OESF Portables Forum
Everything Else => Zaurus Distro Support and Discussion => Distros, Development, and Model Specific Forums => Archived Forums => Angstrom & OpenZaurus => Topic started by: rolf on January 01, 2007, 01:17:37 pm
-
Hi,
apparently, some information has recently become available via reverse engineering about the SD slot on the collie (http://josejx.net/collie). This means we have come a little bit closer to a native 2.6 kernel driver. I have set up a bounty so that people interested in this can make publicly known how much they are willing to contribute financially if such a driver where developed. Take a look at http://wiki.openzaurus.org/FAQ#SD_Card_related_questions (http://wiki.openzaurus.org/FAQ#SD_Card_related_questions) and pledge, develop and discuss about the bounty in general.
Regards
Rolf
-
As soon as I get some more time(within a couple of weeks), and my new 3200(tomorrow), I'm planning on working on a driver. Granted, I'm not the most skilled coder, and I've never done kernel stuff, but I'm going to attempt it.
Just thought that might be relevant, I'm not looking to collect this 'bounty'
-
As soon as I get some more time(within a couple of weeks), and my new 3200(tomorrow), I'm planning on working on a driver.
microsoft/linux, glad to hear you are planning to code for the Z. Looking forward to your contributions. I guess you are aware that this work is not necessary for the 3200? There are drivers in the 2.6 kernel for that model (not that I'd want to discourage you from writing that driver despite of this )
-
ah yes, you misunderstand what I was saying. I've currently got a 5500, and was using it, so I didn't want to mess up my flash by doing dev work for it. I just got my 3200 now, so I'm planning on using the 5500 only for dev work, and now I'm not worried about losing anything, or whatever.
-
I've currently got a 5500
Nice. Have you ever done any kernel work before (not necessarily Z)?
-
none whatsoever. This would be a learning project, with the hopeful outcome of helping the community as well. I don't want to see collie go by the wayside because of a driver issue. OZ is the only rom that still actively supports 5500, and now, with the upcoming Angstrom, the plan is to drop support there...
-
none whatsoever. This would be a learning project
I guess the learning curve is going to be quite steep and lengthy then. I wish you good luck, nonetheless. Let us know about your progress.
-
I've been told that people will help me in any way they can, just that some people don't have enough time to do it themselves
-
I've been told that people will help me in any way they can, just that some people don't have enough time to do it themselves
[div align=\"right\"][a href=\"index.php?act=findpost&pid=150402\"][{POST_SNAPBACK}][/a][/div]
The link provided in your earlier post is down at the moment. However, if what you refer to a bounty is financial pledges, sign me up. I need a modern OS that can be stable. It seems that the 2.6 kernel holds much promise.
Chris--
Update:
I finally was able to get onto the FAQ to add my name to the list for a bounty and for some reason it would not. So I am willing to pay $25.00 for a Collie driver that works with the 2.6 kernel.
CL-
-
I finally was able to get onto the FAQ to add my name to the list for a bounty and for some reason it would not. So I am willing to pay $25.00 for a Collie driver that works with the 2.6 kernel.
OZ wiki is quite unstable these days
Thank you for your pledge, I added it to the wiki for you. I hope you do not mind.
Best, Rolf
-
Unstable or just hard to access - I think this is an issue with its being hosted on berlios.
Si
-
I read that the SD Card Association has opened up their specs for both the physical layer and the host controller. And the Linux and FreeBSD folks are now working on drivers:
spec: http://www.sdcard.org/HostController/index.html (http://www.sdcard.org/HostController/index.html)
Linux driver: http://mmc.drzeus.cx/wiki/Linux/Drivers/sdhci (http://mmc.drzeus.cx/wiki/Linux/Drivers/sdhci)
FreeBSD driver: http://bsdimp.blogspot.com/2006/09/sdmmc-f...er-outline.html (http://bsdimp.blogspot.com/2006/09/sdmmc-freebsd-driver-outline.html)
On the strength of this, could Sharp now be persuaded to release their own driver source?
--Dan
-
I read that the SD Card Association has opened up their specs for both the physical layer and the host controller. And the Linux and FreeBSD folks are now working on drivers:
spec: http://www.sdcard.org/HostController/index.html (http://www.sdcard.org/HostController/index.html)
Linux driver: http://mmc.drzeus.cx/wiki/Linux/Drivers/sdhci (http://mmc.drzeus.cx/wiki/Linux/Drivers/sdhci)
FreeBSD driver: http://bsdimp.blogspot.com/2006/09/sdmmc-f...er-outline.html (http://bsdimp.blogspot.com/2006/09/sdmmc-freebsd-driver-outline.html)
On the strength of this, could Sharp now be persuaded to release their own driver source?
--Dan
[div align=\"right\"][a href=\"index.php?act=findpost&pid=154004\"][{POST_SNAPBACK}][/a][/div]
Doubtful. The linux devs over at kernel.org have set up a program where the dev enters into an NDA, and then will write a driver for the kernel, without the companies needing to do anything, and I think sharp has refused stuff like that in the past. Sharp is very much against the opening up of this chip, I'm not sure why. I need to get openembedded set up, and then I'm gonna start looking at existing sd code, see if I can't figure it out. It's very likely I won't get anywhere, but it's worth a shot.
-
I read that the SD Card Association has opened up their specs for both the physical layer and the host controller. And the Linux and FreeBSD folks are now working on drivers:
spec: http://www.sdcard.org/HostController/index.html (http://www.sdcard.org/HostController/index.html)
Linux driver: http://mmc.drzeus.cx/wiki/Linux/Drivers/sdhci (http://mmc.drzeus.cx/wiki/Linux/Drivers/sdhci)
FreeBSD driver: http://bsdimp.blogspot.com/2006/09/sdmmc-f...er-outline.html (http://bsdimp.blogspot.com/2006/09/sdmmc-freebsd-driver-outline.html)
On the strength of this, could Sharp now be persuaded to release their own driver source?
--Dan
[div align=\"right\"][a href=\"index.php?act=findpost&pid=154004\"][{POST_SNAPBACK}][/a][/div]
Doubtful. The linux devs over at kernel.org have set up a program where the dev enters into an NDA, and then will write a driver for the kernel, without the companies needing to do anything, and I think sharp has refused stuff like that in the past. Sharp is very much against the opening up of this chip, I'm not sure why. I need to get openembedded set up, and then I'm gonna start looking at existing sd code, see if I can't figure it out. It's very likely I won't get anywhere, but it's worth a shot.
[div align=\"right\"][a href=\"index.php?act=findpost&pid=154429\"][{POST_SNAPBACK}][/a][/div]
Sharp is a hardware company. They own the designs for their hardware, but as you all can see, software is not one of sharp's forte. they license it from their partners which are smaller software companies who own the code. so in order for sharp to share the code with anyone, they would firstly need to buy the code (or at least renegotiate their existing contract/license agreement which still costs a lot of money, ie lawyers need to be paid) from their software vendor before they can give it to you.
so simply put, it would cost sharp lots of money to give you the source code. they arent a charity organisation. they are in business to make profit. so unless you can find a suicidal manager inside sharp willing to do this you wont be getting any free source code...