On Thu, Aug 13, 2009 at 02:23:33PM -0700, Garrett D'Amore wrote: > Dave Plauger wrote: > > > > > > > > > >Ivek Szczesniak wrote: > >>The stdio implementation in libc is among the slowest stdio > >>versions out there. If you want to archive better performance you > >>should use the stdio implementation in libast or use mmap(2). > >This is an interesting implementation suggestion, but is outside the > >scope of PSARC because it does not affect the interfaces > >being proposed. We did achieve quite a speedup over the old method. > >We'll take another look. > > Using libast might well incur extra PSARC oversight -- are the libast > interfaces public? Consolidation private? If they are *project > private* then you'll need to get a contract for them. Using mmap() > would be free of those issues, and is likely to be the fastest without > imposing any new interdependencies.
Without knowing about this project, yesterday I prototyped a simple change which buffered input and output from savecore(), and it achieved a 2.5x speedup on an pessimal setup (/var/crash on UFS, dump device on the same disk). I'm codereviewing this wad, so I'll talk to the developer about this. I agree that this doesn't seem architectural to me. Cheers, - jonathan
