Discussion:
[Bug 796499] Changing the returned data from the buffer so that it can be modified and read without copying
"GStreamer" (GNOME Bugzilla)
2018-06-06 06:34:15 UTC
Permalink
https://bugzilla.gnome.org/show_bug.cgi?id=796499

Sebastian Dröge (slomo) <***@coaxion.net> changed:

What |Removed |Added
----------------------------------------------------------------------------
CC| |***@coaxion.net

--- Comment #4 from Sebastian Dröge (slomo) <***@coaxion.net> ---
Would it be possible to a) make sure that the buffer can't go out of scope (and
get destroyed) while there's a map, and b) to unmap the mapinfo if it was not
unmapped explicitly once it goes out of scope?
--
You are receiving this mail because:
You are the QA Contact for the bug.
You are the assignee for the bug.
"GStreamer" (GNOME Bugzilla)
2018-06-05 12:29:04 UTC
Permalink
https://bugzilla.gnome.org/show_bug.cgi?id=796499

Daniel Klamt <***@pengutronix.de> changed:

What |Removed |Added
----------------------------------------------------------------------------
CC| |***@pengutronix.de
--
You are receiving this mail because:
You are the QA Contact for the bug.
You are the assignee for the bug.
"GStreamer" (GNOME Bugzilla)
2018-06-05 12:23:53 UTC
Permalink
https://bugzilla.gnome.org/show_bug.cgi?id=796499

--- Comment #2 from Daniel Klamt <***@pengutronix.de> ---
Created attachment 372555
--> https://bugzilla.gnome.org/attachment.cgi?id=372555&action=edit
A gst-python element that uses the patch

gst-launch-1.0 videotestsrc ! videoconvert ! numpye ! videoconvert !
xvimagesink
--
You are receiving this mail because:
You are the QA Contact for the bug.
You are the assignee for the bug.
"GStreamer" (GNOME Bugzilla)
2018-06-05 12:21:30 UTC
Permalink
https://bugzilla.gnome.org/show_bug.cgi?id=796499

--- Comment #1 from Daniel Klamt <***@pengutronix.de> ---
Created attachment 372554
--> https://bugzilla.gnome.org/attachment.cgi?id=372554&action=edit
The Patch
--
You are receiving this mail because:
You are the QA Contact for the bug.
You are the assignee for the bug.
"GStreamer" (GNOME Bugzilla)
2018-06-05 12:45:21 UTC
Permalink
https://bugzilla.gnome.org/show_bug.cgi?id=796499

Tim-Philipp Müller <***@zen.co.uk> changed:

What |Removed |Added
----------------------------------------------------------------------------
Summary|Changing the returned data |Changing the returned data
|from the buffer so that it |from the buffer so that it
|can be modified an read |can be modified and read
|without copying |without copying
--
You are receiving this mail because:
You are the QA Contact for the bug.
You are the assignee for the bug.
"GStreamer" (GNOME Bugzilla)
2018-06-05 12:45:03 UTC
Permalink
https://bugzilla.gnome.org/show_bug.cgi?id=796499

Tim-Philipp Müller <***@zen.co.uk> changed:

What |Removed |Added
----------------------------------------------------------------------------
CC| |***@zen.co.uk
See Also| |https://bugzilla.gnome.org/
| |show_bug.cgi?id=678663
Severity|normal |enhancement

--- Comment #3 from Tim-Philipp Müller <***@zen.co.uk> ---
I'm guessing this is for version 1.14.1 not 1.4.1?
--
You are receiving this mail because:
You are the QA Contact for the bug.
You are the assignee for the bug.
"GStreamer" (GNOME Bugzilla)
2018-06-05 14:10:05 UTC
Permalink
https://bugzilla.gnome.org/show_bug.cgi?id=796499

Daniel Klamt <***@pengutronix.de> changed:

What |Removed |Added
----------------------------------------------------------------------------
Version|1.4.1 |1.14.1
--
You are receiving this mail because:
You are the QA Contact for the bug.
You are the assignee for the bug.
"GStreamer" (GNOME Bugzilla)
2018-07-11 13:35:01 UTC
Permalink
https://bugzilla.gnome.org/show_bug.cgi?id=796499

--- Comment #5 from Daniel Klamt <***@pengutronix.de> ---
As far as my understanding of python goes both a and b are handled
automatically by the python garbage collector.
--
You are receiving this mail because:
You are the QA Contact for the bug.
You are the assignee for the bug.
"GStreamer" (GNOME Bugzilla)
2018-11-02 13:26:13 UTC
Permalink
https://bugzilla.gnome.org/show_bug.cgi?id=796499

--- Comment #6 from Daniel Klamt <***@pengutronix.de> ---
The issue is that the Gst.MapInfo returned by Gst.Buffer.map would have a
bytes() object as .data member, not a memoryview().
And the original pointer would get lost when the GstMemoryInfo c structure was
recreated from the python object to be passed to gst_buffer_unmap.
https://paste.debian.net/1049985/

This would be a dirty fix
https://pastebin.com/NmUw1F3A
--
You are receiving this mail because:
You are the QA Contact for the bug.
You are the assignee for the bug.
"GStreamer" (GNOME Bugzilla)
2018-11-03 15:37:31 UTC
Permalink
https://bugzilla.gnome.org/show_bug.cgi?id=796499

GStreamer system administrator <***@gstreamer.net> changed:

What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |RESOLVED
Resolution|--- |OBSOLETE

--- Comment #7 from GStreamer system administrator <***@gstreamer.net> ---
-- GitLab Migration Automatic Message --

This bug has been migrated to freedesktop.org's GitLab instance and has been
closed from further activity.

You can subscribe and participate further through the new bug through this link
to our GitLab instance:
https://gitlab.freedesktop.org/gstreamer/gst-python/issues/13.
--
You are receiving this mail because:
You are the QA Contact for the bug.
You are the assignee for the bug.
Loading...