I will provide another update regarding OpenChrome Version 0.6.
In the previous announce e-mail I sent, I may have wrote that OpenChrome 
Version 0.6 will be released sometime in September.
I now feel like I may need additional 1 to 2 months to improve the code quality 
of FP (Flat Panel) detection and initialization code.
I do not necessarily want to use this e-mail to again criticize past developers 
who worked on OpenChrome, but that being said, the code quality of FP and TV 
code remains very poor.
I feel like I will need to spend more time fixing the code for FP before 
    I do not necessarily want to get into my personal life, but in the middle 
of July, the place I lived (an apartment somewhere in the United States of 
America) had a fire at 4 A.M., and as a result, I was not allowed to return to 
my place for about 7 weeks.
Fortunately, I happened to have a monitor capable of dual VGA and DVI input 
inside my car, so I was able to continue the development for the next 2 months 
inside a nice hotel. (Refer to the numerous code commits I made related to DVI 
support during this period.)
I am now back at my place, but the contractor the building owner hired to fix 
up my unit really messed up my living quarters to the point where they piled up 
my belongings like how trash ends up in the dumpster . . . (They stuffed most 
of my belongings in black trash bags. They also did this to my same floor 
neighbors as well.)
I definitely had stuff stolen, and several of my neighbors have also reported 
they valuables getting stolen by whoever that had access to the rooms.
It will likely take another 3 weeks of intense cleaning and organizing to get 
to the point where I can perform any development works at my place.
My numerous VIA Technologies hardware I use for validation before code commit 
is hiding somewhere in the pile, so I will likely not have access to them for 
several more weeks.
I was planning to start to work on supporting HDMI for VX900H chipset, and 
recently acquired Zotac ZBOX nano VD01 for this purpose.
I hope it was not stolen by some of the crooked employees of the contractor 
tasked to repair my unit.
    I also started to work with the new DRM code that is meant to support KMS / 
I managed to find an OS environment where I can compile the Linux kernel and 
install it.
Unfortunately, there is a bug with the code that prevents UniChrome IGP from 
displaying something on the monitor at this point.
I put a few fixes to the code (analog VGA related fixes), but so far, it has 
not fixed the problem.
I believe this is a UniChrome IGP specific bug since the new DRM has been 
observed to work with P4M890 (UniChrome Pro) and VX800 (Chrome9 HC3) chipsets.
One thing I noticed about the code is that it appears that it needs more than 
512 MB of RAM just to boot the OS.
I really consider this a bug since the older DRI1-based DRM can boot with far 
less RAM.
In case of the test system I set up (HP Pavilion a800n - KM400A chipset), it 
was not working with 512 MB of RAM (it will result in a kernel panic), but it 
can boot with 1 GB of RAM. (Of course, the VGA monitor is not working yet.)
I will continue to work with the code to improve it so that someday it can be 
mainlined with the Linux kernel.
Unfortunately, number of more senior Linux developers were not too receptive 
(my personal opinion) to the VIA Technologies Chrome IGP DRM project, so at 
this point, I feel like I will be on my own during the entire development 


As far as I am concerned, throwing out the existing DRM code advised by the 
above developers is not practical since VIA Technologies Chrome IGP was 
comparable in terms of functionality to Intel IGP up to the 3 or 4 Series 
chipsets, so analog VGA / DVI / FP support will be the baseline hardware 
support for VIA Technologies Chrome IGP DRM to be usable.


Kevin Brace
The OpenChrome Project maintainer / developer
Openchrome-devel mailing list

Reply via email to