This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

check for valid location of zero length dwarf block forms?


   We have a clarification from Apple on the problem of
dsymutils asserting since the VTA merge changes...

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=41473#c27
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=41473#c29

Does the code in gcc/dwarfout.c currently check if
any block form with zero length actually has a 
valid location before actually emitting a DW_AT_location attribute?
Apple believes that these shouldn't be emitted in that case.
This issue is currently preventing darwin from emitting
dwarf debug information (dSYM) in parts of gcc trunk (PR41473).
Thanks in advance for any information on this issue.
              Jack


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]