Anyone able to reproduce this? I can't imagine why we would just hang ...
- Bert -
Begin forwarded message:
From: "Zarro Boogs per Child" bugtracker@laptop.org Date: November 11, 2006 3:17:24 GMT+01:00 Cc: bugs@laptop.org Subject: [Trac #402] TamTam blocks EToys from starting Reply-To: devel@laptop.org
#402: TamTam blocks EToys from starting
+------------------------------------------------------- Reporter: jg | Owner: sean Type: defect | Status: new Priority: high | Milestone: BTest-1 Component: tamtam | Keywords: relnote
+------------------------------------------------------- If you start TamTam up, and then start etoys, etoys starts up and you get a black screen; ethttp://dev.laptop.org/queryoys can't run, presumably because TamTam is not releasing the audio device when it is not visible.
If you kill etoys, and then tamtam, you can start up etoys, and audio works for it.
-- Ticket URL: http://dev.laptop.org/ticket/402 One Laptop Per Child http://laptop.org/
Bert,
I tried this on the A-board with build 168 image. eToys starts up fine. However, as long as either TamTam or the game running in background, eToys cannot play sound. (But otherwise it is fine.)
Anyone able to reproduce this? I can't imagine why we would just hang ...
One possible reason is that he has some more data or apps that eat memory so that eToys couldn't allocate enough memory. (I had that problem.)
-- Yoshiki
Is OLPC OS doing fixed allocations/partitions? I guess they must be. How much elbow room do we really need here?
Cheers,
Alan
--------------
At 09:29 AM 11/11/2006, Yoshiki Ohshima wrote:
Bert,
I tried this on the A-board with build 168 image. eToys starts up fine. However, as long as either TamTam or the game running in background, eToys cannot play sound. (But otherwise it is fine.)
Anyone able to reproduce this? I can't imagine why we would just hang ...
One possible reason is that he has some more data or apps that eat memory so that eToys couldn't allocate enough memory. (I had that problem.)
-- Yoshiki _______________________________________________ Etoys mailing list Etoys@laptop.org http://mailman.laptop.org/mailman/listinfo/etoys
Is OLPC OS doing fixed allocations/partitions? I guess they must be. How much elbow room do we really need here?
Currently, they have 30MB or such amount of "cached swap" and, eToys can grow with that area. The object memory size can be 60MB or such. (When I nil-out some big buffers, the VM statistics shows a smaller number, but "top" or the kernel doesn't see it somehow...)
We should probably limit ourselves and make sure the VM returnn free memory.
-- Yoshiki
Thanks Yoshiki!
Cheers,
Alan
------------
At 10:34 AM 11/11/2006, Yoshiki Ohshima wrote:
Is OLPC OS doing fixed allocations/partitions? I guess they must be. How much elbow room do we really need here?
Currently, they have 30MB or such amount of "cached swap" and, eToys can grow with that area. The object memory size can be 60MB or such. (When I nil-out some big buffers, the VM statistics shows a smaller number, but "top" or the kernel doesn't see it somehow...)
We should probably limit ourselves and make sure the VM returnn free memory.
-- Yoshiki _______________________________________________ Etoys mailing list Etoys@laptop.org http://mailman.laptop.org/mailman/listinfo/etoys
Hi, Brad,
One possible reason is that he has some more data or apps that eat memory so that eToys couldn't allocate enough memory. (I had that problem.)
I've never written for ALSA, only for PortAudio that interfaces with ALSA (among other drivers). I believe once the device is open, others cannot access the device. Which is why there might be a blank screen when opening eToys while TamTam is running. Maybe it's trying to open the device and doesn't fail gracefully.
A few days ago, this (or this kind of) conflict was reported between eToys and other other apps. We set #soundStopWhenDone preference true after that, and it seems to be ok now.
The bug report might have been from the older disk image.
Thank you!
-- Yoshiki
etoys-dev@lists.squeakfoundation.org