Christoph Hellwig a9ab5e8406 nfs: page cache invalidation for dio ...

Make sure to properly invalidate the pagecache before performing direct I/O,
so that no stale pages are left around.  This matches what the generic
direct I/O code does.  Also take the i_mutex over the direct write submission
to avoid the lifelock vs truncate waiting for i_dio_count to decrease, and
to avoid having the pagecache easily repopulated while direct I/O is in
progrss.  Again matching the generic direct I/O code.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>

2014-01-13 17:29:50 -05:00

..

blocklayout

nfs: fix do_div() warning by instead using sector_div()

2013-12-04 12:57:37 -05:00

objlayout

…

cache_lib.c

…

cache_lib.h

…

callback_proc.c

…

callback_xdr.c

…

callback.c

…

callback.h

…

client.c

…

delegation.c

…

delegation.h

…

dir.c

NFS: dprintk() should not print negative fileids and inode numbers

2014-01-05 15:51:23 -05:00

direct.c

nfs: page cache invalidation for dio

2014-01-13 17:29:50 -05:00

dns_resolve.c

…

dns_resolve.h

…

file.c

NFS: dprintk() should not print negative fileids and inode numbers

2014-01-05 15:51:23 -05:00

fscache-index.c

…

fscache.c

…

fscache.h

…

getroot.c

…

idmap.c

…

inode.c

NFSv4.1: Don't trust attributes if a pNFS LAYOUTCOMMIT is outstanding

2014-01-13 12:08:11 -05:00

internal.h

…

iostat.h

…

Kconfig

…

Makefile

…

mount_clnt.c

…

namespace.c

…

netns.h

…

nfs2super.c

…

nfs2xdr.c

…

nfs3acl.c

NFS: dprintk() should not print negative fileids and inode numbers

2014-01-05 15:51:23 -05:00

nfs3client.c

…

nfs3proc.c

…

nfs3super.c

…

nfs3xdr.c

…

nfs4_fs.h

…

nfs4client.c

nfs: check if gssd is running before attempting to use krb5i auth in SETCLIENTID call

2013-12-06 13:06:32 -05:00

nfs4file.c

…

nfs4filelayout.c

NFS: dprintk() should not print negative fileids and inode numbers

2014-01-05 15:51:23 -05:00

nfs4filelayout.h

…

nfs4filelayoutdev.c

nfs: fix dead code of ipv6_addr_scope

2014-01-05 15:38:21 -05:00

nfs4getroot.c

…

nfs4namespace.c

…

nfs4proc.c

nfs4.1: properly handle ENOTSUP in SECINFO_NO_NAME

2014-01-13 17:29:47 -05:00

nfs4renewd.c

…

nfs4session.c

…

nfs4session.h

…

nfs4state.c

point to the right include file in a comment (left over from a9004abc3)

2014-01-05 15:51:35 -05:00

nfs4super.c

NFSv4.1: Fix a race in nfs4_write_inode

2014-01-13 13:34:36 -05:00

nfs4sysctl.c

…

nfs4trace.c

…

nfs4trace.h

…

nfs4xdr.c

NFSv4: OPEN must handle the NFS4ERR_IO return code correctly

2013-12-06 13:06:30 -05:00

nfs.h

…

nfsroot.c

…

nfstrace.c

…

nfstrace.h

…

pagelist.c

…

pnfs_dev.c

…

pnfs.c

NFSv4.1: Fix a race in nfs4_write_inode

2014-01-13 13:34:36 -05:00

pnfs.h

NFSv4.1: Don't trust attributes if a pNFS LAYOUTCOMMIT is outstanding

2014-01-13 12:08:11 -05:00

proc.c

…

read.c

NFS: dprintk() should not print negative fileids and inode numbers

2014-01-05 15:51:23 -05:00

super.c

…

symlink.c

…

sysctl.c

…

unlink.c

…

write.c

NFS: dprintk() should not print negative fileids and inode numbers

2014-01-05 15:51:23 -05:00