sys.wait not in ms on mbed platform (patch included)

[GoogleCodeExporter]

sys.wait(1000) should wait for a second but it is much much longer.

in plat.cpp pm_vmPeriodic is being called with the wrong number.

I think this might also be an issue in at91sam7s-ek, judging from the comment, 
but I didn't check the data sheet to establish exactly what the timer is really 
set to.

Original issue reported on code.google.com by michae...@gmail.com on 7 Aug 2010 at 5:44

Attachments:

• mbed_plat.patch

[GoogleCodeExporter]

I agree the value is wrong.  This patch looks correct.  I will apply it when I 
can get the current sys.wait() to execute without errors on the mbed:

ipm> sys.wait(100)
Connection read error, type Ctrl+D to quit.

(VM no longer replies)

Original comment by dwhall...@gmail.com on 7 Aug 2010 at 10:44

[GoogleCodeExporter]

I think the current sys.wait causes timeouts on ipm as it is 100 times longer 
than it should be.

Original comment by michae...@gmail.com on 7 Aug 2010 at 11:27

[GoogleCodeExporter]

Michael, thanks for the report and the patch.

r574
- Manually applied changes based on the patch.
- Tested manually.  Times and delays work properly

Any sys.wait() over roughly 3000ms causes ipm-host to timeout and think the 
ipm-target is not responding.  Calling sys.wait() with values larger than 3000 
works as expected on the target in interactive and non-interactive modes, but 
it will break communications of ipm.

No vm changes were made, only platform.  
Tested manually on ipm.
Mainlined directly.

Original comment by dwhall...@gmail.com on 14 Aug 2010 at 6:48

• Changed state: Fixed
• Added labels: Component-platform, Milestone-09
• Removed labels: Component-all

[GoogleCodeExporter]

Original comment by dwhall...@gmail.com on 29 Sep 2010 at 7:24