* AFS volume synchronisation information
*/
struct afs_volsync {
- time64_t creation; /* volume creation time */
+ time64_t creation; /* Volume creation time (or TIME64_MIN) */
+ time64_t update; /* Volume update time (or TIME64_MIN) */
};
/*
clear_bit(AFS_VNODE_NEW_CONTENT, &vnode->flags);
if (test_and_clear_bit(AFS_VNODE_CB_PROMISED, &vnode->flags)) {
vnode->cb_break++;
- vnode->cb_v_break = vnode->volume->cb_v_break;
+ vnode->cb_v_break = atomic_read(&vnode->volume->cb_v_break);
afs_clear_permits(vnode);
if (vnode->lock_state == AFS_VNODE_LOCK_WAITING_FOR_CB)
struct super_block *sb;
struct afs_vnode *vnode;
struct inode *inode;
+ unsigned int cb_v_break;
if (fid->vnode == 0 && fid->unique == 0) {
/* The callback break applies to an entire volume. */
write_lock(&volume->cb_v_break_lock);
- volume->cb_v_break++;
- trace_afs_cb_break(fid, volume->cb_v_break,
+ cb_v_break = atomic_inc_return(&volume->cb_v_break);
+ trace_afs_cb_break(fid, cb_v_break,
afs_cb_break_for_volume_callback, false);
write_unlock(&volume->cb_v_break_lock);
return;
key_get(key);
}
- op->key = key;
- op->volume = afs_get_volume(volume, afs_volume_trace_get_new_op);
- op->net = volume->cell->net;
- op->cb_v_break = volume->cb_v_break;
- op->debug_id = atomic_inc_return(&afs_operation_debug_counter);
+ op->key = key;
+ op->volume = afs_get_volume(volume, afs_volume_trace_get_new_op);
+ op->net = volume->cell->net;
+ op->cb_v_break = atomic_read(&volume->cb_v_break);
+ op->pre_volsync.creation = volume->creation_time;
+ op->pre_volsync.update = volume->update_time;
+ op->debug_id = atomic_inc_return(&afs_operation_debug_counter);
op->nr_iterations = -1;
afs_op_set_error(op, -EDESTADDRREQ);
afs_prepare_vnode(op, &op->file[0], 0);
afs_prepare_vnode(op, &op->file[1], 1);
- op->cb_v_break = op->volume->cb_v_break;
+ op->cb_v_break = atomic_read(&op->volume->cb_v_break);
_leave(" = true");
return true;
}
return ret;
bp = call->buffer;
- xdr_decode_AFSVolSync(&bp, &op->volsync);
+ /* Unfortunately, prior to OpenAFS-1.6, volsync here is filled
+ * with rubbish.
+ */
+ xdr_decode_AFSVolSync(&bp, NULL);
call->unmarshall++;
fallthrough;
BUG_ON(!(inode->i_state & I_NEW));
vnode = AFS_FS_I(inode);
- vnode->cb_v_break = as->volume->cb_v_break,
+ vnode->cb_v_break = atomic_read(&as->volume->cb_v_break),
afs_set_netfs_context(vnode);
op = afs_alloc_operation(key, as->volume);
rwlock_t servers_lock; /* Lock for ->servers */
unsigned int servers_seq; /* Incremented each time ->servers changes */
- unsigned cb_v_break; /* Break-everything counter. */
+ /* RO release tracking */
+ struct mutex volsync_lock; /* Time/state evaluation lock */
+ time64_t creation_time; /* Volume creation time (or TIME64_MIN) */
+ time64_t update_time; /* Volume update time (or TIME64_MIN) */
+
+ /* Callback management */
+ atomic_t cb_ro_snapshot; /* RO volume update-from-snapshot counter */
+ atomic_t cb_v_break; /* Volume-break event counter. */
+ atomic_t cb_scrub; /* Scrub-all-data event counter. */
rwlock_t cb_v_break_lock;
afs_voltype_t type; /* type of volume */
struct afs_volume *volume; /* Volume being accessed */
struct afs_vnode_param file[2];
struct afs_vnode_param *more_files;
- struct afs_volsync volsync;
+ struct afs_volsync pre_volsync; /* Volsync before op */
+ struct afs_volsync volsync; /* Volsync returned by op */
struct dentry *dentry; /* Dentry to be altered */
struct dentry *dentry_2; /* Second dentry to be altered */
struct timespec64 mtime; /* Modification time to record */
static inline bool afs_cb_is_broken(unsigned int cb_break,
const struct afs_vnode *vnode)
{
- return cb_break != (vnode->cb_break + vnode->volume->cb_v_break);
+ return cb_break != (vnode->cb_break + atomic_read(&vnode->volume->cb_v_break));
}
/*
/*
* validation.c
*/
+int afs_update_volume_state(struct afs_operation *op);
bool afs_check_validity(struct afs_vnode *vnode);
bool afs_pagecache_valid(struct afs_vnode *vnode);
int afs_validate(struct afs_vnode *vnode, struct key *key);
vnode->cb_server = server;
vnode->cb_s_break = server->cb_s_break;
vnode->cb_fs_s_break = atomic_read(&server->cell->fs_s_break);
- vnode->cb_v_break = vnode->volume->cb_v_break;
+ vnode->cb_v_break = atomic_read(&vnode->volume->cb_v_break);
clear_bit(AFS_VNODE_CB_PROMISED, &vnode->flags);
}
op->addr_index = addr_index;
set_bit(addr_index, &op->addr_tried);
+ op->volsync.creation = TIME64_MIN;
+ op->volsync.update = TIME64_MIN;
op->call_responded = false;
_debug("address [%u] %u/%u %pISp",
op->server_index, addr_index, alist->nr_addrs,
#include <linux/sched.h>
#include "internal.h"
+/*
+ * See if the server we've just talked to is currently excluded.
+ */
+static bool __afs_is_server_excluded(struct afs_operation *op, struct afs_volume *volume)
+{
+ const struct afs_server_entry *se;
+ const struct afs_server_list *slist;
+ bool is_excluded = true;
+ int i;
+
+ rcu_read_lock();
+
+ slist = rcu_dereference(volume->servers);
+ for (i = 0; i < slist->nr_servers; i++) {
+ se = &slist->servers[i];
+ if (op->server == se->server) {
+ is_excluded = test_bit(AFS_SE_EXCLUDED, &se->flags);
+ break;
+ }
+ }
+
+ rcu_read_unlock();
+ return is_excluded;
+}
+
+/*
+ * Update the volume's server list when the creation time changes and see if
+ * the server we've just talked to is currently excluded.
+ */
+static int afs_is_server_excluded(struct afs_operation *op, struct afs_volume *volume)
+{
+ int ret;
+
+ if (__afs_is_server_excluded(op, volume))
+ return 1;
+
+ set_bit(AFS_VOLUME_NEEDS_UPDATE, &volume->flags);
+ ret = afs_check_volume_status(op->volume, op);
+ if (ret < 0)
+ return ret;
+
+ return __afs_is_server_excluded(op, volume);
+}
+
+/*
+ * Handle a change to the volume creation time in the VolSync record.
+ */
+static int afs_update_volume_creation_time(struct afs_operation *op, struct afs_volume *volume)
+{
+ unsigned int snap;
+ time64_t cur = volume->creation_time;
+ time64_t old = op->pre_volsync.creation;
+ time64_t new = op->volsync.creation;
+ int ret;
+
+ _enter("%llx,%llx,%llx->%llx", volume->vid, cur, old, new);
+
+ if (cur == TIME64_MIN) {
+ volume->creation_time = new;
+ return 0;
+ }
+
+ if (new == cur)
+ return 0;
+
+ /* Try to advance the creation timestamp from what we had before the
+ * operation to what we got back from the server. This should
+ * hopefully ensure that in a race between multiple operations only one
+ * of them will do this.
+ */
+ if (cur != old)
+ return 0;
+
+ /* If the creation time changes in an unexpected way, we need to scrub
+ * our caches. For a RW vol, this will only change if the volume is
+ * restored from a backup; for a RO/Backup vol, this will advance when
+ * the volume is updated to a new snapshot (eg. "vos release").
+ */
+ if (volume->type == AFSVL_RWVOL)
+ goto regressed;
+ if (volume->type == AFSVL_BACKVOL) {
+ if (new < old)
+ goto regressed;
+ goto advance;
+ }
+
+ /* We have an RO volume, we need to query the VL server and look at the
+ * server flags to see if RW->RO replication is in progress.
+ */
+ ret = afs_is_server_excluded(op, volume);
+ if (ret < 0)
+ return ret;
+ if (ret > 0) {
+ snap = atomic_read(&volume->cb_ro_snapshot);
+ trace_afs_cb_v_break(volume->vid, snap, afs_cb_break_volume_excluded);
+ return ret;
+ }
+
+advance:
+ snap = atomic_inc_return(&volume->cb_ro_snapshot);
+ trace_afs_cb_v_break(volume->vid, snap, afs_cb_break_for_vos_release);
+ volume->creation_time = new;
+ return 0;
+
+regressed:
+ atomic_inc(&volume->cb_scrub);
+ trace_afs_cb_v_break(volume->vid, 0, afs_cb_break_for_creation_regress);
+ volume->creation_time = new;
+ return 0;
+}
+
+/*
+ * Handle a change to the volume update time in the VolSync record.
+ */
+static void afs_update_volume_update_time(struct afs_operation *op, struct afs_volume *volume)
+{
+ enum afs_cb_break_reason reason = afs_cb_break_no_break;
+ time64_t cur = volume->update_time;
+ time64_t old = op->pre_volsync.update;
+ time64_t new = op->volsync.update;
+
+ _enter("%llx,%llx,%llx->%llx", volume->vid, cur, old, new);
+
+ if (cur == TIME64_MIN) {
+ volume->update_time = new;
+ return;
+ }
+
+ if (new == cur)
+ return;
+
+ /* If the volume update time changes in an unexpected way, we need to
+ * scrub our caches. For a RW vol, this will advance on every
+ * modification op; for a RO/Backup vol, this will advance when the
+ * volume is updated to a new snapshot (eg. "vos release").
+ */
+ if (new < old)
+ reason = afs_cb_break_for_update_regress;
+
+ /* Try to advance the update timestamp from what we had before the
+ * operation to what we got back from the server. This should
+ * hopefully ensure that in a race between multiple operations only one
+ * of them will do this.
+ */
+ if (cur == old) {
+ if (reason == afs_cb_break_for_update_regress) {
+ atomic_inc(&volume->cb_scrub);
+ trace_afs_cb_v_break(volume->vid, 0, reason);
+ }
+ volume->update_time = new;
+ }
+}
+
+static int afs_update_volume_times(struct afs_operation *op, struct afs_volume *volume)
+{
+ int ret = 0;
+
+ if (likely(op->volsync.creation == volume->creation_time &&
+ op->volsync.update == volume->update_time))
+ return 0;
+
+ mutex_lock(&volume->volsync_lock);
+ if (op->volsync.creation != volume->creation_time) {
+ ret = afs_update_volume_creation_time(op, volume);
+ if (ret < 0)
+ goto out;
+ }
+ if (op->volsync.update != volume->update_time)
+ afs_update_volume_update_time(op, volume);
+out:
+ mutex_unlock(&volume->volsync_lock);
+ return ret;
+}
+
+/*
+ * Update the state of a volume. Returns 1 to redo the operation from the start.
+ */
+int afs_update_volume_state(struct afs_operation *op)
+{
+ struct afs_volume *volume = op->volume;
+ int ret;
+
+ _enter("%llx", op->volume->vid);
+
+ if (op->volsync.creation != TIME64_MIN || op->volsync.update != TIME64_MIN) {
+ ret = afs_update_volume_times(op, volume);
+ if (ret != 0) {
+ _leave(" = %d", ret);
+ return ret;
+ }
+ }
+
+ return 0;
+}
+
/*
* mark the data attached to an inode as obsolete due to a write on the server
* - might also want to ditch all the outstanding writes and dirty pages
cb_break = vnode->cb_break;
if (test_bit(AFS_VNODE_CB_PROMISED, &vnode->flags)) {
- if (vnode->cb_v_break != vnode->volume->cb_v_break)
+ if (vnode->cb_v_break != atomic_read(&vnode->volume->cb_v_break))
need_clear = afs_cb_break_for_v_break;
else if (!afs_check_server_good(vnode))
need_clear = afs_cb_break_for_s_reinit;
write_seqlock(&vnode->cb_lock);
if (need_clear == afs_cb_break_no_promise)
- vnode->cb_v_break = vnode->volume->cb_v_break;
+ vnode->cb_v_break = atomic_read(&vnode->volume->cb_v_break);
else if (cb_break == vnode->cb_break)
__afs_break_callback(vnode, need_clear);
else
volume->type = params->type;
volume->type_force = params->force;
volume->name_len = vldb->name_len;
+ volume->creation_time = TIME64_MIN;
+ volume->update_time = TIME64_MIN;
refcount_set(&volume->ref, 1);
INIT_HLIST_NODE(&volume->proc_link);
INIT_WORK(&volume->destructor, afs_destroy_volume);
rwlock_init(&volume->servers_lock);
+ mutex_init(&volume->volsync_lock);
rwlock_init(&volume->cb_v_break_lock);
memcpy(volume->name, vldb->name, vldb->name_len + 1);
struct afs_volsync *volsync)
{
struct yfs_xdr_YFSVolSync *x = (void *)*_bp;
- u64 creation;
+ u64 creation, update;
if (volsync) {
creation = xdr_to_u64(x->vol_creation_date);
do_div(creation, 10 * 1000 * 1000);
volsync->creation = creation;
+ update = xdr_to_u64(x->vol_update_date);
+ do_div(update, 10 * 1000 * 1000);
+ volsync->update = update;
}
*_bp += xdr_size(x);
EM(afs_cb_break_no_break, "no-break") \
EM(afs_cb_break_no_promise, "no-promise") \
EM(afs_cb_break_for_callback, "break-cb") \
+ EM(afs_cb_break_for_creation_regress, "creation-regress") \
EM(afs_cb_break_for_deleted, "break-del") \
EM(afs_cb_break_for_lapsed, "break-lapsed") \
EM(afs_cb_break_for_s_reinit, "s-reinit") \
EM(afs_cb_break_for_unlink, "break-unlink") \
+ EM(afs_cb_break_for_update_regress, "update-regress") \
EM(afs_cb_break_for_v_break, "break-v") \
EM(afs_cb_break_for_volume_callback, "break-v-cb") \
- E_(afs_cb_break_for_zap, "break-zap")
+ EM(afs_cb_break_for_vos_release, "break-vos-release") \
+ EM(afs_cb_break_for_zap, "break-zap") \
+ E_(afs_cb_break_volume_excluded, "vol-excluded")
/*
* Generate enums for tracing information.
__entry->cell, __entry->volume, __entry->vid)
);
+TRACE_EVENT(afs_cb_v_break,
+ TP_PROTO(afs_volid_t vid, unsigned int cb_v_break,
+ enum afs_cb_break_reason reason),
+
+ TP_ARGS(vid, cb_v_break, reason),
+
+ TP_STRUCT__entry(
+ __field(afs_volid_t, vid)
+ __field(unsigned int, cb_v_break)
+ __field(enum afs_cb_break_reason, reason)
+ ),
+
+ TP_fast_assign(
+ __entry->vid = vid;
+ __entry->cb_v_break = cb_v_break;
+ __entry->reason = reason;
+ ),
+
+ TP_printk("%llx vb=%x %s",
+ __entry->vid,
+ __entry->cb_v_break,
+ __print_symbolic(__entry->reason, afs_cb_break_reasons))
+ );
+
TRACE_EVENT(afs_cb_break,
TP_PROTO(struct afs_fid *fid, unsigned int cb_break,
enum afs_cb_break_reason reason, bool skipped),
projects / linux.git / commitdiff