Hello! > Do the semantics (I'm not talking about bugs) allow skb passed > to dev->hard_header() (if defined)
No. dev->hard_header() should get enough of space, which is dev->hard_header_len. Actually, it is historical hole in design, inherited from ancient times. Calling conventions of dev->hard_header() just did not allow to reallocate. BTW in 2.6 it can, if it uses pskb_expand_head(). > and then to dev->hard_start_xmit() > to have less link layer header space than dev->hard_header_len? Absolutely. It used to happen all the time. All those devices, which occasionally forget to check for space must be fixed. > I.e., is dev->hard_header_len only advisory? For initial allocator it is an advice. For layers, which add something at head, it is just nothing, if there is enough space. And it is again an advice, when skb is reallocated. > Anyway, the issue with kernel panic is real so I think we better > fix it before 2.6.18, and propagate to "stable" series as well. :-) Know what? This problem followed us since prehistoric times. It happened in 2.4-stablest, 2.2-stable, 2.0... The same devices, the same problem, no matter how much of space it is given to them, they managed to find a hole and to crash. :-) Alexey - To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
