Texas Instruments Finland Oy, Porkkalankatu 22, 00180 Helsinki. Y-tunnus/Business ID: 0615521-4. Kotipaikka/Domicile: Helsinki On 28/09/17 11:56, Laurent Pinchart wrote: >> Was there a reason to create the omapdrm device in omap_dss_init (i.e. >> module_init), instead of omapdss's probe? Now the omapdrm device is >> always created even if there's no DSS device on the SoC. > > Not really, no. > > Do you think this requires a quick -stable fix ? I'm working on a patch series > that will remove the omapdrm device completely. With -stable, you mean 4.14? Afaics, the only downside is a small amount of memory that's not used. It does behave differently than before, as the now-removed platform code did only add the omapdrm device if there's a dss device in the DT data. I don't think that really matters, though. But still, maybe it's nice to fix for 4.14, as we're still only at -rc2. We encountered this with TI kernel, which has DSS6 driver. The omapdrm side is shared, but the dss side is different than with the current dss. So, the dss6 driver needs to create omapdrm device, but old dss driver already creates it... We'll hack around it, as it's only an issue in a development kernel. Tomi _______________________________________________ dri-devel mailing list dri-devel@xxxxxxxxxxxxxxxxxxxxx https://lists.freedesktop.org/mailman/listinfo/dri-devel
