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

hash count 0 (problem or NOT?) & DAG file speed up #4

Open
RezaAhmadi0117 opened this issue Jun 24, 2021 · 2 comments
Open

hash count 0 (problem or NOT?) & DAG file speed up #4

RezaAhmadi0117 opened this issue Jun 24, 2021 · 2 comments

Comments

@RezaAhmadi0117
Copy link

Hi Ed, I've 2 question.
when I try to run search kernel on u50, I get hash count 0 in each step on searching process. it that ok ? look like below:

Search: target 0x15798ee230
Search: start nonce            0, hash count 0
Search: Hash rate  0.03 Mh 
Search: start nonce      8388608, hash count 0
Search: Hash rate  0.07 Mh 
Search: start nonce     16777216, hash count 0
Search: Hash rate  0.07 Mh 
Search: start nonce     25165824, hash count 0
Search: Hash rate  0.07 Mh 
Search: start nonce     33554432, hash count 0
Search: Hash rate  0.07 Mh 
Search: start nonce     41943040, hash count 0
Search: Hash rate  0.07 Mh 
Search: start nonce     50331648, hash count 0

and I made a few change in your code and pass DAG generation to CPU, now it going soo faster that u50 to generate DAG. now how can I send DAG in m_context to another context?

i put my issue and files in this link

@Ed-Yang
Copy link
Owner

Ed-Yang commented Jun 26, 2021

For Xilinx, I cannot find "atomic_inc" equivalent function, so the hashCount is not increased.

#ifdef FAST_EXIT

@Ed-Yang
Copy link
Owner

Ed-Yang commented Jun 26, 2021

For transferring the DAG, it looks like programming on host memory and pass the DAG memory to HBM on U50. It is suggested to check the OpenCL API, it seems could be achieved by passing different parameters.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants