[Bug 678181] Compilation error when using GST_TIME_TO_TIMEVAL()...

[GStreamer (bugzilla.gnome.org)]

https://bugzilla.gnome.org/show_bug.cgi?id=678181
  GStreamer | gstreamer (core) | git

--- Comment #5 from Sebastian Rasmussen <sebras at hotmail.com> 2012-06-18 14:26:33 UTC ---
-> Wim: No, I believe that G_MAXUINT32 is the correct value. The reason being
that if the assert would fail on a 32-bit machine you want it to fail on a
64-bit machine too so that bugs on 32-bit machines will not go unnoticed for
people only compiling for 64-bit machines. G_MAXLONG would have different
values depending on if it is compiled for 32- or 64-bit machines in case I'm
haven't misunderstood things.

-> Tim: My employer uses it in a unit test for a proprietary (so far at least)
element. So yeah, I could absolutely see why this patch would only make
marginal sense to you. I could only see that this macro is used in the
implementation of gstpoll(), and there you might just as well use the struct
timeval directly.

So what I'm trying to say is that if you decide to keep G_TIME_TO_TIMEVAL()
around it ought to complain (on all architectures) when it received bogus
arguments, or to remove it completely which is also fine although it requires a
little bit of work for us (let me know if you want me to try to submit a
removal patch).

-- 
Configure bugmail: https://bugzilla.gnome.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the QA contact for the bug.
You are the assignee for the bug.