Re: [lwip-devel] lwip1.4.0 "netif->hwaddr_len must be the same as ETHARP_HWADDR_LEN for etharp!" failed

[Grant Erickson]

On Oct 9, 2013, at 9:00 AM, address@hidden wrote:
> Date: Wed, 09 Oct 2013 09:33:15 +0800
> From: shenhuan <address@hidden>
> To: address@hidden
> Subject: [lwip-devel] lwip1.4.0 "netif->hwaddr_len must be the same as
>       ETHARP_HWADDR_LEN for etharp!" failed
> Message-ID: <address@hidden>
> Content-Type: text/plain; charset=GB2312
> 
> Hi Everyone!
> 
> Now I use lwip(1.4.0) on Micorblaze(Spartan6) platform,and use network
> device DM9000BI as network interface.
> Lwip run with uCOS-II(2.9.0).
> When I test the speed(TCP/IP) of this system, after a moment past,
> assertion happen =><"netif->hwaddr_len must be the same as
> ETHARP_HWADDR_LEN for etharp!" failed at line 1148 in
> ../src/LWIP/src/netif/etharp.c>.
> Test mode => Continue send data packet from PC to my platform
> 
> Is this the bug of 1.4.0 of LWIP core???
> 
> Someone know something about it,please help me...
> 
> Thanks
> 
> Pofeng
> 
> ------------------------------
> Date: Wed, 9 Oct 2013 07:58:22 -0400
> From: "Bill Auerbach" <address@hidden>
> To: "'lwip-devel'" <address@hidden>
> Subject: Re: [lwip-devel] lwip1.4.0 "netif->hwaddr_len must be the
>       same as ETHARP_HWADDR_LEN for etharp!" failed
> Message-ID: <address@hidden>
> Content-Type: text/plain;     charset="us-ascii"
> 
> Application code (i.e., your driver) is responsible for setting hwaddr_len.
> Check that it is.
> 
> -----Original Message-----
> From: address@hidden
> [mailto:address@hidden On Behalf
> Of shenhuan
> Sent: Tuesday, October 08, 2013 9:33 PM
> To: address@hidden
> Subject: [lwip-devel] lwip1.4.0 "netif->hwaddr_len must be the same as
> ETHARP_HWADDR_LEN for etharp!" failed
> 
> Hi Everyone!
> 
> Now I use lwip(1.4.0) on Micorblaze(Spartan6) platform,and use network
> device DM9000BI as network interface.
> Lwip run with uCOS-II(2.9.0).
> When I test the speed(TCP/IP) of this system, after a moment past, assertion
> happen =><"netif->hwaddr_len must be the same as ETHARP_HWADDR_LEN for
> etharp!" failed at line 1148 in ../src/LWIP/src/netif/etharp.c>.
> Test mode => Continue send data packet from PC to my platform
> 
> Is this the bug of 1.4.0 of LWIP core???
> 
> Someone know something about it,please help me...
> 
> Thanks
> 
> Pofeng

Shenhuan:

As Bill indicated, it is up to drivers to initialize this value. However, you 
may be interested in the following patches if your length is not 6:

        https://savannah.nongnu.org/bugs/?38748
        
Best,

Grant