qnx4 wrte support has never been fully implement, is broken since the dawn
of time and hasn't been actively developed since before git history
started.
Instead of letting it further bitrot and complicate API transition (like
the new truncate code) remove it.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Cc: Anders Larsen <al@alarsen.net>
Cc: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
do_sync_write() does the right thing for turning the aio_writev method
into a normal non-vectored synchronous write, no need to duplicate it in
ntfs.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Anton Altaparmakov <aia21@cantab.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Split the anonfd interface into a bare file pointer creation one, and a
file pointer creation plus install one.
There are cases, like the usage of eventfds inside other kernel
interfaces, where the file pointer created by anonfd needs to be used
inside the initialization of other structures.
As it is right now, as soon as anon_inode_getfd() returns, the kenrle can
race with userspace closing the newly installed file descriptor.
This patch, while keeping the old anon_inode_getfd(), introduces a new
anon_inode_getfile() (whose services are reused in anon_inode_getfd())
that allows to split the file creation phase and the fd install one.
Once all the kernel structures are initialized, the code can call the
proper fd_install().
Gregory manifested the need for something like this inside KVM.
Signed-off-by: Davide Libenzi <davidel@xmailserver.org>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: James Morris <jmorris@namei.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Gregory Haskins <ghaskins@novell.com>
Acked-by: Serge Hallyn <serue@us.ibm.com>
Acked-by: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
As mentioned in Documentation/CodingStyle, move EXPORT* macro's
to the line immediately after the closing function brace line.
Also, move the __initcall() similarly.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Zach Brown <zach.brown@oracle.com>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
According to Documentation/CodingStyle the EXPORT* macro should follow
immediately after the closing function brace line.
Also, mark_buffer_async_write_endio() and do_thaw_all() are not used
elsewhere so they should be marked as static.
In addition, file_fsync() is actually in fs/sync.c so move the EXPORT* to
that file.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
We have had a report of bad memory allocation latency during DVD-RAM (UDF)
writing. This is causing the user's desktop session to become unusable.
Jan tracked the cause of this down to UDF inode reclaim blocking:
gnome-screens D ffff810006d1d598 0 20686 1
ffff810006d1d508 0000000000000082 ffff810037db6718 0000000000000800
ffff810006d1d488 ffffffff807e4280 ffffffff807e4280 ffff810006d1a580
ffff8100bccbc140 ffff810006d1a8c0 0000000006d1d4e8 ffff810006d1a8c0
Call Trace:
[<ffffffff804477f3>] io_schedule+0x63/0xa5
[<ffffffff802c2587>] sync_buffer+0x3b/0x3f
[<ffffffff80447d2a>] __wait_on_bit+0x47/0x79
[<ffffffff80447dc6>] out_of_line_wait_on_bit+0x6a/0x77
[<ffffffff802c24f6>] __wait_on_buffer+0x1f/0x21
[<ffffffff802c442a>] __bread+0x70/0x86
[<ffffffff88de9ec7>] :udf:udf_tread+0x38/0x3a
[<ffffffff88de0fcf>] :udf:udf_update_inode+0x4d/0x68c
[<ffffffff88de26e1>] :udf:udf_write_inode+0x1d/0x2b
[<ffffffff802bcf85>] __writeback_single_inode+0x1c0/0x394
[<ffffffff802bd205>] write_inode_now+0x7d/0xc4
[<ffffffff88de2e76>] :udf:udf_clear_inode+0x3d/0x53
[<ffffffff802b39ae>] clear_inode+0xc2/0x11b
[<ffffffff802b3ab1>] dispose_list+0x5b/0x102
[<ffffffff802b3d35>] shrink_icache_memory+0x1dd/0x213
[<ffffffff8027ede3>] shrink_slab+0xe3/0x158
[<ffffffff8027fbab>] try_to_free_pages+0x177/0x232
[<ffffffff8027a578>] __alloc_pages+0x1fa/0x392
[<ffffffff802951fa>] alloc_page_vma+0x176/0x189
[<ffffffff802822d8>] __do_fault+0x10c/0x417
[<ffffffff80284232>] handle_mm_fault+0x466/0x940
[<ffffffff8044b922>] do_page_fault+0x676/0xabf
This blocks with iprune_mutex held, which then blocks other reclaimers:
X D ffff81009d47c400 0 17285 14831
ffff8100844f3728 0000000000000086 0000000000000000 ffff81000000e288
ffff81000000da00 ffffffff807e4280 ffffffff807e4280 ffff81009d47c400
ffffffff805ff890 ffff81009d47c740 00000000844f3808 ffff81009d47c740
Call Trace:
[<ffffffff80447f8c>] __mutex_lock_slowpath+0x72/0xa9
[<ffffffff80447e1a>] mutex_lock+0x1e/0x22
[<ffffffff802b3ba1>] shrink_icache_memory+0x49/0x213
[<ffffffff8027ede3>] shrink_slab+0xe3/0x158
[<ffffffff8027fbab>] try_to_free_pages+0x177/0x232
[<ffffffff8027a578>] __alloc_pages+0x1fa/0x392
[<ffffffff8029507f>] alloc_pages_current+0xd1/0xd6
[<ffffffff80279ac0>] __get_free_pages+0xe/0x4d
[<ffffffff802ae1b7>] __pollwait+0x5e/0xdf
[<ffffffff8860f2b4>] :nvidia:nv_kern_poll+0x2e/0x73
[<ffffffff802ad949>] do_select+0x308/0x506
[<ffffffff802adced>] core_sys_select+0x1a6/0x254
[<ffffffff802ae0b7>] sys_select+0xb5/0x157
Now I think the main problem is having the filesystem block (and do IO) in
inode reclaim. The problem is that this doesn't get accounted well and
penalizes a random allocator with a big latency spike caused by work
generated from elsewhere.
I think the best idea would be to avoid this. By design if possible, or
by deferring the hard work to an asynchronous context. If the latter,
then the fs would probably want to throttle creation of new work with
queue size of the deferred work, but let's not get into those details.
Anyway, the other obvious thing we looked at is the iprune_mutex which is
causing the cascading blocking. We could turn this into an rwsem to
improve concurrency. It is unreasonable to totally ban all potentially
slow or blocking operations in inode reclaim, so I think this is a cheap
way to get a small improvement.
This doesn't solve the whole problem of course. The process doing inode
reclaim will still take the latency hit, and concurrent processes may end
up contending on filesystem locks. So fs developers should keep these
problems in mind.
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Jan Kara <jack@ucw.cz>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Make all seq_operations structs const, to help mitigate against
revectoring user-triggerable function pointers.
This is derived from the grsecurity patch, although generated from scratch
because it's simpler than extracting the changes from there.
Signed-off-by: James Morris <jmorris@namei.org>
Acked-by: Serge Hallyn <serue@us.ibm.com>
Acked-by: Casey Schaufler <casey@schaufler-ca.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Move various magic-number definitions into magic.h.
Signed-off-by: Nick Black <dank@qemfd.net>
Acked-by: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Casey Schaufler <casey@schaufler-ca.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
__estimate_accuracy() was prone to integer overflow, for example if *tv ==
{2147, 483648000} on a 32 bit computer (or even for delays as small as
{429, 500000000} if the task is niced).
Because the result was already forced between 0 and 100ms, the effect of
the overflow was not too problematic, but the use of the hrtimer range
feature was not optimal in overflow cases.
This patch ensures that there can not be an integer overflow in this
function.
Signed-off-by: Guillaume Knispel <gknispel@proformatique.com>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Arjan van de Ven <arjan@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Tejun Heo <tj@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This function uses signed integers for the unix_date and local variables -
if a negative number is supplied and the leap-year condition is not met,
month will be 0, leading to a read of day_n[-1]
Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When calling vfs_unlink() on the lower dentry, d_delete() turns the
dentry into a negative dentry when the d_count is 1. This eventually
caused a NULL pointer deref when a read() or write() was done and the
negative dentry's d_inode was dereferenced in
ecryptfs_read_update_atime() or ecryptfs_getxattr().
Placing mutt's tmpdir in an eCryptfs mount is what initially triggered
the oops and I was able to reproduce it with the following sequence:
open("/tmp/upper/foo", O_RDWR|O_CREAT|O_EXCL|O_NOFOLLOW, 0600) = 3
link("/tmp/upper/foo", "/tmp/upper/bar") = 0
unlink("/tmp/upper/foo") = 0
open("/tmp/upper/bar", O_RDWR|O_CREAT|O_NOFOLLOW, 0600) = 4
unlink("/tmp/upper/bar") = 0
write(4, "eCryptfs test\n"..., 14 <unfinished ...>
+++ killed by SIGKILL +++
https://bugs.launchpad.net/ecryptfs/+bug/387073
Reported-by: Loïc Minier <loic.minier@canonical.com>
Cc: Serge Hallyn <serue@us.ibm.com>
Cc: Dave Kleikamp <shaggy@linux.vnet.ibm.com>
Cc: ecryptfs-devel@lists.launchpad.net
Cc: stable <stable@kernel.org>
Signed-off-by: Tyler Hicks <tyhicks@linux.vnet.ibm.com>
Errors returned from vfs_read() and vfs_write() calls to the lower
filesystem were being masked as -EINVAL. This caused some confusion to
users who saw EINVAL instead of ENOSPC when the disk was full, for
instance.
Also, the actual bytes read or written were not accessible by callers to
ecryptfs_read_lower() and ecryptfs_write_lower(), which may be useful in
some cases. This patch updates the error handling logic where those
functions are called in order to accept positive return codes indicating
success.
Cc: Eric Sandeen <esandeen@redhat.com>
Acked-by: Serge Hallyn <serue@us.ibm.com>
Cc: ecryptfs-devel@lists.launchpad.net
Signed-off-by: Tyler Hicks <tyhicks@linux.vnet.ibm.com>
When searching through the global authentication tokens for a given key
signature, verify that a matching key has not been revoked and has not
expired. This allows the `keyctl revoke` command to be properly used on
keys in use by eCryptfs.
Acked-by: Serge Hallyn <serue@us.ibm.com>
Cc: ecryptfs-devel@lists.launchpad.net
Cc: stable <stable@kernel.org>
Signed-off-by: Tyler Hicks <tyhicks@linux.vnet.ibm.com>
Returns -ENOTSUPP when attempting to use filename encryption with
something other than a password authentication token, such as a private
token from openssl. Using filename encryption with a userspace eCryptfs
key module is a future goal. Until then, this patch handles the
situation a little better than simply using a BUG_ON().
Acked-by: Serge Hallyn <serue@us.ibm.com>
Cc: ecryptfs-devel@lists.launchpad.net
Cc: stable <stable@kernel.org>
Signed-off-by: Tyler Hicks <tyhicks@linux.vnet.ibm.com>
If the lower inode is read-only, don't attempt to open the lower file
read/write and don't hand off the open request to the privileged
eCryptfs kthread for opening it read/write. Instead, only try an
unprivileged, read-only open of the file and give up if that fails.
This patch fixes an oops when eCryptfs is mounted on top of a read-only
mount.
Acked-by: Serge Hallyn <serue@us.ibm.com>
Cc: Eric Sandeen <esandeen@redhat.com>
Cc: Dave Kleikamp <shaggy@linux.vnet.ibm.com>
Cc: ecryptfs-devel@lists.launchpad.net
Cc: stable <stable@kernel.org>
Signed-off-by: Tyler Hicks <tyhicks@linux.vnet.ibm.com>
Returns an error when an unrecognized cipher code is present in a tag 3
packet or an ecryptfs_crypt_stat cannot be initialized. Also sets an
crypt_stat->tfm error pointer to NULL to ensure that it will not be
incorrectly freed in ecryptfs_destroy_crypt_stat().
Acked-by: Serge Hallyn <serue@us.ibm.com>
Cc: ecryptfs-devel@lists.launchpad.net
Cc: stable <stable@kernel.org>
Signed-off-by: Tyler Hicks <tyhicks@linux.vnet.ibm.com>
So, I compiled a 2.6.31-rc5 kernel with ecryptfs and loaded its module.
When it came time to mount my filesystem, I got this in dmesg, and it
refused to mount:
[93577.776637] Unable to allocate crypto cipher with name [aes]; rc = [-2]
[93577.783280] Error attempting to initialize key TFM cipher with name = [aes]; rc = [-2]
[93577.791183] Error attempting to initialize cipher with name = [aes] and key size = [32]; rc = [-2]
[93577.800113] Error parsing options; rc = [-22]
I figured from the error message that I'd either forgotten to load "aes"
or that my key size was bogus. Neither one of those was the case. In
fact, I was missing the CRYPTO_ECB config option and the 'ecb' module.
Unfortunately, there's no trace of 'ecb' in that error message.
I've done two things to fix this. First, I've modified ecryptfs's
Kconfig entry to select CRYPTO_ECB and CRYPTO_CBC. I also took CRYPTO
out of the dependencies since the 'select' will take care of it for us.
I've also modified the error messages to print a string that should
contain both 'ecb' and 'aes' in my error case. That will give any
future users a chance of finding the right modules and Kconfig options.
I also wonder if we should:
select CRYPTO_AES if !EMBEDDED
since I think most ecryptfs users are using AES like me.
Cc: ecryptfs-devel@lists.launchpad.net
Cc: linux-fsdevel@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Cc: Dustin Kirkland <kirkland@canonical.com>
Signed-off-by: Dave Hansen <dave@linux.vnet.ibm.com>
[tyhicks@linux.vnet.ibm.com: Removed extra newline, 80-char violation]
Signed-off-by: Tyler Hicks <tyhicks@linux.vnet.ibm.com>
Lockdep reports the following valid-looking possible AB-BA deadlock with
global_auth_tok_list_mutex and keysig_list_mutex:
ecryptfs_new_file_context() ->
ecryptfs_copy_mount_wide_sigs_to_inode_sigs() ->
mutex_lock(&mount_crypt_stat->global_auth_tok_list_mutex);
-> ecryptfs_add_keysig() ->
mutex_lock(&crypt_stat->keysig_list_mutex);
vs
ecryptfs_generate_key_packet_set() ->
mutex_lock(&crypt_stat->keysig_list_mutex);
-> ecryptfs_find_global_auth_tok_for_sig() ->
mutex_lock(&mount_crypt_stat->global_auth_tok_list_mutex);
ie the two mutexes are taken in opposite orders in the two different
code paths. I'm not sure if this is a real bug where two threads could
actually hit the two paths in parallel and deadlock, but it at least
makes lockdep impossible to use with ecryptfs since this report triggers
every time and disables future lockdep reporting.
Since ecryptfs_add_keysig() is called only from the single callsite in
ecryptfs_copy_mount_wide_sigs_to_inode_sigs(), the simplest fix seems to
be to move the lock of keysig_list_mutex back up outside of the where
global_auth_tok_list_mutex is taken. This patch does that, and fixes
the lockdep report on my system (and ecryptfs still works OK).
The full output of lockdep fixed by this patch is:
=======================================================
[ INFO: possible circular locking dependency detected ]
2.6.31-2-generic #14~rbd2
-------------------------------------------------------
gdm/2640 is trying to acquire lock:
(&mount_crypt_stat->global_auth_tok_list_mutex){+.+.+.}, at: [<ffffffff8121591e>] ecryptfs_find_global_auth_tok_for_sig+0x2e/0x90
but task is already holding lock:
(&crypt_stat->keysig_list_mutex){+.+.+.}, at: [<ffffffff81217728>] ecryptfs_generate_key_packet_set+0x58/0x2b0
which lock already depends on the new lock.
the existing dependency chain (in reverse order) is:
-> #1 (&crypt_stat->keysig_list_mutex){+.+.+.}:
[<ffffffff8108c897>] check_prev_add+0x2a7/0x370
[<ffffffff8108cfc1>] validate_chain+0x661/0x750
[<ffffffff8108d2e7>] __lock_acquire+0x237/0x430
[<ffffffff8108d585>] lock_acquire+0xa5/0x150
[<ffffffff815526cd>] __mutex_lock_common+0x4d/0x3d0
[<ffffffff81552b56>] mutex_lock_nested+0x46/0x60
[<ffffffff8121526a>] ecryptfs_add_keysig+0x5a/0xb0
[<ffffffff81213299>] ecryptfs_copy_mount_wide_sigs_to_inode_sigs+0x59/0xb0
[<ffffffff81214b06>] ecryptfs_new_file_context+0xa6/0x1a0
[<ffffffff8120e42a>] ecryptfs_initialize_file+0x4a/0x140
[<ffffffff8120e54d>] ecryptfs_create+0x2d/0x60
[<ffffffff8113a7d4>] vfs_create+0xb4/0xe0
[<ffffffff8113a8c4>] __open_namei_create+0xc4/0x110
[<ffffffff8113d1c1>] do_filp_open+0xa01/0xae0
[<ffffffff8112d8d9>] do_sys_open+0x69/0x140
[<ffffffff8112d9f0>] sys_open+0x20/0x30
[<ffffffff81013132>] system_call_fastpath+0x16/0x1b
[<ffffffffffffffff>] 0xffffffffffffffff
-> #0 (&mount_crypt_stat->global_auth_tok_list_mutex){+.+.+.}:
[<ffffffff8108c675>] check_prev_add+0x85/0x370
[<ffffffff8108cfc1>] validate_chain+0x661/0x750
[<ffffffff8108d2e7>] __lock_acquire+0x237/0x430
[<ffffffff8108d585>] lock_acquire+0xa5/0x150
[<ffffffff815526cd>] __mutex_lock_common+0x4d/0x3d0
[<ffffffff81552b56>] mutex_lock_nested+0x46/0x60
[<ffffffff8121591e>] ecryptfs_find_global_auth_tok_for_sig+0x2e/0x90
[<ffffffff812177d5>] ecryptfs_generate_key_packet_set+0x105/0x2b0
[<ffffffff81212f49>] ecryptfs_write_headers_virt+0xc9/0x120
[<ffffffff8121306d>] ecryptfs_write_metadata+0xcd/0x200
[<ffffffff8120e44b>] ecryptfs_initialize_file+0x6b/0x140
[<ffffffff8120e54d>] ecryptfs_create+0x2d/0x60
[<ffffffff8113a7d4>] vfs_create+0xb4/0xe0
[<ffffffff8113a8c4>] __open_namei_create+0xc4/0x110
[<ffffffff8113d1c1>] do_filp_open+0xa01/0xae0
[<ffffffff8112d8d9>] do_sys_open+0x69/0x140
[<ffffffff8112d9f0>] sys_open+0x20/0x30
[<ffffffff81013132>] system_call_fastpath+0x16/0x1b
[<ffffffffffffffff>] 0xffffffffffffffff
other info that might help us debug this:
2 locks held by gdm/2640:
#0: (&sb->s_type->i_mutex_key#11){+.+.+.}, at: [<ffffffff8113cb8b>] do_filp_open+0x3cb/0xae0
#1: (&crypt_stat->keysig_list_mutex){+.+.+.}, at: [<ffffffff81217728>] ecryptfs_generate_key_packet_set+0x58/0x2b0
stack backtrace:
Pid: 2640, comm: gdm Tainted: G C 2.6.31-2-generic #14~rbd2
Call Trace:
[<ffffffff8108b988>] print_circular_bug_tail+0xa8/0xf0
[<ffffffff8108c675>] check_prev_add+0x85/0x370
[<ffffffff81094912>] ? __module_text_address+0x12/0x60
[<ffffffff8108cfc1>] validate_chain+0x661/0x750
[<ffffffff81017275>] ? print_context_stack+0x85/0x140
[<ffffffff81089c68>] ? find_usage_backwards+0x38/0x160
[<ffffffff8108d2e7>] __lock_acquire+0x237/0x430
[<ffffffff8108d585>] lock_acquire+0xa5/0x150
[<ffffffff8121591e>] ? ecryptfs_find_global_auth_tok_for_sig+0x2e/0x90
[<ffffffff8108b0b0>] ? check_usage_backwards+0x0/0xb0
[<ffffffff815526cd>] __mutex_lock_common+0x4d/0x3d0
[<ffffffff8121591e>] ? ecryptfs_find_global_auth_tok_for_sig+0x2e/0x90
[<ffffffff8121591e>] ? ecryptfs_find_global_auth_tok_for_sig+0x2e/0x90
[<ffffffff8108c02c>] ? mark_held_locks+0x6c/0xa0
[<ffffffff81125b0d>] ? kmem_cache_alloc+0xfd/0x1a0
[<ffffffff8108c34d>] ? trace_hardirqs_on_caller+0x14d/0x190
[<ffffffff81552b56>] mutex_lock_nested+0x46/0x60
[<ffffffff8121591e>] ecryptfs_find_global_auth_tok_for_sig+0x2e/0x90
[<ffffffff812177d5>] ecryptfs_generate_key_packet_set+0x105/0x2b0
[<ffffffff81212f49>] ecryptfs_write_headers_virt+0xc9/0x120
[<ffffffff8121306d>] ecryptfs_write_metadata+0xcd/0x200
[<ffffffff81210240>] ? ecryptfs_init_persistent_file+0x60/0xe0
[<ffffffff8120e44b>] ecryptfs_initialize_file+0x6b/0x140
[<ffffffff8120e54d>] ecryptfs_create+0x2d/0x60
[<ffffffff8113a7d4>] vfs_create+0xb4/0xe0
[<ffffffff8113a8c4>] __open_namei_create+0xc4/0x110
[<ffffffff8113d1c1>] do_filp_open+0xa01/0xae0
[<ffffffff8129a93e>] ? _raw_spin_unlock+0x5e/0xb0
[<ffffffff8155410b>] ? _spin_unlock+0x2b/0x40
[<ffffffff81139e9b>] ? getname+0x3b/0x240
[<ffffffff81148a5a>] ? alloc_fd+0xfa/0x140
[<ffffffff8112d8d9>] do_sys_open+0x69/0x140
[<ffffffff81553b8f>] ? trace_hardirqs_on_thunk+0x3a/0x3f
[<ffffffff8112d9f0>] sys_open+0x20/0x30
[<ffffffff81013132>] system_call_fastpath+0x16/0x1b
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Tyler Hicks <tyhicks@linux.vnet.ibm.com>
In ecryptfs_destroy_inode(), inode_info->lower_file_mutex is locked,
and just after the mutex is unlocked, the code does:
kmem_cache_free(ecryptfs_inode_info_cache, inode_info);
This means that if another context could possibly try to take the same
mutex as ecryptfs_destroy_inode(), then it could end up getting the
mutex just before the data structure containing the mutex is freed.
So any such use would be an obvious use-after-free bug (catchable with
slab poisoning or mutex debugging), and therefore the locking in
ecryptfs_destroy_inode() is not needed and can be dropped.
Similarly, in ecryptfs_destroy_crypt_stat(), crypt_stat->keysig_list_mutex
is locked, and then the mutex is unlocked just before the code does:
memset(crypt_stat, 0, sizeof(struct ecryptfs_crypt_stat));
Therefore taking this mutex is similarly not necessary.
Removing this locking fixes false-positive lockdep reports such as the
following (and they are false-positives for exactly the same reason
that the locking is not needed):
=================================
[ INFO: inconsistent lock state ]
2.6.31-2-generic #14~rbd3
---------------------------------
inconsistent {RECLAIM_FS-ON-W} -> {IN-RECLAIM_FS-W} usage.
kswapd0/323 [HC0[0]:SC0[0]:HE1:SE1] takes:
(&inode_info->lower_file_mutex){+.+.?.}, at: [<ffffffff81210d34>] ecryptfs_destroy_inode+0x34/0x100
{RECLAIM_FS-ON-W} state was registered at:
[<ffffffff8108c02c>] mark_held_locks+0x6c/0xa0
[<ffffffff8108c10f>] lockdep_trace_alloc+0xaf/0xe0
[<ffffffff81125a51>] kmem_cache_alloc+0x41/0x1a0
[<ffffffff8113117a>] get_empty_filp+0x7a/0x1a0
[<ffffffff8112dd46>] dentry_open+0x36/0xc0
[<ffffffff8121a36c>] ecryptfs_privileged_open+0x5c/0x2e0
[<ffffffff81210283>] ecryptfs_init_persistent_file+0xa3/0xe0
[<ffffffff8120e838>] ecryptfs_lookup_and_interpose_lower+0x278/0x380
[<ffffffff8120f97a>] ecryptfs_lookup+0x12a/0x250
[<ffffffff8113930a>] real_lookup+0xea/0x160
[<ffffffff8113afc8>] do_lookup+0xb8/0xf0
[<ffffffff8113b518>] __link_path_walk+0x518/0x870
[<ffffffff8113bd9c>] path_walk+0x5c/0xc0
[<ffffffff8113be5b>] do_path_lookup+0x5b/0xa0
[<ffffffff8113bfe7>] user_path_at+0x57/0xa0
[<ffffffff811340dc>] vfs_fstatat+0x3c/0x80
[<ffffffff8113424b>] vfs_stat+0x1b/0x20
[<ffffffff81134274>] sys_newstat+0x24/0x50
[<ffffffff81013132>] system_call_fastpath+0x16/0x1b
[<ffffffffffffffff>] 0xffffffffffffffff
irq event stamp: 7811
hardirqs last enabled at (7811): [<ffffffff810c037f>] call_rcu+0x5f/0x90
hardirqs last disabled at (7810): [<ffffffff810c0353>] call_rcu+0x33/0x90
softirqs last enabled at (3764): [<ffffffff810631da>] __do_softirq+0x14a/0x220
softirqs last disabled at (3751): [<ffffffff8101440c>] call_softirq+0x1c/0x30
other info that might help us debug this:
2 locks held by kswapd0/323:
#0: (shrinker_rwsem){++++..}, at: [<ffffffff810f67ed>] shrink_slab+0x3d/0x190
#1: (&type->s_umount_key#35){.+.+..}, at: [<ffffffff811429a1>] prune_dcache+0xd1/0x1b0
stack backtrace:
Pid: 323, comm: kswapd0 Tainted: G C 2.6.31-2-generic #14~rbd3
Call Trace:
[<ffffffff8108ad6c>] print_usage_bug+0x18c/0x1a0
[<ffffffff8108aff0>] ? check_usage_forwards+0x0/0xc0
[<ffffffff8108bac2>] mark_lock_irq+0xf2/0x280
[<ffffffff8108bd87>] mark_lock+0x137/0x1d0
[<ffffffff81164710>] ? fsnotify_clear_marks_by_inode+0x30/0xf0
[<ffffffff8108bee6>] mark_irqflags+0xc6/0x1a0
[<ffffffff8108d337>] __lock_acquire+0x287/0x430
[<ffffffff8108d585>] lock_acquire+0xa5/0x150
[<ffffffff81210d34>] ? ecryptfs_destroy_inode+0x34/0x100
[<ffffffff8108d2e7>] ? __lock_acquire+0x237/0x430
[<ffffffff815526ad>] __mutex_lock_common+0x4d/0x3d0
[<ffffffff81210d34>] ? ecryptfs_destroy_inode+0x34/0x100
[<ffffffff81164710>] ? fsnotify_clear_marks_by_inode+0x30/0xf0
[<ffffffff81210d34>] ? ecryptfs_destroy_inode+0x34/0x100
[<ffffffff8129a91e>] ? _raw_spin_unlock+0x5e/0xb0
[<ffffffff81552b36>] mutex_lock_nested+0x46/0x60
[<ffffffff81210d34>] ecryptfs_destroy_inode+0x34/0x100
[<ffffffff81145d27>] destroy_inode+0x87/0xd0
[<ffffffff81146b4c>] generic_delete_inode+0x12c/0x1a0
[<ffffffff81145832>] iput+0x62/0x70
[<ffffffff811423c8>] dentry_iput+0x98/0x110
[<ffffffff81142550>] d_kill+0x50/0x80
[<ffffffff81142623>] prune_one_dentry+0xa3/0xc0
[<ffffffff811428b1>] __shrink_dcache_sb+0x271/0x290
[<ffffffff811429d9>] prune_dcache+0x109/0x1b0
[<ffffffff81142abf>] shrink_dcache_memory+0x3f/0x50
[<ffffffff810f68dd>] shrink_slab+0x12d/0x190
[<ffffffff810f9377>] balance_pgdat+0x4d7/0x640
[<ffffffff8104c4c0>] ? finish_task_switch+0x40/0x150
[<ffffffff810f63c0>] ? isolate_pages_global+0x0/0x60
[<ffffffff810f95f7>] kswapd+0x117/0x170
[<ffffffff810777a0>] ? autoremove_wake_function+0x0/0x40
[<ffffffff810f94e0>] ? kswapd+0x0/0x170
[<ffffffff810773be>] kthread+0x9e/0xb0
[<ffffffff8101430a>] child_rip+0xa/0x20
[<ffffffff81013c90>] ? restore_args+0x0/0x30
[<ffffffff81077320>] ? kthread+0x0/0xb0
[<ffffffff81014300>] ? child_rip+0x0/0x20
Signed-off-by: Roland Dreier <roland@digitalvampire.org>
Signed-off-by: Tyler Hicks <tyhicks@linux.vnet.ibm.com>
In ocfs2_file_aio_write, we will prevent direct io if
we find that we are appending(changing i_size) and call
generic_file_aio_write_nolock. But actually O_DIRECT flag
is there and this function will call generic_file_direct_write
eventually which will update i_size and leave di->i_size
alone. The bug is
http://oss.oracle.com/bugzilla/show_bug.cgi?id=1173.
So this patch let ocfs2_direct_IO returns 0 directly if we
are appending so that buffered write will be called and
di->i_size get updated successfully. And this is also
what we want in ocfs2_file_aio_write.
Signed-off-by: Tao Ma <tao.ma@oracle.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
when we check/modify lockres->purge, we should with the protection of lockres->spinlock.
in dlm_purge_lockres(), the checking/modifying is not with the protectin.
this patch fixes it.
Signed-off-by: Wengang Wang <wen.gang.wang@oracle.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
This patch adds the missed mlog_exit() and mlog_exit_void() lines when routines
return.
Signed-off-by: Coly Li <coly.li@suse.de>
Acked-by: Mark Fasheh <mfasheh@suse.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
In a clustered setup, we have to panic the box on journal abort. This is
because we don't have the facility to go hard readonly. With hard ro, another
node would detect node failure and initiate recovery.
Having said that, we shouldn't force panic if the volume is mounted locally.
This patch defers the handling to the mount option, errors.
Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
The ioctl will take 3 parameters: old_path, new_path and
preserve and call vfs_reflink. It is useful when we backport
reflink features to old kernels.
Signed-off-by: Tao Ma <tao.ma@oracle.com>
reflink has 2 options for the destination file:
1. snapshot: reflink will attempt to preserve ownership, permissions,
and all other security state in order to create a full snapshot.
2. new file: it will acquire the data extent sharing but will see the
file's security state and attributes initialized as a new file.
So add the option to ocfs2.
Signed-off-by: Tao Ma <tao.ma@oracle.com>
reflink is a very complicated process, so it can't be integrated
into one transaction. So if the system panic in the operation, we
may leave a unfinished inode in the destication directory.
So we will try to create an inode in orphan_dir first, reflink it
to the src file and then move it to the destication file in the end.
In that way we won't be afraid of any corruption during the reflink.
This patch adds 2 functions for orphan_dir operation:
1. Create a new inode in orphand dir.
2. Move an inode to a target dir.
Note:
fsck.ocfs2 should work for us to remove the unfinished file in the
orphan_dir.
Signed-off-by: Tao Ma <tao.ma@oracle.com>
In order to make the original function more suitable for reflink,
we modify the following inode operations. Both are tiny.
1. ocfs2_mknod_locked only use dentry for mlog, so move it to
the caller so that reflink can use it without dentry.
2. ocfs2_prepare_orphan_dir only want inode to get its ip_blkno.
So use ip_blkno instead.
Signed-off-by: Tao Ma <tao.ma@oracle.com>
In ocfs2_extend_rotate_transaction, op_credits is the orignal
credits in the handle and we only want to extend the credits
for the rotation, but the old solution always double it. It
is harmless for some minor operations, but for actions like
reflink we may rotate tree many times and cause the credits
increase dramatically. So this patch try to only increase
the desired credits.
Signed-off-by: Tao Ma <tao.ma@oracle.com>
Actually the whole reflink will touch refcount tree 2 times:
1. It will add the clusters in the extent record to the tree if it
isn't refcounted before.
2. It will add 1 refcount to these clusters when it add these
extent records to the tree.
So actually we shouldn't do merge in the 1st operation since the 2nd
one will soon be called and we may have to split it again. Do a merge
first and split soon is a waste of time. So we only merge in the 2nd
round. This is done by adding a new internal __ocfs2_increase_refcount
and call it with "not-merge" for 1st refcount operation in reflink.
This also has a side-effect that we don't need to worry too much about
the metadata allocation in the 2nd round since it will only merge and
no split will happen for those records.
Signed-off-by: Tao Ma <tao.ma@oracle.com>
The old xattr value remove is quite simple, it just erase the
tree and free the clusters. But as we have added refcount support,
The process is a little complicated.
We have to lock the refcount tree at the beginning, what's more,
we may split the refcount tree in some cases, so meta/credits are
needed.
Signed-off-by: Tao Ma <tao.ma@oracle.com>
With reflink, there is a need that we create a new xattr indexed
block from the very beginning. So add a new parameter for
ocfs2_create_xattr_block.
Signed-off-by: Tao Ma <tao.ma@oracle.com>
Now with xattr refcount support, we need to check whether
we have xattr refcounted before we remove the refcount tree.
Now the mechanism is:
1) Check whether i_clusters == 0, if no, exit.
2) check whether we have i_xattr_loc in dinode. if yes, exit.
2) Check whether we have inline xattr stored outside, if yes, exit.
4) Remove the tree.
Signed-off-by: Tao Ma <tao.ma@oracle.com>
In ocfs2, when xattr's value is larger than OCFS2_XATTR_INLINE_SIZE,
it will be kept outside of the blocks we store xattr entry. And they
are stored in a b-tree also. So this patch try to attach all these
clusters to refcount tree also.
Signed-off-by: Tao Ma <tao.ma@oracle.com>
Currently we have ocfs2_iterate_xattr_buckets which can receive
a para and a callback to iterate a series of bucket. It is good.
But actually the 2 callers ocfs2_xattr_tree_list_index_block and
ocfs2_delete_xattr_index_block are almost the same. The only
difference is that the latter need to handle the extent record
also. So add a new function named ocfs2_iterate_xattr_index_block.
It can be given func callback which are used for exten record.
So now we only have one iteration function for the xattr index
block. Ane what's more, it is useful for our future reflink
operations.
Signed-off-by: Tao Ma <tao.ma@oracle.com>
In order to make 2 transcation(xattr and cow) independent with each other,
we CoW the whole xattr out in case we are setting them.
Signed-off-by: Tao Ma <tao.ma@oracle.com>
We currently use pagecache to duplicate clusters in CoW,
but it isn't suitable for xattr case. So abstract it out
so that the caller can decide which method it use.
Signed-off-by: Tao Ma <tao.ma@oracle.com>
With the new refcount tree, xattr value can also be refcounted
among multiple files. So return the appropriate extent flags
so that CoW can used it later.
Signed-off-by: Tao Ma <tao.ma@oracle.com>
A reflink creates a snapshot of a file, that means the attributes
must be identical except for three exceptions - nlink, ino, and ctime.
As for time changes, Here is a brief description:
1. Source file:
1) atime: Ignore. Let the lazy atime code handle that.
2) mtime: don't touch.
3) ctime: If we change the tree (adding REFCOUNTED to at least one
extent), update it.
2. Destination file:
1) atime: ignore.
2) mtime: we want it to appear identical to the source.
3) ctime: update.
The idea here is that an ls -l will show the same time for the
src and target - it shows mtime. Backup software like rsync and tar
will treat the new file correctly too.
Signed-off-by: Tao Ma <tao.ma@oracle.com>
2 major functions are added in this patch.
ocfs2_attach_refcount_tree will create a new refcount tree to the
old file if it doesn't have one and insert all the extent records
to the tree if they are not refcounted.
ocfs2_create_reflink_node will:
1. set the refcount tree to the new file.
2. call ocfs2_duplicate_extent_list which will iterate all the
extents for the old file, insert it to the new file and increase
the corresponding referennce count.
Signed-off-by: Tao Ma <tao.ma@oracle.com>
When we truncate a file to a specific size which resides in a reflinked
cluster, we need to CoW it since ocfs2_zero_range_for_truncate will
zero the space after the size(just another type of write).
So we add a "max_cpos" in ocfs2_refcount_cow so that it will stop when
it hit the max cluster offset.
Signed-off-by: Tao Ma <tao.ma@oracle.com>
When we use mmap, we CoW the refcountd clusters in
ocfs2_write_begin_nolock. While for normal file
io(including directio), we do CoW in
ocfs2_prepare_inode_for_write.
Signed-off-by: Tao Ma <tao.ma@oracle.com>
During CoW, if the old extent record is refcounted, we allocate
som new clusters and do CoW. Actually we can have some improvement
here. If the old extent has refcount=1, that means now it is only
used by this file. So we don't need to allocate new clusters, just
remove the refcounted flag and it is OK. We also have to remove
it from the refcount tree while not deleting it.
Signed-off-by: Tao Ma <tao.ma@oracle.com>
This patch try CoW support for a refcounted record.
the whole process will be:
1. Calculate how many clusters we need to CoW and where we start.
Extents that are not completely encompassed by the write will
be broken on 1MB boundaries.
2. Do CoW for the clusters with the help of page cache.
3. Change the b-tree structure with the new allocated clusters.
Signed-off-by: Tao Ma <tao.ma@oracle.com>
Add 'Decrement refcount for delete' in to the normal truncate
process. So for a refcounted extent record, call refcount rec
decrementation instead of cluster free.
Signed-off-by: Tao Ma <tao.ma@oracle.com>
