Re: the " 'official' point of view" expressed by kernelnewbies.orgregarding reiser4 inclusion

From: Jeff Garzik
Date: Mon Jul 24 2006 - 14:27:12 EST

Olivier Galibert wrote:
The fact that the ext maintainers are very, very good helps quite a
lot too. But I think it doesn't change the fact that if r4 has been a
set of patches through time to r3, good or not, there wouldn't be a

That's a huuuuuge leap of logic.

Metadata plugins found in reiser4 are far better done at the VFS level, than burying "reiser4 can look like ext2, if it wishes" functionality inside the filesystem.

I guarantee such a patch to reiser3 would get rejected.

It's maybe the lack of an official development branch, but it looks
like the kernel development has become very risk-averse, and the bar
is set much higher to accept anything that looks relatively new. Any
reason is good to have it dropped, cosmetic or not.

New stuff goes in all the time.

The bar is set too high in some cases (read: SCSI subsystem submissions), but reiser4 submission cannot be generalized as you have done here. There are very real issues present, that need to be dealt with.

Just to give you an idea, if the criteria applied to suspend2 or
reiser4 had been applied to everything else, we wouldn't have at least
XFS[1], ALSA[2], sysfs[3] and DRM[4]. Whether it is good or bad is an
interesting question itself. But before, code just had to be
reasonably sane, and it was expected to be fixed through time. Some
even has been (sysfs got better). Now it has to attain an ever moving
level of perfection before it has a chanc to be accepted.

reiser4 tries to be another VFS. That's a bit more than needing additional minor fixes over time.


To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at
Please read the FAQ at