This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]
Other format:	[Raw text]

Re: Deprecating -f options?

From: Andrew MacLeod <amacleod at redhat dot com>
To: Joern RENNECKE <joern dot rennecke at st dot com>
Cc: gcc mailing list <gcc at gcc dot gnu dot org>
Date: Thu, 15 Jun 2006 14:21:27 -0400
Subject: Re: Deprecating -f options?
References: <4491A18F.1040903@st.com>

On Thu, 2006-06-15 at 19:06 +0100, Joern RENNECKE wrote:
> In http://gcc.gnu.org/ml/gcc/2006-06/msg00549.html, you wrote:
> 
> > -ftree-live-range-split  splits live ranges on exit from ssa.  This is
> > on by default, and in fact it is more work to NOT split unrelated live
> > ranges, and creates extra register pressure. 
> 
> I've found that in 4.1, -fno-web is often needed in order to generate auto-increment addressing.
> I suppose that will be similar for any live range splitting that happens before flow.
> 

I assume you are referring to and rtl flow pass...
-ftree-live-range-split happens on trees before expand and only
determines whether non-interfering ssa-names get recombined to a single
variable, or are left as different variables. I suspect if you disable
it you would get worse performance results, and I highly doubt it has
any effect on auto increment addressing.

 In fact, the option is rather misnamed. It would be more correct to
call it -ftree-combine-live-ranges and have it turned off by default.   

Im not aware of anyone turning it off for anything practical. 

Andrew

References:
- Re: Deprecating -f options?
  - From: Joern RENNECKE

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]