On 06/09/2011 09:32 PM, Andrew Morton wrote:
On Thu, 9 Jun 2011 14:49:47 +0100
Andy Whitcroft<apw@xxxxxxxxxxxxx> wrote:
On Wed, Jun 08, 2011 at 08:52:33PM -0700, Andrew Morton wrote:
The problem with the userspace alternative is that it does not work. I
tried to run my live CD on top of unionfs-fuse and the filesystem
would fail intermittently leading to random errors during boot.
If the implementation is slow or buggy then the appropriate action is
to speed it up and to fix the bugs, so these are just non-arguments,
IMO.
Exactly. It is rather sad that people never bothered to file bug reports
about slow performance issues. I'm one of the upstream authors of
unionfs-fuse and also use it on my own for live-USB sticks and
NFS-booted systems and do not have such problems.
If it is demonstrated that the userspace implementation simply cannot
ever have acceptable performance then OK, we have an argument for a
kernel driver.
Well, I know rather well were most of the performance issues in
unionfs-fuse come from (at least the normal issues, not those above
where it seems to fail at all). And those issues are mostly unrelated to
user-space, but just due to very simple approaches to get a working
union implementation at all (neither kernel solution actually does what
I need...). The main problem is that I barely find spare time to further
improve unionfs-fuse...
Another real difficulty with user-space implementations are init scripts
- while it works perfectly with Debian, I always have the feeling that
RedHat does everything they to make it impossible to use fuse as root
file system (eventually I always get it working, till the next minor
release, which breaks all workarounds again).
Anyway, given my lack of time and the resulting slow development of
unionfs-fuse, distros probably need another solution for their live CDs.
Cheers,
Bernd
--
To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
