https://bugs.kde.org/show_bug.cgi?id=471807
Mark Wielaard <m...@klomp.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Ever confirmed|0 |1
Status|REPORTED |CONFIRMED
--- Comment #11 from Mark Wielaard <m...@klomp.org> ---
Thanks for those changes, they look good. Some minor comments/thoughts:
The move_DebugInfo_one_step_forward in VG_(load_di) and VG_(addr_load_di) seem
unnecessary, but mostly harmless.
Those are optimizations for VG_(find_DebugInfo) and find_DiCfSI, should
probably only be done there.
ML_(canonicaliseTables) looks correct, you don't want to freeze if deferred,
there could be a separate .symtab show up.
I don't fully understand the ML_(finish_CFSI_arrays) early return. Does that
ever happen? Is .eh_frame not always read early because it is always in the
main ELF image? Or do we also handle .debug_frame? Ah, we do. OK then we don't
want to free cfsi_rd just yet.
This looks good. I think we should remove those two
move_DebugInfo_one_step_forward calls. I like to commit this with that small
change so we can test it a bit more widely.
--
You are receiving this mail because:
You are watching all bug changes.
