Bug 173048

Summary:	daap stresses cpu and takes very long to load
Product:	[Applications] amarok	Reporter:	Christian Weilbach <christian_weilbach>
Component:	Collections/DAAP	Assignee:	Amarok Developers <amarok-bugs-dist>
Status:	RESOLVED FIXED
Severity:	minor	CC:	hydrogen, ian.monroe, rdieter
Priority:	LO	Keywords:	needs_verification
Version:	unspecified
Target Milestone:	---
Platform:	Compiled Sources
OS:	Linux
Latest Commit:		Version Fixed In:
Attachments:	Callgrind snippet during load of daap database

Description Christian Weilbach 2008-10-17 17:58:16 UTC

Version:            (using Devel)
Compiler:          gcc 4.3.2 (Debian 4.3.2-1)
 
OS:                Linux
Installed from:    Compiled sources

Recent amarok 2 version (rev. 871961) takes very long to load daap resources without any progress feedback or information, but stressing one cpu completely. This is especially confusing for users who have a zeroconfed daap resource on the network since they might not know where the load comes from.
Amarok 2 takes much longer to load the resource than amarok 1.4 or rhythmbox.

Comment 1 Dan Meltzer 2008-12-02 18:10:27 UTC

The daap code has not seen much attention since 1.4... Someone will try to take a look at this after 2.0, however.  Can you provide any more information? (valgrind would probably be useful here)

Comment 2 Christian Weilbach 2009-01-19 13:26:29 UTC

I have chatted with eean and with mmh can't remember the nick. And they have fixed it, so it works at the moment if you edit the amarokrc by hand. But it takes much longer to load it than in 1.4 or in rhythmbox and it stresses my cpu. It might be extrem with me since it is 370gb of music data in the daap db. You will most likely notice the issue as well if you test it and try to fix it. Sorry I am back to 3.5 on lenny stable for now, so at the moment I cannot reproduce it. You can close the bug, if that is not enough input for you.

Comment 3 Christian Weilbach 2009-01-30 16:41:06 UTC

I am retesting 4.2 with amarok trunk here now and this is really bad since it autoloads the daap server since it is broadcasting itself. This means up to an hour of full cpu load here. I guess I disable daap in the code base now.

Comment 4 Christian Weilbach 2009-02-09 03:55:02 UTC

Created attachment 31138 [details]
Callgrind snippet during load of daap database

Here you are. Hope you can get the hungry bastard.

Comment 5 Myriam Schweingruber 2009-08-02 20:21:54 UTC

Sorry, Christian, this report slipped from our attention. I haven't seen any other reports about this since, so I guess this is solved with current 2.2-git. Feel free to reopen the bug if necessary.

Comment 6 Seb Ruiz 2009-08-29 07:22:48 UTC

This is now *actually* fixed for 2.2