Personal thoughts of Reiser 4 and the Kernel

Posted Jun 8, 2006 7:39 UTC (Thu) by Los__D (guest, #15263)
In reply to: Personal thoughts of Reiser 4 and the Kernel by cventers
Parent article: What's not going into 2.6.18

If it's just stylistic I'd say that it's silly to keep it out as it's a filesystem (not a piece of core code).

Wasn't one of the gripes that it touches major pieces of the core VFS, not just it's own? If it's still so, I can see a VERY good reason to be perfectionist about this.

It's quite a while since I followed the discussion though, so they might already have been forced to pull the VFS parts out.

Personal thoughts of Reiser 4 and the Kernel

Posted Jun 8, 2006 9:11 UTC (Thu) by cventers (guest, #31465) [Link] (1 responses)

I think you may be misremembering this a bit. To my recollection, one of
the key friction points was that reiser4 developed this whole new concept
of plugins, which was _very_ similar to the VFS. Kernel developers don't
want abstraction layers hanging off of abstraction layers. So it really
wasn't that they were abusing the VFS, it was just that they were building
their own VFS under the VFS.

It was suggested that they take the differences in their plugin system and
build those capabilities into the VFS itself, so that it would be
available to every filesystem.

I'm not sure what the status of the code is these days. I would imagine
that enough time has gone by that many of these things have been
addressed, and reiser4 is probably a lot closer to being ready for
mainline (if it isn't already).

Personal thoughts of Reiser 4 and the Kernel

Posted Jun 8, 2006 17:08 UTC (Thu) by Los__D (guest, #15263) [Link]

*plonk*

Yep, it was that way around! :)