From 8aa7e847d834ed937a9ad37a0f2ad5b8584c1ab0 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Thu, 9 Jul 2009 14:52:32 +0200 Subject: Fix congestion_wait() sync/async vs read/write confusion Commit 1faa16d22877f4839bd433547d770c676d1d964c accidentally broke the bdi congestion wait queue logic, causing us to wait on congestion for WRITE (== 1) when we really wanted BLK_RW_ASYNC (== 0) instead. Signed-off-by: Jens Axboe --- fs/xfs/linux-2.6/kmem.c | 4 ++-- fs/xfs/linux-2.6/xfs_buf.c | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) (limited to 'fs/xfs') diff --git a/fs/xfs/linux-2.6/kmem.c b/fs/xfs/linux-2.6/kmem.c index 1cd3b55ee3d2..2d3f90afe5f1 100644 --- a/fs/xfs/linux-2.6/kmem.c +++ b/fs/xfs/linux-2.6/kmem.c @@ -53,7 +53,7 @@ kmem_alloc(size_t size, unsigned int __nocast flags) printk(KERN_ERR "XFS: possible memory allocation " "deadlock in %s (mode:0x%x)\n", __func__, lflags); - congestion_wait(WRITE, HZ/50); + congestion_wait(BLK_RW_ASYNC, HZ/50); } while (1); } @@ -130,7 +130,7 @@ kmem_zone_alloc(kmem_zone_t *zone, unsigned int __nocast flags) printk(KERN_ERR "XFS: possible memory allocation " "deadlock in %s (mode:0x%x)\n", __func__, lflags); - congestion_wait(WRITE, HZ/50); + congestion_wait(BLK_RW_ASYNC, HZ/50); } while (1); } diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c index 1418b916fc27..0c93c7ef3d18 100644 --- a/fs/xfs/linux-2.6/xfs_buf.c +++ b/fs/xfs/linux-2.6/xfs_buf.c @@ -412,7 +412,7 @@ _xfs_buf_lookup_pages( XFS_STATS_INC(xb_page_retries); xfsbufd_wakeup(0, gfp_mask); - congestion_wait(WRITE, HZ/50); + congestion_wait(BLK_RW_ASYNC, HZ/50); goto retry; } -- cgit v1.2.3 From 405f55712dfe464b3240d7816cc4fe4174831be2 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Sat, 11 Jul 2009 22:08:37 +0400 Subject: headers: smp_lock.h redux * Remove smp_lock.h from files which don't need it (including some headers!) * Add smp_lock.h to files which do need it * Make smp_lock.h include conditional in hardirq.h It's needed only for one kernel_locked() usage which is under CONFIG_PREEMPT This will make hardirq.h inclusion cheaper for every PREEMPT=n config (which includes allmodconfig/allyesconfig, BTW) Signed-off-by: Alexey Dobriyan Signed-off-by: Linus Torvalds --- fs/xfs/linux-2.6/xfs_file.c | 1 - 1 file changed, 1 deletion(-) (limited to 'fs/xfs') diff --git a/fs/xfs/linux-2.6/xfs_file.c b/fs/xfs/linux-2.6/xfs_file.c index f4e255441574..0542fd507649 100644 --- a/fs/xfs/linux-2.6/xfs_file.c +++ b/fs/xfs/linux-2.6/xfs_file.c @@ -41,7 +41,6 @@ #include "xfs_ioctl.h" #include -#include static struct vm_operations_struct xfs_file_vm_ops; -- cgit v1.2.3 From 97db39a1f6f69e906e98118392400de5217aa33a Mon Sep 17 00:00:00 2001 From: Eric Sandeen Date: Sun, 26 Jul 2009 21:52:01 -0500 Subject: xfs: reduce bmv_count in xfs_vn_fiemap commit 6321e3ed2acf3ee9643cdd403e1c88605d7944ba caused the full bmv_count's worth of getbmapx structures to get allocated; telling it to do MAXEXTNUM was a bit insane, resulting in ENOMEM every time. Chop it down to something reasonable, the number of slots in the caller's input buffer. If this is too large the caller may get ENOMEM but the reason should not be a mystery, and they can try again with something smaller. We add 1 to the value because in the normal getbmap world, bmv_count includes the header and xfs_getbmap does: nex = bmv->bmv_count - 1; if (nex <= 0) return XFS_ERROR(EINVAL); Signed-off-by: Eric Sandeen Reviewed-by: Olaf Weber Reviewed-by: Christoph Hellwig Signed-off-by: Felix Blyakher --- fs/xfs/linux-2.6/xfs_iops.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs/xfs') diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c index 58973bb46038..8070b34cc287 100644 --- a/fs/xfs/linux-2.6/xfs_iops.c +++ b/fs/xfs/linux-2.6/xfs_iops.c @@ -680,8 +680,8 @@ xfs_vn_fiemap( else bm.bmv_length = BTOBB(length); - /* our formatter will tell xfs_getbmap when to stop. */ - bm.bmv_count = MAXEXTNUM; + /* We add one because in getbmap world count includes the header */ + bm.bmv_count = fieinfo->fi_extents_max + 1; bm.bmv_iflags = BMV_IF_PREALLOC; if (fieinfo->fi_flags & FIEMAP_FLAG_XATTR) bm.bmv_iflags |= BMV_IF_ATTRFORK; -- cgit v1.2.3 From c8a4051c3731b6db224482218cfd535ab9393ff8 Mon Sep 17 00:00:00 2001 From: Eric Sandeen Date: Fri, 31 Jul 2009 00:02:17 -0500 Subject: xfs: bump up nr_to_write in xfs_vm_writepage VM calculation for nr_to_write seems off. Bump it way up, this gets simple streaming writes zippy again. To be reviewed again after Jens' writeback changes. Signed-off-by: Christoph Hellwig Signed-off-by: Eric Sandeen Cc: Chris Mason Reviewed-by: Felix Blyakher Signed-off-by: Felix Blyakher --- fs/xfs/linux-2.6/xfs_aops.c | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'fs/xfs') diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c index 7ec89fc05b2b..aecf2519db76 100644 --- a/fs/xfs/linux-2.6/xfs_aops.c +++ b/fs/xfs/linux-2.6/xfs_aops.c @@ -1268,6 +1268,14 @@ xfs_vm_writepage( if (!page_has_buffers(page)) create_empty_buffers(page, 1 << inode->i_blkbits, 0); + + /* + * VM calculation for nr_to_write seems off. Bump it way + * up, this gets simple streaming writes zippy again. + * To be reviewed again after Jens' writeback changes. + */ + wbc->nr_to_write *= 4; + /* * Convert delayed allocate, unwritten or unmapped space * to real space and flush out to disk. -- cgit v1.2.3 From 54e346215e4fe2ca8c94c54e546cc61902060510 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Fri, 7 Aug 2009 14:38:25 -0300 Subject: vfs: fix inode_init_always calling convention Currently inode_init_always calls into ->destroy_inode if the additional initialization fails. That's not only counter-intuitive because inode_init_always did not allocate the inode structure, but in case of XFS it's actively harmful as ->destroy_inode might delete the inode from a radix-tree that has never been added. This in turn might end up deleting the inode for the same inum that has been instanciated by another process and cause lots of cause subtile problems. Also in the case of re-initializing a reclaimable inode in XFS it would free an inode we still want to keep alive. Signed-off-by: Christoph Hellwig Reviewed-by: Eric Sandeen --- fs/xfs/xfs_iget.c | 17 +++++------------ 1 file changed, 5 insertions(+), 12 deletions(-) (limited to 'fs/xfs') diff --git a/fs/xfs/xfs_iget.c b/fs/xfs/xfs_iget.c index 5fcec6f020a7..719c85b155f4 100644 --- a/fs/xfs/xfs_iget.c +++ b/fs/xfs/xfs_iget.c @@ -64,6 +64,10 @@ xfs_inode_alloc( ip = kmem_zone_alloc(xfs_inode_zone, KM_SLEEP); if (!ip) return NULL; + if (inode_init_always(mp->m_super, VFS_I(ip))) { + kmem_zone_free(xfs_inode_zone, ip); + return NULL; + } ASSERT(atomic_read(&ip->i_iocount) == 0); ASSERT(atomic_read(&ip->i_pincount) == 0); @@ -105,17 +109,6 @@ xfs_inode_alloc( #ifdef XFS_DIR2_TRACE ip->i_dir_trace = ktrace_alloc(XFS_DIR2_KTRACE_SIZE, KM_NOFS); #endif - /* - * Now initialise the VFS inode. We do this after the xfs_inode - * initialisation as internal failures will result in ->destroy_inode - * being called and that will pass down through the reclaim path and - * free the XFS inode. This path requires the XFS inode to already be - * initialised. Hence if this call fails, the xfs_inode has already - * been freed and we should not reference it at all in the error - * handling. - */ - if (!inode_init_always(mp->m_super, VFS_I(ip))) - return NULL; /* prevent anyone from using this yet */ VFS_I(ip)->i_state = I_NEW|I_LOCK; @@ -167,7 +160,7 @@ xfs_iget_cache_hit( * errors cleanly, then tag it so it can be set up correctly * later. */ - if (!inode_init_always(mp->m_super, VFS_I(ip))) { + if (inode_init_always(mp->m_super, VFS_I(ip))) { error = ENOMEM; goto out_error; } -- cgit v1.2.3 From b36ec0428a06fcbdb67d61e9e664154e5dd9a8c7 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Fri, 7 Aug 2009 14:38:34 -0300 Subject: xfs: fix freeing of inodes not yet added to the inode cache When freeing an inode that lost race getting added to the inode cache we must not call into ->destroy_inode, because that would delete the inode that won the race from the inode cache radix tree. This patch uses splits a new xfs_inode_free helper out of xfs_ireclaim and uses that plus __destroy_inode to make sure we really only free the memory allocted for the inode that lost the race, and not mess with the inode cache state. Signed-off-by: Christoph Hellwig Reviewed-by: Eric Sandeen Reported-by: Alex Samad Reported-by: Andrew Randrianasulu Reported-by: Stephane Reported-by: Tommy Reported-by: Miah Gregory Reported-by: Gabriel Barazer Reported-by: Leandro Lucarella Reported-by: Daniel Burr Reported-by: Nickolay Reported-by: Michael Guntsche Reported-by: Dan Carley Reported-by: Michael Ole Olsen Reported-by: Michael Weissenbacher Reported-by: Martin Spott Reported-by: Christian Kujau Tested-by: Michael Guntsche Tested-by: Dan Carley Tested-by: Christian Kujau --- fs/xfs/xfs_iget.c | 125 +++++++++++++++++++++++++++++------------------------ fs/xfs/xfs_inode.h | 17 -------- 2 files changed, 68 insertions(+), 74 deletions(-) (limited to 'fs/xfs') diff --git a/fs/xfs/xfs_iget.c b/fs/xfs/xfs_iget.c index 719c85b155f4..34ec86923f7e 100644 --- a/fs/xfs/xfs_iget.c +++ b/fs/xfs/xfs_iget.c @@ -116,6 +116,71 @@ xfs_inode_alloc( return ip; } +STATIC void +xfs_inode_free( + struct xfs_inode *ip) +{ + switch (ip->i_d.di_mode & S_IFMT) { + case S_IFREG: + case S_IFDIR: + case S_IFLNK: + xfs_idestroy_fork(ip, XFS_DATA_FORK); + break; + } + + if (ip->i_afp) + xfs_idestroy_fork(ip, XFS_ATTR_FORK); + +#ifdef XFS_INODE_TRACE + ktrace_free(ip->i_trace); +#endif +#ifdef XFS_BMAP_TRACE + ktrace_free(ip->i_xtrace); +#endif +#ifdef XFS_BTREE_TRACE + ktrace_free(ip->i_btrace); +#endif +#ifdef XFS_RW_TRACE + ktrace_free(ip->i_rwtrace); +#endif +#ifdef XFS_ILOCK_TRACE + ktrace_free(ip->i_lock_trace); +#endif +#ifdef XFS_DIR2_TRACE + ktrace_free(ip->i_dir_trace); +#endif + + if (ip->i_itemp) { + /* + * Only if we are shutting down the fs will we see an + * inode still in the AIL. If it is there, we should remove + * it to prevent a use-after-free from occurring. + */ + xfs_log_item_t *lip = &ip->i_itemp->ili_item; + struct xfs_ail *ailp = lip->li_ailp; + + ASSERT(((lip->li_flags & XFS_LI_IN_AIL) == 0) || + XFS_FORCED_SHUTDOWN(ip->i_mount)); + if (lip->li_flags & XFS_LI_IN_AIL) { + spin_lock(&ailp->xa_lock); + if (lip->li_flags & XFS_LI_IN_AIL) + xfs_trans_ail_delete(ailp, lip); + else + spin_unlock(&ailp->xa_lock); + } + xfs_inode_item_destroy(ip); + ip->i_itemp = NULL; + } + + /* asserts to verify all state is correct here */ + ASSERT(atomic_read(&ip->i_iocount) == 0); + ASSERT(atomic_read(&ip->i_pincount) == 0); + ASSERT(!spin_is_locked(&ip->i_flags_lock)); + ASSERT(completion_done(&ip->i_flush)); + + kmem_zone_free(xfs_inode_zone, ip); +} + /* * Check the validity of the inode we just found it the cache */ @@ -292,7 +357,8 @@ out_preload_end: if (lock_flags) xfs_iunlock(ip, lock_flags); out_destroy: - xfs_destroy_inode(ip); + __destroy_inode(VFS_I(ip)); + xfs_inode_free(ip); return error; } @@ -497,62 +563,7 @@ xfs_ireclaim( xfs_qm_dqdetach(ip); xfs_iunlock(ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL); - switch (ip->i_d.di_mode & S_IFMT) { - case S_IFREG: - case S_IFDIR: - case S_IFLNK: - xfs_idestroy_fork(ip, XFS_DATA_FORK); - break; - } - - if (ip->i_afp) - xfs_idestroy_fork(ip, XFS_ATTR_FORK); - -#ifdef XFS_INODE_TRACE - ktrace_free(ip->i_trace); -#endif -#ifdef XFS_BMAP_TRACE - ktrace_free(ip->i_xtrace); -#endif -#ifdef XFS_BTREE_TRACE - ktrace_free(ip->i_btrace); -#endif -#ifdef XFS_RW_TRACE - ktrace_free(ip->i_rwtrace); -#endif -#ifdef XFS_ILOCK_TRACE - ktrace_free(ip->i_lock_trace); -#endif -#ifdef XFS_DIR2_TRACE - ktrace_free(ip->i_dir_trace); -#endif - if (ip->i_itemp) { - /* - * Only if we are shutting down the fs will we see an - * inode still in the AIL. If it is there, we should remove - * it to prevent a use-after-free from occurring. - */ - xfs_log_item_t *lip = &ip->i_itemp->ili_item; - struct xfs_ail *ailp = lip->li_ailp; - - ASSERT(((lip->li_flags & XFS_LI_IN_AIL) == 0) || - XFS_FORCED_SHUTDOWN(ip->i_mount)); - if (lip->li_flags & XFS_LI_IN_AIL) { - spin_lock(&ailp->xa_lock); - if (lip->li_flags & XFS_LI_IN_AIL) - xfs_trans_ail_delete(ailp, lip); - else - spin_unlock(&ailp->xa_lock); - } - xfs_inode_item_destroy(ip); - ip->i_itemp = NULL; - } - /* asserts to verify all state is correct here */ - ASSERT(atomic_read(&ip->i_iocount) == 0); - ASSERT(atomic_read(&ip->i_pincount) == 0); - ASSERT(!spin_is_locked(&ip->i_flags_lock)); - ASSERT(completion_done(&ip->i_flush)); - kmem_zone_free(xfs_inode_zone, ip); + xfs_inode_free(ip); } /* diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h index 1804f866a71d..65f24a3cc992 100644 --- a/fs/xfs/xfs_inode.h +++ b/fs/xfs/xfs_inode.h @@ -309,23 +309,6 @@ static inline struct inode *VFS_I(struct xfs_inode *ip) return &ip->i_vnode; } -/* - * Get rid of a partially initialized inode. - * - * We have to go through destroy_inode to make sure allocations - * from init_inode_always like the security data are undone. - * - * We mark the inode bad so that it takes the short cut in - * the reclaim path instead of going through the flush path - * which doesn't make sense for an inode that has never seen the - * light of day. - */ -static inline void xfs_destroy_inode(struct xfs_inode *ip) -{ - make_bad_inode(VFS_I(ip)); - return destroy_inode(VFS_I(ip)); -} - /* * i_flags helper functions */ -- cgit v1.2.3 From 0cc6eee130b0c062feec8446d9cecdb17d2cfad3 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Sat, 18 Jul 2009 18:14:53 -0400 Subject: xfs: avoid memory allocation under m_peraglock in growfs code Allocate the memory for the larger m_perag array before taking the per-AG lock as the per-AG lock can be taken under the i_lock which can be taken from reclaim context. Reported by the new reclaim context tracing in lockdep. Signed-off-by: Christoph Hellwig Reviewed-by: Felix Blyakher Signed-off-by: Felix Blyakher --- fs/xfs/xfs_fsops.c | 20 ++++++++++++++------ 1 file changed, 14 insertions(+), 6 deletions(-) (limited to 'fs/xfs') diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c index cbd451bb4848..2d0b3e1da9e6 100644 --- a/fs/xfs/xfs_fsops.c +++ b/fs/xfs/xfs_fsops.c @@ -167,17 +167,25 @@ xfs_growfs_data_private( new = nb - mp->m_sb.sb_dblocks; oagcount = mp->m_sb.sb_agcount; if (nagcount > oagcount) { + void *new_perag, *old_perag; + xfs_filestream_flush(mp); + + new_perag = kmem_zalloc(sizeof(xfs_perag_t) * nagcount, + KM_MAYFAIL); + if (!new_perag) + return XFS_ERROR(ENOMEM); + down_write(&mp->m_peraglock); - mp->m_perag = kmem_realloc(mp->m_perag, - sizeof(xfs_perag_t) * nagcount, - sizeof(xfs_perag_t) * oagcount, - KM_SLEEP); - memset(&mp->m_perag[oagcount], 0, - (nagcount - oagcount) * sizeof(xfs_perag_t)); + memcpy(new_perag, mp->m_perag, sizeof(xfs_perag_t) * oagcount); + old_perag = mp->m_perag; + mp->m_perag = new_perag; + mp->m_flags |= XFS_MOUNT_32BITINODES; nagimax = xfs_initialize_perag(mp, nagcount); up_write(&mp->m_peraglock); + + kmem_free(old_perag); } tp = xfs_trans_alloc(mp, XFS_TRANS_GROWFS); tp->t_flags |= XFS_TRANS_RESERVE; -- cgit v1.2.3 From ca35dcd6cae7d4a780c484c53f45548c4719f82c Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Sat, 18 Jul 2009 18:14:54 -0400 Subject: xfs: switch to NOFS allocation under i_lock in xfs_getbmap xfs_getbmap allocates memory with i_lock held, but i_lock is taken in reclaim context so all allocations under it must avoid recursions into the filesystem. Reported by the new reclaim context tracing in lockdep. Signed-off-by: Christoph Hellwig Reviewed-by: Felix Blyakher Signed-off-by: Felix Blyakher --- fs/xfs/xfs_bmap.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs/xfs') diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c index 7928b9983c1d..8ee5b5a76a2a 100644 --- a/fs/xfs/xfs_bmap.c +++ b/fs/xfs/xfs_bmap.c @@ -6009,7 +6009,7 @@ xfs_getbmap( */ error = ENOMEM; subnex = 16; - map = kmem_alloc(subnex * sizeof(*map), KM_MAYFAIL); + map = kmem_alloc(subnex * sizeof(*map), KM_MAYFAIL | KM_NOFS); if (!map) goto out_unlock_ilock; -- cgit v1.2.3 From f41d7fb9da05b604f8a69fb6cac2a0563c8ede4e Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Sat, 18 Jul 2009 18:14:55 -0400 Subject: xfs: switch to NOFS allocation under i_lock in xfs_da_state_alloc xfs_da_state_alloc is always called with i_lock held, but i_lock is taken in reclaim context so all allocations under it must avoid recursions into the filesystem. Reported by the new reclaim context tracing in lockdep. Signed-off-by: Christoph Hellwig Reviewed-by: Felix Blyakher Signed-off-by: Felix Blyakher --- fs/xfs/xfs_da_btree.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs/xfs') diff --git a/fs/xfs/xfs_da_btree.c b/fs/xfs/xfs_da_btree.c index 9ff6e57a5075..bd0bb6dfcdfa 100644 --- a/fs/xfs/xfs_da_btree.c +++ b/fs/xfs/xfs_da_btree.c @@ -2201,7 +2201,7 @@ kmem_zone_t *xfs_dabuf_zone; /* dabuf zone */ xfs_da_state_t * xfs_da_state_alloc(void) { - return kmem_zone_zalloc(xfs_da_state_zone, KM_SLEEP); + return kmem_zone_zalloc(xfs_da_state_zone, KM_NOFS); } /* -- cgit v1.2.3 From 73195ed7864ae4a1fb0bea2ed9df59d19b4fde90 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Sat, 18 Jul 2009 18:14:56 -0400 Subject: xfs: switch to NOFS allocation under i_lock in xfs_da_buf_make i_lock is taken in the reclaim context so all allocations under it must avoid recursions into the filesystem. Reported by the new reclaim context tracing in lockdep. Signed-off-by: Christoph Hellwig Reviewed-by: Felix Blyakher Signed-off-by: Felix Blyakher --- fs/xfs/xfs_da_btree.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs/xfs') diff --git a/fs/xfs/xfs_da_btree.c b/fs/xfs/xfs_da_btree.c index bd0bb6dfcdfa..2847bbc1c534 100644 --- a/fs/xfs/xfs_da_btree.c +++ b/fs/xfs/xfs_da_btree.c @@ -2261,9 +2261,9 @@ xfs_da_buf_make(int nbuf, xfs_buf_t **bps, inst_t *ra) int off; if (nbuf == 1) - dabuf = kmem_zone_alloc(xfs_dabuf_zone, KM_SLEEP); + dabuf = kmem_zone_alloc(xfs_dabuf_zone, KM_NOFS); else - dabuf = kmem_alloc(XFS_DA_BUF_SIZE(nbuf), KM_SLEEP); + dabuf = kmem_alloc(XFS_DA_BUF_SIZE(nbuf), KM_NOFS); dabuf->dirty = 0; #ifdef XFS_DABUF_DEBUG dabuf->ra = ra; -- cgit v1.2.3 From 3f52c2f0a07c23771909cc53f2e9451a7f1bf253 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Sat, 18 Jul 2009 18:14:57 -0400 Subject: xfs: switch to NOFS allocation under i_lock in xfs_dir_cilookup_result xfs_dir_cilookup_result is always called with i_lock held, but i_lock is taken in reclaim context so all allocations under it must avoid recursions into the filesystem. Reported by the new reclaim context tracing in lockdep. Signed-off-by: Christoph Hellwig Reviewed-by: Felix Blyakher Signed-off-by: Felix Blyakher --- fs/xfs/xfs_dir2.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs/xfs') diff --git a/fs/xfs/xfs_dir2.c b/fs/xfs/xfs_dir2.c index c657bec6d951..bb1d58eb3982 100644 --- a/fs/xfs/xfs_dir2.c +++ b/fs/xfs/xfs_dir2.c @@ -256,7 +256,7 @@ xfs_dir_cilookup_result( !(args->op_flags & XFS_DA_OP_CILOOKUP)) return EEXIST; - args->value = kmem_alloc(len, KM_MAYFAIL); + args->value = kmem_alloc(len, KM_NOFS | KM_MAYFAIL); if (!args->value) return ENOMEM; -- cgit v1.2.3 From 36fae17a648e0aee5d9560514d08477ef48dc87f Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Sat, 18 Jul 2009 18:14:58 -0400 Subject: xfs: switch to NOFS allocation under i_lock in xfs_buf_associate_memory xfs_buf_associate_memory is used for setting up the spare buffer for the log wrap case in xlog_sync which can happen under i_lock when called from xfs_fsync. The i_lock mutex is taken in reclaim context so all allocations under it must avoid recursions into the filesystem. There are a couple more uses of xfs_buf_associate_memory in the log recovery code that are also affected by this, but I'd rather keep the code simple than passing on a gfp_mask argument. Longer term we should just stop requiring the memoery allocation in xlog_sync by some smaller rework of the buffer layer. Reported by the new reclaim context tracing in lockdep. Signed-off-by: Christoph Hellwig Reviewed-by: Felix Blyakher Signed-off-by: Felix Blyakher --- fs/xfs/linux-2.6/xfs_buf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs/xfs') diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c index 1418b916fc27..178c20c13e83 100644 --- a/fs/xfs/linux-2.6/xfs_buf.c +++ b/fs/xfs/linux-2.6/xfs_buf.c @@ -770,7 +770,7 @@ xfs_buf_associate_memory( bp->b_pages = NULL; bp->b_addr = mem; - rval = _xfs_buf_get_pages(bp, page_count, 0); + rval = _xfs_buf_get_pages(bp, page_count, XBF_DONT_BLOCK); if (rval) return rval; -- cgit v1.2.3 From 10746e47e722b5688fcd6eba9fbf9b2e64a248a7 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Sat, 18 Jul 2009 18:14:59 -0400 Subject: xfs: switch to NOFS allocation under i_lock in xfs_attr_rmtval_set xfs_attr_rmtval_set is always called with i_lock held, and i_lock is taken in reclaim context so all allocations under it must avoid recursions into the filesystem. Reported by the new reclaim context tracing in lockdep. Signed-off-by: Christoph Hellwig Reviewed-by: Felix Blyakher Signed-off-by: Felix Blyakher --- fs/xfs/xfs_attr.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs/xfs') diff --git a/fs/xfs/xfs_attr.c b/fs/xfs/xfs_attr.c index db15feb906ff..bfb583791fe2 100644 --- a/fs/xfs/xfs_attr.c +++ b/fs/xfs/xfs_attr.c @@ -2141,8 +2141,8 @@ xfs_attr_rmtval_set(xfs_da_args_t *args) dblkno = XFS_FSB_TO_DADDR(mp, map.br_startblock), blkcnt = XFS_FSB_TO_BB(mp, map.br_blockcount); - bp = xfs_buf_get_flags(mp->m_ddev_targp, dblkno, - blkcnt, XFS_BUF_LOCK); + bp = xfs_buf_get_flags(mp->m_ddev_targp, dblkno, blkcnt, + XFS_BUF_LOCK | XBF_DONT_BLOCK); ASSERT(bp); ASSERT(!XFS_BUF_GETERROR(bp)); -- cgit v1.2.3 From 7b02ecb3031b192823bc732ae717febc0a59aa92 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Sat, 18 Jul 2009 18:15:00 -0400 Subject: xfs: switch to NOFS allocation under i_lock in xfs_readlink_bmap xfs_readlink_bmap is called with i_lock held, but i_lock is taken in reclaim context so all allocations under it must avoid recursions into the filesystem. Reported by the new reclaim context tracing in lockdep. Signed-off-by: Christoph Hellwig Reviewed-by: Felix Blyakher Signed-off-by: Felix Blyakher --- fs/xfs/xfs_vnodeops.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'fs/xfs') diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c index c4eca5ed5dab..492d75bae2bf 100644 --- a/fs/xfs/xfs_vnodeops.c +++ b/fs/xfs/xfs_vnodeops.c @@ -538,7 +538,9 @@ xfs_readlink_bmap( d = XFS_FSB_TO_DADDR(mp, mval[n].br_startblock); byte_cnt = XFS_FSB_TO_B(mp, mval[n].br_blockcount); - bp = xfs_buf_read(mp->m_ddev_targp, d, BTOBB(byte_cnt), 0); + bp = xfs_buf_read_flags(mp->m_ddev_targp, d, BTOBB(byte_cnt), + XBF_LOCK | XBF_MAPPED | + XBF_DONT_BLOCK); error = XFS_BUF_GETERROR(bp); if (error) { xfs_ioerror_alert("xfs_readlink", -- cgit v1.2.3 From ddd3a14e0f030f0f7b900621f67532285b8657ef Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Sat, 18 Jul 2009 18:15:01 -0400 Subject: xfs: switch to NOFS allocation under i_lock in xfs_attr_rmtval_get xfs_attr_rmtval_get is always called with i_lock held, but i_lock is taken in reclaim context so all allocations under it must avoid recursions into the filesystem. Reported by the new reclaim context tracing in lockdep. Signed-off-by: Christoph Hellwig Reviewed-by: Felix Blyakher Signed-off-by: Felix Blyakher --- fs/xfs/xfs_attr.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'fs/xfs') diff --git a/fs/xfs/xfs_attr.c b/fs/xfs/xfs_attr.c index bfb583791fe2..4ece1906bd41 100644 --- a/fs/xfs/xfs_attr.c +++ b/fs/xfs/xfs_attr.c @@ -2010,7 +2010,9 @@ xfs_attr_rmtval_get(xfs_da_args_t *args) dblkno = XFS_FSB_TO_DADDR(mp, map[i].br_startblock); blkcnt = XFS_FSB_TO_BB(mp, map[i].br_blockcount); error = xfs_read_buf(mp, mp->m_ddev_targp, dblkno, - blkcnt, XFS_BUF_LOCK, &bp); + blkcnt, + XFS_BUF_LOCK | XBF_DONT_BLOCK, + &bp); if (error) return(error); -- cgit v1.2.3 From e0c222c411e22f086e929cd69fdcc89336164ec1 Mon Sep 17 00:00:00 2001 From: Eric Sandeen Date: Mon, 20 Jul 2009 10:52:15 -0500 Subject: use XFS_CORRUPTION_ERROR in xfs_btree_check_sblock In Red Hat Bug 512552 - Can't write to XFS mount during raid5 resync a user ran into corruption while resyncing a raid, and we failed a consistency test, but didn't get much more info; it'd be nice to call XFS_CORRUPTION_ERROR here so we can see the buffer contents. Signed-off-by: Eric Sandeen Reviewed-by: Christoph Hellwig Signed-off-by: Felix Blyakher --- fs/xfs/xfs_btree.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs/xfs') diff --git a/fs/xfs/xfs_btree.c b/fs/xfs/xfs_btree.c index e9df99574829..26717388acf5 100644 --- a/fs/xfs/xfs_btree.c +++ b/fs/xfs/xfs_btree.c @@ -120,8 +120,8 @@ xfs_btree_check_sblock( XFS_RANDOM_BTREE_CHECK_SBLOCK))) { if (bp) xfs_buftrace("SBTREE ERROR", bp); - XFS_ERROR_REPORT("xfs_btree_check_sblock", XFS_ERRLEVEL_LOW, - cur->bc_mp); + XFS_CORRUPTION_ERROR("xfs_btree_check_sblock", + XFS_ERRLEVEL_LOW, cur->bc_mp, block); return XFS_ERROR(EFSCORRUPTED); } return 0; -- cgit v1.2.3 From b89d4208de3de442c9025919c4261be0b38e79a4 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 10 Aug 2009 11:32:18 -0300 Subject: xfs: check for dinode realtime flag corruption Ramon tested XFS with a modified version of fsfuzzer and hit a NULL pointer dereference in __xfs_get_blocks due to the RT device target pointer being NULL. To fix this reject inode with the realtime bit set on a a filesystem without an RT subvolume during inode read. Signed-off-by: Christoph Hellwig Reviewed-by: Eric Sandeen Reviewed-by: Felix Blyakher Reported-by: Ramon de Carvalho Valle Tested-by: Ramon de Carvalho Valle Signed-off-by: Felix Blyakher --- fs/xfs/xfs_inode.c | 10 ++++++++++ 1 file changed, 10 insertions(+) (limited to 'fs/xfs') diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index 1f22d65fed0a..da428b3fe0f5 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -343,6 +343,16 @@ xfs_iformat( return XFS_ERROR(EFSCORRUPTED); } + if (unlikely((ip->i_d.di_flags & XFS_DIFLAG_REALTIME) && + !ip->i_mount->m_rtdev_targp)) { + xfs_fs_repair_cmn_err(CE_WARN, ip->i_mount, + "corrupt dinode %Lu, has realtime flag set.", + ip->i_ino); + XFS_CORRUPTION_ERROR("xfs_iformat(realtime)", + XFS_ERRLEVEL_LOW, ip->i_mount, dip); + return XFS_ERROR(EFSCORRUPTED); + } + switch (ip->i_d.di_mode & S_IFMT) { case S_IFIFO: case S_IFCHR: -- cgit v1.2.3 From a8914f3a6d72c97328597a556a99daaf5cc288ae Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 10 Aug 2009 11:32:44 -0300 Subject: xfs: fix spin_is_locked assert on uni-processor builds Without SMP or preemption spin_is_locked always returns false, so we can't do an assert with it. Instead use assert_spin_locked, which does the right thing on all builds. Signed-off-by: Christoph Hellwig Reviewed-by: Eric Sandeen Reported-by: Johannes Engel Tested-by: Johannes Engel Signed-off-by: Felix Blyakher --- fs/xfs/xfs_log.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs/xfs') diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c index 3750f04ede0b..9dbdff3ea484 100644 --- a/fs/xfs/xfs_log.c +++ b/fs/xfs/xfs_log.c @@ -3180,7 +3180,7 @@ try_again: STATIC void xlog_state_want_sync(xlog_t *log, xlog_in_core_t *iclog) { - ASSERT(spin_is_locked(&log->l_icloglock)); + assert_spin_locked(&log->l_icloglock); if (iclog->ic_state == XLOG_STATE_ACTIVE) { xlog_state_switch_iclogs(log, iclog, 0); -- cgit v1.2.3 From bc990f5cb424cdca9dda866785d088e2c2110ecc Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Sun, 16 Aug 2009 20:36:34 -0400 Subject: xfs: fix locking in xfs_iget_cache_hit The locking in xfs_iget_cache_hit currently has numerous problems: - we clear the reclaim tag without i_flags_lock which protects modifications to it - we call inode_init_always which can sleep with pag_ici_lock held (this is oss.sgi.com BZ #819) - we acquire and drop i_flags_lock a lot and thus provide no consistency between the various flags we set/clear under it This patch fixes all that with a major revamp of the locking in the function. The new version acquires i_flags_lock early and only drops it once we need to call into inode_init_always or before calling xfs_ilock. This patch fixes a bug seen in the wild where we race modifying the reclaim tag. Signed-off-by: Christoph Hellwig Reviewed-by: Felix Blyakher Reviewed-by: Eric Sandeen Signed-off-by: Felix Blyakher --- fs/xfs/linux-2.6/xfs_sync.c | 13 ++++- fs/xfs/linux-2.6/xfs_sync.h | 1 + fs/xfs/xfs_iget.c | 113 +++++++++++++++++++++++--------------------- 3 files changed, 70 insertions(+), 57 deletions(-) (limited to 'fs/xfs') diff --git a/fs/xfs/linux-2.6/xfs_sync.c b/fs/xfs/linux-2.6/xfs_sync.c index b619d6b8ca43..98ef624d9baf 100644 --- a/fs/xfs/linux-2.6/xfs_sync.c +++ b/fs/xfs/linux-2.6/xfs_sync.c @@ -708,6 +708,16 @@ xfs_reclaim_inode( return 0; } +void +__xfs_inode_set_reclaim_tag( + struct xfs_perag *pag, + struct xfs_inode *ip) +{ + radix_tree_tag_set(&pag->pag_ici_root, + XFS_INO_TO_AGINO(ip->i_mount, ip->i_ino), + XFS_ICI_RECLAIM_TAG); +} + /* * We set the inode flag atomically with the radix tree tag. * Once we get tag lookups on the radix tree, this inode flag @@ -722,8 +732,7 @@ xfs_inode_set_reclaim_tag( read_lock(&pag->pag_ici_lock); spin_lock(&ip->i_flags_lock); - radix_tree_tag_set(&pag->pag_ici_root, - XFS_INO_TO_AGINO(mp, ip->i_ino), XFS_ICI_RECLAIM_TAG); + __xfs_inode_set_reclaim_tag(pag, ip); __xfs_iflags_set(ip, XFS_IRECLAIMABLE); spin_unlock(&ip->i_flags_lock); read_unlock(&pag->pag_ici_lock); diff --git a/fs/xfs/linux-2.6/xfs_sync.h b/fs/xfs/linux-2.6/xfs_sync.h index 2a10301c99c7..59120602588a 100644 --- a/fs/xfs/linux-2.6/xfs_sync.h +++ b/fs/xfs/linux-2.6/xfs_sync.h @@ -48,6 +48,7 @@ int xfs_reclaim_inode(struct xfs_inode *ip, int locked, int sync_mode); int xfs_reclaim_inodes(struct xfs_mount *mp, int mode); void xfs_inode_set_reclaim_tag(struct xfs_inode *ip); +void __xfs_inode_set_reclaim_tag(struct xfs_perag *pag, struct xfs_inode *ip); void xfs_inode_clear_reclaim_tag(struct xfs_inode *ip); void __xfs_inode_clear_reclaim_tag(struct xfs_mount *mp, struct xfs_perag *pag, struct xfs_inode *ip); diff --git a/fs/xfs/xfs_iget.c b/fs/xfs/xfs_iget.c index 34ec86923f7e..ecbf8b4d2e2e 100644 --- a/fs/xfs/xfs_iget.c +++ b/fs/xfs/xfs_iget.c @@ -191,80 +191,82 @@ xfs_iget_cache_hit( int flags, int lock_flags) __releases(pag->pag_ici_lock) { + struct inode *inode = VFS_I(ip); struct xfs_mount *mp = ip->i_mount; - int error = EAGAIN; + int error; + + spin_lock(&ip->i_flags_lock); /* - * If INEW is set this inode is being set up - * If IRECLAIM is set this inode is being torn down - * Pause and try again. + * If we are racing with another cache hit that is currently + * instantiating this inode or currently recycling it out of + * reclaimabe state, wait for the initialisation to complete + * before continuing. + * + * XXX(hch): eventually we should do something equivalent to + * wait_on_inode to wait for these flags to be cleared + * instead of polling for it. */ - if (xfs_iflags_test(ip, (XFS_INEW|XFS_IRECLAIM))) { + if (ip->i_flags & (XFS_INEW|XFS_IRECLAIM)) { XFS_STATS_INC(xs_ig_frecycle); + error = EAGAIN; goto out_error; } - /* If IRECLAIMABLE is set, we've torn down the vfs inode part */ - if (xfs_iflags_test(ip, XFS_IRECLAIMABLE)) { - - /* - * If lookup is racing with unlink, then we should return an - * error immediately so we don't remove it from the reclaim - * list and potentially leak the inode. - */ - if ((ip->i_d.di_mode == 0) && !(flags & XFS_IGET_CREATE)) { - error = ENOENT; - goto out_error; - } + /* + * If lookup is racing with unlink return an error immediately. + */ + if (ip->i_d.di_mode == 0 && !(flags & XFS_IGET_CREATE)) { + error = ENOENT; + goto out_error; + } + /* + * If IRECLAIMABLE is set, we've torn down the VFS inode already. + * Need to carefully get it back into useable state. + */ + if (ip->i_flags & XFS_IRECLAIMABLE) { xfs_itrace_exit_tag(ip, "xfs_iget.alloc"); /* - * We need to re-initialise the VFS inode as it has been - * 'freed' by the VFS. Do this here so we can deal with - * errors cleanly, then tag it so it can be set up correctly - * later. + * We need to set XFS_INEW atomically with clearing the + * reclaimable tag so that we do have an indicator of the + * inode still being initialized. */ - if (inode_init_always(mp->m_super, VFS_I(ip))) { - error = ENOMEM; - goto out_error; - } + ip->i_flags |= XFS_INEW; + ip->i_flags &= ~XFS_IRECLAIMABLE; + __xfs_inode_clear_reclaim_tag(mp, pag, ip); - /* - * We must set the XFS_INEW flag before clearing the - * XFS_IRECLAIMABLE flag so that if a racing lookup does - * not find the XFS_IRECLAIMABLE above but has the igrab() - * below succeed we can safely check XFS_INEW to detect - * that this inode is still being initialised. - */ - xfs_iflags_set(ip, XFS_INEW); - xfs_iflags_clear(ip, XFS_IRECLAIMABLE); + spin_unlock(&ip->i_flags_lock); + read_unlock(&pag->pag_ici_lock); - /* clear the radix tree reclaim flag as well. */ - __xfs_inode_clear_reclaim_tag(mp, pag, ip); - } else if (!igrab(VFS_I(ip))) { + error = -inode_init_always(mp->m_super, inode); + if (error) { + /* + * Re-initializing the inode failed, and we are in deep + * trouble. Try to re-add it to the reclaim list. + */ + read_lock(&pag->pag_ici_lock); + spin_lock(&ip->i_flags_lock); + + ip->i_flags &= ~XFS_INEW; + ip->i_flags |= XFS_IRECLAIMABLE; + __xfs_inode_set_reclaim_tag(pag, ip); + goto out_error; + } + inode->i_state = I_LOCK|I_NEW; + } else { /* If the VFS inode is being torn down, pause and try again. */ - XFS_STATS_INC(xs_ig_frecycle); - goto out_error; - } else if (xfs_iflags_test(ip, XFS_INEW)) { - /* - * We are racing with another cache hit that is - * currently recycling this inode out of the XFS_IRECLAIMABLE - * state. Wait for the initialisation to complete before - * continuing. - */ - wait_on_inode(VFS_I(ip)); - } + if (!igrab(inode)) { + error = EAGAIN; + goto out_error; + } - if (ip->i_d.di_mode == 0 && !(flags & XFS_IGET_CREATE)) { - error = ENOENT; - iput(VFS_I(ip)); - goto out_error; + /* We've got a live one. */ + spin_unlock(&ip->i_flags_lock); + read_unlock(&pag->pag_ici_lock); } - /* We've got a live one. */ - read_unlock(&pag->pag_ici_lock); - if (lock_flags != 0) xfs_ilock(ip, lock_flags); @@ -274,6 +276,7 @@ xfs_iget_cache_hit( return 0; out_error: + spin_unlock(&ip->i_flags_lock); read_unlock(&pag->pag_ici_lock); return error; } -- cgit v1.2.3