Oops, yes the msleep() had been left in. I've attached debug output from the original ds1wm.c. Not sure how the mailing list copes with bzip2 compressed attachments; hopefully it will work.

Regards,
Paul

--- On Sat, 25/6/11, Jean-François Dagenais <dagenaisj@sonatest.com> wrote:
> This seems to be the output after you have restored the
> msleep. Could you provide it without so we can get a better
> idea as to what happened. If indeed this is a major
> regression for many devices, we'll have to submit a patch to
> the main branch.
> /jfd