[Bug symtab/34389] Source code lookup fails for code at or near 0 when linking with garbage collection
tromey at sourceware dot org via Gdb-prs <[email protected]>
| Newsgroups | gmane.comp.gdb.bugs.discuss |
|---|---|
| Message-ID | <[email protected]/bugzilla/> |
https://sourceware.org/bugzilla/show_bug.cgi?id=34389
Tom Tromey <tromey at sourceware dot org> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |tromey at sourceware dot org
--- Comment #2 from Tom Tromey <tromey at sourceware dot org> ---
(In reply to Andrew Burgess from comment #1)
> The current code assumes that the blockvector's address map is "perfect", if
> the address map claims to contain a given address then we assume that this
> is the correct block vector
I think it would be ideal to preserve this behavior; and if any data scrubbing
is needed, we should require that the debuginfo reader do this work.
> So this patch says that, if the PC we're looking for is within an address
> range that starts at zero then this might be a garbage collected range, and
> we shouldn't trust the address map.
I wonder why the existing section-at-zero mechanism isn't working here.
I suspect some spot in the DWARF reader should be filtering out these
bogus address ranges.
--
You are receiving this mail because:
You are on the CC list for the bug.