Hello list,

finally the switch in the TP-Link TL-WR741N(D) (and probably other devices 
based von AR7140) is not blocking broadcasts any more and thus allows to get an 
IP via DHCP.

But my question is left to be answered. Does this bridge which is created over 
the LAN-ports really mean that the CPU is doing the bridging?? Performance-wise 
this would be a bad behaviour.

Regards
Dennis

From: [email protected]
To: [email protected]
Date: Mon, 1 Mar 2010 21:47:44 +0100
Subject: Re: [OpenWrt-Devel] TP-Link TL-WR741ND: broadcasts on ethernet not 
reaching CPU








Hello

> From: [email protected]
> 
> I couldn't find any code in svn r19920 dealing specifically with the AR7240 
> and the built-in switch.

Today in revision 19927 switch support for AG7240 has been committed. I'll try 
that. Buildroot is running at the moment.

> 
> So I guess that the bootloader settings of the switch stay valid. In the 
> u-boot sources from the
> TP-Link tar ball you find in board/ar7240/common/ar7240_s26_phy.c, line 345:
> 
>    /* Enable ARP packets to CPU port */
>     
> athrs26_reg_write(S26_ARL_TBL_CTRL_REG,(athrs26_reg_read(S26_ARL_TBL_CTRL_REG)
>  | 0x100000));
> 
> Maybe other broadcast packets need to be enabled here, too?

could be an explanation.

What I understand from the commit 19930 is, that bridging is now done in CPU on 
this switch? Am I right?


> [...]
> Regards,
> Joerg.

Regards,
Dennis
                                          
Herzig: Verschicke fertige  Liebesgrüße als E-Mail!                             
          
_________________________________________________________________
http://redirect.gimas.net/?n=M1002xHMNoSpam2
Keine Lust auf Spam? Hotmail blockiert Spam automatisch
_______________________________________________
openwrt-devel mailing list
[email protected]
https://lists.openwrt.org/mailman/listinfo/openwrt-devel

Reply via email to