> > Hm. It looks like I might have been wrong about that. Digging a > > little deeper, it seems like plan9port doesn't have any fdisk, > > prep, or fossil. Is that correct? (And if so, why?) I know that > > Linux doesn't have any kernel-level support for Plan 9 partitions > > or the fossil file system. But can't that stuff be done in > > userspace, too? > > You can port it if you want. Plan9port tries to integrate well with > the host operating system. Part of that means using the same > file tree, and the host OS cannot talk to fossil. > > There is disk code that knows how to read Plan 9 partition tables. > The venti server uses it to let you write /dev/sdC0:arenas.
I started a port of Fossil to plan9port few months ago. It seems to be working fine, but: - the console doesn't work, - it doesn't handle Plan 9 partitions tables, - I haven't wrote the manuals. I hope to fix all of these problems in the future. The lack of console is obviously a big issue. Beware this is not exactly the Plan 9 Fossil. This is the modified Fossil I use on Plan 9. This Fossil is based on libventi and libthread, instead of liboventi. This is why liboventi haven't been ported. I considered it is too experimental to be submitted to plan9port, but the patches are available online: - [1] fossil-plan9: import Fossil from Plan 9, - [2] fossil-new: modified Fossil, - [3] fossil-new-p9p: required plan9port changes on Fossil. However, I recently submitted two needed patches to libventi: - [4] libventi-redial: add redial and reconn functions, - [5] libventi-noarchive: add VtEntryNoArchive. These two patches add features, currently part of liboventi, missing from libventi, but useful to Fossil. [1] http://www.9legacy.org/p9p/patch/fossil-plan9.diff [2] http://www.9legacy.org/p9p/patch/fossil-new.diff [3] http://www.9legacy.org/p9p/patch/fossil-new-p9p.diff [4] http://codereview.appspot.com/download/issue5325046_1.diff [5] http://codereview.appspot.com/download/issue5298088_1.diff If you want to contribute, I will be happy to receive patches and bug reports. -- David du Colombier
