Quoting Steve Langasek (vor...@debian.org):

> > Steve, any expertise here?
> 
> I fear this is probably due to the shrink_dead_code patch, whose express
> purpose is to trim functions out of the binaries that aren't used in the
> program - we have no good way to detect that we want to keep these for the
> benefit of external vfs modules, and in this case the external vfs modules
> are the *only* things using this function.


I was fearing something like this...

Could we go back to the reasons that made us use this patch
(apparently the comments have vanished in the patch). IIRC,
this was because of a user mentioning the binaries' "bloat". However,
if reducing that bloat makes us lose some functionality, I would
prefer dropping the patch (which would: 1) make us closer from
upstream 2) save me a lot of time when merging new upstream versions).


Attachment: signature.asc
Description: Digital signature

Reply via email to