Re: Debugging PowerPC 860 embedded target with Abatron BDI 2000: Remote packet too long errors.

[Elena Zannoni <ezannoni at cygnus dot com>]

Hi, have you tried changing the default architecture that gdb thinks
it is dealing with?

If your current architecture (use 'show architecture') is powerpc:common, the
register set is smaller. Try 'set architecture powerpc:860'

Look in rs6000-tdep.c at the 'variants' array.

Hope this helps a bit
Elena

Cristian Alonso writes:
 > Hi,
 > I am using an Abatron (Swiss company that makes BDM's which work whith gdb)
 > BDI2000 to debug an embedded target with a PowerPC 860 CPU using Insight-5.1
 > (last prerelease snapshot available). 
 > 
 > After compiing last Insight-5.1 prerelease available and connecting to the
 > remote target i get lots of the following errors:
 > (gdb) target remote 172.24.78.42:2001
 > Remote debugging using 172.24.78.42:2001
 > Remote packet too long:
 > 000000ff0079800400080bc00000000000000000002be6d8009166100038eb38006f00000000
 > 80000000000a00009032006f000000080bc00000000000000000000000000000000000000000
 > 0000000000000000000000000000000000000000009171e40091686400000000000080000091
 > 65e8009166100000000000798004000000000000000000000000000000000000000000000000
 > 0000000000000000000000000000000000000000000000000000000000000000000000000000
 > 0000000000000000000000000000000000000000000000000000000000000000000000000000
 > 0000000000000000000000000000000000000000000000000000000000000000000000000000
 > 0000000000000000000000000000000000000000000000000000000000000000000000000000
 > 0000000000000000000000000000000000000000000000000000000000000000000000000000
 > 0000000000000000000000000000000000000000000000000000000000000000000000000000
 > 00000000002be8d40000903240000002002be8c00000000000000000deadbeefdeadbeefdead
 > beefdeadbeefdeadbee
 > Ignoring packet error, continuing...
 > 
 > After taking a look at GDB's source files, i found function
 > build_remote_packet_sizes in remote.c, in which 
 > the following lines are executed:
 > .
 > .
 > .
 >   remote_packet_size = 400 - 1;
 >   /* Should REGISTER_BYTES needs more space than the default, adjust
 >      the size accordingly. Remember that each byte is encoded as two
 >      characters. 32 is the overhead for the packet header /
 >      footer. NOTE: cagney/1999-10-26: I suspect that 8
 >      (``$NN:G...#NN'') is a better guess, the below has been padded a
 >      little. */
 >   if (REGISTER_BYTES > ((remote_packet_size - 32) / 2))
 >     remote_packet_size = (REGISTER_BYTES * 2 + 32);
 > .
 > .
 > 
 > Changing the value of remote_packet_size to 5 Kbytes works fine with my
 > BDI2000. It makes my insight work fine and allows me to laugh at all the
 > Windriver Visionclick users who told me i would not be able to debug using
 > "that free software of yours" ;-).
 > 
 > It seems that i should be able to find the definition of REGISTER_BYTES and
 > change it to a different value and submit the patch but i have the following
 > problems:
 >   - I have not been able to find in which file exactly (ppc.h does not seem
 > to work...)do i have to change the definition of REGISTER_BYTES.
 >   - I do not know which value is correct for this #define (it seems that it
 > has to be higher, and it has something to do with how many registers do i
 > have in my arch, but how do i find out?)
 > 
 > Thanks in advance,
 > Cristian Alonso Aldama
 > Development Group Manager
 > R&D Department
 > e-mail: calonso@teldat.es 
 > http://www.teldat.es