Re: [RFC: 2.6 patch] add -fno-tree-scev-cprop to KBUILD_CFLAGS

[Bernd Schmidt <bernds_cb1@t-online.de>]

Adrian Bunk wrote:
> It can be a performance regression, but there are also cases where it 
> can improve performance. If gcc produces lower performance code that
> would be a bug in gcc that should be reported, but using a division is 
> not generally wrong.
> 
> A more clearer example might be:
> 
> <--  snip  -->
> 
> void foo(u64 ns)
> {
> 	if (ns < 10000)
> 		return;
> 
> 	while(ns >= 3) {
> 		ns -= 3;
> #ifdef DEBUG
> 		bar(ns);
> #endif
> 	}
> }
> 
> <--  snip  -->
> 
> With DEBUG not defined you can hardly argue gcc should be fixed to not 
> use a division for performance reasons.

Absent any clear information about the possible values of ns, IMO this
is a case where the compiler should just assume that the programmer
knows best whether to use a loop or a division.  Principle of least
surprise, and all that...


Bernd
-- 
This footer brought to you by insane German lawmakers.
Analog Devices GmbH      Wilhelm-Wagenfeld-Str. 6      80807 Muenchen
Sitz der Gesellschaft Muenchen, Registergericht Muenchen HRB 40368
Geschaeftsfuehrer Thomas Wessel, William A. Martin, Margaret Seif