RE: [ECOS] Still having problems getting networking up

[Gary Thomas <gthomas at cambridge dot redhat dot com>]

On 16-Feb-2001 Lewin A.R.W. Edwards wrote:
> 
>>    Lewin> ecosconfig new edb7xxx
>>     Lewin> ecosconfig add eth_drivers net
>>
>>You should be able to just do "ecosconfig new edb7xxx net" to combine
>>the above two steps, but that is not the problem here.
> 
> Just FYI, it seems that template is currently broken, as eCos won't build 
> successfully with "ecosconfig new edb7xxx net". I'll redo it and send a 
> make.out if you want, but it should be easy to reproduce.
> 

I tried this "six ways from Sunday" against the current [master] sources and 
I could not reproduce your error.

> Anyway, I got everything built OK, but it seems that more configuration is 
> required. When my app starts, it hangs repeating "CS8900: Tx interrupt lost".
> 

If you have an EDB7212 board then you'll need to remove a resistor to get
interrupts to work.  Remove R168 and all jumpers from JP45.

Alternatively, you can try enabling the #define INTS_DONT_WORK in the driver.

> To gather a little more info, I set up a (different from previous) bogus 
> MAC address of 0x00:0x14:0x49:0x18:0x14:0x00 using fconfig, and I enabled 
> boot-time network debugging. Power up and I get several repetitions of this:
> 
> Ethernet send:
> 0x000029E4: FFFF FFFF FFFF 0014  4918 1400 0800        |........I.....  |
> 0x0000C93C: 4500 0148 0002 0000  4011 79A4 0000 0000   |E..H....@.y.....|
> 0x0000C94C: FFFF FFFF 0044 0043  0134 0F6B 0101 0600   |.....D.C.4.k....|
> 0x0000C95C: 5555 3412 0000 0000  0000 0000 0000 0000   |UU4.............|
> 0x0000C96C: 0000 0000 0000 0000  0014 4918 1400 0000   |..........I.....|
> 0x0000C97C: 0000 0000 0000 0000  0000 0000 0000 0000   |................|
> 0x0000C98C: 0000 0000 0000 0000  0000 0000 0000 0000   |................|
> 0x0000C99C: 0000 0000 0000 0000  0000 0000 0000 0000   |................|
> 0x0000C9AC: 0000 0000 0000 0000  0000 0000 0000 0000   |................|
> 0x0000C9BC: 0000 0000 0000 0000  0000 0000 0000 0000   |................|
> 0x0000C9CC: 0000 0000 0000 0000  0000 0000 0000 0000   |................|
> 0x0000C9DC: 0000 0000 0000 0000  0000 0000 0000 0000   |................|
> 0x0000C9EC: 0000 0000 0000 0000  0000 0000 0000 0000   |................|
> 0x0000C9FC: 0000 0000 0000 0000  0000 0000 0000 0000   |................|
> 0x0000CA0C: 0000 0000 0000 0000  0000 0000 0000 0000   |................|
> 0x0000CA1C: 0000 0000 0000 0000  0000 0000 0000 0000   |................|
> 0x0000CA2C: 0000 0000 0000 0000  0000 0000 0000 0000   |................|
> 0x0000CA3C: 0000 0000 0000 0000  0000 0000 0000 0000   |................|
> 0x0000CA4C: 0000 0000 0000 0000  0000 0000 0000 0000   |................|
> 0x0000CA5C: 0000 0000 0000 0000  0000 0000 0000 0000   |................|
> 0x0000CA6C: 0000 0000 0000 0000  0000 0000 0000 0000   |................|
> 0x0000CA7C: 0000 0000 0000 0000                        |........        |
> Tx event: 8
> Can't get BOOTP info - network disabled!
> RedBoot>
> 
> The link LED is on, everything should be working I think. Is there 
> something else I've forgotten?
> 
> One issue that I don't understand here is the following quote from 
> <http://sources.redhat.com/ecos/docs-latest/tcpip/tcpip.3.html#pgfId=1132625>: 
> "This assumes that the MAC address is already defined in the serial EEPROM 
> or however the particular target implements this; there is no (tested) 
> support for setting the MAC address in this release." Does the networking 
> code in eCos [not redboot] read the MAC address set by fconfig and saved in 
> main flash store? If not, then how is it possible to bring up eth0 on the 
> Cirrus board?
> 

Currently, if you run RedBoot and enable the network address (turn on networking
support in RedBoot) then eCos will just use that.  Note: eCos does not get the
ESA from the FLASH, it assumes that RedBoot has set it up.