Auckland Testing Summary 2 March 2013
Tom Parker
tom at carrott.org
Sat Mar 2 03:23:08 EST 2013
Auckland Testing Summary 2 March 2013
Who: John, Hans, Oliver, Tabitha, Tom
Testing 3.1.0 build 30 on XO-1.75 and build 32 on XO-4
In Physics on XO-1.75 the top couple of cm of the screen is white. The
exact line where this happens moves around. This has been reported
before. This doesn’t happen in the same way on the XO-4, with a
complicated enough display, the whole screen flickers every few seconds.
The more complicated the display the worse it gets, to the point where
it becomes unusable (3/4 of the screen is permanently white with only a
few objects on the screen). Also the mouse pointer is very badly
corrupted. This looks like http://dev.laptop.org/ticket/11922
Browse hung immediately on logging in to a Zimbra webmail. The mouse
pointer stopped responding to the page (it is supposed to change shape
over links and so forth), and the display stopped updating. It does not
respond to stop from the frame menu and it does not consume cpu. This
happened 3 times in a row. We reported this as
http://bugs.sugarlabs.org/ticket/3775 a long time ago. This time we
reproduced on XO-4. Unfortunately I am unable to find a working free
zimbra instance to produce a test case with.
Quickly tapping the XO figure in the middle of the activity ring and
then on the white space next to it and then back on the XO figure
eventually caused the touch screen to stop working and also the mouse
pointer to never re-appear. The keyboard still works and it seems
everything else still works, but mouse/touchscreen interaction was
impossible. There was nothing interesting in the logs. I am also able to
reproduce http://bugs.sugarlabs.org/ticket/2184 without breaking
everything using this method.
We had physics exhaust all the memory in the system on XO-1.75. The
system became laggy and started reading a lot from the disk and
eventually the OOM killer killed physics. Creating a scene with an
unclosed freehand object spinning with a motor seems to trigger this
memory leak, it seems to leak more than 1MB or so per minute with this
scene. Physics seems to pause when it is not the visible activity and
the memory leaking stops too. This happened on both XO-1.75 and XO-4.
Raised http://bugs.sugarlabs.org/ticket/4455
Write works well, but inserting an image is strange. On one laptop the
insert image window appeared behind write, giving the impression that
the clicking on the button does nothing. We opened log and during that
process the missing insert image screens appeared. This only happened
once but during this event we opened many instances of the insert image
window with them all appearing behind Write. It took quite some time to
close them all.
Distance doesn’t work, this activity should be removed if it cannot be
fixed. This was previously reported more than a year ago in
http://dev.laptop.org/ticket/11596
Scratch works, and loud? sensor works too! Yay. Is there a way to access
the built in accelerometer and light sensor? The Sensor block appears to
only work with external hardware.
There are no references on the Science page in WikipediaEN (there are
lots on the real wikipedia). There is a heading for references but no
references under it. Interestingly there are references on the Money page.
Measure starts with too much gain on XO-1.75, but has good gain on XO-4.
The trigger works, but occurs at a multiple of the actual trigger value.
We reported this a long time ago, but I can’t find a ticket. Will raise
one next time if I don’t find it before then.
A two year old is reliably able to break things by pressing the rotate
button. He lost the moon (the rest of Moon’s display was still there)
and he caused issues for TamTamMini.
Read & Get Books works.
Calculate activity works.
Tab tested on XO-1.75 memorize, measure (basic), maze, implode, tamtam
mini, help (still no keybd shortcut image), ruler, clock, record, moon,
stopwatch, simplegraph (also know as chart - can we pick one name?),
portfolio, labyrinth.
paint greyscale button doesnt work on this laptop but does on another,
both laptops have log file incorrectly(?) named “oficina” not “paint”.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: power-logs.tar.bz2
Type: application/x-bzip
Size: 18631 bytes
Desc: not available
URL: <http://lists.laptop.org/pipermail/devel/attachments/20130302/f70fdf55/attachment.bin>
More information about the Devel
mailing list