[illumos-Developer] HAL with RBAC

[Garrett D'Amore]

Looks better, although I notice you have some new cstyle exceptions that
you introduced (function<space><lparen>.)  I guess we're not cstyling
any of HAL.

	- Garrett

On Thu, 2011-06-23 at 17:40 +0100, Andrew Stormont wrote:
> Hi,
> 
> Please see updated patch attached.  Copyrights have been updated and
> #ifdef suns's have been removed.
> 
> Thanks,
> Andy
> 
> On 20/06/2011 17:25, "Garrett D'Amore" <garrett at nexenta.com> wrote:
> 
> >On Mon, 2011-06-20 at 16:39 +0100, Andrew Stormont wrote:
> >> Hi,
> >> 
> >> 
> >> > Also, you need to use Nexenta copyrights on this stuff.  One of the
> >> side effects of working for a company is that your work on stuff
> >> associated with your employer
> >> > becomes the property of your employer.
> >> 
> >> 
> >> Yeah I was a little unsure about that as this was a weekend project of
> >> mine.
> >
> >Yeah, sadly US/California law doesn't recognize your own time unless it
> >is on unrelated projects.  I learned that the hard way myself.
> >
> >> 
> >> 
> >> > You did not check the return value from g_strdup for NULL
> >> 
> >> 
> >> The old code didn't check for NULL either but I see your point.  I
> >> will fix this.
> >
> >Thanks.
> >
> >> 
> >> 
> >> > Is the Hal upstream dead?
> >> 
> >> 
> >> HAL is currently in maintenance state which means they're only
> >> accepting bug fixes.  They do have a solaris backend though which is
> >> very broken so maybe they'd still accept fixes to that.  I'm not sure
> >> if it makes a lot of difference to us either way.
> >> 
> >> 
> >> > Because one effect of these changes that I can see is that it would
> >> be harder to merge from the upstream.
> >> 
> >> 
> >> I wouldn't worry about this making it harder to merge from upstream
> >> because the PolicyKit code we already have in place is completely
> >> incompatible with what upstream has.  Essentially what I've just done
> >> is replaced one incompatible thing with another that has one less
> >> dependency, and HAL is the only user of our current PolicyKit
> >> implementation.
> >
> >This is good information.  Given this, I agree that the divergence from
> >the upstream is probably reasonable.  Given that, does it make sense to
> >just yank the #ifdef sun checks?
> >
> >	- Garrett
> >> 
> >> 
> >> Thanks,
> >> Andy
> >> 
> >> 
> >> From: Garrett D'Amore <garrett at nexenta.com>
> >> Date: Mon, 20 Jun 2011 07:19:49 -0700
> >> To: Andrew Stormont <andrew.stormont at nexenta.com>
> >> Subject: Re: [illumos-Developer] HAL with RBAC
> >> 
> >> 
> >> 
> >> You did not check the return value from g_strdup for NULL.  Also, you
> >> need to use Nexenta copyrights on this stuff.  One of the side effects
> >> of working for a company is that your work on stuff associated with
> >> your employer becomes the property of your employer.
> >> 
> >> 
> >> Is the Hal upstream dead?  Because one effect of these changes that I
> >> can see is that it would be harder to merge from the upstream.
> >> 
> >>   -- Garrett D'Amore
> >> 
> >> On Jun 20, 2011, at 4:34 AM, "Andrew Stormont"
> >> <andrew.stormont at nexenta.com> wrote:
> >> 
> >> 
> >> 
> >> > Hi,
> >> > 
> >> > 
> >> > One thing that really bugs me in Illumos is the existence of
> >> > something calling 'libpolkit' which is effectively a shim that
> >> > translates some PolicyKit-ish like calls to some RBAC calls.  This
> >> > shim gets in the way if you want to use a real PolicyKit
> >> > implementation, which is something I do in StormOS for things like
> >> > GNOME, PackageKit, etc.  What the attached patch does is remove HALs
> >> > dependency on this shim and implements the RBAC stuff directly.
> >> > 
> >> > 
> >> > Webrev: http://cr.illumos.org/view/5pizl4rg/
> >> > 
> >> > 
> >> > Thanks,
> >> > Andy
> >> > <illumos-gate.patch>
> >> > _______________________________________________
> >> > Developer mailing list
> >> > Developer at lists.illumos.org
> >> > http://lists.illumos.org/m/listinfo/developer
> >> > 
> >
> >
> 
> _______________________________________________
> Developer mailing list
> Developer at lists.illumos.org
> http://lists.illumos.org/m/listinfo/developer