Skip to content

Commit d2ddc77

Browse files
dhowellsdhowells
committed
afs: Overhaul volume and server record caching and fileserver rotation
The current code assumes that volumes and servers are per-cell and are never shared, but this is not enforced, and, indeed, public cells do exist that are aliases of each other. Further, an organisation can, say, set up a public cell and a private cell with overlapping, but not identical, sets of servers. The difference is purely in the database attached to the VL servers. The current code will malfunction if it sees a server in two cells as it assumes global address -> server record mappings and that each server is in just one cell. Further, each server may have multiple addresses - and may have addresses of different families (IPv4 and IPv6, say). To this end, the following structural changes are made: (1) Server record management is overhauled: (a) Server records are made independent of cell. The namespace keeps track of them, volume records have lists of them and each vnode has a server on which its callback interest currently resides. (b) The cell record no longer keeps a list of servers known to be in that cell. (c) The server records are now kept in a flat list because there's no single address to sort on. (d) Server records are now keyed by their UUID within the namespace. (e) The addresses for a server are obtained with the VL.GetAddrsU rather than with VL.GetEntryByName, using the server's UUID as a parameter. (f) Cached server records are garbage collected after a period of non-use and are counted out of existence before purging is allowed to complete. This protects the work functions against rmmod. (g) The servers list is now in /proc/fs/afs/servers. (2) Volume record management is overhauled: (a) An RCU-replaceable server list is introduced. This tracks both servers and their coresponding callback interests. (b) The superblock is now keyed on cell record and numeric volume ID. (c) The volume record is now tied to the superblock which mounts it, and is activated when mounted and deactivated when unmounted. This makes it easier to handle the cache cookie without causing a double-use in fscache. (d) The volume record is loaded from the VLDB using VL.GetEntryByNameU to get the server UUID list. (e) The volume name is updated if it is seen to have changed when the volume is updated (the update is keyed on the volume ID). (3) The vlocation record is got rid of and VLDB records are no longer cached. Sufficient information is stored in the volume record, though an update to a volume record is now no longer shared between related volumes (volumes come in bundles of three: R/W, R/O and backup). and the following procedural changes are made: (1) The fileserver cursor introduced previously is now fleshed out and used to iterate over fileservers and their addresses. (2) Volume status is checked during iteration, and the server list is replaced if a change is detected. (3) Server status is checked during iteration, and the address list is replaced if a change is detected. (4) The abort code is saved into the address list cursor and -ECONNABORTED returned in afs_make_call() if a remote abort happened rather than translating the abort into an error message. This allows actions to be taken depending on the abort code more easily. (a) If a VMOVED abort is seen then this is handled by rechecking the volume and restarting the iteration. (b) If a VBUSY, VRESTARTING or VSALVAGING abort is seen then this is handled by sleeping for a short period and retrying and/or trying other servers that might serve that volume. A message is also displayed once until the condition has cleared. (c) If a VOFFLINE abort is seen, then this is handled as VBUSY for the moment. (d) If a VNOVOL abort is seen, the volume is rechecked in the VLDB to see if it has been deleted; if not, the fileserver is probably indicating that the volume couldn't be attached and needs salvaging. (e) If statfs() sees one of these aborts, it does not sleep, but rather returns an error, so as not to block the umount program. (5) The fileserver iteration functions in vnode.c are now merged into their callers and more heavily macroised around the cursor. vnode.c is removed. (6) Operations on a particular vnode are serialised on that vnode because the server will lock that vnode whilst it operates on it, so a second op sent will just have to wait. (7) Fileservers are probed with FS.GetCapabilities before being used. This is where service upgrade will be done. (8) A callback interest on a fileserver is set up before an FS operation is performed and passed through to afs_make_call() so that it can be set on the vnode if the operation returns a callback. The callback interest is passed through to afs_iget() also so that it can be set there too. In general, record updating is done on an as-needed basis when we try to access servers, volumes or vnodes rather than offloading it to work items and special threads. Notes: (1) Pre AFS-3.4 servers are no longer supported, though this can be added back if necessary (AFS-3.4 was released in 1998). (2) VBUSY is retried forever for the moment at intervals of 1s. (3) /proc/fs/afs/<cell>/servers no longer exists. Signed-off-by: David Howells <dhowells@redhat.com>
1 parent 9cc6fc5 commit d2ddc77

26 files changed

+2795
-2575
lines changed

‎fs/afs/Makefile

Lines changed: 1 addition & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -24,11 +24,10 @@ kafs-objs := \
2424
rxrpc.o \
2525
security.o \
2626
server.o \
27+
server_list.o \
2728
super.o \
2829
netdevices.o \
2930
vlclient.o \
30-
vlocation.o \
31-
vnode.o \
3231
volume.o \
3332
write.o \
3433
xattr.o

‎fs/afs/addr_list.c

Lines changed: 31 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -227,6 +227,37 @@ struct afs_addr_list *afs_dns_query(struct afs_cell *cell, time64_t *_expiry)
227227
return alist;
228228
}
229229

230+
/*
231+
* Merge an IPv4 entry into a fileserver address list.
232+
*/
233+
void afs_merge_fs_addr4(struct afs_addr_list *alist, __be32 xdr)
234+
{
235+
struct sockaddr_in6 *a;
236+
int i;
237+
238+
for (i = 0; i < alist->nr_ipv4; i++) {
239+
a = &alist->addrs[i].transport.sin6;
240+
if (xdr == a->sin6_addr.s6_addr32[3])
241+
return;
242+
if (xdr < a->sin6_addr.s6_addr32[3])
243+
break;
244+
}
245+
246+
if (i < alist->nr_addrs)
247+
memmove(alist->addrs + i + 1,
248+
alist->addrs + i,
249+
sizeof(alist->addrs[0]) * (alist->nr_addrs - i));
250+
251+
a = &alist->addrs[i].transport.sin6;
252+
a->sin6_port = htons(AFS_FS_PORT);
253+
a->sin6_addr.s6_addr32[0] = 0;
254+
a->sin6_addr.s6_addr32[1] = 0;
255+
a->sin6_addr.s6_addr32[2] = htonl(0xffff);
256+
a->sin6_addr.s6_addr32[3] = xdr;
257+
alist->nr_ipv4++;
258+
alist->nr_addrs++;
259+
}
260+
230261
/*
231262
* Get an address to try.
232263
*/

‎fs/afs/afs_fs.h

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -38,6 +38,7 @@ enum AFS_FS_Operations {
3838
FSFETCHDATA64 = 65537, /* AFS Fetch file data */
3939
FSSTOREDATA64 = 65538, /* AFS Store file data */
4040
FSGIVEUPALLCALLBACKS = 65539, /* AFS Give up all outstanding callbacks on a server */
41+
FSGETCAPABILITIES = 65540, /* Probe and get the capabilities of a fileserver */
4142
};
4243

4344
enum AFS_FS_Errors {

‎fs/afs/afs_vl.h

Lines changed: 42 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -88,4 +88,46 @@ struct afs_vldbentry {
8888

8989
#define AFS_VLDB_MAXNAMELEN 65
9090

91+
92+
struct afs_ListAddrByAttributes__xdr {
93+
__be32 Mask;
94+
#define AFS_VLADDR_IPADDR 0x1 /* Match by ->ipaddr */
95+
#define AFS_VLADDR_INDEX 0x2 /* Match by ->index */
96+
#define AFS_VLADDR_UUID 0x4 /* Match by ->uuid */
97+
__be32 ipaddr;
98+
__be32 index;
99+
__be32 spare;
100+
struct afs_uuid__xdr uuid;
101+
};
102+
103+
struct afs_uvldbentry__xdr {
104+
__be32 name[AFS_VLDB_MAXNAMELEN];
105+
__be32 nServers;
106+
struct afs_uuid__xdr serverNumber[AFS_NMAXNSERVERS];
107+
__be32 serverUnique[AFS_NMAXNSERVERS];
108+
__be32 serverPartition[AFS_NMAXNSERVERS];
109+
__be32 serverFlags[AFS_NMAXNSERVERS];
110+
__be32 volumeId[AFS_MAXTYPES];
111+
__be32 cloneId;
112+
__be32 flags;
113+
__be32 spares1;
114+
__be32 spares2;
115+
__be32 spares3;
116+
__be32 spares4;
117+
__be32 spares5;
118+
__be32 spares6;
119+
__be32 spares7;
120+
__be32 spares8;
121+
__be32 spares9;
122+
};
123+
124+
struct afs_address_list {
125+
refcount_t usage;
126+
unsigned int version;
127+
unsigned int nr_addrs;
128+
struct sockaddr_rxrpc addrs[];
129+
};
130+
131+
extern void afs_put_address_list(struct afs_address_list *alist);
132+
91133
#endif /* AFS_VL_H */

‎fs/afs/callback.c

Lines changed: 10 additions & 10 deletions
Original file line numberDiff line numberDiff line change
@@ -26,10 +26,10 @@
2626
* - Called with volume->server_sem held.
2727
*/
2828
int afs_register_server_cb_interest(struct afs_vnode *vnode,
29-
struct afs_cb_interest **ppcbi,
30-
struct afs_server *server)
29+
struct afs_server_entry *entry)
3130
{
32-
struct afs_cb_interest *cbi = *ppcbi, *vcbi, *new, *x;
31+
struct afs_cb_interest *cbi = entry->cb_interest, *vcbi, *new, *x;
32+
struct afs_server *server = entry->server;
3333

3434
again:
3535
vcbi = vnode->cb_interest;
@@ -47,7 +47,7 @@ int afs_register_server_cb_interest(struct afs_vnode *vnode,
4747

4848
if (!cbi && vcbi->server == server) {
4949
afs_get_cb_interest(vcbi);
50-
x = cmpxchg(ppcbi, cbi, vcbi);
50+
x = cmpxchg(&entry->cb_interest, cbi, vcbi);
5151
if (x != cbi) {
5252
cbi = x;
5353
afs_put_cb_interest(afs_v2net(vnode), vcbi);
@@ -72,7 +72,7 @@ int afs_register_server_cb_interest(struct afs_vnode *vnode,
7272
list_add_tail(&new->cb_link, &server->cb_interests);
7373
write_unlock(&server->cb_break_lock);
7474

75-
x = cmpxchg(ppcbi, cbi, new);
75+
x = cmpxchg(&entry->cb_interest, cbi, new);
7676
if (x == cbi) {
7777
cbi = new;
7878
} else {
@@ -137,7 +137,7 @@ void afs_put_cb_interest(struct afs_net *net, struct afs_cb_interest *cbi)
137137
*/
138138
void afs_init_callback_state(struct afs_server *server)
139139
{
140-
if (!test_and_clear_bit(AFS_SERVER_NEW, &server->flags))
140+
if (!test_and_clear_bit(AFS_SERVER_FL_NEW, &server->flags))
141141
server->cb_s_break++;
142142
}
143143

@@ -233,12 +233,12 @@ void afs_break_callbacks(struct afs_server *server, size_t count,
233233
/*
234234
* Clear the callback interests in a server list.
235235
*/
236-
void afs_clear_callback_interests(struct afs_net *net, struct afs_volume *volume)
236+
void afs_clear_callback_interests(struct afs_net *net, struct afs_server_list *slist)
237237
{
238238
int i;
239239

240-
for (i = 0; i < ARRAY_SIZE(volume->cb_interests); i++) {
241-
afs_put_cb_interest(net, volume->cb_interests[i]);
242-
volume->cb_interests[i] = NULL;
240+
for (i = 0; i < slist->nr_servers; i++) {
241+
afs_put_cb_interest(net, slist->servers[i].cb_interest);
242+
slist->servers[i].cb_interest = NULL;
243243
}
244244
}

‎fs/afs/cell.c

Lines changed: 2 additions & 5 deletions
Original file line numberDiff line numberDiff line change
@@ -146,13 +146,10 @@ static struct afs_cell *afs_alloc_cell(struct afs_net *net,
146146

147147
atomic_set(&cell->usage, 2);
148148
INIT_WORK(&cell->manager, afs_manage_cell);
149-
rwlock_init(&cell->servers_lock);
150-
INIT_LIST_HEAD(&cell->servers);
151-
init_rwsem(&cell->vl_sem);
152-
INIT_LIST_HEAD(&cell->vl_list);
153-
spin_lock_init(&cell->vl_lock);
154149
cell->flags = ((1 << AFS_CELL_FL_NOT_READY) |
155150
(1 << AFS_CELL_FL_NO_LOOKUP_YET));
151+
INIT_LIST_HEAD(&cell->proc_volumes);
152+
rwlock_init(&cell->proc_lock);
156153
rwlock_init(&cell->vl_addrs_lock);
157154

158155
/* Fill in the VL server list if we were given a list of addresses to

0 commit comments

Comments
 (0)