Replying for visibility as open-lldp bounced the original message. On Monday, July 6, 2015 at 5:40:24 PM UTC-7, Anish Bhatt wrote: > > I was trying to use the *sk_cgrp_prioidx* value to find correct dcb > priority for iscsi (using *cgdcbxd* + *cgrulesengd* as recommended by > http://open-lldp.org/dcb_overview). The *sk_cgrp_prioidx* index into * > netprio_map* hangs off of *struck sock*, and the particular iscsi offload > driver I’m working with (cxgb4i) does not actually use struck sock at all. > From a preliminary glance, other open-iscsi offload transports do not seem > to use a *struck sock* either (don’t quote me on this ) leaving them > unable to actually find the priority index into *netprio_map*. > > > > Is there a suggested way for such drivers to use *netprio_map* ? *cgdcbxd*’s > cgroup creation is working fine (using our firmware dcb state machine + > dcbnl_ops, no open-lldp) and I can use I can call *task_netprioidx()* at > connection time provided I use *cgexec* with * iscsid* to get the correct > index into *netprio_map*. Not sure if this is the best approach however, > does anyone have a better suggestion ? Would it make sense for the > *sk_cgrp_prioidx* to hang off the skb or have *libiscsi* provide this > value to open-iscsi transports for a more generic implementation ? > > > > -Anish > > > > One socket to bind them all. > > >
-- You received this message because you are subscribed to the Google Groups "open-iscsi" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/open-iscsi. For more options, visit https://groups.google.com/d/optout.
