Geometry Boost List: [ggl] inline keyword in class definition

Subject: [ggl] inline keyword in class definition
From: Mateusz Loskot (mateusz)
Date: 2009-05-06 15:18:12

Bruno Lalande wrote:
>> Just a side note, I'm not Cygwin nor MinGW expert, but I've heard
>> GCC does not optimize well on Windows. It tells me that these ports
>> are not bets choices to measure performance.
> 
> I think the best choice to measure performance is to use what's used 
> in industry, and those two products are widely used.

Sure, Visual C++ and GCC are well settled in the industry, but
I'm not sure sure about GCC on Windows.

> Besides that, GCC doesn't optimize well on Windows regarding
> Windows-specific stuff only. But the general optimizations mechanics,
> like inlining, remain unchanged.

Right. By the way, I've found an interesting comparison:

C++ Performance Benchmarks (VC++ 2008 vs. MinGW gcc 4.3)

http://social.microsoft.com/Forums/it-IT/vcgeneral/thread/b992fdcc-5ad0-4ae7-b5d6-5bd49eb7e8dc

I've always been sceptic about any form of GCC under Windows
(though I love GCC on Unix). Perhaps I should change my mind
about MinGW :-)

>> Sure, I'm not going to remove. I wanted to check if we want to put
>> must specify inline" as a guideline.
> 
> I think you could state "inline keywork is only needed on free 
> functions" BUT not until we've tested this assumption on several 
> compilers.

Right.

Best regards,
-- 
Mateusz Loskot, http://mateusz.loskot.net
Charter Member of OSGeo, http://osgeo.org