On Wed, Nov 20, 2019 at 07:45:52AM -0800, Matthew Wilcox wrote: > On Wed, Nov 20, 2019 at 10:23:18PM +0800, zhengbin wrote: > > I have tried to change last_ino type to unsigned long, while this was > > rejected, see details on https://patchwork.kernel.org/patch/11023915. > > Did you end up trying sbitmap? > > What I think is fundamentally wrong with this patch is that you've found a > problem in get_next_ino() and decided to use a different scheme for this > one filesystem, leaving every other filesystem which uses get_next_ino() > facing the same problem. > > That could be acceptable if you explained why tmpfs is fundamentally > different from all the other filesystems that use get_next_ino(), but > you haven't (and I don't think there is such a difference. eg pipes, > autofs and ipc mqueue could all have the same problem. If you think that anyone is willing to pay one hell of a price on each pipe(2)... Note that get_next_ino() is pretty careful about staying within per-cpu stuff most of the time; it hits any cross-CPU traffic only in 1/1024th of calls. This, AFAICS, dirties shared cachelines on each call. And there's a plenty of pipe-heavy workloads, for obvious reasons.
