[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-devel
Subject:    Two optimization questions
From:       Adriaan de Groot <adridg () cs ! kun ! nl>
Date:       2004-04-20 21:23:47
Message-ID: 200404202323.52657.adridg () cs ! kun ! nl
[Download RAW message or body]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

[Perhaps better for kde-optimize, I know, although these questions are mostly 
code-clarity vs. extremely minor performance gains.]

(1) Consider

	for (QStringList::constIterator i = l.begin() ; i!=l.end(); ++i) { .. }

versus

	QStringList::constIterator e = l.end();
	for (QStringList::constIterator i = l.begin() ; i!=e; ++i) { .. }

Is there really a big performance hit for the first? I know KMail was looking 
close at this, for huge iterations; I'm wondering what The Right Thing To Do 
is, since the latter is slightly longer code, and possibly harder to 
understand. What's your advice?

(2) Consider function foo(), which spans 96 lines (that's 4 screenfuls). It 
consists of several identifiable blocks of functionality, so I _could_ write 
it like

	T foo() { doThis(); doThat(); doItWithAWhiffleBallBat(); doRest(); }

In languages like Elan, Haskell, Ocaml, I know how to do this with a minimum 
of fuss, keeping the big-picture structure of foo() visible while defining it 
minutely elsewhere. I'm wondering how to do this effectively in KDE C++ code.

Suppose I write:

	static void doThis(.. parameters needed from locals of foo ..) { ... }
	static void doThat( ... ) { ... }
	...

	T foo() { doThis(parms); doThat(...); ... }

What's the performance hit here? I'd hate to have real function-call overhead 
for those functions, and I'd hate for them to be external symbols, since 
they'd slow down linkage to no effect. Can I count on the compiler to inline 
these in foo?

(2a) Alternatively, just for code clarity, when is kate going to get folding 
that is persistent across loads and saves, like the venerable ae editor on 
OS/2 had? Then I could fold up foo() into its logical parts and dispense with 
splitting it out entirely.




- -- 
pub  1024D/FEA2A3FE 2002-06-18 Adriaan de Groot <groot@kde.org>
                     Would you like a freem?
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (FreeBSD)

iD8DBQFAhZTodqzuAf6io/4RAiXRAJ9vBBridl2jITAmn/ekJBYJ5+c1JgCdGiM8
9aaFxms1tGZjqX830HhASqQ=
=tENg
-----END PGP SIGNATURE-----
 
>> Visit http://mail.kde.org/mailman/listinfo/kde-devel#unsub to unsubscribe <<

[prev in list] [next in list] [prev in thread] [next in thread]