After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 579177 - GstInfo: maintain ABI compatibility
GstInfo: maintain ABI compatibility
Status: RESOLVED FIXED
Product: GStreamer
Classification: Platform
Component: gstreamer (core)
git master
Other Linux
: Normal normal
: 0.10.24
Assigned To: René Stadler
GStreamer Maintainers
: 579202 (view as bug list)
Depends on:
Blocks:
 
 
Reported: 2009-04-16 16:01 UTC by Tim-Philipp Müller
Modified: 2009-06-12 15:24 UTC
See Also:
GNOME target: ---
GNOME version: ---


Attachments
compete stubs (4.00 KB, patch)
2009-04-16 19:28 UTC, Stefan Sauer (gstreamer, gtkdoc dev)
none Details | Review
Dirty(?) patch (2.98 KB, patch)
2009-04-20 23:34 UTC, René Stadler
committed Details | Review

Description Tim-Philipp Müller 2009-04-16 16:01:42 UTC
This commit

http://cgit.freedesktop.org/gstreamer/gstreamer/commit/?id=659102e04a7111895cde506e5e3c9e118b0a81f6

removes stub functions that have been added to maintain ABI compatibility even with the debugging system in core disabled.

The reason for the commit is that the no-op #defines in gstinfo.h also apply to gstinfo.c when compiling gstinfo.c, so the preprocessor will replace the function names with the no-ops, causing a build failure ...

We should fix this differently though.
Comment 1 Tim-Philipp Müller 2009-04-16 19:20:58 UTC
*** Bug 579202 has been marked as a duplicate of this bug. ***
Comment 2 Stefan Sauer (gstreamer, gtkdoc dev) 2009-04-16 19:28:36 UTC
Created attachment 132784 [details] [review]
compete stubs
Comment 3 Stefan Sauer (gstreamer, gtkdoc dev) 2009-04-17 12:09:53 UTC
Author: Stefan Kost <ensonic@users.sf.net>
Date:   Thu Apr 16 22:26:00 2009 +0300

    gstdebug: compete stubs. Fixes #579177.
    
    Avoid defines when including gstinfo.h ourself and complete stubs. Sync stub
    returns with the defines.
Comment 4 René Stadler 2009-04-20 23:27:36 UTC
It's still breaking the ABI. At least these symbols are missing:

__gst_debug_enabled
__gst_debug_min
_gst_debug_category_new
_gst_debug_register_funcptr
_gst_debug_nameof_funcptr
GST_CAT_AUTOPLUG
GST_CAT_AUTOPLUG_ATTEMPT
GST_CAT_BUFFER
GST_CAT_BUS
GST_CAT_CALL_TRACE
GST_CAT_CAPS
GST_CAT_CLOCK
GST_CAT_DEFAULT
GST_CAT_ELEMENT_PADS
GST_CAT_ERROR_SYSTEM
GST_CAT_EVENT
GST_CAT_GST_INIT
GST_CAT_MESSAGE
GST_CAT_NEGOTIATION
GST_CAT_PADS
GST_CAT_PARAMS
GST_CAT_PARENTAGE
GST_CAT_PIPELINE
GST_CAT_PLUGIN_INFO
GST_CAT_PLUGIN_LOADING
GST_CAT_PROBE
GST_CAT_PROPERTIES
GST_CAT_QOS
GST_CAT_REFCOUNTING
GST_CAT_REGISTRY
GST_CAT_SCHEDULING
GST_CAT_SIGNAL
GST_CAT_STATES
GST_CAT_TYPES
GST_CAT_XML
Comment 5 René Stadler 2009-04-20 23:34:14 UTC
Created attachment 132999 [details] [review]
Dirty(?) patch

I cooked this patch quickly on Friday evening. Seems to fix it. Maybe there is a more elegant way, but AFAICS gstinfo.h is turning into #ifdef hell if you want to do it without duplicating the global variable declarations :|
Comment 6 Jan Schmidt 2009-04-21 19:45:54 UTC
I don't want to put this patch in tonight for this pre-release, but lets put it or something like it in tomorrow or Thursday, please.
Comment 7 Jan Schmidt 2009-04-28 09:18:29 UTC
On second thought, the followup patch adds symbols that were already missing in previous releases. I think it'd be better to have them so that our ABI is the same regardless of disabling debug, but since it's not a regression maybe not for this release.
Comment 8 René Stadler 2009-04-29 07:48:08 UTC
I agree, but please exclude this bug from the release notes so nobody gets the impression that this is fixed.
Comment 9 Jan Schmidt 2009-04-29 09:27:06 UTC
The release script pulls in RESOLVED bugs with the target milestone set to the release number in question, so that should be fine.
Comment 10 Tobias Mueller 2009-06-05 12:59:39 UTC
Hey folks. It's been a while since the patch 132999 has been set accept-commit_after_freeze. I am a bit confused, as I don't know which freeze is to respect (anymore?). Could anybody confirm that this patch is still to commit?
Comment 11 René Stadler 2009-06-12 15:24:17 UTC
Done. Good to see I'm not the only one who cares about --disable-gst-debug :)

commit c403e4b7f0a66f9387893869754c90ca09de70c9
Author: René Stadler <rene.stadler@nokia.com>
Date:   Fri Apr 17 17:59:38 2009 +0300

    Fix remaining --disable-gst-debug ABI breakage.
    
    Fixes #579177.