diff options
Diffstat (limited to 'fs')
-rw-r--r-- | fs/cifs/readdir.c | 11 | ||||
-rw-r--r-- | fs/ext2/balloc.c | 1 | ||||
-rw-r--r-- | fs/ext2/ialloc.c | 1 | ||||
-rw-r--r-- | fs/ext2/super.c | 41 | ||||
-rw-r--r-- | fs/ext3/inode.c | 13 | ||||
-rw-r--r-- | fs/ext3/super.c | 42 | ||||
-rw-r--r-- | fs/jbd/transaction.c | 2 | ||||
-rw-r--r-- | fs/nfs/direct.c | 50 | ||||
-rw-r--r-- | fs/nfs/read.c | 24 | ||||
-rw-r--r-- | fs/nfs/write.c | 37 | ||||
-rw-r--r-- | fs/super.c | 1 | ||||
-rw-r--r-- | fs/xfs/linux-2.6/xfs_aops.c | 18 | ||||
-rw-r--r-- | fs/xfs/linux-2.6/xfs_lrw.c | 27 | ||||
-rw-r--r-- | fs/xfs/quota/xfs_qm_bhv.c | 2 | ||||
-rw-r--r-- | fs/xfs/xfs_alloc.h | 20 | ||||
-rw-r--r-- | fs/xfs/xfs_bmap.c | 2 | ||||
-rw-r--r-- | fs/xfs/xfs_fsops.c | 16 | ||||
-rw-r--r-- | fs/xfs/xfs_mount.c | 32 | ||||
-rw-r--r-- | fs/xfs/xfs_vfsops.c | 3 |
19 files changed, 217 insertions, 126 deletions
diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c index 105761e..9aeb58a 100644 --- a/fs/cifs/readdir.c +++ b/fs/cifs/readdir.c @@ -82,7 +82,6 @@ static int construct_dentry(struct qstr *qstring, struct file *file, if(*ptmp_inode == NULL) return rc; rc = 1; - d_instantiate(tmp_dentry, *ptmp_inode); } } else { tmp_dentry = d_alloc(file->f_dentry, qstring); @@ -99,9 +98,7 @@ static int construct_dentry(struct qstr *qstring, struct file *file, tmp_dentry->d_op = &cifs_dentry_ops; if(*ptmp_inode == NULL) return rc; - rc = 1; - d_instantiate(tmp_dentry, *ptmp_inode); - d_rehash(tmp_dentry); + rc = 2; } tmp_dentry->d_time = jiffies; @@ -870,6 +867,12 @@ static int cifs_filldir(char *pfindEntry, struct file *file, pfindEntry, &obj_type, rc); else fill_in_inode(tmp_inode, 1 /* NT */, pfindEntry, &obj_type, rc); + + if(rc) /* new inode - needs to be tied to dentry */ { + d_instantiate(tmp_dentry, tmp_inode); + if(rc == 2) + d_rehash(tmp_dentry); + } rc = filldir(direntry,qstring.name,qstring.len,file->f_pos, diff --git a/fs/ext2/balloc.c b/fs/ext2/balloc.c index d487043..b1981d0 100644 --- a/fs/ext2/balloc.c +++ b/fs/ext2/balloc.c @@ -539,7 +539,6 @@ unsigned long ext2_count_free (struct buffer_head * map, unsigned int numchars) #endif /* EXT2FS_DEBUG */ -/* Superblock must be locked */ unsigned long ext2_count_free_blocks (struct super_block * sb) { struct ext2_group_desc * desc; diff --git a/fs/ext2/ialloc.c b/fs/ext2/ialloc.c index de85c61..695f69c 100644 --- a/fs/ext2/ialloc.c +++ b/fs/ext2/ialloc.c @@ -637,7 +637,6 @@ fail: return ERR_PTR(err); } -/* Superblock must be locked */ unsigned long ext2_count_free_inodes (struct super_block * sb) { struct ext2_group_desc *desc; diff --git a/fs/ext2/super.c b/fs/ext2/super.c index 681dea8..4286ff6 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -251,6 +251,44 @@ static struct super_operations ext2_sops = { #endif }; +static struct dentry *ext2_get_dentry(struct super_block *sb, void *vobjp) +{ + __u32 *objp = vobjp; + unsigned long ino = objp[0]; + __u32 generation = objp[1]; + struct inode *inode; + struct dentry *result; + + if (ino < EXT2_FIRST_INO(sb) && ino != EXT2_ROOT_INO) + return ERR_PTR(-ESTALE); + if (ino > le32_to_cpu(EXT2_SB(sb)->s_es->s_inodes_count)) + return ERR_PTR(-ESTALE); + + /* iget isn't really right if the inode is currently unallocated!! + * ext2_read_inode currently does appropriate checks, but + * it might be "neater" to call ext2_get_inode first and check + * if the inode is valid..... + */ + inode = iget(sb, ino); + if (inode == NULL) + return ERR_PTR(-ENOMEM); + if (is_bad_inode(inode) || + (generation && inode->i_generation != generation)) { + /* we didn't find the right inode.. */ + iput(inode); + return ERR_PTR(-ESTALE); + } + /* now to find a dentry. + * If possible, get a well-connected one + */ + result = d_alloc_anon(inode); + if (!result) { + iput(inode); + return ERR_PTR(-ENOMEM); + } + return result; +} + /* Yes, most of these are left as NULL!! * A NULL value implies the default, which works with ext2-like file * systems, but can be improved upon. @@ -258,6 +296,7 @@ static struct super_operations ext2_sops = { */ static struct export_operations ext2_export_ops = { .get_parent = ext2_get_parent, + .get_dentry = ext2_get_dentry, }; static unsigned long get_sb_block(void **data) @@ -1044,7 +1083,6 @@ static int ext2_statfs (struct dentry * dentry, struct kstatfs * buf) unsigned long overhead; int i; - lock_super(sb); if (test_opt (sb, MINIX_DF)) overhead = 0; else { @@ -1085,7 +1123,6 @@ static int ext2_statfs (struct dentry * dentry, struct kstatfs * buf) buf->f_files = le32_to_cpu(sbi->s_es->s_inodes_count); buf->f_ffree = ext2_count_free_inodes (sb); buf->f_namelen = EXT2_NAME_LEN; - unlock_super(sb); return 0; } diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c index c5ee9f0..84be02e 100644 --- a/fs/ext3/inode.c +++ b/fs/ext3/inode.c @@ -925,7 +925,7 @@ int ext3_get_blocks_handle(handle_t *handle, struct inode *inode, set_buffer_new(bh_result); got_it: map_bh(bh_result, inode->i_sb, le32_to_cpu(chain[depth-1].key)); - if (blocks_to_boundary == 0) + if (count > blocks_to_boundary) set_buffer_boundary(bh_result); err = count; /* Clean up and exit */ @@ -1009,11 +1009,14 @@ struct buffer_head *ext3_getblk(handle_t *handle, struct inode *inode, buffer_trace_init(&dummy.b_history); err = ext3_get_blocks_handle(handle, inode, block, 1, &dummy, create, 1); - if (err == 1) { + /* + * ext3_get_blocks_handle() returns number of blocks + * mapped. 0 in case of a HOLE. + */ + if (err > 0) { + if (err > 1) + WARN_ON(1); err = 0; - } else if (err >= 0) { - WARN_ON(1); - err = -EIO; } *errp = err; if (!err && buffer_mapped(&dummy)) { diff --git a/fs/ext3/super.c b/fs/ext3/super.c index 813d589..3559086 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -554,6 +554,47 @@ static int ext3_show_options(struct seq_file *seq, struct vfsmount *vfs) return 0; } + +static struct dentry *ext3_get_dentry(struct super_block *sb, void *vobjp) +{ + __u32 *objp = vobjp; + unsigned long ino = objp[0]; + __u32 generation = objp[1]; + struct inode *inode; + struct dentry *result; + + if (ino < EXT3_FIRST_INO(sb) && ino != EXT3_ROOT_INO) + return ERR_PTR(-ESTALE); + if (ino > le32_to_cpu(EXT3_SB(sb)->s_es->s_inodes_count)) + return ERR_PTR(-ESTALE); + + /* iget isn't really right if the inode is currently unallocated!! + * + * ext3_read_inode will return a bad_inode if the inode had been + * deleted, so we should be safe. + * + * Currently we don't know the generation for parent directory, so + * a generation of 0 means "accept any" + */ + inode = iget(sb, ino); + if (inode == NULL) + return ERR_PTR(-ENOMEM); + if (is_bad_inode(inode) || + (generation && inode->i_generation != generation)) { + iput(inode); + return ERR_PTR(-ESTALE); + } + /* now to find a dentry. + * If possible, get a well-connected one + */ + result = d_alloc_anon(inode); + if (!result) { + iput(inode); + return ERR_PTR(-ENOMEM); + } + return result; +} + #ifdef CONFIG_QUOTA #define QTYPE2NAME(t) ((t)==USRQUOTA?"user":"group") #define QTYPE2MOPT(on, t) ((t)==USRQUOTA?((on)##USRJQUOTA):((on)##GRPJQUOTA)) @@ -622,6 +663,7 @@ static struct super_operations ext3_sops = { static struct export_operations ext3_export_ops = { .get_parent = ext3_get_parent, + .get_dentry = ext3_get_dentry, }; enum { diff --git a/fs/jbd/transaction.c b/fs/jbd/transaction.c index de2e4cb..f5169a9 100644 --- a/fs/jbd/transaction.c +++ b/fs/jbd/transaction.c @@ -727,7 +727,7 @@ done: out: if (unlikely(frozen_buffer)) /* It's usually NULL */ - kfree(frozen_buffer); + jbd_slab_free(frozen_buffer, bh->b_size); JBUFFER_TRACE(jh, "exit"); return error; diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c index fecd3b09..76ca1cb 100644 --- a/fs/nfs/direct.c +++ b/fs/nfs/direct.c @@ -100,25 +100,6 @@ static inline int put_dreq(struct nfs_direct_req *dreq) return atomic_dec_and_test(&dreq->io_count); } -/* - * "size" is never larger than rsize or wsize. - */ -static inline int nfs_direct_count_pages(unsigned long user_addr, size_t size) -{ - int page_count; - - page_count = (user_addr + size + PAGE_SIZE - 1) >> PAGE_SHIFT; - page_count -= user_addr >> PAGE_SHIFT; - BUG_ON(page_count < 0); - - return page_count; -} - -static inline unsigned int nfs_max_pages(unsigned int size) -{ - return (size + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; -} - /** * nfs_direct_IO - NFS address space operation for direct I/O * @rw: direction (read or write) @@ -276,28 +257,24 @@ static ssize_t nfs_direct_read_schedule(struct nfs_direct_req *dreq, unsigned lo struct nfs_open_context *ctx = dreq->ctx; struct inode *inode = ctx->dentry->d_inode; size_t rsize = NFS_SERVER(inode)->rsize; - unsigned int rpages = nfs_max_pages(rsize); unsigned int pgbase; int result; ssize_t started = 0; get_dreq(dreq); - pgbase = user_addr & ~PAGE_MASK; do { struct nfs_read_data *data; size_t bytes; + pgbase = user_addr & ~PAGE_MASK; + bytes = min(rsize,count); + result = -ENOMEM; - data = nfs_readdata_alloc(rpages); + data = nfs_readdata_alloc(pgbase + bytes); if (unlikely(!data)) break; - bytes = rsize; - if (count < rsize) - bytes = count; - - data->npages = nfs_direct_count_pages(user_addr, bytes); down_read(¤t->mm->mmap_sem); result = get_user_pages(current, current->mm, user_addr, data->npages, 1, 0, data->pagevec, NULL); @@ -344,8 +321,10 @@ static ssize_t nfs_direct_read_schedule(struct nfs_direct_req *dreq, unsigned lo started += bytes; user_addr += bytes; pos += bytes; + /* FIXME: Remove this unnecessary math from final patch */ pgbase += bytes; pgbase &= ~PAGE_MASK; + BUG_ON(pgbase != (user_addr & ~PAGE_MASK)); count -= bytes; } while (count != 0); @@ -524,7 +503,7 @@ static void nfs_direct_write_complete(struct nfs_direct_req *dreq, struct inode static void nfs_alloc_commit_data(struct nfs_direct_req *dreq) { - dreq->commit_data = nfs_commit_alloc(0); + dreq->commit_data = nfs_commit_alloc(); if (dreq->commit_data != NULL) dreq->commit_data->req = (struct nfs_page *) dreq; } @@ -605,28 +584,24 @@ static ssize_t nfs_direct_write_schedule(struct nfs_direct_req *dreq, unsigned l struct nfs_open_context *ctx = dreq->ctx; struct inode *inode = ctx->dentry->d_inode; size_t wsize = NFS_SERVER(inode)->wsize; - unsigned int wpages = nfs_max_pages(wsize); unsigned int pgbase; int result; ssize_t started = 0; get_dreq(dreq); - pgbase = user_addr & ~PAGE_MASK; do { struct nfs_write_data *data; size_t bytes; + pgbase = user_addr & ~PAGE_MASK; + bytes = min(wsize,count); + result = -ENOMEM; - data = nfs_writedata_alloc(wpages); + data = nfs_writedata_alloc(pgbase + bytes); if (unlikely(!data)) break; - bytes = wsize; - if (count < wsize) - bytes = count; - - data->npages = nfs_direct_count_pages(user_addr, bytes); down_read(¤t->mm->mmap_sem); result = get_user_pages(current, current->mm, user_addr, data->npages, 0, 0, data->pagevec, NULL); @@ -676,8 +651,11 @@ static ssize_t nfs_direct_write_schedule(struct nfs_direct_req *dreq, unsigned l started += bytes; user_addr += bytes; pos += bytes; + + /* FIXME: Remove this useless math from the final patch */ pgbase += bytes; pgbase &= ~PAGE_MASK; + BUG_ON(pgbase != (user_addr & ~PAGE_MASK)); count -= bytes; } while (count != 0); diff --git a/fs/nfs/read.c b/fs/nfs/read.c index da9cf11..7a9ee00 100644 --- a/fs/nfs/read.c +++ b/fs/nfs/read.c @@ -43,13 +43,15 @@ static mempool_t *nfs_rdata_mempool; #define MIN_POOL_READ (32) -struct nfs_read_data *nfs_readdata_alloc(unsigned int pagecount) +struct nfs_read_data *nfs_readdata_alloc(size_t len) { + unsigned int pagecount = (len + PAGE_SIZE - 1) >> PAGE_SHIFT; struct nfs_read_data *p = mempool_alloc(nfs_rdata_mempool, SLAB_NOFS); if (p) { memset(p, 0, sizeof(*p)); INIT_LIST_HEAD(&p->pages); + p->npages = pagecount; if (pagecount <= ARRAY_SIZE(p->page_array)) p->pagevec = p->page_array; else { @@ -140,7 +142,7 @@ static int nfs_readpage_sync(struct nfs_open_context *ctx, struct inode *inode, int result; struct nfs_read_data *rdata; - rdata = nfs_readdata_alloc(1); + rdata = nfs_readdata_alloc(count); if (!rdata) return -ENOMEM; @@ -336,25 +338,25 @@ static int nfs_pagein_multi(struct list_head *head, struct inode *inode) struct nfs_page *req = nfs_list_entry(head->next); struct page *page = req->wb_page; struct nfs_read_data *data; - unsigned int rsize = NFS_SERVER(inode)->rsize; - unsigned int nbytes, offset; + size_t rsize = NFS_SERVER(inode)->rsize, nbytes; + unsigned int offset; int requests = 0; LIST_HEAD(list); nfs_list_remove_request(req); nbytes = req->wb_bytes; - for(;;) { - data = nfs_readdata_alloc(1); + do { + size_t len = min(nbytes,rsize); + + data = nfs_readdata_alloc(len); if (!data) goto out_bad; INIT_LIST_HEAD(&data->pages); list_add(&data->pages, &list); requests++; - if (nbytes <= rsize) - break; - nbytes -= rsize; - } + nbytes -= len; + } while(nbytes != 0); atomic_set(&req->wb_complete, requests); ClearPageError(page); @@ -402,7 +404,7 @@ static int nfs_pagein_one(struct list_head *head, struct inode *inode) if (NFS_SERVER(inode)->rsize < PAGE_CACHE_SIZE) return nfs_pagein_multi(head, inode); - data = nfs_readdata_alloc(NFS_SERVER(inode)->rpages); + data = nfs_readdata_alloc(NFS_SERVER(inode)->rsize); if (!data) goto out_bad; diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 5077499..8ab3cf1 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -90,22 +90,13 @@ static mempool_t *nfs_commit_mempool; static DECLARE_WAIT_QUEUE_HEAD(nfs_write_congestion); -struct nfs_write_data *nfs_commit_alloc(unsigned int pagecount) +struct nfs_write_data *nfs_commit_alloc(void) { struct nfs_write_data *p = mempool_alloc(nfs_commit_mempool, SLAB_NOFS); if (p) { memset(p, 0, sizeof(*p)); INIT_LIST_HEAD(&p->pages); - if (pagecount <= ARRAY_SIZE(p->page_array)) - p->pagevec = p->page_array; - else { - p->pagevec = kcalloc(pagecount, sizeof(struct page *), GFP_NOFS); - if (!p->pagevec) { - mempool_free(p, nfs_commit_mempool); - p = NULL; - } - } } return p; } @@ -117,13 +108,15 @@ void nfs_commit_free(struct nfs_write_data *p) mempool_free(p, nfs_commit_mempool); } -struct nfs_write_data *nfs_writedata_alloc(unsigned int pagecount) +struct nfs_write_data *nfs_writedata_alloc(size_t len) { + unsigned int pagecount = (len + PAGE_SIZE - 1) >> PAGE_SHIFT; struct nfs_write_data *p = mempool_alloc(nfs_wdata_mempool, SLAB_NOFS); if (p) { memset(p, 0, sizeof(*p)); INIT_LIST_HEAD(&p->pages); + p->npages = pagecount; if (pagecount <= ARRAY_SIZE(p->page_array)) p->pagevec = p->page_array; else { @@ -208,7 +201,7 @@ static int nfs_writepage_sync(struct nfs_open_context *ctx, struct inode *inode, int result, written = 0; struct nfs_write_data *wdata; - wdata = nfs_writedata_alloc(1); + wdata = nfs_writedata_alloc(wsize); if (!wdata) return -ENOMEM; @@ -999,24 +992,24 @@ static int nfs_flush_multi(struct inode *inode, struct list_head *head, int how) struct nfs_page *req = nfs_list_entry(head->next); struct page *page = req->wb_page; struct nfs_write_data *data; - unsigned int wsize = NFS_SERVER(inode)->wsize; - unsigned int nbytes, offset; + size_t wsize = NFS_SERVER(inode)->wsize, nbytes; + unsigned int offset; int requests = 0; LIST_HEAD(list); nfs_list_remove_request(req); nbytes = req->wb_bytes; - for (;;) { - data = nfs_writedata_alloc(1); + do { + size_t len = min(nbytes, wsize); + + data = nfs_writedata_alloc(len); if (!data) goto out_bad; list_add(&data->pages, &list); requests++; - if (nbytes <= wsize) - break; - nbytes -= wsize; - } + nbytes -= len; + } while (nbytes != 0); atomic_set(&req->wb_complete, requests); ClearPageError(page); @@ -1070,7 +1063,7 @@ static int nfs_flush_one(struct inode *inode, struct list_head *head, int how) struct nfs_write_data *data; unsigned int count; - data = nfs_writedata_alloc(NFS_SERVER(inode)->wpages); + data = nfs_writedata_alloc(NFS_SERVER(inode)->wsize); if (!data) goto out_bad; @@ -1378,7 +1371,7 @@ nfs_commit_list(struct inode *inode, struct list_head *head, int how) struct nfs_write_data *data; struct nfs_page *req; - data = nfs_commit_alloc(NFS_SERVER(inode)->wpages); + data = nfs_commit_alloc(); if (!data) goto out_bad; @@ -49,6 +49,7 @@ DEFINE_SPINLOCK(sb_lock); /** * alloc_super - create new superblock + * @type: filesystem type superblock should belong to * * Allocates and initializes a new &struct super_block. alloc_super() * returns a pointer new superblock or %NULL if allocation had failed. diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c index c40f81b..34dcb43 100644 --- a/fs/xfs/linux-2.6/xfs_aops.c +++ b/fs/xfs/linux-2.6/xfs_aops.c @@ -1390,11 +1390,19 @@ xfs_vm_direct_IO( iocb->private = xfs_alloc_ioend(inode, IOMAP_UNWRITTEN); - ret = blockdev_direct_IO_own_locking(rw, iocb, inode, - iomap.iomap_target->bt_bdev, - iov, offset, nr_segs, - xfs_get_blocks_direct, - xfs_end_io_direct); + if (rw == WRITE) { + ret = blockdev_direct_IO_own_locking(rw, iocb, inode, + iomap.iomap_target->bt_bdev, + iov, offset, nr_segs, + xfs_get_blocks_direct, + xfs_end_io_direct); + } else { + ret = blockdev_direct_IO_no_locking(rw, iocb, inode, + iomap.iomap_target->bt_bdev, + iov, offset, nr_segs, + xfs_get_blocks_direct, + xfs_end_io_direct); + } if (unlikely(ret <= 0 && iocb->private)) xfs_destroy_ioend(iocb->private); diff --git a/fs/xfs/linux-2.6/xfs_lrw.c b/fs/xfs/linux-2.6/xfs_lrw.c index 5d9cfd9..ee788b1 100644 --- a/fs/xfs/linux-2.6/xfs_lrw.c +++ b/fs/xfs/linux-2.6/xfs_lrw.c @@ -264,7 +264,9 @@ xfs_read( dmflags, &locktype); if (ret) { xfs_iunlock(ip, XFS_IOLOCK_SHARED); - goto unlock_mutex; + if (unlikely(ioflags & IO_ISDIRECT)) + mutex_unlock(&inode->i_mutex); + return ret; } } @@ -272,6 +274,9 @@ xfs_read( bhv_vop_flushinval_pages(vp, ctooff(offtoct(*offset)), -1, FI_REMAPF_LOCKED); + if (unlikely(ioflags & IO_ISDIRECT)) + mutex_unlock(&inode->i_mutex); + xfs_rw_enter_trace(XFS_READ_ENTER, &ip->i_iocore, (void *)iovp, segs, *offset, ioflags); ret = __generic_file_aio_read(iocb, iovp, segs, offset); @@ -281,10 +286,6 @@ xfs_read( XFS_STATS_ADD(xs_read_bytes, ret); xfs_iunlock(ip, XFS_IOLOCK_SHARED); - -unlock_mutex: - if (unlikely(ioflags & IO_ISDIRECT)) - mutex_unlock(&inode->i_mutex); return ret; } @@ -390,6 +391,8 @@ xfs_splice_write( xfs_inode_t *ip = XFS_BHVTOI(bdp); xfs_mount_t *mp = ip->i_mount; ssize_t ret; + struct inode *inode = outfilp->f_mapping->host; + xfs_fsize_t isize; XFS_STATS_INC(xs_write_calls); if (XFS_FORCED_SHUTDOWN(ip->i_mount)) @@ -416,6 +419,20 @@ xfs_splice_write( if (ret > 0) XFS_STATS_ADD(xs_write_bytes, ret); + isize = i_size_read(inode); + if (unlikely(ret < 0 && ret != -EFAULT && *ppos > isize)) + *ppos = isize; + + if (*ppos > ip->i_d.di_size) { + xfs_ilock(ip, XFS_ILOCK_EXCL); + if (*ppos > ip->i_d.di_size) { + ip->i_d.di_size = *ppos; + i_size_write(inode, *ppos); + ip->i_update_core = 1; + ip->i_update_size = 1; + } + xfs_iunlock(ip, XFS_ILOCK_EXCL); + } xfs_iunlock(ip, XFS_IOLOCK_EXCL); return ret; } diff --git a/fs/xfs/quota/xfs_qm_bhv.c b/fs/xfs/quota/xfs_qm_bhv.c index f137856..db8872b 100644 --- a/fs/xfs/quota/xfs_qm_bhv.c +++ b/fs/xfs/quota/xfs_qm_bhv.c @@ -203,7 +203,7 @@ xfs_qm_statvfs( if (error || !vnode) return error; - mp = XFS_BHVTOM(bhv); + mp = xfs_vfstom(bhvtovfs(bhv)); ip = xfs_vtoi(vnode); if (!(ip->i_d.di_flags & XFS_DIFLAG_PROJINHERIT)) diff --git a/fs/xfs/xfs_alloc.h b/fs/xfs/xfs_alloc.h index 650591f..5a42561 100644 --- a/fs/xfs/xfs_alloc.h +++ b/fs/xfs/xfs_alloc.h @@ -44,6 +44,26 @@ typedef enum xfs_alloctype #define XFS_ALLOC_FLAG_FREEING 0x00000002 /* indicate caller is freeing extents*/ /* + * In order to avoid ENOSPC-related deadlock caused by + * out-of-order locking of AGF buffer (PV 947395), we place + * constraints on the relationship among actual allocations for + * data blocks, freelist blocks, and potential file data bmap + * btree blocks. However, these restrictions may result in no + * actual space allocated for a delayed extent, for example, a data + * block in a certain AG is allocated but there is no additional + * block for the additional bmap btree block due to a split of the + * bmap btree of the file. The result of this may lead to an + * infinite loop in xfssyncd when the file gets flushed to disk and + * all delayed extents need to be actually allocated. To get around + * this, we explicitly set aside a few blocks which will not be + * reserved in delayed allocation. Considering the minimum number of + * needed freelist blocks is 4 fsbs _per AG_, a potential split of file's bmap + * btree requires 1 fsb, so we set the number of set-aside blocks + * to 4 + 4*agcount. + */ +#define XFS_ALLOC_SET_ASIDE(mp) (4 + ((mp)->m_sb.sb_agcount * 4)) + +/* * Argument structure for xfs_alloc routines. * This is turned into a structure to avoid having 20 arguments passed * down several levels of the stack. diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c index 3a61375..bf46fae 100644 --- a/fs/xfs/xfs_bmap.c +++ b/fs/xfs/xfs_bmap.c @@ -4993,7 +4993,7 @@ xfs_bmapi( bma.firstblock = *firstblock; bma.alen = alen; bma.off = aoff; - bma.conv = (flags & XFS_BMAPI_CONVERT); + bma.conv = !!(flags & XFS_BMAPI_CONVERT); bma.wasdel = wasdelay; bma.minlen = minlen; bma.low = flist->xbf_low; diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c index 077629b..c064e72 100644 --- a/fs/xfs/xfs_fsops.c +++ b/fs/xfs/xfs_fsops.c @@ -462,7 +462,7 @@ xfs_fs_counts( xfs_icsb_sync_counters_lazy(mp); s = XFS_SB_LOCK(mp); - cnt->freedata = mp->m_sb.sb_fdblocks; + cnt->freedata = mp->m_sb.sb_fdblocks - XFS_ALLOC_SET_ASIDE(mp); cnt->freertx = mp->m_sb.sb_frextents; cnt->freeino = mp->m_sb.sb_ifree; cnt->allocino = mp->m_sb.sb_icount; @@ -519,15 +519,19 @@ xfs_reserve_blocks( } mp->m_resblks = request; } else { + __int64_t free; + + free = mp->m_sb.sb_fdblocks - XFS_ALLOC_SET_ASIDE(mp); delta = request - mp->m_resblks; - lcounter = mp->m_sb.sb_fdblocks - delta; + lcounter = free - delta; if (lcounter < 0) { /* We can't satisfy the request, just get what we can */ - mp->m_resblks += mp->m_sb.sb_fdblocks; - mp->m_resblks_avail += mp->m_sb.sb_fdblocks; - mp->m_sb.sb_fdblocks = 0; + mp->m_resblks += free; + mp->m_resblks_avail += free; + mp->m_sb.sb_fdblocks = XFS_ALLOC_SET_ASIDE(mp); } else { - mp->m_sb.sb_fdblocks = lcounter; + mp->m_sb.sb_fdblocks = + lcounter + XFS_ALLOC_SET_ASIDE(mp); mp->m_resblks = request; mp->m_resblks_avail += delta; } diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index 4be5c0b..9dfae18 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -1243,24 +1243,6 @@ xfs_mod_sb(xfs_trans_t *tp, __int64_t fields) xfs_trans_log_buf(tp, bp, first, last); } -/* - * In order to avoid ENOSPC-related deadlock caused by - * out-of-order locking of AGF buffer (PV 947395), we place - * constraints on the relationship among actual allocations for - * data blocks, freelist blocks, and potential file data bmap - * btree blocks. However, these restrictions may result in no - * actual space allocated for a delayed extent, for example, a data - * block in a certain AG is allocated but there is no additional - * block for the additional bmap btree block due to a split of the - * bmap btree of the file. The result of this may lead to an - * infinite loop in xfssyncd when the file gets flushed to disk and - * all delayed extents need to be actually allocated. To get around - * this, we explicitly set aside a few blocks which will not be - * reserved in delayed allocation. Considering the minimum number of - * needed freelist blocks is 4 fsbs, a potential split of file's bmap - * btree requires 1 fsb, so we set the number of set-aside blocks to 8. -*/ -#define SET_ASIDE_BLOCKS 8 /* * xfs_mod_incore_sb_unlocked() is a utility routine common used to apply @@ -1306,7 +1288,8 @@ xfs_mod_incore_sb_unlocked(xfs_mount_t *mp, xfs_sb_field_t field, return 0; case XFS_SBS_FDBLOCKS: - lcounter = (long long)mp->m_sb.sb_fdblocks - SET_ASIDE_BLOCKS; + lcounter = (long long) + mp->m_sb.sb_fdblocks - XFS_ALLOC_SET_ASIDE(mp); res_used = (long long)(mp->m_resblks - mp->m_resblks_avail); if (delta > 0) { /* Putting blocks back */ @@ -1340,7 +1323,7 @@ xfs_mod_incore_sb_unlocked(xfs_mount_t *mp, xfs_sb_field_t field, } } - mp->m_sb.sb_fdblocks = lcounter + SET_ASIDE_BLOCKS; + mp->m_sb.sb_fdblocks = lcounter + XFS_ALLOC_SET_ASIDE(mp); return 0; case XFS_SBS_FREXTENTS: lcounter = (long long)mp->m_sb.sb_frextents; @@ -2021,7 +2004,8 @@ xfs_icsb_sync_counters_lazy( * when we get near ENOSPC. */ #define XFS_ICSB_INO_CNTR_REENABLE 64 -#define XFS_ICSB_FDBLK_CNTR_REENABLE 512 +#define XFS_ICSB_FDBLK_CNTR_REENABLE(mp) \ + (512 + XFS_ALLOC_SET_ASIDE(mp)) STATIC void xfs_icsb_balance_counter( xfs_mount_t *mp, @@ -2055,7 +2039,7 @@ xfs_icsb_balance_counter( case XFS_SBS_FDBLOCKS: count = mp->m_sb.sb_fdblocks; resid = do_div(count, weight); - if (count < XFS_ICSB_FDBLK_CNTR_REENABLE) + if (count < XFS_ICSB_FDBLK_CNTR_REENABLE(mp)) goto out; break; default: @@ -2110,11 +2094,11 @@ again: case XFS_SBS_FDBLOCKS: BUG_ON((mp->m_resblks - mp->m_resblks_avail) != 0); - lcounter = icsbp->icsb_fdblocks; + lcounter = icsbp->icsb_fdblocks - XFS_ALLOC_SET_ASIDE(mp); lcounter += delta; if (unlikely(lcounter < 0)) goto slow_path; - icsbp->icsb_fdblocks = lcounter; + icsbp->icsb_fdblocks = lcounter + XFS_ALLOC_SET_ASIDE(mp); break; default: BUG(); diff --git a/fs/xfs/xfs_vfsops.c b/fs/xfs/xfs_vfsops.c index b427d22..a34796e 100644 --- a/fs/xfs/xfs_vfsops.c +++ b/fs/xfs/xfs_vfsops.c @@ -811,7 +811,8 @@ xfs_statvfs( statp->f_bsize = sbp->sb_blocksize; lsize = sbp->sb_logstart ? sbp->sb_logblocks : 0; statp->f_blocks = sbp->sb_dblocks - lsize; - statp->f_bfree = statp->f_bavail = sbp->sb_fdblocks; + statp->f_bfree = statp->f_bavail = + sbp->sb_fdblocks - XFS_ALLOC_SET_ASIDE(mp); fakeinos = statp->f_bfree << sbp->sb_inopblog; #if XFS_BIG_INUMS fakeinos += mp->m_inoadd; |