This is the mail archive of the
gdb@sourceware.org
mailing list for the GDB project.
Re: m68k gdb issue with skyeye
- From: Michael Snyder <msnyder at specifix dot com>
- To: Joel Sherrill <joel dot sherrill at oarcorp dot com>
- Cc: gdb at sourceware dot org, kang shuo <blackfin dot kang at gmail dot com>
- Date: Tue, 12 Aug 2008 18:01:42 -0700
- Subject: Re: m68k gdb issue with skyeye
- References: <48A0F96E.7000701@oarcorp.com>
On Mon, 2008-08-11 at 21:46 -0500, Joel Sherrill wrote:
> Hi,
>
> Michael Kang and I have run into something with
> m68k gdb and don't know how to fix it. The executable
> is for a coldfire mc5272.
>
> With m68k-rtems* (basically m68k-elf) on gdb 6.8,
> connecting to the Skyeye gdbserver gives this.
>
> (gdb) tar remote :12345
> Remote debugging using :12345
> Remote 'g' packet reply is too long:
> 000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000200000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000
>
> Michael tried gdb 6.8 targeting m68k-linux and
> it connects OK.
>
> I tried gdb 6.6 m68k-rtems and it connects and
> appears to work OK.
>
> It looks like the way the maximum packet size is
> calculated has changed since 6.6 and is dependent
> on number of registers.
>
> Does anyone see what the subtle difference is that
> is breaking it for m68k-rtems?
I have run into something similar -- yes there was a change in the way
this was computed, in fact I don't think it used to ever check if the
packet was too long (only if it was too short).