On Thu, Dec 23, 2010 at 01:30:15PM +0100, Thomas Pfaff wrote: > On Thu, 23 Dec 2010 00:36:26 +0000 > Jacob Meuser <[email protected]> wrote: > > > no feedback yet. anyone care to comment on this? > > > > > this diff covers rum(4), run(4), ural(4) and urtw(4). without the > > > diff, I can get the kernel to crash by starting a scan with > > > the deice, then ejecting it while the scan is running. with this > > > diff, I cannot get a crash. > > I cannot reproduce this problem with my rum(4) device. I've not tried > your diff to look for any regressions, but I'll try to do that later.
hmm. indeed, it is difficult to make rum(4) crash in this situation with -current. but not so hard if the rum(4) is attached to a cardbus usb adapter, and that is ejected (the patch addresses that situation too). I did just get crashes on the first try with run(4), ural(4) and utrw(4) though. this patch does the same thing in all these drivers. it shouldn't be hard to use the concepts in other drivers. that was the plan ... do a group of drivers to find the pattern. -- [email protected] SDF Public Access UNIX System - http://sdf.lonestar.org
