On 07/10/2012 10:58 PM, Simon Glass wrote: > Hi Stephen, > > On Fri, Jun 15, 2012 at 1:47 AM, Stephen Warren <swar...@wwwdotorg.org > <mailto:swar...@wwwdotorg.org>> wrote: > > On 06/13/2012 10:19 AM, Simon Glass wrote: > > Add calls to the LCD driver from Nvidia board code. > > > diff --git a/board/nvidia/common/board.c b/board/nvidia/common/board.c > > > @@ -87,6 +88,9 @@ int board_init(void) > > > +#ifdef CONFIG_VIDEO_TEGRA2 > > + tegra_lcd_check_next_stage(gd->blob, 0); > > +#endif > > This seems to be conflating video support with LCD support. It would be > quite possible to have a board with no LCD, yet supporting display over > HDMI for example. In other words, shouldn't the ifdef above be something > more like: > > #if define(CONFIG_LCD_SUPPORT) > register_lcd_driver(); > #endif > #if defined(CONFIG_VIDEO_TEGRA2) > tegra_display_init(...); > #endif > > and internal to tegra_display_init(), the DT is searched for LCD > controller nodes, and if any are found, they're matched to the LCD > driver registered by the first call above. > > > Yes that sounds great, but again we don't really have this > infrastructure in U-Boot. We would be inventing it just for Tegra, and I > would prefer to wait until the device model stuff is done before being > too fancy.
What is "the device model stuff" you mention? > We don't have an HDMI driver at present, so perhaps if/when that appears > in U-Boot it would be a good time to add support for that? Certainly adding code for HDMI can wait until later. However, I do think that we should use the correct ifdefs up-front, so that e.g. adding HDMI support only means adding a bunch of code, not going through the existing code and untangling conflated ifdefs. _______________________________________________ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
