Auckland Testing Summary 23 March 2013

classic Classic list List threaded Threaded
8 messages Options
Reply | Threaded
Open this post in threaded view
|

Auckland Testing Summary 23 March 2013

carrott
Auckland Testing Summary 23 March 2013
Who: Fabiana, John, Hans, Oliver, Tabitha, Tom

Testing 13.2.0 Build 1 on XO-1.75 and XO-4

No sound in TamTam or Speak on 3 out 5 XO-1.75s. Speak hung when saying
a sentence, scratch failed to start with an error in it’s log about
missing sound driver. Record hangs when you take a photo on those
laptops that have broken sound. We have discovered that this is related
to firmware Q4D27, as the two laptops that worked were still on Q4D24.
Updating them caused the sound to stop working.

On Karearea, the UI did not respond to mouse clicks or keyboard input,
but the neighbourhood/friends/activity list keys worked. After 3 reboots
this started working and we discovered the wifi doesn’t work, eth0
exists but is not configured. This laptop is one of the ones on which
sound does not work.

I have the /var/log/messages from a few laptops including Karearea's bad
wifi boot if anyone is interested.

Measure seems fine
Physics still has the top of screen problem on XO-1.75 but is working
very well on XO-4.

The context menu on the activity ring is annoying. It takes a long time
to appear and disappear.

WikipediaEN works, WikipediaES works

Browse still hangs on Zimbra. The PDF viewer embedded in browse doesn’t
have a jump to page option, and in touch mode, you can’t use the scroll
bar because you can’t reach it with your finger. If you open a pdf file,
save it to the journal, close browse and re-open then you can’t close
the pdf tab, raised http://bugs.sugarlabs.org/ticket/4471

Collaboration doesn’t work in write, does in chat. Access point, no
school server. Nothing interesting in the log.

Battery life on XO-4 C2 seems significantly improved over previous
builds. We would be interested in a confirmation or otherwise of this in
the power log analysis.

_______________________________________________
Testing mailing list
[hidden email]
http://lists.laptop.org/listinfo/testing

powerlogs.tar.bz2 (33K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Auckland Testing Summary 23 March 2013

Gary Martin-2
Hi Tom,

On 23 Mar 2013, at 08:43, Tom Parker <[hidden email]> wrote:

> Auckland Testing Summary 23 March 2013
> Who: Fabiana, John, Hans, Oliver, Tabitha, Tom
>
> Testing 13.2.0 Build 1 on XO-1.75 and XO-4
>
> No sound in TamTam or Speak on 3 out 5 XO-1.75s. Speak hung when saying a sentence, scratch failed to start with an error in it’s log about missing sound driver. Record hangs when you take a photo on those laptops that have broken sound. We have discovered that this is related to firmware Q4D27, as the two laptops that worked were still on Q4D24. Updating them caused the sound to stop working.
>
> On Karearea, the UI did not respond to mouse clicks or keyboard input, but the neighbourhood/friends/activity list keys worked. After 3 reboots this started working and we discovered the wifi doesn’t work, eth0 exists but is not configured. This laptop is one of the ones on which sound does not work.
>
> I have the /var/log/messages from a few laptops including Karearea's bad wifi boot if anyone is interested.
>
> Measure seems fine
> Physics still has the top of screen problem on XO-1.75 but is working very well on XO-4.

How critical do you think this one is? I have a patch from John that fixes this, it's related to hardware driver frame buffer support issues, but at the cost of some memory & performance cycles. Basically it removes pygame default screen buffering feature and does the buffering manually – no more white flicker at the top of the XO-1.75 redraw cycle, but not for free.

Regards,
--Gary

> The context menu on the activity ring is annoying. It takes a long time to appear and disappear.
>
> WikipediaEN works, WikipediaES works
>
> Browse still hangs on Zimbra. The PDF viewer embedded in browse doesn’t have a jump to page option, and in touch mode, you can’t use the scroll bar because you can’t reach it with your finger. If you open a pdf file, save it to the journal, close browse and re-open then you can’t close the pdf tab, raised http://bugs.sugarlabs.org/ticket/4471
>
> Collaboration doesn’t work in write, does in chat. Access point, no school server. Nothing interesting in the log.
>
> Battery life on XO-4 C2 seems significantly improved over previous builds. We would be interested in a confirmation or otherwise of this in the power log analysis.
> <powerlogs.tar.bz2>_______________________________________________
> Devel mailing list
> [hidden email]
> http://lists.laptop.org/listinfo/devel

_______________________________________________
Testing mailing list
[hidden email]
http://lists.laptop.org/listinfo/testing
Reply | Threaded
Open this post in threaded view
|

Re: Auckland Testing Summary 23 March 2013

Daniel Drake-5
In reply to this post by carrott
Thanks for testing.

On Sat, Mar 23, 2013 at 2:43 AM, Tom Parker <[hidden email]> wrote:
> No sound in TamTam or Speak on 3 out 5 XO-1.75s. Speak hung when saying a
> sentence, scratch failed to start with an error in it’s log about missing
> sound driver. Record hangs when you take a photo on those laptops that have
> broken sound. We have discovered that this is related to firmware Q4D27, as
> the two laptops that worked were still on Q4D24. Updating them caused the
> sound to stop working.

Thanks, I've filed http://dev.laptop.org/ticket/12637

> On Karearea, the UI did not respond to mouse clicks or keyboard input, but
> the neighbourhood/friends/activity list keys worked. After 3 reboots this
> started working and we discovered the wifi doesn’t work, eth0 exists but is
> not configured. This laptop is one of the ones on which sound does not work.
>
> I have the /var/log/messages from a few laptops including Karearea's bad
> wifi boot if anyone is interested.

I would be interested in that log.

> The context menu on the activity ring is annoying. It takes a long time to
> appear and disappear.

Has this changed since 13.1 (or some other release?).

> Battery life on XO-4 C2 seems significantly improved over previous builds.
> We would be interested in a confirmation or otherwise of this in the power
> log analysis.

I guess you are comparing to 13.1.0 build 36?
The big difference here is that 13.1.0 build 36 has automatic power
management disabled, it is now enabled in 13.2.0 (not 100% stable just
yet, but we continue to shake out the bugs).

Daniel
_______________________________________________
Testing mailing list
[hidden email]
http://lists.laptop.org/listinfo/testing
Reply | Threaded
Open this post in threaded view
|

Re: Auckland Testing Summary 23 March 2013

Richard A. Smith
>> Battery life on XO-4 C2 seems significantly improved over previous builds.
>> We would be interested in a confirmation or otherwise of this in the power
>> log analysis.
>
> I guess you are comparing to 13.1.0 build 36?
> The big difference here is that 13.1.0 build 36 has automatic power
> management disabled, it is now enabled in 13.2.0 (not 100% stable just
> yet, but we continue to shake out the bugs).

Its probably the firmware change rather than a build specific change.
We turned off unused cores and in q7b23 and that made almost 800mW
worth of difference across the board in operational power draw.  I'll
do a plot of your power logs though and confirm.

--
Richard A. Smith
One Laptop per Child
_______________________________________________
Testing mailing list
[hidden email]
http://lists.laptop.org/listinfo/testing
Reply | Threaded
Open this post in threaded view
|

Re: Auckland Testing Summary 23 March 2013

carrott
In reply to this post by Daniel Drake-5
On 26/03/13 05:00, Daniel Drake wrote:

> On Sat, Mar 23, 2013 at 2:43 AM, Tom Parker <[hidden email]> wrote:
>
>> On Karearea, the UI did not respond to mouse clicks or keyboard input, but
>> the neighbourhood/friends/activity list keys worked. After 3 reboots this
>> started working and we discovered the wifi doesn’t work, eth0 exists but is
>> not configured. This laptop is one of the ones on which sound does not work.
>>
>> I have the /var/log/messages from a few laptops including Karearea's bad
>> wifi boot if anyone is interested.
>
> I would be interested in that log.
Attached all the messages logs I collected. Karearea was the laptop with
the problems in addition to the sound. There is at least one logfile
from a boot on the old firmware with working sound so don't be confused
by that.

> I guess you are comparing to 13.1.0 build 36?
> The big difference here is that 13.1.0 build 36 has automatic power
> management disabled, it is now enabled in 13.2.0 (not 100% stable just
> yet, but we continue to shake out the bugs).

We didn't notice any problems with suspend-resume. I remember thinking
that the wifi took a long time to come back but I didn't investigate this.


_______________________________________________
Testing mailing list
[hidden email]
http://lists.laptop.org/listinfo/testing

1320-crash.txt.tar.bz2 (41K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Auckland Testing Summary 23 March 2013

carrott
In reply to this post by Gary Martin-2
On 24/03/13 17:56, Gary Martin wrote:
>> Physics still has the top of screen problem on XO-1.75 but is working very well on XO-4.
>
> How critical do you think this one is?

It's quite annoying if you happen to build something in physics that
extends to the top of the screen and certainly doesn't look good when
you show people the laptop.

> I have a patch from John that fixes this, it's related to hardware driver frame buffer support issues, but at the cost of some memory & performance cycles. Basically it removes pygame default screen buffering feature and does the buffering manually – no more white flicker at the top of the XO-1.75 redraw cycle, but not for free.

Is this as a workaround for a bug in pygame and/or the graphics drivers?
Can you isolate the change to XO-1.75 only? The other hardware seems to
work fine and doesn't need a performance hit.

Are you able to synchronize with the vertical blanking interrupt? I once
wrote a hack for an Amiga with an add-on graphics card that didn't drive
the vertical blank interrupt to busy wait until the vertical blank
started. It had a timer and kept the busy wait period small but not so
small that it would miss the start of the vblank.. This stopped icons
flickering when you dragged them.
_______________________________________________
Testing mailing list
[hidden email]
http://lists.laptop.org/listinfo/testing
Reply | Threaded
Open this post in threaded view
|

Re: Auckland Testing Summary 23 March 2013

Richard A. Smith
In reply to this post by Richard A. Smith
On 03/25/2013 02:03 PM, Richard Smith wrote:

>>> Battery life on XO-4 C2 seems significantly improved over previous builds.
>>> We would be interested in a confirmation or otherwise of this in the power
>>> log analysis.
>>
>> I guess you are comparing to 13.1.0 build 36?
>> The big difference here is that 13.1.0 build 36 has automatic power
>> management disabled, it is now enabled in 13.2.0 (not 100% stable just
>> yet, but we continue to shake out the bugs).
>
> Its probably the firmware change rather than a build specific change.
> We turned off unused cores and in q7b23 and that made almost 800mW
> worth of difference across the board in operational power draw.  I'll
> do a plot of your power logs though and confirm.

It appears there is considerable difference between your previous tests
and now.  Last round of logs I have from you prior to this one was March
2nd and prior to that was November 15, 2012 is that correct or did I
miss a batch between November and March?

Here's a ploy of XO-4 only 13.1.0-32 vs 13.2.0-1 all your other XO-4
13.1.0 builds I have were very early and not really interesting from a
power standpoint.

http://dev.laptop.org/~rsmith/auckland_13.1_vs_13.2.png

Your peak decreased which I believe is from your firmware upgrade but
does not explain the whole story.  A closer look at your log file shows
a much, much higher suspend frequency in 13.2.  In your 13.1 tests you
had perhaps 1 or 2 suspends whereas in the 13.2 log files almost every
other line is a suspend event.

You didn't change your testing procedure did you?  Something that would
allow for more idleness?

--
Richard A. Smith  <[hidden email]>
One Laptop per Child
_______________________________________________
Testing mailing list
[hidden email]
http://lists.laptop.org/listinfo/testing
Reply | Threaded
Open this post in threaded view
|

Re: Auckland Testing Summary 23 March 2013

Jerry Vonau
On Mon, 2013-03-25 at 22:19 -0400, Richard A. Smith wrote:

> On 03/25/2013 02:03 PM, Richard Smith wrote:
> >>> Battery life on XO-4 C2 seems significantly improved over previous builds.
> >>> We would be interested in a confirmation or otherwise of this in the power
> >>> log analysis.
> >>
> >> I guess you are comparing to 13.1.0 build 36?
> >> The big difference here is that 13.1.0 build 36 has automatic power
> >> management disabled, it is now enabled in 13.2.0 (not 100% stable just
> >> yet, but we continue to shake out the bugs).
> >
> > Its probably the firmware change rather than a build specific change.
> > We turned off unused cores and in q7b23 and that made almost 800mW
> > worth of difference across the board in operational power draw.  I'll
> > do a plot of your power logs though and confirm.
>
> It appears there is considerable difference between your previous tests
> and now.  Last round of logs I have from you prior to this one was March
> 2nd and prior to that was November 15, 2012 is that correct or did I
> miss a batch between November and March?
>
> Here's a ploy of XO-4 only 13.1.0-32 vs 13.2.0-1 all your other XO-4
> 13.1.0 builds I have were very early and not really interesting from a
> power standpoint.
>
> http://dev.laptop.org/~rsmith/auckland_13.1_vs_13.2.png
>
> Your peak decreased which I believe is from your firmware upgrade but
> does not explain the whole story.  A closer look at your log file shows
> a much, much higher suspend frequency in 13.2.  In your 13.1 tests you
> had perhaps 1 or 2 suspends whereas in the 13.2 log files almost every
> other line is a suspend event.
>
> You didn't change your testing procedure did you?  Something that would
> allow for more idleness?
>

I was seeing high cpu usage that was present preventing suspend with
earlier builds.

Jerry




_______________________________________________
Testing mailing list
[hidden email]
http://lists.laptop.org/listinfo/testing