SChernykh
1cdab71613
Always reset nonce on RandomX dataset change
...
Also never get a new job when mining is paused
2024-08-14 16:41:03 +02:00
SChernykh
3308b1ace4
Added rx/yada OpenCL support
2024-08-07 00:18:51 +02:00
SChernykh
637c7c9209
Removed deprecated AstroBWTv1 and v2
2022-06-12 11:47:36 +02:00
SChernykh
f9d8461dbd
Dero HE (astrobwt/v2) OpenCL support
2022-03-14 20:13:31 +01:00
SChernykh
3cdfe6b30e
Fixed unaligned memory accesses
2022-01-26 17:18:18 +01:00
XMRig
22f68bda8d
Code cleanup based on Clang-Tidy.
2021-08-25 18:45:15 +07:00
SChernykh
01d731ea5a
AstroBWT OpenCL fixes
...
- Rewrote main BWT kernel to work properly on Navi
- Fixed nonce iterations in OclWorker
- Fixed memory allocation for AstroBWT
2020-12-27 16:44:35 +01:00
XMRig
9a120b979b
Added GpuWorker class.
2020-12-04 22:25:28 +07:00
cohcho
9c3101b419
WorkerJob: fix calls of add/nextRound
...
WorkerJob::nextRound() doesn't require roundSize to be power of 2
Use CudaWorker::intensity(), OclWorker::intensity() to get device batch size
Sync nonce and device iteration
2020-10-15 14:56:37 +00:00
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
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
5730b7fd27
Show GPU # when compute error happens
2020-06-11 19:48:53 +02:00
SChernykh
afdf26616e
KawPow: reduced stale/expired shares
2020-05-31 18:22:21 +02:00
SChernykh
f24da62db0
KawPow WIP
2020-05-27 16:19:57 +02:00
SChernykh
c2260215b3
AstroBWT OpenCL code
2020-03-22 22:36:21 +01:00
XMRig
7828951b51
Fixed nicehash nonce overflow for GPU backends.
2020-02-06 22:00:03 +07:00
XMRig
2872c4737a
Display backend for shares.
2019-10-29 15:43:13 +07:00
XMRig
768f1d9269
Class OclInterleave renamed to OclSharedData and added class OclSharedState.
2019-09-30 05:18:53 +07:00
XMRig
d1d0ed7fe2
Allow partially started threads.
2019-09-17 02:22:59 +07:00
XMRig
c4b49d4213
OpenCL RandomX WIP
2019-09-11 15:48:02 +07:00
XMRig
7c772d733b
Added basic cn/gpu support.
2019-09-08 08:59:17 +07:00
XMRig
a05c40b7e2
Improved OpenCL startup time.
2019-09-06 21:48:15 +07:00
XMRig
61d8686ca5
Restored OpenCL interleave.
2019-09-06 11:43:02 +07:00
XMRig
9b54e4b827
Define double OpenCL threads in simple way.
2019-09-05 09:27:29 +07:00
XMRig
aad2f63362
Implemented verification on CPU.
2019-09-01 19:31:25 +07:00
XMRig
65f5806367
Implemented cn0 kernel launch.
2019-09-01 07:05:49 +07:00
XMRig
18988f5ff0
Added Cn0Kernel and OclKernel classes.
2019-08-28 08:57:55 +07:00
XMRig
1d2bff4de0
Initial stub for setting job in OpenCL backend.
2019-08-28 02:05:19 +07:00
XMRig
897f08183b
Restored OclCache.
2019-08-27 06:31:40 +07:00
XMRig
684321a7f2
Added classes IOclRunner, OclBaseRunner, OclCnRunner, OclRxRunner.
2019-08-26 04:44:01 +07:00
XMRig
fe5832bfc6
Added OclWorker class.
2019-08-25 03:50:49 +07:00
