> What I meant was that, if a filesystem requires vma walks, we need to do > it VFS level with something like the following patch.
I don't think this patch is the way to go at all. It imposes an allocation and vma walking overhead for the vast majority of IOs that aren't interested. It doesn't look like it will get a consistent ordering when multiple file systems are concerned. It doesn't record the ranges of the mappings involved so Lustre can't properly use its range locks. And finally, it doesn't prohibit mapping operations for the duration of the IO -- the whole reason we ended up in this thread in the first place :) Christoph, would you be interested in looking at a more thorough patch if I threw one together? - z - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/