nfs: don't sleep with inode lock in lock_and_join_requests

This handles the 'nonblock=false' case in nfs_lock_and_join_requests.
If the group is already locked and blocking is allowed, drop the inode lock
and wait for the group lock to be cleared before trying it all again.
This should fix warnings found in peterz's tree (sched/wait branch), where
might_sleep() checks are added to wait.[ch].

Reported-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Weston Andros Adamson <dros@primarydata.com>
Reviewed-by: Peng Tao <tao.peng@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>

3 files changed, 29 insertions, 1 deletions

diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c
index 30c9626f96b0..4ec67f8d70aa 100644
--- a/fs/nfs/pagelist.c
+++ b/fs/nfs/pagelist.c
@@ -168,6 +168,23 @@ nfs_page_group_lock(struct nfs_page *req, bool nonblock)
 }
 
 /*
+ * nfs_page_group_lock_wait - wait for the lock to clear, but don't grab it
+ * @req - a request in the group
+ *
+ * This is a blocking call to wait for the group lock to be cleared.
+ */
+void
+nfs_page_group_lock_wait(struct nfs_page *req)
+{
+        struct nfs_page *head = req->wb_head;
+
+        WARN_ON_ONCE(head != head->wb_head);
+
+        wait_on_bit(&head->wb_flags, PG_HEADLOCK,
+                TASK_UNINTERRUPTIBLE);
+}
+
+/*
  * nfs_page_group_unlock - unlock the head of the page group
  * @req - request in group that is to be unlocked
  */
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index e056f617adf2..175d5d073ccf 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -478,13 +478,23 @@ try_again:
                 return NULL;
         }
 
-        /* lock each request in the page group */
+        /* holding inode lock, so always make a non-blocking call to try the
+         * page group lock */
         ret = nfs_page_group_lock(head, true);
         if (ret < 0) {
                 spin_unlock(&inode->i_lock);
+
+                if (!nonblock && ret == -EAGAIN) {
+                        nfs_page_group_lock_wait(head);
+                        nfs_release_request(head);
+                        goto try_again;
+                }
+
                 nfs_release_request(head);
                 return ERR_PTR(ret);
         }
+
+        /* lock each request in the page group */
         subreq = head;
         do {
                 /*
diff --git a/include/linux/nfs_page.h b/include/linux/nfs_page.h
index 6ad2bbcad405..6c3e06ee2fb7 100644
--- a/include/linux/nfs_page.h
+++ b/include/linux/nfs_page.h
@@ -123,6 +123,7 @@ extern  int nfs_wait_on_request(struct nfs_page *);
 extern  void nfs_unlock_request(struct nfs_page *req);
 extern  void nfs_unlock_and_release_request(struct nfs_page *);
 extern int nfs_page_group_lock(struct nfs_page *, bool);
+extern void nfs_page_group_lock_wait(struct nfs_page *);
 extern void nfs_page_group_unlock(struct nfs_page *);
 extern bool nfs_page_group_sync_on_bit(struct nfs_page *, unsigned int);