = ctx->h[5]; + F[6] = ctx->h[6]; F[7] = ctx->h[7]; + + for (i=0;i<16;i++,F--) + { +#ifdef B_ENDIAN + T = W[i]; +#else + T = PULL64(W[i]); #endif + F[0] = A; + F[4] = E; + F[8] = T; + T += F[7] + Sigma1(E) + Ch(E,F[5],F[6]) + K512[i]; + E = F[3] + T; + A = T + Sigma0(A) + Maj(A,F[1],F[2]); + } + + for (;i<80;i++,F--) + { + T = sigma0(F[8+16-1]); + T += sigma1(F[8+16-14]); + T += F[8+16] + F[8+16-9]; + + F[0] = A; + F[4] = E; + F[8] = T; + T += F[7] + Sigma1(E) + Ch(E,F[5],F[6]) + K512[i]; + E = F[3] + T; + A = T + Sigma0(A) + Maj(A,F[1],F[2]); + } + + ctx->h[0] += A; ctx->h[1] += F[1]; + ctx->h[2] += F[2]; ctx->h[3] += F[3]; + ctx->h[4] += E; ctx->h[5] += F[5]; + ctx->h[6] += F[6]; ctx->h[7] += F[7]; + + W+=SHA_LBLOCK; + } + } -#ifdef OPENSSL_SMALL_FOOTPRINT +#elif defined(OPENSSL_SMALL_FOOTPRINT) -static void sha512_block (SHA512_CTX *ctx, const void *in, size_t num) +static void sha512_block_data_order (SHA512_CTX *ctx, const void *in, size_t num) { const SHA_LONG64 *W=in; SHA_LONG64 a,b,c,d,e,f,g,h,s0,s1,T1,T2; SHA_LONG64 X[16]; int i; -#ifdef GO_FOR_SSE2 - GO_FOR_SSE2(ctx,in,num); -#endif - while (num--) { a = ctx->h[0]; b = ctx->h[1]; c = ctx->h[2]; d = ctx->h[3]; @@ -380,23 +505,19 @@ static void sha512_block (SHA512_CTX *ctx, const void *in, size_t num) h = Sigma0(a) + Maj(a,b,c); \ d += T1; h += T1; } while (0) -#define ROUND_16_80(i,a,b,c,d,e,f,g,h,X) do { \ - s0 = X[(i+1)&0x0f]; s0 = sigma0(s0); \ - s1 = X[(i+14)&0x0f]; s1 = sigma1(s1); \ - T1 = X[(i)&0x0f] += s0 + s1 + X[(i+9)&0x0f]; \ - ROUND_00_15(i,a,b,c,d,e,f,g,h); } while (0) +#define ROUND_16_80(i,j,a,b,c,d,e,f,g,h,X) do { \ + s0 = X[(j+1)&0x0f]; s0 = sigma0(s0); \ + s1 = X[(j+14)&0x0f]; s1 = sigma1(s1); \ + T1 = X[(j)&0x0f] += s0 + s1 + X[(j+9)&0x0f]; \ + ROUND_00_15(i+j,a,b,c,d,e,f,g,h); } while (0) -static void sha512_block (SHA512_CTX *ctx, const void *in, size_t num) +static void sha512_block_data_order (SHA512_CTX *ctx, const void *in, size_t num) { const SHA_LONG64 *W=in; SHA_LONG64 a,b,c,d,e,f,g,h,s0,s1,T1; SHA_LONG64 X[16]; int i; -#ifdef GO_FOR_SSE2 - GO_FOR_SSE2(ctx,in,num); -#endif - while (num--) { a = ctx->h[0]; b = ctx->h[1]; c = ctx->h[2]; d = ctx->h[3]; @@ -438,16 +559,24 @@ static void sha512_block (SHA512_CTX *ctx, const void *in, size_t num) T1 = X[15] = PULL64(W[15]); ROUND_00_15(15,b,c,d,e,f,g,h,a); #endif - for (i=16;i<80;i+=8) + for (i=16;i<80;i+=16) { - ROUND_16_80(i+0,a,b,c,d,e,f,g,h,X); - ROUND_16_80(i+1,h,a,b,c,d,e,f,g,X); - ROUND_16_80(i+2,g,h,a,b,c,d,e,f,X); - ROUND_16_80(i+3,f,g,h,a,b,c,d,e,X); - ROUND_16_80(i+4,e,f,g,h,a,b,c,d,X); - ROUND_16_80(i+5,d,e,f,g,h,a,b,c,X); - ROUND_16_80(i+6,c,d,e,f,g,h,a,b,X); - ROUND_16_80(i+7,b,c,d,e,f,g,h,a,X); + ROUND_16_80(i, 0,a,b,c,d,e,f,g,h,X); + ROUND_16_80(i, 1,h,a,b,c,d,e,f,g,X); + ROUND_16_80(i, 2,g,h,a,b,c,d,e,f,X); + ROUND_16_80(i, 3,f,g,h,a,b,c,d,e,X); + ROUND_16_80(i, 4,e,f,g,h,a,b,c,d,X); + ROUND_16_80(i, 5,d,e,f,g,h,a,b,c,X); + ROUND_16_80(i, 6,c,d,e,f,g,h,a,b,X); + ROUND_16_80(i, 7,b,c,d,e,f,g,h,a,X); + ROUND_16_80(i, 8,a,b,c,d,e,f,g,h,X); + ROUND_16_80(i, 9,h,a,b,c,d,e,f,g,X); + ROUND_16_80(i,10,g,h,a,b,c,d,e,f,X); + ROUND_16_80(i,11,f,g,h,a,b,c,d,e,X); + ROUND_16_80(i,12,e,f,g,h,a,b,c,d,X); + ROUND_16_80(i,13,d,e,f,g,h,a,b,c,X); + ROUND_16_80(i,14,c,d,e,f,g,h,a,b,X); + ROUND_16_80(i,15,b,c,d,e,f,g,h,a,X); } ctx->h[0] += a; ctx->h[1] += b; ctx->h[2] += c; ctx->h[3] += d; @@ -458,3 +587,13 @@ static void sha512_block (SHA512_CTX *ctx, const void *in, size_t num) } #endif + +#endif /* SHA512_ASM */ + +#else /* !OPENSSL_NO_SHA512 */ + +#if defined(PEDANTIC) || defined(__DECC) || defined(OPENSSL_SYS_MACOSX) +static void *dummy=&dummy; +#endif + +#endif /* !OPENSSL_NO_SHA512 */