I am totally disgusted with the Min2451, the successor to the popular Mini2440. While the Mini2440 used a highly respected open source boot loader, u-boot, the Mini2451 uses this piece of crap called Superboot that has no documentation and as far as anyone can tell has no features. I've had it. I got sucked into buying 20 of these bricks and I'm going to find a replacement board for my Embedded Linux Learning Kit.
Dear Abbott, Superboot is more simpler and more efficient than supervivi on older boards. And Yes, Mini2451 is really good upgrade for Mini2440.
This thread has gone quiet: http://comments.gmane.org/gmane.comp.embedded.ptxdist.devel/12338 however, I wonder if it is worth getting in touch with them.
Well, you're absolutely right that superboot is "simpler" because it has no features! There's no way to control the boot process, you can't boot the kernel over TFTP, and you can't mount the root file system over NFS. Mounting the RFS over NFS is how you do development! What are you people thinking?!?! Yes, the Mini2451 is an improvement in hardware, but that's useless if the software isn't usable. Seriously, has anyone else been suckered into using the Mini2451? Oh, and I'm also very disappointed in your distribution of a fork of the kernel. Why didn't you submit your changes to kernel.org?
Hi Doug I think they do themselves a disservice, for some very useful hardware, by the binary blob approach they are using. There is a lot of competition in this space, with lots of support, and getting better and better all the time. Sean
I completely agree, the hardware is very good. If they just followed mainstream kernel, uboot and samsung sources, it would give them less work and would make their platform great and better. They would just have upstream features included in their product effortlessly. I give the example of nanopc-t1: It could just put raspberry and others in the pocket if they just didnt do illegal stuff like not releasing uboot code that they are using mixed with the superboot blob(take the ubuntu t1 image for example). They release the source for tiny4412, and not for nanopc. What is the problem? They already did modify uboot to run on the nano pc for the ubuntu image... Plus, it is illegal not to release it. The result is that we developers that buy their products, if we want to build an industrial linux solution for it, we obly have a old and undupported image of ubuntu, a stalled kernel with some bugs on the mfc component, and no uboot to do the normal development process (tftp,etc) There is awesome support for samsung soc related stuff on the mainstream kernel That makes me think that someone at friendlyarm (the one that makes decisions) is kind of stupid.