This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: minor code-quality regression vs. 2.95
- To: Clinton Popetz <cpopetz at cygnus dot com>
- Subject: Re: minor code-quality regression vs. 2.95
- From: Jeffrey A Law <law at cygnus dot com>
- Date: Thu, 13 Apr 2000 08:17:32 -0600
- cc: Richard Henderson <rth at cygnus dot com>, Zack Weinberg <zack at wolery dot cumb dot org>, gcc at gcc dot gnu dot org
- Reply-To: law at cygnus dot com
In message <20000413085015.A15135@cpopetz.com>you write:
> Ok, I'll remove this from delete_computation:
>
> #ifdef INSN_SCHEDULING
> /* ?!? The schedulers do not keep REG_DEAD notes accurate after
> reload has completed. The schedulers need to be fixed. Until
> they are, we must not rely on the death notes here. */
> if (reload_completed && flag_schedule_insns_after_reload)
> {
> delete_insn (insn);
> return;
> }
> #endif
>
>
> and see if we fail regression anywhere.
I discussed this with Clint this morning, but wanted to get the
info straight for archival purposes.
the code needs to change since we now have accurate life info after
sched2, but cross jumping, reg-stack & reorg can confuse things.
jeff