Skip to content

Commit

Permalink
v3.8.1
Browse files Browse the repository at this point in the history
  • Loading branch information
@JayDDee
JayDDee committed Feb 7, 2018
1 parent 54b8fd7 commit a28daca
Show file tree
Hide file tree
Showing 83 changed files with 5,119 additions and 3,890 deletions.
12 changes: 9 additions & 3 deletions Makefile.am
View file
Original file line number Diff line number Diff line change
Expand Up @@ -100,7 +100,8 @@ cpuminer_SOURCES = \
algo/lbry.c \
algo/luffa/sph_luffa.c \
algo/luffa/luffa.c \
algo/luffa/sse2/luffa_for_sse2.c \
algo/luffa/luffa_for_sse2.c \
algo/luffa/luffa-hash-2way.c \
algo/lyra2/lyra2.c \
algo/lyra2/sponge.c \
algo/lyra2/lyra2rev2-gate.c \
Expand All @@ -127,7 +128,11 @@ cpuminer_SOURCES = \
algo/quark/anime-gate.c \
algo/quark/anime.c \
algo/quark/anime-4way.c \
algo/qubit/qubit-gate.c \
algo/qubit/qubit.c \
algo/qubit/qubit-2way.c \
algo/qubit/deep-gate.c \
algo/qubit/deep-2way.c \
algo/qubit/deep.c \
algo/ripemd/sph_ripemd.c \
algo/scrypt.c \
Expand All @@ -143,8 +148,9 @@ cpuminer_SOURCES = \
algo/shavite/sph-shavite-aesni.c \
algo/shavite/shavite.c \
algo/simd/sph_simd.c \
algo/simd/sse2/nist.c \
algo/simd/sse2/vector.c \
algo/simd/nist.c \
algo/simd/vector.c \
algo/simd/simd-hash-2way.c \
algo/skein/sph_skein.c \
algo/skein/skein-hash-4way.c \
algo/skein/skein.c \
Expand Down
11 changes: 8 additions & 3 deletions README.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -16,6 +16,7 @@ See file RELEASE_NOTES for change log and compile instructions.
Supported Algorithms
--------------------

anime Animecoin
argon2
axiom Shabal-256 MemoHash
bastion
Expand Down Expand Up @@ -78,6 +79,7 @@ Supported Algorithms
x13sm3 hsr (Hshare)
x14 X14
x15 X15
x16r Ravencoin
x17
xevan Bitsend
yescrypt Globalboost-Y (BSTY)
Expand Down Expand Up @@ -136,10 +138,13 @@ output from the miner showing the startup and any errors.
Donations
---------

I do not do this for money but I have a donation address if users
are so inclined.
cpuminer-opt has no fees of any kind but donations are accepted.

bitcoin:12tdvfF7KmAsihBXQXynT6E6th2c2pByTT?label=donations
BTC: 12tdvfF7KmAsihBXQXynT6E6th2c2pByTT
ETH: 0x72122edabcae9d3f57eab0729305a425f6fef6d0
LTC: LdUwoHJnux9r9EKqFWNvAi45kQompHk6e8
BCH: 1QKYkB6atn4P7RFozyziAXLEnurwnUM1cQ
BTG: GVUyECtRHeC5D58z9F3nGGfVQndwnsPnHQ

Happy mining!

9 changes: 9 additions & 0 deletions README.txt
View file
Original file line number Diff line number Diff line change
Expand Up @@ -25,3 +25,12 @@ cpuminer-aes-avx.exe "-march=corei7-avx" Sandybridge, Ivybridge
cpuminer-avx2.exe "-march=core-avx2" Haswell...
cpuminer-avx2-sha.exe "-march=core-avx2 -msha" Ryzen

If you like this software feel free to donate:

BTC: 12tdvfF7KmAsihBXQXynT6E6th2c2pByTT
ETH: 0x72122edabcae9d3f57eab0729305a425f6fef6d0
LTC: LdUwoHJnux9r9EKqFWNvAi45kQompHk6e8
BCH: 1QKYkB6atn4P7RFozyziAXLEnurwnUM1cQ
BTG: GVUyECtRHeC5D58z9F3nGGfVQndwnsPnHQ


16 changes: 13 additions & 3 deletions RELEASE_NOTES
View file
Original file line number Diff line number Diff line change
Expand Up @@ -98,8 +98,8 @@ Start mining.

Windows

The following in how the Windows binary releases are built. It's old and
not very good but it works, for me anyway.
Precompiled Windows binaries are built on a Linux host using Mingw
with a more recent compiler than the following Windows hosted procedure.

Building on Windows prerequisites:

Expand Down Expand Up @@ -131,7 +131,7 @@ or similar Windows program.
In msys shell cd to miner directory.
cd /c/path/to/cpuminer-opt

Run winbuild.sh to build on Windows or execute the following commands.
Run build.sh to build on Windows or execute the following commands.

./autogen.sh
CFLAGS="-O3 -march=native -Wall" CXXFLAGS="$CFLAGS -std=gnu++11 -fpermissive" ./configure --with-curl
Expand Down Expand Up @@ -159,6 +159,16 @@ Support for even older x86_64 without AES_NI or SSE2 is not availble.
Change Log
----------

v3.8.1

Fixes x16r on CPUs with only SSE2.
More Optimizations for X algos, qubit & deep.
Corrected algo optimizations for scrypt and yescrypt, no new optimizations.

v3.8.0.1

Fixed x16r AVX2 low hash rate.

v3.8.0

4way no longer a seperate feature, included in AVX2.
Expand Down
190 changes: 95 additions & 95 deletions algo/blake/blake-hash-4way.c
View file
Original file line number Diff line number Diff line change
Expand Up @@ -553,22 +553,22 @@ do { \
, _mm_set_epi32( CS6, CS6, CS6, CS6 ) ); \
VF = _mm_xor_si128( _mm_set_epi32( T1, T1, T1, T1 ), \
_mm_set_epi32( CS7, CS7, CS7, CS7 ) ); \
M[0x0] = mm_byteswap_32( *(buf + 0) ); \
M[0x1] = mm_byteswap_32( *(buf + 1) ); \
M[0x2] = mm_byteswap_32( *(buf + 2) ); \
M[0x3] = mm_byteswap_32( *(buf + 3) ); \
M[0x4] = mm_byteswap_32( *(buf + 4) ); \
M[0x5] = mm_byteswap_32( *(buf + 5) ); \
M[0x6] = mm_byteswap_32( *(buf + 6) ); \
M[0x7] = mm_byteswap_32( *(buf + 7) ); \
M[0x8] = mm_byteswap_32( *(buf + 8) ); \
M[0x9] = mm_byteswap_32( *(buf + 9) ); \
M[0xA] = mm_byteswap_32( *(buf + 10) ); \
M[0xB] = mm_byteswap_32( *(buf + 11) ); \
M[0xC] = mm_byteswap_32( *(buf + 12) ); \
M[0xD] = mm_byteswap_32( *(buf + 13) ); \
M[0xE] = mm_byteswap_32( *(buf + 14) ); \
M[0xF] = mm_byteswap_32( *(buf + 15) ); \
M[0x0] = mm_bswap_32( *(buf + 0) ); \
M[0x1] = mm_bswap_32( *(buf + 1) ); \
M[0x2] = mm_bswap_32( *(buf + 2) ); \
M[0x3] = mm_bswap_32( *(buf + 3) ); \
M[0x4] = mm_bswap_32( *(buf + 4) ); \
M[0x5] = mm_bswap_32( *(buf + 5) ); \
M[0x6] = mm_bswap_32( *(buf + 6) ); \
M[0x7] = mm_bswap_32( *(buf + 7) ); \
M[0x8] = mm_bswap_32( *(buf + 8) ); \
M[0x9] = mm_bswap_32( *(buf + 9) ); \
M[0xA] = mm_bswap_32( *(buf + 10) ); \
M[0xB] = mm_bswap_32( *(buf + 11) ); \
M[0xC] = mm_bswap_32( *(buf + 12) ); \
M[0xD] = mm_bswap_32( *(buf + 13) ); \
M[0xE] = mm_bswap_32( *(buf + 14) ); \
M[0xF] = mm_bswap_32( *(buf + 15) ); \
for (r = 0; r < rounds; r ++) \
ROUND_S_4WAY(r); \
H0 = _mm_xor_si128( _mm_xor_si128( \
Expand Down Expand Up @@ -615,22 +615,22 @@ do { \
VD = _mm_xor_si128( _mm_set1_epi32( T0 ), _mm_set1_epi32( CS5 ) ); \
VE = _mm_xor_si128( _mm_set1_epi32( T1 ), _mm_set1_epi32( CS6 ) ); \
VF = _mm_xor_si128( _mm_set1_epi32( T1 ), _mm_set1_epi32( CS7 ) ); \
M0 = mm_byteswap_32( * buf ); \
M1 = mm_byteswap_32( *(buf+1) ); \
M2 = mm_byteswap_32( *(buf+2) ); \
M3 = mm_byteswap_32( *(buf+3) ); \
M4 = mm_byteswap_32( *(buf+4) ); \
M5 = mm_byteswap_32( *(buf+5) ); \
M6 = mm_byteswap_32( *(buf+6) ); \
M7 = mm_byteswap_32( *(buf+7) ); \
M8 = mm_byteswap_32( *(buf+8) ); \
M9 = mm_byteswap_32( *(buf+9) ); \
MA = mm_byteswap_32( *(buf+10) ); \
MB = mm_byteswap_32( *(buf+11) ); \
MC = mm_byteswap_32( *(buf+12) ); \
MD = mm_byteswap_32( *(buf+13) ); \
ME = mm_byteswap_32( *(buf+14) ); \
MF = mm_byteswap_32( *(buf+15) ); \
M0 = mm_bswap_32( * buf ); \
M1 = mm_bswap_32( *(buf+1) ); \
M2 = mm_bswap_32( *(buf+2) ); \
M3 = mm_bswap_32( *(buf+3) ); \
M4 = mm_bswap_32( *(buf+4) ); \
M5 = mm_bswap_32( *(buf+5) ); \
M6 = mm_bswap_32( *(buf+6) ); \
M7 = mm_bswap_32( *(buf+7) ); \
M8 = mm_bswap_32( *(buf+8) ); \
M9 = mm_bswap_32( *(buf+9) ); \
MA = mm_bswap_32( *(buf+10) ); \
MB = mm_bswap_32( *(buf+11) ); \
MC = mm_bswap_32( *(buf+12) ); \
MD = mm_bswap_32( *(buf+13) ); \
ME = mm_bswap_32( *(buf+14) ); \
MF = mm_bswap_32( *(buf+15) ); \
ROUND_S_4WAY(0); \
ROUND_S_4WAY(1); \
ROUND_S_4WAY(2); \
Expand Down Expand Up @@ -727,22 +727,22 @@ do { \
VD = _mm256_xor_si256( _mm256_set1_epi32( T0 ), _mm256_set1_epi32( CS5 ) ); \
VE = _mm256_xor_si256( _mm256_set1_epi32( T1 ), _mm256_set1_epi32( CS6 ) ); \
VF = _mm256_xor_si256( _mm256_set1_epi32( T1 ), _mm256_set1_epi32( CS7 ) ); \
M0 = mm256_byteswap_32( * buf ); \
M1 = mm256_byteswap_32( *(buf+1) ); \
M2 = mm256_byteswap_32( *(buf+2) ); \
M3 = mm256_byteswap_32( *(buf+3) ); \
M4 = mm256_byteswap_32( *(buf+4) ); \
M5 = mm256_byteswap_32( *(buf+5) ); \
M6 = mm256_byteswap_32( *(buf+6) ); \
M7 = mm256_byteswap_32( *(buf+7) ); \
M8 = mm256_byteswap_32( *(buf+8) ); \
M9 = mm256_byteswap_32( *(buf+9) ); \
MA = mm256_byteswap_32( *(buf+10) ); \
MB = mm256_byteswap_32( *(buf+11) ); \
MC = mm256_byteswap_32( *(buf+12) ); \
MD = mm256_byteswap_32( *(buf+13) ); \
ME = mm256_byteswap_32( *(buf+14) ); \
MF = mm256_byteswap_32( *(buf+15) ); \
M0 = mm256_bswap_32( * buf ); \
M1 = mm256_bswap_32( *(buf+1) ); \
M2 = mm256_bswap_32( *(buf+2) ); \
M3 = mm256_bswap_32( *(buf+3) ); \
M4 = mm256_bswap_32( *(buf+4) ); \
M5 = mm256_bswap_32( *(buf+5) ); \
M6 = mm256_bswap_32( *(buf+6) ); \
M7 = mm256_bswap_32( *(buf+7) ); \
M8 = mm256_bswap_32( *(buf+8) ); \
M9 = mm256_bswap_32( *(buf+9) ); \
MA = mm256_bswap_32( *(buf+10) ); \
MB = mm256_bswap_32( *(buf+11) ); \
MC = mm256_bswap_32( *(buf+12) ); \
MD = mm256_bswap_32( *(buf+13) ); \
ME = mm256_bswap_32( *(buf+14) ); \
MF = mm256_bswap_32( *(buf+15) ); \
ROUND_S_8WAY(0); \
ROUND_S_8WAY(1); \
ROUND_S_8WAY(2); \
Expand Down Expand Up @@ -848,22 +848,22 @@ do { \
_mm256_set_epi64x( CB6, CB6, CB6, CB6 ) ); \
VF = _mm256_xor_si256( _mm256_set_epi64x( T1, T1, T1, T1 ), \
_mm256_set_epi64x( CB7, CB7, CB7, CB7 ) ); \
M[0x0] = mm256_byteswap_64( *(buf+0) ); \
M[0x1] = mm256_byteswap_64( *(buf+1) ); \
M[0x2] = mm256_byteswap_64( *(buf+2) ); \
M[0x3] = mm256_byteswap_64( *(buf+3) ); \
M[0x4] = mm256_byteswap_64( *(buf+4) ); \
M[0x5] = mm256_byteswap_64( *(buf+5) ); \
M[0x6] = mm256_byteswap_64( *(buf+6) ); \
M[0x7] = mm256_byteswap_64( *(buf+7) ); \
M[0x8] = mm256_byteswap_64( *(buf+8) ); \
M[0x9] = mm256_byteswap_64( *(buf+9) ); \
M[0xA] = mm256_byteswap_64( *(buf+10) ); \
M[0xB] = mm256_byteswap_64( *(buf+11) ); \
M[0xC] = mm256_byteswap_64( *(buf+12) ); \
M[0xD] = mm256_byteswap_64( *(buf+13) ); \
M[0xE] = mm256_byteswap_64( *(buf+14) ); \
M[0xF] = mm256_byteswap_64( *(buf+15) ); \
M[0x0] = mm256_bswap_64( *(buf+0) ); \
M[0x1] = mm256_bswap_64( *(buf+1) ); \
M[0x2] = mm256_bswap_64( *(buf+2) ); \
M[0x3] = mm256_bswap_64( *(buf+3) ); \
M[0x4] = mm256_bswap_64( *(buf+4) ); \
M[0x5] = mm256_bswap_64( *(buf+5) ); \
M[0x6] = mm256_bswap_64( *(buf+6) ); \
M[0x7] = mm256_bswap_64( *(buf+7) ); \
M[0x8] = mm256_bswap_64( *(buf+8) ); \
M[0x9] = mm256_bswap_64( *(buf+9) ); \
M[0xA] = mm256_bswap_64( *(buf+10) ); \
M[0xB] = mm256_bswap_64( *(buf+11) ); \
M[0xC] = mm256_bswap_64( *(buf+12) ); \
M[0xD] = mm256_bswap_64( *(buf+13) ); \
M[0xE] = mm256_bswap_64( *(buf+14) ); \
M[0xF] = mm256_bswap_64( *(buf+15) ); \
for (r = 0; r < 16; r ++) \
ROUND_B_4WAY(r); \
H0 = _mm256_xor_si256( _mm256_xor_si256( \
Expand Down Expand Up @@ -913,22 +913,22 @@ do { \
_mm256_set_epi64x( CB6, CB6, CB6, CB6 ) ); \
VF = _mm256_xor_si256( _mm256_set_epi64x( T1, T1, T1, T1 ), \
_mm256_set_epi64x( CB7, CB7, CB7, CB7 ) ); \
M0 = mm256_byteswap_64( *(buf + 0) ); \
M1 = mm256_byteswap_64( *(buf + 1) ); \
M2 = mm256_byteswap_64( *(buf + 2) ); \
M3 = mm256_byteswap_64( *(buf + 3) ); \
M4 = mm256_byteswap_64( *(buf + 4) ); \
M5 = mm256_byteswap_64( *(buf + 5) ); \
M6 = mm256_byteswap_64( *(buf + 6) ); \
M7 = mm256_byteswap_64( *(buf + 7) ); \
M8 = mm256_byteswap_64( *(buf + 8) ); \
M9 = mm256_byteswap_64( *(buf + 9) ); \
MA = mm256_byteswap_64( *(buf + 10) ); \
MB = mm256_byteswap_64( *(buf + 11) ); \
MC = mm256_byteswap_64( *(buf + 12) ); \
MD = mm256_byteswap_64( *(buf + 13) ); \
ME = mm256_byteswap_64( *(buf + 14) ); \
MF = mm256_byteswap_64( *(buf + 15) ); \
M0 = mm256_bswap_64( *(buf + 0) ); \
M1 = mm256_bswap_64( *(buf + 1) ); \
M2 = mm256_bswap_64( *(buf + 2) ); \
M3 = mm256_bswap_64( *(buf + 3) ); \
M4 = mm256_bswap_64( *(buf + 4) ); \
M5 = mm256_bswap_64( *(buf + 5) ); \
M6 = mm256_bswap_64( *(buf + 6) ); \
M7 = mm256_bswap_64( *(buf + 7) ); \
M8 = mm256_bswap_64( *(buf + 8) ); \
M9 = mm256_bswap_64( *(buf + 9) ); \
MA = mm256_bswap_64( *(buf + 10) ); \
MB = mm256_bswap_64( *(buf + 11) ); \
MC = mm256_bswap_64( *(buf + 12) ); \
MD = mm256_bswap_64( *(buf + 13) ); \
ME = mm256_bswap_64( *(buf + 14) ); \
MF = mm256_bswap_64( *(buf + 15) ); \
ROUND_B_4WAY(0); \
ROUND_B_4WAY(1); \
ROUND_B_4WAY(2); \
Expand Down Expand Up @@ -1064,8 +1064,8 @@ blake32_4way_close( blake_4way_small_context *sc, unsigned ub, unsigned n,
if (out_size_w32 == 8)
u.buf[52>>2] = _mm_or_si128( u.buf[52>>2],
_mm_set1_epi32( 0x01000000UL ) );
*(u.buf+(56>>2)) = mm_byteswap_32( _mm_set1_epi32( th ) );
*(u.buf+(60>>2)) = mm_byteswap_32( _mm_set1_epi32( tl ) );
*(u.buf+(56>>2)) = mm_bswap_32( _mm_set1_epi32( th ) );
*(u.buf+(60>>2)) = mm_bswap_32( _mm_set1_epi32( tl ) );
blake32_4way( sc, u.buf + (ptr>>2), 64 - ptr );
}
else
Expand All @@ -1077,13 +1077,13 @@ blake32_4way_close( blake_4way_small_context *sc, unsigned ub, unsigned n,
memset_zero_128( u.buf, 56>>2 );
if (out_size_w32 == 8)
u.buf[52>>2] = _mm_set1_epi32( 0x01000000UL );
*(u.buf+(56>>2)) = mm_byteswap_32( _mm_set1_epi32( th ) );
*(u.buf+(60>>2)) = mm_byteswap_32( _mm_set1_epi32( tl ) );
*(u.buf+(56>>2)) = mm_bswap_32( _mm_set1_epi32( th ) );
*(u.buf+(60>>2)) = mm_bswap_32( _mm_set1_epi32( tl ) );
blake32_4way( sc, u.buf, 64 );
}
out = (__m128i*)dst;
for ( k = 0; k < out_size_w32; k++ )
out[k] = mm_byteswap_32( sc->H[k] );
out[k] = mm_bswap_32( sc->H[k] );
}

#if defined (__AVX2__)
Expand Down Expand Up @@ -1187,8 +1187,8 @@ blake32_8way_close( blake_8way_small_context *sc, unsigned ub, unsigned n,
if (out_size_w32 == 8)
u.buf[52>>2] = _mm256_or_si256( u.buf[52>>2],
_mm256_set1_epi32( 0x01000000UL ) );
*(u.buf+(56>>2)) = mm256_byteswap_32( _mm256_set1_epi32( th ) );
*(u.buf+(60>>2)) = mm256_byteswap_32( _mm256_set1_epi32( tl ) );
*(u.buf+(56>>2)) = mm256_bswap_32( _mm256_set1_epi32( th ) );
*(u.buf+(60>>2)) = mm256_bswap_32( _mm256_set1_epi32( tl ) );
blake32_8way( sc, u.buf + (ptr>>2), 64 - ptr );
}
else
Expand All @@ -1200,13 +1200,13 @@ blake32_8way_close( blake_8way_small_context *sc, unsigned ub, unsigned n,
memset_zero_256( u.buf, 56>>2 );
if (out_size_w32 == 8)
u.buf[52>>2] = _mm256_set1_epi32( 0x01000000UL );
*(u.buf+(56>>2)) = mm256_byteswap_32( _mm256_set1_epi32( th ) );
*(u.buf+(60>>2)) = mm256_byteswap_32( _mm256_set1_epi32( tl ) );
*(u.buf+(56>>2)) = mm256_bswap_32( _mm256_set1_epi32( th ) );
*(u.buf+(60>>2)) = mm256_bswap_32( _mm256_set1_epi32( tl ) );
blake32_8way( sc, u.buf, 64 );
}
out = (__m256i*)dst;
for ( k = 0; k < out_size_w32; k++ )
out[k] = mm256_byteswap_32( sc->H[k] );
out[k] = mm256_bswap_32( sc->H[k] );
}

// Blake-512 4 way
Expand Down Expand Up @@ -1311,9 +1311,9 @@ blake64_4way_close( blake_4way_big_context *sc,
if ( out_size_w64 == 8 )
u.buf[(104>>3)] = _mm256_or_si256( u.buf[(104>>3)],
_mm256_set1_epi64x( 0x0100000000000000ULL ) );
*(u.buf+(112>>3)) = mm256_byteswap_64(
*(u.buf+(112>>3)) = mm256_bswap_64(
_mm256_set_epi64x( th, th, th, th ) );
*(u.buf+(120>>3)) = mm256_byteswap_64(
*(u.buf+(120>>3)) = mm256_bswap_64(
_mm256_set_epi64x( tl, tl, tl, tl ) );

blake64_4way( sc, u.buf + (ptr>>3), 128 - ptr );
Expand All @@ -1328,16 +1328,16 @@ blake64_4way_close( blake_4way_big_context *sc,
memset_zero_256( u.buf, 112>>3 );
if ( out_size_w64 == 8 )
u.buf[104>>3] = _mm256_set1_epi64x( 0x0100000000000000ULL );
*(u.buf+(112>>3)) = mm256_byteswap_64(
*(u.buf+(112>>3)) = mm256_bswap_64(
_mm256_set_epi64x( th, th, th, th ) );
*(u.buf+(120>>3)) = mm256_byteswap_64(
*(u.buf+(120>>3)) = mm256_bswap_64(
_mm256_set_epi64x( tl, tl, tl, tl ) );

blake64_4way( sc, u.buf, 128 );
}
out = (__m256i*)dst;
for ( k = 0; k < out_size_w64; k++ )
out[k] = mm256_byteswap_64( sc->H[k] );
out[k] = mm256_bswap_64( sc->H[k] );
}

#endif
Expand Down
Loading

0 comments on commit a28daca

Please sign in to comment.