Re: [microblaze-uclinux] Sleep time issue

[Giulio Mazzoleni <giulio.mazzoleni@xxxxxxxxxxxxxxxxxxxxxxx>]

This is the output of a simple debug application I wrote to test the
sleeping issue.

I'm using a custom board with a gps disciplined OCXO, so the timebase
should be ok.

When I run my application I get:

# ./sleeptest 
--> cpu-gsp diff: -15.954 s
gps time: 1231951464.586947000 - Wed Jan 14 16:44:24 2009
cpu time: 1231951450.270515000 - Wed Jan 14 16:44:10 2009

-- usleep --
  asked to sleep    0 ms --> cpu:    2 ms --> gps:  273 ms
  asked to sleep  250 ms --> cpu:  255 ms --> gps:  526 ms
  asked to sleep  500 ms --> cpu:  505 ms --> gps:  776 ms
  asked to sleep  750 ms --> cpu:  755 ms --> gps: 1026 ms
  asked to sleep 1000 ms --> cpu: 1005 ms --> gps: 1276 ms

-- sleep --
  asked to sleep 0 s --> cpu: 0.000 s --> gps: 0.270 s
  asked to sleep 1 s --> cpu: 1.006 s --> gps: 1.277 s
  asked to sleep 2 s --> cpu: 2.004 s --> gps: 2.276 s
  asked to sleep 3 s --> cpu: 3.003 s --> gps: 3.274 s
  asked to sleep 4 s --> cpu: 4.005 s --> gps: 4.547 s
  asked to sleep 5 s --> cpu: 5.004 s --> gps: 5.275 s

--> cpu-gsp diff: -20.844 s
gps time: 1231951489.497964900 - Wed Jan 14 16:44:49 2009
cpu time: 1231951470.071512000 - Wed Jan 14 16:44:30 2009

Keep investigating..
Best regards.
Giulio

___________________________
microblaze-uclinux mailing list
microblaze-uclinux@xxxxxxxxxxxxxx
Project Home Page : http://www.itee.uq.edu.au/~jwilliams/mblaze-uclinux
Mailing List Archive : http://www.itee.uq.edu.au/~listarch/microblaze-uclinux/