On Wed, Oct 06, 2010 at 11:40:45PM -0400, Ian Darwin wrote: > > > Also found a new bug: > > > wsmoused + X don't work (suspend/resume) > > > > > > 1) wsmoused > > > 2) startx > > > 3) suspend/resume > > > 4) in X mouse don't work > > > 5) suspend/resume > > > 6) in X mouse work > > Urm, I did say I'd test it later today, and I did so - just after > the diff got committed :-( So, I just updated pms.c to the committed > version and built a kernel. Fortunately, as reported in my earlier test, > I do not have the above problems on the Dell Studio 1558 laptop. > > However, in the process of this testing, I found two problems, > one related and one maybe related. Both have to do with running > two X sessions one after another. > > 1) (might or might not be related to pms) When I terminate X and > try to re-start it (either under XDM or by running sudo startx > twice), the second X does not always start - it tries, but it hangs > just after the basket weave background and the X cursor, before any > apps or the XDM login screen. Other times it works fine. Since > restarting X is not something I normally do (I'm usually the only > user of this machine, and I usually either suspend or power it off > when not using), I cannot say if this started when the pms/pmsi > merger went in. The last time I can be sure that I had several X's > in a row without rebooting is Sept 20, according to last(1). So I > can try triaging (actually we're bisecting, not triaging, when we > do this :-) ) on the weekend when I am home and have a cvsync repo > to build multiple kernels against, and time to do so. > > 2) (probably related to pms) With wsmoused running, the second X session > (as above) fails to get /dev/wsmouse. Sadly I captured the wrong > Xorg.log on this one so I'll have to replicate, but I'm out of time > for tonight. >
this error does not pms driver, but failure mechanism is similar to one that I found. At first I will finish with pms, and then take over these problems. I think the objection would not be? :) > This machine is Intel core i7, amd64 mode, and running -current > (userland + X updated earlier today to Oct 6 userland, Oct 3 X > snapshot). BIOS is up to date. Dmesg below; longer dmesg with > ACPI_DEBUG, and acpidump, on request. > > More later, sorry... > > Ian > -- Alexandr Shadchin
