Hello.
+static int omap34xxcam_open(struct inode *inode, struct file *file)
+{
<snip>
+ if (atomic_inc_return(&vdev->users) == 1) {
+ isp_get();
+ if (omap34xxcam_slave_power_set(vdev, V4L2_POWER_ON,
+ OMAP34XXCAM_SLAVE_POWER_ALL))
+ goto out_slave_power_set_standby;
+ omap34xxcam_slave_power_set(
+ vdev, V4L2_POWER_STANDBY,
+ OMAP34XXCAM_SLAVE_POWER_SENSOR);
+ omap34xxcam_slave_power_suggest(
+ vdev, V4L2_POWER_STANDBY,
+ OMAP34XXCAM_SLAVE_POWER_LENS);
+ }
I'm wondering whether this V4L2_POWER_STANDBY operation for sensor
device is really necessary.
Because if that makes sensor device in standby mode, we do S_FMT and
bunch of V4L2 APIs while the camera module is in standby mode.
In most cases of "sensor + ISP" SOC camera modules, I2C command is not
working while the camera module is in standby mode.
Following the camera interface source code, sensor goes down to
standby mode until VIDIOC_STREAMON is called.
If this power up timing depends on sensor device, then I think we need
a conditional power on sequence.
As you defined slave devices as SENSOR, LENS, FLASH, then how about
making a new slave category like "ISP" for "sensor+ISP" SOC modules?
Then we could make slave devices with conditional cases.
Cheers.
Regards,
Nate
--
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
