xen/manage: Fix USB interaction issues when resuming

commit 72978b2fe2f2cdf9f319c6c6dcdbe92b38de2be2 upstream.

Commit 61a734d305e1 ("xen/manage: Always freeze/thaw processes when
suspend/resuming") ensured that userspace processes were always frozen
before suspending to reduce interaction issues when resuming devices.
However, freeze_processes() does not freeze kernel threads.  Freeze
kernel threads as well to prevent deadlocks with the khubd thread when
resuming devices.

This is what native suspend and resume does.

Example deadlock:
[ 7279.648010]  [<ffffffff81446bde>] ? xen_poll_irq_timeout+0x3e/0x50
[ 7279.648010]  [<ffffffff81448d60>] xen_poll_irq+0x10/0x20
[ 7279.648010]  [<ffffffff81011723>] xen_lock_spinning+0xb3/0x120
[ 7279.648010]  [<ffffffff810115d1>] __raw_callee_save_xen_lock_spinning+0x11/0x20
[ 7279.648010]  [<ffffffff815620b6>] ? usb_control_msg+0xe6/0x120
[ 7279.648010]  [<ffffffff81747e50>] ? _raw_spin_lock_irq+0x50/0x60
[ 7279.648010]  [<ffffffff8174522c>] wait_for_completion+0xac/0x160
[ 7279.648010]  [<ffffffff8109c520>] ? try_to_wake_up+0x2c0/0x2c0
[ 7279.648010]  [<ffffffff814b60f2>] dpm_wait+0x32/0x40
[ 7279.648010]  [<ffffffff814b6eb0>] device_resume+0x90/0x210
[ 7279.648010]  [<ffffffff814b7d71>] dpm_resume+0x121/0x250
[ 7279.648010]  [<ffffffff8144c570>] ? xenbus_dev_request_and_reply+0xc0/0xc0
[ 7279.648010]  [<ffffffff814b80d5>] dpm_resume_end+0x15/0x30
[ 7279.648010]  [<ffffffff81449fba>] do_suspend+0x10a/0x200
[ 7279.648010]  [<ffffffff8144a2f0>] ? xen_pre_suspend+0x20/0x20
[ 7279.648010]  [<ffffffff8144a1d0>] shutdown_handler+0x120/0x150
[ 7279.648010]  [<ffffffff8144c60f>] xenwatch_thread+0x9f/0x160
[ 7279.648010]  [<ffffffff810ac510>] ? finish_wait+0x80/0x80
[ 7279.648010]  [<ffffffff8108d189>] kthread+0xc9/0xe0
[ 7279.648010]  [<ffffffff8108d0c0>] ? flush_kthread_worker+0x80/0x80
[ 7279.648010]  [<ffffffff8175087c>] ret_from_fork+0x7c/0xb0
[ 7279.648010]  [<ffffffff8108d0c0>] ? flush_kthread_worker+0x80/0x80

[ 7441.216287] INFO: task khubd:89 blocked for more than 120 seconds.
[ 7441.219457]       Tainted: G            X 3.13.11-ckt12.kz #1
[ 7441.222176] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[ 7441.225827] khubd           D ffff88003f433440     0    89      2 0x00000000
[ 7441.229258]  ffff88003ceb9b98 0000000000000046 ffff88003ce83000 0000000000013440
[ 7441.232959]  ffff88003ceb9fd8 0000000000013440 ffff88003cd13000 ffff88003ce83000
[ 7441.236658]  0000000000000286 ffff88003d3e0000 ffff88003ceb9bd0 00000001001aa01e
[ 7441.240415] Call Trace:
[ 7441.241614]  [<ffffffff817442f9>] schedule+0x29/0x70
[ 7441.243930]  [<ffffffff81743406>] schedule_timeout+0x166/0x2c0
[ 7441.246681]  [<ffffffff81075b80>] ? call_timer_fn+0x110/0x110
[ 7441.249339]  [<ffffffff8174357e>] schedule_timeout_uninterruptible+0x1e/0x20
[ 7441.252644]  [<ffffffff81077710>] msleep+0x20/0x30
[ 7441.254812]  [<ffffffff81555f00>] hub_port_reset+0xf0/0x580
[ 7441.257400]  [<ffffffff81558465>] hub_port_init+0x75/0xb40
[ 7441.259981]  [<ffffffff814bb3c9>] ? update_autosuspend+0x39/0x60
[ 7441.262817]  [<ffffffff814bb4f0>] ? pm_runtime_set_autosuspend_delay+0x50/0xa0
[ 7441.266212]  [<ffffffff8155a64a>] hub_thread+0x71a/0x1750
[ 7441.268728]  [<ffffffff810ac510>] ? finish_wait+0x80/0x80
[ 7441.271272]  [<ffffffff81559f30>] ? usb_port_resume+0x670/0x670
[ 7441.274067]  [<ffffffff8108d189>] kthread+0xc9/0xe0
[ 7441.276305]  [<ffffffff8108d0c0>] ? flush_kthread_worker+0x80/0x80
[ 7441.279131]  [<ffffffff8175087c>] ret_from_fork+0x7c/0xb0
[ 7441.281659]  [<ffffffff8108d0c0>] ? flush_kthread_worker+0x80/0x80

Signed-off-by: Ross Lagerwall <ross.lagerwall@citrix.com>
Signed-off-by: David Vrabel <david.vrabel@citrix.com>
Signed-off-by: Luis Henriques <luis.henriques@canonical.com>

diff --git a/drivers/xen/manage.c b/drivers/xen/manage.c
index f8bb36f9d9cef862c6c502b5ba25b7d38476cd85..bf1940706422fe1c2dd334cf28096294ab74e542 100644 (file)
--- a/drivers/xen/manage.c
+++ b/drivers/xen/manage.c
@@ -105,10 +105,16 @@ static void do_suspend(void)
 
        err = freeze_processes();
        if (err) {
-               pr_err("%s: freeze failed %d\n", __func__, err);
+               pr_err("%s: freeze processes failed %d\n", __func__, err);
                goto out;
        }
 
+       err = freeze_kernel_threads();
+       if (err) {
+               pr_err("%s: freeze kernel threads failed %d\n", __func__, err);
+               goto out_thaw;
+       }
+
        err = dpm_suspend_start(PMSG_FREEZE);
        if (err) {
                pr_err("%s: dpm_suspend_start %d\n", __func__, err);