struct ceph_msg *ceph_msgpool_get(struct ceph_msgpool *pool, int front_len)
{
wait_queue_t wait;
struct ceph_msg *msg;
if (front_len && front_len > pool->front_len) {
pr_err("msgpool_get pool %p need front %d, pool size is %d\n",
pool, front_len, pool->front_len);
WARN_ON(1);
/* try to alloc a fresh message */
msg = ceph_msg_new(0, front_len, 0, 0, NULL);
if (!IS_ERR(msg))
return msg;
}
if (!front_len)
front_len = pool->front_len;
if (pool->blocking) {
/* mempool_t behavior; first try to alloc */
msg = ceph_msg_new(0, front_len, 0, 0, NULL);
if (!IS_ERR(msg))
return msg;
}
while (1) {
spin_lock(&pool->lock);
if (likely(pool->num)) {
msg = list_entry(pool->msgs.next, struct ceph_msg,
list_head);
list_del_init(&msg->list_head);
pool->num--;
dout("msgpool_get %p got %p, now %d/%d\n", pool, msg,
pool->num, pool->min);
spin_unlock(&pool->lock);
return msg;
}
pr_err("msgpool_get %p now %d/%d, %s\n", pool, pool->num,
pool->min, pool->blocking ? "waiting" : "may fail");
spin_unlock(&pool->lock);
if (!pool->blocking) {
WARN_ON(1);
/* maybe we can allocate it now? */
msg = ceph_msg_new(0, front_len, 0, 0, NULL);
if (!IS_ERR(msg))
return msg;
pr_err("msgpool_get %p empty + alloc failed\n", pool);
return ERR_PTR(-ENOMEM);
}
init_wait(&wait);
prepare_to_wait(&pool->wait, &wait, TASK_UNINTERRUPTIBLE);
schedule();
finish_wait(&pool->wait, &wait);
}
/*
* Allocate or release as necessary to meet our target pool size.
*/
static int __fill_msgpool(struct ceph_msgpool *pool)
{
struct ceph_msg *msg;
while (pool->num < pool->min) {
dout("fill_msgpool %p %d/%d allocating\n", pool, pool->num,
pool->min);
spin_unlock(&pool->lock);
msg = ceph_msg_new(0, pool->front_len, 0, 0, NULL);
spin_lock(&pool->lock);
if (IS_ERR(msg))
return PTR_ERR(msg);
msg->pool = pool;
list_add(&msg->list_head, &pool->msgs);
pool->num++;
}
while (pool->num > pool->min) {
msg = list_first_entry(&pool->msgs, struct ceph_msg, list_head);
dout("fill_msgpool %p %d/%d releasing %p\n", pool, pool->num,
pool->min, msg);
list_del_init(&msg->list_head);
pool->num--;
ceph_msg_kfree(msg);
}
return 0;
}
static void *alloc_fn(gfp_t gfp_mask, void *arg)
{
struct ceph_msgpool *pool = arg;
void *p;
p = ceph_msg_new(0, pool->front_len, gfp_mask);
if (!p)
pr_err("msgpool %s alloc failed\n", pool->name);
return p;
}
static void *msgpool_alloc(gfp_t gfp_mask, void *arg)
{
struct ceph_msgpool *pool = arg;
struct ceph_msg *msg;
msg = ceph_msg_new(0, pool->front_len, gfp_mask, true);
if (!msg) {
dout("msgpool_alloc %s failed\n", pool->name);
} else {
dout("msgpool_alloc %s %p\n", pool->name, msg);
msg->pool = pool;
}
return msg;
}
struct ceph_msg *ceph_msgpool_get(struct ceph_msgpool *pool,
int front_len)
{
if (front_len > pool->front_len) {
pr_err("msgpool_get pool %s need front %d, pool size is %d\n",
pool->name, front_len, pool->front_len);
WARN_ON(1);
/* try to alloc a fresh message */
return ceph_msg_new(0, front_len, GFP_NOFS);
}
return mempool_alloc(pool->pool, GFP_NOFS);
}
struct ceph_msg *ceph_msgpool_get(struct ceph_msgpool *pool,
int front_len)
{
struct ceph_msg *msg;
if (front_len > pool->front_len) {
dout("msgpool_get %s need front %d, pool size is %d\n",
pool->name, front_len, pool->front_len);
WARN_ON(1);
/* try to alloc a fresh message */
return ceph_msg_new(0, front_len, GFP_NOFS, false);
}
msg = mempool_alloc(pool->pool, GFP_NOFS);
dout("msgpool_get %s %p\n", pool->name, msg);
return msg;
}
struct ceph_osd_request *ceph_osdc_alloc_request(struct ceph_osd_client *osdc,
int flags,
struct ceph_snap_context *snapc,
struct ceph_osd_req_op *ops,
bool use_mempool,
gfp_t gfp_flags,
struct page **pages,
struct bio *bio)
{
struct ceph_osd_request *req;
struct ceph_msg *msg;
int needs_trail;
int num_op = get_num_ops(ops, &needs_trail);
size_t msg_size = sizeof(struct ceph_osd_request_head);
msg_size += num_op*sizeof(struct ceph_osd_op);
if (use_mempool) {
req = mempool_alloc(osdc->req_mempool, gfp_flags);
memset(req, 0, sizeof(*req));
} else {
req = kzalloc(sizeof(*req), gfp_flags);
}
if (req == NULL)
return NULL;
req->r_osdc = osdc;
req->r_mempool = use_mempool;
kref_init(&req->r_kref);
init_completion(&req->r_completion);
init_completion(&req->r_safe_completion);
rb_init_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);
INIT_LIST_HEAD(&req->r_req_lru_item);
INIT_LIST_HEAD(&req->r_osd_item);
req->r_flags = flags;
WARN_ON((flags & (CEPH_OSD_FLAG_READ|CEPH_OSD_FLAG_WRITE)) == 0);
/* create reply message */
if (use_mempool)
msg = ceph_msgpool_get(&osdc->msgpool_op_reply, 0);
else
msg = ceph_msg_new(CEPH_MSG_OSD_OPREPLY,
OSD_OPREPLY_FRONT_LEN, gfp_flags, true);
if (!msg) {
ceph_osdc_put_request(req);
return NULL;
}
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);
}
/* create request message; allow space for oid */
msg_size += MAX_OBJ_NAME_SIZE;
if (snapc)
msg_size += sizeof(u64) * snapc->num_snaps;
if (use_mempool)
msg = ceph_msgpool_get(&osdc->msgpool_op, 0);
else
msg = ceph_msg_new(CEPH_MSG_OSD_OP, msg_size, gfp_flags, true);
if (!msg) {
ceph_osdc_put_request(req);
return NULL;
}
memset(msg->front.iov_base, 0, msg->front.iov_len);
req->r_request = msg;
req->r_pages = pages;
#ifdef CONFIG_BLOCK
if (bio) {
req->r_bio = bio;
bio_get(req->r_bio);
}
#endif
return req;
}
/*
* build new request AND message, calculate layout, and adjust file
* extent as needed.
*
* if the file was recently truncated, we include information about its
* old and new size so that the object can be updated appropriately. (we
* avoid synchronously deleting truncated objects because it's slow.)
*
* if @do_sync, include a 'startsync' command so that the osd will flush
* data quickly.
*/
struct ceph_osd_request *ceph_osdc_new_request(struct ceph_osd_client *osdc,
struct ceph_file_layout *layout,
struct ceph_vino vino,
u64 off, u64 *plen,
int opcode, int flags,
struct ceph_snap_context *snapc,
int do_sync,
u32 truncate_seq,
u64 truncate_size,
struct timespec *mtime,
bool use_mempool, int num_reply)
{
struct ceph_osd_request *req;
struct ceph_msg *msg;
struct ceph_osd_request_head *head;
struct ceph_osd_op *op;
void *p;
int num_op = 1 + do_sync;
size_t msg_size = sizeof(*head) + num_op*sizeof(*op);
int i;
if (use_mempool) {
req = mempool_alloc(osdc->req_mempool, GFP_NOFS);
memset(req, 0, sizeof(*req));
} else {
req = kzalloc(sizeof(*req), GFP_NOFS);
}
if (req == NULL)
return NULL;
req->r_osdc = osdc;
req->r_mempool = use_mempool;
kref_init(&req->r_kref);
init_completion(&req->r_completion);
init_completion(&req->r_safe_completion);
INIT_LIST_HEAD(&req->r_unsafe_item);
req->r_flags = flags;
WARN_ON((flags & (CEPH_OSD_FLAG_READ|CEPH_OSD_FLAG_WRITE)) == 0);
/* create reply message */
if (use_mempool)
msg = ceph_msgpool_get(&osdc->msgpool_op_reply, 0);
else
msg = ceph_msg_new(CEPH_MSG_OSD_OPREPLY,
OSD_OPREPLY_FRONT_LEN, GFP_NOFS);
if (!msg) {
ceph_osdc_put_request(req);
return NULL;
}
req->r_reply = msg;
/* create request message; allow space for oid */
msg_size += 40;
if (snapc)
msg_size += sizeof(u64) * snapc->num_snaps;
if (use_mempool)
msg = ceph_msgpool_get(&osdc->msgpool_op, 0);
else
msg = ceph_msg_new(CEPH_MSG_OSD_OP, msg_size, GFP_NOFS);
if (!msg) {
ceph_osdc_put_request(req);
return NULL;
}
msg->hdr.type = cpu_to_le16(CEPH_MSG_OSD_OP);
memset(msg->front.iov_base, 0, msg->front.iov_len);
head = msg->front.iov_base;
op = (void *)(head + 1);
p = (void *)(op + num_op);
req->r_request = msg;
req->r_snapc = ceph_get_snap_context(snapc);
head->client_inc = cpu_to_le32(1); /* always, for now. */
head->flags = cpu_to_le32(flags);
if (flags & CEPH_OSD_FLAG_WRITE)
ceph_encode_timespec(&head->mtime, mtime);
head->num_ops = cpu_to_le16(num_op);
op->op = cpu_to_le16(opcode);
/* calculate max write size */
calc_layout(osdc, vino, layout, off, plen, req);
req->r_file_layout = *layout; /* keep a copy */
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);
op->payload_len = cpu_to_le32(*plen);
}
op->extent.truncate_size = cpu_to_le64(truncate_size);
op->extent.truncate_seq = cpu_to_le32(truncate_seq);
/* fill in oid */
head->object_len = cpu_to_le32(req->r_oid_len);
memcpy(p, req->r_oid, req->r_oid_len);
p += req->r_oid_len;
if (do_sync) {
op++;
op->op = cpu_to_le16(CEPH_OSD_OP_STARTSYNC);
}
if (snapc) {
head->snap_seq = cpu_to_le64(snapc->seq);
head->num_snaps = cpu_to_le32(snapc->num_snaps);
for (i = 0; i < snapc->num_snaps; i++) {
put_unaligned_le64(snapc->snaps[i], p);
p += sizeof(u64);
}
}
BUG_ON(p > msg->front.iov_base + msg->front.iov_len);
msg_size = p - msg->front.iov_base;
msg->front.iov_len = msg_size;
msg->hdr.front_len = cpu_to_le32(msg_size);
return req;
}
