Re: commit 52ef73ac8c40ea2a8ca80cf96d52a836bcaf76c2 breaks vmware

[Christian Hammond <chipx86 chipx86 com>]

Okay, just to keep people in the loop.

We are indeed doing a couple stupid things with flags. Why it's breaking us, I don't know yet (hard to say whether it's specific to a certain version of our product, or if it's the fact that I had to just backport this patch to test, or that I'm in a devel environment for the product).

So, we have code that wraps some of the option parsing stuff, and it looks like we're stupidly defaulting to G_OPTION_FLAG_NO_ARG for a bunch of non-callback options. We're also in one place using G_OPTION_FLAG_OPTIONAL_ARG without a callback. Clearly wrong, and I'll work on a fix for this, but it's still important we identify why exactly it's hanging and hopefully keep that from happening (breaking old versions of Workstation would be very problematic).

Jeff, what version of Workstation are you running?

Christian

-- 
Christian Hammond - chipx86 chipx86 com
Review Board - http://www.reviewboard.org

VMware, Inc. - http://www.vmware.com

On Tue, Mar 8, 2011 at 4:43 PM, Christian Hammond <chipx86 chipx86 com> wrote:

Jeff, would you mind temporarily commenting out the last conditional involving G_OPTION_ARG_CALLBACK and see if things work again? If not, it would help me to know which of the conditionals is breaking us, so going through and testing each would be awesome. I suspect it's the last check, though.

We may be capturing any g_warnings and directing them to the log file. If the log file is in fact not being generated yet, or not being shown, then maybe going through and temporarily replacing the g_warnings with printfs would help.

It'll be interesting to find out if this is a bug in our software or in the new GTK+ change. Either way, though, hopefully the new change won't blow up old versions :(

Christian

--
Christian Hammond - chipx86 chipx86 com
Review Board - http://www.reviewboard.org
VMware, Inc. - http://www.vmware.com

On Tue, Mar 8, 2011 at 4:30 PM, Christian Hammond <chipx86 chipx86 com> wrote:

Hi Jeff,

It's possible our logging code hasn't been reached at this point, but if you run vmware from the command line, it would normally tell you the path to the UI log file. Having that would be awesome, if it does show it.

I'm looking through this commit and trying to figure out why we break. We're kind of in heads-down deadline mode here right now so I can't do a lot of investigation, but I'll report back if anything looks obvious.

Christian

--
Christian Hammond - chipx86 chipx86 com
Review Board - http://www.reviewboard.org
VMware, Inc. - http://www.vmware.com

On Tue, Mar 8, 2011 at 4:17 PM, Jeff Chua <jeff chua linux gmail com> wrote:

Commit 52ef73ac8c40ea2a8ca80cf96d52a836bcaf76c2 breaks vmware 7.1.3

VMware hangs on startup. Nothing displayed. Reverting the commit solves the problem. Sorry, I couldn't find any logs in vmware.

Thanks,
Jeff



commit 52ef73ac8c40ea2a8ca80cf96d52a836bcaf76c2
Author: Matthias Clasen <mclasen redhat com>
Date:   Fri Feb 25 10:10:37 2011 -0500

   GOptionContext: Warn about invalid arg/flag combinations

   This was proposed by Kjell Ahlstedt in bug 642825.
   Also adding a few tests for this new behaviour.


_______________________________________________
gtk-devel-list mailing list
gtk-devel-list gnome org
http://mail.gnome.org/mailman/listinfo/gtk-devel-list