From mboxrd@z Thu Jan 1 00:00:00 1970 From: Stephen Warren Subject: Re: Trimslice Barebone fails to detect HDMI connection, no DDC i2c channel present Date: Wed, 11 Dec 2013 15:17:23 -0700 Message-ID: <52A8E473.8050502@wwwdotorg.org> References: <20131211214911.GA24176@bart.dudau.co.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20131211214911.GA24176-hOhETlTuV5niMG9XS5x8Mg@public.gmane.org> Sender: linux-tegra-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Liviu Dudau , linux-tegra-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-Id: linux-tegra@vger.kernel.org On 12/11/2013 02:49 PM, Liviu Dudau wrote: > Hello, > > Prompted by a discussion on G+, here is a report of an issue I'm seeing > with mainline kernel (commit 78fd82238d0e57). > > The kernel fails to detect the DDC i2c interface (on Trimslice that's > on Tegra2 I2C-0 interface), so I'm getting no output on the HDMI That probably should be I2C-1 for HDMI; I2C-0 is likely the DVI port's DDC interface. But since you tested all ports, this isn't the cause of the problem. > U-Boot 2014.01-rc1-dirty (Nov 26 2013 - 22:29:37) Are you using an unmodified upstream U-Boot? The environment you posted doesn't match upstream's environment, and the "-dirty" implies you have some local modifications. If you break into U-Boot, can you read the EDID using the "i2c" command? In Linux, can you "i2cget" from bus 1 device 0x50 even if "i2cdetect" didn't see the EDID? Is the monitor you have plugged in known-working without any EDID-related issues?