Is it worth removing the WRT "architecture" and just keeping mips, x86 and arm? I think so, what does the list think?
For the embedded book, the WRT arch is the same as MIPS (it's a MIPS processor inside) but WRT has some extra steps to build firmware, setup to flash images onto a wrt54, and there's an older kenel config provided. But WRT hasn't been maintained in a long time and is getting quite out of date compared to the core parts of the embedded book (for x86, mips and arm). The embedded book is getting little love, as of late, but I hope to correct that. Removing WRT will make loving the embedded book easier. I understand the desire to have a target that's cheap and easy to buy, but the WRT54 series of routers are no longer cheap compared to things like a BeagleBone Black or a Raspberry Pi. A few years ago, the WRT54 was the cheap dev kit, it's less so now. Besides that there have been circuit changes to it over the years and I'm not even sure that the WRT54GL you can buy today is the same system as it was a few years ago. Supporting the unique hardware setup of the WRT, or any other dev kit, is something I'd like to get away from (see my bootloaders / kernel config question [1]). [1]:http://lists.cross-lfs.org/pipermail/clfs-dev-cross-lfs.org/2013-June/001443.html I'd prefer to help people build an embedded filesystem, possibly with QEMU configuration for the kernel, rather than try to support lots of hardware. Transitioning to specific hardware would be left up to the user (mainly that consists of bootloader, kernel, and device tree). Thanks, Andrew _______________________________________________ Clfs-dev mailing list [email protected] http://lists.cross-lfs.org/listinfo.cgi/clfs-dev-cross-lfs.org
