On Fri, Oct 06, 2017 at 06:04:42PM +0800, Anand Jain wrote: > EIO is for IO (RW) failures, use correct error return > codes at two locations with in read_one_chunk(). Also > adds a small cleanup. This patch does 3 things, please split it, and drop the cleanup part. 1) switch add_missing_dev to ERR_PTR/IS_ERR return value and update callers 2) change return value from EIO to EINVAL in case a device is missing and mount is not degraded I'm not sure that EINVAL is the right error code, as it's used when the mount options are unrecognized or wrong. I agree that EIO is not the right code either, as it's reserved for "true" IO errors. Possibly it could be ENOENT or EAGAIN followed by the syslog message. -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
