From 370f048214b4e9aa2102fa3c454ae1374da287c5 Mon Sep 17 00:00:00 2001 From: Eric Sandeen Date: Thu, 2 Jul 2009 00:09:33 -0500 Subject: [PATCH 01/19] xfs: add more statics & drop some unused functions A lot more functions could be made static, but they need forward declarations; this does some easy ones, and also found a few unused functions in the process. Signed-off-by: Eric Sandeen Reviewed-by: Christoph Hellwig Signed-off-by: Felix Blyakher --- fs/xfs/linux-2.6/xfs_super.c | 2 +- fs/xfs/linux-2.6/xfs_sync.c | 15 ------------- fs/xfs/linux-2.6/xfs_sync.h | 1 - fs/xfs/xfs_ag.h | 3 --- fs/xfs/xfs_alloc.c | 2 +- fs/xfs/xfs_bmap.c | 2 +- fs/xfs/xfs_bmap.h | 11 ---------- fs/xfs/xfs_bmap_btree.c | 20 ++++++++--------- fs/xfs/xfs_bmap_btree.h | 1 - fs/xfs/xfs_btree.c | 42 +----------------------------------- fs/xfs/xfs_btree.h | 15 ------------- fs/xfs/xfs_inode.c | 8 +++---- fs/xfs/xfs_inode.h | 5 ----- fs/xfs/xfs_itable.c | 2 +- fs/xfs/xfs_itable.h | 5 ----- fs/xfs/xfs_log_priv.h | 2 -- fs/xfs/xfs_log_recover.c | 2 +- fs/xfs/xfs_mount.c | 2 +- fs/xfs/xfs_mount.h | 3 --- fs/xfs/xfs_mru_cache.c | 29 ------------------------- fs/xfs/xfs_mru_cache.h | 1 - fs/xfs/xfs_rw.h | 6 ------ fs/xfs/xfs_vnodeops.c | 2 +- 23 files changed, 22 insertions(+), 159 deletions(-) diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c index 36fb8a657c5..de593787232 100644 --- a/fs/xfs/linux-2.6/xfs_super.c +++ b/fs/xfs/linux-2.6/xfs_super.c @@ -687,7 +687,7 @@ xfs_barrier_test( return error; } -void +STATIC void xfs_mountfs_check_barriers(xfs_mount_t *mp) { int error; diff --git a/fs/xfs/linux-2.6/xfs_sync.c b/fs/xfs/linux-2.6/xfs_sync.c index b619d6b8ca4..fbf3e0288b3 100644 --- a/fs/xfs/linux-2.6/xfs_sync.c +++ b/fs/xfs/linux-2.6/xfs_sync.c @@ -740,21 +740,6 @@ __xfs_inode_clear_reclaim_tag( XFS_INO_TO_AGINO(mp, ip->i_ino), XFS_ICI_RECLAIM_TAG); } -void -xfs_inode_clear_reclaim_tag( - xfs_inode_t *ip) -{ - xfs_mount_t *mp = ip->i_mount; - xfs_perag_t *pag = xfs_get_perag(mp, ip->i_ino); - - read_lock(&pag->pag_ici_lock); - spin_lock(&ip->i_flags_lock); - __xfs_inode_clear_reclaim_tag(mp, pag, ip); - spin_unlock(&ip->i_flags_lock); - read_unlock(&pag->pag_ici_lock); - xfs_put_perag(mp, pag); -} - STATIC int xfs_reclaim_inode_now( struct xfs_inode *ip, diff --git a/fs/xfs/linux-2.6/xfs_sync.h b/fs/xfs/linux-2.6/xfs_sync.h index 2a10301c99c..23e7e7e6e13 100644 --- a/fs/xfs/linux-2.6/xfs_sync.h +++ b/fs/xfs/linux-2.6/xfs_sync.h @@ -48,7 +48,6 @@ 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_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_ag.h b/fs/xfs/xfs_ag.h index f24b50b68d0..af3cfeb18b0 100644 --- a/fs/xfs/xfs_ag.h +++ b/fs/xfs/xfs_ag.h @@ -91,9 +91,6 @@ typedef struct xfs_agf { #define XFS_AGF_BLOCK(mp) XFS_HDR_BLOCK(mp, XFS_AGF_DADDR(mp)) #define XFS_BUF_TO_AGF(bp) ((xfs_agf_t *)XFS_BUF_PTR(bp)) -extern int xfs_read_agf(struct xfs_mount *mp, struct xfs_trans *tp, - xfs_agnumber_t agno, int flags, struct xfs_buf **bpp); - /* * Size of the unlinked inode hash table in the agi. */ diff --git a/fs/xfs/xfs_alloc.c b/fs/xfs/xfs_alloc.c index 2cf944eb796..6316004922d 100644 --- a/fs/xfs/xfs_alloc.c +++ b/fs/xfs/xfs_alloc.c @@ -2248,7 +2248,7 @@ xfs_alloc_put_freelist( /* * Read in the allocation group header (free/alloc section). */ -int /* error */ +STATIC int /* error */ xfs_read_agf( struct xfs_mount *mp, /* mount point structure */ struct xfs_trans *tp, /* transaction pointer */ diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c index 7928b9983c1..975972482e8 100644 --- a/fs/xfs/xfs_bmap.c +++ b/fs/xfs/xfs_bmap.c @@ -3713,7 +3713,7 @@ done: * entry (null if none). Else, *lastxp will be set to the index * of the found entry; *gotp will contain the entry. */ -xfs_bmbt_rec_host_t * /* pointer to found extent entry */ +STATIC xfs_bmbt_rec_host_t * /* pointer to found extent entry */ xfs_bmap_search_multi_extents( xfs_ifork_t *ifp, /* inode fork pointer */ xfs_fileoff_t bno, /* block number searched for */ diff --git a/fs/xfs/xfs_bmap.h b/fs/xfs/xfs_bmap.h index 1b8ff9256bd..56f62d2edc3 100644 --- a/fs/xfs/xfs_bmap.h +++ b/fs/xfs/xfs_bmap.h @@ -392,17 +392,6 @@ xfs_bmap_count_blocks( int whichfork, int *count); -/* - * Search the extent records for the entry containing block bno. - * If bno lies in a hole, point to the next entry. If bno lies - * past eof, *eofp will be set, and *prevp will contain the last - * entry (null if none). Else, *lastxp will be set to the index - * of the found entry; *gotp will contain the entry. - */ -xfs_bmbt_rec_host_t * -xfs_bmap_search_multi_extents(struct xfs_ifork *, xfs_fileoff_t, int *, - xfs_extnum_t *, xfs_bmbt_irec_t *, xfs_bmbt_irec_t *); - #endif /* __KERNEL__ */ #endif /* __XFS_BMAP_H__ */ diff --git a/fs/xfs/xfs_bmap_btree.c b/fs/xfs/xfs_bmap_btree.c index 5c1ade06578..eb7b702d069 100644 --- a/fs/xfs/xfs_bmap_btree.c +++ b/fs/xfs/xfs_bmap_btree.c @@ -202,16 +202,6 @@ xfs_bmbt_get_state( ext_flag); } -/* Endian flipping versions of the bmbt extraction functions */ -void -xfs_bmbt_disk_get_all( - xfs_bmbt_rec_t *r, - xfs_bmbt_irec_t *s) -{ - __xfs_bmbt_get_all(get_unaligned_be64(&r->l0), - get_unaligned_be64(&r->l1), s); -} - /* * Extract the blockcount field from an on disk bmap extent record. */ @@ -816,6 +806,16 @@ xfs_bmbt_trace_key( *l1 = 0; } +/* Endian flipping versions of the bmbt extraction functions */ +STATIC void +xfs_bmbt_disk_get_all( + xfs_bmbt_rec_t *r, + xfs_bmbt_irec_t *s) +{ + __xfs_bmbt_get_all(get_unaligned_be64(&r->l0), + get_unaligned_be64(&r->l1), s); +} + STATIC void xfs_bmbt_trace_record( struct xfs_btree_cur *cur, diff --git a/fs/xfs/xfs_bmap_btree.h b/fs/xfs/xfs_bmap_btree.h index 0e8df007615..5549d495947 100644 --- a/fs/xfs/xfs_bmap_btree.h +++ b/fs/xfs/xfs_bmap_btree.h @@ -220,7 +220,6 @@ extern xfs_fsblock_t xfs_bmbt_get_startblock(xfs_bmbt_rec_host_t *r); extern xfs_fileoff_t xfs_bmbt_get_startoff(xfs_bmbt_rec_host_t *r); extern xfs_exntst_t xfs_bmbt_get_state(xfs_bmbt_rec_host_t *r); -extern void xfs_bmbt_disk_get_all(xfs_bmbt_rec_t *r, xfs_bmbt_irec_t *s); extern xfs_filblks_t xfs_bmbt_disk_get_blockcount(xfs_bmbt_rec_t *r); extern xfs_fileoff_t xfs_bmbt_disk_get_startoff(xfs_bmbt_rec_t *r); diff --git a/fs/xfs/xfs_btree.c b/fs/xfs/xfs_btree.c index e9df9957482..cde5a266829 100644 --- a/fs/xfs/xfs_btree.c +++ b/fs/xfs/xfs_btree.c @@ -645,46 +645,6 @@ xfs_btree_read_bufl( return 0; } -/* - * Get a buffer for the block, return it read in. - * Short-form addressing. - */ -int /* error */ -xfs_btree_read_bufs( - xfs_mount_t *mp, /* file system mount point */ - xfs_trans_t *tp, /* transaction pointer */ - xfs_agnumber_t agno, /* allocation group number */ - xfs_agblock_t agbno, /* allocation group block number */ - uint lock, /* lock flags for read_buf */ - xfs_buf_t **bpp, /* buffer for agno/agbno */ - int refval) /* ref count value for buffer */ -{ - xfs_buf_t *bp; /* return value */ - xfs_daddr_t d; /* real disk block address */ - int error; - - ASSERT(agno != NULLAGNUMBER); - ASSERT(agbno != NULLAGBLOCK); - d = XFS_AGB_TO_DADDR(mp, agno, agbno); - if ((error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp, d, - mp->m_bsize, lock, &bp))) { - return error; - } - ASSERT(!bp || !XFS_BUF_GETERROR(bp)); - if (bp != NULL) { - switch (refval) { - case XFS_ALLOC_BTREE_REF: - XFS_BUF_SET_VTYPE_REF(bp, B_FS_MAP, refval); - break; - case XFS_INO_BTREE_REF: - XFS_BUF_SET_VTYPE_REF(bp, B_FS_INOMAP, refval); - break; - } - } - *bpp = bp; - return 0; -} - /* * Read-ahead the block, don't wait for it, don't return a buffer. * Long-form addressing. @@ -2951,7 +2911,7 @@ error0: * inode we have to copy the single block it was pointing to into the * inode. */ -int +STATIC int xfs_btree_kill_iroot( struct xfs_btree_cur *cur) { diff --git a/fs/xfs/xfs_btree.h b/fs/xfs/xfs_btree.h index 4f852b735b9..7fa07062bdd 100644 --- a/fs/xfs/xfs_btree.h +++ b/fs/xfs/xfs_btree.h @@ -378,20 +378,6 @@ xfs_btree_read_bufl( struct xfs_buf **bpp, /* buffer for fsbno */ int refval);/* ref count value for buffer */ -/* - * Get a buffer for the block, return it read in. - * Short-form addressing. - */ -int /* error */ -xfs_btree_read_bufs( - struct xfs_mount *mp, /* file system mount point */ - struct xfs_trans *tp, /* transaction pointer */ - xfs_agnumber_t agno, /* allocation group number */ - xfs_agblock_t agbno, /* allocation group block number */ - uint lock, /* lock flags for read_buf */ - struct xfs_buf **bpp, /* buffer for agno/agbno */ - int refval);/* ref count value for buffer */ - /* * Read-ahead the block, don't wait for it, don't return a buffer. * Long-form addressing. @@ -432,7 +418,6 @@ int xfs_btree_decrement(struct xfs_btree_cur *, int, int *); int xfs_btree_lookup(struct xfs_btree_cur *, xfs_lookup_t, int *); int xfs_btree_update(struct xfs_btree_cur *, union xfs_btree_rec *); int xfs_btree_new_iroot(struct xfs_btree_cur *, int *, int *); -int xfs_btree_kill_iroot(struct xfs_btree_cur *); int xfs_btree_insert(struct xfs_btree_cur *, int *); int xfs_btree_delete(struct xfs_btree_cur *, int *); int xfs_btree_get_rec(struct xfs_btree_cur *, union xfs_btree_rec **, int *); diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index 1f22d65fed0..2dcb3d781ae 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -641,7 +641,7 @@ xfs_iformat_btree( return 0; } -void +STATIC void xfs_dinode_from_disk( xfs_icdinode_t *to, xfs_dinode_t *from) @@ -1237,7 +1237,7 @@ xfs_isize_check( * In that case the pages will still be in memory, but the inode size * will never have been updated. */ -xfs_fsize_t +STATIC xfs_fsize_t xfs_file_last_byte( xfs_inode_t *ip) { @@ -3827,7 +3827,7 @@ xfs_iext_inline_to_direct( /* * Resize an extent indirection array to new_size bytes. */ -void +STATIC void xfs_iext_realloc_indirect( xfs_ifork_t *ifp, /* inode fork pointer */ int new_size) /* new indirection array size */ @@ -3852,7 +3852,7 @@ xfs_iext_realloc_indirect( /* * Switch from indirection array to linear (direct) extent allocations. */ -void +STATIC void xfs_iext_indirect_to_direct( xfs_ifork_t *ifp) /* inode fork pointer */ { diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h index 77016702938..a0b2da66df3 100644 --- a/fs/xfs/xfs_inode.h +++ b/fs/xfs/xfs_inode.h @@ -526,7 +526,6 @@ void xfs_ipin(xfs_inode_t *); void xfs_iunpin(xfs_inode_t *); int xfs_iflush(xfs_inode_t *, uint); void xfs_ichgtime(xfs_inode_t *, int); -xfs_fsize_t xfs_file_last_byte(xfs_inode_t *); void xfs_lock_inodes(xfs_inode_t **, int, uint); void xfs_lock_two_inodes(xfs_inode_t *, xfs_inode_t *, uint); @@ -594,8 +593,6 @@ int xfs_itobp(struct xfs_mount *, struct xfs_trans *, struct xfs_buf **, uint); int xfs_iread(struct xfs_mount *, struct xfs_trans *, struct xfs_inode *, xfs_daddr_t, uint); -void xfs_dinode_from_disk(struct xfs_icdinode *, - struct xfs_dinode *); void xfs_dinode_to_disk(struct xfs_dinode *, struct xfs_icdinode *); void xfs_idestroy_fork(struct xfs_inode *, int); @@ -614,8 +611,6 @@ void xfs_iext_remove_inline(xfs_ifork_t *, xfs_extnum_t, int); void xfs_iext_remove_direct(xfs_ifork_t *, xfs_extnum_t, int); void xfs_iext_remove_indirect(xfs_ifork_t *, xfs_extnum_t, int); void xfs_iext_realloc_direct(xfs_ifork_t *, int); -void xfs_iext_realloc_indirect(xfs_ifork_t *, int); -void xfs_iext_indirect_to_direct(xfs_ifork_t *); void xfs_iext_direct_to_inline(xfs_ifork_t *, xfs_extnum_t); void xfs_iext_inline_to_direct(xfs_ifork_t *, int); void xfs_iext_destroy(xfs_ifork_t *); diff --git a/fs/xfs/xfs_itable.c b/fs/xfs/xfs_itable.c index aeb2d2221c7..c471122d223 100644 --- a/fs/xfs/xfs_itable.c +++ b/fs/xfs/xfs_itable.c @@ -39,7 +39,7 @@ #include "xfs_error.h" #include "xfs_btree.h" -int +STATIC int xfs_internal_inum( xfs_mount_t *mp, xfs_ino_t ino) diff --git a/fs/xfs/xfs_itable.h b/fs/xfs/xfs_itable.h index 1fb04e7deb6..20792bf4594 100644 --- a/fs/xfs/xfs_itable.h +++ b/fs/xfs/xfs_itable.h @@ -99,11 +99,6 @@ xfs_bulkstat_one( void *dibuff, int *stat); -int -xfs_internal_inum( - xfs_mount_t *mp, - xfs_ino_t ino); - typedef int (*inumbers_fmt_pf)( void __user *ubuffer, /* buffer to write to */ const xfs_inogrp_t *buffer, /* buffer to read from */ diff --git a/fs/xfs/xfs_log_priv.h b/fs/xfs/xfs_log_priv.h index bcad5f4c1fd..679c7c4926a 100644 --- a/fs/xfs/xfs_log_priv.h +++ b/fs/xfs/xfs_log_priv.h @@ -451,8 +451,6 @@ extern int xlog_find_tail(xlog_t *log, extern int xlog_recover(xlog_t *log); extern int xlog_recover_finish(xlog_t *log); extern void xlog_pack_data(xlog_t *log, xlog_in_core_t *iclog, int); -extern void xlog_recover_process_iunlinks(xlog_t *log); - extern struct xfs_buf *xlog_get_bp(xlog_t *, int); extern void xlog_put_bp(struct xfs_buf *); diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c index 47da2fb4537..1099395d7d6 100644 --- a/fs/xfs/xfs_log_recover.c +++ b/fs/xfs/xfs_log_recover.c @@ -3263,7 +3263,7 @@ xlog_recover_process_one_iunlink( * freeing of the inode and its removal from the list must be * atomic. */ -void +STATIC void xlog_recover_process_iunlinks( xlog_t *log) { diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index 5c6f092659c..8b6c9e807ef 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -1568,7 +1568,7 @@ xfs_mod_sb(xfs_trans_t *tp, __int64_t fields) * * The m_sb_lock must be held when this routine is called. */ -int +STATIC int xfs_mod_incore_sb_unlocked( xfs_mount_t *mp, xfs_sb_field_t field, diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h index a5122382afd..a6c023bc0fb 100644 --- a/fs/xfs/xfs_mount.h +++ b/fs/xfs/xfs_mount.h @@ -414,13 +414,10 @@ typedef struct xfs_mod_sb { extern int xfs_log_sbcount(xfs_mount_t *, uint); extern int xfs_mountfs(xfs_mount_t *mp); -extern void xfs_mountfs_check_barriers(xfs_mount_t *mp); extern void xfs_unmountfs(xfs_mount_t *); extern int xfs_unmountfs_writesb(xfs_mount_t *); extern int xfs_mod_incore_sb(xfs_mount_t *, xfs_sb_field_t, int64_t, int); -extern int xfs_mod_incore_sb_unlocked(xfs_mount_t *, xfs_sb_field_t, - int64_t, int); extern int xfs_mod_incore_sb_batch(xfs_mount_t *, xfs_mod_sb_t *, uint, int); extern int xfs_mount_log_sb(xfs_mount_t *, __int64_t); diff --git a/fs/xfs/xfs_mru_cache.c b/fs/xfs/xfs_mru_cache.c index afee7eb2432..4b0613d99fa 100644 --- a/fs/xfs/xfs_mru_cache.c +++ b/fs/xfs/xfs_mru_cache.c @@ -563,35 +563,6 @@ xfs_mru_cache_lookup( return elem ? elem->value : NULL; } -/* - * To look up an element using its key, but leave its location in the internal - * lists alone, call xfs_mru_cache_peek(). If the element isn't found, this - * function returns NULL. - * - * See the comments above the declaration of the xfs_mru_cache_lookup() function - * for important locking information pertaining to this call. - */ -void * -xfs_mru_cache_peek( - xfs_mru_cache_t *mru, - unsigned long key) -{ - xfs_mru_cache_elem_t *elem; - - ASSERT(mru && mru->lists); - if (!mru || !mru->lists) - return NULL; - - spin_lock(&mru->lock); - elem = radix_tree_lookup(&mru->store, key); - if (!elem) - spin_unlock(&mru->lock); - else - __release(mru_lock); /* help sparse not be stupid */ - - return elem ? elem->value : NULL; -} - /* * To release the internal data structure spinlock after having performed an * xfs_mru_cache_lookup() or an xfs_mru_cache_peek(), call xfs_mru_cache_done() diff --git a/fs/xfs/xfs_mru_cache.h b/fs/xfs/xfs_mru_cache.h index dd58ea1bbeb..5d439f34b0c 100644 --- a/fs/xfs/xfs_mru_cache.h +++ b/fs/xfs/xfs_mru_cache.h @@ -49,7 +49,6 @@ int xfs_mru_cache_insert(struct xfs_mru_cache *mru, unsigned long key, void * xfs_mru_cache_remove(struct xfs_mru_cache *mru, unsigned long key); void xfs_mru_cache_delete(struct xfs_mru_cache *mru, unsigned long key); void *xfs_mru_cache_lookup(struct xfs_mru_cache *mru, unsigned long key); -void *xfs_mru_cache_peek(struct xfs_mru_cache *mru, unsigned long key); void xfs_mru_cache_done(struct xfs_mru_cache *mru); #endif /* __XFS_MRU_CACHE_H__ */ diff --git a/fs/xfs/xfs_rw.h b/fs/xfs/xfs_rw.h index f76c003ec55..ae65f0df87d 100644 --- a/fs/xfs/xfs_rw.h +++ b/fs/xfs/xfs_rw.h @@ -78,10 +78,4 @@ extern int xfs_read_buf(struct xfs_mount *mp, xfs_buftarg_t *btp, extern void xfs_ioerror_alert(char *func, struct xfs_mount *mp, xfs_buf_t *bp, xfs_daddr_t blkno); -/* - * Prototypes for functions in xfs_vnodeops.c. - */ -extern int xfs_free_eofblocks(struct xfs_mount *mp, struct xfs_inode *ip, - int flags); - #endif /* __XFS_RW_H__ */ diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c index c4eca5ed5da..1dd70688715 100644 --- a/fs/xfs/xfs_vnodeops.c +++ b/fs/xfs/xfs_vnodeops.c @@ -716,7 +716,7 @@ xfs_fsync( * when the link count isn't zero and by xfs_dm_punch_hole() when * punching a hole to EOF. */ -int +STATIC int xfs_free_eofblocks( xfs_mount_t *mp, xfs_inode_t *ip, From ab8b9baac3f48fb3fc3d47790950c0eec134e678 Mon Sep 17 00:00:00 2001 From: Eric Sandeen Date: Thu, 2 Jul 2009 21:35:43 -0500 Subject: [PATCH 02/19] un-static xfs_read_agf CONFIG_XFS_DEBUG builds still need xfs_read_agf to be non-static, oops. Signed-off-by: Eric Sandeen Reviewed-by: Felix Blyakher Signed-off-by: Felix Blyakher --- fs/xfs/xfs_ag.h | 3 +++ fs/xfs/xfs_alloc.c | 2 +- 2 files changed, 4 insertions(+), 1 deletion(-) diff --git a/fs/xfs/xfs_ag.h b/fs/xfs/xfs_ag.h index af3cfeb18b0..f24b50b68d0 100644 --- a/fs/xfs/xfs_ag.h +++ b/fs/xfs/xfs_ag.h @@ -91,6 +91,9 @@ typedef struct xfs_agf { #define XFS_AGF_BLOCK(mp) XFS_HDR_BLOCK(mp, XFS_AGF_DADDR(mp)) #define XFS_BUF_TO_AGF(bp) ((xfs_agf_t *)XFS_BUF_PTR(bp)) +extern int xfs_read_agf(struct xfs_mount *mp, struct xfs_trans *tp, + xfs_agnumber_t agno, int flags, struct xfs_buf **bpp); + /* * Size of the unlinked inode hash table in the agi. */ diff --git a/fs/xfs/xfs_alloc.c b/fs/xfs/xfs_alloc.c index 6316004922d..2cf944eb796 100644 --- a/fs/xfs/xfs_alloc.c +++ b/fs/xfs/xfs_alloc.c @@ -2248,7 +2248,7 @@ xfs_alloc_put_freelist( /* * Read in the allocation group header (free/alloc section). */ -STATIC int /* error */ +int /* error */ xfs_read_agf( struct xfs_mount *mp, /* mount point structure */ struct xfs_trans *tp, /* transaction pointer */ From b56063453881a6d94cf5718c6769de6e35e67753 Mon Sep 17 00:00:00 2001 From: Eric Sandeen Date: Sun, 5 Jul 2009 12:23:35 -0500 Subject: [PATCH 03/19] xfs: remove XFS_INO64_OFFSET Commit a19d9f887d81106d52cacbc9930207b487e07e0e removed the ino64 option but left the XFS_INO64_OFFSET define it used in place - just remove it. Signed-off-by: Eric Sandeen Reviewed-by: Christoph Hellwig Reviewed-by: Felix Blyakher Signed-off-by: Felix Blyakher --- fs/xfs/xfs_inum.h | 1 - 1 file changed, 1 deletion(-) diff --git a/fs/xfs/xfs_inum.h b/fs/xfs/xfs_inum.h index 7a28191cb0d..b8e4ee4e89a 100644 --- a/fs/xfs/xfs_inum.h +++ b/fs/xfs/xfs_inum.h @@ -72,7 +72,6 @@ struct xfs_mount; #if XFS_BIG_INUMS #define XFS_MAXINUMBER ((xfs_ino_t)((1ULL << 56) - 1ULL)) -#define XFS_INO64_OFFSET ((xfs_ino_t)(1ULL << 32)) #else #define XFS_MAXINUMBER ((xfs_ino_t)((1ULL << 32) - 1ULL)) #endif From 5c04c78afba4805846519f29f0b55ac8759e0d48 Mon Sep 17 00:00:00 2001 From: Eric Sandeen Date: Sun, 26 Jul 2009 21:52:01 -0500 Subject: [PATCH 04/19] 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(-) diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c index 58973bb4603..8070b34cc28 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; From ede58517ca93277547a0d054728c352618212d85 Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Mon, 27 Jul 2009 18:15:25 +0200 Subject: [PATCH 05/19] fs/xfs: Correct redundant test bp was tested for NULL a few lines before, followed by a return, and there is no intervening modification of its value. A simplified version of the semantic match that finds this problem is as follows: (http://www.emn.fr/x-info/coccinelle/) // @r exists@ local idexpression x; expression E; position p1,p2; @@ if (x == NULL || ...) { ... when forall return ...; } ... when != \(x=E\|x--\|x++\|--x\|++x\|x-=E\|x+=E\|x|=E\|x&=E\|&x\) ( *x == NULL | *x != NULL ) // Signed-off-by: Julia Lawall Acked-by: Felix Blyakher Signed-off-by: Felix Blyakher --- fs/xfs/xfs_trans_buf.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/xfs/xfs_trans_buf.c b/fs/xfs/xfs_trans_buf.c index 8ee2f8c8b0a..218829e6a15 100644 --- a/fs/xfs/xfs_trans_buf.c +++ b/fs/xfs/xfs_trans_buf.c @@ -307,7 +307,7 @@ xfs_trans_read_buf( return (flags & XFS_BUF_TRYLOCK) ? EAGAIN : XFS_ERROR(ENOMEM); - if ((bp != NULL) && (XFS_BUF_GETERROR(bp) != 0)) { + if (XFS_BUF_GETERROR(bp) != 0) { xfs_ioerror_alert("xfs_trans_read_buf", mp, bp, blkno); error = XFS_BUF_GETERROR(bp); @@ -315,7 +315,7 @@ xfs_trans_read_buf( return error; } #ifdef DEBUG - if (xfs_do_error && (bp != NULL)) { + if (xfs_do_error) { if (xfs_error_target == target) { if (((xfs_req_num++) % xfs_error_mod) == 0) { xfs_buf_relse(bp); From a81655ae5a9da16a08ba14637dc6c10217e57492 Mon Sep 17 00:00:00 2001 From: Eric Sandeen Date: Fri, 31 Jul 2009 00:02:17 -0500 Subject: [PATCH 06/19] 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(+) diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c index 7ec89fc05b2..aecf2519db7 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. From efe1f30e8f05a3836dfcf6302774707b3b47193f Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Sat, 18 Jul 2009 18:14:53 -0400 Subject: [PATCH 07/19] 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(-) diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c index cbd451bb484..2d0b3e1da9e 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; From 1ec2b8a777d6ac5464a836b3c5520b1a5190194c Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Sat, 18 Jul 2009 18:14:54 -0400 Subject: [PATCH 08/19] 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(-) diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c index 975972482e8..8971fb09d38 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; From 0043dd88af762d3a9154317d351400b5b941ec8d Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Sat, 18 Jul 2009 18:14:55 -0400 Subject: [PATCH 09/19] 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(-) diff --git a/fs/xfs/xfs_da_btree.c b/fs/xfs/xfs_da_btree.c index 9ff6e57a507..bd0bb6dfcdf 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); } /* From eeb6f67ba1176a141b6802b2ec8505835482f631 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Sat, 18 Jul 2009 18:14:56 -0400 Subject: [PATCH 10/19] 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(-) diff --git a/fs/xfs/xfs_da_btree.c b/fs/xfs/xfs_da_btree.c index bd0bb6dfcdf..2847bbc1c53 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; From ea78cd5e6b6ebe123641382d5040252d19829332 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Sat, 18 Jul 2009 18:14:57 -0400 Subject: [PATCH 11/19] 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(-) diff --git a/fs/xfs/xfs_dir2.c b/fs/xfs/xfs_dir2.c index c657bec6d95..bb1d58eb398 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; From 837273b8a150a3761467573c5eb477791cf19268 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Sat, 18 Jul 2009 18:14:58 -0400 Subject: [PATCH 12/19] 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(-) diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c index e28800a9f2b..da3b666f1ac 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; From f56ebd66df411446030e2cc820554c6ce4ba0836 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Sat, 18 Jul 2009 18:14:59 -0400 Subject: [PATCH 13/19] 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(-) diff --git a/fs/xfs/xfs_attr.c b/fs/xfs/xfs_attr.c index db15feb906f..bfb583791fe 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)); From 5a25f1abb26f5ab7c6c72fe0b3807c137ee484bb Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Sat, 18 Jul 2009 18:15:00 -0400 Subject: [PATCH 14/19] 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(-) diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c index 1dd70688715..ceecafd1f9c 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", From 7509ee34a365c041718d0b676e5ee255b58b5dda Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Sat, 18 Jul 2009 18:15:01 -0400 Subject: [PATCH 15/19] 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(-) diff --git a/fs/xfs/xfs_attr.c b/fs/xfs/xfs_attr.c index bfb583791fe..4ece1906bd4 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); From 583e0e41ee29478e2147aeffeb772657f8db767a Mon Sep 17 00:00:00 2001 From: Eric Sandeen Date: Mon, 20 Jul 2009 10:52:15 -0500 Subject: [PATCH 16/19] 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(-) diff --git a/fs/xfs/xfs_btree.c b/fs/xfs/xfs_btree.c index cde5a266829..52b5f14d0c3 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; From 16715dbe64ccac265010ab8b60848d212d002521 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 10 Aug 2009 11:32:18 -0300 Subject: [PATCH 17/19] 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(+) diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index 2dcb3d781ae..c1dc7ef5a1d 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: From 79dd43bb85d64ba14a781f940c858d7bbe8c9a6d Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 10 Aug 2009 11:32:44 -0300 Subject: [PATCH 18/19] 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(-) diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c index 3750f04ede0..9dbdff3ea48 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); From a022fe09700365c51d1f55884bca9754eb96a802 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Sun, 16 Aug 2009 20:36:34 -0400 Subject: [PATCH 19/19] 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 | 115 ++++++++++++++++++------------------ 3 files changed, 71 insertions(+), 58 deletions(-) diff --git a/fs/xfs/linux-2.6/xfs_sync.c b/fs/xfs/linux-2.6/xfs_sync.c index fbf3e0288b3..320be6aea49 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 23e7e7e6e13..27920eb7a82 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_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 76c540f719e..91adfab2f45 100644 --- a/fs/xfs/xfs_iget.c +++ b/fs/xfs/xfs_iget.c @@ -134,80 +134,82 @@ 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; + ip->i_flags |= XFS_INEW; + ip->i_flags &= ~XFS_IRECLAIMABLE; + __xfs_inode_clear_reclaim_tag(mp, pag, ip); + + spin_unlock(&ip->i_flags_lock); + read_unlock(&pag->pag_ici_lock); + + 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. */ + if (!igrab(inode)) { + error = EAGAIN; goto out_error; } - /* - * 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); - - /* clear the radix tree reclaim flag as well. */ - __xfs_inode_clear_reclaim_tag(mp, pag, ip); - } else if (!igrab(VFS_I(ip))) { - /* 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)); + /* We've got a live one. */ + spin_unlock(&ip->i_flags_lock); + read_unlock(&pag->pag_ici_lock); } - 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. */ - read_unlock(&pag->pag_ici_lock); - if (lock_flags != 0) xfs_ilock(ip, lock_flags); @@ -217,6 +219,7 @@ xfs_iget_cache_hit( return 0; out_error: + spin_unlock(&ip->i_flags_lock); read_unlock(&pag->pag_ici_lock); return error; }