Re: [PATCH -next] drm/komeda: Convert to DEFINE_SHOW_ATTRIBUTE

[james qian wang (Arm Technology China)]

On Fri, Jul 17, 2020 at 09:06:57AM +0200, Daniel Vetter wrote:
> On Fri, Jul 17, 2020 at 8:40 AM james qian wang (Arm Technology China)
> <james.qian.wang@xxxxxxx> wrote:
> >
> > On Thu, Jul 16, 2020 at 05:03:33PM +0800, Qinglang Miao wrote:
> > > From: Liu Shixin <liushixin2@xxxxxxxxxx>
> > >
> > > Use DEFINE_SHOW_ATTRIBUTE macro to simplify the code.
> > >
> > > Signed-off-by: Liu Shixin <liushixin2@xxxxxxxxxx>
> > > ---
> > >  drivers/gpu/drm/arm/display/komeda/komeda_dev.c | 13 +------------
> > >  1 file changed, 1 insertion(+), 12 deletions(-)
> > >
> > > diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_dev.c b/drivers/gpu/drm/arm/display/komeda/komeda_dev.c
> > > index 0246b2e94..4a10e6b9e 100644
> > > --- a/drivers/gpu/drm/arm/display/komeda/komeda_dev.c
> > > +++ b/drivers/gpu/drm/arm/display/komeda/komeda_dev.c
> > > @@ -41,18 +41,7 @@ static int komeda_register_show(struct seq_file *sf, void *x)
> > >       return 0;
> > >  }
> > >
> > > -static int komeda_register_open(struct inode *inode, struct file *filp)
> > > -{
> > > -     return single_open(filp, komeda_register_show, inode->i_private);
> > > -}
> > > -
> > > -static const struct file_operations komeda_register_fops = {
> > > -     .owner          = THIS_MODULE,
> > > -     .open           = komeda_register_open,
> > > -     .read_iter              = seq_read_iter,

-	.read		= seq_read,
+	.read_iter		= seq_read_iter,

> > > -     .llseek         = seq_lseek,
> > > -     .release        = single_release,
> > > -};
> > > +DEFINE_SHOW_ATTRIBUTE(komeda_register);
> > >
> >
> > Hi Shixin & Qinglang
> >
> > Thanks for your patch.
> >
> > Reviewed-by: James Qian Wang <james.qian.wang@xxxxxxx>
> >
> > Since your patch is not for drm-misc-next, so seems better
> > to leave it to you to merge it. :)
> 
> I do think it's for drm-misc-next, what other tree would it be for?
> Some people put -next in their patch tag to differentiate from -fixes,
> so maintainers know what to do with the patch. It's also not part of a
> series, hence I think this is on you to apply it.
> 
> Cheers, Daniel

Hi Daniel:

I tried to apply this patch to drm-misc-next, but failed, and found
this patch is actually based on linux-next, and the code base of
linux-next is a little different with our drm-misc-next.
and one of the difference is linux-next has a patch (call it patch-A):

seq_file: switch over direct seq_read method calls to seq_read_iter
https://lkml.org/lkml/2020/7/7/1267

which changed code like below:

diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_dev.c b/drivers/gpu/drm/arm/display/komeda/komeda_dev.c
index 1d767473ba8a06..0246b2e94d8cbd 100644
--- a/drivers/gpu/drm/arm/display/komeda/komeda_dev.c
+++ b/drivers/gpu/drm/arm/display/komeda/komeda_dev.c
@@ -49,7 +49,7 @@ static int komeda_register_open(struct inode *inode, struct file *filp)
 static const struct file_operations komeda_register_fops = {
 	.owner		= THIS_MODULE,
 	.open		= komeda_register_open,
-	.read		= seq_read,
+	.read_iter		= seq_read_iter,
 	.llseek		= seq_lseek,
 	.release	= single_release,
 };

And these code will be deleted by this patch, if we merge this patch into
drm-misc-next firstly before the patch-A, that may import a conflict when
we merge our misc into upstreams.

if we want it to be merged into drm-misc, I think we'd better to wait the
upstream (the patch-A) has been synced back to drm-misc.

And what's your opinion ?

Thanks
James

> >
> > Thanks
> > James
> >
> > >  #ifdef CONFIG_DEBUG_FS
> > >  static void komeda_debugfs_init(struct komeda_dev *mdev)
> > > --
> > > 2.17.1
> > _______________________________________________
> > dri-devel mailing list
> > dri-devel@xxxxxxxxxxxxxxxxxxxxx
> > https://lists.freedesktop.org/mailman/listinfo/dri-devel
> 
> 
> 
> -- 
> Daniel Vetter
> Software Engineer, Intel Corporation
> http://blog.ffwll.ch