How to waste 3x as much time when you get an unrelated build failure. 
-
-
-
It's my 5s solution for unsupervised build on a system where you expect OOM.
-
And only 3x on the final failing command not the whole build.
-
Ah, good point.
End of conversation
New conversation -
-
-
Ah, nice, but not to be confused for a solution to the problem where make -j>1 leaves corrupted files that seem up-to-date and prevent future compilations.
-
AIUI GNU make mitigates that by actively deleting the target when the command fails, and gcc etc avoid leaving bad partial output too. But yeah you can break this if you try...
-
Easy to get wrong with OCaml: - bytecode compiler generates t.cmi and t.cmo from http://t.ml d.cmi - native compiler generates t.cmi and t.cmx from http://t.ml d.cmi d.cmx
-
In fact we eventually settled on a building discipline that allows “make -j” but it took several tries to get right.
End of conversation
New conversation -
-
-
|| make -j0.5 || make -j0.25 || ...
-
zeno’s build failover
-
I mean, there should be a script here that accepts a (sparse?) array of coefficients to define a polynomial that defines the desired backoff behavior... ... ... *code code code*
End of conversation
New conversation -
-
-
Start the day off with a forkbomb!
make -jThanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.