On 5/13/23 21:44, Jonathan Cameron wrote:
On Fri, 12 May 2023 10:53:41 +0300
Matti Vaittinen <mazziesaccount@xxxxxxxxx> wrote:
The fwnode_irq_get_byname() was returning 0 at device-tree mapping
error. If this occurred, the KX022A driver did abort the probe but
errorneously directly returned the return value from
fwnode_irq_get_byname() from probe. In case of a device-tree mapping
error this indicated success.
The fwnode_irq_get_byname() has since been fixed to not return zero on
error so the check for fwnode_irq_get_byname() can be relaxed to only
treat negative values as errors. This will also do decent fix even when
backported to branches where fwnode_irq_get_byname() can still return
zero on error because KX022A probe should later fail at IRQ requesting
and a prober error handling should follow.
On that basis I've picked this one up directly for the fixes-togreg branch of
iio.git and marked it for stable.
Thanks for picking this up Jonathan. Although, the commit message is
slightly misleading w/o the previous patches in this series because the
fwnode_irq_get_byname() is fixed in the first patch.
Yours,
-- Matti
--
Matti Vaittinen
Linux kernel developer at ROHM Semiconductors
Oulu Finland
~~ When things go utterly wrong vim users can always type :help! ~~
