DragonX/drg-xmrig
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases 2 Wiki Activity
2,472 Commits 1 Branch 3 Tags
21022fe226405c4b7b6c6c6deaaa58814e655021
Commit Graph
100 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
SChernykh 21022fe226 Also fix RelWithDebIfno build in Visual Studio 2020-10-27 14:25:43 +01:00
SChernykh d822ffef36 Fixed Debug build in Visual Studio 2020-10-27 14:08:36 +01:00
SChernykh ccf13673aa Fixed roundSize() usage in OCL/CUDA 2020-10-15 14:39:54 +02:00
SChernykh 2ffbf20f20 Make round size power of 2 2020-10-15 14:29:15 +02:00
SChernykh 6078052780 Benchmark: added more check hashes and a progress indicator 2020-10-15 08:23:47 +02:00
SChernykh 04ed885a65 Fixed compile errors in Linux 2020-10-14 21:03:21 +02:00
SChernykh a046437d5f Added benchmark and stress test 
Easy to use and zero configuration embedded benchmark/stress test.
 2020-10-14 19:45:05 +02:00
SChernykh f8a0aa06c8 Added argon2/chukwav2 algorithm 
New Turtlecoin algorithm. Source: https://github.com/turtlecoin/turtlecoin/blob/development/src/crypto/hash.h#L57
 2020-10-12 08:26:57 +02:00
SChernykh 1f489ee378 Fixed total hashrate update 
Don't add data points where one of the threads doesn't have hashrate data yet.
 2020-10-10 17:00:30 +02:00
SChernykh d21a911dd0 Fix HashrateInterpolator::addDataPoint 2020-10-10 11:22:19 +02:00
SChernykh 2af99c3d83 More precise hashrate calculation 
- Use only steady timestamp counters to guarantee correctness
- CPU backend: directly measure total hashrate using raw hash counters from each thread; update data more often on ARM CPUs because they're slower
- GPU backends: directly measure total hashrate too, but use interpolator with 4 second lag to fix variance from batches of hashes

Total hashrate is now measured directly (realtime for CPU, 4 seconds lag for GPU), so it might differ a bit from the sum of all thread hashrates because data points are taken at different moments in time.

Overhead is reduced a lot since it doesn't have to go through all threads to calculate max total hashrate on every timer tick (2 times a second).
 2020-10-10 11:18:01 +02:00
SChernykh 25d4142e39 Change to fetch_xor to make code simpler 2020-10-09 11:50:11 +02:00
SChernykh 151e1f7d77 Fix possible race condition in hashrate counting code 
Use single atomic operation to switch between data points.
 2020-10-09 10:29:18 +02:00
SChernykh c370d54e74 Fix 32-bit compilation 2020-10-07 18:19:35 +02:00
SChernykh 7464c4df7c Fixed ARM compilation 2020-10-07 18:09:42 +02:00
SChernykh ba74d14728 Fix FreeBSD compilation 2020-10-07 18:00:36 +02:00
SChernykh 1865d31d0c RandomX: added huge-pages-jit config parameter 
Set to false by default, gives 0.2% boost on Ryzen 7 3700X with 16 threads, but hashrate might be unstable on Ryzen between launches. Use with caution.
 2020-10-07 17:42:55 +02:00
SChernykh 6a25fa75c7 RandomX: allocate 2 MB pages for generated code, if possible 
+0.2% boost on Ryzen 7 3700X
 2020-10-07 10:35:10 +02:00
SChernykh d5dd4ce89f RandomX: removed rx/loki 
Loki forks to PoS on October 9th.
 2020-10-02 17:02:52 +02:00
SChernykh d498274ca9 RandomX: optimized soft AES code 
Unrolled loop was 5-10% slower depending on CPU.
 2020-09-29 21:22:11 +02:00
SChernykh fbb8e955aa RandomX: removed duplicate constatns in Blake2b 2020-09-27 11:50:08 +02:00
SChernykh 36592c2fe0 RandomX: isolate SSE4.1 code to fix crashes on old CPUs 2020-09-27 11:46:32 +02:00
SChernykh 411ba9f80d Fixed SSE4.1 for old CPUs 
Enable SSE4.1 only where it's needed.
 2020-09-27 08:55:57 +02:00
SChernykh c07ee768ed KawPow: fixed OpenCL memory leak 2020-09-25 10:53:24 +02:00
SChernykh c4a083aa8c Fix for ARM compilation 2020-09-23 11:44:08 +02:00
SChernykh b8d604bec4 RandomX improved performance of GCC compiled binaries 
JIT compilator was slower compared to MSVC compiled binary. Up to +0.1% speedup on rx/wow in Linux.
 2020-09-22 13:48:11 +02:00
SChernykh 60faba76e8 RandomX: AES improvements 
- A bit faster hardware AES code when compiled with MSVC
- More reliable software AES benchmark
 2020-09-21 17:51:08 +02:00
SChernykh e94b92b2aa Fixed Cryptonight OpenCL for AMD 20.7.2 drivers 
Vega 64 + Windows 10 + AMD 20.7.2 drivers were broken on Cryptonight algorithms.
 2020-09-19 23:12:05 +02:00
SChernykh 1e38eb3f4f RandomX refactoring, moved more stuff to compile time 
Small x86 JIT compiler speedup.
 2020-09-18 20:51:25 +02:00
SChernykh 3ef7c9b5e6 RandomX: returned old soft AES impl and auto-select between the two 2020-09-15 20:48:27 +02:00
SChernykh f126fa112e RandomX: added performance profiler (for developers) 
Also optimized Blake2b SSE4.1 code size to avoid code cache pollution.
 2020-09-12 23:07:52 +02:00
SChernykh 9a53285c5f RandomX: added SSE4.1-optimized Blake2b 
+0.15% on `rx/0`
+0.3% on `rx/wow`
 2020-09-10 14:28:40 +02:00
SChernykh f26d4274fc RandomX: added parameter for scratchpad prefetch mode 
`scratchpad_prefetch_mode` can have 4 values:
0: off
1: use `prefetcht0` instruction (default, same as previous XMRig versions)
2: use `prefetchnta` instruction (faster on Coffee Lake and a few other CPUs)
3: use `mov` instruction
 2020-09-04 16:16:07 +02:00
SChernykh 1dca648537 Fixed self-select login sequence 
In self-select mode, we only have pool wallet right after login.
 2020-08-25 16:17:48 +02:00
SChernykh 07d9abc90e RandomX JIT: optimized address mask calculation 2020-08-12 16:45:16 +02:00
SChernykh 3d78f30171 Try to allocate scratchpad from dataset's 1 GB huge pages, if normal huge pages are not available 2020-07-31 13:37:22 +02:00
SChernykh 5aae95eddc Force 2 MB pages size in allocateLargePagesMemory() on Linux 2020-07-31 09:55:49 +02:00
SChernykh 3c383c59dc Fixed crash in RelWithDbgInfo MSVC build 
Same problem as in https://github.com/xmrig/xmrig/pull/1784 , fixed with compiler flags this time.
 2020-07-29 10:39:58 +02:00
SChernykh 8c8a46acb8 Added mining on battery setting 2020-07-22 20:21:42 +02:00
SChernykh 51370682f3 Fixed RandomX initialization for VS debug builds 2020-07-21 10:10:07 +02:00
SChernykh 3180e242a9 Cryptonight OpenCL: fix for long input data 2020-07-16 10:39:32 +02:00
SChernykh e8dbb8ca9f Removed cache QoS warning at exit on unsupported CPUs 2020-07-13 20:43:49 +02:00
SChernykh a3f4d91ad7 Cache QoS: fix for seting MSR 2020-07-13 20:30:44 +02:00
SChernykh 5af50b854c RandomX: added cache QoS support 
False by default. If set to true, all non-mining CPU cores will not have access to L3 cache.
 2020-07-13 17:23:18 +02:00
SChernykh 08d71d5144 RandomX: tweaked Ryzen code 
Very small speedup
 2020-07-05 16:06:59 +02:00
SChernykh bb87e9008c KawPow: fixed more duplicate share errors 2020-07-03 21:33:44 +02:00
SChernykh 119a065424 Fix typo 2020-07-02 14:29:52 +02:00
SChernykh d2d8fd48b9 KawPow: fixed rare duplicate share errors 2020-07-02 11:36:31 +02:00
SChernykh aa48c64a05 Fixed DAG initialization on slower AMD GPUs 
Display driver could reset on GPUs with screen connected.
 2020-06-30 16:53:56 +02:00
SChernykh 038629c093 Print error message when MSR mod fails 
Make sure user knows that hashrate is worse than it could be.
 2020-06-26 19:54:06 +02:00
SChernykh 683ca745a8 KawPow: optimized CPU share verification 
- 2 times faster CPU share verification (11 -> 5 ms)
- 1.5 times faster light cache initialization
 2020-06-26 12:31:26 +02:00
SChernykh cf1ee9b3d4 Fixed AstroBWT OpenCL compilation on some systems 2020-06-23 13:55:43 +02:00
SChernykh ccb6cce3f0 Fixed NiceHash disconnects for KawPow 2020-06-12 14:08:00 +02:00
SChernykh 5730b7fd27 Show GPU # when compute error happens 2020-06-11 19:48:53 +02:00
SChernykh 3292679e66 KawPow: fixed crash on old CPUs 
- Use `popcnt` instruction only when it's supported
 2020-06-10 21:49:43 +02:00
SChernykh 82e6e3df6f Merge remote-tracking branch 'upstream/evo' into evo 2020-06-10 20:55:59 +02:00
SChernykh c648e79c97 Merge remote-tracking branch 'upstream/evo' into evo 2020-06-10 08:19:36 +02:00
SChernykh e6f454f37d Fixed detection of AVX2/AVX512 2020-06-09 17:47:23 +02:00
SChernykh 4f0ff0c6b9 Disabled AVX-512F for Argon2 
See #1722
 2020-06-09 13:53:14 +02:00
SChernykh 63dd56831e Fixed GCC 10.1 issues 
- Fixed uninitialized `state->x` warning
- Fixed broken code with `-O3` or `-Ofast`
 2020-06-07 16:23:17 +02:00
SChernykh 230b2a4b63 Fixed GCC 10.1 issues 
- Fixed uninitialized `state->x` warning
- Fixed broken code with `-O3` or `-Ofast`
 2020-06-07 16:16:09 +02:00
SChernykh 2ab164caf2 Conceal (CCX) support 2020-06-07 01:01:45 +02:00
SChernykh 9fd7f42e95 Reduced memory for KawPow 2020-06-05 14:01:49 +02:00
SChernykh 25685ebab1 Print errors from KawPow DAG initialization 2020-06-05 09:33:36 +02:00
SChernykh afdf26616e KawPow: reduced stale/expired shares 2020-05-31 18:22:21 +02:00
SChernykh 3723fa777d Tuned KawPow OpenCL kernel 
Removed unnecessary memory barrier.
 2020-05-29 13:10:24 +02:00
SChernykh 9725e14a02 KawPow: fixed switch back from dev donate 2020-05-29 10:49:22 +02:00
SChernykh 50e1f7217e Update EthStratumClient.cpp 2020-05-29 08:39:03 +02:00
SChernykh d7c9d08614 Added listener verify callback 2020-05-29 08:36:59 +02:00
SChernykh bf45453a0f Fixed stratum login notification 2020-05-29 08:31:53 +02:00
SChernykh 474f97695e Hashrate display fixes 2020-05-28 22:23:31 +02:00
SChernykh ee4c5712e9 Merge remote-tracking branch 'upstream/evo' into evo 2020-05-28 22:06:10 +02:00
SChernykh 2f1940ca58 Fixed hashrate and diff display for KawPow 2020-05-28 22:03:28 +02:00
SChernykh 3d622ff2a9 KawPow fix for retarted AMD OpenCL compiler 2020-05-28 20:27:25 +02:00
SChernykh 7056f67726 KawPow performance fix for AMD Navi 2020-05-28 18:28:23 +02:00
SChernykh 3168919b83 KawPow: tuned work group size for OpenCL 2020-05-28 10:58:06 +02:00
SChernykh f24da62db0 KawPow WIP 2020-05-27 16:19:57 +02:00
SChernykh b8a706948a Optimized RandomX dataset initialization 
- Use single Argon2 implemenation
- Auto-select the fastest Argon2 implementation for RandomX
 2020-05-03 20:44:59 +02:00
SChernykh b2dcacc679 Improved JSON config error reporting 
Show incorrect lines in config.json together with line number and position.
 2020-04-29 11:08:45 +02:00
SChernykh 9a6baa375d Refactored CFROUND 2020-04-21 15:44:04 +02:00
SChernykh c43a848415 Fix code style 2020-04-09 14:31:42 +02:00
SChernykh fe96036607 RandomX JIT refactoring 
- Smaller memory footprint
- A bit faster overall
 2020-04-09 14:24:54 +02:00
SChernykh 3d5a35885e Fixed VM destruction 2020-04-08 08:31:53 +02:00
SChernykh 73cb010b5d Fix off-by-one error 2020-04-07 18:53:08 +02:00
SChernykh 9bd78a9603 Fixed pool lock 2020-04-07 18:48:02 +02:00
SChernykh 5449dfbc91 Use node number instead of affinity 2020-04-07 18:46:22 +02:00
SChernykh e5d43089df Pooled allocation of RandomX VMs 
+0.5% speedup on Zen2 when the whole L3 cache is used (16 threads on 3700X/3800X, 32 threads on 3950X).
 2020-04-07 18:31:35 +02:00
SChernykh e793dd3be2 AstroBWT CUDA support 2020-04-04 11:05:44 +02:00
SChernykh 2ac0f7fa24 Fixed AstroBWT OpenCL for NVIDIA GPUs 2020-03-24 15:55:54 +01:00
SChernykh c2260215b3 AstroBWT OpenCL code 2020-03-22 22:36:21 +01:00
SChernykh f9c84ac456 Fix MacOS compilation 2020-03-11 16:35:52 +01:00
SChernykh 838bd14948 AVX2 optimized code for AstroBWT 
Added "astrobwt-avx2" parameter in config.json, it's turned off ("false") by default.

4-5% speedup on CPUs with proper AVX2 support (AMD Ryzen starting with Zen2, Intel Core starting with Haswell).

There will be no speedup on the following CPUs:

- Intel Pentium/Celeron don't support AVX2
- AMD Zen/Zen+ have only half-speed AVX

GCC compiled version is faster without AVX2, MSVC compiled version is faster with AVX2
 2020-03-10 22:35:14 +01:00
SChernykh ea14d937f8 Activate MSR mod only for RandomX algorithms 2020-03-09 19:10:26 +01:00
SChernykh c565ed93cd Fixed invalid AstroBWT hashes after algo switching 2020-03-07 16:41:33 +01:00
SChernykh 1db5a6d0b8 Fixed compilation withut randomx/argon2 2020-03-06 07:22:57 +01:00
SChernykh 19fc7b6f43 AstroBWT 20-50% speedup 
Skips hashes with large stage 2 size. Added configurable `astrobwt-max-size` parameter, default value is 550, min 400, max 1200, optimal value ranges from 500 to 600 depending on CPU.

- Intel CPUs get 20-25% speedup
- 1st- and 2nd-gen Ryzens get 30% speedup
- 3rd-gen Ryzens get up to 50% speedup
 2020-03-05 12:20:21 +01:00
SChernykh c76bd4fc63 Fixed new block detection for Dero solo mining 2020-03-03 17:53:19 +01:00
SChernykh 1ccf408ba7 AstroBWT algorithm (DERO) support 
To test:

- Download https://github.com/deroproject/derosuite/releases/tag/AstroBWT
- Run daemon with `--testnet` in command line

In config.json:
- "coin":"dero"
- "url":"127.0.0.1:30306"
- "daemon:"true"
 2020-02-29 22:41:24 +01:00
SChernykh c6e9e2308d Optimized CFROUND 
Shorter version using BMI2 instructionns
 2020-02-21 19:00:58 +01:00
SChernykh 6f66a0375f Fixed 32-bit compilation 2020-02-21 16:08:23 +01:00