Em Mon, 07 Sep 2009 11:13:36 -0700 Brian Rogers <brian@xxxxxxxx> escreveu: > Mauro Carvalho Chehab wrote: > > Em Sat, 5 Sep 2009 20:46:12 +0200 > > Jean Delvare <khali@xxxxxxxxxxxx> escreveu: > > > > > >> As far as I can see, the em28xx and saa7134 drivers have the exact same > >> problem. Is there anyone working on this? > >> > > > > I tested it here with an em28xx device and I got the trouble. I've committed a > > patch fixing it with this driver, with a different strategy, using dynamic > > memory. I did a similar patch for saa7134, although I can't test it ATM. Tests with > > saa7134 devices with i2c keyboards are welcome. > > > > Works fine with the MSI TV@nywhere Plus that prompted my version of the > patch. > > A couple comments. It looks like since the saa7134_dev is allocated with > kzalloc, you shouldn't have to zero out dev->info and dev->init_data. Yes, that's true. However, as it were allocated on another routine, and we're adding a callback pointer to be called for i2c IR handling, I opted to be a little paranoid here, since a later IR redesign could insert some code in-between. > Secondly, you misspelled my name in your commit messages. :) OK, I fixed your names at the commit messages on my -git. They'll be correct at the time I send them upstream. > Are you going to send this in for 2.6.31, or is it likely to wait until > 2.6.31.1? I filed a bug to track the regression: > http://bugzilla.kernel.org/show_bug.cgi?id=14142 They should be backported to be added at 2.6.31 -stable, since there were some IR struct changes that renamed the IR keycode tables. As the patches were applied after those changes, they'll require backporting. Cheers, Mauro -- To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
