This is the mail archive of the
mailing list for the GCC project.
Re: [Fwd: The new scheduler and x86 CPUs]
- To: David Edelsohn <dje at watson dot ibm dot com>, Vladimir Makarov <vmakarov at redhat dot com>
- Subject: Re: [Fwd: The new scheduler and x86 CPUs]
- From: Dan Nicolaescu <dann at godzilla dot ICS dot UCI dot EDU>
- Date: Thu, 30 Aug 2001 02:01:37 -0700
- Cc: gcc at gcc dot gnu dot org
First of all congratulation and thanks for this great work!
It probably deserves a NEWS entry on the GCC website.
Vladimir Makarov wrote:
> As for software pipelining, I am not going to add RCSP soon. I wrote
> in my comments in RCSP at least 7 places where RCSP could be and should
> be improved. But even now I have some Fujitsu tests (these are small
> tests like sorting) where the current RCSP for a Fujitsu VLIW processors
> gave 20-40% improvements (on simulator).
Please consider submiting RCSP as soon as it's convenient to you.
Software pipelining is a very important missing feature in gcc.
If it's not turned on by default, it's functional, gives
improvements in some situations it can be put in gcc (preferably on
the mainline or on a branch) even if there's a lot of room for improvement.
It would be in a situation analogous to the SSA code.
Obviously more people can work on improving the software pipelining
implementation if it's available...