You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Debian's CI for arm64 has recently started failing this test more often than not.
I increased the sleep to 100 and some local tests show that to be more reliable, but it would be better if there were a deterministic way to know when the async callback has been run.
I was thinking of a semaphore, but I noticed that uv API doesn't appear to be exposed in luv. Open to ideas, but for now I'll probably patch the test to use a longer sleep.
The text was updated successfully, but these errors were encountered:
…en async work is done
On slow / busy systems, it may take longer than 10ms for the async
callback to run. When this happens, the test process typically crashes
since the async context has already been freed by the time it finally
gets scheduled to run.
Using a semaphore ensures the work is done before continuing through the
test.
Closes: luvit#715
…en async work is done
On slow / busy systems, it may take longer than 10ms for the async
callback to run. When this happens, the test process typically crashes
since the async context has already been freed by the time it finally
gets scheduled to run.
Using a semaphore ensures the work is done before continuing through the
test.
Closes: luvit#715
Debian's CI for arm64 has recently started failing this test more often than not.
I increased the sleep to 100 and some local tests show that to be more reliable, but it would be better if there were a deterministic way to know when the async callback has been run.
I was thinking of a semaphore, but I noticed that uv API doesn't appear to be exposed in luv. Open to ideas, but for now I'll probably patch the test to use a longer sleep.
The text was updated successfully, but these errors were encountered: