From mboxrd@z Thu Jan 1 00:00:00 1970 From: Arno Willig Subject: I2C writes produce duplicated bytes Date: Tue, 29 Apr 2014 12:44:23 +0000 (UTC) Message-ID: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Received: from plane.gmane.org (plane.gmane.org [80.91.229.3]) by gabe.freedesktop.org (Postfix) with ESMTP id 7C8646E9F1 for ; Tue, 29 Apr 2014 05:50:10 -0700 (PDT) Received: from list by plane.gmane.org with local (Exim 4.69) (envelope-from ) id 1Wf7U9-0000e1-0c for dri-devel@lists.freedesktop.org; Tue, 29 Apr 2014 14:50:09 +0200 Received: from bdit.hertznet.de ([80.149.83.93]) by main.gmane.org with esmtp (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Tue, 29 Apr 2014 14:50:09 +0200 Received: from arno.willig by bdit.hertznet.de with local (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Tue, 29 Apr 2014 14:50:09 +0200 List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dri-devel-bounces@lists.freedesktop.org Sender: "dri-devel" To: dri-devel@lists.freedesktop.org List-Id: dri-devel@lists.freedesktop.org Hi, I am working with "ASRock FM2A75M-HD+" mainboards equipped with "AMD A4-4000 APU"s. For some reasons I am required to write EDID data to VGA-connected TFT panels which worked fine with an older mainboard with nvidia gpu. With the ASRock board the smbus i2c DDC channel was recognized as "Radeon aux bus DP-auxch" in linux 3.13.5 by i2cdetect. However this seems to be fixed in 3.15.0-rc3, now it's "card0-VGA-1", which looks correct. My problem is the following: I tried to write to address 0x10 in the EDID block (val==0xaa) i2cset -y 6 0x50 0x10 0xaa Unfortunately the byte value is also copied to address 0x11, which is bad. Some experiments showed that bytes written to all addresses are duplicated to the next address. (Apparently a byte value written to address 0x0f is duplicated to 0x00 and 0x1f is duplicated to 0x10 and so ). This behaviour also seems to render ddc/ci commands unusable on this hardware. Could this be a kernel bug in the radeon driver or is it an issue with the mainboards? Thanks for helping! Ciao, Arno