[PATCH] dbus-glib patches
david at fubar.dk
Mon Mar 17 13:14:46 PDT 2008
On Mon, 2008-03-17 at 15:39 -0400, David Zeuthen wrote:
> > Agreed, I don't understand why the timeout is there.
> Havoc? Can we change this?
I talked to Havoc on IRC and he found some useful links to earlier
So the conclusion is that this is something we need to fix and we won't
break ABI doing so. The proposed solution
- make INT_MAX special to avoid 49.52 days of timeout
- remove bus timeout completely unless set in the config file
and for bonus
- audit the code to ensure we use a monotonically increasing clock
I'll try to write a patch for the first two items this week.
(Also, my cursory testing over the weekend shows that the five minute
bus timeout isn't enforced. That's probably another bug.)
More information about the dbus