On Fri, Jul 25, 2014 at 1:45 PM, Andy Zhou wrote:
> The fix looks good. It would be good to mention that concurrent
> accessing ma->count
> without proper locking is the actual root cause in the commit message.
>
> It would help if we later make more proper fix later on and reference
> back to th
The fix looks good. It would be good to mention that concurrent
accessing ma->count
without proper locking is the actual root cause in the commit message.
It would help if we later make more proper fix later on and reference
back to this commit.
One minor thing: In ovs_flow_tbl_lookup_stats() th
There is race in datapath when last mask in mask array deleted can
result in NULL pointer dereference in datapath.
datapath lookup does not check mask pointer if its index is less than
mask-array count. That is safe because delete operation moves last valid
pointer to the deleted element. But this
