From: Christoph Hellwig <hch@infradead.org>
To: Dave Chinner <david@fromorbit.com>
Cc: linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org,
axboe@kernel.dk
Subject: Re: fs: Inode cache scalability V3
Date: Wed, 13 Oct 2010 19:36:48 -0400 [thread overview]
Message-ID: <20101013233647.GA18691@infradead.org> (raw)
In-Reply-To: <20101013214609.GA24695@infradead.org>
On Wed, Oct 13, 2010 at 05:46:09PM -0400, Christoph Hellwig wrote:
> On Wed, Oct 13, 2010 at 11:58:45AM -0400, Christoph Hellwig wrote:
> >
> > It's 100% reproducible on my kvm VM. The bug is the assert_spin_locked
> > in redirty_tail. I really can't find a way how we reach it without
> > d_lock so this really confuses me.
>
> We are for some reason getting a block device inode that is on the
> dirty list of a bdi that it doesn't point to. Still trying to figure
> out how exactly that happens.
It's because __blkdev_put reset the bdi on the mapping, and bdev inodes
are still special cased to not use s_bdi unlike everybody else. So
we keep switch between different bdis that get locked.
I wonder what's a good workaround for that. Just flushing out all
dirty state of a block device inode on last close would fix, but we'd
still have all the dragons hidden underneath until we finally sort
out the bdi reference mess.
next prev parent reply other threads:[~2010-10-13 23:36 UTC|newest]
Thread overview: 49+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-10-13 0:15 fs: Inode cache scalability V3 Dave Chinner
2010-10-13 0:15 ` [PATCH 01/18] kernel: add bl_list Dave Chinner
2010-10-13 0:15 ` [PATCH 02/18] fs: Convert nr_inodes and nr_unused to per-cpu counters Dave Chinner
2010-10-13 0:15 ` [PATCH 03/18] fs: Implement lazy LRU updates for inodes Dave Chinner
2010-10-13 13:32 ` Christoph Hellwig
2010-10-16 0:11 ` Dave Chinner
2010-10-16 7:56 ` Nick Piggin
2010-10-13 0:15 ` [PATCH 04/18] fs: inode split IO and LRU lists Dave Chinner
2010-10-13 11:31 ` Christoph Hellwig
2010-10-13 0:15 ` [PATCH 05/18] fs: Clean up inode reference counting Dave Chinner
2010-10-13 11:33 ` Christoph Hellwig
2010-10-13 0:15 ` [PATCH 06/18] exofs: use iput() for inode reference count decrements Dave Chinner
2010-10-13 11:34 ` Christoph Hellwig
2010-10-13 14:49 ` Boaz Harrosh
2010-10-17 1:24 ` Christoph Hellwig
2010-10-24 18:06 ` Boaz Harrosh
2010-10-13 0:15 ` [PATCH 07/18] fs: rework icount to be a locked variable Dave Chinner
2010-10-13 11:36 ` Christoph Hellwig
2010-10-16 0:15 ` Dave Chinner
2010-10-16 0:20 ` Dave Chinner
2010-10-16 0:23 ` Christoph Hellwig
2010-10-13 0:15 ` [PATCH 08/18] fs: Factor inode hash operations into functions Dave Chinner
2010-10-13 0:15 ` [PATCH 09/18] fs: Introduce per-bucket inode hash locks Dave Chinner
2010-10-13 11:41 ` Christoph Hellwig
2010-10-13 15:05 ` Christoph Hellwig
2010-10-13 0:15 ` [PATCH 10/18] fs: add a per-superblock lock for the inode list Dave Chinner
2010-10-13 0:15 ` [PATCH 11/18] fs: split locking of inode writeback and LRU lists Dave Chinner
2010-10-13 3:26 ` Lin Ming
2010-10-13 13:18 ` Christoph Hellwig
2010-10-13 0:15 ` [PATCH 12/18] fs: Protect inode->i_state with the inode->i_lock Dave Chinner
2010-10-13 13:27 ` Christoph Hellwig
2010-10-13 0:15 ` [PATCH 13/18] fs: introduce a per-cpu last_ino allocator Dave Chinner
2010-10-13 0:15 ` [PATCH 14/18] fs: Make iunique independent of inode_lock Dave Chinner
2010-10-13 0:15 ` [PATCH 15/18] fs: icache remove inode_lock Dave Chinner
2010-10-13 2:09 ` Dave Chinner
2010-10-13 13:42 ` Christoph Hellwig
2010-10-13 0:15 ` [PATCH 16/18] fs: Reduce inode I_FREEING and factor inode disposal Dave Chinner
2010-10-13 13:51 ` Christoph Hellwig
2010-10-13 0:16 ` [PATCH 17/18] fs: split __inode_add_to_list Dave Chinner
2010-10-13 15:08 ` Christoph Hellwig
2010-10-13 0:16 ` [PATCH 18/18] fs: do not assign default i_ino in new_inode Dave Chinner
2010-10-16 7:57 ` Nick Piggin
2010-10-16 16:30 ` Christoph Hellwig
2010-10-13 14:51 ` fs: Inode cache scalability V3 Christoph Hellwig
2010-10-13 15:58 ` Christoph Hellwig
2010-10-13 21:46 ` Christoph Hellwig
2010-10-13 23:36 ` Christoph Hellwig [this message]
2010-10-13 23:55 ` Dave Chinner
2010-10-14 0:06 ` Christoph Hellwig
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20101013233647.GA18691@infradead.org \
--to=hch@infradead.org \
--cc=axboe@kernel.dk \
--cc=david@fromorbit.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).