From: Glen Miner Newsgroups: comp.os.msdos.djgpp Subject: Re: Optimization Date: Thu, 28 Nov 1996 08:13:20 -0500 (EST) Organization: Newbridge Networks Corporation Lines: 33 Message-ID: References: <329D6012 DOT D0E AT techfak DOT uni-kiel DOT de> NNTP-Posting-Host: 138.120.136.238 Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII In-Reply-To: <329D6012.D0E@techfak.uni-kiel.de> To: djgpp AT delorie DOT com DJ-Gateway: from newsgroup comp.os.msdos.djgpp > Keep all these functions in the same file as your main recursion loop, > mark them as 'static inline ...' and put a declaration of all these > functions before any real code. GCC will happily insert them into the > calling function if possible. Hmm, I think that due to their size this may be a liability, but I'll try it. > > How well will djgpp deal with two dimentional arrays? Say I'm stepping > > through char MyArray[9][9] like so: > > > > for (x = 0; x < 6; x++) > > for (y = 0; y < 7; y++) > > MyArray [x][y] = SomeValue; > > On a P133: > > MyArray GCC Options P-133 486-33 > char -O3 1005 7600 > char -O6 -fomit-frame-pointer -funroll-loops 975 3000 > short -O6 -fomit-frame-pointer -funroll-loops 925 3450 > int -O6 -fomit-frame-pointer -funroll-loops 905 2600 > > -> Use int and -funroll-loops in such a case ! Can this be? Is manipulating 32bit data _REALLY_ faster then 8bit? Peace ===[ Gabo / [ABC] : gaminer AT undergrad DOT math DOT uwaterloo DOT ca ]=================== Latest ABC Shogi: http://www.undergrad.math.uwaterloo.ca/~gaminer/shogi.html "What Greenpeace spends in a year General Motors spends in four hours" -Moby