On Fri, 9 Nov 2007, Mario. wrote: > No, standard optimisation is -O2, the -O3 is unrolling all loops so > that CPUs do less guessing and more crunching more like a RISC. > There is more about the K7 opts... pplus there is some errata teh > compiler may need manually be notified of. The K7 has also some power > management, you want probably to disable it in the kernel too. Some > bugs manifest as a combination of a specific code features, compiler > features and processor features. > > And BTW, there is this 586 and i686 processor option difference, are > you sure you want to have 586 only? (without MMX etc...)
gcc using mmx is famous for giving up because it has too few registers to work with :P Maybe I could do some more research but recompiling all the kernel, rtai and emc2 even with 1GHz it's a little slow... And it gives the problem both with gcc 4.0.2 and 4.2.2. So if someone else want to make a tailored system like myself, should be aware of there 'troubles' (nasty stuff: inserting a rtapi_print between some assignment makes it work because it interrupts the flow of the optimizer). Still, it would be useful to know if it's a compiler bug or just some misalignment in compilation parameters... ------------------------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/ _______________________________________________ Emc-developers mailing list Emc-developers@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-developers
