aboutsummaryrefslogtreecommitdiff
path: root/net/ceph/messenger.c
diff options
context:
space:
mode:
Diffstat (limited to 'net/ceph/messenger.c')
-rw-r--r--net/ceph/messenger.c109
1 files changed, 61 insertions, 48 deletions
diff --git a/net/ceph/messenger.c b/net/ceph/messenger.c
index 4a5df7b1cc9..1948d592aa5 100644
--- a/net/ceph/messenger.c
+++ b/net/ceph/messenger.c
@@ -15,6 +15,7 @@
#include <linux/dns_resolver.h>
#include <net/tcp.h>
+#include <linux/ceph/ceph_features.h>
#include <linux/ceph/libceph.h>
#include <linux/ceph/messenger.h>
#include <linux/ceph/decode.h>
@@ -382,7 +383,7 @@ static void con_sock_state_closed(struct ceph_connection *con)
*/
/* data available on socket, or listen socket received a connect */
-static void ceph_sock_data_ready(struct sock *sk, int count_unused)
+static void ceph_sock_data_ready(struct sock *sk)
{
struct ceph_connection *con = sk->sk_user_data;
if (atomic_read(&con->msgr->stopping)) {
@@ -556,7 +557,7 @@ static int ceph_tcp_sendmsg(struct socket *sock, struct kvec *iov,
return r;
}
-static int ceph_tcp_sendpage(struct socket *sock, struct page *page,
+static int __ceph_tcp_sendpage(struct socket *sock, struct page *page,
int offset, size_t size, bool more)
{
int flags = MSG_DONTWAIT | MSG_NOSIGNAL | (more ? MSG_MORE : MSG_EOR);
@@ -569,6 +570,24 @@ static int ceph_tcp_sendpage(struct socket *sock, struct page *page,
return ret;
}
+static int ceph_tcp_sendpage(struct socket *sock, struct page *page,
+ int offset, size_t size, bool more)
+{
+ int ret;
+ struct kvec iov;
+
+ /* sendpage cannot properly handle pages with page_count == 0,
+ * we need to fallback to sendmsg if that's the case */
+ if (page_count(page) >= 1)
+ return __ceph_tcp_sendpage(sock, page, offset, size, more);
+
+ iov.iov_base = kmap(page) + offset;
+ iov.iov_len = size;
+ ret = ceph_tcp_sendmsg(sock, &iov, 1, size, more);
+ kunmap(page);
+
+ return ret;
+}
/*
* Shutdown/close the socket for the given connection.
@@ -777,13 +796,12 @@ static void ceph_msg_data_bio_cursor_init(struct ceph_msg_data_cursor *cursor,
bio = data->bio;
BUG_ON(!bio);
- BUG_ON(!bio->bi_vcnt);
cursor->resid = min(length, data->bio_length);
cursor->bio = bio;
- cursor->vector_index = 0;
- cursor->vector_offset = 0;
- cursor->last_piece = length <= bio->bi_io_vec[0].bv_len;
+ cursor->bvec_iter = bio->bi_iter;
+ cursor->last_piece =
+ cursor->resid <= bio_iter_len(bio, cursor->bvec_iter);
}
static struct page *ceph_msg_data_bio_next(struct ceph_msg_data_cursor *cursor,
@@ -792,71 +810,67 @@ static struct page *ceph_msg_data_bio_next(struct ceph_msg_data_cursor *cursor,
{
struct ceph_msg_data *data = cursor->data;
struct bio *bio;
- struct bio_vec *bio_vec;
- unsigned int index;
+ struct bio_vec bio_vec;
BUG_ON(data->type != CEPH_MSG_DATA_BIO);
bio = cursor->bio;
BUG_ON(!bio);
- index = cursor->vector_index;
- BUG_ON(index >= (unsigned int) bio->bi_vcnt);
+ bio_vec = bio_iter_iovec(bio, cursor->bvec_iter);
- bio_vec = &bio->bi_io_vec[index];
- BUG_ON(cursor->vector_offset >= bio_vec->bv_len);
- *page_offset = (size_t) (bio_vec->bv_offset + cursor->vector_offset);
+ *page_offset = (size_t) bio_vec.bv_offset;
BUG_ON(*page_offset >= PAGE_SIZE);
if (cursor->last_piece) /* pagelist offset is always 0 */
*length = cursor->resid;
else
- *length = (size_t) (bio_vec->bv_len - cursor->vector_offset);
+ *length = (size_t) bio_vec.bv_len;
BUG_ON(*length > cursor->resid);
BUG_ON(*page_offset + *length > PAGE_SIZE);
- return bio_vec->bv_page;
+ return bio_vec.bv_page;
}
static bool ceph_msg_data_bio_advance(struct ceph_msg_data_cursor *cursor,
size_t bytes)
{
struct bio *bio;
- struct bio_vec *bio_vec;
- unsigned int index;
+ struct bio_vec bio_vec;
BUG_ON(cursor->data->type != CEPH_MSG_DATA_BIO);
bio = cursor->bio;
BUG_ON(!bio);
- index = cursor->vector_index;
- BUG_ON(index >= (unsigned int) bio->bi_vcnt);
- bio_vec = &bio->bi_io_vec[index];
+ bio_vec = bio_iter_iovec(bio, cursor->bvec_iter);
/* Advance the cursor offset */
BUG_ON(cursor->resid < bytes);
cursor->resid -= bytes;
- cursor->vector_offset += bytes;
- if (cursor->vector_offset < bio_vec->bv_len)
+
+ bio_advance_iter(bio, &cursor->bvec_iter, bytes);
+
+ if (bytes < bio_vec.bv_len)
return false; /* more bytes to process in this segment */
- BUG_ON(cursor->vector_offset != bio_vec->bv_len);
/* Move on to the next segment, and possibly the next bio */
- if (++index == (unsigned int) bio->bi_vcnt) {
+ if (!cursor->bvec_iter.bi_size) {
bio = bio->bi_next;
- index = 0;
+ cursor->bio = bio;
+ if (bio)
+ cursor->bvec_iter = bio->bi_iter;
+ else
+ memset(&cursor->bvec_iter, 0,
+ sizeof(cursor->bvec_iter));
}
- cursor->bio = bio;
- cursor->vector_index = index;
- cursor->vector_offset = 0;
if (!cursor->last_piece) {
BUG_ON(!cursor->resid);
BUG_ON(!bio);
/* A short read is OK, so use <= rather than == */
- if (cursor->resid <= bio->bi_io_vec[index].bv_len)
+ if (cursor->resid <= bio_iter_len(bio, cursor->bvec_iter))
cursor->last_piece = true;
}
@@ -923,6 +937,9 @@ static bool ceph_msg_data_pages_advance(struct ceph_msg_data_cursor *cursor,
if (!bytes || cursor->page_offset)
return false; /* more bytes to process in the current page */
+ if (!cursor->resid)
+ return false; /* no more data */
+
/* Move on to the next page; offset is already at 0 */
BUG_ON(cursor->page_index >= cursor->page_count);
@@ -1008,6 +1025,9 @@ static bool ceph_msg_data_pagelist_advance(struct ceph_msg_data_cursor *cursor,
if (!bytes || cursor->offset & ~PAGE_MASK)
return false; /* more bytes to process in the current page */
+ if (!cursor->resid)
+ return false; /* no more data */
+
/* Move on to the next page */
BUG_ON(list_is_last(&cursor->page->lru, &pagelist->head));
@@ -1865,7 +1885,9 @@ int ceph_parse_ips(const char *c, const char *end,
port = (port * 10) + (*p - '0');
p++;
}
- if (port > 65535 || port == 0)
+ if (port == 0)
+ port = CEPH_MON_PORT;
+ else if (port > 65535)
goto bad;
} else {
port = CEPH_MON_PORT;
@@ -1945,7 +1967,8 @@ static int process_connect(struct ceph_connection *con)
{
u64 sup_feat = con->msgr->supported_features;
u64 req_feat = con->msgr->required_features;
- u64 server_feat = le64_to_cpu(con->in_reply.features);
+ u64 server_feat = ceph_sanitize_features(
+ le64_to_cpu(con->in_reply.features));
int ret;
dout("process_connect on %p tag %d\n", con, (int)con->in_tag);
@@ -2853,8 +2876,8 @@ static void con_fault(struct ceph_connection *con)
*/
void ceph_messenger_init(struct ceph_messenger *msgr,
struct ceph_entity_addr *myaddr,
- u32 supported_features,
- u32 required_features,
+ u64 supported_features,
+ u64 required_features,
bool nocrc)
{
msgr->supported_features = supported_features;
@@ -3126,15 +3149,8 @@ struct ceph_msg *ceph_msg_new(int type, int front_len, gfp_t flags,
INIT_LIST_HEAD(&m->data);
/* front */
- m->front_max = front_len;
if (front_len) {
- if (front_len > PAGE_CACHE_SIZE) {
- m->front.iov_base = __vmalloc(front_len, flags,
- PAGE_KERNEL);
- m->front_is_vmalloc = true;
- } else {
- m->front.iov_base = kmalloc(front_len, flags);
- }
+ m->front.iov_base = ceph_kvmalloc(front_len, flags);
if (m->front.iov_base == NULL) {
dout("ceph_msg_new can't allocate %d bytes\n",
front_len);
@@ -3143,7 +3159,7 @@ struct ceph_msg *ceph_msg_new(int type, int front_len, gfp_t flags,
} else {
m->front.iov_base = NULL;
}
- m->front.iov_len = front_len;
+ m->front_alloc_len = m->front.iov_len = front_len;
dout("ceph_msg_new %p front %d\n", m, front_len);
return m;
@@ -3256,10 +3272,7 @@ static int ceph_con_in_msg_alloc(struct ceph_connection *con, int *skip)
void ceph_msg_kfree(struct ceph_msg *m)
{
dout("msg_kfree %p\n", m);
- if (m->front_is_vmalloc)
- vfree(m->front.iov_base);
- else
- kfree(m->front.iov_base);
+ ceph_kvfree(m->front.iov_base);
kmem_cache_free(ceph_msg_cache, m);
}
@@ -3301,8 +3314,8 @@ EXPORT_SYMBOL(ceph_msg_last_put);
void ceph_msg_dump(struct ceph_msg *msg)
{
- pr_debug("msg_dump %p (front_max %d length %zd)\n", msg,
- msg->front_max, msg->data_length);
+ pr_debug("msg_dump %p (front_alloc_len %d length %zd)\n", msg,
+ msg->front_alloc_len, msg->data_length);
print_hex_dump(KERN_DEBUG, "header: ",
DUMP_PREFIX_OFFSET, 16, 1,
&msg->hdr, sizeof(msg->hdr), true);