> Out of sheer curiosity: in what situations does that happen for you? *If* you > are working on a single branch, ie. you're not switching back and forth > between master and your feature branches, this should not happen (and even if > you switch between branches it should still be safe for most of the time).
I'm actually swtiching branches quite a lot, because most of the time I have multiple tasks at hand, working on multiple tickets etc. But I think there's another way to introduce this kind of failures. Since compilation take very long, I never start the build process and wait until it's done, I keep coding in the meantime. Maybe I should rsync changes manually to another directory and run build there while I keep working on current tree etc. I just tried `make 1` in ghc directory and I think it worked(I'll try on a clean tree once I have to `make clean` and see if it's really building only stage-1). But my next question is why marking it as such in build.mk doesn't work.. I'd expect `devel1` to build stage1 only. 2015-10-20 3:12 GMT-04:00 Jan Stolarek <jan.stola...@p.lodz.pl>: >> it's very easy to end up in a state where `make` simply can't build, and only >> easy fix is a `make clean` > Out of sheer curiosity: in what situations does that happen for you? *If* you > are working on a > single branch, ie. you're not switching back and forth between master and > your feature branches, > this should not happen (and even if you switch between branches it should > still be safe for most > of the time). Richard already gave you a trick (cd ghc/ && make 2) that > allows to build only > stage 2 compiler without rebuilding stage 1 or any of the libraries. One > situation that comes to > my mind that can really require a full rebuild is changing format of > interface files. But other > than that you shouldn't have to do a full rebuild too often. > > I know how you feel. When I started working on GHC I had to do a full rebuild > very often. But now > I improved my workflow and only do a full build after a major rebase of my > branch against master. > > Janek > > --- > Politechnika Łódzka > Lodz University of Technology > > Treść tej wiadomości zawiera informacje przeznaczone tylko dla adresata. > Jeżeli nie jesteście Państwo jej adresatem, bądź otrzymaliście ją przez > pomyłkę > prosimy o powiadomienie o tym nadawcy oraz trwałe jej usunięcie. > > This email contains information intended solely for the use of the individual > to whom it is addressed. > If you are not the intended recipient or if you have received this message in > error, > please notify the sender and delete it from your system. _______________________________________________ ghc-devs mailing list ghc-devs@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs