btrfs: merge ordered work callbacks in btrfs_work into one

There are two callbacks defined in btrfs_work but only two actually make
use of them, otherwise there are NULLs. We can get rid of the freeing
callback making it a special case of the normal work. This reduces the
size of btrfs_work by 8 bytes, final layout:

struct btrfs_work {
        btrfs_func_t               func;                 /*     0     8 */
        btrfs_ordered_func_t       ordered_func;         /*     8     8 */
        struct work_struct         normal_work;          /*    16    32 */
        struct list_head           ordered_list;         /*    48    16 */
        /* --- cacheline 1 boundary (64 bytes) --- */
        struct btrfs_workqueue *   wq;                   /*    64     8 */
        long unsigned int          flags;                /*    72     8 */

        /* size: 80, cachelines: 2, members: 6 */
        /* last cacheline: 16 bytes */
};

This in turn reduces size of other structures (on a release config):

- async_chunk			 160 ->  152
- async_submit_bio		 152 ->  144
- btrfs_async_delayed_work	 104 ->   96
- btrfs_caching_control		 176 ->  168
- btrfs_delalloc_work		 144 ->  136
- btrfs_fs_info			3608 -> 3600
- btrfs_ordered_extent		 440 ->  424
- btrfs_writepage_fixup		 104 ->   96

Signed-off-by: David Sterba <dsterba@suse.com>

diff --git a/fs/btrfs/async-thread.c b/fs/btrfs/async-thread.c
index 714ca74b66bf98650a2b65affd819265f73883dc..9e261aac671e629c8ed6d3ceb85ce9aa416101e4 100644 (file)
--- a/fs/btrfs/async-thread.c
+++ b/fs/btrfs/async-thread.c
@@ -243,7 +243,7 @@ static void run_ordered_work(struct btrfs_workqueue *wq,
                        break;
                trace_btrfs_ordered_sched(work);
                spin_unlock_irqrestore(lock, flags);
-               work->ordered_func(work);
+               work->ordered_func(work, false);
 
                /* now take the lock again and drop our item from the list */
                spin_lock_irqsave(lock, flags);
@@ -278,7 +278,7 @@ static void run_ordered_work(struct btrfs_workqueue *wq,
                         * We don't want to call the ordered free functions with
                         * the lock held.
                         */
-                       work->ordered_free(work);
+                       work->ordered_func(work, true);
                        /* NB: work must not be dereferenced past this point. */
                        trace_btrfs_all_work_done(wq->fs_info, work);
                }
@@ -286,7 +286,7 @@ static void run_ordered_work(struct btrfs_workqueue *wq,
        spin_unlock_irqrestore(lock, flags);
 
        if (free_self) {
-               self->ordered_free(self);
+               self->ordered_func(self, true);
                /* NB: self must not be dereferenced past this point. */
                trace_btrfs_all_work_done(wq->fs_info, self);
        }
@@ -301,7 +301,7 @@ static void btrfs_work_helper(struct work_struct *normal_work)
 
        /*
         * We should not touch things inside work in the following cases:
-        * 1) after work->func() if it has no ordered_free
+        * 1) after work->func() if it has no ordered_func(..., true) to free
         *    Since the struct is freed in work->func().
         * 2) after setting WORK_DONE_BIT
         *    The work may be freed in other threads almost instantly.
@@ -330,11 +330,10 @@ static void btrfs_work_helper(struct work_struct *normal_work)
 }
 
 void btrfs_init_work(struct btrfs_work *work, btrfs_func_t func,
-                    btrfs_func_t ordered_func, btrfs_func_t ordered_free)
+                    btrfs_ordered_func_t ordered_func)
 {
        work->func = func;
        work->ordered_func = ordered_func;
-       work->ordered_free = ordered_free;
        INIT_WORK(&work->normal_work, btrfs_work_helper);
        INIT_LIST_HEAD(&work->ordered_list);
        work->flags = 0;

diff --git a/fs/btrfs/async-thread.h b/fs/btrfs/async-thread.h
index 30f66c5e2e6e4194b11cc800a3b24fabbe3d3f6c..62b8a0d5789865378ac1cce504296eb13a3feb30 100644 (file)
--- a/fs/btrfs/async-thread.h
+++ b/fs/btrfs/async-thread.h
@@ -13,11 +13,11 @@ struct btrfs_fs_info;
 struct btrfs_workqueue;
 struct btrfs_work;
 typedef void (*btrfs_func_t)(struct btrfs_work *arg);
+typedef void (*btrfs_ordered_func_t)(struct btrfs_work *arg, bool);
 
 struct btrfs_work {
        btrfs_func_t func;
-       btrfs_func_t ordered_func;
-       btrfs_func_t ordered_free;
+       btrfs_ordered_func_t ordered_func;
 
        /* Don't touch things below */
        struct work_struct normal_work;
@@ -35,7 +35,7 @@ struct btrfs_workqueue *btrfs_alloc_ordered_workqueue(
                                struct btrfs_fs_info *fs_info, const char *name,
                                unsigned int flags);
 void btrfs_init_work(struct btrfs_work *work, btrfs_func_t func,
-                    btrfs_func_t ordered_func, btrfs_func_t ordered_free);
+                    btrfs_ordered_func_t ordered_func);
 void btrfs_queue_work(struct btrfs_workqueue *wq,
                      struct btrfs_work *work);
 void btrfs_destroy_workqueue(struct btrfs_workqueue *wq);

diff --git a/fs/btrfs/bio.c b/fs/btrfs/bio.c
index ab7ee090b6067955405fa4e447a788b06202bd88..4f3b693a16b137e3ed40cc1ca0e7dbe4112cc66b 100644 (file)
--- a/fs/btrfs/bio.c
+++ b/fs/btrfs/bio.c
@@ -574,13 +574,20 @@ static void run_one_async_start(struct btrfs_work *work)
  *
  * At IO completion time the csums attached on the ordered extent record are
  * inserted into the tree.
+ *
+ * If called with @do_free == true, then it will free the work struct.
  */
-static void run_one_async_done(struct btrfs_work *work)
+static void run_one_async_done(struct btrfs_work *work, bool do_free)
 {
        struct async_submit_bio *async =
                container_of(work, struct async_submit_bio, work);
        struct bio *bio = &async->bbio->bio;
 
+       if (do_free) {
+               kfree(container_of(work, struct async_submit_bio, work));
+               return;
+       }
+
        /* If an error occurred we just want to clean up the bio and move on. */
        if (bio->bi_status) {
                btrfs_orig_bbio_end_io(async->bbio);
@@ -596,11 +603,6 @@ static void run_one_async_done(struct btrfs_work *work)
        __btrfs_submit_bio(bio, async->bioc, &async->smap, async->mirror_num);
 }
 
-static void run_one_async_free(struct btrfs_work *work)
-{
-       kfree(container_of(work, struct async_submit_bio, work));
-}
-
 static bool should_async_write(struct btrfs_bio *bbio)
 {
        /* Submit synchronously if the checksum implementation is fast. */
@@ -642,8 +644,7 @@ static bool btrfs_wq_submit_bio(struct btrfs_bio *bbio,
        async->smap = *smap;
        async->mirror_num = mirror_num;
 
-       btrfs_init_work(&async->work, run_one_async_start, run_one_async_done,
-                       run_one_async_free);
+       btrfs_init_work(&async->work, run_one_async_start, run_one_async_done);
        btrfs_queue_work(fs_info->workers, &async->work);
        return true;
 }

diff --git a/fs/btrfs/block-group.c b/fs/btrfs/block-group.c
index 72dbfb410e42ed190adf75f003ba9b6afb5b2485..2b9aaeefaf76be3f426ba25e5fb75eb1a4ffa528 100644 (file)
--- a/fs/btrfs/block-group.c
+++ b/fs/btrfs/block-group.c
@@ -935,7 +935,7 @@ int btrfs_cache_block_group(struct btrfs_block_group *cache, bool wait)
        caching_ctl->block_group = cache;
        refcount_set(&caching_ctl->count, 2);
        atomic_set(&caching_ctl->progress, 0);
-       btrfs_init_work(&caching_ctl->work, caching_thread, NULL, NULL);
+       btrfs_init_work(&caching_ctl->work, caching_thread, NULL);
 
        spin_lock(&cache->lock);
        if (cache->cached != BTRFS_CACHE_NO) {

diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c
index 23770db1e109d3c2a5099b5b6bb3c3d53980e21e..fc18a53d6584242cd6cbcc5882abc5a31f3b2e1d 100644 (file)
--- a/fs/btrfs/delayed-inode.c
+++ b/fs/btrfs/delayed-inode.c
@@ -1379,8 +1379,7 @@ static int btrfs_wq_run_delayed_node(struct btrfs_delayed_root *delayed_root,
                return -ENOMEM;
 
        async_work->delayed_root = delayed_root;
-       btrfs_init_work(&async_work->work, btrfs_async_run_delayed_root, NULL,
-                       NULL);
+       btrfs_init_work(&async_work->work, btrfs_async_run_delayed_root, NULL);
        async_work->nr = nr;
 
        btrfs_queue_work(fs_info->delayed_workers, &async_work->work);

diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index b5e0ed3a36f78d2f54165a0e3e80798ab5daabf2..514d2e8a4f527dd7efa8af605f5ae6f7f53eefb9 100644 (file)
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -1566,8 +1566,11 @@ out_unlock:
  * Phase two of compressed writeback.  This is the ordered portion of the code,
  * which only gets called in the order the work was queued.  We walk all the
  * async extents created by compress_file_range and send them down to the disk.
+ *
+ * If called with @do_free == true then it'll try to finish the work and free
+ * the work struct eventually.
  */
-static noinline void submit_compressed_extents(struct btrfs_work *work)
+static noinline void submit_compressed_extents(struct btrfs_work *work, bool do_free)
 {
        struct async_chunk *async_chunk = container_of(work, struct async_chunk,
                                                     work);
@@ -1576,6 +1579,21 @@ static noinline void submit_compressed_extents(struct btrfs_work *work)
        unsigned long nr_pages;
        u64 alloc_hint = 0;
 
+       if (do_free) {
+               struct async_chunk *async_chunk;
+               struct async_cow *async_cow;
+
+               async_chunk = container_of(work, struct async_chunk, work);
+               btrfs_add_delayed_iput(async_chunk->inode);
+               if (async_chunk->blkcg_css)
+                       css_put(async_chunk->blkcg_css);
+
+               async_cow = async_chunk->async_cow;
+               if (atomic_dec_and_test(&async_cow->num_chunks))
+                       kvfree(async_cow);
+               return;
+       }
+
        nr_pages = (async_chunk->end - async_chunk->start + PAGE_SIZE) >>
                PAGE_SHIFT;
 
@@ -1592,21 +1610,6 @@ static noinline void submit_compressed_extents(struct btrfs_work *work)
                cond_wake_up_nomb(&fs_info->async_submit_wait);
 }
 
-static noinline void async_cow_free(struct btrfs_work *work)
-{
-       struct async_chunk *async_chunk;
-       struct async_cow *async_cow;
-
-       async_chunk = container_of(work, struct async_chunk, work);
-       btrfs_add_delayed_iput(async_chunk->inode);
-       if (async_chunk->blkcg_css)
-               css_put(async_chunk->blkcg_css);
-
-       async_cow = async_chunk->async_cow;
-       if (atomic_dec_and_test(&async_cow->num_chunks))
-               kvfree(async_cow);
-}
-
 static bool run_delalloc_compressed(struct btrfs_inode *inode,
                                    struct page *locked_page, u64 start,
                                    u64 end, struct writeback_control *wbc)
@@ -1684,7 +1687,7 @@ static bool run_delalloc_compressed(struct btrfs_inode *inode,
                }
 
                btrfs_init_work(&async_chunk[i].work, compress_file_range,
-                               submit_compressed_extents, async_cow_free);
+                               submit_compressed_extents);
 
                nr_pages = DIV_ROUND_UP(cur_end - start, PAGE_SIZE);
                atomic_add(nr_pages, &fs_info->async_delalloc_pages);
@@ -2848,7 +2851,7 @@ int btrfs_writepage_cow_fixup(struct page *page)
        ihold(inode);
        btrfs_page_set_checked(fs_info, page, page_offset(page), PAGE_SIZE);
        get_page(page);
-       btrfs_init_work(&fixup->work, btrfs_writepage_fixup_worker, NULL, NULL);
+       btrfs_init_work(&fixup->work, btrfs_writepage_fixup_worker, NULL);
        fixup->page = page;
        fixup->inode = BTRFS_I(inode);
        btrfs_queue_work(fs_info->fixup_workers, &fixup->work);
@@ -9215,7 +9218,7 @@ static struct btrfs_delalloc_work *btrfs_alloc_delalloc_work(struct inode *inode
        init_completion(&work->completion);
        INIT_LIST_HEAD(&work->list);
        work->inode = inode;
-       btrfs_init_work(&work->work, btrfs_run_delalloc_work, NULL, NULL);
+       btrfs_init_work(&work->work, btrfs_run_delalloc_work, NULL);
 
        return work;
 }

diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c
index 55c7d5543265c4df8830018b89d3bbabe835058c..b133ea0bc45925a78fc74d73a2eededa5baf352b 100644 (file)
--- a/fs/btrfs/ordered-data.c
+++ b/fs/btrfs/ordered-data.c
@@ -365,7 +365,7 @@ static void btrfs_queue_ordered_fn(struct btrfs_ordered_extent *ordered)
        struct btrfs_workqueue *wq = btrfs_is_free_space_inode(inode) ?
                fs_info->endio_freespace_worker : fs_info->endio_write_workers;
 
-       btrfs_init_work(&ordered->work, finish_ordered_fn, NULL, NULL);
+       btrfs_init_work(&ordered->work, finish_ordered_fn, NULL);
        btrfs_queue_work(wq, &ordered->work);
 }
 
@@ -712,7 +712,7 @@ u64 btrfs_wait_ordered_extents(struct btrfs_root *root, u64 nr,
                spin_unlock(&root->ordered_extent_lock);
 
                btrfs_init_work(&ordered->flush_work,
-                               btrfs_run_ordered_extent_work, NULL, NULL);
+                               btrfs_run_ordered_extent_work, NULL);
                list_add_tail(&ordered->work_list, &works);
                btrfs_queue_work(fs_info->flush_workers, &ordered->flush_work);
 

diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c
index ae7e2d79209fe177e704fe28e25f8d6c55e01ab9..07716332603b55e5b7f045c2d399222163ad75b5 100644 (file)
--- a/fs/btrfs/qgroup.c
+++ b/fs/btrfs/qgroup.c
@@ -3583,7 +3583,7 @@ qgroup_rescan_init(struct btrfs_fs_info *fs_info, u64 progress_objectid,
        mutex_unlock(&fs_info->qgroup_rescan_lock);
 
        btrfs_init_work(&fs_info->qgroup_rescan_work,
-                       btrfs_qgroup_rescan_worker, NULL, NULL);
+                       btrfs_qgroup_rescan_worker, NULL);
        return 0;
 }
 

diff --git a/include/trace/events/btrfs.h b/include/trace/events/btrfs.h
index 638412ab090d2a00ff87a707dc225263cf8fd4fb..279a7a0c90c09cecb0aefcb71509df5d424e5b20 100644 (file)
--- a/include/trace/events/btrfs.h
+++ b/include/trace/events/btrfs.h
@@ -1561,7 +1561,6 @@ DECLARE_EVENT_CLASS(btrfs__work,
                __field(        const void *,   wq                      )
                __field(        const void *,   func                    )
                __field(        const void *,   ordered_func            )
-               __field(        const void *,   ordered_free            )
                __field(        const void *,   normal_work             )
        ),
 
@@ -1570,14 +1569,12 @@ DECLARE_EVENT_CLASS(btrfs__work,
                __entry->wq             = work->wq;
                __entry->func           = work->func;
                __entry->ordered_func   = work->ordered_func;
-               __entry->ordered_free   = work->ordered_free;
                __entry->normal_work    = &work->normal_work;
        ),
 
-       TP_printk_btrfs("work=%p (normal_work=%p) wq=%p func=%ps ordered_func=%p "
-                 "ordered_free=%p",
+       TP_printk_btrfs("work=%p (normal_work=%p) wq=%p func=%ps ordered_func=%p",
                  __entry->work, __entry->normal_work, __entry->wq,
-                  __entry->func, __entry->ordered_func, __entry->ordered_free)
+                  __entry->func, __entry->ordered_func)
 );
 
 /*