[PATCH v9 5/8] drm: Remove usage of deprecated DRM_ERROR

Siddh Raman Pant code at siddh.me
Tue Jun 6 17:06:25 UTC 2023


On Tue, 06 Jun 2023 20:14:52 +0530, Laurent Pinchart wrote:
> Hi Siddh,
> 
> Thank you for the patch.

Anytime :)

> >       if (!crtcs || !modes || !enabled || !offsets) {
> > -             DRM_ERROR("Memory allocation failed\n");
> > +             drm_err(client->dev, "Memory allocation failed\n");
> 
> We could probably drop this message as memory allocation functions are
> already vocal on failure, but that's a separate fix.

Okay. Should I send a patch at the end of the series removing the
superfluous messages you pointed out in drm core?

> >       if (!drm_core_init_complete) {
> > -             DRM_ERROR("DRM core is not initialized\n");
> > +             drm_err(NULL, "DRM core is not initialized\n");
> 
> Could this use dev ?

No, the drm_device's dev pointer is assigned later. See line 621.

> >       if (!vma_offset_manager) {
> > -             DRM_ERROR("out of memory\n");
> > +             drm_err(dev, "out of memory\n");
> 
> Same here, I think the message could be dropped.

Okay.

> >       if (!ht->table) {
> > -             DRM_ERROR("Out of memory for hash table\n");
> > +             drm_err(NULL, "Out of memory for hash table\n");
> 
> Same.

Okay.
 
> With the commit message fixed as mentioned in the review of an earlier
> patch in this series, and the issue in drm_dev_init() addressed if
> needed,
> 
> Reviewed-by: Laurent Pinchart laurent.pinchart+renesas at ideasonboard.com>

Thanks,
Siddh


More information about the dri-devel mailing list