Closed gapisback closed 1 year ago
Name | Link |
---|---|
Latest commit | f65c2b5ab8692d84f4923df5f5f790a305e9da6d |
Latest deploy log | https://app.netlify.com/sites/splinterdb/deploys/643953b5c6fca00008453a09 |
HI, @rosenhouse - I re-ran my /shmem
stabilization runs with this increased time out -plus- commented out a few new longish-running unit-test cases. (The problem seems to be on Splinter-side.)
Now the test runs have finished in just less than 2 hours:
See: no. 27: debug-clang: All Tests: [ 1h 52m 41s ] See: no. 27: debug-gcc: All Tests: [ 1h 52m 28s ]
It's just under 2 hours, so I'd still like to bump up the timeout to be safe & clear.
I suspect we will eventually need something 2+ hrs as quite a bit of new tests are coming in.
In-flight stabilization of shared memory support in Splinter is bringin along with tons more additional tests. We are effetively running most of the existing twice; once w/o and once w/ shared memory configured. Debug-build test runs are timing out at 2 hours. Bump timeout to 3hs, and once stabilized, we can look into dropping this to 2h.