The following set of patches fix known issues with the 2.6.22-rc3 NFS client
code, and significantly enhance the support for NFSv4.
- linux-2.6.22-001-odirect_refcount_fix.dif:
-
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Wed, 30 May 2007 12:58:00 -0400
NFS: Fix a refcount leakage in O_DIRECT
The current code is leaking a reference to dreq->kref when the calls to
nfs_direct_read_schedule() and nfs_direct_write_schedule() return an error. This patch
moves the call to kref_put() from nfs_direct_wait() back into nfs_direct_read() and
nfs_direct_write() (which are the functions that actually took the reference in the
first place) fixing the leak.
Thanks to Denis V. Lunev for spotting the bug and proposing the original fix.
Acked-by: Denis V. Lunev <dlunev@gmail.com>
Acked-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
- linux-2.6.22-002-odirect_cleanup.dif:
-
From: Chuck Lever <chuck.lever@oracle.com>
Date: Sat, 19 May 2007 17:22:52 -0400
NFS: Clean ups in fs/nfs/direct.c
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
- linux-2.6.22-003-odirect_accept_short_get_user_pages.dif:
-
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Tue, 22 May 2007 10:22:20 -0400
NFS: Don't fail an O_DIRECT read/write if get_user_pages() returns pages
There is no need to fail the entire O_DIRECT read/write just because
get_user_pages() returned fewer pages than we requested.
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
- linux-2.6.22-004-fix_uptodate_write.dif:
-
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Sat, 19 May 2007 11:58:03 -0400
NFS: Don't mark a written page as uptodate until it is on disk
The write may fail, so we should not mark the page as uptodate until we are certain
that the data has been accepted and written to disk by the server.
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
- linux-2.6.22-005-read_optimisation.dif:
-
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Sun, 20 May 2007 13:05:05 -0400
NFS: Minor read optimisation...
Since PG_uptodate may now end up getting set during the call to nfs_wb_page(), we
can avoid putting a read request on the wire in those situations.
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
- linux-2.6.22-006-mount_noshared.dif:
-
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Wed, 16 May 2007 16:53:28 -0400
NFS: Add the mount option "nosharecache"
Prior to David Howell's mount changes in 2.6.18, users who mounted different
directories which happened to be from the same filesystem on the server would get
different super blocks, and hence could choose different mount options. As long as
there were no hard linked files that crossed from one subtree to another, this was
quite safe. Post the changes, if the two directories are on the same filesystem (have
the same 'fsid'), they will share the same super block, and hence the same mount
options.
Add a flag to allow users to elect not to share the NFS super block with another
mount point, even if the fsids are the same. This will allow users to set different
mount options for the two different super blocks, as was previously possible. It is
still up to the user to ensure that there are no cache coherency issues when doing
this, however the default behaviour will be to share super blocks whenever two paths
result in the same fsid.
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
- linux-2.6.22-007-mount_shared_return_error.dif:
-
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Wed, 16 May 2007 16:53:28 -0400
NFS: Error when mounting the same filesystem with different options
Unless the user sets the NFS_MOUNT_NOSHAREDCACHE mount flag, we should return EBUSY
if the filesystem is already mounted on a superblock that has set conflicting mount
options.
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
- linux-2.6.22-008-gss_triple_to_pseudoflavor.dif:
-
From: Andy Adamson <andros@umich.edu>
Date: Fri, 18 May 2007 16:52:39 -0400
SUNRPC: Add a helper function to convert a GSS triple into pseudoflavor
Helper function for OP_SECINFO. Return the pseudoflavor for a supported security
mechanism OID, qop, and service. Return RPC_AUTH_MAXFLAVOR upon failure.
Signed-off-by: Andy Adamson <andros@citi.umich.edu>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
- linux-2.6.22-009-nfs_secinfo_xdr.dif:
-
From: Andy Adamson <andros@umich.edu>
Date: Fri, 18 May 2007 16:52:40 -0400
NFSv4: Add the NFS SECINFO xdr encode/decode routines
Encode and decode routines for OP_SECINFO. Decode routine saves an orderd list of up
to NFS4_SECINFO_MAXFLAVORS server supplied pseudoflavors supported on the client.
Signed-off-by: Andy Adamson <andros@citi.umich.edu>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
- linux-2.6.22-010-return_untried_security_flavor.dif:
-
From: Andy Adamson <andros@umich.edu>
Date: Fri, 18 May 2007 16:52:41 -0400
NFSv4: Add a helper to return the next untried security flavor
Helper function for NFS4ERR_WRONGSEC handler Returns RPC_AUTH_MAXFLAVOR when all
flavors have been tried.
Signed-off-by: Andy Adamson <andros@citi.umich.edu>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
- linux-2.6.22-011-obtain_an_rpc_credential_for_nfs4err_wrongsec_handler.dif:
-
From: Andy Adamson <andros@umich.edu>
Date: Fri, 18 May 2007 16:52:42 -0400
NFSv4: Obtain an RPC credential for NFS4ERR_WRONGSEC handler
Helper function for WRONGSEC handler
Signed-off-by: Andy Adamson <andros@citi.umich.edu>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
- linux-2.6.22-012-try_each_security_flavor_in_wrongsec_list.dif:
-
From: Andy Adamson <andros@umich.edu>
Date: Fri, 18 May 2007 16:52:43 -0400
NFSv4: Try each security flavor in WRONGSEC list
Main function for NFS4ERR_WRONGSEC handling. Success returns zero with ex_cred set
and ready for exception retry.
Use a private rpc_clnt to obtain credentials for a flavor in the list.
An rpc_clnt and not an rpc_auth is created and destroyed upon each attempt to get a
credential due to the RPC_AUTH_GSS upcall pipe creation/destruction and rpc.gssd
notification.
NOTE: We could update rpc.gssd to do the right thing when the pipe file goes away,
and then we could keep the ex_clnt around and get creds with new rpc_auth's.
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
- linux-2.6.22-013-nfs_secinfo_proc.dif:
-
From: Andy Adamson <andros@umich.edu>
Date: Fri, 18 May 2007 16:52:44 -0400
NFSv4: Add the NFS SECINFO proc
Place the secinfo call on the wire
Signed-off-by: Andy Adamson <andros@citi.umich.edu>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
- linux-2.6.22-014-collect_secinfo_parameters_for_nfs_lookup_wrongsec_handling.dif:
-
From: Andy Adamson <andros@umich.edu>
Date: Fri, 18 May 2007 16:52:45 -0400
NFSv4: Collect SECINFO parameters for nfs_lookup WRONGSEC handling
Add an exception structure to __nfs4_proc_lookup and to the lookup XDR result
structure and collect the operation, directory filehandle, and name needed to call
SECINFO on WRONGSEC error.
NOTE: The WRONGSEC error could have (incorrectly) been on the PUTFH, but we don't
care.
Signed-off-by: Andy Adamson <andros@citi.umich.edu>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
- linux-2.6.22-015-handle_nfs4err_wrongsec_for_nfs4_proc_lookup.dif:
-
From: Andy Adamson <andros@umich.edu>
Date: Fri, 18 May 2007 16:52:46 -0400
NFSv4: Handle NFS4ERR_WRONGSEC in nfs4_proc_lookup
The first time through the wrongsec handler, setup the ex_list via a secinfo call.
The existance of ex_list signals to skip setup and try the next flavor on the list,
which will only occur if the server secinfo list is incorret, or if we are trying a
list of flavors (from mount) on a PUTROOTFH/PUTFH WRONGSEC.
Free the wrongsec handler resources upon success. NOTE: future patches will pass a
successful flavor to nfs_clone_server() for new fsid nfs_server setup.
Signed-off-by: Andy Adamson <andros@citi.umich.edu>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
- linux-2.6.22-016-add_security_flavor_to_struct_nfs_fattr.dif:
-
From: Andy Adamson <andros@umich.edu>
Date: Fri, 18 May 2007 16:52:47 -0400
NFSv4: Add security flavor to struct nfs_fattr
Set fattr->flavor upon successful nfs4_proc_lookup retry in WRONGSEC error
processing. This will ignored in all nfs4_proc_lookup returns except in
nfs_follow_mount where the fattr->flavor is passed through to nfs_clone_server via
the nfs_mount_data->fattr field.
Signed-off-by: Andy Adamson <andros@citi.umich.edu>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
- linux-2.6.22-NFS_ALL.dif:
-
All of the above
![[ICO]](/icons/blank.gif){kind=icon}
![[PARENTDIR]](/icons/back.gif){kind=icon}
![[TXT]](/icons/text.gif){kind=icon}
![[ ]](/icons/unknown.gif){kind=icon}