On Tue, May 21, 2019 at 02:20:51PM -0500, Eric Sandeen wrote: > > On Tue, May 21, 2019 at 01:59:58PM -0500, Eric Sandeen wrote: > >> So yeah I'm of the opinion that unless it's kernel(-ish?) code it should be > >> positive, and I can send a patch to clean up stuff that's not. > >> > >> I can be swayed by counterarguments if you have them. :) > > > > What speaks against everything is negative? It isn't like returning > > positive errors really is a traditional userspace convention, as that > > is return -1 (negative!) and look at errno.. > > Sorry, I wasn't clear - I meant returning negative errnos. That's > the part that's not consistent. Yes. And for libxfs/libfrog/etc stuff I think sticking to that and always returning negative error values sounds sanest to me. Note that some userspace libraries have adopted that calling convention, for example libaio.
