On Tue, 19 Dec 2017 09:52:50 +0100
Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx> wrote:
> On Thu, Dec 14, 2017 at 04:16:05PM +0100, Boris Brezillon wrote:
> > +/**
> > + * i3c_device_match_id() - Find the I3C device ID entry matching an I3C dev
> > + * @i3cdev: the I3C device we're searching a match for
> > + * @id_table: the I3C device ID table
> > + *
> > + * Return: a pointer to the first entry matching @i3cdev, or NULL if there's
> > + * no match.
> > + */
> > +const struct i3c_device_id *
> > +i3c_device_match_id(struct i3c_device *i3cdev,
> > + const struct i3c_device_id *id_table)
> > +{
> > + const struct i3c_device_id *id;
> > +
> > + /*
> > + * The lower 32bits of the provisional ID is just filled with a random
> > + * value, try to match using DCR info.
> > + */
> > + if (!I3C_PID_RND_LOWER_32BITS(i3cdev->info.pid)) {
> > + u16 manuf = I3C_PID_MANUF_ID(i3cdev->info.pid);
> > + u16 part = I3C_PID_PART_ID(i3cdev->info.pid);
> > + u16 ext_info = I3C_PID_EXTRA_INFO(i3cdev->info.pid);
> > +
> > + /* First try to match by manufacturer/part ID. */
> > + for (id = id_table; id->match_flags != 0; id++) {
> > + if ((id->match_flags & I3C_MATCH_MANUF_AND_PART) !=
> > + I3C_MATCH_MANUF_AND_PART)
> > + continue;
> > +
> > + if (manuf != id->manuf_id || part != id->part_id)
> > + continue;
> > +
> > + if ((id->match_flags & I3C_MATCH_EXTRA_INFO) &&
> > + ext_info != id->extra_info)
> > + continue;
> > +
> > + return id;
> > + }
> > + }
> > +
> > + /* Fallback to DCR match. */
> > + for (id = id_table; id->match_flags != 0; id++) {
> > + if ((id->match_flags & I3C_MATCH_DCR) &&
> > + id->dcr == i3cdev->info.dcr)
> > + return id;
> > + }
> > +
> > + return NULL;
> > +}
> > +EXPORT_SYMBOL_GPL(i3c_device_match_id);
>
> I just picked one random export here, but it feels like you are
> exporting a bunch of symbols you don't need to. Why would something
> outside of the i3c "core" need to call this function?
Because I'm not passing the i3c_device_id to the ->probe() method, and
if the driver is supporting different variants of the device, it may
want to know which one is being probed.
I considered retrieving this information in the core just before probing
the driver and passing it to the ->probe() function, but it means
having an extra i3c_device_match_id() call for everyone even those who
don't care about the device_id information, so I thought exporting this
function was a good alternative (device drivers can use it when they
actually need to retrieve the device_id).
Anyway, that's something I can change if you think passing the
i3c_device_id to the ->probe() method is preferable.
> Have you looked
> to see if you really have callers for everything you are exporting?
Yes, I tried to only export functions that I think will be needed by
I3C device drivers and I3C master drivers. Note that I didn't post the
dummy device driver I developed to test the framework (partly because
this is
>
> Other than that, the driver core interaction looks good now, nice job.
Thanks.
Boris
