On Wed, 2012-07-25 at 18:42 -0800, firstname.lastname@example.org wrote: > From: Ben Hutchings <email@example.com> > Date: Sat, 14 Jul 2012 05:43:54 +0100 > > Does the IEEE 1394 port work properly after this or not? > > It does not. The offending 1394 appears to be on an > Adaptec PCI card. If the camera is connected there, > Coriander always notifies that no camera is found. > > Luckily there is another 1394 socket on the front panel. > With the camera connected there, Coriander immediately > gives an image. So, this complaint comes from one > particular chip. > > > If not, can you test Linux 3.2 from testing, unstable > > or squeeze-backports? > > I have a vow never to monkey with unstable again but > the squeeze backport kernel might be tolerable. > > > The time limit for this has been increased in later versions of > > the driver, and it may be that this would fix the problem. > > Rather than an adjustment to allow the timeout to pass, > wouldn't a genuine repair of the driver be desirable? I didn't say the timeout was ignored; I said the time limit was increased. If a driver sets a time limit for some hardware operation which is shorter than the hardware may require to complete it successfully, increasing the time limit may well be the correct fix. Ben. -- Ben Hutchings Humans are not rational beings; they are rationalising beings.
Description: This is a digitally signed message part