Re: [Unichrome-devel] VIA VT1631 LVDS
Brought to you by:
dwdeath
From: Philip P. <phi...@re...> - 2007-04-23 20:17:58
|
Luc Verhaegen. wrote: > On Sat, Apr 21, 2007 at 12:13:45AM -0600, Philip Prindeville wrote: > >> Or you could consider sharing what you've learned over >> those years for the collective good... >> > > Do you mean like writing an absolutely solid driver and making it > available for everyone to use? > > Maybe you should try looking at this driver. > > Luc Verhaegen. > Well, I did exactly that. I took patches from diffing the Unichrome and Openchrome versions of the drivers, and built and loaded the Openchrome code with the following Unichrome via_i2c.c diffs, and retested my hacked up interface to the sil164 submodule. No joy. The device is detected (via xf86i2cProbeAddress()), but fails to respond to a ReadByte() from sub-address 0 (and just about every I2C device has something at sub-address 0). So I'm well and truly befuddled. And so as to not spark another Unichrome vs. Openchrome flame war, I'm working with the Openchrome driver for now because that's what the current xorg-x11-drv-via is based on. The Intel and Radeon drivers are trying to come up with a standard for an API for TMDS encoders as sub-modules, and I'm trying to support that effort by adding the Via driver to the list of TMDS-encoder-aware drivers. Once that works and the API is solidified, I will backport the changes to the Unichrome-tree as well. But at this rate... not even being able to read a byte from the TMDS chip... I'm wondering if that's every going to happen. I was also looking at the Via Tech drivers, and couldn't figure out the whole GPIO_I2C* stuff. Anyone know what that's about? Or where/when it's relevant? -Philip |