release 2.8.3

set cn_gpu for Ryo

- remove fork for Ryo from cn_heavy to cn_gpu (now always cn_gpu)
- add example pool for ryo

increase version to 2.8.3

OpenCL: fix Vega invalid shares (windows)

CMakeLists.txt: Adjust OpenCL probing to find OCL SDK Light

Please fix Zelerius iterations

Implement cn-conceal

Vega GPUs create still invalid shares on windows 19.X and 18.7+ driver for cn_v8 and cn_heavy.

Add `OCL_ROOT` envvar support for Windows OCL SDK Light autodetection and prefer it over AMD APP SDK if both are installed

fix bfactor


fix

release 2.8.2

increase version to 2.8.2

OpenCL: fix the fix for the blockchain driver

 #2229 was not solving the issues

- revert #2229
- introduce the working fix

OpenCl: fix blockchain driver support

The OpenCl version of the blockchain driver is not understanding if
apointer to a pointer points into shared memory and throw an error
during the compilation.

- revert the usage of the struct to group all shared memory arrays

release 2.8.1

update algorithm documentation

- add more information if a algorithm derives from an other
- update `pools.tpl`

update version to 2.8.1

OpenCl: optimize cn_gpu

Add seperate kernel to prepare the scratchpad memory.

CUDA: update cn_gpu auto adjust values

CUDA/OpenCL: use shared mem object

better variable nameining

Optimize the auto adjustment for cn_gpu based on precomputed occupancy
values from CUDA10.

Combine the shared memory for a hash within one struct.
Reduce the shared memory footprint per hash by 64 byte.

- rename variable names like `b` and `bb` to something with a little bit
of meaning.

 add Zelerius coin

add cryptonight_v8_half

OpenCL: optimize cn_gpu

fix groestl, skein and blake

Optimize cn_gpu

based on the suggestion from @xmrig https://github.com/xmrig/xmrig-amd/commit/db4e169f3a78f273abf89ea8cf5bba7eccf1490b