On 04/01/2014, at 10:09 PM, srean wrote: > > Well I don't know that. It would actually require doing some testing > and experiments with performance measurements. For example > splitting the output into separately compiled units for every > function may be faster or slower, depending perhaps on whether > we can organise pre-compiled headers. I just don't know. > > Parallel compilation might help some too, but I die a little when I hear > separate compilation given one of Felix's main strength is whole program > analysis and optimization. I thought you were stressing more on algorithmic > improvements in the compiler.
I'm not planning on any significant rewriting of the compiler in the short term, because it's, er .. significant. The separate compilation stuff is useful. Splitting out routines doesn't necessarily have much impact on performance, depends *what* you split out. In particular, I doubt the Felix build tool performance will be impacted much by making separate pieces, since the main overheads are flxg and C++ anyhow. OTOH it may improve my build times by 10x. At the moment it takes me HOURS to compile a change to flx, and I write lots of bugs so it can take days. I often sleep whilst the builds are running and wake up to find they're still not finished :) -- john skaller skal...@users.sourceforge.net http://felix-lang.org ------------------------------------------------------------------------------ Rapidly troubleshoot problems before they affect your business. Most IT organizations don't have a clear picture of how application performance affects their revenue. With AppDynamics, you get 100% visibility into your Java,.NET, & PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro! http://pubads.g.doubleclick.net/gampad/clk?id=84349831&iu=/4140/ostg.clktrk _______________________________________________ Felix-language mailing list Felix-language@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/felix-language
