Re: [Patch net-next v2] net_sched: move the empty tp check from ->destroy() to ->delete()

2017-04-13 Thread Cong Wang
On Thu, Apr 13, 2017 at 12:28 AM, kbuild test robot wrote: > Hi Cong, > > [auto build test WARNING on net-next/master] > > url: > https://github.com/0day-ci/linux/commits/Cong-Wang/net_sched-move-the-empty-tp-check-from-destroy-to-delete/20170413-145318 > config: x86_64-randconfig-x004-201715

Re: [Patch net-next v2] net_sched: move the empty tp check from ->destroy() to ->delete()

2017-04-13 Thread kbuild test robot
Hi Cong, [auto build test WARNING on net-next/master] url: https://github.com/0day-ci/linux/commits/Cong-Wang/net_sched-move-the-empty-tp-check-from-destroy-to-delete/20170413-145318 config: x86_64-randconfig-x004-201715 (attached as .config) compiler: gcc-6 (Debian 6.2.0-3) 6.2.0 20160901 re

[Patch net-next v2] net_sched: move the empty tp check from ->destroy() to ->delete()

2017-04-12 Thread Cong Wang
Roi reported we could have a race condition where in ->classify() path we dereference tp->root and meanwhile a parallel ->destroy() makes it a NULL. This is possible because ->destroy() could be called when deleting a filter to check if we are the last one in tp, this tp is still linked and visibl