On 4/11/25 19:54, Jakub Sitnicki wrote: > On Fri, Apr 11, 2025 at 09:09 PM +08, Jiayuan Chen wrote: >> On Fri, Apr 11, 2025 at 01:32:41PM +0200, Michal Luczaj wrote: >>> Test redirection logic. All supported and unsupported redirect combinations >>> are tested for success and failure respectively. >>> >>> BPF_MAP_TYPE_SOCKMAP >>> BPF_MAP_TYPE_SOCKHASH >>> x >>> sk_msg-to-egress >>> sk_msg-to-ingress >>> sk_skb-to-egress >>> sk_skb-to-ingress >> >> Could we also add test cases for SK_PASS (and even SK_DROP)? >> Previously, we encountered deadlocks and incorrect sequence issues when >> the program returned SK_PASS, so explicit testing for these cases would >> be helpful. >> >> If implemented, this test would fully exercise all code paths and >> demonstrate a complete example that covers every aspect of >> sockmap's packet steering and connection management capabilities. > > This could easily be a follow up in my mind. > > [...]
Yeah, I wouldn't mind doing this in multiple steps. That said, with SK_PASS/SK_DROP involved, are we sticking with the name "sockmap_redir"?
