On Fri, Sep 23, 2011 at 11:45, Mikhail Gusarov <[email protected]>wrote:
> > Twas brillig at 10:40:06 23.09.2011 UTC-04 when [email protected] did gyre > and gimble: > > MF> i don't see any justification to back up this statement other than > MF> "this is what glibc does". i'm honestly not too concerned by that > MF> as there are places already we deviate from glibc behavior by > MF> design. what i do want to know is what the ELF spec says. could > MF> you look that up (i dont have a copy off hand and the LF is down > MF> still) ? > > As a matter of anecdote, BSD linkers don't use executable's RPATH in > this situation. > config option? As a past embedded guy, I like the idea of this being around as a way to fix dependency issues based on how developers are accustomed to coding on common platforms. Allan -- [email protected] "金鱼" http://linkedin.com/in/goldfish _______________________________________________ uClibc mailing list [email protected] http://lists.busybox.net/mailman/listinfo/uclibc
