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
dc4e379345
fix broken memchunk feature
...
fix double definition of define `MEM_CHUNK`
2018-02-21 22:41:43 +01:00
fireice-uk and GitHub
aa34bac139
Merge pull request #1101 from psychocrypt/topic-mesaSupport
...
add Mesa OpenCL support
2018-02-21 21:31:51 +00:00
psychocrypt
cff6b6cbfb
add OpenCL compatibility mode
...
- add new option `comp_mode` to the amd config
- disable `if guards` within opencl kernel if `comp_mode : false`
2018-02-19 22:13:49 +01:00
psychocrypt
7b3929dbef
add Mesa OpenCL support
...
allow usage of Mesa OpenCl
2018-02-19 21:58:17 +01:00
fireice-uk and GitHub
f785481bb7
Merge pull request #1087 from psychocrypt/topic-blockedStride3
...
AMD: option `mem_chunk`and new `strided_index`
2018-02-19 14:10:36 +00:00
psychocrypt
737185ee82
AMD: mem_chunkand new strided_index
...
- add new option for `strided_index`
- add additional option if `strided_index == 2` to controll the memory chunk with
2018-02-17 20:51:55 +01:00
psychocrypt
b0d03b3302
AMD: reduce register usage
...
reduce usage of registers: based on the suggestion of @enerc77
2018-02-14 21:21:51 +01:00
psychocrypt
ac474caa8b
remove usage of rotate
...
revert the change that the OpenCl function `rotate` is used instead of `SKEIN_ROT`
2018-01-24 22:11:47 +01:00
Grant Galitz and psychocrypt
b9fb744a10
author update
2018-01-24 22:11:47 +01:00
Grant Galitz and psychocrypt
617af4b301
Optimize Skein
...
- Eliminate modulus math (It runs slow inside microcode).
- Convert whatever the hell was going on into a rotate op. Like... Someone kinda reinvented the wheel in order to do a simple rotate.
2018-01-24 22:11:47 +01:00
psychocrypt
58db6082a3
differgence in OpenCL code
...
remove branch differgences in AMD OpenCl code based on #454
a Please enter the commit message for your changes. Lines starting
2017-12-30 21:16:35 +01:00
fireice-uk and GitHub
c6f56b24ec
Merge pull request #332 from psychocrypt/topic-tuneAMD
...
optimize aes round command queue
2017-12-04 15:48:58 +00:00
psychocrypt
efc8dc447a
optimize aes round command queue
...
unroll AES round by hand to help the compiler to optimize the command queue
2017-12-03 16:40:42 +01:00
psychocrypt
d10d20f113
add AMD scratchpad indexing option
...
Allow to change the indexing used to address the hash scratchpad memory.
- add option `strided_index` for each gpu
2017-12-02 21:15:35 +01:00
psychocrypt
2346c8be20
add eon support to amd backend
...
- add compile parameter to support aeon and xmr
- update auto suggestion to handle aeon
2017-10-27 20:12:01 +02:00
psychocrypt
8888a60898
amd: use 64bit target
...
- remove 32bit target value
- use always 64bit target for amd backend
2017-10-24 23:22:28 +02:00
psychocrypt
2c2d05e57a
remove early returns
...
Early returns within a kernel can have unexpected behavior, this strongly depends on the opencl
runtime compiler. To avoid errors all early returns are removed.
There is no negative effect(e.g.performance) if the threads stay alive up to the end of the kernel.
One source for the early return is: http://al-key-opencl.blogspot.de/2014/09/be-careful-not-to-mix-early-return-and.html
This pull request also fix a race condition where all fill the shared memory.
Also a bug introduced with #16 is fixed, because of the early return not was possible that
the last block works with an wrong initilized shared memory (result should be a wrong hash if the result target size is valid).
2017-10-10 21:04:37 +02:00
psychocrypt
611378eb96
avoid possible illegal memory access
...
Is is possible that the number of results are greater than the result output array,
in this case invalid memory can be access within the device and on the host side.
2017-10-06 21:50:51 +02:00
psychocrypt
0304a7b09e
fix invalid kernel call parameter
...
The number of threads within a kernel must be a multiple of the worksize.
If not it can crash on some systems.
2017-10-05 00:27:37 +02:00
psychocrypt
bb3e8840d7
fix windows compile
...
- string literal can only be 16kbyte
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
