[dpdk-dev] [Bug 628] Possible deadlock in testpmd

[dpdk-dev] [Bug 628] Possible deadlock in testpmd

[bugzilla]

https://bugs.dpdk.org/show_bug.cgi?id=628

            Bug ID: 628
           Summary: Possible deadlock in testpmd
           Product: DPDK
           Version: 20.08
          Hardware: x86
                OS: All
            Status: UNCONFIRMED
          Severity: normal
          Priority: High
         Component: testpmd
          Assignee: dev@dpdk.org
          Reporter: maxime.coquelin@redhat.com
                CC: thomas@monjalon.net, xianghao3@jd.com
        Depends on: 611
  Target Milestone: ---

+++ This bug was initially created as a clone of Bug #611 +++

Recently we see a deadlock in vdpa example app if SIGTERM signal is sent at
early start stage.After debuging, we suspect it's due to vhost_user.mutex is
requested twice,once by start sequence start_vdpa(), the other by SIGTERM
signal handler which calling close_vdpa().

Also checked other example apps, seems they all have same issue.

+++

Testpmd seems also impacted. Indeed, pmd_test_exit() gets called, which turns
out to call dev_stop() and dev_close() ethdev ops that almost all take spinlock
or mutexes.


Referenced Bugs:

https://bugs.dpdk.org/show_bug.cgi?id=611
[Bug 611] deadlock in example vdpa app
-- 
You are receiving this mail because:
You are the assignee for the bug.