17 Mar, 2007
1 commit
-
The current NFS client congestion logic is severly broken, it marks the
backing device congested during each nfs_writepages() call but doesn't
mirror this in nfs_writepage() which makes for deadlocks. Also it
implements its own waitqueue.Replace this by a more regular congestion implementation that puts a cap on
the number of active writeback pages and uses the bdi congestion waitqueue.Also always use an interruptible wait since it makes sense to be able to
SIGKILL the process even for mounts without 'intr'.Signed-off-by: Peter Zijlstra
Acked-by: Trond Myklebust
Cc: Christoph Lameter
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
13 Feb, 2007
2 commits
-
Conflicts:
net/sunrpc/auth_gss/gss_krb5_crypto.c
net/sunrpc/auth_gss/gss_spkm3_token.c
net/sunrpc/clnt.cMerge with mainline and fix conflicts.
-
Many struct inode_operations in the kernel can be "const". Marking them const
moves these to the .rodata section, which avoids false sharing with potential
dirty data. In addition it'll catch accidental writes at compile time to
these shared resources.Signed-off-by: Arjan van de Ven
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
04 Feb, 2007
1 commit
-
Trond, looks like the changes to include/linux/nfs_fs.h in 2.6.18
that moved the #include's of sunrpc header files into the #ifdef __KERNEL__ block
disabled nfs debugging for all nfs c file not including any sunrpc header.The following patch moves the definition down, right before its use
for defining ifdebug.Signed-off-by: Benny Halevy
(Moved definition further down into the __KERNEL__ section: Trond)
Signed-off-by: Trond Myklebust
25 Jan, 2007
1 commit
-
Prevent the call to invalidate_inode_pages2() from racing with file writes
by taking the inode->i_mutex across the page cache flush and invalidate.Signed-off-by: Trond Myklebust
Signed-off-by: Linus Torvalds
06 Dec, 2006
6 commits
-
Signed-off-by: Trond Myklebust
-
Signed-off-by: Trond Myklebust
-
We will want to allow nfs_writepage() to distinguish between pages that
have been marked as dirty by the VM, and those that have been marked as
dirty by nfs_updatepage().
In the former case, the entire page will want to be written out, and so any
requests that were pending need to be flushed out first.Signed-off-by: Trond Myklebust
-
Allow callers to directly pass it a struct writeback_control.
Signed-off-by: Trond Myklebust
-
Pass down struct writeback control.
Signed-off-by: Trond Myklebust
-
Use RCU to ensure that we can safely call rpc_finish_wakeup after we've
called __rpc_do_wake_up_task. If not, there is a theoretical race, in which
the rpc_task finishes executing, and gets freed first.Signed-off-by: Trond Myklebust
03 Dec, 2006
1 commit
-
Signed-off-by: Al Viro
Signed-off-by: David S. Miller
21 Oct, 2006
2 commits
-
Signed-off-by: Al Viro
Acked-by: Trond Myklebust
Acked-by: Neil Brown
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
If invalidate_inode_pages2() fails, then it should in principle just be
because the current process was signalled. In that case, we just want to
ensure that the inode's page cache remains marked as invalid.Also add a helper to allow the O_DIRECT code to simply mark the page cache as
invalid once it is finished writing, instead of calling
invalidate_inode_pages2() itself.Signed-off-by: Trond Myklebust
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
01 Oct, 2006
1 commit
-
This patch vectorizes aio_read() and aio_write() methods to prepare for
collapsing all aio & vectored operations into one interface - which is
aio_read()/aio_write().Signed-off-by: Badari Pulavarty
Signed-off-by: Christoph Hellwig
Cc: Michael Holzheu
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
27 Sep, 2006
1 commit
-
Signed-off-by: Adrian Bunk
Acked-by: Trond Myklebust
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
24 Sep, 2006
1 commit
-
Signed-off-by: Jeff Garzik
23 Sep, 2006
5 commits
-
Remove some unused macros related to accessing an RPC peer address
Test plan:
Compile kernel with CONFIG_NFS option enabled.Signed-off-by: Chuck Lever
Signed-off-by: Trond Myklebust -
Move the rpc_ops from the nfs_server struct to the nfs_client struct as they're
common to all server records of a particular NFS protocol version.Signed-Off-By: David Howells
Signed-off-by: Trond Myklebust -
Generalise the nfs_client structure by:
(1) Moving nfs_client to a more general place (nfs_fs_sb.h).
(2) Renaming its maintenance routines to be non-NFS4 specific.
(3) Move those maintenance routines to a new non-NFS4 specific file (client.c)
and move the declarations to internal.h.(4) Make nfs_find/get_client() take a full sockaddr_in to include the port
number (will be required for NFS2/3).(5) Make nfs_find/get_client() take the NFS protocol version (again will be
required to differentiate NFS2, 3 & 4 client records).Also:
(6) Make nfs_client construction proceed akin to inodes, marking them as under
construction and providing a function to indicate completion.(7) Make nfs_get_client() wait interruptibly if it finds a client that it can
share, but that client is currently being constructed.(8) Make nfs4_create_client() use (6) and (7) instead of locking cl_sem.
Signed-Off-By: David Howells
Signed-off-by: Trond Myklebust -
...in order to allow the addition of a memory shrinker.
Signed-off-by: Trond Myklebust
-
The current access cache only allows one entry at a time to be cached for each
inode. Add a per-inode red-black tree in order to allow more than one to
be cached at a time.Should significantly cut down the time spent in path traversal for shared
directories such as ${PATH}, /usr/share, etc.Signed-off-by: Trond Myklebust
13 Sep, 2006
1 commit
-
We don't need any of this crap included from the user-visible part of nfs_fs.h
-- remove it all.In fact, we probably don't need anything but NFS_SUPER_MAGIC to be defined; is
there any need for anything else? And magic numbers should probably move to
rather than being strewn across various fs-specific include
files which exist in userspace for solely that purpose.With this patch, 'make header_check' works again at least on PowerPC.
Signed-off-by: David Woodhouse
Cc: Trond Myklebust
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
09 Sep, 2006
1 commit
-
The logic in nfs_direct_read_schedule and nfs_direct_write_schedule can
allow data->npages to be one larger than rpages. This causes a page
pointer to be written beyond the end of the pagevec in nfs_read_data (or
nfs_write_data).Fix this by making nfs_(read|write)_alloc() calculate the size of the
pagevec array, and initialise data->npages.Also get rid of the redundant argument to nfs_commit_alloc().
Signed-off-by: Trond Myklebust
Cc: Chuck Lever
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
04 Aug, 2006
1 commit
-
nfs_writedata_free() and nfs_readdata_free() can now become static.
Signed-off-by: Adrian Bunk
Cc: Trond Myklebust
Signed-off-by: Andrew Morton
Signed-off-by: Trond Myklebust
(cherry picked from 5e1ce40f0c3c8f67591aff17756930d7a18ceb1a commit)
01 Jul, 2006
1 commit
-
Signed-off-by: Jörn Engel
Signed-off-by: Adrian Bunk
29 Jun, 2006
1 commit
-
Same as with already do with the file operations: keep them in .rodata and
prevents people from doing runtime patching.Signed-off-by: Christoph Hellwig
Cc: Steven French
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
25 Jun, 2006
1 commit
-
Fix various problems with nfs4 disabled. And various other things.
In file included from fs/nfs/inode.c:50:
fs/nfs/internal.h:24: error: static declaration of 'nfs_do_refmount' follows non-static declaration
include/linux/nfs_fs.h:320: error: previous declaration of 'nfs_do_refmount' was here
fs/nfs/internal.h:65: warning: 'struct nfs4_fs_locations' declared inside parameter list
fs/nfs/internal.h:65: warning: its scope is only this definition or declaration, which is probably not what you want
fs/nfs/internal.h: In function 'nfs4_path':
fs/nfs/internal.h:97: error: 'struct nfs_server' has no member named 'mnt_path'
fs/nfs/inode.c: In function 'init_once':
fs/nfs/inode.c:1116: error: 'struct nfs_inode' has no member named 'open_states'
fs/nfs/inode.c:1116: error: 'struct nfs_inode' has no member named 'delegation'
fs/nfs/inode.c:1116: error: 'struct nfs_inode' has no member named 'delegation_state'
fs/nfs/inode.c:1116: error: 'struct nfs_inode' has no member named 'rwsem'
distcc[26452] ERROR: compile fs/nfs/inode.c on g5/64 failed
make[1]: *** [fs/nfs/inode.o] Error 1
make: *** [fs/nfs/inode.o] Error 2
make: *** Waiting for unfinished jobs....
In file included from fs/nfs/nfs3xdr.c:26:
fs/nfs/internal.h:24: error: static declaration of 'nfs_do_refmount' follows non-static declaration
include/linux/nfs_fs.h:320: error: previous declaration of 'nfs_do_refmount' was here
fs/nfs/internal.h:65: warning: 'struct nfs4_fs_locations' declared inside parameter list
fs/nfs/internal.h:65: warning: its scope is only this definition or declaration, which is probably not what you want
fs/nfs/internal.h: In function 'nfs4_path':
fs/nfs/internal.h:97: error: 'struct nfs_server' has no member named 'mnt_path'
distcc[26486] ERROR: compile fs/nfs/nfs3xdr.c on g5/64 failed
make[1]: *** [fs/nfs/nfs3xdr.o] Error 1
make: *** [fs/nfs/nfs3xdr.o] Error 2
In file included from fs/nfs/nfs3proc.c:24:
fs/nfs/internal.h:24: error: static declaration of 'nfs_do_refmount' follows non-static declaration
include/linux/nfs_fs.h:320: error: previous declaration of 'nfs_do_refmount' was here
fs/nfs/internal.h:65: warning: 'struct nfs4_fs_locations' declared inside parameter list
fs/nfs/internal.h:65: warning: its scope is only this definition or declaration, which is probably not what you want
fs/nfs/internal.h: In function 'nfs4_path':
fs/nfs/internal.h:97: error: 'struct nfs_server' has no member named 'mnt_path'
distcc[26469] ERROR: compile fs/nfs/nfs3proc.c on bix/32 failed
make[1]: *** [fs/nfs/nfs3proc.o] Error 1
make: *** [fs/nfs/nfs3proc.o] Error 2
**FAILED**Cc: Alexey Dobriyan
Cc: Andreas Gruenbacher
Cc: Andy Adamson
Cc: Chuck Lever
Cc: David Howells
Cc: J. Bruce Fields
Cc: Manoj Naik
Cc: Marc Eshel
Cc: Trond Myklebust
Signed-off-by: Andrew Morton
Signed-off-by: Trond Myklebust
09 Jun, 2006
8 commits
-
Respond to a moved error on NFS lookup by setting up the referral.
Note: We don't actually follow the referral during lookup/getattr, but
later when we detect fsid mismatch in inode revalidation (similar to the
processing done for cloning submounts). Referrals will have fake attributes
until they are actually followed or traversed.Signed-off-by: Manoj Naik
Signed-off-by: Trond Myklebust -
Set up mountpoint when hitting a referral on moved error by getting
fs_locations.Signed-off-by: Manoj Naik
Signed-off-by: Trond Myklebust -
Make automounted partitions expire using the mark_mounts_for_expiry()
function. The timeout is controlled via a sysctl.Signed-off-by: Trond Myklebust
-
Signed-off-by: Trond Myklebust
-
This should enable us to detect if we are crossing a mountpoint in the
case where the server is exporting "nohide" mounts.Signed-off-by: Trond Myklebust
-
In the case of a call to truncate_inode_pages(), we should really try to
cancel any pending writes on the page.Signed-off-by: Trond Myklebust
-
Separate out the function of revalidating the inode metadata, and
revalidating the mapping. The former may be called by lookup(),
and only really needs to check that permissions, ctime, etc haven't changed
whereas the latter needs only done when we want to read data from the page
cache, and may need to sync and then invalidate the mapping.Signed-off-by: Trond Myklebust
-
Whenever the directory changes, we want to make sure that we always
invalidate its page cache. Fix up update_changeattr() and
nfs_mark_for_revalidate() so that they do so.Signed-off-by: Trond Myklebust
29 Mar, 2006
1 commit
-
This is a conversion to make the various file_operations structs in fs/
const. Basically a regexp job, with a few manual fixupsThe goal is both to increase correctness (harder to accidentally write to
shared datastructures) and reducing the false sharing of cachelines with
things that get dirty in .data (while .rodata is nicely read only and thus
cache clean)Signed-off-by: Arjan van de Ven
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
21 Mar, 2006
3 commits
-
Kudos to Neil Brown for spotting the problem:
"in nfs_sync_inode, there is effectively the sequence:
nfs_wait_on_requests
nfs_flush_inode
nfs_commit_inodeThis seems a bit racy to me as if the only requests are on the
->commit list, and nfs_commit_inode is called separately after
nfs_wait_on_requests completes, and before nfs_commit_inode start
(say: by nfs_write_inode) then none of these function will return
>0, yet there will be some pending request that aren't waited for."The solution is to search for requests to wait upon, search for dirty
requests, and search for uncommitted requests while holding the
nfsi->req_lockThe patch also cleans up nfs_sync_inode(), getting rid of the redundant
FLUSH_WAIT flag. It turns out that we were always setting it.Signed-off-by: Trond Myklebust
-
Signed-off-by: Trond Myklebust
-
Currently NFS O_DIRECT writes use FILE_SYNC so that a COMMIT is not
necessary. This simplifies the internal logic, but this could be a
difficult workload for some servers.Instead, let's send UNSTABLE writes, and after they all complete, send a
COMMIT for the dirty range. After the COMMIT returns successfully, then do
the wake_up or fire off aio_complete().Test plan:
Async direct I/O tests against Solaris (or any server that requires
committed unstable writes). Reboot server during test.Based on an earlier patch by Chuck Lever
Signed-off-by: Trond Myklebust