[Qemu-devel] [Bug 1634852] Re: Qemu VirtFS mounts are not accessible after resuming guest from hibernation

[Greg Kurz]

The root cause lies in the 9pnet_virtio driver in the guest kernel: it
does not support suspend/hibernation...

** Project changed: qemu => linux

-- 
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.launchpad.net/bugs/1634852

Title:
  Qemu VirtFS mounts are not accessible after resuming guest from
  hibernation

Status in Linux:
  New

Bug description:
  Host OS:  Funtoo Linux
  Host Kernel:  4.7.4-gentoo
  Qemu Version:  2.7.0
  Guest OS:  Ubuntu 14.04
  Guest Kernel:  reproduced with both 4.4.0-42-generic and 3.13.0-98-generic

  Qemu command line:

  qemu-system-x86_64 \
      -machine type=pc,accel=kvm \
      -cpu host \
      -smp 3 \
      -m 8G \
      -netdev bridge,id=hn0,vhost=on \
      -device virtio-net-pci,netdev=hn0,mac=52:54:fa:70:35:f7 \
      -drive 
file=/dev/mapper/vg-ubuntu,format=raw,if=virtio,cache=none,discard=on \
      -virtfs 
local,path=/home/dharding/code,security_model=passthrough,mount_tag=code \
      -virtfs 
local,path=/home/dharding/xfer,security_model=passthrough,mount_tag=xfer \
      -display gtk

  
  Relevant lines from guest /etc/fstab:
  code /home/dharding/code 9p 
trans=virtio,version=9p2000.L,msize=262144,_netdev 0 0
  xfer /home/dharding/xfer 9p 
trans=virtio,version=9p2000.L,msize=262144,_netdev 0 0

  
  Steps to reproduce:
  - start qemu using the above command line
  - in the guest, run "sudo pm-hibernate"
  - after qemu exits, run again using the same command line
  - once the guest resumes from hibernation, run "ls /home/dharding/code"
  - the ls command will hang forever

  The ls call stack is:

  [<ffffffffc00743a0>] p9_client_rpc+0x110/0x460 [9pnet]
  [<ffffffffc0076a50>] p9_client_getattr_dotl+0x60/0x160 [9pnet]
  [<ffffffffc009ef77>] v9fs_vfs_getattr_dotl+0x47/0xa0 [9p]
  [<ffffffff81202a5c>] vfs_getattr_nosec+0x2c/0x40
  [<ffffffff81202b26>] vfs_getattr+0x26/0x30
  [<ffffffff81202bf5>] vfs_fstatat+0x65/0xa0
  [<ffffffff8120306f>] SYSC_newstat+0x1f/0x40
  [<ffffffff812032be>] SyS_newstat+0xe/0x10
  [<ffffffff817fa4f6>] entry_SYSCALL_64_fastpath+0x16/0x75

To manage notifications about this bug go to:
https://bugs.launchpad.net/linux/+bug/1634852/+subscriptions