From 1be62dc190ebaca331038962c873e7967de6cc4b Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Fri, 4 Apr 2008 14:38:17 -0700 Subject: Be more careful about marking buffers dirty Mikulas Patocka noted that the optimization where we check if a buffer was already dirty (and we avoid re-dirtying it) was not really SMP-safe. Since the read of the old status was not synchronized with anything, an aggressive CPU re-ordering of memory accesses might have moved that read up to before the data was even written to the buffer, and another CPU that cleaned it again, causing the newly dirty state to never actually hit the disk. Admittedly this would probably never trigger in practice, but it's still wrong. Mikulas sent a patch that fixed the problem, but I dislike the subtlety of the whole optimization, so this is an alternate fix that is more explicit about the particular SMP ordering for the optimization, and separates out the speculative reads of the buffer state into its own conditional (and makes the memory barrier only happen if we are likely to actually hit the optimized case in the first place). I considered removing the optimization entirely, but Andrew argued for it's continued existence. I'm a push-over. Cc: Mikulas Patocka Cc: Andrew Morton Signed-off-by: Linus Torvalds --- fs/buffer.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/buffer.c b/fs/buffer.c index 98196327ddf..39ff14403d1 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -1181,7 +1181,20 @@ __getblk_slow(struct block_device *bdev, sector_t block, int size) void mark_buffer_dirty(struct buffer_head *bh) { WARN_ON_ONCE(!buffer_uptodate(bh)); - if (!buffer_dirty(bh) && !test_set_buffer_dirty(bh)) + + /* + * Very *carefully* optimize the it-is-already-dirty case. + * + * Don't let the final "is it dirty" escape to before we + * perhaps modified the buffer. + */ + if (buffer_dirty(bh)) { + smp_mb(); + if (buffer_dirty(bh)) + return; + } + + if (!test_set_buffer_dirty(bh)) __set_page_dirty(bh->b_page, page_mapping(bh->b_page), 0); } -- cgit v1.2.3 From 66d3aac04175c35564cc5982934b68bc4f89a76d Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Mon, 31 Mar 2008 15:01:58 -0400 Subject: NFS: initialize flags field in nfs_open_context The nfs_open_context struct had a "flags" field added recently, but the allocator isn't initializing it. It also looks like the allocator isn't initializing the mode or list either, but they seem to be overwritten by the caller, so that's less of an issue. Signed-off-by: Jeff Layton Signed-off-by: Trond Myklebust --- fs/nfs/inode.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index a4c7cf2bff3..6f88d7c77ac 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -506,6 +506,7 @@ static struct nfs_open_context *alloc_nfs_open_context(struct vfsmount *mnt, str ctx->cred = get_rpccred(cred); ctx->state = NULL; ctx->lockowner = current->files; + ctx->flags = 0; ctx->error = 0; ctx->dir_cookie = 0; atomic_set(&ctx->count, 1); -- cgit v1.2.3 From 240ee831187e000704cba3208ef6b0a8ad49bd64 Mon Sep 17 00:00:00 2001 From: Bryan Wu Date: Tue, 1 Apr 2008 20:23:39 -0700 Subject: fix bug - executing FDPIC ELF on NFS mount triggers BUG() at mm/nommu.c:862:/do_mmap_private() NFS needs a NOMMU version mmap function to support uClinux on NOMMU machine http://blackfin.uclinux.org/gf/project/uclinux-dist/tracker/?action=TrackerItemEdit&tracker_id=141&tracker_item_id=3992 Signed-off-by: Bryan Wu Cc: Mike Frysinger Signed-off-by: Trond Myklebust --- fs/nfs/file.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'fs') diff --git a/fs/nfs/file.c b/fs/nfs/file.c index ef57a5ae590..5d2e9d9a4e2 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -64,7 +64,11 @@ const struct file_operations nfs_file_operations = { .write = do_sync_write, .aio_read = nfs_file_read, .aio_write = nfs_file_write, +#ifdef CONFIG_MMU .mmap = nfs_file_mmap, +#else + .mmap = generic_file_mmap, +#endif .open = nfs_file_open, .flush = nfs_file_flush, .release = nfs_file_release, -- cgit v1.2.3 From 8191ecd1d14c6914c660dfa007154860a7908857 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Thu, 10 Apr 2008 08:24:25 +0200 Subject: splice: fix infinite loop in generic_file_splice_read() There's a quirky loop in generic_file_splice_read() that could go on indefinitely, if the file splice returns 0 permanently (and not just as a temporary condition). Get rid of the loop and pass back -EAGAIN correctly from __generic_file_splice_read(), so we handle that condition properly as well. Signed-off-by: Jens Axboe --- fs/splice.c | 31 ++++++------------------------- 1 file changed, 6 insertions(+), 25 deletions(-) (limited to 'fs') diff --git a/fs/splice.c b/fs/splice.c index a861bb318ac..eeb1a86a701 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -370,8 +370,10 @@ __generic_file_splice_read(struct file *in, loff_t *ppos, * for an in-flight io page */ if (flags & SPLICE_F_NONBLOCK) { - if (TestSetPageLocked(page)) + if (TestSetPageLocked(page)) { + error = -EAGAIN; break; + } } else lock_page(page); @@ -479,9 +481,8 @@ ssize_t generic_file_splice_read(struct file *in, loff_t *ppos, struct pipe_inode_info *pipe, size_t len, unsigned int flags) { - ssize_t spliced; - int ret; loff_t isize, left; + int ret; isize = i_size_read(in->f_mapping->host); if (unlikely(*ppos >= isize)) @@ -491,29 +492,9 @@ ssize_t generic_file_splice_read(struct file *in, loff_t *ppos, if (unlikely(left < len)) len = left; - ret = 0; - spliced = 0; - while (len && !spliced) { - ret = __generic_file_splice_read(in, ppos, pipe, len, flags); - - if (ret < 0) - break; - else if (!ret) { - if (spliced) - break; - if (flags & SPLICE_F_NONBLOCK) { - ret = -EAGAIN; - break; - } - } - + ret = __generic_file_splice_read(in, ppos, pipe, len, flags); + if (ret > 0) *ppos += ret; - len -= ret; - spliced += ret; - } - - if (spliced) - return spliced; return ret; } -- cgit v1.2.3 From 6211870992502efdccf9cd6bae155c83277835fa Mon Sep 17 00:00:00 2001 From: Eric Sandeen Date: Thu, 6 Mar 2008 13:44:28 +1100 Subject: [XFS] remove shouting-indirection macros from xfs_sb.h Remove macro-to-small-function indirection from xfs_sb.h, and remove some which are completely unused. SGI-PV: 976035 SGI-Modid: xfs-linux-melb:xfs-kern:30528a Signed-off-by: Eric Sandeen Signed-off-by: Donald Douwsma Signed-off-by: Lachlan McIlroy --- fs/xfs/linux-2.6/xfs_ioctl.c | 2 +- fs/xfs/quota/xfs_qm.c | 6 ++-- fs/xfs/quota/xfs_qm_bhv.c | 2 +- fs/xfs/quota/xfs_qm_syscalls.c | 4 +-- fs/xfs/xfs_attr_leaf.c | 6 ++-- fs/xfs/xfs_bmap.c | 18 +++++------ fs/xfs/xfs_bmap_btree.h | 2 +- fs/xfs/xfs_dir2.c | 2 +- fs/xfs/xfs_fsops.c | 24 +++++++-------- fs/xfs/xfs_ialloc.c | 6 ++-- fs/xfs/xfs_inode.c | 6 ++-- fs/xfs/xfs_inode_item.c | 4 +-- fs/xfs/xfs_itable.c | 2 +- fs/xfs/xfs_log.c | 14 ++++----- fs/xfs/xfs_log_priv.h | 4 +-- fs/xfs/xfs_log_recover.c | 16 +++++----- fs/xfs/xfs_mount.c | 10 +++--- fs/xfs/xfs_sb.h | 69 +++--------------------------------------- fs/xfs/xfs_utils.c | 6 ++-- fs/xfs/xfs_vfsops.c | 9 +++--- fs/xfs/xfs_vnodeops.c | 2 +- 21 files changed, 76 insertions(+), 138 deletions(-) (limited to 'fs') diff --git a/fs/xfs/linux-2.6/xfs_ioctl.c b/fs/xfs/linux-2.6/xfs_ioctl.c index a9952e490ac..f34bd010eb5 100644 --- a/fs/xfs/linux-2.6/xfs_ioctl.c +++ b/fs/xfs/linux-2.6/xfs_ioctl.c @@ -732,7 +732,7 @@ xfs_ioctl( * Only allow the sys admin to reserve space unless * unwritten extents are enabled. */ - if (!XFS_SB_VERSION_HASEXTFLGBIT(&mp->m_sb) && + if (!xfs_sb_version_hasextflgbit(&mp->m_sb) && !capable(CAP_SYS_ADMIN)) return -EPERM; diff --git a/fs/xfs/quota/xfs_qm.c b/fs/xfs/quota/xfs_qm.c index 1f3da5b8657..8e9c5ae6504 100644 --- a/fs/xfs/quota/xfs_qm.c +++ b/fs/xfs/quota/xfs_qm.c @@ -1405,13 +1405,13 @@ xfs_qm_qino_alloc( #if defined(DEBUG) && defined(XFS_LOUD_RECOVERY) unsigned oldv = mp->m_sb.sb_versionnum; #endif - ASSERT(!XFS_SB_VERSION_HASQUOTA(&mp->m_sb)); + ASSERT(!xfs_sb_version_hasquota(&mp->m_sb)); ASSERT((sbfields & (XFS_SB_VERSIONNUM | XFS_SB_UQUOTINO | XFS_SB_GQUOTINO | XFS_SB_QFLAGS)) == (XFS_SB_VERSIONNUM | XFS_SB_UQUOTINO | XFS_SB_GQUOTINO | XFS_SB_QFLAGS)); - XFS_SB_VERSION_ADDQUOTA(&mp->m_sb); + xfs_sb_version_addquota(&mp->m_sb); mp->m_sb.sb_uquotino = NULLFSINO; mp->m_sb.sb_gquotino = NULLFSINO; @@ -1954,7 +1954,7 @@ xfs_qm_init_quotainos( /* * Get the uquota and gquota inodes */ - if (XFS_SB_VERSION_HASQUOTA(&mp->m_sb)) { + if (xfs_sb_version_hasquota(&mp->m_sb)) { if (XFS_IS_UQUOTA_ON(mp) && mp->m_sb.sb_uquotino != NULLFSINO) { ASSERT(mp->m_sb.sb_uquotino > 0); diff --git a/fs/xfs/quota/xfs_qm_bhv.c b/fs/xfs/quota/xfs_qm_bhv.c index 97bb3293758..f4f6c4c861d 100644 --- a/fs/xfs/quota/xfs_qm_bhv.c +++ b/fs/xfs/quota/xfs_qm_bhv.c @@ -118,7 +118,7 @@ xfs_qm_newmount( *quotaflags = 0; *needquotamount = B_FALSE; - quotaondisk = XFS_SB_VERSION_HASQUOTA(&mp->m_sb) && + quotaondisk = xfs_sb_version_hasquota(&mp->m_sb) && (mp->m_sb.sb_qflags & XFS_ALL_QUOTA_ACCT); if (quotaondisk) { diff --git a/fs/xfs/quota/xfs_qm_syscalls.c b/fs/xfs/quota/xfs_qm_syscalls.c index 2cc5886cfe8..d2b8be7e75f 100644 --- a/fs/xfs/quota/xfs_qm_syscalls.c +++ b/fs/xfs/quota/xfs_qm_syscalls.c @@ -377,7 +377,7 @@ xfs_qm_scall_trunc_qfiles( if (!capable(CAP_SYS_ADMIN)) return XFS_ERROR(EPERM); error = 0; - if (!XFS_SB_VERSION_HASQUOTA(&mp->m_sb) || flags == 0) { + if (!xfs_sb_version_hasquota(&mp->m_sb) || flags == 0) { qdprintk("qtrunc flags=%x m_qflags=%x\n", flags, mp->m_qflags); return XFS_ERROR(EINVAL); } @@ -522,7 +522,7 @@ xfs_qm_scall_getqstat( memset(out, 0, sizeof(fs_quota_stat_t)); out->qs_version = FS_QSTAT_VERSION; - if (! XFS_SB_VERSION_HASQUOTA(&mp->m_sb)) { + if (!xfs_sb_version_hasquota(&mp->m_sb)) { out->qs_uquota.qfs_ino = NULLFSINO; out->qs_gquota.qfs_ino = NULLFSINO; return (0); diff --git a/fs/xfs/xfs_attr_leaf.c b/fs/xfs/xfs_attr_leaf.c index b08e2a2a8ad..96ba6aa4ed8 100644 --- a/fs/xfs/xfs_attr_leaf.c +++ b/fs/xfs/xfs_attr_leaf.c @@ -227,10 +227,10 @@ STATIC void xfs_sbversion_add_attr2(xfs_mount_t *mp, xfs_trans_t *tp) { if ((mp->m_flags & XFS_MOUNT_ATTR2) && - !(XFS_SB_VERSION_HASATTR2(&mp->m_sb))) { + !(xfs_sb_version_hasattr2(&mp->m_sb))) { spin_lock(&mp->m_sb_lock); - if (!XFS_SB_VERSION_HASATTR2(&mp->m_sb)) { - XFS_SB_VERSION_ADDATTR2(&mp->m_sb); + if (!xfs_sb_version_hasattr2(&mp->m_sb)) { + xfs_sb_version_addattr2(&mp->m_sb); spin_unlock(&mp->m_sb_lock); xfs_mod_sb(tp, XFS_SB_VERSIONNUM | XFS_SB_FEATURES2); } else diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c index 1c0a5a585a8..2def273855a 100644 --- a/fs/xfs/xfs_bmap.c +++ b/fs/xfs/xfs_bmap.c @@ -4047,17 +4047,17 @@ xfs_bmap_add_attrfork( xfs_trans_log_inode(tp, ip, logflags); if (error) goto error2; - if (!XFS_SB_VERSION_HASATTR(&mp->m_sb) || - (!XFS_SB_VERSION_HASATTR2(&mp->m_sb) && version == 2)) { + if (!xfs_sb_version_hasattr(&mp->m_sb) || + (!xfs_sb_version_hasattr2(&mp->m_sb) && version == 2)) { __int64_t sbfields = 0; spin_lock(&mp->m_sb_lock); - if (!XFS_SB_VERSION_HASATTR(&mp->m_sb)) { - XFS_SB_VERSION_ADDATTR(&mp->m_sb); + if (!xfs_sb_version_hasattr(&mp->m_sb)) { + xfs_sb_version_addattr(&mp->m_sb); sbfields |= XFS_SB_VERSIONNUM; } - if (!XFS_SB_VERSION_HASATTR2(&mp->m_sb) && version == 2) { - XFS_SB_VERSION_ADDATTR2(&mp->m_sb); + if (!xfs_sb_version_hasattr2(&mp->m_sb) && version == 2) { + xfs_sb_version_addattr2(&mp->m_sb); sbfields |= (XFS_SB_VERSIONNUM | XFS_SB_FEATURES2); } if (sbfields) { @@ -5043,7 +5043,7 @@ xfs_bmapi( * A wasdelay extent has been initialized, so * shouldn't be flagged as unwritten. */ - if (wr && XFS_SB_VERSION_HASEXTFLGBIT(&mp->m_sb)) { + if (wr && xfs_sb_version_hasextflgbit(&mp->m_sb)) { if (!wasdelay && (flags & XFS_BMAPI_PREALLOC)) got.br_state = XFS_EXT_UNWRITTEN; } @@ -5483,7 +5483,7 @@ xfs_bunmapi( * get rid of part of a realtime extent. */ if (del.br_state == XFS_EXT_UNWRITTEN || - !XFS_SB_VERSION_HASEXTFLGBIT(&mp->m_sb)) { + !xfs_sb_version_hasextflgbit(&mp->m_sb)) { /* * This piece is unwritten, or we're not * using unwritten extents. Skip over it. @@ -5535,7 +5535,7 @@ xfs_bunmapi( } else if ((del.br_startoff == start && (del.br_state == XFS_EXT_UNWRITTEN || xfs_trans_get_block_res(tp) == 0)) || - !XFS_SB_VERSION_HASEXTFLGBIT(&mp->m_sb)) { + !xfs_sb_version_hasextflgbit(&mp->m_sb)) { /* * Can't make it unwritten. There isn't * a full extent here so just skip it. diff --git a/fs/xfs/xfs_bmap_btree.h b/fs/xfs/xfs_bmap_btree.h index 2d950e97591..cd0d4b4bb81 100644 --- a/fs/xfs/xfs_bmap_btree.h +++ b/fs/xfs/xfs_bmap_btree.h @@ -120,7 +120,7 @@ typedef enum { * Extent state and extent format macros. */ #define XFS_EXTFMT_INODE(x) \ - (XFS_SB_VERSION_HASEXTFLGBIT(&((x)->i_mount->m_sb)) ? \ + (xfs_sb_version_hasextflgbit(&((x)->i_mount->m_sb)) ? \ XFS_EXTFMT_HASSTATE : XFS_EXTFMT_NOSTATE) #define ISUNWRITTEN(x) ((x)->br_state == XFS_EXT_UNWRITTEN) diff --git a/fs/xfs/xfs_dir2.c b/fs/xfs/xfs_dir2.c index be7c4251fa6..e92e73f0e6a 100644 --- a/fs/xfs/xfs_dir2.c +++ b/fs/xfs/xfs_dir2.c @@ -49,7 +49,7 @@ void xfs_dir_mount( xfs_mount_t *mp) { - ASSERT(XFS_SB_VERSION_HASDIRV2(&mp->m_sb)); + ASSERT(xfs_sb_version_hasdirv2(&mp->m_sb)); ASSERT((1 << (mp->m_sb.sb_blocklog + mp->m_sb.sb_dirblklog)) <= XFS_MAX_BLOCKSIZE); mp->m_dirblksize = 1 << (mp->m_sb.sb_blocklog + mp->m_sb.sb_dirblklog); diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c index eadc1591c79..d3a0f538d6a 100644 --- a/fs/xfs/xfs_fsops.c +++ b/fs/xfs/xfs_fsops.c @@ -77,36 +77,36 @@ xfs_fs_geometry( if (new_version >= 3) { geo->version = XFS_FSOP_GEOM_VERSION; geo->flags = - (XFS_SB_VERSION_HASATTR(&mp->m_sb) ? + (xfs_sb_version_hasattr(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_ATTR : 0) | - (XFS_SB_VERSION_HASNLINK(&mp->m_sb) ? + (xfs_sb_version_hasnlink(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_NLINK : 0) | - (XFS_SB_VERSION_HASQUOTA(&mp->m_sb) ? + (xfs_sb_version_hasquota(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_QUOTA : 0) | - (XFS_SB_VERSION_HASALIGN(&mp->m_sb) ? + (xfs_sb_version_hasalign(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_IALIGN : 0) | - (XFS_SB_VERSION_HASDALIGN(&mp->m_sb) ? + (xfs_sb_version_hasdalign(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_DALIGN : 0) | - (XFS_SB_VERSION_HASSHARED(&mp->m_sb) ? + (xfs_sb_version_hasshared(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_SHARED : 0) | - (XFS_SB_VERSION_HASEXTFLGBIT(&mp->m_sb) ? + (xfs_sb_version_hasextflgbit(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_EXTFLG : 0) | - (XFS_SB_VERSION_HASDIRV2(&mp->m_sb) ? + (xfs_sb_version_hasdirv2(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_DIRV2 : 0) | - (XFS_SB_VERSION_HASSECTOR(&mp->m_sb) ? + (xfs_sb_version_hassector(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_SECTOR : 0) | (xfs_sb_version_haslazysbcount(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_LAZYSB : 0) | - (XFS_SB_VERSION_HASATTR2(&mp->m_sb) ? + (xfs_sb_version_hasattr2(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_ATTR2 : 0); - geo->logsectsize = XFS_SB_VERSION_HASSECTOR(&mp->m_sb) ? + geo->logsectsize = xfs_sb_version_hassector(&mp->m_sb) ? mp->m_sb.sb_logsectsize : BBSIZE; geo->rtsectsize = mp->m_sb.sb_blocksize; geo->dirblocksize = mp->m_dirblksize; } if (new_version >= 4) { geo->flags |= - (XFS_SB_VERSION_HASLOGV2(&mp->m_sb) ? + (xfs_sb_version_haslogv2(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_LOGV2 : 0); geo->logsunit = mp->m_sb.sb_logsunit; } diff --git a/fs/xfs/xfs_ialloc.c b/fs/xfs/xfs_ialloc.c index db9d5fa600a..5a146cb2298 100644 --- a/fs/xfs/xfs_ialloc.c +++ b/fs/xfs/xfs_ialloc.c @@ -191,7 +191,7 @@ xfs_ialloc_ag_alloc( ASSERT(!(args.mp->m_flags & XFS_MOUNT_NOALIGN)); args.alignment = args.mp->m_dalign; isaligned = 1; - } else if (XFS_SB_VERSION_HASALIGN(&args.mp->m_sb) && + } else if (xfs_sb_version_hasalign(&args.mp->m_sb) && args.mp->m_sb.sb_inoalignmt >= XFS_B_TO_FSBT(args.mp, XFS_INODE_CLUSTER_SIZE(args.mp))) @@ -230,7 +230,7 @@ xfs_ialloc_ag_alloc( args.agbno = be32_to_cpu(agi->agi_root); args.fsbno = XFS_AGB_TO_FSB(args.mp, be32_to_cpu(agi->agi_seqno), args.agbno); - if (XFS_SB_VERSION_HASALIGN(&args.mp->m_sb) && + if (xfs_sb_version_hasalign(&args.mp->m_sb) && args.mp->m_sb.sb_inoalignmt >= XFS_B_TO_FSBT(args.mp, XFS_INODE_CLUSTER_SIZE(args.mp))) args.alignment = args.mp->m_sb.sb_inoalignmt; @@ -271,7 +271,7 @@ xfs_ialloc_ag_alloc( * use the old version so that old kernels will continue to be * able to use the file system. */ - if (XFS_SB_VERSION_HASNLINK(&args.mp->m_sb)) + if (xfs_sb_version_hasnlink(&args.mp->m_sb)) version = XFS_DINODE_VERSION_2; else version = XFS_DINODE_VERSION_1; diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index a550546a708..f43a6e01d68 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -1147,7 +1147,7 @@ xfs_ialloc( * the inode version number now. This way we only do the conversion * here rather than here and in the flush/logging code. */ - if (XFS_SB_VERSION_HASNLINK(&tp->t_mountp->m_sb) && + if (xfs_sb_version_hasnlink(&tp->t_mountp->m_sb) && ip->i_d.di_version == XFS_DINODE_VERSION_1) { ip->i_d.di_version = XFS_DINODE_VERSION_2; /* @@ -3434,9 +3434,9 @@ xfs_iflush_int( * has been updated, then make the conversion permanent. */ ASSERT(ip->i_d.di_version == XFS_DINODE_VERSION_1 || - XFS_SB_VERSION_HASNLINK(&mp->m_sb)); + xfs_sb_version_hasnlink(&mp->m_sb)); if (ip->i_d.di_version == XFS_DINODE_VERSION_1) { - if (!XFS_SB_VERSION_HASNLINK(&mp->m_sb)) { + if (!xfs_sb_version_hasnlink(&mp->m_sb)) { /* * Convert it back. */ diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c index 034ca720229..2c775b4ae9e 100644 --- a/fs/xfs/xfs_inode_item.c +++ b/fs/xfs/xfs_inode_item.c @@ -296,9 +296,9 @@ xfs_inode_item_format( */ mp = ip->i_mount; ASSERT(ip->i_d.di_version == XFS_DINODE_VERSION_1 || - XFS_SB_VERSION_HASNLINK(&mp->m_sb)); + xfs_sb_version_hasnlink(&mp->m_sb)); if (ip->i_d.di_version == XFS_DINODE_VERSION_1) { - if (!XFS_SB_VERSION_HASNLINK(&mp->m_sb)) { + if (!xfs_sb_version_hasnlink(&mp->m_sb)) { /* * Convert it back. */ diff --git a/fs/xfs/xfs_itable.c b/fs/xfs/xfs_itable.c index 658aab6b1bb..f615e04364f 100644 --- a/fs/xfs/xfs_itable.c +++ b/fs/xfs/xfs_itable.c @@ -45,7 +45,7 @@ xfs_internal_inum( xfs_ino_t ino) { return (ino == mp->m_sb.sb_rbmino || ino == mp->m_sb.sb_rsumino || - (XFS_SB_VERSION_HASQUOTA(&mp->m_sb) && + (xfs_sb_version_hasquota(&mp->m_sb) && (ino == mp->m_sb.sb_uquotino || ino == mp->m_sb.sb_gquotino))); } diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c index a75edca1860..31f2b04f2c9 100644 --- a/fs/xfs/xfs_log.c +++ b/fs/xfs/xfs_log.c @@ -1090,7 +1090,7 @@ xlog_get_iclog_buffer_size(xfs_mount_t *mp, size >>= 1; } - if (XFS_SB_VERSION_HASLOGV2(&mp->m_sb)) { + if (xfs_sb_version_haslogv2(&mp->m_sb)) { /* # headers = size / 32K * one header holds cycles from 32K of data */ @@ -1186,13 +1186,13 @@ xlog_alloc_log(xfs_mount_t *mp, log->l_grant_reserve_cycle = 1; log->l_grant_write_cycle = 1; - if (XFS_SB_VERSION_HASSECTOR(&mp->m_sb)) { + if (xfs_sb_version_hassector(&mp->m_sb)) { log->l_sectbb_log = mp->m_sb.sb_logsectlog - BBSHIFT; ASSERT(log->l_sectbb_log <= mp->m_sectbb_log); /* for larger sector sizes, must have v2 or external log */ ASSERT(log->l_sectbb_log == 0 || log->l_logBBstart == 0 || - XFS_SB_VERSION_HASLOGV2(&mp->m_sb)); + xfs_sb_version_haslogv2(&mp->m_sb)); ASSERT(mp->m_sb.sb_logsectlog >= BBSHIFT); } log->l_sectbb_mask = (1 << log->l_sectbb_log) - 1; @@ -1247,7 +1247,7 @@ xlog_alloc_log(xfs_mount_t *mp, memset(head, 0, sizeof(xlog_rec_header_t)); head->h_magicno = cpu_to_be32(XLOG_HEADER_MAGIC_NUM); head->h_version = cpu_to_be32( - XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb) ? 2 : 1); + xfs_sb_version_haslogv2(&log->l_mp->m_sb) ? 2 : 1); head->h_size = cpu_to_be32(log->l_iclog_size); /* new fields */ head->h_fmt = cpu_to_be32(XLOG_FMT); @@ -1402,7 +1402,7 @@ xlog_sync(xlog_t *log, int roundoff; /* roundoff to BB or stripe */ int split = 0; /* split write into two regions */ int error; - int v2 = XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb); + int v2 = xfs_sb_version_haslogv2(&log->l_mp->m_sb); XFS_STATS_INC(xs_log_writes); ASSERT(iclog->ic_refcnt == 0); @@ -2881,7 +2881,7 @@ xlog_state_switch_iclogs(xlog_t *log, log->l_curr_block += BTOBB(eventual_size)+BTOBB(log->l_iclog_hsize); /* Round up to next log-sunit */ - if (XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb) && + if (xfs_sb_version_haslogv2(&log->l_mp->m_sb) && log->l_mp->m_sb.sb_logsunit > 1) { __uint32_t sunit_bb = BTOBB(log->l_mp->m_sb.sb_logsunit); log->l_curr_block = roundup(log->l_curr_block, sunit_bb); @@ -3334,7 +3334,7 @@ xlog_ticket_get(xlog_t *log, unit_bytes += sizeof(xlog_op_header_t) * num_headers; /* for roundoff padding for transaction data and one for commit record */ - if (XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb) && + if (xfs_sb_version_haslogv2(&log->l_mp->m_sb) && log->l_mp->m_sb.sb_logsunit > 1) { /* log su roundoff */ unit_bytes += 2*log->l_mp->m_sb.sb_logsunit; diff --git a/fs/xfs/xfs_log_priv.h b/fs/xfs/xfs_log_priv.h index e008233ee24..c6244cc733c 100644 --- a/fs/xfs/xfs_log_priv.h +++ b/fs/xfs/xfs_log_priv.h @@ -49,10 +49,10 @@ struct xfs_mount; #define XLOG_HEADER_SIZE 512 #define XLOG_REC_SHIFT(log) \ - BTOBB(1 << (XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb) ? \ + BTOBB(1 << (xfs_sb_version_haslogv2(&log->l_mp->m_sb) ? \ XLOG_MAX_RECORD_BSHIFT : XLOG_BIG_RECORD_BSHIFT)) #define XLOG_TOTAL_REC_SHIFT(log) \ - BTOBB(XLOG_MAX_ICLOGS << (XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb) ? \ + BTOBB(XLOG_MAX_ICLOGS << (xfs_sb_version_haslogv2(&log->l_mp->m_sb) ? \ XLOG_MAX_RECORD_BSHIFT : XLOG_BIG_RECORD_BSHIFT)) diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c index b82d5d4d246..b2b70eba282 100644 --- a/fs/xfs/xfs_log_recover.c +++ b/fs/xfs/xfs_log_recover.c @@ -478,7 +478,7 @@ xlog_find_verify_log_record( * reset last_blk. Only when last_blk points in the middle of a log * record do we update last_blk. */ - if (XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb)) { + if (xfs_sb_version_haslogv2(&log->l_mp->m_sb)) { uint h_size = be32_to_cpu(head->h_size); xhdrs = h_size / XLOG_HEADER_CYCLE_SIZE; @@ -888,7 +888,7 @@ xlog_find_tail( * unmount record if there is one, so we pass the lsn of the * unmount record rather than the block after it. */ - if (XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb)) { + if (xfs_sb_version_haslogv2(&log->l_mp->m_sb)) { int h_size = be32_to_cpu(rhead->h_size); int h_version = be32_to_cpu(rhead->h_version); @@ -1101,7 +1101,7 @@ xlog_add_record( recp->h_magicno = cpu_to_be32(XLOG_HEADER_MAGIC_NUM); recp->h_cycle = cpu_to_be32(cycle); recp->h_version = cpu_to_be32( - XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb) ? 2 : 1); + xfs_sb_version_haslogv2(&log->l_mp->m_sb) ? 2 : 1); recp->h_lsn = cpu_to_be64(xlog_assign_lsn(cycle, block)); recp->h_tail_lsn = cpu_to_be64(xlog_assign_lsn(tail_cycle, tail_block)); recp->h_fmt = cpu_to_be32(XLOG_FMT); @@ -3348,7 +3348,7 @@ xlog_pack_data( dp += BBSIZE; } - if (XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb)) { + if (xfs_sb_version_haslogv2(&log->l_mp->m_sb)) { xhdr = (xlog_in_core_2_t *)&iclog->ic_header; for ( ; i < BTOBB(size); i++) { j = i / (XLOG_HEADER_CYCLE_SIZE / BBSIZE); @@ -3388,7 +3388,7 @@ xlog_unpack_data_checksum( be32_to_cpu(rhead->h_chksum), chksum); cmn_err(CE_DEBUG, "XFS: Disregard message if filesystem was created with non-DEBUG kernel"); - if (XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb)) { + if (xfs_sb_version_haslogv2(&log->l_mp->m_sb)) { cmn_err(CE_DEBUG, "XFS: LogR this is a LogV2 filesystem\n"); } @@ -3415,7 +3415,7 @@ xlog_unpack_data( dp += BBSIZE; } - if (XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb)) { + if (xfs_sb_version_haslogv2(&log->l_mp->m_sb)) { xhdr = (xlog_in_core_2_t *)rhead; for ( ; i < BTOBB(be32_to_cpu(rhead->h_len)); i++) { j = i / (XLOG_HEADER_CYCLE_SIZE / BBSIZE); @@ -3494,7 +3494,7 @@ xlog_do_recovery_pass( * Read the header of the tail block and get the iclog buffer size from * h_size. Use this to tell how many sectors make up the log header. */ - if (XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb)) { + if (xfs_sb_version_haslogv2(&log->l_mp->m_sb)) { /* * When using variable length iclogs, read first sector of * iclog header and extract the header size from it. Get a @@ -3838,7 +3838,7 @@ xlog_do_recover( sbp = &log->l_mp->m_sb; xfs_sb_from_disk(sbp, XFS_BUF_TO_SBP(bp)); ASSERT(sbp->sb_magicnum == XFS_SB_MAGIC); - ASSERT(XFS_SB_GOOD_VERSION(sbp)); + ASSERT(xfs_sb_good_version(sbp)); xfs_buf_relse(bp); /* We've re-read the superblock so re-initialize per-cpu counters */ diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index 6409b376299..99bab1e372b 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -225,7 +225,7 @@ xfs_mount_validate_sb( return XFS_ERROR(EWRONGFS); } - if (!XFS_SB_GOOD_VERSION(sbp)) { + if (!xfs_sb_good_version(sbp)) { xfs_fs_mount_cmn_err(flags, "bad version"); return XFS_ERROR(EWRONGFS); } @@ -300,7 +300,7 @@ xfs_mount_validate_sb( /* * Version 1 directory format has never worked on Linux. */ - if (unlikely(!XFS_SB_VERSION_HASDIRV2(sbp))) { + if (unlikely(!xfs_sb_version_hasdirv2(sbp))) { xfs_fs_mount_cmn_err(flags, "file system using version 1 directory format"); return XFS_ERROR(ENOSYS); @@ -781,7 +781,7 @@ xfs_update_alignment(xfs_mount_t *mp, int mfsi_flags, __uint64_t *update_flags) * Update superblock with new values * and log changes */ - if (XFS_SB_VERSION_HASDALIGN(sbp)) { + if (xfs_sb_version_hasdalign(sbp)) { if (sbp->sb_unit != mp->m_dalign) { sbp->sb_unit = mp->m_dalign; *update_flags |= XFS_SB_UNIT; @@ -792,7 +792,7 @@ xfs_update_alignment(xfs_mount_t *mp, int mfsi_flags, __uint64_t *update_flags) } } } else if ((mp->m_flags & XFS_MOUNT_NOALIGN) != XFS_MOUNT_NOALIGN && - XFS_SB_VERSION_HASDALIGN(&mp->m_sb)) { + xfs_sb_version_hasdalign(&mp->m_sb)) { mp->m_dalign = sbp->sb_unit; mp->m_swidth = sbp->sb_width; } @@ -869,7 +869,7 @@ xfs_set_rw_sizes(xfs_mount_t *mp) STATIC void xfs_set_inoalignment(xfs_mount_t *mp) { - if (XFS_SB_VERSION_HASALIGN(&mp->m_sb) && + if (xfs_sb_version_hasalign(&mp->m_sb) && mp->m_sb.sb_inoalignmt >= XFS_B_TO_FSBT(mp, mp->m_inode_cluster_size)) mp->m_inoalign_mask = mp->m_sb.sb_inoalignmt - 1; diff --git a/fs/xfs/xfs_sb.h b/fs/xfs/xfs_sb.h index 94660b1a6cc..d3a59735009 100644 --- a/fs/xfs/xfs_sb.h +++ b/fs/xfs/xfs_sb.h @@ -271,7 +271,6 @@ typedef enum { #define XFS_SB_VERSION_NUM(sbp) ((sbp)->sb_versionnum & XFS_SB_VERSION_NUMBITS) -#define XFS_SB_GOOD_VERSION(sbp) xfs_sb_good_version(sbp) #ifdef __KERNEL__ static inline int xfs_sb_good_version(xfs_sb_t *sbp) { @@ -297,7 +296,6 @@ static inline int xfs_sb_good_version(xfs_sb_t *sbp) } #endif /* __KERNEL__ */ -#define XFS_SB_VERSION_TONEW(v) xfs_sb_version_tonew(v) static inline unsigned xfs_sb_version_tonew(unsigned v) { return ((((v) == XFS_SB_VERSION_1) ? \ @@ -308,7 +306,6 @@ static inline unsigned xfs_sb_version_tonew(unsigned v) XFS_SB_VERSION_4); } -#define XFS_SB_VERSION_TOOLD(v) xfs_sb_version_toold(v) static inline unsigned xfs_sb_version_toold(unsigned v) { return (((v) & (XFS_SB_VERSION_QUOTABIT | XFS_SB_VERSION_ALIGNBIT)) ? \ @@ -320,7 +317,6 @@ static inline unsigned xfs_sb_version_toold(unsigned v) XFS_SB_VERSION_1))); } -#define XFS_SB_VERSION_HASATTR(sbp) xfs_sb_version_hasattr(sbp) static inline int xfs_sb_version_hasattr(xfs_sb_t *sbp) { return ((sbp)->sb_versionnum == XFS_SB_VERSION_2) || \ @@ -329,7 +325,6 @@ static inline int xfs_sb_version_hasattr(xfs_sb_t *sbp) ((sbp)->sb_versionnum & XFS_SB_VERSION_ATTRBIT)); } -#define XFS_SB_VERSION_ADDATTR(sbp) xfs_sb_version_addattr(sbp) static inline void xfs_sb_version_addattr(xfs_sb_t *sbp) { (sbp)->sb_versionnum = (((sbp)->sb_versionnum == XFS_SB_VERSION_1) ? \ @@ -339,7 +334,6 @@ static inline void xfs_sb_version_addattr(xfs_sb_t *sbp) (XFS_SB_VERSION_4 | XFS_SB_VERSION_ATTRBIT))); } -#define XFS_SB_VERSION_HASNLINK(sbp) xfs_sb_version_hasnlink(sbp) static inline int xfs_sb_version_hasnlink(xfs_sb_t *sbp) { return ((sbp)->sb_versionnum == XFS_SB_VERSION_3) || \ @@ -347,7 +341,6 @@ static inline int xfs_sb_version_hasnlink(xfs_sb_t *sbp) ((sbp)->sb_versionnum & XFS_SB_VERSION_NLINKBIT)); } -#define XFS_SB_VERSION_ADDNLINK(sbp) xfs_sb_version_addnlink(sbp) static inline void xfs_sb_version_addnlink(xfs_sb_t *sbp) { (sbp)->sb_versionnum = ((sbp)->sb_versionnum <= XFS_SB_VERSION_2 ? \ @@ -355,115 +348,63 @@ static inline void xfs_sb_version_addnlink(xfs_sb_t *sbp) ((sbp)->sb_versionnum | XFS_SB_VERSION_NLINKBIT)); } -#define XFS_SB_VERSION_HASQUOTA(sbp) xfs_sb_version_hasquota(sbp) static inline int xfs_sb_version_hasquota(xfs_sb_t *sbp) { return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ ((sbp)->sb_versionnum & XFS_SB_VERSION_QUOTABIT); } -#define XFS_SB_VERSION_ADDQUOTA(sbp) xfs_sb_version_addquota(sbp) static inline void xfs_sb_version_addquota(xfs_sb_t *sbp) { (sbp)->sb_versionnum = \ (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4 ? \ ((sbp)->sb_versionnum | XFS_SB_VERSION_QUOTABIT) : \ - (XFS_SB_VERSION_TONEW((sbp)->sb_versionnum) | \ + (xfs_sb_version_tonew((sbp)->sb_versionnum) | \ XFS_SB_VERSION_QUOTABIT)); } -#define XFS_SB_VERSION_HASALIGN(sbp) xfs_sb_version_hasalign(sbp) static inline int xfs_sb_version_hasalign(xfs_sb_t *sbp) { return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ ((sbp)->sb_versionnum & XFS_SB_VERSION_ALIGNBIT); } -#define XFS_SB_VERSION_SUBALIGN(sbp) xfs_sb_version_subalign(sbp) -static inline void xfs_sb_version_subalign(xfs_sb_t *sbp) -{ - (sbp)->sb_versionnum = \ - XFS_SB_VERSION_TOOLD((sbp)->sb_versionnum & ~XFS_SB_VERSION_ALIGNBIT); -} - -#define XFS_SB_VERSION_HASDALIGN(sbp) xfs_sb_version_hasdalign(sbp) static inline int xfs_sb_version_hasdalign(xfs_sb_t *sbp) { return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ ((sbp)->sb_versionnum & XFS_SB_VERSION_DALIGNBIT); } -#define XFS_SB_VERSION_ADDDALIGN(sbp) xfs_sb_version_adddalign(sbp) -static inline int xfs_sb_version_adddalign(xfs_sb_t *sbp) -{ - return (sbp)->sb_versionnum = \ - ((sbp)->sb_versionnum | XFS_SB_VERSION_DALIGNBIT); -} - -#define XFS_SB_VERSION_HASSHARED(sbp) xfs_sb_version_hasshared(sbp) static inline int xfs_sb_version_hasshared(xfs_sb_t *sbp) { return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ ((sbp)->sb_versionnum & XFS_SB_VERSION_SHAREDBIT); } -#define XFS_SB_VERSION_ADDSHARED(sbp) xfs_sb_version_addshared(sbp) -static inline int xfs_sb_version_addshared(xfs_sb_t *sbp) -{ - return (sbp)->sb_versionnum = \ - ((sbp)->sb_versionnum | XFS_SB_VERSION_SHAREDBIT); -} - -#define XFS_SB_VERSION_SUBSHARED(sbp) xfs_sb_version_subshared(sbp) -static inline int xfs_sb_version_subshared(xfs_sb_t *sbp) -{ - return (sbp)->sb_versionnum = \ - ((sbp)->sb_versionnum & ~XFS_SB_VERSION_SHAREDBIT); -} - -#define XFS_SB_VERSION_HASDIRV2(sbp) xfs_sb_version_hasdirv2(sbp) static inline int xfs_sb_version_hasdirv2(xfs_sb_t *sbp) { return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ ((sbp)->sb_versionnum & XFS_SB_VERSION_DIRV2BIT); } -#define XFS_SB_VERSION_HASLOGV2(sbp) xfs_sb_version_haslogv2(sbp) static inline int xfs_sb_version_haslogv2(xfs_sb_t *sbp) { return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ ((sbp)->sb_versionnum & XFS_SB_VERSION_LOGV2BIT); } -#define XFS_SB_VERSION_HASEXTFLGBIT(sbp) xfs_sb_version_hasextflgbit(sbp) static inline int xfs_sb_version_hasextflgbit(xfs_sb_t *sbp) { return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ ((sbp)->sb_versionnum & XFS_SB_VERSION_EXTFLGBIT); } -#define XFS_SB_VERSION_ADDEXTFLGBIT(sbp) xfs_sb_version_addextflgbit(sbp) -static inline int xfs_sb_version_addextflgbit(xfs_sb_t *sbp) -{ - return (sbp)->sb_versionnum = \ - ((sbp)->sb_versionnum | XFS_SB_VERSION_EXTFLGBIT); -} - -#define XFS_SB_VERSION_SUBEXTFLGBIT(sbp) xfs_sb_version_subextflgbit(sbp) -static inline int xfs_sb_version_subextflgbit(xfs_sb_t *sbp) -{ - return (sbp)->sb_versionnum = \ - ((sbp)->sb_versionnum & ~XFS_SB_VERSION_EXTFLGBIT); -} - -#define XFS_SB_VERSION_HASSECTOR(sbp) xfs_sb_version_hassector(sbp) static inline int xfs_sb_version_hassector(xfs_sb_t *sbp) { return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ ((sbp)->sb_versionnum & XFS_SB_VERSION_SECTORBIT); } -#define XFS_SB_VERSION_HASMOREBITS(sbp) xfs_sb_version_hasmorebits(sbp) static inline int xfs_sb_version_hasmorebits(xfs_sb_t *sbp) { return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ @@ -476,24 +417,22 @@ static inline int xfs_sb_version_hasmorebits(xfs_sb_t *sbp) * For example, for a bit defined as XFS_SB_VERSION2_FUNBIT, has a macro: * * SB_VERSION_HASFUNBIT(xfs_sb_t *sbp) - * ((XFS_SB_VERSION_HASMOREBITS(sbp) && + * ((xfs_sb_version_hasmorebits(sbp) && * ((sbp)->sb_features2 & XFS_SB_VERSION2_FUNBIT) */ static inline int xfs_sb_version_haslazysbcount(xfs_sb_t *sbp) { - return (XFS_SB_VERSION_HASMOREBITS(sbp) && \ + return (xfs_sb_version_hasmorebits(sbp) && \ ((sbp)->sb_features2 & XFS_SB_VERSION2_LAZYSBCOUNTBIT)); } -#define XFS_SB_VERSION_HASATTR2(sbp) xfs_sb_version_hasattr2(sbp) static inline int xfs_sb_version_hasattr2(xfs_sb_t *sbp) { - return (XFS_SB_VERSION_HASMOREBITS(sbp)) && \ + return (xfs_sb_version_hasmorebits(sbp)) && \ ((sbp)->sb_features2 & XFS_SB_VERSION2_ATTR2BIT); } -#define XFS_SB_VERSION_ADDATTR2(sbp) xfs_sb_version_addattr2(sbp) static inline void xfs_sb_version_addattr2(xfs_sb_t *sbp) { ((sbp)->sb_versionnum = \ diff --git a/fs/xfs/xfs_utils.c b/fs/xfs/xfs_utils.c index 45d740df53b..18a85e74668 100644 --- a/fs/xfs/xfs_utils.c +++ b/fs/xfs/xfs_utils.c @@ -339,10 +339,10 @@ xfs_bump_ino_vers2( ip->i_d.di_onlink = 0; memset(&(ip->i_d.di_pad[0]), 0, sizeof(ip->i_d.di_pad)); mp = tp->t_mountp; - if (!XFS_SB_VERSION_HASNLINK(&mp->m_sb)) { + if (!xfs_sb_version_hasnlink(&mp->m_sb)) { spin_lock(&mp->m_sb_lock); - if (!XFS_SB_VERSION_HASNLINK(&mp->m_sb)) { - XFS_SB_VERSION_ADDNLINK(&mp->m_sb); + if (!xfs_sb_version_hasnlink(&mp->m_sb)) { + xfs_sb_version_addnlink(&mp->m_sb); spin_unlock(&mp->m_sb_lock); xfs_mod_sb(tp, XFS_SB_VERSIONNUM); } else { diff --git a/fs/xfs/xfs_vfsops.c b/fs/xfs/xfs_vfsops.c index 7321304a69c..7094caff13c 100644 --- a/fs/xfs/xfs_vfsops.c +++ b/fs/xfs/xfs_vfsops.c @@ -330,7 +330,7 @@ xfs_finish_flags( int ronly = (mp->m_flags & XFS_MOUNT_RDONLY); /* Fail a mount where the logbuf is smaller then the log stripe */ - if (XFS_SB_VERSION_HASLOGV2(&mp->m_sb)) { + if (xfs_sb_version_haslogv2(&mp->m_sb)) { if ((ap->logbufsize <= 0) && (mp->m_sb.sb_logsunit > XLOG_BIG_RECORD_BSIZE)) { mp->m_logbsize = mp->m_sb.sb_logsunit; @@ -349,9 +349,8 @@ xfs_finish_flags( } } - if (XFS_SB_VERSION_HASATTR2(&mp->m_sb)) { + if (xfs_sb_version_hasattr2(&mp->m_sb)) mp->m_flags |= XFS_MOUNT_ATTR2; - } /* * prohibit r/w mounts of read-only filesystems @@ -366,7 +365,7 @@ xfs_finish_flags( * check for shared mount. */ if (ap->flags & XFSMNT_SHARED) { - if (!XFS_SB_VERSION_HASSHARED(&mp->m_sb)) + if (!xfs_sb_version_hasshared(&mp->m_sb)) return XFS_ERROR(EINVAL); /* @@ -512,7 +511,7 @@ xfs_mount( if (!error && logdev && logdev != ddev) { unsigned int log_sector_size = BBSIZE; - if (XFS_SB_VERSION_HASSECTOR(&mp->m_sb)) + if (xfs_sb_version_hassector(&mp->m_sb)) log_sector_size = mp->m_sb.sb_logsectsize; error = xfs_setsize_buftarg(mp->m_logdev_targp, mp->m_sb.sb_blocksize, diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c index 51305242ff8..64c5953feca 100644 --- a/fs/xfs/xfs_vnodeops.c +++ b/fs/xfs/xfs_vnodeops.c @@ -4132,7 +4132,7 @@ xfs_free_file_space( * actually need to zero the extent edges. Otherwise xfs_bunmapi * will take care of it for us. */ - if (rt && !XFS_SB_VERSION_HASEXTFLGBIT(&mp->m_sb)) { + if (rt && !xfs_sb_version_hasextflgbit(&mp->m_sb)) { nimap = 1; error = xfs_bmapi(NULL, ip, startoffset_fsb, 1, 0, NULL, 0, &imap, &nimap, NULL, NULL); -- cgit v1.2.3 From ee1c090825bf2f1cb9712489c234eabf69a9a78f Mon Sep 17 00:00:00 2001 From: David Chinner Date: Thu, 6 Mar 2008 13:45:50 +1100 Subject: [XFS] Fix superblock features2 field alignment problem Due to the xfs_dsb_t structure not being 64 bit aligned, the last field of the on-disk superblock can vary in location This causes problems when the filesystem gets moved to a different platform, or there is a 32 bit userspace and 64 bit kernel. This patch detects the defect at mount time, logs a warning such as: XFS: correcting sb_features alignment problem in dmesg and corrects the problem so that everything is OK. it also blacklists the bad field in the superblock so it does not get used for something else later on. SGI-PV: 977636 SGI-Modid: xfs-linux-melb:xfs-kern:30539a Signed-off-by: David Chinner Signed-off-by: Christoph Hellwig Signed-off-by: Eric Sandeen Signed-off-by: Lachlan McIlroy --- fs/xfs/xfs_mount.c | 34 ++++++++++++++++++++++++++++------ fs/xfs/xfs_sb.h | 37 ++++++++++++++++++++++++++++++++++--- 2 files changed, 62 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index 99bab1e372b..4e93c02faf2 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -44,7 +44,7 @@ #include "xfs_quota.h" #include "xfs_fsops.h" -STATIC void xfs_mount_log_sbunit(xfs_mount_t *, __int64_t); +STATIC void xfs_mount_log_sb(xfs_mount_t *, __int64_t); STATIC int xfs_uuid_mount(xfs_mount_t *); STATIC void xfs_uuid_unmount(xfs_mount_t *mp); STATIC void xfs_unmountfs_wait(xfs_mount_t *); @@ -119,6 +119,7 @@ static const struct { { offsetof(xfs_sb_t, sb_logsectsize),0 }, { offsetof(xfs_sb_t, sb_logsunit), 0 }, { offsetof(xfs_sb_t, sb_features2), 0 }, + { offsetof(xfs_sb_t, sb_bad_features2), 0 }, { sizeof(xfs_sb_t), 0 } }; @@ -449,6 +450,7 @@ xfs_sb_from_disk( to->sb_logsectsize = be16_to_cpu(from->sb_logsectsize); to->sb_logsunit = be32_to_cpu(from->sb_logsunit); to->sb_features2 = be32_to_cpu(from->sb_features2); + to->sb_bad_features2 = be32_to_cpu(from->sb_bad_features2); } /* @@ -969,6 +971,26 @@ xfs_mountfs( } xfs_mount_common(mp, sbp); + /* + * Check for a bad features2 field alignment. This happened on + * some platforms due to xfs_sb_t not being 64bit size aligned + * when sb_features was added and hence the compiler put it in + * the wrong place. + * + * If we detect a bad field, we or the set bits into the existing + * features2 field in case it has already been modified and we + * don't want to lose any features. Zero the bad one and mark + * the two fields as needing updates once the transaction subsystem + * is online. + */ + if (xfs_sb_has_bad_features2(sbp)) { + cmn_err(CE_WARN, + "XFS: correcting sb_features alignment problem"); + sbp->sb_features2 |= sbp->sb_bad_features2; + sbp->sb_bad_features2 = 0; + update_flags |= XFS_SB_FEATURES2 | XFS_SB_BAD_FEATURES2; + } + /* * Check if sb_agblocks is aligned at stripe boundary * If sb_agblocks is NOT aligned turn off m_dalign since @@ -1159,11 +1181,10 @@ xfs_mountfs( } /* - * If fs is not mounted readonly, then update the superblock - * unit and width changes. + * If fs is not mounted readonly, then update the superblock changes. */ if (update_flags && !(mp->m_flags & XFS_MOUNT_RDONLY)) - xfs_mount_log_sbunit(mp, update_flags); + xfs_mount_log_sb(mp, update_flags); /* * Initialise the XFS quota management subsystem for this mount @@ -1878,13 +1899,14 @@ xfs_uuid_unmount( * be altered by the mount options. Only the first superblock is updated. */ STATIC void -xfs_mount_log_sbunit( +xfs_mount_log_sb( xfs_mount_t *mp, __int64_t fields) { xfs_trans_t *tp; - ASSERT(fields & (XFS_SB_UNIT|XFS_SB_WIDTH|XFS_SB_UUID)); + ASSERT(fields & (XFS_SB_UNIT | XFS_SB_WIDTH | XFS_SB_UUID | + XFS_SB_FEATURES2 | XFS_SB_BAD_FEATURES2)); tp = xfs_trans_alloc(mp, XFS_TRANS_SB_UNIT); if (xfs_trans_reserve(tp, 0, mp->m_sb.sb_sectsize + 128, 0, 0, diff --git a/fs/xfs/xfs_sb.h b/fs/xfs/xfs_sb.h index d3a59735009..b1a83f8ec04 100644 --- a/fs/xfs/xfs_sb.h +++ b/fs/xfs/xfs_sb.h @@ -89,6 +89,7 @@ struct xfs_mount; /* * Superblock - in core version. Must match the ondisk version below. + * Must be padded to 64 bit alignment. */ typedef struct xfs_sb { __uint32_t sb_magicnum; /* magic number == XFS_SB_MAGIC */ @@ -145,10 +146,21 @@ typedef struct xfs_sb { __uint16_t sb_logsectsize; /* sector size for the log, bytes */ __uint32_t sb_logsunit; /* stripe unit size for the log */ __uint32_t sb_features2; /* additional feature bits */ + + /* + * bad features2 field as a result of failing to pad the sb + * structure to 64 bits. Some machines will be using this field + * for features2 bits. Easiest just to mark it bad and not use + * it for anything else. + */ + __uint32_t sb_bad_features2; + + /* must be padded to 64 bit alignment */ } xfs_sb_t; /* - * Superblock - on disk version. Must match the in core version below. + * Superblock - on disk version. Must match the in core version above. + * Must be padded to 64 bit alignment. */ typedef struct xfs_dsb { __be32 sb_magicnum; /* magic number == XFS_SB_MAGIC */ @@ -205,6 +217,15 @@ typedef struct xfs_dsb { __be16 sb_logsectsize; /* sector size for the log, bytes */ __be32 sb_logsunit; /* stripe unit size for the log */ __be32 sb_features2; /* additional feature bits */ + /* + * bad features2 field as a result of failing to pad the sb + * structure to 64 bits. Some machines will be using this field + * for features2 bits. Easiest just to mark it bad and not use + * it for anything else. + */ + __be32 sb_bad_features2; + + /* must be padded to 64 bit alignment */ } xfs_dsb_t; /* @@ -223,7 +244,7 @@ typedef enum { XFS_SBS_GQUOTINO, XFS_SBS_QFLAGS, XFS_SBS_FLAGS, XFS_SBS_SHARED_VN, XFS_SBS_INOALIGNMT, XFS_SBS_UNIT, XFS_SBS_WIDTH, XFS_SBS_DIRBLKLOG, XFS_SBS_LOGSECTLOG, XFS_SBS_LOGSECTSIZE, XFS_SBS_LOGSUNIT, - XFS_SBS_FEATURES2, + XFS_SBS_FEATURES2, XFS_SBS_BAD_FEATURES2, XFS_SBS_FIELDCOUNT } xfs_sb_field_t; @@ -248,13 +269,15 @@ typedef enum { #define XFS_SB_IFREE XFS_SB_MVAL(IFREE) #define XFS_SB_FDBLOCKS XFS_SB_MVAL(FDBLOCKS) #define XFS_SB_FEATURES2 XFS_SB_MVAL(FEATURES2) +#define XFS_SB_BAD_FEATURES2 XFS_SB_MVAL(BAD_FEATURES2) #define XFS_SB_NUM_BITS ((int)XFS_SBS_FIELDCOUNT) #define XFS_SB_ALL_BITS ((1LL << XFS_SB_NUM_BITS) - 1) #define XFS_SB_MOD_BITS \ (XFS_SB_UUID | XFS_SB_ROOTINO | XFS_SB_RBMINO | XFS_SB_RSUMINO | \ XFS_SB_VERSIONNUM | XFS_SB_UQUOTINO | XFS_SB_GQUOTINO | \ XFS_SB_QFLAGS | XFS_SB_SHARED_VN | XFS_SB_UNIT | XFS_SB_WIDTH | \ - XFS_SB_ICOUNT | XFS_SB_IFREE | XFS_SB_FDBLOCKS | XFS_SB_FEATURES2) + XFS_SB_ICOUNT | XFS_SB_IFREE | XFS_SB_FDBLOCKS | XFS_SB_FEATURES2 | \ + XFS_SB_BAD_FEATURES2) /* @@ -296,6 +319,14 @@ static inline int xfs_sb_good_version(xfs_sb_t *sbp) } #endif /* __KERNEL__ */ +/* + * Detect a bad features2 field + */ +static inline int xfs_sb_has_bad_features2(xfs_sb_t *sbp) +{ + return (sbp->sb_bad_features2 != 0); +} + static inline unsigned xfs_sb_version_tonew(unsigned v) { return ((((v) == XFS_SB_VERSION_1) ? \ -- cgit v1.2.3 From e6957ea48429aeda4f5d51af4238231d44b0dc4a Mon Sep 17 00:00:00 2001 From: Eric Sandeen Date: Thu, 10 Apr 2008 12:19:34 +1000 Subject: [XFS] Ensure "both" features2 slots are consistent Since older kernels may look in the sb_bad_features2 slot for flags, rather than zeroing it out on fixup, we should make it equal to the sb_features2 value. Also, if the ATTR2 flag was not found prior to features2 fixup, it was not set in the mount flags, so re-check after the fixup so that the current session will use the feature. Also fix up the comments to reflect these changes. SGI-PV: 980085 SGI-Modid: xfs-linux-melb:xfs-kern:30778a Signed-off-by: Eric Sandeen Signed-off-by: David Chinner Signed-off-by: Lachlan McIlroy --- fs/xfs/xfs_mount.c | 37 +++++++++++++++++++++++++------------ fs/xfs/xfs_sb.h | 7 ++++--- 2 files changed, 29 insertions(+), 15 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index 4e93c02faf2..8ed164eb954 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -972,23 +972,35 @@ xfs_mountfs( xfs_mount_common(mp, sbp); /* - * Check for a bad features2 field alignment. This happened on - * some platforms due to xfs_sb_t not being 64bit size aligned - * when sb_features was added and hence the compiler put it in - * the wrong place. + * Check for a mismatched features2 values. Older kernels + * read & wrote into the wrong sb offset for sb_features2 + * on some platforms due to xfs_sb_t not being 64bit size aligned + * when sb_features2 was added, which made older superblock + * reading/writing routines swap it as a 64-bit value. * - * If we detect a bad field, we or the set bits into the existing - * features2 field in case it has already been modified and we - * don't want to lose any features. Zero the bad one and mark - * the two fields as needing updates once the transaction subsystem - * is online. + * For backwards compatibility, we make both slots equal. + * + * If we detect a mismatched field, we OR the set bits into the + * existing features2 field in case it has already been modified; we + * don't want to lose any features. We then update the bad location + * with the ORed value so that older kernels will see any features2 + * flags, and mark the two fields as needing updates once the + * transaction subsystem is online. */ - if (xfs_sb_has_bad_features2(sbp)) { + if (xfs_sb_has_mismatched_features2(sbp)) { cmn_err(CE_WARN, "XFS: correcting sb_features alignment problem"); sbp->sb_features2 |= sbp->sb_bad_features2; - sbp->sb_bad_features2 = 0; + sbp->sb_bad_features2 = sbp->sb_features2; update_flags |= XFS_SB_FEATURES2 | XFS_SB_BAD_FEATURES2; + + /* + * Re-check for ATTR2 in case it was found in bad_features2 + * slot. + */ + if (xfs_sb_version_hasattr2(&mp->m_sb)) + mp->m_flags |= XFS_MOUNT_ATTR2; + } /* @@ -1896,7 +1908,8 @@ xfs_uuid_unmount( /* * Used to log changes to the superblock unit and width fields which could - * be altered by the mount options. Only the first superblock is updated. + * be altered by the mount options, as well as any potential sb_features2 + * fixup. Only the first superblock is updated. */ STATIC void xfs_mount_log_sb( diff --git a/fs/xfs/xfs_sb.h b/fs/xfs/xfs_sb.h index b1a83f8ec04..d904efe7f87 100644 --- a/fs/xfs/xfs_sb.h +++ b/fs/xfs/xfs_sb.h @@ -320,11 +320,12 @@ static inline int xfs_sb_good_version(xfs_sb_t *sbp) #endif /* __KERNEL__ */ /* - * Detect a bad features2 field + * Detect a mismatched features2 field. Older kernels read/wrote + * this into the wrong slot, so to be safe we keep them in sync. */ -static inline int xfs_sb_has_bad_features2(xfs_sb_t *sbp) +static inline int xfs_sb_has_mismatched_features2(xfs_sb_t *sbp) { - return (sbp->sb_bad_features2 != 0); + return (sbp->sb_bad_features2 != sbp->sb_features2); } static inline unsigned xfs_sb_version_tonew(unsigned v) -- cgit v1.2.3 From 76b0c26af2736b7e5b87e6ed7ab63901483d5736 Mon Sep 17 00:00:00 2001 From: Roman Zippel Date: Wed, 9 Apr 2008 17:44:07 +0200 Subject: HFS+: fix unlink of links Some time ago while attempting to handle invalid link counts, I botched the unlink of links itself, so this patch fixes this now correctly, so that only the link count of nodes that don't point to links is ignored. Thanks to Vlado Plaga to notify me of this problem. Signed-off-by: Roman Zippel Signed-off-by: Linus Torvalds --- fs/hfsplus/dir.c | 23 +++++++++++++++-------- 1 file changed, 15 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/hfsplus/dir.c b/fs/hfsplus/dir.c index 29683645fa0..5f402367825 100644 --- a/fs/hfsplus/dir.c +++ b/fs/hfsplus/dir.c @@ -340,16 +340,23 @@ static int hfsplus_unlink(struct inode *dir, struct dentry *dentry) if (inode->i_nlink > 0) drop_nlink(inode); - hfsplus_delete_inode(inode); - if (inode->i_ino != cnid && !inode->i_nlink) { - if (!atomic_read(&HFSPLUS_I(inode).opencnt)) { - res = hfsplus_delete_cat(inode->i_ino, HFSPLUS_SB(sb).hidden_dir, NULL); - if (!res) - hfsplus_delete_inode(inode); + if (inode->i_ino == cnid) + clear_nlink(inode); + if (!inode->i_nlink) { + if (inode->i_ino != cnid) { + HFSPLUS_SB(sb).file_count--; + if (!atomic_read(&HFSPLUS_I(inode).opencnt)) { + res = hfsplus_delete_cat(inode->i_ino, + HFSPLUS_SB(sb).hidden_dir, + NULL); + if (!res) + hfsplus_delete_inode(inode); + } else + inode->i_flags |= S_DEAD; } else - inode->i_flags |= S_DEAD; + hfsplus_delete_inode(inode); } else - clear_nlink(inode); + HFSPLUS_SB(sb).file_count--; inode->i_ctime = CURRENT_TIME_SEC; mark_inode_dirty(inode); -- cgit v1.2.3 From 54a015104136974262afa4b8ddd943ea70dec8a2 Mon Sep 17 00:00:00 2001 From: Roland McGrath Date: Thu, 10 Apr 2008 15:37:38 -0700 Subject: asmlinkage_protect replaces prevent_tail_call The prevent_tail_call() macro works around the problem of the compiler clobbering argument words on the stack, which for asmlinkage functions is the caller's (user's) struct pt_regs. The tail/sibling-call optimization is not the only way that the compiler can decide to use stack argument words as scratch space, which we have to prevent. Other optimizations can do it too. Until we have new compiler support to make "asmlinkage" binding on the compiler's own use of the stack argument frame, we have work around all the manifestations of this issue that crop up. More cases seem to be prevented by also keeping the incoming argument variables live at the end of the function. This makes their original stack slots attractive places to leave those variables, so the compiler tends not clobber them for something else. It's still no guarantee, but it handles some observed cases that prevent_tail_call() did not. Signed-off-by: Roland McGrath Signed-off-by: Linus Torvalds --- fs/open.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/open.c b/fs/open.c index a4b12022eda..3fa4e4ffce4 100644 --- a/fs/open.c +++ b/fs/open.c @@ -335,7 +335,7 @@ asmlinkage long sys_ftruncate(unsigned int fd, unsigned long length) { long ret = do_sys_ftruncate(fd, length, 1); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(2, ret, fd, length); return ret; } @@ -350,7 +350,7 @@ asmlinkage long sys_ftruncate64(unsigned int fd, loff_t length) { long ret = do_sys_ftruncate(fd, length, 0); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(2, ret, fd, length); return ret; } #endif @@ -1067,7 +1067,7 @@ asmlinkage long sys_open(const char __user *filename, int flags, int mode) ret = do_sys_open(AT_FDCWD, filename, flags, mode); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(3, ret, filename, flags, mode); return ret; } @@ -1081,7 +1081,7 @@ asmlinkage long sys_openat(int dfd, const char __user *filename, int flags, ret = do_sys_open(dfd, filename, flags, mode); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(4, ret, dfd, filename, flags, mode); return ret; } -- cgit v1.2.3 From 598af051a79d05b751fe793f1fe09fcf74763e02 Mon Sep 17 00:00:00 2001 From: Roland McGrath Date: Thu, 10 Apr 2008 15:38:45 -0700 Subject: asmlinkage_protect sys_io_getevents Use asmlinkage_protect in sys_io_getevents, because GCC for i386 with CONFIG_FRAME_POINTER=n can decide to clobber an argument word on the stack, i.e. the user struct pt_regs. Here the problem is not a tail call, but just the compiler's use of the stack when it inlines and optimizes the body of the called function. This seems to avoid it. Signed-off-by: Roland McGrath Signed-off-by: Linus Torvalds --- fs/aio.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/aio.c b/fs/aio.c index 6af92194062..60a4cd4d113 100644 --- a/fs/aio.c +++ b/fs/aio.c @@ -1790,6 +1790,7 @@ asmlinkage long sys_io_getevents(aio_context_t ctx_id, put_ioctx(ioctx); } + asmlinkage_protect(5, ret, ctx_id, min_nr, nr, events, timeout); return ret; } -- cgit v1.2.3 From 8d1c98b0b5c0148b519c6416e689ef6a89ffcea3 Mon Sep 17 00:00:00 2001 From: Davide Libenzi Date: Thu, 10 Apr 2008 21:29:19 -0700 Subject: eventfd/kaio integration fix Jeff Roberson discovered a race when using kaio eventfd based notifications. When it occurs it can lead tomissed wakeups and hung userspace. This patch fixes the race by moving the notification inside the spinlocked section of kaio. The operation is safe since eventfd spinlock and kaio one are unrelated. Signed-off-by: Davide Libenzi Cc: Zach Brown Cc: Jeff Roberson Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/aio.c | 17 +++++++++-------- 1 file changed, 9 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/aio.c b/fs/aio.c index 60a4cd4d113..228368610df 100644 --- a/fs/aio.c +++ b/fs/aio.c @@ -936,14 +936,6 @@ int aio_complete(struct kiocb *iocb, long res, long res2) return 1; } - /* - * Check if the user asked us to deliver the result through an - * eventfd. The eventfd_signal() function is safe to be called - * from IRQ context. - */ - if (!IS_ERR(iocb->ki_eventfd)) - eventfd_signal(iocb->ki_eventfd, 1); - info = &ctx->ring_info; /* add a completion event to the ring buffer. @@ -992,6 +984,15 @@ int aio_complete(struct kiocb *iocb, long res, long res2) kunmap_atomic(ring, KM_IRQ1); pr_debug("added to ring %p at [%lu]\n", iocb, tail); + + /* + * Check if the user asked us to deliver the result through an + * eventfd. The eventfd_signal() function is safe to be called + * from IRQ context. + */ + if (!IS_ERR(iocb->ki_eventfd)) + eventfd_signal(iocb->ki_eventfd, 1); + put_rq: /* everything turned out well, dispose of the aiocb. */ ret = __aio_put_req(ctx, iocb); -- cgit v1.2.3 From 0859ab59a8a48d2a96b9d2b7100889bcb6bb5818 Mon Sep 17 00:00:00 2001 From: Davide Libenzi Date: Thu, 10 Apr 2008 21:29:29 -0700 Subject: signalfd: fix for incorrect SI_QUEUE user data reporting Michael Kerrisk found out that signalfd was not reporting back user data pushed using sigqueue: http://groups.google.com/group/linux.kernel/msg/9397cab8551e3123 The following patch makes signalfd report back the ssi_ptr and ssi_int members of the signalfd_siginfo structure. Signed-off-by: Davide Libenzi Acked-by: Michael Kerrisk Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/signalfd.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/signalfd.c b/fs/signalfd.c index cb2b63ae0bf..8ead0db3593 100644 --- a/fs/signalfd.c +++ b/fs/signalfd.c @@ -111,9 +111,14 @@ static int signalfd_copyinfo(struct signalfd_siginfo __user *uinfo, err |= __put_user(kinfo->si_uid, &uinfo->ssi_uid); err |= __put_user((long) kinfo->si_ptr, &uinfo->ssi_ptr); break; - default: /* this is just in case for now ... */ + default: + /* + * This case catches also the signals queued by sigqueue(). + */ err |= __put_user(kinfo->si_pid, &uinfo->ssi_pid); err |= __put_user(kinfo->si_uid, &uinfo->ssi_uid); + err |= __put_user((long) kinfo->si_ptr, &uinfo->ssi_ptr); + err |= __put_user(kinfo->si_int, &uinfo->ssi_int); break; } -- cgit v1.2.3 From 6ded55da6be9f186ae1022724a5881b43846c164 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Mon, 7 Apr 2008 15:59:03 -0400 Subject: Documentation: move nfsroot.txt to filesystems/ Documentation/ is a little large, and filesystems/ seems an obvious place for this file. Signed-off-by: J. Bruce Fields Signed-off-by: Jonathan Corbet --- fs/Kconfig | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/Kconfig b/fs/Kconfig index d7312825592..c509123bea4 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -1744,10 +1744,10 @@ config ROOT_NFS If you want your Linux box to mount its whole root file system (the one containing the directory /) from some other computer over the net via NFS (presumably because your box doesn't have a hard disk), - say Y. Read for details. It is - likely that in this case, you also want to say Y to "Kernel level IP - autoconfiguration" so that your box can discover its network address - at boot time. + say Y. Read for + details. It is likely that in this case, you also want to say Y to + "Kernel level IP autoconfiguration" so that your box can discover + its network address at boot time. Most people say N here. -- cgit v1.2.3 From 19e729a928172103e101ffd0829fd13e68c13f78 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Mon, 14 Apr 2008 15:03:02 -0400 Subject: locks: fix possible infinite loop in fcntl(F_SETLKW) over nfs Miklos Szeredi found the bug: "Basically what happens is that on the server nlm_fopen() calls nfsd_open() which returns -EACCES, to which nlm_fopen() returns NLM_LCK_DENIED. "On the client this will turn into a -EAGAIN (nlm_stat_to_errno()), which in will cause fcntl_setlk() to retry forever." So, for example, opening a file on an nfs filesystem, changing permissions to forbid further access, then trying to lock the file, could result in an infinite loop. And Trond Myklebust identified the culprit, from Marc Eshel and I: 7723ec9777d9832849b76475b1a21a2872a40d20 "locks: factor out generic/filesystem switch from setlock code" That commit claimed to just be reshuffling code, but actually introduced a behavioral change by calling the lock method repeatedly as long as it returned -EAGAIN. We assumed this would be safe, since we assumed a lock of type SETLKW would only return with either success or an error other than -EAGAIN. However, nfs does can in fact return -EAGAIN in this situation, and independently of whether that behavior is correct or not, we don't actually need this change, and it seems far safer not to depend on such assumptions about the filesystem's ->lock method. Therefore, revert the problematic part of the original commit. This leaves vfs_lock_file() and its other callers unchanged, while returning fcntl_setlk and fcntl_setlk64 to their former behavior. Signed-off-by: J. Bruce Fields Tested-by: Miklos Szeredi Cc: Trond Myklebust Cc: Marc Eshel Signed-off-by: Linus Torvalds --- fs/locks.c | 48 ++++++++++++++++++++++++++++-------------------- 1 file changed, 28 insertions(+), 20 deletions(-) (limited to 'fs') diff --git a/fs/locks.c b/fs/locks.c index d83fab1b77b..43c0af21a0c 100644 --- a/fs/locks.c +++ b/fs/locks.c @@ -1801,17 +1801,21 @@ again: if (error) goto out; - for (;;) { - error = vfs_lock_file(filp, cmd, file_lock, NULL); - if (error != -EAGAIN || cmd == F_SETLK) - break; - error = wait_event_interruptible(file_lock->fl_wait, - !file_lock->fl_next); - if (!error) - continue; + if (filp->f_op && filp->f_op->lock != NULL) + error = filp->f_op->lock(filp, cmd, file_lock); + else { + for (;;) { + error = posix_lock_file(filp, file_lock, NULL); + if (error != -EAGAIN || cmd == F_SETLK) + break; + error = wait_event_interruptible(file_lock->fl_wait, + !file_lock->fl_next); + if (!error) + continue; - locks_delete_block(file_lock); - break; + locks_delete_block(file_lock); + break; + } } /* @@ -1925,17 +1929,21 @@ again: if (error) goto out; - for (;;) { - error = vfs_lock_file(filp, cmd, file_lock, NULL); - if (error != -EAGAIN || cmd == F_SETLK64) - break; - error = wait_event_interruptible(file_lock->fl_wait, - !file_lock->fl_next); - if (!error) - continue; + if (filp->f_op && filp->f_op->lock != NULL) + error = filp->f_op->lock(filp, cmd, file_lock); + else { + for (;;) { + error = posix_lock_file(filp, file_lock, NULL); + if (error != -EAGAIN || cmd == F_SETLK64) + break; + error = wait_event_interruptible(file_lock->fl_wait, + !file_lock->fl_next); + if (!error) + continue; - locks_delete_block(file_lock); - break; + locks_delete_block(file_lock); + break; + } } /* -- cgit v1.2.3 From abe2f41430770270a1512bf78f489284bfbdbd04 Mon Sep 17 00:00:00 2001 From: Alexey Korolev Date: Mon, 14 Apr 2008 20:45:06 +0100 Subject: JFFS2 Fix of panics caused by wrong condition for hole frag creation in write_begin This fixes a regression introduced in commit 205c109a7a96d9a3d8ffe64c4068b70811fef5e8 when switching to write_begin/write_end operations in JFFS2. The page offset is miscalculated, leading to corruption of the fragment lists and subsequently to memory corruption and panics. [ Side note: the bug is a fairly direct result of the naming. Nick was likely misled by the use of "offs", since we tend to use the notion of "offset" not as an absolute position, but as an offset _within_ a page or allocation. Alternatively, a "pgoff_t" is a page index, but not a byte offset - our VM naming can be a bit confusing. So in this case, a VM person would likely have called this a "pos", not an "offs", or perhaps talked about byte offsets rather than page offsets (since it's counted in bytes, not pages). - Linus ] Signed-off-by: Alexey Korolev Signed-off-by: Vasiliy Leonenko Signed-off-by: David Woodhouse Signed-off-by: Linus Torvalds --- fs/jffs2/file.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/jffs2/file.c b/fs/jffs2/file.c index f9c5dd6f4b6..dcc2734e0b5 100644 --- a/fs/jffs2/file.c +++ b/fs/jffs2/file.c @@ -129,7 +129,7 @@ static int jffs2_write_begin(struct file *filp, struct address_space *mapping, struct inode *inode = mapping->host; struct jffs2_inode_info *f = JFFS2_INODE_INFO(inode); pgoff_t index = pos >> PAGE_CACHE_SHIFT; - uint32_t pageofs = pos & (PAGE_CACHE_SIZE - 1); + uint32_t pageofs = index << PAGE_CACHE_SHIFT; int ret = 0; pg = __grab_cache_page(mapping, index); -- cgit v1.2.3