fixAMD benchmark mode

revert input size change

move option currency down

revert #1198, the block size is limited to 84byte

move option `--currency` down

OpenCL precompiled code message

increase version to 2.3.0

The OpenCl code is only supporting jobs up to 84byte

change warning when precompiled OpenCL code is not found

Fix input buffer size cuda

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>

fix shadowed variable

A redefinition of a variable in a local scope avoid that the intensity is rounded to a multiple of the work size.

Fix #1122 - HTML report violates XML rules

CUDA: reduce startup time

OpenCL compile cache

Comment cleanup

Reduce OpenCL start time by using a self made compiler cache.

- store compiled OpenCL binary
- load OpenCl binary if available

- add project picosha2
- add license notes for picosha2

- reduce startup time for multi gpu systems
- initialize the GPU memory non concurrent

fix broken memchunk feature

fix double definition of define `MEM_CHUNK`

fix compile with OpenCL 1.1

add Mesa OpenCL support

add OpenCL compatibility mode

guard error types those are only defined in OpenCL >1.1

- add new option `comp_mode` to the amd config
- disable `if guards` within opencl kernel if `comp_mode : false`

allow usage of Mesa OpenCl

AMD: option `mem_chunk`and new `strided_index`

AMD: reduce register usage

fix missing message in log file

- add new option for `strided_index`
- add additional option if `strided_index == 2` to controll the memory chunk with

initialize the log file as fast as possible after the start of the miner

reduce usage of registers: based on the suggestion of @enerc77