Re: [5.3] Alpha and that mdebug problem?

[David Carlton <carlton at math dot stanford dot edu>]

On Wed, 20 Nov 2002 13:04:11 -0500, Andrew Cagney <ac131313@redhat.com> said:

> [outch!]

Don't worry, I'm not mad at anybody in particular, I'm just
venting. :-)

> FYI, GDB flushed the os9k reader recently so all is not lost.

Yay!  (My secret goal is to have GDB contain fewer lines of code when
I stop working on it than it was when I started working on it: see my
patches from 2002-11-15, 2002-10-28, 2002-10-23, 2002-10-18, part of
2002-08-19, and even part of 2002-11-05.)

> mdebug, though, really is pretty common (MIPS for instance depends on
> bits of it).

Yeah, so I hear.  It just seems to me that, if it's so common,
somebody should be willing to buildsym-ify it.  So, from my point of
view, slowly pessimizing it until somebody else gets sick enough of it
to take responsibility seems like a valid strategy...

Who knows, this namespace stuff that I'm doing won't take forever, so
maybe I'll get to it myself in the Spring, if somebody else can
provide a machine for me to work on.  (I'm certainly not willing to
fix it working on simulators, especially given that my current machine
is 4+-years old and running out of disk space as is; my department is
remarkably tolerant of the fact that I spend most of my non-teaching
time hacking GDB, but asking them to buy me a better computer would, I
think, be going too far.)  Or Red Hat could hire me next Summer when
I'm done at Stanford, and I can work on it then. :-)

>>> However backwards we may find them, people are still using mdebug,
>>> and making the test suite unusable would make it impossible for a
>>> company like Red Hat to give them a code drop when they need one.

>> If it really is true that Red Hat wishes to be able to support
>> people using that debug format, perhaps somebody at Red Hat should
>> spend some time buildsym-ifying it and fixing PR gdb/676.

> Do you have the URL?

That quote is from
<http://sources.redhat.com/ml/gdb-patches/2002-09/msg00564.html>.  I
should reiterate that I'm not annoyed that the patch in question was
shelved: it was strictly a code cleanup patch whose only observable
effect would be to pessimize mdebugread.c.  But, on the other hand, I
wanted that code cleanup for a reasonably good reason.

David Carlton
carlton@math.stanford.edu