Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

use OR(<<, >>) for all rotations #5

Open
wants to merge 1 commit into
base: master
Choose a base branch
from
Open

use OR(<<, >>) for all rotations #5

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
14 changes: 4 additions & 10 deletions blake2b-common.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -27,22 +27,16 @@ static INLINE uint64_t LOADU64(void const * p) {
return v;
}

#define ROTATE16 _mm256_setr_epi8( 2, 3, 4, 5, 6, 7, 0, 1, 10, 11, 12, 13, 14, 15, 8, 9, \
2, 3, 4, 5, 6, 7, 0, 1, 10, 11, 12, 13, 14, 15, 8, 9 )

#define ROTATE24 _mm256_setr_epi8( 3, 4, 5, 6, 7, 0, 1, 2, 11, 12, 13, 14, 15, 8, 9, 10, \
3, 4, 5, 6, 7, 0, 1, 2, 11, 12, 13, 14, 15, 8, 9, 10 )

#define ADD(a, b) _mm256_add_epi64(a, b)
#define SUB(a, b) _mm256_sub_epi64(a, b)

#define XOR(a, b) _mm256_xor_si256(a, b)
#define AND(a, b) _mm256_and_si256(a, b)
#define OR(a, b) _mm256_or_si256(a, b)

#define ROT32(x) _mm256_shuffle_epi32((x), _MM_SHUFFLE(2, 3, 0, 1))
#define ROT24(x) _mm256_shuffle_epi8((x), ROTATE24)
#define ROT16(x) _mm256_shuffle_epi8((x), ROTATE16)
#define ROT63(x) _mm256_or_si256(_mm256_srli_epi64((x), 63), ADD((x), (x)))
#define ROT32(x) _mm256_or_si256(_mm256_srli_epi64((x), 32), _mm256_slli_epi64((x), 64 -32))
#define ROT24(x) _mm256_or_si256(_mm256_srli_epi64((x), 24), _mm256_slli_epi64((x), 64 -24))
#define ROT16(x) _mm256_or_si256(_mm256_srli_epi64((x), 16), _mm256_slli_epi64((x), 64 -16))
#define ROT63(x) _mm256_or_si256(_mm256_srli_epi64((x), 63), _mm256_slli_epi64((x), 64 -63))

#endif
27 changes: 8 additions & 19 deletions blake2s-common.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -24,45 +24,34 @@ static INLINE uint32_t LOADU32(void const * p) {
#define TOF(reg) _mm_castsi128_ps((reg))
#define TOI(reg) _mm_castps_si128((reg))

#define ROTATE8128 _mm_set_epi8(12, 15, 14, 13, 8, 11, 10, 9, 4, 7, 6, 5, 0, 3, 2, 1)

#define ROTATE16128 _mm_set_epi8(13, 12, 15, 14, 9, 8, 11, 10, 5, 4, 7, 6, 1, 0, 3, 2)


#define ADD128(a, b) _mm_add_epi32(a, b)
#define SUB128(a, b) _mm_sub_epi32(a, b)

#define XOR128(a, b) _mm_xor_si128(a, b)
#define AND128(a, b) _mm_and_si128(a, b)
#define OR128(a, b) _mm_or_si128(a, b)

#define ROT16128(x) _mm_shuffle_epi8((x), ROTATE16128)
#define ROT12128(x) OR128(_mm_srli_epi32((x), 12), _mm_slli_epi32((x), 20))
#define ROT8128(x) _mm_shuffle_epi8((x), ROTATE8128)
#define ROT7128(x) OR128(_mm_srli_epi32((x), 7), _mm_slli_epi32((x), 25))
#define ROT16128(x) OR128(_mm_srli_epi32((x), 16), _mm_slli_epi32((x), 32 - 16))
#define ROT12128(x) OR128(_mm_srli_epi32((x), 12), _mm_slli_epi32((x), 32 - 12))
#define ROT8128(x) OR128(_mm_srli_epi32((x), 8), _mm_slli_epi32((x), 32 - 8))
#define ROT7128(x) OR128(_mm_srli_epi32((x), 7), _mm_slli_epi32((x), 32 - 7))

#define LOAD(p) _mm256_load_si256( (__m256i *)(p) )
#define STORE(p,r) _mm256_store_si256((__m256i *)(p), r)

#define LOADU(p) _mm256_loadu_si256( (__m256i *)(p) )
#define STOREU(p,r) _mm256_storeu_si256((__m256i *)(p), r)

#define ROTATE8 _mm256_set_epi8(12, 15, 14, 13, 8, 11, 10, 9, 4, 7, 6, 5, 0, 3, 2, 1, \
12, 15, 14, 13, 8, 11, 10, 9, 4, 7, 6, 5, 0, 3, 2, 1)

#define ROTATE16 _mm256_set_epi8(13, 12, 15, 14, 9, 8, 11, 10, 5, 4, 7, 6, 1, 0, 3, 2, \
13, 12, 15, 14, 9, 8, 11, 10, 5, 4, 7, 6, 1, 0, 3, 2)

#define ADD(a, b) _mm256_add_epi32(a, b)
#define SUB(a, b) _mm256_sub_epi32(a, b)

#define XOR(a, b) _mm256_xor_si256(a, b)
#define AND(a, b) _mm256_and_si256(a, b)
#define OR(a, b) _mm256_or_si256(a, b)

#define ROT16(x) _mm256_shuffle_epi8((x), ROTATE16)
#define ROT12(x) _mm256_or_si256(_mm256_srli_epi32((x), 12), _mm256_slli_epi32((x), 20))
#define ROT8(x) _mm256_shuffle_epi8((x), ROTATE8)
#define ROT7(x) _mm256_or_si256(_mm256_srli_epi32((x), 7), _mm256_slli_epi32((x), 25))
#define ROT16(x) _mm256_or_si256(_mm256_srli_epi32((x), 16), _mm256_slli_epi32((x), 32 - 16))
#define ROT12(x) _mm256_or_si256(_mm256_srli_epi32((x), 12), _mm256_slli_epi32((x), 32 - 12))
#define ROT8(x) _mm256_or_si256(_mm256_srli_epi32((x), 8), _mm256_slli_epi32((x), 32 - 8))
#define ROT7(x) _mm256_or_si256(_mm256_srli_epi32((x), 7), _mm256_slli_epi32((x), 32 - 7))

#endif