driver core: Flow the return code from ->probe() through to sysfs bind
authorChristoph Hellwig <hch@lst.de>
Thu, 17 Jun 2021 14:22:11 +0000 (16:22 +0200)
committerAlex Williamson <alex.williamson@redhat.com>
Mon, 21 Jun 2021 21:29:24 +0000 (15:29 -0600)
commitef6dcbdd8eb2f44dce70a3abecc32d43cc5f3e64
treedcb0a8002852ae99ea21945002848c958d5e4c52
parente1499647c69c72c4583273e773d8c2786cb4bee9
driver core: Flow the return code from ->probe() through to sysfs bind

Currently really_probe() returns 1 on success and 0 if the probe() call
fails. This return code arrangement is designed to be useful for
__device_attach_driver() which is walking the device list and trying every
driver. 0 means to keep trying.

However, it is not useful for the other places that call through to
really_probe() that do actually want to see the probe() return code.

For instance bind_store() would be better to return the actual error code
from the driver's probe method, not discarding it and returning -ENODEV.

Reorganize things so that really_probe() returns the error code from
->probe as a (inverted) positive number, and 0 for successful attach.

With this, __device_attach_driver can ignore the (positive) probe errors,
return 1 to exit the loop for a successful binding and pass on the
other negative errors, while device_driver_attach simplify inverts the
positive errors and returns all errors to the sysfs code.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Link: https://lore.kernel.org/r/20210617142218.1877096-4-hch@lst.de
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
drivers/base/bus.c
drivers/base/dd.c