On 26 September 2010 22:52, Jonathan Corbet <corbet@xxxxxxx> wrote: > Daniel, 0x2C *has* to be GPIO on the XO - that's the power/reset pair > for the camera sensor. That's a little non-standard; I think normal > configurations use it for (i2c) communications to the monitor. So I > think I understand what you're seeing: if the port is i2c, the driver > tries (unsuccessfully) to talk over it, confusing the camera in the > process. If it's GPIO, those attempts fail immediately and it moves on. You may be right, but I've identified a more prominent reason why via-camera fails to initialize: Linux currently has port 0x2C as VIA_MODE_I2C in via-core.c. During initialization, via-gpio.c iterates over all ports and ignores the ones that are not VIA_MODE_GPIO. So 0x2c is ignored by via-gpio since it is marked as I2C. As you point out, the GPIOs we need are behind 0x2c, so they are never probed/enabled. Florian, the via-camera driver is in linux-next and the maintainer indicates it'll go to Linus within a day or 2. This is the last issue preventing the driver from working on XO-1.5. Do you have any thoughts on our next steps? Thanks, Daniel -- To unsubscribe from this list: send the line "unsubscribe linux-fbdev" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
