Re: [lwip-users] arp question

[Jim Gibbons]

Leon,

I've just noticed your recent flurry of activity.  I know that this must 
take a lot of your time.   I would like to express the gratitude of the 
community for your efforts and those of the other developers.

In the matter of ARP, I think some concurrent access protection might be 
in order in common use scenarios.  Perhaps it should be noted in 
comments so that callers could do it, since the callers are usually 
netif implementors.

I agree that it is a simple matter to get etharp_output and the timer to 
run out of the tcpip task.  I think it is substantially harder to get 
etharp_ip_input and etharp_arp_input to run in that context.  Most of us 
who implement netif's in multitasking environments end up with an 
ethernet receive thread from which the etharp input functions must be 
called.  Other brave souls do this from an interrupt context.  In the 
former case, a mutex would provide the protection required, while in the 
latter case SYS_LIGHTWEIGHT_PROT methods might be used to protect the 
timer and output.

Finally, a very broad question.  Is 1.0.0 ready for prime time?  I have 
two embedded systems currently running 0.7.2, and it has been hard for 
me to tell from all of the e-mail going by whether 1.0.0 would prove to 
be more stable or less so for these products.

Thanks again for all your efforts.

Leon Woestenberg wrote:

> Jim,
>
> ARP is not protected against concurrent access.
>
> Calling the etharp_timer should be possible from within the
> TCP/IP thread though.
>
> I have suggested earlier to rename the tcpip thread name to
> something more generic lwip_timer_thread or so, and do all
> the timeout involved actions there.
>
> Regards,
>
> Leon.
>
>
> _______________________________________________
> lwip-users mailing list
> address@hidden
> http://lists.nongnu.org/mailman/listinfo/lwip-users
>