On Sun, 08 Nov 2009, Jerone Young wrote: > I happen to be looking into the thinkpad-acpi code. The issue is that > currently the wwan card is completely gone from the USB bus after > resume. The only way for it to show up is to enable it via the > thinpad_acpi sys interface. Hmm... just in case, either enable debug or add some printks, and track what happens to wwan rfkill state before suspend and after suspend. It could be a rfkill screwup, as opposed to a bug in thinkpad-apci, or firmware weirdness. It is rfkill's job to reset the kill switch states after resume, through the callbacks. > The current workaround is to have a script write to the sys interface to > enable the card, though the card should actually be enabled after resume > as it is during an initial boot by thinkpad-acpi. thinkpad-acpi doesn't "enable" WWAN on initial boot. It leaves it in whatever state the BIOS set it to. rfkill, however, might enable it at boot. > Bluetooth does this. Why not WWAN? Both bluetooth and wwan especifically tell the firmware to NOT reenable the devices on resume, and want the rfkill core to do it itself. That kills any possibe disable-enable-disable races. If the rfkill core (or userspace!) is screwing up, there will be no WWAN on restore, and that's a possibility. Let's track down what userspace and rfkill are telling thinkpad-acpi to do first, and that might give us some answers... It could be a thinkpad-acpi bug, of course. Or a firmware bug. -- "One disk to rule them all, One disk to find them. One disk to bring them all and in the darkness grind them. In the Land of Redmond where the shadows lie." -- The Silicon Valley Tarot Henrique Holschuh ------------------------------------------------------------------------------ Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day trial. Simplify your report design, integration and deployment - and focus on what you do best, core application coding. Discover what's new with Crystal Reports now. http://p.sf.net/sfu/bobj-july _______________________________________________ ibm-acpi-devel mailing list ibm-acpi-devel@xxxxxxxxxxxxxxxxxxxxx https://lists.sourceforge.net/lists/listinfo/ibm-acpi-devel
