void gcm_gmult_neon(u64 Xi[2],const u128 Htable[16]);
void gcm_ghash_neon(u64 Xi[2],const u128 Htable[16],const u8 *inp,size_t len);
# endif
+# elif defined(__sparc__) || defined(__sparc)
+# include "sparc_arch.h"
+# define GHASH_ASM_SPARC
+# define GCM_FUNCREF_4BIT
+extern unsigned int OPENSSL_sparcv9cap_P[];
+void gcm_init_vis3(u128 Htable[16],const u64 Xi[2]);
+void gcm_gmult_vis3(u64 Xi[2],const u128 Htable[16]);
+void gcm_ghash_vis3(u64 Xi[2],const u128 Htable[16],const u8 *inp,size_t len);
# endif
#endif
ctx->gmult = gcm_gmult_4bit;
ctx->ghash = gcm_ghash_4bit;
}
+# elif defined(GHASH_ASM_SPARC)
+ if (OPENSSL_sparcv9cap_P[0] & SPARCV9_VIS3) {
+ gcm_init_vis3(ctx->Htable,ctx->H.u);
+ ctx->gmult = gcm_gmult_vis3;
+ ctx->ghash = gcm_ghash_vis3;
+ } else {
+ gcm_init_4bit(ctx->Htable,ctx->H.u);
+ ctx->gmult = gcm_gmult_4bit;
+ ctx->ghash = gcm_ghash_4bit;
+ }
# else
gcm_init_4bit(ctx->Htable,ctx->H.u);
# endif
size_t j=GHASH_CHUNK;
while (j) {
+ size_t *out_t=(size_t *)out;
+ const size_t *in_t=(const size_t *)in;
+
(*block)(ctx->Yi.c,ctx->EKi.c,key);
++ctr;
if (is_endian.little)
PUTU32(ctx->Yi.c+12,ctr);
else
ctx->Yi.d[3] = ctr;
- for (i=0; i<16; i+=sizeof(size_t))
- *(size_t *)(out+i) =
- *(size_t *)(in+i)^*(size_t *)(ctx->EKi.c+i);
+ for (i=0; i<16/sizeof(size_t); ++i)
+ out_t[i] = in_t[i] ^ ctx->EKi.t[i];
out += 16;
in += 16;
j -= 16;
size_t j=i;
while (len>=16) {
+ size_t *out_t=(size_t *)out;
+ const size_t *in_t=(const size_t *)in;
+
(*block)(ctx->Yi.c,ctx->EKi.c,key);
++ctr;
if (is_endian.little)
PUTU32(ctx->Yi.c+12,ctr);
else
ctx->Yi.d[3] = ctr;
- for (i=0; i<16; i+=sizeof(size_t))
- *(size_t *)(out+i) =
- *(size_t *)(in+i)^*(size_t *)(ctx->EKi.c+i);
+ for (i=0; i<16/sizeof(size_t); ++i)
+ out_t[i] = in_t[i] ^ ctx->EKi.t[i];
out += 16;
in += 16;
len -= 16;
}
#else
while (len>=16) {
+ size_t *out_t=(size_t *)out;
+ const size_t *in_t=(const size_t *)in;
+
(*block)(ctx->Yi.c,ctx->EKi.c,key);
++ctr;
if (is_endian.little)
PUTU32(ctx->Yi.c+12,ctr);
else
ctx->Yi.d[3] = ctr;
- for (i=0; i<16; i+=sizeof(size_t))
- *(size_t *)(ctx->Xi.c+i) ^=
- *(size_t *)(out+i) =
- *(size_t *)(in+i)^*(size_t *)(ctx->EKi.c+i);
+ for (i=0; i<16/sizeof(size_t); ++i)
+ ctx->Xi.t[i] ^=
+ out_t[i] = in_t[i]^ctx->EKi.t[i];
GCM_MUL(ctx,Xi);
out += 16;
in += 16;
GHASH(ctx,in,GHASH_CHUNK);
while (j) {
+ size_t *out_t=(size_t *)out;
+ const size_t *in_t=(const size_t *)in;
+
(*block)(ctx->Yi.c,ctx->EKi.c,key);
++ctr;
if (is_endian.little)
PUTU32(ctx->Yi.c+12,ctr);
else
ctx->Yi.d[3] = ctr;
- for (i=0; i<16; i+=sizeof(size_t))
- *(size_t *)(out+i) =
- *(size_t *)(in+i)^*(size_t *)(ctx->EKi.c+i);
+ for (i=0; i<16/sizeof(size_t); ++i)
+ out_t[i] = in_t[i]^ctx->EKi.t[i];
out += 16;
in += 16;
j -= 16;
if ((i = (len&(size_t)-16))) {
GHASH(ctx,in,i);
while (len>=16) {
+ size_t *out_t=(size_t *)out;
+ const size_t *in_t=(const size_t *)in;
+
(*block)(ctx->Yi.c,ctx->EKi.c,key);
++ctr;
if (is_endian.little)
PUTU32(ctx->Yi.c+12,ctr);
else
ctx->Yi.d[3] = ctr;
- for (i=0; i<16; i+=sizeof(size_t))
- *(size_t *)(out+i) =
- *(size_t *)(in+i)^*(size_t *)(ctx->EKi.c+i);
+ for (i=0; i<16/sizeof(size_t); ++i)
+ out_t[i] = in_t[i]^ctx->EKi.t[i];
out += 16;
in += 16;
len -= 16;
}
#else
while (len>=16) {
+ size_t *out_t=(size_t *)out;
+ const size_t *in_t=(const size_t *)in;
+
(*block)(ctx->Yi.c,ctx->EKi.c,key);
++ctr;
if (is_endian.little)
PUTU32(ctx->Yi.c+12,ctr);
else
ctx->Yi.d[3] = ctr;
- for (i=0; i<16; i+=sizeof(size_t)) {
- size_t c = *(size_t *)(in+i);
- *(size_t *)(out+i) = c^*(size_t *)(ctx->EKi.c+i);
- *(size_t *)(ctx->Xi.c+i) ^= c;
+ for (i=0; i<16/sizeof(size_t); ++i) {
+ size_t c = in[i];
+ out[i] = c^ctx->EKi.t[i];
+ ctx->Xi.t[i] ^= c;
}
GCM_MUL(ctx,Xi);
out += 16;
void (*gcm_gmult_p)(u64 Xi[2],const u128 Htable[16]) = ctx->gmult;
#endif
- if (ctx->mres)
+ if (ctx->mres || ctx->ares)
GCM_MUL(ctx,Xi);
if (is_endian.little) {