This is the mail archive of the gcc-patches@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: Speedup CSE by 5%


On Mon, 2005-01-17 at 21:09 +0100, Arend Bayer wrote:
> On Mon, 17 Jan 2005, Jeffrey A Law wrote:
> 
> > > If by "inlining for_each_rtx" you include the constant propagation that avoids
> > > the indirect function call to approx_reg_cost_1, then probably yes.
> > Well, if avoiding the indirection is a significant component (or the
> > major component), then another possibility we could look into would be
> > turning for_each_rtx into an inline function itself.
> > 
> > That wouldn't get all the specializations you're doing and could 
> > potentially have bad icache effects.  But it would have the positive
> > effect that everyone using for_each_rtx would avoid the indirect call
> > overhead.
> 
> Again, I don't see how this works with for_each_rtx being a recursive
> function. I think all you could achieve is to get rid of the indirect
> function call on the first recursion level.
You are (of course) correct.  I've been working with gimple too much
lately and needed a reminder of the annoyingly recursive nature of
RTL.  Ugh.

Our of curiosity, do your profiles indicate if there's any call
sites for approx_reg_cost which are particularly abusive?  I also
wonder if we're doing a lot of redundant calls because of the 
redundant processing of insns inherent in our current CSE
implementation.


Jeff


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