The address comparison in the __nfs_find_client() function is deceptive.
It uses a memcmp() to check a pair of u32 fields for equality. Not only is
this inefficient, but usually memcmp() is used for comparing two *whole*
sockaddr_in's (which includes comparisons of the address family and port
number), so it's easy to mistake the comparison here for a whole sockaddr
comparison, which it isn't.
So for clarity and efficiency, we replace the memcmp() with a simple test
for equality between the two s_addr fields. This should have no
behavioral effect.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Clean up: mount option parsing uses kstrndup in several places, rather than
using kzalloc. Replace the few remaining uses of kzalloc with kstrndup,
for consistency.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Remove the mount option that allows users to specify an alternate mountd
program number. The client hasn't support setting an alternate mountd
program number for a very long time.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Remove the mount option that allows users to specify an alternate NFS
program number. The client hasn't support setting an alternate NFS
program number for a very long time.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Text-based mount option parsing introduced a minor regression in the
behavior of NFS version 4 mounts. NFS version 4 is not supposed to require
a running rpcbind service on the server in order for a mount to succeed.
In other words, if the mount options don't specify a port number, the port
number is supposed to default to 2049. For earlier versions of NFS, the
default port number was zero in order to cause the RPC client to autobind
to the server's NFS service.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
POSIX requires that ctime and mtime, as reported by the stat(2) call,
reflect the activity of the most recent write(2). To that end, nfs_getattr()
flushes pending dirty writes to a file before doing a GETATTR to allow the
NFS server to set the file's size, ctime, and mtime properly.
However, nfs_getattr() can be starved when a constant stream of application
writes to a file prevents nfs_wb_nocommit() from completing. This usually
results in hangs of programs doing a stat against an NFS file that is being
written. "ls -l" is a common victim of this behavior.
To prevent starvation, hold the file's i_mutex in nfs_getattr() to
freeze applications writes temporarily so the client can more quickly obtain
clean values for a file's size, mtime, and ctime.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
The nfs_wcc_update_inode() function omits logic to convert the type of
the NFS on-the-wire value of a file's size (__u64) to the type of file
size value stored in struct inode (loff_t, which is signed).
Everywhere else in the NFS client I checked already correctly converts the
file size type.
This effects only very large files.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Replace use of rpc_call_setup() with rpc_init_task(), and in cases where we
need to initialise task->tk_action, with rpc_call_start().
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Move the common code for setting up the nfs_write_data and nfs_read_data
structures into fs/nfs/read.c, fs/nfs/write.c and fs/nfs/direct.c.
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
We want the default scheduling priority (priority == 0) to remain
RPC_PRIORITY_NORMAL.
Also ensure that the priority wait queue scheduling is per process id
instead of sometimes being per thread, and sometimes being per inode.
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Added an active/deactive mechanism to the nfs_server structure
allowing async operations to hold off umount until the
operations are done.
Signed-off-by: Steve Dickson <steved@redhat.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Reduce the time spent locking the rpc_sequence structure by queuing the
nfs_seqid only when we are ready to take the lock (when calling
nfs_wait_on_sequence).
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
The current model locks the page twice for no good reason. Optimise by
inlining the parts of nfs_write_begin()/nfs_write_end() that we care about.
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
If the server returns an ENOENT error, we still need to do a d_delete() in
order to ensure that the dentry is deleted.
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
In nfs_do_call_unlink() we check that we haven't raced, and that lookup()
hasn't created an aliased dentry to our sillydeleted dentry. If somebody
has deleted the file on the server and the lookup() resulted in a negative
dentry, then ignore...
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Ensure that readdir revalidates its data cache after blocking on
sillyrename.
Also fix a typo in nfs_do_call_unlink(): swap the ^= for an |=. The result
is the same, since we've already checked that the flag is unset, but it
makes the code more readable.
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
A bug report on nfsd that states that since it was switched to use
splice instead of sendfile, the atime was no longer being updated
on the input file. do_generic_mapping_read() does this when accessing
the file, make splice do it for the direct splice handler.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6.25: (1470 commits)
[IPV6] ADDRLABEL: Fix double free on label deletion.
[PPP]: Sparse warning fixes.
[IPV4] fib_trie: remove unneeded NULL check
[IPV4] fib_trie: More whitespace cleanup.
[NET_SCHED]: Use nla_policy for attribute validation in ematches
[NET_SCHED]: Use nla_policy for attribute validation in actions
[NET_SCHED]: Use nla_policy for attribute validation in classifiers
[NET_SCHED]: Use nla_policy for attribute validation in packet schedulers
[NET_SCHED]: sch_api: introduce constant for rate table size
[NET_SCHED]: Use typeful attribute parsing helpers
[NET_SCHED]: Use typeful attribute construction helpers
[NET_SCHED]: Use NLA_PUT_STRING for string dumping
[NET_SCHED]: Use nla_nest_start/nla_nest_end
[NET_SCHED]: Propagate nla_parse return value
[NET_SCHED]: act_api: use PTR_ERR in tcf_action_init/tcf_action_get
[NET_SCHED]: act_api: use nlmsg_parse
[NET_SCHED]: act_api: fix netlink API conversion bug
[NET_SCHED]: sch_netem: use nla_parse_nested_compat
[NET_SCHED]: sch_atm: fix format string warning
[NETNS]: Add namespace for ICMP replying code.
...
* git://git.kernel.org/pub/scm/linux/kernel/git/sam/kbuild: (79 commits)
Remove references to "make dep"
kconfig: document use of HAVE_*
Introduce new section reference annotations tags: __ref, __refdata, __refconst
kbuild: warn about ld added unique sections
kbuild: add verbose option to Section mismatch reporting in modpost
kconfig: tristate choices with mixed tristate and boolean values
asm-generic/vmlix.lds.h: simplify __mem{init,exit}* dependencies
remove __attribute_used__
kbuild: support ARCH=x86 in buildtar
kconfig: remove "enable"
kbuild: simplified warning report in modpost
kbuild: introduce a few helpers in modpost
kbuild: use simpler section mismatch warnings in modpost
kbuild: link vmlinux.o before kallsyms passes
kbuild: introduce new option to enhance section mismatch analysis
Use separate sections for __dev/__cpu/__mem code/data
compiler.h: introduce __section()
all archs: consolidate init and exit sections in vmlinux.lds.h
kbuild: check section names consistently in modpost
kbuild: introduce blacklisting in modpost
...
* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4: (50 commits)
jbd2: sparse pointer use of zero as null
jbd2: Use round-jiffies() function for the "5 second" ext4/jbd2 wakeup
jbd2: Mark jbd2 slabs as SLAB_TEMPORARY
jbd2: add lockdep support
ext4: Use the ext4_ext_actual_len() helper function
ext4: fix uniniatilized extent splitting error
ext4: Check for return value from sb_set_blocksize
ext4: Add stripe= option to /proc/mounts
ext4: Enable the multiblock allocator by default
ext4: Add multi block allocator for ext4
ext4: Add new functions for searching extent tree
ext4: Add ext4_find_next_bit()
ext4: fix up EXT4FS_DEBUG builds
ext4: Fix ext4_show_options to show the correct mount options.
ext4: Add EXT4_IOC_MIGRATE ioctl
ext4: Add inode version support in ext4
vfs: Add 64 bit i_version support
ext4: Add the journal checksum feature
jbd2: jbd2 stats through procfs
ext4: Take read lock during overwrite case.
...
Get rid of sparse related warnings from places that use integer as NULL
pointer. (Ported from upstream ext3/jbd changes.)
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
While "every 5 seconds" doesn't sound as a problem, there can be many
of these (and these timers do add up over all the kernel). The "5
second" wakeup isn't really timing sensitive; in addition even with
rounding it'll still happen every 5 seconds (with the exception of the
very first time, which is likely to be rounded up to somewhere closer
to 6 seconds)
(Ported from similar JBD patch made by Arjan van de Ven to
fs/jbd/transaction.c)
Cc: Arjan van de Ven <arjan@linux.intel.com>
Cc: Andrew Morton <akpm@osdl.org>
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
This patch marks slab allocations by jbd2 as short-lived in support of
Mel Gorman's "Group short-lived and reclaimable kernel allocations"
patch. (Ported from similar changes made to fs/jbd/journal.c and
fs/jbd/revoke.c in Mel's patch.)
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
ext4 uses the high bit of the extent length to encode whether the extent
is intialized or not. The helper function ext4_ext_get_actual_len should
be used to get the actual length of the extent.
This addresses the kernel bug documented here:
http://bugzilla.kernel.org/show_bug.cgi?id=9732
kernel BUG at fs/ext4/extents.c:1056!
....
Call Trace:
[<ffffffff88366073>] :ext4dev:ext4_ext_get_blocks+0x5ba/0x8c1
[<ffffffff81053c91>] lock_release_holdtime+0x27/0x49
[<ffffffff812748f6>] _spin_unlock+0x17/0x20
[<ffffffff883400a6>] :jbd2:start_this_handle+0x4e0/0x4fe
[<ffffffff88366564>] :ext4dev:ext4_fallocate+0x175/0x39a
[<ffffffff81053c91>] lock_release_holdtime+0x27/0x49
[<ffffffff81056480>] __lock_acquire+0x4e7/0xc4d
[<ffffffff81053c91>] lock_release_holdtime+0x27/0x49
[<ffffffff810a8de7>] sys_fallocate+0xe4/0x10d
[<ffffffff8100c043>] tracesys+0xd5/0xda
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Fix bug reported by Dmitry Monakhov caused by lost error code
Testcase:
blksize = 0x1000;
fd = open(argv[1], O_RDWR|O_CREAT, 0700);
unsigned long long sz = 0x10000000UL;
/* allocating big blocks chunk */
syscall(__NR_fallocate, fd, 0, 0UL, sz)
/* grab all other available filesystem space */
tfd = open("tmp", O_RDWR|O_CREAT|O_DIRECT, 0700);
while( write(tfd, buf, 4096) > 0); /* loop untill ENOSPC */
fsync(fd); /* just in case */
while (pos < sz) {
/* each seek+ write operation result in splits uninitialized extent
in three extents. Splitting may result in new extent allocation
which probably will fail because of ENOSPC*/
lseek(fd, blksize*2 -1, SEEK_CUR);
if ((ret = write(fd, 'a', 1)) != 1)
exit(1);
pos += blksize * 2;
}
Signed-off-by: Dmitry Monakhov <dmonakhov@openvz.org>
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
sb_set_blocksize validates whether the specfied block size can be used by
the file system. Make sure we fail mounting the file system if the
blocksize specfied cannot be used.
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
Enable the multiblock allocator by default.
Fix ext4_show_options() so if it is not enabled, the nomballoc option
included in /proc/mounts.
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Acked-by: Eric Sandeen <sandeen@redhat.com>
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Add the functions ext4_ext_search_left() and ext4_ext_search_right(),
which are used by mballoc during ext4_ext_get_blocks to decided whether
to merge extent information.
Signed-off-by: Alex Tomas <alex@clusterfs.com>
Signed-off-by: Andreas Dilger <adilger@clusterfs.com>
Signed-off-by: Johann Lombardi <johann@clusterfs.com>
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Builds with EXT4FS_DEBUG defined (to enable ext4_debug()) fail
without these changes. Clean up some format warnings too.
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
We need to look at the default value and make sure
the mount options are not set via default value
before showing them via ext4_show_options
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
This patch adds 64-bit inode version support to ext4. The lower 32 bits
are stored in the osd1.linux1.l_i_version field while the high 32 bits
are stored in the i_version_hi field newly created in the ext4_inode.
This field is incremented in case the ext4_inode is large enough. A
i_version mount option has been added to enable the feature.
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
Signed-off-by: Andreas Dilger <adilger@clusterfs.com>
Signed-off-by: Kalpak Shah <kalpak@clusterfs.com>
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Jean Noel Cordenner <jean-noel.cordenner@bull.net>
The i_version field of the inode is changed to be a 64-bit counter that
is set on every inode creation and that is incremented every time the
inode data is modified (similarly to the "ctime" time-stamp).
The aim is to fulfill a NFSv4 requirement for rfc3530.
This first part concerns the vfs, it converts the 32-bit i_version in
the generic inode to a 64-bit, a flag is added in the super block in
order to check if the feature is enabled and the i_version is
incremented in the vfs.
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
Signed-off-by: Jean Noel Cordenner <jean-noel.cordenner@bull.net>
Signed-off-by: Kalpak Shah <kalpak@clusterfs.com>
The journal checksum feature adds two new flags i.e
JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT and JBD2_FEATURE_COMPAT_CHECKSUM.
JBD2_FEATURE_CHECKSUM flag indicates that the commit block contains the
checksum for the blocks described by the descriptor blocks.
Due to checksums, writing of the commit record no longer needs to be
synchronous. Now commit record can be sent to disk without waiting for
descriptor blocks to be written to disk. This behavior is controlled
using JBD2_FEATURE_ASYNC_COMMIT flag. Older kernels/e2fsck should not be
able to recover the journal with _ASYNC_COMMIT hence it is made
incompat.
The commit header has been extended to hold the checksum along with the
type of the checksum.
For recovery in pass scan checksums are verified to ensure the sanity
and completeness(in case of _ASYNC_COMMIT) of every transaction.
Signed-off-by: Andreas Dilger <adilger@clusterfs.com>
Signed-off-by: Girish Shilamkar <girish@clusterfs.com>
Signed-off-by: Dave Kleikamp <shaggy@linux.vnet.ibm.com>
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
The patch below updates the jbd stats patch to 2.6.20/jbd2.
The initial patch was posted by Alex Tomas in December 2005
(http://marc.info/?l=linux-ext4&m=113538565128617&w=2).
It provides statistics via procfs such as transaction lifetime and size.
Sometimes, investigating performance problems, i find useful to have
stats from jbd about transaction's lifetime, size, etc. here is a
patch for review and inclusion probably.
for example, stats after creation of 3M files in htree directory:
[root@bob ~]# cat /proc/fs/jbd/sda/history
R/C tid wait run lock flush log hndls block inlog ctime write drop close
R 261 8260 2720 0 0 750 9892 8170 8187
C 259 750 0 4885 1
R 262 20 2200 10 0 770 9836 8170 8187
R 263 30 2200 10 0 3070 9812 8170 8187
R 264 0 5000 10 0 1340 0 0 0
C 261 8240 3212 4957 0
R 265 8260 1470 0 0 4640 9854 8170 8187
R 266 0 5000 10 0 1460 0 0 0
C 262 8210 2989 4868 0
R 267 8230 1490 10 0 4440 9875 8171 8188
R 268 0 5000 10 0 1260 0 0 0
C 263 7710 2937 4908 0
R 269 7730 1470 10 0 3330 9841 8170 8187
R 270 0 5000 10 0 830 0 0 0
C 265 8140 3234 4898 0
C 267 720 0 4849 1
R 271 8630 2740 20 0 740 9819 8170 8187
C 269 800 0 4214 1
R 272 40 2170 10 0 830 9716 8170 8187
R 273 40 2280 0 0 3530 9799 8170 8187
R 274 0 5000 10 0 990 0 0 0
where,
R - line for transaction's life from T_RUNNING to T_FINISHED
C - line for transaction's checkpointing
tid - transaction's id
wait - for how long we were waiting for new transaction to start
(the longest period journal_start() took in this transaction)
run - real transaction's lifetime (from T_RUNNING to T_LOCKED
lock - how long we were waiting for all handles to close
(time the transaction was in T_LOCKED)
flush - how long it took to flush all data (data=ordered)
log - how long it took to write the transaction to the log
hndls - how many handles got to the transaction
block - how many blocks got to the transaction
inlog - how many blocks are written to the log (block + descriptors)
ctime - how long it took to checkpoint the transaction
write - how many blocks have been written during checkpointing
drop - how many blocks have been dropped during checkpointing
close - how many running transactions have been closed to checkpoint this one
all times are in msec.
[root@bob ~]# cat /proc/fs/jbd/sda/info
280 transaction, each upto 8192 blocks
average:
1633ms waiting for transaction
3616ms running transaction
5ms transaction was being locked
1ms flushing data (in ordered mode)
1799ms logging transaction
11781 handles per transaction
5629 blocks per transaction
5641 logged blocks per transaction
Signed-off-by: Johann Lombardi <johann.lombardi@bull.net>
Signed-off-by: Mariusz Kozlowski <m.kozlowski@tuxland.pl>
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
When we are overwriting a file and not actually allocating new file system
blocks we need to take only the read lock on i_data_sem.
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
We are currently taking the truncate_mutex for every read. This would have
performance impact on large CPU configuration. Convert the lock to read write
semaphore and take read lock when we are trying to read the file.
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
When doing a migrate from ext3 to ext4 inode we need to make sure the test
for inode type and walking inode data happens inside lock. To make this
happen move truncate_mutex early before checking the i_flags.
This actually should enable us to remove the verify_chain().
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Chuck Lever