This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: C identifier lookup speedups, 2/2
- From: David Edelsohn <dje at watson dot ibm dot com>
- To: Mark Mitchell <mark at codesourcery dot com>
- Cc: Zack Weinberg <zack at codesourcery dot com>, gcc-patches at gcc dot gnu dot org
- Date: Mon, 14 Apr 2003 16:26:33 -0400
- Subject: Re: C identifier lookup speedups, 2/2
>>>>> Mark Mitchell writes:
Mark> OK, I see what's going on. There's really no good way to get the
Mark> previous behavior without a major overhaul of now pushdecl works, which
Mark> is not going to be fun.
Mark> Zack, the problem is that you added:
Mark> DECL_CONTEXT (x) = current_function_decl;
Mark> to pushdecl. That means that the goofy extern declaration in scope2.c
Mark> gets DECL_CONTEXT set here, which isn't right, w.r.t. the original
Mark> interpretation of the code.
Mark> I'll change the pedwarn to an error and move the test to noncompile;
Mark> it's not worth it to do hard work to save this bit of brokenness.
The original code in c-decl.c mentioned that sometime one has to
do this. Who uses this syntax? Glibc? Breakage on AIX probably means
breakage on PPC64 Linux as well.
David