Closed GoogleCodeExporter closed 9 years ago
Jim, mind taking a glance at this one?
Thanks,
Neal
Original comment by neal...@gmail.com
on 8 Jan 2010 at 10:56
Thanks for working this out, Ian. This seems to have been present since before
I
worked on the STABS code.
Your analysis looks right to me, but I wonder if there isn't a nicer way to do
the
patch. In particular, don't N_UNDF stabs appear before the initial N_SO(s) for
a
compilation unit? If that's true, then we could recognize the N_UNDF entries in
StabsReader::Process, which would be a bit cleaner. I'll redo the patch; just
let me
know if this is true.
I'll see if I can reproduce this; if not, I may need to ask Lei for an object
file I
can use for testing.
Original comment by jimbla...@gmail.com
on 9 Jan 2010 at 12:16
Hi Jim. You should be able to reproduce the problem with GNU ld by using the --
traditional-format linker option. Yes, there should normally be an N_UNDF stab
at the
start of the stabs for each compilation unit.
Original comment by ianlance...@gmail.com
on 9 Jan 2010 at 1:11
I've filed a fix for this:
http://breakpad.appspot.com/54003
Original comment by jimbla...@gmail.com
on 13 Jan 2010 at 7:26
Fix reviewed, committed, r490.
Original comment by jimbla...@gmail.com
on 14 Jan 2010 at 5:27
Original issue reported on code.google.com by
ianlance...@gmail.com
on 4 Jan 2010 at 11:15