suspend-to-disk o ThinkPad x60s, problem with USB

[Peter Whittaker]

On Wed, 2006-07-06 at 06:30 +0000, Stefan Wehr wrote:
> Peter Whittaker <pwwnow at gmail.com> wrote::
> 
> > JOOC, if you immediately suspend-to-disk (S4) and restart or
> > suspend-to-ram (S3) and resume, does USB come back?
> 
> Suspend-to-ram works fine, only suspend-to-disk is problematic.

OK - definitely different bugs, then. I was sort of hoping yours was the
same as or similar to mine, it would have meant more data points for the
developers.

> > On my Thinkpad A20m, a long S3 means USB is gone; I have to S3 and
> > resume immediately to get it (and networking) back. FYI, kernel logs,
> > etc., are in bug# 31896.
> 
> It doesn't matter whether I reboot immediately or wait for some time.
> Also, in my case USB does not work at all; your bug suggests that 
> it should work at least sometimes.

Yes, that's the "el bizarro" behaviour I am seeing: If I go through my
little resume-suspend-resume-disableNet-enableNet routine (which is
getting tedious, BTW), USB and all other functions return.

I've more testing to do, primarily to determine if I can reproduce these
"results", but it seems that over time, the system is slightly slower
after each r-s-r-d-e sequence... ...or at least it was until this
morning, when network-manager appeared to die during at "disable" - it
disappeared from the panel, and when it returned, networking was enabled
and system response was zippier than it's been for a few days. Nothing
in syslog to suggest that it really died, but the icon disappearance and
the re-enabling of networking without my action are suggestive.

FWIW, do you know what it seems like? It's as if suspend-and-resume
causes the system to "lose" "resources" or to "gain" "obstacles":
Switching from one app to another is very slow, and overall performance
is sluggish, as if the OS is having difficulty finding the pages, or as
if there is an accumulation of process "crud" getting in the way of IPC
or of memory manipulation.

At least it was until this morning's bizarre n-m hiccup - now things are
zippy again, windows get repainted quickly, etc.

Anyone have any suggestions for where to look for information useful to
the development teams? Like I wrote above, there is nothing new in
syslog, kern.log, daemon.log, to suggest that anything out of the
ordinary happened with n-m, but....

Thanks,

pww

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 191 bytes
Desc: This is a digitally signed message part
Url : https://lists.ubuntu.com/archives/laptop-testing-team/attachments/20060607/32440891/attachment.pgp