On 31 May 2019, at 4:49, Björn Töpel wrote:
> On 2019-05-30 20:57, Jonathan Lemon wrote:
>> Currently, the AF_XDP code uses a separate map in order to
>> determine if an xsk is bound to a queue. Instead of doing this,
>> have bpf_map_lookup_elem() return the queue_id, as a way of
>> indicating that there is a valid entry at the map index.
>>
>> Rearrange some xdp_sock members to eliminate structure holes.
>>
>> Signed-off-by: Jonathan Lemon <jonathan.lemon@xxxxxxxxx>
>> ---
>> include/net/xdp_sock.h | 6 +++---
>> kernel/bpf/verifier.c | 6 +++++-
>> kernel/bpf/xskmap.c | 4 +++-
>> .../selftests/bpf/verifier/prevent_map_lookup.c | 15 ---------------
>> 4 files changed, 11 insertions(+), 20 deletions(-)
>>
>> diff --git a/include/net/xdp_sock.h b/include/net/xdp_sock.h
>> index d074b6d60f8a..7d84b1da43d2 100644
>> --- a/include/net/xdp_sock.h
>> +++ b/include/net/xdp_sock.h
>> @@ -57,12 +57,12 @@ struct xdp_sock {
>> struct net_device *dev;
>> struct xdp_umem *umem;
>> struct list_head flush_node;
>> - u16 queue_id;
>> - struct xsk_queue *tx ____cacheline_aligned_in_smp;
>> - struct list_head list;
>> + u32 queue_id;
>
> Why the increase of size?
Currently xskmaps are defined as size=4, so the returned pointer for
bpf must have that many bytes. I was kicking around the idea of using
a union and returning { qid, zc }, but Alexei pointed out that doesn't
make a good API.
Besides, queue_index in struct xdp_rxq_info is already a u32.
--
Jonathan
