1
0
mirror of https://github.com/GOSTSec/ccminer synced 2025-01-09 14:28:15 +00:00
Commit Graph

108 Commits

Author SHA1 Message Date
Tanguy Pruvot
b9da6c67f5 improve jh512 with vectors (nist5,quark,sib,x11+,zr5)
the main improvement is to reduce asm calls to read global mem

but, a few more regs are used (68 mini vs 64 on SM 5.2)
so reduce the forced launch bounds to allow 80 or 128 regs per thread

Note: cuda 6.5 seems not able to store with v4.u32... (7.5 is fine)
        st.global.v4.u32        [%rd2], {%r3783, %r3824, %r3823, %r3822};
        st.global.v2.u32        [%rd2+16], {%r3821, %r3820};
        st.global.u32   [%rd2+24], %r3819;
        st.global.u32   [%rd2+28], %r3818;
        st.global.u32   [%rd2+44], %r3814;
        st.global.u32   [%rd2+40], %r3815;
        ...

todo, check alexis variant.. but wanted to keep this code before in git...
2017-01-30 07:27:01 +01:00
Tanguy Pruvot
93adb56c8e handle cryptonight light variant
Signed-off-by: Tanguy Pruvot <tanguy.pruvot@gmail.com>
2017-01-09 18:30:59 +01:00
Tanguy Pruvot
066a569357 import xmr, to finish
todo: fix jh cuda and wrong decimal diff (0xffff problem ?)
2017-01-07 01:40:59 +01:00
Tanguy Pruvot
2bbccc5ff4 wildkeccak, basic stratum port of rpc 2.0
scratchpad delete fix and redownload, reduce rejects

(work in progress)
2017-01-06 08:55:15 +01:00
Tanguy Pruvot
1a31d4d2d6 sia: move specific code in a new rpc unit
part 1: longpoll stuff (nanopool)
2016-09-28 01:27:57 +02:00
Tanguy Pruvot
009b013d25 nist5: rename and move source file
build tip: autoreconf && make -j
2016-09-27 00:57:31 +02:00
Tanguy Pruvot
9f2ed5135b lbry maxwell and pascal update (up to 10% on pascal)
Based on alexis78 work and sponsored by LBRY.IO team (thanks)

Release 1.8.2, use cuda 8 for x86
2016-09-22 19:37:17 +02:00
Tanguy Pruvot
683dc0e149 VeltorCoin Streebog based algo (veltor)
also known as "Thor's Riddle"... yes sure ;)

Credits to ocminer who found and "implemented" it.

Note: tested "ok" on x64 and CUDA 6.5 x86, not on 7.5 and 8.0 x86

PS: Don't have the time for a more proper CUDA implementation of Streebog
2016-08-18 18:47:37 +02:00
Tanguy Pruvot
b31fb5316b sia: cuda impl. and suprnova getwork over stratum
Signed-off-by: Tanguy Pruvot <tanguy.pruvot@gmail.com>
2016-08-06 12:56:35 +02:00
Tanguy Pruvot
223077d11a merged sha256d/ripemd
drop ripemd.cu and remove unused kernels
2016-07-16 22:29:12 +02:00
Tanguy Pruvot
5703a74e22 lbry algo (stratum only)
Signed-off-by: Tanguy Pruvot <tanguy.pruvot@gmail.com>
2016-07-13 09:44:13 +02:00
Tanguy Pruvot
85c212eaad implement x11evo algo
Signed-off-by: Tanguy Pruvot <tanguy.pruvot@gmail.com>
2016-05-31 20:05:15 +02:00
Tanguy Pruvot
a7905e6802 some changes to prepare 1.8 and VS2015 support
also move vstudio specific cpuminer-config.h in compat/
2016-05-28 06:27:41 +02:00
Tanguy Pruvot
8658718791 linux: skein2.cu doesnt exist 2016-05-17 22:21:29 +02:00
Tanguy Pruvot
ee042e5381 linux: missed to remove one PCI tag 2016-05-16 21:42:02 +02:00
Tanguy Pruvot
dad0110557 x17 cleanup
haval256 is now 2x faster, but sha512 perf depends a lot on cuda version...
2016-05-09 16:34:18 +02:00
Tanguy Pruvot
7ffe65c262 blake2s algo
Signed-off-by: Tanguy Pruvot <tanguy.pruvot@gmail.com>
2016-03-13 16:50:32 +01:00
Tanguy Pruvot
c3c977da04 fully remove whirlpoolx for builds 2016-02-19 15:52:17 +01:00
Tanguy Pruvot
4944e1a098 mrM4D vnl, with some changes 2016-02-19 11:31:00 +01:00
Tanguy Pruvot
6e95407dcf decred algo for longpoll/getwork
Signed-off-by: Tanguy Pruvot <tanguy.pruvot@gmail.com>
2016-02-11 07:10:46 +01:00
Tanguy Pruvot
e75b26feb4 sib coin algo (X11 + Streebog)
Signed-off-by: Tanguy Pruvot <tanguy.pruvot@gmail.com>
2015-12-31 19:11:48 +01:00
Tanguy Pruvot
61ff92b5b4 never interrupt global benchmark with found nonces
fix some algo weird hashrates (like blake)
and reset device between algos, for better accuracy

but this reset doesnt seems enough to bench all algos correctly...

to test on linux, could be a driver issue...

heavy: fix first alloc and indent with tabs...
2015-11-01 21:12:50 +01:00
Tanguy Pruvot
2308f555c3 simd: cleanup and ignore linux host warning 2015-11-01 13:35:36 +01:00
Tanguy Pruvot
ab5cc7162e refactor: create bench.cpp and algos.h
Also enhance multi-thread benchmark synchro. with pthread barriers
2015-10-11 00:10:27 +02:00
Tanguy Pruvot
3f589cc4db restore the whirlpool algo 2015-10-06 23:37:07 +02:00
Tanguy Pruvot
01f3183c31 bmw algo for MDT, with midstate
which could be extracted from json too

replace a satcoin by another one ;)

Signed-off-by: Tanguy Pruvot <tanguy.pruvot@gmail.com>
2015-08-22 15:01:51 +02:00
Tanguy Pruvot
e3548f46f3 drop animecoin support
no more really minable... just minable in french
2015-08-22 12:35:22 +02:00
Tanguy Pruvot
d4e191610e Import and adapt lyra2v2
not tested on windows and with SM <= 5
2015-08-18 09:27:11 +02:00
Tanguy Pruvot
3788fef9fd Restore whirlpoolx algo 2015-08-15 06:53:40 +02:00
Tanguy Pruvot
c5df142124 Add c11 algo (x11 variant)
Used by Chaincoin and Flaxscript
2015-06-29 11:46:16 +02:00
Tanguy Pruvot
15293d063f remove pluck algo
Supcoin seems.... dead and the algo was not supported on all devices
2015-06-28 20:48:23 +02:00
Tanguy Pruvot
cad3772122 linux: disable arch 5.2 for the noobs 2015-06-23 10:06:48 +02:00
Tanguy Pruvot
6b41234ff1 nvml: get devices vendor names with libpci
made for linux and require libpci-dev (optional)

if libpci is not installed, card's vendor names are not handled...

Note: only a few vendor names were added, common GeForce vendors.

Signed-off-by: Tanguy Pruvot <tanguy.pruvot@gmail.com>
2015-06-22 04:03:22 +02:00
Tanguy Pruvot
576884b199 refactor: move pool functions in pools.cpp
ccminer.cpp begins to be too big...
2015-06-18 04:44:52 +02:00
Tanguy Pruvot
9f5744d4c0 luffa/cube: fine tuning of maxregcount for the 750Ti
This allow to get 69 regs used (tested on linux) 69 or 72 make
the compiler to use 64 regs which is not enough on the 750Ti
for optimal performance...
2015-06-17 03:58:31 +02:00
Tanguy Pruvot
39238f077d diff: import bitcoin bignum class for conversions
BIGNUM is defined in OpenSSL...
2015-06-11 00:03:19 +02:00
Tanguy Pruvot
5636e2be5e remove double reference to luffa algo
doomcoin is dead but the luffa algo is still used (Joincoin)..

keep doom as alias for compat... rename functions...
2015-06-10 22:10:16 +02:00
Tanguy Pruvot
2d5755d1ef linux: add missing scrypt jansson includes 2015-05-26 10:56:23 +02:00
Tanguy Pruvot
298573b45d Drop Whirlpool and whirpoolx algos, no more used...
Vanillacoin is now PoS only
2015-05-19 01:05:38 +02:00
Tanguy Pruvot
4c3892f268 Merged skeincoin algo for SM5+ devices
Should give same or better than SP and klaus versions

Keep old code for older devices and skein2 compat

Linux perf: 750Ti 78MH/s and GTX 970 260MH/s

Signed-off-by: Tanguy Pruvot <tanguy.pruvot@gmail.com>
2015-05-12 21:32:33 +02:00
Tanguy Pruvot
afe57f8341 rename skein2 to c++, no cuda kernel code
and some other changes...
2015-05-12 04:41:11 +02:00
Tanguy Pruvot
73af5c583c neoscrypt: fixes for windows 2015-05-05 16:15:53 +02:00
Tanguy Pruvot
e1b274ef65 neoscrypt: fix bad source file ext (cu/cpp) 2015-05-04 13:37:56 +02:00
Tanguy Pruvot
65c1d787e7 add neoscrypt based on djm34 work
indent, link --intensity, and some clean up

Tested speed on linux ~= 160kH/s on a 750Ti (Black Edition)

To be continued...
2015-05-03 14:38:15 +02:00
Tanguy Pruvot
7edc02633f linux: return to SM 5.0 only default
and tag version 1.6.2
2015-04-21 17:52:09 +02:00
Tanguy Pruvot
ca9ec1b803 scrypt: compile also titan kernel with recent SM 2015-04-21 16:18:21 +02:00
Tanguy Pruvot
9208888c57 scrypt: assign specific compute models to the kernels 2015-04-21 11:12:41 +02:00
Tanguy Pruvot
4f379803d3 scrypt: remove some unused functions/algo checks
there was remains of blake algo
2015-04-21 09:51:09 +02:00
Tanguy Pruvot
9dc78da2ee import and adapt scrypt from cudaminer project
scrypt-jane under work...
2015-04-18 18:24:55 +02:00
Tanguy Pruvot
d58d53f2b2 update README, small changes, prepare release 1.6.1
still need a SM 3.0 fix for skein...
2015-04-14 23:28:00 +02:00