- 18 3月, 2011 8 次提交
-
-
由 Josef Bacik 提交于
While trying to track down some NFS problems with BTRFS, I kept noticing I was getting -EACCESS for no apparent reason. Eric Paris and printk() helped me figure out that it was SELinux that was giving me grief, with the following denial type=AVC msg=audit(1290013638.413:95): avc: denied { 0x800000 } for pid=1772 comm="nfsd" name="" dev=sda1 ino=256 scontext=system_u:system_r:kernel_t:s0 tcontext=system_u:object_r:unlabeled_t:s0 tclass=file Turns out this is because in d_obtain_alias if we can't find an alias we create one and do all the normal instantiation stuff, but we don't do the security_d_instantiate. Usually we are protected from getting a hashed dentry that hasn't yet run security_d_instantiate() by the parent's i_mutex, but obviously this isn't an option there, so in order to deal with the case that a second thread comes in and finds our new dentry before we get to run security_d_instantiate(), we go ahead and call it if we find a dentry already. Eric assures me that this is ok as the code checks to see if the dentry has been initialized already so calling security_d_instantiate() against the same dentry multiple times is ok. With this patch I'm no longer getting errant -EACCESS values. Signed-off-by: NJosef Bacik <josef@redhat.com> Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
-
由 Al Viro 提交于
it's always false... Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
-
由 Al Viro 提交于
it's always false now Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
-
由 Al Viro 提交于
Have it nested inside ->i_mutex. Instead of using follow_down() under namespace_sem, followed by grabbing i_mutex and checking that mountpoint to be is not dead, do the following: grab i_mutex check that it's not dead grab namespace_sem see if anything is mounted there if not, we've won otherwise drop locks put_path on what we had replace with what's mounted retry everything with new mountpoint to be New helper (lock_mount()) does that. do_add_mount(), do_move_mount(), do_loopback() and pivot_root() switched to it; in case of the last two that eliminates a race we used to have - original code didn't do follow_down(). Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
-
由 Al Viro 提交于
Don't hold vfsmount_lock over the loop traversing ->mnt_parent; do check_mnt(new.mnt) under namespace_sem instead; combined with namespace_sem held over all that code it'll guarantee the stability of ->mnt_parent chain all the way to the root. Doing check_mnt() outside of namespace_sem in case of pivot_root() is wrong anyway. Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
-
由 Al Viro 提交于
new function: mount_fs(). Does all work done by vfs_kern_mount() except the allocation and filling of vfsmount; returns root dentry or ERR_PTR(). vfs_kern_mount() switched to using it and taken to fs/namespace.c, along with its wrappers. alloc_vfsmnt()/free_vfsmnt() made static. functions in namespace.c slightly reordered. Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
-
由 Tony Luck 提交于
1) Change from ->get_sb() to ->mount() 2) Use mount_single() instead of mount_nodev() 3) Pulled in ramfs_get_inode() & trimmed to what I need for pstore 4) Drop the ugly pstore_writefile() Just save data using kmalloc() and provide a pstore_file_read() that uses simple_read_from_buffer(). Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
-
由 Al Viro 提交于
not needed anymore, since all users (->get_sb() instances) are gone. Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
-
- 17 3月, 2011 9 次提交
-
-
由 Al Viro 提交于
This is an ex-parrot. Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
-
由 Al Viro 提交于
The last remaining instances of ->get_sb() can be converted ->mount() now - nothing in them uses new vfsmount anymore. Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
-
由 Al Viro 提交于
now we can do that - nobody cares about its value anymore. Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
-
由 Al Viro 提交于
a) ->show_devname(m, mnt) - what to put into devname columns in mounts, mountinfo and mountstats b) ->show_path(m, mnt) - what to put into relative path column in mountinfo Leaving those NULL gives old behaviour. NFS switched to using those. Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
-
由 Al Viro 提交于
It's always equal to dentry->d_sb Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
-
由 Al Viro 提交于
part 3: now we have everything to get nfs_path() just by dentry - just follow to (disconnected) root and pick the rest of the thing there. Start killing propagation of struct vfsmount * on the paths that used to bring it to nfs_path(). Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
-
由 Al Viro 提交于
part 2: make sure that disconnected roots have corresponding mnt_devname values stashed into them. Have nfs*_get_root() stuff a copy of devname into ->d_fsdata of the found root, provided that it is disconnected. Have ->d_release() free it when dentry goes away. Have the places where NFS uses ->d_fsdata for sillyrename (and that can *never* happen to a disconnected root - dentry will be attached to its parent) free old devname copies if they find those. Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
-
由 Al Viro 提交于
step 1 of ->mnt_devname fixes: make sure we have the value of devname available in ..._get_root(). Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
-
由 Miklos Szeredi 提交于
__fput doesn't need a cdev_put() for O_PATH handles. Signed-off-by: mszeredi@suse.cz Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
-
- 16 3月, 2011 19 次提交
-
-
由 Linus Torvalds 提交于
It turns out that while a maximum of 8 partitions may be what people "should" have had, you can actually fit up to 18 entries(*) in a sector. And some people clearly were taking advantage of that, like Michael Cree, who had ten partitions on one of his OSF disks. (*) The OSF partition data starts at byte offset 64 in the first sector, and the array of 16-byte partition entries start at offset 148 in the on-disk partition structure. Reported-by: NMichael Cree <mcree@orcon.net.nz> Cc: stable@kernel.org (v2.6.38) Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
-
由 Christoph Hellwig 提交于
iprune_sem is continously giving us lockdep warnings because we do take it in read mode in the reclaim path, but we're also doing non-NOFS allocations under it taken in write mode. Taking a bit deeper look at it I think it's fixable quite trivially: - for invalidate_inodes we do not need iprune_sem at all. We have an active reference on the superblock, so the filesystem is not going away until it has finished. - for evict_inodes we do need it, to make sure prune_icache has done it's work before we tear down the superblock. But there is no reason to hold it over the actual reclaim operation - it's enough to cycle through it after the actual reclaim to make sure we wait for any pending prune_icache to complete. We just have to remove the WARN_ON for otherwise busy inodes as they can actually happen now. Signed-off-by: NChristoph Hellwig <hch@lst.de> Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
-
由 Artem Bityutskiy 提交于
Clean-up commentaries in debug.h and remove references to non-existing symblols. Signed-off-by: NArtem Bityutskiy <Artem.Bityutskiy@nokia.com>
-
由 Artem Bityutskiy 提交于
When debugging is enabled, we allocate a buffer of PEB size for various debugging purposes. However, now all users of this buffer are gone and we can safely remove it and save 128KiB or more RAM. Signed-off-by: NArtem Bityutskiy <Artem.Bityutskiy@nokia.com>
-
由 Artem Bityutskiy 提交于
Instead of using pre-allocated 'c->dbg->buf' buffer in 'dbg_scan_orphans()', dynamically allocate it when needed. The intend is to get rid of the pre-allocated 'c->dbg->buf' buffer and save 128KiB of RAM (or more if PEB size is larger). Indeed, currently we allocate this memory even if the user never enables any self-check, which is wasteful. Signed-off-by: NArtem Bityutskiy <Artem.Bityutskiy@nokia.com>
-
由 Artem Bityutskiy 提交于
Instead of using pre-allocated 'c->dbg->buf' buffer in 'dump_lpt_leb()', dynamically allocate it when needed. The intend is to get rid of the pre-allocated 'c->dbg->buf' buffer and save 128KiB of RAM (or more if PEB size is larger). Indeed, currently we allocate this memory even if the user never enables any self-check, which is wasteful. Signed-off-by: NArtem Bityutskiy <Artem.Bityutskiy@nokia.com>
-
由 Artem Bityutskiy 提交于
Instead of using pre-allocated 'c->dbg->buf' buffer in 'dbg_check_ltab_lnum()', dynamically allocate it when needed. The intend is to get rid of the pre-allocated 'c->dbg->buf' buffer and save 128KiB of RAM (or more if PEB size is larger). Indeed, currently we allocate this memory even if the user never enables any self-check, which is wasteful. Signed-off-by: NArtem Bityutskiy <Artem.Bityutskiy@nokia.com>
-
由 Artem Bityutskiy 提交于
Instead of using pre-allocated 'c->dbg->buf' buffer in 'scan_check_cb()', dynamically allocate it when needed. The intend is to get rid of the pre-allocated 'c->dbg->buf' buffer and save 128KiB of RAM (or more if PEB size is larger). Indeed, currently we allocate this memory even if the user never enables any self-check, which is wasteful. Signed-off-by: NArtem Bityutskiy <Artem.Bityutskiy@nokia.com>
-
由 Artem Bityutskiy 提交于
Instead of using pre-allocated 'c->dbg->buf' buffer in 'dbg_dump_leb()', dynamically allocate it when needed. The intend is to get rid of the pre-allocated 'c->dbg->buf' buffer and save 128KiB of RAM (or more if PEB size is larger). Indeed, currently we allocate this memory even if the user never enables any self-check, which is wasteful. Signed-off-by: NArtem Bityutskiy <Artem.Bityutskiy@nokia.com>
-
由 Al Viro 提交于
commit 574197e0 had a missing piece, breaking the loop detection ;-/ Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
-
由 Trond Myklebust 提交于
The new vfs locking scheme introduced in 2.6.38 breaks NFS sillyrename because the latter relies on being able to determine the parent directory of the dentry in the ->iput() callback in order to send the appropriate unlink rpc call. Looking at the code that cares about races with dput(), there doesn't seem to be anything that specifically uses d_parent as a test for whether or not there is a race: - __d_lookup_rcu(), __d_lookup() all test for d_hashed() after d_parent - shrink_dcache_for_umount() is safe since nothing else can rearrange the dentries in that super block. - have_submount(), select_parent() and d_genocide() can test for a deletion if we set the DCACHE_DISCONNECTED flag when the dentry is removed from the parent's d_subdirs list. Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com> Cc: stable@kernel.org (2.6.38, needs commit c826cb7d "dcache.c: create helper function for duplicated functionality" ) Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
-
由 Linus Torvalds 提交于
This creates a helper function for he "try to ascend into the parent directory" case, which was written out in triplicate before. With all the locking and subtle sequence number stuff, we really don't want to duplicate that kind of code. Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
-
由 Al Viro 提交于
* pull the handling of current->total_link_count into __do_follow_link() * put the common "do ->put_link() if needed and path_put() the link" stuff into a helper (put_link(nd, link, cookie)) * rename __do_follow_link() to follow_link(), while we are at it Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
-
由 Al Viro 提交于
The last remaining place (resolution of nested symlink) converted to the loop of the same kind we have in path_lookupat() and path_openat(). Note that we still *do* have a recursion in pathname resolution; can't avoid it, really. However, it's strictly for nested symlinks now - i.e. ones in the middle of a pathname. link_path_walk() has lost the tail now - it always walks everything except the last component. do_follow_link() renamed to nested_symlink() and moved down. Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
-
由 Al Viro 提交于
Now that link_path_walk() is called without LOOKUP_PARENT only from do_follow_link(), we can simplify the checks in last component handling. First of all, checking if we'd arrived to a directory is not needed - the caller will check it anyway. And LOOKUP_FOLLOW is guaranteed to be there, since we only get to that place with nd->depth > 0. Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
-
由 Al Viro 提交于
Now the only caller of link_path_walk() that does *not* pass LOOKUP_PARENT is do_follow_link() Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
-
由 Al Viro 提交于
... and note that we only need to do it for LAST_BIND symlinks Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
-
由 Al Viro 提交于
new helper: walk_component(). Handles everything except symlinks; returns negative on error, 0 on success and 1 on symlinks we decided to follow. Drops out of RCU mode on such symlinks. link_path_walk() and do_last() switched to using that. Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
-
由 Aneesh Kumar K.V 提交于
We don't want to allow creation of private hardlinks by different application using the fd passed to them via SCM_RIGHTS. So limit the null relative name usage in linkat syscall to CAP_DAC_READ_SEARCH Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
-
- 15 3月, 2011 4 次提交
-
-
由 Aneesh Kumar K.V 提交于
Changes to make sure writeback fid is owned by root Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: NVenkateswararao Jujjuri <jvrao@linux.vnet.ibm.com> Signed-off-by: NEric Van Hensbergen <ericvh@gmail.com>
-
由 Aneesh Kumar K.V 提交于
change file attribute can result in making the file readonly. So flush the dirty pages before that. Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: NVenkateswararao Jujjuri <jvrao@linux.vnet.ibm.com> Signed-off-by: NEric Van Hensbergen <ericvh@gmail.com>
-
由 Aneesh Kumar K.V 提交于
We need to call vmtruncate before 9p setattr operation, otherwise we could write back some dirty pages between setattr with ATTR_SIZE and vmtruncate causing some truncated pages to be written back to server Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: NVenkateswararao Jujjuri <jvrao@linux.vnet.ibm.com> Signed-off-by: NEric Van Hensbergen <ericvh@gmail.com>
-
由 Aneesh Kumar K.V 提交于
With caching enabled, we need to make sure we don't update inode->i_size via stat2inode because we could have dirty data which is not yet written to the server Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: NVenkateswararao Jujjuri <jvrao@linux.vnet.ibm.com> Signed-off-by: NEric Van Hensbergen <ericvh@gmail.com>
-