Re: [lwip-devel] [bug #43028] IP_MULTICAST_TTL affects unicast datagrams

[Joel Cunningham]

Simon,

I'm not all that familiar with how often new members get added to the pcb, but is it out of the question to add another u8_t for mcast_ttl guarded by #if LWIP_IGMP?

If the bug isn't addressed, then applications are just going to end up having to create a second socket, possibly with SO_REUSEADDR (_PORT) depending on use case, to have unicast and multicast senders with different TTLs.  This seems like it will consume more resources for that use case than u8_t

I can put a patch together for adding a u8_t for multicast TTL if that's an acceptable approach

Thanks,

Joel

On Aug 20, 2014, at 02:49 PM, Simon Goldschmidt <address@hidden> wrote:

Follow-up Comment #1, bug #43028 (project lwip):

The report seems correct, but do you have an idea of how to fix? We don't have
more than one place to store TTL, so this is a bug I'm tempted to accept for
staying lightweight...

_______________________________________________________

Reply to this item at:

<http://savannah.nongnu.org/bugs/?43028        >

_______________________________________________
Nachricht gesendet von/durch Savannah
http://savannah.nongnu.org/