2009/8/5 Hans Petter Selasky <hselasky_at_c2i.net>: > On Wednesday 05 August 2009 16:13:31 Attilio Rao wrote: >> 2009/8/5 Hans Petter Selasky <hselasky_at_c2i.net>: >> > On Wednesday 05 August 2009 13:43:11 Robert Watson wrote: >> >> On Wed, 5 Aug 2009, Ed Schouten wrote: >> >> > * Lucius Windschuh <lwindschuh_at_googlemail.com> wrote: >> >> >> So I updated my machine to CURRENT r196062 and use a USB audio >> >> >> converter. Attaching it to the machine leads to a kernel panic: >> >> >> >> >> >> <snip> >> >> > >> >> > I suspect this has something to do with the Newbus locking, which >> >> > causes some pieces of code to run without Giant held, while they >> >> > previously did. >> >> >> >> There's a patch in the re_at_ queue to re-add Giant around newbus >> >> attachment, per John Baldwin's request. However, committing that patch >> >> is stalled while issues with the svn->cvs export of the new RELENG_8 >> >> branch are resolved. I expect to see the patch go into the tree RSN. >> > >> > Try this patch: >> > >> > http://perforce.freebsd.org/chv.cgi?CH=167030 >> >> Hans, >> I recall of a similar problem in ukbd. I resolved it by acquiring >> Giant earlier in ukbd_attach(), but probabilly I could just bring it >> down and pass a NULL pointer to usbd_do_transfer(), right? > > Yes, correct. > > I've changed the Giant mutex to a NULL one in USB P4. Maybe you can fix the > rest? > > http://perforce.freebsd.org/chv.cgi?CH=167032 Of course, thanks for the submissions. Attilio -- Peace can only be achieved by understanding - A. EinsteinReceived on Wed Aug 05 2009 - 12:20:10 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:39:53 UTC