diff options
Diffstat (limited to 'fs/block_dev.c')
| -rw-r--r-- | fs/block_dev.c | 1392 |
1 files changed, 767 insertions, 625 deletions
diff --git a/fs/block_dev.c b/fs/block_dev.c index 73d6a735b8f..6d7274619bf 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -11,13 +11,14 @@ #include <linux/slab.h> #include <linux/kmod.h> #include <linux/major.h> -#include <linux/smp_lock.h> #include <linux/device_cgroup.h> #include <linux/highmem.h> #include <linux/blkdev.h> #include <linux/module.h> #include <linux/blkpg.h> +#include <linux/magic.h> #include <linux/buffer_head.h> +#include <linux/swap.h> #include <linux/pagevec.h> #include <linux/writeback.h> #include <linux/mpage.h> @@ -25,7 +26,8 @@ #include <linux/uio.h> #include <linux/namei.h> #include <linux/log2.h> -#include <linux/kmemleak.h> +#include <linux/cleancache.h> +#include <linux/aio.h> #include <asm/uaccess.h> #include "internal.h" @@ -45,30 +47,67 @@ inline struct block_device *I_BDEV(struct inode *inode) { return &BDEV_I(inode)->bdev; } - EXPORT_SYMBOL(I_BDEV); -static sector_t max_block(struct block_device *bdev) +/* + * Move the inode from its current bdi to a new bdi. If the inode is dirty we + * need to move it onto the dirty list of @dst so that the inode is always on + * the right list. + */ +static void bdev_inode_switch_bdi(struct inode *inode, + struct backing_dev_info *dst) { - sector_t retval = ~((sector_t)0); - loff_t sz = i_size_read(bdev->bd_inode); + struct backing_dev_info *old = inode->i_data.backing_dev_info; + bool wakeup_bdi = false; - if (sz) { - unsigned int size = block_size(bdev); - unsigned int sizebits = blksize_bits(size); - retval = (sz >> sizebits); + if (unlikely(dst == old)) /* deadlock avoidance */ + return; + bdi_lock_two(&old->wb, &dst->wb); + spin_lock(&inode->i_lock); + inode->i_data.backing_dev_info = dst; + if (inode->i_state & I_DIRTY) { + if (bdi_cap_writeback_dirty(dst) && !wb_has_dirty_io(&dst->wb)) + wakeup_bdi = true; + list_move(&inode->i_wb_list, &dst->wb.b_dirty); } - return retval; + spin_unlock(&inode->i_lock); + spin_unlock(&old->wb.list_lock); + spin_unlock(&dst->wb.list_lock); + + if (wakeup_bdi) + bdi_wakeup_thread_delayed(dst); } /* Kill _all_ buffers and pagecache , dirty or not.. */ -static void kill_bdev(struct block_device *bdev) +void kill_bdev(struct block_device *bdev) { - if (bdev->bd_inode->i_mapping->nrpages == 0) + struct address_space *mapping = bdev->bd_inode->i_mapping; + + if (mapping->nrpages == 0 && mapping->nrshadows == 0) return; + invalidate_bh_lrus(); - truncate_inode_pages(bdev->bd_inode->i_mapping, 0); + truncate_inode_pages(mapping, 0); } +EXPORT_SYMBOL(kill_bdev); + +/* Invalidate clean unused buffers and pagecache. */ +void invalidate_bdev(struct block_device *bdev) +{ + struct address_space *mapping = bdev->bd_inode->i_mapping; + + if (mapping->nrpages == 0) + return; + + invalidate_bh_lrus(); + lru_add_drain_all(); /* make sure all lru add caches are flushed */ + invalidate_mapping_pages(mapping, 0, -1); + /* 99% of the time, we don't need to flush the cleancache on the bdev. + * But, for the strange corners, lets be cautious + */ + cleancache_invalidate_inode(mapping); +} +EXPORT_SYMBOL(invalidate_bdev); int set_blocksize(struct block_device *bdev, int size) { @@ -119,61 +158,22 @@ static int blkdev_get_block(struct inode *inode, sector_t iblock, struct buffer_head *bh, int create) { - if (iblock >= max_block(I_BDEV(inode))) { - if (create) - return -EIO; - - /* - * for reads, we're just trying to fill a partial page. - * return a hole, they will have to call get_block again - * before they can fill it, and they will get -EIO at that - * time - */ - return 0; - } bh->b_bdev = I_BDEV(inode); bh->b_blocknr = iblock; set_buffer_mapped(bh); return 0; } -static int -blkdev_get_blocks(struct inode *inode, sector_t iblock, - struct buffer_head *bh, int create) -{ - sector_t end_block = max_block(I_BDEV(inode)); - unsigned long max_blocks = bh->b_size >> inode->i_blkbits; - - if ((iblock + max_blocks) > end_block) { - max_blocks = end_block - iblock; - if ((long)max_blocks <= 0) { - if (create) - return -EIO; /* write fully beyond EOF */ - /* - * It is a read which is fully beyond EOF. We return - * a !buffer_mapped buffer - */ - max_blocks = 0; - } - } - - bh->b_bdev = I_BDEV(inode); - bh->b_blocknr = iblock; - bh->b_size = max_blocks << inode->i_blkbits; - if (max_blocks) - set_buffer_mapped(bh); - return 0; -} - static ssize_t -blkdev_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov, - loff_t offset, unsigned long nr_segs) +blkdev_direct_IO(int rw, struct kiocb *iocb, struct iov_iter *iter, + loff_t offset) { struct file *file = iocb->ki_filp; struct inode *inode = file->f_mapping->host; - return blockdev_direct_IO_no_locking(rw, iocb, inode, I_BDEV(inode), - iov, offset, nr_segs, blkdev_get_blocks, NULL); + return __blockdev_direct_IO(rw, iocb, inode, I_BDEV(inode), iter, + offset, blkdev_get_block, + NULL, NULL, 0); } int __sync_blockdev(struct block_device *bdev, int wait) @@ -245,36 +245,14 @@ struct super_block *freeze_bdev(struct block_device *bdev) sb = get_active_super(bdev); if (!sb) goto out; - if (sb->s_flags & MS_RDONLY) { - deactivate_locked_super(sb); + error = freeze_super(sb); + if (error) { + deactivate_super(sb); + bdev->bd_fsfreeze_count--; mutex_unlock(&bdev->bd_fsfreeze_mutex); - return sb; - } - - sb->s_frozen = SB_FREEZE_WRITE; - smp_wmb(); - - sync_filesystem(sb); - - sb->s_frozen = SB_FREEZE_TRANS; - smp_wmb(); - - sync_blockdev(sb->s_bdev); - - if (sb->s_op->freeze_fs) { - error = sb->s_op->freeze_fs(sb); - if (error) { - printk(KERN_ERR - "VFS:Filesystem freeze failed\n"); - sb->s_frozen = SB_UNFROZEN; - deactivate_locked_super(sb); - bdev->bd_fsfreeze_count--; - mutex_unlock(&bdev->bd_fsfreeze_mutex); - return ERR_PTR(error); - } + return ERR_PTR(error); } - up_write(&sb->s_umount); - + deactivate_super(sb); out: sync_blockdev(bdev); mutex_unlock(&bdev->bd_fsfreeze_mutex); @@ -295,40 +273,22 @@ int thaw_bdev(struct block_device *bdev, struct super_block *sb) mutex_lock(&bdev->bd_fsfreeze_mutex); if (!bdev->bd_fsfreeze_count) - goto out_unlock; + goto out; error = 0; if (--bdev->bd_fsfreeze_count > 0) - goto out_unlock; + goto out; if (!sb) - goto out_unlock; + goto out; - BUG_ON(sb->s_bdev != bdev); - down_write(&sb->s_umount); - if (sb->s_flags & MS_RDONLY) - goto out_deactivate; - - if (sb->s_op->unfreeze_fs) { - error = sb->s_op->unfreeze_fs(sb); - if (error) { - printk(KERN_ERR - "VFS:Filesystem thaw failed\n"); - sb->s_frozen = SB_FREEZE_TRANS; - bdev->bd_fsfreeze_count++; - mutex_unlock(&bdev->bd_fsfreeze_mutex); - return error; - } + error = thaw_super(sb); + if (error) { + bdev->bd_fsfreeze_count++; + mutex_unlock(&bdev->bd_fsfreeze_mutex); + return error; } - - sb->s_frozen = SB_UNFROZEN; - smp_wmb(); - wake_up(&sb->s_wait_unfrozen); - -out_deactivate: - if (sb) - deactivate_locked_super(sb); -out_unlock: +out: mutex_unlock(&bdev->bd_fsfreeze_mutex); return 0; } @@ -348,9 +308,8 @@ static int blkdev_write_begin(struct file *file, struct address_space *mapping, loff_t pos, unsigned len, unsigned flags, struct page **pagep, void **fsdata) { - *pagep = NULL; - return block_write_begin(file, mapping, pos, len, flags, pagep, fsdata, - blkdev_get_block); + return block_write_begin(mapping, pos, len, flags, pagep, + blkdev_get_block); } static int blkdev_write_end(struct file *file, struct address_space *mapping, @@ -368,55 +327,105 @@ static int blkdev_write_end(struct file *file, struct address_space *mapping, /* * private llseek: - * for a block special file file->f_path.dentry->d_inode->i_size is zero + * for a block special file file_inode(file)->i_size is zero * so we compute the size by hand (just as in block_read/write above) */ -static loff_t block_llseek(struct file *file, loff_t offset, int origin) +static loff_t block_llseek(struct file *file, loff_t offset, int whence) { struct inode *bd_inode = file->f_mapping->host; - loff_t size; loff_t retval; mutex_lock(&bd_inode->i_mutex); - size = i_size_read(bd_inode); - - switch (origin) { - case 2: - offset += size; - break; - case 1: - offset += file->f_pos; - } - retval = -EINVAL; - if (offset >= 0 && offset <= size) { - if (offset != file->f_pos) { - file->f_pos = offset; - } - retval = offset; - } + retval = fixed_size_llseek(file, offset, whence, i_size_read(bd_inode)); mutex_unlock(&bd_inode->i_mutex); return retval; } -/* - * Filp is never NULL; the only case when ->fsync() is called with - * NULL first argument is nfsd_sync_dir() and that's not a directory. - */ - -static int block_fsync(struct file *filp, struct dentry *dentry, int datasync) +int blkdev_fsync(struct file *filp, loff_t start, loff_t end, int datasync) { - struct block_device *bdev = I_BDEV(filp->f_mapping->host); + struct inode *bd_inode = filp->f_mapping->host; + struct block_device *bdev = I_BDEV(bd_inode); int error; - - error = sync_blockdev(bdev); + + error = filemap_write_and_wait_range(filp->f_mapping, start, end); if (error) return error; - - error = blkdev_issue_flush(bdev, NULL); + + /* + * There is no need to serialise calls to blkdev_issue_flush with + * i_mutex and doing so causes performance issues with concurrent + * O_SYNC writers to a block device. + */ + error = blkdev_issue_flush(bdev, GFP_KERNEL, NULL); if (error == -EOPNOTSUPP) error = 0; + return error; } +EXPORT_SYMBOL(blkdev_fsync); + +/** + * bdev_read_page() - Start reading a page from a block device + * @bdev: The device to read the page from + * @sector: The offset on the device to read the page to (need not be aligned) + * @page: The page to read + * + * On entry, the page should be locked. It will be unlocked when the page + * has been read. If the block driver implements rw_page synchronously, + * that will be true on exit from this function, but it need not be. + * + * Errors returned by this function are usually "soft", eg out of memory, or + * queue full; callers should try a different route to read this page rather + * than propagate an error back up the stack. + * + * Return: negative errno if an error occurs, 0 if submission was successful. + */ +int bdev_read_page(struct block_device *bdev, sector_t sector, + struct page *page) +{ + const struct block_device_operations *ops = bdev->bd_disk->fops; + if (!ops->rw_page) + return -EOPNOTSUPP; + return ops->rw_page(bdev, sector + get_start_sect(bdev), page, READ); +} +EXPORT_SYMBOL_GPL(bdev_read_page); + +/** + * bdev_write_page() - Start writing a page to a block device + * @bdev: The device to write the page to + * @sector: The offset on the device to write the page to (need not be aligned) + * @page: The page to write + * @wbc: The writeback_control for the write + * + * On entry, the page should be locked and not currently under writeback. + * On exit, if the write started successfully, the page will be unlocked and + * under writeback. If the write failed already (eg the driver failed to + * queue the page to the device), the page will still be locked. If the + * caller is a ->writepage implementation, it will need to unlock the page. + * + * Errors returned by this function are usually "soft", eg out of memory, or + * queue full; callers should try a different route to write this page rather + * than propagate an error back up the stack. + * + * Return: negative errno if an error occurs, 0 if submission was successful. + */ +int bdev_write_page(struct block_device *bdev, sector_t sector, + struct page *page, struct writeback_control *wbc) +{ + int result; + int rw = (wbc->sync_mode == WB_SYNC_ALL) ? WRITE_SYNC : WRITE; + const struct block_device_operations *ops = bdev->bd_disk->fops; + if (!ops->rw_page) + return -EOPNOTSUPP; + set_page_writeback(page); + result = ops->rw_page(bdev, sector + get_start_sect(bdev), page, rw); + if (result) + end_page_writeback(page); + else + unlock_page(page); + return result; +} +EXPORT_SYMBOL_GPL(bdev_write_page); /* * pseudo-fs @@ -433,13 +442,19 @@ static struct inode *bdev_alloc_inode(struct super_block *sb) return &ei->vfs_inode; } -static void bdev_destroy_inode(struct inode *inode) +static void bdev_i_callback(struct rcu_head *head) { + struct inode *inode = container_of(head, struct inode, i_rcu); struct bdev_inode *bdi = BDEV_I(inode); kmem_cache_free(bdev_cachep, bdi); } +static void bdev_destroy_inode(struct inode *inode) +{ + call_rcu(&inode->i_rcu, bdev_i_callback); +} + static void init_once(void *foo) { struct bdev_inode *ei = (struct bdev_inode *) foo; @@ -450,7 +465,7 @@ static void init_once(void *foo) INIT_LIST_HEAD(&bdev->bd_inodes); INIT_LIST_HEAD(&bdev->bd_list); #ifdef CONFIG_SYSFS - INIT_LIST_HEAD(&bdev->bd_holder_list); + INIT_LIST_HEAD(&bdev->bd_holder_disks); #endif inode_init_once(&ei->vfs_inode); /* Initialize mutex for freeze. */ @@ -464,10 +479,13 @@ static inline void __bd_forget(struct inode *inode) inode->i_mapping = &inode->i_data; } -static void bdev_clear_inode(struct inode *inode) +static void bdev_evict_inode(struct inode *inode) { struct block_device *bdev = &BDEV_I(inode)->bdev; struct list_head *p; + truncate_inode_pages_final(&inode->i_data); + invalidate_inode_buffers(inode); /* is it needed here? */ + clear_inode(inode); spin_lock(&bdev_lock); while ( (p = bdev->bd_inodes.next) != &bdev->bd_inodes ) { __bd_forget(list_entry(p, struct inode, i_devices)); @@ -481,27 +499,27 @@ static const struct super_operations bdev_sops = { .alloc_inode = bdev_alloc_inode, .destroy_inode = bdev_destroy_inode, .drop_inode = generic_delete_inode, - .clear_inode = bdev_clear_inode, + .evict_inode = bdev_evict_inode, }; -static int bd_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *bd_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_pseudo(fs_type, "bdev:", &bdev_sops, 0x62646576, mnt); + return mount_pseudo(fs_type, "bdev:", &bdev_sops, NULL, BDEVFS_MAGIC); } static struct file_system_type bd_type = { .name = "bdev", - .get_sb = bd_get_sb, + .mount = bd_mount, .kill_sb = kill_anon_super, }; -struct super_block *blockdev_superblock __read_mostly; +static struct super_block *blockdev_superblock __read_mostly; void __init bdev_cache_init(void) { int err; - struct vfsmount *bd_mnt; + static struct vfsmount *bd_mnt; bdev_cachep = kmem_cache_create("bdev_cache", sizeof(struct bdev_inode), 0, (SLAB_HWCACHE_ALIGN|SLAB_RECLAIM_ACCOUNT| @@ -513,12 +531,7 @@ void __init bdev_cache_init(void) bd_mnt = kern_mount(&bd_type); if (IS_ERR(bd_mnt)) panic("Cannot create bdev pseudo-fs"); - /* - * This vfsmount structure is only used to obtain the - * blockdev_superblock, so tell kmemleak not to report it. - */ - kmemleak_not_leak(bd_mnt); - blockdev_superblock = bd_mnt->mnt_sb; /* For writeback */ + blockdev_superblock = bd_mnt->mnt_sb; /* For writeback */ } /* @@ -559,6 +572,7 @@ struct block_device *bdget(dev_t dev) if (inode->i_state & I_NEW) { bdev->bd_contains = NULL; + bdev->bd_super = NULL; bdev->bd_inode = inode; bdev->bd_block_size = (1 << inode->i_blkbits); bdev->bd_part_count = 0; @@ -585,9 +599,10 @@ EXPORT_SYMBOL(bdget); */ struct block_device *bdgrab(struct block_device *bdev) { - atomic_inc(&bdev->bd_inode->i_count); + ihold(bdev->bd_inode); return bdev; } +EXPORT_SYMBOL(bdgrab); long nr_blockdev_pages(void) { @@ -615,7 +630,7 @@ static struct block_device *bd_acquire(struct inode *inode) spin_lock(&bdev_lock); bdev = inode->i_bdev; if (bdev) { - atomic_inc(&bdev->bd_inode->i_count); + ihold(bdev->bd_inode); spin_unlock(&bdev_lock); return bdev; } @@ -626,12 +641,12 @@ static struct block_device *bd_acquire(struct inode *inode) spin_lock(&bdev_lock); if (!inode->i_bdev) { /* - * We take an additional bd_inode->i_count for inode, + * We take an additional reference to bd_inode, * and it's released in clear_inode() of inode. * So, we can access it via ->i_mapping always * without igrab(). */ - atomic_inc(&bdev->bd_inode->i_count); + ihold(bdev->bd_inode); inode->i_bdev = bdev; inode->i_mapping = bdev->bd_inode->i_mapping; list_add(&inode->i_devices, &bdev->bd_inodes); @@ -641,6 +656,11 @@ static struct block_device *bd_acquire(struct inode *inode) return bdev; } +int sb_is_blkdev_sb(struct super_block *sb) +{ + return sb == blockdev_superblock; +} + /* Call when you free inode */ void bd_forget(struct inode *inode) @@ -648,407 +668,319 @@ void bd_forget(struct inode *inode) struct block_device *bdev = NULL; spin_lock(&bdev_lock); - if (inode->i_bdev) { - if (!sb_is_blkdev_sb(inode->i_sb)) - bdev = inode->i_bdev; - __bd_forget(inode); - } + if (!sb_is_blkdev_sb(inode->i_sb)) + bdev = inode->i_bdev; + __bd_forget(inode); spin_unlock(&bdev_lock); if (bdev) iput(bdev->bd_inode); } -int bd_claim(struct block_device *bdev, void *holder) +/** + * bd_may_claim - test whether a block device can be claimed + * @bdev: block device of interest + * @whole: whole block device containing @bdev, may equal @bdev + * @holder: holder trying to claim @bdev + * + * Test whether @bdev can be claimed by @holder. + * + * CONTEXT: + * spin_lock(&bdev_lock). + * + * RETURNS: + * %true if @bdev can be claimed, %false otherwise. + */ +static bool bd_may_claim(struct block_device *bdev, struct block_device *whole, + void *holder) { - int res; - spin_lock(&bdev_lock); - - /* first decide result */ if (bdev->bd_holder == holder) - res = 0; /* already a holder */ + return true; /* already a holder */ else if (bdev->bd_holder != NULL) - res = -EBUSY; /* held by someone else */ + return false; /* held by someone else */ else if (bdev->bd_contains == bdev) - res = 0; /* is a whole device which isn't held */ + return true; /* is a whole device which isn't held */ - else if (bdev->bd_contains->bd_holder == bd_claim) - res = 0; /* is a partition of a device that is being partitioned */ - else if (bdev->bd_contains->bd_holder != NULL) - res = -EBUSY; /* is a partition of a held device */ + else if (whole->bd_holder == bd_may_claim) + return true; /* is a partition of a device that is being partitioned */ + else if (whole->bd_holder != NULL) + return false; /* is a partition of a held device */ else - res = 0; /* is a partition of an un-held device */ - - /* now impose change */ - if (res==0) { - /* note that for a whole device bd_holders - * will be incremented twice, and bd_holder will - * be set to bd_claim before being set to holder - */ - bdev->bd_contains->bd_holders ++; - bdev->bd_contains->bd_holder = bd_claim; - bdev->bd_holders++; - bdev->bd_holder = holder; - } - spin_unlock(&bdev_lock); - return res; -} - -EXPORT_SYMBOL(bd_claim); - -void bd_release(struct block_device *bdev) -{ - spin_lock(&bdev_lock); - if (!--bdev->bd_contains->bd_holders) - bdev->bd_contains->bd_holder = NULL; - if (!--bdev->bd_holders) - bdev->bd_holder = NULL; - spin_unlock(&bdev_lock); + return true; /* is a partition of an un-held device */ } -EXPORT_SYMBOL(bd_release); - -#ifdef CONFIG_SYSFS -/* - * Functions for bd_claim_by_kobject / bd_release_from_kobject +/** + * bd_prepare_to_claim - prepare to claim a block device + * @bdev: block device of interest + * @whole: the whole device containing @bdev, may equal @bdev + * @holder: holder trying to claim @bdev + * + * Prepare to claim @bdev. This function fails if @bdev is already + * claimed by another holder and waits if another claiming is in + * progress. This function doesn't actually claim. On successful + * return, the caller has ownership of bd_claiming and bd_holder[s]. * - * If a kobject is passed to bd_claim_by_kobject() - * and the kobject has a parent directory, - * following symlinks are created: - * o from the kobject to the claimed bdev - * o from "holders" directory of the bdev to the parent of the kobject - * bd_release_from_kobject() removes these symlinks. + * CONTEXT: + * spin_lock(&bdev_lock). Might release bdev_lock, sleep and regrab + * it multiple times. * - * Example: - * If /dev/dm-0 maps to /dev/sda, kobject corresponding to - * /sys/block/dm-0/slaves is passed to bd_claim_by_kobject(), then: - * /sys/block/dm-0/slaves/sda --> /sys/block/sda - * /sys/block/sda/holders/dm-0 --> /sys/block/dm-0 + * RETURNS: + * 0 if @bdev can be claimed, -EBUSY otherwise. */ - -static int add_symlink(struct kobject *from, struct kobject *to) +static int bd_prepare_to_claim(struct block_device *bdev, + struct block_device *whole, void *holder) { - if (!from || !to) - return 0; - return sysfs_create_link(from, to, kobject_name(to)); -} +retry: + /* if someone else claimed, fail */ + if (!bd_may_claim(bdev, whole, holder)) + return -EBUSY; -static void del_symlink(struct kobject *from, struct kobject *to) -{ - if (!from || !to) - return; - sysfs_remove_link(from, kobject_name(to)); -} + /* if claiming is already in progress, wait for it to finish */ + if (whole->bd_claiming) { + wait_queue_head_t *wq = bit_waitqueue(&whole->bd_claiming, 0); + DEFINE_WAIT(wait); -/* - * 'struct bd_holder' contains pointers to kobjects symlinked by - * bd_claim_by_kobject. - * It's connected to bd_holder_list which is protected by bdev->bd_sem. - */ -struct bd_holder { - struct list_head list; /* chain of holders of the bdev */ - int count; /* references from the holder */ - struct kobject *sdir; /* holder object, e.g. "/block/dm-0/slaves" */ - struct kobject *hdev; /* e.g. "/block/dm-0" */ - struct kobject *hdir; /* e.g. "/block/sda/holders" */ - struct kobject *sdev; /* e.g. "/block/sda" */ -}; + prepare_to_wait(wq, &wait, TASK_UNINTERRUPTIBLE); + spin_unlock(&bdev_lock); + schedule(); + finish_wait(wq, &wait); + spin_lock(&bdev_lock); + goto retry; + } -/* - * Get references of related kobjects at once. - * Returns 1 on success. 0 on failure. + /* yay, all mine */ + return 0; +} + +/** + * bd_start_claiming - start claiming a block device + * @bdev: block device of interest + * @holder: holder trying to claim @bdev + * + * @bdev is about to be opened exclusively. Check @bdev can be opened + * exclusively and mark that an exclusive open is in progress. Each + * successful call to this function must be matched with a call to + * either bd_finish_claiming() or bd_abort_claiming() (which do not + * fail). + * + * This function is used to gain exclusive access to the block device + * without actually causing other exclusive open attempts to fail. It + * should be used when the open sequence itself requires exclusive + * access but may subsequently fail. + * + * CONTEXT: + * Might sleep. * - * Should call bd_holder_release_dirs() after successful use. + * RETURNS: + * Pointer to the block device containing @bdev on success, ERR_PTR() + * value on failure. */ -static int bd_holder_grab_dirs(struct block_device *bdev, - struct bd_holder *bo) +static struct block_device *bd_start_claiming(struct block_device *bdev, + void *holder) { - if (!bdev || !bo) - return 0; - - bo->sdir = kobject_get(bo->sdir); - if (!bo->sdir) - return 0; + struct gendisk *disk; + struct block_device *whole; + int partno, err; - bo->hdev = kobject_get(bo->sdir->parent); - if (!bo->hdev) - goto fail_put_sdir; + might_sleep(); - bo->sdev = kobject_get(&part_to_dev(bdev->bd_part)->kobj); - if (!bo->sdev) - goto fail_put_hdev; + /* + * @bdev might not have been initialized properly yet, look up + * and grab the outer block device the hard way. + */ + disk = get_gendisk(bdev->bd_dev, &partno); + if (!disk) + return ERR_PTR(-ENXIO); - bo->hdir = kobject_get(bdev->bd_part->holder_dir); - if (!bo->hdir) - goto fail_put_sdev; + /* + * Normally, @bdev should equal what's returned from bdget_disk() + * if partno is 0; however, some drivers (floppy) use multiple + * bdev's for the same physical device and @bdev may be one of the + * aliases. Keep @bdev if partno is 0. This means claimer + * tracking is broken for those devices but it has always been that + * way. + */ + if (partno) + whole = bdget_disk(disk, 0); + else + whole = bdgrab(bdev); - return 1; + module_put(disk->fops->owner); + put_disk(disk); + if (!whole) + return ERR_PTR(-ENOMEM); -fail_put_sdev: - kobject_put(bo->sdev); -fail_put_hdev: - kobject_put(bo->hdev); -fail_put_sdir: - kobject_put(bo->sdir); + /* prepare to claim, if successful, mark claiming in progress */ + spin_lock(&bdev_lock); - return 0; + err = bd_prepare_to_claim(bdev, whole, holder); + if (err == 0) { + whole->bd_claiming = holder; + spin_unlock(&bdev_lock); + return whole; + } else { + spin_unlock(&bdev_lock); + bdput(whole); + return ERR_PTR(err); + } } -/* Put references of related kobjects at once. */ -static void bd_holder_release_dirs(struct bd_holder *bo) -{ - kobject_put(bo->hdir); - kobject_put(bo->sdev); - kobject_put(bo->hdev); - kobject_put(bo->sdir); -} +#ifdef CONFIG_SYSFS +struct bd_holder_disk { + struct list_head list; + struct gendisk *disk; + int refcnt; +}; -static struct bd_holder *alloc_bd_holder(struct kobject *kobj) +static struct bd_holder_disk *bd_find_holder_disk(struct block_device *bdev, + struct gendisk *disk) { - struct bd_holder *bo; + struct bd_holder_disk *holder; - bo = kzalloc(sizeof(*bo), GFP_KERNEL); - if (!bo) - return NULL; - - bo->count = 1; - bo->sdir = kobj; - - return bo; + list_for_each_entry(holder, &bdev->bd_holder_disks, list) + if (holder->disk == disk) + return holder; + return NULL; } -static void free_bd_holder(struct bd_holder *bo) +static int add_symlink(struct kobject *from, struct kobject *to) { - kfree(bo); + return sysfs_create_link(from, to, kobject_name(to)); } -/** - * find_bd_holder - find matching struct bd_holder from the block device - * - * @bdev: struct block device to be searched - * @bo: target struct bd_holder - * - * Returns matching entry with @bo in @bdev->bd_holder_list. - * If found, increment the reference count and return the pointer. - * If not found, returns NULL. - */ -static struct bd_holder *find_bd_holder(struct block_device *bdev, - struct bd_holder *bo) +static void del_symlink(struct kobject *from, struct kobject *to) { - struct bd_holder *tmp; - - list_for_each_entry(tmp, &bdev->bd_holder_list, list) - if (tmp->sdir == bo->sdir) { - tmp->count++; - return tmp; - } - - return NULL; + sysfs_remove_link(from, kobject_name(to)); } /** - * add_bd_holder - create sysfs symlinks for bd_claim() relationship + * bd_link_disk_holder - create symlinks between holding disk and slave bdev + * @bdev: the claimed slave bdev + * @disk: the holding disk + * + * DON'T USE THIS UNLESS YOU'RE ALREADY USING IT. + * + * This functions creates the following sysfs symlinks. + * + * - from "slaves" directory of the holder @disk to the claimed @bdev + * - from "holders" directory of the @bdev to the holder @disk + * + * For example, if /dev/dm-0 maps to /dev/sda and disk for dm-0 is + * passed to bd_link_disk_holder(), then: + * + * /sys/block/dm-0/slaves/sda --> /sys/block/sda + * /sys/block/sda/holders/dm-0 --> /sys/block/dm-0 * - * @bdev: block device to be bd_claimed - * @bo: preallocated and initialized by alloc_bd_holder() + * The caller must have claimed @bdev before calling this function and + * ensure that both @bdev and @disk are valid during the creation and + * lifetime of these symlinks. * - * Add @bo to @bdev->bd_holder_list, create symlinks. + * CONTEXT: + * Might sleep. * - * Returns 0 if symlinks are created. - * Returns -ve if something fails. + * RETURNS: + * 0 on success, -errno on failure. */ -static int add_bd_holder(struct block_device *bdev, struct bd_holder *bo) +int bd_link_disk_holder(struct block_device *bdev, struct gendisk *disk) { - int err; + struct bd_holder_disk *holder; + int ret = 0; - if (!bo) - return -EINVAL; + mutex_lock(&bdev->bd_mutex); - if (!bd_holder_grab_dirs(bdev, bo)) - return -EBUSY; + WARN_ON_ONCE(!bdev->bd_holder); - err = add_symlink(bo->sdir, bo->sdev); - if (err) - return err; + /* FIXME: remove the following once add_disk() handles errors */ + if (WARN_ON(!disk->slave_dir || !bdev->bd_part->holder_dir)) + goto out_unlock; - err = add_symlink(bo->hdir, bo->hdev); - if (err) { - del_symlink(bo->sdir, bo->sdev); - return err; + holder = bd_find_holder_disk(bdev, disk); + if (holder) { + holder->refcnt++; + goto out_unlock; } - list_add_tail(&bo->list, &bdev->bd_holder_list); - return 0; -} - -/** - * del_bd_holder - delete sysfs symlinks for bd_claim() relationship - * - * @bdev: block device to be bd_claimed - * @kobj: holder's kobject - * - * If there is matching entry with @kobj in @bdev->bd_holder_list - * and no other bd_claim() from the same kobject, - * remove the struct bd_holder from the list, delete symlinks for it. - * - * Returns a pointer to the struct bd_holder when it's removed from the list - * and ready to be freed. - * Returns NULL if matching claim isn't found or there is other bd_claim() - * by the same kobject. - */ -static struct bd_holder *del_bd_holder(struct block_device *bdev, - struct kobject *kobj) -{ - struct bd_holder *bo; - - list_for_each_entry(bo, &bdev->bd_holder_list, list) { - if (bo->sdir == kobj) { - bo->count--; - BUG_ON(bo->count < 0); - if (!bo->count) { - list_del(&bo->list); - del_symlink(bo->sdir, bo->sdev); - del_symlink(bo->hdir, bo->hdev); - bd_holder_release_dirs(bo); - return bo; - } - break; - } + holder = kzalloc(sizeof(*holder), GFP_KERNEL); + if (!holder) { + ret = -ENOMEM; + goto out_unlock; } - return NULL; -} - -/** - * bd_claim_by_kobject - bd_claim() with additional kobject signature - * - * @bdev: block device to be claimed - * @holder: holder's signature - * @kobj: holder's kobject - * - * Do bd_claim() and if it succeeds, create sysfs symlinks between - * the bdev and the holder's kobject. - * Use bd_release_from_kobject() when relesing the claimed bdev. - * - * Returns 0 on success. (same as bd_claim()) - * Returns errno on failure. - */ -static int bd_claim_by_kobject(struct block_device *bdev, void *holder, - struct kobject *kobj) -{ - int err; - struct bd_holder *bo, *found; - - if (!kobj) - return -EINVAL; - - bo = alloc_bd_holder(kobj); - if (!bo) - return -ENOMEM; + INIT_LIST_HEAD(&holder->list); + holder->disk = disk; + holder->refcnt = 1; - mutex_lock(&bdev->bd_mutex); + ret = add_symlink(disk->slave_dir, &part_to_dev(bdev->bd_part)->kobj); + if (ret) + goto out_free; - err = bd_claim(bdev, holder); - if (err) - goto fail; + ret = add_symlink(bdev->bd_part->holder_dir, &disk_to_dev(disk)->kobj); + if (ret) + goto out_del; + /* + * bdev could be deleted beneath us which would implicitly destroy + * the holder directory. Hold on to it. + */ + kobject_get(bdev->bd_part->holder_dir); - found = find_bd_holder(bdev, bo); - if (found) - goto fail; + list_add(&holder->list, &bdev->bd_holder_disks); + goto out_unlock; - err = add_bd_holder(bdev, bo); - if (err) - bd_release(bdev); - else - bo = NULL; -fail: +out_del: + del_symlink(disk->slave_dir, &part_to_dev(bdev->bd_part)->kobj); +out_free: + kfree(holder); +out_unlock: mutex_unlock(&bdev->bd_mutex); - free_bd_holder(bo); - return err; + return ret; } +EXPORT_SYMBOL_GPL(bd_link_disk_holder); /** - * bd_release_from_kobject - bd_release() with additional kobject signature + * bd_unlink_disk_holder - destroy symlinks created by bd_link_disk_holder() + * @bdev: the calimed slave bdev + * @disk: the holding disk * - * @bdev: block device to be released - * @kobj: holder's kobject + * DON'T USE THIS UNLESS YOU'RE ALREADY USING IT. * - * Do bd_release() and remove sysfs symlinks created by bd_claim_by_kobject(). + * CONTEXT: + * Might sleep. */ -static void bd_release_from_kobject(struct block_device *bdev, - struct kobject *kobj) +void bd_unlink_disk_holder(struct block_device *bdev, struct gendisk *disk) { - if (!kobj) - return; + struct bd_holder_disk *holder; mutex_lock(&bdev->bd_mutex); - bd_release(bdev); - free_bd_holder(del_bd_holder(bdev, kobj)); - mutex_unlock(&bdev->bd_mutex); -} -/** - * bd_claim_by_disk - wrapper function for bd_claim_by_kobject() - * - * @bdev: block device to be claimed - * @holder: holder's signature - * @disk: holder's gendisk - * - * Call bd_claim_by_kobject() with getting @disk->slave_dir. - */ -int bd_claim_by_disk(struct block_device *bdev, void *holder, - struct gendisk *disk) -{ - return bd_claim_by_kobject(bdev, holder, kobject_get(disk->slave_dir)); -} -EXPORT_SYMBOL_GPL(bd_claim_by_disk); + holder = bd_find_holder_disk(bdev, disk); -/** - * bd_release_from_disk - wrapper function for bd_release_from_kobject() - * - * @bdev: block device to be claimed - * @disk: holder's gendisk - * - * Call bd_release_from_kobject() and put @disk->slave_dir. - */ -void bd_release_from_disk(struct block_device *bdev, struct gendisk *disk) -{ - bd_release_from_kobject(bdev, disk->slave_dir); - kobject_put(disk->slave_dir); -} -EXPORT_SYMBOL_GPL(bd_release_from_disk); -#endif + if (!WARN_ON_ONCE(holder == NULL) && !--holder->refcnt) { + del_symlink(disk->slave_dir, &part_to_dev(bdev->bd_part)->kobj); + del_symlink(bdev->bd_part->holder_dir, + &disk_to_dev(disk)->kobj); + kobject_put(bdev->bd_part->holder_dir); + list_del_init(&holder->list); + kfree(holder); + } -/* - * Tries to open block device by device number. Use it ONLY if you - * really do not have anything better - i.e. when you are behind a - * truly sucky interface and all you are given is a device number. _Never_ - * to be used for internal purposes. If you ever need it - reconsider - * your API. - */ -struct block_device *open_by_devnum(dev_t dev, fmode_t mode) -{ - struct block_device *bdev = bdget(dev); - int err = -ENOMEM; - if (bdev) - err = blkdev_get(bdev, mode); - return err ? ERR_PTR(err) : bdev; + mutex_unlock(&bdev->bd_mutex); } - -EXPORT_SYMBOL(open_by_devnum); +EXPORT_SYMBOL_GPL(bd_unlink_disk_holder); +#endif /** * flush_disk - invalidates all buffer-cache entries on a disk * * @bdev: struct block device to be flushed + * @kill_dirty: flag to guide handling of dirty inodes * * Invalidates all buffer-cache entries on a disk. It should be called * when a disk has been changed -- either by a media change or online * resize. */ -static void flush_disk(struct block_device *bdev) +static void flush_disk(struct block_device *bdev, bool kill_dirty) { - if (__invalidate_device(bdev)) { + if (__invalidate_device(bdev, kill_dirty)) { char name[BDEVNAME_SIZE] = ""; if (bdev->bd_disk) @@ -1059,7 +991,7 @@ static void flush_disk(struct block_device *bdev) if (!bdev->bd_disk) return; - if (disk_partitionable(bdev->bd_disk)) + if (disk_part_scan_enabled(bdev->bd_disk)) bdev->bd_invalidated = 1; } @@ -1085,7 +1017,7 @@ void check_disk_size_change(struct gendisk *disk, struct block_device *bdev) "%s: detected capacity change from %lld to %lld\n", name, bdev_size, disk_size); i_size_write(bdev->bd_inode, disk_size); - flush_disk(bdev); + flush_disk(bdev, false); } } EXPORT_SYMBOL(check_disk_size_change); @@ -1112,6 +1044,7 @@ int revalidate_disk(struct gendisk *disk) mutex_lock(&bdev->bd_mutex); check_disk_size_change(disk, bdev); + bdev->bd_invalidated = 0; mutex_unlock(&bdev->bd_mutex); bdput(bdev); return ret; @@ -1131,13 +1064,14 @@ int check_disk_change(struct block_device *bdev) { struct gendisk *disk = bdev->bd_disk; const struct block_device_operations *bdops = disk->fops; + unsigned int events; - if (!bdops->media_changed) - return 0; - if (!bdops->media_changed(bdev->bd_disk)) + events = disk_clear_events(disk, DISK_EVENT_MEDIA_CHANGE | + DISK_EVENT_EJECT_REQUEST); + if (!(events & DISK_EVENT_MEDIA_CHANGE)) return 0; - flush_disk(bdev); + flush_disk(bdev, true); if (bdops->revalidate_disk) bdops->revalidate_disk(bdev->bd_disk); return 1; @@ -1149,7 +1083,9 @@ void bd_set_size(struct block_device *bdev, loff_t size) { unsigned bsize = bdev_logical_block_size(bdev); - bdev->bd_inode->i_size = size; + mutex_lock(&bdev->bd_inode->i_mutex); + i_size_write(bdev->bd_inode, size); + mutex_unlock(&bdev->bd_inode->i_mutex); while (bsize < PAGE_CACHE_SIZE) { if (size & bsize) break; @@ -1160,7 +1096,7 @@ void bd_set_size(struct block_device *bdev, loff_t size) } EXPORT_SYMBOL(bd_set_size); -static int __blkdev_put(struct block_device *bdev, fmode_t mode, int for_part); +static void __blkdev_put(struct block_device *bdev, fmode_t mode, int for_part); /* * bd_mutex locking: @@ -1172,6 +1108,7 @@ static int __blkdev_put(struct block_device *bdev, fmode_t mode, int for_part); static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part) { struct gendisk *disk; + struct module *owner; int ret; int partno; int perm = 0; @@ -1183,23 +1120,27 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part) /* * hooks: /n/, see "layering violations". */ - ret = devcgroup_inode_permission(bdev->bd_inode, perm); - if (ret != 0) { - bdput(bdev); - return ret; + if (!for_part) { + ret = devcgroup_inode_permission(bdev->bd_inode, perm); + if (ret != 0) { + bdput(bdev); + return ret; + } } - lock_kernel(); restart: ret = -ENXIO; disk = get_gendisk(bdev->bd_dev, &partno); if (!disk) - goto out_unlock_kernel; + goto out; + owner = disk->fops->owner; + disk_block_events(disk); mutex_lock_nested(&bdev->bd_mutex, for_part); if (!bdev->bd_openers) { bdev->bd_disk = disk; + bdev->bd_queue = disk->queue; bdev->bd_contains = bdev; if (!partno) { struct backing_dev_info *bdi; @@ -1209,6 +1150,7 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part) if (!bdev->bd_part) goto out_clear; + ret = 0; if (disk->fops->open) { ret = disk->fops->open(bdev, mode); if (ret == -ERESTARTSYS) { @@ -1218,24 +1160,38 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part) */ disk_put_part(bdev->bd_part); bdev->bd_part = NULL; - module_put(disk->fops->owner); - put_disk(disk); bdev->bd_disk = NULL; + bdev->bd_queue = NULL; mutex_unlock(&bdev->bd_mutex); + disk_unblock_events(disk); + put_disk(disk); + module_put(owner); goto restart; } - if (ret) - goto out_clear; } - if (!bdev->bd_openers) { + + if (!ret) { bd_set_size(bdev,(loff_t)get_capacity(disk)<<9); bdi = blk_get_backing_dev_info(bdev); if (bdi == NULL) bdi = &default_backing_dev_info; - bdev->bd_inode->i_data.backing_dev_info = bdi; + bdev_inode_switch_bdi(bdev->bd_inode, bdi); } - if (bdev->bd_invalidated) - rescan_partitions(disk, bdev); + + /* + * If the device is invalidated, rescan partition + * if open succeeded or failed with -ENOMEDIUM. + * The latter is necessary to prevent ghost + * partitions on a removed medium. + */ + if (bdev->bd_invalidated) { + if (!ret) + rescan_partitions(disk, bdev); + else if (ret == -ENOMEDIUM) + invalidate_partitions(disk, bdev); + } + if (ret) + goto out_clear; } else { struct block_device *whole; whole = bdget_disk(disk, 0); @@ -1247,8 +1203,8 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part) if (ret) goto out_clear; bdev->bd_contains = whole; - bdev->bd_inode->i_data.backing_dev_info = - whole->bd_inode->i_data.backing_dev_info; + bdev_inode_switch_bdi(bdev->bd_inode, + whole->bd_inode->i_data.backing_dev_info); bdev->bd_part = disk_get_part(disk, partno); if (!(disk->flags & GENHD_FL_UP) || !bdev->bd_part || !bdev->bd_part->nr_sects) { @@ -1258,57 +1214,218 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part) bd_set_size(bdev, (loff_t)bdev->bd_part->nr_sects << 9); } } else { - module_put(disk->fops->owner); - put_disk(disk); - disk = NULL; if (bdev->bd_contains == bdev) { - if (bdev->bd_disk->fops->open) { + ret = 0; + if (bdev->bd_disk->fops->open) ret = bdev->bd_disk->fops->open(bdev, mode); - if (ret) - goto out_unlock_bdev; + /* the same as first opener case, read comment there */ + if (bdev->bd_invalidated) { + if (!ret) + rescan_partitions(bdev->bd_disk, bdev); + else if (ret == -ENOMEDIUM) + invalidate_partitions(bdev->bd_disk, bdev); } - if (bdev->bd_invalidated) - rescan_partitions(bdev->bd_disk, bdev); + if (ret) + goto out_unlock_bdev; } + /* only one opener holds refs to the module and disk */ + put_disk(disk); + module_put(owner); } bdev->bd_openers++; if (for_part) bdev->bd_part_count++; mutex_unlock(&bdev->bd_mutex); - unlock_kernel(); + disk_unblock_events(disk); return 0; out_clear: disk_put_part(bdev->bd_part); bdev->bd_disk = NULL; bdev->bd_part = NULL; - bdev->bd_inode->i_data.backing_dev_info = &default_backing_dev_info; + bdev->bd_queue = NULL; + bdev_inode_switch_bdi(bdev->bd_inode, &default_backing_dev_info); if (bdev != bdev->bd_contains) __blkdev_put(bdev->bd_contains, mode, 1); bdev->bd_contains = NULL; out_unlock_bdev: mutex_unlock(&bdev->bd_mutex); - out_unlock_kernel: - unlock_kernel(); - - if (disk) - module_put(disk->fops->owner); + disk_unblock_events(disk); put_disk(disk); + module_put(owner); + out: bdput(bdev); return ret; } -int blkdev_get(struct block_device *bdev, fmode_t mode) +/** + * blkdev_get - open a block device + * @bdev: block_device to open + * @mode: FMODE_* mask + * @holder: exclusive holder identifier + * + * Open @bdev with @mode. If @mode includes %FMODE_EXCL, @bdev is + * open with exclusive access. Specifying %FMODE_EXCL with %NULL + * @holder is invalid. Exclusive opens may nest for the same @holder. + * + * On success, the reference count of @bdev is unchanged. On failure, + * @bdev is put. + * + * CONTEXT: + * Might sleep. + * + * RETURNS: + * 0 on success, -errno on failure. + */ +int blkdev_get(struct block_device *bdev, fmode_t mode, void *holder) { - return __blkdev_get(bdev, mode, 0); + struct block_device *whole = NULL; + int res; + + WARN_ON_ONCE((mode & FMODE_EXCL) && !holder); + + if ((mode & FMODE_EXCL) && holder) { + whole = bd_start_claiming(bdev, holder); + if (IS_ERR(whole)) { + bdput(bdev); + return PTR_ERR(whole); + } + } + + res = __blkdev_get(bdev, mode, 0); + + if (whole) { + struct gendisk *disk = whole->bd_disk; + + /* finish claiming */ + mutex_lock(&bdev->bd_mutex); + spin_lock(&bdev_lock); + + if (!res) { + BUG_ON(!bd_may_claim(bdev, whole, holder)); + /* + * Note that for a whole device bd_holders + * will be incremented twice, and bd_holder + * will be set to bd_may_claim before being + * set to holder + */ + whole->bd_holders++; + whole->bd_holder = bd_may_claim; + bdev->bd_holders++; + bdev->bd_holder = holder; + } + + /* tell others that we're done */ + BUG_ON(whole->bd_claiming != holder); + whole->bd_claiming = NULL; + wake_up_bit(&whole->bd_claiming, 0); + + spin_unlock(&bdev_lock); + + /* + * Block event polling for write claims if requested. Any + * write holder makes the write_holder state stick until + * all are released. This is good enough and tracking + * individual writeable reference is too fragile given the + * way @mode is used in blkdev_get/put(). + */ + if (!res && (mode & FMODE_WRITE) && !bdev->bd_write_holder && + (disk->flags & GENHD_FL_BLOCK_EVENTS_ON_EXCL_WRITE)) { + bdev->bd_write_holder = true; + disk_block_events(disk); + } + + mutex_unlock(&bdev->bd_mutex); + bdput(whole); + } + + return res; } EXPORT_SYMBOL(blkdev_get); +/** + * blkdev_get_by_path - open a block device by name + * @path: path to the block device to open + * @mode: FMODE_* mask + * @holder: exclusive holder identifier + * + * Open the blockdevice described by the device file at @path. @mode + * and @holder are identical to blkdev_get(). + * + * On success, the returned block_device has reference count of one. + * + * CONTEXT: + * Might sleep. + * + * RETURNS: + * Pointer to block_device on success, ERR_PTR(-errno) on failure. + */ +struct block_device *blkdev_get_by_path(const char *path, fmode_t mode, + void *holder) +{ + struct block_device *bdev; + int err; + + bdev = lookup_bdev(path); + if (IS_ERR(bdev)) + return bdev; + + err = blkdev_get(bdev, mode, holder); + if (err) + return ERR_PTR(err); + + if ((mode & FMODE_WRITE) && bdev_read_only(bdev)) { + blkdev_put(bdev, mode); + return ERR_PTR(-EACCES); + } + + return bdev; +} +EXPORT_SYMBOL(blkdev_get_by_path); + +/** + * blkdev_get_by_dev - open a block device by device number + * @dev: device number of block device to open + * @mode: FMODE_* mask + * @holder: exclusive holder identifier + * + * Open the blockdevice described by device number @dev. @mode and + * @holder are identical to blkdev_get(). + * + * Use it ONLY if you really do not have anything better - i.e. when + * you are behind a truly sucky interface and all you are given is a + * device number. _Never_ to be used for internal purposes. If you + * ever need it - reconsider your API. + * + * On success, the returned block_device has reference count of one. + * + * CONTEXT: + * Might sleep. + * + * RETURNS: + * Pointer to block_device on success, ERR_PTR(-errno) on failure. + */ +struct block_device *blkdev_get_by_dev(dev_t dev, fmode_t mode, void *holder) +{ + struct block_device *bdev; + int err; + + bdev = bdget(dev); + if (!bdev) + return ERR_PTR(-ENOMEM); + + err = blkdev_get(bdev, mode, holder); + if (err) + return ERR_PTR(err); + + return bdev; +} +EXPORT_SYMBOL(blkdev_get_by_dev); + static int blkdev_open(struct inode * inode, struct file * filp) { struct block_device *bdev; - int res; /* * Preserve backwards compatibility and allow large file access @@ -1331,75 +1448,104 @@ static int blkdev_open(struct inode * inode, struct file * filp) filp->f_mapping = bdev->bd_inode->i_mapping; - res = blkdev_get(bdev, filp->f_mode); - if (res) - return res; - - if (filp->f_mode & FMODE_EXCL) { - res = bd_claim(bdev, filp); - if (res) - goto out_blkdev_put; - } - - return 0; - - out_blkdev_put: - blkdev_put(bdev, filp->f_mode); - return res; + return blkdev_get(bdev, filp->f_mode, filp); } -static int __blkdev_put(struct block_device *bdev, fmode_t mode, int for_part) +static void __blkdev_put(struct block_device *bdev, fmode_t mode, int for_part) { - int ret = 0; struct gendisk *disk = bdev->bd_disk; struct block_device *victim = NULL; mutex_lock_nested(&bdev->bd_mutex, for_part); - lock_kernel(); if (for_part) bdev->bd_part_count--; if (!--bdev->bd_openers) { + WARN_ON_ONCE(bdev->bd_holders); sync_blockdev(bdev); kill_bdev(bdev); + /* ->release can cause the old bdi to disappear, + * so must switch it out first + */ + bdev_inode_switch_bdi(bdev->bd_inode, + &default_backing_dev_info); } if (bdev->bd_contains == bdev) { if (disk->fops->release) - ret = disk->fops->release(disk, mode); + disk->fops->release(disk, mode); } if (!bdev->bd_openers) { struct module *owner = disk->fops->owner; - put_disk(disk); - module_put(owner); disk_put_part(bdev->bd_part); bdev->bd_part = NULL; bdev->bd_disk = NULL; - bdev->bd_inode->i_data.backing_dev_info = &default_backing_dev_info; if (bdev != bdev->bd_contains) victim = bdev->bd_contains; bdev->bd_contains = NULL; + + put_disk(disk); + module_put(owner); } - unlock_kernel(); mutex_unlock(&bdev->bd_mutex); bdput(bdev); if (victim) __blkdev_put(victim, mode, 1); - return ret; } -int blkdev_put(struct block_device *bdev, fmode_t mode) +void blkdev_put(struct block_device *bdev, fmode_t mode) { - return __blkdev_put(bdev, mode, 0); + mutex_lock(&bdev->bd_mutex); + + if (mode & FMODE_EXCL) { + bool bdev_free; + + /* + * Release a claim on the device. The holder fields + * are protected with bdev_lock. bd_mutex is to + * synchronize disk_holder unlinking. + */ + spin_lock(&bdev_lock); + + WARN_ON_ONCE(--bdev->bd_holders < 0); + WARN_ON_ONCE(--bdev->bd_contains->bd_holders < 0); + + /* bd_contains might point to self, check in a separate step */ + if ((bdev_free = !bdev->bd_holders)) + bdev->bd_holder = NULL; + if (!bdev->bd_contains->bd_holders) + bdev->bd_contains->bd_holder = NULL; + + spin_unlock(&bdev_lock); + + /* + * If this was the last claim, remove holder link and + * unblock evpoll if it was a write holder. + */ + if (bdev_free && bdev->bd_write_holder) { + disk_unblock_events(bdev->bd_disk); + bdev->bd_write_holder = false; + } + } + + /* + * Trigger event checking and tell drivers to flush MEDIA_CHANGE + * event. This is to ensure detection of media removal commanded + * from userland - e.g. eject(1). + */ + disk_flush_events(bdev->bd_disk, DISK_EVENT_MEDIA_CHANGE); + + mutex_unlock(&bdev->bd_mutex); + + __blkdev_put(bdev, mode, 0); } EXPORT_SYMBOL(blkdev_put); static int blkdev_close(struct inode * inode, struct file * filp) { struct block_device *bdev = I_BDEV(filp->f_mapping->host); - if (bdev->bd_holder == filp) - bd_release(bdev); - return blkdev_put(bdev, filp->f_mode); + blkdev_put(bdev, filp->f_mode); + return 0; } static long block_ioctl(struct file *file, unsigned cmd, unsigned long arg) @@ -1426,25 +1572,39 @@ static long block_ioctl(struct file *file, unsigned cmd, unsigned long arg) * Does not take i_mutex for the write and thus is not for general purpose * use. */ -ssize_t blkdev_aio_write(struct kiocb *iocb, const struct iovec *iov, - unsigned long nr_segs, loff_t pos) +ssize_t blkdev_write_iter(struct kiocb *iocb, struct iov_iter *from) { struct file *file = iocb->ki_filp; + struct blk_plug plug; ssize_t ret; - BUG_ON(iocb->ki_pos != pos); - - ret = __generic_file_aio_write(iocb, iov, nr_segs, &iocb->ki_pos); - if (ret > 0 || ret == -EIOCBQUEUED) { + blk_start_plug(&plug); + ret = __generic_file_write_iter(iocb, from); + if (ret > 0) { ssize_t err; - - err = generic_write_sync(file, pos, ret); - if (err < 0 && ret > 0) + err = generic_write_sync(file, iocb->ki_pos - ret, ret); + if (err < 0) ret = err; } + blk_finish_plug(&plug); return ret; } -EXPORT_SYMBOL_GPL(blkdev_aio_write); +EXPORT_SYMBOL_GPL(blkdev_write_iter); + +static ssize_t blkdev_read_iter(struct kiocb *iocb, struct iov_iter *to) +{ + struct file *file = iocb->ki_filp; + struct inode *bd_inode = file->f_mapping->host; + loff_t size = i_size_read(bd_inode); + loff_t pos = iocb->ki_pos; + + if (pos >= size) + return 0; + + size -= pos; + iov_iter_truncate(to, size); + return generic_file_read_iter(iocb, to); +} /* * Try to release a page associated with block device when the system @@ -1463,30 +1623,30 @@ static int blkdev_releasepage(struct page *page, gfp_t wait) static const struct address_space_operations def_blk_aops = { .readpage = blkdev_readpage, .writepage = blkdev_writepage, - .sync_page = block_sync_page, .write_begin = blkdev_write_begin, .write_end = blkdev_write_end, .writepages = generic_writepages, .releasepage = blkdev_releasepage, .direct_IO = blkdev_direct_IO, + .is_dirty_writeback = buffer_check_dirty_writeback, }; const struct file_operations def_blk_fops = { .open = blkdev_open, .release = blkdev_close, .llseek = block_llseek, - .read = do_sync_read, - .write = do_sync_write, - .aio_read = generic_file_aio_read, - .aio_write = blkdev_aio_write, + .read = new_sync_read, + .write = new_sync_write, + .read_iter = blkdev_read_iter, + .write_iter = blkdev_write_iter, .mmap = generic_file_mmap, - .fsync = block_fsync, + .fsync = blkdev_fsync, .unlocked_ioctl = block_ioctl, #ifdef CONFIG_COMPAT .compat_ioctl = compat_blkdev_ioctl, #endif .splice_read = generic_file_splice_read, - .splice_write = generic_file_splice_write, + .splice_write = iter_file_splice_write, }; int ioctl_by_bdev(struct block_device *bdev, unsigned cmd, unsigned long arg) @@ -1543,61 +1703,7 @@ fail: } EXPORT_SYMBOL(lookup_bdev); -/** - * open_bdev_exclusive - open a block device by name and set it up for use - * - * @path: special file representing the block device - * @mode: FMODE_... combination to pass be used - * @holder: owner for exclusion - * - * Open the blockdevice described by the special file at @path, claim it - * for the @holder. - */ -struct block_device *open_bdev_exclusive(const char *path, fmode_t mode, void *holder) -{ - struct block_device *bdev; - int error = 0; - - bdev = lookup_bdev(path); - if (IS_ERR(bdev)) - return bdev; - - error = blkdev_get(bdev, mode); - if (error) - return ERR_PTR(error); - error = -EACCES; - if ((mode & FMODE_WRITE) && bdev_read_only(bdev)) - goto blkdev_put; - error = bd_claim(bdev, holder); - if (error) - goto blkdev_put; - - return bdev; - -blkdev_put: - blkdev_put(bdev, mode); - return ERR_PTR(error); -} - -EXPORT_SYMBOL(open_bdev_exclusive); - -/** - * close_bdev_exclusive - close a blockdevice opened by open_bdev_exclusive() - * - * @bdev: blockdevice to close - * @mode: mode, must match that used to open. - * - * This is the counterpart to open_bdev_exclusive(). - */ -void close_bdev_exclusive(struct block_device *bdev, fmode_t mode) -{ - bd_release(bdev); - blkdev_put(bdev, mode); -} - -EXPORT_SYMBOL(close_bdev_exclusive); - -int __invalidate_device(struct block_device *bdev) +int __invalidate_device(struct block_device *bdev, bool kill_dirty) { struct super_block *sb = get_super(bdev); int res = 0; @@ -1610,10 +1716,46 @@ int __invalidate_device(struct block_device *bdev) * hold). */ shrink_dcache_sb(sb); - res = invalidate_inodes(sb); + res = invalidate_inodes(sb, kill_dirty); drop_super(sb); } invalidate_bdev(bdev); return res; } EXPORT_SYMBOL(__invalidate_device); + +void iterate_bdevs(void (*func)(struct block_device *, void *), void *arg) +{ + struct inode *inode, *old_inode = NULL; + + spin_lock(&inode_sb_list_lock); + list_for_each_entry(inode, &blockdev_superblock->s_inodes, i_sb_list) { + struct address_space *mapping = inode->i_mapping; + + spin_lock(&inode->i_lock); + if (inode->i_state & (I_FREEING|I_WILL_FREE|I_NEW) || + mapping->nrpages == 0) { + spin_unlock(&inode->i_lock); + continue; + } + __iget(inode); + spin_unlock(&inode->i_lock); + spin_unlock(&inode_sb_list_lock); + /* + * We hold a reference to 'inode' so it couldn't have been + * removed from s_inodes list while we dropped the + * inode_sb_list_lock. We cannot iput the inode now as we can + * be holding the last reference and we cannot iput it under + * inode_sb_list_lock. So we keep the reference and iput it + * later. + */ + iput(old_inode); + old_inode = inode; + + func(I_BDEV(inode), arg); + + spin_lock(&inode_sb_list_lock); + } + spin_unlock(&inode_sb_list_lock); + iput(old_inode); +} |