On Mon, 2009-11-23 at 09:54 +0100, Jean Delvare wrote: > On Sun, 22 Nov 2009 19:17:59 -0500, Andy Walls wrote: > > On Sun, 2009-11-22 at 21:32 +0100, Jean Delvare wrote: > > > The fact that 0x30-0x37 and 0x50-0x5f all reply suggest that the bus > > > driver erroneously returns success to "SMBus receive byte" transactions > > > even when no device acks. This is a bug which should get fixed. If you > > > point me to the I2C adapter driver code, I can take a look. > > > > Although Igor's information makes the original need for this moot, here > > is the i2c adapter driver code: > > > > http://linuxtv.org/hg/v4l-dvb/file/8bff7e6c44d4/linux/drivers/media/video/cx23885/cx23885-i2c.c > > The results are not surprising: i2c_slave_did_ack() is only called for > zero-length transactions. For all other transactions, no check is done. > This is incorrect. > > I have written 3 patches for cx23885-i2c.c, the second one should fix > this particular issue. The other two are cleanups. Patches are there if > you want to take a look / give them a try: > http://khali.linux-fr.org/devel/misc/cx23885/ > > These are totally untested, and I don't know anything about the > hardware, so they might need some more work. But at least you should > get the idea of what's missing. Jean, (adding Steven Toth to the Cc: list) Thanks! I will inspect and test these with my HVR-1850 (CX23888) loaner board this weekend (hopefully). Regards, Andy -- 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
