diff options
Diffstat (limited to 'fs')
118 files changed, 4835 insertions, 4574 deletions
diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c index c95295c6504..e83aa5ebe86 100644 --- a/fs/9p/vfs_inode.c +++ b/fs/9p/vfs_inode.c @@ -626,8 +626,7 @@ static struct dentry *v9fs_vfs_lookup(struct inode *dir, struct dentry *dentry, return NULL; error: - if (fid) - p9_client_clunk(fid); + p9_client_clunk(fid); return ERR_PTR(result); } diff --git a/fs/Kconfig b/fs/Kconfig index abccb5dab9a..f54a157a029 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -136,37 +136,51 @@ config EXT3_FS_SECURITY If you are not using a security module that requires using extended attributes for file security labels, say N. -config EXT4DEV_FS - tristate "Ext4dev/ext4 extended fs support development (EXPERIMENTAL)" - depends on EXPERIMENTAL +config EXT4_FS + tristate "The Extended 4 (ext4) filesystem" select JBD2 select CRC16 help - Ext4dev is a predecessor filesystem of the next generation - extended fs ext4, based on ext3 filesystem code. It will be - renamed ext4 fs later, once ext4dev is mature and stabilized. + This is the next generation of the ext3 filesystem. Unlike the change from ext2 filesystem to ext3 filesystem, - the on-disk format of ext4dev is not the same as ext3 any more: - it is based on extent maps and it supports 48-bit physical block - numbers. These combined on-disk format changes will allow - ext4dev/ext4 to handle more than 16 TB filesystem volumes -- - a hard limit that ext3 cannot overcome without changing the - on-disk format. - - Other than extent maps and 48-bit block numbers, ext4dev also is - likely to have other new features such as persistent preallocation, - high resolution time stamps, and larger file support etc. These - features will be added to ext4dev gradually. + the on-disk format of ext4 is not forwards compatible with + ext3; it is based on extent maps and it supports 48-bit + physical block numbers. The ext4 filesystem also supports delayed + allocation, persistent preallocation, high resolution time stamps, + and a number of other features to improve performance and speed + up fsck time. For more information, please see the web pages at + http://ext4.wiki.kernel.org. + + The ext4 filesystem will support mounting an ext3 + filesystem; while there will be some performance gains from + the delayed allocation and inode table readahead, the best + performance gains will require enabling ext4 features in the + filesystem, or formating a new filesystem as an ext4 + filesystem initially. To compile this file system support as a module, choose M here. The module will be called ext4dev. If unsure, say N. -config EXT4DEV_FS_XATTR - bool "Ext4dev extended attributes" - depends on EXT4DEV_FS +config EXT4DEV_COMPAT + bool "Enable ext4dev compatibility" + depends on EXT4_FS + help + Starting with 2.6.28, the name of the ext4 filesystem was + renamed from ext4dev to ext4. Unfortunately there are some + legacy userspace programs (such as klibc's fstype) have + "ext4dev" hardcoded. + + To enable backwards compatibility so that systems that are + still expecting to mount ext4 filesystems using ext4dev, + chose Y here. This feature will go away by 2.6.31, so + please arrange to get your userspace programs fixed! + +config EXT4_FS_XATTR + bool "Ext4 extended attributes" + depends on EXT4_FS default y help Extended attributes are name:value pairs associated with inodes by @@ -175,11 +189,11 @@ config EXT4DEV_FS_XATTR If unsure, say N. - You need this for POSIX ACL support on ext4dev/ext4. + You need this for POSIX ACL support on ext4. -config EXT4DEV_FS_POSIX_ACL - bool "Ext4dev POSIX Access Control Lists" - depends on EXT4DEV_FS_XATTR +config EXT4_FS_POSIX_ACL + bool "Ext4 POSIX Access Control Lists" + depends on EXT4_FS_XATTR select FS_POSIX_ACL help POSIX Access Control Lists (ACLs) support permissions for users and @@ -190,14 +204,14 @@ config EXT4DEV_FS_POSIX_ACL If you don't know what Access Control Lists are, say N -config EXT4DEV_FS_SECURITY - bool "Ext4dev Security Labels" - depends on EXT4DEV_FS_XATTR +config EXT4_FS_SECURITY + bool "Ext4 Security Labels" + depends on EXT4_FS_XATTR help Security labels support alternative access control models implemented by security modules like SELinux. This option enables an extended attribute handler for file security - labels in the ext4dev/ext4 filesystem. + labels in the ext4 filesystem. If you are not using a security module that requires using extended attributes for file security labels, say N. @@ -240,22 +254,22 @@ config JBD2 help This is a generic journaling layer for block devices that support both 32-bit and 64-bit block numbers. It is currently used by - the ext4dev/ext4 filesystem, but it could also be used to add + the ext4 filesystem, but it could also be used to add journal support to other file systems or block devices such as RAID or LVM. - If you are using ext4dev/ext4, you need to say Y here. If you are not - using ext4dev/ext4 then you will probably want to say N. + If you are using ext4, you need to say Y here. If you are not + using ext4 then you will probably want to say N. To compile this device as a module, choose M here. The module will be - called jbd2. If you are compiling ext4dev/ext4 into the kernel, + called jbd2. If you are compiling ext4 into the kernel, you cannot compile this code as a module. config JBD2_DEBUG - bool "JBD2 (ext4dev/ext4) debugging support" + bool "JBD2 (ext4) debugging support" depends on JBD2 && DEBUG_FS help - If you are using the ext4dev/ext4 journaled file system (or + If you are using the ext4 journaled file system (or potentially any other filesystem/device using JBD2), this option allows you to enable debugging output while the system is running, in order to help track down any problems you are having. @@ -270,9 +284,9 @@ config JBD2_DEBUG config FS_MBCACHE # Meta block cache for Extended Attributes (ext2/ext3/ext4) tristate - depends on EXT2_FS_XATTR || EXT3_FS_XATTR || EXT4DEV_FS_XATTR - default y if EXT2_FS=y || EXT3_FS=y || EXT4DEV_FS=y - default m if EXT2_FS=m || EXT3_FS=m || EXT4DEV_FS=m + depends on EXT2_FS_XATTR || EXT3_FS_XATTR || EXT4_FS_XATTR + default y if EXT2_FS=y || EXT3_FS=y || EXT4_FS=y + default m if EXT2_FS=m || EXT3_FS=m || EXT4_FS=m config REISERFS_FS tristate "Reiserfs support" diff --git a/fs/Makefile b/fs/Makefile index a1482a5eff1..de404b00eb0 100644 --- a/fs/Makefile +++ b/fs/Makefile @@ -69,7 +69,7 @@ obj-$(CONFIG_DLM) += dlm/ # Do not add any filesystems before this line obj-$(CONFIG_REISERFS_FS) += reiserfs/ obj-$(CONFIG_EXT3_FS) += ext3/ # Before ext2 so root fs can be ext3 -obj-$(CONFIG_EXT4DEV_FS) += ext4/ # Before ext2 so root fs can be ext4dev +obj-$(CONFIG_EXT4_FS) += ext4/ # Before ext2 so root fs can be ext4dev obj-$(CONFIG_JBD) += jbd/ obj-$(CONFIG_JBD2) += jbd2/ obj-$(CONFIG_EXT2_FS) += ext2/ diff --git a/fs/bfs/dir.c b/fs/bfs/dir.c index 87ee5ccee34..ed8feb052df 100644 --- a/fs/bfs/dir.c +++ b/fs/bfs/dir.c @@ -125,8 +125,8 @@ static int bfs_create(struct inode *dir, struct dentry *dentry, int mode, inode->i_ino); if (err) { inode_dec_link_count(inode); - iput(inode); mutex_unlock(&info->bfs_lock); + iput(inode); return err; } mutex_unlock(&info->bfs_lock); diff --git a/fs/bio-integrity.c b/fs/bio-integrity.c index c3e174b35fe..19caf7c962a 100644 --- a/fs/bio-integrity.c +++ b/fs/bio-integrity.c @@ -107,7 +107,8 @@ void bio_integrity_free(struct bio *bio, struct bio_set *bs) BUG_ON(bip == NULL); /* A cloned bio doesn't own the integrity metadata */ - if (!bio_flagged(bio, BIO_CLONED) && bip->bip_buf != NULL) + if (!bio_flagged(bio, BIO_CLONED) && !bio_flagged(bio, BIO_FS_INTEGRITY) + && bip->bip_buf != NULL) kfree(bip->bip_buf); mempool_free(bip->bip_vec, bs->bvec_pools[bip->bip_pool]); @@ -150,6 +151,24 @@ int bio_integrity_add_page(struct bio *bio, struct page *page, } EXPORT_SYMBOL(bio_integrity_add_page); +static int bdev_integrity_enabled(struct block_device *bdev, int rw) +{ + struct blk_integrity *bi = bdev_get_integrity(bdev); + + if (bi == NULL) + return 0; + + if (rw == READ && bi->verify_fn != NULL && + (bi->flags & INTEGRITY_FLAG_READ)) + return 1; + + if (rw == WRITE && bi->generate_fn != NULL && + (bi->flags & INTEGRITY_FLAG_WRITE)) + return 1; + + return 0; +} + /** * bio_integrity_enabled - Check whether integrity can be passed * @bio: bio to check @@ -313,6 +332,14 @@ static void bio_integrity_generate(struct bio *bio) } } +static inline unsigned short blk_integrity_tuple_size(struct blk_integrity *bi) +{ + if (bi) + return bi->tuple_size; + + return 0; +} + /** * bio_integrity_prep - Prepare bio for integrity I/O * @bio: bio to prepare @@ -30,7 +30,7 @@ static struct kmem_cache *bio_slab __read_mostly; -mempool_t *bio_split_pool __read_mostly; +static mempool_t *bio_split_pool __read_mostly; /* * if you change this list, also change bvec_alloc or things will @@ -60,25 +60,46 @@ struct bio_vec *bvec_alloc_bs(gfp_t gfp_mask, int nr, unsigned long *idx, struct struct bio_vec *bvl; /* - * see comment near bvec_array define! + * If 'bs' is given, lookup the pool and do the mempool alloc. + * If not, this is a bio_kmalloc() allocation and just do a + * kzalloc() for the exact number of vecs right away. */ - switch (nr) { - case 1 : *idx = 0; break; - case 2 ... 4: *idx = 1; break; - case 5 ... 16: *idx = 2; break; - case 17 ... 64: *idx = 3; break; - case 65 ... 128: *idx = 4; break; - case 129 ... BIO_MAX_PAGES: *idx = 5; break; + if (bs) { + /* + * see comment near bvec_array define! + */ + switch (nr) { + case 1: + *idx = 0; + break; + case 2 ... 4: + *idx = 1; + break; + case 5 ... 16: + *idx = 2; + break; + case 17 ... 64: + *idx = 3; + break; + case 65 ... 128: + *idx = 4; + break; + case 129 ... BIO_MAX_PAGES: + *idx = 5; + break; default: return NULL; - } - /* - * idx now points to the pool we want to allocate from - */ + } - bvl = mempool_alloc(bs->bvec_pools[*idx], gfp_mask); - if (bvl) - memset(bvl, 0, bvec_nr_vecs(*idx) * sizeof(struct bio_vec)); + /* + * idx now points to the pool we want to allocate from + */ + bvl = mempool_alloc(bs->bvec_pools[*idx], gfp_mask); + if (bvl) + memset(bvl, 0, + bvec_nr_vecs(*idx) * sizeof(struct bio_vec)); + } else + bvl = kzalloc(nr * sizeof(struct bio_vec), gfp_mask); return bvl; } @@ -107,10 +128,17 @@ static void bio_fs_destructor(struct bio *bio) bio_free(bio, fs_bio_set); } +static void bio_kmalloc_destructor(struct bio *bio) +{ + kfree(bio->bi_io_vec); + kfree(bio); +} + void bio_init(struct bio *bio) { memset(bio, 0, sizeof(*bio)); bio->bi_flags = 1 << BIO_UPTODATE; + bio->bi_comp_cpu = -1; atomic_set(&bio->bi_cnt, 1); } @@ -118,19 +146,25 @@ void bio_init(struct bio *bio) * bio_alloc_bioset - allocate a bio for I/O * @gfp_mask: the GFP_ mask given to the slab allocator * @nr_iovecs: number of iovecs to pre-allocate - * @bs: the bio_set to allocate from + * @bs: the bio_set to allocate from. If %NULL, just use kmalloc * * Description: - * bio_alloc_bioset will first try it's on mempool to satisfy the allocation. + * bio_alloc_bioset will first try its own mempool to satisfy the allocation. * If %__GFP_WAIT is set then we will block on the internal pool waiting - * for a &struct bio to become free. + * for a &struct bio to become free. If a %NULL @bs is passed in, we will + * fall back to just using @kmalloc to allocate the required memory. * * allocate bio and iovecs from the memory pools specified by the - * bio_set structure. + * bio_set structure, or @kmalloc if none given. **/ struct bio *bio_alloc_bioset(gfp_t gfp_mask, int nr_iovecs, struct bio_set *bs) { - struct bio *bio = mempool_alloc(bs->bio_pool, gfp_mask); + struct bio *bio; + + if (bs) + bio = mempool_alloc(bs->bio_pool, gfp_mask); + else + bio = kmalloc(sizeof(*bio), gfp_mask); if (likely(bio)) { struct bio_vec *bvl = NULL; @@ -141,7 +175,10 @@ struct bio *bio_alloc_bioset(gfp_t gfp_mask, int nr_iovecs, struct bio_set *bs) bvl = bvec_alloc_bs(gfp_mask, nr_iovecs, &idx, bs); if (unlikely(!bvl)) { - mempool_free(bio, bs->bio_pool); + if (bs) + mempool_free(bio, bs->bio_pool); + else + kfree(bio); bio = NULL; goto out; } @@ -164,6 +201,23 @@ struct bio *bio_alloc(gfp_t gfp_mask, int nr_iovecs) return bio; } +/* + * Like bio_alloc(), but doesn't use a mempool backing. This means that + * it CAN fail, but while bio_alloc() can only be used for allocations + * that have a short (finite) life span, bio_kmalloc() should be used + * for more permanent bio allocations (like allocating some bio's for + * initalization or setup purposes). + */ +struct bio *bio_kmalloc(gfp_t gfp_mask, int nr_iovecs) +{ + struct bio *bio = bio_alloc_bioset(gfp_mask, nr_iovecs, NULL); + + if (bio) + bio->bi_destructor = bio_kmalloc_destructor; + + return bio; +} + void zero_fill_bio(struct bio *bio) { unsigned long flags; @@ -208,14 +262,6 @@ inline int bio_phys_segments(struct request_queue *q, struct bio *bio) return bio->bi_phys_segments; } -inline int bio_hw_segments(struct request_queue *q, struct bio *bio) -{ - if (unlikely(!bio_flagged(bio, BIO_SEG_VALID))) - blk_recount_segments(q, bio); - - return bio->bi_hw_segments; -} - /** * __bio_clone - clone a bio * @bio: destination bio @@ -350,8 +396,7 @@ static int __bio_add_page(struct request_queue *q, struct bio *bio, struct page */ while (bio->bi_phys_segments >= q->max_phys_segments - || bio->bi_hw_segments >= q->max_hw_segments - || BIOVEC_VIRT_OVERSIZE(bio->bi_size)) { + || bio->bi_phys_segments >= q->max_hw_segments) { if (retried_segments) return 0; @@ -395,13 +440,11 @@ static int __bio_add_page(struct request_queue *q, struct bio *bio, struct page } /* If we may be able to merge these biovecs, force a recount */ - if (bio->bi_vcnt && (BIOVEC_PHYS_MERGEABLE(bvec-1, bvec) || - BIOVEC_VIRT_MERGEABLE(bvec-1, bvec))) + if (bio->bi_vcnt && (BIOVEC_PHYS_MERGEABLE(bvec-1, bvec))) bio->bi_flags &= ~(1 << BIO_SEG_VALID); bio->bi_vcnt++; bio->bi_phys_segments++; - bio->bi_hw_segments++; done: bio->bi_size += len; return len; @@ -449,16 +492,19 @@ int bio_add_page(struct bio *bio, struct page *page, unsigned int len, struct bio_map_data { struct bio_vec *iovecs; - int nr_sgvecs; struct sg_iovec *sgvecs; + int nr_sgvecs; + int is_our_pages; }; static void bio_set_map_data(struct bio_map_data *bmd, struct bio *bio, - struct sg_iovec *iov, int iov_count) + struct sg_iovec *iov, int iov_count, + int is_our_pages) { memcpy(bmd->iovecs, bio->bi_io_vec, sizeof(struct bio_vec) * bio->bi_vcnt); memcpy(bmd->sgvecs, iov, sizeof(struct sg_iovec) * iov_count); bmd->nr_sgvecs = iov_count; + bmd->is_our_pages = is_our_pages; bio->bi_private = bmd; } @@ -493,7 +539,8 @@ static struct bio_map_data *bio_alloc_map_data(int nr_segs, int iov_count, } static int __bio_copy_iov(struct bio *bio, struct bio_vec *iovecs, - struct sg_iovec *iov, int iov_count, int uncopy) + struct sg_iovec *iov, int iov_count, int uncopy, + int do_free_page) { int ret = 0, i; struct bio_vec *bvec; @@ -536,7 +583,7 @@ static int __bio_copy_iov(struct bio *bio, struct bio_vec *iovecs, } } - if (uncopy) + if (do_free_page) __free_page(bvec->bv_page); } @@ -553,10 +600,11 @@ static int __bio_copy_iov(struct bio *bio, struct bio_vec *iovecs, int bio_uncopy_user(struct bio *bio) { struct bio_map_data *bmd = bio->bi_private; - int ret; - - ret = __bio_copy_iov(bio, bmd->iovecs, bmd->sgvecs, bmd->nr_sgvecs, 1); + int ret = 0; + if (!bio_flagged(bio, BIO_NULL_MAPPED)) + ret = __bio_copy_iov(bio, bmd->iovecs, bmd->sgvecs, + bmd->nr_sgvecs, 1, bmd->is_our_pages); bio_free_map_data(bmd); bio_put(bio); return ret; @@ -565,16 +613,20 @@ int bio_uncopy_user(struct bio *bio) /** * bio_copy_user_iov - copy user data to bio * @q: destination block queue + * @map_data: pointer to the rq_map_data holding pages (if necessary) * @iov: the iovec. * @iov_count: number of elements in the iovec * @write_to_vm: bool indicating writing to pages or not + * @gfp_mask: memory allocation flags * * Prepares and returns a bio for indirect user io, bouncing data * to/from kernel pages as necessary. Must be paired with * call bio_uncopy_user() on io completion. */ -struct bio *bio_copy_user_iov(struct request_queue *q, struct sg_iovec *iov, - int iov_count, int write_to_vm) +struct bio *bio_copy_user_iov(struct request_queue *q, + struct rq_map_data *map_data, + struct sg_iovec *iov, int iov_count, + int write_to_vm, gfp_t gfp_mask) { struct bio_map_data *bmd; struct bio_vec *bvec; @@ -597,25 +649,38 @@ struct bio *bio_copy_user_iov(struct request_queue *q, struct sg_iovec *iov, len += iov[i].iov_len; } - bmd = bio_alloc_map_data(nr_pages, iov_count, GFP_KERNEL); + bmd = bio_alloc_map_data(nr_pages, iov_count, gfp_mask); if (!bmd) return ERR_PTR(-ENOMEM); ret = -ENOMEM; - bio = bio_alloc(GFP_KERNEL, nr_pages); + bio = bio_alloc(gfp_mask, nr_pages); if (!bio) goto out_bmd; bio->bi_rw |= (!write_to_vm << BIO_RW); ret = 0; + i = 0; while (len) { - unsigned int bytes = PAGE_SIZE; + unsigned int bytes; + + if (map_data) + bytes = 1U << (PAGE_SHIFT + map_data->page_order); + else + bytes = PAGE_SIZE; if (bytes > len) bytes = len; - page = alloc_page(q->bounce_gfp | GFP_KERNEL); + if (map_data) { + if (i == map_data->nr_entries) { + ret = -ENOMEM; + break; + } + page = map_data->pages[i++]; + } else + page = alloc_page(q->bounce_gfp | gfp_mask); if (!page) { ret = -ENOMEM; break; @@ -634,16 +699,17 @@ struct bio *bio_copy_user_iov(struct request_queue *q, struct sg_iovec *iov, * success */ if (!write_to_vm) { - ret = __bio_copy_iov(bio, bio->bi_io_vec, iov, iov_count, 0); + ret = __bio_copy_iov(bio, bio->bi_io_vec, iov, iov_count, 0, 0); if (ret) goto cleanup; } - bio_set_map_data(bmd, bio, iov, iov_count); + bio_set_map_data(bmd, bio, iov, iov_count, map_data ? 0 : 1); return bio; cleanup: - bio_for_each_segment(bvec, bio, i) - __free_page(bvec->bv_page); + if (!map_data) + bio_for_each_segment(bvec, bio, i) + __free_page(bvec->bv_page); bio_put(bio); out_bmd: @@ -654,29 +720,32 @@ out_bmd: /** * bio_copy_user - copy user data to bio * @q: destination block queue + * @map_data: pointer to the rq_map_data holding pages (if necessary) * @uaddr: start of user address * @len: length in bytes * @write_to_vm: bool indicating writing to pages or not + * @gfp_mask: memory allocation flags * * Prepares and returns a bio for indirect user io, bouncing data * to/from kernel pages as necessary. Must be paired with * call bio_uncopy_user() on io completion. */ -struct bio *bio_copy_user(struct request_queue *q, unsigned long uaddr, - unsigned int len, int write_to_vm) +struct bio *bio_copy_user(struct request_queue *q, struct rq_map_data *map_data, + unsigned long uaddr, unsigned int len, + int write_to_vm, gfp_t gfp_mask) { struct sg_iovec iov; iov.iov_base = (void __user *)uaddr; iov.iov_len = len; - return bio_copy_user_iov(q, &iov, 1, write_to_vm); + return bio_copy_user_iov(q, map_data, &iov, 1, write_to_vm, gfp_mask); } static struct bio *__bio_map_user_iov(struct request_queue *q, struct block_device *bdev, struct sg_iovec *iov, int iov_count, - int write_to_vm) + int write_to_vm, gfp_t gfp_mask) { int i, j; int nr_pages = 0; @@ -702,12 +771,12 @@ static struct bio *__bio_map_user_iov(struct request_queue *q, if (!nr_pages) return ERR_PTR(-EINVAL); - bio = bio_alloc(GFP_KERNEL, nr_pages); + bio = bio_alloc(gfp_mask, nr_pages); if (!bio) return ERR_PTR(-ENOMEM); ret = -ENOMEM; - pages = kcalloc(nr_pages, sizeof(struct page *), GFP_KERNEL); + pages = kcalloc(nr_pages, sizeof(struct page *), gfp_mask); if (!pages) goto out; @@ -786,19 +855,21 @@ static struct bio *__bio_map_user_iov(struct request_queue *q, * @uaddr: start of user address * @len: length in bytes * @write_to_vm: bool indicating writing to pages or not + * @gfp_mask: memory allocation flags * * Map the user space address into a bio suitable for io to a block * device. Returns an error pointer in case of error. */ struct bio *bio_map_user(struct request_queue *q, struct block_device *bdev, - unsigned long uaddr, unsigned int len, int write_to_vm) + unsigned long uaddr, unsigned int len, int write_to_vm, + gfp_t gfp_mask) { struct sg_iovec iov; iov.iov_base = (void __user *)uaddr; iov.iov_len = len; - return bio_map_user_iov(q, bdev, &iov, 1, write_to_vm); + return bio_map_user_iov(q, bdev, &iov, 1, write_to_vm, gfp_mask); } /** @@ -808,18 +879,19 @@ struct bio *bio_map_user(struct request_queue *q, struct block_device *bdev, * @iov: the iovec. * @iov_count: number of elements in the iovec * @write_to_vm: bool indicating writing to pages or not + * @gfp_mask: memory allocation flags * * Map the user space address into a bio suitable for io to a block * device. Returns an error pointer in case of error. */ struct bio *bio_map_user_iov(struct request_queue *q, struct block_device *bdev, struct sg_iovec *iov, int iov_count, - int write_to_vm) + int write_to_vm, gfp_t gfp_mask) { struct bio *bio; - bio = __bio_map_user_iov(q, bdev, iov, iov_count, write_to_vm); - + bio = __bio_map_user_iov(q, bdev, iov, iov_count, write_to_vm, + gfp_mask); if (IS_ERR(bio)) return bio; @@ -976,48 +1048,13 @@ static void bio_copy_kern_endio(struct bio *bio, int err) struct bio *bio_copy_kern(struct request_queue *q, void *data, unsigned int len, gfp_t gfp_mask, int reading) { - unsigned long kaddr = (unsigned long)data; - unsigned long end = (kaddr + len + PAGE_SIZE - 1) >> PAGE_SHIFT; - unsigned long start = kaddr >> PAGE_SHIFT; - const int nr_pages = end - start; struct bio *bio; struct bio_vec *bvec; - struct bio_map_data *bmd; - int i, ret; - struct sg_iovec iov; - - iov.iov_base = data; - iov.iov_len = len; - - bmd = bio_alloc_map_data(nr_pages, 1, gfp_mask); - if (!bmd) - return ERR_PTR(-ENOMEM); - - ret = -ENOMEM; - bio = bio_alloc(gfp_mask, nr_pages); - if (!bio) - goto out_bmd; - - while (len) { - struct page *page; - unsigned int bytes = PAGE_SIZE; - - if (bytes > len) - bytes = len; - - page = alloc_page(q->bounce_gfp | gfp_mask); - if (!page) { - ret = -ENOMEM; - goto cleanup; - } - - if (bio_add_pc_page(q, bio, page, bytes, 0) < bytes) { - ret = -EINVAL; - goto cleanup; - } + int i; - len -= bytes; - } + bio = bio_copy_user(q, NULL, (unsigned long)data, len, 1, gfp_mask); + if (IS_ERR(bio)) + return bio; if (!reading) { void *p = data; @@ -1030,20 +1067,9 @@ struct bio *bio_copy_kern(struct request_queue *q, void *data, unsigned int len, } } - bio->bi_private = bmd; bio->bi_end_io = bio_copy_kern_endio; - bio_set_map_data(bmd, bio, &iov, 1); return bio; -cleanup: - bio_for_each_segment(bvec, bio, i) - __free_page(bvec->bv_page); - - bio_put(bio); -out_bmd: - bio_free_map_data(bmd); - - return ERR_PTR(ret); } /* @@ -1230,9 +1256,9 @@ static void bio_pair_end_2(struct bio *bi, int err) * split a bio - only worry about a bio with a single page * in it's iovec */ -struct bio_pair *bio_split(struct bio *bi, mempool_t *pool, int first_sectors) +struct bio_pair *bio_split(struct bio *bi, int first_sectors) { - struct bio_pair *bp = mempool_alloc(pool, GFP_NOIO); + struct bio_pair *bp = mempool_alloc(bio_split_pool, GFP_NOIO); if (!bp) return bp; @@ -1266,7 +1292,7 @@ struct bio_pair *bio_split(struct bio *bi, mempool_t *pool, int first_sectors) bp->bio2.bi_end_io = bio_pair_end_2; bp->bio1.bi_private = bi; - bp->bio2.bi_private = pool; + bp->bio2.bi_private = bio_split_pool; if (bio_integrity(bi)) bio_integrity_split(bi, bp, first_sectors); @@ -1274,6 +1300,42 @@ struct bio_pair *bio_split(struct bio *bi, mempool_t *pool, int first_sectors) return bp; } +/** + * bio_sector_offset - Find hardware sector offset in bio + * @bio: bio to inspect + * @index: bio_vec index + * @offset: offset in bv_page + * + * Return the number of hardware sectors between beginning of bio + * and an end point indicated by a bio_vec index and an offset + * within that vector's page. + */ +sector_t bio_sector_offset(struct bio *bio, unsigned short index, + unsigned int offset) +{ + unsigned int sector_sz = queue_hardsect_size(bio->bi_bdev->bd_disk->queue); + struct bio_vec *bv; + sector_t sectors; + int i; + + sectors = 0; + + if (index >= bio->bi_idx) + index = bio->bi_vcnt - 1; + + __bio_for_each_segment(bv, bio, i, 0) { + if (i == index) { + if (offset > bv->bv_offset) + sectors += (offset - bv->bv_offset) / sector_sz; + break; + } + + sectors += bv->bv_len / sector_sz; + } + + return sectors; +} +EXPORT_SYMBOL(bio_sector_offset); /* * create memory pools for biovec's in a bio_set. @@ -1376,6 +1438,7 @@ static int __init init_bio(void) subsys_initcall(init_bio); EXPORT_SYMBOL(bio_alloc); +EXPORT_SYMBOL(bio_kmalloc); EXPORT_SYMBOL(bio_put); EXPORT_SYMBOL(bio_free); EXPORT_SYMBOL(bio_endio); @@ -1383,7 +1446,6 @@ EXPORT_SYMBOL(bio_init); EXPORT_SYMBOL(__bio_clone); EXPORT_SYMBOL(bio_clone); EXPORT_SYMBOL(bio_phys_segments); -EXPORT_SYMBOL(bio_hw_segments); EXPORT_SYMBOL(bio_add_page); EXPORT_SYMBOL(bio_add_pc_page); EXPORT_SYMBOL(bio_get_nr_vecs); @@ -1393,7 +1455,6 @@ EXPORT_SYMBOL(bio_map_kern); EXPORT_SYMBOL(bio_copy_kern); EXPORT_SYMBOL(bio_pair_release); EXPORT_SYMBOL(bio_split); -EXPORT_SYMBOL(bio_split_pool); EXPORT_SYMBOL(bio_copy_user); EXPORT_SYMBOL(bio_uncopy_user); EXPORT_SYMBOL(bioset_create); diff --git a/fs/block_dev.c b/fs/block_dev.c index aff54219e04..d84f0469a01 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -540,22 +540,6 @@ EXPORT_SYMBOL(bd_release); * /sys/block/sda/holders/dm-0 --> /sys/block/dm-0 */ -static struct kobject *bdev_get_kobj(struct block_device *bdev) -{ - if (bdev->bd_contains != bdev) - return kobject_get(&bdev->bd_part->dev.kobj); - else - return kobject_get(&bdev->bd_disk->dev.kobj); -} - -static struct kobject *bdev_get_holder(struct block_device *bdev) -{ - if (bdev->bd_contains != bdev) - return kobject_get(bdev->bd_part->holder_dir); - else - return kobject_get(bdev->bd_disk->holder_dir); -} - static int add_symlink(struct kobject *from, struct kobject *to) { if (!from || !to) @@ -604,11 +588,11 @@ static int bd_holder_grab_dirs(struct block_device *bdev, if (!bo->hdev) goto fail_put_sdir; - bo->sdev = bdev_get_kobj(bdev); + bo->sdev = kobject_get(&part_to_dev(bdev->bd_part)->kobj); if (!bo->sdev) goto fail_put_hdev; - bo->hdir = bdev_get_holder(bdev); + bo->hdir = kobject_get(bdev->bd_part->holder_dir); if (!bo->hdir) goto fail_put_sdev; @@ -868,6 +852,87 @@ struct block_device *open_by_devnum(dev_t dev, unsigned mode) EXPORT_SYMBOL(open_by_devnum); +/** + * flush_disk - invalidates all buffer-cache entries on a disk + * + * @bdev: struct block device to be flushed + * + * 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) +{ + if (__invalidate_device(bdev)) { + char name[BDEVNAME_SIZE] = ""; + + if (bdev->bd_disk) + disk_name(bdev->bd_disk, 0, name); + printk(KERN_WARNING "VFS: busy inodes on changed media or " + "resized disk %s\n", name); + } + + if (!bdev->bd_disk) + return; + if (disk_partitionable(bdev->bd_disk)) + bdev->bd_invalidated = 1; +} + +/** + * check_disk_size_change - checks for disk size change and adjusts bdev size. + * @disk: struct gendisk to check + * @bdev: struct bdev to adjust. + * + * This routine checks to see if the bdev size does not match the disk size + * and adjusts it if it differs. + */ +void check_disk_size_change(struct gendisk *disk, struct block_device *bdev) +{ + loff_t disk_size, bdev_size; + + disk_size = (loff_t)get_capacity(disk) << 9; + bdev_size = i_size_read(bdev->bd_inode); + if (disk_size != bdev_size) { + char name[BDEVNAME_SIZE]; + + disk_name(disk, 0, name); + printk(KERN_INFO + "%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); + } +} +EXPORT_SYMBOL(check_disk_size_change); + +/** + * revalidate_disk - wrapper for lower-level driver's revalidate_disk call-back + * @disk: struct gendisk to be revalidated + * + * This routine is a wrapper for lower-level driver's revalidate_disk + * call-backs. It is used to do common pre and post operations needed + * for all revalidate_disk operations. + */ +int revalidate_disk(struct gendisk *disk) +{ + struct block_device *bdev; + int ret = 0; + + if (disk->fops->revalidate_disk) + ret = disk->fops->revalidate_disk(disk); + + bdev = bdget_disk(disk, 0); + if (!bdev) + return ret; + + mutex_lock(&bdev->bd_mutex); + check_disk_size_change(disk, bdev); + mutex_unlock(&bdev->bd_mutex); + bdput(bdev); + return ret; +} +EXPORT_SYMBOL(revalidate_disk); + /* * This routine checks whether a removable media has been changed, * and invalidates all buffer-cache-entries in that case. This @@ -887,13 +952,9 @@ int check_disk_change(struct block_device *bdev) if (!bdops->media_changed(bdev->bd_disk)) return 0; - if (__invalidate_device(bdev)) - printk("VFS: busy inodes on changed media.\n"); - + flush_disk(bdev); if (bdops->revalidate_disk) bdops->revalidate_disk(bdev->bd_disk); - if (bdev->bd_disk->minors > 1) - bdev->bd_invalidated = 1; return 1; } @@ -927,10 +988,10 @@ static int __blkdev_put(struct block_device *bdev, int for_part); static int do_open(struct block_device *bdev, struct file *file, int for_part) { - struct module *owner = NULL; struct gendisk *disk; + struct hd_struct *part = NULL; int ret; - int part; + int partno; int perm = 0; if (file->f_mode & FMODE_READ) @@ -948,25 +1009,27 @@ static int do_open(struct block_device *bdev, struct file *file, int for_part) ret = -ENXIO; file->f_mapping = bdev->bd_inode->i_mapping; + lock_kernel(); - disk = get_gendisk(bdev->bd_dev, &part); - if (!disk) { - unlock_kernel(); - bdput(bdev); - return ret; - } - owner = disk->fops->owner; + + disk = get_gendisk(bdev->bd_dev, &partno); + if (!disk) + goto out_unlock_kernel; + part = disk_get_part(disk, partno); + if (!part) + goto out_unlock_kernel; mutex_lock_nested(&bdev->bd_mutex, for_part); if (!bdev->bd_openers) { bdev->bd_disk = disk; + bdev->bd_part = part; bdev->bd_contains = bdev; - if (!part) { + if (!partno) { struct backing_dev_info *bdi; if (disk->fops->open) { ret = disk->fops->open(bdev->bd_inode, file); if (ret) - goto out_first; + goto out_clear; } if (!bdev->bd_openers) { bd_set_size(bdev,(loff_t)get_capacity(disk)<<9); @@ -978,36 +1041,36 @@ static int do_open(struct block_device *bdev, struct file *file, int for_part) if (bdev->bd_invalidated) rescan_partitions(disk, bdev); } else { - struct hd_struct *p; struct block_device *whole; whole = bdget_disk(disk, 0); ret = -ENOMEM; if (!whole) - goto out_first; + goto out_clear; BUG_ON(for_part); ret = __blkdev_get(whole, file->f_mode, file->f_flags, 1); if (ret) - goto out_first; + goto out_clear; bdev->bd_contains = whole; - p = disk->part[part - 1]; bdev->bd_inode->i_data.backing_dev_info = whole->bd_inode->i_data.backing_dev_info; - if (!(disk->flags & GENHD_FL_UP) || !p || !p->nr_sects) { + if (!(disk->flags & GENHD_FL_UP) || + !part || !part->nr_sects) { ret = -ENXIO; - goto out_first; + goto out_clear; } - kobject_get(&p->dev.kobj); - bdev->bd_part = p; - bd_set_size(bdev, (loff_t) p->nr_sects << 9); + bd_set_size(bdev, (loff_t)part->nr_sects << 9); } } else { + disk_put_part(part); put_disk(disk); - module_put(owner); + module_put(disk->fops->owner); + part = NULL; + disk = NULL; if (bdev->bd_contains == bdev) { if (bdev->bd_disk->fops->open) { ret = bdev->bd_disk->fops->open(bdev->bd_inode, file); if (ret) - goto out; + goto out_unlock_bdev; } if (bdev->bd_invalidated) rescan_partitions(bdev->bd_disk, bdev); @@ -1020,19 +1083,24 @@ static int do_open(struct block_device *bdev, struct file *file, int for_part) unlock_kernel(); return 0; -out_first: + out_clear: bdev->bd_disk = NULL; + bdev->bd_part = NULL; bdev->bd_inode->i_data.backing_dev_info = &default_backing_dev_info; if (bdev != bdev->bd_contains) __blkdev_put(bdev->bd_contains, 1); bdev->bd_contains = NULL; - put_disk(disk); - module_put(owner); -out: + out_unlock_bdev: mutex_unlock(&bdev->bd_mutex); + out_unlock_kernel: unlock_kernel(); - if (ret) - bdput(bdev); + + disk_put_part(part); + if (disk) + module_put(disk->fops->owner); + put_disk(disk); + bdput(bdev); + return ret; } @@ -1117,11 +1185,8 @@ static int __blkdev_put(struct block_device *bdev, int for_part) put_disk(disk); module_put(owner); - - if (bdev->bd_contains != bdev) { - kobject_put(&bdev->bd_part->dev.kobj); - bdev->bd_part = NULL; - } + 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) @@ -1197,10 +1262,9 @@ EXPORT_SYMBOL(ioctl_by_bdev); /** * lookup_bdev - lookup a struct block_device by name + * @pathname: special file representing the block device * - * @path: special file representing the block device - * - * Get a reference to the blockdevice at @path in the current + * Get a reference to the blockdevice at @pathname in the current * namespace if possible and return it. Return ERR_PTR(error) * otherwise. */ diff --git a/fs/cifs/cifs_spnego.c b/fs/cifs/cifs_spnego.c index 117ef4bba68..fcee9298b62 100644 --- a/fs/cifs/cifs_spnego.c +++ b/fs/cifs/cifs_spnego.c @@ -66,11 +66,28 @@ struct key_type cifs_spnego_key_type = { .describe = user_describe, }; -#define MAX_VER_STR_LEN 8 /* length of longest version string e.g. - strlen("ver=0xFF") */ -#define MAX_MECH_STR_LEN 13 /* length of longest security mechanism name, eg - in future could have strlen(";sec=ntlmsspi") */ -#define MAX_IPV6_ADDR_LEN 42 /* eg FEDC:BA98:7654:3210:FEDC:BA98:7654:3210/60 */ +/* length of longest version string e.g. strlen("ver=0xFF") */ +#define MAX_VER_STR_LEN 8 + +/* length of longest security mechanism name, eg in future could have + * strlen(";sec=ntlmsspi") */ +#define MAX_MECH_STR_LEN 13 + +/* max possible addr len eg FEDC:BA98:7654:3210:FEDC:BA98:7654:3210/60 */ +#define MAX_IPV6_ADDR_LEN 42 + +/* strlen of "host=" */ +#define HOST_KEY_LEN 5 + +/* strlen of ";ip4=" or ";ip6=" */ +#define IP_KEY_LEN 5 + +/* strlen of ";uid=0x" */ +#define UID_KEY_LEN 7 + +/* strlen of ";user=" */ +#define USER_KEY_LEN 6 + /* get a key struct with a SPNEGO security blob, suitable for session setup */ struct key * cifs_get_spnego_key(struct cifsSesInfo *sesInfo) @@ -84,11 +101,11 @@ cifs_get_spnego_key(struct cifsSesInfo *sesInfo) /* length of fields (with semicolons): ver=0xyz ip4=ipaddress host=hostname sec=mechanism uid=0xFF user=username */ desc_len = MAX_VER_STR_LEN + - 6 /* len of "host=" */ + strlen(hostname) + - 5 /* len of ";ipv4=" */ + MAX_IPV6_ADDR_LEN + + HOST_KEY_LEN + strlen(hostname) + + IP_KEY_LEN + MAX_IPV6_ADDR_LEN + MAX_MECH_STR_LEN + - 7 /* len of ";uid=0x" */ + (sizeof(uid_t) * 2) + - 6 /* len of ";user=" */ + strlen(sesInfo->userName) + 1; + UID_KEY_LEN + (sizeof(uid_t) * 2) + + USER_KEY_LEN + strlen(sesInfo->userName) + 1; spnego_key = ERR_PTR(-ENOMEM); description = kzalloc(desc_len, GFP_KERNEL); diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h index 135c965c413..f7b4a5cd837 100644 --- a/fs/cifs/cifsfs.h +++ b/fs/cifs/cifsfs.h @@ -41,7 +41,7 @@ extern int cifs_create(struct inode *, struct dentry *, int, struct nameidata *); extern struct dentry *cifs_lookup(struct inode *, struct dentry *, struct nameidata *); -extern int cifs_unlink(struct inode *, struct dentry *); +extern int cifs_unlink(struct inode *dir, struct dentry *dentry); extern int cifs_hardlink(struct dentry *, struct inode *, struct dentry *); extern int cifs_mknod(struct inode *, struct dentry *, int, dev_t); extern int cifs_mkdir(struct inode *, struct dentry *, int); diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index 8dfd6f24d48..0d22479d99b 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -309,6 +309,7 @@ struct cifs_search_info { __u32 resume_key; char *ntwrk_buf_start; char *srch_entries_start; + char *last_entry; char *presume_name; unsigned int resume_name_len; bool endOfSearch:1; diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h index a729d083e6f..0cff7fe986e 100644 --- a/fs/cifs/cifsproto.h +++ b/fs/cifs/cifsproto.h @@ -179,6 +179,8 @@ extern int CIFSSMBSetPathInfo(const int xid, struct cifsTconInfo *tcon, extern int CIFSSMBSetFileInfo(const int xid, struct cifsTconInfo *tcon, const FILE_BASIC_INFO *data, __u16 fid, __u32 pid_of_opener); +extern int CIFSSMBSetFileDisposition(const int xid, struct cifsTconInfo *tcon, + bool delete_file, __u16 fid, __u32 pid_of_opener); #if 0 extern int CIFSSMBSetAttrLegacy(int xid, struct cifsTconInfo *tcon, char *fileName, __u16 dos_attributes, @@ -229,7 +231,7 @@ extern int CIFSSMBRename(const int xid, struct cifsTconInfo *tcon, const struct nls_table *nls_codepage, int remap_special_chars); extern int CIFSSMBRenameOpenFile(const int xid, struct cifsTconInfo *pTcon, - int netfid, char *target_name, + int netfid, const char *target_name, const struct nls_table *nls_codepage, int remap_special_chars); extern int CIFSCreateHardLink(const int xid, diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index 994de7c9047..6f4ffe15d68 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -2017,7 +2017,7 @@ renameRetry: } int CIFSSMBRenameOpenFile(const int xid, struct cifsTconInfo *pTcon, - int netfid, char *target_name, + int netfid, const char *target_name, const struct nls_table *nls_codepage, int remap) { struct smb_com_transaction2_sfi_req *pSMB = NULL; @@ -2071,7 +2071,7 @@ int CIFSSMBRenameOpenFile(const int xid, struct cifsTconInfo *pTcon, remap); } rename_info->target_name_len = cpu_to_le32(2 * len_of_str); - count = 12 /* sizeof(struct set_file_rename) */ + (2 * len_of_str) + 2; + count = 12 /* sizeof(struct set_file_rename) */ + (2 * len_of_str); byte_count += count; pSMB->DataCount = cpu_to_le16(count); pSMB->TotalDataCount = pSMB->DataCount; @@ -3614,6 +3614,8 @@ findFirstRetry: /* BB remember to free buffer if error BB */ rc = validate_t2((struct smb_t2_rsp *)pSMBr); if (rc == 0) { + unsigned int lnoff; + if (pSMBr->hdr.Flags2 & SMBFLG2_UNICODE) psrch_inf->unicode = true; else @@ -3636,6 +3638,17 @@ findFirstRetry: le16_to_cpu(parms->SearchCount); psrch_inf->index_of_last_entry = 2 /* skip . and .. */ + psrch_inf->entries_in_buffer; + lnoff = le16_to_cpu(parms->LastNameOffset); + if (tcon->ses->server->maxBuf - MAX_CIFS_HDR_SIZE < + lnoff) { + cERROR(1, ("ignoring corrupt resume name")); + psrch_inf->last_entry = NULL; + return rc; + } + + psrch_inf->last_entry = psrch_inf->srch_entries_start + + lnoff; + *pnetfid = parms->SearchHandle; } else { cifs_buf_release(pSMB); @@ -3725,6 +3738,8 @@ int CIFSFindNext(const int xid, struct cifsTconInfo *tcon, rc = validate_t2((struct smb_t2_rsp *)pSMBr); if (rc == 0) { + unsigned int lnoff; + /* BB fixme add lock for file (srch_info) struct here */ if (pSMBr->hdr.Flags2 & SMBFLG2_UNICODE) psrch_inf->unicode = true; @@ -3751,6 +3766,16 @@ int CIFSFindNext(const int xid, struct cifsTconInfo *tcon, le16_to_cpu(parms->SearchCount); psrch_inf->index_of_last_entry += psrch_inf->entries_in_buffer; + lnoff = le16_to_cpu(parms->LastNameOffset); + if (tcon->ses->server->maxBuf - MAX_CIFS_HDR_SIZE < + lnoff) { + cERROR(1, ("ignoring corrupt resume name")); + psrch_inf->last_entry = NULL; + return rc; + } else + psrch_inf->last_entry = + psrch_inf->srch_entries_start + lnoff; + /* cFYI(1,("fnxt2 entries in buf %d index_of_last %d", psrch_inf->entries_in_buffer, psrch_inf->index_of_last_entry)); */ @@ -4876,6 +4901,61 @@ CIFSSMBSetFileInfo(const int xid, struct cifsTconInfo *tcon, return rc; } +int +CIFSSMBSetFileDisposition(const int xid, struct cifsTconInfo *tcon, + bool delete_file, __u16 fid, __u32 pid_of_opener) +{ + struct smb_com_transaction2_sfi_req *pSMB = NULL; + char *data_offset; + int rc = 0; + __u16 params, param_offset, offset, byte_count, count; + + cFYI(1, ("Set File Disposition (via SetFileInfo)")); + rc = small_smb_init(SMB_COM_TRANSACTION2, 15, tcon, (void **) &pSMB); + + if (rc) + return rc; + + pSMB->hdr.Pid = cpu_to_le16((__u16)pid_of_opener); + pSMB->hdr.PidHigh = cpu_to_le16((__u16)(pid_of_opener >> 16)); + + params = 6; + pSMB->MaxSetupCount = 0; + pSMB->Reserved = 0; + pSMB->Flags = 0; + pSMB->Timeout = 0; + pSMB->Reserved2 = 0; + param_offset = offsetof(struct smb_com_transaction2_sfi_req, Fid) - 4; + offset = param_offset + params; + + data_offset = (char *) (&pSMB->hdr.Protocol) + offset; + + count = 1; + pSMB->MaxParameterCount = cpu_to_le16(2); + /* BB find max SMB PDU from sess */ + pSMB->MaxDataCount = cpu_to_le16(1000); + pSMB->SetupCount = 1; + pSMB->Reserved3 = 0; + pSMB->SubCommand = cpu_to_le16(TRANS2_SET_FILE_INFORMATION); + byte_count = 3 /* pad */ + params + count; + pSMB->DataCount = cpu_to_le16(count); + pSMB->ParameterCount = cpu_to_le16(params); + pSMB->TotalDataCount = pSMB->DataCount; + pSMB->TotalParameterCount = pSMB->ParameterCount; + pSMB->ParameterOffset = cpu_to_le16(param_offset); + pSMB->DataOffset = cpu_to_le16(offset); + pSMB->Fid = fid; + pSMB->InformationLevel = cpu_to_le16(SMB_SET_FILE_DISPOSITION_INFO); + pSMB->Reserved4 = 0; + pSMB->hdr.smb_buf_length += byte_count; + pSMB->ByteCount = cpu_to_le16(byte_count); + *data_offset = delete_file ? 1 : 0; + rc = SendReceiveNoRsp(xid, tcon->ses, (struct smb_hdr *) pSMB, 0); + if (rc) + cFYI(1, ("Send error in SetFileDisposition = %d", rc)); + + return rc; +} int CIFSSMBSetPathInfo(const int xid, struct cifsTconInfo *tcon, diff --git a/fs/cifs/dns_resolve.c b/fs/cifs/dns_resolve.c index a2e0673e1b0..1e0c1bd8f2e 100644 --- a/fs/cifs/dns_resolve.c +++ b/fs/cifs/dns_resolve.c @@ -29,19 +29,55 @@ #include "cifsproto.h" #include "cifs_debug.h" -static int dns_resolver_instantiate(struct key *key, const void *data, +/* Checks if supplied name is IP address + * returns: + * 1 - name is IP + * 0 - name is not IP + */ +static int +is_ip(const char *name) +{ + int rc; + struct sockaddr_in sin_server; + struct sockaddr_in6 sin_server6; + + rc = cifs_inet_pton(AF_INET, name, + &sin_server.sin_addr.s_addr); + + if (rc <= 0) { + /* not ipv4 address, try ipv6 */ + rc = cifs_inet_pton(AF_INET6, name, + &sin_server6.sin6_addr.in6_u); + if (rc > 0) + return 1; + } else { + return 1; + } + /* we failed translating address */ + return 0; +} + +static int +dns_resolver_instantiate(struct key *key, const void *data, size_t datalen) { int rc = 0; char *ip; - ip = kmalloc(datalen+1, GFP_KERNEL); + ip = kmalloc(datalen + 1, GFP_KERNEL); if (!ip) return -ENOMEM; memcpy(ip, data, datalen); ip[datalen] = '\0'; + /* make sure this looks like an address */ + if (!is_ip((const char *) ip)) { + kfree(ip); + return -EINVAL; + } + + key->type_data.x[0] = datalen; rcu_assign_pointer(key->payload.data, ip); return rc; @@ -62,33 +98,6 @@ struct key_type key_type_dns_resolver = { .match = user_match, }; -/* Checks if supplied name is IP address - * returns: - * 1 - name is IP - * 0 - name is not IP - */ -static int is_ip(const char *name) -{ - int rc; - struct sockaddr_in sin_server; - struct sockaddr_in6 sin_server6; - - rc = cifs_inet_pton(AF_INET, name, - &sin_server.sin_addr.s_addr); - - if (rc <= 0) { - /* not ipv4 address, try ipv6 */ - rc = cifs_inet_pton(AF_INET6, name, - &sin_server6.sin6_addr.in6_u); - if (rc > 0) - return 1; - } else { - return 1; - } - /* we failed translating address */ - return 0; -} - /* Resolves server name to ip address. * input: * unc - server UNC @@ -140,6 +149,7 @@ dns_resolve_server_name_to_ip(const char *unc, char **ip_addr) rkey = request_key(&key_type_dns_resolver, name, ""); if (!IS_ERR(rkey)) { + len = rkey->type_data.x[0]; data = rkey->payload.data; } else { cERROR(1, ("%s: unable to resolve: %s", __func__, name)); @@ -148,11 +158,9 @@ dns_resolve_server_name_to_ip(const char *unc, char **ip_addr) skip_upcall: if (data) { - len = strlen(data); - *ip_addr = kmalloc(len+1, GFP_KERNEL); + *ip_addr = kmalloc(len + 1, GFP_KERNEL); if (*ip_addr) { - memcpy(*ip_addr, data, len); - (*ip_addr)[len] = '\0'; + memcpy(*ip_addr, data, len + 1); if (!IS_ERR(rkey)) cFYI(1, ("%s: resolved: %s to %s", __func__, name, diff --git a/fs/cifs/file.c b/fs/cifs/file.c index cbefe1f1f9f..c4a8a060512 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -107,7 +107,7 @@ static inline int cifs_open_inode_helper(struct inode *inode, struct file *file, /* want handles we can use to read with first in the list so we do not have to walk the - list to search for one in prepare_write */ + list to search for one in write_begin */ if ((file->f_flags & O_ACCMODE) == O_WRONLY) { list_add_tail(&pCifsFile->flist, &pCifsInode->openFileList); @@ -915,7 +915,7 @@ ssize_t cifs_user_write(struct file *file, const char __user *write_data, } static ssize_t cifs_write(struct file *file, const char *write_data, - size_t write_size, loff_t *poffset) + size_t write_size, loff_t *poffset) { int rc = 0; unsigned int bytes_written = 0; @@ -1065,6 +1065,7 @@ struct cifsFileInfo *find_readable_file(struct cifsInodeInfo *cifs_inode) struct cifsFileInfo *find_writable_file(struct cifsInodeInfo *cifs_inode) { struct cifsFileInfo *open_file; + bool any_available = false; int rc; /* Having a null inode here (because mapping->host was set to zero by @@ -1080,8 +1081,10 @@ struct cifsFileInfo *find_writable_file(struct cifsInodeInfo *cifs_inode) read_lock(&GlobalSMBSeslock); refind_writable: list_for_each_entry(open_file, &cifs_inode->openFileList, flist) { - if (open_file->closePend) + if (open_file->closePend || + (!any_available && open_file->pid != current->tgid)) continue; + if (open_file->pfile && ((open_file->pfile->f_flags & O_RDWR) || (open_file->pfile->f_flags & O_WRONLY))) { @@ -1131,6 +1134,11 @@ refind_writable: of the loop here. */ } } + /* couldn't find useable FH with same pid, try any available */ + if (!any_available) { + any_available = true; + goto refind_writable; + } read_unlock(&GlobalSMBSeslock); return NULL; } @@ -1447,49 +1455,52 @@ static int cifs_writepage(struct page *page, struct writeback_control *wbc) return rc; } -static int cifs_commit_write(struct file *file, struct page *page, - unsigned offset, unsigned to) +static int cifs_write_end(struct file *file, struct address_space *mapping, + loff_t pos, unsigned len, unsigned copied, + struct page *page, void *fsdata) { - int xid; - int rc = 0; - struct inode *inode = page->mapping->host; - loff_t position = ((loff_t)page->index << PAGE_CACHE_SHIFT) + to; - char *page_data; + int rc; + struct inode *inode = mapping->host; - xid = GetXid(); - cFYI(1, ("commit write for page %p up to position %lld for %d", - page, position, to)); - spin_lock(&inode->i_lock); - if (position > inode->i_size) - i_size_write(inode, position); + cFYI(1, ("write_end for page %p from pos %lld with %d bytes", + page, pos, copied)); + + if (!PageUptodate(page) && copied == PAGE_CACHE_SIZE) + SetPageUptodate(page); - spin_unlock(&inode->i_lock); if (!PageUptodate(page)) { - position = ((loff_t)page->index << PAGE_CACHE_SHIFT) + offset; - /* can not rely on (or let) writepage write this data */ - if (to < offset) { - cFYI(1, ("Illegal offsets, can not copy from %d to %d", - offset, to)); - FreeXid(xid); - return rc; - } + char *page_data; + unsigned offset = pos & (PAGE_CACHE_SIZE - 1); + int xid; + + xid = GetXid(); /* this is probably better than directly calling partialpage_write since in this function the file handle is known which we might as well leverage */ /* BB check if anything else missing out of ppw such as updating last write time */ page_data = kmap(page); - rc = cifs_write(file, page_data + offset, to-offset, - &position); - if (rc > 0) - rc = 0; - /* else if (rc < 0) should we set writebehind rc? */ + rc = cifs_write(file, page_data + offset, copied, &pos); + /* if (rc < 0) should we set writebehind rc? */ kunmap(page); + + FreeXid(xid); } else { + rc = copied; + pos += copied; set_page_dirty(page); } - FreeXid(xid); + if (rc > 0) { + spin_lock(&inode->i_lock); + if (pos > inode->i_size) + i_size_write(inode, pos); + spin_unlock(&inode->i_lock); + } + + unlock_page(page); + page_cache_release(page); + return rc; } @@ -2035,49 +2046,44 @@ bool is_size_safe_to_change(struct cifsInodeInfo *cifsInode, __u64 end_of_file) return true; } -static int cifs_prepare_write(struct file *file, struct page *page, - unsigned from, unsigned to) +static int cifs_write_begin(struct file *file, struct address_space *mapping, + loff_t pos, unsigned len, unsigned flags, + struct page **pagep, void **fsdata) { - int rc = 0; - loff_t i_size; - loff_t offset; + pgoff_t index = pos >> PAGE_CACHE_SHIFT; + loff_t offset = pos & (PAGE_CACHE_SIZE - 1); - cFYI(1, ("prepare write for page %p from %d to %d", page, from, to)); - if (PageUptodate(page)) + cFYI(1, ("write_begin from %lld len %d", (long long)pos, len)); + + *pagep = __grab_cache_page(mapping, index); + if (!*pagep) + return -ENOMEM; + + if (PageUptodate(*pagep)) return 0; /* If we are writing a full page it will be up to date, no need to read from the server */ - if ((to == PAGE_CACHE_SIZE) && (from == 0)) { - SetPageUptodate(page); + if (len == PAGE_CACHE_SIZE && flags & AOP_FLAG_UNINTERRUPTIBLE) return 0; - } - offset = (loff_t)page->index << PAGE_CACHE_SHIFT; - i_size = i_size_read(page->mapping->host); + if ((file->f_flags & O_ACCMODE) != O_WRONLY) { + int rc; - if ((offset >= i_size) || - ((from == 0) && (offset + to) >= i_size)) { - /* - * We don't need to read data beyond the end of the file. - * zero it, and set the page uptodate - */ - simple_prepare_write(file, page, from, to); - SetPageUptodate(page); - } else if ((file->f_flags & O_ACCMODE) != O_WRONLY) { /* might as well read a page, it is fast enough */ - rc = cifs_readpage_worker(file, page, &offset); + rc = cifs_readpage_worker(file, *pagep, &offset); + + /* we do not need to pass errors back + e.g. if we do not have read access to the file + because cifs_write_end will attempt synchronous writes + -- shaggy */ } else { /* we could try using another file handle if there is one - but how would we lock it to prevent close of that handle racing with this read? In any case - this will be written out by commit_write so is fine */ + this will be written out by write_end so is fine */ } - /* we do not need to pass errors back - e.g. if we do not have read access to the file - because cifs_commit_write will do the right thing. -- shaggy */ - return 0; } @@ -2086,8 +2092,8 @@ const struct address_space_operations cifs_addr_ops = { .readpages = cifs_readpages, .writepage = cifs_writepage, .writepages = cifs_writepages, - .prepare_write = cifs_prepare_write, - .commit_write = cifs_commit_write, + .write_begin = cifs_write_begin, + .write_end = cifs_write_end, .set_page_dirty = __set_page_dirty_nobuffers, /* .sync_page = cifs_sync_page, */ /* .direct_IO = */ @@ -2102,8 +2108,8 @@ const struct address_space_operations cifs_addr_ops_smallbuf = { .readpage = cifs_readpage, .writepage = cifs_writepage, .writepages = cifs_writepages, - .prepare_write = cifs_prepare_write, - .commit_write = cifs_commit_write, + .write_begin = cifs_write_begin, + .write_end = cifs_write_end, .set_page_dirty = __set_page_dirty_nobuffers, /* .sync_page = cifs_sync_page, */ /* .direct_IO = */ diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index 9c548f11010..a8c833345fc 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -665,40 +665,201 @@ struct inode *cifs_iget(struct super_block *sb, unsigned long ino) return inode; } -int cifs_unlink(struct inode *inode, struct dentry *direntry) +static int +cifs_set_file_info(struct inode *inode, struct iattr *attrs, int xid, + char *full_path, __u32 dosattr) +{ + int rc; + int oplock = 0; + __u16 netfid; + __u32 netpid; + bool set_time = false; + struct cifsFileInfo *open_file; + struct cifsInodeInfo *cifsInode = CIFS_I(inode); + struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); + struct cifsTconInfo *pTcon = cifs_sb->tcon; + FILE_BASIC_INFO info_buf; + + if (attrs->ia_valid & ATTR_ATIME) { + set_time = true; + info_buf.LastAccessTime = + cpu_to_le64(cifs_UnixTimeToNT(attrs->ia_atime)); + } else + info_buf.LastAccessTime = 0; + + if (attrs->ia_valid & ATTR_MTIME) { + set_time = true; + info_buf.LastWriteTime = + cpu_to_le64(cifs_UnixTimeToNT(attrs->ia_mtime)); + } else + info_buf.LastWriteTime = 0; + + /* + * Samba throws this field away, but windows may actually use it. + * Do not set ctime unless other time stamps are changed explicitly + * (i.e. by utimes()) since we would then have a mix of client and + * server times. + */ + if (set_time && (attrs->ia_valid & ATTR_CTIME)) { + cFYI(1, ("CIFS - CTIME changed")); + info_buf.ChangeTime = + cpu_to_le64(cifs_UnixTimeToNT(attrs->ia_ctime)); + } else + info_buf.ChangeTime = 0; + + info_buf.CreationTime = 0; /* don't change */ + info_buf.Attributes = cpu_to_le32(dosattr); + + /* + * If the file is already open for write, just use that fileid + */ + open_file = find_writable_file(cifsInode); + if (open_file) { + netfid = open_file->netfid; + netpid = open_file->pid; + goto set_via_filehandle; + } + + /* + * NT4 apparently returns success on this call, but it doesn't + * really work. + */ + if (!(pTcon->ses->flags & CIFS_SES_NT4)) { + rc = CIFSSMBSetPathInfo(xid, pTcon, full_path, + &info_buf, cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); + if (rc == 0) { + cifsInode->cifsAttrs = dosattr; + goto out; + } else if (rc != -EOPNOTSUPP && rc != -EINVAL) + goto out; + } + + cFYI(1, ("calling SetFileInfo since SetPathInfo for " + "times not supported by this server")); + rc = CIFSSMBOpen(xid, pTcon, full_path, FILE_OPEN, + SYNCHRONIZE | FILE_WRITE_ATTRIBUTES, + CREATE_NOT_DIR, &netfid, &oplock, + NULL, cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); + + if (rc != 0) { + if (rc == -EIO) + rc = -EINVAL; + goto out; + } + + netpid = current->tgid; + +set_via_filehandle: + rc = CIFSSMBSetFileInfo(xid, pTcon, &info_buf, netfid, netpid); + if (!rc) + cifsInode->cifsAttrs = dosattr; + + if (open_file == NULL) + CIFSSMBClose(xid, pTcon, netfid); + else + atomic_dec(&open_file->wrtPending); +out: + return rc; +} + +/* + * open the given file (if it isn't already), set the DELETE_ON_CLOSE bit + * and rename it to a random name that hopefully won't conflict with + * anything else. + */ +static int +cifs_rename_pending_delete(char *full_path, struct inode *inode, int xid) +{ + int oplock = 0; + int rc; + __u16 netfid; + struct cifsInodeInfo *cifsInode = CIFS_I(inode); + struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); + struct cifsTconInfo *tcon = cifs_sb->tcon; + __u32 dosattr; + FILE_BASIC_INFO *info_buf; + + rc = CIFSSMBOpen(xid, tcon, full_path, FILE_OPEN, + DELETE|FILE_WRITE_ATTRIBUTES, + CREATE_NOT_DIR|CREATE_DELETE_ON_CLOSE, + &netfid, &oplock, NULL, cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); + if (rc != 0) + goto out; + + /* set ATTR_HIDDEN and clear ATTR_READONLY */ + cifsInode = CIFS_I(inode); + dosattr = cifsInode->cifsAttrs & ~ATTR_READONLY; + if (dosattr == 0) + dosattr |= ATTR_NORMAL; + dosattr |= ATTR_HIDDEN; + + info_buf = kzalloc(sizeof(*info_buf), GFP_KERNEL); + if (info_buf == NULL) { + rc = -ENOMEM; + goto out_close; + } + info_buf->Attributes = cpu_to_le32(dosattr); + rc = CIFSSMBSetFileInfo(xid, tcon, info_buf, netfid, current->tgid); + kfree(info_buf); + if (rc != 0) + goto out_close; + cifsInode->cifsAttrs = dosattr; + + /* silly-rename the file */ + CIFSSMBRenameOpenFile(xid, tcon, netfid, NULL, cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); + + /* set DELETE_ON_CLOSE */ + rc = CIFSSMBSetFileDisposition(xid, tcon, true, netfid, current->tgid); + + /* + * some samba versions return -ENOENT when we try to set the file + * disposition here. Likely a samba bug, but work around it for now + */ + if (rc == -ENOENT) + rc = 0; + +out_close: + CIFSSMBClose(xid, tcon, netfid); +out: + return rc; +} + +int cifs_unlink(struct inode *dir, struct dentry *dentry) { int rc = 0; int xid; - struct cifs_sb_info *cifs_sb; - struct cifsTconInfo *pTcon; char *full_path = NULL; - struct cifsInodeInfo *cifsInode; - FILE_BASIC_INFO *pinfo_buf; + struct inode *inode = dentry->d_inode; + struct cifsInodeInfo *cifsInode = CIFS_I(inode); + struct super_block *sb = dir->i_sb; + struct cifs_sb_info *cifs_sb = CIFS_SB(sb); + struct cifsTconInfo *tcon = cifs_sb->tcon; + struct iattr *attrs = NULL; + __u32 dosattr = 0, origattr = 0; - cFYI(1, ("cifs_unlink, inode = 0x%p", inode)); + cFYI(1, ("cifs_unlink, dir=0x%p, dentry=0x%p", dir, dentry)); xid = GetXid(); - if (inode) - cifs_sb = CIFS_SB(inode->i_sb); - else - cifs_sb = CIFS_SB(direntry->d_sb); - pTcon = cifs_sb->tcon; - - /* Unlink can be called from rename so we can not grab the sem here - since we deadlock otherwise */ -/* mutex_lock(&direntry->d_sb->s_vfs_rename_mutex);*/ - full_path = build_path_from_dentry(direntry); -/* mutex_unlock(&direntry->d_sb->s_vfs_rename_mutex);*/ + /* Unlink can be called from rename so we can not take the + * sb->s_vfs_rename_mutex here */ + full_path = build_path_from_dentry(dentry); if (full_path == NULL) { FreeXid(xid); return -ENOMEM; } - if ((pTcon->ses->capabilities & CAP_UNIX) && + if ((tcon->ses->capabilities & CAP_UNIX) && (CIFS_UNIX_POSIX_PATH_OPS_CAP & - le64_to_cpu(pTcon->fsUnixInfo.Capability))) { - rc = CIFSPOSIXDelFile(xid, pTcon, full_path, + le64_to_cpu(tcon->fsUnixInfo.Capability))) { + rc = CIFSPOSIXDelFile(xid, tcon, full_path, SMB_POSIX_UNLINK_FILE_TARGET, cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); cFYI(1, ("posix del rc %d", rc)); @@ -706,125 +867,60 @@ int cifs_unlink(struct inode *inode, struct dentry *direntry) goto psx_del_no_retry; } - rc = CIFSSMBDelFile(xid, pTcon, full_path, cifs_sb->local_nls, +retry_std_delete: + rc = CIFSSMBDelFile(xid, tcon, full_path, cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); + psx_del_no_retry: if (!rc) { - if (direntry->d_inode) - drop_nlink(direntry->d_inode); + if (inode) + drop_nlink(inode); } else if (rc == -ENOENT) { - d_drop(direntry); + d_drop(dentry); } else if (rc == -ETXTBSY) { - int oplock = 0; - __u16 netfid; - - rc = CIFSSMBOpen(xid, pTcon, full_path, FILE_OPEN, DELETE, - CREATE_NOT_DIR | CREATE_DELETE_ON_CLOSE, - &netfid, &oplock, NULL, cifs_sb->local_nls, - cifs_sb->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); - if (rc == 0) { - CIFSSMBRenameOpenFile(xid, pTcon, netfid, NULL, - cifs_sb->local_nls, - cifs_sb->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); - CIFSSMBClose(xid, pTcon, netfid); - if (direntry->d_inode) - drop_nlink(direntry->d_inode); + rc = cifs_rename_pending_delete(full_path, inode, xid); + if (rc == 0) + drop_nlink(inode); + } else if (rc == -EACCES && dosattr == 0) { + attrs = kzalloc(sizeof(*attrs), GFP_KERNEL); + if (attrs == NULL) { + rc = -ENOMEM; + goto out_reval; } - } else if (rc == -EACCES) { - /* try only if r/o attribute set in local lookup data? */ - pinfo_buf = kzalloc(sizeof(FILE_BASIC_INFO), GFP_KERNEL); - if (pinfo_buf) { - /* ATTRS set to normal clears r/o bit */ - pinfo_buf->Attributes = cpu_to_le32(ATTR_NORMAL); - if (!(pTcon->ses->flags & CIFS_SES_NT4)) - rc = CIFSSMBSetPathInfo(xid, pTcon, full_path, - pinfo_buf, - cifs_sb->local_nls, - cifs_sb->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); - else - rc = -EOPNOTSUPP; - if (rc == -EOPNOTSUPP) { - int oplock = 0; - __u16 netfid; - /* rc = CIFSSMBSetAttrLegacy(xid, pTcon, - full_path, - (__u16)ATTR_NORMAL, - cifs_sb->local_nls); - For some strange reason it seems that NT4 eats the - old setattr call without actually setting the - attributes so on to the third attempted workaround - */ - - /* BB could scan to see if we already have it open - and pass in pid of opener to function */ - rc = CIFSSMBOpen(xid, pTcon, full_path, - FILE_OPEN, SYNCHRONIZE | - FILE_WRITE_ATTRIBUTES, 0, - &netfid, &oplock, NULL, - cifs_sb->local_nls, - cifs_sb->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); - if (rc == 0) { - rc = CIFSSMBSetFileInfo(xid, pTcon, - pinfo_buf, - netfid, - current->tgid); - CIFSSMBClose(xid, pTcon, netfid); - } - } - kfree(pinfo_buf); - } - if (rc == 0) { - rc = CIFSSMBDelFile(xid, pTcon, full_path, - cifs_sb->local_nls, - cifs_sb->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); - if (!rc) { - if (direntry->d_inode) - drop_nlink(direntry->d_inode); - } else if (rc == -ETXTBSY) { - int oplock = 0; - __u16 netfid; - - rc = CIFSSMBOpen(xid, pTcon, full_path, - FILE_OPEN, DELETE, - CREATE_NOT_DIR | - CREATE_DELETE_ON_CLOSE, - &netfid, &oplock, NULL, - cifs_sb->local_nls, - cifs_sb->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); - if (rc == 0) { - CIFSSMBRenameOpenFile(xid, pTcon, - netfid, NULL, - cifs_sb->local_nls, - cifs_sb->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); - CIFSSMBClose(xid, pTcon, netfid); - if (direntry->d_inode) - drop_nlink(direntry->d_inode); - } - /* BB if rc = -ETXTBUSY goto the rename logic BB */ - } - } - } - if (direntry->d_inode) { - cifsInode = CIFS_I(direntry->d_inode); - cifsInode->time = 0; /* will force revalidate to get info - when needed */ - direntry->d_inode->i_ctime = current_fs_time(inode->i_sb); + /* try to reset dos attributes */ + origattr = cifsInode->cifsAttrs; + if (origattr == 0) + origattr |= ATTR_NORMAL; + dosattr = origattr & ~ATTR_READONLY; + if (dosattr == 0) + dosattr |= ATTR_NORMAL; + dosattr |= ATTR_HIDDEN; + + rc = cifs_set_file_info(inode, attrs, xid, full_path, dosattr); + if (rc != 0) + goto out_reval; + + goto retry_std_delete; } + + /* undo the setattr if we errored out and it's needed */ + if (rc != 0 && dosattr != 0) + cifs_set_file_info(inode, attrs, xid, full_path, origattr); + +out_reval: if (inode) { - inode->i_ctime = inode->i_mtime = current_fs_time(inode->i_sb); cifsInode = CIFS_I(inode); - cifsInode->time = 0; /* force revalidate of dir as well */ + cifsInode->time = 0; /* will force revalidate to get info + when needed */ + inode->i_ctime = current_fs_time(sb); } + dir->i_ctime = dir->i_mtime = current_fs_time(sb); + cifsInode = CIFS_I(dir); + CIFS_I(dir)->time = 0; /* force revalidate of dir as well */ kfree(full_path); + kfree(attrs); FreeXid(xid); return rc; } @@ -869,7 +965,7 @@ static void posix_fill_in_inode(struct inode *tmp_inode, int cifs_mkdir(struct inode *inode, struct dentry *direntry, int mode) { - int rc = 0; + int rc = 0, tmprc; int xid; struct cifs_sb_info *cifs_sb; struct cifsTconInfo *pTcon; @@ -931,6 +1027,7 @@ int cifs_mkdir(struct inode *inode, struct dentry *direntry, int mode) kfree(pInfo); goto mkdir_get_info; } + /* Is an i_ino of zero legal? */ /* Are there sanity checks we can use to ensure that the server is really filling in that field? */ @@ -1019,12 +1116,20 @@ mkdir_get_info: if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL) && (mode & S_IWUGO) == 0) { FILE_BASIC_INFO pInfo; + struct cifsInodeInfo *cifsInode; + u32 dosattrs; + memset(&pInfo, 0, sizeof(pInfo)); - pInfo.Attributes = cpu_to_le32(ATTR_READONLY); - CIFSSMBSetPathInfo(xid, pTcon, full_path, - &pInfo, cifs_sb->local_nls, + cifsInode = CIFS_I(newinode); + dosattrs = cifsInode->cifsAttrs|ATTR_READONLY; + pInfo.Attributes = cpu_to_le32(dosattrs); + tmprc = CIFSSMBSetPathInfo(xid, pTcon, + full_path, &pInfo, + cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); + if (tmprc == 0) + cifsInode->cifsAttrs = dosattrs; } if (direntry->d_inode) { if (cifs_sb->mnt_cifs_flags & @@ -1096,117 +1201,141 @@ int cifs_rmdir(struct inode *inode, struct dentry *direntry) return rc; } +static int +cifs_do_rename(int xid, struct dentry *from_dentry, const char *fromPath, + struct dentry *to_dentry, const char *toPath) +{ + struct cifs_sb_info *cifs_sb = CIFS_SB(from_dentry->d_sb); + struct cifsTconInfo *pTcon = cifs_sb->tcon; + __u16 srcfid; + int oplock, rc; + + /* try path-based rename first */ + rc = CIFSSMBRename(xid, pTcon, fromPath, toPath, cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); + + /* + * don't bother with rename by filehandle unless file is busy and + * source Note that cross directory moves do not work with + * rename by filehandle to various Windows servers. + */ + if (rc == 0 || rc != -ETXTBSY) + return rc; + + /* open the file to be renamed -- we need DELETE perms */ + rc = CIFSSMBOpen(xid, pTcon, fromPath, FILE_OPEN, DELETE, + CREATE_NOT_DIR, &srcfid, &oplock, NULL, + cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); + + if (rc == 0) { + rc = CIFSSMBRenameOpenFile(xid, pTcon, srcfid, + (const char *) to_dentry->d_name.name, + cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); + + CIFSSMBClose(xid, pTcon, srcfid); + } + + return rc; +} + int cifs_rename(struct inode *source_inode, struct dentry *source_direntry, struct inode *target_inode, struct dentry *target_direntry) { - char *fromName; - char *toName; + char *fromName = NULL; + char *toName = NULL; struct cifs_sb_info *cifs_sb_source; struct cifs_sb_info *cifs_sb_target; struct cifsTconInfo *pTcon; + FILE_UNIX_BASIC_INFO *info_buf_source = NULL; + FILE_UNIX_BASIC_INFO *info_buf_target; int xid; - int rc = 0; - - xid = GetXid(); + int rc; cifs_sb_target = CIFS_SB(target_inode->i_sb); cifs_sb_source = CIFS_SB(source_inode->i_sb); pTcon = cifs_sb_source->tcon; + xid = GetXid(); + + /* + * BB: this might be allowed if same server, but different share. + * Consider adding support for this + */ if (pTcon != cifs_sb_target->tcon) { - FreeXid(xid); - return -EXDEV; /* BB actually could be allowed if same server, - but different share. - Might eventually add support for this */ + rc = -EXDEV; + goto cifs_rename_exit; } - /* we already have the rename sem so we do not need to grab it again - here to protect the path integrity */ + /* + * we already have the rename sem so we do not need to + * grab it again here to protect the path integrity + */ fromName = build_path_from_dentry(source_direntry); + if (fromName == NULL) { + rc = -ENOMEM; + goto cifs_rename_exit; + } + toName = build_path_from_dentry(target_direntry); - if ((fromName == NULL) || (toName == NULL)) { + if (toName == NULL) { rc = -ENOMEM; goto cifs_rename_exit; } - rc = CIFSSMBRename(xid, pTcon, fromName, toName, - cifs_sb_source->local_nls, - cifs_sb_source->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); + rc = cifs_do_rename(xid, source_direntry, fromName, + target_direntry, toName); + if (rc == -EEXIST) { - /* check if they are the same file because rename of hardlinked - files is a noop */ - FILE_UNIX_BASIC_INFO *info_buf_source; - FILE_UNIX_BASIC_INFO *info_buf_target; - - info_buf_source = - kmalloc(2 * sizeof(FILE_UNIX_BASIC_INFO), GFP_KERNEL); - if (info_buf_source != NULL) { + if (pTcon->unix_ext) { + /* + * Are src and dst hardlinks of same inode? We can + * only tell with unix extensions enabled + */ + info_buf_source = + kmalloc(2 * sizeof(FILE_UNIX_BASIC_INFO), + GFP_KERNEL); + if (info_buf_source == NULL) + goto unlink_target; + info_buf_target = info_buf_source + 1; - if (pTcon->unix_ext) - rc = CIFSSMBUnixQPathInfo(xid, pTcon, fromName, - info_buf_source, - cifs_sb_source->local_nls, - cifs_sb_source->mnt_cifs_flags & + rc = CIFSSMBUnixQPathInfo(xid, pTcon, fromName, + info_buf_source, + cifs_sb_source->local_nls, + cifs_sb_source->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); - /* else rc is still EEXIST so will fall through to - unlink the target and retry rename */ - if (rc == 0) { - rc = CIFSSMBUnixQPathInfo(xid, pTcon, toName, - info_buf_target, + if (rc != 0) + goto unlink_target; + + rc = CIFSSMBUnixQPathInfo(xid, pTcon, + toName, info_buf_target, cifs_sb_target->local_nls, /* remap based on source sb */ cifs_sb_source->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); - } - if ((rc == 0) && - (info_buf_source->UniqueId == - info_buf_target->UniqueId)) { - /* do not rename since the files are hardlinked which - is a noop */ - } else { - /* we either can not tell the files are hardlinked - (as with Windows servers) or files are not - hardlinked so delete the target manually before - renaming to follow POSIX rather than Windows - semantics */ - cifs_unlink(target_inode, target_direntry); - rc = CIFSSMBRename(xid, pTcon, fromName, - toName, - cifs_sb_source->local_nls, - cifs_sb_source->mnt_cifs_flags - & CIFS_MOUNT_MAP_SPECIAL_CHR); - } - kfree(info_buf_source); - } /* if we can not get memory just leave rc as EEXIST */ - } - - if (rc) - cFYI(1, ("rename rc %d", rc)); - - if ((rc == -EIO) || (rc == -EEXIST)) { - int oplock = 0; - __u16 netfid; - - /* BB FIXME Is Generic Read correct for rename? */ - /* if renaming directory - we should not say CREATE_NOT_DIR, - need to test renaming open directory, also GENERIC_READ - might not right be right access to request */ - rc = CIFSSMBOpen(xid, pTcon, fromName, FILE_OPEN, GENERIC_READ, - CREATE_NOT_DIR, &netfid, &oplock, NULL, - cifs_sb_source->local_nls, - cifs_sb_source->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); - if (rc == 0) { - rc = CIFSSMBRenameOpenFile(xid, pTcon, netfid, toName, - cifs_sb_source->local_nls, - cifs_sb_source->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); - CIFSSMBClose(xid, pTcon, netfid); - } + + if (rc == 0 && (info_buf_source->UniqueId == + info_buf_target->UniqueId)) + /* same file, POSIX says that this is a noop */ + goto cifs_rename_exit; + } /* else ... BB we could add the same check for Windows by + checking the UniqueId via FILE_INTERNAL_INFO */ +unlink_target: + /* + * we either can not tell the files are hardlinked (as with + * Windows servers) or files are not hardlinked. Delete the + * target manually before renaming to follow POSIX rather than + * Windows semantics + */ + cifs_unlink(target_inode, target_direntry); + rc = cifs_do_rename(xid, source_direntry, fromName, + target_direntry, toName); } cifs_rename_exit: + kfree(info_buf_source); kfree(fromName); kfree(toName); FreeXid(xid); @@ -1507,101 +1636,6 @@ cifs_set_file_size(struct inode *inode, struct iattr *attrs, } static int -cifs_set_file_info(struct inode *inode, struct iattr *attrs, int xid, - char *full_path, __u32 dosattr) -{ - int rc; - int oplock = 0; - __u16 netfid; - __u32 netpid; - bool set_time = false; - struct cifsFileInfo *open_file; - struct cifsInodeInfo *cifsInode = CIFS_I(inode); - struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); - struct cifsTconInfo *pTcon = cifs_sb->tcon; - FILE_BASIC_INFO info_buf; - - if (attrs->ia_valid & ATTR_ATIME) { - set_time = true; - info_buf.LastAccessTime = - cpu_to_le64(cifs_UnixTimeToNT(attrs->ia_atime)); - } else - info_buf.LastAccessTime = 0; - - if (attrs->ia_valid & ATTR_MTIME) { - set_time = true; - info_buf.LastWriteTime = - cpu_to_le64(cifs_UnixTimeToNT(attrs->ia_mtime)); - } else - info_buf.LastWriteTime = 0; - - /* - * Samba throws this field away, but windows may actually use it. - * Do not set ctime unless other time stamps are changed explicitly - * (i.e. by utimes()) since we would then have a mix of client and - * server times. - */ - if (set_time && (attrs->ia_valid & ATTR_CTIME)) { - cFYI(1, ("CIFS - CTIME changed")); - info_buf.ChangeTime = - cpu_to_le64(cifs_UnixTimeToNT(attrs->ia_ctime)); - } else - info_buf.ChangeTime = 0; - - info_buf.CreationTime = 0; /* don't change */ - info_buf.Attributes = cpu_to_le32(dosattr); - - /* - * If the file is already open for write, just use that fileid - */ - open_file = find_writable_file(cifsInode); - if (open_file) { - netfid = open_file->netfid; - netpid = open_file->pid; - goto set_via_filehandle; - } - - /* - * NT4 apparently returns success on this call, but it doesn't - * really work. - */ - if (!(pTcon->ses->flags & CIFS_SES_NT4)) { - rc = CIFSSMBSetPathInfo(xid, pTcon, full_path, - &info_buf, cifs_sb->local_nls, - cifs_sb->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); - if (rc != -EOPNOTSUPP && rc != -EINVAL) - goto out; - } - - cFYI(1, ("calling SetFileInfo since SetPathInfo for " - "times not supported by this server")); - rc = CIFSSMBOpen(xid, pTcon, full_path, FILE_OPEN, - SYNCHRONIZE | FILE_WRITE_ATTRIBUTES, - CREATE_NOT_DIR, &netfid, &oplock, - NULL, cifs_sb->local_nls, - cifs_sb->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); - - if (rc != 0) { - if (rc == -EIO) - rc = -EINVAL; - goto out; - } - - netpid = current->tgid; - -set_via_filehandle: - rc = CIFSSMBSetFileInfo(xid, pTcon, &info_buf, netfid, netpid); - if (open_file == NULL) - CIFSSMBClose(xid, pTcon, netfid); - else - atomic_dec(&open_file->wrtPending); -out: - return rc; -} - -static int cifs_setattr_unix(struct dentry *direntry, struct iattr *attrs) { int rc; diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c index 4b17f8fe315..88786ba02d2 100644 --- a/fs/cifs/misc.c +++ b/fs/cifs/misc.c @@ -150,8 +150,7 @@ cifs_buf_get(void) but it may be more efficient to always alloc same size albeit slightly larger than necessary and maxbuffersize defaults to this and can not be bigger */ - ret_buf = (struct smb_hdr *) mempool_alloc(cifs_req_poolp, - GFP_KERNEL | GFP_NOFS); + ret_buf = mempool_alloc(cifs_req_poolp, GFP_NOFS); /* clear the first few header bytes */ /* for most paths, more is cleared in header_assemble */ @@ -188,8 +187,7 @@ cifs_small_buf_get(void) but it may be more efficient to always alloc same size albeit slightly larger than necessary and maxbuffersize defaults to this and can not be bigger */ - ret_buf = (struct smb_hdr *) mempool_alloc(cifs_sm_req_poolp, - GFP_KERNEL | GFP_NOFS); + ret_buf = mempool_alloc(cifs_sm_req_poolp, GFP_NOFS); if (ret_buf) { /* No need to clear memory here, cleared in header assemble */ /* memset(ret_buf, 0, sizeof(struct smb_hdr) + 27);*/ @@ -313,8 +311,6 @@ header_assemble(struct smb_hdr *buffer, char smb_command /* command */ , buffer->Flags2 = SMBFLG2_KNOWS_LONG_NAMES; buffer->Pid = cpu_to_le16((__u16)current->tgid); buffer->PidHigh = cpu_to_le16((__u16)(current->tgid >> 16)); - spin_lock(&GlobalMid_Lock); - spin_unlock(&GlobalMid_Lock); if (treeCon) { buffer->Tid = treeCon->tid; if (treeCon->ses) { diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c index 5f40ed3473f..765adf12d54 100644 --- a/fs/cifs/readdir.c +++ b/fs/cifs/readdir.c @@ -640,6 +640,70 @@ static int is_dir_changed(struct file *file) } +static int cifs_save_resume_key(const char *current_entry, + struct cifsFileInfo *cifsFile) +{ + int rc = 0; + unsigned int len = 0; + __u16 level; + char *filename; + + if ((cifsFile == NULL) || (current_entry == NULL)) + return -EINVAL; + + level = cifsFile->srch_inf.info_level; + + if (level == SMB_FIND_FILE_UNIX) { + FILE_UNIX_INFO *pFindData = (FILE_UNIX_INFO *)current_entry; + + filename = &pFindData->FileName[0]; + if (cifsFile->srch_inf.unicode) { + len = cifs_unicode_bytelen(filename); + } else { + /* BB should we make this strnlen of PATH_MAX? */ + len = strnlen(filename, PATH_MAX); + } + cifsFile->srch_inf.resume_key = pFindData->ResumeKey; + } else if (level == SMB_FIND_FILE_DIRECTORY_INFO) { + FILE_DIRECTORY_INFO *pFindData = + (FILE_DIRECTORY_INFO *)current_entry; + filename = &pFindData->FileName[0]; + len = le32_to_cpu(pFindData->FileNameLength); + cifsFile->srch_inf.resume_key = pFindData->FileIndex; + } else if (level == SMB_FIND_FILE_FULL_DIRECTORY_INFO) { + FILE_FULL_DIRECTORY_INFO *pFindData = + (FILE_FULL_DIRECTORY_INFO *)current_entry; + filename = &pFindData->FileName[0]; + len = le32_to_cpu(pFindData->FileNameLength); + cifsFile->srch_inf.resume_key = pFindData->FileIndex; + } else if (level == SMB_FIND_FILE_ID_FULL_DIR_INFO) { + SEARCH_ID_FULL_DIR_INFO *pFindData = + (SEARCH_ID_FULL_DIR_INFO *)current_entry; + filename = &pFindData->FileName[0]; + len = le32_to_cpu(pFindData->FileNameLength); + cifsFile->srch_inf.resume_key = pFindData->FileIndex; + } else if (level == SMB_FIND_FILE_BOTH_DIRECTORY_INFO) { + FILE_BOTH_DIRECTORY_INFO *pFindData = + (FILE_BOTH_DIRECTORY_INFO *)current_entry; + filename = &pFindData->FileName[0]; + len = le32_to_cpu(pFindData->FileNameLength); + cifsFile->srch_inf.resume_key = pFindData->FileIndex; + } else if (level == SMB_FIND_FILE_INFO_STANDARD) { + FIND_FILE_STANDARD_INFO *pFindData = + (FIND_FILE_STANDARD_INFO *)current_entry; + filename = &pFindData->FileName[0]; + /* one byte length, no name conversion */ + len = (unsigned int)pFindData->FileNameLength; + cifsFile->srch_inf.resume_key = pFindData->ResumeKey; + } else { + cFYI(1, ("Unknown findfirst level %d", level)); + return -EINVAL; + } + cifsFile->srch_inf.resume_name_len = len; + cifsFile->srch_inf.presume_name = filename; + return rc; +} + /* find the corresponding entry in the search */ /* Note that the SMB server returns search entries for . and .. which complicates logic here if we choose to parse for them and we do not @@ -703,6 +767,7 @@ static int find_cifs_entry(const int xid, struct cifsTconInfo *pTcon, while ((index_to_find >= cifsFile->srch_inf.index_of_last_entry) && (rc == 0) && !cifsFile->srch_inf.endOfSearch) { cFYI(1, ("calling findnext2")); + cifs_save_resume_key(cifsFile->srch_inf.last_entry, cifsFile); rc = CIFSFindNext(xid, pTcon, cifsFile->netfid, &cifsFile->srch_inf); if (rc) @@ -919,69 +984,6 @@ static int cifs_filldir(char *pfindEntry, struct file *file, return rc; } -static int cifs_save_resume_key(const char *current_entry, - struct cifsFileInfo *cifsFile) -{ - int rc = 0; - unsigned int len = 0; - __u16 level; - char *filename; - - if ((cifsFile == NULL) || (current_entry == NULL)) - return -EINVAL; - - level = cifsFile->srch_inf.info_level; - - if (level == SMB_FIND_FILE_UNIX) { - FILE_UNIX_INFO *pFindData = (FILE_UNIX_INFO *)current_entry; - - filename = &pFindData->FileName[0]; - if (cifsFile->srch_inf.unicode) { - len = cifs_unicode_bytelen(filename); - } else { - /* BB should we make this strnlen of PATH_MAX? */ - len = strnlen(filename, PATH_MAX); - } - cifsFile->srch_inf.resume_key = pFindData->ResumeKey; - } else if (level == SMB_FIND_FILE_DIRECTORY_INFO) { - FILE_DIRECTORY_INFO *pFindData = - (FILE_DIRECTORY_INFO *)current_entry; - filename = &pFindData->FileName[0]; - len = le32_to_cpu(pFindData->FileNameLength); - cifsFile->srch_inf.resume_key = pFindData->FileIndex; - } else if (level == SMB_FIND_FILE_FULL_DIRECTORY_INFO) { - FILE_FULL_DIRECTORY_INFO *pFindData = - (FILE_FULL_DIRECTORY_INFO *)current_entry; - filename = &pFindData->FileName[0]; - len = le32_to_cpu(pFindData->FileNameLength); - cifsFile->srch_inf.resume_key = pFindData->FileIndex; - } else if (level == SMB_FIND_FILE_ID_FULL_DIR_INFO) { - SEARCH_ID_FULL_DIR_INFO *pFindData = - (SEARCH_ID_FULL_DIR_INFO *)current_entry; - filename = &pFindData->FileName[0]; - len = le32_to_cpu(pFindData->FileNameLength); - cifsFile->srch_inf.resume_key = pFindData->FileIndex; - } else if (level == SMB_FIND_FILE_BOTH_DIRECTORY_INFO) { - FILE_BOTH_DIRECTORY_INFO *pFindData = - (FILE_BOTH_DIRECTORY_INFO *)current_entry; - filename = &pFindData->FileName[0]; - len = le32_to_cpu(pFindData->FileNameLength); - cifsFile->srch_inf.resume_key = pFindData->FileIndex; - } else if (level == SMB_FIND_FILE_INFO_STANDARD) { - FIND_FILE_STANDARD_INFO *pFindData = - (FIND_FILE_STANDARD_INFO *)current_entry; - filename = &pFindData->FileName[0]; - /* one byte length, no name conversion */ - len = (unsigned int)pFindData->FileNameLength; - cifsFile->srch_inf.resume_key = pFindData->ResumeKey; - } else { - cFYI(1, ("Unknown findfirst level %d", level)); - return -EINVAL; - } - cifsFile->srch_inf.resume_name_len = len; - cifsFile->srch_inf.presume_name = filename; - return rc; -} int cifs_readdir(struct file *file, void *direntry, filldir_t filldir) { diff --git a/fs/cifs/sess.c b/fs/cifs/sess.c index 252fdc0567f..2851d5da0c8 100644 --- a/fs/cifs/sess.c +++ b/fs/cifs/sess.c @@ -624,8 +624,10 @@ CIFS_SessSetup(unsigned int xid, struct cifsSesInfo *ses, int first_time, ses, nls_cp); ssetup_exit: - if (spnego_key) + if (spnego_key) { + key_revoke(spnego_key); key_put(spnego_key); + } kfree(str_area); if (resp_buf_type == CIFS_SMALL_BUFFER) { cFYI(1, ("ssetup freeing small buf %p", iov[0].iov_base)); diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c index e286db9f5ee..bf0e6d8e382 100644 --- a/fs/cifs/transport.c +++ b/fs/cifs/transport.c @@ -50,8 +50,7 @@ AllocMidQEntry(const struct smb_hdr *smb_buffer, struct cifsSesInfo *ses) return NULL; } - temp = (struct mid_q_entry *) mempool_alloc(cifs_mid_poolp, - GFP_KERNEL | GFP_NOFS); + temp = mempool_alloc(cifs_mid_poolp, GFP_NOFS); if (temp == NULL) return temp; else { diff --git a/fs/dcache.c b/fs/dcache.c index 80e93956ace..e7a1a99b746 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -1395,6 +1395,10 @@ struct dentry * __d_lookup(struct dentry * parent, struct qstr * name) if (dentry->d_parent != parent) goto next; + /* non-existing due to RCU? */ + if (d_unhashed(dentry)) + goto next; + /* * It is safe to compare names since d_move() cannot * change the qstr (protected by d_lock). @@ -1410,10 +1414,8 @@ struct dentry * __d_lookup(struct dentry * parent, struct qstr * name) goto next; } - if (!d_unhashed(dentry)) { - atomic_inc(&dentry->d_count); - found = dentry; - } + atomic_inc(&dentry->d_count); + found = dentry; spin_unlock(&dentry->d_lock); break; next: diff --git a/fs/dlm/config.c b/fs/dlm/config.c index 89d2fb7b991..fd9859f92fa 100644 --- a/fs/dlm/config.c +++ b/fs/dlm/config.c @@ -14,6 +14,9 @@ #include <linux/kernel.h> #include <linux/module.h> #include <linux/configfs.h> +#include <linux/in.h> +#include <linux/in6.h> +#include <net/ipv6.h> #include <net/sock.h> #include "config.h" @@ -377,24 +380,24 @@ static struct config_item_type node_type = { .ct_owner = THIS_MODULE, }; -static struct dlm_cluster *to_cluster(struct config_item *i) +static struct dlm_cluster *config_item_to_cluster(struct config_item *i) { return i ? container_of(to_config_group(i), struct dlm_cluster, group) : NULL; } -static struct dlm_space *to_space(struct config_item *i) +static struct dlm_space *config_item_to_space(struct config_item *i) { return i ? container_of(to_config_group(i), struct dlm_space, group) : NULL; } -static struct dlm_comm *to_comm(struct config_item *i) +static struct dlm_comm *config_item_to_comm(struct config_item *i) { return i ? container_of(i, struct dlm_comm, item) : NULL; } -static struct dlm_node *to_node(struct config_item *i) +static struct dlm_node *config_item_to_node(struct config_item *i) { return i ? container_of(i, struct dlm_node, item) : NULL; } @@ -450,7 +453,7 @@ static struct config_group *make_cluster(struct config_group *g, static void drop_cluster(struct config_group *g, struct config_item *i) { - struct dlm_cluster *cl = to_cluster(i); + struct dlm_cluster *cl = config_item_to_cluster(i); struct config_item *tmp; int j; @@ -468,7 +471,7 @@ static void drop_cluster(struct config_group *g, struct config_item *i) static void release_cluster(struct config_item *i) { - struct dlm_cluster *cl = to_cluster(i); + struct dlm_cluster *cl = config_item_to_cluster(i); kfree(cl->group.default_groups); kfree(cl); } @@ -507,7 +510,7 @@ static struct config_group *make_space(struct config_group *g, const char *name) static void drop_space(struct config_group *g, struct config_item *i) { - struct dlm_space *sp = to_space(i); + struct dlm_space *sp = config_item_to_space(i); struct config_item *tmp; int j; @@ -524,7 +527,7 @@ static void drop_space(struct config_group *g, struct config_item *i) static void release_space(struct config_item *i) { - struct dlm_space *sp = to_space(i); + struct dlm_space *sp = config_item_to_space(i); kfree(sp->group.default_groups); kfree(sp); } @@ -546,7 +549,7 @@ static struct config_item *make_comm(struct config_group *g, const char *name) static void drop_comm(struct config_group *g, struct config_item *i) { - struct dlm_comm *cm = to_comm(i); + struct dlm_comm *cm = config_item_to_comm(i); if (local_comm == cm) local_comm = NULL; dlm_lowcomms_close(cm->nodeid); @@ -557,13 +560,13 @@ static void drop_comm(struct config_group *g, struct config_item *i) static void release_comm(struct config_item *i) { - struct dlm_comm *cm = to_comm(i); + struct dlm_comm *cm = config_item_to_comm(i); kfree(cm); } static struct config_item *make_node(struct config_group *g, const char *name) { - struct dlm_space *sp = to_space(g->cg_item.ci_parent); + struct dlm_space *sp = config_item_to_space(g->cg_item.ci_parent); struct dlm_node *nd; nd = kzalloc(sizeof(struct dlm_node), GFP_KERNEL); @@ -585,8 +588,8 @@ static struct config_item *make_node(struct config_group *g, const char *name) static void drop_node(struct config_group *g, struct config_item *i) { - struct dlm_space *sp = to_space(g->cg_item.ci_parent); - struct dlm_node *nd = to_node(i); + struct dlm_space *sp = config_item_to_space(g->cg_item.ci_parent); + struct dlm_node *nd = config_item_to_node(i); mutex_lock(&sp->members_lock); list_del(&nd->list); @@ -598,7 +601,7 @@ static void drop_node(struct config_group *g, struct config_item *i) static void release_node(struct config_item *i) { - struct dlm_node *nd = to_node(i); + struct dlm_node *nd = config_item_to_node(i); kfree(nd); } @@ -632,7 +635,7 @@ void dlm_config_exit(void) static ssize_t show_cluster(struct config_item *i, struct configfs_attribute *a, char *buf) { - struct dlm_cluster *cl = to_cluster(i); + struct dlm_cluster *cl = config_item_to_cluster(i); struct cluster_attribute *cla = container_of(a, struct cluster_attribute, attr); return cla->show ? cla->show(cl, buf) : 0; @@ -642,7 +645,7 @@ static ssize_t store_cluster(struct config_item *i, struct configfs_attribute *a, const char *buf, size_t len) { - struct dlm_cluster *cl = to_cluster(i); + struct dlm_cluster *cl = config_item_to_cluster(i); struct cluster_attribute *cla = container_of(a, struct cluster_attribute, attr); return cla->store ? cla->store(cl, buf, len) : -EINVAL; @@ -651,7 +654,7 @@ static ssize_t store_cluster(struct config_item *i, static ssize_t show_comm(struct config_item *i, struct configfs_attribute *a, char *buf) { - struct dlm_comm *cm = to_comm(i); + struct dlm_comm *cm = config_item_to_comm(i); struct comm_attribute *cma = container_of(a, struct comm_attribute, attr); return cma->show ? cma->show(cm, buf) : 0; @@ -660,7 +663,7 @@ static ssize_t show_comm(struct config_item *i, struct configfs_attribute *a, static ssize_t store_comm(struct config_item *i, struct configfs_attribute *a, const char *buf, size_t len) { - struct dlm_comm *cm = to_comm(i); + struct dlm_comm *cm = config_item_to_comm(i); struct comm_attribute *cma = container_of(a, struct comm_attribute, attr); return cma->store ? cma->store(cm, buf, len) : -EINVAL; @@ -714,7 +717,7 @@ static ssize_t comm_addr_write(struct dlm_comm *cm, const char *buf, size_t len) static ssize_t show_node(struct config_item *i, struct configfs_attribute *a, char *buf) { - struct dlm_node *nd = to_node(i); + struct dlm_node *nd = config_item_to_node(i); struct node_attribute *nda = container_of(a, struct node_attribute, attr); return nda->show ? nda->show(nd, buf) : 0; @@ -723,7 +726,7 @@ static ssize_t show_node(struct config_item *i, struct configfs_attribute *a, static ssize_t store_node(struct config_item *i, struct configfs_attribute *a, const char *buf, size_t len) { - struct dlm_node *nd = to_node(i); + struct dlm_node *nd = config_item_to_node(i); struct node_attribute *nda = container_of(a, struct node_attribute, attr); return nda->store ? nda->store(nd, buf, len) : -EINVAL; @@ -768,7 +771,7 @@ static struct dlm_space *get_space(char *name) i = config_group_find_item(space_list, name); mutex_unlock(&space_list->cg_subsys->su_mutex); - return to_space(i); + return config_item_to_space(i); } static void put_space(struct dlm_space *sp) @@ -776,6 +779,33 @@ static void put_space(struct dlm_space *sp) config_item_put(&sp->group.cg_item); } +static int addr_compare(struct sockaddr_storage *x, struct sockaddr_storage *y) +{ + switch (x->ss_family) { + case AF_INET: { + struct sockaddr_in *sinx = (struct sockaddr_in *)x; + struct sockaddr_in *siny = (struct sockaddr_in *)y; + if (sinx->sin_addr.s_addr != siny->sin_addr.s_addr) + return 0; + if (sinx->sin_port != siny->sin_port) + return 0; + break; + } + case AF_INET6: { + struct sockaddr_in6 *sinx = (struct sockaddr_in6 *)x; + struct sockaddr_in6 *siny = (struct sockaddr_in6 *)y; + if (!ipv6_addr_equal(&sinx->sin6_addr, &siny->sin6_addr)) + return 0; + if (sinx->sin6_port != siny->sin6_port) + return 0; + break; + } + default: + return 0; + } + return 1; +} + static struct dlm_comm *get_comm(int nodeid, struct sockaddr_storage *addr) { struct config_item *i; @@ -788,7 +818,7 @@ static struct dlm_comm *get_comm(int nodeid, struct sockaddr_storage *addr) mutex_lock(&clusters_root.subsys.su_mutex); list_for_each_entry(i, &comm_list->cg_children, ci_entry) { - cm = to_comm(i); + cm = config_item_to_comm(i); if (nodeid) { if (cm->nodeid != nodeid) @@ -797,8 +827,7 @@ static struct dlm_comm *get_comm(int nodeid, struct sockaddr_storage *addr) config_item_get(i); break; } else { - if (!cm->addr_count || - memcmp(cm->addr[0], addr, sizeof(*addr))) + if (!cm->addr_count || !addr_compare(cm->addr[0], addr)) continue; found = 1; config_item_get(i); diff --git a/fs/dlm/dlm_internal.h b/fs/dlm/dlm_internal.h index 5a7ac33b629..868e4c9ef12 100644 --- a/fs/dlm/dlm_internal.h +++ b/fs/dlm/dlm_internal.h @@ -2,7 +2,7 @@ ******************************************************************************* ** ** Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved. -** Copyright (C) 2004-2007 Red Hat, Inc. All rights reserved. +** Copyright (C) 2004-2008 Red Hat, Inc. All rights reserved. ** ** This copyrighted material is made available to anyone wishing to use, ** modify, copy, or redistribute it subject to the terms and conditions @@ -441,8 +441,11 @@ struct dlm_ls { uint32_t ls_global_id; /* global unique lockspace ID */ uint32_t ls_exflags; int ls_lvblen; - int ls_count; /* reference count */ + int ls_count; /* refcount of processes in + the dlm using this ls */ + int ls_create_count; /* create/release refcount */ unsigned long ls_flags; /* LSFL_ */ + unsigned long ls_scan_time; struct kobject ls_kobj; struct dlm_rsbtable *ls_rsbtbl; diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c index 499e16759e9..d910501de6d 100644 --- a/fs/dlm/lockspace.c +++ b/fs/dlm/lockspace.c @@ -2,7 +2,7 @@ ******************************************************************************* ** ** Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved. -** Copyright (C) 2004-2007 Red Hat, Inc. All rights reserved. +** Copyright (C) 2004-2008 Red Hat, Inc. All rights reserved. ** ** This copyrighted material is made available to anyone wishing to use, ** modify, copy, or redistribute it subject to the terms and conditions @@ -23,6 +23,7 @@ #include "lock.h" #include "recover.h" #include "requestqueue.h" +#include "user.h" static int ls_count; static struct mutex ls_lock; @@ -211,19 +212,41 @@ void dlm_lockspace_exit(void) kset_unregister(dlm_kset); } +static struct dlm_ls *find_ls_to_scan(void) +{ + struct dlm_ls *ls; + + spin_lock(&lslist_lock); + list_for_each_entry(ls, &lslist, ls_list) { + if (time_after_eq(jiffies, ls->ls_scan_time + + dlm_config.ci_scan_secs * HZ)) { + spin_unlock(&lslist_lock); + return ls; + } + } + spin_unlock(&lslist_lock); + return NULL; +} + static int dlm_scand(void *data) { struct dlm_ls *ls; + int timeout_jiffies = dlm_config.ci_scan_secs * HZ; while (!kthread_should_stop()) { - list_for_each_entry(ls, &lslist, ls_list) { + ls = find_ls_to_scan(); + if (ls) { if (dlm_lock_recovery_try(ls)) { + ls->ls_scan_time = jiffies; dlm_scan_rsbs(ls); dlm_scan_timeout(ls); dlm_unlock_recovery(ls); + } else { + ls->ls_scan_time += HZ; } + } else { + schedule_timeout_interruptible(timeout_jiffies); } - schedule_timeout_interruptible(dlm_config.ci_scan_secs * HZ); } return 0; } @@ -246,23 +269,6 @@ static void dlm_scand_stop(void) kthread_stop(scand_task); } -static struct dlm_ls *dlm_find_lockspace_name(char *name, int namelen) -{ - struct dlm_ls *ls; - - spin_lock(&lslist_lock); - - list_for_each_entry(ls, &lslist, ls_list) { - if (ls->ls_namelen == namelen && - memcmp(ls->ls_name, name, namelen) == 0) - goto out; - } - ls = NULL; - out: - spin_unlock(&lslist_lock); - return ls; -} - struct dlm_ls *dlm_find_lockspace_global(uint32_t id) { struct dlm_ls *ls; @@ -327,6 +333,7 @@ static void remove_lockspace(struct dlm_ls *ls) for (;;) { spin_lock(&lslist_lock); if (ls->ls_count == 0) { + WARN_ON(ls->ls_create_count != 0); list_del(&ls->ls_list); spin_unlock(&lslist_lock); return; @@ -381,7 +388,7 @@ static int new_lockspace(char *name, int namelen, void **lockspace, uint32_t flags, int lvblen) { struct dlm_ls *ls; - int i, size, error = -ENOMEM; + int i, size, error; int do_unreg = 0; if (namelen > DLM_LOCKSPACE_LEN) @@ -393,12 +400,37 @@ static int new_lockspace(char *name, int namelen, void **lockspace, if (!try_module_get(THIS_MODULE)) return -EINVAL; - ls = dlm_find_lockspace_name(name, namelen); - if (ls) { - *lockspace = ls; + if (!dlm_user_daemon_available()) { + module_put(THIS_MODULE); + return -EUNATCH; + } + + error = 0; + + spin_lock(&lslist_lock); + list_for_each_entry(ls, &lslist, ls_list) { + WARN_ON(ls->ls_create_count <= 0); + if (ls->ls_namelen != namelen) + continue; + if (memcmp(ls->ls_name, name, namelen)) + continue; + if (flags & DLM_LSFL_NEWEXCL) { + error = -EEXIST; + break; + } + ls->ls_create_count++; module_put(THIS_MODULE); - return -EEXIST; + error = 1; /* not an error, return 0 */ + break; } + spin_unlock(&lslist_lock); + + if (error < 0) + goto out; + if (error) + goto ret_zero; + + error = -ENOMEM; ls = kzalloc(sizeof(struct dlm_ls) + namelen, GFP_KERNEL); if (!ls) @@ -408,6 +440,7 @@ static int new_lockspace(char *name, int namelen, void **lockspace, ls->ls_lvblen = lvblen; ls->ls_count = 0; ls->ls_flags = 0; + ls->ls_scan_time = jiffies; if (flags & DLM_LSFL_TIMEWARN) set_bit(LSFL_TIMEWARN, &ls->ls_flags); @@ -418,8 +451,9 @@ static int new_lockspace(char *name, int namelen, void **lockspace, ls->ls_allocation = GFP_KERNEL; /* ls_exflags are forced to match among nodes, and we don't - need to require all nodes to have TIMEWARN or FS set */ - ls->ls_exflags = (flags & ~(DLM_LSFL_TIMEWARN | DLM_LSFL_FS)); + need to require all nodes to have some flags set */ + ls->ls_exflags = (flags & ~(DLM_LSFL_TIMEWARN | DLM_LSFL_FS | + DLM_LSFL_NEWEXCL)); size = dlm_config.ci_rsbtbl_size; ls->ls_rsbtbl_size = size; @@ -510,6 +544,7 @@ static int new_lockspace(char *name, int namelen, void **lockspace, down_write(&ls->ls_in_recovery); spin_lock(&lslist_lock); + ls->ls_create_count = 1; list_add(&ls->ls_list, &lslist); spin_unlock(&lslist_lock); @@ -548,7 +583,7 @@ static int new_lockspace(char *name, int namelen, void **lockspace, dlm_create_debug_file(ls); log_debug(ls, "join complete"); - + ret_zero: *lockspace = ls; return 0; @@ -635,13 +670,34 @@ static int release_lockspace(struct dlm_ls *ls, int force) struct dlm_lkb *lkb; struct dlm_rsb *rsb; struct list_head *head; - int i; - int busy = lockspace_busy(ls); + int i, busy, rv; + + busy = lockspace_busy(ls); + + spin_lock(&lslist_lock); + if (ls->ls_create_count == 1) { + if (busy > force) + rv = -EBUSY; + else { + /* remove_lockspace takes ls off lslist */ + ls->ls_create_count = 0; + rv = 0; + } + } else if (ls->ls_create_count > 1) { + rv = --ls->ls_create_count; + } else { + rv = -EINVAL; + } + spin_unlock(&lslist_lock); - if (busy > force) - return -EBUSY; + if (rv) { + log_debug(ls, "release_lockspace no remove %d", rv); + return rv; + } + + dlm_device_deregister(ls); - if (force < 3) + if (force < 3 && dlm_user_daemon_available()) do_uevent(ls, 0); dlm_recoverd_stop(ls); @@ -720,15 +776,10 @@ static int release_lockspace(struct dlm_ls *ls, int force) dlm_clear_members(ls); dlm_clear_members_gone(ls); kfree(ls->ls_node_array); + log_debug(ls, "release_lockspace final free"); kobject_put(&ls->ls_kobj); /* The ls structure will be freed when the kobject is done with */ - mutex_lock(&ls_lock); - ls_count--; - if (!ls_count) - threads_stop(); - mutex_unlock(&ls_lock); - module_put(THIS_MODULE); return 0; } @@ -750,11 +801,38 @@ static int release_lockspace(struct dlm_ls *ls, int force) int dlm_release_lockspace(void *lockspace, int force) { struct dlm_ls *ls; + int error; ls = dlm_find_lockspace_local(lockspace); if (!ls) return -EINVAL; dlm_put_lockspace(ls); - return release_lockspace(ls, force); + + mutex_lock(&ls_lock); + error = release_lockspace(ls, force); + if (!error) + ls_count--; + else if (!ls_count) + threads_stop(); + mutex_unlock(&ls_lock); + + return error; +} + +void dlm_stop_lockspaces(void) +{ + struct dlm_ls *ls; + + restart: + spin_lock(&lslist_lock); + list_for_each_entry(ls, &lslist, ls_list) { + if (!test_bit(LSFL_RUNNING, &ls->ls_flags)) + continue; + spin_unlock(&lslist_lock); + log_error(ls, "no userland control daemon, stopping lockspace"); + dlm_ls_stop(ls); + goto restart; + } + spin_unlock(&lslist_lock); } diff --git a/fs/dlm/lockspace.h b/fs/dlm/lockspace.h index 891eabbdd02..f879f87901f 100644 --- a/fs/dlm/lockspace.h +++ b/fs/dlm/lockspace.h @@ -20,6 +20,7 @@ struct dlm_ls *dlm_find_lockspace_global(uint32_t id); struct dlm_ls *dlm_find_lockspace_local(void *id); struct dlm_ls *dlm_find_lockspace_device(int minor); void dlm_put_lockspace(struct dlm_ls *ls); +void dlm_stop_lockspaces(void); #endif /* __LOCKSPACE_DOT_H__ */ diff --git a/fs/dlm/user.c b/fs/dlm/user.c index 34f14a14fb4..b3832c67194 100644 --- a/fs/dlm/user.c +++ b/fs/dlm/user.c @@ -1,5 +1,5 @@ /* - * Copyright (C) 2006-2007 Red Hat, Inc. All rights reserved. + * Copyright (C) 2006-2008 Red Hat, Inc. All rights reserved. * * This copyrighted material is made available to anyone wishing to use, * modify, copy, or redistribute it subject to the terms and conditions @@ -15,7 +15,6 @@ #include <linux/poll.h> #include <linux/signal.h> #include <linux/spinlock.h> -#include <linux/smp_lock.h> #include <linux/dlm.h> #include <linux/dlm_device.h> @@ -27,6 +26,8 @@ static const char name_prefix[] = "dlm"; static const struct file_operations device_fops; +static atomic_t dlm_monitor_opened; +static int dlm_monitor_unused = 1; #ifdef CONFIG_COMPAT @@ -340,10 +341,15 @@ static int device_user_deadlock(struct dlm_user_proc *proc, return error; } -static int create_misc_device(struct dlm_ls *ls, char *name) +static int dlm_device_register(struct dlm_ls *ls, char *name) { int error, len; + /* The device is already registered. This happens when the + lockspace is created multiple times from userspace. */ + if (ls->ls_device.name) + return 0; + error = -ENOMEM; len = strlen(name) + strlen(name_prefix) + 2; ls->ls_device.name = kzalloc(len, GFP_KERNEL); @@ -363,6 +369,22 @@ fail: return error; } +int dlm_device_deregister(struct dlm_ls *ls) +{ + int error; + + /* The device is not registered. This happens when the lockspace + was never used from userspace, or when device_create_lockspace() + calls dlm_release_lockspace() after the register fails. */ + if (!ls->ls_device.name) + return 0; + + error = misc_deregister(&ls->ls_device); + if (!error) + kfree(ls->ls_device.name); + return error; +} + static int device_user_purge(struct dlm_user_proc *proc, struct dlm_purge_params *params) { @@ -397,7 +419,7 @@ static int device_create_lockspace(struct dlm_lspace_params *params) if (!ls) return -ENOENT; - error = create_misc_device(ls, params->name); + error = dlm_device_register(ls, params->name); dlm_put_lockspace(ls); if (error) @@ -421,31 +443,22 @@ static int device_remove_lockspace(struct dlm_lspace_params *params) if (!ls) return -ENOENT; - /* Deregister the misc device first, so we don't have - * a device that's not attached to a lockspace. If - * dlm_release_lockspace fails then we can recreate it - */ - error = misc_deregister(&ls->ls_device); - if (error) { - dlm_put_lockspace(ls); - goto out; - } - kfree(ls->ls_device.name); - if (params->flags & DLM_USER_LSFLG_FORCEFREE) force = 2; lockspace = ls->ls_local_handle; + dlm_put_lockspace(ls); - /* dlm_release_lockspace waits for references to go to zero, - so all processes will need to close their device for the ls - before the release will procede */ + /* The final dlm_release_lockspace waits for references to go to + zero, so all processes will need to close their device for the + ls before the release will proceed. release also calls the + device_deregister above. Converting a positive return value + from release to zero means that userspace won't know when its + release was the final one, but it shouldn't need to know. */ - dlm_put_lockspace(ls); error = dlm_release_lockspace(lockspace, force); - if (error) - create_misc_device(ls, ls->ls_name); - out: + if (error > 0) + error = 0; return error; } @@ -623,17 +636,13 @@ static int device_open(struct inode *inode, struct file *file) struct dlm_user_proc *proc; struct dlm_ls *ls; - lock_kernel(); ls = dlm_find_lockspace_device(iminor(inode)); - if (!ls) { - unlock_kernel(); + if (!ls) return -ENOENT; - } proc = kzalloc(sizeof(struct dlm_user_proc), GFP_KERNEL); if (!proc) { dlm_put_lockspace(ls); - unlock_kernel(); return -ENOMEM; } @@ -645,7 +654,6 @@ static int device_open(struct inode *inode, struct file *file) spin_lock_init(&proc->locks_spin); init_waitqueue_head(&proc->wait); file->private_data = proc; - unlock_kernel(); return 0; } @@ -878,9 +886,28 @@ static unsigned int device_poll(struct file *file, poll_table *wait) return 0; } +int dlm_user_daemon_available(void) +{ + /* dlm_controld hasn't started (or, has started, but not + properly populated configfs) */ + + if (!dlm_our_nodeid()) + return 0; + + /* This is to deal with versions of dlm_controld that don't + know about the monitor device. We assume that if the + dlm_controld was started (above), but the monitor device + was never opened, that it's an old version. dlm_controld + should open the monitor device before populating configfs. */ + + if (dlm_monitor_unused) + return 1; + + return atomic_read(&dlm_monitor_opened) ? 1 : 0; +} + static int ctl_device_open(struct inode *inode, struct file *file) { - cycle_kernel_lock(); file->private_data = NULL; return 0; } @@ -890,6 +917,20 @@ static int ctl_device_close(struct inode *inode, struct file *file) return 0; } +static int monitor_device_open(struct inode *inode, struct file *file) +{ + atomic_inc(&dlm_monitor_opened); + dlm_monitor_unused = 0; + return 0; +} + +static int monitor_device_close(struct inode *inode, struct file *file) +{ + if (atomic_dec_and_test(&dlm_monitor_opened)) + dlm_stop_lockspaces(); + return 0; +} + static const struct file_operations device_fops = { .open = device_open, .release = device_close, @@ -913,19 +954,42 @@ static struct miscdevice ctl_device = { .minor = MISC_DYNAMIC_MINOR, }; +static const struct file_operations monitor_device_fops = { + .open = monitor_device_open, + .release = monitor_device_close, + .owner = THIS_MODULE, +}; + +static struct miscdevice monitor_device = { + .name = "dlm-monitor", + .fops = &monitor_device_fops, + .minor = MISC_DYNAMIC_MINOR, +}; + int __init dlm_user_init(void) { int error; + atomic_set(&dlm_monitor_opened, 0); + error = misc_register(&ctl_device); - if (error) + if (error) { log_print("misc_register failed for control device"); + goto out; + } + error = misc_register(&monitor_device); + if (error) { + log_print("misc_register failed for monitor device"); + misc_deregister(&ctl_device); + } + out: return error; } void dlm_user_exit(void) { misc_deregister(&ctl_device); + misc_deregister(&monitor_device); } diff --git a/fs/dlm/user.h b/fs/dlm/user.h index d38e9f3e415..35eb6a13d61 100644 --- a/fs/dlm/user.h +++ b/fs/dlm/user.h @@ -1,5 +1,5 @@ /* - * Copyright (C) 2006 Red Hat, Inc. All rights reserved. + * Copyright (C) 2006-2008 Red Hat, Inc. All rights reserved. * * This copyrighted material is made available to anyone wishing to use, * modify, copy, or redistribute it subject to the terms and conditions @@ -12,5 +12,7 @@ void dlm_user_add_ast(struct dlm_lkb *lkb, int type); int dlm_user_init(void); void dlm_user_exit(void); +int dlm_device_deregister(struct dlm_ls *ls); +int dlm_user_daemon_available(void); #endif diff --git a/fs/exec.c b/fs/exec.c index 32993beecbe..cecee501ce7 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -752,11 +752,11 @@ static int exec_mmap(struct mm_struct *mm) tsk->active_mm = mm; activate_mm(active_mm, mm); task_unlock(tsk); - mm_update_next_owner(old_mm); arch_pick_mmap_layout(mm); if (old_mm) { up_read(&old_mm->mmap_sem); BUG_ON(active_mm != old_mm); + mm_update_next_owner(old_mm); mmput(old_mm); return 0; } diff --git a/fs/ext2/ext2.h b/fs/ext2/ext2.h index 47d88da2d33..bae998c1e44 100644 --- a/fs/ext2/ext2.h +++ b/fs/ext2/ext2.h @@ -133,6 +133,8 @@ extern void ext2_truncate (struct inode *); extern int ext2_setattr (struct dentry *, struct iattr *); extern void ext2_set_inode_flags(struct inode *inode); extern void ext2_get_inode_flags(struct ext2_inode_info *); +extern int ext2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, + u64 start, u64 len); int __ext2_write_begin(struct file *file, struct address_space *mapping, loff_t pos, unsigned len, unsigned flags, struct page **pagep, void **fsdata); diff --git a/fs/ext2/file.c b/fs/ext2/file.c index 5f2fa9c3629..45ed0712218 100644 --- a/fs/ext2/file.c +++ b/fs/ext2/file.c @@ -86,4 +86,5 @@ const struct inode_operations ext2_file_inode_operations = { #endif .setattr = ext2_setattr, .permission = ext2_permission, + .fiemap = ext2_fiemap, }; diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c index 991d6dfeb51..7658b33e265 100644 --- a/fs/ext2/inode.c +++ b/fs/ext2/inode.c @@ -31,6 +31,7 @@ #include <linux/writeback.h> #include <linux/buffer_head.h> #include <linux/mpage.h> +#include <linux/fiemap.h> #include "ext2.h" #include "acl.h" #include "xip.h" @@ -704,6 +705,13 @@ int ext2_get_block(struct inode *inode, sector_t iblock, struct buffer_head *bh_ } +int ext2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, + u64 start, u64 len) +{ + return generic_block_fiemap(inode, fieinfo, start, len, + ext2_get_block); +} + static int ext2_writepage(struct page *page, struct writeback_control *wbc) { return block_write_full_page(page, ext2_get_block, wbc); diff --git a/fs/ext3/file.c b/fs/ext3/file.c index acc4913d301..3be1e0689c9 100644 --- a/fs/ext3/file.c +++ b/fs/ext3/file.c @@ -134,5 +134,6 @@ const struct inode_operations ext3_file_inode_operations = { .removexattr = generic_removexattr, #endif .permission = ext3_permission, + .fiemap = ext3_fiemap, }; diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c index 507d8689b11..ebfec4d0148 100644 --- a/fs/ext3/inode.c +++ b/fs/ext3/inode.c @@ -36,6 +36,7 @@ #include <linux/mpage.h> #include <linux/uio.h> #include <linux/bio.h> +#include <linux/fiemap.h> #include "xattr.h" #include "acl.h" @@ -981,6 +982,13 @@ out: return ret; } +int ext3_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, + u64 start, u64 len) +{ + return generic_block_fiemap(inode, fieinfo, start, len, + ext3_get_block); +} + /* * `handle' can be NULL if create is zero */ diff --git a/fs/ext4/Makefile b/fs/ext4/Makefile index ac6fa8ca0a2..a8ff003a00f 100644 --- a/fs/ext4/Makefile +++ b/fs/ext4/Makefile @@ -2,12 +2,12 @@ # Makefile for the linux ext4-filesystem routines. # -obj-$(CONFIG_EXT4DEV_FS) += ext4dev.o +obj-$(CONFIG_EXT4_FS) += ext4.o -ext4dev-y := balloc.o bitmap.o dir.o file.o fsync.o ialloc.o inode.o \ +ext4-y := balloc.o bitmap.o dir.o file.o fsync.o ialloc.o inode.o \ ioctl.o namei.o super.o symlink.o hash.o resize.o extents.o \ ext4_jbd2.o migrate.o mballoc.o -ext4dev-$(CONFIG_EXT4DEV_FS_XATTR) += xattr.o xattr_user.o xattr_trusted.o -ext4dev-$(CONFIG_EXT4DEV_FS_POSIX_ACL) += acl.o -ext4dev-$(CONFIG_EXT4DEV_FS_SECURITY) += xattr_security.o +ext4-$(CONFIG_EXT4_FS_XATTR) += xattr.o xattr_user.o xattr_trusted.o +ext4-$(CONFIG_EXT4_FS_POSIX_ACL) += acl.o +ext4-$(CONFIG_EXT4_FS_SECURITY) += xattr_security.o diff --git a/fs/ext4/acl.h b/fs/ext4/acl.h index cd2b855a07d..cb45257a246 100644 --- a/fs/ext4/acl.h +++ b/fs/ext4/acl.h @@ -51,18 +51,18 @@ static inline int ext4_acl_count(size_t size) } } -#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL +#ifdef CONFIG_EXT4_FS_POSIX_ACL /* Value for inode->u.ext4_i.i_acl and inode->u.ext4_i.i_default_acl if the ACL has not been cached */ #define EXT4_ACL_NOT_CACHED ((void *)-1) /* acl.c */ -extern int ext4_permission (struct inode *, int); -extern int ext4_acl_chmod (struct inode *); -extern int ext4_init_acl (handle_t *, struct inode *, struct inode *); +extern int ext4_permission(struct inode *, int); +extern int ext4_acl_chmod(struct inode *); +extern int ext4_init_acl(handle_t *, struct inode *, struct inode *); -#else /* CONFIG_EXT4DEV_FS_POSIX_ACL */ +#else /* CONFIG_EXT4_FS_POSIX_ACL */ #include <linux/sched.h> #define ext4_permission NULL @@ -77,5 +77,5 @@ ext4_init_acl(handle_t *handle, struct inode *inode, struct inode *dir) { return 0; } -#endif /* CONFIG_EXT4DEV_FS_POSIX_ACL */ +#endif /* CONFIG_EXT4_FS_POSIX_ACL */ diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c index e9fa960ba6d..bd2ece22882 100644 --- a/fs/ext4/balloc.c +++ b/fs/ext4/balloc.c @@ -83,6 +83,7 @@ static int ext4_group_used_meta_blocks(struct super_block *sb, } return used_blocks; } + /* Initializes an uninitialized block bitmap if given, and returns the * number of blocks free in the group. */ unsigned ext4_init_block_bitmap(struct super_block *sb, struct buffer_head *bh, @@ -132,7 +133,7 @@ unsigned ext4_init_block_bitmap(struct super_block *sb, struct buffer_head *bh, */ group_blocks = ext4_blocks_count(sbi->s_es) - le32_to_cpu(sbi->s_es->s_first_data_block) - - (EXT4_BLOCKS_PER_GROUP(sb) * (sbi->s_groups_count -1)); + (EXT4_BLOCKS_PER_GROUP(sb) * (sbi->s_groups_count - 1)); } else { group_blocks = EXT4_BLOCKS_PER_GROUP(sb); } @@ -200,20 +201,20 @@ unsigned ext4_init_block_bitmap(struct super_block *sb, struct buffer_head *bh, * @bh: pointer to the buffer head to store the block * group descriptor */ -struct ext4_group_desc * ext4_get_group_desc(struct super_block * sb, +struct ext4_group_desc * ext4_get_group_desc(struct super_block *sb, ext4_group_t block_group, - struct buffer_head ** bh) + struct buffer_head **bh) { unsigned long group_desc; unsigned long offset; - struct ext4_group_desc * desc; + struct ext4_group_desc *desc; struct ext4_sb_info *sbi = EXT4_SB(sb); if (block_group >= sbi->s_groups_count) { - ext4_error (sb, "ext4_get_group_desc", - "block_group >= groups_count - " - "block_group = %lu, groups_count = %lu", - block_group, sbi->s_groups_count); + ext4_error(sb, "ext4_get_group_desc", + "block_group >= groups_count - " + "block_group = %lu, groups_count = %lu", + block_group, sbi->s_groups_count); return NULL; } @@ -222,10 +223,10 @@ struct ext4_group_desc * ext4_get_group_desc(struct super_block * sb, group_desc = block_group >> EXT4_DESC_PER_BLOCK_BITS(sb); offset = block_group & (EXT4_DESC_PER_BLOCK(sb) - 1); if (!sbi->s_group_desc[group_desc]) { - ext4_error (sb, "ext4_get_group_desc", - "Group descriptor not loaded - " - "block_group = %lu, group_desc = %lu, desc = %lu", - block_group, group_desc, offset); + ext4_error(sb, "ext4_get_group_desc", + "Group descriptor not loaded - " + "block_group = %lu, group_desc = %lu, desc = %lu", + block_group, group_desc, offset); return NULL; } @@ -302,8 +303,8 @@ err_out: struct buffer_head * ext4_read_block_bitmap(struct super_block *sb, ext4_group_t block_group) { - struct ext4_group_desc * desc; - struct buffer_head * bh = NULL; + struct ext4_group_desc *desc; + struct buffer_head *bh = NULL; ext4_fsblk_t bitmap_blk; desc = ext4_get_group_desc(sb, block_group, NULL); @@ -318,9 +319,11 @@ ext4_read_block_bitmap(struct super_block *sb, ext4_group_t block_group) block_group, bitmap_blk); return NULL; } - if (bh_uptodate_or_lock(bh)) + if (buffer_uptodate(bh) && + !(desc->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT))) return bh; + lock_buffer(bh); spin_lock(sb_bgl_lock(EXT4_SB(sb), block_group)); if (desc->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) { ext4_init_block_bitmap(sb, bh, block_group, desc); @@ -345,301 +348,6 @@ ext4_read_block_bitmap(struct super_block *sb, ext4_group_t block_group) */ return bh; } -/* - * The reservation window structure operations - * -------------------------------------------- - * Operations include: - * dump, find, add, remove, is_empty, find_next_reservable_window, etc. - * - * We use a red-black tree to represent per-filesystem reservation - * windows. - * - */ - -/** - * __rsv_window_dump() -- Dump the filesystem block allocation reservation map - * @rb_root: root of per-filesystem reservation rb tree - * @verbose: verbose mode - * @fn: function which wishes to dump the reservation map - * - * If verbose is turned on, it will print the whole block reservation - * windows(start, end). Otherwise, it will only print out the "bad" windows, - * those windows that overlap with their immediate neighbors. - */ -#if 1 -static void __rsv_window_dump(struct rb_root *root, int verbose, - const char *fn) -{ - struct rb_node *n; - struct ext4_reserve_window_node *rsv, *prev; - int bad; - -restart: - n = rb_first(root); - bad = 0; - prev = NULL; - - printk("Block Allocation Reservation Windows Map (%s):\n", fn); - while (n) { - rsv = rb_entry(n, struct ext4_reserve_window_node, rsv_node); - if (verbose) - printk("reservation window 0x%p " - "start: %llu, end: %llu\n", - rsv, rsv->rsv_start, rsv->rsv_end); - if (rsv->rsv_start && rsv->rsv_start >= rsv->rsv_end) { - printk("Bad reservation %p (start >= end)\n", - rsv); - bad = 1; - } - if (prev && prev->rsv_end >= rsv->rsv_start) { - printk("Bad reservation %p (prev->end >= start)\n", - rsv); - bad = 1; - } - if (bad) { - if (!verbose) { - printk("Restarting reservation walk in verbose mode\n"); - verbose = 1; - goto restart; - } - } - n = rb_next(n); - prev = rsv; - } - printk("Window map complete.\n"); - BUG_ON(bad); -} -#define rsv_window_dump(root, verbose) \ - __rsv_window_dump((root), (verbose), __func__) -#else -#define rsv_window_dump(root, verbose) do {} while (0) -#endif - -/** - * goal_in_my_reservation() - * @rsv: inode's reservation window - * @grp_goal: given goal block relative to the allocation block group - * @group: the current allocation block group - * @sb: filesystem super block - * - * Test if the given goal block (group relative) is within the file's - * own block reservation window range. - * - * If the reservation window is outside the goal allocation group, return 0; - * grp_goal (given goal block) could be -1, which means no specific - * goal block. In this case, always return 1. - * If the goal block is within the reservation window, return 1; - * otherwise, return 0; - */ -static int -goal_in_my_reservation(struct ext4_reserve_window *rsv, ext4_grpblk_t grp_goal, - ext4_group_t group, struct super_block *sb) -{ - ext4_fsblk_t group_first_block, group_last_block; - - group_first_block = ext4_group_first_block_no(sb, group); - group_last_block = group_first_block + (EXT4_BLOCKS_PER_GROUP(sb) - 1); - - if ((rsv->_rsv_start > group_last_block) || - (rsv->_rsv_end < group_first_block)) - return 0; - if ((grp_goal >= 0) && ((grp_goal + group_first_block < rsv->_rsv_start) - || (grp_goal + group_first_block > rsv->_rsv_end))) - return 0; - return 1; -} - -/** - * search_reserve_window() - * @rb_root: root of reservation tree - * @goal: target allocation block - * - * Find the reserved window which includes the goal, or the previous one - * if the goal is not in any window. - * Returns NULL if there are no windows or if all windows start after the goal. - */ -static struct ext4_reserve_window_node * -search_reserve_window(struct rb_root *root, ext4_fsblk_t goal) -{ - struct rb_node *n = root->rb_node; - struct ext4_reserve_window_node *rsv; - - if (!n) - return NULL; - - do { - rsv = rb_entry(n, struct ext4_reserve_window_node, rsv_node); - - if (goal < rsv->rsv_start) - n = n->rb_left; - else if (goal > rsv->rsv_end) - n = n->rb_right; - else - return rsv; - } while (n); - /* - * We've fallen off the end of the tree: the goal wasn't inside - * any particular node. OK, the previous node must be to one - * side of the interval containing the goal. If it's the RHS, - * we need to back up one. - */ - if (rsv->rsv_start > goal) { - n = rb_prev(&rsv->rsv_node); - rsv = rb_entry(n, struct ext4_reserve_window_node, rsv_node); - } - return rsv; -} - -/** - * ext4_rsv_window_add() -- Insert a window to the block reservation rb tree. - * @sb: super block - * @rsv: reservation window to add - * - * Must be called with rsv_lock hold. - */ -void ext4_rsv_window_add(struct super_block *sb, - struct ext4_reserve_window_node *rsv) -{ - struct rb_root *root = &EXT4_SB(sb)->s_rsv_window_root; - struct rb_node *node = &rsv->rsv_node; - ext4_fsblk_t start = rsv->rsv_start; - - struct rb_node ** p = &root->rb_node; - struct rb_node * parent = NULL; - struct ext4_reserve_window_node *this; - - while (*p) - { - parent = *p; - this = rb_entry(parent, struct ext4_reserve_window_node, rsv_node); - - if (start < this->rsv_start) - p = &(*p)->rb_left; - else if (start > this->rsv_end) - p = &(*p)->rb_right; - else { - rsv_window_dump(root, 1); - BUG(); - } - } - - rb_link_node(node, parent, p); - rb_insert_color(node, root); -} - -/** - * ext4_rsv_window_remove() -- unlink a window from the reservation rb tree - * @sb: super block - * @rsv: reservation window to remove - * - * Mark the block reservation window as not allocated, and unlink it - * from the filesystem reservation window rb tree. Must be called with - * rsv_lock hold. - */ -static void rsv_window_remove(struct super_block *sb, - struct ext4_reserve_window_node *rsv) -{ - rsv->rsv_start = EXT4_RESERVE_WINDOW_NOT_ALLOCATED; - rsv->rsv_end = EXT4_RESERVE_WINDOW_NOT_ALLOCATED; - rsv->rsv_alloc_hit = 0; - rb_erase(&rsv->rsv_node, &EXT4_SB(sb)->s_rsv_window_root); -} - -/* - * rsv_is_empty() -- Check if the reservation window is allocated. - * @rsv: given reservation window to check - * - * returns 1 if the end block is EXT4_RESERVE_WINDOW_NOT_ALLOCATED. - */ -static inline int rsv_is_empty(struct ext4_reserve_window *rsv) -{ - /* a valid reservation end block could not be 0 */ - return rsv->_rsv_end == EXT4_RESERVE_WINDOW_NOT_ALLOCATED; -} - -/** - * ext4_init_block_alloc_info() - * @inode: file inode structure - * - * Allocate and initialize the reservation window structure, and - * link the window to the ext4 inode structure at last - * - * The reservation window structure is only dynamically allocated - * and linked to ext4 inode the first time the open file - * needs a new block. So, before every ext4_new_block(s) call, for - * regular files, we should check whether the reservation window - * structure exists or not. In the latter case, this function is called. - * Fail to do so will result in block reservation being turned off for that - * open file. - * - * This function is called from ext4_get_blocks_handle(), also called - * when setting the reservation window size through ioctl before the file - * is open for write (needs block allocation). - * - * Needs down_write(i_data_sem) protection prior to call this function. - */ -void ext4_init_block_alloc_info(struct inode *inode) -{ - struct ext4_inode_info *ei = EXT4_I(inode); - struct ext4_block_alloc_info *block_i = ei->i_block_alloc_info; - struct super_block *sb = inode->i_sb; - - block_i = kmalloc(sizeof(*block_i), GFP_NOFS); - if (block_i) { - struct ext4_reserve_window_node *rsv = &block_i->rsv_window_node; - - rsv->rsv_start = EXT4_RESERVE_WINDOW_NOT_ALLOCATED; - rsv->rsv_end = EXT4_RESERVE_WINDOW_NOT_ALLOCATED; - - /* - * if filesystem is mounted with NORESERVATION, the goal - * reservation window size is set to zero to indicate - * block reservation is off - */ - if (!test_opt(sb, RESERVATION)) - rsv->rsv_goal_size = 0; - else - rsv->rsv_goal_size = EXT4_DEFAULT_RESERVE_BLOCKS; - rsv->rsv_alloc_hit = 0; - block_i->last_alloc_logical_block = 0; - block_i->last_alloc_physical_block = 0; - } - ei->i_block_alloc_info = block_i; -} - -/** - * ext4_discard_reservation() - * @inode: inode - * - * Discard(free) block reservation window on last file close, or truncate - * or at last iput(). - * - * It is being called in three cases: - * ext4_release_file(): last writer close the file - * ext4_clear_inode(): last iput(), when nobody link to this file. - * ext4_truncate(): when the block indirect map is about to change. - * - */ -void ext4_discard_reservation(struct inode *inode) -{ - struct ext4_inode_info *ei = EXT4_I(inode); - struct ext4_block_alloc_info *block_i = ei->i_block_alloc_info; - struct ext4_reserve_window_node *rsv; - spinlock_t *rsv_lock = &EXT4_SB(inode->i_sb)->s_rsv_window_lock; - - ext4_mb_discard_inode_preallocations(inode); - - if (!block_i) - return; - - rsv = &block_i->rsv_window_node; - if (!rsv_is_empty(&rsv->rsv_window)) { - spin_lock(rsv_lock); - if (!rsv_is_empty(&rsv->rsv_window)) - rsv_window_remove(inode->i_sb, rsv); - spin_unlock(rsv_lock); - } -} /** * ext4_free_blocks_sb() -- Free given blocks and update quota @@ -648,6 +356,13 @@ void ext4_discard_reservation(struct inode *inode) * @block: start physcial block to free * @count: number of blocks to free * @pdquot_freed_blocks: pointer to quota + * + * XXX This function is only used by the on-line resizing code, which + * should probably be fixed up to call the mballoc variant. There + * this needs to be cleaned up later; in fact, I'm not convinced this + * is 100% correct in the face of the mballoc code. The online resizing + * code needs to be fixed up to more tightly (and correctly) interlock + * with the mballoc code. */ void ext4_free_blocks_sb(handle_t *handle, struct super_block *sb, ext4_fsblk_t block, unsigned long count, @@ -659,8 +374,8 @@ void ext4_free_blocks_sb(handle_t *handle, struct super_block *sb, ext4_grpblk_t bit; unsigned long i; unsigned long overflow; - struct ext4_group_desc * desc; - struct ext4_super_block * es; + struct ext4_group_desc *desc; + struct ext4_super_block *es; struct ext4_sb_info *sbi; int err = 0, ret; ext4_grpblk_t group_freed; @@ -671,13 +386,13 @@ void ext4_free_blocks_sb(handle_t *handle, struct super_block *sb, if (block < le32_to_cpu(es->s_first_data_block) || block + count < block || block + count > ext4_blocks_count(es)) { - ext4_error (sb, "ext4_free_blocks", - "Freeing blocks not in datazone - " - "block = %llu, count = %lu", block, count); + ext4_error(sb, "ext4_free_blocks", + "Freeing blocks not in datazone - " + "block = %llu, count = %lu", block, count); goto error_return; } - ext4_debug ("freeing block(s) %llu-%llu\n", block, block + count - 1); + ext4_debug("freeing block(s) %llu-%llu\n", block, block + count - 1); do_more: overflow = 0; @@ -694,7 +409,7 @@ do_more: bitmap_bh = ext4_read_block_bitmap(sb, block_group); if (!bitmap_bh) goto error_return; - desc = ext4_get_group_desc (sb, block_group, &gd_bh); + desc = ext4_get_group_desc(sb, block_group, &gd_bh); if (!desc) goto error_return; @@ -703,10 +418,10 @@ do_more: in_range(block, ext4_inode_table(sb, desc), sbi->s_itb_per_group) || in_range(block + count - 1, ext4_inode_table(sb, desc), sbi->s_itb_per_group)) { - ext4_error (sb, "ext4_free_blocks", - "Freeing blocks in system zones - " - "Block = %llu, count = %lu", - block, count); + ext4_error(sb, "ext4_free_blocks", + "Freeing blocks in system zones - " + "Block = %llu, count = %lu", + block, count); goto error_return; } @@ -848,7 +563,7 @@ void ext4_free_blocks(handle_t *handle, struct inode *inode, ext4_fsblk_t block, unsigned long count, int metadata) { - struct super_block * sb; + struct super_block *sb; unsigned long dquot_freed_blocks; /* this isn't the right place to decide whether block is metadata @@ -859,748 +574,52 @@ void ext4_free_blocks(handle_t *handle, struct inode *inode, sb = inode->i_sb; - if (!test_opt(sb, MBALLOC) || !EXT4_SB(sb)->s_group_info) - ext4_free_blocks_sb(handle, sb, block, count, - &dquot_freed_blocks); - else - ext4_mb_free_blocks(handle, inode, block, count, - metadata, &dquot_freed_blocks); + ext4_mb_free_blocks(handle, inode, block, count, + metadata, &dquot_freed_blocks); if (dquot_freed_blocks) DQUOT_FREE_BLOCK(inode, dquot_freed_blocks); return; } -/** - * ext4_test_allocatable() - * @nr: given allocation block group - * @bh: bufferhead contains the bitmap of the given block group - * - * For ext4 allocations, we must not reuse any blocks which are - * allocated in the bitmap buffer's "last committed data" copy. This - * prevents deletes from freeing up the page for reuse until we have - * committed the delete transaction. - * - * If we didn't do this, then deleting something and reallocating it as - * data would allow the old block to be overwritten before the - * transaction committed (because we force data to disk before commit). - * This would lead to corruption if we crashed between overwriting the - * data and committing the delete. - * - * @@@ We may want to make this allocation behaviour conditional on - * data-writes at some point, and disable it for metadata allocations or - * sync-data inodes. - */ -static int ext4_test_allocatable(ext4_grpblk_t nr, struct buffer_head *bh) -{ - int ret; - struct journal_head *jh = bh2jh(bh); - - if (ext4_test_bit(nr, bh->b_data)) - return 0; - - jbd_lock_bh_state(bh); - if (!jh->b_committed_data) - ret = 1; - else - ret = !ext4_test_bit(nr, jh->b_committed_data); - jbd_unlock_bh_state(bh); - return ret; -} - -/** - * bitmap_search_next_usable_block() - * @start: the starting block (group relative) of the search - * @bh: bufferhead contains the block group bitmap - * @maxblocks: the ending block (group relative) of the reservation - * - * The bitmap search --- search forward alternately through the actual - * bitmap on disk and the last-committed copy in journal, until we find a - * bit free in both bitmaps. - */ -static ext4_grpblk_t -bitmap_search_next_usable_block(ext4_grpblk_t start, struct buffer_head *bh, - ext4_grpblk_t maxblocks) +int ext4_claim_free_blocks(struct ext4_sb_info *sbi, + s64 nblocks) { - ext4_grpblk_t next; - struct journal_head *jh = bh2jh(bh); - - while (start < maxblocks) { - next = ext4_find_next_zero_bit(bh->b_data, maxblocks, start); - if (next >= maxblocks) - return -1; - if (ext4_test_allocatable(next, bh)) - return next; - jbd_lock_bh_state(bh); - if (jh->b_committed_data) - start = ext4_find_next_zero_bit(jh->b_committed_data, - maxblocks, next); - jbd_unlock_bh_state(bh); - } - return -1; -} + s64 free_blocks, dirty_blocks; + s64 root_blocks = 0; + struct percpu_counter *fbc = &sbi->s_freeblocks_counter; + struct percpu_counter *dbc = &sbi->s_dirtyblocks_counter; -/** - * find_next_usable_block() - * @start: the starting block (group relative) to find next - * allocatable block in bitmap. - * @bh: bufferhead contains the block group bitmap - * @maxblocks: the ending block (group relative) for the search - * - * Find an allocatable block in a bitmap. We honor both the bitmap and - * its last-committed copy (if that exists), and perform the "most - * appropriate allocation" algorithm of looking for a free block near - * the initial goal; then for a free byte somewhere in the bitmap; then - * for any free bit in the bitmap. - */ -static ext4_grpblk_t -find_next_usable_block(ext4_grpblk_t start, struct buffer_head *bh, - ext4_grpblk_t maxblocks) -{ - ext4_grpblk_t here, next; - char *p, *r; - - if (start > 0) { - /* - * The goal was occupied; search forward for a free - * block within the next XX blocks. - * - * end_goal is more or less random, but it has to be - * less than EXT4_BLOCKS_PER_GROUP. Aligning up to the - * next 64-bit boundary is simple.. - */ - ext4_grpblk_t end_goal = (start + 63) & ~63; - if (end_goal > maxblocks) - end_goal = maxblocks; - here = ext4_find_next_zero_bit(bh->b_data, end_goal, start); - if (here < end_goal && ext4_test_allocatable(here, bh)) - return here; - ext4_debug("Bit not found near goal\n"); - } - - here = start; - if (here < 0) - here = 0; - - p = ((char *)bh->b_data) + (here >> 3); - r = memscan(p, 0, ((maxblocks + 7) >> 3) - (here >> 3)); - next = (r - ((char *)bh->b_data)) << 3; - - if (next < maxblocks && next >= start && ext4_test_allocatable(next, bh)) - return next; - - /* - * The bitmap search --- search forward alternately through the actual - * bitmap and the last-committed copy until we find a bit free in - * both - */ - here = bitmap_search_next_usable_block(here, bh, maxblocks); - return here; -} - -/** - * claim_block() - * @block: the free block (group relative) to allocate - * @bh: the bufferhead containts the block group bitmap - * - * We think we can allocate this block in this bitmap. Try to set the bit. - * If that succeeds then check that nobody has allocated and then freed the - * block since we saw that is was not marked in b_committed_data. If it _was_ - * allocated and freed then clear the bit in the bitmap again and return - * zero (failure). - */ -static inline int -claim_block(spinlock_t *lock, ext4_grpblk_t block, struct buffer_head *bh) -{ - struct journal_head *jh = bh2jh(bh); - int ret; - - if (ext4_set_bit_atomic(lock, block, bh->b_data)) - return 0; - jbd_lock_bh_state(bh); - if (jh->b_committed_data && ext4_test_bit(block,jh->b_committed_data)) { - ext4_clear_bit_atomic(lock, block, bh->b_data); - ret = 0; - } else { - ret = 1; - } - jbd_unlock_bh_state(bh); - return ret; -} + free_blocks = percpu_counter_read_positive(fbc); + dirty_blocks = percpu_counter_read_positive(dbc); -/** - * ext4_try_to_allocate() - * @sb: superblock - * @handle: handle to this transaction - * @group: given allocation block group - * @bitmap_bh: bufferhead holds the block bitmap - * @grp_goal: given target block within the group - * @count: target number of blocks to allocate - * @my_rsv: reservation window - * - * Attempt to allocate blocks within a give range. Set the range of allocation - * first, then find the first free bit(s) from the bitmap (within the range), - * and at last, allocate the blocks by claiming the found free bit as allocated. - * - * To set the range of this allocation: - * if there is a reservation window, only try to allocate block(s) from the - * file's own reservation window; - * Otherwise, the allocation range starts from the give goal block, ends at - * the block group's last block. - * - * If we failed to allocate the desired block then we may end up crossing to a - * new bitmap. In that case we must release write access to the old one via - * ext4_journal_release_buffer(), else we'll run out of credits. - */ -static ext4_grpblk_t -ext4_try_to_allocate(struct super_block *sb, handle_t *handle, - ext4_group_t group, struct buffer_head *bitmap_bh, - ext4_grpblk_t grp_goal, unsigned long *count, - struct ext4_reserve_window *my_rsv) -{ - ext4_fsblk_t group_first_block; - ext4_grpblk_t start, end; - unsigned long num = 0; - - /* we do allocation within the reservation window if we have a window */ - if (my_rsv) { - group_first_block = ext4_group_first_block_no(sb, group); - if (my_rsv->_rsv_start >= group_first_block) - start = my_rsv->_rsv_start - group_first_block; - else - /* reservation window cross group boundary */ - start = 0; - end = my_rsv->_rsv_end - group_first_block + 1; - if (end > EXT4_BLOCKS_PER_GROUP(sb)) - /* reservation window crosses group boundary */ - end = EXT4_BLOCKS_PER_GROUP(sb); - if ((start <= grp_goal) && (grp_goal < end)) - start = grp_goal; - else - grp_goal = -1; - } else { - if (grp_goal > 0) - start = grp_goal; - else - start = 0; - end = EXT4_BLOCKS_PER_GROUP(sb); - } - - BUG_ON(start > EXT4_BLOCKS_PER_GROUP(sb)); - -repeat: - if (grp_goal < 0 || !ext4_test_allocatable(grp_goal, bitmap_bh)) { - grp_goal = find_next_usable_block(start, bitmap_bh, end); - if (grp_goal < 0) - goto fail_access; - if (!my_rsv) { - int i; - - for (i = 0; i < 7 && grp_goal > start && - ext4_test_allocatable(grp_goal - 1, - bitmap_bh); - i++, grp_goal--) - ; - } - } - start = grp_goal; - - if (!claim_block(sb_bgl_lock(EXT4_SB(sb), group), - grp_goal, bitmap_bh)) { - /* - * The block was allocated by another thread, or it was - * allocated and then freed by another thread - */ - start++; - grp_goal++; - if (start >= end) - goto fail_access; - goto repeat; - } - num++; - grp_goal++; - while (num < *count && grp_goal < end - && ext4_test_allocatable(grp_goal, bitmap_bh) - && claim_block(sb_bgl_lock(EXT4_SB(sb), group), - grp_goal, bitmap_bh)) { - num++; - grp_goal++; - } - *count = num; - return grp_goal - num; -fail_access: - *count = num; - return -1; -} - -/** - * find_next_reservable_window(): - * find a reservable space within the given range. - * It does not allocate the reservation window for now: - * alloc_new_reservation() will do the work later. - * - * @search_head: the head of the searching list; - * This is not necessarily the list head of the whole filesystem - * - * We have both head and start_block to assist the search - * for the reservable space. The list starts from head, - * but we will shift to the place where start_block is, - * then start from there, when looking for a reservable space. - * - * @size: the target new reservation window size - * - * @group_first_block: the first block we consider to start - * the real search from - * - * @last_block: - * the maximum block number that our goal reservable space - * could start from. This is normally the last block in this - * group. The search will end when we found the start of next - * possible reservable space is out of this boundary. - * This could handle the cross boundary reservation window - * request. - * - * basically we search from the given range, rather than the whole - * reservation double linked list, (start_block, last_block) - * to find a free region that is of my size and has not - * been reserved. - * - */ -static int find_next_reservable_window( - struct ext4_reserve_window_node *search_head, - struct ext4_reserve_window_node *my_rsv, - struct super_block * sb, - ext4_fsblk_t start_block, - ext4_fsblk_t last_block) -{ - struct rb_node *next; - struct ext4_reserve_window_node *rsv, *prev; - ext4_fsblk_t cur; - int size = my_rsv->rsv_goal_size; - - /* TODO: make the start of the reservation window byte-aligned */ - /* cur = *start_block & ~7;*/ - cur = start_block; - rsv = search_head; - if (!rsv) - return -1; - - while (1) { - if (cur <= rsv->rsv_end) - cur = rsv->rsv_end + 1; - - /* TODO? - * in the case we could not find a reservable space - * that is what is expected, during the re-search, we could - * remember what's the largest reservable space we could have - * and return that one. - * - * For now it will fail if we could not find the reservable - * space with expected-size (or more)... - */ - if (cur > last_block) - return -1; /* fail */ - - prev = rsv; - next = rb_next(&rsv->rsv_node); - rsv = rb_entry(next,struct ext4_reserve_window_node,rsv_node); + if (!capable(CAP_SYS_RESOURCE) && + sbi->s_resuid != current->fsuid && + (sbi->s_resgid == 0 || !in_group_p(sbi->s_resgid))) + root_blocks = ext4_r_blocks_count(sbi->s_es); - /* - * Reached the last reservation, we can just append to the - * previous one. - */ - if (!next) - break; - - if (cur + size <= rsv->rsv_start) { - /* - * Found a reserveable space big enough. We could - * have a reservation across the group boundary here - */ - break; + if (free_blocks - (nblocks + root_blocks + dirty_blocks) < + EXT4_FREEBLOCKS_WATERMARK) { + free_blocks = percpu_counter_sum(fbc); + dirty_blocks = percpu_counter_sum(dbc); + if (dirty_blocks < 0) { + printk(KERN_CRIT "Dirty block accounting " + "went wrong %lld\n", + dirty_blocks); } } - /* - * we come here either : - * when we reach the end of the whole list, - * and there is empty reservable space after last entry in the list. - * append it to the end of the list. - * - * or we found one reservable space in the middle of the list, - * return the reservation window that we could append to. - * succeed. + /* Check whether we have space after + * accounting for current dirty blocks */ + if (free_blocks < ((root_blocks + nblocks) + dirty_blocks)) + /* we don't have free space */ + return -ENOSPC; - if ((prev != my_rsv) && (!rsv_is_empty(&my_rsv->rsv_window))) - rsv_window_remove(sb, my_rsv); - - /* - * Let's book the whole avaliable window for now. We will check the - * disk bitmap later and then, if there are free blocks then we adjust - * the window size if it's larger than requested. - * Otherwise, we will remove this node from the tree next time - * call find_next_reservable_window. - */ - my_rsv->rsv_start = cur; - my_rsv->rsv_end = cur + size - 1; - my_rsv->rsv_alloc_hit = 0; - - if (prev != my_rsv) - ext4_rsv_window_add(sb, my_rsv); - + /* Add the blocks to nblocks */ + percpu_counter_add(dbc, nblocks); return 0; } /** - * alloc_new_reservation()--allocate a new reservation window - * - * To make a new reservation, we search part of the filesystem - * reservation list (the list that inside the group). We try to - * allocate a new reservation window near the allocation goal, - * or the beginning of the group, if there is no goal. - * - * We first find a reservable space after the goal, then from - * there, we check the bitmap for the first free block after - * it. If there is no free block until the end of group, then the - * whole group is full, we failed. Otherwise, check if the free - * block is inside the expected reservable space, if so, we - * succeed. - * If the first free block is outside the reservable space, then - * start from the first free block, we search for next available - * space, and go on. - * - * on succeed, a new reservation will be found and inserted into the list - * It contains at least one free block, and it does not overlap with other - * reservation windows. - * - * failed: we failed to find a reservation window in this group - * - * @rsv: the reservation - * - * @grp_goal: The goal (group-relative). It is where the search for a - * free reservable space should start from. - * if we have a grp_goal(grp_goal >0 ), then start from there, - * no grp_goal(grp_goal = -1), we start from the first block - * of the group. - * - * @sb: the super block - * @group: the group we are trying to allocate in - * @bitmap_bh: the block group block bitmap - * - */ -static int alloc_new_reservation(struct ext4_reserve_window_node *my_rsv, - ext4_grpblk_t grp_goal, struct super_block *sb, - ext4_group_t group, struct buffer_head *bitmap_bh) -{ - struct ext4_reserve_window_node *search_head; - ext4_fsblk_t group_first_block, group_end_block, start_block; - ext4_grpblk_t first_free_block; - struct rb_root *fs_rsv_root = &EXT4_SB(sb)->s_rsv_window_root; - unsigned long size; - int ret; - spinlock_t *rsv_lock = &EXT4_SB(sb)->s_rsv_window_lock; - - group_first_block = ext4_group_first_block_no(sb, group); - group_end_block = group_first_block + (EXT4_BLOCKS_PER_GROUP(sb) - 1); - - if (grp_goal < 0) - start_block = group_first_block; - else - start_block = grp_goal + group_first_block; - - size = my_rsv->rsv_goal_size; - - if (!rsv_is_empty(&my_rsv->rsv_window)) { - /* - * if the old reservation is cross group boundary - * and if the goal is inside the old reservation window, - * we will come here when we just failed to allocate from - * the first part of the window. We still have another part - * that belongs to the next group. In this case, there is no - * point to discard our window and try to allocate a new one - * in this group(which will fail). we should - * keep the reservation window, just simply move on. - * - * Maybe we could shift the start block of the reservation - * window to the first block of next group. - */ - - if ((my_rsv->rsv_start <= group_end_block) && - (my_rsv->rsv_end > group_end_block) && - (start_block >= my_rsv->rsv_start)) - return -1; - - if ((my_rsv->rsv_alloc_hit > - (my_rsv->rsv_end - my_rsv->rsv_start + 1) / 2)) { - /* - * if the previously allocation hit ratio is - * greater than 1/2, then we double the size of - * the reservation window the next time, - * otherwise we keep the same size window - */ - size = size * 2; - if (size > EXT4_MAX_RESERVE_BLOCKS) - size = EXT4_MAX_RESERVE_BLOCKS; - my_rsv->rsv_goal_size= size; - } - } - - spin_lock(rsv_lock); - /* - * shift the search start to the window near the goal block - */ - search_head = search_reserve_window(fs_rsv_root, start_block); - - /* - * find_next_reservable_window() simply finds a reservable window - * inside the given range(start_block, group_end_block). - * - * To make sure the reservation window has a free bit inside it, we - * need to check the bitmap after we found a reservable window. - */ -retry: - ret = find_next_reservable_window(search_head, my_rsv, sb, - start_block, group_end_block); - - if (ret == -1) { - if (!rsv_is_empty(&my_rsv->rsv_window)) - rsv_window_remove(sb, my_rsv); - spin_unlock(rsv_lock); - return -1; - } - - /* - * On success, find_next_reservable_window() returns the - * reservation window where there is a reservable space after it. - * Before we reserve this reservable space, we need - * to make sure there is at least a free block inside this region. - * - * searching the first free bit on the block bitmap and copy of - * last committed bitmap alternatively, until we found a allocatable - * block. Search start from the start block of the reservable space - * we just found. - */ - spin_unlock(rsv_lock); - first_free_block = bitmap_search_next_usable_block( - my_rsv->rsv_start - group_first_block, - bitmap_bh, group_end_block - group_first_block + 1); - - if (first_free_block < 0) { - /* - * no free block left on the bitmap, no point - * to reserve the space. return failed. - */ - spin_lock(rsv_lock); - if (!rsv_is_empty(&my_rsv->rsv_window)) - rsv_window_remove(sb, my_rsv); - spin_unlock(rsv_lock); - return -1; /* failed */ - } - - start_block = first_free_block + group_first_block; - /* - * check if the first free block is within the - * free space we just reserved - */ - if (start_block >= my_rsv->rsv_start && start_block <= my_rsv->rsv_end) - return 0; /* success */ - /* - * if the first free bit we found is out of the reservable space - * continue search for next reservable space, - * start from where the free block is, - * we also shift the list head to where we stopped last time - */ - search_head = my_rsv; - spin_lock(rsv_lock); - goto retry; -} - -/** - * try_to_extend_reservation() - * @my_rsv: given reservation window - * @sb: super block - * @size: the delta to extend - * - * Attempt to expand the reservation window large enough to have - * required number of free blocks - * - * Since ext4_try_to_allocate() will always allocate blocks within - * the reservation window range, if the window size is too small, - * multiple blocks allocation has to stop at the end of the reservation - * window. To make this more efficient, given the total number of - * blocks needed and the current size of the window, we try to - * expand the reservation window size if necessary on a best-effort - * basis before ext4_new_blocks() tries to allocate blocks, - */ -static void try_to_extend_reservation(struct ext4_reserve_window_node *my_rsv, - struct super_block *sb, int size) -{ - struct ext4_reserve_window_node *next_rsv; - struct rb_node *next; - spinlock_t *rsv_lock = &EXT4_SB(sb)->s_rsv_window_lock; - - if (!spin_trylock(rsv_lock)) - return; - - next = rb_next(&my_rsv->rsv_node); - - if (!next) - my_rsv->rsv_end += size; - else { - next_rsv = rb_entry(next, struct ext4_reserve_window_node, rsv_node); - - if ((next_rsv->rsv_start - my_rsv->rsv_end - 1) >= size) - my_rsv->rsv_end += size; - else - my_rsv->rsv_end = next_rsv->rsv_start - 1; - } - spin_unlock(rsv_lock); -} - -/** - * ext4_try_to_allocate_with_rsv() - * @sb: superblock - * @handle: handle to this transaction - * @group: given allocation block group - * @bitmap_bh: bufferhead holds the block bitmap - * @grp_goal: given target block within the group - * @count: target number of blocks to allocate - * @my_rsv: reservation window - * @errp: pointer to store the error code - * - * This is the main function used to allocate a new block and its reservation - * window. - * - * Each time when a new block allocation is need, first try to allocate from - * its own reservation. If it does not have a reservation window, instead of - * looking for a free bit on bitmap first, then look up the reservation list to - * see if it is inside somebody else's reservation window, we try to allocate a - * reservation window for it starting from the goal first. Then do the block - * allocation within the reservation window. - * - * This will avoid keeping on searching the reservation list again and - * again when somebody is looking for a free block (without - * reservation), and there are lots of free blocks, but they are all - * being reserved. - * - * We use a red-black tree for the per-filesystem reservation list. - * - */ -static ext4_grpblk_t -ext4_try_to_allocate_with_rsv(struct super_block *sb, handle_t *handle, - ext4_group_t group, struct buffer_head *bitmap_bh, - ext4_grpblk_t grp_goal, - struct ext4_reserve_window_node * my_rsv, - unsigned long *count, int *errp) -{ - ext4_fsblk_t group_first_block, group_last_block; - ext4_grpblk_t ret = 0; - int fatal; - unsigned long num = *count; - - *errp = 0; - - /* - * Make sure we use undo access for the bitmap, because it is critical - * that we do the frozen_data COW on bitmap buffers in all cases even - * if the buffer is in BJ_Forget state in the committing transaction. - */ - BUFFER_TRACE(bitmap_bh, "get undo access for new block"); - fatal = ext4_journal_get_undo_access(handle, bitmap_bh); - if (fatal) { - *errp = fatal; - return -1; - } - - /* - * we don't deal with reservation when - * filesystem is mounted without reservation - * or the file is not a regular file - * or last attempt to allocate a block with reservation turned on failed - */ - if (my_rsv == NULL ) { - ret = ext4_try_to_allocate(sb, handle, group, bitmap_bh, - grp_goal, count, NULL); - goto out; - } - /* - * grp_goal is a group relative block number (if there is a goal) - * 0 <= grp_goal < EXT4_BLOCKS_PER_GROUP(sb) - * first block is a filesystem wide block number - * first block is the block number of the first block in this group - */ - group_first_block = ext4_group_first_block_no(sb, group); - group_last_block = group_first_block + (EXT4_BLOCKS_PER_GROUP(sb) - 1); - - /* - * Basically we will allocate a new block from inode's reservation - * window. - * - * We need to allocate a new reservation window, if: - * a) inode does not have a reservation window; or - * b) last attempt to allocate a block from existing reservation - * failed; or - * c) we come here with a goal and with a reservation window - * - * We do not need to allocate a new reservation window if we come here - * at the beginning with a goal and the goal is inside the window, or - * we don't have a goal but already have a reservation window. - * then we could go to allocate from the reservation window directly. - */ - while (1) { - if (rsv_is_empty(&my_rsv->rsv_window) || (ret < 0) || - !goal_in_my_reservation(&my_rsv->rsv_window, - grp_goal, group, sb)) { - if (my_rsv->rsv_goal_size < *count) - my_rsv->rsv_goal_size = *count; - ret = alloc_new_reservation(my_rsv, grp_goal, sb, - group, bitmap_bh); - if (ret < 0) - break; /* failed */ - - if (!goal_in_my_reservation(&my_rsv->rsv_window, - grp_goal, group, sb)) - grp_goal = -1; - } else if (grp_goal >= 0) { - int curr = my_rsv->rsv_end - - (grp_goal + group_first_block) + 1; - - if (curr < *count) - try_to_extend_reservation(my_rsv, sb, - *count - curr); - } - - if ((my_rsv->rsv_start > group_last_block) || - (my_rsv->rsv_end < group_first_block)) { - rsv_window_dump(&EXT4_SB(sb)->s_rsv_window_root, 1); - BUG(); - } - ret = ext4_try_to_allocate(sb, handle, group, bitmap_bh, - grp_goal, &num, &my_rsv->rsv_window); - if (ret >= 0) { - my_rsv->rsv_alloc_hit += num; - *count = num; - break; /* succeed */ - } - num = *count; - } -out: - if (ret >= 0) { - BUFFER_TRACE(bitmap_bh, "journal_dirty_metadata for " - "bitmap block"); - fatal = ext4_journal_dirty_metadata(handle, bitmap_bh); - if (fatal) { - *errp = fatal; - return -1; - } - return ret; - } - - BUFFER_TRACE(bitmap_bh, "journal_release_buffer"); - ext4_journal_release_buffer(handle, bitmap_bh); - return ret; -} - -/** * ext4_has_free_blocks() * @sbi: in-core super block structure. * @nblocks: number of neeed blocks @@ -1610,29 +629,34 @@ out: * On success, return nblocks */ ext4_fsblk_t ext4_has_free_blocks(struct ext4_sb_info *sbi, - ext4_fsblk_t nblocks) + s64 nblocks) { - ext4_fsblk_t free_blocks; - ext4_fsblk_t root_blocks = 0; + s64 free_blocks, dirty_blocks; + s64 root_blocks = 0; + struct percpu_counter *fbc = &sbi->s_freeblocks_counter; + struct percpu_counter *dbc = &sbi->s_dirtyblocks_counter; - free_blocks = percpu_counter_read_positive(&sbi->s_freeblocks_counter); + free_blocks = percpu_counter_read_positive(fbc); + dirty_blocks = percpu_counter_read_positive(dbc); if (!capable(CAP_SYS_RESOURCE) && sbi->s_resuid != current->fsuid && (sbi->s_resgid == 0 || !in_group_p(sbi->s_resgid))) root_blocks = ext4_r_blocks_count(sbi->s_es); -#ifdef CONFIG_SMP - if (free_blocks - root_blocks < FBC_BATCH) - free_blocks = - percpu_counter_sum_and_set(&sbi->s_freeblocks_counter); -#endif - if (free_blocks <= root_blocks) + + if (free_blocks - (nblocks + root_blocks + dirty_blocks) < + EXT4_FREEBLOCKS_WATERMARK) { + free_blocks = percpu_counter_sum(fbc); + dirty_blocks = percpu_counter_sum(dbc); + } + if (free_blocks <= (root_blocks + dirty_blocks)) /* we don't have free space */ return 0; - if (free_blocks - root_blocks < nblocks) - return free_blocks - root_blocks; + + if (free_blocks - (root_blocks + dirty_blocks) < nblocks) + return free_blocks - (root_blocks + dirty_blocks); return nblocks; - } +} /** @@ -1657,303 +681,6 @@ int ext4_should_retry_alloc(struct super_block *sb, int *retries) return jbd2_journal_force_commit_nested(EXT4_SB(sb)->s_journal); } -/** - * ext4_old_new_blocks() -- core block bitmap based block allocation function - * - * @handle: handle to this transaction - * @inode: file inode - * @goal: given target block(filesystem wide) - * @count: target number of blocks to allocate - * @errp: error code - * - * ext4_old_new_blocks uses a goal block to assist allocation and look up - * the block bitmap directly to do block allocation. It tries to - * allocate block(s) from the block group contains the goal block first. If - * that fails, it will try to allocate block(s) from other block groups - * without any specific goal block. - * - * This function is called when -o nomballoc mount option is enabled - * - */ -ext4_fsblk_t ext4_old_new_blocks(handle_t *handle, struct inode *inode, - ext4_fsblk_t goal, unsigned long *count, int *errp) -{ - struct buffer_head *bitmap_bh = NULL; - struct buffer_head *gdp_bh; - ext4_group_t group_no; - ext4_group_t goal_group; - ext4_grpblk_t grp_target_blk; /* blockgroup relative goal block */ - ext4_grpblk_t grp_alloc_blk; /* blockgroup-relative allocated block*/ - ext4_fsblk_t ret_block; /* filesyetem-wide allocated block */ - ext4_group_t bgi; /* blockgroup iteration index */ - int fatal = 0, err; - int performed_allocation = 0; - ext4_grpblk_t free_blocks; /* number of free blocks in a group */ - struct super_block *sb; - struct ext4_group_desc *gdp; - struct ext4_super_block *es; - struct ext4_sb_info *sbi; - struct ext4_reserve_window_node *my_rsv = NULL; - struct ext4_block_alloc_info *block_i; - unsigned short windowsz = 0; - ext4_group_t ngroups; - unsigned long num = *count; - - sb = inode->i_sb; - if (!sb) { - *errp = -ENODEV; - printk("ext4_new_block: nonexistent device"); - return 0; - } - - sbi = EXT4_SB(sb); - if (!EXT4_I(inode)->i_delalloc_reserved_flag) { - /* - * With delalloc we already reserved the blocks - */ - *count = ext4_has_free_blocks(sbi, *count); - } - if (*count == 0) { - *errp = -ENOSPC; - return 0; /*return with ENOSPC error */ - } - num = *count; - - /* - * Check quota for allocation of this block. - */ - if (DQUOT_ALLOC_BLOCK(inode, num)) { - *errp = -EDQUOT; - return 0; - } - - sbi = EXT4_SB(sb); - es = EXT4_SB(sb)->s_es; - ext4_debug("goal=%llu.\n", goal); - /* - * Allocate a block from reservation only when - * filesystem is mounted with reservation(default,-o reservation), and - * it's a regular file, and - * the desired window size is greater than 0 (One could use ioctl - * command EXT4_IOC_SETRSVSZ to set the window size to 0 to turn off - * reservation on that particular file) - */ - block_i = EXT4_I(inode)->i_block_alloc_info; - if (block_i && ((windowsz = block_i->rsv_window_node.rsv_goal_size) > 0)) - my_rsv = &block_i->rsv_window_node; - - /* - * First, test whether the goal block is free. - */ - if (goal < le32_to_cpu(es->s_first_data_block) || - goal >= ext4_blocks_count(es)) - goal = le32_to_cpu(es->s_first_data_block); - ext4_get_group_no_and_offset(sb, goal, &group_no, &grp_target_blk); - goal_group = group_no; -retry_alloc: - gdp = ext4_get_group_desc(sb, group_no, &gdp_bh); - if (!gdp) - goto io_error; - - free_blocks = le16_to_cpu(gdp->bg_free_blocks_count); - /* - * if there is not enough free blocks to make a new resevation - * turn off reservation for this allocation - */ - if (my_rsv && (free_blocks < windowsz) - && (rsv_is_empty(&my_rsv->rsv_window))) - my_rsv = NULL; - - if (free_blocks > 0) { - bitmap_bh = ext4_read_block_bitmap(sb, group_no); - if (!bitmap_bh) - goto io_error; - grp_alloc_blk = ext4_try_to_allocate_with_rsv(sb, handle, - group_no, bitmap_bh, grp_target_blk, - my_rsv, &num, &fatal); - if (fatal) - goto out; - if (grp_alloc_blk >= 0) - goto allocated; - } - - ngroups = EXT4_SB(sb)->s_groups_count; - smp_rmb(); - - /* - * Now search the rest of the groups. We assume that - * group_no and gdp correctly point to the last group visited. - */ - for (bgi = 0; bgi < ngroups; bgi++) { - group_no++; - if (group_no >= ngroups) - group_no = 0; - gdp = ext4_get_group_desc(sb, group_no, &gdp_bh); - if (!gdp) - goto io_error; - free_blocks = le16_to_cpu(gdp->bg_free_blocks_count); - /* - * skip this group if the number of - * free blocks is less than half of the reservation - * window size. - */ - if (free_blocks <= (windowsz/2)) - continue; - - brelse(bitmap_bh); - bitmap_bh = ext4_read_block_bitmap(sb, group_no); - if (!bitmap_bh) - goto io_error; - /* - * try to allocate block(s) from this group, without a goal(-1). - */ - grp_alloc_blk = ext4_try_to_allocate_with_rsv(sb, handle, - group_no, bitmap_bh, -1, my_rsv, - &num, &fatal); - if (fatal) - goto out; - if (grp_alloc_blk >= 0) - goto allocated; - } - /* - * We may end up a bogus ealier ENOSPC error due to - * filesystem is "full" of reservations, but - * there maybe indeed free blocks avaliable on disk - * In this case, we just forget about the reservations - * just do block allocation as without reservations. - */ - if (my_rsv) { - my_rsv = NULL; - windowsz = 0; - group_no = goal_group; - goto retry_alloc; - } - /* No space left on the device */ - *errp = -ENOSPC; - goto out; - -allocated: - - ext4_debug("using block group %lu(%d)\n", - group_no, gdp->bg_free_blocks_count); - - BUFFER_TRACE(gdp_bh, "get_write_access"); - fatal = ext4_journal_get_write_access(handle, gdp_bh); - if (fatal) - goto out; - - ret_block = grp_alloc_blk + ext4_group_first_block_no(sb, group_no); - - if (in_range(ext4_block_bitmap(sb, gdp), ret_block, num) || - in_range(ext4_inode_bitmap(sb, gdp), ret_block, num) || - in_range(ret_block, ext4_inode_table(sb, gdp), - EXT4_SB(sb)->s_itb_per_group) || - in_range(ret_block + num - 1, ext4_inode_table(sb, gdp), - EXT4_SB(sb)->s_itb_per_group)) { - ext4_error(sb, "ext4_new_block", - "Allocating block in system zone - " - "blocks from %llu, length %lu", - ret_block, num); - /* - * claim_block marked the blocks we allocated - * as in use. So we may want to selectively - * mark some of the blocks as free - */ - goto retry_alloc; - } - - performed_allocation = 1; - -#ifdef CONFIG_JBD2_DEBUG - { - struct buffer_head *debug_bh; - - /* Record bitmap buffer state in the newly allocated block */ - debug_bh = sb_find_get_block(sb, ret_block); - if (debug_bh) { - BUFFER_TRACE(debug_bh, "state when allocated"); - BUFFER_TRACE2(debug_bh, bitmap_bh, "bitmap state"); - brelse(debug_bh); - } - } - jbd_lock_bh_state(bitmap_bh); - spin_lock(sb_bgl_lock(sbi, group_no)); - if (buffer_jbd(bitmap_bh) && bh2jh(bitmap_bh)->b_committed_data) { - int i; - - for (i = 0; i < num; i++) { - if (ext4_test_bit(grp_alloc_blk+i, - bh2jh(bitmap_bh)->b_committed_data)) { - printk("%s: block was unexpectedly set in " - "b_committed_data\n", __func__); - } - } - } - ext4_debug("found bit %d\n", grp_alloc_blk); - spin_unlock(sb_bgl_lock(sbi, group_no)); - jbd_unlock_bh_state(bitmap_bh); -#endif - - if (ret_block + num - 1 >= ext4_blocks_count(es)) { - ext4_error(sb, "ext4_new_block", - "block(%llu) >= blocks count(%llu) - " - "block_group = %lu, es == %p ", ret_block, - ext4_blocks_count(es), group_no, es); - goto out; - } - - /* - * It is up to the caller to add the new buffer to a journal - * list of some description. We don't know in advance whether - * the caller wants to use it as metadata or data. - */ - spin_lock(sb_bgl_lock(sbi, group_no)); - if (gdp->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) - gdp->bg_flags &= cpu_to_le16(~EXT4_BG_BLOCK_UNINIT); - le16_add_cpu(&gdp->bg_free_blocks_count, -num); - gdp->bg_checksum = ext4_group_desc_csum(sbi, group_no, gdp); - spin_unlock(sb_bgl_lock(sbi, group_no)); - if (!EXT4_I(inode)->i_delalloc_reserved_flag) - percpu_counter_sub(&sbi->s_freeblocks_counter, num); - - if (sbi->s_log_groups_per_flex) { - ext4_group_t flex_group = ext4_flex_group(sbi, group_no); - spin_lock(sb_bgl_lock(sbi, flex_group)); - sbi->s_flex_groups[flex_group].free_blocks -= num; - spin_unlock(sb_bgl_lock(sbi, flex_group)); - } - - BUFFER_TRACE(gdp_bh, "journal_dirty_metadata for group descriptor"); - err = ext4_journal_dirty_metadata(handle, gdp_bh); - if (!fatal) - fatal = err; - - sb->s_dirt = 1; - if (fatal) - goto out; - - *errp = 0; - brelse(bitmap_bh); - DQUOT_FREE_BLOCK(inode, *count-num); - *count = num; - return ret_block; - -io_error: - *errp = -EIO; -out: - if (fatal) { - *errp = fatal; - ext4_std_error(sb, fatal); - } - /* - * Undo the block allocation - */ - if (!performed_allocation) - DQUOT_FREE_BLOCK(inode, *count); - brelse(bitmap_bh); - return 0; -} - #define EXT4_META_BLOCK 0x1 static ext4_fsblk_t do_blk_alloc(handle_t *handle, struct inode *inode, @@ -1963,10 +690,6 @@ static ext4_fsblk_t do_blk_alloc(handle_t *handle, struct inode *inode, struct ext4_allocation_request ar; ext4_fsblk_t ret; - if (!test_opt(inode->i_sb, MBALLOC)) { - return ext4_old_new_blocks(handle, inode, goal, count, errp); - } - memset(&ar, 0, sizeof(ar)); /* Fill with neighbour allocated blocks */ @@ -2008,7 +731,7 @@ ext4_fsblk_t ext4_new_meta_blocks(handle_t *handle, struct inode *inode, /* * Account for the allocated meta blocks */ - if (!(*errp)) { + if (!(*errp) && EXT4_I(inode)->i_delalloc_reserved_flag) { spin_lock(&EXT4_I(inode)->i_block_reservation_lock); EXT4_I(inode)->i_allocated_meta_blocks += *count; spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); @@ -2093,10 +816,9 @@ ext4_fsblk_t ext4_count_free_blocks(struct super_block *sb) bitmap_count += x; } brelse(bitmap_bh); - printk("ext4_count_free_blocks: stored = %llu" - ", computed = %llu, %llu\n", - ext4_free_blocks_count(es), - desc_count, bitmap_count); + printk(KERN_DEBUG "ext4_count_free_blocks: stored = %llu" + ", computed = %llu, %llu\n", ext4_free_blocks_count(es), + desc_count, bitmap_count); return bitmap_count; #else desc_count = 0; @@ -2183,8 +905,9 @@ unsigned long ext4_bg_num_gdb(struct super_block *sb, ext4_group_t group) if (!EXT4_HAS_INCOMPAT_FEATURE(sb,EXT4_FEATURE_INCOMPAT_META_BG) || metagroup < first_meta_bg) - return ext4_bg_num_gdb_nometa(sb,group); + return ext4_bg_num_gdb_nometa(sb, group); return ext4_bg_num_gdb_meta(sb,group); } + diff --git a/fs/ext4/bitmap.c b/fs/ext4/bitmap.c index d37ea675045..0a7a6663c19 100644 --- a/fs/ext4/bitmap.c +++ b/fs/ext4/bitmap.c @@ -15,17 +15,17 @@ static const int nibblemap[] = {4, 3, 3, 2, 3, 2, 2, 1, 3, 2, 2, 1, 2, 1, 1, 0}; -unsigned long ext4_count_free (struct buffer_head * map, unsigned int numchars) +unsigned long ext4_count_free(struct buffer_head *map, unsigned int numchars) { unsigned int i; unsigned long sum = 0; if (!map) - return (0); + return 0; for (i = 0; i < numchars; i++) sum += nibblemap[map->b_data[i] & 0xf] + nibblemap[(map->b_data[i] >> 4) & 0xf]; - return (sum); + return sum; } #endif /* EXT4FS_DEBUG */ diff --git a/fs/ext4/dir.c b/fs/ext4/dir.c index ec8e33b4521..3ca6a2b7632 100644 --- a/fs/ext4/dir.c +++ b/fs/ext4/dir.c @@ -33,10 +33,10 @@ static unsigned char ext4_filetype_table[] = { }; static int ext4_readdir(struct file *, void *, filldir_t); -static int ext4_dx_readdir(struct file * filp, - void * dirent, filldir_t filldir); -static int ext4_release_dir (struct inode * inode, - struct file * filp); +static int ext4_dx_readdir(struct file *filp, + void *dirent, filldir_t filldir); +static int ext4_release_dir(struct inode *inode, + struct file *filp); const struct file_operations ext4_dir_operations = { .llseek = generic_file_llseek, @@ -61,12 +61,12 @@ static unsigned char get_dtype(struct super_block *sb, int filetype) } -int ext4_check_dir_entry (const char * function, struct inode * dir, - struct ext4_dir_entry_2 * de, - struct buffer_head * bh, - unsigned long offset) +int ext4_check_dir_entry(const char *function, struct inode *dir, + struct ext4_dir_entry_2 *de, + struct buffer_head *bh, + unsigned long offset) { - const char * error_msg = NULL; + const char *error_msg = NULL; const int rlen = ext4_rec_len_from_disk(de->rec_len); if (rlen < EXT4_DIR_REC_LEN(1)) @@ -82,7 +82,7 @@ int ext4_check_dir_entry (const char * function, struct inode * dir, error_msg = "inode out of bounds"; if (error_msg != NULL) - ext4_error (dir->i_sb, function, + ext4_error(dir->i_sb, function, "bad entry in directory #%lu: %s - " "offset=%lu, inode=%lu, rec_len=%d, name_len=%d", dir->i_ino, error_msg, offset, @@ -91,8 +91,8 @@ int ext4_check_dir_entry (const char * function, struct inode * dir, return error_msg == NULL ? 1 : 0; } -static int ext4_readdir(struct file * filp, - void * dirent, filldir_t filldir) +static int ext4_readdir(struct file *filp, + void *dirent, filldir_t filldir) { int error = 0; unsigned long offset; @@ -102,6 +102,7 @@ static int ext4_readdir(struct file * filp, int err; struct inode *inode = filp->f_path.dentry->d_inode; int ret = 0; + int dir_has_error = 0; sb = inode->i_sb; @@ -148,9 +149,13 @@ static int ext4_readdir(struct file * filp, * of recovering data when there's a bad sector */ if (!bh) { - ext4_error (sb, "ext4_readdir", - "directory #%lu contains a hole at offset %lu", - inode->i_ino, (unsigned long)filp->f_pos); + if (!dir_has_error) { + ext4_error(sb, __func__, "directory #%lu " + "contains a hole at offset %Lu", + inode->i_ino, + (unsigned long long) filp->f_pos); + dir_has_error = 1; + } /* corrupt size? Maybe no more blocks to read */ if (filp->f_pos > inode->i_blocks << 9) break; @@ -187,14 +192,14 @@ revalidate: while (!error && filp->f_pos < inode->i_size && offset < sb->s_blocksize) { de = (struct ext4_dir_entry_2 *) (bh->b_data + offset); - if (!ext4_check_dir_entry ("ext4_readdir", inode, de, - bh, offset)) { + if (!ext4_check_dir_entry("ext4_readdir", inode, de, + bh, offset)) { /* * On error, skip the f_pos to the next block */ filp->f_pos = (filp->f_pos | (sb->s_blocksize - 1)) + 1; - brelse (bh); + brelse(bh); ret = stored; goto out; } @@ -218,12 +223,12 @@ revalidate: break; if (version != filp->f_version) goto revalidate; - stored ++; + stored++; } filp->f_pos += ext4_rec_len_from_disk(de->rec_len); } offset = 0; - brelse (bh); + brelse(bh); } out: return ret; @@ -290,9 +295,9 @@ static void free_rb_tree_fname(struct rb_root *root) parent = rb_parent(n); fname = rb_entry(n, struct fname, rb_hash); while (fname) { - struct fname * old = fname; + struct fname *old = fname; fname = fname->next; - kfree (old); + kfree(old); } if (!parent) root->rb_node = NULL; @@ -331,7 +336,7 @@ int ext4_htree_store_dirent(struct file *dir_file, __u32 hash, struct ext4_dir_entry_2 *dirent) { struct rb_node **p, *parent = NULL; - struct fname * fname, *new_fn; + struct fname *fname, *new_fn; struct dir_private_info *info; int len; @@ -388,19 +393,20 @@ int ext4_htree_store_dirent(struct file *dir_file, __u32 hash, * for all entres on the fname linked list. (Normally there is only * one entry on the linked list, unless there are 62 bit hash collisions.) */ -static int call_filldir(struct file * filp, void * dirent, +static int call_filldir(struct file *filp, void *dirent, filldir_t filldir, struct fname *fname) { struct dir_private_info *info = filp->private_data; loff_t curr_pos; struct inode *inode = filp->f_path.dentry->d_inode; - struct super_block * sb; + struct super_block *sb; int error; sb = inode->i_sb; if (!fname) { - printk("call_filldir: called with null fname?!?\n"); + printk(KERN_ERR "ext4: call_filldir: called with " + "null fname?!?\n"); return 0; } curr_pos = hash2pos(fname->hash, fname->minor_hash); @@ -419,8 +425,8 @@ static int call_filldir(struct file * filp, void * dirent, return 0; } -static int ext4_dx_readdir(struct file * filp, - void * dirent, filldir_t filldir) +static int ext4_dx_readdir(struct file *filp, + void *dirent, filldir_t filldir) { struct dir_private_info *info = filp->private_data; struct inode *inode = filp->f_path.dentry->d_inode; @@ -511,7 +517,7 @@ finished: return 0; } -static int ext4_release_dir (struct inode * inode, struct file * filp) +static int ext4_release_dir(struct inode *inode, struct file *filp) { if (filp->private_data) ext4_htree_free_dir_info(filp->private_data); diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index 295003241d3..6690a41cdd9 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -44,9 +44,9 @@ #ifdef EXT4FS_DEBUG #define ext4_debug(f, a...) \ do { \ - printk (KERN_DEBUG "EXT4-fs DEBUG (%s, %d): %s:", \ + printk(KERN_DEBUG "EXT4-fs DEBUG (%s, %d): %s:", \ __FILE__, __LINE__, __func__); \ - printk (KERN_DEBUG f, ## a); \ + printk(KERN_DEBUG f, ## a); \ } while (0) #else #define ext4_debug(f, a...) do {} while (0) @@ -128,7 +128,7 @@ struct ext4_allocation_request { #else # define EXT4_BLOCK_SIZE(s) (EXT4_MIN_BLOCK_SIZE << (s)->s_log_block_size) #endif -#define EXT4_ADDR_PER_BLOCK(s) (EXT4_BLOCK_SIZE(s) / sizeof (__u32)) +#define EXT4_ADDR_PER_BLOCK(s) (EXT4_BLOCK_SIZE(s) / sizeof(__u32)) #ifdef __KERNEL__ # define EXT4_BLOCK_SIZE_BITS(s) ((s)->s_blocksize_bits) #else @@ -245,7 +245,7 @@ struct flex_groups { #define EXT4_RESERVED_FL 0x80000000 /* reserved for ext4 lib */ #define EXT4_FL_USER_VISIBLE 0x000BDFFF /* User visible flags */ -#define EXT4_FL_USER_MODIFIABLE 0x000380FF /* User modifiable flags */ +#define EXT4_FL_USER_MODIFIABLE 0x000B80FF /* User modifiable flags */ /* * Inode dynamic state flags @@ -291,8 +291,6 @@ struct ext4_new_group_data { #define EXT4_IOC_SETFLAGS FS_IOC_SETFLAGS #define EXT4_IOC_GETVERSION _IOR('f', 3, long) #define EXT4_IOC_SETVERSION _IOW('f', 4, long) -#define EXT4_IOC_GROUP_EXTEND _IOW('f', 7, unsigned long) -#define EXT4_IOC_GROUP_ADD _IOW('f', 8,struct ext4_new_group_input) #define EXT4_IOC_GETVERSION_OLD FS_IOC_GETVERSION #define EXT4_IOC_SETVERSION_OLD FS_IOC_SETVERSION #ifdef CONFIG_JBD2_DEBUG @@ -300,7 +298,10 @@ struct ext4_new_group_data { #endif #define EXT4_IOC_GETRSVSZ _IOR('f', 5, long) #define EXT4_IOC_SETRSVSZ _IOW('f', 6, long) -#define EXT4_IOC_MIGRATE _IO('f', 7) +#define EXT4_IOC_GROUP_EXTEND _IOW('f', 7, unsigned long) +#define EXT4_IOC_GROUP_ADD _IOW('f', 8, struct ext4_new_group_input) +#define EXT4_IOC_MIGRATE _IO('f', 9) + /* note ioctl 11 reserved for filesystem-independent FIEMAP ioctl */ /* * ioctl commands in 32 bit emulation @@ -538,8 +539,9 @@ do { \ #define EXT4_MOUNT_JOURNAL_CHECKSUM 0x800000 /* Journal checksums */ #define EXT4_MOUNT_JOURNAL_ASYNC_COMMIT 0x1000000 /* Journal Async Commit */ #define EXT4_MOUNT_I_VERSION 0x2000000 /* i_version support */ -#define EXT4_MOUNT_MBALLOC 0x4000000 /* Buddy allocation support */ #define EXT4_MOUNT_DELALLOC 0x8000000 /* Delalloc support */ +#define EXT4_MOUNT_DATA_ERR_ABORT 0x10000000 /* Abort on file data write */ + /* Compatibility, for having both ext2_fs.h and ext4_fs.h included at once */ #ifndef _LINUX_EXT2_FS_H #define clear_opt(o, opt) o &= ~EXT4_MOUNT_##opt @@ -667,7 +669,7 @@ struct ext4_super_block { }; #ifdef __KERNEL__ -static inline struct ext4_sb_info * EXT4_SB(struct super_block *sb) +static inline struct ext4_sb_info *EXT4_SB(struct super_block *sb) { return sb->s_fs_info; } @@ -725,11 +727,11 @@ static inline int ext4_valid_inum(struct super_block *sb, unsigned long ino) */ #define EXT4_HAS_COMPAT_FEATURE(sb,mask) \ - ( EXT4_SB(sb)->s_es->s_feature_compat & cpu_to_le32(mask) ) + (EXT4_SB(sb)->s_es->s_feature_compat & cpu_to_le32(mask)) #define EXT4_HAS_RO_COMPAT_FEATURE(sb,mask) \ - ( EXT4_SB(sb)->s_es->s_feature_ro_compat & cpu_to_le32(mask) ) + (EXT4_SB(sb)->s_es->s_feature_ro_compat & cpu_to_le32(mask)) #define EXT4_HAS_INCOMPAT_FEATURE(sb,mask) \ - ( EXT4_SB(sb)->s_es->s_feature_incompat & cpu_to_le32(mask) ) + (EXT4_SB(sb)->s_es->s_feature_incompat & cpu_to_le32(mask)) #define EXT4_SET_COMPAT_FEATURE(sb,mask) \ EXT4_SB(sb)->s_es->s_feature_compat |= cpu_to_le32(mask) #define EXT4_SET_RO_COMPAT_FEATURE(sb,mask) \ @@ -789,6 +791,8 @@ static inline int ext4_valid_inum(struct super_block *sb, unsigned long ino) #define EXT4_DEF_RESUID 0 #define EXT4_DEF_RESGID 0 +#define EXT4_DEF_INODE_READAHEAD_BLKS 32 + /* * Default mount options */ @@ -954,6 +958,24 @@ ext4_group_first_block_no(struct super_block *sb, ext4_group_t group_no) void ext4_get_group_no_and_offset(struct super_block *sb, ext4_fsblk_t blocknr, unsigned long *blockgrpp, ext4_grpblk_t *offsetp); +extern struct proc_dir_entry *ext4_proc_root; + +#ifdef CONFIG_PROC_FS +extern const struct file_operations ext4_ui_proc_fops; + +#define EXT4_PROC_HANDLER(name, var) \ +do { \ + proc = proc_create_data(name, mode, sbi->s_proc, \ + &ext4_ui_proc_fops, &sbi->s_##var); \ + if (proc == NULL) { \ + printk(KERN_ERR "EXT4-fs: can't create %s\n", name); \ + goto err_out; \ + } \ +} while (0) +#else +#define EXT4_PROC_HANDLER(name, var) +#endif + /* * Function prototypes */ @@ -981,23 +1003,20 @@ extern ext4_fsblk_t ext4_new_meta_blocks(handle_t *handle, struct inode *inode, extern ext4_fsblk_t ext4_new_blocks(handle_t *handle, struct inode *inode, ext4_lblk_t iblock, ext4_fsblk_t goal, unsigned long *count, int *errp); -extern ext4_fsblk_t ext4_old_new_blocks(handle_t *handle, struct inode *inode, - ext4_fsblk_t goal, unsigned long *count, int *errp); +extern int ext4_claim_free_blocks(struct ext4_sb_info *sbi, s64 nblocks); extern ext4_fsblk_t ext4_has_free_blocks(struct ext4_sb_info *sbi, - ext4_fsblk_t nblocks); -extern void ext4_free_blocks (handle_t *handle, struct inode *inode, + s64 nblocks); +extern void ext4_free_blocks(handle_t *handle, struct inode *inode, ext4_fsblk_t block, unsigned long count, int metadata); -extern void ext4_free_blocks_sb (handle_t *handle, struct super_block *sb, - ext4_fsblk_t block, unsigned long count, +extern void ext4_free_blocks_sb(handle_t *handle, struct super_block *sb, + ext4_fsblk_t block, unsigned long count, unsigned long *pdquot_freed_blocks); -extern ext4_fsblk_t ext4_count_free_blocks (struct super_block *); -extern void ext4_check_blocks_bitmap (struct super_block *); +extern ext4_fsblk_t ext4_count_free_blocks(struct super_block *); +extern void ext4_check_blocks_bitmap(struct super_block *); extern struct ext4_group_desc * ext4_get_group_desc(struct super_block * sb, ext4_group_t block_group, struct buffer_head ** bh); extern int ext4_should_retry_alloc(struct super_block *sb, int *retries); -extern void ext4_init_block_alloc_info(struct inode *); -extern void ext4_rsv_window_add(struct super_block *sb, struct ext4_reserve_window_node *rsv); /* dir.c */ extern int ext4_check_dir_entry(const char *, struct inode *, @@ -1009,20 +1028,20 @@ extern int ext4_htree_store_dirent(struct file *dir_file, __u32 hash, extern void ext4_htree_free_dir_info(struct dir_private_info *p); /* fsync.c */ -extern int ext4_sync_file (struct file *, struct dentry *, int); +extern int ext4_sync_file(struct file *, struct dentry *, int); /* hash.c */ extern int ext4fs_dirhash(const char *name, int len, struct dx_hash_info *hinfo); /* ialloc.c */ -extern struct inode * ext4_new_inode (handle_t *, struct inode *, int); -extern void ext4_free_inode (handle_t *, struct inode *); -extern struct inode * ext4_orphan_get (struct super_block *, unsigned long); -extern unsigned long ext4_count_free_inodes (struct super_block *); -extern unsigned long ext4_count_dirs (struct super_block *); -extern void ext4_check_inodes_bitmap (struct super_block *); -extern unsigned long ext4_count_free (struct buffer_head *, unsigned); +extern struct inode * ext4_new_inode(handle_t *, struct inode *, int); +extern void ext4_free_inode(handle_t *, struct inode *); +extern struct inode * ext4_orphan_get(struct super_block *, unsigned long); +extern unsigned long ext4_count_free_inodes(struct super_block *); +extern unsigned long ext4_count_dirs(struct super_block *); +extern void ext4_check_inodes_bitmap(struct super_block *); +extern unsigned long ext4_count_free(struct buffer_head *, unsigned); /* mballoc.c */ extern long ext4_mb_stats; @@ -1032,7 +1051,7 @@ extern int ext4_mb_release(struct super_block *); extern ext4_fsblk_t ext4_mb_new_blocks(handle_t *, struct ext4_allocation_request *, int *); extern int ext4_mb_reserve_blocks(struct super_block *, int); -extern void ext4_mb_discard_inode_preallocations(struct inode *); +extern void ext4_discard_preallocations(struct inode *); extern int __init init_ext4_mballoc(void); extern void exit_ext4_mballoc(void); extern void ext4_mb_free_blocks(handle_t *, struct inode *, @@ -1050,24 +1069,25 @@ struct buffer_head *ext4_getblk(handle_t *, struct inode *, ext4_lblk_t, int, int *); struct buffer_head *ext4_bread(handle_t *, struct inode *, ext4_lblk_t, int, int *); +int ext4_get_block(struct inode *inode, sector_t iblock, + struct buffer_head *bh_result, int create); int ext4_get_blocks_handle(handle_t *handle, struct inode *inode, ext4_lblk_t iblock, unsigned long maxblocks, struct buffer_head *bh_result, int create, int extend_disksize); extern struct inode *ext4_iget(struct super_block *, unsigned long); -extern int ext4_write_inode (struct inode *, int); -extern int ext4_setattr (struct dentry *, struct iattr *); +extern int ext4_write_inode(struct inode *, int); +extern int ext4_setattr(struct dentry *, struct iattr *); extern int ext4_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat); -extern void ext4_delete_inode (struct inode *); -extern int ext4_sync_inode (handle_t *, struct inode *); -extern void ext4_discard_reservation (struct inode *); +extern void ext4_delete_inode(struct inode *); +extern int ext4_sync_inode(handle_t *, struct inode *); extern void ext4_dirty_inode(struct inode *); extern int ext4_change_inode_journal_flag(struct inode *, int); extern int ext4_get_inode_loc(struct inode *, struct ext4_iloc *); extern int ext4_can_truncate(struct inode *inode); -extern void ext4_truncate (struct inode *); +extern void ext4_truncate(struct inode *); extern void ext4_set_inode_flags(struct inode *); extern void ext4_get_inode_flags(struct ext4_inode_info *); extern void ext4_set_aops(struct inode *inode); @@ -1080,11 +1100,10 @@ extern int ext4_page_mkwrite(struct vm_area_struct *vma, struct page *page); /* ioctl.c */ extern long ext4_ioctl(struct file *, unsigned int, unsigned long); -extern long ext4_compat_ioctl (struct file *, unsigned int, unsigned long); +extern long ext4_compat_ioctl(struct file *, unsigned int, unsigned long); /* migrate.c */ -extern int ext4_ext_migrate(struct inode *, struct file *, unsigned int, - unsigned long); +extern int ext4_ext_migrate(struct inode *); /* namei.c */ extern int ext4_orphan_add(handle_t *, struct inode *); extern int ext4_orphan_del(handle_t *, struct inode *); @@ -1099,14 +1118,14 @@ extern int ext4_group_extend(struct super_block *sb, ext4_fsblk_t n_blocks_count); /* super.c */ -extern void ext4_error (struct super_block *, const char *, const char *, ...) +extern void ext4_error(struct super_block *, const char *, const char *, ...) __attribute__ ((format (printf, 3, 4))); -extern void __ext4_std_error (struct super_block *, const char *, int); -extern void ext4_abort (struct super_block *, const char *, const char *, ...) +extern void __ext4_std_error(struct super_block *, const char *, int); +extern void ext4_abort(struct super_block *, const char *, const char *, ...) __attribute__ ((format (printf, 3, 4))); -extern void ext4_warning (struct super_block *, const char *, const char *, ...) +extern void ext4_warning(struct super_block *, const char *, const char *, ...) __attribute__ ((format (printf, 3, 4))); -extern void ext4_update_dynamic_rev (struct super_block *sb); +extern void ext4_update_dynamic_rev(struct super_block *sb); extern int ext4_update_compat_feature(handle_t *handle, struct super_block *sb, __u32 compat); extern int ext4_update_rocompat_feature(handle_t *handle, @@ -1179,7 +1198,7 @@ static inline void ext4_isize_set(struct ext4_inode *raw_inode, loff_t i_size) static inline struct ext4_group_info *ext4_get_group_info(struct super_block *sb, - ext4_group_t group) + ext4_group_t group) { struct ext4_group_info ***grp_info; long indexv, indexh; @@ -1207,6 +1226,28 @@ do { \ __ext4_std_error((sb), __func__, (errno)); \ } while (0) +#ifdef CONFIG_SMP +/* Each CPU can accumulate FBC_BATCH blocks in their local + * counters. So we need to make sure we have free blocks more + * than FBC_BATCH * nr_cpu_ids. Also add a window of 4 times. + */ +#define EXT4_FREEBLOCKS_WATERMARK (4 * (FBC_BATCH * nr_cpu_ids)) +#else +#define EXT4_FREEBLOCKS_WATERMARK 0 +#endif + +static inline void ext4_update_i_disksize(struct inode *inode, loff_t newsize) +{ + /* + * XXX: replace with spinlock if seen contended -bzzz + */ + down_write(&EXT4_I(inode)->i_data_sem); + if (newsize > EXT4_I(inode)->i_disksize) + EXT4_I(inode)->i_disksize = newsize; + up_write(&EXT4_I(inode)->i_data_sem); + return ; +} + /* * Inodes and files operations */ diff --git a/fs/ext4/ext4_extents.h b/fs/ext4/ext4_extents.h index d33dc56d698..bec7ce59fc0 100644 --- a/fs/ext4/ext4_extents.h +++ b/fs/ext4/ext4_extents.h @@ -124,6 +124,19 @@ struct ext4_ext_path { #define EXT4_EXT_CACHE_GAP 1 #define EXT4_EXT_CACHE_EXTENT 2 +/* + * to be called by ext4_ext_walk_space() + * negative retcode - error + * positive retcode - signal for ext4_ext_walk_space(), see below + * callback must return valid extent (passed or newly created) + */ +typedef int (*ext_prepare_callback)(struct inode *, struct ext4_ext_path *, + struct ext4_ext_cache *, + struct ext4_extent *, void *); + +#define EXT_CONTINUE 0 +#define EXT_BREAK 1 +#define EXT_REPEAT 2 #define EXT_MAX_BLOCK 0xffffffff @@ -224,6 +237,8 @@ extern int ext4_ext_try_to_merge(struct inode *inode, struct ext4_extent *); extern unsigned int ext4_ext_check_overlap(struct inode *, struct ext4_extent *, struct ext4_ext_path *); extern int ext4_ext_insert_extent(handle_t *, struct inode *, struct ext4_ext_path *, struct ext4_extent *); +extern int ext4_ext_walk_space(struct inode *, ext4_lblk_t, ext4_lblk_t, + ext_prepare_callback, void *); extern struct ext4_ext_path *ext4_ext_find_extent(struct inode *, ext4_lblk_t, struct ext4_ext_path *); extern int ext4_ext_search_left(struct inode *, struct ext4_ext_path *, diff --git a/fs/ext4/ext4_i.h b/fs/ext4/ext4_i.h index ef7409f0e7e..5c124c0ac6d 100644 --- a/fs/ext4/ext4_i.h +++ b/fs/ext4/ext4_i.h @@ -33,38 +33,6 @@ typedef __u32 ext4_lblk_t; /* data type for block group number */ typedef unsigned long ext4_group_t; -struct ext4_reserve_window { - ext4_fsblk_t _rsv_start; /* First byte reserved */ - ext4_fsblk_t _rsv_end; /* Last byte reserved or 0 */ -}; - -struct ext4_reserve_window_node { - struct rb_node rsv_node; - __u32 rsv_goal_size; - __u32 rsv_alloc_hit; - struct ext4_reserve_window rsv_window; -}; - -struct ext4_block_alloc_info { - /* information about reservation window */ - struct ext4_reserve_window_node rsv_window_node; - /* - * was i_next_alloc_block in ext4_inode_info - * is the logical (file-relative) number of the - * most-recently-allocated block in this file. - * We use this for detecting linearly ascending allocation requests. - */ - ext4_lblk_t last_alloc_logical_block; - /* - * Was i_next_alloc_goal in ext4_inode_info - * is the *physical* companion to i_next_alloc_block. - * it the physical block number of the block which was most-recentl - * allocated to this file. This give us the goal (target) for the next - * allocation when we detect linearly ascending requests. - */ - ext4_fsblk_t last_alloc_physical_block; -}; - #define rsv_start rsv_window._rsv_start #define rsv_end rsv_window._rsv_end @@ -97,11 +65,8 @@ struct ext4_inode_info { ext4_group_t i_block_group; __u32 i_state; /* Dynamic state flags for ext4 */ - /* block reservation info */ - struct ext4_block_alloc_info *i_block_alloc_info; - ext4_lblk_t i_dir_start_lookup; -#ifdef CONFIG_EXT4DEV_FS_XATTR +#ifdef CONFIG_EXT4_FS_XATTR /* * Extended attributes can be read independently of the main file * data. Taking i_mutex even when reading would cause contention @@ -111,7 +76,7 @@ struct ext4_inode_info { */ struct rw_semaphore xattr_sem; #endif -#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL +#ifdef CONFIG_EXT4_FS_POSIX_ACL struct posix_acl *i_acl; struct posix_acl *i_default_acl; #endif diff --git a/fs/ext4/ext4_sb.h b/fs/ext4/ext4_sb.h index 6300226d553..6a0b40d4326 100644 --- a/fs/ext4/ext4_sb.h +++ b/fs/ext4/ext4_sb.h @@ -40,8 +40,8 @@ struct ext4_sb_info { unsigned long s_blocks_last; /* Last seen block count */ loff_t s_bitmap_maxbytes; /* max bytes for bitmap files */ struct buffer_head * s_sbh; /* Buffer containing the super block */ - struct ext4_super_block * s_es; /* Pointer to the super block in the buffer */ - struct buffer_head ** s_group_desc; + struct ext4_super_block *s_es; /* Pointer to the super block in the buffer */ + struct buffer_head **s_group_desc; unsigned long s_mount_opt; ext4_fsblk_t s_sb_block; uid_t s_resuid; @@ -52,6 +52,7 @@ struct ext4_sb_info { int s_desc_per_block_bits; int s_inode_size; int s_first_ino; + unsigned int s_inode_readahead_blks; spinlock_t s_next_gen_lock; u32 s_next_generation; u32 s_hash_seed[4]; @@ -59,16 +60,17 @@ struct ext4_sb_info { struct percpu_counter s_freeblocks_counter; struct percpu_counter s_freeinodes_counter; struct percpu_counter s_dirs_counter; + struct percpu_counter s_dirtyblocks_counter; struct blockgroup_lock s_blockgroup_lock; + struct proc_dir_entry *s_proc; /* root of the per fs reservation window tree */ spinlock_t s_rsv_window_lock; struct rb_root s_rsv_window_root; - struct ext4_reserve_window_node s_rsv_window_head; /* Journaling */ - struct inode * s_journal_inode; - struct journal_s * s_journal; + struct inode *s_journal_inode; + struct journal_s *s_journal; struct list_head s_orphan; unsigned long s_commit_interval; struct block_device *journal_bdev; @@ -106,12 +108,12 @@ struct ext4_sb_info { /* tunables */ unsigned long s_stripe; - unsigned long s_mb_stream_request; - unsigned long s_mb_max_to_scan; - unsigned long s_mb_min_to_scan; - unsigned long s_mb_stats; - unsigned long s_mb_order2_reqs; - unsigned long s_mb_group_prealloc; + unsigned int s_mb_stream_request; + unsigned int s_mb_max_to_scan; + unsigned int s_mb_min_to_scan; + unsigned int s_mb_stats; + unsigned int s_mb_order2_reqs; + unsigned int s_mb_group_prealloc; /* where last allocation was done - for stream allocation */ unsigned long s_mb_last_group; unsigned long s_mb_last_start; @@ -121,7 +123,6 @@ struct ext4_sb_info { int s_mb_history_cur; int s_mb_history_max; int s_mb_history_num; - struct proc_dir_entry *s_mb_proc; spinlock_t s_mb_history_lock; int s_mb_history_filter; diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index b24d3c53f20..ea2ce3c0ae6 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -40,6 +40,7 @@ #include <linux/slab.h> #include <linux/falloc.h> #include <asm/uaccess.h> +#include <linux/fiemap.h> #include "ext4_jbd2.h" #include "ext4_extents.h" @@ -383,8 +384,8 @@ static void ext4_ext_show_leaf(struct inode *inode, struct ext4_ext_path *path) ext_debug("\n"); } #else -#define ext4_ext_show_path(inode,path) -#define ext4_ext_show_leaf(inode,path) +#define ext4_ext_show_path(inode, path) +#define ext4_ext_show_leaf(inode, path) #endif void ext4_ext_drop_refs(struct ext4_ext_path *path) @@ -440,9 +441,10 @@ ext4_ext_binsearch_idx(struct inode *inode, for (k = 0; k < le16_to_cpu(eh->eh_entries); k++, ix++) { if (k != 0 && le32_to_cpu(ix->ei_block) <= le32_to_cpu(ix[-1].ei_block)) { - printk("k=%d, ix=0x%p, first=0x%p\n", k, - ix, EXT_FIRST_INDEX(eh)); - printk("%u <= %u\n", + printk(KERN_DEBUG "k=%d, ix=0x%p, " + "first=0x%p\n", k, + ix, EXT_FIRST_INDEX(eh)); + printk(KERN_DEBUG "%u <= %u\n", le32_to_cpu(ix->ei_block), le32_to_cpu(ix[-1].ei_block)); } @@ -1475,7 +1477,7 @@ int ext4_ext_insert_extent(handle_t *handle, struct inode *inode, struct ext4_ext_path *path, struct ext4_extent *newext) { - struct ext4_extent_header * eh; + struct ext4_extent_header *eh; struct ext4_extent *ex, *fex; struct ext4_extent *nearex; /* nearest extent */ struct ext4_ext_path *npath = NULL; @@ -1625,6 +1627,113 @@ cleanup: return err; } +int ext4_ext_walk_space(struct inode *inode, ext4_lblk_t block, + ext4_lblk_t num, ext_prepare_callback func, + void *cbdata) +{ + struct ext4_ext_path *path = NULL; + struct ext4_ext_cache cbex; + struct ext4_extent *ex; + ext4_lblk_t next, start = 0, end = 0; + ext4_lblk_t last = block + num; + int depth, exists, err = 0; + + BUG_ON(func == NULL); + BUG_ON(inode == NULL); + + while (block < last && block != EXT_MAX_BLOCK) { + num = last - block; + /* find extent for this block */ + path = ext4_ext_find_extent(inode, block, path); + if (IS_ERR(path)) { + err = PTR_ERR(path); + path = NULL; + break; + } + + depth = ext_depth(inode); + BUG_ON(path[depth].p_hdr == NULL); + ex = path[depth].p_ext; + next = ext4_ext_next_allocated_block(path); + + exists = 0; + if (!ex) { + /* there is no extent yet, so try to allocate + * all requested space */ + start = block; + end = block + num; + } else if (le32_to_cpu(ex->ee_block) > block) { + /* need to allocate space before found extent */ + start = block; + end = le32_to_cpu(ex->ee_block); + if (block + num < end) + end = block + num; + } else if (block >= le32_to_cpu(ex->ee_block) + + ext4_ext_get_actual_len(ex)) { + /* need to allocate space after found extent */ + start = block; + end = block + num; + if (end >= next) + end = next; + } else if (block >= le32_to_cpu(ex->ee_block)) { + /* + * some part of requested space is covered + * by found extent + */ + start = block; + end = le32_to_cpu(ex->ee_block) + + ext4_ext_get_actual_len(ex); + if (block + num < end) + end = block + num; + exists = 1; + } else { + BUG(); + } + BUG_ON(end <= start); + + if (!exists) { + cbex.ec_block = start; + cbex.ec_len = end - start; + cbex.ec_start = 0; + cbex.ec_type = EXT4_EXT_CACHE_GAP; + } else { + cbex.ec_block = le32_to_cpu(ex->ee_block); + cbex.ec_len = ext4_ext_get_actual_len(ex); + cbex.ec_start = ext_pblock(ex); + cbex.ec_type = EXT4_EXT_CACHE_EXTENT; + } + + BUG_ON(cbex.ec_len == 0); + err = func(inode, path, &cbex, ex, cbdata); + ext4_ext_drop_refs(path); + + if (err < 0) + break; + + if (err == EXT_REPEAT) + continue; + else if (err == EXT_BREAK) { + err = 0; + break; + } + + if (ext_depth(inode) != depth) { + /* depth was changed. we have to realloc path */ + kfree(path); + path = NULL; + } + + block = cbex.ec_block + cbex.ec_len; + } + + if (path) { + ext4_ext_drop_refs(path); + kfree(path); + } + + return err; +} + static void ext4_ext_put_in_cache(struct inode *inode, ext4_lblk_t block, __u32 len, ext4_fsblk_t start, int type) @@ -2142,7 +2251,7 @@ void ext4_ext_init(struct super_block *sb) */ if (test_opt(sb, EXTENTS)) { - printk("EXT4-fs: file extents enabled"); + printk(KERN_INFO "EXT4-fs: file extents enabled"); #ifdef AGGRESSIVE_TEST printk(", aggressive tests"); #endif @@ -2696,11 +2805,8 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, goto out2; } /* - * Okay, we need to do block allocation. Lazily initialize the block - * allocation info here if necessary. + * Okay, we need to do block allocation. */ - if (S_ISREG(inode->i_mode) && (!EXT4_I(inode)->i_block_alloc_info)) - ext4_init_block_alloc_info(inode); /* find neighbour allocated blocks */ ar.lleft = iblock; @@ -2760,7 +2866,7 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, /* free data blocks we just allocated */ /* not a good idea to call discard here directly, * but otherwise we'd need to call it every free() */ - ext4_mb_discard_inode_preallocations(inode); + ext4_discard_preallocations(inode); ext4_free_blocks(handle, inode, ext_pblock(&newex), ext4_ext_get_actual_len(&newex), 0); goto out2; @@ -2824,7 +2930,7 @@ void ext4_ext_truncate(struct inode *inode) down_write(&EXT4_I(inode)->i_data_sem); ext4_ext_invalidate_cache(inode); - ext4_discard_reservation(inode); + ext4_discard_preallocations(inode); /* * TODO: optimization is possible here. @@ -2877,10 +2983,11 @@ static void ext4_falloc_update_inode(struct inode *inode, * Update only when preallocation was requested beyond * the file size. */ - if (!(mode & FALLOC_FL_KEEP_SIZE) && - new_size > i_size_read(inode)) { - i_size_write(inode, new_size); - EXT4_I(inode)->i_disksize = new_size; + if (!(mode & FALLOC_FL_KEEP_SIZE)) { + if (new_size > i_size_read(inode)) + i_size_write(inode, new_size); + if (new_size > EXT4_I(inode)->i_disksize) + ext4_update_i_disksize(inode, new_size); } } @@ -2972,3 +3079,143 @@ retry: mutex_unlock(&inode->i_mutex); return ret > 0 ? ret2 : ret; } + +/* + * Callback function called for each extent to gather FIEMAP information. + */ +int ext4_ext_fiemap_cb(struct inode *inode, struct ext4_ext_path *path, + struct ext4_ext_cache *newex, struct ext4_extent *ex, + void *data) +{ + struct fiemap_extent_info *fieinfo = data; + unsigned long blksize_bits = inode->i_sb->s_blocksize_bits; + __u64 logical; + __u64 physical; + __u64 length; + __u32 flags = 0; + int error; + + logical = (__u64)newex->ec_block << blksize_bits; + + if (newex->ec_type == EXT4_EXT_CACHE_GAP) { + pgoff_t offset; + struct page *page; + struct buffer_head *bh = NULL; + + offset = logical >> PAGE_SHIFT; + page = find_get_page(inode->i_mapping, offset); + if (!page || !page_has_buffers(page)) + return EXT_CONTINUE; + + bh = page_buffers(page); + + if (!bh) + return EXT_CONTINUE; + + if (buffer_delay(bh)) { + flags |= FIEMAP_EXTENT_DELALLOC; + page_cache_release(page); + } else { + page_cache_release(page); + return EXT_CONTINUE; + } + } + + physical = (__u64)newex->ec_start << blksize_bits; + length = (__u64)newex->ec_len << blksize_bits; + + if (ex && ext4_ext_is_uninitialized(ex)) + flags |= FIEMAP_EXTENT_UNWRITTEN; + + /* + * If this extent reaches EXT_MAX_BLOCK, it must be last. + * + * Or if ext4_ext_next_allocated_block is EXT_MAX_BLOCK, + * this also indicates no more allocated blocks. + * + * XXX this might miss a single-block extent at EXT_MAX_BLOCK + */ + if (logical + length - 1 == EXT_MAX_BLOCK || + ext4_ext_next_allocated_block(path) == EXT_MAX_BLOCK) + flags |= FIEMAP_EXTENT_LAST; + + error = fiemap_fill_next_extent(fieinfo, logical, physical, + length, flags); + if (error < 0) + return error; + if (error == 1) + return EXT_BREAK; + + return EXT_CONTINUE; +} + +/* fiemap flags we can handle specified here */ +#define EXT4_FIEMAP_FLAGS (FIEMAP_FLAG_SYNC|FIEMAP_FLAG_XATTR) + +int ext4_xattr_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo) +{ + __u64 physical = 0; + __u64 length; + __u32 flags = FIEMAP_EXTENT_LAST; + int blockbits = inode->i_sb->s_blocksize_bits; + int error = 0; + + /* in-inode? */ + if (EXT4_I(inode)->i_state & EXT4_STATE_XATTR) { + struct ext4_iloc iloc; + int offset; /* offset of xattr in inode */ + + error = ext4_get_inode_loc(inode, &iloc); + if (error) + return error; + physical = iloc.bh->b_blocknr << blockbits; + offset = EXT4_GOOD_OLD_INODE_SIZE + + EXT4_I(inode)->i_extra_isize; + physical += offset; + length = EXT4_SB(inode->i_sb)->s_inode_size - offset; + flags |= FIEMAP_EXTENT_DATA_INLINE; + } else { /* external block */ + physical = EXT4_I(inode)->i_file_acl << blockbits; + length = inode->i_sb->s_blocksize; + } + + if (physical) + error = fiemap_fill_next_extent(fieinfo, 0, physical, + length, flags); + return (error < 0 ? error : 0); +} + +int ext4_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, + __u64 start, __u64 len) +{ + ext4_lblk_t start_blk; + ext4_lblk_t len_blks; + int error = 0; + + /* fallback to generic here if not in extents fmt */ + if (!(EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL)) + return generic_block_fiemap(inode, fieinfo, start, len, + ext4_get_block); + + if (fiemap_check_flags(fieinfo, EXT4_FIEMAP_FLAGS)) + return -EBADR; + + if (fieinfo->fi_flags & FIEMAP_FLAG_XATTR) { + error = ext4_xattr_fiemap(inode, fieinfo); + } else { + start_blk = start >> inode->i_sb->s_blocksize_bits; + len_blks = len >> inode->i_sb->s_blocksize_bits; + + /* + * Walk the extent tree gathering extent information. + * ext4_ext_fiemap_cb will push extents back to user. + */ + down_write(&EXT4_I(inode)->i_data_sem); + error = ext4_ext_walk_space(inode, start_blk, len_blks, + ext4_ext_fiemap_cb, fieinfo); + up_write(&EXT4_I(inode)->i_data_sem); + } + + return error; +} + diff --git a/fs/ext4/file.c b/fs/ext4/file.c index 430eb7978db..6bd11fba71f 100644 --- a/fs/ext4/file.c +++ b/fs/ext4/file.c @@ -31,14 +31,14 @@ * from ext4_file_open: open gets called at every open, but release * gets called only when /all/ the files are closed. */ -static int ext4_release_file (struct inode * inode, struct file * filp) +static int ext4_release_file(struct inode *inode, struct file *filp) { /* if we are the last writer on the inode, drop the block reservation */ if ((filp->f_mode & FMODE_WRITE) && (atomic_read(&inode->i_writecount) == 1)) { down_write(&EXT4_I(inode)->i_data_sem); - ext4_discard_reservation(inode); + ext4_discard_preallocations(inode); up_write(&EXT4_I(inode)->i_data_sem); } if (is_dx(inode) && filp->private_data) @@ -140,6 +140,9 @@ static int ext4_file_mmap(struct file *file, struct vm_area_struct *vma) return 0; } +extern int ext4_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, + __u64 start, __u64 len); + const struct file_operations ext4_file_operations = { .llseek = generic_file_llseek, .read = do_sync_read, @@ -162,7 +165,7 @@ const struct inode_operations ext4_file_inode_operations = { .truncate = ext4_truncate, .setattr = ext4_setattr, .getattr = ext4_getattr, -#ifdef CONFIG_EXT4DEV_FS_XATTR +#ifdef CONFIG_EXT4_FS_XATTR .setxattr = generic_setxattr, .getxattr = generic_getxattr, .listxattr = ext4_listxattr, @@ -170,5 +173,6 @@ const struct inode_operations ext4_file_inode_operations = { #endif .permission = ext4_permission, .fallocate = ext4_fallocate, + .fiemap = ext4_fiemap, }; diff --git a/fs/ext4/fsync.c b/fs/ext4/fsync.c index a45c3737ad3..5afe4370840 100644 --- a/fs/ext4/fsync.c +++ b/fs/ext4/fsync.c @@ -28,6 +28,7 @@ #include <linux/writeback.h> #include <linux/jbd2.h> #include <linux/blkdev.h> +#include <linux/marker.h> #include "ext4.h" #include "ext4_jbd2.h" @@ -43,7 +44,7 @@ * inode to disk. */ -int ext4_sync_file(struct file * file, struct dentry *dentry, int datasync) +int ext4_sync_file(struct file *file, struct dentry *dentry, int datasync) { struct inode *inode = dentry->d_inode; journal_t *journal = EXT4_SB(inode->i_sb)->s_journal; @@ -51,6 +52,10 @@ int ext4_sync_file(struct file * file, struct dentry *dentry, int datasync) J_ASSERT(ext4_journal_current_handle() == NULL); + trace_mark(ext4_sync_file, "dev %s datasync %d ino %ld parent %ld", + inode->i_sb->s_id, datasync, inode->i_ino, + dentry->d_parent->d_inode->i_ino); + /* * data=writeback: * The caller's filemap_fdatawrite()/wait will sync the data. diff --git a/fs/ext4/hash.c b/fs/ext4/hash.c index 1d6329dbe39..556ca8eba3d 100644 --- a/fs/ext4/hash.c +++ b/fs/ext4/hash.c @@ -27,7 +27,7 @@ static void TEA_transform(__u32 buf[4], __u32 const in[]) sum += DELTA; b0 += ((b1 << 4)+a) ^ (b1+sum) ^ ((b1 >> 5)+b); b1 += ((b0 << 4)+c) ^ (b0+sum) ^ ((b0 >> 5)+d); - } while(--n); + } while (--n); buf[0] += b0; buf[1] += b1; @@ -35,7 +35,7 @@ static void TEA_transform(__u32 buf[4], __u32 const in[]) /* The old legacy hash */ -static __u32 dx_hack_hash (const char *name, int len) +static __u32 dx_hack_hash(const char *name, int len) { __u32 hash0 = 0x12a3fe2d, hash1 = 0x37abe8f9; while (len--) { @@ -59,7 +59,7 @@ static void str2hashbuf(const char *msg, int len, __u32 *buf, int num) val = pad; if (len > num*4) len = num * 4; - for (i=0; i < len; i++) { + for (i = 0; i < len; i++) { if ((i % 4) == 0) val = pad; val = msg[i] + (val << 8); @@ -104,7 +104,7 @@ int ext4fs_dirhash(const char *name, int len, struct dx_hash_info *hinfo) /* Check to see if the seed is all zero's */ if (hinfo->seed) { - for (i=0; i < 4; i++) { + for (i = 0; i < 4; i++) { if (hinfo->seed[i]) break; } diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c index f344834bbf5..fe34d74cfb1 100644 --- a/fs/ext4/ialloc.c +++ b/fs/ext4/ialloc.c @@ -115,9 +115,11 @@ ext4_read_inode_bitmap(struct super_block *sb, ext4_group_t block_group) block_group, bitmap_blk); return NULL; } - if (bh_uptodate_or_lock(bh)) + if (buffer_uptodate(bh) && + !(desc->bg_flags & cpu_to_le16(EXT4_BG_INODE_UNINIT))) return bh; + lock_buffer(bh); spin_lock(sb_bgl_lock(EXT4_SB(sb), block_group)); if (desc->bg_flags & cpu_to_le16(EXT4_BG_INODE_UNINIT)) { ext4_init_inode_bitmap(sb, bh, block_group, desc); @@ -154,39 +156,40 @@ ext4_read_inode_bitmap(struct super_block *sb, ext4_group_t block_group) * though), and then we'd have two inodes sharing the * same inode number and space on the harddisk. */ -void ext4_free_inode (handle_t *handle, struct inode * inode) +void ext4_free_inode(handle_t *handle, struct inode *inode) { - struct super_block * sb = inode->i_sb; + struct super_block *sb = inode->i_sb; int is_directory; unsigned long ino; struct buffer_head *bitmap_bh = NULL; struct buffer_head *bh2; ext4_group_t block_group; unsigned long bit; - struct ext4_group_desc * gdp; - struct ext4_super_block * es; + struct ext4_group_desc *gdp; + struct ext4_super_block *es; struct ext4_sb_info *sbi; int fatal = 0, err; ext4_group_t flex_group; if (atomic_read(&inode->i_count) > 1) { - printk ("ext4_free_inode: inode has count=%d\n", - atomic_read(&inode->i_count)); + printk(KERN_ERR "ext4_free_inode: inode has count=%d\n", + atomic_read(&inode->i_count)); return; } if (inode->i_nlink) { - printk ("ext4_free_inode: inode has nlink=%d\n", - inode->i_nlink); + printk(KERN_ERR "ext4_free_inode: inode has nlink=%d\n", + inode->i_nlink); return; } if (!sb) { - printk("ext4_free_inode: inode on nonexistent device\n"); + printk(KERN_ERR "ext4_free_inode: inode on " + "nonexistent device\n"); return; } sbi = EXT4_SB(sb); ino = inode->i_ino; - ext4_debug ("freeing inode %lu\n", ino); + ext4_debug("freeing inode %lu\n", ino); /* * Note: we must free any quota before locking the superblock, @@ -200,12 +203,12 @@ void ext4_free_inode (handle_t *handle, struct inode * inode) is_directory = S_ISDIR(inode->i_mode); /* Do this BEFORE marking the inode not in use or returning an error */ - clear_inode (inode); + clear_inode(inode); es = EXT4_SB(sb)->s_es; if (ino < EXT4_FIRST_INO(sb) || ino > le32_to_cpu(es->s_inodes_count)) { - ext4_error (sb, "ext4_free_inode", - "reserved or nonexistent inode %lu", ino); + ext4_error(sb, "ext4_free_inode", + "reserved or nonexistent inode %lu", ino); goto error_return; } block_group = (ino - 1) / EXT4_INODES_PER_GROUP(sb); @@ -222,10 +225,10 @@ void ext4_free_inode (handle_t *handle, struct inode * inode) /* Ok, now we can actually update the inode bitmaps.. */ if (!ext4_clear_bit_atomic(sb_bgl_lock(sbi, block_group), bit, bitmap_bh->b_data)) - ext4_error (sb, "ext4_free_inode", - "bit already cleared for inode %lu", ino); + ext4_error(sb, "ext4_free_inode", + "bit already cleared for inode %lu", ino); else { - gdp = ext4_get_group_desc (sb, block_group, &bh2); + gdp = ext4_get_group_desc(sb, block_group, &bh2); BUFFER_TRACE(bh2, "get_write_access"); fatal = ext4_journal_get_write_access(handle, bh2); @@ -287,7 +290,7 @@ static int find_group_dir(struct super_block *sb, struct inode *parent, avefreei = freei / ngroups; for (group = 0; group < ngroups; group++) { - desc = ext4_get_group_desc (sb, group, NULL); + desc = ext4_get_group_desc(sb, group, NULL); if (!desc || !desc->bg_free_inodes_count) continue; if (le16_to_cpu(desc->bg_free_inodes_count) < avefreei) @@ -576,16 +579,16 @@ static int find_group_other(struct super_block *sb, struct inode *parent, * For other inodes, search forward from the parent directory's block * group to find a free inode. */ -struct inode *ext4_new_inode(handle_t *handle, struct inode * dir, int mode) +struct inode *ext4_new_inode(handle_t *handle, struct inode *dir, int mode) { struct super_block *sb; struct buffer_head *bitmap_bh = NULL; struct buffer_head *bh2; ext4_group_t group = 0; unsigned long ino = 0; - struct inode * inode; - struct ext4_group_desc * gdp = NULL; - struct ext4_super_block * es; + struct inode *inode; + struct ext4_group_desc *gdp = NULL; + struct ext4_super_block *es; struct ext4_inode_info *ei; struct ext4_sb_info *sbi; int ret2, err = 0; @@ -613,7 +616,7 @@ struct inode *ext4_new_inode(handle_t *handle, struct inode * dir, int mode) } if (S_ISDIR(mode)) { - if (test_opt (sb, OLDALLOC)) + if (test_opt(sb, OLDALLOC)) ret2 = find_group_dir(sb, dir, &group); else ret2 = find_group_orlov(sb, dir, &group); @@ -783,7 +786,7 @@ got: } inode->i_uid = current->fsuid; - if (test_opt (sb, GRPID)) + if (test_opt(sb, GRPID)) inode->i_gid = dir->i_gid; else if (dir->i_mode & S_ISGID) { inode->i_gid = dir->i_gid; @@ -816,7 +819,6 @@ got: ei->i_flags &= ~EXT4_DIRSYNC_FL; ei->i_file_acl = 0; ei->i_dtime = 0; - ei->i_block_alloc_info = NULL; ei->i_block_group = group; ext4_set_inode_flags(inode); @@ -832,7 +834,7 @@ got: ei->i_extra_isize = EXT4_SB(sb)->s_want_extra_isize; ret = inode; - if(DQUOT_ALLOC_INODE(inode)) { + if (DQUOT_ALLOC_INODE(inode)) { err = -EDQUOT; goto fail_drop; } @@ -841,7 +843,7 @@ got: if (err) goto fail_free_drop; - err = ext4_init_security(handle,inode, dir); + err = ext4_init_security(handle, inode, dir); if (err) goto fail_free_drop; @@ -959,7 +961,7 @@ error: return ERR_PTR(err); } -unsigned long ext4_count_free_inodes (struct super_block * sb) +unsigned long ext4_count_free_inodes(struct super_block *sb) { unsigned long desc_count; struct ext4_group_desc *gdp; @@ -974,7 +976,7 @@ unsigned long ext4_count_free_inodes (struct super_block * sb) bitmap_count = 0; gdp = NULL; for (i = 0; i < EXT4_SB(sb)->s_groups_count; i++) { - gdp = ext4_get_group_desc (sb, i, NULL); + gdp = ext4_get_group_desc(sb, i, NULL); if (!gdp) continue; desc_count += le16_to_cpu(gdp->bg_free_inodes_count); @@ -989,13 +991,14 @@ unsigned long ext4_count_free_inodes (struct super_block * sb) bitmap_count += x; } brelse(bitmap_bh); - printk("ext4_count_free_inodes: stored = %u, computed = %lu, %lu\n", - le32_to_cpu(es->s_free_inodes_count), desc_count, bitmap_count); + printk(KERN_DEBUG "ext4_count_free_inodes: " + "stored = %u, computed = %lu, %lu\n", + le32_to_cpu(es->s_free_inodes_count), desc_count, bitmap_count); return desc_count; #else desc_count = 0; for (i = 0; i < EXT4_SB(sb)->s_groups_count; i++) { - gdp = ext4_get_group_desc (sb, i, NULL); + gdp = ext4_get_group_desc(sb, i, NULL); if (!gdp) continue; desc_count += le16_to_cpu(gdp->bg_free_inodes_count); @@ -1006,13 +1009,13 @@ unsigned long ext4_count_free_inodes (struct super_block * sb) } /* Called at mount-time, super-block is locked */ -unsigned long ext4_count_dirs (struct super_block * sb) +unsigned long ext4_count_dirs(struct super_block * sb) { unsigned long count = 0; ext4_group_t i; for (i = 0; i < EXT4_SB(sb)->s_groups_count; i++) { - struct ext4_group_desc *gdp = ext4_get_group_desc (sb, i, NULL); + struct ext4_group_desc *gdp = ext4_get_group_desc(sb, i, NULL); if (!gdp) continue; count += le16_to_cpu(gdp->bg_used_dirs_count); diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 7e91913e325..9b4ec9decfd 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -190,7 +190,7 @@ static int ext4_journal_test_restart(handle_t *handle, struct inode *inode) /* * Called at the last iput() if i_nlink is zero. */ -void ext4_delete_inode (struct inode * inode) +void ext4_delete_inode(struct inode *inode) { handle_t *handle; int err; @@ -330,11 +330,11 @@ static int ext4_block_to_path(struct inode *inode, int final = 0; if (i_block < 0) { - ext4_warning (inode->i_sb, "ext4_block_to_path", "block < 0"); + ext4_warning(inode->i_sb, "ext4_block_to_path", "block < 0"); } else if (i_block < direct_blocks) { offsets[n++] = i_block; final = direct_blocks; - } else if ( (i_block -= direct_blocks) < indirect_blocks) { + } else if ((i_block -= direct_blocks) < indirect_blocks) { offsets[n++] = EXT4_IND_BLOCK; offsets[n++] = i_block; final = ptrs; @@ -400,14 +400,14 @@ static Indirect *ext4_get_branch(struct inode *inode, int depth, *err = 0; /* i_data is not going away, no lock needed */ - add_chain (chain, NULL, EXT4_I(inode)->i_data + *offsets); + add_chain(chain, NULL, EXT4_I(inode)->i_data + *offsets); if (!p->key) goto no_block; while (--depth) { bh = sb_bread(sb, le32_to_cpu(p->key)); if (!bh) goto failure; - add_chain(++p, bh, (__le32*)bh->b_data + *++offsets); + add_chain(++p, bh, (__le32 *)bh->b_data + *++offsets); /* Reader: end */ if (!p->key) goto no_block; @@ -443,7 +443,7 @@ no_block: static ext4_fsblk_t ext4_find_near(struct inode *inode, Indirect *ind) { struct ext4_inode_info *ei = EXT4_I(inode); - __le32 *start = ind->bh ? (__le32*) ind->bh->b_data : ei->i_data; + __le32 *start = ind->bh ? (__le32 *) ind->bh->b_data : ei->i_data; __le32 *p; ext4_fsblk_t bg_start; ext4_fsblk_t last_block; @@ -486,18 +486,9 @@ static ext4_fsblk_t ext4_find_near(struct inode *inode, Indirect *ind) static ext4_fsblk_t ext4_find_goal(struct inode *inode, ext4_lblk_t block, Indirect *partial) { - struct ext4_block_alloc_info *block_i; - - block_i = EXT4_I(inode)->i_block_alloc_info; - /* - * try the heuristic for sequential allocation, - * failing that at least try to get decent locality. + * XXX need to get goal block from mballoc's data structures */ - if (block_i && (block == block_i->last_alloc_logical_block + 1) - && (block_i->last_alloc_physical_block != 0)) { - return block_i->last_alloc_physical_block + 1; - } return ext4_find_near(inode, partial); } @@ -630,7 +621,7 @@ allocated: *err = 0; return ret; failed_out: - for (i = 0; i <index; i++) + for (i = 0; i < index; i++) ext4_free_blocks(handle, inode, new_blocks[i], 1, 0); return ret; } @@ -703,7 +694,7 @@ static int ext4_alloc_branch(handle_t *handle, struct inode *inode, branch[n].p = (__le32 *) bh->b_data + offsets[n]; branch[n].key = cpu_to_le32(new_blocks[n]); *branch[n].p = branch[n].key; - if ( n == indirect_blks) { + if (n == indirect_blks) { current_block = new_blocks[n]; /* * End of chain, update the last new metablock of @@ -730,7 +721,7 @@ failed: BUFFER_TRACE(branch[i].bh, "call jbd2_journal_forget"); ext4_journal_forget(handle, branch[i].bh); } - for (i = 0; i <indirect_blks; i++) + for (i = 0; i < indirect_blks; i++) ext4_free_blocks(handle, inode, new_blocks[i], 1, 0); ext4_free_blocks(handle, inode, new_blocks[i], num, 0); @@ -757,10 +748,8 @@ static int ext4_splice_branch(handle_t *handle, struct inode *inode, { int i; int err = 0; - struct ext4_block_alloc_info *block_i; ext4_fsblk_t current_block; - block_i = EXT4_I(inode)->i_block_alloc_info; /* * If we're splicing into a [td]indirect block (as opposed to the * inode) then we need to get write access to the [td]indirect block @@ -783,18 +772,7 @@ static int ext4_splice_branch(handle_t *handle, struct inode *inode, if (num == 0 && blks > 1) { current_block = le32_to_cpu(where->key) + 1; for (i = 1; i < blks; i++) - *(where->p + i ) = cpu_to_le32(current_block++); - } - - /* - * update the most recently allocated logical & physical block - * in i_block_alloc_info, to assist find the proper goal block for next - * allocation - */ - if (block_i) { - block_i->last_alloc_logical_block = block + blks - 1; - block_i->last_alloc_physical_block = - le32_to_cpu(where[num].key) + blks - 1; + *(where->p + i) = cpu_to_le32(current_block++); } /* We are done with atomic stuff, now do the rest of housekeeping */ @@ -914,12 +892,8 @@ int ext4_get_blocks_handle(handle_t *handle, struct inode *inode, goto cleanup; /* - * Okay, we need to do block allocation. Lazily initialize the block - * allocation info here if necessary + * Okay, we need to do block allocation. */ - if (S_ISREG(inode->i_mode) && (!ei->i_block_alloc_info)) - ext4_init_block_alloc_info(inode); - goal = ext4_find_goal(inode, iblock, partial); /* the number of blocks need to allocate for [d,t]indirect blocks */ @@ -1030,19 +1004,20 @@ static void ext4_da_update_reserve_space(struct inode *inode, int used) BUG_ON(mdb > EXT4_I(inode)->i_reserved_meta_blocks); mdb_free = EXT4_I(inode)->i_reserved_meta_blocks - mdb; - /* Account for allocated meta_blocks */ - mdb_free -= EXT4_I(inode)->i_allocated_meta_blocks; + if (mdb_free) { + /* Account for allocated meta_blocks */ + mdb_free -= EXT4_I(inode)->i_allocated_meta_blocks; - /* update fs free blocks counter for truncate case */ - percpu_counter_add(&sbi->s_freeblocks_counter, mdb_free); + /* update fs dirty blocks counter */ + percpu_counter_sub(&sbi->s_dirtyblocks_counter, mdb_free); + EXT4_I(inode)->i_allocated_meta_blocks = 0; + EXT4_I(inode)->i_reserved_meta_blocks = mdb; + } /* update per-inode reservations */ BUG_ON(used > EXT4_I(inode)->i_reserved_data_blocks); EXT4_I(inode)->i_reserved_data_blocks -= used; - BUG_ON(mdb > EXT4_I(inode)->i_reserved_meta_blocks); - EXT4_I(inode)->i_reserved_meta_blocks = mdb; - EXT4_I(inode)->i_allocated_meta_blocks = 0; spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); } @@ -1160,8 +1135,8 @@ int ext4_get_blocks_wrap(handle_t *handle, struct inode *inode, sector_t block, /* Maximum number of blocks we map for direct IO at once. */ #define DIO_MAX_BLOCKS 4096 -static int ext4_get_block(struct inode *inode, sector_t iblock, - struct buffer_head *bh_result, int create) +int ext4_get_block(struct inode *inode, sector_t iblock, + struct buffer_head *bh_result, int create) { handle_t *handle = ext4_journal_current_handle(); int ret = 0, started = 0; @@ -1241,7 +1216,7 @@ struct buffer_head *ext4_getblk(handle_t *handle, struct inode *inode, BUFFER_TRACE(bh, "call get_create_access"); fatal = ext4_journal_get_create_access(handle, bh); if (!fatal && !buffer_uptodate(bh)) { - memset(bh->b_data,0,inode->i_sb->s_blocksize); + memset(bh->b_data, 0, inode->i_sb->s_blocksize); set_buffer_uptodate(bh); } unlock_buffer(bh); @@ -1266,7 +1241,7 @@ err: struct buffer_head *ext4_bread(handle_t *handle, struct inode *inode, ext4_lblk_t block, int create, int *err) { - struct buffer_head * bh; + struct buffer_head *bh; bh = ext4_getblk(handle, inode, block, create, err); if (!bh) @@ -1282,13 +1257,13 @@ struct buffer_head *ext4_bread(handle_t *handle, struct inode *inode, return NULL; } -static int walk_page_buffers( handle_t *handle, - struct buffer_head *head, - unsigned from, - unsigned to, - int *partial, - int (*fn)( handle_t *handle, - struct buffer_head *bh)) +static int walk_page_buffers(handle_t *handle, + struct buffer_head *head, + unsigned from, + unsigned to, + int *partial, + int (*fn)(handle_t *handle, + struct buffer_head *bh)) { struct buffer_head *bh; unsigned block_start, block_end; @@ -1296,9 +1271,9 @@ static int walk_page_buffers( handle_t *handle, int err, ret = 0; struct buffer_head *next; - for ( bh = head, block_start = 0; - ret == 0 && (bh != head || !block_start); - block_start = block_end, bh = next) + for (bh = head, block_start = 0; + ret == 0 && (bh != head || !block_start); + block_start = block_end, bh = next) { next = bh->b_this_page; block_end = block_start + blocksize; @@ -1351,23 +1326,23 @@ static int ext4_write_begin(struct file *file, struct address_space *mapping, loff_t pos, unsigned len, unsigned flags, struct page **pagep, void **fsdata) { - struct inode *inode = mapping->host; + struct inode *inode = mapping->host; int ret, needed_blocks = ext4_writepage_trans_blocks(inode); handle_t *handle; int retries = 0; - struct page *page; + struct page *page; pgoff_t index; - unsigned from, to; + unsigned from, to; index = pos >> PAGE_CACHE_SHIFT; - from = pos & (PAGE_CACHE_SIZE - 1); - to = from + len; + from = pos & (PAGE_CACHE_SIZE - 1); + to = from + len; retry: - handle = ext4_journal_start(inode, needed_blocks); - if (IS_ERR(handle)) { - ret = PTR_ERR(handle); - goto out; + handle = ext4_journal_start(inode, needed_blocks); + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + goto out; } page = __grab_cache_page(mapping, index); @@ -1387,9 +1362,16 @@ retry: } if (ret) { - unlock_page(page); + unlock_page(page); ext4_journal_stop(handle); - page_cache_release(page); + page_cache_release(page); + /* + * block_write_begin may have instantiated a few blocks + * outside i_size. Trim these off again. Don't need + * i_size_read because we hold i_mutex. + */ + if (pos + len > inode->i_size) + vmtruncate(inode, inode->i_size); } if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries)) @@ -1426,16 +1408,18 @@ static int ext4_ordered_write_end(struct file *file, ret = ext4_jbd2_file_inode(handle, inode); if (ret == 0) { - /* - * generic_write_end() will run mark_inode_dirty() if i_size - * changes. So let's piggyback the i_disksize mark_inode_dirty - * into that. - */ loff_t new_i_size; new_i_size = pos + copied; - if (new_i_size > EXT4_I(inode)->i_disksize) - EXT4_I(inode)->i_disksize = new_i_size; + if (new_i_size > EXT4_I(inode)->i_disksize) { + ext4_update_i_disksize(inode, new_i_size); + /* We need to mark inode dirty even if + * new_i_size is less that inode->i_size + * bu greater than i_disksize.(hint delalloc) + */ + ext4_mark_inode_dirty(handle, inode); + } + ret2 = generic_write_end(file, mapping, pos, len, copied, page, fsdata); copied = ret2; @@ -1460,8 +1444,14 @@ static int ext4_writeback_write_end(struct file *file, loff_t new_i_size; new_i_size = pos + copied; - if (new_i_size > EXT4_I(inode)->i_disksize) - EXT4_I(inode)->i_disksize = new_i_size; + if (new_i_size > EXT4_I(inode)->i_disksize) { + ext4_update_i_disksize(inode, new_i_size); + /* We need to mark inode dirty even if + * new_i_size is less that inode->i_size + * bu greater than i_disksize.(hint delalloc) + */ + ext4_mark_inode_dirty(handle, inode); + } ret2 = generic_write_end(file, mapping, pos, len, copied, page, fsdata); @@ -1486,6 +1476,7 @@ static int ext4_journalled_write_end(struct file *file, int ret = 0, ret2; int partial = 0; unsigned from, to; + loff_t new_i_size; from = pos & (PAGE_CACHE_SIZE - 1); to = from + len; @@ -1500,11 +1491,12 @@ static int ext4_journalled_write_end(struct file *file, to, &partial, write_end_fn); if (!partial) SetPageUptodate(page); - if (pos+copied > inode->i_size) + new_i_size = pos + copied; + if (new_i_size > inode->i_size) i_size_write(inode, pos+copied); EXT4_I(inode)->i_state |= EXT4_STATE_JDATA; - if (inode->i_size > EXT4_I(inode)->i_disksize) { - EXT4_I(inode)->i_disksize = inode->i_size; + if (new_i_size > EXT4_I(inode)->i_disksize) { + ext4_update_i_disksize(inode, new_i_size); ret2 = ext4_mark_inode_dirty(handle, inode); if (!ret) ret = ret2; @@ -1521,6 +1513,7 @@ static int ext4_journalled_write_end(struct file *file, static int ext4_da_reserve_space(struct inode *inode, int nrblocks) { + int retries = 0; struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); unsigned long md_needed, mdblocks, total = 0; @@ -1529,6 +1522,7 @@ static int ext4_da_reserve_space(struct inode *inode, int nrblocks) * in order to allocate nrblocks * worse case is one extent per block */ +repeat: spin_lock(&EXT4_I(inode)->i_block_reservation_lock); total = EXT4_I(inode)->i_reserved_data_blocks + nrblocks; mdblocks = ext4_calc_metadata_amount(inode, total); @@ -1537,13 +1531,14 @@ static int ext4_da_reserve_space(struct inode *inode, int nrblocks) md_needed = mdblocks - EXT4_I(inode)->i_reserved_meta_blocks; total = md_needed + nrblocks; - if (ext4_has_free_blocks(sbi, total) < total) { + if (ext4_claim_free_blocks(sbi, total)) { spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); + if (ext4_should_retry_alloc(inode->i_sb, &retries)) { + yield(); + goto repeat; + } return -ENOSPC; } - /* reduce fs free blocks counter */ - percpu_counter_sub(&sbi->s_freeblocks_counter, total); - EXT4_I(inode)->i_reserved_data_blocks += nrblocks; EXT4_I(inode)->i_reserved_meta_blocks = mdblocks; @@ -1585,8 +1580,8 @@ static void ext4_da_release_space(struct inode *inode, int to_free) release = to_free + mdb_free; - /* update fs free blocks counter for truncate case */ - percpu_counter_add(&sbi->s_freeblocks_counter, release); + /* update fs dirty blocks counter for truncate case */ + percpu_counter_sub(&sbi->s_dirtyblocks_counter, release); /* update per-inode reservations */ BUG_ON(to_free > EXT4_I(inode)->i_reserved_data_blocks); @@ -1630,6 +1625,7 @@ struct mpage_da_data { struct writeback_control *wbc; int io_done; long pages_written; + int retval; }; /* @@ -1783,6 +1779,57 @@ static inline void __unmap_underlying_blocks(struct inode *inode, unmap_underlying_metadata(bdev, bh->b_blocknr + i); } +static void ext4_da_block_invalidatepages(struct mpage_da_data *mpd, + sector_t logical, long blk_cnt) +{ + int nr_pages, i; + pgoff_t index, end; + struct pagevec pvec; + struct inode *inode = mpd->inode; + struct address_space *mapping = inode->i_mapping; + + index = logical >> (PAGE_CACHE_SHIFT - inode->i_blkbits); + end = (logical + blk_cnt - 1) >> + (PAGE_CACHE_SHIFT - inode->i_blkbits); + while (index <= end) { + nr_pages = pagevec_lookup(&pvec, mapping, index, PAGEVEC_SIZE); + if (nr_pages == 0) + break; + for (i = 0; i < nr_pages; i++) { + struct page *page = pvec.pages[i]; + index = page->index; + if (index > end) + break; + index++; + + BUG_ON(!PageLocked(page)); + BUG_ON(PageWriteback(page)); + block_invalidatepage(page, 0); + ClearPageUptodate(page); + unlock_page(page); + } + } + return; +} + +static void ext4_print_free_blocks(struct inode *inode) +{ + struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); + printk(KERN_EMERG "Total free blocks count %lld\n", + ext4_count_free_blocks(inode->i_sb)); + printk(KERN_EMERG "Free/Dirty block details\n"); + printk(KERN_EMERG "free_blocks=%lld\n", + percpu_counter_sum(&sbi->s_freeblocks_counter)); + printk(KERN_EMERG "dirty_blocks=%lld\n", + percpu_counter_sum(&sbi->s_dirtyblocks_counter)); + printk(KERN_EMERG "Block reservation details\n"); + printk(KERN_EMERG "i_reserved_data_blocks=%lu\n", + EXT4_I(inode)->i_reserved_data_blocks); + printk(KERN_EMERG "i_reserved_meta_blocks=%lu\n", + EXT4_I(inode)->i_reserved_meta_blocks); + return; +} + /* * mpage_da_map_blocks - go through given space * @@ -1792,32 +1839,69 @@ static inline void __unmap_underlying_blocks(struct inode *inode, * The function skips space we know is already mapped to disk blocks. * */ -static void mpage_da_map_blocks(struct mpage_da_data *mpd) +static int mpage_da_map_blocks(struct mpage_da_data *mpd) { int err = 0; - struct buffer_head *lbh = &mpd->lbh; - sector_t next = lbh->b_blocknr; struct buffer_head new; + struct buffer_head *lbh = &mpd->lbh; + sector_t next; /* * We consider only non-mapped and non-allocated blocks */ if (buffer_mapped(lbh) && !buffer_delay(lbh)) - return; - + return 0; new.b_state = lbh->b_state; new.b_blocknr = 0; new.b_size = lbh->b_size; - + next = lbh->b_blocknr; /* * If we didn't accumulate anything * to write simply return */ if (!new.b_size) - return; + return 0; err = mpd->get_block(mpd->inode, next, &new, 1); - if (err) - return; + if (err) { + + /* If get block returns with error + * we simply return. Later writepage + * will redirty the page and writepages + * will find the dirty page again + */ + if (err == -EAGAIN) + return 0; + + if (err == -ENOSPC && + ext4_count_free_blocks(mpd->inode->i_sb)) { + mpd->retval = err; + return 0; + } + + /* + * get block failure will cause us + * to loop in writepages. Because + * a_ops->writepage won't be able to + * make progress. The page will be redirtied + * by writepage and writepages will again + * try to write the same. + */ + printk(KERN_EMERG "%s block allocation failed for inode %lu " + "at logical offset %llu with max blocks " + "%zd with error %d\n", + __func__, mpd->inode->i_ino, + (unsigned long long)next, + lbh->b_size >> mpd->inode->i_blkbits, err); + printk(KERN_EMERG "This should not happen.!! " + "Data will be lost\n"); + if (err == -ENOSPC) { + ext4_print_free_blocks(mpd->inode); + } + /* invlaidate all the pages */ + ext4_da_block_invalidatepages(mpd, next, + lbh->b_size >> mpd->inode->i_blkbits); + return err; + } BUG_ON(new.b_size == 0); if (buffer_new(&new)) @@ -1830,7 +1914,7 @@ static void mpage_da_map_blocks(struct mpage_da_data *mpd) if (buffer_delay(lbh) || buffer_unwritten(lbh)) mpage_put_bnr_to_bhs(mpd, next, &new); - return; + return 0; } #define BH_FLAGS ((1 << BH_Uptodate) | (1 << BH_Mapped) | \ @@ -1899,8 +1983,8 @@ flush_it: * We couldn't merge the block to our extent, so we * need to flush current extent and start new one */ - mpage_da_map_blocks(mpd); - mpage_da_submit_io(mpd); + if (mpage_da_map_blocks(mpd) == 0) + mpage_da_submit_io(mpd); mpd->io_done = 1; return; } @@ -1942,8 +2026,8 @@ static int __mpage_da_writepage(struct page *page, * and start IO on them using writepage() */ if (mpd->next_page != mpd->first_page) { - mpage_da_map_blocks(mpd); - mpage_da_submit_io(mpd); + if (mpage_da_map_blocks(mpd) == 0) + mpage_da_submit_io(mpd); /* * skip rest of the page in the page_vec */ @@ -2018,39 +2102,36 @@ static int __mpage_da_writepage(struct page *page, */ static int mpage_da_writepages(struct address_space *mapping, struct writeback_control *wbc, - get_block_t get_block) + struct mpage_da_data *mpd) { - struct mpage_da_data mpd; long to_write; int ret; - if (!get_block) + if (!mpd->get_block) return generic_writepages(mapping, wbc); - mpd.wbc = wbc; - mpd.inode = mapping->host; - mpd.lbh.b_size = 0; - mpd.lbh.b_state = 0; - mpd.lbh.b_blocknr = 0; - mpd.first_page = 0; - mpd.next_page = 0; - mpd.get_block = get_block; - mpd.io_done = 0; - mpd.pages_written = 0; + mpd->lbh.b_size = 0; + mpd->lbh.b_state = 0; + mpd->lbh.b_blocknr = 0; + mpd->first_page = 0; + mpd->next_page = 0; + mpd->io_done = 0; + mpd->pages_written = 0; + mpd->retval = 0; to_write = wbc->nr_to_write; - ret = write_cache_pages(mapping, wbc, __mpage_da_writepage, &mpd); + ret = write_cache_pages(mapping, wbc, __mpage_da_writepage, mpd); /* * Handle last extent of pages */ - if (!mpd.io_done && mpd.next_page != mpd.first_page) { - mpage_da_map_blocks(&mpd); - mpage_da_submit_io(&mpd); + if (!mpd->io_done && mpd->next_page != mpd->first_page) { + if (mpage_da_map_blocks(mpd) == 0) + mpage_da_submit_io(mpd); } - wbc->nr_to_write = to_write - mpd.pages_written; + wbc->nr_to_write = to_write - mpd->pages_written; return ret; } @@ -2103,18 +2184,24 @@ static int ext4_da_get_block_write(struct inode *inode, sector_t iblock, handle_t *handle = NULL; handle = ext4_journal_current_handle(); - if (!handle) { - ret = ext4_get_blocks_wrap(handle, inode, iblock, max_blocks, - bh_result, 0, 0, 0); - BUG_ON(!ret); - } else { - ret = ext4_get_blocks_wrap(handle, inode, iblock, max_blocks, - bh_result, create, 0, EXT4_DELALLOC_RSVED); - } - + BUG_ON(!handle); + ret = ext4_get_blocks_wrap(handle, inode, iblock, max_blocks, + bh_result, create, 0, EXT4_DELALLOC_RSVED); if (ret > 0) { + bh_result->b_size = (ret << inode->i_blkbits); + if (ext4_should_order_data(inode)) { + int retval; + retval = ext4_jbd2_file_inode(handle, inode); + if (retval) + /* + * Failed to add inode for ordered + * mode. Don't update file size + */ + return retval; + } + /* * Update on-disk size along with block allocation * we don't use 'extend_disksize' as size may change @@ -2124,18 +2211,9 @@ static int ext4_da_get_block_write(struct inode *inode, sector_t iblock, if (disksize > i_size_read(inode)) disksize = i_size_read(inode); if (disksize > EXT4_I(inode)->i_disksize) { - /* - * XXX: replace with spinlock if seen contended -bzzz - */ - down_write(&EXT4_I(inode)->i_data_sem); - if (disksize > EXT4_I(inode)->i_disksize) - EXT4_I(inode)->i_disksize = disksize; - up_write(&EXT4_I(inode)->i_data_sem); - - if (EXT4_I(inode)->i_disksize == disksize) { - ret = ext4_mark_inode_dirty(handle, inode); - return ret; - } + ext4_update_i_disksize(inode, disksize); + ret = ext4_mark_inode_dirty(handle, inode); + return ret; } ret = 0; } @@ -2284,6 +2362,7 @@ static int ext4_da_writepages(struct address_space *mapping, { handle_t *handle = NULL; loff_t range_start = 0; + struct mpage_da_data mpd; struct inode *inode = mapping->host; int needed_blocks, ret = 0, nr_to_writebump = 0; long to_write, pages_skipped = 0; @@ -2317,6 +2396,9 @@ static int ext4_da_writepages(struct address_space *mapping, range_start = wbc->range_start; pages_skipped = wbc->pages_skipped; + mpd.wbc = wbc; + mpd.inode = mapping->host; + restart_loop: to_write = wbc->nr_to_write; while (!ret && to_write > 0) { @@ -2340,23 +2422,17 @@ restart_loop: dump_stack(); goto out_writepages; } - if (ext4_should_order_data(inode)) { - /* - * With ordered mode we need to add - * the inode to the journal handl - * when we do block allocation. - */ - ret = ext4_jbd2_file_inode(handle, inode); - if (ret) { - ext4_journal_stop(handle); - goto out_writepages; - } - } - to_write -= wbc->nr_to_write; - ret = mpage_da_writepages(mapping, wbc, - ext4_da_get_block_write); + + mpd.get_block = ext4_da_get_block_write; + ret = mpage_da_writepages(mapping, wbc, &mpd); + ext4_journal_stop(handle); + + if (mpd.retval == -ENOSPC) + jbd2_journal_force_commit_nested(sbi->s_journal); + + /* reset the retry count */ if (ret == MPAGE_DA_EXTENT_TAIL) { /* * got one extent now try with @@ -2391,6 +2467,33 @@ out_writepages: return ret; } +#define FALL_BACK_TO_NONDELALLOC 1 +static int ext4_nonda_switch(struct super_block *sb) +{ + s64 free_blocks, dirty_blocks; + struct ext4_sb_info *sbi = EXT4_SB(sb); + + /* + * switch to non delalloc mode if we are running low + * on free block. The free block accounting via percpu + * counters can get slightly wrong with FBC_BATCH getting + * accumulated on each CPU without updating global counters + * Delalloc need an accurate free block accounting. So switch + * to non delalloc when we are near to error range. + */ + free_blocks = percpu_counter_read_positive(&sbi->s_freeblocks_counter); + dirty_blocks = percpu_counter_read_positive(&sbi->s_dirtyblocks_counter); + if (2 * free_blocks < 3 * dirty_blocks || + free_blocks < (dirty_blocks + EXT4_FREEBLOCKS_WATERMARK)) { + /* + * free block count is less that 150% of dirty blocks + * or free blocks is less that watermark + */ + return 1; + } + return 0; +} + static int ext4_da_write_begin(struct file *file, struct address_space *mapping, loff_t pos, unsigned len, unsigned flags, struct page **pagep, void **fsdata) @@ -2406,6 +2509,12 @@ static int ext4_da_write_begin(struct file *file, struct address_space *mapping, from = pos & (PAGE_CACHE_SIZE - 1); to = from + len; + if (ext4_nonda_switch(inode->i_sb)) { + *fsdata = (void *)FALL_BACK_TO_NONDELALLOC; + return ext4_write_begin(file, mapping, pos, + len, flags, pagep, fsdata); + } + *fsdata = (void *)0; retry: /* * With delayed allocation, we don't log the i_disksize update @@ -2433,6 +2542,13 @@ retry: unlock_page(page); ext4_journal_stop(handle); page_cache_release(page); + /* + * block_write_begin may have instantiated a few blocks + * outside i_size. Trim these off again. Don't need + * i_size_read because we hold i_mutex. + */ + if (pos + len > inode->i_size) + vmtruncate(inode, inode->i_size); } if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries)) @@ -2456,7 +2572,7 @@ static int ext4_da_should_update_i_disksize(struct page *page, bh = page_buffers(page); idx = offset >> inode->i_blkbits; - for (i=0; i < idx; i++) + for (i = 0; i < idx; i++) bh = bh->b_this_page; if (!buffer_mapped(bh) || (buffer_delay(bh))) @@ -2474,9 +2590,22 @@ static int ext4_da_write_end(struct file *file, handle_t *handle = ext4_journal_current_handle(); loff_t new_i_size; unsigned long start, end; + int write_mode = (int)(unsigned long)fsdata; + + if (write_mode == FALL_BACK_TO_NONDELALLOC) { + if (ext4_should_order_data(inode)) { + return ext4_ordered_write_end(file, mapping, pos, + len, copied, page, fsdata); + } else if (ext4_should_writeback_data(inode)) { + return ext4_writeback_write_end(file, mapping, pos, + len, copied, page, fsdata); + } else { + BUG(); + } + } start = pos & (PAGE_CACHE_SIZE - 1); - end = start + copied -1; + end = start + copied - 1; /* * generic_write_end() will run mark_inode_dirty() if i_size @@ -2500,6 +2629,11 @@ static int ext4_da_write_end(struct file *file, EXT4_I(inode)->i_disksize = new_i_size; } up_write(&EXT4_I(inode)->i_data_sem); + /* We need to mark inode dirty even if + * new_i_size is less that inode->i_size + * bu greater than i_disksize.(hint delalloc) + */ + ext4_mark_inode_dirty(handle, inode); } } ret2 = generic_write_end(file, mapping, pos, len, copied, @@ -2591,7 +2725,7 @@ static sector_t ext4_bmap(struct address_space *mapping, sector_t block) return 0; } - return generic_block_bmap(mapping,block,ext4_get_block); + return generic_block_bmap(mapping, block, ext4_get_block); } static int bget_one(handle_t *handle, struct buffer_head *bh) @@ -3197,7 +3331,7 @@ static Indirect *ext4_find_shared(struct inode *inode, int depth, if (!partial->key && *partial->p) /* Writer: end */ goto no_top; - for (p=partial; p>chain && all_zeroes((__le32*)p->bh->b_data,p->p); p--) + for (p = partial; (p > chain) && all_zeroes((__le32 *) p->bh->b_data, p->p); p--) ; /* * OK, we've found the last block that must survive. The rest of our @@ -3216,7 +3350,7 @@ static Indirect *ext4_find_shared(struct inode *inode, int depth, } /* Writer: end */ - while(partial > p) { + while (partial > p) { brelse(partial->bh); partial--; } @@ -3408,9 +3542,9 @@ static void ext4_free_branches(handle_t *handle, struct inode *inode, /* This zaps the entire block. Bottom up. */ BUFFER_TRACE(bh, "free child branches"); ext4_free_branches(handle, inode, bh, - (__le32*)bh->b_data, - (__le32*)bh->b_data + addr_per_block, - depth); + (__le32 *) bh->b_data, + (__le32 *) bh->b_data + addr_per_block, + depth); /* * We've probably journalled the indirect block several @@ -3578,7 +3712,7 @@ void ext4_truncate(struct inode *inode) */ down_write(&ei->i_data_sem); - ext4_discard_reservation(inode); + ext4_discard_preallocations(inode); /* * The orphan list entry will now protect us from any crash which @@ -3673,41 +3807,6 @@ out_stop: ext4_journal_stop(handle); } -static ext4_fsblk_t ext4_get_inode_block(struct super_block *sb, - unsigned long ino, struct ext4_iloc *iloc) -{ - ext4_group_t block_group; - unsigned long offset; - ext4_fsblk_t block; - struct ext4_group_desc *gdp; - - if (!ext4_valid_inum(sb, ino)) { - /* - * This error is already checked for in namei.c unless we are - * looking at an NFS filehandle, in which case no error - * report is needed - */ - return 0; - } - - block_group = (ino - 1) / EXT4_INODES_PER_GROUP(sb); - gdp = ext4_get_group_desc(sb, block_group, NULL); - if (!gdp) - return 0; - - /* - * Figure out the offset within the block group inode table - */ - offset = ((ino - 1) % EXT4_INODES_PER_GROUP(sb)) * - EXT4_INODE_SIZE(sb); - block = ext4_inode_table(sb, gdp) + - (offset >> EXT4_BLOCK_SIZE_BITS(sb)); - - iloc->block_group = block_group; - iloc->offset = offset & (EXT4_BLOCK_SIZE(sb) - 1); - return block; -} - /* * ext4_get_inode_loc returns with an extra refcount against the inode's * underlying buffer_head on success. If 'in_mem' is true, we have all @@ -3717,19 +3816,35 @@ static ext4_fsblk_t ext4_get_inode_block(struct super_block *sb, static int __ext4_get_inode_loc(struct inode *inode, struct ext4_iloc *iloc, int in_mem) { - ext4_fsblk_t block; - struct buffer_head *bh; + struct ext4_group_desc *gdp; + struct buffer_head *bh; + struct super_block *sb = inode->i_sb; + ext4_fsblk_t block; + int inodes_per_block, inode_offset; + + iloc->bh = 0; + if (!ext4_valid_inum(sb, inode->i_ino)) + return -EIO; - block = ext4_get_inode_block(inode->i_sb, inode->i_ino, iloc); - if (!block) + iloc->block_group = (inode->i_ino - 1) / EXT4_INODES_PER_GROUP(sb); + gdp = ext4_get_group_desc(sb, iloc->block_group, NULL); + if (!gdp) return -EIO; - bh = sb_getblk(inode->i_sb, block); + /* + * Figure out the offset within the block group inode table + */ + inodes_per_block = (EXT4_BLOCK_SIZE(sb) / EXT4_INODE_SIZE(sb)); + inode_offset = ((inode->i_ino - 1) % + EXT4_INODES_PER_GROUP(sb)); + block = ext4_inode_table(sb, gdp) + (inode_offset / inodes_per_block); + iloc->offset = (inode_offset % inodes_per_block) * EXT4_INODE_SIZE(sb); + + bh = sb_getblk(sb, block); if (!bh) { - ext4_error (inode->i_sb, "ext4_get_inode_loc", - "unable to read inode block - " - "inode=%lu, block=%llu", - inode->i_ino, block); + ext4_error(sb, "ext4_get_inode_loc", "unable to read " + "inode block - inode=%lu, block=%llu", + inode->i_ino, block); return -EIO; } if (!buffer_uptodate(bh)) { @@ -3757,28 +3872,12 @@ static int __ext4_get_inode_loc(struct inode *inode, */ if (in_mem) { struct buffer_head *bitmap_bh; - struct ext4_group_desc *desc; - int inodes_per_buffer; - int inode_offset, i; - ext4_group_t block_group; - int start; - - block_group = (inode->i_ino - 1) / - EXT4_INODES_PER_GROUP(inode->i_sb); - inodes_per_buffer = bh->b_size / - EXT4_INODE_SIZE(inode->i_sb); - inode_offset = ((inode->i_ino - 1) % - EXT4_INODES_PER_GROUP(inode->i_sb)); - start = inode_offset & ~(inodes_per_buffer - 1); + int i, start; - /* Is the inode bitmap in cache? */ - desc = ext4_get_group_desc(inode->i_sb, - block_group, NULL); - if (!desc) - goto make_io; + start = inode_offset & ~(inodes_per_block - 1); - bitmap_bh = sb_getblk(inode->i_sb, - ext4_inode_bitmap(inode->i_sb, desc)); + /* Is the inode bitmap in cache? */ + bitmap_bh = sb_getblk(sb, ext4_inode_bitmap(sb, gdp)); if (!bitmap_bh) goto make_io; @@ -3791,14 +3890,14 @@ static int __ext4_get_inode_loc(struct inode *inode, brelse(bitmap_bh); goto make_io; } - for (i = start; i < start + inodes_per_buffer; i++) { + for (i = start; i < start + inodes_per_block; i++) { if (i == inode_offset) continue; if (ext4_test_bit(i, bitmap_bh->b_data)) break; } brelse(bitmap_bh); - if (i == start + inodes_per_buffer) { + if (i == start + inodes_per_block) { /* all other inodes are free, so skip I/O */ memset(bh->b_data, 0, bh->b_size); set_buffer_uptodate(bh); @@ -3809,6 +3908,36 @@ static int __ext4_get_inode_loc(struct inode *inode, make_io: /* + * If we need to do any I/O, try to pre-readahead extra + * blocks from the inode table. + */ + if (EXT4_SB(sb)->s_inode_readahead_blks) { + ext4_fsblk_t b, end, table; + unsigned num; + + table = ext4_inode_table(sb, gdp); + /* Make sure s_inode_readahead_blks is a power of 2 */ + while (EXT4_SB(sb)->s_inode_readahead_blks & + (EXT4_SB(sb)->s_inode_readahead_blks-1)) + EXT4_SB(sb)->s_inode_readahead_blks = + (EXT4_SB(sb)->s_inode_readahead_blks & + (EXT4_SB(sb)->s_inode_readahead_blks-1)); + b = block & ~(EXT4_SB(sb)->s_inode_readahead_blks-1); + if (table > b) + b = table; + end = b + EXT4_SB(sb)->s_inode_readahead_blks; + num = EXT4_INODES_PER_GROUP(sb); + if (EXT4_HAS_RO_COMPAT_FEATURE(sb, + EXT4_FEATURE_RO_COMPAT_GDT_CSUM)) + num -= le16_to_cpu(gdp->bg_itable_unused); + table += num / inodes_per_block; + if (end > table) + end = table; + while (b <= end) + sb_breadahead(sb, b++); + } + + /* * There are other valid inodes in the buffer, this inode * has in-inode xattrs, or we don't have this inode in memory. * Read the block from disk. @@ -3818,10 +3947,9 @@ make_io: submit_bh(READ_META, bh); wait_on_buffer(bh); if (!buffer_uptodate(bh)) { - ext4_error(inode->i_sb, "ext4_get_inode_loc", - "unable to read inode block - " - "inode=%lu, block=%llu", - inode->i_ino, block); + ext4_error(sb, __func__, + "unable to read inode block - inode=%lu, " + "block=%llu", inode->i_ino, block); brelse(bh); return -EIO; } @@ -3913,11 +4041,10 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino) return inode; ei = EXT4_I(inode); -#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL +#ifdef CONFIG_EXT4_FS_POSIX_ACL ei->i_acl = EXT4_ACL_NOT_CACHED; ei->i_default_acl = EXT4_ACL_NOT_CACHED; #endif - ei->i_block_alloc_info = NULL; ret = __ext4_get_inode_loc(inode, &iloc, 0); if (ret < 0) @@ -3927,7 +4054,7 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino) inode->i_mode = le16_to_cpu(raw_inode->i_mode); inode->i_uid = (uid_t)le16_to_cpu(raw_inode->i_uid_low); inode->i_gid = (gid_t)le16_to_cpu(raw_inode->i_gid_low); - if(!(test_opt (inode->i_sb, NO_UID32))) { + if (!(test_opt(inode->i_sb, NO_UID32))) { inode->i_uid |= le16_to_cpu(raw_inode->i_uid_high) << 16; inode->i_gid |= le16_to_cpu(raw_inode->i_gid_high) << 16; } @@ -3945,7 +4072,7 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino) if (inode->i_mode == 0 || !(EXT4_SB(inode->i_sb)->s_mount_state & EXT4_ORPHAN_FS)) { /* this inode is deleted */ - brelse (bh); + brelse(bh); ret = -ESTALE; goto bad_inode; } @@ -3978,7 +4105,7 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino) ei->i_extra_isize = le16_to_cpu(raw_inode->i_extra_isize); if (EXT4_GOOD_OLD_INODE_SIZE + ei->i_extra_isize > EXT4_INODE_SIZE(inode->i_sb)) { - brelse (bh); + brelse(bh); ret = -EIO; goto bad_inode; } @@ -4031,7 +4158,7 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino) init_special_inode(inode, inode->i_mode, new_decode_dev(le32_to_cpu(raw_inode->i_block[1]))); } - brelse (iloc.bh); + brelse(iloc.bh); ext4_set_inode_flags(inode); unlock_new_inode(inode); return inode; @@ -4113,14 +4240,14 @@ static int ext4_do_update_inode(handle_t *handle, ext4_get_inode_flags(ei); raw_inode->i_mode = cpu_to_le16(inode->i_mode); - if(!(test_opt(inode->i_sb, NO_UID32))) { + if (!(test_opt(inode->i_sb, NO_UID32))) { raw_inode->i_uid_low = cpu_to_le16(low_16_bits(inode->i_uid)); raw_inode->i_gid_low = cpu_to_le16(low_16_bits(inode->i_gid)); /* * Fix up interoperability with old kernels. Otherwise, old inodes get * re-used with the upper 16 bits of the uid/gid intact */ - if(!ei->i_dtime) { + if (!ei->i_dtime) { raw_inode->i_uid_high = cpu_to_le16(high_16_bits(inode->i_uid)); raw_inode->i_gid_high = @@ -4208,7 +4335,7 @@ static int ext4_do_update_inode(handle_t *handle, ei->i_state &= ~EXT4_STATE_NEW; out_brelse: - brelse (bh); + brelse(bh); ext4_std_error(inode->i_sb, err); return err; } @@ -4811,6 +4938,7 @@ int ext4_page_mkwrite(struct vm_area_struct *vma, struct page *page) loff_t size; unsigned long len; int ret = -EINVAL; + void *fsdata; struct file *file = vma->vm_file; struct inode *inode = file->f_path.dentry->d_inode; struct address_space *mapping = inode->i_mapping; @@ -4849,11 +4977,11 @@ int ext4_page_mkwrite(struct vm_area_struct *vma, struct page *page) * on the same page though */ ret = mapping->a_ops->write_begin(file, mapping, page_offset(page), - len, AOP_FLAG_UNINTERRUPTIBLE, &page, NULL); + len, AOP_FLAG_UNINTERRUPTIBLE, &page, &fsdata); if (ret < 0) goto out_unlock; ret = mapping->a_ops->write_end(file, mapping, page_offset(page), - len, len, page, NULL); + len, len, page, fsdata); if (ret < 0) goto out_unlock; ret = 0; diff --git a/fs/ext4/ioctl.c b/fs/ext4/ioctl.c index 7a6c2f1faba..dc99b4776d5 100644 --- a/fs/ext4/ioctl.c +++ b/fs/ext4/ioctl.c @@ -23,9 +23,8 @@ long ext4_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) struct inode *inode = filp->f_dentry->d_inode; struct ext4_inode_info *ei = EXT4_I(inode); unsigned int flags; - unsigned short rsv_window_size; - ext4_debug ("cmd = %u, arg = %lu\n", cmd, arg); + ext4_debug("cmd = %u, arg = %lu\n", cmd, arg); switch (cmd) { case EXT4_IOC_GETFLAGS: @@ -34,7 +33,7 @@ long ext4_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) return put_user(flags, (int __user *) arg); case EXT4_IOC_SETFLAGS: { handle_t *handle = NULL; - int err; + int err, migrate = 0; struct ext4_iloc iloc; unsigned int oldflags; unsigned int jflag; @@ -82,6 +81,17 @@ long ext4_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) if (!capable(CAP_SYS_RESOURCE)) goto flags_out; } + if (oldflags & EXT4_EXTENTS_FL) { + /* We don't support clearning extent flags */ + if (!(flags & EXT4_EXTENTS_FL)) { + err = -EOPNOTSUPP; + goto flags_out; + } + } else if (flags & EXT4_EXTENTS_FL) { + /* migrate the file */ + migrate = 1; + flags &= ~EXT4_EXTENTS_FL; + } handle = ext4_journal_start(inode, 1); if (IS_ERR(handle)) { @@ -109,6 +119,10 @@ flags_err: if ((jflag ^ oldflags) & (EXT4_JOURNAL_DATA_FL)) err = ext4_change_inode_journal_flag(inode, jflag); + if (err) + goto flags_out; + if (migrate) + err = ext4_ext_migrate(inode); flags_out: mutex_unlock(&inode->i_mutex); mnt_drop_write(filp->f_path.mnt); @@ -175,53 +189,10 @@ setversion_out: return ret; } #endif - case EXT4_IOC_GETRSVSZ: - if (test_opt(inode->i_sb, RESERVATION) - && S_ISREG(inode->i_mode) - && ei->i_block_alloc_info) { - rsv_window_size = ei->i_block_alloc_info->rsv_window_node.rsv_goal_size; - return put_user(rsv_window_size, (int __user *)arg); - } - return -ENOTTY; - case EXT4_IOC_SETRSVSZ: { - int err; - - if (!test_opt(inode->i_sb, RESERVATION) ||!S_ISREG(inode->i_mode)) - return -ENOTTY; - - if (!is_owner_or_cap(inode)) - return -EACCES; - - if (get_user(rsv_window_size, (int __user *)arg)) - return -EFAULT; - - err = mnt_want_write(filp->f_path.mnt); - if (err) - return err; - - if (rsv_window_size > EXT4_MAX_RESERVE_BLOCKS) - rsv_window_size = EXT4_MAX_RESERVE_BLOCKS; - - /* - * need to allocate reservation structure for this inode - * before set the window size - */ - down_write(&ei->i_data_sem); - if (!ei->i_block_alloc_info) - ext4_init_block_alloc_info(inode); - - if (ei->i_block_alloc_info){ - struct ext4_reserve_window_node *rsv = &ei->i_block_alloc_info->rsv_window_node; - rsv->rsv_goal_size = rsv_window_size; - } - up_write(&ei->i_data_sem); - mnt_drop_write(filp->f_path.mnt); - return 0; - } case EXT4_IOC_GROUP_EXTEND: { ext4_fsblk_t n_blocks_count; struct super_block *sb = inode->i_sb; - int err; + int err, err2; if (!capable(CAP_SYS_RESOURCE)) return -EPERM; @@ -235,8 +206,10 @@ setversion_out: err = ext4_group_extend(sb, EXT4_SB(sb)->s_es, n_blocks_count); jbd2_journal_lock_updates(EXT4_SB(sb)->s_journal); - jbd2_journal_flush(EXT4_SB(sb)->s_journal); + err2 = jbd2_journal_flush(EXT4_SB(sb)->s_journal); jbd2_journal_unlock_updates(EXT4_SB(sb)->s_journal); + if (err == 0) + err = err2; mnt_drop_write(filp->f_path.mnt); return err; @@ -244,7 +217,7 @@ setversion_out: case EXT4_IOC_GROUP_ADD: { struct ext4_new_group_data input; struct super_block *sb = inode->i_sb; - int err; + int err, err2; if (!capable(CAP_SYS_RESOURCE)) return -EPERM; @@ -259,15 +232,36 @@ setversion_out: err = ext4_group_add(sb, &input); jbd2_journal_lock_updates(EXT4_SB(sb)->s_journal); - jbd2_journal_flush(EXT4_SB(sb)->s_journal); + err2 = jbd2_journal_flush(EXT4_SB(sb)->s_journal); jbd2_journal_unlock_updates(EXT4_SB(sb)->s_journal); + if (err == 0) + err = err2; mnt_drop_write(filp->f_path.mnt); return err; } case EXT4_IOC_MIGRATE: - return ext4_ext_migrate(inode, filp, cmd, arg); + { + int err; + if (!is_owner_or_cap(inode)) + return -EACCES; + + err = mnt_want_write(filp->f_path.mnt); + if (err) + return err; + /* + * inode_mutex prevent write and truncate on the file. + * Read still goes through. We take i_data_sem in + * ext4_ext_swap_inode_data before we switch the + * inode format to prevent read. + */ + mutex_lock(&(inode->i_mutex)); + err = ext4_ext_migrate(inode); + mutex_unlock(&(inode->i_mutex)); + mnt_drop_write(filp->f_path.mnt); + return err; + } default: return -ENOTTY; diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index e0e3a5eb1dd..b580714f0d8 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -477,9 +477,10 @@ static void mb_cmp_bitmaps(struct ext4_buddy *e4b, void *bitmap) b2 = (unsigned char *) bitmap; for (i = 0; i < e4b->bd_sb->s_blocksize; i++) { if (b1[i] != b2[i]) { - printk("corruption in group %lu at byte %u(%u):" - " %x in copy != %x on disk/prealloc\n", - e4b->bd_group, i, i * 8, b1[i], b2[i]); + printk(KERN_ERR "corruption in group %lu " + "at byte %u(%u): %x in copy != %x " + "on disk/prealloc\n", + e4b->bd_group, i, i * 8, b1[i], b2[i]); BUG(); } } @@ -533,9 +534,6 @@ static int __mb_check_buddy(struct ext4_buddy *e4b, char *file, void *buddy; void *buddy2; - if (!test_opt(sb, MBALLOC)) - return 0; - { static int mb_check_counter; if (mb_check_counter++ % 100 != 0) @@ -784,9 +782,11 @@ static int ext4_mb_init_cache(struct page *page, char *incore) if (bh[i] == NULL) goto out; - if (bh_uptodate_or_lock(bh[i])) + if (buffer_uptodate(bh[i]) && + !(desc->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT))) continue; + lock_buffer(bh[i]); spin_lock(sb_bgl_lock(EXT4_SB(sb), first_group + i)); if (desc->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) { ext4_init_block_bitmap(sb, bh[i], @@ -2169,9 +2169,10 @@ static void ext4_mb_history_release(struct super_block *sb) { struct ext4_sb_info *sbi = EXT4_SB(sb); - remove_proc_entry("mb_groups", sbi->s_mb_proc); - remove_proc_entry("mb_history", sbi->s_mb_proc); - + if (sbi->s_proc != NULL) { + remove_proc_entry("mb_groups", sbi->s_proc); + remove_proc_entry("mb_history", sbi->s_proc); + } kfree(sbi->s_mb_history); } @@ -2180,10 +2181,10 @@ static void ext4_mb_history_init(struct super_block *sb) struct ext4_sb_info *sbi = EXT4_SB(sb); int i; - if (sbi->s_mb_proc != NULL) { - proc_create_data("mb_history", S_IRUGO, sbi->s_mb_proc, + if (sbi->s_proc != NULL) { + proc_create_data("mb_history", S_IRUGO, sbi->s_proc, &ext4_mb_seq_history_fops, sb); - proc_create_data("mb_groups", S_IRUGO, sbi->s_mb_proc, + proc_create_data("mb_groups", S_IRUGO, sbi->s_proc, &ext4_mb_seq_groups_fops, sb); } @@ -2485,19 +2486,14 @@ int ext4_mb_init(struct super_block *sb, int needs_recovery) unsigned max; int ret; - if (!test_opt(sb, MBALLOC)) - return 0; - i = (sb->s_blocksize_bits + 2) * sizeof(unsigned short); sbi->s_mb_offsets = kmalloc(i, GFP_KERNEL); if (sbi->s_mb_offsets == NULL) { - clear_opt(sbi->s_mount_opt, MBALLOC); return -ENOMEM; } sbi->s_mb_maxs = kmalloc(i, GFP_KERNEL); if (sbi->s_mb_maxs == NULL) { - clear_opt(sbi->s_mount_opt, MBALLOC); kfree(sbi->s_mb_maxs); return -ENOMEM; } @@ -2520,7 +2516,6 @@ int ext4_mb_init(struct super_block *sb, int needs_recovery) /* init file for buddy data */ ret = ext4_mb_init_backend(sb); if (ret != 0) { - clear_opt(sbi->s_mount_opt, MBALLOC); kfree(sbi->s_mb_offsets); kfree(sbi->s_mb_maxs); return ret; @@ -2540,17 +2535,15 @@ int ext4_mb_init(struct super_block *sb, int needs_recovery) sbi->s_mb_history_filter = EXT4_MB_HISTORY_DEFAULT; sbi->s_mb_group_prealloc = MB_DEFAULT_GROUP_PREALLOC; - i = sizeof(struct ext4_locality_group) * nr_cpu_ids; - sbi->s_locality_groups = kmalloc(i, GFP_KERNEL); + sbi->s_locality_groups = alloc_percpu(struct ext4_locality_group); if (sbi->s_locality_groups == NULL) { - clear_opt(sbi->s_mount_opt, MBALLOC); kfree(sbi->s_mb_offsets); kfree(sbi->s_mb_maxs); return -ENOMEM; } - for (i = 0; i < nr_cpu_ids; i++) { + for_each_possible_cpu(i) { struct ext4_locality_group *lg; - lg = &sbi->s_locality_groups[i]; + lg = per_cpu_ptr(sbi->s_locality_groups, i); mutex_init(&lg->lg_mutex); for (j = 0; j < PREALLOC_TB_SIZE; j++) INIT_LIST_HEAD(&lg->lg_prealloc_list[j]); @@ -2560,7 +2553,7 @@ int ext4_mb_init(struct super_block *sb, int needs_recovery) ext4_mb_init_per_dev_proc(sb); ext4_mb_history_init(sb); - printk("EXT4-fs: mballoc enabled\n"); + printk(KERN_INFO "EXT4-fs: mballoc enabled\n"); return 0; } @@ -2589,9 +2582,6 @@ int ext4_mb_release(struct super_block *sb) struct ext4_group_info *grinfo; struct ext4_sb_info *sbi = EXT4_SB(sb); - if (!test_opt(sb, MBALLOC)) - return 0; - /* release freed, non-committed blocks */ spin_lock(&sbi->s_md_lock); list_splice_init(&sbi->s_closed_transaction, @@ -2647,8 +2637,7 @@ int ext4_mb_release(struct super_block *sb) atomic_read(&sbi->s_mb_discarded)); } - kfree(sbi->s_locality_groups); - + free_percpu(sbi->s_locality_groups); ext4_mb_history_release(sb); ext4_mb_destroy_per_dev_proc(sb); @@ -2721,118 +2710,46 @@ ext4_mb_free_committed_blocks(struct super_block *sb) #define EXT4_MB_STREAM_REQ "stream_req" #define EXT4_MB_GROUP_PREALLOC "group_prealloc" - - -#define MB_PROC_FOPS(name) \ -static int ext4_mb_##name##_proc_show(struct seq_file *m, void *v) \ -{ \ - struct ext4_sb_info *sbi = m->private; \ - \ - seq_printf(m, "%ld\n", sbi->s_mb_##name); \ - return 0; \ -} \ - \ -static int ext4_mb_##name##_proc_open(struct inode *inode, struct file *file)\ -{ \ - return single_open(file, ext4_mb_##name##_proc_show, PDE(inode)->data);\ -} \ - \ -static ssize_t ext4_mb_##name##_proc_write(struct file *file, \ - const char __user *buf, size_t cnt, loff_t *ppos) \ -{ \ - struct ext4_sb_info *sbi = PDE(file->f_path.dentry->d_inode)->data;\ - char str[32]; \ - long value; \ - if (cnt >= sizeof(str)) \ - return -EINVAL; \ - if (copy_from_user(str, buf, cnt)) \ - return -EFAULT; \ - value = simple_strtol(str, NULL, 0); \ - if (value <= 0) \ - return -ERANGE; \ - sbi->s_mb_##name = value; \ - return cnt; \ -} \ - \ -static const struct file_operations ext4_mb_##name##_proc_fops = { \ - .owner = THIS_MODULE, \ - .open = ext4_mb_##name##_proc_open, \ - .read = seq_read, \ - .llseek = seq_lseek, \ - .release = single_release, \ - .write = ext4_mb_##name##_proc_write, \ -}; - -MB_PROC_FOPS(stats); -MB_PROC_FOPS(max_to_scan); -MB_PROC_FOPS(min_to_scan); -MB_PROC_FOPS(order2_reqs); -MB_PROC_FOPS(stream_request); -MB_PROC_FOPS(group_prealloc); - -#define MB_PROC_HANDLER(name, var) \ -do { \ - proc = proc_create_data(name, mode, sbi->s_mb_proc, \ - &ext4_mb_##var##_proc_fops, sbi); \ - if (proc == NULL) { \ - printk(KERN_ERR "EXT4-fs: can't to create %s\n", name); \ - goto err_out; \ - } \ -} while (0) - static int ext4_mb_init_per_dev_proc(struct super_block *sb) { mode_t mode = S_IFREG | S_IRUGO | S_IWUSR; struct ext4_sb_info *sbi = EXT4_SB(sb); struct proc_dir_entry *proc; - char devname[64]; - if (proc_root_ext4 == NULL) { - sbi->s_mb_proc = NULL; + if (sbi->s_proc == NULL) return -EINVAL; - } - bdevname(sb->s_bdev, devname); - sbi->s_mb_proc = proc_mkdir(devname, proc_root_ext4); - - MB_PROC_HANDLER(EXT4_MB_STATS_NAME, stats); - MB_PROC_HANDLER(EXT4_MB_MAX_TO_SCAN_NAME, max_to_scan); - MB_PROC_HANDLER(EXT4_MB_MIN_TO_SCAN_NAME, min_to_scan); - MB_PROC_HANDLER(EXT4_MB_ORDER2_REQ, order2_reqs); - MB_PROC_HANDLER(EXT4_MB_STREAM_REQ, stream_request); - MB_PROC_HANDLER(EXT4_MB_GROUP_PREALLOC, group_prealloc); + EXT4_PROC_HANDLER(EXT4_MB_STATS_NAME, mb_stats); + EXT4_PROC_HANDLER(EXT4_MB_MAX_TO_SCAN_NAME, mb_max_to_scan); + EXT4_PROC_HANDLER(EXT4_MB_MIN_TO_SCAN_NAME, mb_min_to_scan); + EXT4_PROC_HANDLER(EXT4_MB_ORDER2_REQ, mb_order2_reqs); + EXT4_PROC_HANDLER(EXT4_MB_STREAM_REQ, mb_stream_request); + EXT4_PROC_HANDLER(EXT4_MB_GROUP_PREALLOC, mb_group_prealloc); return 0; err_out: - printk(KERN_ERR "EXT4-fs: Unable to create %s\n", devname); - remove_proc_entry(EXT4_MB_GROUP_PREALLOC, sbi->s_mb_proc); - remove_proc_entry(EXT4_MB_STREAM_REQ, sbi->s_mb_proc); - remove_proc_entry(EXT4_MB_ORDER2_REQ, sbi->s_mb_proc); - remove_proc_entry(EXT4_MB_MIN_TO_SCAN_NAME, sbi->s_mb_proc); - remove_proc_entry(EXT4_MB_MAX_TO_SCAN_NAME, sbi->s_mb_proc); - remove_proc_entry(EXT4_MB_STATS_NAME, sbi->s_mb_proc); - remove_proc_entry(devname, proc_root_ext4); - sbi->s_mb_proc = NULL; - + remove_proc_entry(EXT4_MB_GROUP_PREALLOC, sbi->s_proc); + remove_proc_entry(EXT4_MB_STREAM_REQ, sbi->s_proc); + remove_proc_entry(EXT4_MB_ORDER2_REQ, sbi->s_proc); + remove_proc_entry(EXT4_MB_MIN_TO_SCAN_NAME, sbi->s_proc); + remove_proc_entry(EXT4_MB_MAX_TO_SCAN_NAME, sbi->s_proc); + remove_proc_entry(EXT4_MB_STATS_NAME, sbi->s_proc); return -ENOMEM; } static int ext4_mb_destroy_per_dev_proc(struct super_block *sb) { struct ext4_sb_info *sbi = EXT4_SB(sb); - char devname[64]; - if (sbi->s_mb_proc == NULL) + if (sbi->s_proc == NULL) return -EINVAL; - bdevname(sb->s_bdev, devname); - remove_proc_entry(EXT4_MB_GROUP_PREALLOC, sbi->s_mb_proc); - remove_proc_entry(EXT4_MB_STREAM_REQ, sbi->s_mb_proc); - remove_proc_entry(EXT4_MB_ORDER2_REQ, sbi->s_mb_proc); - remove_proc_entry(EXT4_MB_MIN_TO_SCAN_NAME, sbi->s_mb_proc); - remove_proc_entry(EXT4_MB_MAX_TO_SCAN_NAME, sbi->s_mb_proc); - remove_proc_entry(EXT4_MB_STATS_NAME, sbi->s_mb_proc); - remove_proc_entry(devname, proc_root_ext4); + remove_proc_entry(EXT4_MB_GROUP_PREALLOC, sbi->s_proc); + remove_proc_entry(EXT4_MB_STREAM_REQ, sbi->s_proc); + remove_proc_entry(EXT4_MB_ORDER2_REQ, sbi->s_proc); + remove_proc_entry(EXT4_MB_MIN_TO_SCAN_NAME, sbi->s_proc); + remove_proc_entry(EXT4_MB_MAX_TO_SCAN_NAME, sbi->s_proc); + remove_proc_entry(EXT4_MB_STATS_NAME, sbi->s_proc); return 0; } @@ -2854,11 +2771,6 @@ int __init init_ext4_mballoc(void) kmem_cache_destroy(ext4_pspace_cachep); return -ENOMEM; } -#ifdef CONFIG_PROC_FS - proc_root_ext4 = proc_mkdir("fs/ext4", NULL); - if (proc_root_ext4 == NULL) - printk(KERN_ERR "EXT4-fs: Unable to create fs/ext4\n"); -#endif return 0; } @@ -2867,9 +2779,6 @@ void exit_ext4_mballoc(void) /* XXX: synchronize_rcu(); */ kmem_cache_destroy(ext4_pspace_cachep); kmem_cache_destroy(ext4_ac_cachep); -#ifdef CONFIG_PROC_FS - remove_proc_entry("fs/ext4", NULL); -#endif } @@ -2879,7 +2788,7 @@ void exit_ext4_mballoc(void) */ static noinline_for_stack int ext4_mb_mark_diskspace_used(struct ext4_allocation_context *ac, - handle_t *handle) + handle_t *handle, unsigned long reserv_blks) { struct buffer_head *bitmap_bh = NULL; struct ext4_super_block *es; @@ -2968,15 +2877,16 @@ ext4_mb_mark_diskspace_used(struct ext4_allocation_context *ac, le16_add_cpu(&gdp->bg_free_blocks_count, -ac->ac_b_ex.fe_len); gdp->bg_checksum = ext4_group_desc_csum(sbi, ac->ac_b_ex.fe_group, gdp); spin_unlock(sb_bgl_lock(sbi, ac->ac_b_ex.fe_group)); - + percpu_counter_sub(&sbi->s_freeblocks_counter, ac->ac_b_ex.fe_len); /* - * free blocks account has already be reduced/reserved - * at write_begin() time for delayed allocation - * do not double accounting + * Now reduce the dirty block count also. Should not go negative */ if (!(ac->ac_flags & EXT4_MB_DELALLOC_RESERVED)) - percpu_counter_sub(&sbi->s_freeblocks_counter, - ac->ac_b_ex.fe_len); + /* release all the reserved blocks if non delalloc */ + percpu_counter_sub(&sbi->s_dirtyblocks_counter, reserv_blks); + else + percpu_counter_sub(&sbi->s_dirtyblocks_counter, + ac->ac_b_ex.fe_len); if (sbi->s_log_groups_per_flex) { ext4_group_t flex_group = ext4_flex_group(sbi, @@ -3884,7 +3794,7 @@ out: * * FIXME!! Make sure it is valid at all the call sites */ -void ext4_mb_discard_inode_preallocations(struct inode *inode) +void ext4_discard_preallocations(struct inode *inode) { struct ext4_inode_info *ei = EXT4_I(inode); struct super_block *sb = inode->i_sb; @@ -3896,7 +3806,7 @@ void ext4_mb_discard_inode_preallocations(struct inode *inode) struct ext4_buddy e4b; int err; - if (!test_opt(sb, MBALLOC) || !S_ISREG(inode->i_mode)) { + if (!S_ISREG(inode->i_mode)) { /*BUG_ON(!list_empty(&ei->i_prealloc_list));*/ return; } @@ -4094,8 +4004,7 @@ static void ext4_mb_group_or_file(struct ext4_allocation_context *ac) * per cpu locality group is to reduce the contention between block * request from multiple CPUs. */ - ac->ac_lg = &sbi->s_locality_groups[get_cpu()]; - put_cpu(); + ac->ac_lg = per_cpu_ptr(sbi->s_locality_groups, raw_smp_processor_id()); /* we're going to use group allocation */ ac->ac_flags |= EXT4_MB_HINT_GROUP_ALLOC; @@ -4369,33 +4278,32 @@ static int ext4_mb_discard_preallocations(struct super_block *sb, int needed) ext4_fsblk_t ext4_mb_new_blocks(handle_t *handle, struct ext4_allocation_request *ar, int *errp) { + int freed; struct ext4_allocation_context *ac = NULL; struct ext4_sb_info *sbi; struct super_block *sb; ext4_fsblk_t block = 0; - int freed; - int inquota; + unsigned long inquota; + unsigned long reserv_blks = 0; sb = ar->inode->i_sb; sbi = EXT4_SB(sb); - if (!test_opt(sb, MBALLOC)) { - block = ext4_old_new_blocks(handle, ar->inode, ar->goal, - &(ar->len), errp); - return block; - } if (!EXT4_I(ar->inode)->i_delalloc_reserved_flag) { /* * With delalloc we already reserved the blocks */ - ar->len = ext4_has_free_blocks(sbi, ar->len); - } - - if (ar->len == 0) { - *errp = -ENOSPC; - return 0; + while (ar->len && ext4_claim_free_blocks(sbi, ar->len)) { + /* let others to free the space */ + yield(); + ar->len = ar->len >> 1; + } + if (!ar->len) { + *errp = -ENOSPC; + return 0; + } + reserv_blks = ar->len; } - while (ar->len && DQUOT_ALLOC_BLOCK(ar->inode, ar->len)) { ar->flags |= EXT4_MB_HINT_NOPREALLOC; ar->len--; @@ -4441,7 +4349,7 @@ repeat: } if (likely(ac->ac_status == AC_STATUS_FOUND)) { - *errp = ext4_mb_mark_diskspace_used(ac, handle); + *errp = ext4_mb_mark_diskspace_used(ac, handle, reserv_blks); if (*errp == -EAGAIN) { ac->ac_b_ex.fe_group = 0; ac->ac_b_ex.fe_start = 0; diff --git a/fs/ext4/mballoc.h b/fs/ext4/mballoc.h index c7c9906c2a7..b3b4828f8b8 100644 --- a/fs/ext4/mballoc.h +++ b/fs/ext4/mballoc.h @@ -257,7 +257,6 @@ static void ext4_mb_store_history(struct ext4_allocation_context *ac); #define in_range(b, first, len) ((b) >= (first) && (b) <= (first) + (len) - 1) -static struct proc_dir_entry *proc_root_ext4; struct buffer_head *read_block_bitmap(struct super_block *, ext4_group_t); static void ext4_mb_generate_from_pa(struct super_block *sb, void *bitmap, diff --git a/fs/ext4/migrate.c b/fs/ext4/migrate.c index 46fc0b5b12b..f2a9cf498ec 100644 --- a/fs/ext4/migrate.c +++ b/fs/ext4/migrate.c @@ -447,8 +447,7 @@ static int free_ext_block(handle_t *handle, struct inode *inode) } -int ext4_ext_migrate(struct inode *inode, struct file *filp, - unsigned int cmd, unsigned long arg) +int ext4_ext_migrate(struct inode *inode) { handle_t *handle; int retval = 0, i; @@ -516,12 +515,6 @@ int ext4_ext_migrate(struct inode *inode, struct file *filp, * when we add extents we extent the journal */ /* - * inode_mutex prevent write and truncate on the file. Read still goes - * through. We take i_data_sem in ext4_ext_swap_inode_data before we - * switch the inode format to prevent read. - */ - mutex_lock(&(inode->i_mutex)); - /* * Even though we take i_mutex we can still cause block allocation * via mmap write to holes. If we have allocated new blocks we fail * migrate. New block allocation will clear EXT4_EXT_MIGRATE flag. @@ -623,7 +616,6 @@ err_out: tmp_inode->i_nlink = 0; ext4_journal_stop(handle); - mutex_unlock(&(inode->i_mutex)); if (tmp_inode) iput(tmp_inode); diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index 387ad98350c..92db9e94514 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -151,34 +151,36 @@ struct dx_map_entry static inline ext4_lblk_t dx_get_block(struct dx_entry *entry); static void dx_set_block(struct dx_entry *entry, ext4_lblk_t value); -static inline unsigned dx_get_hash (struct dx_entry *entry); -static void dx_set_hash (struct dx_entry *entry, unsigned value); -static unsigned dx_get_count (struct dx_entry *entries); -static unsigned dx_get_limit (struct dx_entry *entries); -static void dx_set_count (struct dx_entry *entries, unsigned value); -static void dx_set_limit (struct dx_entry *entries, unsigned value); -static unsigned dx_root_limit (struct inode *dir, unsigned infosize); -static unsigned dx_node_limit (struct inode *dir); -static struct dx_frame *dx_probe(struct dentry *dentry, +static inline unsigned dx_get_hash(struct dx_entry *entry); +static void dx_set_hash(struct dx_entry *entry, unsigned value); +static unsigned dx_get_count(struct dx_entry *entries); +static unsigned dx_get_limit(struct dx_entry *entries); +static void dx_set_count(struct dx_entry *entries, unsigned value); +static void dx_set_limit(struct dx_entry *entries, unsigned value); +static unsigned dx_root_limit(struct inode *dir, unsigned infosize); +static unsigned dx_node_limit(struct inode *dir); +static struct dx_frame *dx_probe(const struct qstr *d_name, struct inode *dir, struct dx_hash_info *hinfo, struct dx_frame *frame, int *err); -static void dx_release (struct dx_frame *frames); -static int dx_make_map (struct ext4_dir_entry_2 *de, int size, - struct dx_hash_info *hinfo, struct dx_map_entry map[]); +static void dx_release(struct dx_frame *frames); +static int dx_make_map(struct ext4_dir_entry_2 *de, int size, + struct dx_hash_info *hinfo, struct dx_map_entry map[]); static void dx_sort_map(struct dx_map_entry *map, unsigned count); -static struct ext4_dir_entry_2 *dx_move_dirents (char *from, char *to, +static struct ext4_dir_entry_2 *dx_move_dirents(char *from, char *to, struct dx_map_entry *offsets, int count); -static struct ext4_dir_entry_2* dx_pack_dirents (char *base, int size); +static struct ext4_dir_entry_2* dx_pack_dirents(char *base, int size); static void dx_insert_block(struct dx_frame *frame, u32 hash, ext4_lblk_t block); static int ext4_htree_next_block(struct inode *dir, __u32 hash, struct dx_frame *frame, struct dx_frame *frames, __u32 *start_hash); -static struct buffer_head * ext4_dx_find_entry(struct dentry *dentry, - struct ext4_dir_entry_2 **res_dir, int *err); +static struct buffer_head * ext4_dx_find_entry(struct inode *dir, + const struct qstr *d_name, + struct ext4_dir_entry_2 **res_dir, + int *err); static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry, struct inode *inode); @@ -207,44 +209,44 @@ static inline void dx_set_block(struct dx_entry *entry, ext4_lblk_t value) entry->block = cpu_to_le32(value); } -static inline unsigned dx_get_hash (struct dx_entry *entry) +static inline unsigned dx_get_hash(struct dx_entry *entry) { return le32_to_cpu(entry->hash); } -static inline void dx_set_hash (struct dx_entry *entry, unsigned value) +static inline void dx_set_hash(struct dx_entry *entry, unsigned value) { entry->hash = cpu_to_le32(value); } -static inline unsigned dx_get_count (struct dx_entry *entries) +static inline unsigned dx_get_count(struct dx_entry *entries) { return le16_to_cpu(((struct dx_countlimit *) entries)->count); } -static inline unsigned dx_get_limit (struct dx_entry *entries) +static inline unsigned dx_get_limit(struct dx_entry *entries) { return le16_to_cpu(((struct dx_countlimit *) entries)->limit); } -static inline void dx_set_count (struct dx_entry *entries, unsigned value) +static inline void dx_set_count(struct dx_entry *entries, unsigned value) { ((struct dx_countlimit *) entries)->count = cpu_to_le16(value); } -static inline void dx_set_limit (struct dx_entry *entries, unsigned value) +static inline void dx_set_limit(struct dx_entry *entries, unsigned value) { ((struct dx_countlimit *) entries)->limit = cpu_to_le16(value); } -static inline unsigned dx_root_limit (struct inode *dir, unsigned infosize) +static inline unsigned dx_root_limit(struct inode *dir, unsigned infosize) { unsigned entry_space = dir->i_sb->s_blocksize - EXT4_DIR_REC_LEN(1) - EXT4_DIR_REC_LEN(2) - infosize; return entry_space / sizeof(struct dx_entry); } -static inline unsigned dx_node_limit (struct inode *dir) +static inline unsigned dx_node_limit(struct inode *dir) { unsigned entry_space = dir->i_sb->s_blocksize - EXT4_DIR_REC_LEN(0); return entry_space / sizeof(struct dx_entry); @@ -254,12 +256,12 @@ static inline unsigned dx_node_limit (struct inode *dir) * Debug */ #ifdef DX_DEBUG -static void dx_show_index (char * label, struct dx_entry *entries) +static void dx_show_index(char * label, struct dx_entry *entries) { int i, n = dx_get_count (entries); - printk("%s index ", label); + printk(KERN_DEBUG "%s index ", label); for (i = 0; i < n; i++) { - printk("%x->%lu ", i? dx_get_hash(entries + i) : + printk("%x->%lu ", i ? dx_get_hash(entries + i) : 0, (unsigned long)dx_get_block(entries + i)); } printk("\n"); @@ -306,7 +308,7 @@ struct stats dx_show_entries(struct dx_hash_info *hinfo, struct inode *dir, struct dx_entry *entries, int levels) { unsigned blocksize = dir->i_sb->s_blocksize; - unsigned count = dx_get_count (entries), names = 0, space = 0, i; + unsigned count = dx_get_count(entries), names = 0, space = 0, i; unsigned bcount = 0; struct buffer_head *bh; int err; @@ -325,11 +327,12 @@ struct stats dx_show_entries(struct dx_hash_info *hinfo, struct inode *dir, names += stats.names; space += stats.space; bcount += stats.bcount; - brelse (bh); + brelse(bh); } if (bcount) - printk("%snames %u, fullness %u (%u%%)\n", levels?"":" ", - names, space/bcount,(space/bcount)*100/blocksize); + printk(KERN_DEBUG "%snames %u, fullness %u (%u%%)\n", + levels ? "" : " ", names, space/bcount, + (space/bcount)*100/blocksize); return (struct stats) { names, space, bcount}; } #endif /* DX_DEBUG */ @@ -344,7 +347,7 @@ struct stats dx_show_entries(struct dx_hash_info *hinfo, struct inode *dir, * back to userspace. */ static struct dx_frame * -dx_probe(struct dentry *dentry, struct inode *dir, +dx_probe(const struct qstr *d_name, struct inode *dir, struct dx_hash_info *hinfo, struct dx_frame *frame_in, int *err) { unsigned count, indirect; @@ -355,8 +358,6 @@ dx_probe(struct dentry *dentry, struct inode *dir, u32 hash; frame->bh = NULL; - if (dentry) - dir = dentry->d_parent->d_inode; if (!(bh = ext4_bread (NULL,dir, 0, 0, err))) goto fail; root = (struct dx_root *) bh->b_data; @@ -372,8 +373,8 @@ dx_probe(struct dentry *dentry, struct inode *dir, } hinfo->hash_version = root->info.hash_version; hinfo->seed = EXT4_SB(dir->i_sb)->s_hash_seed; - if (dentry) - ext4fs_dirhash(dentry->d_name.name, dentry->d_name.len, hinfo); + if (d_name) + ext4fs_dirhash(d_name->name, d_name->len, hinfo); hash = hinfo->hash; if (root->info.unused_flags & 1) { @@ -406,7 +407,7 @@ dx_probe(struct dentry *dentry, struct inode *dir, goto fail; } - dxtrace (printk("Look up %x", hash)); + dxtrace(printk("Look up %x", hash)); while (1) { count = dx_get_count(entries); @@ -555,7 +556,7 @@ static int ext4_htree_next_block(struct inode *dir, __u32 hash, 0, &err))) return err; /* Failure */ p++; - brelse (p->bh); + brelse(p->bh); p->bh = bh; p->at = p->entries = ((struct dx_node *) bh->b_data)->entries; } @@ -593,7 +594,7 @@ static int htree_dirblock_to_tree(struct file *dir_file, /* On error, skip the f_pos to the next block. */ dir_file->f_pos = (dir_file->f_pos | (dir->i_sb->s_blocksize - 1)) + 1; - brelse (bh); + brelse(bh); return count; } ext4fs_dirhash(de->name, de->name_len, hinfo); @@ -635,8 +636,8 @@ int ext4_htree_fill_tree(struct file *dir_file, __u32 start_hash, int ret, err; __u32 hashval; - dxtrace(printk("In htree_fill_tree, start hash: %x:%x\n", start_hash, - start_minor_hash)); + dxtrace(printk(KERN_DEBUG "In htree_fill_tree, start hash: %x:%x\n", + start_hash, start_minor_hash)); dir = dir_file->f_path.dentry->d_inode; if (!(EXT4_I(dir)->i_flags & EXT4_INDEX_FL)) { hinfo.hash_version = EXT4_SB(dir->i_sb)->s_def_hash_version; @@ -648,7 +649,7 @@ int ext4_htree_fill_tree(struct file *dir_file, __u32 start_hash, } hinfo.hash = start_hash; hinfo.minor_hash = 0; - frame = dx_probe(NULL, dir_file->f_path.dentry->d_inode, &hinfo, frames, &err); + frame = dx_probe(NULL, dir, &hinfo, frames, &err); if (!frame) return err; @@ -694,8 +695,8 @@ int ext4_htree_fill_tree(struct file *dir_file, __u32 start_hash, break; } dx_release(frames); - dxtrace(printk("Fill tree: returned %d entries, next hash: %x\n", - count, *next_hash)); + dxtrace(printk(KERN_DEBUG "Fill tree: returned %d entries, " + "next hash: %x\n", count, *next_hash)); return count; errout: dx_release(frames); @@ -802,17 +803,17 @@ static inline int ext4_match (int len, const char * const name, /* * Returns 0 if not found, -1 on failure, and 1 on success */ -static inline int search_dirblock(struct buffer_head * bh, +static inline int search_dirblock(struct buffer_head *bh, struct inode *dir, - struct dentry *dentry, + const struct qstr *d_name, unsigned long offset, struct ext4_dir_entry_2 ** res_dir) { struct ext4_dir_entry_2 * de; char * dlimit; int de_len; - const char *name = dentry->d_name.name; - int namelen = dentry->d_name.len; + const char *name = d_name->name; + int namelen = d_name->len; de = (struct ext4_dir_entry_2 *) bh->b_data; dlimit = bh->b_data + dir->i_sb->s_blocksize; @@ -851,12 +852,13 @@ static inline int search_dirblock(struct buffer_head * bh, * The returned buffer_head has ->b_count elevated. The caller is expected * to brelse() it when appropriate. */ -static struct buffer_head * ext4_find_entry (struct dentry *dentry, +static struct buffer_head * ext4_find_entry (struct inode *dir, + const struct qstr *d_name, struct ext4_dir_entry_2 ** res_dir) { - struct super_block * sb; - struct buffer_head * bh_use[NAMEI_RA_SIZE]; - struct buffer_head * bh, *ret = NULL; + struct super_block *sb; + struct buffer_head *bh_use[NAMEI_RA_SIZE]; + struct buffer_head *bh, *ret = NULL; ext4_lblk_t start, block, b; int ra_max = 0; /* Number of bh's in the readahead buffer, bh_use[] */ @@ -865,16 +867,15 @@ static struct buffer_head * ext4_find_entry (struct dentry *dentry, int num = 0; ext4_lblk_t nblocks; int i, err; - struct inode *dir = dentry->d_parent->d_inode; int namelen; *res_dir = NULL; sb = dir->i_sb; - namelen = dentry->d_name.len; + namelen = d_name->len; if (namelen > EXT4_NAME_LEN) return NULL; if (is_dx(dir)) { - bh = ext4_dx_find_entry(dentry, res_dir, &err); + bh = ext4_dx_find_entry(dir, d_name, res_dir, &err); /* * On success, or if the error was file not found, * return. Otherwise, fall back to doing a search the @@ -882,7 +883,8 @@ static struct buffer_head * ext4_find_entry (struct dentry *dentry, */ if (bh || (err != ERR_BAD_DX_DIR)) return bh; - dxtrace(printk("ext4_find_entry: dx failed, falling back\n")); + dxtrace(printk(KERN_DEBUG "ext4_find_entry: dx failed, " + "falling back\n")); } nblocks = dir->i_size >> EXT4_BLOCK_SIZE_BITS(sb); start = EXT4_I(dir)->i_dir_start_lookup; @@ -926,7 +928,7 @@ restart: brelse(bh); goto next; } - i = search_dirblock(bh, dir, dentry, + i = search_dirblock(bh, dir, d_name, block << EXT4_BLOCK_SIZE_BITS(sb), res_dir); if (i == 1) { EXT4_I(dir)->i_dir_start_lookup = block; @@ -956,11 +958,11 @@ restart: cleanup_and_exit: /* Clean up the read-ahead blocks */ for (; ra_ptr < ra_max; ra_ptr++) - brelse (bh_use[ra_ptr]); + brelse(bh_use[ra_ptr]); return ret; } -static struct buffer_head * ext4_dx_find_entry(struct dentry *dentry, +static struct buffer_head * ext4_dx_find_entry(struct inode *dir, const struct qstr *d_name, struct ext4_dir_entry_2 **res_dir, int *err) { struct super_block * sb; @@ -971,14 +973,13 @@ static struct buffer_head * ext4_dx_find_entry(struct dentry *dentry, struct buffer_head *bh; ext4_lblk_t block; int retval; - int namelen = dentry->d_name.len; - const u8 *name = dentry->d_name.name; - struct inode *dir = dentry->d_parent->d_inode; + int namelen = d_name->len; + const u8 *name = d_name->name; sb = dir->i_sb; /* NFS may look up ".." - look at dx_root directory block */ if (namelen > 2 || name[0] != '.'||(name[1] != '.' && name[1] != '\0')){ - if (!(frame = dx_probe(dentry, NULL, &hinfo, frames, err))) + if (!(frame = dx_probe(d_name, dir, &hinfo, frames, err))) return NULL; } else { frame = frames; @@ -1010,7 +1011,7 @@ static struct buffer_head * ext4_dx_find_entry(struct dentry *dentry, return bh; } } - brelse (bh); + brelse(bh); /* Check to see if we should continue to search */ retval = ext4_htree_next_block(dir, hash, frame, frames, NULL); @@ -1025,25 +1026,25 @@ static struct buffer_head * ext4_dx_find_entry(struct dentry *dentry, *err = -ENOENT; errout: - dxtrace(printk("%s not found\n", name)); + dxtrace(printk(KERN_DEBUG "%s not found\n", name)); dx_release (frames); return NULL; } -static struct dentry *ext4_lookup(struct inode * dir, struct dentry *dentry, struct nameidata *nd) +static struct dentry *ext4_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) { - struct inode * inode; - struct ext4_dir_entry_2 * de; - struct buffer_head * bh; + struct inode *inode; + struct ext4_dir_entry_2 *de; + struct buffer_head *bh; if (dentry->d_name.len > EXT4_NAME_LEN) return ERR_PTR(-ENAMETOOLONG); - bh = ext4_find_entry(dentry, &de); + bh = ext4_find_entry(dir, &dentry->d_name, &de); inode = NULL; if (bh) { unsigned long ino = le32_to_cpu(de->inode); - brelse (bh); + brelse(bh); if (!ext4_valid_inum(dir->i_sb, ino)) { ext4_error(dir->i_sb, "ext4_lookup", "bad inode number: %lu", ino); @@ -1062,15 +1063,14 @@ struct dentry *ext4_get_parent(struct dentry *child) unsigned long ino; struct dentry *parent; struct inode *inode; - struct dentry dotdot; + static const struct qstr dotdot = { + .name = "..", + .len = 2, + }; struct ext4_dir_entry_2 * de; struct buffer_head *bh; - dotdot.d_name.name = ".."; - dotdot.d_name.len = 2; - dotdot.d_parent = child; /* confusing, isn't it! */ - - bh = ext4_find_entry(&dotdot, &de); + bh = ext4_find_entry(child->d_inode, &dotdot, &de); inode = NULL; if (!bh) return ERR_PTR(-ENOENT); @@ -1201,10 +1201,10 @@ static struct ext4_dir_entry_2 *do_split(handle_t *handle, struct inode *dir, /* create map in the end of data2 block */ map = (struct dx_map_entry *) (data2 + blocksize); - count = dx_make_map ((struct ext4_dir_entry_2 *) data1, + count = dx_make_map((struct ext4_dir_entry_2 *) data1, blocksize, hinfo, map); map -= count; - dx_sort_map (map, count); + dx_sort_map(map, count); /* Split the existing block in the middle, size-wise */ size = 0; move = 0; @@ -1225,7 +1225,7 @@ static struct ext4_dir_entry_2 *do_split(handle_t *handle, struct inode *dir, /* Fancy dance to stay within two buffers */ de2 = dx_move_dirents(data1, data2, map + split, count - split); - de = dx_pack_dirents(data1,blocksize); + de = dx_pack_dirents(data1, blocksize); de->rec_len = ext4_rec_len_to_disk(data1 + blocksize - (char *) de); de2->rec_len = ext4_rec_len_to_disk(data2 + blocksize - (char *) de2); dxtrace(dx_show_leaf (hinfo, (struct ext4_dir_entry_2 *) data1, blocksize, 1)); @@ -1237,15 +1237,15 @@ static struct ext4_dir_entry_2 *do_split(handle_t *handle, struct inode *dir, swap(*bh, bh2); de = de2; } - dx_insert_block (frame, hash2 + continued, newblock); - err = ext4_journal_dirty_metadata (handle, bh2); + dx_insert_block(frame, hash2 + continued, newblock); + err = ext4_journal_dirty_metadata(handle, bh2); if (err) goto journal_error; - err = ext4_journal_dirty_metadata (handle, frame->bh); + err = ext4_journal_dirty_metadata(handle, frame->bh); if (err) goto journal_error; - brelse (bh2); - dxtrace(dx_show_index ("frame", frame->entries)); + brelse(bh2); + dxtrace(dx_show_index("frame", frame->entries)); return de; journal_error: @@ -1271,7 +1271,7 @@ errout: */ static int add_dirent_to_buf(handle_t *handle, struct dentry *dentry, struct inode *inode, struct ext4_dir_entry_2 *de, - struct buffer_head * bh) + struct buffer_head *bh) { struct inode *dir = dentry->d_parent->d_inode; const char *name = dentry->d_name.name; @@ -1288,11 +1288,11 @@ static int add_dirent_to_buf(handle_t *handle, struct dentry *dentry, while ((char *) de <= top) { if (!ext4_check_dir_entry("ext4_add_entry", dir, de, bh, offset)) { - brelse (bh); + brelse(bh); return -EIO; } - if (ext4_match (namelen, name, de)) { - brelse (bh); + if (ext4_match(namelen, name, de)) { + brelse(bh); return -EEXIST; } nlen = EXT4_DIR_REC_LEN(de->name_len); @@ -1329,7 +1329,7 @@ static int add_dirent_to_buf(handle_t *handle, struct dentry *dentry, } else de->inode = 0; de->name_len = namelen; - memcpy (de->name, name, namelen); + memcpy(de->name, name, namelen); /* * XXX shouldn't update any times until successful * completion of syscall, but too many callers depend @@ -1377,7 +1377,7 @@ static int make_indexed_dir(handle_t *handle, struct dentry *dentry, struct fake_dirent *fde; blocksize = dir->i_sb->s_blocksize; - dxtrace(printk("Creating index\n")); + dxtrace(printk(KERN_DEBUG "Creating index\n")); retval = ext4_journal_get_write_access(handle, bh); if (retval) { ext4_std_error(dir->i_sb, retval); @@ -1386,7 +1386,7 @@ static int make_indexed_dir(handle_t *handle, struct dentry *dentry, } root = (struct dx_root *) bh->b_data; - bh2 = ext4_append (handle, dir, &block, &retval); + bh2 = ext4_append(handle, dir, &block, &retval); if (!(bh2)) { brelse(bh); return retval; @@ -1412,9 +1412,9 @@ static int make_indexed_dir(handle_t *handle, struct dentry *dentry, root->info.info_length = sizeof(root->info); root->info.hash_version = EXT4_SB(dir->i_sb)->s_def_hash_version; entries = root->entries; - dx_set_block (entries, 1); - dx_set_count (entries, 1); - dx_set_limit (entries, dx_root_limit(dir, sizeof(root->info))); + dx_set_block(entries, 1); + dx_set_count(entries, 1); + dx_set_limit(entries, dx_root_limit(dir, sizeof(root->info))); /* Initialize as for dx_probe */ hinfo.hash_version = root->info.hash_version; @@ -1443,14 +1443,14 @@ static int make_indexed_dir(handle_t *handle, struct dentry *dentry, * may not sleep between calling this and putting something into * the entry, as someone else might have used it while you slept. */ -static int ext4_add_entry (handle_t *handle, struct dentry *dentry, - struct inode *inode) +static int ext4_add_entry(handle_t *handle, struct dentry *dentry, + struct inode *inode) { struct inode *dir = dentry->d_parent->d_inode; unsigned long offset; - struct buffer_head * bh; + struct buffer_head *bh; struct ext4_dir_entry_2 *de; - struct super_block * sb; + struct super_block *sb; int retval; int dx_fallback=0; unsigned blocksize; @@ -1500,13 +1500,13 @@ static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry, struct dx_frame frames[2], *frame; struct dx_entry *entries, *at; struct dx_hash_info hinfo; - struct buffer_head * bh; + struct buffer_head *bh; struct inode *dir = dentry->d_parent->d_inode; - struct super_block * sb = dir->i_sb; + struct super_block *sb = dir->i_sb; struct ext4_dir_entry_2 *de; int err; - frame = dx_probe(dentry, NULL, &hinfo, frames, &err); + frame = dx_probe(&dentry->d_name, dir, &hinfo, frames, &err); if (!frame) return err; entries = frame->entries; @@ -1527,7 +1527,7 @@ static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry, } /* Block full, should compress but for now just split */ - dxtrace(printk("using %u of %u node entries\n", + dxtrace(printk(KERN_DEBUG "using %u of %u node entries\n", dx_get_count(entries), dx_get_limit(entries))); /* Need to split index? */ if (dx_get_count(entries) == dx_get_limit(entries)) { @@ -1559,7 +1559,8 @@ static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry, if (levels) { unsigned icount1 = icount/2, icount2 = icount - icount1; unsigned hash2 = dx_get_hash(entries + icount1); - dxtrace(printk("Split index %i/%i\n", icount1, icount2)); + dxtrace(printk(KERN_DEBUG "Split index %i/%i\n", + icount1, icount2)); BUFFER_TRACE(frame->bh, "get_write_access"); /* index root */ err = ext4_journal_get_write_access(handle, @@ -1567,11 +1568,11 @@ static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry, if (err) goto journal_error; - memcpy ((char *) entries2, (char *) (entries + icount1), - icount2 * sizeof(struct dx_entry)); - dx_set_count (entries, icount1); - dx_set_count (entries2, icount2); - dx_set_limit (entries2, dx_node_limit(dir)); + memcpy((char *) entries2, (char *) (entries + icount1), + icount2 * sizeof(struct dx_entry)); + dx_set_count(entries, icount1); + dx_set_count(entries2, icount2); + dx_set_limit(entries2, dx_node_limit(dir)); /* Which index block gets the new entry? */ if (at - entries >= icount1) { @@ -1579,16 +1580,17 @@ static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry, frame->entries = entries = entries2; swap(frame->bh, bh2); } - dx_insert_block (frames + 0, hash2, newblock); - dxtrace(dx_show_index ("node", frames[1].entries)); - dxtrace(dx_show_index ("node", + dx_insert_block(frames + 0, hash2, newblock); + dxtrace(dx_show_index("node", frames[1].entries)); + dxtrace(dx_show_index("node", ((struct dx_node *) bh2->b_data)->entries)); err = ext4_journal_dirty_metadata(handle, bh2); if (err) goto journal_error; brelse (bh2); } else { - dxtrace(printk("Creating second level index...\n")); + dxtrace(printk(KERN_DEBUG + "Creating second level index...\n")); memcpy((char *) entries2, (char *) entries, icount * sizeof(struct dx_entry)); dx_set_limit(entries2, dx_node_limit(dir)); @@ -1630,12 +1632,12 @@ cleanup: * ext4_delete_entry deletes a directory entry by merging it with the * previous entry */ -static int ext4_delete_entry (handle_t *handle, - struct inode * dir, - struct ext4_dir_entry_2 * de_del, - struct buffer_head * bh) +static int ext4_delete_entry(handle_t *handle, + struct inode *dir, + struct ext4_dir_entry_2 *de_del, + struct buffer_head *bh) { - struct ext4_dir_entry_2 * de, * pde; + struct ext4_dir_entry_2 *de, *pde; int i; i = 0; @@ -1716,11 +1718,11 @@ static int ext4_add_nondir(handle_t *handle, * If the create succeeds, we fill in the inode information * with d_instantiate(). */ -static int ext4_create (struct inode * dir, struct dentry * dentry, int mode, - struct nameidata *nd) +static int ext4_create(struct inode *dir, struct dentry *dentry, int mode, + struct nameidata *nd) { handle_t *handle; - struct inode * inode; + struct inode *inode; int err, retries = 0; retry: @@ -1747,8 +1749,8 @@ retry: return err; } -static int ext4_mknod (struct inode * dir, struct dentry *dentry, - int mode, dev_t rdev) +static int ext4_mknod(struct inode *dir, struct dentry *dentry, + int mode, dev_t rdev) { handle_t *handle; struct inode *inode; @@ -1767,11 +1769,11 @@ retry: if (IS_DIRSYNC(dir)) handle->h_sync = 1; - inode = ext4_new_inode (handle, dir, mode); + inode = ext4_new_inode(handle, dir, mode); err = PTR_ERR(inode); if (!IS_ERR(inode)) { init_special_inode(inode, inode->i_mode, rdev); -#ifdef CONFIG_EXT4DEV_FS_XATTR +#ifdef CONFIG_EXT4_FS_XATTR inode->i_op = &ext4_special_inode_operations; #endif err = ext4_add_nondir(handle, dentry, inode); @@ -1782,12 +1784,12 @@ retry: return err; } -static int ext4_mkdir(struct inode * dir, struct dentry * dentry, int mode) +static int ext4_mkdir(struct inode *dir, struct dentry *dentry, int mode) { handle_t *handle; - struct inode * inode; - struct buffer_head * dir_block; - struct ext4_dir_entry_2 * de; + struct inode *inode; + struct buffer_head *dir_block; + struct ext4_dir_entry_2 *de; int err, retries = 0; if (EXT4_DIR_LINK_MAX(dir)) @@ -1803,7 +1805,7 @@ retry: if (IS_DIRSYNC(dir)) handle->h_sync = 1; - inode = ext4_new_inode (handle, dir, S_IFDIR | mode); + inode = ext4_new_inode(handle, dir, S_IFDIR | mode); err = PTR_ERR(inode); if (IS_ERR(inode)) goto out_stop; @@ -1811,7 +1813,7 @@ retry: inode->i_op = &ext4_dir_inode_operations; inode->i_fop = &ext4_dir_operations; inode->i_size = EXT4_I(inode)->i_disksize = inode->i_sb->s_blocksize; - dir_block = ext4_bread (handle, inode, 0, 1, &err); + dir_block = ext4_bread(handle, inode, 0, 1, &err); if (!dir_block) goto out_clear_inode; BUFFER_TRACE(dir_block, "get_write_access"); @@ -1820,26 +1822,26 @@ retry: de->inode = cpu_to_le32(inode->i_ino); de->name_len = 1; de->rec_len = ext4_rec_len_to_disk(EXT4_DIR_REC_LEN(de->name_len)); - strcpy (de->name, "."); + strcpy(de->name, "."); ext4_set_de_type(dir->i_sb, de, S_IFDIR); de = ext4_next_entry(de); de->inode = cpu_to_le32(dir->i_ino); de->rec_len = ext4_rec_len_to_disk(inode->i_sb->s_blocksize - EXT4_DIR_REC_LEN(1)); de->name_len = 2; - strcpy (de->name, ".."); + strcpy(de->name, ".."); ext4_set_de_type(dir->i_sb, de, S_IFDIR); inode->i_nlink = 2; BUFFER_TRACE(dir_block, "call ext4_journal_dirty_metadata"); ext4_journal_dirty_metadata(handle, dir_block); - brelse (dir_block); + brelse(dir_block); ext4_mark_inode_dirty(handle, inode); - err = ext4_add_entry (handle, dentry, inode); + err = ext4_add_entry(handle, dentry, inode); if (err) { out_clear_inode: clear_nlink(inode); ext4_mark_inode_dirty(handle, inode); - iput (inode); + iput(inode); goto out_stop; } ext4_inc_count(handle, dir); @@ -1856,17 +1858,17 @@ out_stop: /* * routine to check that the specified directory is empty (for rmdir) */ -static int empty_dir (struct inode * inode) +static int empty_dir(struct inode *inode) { unsigned long offset; - struct buffer_head * bh; - struct ext4_dir_entry_2 * de, * de1; - struct super_block * sb; + struct buffer_head *bh; + struct ext4_dir_entry_2 *de, *de1; + struct super_block *sb; int err = 0; sb = inode->i_sb; if (inode->i_size < EXT4_DIR_REC_LEN(1) + EXT4_DIR_REC_LEN(2) || - !(bh = ext4_bread (NULL, inode, 0, 0, &err))) { + !(bh = ext4_bread(NULL, inode, 0, 0, &err))) { if (err) ext4_error(inode->i_sb, __func__, "error %d reading directory #%lu offset 0", @@ -1881,23 +1883,23 @@ static int empty_dir (struct inode * inode) de1 = ext4_next_entry(de); if (le32_to_cpu(de->inode) != inode->i_ino || !le32_to_cpu(de1->inode) || - strcmp (".", de->name) || - strcmp ("..", de1->name)) { - ext4_warning (inode->i_sb, "empty_dir", - "bad directory (dir #%lu) - no `.' or `..'", - inode->i_ino); - brelse (bh); + strcmp(".", de->name) || + strcmp("..", de1->name)) { + ext4_warning(inode->i_sb, "empty_dir", + "bad directory (dir #%lu) - no `.' or `..'", + inode->i_ino); + brelse(bh); return 1; } offset = ext4_rec_len_from_disk(de->rec_len) + ext4_rec_len_from_disk(de1->rec_len); de = ext4_next_entry(de1); - while (offset < inode->i_size ) { + while (offset < inode->i_size) { if (!bh || (void *) de >= (void *) (bh->b_data+sb->s_blocksize)) { err = 0; - brelse (bh); - bh = ext4_bread (NULL, inode, + brelse(bh); + bh = ext4_bread(NULL, inode, offset >> EXT4_BLOCK_SIZE_BITS(sb), 0, &err); if (!bh) { if (err) @@ -1917,13 +1919,13 @@ static int empty_dir (struct inode * inode) continue; } if (le32_to_cpu(de->inode)) { - brelse (bh); + brelse(bh); return 0; } offset += ext4_rec_len_from_disk(de->rec_len); de = ext4_next_entry(de); } - brelse (bh); + brelse(bh); return 1; } @@ -1954,8 +1956,8 @@ int ext4_orphan_add(handle_t *handle, struct inode *inode) * ->i_nlink. For, say it, character device. Not a regular file, * not a directory, not a symlink and ->i_nlink > 0. */ - J_ASSERT ((S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) || - S_ISLNK(inode->i_mode)) || inode->i_nlink == 0); + J_ASSERT((S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) || + S_ISLNK(inode->i_mode)) || inode->i_nlink == 0); BUFFER_TRACE(EXT4_SB(sb)->s_sbh, "get_write_access"); err = ext4_journal_get_write_access(handle, EXT4_SB(sb)->s_sbh); @@ -2069,12 +2071,12 @@ out_brelse: goto out_err; } -static int ext4_rmdir (struct inode * dir, struct dentry *dentry) +static int ext4_rmdir(struct inode *dir, struct dentry *dentry) { int retval; - struct inode * inode; - struct buffer_head * bh; - struct ext4_dir_entry_2 * de; + struct inode *inode; + struct buffer_head *bh; + struct ext4_dir_entry_2 *de; handle_t *handle; /* Initialize quotas before so that eventual writes go in @@ -2085,7 +2087,7 @@ static int ext4_rmdir (struct inode * dir, struct dentry *dentry) return PTR_ERR(handle); retval = -ENOENT; - bh = ext4_find_entry (dentry, &de); + bh = ext4_find_entry(dir, &dentry->d_name, &de); if (!bh) goto end_rmdir; @@ -2099,16 +2101,16 @@ static int ext4_rmdir (struct inode * dir, struct dentry *dentry) goto end_rmdir; retval = -ENOTEMPTY; - if (!empty_dir (inode)) + if (!empty_dir(inode)) goto end_rmdir; retval = ext4_delete_entry(handle, dir, de, bh); if (retval) goto end_rmdir; if (!EXT4_DIR_LINK_EMPTY(inode)) - ext4_warning (inode->i_sb, "ext4_rmdir", - "empty directory has too many links (%d)", - inode->i_nlink); + ext4_warning(inode->i_sb, "ext4_rmdir", + "empty directory has too many links (%d)", + inode->i_nlink); inode->i_version++; clear_nlink(inode); /* There's no need to set i_disksize: the fact that i_nlink is @@ -2124,16 +2126,16 @@ static int ext4_rmdir (struct inode * dir, struct dentry *dentry) end_rmdir: ext4_journal_stop(handle); - brelse (bh); + brelse(bh); return retval; } -static int ext4_unlink(struct inode * dir, struct dentry *dentry) +static int ext4_unlink(struct inode *dir, struct dentry *dentry) { int retval; - struct inode * inode; - struct buffer_head * bh; - struct ext4_dir_entry_2 * de; + struct inode *inode; + struct buffer_head *bh; + struct ext4_dir_entry_2 *de; handle_t *handle; /* Initialize quotas before so that eventual writes go @@ -2147,7 +2149,7 @@ static int ext4_unlink(struct inode * dir, struct dentry *dentry) handle->h_sync = 1; retval = -ENOENT; - bh = ext4_find_entry (dentry, &de); + bh = ext4_find_entry(dir, &dentry->d_name, &de); if (!bh) goto end_unlink; @@ -2158,9 +2160,9 @@ static int ext4_unlink(struct inode * dir, struct dentry *dentry) goto end_unlink; if (!inode->i_nlink) { - ext4_warning (inode->i_sb, "ext4_unlink", - "Deleting nonexistent file (%lu), %d", - inode->i_ino, inode->i_nlink); + ext4_warning(inode->i_sb, "ext4_unlink", + "Deleting nonexistent file (%lu), %d", + inode->i_ino, inode->i_nlink); inode->i_nlink = 1; } retval = ext4_delete_entry(handle, dir, de, bh); @@ -2178,15 +2180,15 @@ static int ext4_unlink(struct inode * dir, struct dentry *dentry) end_unlink: ext4_journal_stop(handle); - brelse (bh); + brelse(bh); return retval; } -static int ext4_symlink (struct inode * dir, - struct dentry *dentry, const char * symname) +static int ext4_symlink(struct inode *dir, + struct dentry *dentry, const char *symname) { handle_t *handle; - struct inode * inode; + struct inode *inode; int l, err, retries = 0; l = strlen(symname)+1; @@ -2203,12 +2205,12 @@ retry: if (IS_DIRSYNC(dir)) handle->h_sync = 1; - inode = ext4_new_inode (handle, dir, S_IFLNK|S_IRWXUGO); + inode = ext4_new_inode(handle, dir, S_IFLNK|S_IRWXUGO); err = PTR_ERR(inode); if (IS_ERR(inode)) goto out_stop; - if (l > sizeof (EXT4_I(inode)->i_data)) { + if (l > sizeof(EXT4_I(inode)->i_data)) { inode->i_op = &ext4_symlink_inode_operations; ext4_set_aops(inode); /* @@ -2221,14 +2223,14 @@ retry: if (err) { clear_nlink(inode); ext4_mark_inode_dirty(handle, inode); - iput (inode); + iput(inode); goto out_stop; } } else { /* clear the extent format for fast symlink */ EXT4_I(inode)->i_flags &= ~EXT4_EXTENTS_FL; inode->i_op = &ext4_fast_symlink_inode_operations; - memcpy((char*)&EXT4_I(inode)->i_data,symname,l); + memcpy((char *)&EXT4_I(inode)->i_data, symname, l); inode->i_size = l-1; } EXT4_I(inode)->i_disksize = inode->i_size; @@ -2240,8 +2242,8 @@ out_stop: return err; } -static int ext4_link (struct dentry * old_dentry, - struct inode * dir, struct dentry *dentry) +static int ext4_link(struct dentry *old_dentry, + struct inode *dir, struct dentry *dentry) { handle_t *handle; struct inode *inode = old_dentry->d_inode; @@ -2284,13 +2286,13 @@ retry: * Anybody can rename anything with this: the permission checks are left to the * higher-level routines. */ -static int ext4_rename (struct inode * old_dir, struct dentry *old_dentry, - struct inode * new_dir,struct dentry *new_dentry) +static int ext4_rename(struct inode *old_dir, struct dentry *old_dentry, + struct inode *new_dir, struct dentry *new_dentry) { handle_t *handle; - struct inode * old_inode, * new_inode; - struct buffer_head * old_bh, * new_bh, * dir_bh; - struct ext4_dir_entry_2 * old_de, * new_de; + struct inode *old_inode, *new_inode; + struct buffer_head *old_bh, *new_bh, *dir_bh; + struct ext4_dir_entry_2 *old_de, *new_de; int retval; old_bh = new_bh = dir_bh = NULL; @@ -2308,7 +2310,7 @@ static int ext4_rename (struct inode * old_dir, struct dentry *old_dentry, if (IS_DIRSYNC(old_dir) || IS_DIRSYNC(new_dir)) handle->h_sync = 1; - old_bh = ext4_find_entry (old_dentry, &old_de); + old_bh = ext4_find_entry(old_dir, &old_dentry->d_name, &old_de); /* * Check for inode number is _not_ due to possible IO errors. * We might rmdir the source, keep it as pwd of some process @@ -2321,32 +2323,32 @@ static int ext4_rename (struct inode * old_dir, struct dentry *old_dentry, goto end_rename; new_inode = new_dentry->d_inode; - new_bh = ext4_find_entry (new_dentry, &new_de); + new_bh = ext4_find_entry(new_dir, &new_dentry->d_name, &new_de); if (new_bh) { if (!new_inode) { - brelse (new_bh); + brelse(new_bh); new_bh = NULL; } } if (S_ISDIR(old_inode->i_mode)) { if (new_inode) { retval = -ENOTEMPTY; - if (!empty_dir (new_inode)) + if (!empty_dir(new_inode)) goto end_rename; } retval = -EIO; - dir_bh = ext4_bread (handle, old_inode, 0, 0, &retval); + dir_bh = ext4_bread(handle, old_inode, 0, 0, &retval); if (!dir_bh) goto end_rename; if (le32_to_cpu(PARENT_INO(dir_bh->b_data)) != old_dir->i_ino) goto end_rename; retval = -EMLINK; - if (!new_inode && new_dir!=old_dir && + if (!new_inode && new_dir != old_dir && new_dir->i_nlink >= EXT4_LINK_MAX) goto end_rename; } if (!new_bh) { - retval = ext4_add_entry (handle, new_dentry, old_inode); + retval = ext4_add_entry(handle, new_dentry, old_inode); if (retval) goto end_rename; } else { @@ -2388,7 +2390,7 @@ static int ext4_rename (struct inode * old_dir, struct dentry *old_dentry, struct buffer_head *old_bh2; struct ext4_dir_entry_2 *old_de2; - old_bh2 = ext4_find_entry(old_dentry, &old_de2); + old_bh2 = ext4_find_entry(old_dir, &old_dentry->d_name, &old_de2); if (old_bh2) { retval = ext4_delete_entry(handle, old_dir, old_de2, old_bh2); @@ -2433,9 +2435,9 @@ static int ext4_rename (struct inode * old_dir, struct dentry *old_dentry, retval = 0; end_rename: - brelse (dir_bh); - brelse (old_bh); - brelse (new_bh); + brelse(dir_bh); + brelse(old_bh); + brelse(new_bh); ext4_journal_stop(handle); return retval; } @@ -2454,7 +2456,7 @@ const struct inode_operations ext4_dir_inode_operations = { .mknod = ext4_mknod, .rename = ext4_rename, .setattr = ext4_setattr, -#ifdef CONFIG_EXT4DEV_FS_XATTR +#ifdef CONFIG_EXT4_FS_XATTR .setxattr = generic_setxattr, .getxattr = generic_getxattr, .listxattr = ext4_listxattr, @@ -2465,7 +2467,7 @@ const struct inode_operations ext4_dir_inode_operations = { const struct inode_operations ext4_special_inode_operations = { .setattr = ext4_setattr, -#ifdef CONFIG_EXT4DEV_FS_XATTR +#ifdef CONFIG_EXT4_FS_XATTR .setxattr = generic_setxattr, .getxattr = generic_getxattr, .listxattr = ext4_listxattr, diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c index b3d35604ea1..b6ec1843a01 100644 --- a/fs/ext4/resize.c +++ b/fs/ext4/resize.c @@ -416,8 +416,8 @@ static int add_new_gdb(handle_t *handle, struct inode *inode, "EXT4-fs: ext4_add_new_gdb: adding group block %lu\n", gdb_num); - /* - * If we are not using the primary superblock/GDT copy don't resize, + /* + * If we are not using the primary superblock/GDT copy don't resize, * because the user tools have no way of handling this. Probably a * bad time to do it anyways. */ @@ -870,11 +870,10 @@ int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input) * We can allocate memory for mb_alloc based on the new group * descriptor */ - if (test_opt(sb, MBALLOC)) { - err = ext4_mb_add_more_groupinfo(sb, input->group, gdp); - if (err) - goto exit_journal; - } + err = ext4_mb_add_more_groupinfo(sb, input->group, gdp); + if (err) + goto exit_journal; + /* * Make the new blocks and inodes valid next. We do this before * increasing the group count so that once the group is enabled, @@ -929,6 +928,15 @@ int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input) percpu_counter_add(&sbi->s_freeinodes_counter, EXT4_INODES_PER_GROUP(sb)); + if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_FLEX_BG)) { + ext4_group_t flex_group; + flex_group = ext4_flex_group(sbi, input->group); + sbi->s_flex_groups[flex_group].free_blocks += + input->free_blocks_count; + sbi->s_flex_groups[flex_group].free_inodes += + EXT4_INODES_PER_GROUP(sb); + } + ext4_journal_dirty_metadata(handle, sbi->s_sbh); sb->s_dirt = 1; @@ -964,7 +972,7 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es, ext4_group_t o_groups_count; ext4_grpblk_t last; ext4_grpblk_t add; - struct buffer_head * bh; + struct buffer_head *bh; handle_t *handle; int err; unsigned long freed_blocks; @@ -1077,8 +1085,15 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es, /* * Mark mballoc pages as not up to date so that they will be updated * next time they are loaded by ext4_mb_load_buddy. + * + * XXX Bad, Bad, BAD!!! We should not be overloading the + * Uptodate flag, particularly on thte bitmap bh, as way of + * hinting to ext4_mb_load_buddy() that it needs to be + * overloaded. A user could take a LVM snapshot, then do an + * on-line fsck, and clear the uptodate flag, and this would + * not be a bug in userspace, but a bug in the kernel. FIXME!!! */ - if (test_opt(sb, MBALLOC)) { + { struct ext4_sb_info *sbi = EXT4_SB(sb); struct inode *inode = sbi->s_buddy_cache; int blocks_per_page; diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 566344b926b..fb940c22ab0 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -34,6 +34,8 @@ #include <linux/namei.h> #include <linux/quotaops.h> #include <linux/seq_file.h> +#include <linux/proc_fs.h> +#include <linux/marker.h> #include <linux/log2.h> #include <linux/crc16.h> #include <asm/uaccess.h> @@ -45,6 +47,8 @@ #include "namei.h" #include "group.h" +struct proc_dir_entry *ext4_proc_root; + static int ext4_load_journal(struct super_block *, struct ext4_super_block *, unsigned long journal_devnum); static int ext4_create_journal(struct super_block *, struct ext4_super_block *, @@ -503,15 +507,18 @@ static void ext4_put_super(struct super_block *sb) ext4_mb_release(sb); ext4_ext_release(sb); ext4_xattr_put_super(sb); - jbd2_journal_destroy(sbi->s_journal); + if (jbd2_journal_destroy(sbi->s_journal) < 0) + ext4_abort(sb, __func__, "Couldn't clean up the journal"); sbi->s_journal = NULL; if (!(sb->s_flags & MS_RDONLY)) { EXT4_CLEAR_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER); es->s_state = cpu_to_le16(sbi->s_mount_state); - BUFFER_TRACE(sbi->s_sbh, "marking dirty"); - mark_buffer_dirty(sbi->s_sbh); ext4_commit_super(sb, es, 1); } + if (sbi->s_proc) { + remove_proc_entry("inode_readahead_blks", sbi->s_proc); + remove_proc_entry(sb->s_id, ext4_proc_root); + } for (i = 0; i < sbi->s_gdb_count; i++) brelse(sbi->s_group_desc[i]); @@ -520,6 +527,7 @@ static void ext4_put_super(struct super_block *sb) percpu_counter_destroy(&sbi->s_freeblocks_counter); percpu_counter_destroy(&sbi->s_freeinodes_counter); percpu_counter_destroy(&sbi->s_dirs_counter); + percpu_counter_destroy(&sbi->s_dirtyblocks_counter); brelse(sbi->s_sbh); #ifdef CONFIG_QUOTA for (i = 0; i < MAXQUOTAS; i++) @@ -562,11 +570,10 @@ static struct inode *ext4_alloc_inode(struct super_block *sb) ei = kmem_cache_alloc(ext4_inode_cachep, GFP_NOFS); if (!ei) return NULL; -#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL +#ifdef CONFIG_EXT4_FS_POSIX_ACL ei->i_acl = EXT4_ACL_NOT_CACHED; ei->i_default_acl = EXT4_ACL_NOT_CACHED; #endif - ei->i_block_alloc_info = NULL; ei->vfs_inode.i_version = 1; ei->vfs_inode.i_data.writeback_index = 0; memset(&ei->i_cached_extent, 0, sizeof(struct ext4_ext_cache)); @@ -599,7 +606,7 @@ static void init_once(void *foo) struct ext4_inode_info *ei = (struct ext4_inode_info *) foo; INIT_LIST_HEAD(&ei->i_orphan); -#ifdef CONFIG_EXT4DEV_FS_XATTR +#ifdef CONFIG_EXT4_FS_XATTR init_rwsem(&ei->xattr_sem); #endif init_rwsem(&ei->i_data_sem); @@ -625,8 +632,7 @@ static void destroy_inodecache(void) static void ext4_clear_inode(struct inode *inode) { - struct ext4_block_alloc_info *rsv = EXT4_I(inode)->i_block_alloc_info; -#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL +#ifdef CONFIG_EXT4_FS_POSIX_ACL if (EXT4_I(inode)->i_acl && EXT4_I(inode)->i_acl != EXT4_ACL_NOT_CACHED) { posix_acl_release(EXT4_I(inode)->i_acl); @@ -638,10 +644,7 @@ static void ext4_clear_inode(struct inode *inode) EXT4_I(inode)->i_default_acl = EXT4_ACL_NOT_CACHED; } #endif - ext4_discard_reservation(inode); - EXT4_I(inode)->i_block_alloc_info = NULL; - if (unlikely(rsv)) - kfree(rsv); + ext4_discard_preallocations(inode); jbd2_journal_release_jbd_inode(EXT4_SB(inode->i_sb)->s_journal, &EXT4_I(inode)->jinode); } @@ -654,7 +657,7 @@ static inline void ext4_show_quota_options(struct seq_file *seq, if (sbi->s_jquota_fmt) seq_printf(seq, ",jqfmt=%s", - (sbi->s_jquota_fmt == QFMT_VFS_OLD) ? "vfsold": "vfsv0"); + (sbi->s_jquota_fmt == QFMT_VFS_OLD) ? "vfsold" : "vfsv0"); if (sbi->s_qf_names[USRQUOTA]) seq_printf(seq, ",usrjquota=%s", sbi->s_qf_names[USRQUOTA]); @@ -718,7 +721,7 @@ static int ext4_show_options(struct seq_file *seq, struct vfsmount *vfs) seq_puts(seq, ",debug"); if (test_opt(sb, OLDALLOC)) seq_puts(seq, ",oldalloc"); -#ifdef CONFIG_EXT4DEV_FS_XATTR +#ifdef CONFIG_EXT4_FS_XATTR if (test_opt(sb, XATTR_USER) && !(def_mount_opts & EXT4_DEFM_XATTR_USER)) seq_puts(seq, ",user_xattr"); @@ -727,7 +730,7 @@ static int ext4_show_options(struct seq_file *seq, struct vfsmount *vfs) seq_puts(seq, ",nouser_xattr"); } #endif -#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL +#ifdef CONFIG_EXT4_FS_POSIX_ACL if (test_opt(sb, POSIX_ACL) && !(def_mount_opts & EXT4_DEFM_ACL)) seq_puts(seq, ",acl"); if (!test_opt(sb, POSIX_ACL) && (def_mount_opts & EXT4_DEFM_ACL)) @@ -752,8 +755,6 @@ static int ext4_show_options(struct seq_file *seq, struct vfsmount *vfs) seq_puts(seq, ",nobh"); if (!test_opt(sb, EXTENTS)) seq_puts(seq, ",noextents"); - if (!test_opt(sb, MBALLOC)) - seq_puts(seq, ",nomballoc"); if (test_opt(sb, I_VERSION)) seq_puts(seq, ",i_version"); if (!test_opt(sb, DELALLOC)) @@ -773,6 +774,13 @@ static int ext4_show_options(struct seq_file *seq, struct vfsmount *vfs) else if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_WRITEBACK_DATA) seq_puts(seq, ",data=writeback"); + if (sbi->s_inode_readahead_blks != EXT4_DEF_INODE_READAHEAD_BLKS) + seq_printf(seq, ",inode_readahead_blks=%u", + sbi->s_inode_readahead_blks); + + if (test_opt(sb, DATA_ERR_ABORT)) + seq_puts(seq, ",data_err=abort"); + ext4_show_quota_options(seq, sb); return 0; } @@ -822,7 +830,7 @@ static struct dentry *ext4_fh_to_parent(struct super_block *sb, struct fid *fid, } #ifdef CONFIG_QUOTA -#define QTYPE2NAME(t) ((t) == USRQUOTA?"user":"group") +#define QTYPE2NAME(t) ((t) == USRQUOTA ? "user" : "group") #define QTYPE2MOPT(on, t) ((t) == USRQUOTA?((on)##USRJQUOTA):((on)##GRPJQUOTA)) static int ext4_dquot_initialize(struct inode *inode, int type); @@ -902,11 +910,13 @@ enum { Opt_commit, Opt_journal_update, Opt_journal_inum, Opt_journal_dev, Opt_journal_checksum, Opt_journal_async_commit, Opt_abort, Opt_data_journal, Opt_data_ordered, Opt_data_writeback, + Opt_data_err_abort, Opt_data_err_ignore, Opt_usrjquota, Opt_grpjquota, Opt_offusrjquota, Opt_offgrpjquota, Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0, Opt_quota, Opt_noquota, Opt_ignore, Opt_barrier, Opt_err, Opt_resize, Opt_usrquota, Opt_grpquota, Opt_extents, Opt_noextents, Opt_i_version, Opt_mballoc, Opt_nomballoc, Opt_stripe, Opt_delalloc, Opt_nodelalloc, + Opt_inode_readahead_blks }; static match_table_t tokens = { @@ -947,6 +957,8 @@ static match_table_t tokens = { {Opt_data_journal, "data=journal"}, {Opt_data_ordered, "data=ordered"}, {Opt_data_writeback, "data=writeback"}, + {Opt_data_err_abort, "data_err=abort"}, + {Opt_data_err_ignore, "data_err=ignore"}, {Opt_offusrjquota, "usrjquota="}, {Opt_usrjquota, "usrjquota=%s"}, {Opt_offgrpjquota, "grpjquota="}, @@ -967,6 +979,7 @@ static match_table_t tokens = { {Opt_resize, "resize"}, {Opt_delalloc, "delalloc"}, {Opt_nodelalloc, "nodelalloc"}, + {Opt_inode_readahead_blks, "inode_readahead_blks=%u"}, {Opt_err, NULL}, }; @@ -981,7 +994,7 @@ static ext4_fsblk_t get_sb_block(void **data) /*todo: use simple_strtoll with >32bit ext4 */ sb_block = simple_strtoul(options, &options, 0); if (*options && *options != ',') { - printk("EXT4-fs: Invalid sb specification: %s\n", + printk(KERN_ERR "EXT4-fs: Invalid sb specification: %s\n", (char *) *data); return 1; } @@ -1072,7 +1085,7 @@ static int parse_options(char *options, struct super_block *sb, case Opt_orlov: clear_opt(sbi->s_mount_opt, OLDALLOC); break; -#ifdef CONFIG_EXT4DEV_FS_XATTR +#ifdef CONFIG_EXT4_FS_XATTR case Opt_user_xattr: set_opt(sbi->s_mount_opt, XATTR_USER); break; @@ -1082,10 +1095,11 @@ static int parse_options(char *options, struct super_block *sb, #else case Opt_user_xattr: case Opt_nouser_xattr: - printk("EXT4 (no)user_xattr options not supported\n"); + printk(KERN_ERR "EXT4 (no)user_xattr options " + "not supported\n"); break; #endif -#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL +#ifdef CONFIG_EXT4_FS_POSIX_ACL case Opt_acl: set_opt(sbi->s_mount_opt, POSIX_ACL); break; @@ -1095,7 +1109,8 @@ static int parse_options(char *options, struct super_block *sb, #else case Opt_acl: case Opt_noacl: - printk("EXT4 (no)acl options not supported\n"); + printk(KERN_ERR "EXT4 (no)acl options " + "not supported\n"); break; #endif case Opt_reservation: @@ -1178,6 +1193,12 @@ static int parse_options(char *options, struct super_block *sb, sbi->s_mount_opt |= data_opt; } break; + case Opt_data_err_abort: + set_opt(sbi->s_mount_opt, DATA_ERR_ABORT); + break; + case Opt_data_err_ignore: + clear_opt(sbi->s_mount_opt, DATA_ERR_ABORT); + break; #ifdef CONFIG_QUOTA case Opt_usrjquota: qtype = USRQUOTA; @@ -1189,8 +1210,8 @@ set_qf_name: sb_any_quota_suspended(sb)) && !sbi->s_qf_names[qtype]) { printk(KERN_ERR - "EXT4-fs: Cannot change journaled " - "quota options when quota turned on.\n"); + "EXT4-fs: Cannot change journaled " + "quota options when quota turned on.\n"); return 0; } qname = match_strdup(&args[0]); @@ -1357,12 +1378,6 @@ set_qf_format: case Opt_nodelalloc: clear_opt(sbi->s_mount_opt, DELALLOC); break; - case Opt_mballoc: - set_opt(sbi->s_mount_opt, MBALLOC); - break; - case Opt_nomballoc: - clear_opt(sbi->s_mount_opt, MBALLOC); - break; case Opt_stripe: if (match_int(&args[0], &option)) return 0; @@ -1373,6 +1388,13 @@ set_qf_format: case Opt_delalloc: set_opt(sbi->s_mount_opt, DELALLOC); break; + case Opt_inode_readahead_blks: + if (match_int(&args[0], &option)) + return 0; + if (option < 0 || option > (1 << 30)) + return 0; + sbi->s_inode_readahead_blks = option; + break; default: printk(KERN_ERR "EXT4-fs: Unrecognized mount option \"%s\" " @@ -1473,15 +1495,9 @@ static int ext4_setup_super(struct super_block *sb, struct ext4_super_block *es, EXT4_INODES_PER_GROUP(sb), sbi->s_mount_opt); - printk(KERN_INFO "EXT4 FS on %s, ", sb->s_id); - if (EXT4_SB(sb)->s_journal->j_inode == NULL) { - char b[BDEVNAME_SIZE]; - - printk("external journal on %s\n", - bdevname(EXT4_SB(sb)->s_journal->j_dev, b)); - } else { - printk("internal journal\n"); - } + printk(KERN_INFO "EXT4 FS on %s, %s journal on %s\n", + sb->s_id, EXT4_SB(sb)->s_journal->j_inode ? "internal" : + "external", EXT4_SB(sb)->s_journal->j_devname); return res; } @@ -1504,8 +1520,11 @@ static int ext4_fill_flex_info(struct super_block *sb) sbi->s_log_groups_per_flex = sbi->s_es->s_log_groups_per_flex; groups_per_flex = 1 << sbi->s_log_groups_per_flex; - flex_group_count = (sbi->s_groups_count + groups_per_flex - 1) / - groups_per_flex; + /* We allocate both existing and potentially added groups */ + flex_group_count = ((sbi->s_groups_count + groups_per_flex - 1) + + ((sbi->s_es->s_reserved_gdt_blocks +1 ) << + EXT4_DESC_PER_BLOCK_BITS(sb))) / + groups_per_flex; sbi->s_flex_groups = kzalloc(flex_group_count * sizeof(struct flex_groups), GFP_KERNEL); if (sbi->s_flex_groups == NULL) { @@ -1584,7 +1603,7 @@ static int ext4_check_descriptors(struct super_block *sb) if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_FLEX_BG)) flexbg_flag = 1; - ext4_debug ("Checking group descriptors"); + ext4_debug("Checking group descriptors"); for (i = 0; i < sbi->s_groups_count; i++) { struct ext4_group_desc *gdp = ext4_get_group_desc(sb, i, NULL); @@ -1623,8 +1642,10 @@ static int ext4_check_descriptors(struct super_block *sb) "Checksum for group %lu failed (%u!=%u)\n", i, le16_to_cpu(ext4_group_desc_csum(sbi, i, gdp)), le16_to_cpu(gdp->bg_checksum)); - if (!(sb->s_flags & MS_RDONLY)) + if (!(sb->s_flags & MS_RDONLY)) { + spin_unlock(sb_bgl_lock(sbi, i)); return 0; + } } spin_unlock(sb_bgl_lock(sbi, i)); if (!flexbg_flag) @@ -1714,9 +1735,9 @@ static void ext4_orphan_cleanup(struct super_block *sb, DQUOT_INIT(inode); if (inode->i_nlink) { printk(KERN_DEBUG - "%s: truncating inode %lu to %Ld bytes\n", + "%s: truncating inode %lu to %lld bytes\n", __func__, inode->i_ino, inode->i_size); - jbd_debug(2, "truncating inode %lu to %Ld bytes\n", + jbd_debug(2, "truncating inode %lu to %lld bytes\n", inode->i_ino, inode->i_size); ext4_truncate(inode); nr_truncates++; @@ -1914,6 +1935,7 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) unsigned long journal_devnum = 0; unsigned long def_mount_opts; struct inode *root; + char *cp; int ret = -EINVAL; int blocksize; int db_count; @@ -1930,10 +1952,15 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) sbi->s_mount_opt = 0; sbi->s_resuid = EXT4_DEF_RESUID; sbi->s_resgid = EXT4_DEF_RESGID; + sbi->s_inode_readahead_blks = EXT4_DEF_INODE_READAHEAD_BLKS; sbi->s_sb_block = sb_block; unlock_kernel(); + /* Cleanup superblock name */ + for (cp = sb->s_id; (cp = strchr(cp, '/'));) + *cp = '!'; + blocksize = sb_min_blocksize(sb, EXT4_MIN_BLOCK_SIZE); if (!blocksize) { printk(KERN_ERR "EXT4-fs: unable to set blocksize\n"); @@ -1973,11 +2000,11 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) set_opt(sbi->s_mount_opt, GRPID); if (def_mount_opts & EXT4_DEFM_UID16) set_opt(sbi->s_mount_opt, NO_UID32); -#ifdef CONFIG_EXT4DEV_FS_XATTR +#ifdef CONFIG_EXT4_FS_XATTR if (def_mount_opts & EXT4_DEFM_XATTR_USER) set_opt(sbi->s_mount_opt, XATTR_USER); #endif -#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL +#ifdef CONFIG_EXT4_FS_POSIX_ACL if (def_mount_opts & EXT4_DEFM_ACL) set_opt(sbi->s_mount_opt, POSIX_ACL); #endif @@ -2012,11 +2039,6 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) ext4_warning(sb, __func__, "extents feature not enabled on this filesystem, " "use tune2fs.\n"); - /* - * turn on mballoc code by default in ext4 filesystem - * Use -o nomballoc to turn it off - */ - set_opt(sbi->s_mount_opt, MBALLOC); /* * enable delayed allocation by default @@ -2041,16 +2063,6 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) "running e2fsck is recommended\n"); /* - * Since ext4 is still considered development code, we require - * that the TEST_FILESYS flag in s->flags be set. - */ - if (!(le32_to_cpu(es->s_flags) & EXT2_FLAGS_TEST_FILESYS)) { - printk(KERN_WARNING "EXT4-fs: %s: not marked " - "OK to use with test code.\n", sb->s_id); - goto failed_mount; - } - - /* * Check feature flags regardless of the revision level, since we * previously didn't change the revision level when setting the flags, * so there is a chance incompat flags are set on a rev 0 filesystem. @@ -2219,6 +2231,16 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) goto failed_mount; } +#ifdef CONFIG_PROC_FS + if (ext4_proc_root) + sbi->s_proc = proc_mkdir(sb->s_id, ext4_proc_root); + + if (sbi->s_proc) + proc_create_data("inode_readahead_blks", 0644, sbi->s_proc, + &ext4_ui_proc_fops, + &sbi->s_inode_readahead_blks); +#endif + bgl_lock_init(&sbi->s_blockgroup_lock); for (i = 0; i < db_count; i++) { @@ -2257,24 +2279,14 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) err = percpu_counter_init(&sbi->s_dirs_counter, ext4_count_dirs(sb)); } + if (!err) { + err = percpu_counter_init(&sbi->s_dirtyblocks_counter, 0); + } if (err) { printk(KERN_ERR "EXT4-fs: insufficient memory\n"); goto failed_mount3; } - /* per fileystem reservation list head & lock */ - spin_lock_init(&sbi->s_rsv_window_lock); - sbi->s_rsv_window_root = RB_ROOT; - /* Add a single, static dummy reservation to the start of the - * reservation window list --- it gives us a placeholder for - * append-at-start-of-list which makes the allocation logic - * _much_ simpler. */ - sbi->s_rsv_window_head.rsv_start = EXT4_RESERVE_WINDOW_NOT_ALLOCATED; - sbi->s_rsv_window_head.rsv_end = EXT4_RESERVE_WINDOW_NOT_ALLOCATED; - sbi->s_rsv_window_head.rsv_alloc_hit = 0; - sbi->s_rsv_window_head.rsv_goal_size = 0; - ext4_rsv_window_add(sb, &sbi->s_rsv_window_head); - sbi->s_stripe = ext4_get_stripe_size(sbi); /* @@ -2471,7 +2483,12 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) printk(KERN_INFO "EXT4-fs: delayed allocation enabled\n"); ext4_ext_init(sb); - ext4_mb_init(sb, needs_recovery); + err = ext4_mb_init(sb, needs_recovery); + if (err) { + printk(KERN_ERR "EXT4-fs: failed to initalize mballoc (%d)\n", + err); + goto failed_mount4; + } lock_kernel(); return 0; @@ -2489,11 +2506,16 @@ failed_mount3: percpu_counter_destroy(&sbi->s_freeblocks_counter); percpu_counter_destroy(&sbi->s_freeinodes_counter); percpu_counter_destroy(&sbi->s_dirs_counter); + percpu_counter_destroy(&sbi->s_dirtyblocks_counter); failed_mount2: for (i = 0; i < db_count; i++) brelse(sbi->s_group_desc[i]); kfree(sbi->s_group_desc); failed_mount: + if (sbi->s_proc) { + remove_proc_entry("inode_readahead_blks", sbi->s_proc); + remove_proc_entry(sb->s_id, ext4_proc_root); + } #ifdef CONFIG_QUOTA for (i = 0; i < MAXQUOTAS; i++) kfree(sbi->s_qf_names[i]); @@ -2527,6 +2549,10 @@ static void ext4_init_journal_params(struct super_block *sb, journal_t *journal) journal->j_flags |= JBD2_BARRIER; else journal->j_flags &= ~JBD2_BARRIER; + if (test_opt(sb, DATA_ERR_ABORT)) + journal->j_flags |= JBD2_ABORT_ON_SYNCDATA_ERR; + else + journal->j_flags &= ~JBD2_ABORT_ON_SYNCDATA_ERR; spin_unlock(&journal->j_state_lock); } @@ -2552,7 +2578,7 @@ static journal_t *ext4_get_journal(struct super_block *sb, return NULL; } - jbd_debug(2, "Journal inode found at %p: %Ld bytes\n", + jbd_debug(2, "Journal inode found at %p: %lld bytes\n", journal_inode, journal_inode->i_size); if (!S_ISREG(journal_inode->i_mode)) { printk(KERN_ERR "EXT4-fs: invalid journal inode.\n"); @@ -2715,6 +2741,11 @@ static int ext4_load_journal(struct super_block *sb, return -EINVAL; } + if (journal->j_flags & JBD2_BARRIER) + printk(KERN_INFO "EXT4-fs: barriers enabled\n"); + else + printk(KERN_INFO "EXT4-fs: barriers disabled\n"); + if (!really_read_only && test_opt(sb, UPDATE_JOURNAL)) { err = jbd2_journal_update_format(journal); if (err) { @@ -2799,13 +2830,34 @@ static void ext4_commit_super(struct super_block *sb, if (!sbh) return; + if (buffer_write_io_error(sbh)) { + /* + * Oh, dear. A previous attempt to write the + * superblock failed. This could happen because the + * USB device was yanked out. Or it could happen to + * be a transient write error and maybe the block will + * be remapped. Nothing we can do but to retry the + * write and hope for the best. + */ + printk(KERN_ERR "ext4: previous I/O error to " + "superblock detected for %s.\n", sb->s_id); + clear_buffer_write_io_error(sbh); + set_buffer_uptodate(sbh); + } es->s_wtime = cpu_to_le32(get_seconds()); ext4_free_blocks_count_set(es, ext4_count_free_blocks(sb)); es->s_free_inodes_count = cpu_to_le32(ext4_count_free_inodes(sb)); BUFFER_TRACE(sbh, "marking dirty"); mark_buffer_dirty(sbh); - if (sync) + if (sync) { sync_dirty_buffer(sbh); + if (buffer_write_io_error(sbh)) { + printk(KERN_ERR "ext4: I/O error while writing " + "superblock for %s.\n", sb->s_id); + clear_buffer_write_io_error(sbh); + set_buffer_uptodate(sbh); + } + } } @@ -2820,7 +2872,9 @@ static void ext4_mark_recovery_complete(struct super_block *sb, journal_t *journal = EXT4_SB(sb)->s_journal; jbd2_journal_lock_updates(journal); - jbd2_journal_flush(journal); + if (jbd2_journal_flush(journal) < 0) + goto out; + lock_super(sb); if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER) && sb->s_flags & MS_RDONLY) { @@ -2829,6 +2883,8 @@ static void ext4_mark_recovery_complete(struct super_block *sb, ext4_commit_super(sb, es, 1); } unlock_super(sb); + +out: jbd2_journal_unlock_updates(journal); } @@ -2907,6 +2963,7 @@ static int ext4_sync_fs(struct super_block *sb, int wait) { tid_t target; + trace_mark(ext4_sync_fs, "dev %s wait %d", sb->s_id, wait); sb->s_dirt = 0; if (jbd2_journal_start_commit(EXT4_SB(sb)->s_journal, &target)) { if (wait) @@ -2928,7 +2985,13 @@ static void ext4_write_super_lockfs(struct super_block *sb) /* Now we set up the journal barrier. */ jbd2_journal_lock_updates(journal); - jbd2_journal_flush(journal); + + /* + * We don't want to clear needs_recovery flag when we failed + * to flush the journal. + */ + if (jbd2_journal_flush(journal) < 0) + return; /* Journal blocked and flushed, clear needs_recovery flag. */ EXT4_CLEAR_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER); @@ -3162,7 +3225,8 @@ static int ext4_statfs(struct dentry *dentry, struct kstatfs *buf) buf->f_type = EXT4_SUPER_MAGIC; buf->f_bsize = sb->s_blocksize; buf->f_blocks = ext4_blocks_count(es) - sbi->s_overhead_last; - buf->f_bfree = percpu_counter_sum_positive(&sbi->s_freeblocks_counter); + buf->f_bfree = percpu_counter_sum_positive(&sbi->s_freeblocks_counter) - + percpu_counter_sum_positive(&sbi->s_dirtyblocks_counter); ext4_free_blocks_count_set(es, buf->f_bfree); buf->f_bavail = buf->f_bfree - ext4_r_blocks_count(es); if (buf->f_bfree < ext4_r_blocks_count(es)) @@ -3367,8 +3431,12 @@ static int ext4_quota_on(struct super_block *sb, int type, int format_id, * otherwise be livelocked... */ jbd2_journal_lock_updates(EXT4_SB(sb)->s_journal); - jbd2_journal_flush(EXT4_SB(sb)->s_journal); + err = jbd2_journal_flush(EXT4_SB(sb)->s_journal); jbd2_journal_unlock_updates(EXT4_SB(sb)->s_journal); + if (err) { + path_put(&nd.path); + return err; + } } err = vfs_quota_on_path(sb, type, format_id, &nd.path); @@ -3432,7 +3500,7 @@ static ssize_t ext4_quota_write(struct super_block *sb, int type, handle_t *handle = journal_current_handle(); if (!handle) { - printk(KERN_WARNING "EXT4-fs: Quota write (off=%Lu, len=%Lu)" + printk(KERN_WARNING "EXT4-fs: Quota write (off=%llu, len=%llu)" " cancelled because transaction is not started.\n", (unsigned long long)off, (unsigned long long)len); return -EIO; @@ -3493,18 +3561,82 @@ static int ext4_get_sb(struct file_system_type *fs_type, return get_sb_bdev(fs_type, flags, dev_name, data, ext4_fill_super, mnt); } +#ifdef CONFIG_PROC_FS +static int ext4_ui_proc_show(struct seq_file *m, void *v) +{ + unsigned int *p = m->private; + + seq_printf(m, "%u\n", *p); + return 0; +} + +static int ext4_ui_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, ext4_ui_proc_show, PDE(inode)->data); +} + +static ssize_t ext4_ui_proc_write(struct file *file, const char __user *buf, + size_t cnt, loff_t *ppos) +{ + unsigned int *p = PDE(file->f_path.dentry->d_inode)->data; + char str[32]; + unsigned long value; + + if (cnt >= sizeof(str)) + return -EINVAL; + if (copy_from_user(str, buf, cnt)) + return -EFAULT; + value = simple_strtol(str, NULL, 0); + if (value < 0) + return -ERANGE; + *p = value; + return cnt; +} + +const struct file_operations ext4_ui_proc_fops = { + .owner = THIS_MODULE, + .open = ext4_ui_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, + .write = ext4_ui_proc_write, +}; +#endif + +static struct file_system_type ext4_fs_type = { + .owner = THIS_MODULE, + .name = "ext4", + .get_sb = ext4_get_sb, + .kill_sb = kill_block_super, + .fs_flags = FS_REQUIRES_DEV, +}; + +#ifdef CONFIG_EXT4DEV_COMPAT +static int ext4dev_get_sb(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data, struct vfsmount *mnt) +{ + printk(KERN_WARNING "EXT4-fs: Update your userspace programs " + "to mount using ext4\n"); + printk(KERN_WARNING "EXT4-fs: ext4dev backwards compatibility " + "will go away by 2.6.31\n"); + return get_sb_bdev(fs_type, flags, dev_name, data, ext4_fill_super, mnt); +} + static struct file_system_type ext4dev_fs_type = { .owner = THIS_MODULE, .name = "ext4dev", - .get_sb = ext4_get_sb, + .get_sb = ext4dev_get_sb, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV, }; +MODULE_ALIAS("ext4dev"); +#endif static int __init init_ext4_fs(void) { int err; + ext4_proc_root = proc_mkdir("fs/ext4", NULL); err = init_ext4_mballoc(); if (err) return err; @@ -3515,9 +3647,16 @@ static int __init init_ext4_fs(void) err = init_inodecache(); if (err) goto out1; - err = register_filesystem(&ext4dev_fs_type); + err = register_filesystem(&ext4_fs_type); if (err) goto out; +#ifdef CONFIG_EXT4DEV_COMPAT + err = register_filesystem(&ext4dev_fs_type); + if (err) { + unregister_filesystem(&ext4_fs_type); + goto out; + } +#endif return 0; out: destroy_inodecache(); @@ -3530,10 +3669,14 @@ out2: static void __exit exit_ext4_fs(void) { + unregister_filesystem(&ext4_fs_type); +#ifdef CONFIG_EXT4DEV_COMPAT unregister_filesystem(&ext4dev_fs_type); +#endif destroy_inodecache(); exit_ext4_xattr(); exit_ext4_mballoc(); + remove_proc_entry("fs/ext4", NULL); } MODULE_AUTHOR("Remy Card, Stephen Tweedie, Andrew Morton, Andreas Dilger, Theodore Ts'o and others"); diff --git a/fs/ext4/symlink.c b/fs/ext4/symlink.c index e9178643dc0..00740cb32be 100644 --- a/fs/ext4/symlink.c +++ b/fs/ext4/symlink.c @@ -23,10 +23,10 @@ #include "ext4.h" #include "xattr.h" -static void * ext4_follow_link(struct dentry *dentry, struct nameidata *nd) +static void *ext4_follow_link(struct dentry *dentry, struct nameidata *nd) { struct ext4_inode_info *ei = EXT4_I(dentry->d_inode); - nd_set_link(nd, (char*)ei->i_data); + nd_set_link(nd, (char *) ei->i_data); return NULL; } @@ -34,7 +34,7 @@ const struct inode_operations ext4_symlink_inode_operations = { .readlink = generic_readlink, .follow_link = page_follow_link_light, .put_link = page_put_link, -#ifdef CONFIG_EXT4DEV_FS_XATTR +#ifdef CONFIG_EXT4_FS_XATTR .setxattr = generic_setxattr, .getxattr = generic_getxattr, .listxattr = ext4_listxattr, @@ -45,7 +45,7 @@ const struct inode_operations ext4_symlink_inode_operations = { const struct inode_operations ext4_fast_symlink_inode_operations = { .readlink = generic_readlink, .follow_link = ext4_follow_link, -#ifdef CONFIG_EXT4DEV_FS_XATTR +#ifdef CONFIG_EXT4_FS_XATTR .setxattr = generic_setxattr, .getxattr = generic_getxattr, .listxattr = ext4_listxattr, diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c index 8954208b489..80626d516fe 100644 --- a/fs/ext4/xattr.c +++ b/fs/ext4/xattr.c @@ -99,12 +99,12 @@ static struct mb_cache *ext4_xattr_cache; static struct xattr_handler *ext4_xattr_handler_map[] = { [EXT4_XATTR_INDEX_USER] = &ext4_xattr_user_handler, -#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL +#ifdef CONFIG_EXT4_FS_POSIX_ACL [EXT4_XATTR_INDEX_POSIX_ACL_ACCESS] = &ext4_xattr_acl_access_handler, [EXT4_XATTR_INDEX_POSIX_ACL_DEFAULT] = &ext4_xattr_acl_default_handler, #endif [EXT4_XATTR_INDEX_TRUSTED] = &ext4_xattr_trusted_handler, -#ifdef CONFIG_EXT4DEV_FS_SECURITY +#ifdef CONFIG_EXT4_FS_SECURITY [EXT4_XATTR_INDEX_SECURITY] = &ext4_xattr_security_handler, #endif }; @@ -112,11 +112,11 @@ static struct xattr_handler *ext4_xattr_handler_map[] = { struct xattr_handler *ext4_xattr_handlers[] = { &ext4_xattr_user_handler, &ext4_xattr_trusted_handler, -#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL +#ifdef CONFIG_EXT4_FS_POSIX_ACL &ext4_xattr_acl_access_handler, &ext4_xattr_acl_default_handler, #endif -#ifdef CONFIG_EXT4DEV_FS_SECURITY +#ifdef CONFIG_EXT4_FS_SECURITY &ext4_xattr_security_handler, #endif NULL @@ -959,6 +959,7 @@ ext4_xattr_set_handle(handle_t *handle, struct inode *inode, int name_index, struct ext4_xattr_block_find bs = { .s = { .not_found = -ENODATA, }, }; + unsigned long no_expand; int error; if (!name) @@ -966,6 +967,9 @@ ext4_xattr_set_handle(handle_t *handle, struct inode *inode, int name_index, if (strlen(name) > 255) return -ERANGE; down_write(&EXT4_I(inode)->xattr_sem); + no_expand = EXT4_I(inode)->i_state & EXT4_STATE_NO_EXPAND; + EXT4_I(inode)->i_state |= EXT4_STATE_NO_EXPAND; + error = ext4_get_inode_loc(inode, &is.iloc); if (error) goto cleanup; @@ -1042,6 +1046,8 @@ ext4_xattr_set_handle(handle_t *handle, struct inode *inode, int name_index, cleanup: brelse(is.iloc.bh); brelse(bs.bh); + if (no_expand == 0) + EXT4_I(inode)->i_state &= ~EXT4_STATE_NO_EXPAND; up_write(&EXT4_I(inode)->xattr_sem); return error; } diff --git a/fs/ext4/xattr.h b/fs/ext4/xattr.h index 5992fe979bb..8ede88b18c2 100644 --- a/fs/ext4/xattr.h +++ b/fs/ext4/xattr.h @@ -51,8 +51,8 @@ struct ext4_xattr_entry { (((name_len) + EXT4_XATTR_ROUND + \ sizeof(struct ext4_xattr_entry)) & ~EXT4_XATTR_ROUND) #define EXT4_XATTR_NEXT(entry) \ - ( (struct ext4_xattr_entry *)( \ - (char *)(entry) + EXT4_XATTR_LEN((entry)->e_name_len)) ) + ((struct ext4_xattr_entry *)( \ + (char *)(entry) + EXT4_XATTR_LEN((entry)->e_name_len))) #define EXT4_XATTR_SIZE(size) \ (((size) + EXT4_XATTR_ROUND) & ~EXT4_XATTR_ROUND) @@ -63,7 +63,7 @@ struct ext4_xattr_entry { EXT4_I(inode)->i_extra_isize)) #define IFIRST(hdr) ((struct ext4_xattr_entry *)((hdr)+1)) -# ifdef CONFIG_EXT4DEV_FS_XATTR +# ifdef CONFIG_EXT4_FS_XATTR extern struct xattr_handler ext4_xattr_user_handler; extern struct xattr_handler ext4_xattr_trusted_handler; @@ -88,7 +88,7 @@ extern void exit_ext4_xattr(void); extern struct xattr_handler *ext4_xattr_handlers[]; -# else /* CONFIG_EXT4DEV_FS_XATTR */ +# else /* CONFIG_EXT4_FS_XATTR */ static inline int ext4_xattr_get(struct inode *inode, int name_index, const char *name, @@ -141,9 +141,9 @@ ext4_expand_extra_isize_ea(struct inode *inode, int new_extra_isize, #define ext4_xattr_handlers NULL -# endif /* CONFIG_EXT4DEV_FS_XATTR */ +# endif /* CONFIG_EXT4_FS_XATTR */ -#ifdef CONFIG_EXT4DEV_FS_SECURITY +#ifdef CONFIG_EXT4_FS_SECURITY extern int ext4_init_security(handle_t *handle, struct inode *inode, struct inode *dir); #else diff --git a/fs/fat/fatent.c b/fs/fat/fatent.c index 302e95c4af7..fb98b3d847e 100644 --- a/fs/fat/fatent.c +++ b/fs/fat/fatent.c @@ -6,6 +6,7 @@ #include <linux/module.h> #include <linux/fs.h> #include <linux/msdos_fs.h> +#include <linux/blkdev.h> struct fatent_operations { void (*ent_blocknr)(struct super_block *, int, int *, sector_t *); @@ -535,6 +536,7 @@ int fat_free_clusters(struct inode *inode, int cluster) struct fat_entry fatent; struct buffer_head *bhs[MAX_BUF_PER_PAGE]; int i, err, nr_bhs; + int first_cl = cluster; nr_bhs = 0; fatent_init(&fatent); @@ -551,6 +553,18 @@ int fat_free_clusters(struct inode *inode, int cluster) goto error; } + /* + * Issue discard for the sectors we no longer care about, + * batching contiguous clusters into one request + */ + if (cluster != fatent.entry + 1) { + int nr_clus = fatent.entry - first_cl + 1; + + sb_issue_discard(sb, fat_clus_to_blknr(sbi, first_cl), + nr_clus * sbi->sec_per_clus); + first_cl = cluster; + } + ops->ent_put(&fatent, FAT_ENT_FREE); if (sbi->free_clusters != -1) { sbi->free_clusters++; diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index 13391e54661..c962283d4e7 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -1265,6 +1265,8 @@ static void blocking_cb(struct gfs2_sbd *sdp, struct lm_lockname *name, holdtime = gl->gl_tchange + gl->gl_ops->go_min_hold_time; if (time_before(now, holdtime)) delay = holdtime - now; + if (test_bit(GLF_REPLY_PENDING, &gl->gl_flags)) + delay = gl->gl_ops->go_min_hold_time; spin_lock(&gl->gl_spin); handle_callback(gl, state, 1, delay); @@ -1578,8 +1580,6 @@ static const char *hflags2str(char *buf, unsigned flags, unsigned long iflags) *p++ = 'a'; if (flags & GL_EXACT) *p++ = 'E'; - if (flags & GL_ATIME) - *p++ = 'a'; if (flags & GL_NOCACHE) *p++ = 'c'; if (test_bit(HIF_HOLDER, &iflags)) @@ -1816,15 +1816,17 @@ restart: if (gl) { gi->gl = hlist_entry(gl->gl_list.next, struct gfs2_glock, gl_list); - if (gi->gl) - gfs2_glock_hold(gi->gl); + } else { + gi->gl = hlist_entry(gl_hash_table[gi->hash].hb_list.first, + struct gfs2_glock, gl_list); } + if (gi->gl) + gfs2_glock_hold(gi->gl); read_unlock(gl_lock_addr(gi->hash)); if (gl) gfs2_glock_put(gl); - if (gl && gi->gl == NULL) - gi->hash++; while (gi->gl == NULL) { + gi->hash++; if (gi->hash >= GFS2_GL_HASH_SIZE) return 1; read_lock(gl_lock_addr(gi->hash)); @@ -1833,7 +1835,6 @@ restart: if (gi->gl) gfs2_glock_hold(gi->gl); read_unlock(gl_lock_addr(gi->hash)); - gi->hash++; } if (gi->sdp != gi->gl->gl_sbd) diff --git a/fs/gfs2/glock.h b/fs/gfs2/glock.h index 971d92af70f..695c6b19361 100644 --- a/fs/gfs2/glock.h +++ b/fs/gfs2/glock.h @@ -24,7 +24,6 @@ #define GL_ASYNC 0x00000040 #define GL_EXACT 0x00000080 #define GL_SKIP 0x00000100 -#define GL_ATIME 0x00000200 #define GL_NOCACHE 0x00000400 #define GLR_TRYFAILED 13 diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h index 448697a5c46..f566ec1b4e8 100644 --- a/fs/gfs2/incore.h +++ b/fs/gfs2/incore.h @@ -386,20 +386,21 @@ struct gfs2_statfs_change_host { #define GFS2_DATA_ORDERED 2 struct gfs2_args { - char ar_lockproto[GFS2_LOCKNAME_LEN]; /* Name of the Lock Protocol */ - char ar_locktable[GFS2_LOCKNAME_LEN]; /* Name of the Lock Table */ - char ar_hostdata[GFS2_LOCKNAME_LEN]; /* Host specific data */ - int ar_spectator; /* Don't get a journal because we're always RO */ - int ar_ignore_local_fs; /* Don't optimize even if local_fs is 1 */ - int ar_localflocks; /* Let the VFS do flock|fcntl locks for us */ - int ar_localcaching; /* Local-style caching (dangerous on multihost) */ - int ar_debug; /* Oops on errors instead of trying to be graceful */ - int ar_upgrade; /* Upgrade ondisk/multihost format */ - unsigned int ar_num_glockd; /* Number of glockd threads */ - int ar_posix_acl; /* Enable posix acls */ - int ar_quota; /* off/account/on */ - int ar_suiddir; /* suiddir support */ - int ar_data; /* ordered/writeback */ + char ar_lockproto[GFS2_LOCKNAME_LEN]; /* Name of the Lock Protocol */ + char ar_locktable[GFS2_LOCKNAME_LEN]; /* Name of the Lock Table */ + char ar_hostdata[GFS2_LOCKNAME_LEN]; /* Host specific data */ + unsigned int ar_spectator:1; /* Don't get a journal */ + unsigned int ar_ignore_local_fs:1; /* Ignore optimisations */ + unsigned int ar_localflocks:1; /* Let the VFS do flock|fcntl */ + unsigned int ar_localcaching:1; /* Local caching */ + unsigned int ar_debug:1; /* Oops on errors */ + unsigned int ar_upgrade:1; /* Upgrade ondisk format */ + unsigned int ar_posix_acl:1; /* Enable posix acls */ + unsigned int ar_quota:2; /* off/account/on */ + unsigned int ar_suiddir:1; /* suiddir support */ + unsigned int ar_data:2; /* ordered/writeback */ + unsigned int ar_meta:1; /* mount metafs */ + unsigned int ar_num_glockd; /* Number of glockd threads */ }; struct gfs2_tune { @@ -419,7 +420,6 @@ struct gfs2_tune { unsigned int gt_quota_scale_den; /* Denominator */ unsigned int gt_quota_cache_secs; unsigned int gt_quota_quantum; /* Secs between syncs to quota file */ - unsigned int gt_atime_quantum; /* Min secs between atime updates */ unsigned int gt_new_files_jdata; unsigned int gt_max_readahead; /* Max bytes to read-ahead from disk */ unsigned int gt_stall_secs; /* Detects trouble! */ @@ -432,7 +432,7 @@ enum { SDF_JOURNAL_CHECKED = 0, SDF_JOURNAL_LIVE = 1, SDF_SHUTDOWN = 2, - SDF_NOATIME = 3, + SDF_NOBARRIERS = 3, }; #define GFS2_FSNAME_LEN 256 @@ -461,7 +461,6 @@ struct gfs2_sb_host { struct gfs2_sbd { struct super_block *sd_vfs; - struct super_block *sd_vfs_meta; struct kobject sd_kobj; unsigned long sd_flags; /* SDF_... */ struct gfs2_sb_host sd_sb; @@ -499,7 +498,9 @@ struct gfs2_sbd { /* Inode Stuff */ - struct inode *sd_master_dir; + struct dentry *sd_master_dir; + struct dentry *sd_root_dir; + struct inode *sd_jindex; struct inode *sd_inum_inode; struct inode *sd_statfs_inode; @@ -634,7 +635,6 @@ struct gfs2_sbd { /* Debugging crud */ unsigned long sd_last_warning; - struct vfsmount *sd_gfs2mnt; struct dentry *debugfs_dir; /* debugfs directory */ struct dentry *debugfs_dentry_glocks; /* for debugfs */ }; diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index 8b0806a3294..7cee695fa44 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -18,6 +18,7 @@ #include <linux/crc32.h> #include <linux/lm_interface.h> #include <linux/security.h> +#include <linux/time.h> #include "gfs2.h" #include "incore.h" @@ -249,6 +250,7 @@ static int gfs2_dinode_in(struct gfs2_inode *ip, const void *buf) { struct gfs2_dinode_host *di = &ip->i_di; const struct gfs2_dinode *str = buf; + struct timespec atime; u16 height, depth; if (unlikely(ip->i_no_addr != be64_to_cpu(str->di_num.no_addr))) @@ -275,8 +277,10 @@ static int gfs2_dinode_in(struct gfs2_inode *ip, const void *buf) di->di_size = be64_to_cpu(str->di_size); i_size_write(&ip->i_inode, di->di_size); gfs2_set_inode_blocks(&ip->i_inode, be64_to_cpu(str->di_blocks)); - ip->i_inode.i_atime.tv_sec = be64_to_cpu(str->di_atime); - ip->i_inode.i_atime.tv_nsec = be32_to_cpu(str->di_atime_nsec); + atime.tv_sec = be64_to_cpu(str->di_atime); + atime.tv_nsec = be32_to_cpu(str->di_atime_nsec); + if (timespec_compare(&ip->i_inode.i_atime, &atime) < 0) + ip->i_inode.i_atime = atime; ip->i_inode.i_mtime.tv_sec = be64_to_cpu(str->di_mtime); ip->i_inode.i_mtime.tv_nsec = be32_to_cpu(str->di_mtime_nsec); ip->i_inode.i_ctime.tv_sec = be64_to_cpu(str->di_ctime); @@ -1033,13 +1037,11 @@ struct inode *gfs2_createi(struct gfs2_holder *ghs, const struct qstr *name, if (bh) brelse(bh); - if (!inode) - return ERR_PTR(-ENOMEM); return inode; fail_gunlock2: gfs2_glock_dq_uninit(ghs + 1); - if (inode) + if (inode && !IS_ERR(inode)) iput(inode); fail_gunlock: gfs2_glock_dq(ghs); @@ -1140,54 +1142,6 @@ int gfs2_unlink_ok(struct gfs2_inode *dip, const struct qstr *name, return 0; } -/* - * gfs2_ok_to_move - check if it's ok to move a directory to another directory - * @this: move this - * @to: to here - * - * Follow @to back to the root and make sure we don't encounter @this - * Assumes we already hold the rename lock. - * - * Returns: errno - */ - -int gfs2_ok_to_move(struct gfs2_inode *this, struct gfs2_inode *to) -{ - struct inode *dir = &to->i_inode; - struct super_block *sb = dir->i_sb; - struct inode *tmp; - struct qstr dotdot; - int error = 0; - - gfs2_str2qstr(&dotdot, ".."); - - igrab(dir); - - for (;;) { - if (dir == &this->i_inode) { - error = -EINVAL; - break; - } - if (dir == sb->s_root->d_inode) { - error = 0; - break; - } - - tmp = gfs2_lookupi(dir, &dotdot, 1); - if (IS_ERR(tmp)) { - error = PTR_ERR(tmp); - break; - } - - iput(dir); - dir = tmp; - } - - iput(dir); - - return error; -} - /** * gfs2_readlinki - return the contents of a symlink * @ip: the symlink's inode @@ -1207,8 +1161,8 @@ int gfs2_readlinki(struct gfs2_inode *ip, char **buf, unsigned int *len) unsigned int x; int error; - gfs2_holder_init(ip->i_gl, LM_ST_SHARED, GL_ATIME, &i_gh); - error = gfs2_glock_nq_atime(&i_gh); + gfs2_holder_init(ip->i_gl, LM_ST_SHARED, 0, &i_gh); + error = gfs2_glock_nq(&i_gh); if (error) { gfs2_holder_uninit(&i_gh); return error; @@ -1243,101 +1197,6 @@ out: return error; } -/** - * gfs2_glock_nq_atime - Acquire a hold on an inode's glock, and - * conditionally update the inode's atime - * @gh: the holder to acquire - * - * Tests atime (access time) for gfs2_read, gfs2_readdir and gfs2_mmap - * Update if the difference between the current time and the inode's current - * atime is greater than an interval specified at mount. - * - * Returns: errno - */ - -int gfs2_glock_nq_atime(struct gfs2_holder *gh) -{ - struct gfs2_glock *gl = gh->gh_gl; - struct gfs2_sbd *sdp = gl->gl_sbd; - struct gfs2_inode *ip = gl->gl_object; - s64 quantum = gfs2_tune_get(sdp, gt_atime_quantum); - unsigned int state; - int flags; - int error; - struct timespec tv = CURRENT_TIME; - - if (gfs2_assert_warn(sdp, gh->gh_flags & GL_ATIME) || - gfs2_assert_warn(sdp, !(gh->gh_flags & GL_ASYNC)) || - gfs2_assert_warn(sdp, gl->gl_ops == &gfs2_inode_glops)) - return -EINVAL; - - state = gh->gh_state; - flags = gh->gh_flags; - - error = gfs2_glock_nq(gh); - if (error) - return error; - - if (test_bit(SDF_NOATIME, &sdp->sd_flags) || - (sdp->sd_vfs->s_flags & MS_RDONLY)) - return 0; - - if (tv.tv_sec - ip->i_inode.i_atime.tv_sec >= quantum) { - gfs2_glock_dq(gh); - gfs2_holder_reinit(LM_ST_EXCLUSIVE, gh->gh_flags & ~LM_FLAG_ANY, - gh); - error = gfs2_glock_nq(gh); - if (error) - return error; - - /* Verify that atime hasn't been updated while we were - trying to get exclusive lock. */ - - tv = CURRENT_TIME; - if (tv.tv_sec - ip->i_inode.i_atime.tv_sec >= quantum) { - struct buffer_head *dibh; - struct gfs2_dinode *di; - - error = gfs2_trans_begin(sdp, RES_DINODE, 0); - if (error == -EROFS) - return 0; - if (error) - goto fail; - - error = gfs2_meta_inode_buffer(ip, &dibh); - if (error) - goto fail_end_trans; - - ip->i_inode.i_atime = tv; - - gfs2_trans_add_bh(ip->i_gl, dibh, 1); - di = (struct gfs2_dinode *)dibh->b_data; - di->di_atime = cpu_to_be64(ip->i_inode.i_atime.tv_sec); - di->di_atime_nsec = cpu_to_be32(ip->i_inode.i_atime.tv_nsec); - brelse(dibh); - - gfs2_trans_end(sdp); - } - - /* If someone else has asked for the glock, - unlock and let them have it. Then reacquire - in the original state. */ - if (gfs2_glock_is_blocking(gl)) { - gfs2_glock_dq(gh); - gfs2_holder_reinit(state, flags, gh); - return gfs2_glock_nq(gh); - } - } - - return 0; - -fail_end_trans: - gfs2_trans_end(sdp); -fail: - gfs2_glock_dq(gh); - return error; -} - static int __gfs2_setattr_simple(struct gfs2_inode *ip, struct iattr *attr) { diff --git a/fs/gfs2/inode.h b/fs/gfs2/inode.h index 58f9607d6a8..2d43f69610a 100644 --- a/fs/gfs2/inode.h +++ b/fs/gfs2/inode.h @@ -91,9 +91,7 @@ int gfs2_rmdiri(struct gfs2_inode *dip, const struct qstr *name, int gfs2_unlink_ok(struct gfs2_inode *dip, const struct qstr *name, const struct gfs2_inode *ip); int gfs2_permission(struct inode *inode, int mask); -int gfs2_ok_to_move(struct gfs2_inode *this, struct gfs2_inode *to); int gfs2_readlinki(struct gfs2_inode *ip, char **buf, unsigned int *len); -int gfs2_glock_nq_atime(struct gfs2_holder *gh); int gfs2_setattr_simple(struct gfs2_inode *ip, struct iattr *attr); struct inode *gfs2_lookup_simple(struct inode *dip, const char *name); void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf); diff --git a/fs/gfs2/locking/dlm/mount.c b/fs/gfs2/locking/dlm/mount.c index 09d78c216f4..0c4cbe6c828 100644 --- a/fs/gfs2/locking/dlm/mount.c +++ b/fs/gfs2/locking/dlm/mount.c @@ -144,7 +144,8 @@ static int gdlm_mount(char *table_name, char *host_data, error = dlm_new_lockspace(ls->fsname, strlen(ls->fsname), &ls->dlm_lockspace, - DLM_LSFL_FS | (nodir ? DLM_LSFL_NODIR : 0), + DLM_LSFL_FS | DLM_LSFL_NEWEXCL | + (nodir ? DLM_LSFL_NODIR : 0), GDLM_LVB_SIZE); if (error) { log_error("dlm_new_lockspace error %d", error); diff --git a/fs/gfs2/log.c b/fs/gfs2/log.c index 6c6af9f5e3a..ad305854bdc 100644 --- a/fs/gfs2/log.c +++ b/fs/gfs2/log.c @@ -18,6 +18,7 @@ #include <linux/delay.h> #include <linux/kthread.h> #include <linux/freezer.h> +#include <linux/bio.h> #include "gfs2.h" #include "incore.h" @@ -584,7 +585,6 @@ static void log_write_header(struct gfs2_sbd *sdp, u32 flags, int pull) memset(bh->b_data, 0, bh->b_size); set_buffer_uptodate(bh); clear_buffer_dirty(bh); - unlock_buffer(bh); gfs2_ail1_empty(sdp, 0); tail = current_tail(sdp); @@ -601,8 +601,23 @@ static void log_write_header(struct gfs2_sbd *sdp, u32 flags, int pull) hash = gfs2_disk_hash(bh->b_data, sizeof(struct gfs2_log_header)); lh->lh_hash = cpu_to_be32(hash); - set_buffer_dirty(bh); - if (sync_dirty_buffer(bh)) + bh->b_end_io = end_buffer_write_sync; + if (test_bit(SDF_NOBARRIERS, &sdp->sd_flags)) + goto skip_barrier; + get_bh(bh); + submit_bh(WRITE_BARRIER | (1 << BIO_RW_META), bh); + wait_on_buffer(bh); + if (buffer_eopnotsupp(bh)) { + clear_buffer_eopnotsupp(bh); + set_buffer_uptodate(bh); + set_bit(SDF_NOBARRIERS, &sdp->sd_flags); + lock_buffer(bh); +skip_barrier: + get_bh(bh); + submit_bh(WRITE_SYNC | (1 << BIO_RW_META), bh); + wait_on_buffer(bh); + } + if (!buffer_uptodate(bh)) gfs2_io_error_bh(sdp, bh); brelse(bh); diff --git a/fs/gfs2/mount.c b/fs/gfs2/mount.c index b941f9f9f95..df48333e6f0 100644 --- a/fs/gfs2/mount.c +++ b/fs/gfs2/mount.c @@ -42,6 +42,7 @@ enum { Opt_nosuiddir, Opt_data_writeback, Opt_data_ordered, + Opt_meta, Opt_err, }; @@ -66,6 +67,7 @@ static match_table_t tokens = { {Opt_nosuiddir, "nosuiddir"}, {Opt_data_writeback, "data=writeback"}, {Opt_data_ordered, "data=ordered"}, + {Opt_meta, "meta"}, {Opt_err, NULL} }; @@ -239,6 +241,11 @@ int gfs2_mount_args(struct gfs2_sbd *sdp, char *data_arg, int remount) case Opt_data_ordered: args->ar_data = GFS2_DATA_ORDERED; break; + case Opt_meta: + if (remount && args->ar_meta != 1) + goto cant_remount; + args->ar_meta = 1; + break; case Opt_err: default: fs_info(sdp, "unknown option: %s\n", o); diff --git a/fs/gfs2/ops_address.c b/fs/gfs2/ops_address.c index e64a1b04117..27563816e1c 100644 --- a/fs/gfs2/ops_address.c +++ b/fs/gfs2/ops_address.c @@ -512,8 +512,8 @@ static int gfs2_readpage(struct file *file, struct page *page) int error; unlock_page(page); - gfs2_holder_init(ip->i_gl, LM_ST_SHARED, GL_ATIME, &gh); - error = gfs2_glock_nq_atime(&gh); + gfs2_holder_init(ip->i_gl, LM_ST_SHARED, 0, &gh); + error = gfs2_glock_nq(&gh); if (unlikely(error)) goto out; error = AOP_TRUNCATED_PAGE; @@ -594,8 +594,8 @@ static int gfs2_readpages(struct file *file, struct address_space *mapping, struct gfs2_holder gh; int ret; - gfs2_holder_init(ip->i_gl, LM_ST_SHARED, GL_ATIME, &gh); - ret = gfs2_glock_nq_atime(&gh); + gfs2_holder_init(ip->i_gl, LM_ST_SHARED, 0, &gh); + ret = gfs2_glock_nq(&gh); if (unlikely(ret)) goto out_uninit; if (!gfs2_is_stuffed(ip)) @@ -636,8 +636,8 @@ static int gfs2_write_begin(struct file *file, struct address_space *mapping, unsigned to = from + len; struct page *page; - gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, GL_ATIME, &ip->i_gh); - error = gfs2_glock_nq_atime(&ip->i_gh); + gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &ip->i_gh); + error = gfs2_glock_nq(&ip->i_gh); if (unlikely(error)) goto out_uninit; @@ -975,7 +975,7 @@ static int gfs2_ok_for_dio(struct gfs2_inode *ip, int rw, loff_t offset) if (gfs2_is_stuffed(ip)) return 0; - if (offset > i_size_read(&ip->i_inode)) + if (offset >= i_size_read(&ip->i_inode)) return 0; return 1; } @@ -1000,8 +1000,8 @@ static ssize_t gfs2_direct_IO(int rw, struct kiocb *iocb, * unfortunately have the option of only flushing a range like * the VFS does. */ - gfs2_holder_init(ip->i_gl, LM_ST_DEFERRED, GL_ATIME, &gh); - rv = gfs2_glock_nq_atime(&gh); + gfs2_holder_init(ip->i_gl, LM_ST_DEFERRED, 0, &gh); + rv = gfs2_glock_nq(&gh); if (rv) return rv; rv = gfs2_ok_for_dio(ip, rw, offset); diff --git a/fs/gfs2/ops_file.c b/fs/gfs2/ops_file.c index e9a366d4411..3a747f8e218 100644 --- a/fs/gfs2/ops_file.c +++ b/fs/gfs2/ops_file.c @@ -89,8 +89,8 @@ static int gfs2_readdir(struct file *file, void *dirent, filldir_t filldir) u64 offset = file->f_pos; int error; - gfs2_holder_init(dip->i_gl, LM_ST_SHARED, GL_ATIME, &d_gh); - error = gfs2_glock_nq_atime(&d_gh); + gfs2_holder_init(dip->i_gl, LM_ST_SHARED, 0, &d_gh); + error = gfs2_glock_nq(&d_gh); if (error) { gfs2_holder_uninit(&d_gh); return error; @@ -153,8 +153,8 @@ static int gfs2_get_flags(struct file *filp, u32 __user *ptr) int error; u32 fsflags; - gfs2_holder_init(ip->i_gl, LM_ST_SHARED, GL_ATIME, &gh); - error = gfs2_glock_nq_atime(&gh); + gfs2_holder_init(ip->i_gl, LM_ST_SHARED, 0, &gh); + error = gfs2_glock_nq(&gh); if (error) return error; @@ -351,8 +351,8 @@ static int gfs2_page_mkwrite(struct vm_area_struct *vma, struct page *page) struct gfs2_alloc *al; int ret; - gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, GL_ATIME, &gh); - ret = gfs2_glock_nq_atime(&gh); + gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh); + ret = gfs2_glock_nq(&gh); if (ret) goto out; @@ -434,8 +434,8 @@ static int gfs2_mmap(struct file *file, struct vm_area_struct *vma) struct gfs2_holder i_gh; int error; - gfs2_holder_init(ip->i_gl, LM_ST_SHARED, GL_ATIME, &i_gh); - error = gfs2_glock_nq_atime(&i_gh); + gfs2_holder_init(ip->i_gl, LM_ST_SHARED, 0, &i_gh); + error = gfs2_glock_nq(&i_gh); if (error) { gfs2_holder_uninit(&i_gh); return error; diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c index b4d1d649063..b117fcf2c4f 100644 --- a/fs/gfs2/ops_fstype.c +++ b/fs/gfs2/ops_fstype.c @@ -40,6 +40,44 @@ #define DO 0 #define UNDO 1 +static const u32 gfs2_old_fs_formats[] = { + 0 +}; + +static const u32 gfs2_old_multihost_formats[] = { + 0 +}; + +/** + * gfs2_tune_init - Fill a gfs2_tune structure with default values + * @gt: tune + * + */ + +static void gfs2_tune_init(struct gfs2_tune *gt) +{ + spin_lock_init(>->gt_spin); + + gt->gt_demote_secs = 300; + gt->gt_incore_log_blocks = 1024; + gt->gt_log_flush_secs = 60; + gt->gt_recoverd_secs = 60; + gt->gt_logd_secs = 1; + gt->gt_quotad_secs = 5; + gt->gt_quota_simul_sync = 64; + gt->gt_quota_warn_period = 10; + gt->gt_quota_scale_num = 1; + gt->gt_quota_scale_den = 1; + gt->gt_quota_cache_secs = 300; + gt->gt_quota_quantum = 60; + gt->gt_new_files_jdata = 0; + gt->gt_max_readahead = 1 << 18; + gt->gt_stall_secs = 600; + gt->gt_complain_secs = 10; + gt->gt_statfs_quantum = 30; + gt->gt_statfs_slow = 0; +} + static struct gfs2_sbd *init_sbd(struct super_block *sb) { struct gfs2_sbd *sdp; @@ -96,21 +134,271 @@ static struct gfs2_sbd *init_sbd(struct super_block *sb) return sdp; } -static void init_vfs(struct super_block *sb, unsigned noatime) + +/** + * gfs2_check_sb - Check superblock + * @sdp: the filesystem + * @sb: The superblock + * @silent: Don't print a message if the check fails + * + * Checks the version code of the FS is one that we understand how to + * read and that the sizes of the various on-disk structures have not + * changed. + */ + +static int gfs2_check_sb(struct gfs2_sbd *sdp, struct gfs2_sb_host *sb, int silent) { - struct gfs2_sbd *sdp = sb->s_fs_info; + unsigned int x; - sb->s_magic = GFS2_MAGIC; - sb->s_op = &gfs2_super_ops; - sb->s_export_op = &gfs2_export_ops; - sb->s_time_gran = 1; - sb->s_maxbytes = MAX_LFS_FILESIZE; + if (sb->sb_magic != GFS2_MAGIC || + sb->sb_type != GFS2_METATYPE_SB) { + if (!silent) + printk(KERN_WARNING "GFS2: not a GFS2 filesystem\n"); + return -EINVAL; + } + + /* If format numbers match exactly, we're done. */ + + if (sb->sb_fs_format == GFS2_FORMAT_FS && + sb->sb_multihost_format == GFS2_FORMAT_MULTI) + return 0; + + if (sb->sb_fs_format != GFS2_FORMAT_FS) { + for (x = 0; gfs2_old_fs_formats[x]; x++) + if (gfs2_old_fs_formats[x] == sb->sb_fs_format) + break; + + if (!gfs2_old_fs_formats[x]) { + printk(KERN_WARNING + "GFS2: code version (%u, %u) is incompatible " + "with ondisk format (%u, %u)\n", + GFS2_FORMAT_FS, GFS2_FORMAT_MULTI, + sb->sb_fs_format, sb->sb_multihost_format); + printk(KERN_WARNING + "GFS2: I don't know how to upgrade this FS\n"); + return -EINVAL; + } + } + + if (sb->sb_multihost_format != GFS2_FORMAT_MULTI) { + for (x = 0; gfs2_old_multihost_formats[x]; x++) + if (gfs2_old_multihost_formats[x] == + sb->sb_multihost_format) + break; + + if (!gfs2_old_multihost_formats[x]) { + printk(KERN_WARNING + "GFS2: code version (%u, %u) is incompatible " + "with ondisk format (%u, %u)\n", + GFS2_FORMAT_FS, GFS2_FORMAT_MULTI, + sb->sb_fs_format, sb->sb_multihost_format); + printk(KERN_WARNING + "GFS2: I don't know how to upgrade this FS\n"); + return -EINVAL; + } + } + + if (!sdp->sd_args.ar_upgrade) { + printk(KERN_WARNING + "GFS2: code version (%u, %u) is incompatible " + "with ondisk format (%u, %u)\n", + GFS2_FORMAT_FS, GFS2_FORMAT_MULTI, + sb->sb_fs_format, sb->sb_multihost_format); + printk(KERN_INFO + "GFS2: Use the \"upgrade\" mount option to upgrade " + "the FS\n"); + printk(KERN_INFO "GFS2: See the manual for more details\n"); + return -EINVAL; + } + + return 0; +} + +static void end_bio_io_page(struct bio *bio, int error) +{ + struct page *page = bio->bi_private; - if (sb->s_flags & (MS_NOATIME | MS_NODIRATIME)) - set_bit(noatime, &sdp->sd_flags); + if (!error) + SetPageUptodate(page); + else + printk(KERN_WARNING "gfs2: error %d reading superblock\n", error); + unlock_page(page); +} + +static void gfs2_sb_in(struct gfs2_sb_host *sb, const void *buf) +{ + const struct gfs2_sb *str = buf; + + sb->sb_magic = be32_to_cpu(str->sb_header.mh_magic); + sb->sb_type = be32_to_cpu(str->sb_header.mh_type); + sb->sb_format = be32_to_cpu(str->sb_header.mh_format); + sb->sb_fs_format = be32_to_cpu(str->sb_fs_format); + sb->sb_multihost_format = be32_to_cpu(str->sb_multihost_format); + sb->sb_bsize = be32_to_cpu(str->sb_bsize); + sb->sb_bsize_shift = be32_to_cpu(str->sb_bsize_shift); + sb->sb_master_dir.no_addr = be64_to_cpu(str->sb_master_dir.no_addr); + sb->sb_master_dir.no_formal_ino = be64_to_cpu(str->sb_master_dir.no_formal_ino); + sb->sb_root_dir.no_addr = be64_to_cpu(str->sb_root_dir.no_addr); + sb->sb_root_dir.no_formal_ino = be64_to_cpu(str->sb_root_dir.no_formal_ino); + + memcpy(sb->sb_lockproto, str->sb_lockproto, GFS2_LOCKNAME_LEN); + memcpy(sb->sb_locktable, str->sb_locktable, GFS2_LOCKNAME_LEN); +} + +/** + * gfs2_read_super - Read the gfs2 super block from disk + * @sdp: The GFS2 super block + * @sector: The location of the super block + * @error: The error code to return + * + * This uses the bio functions to read the super block from disk + * because we want to be 100% sure that we never read cached data. + * A super block is read twice only during each GFS2 mount and is + * never written to by the filesystem. The first time its read no + * locks are held, and the only details which are looked at are those + * relating to the locking protocol. Once locking is up and working, + * the sb is read again under the lock to establish the location of + * the master directory (contains pointers to journals etc) and the + * root directory. + * + * Returns: 0 on success or error + */ + +static int gfs2_read_super(struct gfs2_sbd *sdp, sector_t sector) +{ + struct super_block *sb = sdp->sd_vfs; + struct gfs2_sb *p; + struct page *page; + struct bio *bio; + + page = alloc_page(GFP_NOFS); + if (unlikely(!page)) + return -ENOBUFS; + + ClearPageUptodate(page); + ClearPageDirty(page); + lock_page(page); + + bio = bio_alloc(GFP_NOFS, 1); + if (unlikely(!bio)) { + __free_page(page); + return -ENOBUFS; + } - /* Don't let the VFS update atimes. GFS2 handles this itself. */ - sb->s_flags |= MS_NOATIME | MS_NODIRATIME; + bio->bi_sector = sector * (sb->s_blocksize >> 9); + bio->bi_bdev = sb->s_bdev; + bio_add_page(bio, page, PAGE_SIZE, 0); + + bio->bi_end_io = end_bio_io_page; + bio->bi_private = page; + submit_bio(READ_SYNC | (1 << BIO_RW_META), bio); + wait_on_page_locked(page); + bio_put(bio); + if (!PageUptodate(page)) { + __free_page(page); + return -EIO; + } + p = kmap(page); + gfs2_sb_in(&sdp->sd_sb, p); + kunmap(page); + __free_page(page); + return 0; +} +/** + * gfs2_read_sb - Read super block + * @sdp: The GFS2 superblock + * @gl: the glock for the superblock (assumed to be held) + * @silent: Don't print message if mount fails + * + */ + +static int gfs2_read_sb(struct gfs2_sbd *sdp, struct gfs2_glock *gl, int silent) +{ + u32 hash_blocks, ind_blocks, leaf_blocks; + u32 tmp_blocks; + unsigned int x; + int error; + + error = gfs2_read_super(sdp, GFS2_SB_ADDR >> sdp->sd_fsb2bb_shift); + if (error) { + if (!silent) + fs_err(sdp, "can't read superblock\n"); + return error; + } + + error = gfs2_check_sb(sdp, &sdp->sd_sb, silent); + if (error) + return error; + + sdp->sd_fsb2bb_shift = sdp->sd_sb.sb_bsize_shift - + GFS2_BASIC_BLOCK_SHIFT; + sdp->sd_fsb2bb = 1 << sdp->sd_fsb2bb_shift; + sdp->sd_diptrs = (sdp->sd_sb.sb_bsize - + sizeof(struct gfs2_dinode)) / sizeof(u64); + sdp->sd_inptrs = (sdp->sd_sb.sb_bsize - + sizeof(struct gfs2_meta_header)) / sizeof(u64); + sdp->sd_jbsize = sdp->sd_sb.sb_bsize - sizeof(struct gfs2_meta_header); + sdp->sd_hash_bsize = sdp->sd_sb.sb_bsize / 2; + sdp->sd_hash_bsize_shift = sdp->sd_sb.sb_bsize_shift - 1; + sdp->sd_hash_ptrs = sdp->sd_hash_bsize / sizeof(u64); + sdp->sd_qc_per_block = (sdp->sd_sb.sb_bsize - + sizeof(struct gfs2_meta_header)) / + sizeof(struct gfs2_quota_change); + + /* Compute maximum reservation required to add a entry to a directory */ + + hash_blocks = DIV_ROUND_UP(sizeof(u64) * (1 << GFS2_DIR_MAX_DEPTH), + sdp->sd_jbsize); + + ind_blocks = 0; + for (tmp_blocks = hash_blocks; tmp_blocks > sdp->sd_diptrs;) { + tmp_blocks = DIV_ROUND_UP(tmp_blocks, sdp->sd_inptrs); + ind_blocks += tmp_blocks; + } + + leaf_blocks = 2 + GFS2_DIR_MAX_DEPTH; + + sdp->sd_max_dirres = hash_blocks + ind_blocks + leaf_blocks; + + sdp->sd_heightsize[0] = sdp->sd_sb.sb_bsize - + sizeof(struct gfs2_dinode); + sdp->sd_heightsize[1] = sdp->sd_sb.sb_bsize * sdp->sd_diptrs; + for (x = 2;; x++) { + u64 space, d; + u32 m; + + space = sdp->sd_heightsize[x - 1] * sdp->sd_inptrs; + d = space; + m = do_div(d, sdp->sd_inptrs); + + if (d != sdp->sd_heightsize[x - 1] || m) + break; + sdp->sd_heightsize[x] = space; + } + sdp->sd_max_height = x; + sdp->sd_heightsize[x] = ~0; + gfs2_assert(sdp, sdp->sd_max_height <= GFS2_MAX_META_HEIGHT); + + sdp->sd_jheightsize[0] = sdp->sd_sb.sb_bsize - + sizeof(struct gfs2_dinode); + sdp->sd_jheightsize[1] = sdp->sd_jbsize * sdp->sd_diptrs; + for (x = 2;; x++) { + u64 space, d; + u32 m; + + space = sdp->sd_jheightsize[x - 1] * sdp->sd_inptrs; + d = space; + m = do_div(d, sdp->sd_inptrs); + + if (d != sdp->sd_jheightsize[x - 1] || m) + break; + sdp->sd_jheightsize[x] = space; + } + sdp->sd_max_jheight = x; + sdp->sd_jheightsize[x] = ~0; + gfs2_assert(sdp, sdp->sd_max_jheight <= GFS2_MAX_META_HEIGHT); + + return 0; } static int init_names(struct gfs2_sbd *sdp, int silent) @@ -224,51 +512,59 @@ fail: return error; } -static inline struct inode *gfs2_lookup_root(struct super_block *sb, - u64 no_addr) +static int gfs2_lookup_root(struct super_block *sb, struct dentry **dptr, + u64 no_addr, const char *name) { - return gfs2_inode_lookup(sb, DT_DIR, no_addr, 0, 0); + struct gfs2_sbd *sdp = sb->s_fs_info; + struct dentry *dentry; + struct inode *inode; + + inode = gfs2_inode_lookup(sb, DT_DIR, no_addr, 0, 0); + if (IS_ERR(inode)) { + fs_err(sdp, "can't read in %s inode: %ld\n", name, PTR_ERR(inode)); + return PTR_ERR(inode); + } + dentry = d_alloc_root(inode); + if (!dentry) { + fs_err(sdp, "can't alloc %s dentry\n", name); + iput(inode); + return -ENOMEM; + } + dentry->d_op = &gfs2_dops; + *dptr = dentry; + return 0; } -static int init_sb(struct gfs2_sbd *sdp, int silent, int undo) +static int init_sb(struct gfs2_sbd *sdp, int silent) { struct super_block *sb = sdp->sd_vfs; struct gfs2_holder sb_gh; u64 no_addr; - struct inode *inode; - int error = 0; + int ret; - if (undo) { - if (sb->s_root) { - dput(sb->s_root); - sb->s_root = NULL; - } - return 0; + ret = gfs2_glock_nq_num(sdp, GFS2_SB_LOCK, &gfs2_meta_glops, + LM_ST_SHARED, 0, &sb_gh); + if (ret) { + fs_err(sdp, "can't acquire superblock glock: %d\n", ret); + return ret; } - error = gfs2_glock_nq_num(sdp, GFS2_SB_LOCK, &gfs2_meta_glops, - LM_ST_SHARED, 0, &sb_gh); - if (error) { - fs_err(sdp, "can't acquire superblock glock: %d\n", error); - return error; - } - - error = gfs2_read_sb(sdp, sb_gh.gh_gl, silent); - if (error) { - fs_err(sdp, "can't read superblock: %d\n", error); + ret = gfs2_read_sb(sdp, sb_gh.gh_gl, silent); + if (ret) { + fs_err(sdp, "can't read superblock: %d\n", ret); goto out; } /* Set up the buffer cache and SB for real */ if (sdp->sd_sb.sb_bsize < bdev_hardsect_size(sb->s_bdev)) { - error = -EINVAL; + ret = -EINVAL; fs_err(sdp, "FS block size (%u) is too small for device " "block size (%u)\n", sdp->sd_sb.sb_bsize, bdev_hardsect_size(sb->s_bdev)); goto out; } if (sdp->sd_sb.sb_bsize > PAGE_SIZE) { - error = -EINVAL; + ret = -EINVAL; fs_err(sdp, "FS block size (%u) is too big for machine " "page size (%u)\n", sdp->sd_sb.sb_bsize, (unsigned int)PAGE_SIZE); @@ -278,26 +574,21 @@ static int init_sb(struct gfs2_sbd *sdp, int silent, int undo) /* Get the root inode */ no_addr = sdp->sd_sb.sb_root_dir.no_addr; - if (sb->s_type == &gfs2meta_fs_type) - no_addr = sdp->sd_sb.sb_master_dir.no_addr; - inode = gfs2_lookup_root(sb, no_addr); - if (IS_ERR(inode)) { - error = PTR_ERR(inode); - fs_err(sdp, "can't read in root inode: %d\n", error); + ret = gfs2_lookup_root(sb, &sdp->sd_root_dir, no_addr, "root"); + if (ret) goto out; - } - sb->s_root = d_alloc_root(inode); - if (!sb->s_root) { - fs_err(sdp, "can't get root dentry\n"); - error = -ENOMEM; - iput(inode); - } else - sb->s_root->d_op = &gfs2_dops; - + /* Get the master inode */ + no_addr = sdp->sd_sb.sb_master_dir.no_addr; + ret = gfs2_lookup_root(sb, &sdp->sd_master_dir, no_addr, "master"); + if (ret) { + dput(sdp->sd_root_dir); + goto out; + } + sb->s_root = dget(sdp->sd_args.ar_meta ? sdp->sd_master_dir : sdp->sd_root_dir); out: gfs2_glock_dq_uninit(&sb_gh); - return error; + return ret; } /** @@ -372,6 +663,7 @@ static void gfs2_lm_others_may_mount(struct gfs2_sbd *sdp) static int init_journal(struct gfs2_sbd *sdp, int undo) { + struct inode *master = sdp->sd_master_dir->d_inode; struct gfs2_holder ji_gh; struct task_struct *p; struct gfs2_inode *ip; @@ -383,7 +675,7 @@ static int init_journal(struct gfs2_sbd *sdp, int undo) goto fail_recoverd; } - sdp->sd_jindex = gfs2_lookup_simple(sdp->sd_master_dir, "jindex"); + sdp->sd_jindex = gfs2_lookup_simple(master, "jindex"); if (IS_ERR(sdp->sd_jindex)) { fs_err(sdp, "can't lookup journal index: %d\n", error); return PTR_ERR(sdp->sd_jindex); @@ -506,25 +798,17 @@ static int init_inodes(struct gfs2_sbd *sdp, int undo) { int error = 0; struct gfs2_inode *ip; - struct inode *inode; + struct inode *master = sdp->sd_master_dir->d_inode; if (undo) goto fail_qinode; - inode = gfs2_lookup_root(sdp->sd_vfs, sdp->sd_sb.sb_master_dir.no_addr); - if (IS_ERR(inode)) { - error = PTR_ERR(inode); - fs_err(sdp, "can't read in master directory: %d\n", error); - goto fail; - } - sdp->sd_master_dir = inode; - error = init_journal(sdp, undo); if (error) - goto fail_master; + goto fail; /* Read in the master inode number inode */ - sdp->sd_inum_inode = gfs2_lookup_simple(sdp->sd_master_dir, "inum"); + sdp->sd_inum_inode = gfs2_lookup_simple(master, "inum"); if (IS_ERR(sdp->sd_inum_inode)) { error = PTR_ERR(sdp->sd_inum_inode); fs_err(sdp, "can't read in inum inode: %d\n", error); @@ -533,7 +817,7 @@ static int init_inodes(struct gfs2_sbd *sdp, int undo) /* Read in the master statfs inode */ - sdp->sd_statfs_inode = gfs2_lookup_simple(sdp->sd_master_dir, "statfs"); + sdp->sd_statfs_inode = gfs2_lookup_simple(master, "statfs"); if (IS_ERR(sdp->sd_statfs_inode)) { error = PTR_ERR(sdp->sd_statfs_inode); fs_err(sdp, "can't read in statfs inode: %d\n", error); @@ -541,7 +825,7 @@ static int init_inodes(struct gfs2_sbd *sdp, int undo) } /* Read in the resource index inode */ - sdp->sd_rindex = gfs2_lookup_simple(sdp->sd_master_dir, "rindex"); + sdp->sd_rindex = gfs2_lookup_simple(master, "rindex"); if (IS_ERR(sdp->sd_rindex)) { error = PTR_ERR(sdp->sd_rindex); fs_err(sdp, "can't get resource index inode: %d\n", error); @@ -552,7 +836,7 @@ static int init_inodes(struct gfs2_sbd *sdp, int undo) sdp->sd_rindex_uptodate = 0; /* Read in the quota inode */ - sdp->sd_quota_inode = gfs2_lookup_simple(sdp->sd_master_dir, "quota"); + sdp->sd_quota_inode = gfs2_lookup_simple(master, "quota"); if (IS_ERR(sdp->sd_quota_inode)) { error = PTR_ERR(sdp->sd_quota_inode); fs_err(sdp, "can't get quota file inode: %d\n", error); @@ -571,8 +855,6 @@ fail_inum: iput(sdp->sd_inum_inode); fail_journal: init_journal(sdp, UNDO); -fail_master: - iput(sdp->sd_master_dir); fail: return error; } @@ -583,6 +865,7 @@ static int init_per_node(struct gfs2_sbd *sdp, int undo) char buf[30]; int error = 0; struct gfs2_inode *ip; + struct inode *master = sdp->sd_master_dir->d_inode; if (sdp->sd_args.ar_spectator) return 0; @@ -590,7 +873,7 @@ static int init_per_node(struct gfs2_sbd *sdp, int undo) if (undo) goto fail_qc_gh; - pn = gfs2_lookup_simple(sdp->sd_master_dir, "per_node"); + pn = gfs2_lookup_simple(master, "per_node"); if (IS_ERR(pn)) { error = PTR_ERR(pn); fs_err(sdp, "can't find per_node directory: %d\n", error); @@ -800,7 +1083,11 @@ static int fill_super(struct super_block *sb, void *data, int silent) goto fail; } - init_vfs(sb, SDF_NOATIME); + sb->s_magic = GFS2_MAGIC; + sb->s_op = &gfs2_super_ops; + sb->s_export_op = &gfs2_export_ops; + sb->s_time_gran = 1; + sb->s_maxbytes = MAX_LFS_FILESIZE; /* Set up the buffer cache and fill in some fake block size values to allow us to read-in the on-disk superblock. */ @@ -828,7 +1115,7 @@ static int fill_super(struct super_block *sb, void *data, int silent) if (error) goto fail_lm; - error = init_sb(sdp, silent, DO); + error = init_sb(sdp, silent); if (error) goto fail_locking; @@ -869,7 +1156,11 @@ fail_per_node: fail_inodes: init_inodes(sdp, UNDO); fail_sb: - init_sb(sdp, 0, UNDO); + if (sdp->sd_root_dir) + dput(sdp->sd_root_dir); + if (sdp->sd_master_dir) + dput(sdp->sd_master_dir); + sb->s_root = NULL; fail_locking: init_locking(sdp, &mount_gh, UNDO); fail_lm: @@ -887,151 +1178,63 @@ fail: } static int gfs2_get_sb(struct file_system_type *fs_type, int flags, - const char *dev_name, void *data, struct vfsmount *mnt) + const char *dev_name, void *data, struct vfsmount *mnt) { - struct super_block *sb; - struct gfs2_sbd *sdp; - int error = get_sb_bdev(fs_type, flags, dev_name, data, fill_super, mnt); - if (error) - goto out; - sb = mnt->mnt_sb; - sdp = sb->s_fs_info; - sdp->sd_gfs2mnt = mnt; -out: - return error; + return get_sb_bdev(fs_type, flags, dev_name, data, fill_super, mnt); } -static int fill_super_meta(struct super_block *sb, struct super_block *new, - void *data, int silent) +static struct super_block *get_gfs2_sb(const char *dev_name) { - struct gfs2_sbd *sdp = sb->s_fs_info; - struct inode *inode; - int error = 0; - - new->s_fs_info = sdp; - sdp->sd_vfs_meta = sb; - - init_vfs(new, SDF_NOATIME); - - /* Get the master inode */ - inode = igrab(sdp->sd_master_dir); - - new->s_root = d_alloc_root(inode); - if (!new->s_root) { - fs_err(sdp, "can't get root dentry\n"); - error = -ENOMEM; - iput(inode); - } else - new->s_root->d_op = &gfs2_dops; - - return error; -} - -static int set_bdev_super(struct super_block *s, void *data) -{ - s->s_bdev = data; - s->s_dev = s->s_bdev->bd_dev; - return 0; -} - -static int test_bdev_super(struct super_block *s, void *data) -{ - return s->s_bdev == data; -} - -static struct super_block* get_gfs2_sb(const char *dev_name) -{ - struct kstat stat; + struct super_block *sb; struct nameidata nd; - struct super_block *sb = NULL, *s; int error; error = path_lookup(dev_name, LOOKUP_FOLLOW, &nd); if (error) { - printk(KERN_WARNING "GFS2: path_lookup on %s returned error\n", - dev_name); - goto out; - } - error = vfs_getattr(nd.path.mnt, nd.path.dentry, &stat); - - list_for_each_entry(s, &gfs2_fs_type.fs_supers, s_instances) { - if ((S_ISBLK(stat.mode) && s->s_dev == stat.rdev) || - (S_ISDIR(stat.mode) && - s == nd.path.dentry->d_inode->i_sb)) { - sb = s; - goto free_nd; - } + printk(KERN_WARNING "GFS2: path_lookup on %s returned error %d\n", + dev_name, error); + return NULL; } - - printk(KERN_WARNING "GFS2: Unrecognized block device or " - "mount point %s\n", dev_name); - -free_nd: + sb = nd.path.dentry->d_inode->i_sb; + if (sb && (sb->s_type == &gfs2_fs_type)) + atomic_inc(&sb->s_active); + else + sb = NULL; path_put(&nd.path); -out: return sb; } static int gfs2_get_sb_meta(struct file_system_type *fs_type, int flags, const char *dev_name, void *data, struct vfsmount *mnt) { - int error = 0; - struct super_block *sb = NULL, *new; + struct super_block *sb = NULL; struct gfs2_sbd *sdp; sb = get_gfs2_sb(dev_name); if (!sb) { printk(KERN_WARNING "GFS2: gfs2 mount does not exist\n"); - error = -ENOENT; - goto error; + return -ENOENT; } sdp = sb->s_fs_info; - if (sdp->sd_vfs_meta) { - printk(KERN_WARNING "GFS2: gfs2meta mount already exists\n"); - error = -EBUSY; - goto error; - } - down(&sb->s_bdev->bd_mount_sem); - new = sget(fs_type, test_bdev_super, set_bdev_super, sb->s_bdev); - up(&sb->s_bdev->bd_mount_sem); - if (IS_ERR(new)) { - error = PTR_ERR(new); - goto error; - } - new->s_flags = flags; - strlcpy(new->s_id, sb->s_id, sizeof(new->s_id)); - sb_set_blocksize(new, sb->s_blocksize); - error = fill_super_meta(sb, new, data, flags & MS_SILENT ? 1 : 0); - if (error) { - up_write(&new->s_umount); - deactivate_super(new); - goto error; - } - - new->s_flags |= MS_ACTIVE; - - /* Grab a reference to the gfs2 mount point */ - atomic_inc(&sdp->sd_gfs2mnt->mnt_count); - return simple_set_mnt(mnt, new); -error: - return error; + mnt->mnt_sb = sb; + mnt->mnt_root = dget(sdp->sd_master_dir); + return 0; } static void gfs2_kill_sb(struct super_block *sb) { - if (sb->s_fs_info) { - gfs2_delete_debugfs_file(sb->s_fs_info); - gfs2_meta_syncfs(sb->s_fs_info); + struct gfs2_sbd *sdp = sb->s_fs_info; + if (sdp) { + gfs2_meta_syncfs(sdp); + dput(sdp->sd_root_dir); + dput(sdp->sd_master_dir); + sdp->sd_root_dir = NULL; + sdp->sd_master_dir = NULL; } + shrink_dcache_sb(sb); kill_block_super(sb); -} - -static void gfs2_kill_sb_meta(struct super_block *sb) -{ - struct gfs2_sbd *sdp = sb->s_fs_info; - generic_shutdown_super(sb); - sdp->sd_vfs_meta = NULL; - atomic_dec(&sdp->sd_gfs2mnt->mnt_count); + if (sdp) + gfs2_delete_debugfs_file(sdp); } struct file_system_type gfs2_fs_type = { @@ -1046,7 +1249,6 @@ struct file_system_type gfs2meta_fs_type = { .name = "gfs2meta", .fs_flags = FS_REQUIRES_DEV, .get_sb = gfs2_get_sb_meta, - .kill_sb = gfs2_kill_sb_meta, .owner = THIS_MODULE, }; diff --git a/fs/gfs2/ops_inode.c b/fs/gfs2/ops_inode.c index e2c62f73a77..534e1e2c65c 100644 --- a/fs/gfs2/ops_inode.c +++ b/fs/gfs2/ops_inode.c @@ -159,9 +159,13 @@ static int gfs2_link(struct dentry *old_dentry, struct inode *dir, gfs2_holder_init(dip->i_gl, LM_ST_EXCLUSIVE, 0, ghs); gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, ghs + 1); - error = gfs2_glock_nq_m(2, ghs); + error = gfs2_glock_nq(ghs); /* parent */ if (error) - goto out; + goto out_parent; + + error = gfs2_glock_nq(ghs + 1); /* child */ + if (error) + goto out_child; error = gfs2_permission(dir, MAY_WRITE | MAY_EXEC); if (error) @@ -245,8 +249,10 @@ out_alloc: if (alloc_required) gfs2_alloc_put(dip); out_gunlock: - gfs2_glock_dq_m(2, ghs); -out: + gfs2_glock_dq(ghs + 1); +out_child: + gfs2_glock_dq(ghs); +out_parent: gfs2_holder_uninit(ghs); gfs2_holder_uninit(ghs + 1); if (!error) { @@ -302,7 +308,7 @@ static int gfs2_unlink(struct inode *dir, struct dentry *dentry) error = gfs2_unlink_ok(dip, &dentry->d_name, ip); if (error) - goto out_rgrp; + goto out_gunlock; error = gfs2_trans_begin(sdp, 2*RES_DINODE + RES_LEAF + RES_RG_BIT, 0); if (error) @@ -316,6 +322,7 @@ static int gfs2_unlink(struct inode *dir, struct dentry *dentry) out_end_trans: gfs2_trans_end(sdp); +out_gunlock: gfs2_glock_dq(ghs + 2); out_rgrp: gfs2_holder_uninit(ghs + 2); @@ -485,7 +492,6 @@ static int gfs2_rmdir(struct inode *dir, struct dentry *dentry) struct gfs2_holder ri_gh; int error; - error = gfs2_rindex_hold(sdp, &ri_gh); if (error) return error; @@ -495,9 +501,17 @@ static int gfs2_rmdir(struct inode *dir, struct dentry *dentry) rgd = gfs2_blk2rgrpd(sdp, ip->i_no_addr); gfs2_holder_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0, ghs + 2); - error = gfs2_glock_nq_m(3, ghs); + error = gfs2_glock_nq(ghs); /* parent */ if (error) - goto out; + goto out_parent; + + error = gfs2_glock_nq(ghs + 1); /* child */ + if (error) + goto out_child; + + error = gfs2_glock_nq(ghs + 2); /* rgrp */ + if (error) + goto out_rgrp; error = gfs2_unlink_ok(dip, &dentry->d_name, ip); if (error) @@ -523,11 +537,15 @@ static int gfs2_rmdir(struct inode *dir, struct dentry *dentry) gfs2_trans_end(sdp); out_gunlock: - gfs2_glock_dq_m(3, ghs); -out: - gfs2_holder_uninit(ghs); - gfs2_holder_uninit(ghs + 1); + gfs2_glock_dq(ghs + 2); +out_rgrp: gfs2_holder_uninit(ghs + 2); + gfs2_glock_dq(ghs + 1); +out_child: + gfs2_holder_uninit(ghs + 1); + gfs2_glock_dq(ghs); +out_parent: + gfs2_holder_uninit(ghs); gfs2_glock_dq_uninit(&ri_gh); return error; } @@ -571,6 +589,54 @@ static int gfs2_mknod(struct inode *dir, struct dentry *dentry, int mode, return 0; } +/* + * gfs2_ok_to_move - check if it's ok to move a directory to another directory + * @this: move this + * @to: to here + * + * Follow @to back to the root and make sure we don't encounter @this + * Assumes we already hold the rename lock. + * + * Returns: errno + */ + +static int gfs2_ok_to_move(struct gfs2_inode *this, struct gfs2_inode *to) +{ + struct inode *dir = &to->i_inode; + struct super_block *sb = dir->i_sb; + struct inode *tmp; + struct qstr dotdot; + int error = 0; + + gfs2_str2qstr(&dotdot, ".."); + + igrab(dir); + + for (;;) { + if (dir == &this->i_inode) { + error = -EINVAL; + break; + } + if (dir == sb->s_root->d_inode) { + error = 0; + break; + } + + tmp = gfs2_lookupi(dir, &dotdot, 1); + if (IS_ERR(tmp)) { + error = PTR_ERR(tmp); + break; + } + + iput(dir); + dir = tmp; + } + + iput(dir); + + return error; +} + /** * gfs2_rename - Rename a file * @odir: Parent directory of old file name @@ -589,7 +655,7 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry, struct gfs2_inode *ip = GFS2_I(odentry->d_inode); struct gfs2_inode *nip = NULL; struct gfs2_sbd *sdp = GFS2_SB(odir); - struct gfs2_holder ghs[5], r_gh; + struct gfs2_holder ghs[5], r_gh = { .gh_gl = NULL, }; struct gfs2_rgrpd *nrgd; unsigned int num_gh; int dir_rename = 0; @@ -603,19 +669,20 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry, return 0; } - /* Make sure we aren't trying to move a dirctory into it's subdir */ - - if (S_ISDIR(ip->i_inode.i_mode) && odip != ndip) { - dir_rename = 1; - error = gfs2_glock_nq_init(sdp->sd_rename_gl, LM_ST_EXCLUSIVE, 0, - &r_gh); + if (odip != ndip) { + error = gfs2_glock_nq_init(sdp->sd_rename_gl, LM_ST_EXCLUSIVE, + 0, &r_gh); if (error) goto out; - error = gfs2_ok_to_move(ip, ndip); - if (error) - goto out_gunlock_r; + if (S_ISDIR(ip->i_inode.i_mode)) { + dir_rename = 1; + /* don't move a dirctory into it's subdir */ + error = gfs2_ok_to_move(ip, ndip); + if (error) + goto out_gunlock_r; + } } num_gh = 1; @@ -639,9 +706,11 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry, gfs2_holder_init(nrgd->rd_gl, LM_ST_EXCLUSIVE, 0, ghs + num_gh++); } - error = gfs2_glock_nq_m(num_gh, ghs); - if (error) - goto out_uninit; + for (x = 0; x < num_gh; x++) { + error = gfs2_glock_nq(ghs + x); + if (error) + goto out_gunlock; + } /* Check out the old directory */ @@ -804,12 +873,12 @@ out_alloc: if (alloc_required) gfs2_alloc_put(ndip); out_gunlock: - gfs2_glock_dq_m(num_gh, ghs); -out_uninit: - for (x = 0; x < num_gh; x++) + while (x--) { + gfs2_glock_dq(ghs + x); gfs2_holder_uninit(ghs + x); + } out_gunlock_r: - if (dir_rename) + if (r_gh.gh_gl) gfs2_glock_dq_uninit(&r_gh); out: return error; diff --git a/fs/gfs2/ops_super.c b/fs/gfs2/ops_super.c index f66ea0f7a35..d5355d9b592 100644 --- a/fs/gfs2/ops_super.c +++ b/fs/gfs2/ops_super.c @@ -20,6 +20,7 @@ #include <linux/gfs2_ondisk.h> #include <linux/crc32.h> #include <linux/lm_interface.h> +#include <linux/time.h> #include "gfs2.h" #include "incore.h" @@ -38,6 +39,7 @@ #include "dir.h" #include "eattr.h" #include "bmap.h" +#include "meta_io.h" /** * gfs2_write_inode - Make sure the inode is stable on the disk @@ -50,16 +52,74 @@ static int gfs2_write_inode(struct inode *inode, int sync) { struct gfs2_inode *ip = GFS2_I(inode); - - /* Check this is a "normal" inode */ - if (test_bit(GIF_USER, &ip->i_flags)) { - if (current->flags & PF_MEMALLOC) - return 0; - if (sync) - gfs2_log_flush(GFS2_SB(inode), ip->i_gl); + struct gfs2_sbd *sdp = GFS2_SB(inode); + struct gfs2_holder gh; + struct buffer_head *bh; + struct timespec atime; + struct gfs2_dinode *di; + int ret = 0; + + /* Check this is a "normal" inode, etc */ + if (!test_bit(GIF_USER, &ip->i_flags) || + (current->flags & PF_MEMALLOC)) + return 0; + ret = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh); + if (ret) + goto do_flush; + ret = gfs2_trans_begin(sdp, RES_DINODE, 0); + if (ret) + goto do_unlock; + ret = gfs2_meta_inode_buffer(ip, &bh); + if (ret == 0) { + di = (struct gfs2_dinode *)bh->b_data; + atime.tv_sec = be64_to_cpu(di->di_atime); + atime.tv_nsec = be32_to_cpu(di->di_atime_nsec); + if (timespec_compare(&inode->i_atime, &atime) > 0) { + gfs2_trans_add_bh(ip->i_gl, bh, 1); + gfs2_dinode_out(ip, bh->b_data); + } + brelse(bh); } + gfs2_trans_end(sdp); +do_unlock: + gfs2_glock_dq_uninit(&gh); +do_flush: + if (sync != 0) + gfs2_log_flush(GFS2_SB(inode), ip->i_gl); + return ret; +} - return 0; +/** + * gfs2_make_fs_ro - Turn a Read-Write FS into a Read-Only one + * @sdp: the filesystem + * + * Returns: errno + */ + +static int gfs2_make_fs_ro(struct gfs2_sbd *sdp) +{ + struct gfs2_holder t_gh; + int error; + + gfs2_quota_sync(sdp); + gfs2_statfs_sync(sdp); + + error = gfs2_glock_nq_init(sdp->sd_trans_gl, LM_ST_SHARED, GL_NOCACHE, + &t_gh); + if (error && !test_bit(SDF_SHUTDOWN, &sdp->sd_flags)) + return error; + + gfs2_meta_syncfs(sdp); + gfs2_log_shutdown(sdp); + + clear_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags); + + if (t_gh.gh_gl) + gfs2_glock_dq_uninit(&t_gh); + + gfs2_quota_cleanup(sdp); + + return error; } /** @@ -73,12 +133,6 @@ static void gfs2_put_super(struct super_block *sb) struct gfs2_sbd *sdp = sb->s_fs_info; int error; - if (!sdp) - return; - - if (!strncmp(sb->s_type->name, "gfs2meta", 8)) - return; /* Nothing to do */ - /* Unfreeze the filesystem, if we need to */ mutex_lock(&sdp->sd_freeze_lock); @@ -101,7 +155,6 @@ static void gfs2_put_super(struct super_block *sb) /* Release stuff */ - iput(sdp->sd_master_dir); iput(sdp->sd_jindex); iput(sdp->sd_inum_inode); iput(sdp->sd_statfs_inode); @@ -152,6 +205,7 @@ static void gfs2_write_super(struct super_block *sb) * * Flushes the log to disk. */ + static int gfs2_sync_fs(struct super_block *sb, int wait) { sb->s_dirt = 0; @@ -270,14 +324,6 @@ static int gfs2_remount_fs(struct super_block *sb, int *flags, char *data) } } - if (*flags & (MS_NOATIME | MS_NODIRATIME)) - set_bit(SDF_NOATIME, &sdp->sd_flags); - else - clear_bit(SDF_NOATIME, &sdp->sd_flags); - - /* Don't let the VFS update atimes. GFS2 handles this itself. */ - *flags |= MS_NOATIME | MS_NODIRATIME; - return error; } @@ -295,6 +341,7 @@ static int gfs2_remount_fs(struct super_block *sb, int *flags, char *data) * inode's blocks, or alternatively pass the baton on to another * node for later deallocation. */ + static void gfs2_drop_inode(struct inode *inode) { struct gfs2_inode *ip = GFS2_I(inode); @@ -333,6 +380,16 @@ static void gfs2_clear_inode(struct inode *inode) } } +static int is_ancestor(const struct dentry *d1, const struct dentry *d2) +{ + do { + if (d1 == d2) + return 1; + d1 = d1->d_parent; + } while (!IS_ROOT(d1)); + return 0; +} + /** * gfs2_show_options - Show mount options for /proc/mounts * @s: seq_file structure @@ -346,6 +403,8 @@ static int gfs2_show_options(struct seq_file *s, struct vfsmount *mnt) struct gfs2_sbd *sdp = mnt->mnt_sb->s_fs_info; struct gfs2_args *args = &sdp->sd_args; + if (is_ancestor(mnt->mnt_root, sdp->sd_master_dir)) + seq_printf(s, ",meta"); if (args->ar_lockproto[0]) seq_printf(s, ",lockproto=%s", args->ar_lockproto); if (args->ar_locktable[0]) @@ -414,6 +473,7 @@ static int gfs2_show_options(struct seq_file *s, struct vfsmount *mnt) * conversion on the iopen lock, but we can change that later. This * is safe, just less efficient. */ + static void gfs2_delete_inode(struct inode *inode) { struct gfs2_sbd *sdp = inode->i_sb->s_fs_info; @@ -478,8 +538,6 @@ out: clear_inode(inode); } - - static struct inode *gfs2_alloc_inode(struct super_block *sb) { struct gfs2_inode *ip; diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c index ca831991cbc..c3ba3d9d0aa 100644 --- a/fs/gfs2/super.c +++ b/fs/gfs2/super.c @@ -33,313 +33,6 @@ #include "trans.h" #include "util.h" -static const u32 gfs2_old_fs_formats[] = { - 0 -}; - -static const u32 gfs2_old_multihost_formats[] = { - 0 -}; - -/** - * gfs2_tune_init - Fill a gfs2_tune structure with default values - * @gt: tune - * - */ - -void gfs2_tune_init(struct gfs2_tune *gt) -{ - spin_lock_init(>->gt_spin); - - gt->gt_demote_secs = 300; - gt->gt_incore_log_blocks = 1024; - gt->gt_log_flush_secs = 60; - gt->gt_recoverd_secs = 60; - gt->gt_logd_secs = 1; - gt->gt_quotad_secs = 5; - gt->gt_quota_simul_sync = 64; - gt->gt_quota_warn_period = 10; - gt->gt_quota_scale_num = 1; - gt->gt_quota_scale_den = 1; - gt->gt_quota_cache_secs = 300; - gt->gt_quota_quantum = 60; - gt->gt_atime_quantum = 3600; - gt->gt_new_files_jdata = 0; - gt->gt_max_readahead = 1 << 18; - gt->gt_stall_secs = 600; - gt->gt_complain_secs = 10; - gt->gt_statfs_quantum = 30; - gt->gt_statfs_slow = 0; -} - -/** - * gfs2_check_sb - Check superblock - * @sdp: the filesystem - * @sb: The superblock - * @silent: Don't print a message if the check fails - * - * Checks the version code of the FS is one that we understand how to - * read and that the sizes of the various on-disk structures have not - * changed. - */ - -int gfs2_check_sb(struct gfs2_sbd *sdp, struct gfs2_sb_host *sb, int silent) -{ - unsigned int x; - - if (sb->sb_magic != GFS2_MAGIC || - sb->sb_type != GFS2_METATYPE_SB) { - if (!silent) - printk(KERN_WARNING "GFS2: not a GFS2 filesystem\n"); - return -EINVAL; - } - - /* If format numbers match exactly, we're done. */ - - if (sb->sb_fs_format == GFS2_FORMAT_FS && - sb->sb_multihost_format == GFS2_FORMAT_MULTI) - return 0; - - if (sb->sb_fs_format != GFS2_FORMAT_FS) { - for (x = 0; gfs2_old_fs_formats[x]; x++) - if (gfs2_old_fs_formats[x] == sb->sb_fs_format) - break; - - if (!gfs2_old_fs_formats[x]) { - printk(KERN_WARNING - "GFS2: code version (%u, %u) is incompatible " - "with ondisk format (%u, %u)\n", - GFS2_FORMAT_FS, GFS2_FORMAT_MULTI, - sb->sb_fs_format, sb->sb_multihost_format); - printk(KERN_WARNING - "GFS2: I don't know how to upgrade this FS\n"); - return -EINVAL; - } - } - - if (sb->sb_multihost_format != GFS2_FORMAT_MULTI) { - for (x = 0; gfs2_old_multihost_formats[x]; x++) - if (gfs2_old_multihost_formats[x] == - sb->sb_multihost_format) - break; - - if (!gfs2_old_multihost_formats[x]) { - printk(KERN_WARNING - "GFS2: code version (%u, %u) is incompatible " - "with ondisk format (%u, %u)\n", - GFS2_FORMAT_FS, GFS2_FORMAT_MULTI, - sb->sb_fs_format, sb->sb_multihost_format); - printk(KERN_WARNING - "GFS2: I don't know how to upgrade this FS\n"); - return -EINVAL; - } - } - - if (!sdp->sd_args.ar_upgrade) { - printk(KERN_WARNING - "GFS2: code version (%u, %u) is incompatible " - "with ondisk format (%u, %u)\n", - GFS2_FORMAT_FS, GFS2_FORMAT_MULTI, - sb->sb_fs_format, sb->sb_multihost_format); - printk(KERN_INFO - "GFS2: Use the \"upgrade\" mount option to upgrade " - "the FS\n"); - printk(KERN_INFO "GFS2: See the manual for more details\n"); - return -EINVAL; - } - - return 0; -} - - -static void end_bio_io_page(struct bio *bio, int error) -{ - struct page *page = bio->bi_private; - - if (!error) - SetPageUptodate(page); - else - printk(KERN_WARNING "gfs2: error %d reading superblock\n", error); - unlock_page(page); -} - -static void gfs2_sb_in(struct gfs2_sb_host *sb, const void *buf) -{ - const struct gfs2_sb *str = buf; - - sb->sb_magic = be32_to_cpu(str->sb_header.mh_magic); - sb->sb_type = be32_to_cpu(str->sb_header.mh_type); - sb->sb_format = be32_to_cpu(str->sb_header.mh_format); - sb->sb_fs_format = be32_to_cpu(str->sb_fs_format); - sb->sb_multihost_format = be32_to_cpu(str->sb_multihost_format); - sb->sb_bsize = be32_to_cpu(str->sb_bsize); - sb->sb_bsize_shift = be32_to_cpu(str->sb_bsize_shift); - sb->sb_master_dir.no_addr = be64_to_cpu(str->sb_master_dir.no_addr); - sb->sb_master_dir.no_formal_ino = be64_to_cpu(str->sb_master_dir.no_formal_ino); - sb->sb_root_dir.no_addr = be64_to_cpu(str->sb_root_dir.no_addr); - sb->sb_root_dir.no_formal_ino = be64_to_cpu(str->sb_root_dir.no_formal_ino); - - memcpy(sb->sb_lockproto, str->sb_lockproto, GFS2_LOCKNAME_LEN); - memcpy(sb->sb_locktable, str->sb_locktable, GFS2_LOCKNAME_LEN); -} - -/** - * gfs2_read_super - Read the gfs2 super block from disk - * @sdp: The GFS2 super block - * @sector: The location of the super block - * @error: The error code to return - * - * This uses the bio functions to read the super block from disk - * because we want to be 100% sure that we never read cached data. - * A super block is read twice only during each GFS2 mount and is - * never written to by the filesystem. The first time its read no - * locks are held, and the only details which are looked at are those - * relating to the locking protocol. Once locking is up and working, - * the sb is read again under the lock to establish the location of - * the master directory (contains pointers to journals etc) and the - * root directory. - * - * Returns: 0 on success or error - */ - -int gfs2_read_super(struct gfs2_sbd *sdp, sector_t sector) -{ - struct super_block *sb = sdp->sd_vfs; - struct gfs2_sb *p; - struct page *page; - struct bio *bio; - - page = alloc_page(GFP_NOFS); - if (unlikely(!page)) - return -ENOBUFS; - - ClearPageUptodate(page); - ClearPageDirty(page); - lock_page(page); - - bio = bio_alloc(GFP_NOFS, 1); - if (unlikely(!bio)) { - __free_page(page); - return -ENOBUFS; - } - - bio->bi_sector = sector * (sb->s_blocksize >> 9); - bio->bi_bdev = sb->s_bdev; - bio_add_page(bio, page, PAGE_SIZE, 0); - - bio->bi_end_io = end_bio_io_page; - bio->bi_private = page; - submit_bio(READ_SYNC | (1 << BIO_RW_META), bio); - wait_on_page_locked(page); - bio_put(bio); - if (!PageUptodate(page)) { - __free_page(page); - return -EIO; - } - p = kmap(page); - gfs2_sb_in(&sdp->sd_sb, p); - kunmap(page); - __free_page(page); - return 0; -} - -/** - * gfs2_read_sb - Read super block - * @sdp: The GFS2 superblock - * @gl: the glock for the superblock (assumed to be held) - * @silent: Don't print message if mount fails - * - */ - -int gfs2_read_sb(struct gfs2_sbd *sdp, struct gfs2_glock *gl, int silent) -{ - u32 hash_blocks, ind_blocks, leaf_blocks; - u32 tmp_blocks; - unsigned int x; - int error; - - error = gfs2_read_super(sdp, GFS2_SB_ADDR >> sdp->sd_fsb2bb_shift); - if (error) { - if (!silent) - fs_err(sdp, "can't read superblock\n"); - return error; - } - - error = gfs2_check_sb(sdp, &sdp->sd_sb, silent); - if (error) - return error; - - sdp->sd_fsb2bb_shift = sdp->sd_sb.sb_bsize_shift - - GFS2_BASIC_BLOCK_SHIFT; - sdp->sd_fsb2bb = 1 << sdp->sd_fsb2bb_shift; - sdp->sd_diptrs = (sdp->sd_sb.sb_bsize - - sizeof(struct gfs2_dinode)) / sizeof(u64); - sdp->sd_inptrs = (sdp->sd_sb.sb_bsize - - sizeof(struct gfs2_meta_header)) / sizeof(u64); - sdp->sd_jbsize = sdp->sd_sb.sb_bsize - sizeof(struct gfs2_meta_header); - sdp->sd_hash_bsize = sdp->sd_sb.sb_bsize / 2; - sdp->sd_hash_bsize_shift = sdp->sd_sb.sb_bsize_shift - 1; - sdp->sd_hash_ptrs = sdp->sd_hash_bsize / sizeof(u64); - sdp->sd_qc_per_block = (sdp->sd_sb.sb_bsize - - sizeof(struct gfs2_meta_header)) / - sizeof(struct gfs2_quota_change); - - /* Compute maximum reservation required to add a entry to a directory */ - - hash_blocks = DIV_ROUND_UP(sizeof(u64) * (1 << GFS2_DIR_MAX_DEPTH), - sdp->sd_jbsize); - - ind_blocks = 0; - for (tmp_blocks = hash_blocks; tmp_blocks > sdp->sd_diptrs;) { - tmp_blocks = DIV_ROUND_UP(tmp_blocks, sdp->sd_inptrs); - ind_blocks += tmp_blocks; - } - - leaf_blocks = 2 + GFS2_DIR_MAX_DEPTH; - - sdp->sd_max_dirres = hash_blocks + ind_blocks + leaf_blocks; - - sdp->sd_heightsize[0] = sdp->sd_sb.sb_bsize - - sizeof(struct gfs2_dinode); - sdp->sd_heightsize[1] = sdp->sd_sb.sb_bsize * sdp->sd_diptrs; - for (x = 2;; x++) { - u64 space, d; - u32 m; - - space = sdp->sd_heightsize[x - 1] * sdp->sd_inptrs; - d = space; - m = do_div(d, sdp->sd_inptrs); - - if (d != sdp->sd_heightsize[x - 1] || m) - break; - sdp->sd_heightsize[x] = space; - } - sdp->sd_max_height = x; - sdp->sd_heightsize[x] = ~0; - gfs2_assert(sdp, sdp->sd_max_height <= GFS2_MAX_META_HEIGHT); - - sdp->sd_jheightsize[0] = sdp->sd_sb.sb_bsize - - sizeof(struct gfs2_dinode); - sdp->sd_jheightsize[1] = sdp->sd_jbsize * sdp->sd_diptrs; - for (x = 2;; x++) { - u64 space, d; - u32 m; - - space = sdp->sd_jheightsize[x - 1] * sdp->sd_inptrs; - d = space; - m = do_div(d, sdp->sd_inptrs); - - if (d != sdp->sd_jheightsize[x - 1] || m) - break; - sdp->sd_jheightsize[x] = space; - } - sdp->sd_max_jheight = x; - sdp->sd_jheightsize[x] = ~0; - gfs2_assert(sdp, sdp->sd_max_jheight <= GFS2_MAX_META_HEIGHT); - - return 0; -} - /** * gfs2_jindex_hold - Grab a lock on the jindex * @sdp: The GFS2 superblock @@ -581,39 +274,6 @@ fail: return error; } -/** - * gfs2_make_fs_ro - Turn a Read-Write FS into a Read-Only one - * @sdp: the filesystem - * - * Returns: errno - */ - -int gfs2_make_fs_ro(struct gfs2_sbd *sdp) -{ - struct gfs2_holder t_gh; - int error; - - gfs2_quota_sync(sdp); - gfs2_statfs_sync(sdp); - - error = gfs2_glock_nq_init(sdp->sd_trans_gl, LM_ST_SHARED, GL_NOCACHE, - &t_gh); - if (error && !test_bit(SDF_SHUTDOWN, &sdp->sd_flags)) - return error; - - gfs2_meta_syncfs(sdp); - gfs2_log_shutdown(sdp); - - clear_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags); - - if (t_gh.gh_gl) - gfs2_glock_dq_uninit(&t_gh); - - gfs2_quota_cleanup(sdp); - - return error; -} - static void gfs2_statfs_change_in(struct gfs2_statfs_change_host *sc, const void *buf) { const struct gfs2_statfs_change *str = buf; diff --git a/fs/gfs2/super.h b/fs/gfs2/super.h index 44361ecc44f..50a4c9b1215 100644 --- a/fs/gfs2/super.h +++ b/fs/gfs2/super.h @@ -12,11 +12,6 @@ #include "incore.h" -void gfs2_tune_init(struct gfs2_tune *gt); - -int gfs2_check_sb(struct gfs2_sbd *sdp, struct gfs2_sb_host *sb, int silent); -int gfs2_read_sb(struct gfs2_sbd *sdp, struct gfs2_glock *gl, int silent); -int gfs2_read_super(struct gfs2_sbd *sdp, sector_t sector); void gfs2_lm_unmount(struct gfs2_sbd *sdp); static inline unsigned int gfs2_jindex_size(struct gfs2_sbd *sdp) @@ -40,7 +35,6 @@ int gfs2_lookup_in_master_dir(struct gfs2_sbd *sdp, char *filename, struct gfs2_inode **ipp); int gfs2_make_fs_rw(struct gfs2_sbd *sdp); -int gfs2_make_fs_ro(struct gfs2_sbd *sdp); int gfs2_statfs_init(struct gfs2_sbd *sdp); void gfs2_statfs_change(struct gfs2_sbd *sdp, diff --git a/fs/gfs2/sys.c b/fs/gfs2/sys.c index 74846559fc3..7e1879f1a02 100644 --- a/fs/gfs2/sys.c +++ b/fs/gfs2/sys.c @@ -269,14 +269,6 @@ ARGS_ATTR(quota, "%u\n"); ARGS_ATTR(suiddir, "%d\n"); ARGS_ATTR(data, "%d\n"); -/* one oddball doesn't fit the macro mold */ -static ssize_t noatime_show(struct gfs2_sbd *sdp, char *buf) -{ - return snprintf(buf, PAGE_SIZE, "%d\n", - !!test_bit(SDF_NOATIME, &sdp->sd_flags)); -} -static struct args_attr args_attr_noatime = __ATTR_RO(noatime); - static struct attribute *args_attrs[] = { &args_attr_lockproto.attr, &args_attr_locktable.attr, @@ -292,7 +284,6 @@ static struct attribute *args_attrs[] = { &args_attr_quota.attr, &args_attr_suiddir.attr, &args_attr_data.attr, - &args_attr_noatime.attr, NULL, }; @@ -407,7 +398,6 @@ TUNE_ATTR(incore_log_blocks, 0); TUNE_ATTR(log_flush_secs, 0); TUNE_ATTR(quota_warn_period, 0); TUNE_ATTR(quota_quantum, 0); -TUNE_ATTR(atime_quantum, 0); TUNE_ATTR(max_readahead, 0); TUNE_ATTR(complain_secs, 0); TUNE_ATTR(statfs_slow, 0); @@ -427,7 +417,6 @@ static struct attribute *tune_attrs[] = { &tune_attr_log_flush_secs.attr, &tune_attr_quota_warn_period.attr, &tune_attr_quota_quantum.attr, - &tune_attr_atime_quantum.attr, &tune_attr_max_readahead.attr, &tune_attr_complain_secs.attr, &tune_attr_statfs_slow.attr, diff --git a/fs/inotify_user.c b/fs/inotify_user.c index 60249429a25..d85c7d931cd 100644 --- a/fs/inotify_user.c +++ b/fs/inotify_user.c @@ -323,7 +323,7 @@ out: } /* - * remove_kevent - cleans up and ultimately frees the given kevent + * remove_kevent - cleans up the given kevent * * Caller must hold dev->ev_mutex. */ @@ -334,7 +334,13 @@ static void remove_kevent(struct inotify_device *dev, dev->event_count--; dev->queue_size -= sizeof(struct inotify_event) + kevent->event.len; +} +/* + * free_kevent - frees the given kevent. + */ +static void free_kevent(struct inotify_kernel_event *kevent) +{ kfree(kevent->name); kmem_cache_free(event_cachep, kevent); } @@ -350,6 +356,7 @@ static void inotify_dev_event_dequeue(struct inotify_device *dev) struct inotify_kernel_event *kevent; kevent = inotify_dev_get_event(dev); remove_kevent(dev, kevent); + free_kevent(kevent); } } @@ -433,17 +440,15 @@ static ssize_t inotify_read(struct file *file, char __user *buf, dev = file->private_data; while (1) { - int events; prepare_to_wait(&dev->wq, &wait, TASK_INTERRUPTIBLE); mutex_lock(&dev->ev_mutex); - events = !list_empty(&dev->events); - mutex_unlock(&dev->ev_mutex); - if (events) { + if (!list_empty(&dev->events)) { ret = 0; break; } + mutex_unlock(&dev->ev_mutex); if (file->f_flags & O_NONBLOCK) { ret = -EAGAIN; @@ -462,7 +467,6 @@ static ssize_t inotify_read(struct file *file, char __user *buf, if (ret) return ret; - mutex_lock(&dev->ev_mutex); while (1) { struct inotify_kernel_event *kevent; @@ -481,6 +485,13 @@ static ssize_t inotify_read(struct file *file, char __user *buf, } break; } + remove_kevent(dev, kevent); + + /* + * Must perform the copy_to_user outside the mutex in order + * to avoid a lock order reversal with mmap_sem. + */ + mutex_unlock(&dev->ev_mutex); if (copy_to_user(buf, &kevent->event, event_size)) { ret = -EFAULT; @@ -498,7 +509,9 @@ static ssize_t inotify_read(struct file *file, char __user *buf, count -= kevent->event.len; } - remove_kevent(dev, kevent); + free_kevent(kevent); + + mutex_lock(&dev->ev_mutex); } mutex_unlock(&dev->ev_mutex); diff --git a/fs/ioctl.c b/fs/ioctl.c index 7db32b3382d..d152856c371 100644 --- a/fs/ioctl.c +++ b/fs/ioctl.c @@ -13,9 +13,14 @@ #include <linux/security.h> #include <linux/module.h> #include <linux/uaccess.h> +#include <linux/writeback.h> +#include <linux/buffer_head.h> #include <asm/ioctls.h> +/* So that the fiemap access checks can't overflow on 32 bit machines. */ +#define FIEMAP_MAX_EXTENTS (UINT_MAX / sizeof(struct fiemap_extent)) + /** * vfs_ioctl - call filesystem specific ioctl methods * @filp: open file to invoke ioctl method on @@ -71,6 +76,276 @@ static int ioctl_fibmap(struct file *filp, int __user *p) return put_user(res, p); } +/** + * fiemap_fill_next_extent - Fiemap helper function + * @fieinfo: Fiemap context passed into ->fiemap + * @logical: Extent logical start offset, in bytes + * @phys: Extent physical start offset, in bytes + * @len: Extent length, in bytes + * @flags: FIEMAP_EXTENT flags that describe this extent + * + * Called from file system ->fiemap callback. Will populate extent + * info as passed in via arguments and copy to user memory. On + * success, extent count on fieinfo is incremented. + * + * Returns 0 on success, -errno on error, 1 if this was the last + * extent that will fit in user array. + */ +#define SET_UNKNOWN_FLAGS (FIEMAP_EXTENT_DELALLOC) +#define SET_NO_UNMOUNTED_IO_FLAGS (FIEMAP_EXTENT_DATA_ENCRYPTED) +#define SET_NOT_ALIGNED_FLAGS (FIEMAP_EXTENT_DATA_TAIL|FIEMAP_EXTENT_DATA_INLINE) +int fiemap_fill_next_extent(struct fiemap_extent_info *fieinfo, u64 logical, + u64 phys, u64 len, u32 flags) +{ + struct fiemap_extent extent; + struct fiemap_extent *dest = fieinfo->fi_extents_start; + + /* only count the extents */ + if (fieinfo->fi_extents_max == 0) { + fieinfo->fi_extents_mapped++; + return (flags & FIEMAP_EXTENT_LAST) ? 1 : 0; + } + + if (fieinfo->fi_extents_mapped >= fieinfo->fi_extents_max) + return 1; + + if (flags & SET_UNKNOWN_FLAGS) + flags |= FIEMAP_EXTENT_UNKNOWN; + if (flags & SET_NO_UNMOUNTED_IO_FLAGS) + flags |= FIEMAP_EXTENT_ENCODED; + if (flags & SET_NOT_ALIGNED_FLAGS) + flags |= FIEMAP_EXTENT_NOT_ALIGNED; + + memset(&extent, 0, sizeof(extent)); + extent.fe_logical = logical; + extent.fe_physical = phys; + extent.fe_length = len; + extent.fe_flags = flags; + + dest += fieinfo->fi_extents_mapped; + if (copy_to_user(dest, &extent, sizeof(extent))) + return -EFAULT; + + fieinfo->fi_extents_mapped++; + if (fieinfo->fi_extents_mapped == fieinfo->fi_extents_max) + return 1; + return (flags & FIEMAP_EXTENT_LAST) ? 1 : 0; +} +EXPORT_SYMBOL(fiemap_fill_next_extent); + +/** + * fiemap_check_flags - check validity of requested flags for fiemap + * @fieinfo: Fiemap context passed into ->fiemap + * @fs_flags: Set of fiemap flags that the file system understands + * + * Called from file system ->fiemap callback. This will compute the + * intersection of valid fiemap flags and those that the fs supports. That + * value is then compared against the user supplied flags. In case of bad user + * flags, the invalid values will be written into the fieinfo structure, and + * -EBADR is returned, which tells ioctl_fiemap() to return those values to + * userspace. For this reason, a return code of -EBADR should be preserved. + * + * Returns 0 on success, -EBADR on bad flags. + */ +int fiemap_check_flags(struct fiemap_extent_info *fieinfo, u32 fs_flags) +{ + u32 incompat_flags; + + incompat_flags = fieinfo->fi_flags & ~(FIEMAP_FLAGS_COMPAT & fs_flags); + if (incompat_flags) { + fieinfo->fi_flags = incompat_flags; + return -EBADR; + } + return 0; +} +EXPORT_SYMBOL(fiemap_check_flags); + +static int fiemap_check_ranges(struct super_block *sb, + u64 start, u64 len, u64 *new_len) +{ + *new_len = len; + + if (len == 0) + return -EINVAL; + + if (start > sb->s_maxbytes) + return -EFBIG; + + /* + * Shrink request scope to what the fs can actually handle. + */ + if ((len > sb->s_maxbytes) || + (sb->s_maxbytes - len) < start) + *new_len = sb->s_maxbytes - start; + + return 0; +} + +static int ioctl_fiemap(struct file *filp, unsigned long arg) +{ + struct fiemap fiemap; + struct fiemap_extent_info fieinfo = { 0, }; + struct inode *inode = filp->f_path.dentry->d_inode; + struct super_block *sb = inode->i_sb; + u64 len; + int error; + + if (!inode->i_op->fiemap) + return -EOPNOTSUPP; + + if (copy_from_user(&fiemap, (struct fiemap __user *)arg, + sizeof(struct fiemap))) + return -EFAULT; + + if (fiemap.fm_extent_count > FIEMAP_MAX_EXTENTS) + return -EINVAL; + + error = fiemap_check_ranges(sb, fiemap.fm_start, fiemap.fm_length, + &len); + if (error) + return error; + + fieinfo.fi_flags = fiemap.fm_flags; + fieinfo.fi_extents_max = fiemap.fm_extent_count; + fieinfo.fi_extents_start = (struct fiemap_extent *)(arg + sizeof(fiemap)); + + if (fiemap.fm_extent_count != 0 && + !access_ok(VERIFY_WRITE, fieinfo.fi_extents_start, + fieinfo.fi_extents_max * sizeof(struct fiemap_extent))) + return -EFAULT; + + if (fieinfo.fi_flags & FIEMAP_FLAG_SYNC) + filemap_write_and_wait(inode->i_mapping); + + error = inode->i_op->fiemap(inode, &fieinfo, fiemap.fm_start, len); + fiemap.fm_flags = fieinfo.fi_flags; + fiemap.fm_mapped_extents = fieinfo.fi_extents_mapped; + if (copy_to_user((char *)arg, &fiemap, sizeof(fiemap))) + error = -EFAULT; + + return error; +} + +#ifdef CONFIG_BLOCK + +#define blk_to_logical(inode, blk) (blk << (inode)->i_blkbits) +#define logical_to_blk(inode, offset) (offset >> (inode)->i_blkbits); + +/* + * @inode - the inode to map + * @arg - the pointer to userspace where we copy everything to + * @get_block - the fs's get_block function + * + * This does FIEMAP for block based inodes. Basically it will just loop + * through get_block until we hit the number of extents we want to map, or we + * go past the end of the file and hit a hole. + * + * If it is possible to have data blocks beyond a hole past @inode->i_size, then + * please do not use this function, it will stop at the first unmapped block + * beyond i_size + */ +int generic_block_fiemap(struct inode *inode, + struct fiemap_extent_info *fieinfo, u64 start, + u64 len, get_block_t *get_block) +{ + struct buffer_head tmp; + unsigned int start_blk; + long long length = 0, map_len = 0; + u64 logical = 0, phys = 0, size = 0; + u32 flags = FIEMAP_EXTENT_MERGED; + int ret = 0; + + if ((ret = fiemap_check_flags(fieinfo, FIEMAP_FLAG_SYNC))) + return ret; + + start_blk = logical_to_blk(inode, start); + + /* guard against change */ + mutex_lock(&inode->i_mutex); + + length = (long long)min_t(u64, len, i_size_read(inode)); + map_len = length; + + do { + /* + * we set b_size to the total size we want so it will map as + * many contiguous blocks as possible at once + */ + memset(&tmp, 0, sizeof(struct buffer_head)); + tmp.b_size = map_len; + + ret = get_block(inode, start_blk, &tmp, 0); + if (ret) + break; + + /* HOLE */ + if (!buffer_mapped(&tmp)) { + /* + * first hole after going past the EOF, this is our + * last extent + */ + if (length <= 0) { + flags = FIEMAP_EXTENT_MERGED|FIEMAP_EXTENT_LAST; + ret = fiemap_fill_next_extent(fieinfo, logical, + phys, size, + flags); + break; + } + + length -= blk_to_logical(inode, 1); + + /* if we have holes up to/past EOF then we're done */ + if (length <= 0) + break; + + start_blk++; + } else { + if (length <= 0 && size) { + ret = fiemap_fill_next_extent(fieinfo, logical, + phys, size, + flags); + if (ret) + break; + } + + logical = blk_to_logical(inode, start_blk); + phys = blk_to_logical(inode, tmp.b_blocknr); + size = tmp.b_size; + flags = FIEMAP_EXTENT_MERGED; + + length -= tmp.b_size; + start_blk += logical_to_blk(inode, size); + + /* + * if we are past the EOF we need to loop again to see + * if there is a hole so we can mark this extent as the + * last one, and if not keep mapping things until we + * find a hole, or we run out of slots in the extent + * array + */ + if (length <= 0) + continue; + + ret = fiemap_fill_next_extent(fieinfo, logical, phys, + size, flags); + if (ret) + break; + } + cond_resched(); + } while (1); + + mutex_unlock(&inode->i_mutex); + + /* if ret is 1 then we just hit the end of the extent array */ + if (ret == 1) + ret = 0; + + return ret; +} +EXPORT_SYMBOL(generic_block_fiemap); + +#endif /* CONFIG_BLOCK */ + static int file_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) { @@ -80,6 +355,8 @@ static int file_ioctl(struct file *filp, unsigned int cmd, switch (cmd) { case FIBMAP: return ioctl_fibmap(filp, p); + case FS_IOC_FIEMAP: + return ioctl_fiemap(filp, arg); case FIGETBSZ: return put_user(inode->i_sb->s_blocksize, p); case FIONREAD: diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c index 91389c8aee8..9203c3332f1 100644 --- a/fs/jbd2/checkpoint.c +++ b/fs/jbd2/checkpoint.c @@ -20,6 +20,7 @@ #include <linux/time.h> #include <linux/fs.h> #include <linux/jbd2.h> +#include <linux/marker.h> #include <linux/errno.h> #include <linux/slab.h> @@ -93,7 +94,8 @@ static int __try_to_free_cp_buf(struct journal_head *jh) int ret = 0; struct buffer_head *bh = jh2bh(jh); - if (jh->b_jlist == BJ_None && !buffer_locked(bh) && !buffer_dirty(bh)) { + if (jh->b_jlist == BJ_None && !buffer_locked(bh) && + !buffer_dirty(bh) && !buffer_write_io_error(bh)) { JBUFFER_TRACE(jh, "remove from checkpoint list"); ret = __jbd2_journal_remove_checkpoint(jh) + 1; jbd_unlock_bh_state(bh); @@ -126,14 +128,29 @@ void __jbd2_log_wait_for_space(journal_t *journal) /* * Test again, another process may have checkpointed while we - * were waiting for the checkpoint lock + * were waiting for the checkpoint lock. If there are no + * outstanding transactions there is nothing to checkpoint and + * we can't make progress. Abort the journal in this case. */ spin_lock(&journal->j_state_lock); + spin_lock(&journal->j_list_lock); nblocks = jbd_space_needed(journal); if (__jbd2_log_space_left(journal) < nblocks) { + int chkpt = journal->j_checkpoint_transactions != NULL; + + spin_unlock(&journal->j_list_lock); spin_unlock(&journal->j_state_lock); - jbd2_log_do_checkpoint(journal); + if (chkpt) { + jbd2_log_do_checkpoint(journal); + } else { + printk(KERN_ERR "%s: no transactions\n", + __func__); + jbd2_journal_abort(journal, 0); + } + spin_lock(&journal->j_state_lock); + } else { + spin_unlock(&journal->j_list_lock); } mutex_unlock(&journal->j_checkpoint_mutex); } @@ -160,21 +177,25 @@ static void jbd_sync_bh(journal_t *journal, struct buffer_head *bh) * buffers. Note that we take the buffers in the opposite ordering * from the one in which they were submitted for IO. * + * Return 0 on success, and return <0 if some buffers have failed + * to be written out. + * * Called with j_list_lock held. */ -static void __wait_cp_io(journal_t *journal, transaction_t *transaction) +static int __wait_cp_io(journal_t *journal, transaction_t *transaction) { struct journal_head *jh; struct buffer_head *bh; tid_t this_tid; int released = 0; + int ret = 0; this_tid = transaction->t_tid; restart: /* Did somebody clean up the transaction in the meanwhile? */ if (journal->j_checkpoint_transactions != transaction || transaction->t_tid != this_tid) - return; + return ret; while (!released && transaction->t_checkpoint_io_list) { jh = transaction->t_checkpoint_io_list; bh = jh2bh(jh); @@ -194,6 +215,9 @@ restart: spin_lock(&journal->j_list_lock); goto restart; } + if (unlikely(buffer_write_io_error(bh))) + ret = -EIO; + /* * Now in whatever state the buffer currently is, we know that * it has been written out and so we can drop it from the list @@ -203,6 +227,8 @@ restart: jbd2_journal_remove_journal_head(bh); __brelse(bh); } + + return ret; } #define NR_BATCH 64 @@ -226,7 +252,8 @@ __flush_batch(journal_t *journal, struct buffer_head **bhs, int *batch_count) * Try to flush one buffer from the checkpoint list to disk. * * Return 1 if something happened which requires us to abort the current - * scan of the checkpoint list. + * scan of the checkpoint list. Return <0 if the buffer has failed to + * be written out. * * Called with j_list_lock held and drops it if 1 is returned * Called under jbd_lock_bh_state(jh2bh(jh)), and drops it @@ -258,6 +285,9 @@ static int __process_buffer(journal_t *journal, struct journal_head *jh, jbd2_log_wait_commit(journal, tid); ret = 1; } else if (!buffer_dirty(bh)) { + ret = 1; + if (unlikely(buffer_write_io_error(bh))) + ret = -EIO; J_ASSERT_JH(jh, !buffer_jbddirty(bh)); BUFFER_TRACE(bh, "remove from checkpoint"); __jbd2_journal_remove_checkpoint(jh); @@ -265,7 +295,6 @@ static int __process_buffer(journal_t *journal, struct journal_head *jh, jbd_unlock_bh_state(bh); jbd2_journal_remove_journal_head(bh); __brelse(bh); - ret = 1; } else { /* * Important: we are about to write the buffer, and @@ -298,6 +327,7 @@ static int __process_buffer(journal_t *journal, struct journal_head *jh, * to disk. We submit larger chunks of data at once. * * The journal should be locked before calling this function. + * Called with j_checkpoint_mutex held. */ int jbd2_log_do_checkpoint(journal_t *journal) { @@ -313,6 +343,8 @@ int jbd2_log_do_checkpoint(journal_t *journal) * journal straight away. */ result = jbd2_cleanup_journal_tail(journal); + trace_mark(jbd2_checkpoint, "dev %s need_checkpoint %d", + journal->j_devname, result); jbd_debug(1, "cleanup_journal_tail returned %d\n", result); if (result <= 0) return result; @@ -321,6 +353,7 @@ int jbd2_log_do_checkpoint(journal_t *journal) * OK, we need to start writing disk blocks. Take one transaction * and write it. */ + result = 0; spin_lock(&journal->j_list_lock); if (!journal->j_checkpoint_transactions) goto out; @@ -339,7 +372,7 @@ restart: int batch_count = 0; struct buffer_head *bhs[NR_BATCH]; struct journal_head *jh; - int retry = 0; + int retry = 0, err; while (!retry && transaction->t_checkpoint_list) { struct buffer_head *bh; @@ -353,6 +386,8 @@ restart: } retry = __process_buffer(journal, jh, bhs, &batch_count, transaction); + if (retry < 0 && !result) + result = retry; if (!retry && (need_resched() || spin_needbreak(&journal->j_list_lock))) { spin_unlock(&journal->j_list_lock); @@ -377,14 +412,18 @@ restart: * Now we have cleaned up the first transaction's checkpoint * list. Let's clean up the second one */ - __wait_cp_io(journal, transaction); + err = __wait_cp_io(journal, transaction); + if (!result) + result = err; } out: spin_unlock(&journal->j_list_lock); - result = jbd2_cleanup_journal_tail(journal); if (result < 0) - return result; - return 0; + jbd2_journal_abort(journal, result); + else + result = jbd2_cleanup_journal_tail(journal); + + return (result < 0) ? result : 0; } /* @@ -400,8 +439,9 @@ out: * This is the only part of the journaling code which really needs to be * aware of transaction aborts. Checkpointing involves writing to the * main filesystem area rather than to the journal, so it can proceed - * even in abort state, but we must not update the journal superblock if - * we have an abort error outstanding. + * even in abort state, but we must not update the super block if + * checkpointing may have failed. Otherwise, we would lose some metadata + * buffers which should be written-back to the filesystem. */ int jbd2_cleanup_journal_tail(journal_t *journal) @@ -410,6 +450,9 @@ int jbd2_cleanup_journal_tail(journal_t *journal) tid_t first_tid; unsigned long blocknr, freed; + if (is_journal_aborted(journal)) + return 1; + /* OK, work out the oldest transaction remaining in the log, and * the log block it starts at. * diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index f2ad061e95e..0abe02c4242 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -16,6 +16,7 @@ #include <linux/time.h> #include <linux/fs.h> #include <linux/jbd2.h> +#include <linux/marker.h> #include <linux/errno.h> #include <linux/slab.h> #include <linux/mm.h> @@ -126,8 +127,7 @@ static int journal_submit_commit_record(journal_t *journal, JBUFFER_TRACE(descriptor, "submit commit block"); lock_buffer(bh); - get_bh(bh); - set_buffer_dirty(bh); + clear_buffer_dirty(bh); set_buffer_uptodate(bh); bh->b_end_io = journal_end_buffer_io_sync; @@ -147,12 +147,9 @@ static int journal_submit_commit_record(journal_t *journal, * to remember if we sent a barrier request */ if (ret == -EOPNOTSUPP && barrier_done) { - char b[BDEVNAME_SIZE]; - printk(KERN_WARNING - "JBD: barrier-based sync failed on %s - " - "disabling barriers\n", - bdevname(journal->j_dev, b)); + "JBD: barrier-based sync failed on %s - " + "disabling barriers\n", journal->j_devname); spin_lock(&journal->j_state_lock); journal->j_flags &= ~JBD2_BARRIER; spin_unlock(&journal->j_state_lock); @@ -160,7 +157,7 @@ static int journal_submit_commit_record(journal_t *journal, /* And try again, without the barrier */ lock_buffer(bh); set_buffer_uptodate(bh); - set_buffer_dirty(bh); + clear_buffer_dirty(bh); ret = submit_bh(WRITE, bh); } *cbh = bh; @@ -371,6 +368,8 @@ void jbd2_journal_commit_transaction(journal_t *journal) commit_transaction = journal->j_running_transaction; J_ASSERT(commit_transaction->t_state == T_RUNNING); + trace_mark(jbd2_start_commit, "dev %s transaction %d", + journal->j_devname, commit_transaction->t_tid); jbd_debug(1, "JBD: starting commit of transaction %d\n", commit_transaction->t_tid); @@ -505,9 +504,10 @@ void jbd2_journal_commit_transaction(journal_t *journal) jh = commit_transaction->t_buffers; /* If we're in abort mode, we just un-journal the buffer and - release it for background writing. */ + release it. */ if (is_journal_aborted(journal)) { + clear_buffer_jbddirty(jh2bh(jh)); JBUFFER_TRACE(jh, "journal is aborting: refile"); jbd2_journal_refile_buffer(journal, jh); /* If that was the last one, we need to clean up @@ -681,11 +681,11 @@ start_journal_io: */ err = journal_finish_inode_data_buffers(journal, commit_transaction); if (err) { - char b[BDEVNAME_SIZE]; - printk(KERN_WARNING "JBD2: Detected IO errors while flushing file data " - "on %s\n", bdevname(journal->j_fs_dev, b)); + "on %s\n", journal->j_devname); + if (journal->j_flags & JBD2_ABORT_ON_SYNCDATA_ERR) + jbd2_journal_abort(journal, err); err = 0; } @@ -786,6 +786,9 @@ wait_for_iobuf: /* AKPM: bforget here */ } + if (err) + jbd2_journal_abort(journal, err); + jbd_debug(3, "JBD: commit phase 5\n"); if (!JBD2_HAS_INCOMPAT_FEATURE(journal, @@ -884,6 +887,8 @@ restart_loop: if (buffer_jbddirty(bh)) { JBUFFER_TRACE(jh, "add to new checkpointing trans"); __jbd2_journal_insert_checkpoint(jh, commit_transaction); + if (is_journal_aborted(journal)) + clear_buffer_jbddirty(bh); JBUFFER_TRACE(jh, "refile for checkpoint writeback"); __jbd2_journal_refile_buffer(jh); jbd_unlock_bh_state(bh); @@ -990,6 +995,9 @@ restart_loop: } spin_unlock(&journal->j_list_lock); + trace_mark(jbd2_end_commit, "dev %s transaction %d head %d", + journal->j_devname, commit_transaction->t_tid, + journal->j_tail_sequence); jbd_debug(1, "JBD: commit %d complete, head %d\n", journal->j_commit_sequence, journal->j_tail_sequence); diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index 8207a01c4ed..783de118de9 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -597,13 +597,9 @@ int jbd2_journal_bmap(journal_t *journal, unsigned long blocknr, if (ret) *retp = ret; else { - char b[BDEVNAME_SIZE]; - printk(KERN_ALERT "%s: journal block not found " "at offset %lu on %s\n", - __func__, - blocknr, - bdevname(journal->j_dev, b)); + __func__, blocknr, journal->j_devname); err = -EIO; __journal_abort_soft(journal, err); } @@ -901,10 +897,7 @@ static struct proc_dir_entry *proc_jbd2_stats; static void jbd2_stats_proc_init(journal_t *journal) { - char name[BDEVNAME_SIZE]; - - bdevname(journal->j_dev, name); - journal->j_proc_entry = proc_mkdir(name, proc_jbd2_stats); + journal->j_proc_entry = proc_mkdir(journal->j_devname, proc_jbd2_stats); if (journal->j_proc_entry) { proc_create_data("history", S_IRUGO, journal->j_proc_entry, &jbd2_seq_history_fops, journal); @@ -915,12 +908,9 @@ static void jbd2_stats_proc_init(journal_t *journal) static void jbd2_stats_proc_exit(journal_t *journal) { - char name[BDEVNAME_SIZE]; - - bdevname(journal->j_dev, name); remove_proc_entry("info", journal->j_proc_entry); remove_proc_entry("history", journal->j_proc_entry); - remove_proc_entry(name, proc_jbd2_stats); + remove_proc_entry(journal->j_devname, proc_jbd2_stats); } static void journal_init_stats(journal_t *journal) @@ -1018,6 +1008,7 @@ journal_t * jbd2_journal_init_dev(struct block_device *bdev, { journal_t *journal = journal_init_common(); struct buffer_head *bh; + char *p; int n; if (!journal) @@ -1039,6 +1030,10 @@ journal_t * jbd2_journal_init_dev(struct block_device *bdev, journal->j_fs_dev = fs_dev; journal->j_blk_offset = start; journal->j_maxlen = len; + bdevname(journal->j_dev, journal->j_devname); + p = journal->j_devname; + while ((p = strchr(p, '/'))) + *p = '!'; jbd2_stats_proc_init(journal); bh = __getblk(journal->j_dev, start, journal->j_blocksize); @@ -1061,6 +1056,7 @@ journal_t * jbd2_journal_init_inode (struct inode *inode) { struct buffer_head *bh; journal_t *journal = journal_init_common(); + char *p; int err; int n; unsigned long long blocknr; @@ -1070,6 +1066,12 @@ journal_t * jbd2_journal_init_inode (struct inode *inode) journal->j_dev = journal->j_fs_dev = inode->i_sb->s_bdev; journal->j_inode = inode; + bdevname(journal->j_dev, journal->j_devname); + p = journal->j_devname; + while ((p = strchr(p, '/'))) + *p = '!'; + p = journal->j_devname + strlen(journal->j_devname); + sprintf(p, ":%lu", journal->j_inode->i_ino); jbd_debug(1, "journal %p: inode %s/%ld, size %Ld, bits %d, blksize %ld\n", journal, inode->i_sb->s_id, inode->i_ino, @@ -1253,6 +1255,22 @@ void jbd2_journal_update_superblock(journal_t *journal, int wait) goto out; } + if (buffer_write_io_error(bh)) { + /* + * Oh, dear. A previous attempt to write the journal + * superblock failed. This could happen because the + * USB device was yanked out. Or it could happen to + * be a transient write error and maybe the block will + * be remapped. Nothing we can do but to retry the + * write and hope for the best. + */ + printk(KERN_ERR "JBD2: previous I/O error detected " + "for journal superblock update for %s.\n", + journal->j_devname); + clear_buffer_write_io_error(bh); + set_buffer_uptodate(bh); + } + spin_lock(&journal->j_state_lock); jbd_debug(1,"JBD: updating superblock (start %ld, seq %d, errno %d)\n", journal->j_tail, journal->j_tail_sequence, journal->j_errno); @@ -1264,9 +1282,16 @@ void jbd2_journal_update_superblock(journal_t *journal, int wait) BUFFER_TRACE(bh, "marking dirty"); mark_buffer_dirty(bh); - if (wait) + if (wait) { sync_dirty_buffer(bh); - else + if (buffer_write_io_error(bh)) { + printk(KERN_ERR "JBD2: I/O error detected " + "when updating journal superblock for %s.\n", + journal->j_devname); + clear_buffer_write_io_error(bh); + set_buffer_uptodate(bh); + } + } else ll_rw_block(SWRITE, 1, &bh); out: @@ -1426,9 +1451,12 @@ recovery_error: * * Release a journal_t structure once it is no longer in use by the * journaled object. + * Return <0 if we couldn't clean up the journal. */ -void jbd2_journal_destroy(journal_t *journal) +int jbd2_journal_destroy(journal_t *journal) { + int err = 0; + /* Wait for the commit thread to wake up and die. */ journal_kill_thread(journal); @@ -1451,11 +1479,16 @@ void jbd2_journal_destroy(journal_t *journal) J_ASSERT(journal->j_checkpoint_transactions == NULL); spin_unlock(&journal->j_list_lock); - /* We can now mark the journal as empty. */ - journal->j_tail = 0; - journal->j_tail_sequence = ++journal->j_transaction_sequence; if (journal->j_sb_buffer) { - jbd2_journal_update_superblock(journal, 1); + if (!is_journal_aborted(journal)) { + /* We can now mark the journal as empty. */ + journal->j_tail = 0; + journal->j_tail_sequence = + ++journal->j_transaction_sequence; + jbd2_journal_update_superblock(journal, 1); + } else { + err = -EIO; + } brelse(journal->j_sb_buffer); } @@ -1467,6 +1500,8 @@ void jbd2_journal_destroy(journal_t *journal) jbd2_journal_destroy_revoke(journal); kfree(journal->j_wbuf); kfree(journal); + + return err; } @@ -1692,10 +1727,16 @@ int jbd2_journal_flush(journal_t *journal) spin_lock(&journal->j_list_lock); while (!err && journal->j_checkpoint_transactions != NULL) { spin_unlock(&journal->j_list_lock); + mutex_lock(&journal->j_checkpoint_mutex); err = jbd2_log_do_checkpoint(journal); + mutex_unlock(&journal->j_checkpoint_mutex); spin_lock(&journal->j_list_lock); } spin_unlock(&journal->j_list_lock); + + if (is_journal_aborted(journal)) + return -EIO; + jbd2_cleanup_journal_tail(journal); /* Finally, mark the journal as really needing no recovery. @@ -1717,7 +1758,7 @@ int jbd2_journal_flush(journal_t *journal) J_ASSERT(journal->j_head == journal->j_tail); J_ASSERT(journal->j_tail_sequence == journal->j_transaction_sequence); spin_unlock(&journal->j_state_lock); - return err; + return 0; } /** @@ -1761,23 +1802,6 @@ int jbd2_journal_wipe(journal_t *journal, int write) } /* - * journal_dev_name: format a character string to describe on what - * device this journal is present. - */ - -static const char *journal_dev_name(journal_t *journal, char *buffer) -{ - struct block_device *bdev; - - if (journal->j_inode) - bdev = journal->j_inode->i_sb->s_bdev; - else - bdev = journal->j_dev; - - return bdevname(bdev, buffer); -} - -/* * Journal abort has very specific semantics, which we describe * for journal abort. * @@ -1793,13 +1817,12 @@ static const char *journal_dev_name(journal_t *journal, char *buffer) void __jbd2_journal_abort_hard(journal_t *journal) { transaction_t *transaction; - char b[BDEVNAME_SIZE]; if (journal->j_flags & JBD2_ABORT) return; printk(KERN_ERR "Aborting journal on device %s.\n", - journal_dev_name(journal, b)); + journal->j_devname); spin_lock(&journal->j_state_lock); journal->j_flags |= JBD2_ABORT; diff --git a/fs/jbd2/recovery.c b/fs/jbd2/recovery.c index 058f50f65b7..73063285b13 100644 --- a/fs/jbd2/recovery.c +++ b/fs/jbd2/recovery.c @@ -225,7 +225,7 @@ do { \ */ int jbd2_journal_recover(journal_t *journal) { - int err; + int err, err2; journal_superblock_t * sb; struct recovery_info info; @@ -263,7 +263,10 @@ int jbd2_journal_recover(journal_t *journal) journal->j_transaction_sequence = ++info.end_transaction; jbd2_journal_clear_revoke(journal); - sync_blockdev(journal->j_fs_dev); + err2 = sync_blockdev(journal->j_fs_dev); + if (!err) + err = err2; + return err; } diff --git a/fs/nfs/super.c b/fs/nfs/super.c index 9abcd2b329f..e9b20173fef 100644 --- a/fs/nfs/super.c +++ b/fs/nfs/super.c @@ -1279,6 +1279,12 @@ static int nfs_parse_mount_options(char *raw, } } + if (errors > 0) { + dfprintk(MOUNT, "NFS: parsing encountered %d error%s\n", + errors, (errors == 1 ? "" : "s")); + if (!sloppy) + return 0; + } return 1; out_nomem: diff --git a/fs/ntfs/usnjrnl.h b/fs/ntfs/usnjrnl.h index 3a8af75351e..4087fbdac32 100644 --- a/fs/ntfs/usnjrnl.h +++ b/fs/ntfs/usnjrnl.h @@ -113,7 +113,7 @@ typedef struct { * Reason flags (32-bit). Cumulative flags describing the change(s) to the * file since it was last opened. I think the names speak for themselves but * if you disagree check out the descriptions in the Linux NTFS project NTFS - * documentation: http://linux-ntfs.sourceforge.net/ntfs/files/usnjrnl.html + * documentation: http://www.linux-ntfs.org/ */ enum { USN_REASON_DATA_OVERWRITE = const_cpu_to_le32(0x00000001), @@ -145,7 +145,7 @@ typedef le32 USN_REASON_FLAGS; * Source info flags (32-bit). Information about the source of the change(s) * to the file. For detailed descriptions of what these mean, see the Linux * NTFS project NTFS documentation: - * http://linux-ntfs.sourceforge.net/ntfs/files/usnjrnl.html + * http://www.linux-ntfs.org/ */ enum { USN_SOURCE_DATA_MANAGEMENT = const_cpu_to_le32(0x00000001), diff --git a/fs/ocfs2/alloc.c b/fs/ocfs2/alloc.c index 10bfb466e06..29ff57ec5d1 100644 --- a/fs/ocfs2/alloc.c +++ b/fs/ocfs2/alloc.c @@ -990,15 +990,6 @@ out: } /* - * This is only valid for leaf nodes, which are the only ones that can - * have empty extents anyway. - */ -static inline int ocfs2_is_empty_extent(struct ocfs2_extent_rec *rec) -{ - return !rec->e_leaf_clusters; -} - -/* * This function will discard the rightmost extent record. */ static void ocfs2_shift_records_right(struct ocfs2_extent_list *el) diff --git a/fs/ocfs2/alloc.h b/fs/ocfs2/alloc.h index 42ff94bd801..60cd3d59230 100644 --- a/fs/ocfs2/alloc.h +++ b/fs/ocfs2/alloc.h @@ -146,4 +146,13 @@ static inline unsigned int ocfs2_rec_clusters(struct ocfs2_extent_list *el, return le16_to_cpu(rec->e_leaf_clusters); } +/* + * This is only valid for leaf nodes, which are the only ones that can + * have empty extents anyway. + */ +static inline int ocfs2_is_empty_extent(struct ocfs2_extent_rec *rec) +{ + return !rec->e_leaf_clusters; +} + #endif /* OCFS2_ALLOC_H */ diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c index 506c24fb507..a53da146627 100644 --- a/fs/ocfs2/aops.c +++ b/fs/ocfs2/aops.c @@ -594,7 +594,7 @@ static int ocfs2_direct_IO_get_blocks(struct inode *inode, sector_t iblock, goto bail; } - if (!ocfs2_sparse_alloc(OCFS2_SB(inode->i_sb)) && !p_blkno) { + if (!ocfs2_sparse_alloc(OCFS2_SB(inode->i_sb)) && !p_blkno && create) { ocfs2_error(inode->i_sb, "Inode %llu has a hole at block %llu\n", (unsigned long long)OCFS2_I(inode)->ip_blkno, diff --git a/fs/ocfs2/extent_map.c b/fs/ocfs2/extent_map.c index c58668a326f..aed268e80b4 100644 --- a/fs/ocfs2/extent_map.c +++ b/fs/ocfs2/extent_map.c @@ -25,6 +25,7 @@ #include <linux/fs.h> #include <linux/init.h> #include <linux/types.h> +#include <linux/fiemap.h> #define MLOG_MASK_PREFIX ML_EXTENT_MAP #include <cluster/masklog.h> @@ -32,6 +33,7 @@ #include "ocfs2.h" #include "alloc.h" +#include "dlmglue.h" #include "extent_map.h" #include "inode.h" #include "super.h" @@ -282,6 +284,51 @@ out: kfree(new_emi); } +static int ocfs2_last_eb_is_empty(struct inode *inode, + struct ocfs2_dinode *di) +{ + int ret, next_free; + u64 last_eb_blk = le64_to_cpu(di->i_last_eb_blk); + struct buffer_head *eb_bh = NULL; + struct ocfs2_extent_block *eb; + struct ocfs2_extent_list *el; + + ret = ocfs2_read_block(OCFS2_SB(inode->i_sb), last_eb_blk, + &eb_bh, OCFS2_BH_CACHED, inode); + if (ret) { + mlog_errno(ret); + goto out; + } + + eb = (struct ocfs2_extent_block *) eb_bh->b_data; + el = &eb->h_list; + + if (!OCFS2_IS_VALID_EXTENT_BLOCK(eb)) { + ret = -EROFS; + OCFS2_RO_ON_INVALID_EXTENT_BLOCK(inode->i_sb, eb); + goto out; + } + + if (el->l_tree_depth) { + ocfs2_error(inode->i_sb, + "Inode %lu has non zero tree depth in " + "leaf block %llu\n", inode->i_ino, + (unsigned long long)eb_bh->b_blocknr); + ret = -EROFS; + goto out; + } + + next_free = le16_to_cpu(el->l_next_free_rec); + + if (next_free == 0 || + (next_free == 1 && ocfs2_is_empty_extent(&el->l_recs[0]))) + ret = 1; + +out: + brelse(eb_bh); + return ret; +} + /* * Return the 1st index within el which contains an extent start * larger than v_cluster. @@ -373,42 +420,28 @@ out: return ret; } -int ocfs2_get_clusters(struct inode *inode, u32 v_cluster, - u32 *p_cluster, u32 *num_clusters, - unsigned int *extent_flags) +static int ocfs2_get_clusters_nocache(struct inode *inode, + struct buffer_head *di_bh, + u32 v_cluster, unsigned int *hole_len, + struct ocfs2_extent_rec *ret_rec, + unsigned int *is_last) { - int ret, i; - unsigned int flags = 0; - struct buffer_head *di_bh = NULL; - struct buffer_head *eb_bh = NULL; + int i, ret, tree_height, len; struct ocfs2_dinode *di; - struct ocfs2_extent_block *eb; + struct ocfs2_extent_block *uninitialized_var(eb); struct ocfs2_extent_list *el; struct ocfs2_extent_rec *rec; - u32 coff; - - if (OCFS2_I(inode)->ip_dyn_features & OCFS2_INLINE_DATA_FL) { - ret = -ERANGE; - mlog_errno(ret); - goto out; - } - - ret = ocfs2_extent_map_lookup(inode, v_cluster, p_cluster, - num_clusters, extent_flags); - if (ret == 0) - goto out; + struct buffer_head *eb_bh = NULL; - ret = ocfs2_read_block(OCFS2_SB(inode->i_sb), OCFS2_I(inode)->ip_blkno, - &di_bh, OCFS2_BH_CACHED, inode); - if (ret) { - mlog_errno(ret); - goto out; - } + memset(ret_rec, 0, sizeof(*ret_rec)); + if (is_last) + *is_last = 0; di = (struct ocfs2_dinode *) di_bh->b_data; el = &di->id2.i_list; + tree_height = le16_to_cpu(el->l_tree_depth); - if (el->l_tree_depth) { + if (tree_height > 0) { ret = ocfs2_find_leaf(inode, el, v_cluster, &eb_bh); if (ret) { mlog_errno(ret); @@ -431,46 +464,143 @@ int ocfs2_get_clusters(struct inode *inode, u32 v_cluster, i = ocfs2_search_extent_list(el, v_cluster); if (i == -1) { /* - * A hole was found. Return some canned values that - * callers can key on. If asked for, num_clusters will - * be populated with the size of the hole. + * Holes can be larger than the maximum size of an + * extent, so we return their lengths in a seperate + * field. */ - *p_cluster = 0; - if (num_clusters) { + if (hole_len) { ret = ocfs2_figure_hole_clusters(inode, el, eb_bh, - v_cluster, - num_clusters); + v_cluster, &len); if (ret) { mlog_errno(ret); goto out; } + + *hole_len = len; } - } else { - rec = &el->l_recs[i]; + goto out_hole; + } - BUG_ON(v_cluster < le32_to_cpu(rec->e_cpos)); + rec = &el->l_recs[i]; - if (!rec->e_blkno) { - ocfs2_error(inode->i_sb, "Inode %lu has bad extent " - "record (%u, %u, 0)", inode->i_ino, - le32_to_cpu(rec->e_cpos), - ocfs2_rec_clusters(el, rec)); - ret = -EROFS; - goto out; + BUG_ON(v_cluster < le32_to_cpu(rec->e_cpos)); + + if (!rec->e_blkno) { + ocfs2_error(inode->i_sb, "Inode %lu has bad extent " + "record (%u, %u, 0)", inode->i_ino, + le32_to_cpu(rec->e_cpos), + ocfs2_rec_clusters(el, rec)); + ret = -EROFS; + goto out; + } + + *ret_rec = *rec; + + /* + * Checking for last extent is potentially expensive - we + * might have to look at the next leaf over to see if it's + * empty. + * + * The first two checks are to see whether the caller even + * cares for this information, and if the extent is at least + * the last in it's list. + * + * If those hold true, then the extent is last if any of the + * additional conditions hold true: + * - Extent list is in-inode + * - Extent list is right-most + * - Extent list is 2nd to rightmost, with empty right-most + */ + if (is_last) { + if (i == (le16_to_cpu(el->l_next_free_rec) - 1)) { + if (tree_height == 0) + *is_last = 1; + else if (eb->h_blkno == di->i_last_eb_blk) + *is_last = 1; + else if (eb->h_next_leaf_blk == di->i_last_eb_blk) { + ret = ocfs2_last_eb_is_empty(inode, di); + if (ret < 0) { + mlog_errno(ret); + goto out; + } + if (ret == 1) + *is_last = 1; + } } + } + +out_hole: + ret = 0; +out: + brelse(eb_bh); + return ret; +} + +static void ocfs2_relative_extent_offsets(struct super_block *sb, + u32 v_cluster, + struct ocfs2_extent_rec *rec, + u32 *p_cluster, u32 *num_clusters) + +{ + u32 coff = v_cluster - le32_to_cpu(rec->e_cpos); + + *p_cluster = ocfs2_blocks_to_clusters(sb, le64_to_cpu(rec->e_blkno)); + *p_cluster = *p_cluster + coff; + + if (num_clusters) + *num_clusters = le16_to_cpu(rec->e_leaf_clusters) - coff; +} + +int ocfs2_get_clusters(struct inode *inode, u32 v_cluster, + u32 *p_cluster, u32 *num_clusters, + unsigned int *extent_flags) +{ + int ret; + unsigned int uninitialized_var(hole_len), flags = 0; + struct buffer_head *di_bh = NULL; + struct ocfs2_extent_rec rec; - coff = v_cluster - le32_to_cpu(rec->e_cpos); + if (OCFS2_I(inode)->ip_dyn_features & OCFS2_INLINE_DATA_FL) { + ret = -ERANGE; + mlog_errno(ret); + goto out; + } - *p_cluster = ocfs2_blocks_to_clusters(inode->i_sb, - le64_to_cpu(rec->e_blkno)); - *p_cluster = *p_cluster + coff; + ret = ocfs2_extent_map_lookup(inode, v_cluster, p_cluster, + num_clusters, extent_flags); + if (ret == 0) + goto out; - if (num_clusters) - *num_clusters = ocfs2_rec_clusters(el, rec) - coff; + ret = ocfs2_read_block(OCFS2_SB(inode->i_sb), OCFS2_I(inode)->ip_blkno, + &di_bh, OCFS2_BH_CACHED, inode); + if (ret) { + mlog_errno(ret); + goto out; + } - flags = rec->e_flags; + ret = ocfs2_get_clusters_nocache(inode, di_bh, v_cluster, &hole_len, + &rec, NULL); + if (ret) { + mlog_errno(ret); + goto out; + } - ocfs2_extent_map_insert_rec(inode, rec); + if (rec.e_blkno == 0ULL) { + /* + * A hole was found. Return some canned values that + * callers can key on. If asked for, num_clusters will + * be populated with the size of the hole. + */ + *p_cluster = 0; + if (num_clusters) { + *num_clusters = hole_len; + } + } else { + ocfs2_relative_extent_offsets(inode->i_sb, v_cluster, &rec, + p_cluster, num_clusters); + flags = rec.e_flags; + + ocfs2_extent_map_insert_rec(inode, &rec); } if (extent_flags) @@ -478,7 +608,6 @@ int ocfs2_get_clusters(struct inode *inode, u32 v_cluster, out: brelse(di_bh); - brelse(eb_bh); return ret; } @@ -521,3 +650,114 @@ int ocfs2_extent_map_get_blocks(struct inode *inode, u64 v_blkno, u64 *p_blkno, out: return ret; } + +static int ocfs2_fiemap_inline(struct inode *inode, struct buffer_head *di_bh, + struct fiemap_extent_info *fieinfo, + u64 map_start) +{ + int ret; + unsigned int id_count; + struct ocfs2_dinode *di; + u64 phys; + u32 flags = FIEMAP_EXTENT_DATA_INLINE|FIEMAP_EXTENT_LAST; + struct ocfs2_inode_info *oi = OCFS2_I(inode); + + di = (struct ocfs2_dinode *)di_bh->b_data; + id_count = le16_to_cpu(di->id2.i_data.id_count); + + if (map_start < id_count) { + phys = oi->ip_blkno << inode->i_sb->s_blocksize_bits; + phys += offsetof(struct ocfs2_dinode, id2.i_data.id_data); + + ret = fiemap_fill_next_extent(fieinfo, 0, phys, id_count, + flags); + if (ret < 0) + return ret; + } + + return 0; +} + +#define OCFS2_FIEMAP_FLAGS (FIEMAP_FLAG_SYNC) + +int ocfs2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, + u64 map_start, u64 map_len) +{ + int ret, is_last; + u32 mapping_end, cpos; + unsigned int hole_size; + struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); + u64 len_bytes, phys_bytes, virt_bytes; + struct buffer_head *di_bh = NULL; + struct ocfs2_extent_rec rec; + + ret = fiemap_check_flags(fieinfo, OCFS2_FIEMAP_FLAGS); + if (ret) + return ret; + + ret = ocfs2_inode_lock(inode, &di_bh, 0); + if (ret) { + mlog_errno(ret); + goto out; + } + + down_read(&OCFS2_I(inode)->ip_alloc_sem); + + /* + * Handle inline-data separately. + */ + if (OCFS2_I(inode)->ip_dyn_features & OCFS2_INLINE_DATA_FL) { + ret = ocfs2_fiemap_inline(inode, di_bh, fieinfo, map_start); + goto out_unlock; + } + + cpos = map_start >> osb->s_clustersize_bits; + mapping_end = ocfs2_clusters_for_bytes(inode->i_sb, + map_start + map_len); + mapping_end -= cpos; + is_last = 0; + while (cpos < mapping_end && !is_last) { + u32 fe_flags; + + ret = ocfs2_get_clusters_nocache(inode, di_bh, cpos, + &hole_size, &rec, &is_last); + if (ret) { + mlog_errno(ret); + goto out; + } + + if (rec.e_blkno == 0ULL) { + cpos += hole_size; + continue; + } + + fe_flags = 0; + if (rec.e_flags & OCFS2_EXT_UNWRITTEN) + fe_flags |= FIEMAP_EXTENT_UNWRITTEN; + if (is_last) + fe_flags |= FIEMAP_EXTENT_LAST; + len_bytes = (u64)le16_to_cpu(rec.e_leaf_clusters) << osb->s_clustersize_bits; + phys_bytes = le64_to_cpu(rec.e_blkno) << osb->sb->s_blocksize_bits; + virt_bytes = (u64)le32_to_cpu(rec.e_cpos) << osb->s_clustersize_bits; + + ret = fiemap_fill_next_extent(fieinfo, virt_bytes, phys_bytes, + len_bytes, fe_flags); + if (ret) + break; + + cpos = le32_to_cpu(rec.e_cpos)+ le16_to_cpu(rec.e_leaf_clusters); + } + + if (ret > 0) + ret = 0; + +out_unlock: + brelse(di_bh); + + up_read(&OCFS2_I(inode)->ip_alloc_sem); + + ocfs2_inode_unlock(inode, 0); +out: + + return ret; +} diff --git a/fs/ocfs2/extent_map.h b/fs/ocfs2/extent_map.h index de91e3e41a2..1b97490e1ea 100644 --- a/fs/ocfs2/extent_map.h +++ b/fs/ocfs2/extent_map.h @@ -50,4 +50,7 @@ int ocfs2_get_clusters(struct inode *inode, u32 v_cluster, u32 *p_cluster, int ocfs2_extent_map_get_blocks(struct inode *inode, u64 v_blkno, u64 *p_blkno, u64 *ret_count, unsigned int *extent_flags); +int ocfs2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, + u64 map_start, u64 map_len); + #endif /* _EXTENT_MAP_H */ diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index ec2ed15c3da..ed38796052d 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -2228,6 +2228,7 @@ const struct inode_operations ocfs2_file_iops = { .getattr = ocfs2_getattr, .permission = ocfs2_permission, .fallocate = ocfs2_fallocate, + .fiemap = ocfs2_fiemap, }; const struct inode_operations ocfs2_special_file_iops = { diff --git a/fs/partitions/check.c b/fs/partitions/check.c index 7d6b34e201d..7408227c49c 100644 --- a/fs/partitions/check.c +++ b/fs/partitions/check.c @@ -120,22 +120,21 @@ static int (*check_part[])(struct parsed_partitions *, struct block_device *) = * a pointer to that same buffer (for convenience). */ -char *disk_name(struct gendisk *hd, int part, char *buf) +char *disk_name(struct gendisk *hd, int partno, char *buf) { - if (!part) + if (!partno) snprintf(buf, BDEVNAME_SIZE, "%s", hd->disk_name); else if (isdigit(hd->disk_name[strlen(hd->disk_name)-1])) - snprintf(buf, BDEVNAME_SIZE, "%sp%d", hd->disk_name, part); + snprintf(buf, BDEVNAME_SIZE, "%sp%d", hd->disk_name, partno); else - snprintf(buf, BDEVNAME_SIZE, "%s%d", hd->disk_name, part); + snprintf(buf, BDEVNAME_SIZE, "%s%d", hd->disk_name, partno); return buf; } const char *bdevname(struct block_device *bdev, char *buf) { - int part = MINOR(bdev->bd_dev) - bdev->bd_disk->first_minor; - return disk_name(bdev->bd_disk, part, buf); + return disk_name(bdev->bd_disk, bdev->bd_part->partno, buf); } EXPORT_SYMBOL(bdevname); @@ -169,7 +168,7 @@ check_partition(struct gendisk *hd, struct block_device *bdev) if (isdigit(state->name[strlen(state->name)-1])) sprintf(state->name, "p"); - state->limit = hd->minors; + state->limit = disk_max_parts(hd); i = res = err = 0; while (!res && check_part[i]) { memset(&state->parts, 0, sizeof(state->parts)); @@ -204,21 +203,22 @@ static ssize_t part_start_show(struct device *dev, return sprintf(buf, "%llu\n",(unsigned long long)p->start_sect); } -static ssize_t part_size_show(struct device *dev, - struct device_attribute *attr, char *buf) +ssize_t part_size_show(struct device *dev, + struct device_attribute *attr, char *buf) { struct hd_struct *p = dev_to_part(dev); return sprintf(buf, "%llu\n",(unsigned long long)p->nr_sects); } -static ssize_t part_stat_show(struct device *dev, - struct device_attribute *attr, char *buf) +ssize_t part_stat_show(struct device *dev, + struct device_attribute *attr, char *buf) { struct hd_struct *p = dev_to_part(dev); + int cpu; - preempt_disable(); - part_round_stats(p); - preempt_enable(); + cpu = part_stat_lock(); + part_round_stats(cpu, p); + part_stat_unlock(); return sprintf(buf, "%8lu %8lu %8llu %8u " "%8lu %8lu %8llu %8u " @@ -238,17 +238,17 @@ static ssize_t part_stat_show(struct device *dev, } #ifdef CONFIG_FAIL_MAKE_REQUEST -static ssize_t part_fail_show(struct device *dev, - struct device_attribute *attr, char *buf) +ssize_t part_fail_show(struct device *dev, + struct device_attribute *attr, char *buf) { struct hd_struct *p = dev_to_part(dev); return sprintf(buf, "%d\n", p->make_it_fail); } -static ssize_t part_fail_store(struct device *dev, - struct device_attribute *attr, - const char *buf, size_t count) +ssize_t part_fail_store(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) { struct hd_struct *p = dev_to_part(dev); int i; @@ -300,40 +300,34 @@ struct device_type part_type = { .release = part_release, }; -static inline void partition_sysfs_add_subdir(struct hd_struct *p) -{ - struct kobject *k; - - k = kobject_get(&p->dev.kobj); - p->holder_dir = kobject_create_and_add("holders", k); - kobject_put(k); -} - -static inline void disk_sysfs_add_subdirs(struct gendisk *disk) +static void delete_partition_rcu_cb(struct rcu_head *head) { - struct kobject *k; + struct hd_struct *part = container_of(head, struct hd_struct, rcu_head); - k = kobject_get(&disk->dev.kobj); - disk->holder_dir = kobject_create_and_add("holders", k); - disk->slave_dir = kobject_create_and_add("slaves", k); - kobject_put(k); + part->start_sect = 0; + part->nr_sects = 0; + part_stat_set_all(part, 0); + put_device(part_to_dev(part)); } -void delete_partition(struct gendisk *disk, int part) +void delete_partition(struct gendisk *disk, int partno) { - struct hd_struct *p = disk->part[part-1]; + struct disk_part_tbl *ptbl = disk->part_tbl; + struct hd_struct *part; - if (!p) + if (partno >= ptbl->len) return; - if (!p->nr_sects) + + part = ptbl->part[partno]; + if (!part) return; - disk->part[part-1] = NULL; - p->start_sect = 0; - p->nr_sects = 0; - part_stat_set_all(p, 0); - kobject_put(p->holder_dir); - device_del(&p->dev); - put_device(&p->dev); + + blk_free_devt(part_devt(part)); + rcu_assign_pointer(ptbl->part[partno], NULL); + kobject_put(part->holder_dir); + device_del(part_to_dev(part)); + + call_rcu(&part->rcu_head, delete_partition_rcu_cb); } static ssize_t whole_disk_show(struct device *dev, @@ -344,102 +338,132 @@ static ssize_t whole_disk_show(struct device *dev, static DEVICE_ATTR(whole_disk, S_IRUSR | S_IRGRP | S_IROTH, whole_disk_show, NULL); -int add_partition(struct gendisk *disk, int part, sector_t start, sector_t len, int flags) +int add_partition(struct gendisk *disk, int partno, + sector_t start, sector_t len, int flags) { struct hd_struct *p; + dev_t devt = MKDEV(0, 0); + struct device *ddev = disk_to_dev(disk); + struct device *pdev; + struct disk_part_tbl *ptbl; + const char *dname; int err; + err = disk_expand_part_tbl(disk, partno); + if (err) + return err; + ptbl = disk->part_tbl; + + if (ptbl->part[partno]) + return -EBUSY; + p = kzalloc(sizeof(*p), GFP_KERNEL); if (!p) return -ENOMEM; if (!init_part_stats(p)) { err = -ENOMEM; - goto out0; + goto out_free; } + pdev = part_to_dev(p); + p->start_sect = start; p->nr_sects = len; - p->partno = part; - p->policy = disk->policy; + p->partno = partno; + p->policy = get_disk_ro(disk); - if (isdigit(disk->dev.bus_id[strlen(disk->dev.bus_id)-1])) - snprintf(p->dev.bus_id, BUS_ID_SIZE, - "%sp%d", disk->dev.bus_id, part); + dname = dev_name(ddev); + if (isdigit(dname[strlen(dname) - 1])) + snprintf(pdev->bus_id, BUS_ID_SIZE, "%sp%d", dname, partno); else - snprintf(p->dev.bus_id, BUS_ID_SIZE, - "%s%d", disk->dev.bus_id, part); + snprintf(pdev->bus_id, BUS_ID_SIZE, "%s%d", dname, partno); + + device_initialize(pdev); + pdev->class = &block_class; + pdev->type = &part_type; + pdev->parent = ddev; - device_initialize(&p->dev); - p->dev.devt = MKDEV(disk->major, disk->first_minor + part); - p->dev.class = &block_class; - p->dev.type = &part_type; - p->dev.parent = &disk->dev; - disk->part[part-1] = p; + err = blk_alloc_devt(p, &devt); + if (err) + goto out_free; + pdev->devt = devt; /* delay uevent until 'holders' subdir is created */ - p->dev.uevent_suppress = 1; - err = device_add(&p->dev); + pdev->uevent_suppress = 1; + err = device_add(pdev); if (err) - goto out1; - partition_sysfs_add_subdir(p); - p->dev.uevent_suppress = 0; + goto out_put; + + err = -ENOMEM; + p->holder_dir = kobject_create_and_add("holders", &pdev->kobj); + if (!p->holder_dir) + goto out_del; + + pdev->uevent_suppress = 0; if (flags & ADDPART_FLAG_WHOLEDISK) { - err = device_create_file(&p->dev, &dev_attr_whole_disk); + err = device_create_file(pdev, &dev_attr_whole_disk); if (err) - goto out2; + goto out_del; } + /* everything is up and running, commence */ + INIT_RCU_HEAD(&p->rcu_head); + rcu_assign_pointer(ptbl->part[partno], p); + /* suppress uevent if the disk supresses it */ - if (!disk->dev.uevent_suppress) - kobject_uevent(&p->dev.kobj, KOBJ_ADD); + if (!ddev->uevent_suppress) + kobject_uevent(&pdev->kobj, KOBJ_ADD); return 0; -out2: - device_del(&p->dev); -out1: - put_device(&p->dev); - free_part_stats(p); -out0: +out_free: kfree(p); return err; +out_del: + kobject_put(p->holder_dir); + device_del(pdev); +out_put: + put_device(pdev); + blk_free_devt(devt); + return err; } /* Not exported, helper to add_disk(). */ void register_disk(struct gendisk *disk) { + struct device *ddev = disk_to_dev(disk); struct block_device *bdev; + struct disk_part_iter piter; + struct hd_struct *part; char *s; - int i; - struct hd_struct *p; int err; - disk->dev.parent = disk->driverfs_dev; - disk->dev.devt = MKDEV(disk->major, disk->first_minor); + ddev->parent = disk->driverfs_dev; - strlcpy(disk->dev.bus_id, disk->disk_name, BUS_ID_SIZE); + strlcpy(ddev->bus_id, disk->disk_name, BUS_ID_SIZE); /* ewww... some of these buggers have / in the name... */ - s = strchr(disk->dev.bus_id, '/'); + s = strchr(ddev->bus_id, '/'); if (s) *s = '!'; /* delay uevents, until we scanned partition table */ - disk->dev.uevent_suppress = 1; + ddev->uevent_suppress = 1; - if (device_add(&disk->dev)) + if (device_add(ddev)) return; #ifndef CONFIG_SYSFS_DEPRECATED - err = sysfs_create_link(block_depr, &disk->dev.kobj, - kobject_name(&disk->dev.kobj)); + err = sysfs_create_link(block_depr, &ddev->kobj, + kobject_name(&ddev->kobj)); if (err) { - device_del(&disk->dev); + device_del(ddev); return; } #endif - disk_sysfs_add_subdirs(disk); + disk->part0.holder_dir = kobject_create_and_add("holders", &ddev->kobj); + disk->slave_dir = kobject_create_and_add("slaves", &ddev->kobj); /* No minors to use for partitions */ - if (disk->minors == 1) + if (!disk_partitionable(disk)) goto exit; /* No such device (e.g., media were just removed) */ @@ -458,50 +482,66 @@ void register_disk(struct gendisk *disk) exit: /* announce disk after possible partitions are created */ - disk->dev.uevent_suppress = 0; - kobject_uevent(&disk->dev.kobj, KOBJ_ADD); + ddev->uevent_suppress = 0; + kobject_uevent(&ddev->kobj, KOBJ_ADD); /* announce possible partitions */ - for (i = 1; i < disk->minors; i++) { - p = disk->part[i-1]; - if (!p || !p->nr_sects) - continue; - kobject_uevent(&p->dev.kobj, KOBJ_ADD); - } + disk_part_iter_init(&piter, disk, 0); + while ((part = disk_part_iter_next(&piter))) + kobject_uevent(&part_to_dev(part)->kobj, KOBJ_ADD); + disk_part_iter_exit(&piter); } int rescan_partitions(struct gendisk *disk, struct block_device *bdev) { + struct disk_part_iter piter; + struct hd_struct *part; struct parsed_partitions *state; - int p, res; + int p, highest, res; if (bdev->bd_part_count) return -EBUSY; res = invalidate_partition(disk, 0); if (res) return res; - bdev->bd_invalidated = 0; - for (p = 1; p < disk->minors; p++) - delete_partition(disk, p); + + disk_part_iter_init(&piter, disk, DISK_PITER_INCL_EMPTY); + while ((part = disk_part_iter_next(&piter))) + delete_partition(disk, part->partno); + disk_part_iter_exit(&piter); + if (disk->fops->revalidate_disk) disk->fops->revalidate_disk(disk); + check_disk_size_change(disk, bdev); + bdev->bd_invalidated = 0; if (!get_capacity(disk) || !(state = check_partition(disk, bdev))) return 0; if (IS_ERR(state)) /* I/O error reading the partition table */ return -EIO; /* tell userspace that the media / partition table may have changed */ - kobject_uevent(&disk->dev.kobj, KOBJ_CHANGE); + kobject_uevent(&disk_to_dev(disk)->kobj, KOBJ_CHANGE); + /* Detect the highest partition number and preallocate + * disk->part_tbl. This is an optimization and not strictly + * necessary. + */ + for (p = 1, highest = 0; p < state->limit; p++) + if (state->parts[p].size) + highest = p; + + disk_expand_part_tbl(disk, highest); + + /* add partitions */ for (p = 1; p < state->limit; p++) { sector_t size = state->parts[p].size; sector_t from = state->parts[p].from; if (!size) continue; if (from + size > get_capacity(disk)) { - printk(KERN_ERR " %s: p%d exceeds device capacity\n", + printk(KERN_WARNING + "%s: p%d exceeds device capacity\n", disk->disk_name, p); - continue; } res = add_partition(disk, p, from, size, state->parts[p].flags); if (res) { @@ -541,25 +581,31 @@ EXPORT_SYMBOL(read_dev_sector); void del_gendisk(struct gendisk *disk) { - int p; + struct disk_part_iter piter; + struct hd_struct *part; /* invalidate stuff */ - for (p = disk->minors - 1; p > 0; p--) { - invalidate_partition(disk, p); - delete_partition(disk, p); + disk_part_iter_init(&piter, disk, + DISK_PITER_INCL_EMPTY | DISK_PITER_REVERSE); + while ((part = disk_part_iter_next(&piter))) { + invalidate_partition(disk, part->partno); + delete_partition(disk, part->partno); } + disk_part_iter_exit(&piter); + invalidate_partition(disk, 0); - disk->capacity = 0; + blk_free_devt(disk_to_dev(disk)->devt); + set_capacity(disk, 0); disk->flags &= ~GENHD_FL_UP; unlink_gendisk(disk); - disk_stat_set_all(disk, 0); - disk->stamp = 0; + part_stat_set_all(&disk->part0, 0); + disk->part0.stamp = 0; - kobject_put(disk->holder_dir); + kobject_put(disk->part0.holder_dir); kobject_put(disk->slave_dir); disk->driverfs_dev = NULL; #ifndef CONFIG_SYSFS_DEPRECATED - sysfs_remove_link(block_depr, disk->dev.bus_id); + sysfs_remove_link(block_depr, dev_name(disk_to_dev(disk))); #endif - device_del(&disk->dev); + device_del(disk_to_dev(disk)); } diff --git a/fs/partitions/check.h b/fs/partitions/check.h index 17ae8ecd9e8..98dbe1a8452 100644 --- a/fs/partitions/check.h +++ b/fs/partitions/check.h @@ -5,15 +5,13 @@ * add_gd_partition adds a partitions details to the devices partition * description. */ -enum { MAX_PART = 256 }; - struct parsed_partitions { char name[BDEVNAME_SIZE]; struct { sector_t from; sector_t size; int flags; - } parts[MAX_PART]; + } parts[DISK_MAX_PARTS]; int next; int limit; }; diff --git a/fs/proc/array.c b/fs/proc/array.c index 0d6eb33597c..71c9be59c9c 100644 --- a/fs/proc/array.c +++ b/fs/proc/array.c @@ -337,65 +337,6 @@ int proc_pid_status(struct seq_file *m, struct pid_namespace *ns, return 0; } -/* - * Use precise platform statistics if available: - */ -#ifdef CONFIG_VIRT_CPU_ACCOUNTING -static cputime_t task_utime(struct task_struct *p) -{ - return p->utime; -} - -static cputime_t task_stime(struct task_struct *p) -{ - return p->stime; -} -#else -static cputime_t task_utime(struct task_struct *p) -{ - clock_t utime = cputime_to_clock_t(p->utime), - total = utime + cputime_to_clock_t(p->stime); - u64 temp; - - /* - * Use CFS's precise accounting: - */ - temp = (u64)nsec_to_clock_t(p->se.sum_exec_runtime); - - if (total) { - temp *= utime; - do_div(temp, total); - } - utime = (clock_t)temp; - - p->prev_utime = max(p->prev_utime, clock_t_to_cputime(utime)); - return p->prev_utime; -} - -static cputime_t task_stime(struct task_struct *p) -{ - clock_t stime; - - /* - * Use CFS's precise accounting. (we subtract utime from - * the total, to make sure the total observed by userspace - * grows monotonically - apps rely on that): - */ - stime = nsec_to_clock_t(p->se.sum_exec_runtime) - - cputime_to_clock_t(task_utime(p)); - - if (stime >= 0) - p->prev_stime = max(p->prev_stime, clock_t_to_cputime(stime)); - - return p->prev_stime; -} -#endif - -static cputime_t task_gtime(struct task_struct *p) -{ - return p->gtime; -} - static int do_task_stat(struct seq_file *m, struct pid_namespace *ns, struct pid *pid, struct task_struct *task, int whole) { diff --git a/fs/proc/generic.c b/fs/proc/generic.c index bca0f81eb68..7821589a17d 100644 --- a/fs/proc/generic.c +++ b/fs/proc/generic.c @@ -547,8 +547,8 @@ static int proc_register(struct proc_dir_entry * dir, struct proc_dir_entry * dp for (tmp = dir->subdir; tmp; tmp = tmp->next) if (strcmp(tmp->name, dp->name) == 0) { - printk(KERN_WARNING "proc_dir_entry '%s' already " - "registered\n", dp->name); + printk(KERN_WARNING "proc_dir_entry '%s/%s' already registered\n", + dir->name, dp->name); dump_stack(); break; } diff --git a/fs/proc/proc_misc.c b/fs/proc/proc_misc.c index ded96986296..29e20c6b1f7 100644 --- a/fs/proc/proc_misc.c +++ b/fs/proc/proc_misc.c @@ -24,6 +24,7 @@ #include <linux/tty.h> #include <linux/string.h> #include <linux/mman.h> +#include <linux/quicklist.h> #include <linux/proc_fs.h> #include <linux/ioport.h> #include <linux/mm.h> @@ -182,6 +183,9 @@ static int meminfo_read_proc(char *page, char **start, off_t off, "SReclaimable: %8lu kB\n" "SUnreclaim: %8lu kB\n" "PageTables: %8lu kB\n" +#ifdef CONFIG_QUICKLIST + "Quicklists: %8lu kB\n" +#endif "NFS_Unstable: %8lu kB\n" "Bounce: %8lu kB\n" "WritebackTmp: %8lu kB\n" @@ -214,6 +218,9 @@ static int meminfo_read_proc(char *page, char **start, off_t off, K(global_page_state(NR_SLAB_RECLAIMABLE)), K(global_page_state(NR_SLAB_UNRECLAIMABLE)), K(global_page_state(NR_PAGETABLE)), +#ifdef CONFIG_QUICKLIST + K(quicklist_total_size()), +#endif K(global_page_state(NR_UNSTABLE_NFS)), K(global_page_state(NR_BOUNCE)), K(global_page_state(NR_WRITEBACK_TEMP)), diff --git a/fs/ramfs/file-nommu.c b/fs/ramfs/file-nommu.c index 52312ec93ff..5145cb9125a 100644 --- a/fs/ramfs/file-nommu.c +++ b/fs/ramfs/file-nommu.c @@ -58,7 +58,7 @@ const struct inode_operations ramfs_file_inode_operations = { * size 0 on the assumption that it's going to be used for an mmap of shared * memory */ -static int ramfs_nommu_expand_for_mapping(struct inode *inode, size_t newsize) +int ramfs_nommu_expand_for_mapping(struct inode *inode, size_t newsize) { struct pagevec lru_pvec; unsigned long npages, xpages, loop, limit; diff --git a/fs/splice.c b/fs/splice.c index 1bbc6f4bb09..a1e701c2715 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -898,6 +898,9 @@ static long do_splice_from(struct pipe_inode_info *pipe, struct file *out, if (unlikely(!(out->f_mode & FMODE_WRITE))) return -EBADF; + if (unlikely(out->f_flags & O_APPEND)) + return -EINVAL; + ret = rw_verify_area(WRITE, out, ppos, len); if (unlikely(ret < 0)) return ret; diff --git a/fs/ubifs/budget.c b/fs/ubifs/budget.c index 15409815747..73db464cd08 100644 --- a/fs/ubifs/budget.c +++ b/fs/ubifs/budget.c @@ -302,18 +302,6 @@ long long ubifs_calc_available(const struct ubifs_info *c, int min_idx_lebs) int subtract_lebs; long long available; - /* - * Force the amount available to the total size reported if the used - * space is zero. - */ - if (c->lst.total_used <= UBIFS_INO_NODE_SZ && - c->budg_data_growth + c->budg_dd_growth == 0) { - /* Do the same calculation as for c->block_cnt */ - available = c->main_lebs - 2; - available *= c->leb_size - c->dark_wm; - return available; - } - available = c->main_bytes - c->lst.total_used; /* @@ -714,34 +702,106 @@ void ubifs_release_dirty_inode_budget(struct ubifs_info *c, } /** - * ubifs_budg_get_free_space - return amount of free space. + * ubifs_reported_space - calculate reported free space. + * @c: the UBIFS file-system description object + * @free: amount of free space + * + * This function calculates amount of free space which will be reported to + * user-space. User-space application tend to expect that if the file-system + * (e.g., via the 'statfs()' call) reports that it has N bytes available, they + * are able to write a file of size N. UBIFS attaches node headers to each data + * node and it has to write indexind nodes as well. This introduces additional + * overhead, and UBIFS it has to report sligtly less free space to meet the + * above expectetion. + * + * This function assumes free space is made up of uncompressed data nodes and + * full index nodes (one per data node, tripled because we always allow enough + * space to write the index thrice). + * + * Note, the calculation is pessimistic, which means that most of the time + * UBIFS reports less space than it actually has. + */ +long long ubifs_reported_space(const struct ubifs_info *c, uint64_t free) +{ + int divisor, factor, f; + + /* + * Reported space size is @free * X, where X is UBIFS block size + * divided by UBIFS block size + all overhead one data block + * introduces. The overhead is the node header + indexing overhead. + * + * Indexing overhead calculations are based on the following formula: + * I = N/(f - 1) + 1, where I - number of indexing nodes, N - number + * of data nodes, f - fanout. Because effective UBIFS fanout is twice + * as less than maximum fanout, we assume that each data node + * introduces 3 * @c->max_idx_node_sz / (@c->fanout/2 - 1) bytes. + * Note, the multiplier 3 is because UBIFS reseves thrice as more space + * for the index. + */ + f = c->fanout > 3 ? c->fanout >> 1 : 2; + factor = UBIFS_BLOCK_SIZE; + divisor = UBIFS_MAX_DATA_NODE_SZ; + divisor += (c->max_idx_node_sz * 3) / (f - 1); + free *= factor; + do_div(free, divisor); + return free; +} + +/** + * ubifs_get_free_space - return amount of free space. * @c: UBIFS file-system description object * - * This function returns amount of free space on the file-system. + * This function calculates amount of free space to report to user-space. + * + * Because UBIFS may introduce substantial overhead (the index, node headers, + * alighment, wastage at the end of eraseblocks, etc), it cannot report real + * amount of free flash space it has (well, because not all dirty space is + * reclamable, UBIFS does not actually know the real amount). If UBIFS did so, + * it would bread user expectetion about what free space is. Users seem to + * accustomed to assume that if the file-system reports N bytes of free space, + * they would be able to fit a file of N bytes to the FS. This almost works for + * traditional file-systems, because they have way less overhead than UBIFS. + * So, to keep users happy, UBIFS tries to take the overhead into account. */ -long long ubifs_budg_get_free_space(struct ubifs_info *c) +long long ubifs_get_free_space(struct ubifs_info *c) { - int min_idx_lebs, rsvd_idx_lebs; + int min_idx_lebs, rsvd_idx_lebs, lebs; long long available, outstanding, free; - /* Do exactly the same calculations as in 'do_budget_space()' */ spin_lock(&c->space_lock); min_idx_lebs = ubifs_calc_min_idx_lebs(c); + outstanding = c->budg_data_growth + c->budg_dd_growth; - if (min_idx_lebs > c->lst.idx_lebs) - rsvd_idx_lebs = min_idx_lebs - c->lst.idx_lebs; - else - rsvd_idx_lebs = 0; - - if (rsvd_idx_lebs > c->lst.empty_lebs + c->freeable_cnt + c->idx_gc_cnt - - c->lst.taken_empty_lebs) { + /* + * Force the amount available to the total size reported if the used + * space is zero. + */ + if (c->lst.total_used <= UBIFS_INO_NODE_SZ && !outstanding) { spin_unlock(&c->space_lock); - return 0; + return (long long)c->block_cnt << UBIFS_BLOCK_SHIFT; } available = ubifs_calc_available(c, min_idx_lebs); - outstanding = c->budg_data_growth + c->budg_dd_growth; - c->min_idx_lebs = min_idx_lebs; + + /* + * When reporting free space to user-space, UBIFS guarantees that it is + * possible to write a file of free space size. This means that for + * empty LEBs we may use more precise calculations than + * 'ubifs_calc_available()' is using. Namely, we know that in empty + * LEBs we would waste only @c->leb_overhead bytes, not @c->dark_wm. + * Thus, amend the available space. + * + * Note, the calculations below are similar to what we have in + * 'do_budget_space()', so refer there for comments. + */ + if (min_idx_lebs > c->lst.idx_lebs) + rsvd_idx_lebs = min_idx_lebs - c->lst.idx_lebs; + else + rsvd_idx_lebs = 0; + lebs = c->lst.empty_lebs + c->freeable_cnt + c->idx_gc_cnt - + c->lst.taken_empty_lebs; + lebs -= rsvd_idx_lebs; + available += lebs * (c->dark_wm - c->leb_overhead); spin_unlock(&c->space_lock); if (available > outstanding) diff --git a/fs/ubifs/debug.c b/fs/ubifs/debug.c index b9cb7747375..d7f7645779f 100644 --- a/fs/ubifs/debug.c +++ b/fs/ubifs/debug.c @@ -538,7 +538,7 @@ void dbg_dump_node(const struct ubifs_info *c, const void *node) printk(KERN_DEBUG "\t%d orphan inode numbers:\n", n); for (i = 0; i < n; i++) printk(KERN_DEBUG "\t ino %llu\n", - le64_to_cpu(orph->inos[i])); + (unsigned long long)le64_to_cpu(orph->inos[i])); break; } default: diff --git a/fs/ubifs/dir.c b/fs/ubifs/dir.c index 5c96f1fb701..526c01ec800 100644 --- a/fs/ubifs/dir.c +++ b/fs/ubifs/dir.c @@ -426,7 +426,7 @@ static int ubifs_readdir(struct file *file, void *dirent, filldir_t filldir) while (1) { dbg_gen("feed '%s', ino %llu, new f_pos %#x", - dent->name, le64_to_cpu(dent->inum), + dent->name, (unsigned long long)le64_to_cpu(dent->inum), key_hash_flash(c, &dent->key)); ubifs_assert(dent->ch.sqnum > ubifs_inode(dir)->creat_sqnum); @@ -587,7 +587,6 @@ static int ubifs_unlink(struct inode *dir, struct dentry *dentry) if (err) { if (err != -ENOSPC) return err; - err = 0; budgeted = 0; } diff --git a/fs/ubifs/file.c b/fs/ubifs/file.c index 4071d1cae29..3d698e2022b 100644 --- a/fs/ubifs/file.c +++ b/fs/ubifs/file.c @@ -793,7 +793,7 @@ static int do_truncation(struct ubifs_info *c, struct inode *inode, int err; struct ubifs_budget_req req; loff_t old_size = inode->i_size, new_size = attr->ia_size; - int offset = new_size & (UBIFS_BLOCK_SIZE - 1); + int offset = new_size & (UBIFS_BLOCK_SIZE - 1), budgeted = 1; struct ubifs_inode *ui = ubifs_inode(inode); dbg_gen("ino %lu, size %lld -> %lld", inode->i_ino, old_size, new_size); @@ -811,8 +811,15 @@ static int do_truncation(struct ubifs_info *c, struct inode *inode, /* A funny way to budget for truncation node */ req.dirtied_ino_d = UBIFS_TRUN_NODE_SZ; err = ubifs_budget_space(c, &req); - if (err) - return err; + if (err) { + /* + * Treat truncations to zero as deletion and always allow them, + * just like we do for '->unlink()'. + */ + if (new_size || err != -ENOSPC) + return err; + budgeted = 0; + } err = vmtruncate(inode, new_size); if (err) @@ -869,7 +876,12 @@ static int do_truncation(struct ubifs_info *c, struct inode *inode, err = ubifs_jnl_truncate(c, inode, old_size, new_size); mutex_unlock(&ui->ui_mutex); out_budg: - ubifs_release_budget(c, &req); + if (budgeted) + ubifs_release_budget(c, &req); + else { + c->nospace = c->nospace_rp = 0; + smp_wmb(); + } return err; } diff --git a/fs/ubifs/find.c b/fs/ubifs/find.c index adee7b5ddea..47814cde240 100644 --- a/fs/ubifs/find.c +++ b/fs/ubifs/find.c @@ -211,14 +211,8 @@ static const struct ubifs_lprops *scan_for_dirty(struct ubifs_info *c, * dirty index heap, and it falls-back to LPT scanning if the heaps are empty * or do not have an LEB which satisfies the @min_space criteria. * - * Note: - * o LEBs which have less than dead watermark of dirty space are never picked - * by this function; - * - * Returns zero and the LEB properties of - * found dirty LEB in case of success, %-ENOSPC if no dirty LEB was found and a - * negative error code in case of other failures. The returned LEB is marked as - * "taken". + * Note, LEBs which have less than dead watermark of free + dirty space are + * never picked by this function. * * The additional @pick_free argument controls if this function has to return a * free or freeable LEB if one is present. For example, GC must to set it to %1, @@ -231,6 +225,10 @@ static const struct ubifs_lprops *scan_for_dirty(struct ubifs_info *c, * * In addition @pick_free is set to %2 by the recovery process in order to * recover gc_lnum in which case an index LEB must not be returned. + * + * This function returns zero and the LEB properties of found dirty LEB in case + * of success, %-ENOSPC if no dirty LEB was found and a negative error code in + * case of other failures. The returned LEB is marked as "taken". */ int ubifs_find_dirty_leb(struct ubifs_info *c, struct ubifs_lprops *ret_lp, int min_space, int pick_free) @@ -245,7 +243,7 @@ int ubifs_find_dirty_leb(struct ubifs_info *c, struct ubifs_lprops *ret_lp, int lebs, rsvd_idx_lebs = 0; spin_lock(&c->space_lock); - lebs = c->lst.empty_lebs; + lebs = c->lst.empty_lebs + c->idx_gc_cnt; lebs += c->freeable_cnt - c->lst.taken_empty_lebs; /* @@ -317,7 +315,7 @@ int ubifs_find_dirty_leb(struct ubifs_info *c, struct ubifs_lprops *ret_lp, lp = idx_lp; if (lp) { - ubifs_assert(lp->dirty >= c->dead_wm); + ubifs_assert(lp->free + lp->dirty >= c->dead_wm); goto found; } @@ -509,7 +507,6 @@ int ubifs_find_free_space(struct ubifs_info *c, int min_space, int *free, rsvd_idx_lebs = 0; lebs = c->lst.empty_lebs + c->freeable_cnt + c->idx_gc_cnt - c->lst.taken_empty_lebs; - ubifs_assert(lebs + c->lst.idx_lebs >= c->min_idx_lebs); if (rsvd_idx_lebs < lebs) /* * OK to allocate an empty LEB, but we still don't want to go diff --git a/fs/ubifs/gc.c b/fs/ubifs/gc.c index d0f3dac2908..02aba36fe3d 100644 --- a/fs/ubifs/gc.c +++ b/fs/ubifs/gc.c @@ -334,15 +334,21 @@ int ubifs_garbage_collect_leb(struct ubifs_info *c, struct ubifs_lprops *lp) err = move_nodes(c, sleb); if (err) - goto out; + goto out_inc_seq; err = gc_sync_wbufs(c); if (err) - goto out; + goto out_inc_seq; err = ubifs_change_one_lp(c, lnum, c->leb_size, 0, 0, 0, 0); if (err) - goto out; + goto out_inc_seq; + + /* Allow for races with TNC */ + c->gced_lnum = lnum; + smp_wmb(); + c->gc_seq += 1; + smp_wmb(); if (c->gc_lnum == -1) { c->gc_lnum = lnum; @@ -363,6 +369,14 @@ int ubifs_garbage_collect_leb(struct ubifs_info *c, struct ubifs_lprops *lp) out: ubifs_scan_destroy(sleb); return err; + +out_inc_seq: + /* We may have moved at least some nodes so allow for races with TNC */ + c->gced_lnum = lnum; + smp_wmb(); + c->gc_seq += 1; + smp_wmb(); + goto out; } /** diff --git a/fs/ubifs/misc.h b/fs/ubifs/misc.h index 87dabf9fe74..4c12a9215d7 100644 --- a/fs/ubifs/misc.h +++ b/fs/ubifs/misc.h @@ -284,38 +284,6 @@ static inline void *ubifs_idx_key(const struct ubifs_info *c, } /** - * ubifs_reported_space - calculate reported free space. - * @c: the UBIFS file-system description object - * @free: amount of free space - * - * This function calculates amount of free space which will be reported to - * user-space. User-space application tend to expect that if the file-system - * (e.g., via the 'statfs()' call) reports that it has N bytes available, they - * are able to write a file of size N. UBIFS attaches node headers to each data - * node and it has to write indexind nodes as well. This introduces additional - * overhead, and UBIFS it has to report sligtly less free space to meet the - * above expectetion. - * - * This function assumes free space is made up of uncompressed data nodes and - * full index nodes (one per data node, doubled because we always allow enough - * space to write the index twice). - * - * Note, the calculation is pessimistic, which means that most of the time - * UBIFS reports less space than it actually has. - */ -static inline long long ubifs_reported_space(const struct ubifs_info *c, - uint64_t free) -{ - int divisor, factor; - - divisor = UBIFS_MAX_DATA_NODE_SZ + (c->max_idx_node_sz * 3); - factor = UBIFS_MAX_DATA_NODE_SZ - UBIFS_DATA_NODE_SZ; - do_div(free, divisor); - - return free * factor; -} - -/** * ubifs_current_time - round current time to time granularity. * @inode: inode */ @@ -325,4 +293,21 @@ static inline struct timespec ubifs_current_time(struct inode *inode) current_fs_time(inode->i_sb) : CURRENT_TIME_SEC; } +/** + * ubifs_tnc_lookup - look up a file-system node. + * @c: UBIFS file-system description object + * @key: node key to lookup + * @node: the node is returned here + * + * This function look up and reads node with key @key. The caller has to make + * sure the @node buffer is large enough to fit the node. Returns zero in case + * of success, %-ENOENT if the node was not found, and a negative error code in + * case of failure. + */ +static inline int ubifs_tnc_lookup(struct ubifs_info *c, + const union ubifs_key *key, void *node) +{ + return ubifs_tnc_locate(c, key, node, NULL, NULL); +} + #endif /* __UBIFS_MISC_H__ */ diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c index f71e6b8822c..3f4902060c7 100644 --- a/fs/ubifs/super.c +++ b/fs/ubifs/super.c @@ -370,8 +370,9 @@ static int ubifs_statfs(struct dentry *dentry, struct kstatfs *buf) { struct ubifs_info *c = dentry->d_sb->s_fs_info; unsigned long long free; + __le32 *uuid = (__le32 *)c->uuid; - free = ubifs_budg_get_free_space(c); + free = ubifs_get_free_space(c); dbg_gen("free space %lld bytes (%lld blocks)", free, free >> UBIFS_BLOCK_SHIFT); @@ -386,7 +387,8 @@ static int ubifs_statfs(struct dentry *dentry, struct kstatfs *buf) buf->f_files = 0; buf->f_ffree = 0; buf->f_namelen = UBIFS_MAX_NLEN; - + buf->f_fsid.val[0] = le32_to_cpu(uuid[0]) ^ le32_to_cpu(uuid[2]); + buf->f_fsid.val[1] = le32_to_cpu(uuid[1]) ^ le32_to_cpu(uuid[3]); return 0; } @@ -530,6 +532,12 @@ static int init_constants_early(struct ubifs_info *c) c->dead_wm = ALIGN(MIN_WRITE_SZ, c->min_io_size); c->dark_wm = ALIGN(UBIFS_MAX_NODE_SZ, c->min_io_size); + /* + * Calculate how many bytes would be wasted at the end of LEB if it was + * fully filled with data nodes of maximum size. This is used in + * calculations when reporting free space. + */ + c->leb_overhead = c->leb_size % UBIFS_MAX_DATA_NODE_SZ; return 0; } @@ -647,13 +655,11 @@ static int init_constants_late(struct ubifs_info *c) * internally because it does not make much sense for UBIFS, but it is * necessary to report something for the 'statfs()' call. * - * Subtract the LEB reserved for GC and the LEB which is reserved for - * deletions. - * - * Review 'ubifs_calc_available()' if changing this calculation. + * Subtract the LEB reserved for GC, the LEB which is reserved for + * deletions, and assume only one journal head is available. */ - tmp64 = c->main_lebs - 2; - tmp64 *= (uint64_t)c->leb_size - c->dark_wm; + tmp64 = c->main_lebs - 2 - c->jhead_cnt + 1; + tmp64 *= (uint64_t)c->leb_size - c->leb_overhead; tmp64 = ubifs_reported_space(c, tmp64); c->block_cnt = tmp64 >> UBIFS_BLOCK_SHIFT; @@ -1018,14 +1024,13 @@ static int mount_ubifs(struct ubifs_info *c) goto out_dereg; } + sprintf(c->bgt_name, BGT_NAME_PATTERN, c->vi.ubi_num, c->vi.vol_id); if (!mounted_read_only) { err = alloc_wbufs(c); if (err) goto out_cbuf; /* Create background thread */ - sprintf(c->bgt_name, BGT_NAME_PATTERN, c->vi.ubi_num, - c->vi.vol_id); c->bgt = kthread_create(ubifs_bg_thread, c, c->bgt_name); if (!c->bgt) c->bgt = ERR_PTR(-EINVAL); diff --git a/fs/ubifs/tnc.c b/fs/ubifs/tnc.c index e909f4a9644..7634c597088 100644 --- a/fs/ubifs/tnc.c +++ b/fs/ubifs/tnc.c @@ -506,7 +506,7 @@ static int fallible_read_node(struct ubifs_info *c, const union ubifs_key *key, if (keys_cmp(c, key, &node_key) != 0) ret = 0; } - if (ret == 0) + if (ret == 0 && c->replaying) dbg_mnt("dangling branch LEB %d:%d len %d, key %s", zbr->lnum, zbr->offs, zbr->len, DBGKEY(key)); return ret; @@ -1382,50 +1382,39 @@ static int lookup_level0_dirty(struct ubifs_info *c, const union ubifs_key *key, } /** - * ubifs_tnc_lookup - look up a file-system node. + * maybe_leb_gced - determine if a LEB may have been garbage collected. * @c: UBIFS file-system description object - * @key: node key to lookup - * @node: the node is returned here + * @lnum: LEB number + * @gc_seq1: garbage collection sequence number * - * This function look up and reads node with key @key. The caller has to make - * sure the @node buffer is large enough to fit the node. Returns zero in case - * of success, %-ENOENT if the node was not found, and a negative error code in - * case of failure. + * This function determines if @lnum may have been garbage collected since + * sequence number @gc_seq1. If it may have been then %1 is returned, otherwise + * %0 is returned. */ -int ubifs_tnc_lookup(struct ubifs_info *c, const union ubifs_key *key, - void *node) +static int maybe_leb_gced(struct ubifs_info *c, int lnum, int gc_seq1) { - int found, n, err; - struct ubifs_znode *znode; - struct ubifs_zbranch zbr, *zt; + int gc_seq2, gced_lnum; - mutex_lock(&c->tnc_mutex); - found = ubifs_lookup_level0(c, key, &znode, &n); - if (!found) { - err = -ENOENT; - goto out; - } else if (found < 0) { - err = found; - goto out; - } - zt = &znode->zbranch[n]; - if (is_hash_key(c, key)) { - /* - * In this case the leaf node cache gets used, so we pass the - * address of the zbranch and keep the mutex locked - */ - err = tnc_read_node_nm(c, zt, node); - goto out; - } - zbr = znode->zbranch[n]; - mutex_unlock(&c->tnc_mutex); - - err = ubifs_tnc_read_node(c, &zbr, node); - return err; - -out: - mutex_unlock(&c->tnc_mutex); - return err; + gced_lnum = c->gced_lnum; + smp_rmb(); + gc_seq2 = c->gc_seq; + /* Same seq means no GC */ + if (gc_seq1 == gc_seq2) + return 0; + /* Different by more than 1 means we don't know */ + if (gc_seq1 + 1 != gc_seq2) + return 1; + /* + * We have seen the sequence number has increased by 1. Now we need to + * be sure we read the right LEB number, so read it again. + */ + smp_rmb(); + if (gced_lnum != c->gced_lnum) + return 1; + /* Finally we can check lnum */ + if (gced_lnum == lnum) + return 1; + return 0; } /** @@ -1436,16 +1425,19 @@ out: * @lnum: LEB number is returned here * @offs: offset is returned here * - * This function is the same as 'ubifs_tnc_lookup()' but it returns the node - * location also. See 'ubifs_tnc_lookup()'. + * This function look up and reads node with key @key. The caller has to make + * sure the @node buffer is large enough to fit the node. Returns zero in case + * of success, %-ENOENT if the node was not found, and a negative error code in + * case of failure. The node location can be returned in @lnum and @offs. */ int ubifs_tnc_locate(struct ubifs_info *c, const union ubifs_key *key, void *node, int *lnum, int *offs) { - int found, n, err; + int found, n, err, safely = 0, gc_seq1; struct ubifs_znode *znode; struct ubifs_zbranch zbr, *zt; +again: mutex_lock(&c->tnc_mutex); found = ubifs_lookup_level0(c, key, &znode, &n); if (!found) { @@ -1456,24 +1448,43 @@ int ubifs_tnc_locate(struct ubifs_info *c, const union ubifs_key *key, goto out; } zt = &znode->zbranch[n]; + if (lnum) { + *lnum = zt->lnum; + *offs = zt->offs; + } if (is_hash_key(c, key)) { /* * In this case the leaf node cache gets used, so we pass the * address of the zbranch and keep the mutex locked */ - *lnum = zt->lnum; - *offs = zt->offs; err = tnc_read_node_nm(c, zt, node); goto out; } + if (safely) { + err = ubifs_tnc_read_node(c, zt, node); + goto out; + } + /* Drop the TNC mutex prematurely and race with garbage collection */ zbr = znode->zbranch[n]; + gc_seq1 = c->gc_seq; mutex_unlock(&c->tnc_mutex); - *lnum = zbr.lnum; - *offs = zbr.offs; + if (ubifs_get_wbuf(c, zbr.lnum)) { + /* We do not GC journal heads */ + err = ubifs_tnc_read_node(c, &zbr, node); + return err; + } - err = ubifs_tnc_read_node(c, &zbr, node); - return err; + err = fallible_read_node(c, key, &zbr, node); + if (err <= 0 || maybe_leb_gced(c, zbr.lnum, gc_seq1)) { + /* + * The node may have been GC'ed out from under us so try again + * while keeping the TNC mutex locked. + */ + safely = 1; + goto again; + } + return 0; out: mutex_unlock(&c->tnc_mutex); @@ -1498,7 +1509,6 @@ static int do_lookup_nm(struct ubifs_info *c, const union ubifs_key *key, { int found, n, err; struct ubifs_znode *znode; - struct ubifs_zbranch zbr; dbg_tnc("name '%.*s' key %s", nm->len, nm->name, DBGKEY(key)); mutex_lock(&c->tnc_mutex); @@ -1522,11 +1532,7 @@ static int do_lookup_nm(struct ubifs_info *c, const union ubifs_key *key, goto out_unlock; } - zbr = znode->zbranch[n]; - mutex_unlock(&c->tnc_mutex); - - err = tnc_read_node_nm(c, &zbr, node); - return err; + err = tnc_read_node_nm(c, &znode->zbranch[n], node); out_unlock: mutex_unlock(&c->tnc_mutex); diff --git a/fs/ubifs/ubifs-media.h b/fs/ubifs/ubifs-media.h index bd2121f3426..a9ecbd9af20 100644 --- a/fs/ubifs/ubifs-media.h +++ b/fs/ubifs/ubifs-media.h @@ -87,7 +87,7 @@ #define UBIFS_SK_LEN 8 /* Minimum index tree fanout */ -#define UBIFS_MIN_FANOUT 2 +#define UBIFS_MIN_FANOUT 3 /* Maximum number of levels in UBIFS indexing B-tree */ #define UBIFS_MAX_LEVELS 512 diff --git a/fs/ubifs/ubifs.h b/fs/ubifs/ubifs.h index d7f706f7a30..17c620b93ee 100644 --- a/fs/ubifs/ubifs.h +++ b/fs/ubifs/ubifs.h @@ -995,6 +995,9 @@ struct ubifs_mount_opts { * @max_idx_node_sz: maximum indexing node aligned on 8-bytes boundary * @max_inode_sz: maximum possible inode size in bytes * @max_znode_sz: size of znode in bytes + * + * @leb_overhead: how many bytes are wasted in an LEB when it is filled with + * data nodes of maximum size - used in free space reporting * @dead_wm: LEB dead space watermark * @dark_wm: LEB dark space watermark * @block_cnt: count of 4KiB blocks on the FS @@ -1028,6 +1031,8 @@ struct ubifs_mount_opts { * @sbuf: a buffer of LEB size used by GC and replay for scanning * @idx_gc: list of index LEBs that have been garbage collected * @idx_gc_cnt: number of elements on the idx_gc list + * @gc_seq: incremented for every non-index LEB garbage collected + * @gced_lnum: last non-index LEB that was garbage collected * * @infos_list: links all 'ubifs_info' objects * @umount_mutex: serializes shrinker and un-mount @@ -1224,6 +1229,8 @@ struct ubifs_info { int max_idx_node_sz; long long max_inode_sz; int max_znode_sz; + + int leb_overhead; int dead_wm; int dark_wm; int block_cnt; @@ -1257,6 +1264,8 @@ struct ubifs_info { void *sbuf; struct list_head idx_gc; int idx_gc_cnt; + volatile int gc_seq; + volatile int gced_lnum; struct list_head infos_list; struct mutex umount_mutex; @@ -1434,9 +1443,10 @@ void ubifs_release_ino_dirty(struct ubifs_info *c, struct inode *inode, struct ubifs_budget_req *req); void ubifs_cancel_ino_op(struct ubifs_info *c, struct inode *inode, struct ubifs_budget_req *req); -long long ubifs_budg_get_free_space(struct ubifs_info *c); +long long ubifs_get_free_space(struct ubifs_info *c); int ubifs_calc_min_idx_lebs(struct ubifs_info *c); void ubifs_convert_page_budget(struct ubifs_info *c); +long long ubifs_reported_space(const struct ubifs_info *c, uint64_t free); long long ubifs_calc_available(const struct ubifs_info *c, int min_idx_lebs); /* find.c */ @@ -1451,8 +1461,6 @@ int ubifs_save_dirty_idx_lnums(struct ubifs_info *c); /* tnc.c */ int ubifs_lookup_level0(struct ubifs_info *c, const union ubifs_key *key, struct ubifs_znode **zn, int *n); -int ubifs_tnc_lookup(struct ubifs_info *c, const union ubifs_key *key, - void *node); int ubifs_tnc_lookup_nm(struct ubifs_info *c, const union ubifs_key *key, void *node, const struct qstr *nm); int ubifs_tnc_locate(struct ubifs_info *c, const union ubifs_key *key, diff --git a/fs/udf/file.c b/fs/udf/file.c index 0ed6e146a0d..eb91f3b7032 100644 --- a/fs/udf/file.c +++ b/fs/udf/file.c @@ -211,6 +211,7 @@ const struct file_operations udf_file_operations = { .release = udf_release_file, .fsync = udf_fsync_file, .splice_read = generic_file_splice_read, + .llseek = generic_file_llseek, }; const struct inode_operations udf_file_inode_operations = { diff --git a/fs/udf/ialloc.c b/fs/udf/ialloc.c index eb9cfa23dc3..a4f2b3ce45b 100644 --- a/fs/udf/ialloc.c +++ b/fs/udf/ialloc.c @@ -76,11 +76,24 @@ struct inode *udf_new_inode(struct inode *dir, int mode, int *err) *err = -ENOSPC; iinfo = UDF_I(inode); - iinfo->i_unique = 0; - iinfo->i_lenExtents = 0; - iinfo->i_next_alloc_block = 0; - iinfo->i_next_alloc_goal = 0; - iinfo->i_strat4096 = 0; + if (UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_USE_EXTENDED_FE)) { + iinfo->i_efe = 1; + if (UDF_VERS_USE_EXTENDED_FE > sbi->s_udfrev) + sbi->s_udfrev = UDF_VERS_USE_EXTENDED_FE; + iinfo->i_ext.i_data = kzalloc(inode->i_sb->s_blocksize - + sizeof(struct extendedFileEntry), + GFP_KERNEL); + } else { + iinfo->i_efe = 0; + iinfo->i_ext.i_data = kzalloc(inode->i_sb->s_blocksize - + sizeof(struct fileEntry), + GFP_KERNEL); + } + if (!iinfo->i_ext.i_data) { + iput(inode); + *err = -ENOMEM; + return NULL; + } block = udf_new_block(dir->i_sb, NULL, dinfo->i_location.partitionReferenceNum, @@ -111,6 +124,7 @@ struct inode *udf_new_inode(struct inode *dir, int mode, int *err) lvhd->uniqueID = cpu_to_le64(uniqueID); mark_buffer_dirty(sbi->s_lvid_bh); } + mutex_unlock(&sbi->s_alloc_mutex); inode->i_mode = mode; inode->i_uid = current->fsuid; if (dir->i_mode & S_ISGID) { @@ -129,25 +143,6 @@ struct inode *udf_new_inode(struct inode *dir, int mode, int *err) iinfo->i_lenEAttr = 0; iinfo->i_lenAlloc = 0; iinfo->i_use = 0; - if (UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_USE_EXTENDED_FE)) { - iinfo->i_efe = 1; - if (UDF_VERS_USE_EXTENDED_FE > sbi->s_udfrev) - sbi->s_udfrev = UDF_VERS_USE_EXTENDED_FE; - iinfo->i_ext.i_data = kzalloc(inode->i_sb->s_blocksize - - sizeof(struct extendedFileEntry), - GFP_KERNEL); - } else { - iinfo->i_efe = 0; - iinfo->i_ext.i_data = kzalloc(inode->i_sb->s_blocksize - - sizeof(struct fileEntry), - GFP_KERNEL); - } - if (!iinfo->i_ext.i_data) { - iput(inode); - *err = -ENOMEM; - mutex_unlock(&sbi->s_alloc_mutex); - return NULL; - } if (UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_USE_AD_IN_ICB)) iinfo->i_alloc_type = ICBTAG_FLAG_AD_IN_ICB; else if (UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_USE_SHORT_AD)) @@ -158,7 +153,6 @@ struct inode *udf_new_inode(struct inode *dir, int mode, int *err) iinfo->i_crtime = current_fs_time(inode->i_sb); insert_inode_hash(inode); mark_inode_dirty(inode); - mutex_unlock(&sbi->s_alloc_mutex); if (DQUOT_ALLOC_INODE(inode)) { DQUOT_DROP(inode); diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c index f42f80a3b1f..a44d68eb50b 100644 --- a/fs/xfs/linux-2.6/xfs_aops.c +++ b/fs/xfs/linux-2.6/xfs_aops.c @@ -1338,6 +1338,10 @@ __xfs_get_blocks( offset = (xfs_off_t)iblock << inode->i_blkbits; ASSERT(bh_result->b_size >= (1 << inode->i_blkbits)); size = bh_result->b_size; + + if (!create && direct && offset >= i_size_read(inode)) + return 0; + error = xfs_iomap(XFS_I(inode), offset, size, create ? flags : BMAPI_READ, &iomap, &niomap); if (error) diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c index 986061ae1b9..36d5fcd3f59 100644 --- a/fs/xfs/linux-2.6/xfs_buf.c +++ b/fs/xfs/linux-2.6/xfs_buf.c @@ -1001,12 +1001,13 @@ xfs_buf_iodone_work( * We can get an EOPNOTSUPP to ordered writes. Here we clear the * ordered flag and reissue them. Because we can't tell the higher * layers directly that they should not issue ordered I/O anymore, they - * need to check if the ordered flag was cleared during I/O completion. + * need to check if the _XFS_BARRIER_FAILED flag was set during I/O completion. */ if ((bp->b_error == EOPNOTSUPP) && (bp->b_flags & (XBF_ORDERED|XBF_ASYNC)) == (XBF_ORDERED|XBF_ASYNC)) { XB_TRACE(bp, "ordered_retry", bp->b_iodone); bp->b_flags &= ~XBF_ORDERED; + bp->b_flags |= _XFS_BARRIER_FAILED; xfs_buf_iorequest(bp); } else if (bp->b_iodone) (*(bp->b_iodone))(bp); diff --git a/fs/xfs/linux-2.6/xfs_buf.h b/fs/xfs/linux-2.6/xfs_buf.h index fe010995665..456519a088c 100644 --- a/fs/xfs/linux-2.6/xfs_buf.h +++ b/fs/xfs/linux-2.6/xfs_buf.h @@ -85,6 +85,14 @@ typedef enum { * modifications being lost. */ _XBF_PAGE_LOCKED = (1 << 22), + + /* + * If we try a barrier write, but it fails we have to communicate + * this to the upper layers. Unfortunately b_error gets overwritten + * when the buffer is re-issued so we have to add another flag to + * keep this information. + */ + _XFS_BARRIER_FAILED = (1 << 23), } xfs_buf_flags_t; typedef enum { diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c index 73c65f19e54..18d3c848783 100644 --- a/fs/xfs/linux-2.6/xfs_super.c +++ b/fs/xfs/linux-2.6/xfs_super.c @@ -1302,9 +1302,29 @@ xfs_fs_remount( mp->m_flags &= ~XFS_MOUNT_BARRIER; break; default: + /* + * Logically we would return an error here to prevent + * users from believing they might have changed + * mount options using remount which can't be changed. + * + * But unfortunately mount(8) adds all options from + * mtab and fstab to the mount arguments in some cases + * so we can't blindly reject options, but have to + * check for each specified option if it actually + * differs from the currently set option and only + * reject it if that's the case. + * + * Until that is implemented we return success for + * every remount request, and silently ignore all + * options that we can't actually change. + */ +#if 0 printk(KERN_INFO "XFS: mount option \"%s\" not supported for remount\n", p); return -EINVAL; +#else + return 0; +#endif } } diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c index 608c30c3f76..002fc2617c8 100644 --- a/fs/xfs/xfs_buf_item.c +++ b/fs/xfs/xfs_buf_item.c @@ -732,6 +732,7 @@ xfs_buf_item_init( bip->bli_item.li_ops = &xfs_buf_item_ops; bip->bli_item.li_mountp = mp; bip->bli_buf = bp; + xfs_buf_hold(bp); bip->bli_format.blf_type = XFS_LI_BUF; bip->bli_format.blf_blkno = (__int64_t)XFS_BUF_ADDR(bp); bip->bli_format.blf_len = (ushort)BTOBB(XFS_BUF_COUNT(bp)); @@ -867,6 +868,21 @@ xfs_buf_item_dirty( return (bip->bli_flags & XFS_BLI_DIRTY); } +STATIC void +xfs_buf_item_free( + xfs_buf_log_item_t *bip) +{ +#ifdef XFS_TRANS_DEBUG + kmem_free(bip->bli_orig); + kmem_free(bip->bli_logged); +#endif /* XFS_TRANS_DEBUG */ + +#ifdef XFS_BLI_TRACE + ktrace_free(bip->bli_trace); +#endif + kmem_zone_free(xfs_buf_item_zone, bip); +} + /* * This is called when the buf log item is no longer needed. It should * free the buf log item associated with the given buffer and clear @@ -887,18 +903,8 @@ xfs_buf_item_relse( (XFS_BUF_IODONE_FUNC(bp) != NULL)) { XFS_BUF_CLR_IODONE_FUNC(bp); } - -#ifdef XFS_TRANS_DEBUG - kmem_free(bip->bli_orig); - bip->bli_orig = NULL; - kmem_free(bip->bli_logged); - bip->bli_logged = NULL; -#endif /* XFS_TRANS_DEBUG */ - -#ifdef XFS_BLI_TRACE - ktrace_free(bip->bli_trace); -#endif - kmem_zone_free(xfs_buf_item_zone, bip); + xfs_buf_rele(bp); + xfs_buf_item_free(bip); } @@ -1120,6 +1126,7 @@ xfs_buf_iodone( ASSERT(bip->bli_buf == bp); + xfs_buf_rele(bp); mp = bip->bli_item.li_mountp; /* @@ -1136,18 +1143,7 @@ xfs_buf_iodone( * xfs_trans_delete_ail() drops the AIL lock. */ xfs_trans_delete_ail(mp, (xfs_log_item_t *)bip); - -#ifdef XFS_TRANS_DEBUG - kmem_free(bip->bli_orig); - bip->bli_orig = NULL; - kmem_free(bip->bli_logged); - bip->bli_logged = NULL; -#endif /* XFS_TRANS_DEBUG */ - -#ifdef XFS_BLI_TRACE - ktrace_free(bip->bli_trace); -#endif - kmem_zone_free(xfs_buf_item_zone, bip); + xfs_buf_item_free(bip); } #if defined(XFS_BLI_TRACE) diff --git a/fs/xfs/xfs_dfrag.c b/fs/xfs/xfs_dfrag.c index 760f4c5b516..75b0cd4da0e 100644 --- a/fs/xfs/xfs_dfrag.c +++ b/fs/xfs/xfs_dfrag.c @@ -149,7 +149,14 @@ xfs_swap_extents( sbp = &sxp->sx_stat; - xfs_lock_two_inodes(ip, tip, lock_flags); + /* + * we have to do two separate lock calls here to keep lockdep + * happy. If we try to get all the locks in one call, lock will + * report false positives when we drop the ILOCK and regain them + * below. + */ + xfs_lock_two_inodes(ip, tip, XFS_IOLOCK_EXCL); + xfs_lock_two_inodes(ip, tip, XFS_ILOCK_EXCL); locked = 1; /* Verify that both files have the same format */ diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index 00e80df9dd9..dbd9cef852e 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -4118,7 +4118,7 @@ xfs_iext_indirect_to_direct( ASSERT(nextents <= XFS_LINEAR_EXTS); size = nextents * sizeof(xfs_bmbt_rec_t); - xfs_iext_irec_compact_full(ifp); + xfs_iext_irec_compact_pages(ifp); ASSERT(ifp->if_real_bytes == XFS_IEXT_BUFSZ); ep = ifp->if_u1.if_ext_irec->er_extbuf; @@ -4449,8 +4449,7 @@ xfs_iext_irec_remove( * compaction policy is as follows: * * Full Compaction: Extents fit into a single page (or inline buffer) - * Full Compaction: Extents occupy less than 10% of allocated space - * Partial Compaction: Extents occupy > 10% and < 50% of allocated space + * Partial Compaction: Extents occupy less than 50% of allocated space * No Compaction: Extents occupy at least 50% of allocated space */ void @@ -4471,8 +4470,6 @@ xfs_iext_irec_compact( xfs_iext_direct_to_inline(ifp, nextents); } else if (nextents <= XFS_LINEAR_EXTS) { xfs_iext_indirect_to_direct(ifp); - } else if (nextents < (nlists * XFS_LINEAR_EXTS) >> 3) { - xfs_iext_irec_compact_full(ifp); } else if (nextents < (nlists * XFS_LINEAR_EXTS) >> 1) { xfs_iext_irec_compact_pages(ifp); } @@ -4496,7 +4493,7 @@ xfs_iext_irec_compact_pages( erp_next = erp + 1; if (erp_next->er_extcount <= (XFS_LINEAR_EXTS - erp->er_extcount)) { - memmove(&erp->er_extbuf[erp->er_extcount], + memcpy(&erp->er_extbuf[erp->er_extcount], erp_next->er_extbuf, erp_next->er_extcount * sizeof(xfs_bmbt_rec_t)); erp->er_extcount += erp_next->er_extcount; @@ -4516,91 +4513,6 @@ xfs_iext_irec_compact_pages( } /* - * Fully compact the extent records managed by the indirection array. - */ -void -xfs_iext_irec_compact_full( - xfs_ifork_t *ifp) /* inode fork pointer */ -{ - xfs_bmbt_rec_host_t *ep, *ep_next; /* extent record pointers */ - xfs_ext_irec_t *erp, *erp_next; /* extent irec pointers */ - int erp_idx = 0; /* extent irec index */ - int ext_avail; /* empty entries in ex list */ - int ext_diff; /* number of exts to add */ - int nlists; /* number of irec's (ex lists) */ - - ASSERT(ifp->if_flags & XFS_IFEXTIREC); - - nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ; - erp = ifp->if_u1.if_ext_irec; - ep = &erp->er_extbuf[erp->er_extcount]; - erp_next = erp + 1; - ep_next = erp_next->er_extbuf; - - while (erp_idx < nlists - 1) { - /* - * Check how many extent records are available in this irec. - * If there is none skip the whole exercise. - */ - ext_avail = XFS_LINEAR_EXTS - erp->er_extcount; - if (ext_avail) { - - /* - * Copy over as many as possible extent records into - * the previous page. - */ - ext_diff = MIN(ext_avail, erp_next->er_extcount); - memcpy(ep, ep_next, ext_diff * sizeof(xfs_bmbt_rec_t)); - erp->er_extcount += ext_diff; - erp_next->er_extcount -= ext_diff; - - /* - * If the next irec is empty now we can simply - * remove it. - */ - if (erp_next->er_extcount == 0) { - /* - * Free page before removing extent record - * so er_extoffs don't get modified in - * xfs_iext_irec_remove. - */ - kmem_free(erp_next->er_extbuf); - erp_next->er_extbuf = NULL; - xfs_iext_irec_remove(ifp, erp_idx + 1); - erp = &ifp->if_u1.if_ext_irec[erp_idx]; - nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ; - - /* - * If the next irec is not empty move up the content - * that has not been copied to the previous page to - * the beggining of this one. - */ - } else { - memmove(erp_next->er_extbuf, &ep_next[ext_diff], - erp_next->er_extcount * - sizeof(xfs_bmbt_rec_t)); - ep_next = erp_next->er_extbuf; - memset(&ep_next[erp_next->er_extcount], 0, - (XFS_LINEAR_EXTS - - erp_next->er_extcount) * - sizeof(xfs_bmbt_rec_t)); - } - } - - if (erp->er_extcount == XFS_LINEAR_EXTS) { - erp_idx++; - if (erp_idx < nlists) - erp = &ifp->if_u1.if_ext_irec[erp_idx]; - else - break; - } - ep = &erp->er_extbuf[erp->er_extcount]; - erp_next = erp + 1; - ep_next = erp_next->er_extbuf; - } -} - -/* * This is called to update the er_extoff field in the indirection * array when extents have been added or removed from one of the * extent lists. erp_idx contains the irec index to begin updating diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c index ccba14eb9db..0b02c644355 100644 --- a/fs/xfs/xfs_log.c +++ b/fs/xfs/xfs_log.c @@ -124,16 +124,27 @@ STATIC void xlog_verify_tail_lsn(xlog_t *log, xlog_in_core_t *iclog, STATIC int xlog_iclogs_empty(xlog_t *log); #if defined(XFS_LOG_TRACE) + +#define XLOG_TRACE_LOGGRANT_SIZE 2048 +#define XLOG_TRACE_ICLOG_SIZE 256 + +void +xlog_trace_loggrant_alloc(xlog_t *log) +{ + log->l_grant_trace = ktrace_alloc(XLOG_TRACE_LOGGRANT_SIZE, KM_NOFS); +} + +void +xlog_trace_loggrant_dealloc(xlog_t *log) +{ + ktrace_free(log->l_grant_trace); +} + void xlog_trace_loggrant(xlog_t *log, xlog_ticket_t *tic, xfs_caddr_t string) { unsigned long cnts; - if (!log->l_grant_trace) { - log->l_grant_trace = ktrace_alloc(2048, KM_NOSLEEP); - if (!log->l_grant_trace) - return; - } /* ticket counts are 1 byte each */ cnts = ((unsigned long)tic->t_ocnt) | ((unsigned long)tic->t_cnt) << 8; @@ -157,10 +168,20 @@ xlog_trace_loggrant(xlog_t *log, xlog_ticket_t *tic, xfs_caddr_t string) } void +xlog_trace_iclog_alloc(xlog_in_core_t *iclog) +{ + iclog->ic_trace = ktrace_alloc(XLOG_TRACE_ICLOG_SIZE, KM_NOFS); +} + +void +xlog_trace_iclog_dealloc(xlog_in_core_t *iclog) +{ + ktrace_free(iclog->ic_trace); +} + +void xlog_trace_iclog(xlog_in_core_t *iclog, uint state) { - if (!iclog->ic_trace) - iclog->ic_trace = ktrace_alloc(256, KM_NOFS); ktrace_enter(iclog->ic_trace, (void *)((unsigned long)state), (void *)((unsigned long)current_pid()), @@ -170,8 +191,15 @@ xlog_trace_iclog(xlog_in_core_t *iclog, uint state) (void *)NULL, (void *)NULL); } #else + +#define xlog_trace_loggrant_alloc(log) +#define xlog_trace_loggrant_dealloc(log) #define xlog_trace_loggrant(log,tic,string) + +#define xlog_trace_iclog_alloc(iclog) +#define xlog_trace_iclog_dealloc(iclog) #define xlog_trace_iclog(iclog,state) + #endif /* XFS_LOG_TRACE */ @@ -1005,11 +1033,12 @@ xlog_iodone(xfs_buf_t *bp) l = iclog->ic_log; /* - * If the ordered flag has been removed by a lower - * layer, it means the underlyin device no longer supports + * If the _XFS_BARRIER_FAILED flag was set by a lower + * layer, it means the underlying device no longer supports * barrier I/O. Warn loudly and turn off barriers. */ - if ((l->l_mp->m_flags & XFS_MOUNT_BARRIER) && !XFS_BUF_ORDERED(bp)) { + if (bp->b_flags & _XFS_BARRIER_FAILED) { + bp->b_flags &= ~_XFS_BARRIER_FAILED; l->l_mp->m_flags &= ~XFS_MOUNT_BARRIER; xfs_fs_cmn_err(CE_WARN, l->l_mp, "xlog_iodone: Barriers are no longer supported" @@ -1231,6 +1260,7 @@ xlog_alloc_log(xfs_mount_t *mp, spin_lock_init(&log->l_grant_lock); sv_init(&log->l_flush_wait, 0, "flush_wait"); + xlog_trace_loggrant_alloc(log); /* log record size must be multiple of BBSIZE; see xlog_rec_header_t */ ASSERT((XFS_BUF_SIZE(bp) & BBMASK) == 0); @@ -1285,6 +1315,8 @@ xlog_alloc_log(xfs_mount_t *mp, sv_init(&iclog->ic_force_wait, SV_DEFAULT, "iclog-force"); sv_init(&iclog->ic_write_wait, SV_DEFAULT, "iclog-write"); + xlog_trace_iclog_alloc(iclog); + iclogp = &iclog->ic_next; } *iclogp = log->l_iclog; /* complete ring */ @@ -1565,11 +1597,7 @@ xlog_dealloc_log(xlog_t *log) sv_destroy(&iclog->ic_force_wait); sv_destroy(&iclog->ic_write_wait); xfs_buf_free(iclog->ic_bp); -#ifdef XFS_LOG_TRACE - if (iclog->ic_trace != NULL) { - ktrace_free(iclog->ic_trace); - } -#endif + xlog_trace_iclog_dealloc(iclog); next_iclog = iclog->ic_next; kmem_free(iclog); iclog = next_iclog; @@ -1578,14 +1606,7 @@ xlog_dealloc_log(xlog_t *log) spinlock_destroy(&log->l_grant_lock); xfs_buf_free(log->l_xbuf); -#ifdef XFS_LOG_TRACE - if (log->l_trace != NULL) { - ktrace_free(log->l_trace); - } - if (log->l_grant_trace != NULL) { - ktrace_free(log->l_grant_trace); - } -#endif + xlog_trace_loggrant_dealloc(log); log->l_mp->m_log = NULL; kmem_free(log); } /* xlog_dealloc_log */ diff --git a/fs/xfs/xfs_log_priv.h b/fs/xfs/xfs_log_priv.h index c8a5b22ee3e..e7d8f84443f 100644 --- a/fs/xfs/xfs_log_priv.h +++ b/fs/xfs/xfs_log_priv.h @@ -448,7 +448,6 @@ typedef struct log { int l_grant_write_bytes; #ifdef XFS_LOG_TRACE - struct ktrace *l_trace; struct ktrace *l_grant_trace; #endif diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c index aa238c8fbd7..8b6812f66a1 100644 --- a/fs/xfs/xfs_vnodeops.c +++ b/fs/xfs/xfs_vnodeops.c @@ -1838,6 +1838,12 @@ again: #endif } +/* + * xfs_lock_two_inodes() can only be used to lock one type of lock + * at a time - the iolock or the ilock, but not both at once. If + * we lock both at once, lockdep will report false positives saying + * we have violated locking orders. + */ void xfs_lock_two_inodes( xfs_inode_t *ip0, @@ -1848,6 +1854,8 @@ xfs_lock_two_inodes( int attempts = 0; xfs_log_item_t *lp; + if (lock_mode & (XFS_IOLOCK_SHARED|XFS_IOLOCK_EXCL)) + ASSERT((lock_mode & (XFS_ILOCK_SHARED|XFS_ILOCK_EXCL)) == 0); ASSERT(ip0->i_ino != ip1->i_ino); if (ip0->i_ino > ip1->i_ino) { @@ -3152,6 +3160,13 @@ error1: /* Just cancel transaction */ /* * Zero file bytes between startoff and endoff inclusive. * The iolock is held exclusive and no blocks are buffered. + * + * This function is used by xfs_free_file_space() to zero + * partial blocks when the range to free is not block aligned. + * When unreserving space with boundaries that are not block + * aligned we round up the start and round down the end + * boundaries and then use this function to zero the parts of + * the blocks that got dropped during the rounding. */ STATIC int xfs_zero_remaining_bytes( @@ -3168,6 +3183,17 @@ xfs_zero_remaining_bytes( int nimap; int error = 0; + /* + * Avoid doing I/O beyond eof - it's not necessary + * since nothing can read beyond eof. The space will + * be zeroed when the file is extended anyway. + */ + if (startoff >= ip->i_size) + return 0; + + if (endoff > ip->i_size) + endoff = ip->i_size; + bp = xfs_buf_get_noaddr(mp->m_sb.sb_blocksize, XFS_IS_REALTIME_INODE(ip) ? mp->m_rtdev_targp : mp->m_ddev_targp); |