From: Zheng Liu <gnehzuil.liu@gmail.com>
To: Theodore Ts'o <tytso@mit.edu>
Cc: linux-ext4@vger.kernel.org, linux-fsdevel@vger.kernel.org,
Yongqiang Yang <xiaoqiangnk@gmail.com>,
Allison Henderson <achender@linux.vnet.ibm.com>,
Zheng Liu <wenqing.lz@taobao.com>
Subject: Re: [RFC][PATCH 3/8 v2] ext4: initialize extent status tree
Date: Wed, 26 Sep 2012 10:09:55 +0800 [thread overview]
Message-ID: <20120926020955.GA4101@gmail.com> (raw)
In-Reply-To: <20120925205921.GA8625@thunk.org>
On Tue, Sep 25, 2012 at 04:59:21PM -0400, Theodore Ts'o wrote:
> On Tue, Sep 25, 2012 at 08:42:52PM +0800, Zheng Liu wrote:
> > > If so, we might want to think about adding a sanity check to make sure
> > > that by the time we are done with the inode in ext4_evict_inode()
> > > (after we have forced writeback), the ext4_es_tree is empty. Agreed?
> >
> > Today I revise this patch again, and I find extent_status_tree is freed
> > in ext4_clear_inode(). So maybe I don't think that we need to check
> > this tree to be freed in ext4_evict_inode(). This change is in this
> > patch '[RFC][PATCH 4/8 v2] ext4: let ext4 maintain extent status tree'.
> > What's your opinion?
>
> When you say "revise this patch again", does that mean that you would
> like to submit a new set of patch series with changes? Or just that
> you are looking at this patch set again?
Yes, I prepare to submit a new patch set.
>
> It's certainly true that ext4_evict_inode() will call
> ext4_clear_inode(), so it's not a question of worrying about a memory
> leak. I was thinking more about doing this as a cheap sanity check
> for the data structure. By the time we call ext4_evict_inode(), the
> mm layer all writeback should be complete. Hence, all of the entries
> to the tree _should_ have been removed by the time we call
> ext4_evict_inode().
>
> I don't know if this is going to change as you start using this data
> structure for other purposes (such as locking a range of pages), but
> if I understand how things are currently working, it _should_ be the
> case that when ext4_evict_inode() calls ext4_clear_inode(), the call
> to ext4_es_remove_extent() should be a no-op, since all of the nodes
> in the extent status tree should have been released by then. If it
> isn't, then either I'm not understanding the code, or there's a bug in
> the code.
Yes, you are right. In currently implementation, extent status tree
only maintains the status of delay extents. So in ext4_evict_inode()
extent status tree should be an empty tree. I will add a sanity check
to ensure it. Thanks for your explanation.
Regards,
Zheng
next prev parent reply other threads:[~2012-09-26 2:09 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-08-22 6:05 [RFC][PATCH 0/8 v2] ext4: extent status tree (step 1) Zheng Liu
2012-08-22 6:05 ` [RFC][PATCH 1/8 v2] ext4: add two structures supporting extent status tree Zheng Liu
2012-08-22 6:05 ` [RFC][PATCH 2/8 v2] ext4: add operations on " Zheng Liu
2012-09-19 18:34 ` Theodore Ts'o
2012-09-24 4:25 ` Zheng Liu
2012-09-19 18:41 ` Theodore Ts'o
2012-09-24 3:35 ` Zheng Liu
2012-08-22 6:05 ` [RFC][PATCH 3/8 v2] ext4: initialize " Zheng Liu
2012-09-19 18:53 ` Theodore Ts'o
2012-09-19 19:05 ` Theodore Ts'o
2012-09-24 4:45 ` Zheng Liu
2012-09-25 12:42 ` Zheng Liu
2012-09-25 20:59 ` Theodore Ts'o
2012-09-26 2:09 ` Zheng Liu [this message]
2012-09-26 2:47 ` Theodore Ts'o
2012-09-26 3:24 ` Zheng Liu
2012-09-26 3:37 ` Theodore Ts'o
2012-09-26 3:54 ` Zheng Liu
2012-09-26 3:46 ` Theodore Ts'o
2012-09-26 3:59 ` Zheng Liu
2012-09-28 7:27 ` Zheng Liu
2012-09-28 17:42 ` Theodore Ts'o
[not found] ` <CANWLp02F=qDXY_KbhAziPkBe8qGKV6oN3XtSdc5bWOkbDG684g@mail.gmail.com>
2012-09-29 3:07 ` Theodore Ts'o
2012-09-29 13:26 ` Zheng Liu
2012-09-30 14:00 ` Dmitry Monakhov
2012-09-29 13:24 ` Fwd: " Zheng Liu
2012-09-26 8:00 ` Zheng Liu
2012-08-22 6:05 ` [RFC][PATCH 4/8 v2] ext4: let ext4 maintain " Zheng Liu
2012-08-22 6:05 ` [RFC][PATCH 5/8 v2] ext4: add some tracepoints in " Zheng Liu
2012-08-22 6:05 ` [RFC][PATCH 6/8 v2] ext4: reimplement fiemap on " Zheng Liu
2012-08-22 6:05 ` [RFC][PATCH 7/8 v2] ext4: reimplement ext4_find_delay_alloc_range " Zheng Liu
2012-08-22 6:05 ` [RFC][PATCH 8/8 v2] ext4: introduce lseek SEEK_DATA/SEEK_HOLE support Zheng Liu
2012-09-20 14:41 ` [RFC][PATCH 0/8 v2] ext4: extent status tree (step 1) Theodore Ts'o
2012-09-21 1:51 ` Zheng Liu
2012-09-21 3:19 ` Yongqiang Yang
2012-09-22 0:02 ` Theodore Ts'o
2012-09-24 3:16 ` Zheng Liu
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=20120926020955.GA4101@gmail.com \
--to=gnehzuil.liu@gmail.com \
--cc=achender@linux.vnet.ibm.com \
--cc=linux-ext4@vger.kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=tytso@mit.edu \
--cc=wenqing.lz@taobao.com \
--cc=xiaoqiangnk@gmail.com \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.