On Fri, Apr 25, 2014 at 08:25:36PM +0200, Sebastian Reichel wrote:
> On Sat, Apr 19, 2014 at 09:16:12PM +0200, Pavel Machek wrote:
> > On Sat 2014-03-29 01:31:43, Sebastian Reichel wrote:
> > > Implement and document generic DT bindings for HSI clients.
> > >
> > > Signed-off-by: Sebastian Reichel <sre@xxxxxxxxxx>
> >
> > Reviewed-by: Pavel Machek <pavel@xxxxxx>
> >
> > > diff --git a/drivers/hsi/hsi.c b/drivers/hsi/hsi.c
> > > index 07e1639..5973906 100644
> > > --- a/drivers/hsi/hsi.c
> > > +++ b/drivers/hsi/hsi.c
> > > @@ -26,8 +26,14 @@
> > > #include <linux/slab.h>
> > > #include <linux/string.h>
> > > #include <linux/notifier.h>
> > > +#include <linux/of.h>
> > > +#include <linux/of_device.h>
> > > #include "hsi_core.h"
> >
> >
> > > + err = hsi_of_property_parse_mode(client, "hsi-mode", &mode);
> > > + if (err) {
> > > + err = hsi_of_property_parse_mode(client, "hsi-rx-mode",
> > > + &cl->rx_cfg.mode);
> > > + if (err)
> > > + goto err;
> > > +
> > > + err = hsi_of_property_parse_mode(client, "hsi-tx-mode",
> > > + &cl->tx_cfg.mode);
> > > + if (err)
> > > + goto err;
> >
> > Will this need some #ifdef CONFIG_OF?
>
> It would only be needed to reduce the amount of kernel code for
> disabled Device Tree. I don't think its worth it, since there is no
> platform in the mainline kernel, which uses HSI without DT.
mh actually it is needed, since there is no stub provided for
of_modalias_node in non DT mode (there are stubs for all other
of_* methods used). I will include the ifdef in PATCHv5.
-- Sebastian
Attachment:
signature.asc
Description: Digital signature
