于 2013/9/25 星期三 0:49, niXman 写道: > 2013/9/24 xunxun: >> 400% for the builder, but 10% for the user > Yes, you're right. But far as I understood, by 10 percent will be > faster only the compiler itself, but not the programs built by it. Is > really compilation time reduction by 10 percent so significant? > I suggest you to try to build MinGW using PGO+LTO, solve all the > problems that you will face, and report the result. After that, I > promise to think about adding the corresponding option to the build > scripts ;) > > On the normal build, PGO and LTO will cause some compiler interal errors, but after my test, -march=pentium4 (specially using -msse2) will reduce the state.
-- Best Regards, xunxun ------------------------------------------------------------------------------ October Webinars: Code for Performance Free Intel webinars can help you accelerate application performance. Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from the latest Intel processors and coprocessors. See abstracts and register > http://pubads.g.doubleclick.net/gampad/clk?id=60133471&iu=/4140/ostg.clktrk _______________________________________________ Mingw-w64-public mailing list Mingw-w64-public@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mingw-w64-public
