|
From: | address@hidden |
Subject: | Re: [lwip-users] low level output question |
Date: | Sat, 09 May 2009 09:09:11 +0200 |
User-agent: | Thunderbird 2.0.0.21 (Macintosh/20090302) |
John Kennedy wrote:
Hm, that does sound strange. Without breakpoints, the packets shown in wireshark are correct? Event the ping responses? If so, I don't have a clue what's going wrong: there should be no packets with lenght=1518 involved in these packets. And 1518 is even too long for an ethernet packet, 1514 should be the maximum (if you didn't misconfigured lwIP).Yes, I'm sending a ping from my PC to my embedded system running lwip. With no breakpoints set wireshark shows an ARP request going out and the ARP response returning from lwip. Then the PC sends an ICMP (ping request) to the embedded system. So the ARP response packet generated by lwip is correct. I was just puzzled because when I set the breakpoint the pbuf length is 1518?
I think the pbufs in the pool (PBUF_POOL type) have their length set to the maximum (which is 1518 because of memory alignment) if they are not used (i.e. currently *in* the pool). Could it be that you are seeing such a pbuf pointer? You didn't yet write what's in the pbuf's payload.
Simon
[Prev in Thread] | Current Thread | [Next in Thread] |