On Fri, May 17, 2013 at 08:38:03AM +0200, Jean-Francois Moine wrote: > The signals SIGALRM (most often) and SIGPOLL (sometimes) may be raised > after irq or some parameter change and this aborts the i2c exchanges. > > The problem is solved by masking these signals. NAK. The real problem is in the I2C driver, which actually suffers from multiple problems. See the set of I2C patches I posted yesterday. _______________________________________________ dri-devel mailing list dri-devel@xxxxxxxxxxxxxxxxxxxxx http://lists.freedesktop.org/mailman/listinfo/dri-devel
