On 05/05, Stanislav Fomichev wrote: > On Fri, May 5, 2023 at 11:45 AM Stanislav Fomichev <sdf@xxxxxxxxxx> wrote: > > > > We're observing some stalls on the heavily loaded machines > > in the cgroup_bpf_prog_query path. This is likely due to > > being blocked on cgroup_mutex. > > > > IIUC, the cgroup_mutex is there mostly to protect the non-effective > > fields (cgrp->bpf.progs) which might be changed by the update path. > > For the BPF_F_QUERY_EFFECTIVE case, all we need is to rcu_dereference > > a bunch of pointers (and keep them around for consistency), so > > let's do it. > > > > Sending out as an RFC because it looks a bit ugly. It would also > > be nice to handle non-effective case locklessly as well, but it > > might require a larger rework. > > > > Signed-off-by: Stanislav Fomichev <sdf@xxxxxxxxxx> Ah, that wont work, can't copy to user while holding rcu. Maybe we can even go as far as converting the progs list to rcu.. This should allow as to drop mutex altogether during the query path. As long as we return EAGAIN on racy deatach we should be fine it seems. Will try to play with this idea.
