linux-btrfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Liu Bo <bo.li.liu@oracle.com>
To: Filipe Manana <fdmanana@kernel.org>
Cc: "linux-btrfs@vger.kernel.org" <linux-btrfs@vger.kernel.org>
Subject: Re: [PATCH] Btrfs: try harder to migrate items to left sibling before splitting a leaf
Date: Wed, 22 Feb 2017 09:37:23 -0800	[thread overview]
Message-ID: <20170222173723.GA6018@localhost.localdomain> (raw)
In-Reply-To: <CAL3q7H6H3zbvLPp0-7ttytM0yNCjz==Ra=K=s5wq6ThPGc7JVQ@mail.gmail.com>

On Wed, Feb 22, 2017 at 02:05:05PM +0000, Filipe Manana wrote:
> On Wed, Feb 22, 2017 at 12:07 AM, Liu Bo <bo.li.liu@oracle.com> wrote:
> > On Sun, Feb 19, 2017 at 08:56:39PM +0000, fdmanana@kernel.org wrote:
> >> From: Filipe Manana <fdmanana@suse.com>
> >>
> >> Before attempting to split a leaf we try to migrate items from the leaf to
> >> its right and left siblings. We start by trying to move items into the
> >> rigth sibling and, if the new item is meant to be inserted at the end of
> >> our leaf, we try to free from our leaf an amount of bytes equal to the
> >> number of bytes used by the new item, by setting the variable space_needed
> >> to the byte size of that new item. However if we fail to move enough items
> >> to the right sibling due to lack of space in that sibling, we then try
> >> to move items into the left sibling, and in that case we try to free
> >> an amount equal to the size of the new item from our leaf, when we need
> >> only to free an amount corresponding to the size of the new item minus
> >> the current free space of our leaf. So make sure that before we try to
> >> move items to the left sibling we do set the variable space_needed with
> >> a value corresponding to the new item's size minus the leaf's current
> >> free space.
> >>
> >> Signed-off-by: Filipe Manana <fdmanana@suse.com>
> >> ---
> >>  fs/btrfs/ctree.c | 7 +++++++
> >>  1 file changed, 7 insertions(+)
> >>
> >> diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
> >> index a426dc8..1d66761 100644
> >> --- a/fs/btrfs/ctree.c
> >> +++ b/fs/btrfs/ctree.c
> >> @@ -4160,6 +4160,9 @@ static noinline int push_for_double_split(struct btrfs_trans_handle *trans,
> >>
> >>       /* try to push all the items before our slot into the next leaf */
> >>       slot = path->slots[0];
> >> +     space_needed = data_size;
> >> +     if (slot > 0)
> >> +             space_needed -= btrfs_leaf_free_space(fs_info, path->nodes[0]);
> >
> > Good point.
> >
> >>       ret = push_leaf_left(trans, root, path, 1, space_needed, 0, slot);
> >>       if (ret < 0)
> >>               return ret;
> >> @@ -4215,6 +4218,10 @@ static noinline int split_leaf(struct btrfs_trans_handle *trans,
> >>               if (wret < 0)
> >>                       return wret;
> >>               if (wret) {
> >> +                     space_needed = data_size;
> >> +                     if (slot > 0)
> >> +                             space_needed -= btrfs_leaf_free_space(fs_info,
> >> +                                                                   l);
> >
> > Not sure if we need this, the above push_leaf_right() was called with
> > @min_data_size == space_needed, thus if @wret == 1, no items have been moved in
> > push_leaf_right() so that leaf 'l' remains unchanged.
> 
> That's exactly the point. If the item is meant to be appended and we
> failed to move it into the right sibling, we want to try to move items
> from the leaf into its left sibling with the goal of freeing
> "new_item_size - leaf_free_space(l)" bytes from our leaf. Without this
> new assignment, we try to free new_item_size bytes, which has higher
> chances of not being successful at migrating items into the left
> sibling (example, the left sibling has less than new_item_size bytes
> free, but has at least "new_item_size - leaf_free_space(l)" bytes
> free).
> 

I see, I missed that we didn't check (slot > 0) before
push_leaf_right(), you may add

Reviewed-by: Liu Bo <bo.li.liu@oracle.com>

Thanks,

-liubo
> >
> > Thanks,
> >
> > -liubo
> >>                       wret = push_leaf_left(trans, root, path, space_needed,
> >>                                             space_needed, 0, (u32)-1);
> >>                       if (wret < 0)
> >> --
> >> 2.7.0.rc3
> >>
> >> --
> >> To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
> >> the body of a message to majordomo@vger.kernel.org
> >> More majordomo info at  http://vger.kernel.org/majordomo-info.html

      reply	other threads:[~2017-02-22 17:38 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-02-19 20:56 [PATCH] Btrfs: try harder to migrate items to left sibling before splitting a leaf fdmanana
2017-02-22  0:07 ` Liu Bo
2017-02-22 14:05   ` Filipe Manana
2017-02-22 17:37     ` Liu Bo [this message]

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=20170222173723.GA6018@localhost.localdomain \
    --to=bo.li.liu@oracle.com \
    --cc=fdmanana@kernel.org \
    --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).