Re: [PATCH bpf-next] bpf/devmap: convert remaining READ_ONCE() to rcu_dereference()

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Martin KaFai Lau <kafai@xxxxxx> writes:

> On Tue, Jun 29, 2021 at 01:00:51AM +0200, Toke Høiland-Jørgensen wrote:
>> There were a couple of READ_ONCE()-invocations left-over by the devmap RCU
>> conversion. Convert these to rcu_dereference() as well to avoid complaints
>> from sparse.
>> 
>> Fixes: 782347b6bcad ("xdp: Add proper __rcu annotations to redirect map entries")
>> Reported-by: kernel test robot <lkp@xxxxxxxxx>
>> Signed-off-by: Toke Høiland-Jørgensen <toke@xxxxxxxxxx>
>> ---
>>  kernel/bpf/devmap.c | 4 ++--
>>  1 file changed, 2 insertions(+), 2 deletions(-)
>> 
>> diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
>> index 2f6bd75cd682..7a0c008f751b 100644
>> --- a/kernel/bpf/devmap.c
>> +++ b/kernel/bpf/devmap.c
>> @@ -558,7 +558,7 @@ int dev_map_enqueue_multi(struct xdp_buff *xdp, struct net_device *dev_rx,
>>  
>>  	if (map->map_type == BPF_MAP_TYPE_DEVMAP) {
>>  		for (i = 0; i < map->max_entries; i++) {
>> -			dst = READ_ONCE(dtab->netdev_map[i]);
>> +			dst = rcu_dereference(dtab->netdev_map[i]);
> __dev_map_lookup_elem() uses rcu_dereference_check(dtab->netdev_map[key], rcu_read_lock_bh_held()).
> It is not needed here?

Hmm, I somehow managed to convince myself last night that it wasn't, but
looking at it again now I think you're right. Will send a v2.

-Toke





[Index of Archives]     [Linux Samsung SoC]     [Linux Rockchip SoC]     [Linux Actions SoC]     [Linux for Synopsys ARC Processors]     [Linux NFS]     [Linux NILFS]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]


  Powered by Linux