On Tue, Jun 11, 2024 at 07:38:37PM -0700, Jakub Kicinski wrote: > On Mon, 10 Jun 2024 17:37:12 +0200 Larysa Zaremba wrote: > > Fix the problems that are triggered by tx_timeout and ice_xdp() calls, > > including both pool and program operations. > > Is there really no way for ice to fix the locking? :( > The busy loops and trylocks() are not great, and seem like duct tape. >
The locking mechanisms I use here do not look pretty, but if I am not missing anything, the synchronization they provide must be robust. A prettier way of protecting the same critical sections would be replacing ICE_CFG_BUSY around ice_vsi_rebuild() with rtnl_lock(), this would eliminate locking code from .ndo_bpf() altogether, ice_rebuild_pending() logic will have to stay. At some point I have decided to avoid using rtnl_lock(), if I do not have to. I think this is a goal worth pursuing?
