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
A long time ago processing was insert to induce an early seg-fault if a port name was misspelled. (This was protection against this error being encounter later in a less obvious way.) Also a test was created to verify the non-matching port was thusly detected.
Currently a bad port name does not induce a seg-fault on 2x2 multi on El Capitan.
It is not clear when a code update removed this failure, because until quite recently 2x2 on all platforms had so test failures that this low priority error wasn't noticed.
The text was updated successfully, but these errors were encountered:
I do not recall ever forcing a segfault in the code. The ConfigGraph's checkForStructuralErrors should be printing a warning about invalid ports. I would be OK upgrading that warning to a Output.fatal() call.
A long time ago processing was insert to induce an early seg-fault if a port name was misspelled. (This was protection against this error being encounter later in a less obvious way.) Also a test was created to verify the non-matching port was thusly detected.
Currently a bad port name does not induce a seg-fault on 2x2 multi on El Capitan.
It is not clear when a code update removed this failure, because until quite recently 2x2 on all platforms had so test failures that this low priority error wasn't noticed.
The text was updated successfully, but these errors were encountered: