[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <415fbefb-1a69-5b80-02c4-aab8b204375c@suse.com>
Date: Tue, 29 Jan 2019 09:54:20 +0200
From: Nikolay Borisov <nborisov@...e.com>
To: Dennis Zhou <dennis@...nel.org>, David Sterba <dsterba@...e.com>,
Josef Bacik <josef@...icpanda.com>, Chris Mason <clm@...com>,
Omar Sandoval <osandov@...ndov.com>,
Nick Terrell <terrelln@...com>
Cc: kernel-team@...com, linux-btrfs@...r.kernel.org,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH 04/11] btrfs: unify compression ops with workspace_manager
On 28.01.19 г. 23:24 ч., Dennis Zhou wrote:
> Make the workspace_manager own the interface operations rather than
> managing index-paired arrays for the workspace_manager and compression
> operations.
>
> Signed-off-by: Dennis Zhou <dennis@...nel.org>
Reviewed-by: Nikolay Borisov <nborisov@...e.com>
> ---
> fs/btrfs/compression.c | 11 +++++++----
> 1 file changed, 7 insertions(+), 4 deletions(-)
>
> diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c
> index bda7e8d2cbc7..b7e986e16640 100644
> --- a/fs/btrfs/compression.c
> +++ b/fs/btrfs/compression.c
> @@ -777,6 +777,7 @@ const struct btrfs_compress_op btrfs_heuristic_compress = {
> };
>
> struct workspace_manager {
> + const struct btrfs_compress_op *ops;
> struct list_head idle_ws;
> spinlock_t ws_lock;
> /* Number of free workspaces */
> @@ -802,6 +803,8 @@ void __init btrfs_init_compress(void)
> int i;
>
> for (i = 0; i < BTRFS_COMPRESS_TYPES; i++) {
> + wsm[i].ops = btrfs_compress_op[i];
> +
> INIT_LIST_HEAD(&wsm[i].idle_ws);
> spin_lock_init(&wsm[i].ws_lock);
> atomic_set(&wsm[i].total_ws, 0);
> @@ -811,7 +814,7 @@ void __init btrfs_init_compress(void)
> * Preallocate one workspace for each compression type so
> * we can guarantee forward progress in the worst case
> */
> - workspace = btrfs_compress_op[i]->alloc_workspace();
> + workspace = wsm[i].ops->alloc_workspace();
> if (IS_ERR(workspace)) {
> pr_warn("BTRFS: cannot preallocate compression workspace, will try later\n");
> } else {
> @@ -874,7 +877,7 @@ static struct list_head *find_workspace(int type)
> * context of btrfs_compress_bio/btrfs_compress_pages
> */
> nofs_flag = memalloc_nofs_save();
> - workspace = btrfs_compress_op[type]->alloc_workspace();
> + workspace = wsm[type].ops->alloc_workspace();
> memalloc_nofs_restore(nofs_flag);
>
> if (IS_ERR(workspace)) {
> @@ -932,7 +935,7 @@ static void free_workspace(int type, struct list_head *workspace)
> }
> spin_unlock(ws_lock);
>
> - btrfs_compress_op[type]->free_workspace(workspace);
> + wsm[type].ops->free_workspace(workspace);
> atomic_dec(total_ws);
> wake:
> cond_wake_up(ws_wait);
> @@ -950,7 +953,7 @@ static void free_workspaces(void)
> while (!list_empty(&wsm[i].idle_ws)) {
> workspace = wsm[i].idle_ws.next;
> list_del(workspace);
> - btrfs_compress_op[i]->free_workspace(workspace);
> + wsm[i].ops->free_workspace(workspace);
> atomic_dec(&wsm[i].total_ws);
> }
> }
>
Powered by blists - more mailing lists