Hi dd,

I already found a solution for my problem. I just set the physical address 
to 2 that it fits the condition and I used static ip now instead of dhcp. 
Now I can ping to the beagle and get a response. Before 
"CPSW0_PORT_1_PHY_ADDR" was set to zero. It is not explained anywhere why I 
needed to set it to 2. I was just try and error by myself.

Snippet of "cpswif.h":

#define CPSW0_PORT_1_PHY_ADDR           2
#define CPSW0_PORT_1_PHY_GIGABIT        FALSE

regards,

DS


On Wednesday, February 7, 2018 at 2:05:34 PM UTC+1, [email protected] 
wrote:
>
> Hi dd,
>
> First I want to thank you for your contribution in efficient bare metal 
> programming. I am programming the beaglebone black in bearmetal as well at 
> the moment and I use Starterware 02.00.01.01 as a basis. I tried executing 
> the enetEcho example but I always run into an issue on initialization. I 
> didn't change anything and run the project with DHCP enabled. In function 
> "cpswif_phylink_config" of "cpswif.c" from the LWIP Stack the issue occurs 
> (No PHY found at address 0 for Port 0. MDIOSLAVE register value is "4". Do 
> you have any hints? Is perhaps my gel file already enabling the PHY? 
> Problem is I found no way of disabling the GEL Files in Code Composer 
> Studio. When I remove or comment the files, the IDE ignores that.
>
> Code Snippet:
>
> static err_t
> cpswif_phylink_config(struct cpswportif * cpswif, u32_t slv_port_num) {
>   struct cpswinst *cpswinst = &cpsw_inst_data[cpswif->inst_num];
>   err_t err;
>
>   /* Check if ethernet PHY is present or not */
>   if(0 == (MDIOPhyAliveStatusGet(cpswinst->mdio_base)
>         & (1 << cpswinst->port[slv_port_num - 1].phy_addr))){
>     LWIP_PRINTF("\n\rNo PHY found at address %d for  Port %d of Instance 
> %d.",
>                 cpswinst->port[slv_port_num - 1].phy_addr, slv_port_num,
>                 cpswif->inst_num);
>     return ERR_CONN;
>   }
>
> With kind regards,
>
> DS
>
>
> -----------------------------------------------------------------------------------------------------------
>
> On Tuesday, February 6, 2018 at 10:44:51 AM UTC+1, dd wrote:
>>
>> Hi all.  I sort-of finished the Ethernet driver port from Starterware.   
>> It employs the full FAT32 filesystem by Elm-Chan and lwIP by Adam Dunkels. 
>>  The all the code, makefile stuff, bootloader etc... required to build an 
>> image with interrupts, MMC, TCP, UDP, DMA, Cache, MMU, IRQ, RTC, httpserver 
>> and other peripherals.  Now you can use the BeagleBone to control IoT 
>> projects, your own tight server or whatever.  Written under GNU/GCC, for 
>> the serious noob.  Sink your teeth into ARM Bare Metal, FAT32  and Internet 
>> Protocol here.  The Bone' is very fast and economical when running without 
>> the burden of an insecure Operating System............dd
>>
>> PS  it runs but it needs more streamlining, and I need more help!    
>> www.github.com/ddlawrence    www.baremetal.tech
>>
>> I am an old engineer/programmer that is very tired of running on 
>> dumbed-down blackbox and otherwise limiting operating systems.  Linux is 
>> not the answer.  It should be, but it is too big/complicated. It really has 
>> not improved in 15 years.  It is much more intricate than the ARM hardware 
>> it is running on.  Linus refuses to migrate to microKernel.  So do not get 
>> old and obsolete like me, watching corporations fuddle with fickle tools, 
>> capitalizing Net Neutrality and restricting your freedom as a programmer.  
>> Learn the basics and build upward, on software that YOU know and own.  The 
>> 'Bone has unchallenged speed and I/O.  This is a worthwhile platform to 
>> invest your effort.  The attached zipfile will give you a big head start.  
>>
>

-- 
For more options, visit http://beagleboard.org/discuss
--- 
You received this message because you are subscribed to the Google Groups 
"BeagleBoard" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/beagleboard/4bd7330a-92bb-4fe6-a998-50ebaff09644%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to