On Tue, 2020-10-06 at 08:37 +0200, Johannes Berg wrote: > On Mon, 2020-10-05 at 15:21 -0700, Jakub Kicinski wrote: > > > > > Nice, easy & useful, maybe I'll code it up tomorrow. > > > > > > OK I thought about it a bit more and looked at the code, and it's not > > > actually possible to do easily right now, because we can't actually > > > point to the bad attribute from the general lib/nlattr.c code ... > > > > > > Why? Because we don't know right now, e.g. for nla_validate(), where in > > > the message we started validation, i.e. the offset of the "head" inside > > > the particular message. > > > > > > For nlmsg_parse() and friends that's a bit easier, but it needs more > > > rejiggering than I'm willing to do tonight ;) > > > > I thought we'd record the const struct nla_policy *tp for the failing > > attr in struct netlink_ext_ack and output based on that. > > We could, but it's a bit useless if you know "which" attribute caused > the issue, but you don't know where it was in the message? That way you > wouldn't know the nesting level etc. > > I mean, we actually have that problem today - the generic lib/nlattr.c > policy violation doesn't tell you where exactly the problem occurred, so > it'd be good to fix that regardless.
Just for the record: I'm obviously *completely* confused, of course we do say which attribute was bad, I was just looking for the offset, but we carry around a pointer and calculate the offset later, with NL_SET_ERR_MSG_ATTR() or NL_SET_BAD_ATTR(). Which means this isn't so hard ... johannes
