X-Git-Url: http://git.onelab.eu/?a=blobdiff_plain;f=fs%2Fbuffer.c;h=81f31297b81e5c99057737ddf534d89c30f481ae;hb=c7b5ebbddf7bcd3651947760f423e3783bbe6573;hp=033773300a5ad0465a6f92c51d1c9b0bc0fb8f94;hpb=5273a3df6485dc2ad6aa7ddd441b9a21970f003b;p=linux-2.6.git diff --git a/fs/buffer.c b/fs/buffer.c index 033773300..81f31297b 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -39,10 +39,31 @@ #include #include +static int fsync_buffers_list(spinlock_t *lock, struct list_head *list); static void invalidate_bh_lrus(void); #define BH_ENTRY(list) list_entry((list), struct buffer_head, b_assoc_buffers) +struct bh_wait_queue { + struct buffer_head *bh; + wait_queue_t wait; +}; + +#define __DEFINE_BH_WAIT(name, b, f) \ + struct bh_wait_queue name = { \ + .bh = b, \ + .wait = { \ + .task = current, \ + .flags = f, \ + .func = bh_wake_function, \ + .task_list = \ + LIST_HEAD_INIT(name.wait.task_list),\ + }, \ + } +#define DEFINE_BH_WAIT(name, bh) __DEFINE_BH_WAIT(name, bh, 0) +#define DEFINE_BH_WAIT_EXCLUSIVE(name, bh) \ + __DEFINE_BH_WAIT(name, bh, WQ_FLAG_EXCLUSIVE) + /* * Hashed waitqueue_head's for wait_on_buffer() */ @@ -74,10 +95,50 @@ void wake_up_buffer(struct buffer_head *bh) smp_mb(); if (waitqueue_active(wq)) - wake_up_all(wq); + __wake_up(wq, TASK_INTERRUPTIBLE|TASK_UNINTERRUPTIBLE, 1, bh); } EXPORT_SYMBOL(wake_up_buffer); +static int bh_wake_function(wait_queue_t *wait, unsigned mode, + int sync, void *key) +{ + struct buffer_head *bh = key; + struct bh_wait_queue *wq; + + wq = container_of(wait, struct bh_wait_queue, wait); + if (wq->bh != bh || buffer_locked(bh)) + return 0; + else + return autoremove_wake_function(wait, mode, sync, key); +} + +static void sync_buffer(struct buffer_head *bh) +{ + struct block_device *bd; + + smp_mb(); + bd = bh->b_bdev; + if (bd) + blk_run_address_space(bd->bd_inode->i_mapping); +} + +void fastcall __lock_buffer(struct buffer_head *bh) +{ + wait_queue_head_t *wqh = bh_waitq_head(bh); + DEFINE_BH_WAIT_EXCLUSIVE(wait, bh); + + do { + prepare_to_wait_exclusive(wqh, &wait.wait, + TASK_UNINTERRUPTIBLE); + if (buffer_locked(bh)) { + sync_buffer(bh); + io_schedule(); + } + } while (test_set_buffer_locked(bh)); + finish_wait(wqh, &wait.wait); +} +EXPORT_SYMBOL(__lock_buffer); + void fastcall unlock_buffer(struct buffer_head *bh) { clear_buffer_locked(bh); @@ -93,20 +154,16 @@ void fastcall unlock_buffer(struct buffer_head *bh) void __wait_on_buffer(struct buffer_head * bh) { wait_queue_head_t *wqh = bh_waitq_head(bh); - DEFINE_WAIT(wait); + DEFINE_BH_WAIT(wait, bh); do { - prepare_to_wait(wqh, &wait, TASK_UNINTERRUPTIBLE); + prepare_to_wait(wqh, &wait.wait, TASK_UNINTERRUPTIBLE); if (buffer_locked(bh)) { - struct block_device *bd; - smp_mb(); - bd = bh->b_bdev; - if (bd) - blk_run_address_space(bd->bd_inode->i_mapping); + sync_buffer(bh); io_schedule(); } } while (buffer_locked(bh)); - finish_wait(wqh, &wait); + finish_wait(wqh, &wait.wait); } static void @@ -157,7 +214,7 @@ void end_buffer_write_sync(struct buffer_head *bh, int uptodate) if (uptodate) { set_buffer_uptodate(bh); } else { - if (printk_ratelimit()) { + if (!buffer_eopnotsupp(bh) && printk_ratelimit()) { buffer_io_error(bh); printk(KERN_WARNING "lost page write due to " "I/O error on %s\n", @@ -669,12 +726,11 @@ still_busy: * PageLocked prevents anyone from starting writeback of a page which is * under read I/O (PageWriteback is only ever set against a locked page). */ -void mark_buffer_async_read(struct buffer_head *bh) +static void mark_buffer_async_read(struct buffer_head *bh) { bh->b_end_io = end_buffer_async_read; set_buffer_async_read(bh); } -EXPORT_SYMBOL(mark_buffer_async_read); void mark_buffer_async_write(struct buffer_head *bh) { @@ -733,14 +789,6 @@ EXPORT_SYMBOL(mark_buffer_async_write); * b_inode back. */ -void buffer_insert_list(spinlock_t *lock, - struct buffer_head *bh, struct list_head *list) -{ - spin_lock(lock); - list_move_tail(&bh->b_assoc_buffers, list); - spin_unlock(lock); -} - /* * The buffer's backing address_space's private_lock must be held */ @@ -843,9 +891,12 @@ void mark_buffer_dirty_inode(struct buffer_head *bh, struct inode *inode) if (mapping->assoc_mapping != buffer_mapping) BUG(); } - if (list_empty(&bh->b_assoc_buffers)) - buffer_insert_list(&buffer_mapping->private_lock, - bh, &mapping->private_list); + if (list_empty(&bh->b_assoc_buffers)) { + spin_lock(&buffer_mapping->private_lock); + list_move_tail(&bh->b_assoc_buffers, + &mapping->private_list); + spin_unlock(&buffer_mapping->private_lock); + } } EXPORT_SYMBOL(mark_buffer_dirty_inode); @@ -895,7 +946,8 @@ int __set_page_dirty_buffers(struct page *page) if (page->mapping) { /* Race with truncate? */ if (!mapping->backing_dev_info->memory_backed) inc_page_state(nr_dirty); - radix_tree_tag_set(&mapping->page_tree, page->index, + radix_tree_tag_set(&mapping->page_tree, + page_index(page), PAGECACHE_TAG_DIRTY); } spin_unlock_irq(&mapping->tree_lock); @@ -925,7 +977,7 @@ EXPORT_SYMBOL(__set_page_dirty_buffers); * the osync code to catch these locked, dirty buffers without requeuing * any newly dirty buffers for write. */ -int fsync_buffers_list(spinlock_t *lock, struct list_head *list) +static int fsync_buffers_list(spinlock_t *lock, struct list_head *list) { struct buffer_head *bh; struct list_head tmp; @@ -1207,12 +1259,6 @@ grow_buffers(struct block_device *bdev, sector_t block, int size) pgoff_t index; int sizebits; - /* Size must be multiple of hard sectorsize */ - if (size & (bdev_hardsect_size(bdev)-1)) - BUG(); - if (size < 512 || size > PAGE_SIZE) - BUG(); - sizebits = -1; do { sizebits++; @@ -1233,6 +1279,18 @@ grow_buffers(struct block_device *bdev, sector_t block, int size) struct buffer_head * __getblk_slow(struct block_device *bdev, sector_t block, int size) { + /* Size must be multiple of hard sectorsize */ + if (unlikely(size & (bdev_hardsect_size(bdev)-1) || + (size < 512 || size > PAGE_SIZE))) { + printk(KERN_ERR "getblk(): invalid block size %d requested\n", + size); + printk(KERN_ERR "hardsect size: %d\n", + bdev_hardsect_size(bdev)); + + dump_stack(); + return NULL; + } + for (;;) { struct buffer_head * bh; @@ -1357,7 +1415,7 @@ struct bh_lru { struct buffer_head *bhs[BH_LRU_SIZE]; }; -static DEFINE_PER_CPU(struct bh_lru, bh_lrus) = {{0}}; +static DEFINE_PER_CPU(struct bh_lru, bh_lrus) = {{ NULL }}; #ifdef CONFIG_SMP #define bh_lru_lock() local_irq_disable() @@ -1488,6 +1546,7 @@ __getblk(struct block_device *bdev, sector_t block, int size) { struct buffer_head *bh = __find_get_block(bdev, block, size); + might_sleep(); if (bh == NULL) bh = __getblk_slow(bdev, block, size); return bh; @@ -1525,10 +1584,9 @@ __bread(struct block_device *bdev, sector_t block, int size) EXPORT_SYMBOL(__bread); /* - * invalidate_bh_lrus() is called rarely - at unmount. Because it is only for - * unmount it only needs to ensure that all buffers from the target device are - * invalidated on return and it doesn't need to worry about new buffers from - * that device being added - the unmount code has to prevent that. + * invalidate_bh_lrus() is called rarely - but not only at unmount. + * This doesn't race because it runs in each cpu either in irq + * or with preempt disabled. */ static void invalidate_bh_lru(void *arg) { @@ -1714,6 +1772,8 @@ void unmap_underlying_metadata(struct block_device *bdev, sector_t block) { struct buffer_head *old_bh; + might_sleep(); + old_bh = __find_get_block_slow(bdev, block, 0); if (old_bh) { clear_buffer_dirty(old_bh); @@ -1744,10 +1804,10 @@ EXPORT_SYMBOL(unmap_underlying_metadata); * state inside lock_buffer(). * * If block_write_full_page() is called for regular writeback - * (called_for_sync() is false) then it will redirty a page which has a locked - * buffer. This only can happen if someone has written the buffer directly, - * with submit_bh(). At the address_space level PageWriteback prevents this - * contention from occurring. + * (wbc->sync_mode == WB_SYNC_NONE) then it will redirty a page which has a + * locked buffer. This only can happen if someone has written the buffer + * directly, with submit_bh(). At the address_space level PageWriteback + * prevents this contention from occurring. */ static int __block_write_full_page(struct inode *inode, struct page *page, get_block_t *get_block, struct writeback_control *wbc) @@ -1836,14 +1896,14 @@ static int __block_write_full_page(struct inode *inode, struct page *page, } } while ((bh = bh->b_this_page) != head); + /* + * The page and its buffers are protected by PageWriteback(), so we can + * drop the bh refcounts early. + */ BUG_ON(PageWriteback(page)); - set_page_writeback(page); /* Keeps try_to_free_buffers() away */ + set_page_writeback(page); unlock_page(page); - /* - * The page may come unlocked any time after the *first* submit_bh() - * call. Be careful with its buffers. - */ do { struct buffer_head *next = bh->b_this_page; if (buffer_async_write(bh)) { @@ -1873,6 +1933,10 @@ done: if (uptodate) SetPageUptodate(page); end_page_writeback(page); + /* + * The page and buffer_heads can be released at any time from + * here on. + */ wbc->pages_skipped++; /* We didn't write this page */ } return err; @@ -2424,7 +2488,7 @@ int nobh_prepare_write(struct page *page, unsigned from, unsigned to, } bh->b_state = map_bh.b_state; atomic_set(&bh->b_count, 0); - bh->b_this_page = 0; + bh->b_this_page = NULL; bh->b_page = page; bh->b_blocknr = map_bh.b_blocknr; bh->b_size = blocksize; @@ -2660,7 +2724,7 @@ int block_write_full_page(struct page *page, get_block_t *get_block, /* * The page straddles i_size. It must be zeroed out on each and every - * writepage invocation because it may be mmapped. "A file is mapped + * writepage invokation because it may be mmapped. "A file is mapped * in multiples of the page size. For a file that is not a multiple of * the page size, the remaining memory is zeroed when mapped, and * writes to that region are not written out to the file." @@ -2690,21 +2754,33 @@ static int end_bio_bh_io_sync(struct bio *bio, unsigned int bytes_done, int err) if (bio->bi_size) return 1; + if (err == -EOPNOTSUPP) { + set_bit(BIO_EOPNOTSUPP, &bio->bi_flags); + set_bit(BH_Eopnotsupp, &bh->b_state); + } + bh->b_end_io(bh, test_bit(BIO_UPTODATE, &bio->bi_flags)); bio_put(bio); return 0; } -void submit_bh(int rw, struct buffer_head * bh) +int submit_bh(int rw, struct buffer_head * bh) { struct bio *bio; + int ret = 0; BUG_ON(!buffer_locked(bh)); BUG_ON(!buffer_mapped(bh)); BUG_ON(!bh->b_end_io); - /* Only clear out a write error when rewriting */ - if (test_set_buffer_req(bh) && rw == WRITE) + if (buffer_ordered(bh) && (rw == WRITE)) + rw = WRITE_BARRIER; + + /* + * Only clear out a write error when rewriting, should this + * include WRITE_SYNC as well? + */ + if (test_set_buffer_req(bh) && (rw == WRITE || rw == WRITE_BARRIER)) clear_buffer_write_io_error(bh); /* @@ -2726,7 +2802,14 @@ void submit_bh(int rw, struct buffer_head * bh) bio->bi_end_io = end_bio_bh_io_sync; bio->bi_private = bh; + bio_get(bio); submit_bio(rw, bio); + + if (bio_flagged(bio, BIO_EOPNOTSUPP)) + ret = -EOPNOTSUPP; + + bio_put(bio); + return ret; } /** @@ -2785,20 +2868,30 @@ void ll_rw_block(int rw, int nr, struct buffer_head *bhs[]) /* * For a data-integrity writeout, we need to wait upon any in-progress I/O - * and then start new I/O and then wait upon it. + * and then start new I/O and then wait upon it. The caller must have a ref on + * the buffer_head. */ -void sync_dirty_buffer(struct buffer_head *bh) +int sync_dirty_buffer(struct buffer_head *bh) { + int ret = 0; + WARN_ON(atomic_read(&bh->b_count) < 1); lock_buffer(bh); if (test_clear_buffer_dirty(bh)) { get_bh(bh); bh->b_end_io = end_buffer_write_sync; - submit_bh(WRITE, bh); + ret = submit_bh(WRITE, bh); wait_on_buffer(bh); + if (buffer_eopnotsupp(bh)) { + clear_buffer_eopnotsupp(bh); + ret = -EOPNOTSUPP; + } + if (!ret && !buffer_uptodate(bh)) + ret = -EIO; } else { unlock_buffer(bh); } + return ret; } /* @@ -2832,7 +2925,6 @@ drop_buffers(struct page *page, struct buffer_head **buffers_to_free) { struct buffer_head *head = page_buffers(page); struct buffer_head *bh; - int was_uptodate = 1; bh = head; do { @@ -2840,8 +2932,6 @@ drop_buffers(struct page *page, struct buffer_head **buffers_to_free) set_bit(AS_EIO, &page->mapping->flags); if (buffer_busy(bh)) goto failed; - if (!buffer_uptodate(bh) && !buffer_req(bh)) - was_uptodate = 0; bh = bh->b_this_page; } while (bh != head); @@ -2905,9 +2995,11 @@ EXPORT_SYMBOL(try_to_free_buffers); int block_sync_page(struct page *page) { struct address_space *mapping; + smp_mb(); - mapping = page->mapping; - blk_run_address_space(mapping); + mapping = page_mapping(page); + if (mapping) + blk_run_backing_dev(mapping->backing_dev_info, page); return 0; } @@ -3035,7 +3127,7 @@ void __init buffer_init(void) bh_cachep = kmem_cache_create("buffer_head", sizeof(struct buffer_head), 0, - 0, init_buffer_head, NULL); + SLAB_PANIC, init_buffer_head, NULL); for (i = 0; i < ARRAY_SIZE(bh_wait_queue_heads); i++) init_waitqueue_head(&bh_wait_queue_heads[i].wqh); @@ -3056,14 +3148,12 @@ EXPORT_SYMBOL(block_read_full_page); EXPORT_SYMBOL(block_sync_page); EXPORT_SYMBOL(block_truncate_page); EXPORT_SYMBOL(block_write_full_page); -EXPORT_SYMBOL(buffer_insert_list); EXPORT_SYMBOL(cont_prepare_write); EXPORT_SYMBOL(end_buffer_async_write); EXPORT_SYMBOL(end_buffer_read_sync); EXPORT_SYMBOL(end_buffer_write_sync); EXPORT_SYMBOL(file_fsync); EXPORT_SYMBOL(fsync_bdev); -EXPORT_SYMBOL(fsync_buffers_list); EXPORT_SYMBOL(generic_block_bmap); EXPORT_SYMBOL(generic_commit_write); EXPORT_SYMBOL(generic_cont_expand);