Hi, Joel!

Thank you for having a look at this.

On 03/21/2014 10:14 AM, Joel Brobecker wrote:
> Use "(void)" instead of "()". There was a recent policy clarification
> regarding the CS to be using with testcases, and basically we decided
> to try to follow the GCS as much as we reasonably could.

Cut-n-paste-o. Fixed.


>> +# If we get here and gdb hasn't crashed, the tests pass.
>> +pass "corrupt DWARF"
>
> That's just me but I usually do a "print 1" test, just to make sure
> that even if the testing framework did not detect the GDB process
> dying, the "print 1" test definitely will. Not important on most,
> if not all platforms, but ISTR some odd platforms where this helped.
> That's just a suggestion, you don't have to follow it.

Actually, I think that's a very good idea (which did not occur to me). 
My big hesitation with this is that this failure gets reported as 
UNRESOLVED. While I try to be studious about checking 
XFAIL/UNRESOLVED/ERROR, I sometimes overlook these in favor of a raw 
PASS/FAIL check in gdb.sum.

I've attached a revision with those two changes (ChangeLog remains 
unchanged).

Keith

ChangeLog
2014-03-20  Keith Seitz  <keiths@redhat.com>

	PR gdb/15827
	* dwarf2read.c (skip_one_die): Check that all relative-offset
	sibling DIEs fall within range of the current reader's buffer.
	(read_partial_die): Likewise.

testsuite/ChangeLog
2014-03-20  Keith Seitz  <keiths@redhat.com>

	PR gdb/15827
	* gdb.dwarf2/corrupt.c: New file.
	* gdb.dwarf2/corrupt.exp: New file.