This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: The second pass of Haifa instruction scheduler
- From: Jim Wilson <wilson at specifixinc dot com>
- To: Hsin-Hsien Wu <sinsen dot wu at msa dot hinet dot net>
- Cc: gcc at gcc dot gnu dot org
- Date: Tue, 06 Jul 2004 12:26:10 -0700
- Subject: Re: The second pass of Haifa instruction scheduler
- References: <200407040210.KAA02167@msr39.hinet.net>
Hsin-Hsien Wu wrote:
> What if the haifa scheduler doesn't set an insn to TImode or to VOIDmode?
If we set TImode, then an instruction is the start of an issue group.
If we do not set TImode, then an instruction is not the start of an
issue group. There is nothing complicated here.
If an instruction was neither set to TImode or VOIDmode, then it means
that we have no issue groups at all. I.e., issue_rate == 1. See
haifa-sched.c, and grep for TImode.
--
Jim Wilson, GNU Tools Support, http://www.SpecifixInc.com