This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [java] RFA: builtins vs. -findirect-dispatch
- From: Bryce McKinlay <mckinlay at redhat dot com>
- To: Andrew Haley <aph at redhat dot com>
- Cc: Java Patches <java-patches at gcc dot gnu dot org>, GCC Patches <gcc-patches at gcc dot gnu dot org>
- Date: Thu, 13 Jul 2006 10:18:13 -0400
- Subject: Re: [java] RFA: builtins vs. -findirect-dispatch
- References: <44B5CB9B.6010106@redhat.com> <17590.3419.701765.956118@zebedee.pink>
Andrew Haley wrote:
Bryce McKinlay writes:
> Currently, some of GCJ's Math builtins can generate non-indirect
> function calls even when -findirect-dispatch is set. The simplest fix
> for this seems to be to simply disable builtins where a function call
> could be generated and -findirect-dispatch in use.
LOL! I remember suggesting this patch before and being told that it
was a bad idea as it would affect performance!
Yeah :). This fix is slightly better, though, as it doesn't disable the
min/max/abs/etc builtins, which map directly to tree EXPR's.
> This is needed for the upcoming libgcj_bc patch, since that will ensure
> at link time that only known BC-ABI symbols are used by an
> -findirect-dispatch binary.
>
> OK to commit?
OK. It's not ideal, but I suppose it has the virtue of simplicity.
Yes, I agree we will want to optimize this eventually. But for now, this
way at least we don't complicate the ABI and mess things up for future
generations.
Bryce