leandrof/linux
1
0
Fork 0
forked from Minki/linux
Code Issues Pull Requests Packages Projects Releases Wiki Activity

AFS development

Browse Source 
-----BEGIN PGP SIGNATURE-----
 
 iQIVAwUAWgm9V/Sw1s6N8H32AQK5mQ//QGUDZLXsUPCtq0XJq0V+r4MUjNp9tCZR
 htiuNrEkHSyPpYgCcQ2Aqdl9kndwVXcE7lWT99mp/a0zwNAsp9GOGVhCXUd5R86G
 XlrBuUYVvBJk18tDsUNWdjRQ0gMHgQSlEnEbsaGiU1bVrpXatI9hL8qoeO78Iy7+
 eaJUQLCuCVJq7qMQGhC0hg338vmHVeYhnViXIxq+HFjsMmR9IVanuK+sQr6NSJxS
 F6RkPxBUPWkRVMHmxTLWj/XSHZwtwu+Mnc/UFYsAPLKEbY0cIohsI8EgfE8U7geU
 yRVnu3MIOXUXUrZizj9SwVYWdJfneRlINqMbHIO8QXMKR38tnQ0C2/7bgBsXiNPv
 YdiAyeqL4nM+JthV/rgA3hWgupwBlSb4ubclTphDNxMs5MBIUIK3XUt9GOXDDUZz
 2FT/FdrphM2UORaI2AEOi4Q0/nHdin+3rld8fjV0Ree/TPNXwcrOmvy8yGnxFCEp
 5b7YLwKrffZGnnS965dhZlnFR6hjndmzFgHdyRrJwc80hXi1Q/+W4F19MoYkkoVK
 G/gLvD3FbmygmFnjCik9TjUrro6vQxo56H/TuWgHTvYriNGH+D/D7EGUwg4GiXZZ
 +7vrNw660uXmZiu9i0YacCRyD8lvm7QpmWLb+uHwzfsBE1+C8UetyQ+egSWVdWJO
 KwPspygWXD4=
 =3vy0
 -----END PGP SIGNATURE-----

Merge tag 'afs-next-20171113' of git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs

Pull AFS updates from David Howells:
 "kAFS filesystem driver overhaul.

  The major points of the overhaul are:

   (1) Preliminary groundwork is laid for supporting network-namespacing
       of kAFS. The remainder of the namespacing work requires some way
       to pass namespace information to submounts triggered by an
       automount. This requires something like the mount overhaul that's
       in progress.

   (2) sockaddr_rxrpc is used in preference to in_addr for holding
       addresses internally and add support for talking to the YFS VL
       server. With this, kAFS can do everything over IPv6 as well as
       IPv4 if it's talking to servers that support it.

   (3) Callback handling is overhauled to be generally passive rather
       than active. 'Callbacks' are promises by the server to tell us
       about data and metadata changes. Callbacks are now checked when
       we next touch an inode rather than actively going and looking for
       it where possible.

   (4) File access permit caching is overhauled to store the caching
       information per-inode rather than per-directory, shared over
       subordinate files. Whilst older AFS servers only allow ACLs on
       directories (shared to the files in that directory), newer AFS
       servers break that restriction.

       To improve memory usage and to make it easier to do mass-key
       removal, permit combinations are cached and shared.

   (5) Cell database management is overhauled to allow lighter locks to
       be used and to make cell records autonomous state machines that
       look after getting their own DNS records and cleaning themselves
       up, in particular preventing races in acquiring and relinquishing
       the fscache token for the cell.

   (6) Volume caching is overhauled. The afs_vlocation record is got rid
       of to simplify things and the superblock is now keyed on the cell
       and the numeric volume ID only. The volume record is tied to a
       superblock and normal superblock management is used to mediate
       the lifetime of the volume fscache token.

   (7) File server record caching is overhauled to make server records
       independent of cells and volumes. A server can be in multiple
       cells (in such a case, the administrator must make sure that the
       VL services for all cells correctly reflect the volumes shared
       between those cells).

       Server records are now indexed using the UUID of the server
       rather than the address since a server can have multiple
       addresses.

   (8) File server rotation is overhauled to handle VMOVED, VBUSY (and
       similar), VOFFLINE and VNOVOL indications and to handle rotation
       both of servers and addresses of those servers. The rotation will
       also wait and retry if the server says it is busy.

   (9) Data writeback is overhauled. Each inode no longer stores a list
       of modified sections tagged with the key that authorised it in
       favour of noting the modified region of a page in page->private
       and storing a list of keys that made modifications in the inode.

       This simplifies things and allows other keys to be used to
       actually write to the server if a key that made a modification
       becomes useless.

  (10) Writable mmap() is implemented. This allows a kernel to be build
       entirely on AFS.

  Note that Pre AFS-3.4 servers are no longer supported, though this can
  be added back if necessary (AFS-3.4 was released in 1998)"

* tag 'afs-next-20171113' of git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs: (35 commits)
  afs: Protect call->state changes against signals
  afs: Trace page dirty/clean
  afs: Implement shared-writeable mmap
  afs: Get rid of the afs_writeback record
  afs: Introduce a file-private data record
  afs: Use a dynamic port if 7001 is in use
  afs: Fix directory read/modify race
  afs: Trace the sending of pages
  afs: Trace the initiation and completion of client calls
  afs: Fix documentation on # vs % prefix in mount source specification
  afs: Fix total-length calculation for multiple-page send
  afs: Only progress call state at end of Tx phase from rxrpc callback
  afs: Make use of the YFS service upgrade to fully support IPv6
  afs: Overhaul volume and server record caching and fileserver rotation
  afs: Move server rotation code into its own file
  afs: Add an address list concept
  afs: Overhaul cell database management
  afs: Overhaul permit caching
  afs: Overhaul the callback handling
  afs: Rename struct afs_call server member to cm_server
  ...
This commit is contained in:
Linus Torvalds 2017-11-16 11:41:22 -08:00
commit 487e2c9f44
47 changed files with 6875 additions and 5430 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
Documentation/filesystems/afs.txt
View File

@ -91,8 +91,8 @@ Filesystems can be mounted anywhere by commands similar to the following:
mount -t afs "#root.cell." /afs/cambridge
Where the initial character is either a hash or a percent symbol depending on
whether you definitely want a R/W volume (hash) or whether you'd prefer a R/O
volume, but are willing to use a R/W volume instead (percent).
whether you definitely want a R/W volume (percent) or whether you'd prefer a
R/O volume, but are willing to use a R/W volume instead (hash).
The name of the volume can be suffixes with ".backup" or ".readonly" to
specify connection to only volumes of those types.

14
arch/mips/kernel/traps.c
View File

@ -1233,18 +1233,6 @@ static int default_cu2_call(struct notifier_block *nfb, unsigned long action,
return NOTIFY_OK;
}
static int wait_on_fp_mode_switch(atomic_t *p)
{
/*
* The FP mode for this task is currently being switched. That may
* involve modifications to the format of this tasks FP context which
* make it unsafe to proceed with execution for the moment. Instead,
* schedule some other task.
*/
schedule();
return 0;
}
static int enable_restore_fp_context(int msa)
{
int err, was_fpu_owner, prior_msa;
@ -1254,7 +1242,7 @@ static int enable_restore_fp_context(int msa)
* complete before proceeding.
*/
wait_on_atomic_t(&current->mm->context.fp_mode_switching,
wait_on_fp_mode_switch, TASK_KILLABLE);
atomic_t_wait, TASK_KILLABLE);
if (!used_math()) {
/* First time FP context user. */

8
drivers/gpu/drm/drm_dp_aux_dev.c
View File

@ -263,12 +263,6 @@ static struct drm_dp_aux_dev *drm_dp_aux_dev_get_by_aux(struct drm_dp_aux *aux)
return aux_dev;
}
static int auxdev_wait_atomic_t(atomic_t *p)
{
schedule();
return 0;
}
void drm_dp_aux_unregister_devnode(struct drm_dp_aux *aux)
{
struct drm_dp_aux_dev *aux_dev;
@ -283,7 +277,7 @@ void drm_dp_aux_unregister_devnode(struct drm_dp_aux *aux)
mutex_unlock(&aux_idr_mutex);
atomic_dec(&aux_dev->usecount);
wait_on_atomic_t(&aux_dev->usecount, auxdev_wait_atomic_t,
wait_on_atomic_t(&aux_dev->usecount, atomic_t_wait,
TASK_UNINTERRUPTIBLE);
minor = aux_dev->index;

10
drivers/gpu/drm/i915/selftests/intel_breadcrumbs.c
View File

@ -271,13 +271,7 @@ struct igt_wakeup {
u32 seqno;
};
static int wait_atomic(atomic_t *p)
{
schedule();
return 0;
}
static int wait_atomic_timeout(atomic_t *p)
static int wait_atomic_timeout(atomic_t *p, unsigned int mode)
{
return schedule_timeout(10 * HZ) ? 0 : -ETIMEDOUT;
}
@ -348,7 +342,7 @@ static void igt_wake_all_sync(atomic_t *ready,
atomic_set(ready, 0);
wake_up_all(wq);
wait_on_atomic_t(set, wait_atomic, TASK_UNINTERRUPTIBLE);
wait_on_atomic_t(set, atomic_t_wait, TASK_UNINTERRUPTIBLE);
atomic_set(ready, count);
atomic_set(done, count);
}

8
drivers/media/platform/qcom/venus/hfi.c
View File

@ -88,12 +88,6 @@ unlock:
return ret;
}
static int core_deinit_wait_atomic_t(atomic_t *p)
{
schedule();
return 0;
}
int hfi_core_deinit(struct venus_core *core, bool blocking)
{
int ret = 0, empty;
@ -112,7 +106,7 @@ int hfi_core_deinit(struct venus_core *core, bool blocking)
if (!empty) {
mutex_unlock(&core->lock);
wait_on_atomic_t(&core->insts_count, core_deinit_wait_atomic_t,
wait_on_atomic_t(&core->insts_count, atomic_t_wait,
TASK_UNINTERRUPTIBLE);
mutex_lock(&core->lock);
}

5
fs/afs/Makefile
View File

@ -7,6 +7,7 @@ afs-cache-$(CONFIG_AFS_FSCACHE) := cache.o
kafs-objs := \
$(afs-cache-y) \
addr_list.o \
callback.o \
cell.o \
cmservice.o \
@ -19,14 +20,14 @@ kafs-objs := \
misc.o \
mntpt.o \
proc.o \
rotate.o \
rxrpc.o \
security.o \
server.o \
server_list.o \
super.o \
netdevices.o \
vlclient.o \
vlocation.o \
vnode.o \
volume.o \
write.o \
xattr.o

381
fs/afs/addr_list.c Normal file
View File

@ -0,0 +1,381 @@
/* Server address list management
*
* Copyright (C) 2017 Red Hat, Inc. All Rights Reserved.
* Written by David Howells (dhowells@redhat.com)
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public Licence
* as published by the Free Software Foundation; either version
* 2 of the Licence, or (at your option) any later version.
*/
#include <linux/slab.h>
#include <linux/ctype.h>
#include <linux/dns_resolver.h>
#include <linux/inet.h>
#include <keys/rxrpc-type.h>
#include "internal.h"
#include "afs_fs.h"
//#define AFS_MAX_ADDRESSES
// ((unsigned int)((PAGE_SIZE - sizeof(struct afs_addr_list)) /
// sizeof(struct sockaddr_rxrpc)))
#define AFS_MAX_ADDRESSES ((unsigned int)(sizeof(unsigned long) * 8))
/*
* Release an address list.
*/
void afs_put_addrlist(struct afs_addr_list *alist)
{
if (alist && refcount_dec_and_test(&alist->usage))
call_rcu(&alist->rcu, (rcu_callback_t)kfree);
}
/*
* Allocate an address list.
*/
struct afs_addr_list *afs_alloc_addrlist(unsigned int nr,
unsigned short service,
unsigned short port)
{
struct afs_addr_list *alist;
unsigned int i;
_enter("%u,%u,%u", nr, service, port);
alist = kzalloc(sizeof(*alist) + sizeof(alist->addrs[0]) * nr,
GFP_KERNEL);
if (!alist)
return NULL;
refcount_set(&alist->usage, 1);
for (i = 0; i < nr; i++) {
struct sockaddr_rxrpc *srx = &alist->addrs[i];
srx->srx_family = AF_RXRPC;
srx->srx_service = service;
srx->transport_type = SOCK_DGRAM;
srx->transport_len = sizeof(srx->transport.sin6);
srx->transport.sin6.sin6_family = AF_INET6;
srx->transport.sin6.sin6_port = htons(port);
}
return alist;
}
/*
* Parse a text string consisting of delimited addresses.
*/
struct afs_addr_list *afs_parse_text_addrs(const char *text, size_t len,
char delim,
unsigned short service,
unsigned short port)
{
struct afs_addr_list *alist;
const char *p, *end = text + len;
unsigned int nr = 0;
_enter("%*.*s,%c", (int)len, (int)len, text, delim);
if (!len)
return ERR_PTR(-EDESTADDRREQ);
if (delim == ':' && (memchr(text, ',', len) || !memchr(text, '.', len)))
delim = ',';
/* Count the addresses */
p = text;
do {
if (!*p)
return ERR_PTR(-EINVAL);
if (*p == delim)
continue;
nr++;
if (*p == '[') {
p++;
if (p == end)
return ERR_PTR(-EINVAL);
p = memchr(p, ']', end - p);
if (!p)
return ERR_PTR(-EINVAL);
p++;
if (p >= end)
break;
}
p = memchr(p, delim, end - p);
if (!p)
break;
p++;
} while (p < end);
_debug("%u/%u addresses", nr, AFS_MAX_ADDRESSES);
if (nr > AFS_MAX_ADDRESSES)
nr = AFS_MAX_ADDRESSES;
alist = afs_alloc_addrlist(nr, service, port);
if (!alist)
return ERR_PTR(-ENOMEM);
/* Extract the addresses */
p = text;
do {
struct sockaddr_rxrpc *srx = &alist->addrs[alist->nr_addrs];
char tdelim = delim;
if (*p == delim) {
p++;
continue;
}
if (*p == '[') {
p++;
tdelim = ']';
}
if (in4_pton(p, end - p,
(u8 *)&srx->transport.sin6.sin6_addr.s6_addr32[3],
tdelim, &p)) {
srx->transport.sin6.sin6_addr.s6_addr32[0] = 0;
srx->transport.sin6.sin6_addr.s6_addr32[1] = 0;
srx->transport.sin6.sin6_addr.s6_addr32[2] = htonl(0xffff);
} else if (in6_pton(p, end - p,
srx->transport.sin6.sin6_addr.s6_addr,
tdelim, &p)) {
/* Nothing to do */
} else {
goto bad_address;
}
if (tdelim == ']') {
if (p == end || *p != ']')
goto bad_address;
p++;
}
if (p < end) {
if (*p == '+') {
/* Port number specification "+1234" */
unsigned int xport = 0;
p++;
if (p >= end || !isdigit(*p))
goto bad_address;
do {
xport *= 10;
xport += *p - '0';
if (xport > 65535)
goto bad_address;
p++;
} while (p < end && isdigit(*p));
srx->transport.sin6.sin6_port = htons(xport);
} else if (*p == delim) {
p++;
} else {
goto bad_address;
}
}
alist->nr_addrs++;
} while (p < end && alist->nr_addrs < AFS_MAX_ADDRESSES);
_leave(" = [nr %u]", alist->nr_addrs);
return alist;
bad_address:
kfree(alist);
return ERR_PTR(-EINVAL);
}
/*
* Compare old and new address lists to see if there's been any change.
* - How to do this in better than O(Nlog(N)) time?
* - We don't really want to sort the address list, but would rather take the
* list as we got it so as not to undo record rotation by the DNS server.
*/
#if 0
static int afs_cmp_addr_list(const struct afs_addr_list *a1,
const struct afs_addr_list *a2)
{
}
#endif
/*
* Perform a DNS query for VL servers and build a up an address list.
*/
struct afs_addr_list *afs_dns_query(struct afs_cell *cell, time64_t *_expiry)
{
struct afs_addr_list *alist;
char *vllist = NULL;
int ret;
_enter("%s", cell->name);
ret = dns_query("afsdb", cell->name, cell->name_len,
"ipv4", &vllist, _expiry);
if (ret < 0)
return ERR_PTR(ret);
alist = afs_parse_text_addrs(vllist, strlen(vllist), ',',
VL_SERVICE, AFS_VL_PORT);
if (IS_ERR(alist)) {
kfree(vllist);
if (alist != ERR_PTR(-ENOMEM))
pr_err("Failed to parse DNS data\n");
return alist;
}
kfree(vllist);
return alist;
}
/*
* Merge an IPv4 entry into a fileserver address list.
*/
void afs_merge_fs_addr4(struct afs_addr_list *alist, __be32 xdr, u16 port)
{
struct sockaddr_in6 *a;
__be16 xport = htons(port);
int i;
for (i = 0; i < alist->nr_ipv4; i++) {
a = &alist->addrs[i].transport.sin6;
if (xdr == a->sin6_addr.s6_addr32[3] &&
xport == a->sin6_port)
return;
if (xdr == a->sin6_addr.s6_addr32[3] &&
xport < a->sin6_port)
break;
if (xdr < a->sin6_addr.s6_addr32[3])
break;
}
if (i < alist->nr_addrs)
memmove(alist->addrs + i + 1,
alist->addrs + i,
sizeof(alist->addrs[0]) * (alist->nr_addrs - i));
a = &alist->addrs[i].transport.sin6;
a->sin6_port = xport;
a->sin6_addr.s6_addr32[0] = 0;
a->sin6_addr.s6_addr32[1] = 0;
a->sin6_addr.s6_addr32[2] = htonl(0xffff);
a->sin6_addr.s6_addr32[3] = xdr;
alist->nr_ipv4++;
alist->nr_addrs++;
}
/*
* Merge an IPv6 entry into a fileserver address list.
*/
void afs_merge_fs_addr6(struct afs_addr_list *alist, __be32 *xdr, u16 port)
{
struct sockaddr_in6 *a;
__be16 xport = htons(port);
int i, diff;
for (i = alist->nr_ipv4; i < alist->nr_addrs; i++) {
a = &alist->addrs[i].transport.sin6;
diff = memcmp(xdr, &a->sin6_addr, 16);
if (diff == 0 &&
xport == a->sin6_port)
return;
if (diff == 0 &&
xport < a->sin6_port)
break;
if (diff < 0)
break;
}
if (i < alist->nr_addrs)
memmove(alist->addrs + i + 1,
alist->addrs + i,
sizeof(alist->addrs[0]) * (alist->nr_addrs - i));
a = &alist->addrs[i].transport.sin6;
a->sin6_port = xport;
a->sin6_addr.s6_addr32[0] = xdr[0];
a->sin6_addr.s6_addr32[1] = xdr[1];
a->sin6_addr.s6_addr32[2] = xdr[2];
a->sin6_addr.s6_addr32[3] = xdr[3];
alist->nr_addrs++;
}
/*
* Get an address to try.
*/
bool afs_iterate_addresses(struct afs_addr_cursor *ac)
{
_enter("%hu+%hd", ac->start, (short)ac->index);
if (!ac->alist)
return false;
if (ac->begun) {
ac->index++;
if (ac->index == ac->alist->nr_addrs)
ac->index = 0;
if (ac->index == ac->start) {
ac->error = -EDESTADDRREQ;
return false;
}
}
ac->begun = true;
ac->responded = false;
ac->addr = &ac->alist->addrs[ac->index];
return true;
}
/*
* Release an address list cursor.
*/
int afs_end_cursor(struct afs_addr_cursor *ac)
{
if (ac->responded && ac->index != ac->start)
WRITE_ONCE(ac->alist->index, ac->index);
afs_put_addrlist(ac->alist);
ac->alist = NULL;
return ac->error;
}
/*
* Set the address cursor for iterating over VL servers.
*/
int afs_set_vl_cursor(struct afs_addr_cursor *ac, struct afs_cell *cell)
{
struct afs_addr_list *alist;
int ret;
if (!rcu_access_pointer(cell->vl_addrs)) {
ret = wait_on_bit(&cell->flags, AFS_CELL_FL_NO_LOOKUP_YET,
TASK_INTERRUPTIBLE);
if (ret < 0)
return ret;
if (!rcu_access_pointer(cell->vl_addrs) &&
ktime_get_real_seconds() < cell->dns_expiry)
return cell->error;
}
read_lock(&cell->vl_addrs_lock);
alist = rcu_dereference_protected(cell->vl_addrs,
lockdep_is_held(&cell->vl_addrs_lock));
if (alist->nr_addrs > 0)
afs_get_addrlist(alist);
else
alist = NULL;
read_unlock(&cell->vl_addrs_lock);
if (!alist)
return -EDESTADDRREQ;
ac->alist = alist;
ac->addr = NULL;
ac->start = READ_ONCE(alist->index);
ac->index = ac->start;
ac->error = 0;
ac->begun = false;
return 0;
}

35
fs/afs/afs.h
View File

@ -14,11 +14,14 @@
#include <linux/in.h>
#define AFS_MAXCELLNAME 64 /* maximum length of a cell name */
#define AFS_MAXVOLNAME 64 /* maximum length of a volume name */
#define AFSNAMEMAX 256 /* maximum length of a filename plus NUL */
#define AFSPATHMAX 1024 /* maximum length of a pathname plus NUL */
#define AFSOPAQUEMAX 1024 /* maximum length of an opaque field */
#define AFS_MAXCELLNAME 64 /* Maximum length of a cell name */
#define AFS_MAXVOLNAME 64 /* Maximum length of a volume name */
#define AFS_MAXNSERVERS 8 /* Maximum servers in a basic volume record */
#define AFS_NMAXNSERVERS 13 /* Maximum servers in a N/U-class volume record */
#define AFS_MAXTYPES 3 /* Maximum number of volume types */
#define AFSNAMEMAX 256 /* Maximum length of a filename plus NUL */
#define AFSPATHMAX 1024 /* Maximum length of a pathname plus NUL */
#define AFSOPAQUEMAX 1024 /* Maximum length of an opaque field */
typedef unsigned afs_volid_t;
typedef unsigned afs_vnodeid_t;
@ -72,6 +75,15 @@ struct afs_callback {
#define AFSCBMAX 50 /* maximum callbacks transferred per bulk op */
struct afs_uuid {
__be32 time_low; /* low part of timestamp */
__be16 time_mid; /* mid part of timestamp */
__be16 time_hi_and_version; /* high part of timestamp and version */
__s8 clock_seq_hi_and_reserved; /* clock seq hi and variant */
__s8 clock_seq_low; /* clock seq low */
__s8 node[6]; /* spatially unique node ID (MAC addr) */
};
/*
* AFS volume information
*/
@ -124,7 +136,6 @@ struct afs_file_status {
afs_access_t caller_access; /* access rights for authenticated caller */
afs_access_t anon_access; /* access rights for unauthenticated caller */
umode_t mode; /* UNIX mode */
struct afs_fid parent; /* parent dir ID for non-dirs only */
time_t mtime_client; /* last time client changed data */
time_t mtime_server; /* last time server changed data */
s32 lock_count; /* file lock count (0=UNLK -1=WRLCK +ve=#RDLCK */
@ -167,4 +178,16 @@ struct afs_volume_status {
#define AFS_BLOCK_SIZE 1024
/*
* XDR encoding of UUID in AFS.
*/
struct afs_uuid__xdr {
__be32 time_low;
__be32 time_mid;
__be32 time_hi_and_version;
__be32 clock_seq_hi_and_reserved;
__be32 clock_seq_low;
__be32 node[6];
};
#endif /* AFS_H */

6
fs/afs/afs_fs.h
View File

@ -37,9 +37,12 @@ enum AFS_FS_Operations {
FSLOOKUP = 161, /* AFS lookup file in directory */
FSFETCHDATA64 = 65537, /* AFS Fetch file data */
FSSTOREDATA64 = 65538, /* AFS Store file data */
FSGIVEUPALLCALLBACKS = 65539, /* AFS Give up all outstanding callbacks on a server */
FSGETCAPABILITIES = 65540, /* Probe and get the capabilities of a fileserver */
};
enum AFS_FS_Errors {
VRESTARTING = -100, /* Server is restarting */
VSALVAGE = 101, /* volume needs salvaging */
VNOVNODE = 102, /* no such file/dir (vnode) */
VNOVOL = 103, /* no such volume or volume unavailable */
@ -51,6 +54,9 @@ enum AFS_FS_Errors {
VOVERQUOTA = 109, /* volume's maximum quota exceeded */
VBUSY = 110, /* volume is temporarily unavailable */
VMOVED = 111, /* volume moved to new server - ask this FS where */
VIO = 112, /* I/O error in volume */
VSALVAGING = 113, /* Volume is being salvaged */
VRESTRICTED = 120, /* Volume is restricted from using */
};
#endif /* AFS_FS_H */

73
fs/afs/afs_vl.h
View File

@ -16,11 +16,17 @@
#define AFS_VL_PORT 7003 /* volume location service port */
#define VL_SERVICE 52 /* RxRPC service ID for the Volume Location service */
#define YFS_VL_SERVICE 2503 /* Service ID for AuriStor upgraded VL service */
enum AFSVL_Operations {
VLGETENTRYBYID = 503, /* AFS Get Cache Entry By ID operation ID */
VLGETENTRYBYNAME = 504, /* AFS Get Cache Entry By Name operation ID */
VLPROBE = 514, /* AFS Probe Volume Location Service operation ID */
VLGETENTRYBYID = 503, /* AFS Get VLDB entry by ID */
VLGETENTRYBYNAME = 504, /* AFS Get VLDB entry by name */
VLPROBE = 514, /* AFS probe VL service */
VLGETENTRYBYIDU = 526, /* AFS Get VLDB entry by ID (UUID-variant) */
VLGETENTRYBYNAMEU = 527, /* AFS Get VLDB entry by name (UUID-variant) */
VLGETADDRSU = 533, /* AFS Get addrs for fileserver */
YVLGETENDPOINTS = 64002, /* YFS Get endpoints for file/volume server */
VLGETCAPABILITIES = 65537, /* AFS Get server capabilities */
};
enum AFSVL_Errors {
@ -54,6 +60,19 @@ enum AFSVL_Errors {
AFSVL_NOMEM = 363547, /* malloc/realloc failed to alloc enough memory */
};
enum {
YFS_SERVER_INDEX = 0,
YFS_SERVER_UUID = 1,
YFS_SERVER_ENDPOINT = 2,
};
enum {
YFS_ENDPOINT_IPV4 = 0,
YFS_ENDPOINT_IPV6 = 1,
};
#define YFS_MAXENDPOINTS 16
/*
* maps to "struct vldbentry" in vvl-spec.pdf
*/
@ -74,11 +93,57 @@ struct afs_vldbentry {
struct in_addr addr; /* server address */
unsigned partition; /* partition ID on this server */
unsigned flags; /* server specific flags */
#define AFS_VLSF_NEWREPSITE 0x0001 /* unused */
#define AFS_VLSF_NEWREPSITE 0x0001 /* Ignore all 'non-new' servers */
#define AFS_VLSF_ROVOL 0x0002 /* this server holds a R/O instance of the volume */
#define AFS_VLSF_RWVOL 0x0004 /* this server holds a R/W instance of the volume */
#define AFS_VLSF_BACKVOL 0x0008 /* this server holds a backup instance of the volume */
#define AFS_VLSF_UUID 0x0010 /* This server is referred to by its UUID */
#define AFS_VLSF_DONTUSE 0x0020 /* This server ref should be ignored */
} servers[8];
};
#define AFS_VLDB_MAXNAMELEN 65
struct afs_ListAddrByAttributes__xdr {
__be32 Mask;
#define AFS_VLADDR_IPADDR 0x1 /* Match by ->ipaddr */
#define AFS_VLADDR_INDEX 0x2 /* Match by ->index */
#define AFS_VLADDR_UUID 0x4 /* Match by ->uuid */
__be32 ipaddr;
__be32 index;
__be32 spare;
struct afs_uuid__xdr uuid;
};
struct afs_uvldbentry__xdr {
__be32 name[AFS_VLDB_MAXNAMELEN];
__be32 nServers;
struct afs_uuid__xdr serverNumber[AFS_NMAXNSERVERS];
__be32 serverUnique[AFS_NMAXNSERVERS];
__be32 serverPartition[AFS_NMAXNSERVERS];
__be32 serverFlags[AFS_NMAXNSERVERS];
__be32 volumeId[AFS_MAXTYPES];
__be32 cloneId;
__be32 flags;
__be32 spares1;
__be32 spares2;
__be32 spares3;
__be32 spares4;
__be32 spares5;
__be32 spares6;
__be32 spares7;
__be32 spares8;
__be32 spares9;
};
struct afs_address_list {
refcount_t usage;
unsigned int version;
unsigned int nr_addrs;
struct sockaddr_rxrpc addrs[];
};
extern void afs_put_address_list(struct afs_address_list *alist);
#endif /* AFS_VL_H */

239
fs/afs/cache.c
View File

@ -14,19 +14,6 @@
static uint16_t afs_cell_cache_get_key(const void *cookie_netfs_data,
void *buffer, uint16_t buflen);
static uint16_t afs_cell_cache_get_aux(const void *cookie_netfs_data,
void *buffer, uint16_t buflen);
static enum fscache_checkaux afs_cell_cache_check_aux(void *cookie_netfs_data,
const void *buffer,
uint16_t buflen);
static uint16_t afs_vlocation_cache_get_key(const void *cookie_netfs_data,
void *buffer, uint16_t buflen);
static uint16_t afs_vlocation_cache_get_aux(const void *cookie_netfs_data,
void *buffer, uint16_t buflen);
static enum fscache_checkaux afs_vlocation_cache_check_aux(
void *cookie_netfs_data, const void *buffer, uint16_t buflen);
static uint16_t afs_volume_cache_get_key(const void *cookie_netfs_data,
void *buffer, uint16_t buflen);
@ -42,23 +29,13 @@ static enum fscache_checkaux afs_vnode_cache_check_aux(void *cookie_netfs_data,
struct fscache_netfs afs_cache_netfs = {
.name = "afs",
.version = 0,
.version = 1,
};
struct fscache_cookie_def afs_cell_cache_index_def = {
.name = "AFS.cell",
.type = FSCACHE_COOKIE_TYPE_INDEX,
.get_key = afs_cell_cache_get_key,
.get_aux = afs_cell_cache_get_aux,
.check_aux = afs_cell_cache_check_aux,
};
struct fscache_cookie_def afs_vlocation_cache_index_def = {
.name = "AFS.vldb",
.type = FSCACHE_COOKIE_TYPE_INDEX,
.get_key = afs_vlocation_cache_get_key,
.get_aux = afs_vlocation_cache_get_aux,
.check_aux = afs_vlocation_cache_check_aux,
};
struct fscache_cookie_def afs_volume_cache_index_def = {
@ -95,150 +72,26 @@ static uint16_t afs_cell_cache_get_key(const void *cookie_netfs_data,
return klen;
}
/*
* provide new auxiliary cache data
*/
static uint16_t afs_cell_cache_get_aux(const void *cookie_netfs_data,
void *buffer, uint16_t bufmax)
{
const struct afs_cell *cell = cookie_netfs_data;
uint16_t dlen;
_enter("%p,%p,%u", cell, buffer, bufmax);
dlen = cell->vl_naddrs * sizeof(cell->vl_addrs[0]);
dlen = min(dlen, bufmax);
dlen &= ~(sizeof(cell->vl_addrs[0]) - 1);
memcpy(buffer, cell->vl_addrs, dlen);
return dlen;
}
/*
* check that the auxiliary data indicates that the entry is still valid
*/
static enum fscache_checkaux afs_cell_cache_check_aux(void *cookie_netfs_data,
const void *buffer,
uint16_t buflen)
{
_leave(" = OKAY");
return FSCACHE_CHECKAUX_OKAY;
}
/*****************************************************************************/
/*
* set the key for the index entry
*/
static uint16_t afs_vlocation_cache_get_key(const void *cookie_netfs_data,
void *buffer, uint16_t bufmax)
{
const struct afs_vlocation *vlocation = cookie_netfs_data;
uint16_t klen;
_enter("{%s},%p,%u", vlocation->vldb.name, buffer, bufmax);
klen = strnlen(vlocation->vldb.name, sizeof(vlocation->vldb.name));
if (klen > bufmax)
return 0;
memcpy(buffer, vlocation->vldb.name, klen);
_leave(" = %u", klen);
return klen;
}
/*
* provide new auxiliary cache data
*/
static uint16_t afs_vlocation_cache_get_aux(const void *cookie_netfs_data,
void *buffer, uint16_t bufmax)
{
const struct afs_vlocation *vlocation = cookie_netfs_data;
uint16_t dlen;
_enter("{%s},%p,%u", vlocation->vldb.name, buffer, bufmax);
dlen = sizeof(struct afs_cache_vlocation);
dlen -= offsetof(struct afs_cache_vlocation, nservers);
if (dlen > bufmax)
return 0;
memcpy(buffer, (uint8_t *)&vlocation->vldb.nservers, dlen);
_leave(" = %u", dlen);
return dlen;
}
/*
* check that the auxiliary data indicates that the entry is still valid
*/
static
enum fscache_checkaux afs_vlocation_cache_check_aux(void *cookie_netfs_data,
const void *buffer,
uint16_t buflen)
{
const struct afs_cache_vlocation *cvldb;
struct afs_vlocation *vlocation = cookie_netfs_data;
uint16_t dlen;
_enter("{%s},%p,%u", vlocation->vldb.name, buffer, buflen);
/* check the size of the data is what we're expecting */
dlen = sizeof(struct afs_cache_vlocation);
dlen -= offsetof(struct afs_cache_vlocation, nservers);
if (dlen != buflen)
return FSCACHE_CHECKAUX_OBSOLETE;
cvldb = container_of(buffer, struct afs_cache_vlocation, nservers);
/* if what's on disk is more valid than what's in memory, then use the
* VL record from the cache */
if (!vlocation->valid || vlocation->vldb.rtime == cvldb->rtime) {
memcpy((uint8_t *)&vlocation->vldb.nservers, buffer, dlen);
vlocation->valid = 1;
_leave(" = SUCCESS [c->m]");
return FSCACHE_CHECKAUX_OKAY;
}
/* need to update the cache if the cached info differs */
if (memcmp(&vlocation->vldb, buffer, dlen) != 0) {
/* delete if the volume IDs for this name differ */
if (memcmp(&vlocation->vldb.vid, &cvldb->vid,
sizeof(cvldb->vid)) != 0
) {
_leave(" = OBSOLETE");
return FSCACHE_CHECKAUX_OBSOLETE;
}
_leave(" = UPDATE");
return FSCACHE_CHECKAUX_NEEDS_UPDATE;
}
_leave(" = OKAY");
return FSCACHE_CHECKAUX_OKAY;
}
/*****************************************************************************/
/*
* set the key for the volume index entry
*/
static uint16_t afs_volume_cache_get_key(const void *cookie_netfs_data,
void *buffer, uint16_t bufmax)
void *buffer, uint16_t bufmax)
{
const struct afs_volume *volume = cookie_netfs_data;
uint16_t klen;
struct {
u64 volid;
} __packed key;
_enter("{%u},%p,%u", volume->type, buffer, bufmax);
klen = sizeof(volume->type);
if (klen > bufmax)
if (bufmax < sizeof(key))
return 0;
memcpy(buffer, &volume->type, sizeof(volume->type));
_leave(" = %u", klen);
return klen;
key.volid = volume->vid;
memcpy(buffer, &key, sizeof(key));
return sizeof(key);
}
/*****************************************************************************/
@ -249,20 +102,25 @@ static uint16_t afs_vnode_cache_get_key(const void *cookie_netfs_data,
void *buffer, uint16_t bufmax)
{
const struct afs_vnode *vnode = cookie_netfs_data;
uint16_t klen;
struct {
u32 vnode_id[3];
} __packed key;
_enter("{%x,%x,%llx},%p,%u",
vnode->fid.vnode, vnode->fid.unique, vnode->status.data_version,
buffer, bufmax);
klen = sizeof(vnode->fid.vnode);
if (klen > bufmax)
/* Allow for a 96-bit key */
memset(&key, 0, sizeof(key));
key.vnode_id[0] = vnode->fid.vnode;
key.vnode_id[1] = 0;
key.vnode_id[2] = 0;
if (sizeof(key) > bufmax)
return 0;
memcpy(buffer, &vnode->fid.vnode, sizeof(vnode->fid.vnode));
_leave(" = %u", klen);
return klen;
memcpy(buffer, &key, sizeof(key));
return sizeof(key);
}
/*
@ -280,6 +138,11 @@ static void afs_vnode_cache_get_attr(const void *cookie_netfs_data,
*size = vnode->status.size;
}
struct afs_vnode_cache_aux {
u64 data_version;
u32 fid_unique;
} __packed;
/*
* provide new auxiliary cache data
*/
@ -287,23 +150,21 @@ static uint16_t afs_vnode_cache_get_aux(const void *cookie_netfs_data,
void *buffer, uint16_t bufmax)
{
const struct afs_vnode *vnode = cookie_netfs_data;
uint16_t dlen;
struct afs_vnode_cache_aux aux;
_enter("{%x,%x,%Lx},%p,%u",
vnode->fid.vnode, vnode->fid.unique, vnode->status.data_version,
buffer, bufmax);
dlen = sizeof(vnode->fid.unique) + sizeof(vnode->status.data_version);
if (dlen > bufmax)
memset(&aux, 0, sizeof(aux));
aux.data_version = vnode->status.data_version;
aux.fid_unique = vnode->fid.unique;
if (bufmax < sizeof(aux))
return 0;
memcpy(buffer, &vnode->fid.unique, sizeof(vnode->fid.unique));
buffer += sizeof(vnode->fid.unique);
memcpy(buffer, &vnode->status.data_version,
sizeof(vnode->status.data_version));
_leave(" = %u", dlen);
return dlen;
memcpy(buffer, &aux, sizeof(aux));
return sizeof(aux);
}
/*
@ -314,43 +175,29 @@ static enum fscache_checkaux afs_vnode_cache_check_aux(void *cookie_netfs_data,
uint16_t buflen)
{
struct afs_vnode *vnode = cookie_netfs_data;
uint16_t dlen;
struct afs_vnode_cache_aux aux;
_enter("{%x,%x,%llx},%p,%u",
vnode->fid.vnode, vnode->fid.unique, vnode->status.data_version,
buffer, buflen);
memcpy(&aux, buffer, sizeof(aux));
/* check the size of the data is what we're expecting */
dlen = sizeof(vnode->fid.unique) + sizeof(vnode->status.data_version);
if (dlen != buflen) {
_leave(" = OBSOLETE [len %hx != %hx]", dlen, buflen);
if (buflen != sizeof(aux)) {
_leave(" = OBSOLETE [len %hx != %zx]", buflen, sizeof(aux));
return FSCACHE_CHECKAUX_OBSOLETE;
}
if (memcmp(buffer,
&vnode->fid.unique,
sizeof(vnode->fid.unique)
) != 0) {
unsigned unique;
memcpy(&unique, buffer, sizeof(unique));
if (vnode->fid.unique != aux.fid_unique) {
_leave(" = OBSOLETE [uniq %x != %x]",
unique, vnode->fid.unique);
aux.fid_unique, vnode->fid.unique);
return FSCACHE_CHECKAUX_OBSOLETE;
}
if (memcmp(buffer + sizeof(vnode->fid.unique),
&vnode->status.data_version,
sizeof(vnode->status.data_version)
) != 0) {
afs_dataversion_t version;
memcpy(&version, buffer + sizeof(vnode->fid.unique),
sizeof(version));
if (vnode->status.data_version != aux.data_version) {
_leave(" = OBSOLETE [vers %llx != %llx]",
version, vnode->status.data_version);
aux.data_version, vnode->status.data_version);
return FSCACHE_CHECKAUX_OBSOLETE;
}

522
fs/afs/callback.c
View File

@ -20,118 +20,151 @@
#include <linux/sched.h>
#include "internal.h"
#if 0
unsigned afs_vnode_update_timeout = 10;
#endif /* 0 */
/*
* Set up an interest-in-callbacks record for a volume on a server and
* register it with the server.
* - Called with volume->server_sem held.
*/
int afs_register_server_cb_interest(struct afs_vnode *vnode,
struct afs_server_entry *entry)
{
struct afs_cb_interest *cbi = entry->cb_interest, *vcbi, *new, *x;
struct afs_server *server = entry->server;
#define afs_breakring_space(server) \
CIRC_SPACE((server)->cb_break_head, (server)->cb_break_tail, \
ARRAY_SIZE((server)->cb_break))
again:
vcbi = vnode->cb_interest;
if (vcbi) {
if (vcbi == cbi)
return 0;
//static void afs_callback_updater(struct work_struct *);
if (cbi && vcbi->server == cbi->server) {
write_seqlock(&vnode->cb_lock);
vnode->cb_interest = afs_get_cb_interest(cbi);
write_sequnlock(&vnode->cb_lock);
afs_put_cb_interest(afs_v2net(vnode), cbi);
return 0;
}
static struct workqueue_struct *afs_callback_update_worker;
if (!cbi && vcbi->server == server) {
afs_get_cb_interest(vcbi);
x = cmpxchg(&entry->cb_interest, cbi, vcbi);
if (x != cbi) {
cbi = x;
afs_put_cb_interest(afs_v2net(vnode), vcbi);
goto again;
}
return 0;
}
}
if (!cbi) {
new = kzalloc(sizeof(struct afs_cb_interest), GFP_KERNEL);
if (!new)
return -ENOMEM;
refcount_set(&new->usage, 1);
new->sb = vnode->vfs_inode.i_sb;
new->vid = vnode->volume->vid;
new->server = afs_get_server(server);
INIT_LIST_HEAD(&new->cb_link);
write_lock(&server->cb_break_lock);
list_add_tail(&new->cb_link, &server->cb_interests);
write_unlock(&server->cb_break_lock);
x = cmpxchg(&entry->cb_interest, cbi, new);
if (x == cbi) {
cbi = new;
} else {
cbi = x;
afs_put_cb_interest(afs_v2net(vnode), new);
}
}
ASSERT(cbi);
/* Change the server the vnode is using. This entails scrubbing any
* interest the vnode had in the previous server it was using.
*/
write_seqlock(&vnode->cb_lock);
vnode->cb_interest = afs_get_cb_interest(cbi);
vnode->cb_s_break = cbi->server->cb_s_break;
clear_bit(AFS_VNODE_CB_PROMISED, &vnode->flags);
write_sequnlock(&vnode->cb_lock);
return 0;
}
/*
* Set a vnode's interest on a server.
*/
void afs_set_cb_interest(struct afs_vnode *vnode, struct afs_cb_interest *cbi)
{
struct afs_cb_interest *old_cbi = NULL;
if (vnode->cb_interest == cbi)
return;
write_seqlock(&vnode->cb_lock);
if (vnode->cb_interest != cbi) {
afs_get_cb_interest(cbi);
old_cbi = vnode->cb_interest;
vnode->cb_interest = cbi;
}
write_sequnlock(&vnode->cb_lock);
afs_put_cb_interest(afs_v2net(vnode), cbi);
}
/*
* Remove an interest on a server.
*/
void afs_put_cb_interest(struct afs_net *net, struct afs_cb_interest *cbi)
{
if (cbi && refcount_dec_and_test(&cbi->usage)) {
if (!list_empty(&cbi->cb_link)) {
write_lock(&cbi->server->cb_break_lock);
list_del_init(&cbi->cb_link);
write_unlock(&cbi->server->cb_break_lock);
afs_put_server(net, cbi->server);
}
kfree(cbi);
}
}
/*
* allow the fileserver to request callback state (re-)initialisation
*/
void afs_init_callback_state(struct afs_server *server)
{
struct afs_vnode *vnode;
_enter("{%p}", server);
spin_lock(&server->cb_lock);
/* kill all the promises on record from this server */
while (!RB_EMPTY_ROOT(&server->cb_promises)) {
vnode = rb_entry(server->cb_promises.rb_node,
struct afs_vnode, cb_promise);
_debug("UNPROMISE { vid=%x:%u uq=%u}",
vnode->fid.vid, vnode->fid.vnode, vnode->fid.unique);
rb_erase(&vnode->cb_promise, &server->cb_promises);
vnode->cb_promised = false;
}
spin_unlock(&server->cb_lock);
_leave("");
}
/*
* handle the data invalidation side of a callback being broken
*/
void afs_broken_callback_work(struct work_struct *work)
{
struct afs_vnode *vnode =
container_of(work, struct afs_vnode, cb_broken_work);
_enter("");
if (test_bit(AFS_VNODE_DELETED, &vnode->flags))
return;
/* we're only interested in dealing with a broken callback on *this*
* vnode and only if no-one else has dealt with it yet */
if (!mutex_trylock(&vnode->validate_lock))
return; /* someone else is dealing with it */
if (test_bit(AFS_VNODE_CB_BROKEN, &vnode->flags)) {
if (S_ISDIR(vnode->vfs_inode.i_mode))
afs_clear_permits(vnode);
if (afs_vnode_fetch_status(vnode, NULL, NULL) < 0)
goto out;
if (test_bit(AFS_VNODE_DELETED, &vnode->flags))
goto out;
/* if the vnode's data version number changed then its contents
* are different */
if (test_and_clear_bit(AFS_VNODE_ZAP_DATA, &vnode->flags))
afs_zap_data(vnode);
}
out:
mutex_unlock(&vnode->validate_lock);
/* avoid the potential race whereby the mutex_trylock() in this
* function happens again between the clear_bit() and the
* mutex_unlock() */
if (test_bit(AFS_VNODE_CB_BROKEN, &vnode->flags)) {
_debug("requeue");
queue_work(afs_callback_update_worker, &vnode->cb_broken_work);
}
_leave("");
if (!test_and_clear_bit(AFS_SERVER_FL_NEW, &server->flags))
server->cb_s_break++;
}
/*
* actually break a callback
*/
static void afs_break_callback(struct afs_server *server,
struct afs_vnode *vnode)
void afs_break_callback(struct afs_vnode *vnode)
{
_enter("");
set_bit(AFS_VNODE_CB_BROKEN, &vnode->flags);
write_seqlock(&vnode->cb_lock);
if (test_and_clear_bit(AFS_VNODE_CB_PROMISED, &vnode->flags)) {
vnode->cb_break++;
afs_clear_permits(vnode);
if (vnode->cb_promised) {
spin_lock(&vnode->lock);
_debug("break callback");
spin_lock(&server->cb_lock);
if (vnode->cb_promised) {
rb_erase(&vnode->cb_promise, &server->cb_promises);
vnode->cb_promised = false;
}
spin_unlock(&server->cb_lock);
queue_work(afs_callback_update_worker, &vnode->cb_broken_work);
if (list_empty(&vnode->granted_locks) &&
!list_empty(&vnode->pending_locks))
afs_lock_may_be_available(vnode);
spin_unlock(&vnode->lock);
}
write_sequnlock(&vnode->cb_lock);
}
/*
@ -143,49 +176,31 @@ static void afs_break_callback(struct afs_server *server,
static void afs_break_one_callback(struct afs_server *server,
struct afs_fid *fid)
{
struct afs_cb_interest *cbi;
struct afs_iget_data data;
struct afs_vnode *vnode;
struct rb_node *p;
struct inode *inode;
_debug("find");
spin_lock(&server->fs_lock);
p = server->fs_vnodes.rb_node;
while (p) {
vnode = rb_entry(p, struct afs_vnode, server_rb);
if (fid->vid < vnode->fid.vid)
p = p->rb_left;
else if (fid->vid > vnode->fid.vid)
p = p->rb_right;
else if (fid->vnode < vnode->fid.vnode)
p = p->rb_left;
else if (fid->vnode > vnode->fid.vnode)
p = p->rb_right;
else if (fid->unique < vnode->fid.unique)
p = p->rb_left;
else if (fid->unique > vnode->fid.unique)
p = p->rb_right;
else
goto found;
read_lock(&server->cb_break_lock);
/* Step through all interested superblocks. There may be more than one
* because of cell aliasing.
*/
list_for_each_entry(cbi, &server->cb_interests, cb_link) {
if (cbi->vid != fid->vid)
continue;
data.volume = NULL;
data.fid = *fid;
inode = ilookup5_nowait(cbi->sb, fid->vnode, afs_iget5_test, &data);
if (inode) {
vnode = AFS_FS_I(inode);
afs_break_callback(vnode);
iput(inode);
}
}
/* not found so we just ignore it (it may have moved to another
* server) */
not_available:
_debug("not avail");
spin_unlock(&server->fs_lock);
_leave("");
return;
found:
_debug("found");
ASSERTCMP(server, ==, vnode->server);
if (!igrab(AFS_VNODE_TO_I(vnode)))
goto not_available;
spin_unlock(&server->fs_lock);
afs_break_callback(server, vnode);
iput(&vnode->vfs_inode);
_leave("");
read_unlock(&server->cb_break_lock);
}
/*
@ -216,261 +231,14 @@ void afs_break_callbacks(struct afs_server *server, size_t count,
}
/*
* record the callback for breaking
* - the caller must hold server->cb_lock
* Clear the callback interests in a server list.
*/
static void afs_do_give_up_callback(struct afs_server *server,
struct afs_vnode *vnode)
void afs_clear_callback_interests(struct afs_net *net, struct afs_server_list *slist)
{
struct afs_callback *cb;
int i;
_enter("%p,%p", server, vnode);
cb = &server->cb_break[server->cb_break_head];
cb->fid = vnode->fid;
cb->version = vnode->cb_version;
cb->expiry = vnode->cb_expiry;
cb->type = vnode->cb_type;
smp_wmb();
server->cb_break_head =
(server->cb_break_head + 1) &
(ARRAY_SIZE(server->cb_break) - 1);
/* defer the breaking of callbacks to try and collect as many as
* possible to ship in one operation */
switch (atomic_inc_return(&server->cb_break_n)) {
case 1 ... AFSCBMAX - 1:
queue_delayed_work(afs_callback_update_worker,
&server->cb_break_work, HZ * 2);
break;
case AFSCBMAX:
afs_flush_callback_breaks(server);
break;
default:
break;
for (i = 0; i < slist->nr_servers; i++) {
afs_put_cb_interest(net, slist->servers[i].cb_interest);
slist->servers[i].cb_interest = NULL;
}
ASSERT(server->cb_promises.rb_node != NULL);
rb_erase(&vnode->cb_promise, &server->cb_promises);
vnode->cb_promised = false;
_leave("");
}
/*
* discard the callback on a deleted item
*/
void afs_discard_callback_on_delete(struct afs_vnode *vnode)
{
struct afs_server *server = vnode->server;
_enter("%d", vnode->cb_promised);
if (!vnode->cb_promised) {
_leave(" [not promised]");
return;
}
ASSERT(server != NULL);
spin_lock(&server->cb_lock);
if (vnode->cb_promised) {
ASSERT(server->cb_promises.rb_node != NULL);
rb_erase(&vnode->cb_promise, &server->cb_promises);
vnode->cb_promised = false;
}
spin_unlock(&server->cb_lock);
_leave("");
}
/*
* give up the callback registered for a vnode on the file server when the
* inode is being cleared
*/
void afs_give_up_callback(struct afs_vnode *vnode)
{
struct afs_server *server = vnode->server;
DECLARE_WAITQUEUE(myself, current);
_enter("%d", vnode->cb_promised);
_debug("GIVE UP INODE %p", &vnode->vfs_inode);
if (!vnode->cb_promised) {
_leave(" [not promised]");
return;
}
ASSERT(server != NULL);
spin_lock(&server->cb_lock);
if (vnode->cb_promised && afs_breakring_space(server) == 0) {
add_wait_queue(&server->cb_break_waitq, &myself);
for (;;) {
set_current_state(TASK_UNINTERRUPTIBLE);
if (!vnode->cb_promised ||
afs_breakring_space(server) != 0)
break;
spin_unlock(&server->cb_lock);
schedule();
spin_lock(&server->cb_lock);
}
remove_wait_queue(&server->cb_break_waitq, &myself);
__set_current_state(TASK_RUNNING);
}
/* of course, it's always possible for the server to break this vnode's
* callback first... */
if (vnode->cb_promised)
afs_do_give_up_callback(server, vnode);
spin_unlock(&server->cb_lock);
_leave("");
}
/*
* dispatch a deferred give up callbacks operation
*/
void afs_dispatch_give_up_callbacks(struct work_struct *work)
{
struct afs_server *server =
container_of(work, struct afs_server, cb_break_work.work);
_enter("");
/* tell the fileserver to discard the callback promises it has
* - in the event of ENOMEM or some other error, we just forget that we
* had callbacks entirely, and the server will call us later to break
* them
*/
afs_fs_give_up_callbacks(server, true);
}
/*
* flush the outstanding callback breaks on a server
*/
void afs_flush_callback_breaks(struct afs_server *server)
{
mod_delayed_work(afs_callback_update_worker, &server->cb_break_work, 0);
}
#if 0
/*
* update a bunch of callbacks
*/
static void afs_callback_updater(struct work_struct *work)
{
struct afs_server *server;
struct afs_vnode *vnode, *xvnode;
time64_t now;
long timeout;
int ret;
server = container_of(work, struct afs_server, updater);
_enter("");
now = ktime_get_real_seconds();
/* find the first vnode to update */
spin_lock(&server->cb_lock);
for (;;) {
if (RB_EMPTY_ROOT(&server->cb_promises)) {
spin_unlock(&server->cb_lock);
_leave(" [nothing]");
return;
}
vnode = rb_entry(rb_first(&server->cb_promises),
struct afs_vnode, cb_promise);
if (atomic_read(&vnode->usage) > 0)
break;
rb_erase(&vnode->cb_promise, &server->cb_promises);
vnode->cb_promised = false;
}
timeout = vnode->update_at - now;
if (timeout > 0) {
queue_delayed_work(afs_vnode_update_worker,
&afs_vnode_update, timeout * HZ);
spin_unlock(&server->cb_lock);
_leave(" [nothing]");
return;
}
list_del_init(&vnode->update);
atomic_inc(&vnode->usage);
spin_unlock(&server->cb_lock);
/* we can now perform the update */
_debug("update %s", vnode->vldb.name);
vnode->state = AFS_VL_UPDATING;
vnode->upd_rej_cnt = 0;
vnode->upd_busy_cnt = 0;
ret = afs_vnode_update_record(vl, &vldb);
switch (ret) {
case 0:
afs_vnode_apply_update(vl, &vldb);
vnode->state = AFS_VL_UPDATING;
break;
case -ENOMEDIUM:
vnode->state = AFS_VL_VOLUME_DELETED;
break;
default:
vnode->state = AFS_VL_UNCERTAIN;
break;
}
/* and then reschedule */
_debug("reschedule");
vnode->update_at = ktime_get_real_seconds() +
afs_vnode_update_timeout;
spin_lock(&server->cb_lock);
if (!list_empty(&server->cb_promises)) {
/* next update in 10 minutes, but wait at least 1 second more
* than the newest record already queued so that we don't spam
* the VL server suddenly with lots of requests
*/
xvnode = list_entry(server->cb_promises.prev,
struct afs_vnode, update);
if (vnode->update_at <= xvnode->update_at)
vnode->update_at = xvnode->update_at + 1;
xvnode = list_entry(server->cb_promises.next,
struct afs_vnode, update);
timeout = xvnode->update_at - now;
if (timeout < 0)
timeout = 0;
} else {
timeout = afs_vnode_update_timeout;
}
list_add_tail(&vnode->update, &server->cb_promises);
_debug("timeout %ld", timeout);
queue_delayed_work(afs_vnode_update_worker,
&afs_vnode_update, timeout * HZ);
spin_unlock(&server->cb_lock);
afs_put_vnode(vl);
}
#endif
/*
* initialise the callback update process
*/
int __init afs_callback_update_init(void)
{
afs_callback_update_worker = alloc_ordered_workqueue("kafs_callbackd",
WQ_MEM_RECLAIM);
return afs_callback_update_worker ? 0 : -ENOMEM;
}
/*
* shut down the callback update process
*/
void afs_callback_update_kill(void)
{
destroy_workqueue(afs_callback_update_worker);
}

951
fs/afs/cell.c
View File

File diff suppressed because it is too large Load Diff

77
fs/afs/cmservice.c
View File

@ -41,7 +41,6 @@ static CM_NAME(CallBack);
static const struct afs_call_type afs_SRXCBCallBack = {
.name = afs_SRXCBCallBack_name,
.deliver = afs_deliver_cb_callback,
.abort_to_error = afs_abort_to_error,
.destructor = afs_cm_destructor,
.work = SRXAFSCB_CallBack,
};
@ -53,7 +52,6 @@ static CM_NAME(InitCallBackState);
static const struct afs_call_type afs_SRXCBInitCallBackState = {
.name = afs_SRXCBInitCallBackState_name,
.deliver = afs_deliver_cb_init_call_back_state,
.abort_to_error = afs_abort_to_error,
.destructor = afs_cm_destructor,
.work = SRXAFSCB_InitCallBackState,
};
@ -65,7 +63,6 @@ static CM_NAME(InitCallBackState3);
static const struct afs_call_type afs_SRXCBInitCallBackState3 = {
.name = afs_SRXCBInitCallBackState3_name,
.deliver = afs_deliver_cb_init_call_back_state3,
.abort_to_error = afs_abort_to_error,
.destructor = afs_cm_destructor,
.work = SRXAFSCB_InitCallBackState,
};
@ -77,7 +74,6 @@ static CM_NAME(Probe);
static const struct afs_call_type afs_SRXCBProbe = {
.name = afs_SRXCBProbe_name,
.deliver = afs_deliver_cb_probe,
.abort_to_error = afs_abort_to_error,
.destructor = afs_cm_destructor,
.work = SRXAFSCB_Probe,
};
@ -89,7 +85,6 @@ static CM_NAME(ProbeUuid);
static const struct afs_call_type afs_SRXCBProbeUuid = {
.name = afs_SRXCBProbeUuid_name,
.deliver = afs_deliver_cb_probe_uuid,
.abort_to_error = afs_abort_to_error,
.destructor = afs_cm_destructor,
.work = SRXAFSCB_ProbeUuid,
};
@ -101,7 +96,6 @@ static CM_NAME(TellMeAboutYourself);
static const struct afs_call_type afs_SRXCBTellMeAboutYourself = {
.name = afs_SRXCBTellMeAboutYourself_name,
.deliver = afs_deliver_cb_tell_me_about_yourself,
.abort_to_error = afs_abort_to_error,
.destructor = afs_cm_destructor,
.work = SRXAFSCB_TellMeAboutYourself,
};
@ -127,6 +121,9 @@ bool afs_cm_incoming_call(struct afs_call *call)
case CBProbe:
call->type = &afs_SRXCBProbe;
return true;
case CBProbeUuid:
call->type = &afs_SRXCBProbeUuid;
return true;
case CBTellMeAboutYourself:
call->type = &afs_SRXCBTellMeAboutYourself;
return true;
@ -147,18 +144,16 @@ static void afs_cm_destructor(struct afs_call *call)
* afs_deliver_cb_callback().
*/
if (call->unmarshall == 5) {
ASSERT(call->server && call->count && call->request);
afs_break_callbacks(call->server, call->count, call->request);
ASSERT(call->cm_server && call->count && call->request);
afs_break_callbacks(call->cm_server, call->count, call->request);
}
afs_put_server(call->server);
call->server = NULL;
kfree(call->buffer);
call->buffer = NULL;
}
/*
* allow the fileserver to see if the cache manager is still alive
* The server supplied a list of callbacks that it wanted to break.
*/
static void SRXAFSCB_CallBack(struct work_struct *work)
{
@ -173,7 +168,7 @@ static void SRXAFSCB_CallBack(struct work_struct *work)
* yet */
afs_send_empty_reply(call);
afs_break_callbacks(call->server, call->count, call->request);
afs_break_callbacks(call->cm_server, call->count, call->request);
afs_put_call(call);
_leave("");
}
@ -193,7 +188,6 @@ static int afs_deliver_cb_callback(struct afs_call *call)
switch (call->unmarshall) {
case 0:
rxrpc_kernel_get_peer(afs_socket, call->rxcall, &srx);
call->offset = 0;
call->unmarshall++;
@ -286,14 +280,16 @@ static int afs_deliver_cb_callback(struct afs_call *call)
break;
}
call->state = AFS_CALL_REPLYING;
if (!afs_check_call_state(call, AFS_CALL_SV_REPLYING))
return -EIO;
/* we'll need the file server record as that tells us which set of
* vnodes to operate upon */
server = afs_find_server(&srx);
rxrpc_kernel_get_peer(call->net->socket, call->rxcall, &srx);
server = afs_find_server(call->net, &srx);
if (!server)
return -ENOTCONN;
call->server = server;
call->cm_server = server;
return afs_queue_call_work(call);
}
@ -305,9 +301,9 @@ static void SRXAFSCB_InitCallBackState(struct work_struct *work)
{
struct afs_call *call = container_of(work, struct afs_call, work);
_enter("{%p}", call->server);
_enter("{%p}", call->cm_server);
afs_init_callback_state(call->server);
afs_init_callback_state(call->cm_server);
afs_send_empty_reply(call);
afs_put_call(call);
_leave("");
@ -324,21 +320,18 @@ static int afs_deliver_cb_init_call_back_state(struct afs_call *call)
_enter("");
rxrpc_kernel_get_peer(afs_socket, call->rxcall, &srx);
rxrpc_kernel_get_peer(call->net->socket, call->rxcall, &srx);
ret = afs_extract_data(call, NULL, 0, false);
if (ret < 0)
return ret;
/* no unmarshalling required */
call->state = AFS_CALL_REPLYING;
/* we'll need the file server record as that tells us which set of
* vnodes to operate upon */
server = afs_find_server(&srx);
server = afs_find_server(call->net, &srx);
if (!server)
return -ENOTCONN;
call->server = server;
call->cm_server = server;
return afs_queue_call_work(call);
}
@ -357,8 +350,6 @@ static int afs_deliver_cb_init_call_back_state3(struct afs_call *call)
_enter("");
rxrpc_kernel_get_peer(afs_socket, call->rxcall, &srx);
_enter("{%u}", call->unmarshall);
switch (call->unmarshall) {
@ -402,15 +393,16 @@ static int afs_deliver_cb_init_call_back_state3(struct afs_call *call)
break;
}
/* no unmarshalling required */
call->state = AFS_CALL_REPLYING;
if (!afs_check_call_state(call, AFS_CALL_SV_REPLYING))
return -EIO;
/* we'll need the file server record as that tells us which set of
* vnodes to operate upon */
server = afs_find_server(&srx);
rxrpc_kernel_get_peer(call->net->socket, call->rxcall, &srx);
server = afs_find_server(call->net, &srx);
if (!server)
return -ENOTCONN;
call->server = server;
call->cm_server = server;
return afs_queue_call_work(call);
}
@ -441,8 +433,8 @@ static int afs_deliver_cb_probe(struct afs_call *call)
if (ret < 0)
return ret;
/* no unmarshalling required */
call->state = AFS_CALL_REPLYING;
if (!afs_check_call_state(call, AFS_CALL_SV_REPLYING))
return -EIO;
return afs_queue_call_work(call);
}
@ -461,7 +453,7 @@ static void SRXAFSCB_ProbeUuid(struct work_struct *work)
_enter("");
if (memcmp(r, &afs_uuid, sizeof(afs_uuid)) == 0)
if (memcmp(r, &call->net->uuid, sizeof(call->net->uuid)) == 0)
reply.match = htonl(0);
else
reply.match = htonl(1);
@ -524,7 +516,8 @@ static int afs_deliver_cb_probe_uuid(struct afs_call *call)
break;
}
call->state = AFS_CALL_REPLYING;
if (!afs_check_call_state(call, AFS_CALL_SV_REPLYING))
return -EIO;
return afs_queue_call_work(call);
}
@ -568,13 +561,13 @@ static void SRXAFSCB_TellMeAboutYourself(struct work_struct *work)
memset(&reply, 0, sizeof(reply));
reply.ia.nifs = htonl(nifs);
reply.ia.uuid[0] = afs_uuid.time_low;
reply.ia.uuid[1] = htonl(ntohs(afs_uuid.time_mid));
reply.ia.uuid[2] = htonl(ntohs(afs_uuid.time_hi_and_version));
reply.ia.uuid[3] = htonl((s8) afs_uuid.clock_seq_hi_and_reserved);
reply.ia.uuid[4] = htonl((s8) afs_uuid.clock_seq_low);
reply.ia.uuid[0] = call->net->uuid.time_low;
reply.ia.uuid[1] = htonl(ntohs(call->net->uuid.time_mid));
reply.ia.uuid[2] = htonl(ntohs(call->net->uuid.time_hi_and_version));
reply.ia.uuid[3] = htonl((s8) call->net->uuid.clock_seq_hi_and_reserved);
reply.ia.uuid[4] = htonl((s8) call->net->uuid.clock_seq_low);
for (loop = 0; loop < 6; loop++)
reply.ia.uuid[loop + 5] = htonl((s8) afs_uuid.node[loop]);
reply.ia.uuid[loop + 5] = htonl((s8) call->net->uuid.node[loop]);
if (ifs) {
for (loop = 0; loop < nifs; loop++) {
@ -605,8 +598,8 @@ static int afs_deliver_cb_tell_me_about_yourself(struct afs_call *call)
if (ret < 0)
return ret;
/* no unmarshalling required */
call->state = AFS_CALL_REPLYING;
if (!afs_check_call_state(call, AFS_CALL_SV_REPLYING))
return -EIO;
return afs_queue_call_work(call);
}

461
fs/afs/dir.c
View File

@ -130,10 +130,11 @@ struct afs_lookup_cookie {
/*
* check that a directory page is valid
*/
static inline bool afs_dir_check_page(struct inode *dir, struct page *page)
bool afs_dir_check_page(struct inode *dir, struct page *page)
{
struct afs_dir_page *dbuf;
loff_t latter;
struct afs_vnode *vnode = AFS_FS_I(dir);
loff_t latter, i_size, off;
int tmp, qty;
#if 0
@ -150,8 +151,15 @@ static inline bool afs_dir_check_page(struct inode *dir, struct page *page)
}
#endif
/* determine how many magic numbers there should be in this page */
latter = dir->i_size - page_offset(page);
/* Determine how many magic numbers there should be in this page, but
* we must take care because the directory may change size under us.
*/
off = page_offset(page);
i_size = i_size_read(dir);
if (i_size <= off)
goto checked;
latter = i_size - off;
if (latter >= PAGE_SIZE)
qty = PAGE_SIZE;
else
@ -162,13 +170,15 @@ static inline bool afs_dir_check_page(struct inode *dir, struct page *page)
dbuf = page_address(page);
for (tmp = 0; tmp < qty; tmp++) {
if (dbuf->blocks[tmp].pagehdr.magic != AFS_DIR_MAGIC) {
printk("kAFS: %s(%lu): bad magic %d/%d is %04hx\n",
printk("kAFS: %s(%lx): bad magic %d/%d is %04hx\n",
__func__, dir->i_ino, tmp, qty,
ntohs(dbuf->blocks[tmp].pagehdr.magic));
trace_afs_dir_check_failed(vnode, off, i_size);
goto error;
}
}
checked:
SetPageChecked(page);
return true;
@ -183,6 +193,7 @@ error:
static inline void afs_dir_put_page(struct page *page)
{
kunmap(page);
unlock_page(page);
put_page(page);
}
@ -197,9 +208,10 @@ static struct page *afs_dir_get_page(struct inode *dir, unsigned long index,
page = read_cache_page(dir->i_mapping, index, afs_page_filler, key);
if (!IS_ERR(page)) {
lock_page(page);
kmap(page);
if (unlikely(!PageChecked(page))) {
if (PageError(page) || !afs_dir_check_page(dir, page))
if (PageError(page))
goto fail;
}
}
@ -384,8 +396,7 @@ out:
*/
static int afs_readdir(struct file *file, struct dir_context *ctx)
{
return afs_dir_iterate(file_inode(file),
ctx, file->private_data);
return afs_dir_iterate(file_inode(file), ctx, afs_file_key(file));
}
/*
@ -553,7 +564,7 @@ static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry,
dentry->d_fsdata = (void *)(unsigned long) vnode->status.data_version;
/* instantiate the dentry */
inode = afs_iget(dir->i_sb, key, &fid, NULL, NULL);
inode = afs_iget(dir->i_sb, key, &fid, NULL, NULL, NULL);
key_put(key);
if (IS_ERR(inode)) {
_leave(" = %ld", PTR_ERR(inode));
@ -581,6 +592,7 @@ static int afs_d_revalidate(struct dentry *dentry, unsigned int flags)
struct afs_vnode *vnode, *dir;
struct afs_fid uninitialized_var(fid);
struct dentry *parent;
struct inode *inode;
struct key *key;
void *dir_version;
int ret;
@ -588,30 +600,39 @@ static int afs_d_revalidate(struct dentry *dentry, unsigned int flags)
if (flags & LOOKUP_RCU)
return -ECHILD;
vnode = AFS_FS_I(d_inode(dentry));
if (d_really_is_positive(dentry))
if (d_really_is_positive(dentry)) {
vnode = AFS_FS_I(d_inode(dentry));
_enter("{v={%x:%u} n=%pd fl=%lx},",
vnode->fid.vid, vnode->fid.vnode, dentry,
vnode->flags);
else
} else {
_enter("{neg n=%pd}", dentry);
}
key = afs_request_key(AFS_FS_S(dentry->d_sb)->volume->cell);
if (IS_ERR(key))
key = NULL;
if (d_really_is_positive(dentry)) {
inode = d_inode(dentry);
if (inode) {
vnode = AFS_FS_I(inode);
afs_validate(vnode, key);
if (test_bit(AFS_VNODE_DELETED, &vnode->flags))
goto out_bad;
}
}
/* lock down the parent dentry so we can peer at it */
parent = dget_parent(dentry);
dir = AFS_FS_I(d_inode(parent));
/* validate the parent directory */
if (test_bit(AFS_VNODE_MODIFIED, &dir->flags))
afs_validate(dir, key);
afs_validate(dir, key);
if (test_bit(AFS_VNODE_DELETED, &dir->flags)) {
_debug("%pd: parent dir deleted", dentry);
goto out_bad;
goto out_bad_parent;
}
dir_version = (void *) (unsigned long) dir->status.data_version;
@ -626,13 +647,16 @@ static int afs_d_revalidate(struct dentry *dentry, unsigned int flags)
case 0:
/* the filename maps to something */
if (d_really_is_negative(dentry))
goto out_bad;
if (is_bad_inode(d_inode(dentry))) {
goto out_bad_parent;
inode = d_inode(dentry);
if (is_bad_inode(inode)) {
printk("kAFS: afs_d_revalidate: %pd2 has bad inode\n",
dentry);
goto out_bad;
goto out_bad_parent;
}
vnode = AFS_FS_I(inode);
/* if the vnode ID has changed, then the dirent points to a
* different file */
if (fid.vnode != vnode->fid.vnode) {
@ -649,10 +673,10 @@ static int afs_d_revalidate(struct dentry *dentry, unsigned int flags)
_debug("%pd: file deleted (uq %u -> %u I:%u)",
dentry, fid.unique,
vnode->fid.unique,
d_inode(dentry)->i_generation);
spin_lock(&vnode->lock);
vnode->vfs_inode.i_generation);
write_seqlock(&vnode->cb_lock);
set_bit(AFS_VNODE_DELETED, &vnode->flags);
spin_unlock(&vnode->lock);
write_sequnlock(&vnode->cb_lock);
goto not_found;
}
goto out_valid;
@ -667,7 +691,7 @@ static int afs_d_revalidate(struct dentry *dentry, unsigned int flags)
default:
_debug("failed to iterate dir %pd: %d",
parent, ret);
goto out_bad;
goto out_bad_parent;
}
out_valid:
@ -683,9 +707,10 @@ not_found:
dentry->d_flags |= DCACHE_NFSFS_RENAMED;
spin_unlock(&dentry->d_lock);
out_bad:
out_bad_parent:
_debug("dropping dentry %pd2", dentry);
dput(parent);
out_bad:
key_put(key);
_leave(" = 0 [bad]");
@ -726,21 +751,49 @@ static void afs_d_release(struct dentry *dentry)
_enter("%pd", dentry);
}
/*
* Create a new inode for create/mkdir/symlink
*/
static void afs_vnode_new_inode(struct afs_fs_cursor *fc,
struct dentry *new_dentry,
struct afs_fid *newfid,
struct afs_file_status *newstatus,
struct afs_callback *newcb)
{
struct inode *inode;
if (fc->ac.error < 0)
return;
inode = afs_iget(fc->vnode->vfs_inode.i_sb, fc->key,
newfid, newstatus, newcb, fc->cbi);
if (IS_ERR(inode)) {
/* ENOMEM or EINTR at a really inconvenient time - just abandon
* the new directory on the server.
*/
fc->ac.error = PTR_ERR(inode);
return;
}
d_instantiate(new_dentry, inode);
if (d_unhashed(new_dentry))
d_rehash(new_dentry);
}
/*
* create a directory on an AFS filesystem
*/
static int afs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
{
struct afs_file_status status;
struct afs_callback cb;
struct afs_server *server;
struct afs_vnode *dvnode, *vnode;
struct afs_fid fid;
struct inode *inode;
struct afs_file_status newstatus;
struct afs_fs_cursor fc;
struct afs_callback newcb;
struct afs_vnode *dvnode = AFS_FS_I(dir);
struct afs_fid newfid;
struct key *key;
int ret;
dvnode = AFS_FS_I(dir);
mode |= S_IFDIR;
_enter("{%x:%u},{%pd},%ho",
dvnode->fid.vid, dvnode->fid.vnode, dentry, mode);
@ -751,40 +804,27 @@ static int afs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
goto error;
}
mode |= S_IFDIR;
ret = afs_vnode_create(dvnode, key, dentry->d_name.name,
mode, &fid, &status, &cb, &server);
if (ret < 0)
goto mkdir_error;
ret = -ERESTARTSYS;
if (afs_begin_vnode_operation(&fc, dvnode, key)) {
while (afs_select_fileserver(&fc)) {
fc.cb_break = dvnode->cb_break + dvnode->cb_s_break;
afs_fs_create(&fc, dentry->d_name.name, mode,
&newfid, &newstatus, &newcb);
}
inode = afs_iget(dir->i_sb, key, &fid, &status, &cb);
if (IS_ERR(inode)) {
/* ENOMEM at a really inconvenient time - just abandon the new
* directory on the server */
ret = PTR_ERR(inode);
goto iget_error;
afs_check_for_remote_deletion(&fc, fc.vnode);
afs_vnode_commit_status(&fc, dvnode, fc.cb_break);
afs_vnode_new_inode(&fc, dentry, &newfid, &newstatus, &newcb);
ret = afs_end_vnode_operation(&fc);
if (ret < 0)
goto error_key;
}
/* apply the status report we've got for the new vnode */
vnode = AFS_FS_I(inode);
spin_lock(&vnode->lock);
vnode->update_cnt++;
spin_unlock(&vnode->lock);
afs_vnode_finalise_status_update(vnode, server);
afs_put_server(server);
d_instantiate(dentry, inode);
if (d_unhashed(dentry)) {
_debug("not hashed");
d_rehash(dentry);
}
key_put(key);
_leave(" = 0");
return 0;
iget_error:
afs_put_server(server);
mkdir_error:
error_key:
key_put(key);
error:
d_drop(dentry);
@ -792,17 +832,30 @@ error:
return ret;
}
/*
* Remove a subdir from a directory.
*/
static void afs_dir_remove_subdir(struct dentry *dentry)
{
if (d_really_is_positive(dentry)) {
struct afs_vnode *vnode = AFS_FS_I(d_inode(dentry));
clear_nlink(&vnode->vfs_inode);
set_bit(AFS_VNODE_DELETED, &vnode->flags);
clear_bit(AFS_VNODE_CB_PROMISED, &vnode->flags);
}
}
/*
* remove a directory from an AFS filesystem
*/
static int afs_rmdir(struct inode *dir, struct dentry *dentry)
{
struct afs_vnode *dvnode, *vnode;
struct afs_fs_cursor fc;
struct afs_vnode *dvnode = AFS_FS_I(dir);
struct key *key;
int ret;
dvnode = AFS_FS_I(dir);
_enter("{%x:%u},{%pd}",
dvnode->fid.vid, dvnode->fid.vnode, dentry);
@ -812,45 +865,69 @@ static int afs_rmdir(struct inode *dir, struct dentry *dentry)
goto error;
}
ret = afs_vnode_remove(dvnode, key, dentry->d_name.name, true);
if (ret < 0)
goto rmdir_error;
ret = -ERESTARTSYS;
if (afs_begin_vnode_operation(&fc, dvnode, key)) {
while (afs_select_fileserver(&fc)) {
fc.cb_break = dvnode->cb_break + dvnode->cb_s_break;
afs_fs_remove(&fc, dentry->d_name.name, true);
}
if (d_really_is_positive(dentry)) {
vnode = AFS_FS_I(d_inode(dentry));
clear_nlink(&vnode->vfs_inode);
set_bit(AFS_VNODE_DELETED, &vnode->flags);
afs_discard_callback_on_delete(vnode);
afs_vnode_commit_status(&fc, dvnode, fc.cb_break);
ret = afs_end_vnode_operation(&fc);
if (ret == 0)
afs_dir_remove_subdir(dentry);
}
key_put(key);
_leave(" = 0");
return 0;
rmdir_error:
key_put(key);
error:
_leave(" = %d", ret);
return ret;
}
/*
* remove a file from an AFS filesystem
* Remove a link to a file or symlink from a directory.
*
* If the file was not deleted due to excess hard links, the fileserver will
* break the callback promise on the file - if it had one - before it returns
* to us, and if it was deleted, it won't
*
* However, if we didn't have a callback promise outstanding, or it was
* outstanding on a different server, then it won't break it either...
*/
static int afs_dir_remove_link(struct dentry *dentry, struct key *key)
{
int ret = 0;
if (d_really_is_positive(dentry)) {
struct afs_vnode *vnode = AFS_FS_I(d_inode(dentry));
if (test_bit(AFS_VNODE_DELETED, &vnode->flags))
kdebug("AFS_VNODE_DELETED");
clear_bit(AFS_VNODE_CB_PROMISED, &vnode->flags);
ret = afs_validate(vnode, key);
if (ret == -ESTALE)
ret = 0;
_debug("nlink %d [val %d]", vnode->vfs_inode.i_nlink, ret);
}
return ret;
}
/*
* Remove a file or symlink from an AFS filesystem.
*/
static int afs_unlink(struct inode *dir, struct dentry *dentry)
{
struct afs_vnode *dvnode, *vnode;
struct afs_fs_cursor fc;
struct afs_vnode *dvnode = AFS_FS_I(dir), *vnode;
struct key *key;
int ret;
dvnode = AFS_FS_I(dir);
_enter("{%x:%u},{%pd}",
dvnode->fid.vid, dvnode->fid.vnode, dentry);
ret = -ENAMETOOLONG;
if (dentry->d_name.len >= AFSNAMEMAX)
goto error;
return -ENAMETOOLONG;
key = afs_request_key(dvnode->volume->cell);
if (IS_ERR(key)) {
@ -858,44 +935,28 @@ static int afs_unlink(struct inode *dir, struct dentry *dentry)
goto error;
}
/* Try to make sure we have a callback promise on the victim. */
if (d_really_is_positive(dentry)) {
vnode = AFS_FS_I(d_inode(dentry));
/* make sure we have a callback promise on the victim */
ret = afs_validate(vnode, key);
if (ret < 0)
goto error;
goto error_key;
}
ret = afs_vnode_remove(dvnode, key, dentry->d_name.name, false);
if (ret < 0)
goto remove_error;
ret = -ERESTARTSYS;
if (afs_begin_vnode_operation(&fc, dvnode, key)) {
while (afs_select_fileserver(&fc)) {
fc.cb_break = dvnode->cb_break + dvnode->cb_s_break;
afs_fs_remove(&fc, dentry->d_name.name, false);
}
if (d_really_is_positive(dentry)) {
/* if the file wasn't deleted due to excess hard links, the
* fileserver will break the callback promise on the file - if
* it had one - before it returns to us, and if it was deleted,
* it won't
*
* however, if we didn't have a callback promise outstanding,
* or it was outstanding on a different server, then it won't
* break it either...
*/
vnode = AFS_FS_I(d_inode(dentry));
if (test_bit(AFS_VNODE_DELETED, &vnode->flags))
_debug("AFS_VNODE_DELETED");
if (test_bit(AFS_VNODE_CB_BROKEN, &vnode->flags))
_debug("AFS_VNODE_CB_BROKEN");
set_bit(AFS_VNODE_CB_BROKEN, &vnode->flags);
ret = afs_validate(vnode, key);
_debug("nlink %d [val %d]", vnode->vfs_inode.i_nlink, ret);
afs_vnode_commit_status(&fc, dvnode, fc.cb_break);
ret = afs_end_vnode_operation(&fc);
if (ret == 0)
ret = afs_dir_remove_link(dentry, key);
}
key_put(key);
_leave(" = 0");
return 0;
remove_error:
error_key:
key_put(key);
error:
_leave(" = %d", ret);
@ -908,60 +969,50 @@ error:
static int afs_create(struct inode *dir, struct dentry *dentry, umode_t mode,
bool excl)
{
struct afs_file_status status;
struct afs_callback cb;
struct afs_server *server;
struct afs_vnode *dvnode, *vnode;
struct afs_fid fid;
struct inode *inode;
struct afs_fs_cursor fc;
struct afs_file_status newstatus;
struct afs_callback newcb;
struct afs_vnode *dvnode = dvnode = AFS_FS_I(dir);
struct afs_fid newfid;
struct key *key;
int ret;
dvnode = AFS_FS_I(dir);
mode |= S_IFREG;
_enter("{%x:%u},{%pd},%ho,",
dvnode->fid.vid, dvnode->fid.vnode, dentry, mode);
ret = -ENAMETOOLONG;
if (dentry->d_name.len >= AFSNAMEMAX)
goto error;
key = afs_request_key(dvnode->volume->cell);
if (IS_ERR(key)) {
ret = PTR_ERR(key);
goto error;
}
mode |= S_IFREG;
ret = afs_vnode_create(dvnode, key, dentry->d_name.name,
mode, &fid, &status, &cb, &server);
if (ret < 0)
goto create_error;
ret = -ERESTARTSYS;
if (afs_begin_vnode_operation(&fc, dvnode, key)) {
while (afs_select_fileserver(&fc)) {
fc.cb_break = dvnode->cb_break + dvnode->cb_s_break;
afs_fs_create(&fc, dentry->d_name.name, mode,
&newfid, &newstatus, &newcb);
}
inode = afs_iget(dir->i_sb, key, &fid, &status, &cb);
if (IS_ERR(inode)) {
/* ENOMEM at a really inconvenient time - just abandon the new
* directory on the server */
ret = PTR_ERR(inode);
goto iget_error;
afs_check_for_remote_deletion(&fc, fc.vnode);
afs_vnode_commit_status(&fc, dvnode, fc.cb_break);
afs_vnode_new_inode(&fc, dentry, &newfid, &newstatus, &newcb);
ret = afs_end_vnode_operation(&fc);
if (ret < 0)
goto error_key;
}
/* apply the status report we've got for the new vnode */
vnode = AFS_FS_I(inode);
spin_lock(&vnode->lock);
vnode->update_cnt++;
spin_unlock(&vnode->lock);
afs_vnode_finalise_status_update(vnode, server);
afs_put_server(server);
d_instantiate(dentry, inode);
if (d_unhashed(dentry)) {
_debug("not hashed");
d_rehash(dentry);
}
key_put(key);
_leave(" = 0");
return 0;
iget_error:
afs_put_server(server);
create_error:
error_key:
key_put(key);
error:
d_drop(dentry);
@ -975,6 +1026,7 @@ error:
static int afs_link(struct dentry *from, struct inode *dir,
struct dentry *dentry)
{
struct afs_fs_cursor fc;
struct afs_vnode *dvnode, *vnode;
struct key *key;
int ret;
@ -987,23 +1039,45 @@ static int afs_link(struct dentry *from, struct inode *dir,
dvnode->fid.vid, dvnode->fid.vnode,
dentry);
ret = -ENAMETOOLONG;
if (dentry->d_name.len >= AFSNAMEMAX)
goto error;
key = afs_request_key(dvnode->volume->cell);
if (IS_ERR(key)) {
ret = PTR_ERR(key);
goto error;
}
ret = afs_vnode_link(dvnode, vnode, key, dentry->d_name.name);
if (ret < 0)
goto link_error;
ret = -ERESTARTSYS;
if (afs_begin_vnode_operation(&fc, dvnode, key)) {
if (mutex_lock_interruptible_nested(&vnode->io_lock, 1) < 0) {
afs_end_vnode_operation(&fc);
return -ERESTARTSYS;
}
while (afs_select_fileserver(&fc)) {
fc.cb_break = dvnode->cb_break + dvnode->cb_s_break;
fc.cb_break_2 = vnode->cb_break + vnode->cb_s_break;
afs_fs_link(&fc, vnode, dentry->d_name.name);
}
afs_vnode_commit_status(&fc, dvnode, fc.cb_break);
afs_vnode_commit_status(&fc, vnode, fc.cb_break_2);
ihold(&vnode->vfs_inode);
d_instantiate(dentry, &vnode->vfs_inode);
mutex_unlock(&vnode->io_lock);
ret = afs_end_vnode_operation(&fc);
if (ret < 0)
goto error_key;
}
ihold(&vnode->vfs_inode);
d_instantiate(dentry, &vnode->vfs_inode);
key_put(key);
_leave(" = 0");
return 0;
link_error:
error_key:
key_put(key);
error:
d_drop(dentry);
@ -1017,20 +1091,21 @@ error:
static int afs_symlink(struct inode *dir, struct dentry *dentry,
const char *content)
{
struct afs_file_status status;
struct afs_server *server;
struct afs_vnode *dvnode, *vnode;
struct afs_fid fid;
struct inode *inode;
struct afs_fs_cursor fc;
struct afs_file_status newstatus;
struct afs_vnode *dvnode = AFS_FS_I(dir);
struct afs_fid newfid;
struct key *key;
int ret;
dvnode = AFS_FS_I(dir);
_enter("{%x:%u},{%pd},%s",
dvnode->fid.vid, dvnode->fid.vnode, dentry,
content);
ret = -ENAMETOOLONG;
if (dentry->d_name.len >= AFSNAMEMAX)
goto error;
ret = -EINVAL;
if (strlen(content) >= AFSPATHMAX)
goto error;
@ -1041,39 +1116,27 @@ static int afs_symlink(struct inode *dir, struct dentry *dentry,
goto error;
}
ret = afs_vnode_symlink(dvnode, key, dentry->d_name.name, content,
&fid, &status, &server);
if (ret < 0)
goto create_error;
ret = -ERESTARTSYS;
if (afs_begin_vnode_operation(&fc, dvnode, key)) {
while (afs_select_fileserver(&fc)) {
fc.cb_break = dvnode->cb_break + dvnode->cb_s_break;
afs_fs_symlink(&fc, dentry->d_name.name, content,
&newfid, &newstatus);
}
inode = afs_iget(dir->i_sb, key, &fid, &status, NULL);
if (IS_ERR(inode)) {
/* ENOMEM at a really inconvenient time - just abandon the new
* directory on the server */
ret = PTR_ERR(inode);
goto iget_error;
afs_check_for_remote_deletion(&fc, fc.vnode);
afs_vnode_commit_status(&fc, dvnode, fc.cb_break);
afs_vnode_new_inode(&fc, dentry, &newfid, &newstatus, NULL);
ret = afs_end_vnode_operation(&fc);
if (ret < 0)
goto error_key;
}
/* apply the status report we've got for the new vnode */
vnode = AFS_FS_I(inode);
spin_lock(&vnode->lock);
vnode->update_cnt++;
spin_unlock(&vnode->lock);
afs_vnode_finalise_status_update(vnode, server);
afs_put_server(server);
d_instantiate(dentry, inode);
if (d_unhashed(dentry)) {
_debug("not hashed");
d_rehash(dentry);
}
key_put(key);
_leave(" = 0");
return 0;
iget_error:
afs_put_server(server);
create_error:
error_key:
key_put(key);
error:
d_drop(dentry);
@ -1088,6 +1151,7 @@ static int afs_rename(struct inode *old_dir, struct dentry *old_dentry,
struct inode *new_dir, struct dentry *new_dentry,
unsigned int flags)
{
struct afs_fs_cursor fc;
struct afs_vnode *orig_dvnode, *new_dvnode, *vnode;
struct key *key;
int ret;
@ -1111,16 +1175,35 @@ static int afs_rename(struct inode *old_dir, struct dentry *old_dentry,
goto error;
}
ret = afs_vnode_rename(orig_dvnode, new_dvnode, key,
old_dentry->d_name.name,
new_dentry->d_name.name);
if (ret < 0)
goto rename_error;
ret = -ERESTARTSYS;
if (afs_begin_vnode_operation(&fc, orig_dvnode, key)) {
if (orig_dvnode != new_dvnode) {
if (mutex_lock_interruptible_nested(&new_dvnode->io_lock, 1) < 0) {
afs_end_vnode_operation(&fc);
return -ERESTARTSYS;
}
}
while (afs_select_fileserver(&fc)) {
fc.cb_break = orig_dvnode->cb_break + orig_dvnode->cb_s_break;
fc.cb_break_2 = new_dvnode->cb_break + new_dvnode->cb_s_break;
afs_fs_rename(&fc, old_dentry->d_name.name,
new_dvnode, new_dentry->d_name.name);
}
afs_vnode_commit_status(&fc, orig_dvnode, fc.cb_break);
afs_vnode_commit_status(&fc, new_dvnode, fc.cb_break_2);
if (orig_dvnode != new_dvnode)
mutex_unlock(&new_dvnode->io_lock);
ret = afs_end_vnode_operation(&fc);
if (ret < 0)
goto error_key;
}
key_put(key);
_leave(" = 0");
return 0;
rename_error:
error_key:
key_put(key);
error:
d_drop(new_dentry);

194
fs/afs/file.c
View File

@ -19,11 +19,11 @@
#include <linux/task_io_accounting_ops.h>
#include "internal.h"
static int afs_file_mmap(struct file *file, struct vm_area_struct *vma);
static int afs_readpage(struct file *file, struct page *page);
static void afs_invalidatepage(struct page *page, unsigned int offset,
unsigned int length);
static int afs_releasepage(struct page *page, gfp_t gfp_flags);
static int afs_launder_page(struct page *page);
static int afs_readpages(struct file *filp, struct address_space *mapping,
struct list_head *pages, unsigned nr_pages);
@ -35,7 +35,7 @@ const struct file_operations afs_file_operations = {
.llseek = generic_file_llseek,
.read_iter = generic_file_read_iter,
.write_iter = afs_file_write,
.mmap = generic_file_readonly_mmap,
.mmap = afs_file_mmap,
.splice_read = generic_file_splice_read,
.fsync = afs_fsync,
.lock = afs_lock,
@ -62,12 +62,63 @@ const struct address_space_operations afs_fs_aops = {
.writepages = afs_writepages,
};
static const struct vm_operations_struct afs_vm_ops = {
.fault = filemap_fault,
.map_pages = filemap_map_pages,
.page_mkwrite = afs_page_mkwrite,
};
/*
* Discard a pin on a writeback key.
*/
void afs_put_wb_key(struct afs_wb_key *wbk)
{
if (refcount_dec_and_test(&wbk->usage)) {
key_put(wbk->key);
kfree(wbk);
}
}
/*
* Cache key for writeback.
*/
int afs_cache_wb_key(struct afs_vnode *vnode, struct afs_file *af)
{
struct afs_wb_key *wbk, *p;
wbk = kzalloc(sizeof(struct afs_wb_key), GFP_KERNEL);
if (!wbk)
return -ENOMEM;
refcount_set(&wbk->usage, 2);
wbk->key = af->key;
spin_lock(&vnode->wb_lock);
list_for_each_entry(p, &vnode->wb_keys, vnode_link) {
if (p->key == wbk->key)
goto found;
}
key_get(wbk->key);
list_add_tail(&wbk->vnode_link, &vnode->wb_keys);
spin_unlock(&vnode->wb_lock);
af->wb = wbk;
return 0;
found:
refcount_inc(&p->usage);
spin_unlock(&vnode->wb_lock);
af->wb = p;
kfree(wbk);
return 0;
}
/*
* open an AFS file or directory and attach a key to it
*/
int afs_open(struct inode *inode, struct file *file)
{
struct afs_vnode *vnode = AFS_FS_I(inode);
struct afs_file *af;
struct key *key;
int ret;
@ -75,19 +126,38 @@ int afs_open(struct inode *inode, struct file *file)
key = afs_request_key(vnode->volume->cell);
if (IS_ERR(key)) {
_leave(" = %ld [key]", PTR_ERR(key));
return PTR_ERR(key);
ret = PTR_ERR(key);
goto error;
}
af = kzalloc(sizeof(*af), GFP_KERNEL);
if (!af) {
ret = -ENOMEM;
goto error_key;
}
af->key = key;
ret = afs_validate(vnode, key);
if (ret < 0) {
_leave(" = %d [val]", ret);
return ret;
}
if (ret < 0)
goto error_af;
file->private_data = key;
if (file->f_mode & FMODE_WRITE) {
ret = afs_cache_wb_key(vnode, af);
if (ret < 0)
goto error_af;
}
file->private_data = af;
_leave(" = 0");
return 0;
error_af:
kfree(af);
error_key:
key_put(key);
error:
_leave(" = %d", ret);
return ret;
}
/*
@ -96,10 +166,16 @@ int afs_open(struct inode *inode, struct file *file)
int afs_release(struct inode *inode, struct file *file)
{
struct afs_vnode *vnode = AFS_FS_I(inode);
struct afs_file *af = file->private_data;
_enter("{%x:%u},", vnode->fid.vid, vnode->fid.vnode);
key_put(file->private_data);
file->private_data = NULL;
if (af->wb)
afs_put_wb_key(af->wb);
key_put(af->key);
kfree(af);
afs_prune_wb_keys(vnode);
_leave(" = 0");
return 0;
}
@ -137,6 +213,37 @@ static void afs_file_readpage_read_complete(struct page *page,
}
#endif
/*
* Fetch file data from the volume.
*/
int afs_fetch_data(struct afs_vnode *vnode, struct key *key, struct afs_read *desc)
{
struct afs_fs_cursor fc;
int ret;
_enter("%s{%x:%u.%u},%x,,,",
vnode->volume->name,
vnode->fid.vid,
vnode->fid.vnode,
vnode->fid.unique,
key_serial(key));
ret = -ERESTARTSYS;
if (afs_begin_vnode_operation(&fc, vnode, key)) {
while (afs_select_fileserver(&fc)) {
fc.cb_break = vnode->cb_break + vnode->cb_s_break;
afs_fs_fetch_data(&fc, desc);
}
afs_check_for_remote_deletion(&fc, fc.vnode);
afs_vnode_commit_status(&fc, vnode, fc.cb_break);
ret = afs_end_vnode_operation(&fc);
}
_leave(" = %d", ret);
return ret;
}
/*
* read page from file, directory or symlink, given a key to use
*/
@ -199,8 +306,13 @@ int afs_page_filler(void *data, struct page *page)
/* read the contents of the file from the server into the
* page */
ret = afs_vnode_fetch_data(vnode, key, req);
ret = afs_fetch_data(vnode, key, req);
afs_put_read(req);
if (ret >= 0 && S_ISDIR(inode->i_mode) &&
!afs_dir_check_page(inode, page))
ret = -EIO;
if (ret < 0) {
if (ret == -ENOENT) {
_debug("got NOENT from server"
@ -259,12 +371,12 @@ static int afs_readpage(struct file *file, struct page *page)
int ret;
if (file) {
key = file->private_data;
key = afs_file_key(file);
ASSERT(key != NULL);
ret = afs_page_filler(key, page);
} else {
struct inode *inode = page->mapping->host;
key = afs_request_key(AFS_FS_S(inode->i_sb)->volume->cell);
key = afs_request_key(AFS_FS_S(inode->i_sb)->cell);
if (IS_ERR(key)) {
ret = PTR_ERR(key);
} else {
@ -281,7 +393,7 @@ static int afs_readpage(struct file *file, struct page *page)
static void afs_readpages_page_done(struct afs_call *call, struct afs_read *req)
{
#ifdef CONFIG_AFS_FSCACHE
struct afs_vnode *vnode = call->reply;
struct afs_vnode *vnode = call->reply[0];
#endif
struct page *page = req->pages[req->index];
@ -310,7 +422,7 @@ static int afs_readpages_one(struct file *file, struct address_space *mapping,
struct afs_read *req;
struct list_head *p;
struct page *first, *page;
struct key *key = file->private_data;
struct key *key = afs_file_key(file);
pgoff_t index;
int ret, n, i;
@ -369,7 +481,7 @@ static int afs_readpages_one(struct file *file, struct address_space *mapping,
return 0;
}
ret = afs_vnode_fetch_data(vnode, key, req);
ret = afs_fetch_data(vnode, key, req);
if (ret < 0)
goto error;
@ -406,7 +518,7 @@ error:
static int afs_readpages(struct file *file, struct address_space *mapping,
struct list_head *pages, unsigned nr_pages)
{
struct key *key = file->private_data;
struct key *key = afs_file_key(file);
struct afs_vnode *vnode;
int ret = 0;
@ -463,16 +575,6 @@ static int afs_readpages(struct file *file, struct address_space *mapping,
return ret;
}
/*
* write back a dirty page
*/
static int afs_launder_page(struct page *page)
{
_enter("{%lu}", page->index);
return 0;
}
/*
* invalidate part or all of a page
* - release a page and clean up its private data if offset is 0 (indicating
@ -481,7 +583,8 @@ static int afs_launder_page(struct page *page)
static void afs_invalidatepage(struct page *page, unsigned int offset,
unsigned int length)
{
struct afs_writeback *wb = (struct afs_writeback *) page_private(page);
struct afs_vnode *vnode = AFS_FS_I(page->mapping->host);
unsigned long priv;
_enter("{%lu},%u,%u", page->index, offset, length);
@ -498,13 +601,11 @@ static void afs_invalidatepage(struct page *page, unsigned int offset,
#endif
if (PagePrivate(page)) {
if (wb && !PageWriteback(page)) {
set_page_private(page, 0);
afs_put_writeback(wb);
}
if (!page_private(page))
ClearPagePrivate(page);
priv = page_private(page);
trace_afs_page_dirty(vnode, tracepoint_string("inval"),
page->index, priv);
set_page_private(page, 0);
ClearPagePrivate(page);
}
}
@ -517,8 +618,8 @@ static void afs_invalidatepage(struct page *page, unsigned int offset,
*/
static int afs_releasepage(struct page *page, gfp_t gfp_flags)
{
struct afs_writeback *wb = (struct afs_writeback *) page_private(page);
struct afs_vnode *vnode = AFS_FS_I(page->mapping->host);
unsigned long priv;
_enter("{{%x:%u}[%lu],%lx},%x",
vnode->fid.vid, vnode->fid.vnode, page->index, page->flags,
@ -534,10 +635,10 @@ static int afs_releasepage(struct page *page, gfp_t gfp_flags)
#endif
if (PagePrivate(page)) {
if (wb) {
set_page_private(page, 0);
afs_put_writeback(wb);
}
priv = page_private(page);
trace_afs_page_dirty(vnode, tracepoint_string("rel"),
page->index, priv);
set_page_private(page, 0);
ClearPagePrivate(page);
}
@ -545,3 +646,16 @@ static int afs_releasepage(struct page *page, gfp_t gfp_flags)
_leave(" = T");
return 1;
}
/*
* Handle setting up a memory mapping on an AFS file.
*/
static int afs_file_mmap(struct file *file, struct vm_area_struct *vma)
{
int ret;
ret = generic_file_mmap(file, vma);
if (ret == 0)
vma->vm_ops = &afs_vm_ops;
return ret;
}

159
fs/afs/flock.c
View File

@ -14,47 +14,16 @@
#define AFS_LOCK_GRANTED 0
#define AFS_LOCK_PENDING 1
struct workqueue_struct *afs_lock_manager;
static void afs_fl_copy_lock(struct file_lock *new, struct file_lock *fl);
static void afs_fl_release_private(struct file_lock *fl);
static struct workqueue_struct *afs_lock_manager;
static DEFINE_MUTEX(afs_lock_manager_mutex);
static const struct file_lock_operations afs_lock_ops = {
.fl_copy_lock = afs_fl_copy_lock,
.fl_release_private = afs_fl_release_private,
};
/*
* initialise the lock manager thread if it isn't already running
*/
static int afs_init_lock_manager(void)
{
int ret;
ret = 0;
if (!afs_lock_manager) {
mutex_lock(&afs_lock_manager_mutex);
if (!afs_lock_manager) {
afs_lock_manager = alloc_workqueue("kafs_lockd",
WQ_MEM_RECLAIM, 0);
if (!afs_lock_manager)
ret = -ENOMEM;
}
mutex_unlock(&afs_lock_manager_mutex);
}
return ret;
}
/*
* destroy the lock manager thread if it's running
*/
void __exit afs_kill_lock_manager(void)
{
if (afs_lock_manager)
destroy_workqueue(afs_lock_manager);
}
/*
* if the callback is broken on this vnode, then the lock may now be available
*/
@ -98,6 +67,100 @@ static void afs_grant_locks(struct afs_vnode *vnode, struct file_lock *fl)
}
}
/*
* Get a lock on a file
*/
static int afs_set_lock(struct afs_vnode *vnode, struct key *key,
afs_lock_type_t type)
{
struct afs_fs_cursor fc;
int ret;
_enter("%s{%x:%u.%u},%x,%u",
vnode->volume->name,
vnode->fid.vid,
vnode->fid.vnode,
vnode->fid.unique,
key_serial(key), type);
ret = -ERESTARTSYS;
if (afs_begin_vnode_operation(&fc, vnode, key)) {
while (afs_select_fileserver(&fc)) {
fc.cb_break = vnode->cb_break + vnode->cb_s_break;
afs_fs_set_lock(&fc, type);
}
afs_check_for_remote_deletion(&fc, fc.vnode);
afs_vnode_commit_status(&fc, vnode, fc.cb_break);
ret = afs_end_vnode_operation(&fc);
}
_leave(" = %d", ret);
return ret;
}
/*
* Extend a lock on a file
*/
static int afs_extend_lock(struct afs_vnode *vnode, struct key *key)
{
struct afs_fs_cursor fc;
int ret;
_enter("%s{%x:%u.%u},%x",
vnode->volume->name,
vnode->fid.vid,
vnode->fid.vnode,
vnode->fid.unique,
key_serial(key));
ret = -ERESTARTSYS;
if (afs_begin_vnode_operation(&fc, vnode, key)) {
while (afs_select_current_fileserver(&fc)) {
fc.cb_break = vnode->cb_break + vnode->cb_s_break;
afs_fs_extend_lock(&fc);
}
afs_check_for_remote_deletion(&fc, fc.vnode);
afs_vnode_commit_status(&fc, vnode, fc.cb_break);
ret = afs_end_vnode_operation(&fc);
}
_leave(" = %d", ret);
return ret;
}
/*
* Release a lock on a file
*/
static int afs_release_lock(struct afs_vnode *vnode, struct key *key)
{
struct afs_fs_cursor fc;
int ret;
_enter("%s{%x:%u.%u},%x",
vnode->volume->name,
vnode->fid.vid,
vnode->fid.vnode,
vnode->fid.unique,
key_serial(key));
ret = -ERESTARTSYS;
if (afs_begin_vnode_operation(&fc, vnode, key)) {
while (afs_select_current_fileserver(&fc)) {
fc.cb_break = vnode->cb_break + vnode->cb_s_break;
afs_fs_release_lock(&fc);
}
afs_check_for_remote_deletion(&fc, fc.vnode);
afs_vnode_commit_status(&fc, vnode, fc.cb_break);
ret = afs_end_vnode_operation(&fc);
}
_leave(" = %d", ret);
return ret;
}
/*
* do work for a lock, including:
* - probing for a lock we're waiting on but didn't get immediately
@ -122,7 +185,7 @@ void afs_lock_work(struct work_struct *work)
/* attempt to release the server lock; if it fails, we just
* wait 5 minutes and it'll time out anyway */
ret = afs_vnode_release_lock(vnode, vnode->unlock_key);
ret = afs_release_lock(vnode, vnode->unlock_key);
if (ret < 0)
printk(KERN_WARNING "AFS:"
" Failed to release lock on {%x:%x} error %d\n",
@ -143,10 +206,10 @@ void afs_lock_work(struct work_struct *work)
BUG();
fl = list_entry(vnode->granted_locks.next,
struct file_lock, fl_u.afs.link);
key = key_get(fl->fl_file->private_data);
key = key_get(afs_file_key(fl->fl_file));
spin_unlock(&vnode->lock);
ret = afs_vnode_extend_lock(vnode, key);
ret = afs_extend_lock(vnode, key);
clear_bit(AFS_VNODE_LOCKING, &vnode->flags);
key_put(key);
switch (ret) {
@ -177,12 +240,12 @@ void afs_lock_work(struct work_struct *work)
BUG();
fl = list_entry(vnode->pending_locks.next,
struct file_lock, fl_u.afs.link);
key = key_get(fl->fl_file->private_data);
key = key_get(afs_file_key(fl->fl_file));
type = (fl->fl_type == F_RDLCK) ?
AFS_LOCK_READ : AFS_LOCK_WRITE;
spin_unlock(&vnode->lock);
ret = afs_vnode_set_lock(vnode, key, type);
ret = afs_set_lock(vnode, key, type);
clear_bit(AFS_VNODE_LOCKING, &vnode->flags);
switch (ret) {
case -EWOULDBLOCK:
@ -213,7 +276,7 @@ void afs_lock_work(struct work_struct *work)
clear_bit(AFS_VNODE_READLOCKED, &vnode->flags);
clear_bit(AFS_VNODE_WRITELOCKED, &vnode->flags);
spin_unlock(&vnode->lock);
afs_vnode_release_lock(vnode, key);
afs_release_lock(vnode, key);
if (!list_empty(&vnode->pending_locks))
afs_lock_may_be_available(vnode);
}
@ -255,7 +318,7 @@ static int afs_do_setlk(struct file *file, struct file_lock *fl)
struct inode *inode = file_inode(file);
struct afs_vnode *vnode = AFS_FS_I(inode);
afs_lock_type_t type;
struct key *key = file->private_data;
struct key *key = afs_file_key(file);
int ret;
_enter("{%x:%u},%u", vnode->fid.vid, vnode->fid.vnode, fl->fl_type);
@ -264,10 +327,6 @@ static int afs_do_setlk(struct file *file, struct file_lock *fl)
if (fl->fl_start != 0 || fl->fl_end != OFFSET_MAX)
return -EINVAL;
ret = afs_init_lock_manager();
if (ret < 0)
return ret;
fl->fl_ops = &afs_lock_ops;
INIT_LIST_HEAD(&fl->fl_u.afs.link);
fl->fl_u.afs.state = AFS_LOCK_PENDING;
@ -278,7 +337,7 @@ static int afs_do_setlk(struct file *file, struct file_lock *fl)
/* make sure we've got a callback on this file and that our view of the
* data version is up to date */
ret = afs_vnode_fetch_status(vnode, NULL, key);
ret = afs_validate(vnode, key);
if (ret < 0)
goto error;
@ -315,7 +374,7 @@ static int afs_do_setlk(struct file *file, struct file_lock *fl)
set_bit(AFS_VNODE_LOCKING, &vnode->flags);
spin_unlock(&vnode->lock);
ret = afs_vnode_set_lock(vnode, key, type);
ret = afs_set_lock(vnode, key, type);
clear_bit(AFS_VNODE_LOCKING, &vnode->flags);
switch (ret) {
case 0:
@ -418,7 +477,7 @@ given_lock:
/* again, make sure we've got a callback on this file and, again, make
* sure that our view of the data version is up to date (we ignore
* errors incurred here and deal with the consequences elsewhere) */
afs_vnode_fetch_status(vnode, NULL, key);
afs_validate(vnode, key);
error:
spin_unlock(&inode->i_lock);
@ -441,7 +500,7 @@ vfs_rejected_lock:
static int afs_do_unlk(struct file *file, struct file_lock *fl)
{
struct afs_vnode *vnode = AFS_FS_I(file->f_mapping->host);
struct key *key = file->private_data;
struct key *key = afs_file_key(file);
int ret;
_enter("{%x:%u},%u", vnode->fid.vid, vnode->fid.vnode, fl->fl_type);
@ -476,7 +535,7 @@ static int afs_do_unlk(struct file *file, struct file_lock *fl)
static int afs_do_getlk(struct file *file, struct file_lock *fl)
{
struct afs_vnode *vnode = AFS_FS_I(file->f_mapping->host);
struct key *key = file->private_data;
struct key *key = afs_file_key(file);
int ret, lock_count;
_enter("");
@ -490,7 +549,7 @@ static int afs_do_getlk(struct file *file, struct file_lock *fl)
posix_test_lock(file, fl);
if (fl->fl_type == F_UNLCK) {
/* no local locks; consult the server */
ret = afs_vnode_fetch_status(vnode, NULL, key);
ret = afs_fetch_status(vnode, key);
if (ret < 0)
goto error;
lock_count = vnode->status.lock_count;

830
fs/afs/fsclient.c
View File

File diff suppressed because it is too large Load Diff

177
fs/afs/inode.c
View File

@ -23,11 +23,6 @@
#include <linux/namei.h>
#include "internal.h"
struct afs_iget_data {
struct afs_fid fid;
struct afs_volume *volume; /* volume on which resides */
};
static const struct inode_operations afs_symlink_inode_operations = {
.get_link = page_get_link,
.listxattr = afs_listxattr,
@ -39,6 +34,7 @@ static const struct inode_operations afs_symlink_inode_operations = {
static int afs_inode_map_status(struct afs_vnode *vnode, struct key *key)
{
struct inode *inode = AFS_VNODE_TO_I(vnode);
bool changed;
_debug("FS: ft=%d lk=%d sz=%llu ver=%Lu mod=%hu",
vnode->status.type,
@ -47,6 +43,8 @@ static int afs_inode_map_status(struct afs_vnode *vnode, struct key *key)
vnode->status.data_version,
vnode->status.mode);
read_seqlock_excl(&vnode->cb_lock);
switch (vnode->status.type) {
case AFS_FTYPE_FILE:
inode->i_mode = S_IFREG | vnode->status.mode;
@ -63,9 +61,7 @@ static int afs_inode_map_status(struct afs_vnode *vnode, struct key *key)
if ((vnode->status.mode & 0777) == 0644) {
inode->i_flags |= S_AUTOMOUNT;
spin_lock(&vnode->lock);
set_bit(AFS_VNODE_MOUNTPOINT, &vnode->flags);
spin_unlock(&vnode->lock);
inode->i_mode = S_IFDIR | 0555;
inode->i_op = &afs_mntpt_inode_operations;
@ -78,13 +74,11 @@ static int afs_inode_map_status(struct afs_vnode *vnode, struct key *key)
break;
default:
printk("kAFS: AFS vnode with undefined type\n");
read_sequnlock_excl(&vnode->cb_lock);
return -EBADMSG;
}
#ifdef CONFIG_AFS_FSCACHE
if (vnode->status.size != inode->i_size)
fscache_attr_changed(vnode->cache);
#endif
changed = (vnode->status.size != inode->i_size);
set_nlink(inode, vnode->status.nlink);
inode->i_uid = vnode->status.owner;
@ -97,13 +91,49 @@ static int afs_inode_map_status(struct afs_vnode *vnode, struct key *key)
inode->i_generation = vnode->fid.unique;
inode->i_version = vnode->status.data_version;
inode->i_mapping->a_ops = &afs_fs_aops;
read_sequnlock_excl(&vnode->cb_lock);
#ifdef CONFIG_AFS_FSCACHE
if (changed)
fscache_attr_changed(vnode->cache);
#endif
return 0;
}
/*
* Fetch file status from the volume.
*/
int afs_fetch_status(struct afs_vnode *vnode, struct key *key)
{
struct afs_fs_cursor fc;
int ret;
_enter("%s,{%x:%u.%u,S=%lx}",
vnode->volume->name,
vnode->fid.vid, vnode->fid.vnode, vnode->fid.unique,
vnode->flags);
ret = -ERESTARTSYS;
if (afs_begin_vnode_operation(&fc, vnode, key)) {
while (afs_select_fileserver(&fc)) {
fc.cb_break = vnode->cb_break + vnode->cb_s_break;
afs_fs_fetch_file_status(&fc, NULL);
}
afs_check_for_remote_deletion(&fc, fc.vnode);
afs_vnode_commit_status(&fc, vnode, fc.cb_break);
ret = afs_end_vnode_operation(&fc);
}
_leave(" = %d", ret);
return ret;
}
/*
* iget5() comparator
*/
static int afs_iget5_test(struct inode *inode, void *opaque)
int afs_iget5_test(struct inode *inode, void *opaque)
{
struct afs_iget_data *data = opaque;
@ -204,7 +234,7 @@ struct inode *afs_iget_autocell(struct inode *dir, const char *dev_name,
*/
struct inode *afs_iget(struct super_block *sb, struct key *key,
struct afs_fid *fid, struct afs_file_status *status,
struct afs_callback *cb)
struct afs_callback *cb, struct afs_cb_interest *cbi)
{
struct afs_iget_data data = { .fid = *fid };
struct afs_super_info *as;
@ -237,8 +267,7 @@ struct inode *afs_iget(struct super_block *sb, struct key *key,
if (!status) {
/* it's a remotely extant inode */
set_bit(AFS_VNODE_CB_BROKEN, &vnode->flags);
ret = afs_vnode_fetch_status(vnode, NULL, key);
ret = afs_fetch_status(vnode, key);
if (ret < 0)
goto bad_inode;
} else {
@ -249,16 +278,17 @@ struct inode *afs_iget(struct super_block *sb, struct key *key,
/* it's a symlink we just created (the fileserver
* didn't give us a callback) */
vnode->cb_version = 0;
vnode->cb_expiry = 0;
vnode->cb_type = 0;
vnode->cb_expires = ktime_get_real_seconds();
vnode->cb_expires_at = 0;
} else {
vnode->cb_version = cb->version;
vnode->cb_expiry = cb->expiry;
vnode->cb_type = cb->type;
vnode->cb_expires = vnode->cb_expiry +
ktime_get_real_seconds();
vnode->cb_expires_at = cb->expiry;
vnode->cb_interest = afs_get_cb_interest(cbi);
set_bit(AFS_VNODE_CB_PROMISED, &vnode->flags);
}
vnode->cb_expires_at += ktime_get_real_seconds();
}
/* set up caching before mapping the status, as map-status reads the
@ -320,25 +350,34 @@ void afs_zap_data(struct afs_vnode *vnode)
*/
int afs_validate(struct afs_vnode *vnode, struct key *key)
{
time64_t now = ktime_get_real_seconds();
bool valid = false;
int ret;
_enter("{v={%x:%u} fl=%lx},%x",
vnode->fid.vid, vnode->fid.vnode, vnode->flags,
key_serial(key));
if (vnode->cb_promised &&
!test_bit(AFS_VNODE_CB_BROKEN, &vnode->flags) &&
!test_bit(AFS_VNODE_MODIFIED, &vnode->flags) &&
!test_bit(AFS_VNODE_ZAP_DATA, &vnode->flags)) {
if (vnode->cb_expires < ktime_get_real_seconds() + 10) {
_debug("callback expired");
set_bit(AFS_VNODE_CB_BROKEN, &vnode->flags);
} else {
goto valid;
/* Quickly check the callback state. Ideally, we'd use read_seqbegin
* here, but we have no way to pass the net namespace to the RCU
* cleanup for the server record.
*/
read_seqlock_excl(&vnode->cb_lock);
if (test_bit(AFS_VNODE_CB_PROMISED, &vnode->flags)) {
if (vnode->cb_s_break != vnode->cb_interest->server->cb_s_break) {
vnode->cb_s_break = vnode->cb_interest->server->cb_s_break;
} else if (!test_bit(AFS_VNODE_DIR_MODIFIED, &vnode->flags) &&
!test_bit(AFS_VNODE_ZAP_DATA, &vnode->flags) &&
vnode->cb_expires_at - 10 > now) {
valid = true;
}
} else if (test_bit(AFS_VNODE_DELETED, &vnode->flags)) {
valid = true;
}
if (test_bit(AFS_VNODE_DELETED, &vnode->flags))
read_sequnlock_excl(&vnode->cb_lock);
if (valid)
goto valid;
mutex_lock(&vnode->validate_lock);
@ -347,12 +386,16 @@ int afs_validate(struct afs_vnode *vnode, struct key *key)
* a new promise - note that if the (parent) directory's metadata was
* changed then the security may be different and we may no longer have
* access */
if (!vnode->cb_promised ||
test_bit(AFS_VNODE_CB_BROKEN, &vnode->flags)) {
if (!test_bit(AFS_VNODE_CB_PROMISED, &vnode->flags)) {
_debug("not promised");
ret = afs_vnode_fetch_status(vnode, NULL, key);
if (ret < 0)
ret = afs_fetch_status(vnode, key);
if (ret < 0) {
if (ret == -ENOENT) {
set_bit(AFS_VNODE_DELETED, &vnode->flags);
ret = -ESTALE;
}
goto error_unlock;
}
_debug("new promise [fl=%lx]", vnode->flags);
}
@ -367,7 +410,7 @@ int afs_validate(struct afs_vnode *vnode, struct key *key)
if (test_and_clear_bit(AFS_VNODE_ZAP_DATA, &vnode->flags))
afs_zap_data(vnode);
clear_bit(AFS_VNODE_MODIFIED, &vnode->flags);
clear_bit(AFS_VNODE_DIR_MODIFIED, &vnode->flags);
mutex_unlock(&vnode->validate_lock);
valid:
_leave(" = 0");
@ -386,10 +429,17 @@ int afs_getattr(const struct path *path, struct kstat *stat,
u32 request_mask, unsigned int query_flags)
{
struct inode *inode = d_inode(path->dentry);
struct afs_vnode *vnode = AFS_FS_I(inode);
int seq = 0;
_enter("{ ino=%lu v=%u }", inode->i_ino, inode->i_generation);
generic_fillattr(inode, stat);
do {
read_seqbegin_or_lock(&vnode->cb_lock, &seq);
generic_fillattr(inode, stat);
} while (need_seqretry(&vnode->cb_lock, seq));
done_seqretry(&vnode->cb_lock, seq);
return 0;
}
@ -411,18 +461,14 @@ int afs_drop_inode(struct inode *inode)
*/
void afs_evict_inode(struct inode *inode)
{
struct afs_permits *permits;
struct afs_vnode *vnode;
vnode = AFS_FS_I(inode);
_enter("{%x:%u.%d} v=%u x=%u t=%u }",
_enter("{%x:%u.%d}",
vnode->fid.vid,
vnode->fid.vnode,
vnode->fid.unique,
vnode->cb_version,
vnode->cb_expiry,
vnode->cb_type);
vnode->fid.unique);
_debug("CLEAR INODE %p", inode);
@ -431,31 +477,24 @@ void afs_evict_inode(struct inode *inode)
truncate_inode_pages_final(&inode->i_data);
clear_inode(inode);
afs_give_up_callback(vnode);
if (vnode->server) {
spin_lock(&vnode->server->fs_lock);
rb_erase(&vnode->server_rb, &vnode->server->fs_vnodes);
spin_unlock(&vnode->server->fs_lock);
afs_put_server(vnode->server);
vnode->server = NULL;
if (vnode->cb_interest) {
afs_put_cb_interest(afs_i2net(inode), vnode->cb_interest);
vnode->cb_interest = NULL;
}
ASSERT(list_empty(&vnode->writebacks));
ASSERT(!vnode->cb_promised);
while (!list_empty(&vnode->wb_keys)) {
struct afs_wb_key *wbk = list_entry(vnode->wb_keys.next,
struct afs_wb_key, vnode_link);
list_del(&wbk->vnode_link);
afs_put_wb_key(wbk);
}
#ifdef CONFIG_AFS_FSCACHE
fscache_relinquish_cookie(vnode->cache, 0);
vnode->cache = NULL;
#endif
mutex_lock(&vnode->permits_lock);
permits = vnode->permits;
RCU_INIT_POINTER(vnode->permits, NULL);
mutex_unlock(&vnode->permits_lock);
if (permits)
call_rcu(&permits->rcu, afs_zap_permits);
afs_put_permits(vnode->permit_cache);
_leave("");
}
@ -464,6 +503,7 @@ void afs_evict_inode(struct inode *inode)
*/
int afs_setattr(struct dentry *dentry, struct iattr *attr)
{
struct afs_fs_cursor fc;
struct afs_vnode *vnode = AFS_FS_I(d_inode(dentry));
struct key *key;
int ret;
@ -479,13 +519,11 @@ int afs_setattr(struct dentry *dentry, struct iattr *attr)
}
/* flush any dirty data outstanding on a regular file */
if (S_ISREG(vnode->vfs_inode.i_mode)) {
if (S_ISREG(vnode->vfs_inode.i_mode))
filemap_write_and_wait(vnode->vfs_inode.i_mapping);
afs_writeback_all(vnode);
}
if (attr->ia_valid & ATTR_FILE) {
key = attr->ia_file->private_data;
key = afs_file_key(attr->ia_file);
} else {
key = afs_request_key(vnode->volume->cell);
if (IS_ERR(key)) {
@ -494,7 +532,18 @@ int afs_setattr(struct dentry *dentry, struct iattr *attr)
}
}
ret = afs_vnode_setattr(vnode, key, attr);
ret = -ERESTARTSYS;
if (afs_begin_vnode_operation(&fc, vnode, key)) {
while (afs_select_fileserver(&fc)) {
fc.cb_break = vnode->cb_break + vnode->cb_s_break;
afs_fs_setattr(&fc, attr);
}
afs_check_for_remote_deletion(&fc, fc.vnode);
afs_vnode_commit_status(&fc, vnode, fc.cb_break);
ret = afs_end_vnode_operation(&fc);
}
if (!(attr->ia_valid & ATTR_FILE))
key_put(key);

931
fs/afs/internal.h
View File

File diff suppressed because it is too large Load Diff

145
fs/afs/main.c
View File

@ -31,30 +31,101 @@ static char *rootcell;
module_param(rootcell, charp, 0);
MODULE_PARM_DESC(rootcell, "root AFS cell name and VL server IP addr list");
struct afs_uuid afs_uuid;
struct workqueue_struct *afs_wq;
struct afs_net __afs_net;
/*
* Initialise an AFS network namespace record.
*/
static int __net_init afs_net_init(struct afs_net *net)
{
int ret;
net->live = true;
generate_random_uuid((unsigned char *)&net->uuid);
INIT_WORK(&net->charge_preallocation_work, afs_charge_preallocation);
mutex_init(&net->socket_mutex);
net->cells = RB_ROOT;
seqlock_init(&net->cells_lock);
INIT_WORK(&net->cells_manager, afs_manage_cells);
timer_setup(&net->cells_timer, afs_cells_timer, 0);
spin_lock_init(&net->proc_cells_lock);
INIT_LIST_HEAD(&net->proc_cells);
seqlock_init(&net->fs_lock);
net->fs_servers = RB_ROOT;
INIT_LIST_HEAD(&net->fs_updates);
INIT_HLIST_HEAD(&net->fs_proc);
INIT_HLIST_HEAD(&net->fs_addresses4);
INIT_HLIST_HEAD(&net->fs_addresses6);
seqlock_init(&net->fs_addr_lock);
INIT_WORK(&net->fs_manager, afs_manage_servers);
timer_setup(&net->fs_timer, afs_servers_timer, 0);
/* Register the /proc stuff */
ret = afs_proc_init(net);
if (ret < 0)
goto error_proc;
/* Initialise the cell DB */
ret = afs_cell_init(net, rootcell);
if (ret < 0)
goto error_cell_init;
/* Create the RxRPC transport */
ret = afs_open_socket(net);
if (ret < 0)
goto error_open_socket;
return 0;
error_open_socket:
net->live = false;
afs_cell_purge(net);
afs_purge_servers(net);
error_cell_init:
net->live = false;
afs_proc_cleanup(net);
error_proc:
net->live = false;
return ret;
}
/*
* Clean up and destroy an AFS network namespace record.
*/
static void __net_exit afs_net_exit(struct afs_net *net)
{
net->live = false;
afs_cell_purge(net);
afs_purge_servers(net);
afs_close_socket(net);
afs_proc_cleanup(net);
}
/*
* initialise the AFS client FS module
*/
static int __init afs_init(void)
{
int ret;
int ret = -ENOMEM;
printk(KERN_INFO "kAFS: Red Hat AFS client v0.1 registering.\n");
generate_random_uuid((unsigned char *)&afs_uuid);
/* create workqueue */
ret = -ENOMEM;
afs_wq = alloc_workqueue("afs", 0, 0);
if (!afs_wq)
return ret;
/* register the /proc stuff */
ret = afs_proc_init();
if (ret < 0)
goto error_proc;
goto error_afs_wq;
afs_async_calls = alloc_workqueue("kafsd", WQ_MEM_RECLAIM, 0);
if (!afs_async_calls)
goto error_async;
afs_lock_manager = alloc_workqueue("kafs_lockd", WQ_MEM_RECLAIM, 0);
if (!afs_lock_manager)
goto error_lockmgr;
#ifdef CONFIG_AFS_FSCACHE
/* we want to be able to cache */
@ -63,25 +134,9 @@ static int __init afs_init(void)
goto error_cache;
#endif
/* initialise the cell DB */
ret = afs_cell_init(rootcell);
ret = afs_net_init(&__afs_net);
if (ret < 0)
goto error_cell_init;
/* initialise the VL update process */
ret = afs_vlocation_update_init();
if (ret < 0)
goto error_vl_update_init;
/* initialise the callback update process */
ret = afs_callback_update_init();
if (ret < 0)
goto error_callback_update_init;
/* create the RxRPC transport */
ret = afs_open_socket();
if (ret < 0)
goto error_open_socket;
goto error_net;
/* register the filesystems */
ret = afs_fs_init();
@ -91,21 +146,18 @@ static int __init afs_init(void)
return ret;
error_fs:
afs_close_socket();
error_open_socket:
afs_callback_update_kill();
error_callback_update_init:
afs_vlocation_purge();
error_vl_update_init:
afs_cell_purge();
error_cell_init:
afs_net_exit(&__afs_net);
error_net:
#ifdef CONFIG_AFS_FSCACHE
fscache_unregister_netfs(&afs_cache_netfs);
error_cache:
#endif
afs_proc_cleanup();
error_proc:
destroy_workqueue(afs_lock_manager);
error_lockmgr:
destroy_workqueue(afs_async_calls);
error_async:
destroy_workqueue(afs_wq);
error_afs_wq:
rcu_barrier();
printk(KERN_ERR "kAFS: failed to register: %d\n", ret);
return ret;
@ -124,17 +176,14 @@ static void __exit afs_exit(void)
printk(KERN_INFO "kAFS: Red Hat AFS client v0.1 unregistering.\n");
afs_fs_exit();
afs_kill_lock_manager();
afs_close_socket();
afs_purge_servers();
afs_callback_update_kill();
afs_vlocation_purge();
destroy_workqueue(afs_wq);
afs_cell_purge();
afs_net_exit(&__afs_net);
#ifdef CONFIG_AFS_FSCACHE
fscache_unregister_netfs(&afs_cache_netfs);
#endif
afs_proc_cleanup();
destroy_workqueue(afs_lock_manager);
destroy_workqueue(afs_async_calls);
destroy_workqueue(afs_wq);
afs_clean_up_permit_cache();
rcu_barrier();
}

38
fs/afs/misc.c
View File

@ -21,12 +21,12 @@
int afs_abort_to_error(u32 abort_code)
{
switch (abort_code) {
/* low errno codes inserted into abort namespace */
/* Low errno codes inserted into abort namespace */
case 13: return -EACCES;
case 27: return -EFBIG;
case 30: return -EROFS;
/* VICE "special error" codes; 101 - 111 */
/* VICE "special error" codes; 101 - 111 */
case VSALVAGE: return -EIO;
case VNOVNODE: return -ENOENT;
case VNOVOL: return -ENOMEDIUM;
@ -39,7 +39,37 @@ int afs_abort_to_error(u32 abort_code)
case VBUSY: return -EBUSY;
case VMOVED: return -ENXIO;
/* Unified AFS error table; ET "uae" == 0x2f6df00 */
/* Volume Location server errors */
case AFSVL_IDEXIST: return -EEXIST;
case AFSVL_IO: return -EREMOTEIO;
case AFSVL_NAMEEXIST: return -EEXIST;
case AFSVL_CREATEFAIL: return -EREMOTEIO;
case AFSVL_NOENT: return -ENOMEDIUM;
case AFSVL_EMPTY: return -ENOMEDIUM;
case AFSVL_ENTDELETED: return -ENOMEDIUM;
case AFSVL_BADNAME: return -EINVAL;
case AFSVL_BADINDEX: return -EINVAL;
case AFSVL_BADVOLTYPE: return -EINVAL;
case AFSVL_BADSERVER: return -EINVAL;
case AFSVL_BADPARTITION: return -EINVAL;
case AFSVL_REPSFULL: return -EFBIG;
case AFSVL_NOREPSERVER: return -ENOENT;
case AFSVL_DUPREPSERVER: return -EEXIST;
case AFSVL_RWNOTFOUND: return -ENOENT;
case AFSVL_BADREFCOUNT: return -EINVAL;
case AFSVL_SIZEEXCEEDED: return -EINVAL;
case AFSVL_BADENTRY: return -EINVAL;
case AFSVL_BADVOLIDBUMP: return -EINVAL;
case AFSVL_IDALREADYHASHED: return -EINVAL;
case AFSVL_ENTRYLOCKED: return -EBUSY;
case AFSVL_BADVOLOPER: return -EBADRQC;
case AFSVL_BADRELLOCKTYPE: return -EINVAL;
case AFSVL_RERELEASE: return -EREMOTEIO;
case AFSVL_BADSERVERFLAG: return -EINVAL;
case AFSVL_PERM: return -EACCES;
case AFSVL_NOMEM: return -EREMOTEIO;
/* Unified AFS error table; ET "uae" == 0x2f6df00 */
case 0x2f6df00: return -EPERM;
case 0x2f6df01: return -ENOENT;
case 0x2f6df04: return -EIO;
@ -68,7 +98,7 @@ int afs_abort_to_error(u32 abort_code)
case 0x2f6df6c: return -ETIMEDOUT;
case 0x2f6df78: return -EDQUOT;
/* RXKAD abort codes; from include/rxrpc/packet.h. ET "RXK" == 0x1260B00 */
/* RXKAD abort codes; from include/rxrpc/packet.h. ET "RXK" == 0x1260B00 */
case RXKADINCONSISTENCY: return -EPROTO;
case RXKADPACKETSHORT: return -EPROTO;
case RXKADLEVELFAIL: return -EKEYREJECTED;

245
fs/afs/proc.c
View File

@ -17,8 +17,15 @@
#include <linux/uaccess.h>
#include "internal.h"
static struct proc_dir_entry *proc_afs;
static inline struct afs_net *afs_proc2net(struct file *f)
{
return &__afs_net;
}
static inline struct afs_net *afs_seq2net(struct seq_file *m)
{
return &__afs_net; // TODO: use seq_file_net(m)
}
static int afs_proc_cells_open(struct inode *inode, struct file *file);
static void *afs_proc_cells_start(struct seq_file *p, loff_t *pos);
@ -98,22 +105,22 @@ static const struct file_operations afs_proc_cell_vlservers_fops = {
.release = seq_release,
};
static int afs_proc_cell_servers_open(struct inode *inode, struct file *file);
static void *afs_proc_cell_servers_start(struct seq_file *p, loff_t *pos);
static void *afs_proc_cell_servers_next(struct seq_file *p, void *v,
static int afs_proc_servers_open(struct inode *inode, struct file *file);
static void *afs_proc_servers_start(struct seq_file *p, loff_t *pos);
static void *afs_proc_servers_next(struct seq_file *p, void *v,
loff_t *pos);
static void afs_proc_cell_servers_stop(struct seq_file *p, void *v);
static int afs_proc_cell_servers_show(struct seq_file *m, void *v);
static void afs_proc_servers_stop(struct seq_file *p, void *v);
static int afs_proc_servers_show(struct seq_file *m, void *v);
static const struct seq_operations afs_proc_cell_servers_ops = {
.start = afs_proc_cell_servers_start,
.next = afs_proc_cell_servers_next,
.stop = afs_proc_cell_servers_stop,
.show = afs_proc_cell_servers_show,
static const struct seq_operations afs_proc_servers_ops = {
.start = afs_proc_servers_start,
.next = afs_proc_servers_next,
.stop = afs_proc_servers_stop,
.show = afs_proc_servers_show,
};
static const struct file_operations afs_proc_cell_servers_fops = {
.open = afs_proc_cell_servers_open,
static const struct file_operations afs_proc_servers_fops = {
.open = afs_proc_servers_open,
.read = seq_read,
.llseek = seq_lseek,
.release = seq_release,
@ -122,23 +129,24 @@ static const struct file_operations afs_proc_cell_servers_fops = {
/*
* initialise the /proc/fs/afs/ directory
*/
int afs_proc_init(void)
int afs_proc_init(struct afs_net *net)
{
_enter("");
proc_afs = proc_mkdir("fs/afs", NULL);
if (!proc_afs)
net->proc_afs = proc_mkdir("fs/afs", NULL);
if (!net->proc_afs)
goto error_dir;
if (!proc_create("cells", 0644, proc_afs, &afs_proc_cells_fops) ||
!proc_create("rootcell", 0644, proc_afs, &afs_proc_rootcell_fops))
if (!proc_create("cells", 0644, net->proc_afs, &afs_proc_cells_fops) ||
!proc_create("rootcell", 0644, net->proc_afs, &afs_proc_rootcell_fops) ||
!proc_create("servers", 0644, net->proc_afs, &afs_proc_servers_fops))
goto error_tree;
_leave(" = 0");
return 0;
error_tree:
remove_proc_subtree("fs/afs", NULL);
proc_remove(net->proc_afs);
error_dir:
_leave(" = -ENOMEM");
return -ENOMEM;
@ -147,9 +155,10 @@ error_dir:
/*
* clean up the /proc/fs/afs/ directory
*/
void afs_proc_cleanup(void)
void afs_proc_cleanup(struct afs_net *net)
{
remove_proc_subtree("fs/afs", NULL);
proc_remove(net->proc_afs);
net->proc_afs = NULL;
}
/*
@ -166,7 +175,6 @@ static int afs_proc_cells_open(struct inode *inode, struct file *file)
m = file->private_data;
m->private = PDE_DATA(inode);
return 0;
}
@ -176,25 +184,28 @@ static int afs_proc_cells_open(struct inode *inode, struct file *file)
*/
static void *afs_proc_cells_start(struct seq_file *m, loff_t *_pos)
{
/* lock the list against modification */
down_read(&afs_proc_cells_sem);
return seq_list_start_head(&afs_proc_cells, *_pos);
struct afs_net *net = afs_seq2net(m);
rcu_read_lock();
return seq_list_start_head(&net->proc_cells, *_pos);
}
/*
* move to next cell in cells list
*/
static void *afs_proc_cells_next(struct seq_file *p, void *v, loff_t *pos)
static void *afs_proc_cells_next(struct seq_file *m, void *v, loff_t *pos)
{
return seq_list_next(v, &afs_proc_cells, pos);
struct afs_net *net = afs_seq2net(m);
return seq_list_next(v, &net->proc_cells, pos);
}
/*
* clean up after reading from the cells list
*/
static void afs_proc_cells_stop(struct seq_file *p, void *v)
static void afs_proc_cells_stop(struct seq_file *m, void *v)
{
up_read(&afs_proc_cells_sem);
rcu_read_unlock();
}
/*
@ -203,16 +214,16 @@ static void afs_proc_cells_stop(struct seq_file *p, void *v)
static int afs_proc_cells_show(struct seq_file *m, void *v)
{
struct afs_cell *cell = list_entry(v, struct afs_cell, proc_link);
struct afs_net *net = afs_seq2net(m);
if (v == &afs_proc_cells) {
if (v == &net->proc_cells) {
/* display header on line 1 */
seq_puts(m, "USE NAME\n");
return 0;
}
/* display one cell per line on subsequent lines */
seq_printf(m, "%3d %s\n",
atomic_read(&cell->usage), cell->name);
seq_printf(m, "%3u %s\n", atomic_read(&cell->usage), cell->name);
return 0;
}
@ -223,6 +234,7 @@ static int afs_proc_cells_show(struct seq_file *m, void *v)
static ssize_t afs_proc_cells_write(struct file *file, const char __user *buf,
size_t size, loff_t *_pos)
{
struct afs_net *net = afs_proc2net(file);
char *kbuf, *name, *args;
int ret;
@ -264,13 +276,13 @@ static ssize_t afs_proc_cells_write(struct file *file, const char __user *buf,
if (strcmp(kbuf, "add") == 0) {
struct afs_cell *cell;
cell = afs_cell_create(name, strlen(name), args, false);
cell = afs_lookup_cell(net, name, strlen(name), args, true);
if (IS_ERR(cell)) {
ret = PTR_ERR(cell);
goto done;
}
afs_put_cell(cell);
set_bit(AFS_CELL_FL_NO_GC, &cell->flags);
printk("kAFS: Added new cell '%s'\n", name);
} else {
goto inval;
@ -303,6 +315,7 @@ static ssize_t afs_proc_rootcell_write(struct file *file,
const char __user *buf,
size_t size, loff_t *_pos)
{
struct afs_net *net = afs_proc2net(file);
char *kbuf, *s;
int ret;
@ -322,7 +335,7 @@ static ssize_t afs_proc_rootcell_write(struct file *file,
/* determine command to perform */
_debug("rootcell=%s", kbuf);
ret = afs_cell_init(kbuf);
ret = afs_cell_init(net, kbuf);
if (ret >= 0)
ret = size; /* consume everything, always */
@ -334,29 +347,27 @@ static ssize_t afs_proc_rootcell_write(struct file *file,
/*
* initialise /proc/fs/afs/<cell>/
*/
int afs_proc_cell_setup(struct afs_cell *cell)
int afs_proc_cell_setup(struct afs_net *net, struct afs_cell *cell)
{
struct proc_dir_entry *dir;
_enter("%p{%s}", cell, cell->name);
_enter("%p{%s},%p", cell, cell->name, net->proc_afs);
dir = proc_mkdir(cell->name, proc_afs);
dir = proc_mkdir(cell->name, net->proc_afs);
if (!dir)
goto error_dir;
if (!proc_create_data("servers", 0, dir,
&afs_proc_cell_servers_fops, cell) ||
!proc_create_data("vlservers", 0, dir,
&afs_proc_cell_vlservers_fops, cell) ||
if (!proc_create_data("vlservers", 0, dir,
&afs_proc_cell_vlservers_fops, cell) ||
!proc_create_data("volumes", 0, dir,
&afs_proc_cell_volumes_fops, cell))
&afs_proc_cell_volumes_fops, cell))
goto error_tree;
_leave(" = 0");
return 0;
error_tree:
remove_proc_subtree(cell->name, proc_afs);
remove_proc_subtree(cell->name, net->proc_afs);
error_dir:
_leave(" = -ENOMEM");
return -ENOMEM;
@ -365,11 +376,11 @@ error_dir:
/*
* remove /proc/fs/afs/<cell>/
*/
void afs_proc_cell_remove(struct afs_cell *cell)
void afs_proc_cell_remove(struct afs_net *net, struct afs_cell *cell)
{
_enter("");
remove_proc_subtree(cell->name, proc_afs);
remove_proc_subtree(cell->name, net->proc_afs);
_leave("");
}
@ -407,9 +418,8 @@ static void *afs_proc_cell_volumes_start(struct seq_file *m, loff_t *_pos)
_enter("cell=%p pos=%Ld", cell, *_pos);
/* lock the list against modification */
down_read(&cell->vl_sem);
return seq_list_start_head(&cell->vl_list, *_pos);
read_lock(&cell->proc_lock);
return seq_list_start_head(&cell->proc_volumes, *_pos);
}
/*
@ -421,7 +431,7 @@ static void *afs_proc_cell_volumes_next(struct seq_file *p, void *v,
struct afs_cell *cell = p->private;
_enter("cell=%p pos=%Ld", cell, *_pos);
return seq_list_next(v, &cell->vl_list, _pos);
return seq_list_next(v, &cell->proc_volumes, _pos);
}
/*
@ -431,17 +441,13 @@ static void afs_proc_cell_volumes_stop(struct seq_file *p, void *v)
{
struct afs_cell *cell = p->private;
up_read(&cell->vl_sem);
read_unlock(&cell->proc_lock);
}
static const char afs_vlocation_states[][4] = {
[AFS_VL_NEW] = "New",
[AFS_VL_CREATING] = "Crt",
[AFS_VL_VALID] = "Val",
[AFS_VL_NO_VOLUME] = "NoV",
[AFS_VL_UPDATING] = "Upd",
[AFS_VL_VOLUME_DELETED] = "Del",
[AFS_VL_UNCERTAIN] = "Unc",
static const char afs_vol_types[3][3] = {
[AFSVL_RWVOL] = "RW",
[AFSVL_ROVOL] = "RO",
[AFSVL_BACKVOL] = "BK",
};
/*
@ -450,23 +456,17 @@ static const char afs_vlocation_states[][4] = {
static int afs_proc_cell_volumes_show(struct seq_file *m, void *v)
{
struct afs_cell *cell = m->private;
struct afs_vlocation *vlocation =
list_entry(v, struct afs_vlocation, link);
struct afs_volume *vol = list_entry(v, struct afs_volume, proc_link);
/* display header on line 1 */
if (v == &cell->vl_list) {
seq_puts(m, "USE STT VLID[0] VLID[1] VLID[2] NAME\n");
/* Display header on line 1 */
if (v == &cell->proc_volumes) {
seq_puts(m, "USE VID TY\n");
return 0;
}
/* display one cell per line on subsequent lines */
seq_printf(m, "%3d %s %08x %08x %08x %s\n",
atomic_read(&vlocation->usage),
afs_vlocation_states[vlocation->state],
vlocation->vldb.vid[0],
vlocation->vldb.vid[1],
vlocation->vldb.vid[2],
vlocation->vldb.name);
seq_printf(m, "%3d %08x %s\n",
atomic_read(&vol->usage), vol->vid,
afs_vol_types[vol->type]);
return 0;
}
@ -501,23 +501,23 @@ static int afs_proc_cell_vlservers_open(struct inode *inode, struct file *file)
*/
static void *afs_proc_cell_vlservers_start(struct seq_file *m, loff_t *_pos)
{
struct afs_addr_list *alist;
struct afs_cell *cell = m->private;
loff_t pos = *_pos;
_enter("cell=%p pos=%Ld", cell, *_pos);
rcu_read_lock();
/* lock the list against modification */
down_read(&cell->vl_sem);
alist = rcu_dereference(cell->vl_addrs);
/* allow for the header line */
if (!pos)
return (void *) 1;
pos--;
if (pos >= cell->vl_naddrs)
if (!alist || pos >= alist->nr_addrs)
return NULL;
return &cell->vl_addrs[pos];
return alist->addrs + pos;
}
/*
@ -526,17 +526,18 @@ static void *afs_proc_cell_vlservers_start(struct seq_file *m, loff_t *_pos)
static void *afs_proc_cell_vlservers_next(struct seq_file *p, void *v,
loff_t *_pos)
{
struct afs_addr_list *alist;
struct afs_cell *cell = p->private;
loff_t pos;
_enter("cell=%p{nad=%u} pos=%Ld", cell, cell->vl_naddrs, *_pos);
alist = rcu_dereference(cell->vl_addrs);
pos = *_pos;
(*_pos)++;
if (pos >= cell->vl_naddrs)
if (!alist || pos >= alist->nr_addrs)
return NULL;
return &cell->vl_addrs[pos];
return alist->addrs + pos;
}
/*
@ -544,9 +545,7 @@ static void *afs_proc_cell_vlservers_next(struct seq_file *p, void *v,
*/
static void afs_proc_cell_vlservers_stop(struct seq_file *p, void *v)
{
struct afs_cell *cell = p->private;
up_read(&cell->vl_sem);
rcu_read_unlock();
}
/*
@ -554,100 +553,76 @@ static void afs_proc_cell_vlservers_stop(struct seq_file *p, void *v)
*/
static int afs_proc_cell_vlservers_show(struct seq_file *m, void *v)
{
struct in_addr *addr = v;
struct sockaddr_rxrpc *addr = v;
/* display header on line 1 */
if (v == (struct in_addr *) 1) {
if (v == (void *)1) {
seq_puts(m, "ADDRESS\n");
return 0;
}
/* display one cell per line on subsequent lines */
seq_printf(m, "%pI4\n", &addr->s_addr);
seq_printf(m, "%pISp\n", &addr->transport);
return 0;
}
/*
* open "/proc/fs/afs/<cell>/servers" which provides a summary of active
* open "/proc/fs/afs/servers" which provides a summary of active
* servers
*/
static int afs_proc_cell_servers_open(struct inode *inode, struct file *file)
static int afs_proc_servers_open(struct inode *inode, struct file *file)
{
struct afs_cell *cell;
struct seq_file *m;
int ret;
cell = PDE_DATA(inode);
if (!cell)
return -ENOENT;
ret = seq_open(file, &afs_proc_cell_servers_ops);
if (ret < 0)
return ret;
m = file->private_data;
m->private = cell;
return 0;
return seq_open(file, &afs_proc_servers_ops);
}
/*
* set up the iterator to start reading from the cells list and return the
* first item
* Set up the iterator to start reading from the server list and return the
* first item.
*/
static void *afs_proc_cell_servers_start(struct seq_file *m, loff_t *_pos)
__acquires(m->private->servers_lock)
static void *afs_proc_servers_start(struct seq_file *m, loff_t *_pos)
{
struct afs_cell *cell = m->private;
struct afs_net *net = afs_seq2net(m);
_enter("cell=%p pos=%Ld", cell, *_pos);
/* lock the list against modification */
read_lock(&cell->servers_lock);
return seq_list_start_head(&cell->servers, *_pos);
rcu_read_lock();
return seq_hlist_start_head_rcu(&net->fs_proc, *_pos);
}
/*
* move to next cell in cells list
*/
static void *afs_proc_cell_servers_next(struct seq_file *p, void *v,
loff_t *_pos)
static void *afs_proc_servers_next(struct seq_file *m, void *v, loff_t *_pos)
{
struct afs_cell *cell = p->private;
struct afs_net *net = afs_seq2net(m);
_enter("cell=%p pos=%Ld", cell, *_pos);
return seq_list_next(v, &cell->servers, _pos);
return seq_hlist_next_rcu(v, &net->fs_proc, _pos);
}
/*
* clean up after reading from the cells list
*/
static void afs_proc_cell_servers_stop(struct seq_file *p, void *v)
__releases(p->private->servers_lock)
static void afs_proc_servers_stop(struct seq_file *p, void *v)
{
struct afs_cell *cell = p->private;
read_unlock(&cell->servers_lock);
rcu_read_unlock();
}
/*
* display a header line followed by a load of volume lines
*/
static int afs_proc_cell_servers_show(struct seq_file *m, void *v)
static int afs_proc_servers_show(struct seq_file *m, void *v)
{
struct afs_cell *cell = m->private;
struct afs_server *server = list_entry(v, struct afs_server, link);
char ipaddr[20];
struct afs_server *server;
struct afs_addr_list *alist;
/* display header on line 1 */
if (v == &cell->servers) {
seq_puts(m, "USE ADDR STATE\n");
if (v == SEQ_START_TOKEN) {
seq_puts(m, "UUID USE ADDR\n");
return 0;
}
/* display one cell per line on subsequent lines */
sprintf(ipaddr, "%pI4", &server->addr);
seq_printf(m, "%3d %-15.15s %5d\n",
atomic_read(&server->usage), ipaddr, server->fs_state);
server = list_entry(v, struct afs_server, proc_link);
alist = rcu_dereference(server->addresses);
seq_printf(m, "%pU %3d %pISp\n",
&server->uuid,
atomic_read(&server->usage),
&alist->addrs[alist->index].transport);
return 0;
}

715
fs/afs/rotate.c Normal file
View File

@ -0,0 +1,715 @@
/* Handle fileserver selection and rotation.
*
* Copyright (C) 2017 Red Hat, Inc. All Rights Reserved.
* Written by David Howells (dhowells@redhat.com)
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public Licence
* as published by the Free Software Foundation; either version
* 2 of the Licence, or (at your option) any later version.
*/
#include <linux/kernel.h>
#include <linux/slab.h>
#include <linux/fs.h>
#include <linux/sched.h>
#include <linux/delay.h>
#include <linux/sched/signal.h>
#include "internal.h"
#include "afs_fs.h"
/*
* Initialise a filesystem server cursor for iterating over FS servers.
*/
void afs_init_fs_cursor(struct afs_fs_cursor *fc, struct afs_vnode *vnode)
{
memset(fc, 0, sizeof(*fc));
}
/*
* Begin an operation on the fileserver.
*
* Fileserver operations are serialised on the server by vnode, so we serialise
* them here also using the io_lock.
*/
bool afs_begin_vnode_operation(struct afs_fs_cursor *fc, struct afs_vnode *vnode,
struct key *key)
{
afs_init_fs_cursor(fc, vnode);
fc->vnode = vnode;
fc->key = key;
fc->ac.error = SHRT_MAX;
if (mutex_lock_interruptible(&vnode->io_lock) < 0) {
fc->ac.error = -EINTR;
fc->flags |= AFS_FS_CURSOR_STOP;
return false;
}
if (test_bit(AFS_VNODE_READLOCKED, &vnode->flags) ||
test_bit(AFS_VNODE_WRITELOCKED, &vnode->flags))
fc->flags |= AFS_FS_CURSOR_CUR_ONLY;
return true;
}
/*
* Begin iteration through a server list, starting with the vnode's last used
* server if possible, or the last recorded good server if not.
*/
static bool afs_start_fs_iteration(struct afs_fs_cursor *fc,
struct afs_vnode *vnode)
{
struct afs_cb_interest *cbi;
int i;
read_lock(&vnode->volume->servers_lock);
fc->server_list = afs_get_serverlist(vnode->volume->servers);
read_unlock(&vnode->volume->servers_lock);
cbi = vnode->cb_interest;
if (cbi) {
/* See if the vnode's preferred record is still available */
for (i = 0; i < fc->server_list->nr_servers; i++) {
if (fc->server_list->servers[i].cb_interest == cbi) {
fc->start = i;
goto found_interest;
}
}
/* If we have a lock outstanding on a server that's no longer
* serving this vnode, then we can't switch to another server
* and have to return an error.
*/
if (fc->flags & AFS_FS_CURSOR_CUR_ONLY) {
fc->ac.error = -ESTALE;
return false;
}
/* Note that the callback promise is effectively broken */
write_seqlock(&vnode->cb_lock);
ASSERTCMP(cbi, ==, vnode->cb_interest);
vnode->cb_interest = NULL;
if (test_and_clear_bit(AFS_VNODE_CB_PROMISED, &vnode->flags))
vnode->cb_break++;
write_sequnlock(&vnode->cb_lock);
afs_put_cb_interest(afs_v2net(vnode), cbi);
cbi = NULL;
} else {
fc->start = READ_ONCE(fc->server_list->index);
}
found_interest:
fc->index = fc->start;
return true;
}
/*
* Post volume busy note.
*/
static void afs_busy(struct afs_volume *volume, u32 abort_code)
{
const char *m;
switch (abort_code) {
case VOFFLINE: m = "offline"; break;
case VRESTARTING: m = "restarting"; break;
case VSALVAGING: m = "being salvaged"; break;
default: m = "busy"; break;
}
pr_notice("kAFS: Volume %u '%s' is %s\n", volume->vid, volume->name, m);
}
/*
* Sleep and retry the operation to the same fileserver.
*/
static bool afs_sleep_and_retry(struct afs_fs_cursor *fc)
{
msleep_interruptible(1000);
if (signal_pending(current)) {
fc->ac.error = -ERESTARTSYS;
return false;
}
return true;
}
/*
* Select the fileserver to use. May be called multiple times to rotate
* through the fileservers.
*/
bool afs_select_fileserver(struct afs_fs_cursor *fc)
{
struct afs_addr_list *alist;
struct afs_server *server;
struct afs_vnode *vnode = fc->vnode;
_enter("%u/%u,%u/%u,%d,%d",
fc->index, fc->start,
fc->ac.index, fc->ac.start,
fc->ac.error, fc->ac.abort_code);
if (fc->flags & AFS_FS_CURSOR_STOP) {
_leave(" = f [stopped]");
return false;
}
/* Evaluate the result of the previous operation, if there was one. */
switch (fc->ac.error) {
case SHRT_MAX:
goto start;
case 0:
default:
/* Success or local failure. Stop. */
fc->flags |= AFS_FS_CURSOR_STOP;
_leave(" = f [okay/local %d]", fc->ac.error);
return false;
case -ECONNABORTED:
/* The far side rejected the operation on some grounds. This
* might involve the server being busy or the volume having been moved.
*/
switch (fc->ac.abort_code) {
case VNOVOL:
/* This fileserver doesn't know about the volume.
* - May indicate that the VL is wrong - retry once and compare
* the results.
* - May indicate that the fileserver couldn't attach to the vol.
*/
if (fc->flags & AFS_FS_CURSOR_VNOVOL) {
fc->ac.error = -EREMOTEIO;
goto failed;
}
write_lock(&vnode->volume->servers_lock);
fc->server_list->vnovol_mask |= 1 << fc->index;
write_unlock(&vnode->volume->servers_lock);
set_bit(AFS_VOLUME_NEEDS_UPDATE, &vnode->volume->flags);
fc->ac.error = afs_check_volume_status(vnode->volume, fc->key);
if (fc->ac.error < 0)
goto failed;
if (test_bit(AFS_VOLUME_DELETED, &vnode->volume->flags)) {
fc->ac.error = -ENOMEDIUM;
goto failed;
}
/* If the server list didn't change, then assume that
* it's the fileserver having trouble.
*/
if (vnode->volume->servers == fc->server_list) {
fc->ac.error = -EREMOTEIO;
goto failed;
}
/* Try again */
fc->flags |= AFS_FS_CURSOR_VNOVOL;
_leave(" = t [vnovol]");
return true;
case VSALVAGE: /* TODO: Should this return an error or iterate? */
case VVOLEXISTS:
case VNOSERVICE:
case VONLINE:
case VDISKFULL:
case VOVERQUOTA:
fc->ac.error = afs_abort_to_error(fc->ac.abort_code);
goto next_server;
case VOFFLINE:
if (!test_and_set_bit(AFS_VOLUME_OFFLINE, &vnode->volume->flags)) {
afs_busy(vnode->volume, fc->ac.abort_code);
clear_bit(AFS_VOLUME_BUSY, &vnode->volume->flags);
}
if (fc->flags & AFS_FS_CURSOR_NO_VSLEEP) {
fc->ac.error = -EADV;
goto failed;
}
if (fc->flags & AFS_FS_CURSOR_CUR_ONLY) {
fc->ac.error = -ESTALE;
goto failed;
}
goto busy;
case VSALVAGING:
case VRESTARTING:
case VBUSY:
/* Retry after going round all the servers unless we
* have a file lock we need to maintain.
*/
if (fc->flags & AFS_FS_CURSOR_NO_VSLEEP) {
fc->ac.error = -EBUSY;
goto failed;
}
if (!test_and_set_bit(AFS_VOLUME_BUSY, &vnode->volume->flags)) {
afs_busy(vnode->volume, fc->ac.abort_code);
clear_bit(AFS_VOLUME_OFFLINE, &vnode->volume->flags);
}
busy:
if (fc->flags & AFS_FS_CURSOR_CUR_ONLY) {
if (!afs_sleep_and_retry(fc))
goto failed;
/* Retry with same server & address */
_leave(" = t [vbusy]");
return true;
}
fc->flags |= AFS_FS_CURSOR_VBUSY;
goto next_server;
case VMOVED:
/* The volume migrated to another server. We consider
* consider all locks and callbacks broken and request
* an update from the VLDB.
*
* We also limit the number of VMOVED hops we will
* honour, just in case someone sets up a loop.
*/
if (fc->flags & AFS_FS_CURSOR_VMOVED) {
fc->ac.error = -EREMOTEIO;
goto failed;
}
fc->flags |= AFS_FS_CURSOR_VMOVED;
set_bit(AFS_VOLUME_WAIT, &vnode->volume->flags);
set_bit(AFS_VOLUME_NEEDS_UPDATE, &vnode->volume->flags);
fc->ac.error = afs_check_volume_status(vnode->volume, fc->key);
if (fc->ac.error < 0)
goto failed;
/* If the server list didn't change, then the VLDB is
* out of sync with the fileservers. This is hopefully
* a temporary condition, however, so we don't want to
* permanently block access to the file.
*
* TODO: Try other fileservers if we can.
*
* TODO: Retry a few times with sleeps.
*/
if (vnode->volume->servers == fc->server_list) {
fc->ac.error = -ENOMEDIUM;
goto failed;
}
goto restart_from_beginning;
default:
clear_bit(AFS_VOLUME_OFFLINE, &vnode->volume->flags);
clear_bit(AFS_VOLUME_BUSY, &vnode->volume->flags);
fc->ac.error = afs_abort_to_error(fc->ac.abort_code);
goto failed;
}
case -ENETUNREACH:
case -EHOSTUNREACH:
case -ECONNREFUSED:
case -ETIMEDOUT:
case -ETIME:
_debug("no conn");
goto iterate_address;
}
restart_from_beginning:
_debug("restart");
afs_end_cursor(&fc->ac);
afs_put_cb_interest(afs_v2net(vnode), fc->cbi);
fc->cbi = NULL;
afs_put_serverlist(afs_v2net(vnode), fc->server_list);
fc->server_list = NULL;
start:
_debug("start");
/* See if we need to do an update of the volume record. Note that the
* volume may have moved or even have been deleted.
*/
fc->ac.error = afs_check_volume_status(vnode->volume, fc->key);
if (fc->ac.error < 0)
goto failed;
if (!afs_start_fs_iteration(fc, vnode))
goto failed;
goto use_server;
next_server:
_debug("next");
afs_put_cb_interest(afs_v2net(vnode), fc->cbi);
fc->cbi = NULL;
fc->index++;
if (fc->index >= fc->server_list->nr_servers)
fc->index = 0;
if (fc->index != fc->start)
goto use_server;
/* That's all the servers poked to no good effect. Try again if some
* of them were busy.
*/
if (fc->flags & AFS_FS_CURSOR_VBUSY)
goto restart_from_beginning;
fc->ac.error = -EDESTADDRREQ;
goto failed;
use_server:
_debug("use");
/* We're starting on a different fileserver from the list. We need to
* check it, create a callback intercept, find its address list and
* probe its capabilities before we use it.
*/
ASSERTCMP(fc->ac.alist, ==, NULL);
server = fc->server_list->servers[fc->index].server;
if (!afs_check_server_record(fc, server))
goto failed;
_debug("USING SERVER: %pU", &server->uuid);
/* Make sure we've got a callback interest record for this server. We
* have to link it in before we send the request as we can be sent a
* break request before we've finished decoding the reply and
* installing the vnode.
*/
fc->ac.error = afs_register_server_cb_interest(
vnode, &fc->server_list->servers[fc->index]);
if (fc->ac.error < 0)
goto failed;
fc->cbi = afs_get_cb_interest(vnode->cb_interest);
read_lock(&server->fs_lock);
alist = rcu_dereference_protected(server->addresses,
lockdep_is_held(&server->fs_lock));
afs_get_addrlist(alist);
read_unlock(&server->fs_lock);
/* Probe the current fileserver if we haven't done so yet. */
if (!test_bit(AFS_SERVER_FL_PROBED, &server->flags)) {
fc->ac.alist = afs_get_addrlist(alist);
if (!afs_probe_fileserver(fc))
goto failed;
}
if (!fc->ac.alist)
fc->ac.alist = alist;
else
afs_put_addrlist(alist);
fc->ac.addr = NULL;
fc->ac.start = READ_ONCE(alist->index);
fc->ac.index = fc->ac.start;
fc->ac.error = 0;
fc->ac.begun = false;
goto iterate_address;
iterate_address:
ASSERT(fc->ac.alist);
_debug("iterate %d/%d", fc->ac.index, fc->ac.alist->nr_addrs);
/* Iterate over the current server's address list to try and find an
* address on which it will respond to us.
*/
if (afs_iterate_addresses(&fc->ac)) {
_leave(" = t");
return true;
}
afs_end_cursor(&fc->ac);
goto next_server;
failed:
fc->flags |= AFS_FS_CURSOR_STOP;
_leave(" = f [failed %d]", fc->ac.error);
return false;
}
/*
* Select the same fileserver we used for a vnode before and only that
* fileserver. We use this when we have a lock on that file, which is backed
* only by the fileserver we obtained it from.
*/
bool afs_select_current_fileserver(struct afs_fs_cursor *fc)
{
struct afs_vnode *vnode = fc->vnode;
struct afs_cb_interest *cbi = vnode->cb_interest;
struct afs_addr_list *alist;
_enter("");
if (!cbi) {
fc->ac.error = -ESTALE;
fc->flags |= AFS_FS_CURSOR_STOP;
return false;
}
read_lock(&cbi->server->fs_lock);
alist = afs_get_addrlist(cbi->server->addresses);
read_unlock(&cbi->server->fs_lock);
if (!alist) {
fc->ac.error = -ESTALE;
fc->flags |= AFS_FS_CURSOR_STOP;
return false;
}
fc->ac.alist = alist;
fc->ac.error = 0;
return true;
}
/*
* Tidy up a filesystem cursor and unlock the vnode.
*/
int afs_end_vnode_operation(struct afs_fs_cursor *fc)
{
struct afs_net *net = afs_v2net(fc->vnode);
int ret;
mutex_unlock(&fc->vnode->io_lock);
afs_end_cursor(&fc->ac);
afs_put_cb_interest(net, fc->cbi);
afs_put_serverlist(net, fc->server_list);
ret = fc->ac.error;
if (ret == -ECONNABORTED)
afs_abort_to_error(fc->ac.abort_code);
return fc->ac.error;
}
#if 0
/*
* Set a filesystem server cursor for using a specific FS server.
*/
int afs_set_fs_cursor(struct afs_fs_cursor *fc, struct afs_vnode *vnode)
{
afs_init_fs_cursor(fc, vnode);
read_seqlock_excl(&vnode->cb_lock);
if (vnode->cb_interest) {
if (vnode->cb_interest->server->fs_state == 0)
fc->server = afs_get_server(vnode->cb_interest->server);
else
fc->ac.error = vnode->cb_interest->server->fs_state;
} else {
fc->ac.error = -ESTALE;
}
read_sequnlock_excl(&vnode->cb_lock);
return fc->ac.error;
}
/*
* pick a server to use to try accessing this volume
* - returns with an elevated usage count on the server chosen
*/
bool afs_volume_pick_fileserver(struct afs_fs_cursor *fc, struct afs_vnode *vnode)
{
struct afs_volume *volume = vnode->volume;
struct afs_server *server;
int ret, state, loop;
_enter("%s", volume->vlocation->vldb.name);
/* stick with the server we're already using if we can */
if (vnode->cb_interest && vnode->cb_interest->server->fs_state == 0) {
fc->server = afs_get_server(vnode->cb_interest->server);
goto set_server;
}
down_read(&volume->server_sem);
/* handle the no-server case */
if (volume->nservers == 0) {
fc->ac.error = volume->rjservers ? -ENOMEDIUM : -ESTALE;
up_read(&volume->server_sem);
_leave(" = f [no servers %d]", fc->ac.error);
return false;
}
/* basically, just search the list for the first live server and use
* that */
ret = 0;
for (loop = 0; loop < volume->nservers; loop++) {
server = volume->servers[loop];
state = server->fs_state;
_debug("consider %d [%d]", loop, state);
switch (state) {
case 0:
goto picked_server;
case -ENETUNREACH:
if (ret == 0)
ret = state;
break;
case -EHOSTUNREACH:
if (ret == 0 ||
ret == -ENETUNREACH)
ret = state;
break;
case -ECONNREFUSED:
if (ret == 0 ||
ret == -ENETUNREACH ||
ret == -EHOSTUNREACH)
ret = state;
break;
default:
case -EREMOTEIO:
if (ret == 0 ||
ret == -ENETUNREACH ||
ret == -EHOSTUNREACH ||
ret == -ECONNREFUSED)
ret = state;
break;
}
}
error:
fc->ac.error = ret;
/* no available servers
* - TODO: handle the no active servers case better
*/
up_read(&volume->server_sem);
_leave(" = f [%d]", fc->ac.error);
return false;
picked_server:
/* Found an apparently healthy server. We need to register an interest
* in receiving callbacks before we talk to it.
*/
ret = afs_register_server_cb_interest(vnode,
&volume->cb_interests[loop], server);
if (ret < 0)
goto error;
fc->server = afs_get_server(server);
up_read(&volume->server_sem);
set_server:
fc->ac.alist = afs_get_addrlist(fc->server->addrs);
fc->ac.addr = &fc->ac.alist->addrs[0];
_debug("USING SERVER: %pIS\n", &fc->ac.addr->transport);
_leave(" = t (picked %pIS)", &fc->ac.addr->transport);
return true;
}
/*
* release a server after use
* - releases the ref on the server struct that was acquired by picking
* - records result of using a particular server to access a volume
* - return true to try again, false if okay or to issue error
* - the caller must release the server struct if result was false
*/
bool afs_iterate_fs_cursor(struct afs_fs_cursor *fc,
struct afs_vnode *vnode)
{
struct afs_volume *volume = vnode->volume;
struct afs_server *server = fc->server;
unsigned loop;
_enter("%s,%pIS,%d",
volume->vlocation->vldb.name, &fc->ac.addr->transport,
fc->ac.error);
switch (fc->ac.error) {
/* success */
case 0:
server->fs_state = 0;
_leave(" = f");
return false;
/* the fileserver denied all knowledge of the volume */
case -ENOMEDIUM:
down_write(&volume->server_sem);
/* firstly, find where the server is in the active list (if it
* is) */
for (loop = 0; loop < volume->nservers; loop++)
if (volume->servers[loop] == server)
goto present;
/* no longer there - may have been discarded by another op */
goto try_next_server_upw;
present:
volume->nservers--;
memmove(&volume->servers[loop],
&volume->servers[loop + 1],
sizeof(volume->servers[loop]) *
(volume->nservers - loop));
volume->servers[volume->nservers] = NULL;
afs_put_server(afs_v2net(vnode), server);
volume->rjservers++;
if (volume->nservers > 0)
/* another server might acknowledge its existence */
goto try_next_server_upw;
/* handle the case where all the fileservers have rejected the
* volume
* - TODO: try asking the fileservers for volume information
* - TODO: contact the VL server again to see if the volume is
* no longer registered
*/
up_write(&volume->server_sem);
afs_put_server(afs_v2net(vnode), server);
fc->server = NULL;
_leave(" = f [completely rejected]");
return false;
/* problem reaching the server */
case -ENETUNREACH:
case -EHOSTUNREACH:
case -ECONNREFUSED:
case -ETIME:
case -ETIMEDOUT:
case -EREMOTEIO:
/* mark the server as dead
* TODO: vary dead timeout depending on error
*/
spin_lock(&server->fs_lock);
if (!server->fs_state) {
server->fs_state = fc->ac.error;
printk("kAFS: SERVER DEAD state=%d\n", fc->ac.error);
}
spin_unlock(&server->fs_lock);
goto try_next_server;
/* miscellaneous error */
default:
case -ENOMEM:
case -ENONET:
/* tell the caller to accept the result */
afs_put_server(afs_v2net(vnode), server);
fc->server = NULL;
_leave(" = f [local failure]");
return false;
}
/* tell the caller to loop around and try the next server */
try_next_server_upw:
up_write(&volume->server_sem);
try_next_server:
afs_put_server(afs_v2net(vnode), server);
_leave(" = t [try next server]");
return true;
}
/*
* Clean up a fileserver cursor.
*/
int afs_end_fs_cursor(struct afs_fs_cursor *fc, struct afs_net *net)
{
afs_end_cursor(&fc->ac);
afs_put_server(net, fc->server);
return fc->ac.error;
}
#endif

335
fs/afs/rxrpc.c
View File

@ -17,13 +17,10 @@
#include "internal.h"
#include "afs_cm.h"
struct socket *afs_socket; /* my RxRPC socket */
static struct workqueue_struct *afs_async_calls;
static struct afs_call *afs_spare_incoming_call;
atomic_t afs_outstanding_calls;
struct workqueue_struct *afs_async_calls;
static void afs_wake_up_call_waiter(struct sock *, struct rxrpc_call *, unsigned long);
static int afs_wait_for_call_to_complete(struct afs_call *);
static long afs_wait_for_call_to_complete(struct afs_call *, struct afs_addr_cursor *);
static void afs_wake_up_async_call(struct sock *, struct rxrpc_call *, unsigned long);
static void afs_process_async_call(struct work_struct *);
static void afs_rx_new_call(struct sock *, struct rxrpc_call *, unsigned long);
@ -34,24 +31,13 @@ static int afs_deliver_cm_op_id(struct afs_call *);
static const struct afs_call_type afs_RXCMxxxx = {
.name = "CB.xxxx",
.deliver = afs_deliver_cm_op_id,
.abort_to_error = afs_abort_to_error,
};
static void afs_charge_preallocation(struct work_struct *);
static DECLARE_WORK(afs_charge_preallocation_work, afs_charge_preallocation);
static int afs_wait_atomic_t(atomic_t *p)
{
schedule();
return 0;
}
/*
* open an RxRPC socket and bind it to be a server for callback notifications
* - the socket is left in blocking mode and non-blocking ops use MSG_DONTWAIT
*/
int afs_open_socket(void)
int afs_open_socket(struct afs_net *net)
{
struct sockaddr_rxrpc srx;
struct socket *socket;
@ -59,28 +45,26 @@ int afs_open_socket(void)
_enter("");
ret = -ENOMEM;
afs_async_calls = alloc_workqueue("kafsd", WQ_MEM_RECLAIM, 0);
if (!afs_async_calls)
goto error_0;
ret = sock_create_kern(&init_net, AF_RXRPC, SOCK_DGRAM, PF_INET, &socket);
ret = sock_create_kern(&init_net, AF_RXRPC, SOCK_DGRAM, PF_INET6, &socket);
if (ret < 0)
goto error_1;
socket->sk->sk_allocation = GFP_NOFS;
/* bind the callback manager's address to make this a server socket */
memset(&srx, 0, sizeof(srx));
srx.srx_family = AF_RXRPC;
srx.srx_service = CM_SERVICE;
srx.transport_type = SOCK_DGRAM;
srx.transport_len = sizeof(srx.transport.sin);
srx.transport.sin.sin_family = AF_INET;
srx.transport.sin.sin_port = htons(AFS_CM_PORT);
memset(&srx.transport.sin.sin_addr, 0,
sizeof(srx.transport.sin.sin_addr));
srx.transport_len = sizeof(srx.transport.sin6);
srx.transport.sin6.sin6_family = AF_INET6;
srx.transport.sin6.sin6_port = htons(AFS_CM_PORT);
ret = kernel_bind(socket, (struct sockaddr *) &srx, sizeof(srx));
if (ret == -EADDRINUSE) {
srx.transport.sin6.sin6_port = 0;
ret = kernel_bind(socket, (struct sockaddr *) &srx, sizeof(srx));
}
if (ret < 0)
goto error_2;
@ -91,16 +75,14 @@ int afs_open_socket(void)
if (ret < 0)
goto error_2;
afs_socket = socket;
afs_charge_preallocation(NULL);
net->socket = socket;
afs_charge_preallocation(&net->charge_preallocation_work);
_leave(" = 0");
return 0;
error_2:
sock_release(socket);
error_1:
destroy_workqueue(afs_async_calls);
error_0:
_leave(" = %d", ret);
return ret;
}
@ -108,36 +90,36 @@ error_0:
/*
* close the RxRPC socket AFS was using
*/
void afs_close_socket(void)
void afs_close_socket(struct afs_net *net)
{
_enter("");
kernel_listen(afs_socket, 0);
kernel_listen(net->socket, 0);
flush_workqueue(afs_async_calls);
if (afs_spare_incoming_call) {
afs_put_call(afs_spare_incoming_call);
afs_spare_incoming_call = NULL;
if (net->spare_incoming_call) {
afs_put_call(net->spare_incoming_call);
net->spare_incoming_call = NULL;
}
_debug("outstanding %u", atomic_read(&afs_outstanding_calls));
wait_on_atomic_t(&afs_outstanding_calls, afs_wait_atomic_t,
_debug("outstanding %u", atomic_read(&net->nr_outstanding_calls));
wait_on_atomic_t(&net->nr_outstanding_calls, atomic_t_wait,
TASK_UNINTERRUPTIBLE);
_debug("no outstanding calls");
kernel_sock_shutdown(afs_socket, SHUT_RDWR);
kernel_sock_shutdown(net->socket, SHUT_RDWR);
flush_workqueue(afs_async_calls);
sock_release(afs_socket);
sock_release(net->socket);
_debug("dework");
destroy_workqueue(afs_async_calls);
_leave("");
}
/*
* Allocate a call.
*/
static struct afs_call *afs_alloc_call(const struct afs_call_type *type,
static struct afs_call *afs_alloc_call(struct afs_net *net,
const struct afs_call_type *type,
gfp_t gfp)
{
struct afs_call *call;
@ -148,11 +130,13 @@ static struct afs_call *afs_alloc_call(const struct afs_call_type *type,
return NULL;
call->type = type;
call->net = net;
atomic_set(&call->usage, 1);
INIT_WORK(&call->async_work, afs_process_async_call);
init_waitqueue_head(&call->waitq);
spin_lock_init(&call->state_lock);
o = atomic_inc_return(&afs_outstanding_calls);
o = atomic_inc_return(&net->nr_outstanding_calls);
trace_afs_call(call, afs_call_trace_alloc, 1, o,
__builtin_return_address(0));
return call;
@ -163,8 +147,9 @@ static struct afs_call *afs_alloc_call(const struct afs_call_type *type,
*/
void afs_put_call(struct afs_call *call)
{
struct afs_net *net = call->net;
int n = atomic_dec_return(&call->usage);
int o = atomic_read(&afs_outstanding_calls);
int o = atomic_read(&net->nr_outstanding_calls);
trace_afs_call(call, afs_call_trace_put, n + 1, o,
__builtin_return_address(0));
@ -175,20 +160,22 @@ void afs_put_call(struct afs_call *call)
ASSERT(call->type->name != NULL);
if (call->rxcall) {
rxrpc_kernel_end_call(afs_socket, call->rxcall);
rxrpc_kernel_end_call(net->socket, call->rxcall);
call->rxcall = NULL;
}
if (call->type->destructor)
call->type->destructor(call);
afs_put_server(call->net, call->cm_server);
afs_put_cb_interest(call->net, call->cbi);
kfree(call->request);
kfree(call);
o = atomic_dec_return(&afs_outstanding_calls);
o = atomic_dec_return(&net->nr_outstanding_calls);
trace_afs_call(call, afs_call_trace_free, 0, o,
__builtin_return_address(0));
if (o == 0)
wake_up_atomic_t(&afs_outstanding_calls);
wake_up_atomic_t(&net->nr_outstanding_calls);
}
}
@ -200,7 +187,7 @@ int afs_queue_call_work(struct afs_call *call)
int u = atomic_inc_return(&call->usage);
trace_afs_call(call, afs_call_trace_work, u,
atomic_read(&afs_outstanding_calls),
atomic_read(&call->net->nr_outstanding_calls),
__builtin_return_address(0));
INIT_WORK(&call->work, call->type->work);
@ -213,12 +200,13 @@ int afs_queue_call_work(struct afs_call *call)
/*
* allocate a call with flat request and reply buffers
*/
struct afs_call *afs_alloc_flat_call(const struct afs_call_type *type,
struct afs_call *afs_alloc_flat_call(struct afs_net *net,
const struct afs_call_type *type,
size_t request_size, size_t reply_max)
{
struct afs_call *call;
call = afs_alloc_call(type, GFP_NOFS);
call = afs_alloc_call(net, type, GFP_NOFS);
if (!call)
goto nomem_call;
@ -236,6 +224,7 @@ struct afs_call *afs_alloc_flat_call(const struct afs_call_type *type,
goto nomem_free;
}
call->operation_ID = type->op;
init_waitqueue_head(&call->waitq);
return call;
@ -300,8 +289,7 @@ static void afs_notify_end_request_tx(struct sock *sock,
{
struct afs_call *call = (struct afs_call *)call_user_ID;
if (call->state == AFS_CALL_REQUESTING)
call->state = AFS_CALL_AWAIT_REPLY;
afs_set_call_state(call, AFS_CALL_CL_REQUESTING, AFS_CALL_CL_AWAIT_REPLY);
}
/*
@ -319,11 +307,13 @@ static int afs_send_pages(struct afs_call *call, struct msghdr *msg)
do {
afs_load_bvec(call, msg, bv, first, last, offset);
trace_afs_send_pages(call, msg, first, last, offset);
offset = 0;
bytes = msg->msg_iter.count;
nr = msg->msg_iter.nr_segs;
ret = rxrpc_kernel_send_data(afs_socket, call->rxcall, msg,
ret = rxrpc_kernel_send_data(call->net->socket, call->rxcall, msg,
bytes, afs_notify_end_request_tx);
for (loop = 0; loop < nr; loop++)
put_page(bv[loop].bv_page);
@ -333,63 +323,62 @@ static int afs_send_pages(struct afs_call *call, struct msghdr *msg)
first += nr;
} while (first <= last);
trace_afs_sent_pages(call, call->first, last, first, ret);
return ret;
}
/*
* initiate a call
*/
int afs_make_call(struct in_addr *addr, struct afs_call *call, gfp_t gfp,
bool async)
long afs_make_call(struct afs_addr_cursor *ac, struct afs_call *call,
gfp_t gfp, bool async)
{
struct sockaddr_rxrpc srx;
struct sockaddr_rxrpc *srx = ac->addr;
struct rxrpc_call *rxcall;
struct msghdr msg;
struct kvec iov[1];
size_t offset;
s64 tx_total_len;
u32 abort_code;
int ret;
_enter("%x,{%d},", addr->s_addr, ntohs(call->port));
_enter(",{%pISp},", &srx->transport);
ASSERT(call->type != NULL);
ASSERT(call->type->name != NULL);
_debug("____MAKE %p{%s,%x} [%d]____",
call, call->type->name, key_serial(call->key),
atomic_read(&afs_outstanding_calls));
atomic_read(&call->net->nr_outstanding_calls));
call->async = async;
memset(&srx, 0, sizeof(srx));
srx.srx_family = AF_RXRPC;
srx.srx_service = call->service_id;
srx.transport_type = SOCK_DGRAM;
srx.transport_len = sizeof(srx.transport.sin);
srx.transport.sin.sin_family = AF_INET;
srx.transport.sin.sin_port = call->port;
memcpy(&srx.transport.sin.sin_addr, addr, 4);
/* Work out the length we're going to transmit. This is awkward for
* calls such as FS.StoreData where there's an extra injection of data
* after the initial fixed part.
*/
tx_total_len = call->request_size;
if (call->send_pages) {
tx_total_len += call->last_to - call->first_offset;
tx_total_len += (call->last - call->first) * PAGE_SIZE;
if (call->last == call->first) {
tx_total_len += call->last_to - call->first_offset;
} else {
/* It looks mathematically like you should be able to
* combine the following lines with the ones above, but
* unsigned arithmetic is fun when it wraps...
*/
tx_total_len += PAGE_SIZE - call->first_offset;
tx_total_len += call->last_to;
tx_total_len += (call->last - call->first - 1) * PAGE_SIZE;
}
}
/* create a call */
rxcall = rxrpc_kernel_begin_call(afs_socket, &srx, call->key,
rxcall = rxrpc_kernel_begin_call(call->net->socket, srx, call->key,
(unsigned long)call,
tx_total_len, gfp,
(async ?
afs_wake_up_async_call :
afs_wake_up_call_waiter),
call->upgrade);
call->key = NULL;
if (IS_ERR(rxcall)) {
ret = PTR_ERR(rxcall);
goto error_kill_call;
@ -409,14 +398,7 @@ int afs_make_call(struct in_addr *addr, struct afs_call *call, gfp_t gfp,
msg.msg_controllen = 0;
msg.msg_flags = MSG_WAITALL | (call->send_pages ? MSG_MORE : 0);
/* We have to change the state *before* sending the last packet as
* rxrpc might give us the reply before it returns from sending the
* request. Further, if the send fails, we may already have been given
* a notification and may have collected it.
*/
if (!call->send_pages)
call->state = AFS_CALL_AWAIT_REPLY;
ret = rxrpc_kernel_send_data(afs_socket, rxcall,
ret = rxrpc_kernel_send_data(call->net->socket, rxcall,
&msg, call->request_size,
afs_notify_end_request_tx);
if (ret < 0)
@ -433,22 +415,26 @@ int afs_make_call(struct in_addr *addr, struct afs_call *call, gfp_t gfp,
if (call->async)
return -EINPROGRESS;
return afs_wait_for_call_to_complete(call);
return afs_wait_for_call_to_complete(call, ac);
error_do_abort:
call->state = AFS_CALL_COMPLETE;
if (ret != -ECONNABORTED) {
rxrpc_kernel_abort_call(afs_socket, rxcall, RX_USER_ABORT,
ret, "KSD");
rxrpc_kernel_abort_call(call->net->socket, rxcall,
RX_USER_ABORT, ret, "KSD");
} else {
abort_code = 0;
offset = 0;
rxrpc_kernel_recv_data(afs_socket, rxcall, NULL, 0, &offset,
false, &abort_code, &call->service_id);
ret = call->type->abort_to_error(abort_code);
rxrpc_kernel_recv_data(call->net->socket, rxcall, NULL,
0, &offset, false, &call->abort_code,
&call->service_id);
ac->abort_code = call->abort_code;
ac->responded = true;
}
call->error = ret;
trace_afs_call_done(call);
error_kill_call:
afs_put_call(call);
ac->error = ret;
_leave(" = %d", ret);
return ret;
}
@ -458,88 +444,98 @@ error_kill_call:
*/
static void afs_deliver_to_call(struct afs_call *call)
{
u32 abort_code;
enum afs_call_state state;
u32 abort_code, remote_abort = 0;
int ret;
_enter("%s", call->type->name);
while (call->state == AFS_CALL_AWAIT_REPLY ||
call->state == AFS_CALL_AWAIT_OP_ID ||
call->state == AFS_CALL_AWAIT_REQUEST ||
call->state == AFS_CALL_AWAIT_ACK
while (state = READ_ONCE(call->state),
state == AFS_CALL_CL_AWAIT_REPLY ||
state == AFS_CALL_SV_AWAIT_OP_ID ||
state == AFS_CALL_SV_AWAIT_REQUEST ||
state == AFS_CALL_SV_AWAIT_ACK
) {
if (call->state == AFS_CALL_AWAIT_ACK) {
if (state == AFS_CALL_SV_AWAIT_ACK) {
size_t offset = 0;
ret = rxrpc_kernel_recv_data(afs_socket, call->rxcall,
ret = rxrpc_kernel_recv_data(call->net->socket,
call->rxcall,
NULL, 0, &offset, false,
&call->abort_code,
&remote_abort,
&call->service_id);
trace_afs_recv_data(call, 0, offset, false, ret);
if (ret == -EINPROGRESS || ret == -EAGAIN)
return;
if (ret == 1 || ret < 0) {
call->state = AFS_CALL_COMPLETE;
goto done;
if (ret < 0 || ret == 1) {
if (ret == 1)
ret = 0;
goto call_complete;
}
return;
}
ret = call->type->deliver(call);
state = READ_ONCE(call->state);
switch (ret) {
case 0:
if (call->state == AFS_CALL_AWAIT_REPLY)
call->state = AFS_CALL_COMPLETE;
if (state == AFS_CALL_CL_PROC_REPLY)
goto call_complete;
ASSERTCMP(state, >, AFS_CALL_CL_PROC_REPLY);
goto done;
case -EINPROGRESS:
case -EAGAIN:
goto out;
case -EIO:
case -ECONNABORTED:
goto call_complete;
ASSERTCMP(state, ==, AFS_CALL_COMPLETE);
goto done;
case -ENOTCONN:
abort_code = RX_CALL_DEAD;
rxrpc_kernel_abort_call(afs_socket, call->rxcall,
rxrpc_kernel_abort_call(call->net->socket, call->rxcall,
abort_code, ret, "KNC");
goto save_error;
goto local_abort;
case -ENOTSUPP:
abort_code = RXGEN_OPCODE;
rxrpc_kernel_abort_call(afs_socket, call->rxcall,
rxrpc_kernel_abort_call(call->net->socket, call->rxcall,
abort_code, ret, "KIV");
goto save_error;
goto local_abort;
case -ENODATA:
case -EBADMSG:
case -EMSGSIZE:
default:
abort_code = RXGEN_CC_UNMARSHAL;
if (call->state != AFS_CALL_AWAIT_REPLY)
if (state != AFS_CALL_CL_AWAIT_REPLY)
abort_code = RXGEN_SS_UNMARSHAL;
rxrpc_kernel_abort_call(afs_socket, call->rxcall,
rxrpc_kernel_abort_call(call->net->socket, call->rxcall,
abort_code, -EBADMSG, "KUM");
goto save_error;
goto local_abort;
}
}
done:
if (call->state == AFS_CALL_COMPLETE && call->incoming)
if (state == AFS_CALL_COMPLETE && call->incoming)
afs_put_call(call);
out:
_leave("");
return;
save_error:
call->error = ret;
local_abort:
abort_code = 0;
call_complete:
call->state = AFS_CALL_COMPLETE;
afs_set_call_complete(call, ret, remote_abort);
state = AFS_CALL_COMPLETE;
goto done;
}
/*
* wait synchronously for a call to complete
*/
static int afs_wait_for_call_to_complete(struct afs_call *call)
static long afs_wait_for_call_to_complete(struct afs_call *call,
struct afs_addr_cursor *ac)
{
signed long rtt2, timeout;
int ret;
long ret;
u64 rtt;
u32 life, last_life;
@ -547,30 +543,31 @@ static int afs_wait_for_call_to_complete(struct afs_call *call)
_enter("");
rtt = rxrpc_kernel_get_rtt(afs_socket, call->rxcall);
rtt = rxrpc_kernel_get_rtt(call->net->socket, call->rxcall);
rtt2 = nsecs_to_jiffies64(rtt) * 2;
if (rtt2 < 2)
rtt2 = 2;
timeout = rtt2;
last_life = rxrpc_kernel_check_life(afs_socket, call->rxcall);
last_life = rxrpc_kernel_check_life(call->net->socket, call->rxcall);
add_wait_queue(&call->waitq, &myself);
for (;;) {
set_current_state(TASK_UNINTERRUPTIBLE);
/* deliver any messages that are in the queue */
if (call->state < AFS_CALL_COMPLETE && call->need_attention) {
if (!afs_check_call_state(call, AFS_CALL_COMPLETE) &&
call->need_attention) {
call->need_attention = false;
__set_current_state(TASK_RUNNING);
afs_deliver_to_call(call);
continue;
}
if (call->state == AFS_CALL_COMPLETE)
if (afs_check_call_state(call, AFS_CALL_COMPLETE))
break;
life = rxrpc_kernel_check_life(afs_socket, call->rxcall);
life = rxrpc_kernel_check_life(call->net->socket, call->rxcall);
if (timeout == 0 &&
life == last_life && signal_pending(current))
break;
@ -587,16 +584,34 @@ static int afs_wait_for_call_to_complete(struct afs_call *call)
__set_current_state(TASK_RUNNING);
/* Kill off the call if it's still live. */
if (call->state < AFS_CALL_COMPLETE) {
if (!afs_check_call_state(call, AFS_CALL_COMPLETE)) {
_debug("call interrupted");
rxrpc_kernel_abort_call(afs_socket, call->rxcall,
RX_USER_ABORT, -EINTR, "KWI");
if (rxrpc_kernel_abort_call(call->net->socket, call->rxcall,
RX_USER_ABORT, -EINTR, "KWI"))
afs_set_call_complete(call, -EINTR, 0);
}
spin_lock_bh(&call->state_lock);
ac->abort_code = call->abort_code;
ac->error = call->error;
spin_unlock_bh(&call->state_lock);
ret = ac->error;
switch (ret) {
case 0:
if (call->ret_reply0) {
ret = (long)call->reply[0];
call->reply[0] = NULL;
}
/* Fall through */
case -ECONNABORTED:
ac->responded = true;
break;
}
ret = call->error;
_debug("call complete");
afs_put_call(call);
_leave(" = %d", ret);
_leave(" = %p", (void *)ret);
return ret;
}
@ -627,7 +642,7 @@ static void afs_wake_up_async_call(struct sock *sk, struct rxrpc_call *rxcall,
u = __atomic_add_unless(&call->usage, 1, 0);
if (u != 0) {
trace_afs_call(call, afs_call_trace_wake, u,
atomic_read(&afs_outstanding_calls),
atomic_read(&call->net->nr_outstanding_calls),
__builtin_return_address(0));
if (!queue_work(afs_async_calls, &call->async_work))
@ -666,7 +681,7 @@ static void afs_process_async_call(struct work_struct *work)
}
if (call->state == AFS_CALL_COMPLETE) {
call->reply = NULL;
call->reply[0] = NULL;
/* We have two refs to release - one from the alloc and one
* queued with the work item - and we can't just deallocate the
@ -691,22 +706,24 @@ static void afs_rx_attach(struct rxrpc_call *rxcall, unsigned long user_call_ID)
/*
* Charge the incoming call preallocation.
*/
static void afs_charge_preallocation(struct work_struct *work)
void afs_charge_preallocation(struct work_struct *work)
{
struct afs_call *call = afs_spare_incoming_call;
struct afs_net *net =
container_of(work, struct afs_net, charge_preallocation_work);
struct afs_call *call = net->spare_incoming_call;
for (;;) {
if (!call) {
call = afs_alloc_call(&afs_RXCMxxxx, GFP_KERNEL);
call = afs_alloc_call(net, &afs_RXCMxxxx, GFP_KERNEL);
if (!call)
break;
call->async = true;
call->state = AFS_CALL_AWAIT_OP_ID;
call->state = AFS_CALL_SV_AWAIT_OP_ID;
init_waitqueue_head(&call->waitq);
}
if (rxrpc_kernel_charge_accept(afs_socket,
if (rxrpc_kernel_charge_accept(net->socket,
afs_wake_up_async_call,
afs_rx_attach,
(unsigned long)call,
@ -714,7 +731,7 @@ static void afs_charge_preallocation(struct work_struct *work)
break;
call = NULL;
}
afs_spare_incoming_call = call;
net->spare_incoming_call = call;
}
/*
@ -735,7 +752,9 @@ static void afs_rx_discard_new_call(struct rxrpc_call *rxcall,
static void afs_rx_new_call(struct sock *sk, struct rxrpc_call *rxcall,
unsigned long user_call_ID)
{
queue_work(afs_wq, &afs_charge_preallocation_work);
struct afs_net *net = afs_sock2net(sk);
queue_work(afs_wq, &net->charge_preallocation_work);
}
/*
@ -756,7 +775,7 @@ static int afs_deliver_cm_op_id(struct afs_call *call)
return ret;
call->operation_ID = ntohl(call->tmp);
call->state = AFS_CALL_AWAIT_REQUEST;
afs_set_call_state(call, AFS_CALL_SV_AWAIT_OP_ID, AFS_CALL_SV_AWAIT_REQUEST);
call->offset = 0;
/* ask the cache manager to route the call (it'll change the call type
@ -781,8 +800,7 @@ static void afs_notify_end_reply_tx(struct sock *sock,
{
struct afs_call *call = (struct afs_call *)call_user_ID;
if (call->state == AFS_CALL_REPLYING)
call->state = AFS_CALL_AWAIT_ACK;
afs_set_call_state(call, AFS_CALL_SV_REPLYING, AFS_CALL_SV_AWAIT_ACK);
}
/*
@ -790,11 +808,12 @@ static void afs_notify_end_reply_tx(struct sock *sock,
*/
void afs_send_empty_reply(struct afs_call *call)
{
struct afs_net *net = call->net;
struct msghdr msg;
_enter("");
rxrpc_kernel_set_tx_length(afs_socket, call->rxcall, 0);
rxrpc_kernel_set_tx_length(net->socket, call->rxcall, 0);
msg.msg_name = NULL;
msg.msg_namelen = 0;
@ -803,8 +822,7 @@ void afs_send_empty_reply(struct afs_call *call)
msg.msg_controllen = 0;
msg.msg_flags = 0;
call->state = AFS_CALL_AWAIT_ACK;
switch (rxrpc_kernel_send_data(afs_socket, call->rxcall, &msg, 0,
switch (rxrpc_kernel_send_data(net->socket, call->rxcall, &msg, 0,
afs_notify_end_reply_tx)) {
case 0:
_leave(" [replied]");
@ -812,7 +830,7 @@ void afs_send_empty_reply(struct afs_call *call)
case -ENOMEM:
_debug("oom");
rxrpc_kernel_abort_call(afs_socket, call->rxcall,
rxrpc_kernel_abort_call(net->socket, call->rxcall,
RX_USER_ABORT, -ENOMEM, "KOO");
default:
_leave(" [error]");
@ -825,13 +843,14 @@ void afs_send_empty_reply(struct afs_call *call)
*/
void afs_send_simple_reply(struct afs_call *call, const void *buf, size_t len)
{
struct afs_net *net = call->net;
struct msghdr msg;
struct kvec iov[1];
int n;
_enter("");
rxrpc_kernel_set_tx_length(afs_socket, call->rxcall, len);
rxrpc_kernel_set_tx_length(net->socket, call->rxcall, len);
iov[0].iov_base = (void *) buf;
iov[0].iov_len = len;
@ -842,8 +861,7 @@ void afs_send_simple_reply(struct afs_call *call, const void *buf, size_t len)
msg.msg_controllen = 0;
msg.msg_flags = 0;
call->state = AFS_CALL_AWAIT_ACK;
n = rxrpc_kernel_send_data(afs_socket, call->rxcall, &msg, len,
n = rxrpc_kernel_send_data(net->socket, call->rxcall, &msg, len,
afs_notify_end_reply_tx);
if (n >= 0) {
/* Success */
@ -853,7 +871,7 @@ void afs_send_simple_reply(struct afs_call *call, const void *buf, size_t len)
if (n == -ENOMEM) {
_debug("oom");
rxrpc_kernel_abort_call(afs_socket, call->rxcall,
rxrpc_kernel_abort_call(net->socket, call->rxcall,
RX_USER_ABORT, -ENOMEM, "KOO");
}
_leave(" [error]");
@ -865,6 +883,9 @@ void afs_send_simple_reply(struct afs_call *call, const void *buf, size_t len)
int afs_extract_data(struct afs_call *call, void *buf, size_t count,
bool want_more)
{
struct afs_net *net = call->net;
enum afs_call_state state;
u32 remote_abort;
int ret;
_enter("{%s,%zu},,%zu,%d",
@ -872,32 +893,32 @@ int afs_extract_data(struct afs_call *call, void *buf, size_t count,
ASSERTCMP(call->offset, <=, count);
ret = rxrpc_kernel_recv_data(afs_socket, call->rxcall,
ret = rxrpc_kernel_recv_data(net->socket, call->rxcall,
buf, count, &call->offset,
want_more, &call->abort_code,
want_more, &remote_abort,
&call->service_id);
trace_afs_recv_data(call, count, call->offset, want_more, ret);
if (ret == 0 || ret == -EAGAIN)
return ret;
state = READ_ONCE(call->state);
if (ret == 1) {
switch (call->state) {
case AFS_CALL_AWAIT_REPLY:
call->state = AFS_CALL_COMPLETE;
switch (state) {
case AFS_CALL_CL_AWAIT_REPLY:
afs_set_call_state(call, state, AFS_CALL_CL_PROC_REPLY);
break;
case AFS_CALL_AWAIT_REQUEST:
call->state = AFS_CALL_REPLYING;
case AFS_CALL_SV_AWAIT_REQUEST:
afs_set_call_state(call, state, AFS_CALL_SV_REPLYING);
break;
case AFS_CALL_COMPLETE:
kdebug("prem complete %d", call->error);
return -EIO;
default:
break;
}
return 0;
}
if (ret == -ECONNABORTED)
call->error = call->type->abort_to_error(call->abort_code);
else
call->error = ret;
call->state = AFS_CALL_COMPLETE;
afs_set_call_complete(call, ret, remote_abort);
return ret;
}

378
fs/afs/security.c
View File

@ -1,6 +1,6 @@
/* AFS security handling
*
* Copyright (C) 2007 Red Hat, Inc. All Rights Reserved.
* Copyright (C) 2007, 2017 Red Hat, Inc. All Rights Reserved.
* Written by David Howells (dhowells@redhat.com)
*
* This program is free software; you can redistribute it and/or
@ -14,9 +14,13 @@
#include <linux/fs.h>
#include <linux/ctype.h>
#include <linux/sched.h>
#include <linux/hashtable.h>
#include <keys/rxrpc-type.h>
#include "internal.h"
static DEFINE_HASHTABLE(afs_permits_cache, 10);
static DEFINE_SPINLOCK(afs_permits_lock);
/*
* get a key
*/
@ -46,167 +50,233 @@ struct key *afs_request_key(struct afs_cell *cell)
}
/*
* dispose of a permits list
* Dispose of a list of permits.
*/
void afs_zap_permits(struct rcu_head *rcu)
static void afs_permits_rcu(struct rcu_head *rcu)
{
struct afs_permits *permits =
container_of(rcu, struct afs_permits, rcu);
int loop;
int i;
_enter("{%d}", permits->count);
for (loop = permits->count - 1; loop >= 0; loop--)
key_put(permits->permits[loop].key);
for (i = 0; i < permits->nr_permits; i++)
key_put(permits->permits[i].key);
kfree(permits);
}
/*
* dispose of a permits list in which all the key pointers have been copied
* Discard a permission cache.
*/
static void afs_dispose_of_permits(struct rcu_head *rcu)
void afs_put_permits(struct afs_permits *permits)
{
struct afs_permits *permits =
container_of(rcu, struct afs_permits, rcu);
_enter("{%d}", permits->count);
kfree(permits);
}
/*
* get the authorising vnode - this is the specified inode itself if it's a
* directory or it's the parent directory if the specified inode is a file or
* symlink
* - the caller must release the ref on the inode
*/
static struct afs_vnode *afs_get_auth_inode(struct afs_vnode *vnode,
struct key *key)
{
struct afs_vnode *auth_vnode;
struct inode *auth_inode;
_enter("");
if (S_ISDIR(vnode->vfs_inode.i_mode)) {
auth_inode = igrab(&vnode->vfs_inode);
ASSERT(auth_inode != NULL);
} else {
auth_inode = afs_iget(vnode->vfs_inode.i_sb, key,
&vnode->status.parent, NULL, NULL);
if (IS_ERR(auth_inode))
return ERR_CAST(auth_inode);
if (permits && refcount_dec_and_test(&permits->usage)) {
spin_lock(&afs_permits_lock);
hash_del_rcu(&permits->hash_node);
spin_unlock(&afs_permits_lock);
call_rcu(&permits->rcu, afs_permits_rcu);
}
auth_vnode = AFS_FS_I(auth_inode);
_leave(" = {%x}", auth_vnode->fid.vnode);
return auth_vnode;
}
/*
* clear the permit cache on a directory vnode
* Clear a permit cache on callback break.
*/
void afs_clear_permits(struct afs_vnode *vnode)
{
struct afs_permits *permits;
_enter("{%x:%u}", vnode->fid.vid, vnode->fid.vnode);
mutex_lock(&vnode->permits_lock);
permits = vnode->permits;
RCU_INIT_POINTER(vnode->permits, NULL);
mutex_unlock(&vnode->permits_lock);
spin_lock(&vnode->lock);
permits = rcu_dereference_protected(vnode->permit_cache,
lockdep_is_held(&vnode->lock));
RCU_INIT_POINTER(vnode->permit_cache, NULL);
vnode->cb_break++;
spin_unlock(&vnode->lock);
if (permits)
call_rcu(&permits->rcu, afs_zap_permits);
_leave("");
afs_put_permits(permits);
}
/*
* add the result obtained for a vnode to its or its parent directory's cache
* for the key used to access it
* Hash a list of permits. Use simple addition to make it easy to add an extra
* one at an as-yet indeterminate position in the list.
*/
void afs_cache_permit(struct afs_vnode *vnode, struct key *key, long acl_order)
static void afs_hash_permits(struct afs_permits *permits)
{
struct afs_permits *permits, *xpermits;
struct afs_permit *permit;
struct afs_vnode *auth_vnode;
int count, loop;
unsigned long h = permits->nr_permits;
int i;
_enter("{%x:%u},%x,%lx",
vnode->fid.vid, vnode->fid.vnode, key_serial(key), acl_order);
auth_vnode = afs_get_auth_inode(vnode, key);
if (IS_ERR(auth_vnode)) {
_leave(" [get error %ld]", PTR_ERR(auth_vnode));
return;
for (i = 0; i < permits->nr_permits; i++) {
h += (unsigned long)permits->permits[i].key / sizeof(void *);
h += permits->permits[i].access;
}
mutex_lock(&auth_vnode->permits_lock);
permits->h = h;
}
/* guard against a rename being detected whilst we waited for the
* lock */
if (memcmp(&auth_vnode->fid, &vnode->status.parent,
sizeof(struct afs_fid)) != 0) {
_debug("renamed");
goto out_unlock;
}
/*
* Cache the CallerAccess result obtained from doing a fileserver operation
* that returned a vnode status for a particular key. If a callback break
* occurs whilst the operation was in progress then we have to ditch the cache
* as the ACL *may* have changed.
*/
void afs_cache_permit(struct afs_vnode *vnode, struct key *key,
unsigned int cb_break)
{
struct afs_permits *permits, *xpermits, *replacement, *new = NULL;
afs_access_t caller_access = READ_ONCE(vnode->status.caller_access);
size_t size = 0;
bool changed = false;
int i, j;
/* have to be careful as the directory's callback may be broken between
* us receiving the status we're trying to cache and us getting the
* lock to update the cache for the status */
if (auth_vnode->acl_order - acl_order > 0) {
_debug("ACL changed?");
goto out_unlock;
}
_enter("{%x:%u},%x,%x",
vnode->fid.vid, vnode->fid.vnode, key_serial(key), caller_access);
/* always update the anonymous mask */
_debug("anon access %x", vnode->status.anon_access);
auth_vnode->status.anon_access = vnode->status.anon_access;
if (key == vnode->volume->cell->anonymous_key)
goto out_unlock;
rcu_read_lock();
xpermits = auth_vnode->permits;
count = 0;
if (xpermits) {
/* see if the permit is already in the list
* - if it is then we just amend the list
*/
count = xpermits->count;
permit = xpermits->permits;
for (loop = count; loop > 0; loop--) {
if (permit->key == key) {
permit->access_mask =
vnode->status.caller_access;
goto out_unlock;
/* Check for the common case first: We got back the same access as last
* time we tried and already have it recorded.
*/
permits = rcu_dereference(vnode->permit_cache);
if (permits) {
if (!permits->invalidated) {
for (i = 0; i < permits->nr_permits; i++) {
if (permits->permits[i].key < key)
continue;
if (permits->permits[i].key > key)
break;
if (permits->permits[i].access != caller_access) {
changed = true;
break;
}
if (cb_break != (vnode->cb_break +
vnode->cb_interest->server->cb_s_break)) {
changed = true;
break;
}
/* The cache is still good. */
rcu_read_unlock();
return;
}
permit++;
}
changed |= permits->invalidated;
size = permits->nr_permits;
/* If this set of permits is now wrong, clear the permits
* pointer so that no one tries to use the stale information.
*/
if (changed) {
spin_lock(&vnode->lock);
if (permits != rcu_access_pointer(vnode->permit_cache))
goto someone_else_changed_it_unlock;
RCU_INIT_POINTER(vnode->permit_cache, NULL);
spin_unlock(&vnode->lock);
afs_put_permits(permits);
permits = NULL;
size = 0;
}
}
permits = kmalloc(sizeof(*permits) + sizeof(*permit) * (count + 1),
GFP_NOFS);
if (!permits)
goto out_unlock;
if (cb_break != (vnode->cb_break + vnode->cb_interest->server->cb_s_break)) {
rcu_read_unlock();
goto someone_else_changed_it;
}
if (xpermits)
memcpy(permits->permits, xpermits->permits,
count * sizeof(struct afs_permit));
/* We need a ref on any permits list we want to copy as we'll have to
* drop the lock to do memory allocation.
*/
if (permits && !refcount_inc_not_zero(&permits->usage)) {
rcu_read_unlock();
goto someone_else_changed_it;
}
_debug("key %x access %x",
key_serial(key), vnode->status.caller_access);
permits->permits[count].access_mask = vnode->status.caller_access;
permits->permits[count].key = key_get(key);
permits->count = count + 1;
rcu_read_unlock();
rcu_assign_pointer(auth_vnode->permits, permits);
if (xpermits)
call_rcu(&xpermits->rcu, afs_dispose_of_permits);
/* Speculatively create a new list with the revised permission set. We
* discard this if we find an extant match already in the hash, but
* it's easier to compare with memcmp this way.
*
* We fill in the key pointers at this time, but we don't get the refs
* yet.
*/
size++;
new = kzalloc(sizeof(struct afs_permits) +
sizeof(struct afs_permit) * size, GFP_NOFS);
if (!new)
return;
out_unlock:
mutex_unlock(&auth_vnode->permits_lock);
iput(&auth_vnode->vfs_inode);
_leave("");
refcount_set(&new->usage, 1);
new->nr_permits = size;
i = j = 0;
if (permits) {
for (i = 0; i < permits->nr_permits; i++) {
if (j == i && permits->permits[i].key > key) {
new->permits[j].key = key;
new->permits[j].access = caller_access;
j++;
}
new->permits[j].key = permits->permits[i].key;
new->permits[j].access = permits->permits[i].access;
j++;
}
}
if (j == i) {
new->permits[j].key = key;
new->permits[j].access = caller_access;
}
afs_hash_permits(new);
afs_put_permits(permits);
/* Now see if the permit list we want is actually already available */
spin_lock(&afs_permits_lock);
hash_for_each_possible(afs_permits_cache, xpermits, hash_node, new->h) {
if (xpermits->h != new->h ||
xpermits->invalidated ||
xpermits->nr_permits != new->nr_permits ||
memcmp(xpermits->permits, new->permits,
new->nr_permits * sizeof(struct afs_permit)) != 0)
continue;
if (refcount_inc_not_zero(&xpermits->usage)) {
replacement = xpermits;
goto found;
}
break;
}
for (i = 0; i < new->nr_permits; i++)
key_get(new->permits[i].key);
hash_add_rcu(afs_permits_cache, &new->hash_node, new->h);
replacement = new;
new = NULL;
found:
spin_unlock(&afs_permits_lock);
kfree(new);
spin_lock(&vnode->lock);
if (cb_break != (vnode->cb_break + vnode->cb_interest->server->cb_s_break) ||
permits != rcu_access_pointer(vnode->permit_cache))
goto someone_else_changed_it_unlock;
rcu_assign_pointer(vnode->permit_cache, replacement);
spin_unlock(&vnode->lock);
afs_put_permits(permits);
return;
someone_else_changed_it_unlock:
spin_unlock(&vnode->lock);
someone_else_changed_it:
/* Someone else changed the cache under us - don't recheck at this
* time.
*/
return;
}
/*
@ -218,56 +288,45 @@ static int afs_check_permit(struct afs_vnode *vnode, struct key *key,
afs_access_t *_access)
{
struct afs_permits *permits;
struct afs_permit *permit;
struct afs_vnode *auth_vnode;
bool valid;
int loop, ret;
bool valid = false;
int i, ret;
_enter("{%x:%u},%x",
vnode->fid.vid, vnode->fid.vnode, key_serial(key));
auth_vnode = afs_get_auth_inode(vnode, key);
if (IS_ERR(auth_vnode)) {
*_access = 0;
_leave(" = %ld", PTR_ERR(auth_vnode));
return PTR_ERR(auth_vnode);
}
ASSERT(S_ISDIR(auth_vnode->vfs_inode.i_mode));
permits = vnode->permit_cache;
/* check the permits to see if we've got one yet */
if (key == auth_vnode->volume->cell->anonymous_key) {
if (key == vnode->volume->cell->anonymous_key) {
_debug("anon");
*_access = auth_vnode->status.anon_access;
*_access = vnode->status.anon_access;
valid = true;
} else {
valid = false;
rcu_read_lock();
permits = rcu_dereference(auth_vnode->permits);
permits = rcu_dereference(vnode->permit_cache);
if (permits) {
permit = permits->permits;
for (loop = permits->count; loop > 0; loop--) {
if (permit->key == key) {
_debug("found in cache");
*_access = permit->access_mask;
valid = true;
for (i = 0; i < permits->nr_permits; i++) {
if (permits->permits[i].key < key)
continue;
if (permits->permits[i].key > key)
break;
}
permit++;
*_access = permits->permits[i].access;
valid = !permits->invalidated;
break;
}
}
rcu_read_unlock();
}
if (!valid) {
/* check the status on the file we're actually interested in
* (the post-processing will cache the result on auth_vnode) */
/* Check the status on the file we're actually interested in
* (the post-processing will cache the result).
*/
_debug("no valid permit");
set_bit(AFS_VNODE_CB_BROKEN, &vnode->flags);
ret = afs_vnode_fetch_status(vnode, auth_vnode, key);
ret = afs_fetch_status(vnode, key);
if (ret < 0) {
iput(&auth_vnode->vfs_inode);
*_access = 0;
_leave(" = %d", ret);
return ret;
@ -275,7 +334,6 @@ static int afs_check_permit(struct afs_vnode *vnode, struct key *key,
*_access = vnode->status.caller_access;
}
iput(&auth_vnode->vfs_inode);
_leave(" = 0 [access %x]", *_access);
return 0;
}
@ -304,14 +362,9 @@ int afs_permission(struct inode *inode, int mask)
return PTR_ERR(key);
}
/* if the promise has expired, we need to check the server again */
if (!vnode->cb_promised) {
_debug("not promised");
ret = afs_vnode_fetch_status(vnode, NULL, key);
if (ret < 0)
goto error;
_debug("new promise [fl=%lx]", vnode->flags);
}
ret = afs_validate(vnode, key);
if (ret < 0)
goto error;
/* check the permits to see if we've got one yet */
ret = afs_check_permit(vnode, key, &access);
@ -365,3 +418,12 @@ error:
_leave(" = %d", ret);
return ret;
}
void __exit afs_clean_up_permit_cache(void)
{
int i;
for (i = 0; i < HASH_SIZE(afs_permits_cache); i++)
WARN_ON_ONCE(!hlist_empty(&afs_permits_cache[i]));
}

826
fs/afs/server.c
View File

@ -11,317 +11,689 @@
#include <linux/sched.h>
#include <linux/slab.h>
#include "afs_fs.h"
#include "internal.h"
static unsigned afs_server_timeout = 10; /* server timeout in seconds */
static unsigned afs_server_gc_delay = 10; /* Server record timeout in seconds */
static unsigned afs_server_update_delay = 30; /* Time till VLDB recheck in secs */
static void afs_reap_server(struct work_struct *);
static void afs_inc_servers_outstanding(struct afs_net *net)
{
atomic_inc(&net->servers_outstanding);
}
/* tree of all the servers, indexed by IP address */
static struct rb_root afs_servers = RB_ROOT;
static DEFINE_RWLOCK(afs_servers_lock);
/* LRU list of all the servers not currently in use */
static LIST_HEAD(afs_server_graveyard);
static DEFINE_SPINLOCK(afs_server_graveyard_lock);
static DECLARE_DELAYED_WORK(afs_server_reaper, afs_reap_server);
static void afs_dec_servers_outstanding(struct afs_net *net)
{
if (atomic_dec_and_test(&net->servers_outstanding))
wake_up_atomic_t(&net->servers_outstanding);
}
/*
* install a server record in the master tree
* Find a server by one of its addresses.
*/
static int afs_install_server(struct afs_server *server)
struct afs_server *afs_find_server(struct afs_net *net,
const struct sockaddr_rxrpc *srx)
{
struct afs_server *xserver;
const struct sockaddr_in6 *a = &srx->transport.sin6, *b;
const struct afs_addr_list *alist;
struct afs_server *server = NULL;
unsigned int i;
bool ipv6 = true;
int seq = 0, diff;
if (srx->transport.sin6.sin6_addr.s6_addr32[0] == 0 ||
srx->transport.sin6.sin6_addr.s6_addr32[1] == 0 ||
srx->transport.sin6.sin6_addr.s6_addr32[2] == htonl(0xffff))
ipv6 = false;
rcu_read_lock();
do {
if (server)
afs_put_server(net, server);
server = NULL;
read_seqbegin_or_lock(&net->fs_addr_lock, &seq);
if (ipv6) {
hlist_for_each_entry_rcu(server, &net->fs_addresses6, addr6_link) {
alist = rcu_dereference(server->addresses);
for (i = alist->nr_ipv4; i < alist->nr_addrs; i++) {
b = &alist->addrs[i].transport.sin6;
diff = (u16)a->sin6_port - (u16)b->sin6_port;
if (diff == 0)
diff = memcmp(&a->sin6_addr,
&b->sin6_addr,
sizeof(struct in6_addr));
if (diff == 0)
goto found;
if (diff < 0) {
// TODO: Sort the list
//if (i == alist->nr_ipv4)
// goto not_found;
break;
}
}
}
} else {
hlist_for_each_entry_rcu(server, &net->fs_addresses4, addr4_link) {
alist = rcu_dereference(server->addresses);
for (i = 0; i < alist->nr_ipv4; i++) {
b = &alist->addrs[i].transport.sin6;
diff = (u16)a->sin6_port - (u16)b->sin6_port;
if (diff == 0)
diff = ((u32)a->sin6_addr.s6_addr32[3] -
(u32)b->sin6_addr.s6_addr32[3]);
if (diff == 0)
goto found;
if (diff < 0) {
// TODO: Sort the list
//if (i == 0)
// goto not_found;
break;
}
}
}
}
//not_found:
server = NULL;
found:
if (server && !atomic_inc_not_zero(&server->usage))
server = NULL;
} while (need_seqretry(&net->fs_addr_lock, seq));
done_seqretry(&net->fs_addr_lock, seq);
rcu_read_unlock();
return server;
}
/*
* Look up a server by its UUID
*/
struct afs_server *afs_find_server_by_uuid(struct afs_net *net, const uuid_t *uuid)
{
struct afs_server *server = NULL;
struct rb_node *p;
int diff, seq = 0;
_enter("%pU", uuid);
do {
/* Unfortunately, rbtree walking doesn't give reliable results
* under just the RCU read lock, so we have to check for
* changes.
*/
if (server)
afs_put_server(net, server);
server = NULL;
read_seqbegin_or_lock(&net->fs_lock, &seq);
p = net->fs_servers.rb_node;
while (p) {
server = rb_entry(p, struct afs_server, uuid_rb);
diff = memcmp(uuid, &server->uuid, sizeof(*uuid));
if (diff < 0) {
p = p->rb_left;
} else if (diff > 0) {
p = p->rb_right;
} else {
afs_get_server(server);
break;
}
server = NULL;
}
} while (need_seqretry(&net->fs_lock, seq));
done_seqretry(&net->fs_lock, seq);
_leave(" = %p", server);
return server;
}
/*
* Install a server record in the namespace tree
*/
static struct afs_server *afs_install_server(struct afs_net *net,
struct afs_server *candidate)
{
const struct afs_addr_list *alist;
struct afs_server *server;
struct rb_node **pp, *p;
int ret;
int ret = -EEXIST, diff;
_enter("%p", server);
_enter("%p", candidate);
write_lock(&afs_servers_lock);
write_seqlock(&net->fs_lock);
ret = -EEXIST;
pp = &afs_servers.rb_node;
/* Firstly install the server in the UUID lookup tree */
pp = &net->fs_servers.rb_node;
p = NULL;
while (*pp) {
p = *pp;
_debug("- consider %p", p);
xserver = rb_entry(p, struct afs_server, master_rb);
if (server->addr.s_addr < xserver->addr.s_addr)
server = rb_entry(p, struct afs_server, uuid_rb);
diff = memcmp(&candidate->uuid, &server->uuid, sizeof(uuid_t));
if (diff < 0)
pp = &(*pp)->rb_left;
else if (server->addr.s_addr > xserver->addr.s_addr)
else if (diff > 0)
pp = &(*pp)->rb_right;
else
goto error;
goto exists;
}
rb_link_node(&server->master_rb, p, pp);
rb_insert_color(&server->master_rb, &afs_servers);
server = candidate;
rb_link_node(&server->uuid_rb, p, pp);
rb_insert_color(&server->uuid_rb, &net->fs_servers);
hlist_add_head_rcu(&server->proc_link, &net->fs_proc);
write_seqlock(&net->fs_addr_lock);
alist = rcu_dereference_protected(server->addresses,
lockdep_is_held(&net->fs_addr_lock.lock));
/* Secondly, if the server has any IPv4 and/or IPv6 addresses, install
* it in the IPv4 and/or IPv6 reverse-map lists.
*
* TODO: For speed we want to use something other than a flat list
* here; even sorting the list in terms of lowest address would help a
* bit, but anything we might want to do gets messy and memory
* intensive.
*/
if (alist->nr_ipv4 > 0)
hlist_add_head_rcu(&server->addr4_link, &net->fs_addresses4);
if (alist->nr_addrs > alist->nr_ipv4)
hlist_add_head_rcu(&server->addr6_link, &net->fs_addresses6);
write_sequnlock(&net->fs_addr_lock);
ret = 0;
error:
write_unlock(&afs_servers_lock);
return ret;
exists:
afs_get_server(server);
write_sequnlock(&net->fs_lock);
return server;
}
/*
* allocate a new server record
*/
static struct afs_server *afs_alloc_server(struct afs_cell *cell,
const struct in_addr *addr)
static struct afs_server *afs_alloc_server(struct afs_net *net,
const uuid_t *uuid,
struct afs_addr_list *alist)
{
struct afs_server *server;
_enter("");
server = kzalloc(sizeof(struct afs_server), GFP_KERNEL);
if (server) {
atomic_set(&server->usage, 1);
server->cell = cell;
if (!server)
goto enomem;
INIT_LIST_HEAD(&server->link);
INIT_LIST_HEAD(&server->grave);
init_rwsem(&server->sem);
spin_lock_init(&server->fs_lock);
server->fs_vnodes = RB_ROOT;
server->cb_promises = RB_ROOT;
spin_lock_init(&server->cb_lock);
init_waitqueue_head(&server->cb_break_waitq);
INIT_DELAYED_WORK(&server->cb_break_work,
afs_dispatch_give_up_callbacks);
atomic_set(&server->usage, 1);
RCU_INIT_POINTER(server->addresses, alist);
server->addr_version = alist->version;
server->uuid = *uuid;
server->flags = (1UL << AFS_SERVER_FL_NEW);
server->update_at = ktime_get_real_seconds() + afs_server_update_delay;
rwlock_init(&server->fs_lock);
INIT_LIST_HEAD(&server->cb_interests);
rwlock_init(&server->cb_break_lock);
memcpy(&server->addr, addr, sizeof(struct in_addr));
server->addr.s_addr = addr->s_addr;
_leave(" = %p{%d}", server, atomic_read(&server->usage));
} else {
_leave(" = NULL [nomem]");
}
afs_inc_servers_outstanding(net);
_leave(" = %p", server);
return server;
enomem:
_leave(" = NULL [nomem]");
return NULL;
}
/*
* get an FS-server record for a cell
* Look up an address record for a server
*/
struct afs_server *afs_lookup_server(struct afs_cell *cell,
const struct in_addr *addr)
static struct afs_addr_list *afs_vl_lookup_addrs(struct afs_cell *cell,
struct key *key, const uuid_t *uuid)
{
struct afs_server *server, *candidate;
struct afs_addr_cursor ac;
struct afs_addr_list *alist;
int ret;
_enter("%p,%pI4", cell, &addr->s_addr);
ret = afs_set_vl_cursor(&ac, cell);
if (ret < 0)
return ERR_PTR(ret);
/* quick scan of the list to see if we already have the server */
read_lock(&cell->servers_lock);
list_for_each_entry(server, &cell->servers, link) {
if (server->addr.s_addr == addr->s_addr)
goto found_server_quickly;
}
read_unlock(&cell->servers_lock);
candidate = afs_alloc_server(cell, addr);
if (!candidate) {
_leave(" = -ENOMEM");
return ERR_PTR(-ENOMEM);
}
write_lock(&cell->servers_lock);
/* check the cell's server list again */
list_for_each_entry(server, &cell->servers, link) {
if (server->addr.s_addr == addr->s_addr)
goto found_server;
}
_debug("new");
server = candidate;
if (afs_install_server(server) < 0)
goto server_in_two_cells;
afs_get_cell(cell);
list_add_tail(&server->link, &cell->servers);
write_unlock(&cell->servers_lock);
_leave(" = %p{%d}", server, atomic_read(&server->usage));
return server;
/* found a matching server quickly */
found_server_quickly:
_debug("found quickly");
afs_get_server(server);
read_unlock(&cell->servers_lock);
no_longer_unused:
if (!list_empty(&server->grave)) {
spin_lock(&afs_server_graveyard_lock);
list_del_init(&server->grave);
spin_unlock(&afs_server_graveyard_lock);
}
_leave(" = %p{%d}", server, atomic_read(&server->usage));
return server;
/* found a matching server on the second pass */
found_server:
_debug("found");
afs_get_server(server);
write_unlock(&cell->servers_lock);
kfree(candidate);
goto no_longer_unused;
/* found a server that seems to be in two cells */
server_in_two_cells:
write_unlock(&cell->servers_lock);
kfree(candidate);
printk(KERN_NOTICE "kAFS: Server %pI4 appears to be in two cells\n",
addr);
_leave(" = -EEXIST");
return ERR_PTR(-EEXIST);
}
/*
* look up a server by its IP address
*/
struct afs_server *afs_find_server(const struct sockaddr_rxrpc *srx)
{
struct afs_server *server = NULL;
struct rb_node *p;
struct in_addr addr = srx->transport.sin.sin_addr;
_enter("{%d,%pI4}", srx->transport.family, &addr.s_addr);
if (srx->transport.family != AF_INET) {
WARN(true, "AFS does not yes support non-IPv4 addresses\n");
return NULL;
}
read_lock(&afs_servers_lock);
p = afs_servers.rb_node;
while (p) {
server = rb_entry(p, struct afs_server, master_rb);
_debug("- consider %p", p);
if (addr.s_addr < server->addr.s_addr) {
p = p->rb_left;
} else if (addr.s_addr > server->addr.s_addr) {
p = p->rb_right;
} else {
afs_get_server(server);
goto found;
while (afs_iterate_addresses(&ac)) {
if (test_bit(ac.index, &ac.alist->yfs))
alist = afs_yfsvl_get_endpoints(cell->net, &ac, key, uuid);
else
alist = afs_vl_get_addrs_u(cell->net, &ac, key, uuid);
switch (ac.error) {
case 0:
afs_end_cursor(&ac);
return alist;
case -ECONNABORTED:
ac.error = afs_abort_to_error(ac.abort_code);
goto error;
case -ENOMEM:
case -ENONET:
goto error;
case -ENETUNREACH:
case -EHOSTUNREACH:
case -ECONNREFUSED:
break;
default:
ac.error = -EIO;
goto error;
}
}
server = NULL;
found:
read_unlock(&afs_servers_lock);
ASSERTIFCMP(server, server->addr.s_addr, ==, addr.s_addr);
_leave(" = %p", server);
error:
return ERR_PTR(afs_end_cursor(&ac));
}
/*
* Get or create a fileserver record.
*/
struct afs_server *afs_lookup_server(struct afs_cell *cell, struct key *key,
const uuid_t *uuid)
{
struct afs_addr_list *alist;
struct afs_server *server, *candidate;
_enter("%p,%pU", cell->net, uuid);
server = afs_find_server_by_uuid(cell->net, uuid);
if (server)
return server;
alist = afs_vl_lookup_addrs(cell, key, uuid);
if (IS_ERR(alist))
return ERR_CAST(alist);
candidate = afs_alloc_server(cell->net, uuid, alist);
if (!candidate) {
afs_put_addrlist(alist);
return ERR_PTR(-ENOMEM);
}
server = afs_install_server(cell->net, candidate);
if (server != candidate) {
afs_put_addrlist(alist);
kfree(candidate);
}
_leave(" = %p{%d}", server, atomic_read(&server->usage));
return server;
}
/*
* destroy a server record
* - removes from the cell list
* Set the server timer to fire after a given delay, assuming it's not already
* set for an earlier time.
*/
void afs_put_server(struct afs_server *server)
static void afs_set_server_timer(struct afs_net *net, time64_t delay)
{
if (net->live) {
afs_inc_servers_outstanding(net);
if (timer_reduce(&net->fs_timer, jiffies + delay * HZ))
afs_dec_servers_outstanding(net);
}
}
/*
* Server management timer. We have an increment on fs_outstanding that we
* need to pass along to the work item.
*/
void afs_servers_timer(struct timer_list *timer)
{
struct afs_net *net = container_of(timer, struct afs_net, fs_timer);
_enter("");
if (!queue_work(afs_wq, &net->fs_manager))
afs_dec_servers_outstanding(net);
}
/*
* Release a reference on a server record.
*/
void afs_put_server(struct afs_net *net, struct afs_server *server)
{
unsigned int usage;
if (!server)
return;
_enter("%p{%d}", server, atomic_read(&server->usage));
server->put_time = ktime_get_real_seconds();
_debug("PUT SERVER %d", atomic_read(&server->usage));
usage = atomic_dec_return(&server->usage);
ASSERTCMP(atomic_read(&server->usage), >, 0);
_enter("{%u}", usage);
if (likely(!atomic_dec_and_test(&server->usage))) {
_leave("");
if (likely(usage > 0))
return;
}
afs_flush_callback_breaks(server);
afs_set_server_timer(net, afs_server_gc_delay);
}
spin_lock(&afs_server_graveyard_lock);
if (atomic_read(&server->usage) == 0) {
list_move_tail(&server->grave, &afs_server_graveyard);
server->time_of_death = ktime_get_real_seconds();
queue_delayed_work(afs_wq, &afs_server_reaper,
afs_server_timeout * HZ);
}
spin_unlock(&afs_server_graveyard_lock);
_leave(" [dead]");
static void afs_server_rcu(struct rcu_head *rcu)
{
struct afs_server *server = container_of(rcu, struct afs_server, rcu);
afs_put_addrlist(server->addresses);
kfree(server);
}
/*
* destroy a dead server
*/
static void afs_destroy_server(struct afs_server *server)
static void afs_destroy_server(struct afs_net *net, struct afs_server *server)
{
struct afs_addr_list *alist = server->addresses;
struct afs_addr_cursor ac = {
.alist = alist,
.addr = &alist->addrs[0],
.start = alist->index,
.index = alist->index,
.error = 0,
};
_enter("%p", server);
ASSERTIF(server->cb_break_head != server->cb_break_tail,
delayed_work_pending(&server->cb_break_work));
ASSERTCMP(server->fs_vnodes.rb_node, ==, NULL);
ASSERTCMP(server->cb_promises.rb_node, ==, NULL);
ASSERTCMP(server->cb_break_head, ==, server->cb_break_tail);
ASSERTCMP(atomic_read(&server->cb_break_n), ==, 0);
afs_put_cell(server->cell);
kfree(server);
afs_fs_give_up_all_callbacks(net, server, &ac, NULL);
call_rcu(&server->rcu, afs_server_rcu);
afs_dec_servers_outstanding(net);
}
/*
* reap dead server records
* Garbage collect any expired servers.
*/
static void afs_reap_server(struct work_struct *work)
static void afs_gc_servers(struct afs_net *net, struct afs_server *gc_list)
{
LIST_HEAD(corpses);
struct afs_server *server;
unsigned long delay, expiry;
time64_t now;
bool deleted;
int usage;
now = ktime_get_real_seconds();
spin_lock(&afs_server_graveyard_lock);
while ((server = gc_list)) {
gc_list = server->gc_next;
while (!list_empty(&afs_server_graveyard)) {
server = list_entry(afs_server_graveyard.next,
struct afs_server, grave);
/* the queue is ordered most dead first */
expiry = server->time_of_death + afs_server_timeout;
if (expiry > now) {
delay = (expiry - now) * HZ;
mod_delayed_work(afs_wq, &afs_server_reaper, delay);
break;
write_seqlock(&net->fs_lock);
usage = 1;
deleted = atomic_try_cmpxchg(&server->usage, &usage, 0);
if (deleted) {
rb_erase(&server->uuid_rb, &net->fs_servers);
hlist_del_rcu(&server->proc_link);
}
write_sequnlock(&net->fs_lock);
write_lock(&server->cell->servers_lock);
write_lock(&afs_servers_lock);
if (atomic_read(&server->usage) > 0) {
list_del_init(&server->grave);
} else {
list_move_tail(&server->grave, &corpses);
list_del_init(&server->link);
rb_erase(&server->master_rb, &afs_servers);
}
write_unlock(&afs_servers_lock);
write_unlock(&server->cell->servers_lock);
}
spin_unlock(&afs_server_graveyard_lock);
/* now reap the corpses we've extracted */
while (!list_empty(&corpses)) {
server = list_entry(corpses.next, struct afs_server, grave);
list_del(&server->grave);
afs_destroy_server(server);
if (deleted)
afs_destroy_server(net, server);
}
}
/*
* discard all the server records for rmmod
* Manage the records of servers known to be within a network namespace. This
* includes garbage collecting unused servers.
*
* Note also that we were given an increment on net->servers_outstanding by
* whoever queued us that we need to deal with before returning.
*/
void __exit afs_purge_servers(void)
void afs_manage_servers(struct work_struct *work)
{
afs_server_timeout = 0;
mod_delayed_work(afs_wq, &afs_server_reaper, 0);
struct afs_net *net = container_of(work, struct afs_net, fs_manager);
struct afs_server *gc_list = NULL;
struct rb_node *cursor;
time64_t now = ktime_get_real_seconds(), next_manage = TIME64_MAX;
bool purging = !net->live;
_enter("");
/* Trawl the server list looking for servers that have expired from
* lack of use.
*/
read_seqlock_excl(&net->fs_lock);
for (cursor = rb_first(&net->fs_servers); cursor; cursor = rb_next(cursor)) {
struct afs_server *server =
rb_entry(cursor, struct afs_server, uuid_rb);
int usage = atomic_read(&server->usage);
_debug("manage %pU %u", &server->uuid, usage);
ASSERTCMP(usage, >=, 1);
ASSERTIFCMP(purging, usage, ==, 1);
if (usage == 1) {
time64_t expire_at = server->put_time;
if (!test_bit(AFS_SERVER_FL_VL_FAIL, &server->flags) &&
!test_bit(AFS_SERVER_FL_NOT_FOUND, &server->flags))
expire_at += afs_server_gc_delay;
if (purging || expire_at <= now) {
server->gc_next = gc_list;
gc_list = server;
} else if (expire_at < next_manage) {
next_manage = expire_at;
}
}
}
read_sequnlock_excl(&net->fs_lock);
/* Update the timer on the way out. We have to pass an increment on
* servers_outstanding in the namespace that we are in to the timer or
* the work scheduler.
*/
if (!purging && next_manage < TIME64_MAX) {
now = ktime_get_real_seconds();
if (next_manage - now <= 0) {
if (queue_work(afs_wq, &net->fs_manager))
afs_inc_servers_outstanding(net);
} else {
afs_set_server_timer(net, next_manage - now);
}
}
afs_gc_servers(net, gc_list);
afs_dec_servers_outstanding(net);
_leave(" [%d]", atomic_read(&net->servers_outstanding));
}
static void afs_queue_server_manager(struct afs_net *net)
{
afs_inc_servers_outstanding(net);
if (!queue_work(afs_wq, &net->fs_manager))
afs_dec_servers_outstanding(net);
}
/*
* Purge list of servers.
*/
void afs_purge_servers(struct afs_net *net)
{
_enter("");
if (del_timer_sync(&net->fs_timer))
atomic_dec(&net->servers_outstanding);
afs_queue_server_manager(net);
_debug("wait");
wait_on_atomic_t(&net->servers_outstanding, atomic_t_wait,
TASK_UNINTERRUPTIBLE);
_leave("");
}
/*
* Probe a fileserver to find its capabilities.
*
* TODO: Try service upgrade.
*/
static bool afs_do_probe_fileserver(struct afs_fs_cursor *fc)
{
_enter("");
fc->ac.addr = NULL;
fc->ac.start = READ_ONCE(fc->ac.alist->index);
fc->ac.index = fc->ac.start;
fc->ac.error = 0;
fc->ac.begun = false;
while (afs_iterate_addresses(&fc->ac)) {
afs_fs_get_capabilities(afs_v2net(fc->vnode), fc->cbi->server,
&fc->ac, fc->key);
switch (fc->ac.error) {
case 0:
afs_end_cursor(&fc->ac);
set_bit(AFS_SERVER_FL_PROBED, &fc->cbi->server->flags);
return true;
case -ECONNABORTED:
fc->ac.error = afs_abort_to_error(fc->ac.abort_code);
goto error;
case -ENOMEM:
case -ENONET:
goto error;
case -ENETUNREACH:
case -EHOSTUNREACH:
case -ECONNREFUSED:
case -ETIMEDOUT:
case -ETIME:
break;
default:
fc->ac.error = -EIO;
goto error;
}
}
error:
afs_end_cursor(&fc->ac);
return false;
}
/*
* If we haven't already, try probing the fileserver to get its capabilities.
* We try not to instigate parallel probes, but it's possible that the parallel
* probes will fail due to authentication failure when ours would succeed.
*
* TODO: Try sending an anonymous probe if an authenticated probe fails.
*/
bool afs_probe_fileserver(struct afs_fs_cursor *fc)
{
bool success;
int ret, retries = 0;
_enter("");
retry:
if (test_bit(AFS_SERVER_FL_PROBED, &fc->cbi->server->flags)) {
_leave(" = t");
return true;
}
if (!test_and_set_bit_lock(AFS_SERVER_FL_PROBING, &fc->cbi->server->flags)) {
success = afs_do_probe_fileserver(fc);
clear_bit_unlock(AFS_SERVER_FL_PROBING, &fc->cbi->server->flags);
wake_up_bit(&fc->cbi->server->flags, AFS_SERVER_FL_PROBING);
_leave(" = t");
return success;
}
_debug("wait");
ret = wait_on_bit(&fc->cbi->server->flags, AFS_SERVER_FL_PROBING,
TASK_INTERRUPTIBLE);
if (ret == -ERESTARTSYS) {
fc->ac.error = ret;
_leave(" = f [%d]", ret);
return false;
}
retries++;
if (retries == 4) {
fc->ac.error = -ESTALE;
_leave(" = f [stale]");
return false;
}
_debug("retry");
goto retry;
}
/*
* Get an update for a server's address list.
*/
static noinline bool afs_update_server_record(struct afs_fs_cursor *fc, struct afs_server *server)
{
struct afs_addr_list *alist, *discard;
_enter("");
alist = afs_vl_lookup_addrs(fc->vnode->volume->cell, fc->key,
&server->uuid);
if (IS_ERR(alist)) {
fc->ac.error = PTR_ERR(alist);
_leave(" = f [%d]", fc->ac.error);
return false;
}
discard = alist;
if (server->addr_version != alist->version) {
write_lock(&server->fs_lock);
discard = rcu_dereference_protected(server->addresses,
lockdep_is_held(&server->fs_lock));
rcu_assign_pointer(server->addresses, alist);
server->addr_version = alist->version;
write_unlock(&server->fs_lock);
}
server->update_at = ktime_get_real_seconds() + afs_server_update_delay;
afs_put_addrlist(discard);
_leave(" = t");
return true;
}
/*
* See if a server's address list needs updating.
*/
bool afs_check_server_record(struct afs_fs_cursor *fc, struct afs_server *server)
{
time64_t now = ktime_get_real_seconds();
long diff;
bool success;
int ret, retries = 0;
_enter("");
ASSERT(server);
retry:
diff = READ_ONCE(server->update_at) - now;
if (diff > 0) {
_leave(" = t [not now %ld]", diff);
return true;
}
if (!test_and_set_bit_lock(AFS_SERVER_FL_UPDATING, &server->flags)) {
success = afs_update_server_record(fc, server);
clear_bit_unlock(AFS_SERVER_FL_UPDATING, &server->flags);
wake_up_bit(&server->flags, AFS_SERVER_FL_UPDATING);
_leave(" = %d", success);
return success;
}
ret = wait_on_bit(&server->flags, AFS_SERVER_FL_UPDATING,
TASK_INTERRUPTIBLE);
if (ret == -ERESTARTSYS) {
fc->ac.error = ret;
_leave(" = f [intr]");
return false;
}
retries++;
if (retries == 4) {
_leave(" = f [stale]");
ret = -ESTALE;
return false;
}
goto retry;
}

153
fs/afs/server_list.c Normal file
View File

@ -0,0 +1,153 @@
/* AFS fileserver list management.
*
* Copyright (C) 2017 Red Hat, Inc. All Rights Reserved.
* Written by David Howells (dhowells@redhat.com)
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License
* as published by the Free Software Foundation; either version
* 2 of the License, or (at your option) any later version.
*/
#include <linux/kernel.h>
#include <linux/slab.h>
#include "internal.h"
void afs_put_serverlist(struct afs_net *net, struct afs_server_list *slist)
{
int i;
if (refcount_dec_and_test(&slist->usage)) {
for (i = 0; i < slist->nr_servers; i++) {
afs_put_cb_interest(net, slist->servers[i].cb_interest);
afs_put_server(net, slist->servers[i].server);
}
kfree(slist);
}
}
/*
* Build a server list from a VLDB record.
*/
struct afs_server_list *afs_alloc_server_list(struct afs_cell *cell,
struct key *key,
struct afs_vldb_entry *vldb,
u8 type_mask)
{
struct afs_server_list *slist;
struct afs_server *server;
int ret = -ENOMEM, nr_servers = 0, i, j;
for (i = 0; i < vldb->nr_servers; i++)
if (vldb->fs_mask[i] & type_mask)
nr_servers++;
slist = kzalloc(sizeof(struct afs_server_list) +
sizeof(struct afs_server_entry) * nr_servers,
GFP_KERNEL);
if (!slist)
goto error;
refcount_set(&slist->usage, 1);
/* Make sure a records exists for each server in the list. */
for (i = 0; i < vldb->nr_servers; i++) {
if (!(vldb->fs_mask[i] & type_mask))
continue;
server = afs_lookup_server(cell, key, &vldb->fs_server[i]);
if (IS_ERR(server)) {
ret = PTR_ERR(server);
if (ret == -ENOENT)
continue;
goto error_2;
}
/* Insertion-sort by server pointer */
for (j = 0; j < slist->nr_servers; j++)
if (slist->servers[j].server >= server)
break;
if (j < slist->nr_servers) {
if (slist->servers[j].server == server) {
afs_put_server(cell->net, server);
continue;
}
memmove(slist->servers + j + 1,
slist->servers + j,
(slist->nr_servers - j) * sizeof(struct afs_server_entry));
}
slist->servers[j].server = server;
slist->nr_servers++;
}
if (slist->nr_servers == 0) {
ret = -EDESTADDRREQ;
goto error_2;
}
return slist;
error_2:
afs_put_serverlist(cell->net, slist);
error:
return ERR_PTR(ret);
}
/*
* Copy the annotations from an old server list to its potential replacement.
*/
bool afs_annotate_server_list(struct afs_server_list *new,
struct afs_server_list *old)
{
struct afs_server *cur;
int i, j;
if (old->nr_servers != new->nr_servers)
goto changed;
for (i = 0; i < old->nr_servers; i++)
if (old->servers[i].server != new->servers[i].server)
goto changed;
return false;
changed:
/* Maintain the same current server as before if possible. */
cur = old->servers[old->index].server;
for (j = 0; j < new->nr_servers; j++) {
if (new->servers[j].server == cur) {
new->index = j;
break;
}
}
/* Keep the old callback interest records where possible so that we
* maintain callback interception.
*/
i = 0;
j = 0;
while (i < old->nr_servers && j < new->nr_servers) {
if (new->servers[j].server == old->servers[i].server) {
struct afs_cb_interest *cbi = old->servers[i].cb_interest;
if (cbi) {
new->servers[j].cb_interest = cbi;
refcount_inc(&cbi->usage);
}
i++;
j++;
continue;
}
if (new->servers[j].server < old->servers[i].server) {
j++;
continue;
}
i++;
continue;
}
return true;
}

190
fs/afs/super.c
View File

@ -25,11 +25,10 @@
#include <linux/statfs.h>
#include <linux/sched.h>
#include <linux/nsproxy.h>
#include <linux/magic.h>
#include <net/net_namespace.h>
#include "internal.h"
#define AFS_FS_MAGIC 0x6B414653 /* 'kAFS' */
static void afs_i_init_once(void *foo);
static struct dentry *afs_mount(struct file_system_type *fs_type,
int flags, const char *dev_name, void *data);
@ -143,9 +142,9 @@ void __exit afs_fs_exit(void)
*/
static int afs_show_devname(struct seq_file *m, struct dentry *root)
{
struct afs_super_info *as = root->d_sb->s_fs_info;
struct afs_super_info *as = AFS_FS_S(root->d_sb);
struct afs_volume *volume = as->volume;
struct afs_cell *cell = volume->cell;
struct afs_cell *cell = as->cell;
const char *suf = "";
char pref = '%';
@ -163,7 +162,7 @@ static int afs_show_devname(struct seq_file *m, struct dentry *root)
break;
}
seq_printf(m, "%c%s:%s%s", pref, cell->name, volume->vlocation->vldb.name, suf);
seq_printf(m, "%c%s:%s%s", pref, cell->name, volume->name, suf);
return 0;
}
@ -201,12 +200,14 @@ static int afs_parse_options(struct afs_mount_params *params,
token = match_token(p, afs_options_list, args);
switch (token) {
case afs_opt_cell:
cell = afs_cell_lookup(args[0].from,
args[0].to - args[0].from,
false);
rcu_read_lock();
cell = afs_lookup_cell_rcu(params->net,
args[0].from,
args[0].to - args[0].from);
rcu_read_unlock();
if (IS_ERR(cell))
return PTR_ERR(cell);
afs_put_cell(params->cell);
afs_put_cell(params->net, params->cell);
params->cell = cell;
break;
@ -308,13 +309,14 @@ static int afs_parse_device_name(struct afs_mount_params *params,
/* lookup the cell record */
if (cellname || !params->cell) {
cell = afs_cell_lookup(cellname, cellnamesz, true);
cell = afs_lookup_cell(params->net, cellname, cellnamesz,
NULL, false);
if (IS_ERR(cell)) {
printk(KERN_ERR "kAFS: unable to lookup cell '%*.*s'\n",
cellnamesz, cellnamesz, cellname ?: "");
return PTR_ERR(cell);
}
afs_put_cell(params->cell);
afs_put_cell(params->net, params->cell);
params->cell = cell;
}
@ -332,14 +334,16 @@ static int afs_parse_device_name(struct afs_mount_params *params,
static int afs_test_super(struct super_block *sb, void *data)
{
struct afs_super_info *as1 = data;
struct afs_super_info *as = sb->s_fs_info;
struct afs_super_info *as = AFS_FS_S(sb);
return as->volume == as1->volume;
return as->net == as1->net && as->volume->vid == as1->volume->vid;
}
static int afs_set_super(struct super_block *sb, void *data)
{
sb->s_fs_info = data;
struct afs_super_info *as = data;
sb->s_fs_info = as;
return set_anon_super(sb, NULL);
}
@ -349,7 +353,7 @@ static int afs_set_super(struct super_block *sb, void *data)
static int afs_fill_super(struct super_block *sb,
struct afs_mount_params *params)
{
struct afs_super_info *as = sb->s_fs_info;
struct afs_super_info *as = AFS_FS_S(sb);
struct afs_fid fid;
struct inode *inode = NULL;
int ret;
@ -366,13 +370,15 @@ static int afs_fill_super(struct super_block *sb,
if (ret)
return ret;
sb->s_bdi->ra_pages = VM_MAX_READAHEAD * 1024 / PAGE_SIZE;
strlcpy(sb->s_id, as->volume->vlocation->vldb.name, sizeof(sb->s_id));
sprintf(sb->s_id, "%u", as->volume->vid);
afs_activate_volume(as->volume);
/* allocate the root inode and dentry */
fid.vid = as->volume->vid;
fid.vnode = 1;
fid.unique = 1;
inode = afs_iget(sb, params->key, &fid, NULL, NULL);
inode = afs_iget(sb, params->key, &fid, NULL, NULL, NULL);
if (IS_ERR(inode))
return PTR_ERR(inode);
@ -394,23 +400,45 @@ error:
return ret;
}
static struct afs_super_info *afs_alloc_sbi(struct afs_mount_params *params)
{
struct afs_super_info *as;
as = kzalloc(sizeof(struct afs_super_info), GFP_KERNEL);
if (as) {
as->net = afs_get_net(params->net);
as->cell = afs_get_cell(params->cell);
}
return as;
}
static void afs_destroy_sbi(struct afs_super_info *as)
{
if (as) {
afs_put_volume(as->cell, as->volume);
afs_put_cell(as->net, as->cell);
afs_put_net(as->net);
kfree(as);
}
}
/*
* get an AFS superblock
*/
static struct dentry *afs_mount(struct file_system_type *fs_type,
int flags, const char *dev_name, void *options)
int flags, const char *dev_name, void *options)
{
struct afs_mount_params params;
struct super_block *sb;
struct afs_volume *vol;
struct afs_volume *candidate;
struct key *key;
char *new_opts = kstrdup(options, GFP_KERNEL);
struct afs_super_info *as;
int ret;
_enter(",,%s,%p", dev_name, options);
memset(&params, 0, sizeof(params));
params.net = &__afs_net;
ret = -EINVAL;
if (current->nsproxy->net_ns != &init_net)
@ -436,66 +464,75 @@ static struct dentry *afs_mount(struct file_system_type *fs_type,
}
params.key = key;
/* parse the device name */
vol = afs_volume_lookup(&params);
if (IS_ERR(vol)) {
ret = PTR_ERR(vol);
goto error;
/* allocate a superblock info record */
ret = -ENOMEM;
as = afs_alloc_sbi(&params);
if (!as)
goto error_key;
/* Assume we're going to need a volume record; at the very least we can
* use it to update the volume record if we have one already. This
* checks that the volume exists within the cell.
*/
candidate = afs_create_volume(&params);
if (IS_ERR(candidate)) {
ret = PTR_ERR(candidate);
goto error_as;
}
/* allocate a superblock info record */
as = kzalloc(sizeof(struct afs_super_info), GFP_KERNEL);
if (!as) {
ret = -ENOMEM;
afs_put_volume(vol);
goto error;
}
as->volume = vol;
as->volume = candidate;
/* allocate a deviceless superblock */
sb = sget(fs_type, afs_test_super, afs_set_super, flags, as);
if (IS_ERR(sb)) {
ret = PTR_ERR(sb);
afs_put_volume(vol);
kfree(as);
goto error;
goto error_as;
}
if (!sb->s_root) {
/* initial superblock/root creation */
_debug("create");
ret = afs_fill_super(sb, &params);
if (ret < 0) {
deactivate_locked_super(sb);
goto error;
}
if (ret < 0)
goto error_sb;
as = NULL;
sb->s_flags |= MS_ACTIVE;
} else {
_debug("reuse");
ASSERTCMP(sb->s_flags, &, MS_ACTIVE);
afs_put_volume(vol);
kfree(as);
afs_destroy_sbi(as);
as = NULL;
}
afs_put_cell(params.cell);
kfree(new_opts);
afs_put_cell(params.net, params.cell);
key_put(params.key);
_leave(" = 0 [%p]", sb);
return dget(sb->s_root);
error:
afs_put_cell(params.cell);
error_sb:
deactivate_locked_super(sb);
goto error_key;
error_as:
afs_destroy_sbi(as);
error_key:
key_put(params.key);
kfree(new_opts);
error:
afs_put_cell(params.net, params.cell);
_leave(" = %d", ret);
return ERR_PTR(ret);
}
static void afs_kill_super(struct super_block *sb)
{
struct afs_super_info *as = sb->s_fs_info;
struct afs_super_info *as = AFS_FS_S(sb);
/* Clear the callback interests (which will do ilookup5) before
* deactivating the superblock.
*/
afs_clear_callback_interests(as->net, as->volume->servers);
kill_anon_super(sb);
afs_put_volume(as->volume);
kfree(as);
afs_deactivate_volume(as->volume);
afs_destroy_sbi(as);
}
/*
@ -507,16 +544,15 @@ static void afs_i_init_once(void *_vnode)
memset(vnode, 0, sizeof(*vnode));
inode_init_once(&vnode->vfs_inode);
init_waitqueue_head(&vnode->update_waitq);
mutex_init(&vnode->permits_lock);
mutex_init(&vnode->io_lock);
mutex_init(&vnode->validate_lock);
spin_lock_init(&vnode->writeback_lock);
spin_lock_init(&vnode->wb_lock);
spin_lock_init(&vnode->lock);
INIT_LIST_HEAD(&vnode->writebacks);
INIT_LIST_HEAD(&vnode->wb_keys);
INIT_LIST_HEAD(&vnode->pending_locks);
INIT_LIST_HEAD(&vnode->granted_locks);
INIT_DELAYED_WORK(&vnode->lock_work, afs_lock_work);
INIT_WORK(&vnode->cb_broken_work, afs_broken_callback_work);
seqlock_init(&vnode->cb_lock);
}
/*
@ -536,9 +572,7 @@ static struct inode *afs_alloc_inode(struct super_block *sb)
memset(&vnode->status, 0, sizeof(vnode->status));
vnode->volume = NULL;
vnode->update_cnt = 0;
vnode->flags = 1 << AFS_VNODE_UNSET;
vnode->cb_promised = false;
_leave(" = %p", &vnode->vfs_inode);
return &vnode->vfs_inode;
@ -562,7 +596,7 @@ static void afs_destroy_inode(struct inode *inode)
_debug("DESTROY INODE %p", inode);
ASSERTCMP(vnode->server, ==, NULL);
ASSERTCMP(vnode->cb_interest, ==, NULL);
call_rcu(&inode->i_rcu, afs_i_callback);
atomic_dec(&afs_count_active_inodes);
@ -573,6 +607,7 @@ static void afs_destroy_inode(struct inode *inode)
*/
static int afs_statfs(struct dentry *dentry, struct kstatfs *buf)
{
struct afs_fs_cursor fc;
struct afs_volume_status vs;
struct afs_vnode *vnode = AFS_FS_I(d_inode(dentry));
struct key *key;
@ -582,21 +617,32 @@ static int afs_statfs(struct dentry *dentry, struct kstatfs *buf)
if (IS_ERR(key))
return PTR_ERR(key);
ret = afs_vnode_get_volume_status(vnode, key, &vs);
key_put(key);
if (ret < 0) {
_leave(" = %d", ret);
return ret;
ret = -ERESTARTSYS;
if (afs_begin_vnode_operation(&fc, vnode, key)) {
fc.flags |= AFS_FS_CURSOR_NO_VSLEEP;
while (afs_select_fileserver(&fc)) {
fc.cb_break = vnode->cb_break + vnode->cb_s_break;
afs_fs_get_volume_status(&fc, &vs);
}
afs_check_for_remote_deletion(&fc, fc.vnode);
afs_vnode_commit_status(&fc, vnode, fc.cb_break);
ret = afs_end_vnode_operation(&fc);
}
buf->f_type = dentry->d_sb->s_magic;
buf->f_bsize = AFS_BLOCK_SIZE;
buf->f_namelen = AFSNAMEMAX - 1;
key_put(key);
if (vs.max_quota == 0)
buf->f_blocks = vs.part_max_blocks;
else
buf->f_blocks = vs.max_quota;
buf->f_bavail = buf->f_bfree = buf->f_blocks - vs.blocks_in_use;
return 0;
if (ret == 0) {
buf->f_type = dentry->d_sb->s_magic;
buf->f_bsize = AFS_BLOCK_SIZE;
buf->f_namelen = AFSNAMEMAX - 1;
if (vs.max_quota == 0)
buf->f_blocks = vs.part_max_blocks;
else
buf->f_blocks = vs.max_quota;
buf->f_bavail = buf->f_bfree = buf->f_blocks - vs.blocks_in_use;
}
return ret;
}

732
fs/afs/vlclient.c
View File

@ -12,58 +12,19 @@
#include <linux/gfp.h>
#include <linux/init.h>
#include <linux/sched.h>
#include "afs_fs.h"
#include "internal.h"
/*
* map volume locator abort codes to error codes
* Deliver reply data to a VL.GetEntryByNameU call.
*/
static int afs_vl_abort_to_error(u32 abort_code)
static int afs_deliver_vl_get_entry_by_name_u(struct afs_call *call)
{
_enter("%u", abort_code);
switch (abort_code) {
case AFSVL_IDEXIST: return -EEXIST;
case AFSVL_IO: return -EREMOTEIO;
case AFSVL_NAMEEXIST: return -EEXIST;
case AFSVL_CREATEFAIL: return -EREMOTEIO;
case AFSVL_NOENT: return -ENOMEDIUM;
case AFSVL_EMPTY: return -ENOMEDIUM;
case AFSVL_ENTDELETED: return -ENOMEDIUM;
case AFSVL_BADNAME: return -EINVAL;
case AFSVL_BADINDEX: return -EINVAL;
case AFSVL_BADVOLTYPE: return -EINVAL;
case AFSVL_BADSERVER: return -EINVAL;
case AFSVL_BADPARTITION: return -EINVAL;
case AFSVL_REPSFULL: return -EFBIG;
case AFSVL_NOREPSERVER: return -ENOENT;
case AFSVL_DUPREPSERVER: return -EEXIST;
case AFSVL_RWNOTFOUND: return -ENOENT;
case AFSVL_BADREFCOUNT: return -EINVAL;
case AFSVL_SIZEEXCEEDED: return -EINVAL;
case AFSVL_BADENTRY: return -EINVAL;
case AFSVL_BADVOLIDBUMP: return -EINVAL;
case AFSVL_IDALREADYHASHED: return -EINVAL;
case AFSVL_ENTRYLOCKED: return -EBUSY;
case AFSVL_BADVOLOPER: return -EBADRQC;
case AFSVL_BADRELLOCKTYPE: return -EINVAL;
case AFSVL_RERELEASE: return -EREMOTEIO;
case AFSVL_BADSERVERFLAG: return -EINVAL;
case AFSVL_PERM: return -EACCES;
case AFSVL_NOMEM: return -EREMOTEIO;
default:
return afs_abort_to_error(abort_code);
}
}
/*
* deliver reply data to a VL.GetEntryByXXX call
*/
static int afs_deliver_vl_get_entry_by_xxx(struct afs_call *call)
{
struct afs_cache_vlocation *entry;
__be32 *bp;
struct afs_uvldbentry__xdr *uvldb;
struct afs_vldb_entry *entry;
bool new_only = false;
u32 tmp;
int loop, ret;
int i, ret;
_enter("");
@ -72,144 +33,613 @@ static int afs_deliver_vl_get_entry_by_xxx(struct afs_call *call)
return ret;
/* unmarshall the reply once we've received all of it */
entry = call->reply;
bp = call->buffer;
uvldb = call->buffer;
entry = call->reply[0];
for (loop = 0; loop < 64; loop++)
entry->name[loop] = ntohl(*bp++);
entry->name[loop] = 0;
bp++; /* final NUL */
for (i = 0; i < ARRAY_SIZE(uvldb->name) - 1; i++)
entry->name[i] = (u8)ntohl(uvldb->name[i]);
entry->name[i] = 0;
entry->name_len = strlen(entry->name);
bp++; /* type */
entry->nservers = ntohl(*bp++);
for (loop = 0; loop < 8; loop++)
entry->servers[loop].s_addr = *bp++;
bp += 8; /* partition IDs */
for (loop = 0; loop < 8; loop++) {
tmp = ntohl(*bp++);
entry->srvtmask[loop] = 0;
if (tmp & AFS_VLSF_RWVOL)
entry->srvtmask[loop] |= AFS_VOL_VTM_RW;
if (tmp & AFS_VLSF_ROVOL)
entry->srvtmask[loop] |= AFS_VOL_VTM_RO;
if (tmp & AFS_VLSF_BACKVOL)
entry->srvtmask[loop] |= AFS_VOL_VTM_BAK;
/* If there is a new replication site that we can use, ignore all the
* sites that aren't marked as new.
*/
for (i = 0; i < AFS_NMAXNSERVERS; i++) {
tmp = ntohl(uvldb->serverFlags[i]);
if (!(tmp & AFS_VLSF_DONTUSE) &&
(tmp & AFS_VLSF_NEWREPSITE))
new_only = true;
}
entry->vid[0] = ntohl(*bp++);
entry->vid[1] = ntohl(*bp++);
entry->vid[2] = ntohl(*bp++);
for (i = 0; i < AFS_NMAXNSERVERS; i++) {
struct afs_uuid__xdr *xdr;
struct afs_uuid *uuid;
int j;
bp++; /* clone ID */
tmp = ntohl(uvldb->serverFlags[i]);
if (tmp & AFS_VLSF_DONTUSE ||
(new_only && !(tmp & AFS_VLSF_NEWREPSITE)))
continue;
if (tmp & AFS_VLSF_RWVOL)
entry->fs_mask[i] |= AFS_VOL_VTM_RW;
if (tmp & AFS_VLSF_ROVOL)
entry->fs_mask[i] |= AFS_VOL_VTM_RO;
if (tmp & AFS_VLSF_BACKVOL)
entry->fs_mask[i] |= AFS_VOL_VTM_BAK;
if (!entry->fs_mask[i])
continue;
tmp = ntohl(*bp++); /* flags */
entry->vidmask = 0;
xdr = &uvldb->serverNumber[i];
uuid = (struct afs_uuid *)&entry->fs_server[i];
uuid->time_low = xdr->time_low;
uuid->time_mid = htons(ntohl(xdr->time_mid));
uuid->time_hi_and_version = htons(ntohl(xdr->time_hi_and_version));
uuid->clock_seq_hi_and_reserved = (u8)ntohl(xdr->clock_seq_hi_and_reserved);
uuid->clock_seq_low = (u8)ntohl(xdr->clock_seq_low);
for (j = 0; j < 6; j++)
uuid->node[j] = (u8)ntohl(xdr->node[j]);
entry->nr_servers++;
}
for (i = 0; i < AFS_MAXTYPES; i++)
entry->vid[i] = ntohl(uvldb->volumeId[i]);
tmp = ntohl(uvldb->flags);
if (tmp & AFS_VLF_RWEXISTS)
entry->vidmask |= AFS_VOL_VTM_RW;
__set_bit(AFS_VLDB_HAS_RW, &entry->flags);
if (tmp & AFS_VLF_ROEXISTS)
entry->vidmask |= AFS_VOL_VTM_RO;
__set_bit(AFS_VLDB_HAS_RO, &entry->flags);
if (tmp & AFS_VLF_BACKEXISTS)
entry->vidmask |= AFS_VOL_VTM_BAK;
if (!entry->vidmask)
return -EBADMSG;
__set_bit(AFS_VLDB_HAS_BAK, &entry->flags);
if (!(tmp & (AFS_VLF_RWEXISTS | AFS_VLF_ROEXISTS | AFS_VLF_BACKEXISTS))) {
entry->error = -ENOMEDIUM;
__set_bit(AFS_VLDB_QUERY_ERROR, &entry->flags);
}
__set_bit(AFS_VLDB_QUERY_VALID, &entry->flags);
_leave(" = 0 [done]");
return 0;
}
static void afs_destroy_vl_get_entry_by_name_u(struct afs_call *call)
{
kfree(call->reply[0]);
afs_flat_call_destructor(call);
}
/*
* VL.GetEntryByNameU operation type.
*/
static const struct afs_call_type afs_RXVLGetEntryByNameU = {
.name = "VL.GetEntryByNameU",
.op = afs_VL_GetEntryByNameU,
.deliver = afs_deliver_vl_get_entry_by_name_u,
.destructor = afs_destroy_vl_get_entry_by_name_u,
};
/*
* Dispatch a get volume entry by name or ID operation (uuid variant). If the
* volname is a decimal number then it's a volume ID not a volume name.
*/
struct afs_vldb_entry *afs_vl_get_entry_by_name_u(struct afs_net *net,
struct afs_addr_cursor *ac,
struct key *key,
const char *volname,
int volnamesz)
{
struct afs_vldb_entry *entry;
struct afs_call *call;
size_t reqsz, padsz;
__be32 *bp;
_enter("");
padsz = (4 - (volnamesz & 3)) & 3;
reqsz = 8 + volnamesz + padsz;
entry = kzalloc(sizeof(struct afs_vldb_entry), GFP_KERNEL);
if (!entry)
return ERR_PTR(-ENOMEM);
call = afs_alloc_flat_call(net, &afs_RXVLGetEntryByNameU, reqsz,
sizeof(struct afs_uvldbentry__xdr));
if (!call) {
kfree(entry);
return ERR_PTR(-ENOMEM);
}
call->key = key;
call->reply[0] = entry;
call->ret_reply0 = true;
/* Marshall the parameters */
bp = call->request;
*bp++ = htonl(VLGETENTRYBYNAMEU);
*bp++ = htonl(volnamesz);
memcpy(bp, volname, volnamesz);
if (padsz > 0)
memset((void *)bp + volnamesz, 0, padsz);
trace_afs_make_vl_call(call);
return (struct afs_vldb_entry *)afs_make_call(ac, call, GFP_KERNEL, false);
}
/*
* Deliver reply data to a VL.GetAddrsU call.
*
* GetAddrsU(IN ListAddrByAttributes *inaddr,
* OUT afsUUID *uuidp1,
* OUT uint32_t *uniquifier,
* OUT uint32_t *nentries,
* OUT bulkaddrs *blkaddrs);
*/
static int afs_deliver_vl_get_addrs_u(struct afs_call *call)
{
struct afs_addr_list *alist;
__be32 *bp;
u32 uniquifier, nentries, count;
int i, ret;
_enter("{%u,%zu/%u}", call->unmarshall, call->offset, call->count);
again:
switch (call->unmarshall) {
case 0:
call->offset = 0;
call->unmarshall++;
/* Extract the returned uuid, uniquifier, nentries and blkaddrs size */
case 1:
ret = afs_extract_data(call, call->buffer,
sizeof(struct afs_uuid__xdr) + 3 * sizeof(__be32),
true);
if (ret < 0)
return ret;
bp = call->buffer + sizeof(struct afs_uuid__xdr);
uniquifier = ntohl(*bp++);
nentries = ntohl(*bp++);
count = ntohl(*bp);
nentries = min(nentries, count);
alist = afs_alloc_addrlist(nentries, FS_SERVICE, AFS_FS_PORT);
if (!alist)
return -ENOMEM;
alist->version = uniquifier;
call->reply[0] = alist;
call->count = count;
call->count2 = nentries;
call->offset = 0;
call->unmarshall++;
/* Extract entries */
case 2:
count = min(call->count, 4U);
ret = afs_extract_data(call, call->buffer,
count * sizeof(__be32),
call->count > 4);
if (ret < 0)
return ret;
alist = call->reply[0];
bp = call->buffer;
for (i = 0; i < count; i++)
if (alist->nr_addrs < call->count2)
afs_merge_fs_addr4(alist, *bp++, AFS_FS_PORT);
call->count -= count;
if (call->count > 0)
goto again;
call->offset = 0;
call->unmarshall++;
break;
}
_leave(" = 0 [done]");
return 0;
}
static void afs_vl_get_addrs_u_destructor(struct afs_call *call)
{
afs_put_server(call->net, (struct afs_server *)call->reply[0]);
kfree(call->reply[1]);
return afs_flat_call_destructor(call);
}
/*
* VL.GetAddrsU operation type.
*/
static const struct afs_call_type afs_RXVLGetAddrsU = {
.name = "VL.GetAddrsU",
.op = afs_VL_GetAddrsU,
.deliver = afs_deliver_vl_get_addrs_u,
.destructor = afs_vl_get_addrs_u_destructor,
};
/*
* Dispatch an operation to get the addresses for a server, where the server is
* nominated by UUID.
*/
struct afs_addr_list *afs_vl_get_addrs_u(struct afs_net *net,
struct afs_addr_cursor *ac,
struct key *key,
const uuid_t *uuid)
{
struct afs_ListAddrByAttributes__xdr *r;
const struct afs_uuid *u = (const struct afs_uuid *)uuid;
struct afs_call *call;
__be32 *bp;
int i;
_enter("");
call = afs_alloc_flat_call(net, &afs_RXVLGetAddrsU,
sizeof(__be32) + sizeof(struct afs_ListAddrByAttributes__xdr),
sizeof(struct afs_uuid__xdr) + 3 * sizeof(__be32));
if (!call)
return ERR_PTR(-ENOMEM);
call->key = key;
call->reply[0] = NULL;
call->ret_reply0 = true;
/* Marshall the parameters */
bp = call->request;
*bp++ = htonl(VLGETADDRSU);
r = (struct afs_ListAddrByAttributes__xdr *)bp;
r->Mask = htonl(AFS_VLADDR_UUID);
r->ipaddr = 0;
r->index = 0;
r->spare = 0;
r->uuid.time_low = u->time_low;
r->uuid.time_mid = htonl(ntohs(u->time_mid));
r->uuid.time_hi_and_version = htonl(ntohs(u->time_hi_and_version));
r->uuid.clock_seq_hi_and_reserved = htonl(u->clock_seq_hi_and_reserved);
r->uuid.clock_seq_low = htonl(u->clock_seq_low);
for (i = 0; i < 6; i++)
r->uuid.node[i] = ntohl(u->node[i]);
trace_afs_make_vl_call(call);
return (struct afs_addr_list *)afs_make_call(ac, call, GFP_KERNEL, false);
}
/*
* Deliver reply data to an VL.GetCapabilities operation.
*/
static int afs_deliver_vl_get_capabilities(struct afs_call *call)
{
u32 count;
int ret;
_enter("{%u,%zu/%u}", call->unmarshall, call->offset, call->count);
again:
switch (call->unmarshall) {
case 0:
call->offset = 0;
call->unmarshall++;
/* Extract the capabilities word count */
case 1:
ret = afs_extract_data(call, &call->tmp,
1 * sizeof(__be32),
true);
if (ret < 0)
return ret;
count = ntohl(call->tmp);
call->count = count;
call->count2 = count;
call->offset = 0;
call->unmarshall++;
/* Extract capabilities words */
case 2:
count = min(call->count, 16U);
ret = afs_extract_data(call, call->buffer,
count * sizeof(__be32),
call->count > 16);
if (ret < 0)
return ret;
/* TODO: Examine capabilities */
call->count -= count;
if (call->count > 0)
goto again;
call->offset = 0;
call->unmarshall++;
break;
}
call->reply[0] = (void *)(unsigned long)call->service_id;
_leave(" = 0 [done]");
return 0;
}
/*
* VL.GetEntryByName operation type
* VL.GetCapabilities operation type
*/
static const struct afs_call_type afs_RXVLGetEntryByName = {
.name = "VL.GetEntryByName",
.deliver = afs_deliver_vl_get_entry_by_xxx,
.abort_to_error = afs_vl_abort_to_error,
static const struct afs_call_type afs_RXVLGetCapabilities = {
.name = "VL.GetCapabilities",
.op = afs_VL_GetCapabilities,
.deliver = afs_deliver_vl_get_capabilities,
.destructor = afs_flat_call_destructor,
};
/*
* VL.GetEntryById operation type
* Probe a fileserver for the capabilities that it supports. This can
* return up to 196 words.
*
* We use this to probe for service upgrade to determine what the server at the
* other end supports.
*/
static const struct afs_call_type afs_RXVLGetEntryById = {
.name = "VL.GetEntryById",
.deliver = afs_deliver_vl_get_entry_by_xxx,
.abort_to_error = afs_vl_abort_to_error,
.destructor = afs_flat_call_destructor,
};
/*
* dispatch a get volume entry by name operation
*/
int afs_vl_get_entry_by_name(struct in_addr *addr,
struct key *key,
const char *volname,
struct afs_cache_vlocation *entry,
bool async)
{
struct afs_call *call;
size_t volnamesz, reqsz, padsz;
__be32 *bp;
_enter("");
volnamesz = strlen(volname);
padsz = (4 - (volnamesz & 3)) & 3;
reqsz = 8 + volnamesz + padsz;
call = afs_alloc_flat_call(&afs_RXVLGetEntryByName, reqsz, 384);
if (!call)
return -ENOMEM;
call->key = key;
call->reply = entry;
call->service_id = VL_SERVICE;
call->port = htons(AFS_VL_PORT);
/* marshall the parameters */
bp = call->request;
*bp++ = htonl(VLGETENTRYBYNAME);
*bp++ = htonl(volnamesz);
memcpy(bp, volname, volnamesz);
if (padsz > 0)
memset((void *) bp + volnamesz, 0, padsz);
/* initiate the call */
return afs_make_call(addr, call, GFP_KERNEL, async);
}
/*
* dispatch a get volume entry by ID operation
*/
int afs_vl_get_entry_by_id(struct in_addr *addr,
struct key *key,
afs_volid_t volid,
afs_voltype_t voltype,
struct afs_cache_vlocation *entry,
bool async)
int afs_vl_get_capabilities(struct afs_net *net,
struct afs_addr_cursor *ac,
struct key *key)
{
struct afs_call *call;
__be32 *bp;
_enter("");
call = afs_alloc_flat_call(&afs_RXVLGetEntryById, 12, 384);
call = afs_alloc_flat_call(net, &afs_RXVLGetCapabilities, 1 * 4, 16 * 4);
if (!call)
return -ENOMEM;
call->key = key;
call->reply = entry;
call->service_id = VL_SERVICE;
call->port = htons(AFS_VL_PORT);
call->upgrade = true; /* Let's see if this is a YFS server */
call->reply[0] = (void *)VLGETCAPABILITIES;
call->ret_reply0 = true;
/* marshall the parameters */
bp = call->request;
*bp++ = htonl(VLGETENTRYBYID);
*bp++ = htonl(volid);
*bp = htonl(voltype);
*bp++ = htonl(VLGETCAPABILITIES);
/* initiate the call */
return afs_make_call(addr, call, GFP_KERNEL, async);
/* Can't take a ref on server */
trace_afs_make_vl_call(call);
return afs_make_call(ac, call, GFP_KERNEL, false);
}
/*
* Deliver reply data to a YFSVL.GetEndpoints call.
*
* GetEndpoints(IN yfsServerAttributes *attr,
* OUT opr_uuid *uuid,
* OUT afs_int32 *uniquifier,
* OUT endpoints *fsEndpoints,
* OUT endpoints *volEndpoints)
*/
static int afs_deliver_yfsvl_get_endpoints(struct afs_call *call)
{
struct afs_addr_list *alist;
__be32 *bp;
u32 uniquifier, size;
int ret;
_enter("{%u,%zu/%u,%u}", call->unmarshall, call->offset, call->count, call->count2);
again:
switch (call->unmarshall) {
case 0:
call->offset = 0;
call->unmarshall = 1;
/* Extract the returned uuid, uniquifier, fsEndpoints count and
* either the first fsEndpoint type or the volEndpoints
* count if there are no fsEndpoints. */
case 1:
ret = afs_extract_data(call, call->buffer,
sizeof(uuid_t) +
3 * sizeof(__be32),
true);
if (ret < 0)
return ret;
bp = call->buffer + sizeof(uuid_t);
uniquifier = ntohl(*bp++);
call->count = ntohl(*bp++);
call->count2 = ntohl(*bp); /* Type or next count */
if (call->count > YFS_MAXENDPOINTS)
return -EBADMSG;
alist = afs_alloc_addrlist(call->count, FS_SERVICE, AFS_FS_PORT);
if (!alist)
return -ENOMEM;
alist->version = uniquifier;
call->reply[0] = alist;
call->offset = 0;
if (call->count == 0)
goto extract_volendpoints;
call->unmarshall = 2;
/* Extract fsEndpoints[] entries */
case 2:
switch (call->count2) {
case YFS_ENDPOINT_IPV4:
size = sizeof(__be32) * (1 + 1 + 1);
break;
case YFS_ENDPOINT_IPV6:
size = sizeof(__be32) * (1 + 4 + 1);
break;
default:
return -EBADMSG;
}
size += sizeof(__be32);
ret = afs_extract_data(call, call->buffer, size, true);
if (ret < 0)
return ret;
alist = call->reply[0];
bp = call->buffer;
switch (call->count2) {
case YFS_ENDPOINT_IPV4:
if (ntohl(bp[0]) != sizeof(__be32) * 2)
return -EBADMSG;
afs_merge_fs_addr4(alist, bp[1], ntohl(bp[2]));
bp += 3;
break;
case YFS_ENDPOINT_IPV6:
if (ntohl(bp[0]) != sizeof(__be32) * 5)
return -EBADMSG;
afs_merge_fs_addr6(alist, bp + 1, ntohl(bp[5]));
bp += 6;
break;
default:
return -EBADMSG;
}
/* Got either the type of the next entry or the count of
* volEndpoints if no more fsEndpoints.
*/
call->count2 = htonl(*bp++);
call->offset = 0;
call->count--;
if (call->count > 0)
goto again;
extract_volendpoints:
/* Extract the list of volEndpoints. */
call->count = call->count2;
if (!call->count)
goto end;
if (call->count > YFS_MAXENDPOINTS)
return -EBADMSG;
call->unmarshall = 3;
/* Extract the type of volEndpoints[0]. Normally we would
* extract the type of the next endpoint when we extract the
* data of the current one, but this is the first...
*/
case 3:
ret = afs_extract_data(call, call->buffer, sizeof(__be32), true);
if (ret < 0)
return ret;
bp = call->buffer;
call->count2 = htonl(*bp++);
call->offset = 0;
call->unmarshall = 4;
/* Extract volEndpoints[] entries */
case 4:
switch (call->count2) {
case YFS_ENDPOINT_IPV4:
size = sizeof(__be32) * (1 + 1 + 1);
break;
case YFS_ENDPOINT_IPV6:
size = sizeof(__be32) * (1 + 4 + 1);
break;
default:
return -EBADMSG;
}
if (call->count > 1)
size += sizeof(__be32);
ret = afs_extract_data(call, call->buffer, size, true);
if (ret < 0)
return ret;
bp = call->buffer;
switch (call->count2) {
case YFS_ENDPOINT_IPV4:
if (ntohl(bp[0]) != sizeof(__be32) * 2)
return -EBADMSG;
bp += 3;
break;
case YFS_ENDPOINT_IPV6:
if (ntohl(bp[0]) != sizeof(__be32) * 5)
return -EBADMSG;
bp += 6;
break;
default:
return -EBADMSG;
}
/* Got either the type of the next entry or the count of
* volEndpoints if no more fsEndpoints.
*/
call->offset = 0;
call->count--;
if (call->count > 0) {
call->count2 = htonl(*bp++);
goto again;
}
end:
call->unmarshall = 5;
/* Done */
case 5:
ret = afs_extract_data(call, call->buffer, 0, false);
if (ret < 0)
return ret;
call->unmarshall = 6;
case 6:
break;
}
alist = call->reply[0];
/* Start with IPv6 if available. */
if (alist->nr_ipv4 < alist->nr_addrs)
alist->index = alist->nr_ipv4;
_leave(" = 0 [done]");
return 0;
}
/*
* YFSVL.GetEndpoints operation type.
*/
static const struct afs_call_type afs_YFSVLGetEndpoints = {
.name = "YFSVL.GetEndpoints",
.op = afs_YFSVL_GetEndpoints,
.deliver = afs_deliver_yfsvl_get_endpoints,
.destructor = afs_vl_get_addrs_u_destructor,
};
/*
* Dispatch an operation to get the addresses for a server, where the server is
* nominated by UUID.
*/
struct afs_addr_list *afs_yfsvl_get_endpoints(struct afs_net *net,
struct afs_addr_cursor *ac,
struct key *key,
const uuid_t *uuid)
{
struct afs_call *call;
__be32 *bp;
_enter("");
call = afs_alloc_flat_call(net, &afs_YFSVLGetEndpoints,
sizeof(__be32) * 2 + sizeof(*uuid),
sizeof(struct in6_addr) + sizeof(__be32) * 3);
if (!call)
return ERR_PTR(-ENOMEM);
call->key = key;
call->reply[0] = NULL;
call->ret_reply0 = true;
/* Marshall the parameters */
bp = call->request;
*bp++ = htonl(YVLGETENDPOINTS);
*bp++ = htonl(YFS_SERVER_UUID);
memcpy(bp, uuid, sizeof(*uuid)); /* Type opr_uuid */
trace_afs_make_vl_call(call);
return (struct afs_addr_list *)afs_make_call(ac, call, GFP_KERNEL, false);
}

720
fs/afs/vlocation.c
View File

@ -1,720 +0,0 @@
/* AFS volume location management
*
* Copyright (C) 2002, 2007 Red Hat, Inc. All Rights Reserved.
* Written by David Howells (dhowells@redhat.com)
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License
* as published by the Free Software Foundation; either version
* 2 of the License, or (at your option) any later version.
*/
#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/slab.h>
#include <linux/init.h>
#include <linux/sched.h>
#include "internal.h"
static unsigned afs_vlocation_timeout = 10; /* volume location timeout in seconds */
static unsigned afs_vlocation_update_timeout = 10 * 60;
static void afs_vlocation_reaper(struct work_struct *);
static void afs_vlocation_updater(struct work_struct *);
static LIST_HEAD(afs_vlocation_updates);
static LIST_HEAD(afs_vlocation_graveyard);
static DEFINE_SPINLOCK(afs_vlocation_updates_lock);
static DEFINE_SPINLOCK(afs_vlocation_graveyard_lock);
static DECLARE_DELAYED_WORK(afs_vlocation_reap, afs_vlocation_reaper);
static DECLARE_DELAYED_WORK(afs_vlocation_update, afs_vlocation_updater);
static struct workqueue_struct *afs_vlocation_update_worker;
/*
* iterate through the VL servers in a cell until one of them admits knowing
* about the volume in question
*/
static int afs_vlocation_access_vl_by_name(struct afs_vlocation *vl,
struct key *key,
struct afs_cache_vlocation *vldb)
{
struct afs_cell *cell = vl->cell;
struct in_addr addr;
int count, ret;
_enter("%s,%s", cell->name, vl->vldb.name);
down_write(&vl->cell->vl_sem);
ret = -ENOMEDIUM;
for (count = cell->vl_naddrs; count > 0; count--) {
addr = cell->vl_addrs[cell->vl_curr_svix];
_debug("CellServ[%hu]: %08x", cell->vl_curr_svix, addr.s_addr);
/* attempt to access the VL server */
ret = afs_vl_get_entry_by_name(&addr, key, vl->vldb.name, vldb,
false);
switch (ret) {
case 0:
goto out;
case -ENOMEM:
case -ENONET:
case -ENETUNREACH:
case -EHOSTUNREACH:
case -ECONNREFUSED:
if (ret == -ENOMEM || ret == -ENONET)
goto out;
goto rotate;
case -ENOMEDIUM:
case -EKEYREJECTED:
case -EKEYEXPIRED:
goto out;
default:
ret = -EIO;
goto rotate;
}
/* rotate the server records upon lookup failure */
rotate:
cell->vl_curr_svix++;
cell->vl_curr_svix %= cell->vl_naddrs;
}
out:
up_write(&vl->cell->vl_sem);
_leave(" = %d", ret);
return ret;
}
/*
* iterate through the VL servers in a cell until one of them admits knowing
* about the volume in question
*/
static int afs_vlocation_access_vl_by_id(struct afs_vlocation *vl,
struct key *key,
afs_volid_t volid,
afs_voltype_t voltype,
struct afs_cache_vlocation *vldb)
{
struct afs_cell *cell = vl->cell;
struct in_addr addr;
int count, ret;
_enter("%s,%x,%d,", cell->name, volid, voltype);
down_write(&vl->cell->vl_sem);
ret = -ENOMEDIUM;
for (count = cell->vl_naddrs; count > 0; count--) {
addr = cell->vl_addrs[cell->vl_curr_svix];
_debug("CellServ[%hu]: %08x", cell->vl_curr_svix, addr.s_addr);
/* attempt to access the VL server */
ret = afs_vl_get_entry_by_id(&addr, key, volid, voltype, vldb,
false);
switch (ret) {
case 0:
goto out;
case -ENOMEM:
case -ENONET:
case -ENETUNREACH:
case -EHOSTUNREACH:
case -ECONNREFUSED:
if (ret == -ENOMEM || ret == -ENONET)
goto out;
goto rotate;
case -EBUSY:
vl->upd_busy_cnt++;
if (vl->upd_busy_cnt <= 3) {
if (vl->upd_busy_cnt > 1) {
/* second+ BUSY - sleep a little bit */
set_current_state(TASK_UNINTERRUPTIBLE);
schedule_timeout(1);
}
continue;
}
break;
case -ENOMEDIUM:
vl->upd_rej_cnt++;
goto rotate;
default:
ret = -EIO;
goto rotate;
}
/* rotate the server records upon lookup failure */
rotate:
cell->vl_curr_svix++;
cell->vl_curr_svix %= cell->vl_naddrs;
vl->upd_busy_cnt = 0;
}
out:
if (ret < 0 && vl->upd_rej_cnt > 0) {
printk(KERN_NOTICE "kAFS:"
" Active volume no longer valid '%s'\n",
vl->vldb.name);
vl->valid = 0;
ret = -ENOMEDIUM;
}
up_write(&vl->cell->vl_sem);
_leave(" = %d", ret);
return ret;
}
/*
* allocate a volume location record
*/
static struct afs_vlocation *afs_vlocation_alloc(struct afs_cell *cell,
const char *name,
size_t namesz)
{
struct afs_vlocation *vl;
vl = kzalloc(sizeof(struct afs_vlocation), GFP_KERNEL);
if (vl) {
vl->cell = cell;
vl->state = AFS_VL_NEW;
atomic_set(&vl->usage, 1);
INIT_LIST_HEAD(&vl->link);
INIT_LIST_HEAD(&vl->grave);
INIT_LIST_HEAD(&vl->update);
init_waitqueue_head(&vl->waitq);
spin_lock_init(&vl->lock);
memcpy(vl->vldb.name, name, namesz);
}
_leave(" = %p", vl);
return vl;
}
/*
* update record if we found it in the cache
*/
static int afs_vlocation_update_record(struct afs_vlocation *vl,
struct key *key,
struct afs_cache_vlocation *vldb)
{
afs_voltype_t voltype;
afs_volid_t vid;
int ret;
/* try to look up a cached volume in the cell VL databases by ID */
_debug("Locally Cached: %s %02x { %08x(%x) %08x(%x) %08x(%x) }",
vl->vldb.name,
vl->vldb.vidmask,
ntohl(vl->vldb.servers[0].s_addr),
vl->vldb.srvtmask[0],
ntohl(vl->vldb.servers[1].s_addr),
vl->vldb.srvtmask[1],
ntohl(vl->vldb.servers[2].s_addr),
vl->vldb.srvtmask[2]);
_debug("Vids: %08x %08x %08x",
vl->vldb.vid[0],
vl->vldb.vid[1],
vl->vldb.vid[2]);
if (vl->vldb.vidmask & AFS_VOL_VTM_RW) {
vid = vl->vldb.vid[0];
voltype = AFSVL_RWVOL;
} else if (vl->vldb.vidmask & AFS_VOL_VTM_RO) {
vid = vl->vldb.vid[1];
voltype = AFSVL_ROVOL;
} else if (vl->vldb.vidmask & AFS_VOL_VTM_BAK) {
vid = vl->vldb.vid[2];
voltype = AFSVL_BACKVOL;
} else {
BUG();
vid = 0;
voltype = 0;
}
/* contact the server to make sure the volume is still available
* - TODO: need to handle disconnected operation here
*/
ret = afs_vlocation_access_vl_by_id(vl, key, vid, voltype, vldb);
switch (ret) {
/* net error */
default:
printk(KERN_WARNING "kAFS:"
" failed to update volume '%s' (%x) up in '%s': %d\n",
vl->vldb.name, vid, vl->cell->name, ret);
_leave(" = %d", ret);
return ret;
/* pulled from local cache into memory */
case 0:
_leave(" = 0");
return 0;
/* uh oh... looks like the volume got deleted */
case -ENOMEDIUM:
printk(KERN_ERR "kAFS:"
" volume '%s' (%x) does not exist '%s'\n",
vl->vldb.name, vid, vl->cell->name);
/* TODO: make existing record unavailable */
_leave(" = %d", ret);
return ret;
}
}
/*
* apply the update to a VL record
*/
static void afs_vlocation_apply_update(struct afs_vlocation *vl,
struct afs_cache_vlocation *vldb)
{
_debug("Done VL Lookup: %s %02x { %08x(%x) %08x(%x) %08x(%x) }",
vldb->name, vldb->vidmask,
ntohl(vldb->servers[0].s_addr), vldb->srvtmask[0],
ntohl(vldb->servers[1].s_addr), vldb->srvtmask[1],
ntohl(vldb->servers[2].s_addr), vldb->srvtmask[2]);
_debug("Vids: %08x %08x %08x",
vldb->vid[0], vldb->vid[1], vldb->vid[2]);
if (strcmp(vldb->name, vl->vldb.name) != 0)
printk(KERN_NOTICE "kAFS:"
" name of volume '%s' changed to '%s' on server\n",
vl->vldb.name, vldb->name);
vl->vldb = *vldb;
#ifdef CONFIG_AFS_FSCACHE
fscache_update_cookie(vl->cache);
#endif
}
/*
* fill in a volume location record, consulting the cache and the VL server
* both
*/
static int afs_vlocation_fill_in_record(struct afs_vlocation *vl,
struct key *key)
{
struct afs_cache_vlocation vldb;
int ret;
_enter("");
ASSERTCMP(vl->valid, ==, 0);
memset(&vldb, 0, sizeof(vldb));
/* see if we have an in-cache copy (will set vl->valid if there is) */
#ifdef CONFIG_AFS_FSCACHE
vl->cache = fscache_acquire_cookie(vl->cell->cache,
&afs_vlocation_cache_index_def, vl,
true);
#endif
if (vl->valid) {
/* try to update a known volume in the cell VL databases by
* ID as the name may have changed */
_debug("found in cache");
ret = afs_vlocation_update_record(vl, key, &vldb);
} else {
/* try to look up an unknown volume in the cell VL databases by
* name */
ret = afs_vlocation_access_vl_by_name(vl, key, &vldb);
if (ret < 0) {
printk("kAFS: failed to locate '%s' in cell '%s'\n",
vl->vldb.name, vl->cell->name);
return ret;
}
}
afs_vlocation_apply_update(vl, &vldb);
_leave(" = 0");
return 0;
}
/*
* queue a vlocation record for updates
*/
static void afs_vlocation_queue_for_updates(struct afs_vlocation *vl)
{
struct afs_vlocation *xvl;
/* wait at least 10 minutes before updating... */
vl->update_at = ktime_get_real_seconds() +
afs_vlocation_update_timeout;
spin_lock(&afs_vlocation_updates_lock);
if (!list_empty(&afs_vlocation_updates)) {
/* ... but wait at least 1 second more than the newest record
* already queued so that we don't spam the VL server suddenly
* with lots of requests
*/
xvl = list_entry(afs_vlocation_updates.prev,
struct afs_vlocation, update);
if (vl->update_at <= xvl->update_at)
vl->update_at = xvl->update_at + 1;
} else {
queue_delayed_work(afs_vlocation_update_worker,
&afs_vlocation_update,
afs_vlocation_update_timeout * HZ);
}
list_add_tail(&vl->update, &afs_vlocation_updates);
spin_unlock(&afs_vlocation_updates_lock);
}
/*
* lookup volume location
* - iterate through the VL servers in a cell until one of them admits knowing
* about the volume in question
* - lookup in the local cache if not able to find on the VL server
* - insert/update in the local cache if did get a VL response
*/
struct afs_vlocation *afs_vlocation_lookup(struct afs_cell *cell,
struct key *key,
const char *name,
size_t namesz)
{
struct afs_vlocation *vl;
int ret;
_enter("{%s},{%x},%*.*s,%zu",
cell->name, key_serial(key),
(int) namesz, (int) namesz, name, namesz);
if (namesz >= sizeof(vl->vldb.name)) {
_leave(" = -ENAMETOOLONG");
return ERR_PTR(-ENAMETOOLONG);
}
/* see if we have an in-memory copy first */
down_write(&cell->vl_sem);
spin_lock(&cell->vl_lock);
list_for_each_entry(vl, &cell->vl_list, link) {
if (vl->vldb.name[namesz] != '\0')
continue;
if (memcmp(vl->vldb.name, name, namesz) == 0)
goto found_in_memory;
}
spin_unlock(&cell->vl_lock);
/* not in the cell's in-memory lists - create a new record */
vl = afs_vlocation_alloc(cell, name, namesz);
if (!vl) {
up_write(&cell->vl_sem);
return ERR_PTR(-ENOMEM);
}
afs_get_cell(cell);
list_add_tail(&vl->link, &cell->vl_list);
vl->state = AFS_VL_CREATING;
up_write(&cell->vl_sem);
fill_in_record:
ret = afs_vlocation_fill_in_record(vl, key);
if (ret < 0)
goto error_abandon;
spin_lock(&vl->lock);
vl->state = AFS_VL_VALID;
spin_unlock(&vl->lock);
wake_up(&vl->waitq);
/* update volume entry in local cache */
#ifdef CONFIG_AFS_FSCACHE
fscache_update_cookie(vl->cache);
#endif
/* schedule for regular updates */
afs_vlocation_queue_for_updates(vl);
goto success;
found_in_memory:
/* found in memory */
_debug("found in memory");
atomic_inc(&vl->usage);
spin_unlock(&cell->vl_lock);
if (!list_empty(&vl->grave)) {
spin_lock(&afs_vlocation_graveyard_lock);
list_del_init(&vl->grave);
spin_unlock(&afs_vlocation_graveyard_lock);
}
up_write(&cell->vl_sem);
/* see if it was an abandoned record that we might try filling in */
spin_lock(&vl->lock);
while (vl->state != AFS_VL_VALID) {
afs_vlocation_state_t state = vl->state;
_debug("invalid [state %d]", state);
if (state == AFS_VL_NEW || state == AFS_VL_NO_VOLUME) {
vl->state = AFS_VL_CREATING;
spin_unlock(&vl->lock);
goto fill_in_record;
}
/* must now wait for creation or update by someone else to
* complete */
_debug("wait");
spin_unlock(&vl->lock);
ret = wait_event_interruptible(vl->waitq,
vl->state == AFS_VL_NEW ||
vl->state == AFS_VL_VALID ||
vl->state == AFS_VL_NO_VOLUME);
if (ret < 0)
goto error;
spin_lock(&vl->lock);
}
spin_unlock(&vl->lock);
success:
_leave(" = %p", vl);
return vl;
error_abandon:
spin_lock(&vl->lock);
vl->state = AFS_VL_NEW;
spin_unlock(&vl->lock);
wake_up(&vl->waitq);
error:
ASSERT(vl != NULL);
afs_put_vlocation(vl);
_leave(" = %d", ret);
return ERR_PTR(ret);
}
/*
* finish using a volume location record
*/
void afs_put_vlocation(struct afs_vlocation *vl)
{
if (!vl)
return;
_enter("%s", vl->vldb.name);
ASSERTCMP(atomic_read(&vl->usage), >, 0);
if (likely(!atomic_dec_and_test(&vl->usage))) {
_leave("");
return;
}
spin_lock(&afs_vlocation_graveyard_lock);
if (atomic_read(&vl->usage) == 0) {
_debug("buried");
list_move_tail(&vl->grave, &afs_vlocation_graveyard);
vl->time_of_death = ktime_get_real_seconds();
queue_delayed_work(afs_wq, &afs_vlocation_reap,
afs_vlocation_timeout * HZ);
/* suspend updates on this record */
if (!list_empty(&vl->update)) {
spin_lock(&afs_vlocation_updates_lock);
list_del_init(&vl->update);
spin_unlock(&afs_vlocation_updates_lock);
}
}
spin_unlock(&afs_vlocation_graveyard_lock);
_leave(" [killed?]");
}
/*
* destroy a dead volume location record
*/
static void afs_vlocation_destroy(struct afs_vlocation *vl)
{
_enter("%p", vl);
#ifdef CONFIG_AFS_FSCACHE
fscache_relinquish_cookie(vl->cache, 0);
#endif
afs_put_cell(vl->cell);
kfree(vl);
}
/*
* reap dead volume location records
*/
static void afs_vlocation_reaper(struct work_struct *work)
{
LIST_HEAD(corpses);
struct afs_vlocation *vl;
unsigned long delay, expiry;
time64_t now;
_enter("");
now = ktime_get_real_seconds();
spin_lock(&afs_vlocation_graveyard_lock);
while (!list_empty(&afs_vlocation_graveyard)) {
vl = list_entry(afs_vlocation_graveyard.next,
struct afs_vlocation, grave);
_debug("check %p", vl);
/* the queue is ordered most dead first */
expiry = vl->time_of_death + afs_vlocation_timeout;
if (expiry > now) {
delay = (expiry - now) * HZ;
_debug("delay %lu", delay);
mod_delayed_work(afs_wq, &afs_vlocation_reap, delay);
break;
}
spin_lock(&vl->cell->vl_lock);
if (atomic_read(&vl->usage) > 0) {
_debug("no reap");
list_del_init(&vl->grave);
} else {
_debug("reap");
list_move_tail(&vl->grave, &corpses);
list_del_init(&vl->link);
}
spin_unlock(&vl->cell->vl_lock);
}
spin_unlock(&afs_vlocation_graveyard_lock);
/* now reap the corpses we've extracted */
while (!list_empty(&corpses)) {
vl = list_entry(corpses.next, struct afs_vlocation, grave);
list_del(&vl->grave);
afs_vlocation_destroy(vl);
}
_leave("");
}
/*
* initialise the VL update process
*/
int __init afs_vlocation_update_init(void)
{
afs_vlocation_update_worker = alloc_workqueue("kafs_vlupdated",
WQ_MEM_RECLAIM, 0);
return afs_vlocation_update_worker ? 0 : -ENOMEM;
}
/*
* discard all the volume location records for rmmod
*/
void afs_vlocation_purge(void)
{
afs_vlocation_timeout = 0;
spin_lock(&afs_vlocation_updates_lock);
list_del_init(&afs_vlocation_updates);
spin_unlock(&afs_vlocation_updates_lock);
mod_delayed_work(afs_vlocation_update_worker, &afs_vlocation_update, 0);
destroy_workqueue(afs_vlocation_update_worker);
mod_delayed_work(afs_wq, &afs_vlocation_reap, 0);
}
/*
* update a volume location
*/
static void afs_vlocation_updater(struct work_struct *work)
{
struct afs_cache_vlocation vldb;
struct afs_vlocation *vl, *xvl;
time64_t now;
long timeout;
int ret;
_enter("");
now = ktime_get_real_seconds();
/* find a record to update */
spin_lock(&afs_vlocation_updates_lock);
for (;;) {
if (list_empty(&afs_vlocation_updates)) {
spin_unlock(&afs_vlocation_updates_lock);
_leave(" [nothing]");
return;
}
vl = list_entry(afs_vlocation_updates.next,
struct afs_vlocation, update);
if (atomic_read(&vl->usage) > 0)
break;
list_del_init(&vl->update);
}
timeout = vl->update_at - now;
if (timeout > 0) {
queue_delayed_work(afs_vlocation_update_worker,
&afs_vlocation_update, timeout * HZ);
spin_unlock(&afs_vlocation_updates_lock);
_leave(" [nothing]");
return;
}
list_del_init(&vl->update);
atomic_inc(&vl->usage);
spin_unlock(&afs_vlocation_updates_lock);
/* we can now perform the update */
_debug("update %s", vl->vldb.name);
vl->state = AFS_VL_UPDATING;
vl->upd_rej_cnt = 0;
vl->upd_busy_cnt = 0;
ret = afs_vlocation_update_record(vl, NULL, &vldb);
spin_lock(&vl->lock);
switch (ret) {
case 0:
afs_vlocation_apply_update(vl, &vldb);
vl->state = AFS_VL_VALID;
break;
case -ENOMEDIUM:
vl->state = AFS_VL_VOLUME_DELETED;
break;
default:
vl->state = AFS_VL_UNCERTAIN;
break;
}
spin_unlock(&vl->lock);
wake_up(&vl->waitq);
/* and then reschedule */
_debug("reschedule");
vl->update_at = ktime_get_real_seconds() +
afs_vlocation_update_timeout;
spin_lock(&afs_vlocation_updates_lock);
if (!list_empty(&afs_vlocation_updates)) {
/* next update in 10 minutes, but wait at least 1 second more
* than the newest record already queued so that we don't spam
* the VL server suddenly with lots of requests
*/
xvl = list_entry(afs_vlocation_updates.prev,
struct afs_vlocation, update);
if (vl->update_at <= xvl->update_at)
vl->update_at = xvl->update_at + 1;
xvl = list_entry(afs_vlocation_updates.next,
struct afs_vlocation, update);
timeout = xvl->update_at - now;
if (timeout < 0)
timeout = 0;
} else {
timeout = afs_vlocation_update_timeout;
}
ASSERT(list_empty(&vl->update));
list_add_tail(&vl->update, &afs_vlocation_updates);
_debug("timeout %ld", timeout);
queue_delayed_work(afs_vlocation_update_worker,
&afs_vlocation_update, timeout * HZ);
spin_unlock(&afs_vlocation_updates_lock);
afs_put_vlocation(vl);
}

1025
fs/afs/vnode.c
View File

File diff suppressed because it is too large Load Diff

623
fs/afs/volume.c
View File

@ -10,19 +10,167 @@
*/
#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/init.h>
#include <linux/slab.h>
#include <linux/fs.h>
#include <linux/pagemap.h>
#include <linux/sched.h>
#include "internal.h"
static const char *afs_voltypes[] = { "R/W", "R/O", "BAK" };
unsigned __read_mostly afs_volume_gc_delay = 10;
unsigned __read_mostly afs_volume_record_life = 60 * 60;
static const char *const afs_voltypes[] = { "R/W", "R/O", "BAK" };
/*
* lookup a volume by name
* - this can be one of the following:
* Allocate a volume record and load it up from a vldb record.
*/
static struct afs_volume *afs_alloc_volume(struct afs_mount_params *params,
struct afs_vldb_entry *vldb,
unsigned long type_mask)
{
struct afs_server_list *slist;
struct afs_server *server;
struct afs_volume *volume;
int ret = -ENOMEM, nr_servers = 0, i, j;
for (i = 0; i < vldb->nr_servers; i++)
if (vldb->fs_mask[i] & type_mask)
nr_servers++;
volume = kzalloc(sizeof(struct afs_volume), GFP_KERNEL);
if (!volume)
goto error_0;
volume->vid = vldb->vid[params->type];
volume->update_at = ktime_get_real_seconds() + afs_volume_record_life;
volume->cell = afs_get_cell(params->cell);
volume->type = params->type;
volume->type_force = params->force;
volume->name_len = vldb->name_len;
atomic_set(&volume->usage, 1);
INIT_LIST_HEAD(&volume->proc_link);
rwlock_init(&volume->servers_lock);
memcpy(volume->name, vldb->name, vldb->name_len + 1);
slist = afs_alloc_server_list(params->cell, params->key, vldb, type_mask);
if (IS_ERR(slist)) {
ret = PTR_ERR(slist);
goto error_1;
}
refcount_set(&slist->usage, 1);
volume->servers = slist;
/* Make sure a records exists for each server this volume occupies. */
for (i = 0; i < nr_servers; i++) {
if (!(vldb->fs_mask[i] & type_mask))
continue;
server = afs_lookup_server(params->cell, params->key,
&vldb->fs_server[i]);
if (IS_ERR(server)) {
ret = PTR_ERR(server);
if (ret == -ENOENT)
continue;
goto error_2;
}
/* Insertion-sort by server pointer */
for (j = 0; j < slist->nr_servers; j++)
if (slist->servers[j].server >= server)
break;
if (j < slist->nr_servers) {
if (slist->servers[j].server == server) {
afs_put_server(params->net, server);
continue;
}
memmove(slist->servers + j + 1,
slist->servers + j,
(slist->nr_servers - j) * sizeof(struct afs_server_entry));
}
slist->servers[j].server = server;
slist->nr_servers++;
}
if (slist->nr_servers == 0) {
ret = -EDESTADDRREQ;
goto error_2;
}
return volume;
error_2:
afs_put_serverlist(params->net, slist);
error_1:
kfree(volume);
error_0:
return ERR_PTR(ret);
}
/*
* Look up a VLDB record for a volume.
*/
static struct afs_vldb_entry *afs_vl_lookup_vldb(struct afs_cell *cell,
struct key *key,
const char *volname,
size_t volnamesz)
{
struct afs_addr_cursor ac;
struct afs_vldb_entry *vldb;
int ret;
ret = afs_set_vl_cursor(&ac, cell);
if (ret < 0)
return ERR_PTR(ret);
while (afs_iterate_addresses(&ac)) {
if (!test_bit(ac.index, &ac.alist->probed)) {
ret = afs_vl_get_capabilities(cell->net, &ac, key);
switch (ret) {
case VL_SERVICE:
clear_bit(ac.index, &ac.alist->yfs);
set_bit(ac.index, &ac.alist->probed);
ac.addr->srx_service = ret;
break;
case YFS_VL_SERVICE:
set_bit(ac.index, &ac.alist->yfs);
set_bit(ac.index, &ac.alist->probed);
ac.addr->srx_service = ret;
break;
}
}
vldb = afs_vl_get_entry_by_name_u(cell->net, &ac, key,
volname, volnamesz);
switch (ac.error) {
case 0:
afs_end_cursor(&ac);
return vldb;
case -ECONNABORTED:
ac.error = afs_abort_to_error(ac.abort_code);
goto error;
case -ENOMEM:
case -ENONET:
goto error;
case -ENETUNREACH:
case -EHOSTUNREACH:
case -ECONNREFUSED:
break;
default:
ac.error = -EIO;
goto error;
}
}
error:
return ERR_PTR(afs_end_cursor(&ac));
}
/*
* Look up a volume in the VL server and create a candidate volume record for
* it.
*
* The volume name can be one of the following:
* "%[cell:]volume[.]" R/W volume
* "#[cell:]volume[.]" R/O or R/W volume (rwparent=0),
* or R/W (rwparent=1) volume
@ -42,353 +190,218 @@ static const char *afs_voltypes[] = { "R/W", "R/O", "BAK" };
* - Rule 3: If parent volume is R/W, then only mount R/W volume unless
* explicitly told otherwise
*/
struct afs_volume *afs_volume_lookup(struct afs_mount_params *params)
struct afs_volume *afs_create_volume(struct afs_mount_params *params)
{
struct afs_vlocation *vlocation = NULL;
struct afs_volume *volume = NULL;
struct afs_server *server = NULL;
char srvtmask;
int ret, loop;
struct afs_vldb_entry *vldb;
struct afs_volume *volume;
unsigned long type_mask = 1UL << params->type;
_enter("{%*.*s,%d}",
params->volnamesz, params->volnamesz, params->volname, params->rwpath);
vldb = afs_vl_lookup_vldb(params->cell, params->key,
params->volname, params->volnamesz);
if (IS_ERR(vldb))
return ERR_CAST(vldb);
/* lookup the volume location record */
vlocation = afs_vlocation_lookup(params->cell, params->key,
params->volname, params->volnamesz);
if (IS_ERR(vlocation)) {
ret = PTR_ERR(vlocation);
vlocation = NULL;
if (test_bit(AFS_VLDB_QUERY_ERROR, &vldb->flags)) {
volume = ERR_PTR(vldb->error);
goto error;
}
/* make the final decision on the type we want */
ret = -ENOMEDIUM;
if (params->force && !(vlocation->vldb.vidmask & (1 << params->type)))
goto error;
srvtmask = 0;
for (loop = 0; loop < vlocation->vldb.nservers; loop++)
srvtmask |= vlocation->vldb.srvtmask[loop];
/* Make the final decision on the type we want */
volume = ERR_PTR(-ENOMEDIUM);
if (params->force) {
if (!(srvtmask & (1 << params->type)))
if (!(vldb->flags & type_mask))
goto error;
} else if (srvtmask & AFS_VOL_VTM_RO) {
} else if (test_bit(AFS_VLDB_HAS_RO, &vldb->flags)) {
params->type = AFSVL_ROVOL;
} else if (srvtmask & AFS_VOL_VTM_RW) {
} else if (test_bit(AFS_VLDB_HAS_RW, &vldb->flags)) {
params->type = AFSVL_RWVOL;
} else {
goto error;
}
down_write(&params->cell->vl_sem);
type_mask = 1UL << params->type;
volume = afs_alloc_volume(params, vldb, type_mask);
/* is the volume already active? */
if (vlocation->vols[params->type]) {
/* yes - re-use it */
volume = vlocation->vols[params->type];
afs_get_volume(volume);
goto success;
}
/* create a new volume record */
_debug("creating new volume record");
ret = -ENOMEM;
volume = kzalloc(sizeof(struct afs_volume), GFP_KERNEL);
if (!volume)
goto error_up;
atomic_set(&volume->usage, 1);
volume->type = params->type;
volume->type_force = params->force;
volume->cell = params->cell;
volume->vid = vlocation->vldb.vid[params->type];
init_rwsem(&volume->server_sem);
/* look up all the applicable server records */
for (loop = 0; loop < 8; loop++) {
if (vlocation->vldb.srvtmask[loop] & (1 << volume->type)) {
server = afs_lookup_server(
volume->cell, &vlocation->vldb.servers[loop]);
if (IS_ERR(server)) {
ret = PTR_ERR(server);
goto error_discard;
}
volume->servers[volume->nservers] = server;
volume->nservers++;
}
}
/* attach the cache and volume location */
#ifdef CONFIG_AFS_FSCACHE
volume->cache = fscache_acquire_cookie(vlocation->cache,
&afs_volume_cache_index_def,
volume, true);
#endif
afs_get_vlocation(vlocation);
volume->vlocation = vlocation;
vlocation->vols[volume->type] = volume;
success:
_debug("kAFS selected %s volume %08x",
afs_voltypes[volume->type], volume->vid);
up_write(&params->cell->vl_sem);
afs_put_vlocation(vlocation);
_leave(" = %p", volume);
return volume;
/* clean up */
error_up:
up_write(&params->cell->vl_sem);
error:
afs_put_vlocation(vlocation);
_leave(" = %d", ret);
return ERR_PTR(ret);
error_discard:
up_write(&params->cell->vl_sem);
for (loop = volume->nservers - 1; loop >= 0; loop--)
afs_put_server(volume->servers[loop]);
kfree(volume);
goto error;
kfree(vldb);
return volume;
}
/*
* destroy a volume record
* Destroy a volume record
*/
void afs_put_volume(struct afs_volume *volume)
static void afs_destroy_volume(struct afs_net *net, struct afs_volume *volume)
{
struct afs_vlocation *vlocation;
int loop;
if (!volume)
return;
_enter("%p", volume);
ASSERTCMP(atomic_read(&volume->usage), >, 0);
vlocation = volume->vlocation;
/* to prevent a race, the decrement and the dequeue must be effectively
* atomic */
down_write(&vlocation->cell->vl_sem);
if (likely(!atomic_dec_and_test(&volume->usage))) {
up_write(&vlocation->cell->vl_sem);
_leave("");
return;
}
vlocation->vols[volume->type] = NULL;
up_write(&vlocation->cell->vl_sem);
/* finish cleaning up the volume */
#ifdef CONFIG_AFS_FSCACHE
fscache_relinquish_cookie(volume->cache, 0);
ASSERTCMP(volume->cache, ==, NULL);
#endif
afs_put_vlocation(vlocation);
for (loop = volume->nservers - 1; loop >= 0; loop--)
afs_put_server(volume->servers[loop]);
afs_put_serverlist(net, volume->servers);
afs_put_cell(net, volume->cell);
kfree(volume);
_leave(" [destroyed]");
}
/*
* pick a server to use to try accessing this volume
* - returns with an elevated usage count on the server chosen
* Drop a reference on a volume record.
*/
struct afs_server *afs_volume_pick_fileserver(struct afs_vnode *vnode)
void afs_put_volume(struct afs_cell *cell, struct afs_volume *volume)
{
struct afs_volume *volume = vnode->volume;
struct afs_server *server;
int ret, state, loop;
if (volume) {
_enter("%s", volume->name);
_enter("%s", volume->vlocation->vldb.name);
/* stick with the server we're already using if we can */
if (vnode->server && vnode->server->fs_state == 0) {
afs_get_server(vnode->server);
_leave(" = %p [current]", vnode->server);
return vnode->server;
if (atomic_dec_and_test(&volume->usage))
afs_destroy_volume(cell->net, volume);
}
down_read(&volume->server_sem);
/* handle the no-server case */
if (volume->nservers == 0) {
ret = volume->rjservers ? -ENOMEDIUM : -ESTALE;
up_read(&volume->server_sem);
_leave(" = %d [no servers]", ret);
return ERR_PTR(ret);
}
/* basically, just search the list for the first live server and use
* that */
ret = 0;
for (loop = 0; loop < volume->nservers; loop++) {
server = volume->servers[loop];
state = server->fs_state;
_debug("consider %d [%d]", loop, state);
switch (state) {
/* found an apparently healthy server */
case 0:
afs_get_server(server);
up_read(&volume->server_sem);
_leave(" = %p (picked %08x)",
server, ntohl(server->addr.s_addr));
return server;
case -ENETUNREACH:
if (ret == 0)
ret = state;
break;
case -EHOSTUNREACH:
if (ret == 0 ||
ret == -ENETUNREACH)
ret = state;
break;
case -ECONNREFUSED:
if (ret == 0 ||
ret == -ENETUNREACH ||
ret == -EHOSTUNREACH)
ret = state;
break;
default:
case -EREMOTEIO:
if (ret == 0 ||
ret == -ENETUNREACH ||
ret == -EHOSTUNREACH ||
ret == -ECONNREFUSED)
ret = state;
break;
}
}
/* no available servers
* - TODO: handle the no active servers case better
*/
up_read(&volume->server_sem);
_leave(" = %d", ret);
return ERR_PTR(ret);
}
/*
* release a server after use
* - releases the ref on the server struct that was acquired by picking
* - records result of using a particular server to access a volume
* - return 0 to try again, 1 if okay or to issue error
* - the caller must release the server struct if result was 0
* Activate a volume.
*/
int afs_volume_release_fileserver(struct afs_vnode *vnode,
struct afs_server *server,
int result)
void afs_activate_volume(struct afs_volume *volume)
{
struct afs_volume *volume = vnode->volume;
unsigned loop;
#ifdef CONFIG_AFS_FSCACHE
volume->cache = fscache_acquire_cookie(volume->cell->cache,
&afs_volume_cache_index_def,
volume, true);
#endif
_enter("%s,%08x,%d",
volume->vlocation->vldb.name, ntohl(server->addr.s_addr),
result);
write_lock(&volume->cell->proc_lock);
list_add_tail(&volume->proc_link, &volume->cell->proc_volumes);
write_unlock(&volume->cell->proc_lock);
}
switch (result) {
/* success */
case 0:
server->fs_act_jif = jiffies;
server->fs_state = 0;
_leave("");
return 1;
/*
* Deactivate a volume.
*/
void afs_deactivate_volume(struct afs_volume *volume)
{
_enter("%s", volume->name);
/* the fileserver denied all knowledge of the volume */
case -ENOMEDIUM:
server->fs_act_jif = jiffies;
down_write(&volume->server_sem);
write_lock(&volume->cell->proc_lock);
list_del_init(&volume->proc_link);
write_unlock(&volume->cell->proc_lock);
/* firstly, find where the server is in the active list (if it
* is) */
for (loop = 0; loop < volume->nservers; loop++)
if (volume->servers[loop] == server)
goto present;
#ifdef CONFIG_AFS_FSCACHE
fscache_relinquish_cookie(volume->cache,
test_bit(AFS_VOLUME_DELETED, &volume->flags));
volume->cache = NULL;
#endif
/* no longer there - may have been discarded by another op */
goto try_next_server_upw;
_leave("");
}
present:
volume->nservers--;
memmove(&volume->servers[loop],
&volume->servers[loop + 1],
sizeof(volume->servers[loop]) *
(volume->nservers - loop));
volume->servers[volume->nservers] = NULL;
afs_put_server(server);
volume->rjservers++;
/*
* Query the VL service to update the volume status.
*/
static int afs_update_volume_status(struct afs_volume *volume, struct key *key)
{
struct afs_server_list *new, *old, *discard;
struct afs_vldb_entry *vldb;
char idbuf[16];
int ret, idsz;
if (volume->nservers > 0)
/* another server might acknowledge its existence */
goto try_next_server_upw;
_enter("");
/* handle the case where all the fileservers have rejected the
* volume
* - TODO: try asking the fileservers for volume information
* - TODO: contact the VL server again to see if the volume is
* no longer registered
*/
up_write(&volume->server_sem);
afs_put_server(server);
_leave(" [completely rejected]");
return 1;
/* We look up an ID by passing it as a decimal string in the
* operation's name parameter.
*/
idsz = sprintf(idbuf, "%u", volume->vid);
/* problem reaching the server */
case -ENETUNREACH:
case -EHOSTUNREACH:
case -ECONNREFUSED:
case -ETIME:
case -ETIMEDOUT:
case -EREMOTEIO:
/* mark the server as dead
* TODO: vary dead timeout depending on error
*/
spin_lock(&server->fs_lock);
if (!server->fs_state) {
server->fs_dead_jif = jiffies + HZ * 10;
server->fs_state = result;
printk("kAFS: SERVER DEAD state=%d\n", result);
}
spin_unlock(&server->fs_lock);
goto try_next_server;
/* miscellaneous error */
default:
server->fs_act_jif = jiffies;
case -ENOMEM:
case -ENONET:
/* tell the caller to accept the result */
afs_put_server(server);
_leave(" [local failure]");
return 1;
vldb = afs_vl_lookup_vldb(volume->cell, key, idbuf, idsz);
if (IS_ERR(vldb)) {
ret = PTR_ERR(vldb);
goto error;
}
/* tell the caller to loop around and try the next server */
try_next_server_upw:
up_write(&volume->server_sem);
try_next_server:
afs_put_server(server);
_leave(" [try next server]");
return 0;
/* See if the volume got renamed. */
if (vldb->name_len != volume->name_len ||
memcmp(vldb->name, volume->name, vldb->name_len) != 0) {
/* TODO: Use RCU'd string. */
memcpy(volume->name, vldb->name, AFS_MAXVOLNAME);
volume->name_len = vldb->name_len;
}
/* See if the volume's server list got updated. */
new = afs_alloc_server_list(volume->cell, key,
vldb, (1 << volume->type));
if (IS_ERR(new)) {
ret = PTR_ERR(new);
goto error_vldb;
}
write_lock(&volume->servers_lock);
discard = new;
old = volume->servers;
if (afs_annotate_server_list(new, old)) {
new->seq = volume->servers_seq + 1;
volume->servers = new;
smp_wmb();
volume->servers_seq++;
discard = old;
}
volume->update_at = ktime_get_real_seconds() + afs_volume_record_life;
clear_bit(AFS_VOLUME_NEEDS_UPDATE, &volume->flags);
write_unlock(&volume->servers_lock);
ret = 0;
afs_put_serverlist(volume->cell->net, discard);
error_vldb:
kfree(vldb);
error:
_leave(" = %d", ret);
return ret;
}
/*
* Make sure the volume record is up to date.
*/
int afs_check_volume_status(struct afs_volume *volume, struct key *key)
{
time64_t now = ktime_get_real_seconds();
int ret, retries = 0;
_enter("");
if (volume->update_at <= now)
set_bit(AFS_VOLUME_NEEDS_UPDATE, &volume->flags);
retry:
if (!test_bit(AFS_VOLUME_NEEDS_UPDATE, &volume->flags) &&
!test_bit(AFS_VOLUME_WAIT, &volume->flags)) {
_leave(" = 0");
return 0;
}
if (!test_and_set_bit_lock(AFS_VOLUME_UPDATING, &volume->flags)) {
ret = afs_update_volume_status(volume, key);
clear_bit_unlock(AFS_VOLUME_WAIT, &volume->flags);
clear_bit_unlock(AFS_VOLUME_UPDATING, &volume->flags);
wake_up_bit(&volume->flags, AFS_VOLUME_WAIT);
_leave(" = %d", ret);
return ret;
}
if (!test_bit(AFS_VOLUME_WAIT, &volume->flags)) {
_leave(" = 0 [no wait]");
return 0;
}
ret = wait_on_bit(&volume->flags, AFS_VOLUME_WAIT, TASK_INTERRUPTIBLE);
if (ret == -ERESTARTSYS) {
_leave(" = %d", ret);
return ret;
}
retries++;
if (retries == 4) {
_leave(" = -ESTALE");
return -ESTALE;
}
goto retry;
}

709
fs/afs/write.c
View File

@ -8,6 +8,7 @@
* as published by the Free Software Foundation; either version
* 2 of the License, or (at your option) any later version.
*/
#include <linux/backing-dev.h>
#include <linux/slab.h>
#include <linux/fs.h>
@ -16,9 +17,6 @@
#include <linux/pagevec.h>
#include "internal.h"
static int afs_write_back_from_locked_page(struct afs_writeback *wb,
struct page *page);
/*
* mark a page as having been made dirty and thus needing writeback
*/
@ -28,58 +26,6 @@ int afs_set_page_dirty(struct page *page)
return __set_page_dirty_nobuffers(page);
}
/*
* unlink a writeback record because its usage has reached zero
* - must be called with the wb->vnode->writeback_lock held
*/
static void afs_unlink_writeback(struct afs_writeback *wb)
{
struct afs_writeback *front;
struct afs_vnode *vnode = wb->vnode;
list_del_init(&wb->link);
if (!list_empty(&vnode->writebacks)) {
/* if an fsync rises to the front of the queue then wake it
* up */
front = list_entry(vnode->writebacks.next,
struct afs_writeback, link);
if (front->state == AFS_WBACK_SYNCING) {
_debug("wake up sync");
front->state = AFS_WBACK_COMPLETE;
wake_up(&front->waitq);
}
}
}
/*
* free a writeback record
*/
static void afs_free_writeback(struct afs_writeback *wb)
{
_enter("");
key_put(wb->key);
kfree(wb);
}
/*
* dispose of a reference to a writeback record
*/
void afs_put_writeback(struct afs_writeback *wb)
{
struct afs_vnode *vnode = wb->vnode;
_enter("{%d}", wb->usage);
spin_lock(&vnode->writeback_lock);
if (--wb->usage == 0)
afs_unlink_writeback(wb);
else
wb = NULL;
spin_unlock(&vnode->writeback_lock);
if (wb)
afs_free_writeback(wb);
}
/*
* partly or wholly fill a page that's under preparation for writing
*/
@ -103,7 +49,7 @@ static int afs_fill_page(struct afs_vnode *vnode, struct key *key,
req->pages[0] = page;
get_page(page);
ret = afs_vnode_fetch_data(vnode, key, req);
ret = afs_fetch_data(vnode, key, req);
afs_put_read(req);
if (ret < 0) {
if (ret == -ENOENT) {
@ -125,42 +71,32 @@ int afs_write_begin(struct file *file, struct address_space *mapping,
loff_t pos, unsigned len, unsigned flags,
struct page **pagep, void **fsdata)
{
struct afs_writeback *candidate, *wb;
struct afs_vnode *vnode = AFS_FS_I(file_inode(file));
struct page *page;
struct key *key = file->private_data;
unsigned from = pos & (PAGE_SIZE - 1);
unsigned to = from + len;
struct key *key = afs_file_key(file);
unsigned long priv;
unsigned f, from = pos & (PAGE_SIZE - 1);
unsigned t, to = from + len;
pgoff_t index = pos >> PAGE_SHIFT;
int ret;
_enter("{%x:%u},{%lx},%u,%u",
vnode->fid.vid, vnode->fid.vnode, index, from, to);
candidate = kzalloc(sizeof(*candidate), GFP_KERNEL);
if (!candidate)
return -ENOMEM;
candidate->vnode = vnode;
candidate->first = candidate->last = index;
candidate->offset_first = from;
candidate->to_last = to;
INIT_LIST_HEAD(&candidate->link);
candidate->usage = 1;
candidate->state = AFS_WBACK_PENDING;
init_waitqueue_head(&candidate->waitq);
/* We want to store information about how much of a page is altered in
* page->private.
*/
BUILD_BUG_ON(PAGE_SIZE > 32768 && sizeof(page->private) < 8);
page = grab_cache_page_write_begin(mapping, index, flags);
if (!page) {
kfree(candidate);
if (!page)
return -ENOMEM;
}
if (!PageUptodate(page) && len != PAGE_SIZE) {
ret = afs_fill_page(vnode, key, pos & PAGE_MASK, PAGE_SIZE, page);
if (ret < 0) {
unlock_page(page);
put_page(page);
kfree(candidate);
_leave(" = %d [prep]", ret);
return ret;
}
@ -171,79 +107,54 @@ int afs_write_begin(struct file *file, struct address_space *mapping,
*pagep = page;
try_again:
spin_lock(&vnode->writeback_lock);
/* see if this page is already pending a writeback under a suitable key
* - if so we can just join onto that one */
wb = (struct afs_writeback *) page_private(page);
if (wb) {
if (wb->key == key && wb->state == AFS_WBACK_PENDING)
goto subsume_in_current_wb;
goto flush_conflicting_wb;
/* See if this page is already partially written in a way that we can
* merge the new write with.
*/
t = f = 0;
if (PagePrivate(page)) {
priv = page_private(page);
f = priv & AFS_PRIV_MAX;
t = priv >> AFS_PRIV_SHIFT;
ASSERTCMP(f, <=, t);
}
if (index > 0) {
/* see if we can find an already pending writeback that we can
* append this page to */
list_for_each_entry(wb, &vnode->writebacks, link) {
if (wb->last == index - 1 && wb->key == key &&
wb->state == AFS_WBACK_PENDING)
goto append_to_previous_wb;
}
if (f != t) {
if (to < f || from > t)
goto flush_conflicting_write;
if (from < f)
f = from;
if (to > t)
t = to;
} else {
f = from;
t = to;
}
list_add_tail(&candidate->link, &vnode->writebacks);
candidate->key = key_get(key);
spin_unlock(&vnode->writeback_lock);
priv = (unsigned long)t << AFS_PRIV_SHIFT;
priv |= f;
trace_afs_page_dirty(vnode, tracepoint_string("begin"),
page->index, priv);
SetPagePrivate(page);
set_page_private(page, (unsigned long) candidate);
_leave(" = 0 [new]");
set_page_private(page, priv);
_leave(" = 0");
return 0;
subsume_in_current_wb:
_debug("subsume");
ASSERTRANGE(wb->first, <=, index, <=, wb->last);
if (index == wb->first && from < wb->offset_first)
wb->offset_first = from;
if (index == wb->last && to > wb->to_last)
wb->to_last = to;
spin_unlock(&vnode->writeback_lock);
kfree(candidate);
_leave(" = 0 [sub]");
return 0;
append_to_previous_wb:
_debug("append into %lx-%lx", wb->first, wb->last);
wb->usage++;
wb->last++;
wb->to_last = to;
spin_unlock(&vnode->writeback_lock);
SetPagePrivate(page);
set_page_private(page, (unsigned long) wb);
kfree(candidate);
_leave(" = 0 [app]");
return 0;
/* the page is currently bound to another context, so if it's dirty we
* need to flush it before we can use the new context */
flush_conflicting_wb:
/* The previous write and this write aren't adjacent or overlapping, so
* flush the page out.
*/
flush_conflicting_write:
_debug("flush conflict");
if (wb->state == AFS_WBACK_PENDING)
wb->state = AFS_WBACK_CONFLICTING;
spin_unlock(&vnode->writeback_lock);
if (clear_page_dirty_for_io(page)) {
ret = afs_write_back_from_locked_page(wb, page);
if (ret < 0) {
afs_put_writeback(candidate);
_leave(" = %d", ret);
return ret;
}
ret = write_one_page(page);
if (ret < 0) {
_leave(" = %d", ret);
return ret;
}
/* the page holds a ref on the writeback record */
afs_put_writeback(wb);
set_page_private(page, 0);
ClearPagePrivate(page);
ret = lock_page_killable(page);
if (ret < 0) {
_leave(" = %d", ret);
return ret;
}
goto try_again;
}
@ -255,7 +166,7 @@ int afs_write_end(struct file *file, struct address_space *mapping,
struct page *page, void *fsdata)
{
struct afs_vnode *vnode = AFS_FS_I(file_inode(file));
struct key *key = file->private_data;
struct key *key = afs_file_key(file);
loff_t i_size, maybe_i_size;
int ret;
@ -266,11 +177,11 @@ int afs_write_end(struct file *file, struct address_space *mapping,
i_size = i_size_read(&vnode->vfs_inode);
if (maybe_i_size > i_size) {
spin_lock(&vnode->writeback_lock);
spin_lock(&vnode->wb_lock);
i_size = i_size_read(&vnode->vfs_inode);
if (maybe_i_size > i_size)
i_size_write(&vnode->vfs_inode, maybe_i_size);
spin_unlock(&vnode->writeback_lock);
spin_unlock(&vnode->wb_lock);
}
if (!PageUptodate(page)) {
@ -299,9 +210,10 @@ int afs_write_end(struct file *file, struct address_space *mapping,
/*
* kill all the pages in the given range
*/
static void afs_kill_pages(struct afs_vnode *vnode, bool error,
static void afs_kill_pages(struct address_space *mapping,
pgoff_t first, pgoff_t last)
{
struct afs_vnode *vnode = AFS_FS_I(mapping->host);
struct pagevec pv;
unsigned count, loop;
@ -316,37 +228,157 @@ static void afs_kill_pages(struct afs_vnode *vnode, bool error,
count = last - first + 1;
if (count > PAGEVEC_SIZE)
count = PAGEVEC_SIZE;
pv.nr = find_get_pages_contig(vnode->vfs_inode.i_mapping,
first, count, pv.pages);
pv.nr = find_get_pages_contig(mapping, first, count, pv.pages);
ASSERTCMP(pv.nr, ==, count);
for (loop = 0; loop < count; loop++) {
struct page *page = pv.pages[loop];
ClearPageUptodate(page);
if (error)
SetPageError(page);
if (PageWriteback(page))
end_page_writeback(page);
SetPageError(page);
end_page_writeback(page);
if (page->index >= first)
first = page->index + 1;
lock_page(page);
generic_error_remove_page(mapping, page);
}
__pagevec_release(&pv);
} while (first < last);
} while (first <= last);
_leave("");
}
/*
* synchronously write back the locked page and any subsequent non-locked dirty
* pages also covered by the same writeback record
* Redirty all the pages in a given range.
*/
static int afs_write_back_from_locked_page(struct afs_writeback *wb,
struct page *primary_page)
static void afs_redirty_pages(struct writeback_control *wbc,
struct address_space *mapping,
pgoff_t first, pgoff_t last)
{
struct afs_vnode *vnode = AFS_FS_I(mapping->host);
struct pagevec pv;
unsigned count, loop;
_enter("{%x:%u},%lx-%lx",
vnode->fid.vid, vnode->fid.vnode, first, last);
pagevec_init(&pv);
do {
_debug("redirty %lx-%lx", first, last);
count = last - first + 1;
if (count > PAGEVEC_SIZE)
count = PAGEVEC_SIZE;
pv.nr = find_get_pages_contig(mapping, first, count, pv.pages);
ASSERTCMP(pv.nr, ==, count);
for (loop = 0; loop < count; loop++) {
struct page *page = pv.pages[loop];
redirty_page_for_writepage(wbc, page);
end_page_writeback(page);
if (page->index >= first)
first = page->index + 1;
}
__pagevec_release(&pv);
} while (first <= last);
_leave("");
}
/*
* write to a file
*/
static int afs_store_data(struct address_space *mapping,
pgoff_t first, pgoff_t last,
unsigned offset, unsigned to)
{
struct afs_vnode *vnode = AFS_FS_I(mapping->host);
struct afs_fs_cursor fc;
struct afs_wb_key *wbk = NULL;
struct list_head *p;
int ret = -ENOKEY, ret2;
_enter("%s{%x:%u.%u},%lx,%lx,%x,%x",
vnode->volume->name,
vnode->fid.vid,
vnode->fid.vnode,
vnode->fid.unique,
first, last, offset, to);
spin_lock(&vnode->wb_lock);
p = vnode->wb_keys.next;
/* Iterate through the list looking for a valid key to use. */
try_next_key:
while (p != &vnode->wb_keys) {
wbk = list_entry(p, struct afs_wb_key, vnode_link);
_debug("wbk %u", key_serial(wbk->key));
ret2 = key_validate(wbk->key);
if (ret2 == 0)
goto found_key;
if (ret == -ENOKEY)
ret = ret2;
p = p->next;
}
spin_unlock(&vnode->wb_lock);
afs_put_wb_key(wbk);
_leave(" = %d [no keys]", ret);
return ret;
found_key:
refcount_inc(&wbk->usage);
spin_unlock(&vnode->wb_lock);
_debug("USE WB KEY %u", key_serial(wbk->key));
ret = -ERESTARTSYS;
if (afs_begin_vnode_operation(&fc, vnode, wbk->key)) {
while (afs_select_fileserver(&fc)) {
fc.cb_break = vnode->cb_break + vnode->cb_s_break;
afs_fs_store_data(&fc, mapping, first, last, offset, to);
}
afs_check_for_remote_deletion(&fc, fc.vnode);
afs_vnode_commit_status(&fc, vnode, fc.cb_break);
ret = afs_end_vnode_operation(&fc);
}
switch (ret) {
case -EACCES:
case -EPERM:
case -ENOKEY:
case -EKEYEXPIRED:
case -EKEYREJECTED:
case -EKEYREVOKED:
_debug("next");
spin_lock(&vnode->wb_lock);
p = wbk->vnode_link.next;
afs_put_wb_key(wbk);
goto try_next_key;
}
afs_put_wb_key(wbk);
_leave(" = %d", ret);
return ret;
}
/*
* Synchronously write back the locked page and any subsequent non-locked dirty
* pages.
*/
static int afs_write_back_from_locked_page(struct address_space *mapping,
struct writeback_control *wbc,
struct page *primary_page,
pgoff_t final_page)
{
struct afs_vnode *vnode = AFS_FS_I(mapping->host);
struct page *pages[8], *page;
unsigned long count;
unsigned n, offset, to;
unsigned long count, priv;
unsigned n, offset, to, f, t;
pgoff_t start, first, last;
int loop, ret;
@ -356,20 +388,33 @@ static int afs_write_back_from_locked_page(struct afs_writeback *wb,
if (test_set_page_writeback(primary_page))
BUG();
/* find all consecutive lockable dirty pages, stopping when we find a
* page that is not immediately lockable, is not dirty or is missing,
* or we reach the end of the range */
/* Find all consecutive lockable dirty pages that have contiguous
* written regions, stopping when we find a page that is not
* immediately lockable, is not dirty or is missing, or we reach the
* end of the range.
*/
start = primary_page->index;
if (start >= wb->last)
priv = page_private(primary_page);
offset = priv & AFS_PRIV_MAX;
to = priv >> AFS_PRIV_SHIFT;
trace_afs_page_dirty(vnode, tracepoint_string("store"),
primary_page->index, priv);
WARN_ON(offset == to);
if (offset == to)
trace_afs_page_dirty(vnode, tracepoint_string("WARN"),
primary_page->index, priv);
if (start >= final_page || to < PAGE_SIZE)
goto no_more;
start++;
do {
_debug("more %lx [%lx]", start, count);
n = wb->last - start + 1;
n = final_page - start + 1;
if (n > ARRAY_SIZE(pages))
n = ARRAY_SIZE(pages);
n = find_get_pages_contig(wb->vnode->vfs_inode.i_mapping,
start, n, pages);
n = find_get_pages_contig(mapping, start, ARRAY_SIZE(pages), pages);
_debug("fgpc %u", n);
if (n == 0)
goto no_more;
@ -381,16 +426,30 @@ static int afs_write_back_from_locked_page(struct afs_writeback *wb,
}
for (loop = 0; loop < n; loop++) {
if (to != PAGE_SIZE)
break;
page = pages[loop];
if (page->index > wb->last)
if (page->index > final_page)
break;
if (!trylock_page(page))
break;
if (!PageDirty(page) ||
page_private(page) != (unsigned long) wb) {
if (!PageDirty(page) || PageWriteback(page)) {
unlock_page(page);
break;
}
priv = page_private(page);
f = priv & AFS_PRIV_MAX;
t = priv >> AFS_PRIV_SHIFT;
if (f != 0) {
unlock_page(page);
break;
}
to = t;
trace_afs_page_dirty(vnode, tracepoint_string("store+"),
page->index, priv);
if (!clear_page_dirty_for_io(page))
BUG();
if (test_set_page_writeback(page))
@ -406,50 +465,55 @@ static int afs_write_back_from_locked_page(struct afs_writeback *wb,
}
start += loop;
} while (start <= wb->last && count < 65536);
} while (start <= final_page && count < 65536);
no_more:
/* we now have a contiguous set of dirty pages, each with writeback set
* and the dirty mark cleared; the first page is locked and must remain
* so, all the rest are unlocked */
/* We now have a contiguous set of dirty pages, each with writeback
* set; the first page is still locked at this point, but all the rest
* have been unlocked.
*/
unlock_page(primary_page);
first = primary_page->index;
last = first + count - 1;
offset = (first == wb->first) ? wb->offset_first : 0;
to = (last == wb->last) ? wb->to_last : PAGE_SIZE;
_debug("write back %lx[%u..] to %lx[..%u]", first, offset, last, to);
ret = afs_vnode_store_data(wb, first, last, offset, to);
if (ret < 0) {
switch (ret) {
case -EDQUOT:
case -ENOSPC:
mapping_set_error(wb->vnode->vfs_inode.i_mapping, -ENOSPC);
break;
case -EROFS:
case -EIO:
case -EREMOTEIO:
case -EFBIG:
case -ENOENT:
case -ENOMEDIUM:
case -ENXIO:
afs_kill_pages(wb->vnode, true, first, last);
mapping_set_error(wb->vnode->vfs_inode.i_mapping, -EIO);
break;
case -EACCES:
case -EPERM:
case -ENOKEY:
case -EKEYEXPIRED:
case -EKEYREJECTED:
case -EKEYREVOKED:
afs_kill_pages(wb->vnode, false, first, last);
break;
default:
break;
}
} else {
ret = afs_store_data(mapping, first, last, offset, to);
switch (ret) {
case 0:
ret = count;
break;
default:
pr_notice("kAFS: Unexpected error from FS.StoreData %d\n", ret);
/* Fall through */
case -EACCES:
case -EPERM:
case -ENOKEY:
case -EKEYEXPIRED:
case -EKEYREJECTED:
case -EKEYREVOKED:
afs_redirty_pages(wbc, mapping, first, last);
mapping_set_error(mapping, ret);
break;
case -EDQUOT:
case -ENOSPC:
afs_redirty_pages(wbc, mapping, first, last);
mapping_set_error(mapping, -ENOSPC);
break;
case -EROFS:
case -EIO:
case -EREMOTEIO:
case -EFBIG:
case -ENOENT:
case -ENOMEDIUM:
case -ENXIO:
afs_kill_pages(mapping, first, last);
mapping_set_error(mapping, ret);
break;
}
_leave(" = %d", ret);
@ -462,16 +526,12 @@ no_more:
*/
int afs_writepage(struct page *page, struct writeback_control *wbc)
{
struct afs_writeback *wb;
int ret;
_enter("{%lx},", page->index);
wb = (struct afs_writeback *) page_private(page);
ASSERT(wb != NULL);
ret = afs_write_back_from_locked_page(wb, page);
unlock_page(page);
ret = afs_write_back_from_locked_page(page->mapping, wbc, page,
wbc->range_end >> PAGE_SHIFT);
if (ret < 0) {
_leave(" = %d", ret);
return 0;
@ -490,7 +550,6 @@ static int afs_writepages_region(struct address_space *mapping,
struct writeback_control *wbc,
pgoff_t index, pgoff_t end, pgoff_t *_next)
{
struct afs_writeback *wb;
struct page *page;
int ret, n;
@ -509,7 +568,12 @@ static int afs_writepages_region(struct address_space *mapping,
* (changing page->mapping to NULL), or even swizzled back from
* swapper_space to tmpfs file mapping
*/
lock_page(page);
ret = lock_page_killable(page);
if (ret < 0) {
put_page(page);
_leave(" = %d", ret);
return ret;
}
if (page->mapping != mapping || !PageDirty(page)) {
unlock_page(page);
@ -525,17 +589,9 @@ static int afs_writepages_region(struct address_space *mapping,
continue;
}
wb = (struct afs_writeback *) page_private(page);
ASSERT(wb != NULL);
spin_lock(&wb->vnode->writeback_lock);
wb->state = AFS_WBACK_WRITING;
spin_unlock(&wb->vnode->writeback_lock);
if (!clear_page_dirty_for_io(page))
BUG();
ret = afs_write_back_from_locked_page(wb, page);
unlock_page(page);
ret = afs_write_back_from_locked_page(mapping, wbc, page, end);
put_page(page);
if (ret < 0) {
_leave(" = %d", ret);
@ -591,17 +647,14 @@ int afs_writepages(struct address_space *mapping,
*/
void afs_pages_written_back(struct afs_vnode *vnode, struct afs_call *call)
{
struct afs_writeback *wb = call->wb;
struct pagevec pv;
unsigned long priv;
unsigned count, loop;
pgoff_t first = call->first, last = call->last;
bool free_wb;
_enter("{%x:%u},{%lx-%lx}",
vnode->fid.vid, vnode->fid.vnode, first, last);
ASSERT(wb != NULL);
pagevec_init(&pv);
do {
@ -610,35 +663,22 @@ void afs_pages_written_back(struct afs_vnode *vnode, struct afs_call *call)
count = last - first + 1;
if (count > PAGEVEC_SIZE)
count = PAGEVEC_SIZE;
pv.nr = find_get_pages_contig(call->mapping, first, count,
pv.pages);
pv.nr = find_get_pages_contig(vnode->vfs_inode.i_mapping,
first, count, pv.pages);
ASSERTCMP(pv.nr, ==, count);
spin_lock(&vnode->writeback_lock);
for (loop = 0; loop < count; loop++) {
struct page *page = pv.pages[loop];
end_page_writeback(page);
if (page_private(page) == (unsigned long) wb) {
set_page_private(page, 0);
ClearPagePrivate(page);
wb->usage--;
}
priv = page_private(pv.pages[loop]);
trace_afs_page_dirty(vnode, tracepoint_string("clear"),
pv.pages[loop]->index, priv);
set_page_private(pv.pages[loop], 0);
end_page_writeback(pv.pages[loop]);
}
free_wb = false;
if (wb->usage == 0) {
afs_unlink_writeback(wb);
free_wb = true;
}
spin_unlock(&vnode->writeback_lock);
first += count;
if (free_wb) {
afs_free_writeback(wb);
wb = NULL;
}
__pagevec_release(&pv);
} while (first <= last);
afs_prune_wb_keys(vnode);
_leave("");
}
@ -669,28 +709,6 @@ ssize_t afs_file_write(struct kiocb *iocb, struct iov_iter *from)
return result;
}
/*
* flush the vnode to the fileserver
*/
int afs_writeback_all(struct afs_vnode *vnode)
{
struct address_space *mapping = vnode->vfs_inode.i_mapping;
struct writeback_control wbc = {
.sync_mode = WB_SYNC_ALL,
.nr_to_write = LONG_MAX,
.range_cyclic = 1,
};
int ret;
_enter("");
ret = mapping->a_ops->writepages(mapping, &wbc);
__mark_inode_dirty(mapping->host, I_DIRTY_PAGES);
_leave(" = %d", ret);
return ret;
}
/*
* flush any dirty pages for this process, and check for write errors.
* - the return status from this call provides a reliable indication of
@ -699,61 +717,13 @@ int afs_writeback_all(struct afs_vnode *vnode)
int afs_fsync(struct file *file, loff_t start, loff_t end, int datasync)
{
struct inode *inode = file_inode(file);
struct afs_writeback *wb, *xwb;
struct afs_vnode *vnode = AFS_FS_I(inode);
int ret;
_enter("{%x:%u},{n=%pD},%d",
vnode->fid.vid, vnode->fid.vnode, file,
datasync);
ret = file_write_and_wait_range(file, start, end);
if (ret)
return ret;
inode_lock(inode);
/* use a writeback record as a marker in the queue - when this reaches
* the front of the queue, all the outstanding writes are either
* completed or rejected */
wb = kzalloc(sizeof(*wb), GFP_KERNEL);
if (!wb) {
ret = -ENOMEM;
goto out;
}
wb->vnode = vnode;
wb->first = 0;
wb->last = -1;
wb->offset_first = 0;
wb->to_last = PAGE_SIZE;
wb->usage = 1;
wb->state = AFS_WBACK_SYNCING;
init_waitqueue_head(&wb->waitq);
spin_lock(&vnode->writeback_lock);
list_for_each_entry(xwb, &vnode->writebacks, link) {
if (xwb->state == AFS_WBACK_PENDING)
xwb->state = AFS_WBACK_CONFLICTING;
}
list_add_tail(&wb->link, &vnode->writebacks);
spin_unlock(&vnode->writeback_lock);
/* push all the outstanding writebacks to the server */
ret = afs_writeback_all(vnode);
if (ret < 0) {
afs_put_writeback(wb);
_leave(" = %d [wb]", ret);
goto out;
}
/* wait for the preceding writes to actually complete */
ret = wait_event_interruptible(wb->waitq,
wb->state == AFS_WBACK_COMPLETE ||
vnode->writebacks.next == &wb->link);
afs_put_writeback(wb);
_leave(" = %d", ret);
out:
inode_unlock(inode);
return ret;
return file_write_and_wait_range(file, start, end);
}
/*
@ -774,19 +744,114 @@ int afs_flush(struct file *file, fl_owner_t id)
* notification that a previously read-only page is about to become writable
* - if it returns an error, the caller will deliver a bus error signal
*/
int afs_page_mkwrite(struct vm_area_struct *vma, struct page *page)
int afs_page_mkwrite(struct vm_fault *vmf)
{
struct afs_vnode *vnode = AFS_FS_I(vma->vm_file->f_mapping->host);
struct file *file = vmf->vma->vm_file;
struct inode *inode = file_inode(file);
struct afs_vnode *vnode = AFS_FS_I(inode);
unsigned long priv;
_enter("{{%x:%u}},{%lx}",
vnode->fid.vid, vnode->fid.vnode, page->index);
vnode->fid.vid, vnode->fid.vnode, vmf->page->index);
/* wait for the page to be written to the cache before we allow it to
* be modified */
sb_start_pagefault(inode->i_sb);
/* Wait for the page to be written to the cache before we allow it to
* be modified. We then assume the entire page will need writing back.
*/
#ifdef CONFIG_AFS_FSCACHE
fscache_wait_on_page_write(vnode->cache, page);
fscache_wait_on_page_write(vnode->cache, vmf->page);
#endif
_leave(" = 0");
return 0;
if (PageWriteback(vmf->page) &&
wait_on_page_bit_killable(vmf->page, PG_writeback) < 0)
return VM_FAULT_RETRY;
if (lock_page_killable(vmf->page) < 0)
return VM_FAULT_RETRY;
/* We mustn't change page->private until writeback is complete as that
* details the portion of the page we need to write back and we might
* need to redirty the page if there's a problem.
*/
wait_on_page_writeback(vmf->page);
priv = (unsigned long)PAGE_SIZE << AFS_PRIV_SHIFT; /* To */
priv |= 0; /* From */
trace_afs_page_dirty(vnode, tracepoint_string("mkwrite"),
vmf->page->index, priv);
SetPagePrivate(vmf->page);
set_page_private(vmf->page, priv);
sb_end_pagefault(inode->i_sb);
return VM_FAULT_LOCKED;
}
/*
* Prune the keys cached for writeback. The caller must hold vnode->wb_lock.
*/
void afs_prune_wb_keys(struct afs_vnode *vnode)
{
LIST_HEAD(graveyard);
struct afs_wb_key *wbk, *tmp;
/* Discard unused keys */
spin_lock(&vnode->wb_lock);
if (!mapping_tagged(&vnode->vfs_inode.i_data, PAGECACHE_TAG_WRITEBACK) &&
!mapping_tagged(&vnode->vfs_inode.i_data, PAGECACHE_TAG_DIRTY)) {
list_for_each_entry_safe(wbk, tmp, &vnode->wb_keys, vnode_link) {
if (refcount_read(&wbk->usage) == 1)
list_move(&wbk->vnode_link, &graveyard);
}
}
spin_unlock(&vnode->wb_lock);
while (!list_empty(&graveyard)) {
wbk = list_entry(graveyard.next, struct afs_wb_key, vnode_link);
list_del(&wbk->vnode_link);
afs_put_wb_key(wbk);
}
}
/*
* Clean up a page during invalidation.
*/
int afs_launder_page(struct page *page)
{
struct address_space *mapping = page->mapping;
struct afs_vnode *vnode = AFS_FS_I(mapping->host);
unsigned long priv;
unsigned int f, t;
int ret = 0;
_enter("{%lx}", page->index);
priv = page_private(page);
if (clear_page_dirty_for_io(page)) {
f = 0;
t = PAGE_SIZE;
if (PagePrivate(page)) {
f = priv & AFS_PRIV_MAX;
t = priv >> AFS_PRIV_SHIFT;
}
trace_afs_page_dirty(vnode, tracepoint_string("launder"),
page->index, priv);
ret = afs_store_data(mapping, page->index, page->index, t, f);
}
trace_afs_page_dirty(vnode, tracepoint_string("laundered"),
page->index, priv);
set_page_private(page, 0);
ClearPagePrivate(page);
#ifdef CONFIG_AFS_FSCACHE
if (PageFsCache(page)) {
fscache_wait_on_page_write(vnode->cache, page);
fscache_uncache_page(vnode->cache, page);
}
#endif
return ret;
}

4
fs/afs/xattr.c
View File

@ -45,7 +45,7 @@ static int afs_xattr_get_cell(const struct xattr_handler *handler,
struct afs_cell *cell = vnode->volume->cell;
size_t namelen;
namelen = strlen(cell->name);
namelen = cell->name_len;
if (size == 0)
return namelen;
if (namelen > size)
@ -96,7 +96,7 @@ static int afs_xattr_get_volume(const struct xattr_handler *handler,
void *buffer, size_t size)
{
struct afs_vnode *vnode = AFS_FS_I(inode);
const char *volname = vnode->volume->vlocation->vldb.name;
const char *volname = vnode->volume->name;
size_t namelen;
namelen = strlen(volname);

27
fs/btrfs/extent-tree.c
View File

@ -3992,16 +3992,9 @@ void btrfs_dec_nocow_writers(struct btrfs_fs_info *fs_info, u64 bytenr)
btrfs_put_block_group(bg);
}
static int btrfs_wait_nocow_writers_atomic_t(atomic_t *a)
{
schedule();
return 0;
}
void btrfs_wait_nocow_writers(struct btrfs_block_group_cache *bg)
{
wait_on_atomic_t(&bg->nocow_writers,
btrfs_wait_nocow_writers_atomic_t,
wait_on_atomic_t(&bg->nocow_writers, atomic_t_wait,
TASK_UNINTERRUPTIBLE);
}
@ -6530,12 +6523,6 @@ void btrfs_dec_block_group_reservations(struct btrfs_fs_info *fs_info,
btrfs_put_block_group(bg);
}
static int btrfs_wait_bg_reservations_atomic_t(atomic_t *a)
{
schedule();
return 0;
}
void btrfs_wait_block_group_reservations(struct btrfs_block_group_cache *bg)
{
struct btrfs_space_info *space_info = bg->space_info;
@ -6558,8 +6545,7 @@ void btrfs_wait_block_group_reservations(struct btrfs_block_group_cache *bg)
down_write(&space_info->groups_sem);
up_write(&space_info->groups_sem);
wait_on_atomic_t(&bg->reservations,
btrfs_wait_bg_reservations_atomic_t,
wait_on_atomic_t(&bg->reservations, atomic_t_wait,
TASK_UNINTERRUPTIBLE);
}
@ -11059,12 +11045,6 @@ int btrfs_start_write_no_snapshotting(struct btrfs_root *root)
return 1;
}
static int wait_snapshotting_atomic_t(atomic_t *a)
{
schedule();
return 0;
}
void btrfs_wait_for_snapshot_creation(struct btrfs_root *root)
{
while (true) {
@ -11073,8 +11053,7 @@ void btrfs_wait_for_snapshot_creation(struct btrfs_root *root)
ret = btrfs_start_write_no_snapshotting(root);
if (ret)
break;
wait_on_atomic_t(&root->will_be_snapshotted,
wait_snapshotting_atomic_t,
wait_on_atomic_t(&root->will_be_snapshotted, atomic_t_wait,
TASK_UNINTERRUPTIBLE);
}
}

2
fs/fscache/cookie.c
View File

@ -558,7 +558,7 @@ void __fscache_disable_cookie(struct fscache_cookie *cookie, bool invalidate)
* have completed.
*/
if (!atomic_dec_and_test(&cookie->n_active))
wait_on_atomic_t(&cookie->n_active, fscache_wait_atomic_t,
wait_on_atomic_t(&cookie->n_active, atomic_t_wait,
TASK_UNINTERRUPTIBLE);
/* Make sure any pending writes are cancelled. */

2
fs/fscache/internal.h
View File

@ -97,8 +97,6 @@ static inline bool fscache_object_congested(void)
return workqueue_congested(WORK_CPU_UNBOUND, fscache_object_wq);
}
extern int fscache_wait_atomic_t(atomic_t *);
/*
* object.c
*/

9
fs/fscache/main.c
View File

@ -195,12 +195,3 @@ static void __exit fscache_exit(void)
}
module_exit(fscache_exit);
/*
* wait_on_atomic_t() sleep function for uninterruptible waiting
*/
int fscache_wait_atomic_t(atomic_t *p)
{
schedule();
return 0;
}

4
fs/nfs/inode.c
View File

@ -85,9 +85,9 @@ int nfs_wait_bit_killable(struct wait_bit_key *key, int mode)
}
EXPORT_SYMBOL_GPL(nfs_wait_bit_killable);
int nfs_wait_atomic_killable(atomic_t *p)
int nfs_wait_atomic_killable(atomic_t *p, unsigned int mode)
{
return nfs_wait_killable(TASK_KILLABLE);
return nfs_wait_killable(mode);
}
/**

2
fs/nfs/internal.h
View File

@ -388,7 +388,7 @@ extern void nfs_evict_inode(struct inode *);
void nfs_zap_acl_cache(struct inode *inode);
extern bool nfs_check_cache_invalid(struct inode *, unsigned long);
extern int nfs_wait_bit_killable(struct wait_bit_key *key, int mode);
extern int nfs_wait_atomic_killable(atomic_t *p);
extern int nfs_wait_atomic_killable(atomic_t *p, unsigned int mode);
/* super.c */
extern const struct super_operations nfs_sops;

8
fs/ocfs2/filecheck.c
View File

@ -129,19 +129,13 @@ static struct kobj_attribute ocfs2_attr_filecheck_set =
ocfs2_filecheck_show,
ocfs2_filecheck_store);
static int ocfs2_filecheck_sysfs_wait(atomic_t *p)
{
schedule();
return 0;
}
static void
ocfs2_filecheck_sysfs_free(struct ocfs2_filecheck_sysfs_entry *entry)
{
struct ocfs2_filecheck_entry *p;
if (!atomic_dec_and_test(&entry->fs_count))
wait_on_atomic_t(&entry->fs_count, ocfs2_filecheck_sysfs_wait,
wait_on_atomic_t(&entry->fs_count, atomic_t_wait,
TASK_UNINTERRUPTIBLE);
spin_lock(&entry->fs_fcheck->fc_lock);

15
include/linux/wait_bit.h
View File

@ -26,6 +26,8 @@ struct wait_bit_queue_entry {
{ .flags = p, .bit_nr = WAIT_ATOMIC_T_BIT_NR, }
typedef int wait_bit_action_f(struct wait_bit_key *key, int mode);
typedef int wait_atomic_t_action_f(atomic_t *counter, unsigned int mode);
void __wake_up_bit(struct wait_queue_head *wq_head, void *word, int bit);
int __wait_on_bit(struct wait_queue_head *wq_head, struct wait_bit_queue_entry *wbq_entry, wait_bit_action_f *action, unsigned int mode);
int __wait_on_bit_lock(struct wait_queue_head *wq_head, struct wait_bit_queue_entry *wbq_entry, wait_bit_action_f *action, unsigned int mode);
@ -34,7 +36,7 @@ void wake_up_atomic_t(atomic_t *p);
int out_of_line_wait_on_bit(void *word, int, wait_bit_action_f *action, unsigned int mode);
int out_of_line_wait_on_bit_timeout(void *word, int, wait_bit_action_f *action, unsigned int mode, unsigned long timeout);
int out_of_line_wait_on_bit_lock(void *word, int, wait_bit_action_f *action, unsigned int mode);
int out_of_line_wait_on_atomic_t(atomic_t *p, int (*)(atomic_t *), unsigned int mode);
int out_of_line_wait_on_atomic_t(atomic_t *p, wait_atomic_t_action_f action, unsigned int mode);
struct wait_queue_head *bit_waitqueue(void *word, int bit);
extern void __init wait_bit_init(void);
@ -51,10 +53,11 @@ int wake_bit_function(struct wait_queue_entry *wq_entry, unsigned mode, int sync
}, \
}
extern int bit_wait(struct wait_bit_key *key, int bit);
extern int bit_wait_io(struct wait_bit_key *key, int bit);
extern int bit_wait_timeout(struct wait_bit_key *key, int bit);
extern int bit_wait_io_timeout(struct wait_bit_key *key, int bit);
extern int bit_wait(struct wait_bit_key *key, int mode);
extern int bit_wait_io(struct wait_bit_key *key, int mode);
extern int bit_wait_timeout(struct wait_bit_key *key, int mode);
extern int bit_wait_io_timeout(struct wait_bit_key *key, int mode);
extern int atomic_t_wait(atomic_t *counter, unsigned int mode);
/**
* wait_on_bit - wait for a bit to be cleared
@ -251,7 +254,7 @@ wait_on_bit_lock_action(unsigned long *word, int bit, wait_bit_action_f *action,
* outside of the target 'word'.
*/
static inline
int wait_on_atomic_t(atomic_t *val, int (*action)(atomic_t *), unsigned mode)
int wait_on_atomic_t(atomic_t *val, wait_atomic_t_action_f action, unsigned mode)
{
might_sleep();
if (atomic_read(val) == 0)

293
include/trace/events/afs.h
View File

@ -30,6 +30,38 @@ enum afs_call_trace {
afs_call_trace_work,
};
enum afs_fs_operation {
afs_FS_FetchData = 130, /* AFS Fetch file data */
afs_FS_FetchStatus = 132, /* AFS Fetch file status */
afs_FS_StoreData = 133, /* AFS Store file data */
afs_FS_StoreStatus = 135, /* AFS Store file status */
afs_FS_RemoveFile = 136, /* AFS Remove a file */
afs_FS_CreateFile = 137, /* AFS Create a file */
afs_FS_Rename = 138, /* AFS Rename or move a file or directory */
afs_FS_Symlink = 139, /* AFS Create a symbolic link */
afs_FS_Link = 140, /* AFS Create a hard link */
afs_FS_MakeDir = 141, /* AFS Create a directory */
afs_FS_RemoveDir = 142, /* AFS Remove a directory */
afs_FS_GetVolumeInfo = 148, /* AFS Get information about a volume */
afs_FS_GetVolumeStatus = 149, /* AFS Get volume status information */
afs_FS_GetRootVolume = 151, /* AFS Get root volume name */
afs_FS_SetLock = 156, /* AFS Request a file lock */
afs_FS_ExtendLock = 157, /* AFS Extend a file lock */
afs_FS_ReleaseLock = 158, /* AFS Release a file lock */
afs_FS_Lookup = 161, /* AFS lookup file in directory */
afs_FS_FetchData64 = 65537, /* AFS Fetch file data */
afs_FS_StoreData64 = 65538, /* AFS Store file data */
afs_FS_GiveUpAllCallBacks = 65539, /* AFS Give up all our callbacks on a server */
afs_FS_GetCapabilities = 65540, /* AFS Get FS server capabilities */
};
enum afs_vl_operation {
afs_VL_GetEntryByNameU = 527, /* AFS Get Vol Entry By Name operation ID */
afs_VL_GetAddrsU = 533, /* AFS Get FS server addresses */
afs_YFSVL_GetEndpoints = 64002, /* YFS Get FS & Vol server addresses */
afs_VL_GetCapabilities = 65537, /* AFS Get VL server capabilities */
};
#endif /* end __AFS_DECLARE_TRACE_ENUMS_ONCE_ONLY */
/*
@ -42,6 +74,37 @@ enum afs_call_trace {
EM(afs_call_trace_wake, "WAKE ") \
E_(afs_call_trace_work, "WORK ")
#define afs_fs_operations \
EM(afs_FS_FetchData, "FS.FetchData") \
EM(afs_FS_FetchStatus, "FS.FetchStatus") \
EM(afs_FS_StoreData, "FS.StoreData") \
EM(afs_FS_StoreStatus, "FS.StoreStatus") \
EM(afs_FS_RemoveFile, "FS.RemoveFile") \
EM(afs_FS_CreateFile, "FS.CreateFile") \
EM(afs_FS_Rename, "FS.Rename") \
EM(afs_FS_Symlink, "FS.Symlink") \
EM(afs_FS_Link, "FS.Link") \
EM(afs_FS_MakeDir, "FS.MakeDir") \
EM(afs_FS_RemoveDir, "FS.RemoveDir") \
EM(afs_FS_GetVolumeInfo, "FS.GetVolumeInfo") \
EM(afs_FS_GetVolumeStatus, "FS.GetVolumeStatus") \
EM(afs_FS_GetRootVolume, "FS.GetRootVolume") \
EM(afs_FS_SetLock, "FS.SetLock") \
EM(afs_FS_ExtendLock, "FS.ExtendLock") \
EM(afs_FS_ReleaseLock, "FS.ReleaseLock") \
EM(afs_FS_Lookup, "FS.Lookup") \
EM(afs_FS_FetchData64, "FS.FetchData64") \
EM(afs_FS_StoreData64, "FS.StoreData64") \
EM(afs_FS_GiveUpAllCallBacks, "FS.GiveUpAllCallBacks") \
E_(afs_FS_GetCapabilities, "FS.GetCapabilities")
#define afs_vl_operations \
EM(afs_VL_GetEntryByNameU, "VL.GetEntryByNameU") \
EM(afs_VL_GetAddrsU, "VL.GetAddrsU") \
EM(afs_YFSVL_GetEndpoints, "YFSVL.GetEndpoints") \
E_(afs_VL_GetCapabilities, "VL.GetCapabilities")
/*
* Export enum symbols via userspace.
*/
@ -51,6 +114,8 @@ enum afs_call_trace {
#define E_(a, b) TRACE_DEFINE_ENUM(a);
afs_call_traces;
afs_fs_operations;
afs_vl_operations;
/*
* Now redefine the EM() and E_() macros to map the enums to the strings that
@ -178,6 +243,234 @@ TRACE_EVENT(afs_call,
__entry->where)
);
TRACE_EVENT(afs_make_fs_call,
TP_PROTO(struct afs_call *call, const struct afs_fid *fid),
TP_ARGS(call, fid),
TP_STRUCT__entry(
__field(struct afs_call *, call )
__field(enum afs_fs_operation, op )
__field_struct(struct afs_fid, fid )
),
TP_fast_assign(
__entry->call = call;
__entry->op = call->operation_ID;
if (fid) {
__entry->fid = *fid;
} else {
__entry->fid.vid = 0;
__entry->fid.vnode = 0;
__entry->fid.unique = 0;
}
),
TP_printk("c=%p %06x:%06x:%06x %s",
__entry->call,
__entry->fid.vid,
__entry->fid.vnode,
__entry->fid.unique,
__print_symbolic(__entry->op, afs_fs_operations))
);
TRACE_EVENT(afs_make_vl_call,
TP_PROTO(struct afs_call *call),
TP_ARGS(call),
TP_STRUCT__entry(
__field(struct afs_call *, call )
__field(enum afs_vl_operation, op )
),
TP_fast_assign(
__entry->call = call;
__entry->op = call->operation_ID;
),
TP_printk("c=%p %s",
__entry->call,
__print_symbolic(__entry->op, afs_vl_operations))
);
TRACE_EVENT(afs_call_done,
TP_PROTO(struct afs_call *call),
TP_ARGS(call),
TP_STRUCT__entry(
__field(struct afs_call *, call )
__field(struct rxrpc_call *, rx_call )
__field(int, ret )
__field(u32, abort_code )
),
TP_fast_assign(
__entry->call = call;
__entry->rx_call = call->rxcall;
__entry->ret = call->error;
__entry->abort_code = call->abort_code;
),
TP_printk(" c=%p ret=%d ab=%d [%p]",
__entry->call,
__entry->ret,
__entry->abort_code,
__entry->rx_call)
);
TRACE_EVENT(afs_send_pages,
TP_PROTO(struct afs_call *call, struct msghdr *msg,
pgoff_t first, pgoff_t last, unsigned int offset),
TP_ARGS(call, msg, first, last, offset),
TP_STRUCT__entry(
__field(struct afs_call *, call )
__field(pgoff_t, first )
__field(pgoff_t, last )
__field(unsigned int, nr )
__field(unsigned int, bytes )
__field(unsigned int, offset )
__field(unsigned int, flags )
),
TP_fast_assign(
__entry->call = call;
__entry->first = first;
__entry->last = last;
__entry->nr = msg->msg_iter.nr_segs;
__entry->bytes = msg->msg_iter.count;
__entry->offset = offset;
__entry->flags = msg->msg_flags;
),
TP_printk(" c=%p %lx-%lx-%lx b=%x o=%x f=%x",
__entry->call,
__entry->first, __entry->first + __entry->nr - 1, __entry->last,
__entry->bytes, __entry->offset,
__entry->flags)
);
TRACE_EVENT(afs_sent_pages,
TP_PROTO(struct afs_call *call, pgoff_t first, pgoff_t last,
pgoff_t cursor, int ret),
TP_ARGS(call, first, last, cursor, ret),
TP_STRUCT__entry(
__field(struct afs_call *, call )
__field(pgoff_t, first )
__field(pgoff_t, last )
__field(pgoff_t, cursor )
__field(int, ret )
),
TP_fast_assign(
__entry->call = call;
__entry->first = first;
__entry->last = last;
__entry->cursor = cursor;
__entry->ret = ret;
),
TP_printk(" c=%p %lx-%lx c=%lx r=%d",
__entry->call,
__entry->first, __entry->last,
__entry->cursor, __entry->ret)
);
TRACE_EVENT(afs_dir_check_failed,
TP_PROTO(struct afs_vnode *vnode, loff_t off, loff_t i_size),
TP_ARGS(vnode, off, i_size),
TP_STRUCT__entry(
__field(struct afs_vnode *, vnode )
__field(loff_t, off )
__field(loff_t, i_size )
),
TP_fast_assign(
__entry->vnode = vnode;
__entry->off = off;
__entry->i_size = i_size;
),
TP_printk("vn=%p %llx/%llx",
__entry->vnode, __entry->off, __entry->i_size)
);
/*
* We use page->private to hold the amount of the page that we've written to,
* splitting the field into two parts. However, we need to represent a range
* 0...PAGE_SIZE inclusive, so we can't support 64K pages on a 32-bit system.
*/
#if PAGE_SIZE > 32768
#define AFS_PRIV_MAX 0xffffffff
#define AFS_PRIV_SHIFT 32
#else
#define AFS_PRIV_MAX 0xffff
#define AFS_PRIV_SHIFT 16
#endif
TRACE_EVENT(afs_page_dirty,
TP_PROTO(struct afs_vnode *vnode, const char *where,
pgoff_t page, unsigned long priv),
TP_ARGS(vnode, where, page, priv),
TP_STRUCT__entry(
__field(struct afs_vnode *, vnode )
__field(const char *, where )
__field(pgoff_t, page )
__field(unsigned long, priv )
),
TP_fast_assign(
__entry->vnode = vnode;
__entry->where = where;
__entry->page = page;
__entry->priv = priv;
),
TP_printk("vn=%p %lx %s %lu-%lu",
__entry->vnode, __entry->page, __entry->where,
__entry->priv & AFS_PRIV_MAX,
__entry->priv >> AFS_PRIV_SHIFT)
);
TRACE_EVENT(afs_call_state,
TP_PROTO(struct afs_call *call,
enum afs_call_state from,
enum afs_call_state to,
int ret, u32 remote_abort),
TP_ARGS(call, from, to, ret, remote_abort),
TP_STRUCT__entry(
__field(struct afs_call *, call )
__field(enum afs_call_state, from )
__field(enum afs_call_state, to )
__field(int, ret )
__field(u32, abort )
),
TP_fast_assign(
__entry->call = call;
__entry->from = from;
__entry->to = to;
__entry->ret = ret;
__entry->abort = remote_abort;
),
TP_printk("c=%p %u->%u r=%d ab=%d",
__entry->call,
__entry->from, __entry->to,
__entry->ret, __entry->abort)
);
#endif /* _TRACE_AFS_H */
/* This part must be outside protection */

1
include/uapi/linux/magic.h
View File

@ -47,6 +47,7 @@
#define OPENPROM_SUPER_MAGIC 0x9fa1
#define QNX4_SUPER_MAGIC 0x002f /* qnx4 fs detection */
#define QNX6_SUPER_MAGIC 0x68191122 /* qnx6 fs detection */
#define AFS_FS_MAGIC 0x6B414653
#define REISERFS_SUPER_MAGIC 0x52654973 /* used by gcc */
/* used by file system utilities that

18
kernel/sched/wait_bit.c
View File

@ -183,7 +183,7 @@ static int wake_atomic_t_function(struct wait_queue_entry *wq_entry, unsigned mo
*/
static __sched
int __wait_on_atomic_t(struct wait_queue_head *wq_head, struct wait_bit_queue_entry *wbq_entry,
int (*action)(atomic_t *), unsigned mode)
wait_atomic_t_action_f action, unsigned int mode)
{
atomic_t *val;
int ret = 0;
@ -193,7 +193,7 @@ int __wait_on_atomic_t(struct wait_queue_head *wq_head, struct wait_bit_queue_en
val = wbq_entry->key.flags;
if (atomic_read(val) == 0)
break;
ret = (*action)(val);
ret = (*action)(val, mode);
} while (!ret && atomic_read(val) != 0);
finish_wait(wq_head, &wbq_entry->wq_entry);
return ret;
@ -210,8 +210,9 @@ int __wait_on_atomic_t(struct wait_queue_head *wq_head, struct wait_bit_queue_en
}, \
}
__sched int out_of_line_wait_on_atomic_t(atomic_t *p, int (*action)(atomic_t *),
unsigned mode)
__sched int out_of_line_wait_on_atomic_t(atomic_t *p,
wait_atomic_t_action_f action,
unsigned int mode)
{
struct wait_queue_head *wq_head = atomic_t_waitqueue(p);
DEFINE_WAIT_ATOMIC_T(wq_entry, p);
@ -220,6 +221,15 @@ __sched int out_of_line_wait_on_atomic_t(atomic_t *p, int (*action)(atomic_t *),
}
EXPORT_SYMBOL(out_of_line_wait_on_atomic_t);
__sched int atomic_t_wait(atomic_t *counter, unsigned int mode)
{
schedule();
if (signal_pending_state(mode, current))
return -EINTR;
return 0;
}
EXPORT_SYMBOL(atomic_t_wait);
/**
* wake_up_atomic_t - Wake up a waiter on a atomic_t
* @p: The atomic_t being waited on, a kernel virtual address

1
mm/filemap.c
View File

@ -1139,6 +1139,7 @@ int wait_on_page_bit_killable(struct page *page, int bit_nr)
wait_queue_head_t *q = page_waitqueue(page);
return wait_on_page_bit_common(q, page, bit_nr, TASK_KILLABLE, false);
}
EXPORT_SYMBOL(wait_on_page_bit_killable);
/**
* add_page_wait_queue - Add an arbitrary waiter to a page's wait queue