Date: Sat, 1 Oct 2011 12:43:23 +0000 From: Alexander Best <arundel@freebsd.org> To: Edward Tomasz Napiera?a <trasz@FreeBSD.org> Cc: svn-src-head@freebsd.org, svn-src-all@freebsd.org, src-committers@freebsd.org Subject: Re: svn commit: r225868 - head/bin/ps Message-ID: <20111001124323.GA14050@freebsd.org> In-Reply-To: <B0E90A39-22E0-478A-BA23-5B8E697CA1DE@FreeBSD.org> References: <201109290631.p8T6VgJ3008377@svn.freebsd.org> <20110929121457.GA53600@freebsd.org> <11B87D92-4C1E-4AD9-BEC5-13D28B022FB1@FreeBSD.org> <20110929185700.GA18684@freebsd.org> <B0E90A39-22E0-478A-BA23-5B8E697CA1DE@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Fri Sep 30 11, Edward Tomasz Napiera?a wrote: > Wiadomo?? napisana przez Alexander Best w dniu 29 wrz 2011, o godz. 20:57: > > On Thu Sep 29 11, Edward Tomasz Napiera?a wrote: > >> Wiadomo?? napisana przez Alexander Best w dniu 29 wrz 2011, o godz. 14:14: > >>> On Thu Sep 29 11, Edward Tomasz Napierala wrote: > >>>> Author: trasz > >>>> Date: Thu Sep 29 06:31:42 2011 > >>>> New Revision: 225868 > >>>> URL: http://svn.freebsd.org/changeset/base/225868 > >>>> > >>>> Log: > >>>> Make ps(1) automatically size its column widths. > >>> > >>> cool to see this committed. :) > >>> > >>> i think i wrote you about the two spaces between the "TT" and "TIME" field > >>> some time ago. you said that the "TT" entry can contain a postfix "-". can > >>> you describe a situation, where this can occur? i've never seen the "TT" field > >>> contain an entry larger than 2 chars. > >> > >> Login via ssh, run "nohup sleep 1000 &", then logout. > > > > maybe it would also be possible to introduce unit symbols for the "vsz" and > > "rss" fields. they are getting really huge on systems with a long uptime > > (*cough* mem leaks) and kilobyte granularity isn't really that great when > > those fields are nearing 1GB. > > I thought about it, but the "-h" option is already taken. maybe making unit symbols the default behavior and introducing a -k flag, which toggles between the unit symbol mode and the traditional kilobyte mode? we could then add an envar $PS, so users can set PS=-k to restore the historic ps(1) behavior. a note to the BUGS section could then be added, similar to the note in top(1)'s BUGS section regarding the -I flag, which explains that ps(1)es semantics were changed, but can be restored via PS=-k. cheers. alex > > -- > If you cut off my head, what would I say? Me and my head, or me and my body?
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20111001124323.GA14050>