diff options
| author |   Jason Gunthorpe <jgg@nvidia.com> | 2020-06-25 20:42:19 +0300 |
|---|---|---|
| committer | Jason Gunthorpe <jgg@nvidia.com> | 2020-07-02 10:46:06 -0300 |
| commit | 65936bf25f90fe440bb2d11624c7d10fab266639 (patch) | |
| tree | 4b8c3bd5d8be4706f6b0bf8797778faadf8bc1b1 /drivers/infiniband/ulp/ipoib/ipoib_main.c | |
| parent | IB/hfi1: Convert PCIBIOS_* errors to generic -E* errors (diff) | |
| download | linux-dev-65936bf25f90fe440bb2d11624c7d10fab266639.tar.xz linux-dev-65936bf25f90fe440bb2d11624c7d10fab266639.zip | |
RDMA/ipoib: Fix ABBA deadlock with ipoib_reap_ah()
ipoib_mcast_carrier_on_task() insanely open codes a rtnl_lock() such that
the only time flush_workqueue() can be called is if it also clears
IPOIB_FLAG_OPER_UP.
Thus the flush inside ipoib_flush_ah() will deadlock if it gets unlucky
enough, and lockdep doesn't help us to find it early:
CPU0 CPU1 CPU2
__ipoib_ib_dev_flush()
down_read(vlan_rwsem)
ipoib_vlan_add()
rtnl_trylock()
down_write(vlan_rwsem)
ipoib_mcast_carrier_on_task()
while (!rtnl_trylock())
msleep(20);
ipoib_flush_ah()
flush_workqueue(priv->wq)
Clean up the ah_reaper related functions and lifecycle to make sense:
- Start/Stop of the reaper should only be done in open/stop NDOs, not in
any other places
- cancel and flush of the reaper should only happen in the stop NDO.
cancel is only functional when combined with IPOIB_STOP_REAPER.
- Non-stop places were flushing the AH's just need to flush out dead AH's
synchronously and ignore the background task completely. It is fully
locked and harmless to leave running.
Which ultimately fixes the ABBA deadlock by removing the unnecessary
flush_workqueue() from the problematic place under the vlan_rwsem.
Fixes: efc82eeeae4e ("IB/ipoib: No longer use flush as a parameter")
Link: https://lore.kernel.org/r/20200625174219.290842-1-kamalheib1@gmail.com
Reported-by: Kamal Heib <kheib@redhat.com>
Tested-by: Kamal Heib <kheib@redhat.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Diffstat (limited to 'drivers/infiniband/ulp/ipoib/ipoib_main.c')
| -rw-r--r-- | drivers/infiniband/ulp/ipoib/ipoib_main.c | 2 |
1 files changed, 2 insertions, 0 deletions
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c index 3cfb682b91b0..ef60e8e4ae67 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_main.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c @@ -1976,6 +1976,8 @@ static void ipoib_ndo_uninit(struct net_device *dev) /* no more works over the priv->wq */ if (priv->wq) { + /* See ipoib_mcast_carrier_on_task() */ + WARN_ON(test_bit(IPOIB_FLAG_OPER_UP, &priv->flags)); flush_workqueue(priv->wq); destroy_workqueue(priv->wq); priv->wq = NULL; |