Hi all,
We have both tried all the builds on the pandaboard 4430 but we have encountered some problems. It seems like those builds are depending on the screen on how they will behave.
The problem at my side is that the display doesn't want to show something. I have tried all the builds on my Asus computer screen (HDMI 1920x1080) and my Sony Braviahttp://store.sony.com/webapp/wcs/stores/servlet/ProductDisplay?catalogId=10551&storeId=10151&langId=-1&partNumber=KDL40NX700#features(HDMI 1920x1080). But every time the screen stayed black. This was all tested with the HDMI port to HDMI in the screen.
At Mark, the builds worked on his tv at 720p (also Sony Bravia but older) through HDMI. But when he connected the pandaboard from HDMI to DVI (with an converter) it started in phone mode with a low resolution.
We followed the instructions of the links you send to us carefully and we can confirm that Android was running.
Could there be a problem with the EDID protocol in combination with our screens? And is there a way to override it when the screen doesn't support the EDID protocol?
With kind regards, Mark Kuipers Ion Aalbers
On 19 January 2012 14:19, Andy Green andy.green@linaro.org wrote:
On 01/19/2012 03:37 PM, Somebody in the thread at some point said:
Thanks for all your responses. We will try it as soon as possible :)
So this tells me that the build determines which resolution it has, this can not be changed through any setting. Is this right?
You can try messing with dss kernel commandlines but I dunno you get anywhere good.
Fracture in video status reflects fracture in TI approach to how the work is done.
There's an upstream effort to get definitive mainline support in, and separately there's customer support based approach from TI that is mainly targeted at Android / Omapzoom.
The code quality is best in mainline but support is incomplete, especially about multihead. In OZ and OZ-ish kernels, filter for what goes in is not as aggressive as mainline, enablement is better but basis is always an old release. That means, the old, working 3.0 code diverges from what got into mainline for later releases and can't be brought on easily unless treated as alien code like SGX.
Linaro TI LT kernel is trying to be a third approach, while worshipping mainline HEAD above all, so we have recent development basis, we glue on whatever missing pieces we can scavenge from wherever, keeping them up to date with Linus HEAD with to goal of being a highly enabled "fake upstream" with the missing pieces integrated.
In your case that boils down to: Jassi added code to come out on HDMI connector at 1080p via EDID on our kernel, AOSP omap 3.0-based kernel doesn't have that support.
-Andy
-- Andy Green | TI Landing Team Leader Linaro.org │ Open source software for ARM SoCs | Follow Linaro http://facebook.com/pages/Linaro/155974581091106 - http://twitter.com/#%21/linaroorg - http://linaro.org/linaro-blog