On Mon, 13 Mar 2000, Jamie Lokier wrote:
> Yury Yu. Rupasov wrote:
> > Hello !
> >
> > I could not find new bugs.
> >
> > I did :
> > 1. stress.sh 50 with big source tree (~750 MB)
> > 2. postmark 50000 files, 50000 transactions
> > 3. creating 1 million of small files
> > 4. tar / untar kernel source tree
> > 5. compile 2.3.51 kernel on reiserfs partition
> > 6. bonnie++ with default parameters.
> > 7. mongo.sh with default params and with 56000 files in 1 dir.
> >
> > Reiserfs-3-6-1 for linux-2.3.51 are stable. At least I have not found
> > bugs yet.
>
> Alex Viro's concerns are not about whether reiserfs passes stress
> tests. They're about subtle, and not to subtle deadlock, livelock and
> race conditions, and long term maintainability.
>
Ok, someone else sent a similar reply, but just to the reiserfs list, so
my reply didn't make it to linux kernel. You are correct, the tests Yura
ran do not answer any of the VFS interface concerns brought up here.
They were not meant to. Yura was running some benchmarks and stress tests
that our 2.3.49 patch had problems with. When he said the patch was ready
to go out, he meant it was ready for me to send to the linux kernel list.
The 2.3 port is still beta code.
> As soon as an fs goes into the kernel, even marked as "experimental",
> then every time Alex modifies VFS he has to look at all the kernel
> filesystems and verify that his VFS changes are valid. Sometimes this
> means he has to change the filesystems themselves. Also Alex Viro is
> not the only person who does this.
>
> That can only be done if every fs in a single kernel release makes the
> same set of assumptions about what to test, what locks are held for
> different functions, when to call iput, dput, d_delete and d_move,
> whether to check S_ISDIR etc.
>
> Currently reiserfs does not make the same assumptions as the current
> kernel filesystems. So it cannot be included even though it passes its
> own stress tests: including reiserfs would create too much work for
> people modifying and auditing VFS.
>
These are excellent points, and we are cleaning things up. If someone is
making those changes, and can't get to reiserfs for some reason (even if
the code is perfect, time constraints might kick in), we can help
propogate the change into our source.
-chris
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Wed Mar 15 2000 - 21:00:25 EST