mirror of
https://github.com/JayDDee/cpuminer-opt.git
synced 2025-09-17 23:44:27 +00:00
Compare commits
16 Commits
| Author | SHA1 | Date | |
|---|---|---|---|
|
5e088d00d0 | ||
|
|
972d4d70db | ||
|
|
e96a6bd699 | ||
|
|
fb9163185a | ||
|
|
6e8b8ed34f | ||
|
|
c0aadbcc99 | ||
|
|
3da149418a | ||
|
|
720610cce5 | ||
|
|
cedcf4d070 | ||
|
|
81b50c3c71 | ||
|
|
0e1e88f53e | ||
|
|
45c77a5c81 | ||
|
|
dbce7e0721 | ||
|
|
6d66051de6 | ||
|
|
b93be8816a | ||
|
|
19b0ac6d5c |
@@ -163,6 +163,7 @@ cpuminer_SOURCES = \
|
||||
algo/sha/sha256-hash-4way.c \
|
||||
algo/sha/sha512-hash-4way.c \
|
||||
algo/sha/hmac-sha256-hash.c \
|
||||
algo/sha/hmac-sha256-hash-4way.c \
|
||||
algo/sha/sha2.c \
|
||||
algo/sha/sha256t-gate.c \
|
||||
algo/sha/sha256t-4way.c \
|
||||
|
||||
55
README.md
55
README.md
@@ -12,10 +12,24 @@ a false positive, they are flagged simply because they are cryptocurrency
|
||||
miners. The source code is open for anyone to inspect. If you don't trust
|
||||
the software, don't use it.
|
||||
|
||||
|
||||
New thread:
|
||||
|
||||
https://bitcointalk.org/index.php?topic=5226770.msg53865575#msg53865575
|
||||
|
||||
Old thread:
|
||||
|
||||
https://bitcointalk.org/index.php?topic=1326803.0
|
||||
|
||||
mailto://jayddee246@gmail.com
|
||||
|
||||
This note is to confirm that bitcointalk users JayDDee and joblo are the
|
||||
same person.
|
||||
|
||||
I created a new BCT user JayDDee to match my github user id.
|
||||
The old thread has been locked but still contains useful information for
|
||||
reading.
|
||||
|
||||
See file RELEASE_NOTES for change log and INSTALL_LINUX or INSTALL_WINDOWS
|
||||
for compile instructions.
|
||||
|
||||
@@ -23,25 +37,25 @@ Requirements
|
||||
------------
|
||||
|
||||
1. A x86_64 architecture CPU with a minimum of SSE2 support. This includes
|
||||
Intel Core2 and newer and AMD equivalents. In order to take advantage of AES_NI
|
||||
optimizations a CPU with AES_NI is required. This includes Intel Westmere
|
||||
and newer and AMD equivalents. Further optimizations are available on some
|
||||
algoritms for CPUs with AVX and AVX2, Sandybridge and Haswell respectively.
|
||||
Intel Core2 and newer and AMD equivalents. Further optimizations are available
|
||||
on some algoritms for CPUs with AES, AVX, AVX2, SHA, AVX512 and VAES.
|
||||
|
||||
Older CPUs are supported by cpuminer-multi by TPruvot but at reduced
|
||||
performance.
|
||||
|
||||
ARM CPUs are not supported.
|
||||
ARM and Aarch64 CPUs are not supported.
|
||||
|
||||
2. 64 bit Linux OS. Ubuntu and Fedora based distributions, including Mint and
|
||||
Centos, are known to work and have all dependencies in their repositories.
|
||||
Others may work but may require more effort. Older versions such as Centos 6
|
||||
don't work due to missing features.
|
||||
2. 64 bit Linux or Windows OS. Ubuntu and Fedora based distributions,
|
||||
including Mint and Centos, are known to work and have all dependencies
|
||||
in their repositories. Others may work but may require more effort. Older
|
||||
versions such as Centos 6 don't work due to missing features.
|
||||
64 bit Windows OS is supported with mingw_w64 and msys or pre-built binaries.
|
||||
|
||||
MacOS, OSx and Android are not supported.
|
||||
|
||||
3. Stratum pool. Some algos may work wallet mining using getwork or GBT. YMMV.
|
||||
3. Stratum pool supporting stratum+tcp:// or stratum+ssl:// protocols or
|
||||
RPC getwork using http:// or https://.
|
||||
GBT is YMMV.
|
||||
|
||||
Supported Algorithms
|
||||
--------------------
|
||||
@@ -138,6 +152,27 @@ Supported Algorithms
|
||||
yespower-b2b generic yespower + blake2b
|
||||
zr5 Ziftr
|
||||
|
||||
Many variations of scrypt based algos can be mine by specifying their
|
||||
parameters:
|
||||
|
||||
scryptn2: --algo scrypt --param-n 1048576
|
||||
|
||||
cpupower: --algo yespower --param-key "CPUpower: The number of CPU working or available for proof-of-work mining"
|
||||
|
||||
power2b: --algo yespower-b2b --param-n 2048 --param-r 32 --param-key "Now I am become Death, the destroyer of worlds"
|
||||
|
||||
sugarchain: --algo yespower --param-n 2048 -param-r 32 --param-key "Satoshi Nakamoto 31/Oct/2008 Proof-of-work is essentially one-CPU-one-vote"
|
||||
|
||||
yespoweriots: --algo yespower --param-n 2048 --param-key "Iots is committed to the development of IOT"
|
||||
|
||||
yespowerlitb: --algo yespower --param-n 2048 --param-r 32 --param-key "LITBpower: The number of LITB working or available for proof-of-work mini"
|
||||
|
||||
yespoweric: --algo yespower --param-n 2048 --param-r 32 --param-key "IsotopeC"
|
||||
|
||||
yespowerurx: --algo yespower --param-n 2048 --param-r 32 --param-key "UraniumX"
|
||||
|
||||
yespowerltncg: --algo yespower --param-n 2048 --param-r 32 --param-key "LTNCGYES"
|
||||
|
||||
Errata
|
||||
------
|
||||
|
||||
|
||||
136
RELEASE_NOTES
136
RELEASE_NOTES
@@ -65,6 +65,142 @@ If not what makes it happen or not happen?
|
||||
Change Log
|
||||
----------
|
||||
|
||||
v3.12.8.2
|
||||
|
||||
Fixed x12 AVX2 rejects.
|
||||
Fixed phi AVX2 crash.
|
||||
|
||||
v3.12.8.1
|
||||
|
||||
Issue #261: Fixed yescryptr8g invalid shares.
|
||||
|
||||
v3.12.8
|
||||
|
||||
Yespower sha256 prehash made thread safe.
|
||||
|
||||
Rewrote diff conversion functions from scratch to be simpler and use
|
||||
long double (float80) and int128 arithmetic for improved accuracy and
|
||||
precision.
|
||||
|
||||
Some code cleanup and assorted small changes.
|
||||
|
||||
v3.12.7
|
||||
|
||||
Issue #257: fixed a file descriptor leak which caused the CPU temperature
|
||||
and frequency query to report zeros after mining for a couple of hours.
|
||||
|
||||
Issue #253: stale share reduction for yescrypt, sonoa.
|
||||
|
||||
v3.12.6.1
|
||||
|
||||
Issue #252: Fixed SSL mining (stratum+tcps://)
|
||||
|
||||
Issue #254 Fixed benchmark.
|
||||
|
||||
Issue #253: Implemented stale share reduction for yespower, x25x, x22i, x21s,
|
||||
x16*, scryptn2, more to come.
|
||||
|
||||
v3.12.6
|
||||
|
||||
Issue #246: improved stale share detection for getwork.
|
||||
|
||||
Improved precision of target_to_diff conversion from 4 digits to 20+.
|
||||
|
||||
Display hash and target debug data for all rejected shares.
|
||||
|
||||
A graphical representation of CPU affinity is displayed when using --threads.
|
||||
|
||||
Added highest and lowest accepted share to summary log.
|
||||
|
||||
Other small changes to logs to improve consistency and clarity.
|
||||
|
||||
v3.12.5
|
||||
|
||||
Issues #246 & #251: fixed incorrect share diff for stratum and getwork,
|
||||
fixed incorrect target diff for getwork. Stats should now be correct for
|
||||
getwork as well as stratum.
|
||||
|
||||
Issue #252: Fixed stratum+tcps not using curl ssl.
|
||||
|
||||
Getwork: reduce stale blocks, faster response to new work.
|
||||
|
||||
Added ntime to new job/work logs.
|
||||
|
||||
README.md now lists the parameters for yespower variations that don't have
|
||||
a specific algo name.
|
||||
|
||||
v3.12.4.6
|
||||
|
||||
Issue #246: fixed getwork repeated new block logs with same height. New work
|
||||
for the same block is now reported as "New work" instead of "New block".
|
||||
Also added a check that work is new before generating "New work" log.
|
||||
|
||||
Added target diff to getwork new block log.
|
||||
|
||||
Changed share ratio in share result log to simple fraction, no longer %.
|
||||
|
||||
Added debug log to display mininginfo, use -D.
|
||||
|
||||
v3.12.4.5
|
||||
|
||||
Issue #246: better stale share detection for getwork, and enhanced logging
|
||||
of stale shares for stratum & getwork.
|
||||
|
||||
Issue #251: fixed incorrect share difficulty and share ratio in share
|
||||
result log.
|
||||
|
||||
Changed submit log to include share diff and block height.
|
||||
|
||||
Small cosmetic changes to logs.
|
||||
|
||||
v3.12.4.4
|
||||
|
||||
Issue #246: Fixed net hashrate in getwork block log,
|
||||
removed duplicate getwork block log,
|
||||
other small tweaks to stats logs for getwork.
|
||||
|
||||
Issue #248: Fixed chronic stale shares with scrypt:1048576 (scryptn2).
|
||||
|
||||
v3.12.4.3
|
||||
|
||||
Fixed segfault in new block log for getwork.
|
||||
|
||||
Disabled silent discarding of stale work after the submit is logged.
|
||||
|
||||
v3.12.4.2
|
||||
|
||||
Issue #245: fixed getwork stale shares, solo mining with getwork now works.
|
||||
|
||||
Issue #246: implemented block and summary logs for getwork.
|
||||
|
||||
v3.12.4.1
|
||||
|
||||
Issue #245: fix scantime when mining solo with getwork.
|
||||
|
||||
Added debug logs for creation of stratum and longpoll threads, use -D to
|
||||
enable.
|
||||
|
||||
v3.12.4
|
||||
|
||||
Issue #244: Change longpoll to ignore job id.
|
||||
|
||||
Lyra2rev2 AVX2 +3%, AVX512 +6%.
|
||||
|
||||
v3.12.3.1
|
||||
|
||||
Issue #241: Fixed regression that broke coinbase address in v3.11.7.
|
||||
|
||||
v3.12.3
|
||||
|
||||
Issue #238: Fixed skunk AVX2.
|
||||
|
||||
Issue #239: Faster AVX2 & AVX512 for skein +44%, skein2 +30%, plus marginal
|
||||
increases for skunk, x16r, x16rv2, x16rt, x16rt-veil, x16s, x21s.
|
||||
|
||||
Faster anime VAES +57%, AVX512 +21%, AVX2 +3%.
|
||||
|
||||
Redesigned code reponsible for #236.
|
||||
|
||||
v3.12.2
|
||||
|
||||
Fixed xevan, skein, skein2 AVX2, #238.
|
||||
|
||||
@@ -97,13 +97,10 @@ int null_scanhash()
|
||||
return 0;
|
||||
}
|
||||
|
||||
void null_hash()
|
||||
int null_hash()
|
||||
{
|
||||
applog(LOG_WARNING,"SWERR: null_hash unsafe null function");
|
||||
};
|
||||
void null_hash_suw()
|
||||
{
|
||||
applog(LOG_WARNING,"SWERR: null_hash_suw unsafe null function");
|
||||
return 0;
|
||||
};
|
||||
|
||||
void init_algo_gate( algo_gate_t* gate )
|
||||
@@ -111,12 +108,10 @@ void init_algo_gate( algo_gate_t* gate )
|
||||
gate->miner_thread_init = (void*)&return_true;
|
||||
gate->scanhash = (void*)&null_scanhash;
|
||||
gate->hash = (void*)&null_hash;
|
||||
gate->hash_suw = (void*)&null_hash_suw;
|
||||
gate->get_new_work = (void*)&std_get_new_work;
|
||||
gate->work_decode = (void*)&std_le_work_decode;
|
||||
gate->decode_extra_data = (void*)&do_nothing;
|
||||
gate->gen_merkle_root = (void*)&sha256d_gen_merkle_root;
|
||||
gate->stratum_gen_work = (void*)&std_stratum_gen_work;
|
||||
gate->build_stratum_request = (void*)&std_le_build_stratum_request;
|
||||
gate->malloc_txs_request = (void*)&std_malloc_txs_request;
|
||||
gate->submit_getwork_result = (void*)&std_le_submit_getwork_result;
|
||||
@@ -230,11 +225,6 @@ bool register_algo_gate( int algo, algo_gate_t *gate )
|
||||
case ALGO_X22I: register_x22i_algo ( gate ); break;
|
||||
case ALGO_X25X: register_x25x_algo ( gate ); break;
|
||||
case ALGO_XEVAN: register_xevan_algo ( gate ); break;
|
||||
/* case ALGO_YESCRYPT: register_yescrypt_05_algo ( gate ); break;
|
||||
case ALGO_YESCRYPTR8: register_yescryptr8_05_algo ( gate ); break;
|
||||
case ALGO_YESCRYPTR16: register_yescryptr16_05_algo ( gate ); break;
|
||||
case ALGO_YESCRYPTR32: register_yescryptr32_05_algo ( gate ); break;
|
||||
*/
|
||||
case ALGO_YESCRYPT: register_yescrypt_algo ( gate ); break;
|
||||
case ALGO_YESCRYPTR8: register_yescryptr8_algo ( gate ); break;
|
||||
case ALGO_YESCRYPTR8G: register_yescryptr8g_algo ( gate ); break;
|
||||
@@ -281,39 +271,37 @@ void exec_hash_function( int algo, void *output, const void *pdata )
|
||||
const char* const algo_alias_map[][2] =
|
||||
{
|
||||
// alias proper
|
||||
{ "argon2d-crds", "argon2d250" },
|
||||
{ "argon2d-dyn", "argon2d500" },
|
||||
{ "argon2d-uis", "argon2d4096" },
|
||||
{ "bcd", "x13bcd" },
|
||||
{ "bitcore", "timetravel10" },
|
||||
{ "bitzeny", "yescryptr8" },
|
||||
{ "blake256r8", "blakecoin" },
|
||||
{ "blake256r8vnl", "vanilla" },
|
||||
{ "blake256r14", "blake" },
|
||||
{ "blake256r14dcr", "decred" },
|
||||
{ "cryptonote", "cryptonight" },
|
||||
{ "cryptonight-light", "cryptolight" },
|
||||
{ "diamond", "dmd-gr" },
|
||||
{ "droplp", "drop" },
|
||||
{ "espers", "hmq1725" },
|
||||
{ "flax", "c11" },
|
||||
{ "hsr", "x13sm3" },
|
||||
{ "jackpot", "jha" },
|
||||
{ "jane", "scryptjane" },
|
||||
{ "lyra2", "lyra2re" },
|
||||
{ "lyra2v2", "lyra2rev2" },
|
||||
{ "lyra2v3", "lyra2rev3" },
|
||||
{ "myrgr", "myr-gr" },
|
||||
{ "myriad", "myr-gr" },
|
||||
{ "neo", "neoscrypt" },
|
||||
{ "phi", "phi1612" },
|
||||
{ "sib", "x11gost" },
|
||||
{ "timetravel8", "timetravel" },
|
||||
{ "veil", "x16rt-veil" },
|
||||
{ "x16r-hex", "hex" },
|
||||
{ "yenten", "yescryptr16" },
|
||||
{ "ziftr", "zr5" },
|
||||
{ NULL, NULL }
|
||||
{ "argon2d-crds", "argon2d250" },
|
||||
{ "argon2d-dyn", "argon2d500" },
|
||||
{ "argon2d-uis", "argon2d4096" },
|
||||
{ "bcd", "x13bcd" },
|
||||
{ "bitcore", "timetravel10" },
|
||||
{ "bitzeny", "yescryptr8" },
|
||||
{ "blake256r8", "blakecoin" },
|
||||
{ "blake256r8vnl", "vanilla" },
|
||||
{ "blake256r14", "blake" },
|
||||
{ "blake256r14dcr", "decred" },
|
||||
{ "diamond", "dmd-gr" },
|
||||
{ "espers", "hmq1725" },
|
||||
{ "flax", "c11" },
|
||||
{ "hsr", "x13sm3" },
|
||||
{ "jackpot", "jha" },
|
||||
{ "jane", "scryptjane" },
|
||||
{ "lyra2", "lyra2re" },
|
||||
{ "lyra2v2", "lyra2rev2" },
|
||||
{ "lyra2v3", "lyra2rev3" },
|
||||
{ "myrgr", "myr-gr" },
|
||||
{ "myriad", "myr-gr" },
|
||||
{ "neo", "neoscrypt" },
|
||||
{ "phi", "phi1612" },
|
||||
{ "scryptn2", "scrypt:1048576" },
|
||||
{ "sib", "x11gost" },
|
||||
{ "timetravel8", "timetravel" },
|
||||
{ "veil", "x16rt-veil" },
|
||||
{ "x16r-hex", "hex" },
|
||||
{ "yenten", "yescryptr16" },
|
||||
{ "ziftr", "zr5" },
|
||||
{ NULL, NULL }
|
||||
};
|
||||
|
||||
// if arg is a valid alias for a known algo it is updated with the proper
|
||||
|
||||
@@ -75,7 +75,7 @@
|
||||
|
||||
// my hack at creating a set data type using bit masks. Set inclusion,
|
||||
// exclusion union and intersection operations are provided for convenience. In // some cases it may be desireable to use boolean algebra directly on the
|
||||
// data to perfomr set operations. Sets can be represented as single
|
||||
// data to perform set operations. Sets can be represented as single
|
||||
// elements, a bitwise OR of multiple elements, a bitwise OR of multiple
|
||||
// set variables or constants, or combinations of the above.
|
||||
// Examples:
|
||||
@@ -110,12 +110,11 @@ inline bool set_excl ( set_t a, set_t b ) { return (a & b) == 0; }
|
||||
|
||||
typedef struct
|
||||
{
|
||||
// mandatory functions, must be overwritten
|
||||
// mandatory function, must be overwritten
|
||||
int ( *scanhash ) ( struct work*, uint32_t, uint64_t*, struct thr_info* );
|
||||
|
||||
// optional unsafe, must be overwritten if algo uses function
|
||||
void ( *hash ) ( void*, const void*, uint32_t ) ;
|
||||
void ( *hash_suw ) ( void*, const void* );
|
||||
// Deprecated, will be removed
|
||||
int ( *hash ) ( void*, const void*, uint32_t ) ;
|
||||
|
||||
//optional, safe to use default in most cases
|
||||
|
||||
@@ -123,9 +122,6 @@ void ( *hash_suw ) ( void*, const void* );
|
||||
// threads.
|
||||
bool ( *miner_thread_init ) ( int );
|
||||
|
||||
// Generate global blockheader from stratum data.
|
||||
void ( *stratum_gen_work ) ( struct stratum_ctx*, struct work* );
|
||||
|
||||
// Get thread local copy of blockheader with unique nonce.
|
||||
void ( *get_new_work ) ( struct work*, struct work*, int, uint32_t* );
|
||||
|
||||
@@ -165,7 +161,9 @@ bool ( *do_this_thread ) ( int );
|
||||
// After do_this_thread
|
||||
void ( *resync_threads ) ( struct work* );
|
||||
|
||||
// No longer needed
|
||||
json_t* (*longpoll_rpc_call) ( CURL*, int*, char* );
|
||||
|
||||
set_t optimizations;
|
||||
int ( *get_work_data_size ) ();
|
||||
int ntime_index;
|
||||
@@ -213,16 +211,13 @@ void four_way_not_tested();
|
||||
int null_scanhash();
|
||||
|
||||
// displays warning
|
||||
void null_hash ();
|
||||
void null_hash_suw();
|
||||
int null_hash ();
|
||||
|
||||
// optional safe targets, default listed first unless noted.
|
||||
|
||||
void std_get_new_work( struct work *work, struct work *g_work, int thr_id,
|
||||
uint32_t* end_nonce_ptr );
|
||||
|
||||
void std_stratum_gen_work( struct stratum_ctx *sctx, struct work *work );
|
||||
|
||||
void sha256d_gen_merkle_root( char *merkle_root, struct stratum_ctx *sctx );
|
||||
void SHA256_gen_merkle_root ( char *merkle_root, struct stratum_ctx *sctx );
|
||||
|
||||
@@ -250,10 +245,6 @@ void std_build_block_header( struct work* g_work, uint32_t version,
|
||||
void std_build_extraheader( struct work *work, struct stratum_ctx *sctx );
|
||||
|
||||
json_t* std_longpoll_rpc_call( CURL *curl, int *err, char *lp_url );
|
||||
//json_t* jr2_longpoll_rpc_call( CURL *curl, int *err );
|
||||
|
||||
//bool std_stratum_handle_response( json_t *val );
|
||||
//bool jr2_stratum_handle_response( json_t *val );
|
||||
|
||||
bool std_ready_to_mine( struct work* work, struct stratum_ctx* stratum,
|
||||
int thr_id );
|
||||
@@ -272,11 +263,6 @@ bool register_algo_gate( int algo, algo_gate_t *gate );
|
||||
// compiler warnings but that's just more work for devs adding new algos.
|
||||
bool register_algo( algo_gate_t *gate );
|
||||
|
||||
// Overrides a common set of functions used by RPC2 and other RPC2-specific
|
||||
// init. Called by algo's register function before initializing algo-specific
|
||||
// functions and data.
|
||||
//bool register_json_rpc2( algo_gate_t *gate );
|
||||
|
||||
// use this to call the hash function of an algo directly, ie util.c test.
|
||||
void exec_hash_function( int algo, void *output, const void *pdata );
|
||||
|
||||
|
||||
@@ -48,7 +48,7 @@ int scanhash_blake_4way( struct work *work, uint32_t max_nonce,
|
||||
if ( fulltest( hash+(i<<3), ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n+i;
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
n += 4;
|
||||
|
||||
@@ -107,7 +107,7 @@ int scanhash_blake_8way( struct work *work, uint32_t max_nonce,
|
||||
if ( (hash+i)[7] <= HTarget && fulltest( hash+i, ptarget ) )
|
||||
{
|
||||
pdata[19] = n+i;
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
n += 8;
|
||||
|
||||
|
||||
@@ -45,7 +45,7 @@ int scanhash_blake2b_8way( struct work *work, uint32_t max_nonce,
|
||||
if ( fulltest( lane_hash, ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n + lane;
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
n += 8;
|
||||
@@ -100,7 +100,7 @@ int scanhash_blake2b_4way( struct work *work, uint32_t max_nonce,
|
||||
if ( fulltest( lane_hash, ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n + lane;
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
n += 4;
|
||||
|
||||
@@ -49,7 +49,7 @@ int scanhash_blake2s_16way( struct work *work, uint32_t max_nonce,
|
||||
if ( likely( fulltest( lane_hash, ptarget ) && !opt_benchmark ) )
|
||||
{
|
||||
pdata[19] = n + lane;
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
n += 16;
|
||||
@@ -104,7 +104,7 @@ int scanhash_blake2s_8way( struct work *work, uint32_t max_nonce,
|
||||
if ( likely( fulltest( lane_hash, ptarget ) && !opt_benchmark ) )
|
||||
{
|
||||
pdata[19] = n + lane;
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
n += 8;
|
||||
@@ -157,7 +157,7 @@ int scanhash_blake2s_4way( struct work *work, uint32_t max_nonce,
|
||||
if ( fulltest( lane_hash, ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n + lane;
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
n += 4;
|
||||
|
||||
@@ -49,7 +49,7 @@ int scanhash_blakecoin_4way( struct work *work, uint32_t max_nonce,
|
||||
&& !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n+i;
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
n += 4;
|
||||
|
||||
@@ -108,7 +108,7 @@ int scanhash_blakecoin_8way( struct work *work, uint32_t max_nonce,
|
||||
&& !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n+i;
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
n += 8;
|
||||
} while ( (n < max_nonce) && !work_restart[thr_id].restart );
|
||||
|
||||
@@ -62,7 +62,7 @@ int scanhash_decred_4way( struct work *work, uint32_t max_nonce,
|
||||
if ( fulltest( hash+(i<<3), ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[DECRED_NONCE_INDEX] = n+i;
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
n += 4;
|
||||
} while ( (n < max_nonce) && !work_restart[thr_id].restart );
|
||||
|
||||
@@ -105,7 +105,7 @@ int scanhash_pentablake_4way( struct work *work,
|
||||
&& fulltest( hash+(i<<3), ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n + i;
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
n += 4;
|
||||
|
||||
|
||||
@@ -46,7 +46,7 @@ int scanhash_bmw512_8way( struct work *work, uint32_t max_nonce,
|
||||
if ( fulltest( lane_hash, ptarget ) )
|
||||
{
|
||||
pdata[19] = n + lane;
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
n += 8;
|
||||
@@ -99,7 +99,7 @@ int scanhash_bmw512_4way( struct work *work, uint32_t max_nonce,
|
||||
if ( fulltest( lane_hash, ptarget ) )
|
||||
{
|
||||
pdata[19] = n + lane;
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
n += 4;
|
||||
|
||||
@@ -179,14 +179,6 @@ int cube_4way_full( cube_4way_context *sp, void *output, int hashbitlen,
|
||||
sp->rounds = 16;
|
||||
sp->pos = 0;
|
||||
|
||||
h[ 0] = m512_const1_128( iv[0] );
|
||||
h[ 1] = m512_const1_128( iv[1] );
|
||||
h[ 2] = m512_const1_128( iv[2] );
|
||||
h[ 3] = m512_const1_128( iv[3] );
|
||||
h[ 4] = m512_const1_128( iv[4] );
|
||||
h[ 5] = m512_const1_128( iv[5] );
|
||||
h[ 6] = m512_const1_128( iv[6] );
|
||||
h[ 7] = m512_const1_128( iv[7] );
|
||||
h[ 0] = m512_const1_128( iv[0] );
|
||||
h[ 1] = m512_const1_128( iv[1] );
|
||||
h[ 2] = m512_const1_128( iv[2] );
|
||||
@@ -447,14 +439,6 @@ int cube_2way_full( cube_2way_context *sp, void *output, int hashbitlen,
|
||||
sp->rounds = 16;
|
||||
sp->pos = 0;
|
||||
|
||||
h[ 0] = m256_const1_128( iv[0] );
|
||||
h[ 1] = m256_const1_128( iv[1] );
|
||||
h[ 2] = m256_const1_128( iv[2] );
|
||||
h[ 3] = m256_const1_128( iv[3] );
|
||||
h[ 4] = m256_const1_128( iv[4] );
|
||||
h[ 5] = m256_const1_128( iv[5] );
|
||||
h[ 6] = m256_const1_128( iv[6] );
|
||||
h[ 7] = m256_const1_128( iv[7] );
|
||||
h[ 0] = m256_const1_128( iv[0] );
|
||||
h[ 1] = m256_const1_128( iv[1] );
|
||||
h[ 2] = m256_const1_128( iv[2] );
|
||||
|
||||
@@ -28,6 +28,27 @@ int cube_4way_update_close( cube_4way_context *sp, void *output,
|
||||
int cube_4way_full( cube_4way_context *sp, void *output, int hashbitlen,
|
||||
const void *data, size_t size );
|
||||
|
||||
int cube_4x256_full( cube_4way_context *sp, void *output, int hashbitlen,
|
||||
const void *data, size_t size );
|
||||
|
||||
#define cube512_4way_init( sp ) cube_4way_update( sp, 512 )
|
||||
#define cube512_4way_update cube_4way_update
|
||||
#define cube512_4way_update_close cube_4way_update
|
||||
#define cube512_4way_close cube_4way_update
|
||||
#define cube512_4way_full( sp, output, data, size ) \
|
||||
cube_4way_full( sp, output, 512, data, size )
|
||||
#define cube512_4x256_full( sp, output, data, size ) \
|
||||
cube_4x256_full( sp, output, 512, data, size )
|
||||
|
||||
#define cube256_4way_init( sp ) cube_4way_update( sp, 256 )
|
||||
#define cube256_4way_update cube_4way_update
|
||||
#define cube256_4way_update_close cube_4way_update
|
||||
#define cube256_4way_close cube_4way_update
|
||||
#define cube256_4way_full( sp, output, data, size ) \
|
||||
cube_4way_full( sp, output, 256, data, size )
|
||||
#define cube256_4x256_full( sp, output, data, size ) \
|
||||
cube_4x256_full( sp, output, 256, data, size )
|
||||
|
||||
#endif
|
||||
|
||||
// 2x128, 2 way parallel SSE2
|
||||
|
||||
@@ -22,18 +22,26 @@ typedef struct
|
||||
} echo_4way_context __attribute__ ((aligned (64)));
|
||||
|
||||
int echo_4way_init( echo_4way_context *state, int hashbitlen );
|
||||
|
||||
#define echo512_4way_init( state ) echo_4way_init( state, 512 )
|
||||
#define echo256_4way_init( state ) echo_4way_init( state, 256 )
|
||||
|
||||
int echo_4way_update( echo_4way_context *state, const void *data,
|
||||
unsigned int databitlen);
|
||||
#define echo512_4way_update echo_4way_update
|
||||
|
||||
int echo_close( echo_4way_context *state, void *hashval );
|
||||
#define echo512_4way_close echo_4way_close
|
||||
|
||||
int echo_4way_update_close( echo_4way_context *state, void *hashval,
|
||||
const void *data, int databitlen );
|
||||
#define echo512_4way_update_close echo_4way_update_close
|
||||
|
||||
int echo_4way_full( echo_4way_context *ctx, void *hashval, int nHashSize,
|
||||
const void *data, int datalen );
|
||||
#define echo512_4way_full( state, hashval, data, datalen ) \
|
||||
echo_4way_full( state, hashval, 512, data, datalen )
|
||||
#define echo256_4way_full( state, hashval, data, datalen ) \
|
||||
echo_4way_full( state, hashval, 256, data, datalen )
|
||||
|
||||
#endif
|
||||
#endif
|
||||
|
||||
@@ -74,6 +74,14 @@ void sph_fugue512_close(void *cc, void *dst);
|
||||
void sph_fugue512_addbits_and_close(
|
||||
void *cc, unsigned ub, unsigned n, void *dst);
|
||||
|
||||
#define sph_fugue512_full( cc, dst, data, len ) \
|
||||
do{ \
|
||||
sph_fugue512_init( cc ); \
|
||||
sph_fugue512( cc, data, len ); \
|
||||
sph_fugue512_close( cc, dst ); \
|
||||
}while(0)
|
||||
|
||||
|
||||
#ifdef __cplusplus
|
||||
}
|
||||
#endif
|
||||
|
||||
@@ -53,7 +53,7 @@ int scanhash_groestl_4way( struct work *work, uint32_t max_nonce,
|
||||
if ( fulltest( hash+(lane<<3), ptarget) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n + lane;
|
||||
submit_lane_solution( work, hash+(lane<<3), mythr, lane );
|
||||
submit_solution( work, hash+(lane<<3), mythr );
|
||||
}
|
||||
n += 4;
|
||||
} while ( ( n < last_nonce ) && !work_restart[thr_id].restart );
|
||||
|
||||
@@ -143,7 +143,7 @@ int scanhash_myriad_8way( struct work *work, uint32_t max_nonce,
|
||||
if ( fulltest( lane_hash, ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n + lane;
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
n += 8;
|
||||
@@ -226,7 +226,7 @@ int scanhash_myriad_4way( struct work *work, uint32_t max_nonce,
|
||||
if ( fulltest( lane_hash, ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n + lane;
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
n += 4;
|
||||
|
||||
@@ -129,7 +129,7 @@ int scanhash_jha_4way( struct work *work, uint32_t max_nonce,
|
||||
if ( fulltest( hash+(i<<3), ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n+i;
|
||||
submit_lane_solution( work, lane_hash, mythr, i );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
n += 4;
|
||||
|
||||
@@ -45,7 +45,7 @@ int scanhash_keccak_8way( struct work *work, uint32_t max_nonce,
|
||||
if ( valid_hash( lane_hash, ptarget ) )
|
||||
{
|
||||
pdata[19] = bswap_32( n + lane );
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
*noncev = _mm512_add_epi32( *noncev,
|
||||
@@ -97,7 +97,7 @@ int scanhash_keccak_4way( struct work *work, uint32_t max_nonce,
|
||||
if ( valid_hash( lane_hash, ptarget ))
|
||||
{
|
||||
pdata[19] = bswap_32( n + lane );
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
*noncev = _mm256_add_epi32( *noncev,
|
||||
|
||||
@@ -52,7 +52,7 @@ int scanhash_sha3d_8way( struct work *work, uint32_t max_nonce,
|
||||
if ( valid_hash( lane_hash, ptarget ) )
|
||||
{
|
||||
pdata[19] = bswap_32( n + lane );
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
*noncev = _mm512_add_epi32( *noncev,
|
||||
@@ -111,7 +111,7 @@ int scanhash_sha3d_4way( struct work *work, uint32_t max_nonce,
|
||||
if ( valid_hash( lane_hash, ptarget ) )
|
||||
{
|
||||
pdata[19] = bswap_32( n + lane );
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
*noncev = _mm256_add_epi32( *noncev,
|
||||
|
||||
@@ -245,7 +245,7 @@ int scanhash_allium_16way( struct work *work, uint32_t max_nonce,
|
||||
if ( unlikely( valid_hash( hash+(lane<<3), ptarget ) && !bench ) )
|
||||
{
|
||||
pdata[19] = bswap_32( n + lane );
|
||||
submit_lane_solution( work, hash+(lane<<3), mythr, lane );
|
||||
submit_solution( work, hash+(lane<<3), mythr );
|
||||
}
|
||||
*noncev = _mm512_add_epi32( *noncev, m512_const1_32( 16 ) );
|
||||
n += 16;
|
||||
@@ -394,7 +394,7 @@ int scanhash_allium_8way( struct work *work, uint32_t max_nonce,
|
||||
if ( unlikely( valid_hash( lane_hash, ptarget ) && !bench ) )
|
||||
{
|
||||
pdata[19] = bswap_32( n + lane );
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
n += 8;
|
||||
|
||||
@@ -94,12 +94,12 @@ bool lyra2rev2_thread_init()
|
||||
const int64_t ROW_LEN_BYTES = ROW_LEN_INT64 * 8;
|
||||
|
||||
int size = (int64_t)ROW_LEN_BYTES * 4; // nRows;
|
||||
#if defined (LYRA2REV2_8WAY)
|
||||
#if defined (LYRA2REV2_16WAY)
|
||||
l2v2_wholeMatrix = _mm_malloc( 2 * size, 64 ); // 2 way
|
||||
init_lyra2rev2_8way_ctx();;
|
||||
#elif defined (LYRA2REV2_4WAY)
|
||||
init_lyra2rev2_16way_ctx();;
|
||||
#elif defined (LYRA2REV2_8WAY)
|
||||
l2v2_wholeMatrix = _mm_malloc( size, 64 );
|
||||
init_lyra2rev2_4way_ctx();;
|
||||
init_lyra2rev2_8way_ctx();;
|
||||
#else
|
||||
l2v2_wholeMatrix = _mm_malloc( size, 64 );
|
||||
init_lyra2rev2_ctx();
|
||||
@@ -109,12 +109,12 @@ bool lyra2rev2_thread_init()
|
||||
|
||||
bool register_lyra2rev2_algo( algo_gate_t* gate )
|
||||
{
|
||||
#if defined (LYRA2REV2_8WAY)
|
||||
#if defined (LYRA2REV2_16WAY)
|
||||
gate->scanhash = (void*)&scanhash_lyra2rev2_16way;
|
||||
gate->hash = (void*)&lyra2rev2_16way_hash;
|
||||
#elif defined (LYRA2REV2_8WAY)
|
||||
gate->scanhash = (void*)&scanhash_lyra2rev2_8way;
|
||||
gate->hash = (void*)&lyra2rev2_8way_hash;
|
||||
#elif defined (LYRA2REV2_4WAY)
|
||||
gate->scanhash = (void*)&scanhash_lyra2rev2_4way;
|
||||
gate->hash = (void*)&lyra2rev2_4way_hash;
|
||||
#else
|
||||
gate->scanhash = (void*)&scanhash_lyra2rev2;
|
||||
gate->hash = (void*)&lyra2rev2_hash;
|
||||
|
||||
@@ -51,30 +51,32 @@ bool init_lyra2rev3_ctx();
|
||||
//////////////////////////////////
|
||||
|
||||
#if defined(__AVX512F__) && defined(__AVX512VL__) && defined(__AVX512DQ__) && defined(__AVX512BW__)
|
||||
#define LYRA2REV2_8WAY 1
|
||||
#define LYRA2REV2_16WAY 1
|
||||
#elif defined(__AVX2__)
|
||||
#define LYRA2REV2_4WAY 1
|
||||
#define LYRA2REV2_8WAY 1
|
||||
#endif
|
||||
|
||||
extern __thread uint64_t* l2v2_wholeMatrix;
|
||||
|
||||
bool register_lyra2rev2_algo( algo_gate_t* gate );
|
||||
|
||||
#if defined(LYRA2REV2_8WAY)
|
||||
#if defined(LYRA2REV2_16WAY)
|
||||
|
||||
void lyra2rev2_16way_hash( void *state, const void *input );
|
||||
int scanhash_lyra2rev2_16way( struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr );
|
||||
bool init_lyra2rev2_16way_ctx();
|
||||
|
||||
#elif defined(LYRA2REV2_8WAY)
|
||||
|
||||
void lyra2rev2_8way_hash( void *state, const void *input );
|
||||
int scanhash_lyra2rev2_8way( struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr );
|
||||
bool init_lyra2rev2_8way_ctx();
|
||||
|
||||
#elif defined(LYRA2REV2_4WAY)
|
||||
|
||||
void lyra2rev2_4way_hash( void *state, const void *input );
|
||||
int scanhash_lyra2rev2_4way( struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr );
|
||||
bool init_lyra2rev2_4way_ctx();
|
||||
|
||||
#else
|
||||
|
||||
void lyra2rev2_hash( void *state, const void *input );
|
||||
int scanhash_lyra2rev2( struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr );
|
||||
|
||||
@@ -76,7 +76,7 @@ int scanhash_lyra2h_4way( struct work *work, uint32_t max_nonce,
|
||||
&& !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n+i;
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
n += 4;
|
||||
} while ( (n < max_nonce-4) && !work_restart[thr_id].restart);
|
||||
|
||||
@@ -8,12 +8,30 @@
|
||||
#include "algo/cubehash/cube-hash-2way.h"
|
||||
|
||||
|
||||
#if 0
|
||||
void lyra2rev2_8way_hash( void *state, const void *input )
|
||||
#if defined (LYRA2REV2_16WAY)
|
||||
|
||||
typedef struct {
|
||||
blake256_16way_context blake;
|
||||
keccak256_8way_context keccak;
|
||||
cubehashParam cube;
|
||||
skein256_8way_context skein;
|
||||
bmw256_16way_context bmw;
|
||||
} lyra2v2_16way_ctx_holder __attribute__ ((aligned (64)));
|
||||
|
||||
static lyra2v2_16way_ctx_holder l2v2_16way_ctx;
|
||||
|
||||
bool init_lyra2rev2_16way_ctx()
|
||||
{
|
||||
uint32_t vhash[8*8] __attribute__ ((aligned (128)));
|
||||
uint32_t vhashA[8*8] __attribute__ ((aligned (64)));
|
||||
uint32_t vhashB[8*8] __attribute__ ((aligned (64)));
|
||||
keccak256_8way_init( &l2v2_16way_ctx.keccak );
|
||||
cubehashInit( &l2v2_16way_ctx.cube, 256, 16, 32 );
|
||||
skein256_8way_init( &l2v2_16way_ctx.skein );
|
||||
bmw256_16way_init( &l2v2_16way_ctx.bmw );
|
||||
return true;
|
||||
}
|
||||
|
||||
void lyra2rev2_16way_hash( void *state, const void *input )
|
||||
{
|
||||
uint32_t vhash[8*16] __attribute__ ((aligned (128)));
|
||||
uint32_t hash0[8] __attribute__ ((aligned (64)));
|
||||
uint32_t hash1[8] __attribute__ ((aligned (64)));
|
||||
uint32_t hash2[8] __attribute__ ((aligned (64)));
|
||||
@@ -22,35 +40,60 @@ void lyra2rev2_8way_hash( void *state, const void *input )
|
||||
uint32_t hash5[8] __attribute__ ((aligned (64)));
|
||||
uint32_t hash6[8] __attribute__ ((aligned (64)));
|
||||
uint32_t hash7[8] __attribute__ ((aligned (64)));
|
||||
lyra2v2_8way_ctx_holder ctx __attribute__ ((aligned (64)));
|
||||
memcpy( &ctx, &l2v2_8way_ctx, sizeof(l2v2_8way_ctx) );
|
||||
uint32_t hash8[8] __attribute__ ((aligned (64)));
|
||||
uint32_t hash9[8] __attribute__ ((aligned (64)));
|
||||
uint32_t hash10[8] __attribute__ ((aligned (64)));
|
||||
uint32_t hash11[8] __attribute__ ((aligned (64)));
|
||||
uint32_t hash12[8] __attribute__ ((aligned (64)));
|
||||
uint32_t hash13[8] __attribute__ ((aligned (64)));
|
||||
uint32_t hash14[8] __attribute__ ((aligned (64)));
|
||||
uint32_t hash15[8] __attribute__ ((aligned (64)));
|
||||
lyra2v2_16way_ctx_holder ctx __attribute__ ((aligned (64)));
|
||||
memcpy( &ctx, &l2v2_16way_ctx, sizeof(l2v2_16way_ctx) );
|
||||
|
||||
blake256_8way_update( &ctx.blake, input + (64<<3), 16 );
|
||||
blake256_8way_close( &ctx.blake, vhash );
|
||||
blake256_16way_update( &ctx.blake, input + (64<<4), 16 );
|
||||
blake256_16way_close( &ctx.blake, vhash );
|
||||
|
||||
rintrlv_8x32_8x64( vhashA, vhash, 256 );
|
||||
dintrlv_16x32( hash0, hash1, hash2, hash3,
|
||||
hash4, hash5, hash6, hash7,
|
||||
hash8, hash9, hash10, hash11,
|
||||
hash12, hash13, hash14, hash15, vhash, 256 );
|
||||
|
||||
keccak256_8way_update( &ctx.keccak, vhashA, 32 );
|
||||
intrlv_8x64( vhash, hash0, hash1, hash2, hash3,
|
||||
hash4, hash5, hash6, hash7, 256 );
|
||||
|
||||
keccak256_8way_update( &ctx.keccak, vhash, 32 );
|
||||
keccak256_8way_close( &ctx.keccak, vhash );
|
||||
|
||||
dintrlv_8x64( hash0, hash1, hash2, hash3,
|
||||
hash4, hash5, hash6, hash7, vhash, 256 );
|
||||
intrlv_8x64( vhash, hash8, hash9, hash10, hash11,
|
||||
hash12, hash13, hash14, hash15, 256 );
|
||||
|
||||
cubehash_full( &ctx.cube, (byte*) hash0, 256, (const byte*) hash0, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash1, 256, (const byte*) hash1, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash2, 256, (const byte*) hash2, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash3, 256, (const byte*) hash3, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash4, 256, (const byte*) hash4, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash5, 256, (const byte*) hash5, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash6, 256, (const byte*) hash6, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash7, 256, (const byte*) hash7, 32 );
|
||||
keccak256_8way_init( &ctx.keccak );
|
||||
keccak256_8way_update( &ctx.keccak, vhash, 32 );
|
||||
keccak256_8way_close( &ctx.keccak, vhash );
|
||||
|
||||
dintrlv_8x64( hash8, hash9, hash10, hash11,
|
||||
hash12, hash13, hash14, hash5, vhash, 256 );
|
||||
|
||||
cubehash_full( &ctx.cube, (byte*) hash0, 256, (const byte*) hash0, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash1, 256, (const byte*) hash1, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash2, 256, (const byte*) hash2, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash3, 256, (const byte*) hash3, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash4, 256, (const byte*) hash4, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash5, 256, (const byte*) hash5, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash6, 256, (const byte*) hash6, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash7, 256, (const byte*) hash7, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash8, 256, (const byte*) hash8, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash9, 256, (const byte*) hash9, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash10, 256, (const byte*) hash10, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash11, 256, (const byte*) hash11, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash12, 256, (const byte*) hash12, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash13, 256, (const byte*) hash13, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash14, 256, (const byte*) hash14, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash15, 256, (const byte*) hash15, 32 );
|
||||
|
||||
// cube_4way_update_close( &ctx.cube, vhashA, vhashA, 32 );
|
||||
// cube_4way_init( &ctx.cube, 256, 16, 32 );
|
||||
// cube_4way_update_close( &ctx.cube, vhashB, vhashB, 32 );
|
||||
//
|
||||
// dintrlv_4x128( hash0, hash1, hash2, hash3, vhashA, 256 );
|
||||
// dintrlv_4x128( hash4, hash5, hash6, hash7, vhashB, 256 );
|
||||
|
||||
intrlv_2x256( vhash, hash0, hash1, 256 );
|
||||
LYRA2REV2_2WAY( l2v2_wholeMatrix, vhash, 32, vhash, 32, 1, 4, 4 );
|
||||
@@ -64,61 +107,127 @@ void lyra2rev2_8way_hash( void *state, const void *input )
|
||||
intrlv_2x256( vhash, hash6, hash7, 256 );
|
||||
LYRA2REV2_2WAY( l2v2_wholeMatrix, vhash, 32, vhash, 32, 1, 4, 4 );
|
||||
dintrlv_2x256( hash6, hash7, vhash, 256 );
|
||||
intrlv_2x256( vhash, hash8, hash9, 256 );
|
||||
LYRA2REV2_2WAY( l2v2_wholeMatrix, vhash, 32, vhash, 32, 1, 4, 4 );
|
||||
dintrlv_2x256( hash8, hash9, vhash, 256 );
|
||||
intrlv_2x256( vhash, hash10, hash11, 256 );
|
||||
LYRA2REV2_2WAY( l2v2_wholeMatrix, vhash, 32, vhash, 32, 1, 4, 4 );
|
||||
dintrlv_2x256( hash10, hash11, vhash, 256 );
|
||||
intrlv_2x256( vhash, hash12, hash13, 256 );
|
||||
LYRA2REV2_2WAY( l2v2_wholeMatrix, vhash, 32, vhash, 32, 1, 4, 4 );
|
||||
dintrlv_2x256( hash12, hash13, vhash, 256 );
|
||||
intrlv_2x256( vhash, hash14, hash15, 256 );
|
||||
LYRA2REV2_2WAY( l2v2_wholeMatrix, vhash, 32, vhash, 32, 1, 4, 4 );
|
||||
dintrlv_2x256( hash14, hash15, vhash, 256 );
|
||||
|
||||
intrlv_8x64( vhash, hash0, hash1, hash2, hash3, hash4, hash5, hash6,
|
||||
hash7, 256 );
|
||||
|
||||
intrlv_8x64( vhash, hash0, hash1, hash2, hash3,
|
||||
hash4, hash5, hash6, hash7, 256 );
|
||||
skein256_8way_update( &ctx.skein, vhash, 32 );
|
||||
skein256_8way_close( &ctx.skein, vhash );
|
||||
|
||||
dintrlv_8x64( hash0, hash1, hash2, hash3,
|
||||
hash4, hash5, hash6, hash7, vhash, 256 );
|
||||
intrlv_8x64( vhash, hash8, hash9, hash10, hash11, hash12,
|
||||
hash13, hash14, hash15, 256 );
|
||||
|
||||
cubehash_full( &ctx.cube, (byte*) hash0, 256, (const byte*) hash0, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash1, 256, (const byte*) hash1, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash2, 256, (const byte*) hash2, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash3, 256, (const byte*) hash3, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash4, 256, (const byte*) hash4, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash5, 256, (const byte*) hash5, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash6, 256, (const byte*) hash6, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash7, 256, (const byte*) hash7, 32 );
|
||||
skein256_8way_init( &ctx.skein );
|
||||
skein256_8way_update( &ctx.skein, vhash, 32 );
|
||||
skein256_8way_close( &ctx.skein, vhash );
|
||||
|
||||
// cube_4way_init( &ctx.cube, 256, 16, 32 );
|
||||
// cube_4way_update_close( &ctx.cube, vhashA, vhashA, 32 );
|
||||
// cube_4way_init( &ctx.cube, 256, 16, 32 );
|
||||
// cube_4way_update_close( &ctx.cube, vhashB, vhashB, 32 );
|
||||
//
|
||||
// dintrlv_4x128( hash0, hash1, hash2, hash3, vhashA, 256 );
|
||||
// dintrlv_4x128( hash4, hash5, hash6, hash7, vhashB, 256 );
|
||||
dintrlv_8x64( hash8, hash9, hash10, hash11,
|
||||
hash12, hash13, hash14, hash15, vhash, 256 );
|
||||
|
||||
intrlv_8x32( vhash, hash0, hash1, hash2, hash3, hash4, hash5, hash6,
|
||||
hash7, 256 );
|
||||
|
||||
cubehash_full( &ctx.cube, (byte*) hash0, 256, (const byte*) hash0, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash1, 256, (const byte*) hash1, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash2, 256, (const byte*) hash2, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash3, 256, (const byte*) hash3, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash4, 256, (const byte*) hash4, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash5, 256, (const byte*) hash5, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash6, 256, (const byte*) hash6, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash7, 256, (const byte*) hash7, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash8, 256, (const byte*) hash8, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash9, 256, (const byte*) hash9, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash10, 256, (const byte*) hash10, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash11, 256, (const byte*) hash11, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash12, 256, (const byte*) hash12, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash13, 256, (const byte*) hash13, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash14, 256, (const byte*) hash14, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash15, 256, (const byte*) hash15, 32 );
|
||||
|
||||
bmw256_8way_update( &ctx.bmw, vhash, 32 );
|
||||
bmw256_8way_close( &ctx.bmw, state );
|
||||
intrlv_16x32( vhash, hash0, hash1, hash2, hash3,
|
||||
hash4, hash5, hash6, hash7,
|
||||
hash8, hash9, hash10, hash11,
|
||||
hash12, hash13, hash14, hash15, 256 );
|
||||
|
||||
bmw256_16way_update( &ctx.bmw, vhash, 32 );
|
||||
bmw256_16way_close( &ctx.bmw, state );
|
||||
}
|
||||
#endif
|
||||
|
||||
int scanhash_lyra2rev2_16way( struct work *work, const uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr )
|
||||
{
|
||||
uint32_t hash[8*16] __attribute__ ((aligned (128)));
|
||||
uint32_t vdata[20*16] __attribute__ ((aligned (64)));
|
||||
uint32_t *hashd7 = &hash[7*16];
|
||||
uint32_t lane_hash[8] __attribute__ ((aligned (32)));
|
||||
uint32_t *pdata = work->data;
|
||||
uint32_t *ptarget = work->target;
|
||||
const uint32_t first_nonce = pdata[19];
|
||||
const uint32_t last_nonce = max_nonce - 16;
|
||||
uint32_t n = first_nonce;
|
||||
const uint32_t targ32 = ptarget[7];
|
||||
__m512i *noncev = (__m512i*)vdata + 19;
|
||||
const int thr_id = mythr->id;
|
||||
const bool bench = opt_benchmark;
|
||||
|
||||
if ( bench ) ptarget[7] = 0x0000ff;
|
||||
|
||||
mm512_bswap32_intrlv80_16x32( vdata, pdata );
|
||||
*noncev = _mm512_set_epi32( n+15, n+14, n+13, n+12, n+11, n+10, n+ 9, n+ 8,
|
||||
n+ 7, n+ 6, n+ 5, n+ 4, n+ 3, n+ 2, n+ 1, n );
|
||||
blake256_16way_init( &l2v2_16way_ctx.blake );
|
||||
blake256_16way_update( &l2v2_16way_ctx.blake, vdata, 64 );
|
||||
|
||||
#if defined (LYRA2REV2_8WAY)
|
||||
do
|
||||
{
|
||||
lyra2rev2_16way_hash( hash, vdata );
|
||||
|
||||
for ( int lane = 0; lane < 16; lane++ )
|
||||
if ( unlikely( hashd7[lane] <= targ32 ) )
|
||||
{
|
||||
extr_lane_16x32( lane_hash, hash, lane, 256 );
|
||||
if ( likely( valid_hash( lane_hash, ptarget ) && !bench ) )
|
||||
{
|
||||
pdata[19] = bswap_32( n + lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
*noncev = _mm512_add_epi32( *noncev, m512_const1_32( 16 ) );
|
||||
n += 16;
|
||||
} while ( likely( (n < last_nonce) && !work_restart[thr_id].restart ) );
|
||||
pdata[19] = n;
|
||||
*hashes_done = n - first_nonce;
|
||||
return 0;
|
||||
}
|
||||
|
||||
#elif defined (LYRA2REV2_8WAY)
|
||||
|
||||
typedef struct {
|
||||
blake256_8way_context blake;
|
||||
keccak256_8way_context keccak;
|
||||
keccak256_4way_context keccak;
|
||||
cubehashParam cube;
|
||||
skein256_8way_context skein;
|
||||
bmw256_8way_context bmw;
|
||||
skein256_4way_context skein;
|
||||
bmw256_8way_context bmw;
|
||||
} lyra2v2_8way_ctx_holder __attribute__ ((aligned (64)));
|
||||
|
||||
static lyra2v2_8way_ctx_holder l2v2_8way_ctx;
|
||||
|
||||
bool init_lyra2rev2_8way_ctx()
|
||||
{
|
||||
keccak256_8way_init( &l2v2_8way_ctx.keccak );
|
||||
keccak256_4way_init( &l2v2_8way_ctx.keccak );
|
||||
cubehashInit( &l2v2_8way_ctx.cube, 256, 16, 32 );
|
||||
skein256_8way_init( &l2v2_8way_ctx.skein );
|
||||
skein256_4way_init( &l2v2_8way_ctx.skein );
|
||||
bmw256_8way_init( &l2v2_8way_ctx.bmw );
|
||||
return true;
|
||||
}
|
||||
@@ -126,7 +235,6 @@ bool init_lyra2rev2_8way_ctx()
|
||||
void lyra2rev2_8way_hash( void *state, const void *input )
|
||||
{
|
||||
uint32_t vhash[8*8] __attribute__ ((aligned (128)));
|
||||
uint32_t vhashA[8*8] __attribute__ ((aligned (64)));
|
||||
uint32_t hash0[8] __attribute__ ((aligned (64)));
|
||||
uint32_t hash1[8] __attribute__ ((aligned (64)));
|
||||
uint32_t hash2[8] __attribute__ ((aligned (64)));
|
||||
@@ -141,14 +249,19 @@ void lyra2rev2_8way_hash( void *state, const void *input )
|
||||
blake256_8way_update( &ctx.blake, input + (64<<3), 16 );
|
||||
blake256_8way_close( &ctx.blake, vhash );
|
||||
|
||||
rintrlv_8x32_8x64( vhashA, vhash, 256 );
|
||||
|
||||
keccak256_8way_update( &ctx.keccak, vhashA, 32 );
|
||||
keccak256_8way_close( &ctx.keccak, vhash );
|
||||
|
||||
dintrlv_8x64( hash0, hash1, hash2, hash3,
|
||||
dintrlv_8x32( hash0, hash1, hash2, hash3,
|
||||
hash4, hash5, hash6, hash7, vhash, 256 );
|
||||
|
||||
intrlv_4x64( vhash, hash0, hash1, hash2, hash3, 256 );
|
||||
keccak256_4way_update( &ctx.keccak, vhash, 32 );
|
||||
keccak256_4way_close( &ctx.keccak, vhash );
|
||||
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 256 );
|
||||
intrlv_4x64( vhash, hash4, hash5, hash6, hash7, 256 );
|
||||
keccak256_4way_init( &ctx.keccak );
|
||||
keccak256_4way_update( &ctx.keccak, vhash, 32 );
|
||||
keccak256_4way_close( &ctx.keccak, vhash );
|
||||
dintrlv_4x64( hash4, hash5, hash6, hash7, vhash, 256 );
|
||||
|
||||
cubehash_full( &ctx.cube, (byte*) hash0, 256, (const byte*) hash0, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash1, 256, (const byte*) hash1, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash2, 256, (const byte*) hash2, 32 );
|
||||
@@ -158,27 +271,25 @@ void lyra2rev2_8way_hash( void *state, const void *input )
|
||||
cubehash_full( &ctx.cube, (byte*) hash6, 256, (const byte*) hash6, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash7, 256, (const byte*) hash7, 32 );
|
||||
|
||||
intrlv_2x256( vhash, hash0, hash1, 256 );
|
||||
LYRA2REV2_2WAY( l2v2_wholeMatrix, vhash, 32, vhash, 32, 1, 4, 4 );
|
||||
dintrlv_2x256( hash0, hash1, vhash, 256 );
|
||||
intrlv_2x256( vhash, hash2, hash3, 256 );
|
||||
LYRA2REV2_2WAY( l2v2_wholeMatrix, vhash, 32, vhash, 32, 1, 4, 4 );
|
||||
dintrlv_2x256( hash2, hash3, vhash, 256 );
|
||||
intrlv_2x256( vhash, hash4, hash5, 256 );
|
||||
LYRA2REV2_2WAY( l2v2_wholeMatrix, vhash, 32, vhash, 32, 1, 4, 4 );
|
||||
dintrlv_2x256( hash4, hash5, vhash, 256 );
|
||||
intrlv_2x256( vhash, hash6, hash7, 256 );
|
||||
LYRA2REV2_2WAY( l2v2_wholeMatrix, vhash, 32, vhash, 32, 1, 4, 4 );
|
||||
dintrlv_2x256( hash6, hash7, vhash, 256 );
|
||||
|
||||
intrlv_8x64( vhash, hash0, hash1, hash2, hash3, hash4, hash5, hash6,
|
||||
hash7, 256 );
|
||||
|
||||
skein256_8way_update( &ctx.skein, vhash, 32 );
|
||||
skein256_8way_close( &ctx.skein, vhash );
|
||||
|
||||
dintrlv_8x64( hash0, hash1, hash2, hash3,
|
||||
hash4, hash5, hash6, hash7, vhash, 256 );
|
||||
LYRA2REV2( l2v2_wholeMatrix, hash0, 32, hash0, 32, hash0, 32, 1, 4, 4 );
|
||||
LYRA2REV2( l2v2_wholeMatrix, hash1, 32, hash1, 32, hash1, 32, 1, 4, 4 );
|
||||
LYRA2REV2( l2v2_wholeMatrix, hash2, 32, hash2, 32, hash2, 32, 1, 4, 4 );
|
||||
LYRA2REV2( l2v2_wholeMatrix, hash3, 32, hash3, 32, hash3, 32, 1, 4, 4 );
|
||||
LYRA2REV2( l2v2_wholeMatrix, hash4, 32, hash4, 32, hash4, 32, 1, 4, 4 );
|
||||
LYRA2REV2( l2v2_wholeMatrix, hash5, 32, hash5, 32, hash5, 32, 1, 4, 4 );
|
||||
LYRA2REV2( l2v2_wholeMatrix, hash6, 32, hash6, 32, hash6, 32, 1, 4, 4 );
|
||||
LYRA2REV2( l2v2_wholeMatrix, hash7, 32, hash7, 32, hash7, 32, 1, 4, 4 );
|
||||
|
||||
intrlv_4x64( vhash, hash0, hash1, hash2, hash3, 256 );
|
||||
skein256_4way_update( &ctx.skein, vhash, 32 );
|
||||
skein256_4way_close( &ctx.skein, vhash );
|
||||
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 256 );
|
||||
intrlv_4x64( vhash, hash4, hash5, hash6, hash7, 256 );
|
||||
skein256_4way_init( &ctx.skein );
|
||||
skein256_4way_update( &ctx.skein, vhash, 32 );
|
||||
skein256_4way_close( &ctx.skein, vhash );
|
||||
dintrlv_4x64( hash4, hash5, hash6, hash7, vhash, 256 );
|
||||
|
||||
cubehash_full( &ctx.cube, (byte*) hash0, 256, (const byte*) hash0, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash1, 256, (const byte*) hash1, 32 );
|
||||
@@ -189,8 +300,8 @@ void lyra2rev2_8way_hash( void *state, const void *input )
|
||||
cubehash_full( &ctx.cube, (byte*) hash6, 256, (const byte*) hash6, 32 );
|
||||
cubehash_full( &ctx.cube, (byte*) hash7, 256, (const byte*) hash7, 32 );
|
||||
|
||||
intrlv_8x32( vhash, hash0, hash1, hash2, hash3, hash4, hash5, hash6,
|
||||
hash7, 256 );
|
||||
intrlv_8x32( vhash, hash0, hash1, hash2, hash3,
|
||||
hash4, hash5, hash6, hash7, 256 );
|
||||
|
||||
bmw256_8way_update( &ctx.bmw, vhash, 32 );
|
||||
bmw256_8way_close( &ctx.bmw, state );
|
||||
@@ -223,7 +334,6 @@ int scanhash_lyra2rev2_8way( struct work *work, const uint32_t max_nonce,
|
||||
do
|
||||
{
|
||||
lyra2rev2_8way_hash( hash, vdata );
|
||||
pdata[19] = n;
|
||||
|
||||
for ( int lane = 0; lane < 8; lane++ )
|
||||
if ( unlikely( hashd7[lane] <= targ32 ) )
|
||||
@@ -232,7 +342,7 @@ int scanhash_lyra2rev2_8way( struct work *work, const uint32_t max_nonce,
|
||||
if ( likely( valid_hash( lane_hash, ptarget ) && !bench ) )
|
||||
{
|
||||
pdata[19] = bswap_32( n + lane );
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
*noncev = _mm256_add_epi32( *noncev, m256_const1_32( 8 ) );
|
||||
@@ -243,6 +353,9 @@ int scanhash_lyra2rev2_8way( struct work *work, const uint32_t max_nonce,
|
||||
return 0;
|
||||
}
|
||||
|
||||
#endif
|
||||
|
||||
/*
|
||||
#elif defined (LYRA2REV2_4WAY)
|
||||
|
||||
typedef struct {
|
||||
@@ -356,7 +469,7 @@ int scanhash_lyra2rev2_4way( struct work *work, uint32_t max_nonce,
|
||||
if ( valid_hash( lane_hash, ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n + lane;
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
n += 4;
|
||||
@@ -367,3 +480,4 @@ int scanhash_lyra2rev2_4way( struct work *work, uint32_t max_nonce,
|
||||
}
|
||||
|
||||
#endif
|
||||
*/
|
||||
|
||||
@@ -165,7 +165,7 @@ int scanhash_lyra2rev3_16way( struct work *work, const uint32_t max_nonce,
|
||||
if ( likely( valid_hash( lane_hash, ptarget ) && !opt_benchmark ) )
|
||||
{
|
||||
pdata[19] = n + lane;
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
n += 16;
|
||||
@@ -284,7 +284,7 @@ int scanhash_lyra2rev3_8way( struct work *work, const uint32_t max_nonce,
|
||||
if ( likely( valid_hash( lane_hash, ptarget ) && !bench ) )
|
||||
{
|
||||
pdata[19] = bswap_32( n + lane );
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
*noncev = _mm256_add_epi32( *noncev, m256_const1_32( 8 ) );
|
||||
@@ -386,7 +386,7 @@ int scanhash_lyra2rev3_4way( struct work *work, const uint32_t max_nonce,
|
||||
if ( valid_hash( lane_hash, ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = bswap_32( n + lane );
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
*noncev = _mm_add_epi32( *noncev, m128_const1_32( 4 ) );
|
||||
|
||||
@@ -124,7 +124,7 @@ int scanhash_lyra2z_16way( struct work *work, uint32_t max_nonce,
|
||||
if ( unlikely( valid_hash( lane_hash, ptarget ) && !bench ) )
|
||||
{
|
||||
pdata[19] = bswap_32( n + lane );
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
*noncev = _mm512_add_epi32( *noncev, m512_const1_32( 16 ) );
|
||||
@@ -222,7 +222,7 @@ int scanhash_lyra2z_8way( struct work *work, uint32_t max_nonce,
|
||||
if ( unlikely( valid_hash( lane_hash, ptarget ) && !bench ) )
|
||||
{
|
||||
pdata[19] = bswap_32( n + lane );
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
*noncev = _mm256_add_epi32( *noncev, m256_const1_32( 8 ) );
|
||||
@@ -301,7 +301,7 @@ int scanhash_lyra2z_4way( struct work *work, uint32_t max_nonce,
|
||||
if ( unlikely( valid_hash( lane_hash, ptarget ) && !bench ) )
|
||||
{
|
||||
pdata[19] = bswap_32( n + lane );
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
*noncev = _mm_add_epi32( *noncev, m128_const1_32( 4 ) );
|
||||
|
||||
@@ -68,7 +68,7 @@ bool lyra2z330_thread_init()
|
||||
|
||||
bool register_lyra2z330_algo( algo_gate_t* gate )
|
||||
{
|
||||
gate->optimizations = SSE42_OPT | AVX2_OPT;
|
||||
gate->optimizations = SSE2_OPT | AVX2_OPT;
|
||||
gate->miner_thread_init = (void*)&lyra2z330_thread_init;
|
||||
gate->scanhash = (void*)&scanhash_lyra2z330;
|
||||
gate->hash = (void*)&lyra2z330_hash;
|
||||
|
||||
@@ -302,7 +302,7 @@ int scanhash_phi2_8way( struct work *work, uint32_t max_nonce,
|
||||
if ( valid_hash( lane_hash, ptarget ) )
|
||||
{
|
||||
be32enc( pdata + 19, n + lane );
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
n += 8;
|
||||
@@ -483,7 +483,7 @@ int scanhash_phi2_4way( struct work *work, uint32_t max_nonce,
|
||||
if ( valid_hash( lane_hash, ptarget ) )
|
||||
{
|
||||
be32enc( pdata + 19, n + lane );
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
edata[ 19 ] += 4;
|
||||
|
||||
@@ -311,7 +311,7 @@ bool register_m7m_algo( algo_gate_t *gate )
|
||||
{
|
||||
gate->optimizations = SHA_OPT;
|
||||
init_m7m_ctx();
|
||||
gate->scanhash = (void*)scanhash_m7m_hash;
|
||||
gate->scanhash = (void*)&scanhash_m7m_hash;
|
||||
gate->build_stratum_request = (void*)&std_be_build_stratum_request;
|
||||
gate->work_decode = (void*)&std_be_work_decode;
|
||||
gate->submit_getwork_result = (void*)&std_be_submit_getwork_result;
|
||||
|
||||
@@ -108,7 +108,7 @@ int scanhash_nist5_8way( struct work *work, uint32_t max_nonce,
|
||||
if ( fulltest( lane_hash, ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n + lane;
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
n += 8;
|
||||
@@ -196,7 +196,7 @@ int scanhash_nist5_4way( struct work *work, uint32_t max_nonce,
|
||||
if ( fulltest( lane_hash, ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n + lane;
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
n += 4;
|
||||
|
||||
@@ -1,18 +1,241 @@
|
||||
#include "cpuminer-config.h"
|
||||
#include "anime-gate.h"
|
||||
|
||||
#if defined (ANIME_4WAY)
|
||||
|
||||
#include <stdio.h>
|
||||
#include <string.h>
|
||||
#include <stdint.h>
|
||||
|
||||
#include "algo/blake/blake-hash-4way.h"
|
||||
#include "algo/bmw/bmw-hash-4way.h"
|
||||
#include "algo/skein/skein-hash-4way.h"
|
||||
#include "algo/jh/jh-hash-4way.h"
|
||||
#include "algo/keccak/keccak-hash-4way.h"
|
||||
#include "algo/groestl/aes_ni/hash-groestl.h"
|
||||
#if defined(__VAES__)
|
||||
#include "algo/groestl/groestl512-hash-4way.h"
|
||||
#endif
|
||||
|
||||
#if defined (ANIME_8WAY)
|
||||
|
||||
typedef struct {
|
||||
blake512_8way_context blake;
|
||||
bmw512_8way_context bmw;
|
||||
#if defined(__VAES__)
|
||||
groestl512_4way_context groestl;
|
||||
#else
|
||||
hashState_groestl groestl;
|
||||
#endif
|
||||
jh512_8way_context jh;
|
||||
skein512_8way_context skein;
|
||||
keccak512_8way_context keccak;
|
||||
} anime_8way_ctx_holder;
|
||||
|
||||
anime_8way_ctx_holder anime_8way_ctx __attribute__ ((aligned (64)));
|
||||
|
||||
void init_anime_8way_ctx()
|
||||
{
|
||||
blake512_8way_init( &anime_8way_ctx.blake );
|
||||
bmw512_8way_init( &anime_8way_ctx.bmw );
|
||||
#if defined(__VAES__)
|
||||
groestl512_4way_init( &anime_8way_ctx.groestl, 64 );
|
||||
#else
|
||||
init_groestl( &anime_8way_ctx.groestl, 64 );
|
||||
#endif
|
||||
skein512_8way_init( &anime_8way_ctx.skein );
|
||||
jh512_8way_init( &anime_8way_ctx.jh );
|
||||
keccak512_8way_init( &anime_8way_ctx.keccak );
|
||||
}
|
||||
|
||||
void anime_8way_hash( void *state, const void *input )
|
||||
{
|
||||
uint64_t vhash[8*8] __attribute__ ((aligned (128)));
|
||||
uint64_t vhashA[8*8] __attribute__ ((aligned (64)));
|
||||
uint64_t vhashB[8*8] __attribute__ ((aligned (64)));
|
||||
uint64_t vhashC[8*8] __attribute__ ((aligned (64)));
|
||||
#if !defined(__VAES__)
|
||||
uint64_t hash0[8] __attribute__ ((aligned (64)));
|
||||
uint64_t hash1[8] __attribute__ ((aligned (64)));
|
||||
uint64_t hash2[8] __attribute__ ((aligned (64)));
|
||||
uint64_t hash3[8] __attribute__ ((aligned (64)));
|
||||
uint64_t hash4[8] __attribute__ ((aligned (64)));
|
||||
uint64_t hash5[8] __attribute__ ((aligned (64)));
|
||||
uint64_t hash6[8] __attribute__ ((aligned (64)));
|
||||
uint64_t hash7[8] __attribute__ ((aligned (64)));
|
||||
#endif
|
||||
__m512i* vh = (__m512i*)vhash;
|
||||
__m512i* vhA = (__m512i*)vhashA;
|
||||
__m512i* vhB = (__m512i*)vhashB;
|
||||
__m512i* vhC = (__m512i*)vhashC;
|
||||
const __m512i bit3_mask = m512_const1_64( 8 );
|
||||
const __m512i zero = _mm512_setzero_si512();
|
||||
__mmask8 vh_mask;
|
||||
anime_8way_ctx_holder ctx;
|
||||
memcpy( &ctx, &anime_8way_ctx, sizeof(anime_8way_ctx) );
|
||||
|
||||
bmw512_8way_full( &ctx.bmw, vhash, input, 80 );
|
||||
|
||||
blake512_8way_full( &ctx.blake, vhash, vhash, 64 );
|
||||
|
||||
vh_mask = _mm512_cmpeq_epi64_mask( _mm512_and_si512( vh[0], bit3_mask ),
|
||||
zero );
|
||||
|
||||
#if defined(__VAES__)
|
||||
|
||||
rintrlv_8x64_4x128( vhashA, vhashB, vhash, 512 );
|
||||
|
||||
if ( ( vh_mask & 0x0f ) != 0x0f )
|
||||
groestl512_4way_full( &ctx.groestl, vhashA, vhashA, 64 );
|
||||
if ( ( vh_mask & 0xf0 ) != 0xf0 )
|
||||
groestl512_4way_full( &ctx.groestl, vhashB, vhashB, 64 );
|
||||
|
||||
rintrlv_4x128_8x64( vhashC, vhashA, vhashB, 512 );
|
||||
|
||||
#else
|
||||
|
||||
dintrlv_8x64_512( hash0, hash1, hash2, hash3,
|
||||
hash4, hash5, hash6, hash7, vhash );
|
||||
|
||||
if ( hash0[0] & 8 )
|
||||
groestl512_full( &ctx.groestl, (char*)hash0, (char*)hash0, 512 );
|
||||
if ( hash1[0] & 8 )
|
||||
groestl512_full( &ctx.groestl, (char*)hash1, (char*)hash1, 512 );
|
||||
if ( hash2[0] & 8)
|
||||
groestl512_full( &ctx.groestl, (char*)hash2, (char*)hash2, 512 );
|
||||
if ( hash3[0] & 8 )
|
||||
groestl512_full( &ctx.groestl, (char*)hash3, (char*)hash3, 512 );
|
||||
if ( hash4[0] & 8 )
|
||||
groestl512_full( &ctx.groestl, (char*)hash4, (char*)hash4, 512 );
|
||||
if ( hash5[0] & 8 )
|
||||
groestl512_full( &ctx.groestl, (char*)hash5, (char*)hash5, 512 );
|
||||
if ( hash6[0] & 8 )
|
||||
groestl512_full( &ctx.groestl, (char*)hash6, (char*)hash6, 512 );
|
||||
if ( hash7[0] & 8 )
|
||||
groestl512_full( &ctx.groestl, (char*)hash7, (char*)hash7, 512 );
|
||||
|
||||
intrlv_8x64_512( vhashC, hash0, hash1, hash2, hash3,
|
||||
hash4, hash5, hash6, hash7 );
|
||||
|
||||
#endif
|
||||
|
||||
if ( vh_mask & 0xff )
|
||||
skein512_8way_full( &ctx.skein, vhashB, vhash, 64 );
|
||||
|
||||
mm512_blend_hash_8x64( vh, vhC, vhB, vh_mask );
|
||||
|
||||
#if defined(__VAES__)
|
||||
|
||||
rintrlv_8x64_4x128( vhashA, vhashB, vhash, 512 );
|
||||
|
||||
groestl512_4way_full( &ctx.groestl, vhashA, vhashA, 64 );
|
||||
groestl512_4way_full( &ctx.groestl, vhashB, vhashB, 64 );
|
||||
|
||||
rintrlv_4x128_8x64( vhash, vhashA, vhashB, 512 );
|
||||
|
||||
#else
|
||||
|
||||
dintrlv_8x64_512( hash0, hash1, hash2, hash3,
|
||||
hash4, hash5, hash6, hash7, vhash );
|
||||
|
||||
groestl512_full( &ctx.groestl, (char*)hash0, (char*)hash0, 512 );
|
||||
groestl512_full( &ctx.groestl, (char*)hash1, (char*)hash1, 512 );
|
||||
groestl512_full( &ctx.groestl, (char*)hash2, (char*)hash2, 512 );
|
||||
groestl512_full( &ctx.groestl, (char*)hash3, (char*)hash3, 512 );
|
||||
groestl512_full( &ctx.groestl, (char*)hash4, (char*)hash4, 512 );
|
||||
groestl512_full( &ctx.groestl, (char*)hash5, (char*)hash5, 512 );
|
||||
groestl512_full( &ctx.groestl, (char*)hash6, (char*)hash6, 512 );
|
||||
groestl512_full( &ctx.groestl, (char*)hash7, (char*)hash7, 512 );
|
||||
|
||||
intrlv_8x64_512( vhash, hash0, hash1, hash2, hash3,
|
||||
hash4, hash5, hash6, hash7 );
|
||||
|
||||
#endif
|
||||
|
||||
jh512_8way_init( &ctx.jh );
|
||||
jh512_8way_update( &ctx.jh, vhash, 64 );
|
||||
jh512_8way_close( &ctx.jh, vhash );
|
||||
|
||||
vh_mask = _mm512_cmpeq_epi64_mask( _mm512_and_si512( vh[0], bit3_mask ),
|
||||
zero );
|
||||
|
||||
if ( ( vh_mask & 0xff ) != 0xff )
|
||||
blake512_8way_full( &ctx.blake, vhashA, vhash, 64 );
|
||||
if ( vh_mask & 0xff )
|
||||
bmw512_8way_full( &ctx.bmw, vhashB, vhash, 64 );
|
||||
|
||||
mm512_blend_hash_8x64( vh, vhA, vhB, vh_mask );
|
||||
|
||||
keccak512_8way_init( &ctx.keccak );
|
||||
keccak512_8way_update( &ctx.keccak, vhash, 64 );
|
||||
keccak512_8way_close( &ctx.keccak, vhash );
|
||||
|
||||
skein512_8way_full( &ctx.skein, vhash, vhash, 64 );
|
||||
|
||||
vh_mask = _mm512_cmpeq_epi64_mask( _mm512_and_si512( vh[0], bit3_mask ),
|
||||
zero );
|
||||
|
||||
if ( ( vh_mask & 0xff ) != 0xff )
|
||||
{
|
||||
keccak512_8way_init( &ctx.keccak );
|
||||
keccak512_8way_update( &ctx.keccak, vhash, 64 );
|
||||
keccak512_8way_close( &ctx.keccak, vhashA );
|
||||
}
|
||||
if ( vh_mask & 0xff )
|
||||
{
|
||||
jh512_8way_init( &ctx.jh );
|
||||
jh512_8way_update( &ctx.jh, vhash, 64 );
|
||||
jh512_8way_close( &ctx.jh, vhashB );
|
||||
}
|
||||
|
||||
casti_m512i( state,0 ) = _mm512_mask_blend_epi64( vh_mask, vhA[0], vhB[0] );
|
||||
casti_m512i( state,1 ) = _mm512_mask_blend_epi64( vh_mask, vhA[1], vhB[1] );
|
||||
casti_m512i( state,2 ) = _mm512_mask_blend_epi64( vh_mask, vhA[2], vhB[2] );
|
||||
casti_m512i( state,3 ) = _mm512_mask_blend_epi64( vh_mask, vhA[3], vhB[3] );
|
||||
}
|
||||
|
||||
int scanhash_anime_8way( struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr )
|
||||
{
|
||||
uint64_t hash64[4*8] __attribute__ ((aligned (64)));
|
||||
uint32_t vdata[20*8] __attribute__ ((aligned (64)));
|
||||
uint32_t lane_hash[8] __attribute__ ((aligned (64)));
|
||||
uint64_t *hash64_q3 = &(hash64[3*8]);
|
||||
uint32_t *pdata = work->data;
|
||||
uint32_t *ptarget = work->target;
|
||||
const uint64_t targ64_q3 = ((uint64_t*)ptarget)[3];
|
||||
uint32_t n = pdata[19];
|
||||
const uint32_t first_nonce = pdata[19];
|
||||
const uint32_t last_nonce = max_nonce - 8;
|
||||
__m512i *noncev = (__m512i*)vdata + 9;
|
||||
const int thr_id = mythr->id;
|
||||
const bool bench = opt_benchmark;
|
||||
|
||||
mm512_bswap32_intrlv80_8x64( vdata, pdata );
|
||||
*noncev = mm512_intrlv_blend_32(
|
||||
_mm512_set_epi32( n+7, 0, n+6, 0, n+5, 0, n+4, 0,
|
||||
n+3, 0, n+2, 0, n+1, 0, n , 0 ), *noncev );
|
||||
|
||||
do
|
||||
{
|
||||
anime_8way_hash( hash64, vdata );
|
||||
|
||||
for ( int lane = 0; lane < 8; lane++ )
|
||||
if ( unlikely( hash64_q3[ lane ] <= targ64_q3 && !bench ) )
|
||||
{
|
||||
extr_lane_8x64( lane_hash, hash64, lane, 256 );
|
||||
if ( valid_hash( lane_hash, ptarget ) )
|
||||
{
|
||||
pdata[19] = bswap_32( n + lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
*noncev = _mm512_add_epi32( *noncev,
|
||||
m512_const1_64( 0x0000000800000000 ) );
|
||||
n += 8;
|
||||
} while ( likely( ( n < last_nonce ) && !work_restart[thr_id].restart ) );
|
||||
pdata[19] = n;
|
||||
*hashes_done = n - first_nonce;
|
||||
return 0;
|
||||
}
|
||||
|
||||
#elif defined (ANIME_4WAY)
|
||||
|
||||
typedef struct {
|
||||
blake512_4way_context blake;
|
||||
@@ -23,18 +246,6 @@ typedef struct {
|
||||
keccak512_4way_context keccak;
|
||||
} anime_4way_ctx_holder;
|
||||
|
||||
anime_4way_ctx_holder anime_4way_ctx __attribute__ ((aligned (64)));
|
||||
|
||||
void init_anime_4way_ctx()
|
||||
{
|
||||
blake512_4way_init( &anime_4way_ctx.blake );
|
||||
bmw512_4way_init( &anime_4way_ctx.bmw );
|
||||
init_groestl( &anime_4way_ctx.groestl, 64 );
|
||||
skein512_4way_init( &anime_4way_ctx.skein );
|
||||
jh512_4way_init( &anime_4way_ctx.jh );
|
||||
keccak512_4way_init( &anime_4way_ctx.keccak );
|
||||
}
|
||||
|
||||
void anime_4way_hash( void *state, const void *input )
|
||||
{
|
||||
uint64_t hash0[8] __attribute__ ((aligned (64)));
|
||||
@@ -48,81 +259,61 @@ void anime_4way_hash( void *state, const void *input )
|
||||
__m256i* vhA = (__m256i*)vhashA;
|
||||
__m256i* vhB = (__m256i*)vhashB;
|
||||
__m256i vh_mask;
|
||||
const uint32_t mask = 8;
|
||||
int h_mask;
|
||||
const __m256i bit3_mask = m256_const1_64( 8 );
|
||||
const __m256i zero = _mm256_setzero_si256();
|
||||
anime_4way_ctx_holder ctx;
|
||||
memcpy( &ctx, &anime_4way_ctx, sizeof(anime_4way_ctx) );
|
||||
|
||||
bmw512_4way_init( &ctx.bmw );
|
||||
bmw512_4way_update( &ctx.bmw, input, 80 );
|
||||
bmw512_4way_close( &ctx.bmw, vhash );
|
||||
|
||||
blake512_4way_update( &ctx.blake, vhash, 64 );
|
||||
blake512_4way_close( &ctx.blake, vhash );
|
||||
blake512_4way_full( &ctx.blake, vhash, vhash, 64 );
|
||||
|
||||
vh_mask = _mm256_cmpeq_epi64( _mm256_and_si256( vh[0], bit3_mask ), zero );
|
||||
h_mask = _mm256_movemask_epi8( vh_mask );
|
||||
|
||||
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
|
||||
|
||||
if ( hash0[0] & mask )
|
||||
{
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash0,
|
||||
(char*)hash0, 512 );
|
||||
}
|
||||
if ( hash1[0] & mask )
|
||||
{
|
||||
reinit_groestl( &ctx.groestl );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash1,
|
||||
(char*)hash1, 512 );
|
||||
}
|
||||
if ( hash2[0] & mask )
|
||||
{
|
||||
reinit_groestl( &ctx.groestl );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash2,
|
||||
(char*)hash2, 512 );
|
||||
}
|
||||
if ( hash3[0] & mask )
|
||||
{
|
||||
reinit_groestl( &ctx.groestl );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash3,
|
||||
(char*)hash3, 512 );
|
||||
}
|
||||
// A
|
||||
if ( hash0[0] & 8 )
|
||||
groestl512_full( &ctx.groestl, (char*)hash0, (char*)hash0, 512 );
|
||||
if ( hash1[0] & 8 )
|
||||
groestl512_full( &ctx.groestl, (char*)hash1, (char*)hash1, 512 );
|
||||
if ( hash2[0] & 8)
|
||||
groestl512_full( &ctx.groestl, (char*)hash2, (char*)hash2, 512 );
|
||||
if ( hash3[0] & 8 )
|
||||
groestl512_full( &ctx.groestl, (char*)hash3, (char*)hash3, 512 );
|
||||
|
||||
intrlv_4x64( vhashA, hash0, hash1, hash2, hash3, 512 );
|
||||
|
||||
if ( mm256_anybits0( vh_mask ) )
|
||||
{
|
||||
skein512_4way_update( &ctx.skein, vhash, 64 );
|
||||
skein512_4way_close( &ctx.skein, vhashB );
|
||||
}
|
||||
// B
|
||||
if ( h_mask & 0xffffffff )
|
||||
skein512_4way_full( &ctx.skein, vhashB, vhash, 64 );
|
||||
|
||||
mm256_blend_hash_4x64( vh, vhA, vhB, vh_mask );
|
||||
|
||||
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
|
||||
|
||||
reinit_groestl( &ctx.groestl );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash0, (char*)hash0, 512 );
|
||||
reinit_groestl( &ctx.groestl );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash1, (char*)hash1, 512 );
|
||||
reinit_groestl( &ctx.groestl );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash2, (char*)hash2, 512 );
|
||||
reinit_groestl( &ctx.groestl );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash3, (char*)hash3, 512 );
|
||||
groestl512_full( &ctx.groestl, (char*)hash0, (char*)hash0, 512 );
|
||||
groestl512_full( &ctx.groestl, (char*)hash1, (char*)hash1, 512 );
|
||||
groestl512_full( &ctx.groestl, (char*)hash2, (char*)hash2, 512 );
|
||||
groestl512_full( &ctx.groestl, (char*)hash3, (char*)hash3, 512 );
|
||||
|
||||
intrlv_4x64( vhash, hash0, hash1, hash2, hash3, 512 );
|
||||
|
||||
jh512_4way_init( &ctx.jh );
|
||||
jh512_4way_update( &ctx.jh, vhash, 64 );
|
||||
jh512_4way_close( &ctx.jh, vhash );
|
||||
|
||||
vh_mask = _mm256_cmpeq_epi64( _mm256_and_si256( vh[0], bit3_mask ), zero );
|
||||
h_mask = _mm256_movemask_epi8( vh_mask );
|
||||
|
||||
if ( mm256_anybits1( vh_mask ) )
|
||||
{
|
||||
blake512_4way_init( &ctx.blake );
|
||||
blake512_4way_update( &ctx.blake, vhash, 64 );
|
||||
blake512_4way_close( &ctx.blake, vhashA );
|
||||
}
|
||||
if ( mm256_anybits0( vh_mask ) )
|
||||
// A
|
||||
if ( ( h_mask & 0xffffffff ) != 0xffffffff )
|
||||
blake512_4way_full( &ctx.blake, vhashA, vhash, 64 );
|
||||
// B
|
||||
if ( h_mask & 0xffffffff )
|
||||
{
|
||||
bmw512_4way_init( &ctx.bmw );
|
||||
bmw512_4way_update( &ctx.bmw, vhash, 64 );
|
||||
@@ -131,64 +322,74 @@ void anime_4way_hash( void *state, const void *input )
|
||||
|
||||
mm256_blend_hash_4x64( vh, vhA, vhB, vh_mask );
|
||||
|
||||
keccak512_4way_init( &ctx.keccak );
|
||||
keccak512_4way_update( &ctx.keccak, vhash, 64 );
|
||||
keccak512_4way_close( &ctx.keccak, vhash );
|
||||
|
||||
skein512_4way_init( &ctx.skein );
|
||||
skein512_4way_update( &ctx.skein, vhash, 64 );
|
||||
skein512_4way_close( &ctx.skein, vhash );
|
||||
skein512_4way_full( &ctx.skein, vhash, vhash, 64 );
|
||||
|
||||
vh_mask = _mm256_cmpeq_epi64( _mm256_and_si256( vh[0], bit3_mask ), zero );
|
||||
h_mask = _mm256_movemask_epi8( vh_mask );
|
||||
|
||||
if ( mm256_anybits1( vh_mask ) )
|
||||
// A
|
||||
if ( ( h_mask & 0xffffffff ) != 0xffffffff )
|
||||
{
|
||||
keccak512_4way_init( &ctx.keccak );
|
||||
keccak512_4way_update( &ctx.keccak, vhash, 64 );
|
||||
keccak512_4way_close( &ctx.keccak, vhashA );
|
||||
}
|
||||
if ( mm256_anybits0( vh_mask ) )
|
||||
// B
|
||||
if ( h_mask & 0xffffffff )
|
||||
{
|
||||
jh512_4way_init( &ctx.jh );
|
||||
jh512_4way_update( &ctx.jh, vhash, 64 );
|
||||
jh512_4way_close( &ctx.jh, vhashB );
|
||||
}
|
||||
|
||||
mm256_blend_hash_4x64( vh, vhA, vhB, vh_mask );
|
||||
|
||||
dintrlv_4x64( state, state+32, state+64, state+96, vhash, 256 );
|
||||
casti_m256i( state, 0 ) = _mm256_blendv_epi8( vhA[0], vhB[0], vh_mask );
|
||||
casti_m256i( state, 1 ) = _mm256_blendv_epi8( vhA[1], vhB[1], vh_mask );
|
||||
casti_m256i( state, 2 ) = _mm256_blendv_epi8( vhA[2], vhB[2], vh_mask );
|
||||
casti_m256i( state, 3 ) = _mm256_blendv_epi8( vhA[3], vhB[3], vh_mask );
|
||||
}
|
||||
|
||||
int scanhash_anime_4way( struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr )
|
||||
{
|
||||
uint32_t hash[4*8] __attribute__ ((aligned (64)));
|
||||
uint64_t hash64[4*4] __attribute__ ((aligned (64)));
|
||||
uint32_t vdata[20*4] __attribute__ ((aligned (64)));
|
||||
uint32_t lane_hash[8] __attribute__ ((aligned (64)));
|
||||
uint64_t *hash64_q3 = &(hash64[3*4]);
|
||||
uint32_t *pdata = work->data;
|
||||
uint32_t *ptarget = work->target;
|
||||
const uint64_t targ64_q3 = ((uint64_t*)ptarget)[3];
|
||||
uint32_t n = pdata[19];
|
||||
const uint32_t first_nonce = pdata[19];
|
||||
const uint32_t last_nonce = max_nonce - 4;
|
||||
__m256i *noncev = (__m256i*)vdata + 9; // aligned
|
||||
__m256i *noncev = (__m256i*)vdata + 9;
|
||||
const int thr_id = mythr->id;
|
||||
const bool bench = opt_benchmark;
|
||||
|
||||
mm256_bswap32_intrlv80_4x64( vdata, pdata );
|
||||
*noncev = mm256_intrlv_blend_32(
|
||||
_mm256_set_epi32( n+3, 0, n+2, 0, n+1, 0, n, 0 ), *noncev );
|
||||
|
||||
do
|
||||
{
|
||||
anime_4way_hash( hash, vdata );
|
||||
anime_4way_hash( hash64, vdata );
|
||||
|
||||
for ( int i = 0; i < 4; i++ )
|
||||
if ( valid_hash( hash+(i<<3), ptarget ) && !opt_benchmark )
|
||||
for ( int lane = 0; lane < 4; lane++ )
|
||||
if ( unlikely( hash64_q3[ lane ] <= targ64_q3 && !bench ) )
|
||||
{
|
||||
pdata[19] = bswap_32( n+i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
extr_lane_4x64( lane_hash, hash64, lane, 256 );
|
||||
if ( valid_hash( lane_hash, ptarget ) )
|
||||
{
|
||||
pdata[19] = bswap_32( n + lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
*noncev = _mm256_add_epi32( *noncev,
|
||||
m256_const1_64( 0x0000000400000000 ) );
|
||||
n += 4;
|
||||
} while ( ( n < last_nonce ) && !work_restart[thr_id].restart );
|
||||
} while ( likely( ( n < last_nonce ) && !work_restart[thr_id].restart ) );
|
||||
pdata[19] = n;
|
||||
*hashes_done = n - first_nonce;
|
||||
return 0;
|
||||
|
||||
@@ -2,8 +2,10 @@
|
||||
|
||||
bool register_anime_algo( algo_gate_t* gate )
|
||||
{
|
||||
#if defined (ANIME_4WAY)
|
||||
init_anime_4way_ctx();
|
||||
#if defined (ANIME_8WAY)
|
||||
gate->scanhash = (void*)&scanhash_anime_8way;
|
||||
gate->hash = (void*)&anime_8way_hash;
|
||||
#elif defined (ANIME_4WAY)
|
||||
gate->scanhash = (void*)&scanhash_anime_4way;
|
||||
gate->hash = (void*)&anime_4way_hash;
|
||||
#else
|
||||
@@ -11,7 +13,7 @@ bool register_anime_algo( algo_gate_t* gate )
|
||||
gate->scanhash = (void*)&scanhash_anime;
|
||||
gate->hash = (void*)&anime_hash;
|
||||
#endif
|
||||
gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT;
|
||||
gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT | AVX512_OPT | VAES_OPT;
|
||||
return true;
|
||||
};
|
||||
|
||||
|
||||
@@ -4,18 +4,25 @@
|
||||
#include "algo-gate-api.h"
|
||||
#include <stdint.h>
|
||||
|
||||
#if defined(__AVX2__) && defined(__AES__)
|
||||
#define ANIME_4WAY
|
||||
#if defined(__AVX512F__) && defined(__AVX512VL__) && defined(__AVX512DQ__) && defined(__AVX512BW__)
|
||||
#define ANIME_8WAY 1
|
||||
#elif defined(__AVX2__) && defined(__AES__)
|
||||
#define ANIME_4WAY 1
|
||||
#endif
|
||||
|
||||
bool register_anime_algo( algo_gate_t* gate );
|
||||
|
||||
#if defined(ANIME_4WAY)
|
||||
#if defined(ANIME_8WAY)
|
||||
|
||||
void anime_8way_hash( void *state, const void *input );
|
||||
int scanhash_anime_8way( struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr );
|
||||
|
||||
#elif defined(ANIME_4WAY)
|
||||
|
||||
void anime_4way_hash( void *state, const void *input );
|
||||
int scanhash_anime_4way( struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr );
|
||||
void init_anime_4way_ctx();
|
||||
|
||||
#endif
|
||||
|
||||
|
||||
File diff suppressed because it is too large
Load Diff
@@ -72,12 +72,10 @@ void quark_8way_hash( void *state, const void *input )
|
||||
|
||||
memcpy( &ctx, &quark_8way_ctx, sizeof(quark_8way_ctx) );
|
||||
|
||||
blake512_8way_update( &ctx.blake, input, 80 );
|
||||
blake512_8way_close( &ctx.blake, vhash );
|
||||
|
||||
bmw512_8way_update( &ctx.bmw, vhash, 64 );
|
||||
bmw512_8way_close( &ctx.bmw, vhash );
|
||||
blake512_8way_full( &ctx.blake, vhash, input, 80 );
|
||||
|
||||
bmw512_8way_full( &ctx.bmw, vhash, vhash, 64 );
|
||||
|
||||
vh_mask = _mm512_cmpeq_epi64_mask( _mm512_and_si512( vh[0], bit3_mask ),
|
||||
zero );
|
||||
|
||||
@@ -86,70 +84,34 @@ void quark_8way_hash( void *state, const void *input )
|
||||
|
||||
rintrlv_8x64_4x128( vhashA, vhashB, vhash, 512 );
|
||||
|
||||
if ( ( vh_mask & 0x0f ) != 0x0f )
|
||||
{
|
||||
groestl512_4way_init( &ctx.groestl, 64 );
|
||||
groestl512_4way_update_close( &ctx.groestl, vhashA, vhashA, 512 );
|
||||
}
|
||||
if ( ( vh_mask & 0xf0 ) != 0xf0 )
|
||||
{
|
||||
groestl512_4way_init( &ctx.groestl, 64 );
|
||||
groestl512_4way_update_close( &ctx.groestl, vhashB, vhashB, 512 );
|
||||
}
|
||||
rintrlv_4x128_8x64( vhashC, vhashA, vhashB, 512 );
|
||||
if ( ( vh_mask & 0x0f ) != 0x0f )
|
||||
groestl512_4way_full( &ctx.groestl, vhashA, vhashA, 64 );
|
||||
if ( ( vh_mask & 0xf0 ) != 0xf0 )
|
||||
groestl512_4way_full( &ctx.groestl, vhashB, vhashB, 64 );
|
||||
|
||||
rintrlv_4x128_8x64( vhashC, vhashA, vhashB, 512 );
|
||||
|
||||
#else
|
||||
|
||||
dintrlv_8x64( hash0, hash1, hash2, hash3, hash4, hash5, hash6, hash7,
|
||||
vhash, 512 );
|
||||
|
||||
if ( hash0[0] & mask )
|
||||
{
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash0,
|
||||
(char*)hash0, 512 );
|
||||
}
|
||||
if ( hash1[0] & mask )
|
||||
{
|
||||
reinit_groestl( &ctx.groestl );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash1,
|
||||
(char*)hash1, 512 );
|
||||
}
|
||||
if ( hash2[0] & mask )
|
||||
{
|
||||
reinit_groestl( &ctx.groestl );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash2,
|
||||
(char*)hash2, 512 );
|
||||
}
|
||||
if ( hash3[0] & mask )
|
||||
{
|
||||
reinit_groestl( &ctx.groestl );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash3,
|
||||
(char*)hash3, 512 );
|
||||
}
|
||||
if ( hash4[0] & mask )
|
||||
{
|
||||
reinit_groestl( &ctx.groestl );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash4,
|
||||
(char*)hash4, 512 );
|
||||
}
|
||||
if ( hash5[0] & mask )
|
||||
{
|
||||
reinit_groestl( &ctx.groestl );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash5,
|
||||
(char*)hash5, 512 );
|
||||
}
|
||||
if ( hash6[0] & mask )
|
||||
{
|
||||
reinit_groestl( &ctx.groestl );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash6,
|
||||
(char*)hash6, 512 );
|
||||
}
|
||||
if ( hash7[0] & mask )
|
||||
{
|
||||
reinit_groestl( &ctx.groestl );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash7,
|
||||
(char*)hash7, 512 );
|
||||
}
|
||||
if ( hash0[0] & 8 )
|
||||
groestl512_full( &ctx.groestl, (char*)hash0, (char*)hash0, 512 );
|
||||
if ( hash1[0] & 8 )
|
||||
groestl512_full( &ctx.groestl, (char*)hash1, (char*)hash1, 512 );
|
||||
if ( hash2[0] & 8)
|
||||
groestl512_full( &ctx.groestl, (char*)hash2, (char*)hash2, 512 );
|
||||
if ( hash3[0] & 8 )
|
||||
groestl512_full( &ctx.groestl, (char*)hash3, (char*)hash3, 512 );
|
||||
if ( hash4[0] & 8 )
|
||||
groestl512_full( &ctx.groestl, (char*)hash4, (char*)hash4, 512 );
|
||||
if ( hash5[0] & 8 )
|
||||
groestl512_full( &ctx.groestl, (char*)hash5, (char*)hash5, 512 );
|
||||
if ( hash6[0] & 8 )
|
||||
groestl512_full( &ctx.groestl, (char*)hash6, (char*)hash6, 512 );
|
||||
if ( hash7[0] & 8 )
|
||||
groestl512_full( &ctx.groestl, (char*)hash7, (char*)hash7, 512 );
|
||||
|
||||
intrlv_8x64( vhashC, hash0, hash1, hash2, hash3, hash4, hash5, hash6,
|
||||
hash7, 512 );
|
||||
@@ -157,10 +119,7 @@ void quark_8way_hash( void *state, const void *input )
|
||||
#endif
|
||||
|
||||
if ( vh_mask & 0xff )
|
||||
{
|
||||
skein512_8way_update( &ctx.skein, vhash, 64 );
|
||||
skein512_8way_close( &ctx.skein, vhashB );
|
||||
}
|
||||
skein512_8way_full( &ctx.skein, vhashB, vhash, 64 );
|
||||
|
||||
mm512_blend_hash_8x64( vh, vhC, vhB, vh_mask );
|
||||
|
||||
@@ -168,10 +127,10 @@ void quark_8way_hash( void *state, const void *input )
|
||||
|
||||
rintrlv_8x64_4x128( vhashA, vhashB, vhash, 512 );
|
||||
|
||||
groestl512_4way_init( &ctx.groestl, 64 );
|
||||
groestl512_4way_update_close( &ctx.groestl, vhashA, vhashA, 512 );
|
||||
groestl512_4way_init( &ctx.groestl, 64 );
|
||||
groestl512_4way_update_close( &ctx.groestl, vhashB, vhashB, 512 );
|
||||
if ( ( vh_mask & 0x0f ) != 0x0f )
|
||||
groestl512_4way_full( &ctx.groestl, vhashA, vhashA, 64 );
|
||||
if ( ( vh_mask & 0xf0 ) != 0xf0 )
|
||||
groestl512_4way_full( &ctx.groestl, vhashB, vhashB, 64 );
|
||||
|
||||
rintrlv_4x128_8x64( vhash, vhashA, vhashB, 512 );
|
||||
|
||||
@@ -180,22 +139,22 @@ void quark_8way_hash( void *state, const void *input )
|
||||
dintrlv_8x64( hash0, hash1, hash2, hash3, hash4, hash5, hash6, hash7,
|
||||
vhash, 512 );
|
||||
|
||||
reinit_groestl( &ctx.groestl );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash0, (char*)hash0, 512 );
|
||||
reinit_groestl( &ctx.groestl );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash1, (char*)hash1, 512 );
|
||||
reinit_groestl( &ctx.groestl );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash2, (char*)hash2, 512 );
|
||||
reinit_groestl( &ctx.groestl );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash3, (char*)hash3, 512 );
|
||||
reinit_groestl( &ctx.groestl );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash4, (char*)hash4, 512 );
|
||||
reinit_groestl( &ctx.groestl );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash5, (char*)hash5, 512 );
|
||||
reinit_groestl( &ctx.groestl );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash6, (char*)hash6, 512 );
|
||||
reinit_groestl( &ctx.groestl );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash7, (char*)hash7, 512 );
|
||||
if ( hash0[0] & 8 )
|
||||
groestl512_full( &ctx.groestl, (char*)hash0, (char*)hash0, 512 );
|
||||
if ( hash1[0] & 8 )
|
||||
groestl512_full( &ctx.groestl, (char*)hash1, (char*)hash1, 512 );
|
||||
if ( hash2[0] & 8)
|
||||
groestl512_full( &ctx.groestl, (char*)hash2, (char*)hash2, 512 );
|
||||
if ( hash3[0] & 8 )
|
||||
groestl512_full( &ctx.groestl, (char*)hash3, (char*)hash3, 512 );
|
||||
if ( hash4[0] & 8 )
|
||||
groestl512_full( &ctx.groestl, (char*)hash4, (char*)hash4, 512 );
|
||||
if ( hash5[0] & 8 )
|
||||
groestl512_full( &ctx.groestl, (char*)hash5, (char*)hash5, 512 );
|
||||
if ( hash6[0] & 8 )
|
||||
groestl512_full( &ctx.groestl, (char*)hash6, (char*)hash6, 512 );
|
||||
if ( hash7[0] & 8 )
|
||||
groestl512_full( &ctx.groestl, (char*)hash7, (char*)hash7, 512 );
|
||||
|
||||
intrlv_8x64( vhash, hash0, hash1, hash2, hash3, hash4, hash5, hash6, hash7,
|
||||
512 );
|
||||
@@ -209,27 +168,16 @@ void quark_8way_hash( void *state, const void *input )
|
||||
zero );
|
||||
|
||||
if ( ( vh_mask & 0xff ) != 0xff )
|
||||
{
|
||||
blake512_8way_init( &ctx.blake );
|
||||
blake512_8way_update( &ctx.blake, vhash, 64 );
|
||||
blake512_8way_close( &ctx.blake, vhashA );
|
||||
}
|
||||
|
||||
blake512_8way_full( &ctx.blake, vhashA, vhash, 64 );
|
||||
if ( vh_mask & 0xff )
|
||||
{
|
||||
bmw512_8way_init( &ctx.bmw );
|
||||
bmw512_8way_update( &ctx.bmw, vhash, 64 );
|
||||
bmw512_8way_close( &ctx.bmw, vhashB );
|
||||
}
|
||||
bmw512_8way_full( &ctx.bmw, vhashB, vhash, 64 );
|
||||
|
||||
mm512_blend_hash_8x64( vh, vhA, vhB, vh_mask );
|
||||
|
||||
keccak512_8way_update( &ctx.keccak, vhash, 64 );
|
||||
keccak512_8way_close( &ctx.keccak, vhash );
|
||||
|
||||
skein512_8way_init( &ctx.skein );
|
||||
skein512_8way_update( &ctx.skein, vhash, 64 );
|
||||
skein512_8way_close( &ctx.skein, vhash );
|
||||
skein512_8way_full( &ctx.skein, vhash, vhash, 64 );
|
||||
|
||||
vh_mask = _mm512_cmpeq_epi64_mask( _mm512_and_si512( vh[0], bit3_mask ),
|
||||
zero );
|
||||
@@ -258,41 +206,44 @@ void quark_8way_hash( void *state, const void *input )
|
||||
int scanhash_quark_8way( struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr )
|
||||
{
|
||||
uint32_t hash[8*8] __attribute__ ((aligned (128)));
|
||||
uint32_t vdata[24*8] __attribute__ ((aligned (64)));
|
||||
uint64_t hash64[4*8] __attribute__ ((aligned (128)));
|
||||
uint32_t vdata[20*8] __attribute__ ((aligned (64)));
|
||||
uint32_t lane_hash[8] __attribute__ ((aligned (64)));
|
||||
uint32_t *hash7 = &(hash[49]);
|
||||
uint32_t *pdata = work->data;
|
||||
uint64_t *hash64_q3 = &(hash64[3*8]);
|
||||
uint32_t *ptarget = work->target;
|
||||
const uint64_t targ64_q3 = ((uint64_t*)ptarget)[3];
|
||||
uint32_t *pdata = work->data;
|
||||
uint32_t n = pdata[19];
|
||||
const uint32_t first_nonce = pdata[19];
|
||||
__m512i *noncev = (__m512i*)vdata + 9; // aligned
|
||||
int thr_id = mythr->id;
|
||||
const uint32_t Htarg = ptarget[7];
|
||||
const uint32_t last_nonce = max_nonce - 8;
|
||||
__m512i *noncev = (__m512i*)vdata + 9;
|
||||
const int thr_id = mythr->id;
|
||||
const bool bench = opt_benchmark;
|
||||
|
||||
mm512_bswap32_intrlv80_8x64( vdata, pdata );
|
||||
*noncev = mm512_intrlv_blend_32(
|
||||
_mm512_set_epi32( n+7, 0, n+6, 0, n+5, 0, n+4, 0,
|
||||
n+3, 0, n+2, 0, n+1, 0, n , 0 ), *noncev );
|
||||
do
|
||||
{
|
||||
*noncev = mm512_intrlv_blend_32( mm512_bswap_32(
|
||||
_mm512_set_epi32( n+7, 0, n+6, 0, n+5, 0, n+4, 0,
|
||||
n+3, 0, n+2, 0, n+1, 0, n , 0 ) ), *noncev );
|
||||
quark_8way_hash( hash64, vdata );
|
||||
|
||||
quark_8way_hash( hash, vdata );
|
||||
pdata[19] = n;
|
||||
|
||||
for ( int i = 0; i < 8; i++ )
|
||||
if ( unlikely( hash7[ i<<1 ] <= Htarg ) )
|
||||
for ( int lane = 0; lane < 8; lane++ )
|
||||
if ( unlikely( hash64_q3[ lane ] <= targ64_q3 && !bench ) )
|
||||
{
|
||||
extr_lane_8x64( lane_hash, hash, i, 256 );
|
||||
if ( likely( fulltest( lane_hash, ptarget ) && !opt_benchmark ) )
|
||||
extr_lane_8x64( lane_hash, hash64, lane, 256 );
|
||||
if ( valid_hash( lane_hash, ptarget ) )
|
||||
{
|
||||
pdata[19] = n+i;
|
||||
submit_lane_solution( work, lane_hash, mythr, i );
|
||||
pdata[19] = bswap_32( n + lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
*noncev = _mm512_add_epi32( *noncev,
|
||||
m512_const1_64( 0x0000000800000000 ) );
|
||||
n += 8;
|
||||
} while ( ( n < max_nonce-8 ) && !work_restart[thr_id].restart );
|
||||
} while ( likely( ( n < last_nonce ) && !work_restart[thr_id].restart ) );
|
||||
|
||||
pdata[19] = n;
|
||||
*hashes_done = n - first_nonce;
|
||||
return 0;
|
||||
}
|
||||
@@ -333,67 +284,47 @@ void quark_4way_hash( void *state, const void *input )
|
||||
__m256i* vhA = (__m256i*)vhashA;
|
||||
__m256i* vhB = (__m256i*)vhashB;
|
||||
__m256i vh_mask;
|
||||
int h_mask;
|
||||
quark_4way_ctx_holder ctx;
|
||||
const __m256i bit3_mask = m256_const1_64( 8 );
|
||||
const uint32_t mask = 8;
|
||||
const __m256i zero = _mm256_setzero_si256();
|
||||
|
||||
memcpy( &ctx, &quark_4way_ctx, sizeof(quark_4way_ctx) );
|
||||
|
||||
blake512_4way_update( &ctx.blake, input, 80 );
|
||||
blake512_4way_close( &ctx.blake, vhash );
|
||||
blake512_4way_full( &ctx.blake, vhash, input, 80 );
|
||||
|
||||
bmw512_4way_update( &ctx.bmw, vhash, 64 );
|
||||
bmw512_4way_close( &ctx.bmw, vhash );
|
||||
|
||||
vh_mask = _mm256_cmpeq_epi64( _mm256_and_si256( vh[0], bit3_mask ), zero );
|
||||
h_mask = _mm256_movemask_epi8( vh_mask );
|
||||
|
||||
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
|
||||
|
||||
if ( hash0[0] & mask )
|
||||
{
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash0,
|
||||
(char*)hash0, 512 );
|
||||
}
|
||||
if ( hash1[0] & mask )
|
||||
{
|
||||
reinit_groestl( &ctx.groestl );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash1,
|
||||
(char*)hash1, 512 );
|
||||
}
|
||||
if ( hash2[0] & mask )
|
||||
{
|
||||
reinit_groestl( &ctx.groestl );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash2,
|
||||
(char*)hash2, 512 );
|
||||
}
|
||||
if ( hash3[0] & mask )
|
||||
{
|
||||
reinit_groestl( &ctx.groestl );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash3,
|
||||
(char*)hash3, 512 );
|
||||
}
|
||||
// A
|
||||
if ( hash0[0] & 8 )
|
||||
groestl512_full( &ctx.groestl, (char*)hash0, (char*)hash0, 512 );
|
||||
if ( hash1[0] & 8 )
|
||||
groestl512_full( &ctx.groestl, (char*)hash1, (char*)hash1, 512 );
|
||||
if ( hash2[0] & 8)
|
||||
groestl512_full( &ctx.groestl, (char*)hash2, (char*)hash2, 512 );
|
||||
if ( hash3[0] & 8 )
|
||||
groestl512_full( &ctx.groestl, (char*)hash3, (char*)hash3, 512 );
|
||||
|
||||
intrlv_4x64( vhashA, hash0, hash1, hash2, hash3, 512 );
|
||||
|
||||
if ( mm256_anybits1( vh_mask ) )
|
||||
{
|
||||
skein512_4way_update( &ctx.skein, vhash, 64 );
|
||||
skein512_4way_close( &ctx.skein, vhashB );
|
||||
}
|
||||
// B
|
||||
if ( likely( h_mask & 0xffffffff ) )
|
||||
skein512_4way_full( &ctx.skein, vhashB, vhash, 64 );
|
||||
|
||||
mm256_blend_hash_4x64( vh, vhA, vhB, vh_mask );
|
||||
|
||||
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
|
||||
|
||||
reinit_groestl( &ctx.groestl );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash0, (char*)hash0, 512 );
|
||||
reinit_groestl( &ctx.groestl );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash1, (char*)hash1, 512 );
|
||||
reinit_groestl( &ctx.groestl );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash2, (char*)hash2, 512 );
|
||||
reinit_groestl( &ctx.groestl );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash3, (char*)hash3, 512 );
|
||||
groestl512_full( &ctx.groestl, (char*)hash0, (char*)hash0, 512 );
|
||||
groestl512_full( &ctx.groestl, (char*)hash1, (char*)hash1, 512 );
|
||||
groestl512_full( &ctx.groestl, (char*)hash2, (char*)hash2, 512 );
|
||||
groestl512_full( &ctx.groestl, (char*)hash3, (char*)hash3, 512 );
|
||||
|
||||
intrlv_4x64( vhash, hash0, hash1, hash2, hash3, 512 );
|
||||
|
||||
@@ -401,15 +332,13 @@ void quark_4way_hash( void *state, const void *input )
|
||||
jh512_4way_close( &ctx.jh, vhash );
|
||||
|
||||
vh_mask = _mm256_cmpeq_epi64( _mm256_and_si256( vh[0], bit3_mask ), zero );
|
||||
h_mask = _mm256_movemask_epi8( vh_mask );
|
||||
|
||||
if ( mm256_anybits1( vh_mask ) )
|
||||
{
|
||||
blake512_4way_init( &ctx.blake );
|
||||
blake512_4way_update( &ctx.blake, vhash, 64 );
|
||||
blake512_4way_close( &ctx.blake, vhashA );
|
||||
}
|
||||
|
||||
if ( mm256_anybits0( vh_mask ) )
|
||||
// A
|
||||
if ( likely( ( h_mask & 0xffffffff ) != 0xffffffff ) )
|
||||
blake512_4way_full( &ctx.blake, vhashA, vhash, 64 );
|
||||
// B
|
||||
if ( likely( h_mask & 0xffffffff ) )
|
||||
{
|
||||
bmw512_4way_init( &ctx.bmw );
|
||||
bmw512_4way_update( &ctx.bmw, vhash, 64 );
|
||||
@@ -421,20 +350,20 @@ void quark_4way_hash( void *state, const void *input )
|
||||
keccak512_4way_update( &ctx.keccak, vhash, 64 );
|
||||
keccak512_4way_close( &ctx.keccak, vhash );
|
||||
|
||||
skein512_4way_init( &ctx.skein );
|
||||
skein512_4way_update( &ctx.skein, vhash, 64 );
|
||||
skein512_4way_close( &ctx.skein, vhash );
|
||||
skein512_4way_full( &ctx.skein, vhash, vhash, 64 );
|
||||
|
||||
vh_mask = _mm256_cmpeq_epi64( _mm256_and_si256( vh[0], bit3_mask ), zero );
|
||||
h_mask = _mm256_movemask_epi8( vh_mask );
|
||||
|
||||
if ( mm256_anybits1( vh_mask ) )
|
||||
// A
|
||||
if ( likely( ( h_mask & 0xffffffff ) != 0xffffffff ) )
|
||||
{
|
||||
keccak512_4way_init( &ctx.keccak );
|
||||
keccak512_4way_update( &ctx.keccak, vhash, 64 );
|
||||
keccak512_4way_close( &ctx.keccak, vhashA );
|
||||
}
|
||||
|
||||
if ( mm256_anybits0( vh_mask ) )
|
||||
// B
|
||||
if ( likely( h_mask & 0xffffffff ) )
|
||||
{
|
||||
jh512_4way_init( &ctx.jh );
|
||||
jh512_4way_update( &ctx.jh, vhash, 64 );
|
||||
@@ -451,41 +380,44 @@ void quark_4way_hash( void *state, const void *input )
|
||||
int scanhash_quark_4way( struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr )
|
||||
{
|
||||
uint32_t hash[4*8] __attribute__ ((aligned (64)));
|
||||
uint32_t vdata[24*4] __attribute__ ((aligned (64)));
|
||||
uint64_t hash64[4*4] __attribute__ ((aligned (64)));
|
||||
uint32_t vdata[20*4] __attribute__ ((aligned (64)));
|
||||
uint32_t lane_hash[8] __attribute__ ((aligned (64)));
|
||||
uint32_t *hash7 = &(hash[25]);
|
||||
uint64_t *hash64_q3 = &(hash64[3*4]);
|
||||
uint32_t *pdata = work->data;
|
||||
uint32_t *ptarget = work->target;
|
||||
const uint64_t targ64_q3 = ((uint64_t*)ptarget)[3];
|
||||
uint32_t n = pdata[19];
|
||||
const uint32_t first_nonce = pdata[19];
|
||||
__m256i *noncev = (__m256i*)vdata + 9; // aligned
|
||||
int thr_id = mythr->id;
|
||||
const uint32_t Htarg = ptarget[7];
|
||||
const uint32_t last_nonce = max_nonce - 4;
|
||||
__m256i *noncev = (__m256i*)vdata + 9;
|
||||
const int thr_id = mythr->id;
|
||||
const bool bench = opt_benchmark;
|
||||
|
||||
mm256_bswap32_intrlv80_4x64( vdata, pdata );
|
||||
*noncev = mm256_intrlv_blend_32(
|
||||
_mm256_set_epi32( n+3, 0, n+2, 0, n+1, 0, n, 0 ), *noncev );
|
||||
do
|
||||
{
|
||||
*noncev = mm256_intrlv_blend_32( mm256_bswap_32(
|
||||
_mm256_set_epi32( n+3, 0, n+2, 0, n+1, 0, n, 0 ) ), *noncev );
|
||||
quark_4way_hash( hash64, vdata );
|
||||
|
||||
quark_4way_hash( hash, vdata );
|
||||
pdata[19] = n;
|
||||
|
||||
for ( int i = 0; i < 4; i++ )
|
||||
if ( unlikely( hash7[ i<<1 ] <= Htarg ) )
|
||||
for ( int lane = 0; lane < 4; lane++ )
|
||||
if ( hash64_q3[ lane ] <= targ64_q3 && !bench )
|
||||
{
|
||||
extr_lane_4x64( lane_hash, hash, i, 256 );
|
||||
if ( likely( fulltest( lane_hash, ptarget ) && !opt_benchmark ) )
|
||||
extr_lane_4x64( lane_hash, hash64, lane, 256 );
|
||||
if ( valid_hash( lane_hash, ptarget ) )
|
||||
{
|
||||
pdata[19] = n+i;
|
||||
submit_lane_solution( work, lane_hash, mythr, i );
|
||||
pdata[19] = bswap_32( n + lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
*noncev = _mm256_add_epi32( *noncev,
|
||||
m256_const1_64( 0x0000000400000000 ) );
|
||||
n += 4;
|
||||
} while ( ( n < max_nonce ) && !work_restart[thr_id].restart );
|
||||
} while ( likely( ( n < last_nonce ) && !work_restart[thr_id].restart ) );
|
||||
|
||||
*hashes_done = n - first_nonce + 1;
|
||||
pdata[19] = n;
|
||||
*hashes_done = n - first_nonce;
|
||||
return 0;
|
||||
}
|
||||
|
||||
|
||||
@@ -106,13 +106,13 @@ int scanhash_deep_2way( struct work *work,uint32_t max_nonce,
|
||||
if ( fulltest( hash, ptarget) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n;
|
||||
submit_lane_solution( work, hash, mythr, 0 );
|
||||
submit_solution( work, hash, mythr );
|
||||
}
|
||||
if ( !( (hash+8)[7] & mask ) )
|
||||
if ( fulltest( hash+8, ptarget) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n+1;
|
||||
submit_lane_solution( work, hash+8, mythr, 1 );
|
||||
submit_solution( work, hash+8, mythr );
|
||||
}
|
||||
n += 2;
|
||||
} while ( ( n < max_nonce ) && !work_restart[thr_id].restart );
|
||||
|
||||
@@ -153,7 +153,7 @@ int scanhash_qubit_4way( struct work *work,uint32_t max_nonce,
|
||||
if ( likely( fulltest( hash+(lane<<3), ptarget) && !opt_benchmark ) )
|
||||
{
|
||||
pdata[19] = n + lane;
|
||||
submit_lane_solution( work, hash+(lane<<3), mythr, lane );
|
||||
submit_solution( work, hash+(lane<<3), mythr );
|
||||
}
|
||||
n += 4;
|
||||
} while ( ( n < max_nonce-4 ) && !work_restart[thr_id].restart );
|
||||
@@ -255,13 +255,13 @@ int scanhash_qubit_2way( struct work *work,uint32_t max_nonce,
|
||||
if ( likely( fulltest( hash, ptarget) && !opt_benchmark ) )
|
||||
{
|
||||
pdata[19] = n;
|
||||
submit_lane_solution( work, hash, mythr, 0 );
|
||||
submit_solution( work, hash, mythr );
|
||||
}
|
||||
if ( unlikely( ( (hash+8))[7] <= Htarg ) )
|
||||
if ( likely( fulltest( hash+8, ptarget) && !opt_benchmark ) )
|
||||
{
|
||||
pdata[19] = n+1;
|
||||
submit_lane_solution( work, hash+8, mythr, 1 );
|
||||
submit_solution( work, hash+8, mythr );
|
||||
}
|
||||
n += 2;
|
||||
} while ( ( n < max_nonce ) && !work_restart[thr_id].restart );
|
||||
|
||||
@@ -132,7 +132,7 @@ int scanhash_lbry_16way( struct work *work, uint32_t max_nonce,
|
||||
if ( likely( fulltest( lane_hash, ptarget ) && !opt_benchmark ) )
|
||||
{
|
||||
pdata[27] = n + i;
|
||||
submit_lane_solution( work, lane_hash, mythr, i );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
n += 16;
|
||||
@@ -251,7 +251,7 @@ int scanhash_lbry_8way( struct work *work, uint32_t max_nonce,
|
||||
if ( fulltest( lane_hash, ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[27] = n + i;
|
||||
submit_lane_solution( work, lane_hash, mythr, i );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
n += 8;
|
||||
|
||||
@@ -380,7 +380,7 @@ static inline void PBKDF2_SHA256_128_32_8way(uint32_t *tstate,
|
||||
#endif /* HAVE_SHA256_8WAY */
|
||||
|
||||
|
||||
#if defined(USE_ASM) && defined(__x86_64__)
|
||||
//#if defined(USE_ASM) && defined(__x86_64__)
|
||||
|
||||
#define SCRYPT_MAX_WAYS 12
|
||||
#define HAVE_SCRYPT_3WAY 1
|
||||
@@ -394,113 +394,6 @@ void scrypt_core_3way(uint32_t *X, uint32_t *V, int N);
|
||||
void scrypt_core_6way(uint32_t *X, uint32_t *V, int N);
|
||||
#endif
|
||||
|
||||
#elif defined(USE_ASM) && defined(__i386__)
|
||||
|
||||
#define SCRYPT_MAX_WAYS 4
|
||||
#define scrypt_best_throughput() 1
|
||||
void scrypt_core(uint32_t *X, uint32_t *V, int N);
|
||||
|
||||
#elif defined(USE_ASM) && defined(__arm__) && defined(__APCS_32__)
|
||||
|
||||
void scrypt_core(uint32_t *X, uint32_t *V, int N);
|
||||
#if defined(__ARM_NEON__)
|
||||
#undef HAVE_SHA256_4WAY
|
||||
#define SCRYPT_MAX_WAYS 3
|
||||
#define HAVE_SCRYPT_3WAY 1
|
||||
#define scrypt_best_throughput() 3
|
||||
void scrypt_core_3way(uint32_t *X, uint32_t *V, int N);
|
||||
#endif
|
||||
|
||||
#else
|
||||
|
||||
static inline void xor_salsa8(uint32_t B[16], const uint32_t Bx[16])
|
||||
{
|
||||
uint32_t x00,x01,x02,x03,x04,x05,x06,x07,x08,x09,x10,x11,x12,x13,x14,x15;
|
||||
int i;
|
||||
|
||||
x00 = (B[ 0] ^= Bx[ 0]);
|
||||
x01 = (B[ 1] ^= Bx[ 1]);
|
||||
x02 = (B[ 2] ^= Bx[ 2]);
|
||||
x03 = (B[ 3] ^= Bx[ 3]);
|
||||
x04 = (B[ 4] ^= Bx[ 4]);
|
||||
x05 = (B[ 5] ^= Bx[ 5]);
|
||||
x06 = (B[ 6] ^= Bx[ 6]);
|
||||
x07 = (B[ 7] ^= Bx[ 7]);
|
||||
x08 = (B[ 8] ^= Bx[ 8]);
|
||||
x09 = (B[ 9] ^= Bx[ 9]);
|
||||
x10 = (B[10] ^= Bx[10]);
|
||||
x11 = (B[11] ^= Bx[11]);
|
||||
x12 = (B[12] ^= Bx[12]);
|
||||
x13 = (B[13] ^= Bx[13]);
|
||||
x14 = (B[14] ^= Bx[14]);
|
||||
x15 = (B[15] ^= Bx[15]);
|
||||
for (i = 0; i < 8; i += 2) {
|
||||
#define R(a, b) (((a) << (b)) | ((a) >> (32 - (b))))
|
||||
/* Operate on columns. */
|
||||
x04 ^= R(x00+x12, 7); x09 ^= R(x05+x01, 7);
|
||||
x14 ^= R(x10+x06, 7); x03 ^= R(x15+x11, 7);
|
||||
|
||||
x08 ^= R(x04+x00, 9); x13 ^= R(x09+x05, 9);
|
||||
x02 ^= R(x14+x10, 9); x07 ^= R(x03+x15, 9);
|
||||
|
||||
x12 ^= R(x08+x04,13); x01 ^= R(x13+x09,13);
|
||||
x06 ^= R(x02+x14,13); x11 ^= R(x07+x03,13);
|
||||
|
||||
x00 ^= R(x12+x08,18); x05 ^= R(x01+x13,18);
|
||||
x10 ^= R(x06+x02,18); x15 ^= R(x11+x07,18);
|
||||
|
||||
/* Operate on rows. */
|
||||
x01 ^= R(x00+x03, 7); x06 ^= R(x05+x04, 7);
|
||||
x11 ^= R(x10+x09, 7); x12 ^= R(x15+x14, 7);
|
||||
|
||||
x02 ^= R(x01+x00, 9); x07 ^= R(x06+x05, 9);
|
||||
x08 ^= R(x11+x10, 9); x13 ^= R(x12+x15, 9);
|
||||
|
||||
x03 ^= R(x02+x01,13); x04 ^= R(x07+x06,13);
|
||||
x09 ^= R(x08+x11,13); x14 ^= R(x13+x12,13);
|
||||
|
||||
x00 ^= R(x03+x02,18); x05 ^= R(x04+x07,18);
|
||||
x10 ^= R(x09+x08,18); x15 ^= R(x14+x13,18);
|
||||
#undef R
|
||||
}
|
||||
B[ 0] += x00;
|
||||
B[ 1] += x01;
|
||||
B[ 2] += x02;
|
||||
B[ 3] += x03;
|
||||
B[ 4] += x04;
|
||||
B[ 5] += x05;
|
||||
B[ 6] += x06;
|
||||
B[ 7] += x07;
|
||||
B[ 8] += x08;
|
||||
B[ 9] += x09;
|
||||
B[10] += x10;
|
||||
B[11] += x11;
|
||||
B[12] += x12;
|
||||
B[13] += x13;
|
||||
B[14] += x14;
|
||||
B[15] += x15;
|
||||
}
|
||||
|
||||
static inline void scrypt_core(uint32_t *X, uint32_t *V, int N)
|
||||
{
|
||||
int i;
|
||||
|
||||
for (i = 0; i < N; i++) {
|
||||
memcpy(&V[i * 32], X, 128);
|
||||
xor_salsa8(&X[0], &X[16]);
|
||||
xor_salsa8(&X[16], &X[0]);
|
||||
}
|
||||
for (i = 0; i < N; i++) {
|
||||
uint32_t j = 32 * (X[16] & (N - 1));
|
||||
for (uint8_t k = 0; k < 32; k++)
|
||||
X[k] ^= V[j + k];
|
||||
xor_salsa8(&X[0], &X[16]);
|
||||
xor_salsa8(&X[16], &X[0]);
|
||||
}
|
||||
}
|
||||
|
||||
#endif
|
||||
|
||||
#ifndef SCRYPT_MAX_WAYS
|
||||
#define SCRYPT_MAX_WAYS 1
|
||||
#define scrypt_best_throughput() 1
|
||||
@@ -511,8 +404,8 @@ unsigned char *scrypt_buffer_alloc(int N)
|
||||
return (uchar*) malloc((size_t)N * SCRYPT_MAX_WAYS * 128 + 63);
|
||||
}
|
||||
|
||||
static void scrypt_1024_1_1_256(const uint32_t *input, uint32_t *output,
|
||||
uint32_t *midstate, unsigned char *scratchpad, int N)
|
||||
static bool scrypt_1024_1_1_256(const uint32_t *input, uint32_t *output,
|
||||
uint32_t *midstate, unsigned char *scratchpad, int N, int thr_id )
|
||||
{
|
||||
uint32_t tstate[8], ostate[8];
|
||||
uint32_t X[32];
|
||||
@@ -527,11 +420,13 @@ static void scrypt_1024_1_1_256(const uint32_t *input, uint32_t *output,
|
||||
scrypt_core(X, V, N);
|
||||
|
||||
PBKDF2_SHA256_128_32(tstate, ostate, X, output);
|
||||
return true;
|
||||
}
|
||||
|
||||
#ifdef HAVE_SHA256_4WAY
|
||||
static void scrypt_1024_1_1_256_4way(const uint32_t *input,
|
||||
uint32_t *output, uint32_t *midstate, unsigned char *scratchpad, int N)
|
||||
static int scrypt_1024_1_1_256_4way(const uint32_t *input,
|
||||
uint32_t *output, uint32_t *midstate, unsigned char *scratchpad, int N,
|
||||
int thrid )
|
||||
{
|
||||
uint32_t _ALIGN(128) tstate[4 * 8];
|
||||
uint32_t _ALIGN(128) ostate[4 * 8];
|
||||
@@ -545,32 +440,47 @@ static void scrypt_1024_1_1_256_4way(const uint32_t *input,
|
||||
for (i = 0; i < 20; i++)
|
||||
for (k = 0; k < 4; k++)
|
||||
W[4 * i + k] = input[k * 20 + i];
|
||||
for (i = 0; i < 8; i++)
|
||||
|
||||
for (i = 0; i < 8; i++)
|
||||
for (k = 0; k < 4; k++)
|
||||
tstate[4 * i + k] = midstate[i];
|
||||
HMAC_SHA256_80_init_4way(W, tstate, ostate);
|
||||
PBKDF2_SHA256_80_128_4way(tstate, ostate, W, W);
|
||||
for (i = 0; i < 32; i++)
|
||||
|
||||
HMAC_SHA256_80_init_4way(W, tstate, ostate);
|
||||
|
||||
PBKDF2_SHA256_80_128_4way(tstate, ostate, W, W);
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
for (i = 0; i < 32; i++)
|
||||
for (k = 0; k < 4; k++)
|
||||
X[k * 32 + i] = W[4 * i + k];
|
||||
scrypt_core(X + 0 * 32, V, N);
|
||||
|
||||
scrypt_core(X + 0 * 32, V, N);
|
||||
scrypt_core(X + 1 * 32, V, N);
|
||||
scrypt_core(X + 2 * 32, V, N);
|
||||
scrypt_core(X + 3 * 32, V, N);
|
||||
for (i = 0; i < 32; i++)
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
for (i = 0; i < 32; i++)
|
||||
for (k = 0; k < 4; k++)
|
||||
W[4 * i + k] = X[k * 32 + i];
|
||||
PBKDF2_SHA256_128_32_4way(tstate, ostate, W, W);
|
||||
for (i = 0; i < 8; i++)
|
||||
|
||||
PBKDF2_SHA256_128_32_4way(tstate, ostate, W, W);
|
||||
|
||||
for (i = 0; i < 8; i++)
|
||||
for (k = 0; k < 4; k++)
|
||||
output[k * 8 + i] = W[4 * i + k];
|
||||
|
||||
return 1;
|
||||
}
|
||||
#endif /* HAVE_SHA256_4WAY */
|
||||
|
||||
#ifdef HAVE_SCRYPT_3WAY
|
||||
|
||||
static void scrypt_1024_1_1_256_3way(const uint32_t *input,
|
||||
uint32_t *output, uint32_t *midstate, unsigned char *scratchpad, int N)
|
||||
static int scrypt_1024_1_1_256_3way(const uint32_t *input,
|
||||
uint32_t *output, uint32_t *midstate, unsigned char *scratchpad, int N,
|
||||
int thrid )
|
||||
{
|
||||
uint32_t _ALIGN(64) tstate[3 * 8], ostate[3 * 8];
|
||||
uint32_t _ALIGN(64) X[3 * 32];
|
||||
@@ -581,23 +491,34 @@ static void scrypt_1024_1_1_256_3way(const uint32_t *input,
|
||||
memcpy(tstate + 0, midstate, 32);
|
||||
memcpy(tstate + 8, midstate, 32);
|
||||
memcpy(tstate + 16, midstate, 32);
|
||||
HMAC_SHA256_80_init(input + 0, tstate + 0, ostate + 0);
|
||||
|
||||
HMAC_SHA256_80_init(input + 0, tstate + 0, ostate + 0);
|
||||
HMAC_SHA256_80_init(input + 20, tstate + 8, ostate + 8);
|
||||
HMAC_SHA256_80_init(input + 40, tstate + 16, ostate + 16);
|
||||
PBKDF2_SHA256_80_128(tstate + 0, ostate + 0, input + 0, X + 0);
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
PBKDF2_SHA256_80_128(tstate + 0, ostate + 0, input + 0, X + 0);
|
||||
PBKDF2_SHA256_80_128(tstate + 8, ostate + 8, input + 20, X + 32);
|
||||
PBKDF2_SHA256_80_128(tstate + 16, ostate + 16, input + 40, X + 64);
|
||||
|
||||
scrypt_core_3way(X, V, N);
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
PBKDF2_SHA256_128_32(tstate + 0, ostate + 0, X + 0, output + 0);
|
||||
scrypt_core_3way(X, V, N);
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
PBKDF2_SHA256_128_32(tstate + 0, ostate + 0, X + 0, output + 0);
|
||||
PBKDF2_SHA256_128_32(tstate + 8, ostate + 8, X + 32, output + 8);
|
||||
PBKDF2_SHA256_128_32(tstate + 16, ostate + 16, X + 64, output + 16);
|
||||
|
||||
return 1;
|
||||
}
|
||||
|
||||
#ifdef HAVE_SHA256_4WAY
|
||||
static void scrypt_1024_1_1_256_12way(const uint32_t *input,
|
||||
uint32_t *output, uint32_t *midstate, unsigned char *scratchpad, int N)
|
||||
static bool scrypt_1024_1_1_256_12way(const uint32_t *input,
|
||||
uint32_t *output, uint32_t *midstate, unsigned char *scratchpad, int N,
|
||||
int thrid )
|
||||
{
|
||||
uint32_t _ALIGN(128) tstate[12 * 8];
|
||||
uint32_t _ALIGN(128) ostate[12 * 8];
|
||||
@@ -612,43 +533,60 @@ static void scrypt_1024_1_1_256_12way(const uint32_t *input,
|
||||
for (i = 0; i < 20; i++)
|
||||
for (k = 0; k < 4; k++)
|
||||
W[128 * j + 4 * i + k] = input[80 * j + k * 20 + i];
|
||||
for (j = 0; j < 3; j++)
|
||||
|
||||
for (j = 0; j < 3; j++)
|
||||
for (i = 0; i < 8; i++)
|
||||
for (k = 0; k < 4; k++)
|
||||
tstate[32 * j + 4 * i + k] = midstate[i];
|
||||
HMAC_SHA256_80_init_4way(W + 0, tstate + 0, ostate + 0);
|
||||
|
||||
HMAC_SHA256_80_init_4way(W + 0, tstate + 0, ostate + 0);
|
||||
HMAC_SHA256_80_init_4way(W + 128, tstate + 32, ostate + 32);
|
||||
HMAC_SHA256_80_init_4way(W + 256, tstate + 64, ostate + 64);
|
||||
PBKDF2_SHA256_80_128_4way(tstate + 0, ostate + 0, W + 0, W + 0);
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
PBKDF2_SHA256_80_128_4way(tstate + 0, ostate + 0, W + 0, W + 0);
|
||||
PBKDF2_SHA256_80_128_4way(tstate + 32, ostate + 32, W + 128, W + 128);
|
||||
PBKDF2_SHA256_80_128_4way(tstate + 64, ostate + 64, W + 256, W + 256);
|
||||
for (j = 0; j < 3; j++)
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
for (j = 0; j < 3; j++)
|
||||
for (i = 0; i < 32; i++)
|
||||
for (k = 0; k < 4; k++)
|
||||
X[128 * j + k * 32 + i] = W[128 * j + 4 * i + k];
|
||||
scrypt_core_3way(X + 0 * 96, V, N);
|
||||
|
||||
scrypt_core_3way(X + 0 * 96, V, N);
|
||||
scrypt_core_3way(X + 1 * 96, V, N);
|
||||
scrypt_core_3way(X + 2 * 96, V, N);
|
||||
scrypt_core_3way(X + 3 * 96, V, N);
|
||||
for (j = 0; j < 3; j++)
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
for (j = 0; j < 3; j++)
|
||||
for (i = 0; i < 32; i++)
|
||||
for (k = 0; k < 4; k++)
|
||||
W[128 * j + 4 * i + k] = X[128 * j + k * 32 + i];
|
||||
PBKDF2_SHA256_128_32_4way(tstate + 0, ostate + 0, W + 0, W + 0);
|
||||
|
||||
PBKDF2_SHA256_128_32_4way(tstate + 0, ostate + 0, W + 0, W + 0);
|
||||
PBKDF2_SHA256_128_32_4way(tstate + 32, ostate + 32, W + 128, W + 128);
|
||||
PBKDF2_SHA256_128_32_4way(tstate + 64, ostate + 64, W + 256, W + 256);
|
||||
for (j = 0; j < 3; j++)
|
||||
|
||||
for (j = 0; j < 3; j++)
|
||||
for (i = 0; i < 8; i++)
|
||||
for (k = 0; k < 4; k++)
|
||||
output[32 * j + k * 8 + i] = W[128 * j + 4 * i + k];
|
||||
|
||||
return 1;
|
||||
}
|
||||
#endif /* HAVE_SHA256_4WAY */
|
||||
|
||||
#endif /* HAVE_SCRYPT_3WAY */
|
||||
|
||||
#ifdef HAVE_SCRYPT_6WAY
|
||||
static void scrypt_1024_1_1_256_24way(const uint32_t *input,
|
||||
uint32_t *output, uint32_t *midstate, unsigned char *scratchpad, int N)
|
||||
static int scrypt_1024_1_1_256_24way( const uint32_t *input,
|
||||
uint32_t *output, uint32_t *midstate,
|
||||
unsigned char *scratchpad, int N, int thrid )
|
||||
{
|
||||
uint32_t _ALIGN(128) tstate[24 * 8];
|
||||
uint32_t _ALIGN(128) ostate[24 * 8];
|
||||
@@ -657,41 +595,60 @@ static void scrypt_1024_1_1_256_24way(const uint32_t *input,
|
||||
uint32_t *V;
|
||||
int i, j, k;
|
||||
|
||||
V = (uint32_t *)(((uintptr_t)(scratchpad) + 63) & ~ (uintptr_t)(63));
|
||||
V = (uint32_t *)( ( (uintptr_t)(scratchpad) + 63 ) & ~ (uintptr_t)(63) );
|
||||
|
||||
for (j = 0; j < 3; j++)
|
||||
for (i = 0; i < 20; i++)
|
||||
for (k = 0; k < 8; k++)
|
||||
for ( j = 0; j < 3; j++ )
|
||||
for ( i = 0; i < 20; i++ )
|
||||
for ( k = 0; k < 8; k++ )
|
||||
W[8 * 32 * j + 8 * i + k] = input[8 * 20 * j + k * 20 + i];
|
||||
for (j = 0; j < 3; j++)
|
||||
for (i = 0; i < 8; i++)
|
||||
for (k = 0; k < 8; k++)
|
||||
|
||||
for ( j = 0; j < 3; j++ )
|
||||
for ( i = 0; i < 8; i++ )
|
||||
for ( k = 0; k < 8; k++ )
|
||||
tstate[8 * 8 * j + 8 * i + k] = midstate[i];
|
||||
HMAC_SHA256_80_init_8way(W + 0, tstate + 0, ostate + 0);
|
||||
HMAC_SHA256_80_init_8way(W + 256, tstate + 64, ostate + 64);
|
||||
HMAC_SHA256_80_init_8way(W + 512, tstate + 128, ostate + 128);
|
||||
PBKDF2_SHA256_80_128_8way(tstate + 0, ostate + 0, W + 0, W + 0);
|
||||
PBKDF2_SHA256_80_128_8way(tstate + 64, ostate + 64, W + 256, W + 256);
|
||||
PBKDF2_SHA256_80_128_8way(tstate + 128, ostate + 128, W + 512, W + 512);
|
||||
for (j = 0; j < 3; j++)
|
||||
for (i = 0; i < 32; i++)
|
||||
for (k = 0; k < 8; k++)
|
||||
|
||||
HMAC_SHA256_80_init_8way( W + 0, tstate + 0, ostate + 0 );
|
||||
HMAC_SHA256_80_init_8way( W + 256, tstate + 64, ostate + 64 );
|
||||
HMAC_SHA256_80_init_8way( W + 512, tstate + 128, ostate + 128 );
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
PBKDF2_SHA256_80_128_8way( tstate + 0, ostate + 0, W + 0, W + 0 );
|
||||
PBKDF2_SHA256_80_128_8way( tstate + 64, ostate + 64, W + 256, W + 256 );
|
||||
PBKDF2_SHA256_80_128_8way( tstate + 128, ostate + 128, W + 512, W + 512 );
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
for ( j = 0; j < 3; j++ )
|
||||
for ( i = 0; i < 32; i++ )
|
||||
for ( k = 0; k < 8; k++ )
|
||||
X[8 * 32 * j + k * 32 + i] = W[8 * 32 * j + 8 * i + k];
|
||||
scrypt_core_6way(X + 0 * 32, V, N);
|
||||
scrypt_core_6way(X + 6 * 32, V, N);
|
||||
scrypt_core_6way(X + 12 * 32, V, N);
|
||||
scrypt_core_6way(X + 18 * 32, V, N);
|
||||
for (j = 0; j < 3; j++)
|
||||
for (i = 0; i < 32; i++)
|
||||
for (k = 0; k < 8; k++)
|
||||
|
||||
scrypt_core_6way( X + 0 * 32, V, N );
|
||||
scrypt_core_6way( X + 6 * 32, V, N );
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
scrypt_core_6way( X + 12 * 32, V, N );
|
||||
scrypt_core_6way( X + 18 * 32, V, N );
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
for ( j = 0; j < 3; j++ )
|
||||
for ( i = 0; i < 32; i++ )
|
||||
for ( k = 0; k < 8; k++ )
|
||||
W[8 * 32 * j + 8 * i + k] = X[8 * 32 * j + k * 32 + i];
|
||||
PBKDF2_SHA256_128_32_8way(tstate + 0, ostate + 0, W + 0, W + 0);
|
||||
PBKDF2_SHA256_128_32_8way(tstate + 64, ostate + 64, W + 256, W + 256);
|
||||
PBKDF2_SHA256_128_32_8way(tstate + 128, ostate + 128, W + 512, W + 512);
|
||||
for (j = 0; j < 3; j++)
|
||||
for (i = 0; i < 8; i++)
|
||||
for (k = 0; k < 8; k++)
|
||||
|
||||
PBKDF2_SHA256_128_32_8way( tstate + 0, ostate + 0, W + 0, W + 0 );
|
||||
PBKDF2_SHA256_128_32_8way( tstate + 64, ostate + 64, W + 256, W + 256 );
|
||||
PBKDF2_SHA256_128_32_8way( tstate + 128, ostate + 128, W + 512, W + 512 );
|
||||
|
||||
for ( j = 0; j < 3; j++ )
|
||||
for ( i = 0; i < 8; i++ )
|
||||
for ( k = 0; k < 8; k++ )
|
||||
output[8 * 8 * j + k * 8 + i] = W[8 * 32 * j + 8 * i + k];
|
||||
|
||||
return 1;
|
||||
}
|
||||
#endif /* HAVE_SCRYPT_6WAY */
|
||||
|
||||
@@ -703,16 +660,18 @@ extern int scanhash_scrypt( struct work *work, uint32_t max_nonce,
|
||||
uint32_t data[SCRYPT_MAX_WAYS * 20], hash[SCRYPT_MAX_WAYS * 8];
|
||||
uint32_t midstate[8];
|
||||
uint32_t n = pdata[19] - 1;
|
||||
const uint32_t Htarg = ptarget[7];
|
||||
int thr_id = mythr->id; // thr_id arg is deprecated
|
||||
int throughput = scrypt_best_throughput();
|
||||
int i;
|
||||
volatile uint8_t *restart = &(work_restart[thr_id].restart);
|
||||
|
||||
#ifdef HAVE_SHA256_4WAY
|
||||
if (sha256_use_4way())
|
||||
throughput *= 4;
|
||||
#endif
|
||||
|
||||
// applog(LOG_INFO,"Scrypt thoughput %d",throughput);
|
||||
|
||||
for (i = 0; i < throughput; i++)
|
||||
memcpy(data + i * 20, pdata, 80);
|
||||
|
||||
@@ -720,46 +679,50 @@ extern int scanhash_scrypt( struct work *work, uint32_t max_nonce,
|
||||
sha256_transform(midstate, data, 0);
|
||||
|
||||
do {
|
||||
|
||||
bool rc = true;
|
||||
for (i = 0; i < throughput; i++)
|
||||
data[i * 20 + 19] = ++n;
|
||||
|
||||
#if defined(HAVE_SHA256_4WAY)
|
||||
if (throughput == 4)
|
||||
scrypt_1024_1_1_256_4way(data, hash, midstate,
|
||||
scratchbuf, scratchbuf_size );
|
||||
rc = scrypt_1024_1_1_256_4way(data, hash, midstate,
|
||||
scratchbuf, scratchbuf_size, thr_id );
|
||||
else
|
||||
#endif
|
||||
#if defined(HAVE_SCRYPT_3WAY) && defined(HAVE_SHA256_4WAY)
|
||||
if (throughput == 12)
|
||||
scrypt_1024_1_1_256_12way(data, hash, midstate,
|
||||
scratchbuf, scratchbuf_size );
|
||||
rc = scrypt_1024_1_1_256_12way(data, hash, midstate,
|
||||
scratchbuf, scratchbuf_size, thr_id );
|
||||
else
|
||||
#endif
|
||||
#if defined(HAVE_SCRYPT_6WAY)
|
||||
if (throughput == 24)
|
||||
scrypt_1024_1_1_256_24way(data, hash, midstate,
|
||||
scratchbuf, scratchbuf_size );
|
||||
rc = scrypt_1024_1_1_256_24way(data, hash, midstate,
|
||||
scratchbuf, scratchbuf_size, thr_id );
|
||||
else
|
||||
#endif
|
||||
#if defined(HAVE_SCRYPT_3WAY)
|
||||
if (throughput == 3)
|
||||
scrypt_1024_1_1_256_3way(data, hash, midstate,
|
||||
scratchbuf, scratchbuf_size );
|
||||
rc = scrypt_1024_1_1_256_3way(data, hash, midstate,
|
||||
scratchbuf, scratchbuf_size, thr_id );
|
||||
else
|
||||
#endif
|
||||
scrypt_1024_1_1_256(data, hash, midstate, scratchbuf,
|
||||
scratchbuf_size );
|
||||
rc = scrypt_1024_1_1_256(data, hash, midstate, scratchbuf,
|
||||
scratchbuf_size, thr_id );
|
||||
|
||||
for (i = 0; i < throughput; i++) {
|
||||
if (unlikely(hash[i * 8 + 7] <= Htarg && fulltest(hash + i * 8, ptarget))) {
|
||||
if ( rc )
|
||||
for ( i = 0; i < throughput; i++ )
|
||||
{
|
||||
if ( unlikely( valid_hash( hash + i * 8, ptarget ) ) )
|
||||
{
|
||||
pdata[19] = data[i * 20 + 19];
|
||||
submit_solution( work, hash, mythr );
|
||||
submit_solution( work, hash + i * 8, mythr );
|
||||
}
|
||||
}
|
||||
} while (likely(n < max_nonce && !work_restart[thr_id].restart));
|
||||
|
||||
}
|
||||
} while ( likely( ( n < ( max_nonce - throughput ) ) && !(*restart) ) );
|
||||
|
||||
*hashes_done = n - pdata[19] + 1;
|
||||
*hashes_done = n - pdata[19];
|
||||
pdata[19] = n;
|
||||
return 0;
|
||||
}
|
||||
@@ -778,7 +741,6 @@ bool register_scrypt_algo( algo_gate_t* gate )
|
||||
gate->optimizations = SSE2_OPT | AVX2_OPT;
|
||||
gate->miner_thread_init =(void*)&scrypt_miner_thread_init;
|
||||
gate->scanhash = (void*)&scanhash_scrypt;
|
||||
// gate->hash = (void*)&scrypt_1024_1_1_256_24way;
|
||||
opt_target_factor = 65536.0;
|
||||
|
||||
if ( !opt_param_n )
|
||||
|
||||
440
algo/sha/hmac-sha256-hash-4way.c
Normal file
440
algo/sha/hmac-sha256-hash-4way.c
Normal file
@@ -0,0 +1,440 @@
|
||||
/*-
|
||||
* Copyright 2005,2007,2009 Colin Percival
|
||||
* Copywright 2020 JayDDee246@gmail.com
|
||||
* All rights reserved.
|
||||
*
|
||||
* Redistribution and use in source and binary forms, with or without
|
||||
* modification, are permitted provided that the following conditions
|
||||
* are met:
|
||||
* 1. Redistributions of source code must retain the above copyright
|
||||
* notice, this list of conditions and the following disclaimer.
|
||||
* 2. Redistributions in binary form must reproduce the above copyright
|
||||
* notice, this list of conditions and the following disclaimer in the
|
||||
* documentation and/or other materials provided with the distribution.
|
||||
*
|
||||
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
|
||||
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
||||
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
|
||||
* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
|
||||
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
|
||||
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
|
||||
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
|
||||
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
|
||||
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
|
||||
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
|
||||
* SUCH DAMAGE.
|
||||
*/
|
||||
|
||||
#include <sys/types.h>
|
||||
#include <stdint.h>
|
||||
#include <string.h>
|
||||
#include "hmac-sha256-hash-4way.h"
|
||||
#include "compat.h"
|
||||
|
||||
// HMAC 4-way SSE2
|
||||
|
||||
/**
|
||||
* HMAC_SHA256_Buf(K, Klen, in, len, digest):
|
||||
* Compute the HMAC-SHA256 of ${len} bytes from ${in} using the key ${K} of
|
||||
* length ${Klen}, and write the result to ${digest}.
|
||||
*/
|
||||
void
|
||||
hmac_sha256_4way_full( void *digest, const void *K, size_t Klen,
|
||||
const void *in, size_t len )
|
||||
{
|
||||
hmac_sha256_4way_context ctx;
|
||||
hmac_sha256_4way_init( &ctx, K, Klen );
|
||||
hmac_sha256_4way_update( &ctx, in, len );
|
||||
hmac_sha256_4way_close( &ctx, digest );
|
||||
}
|
||||
|
||||
/* Initialize an HMAC-SHA256 operation with the given key. */
|
||||
void
|
||||
hmac_sha256_4way_init( hmac_sha256_4way_context *ctx, const void *_K,
|
||||
size_t Klen )
|
||||
{
|
||||
unsigned char pad[64*4] __attribute__ ((aligned (64)));
|
||||
unsigned char khash[32*4] __attribute__ ((aligned (64)));
|
||||
const unsigned char * K = _K;
|
||||
size_t i;
|
||||
|
||||
/* If Klen > 64, the key is really SHA256(K). */
|
||||
if ( Klen > 64 )
|
||||
{
|
||||
sha256_4way_init( &ctx->ictx );
|
||||
sha256_4way_update( &ctx->ictx, K, Klen );
|
||||
sha256_4way_close( &ctx->ictx, khash );
|
||||
K = khash;
|
||||
Klen = 32;
|
||||
}
|
||||
|
||||
/* Inner SHA256 operation is SHA256(K xor [block of 0x36] || data). */
|
||||
sha256_4way_init( &ctx->ictx );
|
||||
memset( pad, 0x36, 64*4 );
|
||||
|
||||
for ( i = 0; i < Klen; i++ )
|
||||
casti_m128i( pad, i ) = _mm_xor_si128( casti_m128i( pad, i ),
|
||||
casti_m128i( K, i ) );
|
||||
|
||||
sha256_4way_update( &ctx->ictx, pad, 64 );
|
||||
|
||||
/* Outer SHA256 operation is SHA256(K xor [block of 0x5c] || hash). */
|
||||
sha256_4way_init( &ctx->octx );
|
||||
memset( pad, 0x5c, 64*4 );
|
||||
for ( i = 0; i < Klen/4; i++ )
|
||||
casti_m128i( pad, i ) = _mm_xor_si128( casti_m128i( pad, i ),
|
||||
casti_m128i( K, i ) );
|
||||
sha256_4way_update( &ctx->octx, pad, 64 );
|
||||
}
|
||||
|
||||
/* Add bytes to the HMAC-SHA256 operation. */
|
||||
void
|
||||
hmac_sha256_4way_update( hmac_sha256_4way_context *ctx, const void *in,
|
||||
size_t len )
|
||||
{
|
||||
/* Feed data to the inner SHA256 operation. */
|
||||
sha256_4way_update( &ctx->ictx, in, len );
|
||||
}
|
||||
|
||||
/* Finish an HMAC-SHA256 operation. */
|
||||
void
|
||||
hmac_sha256_4way_close( hmac_sha256_4way_context *ctx, void *digest )
|
||||
{
|
||||
unsigned char ihash[32*4] __attribute__ ((aligned (64)));
|
||||
|
||||
/* Finish the inner SHA256 operation. */
|
||||
sha256_4way_close( &ctx->ictx, ihash );
|
||||
|
||||
/* Feed the inner hash to the outer SHA256 operation. */
|
||||
sha256_4way_update( &ctx->octx, ihash, 32 );
|
||||
|
||||
/* Finish the outer SHA256 operation. */
|
||||
sha256_4way_close( &ctx->octx, digest );
|
||||
}
|
||||
|
||||
/**
|
||||
* PBKDF2_SHA256(passwd, passwdlen, salt, saltlen, c, buf, dkLen):
|
||||
* Compute PBKDF2(passwd, salt, c, dkLen) using HMAC-SHA256 as the PRF, and
|
||||
* write the output to buf. The value dkLen must be at most 32 * (2^32 - 1).
|
||||
*/
|
||||
void
|
||||
pbkdf2_sha256_4way( uint8_t *buf, size_t dkLen,
|
||||
const uint8_t *passwd, size_t passwdlen,
|
||||
const uint8_t *salt, size_t saltlen, uint64_t c )
|
||||
{
|
||||
hmac_sha256_4way_context PShctx, hctx;
|
||||
uint8_t _ALIGN(128) T[32*4];
|
||||
uint8_t _ALIGN(128) U[32*4];
|
||||
__m128i ivec;
|
||||
size_t i, clen;
|
||||
uint64_t j;
|
||||
int k;
|
||||
|
||||
/* Compute HMAC state after processing P and S. */
|
||||
hmac_sha256_4way_init( &PShctx, passwd, passwdlen );
|
||||
hmac_sha256_4way_update( &PShctx, salt, saltlen );
|
||||
|
||||
/* Iterate through the blocks. */
|
||||
for ( i = 0; i * 32 < dkLen; i++ )
|
||||
{
|
||||
/* Generate INT(i + 1). */
|
||||
ivec = _mm_set1_epi32( bswap_32( i+1 ) );
|
||||
|
||||
/* Compute U_1 = PRF(P, S || INT(i)). */
|
||||
memcpy( &hctx, &PShctx, sizeof(hmac_sha256_4way_context) );
|
||||
hmac_sha256_4way_update( &hctx, &ivec, 4 );
|
||||
hmac_sha256_4way_close( &hctx, U );
|
||||
|
||||
/* T_i = U_1 ... */
|
||||
memcpy( T, U, 32*4 );
|
||||
|
||||
for ( j = 2; j <= c; j++ )
|
||||
{
|
||||
/* Compute U_j. */
|
||||
hmac_sha256_4way_init( &hctx, passwd, passwdlen );
|
||||
hmac_sha256_4way_update( &hctx, U, 32 );
|
||||
hmac_sha256_4way_close( &hctx, U );
|
||||
|
||||
/* ... xor U_j ... */
|
||||
for ( k = 0; k < 8; k++ )
|
||||
casti_m128i( T, k ) = _mm_xor_si128( casti_m128i( T, k ),
|
||||
casti_m128i( U, k ) );
|
||||
}
|
||||
|
||||
/* Copy as many bytes as necessary into buf. */
|
||||
clen = dkLen - i * 32;
|
||||
if ( clen > 32 )
|
||||
clen = 32;
|
||||
memcpy( &buf[ i*32*4 ], T, clen*4 );
|
||||
}
|
||||
}
|
||||
|
||||
#if defined(__AVX2__)
|
||||
|
||||
// HMAC 8-way AVX2
|
||||
|
||||
void
|
||||
hmac_sha256_8way_full( void *digest, const void *K, size_t Klen,
|
||||
const void *in, size_t len )
|
||||
{
|
||||
hmac_sha256_8way_context ctx;
|
||||
hmac_sha256_8way_init( &ctx, K, Klen );
|
||||
hmac_sha256_8way_update( &ctx, in, len );
|
||||
hmac_sha256_8way_close( &ctx, digest );
|
||||
}
|
||||
|
||||
/* Initialize an HMAC-SHA256 operation with the given key. */
|
||||
void
|
||||
hmac_sha256_8way_init( hmac_sha256_8way_context *ctx, const void *_K,
|
||||
size_t Klen )
|
||||
{
|
||||
unsigned char pad[64*8] __attribute__ ((aligned (128)));
|
||||
unsigned char khash[32*8] __attribute__ ((aligned (128)));
|
||||
const unsigned char * K = _K;
|
||||
size_t i;
|
||||
|
||||
/* If Klen > 64, the key is really SHA256(K). */
|
||||
if ( Klen > 64 )
|
||||
{
|
||||
sha256_8way_init( &ctx->ictx );
|
||||
sha256_8way_update( &ctx->ictx, K, Klen );
|
||||
sha256_8way_close( &ctx->ictx, khash );
|
||||
K = khash;
|
||||
Klen = 32;
|
||||
}
|
||||
|
||||
/* Inner SHA256 operation is SHA256(K xor [block of 0x36] || data). */
|
||||
sha256_8way_init( &ctx->ictx );
|
||||
memset( pad, 0x36, 64*8);
|
||||
|
||||
for ( i = 0; i < Klen/4; i++ )
|
||||
casti_m256i( pad, i ) = _mm256_xor_si256( casti_m256i( pad, i ),
|
||||
casti_m256i( K, i ) );
|
||||
|
||||
sha256_8way_update( &ctx->ictx, pad, 64 );
|
||||
|
||||
/* Outer SHA256 operation is SHA256(K xor [block of 0x5c] || hash). */
|
||||
sha256_8way_init( &ctx->octx );
|
||||
memset( pad, 0x5c, 64*8 );
|
||||
for ( i = 0; i < Klen/4; i++ )
|
||||
casti_m256i( pad, i ) = _mm256_xor_si256( casti_m256i( pad, i ),
|
||||
casti_m256i( K, i ) );
|
||||
sha256_8way_update( &ctx->octx, pad, 64 );
|
||||
}
|
||||
|
||||
void
|
||||
hmac_sha256_8way_update( hmac_sha256_8way_context *ctx, const void *in,
|
||||
size_t len )
|
||||
{
|
||||
/* Feed data to the inner SHA256 operation. */
|
||||
sha256_8way_update( &ctx->ictx, in, len );
|
||||
}
|
||||
|
||||
/* Finish an HMAC-SHA256 operation. */
|
||||
void
|
||||
hmac_sha256_8way_close( hmac_sha256_8way_context *ctx, void *digest )
|
||||
{
|
||||
unsigned char ihash[32*8] __attribute__ ((aligned (128)));
|
||||
|
||||
/* Finish the inner SHA256 operation. */
|
||||
sha256_8way_close( &ctx->ictx, ihash );
|
||||
|
||||
/* Feed the inner hash to the outer SHA256 operation. */
|
||||
sha256_8way_update( &ctx->octx, ihash, 32 );
|
||||
|
||||
/* Finish the outer SHA256 operation. */
|
||||
sha256_8way_close( &ctx->octx, digest );
|
||||
}
|
||||
|
||||
/**
|
||||
* PBKDF2_SHA256(passwd, passwdlen, salt, saltlen, c, buf, dkLen):
|
||||
* Compute PBKDF2(passwd, salt, c, dkLen) using HMAC-SHA256 as the PRF, and
|
||||
* write the output to buf. The value dkLen must be at most 32 * (2^32 - 1).
|
||||
*/
|
||||
void
|
||||
pbkdf2_sha256_8way( uint8_t *buf, size_t dkLen, const uint8_t *passwd,
|
||||
size_t passwdlen, const uint8_t *salt, size_t saltlen,
|
||||
uint64_t c )
|
||||
{
|
||||
hmac_sha256_8way_context PShctx, hctx;
|
||||
uint8_t _ALIGN(128) T[32*8];
|
||||
uint8_t _ALIGN(128) U[32*8];
|
||||
size_t i, clen;
|
||||
uint64_t j;
|
||||
int k;
|
||||
|
||||
/* Compute HMAC state after processing P and S. */
|
||||
hmac_sha256_8way_init( &PShctx, passwd, passwdlen );
|
||||
|
||||
// saltlen can be odd number of bytes
|
||||
hmac_sha256_8way_update( &PShctx, salt, saltlen );
|
||||
|
||||
/* Iterate through the blocks. */
|
||||
for ( i = 0; i * 32 < dkLen; i++ )
|
||||
{
|
||||
__m256i ivec = _mm256_set1_epi32( bswap_32( i+1 ) );
|
||||
|
||||
/* Compute U_1 = PRF(P, S || INT(i)). */
|
||||
memcpy( &hctx, &PShctx, sizeof(hmac_sha256_8way_context) );
|
||||
hmac_sha256_8way_update( &hctx, &ivec, 4 );
|
||||
hmac_sha256_8way_close( &hctx, U );
|
||||
|
||||
/* T_i = U_1 ... */
|
||||
memcpy( T, U, 32*8 );
|
||||
|
||||
for ( j = 2; j <= c; j++ )
|
||||
{
|
||||
/* Compute U_j. */
|
||||
hmac_sha256_8way_init( &hctx, passwd, passwdlen );
|
||||
hmac_sha256_8way_update( &hctx, U, 32 );
|
||||
hmac_sha256_8way_close( &hctx, U );
|
||||
|
||||
/* ... xor U_j ... */
|
||||
for ( k = 0; k < 8; k++ )
|
||||
casti_m256i( T, k ) = _mm256_xor_si256( casti_m256i( T, k ),
|
||||
casti_m256i( U, k ) );
|
||||
}
|
||||
|
||||
/* Copy as many bytes as necessary into buf. */
|
||||
clen = dkLen - i * 32;
|
||||
if ( clen > 32 )
|
||||
clen = 32;
|
||||
memcpy( &buf[ i*32*8 ], T, clen*8 );
|
||||
}
|
||||
}
|
||||
|
||||
#if defined(__AVX512F__) && defined(__AVX512VL__) && defined(__AVX512DQ__) && defined(__AVX512BW__)
|
||||
|
||||
// HMAC 16-way AVX512
|
||||
|
||||
void
|
||||
hmac_sha256_16way_full( void *digest, const void *K, size_t Klen,
|
||||
const void *in, size_t len )
|
||||
{
|
||||
hmac_sha256_16way_context ctx;
|
||||
hmac_sha256_16way_init( &ctx, K, Klen );
|
||||
hmac_sha256_16way_update( &ctx, in, len );
|
||||
hmac_sha256_16way_close( &ctx, digest );
|
||||
}
|
||||
|
||||
void
|
||||
hmac_sha256_16way_init( hmac_sha256_16way_context *ctx, const void *_K,
|
||||
size_t Klen )
|
||||
{
|
||||
unsigned char pad[64*16] __attribute__ ((aligned (128)));
|
||||
unsigned char khash[32*16] __attribute__ ((aligned (128)));
|
||||
const unsigned char * K = _K;
|
||||
size_t i;
|
||||
|
||||
/* If Klen > 64, the key is really SHA256(K). */
|
||||
if ( Klen > 64 )
|
||||
{
|
||||
sha256_16way_init( &ctx->ictx );
|
||||
sha256_16way_update( &ctx->ictx, K, Klen );
|
||||
sha256_16way_close( &ctx->ictx, khash );
|
||||
K = khash;
|
||||
Klen = 32;
|
||||
}
|
||||
|
||||
/* Inner SHA256 operation is SHA256(K xor [block of 0x36] || data). */
|
||||
sha256_16way_init( &ctx->ictx );
|
||||
memset( pad, 0x36, 64*16 );
|
||||
|
||||
for ( i = 0; i < Klen; i++ )
|
||||
casti_m512i( pad, i ) = _mm512_xor_si512( casti_m512i( pad, i ),
|
||||
casti_m512i( K, i ) );
|
||||
sha256_16way_update( &ctx->ictx, pad, 64 );
|
||||
|
||||
/* Outer SHA256 operation is SHA256(K xor [block of 0x5c] || hash). */
|
||||
sha256_16way_init( &ctx->octx );
|
||||
memset( pad, 0x5c, 64*16 );
|
||||
for ( i = 0; i < Klen/4; i++ )
|
||||
casti_m512i( pad, i ) = _mm512_xor_si512( casti_m512i( pad, i ),
|
||||
casti_m512i( K, i ) );
|
||||
sha256_16way_update( &ctx->octx, pad, 64 );
|
||||
}
|
||||
|
||||
void
|
||||
hmac_sha256_16way_update( hmac_sha256_16way_context *ctx, const void *in,
|
||||
size_t len )
|
||||
{
|
||||
/* Feed data to the inner SHA256 operation. */
|
||||
sha256_16way_update( &ctx->ictx, in, len );
|
||||
}
|
||||
|
||||
/* Finish an HMAC-SHA256 operation. */
|
||||
void
|
||||
hmac_sha256_16way_close( hmac_sha256_16way_context *ctx, void *digest )
|
||||
{
|
||||
unsigned char ihash[32*16] __attribute__ ((aligned (128)));
|
||||
|
||||
/* Finish the inner SHA256 operation. */
|
||||
sha256_16way_close( &ctx->ictx, ihash );
|
||||
|
||||
/* Feed the inner hash to the outer SHA256 operation. */
|
||||
sha256_16way_update( &ctx->octx, ihash, 32 );
|
||||
|
||||
/* Finish the outer SHA256 operation. */
|
||||
sha256_16way_close( &ctx->octx, digest );
|
||||
}
|
||||
|
||||
/**
|
||||
* PBKDF2_SHA256(passwd, passwdlen, salt, saltlen, c, buf, dkLen):
|
||||
* Compute PBKDF2(passwd, salt, c, dkLen) using HMAC-SHA256 as the PRF, and
|
||||
* write the output to buf. The value dkLen must be at most 32 * (2^32 - 1).
|
||||
*/
|
||||
void
|
||||
pbkdf2_sha256_16way( uint8_t *buf, size_t dkLen,
|
||||
const uint8_t *passwd, size_t passwdlen,
|
||||
const uint8_t *salt, size_t saltlen, uint64_t c )
|
||||
{
|
||||
hmac_sha256_16way_context PShctx, hctx;
|
||||
uint8_t _ALIGN(128) T[32*16];
|
||||
uint8_t _ALIGN(128) U[32*16];
|
||||
__m512i ivec;
|
||||
size_t i, clen;
|
||||
uint64_t j;
|
||||
int k;
|
||||
|
||||
/* Compute HMAC state after processing P and S. */
|
||||
hmac_sha256_16way_init( &PShctx, passwd, passwdlen );
|
||||
hmac_sha256_16way_update( &PShctx, salt, saltlen );
|
||||
|
||||
/* Iterate through the blocks. */
|
||||
for ( i = 0; i * 32 < dkLen; i++ )
|
||||
{
|
||||
/* Generate INT(i + 1). */
|
||||
ivec = _mm512_set1_epi32( bswap_32( i+1 ) );
|
||||
|
||||
/* Compute U_1 = PRF(P, S || INT(i)). */
|
||||
memcpy( &hctx, &PShctx, sizeof(hmac_sha256_16way_context) );
|
||||
hmac_sha256_16way_update( &hctx, &ivec, 4 );
|
||||
hmac_sha256_16way_close( &hctx, U );
|
||||
|
||||
/* T_i = U_1 ... */
|
||||
memcpy( T, U, 32*16 );
|
||||
|
||||
for ( j = 2; j <= c; j++ )
|
||||
{
|
||||
/* Compute U_j. */
|
||||
hmac_sha256_16way_init( &hctx, passwd, passwdlen );
|
||||
hmac_sha256_16way_update( &hctx, U, 32 );
|
||||
hmac_sha256_16way_close( &hctx, U );
|
||||
|
||||
/* ... xor U_j ... */
|
||||
for ( k = 0; k < 8; k++ )
|
||||
casti_m512i( T, k ) = _mm512_xor_si512( casti_m512i( T, k ),
|
||||
casti_m512i( U, k ) );
|
||||
}
|
||||
|
||||
/* Copy as many bytes as necessary into buf. */
|
||||
clen = dkLen - i * 32;
|
||||
if ( clen > 32 )
|
||||
clen = 32;
|
||||
memcpy( &buf[ i*32*16 ], T, clen*16 );
|
||||
}
|
||||
}
|
||||
|
||||
#endif // AVX512
|
||||
#endif // AVX2
|
||||
|
||||
107
algo/sha/hmac-sha256-hash-4way.h
Normal file
107
algo/sha/hmac-sha256-hash-4way.h
Normal file
@@ -0,0 +1,107 @@
|
||||
/*-
|
||||
* Copyright 2005,2007,2009 Colin Percival
|
||||
* Copyright 2020 JayDDee@gmailcom
|
||||
* All rights reserved.
|
||||
*
|
||||
* Redistribution and use in source and binary forms, with or without
|
||||
* modification, are permitted provided that the following conditions
|
||||
* are met:
|
||||
* 1. Redistributions of source code must retain the above copyright
|
||||
* notice, this list of conditions and the following disclaimer.
|
||||
* 2. Redistributions in binary form must reproduce the above copyright
|
||||
* notice, this list of conditions and the following disclaimer in the
|
||||
* documentation and/or other materials provided with the distribution.
|
||||
*
|
||||
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
|
||||
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
||||
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
|
||||
* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
|
||||
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
|
||||
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
|
||||
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
|
||||
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
|
||||
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
|
||||
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
|
||||
* SUCH DAMAGE.
|
||||
*
|
||||
* $FreeBSD: src/lib/libmd/sha256_Y.h,v 1.2 2006/01/17 15:35:56 phk Exp $
|
||||
*/
|
||||
|
||||
#ifndef HMAC_SHA256_4WAY_H__
|
||||
#define HMAC_SHA256_4WAY_H__
|
||||
|
||||
|
||||
// Tested only 8-way with null pers
|
||||
|
||||
#include <sys/types.h>
|
||||
#include <stdint.h>
|
||||
#include "simd-utils.h"
|
||||
#include "sha-hash-4way.h"
|
||||
|
||||
typedef struct _hmac_sha256_4way_context
|
||||
{
|
||||
sha256_4way_context ictx;
|
||||
sha256_4way_context octx;
|
||||
} hmac_sha256_4way_context;
|
||||
|
||||
//void SHA256_Buf( const void *, size_t len, uint8_t digest[32] );
|
||||
void hmac_sha256_4way_init( hmac_sha256_4way_context *, const void *, size_t );
|
||||
void hmac_sha256_4way_update( hmac_sha256_4way_context *, const void *,
|
||||
size_t );
|
||||
void hmac_sha256_4way_close( hmac_sha256_4way_context *, void* );
|
||||
void hmac_sha256_4way_full( void*, const void *, size_t Klen, const void *,
|
||||
size_t len );
|
||||
|
||||
/**
|
||||
* PBKDF2_SHA256(passwd, passwdlen, salt, saltlen, c, buf, dkLen):
|
||||
* Compute PBKDF2(passwd, salt, c, dkLen) using HMAC-SHA256 as the PRF, and
|
||||
* write the output to buf. The value dkLen must be at most 32 * (2^32 - 1).
|
||||
*/
|
||||
void pbkdf2_sha256_4way( uint8_t *, size_t, const uint8_t *, size_t,
|
||||
const uint8_t *, size_t, uint64_t );
|
||||
|
||||
#if defined(__AVX2__)
|
||||
|
||||
typedef struct _hmac_sha256_8way_context
|
||||
{
|
||||
sha256_8way_context ictx;
|
||||
sha256_8way_context octx;
|
||||
} hmac_sha256_8way_context;
|
||||
|
||||
//void SHA256_Buf( const void *, size_t len, uint8_t digest[32] );
|
||||
void hmac_sha256_8way_init( hmac_sha256_8way_context *, const void *, size_t );
|
||||
void hmac_sha256_8way_update( hmac_sha256_8way_context *, const void *,
|
||||
size_t );
|
||||
void hmac_sha256_8way_close( hmac_sha256_8way_context *, void* );
|
||||
void hmac_sha256_8way_full( void*, const void *, size_t Klen, const void *,
|
||||
size_t len );
|
||||
|
||||
void pbkdf2_sha256_8way( uint8_t *, size_t, const uint8_t *, size_t,
|
||||
const uint8_t *, size_t, uint64_t );
|
||||
|
||||
#if defined(__AVX512F__) && defined(__AVX512VL__) && defined(__AVX512DQ__) && defined(__AVX512BW__)
|
||||
|
||||
typedef struct _hmac_sha256_16way_context
|
||||
{
|
||||
sha256_16way_context ictx;
|
||||
sha256_16way_context octx;
|
||||
} hmac_sha256_16way_context;
|
||||
|
||||
//void SHA256_Buf( const void *, size_t len, uint8_t digest[32] );
|
||||
void hmac_sha256_16way_init( hmac_sha256_16way_context *,
|
||||
const void *, size_t );
|
||||
void hmac_sha256_16way_update( hmac_sha256_16way_context *, const void *,
|
||||
size_t );
|
||||
void hmac_sha256_16way_close( hmac_sha256_16way_context *, void* );
|
||||
void hmac_sha256_16way_full( void*, const void *, size_t Klen, const void *,
|
||||
size_t len );
|
||||
|
||||
void pbkdf2_sha256_16way( uint8_t *, size_t, const uint8_t *, size_t,
|
||||
const uint8_t *, size_t, uint64_t );
|
||||
|
||||
|
||||
|
||||
#endif // AVX512
|
||||
#endif // AVX2
|
||||
|
||||
#endif // HMAC_SHA256_4WAY_H__
|
||||
@@ -81,16 +81,17 @@ HMAC_SHA256_Init( HMAC_SHA256_CTX *ctx, const void *_K, size_t Klen )
|
||||
|
||||
/* Inner SHA256 operation is SHA256(K xor [block of 0x36] || data). */
|
||||
SHA256_Init( &ctx->ictx );
|
||||
memset( pad, 0x36, 64 );
|
||||
for ( i = 0; i < Klen; i++ )
|
||||
pad[i] ^= K[i];
|
||||
|
||||
|
||||
for ( i = 0; i < Klen; i++ ) pad[i] = K[i] ^ 0x36;
|
||||
memset( pad + Klen, 0x36, 64 - Klen );
|
||||
SHA256_Update( &ctx->ictx, pad, 64 );
|
||||
|
||||
/* Outer SHA256 operation is SHA256(K xor [block of 0x5c] || hash). */
|
||||
SHA256_Init( &ctx->octx );
|
||||
memset(pad, 0x5c, 64);
|
||||
for ( i = 0; i < Klen; i++ )
|
||||
pad[i] ^= K[i];
|
||||
|
||||
for ( i = 0; i < Klen; i++ ) pad[i] = K[i] ^ 0x5c;
|
||||
memset( pad + Klen, 0x5c, 64 - Klen );
|
||||
SHA256_Update( &ctx->octx, pad, 64 );
|
||||
}
|
||||
|
||||
@@ -161,7 +162,13 @@ PBKDF2_SHA256( const uint8_t *passwd, size_t passwdlen, const uint8_t *salt,
|
||||
HMAC_SHA256_Final( U, &hctx );
|
||||
|
||||
/* ... xor U_j ... */
|
||||
for ( k = 0; k < 32; k++ )
|
||||
// _mm256_xor_si256( *(__m256i*)T, *(__m256i*)U );
|
||||
// _mm_xor_si128( ((__m128i*)T)[0], ((__m128i*)U)[0] );
|
||||
// _mm_xor_si128( ((__m128i*)T)[1], ((__m128i*)U)[1] );
|
||||
|
||||
// for ( k = 0; k < 4; k++ ) T[k] ^= U[k];
|
||||
|
||||
for ( k = 0; k < 32; k++ )
|
||||
T[k] ^= U[k];
|
||||
}
|
||||
|
||||
|
||||
@@ -58,6 +58,7 @@ void sha256_4way_init( sha256_4way_context *sc );
|
||||
void sha256_4way_update( sha256_4way_context *sc, const void *data,
|
||||
size_t len );
|
||||
void sha256_4way_close( sha256_4way_context *sc, void *dst );
|
||||
void sha256_4way_full( void *dst, const void *data, size_t len );
|
||||
|
||||
#endif // SSE2
|
||||
|
||||
@@ -75,6 +76,7 @@ typedef struct {
|
||||
void sha256_8way_init( sha256_8way_context *sc );
|
||||
void sha256_8way_update( sha256_8way_context *sc, const void *data, size_t len );
|
||||
void sha256_8way_close( sha256_8way_context *sc, void *dst );
|
||||
void sha256_8way_full( void *dst, const void *data, size_t len );
|
||||
|
||||
#endif // AVX2
|
||||
|
||||
@@ -92,6 +94,7 @@ typedef struct {
|
||||
void sha256_16way_init( sha256_16way_context *sc );
|
||||
void sha256_16way_update( sha256_16way_context *sc, const void *data, size_t len );
|
||||
void sha256_16way_close( sha256_16way_context *sc, void *dst );
|
||||
void sha256_16way_full( void *dst, const void *data, size_t len );
|
||||
|
||||
#endif // AVX512
|
||||
|
||||
@@ -110,6 +113,7 @@ void sha512_4way_init( sha512_4way_context *sc);
|
||||
void sha512_4way_update( sha512_4way_context *sc, const void *data,
|
||||
size_t len );
|
||||
void sha512_4way_close( sha512_4way_context *sc, void *dst );
|
||||
void sha512_4way_full( void *dst, const void *data, size_t len );
|
||||
|
||||
#endif // AVX2
|
||||
|
||||
@@ -128,6 +132,7 @@ void sha512_8way_init( sha512_8way_context *sc);
|
||||
void sha512_8way_update( sha512_8way_context *sc, const void *data,
|
||||
size_t len );
|
||||
void sha512_8way_close( sha512_8way_context *sc, void *dst );
|
||||
void sha512_8way_full( void *dst, const void *data, size_t len );
|
||||
|
||||
#endif // AVX512
|
||||
|
||||
|
||||
@@ -330,6 +330,14 @@ void sha256_4way_close( sha256_4way_context *sc, void *dst )
|
||||
mm128_block_bswap_32( dst, sc->val );
|
||||
}
|
||||
|
||||
void sha256_4way_full( void *dst, const void *data, size_t len )
|
||||
{
|
||||
sha256_4way_context ctx;
|
||||
sha256_4way_init( &ctx );
|
||||
sha256_4way_update( &ctx, data, len );
|
||||
sha256_4way_close( &ctx, dst );
|
||||
}
|
||||
|
||||
#if defined(__AVX2__)
|
||||
|
||||
// SHA-256 8 way
|
||||
@@ -498,6 +506,10 @@ void sha256_8way_init( sha256_8way_context *sc )
|
||||
*/
|
||||
}
|
||||
|
||||
|
||||
// need to handle odd byte length for yespower.
|
||||
// Assume only last update is odd.
|
||||
|
||||
void sha256_8way_update( sha256_8way_context *sc, const void *data, size_t len )
|
||||
{
|
||||
__m256i *vdata = (__m256i*)data;
|
||||
@@ -564,6 +576,13 @@ void sha256_8way_close( sha256_8way_context *sc, void *dst )
|
||||
mm256_block_bswap_32( dst, sc->val );
|
||||
}
|
||||
|
||||
void sha256_8way_full( void *dst, const void *data, size_t len )
|
||||
{
|
||||
sha256_8way_context ctx;
|
||||
sha256_8way_init( &ctx );
|
||||
sha256_8way_update( &ctx, data, len );
|
||||
sha256_8way_close( &ctx, dst );
|
||||
}
|
||||
|
||||
#if defined(__AVX512F__) && defined(__AVX512VL__) && defined(__AVX512DQ__) && defined(__AVX512BW__)
|
||||
|
||||
@@ -791,6 +810,14 @@ void sha256_16way_close( sha256_16way_context *sc, void *dst )
|
||||
mm512_block_bswap_32( dst, sc->val );
|
||||
}
|
||||
|
||||
void sha256_16way_full( void *dst, const void *data, size_t len )
|
||||
{
|
||||
sha256_16way_context ctx;
|
||||
sha256_16way_init( &ctx );
|
||||
sha256_16way_update( &ctx, data, len );
|
||||
sha256_16way_close( &ctx, dst );
|
||||
}
|
||||
|
||||
#endif // AVX512
|
||||
#endif // __AVX2__
|
||||
#endif // __SSE2__
|
||||
|
||||
@@ -85,7 +85,7 @@ int scanhash_sha256q_8way( struct work *work, uint32_t max_nonce,
|
||||
if ( fulltest( lane_hash, ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n + lane;
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
n += 8;
|
||||
@@ -173,7 +173,7 @@ int scanhash_sha256q_4way( struct work *work, uint32_t max_nonce,
|
||||
if ( fulltest( lane_hash, ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n + lane;
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
n += 4;
|
||||
|
||||
@@ -78,7 +78,7 @@ int scanhash_sha256t_8way( struct work *work, const uint32_t max_nonce,
|
||||
if ( fulltest( lane_hash, ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n + lane;
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
n += 8;
|
||||
@@ -161,7 +161,7 @@ int scanhash_sha256t_4way( struct work *work, const uint32_t max_nonce,
|
||||
if ( fulltest( lane_hash, ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n + lane;
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
n += 4;
|
||||
|
||||
@@ -33,7 +33,7 @@
|
||||
#include <stddef.h>
|
||||
#include <string.h>
|
||||
|
||||
#ifdef __AES__
|
||||
#if defined(__AES__)
|
||||
|
||||
#include "sph_shavite.h"
|
||||
#include "simd-utils.h"
|
||||
|
||||
@@ -35,6 +35,8 @@
|
||||
|
||||
#include "sph_shavite.h"
|
||||
|
||||
#if !defined(__AES__)
|
||||
|
||||
#ifdef __cplusplus
|
||||
extern "C"{
|
||||
#endif
|
||||
@@ -1762,3 +1764,6 @@ sph_shavite512_sw_addbits_and_close(void *cc, unsigned ub, unsigned n, void *dst
|
||||
#ifdef __cplusplus
|
||||
}
|
||||
#endif
|
||||
|
||||
#endif // !AES
|
||||
|
||||
|
||||
@@ -262,15 +262,9 @@ void sph_shavite384_close(void *cc, void *dst);
|
||||
void sph_shavite384_addbits_and_close(
|
||||
void *cc, unsigned ub, unsigned n, void *dst);
|
||||
|
||||
// Always define sw but only define aesni when available
|
||||
// Define fptrs for aesni or sw, not both.
|
||||
void sph_shavite512_sw_init(void *cc);
|
||||
void sph_shavite512_sw(void *cc, const void *data, size_t len);
|
||||
void sph_shavite512_sw_close(void *cc, void *dst);
|
||||
void sph_shavite512_sw_addbits_and_close(
|
||||
void *cc, unsigned ub, unsigned n, void *dst);
|
||||
|
||||
//Don't call these directly from application code, use the macros below.
|
||||
#ifdef __AES__
|
||||
|
||||
void sph_shavite512_aesni_init(void *cc);
|
||||
void sph_shavite512_aesni(void *cc, const void *data, size_t len);
|
||||
void sph_shavite512_aesni_close(void *cc, void *dst);
|
||||
@@ -285,6 +279,13 @@ void sph_shavite512_aesni_addbits_and_close(
|
||||
|
||||
#else
|
||||
|
||||
void sph_shavite512_sw_init(void *cc);
|
||||
void sph_shavite512_sw(void *cc, const void *data, size_t len);
|
||||
void sph_shavite512_sw_close(void *cc, void *dst);
|
||||
void sph_shavite512_sw_addbits_and_close(
|
||||
void *cc, unsigned ub, unsigned n, void *dst);
|
||||
|
||||
|
||||
#define sph_shavite512_init sph_shavite512_sw_init
|
||||
#define sph_shavite512 sph_shavite512_sw
|
||||
#define sph_shavite512_close sph_shavite512_sw_close
|
||||
@@ -293,6 +294,20 @@ void sph_shavite512_aesni_addbits_and_close(
|
||||
|
||||
#endif
|
||||
|
||||
// Use these macros from application code.
|
||||
#define shavite512_context sph_shavite512_context
|
||||
|
||||
#define shavite512_init sph_shavite512_init
|
||||
#define shavite512_update sph_shavite512
|
||||
#define shavite512_close sph_shavite512_close
|
||||
|
||||
#define shavite512_full( cc, dst, data, len ) \
|
||||
do{ \
|
||||
shavite512_init( cc ); \
|
||||
shavite512_update( cc, data, len ); \
|
||||
shavite512_close( cc, dst ); \
|
||||
}while(0)
|
||||
|
||||
#ifdef __cplusplus
|
||||
}
|
||||
#endif
|
||||
|
||||
@@ -24,11 +24,7 @@ void skeinhash_8way( void *state, const void *input )
|
||||
uint32_t vhash32[16*8] __attribute__ ((aligned (128)));
|
||||
sha256_8way_context ctx_sha256;
|
||||
|
||||
skein512_8way_full( &ctx_skein, vhash64, input, 80 );
|
||||
|
||||
// skein512_8way_update( &ctx_skein, input + (64*8), 16 );
|
||||
// skein512_8way_close( &ctx_skein, vhash64 );
|
||||
|
||||
skein512_8way_final16( &ctx_skein, vhash64, input + (64*8) );
|
||||
rintrlv_8x64_8x32( vhash32, vhash64, 512 );
|
||||
|
||||
sha256_8way_init( &ctx_sha256 );
|
||||
@@ -57,8 +53,7 @@ int scanhash_skein_8way( struct work *work, uint32_t max_nonce,
|
||||
*noncev = mm512_intrlv_blend_32(
|
||||
_mm512_set_epi32( n+7, 0, n+6, 0, n+5, 0, n+4, 0,
|
||||
n+3, 0, n+2, 0, n+1, 0, n , 0 ), *noncev );
|
||||
// skein512_8way_init( &skein512_8way_ctx );
|
||||
// skein512_8way_update( &skein512_8way_ctx, vdata, 64 );
|
||||
skein512_8way_prehash64( &skein512_8way_ctx, vdata );
|
||||
do
|
||||
{
|
||||
skeinhash_8way( hash, vdata );
|
||||
@@ -70,7 +65,7 @@ int scanhash_skein_8way( struct work *work, uint32_t max_nonce,
|
||||
if ( valid_hash( lane_hash, ptarget ) )
|
||||
{
|
||||
pdata[19] = bswap_32( n + lane );
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
*noncev = _mm512_add_epi32( *noncev,
|
||||
@@ -85,14 +80,14 @@ int scanhash_skein_8way( struct work *work, uint32_t max_nonce,
|
||||
|
||||
#elif defined (SKEIN_4WAY)
|
||||
|
||||
//static __thread skein512_4way_context skein512_4way_ctx
|
||||
// __attribute__ ((aligned (64)));
|
||||
static __thread skein512_4way_context skein512_4way_ctx
|
||||
__attribute__ ((aligned (64)));
|
||||
|
||||
void skeinhash_4way( void *state, const void *input )
|
||||
{
|
||||
uint64_t vhash64[8*4] __attribute__ ((aligned (128)));
|
||||
skein512_4way_context ctx_skein;
|
||||
// memcpy( &ctx_skein, &skein512_4way_ctx, sizeof( ctx_skein ) );
|
||||
memcpy( &ctx_skein, &skein512_4way_ctx, sizeof( ctx_skein ) );
|
||||
#if defined(__SHA__)
|
||||
uint32_t hash0[16] __attribute__ ((aligned (64)));
|
||||
uint32_t hash1[16] __attribute__ ((aligned (64)));
|
||||
@@ -104,10 +99,7 @@ void skeinhash_4way( void *state, const void *input )
|
||||
sha256_4way_context ctx_sha256;
|
||||
#endif
|
||||
|
||||
skein512_4way_full( &ctx_skein, vhash64, input, 80 );
|
||||
|
||||
// skein512_4way_update( &ctx_skein, input + (64*4), 16 );
|
||||
// skein512_4way_close( &ctx_skein, vhash64 );
|
||||
skein512_4way_final16( &ctx_skein, vhash64, input + (64*4) );
|
||||
|
||||
#if defined(__SHA__)
|
||||
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash64, 512 );
|
||||
@@ -156,8 +148,7 @@ int scanhash_skein_4way( struct work *work, uint32_t max_nonce,
|
||||
const bool bench = opt_benchmark;
|
||||
|
||||
mm256_bswap32_intrlv80_4x64( vdata, pdata );
|
||||
// skein512_4way_init( &skein512_4way_ctx );
|
||||
// skein512_4way_update( &skein512_4way_ctx, vdata, 64 );
|
||||
skein512_4way_prehash64( &skein512_4way_ctx, vdata );
|
||||
|
||||
*noncev = mm256_intrlv_blend_32(
|
||||
_mm256_set_epi32( n+3, 0, n+2, 0, n+1, 0, n, 0 ), *noncev );
|
||||
@@ -171,7 +162,7 @@ int scanhash_skein_4way( struct work *work, uint32_t max_nonce,
|
||||
if ( valid_hash( lane_hash, ptarget ) )
|
||||
{
|
||||
pdata[19] = bswap_32( n + lane );
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
*noncev = _mm256_add_epi32( *noncev,
|
||||
|
||||
@@ -728,6 +728,86 @@ void skein512_8way_full( skein512_8way_context *sc, void *out, const void *data,
|
||||
casti_m512i( out, 7 ) = h7;
|
||||
}
|
||||
|
||||
void
|
||||
skein512_8way_prehash64( skein512_8way_context *sc, const void *data )
|
||||
{
|
||||
__m512i *vdata = (__m512*)data;
|
||||
__m512i *buf = sc->buf;
|
||||
buf[0] = vdata[0];
|
||||
buf[1] = vdata[1];
|
||||
buf[2] = vdata[2];
|
||||
buf[3] = vdata[3];
|
||||
buf[4] = vdata[4];
|
||||
buf[5] = vdata[5];
|
||||
buf[6] = vdata[6];
|
||||
buf[7] = vdata[7];
|
||||
register __m512i h0 = m512_const1_64( 0x4903ADFF749C51CE );
|
||||
register __m512i h1 = m512_const1_64( 0x0D95DE399746DF03 );
|
||||
register __m512i h2 = m512_const1_64( 0x8FD1934127C79BCE );
|
||||
register __m512i h3 = m512_const1_64( 0x9A255629FF352CB1 );
|
||||
register __m512i h4 = m512_const1_64( 0x5DB62599DF6CA7B0 );
|
||||
register __m512i h5 = m512_const1_64( 0xEABE394CA9D5C3F4 );
|
||||
register __m512i h6 = m512_const1_64( 0x991112C71A75B523 );
|
||||
register __m512i h7 = m512_const1_64( 0xAE18A40B660FCC33 );
|
||||
uint64_t bcount = 1;
|
||||
|
||||
UBI_BIG_8WAY( 224, 0 );
|
||||
sc->h0 = h0;
|
||||
sc->h1 = h1;
|
||||
sc->h2 = h2;
|
||||
sc->h3 = h3;
|
||||
sc->h4 = h4;
|
||||
sc->h5 = h5;
|
||||
sc->h6 = h6;
|
||||
sc->h7 = h7;
|
||||
}
|
||||
|
||||
void
|
||||
skein512_8way_final16( skein512_8way_context *sc, void *output,
|
||||
const void *data )
|
||||
{
|
||||
__m512i *in = (__m512i*)data;
|
||||
__m512i *buf = sc->buf;
|
||||
__m512i *out = (__m512i*)output;
|
||||
register __m512i h0 = sc->h0;
|
||||
register __m512i h1 = sc->h1;
|
||||
register __m512i h2 = sc->h2;
|
||||
register __m512i h3 = sc->h3;
|
||||
register __m512i h4 = sc->h4;
|
||||
register __m512i h5 = sc->h5;
|
||||
register __m512i h6 = sc->h6;
|
||||
register __m512i h7 = sc->h7;
|
||||
|
||||
const __m512i zero = m512_zero;
|
||||
buf[0] = in[0];
|
||||
buf[1] = in[1];
|
||||
buf[2] = zero;
|
||||
buf[3] = zero;
|
||||
buf[4] = zero;
|
||||
buf[5] = zero;
|
||||
buf[6] = zero;
|
||||
buf[7] = zero;
|
||||
|
||||
uint64_t bcount = 1;
|
||||
UBI_BIG_8WAY( 352, 16 );
|
||||
|
||||
buf[0] = zero;
|
||||
buf[1] = zero;
|
||||
|
||||
bcount = 0;
|
||||
UBI_BIG_8WAY( 510, 8 );
|
||||
|
||||
out[0] = h0;
|
||||
out[1] = h1;
|
||||
out[2] = h2;
|
||||
out[3] = h3;
|
||||
out[4] = h4;
|
||||
out[5] = h5;
|
||||
out[6] = h6;
|
||||
out[7] = h7;
|
||||
}
|
||||
|
||||
|
||||
void
|
||||
skein256_8way_update(void *cc, const void *data, size_t len)
|
||||
{
|
||||
@@ -942,6 +1022,84 @@ skein512_4way_full( skein512_4way_context *sc, void *out, const void *data,
|
||||
casti_m256i( out, 7 ) = h7;
|
||||
}
|
||||
|
||||
void
|
||||
skein512_4way_prehash64( skein512_4way_context *sc, const void *data )
|
||||
{
|
||||
__m256i *vdata = (__m256i*)data;
|
||||
__m256i *buf = sc->buf;
|
||||
buf[0] = vdata[0];
|
||||
buf[1] = vdata[1];
|
||||
buf[2] = vdata[2];
|
||||
buf[3] = vdata[3];
|
||||
buf[4] = vdata[4];
|
||||
buf[5] = vdata[5];
|
||||
buf[6] = vdata[6];
|
||||
buf[7] = vdata[7];
|
||||
register __m256i h0 = m256_const1_64( 0x4903ADFF749C51CE );
|
||||
register __m256i h1 = m256_const1_64( 0x0D95DE399746DF03 );
|
||||
register __m256i h2 = m256_const1_64( 0x8FD1934127C79BCE );
|
||||
register __m256i h3 = m256_const1_64( 0x9A255629FF352CB1 );
|
||||
register __m256i h4 = m256_const1_64( 0x5DB62599DF6CA7B0 );
|
||||
register __m256i h5 = m256_const1_64( 0xEABE394CA9D5C3F4 );
|
||||
register __m256i h6 = m256_const1_64( 0x991112C71A75B523 );
|
||||
register __m256i h7 = m256_const1_64( 0xAE18A40B660FCC33 );
|
||||
uint64_t bcount = 1;
|
||||
|
||||
UBI_BIG_4WAY( 224, 0 );
|
||||
sc->h0 = h0;
|
||||
sc->h1 = h1;
|
||||
sc->h2 = h2;
|
||||
sc->h3 = h3;
|
||||
sc->h4 = h4;
|
||||
sc->h5 = h5;
|
||||
sc->h6 = h6;
|
||||
sc->h7 = h7;
|
||||
}
|
||||
|
||||
void
|
||||
skein512_4way_final16( skein512_4way_context *sc, void *out, const void *data )
|
||||
{
|
||||
__m256i *vdata = (__m256i*)data;
|
||||
__m256i *buf = sc->buf;
|
||||
register __m256i h0 = sc->h0;
|
||||
register __m256i h1 = sc->h1;
|
||||
register __m256i h2 = sc->h2;
|
||||
register __m256i h3 = sc->h3;
|
||||
register __m256i h4 = sc->h4;
|
||||
register __m256i h5 = sc->h5;
|
||||
register __m256i h6 = sc->h6;
|
||||
register __m256i h7 = sc->h7;
|
||||
|
||||
const __m256i zero = m256_zero;
|
||||
buf[0] = vdata[0];
|
||||
buf[1] = vdata[1];
|
||||
buf[2] = zero;
|
||||
buf[3] = zero;
|
||||
buf[4] = zero;
|
||||
buf[5] = zero;
|
||||
buf[6] = zero;
|
||||
buf[7] = zero;
|
||||
|
||||
uint64_t bcount = 1;
|
||||
UBI_BIG_4WAY( 352, 16 );
|
||||
|
||||
buf[0] = zero;
|
||||
buf[1] = zero;
|
||||
|
||||
bcount = 0;
|
||||
UBI_BIG_4WAY( 510, 8 );
|
||||
|
||||
casti_m256i( out, 0 ) = h0;
|
||||
casti_m256i( out, 1 ) = h1;
|
||||
casti_m256i( out, 2 ) = h2;
|
||||
casti_m256i( out, 3 ) = h3;
|
||||
casti_m256i( out, 4 ) = h4;
|
||||
casti_m256i( out, 5 ) = h5;
|
||||
casti_m256i( out, 6 ) = h6;
|
||||
casti_m256i( out, 7 ) = h7;
|
||||
}
|
||||
|
||||
// Broken for 80 bytes, use prehash.
|
||||
void
|
||||
skein256_4way_update(void *cc, const void *data, size_t len)
|
||||
{
|
||||
|
||||
@@ -69,6 +69,10 @@ void skein512_8way_init( skein512_8way_context *sc );
|
||||
void skein512_8way_update( void *cc, const void *data, size_t len );
|
||||
void skein512_8way_close( void *cc, void *dst );
|
||||
|
||||
void skein512_8way_prehash64( skein512_8way_context *sc, const void *data );
|
||||
void skein512_8way_final16( skein512_8way_context *sc, void *out,
|
||||
const void *data );
|
||||
|
||||
void skein256_8way_init( skein256_8way_context *sc );
|
||||
void skein256_8way_update( void *cc, const void *data, size_t len );
|
||||
void skein256_8way_close( void *cc, void *dst );
|
||||
@@ -96,6 +100,10 @@ void skein256_4way_init( skein256_4way_context *sc );
|
||||
void skein256_4way_update( void *cc, const void *data, size_t len );
|
||||
void skein256_4way_close( void *cc, void *dst );
|
||||
|
||||
void skein512_4way_prehash64( skein512_4way_context *sc, const void *data );
|
||||
void skein512_4way_final16( skein512_4way_context *sc, void *out,
|
||||
const void *data );
|
||||
|
||||
#ifdef __cplusplus
|
||||
}
|
||||
#endif
|
||||
|
||||
@@ -5,20 +5,16 @@
|
||||
|
||||
#if defined(SKEIN_8WAY)
|
||||
|
||||
// static __thread skein512_8way_context skein512_8way_ctx
|
||||
// __attribute__ ((aligned (64)));
|
||||
static __thread skein512_8way_context skein512_8way_ctx
|
||||
__attribute__ ((aligned (64)));
|
||||
|
||||
void skein2hash_8way( void *output, const void *input )
|
||||
{
|
||||
uint64_t hash[16*8] __attribute__ ((aligned (128)));
|
||||
skein512_8way_context ctx;
|
||||
// memcpy( &ctx, &skein512_8way_ctx, sizeof( ctx ) );
|
||||
|
||||
skein512_8way_full( &ctx, hash, input, 80 );
|
||||
|
||||
// skein512_8way_update( &ctx, input + (64*8), 16 );
|
||||
// skein512_8way_close( &ctx, hash );
|
||||
memcpy( &ctx, &skein512_8way_ctx, sizeof( ctx ) );
|
||||
|
||||
skein512_8way_final16( &ctx, hash, input + (64*8) );
|
||||
skein512_8way_full( &ctx, output, hash, 64 );
|
||||
}
|
||||
|
||||
@@ -38,16 +34,17 @@ int scanhash_skein2_8way( struct work *work, uint32_t max_nonce,
|
||||
__m512i *noncev = (__m512i*)vdata + 9;
|
||||
const int thr_id = mythr->id;
|
||||
const bool bench = opt_benchmark;
|
||||
skein512_8way_context ctx;
|
||||
|
||||
mm512_bswap32_intrlv80_8x64( vdata, pdata );
|
||||
*noncev = mm512_intrlv_blend_32(
|
||||
_mm512_set_epi32( n+7, 0, n+6, 0, n+5, 0, n+4, 0,
|
||||
n+3, 0, n+2, 0, n+1, 0, n , 0 ), *noncev );
|
||||
// skein512_8way_init( &skein512_8way_ctx );
|
||||
// skein512_8way_update( &skein512_8way_ctx, vdata, 64 );
|
||||
skein512_8way_prehash64( &ctx, vdata );
|
||||
do
|
||||
{
|
||||
skein2hash_8way( hash, vdata );
|
||||
skein512_8way_final16( &ctx, hash, vdata + (16*8) );
|
||||
skein512_8way_full( &ctx, hash, hash, 64 );
|
||||
|
||||
for ( int lane = 0; lane < 8; lane++ )
|
||||
if ( unlikely( hashq3[ lane ] <= targq3 && !bench ) )
|
||||
@@ -56,7 +53,7 @@ int scanhash_skein2_8way( struct work *work, uint32_t max_nonce,
|
||||
if ( valid_hash( lane_hash, ptarget ) && !bench )
|
||||
{
|
||||
pdata[19] = bswap_32( n + lane );
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
*noncev = _mm512_add_epi32( *noncev,
|
||||
@@ -71,19 +68,16 @@ int scanhash_skein2_8way( struct work *work, uint32_t max_nonce,
|
||||
|
||||
#elif defined(SKEIN_4WAY)
|
||||
|
||||
//static __thread skein512_4way_context skein512_4way_ctx
|
||||
// __attribute__ ((aligned (64)));
|
||||
static __thread skein512_4way_context skein512_4way_ctx
|
||||
__attribute__ ((aligned (64)));
|
||||
|
||||
void skein2hash_4way( void *output, const void *input )
|
||||
{
|
||||
skein512_4way_context ctx;
|
||||
// memcpy( &ctx, &skein512_4way_ctx, sizeof( ctx ) );
|
||||
memcpy( &ctx, &skein512_4way_ctx, sizeof( ctx ) );
|
||||
uint64_t hash[16*4] __attribute__ ((aligned (64)));
|
||||
|
||||
// skein512_4way_update( &ctx, input + (64*4), 16 );
|
||||
// skein512_4way_close( &ctx, hash );
|
||||
|
||||
skein512_4way_full( &ctx, hash, input, 80 );
|
||||
skein512_4way_final16( &ctx, hash, input + (64*4) );
|
||||
skein512_4way_full( &ctx, output, hash, 64 );
|
||||
}
|
||||
|
||||
@@ -103,15 +97,16 @@ int scanhash_skein2_4way( struct work *work, uint32_t max_nonce,
|
||||
__m256i *noncev = (__m256i*)vdata + 9;
|
||||
const int thr_id = mythr->id;
|
||||
const bool bench = opt_benchmark;
|
||||
skein512_4way_context ctx;
|
||||
|
||||
mm256_bswap32_intrlv80_4x64( vdata, pdata );
|
||||
// skein512_4way_init( &skein512_4way_ctx );
|
||||
// skein512_4way_update( &skein512_4way_ctx, vdata, 64 );
|
||||
skein512_4way_prehash64( &ctx, vdata );
|
||||
*noncev = mm256_intrlv_blend_32(
|
||||
_mm256_set_epi32( n+3, 0, n+2, 0, n+1, 0, n, 0 ), *noncev );
|
||||
do
|
||||
{
|
||||
skein2hash_4way( hash, vdata );
|
||||
skein512_4way_final16( &ctx, hash, vdata + (16*4) );
|
||||
skein512_4way_full( &ctx, hash, hash, 64 );
|
||||
|
||||
for ( int lane = 0; lane < 4; lane++ )
|
||||
if ( hash_q3[ lane ] <= targ_q3 )
|
||||
@@ -120,7 +115,7 @@ int scanhash_skein2_4way( struct work *work, uint32_t max_nonce,
|
||||
if ( valid_hash( lane_hash, ptarget ) && !bench )
|
||||
{
|
||||
pdata[19] = bswap_32( n + lane );
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
*noncev = _mm256_add_epi32( *noncev,
|
||||
|
||||
@@ -120,6 +120,13 @@ void sph_whirlpool(void *cc, const void *data, size_t len);
|
||||
*/
|
||||
void sph_whirlpool_close(void *cc, void *dst);
|
||||
|
||||
#define sph_whirlpool512_full( cc, dst, data, len ) \
|
||||
do{ \
|
||||
sph_whirlpool_init( cc ); \
|
||||
sph_whirlpool( cc, data, len ); \
|
||||
sph_whirlpool_close( cc, dst ); \
|
||||
}while(0)
|
||||
|
||||
/**
|
||||
* WHIRLPOOL-0 uses the same structure than plain WHIRLPOOL.
|
||||
*/
|
||||
|
||||
@@ -279,7 +279,7 @@ int scanhash_c11_8way( struct work *work, uint32_t max_nonce,
|
||||
&& fulltest( hash+(i<<3), ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n+i;
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
n += 8;
|
||||
} while ( ( n < max_nonce ) && !work_restart[thr_id].restart );
|
||||
@@ -459,7 +459,7 @@ int scanhash_c11_4way( struct work *work, uint32_t max_nonce,
|
||||
&& fulltest( hash+(i<<3), ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n+i;
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
n += 4;
|
||||
} while ( ( n < max_nonce ) && !work_restart[thr_id].restart );
|
||||
|
||||
@@ -221,7 +221,7 @@ int scanhash_timetravel_4way( struct work *work, uint32_t max_nonce,
|
||||
&& !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n+i;
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
n += 4;
|
||||
} while ( ( n < max_nonce ) && !(*restart) );
|
||||
|
||||
@@ -256,7 +256,7 @@ int scanhash_timetravel10_4way( struct work *work,
|
||||
&& !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n+i;
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
n += 4;
|
||||
} while ( ( n < max_nonce ) && !(*restart) );
|
||||
|
||||
@@ -128,7 +128,7 @@ int scanhash_tribus_8way( struct work *work, uint32_t max_nonce,
|
||||
if ( fulltest( hash+(i<<3), ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n+i;
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
n += 8;
|
||||
} while ( ( n < max_nonce-8 ) && !work_restart[thr_id].restart);
|
||||
@@ -213,7 +213,7 @@ int scanhash_tribus_4way( struct work *work, uint32_t max_nonce,
|
||||
if ( fulltest( hash+(i<<3), ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n+i;
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
n += 4;
|
||||
} while ( ( n < max_nonce-4 ) && !work_restart[thr_id].restart);
|
||||
|
||||
@@ -279,7 +279,7 @@ int scanhash_x11_8way( struct work *work, uint32_t max_nonce,
|
||||
&& fulltest( hash+(i<<3), ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n+i;
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
n += 8;
|
||||
} while ( ( n < last_nonce ) && !work_restart[thr_id].restart );
|
||||
@@ -469,7 +469,7 @@ int scanhash_x11_4way( struct work *work, uint32_t max_nonce,
|
||||
&& fulltest( hash+(i<<3), ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n+i;
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
n += 4;
|
||||
} while ( ( n < max_nonce ) && !work_restart[thr_id].restart );
|
||||
|
||||
@@ -269,7 +269,7 @@ int scanhash_x11evo_4way( struct work* work, uint32_t max_nonce,
|
||||
&& fulltest( hash+(i<<3), ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n+i;
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
n += 4;
|
||||
} while ( ( n < max_nonce ) && !work_restart[thr_id].restart );
|
||||
|
||||
@@ -312,7 +312,7 @@ int scanhash_x11gost_8way( struct work *work, uint32_t max_nonce,
|
||||
&& fulltest( hash+(i<<3), ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n+i;
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
n += 8;
|
||||
} while ( ( n < max_nonce ) && !work_restart[thr_id].restart );
|
||||
@@ -498,7 +498,7 @@ int scanhash_x11gost_4way( struct work *work, uint32_t max_nonce,
|
||||
&& fulltest( hash+(i<<3), ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n+i;
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
n += 4;
|
||||
} while ( ( n < max_nonce ) && !work_restart[thr_id].restart );
|
||||
|
||||
@@ -170,6 +170,9 @@ void x12_8way_hash( void *state, const void *input )
|
||||
dintrlv_4x128_512( hash0, hash1, hash2, hash3, vhashA );
|
||||
dintrlv_4x128_512( hash4, hash5, hash6, hash7, vhashB );
|
||||
|
||||
update_final_echo( &ctx.echo, (BitSequence *)hash0,
|
||||
(const BitSequence *) hash0, 512 );
|
||||
memcpy( &ctx.echo, &x12_8way_ctx.echo, sizeof(hashState_echo) );
|
||||
update_final_echo( &ctx.echo, (BitSequence *)hash1,
|
||||
(const BitSequence *) hash1, 512 );
|
||||
memcpy( &ctx.echo, &x12_8way_ctx.echo, sizeof(hashState_echo) );
|
||||
@@ -263,7 +266,7 @@ int scanhash_x12_8way( struct work *work, uint32_t max_nonce,
|
||||
if ( fulltest( lane_hash, ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n + lane;
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
n += 8;
|
||||
@@ -363,6 +366,18 @@ void x12_4way_hash( void *state, const void *input )
|
||||
simd_2way_update_close( &ctx.simd, vhash, vhash, 512 );
|
||||
dintrlv_2x128( hash2, hash3, vhash, 512 );
|
||||
|
||||
update_final_echo( &ctx.echo, (BitSequence *)hash0,
|
||||
(const BitSequence *) hash0, 512 );
|
||||
memcpy( &ctx.echo, &x12_4way_ctx.echo, sizeof(hashState_echo) );
|
||||
update_final_echo( &ctx.echo, (BitSequence *)hash1,
|
||||
(const BitSequence *) hash1, 512 );
|
||||
memcpy( &ctx.echo, &x12_4way_ctx.echo, sizeof(hashState_echo) );
|
||||
update_final_echo( &ctx.echo, (BitSequence *)hash2,
|
||||
(const BitSequence *) hash2, 512 );
|
||||
memcpy( &ctx.echo, &x12_4way_ctx.echo, sizeof(hashState_echo) );
|
||||
update_final_echo( &ctx.echo, (BitSequence *)hash3,
|
||||
(const BitSequence *) hash3, 512 );
|
||||
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash0, (char*)hash0, 512 );
|
||||
memcpy( &ctx.groestl, &x12_4way_ctx.groestl, sizeof(hashState_groestl) );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash1, (char*)hash1, 512 );
|
||||
@@ -431,7 +446,7 @@ int scanhash_x12_4way( struct work *work, uint32_t max_nonce,
|
||||
if ( fulltest( hash+(i<<3), ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n+i;
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
n += 4;
|
||||
} while ( ( n < max_nonce ) && !work_restart[thr_id].restart );
|
||||
|
||||
@@ -208,7 +208,7 @@ int scanhash_phi1612_8way( struct work *work, uint32_t max_nonce,
|
||||
if ( fulltest( hash+(i<<3), ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n+i;
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
n += 8;
|
||||
} while ( ( n < max_nonce-8 ) && !work_restart[thr_id].restart );
|
||||
@@ -251,8 +251,12 @@ void phi1612_4way_hash( void *state, const void *input )
|
||||
memcpy( &ctx, &phi1612_4way_ctx, sizeof(phi1612_4way_ctx) );
|
||||
|
||||
// Skein parallel 4way
|
||||
skein512_4way_update( &ctx.skein, input, 80 );
|
||||
skein512_4way_close( &ctx.skein, vhash );
|
||||
|
||||
// skein 4way is broken for 80 bytes
|
||||
// skein512_4way_update( &ctx.skein, input, 80 );
|
||||
// skein512_4way_close( &ctx.skein, vhash );
|
||||
skein512_4way_prehash64( &ctx.skein, input );
|
||||
skein512_4way_final16( &ctx.skein, vhash, input + (64*4) );
|
||||
|
||||
// JH
|
||||
jh512_4way_update( &ctx.jh, vhash, 64 );
|
||||
@@ -344,7 +348,7 @@ int scanhash_phi1612_4way( struct work *work, uint32_t max_nonce,
|
||||
if ( fulltest( hash+(i<<3), ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n+i;
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
n += 4;
|
||||
} while ( ( n < max_nonce ) && !work_restart[thr_id].restart );
|
||||
|
||||
@@ -35,8 +35,7 @@ void skunk_8way_hash( void *output, const void *input )
|
||||
skunk_8way_ctx_holder ctx __attribute__ ((aligned (64)));
|
||||
memcpy( &ctx, &skunk_8way_ctx, sizeof(skunk_8way_ctx) );
|
||||
|
||||
skein512_8way_update( &ctx.skein, input, 80 );
|
||||
skein512_8way_close( &ctx.skein, vhash );
|
||||
skein512_8way_final16( &ctx.skein, vhash, input );
|
||||
dintrlv_8x64( hash0, hash1, hash2, hash3, hash4, hash5, hash6,
|
||||
hash7, vhash, 512 );
|
||||
|
||||
@@ -104,35 +103,35 @@ int scanhash_skunk_8way( struct work *work, uint32_t max_nonce,
|
||||
uint32_t *pdata = work->data;
|
||||
uint32_t *ptarget = work->target;
|
||||
const uint32_t first_nonce = pdata[19];
|
||||
const uint32_t last_nonce = max_nonce - 8;
|
||||
uint32_t n = first_nonce;
|
||||
__m512i *noncev = (__m512i*)vdata + 9; // aligned
|
||||
const uint32_t Htarg = ptarget[7];
|
||||
int thr_id = mythr->id;
|
||||
__m512i *noncev = (__m512i*)vdata + 9;
|
||||
const int thr_id = mythr->id;
|
||||
volatile uint8_t *restart = &(work_restart[thr_id].restart);
|
||||
const bool bench = opt_benchmark;
|
||||
|
||||
if ( opt_benchmark )
|
||||
((uint32_t*)ptarget)[7] = 0x0cff;
|
||||
if ( bench ) ptarget[7] = 0x0fff;
|
||||
|
||||
mm512_bswap32_intrlv80_8x64( vdata, pdata );
|
||||
skein512_8way_prehash64( &skunk_8way_ctx.skein, vdata );
|
||||
*noncev = mm512_intrlv_blend_32(
|
||||
_mm512_set_epi32( n+7, 0, n+6, 0, n+5, 0, n+4, 0,
|
||||
n+3, 0, n+2, 0, n+1, 0, n , 0 ), *noncev );
|
||||
do
|
||||
{
|
||||
*noncev = mm512_intrlv_blend_32( mm512_bswap_32(
|
||||
_mm512_set_epi32( n+7, 0, n+6, 0, n+5, 0, n+4, 0,
|
||||
n+3, 0, n+2, 0, n+1, 0, n , 0 ) ), *noncev );
|
||||
|
||||
skunk_8way_hash( hash, vdata );
|
||||
pdata[19] = n;
|
||||
|
||||
for ( int i = 0; i < 8; i++ )
|
||||
if ( unlikely( (hash+(i<<3))[7] <= Htarg ) )
|
||||
if ( likely( fulltest( hash+(i<<3), ptarget ) && !opt_benchmark ) )
|
||||
if ( unlikely( valid_hash( hash+(i<<3), ptarget ) && !bench ) )
|
||||
{
|
||||
pdata[19] = n+i;
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
pdata[19] = bswap_32( n+i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
*noncev = _mm512_add_epi32( *noncev,
|
||||
m512_const1_64( 0x0000000800000000 ) );
|
||||
n +=8;
|
||||
} while ( likely( ( n < max_nonce-8 ) && !(*restart) ) );
|
||||
|
||||
} while ( likely( ( n < last_nonce ) && !( *restart ) ) );
|
||||
pdata[19] = n;
|
||||
*hashes_done = n - first_nonce;
|
||||
return 0;
|
||||
}
|
||||
@@ -159,17 +158,16 @@ static __thread skunk_4way_ctx_holder skunk_4way_ctx;
|
||||
|
||||
void skunk_4way_hash( void *output, const void *input )
|
||||
{
|
||||
uint64_t vhash[8*4] __attribute__ ((aligned (128)));
|
||||
uint64_t hash0[8] __attribute__ ((aligned (64)));
|
||||
uint64_t hash1[8] __attribute__ ((aligned (64)));
|
||||
uint64_t hash2[8] __attribute__ ((aligned (64)));
|
||||
uint64_t hash3[8] __attribute__ ((aligned (64)));
|
||||
uint64_t vhash[8*4] __attribute__ ((aligned (64)));
|
||||
|
||||
skunk_4way_ctx_holder ctx __attribute__ ((aligned (64)));
|
||||
memcpy( &ctx, &skunk_4way_ctx, sizeof(skunk_4way_ctx) );
|
||||
|
||||
skein512_4way_update( &ctx.skein, input, 80 );
|
||||
skein512_4way_close( &ctx.skein, vhash );
|
||||
skein512_4way_final16( &ctx.skein, vhash, input + (64*4) );
|
||||
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
|
||||
|
||||
cubehashUpdateDigest( &ctx.cube, (byte*) hash0, (const byte*)hash0, 64 );
|
||||
@@ -213,40 +211,40 @@ void skunk_4way_hash( void *output, const void *input )
|
||||
int scanhash_skunk_4way( struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr )
|
||||
{
|
||||
uint32_t hash[4*8] __attribute__ ((aligned (64)));
|
||||
uint32_t vdata[24*4] __attribute__ ((aligned (64)));
|
||||
uint32_t hash[4*8] __attribute__ ((aligned (128)));
|
||||
uint32_t vdata[20*4] __attribute__ ((aligned (64)));
|
||||
uint32_t *pdata = work->data;
|
||||
uint32_t *ptarget = work->target;
|
||||
const uint32_t first_nonce = pdata[19];
|
||||
const uint32_t last_nonce = max_nonce - 4;
|
||||
uint32_t n = first_nonce;
|
||||
__m256i *noncev = (__m256i*)vdata + 9; // aligned
|
||||
const uint32_t Htarg = ptarget[7];
|
||||
int thr_id = mythr->id; // thr_id arg is deprecated
|
||||
volatile uint8_t *restart = &(work_restart[thr_id].restart);
|
||||
__m256i *noncev = (__m256i*)vdata + 9;
|
||||
const int thr_id = mythr->id;
|
||||
volatile uint8_t *restart = &( work_restart[ thr_id ].restart );
|
||||
const bool bench = opt_benchmark;
|
||||
|
||||
if ( opt_benchmark )
|
||||
((uint32_t*)ptarget)[7] = 0x0cff;
|
||||
if ( bench ) ptarget[7] = 0x0fff;
|
||||
|
||||
mm256_bswap32_intrlv80_4x64( vdata, pdata );
|
||||
skein512_4way_prehash64( &skunk_4way_ctx.skein, vdata );
|
||||
*noncev = mm256_intrlv_blend_32(
|
||||
_mm256_set_epi32( n+3, 0, n+2, 0, n+1, 0, n, 0 ), *noncev );
|
||||
do
|
||||
{
|
||||
*noncev = mm256_intrlv_blend_32( mm256_bswap_32(
|
||||
_mm256_set_epi32( n+3, 0, n+2, 0, n+1, 0, n, 0 ) ), *noncev );
|
||||
|
||||
skunk_4way_hash( hash, vdata );
|
||||
pdata[19] = n;
|
||||
|
||||
for ( int i = 0; i < 4; i++ )
|
||||
if ( (hash+(i<<3))[7] <= Htarg )
|
||||
if ( fulltest( hash+(i<<3), ptarget ) && !opt_benchmark )
|
||||
if ( unlikely( valid_hash( hash+(i<<3), ptarget ) && !bench ) )
|
||||
{
|
||||
pdata[19] = n+i;
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
pdata[19] = bswap_32( n + i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
*noncev = _mm256_add_epi32( *noncev,
|
||||
m256_const1_64( 0x0000000400000000 ) );
|
||||
n +=4;
|
||||
} while ( ( n < max_nonce ) && !(*restart) );
|
||||
|
||||
*hashes_done = n - first_nonce + 1;
|
||||
} while ( likely( ( n < last_nonce ) && !( *restart ) ) );
|
||||
pdata[19] = n;
|
||||
*hashes_done = n - first_nonce;
|
||||
return 0;
|
||||
}
|
||||
|
||||
|
||||
@@ -319,7 +319,7 @@ int scanhash_x13_8way( struct work *work, uint32_t max_nonce,
|
||||
&& fulltest( hash+(i<<3), ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n+i;
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
n += 8;
|
||||
} while ( ( n < last_nonce ) && !work_restart[thr_id].restart );
|
||||
@@ -531,7 +531,7 @@ int scanhash_x13_4way( struct work *work, uint32_t max_nonce,
|
||||
if ( fulltest( hash+(i<<3), ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n+i;
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
n += 4;
|
||||
} while ( ( n < max_nonce ) && !work_restart[thr_id].restart );
|
||||
|
||||
@@ -321,7 +321,7 @@ int scanhash_x13bcd_8way( struct work *work, uint32_t max_nonce,
|
||||
if ( fulltest( hash+(i<<3), ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n+i;
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
n += 8;
|
||||
} while ( ( n < last_nonce ) && !work_restart[thr_id].restart );
|
||||
@@ -541,7 +541,7 @@ int scanhash_x13bcd_4way( struct work *work, uint32_t max_nonce,
|
||||
if ( fulltest( hash+(i<<3), ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n+i;
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
n += 4;
|
||||
} while ( ( n < last_nonce ) && !work_restart[thr_id].restart );
|
||||
|
||||
@@ -246,7 +246,7 @@ int scanhash_x13sm3_4way( struct work *work, uint32_t max_nonce,
|
||||
if ( fulltest( hash+(i<<3), ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n+i;
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
n += 4;
|
||||
} while ( ( n < max_nonce ) && !work_restart[thr_id].restart );
|
||||
|
||||
@@ -129,7 +129,7 @@ int scanhash_polytimos_4way( struct work *work, uint32_t max_nonce,
|
||||
if( fulltest( hash+(i<<3), ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n+i;
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
n += 4;
|
||||
|
||||
|
||||
@@ -108,7 +108,7 @@ int scanhash_veltor_4way( struct work *work, uint32_t max_nonce,
|
||||
if ( (hash+(i<<3))[7] <= Htarg && fulltest( hash+(i<<3), ptarget ) )
|
||||
{
|
||||
pdata[19] = n+i;
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
n += 4;
|
||||
} while ( ( n < max_nonce ) && !(*restart) );
|
||||
|
||||
@@ -324,7 +324,7 @@ int scanhash_x14_8way( struct work *work, uint32_t max_nonce,
|
||||
if ( fulltest( lane_hash, ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n + lane;
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
n += 8;
|
||||
@@ -534,7 +534,7 @@ int scanhash_x14_4way( struct work *work, uint32_t max_nonce,
|
||||
if ( fulltest( lane_hash, ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n + lane;
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
n += 4;
|
||||
|
||||
@@ -364,7 +364,7 @@ int scanhash_x15_8way( struct work *work, uint32_t max_nonce,
|
||||
if ( fulltest( hash+(i<<3), ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n+i;
|
||||
submit_lane_solution( work, hash, mythr, i );
|
||||
submit_solution( work, hash, mythr );
|
||||
}
|
||||
n += 8;
|
||||
} while ( ( n < last_nonce ) && !work_restart[thr_id].restart );
|
||||
@@ -592,7 +592,7 @@ int scanhash_x15_4way( struct work *work, uint32_t max_nonce,
|
||||
if ( fulltest( hash+(i<<3), ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n+i;
|
||||
submit_lane_solution( work, hash, mythr, i );
|
||||
submit_solution( work, hash, mythr );
|
||||
}
|
||||
n += 4;
|
||||
} while ( ( n < last_nonce ) && !work_restart[thr_id].restart );
|
||||
|
||||
@@ -47,6 +47,7 @@ static void hex_getAlgoString(const uint32_t* prevblock, char *output)
|
||||
*sptr = '\0';
|
||||
}
|
||||
|
||||
/*
|
||||
union _hex_context_overlay
|
||||
{
|
||||
#if defined(__AES__)
|
||||
@@ -63,7 +64,7 @@ union _hex_context_overlay
|
||||
sph_keccak512_context keccak;
|
||||
hashState_luffa luffa;
|
||||
cubehashParam cube;
|
||||
sph_shavite512_context shavite;
|
||||
shavite512_context shavite;
|
||||
hashState_sd simd;
|
||||
sph_hamsi512_context hamsi;
|
||||
sph_fugue512_context fugue;
|
||||
@@ -72,13 +73,14 @@ union _hex_context_overlay
|
||||
SHA512_CTX sha512;
|
||||
};
|
||||
typedef union _hex_context_overlay hex_context_overlay;
|
||||
*/
|
||||
|
||||
static __thread hex_context_overlay hex_ctx;
|
||||
static __thread x16r_context_overlay hex_ctx;
|
||||
|
||||
void hex_hash( void* output, const void* input )
|
||||
int hex_hash( void* output, const void* input, int thrid )
|
||||
{
|
||||
uint32_t _ALIGN(128) hash[16];
|
||||
hex_context_overlay ctx;
|
||||
x16r_context_overlay ctx;
|
||||
memcpy( &ctx, &hex_ctx, sizeof(ctx) );
|
||||
void *in = (void*) input;
|
||||
int size = 80;
|
||||
@@ -157,9 +159,7 @@ void hex_hash( void* output, const void* input )
|
||||
}
|
||||
break;
|
||||
case SHAVITE:
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, in, size );
|
||||
sph_shavite512_close( &ctx.shavite, hash );
|
||||
shavite512_full( &ctx.shavite, hash, in, size );
|
||||
break;
|
||||
case SIMD:
|
||||
init_sd( &ctx.simd, 512 );
|
||||
@@ -187,9 +187,7 @@ void hex_hash( void* output, const void* input )
|
||||
sph_hamsi512_close( &ctx.hamsi, hash );
|
||||
break;
|
||||
case FUGUE:
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, in, size );
|
||||
sph_fugue512_close( &ctx.fugue, hash );
|
||||
sph_fugue512_full( &ctx.fugue, hash, in, size );
|
||||
break;
|
||||
case SHABAL:
|
||||
if ( i == 0 )
|
||||
@@ -203,13 +201,12 @@ void hex_hash( void* output, const void* input )
|
||||
break;
|
||||
case WHIRLPOOL:
|
||||
if ( i == 0 )
|
||||
sph_whirlpool( &ctx.whirlpool, in+64, 16 );
|
||||
else
|
||||
{
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, in, size );
|
||||
sph_whirlpool( &ctx.whirlpool, in+64, 16 );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash );
|
||||
}
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash );
|
||||
else
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash, in, size );
|
||||
break;
|
||||
case SHA_512:
|
||||
SHA512_Init( &ctx.sha512 );
|
||||
@@ -217,11 +214,15 @@ void hex_hash( void* output, const void* input )
|
||||
SHA512_Final( (unsigned char*) hash, &ctx.sha512 );
|
||||
break;
|
||||
}
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
algo = (uint8_t)hash[0] % X16R_HASH_FUNC_COUNT;
|
||||
in = (void*) hash;
|
||||
size = 64;
|
||||
}
|
||||
memcpy(output, hash, 32);
|
||||
return 1;
|
||||
}
|
||||
|
||||
int scanhash_hex( struct work *work, uint32_t max_nonce,
|
||||
@@ -289,8 +290,7 @@ int scanhash_hex( struct work *work, uint32_t max_nonce,
|
||||
do
|
||||
{
|
||||
edata[19] = nonce;
|
||||
hex_hash( hash32, edata );
|
||||
|
||||
if ( hex_hash( hash32, edata, thr_id ) );
|
||||
if ( unlikely( valid_hash( hash32, ptarget ) && !bench ) )
|
||||
{
|
||||
be32enc( &pdata[19], nonce );
|
||||
|
||||
@@ -80,7 +80,7 @@ void x16r_8way_prehash( void *vdata, void *pdata )
|
||||
// Called by wrapper hash function to optionally continue hashing and
|
||||
// convert to final hash.
|
||||
|
||||
void x16r_8way_hash_generic( void* output, const void* input )
|
||||
int x16r_8way_hash_generic( void* output, const void* input, int thrid )
|
||||
{
|
||||
uint32_t vhash[20*8] __attribute__ ((aligned (128)));
|
||||
uint32_t hash0[20] __attribute__ ((aligned (64)));
|
||||
@@ -287,30 +287,14 @@ void x16r_8way_hash_generic( void* output, const void* input )
|
||||
shavite512_4way_full( &ctx.shavite, vhash, vhash, size );
|
||||
dintrlv_4x128_512( hash4, hash5, hash6, hash7, vhash );
|
||||
#else
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, in0, size );
|
||||
sph_shavite512_close( &ctx.shavite, hash0 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, in1, size );
|
||||
sph_shavite512_close( &ctx.shavite, hash1 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, in2, size );
|
||||
sph_shavite512_close( &ctx.shavite, hash2 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, in3, size );
|
||||
sph_shavite512_close( &ctx.shavite, hash3 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, in4, size );
|
||||
sph_shavite512_close( &ctx.shavite, hash4 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, in5, size );
|
||||
sph_shavite512_close( &ctx.shavite, hash5 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, in6, size );
|
||||
sph_shavite512_close( &ctx.shavite, hash6 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, in7, size );
|
||||
sph_shavite512_close( &ctx.shavite, hash7 );
|
||||
shavite512_full( &ctx.shavite, hash0, in0, size );
|
||||
shavite512_full( &ctx.shavite, hash1, in1, size );
|
||||
shavite512_full( &ctx.shavite, hash2, in2, size );
|
||||
shavite512_full( &ctx.shavite, hash3, in3, size );
|
||||
shavite512_full( &ctx.shavite, hash4, in4, size );
|
||||
shavite512_full( &ctx.shavite, hash5, in5, size );
|
||||
shavite512_full( &ctx.shavite, hash6, in6, size );
|
||||
shavite512_full( &ctx.shavite, hash7, in7, size );
|
||||
#endif
|
||||
break;
|
||||
case SIMD:
|
||||
@@ -363,30 +347,14 @@ void x16r_8way_hash_generic( void* output, const void* input )
|
||||
hash7, vhash );
|
||||
break;
|
||||
case FUGUE:
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, in0, size );
|
||||
sph_fugue512_close( &ctx.fugue, hash0 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, in1, size );
|
||||
sph_fugue512_close( &ctx.fugue, hash1 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, in2, size );
|
||||
sph_fugue512_close( &ctx.fugue, hash2 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, in3, size );
|
||||
sph_fugue512_close( &ctx.fugue, hash3 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, in4, size );
|
||||
sph_fugue512_close( &ctx.fugue, hash4 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, in5, size );
|
||||
sph_fugue512_close( &ctx.fugue, hash5 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, in6, size );
|
||||
sph_fugue512_close( &ctx.fugue, hash6 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, in7, size );
|
||||
sph_fugue512_close( &ctx.fugue, hash7 );
|
||||
sph_fugue512_full( &ctx.fugue, hash0, in0, size );
|
||||
sph_fugue512_full( &ctx.fugue, hash1, in1, size );
|
||||
sph_fugue512_full( &ctx.fugue, hash2, in2, size );
|
||||
sph_fugue512_full( &ctx.fugue, hash3, in3, size );
|
||||
sph_fugue512_full( &ctx.fugue, hash4, in4, size );
|
||||
sph_fugue512_full( &ctx.fugue, hash5, in5, size );
|
||||
sph_fugue512_full( &ctx.fugue, hash6, in6, size );
|
||||
sph_fugue512_full( &ctx.fugue, hash7, in7, size );
|
||||
break;
|
||||
case SHABAL:
|
||||
intrlv_8x32( vhash, in0, in1, in2, in3, in4, in5, in6, in7,
|
||||
@@ -431,30 +399,14 @@ void x16r_8way_hash_generic( void* output, const void* input )
|
||||
}
|
||||
else
|
||||
{
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, in0, size );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash0 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, in1, size );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash1 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, in2, size );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash2 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, in3, size );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash3 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, in4, size );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash4 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, in5, size );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash5 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, in6, size );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash6 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, in7, size );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash7 );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash0, in0, size );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash1, in1, size );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash2, in2, size );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash3, in3, size );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash4, in4, size );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash5, in5, size );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash6, in6, size );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash7, in7, size );
|
||||
}
|
||||
break;
|
||||
case SHA_512:
|
||||
@@ -472,6 +424,9 @@ void x16r_8way_hash_generic( void* output, const void* input )
|
||||
hash7, vhash );
|
||||
break;
|
||||
}
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
size = 64;
|
||||
}
|
||||
|
||||
@@ -483,14 +438,17 @@ void x16r_8way_hash_generic( void* output, const void* input )
|
||||
memcpy( output+320, hash5, 64 );
|
||||
memcpy( output+384, hash6, 64 );
|
||||
memcpy( output+448, hash7, 64 );
|
||||
|
||||
return 1;
|
||||
}
|
||||
|
||||
// x16-r,-s,-rt wrapper called directly by scanhash to repackage 512 bit
|
||||
// hash to 256 bit final hash.
|
||||
void x16r_8way_hash( void* output, const void* input )
|
||||
int x16r_8way_hash( void* output, const void* input, int thrid )
|
||||
{
|
||||
uint8_t hash[64*8] __attribute__ ((aligned (128)));
|
||||
x16r_8way_hash_generic( hash, input );
|
||||
if ( !x16r_8way_hash_generic( hash, input, thrid ) )
|
||||
return 0;
|
||||
|
||||
memcpy( output, hash, 32 );
|
||||
memcpy( output+32, hash+64, 32 );
|
||||
@@ -500,7 +458,9 @@ void x16r_8way_hash( void* output, const void* input )
|
||||
memcpy( output+160, hash+320, 32 );
|
||||
memcpy( output+192, hash+384, 32 );
|
||||
memcpy( output+224, hash+448, 32 );
|
||||
}
|
||||
|
||||
return 1;
|
||||
}
|
||||
|
||||
// x16r only
|
||||
int scanhash_x16r_8way( struct work *work, uint32_t max_nonce,
|
||||
@@ -540,13 +500,12 @@ int scanhash_x16r_8way( struct work *work, uint32_t max_nonce,
|
||||
n+3, 0, n+2, 0, n+1, 0, n, 0 ), *noncev );
|
||||
do
|
||||
{
|
||||
x16r_8way_hash( hash, vdata );
|
||||
|
||||
if( x16r_8way_hash( hash, vdata, thr_id ) );
|
||||
for ( int i = 0; i < 8; i++ )
|
||||
if ( unlikely( valid_hash( hash + (i<<3), ptarget ) && !bench ) )
|
||||
{
|
||||
pdata[19] = bswap_32( n+i );
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
*noncev = _mm512_add_epi32( *noncev,
|
||||
m512_const1_64( 0x0000000800000000 ) );
|
||||
@@ -576,8 +535,7 @@ void x16r_4way_prehash( void *vdata, void *pdata )
|
||||
break;
|
||||
case SKEIN:
|
||||
mm256_bswap32_intrlv80_4x64( vdata, pdata );
|
||||
skein512_4way_init( &x16r_ctx.skein );
|
||||
skein512_4way_update( &x16r_ctx.skein, vdata, 64 );
|
||||
skein512_4way_prehash64( &x16r_ctx.skein, vdata );
|
||||
break;
|
||||
case LUFFA:
|
||||
mm128_bswap32_80( edata, pdata );
|
||||
@@ -614,7 +572,7 @@ void x16r_4way_prehash( void *vdata, void *pdata )
|
||||
}
|
||||
}
|
||||
|
||||
void x16r_4way_hash_generic( void* output, const void* input )
|
||||
int x16r_4way_hash_generic( void* output, const void* input, int thrid )
|
||||
{
|
||||
uint32_t vhash[20*4] __attribute__ ((aligned (128)));
|
||||
uint32_t hash0[20] __attribute__ ((aligned (64)));
|
||||
@@ -692,10 +650,7 @@ void x16r_4way_hash_generic( void* output, const void* input )
|
||||
break;
|
||||
case SKEIN:
|
||||
if ( i == 0 )
|
||||
{
|
||||
skein512_4way_update( &ctx.skein, input + (64<<2), 16 );
|
||||
skein512_4way_close( &ctx.skein, vhash );
|
||||
}
|
||||
skein512_4way_final16( &ctx.skein, vhash, input + (64*4) );
|
||||
else
|
||||
{
|
||||
intrlv_4x64( vhash, in0, in1, in2, in3, size<<3 );
|
||||
@@ -756,18 +711,10 @@ void x16r_4way_hash_generic( void* output, const void* input )
|
||||
}
|
||||
break;
|
||||
case SHAVITE:
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, in0, size );
|
||||
sph_shavite512_close( &ctx.shavite, hash0 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, in1, size );
|
||||
sph_shavite512_close( &ctx.shavite, hash1 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, in2, size );
|
||||
sph_shavite512_close( &ctx.shavite, hash2 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, in3, size );
|
||||
sph_shavite512_close( &ctx.shavite, hash3 );
|
||||
shavite512_full( &ctx.shavite, hash0, in0, size );
|
||||
shavite512_full( &ctx.shavite, hash1, in1, size );
|
||||
shavite512_full( &ctx.shavite, hash2, in2, size );
|
||||
shavite512_full( &ctx.shavite, hash3, in3, size );
|
||||
break;
|
||||
case SIMD:
|
||||
intrlv_2x128( vhash, in0, in1, size<<3 );
|
||||
@@ -800,18 +747,10 @@ void x16r_4way_hash_generic( void* output, const void* input )
|
||||
dintrlv_4x64_512( hash0, hash1, hash2, hash3, vhash );
|
||||
break;
|
||||
case FUGUE:
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, in0, size );
|
||||
sph_fugue512_close( &ctx.fugue, hash0 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, in1, size );
|
||||
sph_fugue512_close( &ctx.fugue, hash1 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, in2, size );
|
||||
sph_fugue512_close( &ctx.fugue, hash2 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, in3, size );
|
||||
sph_fugue512_close( &ctx.fugue, hash3 );
|
||||
sph_fugue512_full( &ctx.fugue, hash0, in0, size );
|
||||
sph_fugue512_full( &ctx.fugue, hash1, in1, size );
|
||||
sph_fugue512_full( &ctx.fugue, hash2, in2, size );
|
||||
sph_fugue512_full( &ctx.fugue, hash3, in3, size );
|
||||
break;
|
||||
case SHABAL:
|
||||
intrlv_4x32( vhash, in0, in1, in2, in3, size<<3 );
|
||||
@@ -842,18 +781,10 @@ void x16r_4way_hash_generic( void* output, const void* input )
|
||||
}
|
||||
else
|
||||
{
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, in0, size );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash0 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, in1, size );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash1 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, in2, size );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash2 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, in3, size );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash3 );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash0, in0, size );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash1, in1, size );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash2, in2, size );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash3, in3, size );
|
||||
}
|
||||
break;
|
||||
case SHA_512:
|
||||
@@ -870,23 +801,31 @@ void x16r_4way_hash_generic( void* output, const void* input )
|
||||
dintrlv_4x64_512( hash0, hash1, hash2, hash3, vhash );
|
||||
break;
|
||||
}
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
size = 64;
|
||||
}
|
||||
memcpy( output, hash0, 64 );
|
||||
memcpy( output+64, hash1, 64 );
|
||||
memcpy( output+128, hash2, 64 );
|
||||
memcpy( output+192, hash3, 64 );
|
||||
|
||||
return 1;
|
||||
}
|
||||
|
||||
void x16r_4way_hash( void* output, const void* input )
|
||||
int x16r_4way_hash( void* output, const void* input, int thrid )
|
||||
{
|
||||
uint8_t hash[64*4] __attribute__ ((aligned (64)));
|
||||
x16r_4way_hash_generic( hash, input );
|
||||
if ( !x16r_4way_hash_generic( hash, input, thrid ) )
|
||||
return 0;
|
||||
|
||||
memcpy( output, hash, 32 );
|
||||
memcpy( output+32, hash+64, 32 );
|
||||
memcpy( output+64, hash+128, 32 );
|
||||
memcpy( output+96, hash+192, 32 );
|
||||
|
||||
return 1;
|
||||
}
|
||||
|
||||
int scanhash_x16r_4way( struct work *work, uint32_t max_nonce,
|
||||
@@ -925,12 +864,12 @@ int scanhash_x16r_4way( struct work *work, uint32_t max_nonce,
|
||||
_mm256_set_epi32( n+3, 0, n+2, 0, n+1, 0, n, 0 ), *noncev );
|
||||
do
|
||||
{
|
||||
x16r_4way_hash( hash, vdata );
|
||||
if ( x16r_4way_hash( hash, vdata, thr_id ) );
|
||||
for ( int i = 0; i < 4; i++ )
|
||||
if ( unlikely( valid_hash( hash + (i<<3), ptarget ) && !bench ) )
|
||||
{
|
||||
pdata[19] = bswap_32( n+i );
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
*noncev = _mm256_add_epi32( *noncev,
|
||||
m256_const1_64( 0x0000000400000000 ) );
|
||||
|
||||
@@ -121,7 +121,7 @@ union _x16r_8way_context_overlay
|
||||
echo_4way_context echo;
|
||||
#else
|
||||
hashState_groestl groestl;
|
||||
sph_shavite512_context shavite;
|
||||
shavite512_context shavite;
|
||||
hashState_echo echo;
|
||||
#endif
|
||||
} __attribute__ ((aligned (64)));
|
||||
@@ -131,8 +131,8 @@ typedef union _x16r_8way_context_overlay x16r_8way_context_overlay;
|
||||
extern __thread x16r_8way_context_overlay x16r_ctx;
|
||||
|
||||
void x16r_8way_prehash( void *, void * );
|
||||
void x16r_8way_hash_generic( void *, const void * );
|
||||
void x16r_8way_hash( void *, const void * );
|
||||
int x16r_8way_hash_generic( void *, const void *, int );
|
||||
int x16r_8way_hash( void *, const void *, int );
|
||||
int scanhash_x16r_8way( struct work *, uint32_t ,
|
||||
uint64_t *, struct thr_info * );
|
||||
extern __thread x16r_8way_context_overlay x16r_ctx;
|
||||
@@ -152,7 +152,7 @@ union _x16r_4way_context_overlay
|
||||
luffa_2way_context luffa;
|
||||
hashState_luffa luffa1;
|
||||
cubehashParam cube;
|
||||
sph_shavite512_context shavite;
|
||||
shavite512_context shavite;
|
||||
simd_2way_context simd;
|
||||
hamsi512_4way_context hamsi;
|
||||
sph_fugue512_context fugue;
|
||||
@@ -166,8 +166,8 @@ typedef union _x16r_4way_context_overlay x16r_4way_context_overlay;
|
||||
extern __thread x16r_4way_context_overlay x16r_ctx;
|
||||
|
||||
void x16r_4way_prehash( void *, void * );
|
||||
void x16r_4way_hash_generic( void *, const void * );
|
||||
void x16r_4way_hash( void *, const void * );
|
||||
int x16r_4way_hash_generic( void *, const void *, int );
|
||||
int x16r_4way_hash( void *, const void *, int );
|
||||
int scanhash_x16r_4way( struct work *, uint32_t,
|
||||
uint64_t *, struct thr_info * );
|
||||
extern __thread x16r_4way_context_overlay x16r_ctx;
|
||||
@@ -191,7 +191,7 @@ union _x16r_context_overlay
|
||||
sph_keccak512_context keccak;
|
||||
hashState_luffa luffa;
|
||||
cubehashParam cube;
|
||||
sph_shavite512_context shavite;
|
||||
shavite512_context shavite;
|
||||
hashState_sd simd;
|
||||
sph_hamsi512_context hamsi;
|
||||
sph_fugue512_context fugue;
|
||||
@@ -205,26 +205,26 @@ typedef union _x16r_context_overlay x16r_context_overlay;
|
||||
extern __thread x16r_context_overlay x16_ctx;
|
||||
|
||||
void x16r_prehash( void *, void * );
|
||||
void x16r_hash_generic( void *, const void * );
|
||||
void x16r_hash( void *, const void * );
|
||||
int x16r_hash_generic( void *, const void *, int );
|
||||
int x16r_hash( void *, const void *, int );
|
||||
int scanhash_x16r( struct work *, uint32_t, uint64_t *, struct thr_info * );
|
||||
|
||||
// x16Rv2
|
||||
#if defined(X16RV2_8WAY)
|
||||
|
||||
void x16rv2_8way_hash( void *state, const void *input );
|
||||
int x16rv2_8way_hash( void *state, const void *input, int thrid );
|
||||
int scanhash_x16rv2_8way( struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr );
|
||||
|
||||
#elif defined(X16RV2_4WAY)
|
||||
|
||||
void x16rv2_4way_hash( void *state, const void *input );
|
||||
int x16rv2_4way_hash( void *state, const void *input, int thrid );
|
||||
int scanhash_x16rv2_4way( struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr );
|
||||
|
||||
#else
|
||||
|
||||
void x16rv2_hash( void *state, const void *input );
|
||||
int x16rv2_hash( void *state, const void *input, int thr_id );
|
||||
int scanhash_x16rv2( struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr );
|
||||
|
||||
@@ -254,21 +254,21 @@ int scanhash_x16rt( struct work *work, uint32_t max_nonce,
|
||||
// x21s
|
||||
#if defined(X16R_8WAY)
|
||||
|
||||
void x21s_8way_hash( void *state, const void *input );
|
||||
int x21s_8way_hash( void *state, const void *input, int thrid );
|
||||
int scanhash_x21s_8way( struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr );
|
||||
bool x21s_8way_thread_init();
|
||||
|
||||
#elif defined(X16R_4WAY)
|
||||
|
||||
void x21s_4way_hash( void *state, const void *input );
|
||||
int x21s_4way_hash( void *state, const void *input, int thrid );
|
||||
int scanhash_x21s_4way( struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr );
|
||||
bool x21s_4way_thread_init();
|
||||
|
||||
#else
|
||||
|
||||
void x21s_hash( void *state, const void *input );
|
||||
int x21s_hash( void *state, const void *input, int thr_id );
|
||||
int scanhash_x21s( struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr );
|
||||
bool x21s_thread_init();
|
||||
|
||||
@@ -48,7 +48,7 @@ void x16r_prehash( void *edata, void *pdata )
|
||||
}
|
||||
}
|
||||
|
||||
void x16r_hash_generic( void* output, const void* input )
|
||||
int x16r_hash_generic( void* output, const void* input, int thrid )
|
||||
{
|
||||
uint32_t _ALIGN(128) hash[16];
|
||||
x16r_context_overlay ctx;
|
||||
@@ -124,9 +124,7 @@ void x16r_hash_generic( void* output, const void* input )
|
||||
(byte*)in, size );
|
||||
break;
|
||||
case SHAVITE:
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, in, size );
|
||||
sph_shavite512_close( &ctx.shavite, hash );
|
||||
shavite512_full( &ctx.shavite, hash, in, size );
|
||||
break;
|
||||
case SIMD:
|
||||
simd_full( &ctx.simd, (BitSequence *)hash,
|
||||
@@ -153,9 +151,7 @@ void x16r_hash_generic( void* output, const void* input )
|
||||
sph_hamsi512_close( &ctx.hamsi, hash );
|
||||
break;
|
||||
case FUGUE:
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, in, size );
|
||||
sph_fugue512_close( &ctx.fugue, hash );
|
||||
sph_fugue512_full( &ctx.fugue, hash, in, size );
|
||||
break;
|
||||
case SHABAL:
|
||||
if ( i == 0 )
|
||||
@@ -169,13 +165,12 @@ void x16r_hash_generic( void* output, const void* input )
|
||||
break;
|
||||
case WHIRLPOOL:
|
||||
if ( i == 0 )
|
||||
sph_whirlpool( &ctx.whirlpool, in+64, 16 );
|
||||
else
|
||||
{
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, in, size );
|
||||
sph_whirlpool( &ctx.whirlpool, in+64, 16 );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash );
|
||||
}
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash );
|
||||
else
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash, in, size );
|
||||
break;
|
||||
case SHA_512:
|
||||
SHA512_Init( &ctx.sha512 );
|
||||
@@ -183,18 +178,24 @@ void x16r_hash_generic( void* output, const void* input )
|
||||
SHA512_Final( (unsigned char*) hash, &ctx.sha512 );
|
||||
break;
|
||||
}
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
in = (void*) hash;
|
||||
size = 64;
|
||||
}
|
||||
memcpy( output, hash, 64 );
|
||||
return true;
|
||||
}
|
||||
|
||||
void x16r_hash( void* output, const void* input )
|
||||
int x16r_hash( void* output, const void* input, int thrid )
|
||||
{
|
||||
uint8_t hash[64] __attribute__ ((aligned (64)));
|
||||
x16r_hash_generic( hash, input );
|
||||
if ( !x16r_hash_generic( hash, input, thrid ) )
|
||||
return 0;
|
||||
|
||||
memcpy( output, hash, 32 );
|
||||
memcpy( output, hash, 32 );
|
||||
return 1;
|
||||
}
|
||||
|
||||
int scanhash_x16r( struct work *work, uint32_t max_nonce,
|
||||
@@ -228,8 +229,7 @@ int scanhash_x16r( struct work *work, uint32_t max_nonce,
|
||||
do
|
||||
{
|
||||
edata[19] = nonce;
|
||||
x16r_hash( hash32, edata );
|
||||
|
||||
if ( x16r_hash( hash32, edata, thr_id ) )
|
||||
if ( unlikely( valid_hash( hash32, ptarget ) && !bench ) )
|
||||
{
|
||||
pdata[19] = bswap_32( nonce );
|
||||
|
||||
@@ -41,13 +41,12 @@ int scanhash_x16rt_8way( struct work *work, uint32_t max_nonce,
|
||||
n+3, 0, n+2, 0, n+1, 0, n, 0 ), *noncev );
|
||||
do
|
||||
{
|
||||
x16r_8way_hash( hash, vdata );
|
||||
|
||||
if ( x16r_8way_hash( hash, vdata, thr_id ) )
|
||||
for ( int i = 0; i < 8; i++ )
|
||||
if ( unlikely( valid_hash( hash + (i<<3), ptarget ) && !bench ) )
|
||||
{
|
||||
pdata[19] = bswap_32( n+i );
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
*noncev = _mm512_add_epi32( *noncev,
|
||||
m512_const1_64( 0x0000000800000000 ) );
|
||||
@@ -95,12 +94,12 @@ int scanhash_x16rt_4way( struct work *work, uint32_t max_nonce,
|
||||
_mm256_set_epi32( n+3, 0, n+2, 0, n+1, 0, n, 0 ), *noncev );
|
||||
do
|
||||
{
|
||||
x16r_4way_hash( hash, vdata );
|
||||
if ( x16r_4way_hash( hash, vdata, thr_id ) )
|
||||
for ( int i = 0; i < 4; i++ )
|
||||
if ( unlikely( valid_hash( hash + (i<<3), ptarget ) && !bench ) )
|
||||
{
|
||||
pdata[19] = bswap_32( n+i );
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
*noncev = _mm256_add_epi32( *noncev,
|
||||
m256_const1_64( 0x0000000400000000 ) );
|
||||
|
||||
@@ -36,8 +36,7 @@ int scanhash_x16rt( struct work *work, uint32_t max_nonce,
|
||||
do
|
||||
{
|
||||
edata[19] = nonce;
|
||||
x16r_hash( hash32, edata );
|
||||
|
||||
if ( x16r_hash( hash32, edata, thr_id ) )
|
||||
if ( valid_hash( hash32, ptarget ) && !bench )
|
||||
{
|
||||
pdata[19] = bswap_32( nonce );
|
||||
|
||||
@@ -57,7 +57,7 @@ union _x16rv2_8way_context_overlay
|
||||
echo_4way_context echo;
|
||||
#else
|
||||
hashState_groestl groestl;
|
||||
sph_shavite512_context shavite;
|
||||
shavite512_context shavite;
|
||||
hashState_echo echo;
|
||||
#endif
|
||||
} __attribute__ ((aligned (64)));
|
||||
@@ -65,7 +65,7 @@ union _x16rv2_8way_context_overlay
|
||||
typedef union _x16rv2_8way_context_overlay x16rv2_8way_context_overlay;
|
||||
static __thread x16rv2_8way_context_overlay x16rv2_ctx;
|
||||
|
||||
void x16rv2_8way_hash( void* output, const void* input )
|
||||
int x16rv2_8way_hash( void* output, const void* input, int thrid )
|
||||
{
|
||||
uint32_t vhash[24*8] __attribute__ ((aligned (128)));
|
||||
uint32_t hash0[24] __attribute__ ((aligned (64)));
|
||||
@@ -371,30 +371,14 @@ void x16rv2_8way_hash( void* output, const void* input )
|
||||
shavite512_4way_full( &ctx.shavite, vhash, vhash, size );
|
||||
dintrlv_4x128_512( hash4, hash5, hash6, hash7, vhash );
|
||||
#else
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, in0, size );
|
||||
sph_shavite512_close( &ctx.shavite, hash0 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, in1, size );
|
||||
sph_shavite512_close( &ctx.shavite, hash1 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, in2, size );
|
||||
sph_shavite512_close( &ctx.shavite, hash2 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, in3, size );
|
||||
sph_shavite512_close( &ctx.shavite, hash3 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, in4, size );
|
||||
sph_shavite512_close( &ctx.shavite, hash4 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, in5, size );
|
||||
sph_shavite512_close( &ctx.shavite, hash5 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, in6, size );
|
||||
sph_shavite512_close( &ctx.shavite, hash6 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, in7, size );
|
||||
sph_shavite512_close( &ctx.shavite, hash7 );
|
||||
shavite512_full( &ctx.shavite, hash0, in0, size );
|
||||
shavite512_full( &ctx.shavite, hash1, in1, size );
|
||||
shavite512_full( &ctx.shavite, hash2, in2, size );
|
||||
shavite512_full( &ctx.shavite, hash3, in3, size );
|
||||
shavite512_full( &ctx.shavite, hash4, in4, size );
|
||||
shavite512_full( &ctx.shavite, hash5, in5, size );
|
||||
shavite512_full( &ctx.shavite, hash6, in6, size );
|
||||
shavite512_full( &ctx.shavite, hash7, in7, size );
|
||||
#endif
|
||||
break;
|
||||
case SIMD:
|
||||
@@ -448,30 +432,14 @@ void x16rv2_8way_hash( void* output, const void* input )
|
||||
hash7, vhash );
|
||||
break;
|
||||
case FUGUE:
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, in0, size );
|
||||
sph_fugue512_close( &ctx.fugue, hash0 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, in1, size );
|
||||
sph_fugue512_close( &ctx.fugue, hash1 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, in2, size );
|
||||
sph_fugue512_close( &ctx.fugue, hash2 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, in3, size );
|
||||
sph_fugue512_close( &ctx.fugue, hash3 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, in4, size );
|
||||
sph_fugue512_close( &ctx.fugue, hash4 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, in5, size );
|
||||
sph_fugue512_close( &ctx.fugue, hash5 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, in6, size );
|
||||
sph_fugue512_close( &ctx.fugue, hash6 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, in7, size );
|
||||
sph_fugue512_close( &ctx.fugue, hash7 );
|
||||
sph_fugue512_full( &ctx.fugue, hash0, in0, size );
|
||||
sph_fugue512_full( &ctx.fugue, hash1, in1, size );
|
||||
sph_fugue512_full( &ctx.fugue, hash2, in2, size );
|
||||
sph_fugue512_full( &ctx.fugue, hash3, in3, size );
|
||||
sph_fugue512_full( &ctx.fugue, hash4, in4, size );
|
||||
sph_fugue512_full( &ctx.fugue, hash5, in5, size );
|
||||
sph_fugue512_full( &ctx.fugue, hash6, in6, size );
|
||||
sph_fugue512_full( &ctx.fugue, hash7, in7, size );
|
||||
break;
|
||||
case SHABAL:
|
||||
intrlv_8x32( vhash, in0, in1, in2, in3, in4, in5, in6, in7,
|
||||
@@ -516,30 +484,14 @@ void x16rv2_8way_hash( void* output, const void* input )
|
||||
}
|
||||
else
|
||||
{
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, in0, size );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash0 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, in1, size );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash1 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, in2, size );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash2 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, in3, size );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash3 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, in4, size );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash4 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, in5, size );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash5 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, in6, size );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash6 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, in7, size );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash7 );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash0, in0, size );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash1, in1, size );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash2, in2, size );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash3, in3, size );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash4, in4, size );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash5, in5, size );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash6, in6, size );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash7, in7, size );
|
||||
}
|
||||
break;
|
||||
case SHA_512:
|
||||
@@ -611,6 +563,9 @@ void x16rv2_8way_hash( void* output, const void* input )
|
||||
hash7, vhash );
|
||||
break;
|
||||
}
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
size = 64;
|
||||
}
|
||||
|
||||
@@ -622,6 +577,7 @@ void x16rv2_8way_hash( void* output, const void* input )
|
||||
memcpy( output+160, hash5, 32 );
|
||||
memcpy( output+192, hash6, 32 );
|
||||
memcpy( output+224, hash7, 32 );
|
||||
return 1;
|
||||
}
|
||||
|
||||
int scanhash_x16rv2_8way( struct work *work, uint32_t max_nonce,
|
||||
@@ -717,13 +673,12 @@ int scanhash_x16rv2_8way( struct work *work, uint32_t max_nonce,
|
||||
n+3, 0, n+2, 0, n+1, 0, n, 0 ), *noncev );
|
||||
do
|
||||
{
|
||||
x16rv2_8way_hash( hash, vdata );
|
||||
|
||||
if ( x16rv2_8way_hash( hash, vdata, thr_id ) )
|
||||
for ( int i = 0; i < 8; i++ )
|
||||
if ( unlikely( valid_hash( hash + (i<<3), ptarget ) && !bench ) )
|
||||
{
|
||||
pdata[19] = bswap_32( n+i );
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
*noncev = _mm512_add_epi32( *noncev,
|
||||
m512_const1_64( 0x0000000800000000 ) );
|
||||
@@ -747,7 +702,7 @@ union _x16rv2_4way_context_overlay
|
||||
keccak512_4way_context keccak;
|
||||
luffa_2way_context luffa;
|
||||
cubehashParam cube;
|
||||
sph_shavite512_context shavite;
|
||||
shavite512_context shavite;
|
||||
simd_2way_context simd;
|
||||
hamsi512_4way_context hamsi;
|
||||
sph_fugue512_context fugue;
|
||||
@@ -766,7 +721,7 @@ inline void padtiger512( uint32_t* hash )
|
||||
for ( int i = 6; i < 16; i++ ) hash[i] = 0;
|
||||
}
|
||||
|
||||
void x16rv2_4way_hash( void* output, const void* input )
|
||||
int x16rv2_4way_hash( void* output, const void* input, int thrid )
|
||||
{
|
||||
uint32_t hash0[20] __attribute__ ((aligned (64)));
|
||||
uint32_t hash1[20] __attribute__ ((aligned (64)));
|
||||
@@ -831,47 +786,47 @@ void x16rv2_4way_hash( void* output, const void* input )
|
||||
dintrlv_4x64_512( hash0, hash1, hash2, hash3, vhash );
|
||||
break;
|
||||
case KECCAK:
|
||||
if ( i == 0 )
|
||||
{
|
||||
sph_tiger( &ctx.tiger, in0 + 64, 16 );
|
||||
sph_tiger_close( &ctx.tiger, hash0 );
|
||||
memcpy( &ctx, &x16rv2_ctx, sizeof(ctx) );
|
||||
sph_tiger( &ctx.tiger, in1 + 64, 16 );
|
||||
sph_tiger_close( &ctx.tiger, hash1 );
|
||||
memcpy( &ctx, &x16rv2_ctx, sizeof(ctx) );
|
||||
sph_tiger( &ctx.tiger, in2 + 64, 16 );
|
||||
sph_tiger_close( &ctx.tiger, hash2 );
|
||||
memcpy( &ctx, &x16rv2_ctx, sizeof(ctx) );
|
||||
sph_tiger( &ctx.tiger, in3 + 64, 16 );
|
||||
sph_tiger_close( &ctx.tiger, hash3 );
|
||||
}
|
||||
else
|
||||
{
|
||||
sph_tiger_init( &ctx.tiger );
|
||||
sph_tiger( &ctx.tiger, in0, size );
|
||||
sph_tiger_close( &ctx.tiger, hash0 );
|
||||
sph_tiger_init( &ctx.tiger );
|
||||
sph_tiger( &ctx.tiger, in1, size );
|
||||
sph_tiger_close( &ctx.tiger, hash1 );
|
||||
sph_tiger_init( &ctx.tiger );
|
||||
sph_tiger( &ctx.tiger, in2, size );
|
||||
sph_tiger_close( &ctx.tiger, hash2 );
|
||||
sph_tiger_init( &ctx.tiger );
|
||||
sph_tiger( &ctx.tiger, in3, size );
|
||||
sph_tiger_close( &ctx.tiger, hash3 );
|
||||
}
|
||||
for ( int i = (24/4); i < (64/4); i++ )
|
||||
if ( i == 0 )
|
||||
{
|
||||
sph_tiger( &ctx.tiger, in0 + 64, 16 );
|
||||
sph_tiger_close( &ctx.tiger, hash0 );
|
||||
memcpy( &ctx, &x16rv2_ctx, sizeof(ctx) );
|
||||
sph_tiger( &ctx.tiger, in1 + 64, 16 );
|
||||
sph_tiger_close( &ctx.tiger, hash1 );
|
||||
memcpy( &ctx, &x16rv2_ctx, sizeof(ctx) );
|
||||
sph_tiger( &ctx.tiger, in2 + 64, 16 );
|
||||
sph_tiger_close( &ctx.tiger, hash2 );
|
||||
memcpy( &ctx, &x16rv2_ctx, sizeof(ctx) );
|
||||
sph_tiger( &ctx.tiger, in3 + 64, 16 );
|
||||
sph_tiger_close( &ctx.tiger, hash3 );
|
||||
}
|
||||
else
|
||||
{
|
||||
sph_tiger_init( &ctx.tiger );
|
||||
sph_tiger( &ctx.tiger, in0, size );
|
||||
sph_tiger_close( &ctx.tiger, hash0 );
|
||||
sph_tiger_init( &ctx.tiger );
|
||||
sph_tiger( &ctx.tiger, in1, size );
|
||||
sph_tiger_close( &ctx.tiger, hash1 );
|
||||
sph_tiger_init( &ctx.tiger );
|
||||
sph_tiger( &ctx.tiger, in2, size );
|
||||
sph_tiger_close( &ctx.tiger, hash2 );
|
||||
sph_tiger_init( &ctx.tiger );
|
||||
sph_tiger( &ctx.tiger, in3, size );
|
||||
sph_tiger_close( &ctx.tiger, hash3 );
|
||||
}
|
||||
for ( int i = (24/4); i < (64/4); i++ )
|
||||
hash0[i] = hash1[i] = hash2[i] = hash3[i] = 0;
|
||||
|
||||
intrlv_4x64( vhash, hash0, hash1, hash2, hash3, 512 );
|
||||
keccak512_4way_init( &ctx.keccak );
|
||||
keccak512_4way_update( &ctx.keccak, vhash, 64 );
|
||||
keccak512_4way_close( &ctx.keccak, vhash );
|
||||
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
|
||||
intrlv_4x64( vhash, hash0, hash1, hash2, hash3, 512 );
|
||||
keccak512_4way_init( &ctx.keccak );
|
||||
keccak512_4way_update( &ctx.keccak, vhash, 64 );
|
||||
keccak512_4way_close( &ctx.keccak, vhash );
|
||||
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
|
||||
break;
|
||||
case SKEIN:
|
||||
if ( i == 0 )
|
||||
skein512_4way_update( &ctx.skein, input + (64<<2), 16 );
|
||||
skein512_4way_final16( &ctx.skein, vhash, input + (64*4) );
|
||||
else
|
||||
{
|
||||
intrlv_4x64( vhash, in0, in1, in2, in3, size<<3 );
|
||||
@@ -882,46 +837,46 @@ void x16rv2_4way_hash( void* output, const void* input )
|
||||
dintrlv_4x64_512( hash0, hash1, hash2, hash3, vhash );
|
||||
break;
|
||||
case LUFFA:
|
||||
if ( i == 0 )
|
||||
{
|
||||
sph_tiger( &ctx.tiger, in0 + 64, 16 );
|
||||
sph_tiger_close( &ctx.tiger, hash0 );
|
||||
memcpy( &ctx, &x16rv2_ctx, sizeof(ctx) );
|
||||
sph_tiger( &ctx.tiger, in1 + 64, 16 );
|
||||
sph_tiger_close( &ctx.tiger, hash1 );
|
||||
memcpy( &ctx, &x16rv2_ctx, sizeof(ctx) );
|
||||
sph_tiger( &ctx.tiger, in2 + 64, 16 );
|
||||
sph_tiger_close( &ctx.tiger, hash2 );
|
||||
memcpy( &ctx, &x16rv2_ctx, sizeof(ctx) );
|
||||
sph_tiger( &ctx.tiger, in3 + 64, 16 );
|
||||
sph_tiger_close( &ctx.tiger, hash3 );
|
||||
}
|
||||
else
|
||||
{
|
||||
sph_tiger_init( &ctx.tiger );
|
||||
sph_tiger( &ctx.tiger, in0, size );
|
||||
sph_tiger_close( &ctx.tiger, hash0 );
|
||||
sph_tiger_init( &ctx.tiger );
|
||||
sph_tiger( &ctx.tiger, in1, size );
|
||||
sph_tiger_close( &ctx.tiger, hash1 );
|
||||
sph_tiger_init( &ctx.tiger );
|
||||
sph_tiger( &ctx.tiger, in2, size );
|
||||
sph_tiger_close( &ctx.tiger, hash2 );
|
||||
sph_tiger_init( &ctx.tiger );
|
||||
sph_tiger( &ctx.tiger, in3, size );
|
||||
sph_tiger_close( &ctx.tiger, hash3 );
|
||||
}
|
||||
for ( int i = (24/4); i < (64/4); i++ )
|
||||
if ( i == 0 )
|
||||
{
|
||||
sph_tiger( &ctx.tiger, in0 + 64, 16 );
|
||||
sph_tiger_close( &ctx.tiger, hash0 );
|
||||
memcpy( &ctx, &x16rv2_ctx, sizeof(ctx) );
|
||||
sph_tiger( &ctx.tiger, in1 + 64, 16 );
|
||||
sph_tiger_close( &ctx.tiger, hash1 );
|
||||
memcpy( &ctx, &x16rv2_ctx, sizeof(ctx) );
|
||||
sph_tiger( &ctx.tiger, in2 + 64, 16 );
|
||||
sph_tiger_close( &ctx.tiger, hash2 );
|
||||
memcpy( &ctx, &x16rv2_ctx, sizeof(ctx) );
|
||||
sph_tiger( &ctx.tiger, in3 + 64, 16 );
|
||||
sph_tiger_close( &ctx.tiger, hash3 );
|
||||
}
|
||||
else
|
||||
{
|
||||
sph_tiger_init( &ctx.tiger );
|
||||
sph_tiger( &ctx.tiger, in0, size );
|
||||
sph_tiger_close( &ctx.tiger, hash0 );
|
||||
sph_tiger_init( &ctx.tiger );
|
||||
sph_tiger( &ctx.tiger, in1, size );
|
||||
sph_tiger_close( &ctx.tiger, hash1 );
|
||||
sph_tiger_init( &ctx.tiger );
|
||||
sph_tiger( &ctx.tiger, in2, size );
|
||||
sph_tiger_close( &ctx.tiger, hash2 );
|
||||
sph_tiger_init( &ctx.tiger );
|
||||
sph_tiger( &ctx.tiger, in3, size );
|
||||
sph_tiger_close( &ctx.tiger, hash3 );
|
||||
}
|
||||
for ( int i = (24/4); i < (64/4); i++ )
|
||||
hash0[i] = hash1[i] = hash2[i] = hash3[i] = 0;
|
||||
|
||||
intrlv_2x128( vhash, hash0, hash1, 512 );
|
||||
luffa_2way_init( &ctx.luffa, 512 );
|
||||
luffa_2way_update_close( &ctx.luffa, vhash, vhash, 64 );
|
||||
dintrlv_2x128( hash0, hash1, vhash, 512 );
|
||||
intrlv_2x128( vhash, hash2, hash3, 512 );
|
||||
luffa_2way_init( &ctx.luffa, 512 );
|
||||
luffa_2way_update_close( &ctx.luffa, vhash, vhash, 64 );
|
||||
dintrlv_2x128( hash2, hash3, vhash, 512 );
|
||||
intrlv_2x128( vhash, hash0, hash1, 512 );
|
||||
luffa_2way_init( &ctx.luffa, 512 );
|
||||
luffa_2way_update_close( &ctx.luffa, vhash, vhash, 64 );
|
||||
dintrlv_2x128( hash0, hash1, vhash, 512 );
|
||||
intrlv_2x128( vhash, hash2, hash3, 512 );
|
||||
luffa_2way_init( &ctx.luffa, 512 );
|
||||
luffa_2way_update_close( &ctx.luffa, vhash, vhash, 64 );
|
||||
dintrlv_2x128( hash2, hash3, vhash, 512 );
|
||||
break;
|
||||
case CUBEHASH:
|
||||
if ( i == 0 )
|
||||
@@ -955,18 +910,10 @@ void x16rv2_4way_hash( void* output, const void* input )
|
||||
}
|
||||
break;
|
||||
case SHAVITE:
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, in0, size );
|
||||
sph_shavite512_close( &ctx.shavite, hash0 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, in1, size );
|
||||
sph_shavite512_close( &ctx.shavite, hash1 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, in2, size );
|
||||
sph_shavite512_close( &ctx.shavite, hash2 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, in3, size );
|
||||
sph_shavite512_close( &ctx.shavite, hash3 );
|
||||
shavite512_full( &ctx.shavite, hash0, in0, size );
|
||||
shavite512_full( &ctx.shavite, hash1, in1, size );
|
||||
shavite512_full( &ctx.shavite, hash2, in2, size );
|
||||
shavite512_full( &ctx.shavite, hash3, in3, size );
|
||||
break;
|
||||
case SIMD:
|
||||
intrlv_2x128( vhash, in0, in1, size<<3 );
|
||||
@@ -999,18 +946,10 @@ void x16rv2_4way_hash( void* output, const void* input )
|
||||
dintrlv_4x64_512( hash0, hash1, hash2, hash3, vhash );
|
||||
break;
|
||||
case FUGUE:
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, in0, size );
|
||||
sph_fugue512_close( &ctx.fugue, hash0 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, in1, size );
|
||||
sph_fugue512_close( &ctx.fugue, hash1 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, in2, size );
|
||||
sph_fugue512_close( &ctx.fugue, hash2 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, in3, size );
|
||||
sph_fugue512_close( &ctx.fugue, hash3 );
|
||||
sph_fugue512_full( &ctx.fugue, hash0, in0, size );
|
||||
sph_fugue512_full( &ctx.fugue, hash1, in1, size );
|
||||
sph_fugue512_full( &ctx.fugue, hash2, in2, size );
|
||||
sph_fugue512_full( &ctx.fugue, hash3, in3, size );
|
||||
break;
|
||||
case SHABAL:
|
||||
intrlv_4x32( vhash, in0, in1, in2, in3, size<<3 );
|
||||
@@ -1041,18 +980,10 @@ void x16rv2_4way_hash( void* output, const void* input )
|
||||
}
|
||||
else
|
||||
{
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, in0, size );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash0 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, in1, size );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash1 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, in2, size );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash2 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, in3, size );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash3 );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash0, in0, size );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash1, in1, size );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash2, in2, size );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash3, in3, size );
|
||||
}
|
||||
break;
|
||||
case SHA_512:
|
||||
@@ -1095,12 +1026,16 @@ void x16rv2_4way_hash( void* output, const void* input )
|
||||
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
|
||||
break;
|
||||
}
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
size = 64;
|
||||
}
|
||||
memcpy( output, hash0, 32 );
|
||||
memcpy( output+32, hash1, 32 );
|
||||
memcpy( output+64, hash2, 32 );
|
||||
memcpy( output+96, hash3, 32 );
|
||||
return 1;
|
||||
}
|
||||
|
||||
int scanhash_x16rv2_4way( struct work *work, uint32_t max_nonce,
|
||||
@@ -1117,7 +1052,7 @@ int scanhash_x16rv2_4way( struct work *work, uint32_t max_nonce,
|
||||
const uint32_t last_nonce = max_nonce - 4;
|
||||
uint32_t n = first_nonce;
|
||||
const int thr_id = mythr->id;
|
||||
__m256i *noncev = (__m256i*)vdata + 9; // aligned
|
||||
__m256i *noncev = (__m256i*)vdata + 9;
|
||||
volatile uint8_t *restart = &(work_restart[thr_id].restart);
|
||||
const bool bench = opt_benchmark;
|
||||
|
||||
@@ -1134,7 +1069,7 @@ int scanhash_x16rv2_4way( struct work *work, uint32_t max_nonce,
|
||||
x16_r_s_getAlgoString( (const uint8_t*)bedata1, x16r_hash_order );
|
||||
s_ntime = ntime;
|
||||
if ( opt_debug && !thr_id )
|
||||
applog( LOG_DEBUG, "hash order %s (%08x)", x16r_hash_order, ntime );
|
||||
applog( LOG_INFO, "hash order %s (%08x)", x16r_hash_order, ntime );
|
||||
}
|
||||
|
||||
// Do midstate prehash on hash functions with block size <= 64 bytes.
|
||||
@@ -1157,8 +1092,7 @@ int scanhash_x16rv2_4way( struct work *work, uint32_t max_nonce,
|
||||
break;
|
||||
case SKEIN:
|
||||
mm256_bswap32_intrlv80_4x64( vdata, pdata );
|
||||
skein512_4way_init( &x16rv2_ctx.skein );
|
||||
skein512_4way_update( &x16rv2_ctx.skein, vdata, 64 );
|
||||
skein512_4way_prehash64( &x16r_ctx.skein, vdata );
|
||||
break;
|
||||
case CUBEHASH:
|
||||
mm128_bswap32_80( edata, pdata );
|
||||
@@ -1192,12 +1126,12 @@ int scanhash_x16rv2_4way( struct work *work, uint32_t max_nonce,
|
||||
|
||||
do
|
||||
{
|
||||
x16rv2_4way_hash( hash, vdata );
|
||||
if ( x16rv2_4way_hash( hash, vdata, thr_id ) )
|
||||
for ( int i = 0; i < 4; i++ )
|
||||
if ( unlikely( valid_hash( hash + (i<<3), ptarget ) && !bench ) )
|
||||
{
|
||||
pdata[19] = bswap_32( n+i );
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
*noncev = _mm256_add_epi32( *noncev,
|
||||
m256_const1_64( 0x0000000400000000 ) );
|
||||
|
||||
@@ -51,7 +51,7 @@ union _x16rv2_context_overlay
|
||||
sph_keccak512_context keccak;
|
||||
hashState_luffa luffa;
|
||||
cubehashParam cube;
|
||||
sph_shavite512_context shavite;
|
||||
shavite512_context shavite;
|
||||
hashState_sd simd;
|
||||
sph_hamsi512_context hamsi;
|
||||
sph_fugue512_context fugue;
|
||||
@@ -67,7 +67,7 @@ inline void padtiger512(uint32_t* hash) {
|
||||
for (int i = (24/4); i < (64/4); i++) hash[i] = 0;
|
||||
}
|
||||
|
||||
void x16rv2_hash( void* output, const void* input )
|
||||
int x16rv2_hash( void* output, const void* input, int thrid )
|
||||
{
|
||||
uint32_t _ALIGN(128) hash[16];
|
||||
x16rv2_context_overlay ctx;
|
||||
@@ -136,9 +136,7 @@ void x16rv2_hash( void* output, const void* input )
|
||||
(const byte*)in, size );
|
||||
break;
|
||||
case SHAVITE:
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, in, size );
|
||||
sph_shavite512_close( &ctx.shavite, hash );
|
||||
shavite512_full( &ctx.shavite, hash, in, size );
|
||||
break;
|
||||
case SIMD:
|
||||
init_sd( &ctx.simd, 512 );
|
||||
@@ -162,9 +160,7 @@ void x16rv2_hash( void* output, const void* input )
|
||||
sph_hamsi512_close( &ctx.hamsi, hash );
|
||||
break;
|
||||
case FUGUE:
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, in, size );
|
||||
sph_fugue512_close( &ctx.fugue, hash );
|
||||
sph_fugue512_full( &ctx.fugue, hash, in, size );
|
||||
break;
|
||||
case SHABAL:
|
||||
sph_shabal512_init( &ctx.shabal );
|
||||
@@ -172,9 +168,7 @@ void x16rv2_hash( void* output, const void* input )
|
||||
sph_shabal512_close( &ctx.shabal, hash );
|
||||
break;
|
||||
case WHIRLPOOL:
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, in, size );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash, in, size );
|
||||
break;
|
||||
case SHA_512:
|
||||
sph_tiger_init( &ctx.tiger );
|
||||
@@ -186,10 +180,14 @@ void x16rv2_hash( void* output, const void* input )
|
||||
SHA512_Final( (unsigned char*) hash, &ctx.sha512 );
|
||||
break;
|
||||
}
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
in = (void*) hash;
|
||||
size = 64;
|
||||
}
|
||||
memcpy(output, hash, 32);
|
||||
return 1;
|
||||
}
|
||||
|
||||
int scanhash_x16rv2( struct work *work, uint32_t max_nonce,
|
||||
@@ -227,8 +225,7 @@ int scanhash_x16rv2( struct work *work, uint32_t max_nonce,
|
||||
do
|
||||
{
|
||||
edata[19] = nonce;
|
||||
x16rv2_hash( hash32, edata );
|
||||
|
||||
if ( x16rv2_hash( hash32, edata, thr_id ) )
|
||||
if ( unlikely( valid_hash( hash32, ptarget ) && !bench ) )
|
||||
{
|
||||
pdata[19] = bswap_32( nonce );
|
||||
|
||||
@@ -30,7 +30,7 @@ union _x21s_8way_context_overlay
|
||||
|
||||
typedef union _x21s_8way_context_overlay x21s_8way_context_overlay;
|
||||
|
||||
void x21s_8way_hash( void* output, const void* input )
|
||||
int x21s_8way_hash( void* output, const void* input, int thrid )
|
||||
{
|
||||
uint32_t vhash[16*8] __attribute__ ((aligned (128)));
|
||||
uint8_t shash[64*8] __attribute__ ((aligned (64)));
|
||||
@@ -44,7 +44,8 @@ void x21s_8way_hash( void* output, const void* input )
|
||||
uint32_t *hash7 = (uint32_t*)( shash+448 );
|
||||
x21s_8way_context_overlay ctx;
|
||||
|
||||
x16r_8way_hash_generic( shash, input );
|
||||
if ( !x16r_8way_hash_generic( shash, input, thrid ) )
|
||||
return 0;
|
||||
|
||||
intrlv_8x32_512( vhash, hash0, hash1, hash2, hash3, hash4, hash5, hash6,
|
||||
hash7 );
|
||||
@@ -124,6 +125,8 @@ void x21s_8way_hash( void* output, const void* input )
|
||||
sha256_8way_init( &ctx.sha256 );
|
||||
sha256_8way_update( &ctx.sha256, vhash, 64 );
|
||||
sha256_8way_close( &ctx.sha256, output );
|
||||
|
||||
return 1;
|
||||
}
|
||||
|
||||
int scanhash_x21s_8way( struct work *work, uint32_t max_nonce,
|
||||
@@ -166,8 +169,7 @@ int scanhash_x21s_8way( struct work *work, uint32_t max_nonce,
|
||||
n+3, 0, n+2, 0, n+1, 0, n, 0 ), *noncev );
|
||||
do
|
||||
{
|
||||
x21s_8way_hash( hash, vdata );
|
||||
|
||||
if ( x21s_8way_hash( hash, vdata, thr_id ) )
|
||||
for ( int lane = 0; lane < 8; lane++ )
|
||||
if ( unlikely( hash7[lane] <= Htarg ) )
|
||||
{
|
||||
@@ -175,7 +177,7 @@ int scanhash_x21s_8way( struct work *work, uint32_t max_nonce,
|
||||
if ( likely( valid_hash( lane_hash, ptarget ) && !bench ) )
|
||||
{
|
||||
pdata[19] = bswap_32( n + lane );
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
*noncev = _mm512_add_epi32( *noncev,
|
||||
@@ -215,7 +217,7 @@ union _x21s_4way_context_overlay
|
||||
|
||||
typedef union _x21s_4way_context_overlay x21s_4way_context_overlay;
|
||||
|
||||
void x21s_4way_hash( void* output, const void* input )
|
||||
int x21s_4way_hash( void* output, const void* input, int thrid )
|
||||
{
|
||||
uint32_t vhash[16*4] __attribute__ ((aligned (64)));
|
||||
uint8_t shash[64*4] __attribute__ ((aligned (64)));
|
||||
@@ -225,8 +227,9 @@ void x21s_4way_hash( void* output, const void* input )
|
||||
uint32_t *hash2 = (uint32_t*)( shash+128 );
|
||||
uint32_t *hash3 = (uint32_t*)( shash+192 );
|
||||
|
||||
x16r_4way_hash_generic( shash, input );
|
||||
|
||||
if ( !x16r_4way_hash_generic( shash, input, thrid ) )
|
||||
return 0;
|
||||
|
||||
intrlv_4x32( vhash, hash0, hash1, hash2, hash3, 512 );
|
||||
|
||||
haval256_5_4way_init( &ctx.haval );
|
||||
@@ -299,6 +302,8 @@ void x21s_4way_hash( void* output, const void* input )
|
||||
dintrlv_4x32( output, output+32, output+64,output+96, vhash, 256 );
|
||||
|
||||
#endif
|
||||
|
||||
return 1;
|
||||
}
|
||||
|
||||
int scanhash_x21s_4way( struct work *work, uint32_t max_nonce,
|
||||
@@ -337,12 +342,12 @@ int scanhash_x21s_4way( struct work *work, uint32_t max_nonce,
|
||||
_mm256_set_epi32( n+3, 0, n+2, 0, n+1, 0, n, 0 ), *noncev );
|
||||
do
|
||||
{
|
||||
x21s_4way_hash( hash, vdata );
|
||||
if ( x21s_4way_hash( hash, vdata, thr_id ) )
|
||||
for ( int i = 0; i < 4; i++ )
|
||||
if ( unlikely( valid_hash( hash + (i<<3), ptarget ) && !bench ) )
|
||||
{
|
||||
pdata[19] = bswap_32( n+i );
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
submit_solution( work, hash+(i<<3), mythr );
|
||||
}
|
||||
*noncev = _mm256_add_epi32( *noncev,
|
||||
m256_const1_64( 0x0000000400000000 ) );
|
||||
|
||||
@@ -27,12 +27,13 @@ union _x21s_context_overlay
|
||||
};
|
||||
typedef union _x21s_context_overlay x21s_context_overlay;
|
||||
|
||||
void x21s_hash( void* output, const void* input )
|
||||
int x21s_hash( void* output, const void* input, int thrid )
|
||||
{
|
||||
uint32_t _ALIGN(128) hash[16];
|
||||
x21s_context_overlay ctx;
|
||||
|
||||
x16r_hash_generic( hash, input );
|
||||
if ( !x16r_hash_generic( hash, input, thrid ) )
|
||||
return 0;
|
||||
|
||||
sph_haval256_5_init( &ctx.haval );
|
||||
sph_haval256_5( &ctx.haval, (const void*) hash, 64) ;
|
||||
@@ -54,6 +55,8 @@ void x21s_hash( void* output, const void* input )
|
||||
SHA256_Final( (unsigned char*)hash, &ctx.sha256 );
|
||||
|
||||
memcpy( output, hash, 32 );
|
||||
|
||||
return 1;
|
||||
}
|
||||
|
||||
int scanhash_x21s( struct work *work, uint32_t max_nonce,
|
||||
@@ -87,8 +90,7 @@ int scanhash_x21s( struct work *work, uint32_t max_nonce,
|
||||
do
|
||||
{
|
||||
edata[19] = nonce;
|
||||
x21s_hash( hash32, edata );
|
||||
|
||||
if ( x21s_hash( hash32, edata, thr_id ) )
|
||||
if ( unlikely( valid_hash( hash32, ptarget ) && !bench ) )
|
||||
{
|
||||
pdata[19] = bswap_32( nonce );
|
||||
|
||||
File diff suppressed because it is too large
Load Diff
@@ -4,14 +4,14 @@ bool register_sonoa_algo( algo_gate_t* gate )
|
||||
{
|
||||
#if defined (SONOA_8WAY)
|
||||
gate->scanhash = (void*)&scanhash_sonoa_8way;
|
||||
gate->hash = (void*)&sonoa_8way_hash;
|
||||
// gate->hash = (void*)&sonoa_8way_hash;
|
||||
#elif defined (SONOA_4WAY)
|
||||
gate->scanhash = (void*)&scanhash_sonoa_4way;
|
||||
gate->hash = (void*)&sonoa_4way_hash;
|
||||
// gate->hash = (void*)&sonoa_4way_hash;
|
||||
#else
|
||||
init_sonoa_ctx();
|
||||
gate->scanhash = (void*)&scanhash_sonoa;
|
||||
gate->hash = (void*)&sonoa_hash;
|
||||
// gate->hash = (void*)&sonoa_hash;
|
||||
#endif
|
||||
gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT | AVX512_OPT | VAES_OPT;
|
||||
return true;
|
||||
|
||||
@@ -14,19 +14,19 @@ bool register_sonoa_algo( algo_gate_t* gate );
|
||||
|
||||
#if defined(SONOA_8WAY)
|
||||
|
||||
void sonoa_8way_hash( void *state, const void *input );
|
||||
int sonoa_8way_hash( void *state, const void *input, int thrid );
|
||||
int scanhash_sonoa_8way( struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr );
|
||||
|
||||
#elif defined(SONOA_4WAY)
|
||||
|
||||
void sonoa_4way_hash( void *state, const void *input );
|
||||
int sonoa_4way_hash( void *state, const void *input, int thrid );
|
||||
int scanhash_sonoa_4way( struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr );
|
||||
|
||||
#else
|
||||
|
||||
void sonoa_hash( void *state, const void *input );
|
||||
int sonoa_hash( void *state, const void *input, int thrid );
|
||||
int scanhash_sonoa( struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr );
|
||||
void init_sonoa_ctx();
|
||||
|
||||
573
algo/x17/sonoa.c
573
algo/x17/sonoa.c
@@ -83,27 +83,27 @@ void init_sonoa_ctx()
|
||||
sph_haval256_5_init(&sonoa_ctx.haval);
|
||||
};
|
||||
|
||||
void sonoa_hash( void *state, const void *input )
|
||||
int sonoa_hash( void *state, const void *input, int thrid )
|
||||
{
|
||||
uint8_t hash[128] __attribute__ ((aligned (64)));
|
||||
sonoa_ctx_holder ctx __attribute__ ((aligned (64)));
|
||||
memcpy( &ctx, &sonoa_ctx, sizeof(sonoa_ctx) );
|
||||
sonoa_ctx_holder ctx __attribute__ ((aligned (64)));
|
||||
memcpy( &ctx, &sonoa_ctx, sizeof(sonoa_ctx) );
|
||||
|
||||
sph_blake512(&ctx.blake, input, 80);
|
||||
sph_blake512(&ctx.blake, input, 80);
|
||||
sph_blake512_close(&ctx.blake, hash);
|
||||
|
||||
sph_bmw512(&ctx.bmw, hash, 64);
|
||||
sph_bmw512_close(&ctx.bmw, hash);
|
||||
|
||||
#if defined(__AES__)
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash,
|
||||
(const char*)hash, 512 );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash,
|
||||
(const char*)hash, 512 );
|
||||
#else
|
||||
sph_groestl512(&ctx.groestl, hash, 64);
|
||||
sph_groestl512_close(&ctx.groestl, hash);
|
||||
sph_groestl512(&ctx.groestl, hash, 64);
|
||||
sph_groestl512_close(&ctx.groestl, hash);
|
||||
#endif
|
||||
|
||||
sph_skein512(&ctx.skein, hash, 64);
|
||||
sph_skein512(&ctx.skein, hash, 64);
|
||||
sph_skein512_close(&ctx.skein, hash);
|
||||
|
||||
sph_jh512(&ctx.jh, hash, 64);
|
||||
@@ -112,454 +112,461 @@ void sonoa_hash( void *state, const void *input )
|
||||
sph_keccak512(&ctx.keccak, hash, 64);
|
||||
sph_keccak512_close(&ctx.keccak, hash);
|
||||
|
||||
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash,
|
||||
(const BitSequence*)hash, 64 );
|
||||
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash,
|
||||
(const BitSequence*)hash, 64 );
|
||||
|
||||
cubehashUpdateDigest( &ctx.cubehash, (byte*) hash,
|
||||
(const byte*)hash, 64 );
|
||||
cubehashUpdateDigest( &ctx.cubehash, (byte*) hash,
|
||||
(const byte*)hash, 64 );
|
||||
|
||||
sph_shavite512(&ctx.shavite, hash, 64);
|
||||
sph_shavite512_close(&ctx.shavite, hash);
|
||||
|
||||
update_final_sd( &ctx.simd, (BitSequence *)hash,
|
||||
update_final_sd( &ctx.simd, (BitSequence *)hash,
|
||||
(const BitSequence *)hash, 512 );
|
||||
|
||||
#if defined(__AES__)
|
||||
update_final_echo ( &ctx.echo, (BitSequence *)hash,
|
||||
update_final_echo ( &ctx.echo, (BitSequence *)hash,
|
||||
(const BitSequence *)hash, 512 );
|
||||
#else
|
||||
sph_echo512(&ctx.echo, hash, 64);
|
||||
sph_echo512_close(&ctx.echo, hash);
|
||||
sph_echo512(&ctx.echo, hash, 64);
|
||||
sph_echo512_close(&ctx.echo, hash);
|
||||
#endif
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
//
|
||||
|
||||
sph_bmw512_init( &ctx.bmw);
|
||||
sph_bmw512(&ctx.bmw, hash, 64);
|
||||
sph_bmw512_close(&ctx.bmw, hash);
|
||||
sph_bmw512_init( &ctx.bmw);
|
||||
sph_bmw512(&ctx.bmw, hash, 64);
|
||||
sph_bmw512_close(&ctx.bmw, hash);
|
||||
|
||||
#if defined(__AES__)
|
||||
init_groestl( &ctx.groestl, 64 );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash,
|
||||
(const char*)hash, 512 );
|
||||
init_groestl( &ctx.groestl, 64 );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash,
|
||||
(const char*)hash, 512 );
|
||||
#else
|
||||
sph_groestl512_init(&ctx.groestl );
|
||||
sph_groestl512(&ctx.groestl, hash, 64);
|
||||
sph_groestl512_close(&ctx.groestl, hash);
|
||||
sph_groestl512_init(&ctx.groestl );
|
||||
sph_groestl512(&ctx.groestl, hash, 64);
|
||||
sph_groestl512_close(&ctx.groestl, hash);
|
||||
#endif
|
||||
|
||||
sph_skein512_init( &ctx.skein);
|
||||
sph_skein512(&ctx.skein, hash, 64);
|
||||
sph_skein512_close(&ctx.skein, hash);
|
||||
sph_skein512_init( &ctx.skein);
|
||||
sph_skein512(&ctx.skein, hash, 64);
|
||||
sph_skein512_close(&ctx.skein, hash);
|
||||
|
||||
sph_jh512_init( &ctx.jh);
|
||||
sph_jh512(&ctx.jh, hash, 64);
|
||||
sph_jh512_close(&ctx.jh, hash);
|
||||
sph_jh512_init( &ctx.jh);
|
||||
sph_jh512(&ctx.jh, hash, 64);
|
||||
sph_jh512_close(&ctx.jh, hash);
|
||||
|
||||
sph_keccak512_init( &ctx.keccak );
|
||||
sph_keccak512(&ctx.keccak, hash, 64);
|
||||
sph_keccak512_close(&ctx.keccak, hash);
|
||||
sph_keccak512_init( &ctx.keccak );
|
||||
sph_keccak512(&ctx.keccak, hash, 64);
|
||||
sph_keccak512_close(&ctx.keccak, hash);
|
||||
|
||||
init_luffa( &ctx.luffa, 512 );
|
||||
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash,
|
||||
(const BitSequence*)hash, 64 );
|
||||
init_luffa( &ctx.luffa, 512 );
|
||||
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash,
|
||||
(const BitSequence*)hash, 64 );
|
||||
|
||||
cubehashInit( &ctx.cubehash, 512, 16, 32 );
|
||||
cubehashUpdateDigest( &ctx.cubehash, (byte*) hash,
|
||||
(const byte*)hash, 64 );
|
||||
cubehashInit( &ctx.cubehash, 512, 16, 32 );
|
||||
cubehashUpdateDigest( &ctx.cubehash, (byte*) hash,
|
||||
(const byte*)hash, 64 );
|
||||
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512(&ctx.shavite, hash, 64);
|
||||
sph_shavite512_close(&ctx.shavite, hash);
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512(&ctx.shavite, hash, 64);
|
||||
sph_shavite512_close(&ctx.shavite, hash);
|
||||
|
||||
init_sd( &ctx.simd, 512 );
|
||||
update_final_sd( &ctx.simd, (BitSequence *)hash,
|
||||
init_sd( &ctx.simd, 512 );
|
||||
update_final_sd( &ctx.simd, (BitSequence *)hash,
|
||||
(const BitSequence *)hash, 512 );
|
||||
|
||||
#if defined(__AES__)
|
||||
init_echo( &ctx.echo, 512 );
|
||||
update_final_echo ( &ctx.echo, (BitSequence *)hash,
|
||||
init_echo( &ctx.echo, 512 );
|
||||
update_final_echo ( &ctx.echo, (BitSequence *)hash,
|
||||
(const BitSequence *)hash, 512 );
|
||||
#else
|
||||
sph_echo512_init( &ctx.echo );
|
||||
sph_echo512(&ctx.echo, hash, 64);
|
||||
sph_echo512_close(&ctx.echo, hash);
|
||||
sph_echo512_init( &ctx.echo );
|
||||
sph_echo512(&ctx.echo, hash, 64);
|
||||
sph_echo512_close(&ctx.echo, hash);
|
||||
#endif
|
||||
|
||||
sph_hamsi512(&ctx.hamsi, hash, 64);
|
||||
sph_hamsi512_close(&ctx.hamsi, hash);
|
||||
sph_hamsi512(&ctx.hamsi, hash, 64);
|
||||
sph_hamsi512_close(&ctx.hamsi, hash);
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
//
|
||||
|
||||
sph_bmw512_init( &ctx.bmw);
|
||||
sph_bmw512(&ctx.bmw, hash, 64);
|
||||
sph_bmw512_close(&ctx.bmw, hash);
|
||||
sph_bmw512_init( &ctx.bmw);
|
||||
sph_bmw512(&ctx.bmw, hash, 64);
|
||||
sph_bmw512_close(&ctx.bmw, hash);
|
||||
|
||||
#if defined(__AES__)
|
||||
init_groestl( &ctx.groestl, 64 );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash,
|
||||
(const char*)hash, 512 );
|
||||
init_groestl( &ctx.groestl, 64 );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash,
|
||||
(const char*)hash, 512 );
|
||||
#else
|
||||
sph_groestl512_init(&ctx.groestl );
|
||||
sph_groestl512(&ctx.groestl, hash, 64);
|
||||
sph_groestl512_close(&ctx.groestl, hash);
|
||||
sph_groestl512_init(&ctx.groestl );
|
||||
sph_groestl512(&ctx.groestl, hash, 64);
|
||||
sph_groestl512_close(&ctx.groestl, hash);
|
||||
#endif
|
||||
|
||||
sph_skein512_init( &ctx.skein);
|
||||
sph_skein512(&ctx.skein, hash, 64);
|
||||
sph_skein512_close(&ctx.skein, hash);
|
||||
sph_skein512_init( &ctx.skein);
|
||||
sph_skein512(&ctx.skein, hash, 64);
|
||||
sph_skein512_close(&ctx.skein, hash);
|
||||
|
||||
sph_jh512_init( &ctx.jh);
|
||||
sph_jh512(&ctx.jh, hash, 64);
|
||||
sph_jh512_close(&ctx.jh, hash);
|
||||
sph_jh512_init( &ctx.jh);
|
||||
sph_jh512(&ctx.jh, hash, 64);
|
||||
sph_jh512_close(&ctx.jh, hash);
|
||||
|
||||
sph_keccak512_init( &ctx.keccak );
|
||||
sph_keccak512(&ctx.keccak, hash, 64);
|
||||
sph_keccak512_close(&ctx.keccak, hash);
|
||||
sph_keccak512_init( &ctx.keccak );
|
||||
sph_keccak512(&ctx.keccak, hash, 64);
|
||||
sph_keccak512_close(&ctx.keccak, hash);
|
||||
|
||||
init_luffa( &ctx.luffa, 512 );
|
||||
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash,
|
||||
(const BitSequence*)hash, 64 );
|
||||
init_luffa( &ctx.luffa, 512 );
|
||||
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash,
|
||||
(const BitSequence*)hash, 64 );
|
||||
|
||||
cubehashInit( &ctx.cubehash, 512, 16, 32 );
|
||||
cubehashUpdateDigest( &ctx.cubehash, (byte*) hash,
|
||||
(const byte*)hash, 64 );
|
||||
cubehashInit( &ctx.cubehash, 512, 16, 32 );
|
||||
cubehashUpdateDigest( &ctx.cubehash, (byte*)hash,
|
||||
(const byte*)hash, 64 );
|
||||
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512(&ctx.shavite, hash, 64);
|
||||
sph_shavite512_close(&ctx.shavite, hash);
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512(&ctx.shavite, hash, 64);
|
||||
sph_shavite512_close(&ctx.shavite, hash);
|
||||
|
||||
init_sd( &ctx.simd, 512 );
|
||||
update_final_sd( &ctx.simd, (BitSequence *)hash,
|
||||
init_sd( &ctx.simd, 512 );
|
||||
update_final_sd( &ctx.simd, (BitSequence *)hash,
|
||||
(const BitSequence *)hash, 512 );
|
||||
|
||||
#if defined(__AES__)
|
||||
init_echo( &ctx.echo, 512 );
|
||||
update_final_echo ( &ctx.echo, (BitSequence *)hash,
|
||||
init_echo( &ctx.echo, 512 );
|
||||
update_final_echo ( &ctx.echo, (BitSequence *)hash,
|
||||
(const BitSequence *)hash, 512 );
|
||||
#else
|
||||
sph_echo512_init( &ctx.echo );
|
||||
sph_echo512(&ctx.echo, hash, 64);
|
||||
sph_echo512_close(&ctx.echo, hash);
|
||||
sph_echo512_init( &ctx.echo );
|
||||
sph_echo512(&ctx.echo, hash, 64);
|
||||
sph_echo512_close(&ctx.echo, hash);
|
||||
#endif
|
||||
|
||||
sph_hamsi512_init( &ctx.hamsi );
|
||||
sph_hamsi512(&ctx.hamsi, hash, 64);
|
||||
sph_hamsi512_close(&ctx.hamsi, hash);
|
||||
sph_hamsi512_init( &ctx.hamsi );
|
||||
sph_hamsi512(&ctx.hamsi, hash, 64);
|
||||
sph_hamsi512_close(&ctx.hamsi, hash);
|
||||
|
||||
sph_fugue512(&ctx.fugue, hash, 64);
|
||||
sph_fugue512_close(&ctx.fugue, hash);
|
||||
sph_fugue512(&ctx.fugue, hash, 64);
|
||||
sph_fugue512_close(&ctx.fugue, hash);
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
//
|
||||
|
||||
sph_bmw512_init( &ctx.bmw);
|
||||
sph_bmw512(&ctx.bmw, hash, 64);
|
||||
sph_bmw512_close(&ctx.bmw, hash);
|
||||
sph_bmw512_init( &ctx.bmw);
|
||||
sph_bmw512(&ctx.bmw, hash, 64);
|
||||
sph_bmw512_close(&ctx.bmw, hash);
|
||||
|
||||
#if defined(__AES__)
|
||||
init_groestl( &ctx.groestl, 64 );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash,
|
||||
init_groestl( &ctx.groestl, 64 );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash,
|
||||
(const char*)hash, 512 );
|
||||
#else
|
||||
sph_groestl512_init(&ctx.groestl );
|
||||
sph_groestl512(&ctx.groestl, hash, 64);
|
||||
sph_groestl512_close(&ctx.groestl, hash);
|
||||
sph_groestl512_init(&ctx.groestl );
|
||||
sph_groestl512(&ctx.groestl, hash, 64);
|
||||
sph_groestl512_close(&ctx.groestl, hash);
|
||||
#endif
|
||||
|
||||
sph_skein512_init( &ctx.skein);
|
||||
sph_skein512(&ctx.skein, hash, 64);
|
||||
sph_skein512_close(&ctx.skein, hash);
|
||||
sph_skein512_init( &ctx.skein);
|
||||
sph_skein512(&ctx.skein, hash, 64);
|
||||
sph_skein512_close(&ctx.skein, hash);
|
||||
|
||||
sph_jh512_init( &ctx.jh);
|
||||
sph_jh512(&ctx.jh, hash, 64);
|
||||
sph_jh512_close(&ctx.jh, hash);
|
||||
sph_jh512_init( &ctx.jh);
|
||||
sph_jh512(&ctx.jh, hash, 64);
|
||||
sph_jh512_close(&ctx.jh, hash);
|
||||
|
||||
sph_keccak512_init( &ctx.keccak );
|
||||
sph_keccak512(&ctx.keccak, hash, 64);
|
||||
sph_keccak512_close(&ctx.keccak, hash);
|
||||
sph_keccak512_init( &ctx.keccak );
|
||||
sph_keccak512(&ctx.keccak, hash, 64);
|
||||
sph_keccak512_close(&ctx.keccak, hash);
|
||||
|
||||
init_luffa( &ctx.luffa, 512 );
|
||||
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash,
|
||||
init_luffa( &ctx.luffa, 512 );
|
||||
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash,
|
||||
(const BitSequence*)hash, 64 );
|
||||
|
||||
cubehashInit( &ctx.cubehash, 512, 16, 32 );
|
||||
cubehashUpdateDigest( &ctx.cubehash, (byte*) hash,
|
||||
cubehashInit( &ctx.cubehash, 512, 16, 32 );
|
||||
cubehashUpdateDigest( &ctx.cubehash, (byte*) hash,
|
||||
(const byte*)hash, 64 );
|
||||
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512(&ctx.shavite, hash, 64);
|
||||
sph_shavite512_close(&ctx.shavite, hash);
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512(&ctx.shavite, hash, 64);
|
||||
sph_shavite512_close(&ctx.shavite, hash);
|
||||
|
||||
init_sd( &ctx.simd, 512 );
|
||||
update_final_sd( &ctx.simd, (BitSequence *)hash,
|
||||
init_sd( &ctx.simd, 512 );
|
||||
update_final_sd( &ctx.simd, (BitSequence *)hash,
|
||||
(const BitSequence *)hash, 512 );
|
||||
|
||||
#if defined(__AES__)
|
||||
init_echo( &ctx.echo, 512 );
|
||||
update_final_echo ( &ctx.echo, (BitSequence *)hash,
|
||||
init_echo( &ctx.echo, 512 );
|
||||
update_final_echo ( &ctx.echo, (BitSequence *)hash,
|
||||
(const BitSequence *)hash, 512 );
|
||||
#else
|
||||
sph_echo512_init( &ctx.echo );
|
||||
sph_echo512(&ctx.echo, hash, 64);
|
||||
sph_echo512_close(&ctx.echo, hash);
|
||||
sph_echo512_init( &ctx.echo );
|
||||
sph_echo512(&ctx.echo, hash, 64);
|
||||
sph_echo512_close(&ctx.echo, hash);
|
||||
#endif
|
||||
|
||||
sph_hamsi512_init( &ctx.hamsi );
|
||||
sph_hamsi512(&ctx.hamsi, hash, 64);
|
||||
sph_hamsi512_close(&ctx.hamsi, hash);
|
||||
sph_hamsi512_init( &ctx.hamsi );
|
||||
sph_hamsi512(&ctx.hamsi, hash, 64);
|
||||
sph_hamsi512_close(&ctx.hamsi, hash);
|
||||
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512(&ctx.fugue, hash, 64);
|
||||
sph_fugue512_close(&ctx.fugue, hash);
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512(&ctx.fugue, hash, 64);
|
||||
sph_fugue512_close(&ctx.fugue, hash);
|
||||
|
||||
sph_shabal512(&ctx.shabal, hash, 64);
|
||||
sph_shabal512_close(&ctx.shabal, hash);
|
||||
sph_shabal512(&ctx.shabal, hash, 64);
|
||||
sph_shabal512_close(&ctx.shabal, hash);
|
||||
|
||||
sph_hamsi512_init( &ctx.hamsi );
|
||||
sph_hamsi512(&ctx.hamsi, hash, 64);
|
||||
sph_hamsi512_close(&ctx.hamsi, hash);
|
||||
sph_hamsi512_init( &ctx.hamsi );
|
||||
sph_hamsi512(&ctx.hamsi, hash, 64);
|
||||
sph_hamsi512_close(&ctx.hamsi, hash);
|
||||
|
||||
#if defined(__AES__)
|
||||
init_echo( &ctx.echo, 512 );
|
||||
update_final_echo ( &ctx.echo, (BitSequence *)hash,
|
||||
init_echo( &ctx.echo, 512 );
|
||||
update_final_echo ( &ctx.echo, (BitSequence *)hash,
|
||||
(const BitSequence *)hash, 512 );
|
||||
#else
|
||||
sph_echo512_init( &ctx.echo );
|
||||
sph_echo512(&ctx.echo, hash, 64);
|
||||
sph_echo512_close(&ctx.echo, hash);
|
||||
sph_echo512_init( &ctx.echo );
|
||||
sph_echo512(&ctx.echo, hash, 64);
|
||||
sph_echo512_close(&ctx.echo, hash);
|
||||
#endif
|
||||
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512(&ctx.shavite, hash, 64);
|
||||
sph_shavite512_close(&ctx.shavite, hash);
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512(&ctx.shavite, hash, 64);
|
||||
sph_shavite512_close(&ctx.shavite, hash);
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
//
|
||||
|
||||
sph_bmw512_init( &ctx.bmw);
|
||||
sph_bmw512(&ctx.bmw, hash, 64);
|
||||
sph_bmw512_close(&ctx.bmw, hash);
|
||||
sph_bmw512_init( &ctx.bmw);
|
||||
sph_bmw512(&ctx.bmw, hash, 64);
|
||||
sph_bmw512_close(&ctx.bmw, hash);
|
||||
|
||||
sph_shabal512_init( &ctx.shabal );
|
||||
sph_shabal512_init( &ctx.shabal );
|
||||
sph_shabal512(&ctx.shabal, hash, 64);
|
||||
sph_shabal512_close(&ctx.shabal, hash);
|
||||
sph_shabal512_close(&ctx.shabal, hash);
|
||||
|
||||
#if defined(__AES__)
|
||||
init_groestl( &ctx.groestl, 64 );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash,
|
||||
init_groestl( &ctx.groestl, 64 );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash,
|
||||
(const char*)hash, 512 );
|
||||
#else
|
||||
sph_groestl512_init(&ctx.groestl );
|
||||
sph_groestl512(&ctx.groestl, hash, 64);
|
||||
sph_groestl512_close(&ctx.groestl, hash);
|
||||
sph_groestl512_init(&ctx.groestl );
|
||||
sph_groestl512(&ctx.groestl, hash, 64);
|
||||
sph_groestl512_close(&ctx.groestl, hash);
|
||||
#endif
|
||||
|
||||
sph_skein512_init( &ctx.skein);
|
||||
sph_skein512(&ctx.skein, hash, 64);
|
||||
sph_skein512_close(&ctx.skein, hash);
|
||||
sph_skein512_init( &ctx.skein);
|
||||
sph_skein512(&ctx.skein, hash, 64);
|
||||
sph_skein512_close(&ctx.skein, hash);
|
||||
|
||||
sph_jh512_init( &ctx.jh);
|
||||
sph_jh512(&ctx.jh, hash, 64);
|
||||
sph_jh512_close(&ctx.jh, hash);
|
||||
sph_jh512_init( &ctx.jh);
|
||||
sph_jh512(&ctx.jh, hash, 64);
|
||||
sph_jh512_close(&ctx.jh, hash);
|
||||
|
||||
sph_keccak512_init( &ctx.keccak );
|
||||
sph_keccak512(&ctx.keccak, hash, 64);
|
||||
sph_keccak512_close(&ctx.keccak, hash);
|
||||
sph_keccak512_init( &ctx.keccak );
|
||||
sph_keccak512(&ctx.keccak, hash, 64);
|
||||
sph_keccak512_close(&ctx.keccak, hash);
|
||||
|
||||
init_luffa( &ctx.luffa, 512 );
|
||||
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash,
|
||||
init_luffa( &ctx.luffa, 512 );
|
||||
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash,
|
||||
(const BitSequence*)hash, 64 );
|
||||
|
||||
cubehashInit( &ctx.cubehash, 512, 16, 32 );
|
||||
cubehashUpdateDigest( &ctx.cubehash, (byte*) hash,
|
||||
cubehashInit( &ctx.cubehash, 512, 16, 32 );
|
||||
cubehashUpdateDigest( &ctx.cubehash, (byte*) hash,
|
||||
(const byte*)hash, 64 );
|
||||
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512(&ctx.shavite, hash, 64);
|
||||
sph_shavite512_close(&ctx.shavite, hash);
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512(&ctx.shavite, hash, 64);
|
||||
sph_shavite512_close(&ctx.shavite, hash);
|
||||
|
||||
init_sd( &ctx.simd, 512 );
|
||||
update_final_sd( &ctx.simd, (BitSequence *)hash,
|
||||
init_sd( &ctx.simd, 512 );
|
||||
update_final_sd( &ctx.simd, (BitSequence *)hash,
|
||||
(const BitSequence *)hash, 512 );
|
||||
|
||||
#if defined(__AES__)
|
||||
init_echo( &ctx.echo, 512 );
|
||||
update_final_echo ( &ctx.echo, (BitSequence *)hash,
|
||||
init_echo( &ctx.echo, 512 );
|
||||
update_final_echo ( &ctx.echo, (BitSequence *)hash,
|
||||
(const BitSequence *)hash, 512 );
|
||||
#else
|
||||
sph_echo512_init( &ctx.echo );
|
||||
sph_echo512(&ctx.echo, hash, 64);
|
||||
sph_echo512_close(&ctx.echo, hash);
|
||||
sph_echo512_init( &ctx.echo );
|
||||
sph_echo512(&ctx.echo, hash, 64);
|
||||
sph_echo512_close(&ctx.echo, hash);
|
||||
#endif
|
||||
|
||||
sph_hamsi512_init( &ctx.hamsi );
|
||||
sph_hamsi512(&ctx.hamsi, hash, 64);
|
||||
sph_hamsi512_close(&ctx.hamsi, hash);
|
||||
sph_hamsi512_init( &ctx.hamsi );
|
||||
sph_hamsi512(&ctx.hamsi, hash, 64);
|
||||
sph_hamsi512_close(&ctx.hamsi, hash);
|
||||
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512(&ctx.fugue, hash, 64);
|
||||
sph_fugue512_close(&ctx.fugue, hash);
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512(&ctx.fugue, hash, 64);
|
||||
sph_fugue512_close(&ctx.fugue, hash);
|
||||
|
||||
sph_shabal512_init( &ctx.shabal );
|
||||
sph_shabal512(&ctx.shabal, hash, 64);
|
||||
sph_shabal512_close(&ctx.shabal, hash);
|
||||
sph_shabal512_init( &ctx.shabal );
|
||||
sph_shabal512(&ctx.shabal, hash, 64);
|
||||
sph_shabal512_close(&ctx.shabal, hash);
|
||||
|
||||
sph_whirlpool(&ctx.whirlpool, hash, 64);
|
||||
sph_whirlpool_close(&ctx.whirlpool, hash);
|
||||
sph_whirlpool(&ctx.whirlpool, hash, 64);
|
||||
sph_whirlpool_close(&ctx.whirlpool, hash);
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
//
|
||||
sph_bmw512_init( &ctx.bmw);
|
||||
sph_bmw512(&ctx.bmw, hash, 64);
|
||||
sph_bmw512_close(&ctx.bmw, hash);
|
||||
sph_bmw512_init( &ctx.bmw);
|
||||
sph_bmw512(&ctx.bmw, hash, 64);
|
||||
sph_bmw512_close(&ctx.bmw, hash);
|
||||
|
||||
#if defined(__AES__)
|
||||
init_groestl( &ctx.groestl, 64 );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash,
|
||||
init_groestl( &ctx.groestl, 64 );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash,
|
||||
(const char*)hash, 512 );
|
||||
#else
|
||||
sph_groestl512_init(&ctx.groestl );
|
||||
sph_groestl512(&ctx.groestl, hash, 64);
|
||||
sph_groestl512_close(&ctx.groestl, hash);
|
||||
sph_groestl512_init(&ctx.groestl );
|
||||
sph_groestl512(&ctx.groestl, hash, 64);
|
||||
sph_groestl512_close(&ctx.groestl, hash);
|
||||
#endif
|
||||
|
||||
sph_skein512_init( &ctx.skein);
|
||||
sph_skein512(&ctx.skein, hash, 64);
|
||||
sph_skein512_close(&ctx.skein, hash);
|
||||
sph_skein512_init( &ctx.skein);
|
||||
sph_skein512(&ctx.skein, hash, 64);
|
||||
sph_skein512_close(&ctx.skein, hash);
|
||||
|
||||
sph_jh512_init( &ctx.jh);
|
||||
sph_jh512(&ctx.jh, hash, 64);
|
||||
sph_jh512_close(&ctx.jh, hash);
|
||||
sph_jh512_init( &ctx.jh);
|
||||
sph_jh512(&ctx.jh, hash, 64);
|
||||
sph_jh512_close(&ctx.jh, hash);
|
||||
|
||||
sph_keccak512_init( &ctx.keccak );
|
||||
sph_keccak512(&ctx.keccak, hash, 64);
|
||||
sph_keccak512_close(&ctx.keccak, hash);
|
||||
sph_keccak512_init( &ctx.keccak );
|
||||
sph_keccak512(&ctx.keccak, hash, 64);
|
||||
sph_keccak512_close(&ctx.keccak, hash);
|
||||
|
||||
init_luffa( &ctx.luffa, 512 );
|
||||
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash,
|
||||
init_luffa( &ctx.luffa, 512 );
|
||||
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash,
|
||||
(const BitSequence*)hash, 64 );
|
||||
|
||||
cubehashInit( &ctx.cubehash, 512, 16, 32 );
|
||||
cubehashUpdateDigest( &ctx.cubehash, (byte*) hash,
|
||||
cubehashInit( &ctx.cubehash, 512, 16, 32 );
|
||||
cubehashUpdateDigest( &ctx.cubehash, (byte*) hash,
|
||||
(const byte*)hash, 64 );
|
||||
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512(&ctx.shavite, hash, 64);
|
||||
sph_shavite512_close(&ctx.shavite, hash);
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512(&ctx.shavite, hash, 64);
|
||||
sph_shavite512_close(&ctx.shavite, hash);
|
||||
|
||||
init_sd( &ctx.simd, 512 );
|
||||
update_final_sd( &ctx.simd, (BitSequence *)hash,
|
||||
init_sd( &ctx.simd, 512 );
|
||||
update_final_sd( &ctx.simd, (BitSequence *)hash,
|
||||
(const BitSequence *)hash, 512 );
|
||||
|
||||
#if defined(__AES__)
|
||||
init_echo( &ctx.echo, 512 );
|
||||
update_final_echo ( &ctx.echo, (BitSequence *)hash,
|
||||
init_echo( &ctx.echo, 512 );
|
||||
update_final_echo ( &ctx.echo, (BitSequence *)hash,
|
||||
(const BitSequence *)hash, 512 );
|
||||
#else
|
||||
sph_echo512_init( &ctx.echo );
|
||||
sph_echo512(&ctx.echo, hash, 64);
|
||||
sph_echo512_close(&ctx.echo, hash);
|
||||
sph_echo512_init( &ctx.echo );
|
||||
sph_echo512(&ctx.echo, hash, 64);
|
||||
sph_echo512_close(&ctx.echo, hash);
|
||||
#endif
|
||||
|
||||
sph_hamsi512_init( &ctx.hamsi );
|
||||
sph_hamsi512(&ctx.hamsi, hash, 64);
|
||||
sph_hamsi512_close(&ctx.hamsi, hash);
|
||||
sph_hamsi512_init( &ctx.hamsi );
|
||||
sph_hamsi512(&ctx.hamsi, hash, 64);
|
||||
sph_hamsi512_close(&ctx.hamsi, hash);
|
||||
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512(&ctx.fugue, hash, 64);
|
||||
sph_fugue512_close(&ctx.fugue, hash);
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512(&ctx.fugue, hash, 64);
|
||||
sph_fugue512_close(&ctx.fugue, hash);
|
||||
|
||||
sph_shabal512_init( &ctx.shabal );
|
||||
sph_shabal512(&ctx.shabal, hash, 64);
|
||||
sph_shabal512_close(&ctx.shabal, hash);
|
||||
sph_shabal512_init( &ctx.shabal );
|
||||
sph_shabal512(&ctx.shabal, hash, 64);
|
||||
sph_shabal512_close(&ctx.shabal, hash);
|
||||
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool(&ctx.whirlpool, hash, 64);
|
||||
sph_whirlpool_close(&ctx.whirlpool, hash);
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool(&ctx.whirlpool, hash, 64);
|
||||
sph_whirlpool_close(&ctx.whirlpool, hash);
|
||||
|
||||
SHA512_Update( &ctx.sha512, hash, 64 );
|
||||
SHA512_Final( (unsigned char*) hash, &ctx.sha512 );
|
||||
SHA512_Update( &ctx.sha512, hash, 64 );
|
||||
SHA512_Final( (unsigned char*) hash, &ctx.sha512 );
|
||||
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool(&ctx.whirlpool, hash, 64);
|
||||
sph_whirlpool_close(&ctx.whirlpool, hash);
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool(&ctx.whirlpool, hash, 64);
|
||||
sph_whirlpool_close(&ctx.whirlpool, hash);
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
//
|
||||
|
||||
sph_bmw512_init( &ctx.bmw);
|
||||
sph_bmw512(&ctx.bmw, hash, 64);
|
||||
sph_bmw512_close(&ctx.bmw, hash);
|
||||
sph_bmw512_init( &ctx.bmw);
|
||||
sph_bmw512(&ctx.bmw, hash, 64);
|
||||
sph_bmw512_close(&ctx.bmw, hash);
|
||||
|
||||
#if defined(__AES__)
|
||||
init_groestl( &ctx.groestl, 64 );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash,
|
||||
init_groestl( &ctx.groestl, 64 );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash,
|
||||
(const char*)hash, 512 );
|
||||
#else
|
||||
sph_groestl512_init(&ctx.groestl );
|
||||
sph_groestl512(&ctx.groestl, hash, 64);
|
||||
sph_groestl512_close(&ctx.groestl, hash);
|
||||
sph_groestl512_init(&ctx.groestl );
|
||||
sph_groestl512(&ctx.groestl, hash, 64);
|
||||
sph_groestl512_close(&ctx.groestl, hash);
|
||||
#endif
|
||||
|
||||
sph_skein512_init( &ctx.skein);
|
||||
sph_skein512(&ctx.skein, hash, 64);
|
||||
sph_skein512_close(&ctx.skein, hash);
|
||||
sph_skein512_init( &ctx.skein);
|
||||
sph_skein512(&ctx.skein, hash, 64);
|
||||
sph_skein512_close(&ctx.skein, hash);
|
||||
|
||||
sph_jh512_init( &ctx.jh);
|
||||
sph_jh512(&ctx.jh, hash, 64);
|
||||
sph_jh512_close(&ctx.jh, hash);
|
||||
sph_jh512_init( &ctx.jh);
|
||||
sph_jh512(&ctx.jh, hash, 64);
|
||||
sph_jh512_close(&ctx.jh, hash);
|
||||
|
||||
sph_keccak512_init( &ctx.keccak );
|
||||
sph_keccak512(&ctx.keccak, hash, 64);
|
||||
sph_keccak512_close(&ctx.keccak, hash);
|
||||
sph_keccak512_init( &ctx.keccak );
|
||||
sph_keccak512(&ctx.keccak, hash, 64);
|
||||
sph_keccak512_close(&ctx.keccak, hash);
|
||||
|
||||
init_luffa( &ctx.luffa, 512 );
|
||||
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash,
|
||||
init_luffa( &ctx.luffa, 512 );
|
||||
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash,
|
||||
(const BitSequence*)hash, 64 );
|
||||
|
||||
cubehashInit( &ctx.cubehash, 512, 16, 32 );
|
||||
cubehashUpdateDigest( &ctx.cubehash, (byte*) hash,
|
||||
cubehashInit( &ctx.cubehash, 512, 16, 32 );
|
||||
cubehashUpdateDigest( &ctx.cubehash, (byte*) hash,
|
||||
(const byte*)hash, 64 );
|
||||
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512(&ctx.shavite, hash, 64);
|
||||
sph_shavite512_close(&ctx.shavite, hash);
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512(&ctx.shavite, hash, 64);
|
||||
sph_shavite512_close(&ctx.shavite, hash);
|
||||
|
||||
init_sd( &ctx.simd, 512 );
|
||||
update_final_sd( &ctx.simd, (BitSequence *)hash,
|
||||
init_sd( &ctx.simd, 512 );
|
||||
update_final_sd( &ctx.simd, (BitSequence *)hash,
|
||||
(const BitSequence *)hash, 512 );
|
||||
|
||||
#if defined(__AES__)
|
||||
init_echo( &ctx.echo, 512 );
|
||||
update_final_echo ( &ctx.echo, (BitSequence *)hash,
|
||||
init_echo( &ctx.echo, 512 );
|
||||
update_final_echo ( &ctx.echo, (BitSequence *)hash,
|
||||
(const BitSequence *)hash, 512 );
|
||||
#else
|
||||
sph_echo512_init( &ctx.echo );
|
||||
sph_echo512(&ctx.echo, hash, 64);
|
||||
sph_echo512_close(&ctx.echo, hash);
|
||||
sph_echo512_init( &ctx.echo );
|
||||
sph_echo512(&ctx.echo, hash, 64);
|
||||
sph_echo512_close(&ctx.echo, hash);
|
||||
#endif
|
||||
|
||||
sph_hamsi512_init( &ctx.hamsi );
|
||||
sph_hamsi512(&ctx.hamsi, hash, 64);
|
||||
sph_hamsi512_close(&ctx.hamsi, hash);
|
||||
sph_hamsi512_init( &ctx.hamsi );
|
||||
sph_hamsi512(&ctx.hamsi, hash, 64);
|
||||
sph_hamsi512_close(&ctx.hamsi, hash);
|
||||
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512(&ctx.fugue, hash, 64);
|
||||
sph_fugue512_close(&ctx.fugue, hash);
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512(&ctx.fugue, hash, 64);
|
||||
sph_fugue512_close(&ctx.fugue, hash);
|
||||
|
||||
sph_shabal512_init( &ctx.shabal );
|
||||
sph_shabal512(&ctx.shabal, hash, 64);
|
||||
sph_shabal512_close(&ctx.shabal, hash);
|
||||
sph_shabal512_init( &ctx.shabal );
|
||||
sph_shabal512(&ctx.shabal, hash, 64);
|
||||
sph_shabal512_close(&ctx.shabal, hash);
|
||||
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool(&ctx.whirlpool, hash, 64);
|
||||
sph_whirlpool_close(&ctx.whirlpool, hash);
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool(&ctx.whirlpool, hash, 64);
|
||||
sph_whirlpool_close(&ctx.whirlpool, hash);
|
||||
|
||||
SHA512_Init( &ctx.sha512 );
|
||||
SHA512_Update( &ctx.sha512, hash, 64 );
|
||||
SHA512_Final( (unsigned char*) hash, &ctx.sha512 );
|
||||
SHA512_Init( &ctx.sha512 );
|
||||
SHA512_Update( &ctx.sha512, hash, 64 );
|
||||
SHA512_Final( (unsigned char*) hash, &ctx.sha512 );
|
||||
|
||||
sph_haval256_5(&ctx.haval,(const void*) hash, 64);
|
||||
sph_haval256_5_close(&ctx.haval, hash);
|
||||
sph_haval256_5(&ctx.haval,(const void*) hash, 64);
|
||||
sph_haval256_5_close(&ctx.haval, hash);
|
||||
|
||||
memcpy(state, hash, 32);
|
||||
return 1;
|
||||
}
|
||||
|
||||
int scanhash_sonoa( struct work *work, uint32_t max_nonce,
|
||||
@@ -579,7 +586,7 @@ int scanhash_sonoa( struct work *work, uint32_t max_nonce,
|
||||
do
|
||||
{
|
||||
edata[19] = n;
|
||||
sonoa_hash( hash64, edata );
|
||||
if ( sonoa_hash( hash64, edata, thr_id ) )
|
||||
if ( unlikely( valid_hash( hash64, ptarget ) && !bench ) )
|
||||
{
|
||||
pdata[19] = bswap_32( n );
|
||||
|
||||
@@ -132,30 +132,14 @@ void x17_8way_hash( void *state, const void *input )
|
||||
dintrlv_4x128_512( hash0, hash1, hash2, hash3, vhashA );
|
||||
dintrlv_4x128_512( hash4, hash5, hash6, hash7, vhashB );
|
||||
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, hash0, 64 );
|
||||
sph_shavite512_close( &ctx.shavite, hash0 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, hash1, 64 );
|
||||
sph_shavite512_close( &ctx.shavite, hash1 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, hash2, 64 );
|
||||
sph_shavite512_close( &ctx.shavite, hash2 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, hash3, 64 );
|
||||
sph_shavite512_close( &ctx.shavite, hash3 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, hash4, 64 );
|
||||
sph_shavite512_close( &ctx.shavite, hash4 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, hash5, 64 );
|
||||
sph_shavite512_close( &ctx.shavite, hash5 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, hash6, 64 );
|
||||
sph_shavite512_close( &ctx.shavite, hash6 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, hash7, 64 );
|
||||
sph_shavite512_close( &ctx.shavite, hash7 );
|
||||
shavite512_full( &ctx.shavite, hash0, hash0, 64 );
|
||||
shavite512_full( &ctx.shavite, hash1, hash1, 64 );
|
||||
shavite512_full( &ctx.shavite, hash2, hash2, 64 );
|
||||
shavite512_full( &ctx.shavite, hash3, hash3, 64 );
|
||||
shavite512_full( &ctx.shavite, hash4, hash4, 64 );
|
||||
shavite512_full( &ctx.shavite, hash5, hash5, 64 );
|
||||
shavite512_full( &ctx.shavite, hash6, hash6, 64 );
|
||||
shavite512_full( &ctx.shavite, hash7, hash7, 64 );
|
||||
|
||||
intrlv_4x128_512( vhashA, hash0, hash1, hash2, hash3 );
|
||||
intrlv_4x128_512( vhashB, hash4, hash5, hash6, hash7 );
|
||||
@@ -206,30 +190,14 @@ void x17_8way_hash( void *state, const void *input )
|
||||
dintrlv_8x64_512( hash0, hash1, hash2, hash3, hash4, hash5, hash6, hash7,
|
||||
vhash );
|
||||
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash0, 64 );
|
||||
sph_fugue512_close( &ctx.fugue, hash0 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash1, 64 );
|
||||
sph_fugue512_close( &ctx.fugue, hash1 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash2, 64 );
|
||||
sph_fugue512_close( &ctx.fugue, hash2 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash3, 64 );
|
||||
sph_fugue512_close( &ctx.fugue, hash3 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash4, 64 );
|
||||
sph_fugue512_close( &ctx.fugue, hash4 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash5, 64 );
|
||||
sph_fugue512_close( &ctx.fugue, hash5 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash6, 64 );
|
||||
sph_fugue512_close( &ctx.fugue, hash6 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash7, 64 );
|
||||
sph_fugue512_close( &ctx.fugue, hash7 );
|
||||
sph_fugue512_full( &ctx.fugue, hash0, hash0, 64 );
|
||||
sph_fugue512_full( &ctx.fugue, hash1, hash1, 64 );
|
||||
sph_fugue512_full( &ctx.fugue, hash2, hash2, 64 );
|
||||
sph_fugue512_full( &ctx.fugue, hash3, hash3, 64 );
|
||||
sph_fugue512_full( &ctx.fugue, hash4, hash4, 64 );
|
||||
sph_fugue512_full( &ctx.fugue, hash5, hash5, 64 );
|
||||
sph_fugue512_full( &ctx.fugue, hash6, hash6, 64 );
|
||||
sph_fugue512_full( &ctx.fugue, hash7, hash7, 64 );
|
||||
|
||||
intrlv_8x32_512( vhash, hash0, hash1, hash2, hash3, hash4, hash5, hash6,
|
||||
hash7 );
|
||||
@@ -241,30 +209,14 @@ void x17_8way_hash( void *state, const void *input )
|
||||
dintrlv_8x32_512( hash0, hash1, hash2, hash3, hash4, hash5, hash6, hash7,
|
||||
vhash );
|
||||
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash0, 64 );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash0 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash1, 64 );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash1 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash2, 64 );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash2 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash3, 64 );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash3 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash4, 64 );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash4 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash5, 64 );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash5 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash6, 64 );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash6 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash7, 64 );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash7 );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash0, hash0, 64 );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash1, hash1, 64 );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash2, hash2, 64 );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash3, hash3, 64 );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash4, hash4, 64 );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash5, hash5, 64 );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash6, hash6, 64 );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash7, hash7, 64 );
|
||||
|
||||
intrlv_8x64_512( vhash, hash0, hash1, hash2, hash3, hash4, hash5, hash6,
|
||||
hash7 );
|
||||
@@ -283,10 +235,10 @@ void x17_8way_hash( void *state, const void *input )
|
||||
int scanhash_x17_8way( struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr )
|
||||
{
|
||||
uint32_t hash[8*8] __attribute__ ((aligned (128)));
|
||||
uint32_t hash32[8*8] __attribute__ ((aligned (128)));
|
||||
uint32_t vdata[20*8] __attribute__ ((aligned (64)));
|
||||
uint32_t lane_hash[8] __attribute__ ((aligned (64)));
|
||||
uint32_t *hashd7 = &(hash[7*8]);
|
||||
uint32_t *hash32_d7 = &(hash32[7*8]);
|
||||
uint32_t *pdata = work->data;
|
||||
const uint32_t *ptarget = work->target;
|
||||
const uint32_t first_nonce = pdata[19];
|
||||
@@ -294,7 +246,7 @@ int scanhash_x17_8way( struct work *work, uint32_t max_nonce,
|
||||
__m512i *noncev = (__m512i*)vdata + 9;
|
||||
uint32_t n = first_nonce;
|
||||
const int thr_id = mythr->id;
|
||||
const uint32_t targ32 = ptarget[7];
|
||||
const uint32_t targ32_d7 = ptarget[7];
|
||||
const bool bench = opt_benchmark;
|
||||
|
||||
mm512_bswap32_intrlv80_8x64( vdata, pdata );
|
||||
@@ -303,16 +255,16 @@ int scanhash_x17_8way( struct work *work, uint32_t max_nonce,
|
||||
n+3, 0, n+2, 0, n+1, 0, n, 0 ), *noncev );
|
||||
do
|
||||
{
|
||||
x17_8way_hash( hash, vdata );
|
||||
x17_8way_hash( hash32, vdata );
|
||||
|
||||
for ( int lane = 0; lane < 8; lane++ )
|
||||
if ( unlikely( ( hashd7[ lane ] <= targ32 ) && !bench ) )
|
||||
if ( unlikely( ( hash32_d7[ lane ] <= targ32_d7 ) && !bench ) )
|
||||
{
|
||||
extr_lane_8x32( lane_hash, hash, lane, 256 );
|
||||
extr_lane_8x32( lane_hash, hash32, lane, 256 );
|
||||
if ( likely( valid_hash( lane_hash, ptarget ) ) )
|
||||
{
|
||||
pdata[19] = bswap_32( n + lane );
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
*noncev = _mm512_add_epi32( *noncev,
|
||||
@@ -418,18 +370,10 @@ void x17_4way_hash( void *state, const void *input )
|
||||
|
||||
dintrlv_4x64_512( hash0, hash1, hash2, hash3, vhash );
|
||||
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash0, 64 );
|
||||
sph_fugue512_close( &ctx.fugue, hash0 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash1, 64 );
|
||||
sph_fugue512_close( &ctx.fugue, hash1 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash2, 64 );
|
||||
sph_fugue512_close( &ctx.fugue, hash2 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash3, 64 );
|
||||
sph_fugue512_close( &ctx.fugue, hash3 );
|
||||
sph_fugue512_full( &ctx.fugue, hash0, hash0, 64 );
|
||||
sph_fugue512_full( &ctx.fugue, hash1, hash1, 64 );
|
||||
sph_fugue512_full( &ctx.fugue, hash2, hash2, 64 );
|
||||
sph_fugue512_full( &ctx.fugue, hash3, hash3, 64 );
|
||||
|
||||
intrlv_4x32_512( vhash, hash0, hash1, hash2, hash3 );
|
||||
|
||||
@@ -439,18 +383,10 @@ void x17_4way_hash( void *state, const void *input )
|
||||
|
||||
dintrlv_4x32_512( hash0, hash1, hash2, hash3, vhash );
|
||||
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash0, 64 );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash0 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash1, 64 );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash1 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash2, 64 );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash2 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash3, 64 );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash3 );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash0, hash0, 64 );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash1, hash1, 64 );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash2, hash2, 64 );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash3, hash3, 64 );
|
||||
|
||||
intrlv_4x64_512( vhash, hash0, hash1, hash2, hash3 );
|
||||
|
||||
@@ -468,10 +404,10 @@ void x17_4way_hash( void *state, const void *input )
|
||||
int scanhash_x17_4way( struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr )
|
||||
{
|
||||
uint32_t hash[8*4] __attribute__ ((aligned (64)));
|
||||
uint32_t hash32[8*4] __attribute__ ((aligned (64)));
|
||||
uint32_t vdata[20*4] __attribute__ ((aligned (64)));
|
||||
uint32_t lane_hash[8] __attribute__ ((aligned (64)));
|
||||
uint32_t *hashd7 = &(hash[ 7*4 ]);
|
||||
uint32_t *hash32_d7 = &(hash32[ 7*4 ]);
|
||||
uint32_t *pdata = work->data;
|
||||
const uint32_t *ptarget = work->target;
|
||||
const uint32_t first_nonce = pdata[19];
|
||||
@@ -479,7 +415,7 @@ int scanhash_x17_4way( struct work *work, uint32_t max_nonce,
|
||||
__m256i *noncev = (__m256i*)vdata + 9;
|
||||
uint32_t n = first_nonce;
|
||||
const int thr_id = mythr->id;
|
||||
const uint32_t targ32 = ptarget[7];
|
||||
const uint32_t targ32_d7 = ptarget[7];
|
||||
const bool bench = opt_benchmark;
|
||||
|
||||
mm256_bswap32_intrlv80_4x64( vdata, pdata );
|
||||
@@ -487,16 +423,16 @@ int scanhash_x17_4way( struct work *work, uint32_t max_nonce,
|
||||
_mm256_set_epi32( n+3, 0, n+2, 0, n+1, 0, n, 0 ), *noncev );
|
||||
do
|
||||
{
|
||||
x17_4way_hash( hash, vdata );
|
||||
x17_4way_hash( hash32, vdata );
|
||||
|
||||
for ( int lane = 0; lane < 4; lane++ )
|
||||
if ( unlikely( hashd7[ lane ] <= targ32 && !bench ) )
|
||||
if ( unlikely( hash32_d7[ lane ] <= targ32_d7 && !bench ) )
|
||||
{
|
||||
extr_lane_4x32( lane_hash, hash, lane, 256 );
|
||||
extr_lane_4x32( lane_hash, hash32, lane, 256 );
|
||||
if ( valid_hash( lane_hash, ptarget ) )
|
||||
{
|
||||
pdata[19] = bswap_32( n + lane );
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
*noncev = _mm256_add_epi32( *noncev,
|
||||
|
||||
@@ -134,30 +134,14 @@ void xevan_8way_hash( void *output, const void *input )
|
||||
dintrlv_4x128( hash0, hash1, hash2, hash3, vhashA, dataLen<<3 );
|
||||
dintrlv_4x128( hash4, hash5, hash6, hash7, vhashB, dataLen<<3 );
|
||||
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, hash0, dataLen );
|
||||
sph_shavite512_close( &ctx.shavite, hash0 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, hash1, dataLen );
|
||||
sph_shavite512_close( &ctx.shavite, hash1 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, hash2, dataLen );
|
||||
sph_shavite512_close( &ctx.shavite, hash2 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, hash3, dataLen );
|
||||
sph_shavite512_close( &ctx.shavite, hash3 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, hash4, dataLen );
|
||||
sph_shavite512_close( &ctx.shavite, hash4 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, hash5, dataLen );
|
||||
sph_shavite512_close( &ctx.shavite, hash5 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, hash6, dataLen );
|
||||
sph_shavite512_close( &ctx.shavite, hash6 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, hash7, dataLen );
|
||||
sph_shavite512_close( &ctx.shavite, hash7 );
|
||||
shavite512_full( &ctx.shavite, hash0, hash0, dataLen );
|
||||
shavite512_full( &ctx.shavite, hash1, hash1, dataLen );
|
||||
shavite512_full( &ctx.shavite, hash2, hash2, dataLen );
|
||||
shavite512_full( &ctx.shavite, hash3, hash3, dataLen );
|
||||
shavite512_full( &ctx.shavite, hash4, hash4, dataLen );
|
||||
shavite512_full( &ctx.shavite, hash5, hash5, dataLen );
|
||||
shavite512_full( &ctx.shavite, hash6, hash6, dataLen );
|
||||
shavite512_full( &ctx.shavite, hash7, hash7, dataLen );
|
||||
|
||||
intrlv_4x128( vhashA, hash0, hash1, hash2, hash3, dataLen<<3 );
|
||||
intrlv_4x128( vhashB, hash4, hash5, hash6, hash7, dataLen<<3 );
|
||||
@@ -208,30 +192,14 @@ void xevan_8way_hash( void *output, const void *input )
|
||||
dintrlv_8x64( hash0, hash1, hash2, hash3, hash4, hash5, hash6, hash7,
|
||||
vhash, dataLen<<3 );
|
||||
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash0, dataLen );
|
||||
sph_fugue512_close( &ctx.fugue, hash0 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash1, dataLen );
|
||||
sph_fugue512_close( &ctx.fugue, hash1 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash2, dataLen );
|
||||
sph_fugue512_close( &ctx.fugue, hash2 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash3, dataLen );
|
||||
sph_fugue512_close( &ctx.fugue, hash3 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash4, dataLen );
|
||||
sph_fugue512_close( &ctx.fugue, hash4 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash5, dataLen );
|
||||
sph_fugue512_close( &ctx.fugue, hash5 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash6, dataLen );
|
||||
sph_fugue512_close( &ctx.fugue, hash6 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash7, dataLen );
|
||||
sph_fugue512_close( &ctx.fugue, hash7 );
|
||||
sph_fugue512_full( &ctx.fugue, hash0, hash0, dataLen );
|
||||
sph_fugue512_full( &ctx.fugue, hash1, hash1, dataLen );
|
||||
sph_fugue512_full( &ctx.fugue, hash2, hash2, dataLen );
|
||||
sph_fugue512_full( &ctx.fugue, hash3, hash3, dataLen );
|
||||
sph_fugue512_full( &ctx.fugue, hash4, hash4, dataLen );
|
||||
sph_fugue512_full( &ctx.fugue, hash5, hash5, dataLen );
|
||||
sph_fugue512_full( &ctx.fugue, hash6, hash6, dataLen );
|
||||
sph_fugue512_full( &ctx.fugue, hash7, hash7, dataLen );
|
||||
|
||||
intrlv_8x32( vhash, hash0, hash1, hash2, hash3, hash4, hash5, hash6,
|
||||
hash7, dataLen<<3 );
|
||||
@@ -243,30 +211,14 @@ void xevan_8way_hash( void *output, const void *input )
|
||||
dintrlv_8x32( hash0, hash1, hash2, hash3, hash4, hash5, hash6, hash7,
|
||||
vhash, dataLen<<3 );
|
||||
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash0, dataLen );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash0 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash1, dataLen );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash1 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash2, dataLen );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash2 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash3, dataLen );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash3 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash4, dataLen );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash4 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash5, dataLen );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash5 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash6, dataLen );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash6 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash7, dataLen );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash7 );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash0, hash0, dataLen );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash1, hash1, dataLen );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash2, hash2, dataLen );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash3, hash3, dataLen );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash4, hash4, dataLen );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash5, hash5, dataLen );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash6, hash6, dataLen );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash7, hash7, dataLen );
|
||||
|
||||
intrlv_8x64( vhash, hash0, hash1, hash2, hash3, hash4, hash5, hash6,
|
||||
hash7, dataLen<<3 );
|
||||
@@ -345,30 +297,14 @@ void xevan_8way_hash( void *output, const void *input )
|
||||
dintrlv_4x128( hash0, hash1, hash2, hash3, vhashA, dataLen<<3 );
|
||||
dintrlv_4x128( hash4, hash5, hash6, hash7, vhashB, dataLen<<3 );
|
||||
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, hash0, dataLen );
|
||||
sph_shavite512_close( &ctx.shavite, hash0 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, hash1, dataLen );
|
||||
sph_shavite512_close( &ctx.shavite, hash1 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, hash2, dataLen );
|
||||
sph_shavite512_close( &ctx.shavite, hash2 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, hash3, dataLen );
|
||||
sph_shavite512_close( &ctx.shavite, hash3 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, hash4, dataLen );
|
||||
sph_shavite512_close( &ctx.shavite, hash4 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, hash5, dataLen );
|
||||
sph_shavite512_close( &ctx.shavite, hash5 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, hash6, dataLen );
|
||||
sph_shavite512_close( &ctx.shavite, hash6 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, hash7, dataLen );
|
||||
sph_shavite512_close( &ctx.shavite, hash7 );
|
||||
shavite512_full( &ctx.shavite, hash0, hash0, dataLen );
|
||||
shavite512_full( &ctx.shavite, hash1, hash1, dataLen );
|
||||
shavite512_full( &ctx.shavite, hash2, hash2, dataLen );
|
||||
shavite512_full( &ctx.shavite, hash3, hash3, dataLen );
|
||||
shavite512_full( &ctx.shavite, hash4, hash4, dataLen );
|
||||
shavite512_full( &ctx.shavite, hash5, hash5, dataLen );
|
||||
shavite512_full( &ctx.shavite, hash6, hash6, dataLen );
|
||||
shavite512_full( &ctx.shavite, hash7, hash7, dataLen );
|
||||
|
||||
intrlv_4x128( vhashA, hash0, hash1, hash2, hash3, dataLen<<3 );
|
||||
intrlv_4x128( vhashB, hash4, hash5, hash6, hash7, dataLen<<3 );
|
||||
@@ -419,30 +355,14 @@ void xevan_8way_hash( void *output, const void *input )
|
||||
dintrlv_8x64( hash0, hash1, hash2, hash3, hash4, hash5, hash6, hash7,
|
||||
vhash, dataLen<<3 );
|
||||
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash0, dataLen );
|
||||
sph_fugue512_close( &ctx.fugue, hash0 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash1, dataLen );
|
||||
sph_fugue512_close( &ctx.fugue, hash1 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash2, dataLen );
|
||||
sph_fugue512_close( &ctx.fugue, hash2 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash3, dataLen );
|
||||
sph_fugue512_close( &ctx.fugue, hash3 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash4, dataLen );
|
||||
sph_fugue512_close( &ctx.fugue, hash4 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash5, dataLen );
|
||||
sph_fugue512_close( &ctx.fugue, hash5 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash6, dataLen );
|
||||
sph_fugue512_close( &ctx.fugue, hash6 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash7, dataLen );
|
||||
sph_fugue512_close( &ctx.fugue, hash7 );
|
||||
sph_fugue512_full( &ctx.fugue, hash0, hash0, dataLen );
|
||||
sph_fugue512_full( &ctx.fugue, hash1, hash1, dataLen );
|
||||
sph_fugue512_full( &ctx.fugue, hash2, hash2, dataLen );
|
||||
sph_fugue512_full( &ctx.fugue, hash3, hash3, dataLen );
|
||||
sph_fugue512_full( &ctx.fugue, hash4, hash4, dataLen );
|
||||
sph_fugue512_full( &ctx.fugue, hash5, hash5, dataLen );
|
||||
sph_fugue512_full( &ctx.fugue, hash6, hash6, dataLen );
|
||||
sph_fugue512_full( &ctx.fugue, hash7, hash7, dataLen );
|
||||
|
||||
intrlv_8x32( vhash, hash0, hash1, hash2, hash3, hash4, hash5, hash6,
|
||||
hash7, dataLen<<3 );
|
||||
@@ -454,30 +374,14 @@ void xevan_8way_hash( void *output, const void *input )
|
||||
dintrlv_8x32( hash0, hash1, hash2, hash3, hash4, hash5, hash6, hash7,
|
||||
vhash, dataLen<<3 );
|
||||
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash0, dataLen );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash0 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash1, dataLen );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash1 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash2, dataLen );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash2 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash3, dataLen );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash3 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash4, dataLen );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash4 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash5, dataLen );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash5 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash6, dataLen );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash6 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash7, dataLen );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash7 );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash0, hash0, dataLen );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash1, hash1, dataLen );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash2, hash2, dataLen );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash3, hash3, dataLen );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash4, hash4, dataLen );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash5, hash5, dataLen );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash6, hash6, dataLen );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash7, hash7, dataLen );
|
||||
|
||||
intrlv_8x64( vhash, hash0, hash1, hash2, hash3, hash4, hash5, hash6,
|
||||
hash7, dataLen<<3 );
|
||||
@@ -525,7 +429,7 @@ int scanhash_xevan_8way( struct work *work, uint32_t max_nonce,
|
||||
if ( likely( valid_hash( lane_hash, ptarget ) ) )
|
||||
{
|
||||
pdata[19] = bswap_32( n + lane );
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
*noncev = _mm512_add_epi32( *noncev,
|
||||
@@ -636,18 +540,10 @@ void xevan_4way_hash( void *output, const void *input )
|
||||
|
||||
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, dataLen<<3 );
|
||||
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash0, dataLen );
|
||||
sph_fugue512_close( &ctx.fugue, hash0 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash1, dataLen );
|
||||
sph_fugue512_close( &ctx.fugue, hash1 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash2, dataLen );
|
||||
sph_fugue512_close( &ctx.fugue, hash2 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash3, dataLen );
|
||||
sph_fugue512_close( &ctx.fugue, hash3 );
|
||||
sph_fugue512_full( &ctx.fugue, hash0, hash0, 64 );
|
||||
sph_fugue512_full( &ctx.fugue, hash1, hash1, 64 );
|
||||
sph_fugue512_full( &ctx.fugue, hash2, hash2, 64 );
|
||||
sph_fugue512_full( &ctx.fugue, hash3, hash3, 64 );
|
||||
|
||||
// Parallel 4way 32 bit
|
||||
intrlv_4x32( vhash, hash0, hash1, hash2, hash3, dataLen<<3 );
|
||||
@@ -659,18 +555,10 @@ void xevan_4way_hash( void *output, const void *input )
|
||||
dintrlv_4x32( hash0, hash1, hash2, hash3, vhash, dataLen<<3 );
|
||||
|
||||
// Serial
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash0, dataLen );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash0 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash1, dataLen );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash1 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash2, dataLen );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash2 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash3, dataLen );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash3 );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash0, hash0, dataLen );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash1, hash1, dataLen );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash2, hash2, dataLen );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash3, hash3, dataLen );
|
||||
|
||||
intrlv_4x64( vhash, hash0, hash1, hash2, hash3, dataLen<<3 );
|
||||
|
||||
@@ -749,18 +637,10 @@ void xevan_4way_hash( void *output, const void *input )
|
||||
|
||||
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, dataLen<<3 );
|
||||
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash0, dataLen );
|
||||
sph_fugue512_close( &ctx.fugue, hash0 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash1, dataLen );
|
||||
sph_fugue512_close( &ctx.fugue, hash1 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash2, dataLen );
|
||||
sph_fugue512_close( &ctx.fugue, hash2 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash3, dataLen );
|
||||
sph_fugue512_close( &ctx.fugue, hash3 );
|
||||
sph_fugue512_full( &ctx.fugue, hash0, hash0, 64 );
|
||||
sph_fugue512_full( &ctx.fugue, hash1, hash1, 64 );
|
||||
sph_fugue512_full( &ctx.fugue, hash2, hash2, 64 );
|
||||
sph_fugue512_full( &ctx.fugue, hash3, hash3, 64 );
|
||||
|
||||
intrlv_4x32( vhash, hash0, hash1, hash2, hash3, dataLen<<3 );
|
||||
|
||||
@@ -770,18 +650,10 @@ void xevan_4way_hash( void *output, const void *input )
|
||||
|
||||
dintrlv_4x32( hash0, hash1, hash2, hash3, vhash, dataLen<<3 );
|
||||
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash0, dataLen );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash0 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash1, dataLen );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash1 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash2, dataLen );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash2 );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, hash3, dataLen );
|
||||
sph_whirlpool_close( &ctx.whirlpool, hash3 );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash0, hash0, dataLen );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash1, hash1, dataLen );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash2, hash2, dataLen );
|
||||
sph_whirlpool512_full( &ctx.whirlpool, hash3, hash3, dataLen );
|
||||
|
||||
intrlv_4x64( vhash, hash0, hash1, hash2, hash3, dataLen<<3 );
|
||||
|
||||
@@ -827,7 +699,7 @@ int scanhash_xevan_4way( struct work *work, uint32_t max_nonce,
|
||||
if ( valid_hash( lane_hash, ptarget ) )
|
||||
{
|
||||
pdata[19] = bswap_32( n + lane );
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
*noncev = _mm256_add_epi32( *noncev,
|
||||
|
||||
@@ -62,7 +62,7 @@ union _x22i_8way_ctx_overlay
|
||||
};
|
||||
typedef union _x22i_8way_ctx_overlay x22i_8way_ctx_overlay;
|
||||
|
||||
void x22i_8way_hash( void *output, const void *input )
|
||||
int x22i_8way_hash( void *output, const void *input, int thrid )
|
||||
{
|
||||
uint64_t vhash[8*8] __attribute__ ((aligned (128)));
|
||||
uint64_t vhashA[8*8] __attribute__ ((aligned (64)));
|
||||
@@ -129,6 +129,8 @@ void x22i_8way_hash( void *output, const void *input )
|
||||
keccak512_8way_update( &ctx.keccak, vhash, 64 );
|
||||
keccak512_8way_close( &ctx.keccak, vhash );
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
rintrlv_8x64_4x128( vhashA, vhashB, vhash, 512 );
|
||||
|
||||
luffa512_4way_full( &ctx.luffa, vhashA, vhashA, 64 );
|
||||
@@ -214,6 +216,8 @@ void x22i_8way_hash( void *output, const void *input )
|
||||
|
||||
#endif
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
hamsi512_8way_init( &ctx.hamsi );
|
||||
hamsi512_8way_update( &ctx.hamsi, vhash, 64 );
|
||||
hamsi512_8way_close( &ctx.hamsi, vhash );
|
||||
@@ -346,6 +350,8 @@ void x22i_8way_hash( void *output, const void *input )
|
||||
sph_tiger (&ctx.tiger, (const void*) hash7, 64);
|
||||
sph_tiger_close(&ctx.tiger, (void*) hashA7);
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
memset( hash0, 0, 64 );
|
||||
memset( hash1, 0, 64 );
|
||||
memset( hash2, 0, 64 );
|
||||
@@ -399,6 +405,8 @@ void x22i_8way_hash( void *output, const void *input )
|
||||
sha256_8way_init( &ctx.sha256 );
|
||||
sha256_8way_update( &ctx.sha256, vhash, 64 );
|
||||
sha256_8way_close( &ctx.sha256, output );
|
||||
|
||||
return 1;
|
||||
}
|
||||
|
||||
int scanhash_x22i_8way( struct work *work, uint32_t max_nonce,
|
||||
@@ -428,8 +436,7 @@ int scanhash_x22i_8way( struct work *work, uint32_t max_nonce,
|
||||
n+3, 0, n+2, 0, n+1, 0, n, 0 ), *noncev );
|
||||
do
|
||||
{
|
||||
x22i_8way_hash( hash, vdata );
|
||||
|
||||
if ( x22i_8way_hash( hash, vdata, thr_id ) )
|
||||
for ( int lane = 0; lane < 8; lane++ )
|
||||
if ( unlikely( ( hashd7[ lane ] <= targ32 ) && !bench ) )
|
||||
{
|
||||
@@ -437,7 +444,7 @@ int scanhash_x22i_8way( struct work *work, uint32_t max_nonce,
|
||||
if ( likely( valid_hash( lane_hash, ptarget ) ) )
|
||||
{
|
||||
pdata[19] = bswap_32( n + lane );
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
*noncev = _mm512_add_epi32( *noncev,
|
||||
@@ -486,7 +493,7 @@ int scanhash_x22i_8way( struct work* work, uint32_t max_nonce,
|
||||
if ( likely( fulltest( lane_hash, ptarget ) && !opt_benchmark ) )
|
||||
{
|
||||
pdata[19] = n + lane;
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
n += 8;
|
||||
@@ -524,7 +531,7 @@ union _x22i_4way_ctx_overlay
|
||||
};
|
||||
typedef union _x22i_4way_ctx_overlay x22i_ctx_overlay;
|
||||
|
||||
void x22i_4way_hash( void *output, const void *input )
|
||||
int x22i_4way_hash( void *output, const void *input, int thrid )
|
||||
{
|
||||
uint64_t hash0[8*4] __attribute__ ((aligned (64)));
|
||||
uint64_t hash1[8*4] __attribute__ ((aligned (64)));
|
||||
@@ -563,6 +570,8 @@ void x22i_4way_hash( void *output, const void *input )
|
||||
keccak512_4way_update( &ctx.keccak, vhash, 64 );
|
||||
keccak512_4way_close( &ctx.keccak, vhash );
|
||||
|
||||
if ( work_restart[thrid].restart ) return false;
|
||||
|
||||
rintrlv_4x64_2x128( vhashA, vhashB, vhash, 512 );
|
||||
|
||||
luffa512_2way_full( &ctx.luffa, vhashA, vhashA, 64 );
|
||||
@@ -591,6 +600,8 @@ void x22i_4way_hash( void *output, const void *input )
|
||||
|
||||
intrlv_4x64_512( vhash, hash0, hash1, hash2, hash3 );
|
||||
|
||||
if ( work_restart[thrid].restart ) return false;
|
||||
|
||||
hamsi512_4way_init( &ctx.hamsi );
|
||||
hamsi512_4way_update( &ctx.hamsi, vhash, 64 );
|
||||
hamsi512_4way_close( &ctx.hamsi, vhash );
|
||||
@@ -636,6 +647,8 @@ void x22i_4way_hash( void *output, const void *input )
|
||||
sha512_4way_close( &ctx.sha512, vhash );
|
||||
dintrlv_4x64_512( &hash0[24], &hash1[24], &hash2[24], &hash3[24], vhash );
|
||||
|
||||
if ( work_restart[thrid].restart ) return false;
|
||||
|
||||
ComputeSingleSWIFFTX((unsigned char*)hash0, (unsigned char*)hashA0);
|
||||
ComputeSingleSWIFFTX((unsigned char*)hash1, (unsigned char*)hashA1);
|
||||
ComputeSingleSWIFFTX((unsigned char*)hash2, (unsigned char*)hashA2);
|
||||
@@ -668,6 +681,8 @@ void x22i_4way_hash( void *output, const void *input )
|
||||
sph_tiger (&ctx.tiger, (const void*) hash3, 64);
|
||||
sph_tiger_close(&ctx.tiger, (void*) hashA3);
|
||||
|
||||
if ( work_restart[thrid].restart ) return false;
|
||||
|
||||
memset( hash0, 0, 64 );
|
||||
memset( hash1, 0, 64 );
|
||||
memset( hash2, 0, 64 );
|
||||
@@ -700,8 +715,9 @@ void x22i_4way_hash( void *output, const void *input )
|
||||
sha256_4way_init( &ctx.sha256 );
|
||||
sha256_4way_update( &ctx.sha256, vhash, 64 );
|
||||
sha256_4way_close( &ctx.sha256, output );
|
||||
}
|
||||
|
||||
return 1;
|
||||
}
|
||||
|
||||
int scanhash_x22i_4way( struct work* work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr )
|
||||
@@ -729,8 +745,7 @@ int scanhash_x22i_4way( struct work* work, uint32_t max_nonce,
|
||||
_mm256_set_epi32( n+3, 0, n+2, 0, n+1, 0, n, 0 ), *noncev );
|
||||
do
|
||||
{
|
||||
x22i_4way_hash( hash, vdata );
|
||||
|
||||
if ( x22i_4way_hash( hash, vdata, thr_id ) )
|
||||
for ( int lane = 0; lane < 4; lane++ )
|
||||
if ( unlikely( hashd7[ lane ] <= targ32 && !bench ) )
|
||||
{
|
||||
@@ -738,7 +753,7 @@ int scanhash_x22i_4way( struct work* work, uint32_t max_nonce,
|
||||
if ( valid_hash( lane_hash, ptarget ) )
|
||||
{
|
||||
pdata[19] = bswap_32( n + lane );
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
*noncev = _mm256_add_epi32( *noncev,
|
||||
|
||||
@@ -16,19 +16,19 @@ bool register_x22i_algo( algo_gate_t* gate );
|
||||
|
||||
#if defined(X22I_8WAY)
|
||||
|
||||
void x22i_8way_hash( void *state, const void *input );
|
||||
int x22i_8way_hash( void *state, const void *input, int thrid );
|
||||
int scanhash_x22i_8way( struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr );
|
||||
|
||||
#elif defined(X22I_4WAY)
|
||||
|
||||
void x22i_4way_hash( void *state, const void *input );
|
||||
int x22i_4way_hash( void *state, const void *input, int thrid );
|
||||
int scanhash_x22i_4way( struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr );
|
||||
|
||||
#else
|
||||
|
||||
void x22i_hash( void *state, const void *input );
|
||||
int x22i_hash( void *state, const void *input, int thrid );
|
||||
int scanhash_x22i( struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr );
|
||||
|
||||
@@ -44,19 +44,19 @@ bool register_x25i_algo( algo_gate_t* gate );
|
||||
|
||||
#if defined(X25X_8WAY)
|
||||
|
||||
void x25x_8way_hash( void *state, const void *input );
|
||||
int x25x_8way_hash( void *state, const void *input, int thrid );
|
||||
int scanhash_x25x_8way( struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr );
|
||||
|
||||
#elif defined(X25X_4WAY)
|
||||
|
||||
void x25x_4way_hash( void *state, const void *input );
|
||||
int x25x_4way_hash( void *state, const void *input, int thrid );
|
||||
int scanhash_x25x_4way( struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr );
|
||||
|
||||
#else
|
||||
|
||||
void x25x_hash( void *state, const void *input );
|
||||
int x25x_hash( void *state, const void *input, int thrif );
|
||||
int scanhash_x25x( struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr );
|
||||
|
||||
|
||||
@@ -59,7 +59,7 @@ union _x22i_context_overlay
|
||||
};
|
||||
typedef union _x22i_context_overlay x22i_context_overlay;
|
||||
|
||||
void x22i_hash( void *output, const void *input )
|
||||
int x22i_hash( void *output, const void *input, int thrid )
|
||||
{
|
||||
unsigned char hash[64 * 4] __attribute__((aligned(64))) = {0};
|
||||
unsigned char hash2[65] __attribute__((aligned(64))) = {0};
|
||||
@@ -95,6 +95,8 @@ void x22i_hash( void *output, const void *input )
|
||||
sph_keccak512(&ctx.keccak, (const void*) hash, 64);
|
||||
sph_keccak512_close(&ctx.keccak, hash);
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
init_luffa( &ctx.luffa, 512 );
|
||||
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash,
|
||||
(const BitSequence*)hash, 64 );
|
||||
@@ -121,6 +123,8 @@ void x22i_hash( void *output, const void *input )
|
||||
sph_echo512_close( &ctx.echo, hash );
|
||||
#endif
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
sph_hamsi512_init(&ctx.hamsi);
|
||||
sph_hamsi512(&ctx.hamsi, (const void*) hash, 64);
|
||||
sph_hamsi512_close(&ctx.hamsi, hash);
|
||||
@@ -143,6 +147,8 @@ void x22i_hash( void *output, const void *input )
|
||||
|
||||
ComputeSingleSWIFFTX((unsigned char*)hash, (unsigned char*)hash2);
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
memset(hash, 0, 64);
|
||||
sph_haval256_5_init(&ctx.haval);
|
||||
sph_haval256_5(&ctx.haval,(const void*) hash2, 64);
|
||||
@@ -165,6 +171,8 @@ void x22i_hash( void *output, const void *input )
|
||||
SHA256_Final( (unsigned char*) hash, &ctx.sha256 );
|
||||
|
||||
memcpy(output, hash, 32);
|
||||
|
||||
return 1;
|
||||
}
|
||||
|
||||
int scanhash_x22i( struct work *work, uint32_t max_nonce,
|
||||
@@ -188,7 +196,7 @@ int scanhash_x22i( struct work *work, uint32_t max_nonce,
|
||||
do
|
||||
{
|
||||
edata[19] = n;
|
||||
x22i_hash( hash64, edata );
|
||||
if ( x22i_hash( hash64, edata, thr_id ) );
|
||||
if ( unlikely( valid_hash( hash64, ptarget ) && !bench ) )
|
||||
{
|
||||
pdata[19] = bswap_32( n );
|
||||
|
||||
@@ -94,7 +94,7 @@ union _x25x_8way_ctx_overlay
|
||||
};
|
||||
typedef union _x25x_8way_ctx_overlay x25x_8way_ctx_overlay;
|
||||
|
||||
void x25x_8way_hash( void *output, const void *input )
|
||||
int x25x_8way_hash( void *output, const void *input, int thrid )
|
||||
{
|
||||
uint64_t vhash[8*8] __attribute__ ((aligned (128)));
|
||||
unsigned char hash0[25][64] __attribute__((aligned(64))) = {0};
|
||||
@@ -179,13 +179,15 @@ void x25x_8way_hash( void *output, const void *input )
|
||||
jh512_8way_close( &ctx.jh, vhash );
|
||||
dintrlv_8x64_512( hash0[4], hash1[4], hash2[4], hash3[4],
|
||||
hash4[4], hash5[4], hash6[4], hash7[4], vhash );
|
||||
|
||||
|
||||
keccak512_8way_init( &ctx.keccak );
|
||||
keccak512_8way_update( &ctx.keccak, vhash, 64 );
|
||||
keccak512_8way_close( &ctx.keccak, vhash );
|
||||
dintrlv_8x64_512( hash0[5], hash1[5], hash2[5], hash3[5],
|
||||
hash4[5], hash5[5], hash6[5], hash7[5], vhash );
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
rintrlv_8x64_4x128( vhashA, vhashB, vhash, 512 );
|
||||
|
||||
luffa_4way_init( &ctx.luffa, 512 );
|
||||
@@ -261,6 +263,7 @@ void x25x_8way_hash( void *output, const void *input )
|
||||
intrlv_8x64_512( vhash, hash0[10], hash1[10], hash2[10], hash3[10],
|
||||
hash4[10], hash5[10], hash6[10], hash7[10] );
|
||||
|
||||
|
||||
#else
|
||||
|
||||
init_echo( &ctx.echo, 512 );
|
||||
@@ -292,6 +295,8 @@ void x25x_8way_hash( void *output, const void *input )
|
||||
|
||||
#endif
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
hamsi512_8way_init( &ctx.hamsi );
|
||||
hamsi512_8way_update( &ctx.hamsi, vhash, 64 );
|
||||
hamsi512_8way_close( &ctx.hamsi, vhash );
|
||||
@@ -407,6 +412,8 @@ void x25x_8way_hash( void *output, const void *input )
|
||||
sph_tiger (&ctx.tiger, (const void*) hash7[17], 64);
|
||||
sph_tiger_close(&ctx.tiger, (void*) hash7[18]);
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
intrlv_2x256( vhash, hash0[18], hash1[18], 256 );
|
||||
LYRA2X_2WAY( vhash, 32, vhash, 32, 1, 4, 4 );
|
||||
dintrlv_2x256( hash0[19], hash1[19], vhash, 256 );
|
||||
@@ -468,6 +475,8 @@ void x25x_8way_hash( void *output, const void *input )
|
||||
laneHash(512, (const BitSequence*)hash6[22], 512, (BitSequence*)hash6[23]);
|
||||
laneHash(512, (const BitSequence*)hash7[22], 512, (BitSequence*)hash7[23]);
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
x25x_shuffle( hash0 );
|
||||
x25x_shuffle( hash1 );
|
||||
x25x_shuffle( hash2 );
|
||||
@@ -528,6 +537,8 @@ void x25x_8way_hash( void *output, const void *input )
|
||||
|
||||
blake2s_8way_init( &ctx.blake2s, 32 );
|
||||
blake2s_8way_full_blocks( &ctx.blake2s, output, vhashX, 64*24 );
|
||||
|
||||
return 1;
|
||||
}
|
||||
|
||||
int scanhash_x25x_8way( struct work *work, uint32_t max_nonce,
|
||||
@@ -557,7 +568,7 @@ int scanhash_x25x_8way( struct work *work, uint32_t max_nonce,
|
||||
n+3, 0, n+2, 0, n+1, 0, n, 0 ), *noncev );
|
||||
do
|
||||
{
|
||||
x25x_8way_hash( hash, vdata );
|
||||
if ( x25x_8way_hash( hash, vdata, thr_id ) );
|
||||
|
||||
for ( int lane = 0; lane < 8; lane++ )
|
||||
if ( unlikely( ( hashd7[ lane ] <= targ32 ) && !bench ) )
|
||||
@@ -566,7 +577,7 @@ int scanhash_x25x_8way( struct work *work, uint32_t max_nonce,
|
||||
if ( likely( valid_hash( lane_hash, ptarget ) ) )
|
||||
{
|
||||
pdata[19] = bswap_32( n + lane );
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
*noncev = _mm512_add_epi32( *noncev,
|
||||
@@ -614,7 +625,7 @@ int scanhash_x25x_8way( struct work* work, uint32_t max_nonce,
|
||||
if ( fulltest( lane_hash, ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n + lane;
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
n += 8;
|
||||
@@ -654,7 +665,7 @@ union _x25x_4way_ctx_overlay
|
||||
};
|
||||
typedef union _x25x_4way_ctx_overlay x25x_4way_ctx_overlay;
|
||||
|
||||
void x25x_4way_hash( void *output, const void *input )
|
||||
int x25x_4way_hash( void *output, const void *input, int thrid )
|
||||
{
|
||||
uint64_t vhash[8*4] __attribute__ ((aligned (128)));
|
||||
unsigned char hash0[25][64] __attribute__((aligned(64))) = {0};
|
||||
@@ -686,6 +697,8 @@ void x25x_4way_hash( void *output, const void *input )
|
||||
jh512_4way_close( &ctx.jh, vhash );
|
||||
dintrlv_4x64_512( hash0[4], hash1[4], hash2[4], hash3[4], vhash );
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
keccak512_4way_init( &ctx.keccak );
|
||||
keccak512_4way_update( &ctx.keccak, vhash, 64 );
|
||||
keccak512_4way_close( &ctx.keccak, vhash );
|
||||
@@ -738,6 +751,8 @@ void x25x_4way_hash( void *output, const void *input )
|
||||
|
||||
intrlv_4x64_512( vhash, hash0[10], hash1[10], hash2[10], hash3[10] );
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
hamsi512_4way_init( &ctx.hamsi );
|
||||
hamsi512_4way_update( &ctx.hamsi, vhash, 64 );
|
||||
hamsi512_4way_close( &ctx.hamsi, vhash );
|
||||
@@ -819,6 +834,8 @@ void x25x_4way_hash( void *output, const void *input )
|
||||
LYRA2RE( (void*)hash3[19], 32, (const void*)hash3[18], 32,
|
||||
(const void*)hash3[18], 32, 1, 4, 4 );
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
sph_gost512_init(&ctx.gost);
|
||||
sph_gost512 (&ctx.gost, (const void*) hash0[19], 64);
|
||||
sph_gost512_close(&ctx.gost, (void*) hash0[20]);
|
||||
@@ -850,6 +867,8 @@ void x25x_4way_hash( void *output, const void *input )
|
||||
laneHash(512, (const BitSequence*)hash2[22], 512, (BitSequence*)hash2[23]);
|
||||
laneHash(512, (const BitSequence*)hash3[22], 512, (BitSequence*)hash3[23]);
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
x25x_shuffle( hash0 );
|
||||
x25x_shuffle( hash1 );
|
||||
x25x_shuffle( hash2 );
|
||||
@@ -882,6 +901,8 @@ void x25x_4way_hash( void *output, const void *input )
|
||||
|
||||
blake2s_4way_init( &ctx.blake2s, 32 );
|
||||
blake2s_4way_full_blocks( &ctx.blake2s, output, vhashX, 64*24 );
|
||||
|
||||
return 1;
|
||||
}
|
||||
|
||||
int scanhash_x25x_4way( struct work* work, uint32_t max_nonce,
|
||||
@@ -910,8 +931,7 @@ int scanhash_x25x_4way( struct work* work, uint32_t max_nonce,
|
||||
_mm256_set_epi32( n+3, 0, n+2, 0, n+1, 0, n, 0 ), *noncev );
|
||||
do
|
||||
{
|
||||
x25x_4way_hash( hash, vdata );
|
||||
|
||||
if ( x25x_4way_hash( hash, vdata, thr_id ) )
|
||||
for ( int lane = 0; lane < 4; lane++ )
|
||||
if ( unlikely( hashd7[ lane ] <= targ32 && !bench ) )
|
||||
{
|
||||
@@ -919,7 +939,7 @@ int scanhash_x25x_4way( struct work* work, uint32_t max_nonce,
|
||||
if ( valid_hash( lane_hash, ptarget ) )
|
||||
{
|
||||
pdata[19] = bswap_32( n + lane );
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
submit_solution( work, lane_hash, mythr );
|
||||
}
|
||||
}
|
||||
*noncev = _mm256_add_epi32( *noncev,
|
||||
|
||||
@@ -64,7 +64,7 @@ union _x25x_context_overlay
|
||||
};
|
||||
typedef union _x25x_context_overlay x25x_context_overlay;
|
||||
|
||||
void x25x_hash( void *output, const void *input )
|
||||
int x25x_hash( void *output, const void *input, int thrid )
|
||||
{
|
||||
unsigned char hash[25][64] __attribute__((aligned(64))) = {0};
|
||||
x25x_context_overlay ctx;
|
||||
@@ -99,6 +99,8 @@ void x25x_hash( void *output, const void *input )
|
||||
sph_keccak512(&ctx.keccak, (const void*) &hash[4], 64);
|
||||
sph_keccak512_close(&ctx.keccak, &hash[5]);
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
init_luffa( &ctx.luffa, 512 );
|
||||
update_and_final_luffa( &ctx.luffa, (BitSequence*)&hash[6],
|
||||
(const BitSequence*)&hash[5], 64 );
|
||||
@@ -125,7 +127,9 @@ void x25x_hash( void *output, const void *input )
|
||||
sph_echo512_close( &ctx.echo, &hash[10] );
|
||||
#endif
|
||||
|
||||
sph_hamsi512_init(&ctx.hamsi);
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
sph_hamsi512_init(&ctx.hamsi);
|
||||
sph_hamsi512(&ctx.hamsi, (const void*) &hash[10], 64);
|
||||
sph_hamsi512_close(&ctx.hamsi, &hash[11]);
|
||||
|
||||
@@ -151,6 +155,8 @@ void x25x_hash( void *output, const void *input )
|
||||
sph_haval256_5(&ctx.haval,(const void*) &hash[16], 64);
|
||||
sph_haval256_5_close(&ctx.haval,&hash[17]);
|
||||
|
||||
if ( work_restart[thrid].restart ) return 0;
|
||||
|
||||
sph_tiger_init(&ctx.tiger);
|
||||
sph_tiger (&ctx.tiger, (const void*) &hash[17], 64);
|
||||
sph_tiger_close(&ctx.tiger, (void*) &hash[18]);
|
||||
@@ -199,6 +205,8 @@ void x25x_hash( void *output, const void *input )
|
||||
blake2s_simple( (uint8_t*)&hash[24], (const void*)(&hash[0]), 64 * 24 );
|
||||
|
||||
memcpy(output, &hash[24], 32);
|
||||
|
||||
return 1;
|
||||
}
|
||||
|
||||
int scanhash_x25x( struct work *work, uint32_t max_nonce,
|
||||
@@ -222,7 +230,7 @@ int scanhash_x25x( struct work *work, uint32_t max_nonce,
|
||||
do
|
||||
{
|
||||
edata[19] = n;
|
||||
x25x_hash( hash64, edata );
|
||||
if ( x25x_hash( hash64, edata, thr_id ) );
|
||||
if ( unlikely( valid_hash( hash64, ptarget ) && !bench ) )
|
||||
{
|
||||
pdata[19] = bswap_32( n );
|
||||
|
||||
@@ -31,6 +31,7 @@
|
||||
#undef HUGEPAGE_SIZE
|
||||
#endif
|
||||
|
||||
/*
|
||||
static __inline uint32_t
|
||||
le32dec(const void *pp)
|
||||
{
|
||||
@@ -50,6 +51,7 @@ le32enc(void *pp, uint32_t x)
|
||||
p[2] = (x >> 16) & 0xff;
|
||||
p[3] = (x >> 24) & 0xff;
|
||||
}
|
||||
*/
|
||||
|
||||
static void *
|
||||
alloc_region(yescrypt_region_t * region, size_t size)
|
||||
@@ -154,7 +156,7 @@ int yescrypt_init_shared(yescrypt_shared_t * shared, const uint8_t * param, size
|
||||
if (yescrypt_kdf(&dummy, shared1,
|
||||
param, paramlen, NULL, 0, N, r, p, 0,
|
||||
YESCRYPT_RW | YESCRYPT_PARALLEL_SMIX | __YESCRYPT_INIT_SHARED_1,
|
||||
salt, sizeof(salt)))
|
||||
salt, sizeof(salt), 0 ) )
|
||||
goto out;
|
||||
|
||||
half1 = half2 = *shared;
|
||||
@@ -166,19 +168,19 @@ int yescrypt_init_shared(yescrypt_shared_t * shared, const uint8_t * param, size
|
||||
if (p > 1 && yescrypt_kdf(&half1, &half2.shared1,
|
||||
param, paramlen, salt, sizeof(salt), N, r, p, 0,
|
||||
YESCRYPT_RW | YESCRYPT_PARALLEL_SMIX | __YESCRYPT_INIT_SHARED_2,
|
||||
salt, sizeof(salt)))
|
||||
salt, sizeof(salt), 0 ))
|
||||
goto out;
|
||||
|
||||
if (yescrypt_kdf(&half2, &half1.shared1,
|
||||
param, paramlen, salt, sizeof(salt), N, r, p, 0,
|
||||
YESCRYPT_RW | YESCRYPT_PARALLEL_SMIX | __YESCRYPT_INIT_SHARED_1,
|
||||
salt, sizeof(salt)))
|
||||
salt, sizeof(salt), 0))
|
||||
goto out;
|
||||
|
||||
if (yescrypt_kdf(&half1, &half2.shared1,
|
||||
param, paramlen, salt, sizeof(salt), N, r, p, 0,
|
||||
YESCRYPT_RW | YESCRYPT_PARALLEL_SMIX | __YESCRYPT_INIT_SHARED_1,
|
||||
buf, buflen))
|
||||
buf, buflen, 0))
|
||||
goto out;
|
||||
|
||||
shared->mask1 = mask;
|
||||
|
||||
@@ -1149,7 +1149,7 @@ yescrypt_kdf(const yescrypt_shared_t * shared, yescrypt_local_t * local,
|
||||
const uint8_t * passwd, size_t passwdlen,
|
||||
const uint8_t * salt, size_t saltlen,
|
||||
uint64_t N, uint32_t r, uint32_t p, uint32_t t, yescrypt_flags_t flags,
|
||||
uint8_t * buf, size_t buflen)
|
||||
uint8_t * buf, size_t buflen, int thrid )
|
||||
{
|
||||
uint8_t _ALIGN(128) sha256[32];
|
||||
yescrypt_region_t tmp;
|
||||
@@ -1157,6 +1157,7 @@ yescrypt_kdf(const yescrypt_shared_t * shared, yescrypt_local_t * local,
|
||||
size_t B_size, V_size, XY_size, need;
|
||||
uint8_t * B, * S;
|
||||
salsa20_blk_t * V, * XY;
|
||||
int retval = 1;
|
||||
|
||||
/*
|
||||
* YESCRYPT_PARALLEL_SMIX is a no-op at p = 1 for its intended purpose,
|
||||
@@ -1312,6 +1313,12 @@ yescrypt_kdf(const yescrypt_shared_t * shared, yescrypt_local_t * local,
|
||||
/* 1: (B_0 ... B_{p-1}) <-- PBKDF2(P, S, 1, p * MFLen) */
|
||||
PBKDF2_SHA256(passwd, passwdlen, salt, saltlen, 1, B, B_size);
|
||||
|
||||
if ( work_restart[thrid].restart )
|
||||
{
|
||||
retval = 0;
|
||||
goto out;
|
||||
}
|
||||
|
||||
if (t || flags)
|
||||
memcpy(sha256, B, sizeof(sha256));
|
||||
|
||||
@@ -1339,9 +1346,21 @@ yescrypt_kdf(const yescrypt_shared_t * shared, yescrypt_local_t * local,
|
||||
}
|
||||
}
|
||||
|
||||
if ( work_restart[thrid].restart )
|
||||
{
|
||||
retval = 0;
|
||||
goto out;
|
||||
}
|
||||
|
||||
/* 5: DK <-- PBKDF2(P, B, 1, dkLen) */
|
||||
PBKDF2_SHA256(passwd, passwdlen, B, B_size, 1, buf, buflen);
|
||||
|
||||
if ( work_restart[thrid].restart )
|
||||
{
|
||||
retval = 0;
|
||||
goto out;
|
||||
}
|
||||
|
||||
/*
|
||||
* Except when computing classic scrypt, allow all computation so far
|
||||
* to be performed on the client. The final steps below match those of
|
||||
@@ -1370,9 +1389,10 @@ yescrypt_kdf(const yescrypt_shared_t * shared, yescrypt_local_t * local,
|
||||
}
|
||||
}
|
||||
|
||||
out:
|
||||
if (free_region(&tmp))
|
||||
return -1;
|
||||
|
||||
/* Success! */
|
||||
return 0;
|
||||
return retval;
|
||||
}
|
||||
|
||||
@@ -106,7 +106,8 @@ static const uint8_t* decode64_uint32(uint32_t* dst, uint32_t dstbits, const uin
|
||||
}
|
||||
|
||||
uint8_t* yescrypt_r(const yescrypt_shared_t* shared, yescrypt_local_t* local,
|
||||
const uint8_t* passwd, size_t passwdlen, const uint8_t* setting, uint8_t* buf, size_t buflen)
|
||||
const uint8_t* passwd, size_t passwdlen, const uint8_t* setting,
|
||||
uint8_t* buf, size_t buflen, int thrid )
|
||||
{
|
||||
uint8_t hash[HASH_SIZE];
|
||||
const uint8_t * src, * salt;
|
||||
@@ -210,7 +211,9 @@ uint8_t* yescrypt_r(const yescrypt_shared_t* shared, yescrypt_local_t* local,
|
||||
return NULL;
|
||||
}
|
||||
|
||||
if (yescrypt_kdf(shared, local, passwd, passwdlen, salt, saltlen, N, r, p, 0, flags, hash, sizeof(hash))) {
|
||||
if ( yescrypt_kdf( shared, local, passwd, passwdlen, salt, saltlen, N, r, p,
|
||||
0, flags, hash, sizeof(hash), thrid ) == -1 )
|
||||
{
|
||||
printf("died10 ...");
|
||||
fflush(stdout);
|
||||
return NULL;
|
||||
@@ -237,7 +240,7 @@ uint8_t* yescrypt_r(const yescrypt_shared_t* shared, yescrypt_local_t* local,
|
||||
return buf;
|
||||
}
|
||||
|
||||
uint8_t* yescrypt(const uint8_t* passwd, const uint8_t* setting)
|
||||
uint8_t* yescrypt(const uint8_t* passwd, const uint8_t* setting, int thrid )
|
||||
{
|
||||
static uint8_t buf[4 + 1 + 5 + 5 + BYTES2CHARS(32) + 1 + HASH_LEN + 1];
|
||||
yescrypt_shared_t shared;
|
||||
@@ -252,7 +255,7 @@ uint8_t* yescrypt(const uint8_t* passwd, const uint8_t* setting)
|
||||
return NULL;
|
||||
}
|
||||
retval = yescrypt_r(&shared, &local,
|
||||
passwd, 80, setting, buf, sizeof(buf));
|
||||
passwd, 80, setting, buf, sizeof(buf), thrid );
|
||||
//printf("hashse='%s'\n", (char *)retval);
|
||||
if (yescrypt_free_local(&local)) {
|
||||
yescrypt_free_shared(&shared);
|
||||
@@ -329,7 +332,7 @@ uint8_t* yescrypt_gensalt(uint32_t N_log2, uint32_t r, uint32_t p, yescrypt_flag
|
||||
|
||||
static int yescrypt_bsty(const uint8_t * passwd, size_t passwdlen,
|
||||
const uint8_t * salt, size_t saltlen, uint64_t N, uint32_t r, uint32_t p,
|
||||
uint8_t * buf, size_t buflen)
|
||||
uint8_t * buf, size_t buflen, int thrid )
|
||||
{
|
||||
static __thread int initialized = 0;
|
||||
static __thread yescrypt_shared_t shared;
|
||||
@@ -349,7 +352,7 @@ static int yescrypt_bsty(const uint8_t * passwd, size_t passwdlen,
|
||||
}
|
||||
retval = yescrypt_kdf(&shared, &local,
|
||||
passwd, passwdlen, salt, saltlen, N, r, p, 0, YESCRYPT_FLAGS,
|
||||
buf, buflen);
|
||||
buf, buflen, thrid );
|
||||
#if 0
|
||||
if (yescrypt_free_local(&local)) {
|
||||
yescrypt_free_shared(&shared);
|
||||
@@ -370,16 +373,16 @@ char *yescrypt_client_key = NULL;
|
||||
int yescrypt_client_key_len = 0;
|
||||
|
||||
/* main hash 80 bytes input */
|
||||
void yescrypt_hash( const char *input, char *output, uint32_t len )
|
||||
int yescrypt_hash( const char *input, char *output, uint32_t len, int thrid )
|
||||
{
|
||||
yescrypt_bsty( (uint8_t*)input, len, (uint8_t*)input, len, YESCRYPT_N,
|
||||
YESCRYPT_R, YESCRYPT_P, (uint8_t*)output, 32 );
|
||||
return yescrypt_bsty( (uint8_t*)input, len, (uint8_t*)input, len, YESCRYPT_N,
|
||||
YESCRYPT_R, YESCRYPT_P, (uint8_t*)output, 32, thrid );
|
||||
}
|
||||
|
||||
/* for util.c test */
|
||||
void yescrypthash(void *output, const void *input)
|
||||
int yescrypthash(void *output, const void *input, int thrid)
|
||||
{
|
||||
yescrypt_hash((char*) input, (char*) output, 80);
|
||||
return yescrypt_hash((char*) input, (char*) output, 80, thrid);
|
||||
}
|
||||
|
||||
int scanhash_yescrypt( struct work *work, uint32_t max_nonce,
|
||||
@@ -392,13 +395,13 @@ int scanhash_yescrypt( struct work *work, uint32_t max_nonce,
|
||||
const uint32_t first_nonce = pdata[19];
|
||||
const uint32_t last_nonce = max_nonce;
|
||||
uint32_t n = first_nonce;
|
||||
int thr_id = mythr->id; // thr_id arg is deprecated
|
||||
int thr_id = mythr->id;
|
||||
|
||||
for ( int k = 0; k < 19; k++ )
|
||||
be32enc( &endiandata[k], pdata[k] );
|
||||
endiandata[19] = n;
|
||||
do {
|
||||
yescrypt_hash((char*) endiandata, (char*) vhash, 80);
|
||||
if ( yescrypt_hash((char*) endiandata, (char*) vhash, 80, thr_id ) )
|
||||
if unlikely( valid_hash( vhash, ptarget ) && !opt_benchmark )
|
||||
{
|
||||
be32enc( pdata+19, n );
|
||||
|
||||
@@ -38,12 +38,13 @@ extern "C" {
|
||||
#include <stdint.h>
|
||||
#include <stdlib.h> /* for size_t */
|
||||
#include <stdbool.h>
|
||||
#include "miner.h"
|
||||
|
||||
//#define __SSE4_1__
|
||||
|
||||
void yescrypt_hash(const char* input, char* output, uint32_t len);
|
||||
int yescrypt_hash(const char* input, char* output, uint32_t len, int thrid );
|
||||
|
||||
void yescrypthash(void *output, const void *input);
|
||||
int yescrypthash(void *output, const void *input, int thrid );
|
||||
|
||||
/**
|
||||
* crypto_scrypt(passwd, passwdlen, salt, saltlen, N, r, p, buf, buflen):
|
||||
@@ -301,7 +302,7 @@ extern int yescrypt_kdf(const yescrypt_shared_t * __shared,
|
||||
const uint8_t * __salt, size_t __saltlen,
|
||||
uint64_t __N, uint32_t __r, uint32_t __p, uint32_t __t,
|
||||
yescrypt_flags_t __flags,
|
||||
uint8_t * __buf, size_t __buflen);
|
||||
uint8_t * __buf, size_t __buflen, int thrid);
|
||||
|
||||
/**
|
||||
* yescrypt_r(shared, local, passwd, passwdlen, setting, buf, buflen):
|
||||
@@ -321,7 +322,7 @@ extern uint8_t * yescrypt_r(const yescrypt_shared_t * __shared,
|
||||
yescrypt_local_t * __local,
|
||||
const uint8_t * __passwd, size_t __passwdlen,
|
||||
const uint8_t * __setting,
|
||||
uint8_t * __buf, size_t __buflen);
|
||||
uint8_t * __buf, size_t __buflen, int thrid);
|
||||
|
||||
/**
|
||||
* yescrypt(passwd, setting):
|
||||
@@ -339,7 +340,7 @@ extern uint8_t * yescrypt_r(const yescrypt_shared_t * __shared,
|
||||
*
|
||||
* MT-unsafe.
|
||||
*/
|
||||
extern uint8_t * yescrypt(const uint8_t * __passwd, const uint8_t * __setting);
|
||||
extern uint8_t * yescrypt(const uint8_t * __passwd, const uint8_t * __setting, int thrid );
|
||||
|
||||
/**
|
||||
* yescrypt_gensalt_r(N_log2, r, p, flags, src, srclen, buf, buflen):
|
||||
|
||||
@@ -79,7 +79,7 @@ int main(int argc, const char * const *argv)
|
||||
for (i = 0; i < sizeof(src); i++)
|
||||
src.u8[i] = i * 3;
|
||||
|
||||
if (yespower_tls(src.u8, sizeof(src), ¶ms, &dst)) {
|
||||
if (!yespower_tls(src.u8, sizeof(src), ¶ms, &dst)) {
|
||||
puts("FAILED");
|
||||
return 1;
|
||||
}
|
||||
|
||||
@@ -51,9 +51,13 @@ int scanhash_yespower_r8g( struct work *work, uint32_t max_nonce,
|
||||
be32enc( &endiandata[ i], pdata[ i ]);
|
||||
endiandata[19] = n;
|
||||
|
||||
// do sha256 prehash
|
||||
SHA256_Init( &sha256_prehash_ctx );
|
||||
SHA256_Update( &sha256_prehash_ctx, endiandata, 64 );
|
||||
|
||||
do {
|
||||
yespower_tls( (unsigned char *)endiandata, params.perslen,
|
||||
¶ms, (yespower_binary_t*)hash );
|
||||
¶ms, (yespower_binary_t*)hash, thr_id );
|
||||
|
||||
if unlikely( valid_hash( hash, ptarget ) && !opt_benchmark )
|
||||
{
|
||||
@@ -73,6 +77,7 @@ bool register_yescryptr8g_algo( algo_gate_t* gate )
|
||||
gate->optimizations = SSE2_OPT | SHA_OPT;
|
||||
gate->scanhash = (void*)&scanhash_yespower_r8g;
|
||||
gate->hash = (void*)&yespower_tls;
|
||||
pk_buffer_size = 26;
|
||||
opt_sapling = true;
|
||||
opt_target_factor = 65536.0;
|
||||
return true;
|
||||
|
||||
Some files were not shown because too many files have changed in this diff Show More
Reference in New Issue
Block a user