>> [ I just got back from vacation, so my apologies if I've missed some
>> discussion relevant to this. ]
>> 
>> I'm getting a testsuite failure on gdb.mi/mi-cli.exp: the relevant bit
>> of gdb.log is as follows:
>> 
>> 888-interpreter-exec console "set $pc=0x0"
>> &"frame.c:62: internal-error: get_frame_id: Assertion `!legacy_frame_p (current_gdbarch)' failed.\n"
>> &"A problem internal to GDB has been detected.  Further\ndebugging may prove unreliable.\n"
>> ERROR: Process no longer exists
>> UNRESOLVED: gdb.mi/mi-cli.exp: -interpreter-exec console "set $pc=0x0"
>> testcase ./gdb.mi/mi-cli.exp completed in 1 seconds
>> 
>> This is with GDB updated an hour or so ago, with i686-pc-linux-gnu,
>> GCC 3.1, DWARF 2.
>> 

> FYI, my test run from my just committed call_dummy_p change had the usual 21 failures.  Yet, mysteriously, for my next set of patches, I'm seeing the failures jump to 28.
> 
> While I'll note that I'm the prime suspect, did anything else get changed?

David,

Applying the attached to BFD fixes the problem ....

CC'ing bfd.

Andrew