On Tue, 2 Oct 2007 20:42:06 -0300 "Ulisses Furquim" <[EMAIL PROTECTED]> babbled:
> Hi, > > On 10/2/07, Gustavo Sverzut Barbieri <[EMAIL PROTECTED]> wrote: > > > http://staff.get-e.org/?p=e17/libs/eet.git;a=commitdiff;h=f64fe1386f0a1c4a302d8a11d1c9b0c61d0bb401 > > > > > > doursse, can you take a look? > > > > Damn, andré found it: somehow "char" is unsigned in maemo, this was > > bring to eye due this GCC warning: > > > > eet_utils.c:106: warning: comparison is always false due to limited > > range of data type > > > > why? dunno. but changing it to "int" fixes the warning AND the > > slowdown. He will commit a fix. > > I've googled for this and the first page was: > > http://www.arm.linux.org.uk/docs/faqs/signedchar.php > > Basically, the choice to make "char" be like "signed char" or > "unsigned char" is up to the implementation (or even the platform). > Our cross-compiler for the N800 (ARM) is treating "char" as an > "unsigned char" while the same compiler generating code for x86 is > treating it as a "signed char". Let's pay attention to this from now > on. :-) i tend to stick to int's for almost everything - unless i SPECIFICALLY need to store data in a smaller format (or longer - long long) when it comes to integers. and when i use chars i'm either agnostic to the numeric value (it's just a byte of data) or i explicitly use unsigned char as i am likely to be wanting that, not signed. this has worked pretty well for me for a long time. only my structs end up with non-int types (normally) for storage space reasons. otherwise int's as they often are fastest to access on most platforms. > Best regards, > > -- Ulisses > > ------------------------------------------------------------------------- > This SF.net email is sponsored by: Microsoft > Defy all challenges. Microsoft(R) Visual Studio 2005. > http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > _______________________________________________ > enlightenment-devel mailing list > enlightenment-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/enlightenment-devel > -- ------------- Codito, ergo sum - "I code, therefore I am" -------------- The Rasterman (Carsten Haitzler) [EMAIL PROTECTED] 裸好多 Tokyo, Japan (東京 日本) ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
