Unfortunately nothing seems to get this feature working after the 2.6.21kernel. I have tried your latest 24-rc2 kernel tpowa and it acts the same as 2.6.23. I also patched 2.6.23 with tuxonice and that didn't make any different. I also tried the mm kernel to no avail.
What does work is mounting a usb drive, suspending while mounted, keeping the usb drive plugged in during suspend, and resuming. In this case the USB drive keeps working without the need to remount after a resume. More importantly what doesn't work is mounting a usb drive, suspending while mounted, detaching the usb drive during suspend, reattaching usb drive during suspend before resume, and resuming. In this case the mount is lost and the drive has to be remounted after resume. And the persist value in sysfs for the USB drive does not seem to have any effect what so ever. Whether persist is set to 1 or 0 everything acts as described above. So I think my next step should be posting my question to lkml. Any other suggestions are really appreciated. It's really disappointing that a kernel feature that worked flawlessly 2 kernel releases ago now does not work. Seems like there's some "reinventing of the wheel" going on in the kernel. Thanks for your help guys, Raymano On 11/12/07, Raymano Garibaldi <[EMAIL PROTECTED]> wrote: > > Thanks guys. I will give both your recommendations a shot. > > On 11/12/07, Petar Bogdanovic <[EMAIL PROTECTED] > wrote: > > > > On Mon, Nov 12, 2007 at 06:20:39PM +0100, Tobias Powalowski wrote: > > > Am Montag, 12. November 2007 schrieb Raymano Garibaldi: > > > > Hi, > > > > > > > > I'm having problems with USB_PERSIST and suspending USB devices in > > the > > > > latest kernel 2.6.23.1-6. > > > > > > > > Particularly let's say I have a mounted read-only USB drive. If I > > suspend > > > > the computer and while suspended remove and reattach the USB drive, > > and > > > > then resume the computer, the device mount is lost and it has to be > > > > remounted. I would like the device to stay mounted while the > > computer is > > > > suspended even if it's detached and reattached while the computer is > > > > suspended. This was the way it behaved under kernel 2.6.21. Kernel > > 2.6.22 > > > > didn't even have the USB_PERSIST option so I skipped it in FaunOS. > > And now, > > > > USB_PERSIST only works if you leave the device attached to the > > computer > > > > while the computer is suspended. > > > > > > > > Note that the key here is that the USB device is read-only and it > > can't be > > > > changed when it's detached from the suspended computer. > > > > > > > > I'm starting to think that I have to back port the USB kernel > > subsystem in > > > > 2.6.23 to 2.6.21 and I'm trying to avoid that if I can. > > > > > > > > Any help, hacks, patches, ideas are appreciated. > > > > > > > > Raymano > > > Hi > > > > > > you know that this option is marked experimental and that the kernel > > devs warn > > > about the usage and the possibilities it might not work. > > > probably that is the reason that it doesn't work. > > > > > > you could try an .24 rc2 kernel here if it's fixed in latest upstream: > > > http://www.archlinux.org/~tpowa/2.6.24/ > > <http://www.archlinux.org/%7Etpowa/2.6.24/> > > > > Thanks, you just saved me some time. :) > > > > _______________________________________________ > > arch mailing list > > arch@archlinux.org > > http://archlinux.org/mailman/listinfo/arch > > > >
_______________________________________________ arch mailing list arch@archlinux.org http://archlinux.org/mailman/listinfo/arch
