Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

NVIDIA: optimize cryptonight_gpu #2443

Merged
merged 1 commit into from
Jun 1, 2019

Conversation

psychocrypt
Copy link
Collaborator

  • optimize keccak
  • avoid bank conflicts (shared memory) in phase3
  • use __byte_perm for rot operation if maxwell or newer

increase cn-gpu speed by 10%

- optimize keccak
- avoid bank conflicts (shared memory) in phase3
- use `__byte_perm` for rot operation if maxwell or newer
@fireice-uk fireice-uk merged commit c0ea66b into fireice-uk:dev Jun 1, 2019
psychocrypt added a commit to psychocrypt/xmr-stak that referenced this pull request Jun 30, 2019
- fireice-uk#2443 broke cryptonight_v8 and dereived POW e.g. masari,
rcyptonight_v8_half
psychocrypt added a commit to psychocrypt/xmr-stak that referenced this pull request Jun 30, 2019
- with fireice-uk#2443 the possible maximum number of threads per block was
reduced because phase3 used 16 threads per share (is automatically
avoided if thread limit is exeeded)
- warn user if `threads` is to large and adjust to a valid value
@mhmdyj5949
Copy link

خیلی ممنون

pull bot pushed a commit to npwalker/xmr-stak that referenced this pull request Jul 21, 2019
- fireice-uk#2443 broke cryptonight_v8 and dereived POW e.g. masari,
rcyptonight_v8_half
pull bot pushed a commit to npwalker/xmr-stak that referenced this pull request Jul 21, 2019
- with fireice-uk#2443 the possible maximum number of threads per block was
reduced because phase3 used 16 threads per share (is automatically
avoided if thread limit is exeeded)
- warn user if `threads` is to large and adjust to a valid value
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants