- 10 Feb, 2019 11 commits
-
-
psychocrypt authored
-
fireice-uk authored
CUDA: update cn_gpu auto adjust values
-
fireice-uk authored
CUDA/OpenCL: use shared mem object
-
fireice-uk authored
better variable nameining
-
psychocrypt authored
Optimize the auto adjustment for cn_gpu based on precomputed occupancy values from CUDA10.
-
psychocrypt authored
Combine the shared memory for a hash within one struct. Reduce the shared memory footprint per hash by 64 byte.
-
psychocrypt authored
- rename variable names like `b` and `bb` to something with a little bit of meaning.
-
fireice-uk authored
add Zelerius coin
-
fireice-uk authored
add cryptonight_v8_half
-
fireice-uk authored
OpenCL: optimize cn_gpu
-
fireice-uk authored
fix groestl, skein and blake
-
- 09 Feb, 2019 3 commits
-
-
psychocrypt authored
Optimize cn_gpu
-
psychocrypt authored
-
psychocrypt authored
based on the suggestion from @xmrig https://github.com/xmrig/xmrig-amd/commit/db4e169f3a78f273abf89ea8cf5bba7eccf1490b
-
- 08 Feb, 2019 3 commits
-
-
psychocrypt authored
This PR replaces #2196 and add support for zelerius.
-
psychocrypt authored
- add cryptonight_v8_half - re-add stellite and masari to the supported coins
-
fireice-uk authored
remove cn_turtle as native POW
-
- 07 Feb, 2019 9 commits
-
-
fireice-uk authored
refactor POW definition
-
fireice-uk authored
cn_gpu: disable auto suggestion for CPU
-
psychocrypt authored
cryptonight_turtle is only cryptonight_v8 with a different scratchpad, iteration and mask value. We are using now the new machanism to describe such derived POWs.
-
psychocrypt authored
A POW is now defined by a function `f` and three degrees of freedom `f(iteration, scratchpad, mask)`. `f` is the base algorithm like `cryptonight, cryptonight_gpu` An easy to pars snytax to write the full POW definition down is: `cryptonight_gpu:0x0000c000:0x00200000:0x001fffc0` This change make it very easy to integrate the new trend of variate the number of iteations or the scratchpad size without modifying the full code.
-
psychocrypt authored
To avoid that a miner is waisting power with there CPU to mine cn_gpu the auto suggestion will comment out the config.
-
fireice-uk authored
OpenCl: use user defined unroll in cn_gpu
-
fireice-uk authored
OpenCL: fix invalid work group size
-
fireice-uk authored
OpenCL: fix groestl
-
psychocrypt authored
@xmrig provided the information that the driver 19.2.1 for vega also create invalid results if pragma unroll is used for the groestl algo.
-
- 06 Feb, 2019 2 commits
-
-
psychocrypt authored
- use the user defined unroll - auto suggestion: - only tune for cn_gpu if this is the main user currency (after a fork) - set unroll to 1 for cn_gpu
-
psychocrypt authored
OpenCl kernel using a larger work group size than configured by the user to increase the occupancy. Depending on the algorithm and device the size is limited. This PR fixes that the user was able to select a invalid work group size.
-
- 04 Feb, 2019 6 commits
-
-
fireice-uk authored
OpencL: fix cn_gpu
-
fireice-uk authored
fix missing algorithm name
-
fireice-uk authored
CUDA. fix static in global kernel
-
psychocrypt authored
If comp_mode is used the code will not compile. - fix compile issue - fix wrong conditions to handle `comp_mode`
-
psychocrypt authored
During the implementation of `cryptonight_gpu` we missed to add the name of the algorithm to jpsock. - add function to translate an algorithm into a name.
-
psychocrypt authored
Remove `static constexpr` within the global kernel. This is not supported by all CUDA versions.
-
- 03 Feb, 2019 3 commits
-
-
fireice-uk authored
update version to 2.8.0
-
fireice-uk authored
update documentation
-
fireice-uk authored
coin cleanup
-
- 02 Feb, 2019 3 commits
-
-
psychocrypt authored
-
psychocrypt authored
- remove driver warning - add link to supported card y the ROCm driver
-
fireice-uk authored
OpenCL: fix Blake hashing
-