https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115943
--- Comment #9 from Guinevere Larsen <blarsen at redhat dot com> --- (In reply to Andrew Pinski from comment #8) > (In reply to Guinevere Larsen from comment #7) > > I understand that you could have multiple lines related to a single > > instruction (even if I disagree it should happen at -Og). My question is, > > why is it ok for an entry to have is_stmt and !is_stmt for the same > > instruction? > > I don't see that as an issue because they both are different column info. > Why do you think that is an issue? I misspoke in my last message. My question should have used "instruction" instead of "entry", as in, how can a single instruction in the binary at once have is_stmt and !is_stmt? I'm confused because, according to the DWARF spec: [is_stmt indicates] that the current instruction is a recommended breakpoint location. A recommended breakpoint location is intended to “represent” a line, a statement and/or a semantically distinct subpart of a statement. and entries 1 and 2 of my original example were for the same instruction, 0x401126, and they disagreed on if the instruction should or shouldn't be a breakpoint location. > > Note try adding `-gno-as-loc-support ` and seeing if the special opcode is > used because I didn't see it being used by GCC; it is the binutils assembly > which generates it from the `.loc` and such. I still see some special opcodes used when compiling with `g++ -gno-as-loc-support -fPIC -Og -g gcc_reproducer.cpp -o no_loc`, though the output is slightly better than not using it.
