linux-stable/fs/fat/nfs.c

294 lines
7.5 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0-only
fat (exportfs): move NFS support code Under memory pressure, the system may evict dentries from cache. When the FAT driver receives a NFS request involving an evicted dentry, it is unable to reconnect it to the filesystem root. This causes the request to fail, often with ENOENT. This is partially due to ineffectiveness of the current FAT NFS implementation, and partially due to an unimplemented fh_to_parent method. The latter can cause file accesses to fail on shares exported with subtree_check. This patch set provides the FAT driver with the ability to reconnect dentries. NFS file handle generation and lookups are simplified and made congruent with ext2. Testing has involved a memory-starved virtual machine running 3.5-rc5 that exports a ~2 GB vfat filesystem containing a kernel tree (~770 MB, ~40000 files, 9 levels). Both 'cp -r' and 'ls -lR' operations were performed from a client, some overlapping, some consecutive. Exports with 'subtree_check' and 'no_subtree_check' have been tested. Note that while this patch set improves FAT's NFS support, it does not eliminate ESTALE errors completely. The following should be considered for NFS clients who are sensitive to ESTALE: * Mounting with lookupcache=none Unfortunately this can degrade performance severely, particularly for deep filesystems. * Incorporating VFS patches to retry ESTALE failures on the client-side, such as https://lkml.org/lkml/2012/6/29/381 * Handling ESTALE errors in client application code This patch: Move NFS-related code into its own C file. No functional changes. Signed-off-by: Steven J. Magnani <steve@digidescorp.com> Acked-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-10-05 00:14:44 +00:00
/* fs/fat/nfs.c
*/
#include <linux/exportfs.h>
#include "fat.h"
struct fat_fid {
u32 i_gen;
u32 i_pos_low;
u16 i_pos_hi;
u16 parent_i_pos_hi;
u32 parent_i_pos_low;
u32 parent_i_gen;
};
#define FAT_FID_SIZE_WITHOUT_PARENT 3
#define FAT_FID_SIZE_WITH_PARENT (sizeof(struct fat_fid)/sizeof(u32))
/*
* Look up a directory inode given its starting cluster.
fat (exportfs): move NFS support code Under memory pressure, the system may evict dentries from cache. When the FAT driver receives a NFS request involving an evicted dentry, it is unable to reconnect it to the filesystem root. This causes the request to fail, often with ENOENT. This is partially due to ineffectiveness of the current FAT NFS implementation, and partially due to an unimplemented fh_to_parent method. The latter can cause file accesses to fail on shares exported with subtree_check. This patch set provides the FAT driver with the ability to reconnect dentries. NFS file handle generation and lookups are simplified and made congruent with ext2. Testing has involved a memory-starved virtual machine running 3.5-rc5 that exports a ~2 GB vfat filesystem containing a kernel tree (~770 MB, ~40000 files, 9 levels). Both 'cp -r' and 'ls -lR' operations were performed from a client, some overlapping, some consecutive. Exports with 'subtree_check' and 'no_subtree_check' have been tested. Note that while this patch set improves FAT's NFS support, it does not eliminate ESTALE errors completely. The following should be considered for NFS clients who are sensitive to ESTALE: * Mounting with lookupcache=none Unfortunately this can degrade performance severely, particularly for deep filesystems. * Incorporating VFS patches to retry ESTALE failures on the client-side, such as https://lkml.org/lkml/2012/6/29/381 * Handling ESTALE errors in client application code This patch: Move NFS-related code into its own C file. No functional changes. Signed-off-by: Steven J. Magnani <steve@digidescorp.com> Acked-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-10-05 00:14:44 +00:00
*/
static struct inode *fat_dget(struct super_block *sb, int i_logstart)
fat (exportfs): move NFS support code Under memory pressure, the system may evict dentries from cache. When the FAT driver receives a NFS request involving an evicted dentry, it is unable to reconnect it to the filesystem root. This causes the request to fail, often with ENOENT. This is partially due to ineffectiveness of the current FAT NFS implementation, and partially due to an unimplemented fh_to_parent method. The latter can cause file accesses to fail on shares exported with subtree_check. This patch set provides the FAT driver with the ability to reconnect dentries. NFS file handle generation and lookups are simplified and made congruent with ext2. Testing has involved a memory-starved virtual machine running 3.5-rc5 that exports a ~2 GB vfat filesystem containing a kernel tree (~770 MB, ~40000 files, 9 levels). Both 'cp -r' and 'ls -lR' operations were performed from a client, some overlapping, some consecutive. Exports with 'subtree_check' and 'no_subtree_check' have been tested. Note that while this patch set improves FAT's NFS support, it does not eliminate ESTALE errors completely. The following should be considered for NFS clients who are sensitive to ESTALE: * Mounting with lookupcache=none Unfortunately this can degrade performance severely, particularly for deep filesystems. * Incorporating VFS patches to retry ESTALE failures on the client-side, such as https://lkml.org/lkml/2012/6/29/381 * Handling ESTALE errors in client application code This patch: Move NFS-related code into its own C file. No functional changes. Signed-off-by: Steven J. Magnani <steve@digidescorp.com> Acked-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-10-05 00:14:44 +00:00
{
struct msdos_sb_info *sbi = MSDOS_SB(sb);
struct hlist_head *head;
struct msdos_inode_info *i;
struct inode *inode = NULL;
fat (exportfs): move NFS support code Under memory pressure, the system may evict dentries from cache. When the FAT driver receives a NFS request involving an evicted dentry, it is unable to reconnect it to the filesystem root. This causes the request to fail, often with ENOENT. This is partially due to ineffectiveness of the current FAT NFS implementation, and partially due to an unimplemented fh_to_parent method. The latter can cause file accesses to fail on shares exported with subtree_check. This patch set provides the FAT driver with the ability to reconnect dentries. NFS file handle generation and lookups are simplified and made congruent with ext2. Testing has involved a memory-starved virtual machine running 3.5-rc5 that exports a ~2 GB vfat filesystem containing a kernel tree (~770 MB, ~40000 files, 9 levels). Both 'cp -r' and 'ls -lR' operations were performed from a client, some overlapping, some consecutive. Exports with 'subtree_check' and 'no_subtree_check' have been tested. Note that while this patch set improves FAT's NFS support, it does not eliminate ESTALE errors completely. The following should be considered for NFS clients who are sensitive to ESTALE: * Mounting with lookupcache=none Unfortunately this can degrade performance severely, particularly for deep filesystems. * Incorporating VFS patches to retry ESTALE failures on the client-side, such as https://lkml.org/lkml/2012/6/29/381 * Handling ESTALE errors in client application code This patch: Move NFS-related code into its own C file. No functional changes. Signed-off-by: Steven J. Magnani <steve@digidescorp.com> Acked-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-10-05 00:14:44 +00:00
head = sbi->dir_hashtable + fat_dir_hash(i_logstart);
spin_lock(&sbi->dir_hash_lock);
hlist: drop the node parameter from iterators I'm not sure why, but the hlist for each entry iterators were conceived list_for_each_entry(pos, head, member) The hlist ones were greedy and wanted an extra parameter: hlist_for_each_entry(tpos, pos, head, member) Why did they need an extra pos parameter? I'm not quite sure. Not only they don't really need it, it also prevents the iterator from looking exactly like the list iterator, which is unfortunate. Besides the semantic patch, there was some manual work required: - Fix up the actual hlist iterators in linux/list.h - Fix up the declaration of other iterators based on the hlist ones. - A very small amount of places were using the 'node' parameter, this was modified to use 'obj->member' instead. - Coccinelle didn't handle the hlist_for_each_entry_safe iterator properly, so those had to be fixed up manually. The semantic patch which is mostly the work of Peter Senna Tschudin is here: @@ iterator name hlist_for_each_entry, hlist_for_each_entry_continue, hlist_for_each_entry_from, hlist_for_each_entry_rcu, hlist_for_each_entry_rcu_bh, hlist_for_each_entry_continue_rcu_bh, for_each_busy_worker, ax25_uid_for_each, ax25_for_each, inet_bind_bucket_for_each, sctp_for_each_hentry, sk_for_each, sk_for_each_rcu, sk_for_each_from, sk_for_each_safe, sk_for_each_bound, hlist_for_each_entry_safe, hlist_for_each_entry_continue_rcu, nr_neigh_for_each, nr_neigh_for_each_safe, nr_node_for_each, nr_node_for_each_safe, for_each_gfn_indirect_valid_sp, for_each_gfn_sp, for_each_host; type T; expression a,c,d,e; identifier b; statement S; @@ -T b; <+... when != b ( hlist_for_each_entry(a, - b, c, d) S | hlist_for_each_entry_continue(a, - b, c) S | hlist_for_each_entry_from(a, - b, c) S | hlist_for_each_entry_rcu(a, - b, c, d) S | hlist_for_each_entry_rcu_bh(a, - b, c, d) S | hlist_for_each_entry_continue_rcu_bh(a, - b, c) S | for_each_busy_worker(a, c, - b, d) S | ax25_uid_for_each(a, - b, c) S | ax25_for_each(a, - b, c) S | inet_bind_bucket_for_each(a, - b, c) S | sctp_for_each_hentry(a, - b, c) S | sk_for_each(a, - b, c) S | sk_for_each_rcu(a, - b, c) S | sk_for_each_from -(a, b) +(a) S + sk_for_each_from(a) S | sk_for_each_safe(a, - b, c, d) S | sk_for_each_bound(a, - b, c) S | hlist_for_each_entry_safe(a, - b, c, d, e) S | hlist_for_each_entry_continue_rcu(a, - b, c) S | nr_neigh_for_each(a, - b, c) S | nr_neigh_for_each_safe(a, - b, c, d) S | nr_node_for_each(a, - b, c) S | nr_node_for_each_safe(a, - b, c, d) S | - for_each_gfn_sp(a, c, d, b) S + for_each_gfn_sp(a, c, d) S | - for_each_gfn_indirect_valid_sp(a, c, d, b) S + for_each_gfn_indirect_valid_sp(a, c, d) S | for_each_host(a, - b, c) S | for_each_host_safe(a, - b, c, d) S | for_each_mesh_entry(a, - b, c, d) S ) ...+> [akpm@linux-foundation.org: drop bogus change from net/ipv4/raw.c] [akpm@linux-foundation.org: drop bogus hunk from net/ipv6/raw.c] [akpm@linux-foundation.org: checkpatch fixes] [akpm@linux-foundation.org: fix warnings] [akpm@linux-foudnation.org: redo intrusive kvm changes] Tested-by: Peter Senna Tschudin <peter.senna@gmail.com> Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Signed-off-by: Sasha Levin <sasha.levin@oracle.com> Cc: Wu Fengguang <fengguang.wu@intel.com> Cc: Marcelo Tosatti <mtosatti@redhat.com> Cc: Gleb Natapov <gleb@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-28 01:06:00 +00:00
hlist_for_each_entry(i, head, i_dir_hash) {
BUG_ON(i->vfs_inode.i_sb != sb);
if (i->i_logstart != i_logstart)
continue;
inode = igrab(&i->vfs_inode);
if (inode)
break;
fat (exportfs): move NFS support code Under memory pressure, the system may evict dentries from cache. When the FAT driver receives a NFS request involving an evicted dentry, it is unable to reconnect it to the filesystem root. This causes the request to fail, often with ENOENT. This is partially due to ineffectiveness of the current FAT NFS implementation, and partially due to an unimplemented fh_to_parent method. The latter can cause file accesses to fail on shares exported with subtree_check. This patch set provides the FAT driver with the ability to reconnect dentries. NFS file handle generation and lookups are simplified and made congruent with ext2. Testing has involved a memory-starved virtual machine running 3.5-rc5 that exports a ~2 GB vfat filesystem containing a kernel tree (~770 MB, ~40000 files, 9 levels). Both 'cp -r' and 'ls -lR' operations were performed from a client, some overlapping, some consecutive. Exports with 'subtree_check' and 'no_subtree_check' have been tested. Note that while this patch set improves FAT's NFS support, it does not eliminate ESTALE errors completely. The following should be considered for NFS clients who are sensitive to ESTALE: * Mounting with lookupcache=none Unfortunately this can degrade performance severely, particularly for deep filesystems. * Incorporating VFS patches to retry ESTALE failures on the client-side, such as https://lkml.org/lkml/2012/6/29/381 * Handling ESTALE errors in client application code This patch: Move NFS-related code into its own C file. No functional changes. Signed-off-by: Steven J. Magnani <steve@digidescorp.com> Acked-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-10-05 00:14:44 +00:00
}
spin_unlock(&sbi->dir_hash_lock);
return inode;
fat (exportfs): move NFS support code Under memory pressure, the system may evict dentries from cache. When the FAT driver receives a NFS request involving an evicted dentry, it is unable to reconnect it to the filesystem root. This causes the request to fail, often with ENOENT. This is partially due to ineffectiveness of the current FAT NFS implementation, and partially due to an unimplemented fh_to_parent method. The latter can cause file accesses to fail on shares exported with subtree_check. This patch set provides the FAT driver with the ability to reconnect dentries. NFS file handle generation and lookups are simplified and made congruent with ext2. Testing has involved a memory-starved virtual machine running 3.5-rc5 that exports a ~2 GB vfat filesystem containing a kernel tree (~770 MB, ~40000 files, 9 levels). Both 'cp -r' and 'ls -lR' operations were performed from a client, some overlapping, some consecutive. Exports with 'subtree_check' and 'no_subtree_check' have been tested. Note that while this patch set improves FAT's NFS support, it does not eliminate ESTALE errors completely. The following should be considered for NFS clients who are sensitive to ESTALE: * Mounting with lookupcache=none Unfortunately this can degrade performance severely, particularly for deep filesystems. * Incorporating VFS patches to retry ESTALE failures on the client-side, such as https://lkml.org/lkml/2012/6/29/381 * Handling ESTALE errors in client application code This patch: Move NFS-related code into its own C file. No functional changes. Signed-off-by: Steven J. Magnani <steve@digidescorp.com> Acked-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-10-05 00:14:44 +00:00
}
static struct inode *fat_ilookup(struct super_block *sb, u64 ino, loff_t i_pos)
{
if (MSDOS_SB(sb)->options.nfs == FAT_NFS_NOSTALE_RO)
return fat_iget(sb, i_pos);
else {
if ((ino < MSDOS_ROOT_INO) || (ino == MSDOS_FSINFO_INO))
return NULL;
return ilookup(sb, ino);
}
}
static struct inode *__fat_nfs_get_inode(struct super_block *sb,
u64 ino, u32 generation, loff_t i_pos)
fat (exportfs): move NFS support code Under memory pressure, the system may evict dentries from cache. When the FAT driver receives a NFS request involving an evicted dentry, it is unable to reconnect it to the filesystem root. This causes the request to fail, often with ENOENT. This is partially due to ineffectiveness of the current FAT NFS implementation, and partially due to an unimplemented fh_to_parent method. The latter can cause file accesses to fail on shares exported with subtree_check. This patch set provides the FAT driver with the ability to reconnect dentries. NFS file handle generation and lookups are simplified and made congruent with ext2. Testing has involved a memory-starved virtual machine running 3.5-rc5 that exports a ~2 GB vfat filesystem containing a kernel tree (~770 MB, ~40000 files, 9 levels). Both 'cp -r' and 'ls -lR' operations were performed from a client, some overlapping, some consecutive. Exports with 'subtree_check' and 'no_subtree_check' have been tested. Note that while this patch set improves FAT's NFS support, it does not eliminate ESTALE errors completely. The following should be considered for NFS clients who are sensitive to ESTALE: * Mounting with lookupcache=none Unfortunately this can degrade performance severely, particularly for deep filesystems. * Incorporating VFS patches to retry ESTALE failures on the client-side, such as https://lkml.org/lkml/2012/6/29/381 * Handling ESTALE errors in client application code This patch: Move NFS-related code into its own C file. No functional changes. Signed-off-by: Steven J. Magnani <steve@digidescorp.com> Acked-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-10-05 00:14:44 +00:00
{
struct inode *inode = fat_ilookup(sb, ino, i_pos);
if (inode && generation && (inode->i_generation != generation)) {
iput(inode);
inode = NULL;
}
if (inode == NULL && MSDOS_SB(sb)->options.nfs == FAT_NFS_NOSTALE_RO) {
struct buffer_head *bh = NULL;
struct msdos_dir_entry *de ;
sector_t blocknr;
int offset;
fat_get_blknr_offset(MSDOS_SB(sb), i_pos, &blocknr, &offset);
bh = sb_bread(sb, blocknr);
if (!bh) {
fat_msg(sb, KERN_ERR,
"unable to read block(%llu) for building NFS inode",
(llu)blocknr);
return inode;
}
de = (struct msdos_dir_entry *)bh->b_data;
/* If a file is deleted on server and client is not updated
* yet, we must not build the inode upon a lookup call.
*/
if (IS_FREE(de[offset].name))
inode = NULL;
else
inode = fat_build_inode(sb, &de[offset], i_pos);
brelse(bh);
}
return inode;
fat (exportfs): move NFS support code Under memory pressure, the system may evict dentries from cache. When the FAT driver receives a NFS request involving an evicted dentry, it is unable to reconnect it to the filesystem root. This causes the request to fail, often with ENOENT. This is partially due to ineffectiveness of the current FAT NFS implementation, and partially due to an unimplemented fh_to_parent method. The latter can cause file accesses to fail on shares exported with subtree_check. This patch set provides the FAT driver with the ability to reconnect dentries. NFS file handle generation and lookups are simplified and made congruent with ext2. Testing has involved a memory-starved virtual machine running 3.5-rc5 that exports a ~2 GB vfat filesystem containing a kernel tree (~770 MB, ~40000 files, 9 levels). Both 'cp -r' and 'ls -lR' operations were performed from a client, some overlapping, some consecutive. Exports with 'subtree_check' and 'no_subtree_check' have been tested. Note that while this patch set improves FAT's NFS support, it does not eliminate ESTALE errors completely. The following should be considered for NFS clients who are sensitive to ESTALE: * Mounting with lookupcache=none Unfortunately this can degrade performance severely, particularly for deep filesystems. * Incorporating VFS patches to retry ESTALE failures on the client-side, such as https://lkml.org/lkml/2012/6/29/381 * Handling ESTALE errors in client application code This patch: Move NFS-related code into its own C file. No functional changes. Signed-off-by: Steven J. Magnani <steve@digidescorp.com> Acked-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-10-05 00:14:44 +00:00
}
static struct inode *fat_nfs_get_inode(struct super_block *sb,
u64 ino, u32 generation)
{
return __fat_nfs_get_inode(sb, ino, generation, 0);
}
static int
fat_encode_fh_nostale(struct inode *inode, __u32 *fh, int *lenp,
struct inode *parent)
{
int len = *lenp;
struct msdos_sb_info *sbi = MSDOS_SB(inode->i_sb);
struct fat_fid *fid = (struct fat_fid *) fh;
loff_t i_pos;
int type = FILEID_FAT_WITHOUT_PARENT;
if (parent) {
if (len < FAT_FID_SIZE_WITH_PARENT) {
*lenp = FAT_FID_SIZE_WITH_PARENT;
return FILEID_INVALID;
}
} else {
if (len < FAT_FID_SIZE_WITHOUT_PARENT) {
*lenp = FAT_FID_SIZE_WITHOUT_PARENT;
return FILEID_INVALID;
}
}
i_pos = fat_i_pos_read(sbi, inode);
*lenp = FAT_FID_SIZE_WITHOUT_PARENT;
fid->i_gen = inode->i_generation;
fid->i_pos_low = i_pos & 0xFFFFFFFF;
fid->i_pos_hi = (i_pos >> 32) & 0xFFFF;
if (parent) {
i_pos = fat_i_pos_read(sbi, parent);
fid->parent_i_pos_hi = (i_pos >> 32) & 0xFFFF;
fid->parent_i_pos_low = i_pos & 0xFFFFFFFF;
fid->parent_i_gen = parent->i_generation;
type = FILEID_FAT_WITH_PARENT;
*lenp = FAT_FID_SIZE_WITH_PARENT;
}
return type;
}
/*
fat (exportfs): move NFS support code Under memory pressure, the system may evict dentries from cache. When the FAT driver receives a NFS request involving an evicted dentry, it is unable to reconnect it to the filesystem root. This causes the request to fail, often with ENOENT. This is partially due to ineffectiveness of the current FAT NFS implementation, and partially due to an unimplemented fh_to_parent method. The latter can cause file accesses to fail on shares exported with subtree_check. This patch set provides the FAT driver with the ability to reconnect dentries. NFS file handle generation and lookups are simplified and made congruent with ext2. Testing has involved a memory-starved virtual machine running 3.5-rc5 that exports a ~2 GB vfat filesystem containing a kernel tree (~770 MB, ~40000 files, 9 levels). Both 'cp -r' and 'ls -lR' operations were performed from a client, some overlapping, some consecutive. Exports with 'subtree_check' and 'no_subtree_check' have been tested. Note that while this patch set improves FAT's NFS support, it does not eliminate ESTALE errors completely. The following should be considered for NFS clients who are sensitive to ESTALE: * Mounting with lookupcache=none Unfortunately this can degrade performance severely, particularly for deep filesystems. * Incorporating VFS patches to retry ESTALE failures on the client-side, such as https://lkml.org/lkml/2012/6/29/381 * Handling ESTALE errors in client application code This patch: Move NFS-related code into its own C file. No functional changes. Signed-off-by: Steven J. Magnani <steve@digidescorp.com> Acked-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-10-05 00:14:44 +00:00
* Map a NFS file handle to a corresponding dentry.
* The dentry may or may not be connected to the filesystem root.
*/
static struct dentry *fat_fh_to_dentry(struct super_block *sb, struct fid *fid,
fat (exportfs): move NFS support code Under memory pressure, the system may evict dentries from cache. When the FAT driver receives a NFS request involving an evicted dentry, it is unable to reconnect it to the filesystem root. This causes the request to fail, often with ENOENT. This is partially due to ineffectiveness of the current FAT NFS implementation, and partially due to an unimplemented fh_to_parent method. The latter can cause file accesses to fail on shares exported with subtree_check. This patch set provides the FAT driver with the ability to reconnect dentries. NFS file handle generation and lookups are simplified and made congruent with ext2. Testing has involved a memory-starved virtual machine running 3.5-rc5 that exports a ~2 GB vfat filesystem containing a kernel tree (~770 MB, ~40000 files, 9 levels). Both 'cp -r' and 'ls -lR' operations were performed from a client, some overlapping, some consecutive. Exports with 'subtree_check' and 'no_subtree_check' have been tested. Note that while this patch set improves FAT's NFS support, it does not eliminate ESTALE errors completely. The following should be considered for NFS clients who are sensitive to ESTALE: * Mounting with lookupcache=none Unfortunately this can degrade performance severely, particularly for deep filesystems. * Incorporating VFS patches to retry ESTALE failures on the client-side, such as https://lkml.org/lkml/2012/6/29/381 * Handling ESTALE errors in client application code This patch: Move NFS-related code into its own C file. No functional changes. Signed-off-by: Steven J. Magnani <steve@digidescorp.com> Acked-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-10-05 00:14:44 +00:00
int fh_len, int fh_type)
{
return generic_fh_to_dentry(sb, fid, fh_len, fh_type,
fat_nfs_get_inode);
}
fat (exportfs): move NFS support code Under memory pressure, the system may evict dentries from cache. When the FAT driver receives a NFS request involving an evicted dentry, it is unable to reconnect it to the filesystem root. This causes the request to fail, often with ENOENT. This is partially due to ineffectiveness of the current FAT NFS implementation, and partially due to an unimplemented fh_to_parent method. The latter can cause file accesses to fail on shares exported with subtree_check. This patch set provides the FAT driver with the ability to reconnect dentries. NFS file handle generation and lookups are simplified and made congruent with ext2. Testing has involved a memory-starved virtual machine running 3.5-rc5 that exports a ~2 GB vfat filesystem containing a kernel tree (~770 MB, ~40000 files, 9 levels). Both 'cp -r' and 'ls -lR' operations were performed from a client, some overlapping, some consecutive. Exports with 'subtree_check' and 'no_subtree_check' have been tested. Note that while this patch set improves FAT's NFS support, it does not eliminate ESTALE errors completely. The following should be considered for NFS clients who are sensitive to ESTALE: * Mounting with lookupcache=none Unfortunately this can degrade performance severely, particularly for deep filesystems. * Incorporating VFS patches to retry ESTALE failures on the client-side, such as https://lkml.org/lkml/2012/6/29/381 * Handling ESTALE errors in client application code This patch: Move NFS-related code into its own C file. No functional changes. Signed-off-by: Steven J. Magnani <steve@digidescorp.com> Acked-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-10-05 00:14:44 +00:00
static struct dentry *fat_fh_to_dentry_nostale(struct super_block *sb,
struct fid *fh, int fh_len,
int fh_type)
{
struct inode *inode = NULL;
struct fat_fid *fid = (struct fat_fid *)fh;
loff_t i_pos;
switch (fh_type) {
case FILEID_FAT_WITHOUT_PARENT:
if (fh_len < FAT_FID_SIZE_WITHOUT_PARENT)
return NULL;
break;
case FILEID_FAT_WITH_PARENT:
if (fh_len < FAT_FID_SIZE_WITH_PARENT)
return NULL;
break;
default:
return NULL;
}
i_pos = fid->i_pos_hi;
i_pos = (i_pos << 32) | (fid->i_pos_low);
inode = __fat_nfs_get_inode(sb, 0, fid->i_gen, i_pos);
return d_obtain_alias(inode);
}
/*
* Find the parent for a file specified by NFS handle.
* This requires that the handle contain the i_ino of the parent.
*/
static struct dentry *fat_fh_to_parent(struct super_block *sb, struct fid *fid,
int fh_len, int fh_type)
{
return generic_fh_to_parent(sb, fid, fh_len, fh_type,
fat_nfs_get_inode);
fat (exportfs): move NFS support code Under memory pressure, the system may evict dentries from cache. When the FAT driver receives a NFS request involving an evicted dentry, it is unable to reconnect it to the filesystem root. This causes the request to fail, often with ENOENT. This is partially due to ineffectiveness of the current FAT NFS implementation, and partially due to an unimplemented fh_to_parent method. The latter can cause file accesses to fail on shares exported with subtree_check. This patch set provides the FAT driver with the ability to reconnect dentries. NFS file handle generation and lookups are simplified and made congruent with ext2. Testing has involved a memory-starved virtual machine running 3.5-rc5 that exports a ~2 GB vfat filesystem containing a kernel tree (~770 MB, ~40000 files, 9 levels). Both 'cp -r' and 'ls -lR' operations were performed from a client, some overlapping, some consecutive. Exports with 'subtree_check' and 'no_subtree_check' have been tested. Note that while this patch set improves FAT's NFS support, it does not eliminate ESTALE errors completely. The following should be considered for NFS clients who are sensitive to ESTALE: * Mounting with lookupcache=none Unfortunately this can degrade performance severely, particularly for deep filesystems. * Incorporating VFS patches to retry ESTALE failures on the client-side, such as https://lkml.org/lkml/2012/6/29/381 * Handling ESTALE errors in client application code This patch: Move NFS-related code into its own C file. No functional changes. Signed-off-by: Steven J. Magnani <steve@digidescorp.com> Acked-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-10-05 00:14:44 +00:00
}
static struct dentry *fat_fh_to_parent_nostale(struct super_block *sb,
struct fid *fh, int fh_len,
int fh_type)
{
struct inode *inode = NULL;
struct fat_fid *fid = (struct fat_fid *)fh;
loff_t i_pos;
if (fh_len < FAT_FID_SIZE_WITH_PARENT)
return NULL;
switch (fh_type) {
case FILEID_FAT_WITH_PARENT:
i_pos = fid->parent_i_pos_hi;
i_pos = (i_pos << 32) | (fid->parent_i_pos_low);
inode = __fat_nfs_get_inode(sb, 0, fid->parent_i_gen, i_pos);
break;
}
return d_obtain_alias(inode);
}
/*
* Rebuild the parent for a directory that is not connected
* to the filesystem root
*/
static
struct inode *fat_rebuild_parent(struct super_block *sb, int parent_logstart)
{
int search_clus, clus_to_match;
struct msdos_dir_entry *de;
struct inode *parent = NULL;
struct inode *dummy_grand_parent = NULL;
struct fat_slot_info sinfo;
struct msdos_sb_info *sbi = MSDOS_SB(sb);
sector_t blknr = fat_clus_to_blknr(sbi, parent_logstart);
struct buffer_head *parent_bh = sb_bread(sb, blknr);
if (!parent_bh) {
fat_msg(sb, KERN_ERR,
"unable to read cluster of parent directory");
return NULL;
}
de = (struct msdos_dir_entry *) parent_bh->b_data;
clus_to_match = fat_get_start(sbi, &de[0]);
search_clus = fat_get_start(sbi, &de[1]);
dummy_grand_parent = fat_dget(sb, search_clus);
if (!dummy_grand_parent) {
dummy_grand_parent = new_inode(sb);
if (!dummy_grand_parent) {
brelse(parent_bh);
return parent;
}
dummy_grand_parent->i_ino = iunique(sb, MSDOS_ROOT_INO);
fat_fill_inode(dummy_grand_parent, &de[1]);
MSDOS_I(dummy_grand_parent)->i_pos = -1;
}
if (!fat_scan_logstart(dummy_grand_parent, clus_to_match, &sinfo))
parent = fat_build_inode(sb, sinfo.de, sinfo.i_pos);
brelse(parent_bh);
iput(dummy_grand_parent);
return parent;
}
fat (exportfs): move NFS support code Under memory pressure, the system may evict dentries from cache. When the FAT driver receives a NFS request involving an evicted dentry, it is unable to reconnect it to the filesystem root. This causes the request to fail, often with ENOENT. This is partially due to ineffectiveness of the current FAT NFS implementation, and partially due to an unimplemented fh_to_parent method. The latter can cause file accesses to fail on shares exported with subtree_check. This patch set provides the FAT driver with the ability to reconnect dentries. NFS file handle generation and lookups are simplified and made congruent with ext2. Testing has involved a memory-starved virtual machine running 3.5-rc5 that exports a ~2 GB vfat filesystem containing a kernel tree (~770 MB, ~40000 files, 9 levels). Both 'cp -r' and 'ls -lR' operations were performed from a client, some overlapping, some consecutive. Exports with 'subtree_check' and 'no_subtree_check' have been tested. Note that while this patch set improves FAT's NFS support, it does not eliminate ESTALE errors completely. The following should be considered for NFS clients who are sensitive to ESTALE: * Mounting with lookupcache=none Unfortunately this can degrade performance severely, particularly for deep filesystems. * Incorporating VFS patches to retry ESTALE failures on the client-side, such as https://lkml.org/lkml/2012/6/29/381 * Handling ESTALE errors in client application code This patch: Move NFS-related code into its own C file. No functional changes. Signed-off-by: Steven J. Magnani <steve@digidescorp.com> Acked-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-10-05 00:14:44 +00:00
/*
* Find the parent for a directory that is not currently connected to
* the filesystem root.
*
* On entry, the caller holds d_inode(child_dir)->i_mutex.
fat (exportfs): move NFS support code Under memory pressure, the system may evict dentries from cache. When the FAT driver receives a NFS request involving an evicted dentry, it is unable to reconnect it to the filesystem root. This causes the request to fail, often with ENOENT. This is partially due to ineffectiveness of the current FAT NFS implementation, and partially due to an unimplemented fh_to_parent method. The latter can cause file accesses to fail on shares exported with subtree_check. This patch set provides the FAT driver with the ability to reconnect dentries. NFS file handle generation and lookups are simplified and made congruent with ext2. Testing has involved a memory-starved virtual machine running 3.5-rc5 that exports a ~2 GB vfat filesystem containing a kernel tree (~770 MB, ~40000 files, 9 levels). Both 'cp -r' and 'ls -lR' operations were performed from a client, some overlapping, some consecutive. Exports with 'subtree_check' and 'no_subtree_check' have been tested. Note that while this patch set improves FAT's NFS support, it does not eliminate ESTALE errors completely. The following should be considered for NFS clients who are sensitive to ESTALE: * Mounting with lookupcache=none Unfortunately this can degrade performance severely, particularly for deep filesystems. * Incorporating VFS patches to retry ESTALE failures on the client-side, such as https://lkml.org/lkml/2012/6/29/381 * Handling ESTALE errors in client application code This patch: Move NFS-related code into its own C file. No functional changes. Signed-off-by: Steven J. Magnani <steve@digidescorp.com> Acked-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-10-05 00:14:44 +00:00
*/
static struct dentry *fat_get_parent(struct dentry *child_dir)
fat (exportfs): move NFS support code Under memory pressure, the system may evict dentries from cache. When the FAT driver receives a NFS request involving an evicted dentry, it is unable to reconnect it to the filesystem root. This causes the request to fail, often with ENOENT. This is partially due to ineffectiveness of the current FAT NFS implementation, and partially due to an unimplemented fh_to_parent method. The latter can cause file accesses to fail on shares exported with subtree_check. This patch set provides the FAT driver with the ability to reconnect dentries. NFS file handle generation and lookups are simplified and made congruent with ext2. Testing has involved a memory-starved virtual machine running 3.5-rc5 that exports a ~2 GB vfat filesystem containing a kernel tree (~770 MB, ~40000 files, 9 levels). Both 'cp -r' and 'ls -lR' operations were performed from a client, some overlapping, some consecutive. Exports with 'subtree_check' and 'no_subtree_check' have been tested. Note that while this patch set improves FAT's NFS support, it does not eliminate ESTALE errors completely. The following should be considered for NFS clients who are sensitive to ESTALE: * Mounting with lookupcache=none Unfortunately this can degrade performance severely, particularly for deep filesystems. * Incorporating VFS patches to retry ESTALE failures on the client-side, such as https://lkml.org/lkml/2012/6/29/381 * Handling ESTALE errors in client application code This patch: Move NFS-related code into its own C file. No functional changes. Signed-off-by: Steven J. Magnani <steve@digidescorp.com> Acked-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-10-05 00:14:44 +00:00
{
struct super_block *sb = child_dir->d_sb;
struct buffer_head *bh = NULL;
struct msdos_dir_entry *de;
struct inode *parent_inode = NULL;
struct msdos_sb_info *sbi = MSDOS_SB(sb);
fat (exportfs): move NFS support code Under memory pressure, the system may evict dentries from cache. When the FAT driver receives a NFS request involving an evicted dentry, it is unable to reconnect it to the filesystem root. This causes the request to fail, often with ENOENT. This is partially due to ineffectiveness of the current FAT NFS implementation, and partially due to an unimplemented fh_to_parent method. The latter can cause file accesses to fail on shares exported with subtree_check. This patch set provides the FAT driver with the ability to reconnect dentries. NFS file handle generation and lookups are simplified and made congruent with ext2. Testing has involved a memory-starved virtual machine running 3.5-rc5 that exports a ~2 GB vfat filesystem containing a kernel tree (~770 MB, ~40000 files, 9 levels). Both 'cp -r' and 'ls -lR' operations were performed from a client, some overlapping, some consecutive. Exports with 'subtree_check' and 'no_subtree_check' have been tested. Note that while this patch set improves FAT's NFS support, it does not eliminate ESTALE errors completely. The following should be considered for NFS clients who are sensitive to ESTALE: * Mounting with lookupcache=none Unfortunately this can degrade performance severely, particularly for deep filesystems. * Incorporating VFS patches to retry ESTALE failures on the client-side, such as https://lkml.org/lkml/2012/6/29/381 * Handling ESTALE errors in client application code This patch: Move NFS-related code into its own C file. No functional changes. Signed-off-by: Steven J. Magnani <steve@digidescorp.com> Acked-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-10-05 00:14:44 +00:00
if (!fat_get_dotdot_entry(d_inode(child_dir), &bh, &de)) {
int parent_logstart = fat_get_start(sbi, de);
parent_inode = fat_dget(sb, parent_logstart);
if (!parent_inode && sbi->options.nfs == FAT_NFS_NOSTALE_RO)
parent_inode = fat_rebuild_parent(sb, parent_logstart);
fat (exportfs): move NFS support code Under memory pressure, the system may evict dentries from cache. When the FAT driver receives a NFS request involving an evicted dentry, it is unable to reconnect it to the filesystem root. This causes the request to fail, often with ENOENT. This is partially due to ineffectiveness of the current FAT NFS implementation, and partially due to an unimplemented fh_to_parent method. The latter can cause file accesses to fail on shares exported with subtree_check. This patch set provides the FAT driver with the ability to reconnect dentries. NFS file handle generation and lookups are simplified and made congruent with ext2. Testing has involved a memory-starved virtual machine running 3.5-rc5 that exports a ~2 GB vfat filesystem containing a kernel tree (~770 MB, ~40000 files, 9 levels). Both 'cp -r' and 'ls -lR' operations were performed from a client, some overlapping, some consecutive. Exports with 'subtree_check' and 'no_subtree_check' have been tested. Note that while this patch set improves FAT's NFS support, it does not eliminate ESTALE errors completely. The following should be considered for NFS clients who are sensitive to ESTALE: * Mounting with lookupcache=none Unfortunately this can degrade performance severely, particularly for deep filesystems. * Incorporating VFS patches to retry ESTALE failures on the client-side, such as https://lkml.org/lkml/2012/6/29/381 * Handling ESTALE errors in client application code This patch: Move NFS-related code into its own C file. No functional changes. Signed-off-by: Steven J. Magnani <steve@digidescorp.com> Acked-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-10-05 00:14:44 +00:00
}
brelse(bh);
return d_obtain_alias(parent_inode);
fat (exportfs): move NFS support code Under memory pressure, the system may evict dentries from cache. When the FAT driver receives a NFS request involving an evicted dentry, it is unable to reconnect it to the filesystem root. This causes the request to fail, often with ENOENT. This is partially due to ineffectiveness of the current FAT NFS implementation, and partially due to an unimplemented fh_to_parent method. The latter can cause file accesses to fail on shares exported with subtree_check. This patch set provides the FAT driver with the ability to reconnect dentries. NFS file handle generation and lookups are simplified and made congruent with ext2. Testing has involved a memory-starved virtual machine running 3.5-rc5 that exports a ~2 GB vfat filesystem containing a kernel tree (~770 MB, ~40000 files, 9 levels). Both 'cp -r' and 'ls -lR' operations were performed from a client, some overlapping, some consecutive. Exports with 'subtree_check' and 'no_subtree_check' have been tested. Note that while this patch set improves FAT's NFS support, it does not eliminate ESTALE errors completely. The following should be considered for NFS clients who are sensitive to ESTALE: * Mounting with lookupcache=none Unfortunately this can degrade performance severely, particularly for deep filesystems. * Incorporating VFS patches to retry ESTALE failures on the client-side, such as https://lkml.org/lkml/2012/6/29/381 * Handling ESTALE errors in client application code This patch: Move NFS-related code into its own C file. No functional changes. Signed-off-by: Steven J. Magnani <steve@digidescorp.com> Acked-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-10-05 00:14:44 +00:00
}
const struct export_operations fat_export_ops = {
.encode_fh = generic_encode_ino32_fh,
.fh_to_dentry = fat_fh_to_dentry,
.fh_to_parent = fat_fh_to_parent,
.get_parent = fat_get_parent,
};
const struct export_operations fat_export_ops_nostale = {
.encode_fh = fat_encode_fh_nostale,
.fh_to_dentry = fat_fh_to_dentry_nostale,
.fh_to_parent = fat_fh_to_parent_nostale,
.get_parent = fat_get_parent,
};