Le mercredi 23 mai 2007 à 10:26 +0200, Markus Schiltknecht a écrit : > Hi, > > Richard Levitte wrote: > > I just looked at the patch, and although I can see that it fixes > > things for status, I think it could be taken one step further (and > > thereby generalising the solution) by modifying the file_path class to > > have a status cache. > > Shouldn't the OS do that sort of caching? That's probably why there's no > difference with local filesystems. But why does the kernel not cache NFS > stat() calls? Can we somehow teach the kernel to do that caching, so we > don't have to write our own?
Yes it does, but even on a local filesystem, savings 4000 system calls is not a stupid thing. Doing 4x in a row the same syscall on the same file is stupid. I don't think we need caching : we just need to do the things as many times as necessary. -- Benoît Dejean GNOME http://www.gnomefr.org/ LibGTop http://directory.fsf.org/libgtop.html
signature.asc
Description: Ceci est une partie de message numériquement signée
_______________________________________________ Monotone-devel mailing list Monotone-devel@nongnu.org http://lists.nongnu.org/mailman/listinfo/monotone-devel