Closed brucespang closed 10 years ago
See also https://github.com/facebook/libphenom/pull/46 in which we thought that it would make sense to unbundle CK to make this a bit easier, or at least, make these pull requests smaller and easier to review :-)
Do you have the cycles to do the unbundling? I think the ideal path forward here is:
I'm on PTO for a bit so will be online sporadically, but I'm happy to assist with this when I get my hands on a keyboard; that probably means mostly reviewing things but depends on how badly I suffer coding withdrawal ;-)
Regarding benchmarks, there's tests/bench/run-pipes.php
that repeatedly runs some benchmarks with varying concurrency and records the data then produces some nice charts using R
. You can use that with the before/after to see if there are obvious regressions.
and thanks for looking at this! :)
Sounds good, thanks for the feedback! I definitely have time to work on this.
I've create PR #60 to do the unbundling. I'll have another pull request soon with the api change.
I've created PR #61 for the changes to libphenom's use of ck_ring. I'm going to close this pull request, since we're now tracking it in the other two.
I have a libphenom application that uses concurrency-kit extensively. It needs some things from ck that are not available in 0.3.0. This upgrades to ck-0.4.1 to make those things available.
There were some changes to libphenom that were necessary due to breaking api changes in ck (https://github.com/sbahra/ck/commit/3edb523da5f6c9dcea41879ceb6643dcc7bde305). I've updated job.c and job.h to use the new api.
The libphenom tests pass, and I haven't seen any issues in my application level testing.
Some questions: