Hi,
On Sat, Aug 15, 2020 at 12:46:51PM +0200, Sam Ravnborg wrote:
> On Sat, Aug 15, 2020 at 12:40:22PM +0200, Guido Günther wrote:
> > Hi Sam,
> > On Sat, Aug 15, 2020 at 12:02:30PM +0200, Sam Ravnborg wrote:
> > > Hi Guido.
> > >
> > > > +static int mantix_probe(struct mipi_dsi_device *dsi)
> > > > +{
> > > > + struct device *dev = &dsi->dev;
> > > > + struct mantix *ctx;
> > > > + int ret;
> > > > +
> > > > + ctx = devm_kzalloc(dev, sizeof(*ctx), GFP_KERNEL);
> > > > + if (!ctx)
> > > > + return -ENOMEM;
> > > > +
> > > > + ctx->reset_gpio = devm_gpiod_get(dev, "reset", GPIOD_OUT_LOW);
> > > > + if (IS_ERR(ctx->reset_gpio)) {
> > > > + DRM_DEV_ERROR(dev, "cannot get reset gpio\n");
> > > > + return PTR_ERR(ctx->reset_gpio);
> > > > + }
> > > > +
> > > > + mipi_dsi_set_drvdata(dsi, ctx);
> > > > + ctx->dev = dev;
> > > > +
> > > > + dsi->lanes = 4;
> > > > + dsi->format = MIPI_DSI_FMT_RGB888;
> > > > + dsi->mode_flags = MIPI_DSI_MODE_VIDEO |
> > > > + MIPI_DSI_MODE_VIDEO_BURST | MIPI_DSI_MODE_VIDEO_SYNC_PULSE;
> > > > +
> > > > + ctx->avdd = devm_regulator_get(dev, "avdd");
> > > > + if (IS_ERR(ctx->avdd)) {
> > > > + ret = PTR_ERR(ctx->avdd);
> > > > + if (ret != -EPROBE_DEFER)
> > > > + DRM_DEV_ERROR(dev,
> > > > + "Failed to request avdd regulator: %d\n",
> > > > + ret);
> > > > + return ret;
> > > > + }
> > >
> > > Consider to use the recently added dev_err_probe() here and below.
> > > Note: Not part of drm-misc-next yet - but hopefully after -rc1
> > > when a backmerge is done.
> >
> > In fact I did decided against it since i was told that missing dev_* and
> > DRM_* logging shouldn't be done. So is that o.k. nowadays?
> s/missing/mixing/
>
> I often request that logging is consistent - so I recognize the
> argument.
>
> For panel/* I have not made up my mind what I think is the best
> approach. The DRM_DEV_* and DRM_* logging do not add much value.
> So I have been tempted several times to convert all logging in
> panel/ to dev_* and pr_* (when no struct device * is available).
> That would also avoid that we mix up logging.
>
> We have drm_* logging - but they require a valid drm_device * which we
> do not have in the the panel drivers. So they are ruled out here.
>
> Do you have any opinions/comments on this?
I think for panel drivers DRM_* does not give any bonus so moving to
{dev,pr}_* sounds good. I just wonder if other drm parts don't need
`dev_drm_err_probe()` (or similar) anyway. But then maybe dyn_debug
is enough nowadays to not need DRM_DEV_DEBUG_* either?
Cheers,
-- Guido
>
> Sam
>
