From: Nikolay Borisov <nborisov@suse.com>
To: Josef Bacik <josef@toxicpanda.com>
Cc: kernel-team@fb.com, linux-btrfs@vger.kernel.org
Subject: Re: [PATCH 01/21] Btrfs: rework outstanding_extents
Date: Fri, 13 Oct 2017 16:55:58 +0300 [thread overview]
Message-ID: <9dcaa11c-679c-0fda-4e69-327cf85d1ba5@suse.com> (raw)
In-Reply-To: <20171013131050.q5huwryq444utast@destiny>
>
> The outstanding_extents accounting is consistent with only the items needed to
> handle the outstanding extent items. However since changing the inode requires
> updating the inode item as well we have to keep this floating reservation for
> the inode item until we have 0 outstanding extents. The way we do this is with
> the BTRFS_INODE_DELALLOC_META_RESERVED flag. So if it isn't set we will
> allocate nr_exntents + 1 in btrfs_delalloc_reserve_metadata() and then set our
> bit. If we ever steal this reservation we make sure to clear the flag so we
> know we don't have to clean it up when outstanding_extents goes to 0. It's not
> super intuitive but needs to be done under the BTRFS_I(inode)->lock so this was
> the best place to put it. I suppose we could move the logic out of here and put
> it somewhere else to make it more clear.
I think defining this logic in its own, discrete block of code would be
best w.r.t readibility. It's not super obvious.
I'm slowly going through your patchkit so expect more question but
otherwise the delalloc stuff after this and patch 03 really start
looking a lot more obvious !
next prev parent reply other threads:[~2017-10-13 13:56 UTC|newest]
Thread overview: 51+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-09-29 19:43 [PATCH 00/21] My current btrfs patch queue Josef Bacik
2017-09-29 19:43 ` [PATCH 01/21] Btrfs: rework outstanding_extents Josef Bacik
2017-10-13 8:39 ` Nikolay Borisov
2017-10-13 13:10 ` Josef Bacik
2017-10-13 13:33 ` David Sterba
2017-10-13 13:55 ` Nikolay Borisov [this message]
2017-10-19 18:10 ` Josef Bacik
2017-10-19 3:14 ` Edmund Nadolski
2017-09-29 19:43 ` [PATCH 02/21] btrfs: add tracepoints for outstanding extents mods Josef Bacik
2017-09-29 19:43 ` [PATCH 03/21] btrfs: make the delalloc block rsv per inode Josef Bacik
2017-10-13 11:47 ` Nikolay Borisov
2017-10-13 13:18 ` Josef Bacik
2017-09-29 19:43 ` [PATCH 04/21] btrfs: add ref-verify mount option Josef Bacik
2017-10-13 13:53 ` David Sterba
2017-10-13 13:57 ` David Sterba
2017-09-29 19:43 ` [PATCH 05/21] btrfs: pass root to various extent ref mod functions Josef Bacik
2017-10-13 14:01 ` David Sterba
2017-09-29 19:43 ` [PATCH 06/21] Btrfs: add a extent ref verify tool Josef Bacik
2017-10-13 14:23 ` David Sterba
2017-09-29 19:43 ` [PATCH 07/21] Btrfs: only check delayed ref usage in should_end_transaction Josef Bacik
2017-10-13 17:20 ` David Sterba
2017-09-29 19:43 ` [PATCH 08/21] btrfs: add a helper to return a head ref Josef Bacik
2017-10-13 14:39 ` David Sterba
2017-09-29 19:43 ` [PATCH 09/21] btrfs: move extent_op cleanup to a helper Josef Bacik
2017-10-13 14:50 ` David Sterba
2017-10-16 14:05 ` Nikolay Borisov
2017-10-16 15:02 ` David Sterba
2017-09-29 19:43 ` [PATCH 10/21] btrfs: breakout empty head " Josef Bacik
2017-10-13 14:57 ` David Sterba
2017-10-16 14:07 ` Nikolay Borisov
2017-10-16 14:55 ` David Sterba
2017-09-29 19:43 ` [PATCH 11/21] btrfs: move ref_mod modification into the if (ref) logic Josef Bacik
2017-10-13 15:05 ` David Sterba
2017-09-29 19:43 ` [PATCH 12/21] btrfs: move all ref head cleanup to the helper function Josef Bacik
2017-10-13 15:39 ` David Sterba
2017-09-29 19:43 ` [PATCH 13/21] btrfs: remove delayed_ref_node from ref_head Josef Bacik
2017-10-13 16:05 ` David Sterba
2017-10-16 14:41 ` Nikolay Borisov
2017-09-29 19:43 ` [PATCH 14/21] btrfs: remove type argument from comp_tree_refs Josef Bacik
2017-10-13 16:06 ` David Sterba
2017-09-29 19:43 ` [PATCH 15/21] btrfs: switch args for comp_*_refs Josef Bacik
2017-10-13 16:24 ` David Sterba
2017-09-29 19:44 ` [PATCH 16/21] btrfs: add a comp_refs() helper Josef Bacik
2017-09-29 19:44 ` [PATCH 17/21] btrfs: track refs in a rb_tree instead of a list Josef Bacik
2017-09-29 19:44 ` [PATCH 18/21] btrfs: fix send ioctl on 32bit with 64bit kernel Josef Bacik
2017-09-29 19:44 ` [PATCH 19/21] btrfs: don't call btrfs_start_delalloc_roots in flushoncommit Josef Bacik
2017-10-13 17:10 ` David Sterba
2017-09-29 19:44 ` [PATCH 20/21] btrfs: move btrfs_truncate_block out of trans handle Josef Bacik
2017-09-29 19:44 ` [PATCH 21/21] btrfs: add assertions for releasing trans handle reservations Josef Bacik
2017-10-13 17:17 ` David Sterba
2017-10-13 17:28 ` [PATCH 00/21] My current btrfs patch queue David Sterba
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=9dcaa11c-679c-0fda-4e69-327cf85d1ba5@suse.com \
--to=nborisov@suse.com \
--cc=josef@toxicpanda.com \
--cc=kernel-team@fb.com \
--cc=linux-btrfs@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).