- Oct 05, 2018
-
-
psychocrypt authored
Helper functions to select the asm version based on the number of used hashes per threads and the family name of the cpu. - use the noew cpu type functions to fix the wrong AMD family detection in `autoAdjust.hpp` - allow to set the asm version to `auto` - rename asm option `intel` to `intel_avx` - rename asm option `ryzen` to `amd_avx` Co-authored-by:
fireice-uk <fireice-uk@users.noreply.github.com>
-
- Oct 04, 2018
-
-
Tony Butler authored
-
Tony Butler authored
-
- Oct 01, 2018
-
-
psychocrypt authored
I disabled a few algorithms for fatser compile and missed to re-enable them.
-
psychocrypt authored
- restructe asm preparation function - add double hash asm code
-
psychocrypt authored
`uint` is unknown in windows, therefore switch to the better type `uint32_t`
-
- Sep 30, 2018
-
-
psychocrypt authored
In the auto adjust without hwlock the asm entry was missing
-
psychocrypt authored
- introduce a new schema where two threads work together on one hash - update autoadjustment - remove an mistake where shared memory was shrinked for gpus < sm_70
-
psychocrypt authored
apply optimizations Co-authored-by:
SChernykh <sergey.v.chernykh@gmail.com>
-
psychocrypt authored
-
psychocrypt authored
add cpu implementation for the final monero POW
-
- Sep 24, 2018
-
-
Tony Butler authored
-
- Sep 23, 2018
-
-
BBSCoin Developer authored
-
BBSCoin Developer authored
-
- Sep 22, 2018
-
-
Tony Butler authored
-
- Sep 21, 2018
-
-
psychocrypt authored
- remove unused host function (relict from old refactoring) - remove unused OpenCL full div function
-
psychocrypt authored
Avoid branche differegence
-
- Sep 19, 2018
-
-
psychocrypt authored
- fix that shared memory for fast div is always used even if an algorithm is not using it - optimize fast div algo - store `division_result` (64_bit) per thread instead of shuffle around and store it as 32bit
-
psychocrypt authored
- use optimzed div and sqrt - reduce memory footprint
-
SChernykh authored
Add fast version for div and sqrt for the cuda backend
-
psychocrypt authored
- fix code style issues - fix spelling issue - fix asm to support newer clang versions
-
psychocrypt authored
-
psychocrypt authored
- reintroduce monero7 until the POW is final - update docs (add cryptonigh_v8)
-
psychocrypt authored
-
psychocrypt authored
- add special asm version for win64 and linux - add cmake path for MSVC and other systems
-
psychocrypt authored
- fix assembler code to pass the clang compiler - CMake: set asm file language - fix icc with gcc-7 compile issue with `_addcarry_u64`
-
psychocrypt authored
Remove the asm option `auto` by `off`
-
psychocrypt authored
If single hash is used the type of the variable to hold the intermediat sqrt value is changed from `__m128i` to `uint64_t` as suggested by @SChernykh
-
psychocrypt authored
- add new option to `cpu.txt` named `asm` to select the asm code version - extent function selection method to choose assembler code for `cryptonight_v8` - update auto adjustment to add default value for option `asm`
-
psychocrypt authored
Add @SChernykh assembler version for ryzen and intel processors. Co-authored-by:
SChernykh <sergey.v.chernykh@gmail.com>
-
psychocrypt authored
add option `unroll` for OpenCL to allow better tuning the main POW kernel.
-
psychocrypt authored
Create a special pass for NVIDIA GPUs to load memory chunks first into the shared memory. Co-authored-by:
SChernykh <sergey.v.chernykh@gmail.com>
-
psychocrypt authored
- use shared memory to exchange
-
psychocrypt authored
implement `cryptonight_v8`
-
psychocrypt authored
- implement cryptonight_v8 - update auto adjust to fit the special requirements of `cryptonight_v8` - add fast math integer implementation for `sqrt`, `reciprocal` and `division` Co-authored-by:
SChernykh <sergey.v.chernykh@gmail.com>
-
psychocrypt authored
Add support for single hash cryptonight_v8. Co-authored-by:
SChernykh <sergey.v.chernykh@gmail.com>
-
psychocrypt authored
- rmeove currency `monero7` - introduce `cryptonight_v8` and `monero8`
-
psychocrypt authored
During the initialization of the compile parameter for OpenCL it could be that the fixed size buffer is to small. To avoid this we are now using `std::string`. There is no problem by using `std::string` because this part of code is not perfromance critical.
-
psychocrypt authored
If the first bit of the nonce is `1` (this is very often if we use a nicehash pool) than it could be that some OpenCL implementations handle the 64bit representation of the 32bit nonce on the device side as signed integer. During a right bitshift we pull wrong ones from the wrong higher part of the 64bit nonce representation into the 32bit part of the nonce. The result will be that the computed share is invalid. - explicit cast the nonce on the device to `uint` to avoid any side effects
-
- Sep 17, 2018
-
-
psychocrypt authored
Avoid that a OpenCL binary from the cache is used if the driver or xmr-stak version has changed.
-