Date: Fri, 3 May 1996 23:08:39 +0200 (MET DST) From: Wilko Bulte <wilko@yedi.iaf.nl> To: terry@lambert.org (Terry Lambert) Cc: FreeBSD-hackers@FreeBSD.org Subject: Re: used to think I understood ps .. Message-ID: <199605032108.XAA03044@yedi.iaf.nl> In-Reply-To: <199605031838.LAA14600@phaeton.artisoft.com> from "Terry Lambert" at May 3, 96 11:38:32 am
next in thread | previous in thread | raw e-mail | index | archive | help
> > 1000 3055 298 12 -18 0 140 7588 vnread D+ p2 0:20.99 cmp -l two > > > > Here RSS >> VSZ apparantly. I was under the impression VSZ <= RSS ?? > > > > FWIW the cmp compares two 120Mb files (raw satellite images) > > I would think this would be legal for a program with a shared > libc that hasn't used most of libc yet, but the libc is in core > because other programs have? > > Terry Lambert I'm not sure I understand your point. Considering: # size libc.so.2.1 text data bss dec hex 339968 16384 48004 404356 62b84 I'd assume a worst case program would need all of libc. The approx 7.5Mb RSS seems quite steep (to say the least). What makes ps decide to include shared lib code in the RSS calculation? (maybe RTFS ?) Would not it be fair to consider libc.so are shared burden that is system wide desperately needed? Maybe it's just me.. But please try to enlighten me. Wilko _ __________________________________________________________________________ | / o / / _ Wilko Bulte email: wilko@yedi.iaf.nl |/|/ / / /( (_) Private FreeBSD site - Arnhem - The Netherlands --------------------------------------------------------------------------------
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199605032108.XAA03044>