[Linux-cluster] GFS umount hang after heavy read/write from NFS client

[孙俊伟]

>> 	.concurrently read from NFS
>
>What do you mean by this?
My test program is something like iozone, but is multithreading.
On each thread, it reads one file with speed limit about 1MB/s.
And all threads start reading at the same time, so I refer it as concurrent read.

By the way, in spite of how I get it,
every time the umount hangs, I get the same message as:
Glock (2, 26) <<the same
  gl_flags =
  gl_count = 2
  gl_state = 0
  req_gh = no
  req_bh = no
  lvb_count = 0
  object = yes
  new_le = no
  incore_le = no
  reclaim = no
  aspace = 0
  ail_bufs = no
  Inode:
    num = 26/26
    type = 2
    i_count = 1
    i_flags =
    vnode = yes
Glock (5, 26) <<the same
  gl_flags =
  gl_count = 2
  gl_state = 3
  req_gh = no
  req_bh = no
  lvb_count = 0
  object = yes
  new_le = no
  incore_le = no
  reclaim = no
  aspace = no
  ail_bufs = no
  Holder
    owner = -1
    gh_state = 3
    gh_flags = 5 7
    error = 0
    gh_iflags = 1 6 7