[U-Boot] [PATCH v3] misc: fs_loader: Add support for initializing block device

[Chee, Tien Fong <tien.fong.chee@intel.com>]

On Thu, 2019-01-31 at 03:04 -0700, Simon Glass wrote:
> Hi,
> 
> On Thu, 24 Jan 2019 at 03:24, <tien.fong.chee@intel.com> wrote:
> > 
> > 
> > From: Tien Fong Chee <tien.fong.chee@intel.com>
> > 
> > Firmware loader would encounter problem if the block device is
> > accessed
> > before initializing it. This patch would adding the support of
> > probing
> > block device and initializing block before the block device is
> > accessed by
> > firmware loader.
> > 
> > Signed-off-by: Tien Fong Chee <tien.fong.chee@intel.com>
> > 
> > ---
> > 
> > Changes in v3:
> > - Initializing block device through probing the blk device
> > 
> > Changes in v2:
> > - Initializing MMC through probing the blk device
> > ---
> >  drivers/misc/fs_loader.c | 26 ++++++++++++++++++++++++++
> >  1 file changed, 26 insertions(+)
> Looks good, a few nits below.
> 
> Also at some point (separate patch) we really shoud have a test for
> UCLASS_FS_FIRMWARE_LOADER,.
Sure, i will do in later for separate patch. Need explore how to create
a test for firmware loader and also DTS as software policy
implementation.
> 
> > 
> > 
> > diff --git a/drivers/misc/fs_loader.c b/drivers/misc/fs_loader.c
> > index 57a14a3..df35ec6 100644
> > --- a/drivers/misc/fs_loader.c
> > +++ b/drivers/misc/fs_loader.c
> > @@ -12,6 +12,7 @@
> >  #include <linux/string.h>
> >  #include <mapmem.h>
> >  #include <malloc.h>
> > +#include <mmc.h>
> Can you drop this now?
sure.
> 
> > 
> >  #include <spl.h>
> > 
> >  DECLARE_GLOBAL_DATA_PTR;
> > @@ -252,6 +253,31 @@ static int fs_loader_ofdata_to_platdata(struct
> > udevice *dev)
> > 
> >  static int fs_loader_probe(struct udevice *dev)
> >  {
> > +#if CONFIG_IS_ENABLED(DM) && CONFIG_IS_ENABLED(BLK)
> > +       int ret;
> > +       struct device_platdata *plat = dev->platdata;
> > +
> > +       if (plat->phandlepart.phandle) {
> > +               ofnode node = ofnode_get_by_phandle(plat-
> > >phandlepart.phandle);
> > +
> drop extra blank line
sure.
> 
> > 
> > +               struct udevice *parent_dev = NULL;
> > +
> > +               ret = device_get_global_by_ofnode(node,
> > &parent_dev);
> > +
> here too
sure.
> 
> 
> > 
> > +               if (!ret) {
> > +                       struct udevice *dev;
> > +
> > +                       ret = blk_get_from_parent(parent_dev,
> > &dev);
> > +                       if (ret) {
> > +                               debug("fs_loader: No block device:
> > %d\n",
> > +                                       ret);
> > +
> > +                               return ret;
> > +                       }
> > +               }
> > +       }
> > +#endif
> > +
> >         return 0;
> >  };
> > 
> > --
> > 2.2.0
> >