#10828 LOW 1.75-so: Accelerometer driver for 1.75
Zarro Boogs per Child
bugtracker at laptop.org
Wed Jun 8 19:44:43 EDT 2011
#10828: Accelerometer driver for 1.75
------------------------------------+---------------------------------------
Reporter: cjb | Owner: saadia
Type: task | Status: new
Priority: low | Milestone: 1.75-software
Component: kernel | Version: not specified
Resolution: | Keywords:
Next_action: add to build | Verified: 0
Deployment_affected: | Blockedby:
Blocking: 10893, 10914 |
------------------------------------+---------------------------------------
Comment(by Quozl):
Sorry, I described my request incompletely.
By the time we release a software build, we'll be confident that the
driver will function with a properly manufactured laptop. No worries
there. I was asking about manufacturing tests.
In the current manufacturing process, the OpenFirmware runin tests involve
a human, but the Linux runin tests do not ... except to start and stop the
whole test.
The OpenFirmware runin tests will probably verify that the accelerometer
delivers data.
The purpose of the Linux runin tests is to detect hardware faults that
were not present during the OpenFirmware tests, either because
OpenFirmware could not test them, or because they only appear when driven
by a Linux kernel, or because of an intermittent condition, or because of
a deficiency that will only show a fault with elapsed time or thermal
cycles.
So can I design a test for some functions of the accelerometer and kernel
combination that does not depend on movement? For example, is there a way
to prove from user space that the accelerometer is still sending new data
to the kernel? Is there some expectation of jitter in the coordinates
that might be relied upon? Or some data arrival counter? Hope you can
see my interest. ;-)
--
Ticket URL: <http://dev.laptop.org/ticket/10828#comment:7>
One Laptop Per Child <http://laptop.org/>
OLPC bug tracking system
More information about the Bugs
mailing list