On Mon, Oct 04, 2010 at 10:44:18AM +0300, Andriy Gapon wrote: > ... > Interesting!
:-} > So, init thread is blocked in vfs_mountroot()->g_waitidle() waiting for geom > to > complete probing cd device, but the latter is blocked in cam_periph_getccb() > waiting for a ccb. > Can you try to reproduce this again and examine other thread for anything > interesting in geom/cam department? Thread named as "xpt_thrd" might be of > particular interest. Yes; I can do that. Bearing in mind that each of us has a finite number of keystrokes in his (or her) life, typing all that stuff by hand, though, tends to eat into that upper limit a bit faster than most other forms of recreation do. :-} Is there a way I can use USB or firewire in a way similar to a serial console? The laptop has both of those (though I only have connectors for USB at the moment), and does not have a DB9 (let alone a DB25) serial port. I'd welcome a pointer. Anyway: I have a couple of slices where I can re-create it: * Slice 3 uses "standard ATA", and runs head as of r213322. * Slice 4 uses ATA_CAM and is presently running head as of r213400. Symptoms are fairly similar in each, though ?I seem to see the hang less frequently when I use ATA_CAM. In either case, PID 8 is [xpt_thrd], and wmesg is "ccb_scan". I'll hold off on transcribing the backtrace for PID 8, in hope of being able to let the machine do the clerical work. :-} > We urgently need geom and cam experts here :-) Given sufficient pointers on actions to take, patches to apply, or code to hack, I'm willing. Peace, david -- David H. Wolfskill [email protected] Depriving a girl or boy of an opportunity for education is evil. See http://www.catwhisker.org/~david/publickey.gpg for my public key.
pgpGFp1l7LCzX.pgp
Description: PGP signature
