psychocrypt
46663db4d2
revert input size change
...
revert #1198 , the block size is limited to 84byte
2018-03-25 21:42:22 +02:00
psychocrypt
f46d29544f
fix input size on device
2018-03-25 15:07:10 +02:00
Juan Leni and psychocrypt
5d2cae77f9
Fixing allocation issue
2018-03-25 15:06:37 +02:00
1e7911e653
XMR-Stak 2.3.0 RC
...
Co-authored-by: psychocrypt <psychocryptHPC@gmail.com >
Co-authored-by: fireice-uk <fireice-uk@users.noreply.github.com >
Co-authored-by: Lee Clagett <code@leeclagett.com >
Co-authored-by: curie-kief <curie-kief@users.noreply.github.com >
2018-03-25 13:28:40 +01:00
psychocrypt
c528f51a76
speedup Volta
...
- enable L1 cache for Nvidia Volta GPUs and newer
- remove explicit cache controll for Volta GPU and newer
This pull request increases the hash rate for Volta GPUs by ~5%
2018-01-30 20:47:56 +01:00
psychocrypt
ea6c147ab2
reduce memory usage for low end gpus
...
reduce memory usage to 1GiB for NVIDIA devices with <=6 SMX
2018-01-22 21:02:05 +01:00
fireice-uk and GitHub
c0eb229107
Merge pull request #464 from psychocrypt/topic-handleCudaErrorCodes
...
handle cuda error codes
2017-12-15 22:33:14 +00:00
psychocrypt
74378a5ab5
fix cuda9.1 compile
...
- fix cuda9.1 compile (remove includ eof device_functions.hpp/ removed with cuda9.1)
- remove NVIDIA Volta gpus for MAC OSX
2017-12-14 21:27:19 +01:00
psychocrypt
fd0c9db6e8
handle cuda error codes
...
handle all error codes from the cuda api calls.
2017-12-10 20:39:47 +01:00
Andrew Whittle
41cf1e8b91
Beautification edit as per fireice-uk's suggestion
...
Makes casting more explicit.
2017-12-09 11:04:39 -07:00
Andrew Whittle
01d6723d5b
Fix compat_usleep for WIN32
...
When compiled with VS2017, the negative applied to the uint wait
time is ignored. Fixed by casting first.
2017-12-09 10:13:52 -07:00
psychocrypt
6b739ece97
fix intention
...
- fix indention
2017-12-09 09:16:03 +01:00
psychocrypt
a317606af9
conservative NVIDIA auto suggestion
...
Be more conservative with the auto suggestion.
- increase bfactor if `smx <= 6`
- limit memory for pascal < GTX1070 to 2GiB
- limt memory for pascal <= GTX1080 to 4GiB
2017-12-08 23:42:43 +01:00
fireice-uk and GitHub
3bc72ec9c0
Merge pull request #399 from psychocrypt/topic-nvidiaErrorWithMessage
...
add message to `CUDA_CHECK...` macros
2017-12-08 11:04:50 +00:00
psychocrypt
e5127b5ffa
add message to CUDA_CHECK... macros
...
- add macro `CUDA_CHECK_MSG_KERNEL` and `CUDA_CHECK_MSG`
- add suggestion of typicle errors can be solved
2017-12-07 22:29:07 +01:00
psychocrypt
4dca64c1a9
option to controll gpu synchronization
...
- add option `sync_mode`
- update auto suggestion and jconf
2017-12-01 21:37:11 +01:00
fireice-uk and GitHub
7cec1ca842
Merge pull request #221 from psychocrypt/fix-cudaLaunchBounds
...
fix CUDA launch bounds usage
2017-11-24 00:56:19 +00:00
psychocrypt
43812b36d7
fix CUDA launch bounds usage
...
fix #191
lauch bounds must be placed before the return type but after the template paramater
2017-11-23 21:25:25 +01:00
psychocrypt
36d78b48b7
fix auto suggestion for low end devices
...
Increase bfactor for all devices with lesser than 6 multi processors.
2017-11-20 22:06:34 +01:00
fireice-uk and GitHub
629542ae4d
Merge pull request #133 from psychocrypt/fix-cudaArchBinaryDetection
...
fix wrong cuda binary arch detection
2017-11-17 10:51:53 +00:00
psychocrypt
d326644d8b
fix nvidia auto suggestion
...
The lmem is still incalculably and crash the miner very often.
Increase the potential lmem usage to 16kiB to respect lmem alignments, ...
2017-11-17 10:04:16 +01:00
psychocrypt
656c55309f
fix wrong cuda binary arch detection
...
fix wrong arch comparsion
2017-11-17 09:32:27 +01:00
Unknown
cabd46f39f
Remove whitespace lines
2017-11-16 13:42:27 +00:00
psychocrypt
70737c8d1a
check gpu architecture
...
- check if the gpu architecture is supported by the compiled miner binary
- remove not supported gpus from the auto suggestion
- disallow the selection of a not supported gpu by hand tuning the config
2017-11-15 16:29:41 +01:00
psychocrypt
0659458523
fix wrong memory detection
...
Free and total memory is only evaluated on the first device.
To detect the gpu memory the gpu must be selected.
- create context on the gpu before the memory is checked
- add smx to the auto detection
- change the result code of `cuda_get_deviceinfo()`
2017-11-15 16:26:58 +01:00
psychocrypt
5acec3ea8e
optimize NVIDIA autosuggestion
...
- avoid creation of a config with zero threads or blocks
- WINDOWS: reduce the used memory for the auto suggestion by the amount of already used memory
2017-11-03 21:46:27 +01:00
psychocrypt
7ecf4cec6e
fix windows compile and broken aeon
...
- fix windows linker error during compile
- fix wrong parameter to call aeon (nvidia-backend)
2017-10-28 21:20:19 +02:00
psychocrypt
3e091d9251
increase safety memory for autosuggestion
...
- increase safty memory from 64 to 128 MiB
- NVIDIA: increase lmem reserve per thread to 1kiB
2017-10-27 21:40:38 +02:00
psychocrypt
12802f12e3
rename xmr to monero
...
- rename all `xmr` to `monero`
- be insensitive while check for set currency
- add function to compate two strings insensitive
2017-10-27 20:13:19 +02:00
psychocrypt
89749c32b7
add aeon support to backend nvidia
...
- add template parameter to kernel to support aeon and xmr
- update auto suggestion
2017-10-27 20:12:38 +02:00
fireice-uk and GitHub
336f2325f9
Merge pull request #57 from psychocrypt/fix-nvidiaBackendCrash
...
fix illegal memory access
2017-10-22 15:21:24 +01:00
psychocrypt
29363d5a1a
fix illegal memory access
...
remove restricted pointer
2017-10-21 20:44:23 +02:00
psychocrypt
18ba334391
fix CUDA 9 shuffle warning
...
use `__shffl_snyc` if CUDA 9+ is avalable
2017-10-20 21:12:50 +02:00
psychocrypt
8babae3156
cleanup includes
2017-09-30 23:46:08 +02:00
psychocrypt
cc429b68fa
group files
...
- move source code to `src`
- categorize files and move to group folder
- change upper case class files to lower case
- change C++ header to `*.hpp`
2017-09-30 23:46:08 +02:00
psychocrypt