Hmmm... bummer about the scavenged microdrive. Wonder if that's true for all, or just certain mp3 players?
Going to try something different tonight. I have 2 128MB CF cards. Here's my plan:
1. format card "A" as FAT16 (one partition) and put the LX nBkProOs.img on it.
2. format card "B" as ext2 (one partition) and put the LX rootfs on it.
3. put card B in a PCMCIA adapter.
4. Boot card A and plug in card B a few seconds later.
My hope is, this way I'll be able to get more of the LX filesystem than I did when I had to cram everyone onto just one 128MB card.
Well, it didn't work. Figured it would be a long shot. From the serial output, seems that it looks for the img file and the root file system to be on the same device.
Got to be the holidays... at least for those of us in the US.
I just got a hold of a spare hard drive. In the process of trying to convince my wife to let me put it in her PC and make a dual-boot Vista / Linux machine, so I have a decent machine I can put the OE toolchain on and start building Angstrom.
Definitely not for free. More to the point, though, this company doesn't have anything to do with the NBP or anything we're trying to do in this forum. They deal with their own embedded solutions (development boards and related software). Nice info to know, but I don't think it really belongs here.
On another note, I imagine not much will be done here until after the first of the year. I am curious if anyone has been able to play around with building Angstrom (or any of the other flavors).
I've been reading. Classic. The more you know the more you know you don't know. I'm still trying to figure out how this thing works. I've really only been following instructions rather than figuring anything out. I've got to go back to work around christmas so I'll be gone for about 35 days. Maybe I'll have something when I get back.
Posted 2007-12-06 12:40 PM
The PsionLX solution is already running, i don't know much about the process in the Angstrom project but Marc and myself are working on the PsionLX developer archive to split up the classified data. After this is done, i'm sure we can publish the data. Currently we are working on a wiki-page where all the wiki informations from the developers will be present (except of the personal data of course).
I'm sorry that this is a slow process but personaly the time is running through my fingers, school is pure stress at the moment and thats the primary stuff for me at this time. I hope we can publish the files before the new year's eve.
Over the past few years I've been dabbling in tuning my NAS box a KURO BOX pro. The community has an incredible group of devoted linux/hardware people who have done amazing things. U-boot is one of their seminal projects that has catapulted the evolution of this platform to new heights not available elsewhere. I recently flashed u-boot on my server and have been learning about it. In the process, I have found so many similarities between the kurobox and netbook pro platforms. Particularly with the layout of the flash and the load process.
It probably comes as no surprise, but we're not reinventing the wheel here. This platform is well developed and robust. A unique aspect however, is the suspend mode.
On to my point. Trying to understand BooSt has led me to search the internet for sources. I haven't come up with any but I did find a lot of usage with wince and psion. As we know, the loader can load any image once it initializes a rudimentary set of peripherals. It simply passes control at some point to an image(kernel) that is formatted properly. U-boot has a lot of flexibility as to what that ultimately is whether it's a flash image, a disk based image or even a networked image.
I've found the compression stuff to be pretty common too. Four megs seems to be the minimum for loading a robust "real" kernel and various tricks can be done to chop it up and maximize the real estate.
I don't think our project needs that for the time being as we can load any image we put on the cf disk as long as the header is set up properly (don't know how to do that but apparently old man and others do).
As others have indicated, the goal now needs to be setting up a toolchain (not clear on all that entails) and selecting parts that are needed for a compile. My needs are not that great. I want the old netbook experience with better hardware and usb/audio and instant on. I've heard others mention full blown xwindows . . . I think that's a bit ambitious for this platform.
I hate to say it but I see the abiword apps, nice as they are, pretty much too much (not bloatware but definitely taxing the processor capabilities). I may be talking out my butt (perhaps it's my configuration) on this but that has been my experience.
Anyway, I want to slip into the stream of the openembedded people and start learning the language better so that I can start assembling something closer to what I want. This is starting to sound like a job description with the requisite demands on time and effort.
Oh, Eddie, to answer your quesition, 'is it possible to replace or override Boost with one of these other apps?', I would say we don't have to although it would be nice just to have something with documentation and flexibility. However, the mmc card doesn't seem to work. What is the source of that? The bootloader? I don't really think so.
I pretty much enjoyed reading this thread, Good to see there are still people working on making these nice devices usable.
This tool for creating images is great - It would be even better to have it downloadable from some public server. Of course I can host it, but it would be nice to have it under some open cource license then. With that we could create new kernel images for the NBP with OpenEmbedded consistent with the filesystem images easily.
For porting support for the NBP to a more up to date kernel you could get in touch with the Mobile-Linux project at http://projects.linuxtogo.org/projects/mobile-linux/. This project is meant to join efforts for Linux ports to mobile devices.
Keep up the good work!
With all the new linux devices coming out eee, packard bell, via, in the umpc format, I wonder if teklogix will pick up the pro development again.
If nothing else, these new devices present some interesting case studies in form factor and function.
I notice that the via is using the e17 wm while asus has icewm. The fact that they both can run windows xp tells me they have some major power advantages over the netbook. Personally, I consider that cheating. If you can't do it with 32MB, what kind of man/woman are you? But in the end, they have to come clean and confess their relatively meager battery performance.
Reading the numerous reviews, I couldn't help but laugh at comments like, ''if they only made a bigger screen" or "where's the optical drive". It seems like the most misunderstood market segment. Perhaps it's human nature to want to make everything bigger and faster.
So, having lusted over these new linux based machines and gotten over my desire to own one, I resign once again to get what I want on the netbook pro.
There's been no discussion here short of some great encouragement from none other than Florian, one of the stars from open embedded.
Can someone point me to some instruction on assembling the so-called "tool chain"?