Re: [patch] introducing pfi_kif_check() to better support if_pfsync.c

Hello Richard, thanks for taking a look at this. I'm not sure. ioctl() basically handles the insertion/load of rules. Rule might be actually pre-loaded before desired interface will come alive. So in general ioctl() always want to make sure the pfi_kif interface object referenced by rule exists

[patch] introducing pfi_kif_check() to better support if_pfsync.c

Hello, While reviewing patch, which introduces MP-friendly reference counting for pfi_kif objects (PF stuff), we've found some 'interesting detail' in if_pfsync.c, which I think should get addressed by extra patch. It looks like if_pfsync.c module does not realize, the pfi_kif_get() actually alw