Date: Thu, 29 Sep 2011 13:27:36 +0200 From: Bengt Ahlgren <bengta@sics.se> To: Ian Smith <smithi@nimnet.asn.au> Cc: Matthias Apitz <guru@unixarea.de>, freebsd-mobile@freebsd.org Subject: Re: Laptop recommendations? Message-ID: <uh7oby3fv13.fsf@P142.sics.se> In-Reply-To: <20110709002621.R84785@sola.nimnet.asn.au> (Ian Smith's message of "Sat, 9 Jul 2011 00:57:10 %2B1000 (EST)") References: <20110625182502.GF29050@albert.catwhisker.org> <20110625191708.GA1182@tiny> <20110626165210.P72504@sola.nimnet.asn.au> <uh7boxjjxzx.fsf@P142.sics.se> <20110709002621.R84785@sola.nimnet.asn.au>
next in thread | previous in thread | raw e-mail | index | archive | help
Ian Smith <smithi@nimnet.asn.au> writes: > On Mon, 27 Jun 2011, Bengt Ahlgren wrote: > > Ian Smith <smithi@nimnet.asn.au> writes: > [..] > > > How well does suspend/resume work on these? An absolute must for me. > > > > > > My Thinkpad T23s always resumed 100% reliably through 6.x and 7.x, still > > > does at 7.4-R, but since 8.0-R through to 8.2-R they completely freeze > > > for an even 60 seconds on resume, after which I often get a bunch of > > > 'time went backwards' errors - sometimes on every process running - so > > > I've never felt able to really trust it after resuming. Noone seems to > > > know about that issue; probably no developers have i386 laptops anymore. > > > > Me too for a Thinkpad X40 (running 8.2-R). I tracked down the stall on > > resume to uhci.ko. Without that module loaded, there was no stall. > > Other usb modules did not make any difference. I unfortunately did not > > have time to investigate it further. My laptop comes to life reliably > > after the stall. > > I'm VERY happy to report that building a kernel from 8.2-R GENERIC only > excluding device uhci (and ohci and ehci that the T23 don't use anyway), > loading uhci at boot, unloading it in rc.suspend and loading it again in > rc.resume does completly fix the 60 second stall problem on my T23. > > It doesn't mind if I forget and leave a USB memstick attached through > suspend/resume, though of course it gets confused if a slice was mounted > .. even then, it doesn't panic or require more than umount; mount. > > Thanks heaps for this. I'll report back on an old thread in -stable, > find and cc a couple of other people who confirmed the same stall issue, > at least one being on a Thinkpad, T43 I think. > > It should help having narrowed this problem down to uhci, maybe a proper > fix can be found before 9.0 is released, but meanwhile I'm just stoked. See: http://lists.freebsd.org/pipermail/freebsd-usb/2011-June/010341.html for a patch that solves the uhci stall for me! Also discussed here: http://lists.freebsd.org/pipermail/freebsd-stable/2011-September/064002.html Bengt
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?uh7oby3fv13.fsf>