static void __send_request(struct ceph_osd_client *osdc,
struct ceph_osd_request *req);
-static int op_needs_trail(int op)
-{
- switch (op) {
- case CEPH_OSD_OP_GETXATTR:
- case CEPH_OSD_OP_SETXATTR:
- case CEPH_OSD_OP_CMPXATTR:
- case CEPH_OSD_OP_CALL:
- case CEPH_OSD_OP_NOTIFY:
- return 1;
- default:
- return 0;
- }
-}
-
static int op_has_extent(int op)
{
return (op == CEPH_OSD_OP_READ ||
op == CEPH_OSD_OP_WRITE);
}
-void ceph_calc_raw_layout(struct ceph_osd_client *osdc,
+int ceph_calc_raw_layout(struct ceph_osd_client *osdc,
struct ceph_file_layout *layout,
- u64 snapid,
u64 off, u64 *plen, u64 *bno,
struct ceph_osd_request *req,
struct ceph_osd_req_op *op)
{
- struct ceph_osd_request_head *reqhead = req->r_request->front.iov_base;
u64 orig_len = *plen;
u64 objoff, objlen; /* extent in object */
-
- reqhead->snapid = cpu_to_le64(snapid);
+ int r;
/* object extent? */
- ceph_calc_file_object_mapping(layout, off, plen, bno,
- &objoff, &objlen);
- if (*plen < orig_len)
+ r = ceph_calc_file_object_mapping(layout, off, orig_len, bno,
+ &objoff, &objlen);
+ if (r < 0)
+ return r;
+ if (objlen < orig_len) {
+ *plen = objlen;
dout(" skipping last %llu, final file extent %llu~%llu\n",
orig_len - *plen, off, *plen);
+ }
if (op_has_extent(op->op)) {
+ u32 osize = le32_to_cpu(layout->fl_object_size);
op->extent.offset = objoff;
op->extent.length = objlen;
+ if (op->extent.truncate_size <= off - objoff) {
+ op->extent.truncate_size = 0;
+ } else {
+ op->extent.truncate_size -= off - objoff;
+ if (op->extent.truncate_size > osize)
+ op->extent.truncate_size = osize;
+ }
}
req->r_num_pages = calc_pages_for(off, *plen);
req->r_page_alignment = off & ~PAGE_MASK;
dout("calc_layout bno=%llx %llu~%llu (%d pages)\n",
*bno, objoff, objlen, req->r_num_pages);
-
+ return 0;
}
EXPORT_SYMBOL(ceph_calc_raw_layout);
*
* fill osd op in request message.
*/
-static void calc_layout(struct ceph_osd_client *osdc,
- struct ceph_vino vino,
- struct ceph_file_layout *layout,
- u64 off, u64 *plen,
- struct ceph_osd_request *req,
- struct ceph_osd_req_op *op)
+static int calc_layout(struct ceph_osd_client *osdc,
+ struct ceph_vino vino,
+ struct ceph_file_layout *layout,
+ u64 off, u64 *plen,
+ struct ceph_osd_request *req,
+ struct ceph_osd_req_op *op)
{
u64 bno;
+ int r;
- ceph_calc_raw_layout(osdc, layout, vino.snap, off,
- plen, &bno, req, op);
+ r = ceph_calc_raw_layout(osdc, layout, off, plen, &bno, req, op);
+ if (r < 0)
+ return r;
snprintf(req->r_oid, sizeof(req->r_oid), "%llx.%08llx", vino.ino, bno);
req->r_oid_len = strlen(req->r_oid);
+
+ return r;
}
/*
bio_put(req->r_bio);
#endif
ceph_put_snap_context(req->r_snapc);
- if (req->r_trail) {
- ceph_pagelist_release(req->r_trail);
- kfree(req->r_trail);
- }
+ ceph_pagelist_release(&req->r_trail);
if (req->r_mempool)
mempool_free(req, req->r_osdc->req_mempool);
else
}
EXPORT_SYMBOL(ceph_osdc_release_request);
-static int get_num_ops(struct ceph_osd_req_op *ops, int *needs_trail)
+static int get_num_ops(struct ceph_osd_req_op *ops)
{
int i = 0;
- if (needs_trail)
- *needs_trail = 0;
- while (ops[i].op) {
- if (needs_trail && op_needs_trail(ops[i].op))
- *needs_trail = 1;
+ while (ops[i].op)
i++;
- }
return i;
}
{
struct ceph_osd_request *req;
struct ceph_msg *msg;
- int needs_trail;
- int num_op = get_num_ops(ops, &needs_trail);
+ int num_op = get_num_ops(ops);
size_t msg_size = sizeof(struct ceph_osd_request_head);
msg_size += num_op*sizeof(struct ceph_osd_op);
kref_init(&req->r_kref);
init_completion(&req->r_completion);
init_completion(&req->r_safe_completion);
- rb_init_node(&req->r_node);
+ RB_CLEAR_NODE(&req->r_node);
INIT_LIST_HEAD(&req->r_unsafe_item);
INIT_LIST_HEAD(&req->r_linger_item);
INIT_LIST_HEAD(&req->r_linger_osd);
}
req->r_reply = msg;
- /* allocate space for the trailing data */
- if (needs_trail) {
- req->r_trail = kmalloc(sizeof(struct ceph_pagelist), gfp_flags);
- if (!req->r_trail) {
- ceph_osdc_put_request(req);
- return NULL;
- }
- ceph_pagelist_init(req->r_trail);
- }
+ ceph_pagelist_init(&req->r_trail);
/* create request message; allow space for oid */
msg_size += MAX_OBJ_NAME_SIZE;
case CEPH_OSD_OP_GETXATTR:
case CEPH_OSD_OP_SETXATTR:
case CEPH_OSD_OP_CMPXATTR:
- BUG_ON(!req->r_trail);
-
dst->xattr.name_len = cpu_to_le32(src->xattr.name_len);
dst->xattr.value_len = cpu_to_le32(src->xattr.value_len);
dst->xattr.cmp_op = src->xattr.cmp_op;
dst->xattr.cmp_mode = src->xattr.cmp_mode;
- ceph_pagelist_append(req->r_trail, src->xattr.name,
+ ceph_pagelist_append(&req->r_trail, src->xattr.name,
src->xattr.name_len);
- ceph_pagelist_append(req->r_trail, src->xattr.val,
+ ceph_pagelist_append(&req->r_trail, src->xattr.val,
src->xattr.value_len);
break;
case CEPH_OSD_OP_CALL:
- BUG_ON(!req->r_trail);
-
dst->cls.class_len = src->cls.class_len;
dst->cls.method_len = src->cls.method_len;
dst->cls.indata_len = cpu_to_le32(src->cls.indata_len);
- ceph_pagelist_append(req->r_trail, src->cls.class_name,
+ ceph_pagelist_append(&req->r_trail, src->cls.class_name,
src->cls.class_len);
- ceph_pagelist_append(req->r_trail, src->cls.method_name,
+ ceph_pagelist_append(&req->r_trail, src->cls.method_name,
src->cls.method_len);
- ceph_pagelist_append(req->r_trail, src->cls.indata,
+ ceph_pagelist_append(&req->r_trail, src->cls.indata,
src->cls.indata_len);
break;
case CEPH_OSD_OP_ROLLBACK:
__le32 prot_ver = cpu_to_le32(src->watch.prot_ver);
__le32 timeout = cpu_to_le32(src->watch.timeout);
- BUG_ON(!req->r_trail);
-
- ceph_pagelist_append(req->r_trail,
+ ceph_pagelist_append(&req->r_trail,
&prot_ver, sizeof(prot_ver));
- ceph_pagelist_append(req->r_trail,
+ ceph_pagelist_append(&req->r_trail,
&timeout, sizeof(timeout));
}
case CEPH_OSD_OP_NOTIFY_ACK:
*
*/
void ceph_osdc_build_request(struct ceph_osd_request *req,
- u64 off, u64 *plen,
+ u64 off, u64 len,
struct ceph_osd_req_op *src_ops,
- struct ceph_snap_context *snapc,
- struct timespec *mtime,
- const char *oid,
- int oid_len)
+ struct ceph_snap_context *snapc, u64 snap_id,
+ struct timespec *mtime)
{
struct ceph_msg *msg = req->r_request;
struct ceph_osd_request_head *head;
struct ceph_osd_req_op *src_op;
struct ceph_osd_op *op;
void *p;
- int num_op = get_num_ops(src_ops, NULL);
+ int num_op = get_num_ops(src_ops);
size_t msg_size = sizeof(*head) + num_op*sizeof(*op);
int flags = req->r_flags;
u64 data_len = 0;
int i;
head = msg->front.iov_base;
+ head->snapid = cpu_to_le64(snap_id);
op = (void *)(head + 1);
p = (void *)(op + num_op);
/* fill in oid */
- head->object_len = cpu_to_le32(oid_len);
- memcpy(p, oid, oid_len);
- p += oid_len;
+ head->object_len = cpu_to_le32(req->r_oid_len);
+ memcpy(p, req->r_oid, req->r_oid_len);
+ p += req->r_oid_len;
src_op = src_ops;
while (src_op->op) {
op++;
}
- if (req->r_trail)
- data_len += req->r_trail->length;
+ data_len += req->r_trail.length;
if (snapc) {
head->snap_seq = cpu_to_le64(snapc->seq);
if (flags & CEPH_OSD_FLAG_WRITE) {
req->r_request->hdr.data_off = cpu_to_le16(off);
- req->r_request->hdr.data_len = cpu_to_le32(*plen + data_len);
+ req->r_request->hdr.data_len = cpu_to_le32(len + data_len);
} else if (data_len) {
req->r_request->hdr.data_off = 0;
req->r_request->hdr.data_len = cpu_to_le32(data_len);
{
struct ceph_osd_req_op ops[3];
struct ceph_osd_request *req;
+ int r;
ops[0].op = opcode;
ops[0].extent.truncate_seq = truncate_seq;
use_mempool,
GFP_NOFS, NULL, NULL);
if (!req)
- return NULL;
+ return ERR_PTR(-ENOMEM);
/* calculate max write size */
- calc_layout(osdc, vino, layout, off, plen, req, ops);
+ r = calc_layout(osdc, vino, layout, off, plen, req, ops);
+ if (r < 0)
+ return ERR_PTR(r);
req->r_file_layout = *layout; /* keep a copy */
/* in case it differs from natural (file) alignment that
req->r_num_pages = calc_pages_for(page_align, *plen);
req->r_page_alignment = page_align;
- ceph_osdc_build_request(req, off, plen, ops,
- snapc,
- mtime,
- req->r_oid, req->r_oid_len);
+ ceph_osdc_build_request(req, off, *plen, ops, snapc, vino.snap, mtime);
return req;
}
dout("__kick_osd_requests osd%d\n", osd->o_osd);
err = __reset_osd(osdc, osd);
- if (err == -EAGAIN)
+ if (err)
return;
list_for_each_entry(req, &osd->o_requests, r_osd_item) {
}
}
-static void kick_osd_requests(struct ceph_osd_client *osdc,
- struct ceph_osd *kickosd)
-{
- mutex_lock(&osdc->request_mutex);
- __kick_osd_requests(osdc, kickosd);
- mutex_unlock(&osdc->request_mutex);
-}
-
/*
* If the osd connection drops, we need to resubmit all requests.
*/
dout("osd_reset osd%d\n", osd->o_osd);
osdc = osd->o_osdc;
down_read(&osdc->map_sem);
- kick_osd_requests(osdc, osd);
+ mutex_lock(&osdc->request_mutex);
+ __kick_osd_requests(osdc, osd);
+ mutex_unlock(&osdc->request_mutex);
send_queued(osdc);
up_read(&osdc->map_sem);
}
atomic_set(&osd->o_ref, 1);
osd->o_osdc = osdc;
osd->o_osd = onum;
+ RB_CLEAR_NODE(&osd->o_node);
INIT_LIST_HEAD(&osd->o_requests);
INIT_LIST_HEAD(&osd->o_linger_requests);
INIT_LIST_HEAD(&osd->o_osd_lru);
*/
static int __reset_osd(struct ceph_osd_client *osdc, struct ceph_osd *osd)
{
- struct ceph_osd_request *req;
- int ret = 0;
+ struct ceph_entity_addr *peer_addr;
dout("__reset_osd %p osd%d\n", osd, osd->o_osd);
if (list_empty(&osd->o_requests) &&
list_empty(&osd->o_linger_requests)) {
__remove_osd(osdc, osd);
- } else if (memcmp(&osdc->osdmap->osd_addr[osd->o_osd],
- &osd->o_con.peer_addr,
- sizeof(osd->o_con.peer_addr)) == 0 &&
- !ceph_con_opened(&osd->o_con)) {
+
+ return -ENODEV;
+ }
+
+ peer_addr = &osdc->osdmap->osd_addr[osd->o_osd];
+ if (!memcmp(peer_addr, &osd->o_con.peer_addr, sizeof (*peer_addr)) &&
+ !ceph_con_opened(&osd->o_con)) {
+ struct ceph_osd_request *req;
+
dout(" osd addr hasn't changed and connection never opened,"
" letting msgr retry");
/* touch each r_stamp for handle_timeout()'s benfit */
list_for_each_entry(req, &osd->o_requests, r_osd_item)
req->r_stamp = jiffies;
- ret = -EAGAIN;
- } else {
- ceph_con_close(&osd->o_con);
- ceph_con_open(&osd->o_con, CEPH_ENTITY_TYPE_OSD, osd->o_osd,
- &osdc->osdmap->osd_addr[osd->o_osd]);
- osd->o_incarnation++;
+
+ return -EAGAIN;
}
- return ret;
+
+ ceph_con_close(&osd->o_con);
+ ceph_con_open(&osd->o_con, CEPH_ENTITY_TYPE_OSD, osd->o_osd, peer_addr);
+ osd->o_incarnation++;
+
+ return 0;
}
static void __insert_osd(struct ceph_osd_client *osdc, struct ceph_osd *new)
req->r_osd = NULL;
}
+ list_del_init(&req->r_req_lru_item);
ceph_osdc_put_request(req);
- list_del_init(&req->r_req_lru_item);
if (osdc->num_requests == 0) {
dout(" no requests, canceling timeout\n");
__cancel_osd_timeout(osdc);
struct ceph_osd_request *req)
{
dout("__unregister_linger_request %p\n", req);
+ list_del_init(&req->r_linger_item);
if (req->r_osd) {
- list_del_init(&req->r_linger_item);
list_del_init(&req->r_linger_osd);
if (list_empty(&req->r_osd->o_requests) &&
{
struct ceph_osd_client *osdc =
container_of(work, struct ceph_osd_client, timeout_work.work);
- struct ceph_osd_request *req, *last_req = NULL;
+ struct ceph_osd_request *req;
struct ceph_osd *osd;
- unsigned long timeout = osdc->client->options->osd_timeout * HZ;
unsigned long keepalive =
osdc->client->options->osd_keepalive_timeout * HZ;
- unsigned long last_stamp = 0;
struct list_head slow_osds;
dout("timeout\n");
down_read(&osdc->map_sem);
mutex_lock(&osdc->request_mutex);
- /*
- * reset osds that appear to be _really_ unresponsive. this
- * is a failsafe measure.. we really shouldn't be getting to
- * this point if the system is working properly. the monitors
- * should mark the osd as failed and we should find out about
- * it from an updated osd map.
- */
- while (timeout && !list_empty(&osdc->req_lru)) {
- req = list_entry(osdc->req_lru.next, struct ceph_osd_request,
- r_req_lru_item);
-
- /* hasn't been long enough since we sent it? */
- if (time_before(jiffies, req->r_stamp + timeout))
- break;
-
- /* hasn't been long enough since it was acked? */
- if (req->r_request->ack_stamp == 0 ||
- time_before(jiffies, req->r_request->ack_stamp + timeout))
- break;
-
- BUG_ON(req == last_req && req->r_stamp == last_stamp);
- last_req = req;
- last_stamp = req->r_stamp;
-
- osd = req->r_osd;
- BUG_ON(!osd);
- pr_warning(" tid %llu timed out on osd%d, will reset osd\n",
- req->r_tid, osd->o_osd);
- __kick_osd_requests(osdc, osd);
- }
-
/*
* ping osds that are a bit slow. this ensures that if there
* is a break in the TCP connection we will notice, and reopen
* Requeue requests whose mapping to an OSD has changed. If requests map to
* no osd, request a new map.
*
- * Caller should hold map_sem for read and request_mutex.
+ * Caller should hold map_sem for read.
*/
static void kick_requests(struct ceph_osd_client *osdc, int force_resend)
{
for (p = rb_first(&osdc->requests); p; ) {
req = rb_entry(p, struct ceph_osd_request, r_node);
p = rb_next(p);
+
+ /*
+ * For linger requests that have not yet been
+ * registered, move them to the linger list; they'll
+ * be sent to the osd in the loop below. Unregister
+ * the request before re-registering it as a linger
+ * request to ensure the __map_request() below
+ * will decide it needs to be sent.
+ */
+ if (req->r_linger && list_empty(&req->r_linger_item)) {
+ dout("%p tid %llu restart on osd%d\n",
+ req, req->r_tid,
+ req->r_osd ? req->r_osd->o_osd : -1);
+ __unregister_request(osdc, req);
+ __register_linger_request(osdc, req);
+ continue;
+ }
+
err = __map_request(osdc, req, force_resend);
if (err < 0)
continue; /* error */
req->r_flags |= CEPH_OSD_FLAG_RETRY;
}
}
- if (req->r_linger && list_empty(&req->r_linger_item)) {
- /*
- * register as a linger so that we will
- * re-submit below and get a new tid
- */
- dout("%p tid %llu restart on osd%d\n",
- req, req->r_tid,
- req->r_osd ? req->r_osd->o_osd : -1);
- __register_linger_request(osdc, req);
- __unregister_request(osdc, req);
- }
}
list_for_each_entry_safe(req, nreq, &osdc->req_linger,
dout("linger req=%p req->r_osd=%p\n", req, req->r_osd);
err = __map_request(osdc, req, force_resend);
+ dout("__map_request returned %d\n", err);
if (err == 0)
continue; /* no change and no osd was specified */
if (err < 0)
dout("kicking lingering %p tid %llu osd%d\n", req, req->r_tid,
req->r_osd ? req->r_osd->o_osd : -1);
- __unregister_linger_request(osdc, req);
__register_request(osdc, req);
+ __unregister_linger_request(osdc, req);
}
mutex_unlock(&osdc->request_mutex);
dout("%d requests for down osds, need new map\n", needmap);
ceph_monc_request_next_osdmap(&osdc->client->monc);
}
+ reset_changed_osds(osdc);
}
osdc->osdmap = newmap;
}
kick_requests(osdc, 0);
- reset_changed_osds(osdc);
} else {
dout("ignoring incremental map %u len %d\n",
epoch, maplen);
event->data = data;
event->osdc = osdc;
INIT_LIST_HEAD(&event->osd_node);
+ RB_CLEAR_NODE(&event->node);
kref_init(&event->kref); /* one ref for us */
kref_get(&event->kref); /* one ref for the caller */
init_completion(&event->completion);
#ifdef CONFIG_BLOCK
req->r_request->bio = req->r_bio;
#endif
- req->r_request->trail = req->r_trail;
+ req->r_request->trail = &req->r_trail;
register_request(osdc, req);
CEPH_OSD_OP_READ, CEPH_OSD_FLAG_READ,
NULL, 0, truncate_seq, truncate_size, NULL,
false, 1, page_align);
- if (!req)
- return -ENOMEM;
+ if (IS_ERR(req))
+ return PTR_ERR(req);
/* it may be a short read due to an object boundary */
req->r_pages = pages;
snapc, do_sync,
truncate_seq, truncate_size, mtime,
nofail, 1, page_align);
- if (!req)
- return -ENOMEM;
+ if (IS_ERR(req))
+ return PTR_ERR(req);
/* it may be a short write due to an object boundary */
req->r_pages = pages;