On Tue, 15 Jan 2019 21:00:12 +0100, Andrew Lunn wrote:
> > Ack, will do. IMHO it's a bit of an overkill, one could "safely
> > assume" serial number won't be longer than 32 bytes.. :) But no
> > problem, will do!
>
> A git hash is 40 bytes.
It's 20 bytes, this is in binary form. Plus serial
> Ack, will do. IMHO it's a bit of an overkill, one could "safely
> assume" serial number won't be longer than 32 bytes.. :) But no
> problem, will do!
A git hash is 40 bytes.
Andrew
On Tue, 15 Jan 2019 11:15:12 +0100, Jiri Pirko wrote:
> Tue, Jan 15, 2019 at 01:50:03AM CET, jakub.kicin...@netronome.com wrote:
> >+static int devlink_nl_info_sn_fill(struct sk_buff *msg, struct devlink
> >*devlink,
> >+ struct netlink_ext_ack *extack)
> >+{
> >+
Tue, Jan 15, 2019 at 01:50:03AM CET, jakub.kicin...@netronome.com wrote:
[...]
>+static int devlink_nl_info_sn_fill(struct sk_buff *msg, struct devlink
>*devlink,
>+ struct netlink_ext_ack *extack)
>+{
>+ unsigned char sn[32];
:/ Not good to have number dire
ethtool -i has served us well for a long time, but its showing
its limitations more and more. The device information should
also be reported per device not per-netdev.
Lay foundation for a simple devlink-based ethtool -i replacement.
Add device serial number as initial piece of information expose
