On Thu, 2015-08-13 at 01:58 +0200, Markus Trippelsdorf wrote: > On 2015.08.12 at 13:03 -0600, Jeff Law wrote: > > On 08/12/2015 12:57 PM, Richard Sandiford wrote: > > > Jeff Law <l...@redhat.com> writes: > > >> On 08/10/2015 06:05 AM, tbsaunde+...@tbsaunde.org wrote: > > >>> From: Trevor Saunders <tbsaunde+...@tbsaunde.org> > > >>> > > >>> Hi, > > >>> > > >>> In many places gcc puts classes in the anon namespace so the compiler > > >>> can tell > > >>> they do not get inheritted from to enable better devirtualization. > > >>> However > > >>> debugging code in the anon namespace can be a pain, and the same thing > > >>> can be > > >>> accomplished more directly by marking the classes as final. When > > >>> bootstrapping > > >>> stage 3 should always be built in C++14 mode now, and of course will > > >>> always be > > >>> newer than gcc 4.7, so these classes will always be marked as final > > >>> there. > > >>> AIUI cross compilers are supposed to be built with recent gcc, which I > > >>> would > > >>> tend to think implies newer than 4.7, so they should also be built with > > >>> these > > >>> classes marked as final. I believe that means in all important cases > > >>> this works just as well as the anon namespace. > > >>> > > >>> bootstrapped + regtested on x86_64-linux-gnu, ok? > > >>> > > >>> Trev > > >>> > > >>> > > >>> gcc/ChangeLog: > > >>> > > >>> 2015-08-10 Trevor Saunders <tbsau...@tbsaunde.org> > > >>> > > >>> * compare-elim.c, dce.c, dse.c, gimple-ssa-isolate-paths.c, > > >>> gimple-ssa-strength-reduction.c, graphite.c, init-regs.c, > > >>> ipa-pure-const.c, ipa-visibility.c, ipa.c, mode-switching.c, > > >>> omp-low.c, reorg.c, sanopt.c, trans-mem.c, tree-eh.c, > > >>> tree-if-conv.c, tree-ssa-copyrename.c, tree-ssa-dce.c, > > >>> tree-ssa-dom.c, tree-ssa-dse.c, tree-ssa-forwprop.c, > > >>> tree-ssa-sink.c, tree-ssanames.c, tree-stdarg.c, > > >>> tree-tailcall.c, > > >>> tree-vect-generic.c, tree.c, ubsan.c, var-tracking.c, > > >>> vtable-verify.c, web.c: Use GCC_FINAL instead of the anonymous > > >>> namespace. > > >> OK. > > > > > > I was hoping someone else was going to speak up since I seem > > > to have been posting a few negative messages recently, but I think > > > this is really a step in the wrong direction. I think the code > > > was using anonymous namespaces in exactly the way they were > > > intended to be used. > > No need to worry about seeming to be negative. > > > > > > The problem is you can't get to stuff in the anonymous namespace easily > > in the debugger. There was talk of fixing that, but I don't think it > > ever went forward on the gdb side. > > > > If gdb were to get fixed so that debugging this stuff was easier, then > > I'd fully support putting things back into the anonymous namespace. > > For the record here the gdb bug in question: > https://sourceware.org/bugzilla/show_bug.cgi?id=16874 > It even has a patch attached, that improves the issue.
Sadly that seems to have stalled. FWIW, our gcc/gdbhooks.py adds in a "break-on-pass" gdb command to work around this: https://gcc.gnu.org/ml/gcc-patches/2014-07/msg02011.html (gdb) break-on-pass pass_final Breakpoint 6 at 0x8396ba: file ../../src/gcc/final.c, line 4526. (gdb) cont Continuing. Breakpoint 6, (anonymous namespace)::pass_final::execute (this=0x17fb990) at ../../src/gcc/final.c:4526 4526 virtual unsigned int execute (function *) { return rest_of_handle_final (); } (gdb) Though clearly hacking around it this way is suboptimal. Dave
