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

[BOUNTY $25] Allows multiple kernels to run at the same time #68

Open
wants to merge 15 commits into
base: master
Choose a base branch
from

Conversation

freemo
Copy link
Member

@freemo freemo commented Oct 23, 2017

Closes issue #62

@freemo freemo added the enhancement Adds new functionality label Oct 23, 2017
@freemo freemo requested a review from automenta October 23, 2017 02:45
@automenta
Copy link
Collaborator

if im reading the travis report right, it is showing just one test failure the one i mentioned for both opencl sdk's. this is pretty good. is there a way to confgure travis to also run the CPU (java threadpool and 'alternative threading') configurations also? this is where i noticed issues with static arrays - not that i understand their purpose in a kernel context unless it is some represntation of a shared memory. but even in that case i would expect some more careful way of managing that between kernel threads. so maybe its better just to rip out the static array support completely, i dunno

@freemo
Copy link
Member Author

freemo commented Oct 23, 2017

@automenta travis just provides an environment. It can run JTP or GPU or CPU or whatever the test is configured to run as. So if you create a test that runs under both modes then travis will test under both.

@freemo freemo added bounty $$$ Cash reward! help wanted Have a question? Need assistance? labels Oct 23, 2017
@freemo freemo changed the title Allows multiple kernels to run at the same time [BOUNTY $25] Allows multiple kernels to run at the same time Oct 23, 2017
@freemo
Copy link
Member Author

freemo commented Jan 5, 2018

@automenta nice ill check out your most recent commits. Is it fully functional and ready in your opinion?

@automenta
Copy link
Collaborator

it still needs tested in all cases for each device, including both POCL software libs and GPU hardware libs. but if it doesnt pass everything, it's probably just a few exception cases that shouldnt be too much trouble to fix. i havent tested the PR'd SymmetryLabs code but it shouldnt conflict.

this other branch automenta#2 i'd like to pull in also though i dont understand what it does yet.

@freemo
Copy link
Member Author

freemo commented Jan 6, 2018

@automenta lets chat on gitter when you get a chance

automenta and others added 3 commits February 11, 2018 14:53
@freemo
Copy link
Member Author

freemo commented Feb 11, 2018

@automenta let me know when you think this PR is ready to be reviewed for inclusion.

@codecov-io
Copy link

codecov-io commented Feb 12, 2018

Codecov Report

Merging #68 into master will increase coverage by 1.27%.
The diff coverage is 34.93%.

@@             Coverage Diff              @@
##             master      #68      +/-   ##
============================================
+ Coverage     46.93%   48.21%   +1.27%     
- Complexity      849      891      +42     
============================================
  Files            57       57              
  Lines          9390     9191     -199     
  Branches       1527     1487      -40     
============================================
+ Hits           4407     4431      +24     
+ Misses         4540     4325     -215     
+ Partials        443      435       -8

@freemo
Copy link
Member Author

freemo commented Apr 15, 2018

@automenta do you ever think this PR will be ready? I think last we talked you said it was pretty close?

@freemo freemo removed the help wanted Have a question? Need assistance? label Apr 19, 2018
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bounty $$$ Cash reward! enhancement Adds new functionality
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants