hello, for sometime now i've been repeatedly annoyed by the fact that 10G interfaces lie about their ifi_baudrate. i would like to propose simple (hopefuly) change to address this. quick summary of the problem: struct if_data { ... u_char ifi_spare_char1; /* spare byte */ u_char ifi_spare_char2; /* spare byte */ ... u_long ifi_baudrate; /* linespeed */ ... }; as you can see ifi_baudrate is an u_long which is an arch specific type. on 32-bit arch it does not have enough bits to hold 10G line speed value (in bits per second) proposal we reuse one of the ifi_spare_char1 or ifi_spare_char2 bytes and re-purpose it as power factor to be applied to ifi_baudrate, i.e. real_ifi_baudrate = ifi_baudrate * 10 ** ifi_spare_char1 obviously, 10G nic drivers will have to set ifi_spare_char1 to appropriate value, but it should not be a big deal. also, legacy tools that do not know about ifi_spare_char1 would continue to report "wrong" ifi_baudrate as they used to. any objections? thanks, maxReceived on Wed Sep 19 2012 - 19:16:18 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:40:30 UTC