Merge tag 'ceph-for-4.21-rc1' of git://github.com/ceph/ceph-client

Pull ceph updates from Ilya Dryomov:
 "A fairly quiet round: a couple of messenger performance improvements
  from myself and a few cap handling fixes from Zheng"

* tag 'ceph-for-4.21-rc1' of git://github.com/ceph/ceph-client:
  ceph: don't encode inode pathes into reconnect message
  ceph: update wanted caps after resuming stale session
  ceph: skip updating 'wanted' caps if caps are already issued
  ceph: don't request excl caps when mount is readonly
  ceph: don't update importing cap's mseq when handing cap export
  libceph: switch more to bool in ceph_tcp_sendmsg()
  libceph: use MSG_SENDPAGE_NOTLAST with ceph_tcp_sendpage()
  libceph: use sock_no_sendpage() as a fallback in ceph_tcp_sendpage()
  libceph: drop last_piece logic from write_partial_message_data()
  ceph: remove redundant assignment
  ceph: cleanup splice_dentry()

1 files changed, 21 insertions, 34 deletions

diff --git a/net/ceph/messenger.c b/net/ceph/messenger.c
index 2f126eff275d..d5718284db57 100644
--- a/net/ceph/messenger.c
+++ b/net/ceph/messenger.c
@@ -544,7 +544,7 @@ static int ceph_tcp_recvpage(struct socket *sock, struct page *page,
  * shortly.
  */
 static int ceph_tcp_sendmsg(struct socket *sock, struct kvec *iov,
-                     size_t kvlen, size_t len, int more)
+                            size_t kvlen, size_t len, bool more)
 {
         struct msghdr msg = { .msg_flags = MSG_DONTWAIT | MSG_NOSIGNAL };
         int r;
@@ -560,24 +560,15 @@ static int ceph_tcp_sendmsg(struct socket *sock, struct kvec *iov,
         return r;
 }
 
-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);
-        int ret;
-
-        ret = kernel_sendpage(sock, page, offset, size, flags);
-        if (ret == -EAGAIN)
-                ret = 0;
-
-        return ret;
-}
-
+/*
+ * @more: either or both of MSG_MORE and MSG_SENDPAGE_NOTLAST
+ */
 static int ceph_tcp_sendpage(struct socket *sock, struct page *page,
-                     int offset, size_t size, bool more)
+                             int offset, size_t size, int more)
 {
-        struct msghdr msg = { .msg_flags = MSG_DONTWAIT | MSG_NOSIGNAL };
-        struct bio_vec bvec;
+        ssize_t (*sendpage)(struct socket *sock, struct page *page,
+                            int offset, size_t size, int flags);
+        int flags = MSG_DONTWAIT | MSG_NOSIGNAL | more;
         int ret;
 
         /*
@@ -589,19 +580,11 @@ static int ceph_tcp_sendpage(struct socket *sock, struct page *page,
          * triggers one of hardened usercopy checks.
          */
         if (page_count(page) >= 1 && !PageSlab(page))
-                return __ceph_tcp_sendpage(sock, page, offset, size, more);
-
-        bvec.bv_page = page;
-        bvec.bv_offset = offset;
-        bvec.bv_len = size;
-
-        if (more)
-                msg.msg_flags |= MSG_MORE;
+                sendpage = sock->ops->sendpage;
         else
-                msg.msg_flags |= MSG_EOR;  /* superfluous, but what the hell */
+                sendpage = sock_no_sendpage;
 
-        iov_iter_bvec(&msg.msg_iter, WRITE, &bvec, 1, size);
-        ret = sock_sendmsg(sock, &msg);
+        ret = sendpage(sock, page, offset, size, flags);
         if (ret == -EAGAIN)
                 ret = 0;
 
@@ -1572,6 +1555,7 @@ static int write_partial_message_data(struct ceph_connection *con)
         struct ceph_msg *msg = con->out_msg;
         struct ceph_msg_data_cursor *cursor = &msg->cursor;
         bool do_datacrc = !ceph_test_opt(from_msgr(con->msgr), NOCRC);
+        int more = MSG_MORE | MSG_SENDPAGE_NOTLAST;
         u32 crc;
 
         dout("%s %p msg %p\n", __func__, con, msg);
@@ -1592,7 +1576,6 @@ static int write_partial_message_data(struct ceph_connection *con)
                 struct page *page;
                 size_t page_offset;
                 size_t length;
-                bool last_piece;
                 int ret;
 
                 if (!cursor->resid) {
@@ -1600,10 +1583,11 @@ static int write_partial_message_data(struct ceph_connection *con)
                         continue;
                 }
 
-                page = ceph_msg_data_next(cursor, &page_offset, &length,
-                                          &last_piece);
-                ret = ceph_tcp_sendpage(con->sock, page, page_offset,
-                                        length, !last_piece);
+                page = ceph_msg_data_next(cursor, &page_offset, &length, NULL);
+                if (length == cursor->total_resid)
+                        more = MSG_MORE;
+                ret = ceph_tcp_sendpage(con->sock, page, page_offset, length,
+                                        more);
                 if (ret <= 0) {
                         if (do_datacrc)
                                 msg->footer.data_crc = cpu_to_le32(crc);
@@ -1633,13 +1617,16 @@ static int write_partial_message_data(struct ceph_connection *con)
  */
 static int write_partial_skip(struct ceph_connection *con)
 {
+        int more = MSG_MORE | MSG_SENDPAGE_NOTLAST;
         int ret;
 
         dout("%s %p %d left\n", __func__, con, con->out_skip);
         while (con->out_skip > 0) {
                 size_t size = min(con->out_skip, (int) PAGE_SIZE);
 
-                ret = ceph_tcp_sendpage(con->sock, zero_page, 0, size, true);
+                if (size == con->out_skip)
+                        more = MSG_MORE;
+                ret = ceph_tcp_sendpage(con->sock, zero_page, 0, size, more);
                 if (ret <= 0)
                         goto out;
                 con->out_skip -= ret;