On Tue, Feb 28, 2012 at 02:03:30AM +0100, Steinar H. Gunderson wrote: > I'm still not happy with the bit-banging on the I2C interface (as opposed to > dealing with it in the interrupt handler); I long suspected it for causing > the IRQ0 problems, especially as they seem to have a sort-of similar issue > with I2CDONE/I2CRACk never being set, but it seem the DMA transfers is really > what causes it somehow, so I've left it alone. There's one more thing I forgot to mention; deinitialization does not seem to work correctly. If I remove the mantis driver and reload it, the CAM is no longer detected correctly (when the 50211 driver tries to read the identification bits from the CAM memory space, they simply don't match anymore). This is annoying, as it means a reboot is required whenever one wants to test a new fix, but I haven't been able to figure out why it happens. Just wanted to mention so that nobody bangs their head in the wall too much due to hitting that problem :-) /* Steinar */ -- Homepage: http://www.sesse.net/ -- 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
