Re: FYI GDB on-disk .debug cache (mmapcache) [Re: Tasks]

[Tom Tromey <tromey at redhat dot com>]

>>>>> "Jan" == Jan Kratochvil <jan.kratochvil@redhat.com> writes:

Jan> Maybe it was just stupid from me to think it can be somehow incrementally
Jan> cheated by some obstack/xmalloc trick...

Not at all, it is worth looking at.

I was asking my questions just out of the belief that we have some
manpower now to fix things more thoroughly.  So, we should examine
more difficult solutions as well, if they give better results.

Jan> Do we always want to silently hide the definitions outside of the
Jan> local block?

Could you elaborate on this?  I didn't understand.

Tom> What if we modify GCC to emit a new .debug_privnames with the missing
Tom> data?

Jan> Please note that .debug_pubnames is a part of the DWARF3 spec.

Yeah, I know.  We would need an extension if we went this route.

Tom> I tried enabling the debug_pubnames code.  I needed a couple little
Tom> patches to make it compile and not crash... according to gprof it is
Tom> much faster (from 30 seconds down to 5 seconds) but it also used a lot
Tom> more memory (hitting swap, so subjectively it was extremely slow :).
Tom> So, something is a bit rotted here and at least needs a cleanup.

Of course, after sending off a patchlet for this I realized I
basically failed to read the code; it doesn't look to me like it
actually does anything other than read data and throw it away :-(

Jan> Just a question if psymtabs are enough?

I have no idea.
Time for me to learn more :)

Tom