All of lore.kernel.org
 help / color / mirror / Atom feed
From: Nikolay Borisov <nborisov@suse.com>
To: Liu Bo <bo.liu@linux.alibaba.com>, linux-btrfs@vger.kernel.org
Subject: Re: [PATCH 3/6] Btrfs: move get root of btrfs_search_slot to a helper
Date: Wed, 16 May 2018 09:54:58 +0300	[thread overview]
Message-ID: <aef6bb75-af61-3f74-3059-d7900bed1e28@suse.com> (raw)
In-Reply-To: <1526406728-109055-4-git-send-email-bo.liu@linux.alibaba.com>



On 15.05.2018 20:52, Liu Bo wrote:
> It's good to have a helper instead of having all get-root details
> open-coded.
> 
> The new helper locks (if necessary) and sets root node of the path.
> 
> There is no functional change in this commit.
> 
> Signed-off-by: Liu Bo <bo.liu@linux.alibaba.com>

Codewise it looks ok, you've inverted the conditional so as to give a
more linear flow of the code. I've checked all the various branches and
they seem semantically identical to the open coded version. One minor
nit is that I don't think this is the most suitable name for this
function but I don't really have a better suggestions which would be
more specific.

For the code (the changelog should be more explicit):

Reviewed-by: Nikolay Borisov <nborisov@suse.com>

> ---
>  fs/btrfs/ctree.c | 112 +++++++++++++++++++++++++++++++++----------------------
>  1 file changed, 67 insertions(+), 45 deletions(-)
> 
> diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
> index a96d308c51b8..399839df7a8f 100644
> --- a/fs/btrfs/ctree.c
> +++ b/fs/btrfs/ctree.c
> @@ -2598,6 +2598,72 @@ int btrfs_find_item(struct btrfs_root *fs_root, struct btrfs_path *path,
>  	return 0;
>  }
>  
> +static struct extent_buffer *btrfs_search_slot_get_root(struct btrfs_root *root,
> +							struct btrfs_path *p,
> +							int write_lock_level)
> +{
> +	struct btrfs_fs_info *fs_info = root->fs_info;
> +	struct extent_buffer *b;
> +	int root_lock;
> +	int level = 0;
> +
> +	/*
> +	 * we try very hard to do read locks on the root
> +	 */
> +	root_lock = BTRFS_READ_LOCK;
> +
> +	if (p->search_commit_root) {
> +		/*
> +		 * the commit roots are read only so we always do read locks
> +		 */
> +		if (p->need_commit_sem)
> +			down_read(&fs_info->commit_root_sem);
> +		b = root->commit_root;
> +		extent_buffer_get(b);
> +		level = btrfs_header_level(b);
> +		if (p->need_commit_sem)
> +			up_read(&fs_info->commit_root_sem);
> +		if (!p->skip_locking)
> +			btrfs_tree_read_lock(b);
> +
> +		goto out;
> +	}
> +
> +	if (p->skip_locking) {
> +		b = btrfs_root_node(root);
> +		level = btrfs_header_level(b);
> +		goto out;
> +	}
> +
> +	/*
> +	 * we don't know the level of the root node until we actually
> +	 * have it read locked
> +	 */
> +	b = btrfs_read_lock_root_node(root);
> +	level = btrfs_header_level(b);
> +	if (level > write_lock_level)
> +		goto out;
> +
> +	/*
> +	 * whoops, must trade for write lock
> +	 */
> +	btrfs_tree_read_unlock(b);
> +	free_extent_buffer(b);
> +	b = btrfs_lock_root_node(root);
> +	root_lock = BTRFS_WRITE_LOCK;
> +	/*
> +	 * the level might have changed, check again
> +	 */
> +	level = btrfs_header_level(b);
> +
> +out:
> +	p->nodes[level] = b;
> +	if (!p->skip_locking)
> +		p->locks[level] = root_lock;
> +	return b;
> +}
> +
> +
>  /*
>   * btrfs_search_slot - look for a key in a tree and perform necessary
>   * modifications to preserve tree invariants.
> @@ -2634,7 +2700,6 @@ int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root *root,
>  	int err;
>  	int level;
>  	int lowest_unlock = 1;
> -	int root_lock;
>  	/* everything at write_lock_level or lower must be write locked */
>  	int write_lock_level = 0;
>  	u8 lowest_level = 0;
> @@ -2672,50 +2737,7 @@ int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root *root,
>  
>  again:
>  	prev_cmp = -1;
> -	/*
> -	 * we try very hard to do read locks on the root
> -	 */
> -	root_lock = BTRFS_READ_LOCK;
> -	level = 0;
> -	if (p->search_commit_root) {
> -		/*
> -		 * the commit roots are read only
> -		 * so we always do read locks
> -		 */
> -		if (p->need_commit_sem)
> -			down_read(&fs_info->commit_root_sem);
> -		b = root->commit_root;
> -		extent_buffer_get(b);
> -		level = btrfs_header_level(b);
> -		if (p->need_commit_sem)
> -			up_read(&fs_info->commit_root_sem);
> -		if (!p->skip_locking)
> -			btrfs_tree_read_lock(b);
> -	} else {
> -		if (p->skip_locking) {
> -			b = btrfs_root_node(root);
> -			level = btrfs_header_level(b);
> -		} else {
> -			/* we don't know the level of the root node
> -			 * until we actually have it read locked
> -			 */
> -			b = btrfs_read_lock_root_node(root);
> -			level = btrfs_header_level(b);
> -			if (level <= write_lock_level) {
> -				/* whoops, must trade for write lock */
> -				btrfs_tree_read_unlock(b);
> -				free_extent_buffer(b);
> -				b = btrfs_lock_root_node(root);
> -				root_lock = BTRFS_WRITE_LOCK;
> -
> -				/* the level might have changed, check again */
> -				level = btrfs_header_level(b);
> -			}
> -		}
> -	}
> -	p->nodes[level] = b;
> -	if (!p->skip_locking)
> -		p->locks[level] = root_lock;
> +	b = btrfs_search_slot_get_root(root, p, write_lock_level);
>  
>  	while (b) {
>  		level = btrfs_header_level(b);
> 

  reply	other threads:[~2018-05-16  6:55 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-05-15 17:52 [PATCH 0/6] btrfs_search_slot cleanups Liu Bo
2018-05-15 17:52 ` [PATCH 1/6] Btrfs: remove superfluous free_extent_buffer Liu Bo
2018-05-16  6:40   ` Nikolay Borisov
2018-05-15 17:52 ` [PATCH 2/6] Btrfs: use more straightforward extent_buffer_uptodate Liu Bo
2018-05-16  6:43   ` Nikolay Borisov
2018-05-18  2:09     ` Liu Bo
2018-05-15 17:52 ` [PATCH 3/6] Btrfs: move get root of btrfs_search_slot to a helper Liu Bo
2018-05-16  6:54   ` Nikolay Borisov [this message]
2018-05-15 17:52 ` [PATCH 4/6] Btrfs: remove unused check of skip_locking Liu Bo
2018-05-16  7:03   ` Nikolay Borisov
2018-05-16  7:06     ` Qu Wenruo
2018-05-18  2:55     ` Liu Bo
2018-05-18  5:36       ` Nikolay Borisov
2018-05-15 17:52 ` [PATCH 5/6] Btrfs: grab write lock directly if write_lock_level is the max level Liu Bo
2018-05-16  7:09   ` Nikolay Borisov
2018-05-15 17:52 ` [PATCH 6/6] Btrfs: remove always true check in unlock_up Liu Bo
2018-05-16  6:47   ` Nikolay Borisov
2018-05-15 18:15 ` [PATCH 0/6] btrfs_search_slot cleanups 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=aef6bb75-af61-3f74-3059-d7900bed1e28@suse.com \
    --to=nborisov@suse.com \
    --cc=bo.liu@linux.alibaba.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 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.