| Message ID | 20160708145211.6bf83d8f.john@metanate.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On 2016年07月08日 21:52, John Keeping wrote: > On Fri, 8 Jul 2016 17:04:59 +0800, Chris Zhong wrote: > >> At the first time of bind, there is no any panel attach in mipi. Add a >> DRM_CONNECTOR_POLL_HPD porperty to detect the panel status, when panel >> probe, the dw_mipi_dsi_host_attach would be called, then mipi-dsi will >> trigger a event to notify the drm framework. >> >> Signed-off-by: Chris Zhong <zyw@rock-chips.com> > Can we do something like this instead? We know that the panel must > always be attached and this has the advantage that the display size will > be known when the framebuffer console loads. > > -- >8 -- > Subject: [PATCH] drm/rockchip: dw-mipi-dsi: defer probe if panel is not loaded > > This ensures that the output resolution is known before fbcon loads. > > Signed-off-by: John Keeping <john@metanate.com> > --- > drivers/gpu/drm/rockchip/dw-mipi-dsi.c | 11 +++++++++-- > 1 file changed, 9 insertions(+), 2 deletions(-) > > diff --git a/drivers/gpu/drm/rockchip/dw-mipi-dsi.c b/drivers/gpu/drm/rockchip/dw-mipi-dsi.c > index 6ef5f3be8468..c0499266d116 100644 > --- a/drivers/gpu/drm/rockchip/dw-mipi-dsi.c > +++ b/drivers/gpu/drm/rockchip/dw-mipi-dsi.c > @@ -1154,10 +1154,17 @@ static int dw_mipi_dsi_bind(struct device *dev, struct device *master, > > dsi->dsi_host.ops = &dw_mipi_dsi_host_ops; > dsi->dsi_host.dev = dev; > - return mipi_dsi_host_register(&dsi->dsi_host); > + ret = mipi_dsi_host_register(&dsi->dsi_host); > + if (!ret && !dsi->panel) { > + mipi_dsi_host_unregister(&dsi->dsi_host); > + drm_encoder_cleanup(&dsi->encoder); > + drm_connector_cleanup(&dsi->connector); > + ret = -EPROBE_DEFER; Hi John Do you verify this patch? I do the similar change before, but found panel can't probe. mipi_dsi_host_register will call device_add, I think the panel probe need this. Seems that mipi panel probe request mipi_dsi_host_register, mipi_dsi_host_register request panel on your patch, endless loop. Thanks. > + } > > err_pllref: > - clk_disable_unprepare(dsi->pllref_clk); > + if (ret) > + clk_disable_unprepare(dsi->pllref_clk); > return ret; > } > > > >
On Mon, 11 Jul 2016 08:46:53 +0800, Mark yao wrote: > On 2016年07月08日 21:52, John Keeping wrote: > > On Fri, 8 Jul 2016 17:04:59 +0800, Chris Zhong wrote: > > > >> At the first time of bind, there is no any panel attach in mipi. Add a > >> DRM_CONNECTOR_POLL_HPD porperty to detect the panel status, when panel > >> probe, the dw_mipi_dsi_host_attach would be called, then mipi-dsi will > >> trigger a event to notify the drm framework. > >> > >> Signed-off-by: Chris Zhong <zyw@rock-chips.com> > > Can we do something like this instead? We know that the panel must > > always be attached and this has the advantage that the display size will > > be known when the framebuffer console loads. > > > > -- >8 -- > > Subject: [PATCH] drm/rockchip: dw-mipi-dsi: defer probe if panel is not loaded > > > > This ensures that the output resolution is known before fbcon loads. > > > > Signed-off-by: John Keeping <john@metanate.com> > > --- > > drivers/gpu/drm/rockchip/dw-mipi-dsi.c | 11 +++++++++-- > > 1 file changed, 9 insertions(+), 2 deletions(-) > > > > diff --git a/drivers/gpu/drm/rockchip/dw-mipi-dsi.c b/drivers/gpu/drm/rockchip/dw-mipi-dsi.c > > index 6ef5f3be8468..c0499266d116 100644 > > --- a/drivers/gpu/drm/rockchip/dw-mipi-dsi.c > > +++ b/drivers/gpu/drm/rockchip/dw-mipi-dsi.c > > @@ -1154,10 +1154,17 @@ static int dw_mipi_dsi_bind(struct device *dev, struct device *master, > > > > dsi->dsi_host.ops = &dw_mipi_dsi_host_ops; > > dsi->dsi_host.dev = dev; > > - return mipi_dsi_host_register(&dsi->dsi_host); > > + ret = mipi_dsi_host_register(&dsi->dsi_host); > > + if (!ret && !dsi->panel) { > > + mipi_dsi_host_unregister(&dsi->dsi_host); > > + drm_encoder_cleanup(&dsi->encoder); > > + drm_connector_cleanup(&dsi->connector); > > + ret = -EPROBE_DEFER; > > Do you verify this patch? I do the similar change before, but found > panel can't probe. > > mipi_dsi_host_register will call device_add, I think the panel probe > need this. > > Seems that mipi panel probe request mipi_dsi_host_register, > mipi_dsi_host_register request panel on your patch, endless loop. Yes, I've been running this patch for months. It's possible that I have another patch that is needed to get this to work correctly (I have a whole set of Rockchip MIPI patches that I need to find time to submit, for now I've pushed a branch to Github [1]). I don't know what you mean about the panel probe requesting mipi_dsi_host_register, the panel probe calls mipi_dsi_attach() and is itself called from mipi_dsi_host_register(). My patch doesn't change this at all, the only change is that if the panel driver is not available, then the probe fails. Later, once the panel driver has been registered the probe will succeed. This is no different from the current behaviour if the order of probing drivers happens to load the panel before dw-mipi-dsi. [1] https://github.com/johnkeeping/linux/commits/topic/rockchip-mipi > > + } > > > > err_pllref: > > - clk_disable_unprepare(dsi->pllref_clk); > > + if (ret) > > + clk_disable_unprepare(dsi->pllref_clk); > > return ret; > > } > > > > > > > > > >
diff --git a/drivers/gpu/drm/rockchip/dw-mipi-dsi.c b/drivers/gpu/drm/rockchip/dw-mipi-dsi.c index 6ef5f3be8468..c0499266d116 100644 --- a/drivers/gpu/drm/rockchip/dw-mipi-dsi.c +++ b/drivers/gpu/drm/rockchip/dw-mipi-dsi.c @@ -1154,10 +1154,17 @@ static int dw_mipi_dsi_bind(struct device *dev, struct device *master, dsi->dsi_host.ops = &dw_mipi_dsi_host_ops; dsi->dsi_host.dev = dev; - return mipi_dsi_host_register(&dsi->dsi_host); + ret = mipi_dsi_host_register(&dsi->dsi_host); + if (!ret && !dsi->panel) { + mipi_dsi_host_unregister(&dsi->dsi_host); + drm_encoder_cleanup(&dsi->encoder); + drm_connector_cleanup(&dsi->connector); + ret = -EPROBE_DEFER; + } err_pllref: - clk_disable_unprepare(dsi->pllref_clk); + if (ret) + clk_disable_unprepare(dsi->pllref_clk); return ret; }