On Mon 07 Mar 02:05 PST 2022, Andy Shevchenko wrote:
> On Sun, Mar 06, 2022 at 07:40:34PM -0800, Bjorn Andersson wrote:
> > In some cases multiple connections with the same connection id
> > needs to be resolved from a fwnode graph.
> >
> > One such example is when separate hardware is used for performing muxing
> > and/or orientation switching of the SuperSpeed and SBU lines in a USB
> > Type-C connector. In this case the connector needs to belong to a graph
> > with multiple matching remote endpoints, and the Type-C controller needs
> > to be able to resolve them both.
> >
> > Add a new API that allows this kind of lookup.
>
> Thanks for the update!
>
> First of all, I have noticed that subject misses the verb, something like Add
> or Introduce.
>
Will update accordingly.
> ...
>
> > +/**
> > + * fwnode_connection_find_matches - Find connections from a device node
> > + * @fwnode: Device node with the connection
> > + * @con_id: Identifier for the connection
> > + * @data: Data for the match function
> > + * @match: Function to check and convert the connection description
> > + * @matches: Array of pointers to fill with matches
>
> (Optional) array...
>
Ditto.
> > + * @matches_len: Length of @matches
> > + *
> > + * Find up to @matches_len connections with unique identifier @con_id between
> > + * @fwnode and other device nodes. @match will be used to convert the
> > + * connection description to data the caller is expecting to be returned
> > + * through the @matches array.
>
> > + * If @matches is NULL @matches_len is ignored and the total number of resolved
> > + * matches is returned.
>
> I would require matches_len to be 0, see below.
>
> > + * Return: Number of matches resolved, or negative errno.
> > + */
> > +int fwnode_connection_find_matches(struct fwnode_handle *fwnode,
> > + const char *con_id, void *data,
> > + devcon_match_fn_t match,
> > + void **matches, unsigned int matches_len)
> > +{
> > + unsigned int count_graph;
> > + unsigned int count_ref;
> > +
> > + if (!fwnode || !match)
> > + return -EINVAL;
> > +
> > + count_graph = fwnode_graph_devcon_matches(fwnode, con_id, data, match,
> > + matches, matches_len);
>
> > + if (matches) {
> > + matches += count_graph;
> > + matches_len -= count_graph;
> > + }
>
> So, the valid case is matches != NULL and matches_len == 0. For example, when
> we have run something previously on the buffer and it becomes full.
>
> In this case we have carefully handle this case.
>
> if (matches) {
> matches += count_graph;
> if (matches_len)
> matches_len -= count_graph;
When matches is non-NULL, both the sub-functions are limited by
matches_len and as such count_graph <= matches_len.
As such matches_len >= 0.
In the event that the originally passed matches_len was 0, then
count_graph will be 0 and matches_len will remain 0.
I therefor don't see that this additional check changes things.
> }
>
> Seems it can be also
>
> if (matches)
> matches += count_graph;
>
> if (matches_len)
> matches_len -= count_graph;
We covered the case of matches && (matches_len || !matches_len) above.
For the case of !matches && matches_len, this added conditional would
cause matches_len to be extra ignored by keeping it at 0, but per
kernel-doc and implementation we ignore all other values already.
Note that this is in contrast from vsnprintf() where the code will
continue to produce results, only store the first "matches_len"
entires and return the final count.
Unfortunately we can't follow such semantics here, instead it is clearly
documented in the kernel-doc that @matches_len is ignored when @matches
is NULL.
So unless I'm missing something I don't see what you gain over keeping
the check on only matches.
>
> That said, do we have a test cases for this?
>
I looked briefly at adding some kunit tests for this, but was discourage
by the prospect of building up the graphs to run the tests against.
Regards,
Bjorn
> > + count_ref = fwnode_devcon_matches(fwnode, con_id, data, match,
> > + matches, matches_len);
> > +
> > + return count_graph + count_ref;
> > +}
>
> --
> With Best Regards,
> Andy Shevchenko
>
>
