Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 15 Jun 2006 14:35:04 -0400
From:      John Baldwin <jhb@freebsd.org>
To:        Divacky Roman <xdivac02@stud.fit.vutbr.cz>
Cc:        Perforce Change Reviews <perforce@freebsd.org>
Subject:   Re: PERFORCE change 99144 for review
Message-ID:  <200606151435.04624.jhb@freebsd.org>
In-Reply-To: <20060615081925.GB34336@stud.fit.vutbr.cz>
References:  <200606131712.k5DHCIjF063625@repoman.freebsd.org> <200606131344.08736.jhb@freebsd.org> <20060615081925.GB34336@stud.fit.vutbr.cz>

next in thread | previous in thread | raw e-mail | index | archive | help
On Thursday 15 June 2006 04:19, Divacky Roman wrote:
> On Tue, Jun 13, 2006 at 01:44:08PM -0400, John Baldwin wrote:
> > On Tuesday 13 June 2006 13:12, Roman Divacky wrote:
> > > http://perforce.freebsd.org/chv.cgi?CH=99144
> > > 
> > > Change 99144 by rdivacky@rdivacky_witten on 2006/06/13 17:12:02
> > > 
> > > 	Make the LINUXBASE path settable via sysctl. Useful when you have more 
then 
> > one linux bases
> > > 	installed etc.
> > 
> > Since you only check the one global variable, why did you make the 
sysctl's 
> > try to handle per-prison paths?
> 
> well... none of the per-prison variables is checkd against anywhere so I
> thought there's some magic behind and just copied the code from the other
> variables.
> 
> can you explain me how this works? to me it seems either like unfinished 
work
> or nonsense :)

All the other per-prison variables are accessed via the linux_get_foo()
functions (like uname, etc.) and those functions look in the prison.  However,
you are passing linux_emul_path directly to calls to kern_alternate_path()
(see the CHECKALT macros).  You would have to change those macros to instead 
call a linux_get_path() function that looked in the prison if you wanted 
different prisons to have their own compat paths.

-- 
John Baldwin



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200606151435.04624.jhb>