On Wed, Nov 15, 2017 at 03:48:26PM -0800, Andrew Morton wrote: > On Wed, 15 Nov 2017 12:14:30 -0500 Josef Bacik <josef@xxxxxxxxxxxxxx> wrote: > > > Previously we were using the ratio of the number of lru pages scanned to > > the number of eligible lru pages to determine the number of slab objects > > to scan. The problem with this is that these two things have nothing to > > do with each other, so in slab heavy work loads where there is little to > > no page cache we can end up with the pages scanned being a very low > > number. This means that we reclaim next to no slab pages and waste a > > lot of time reclaiming small amounts of space. > > > > ... > > > > Andrew, I noticed you hadn't picked this up yet, so I rebased it on the latest > > linus and updated the ack's, it should be good to go. > > I dropped a previous version of this on Oct 3 due to runtime failures > (I think). What were those and how does this patch fix them (if it > does?) > I went back and looked and you said it didn't apply cleanly, but then I rebased it and it applied fine, so I was confused. What I _think_ happened is Johannes added a cleanup patch in the thread that didn't apply cleanly so you didn't apply either of them. I'll dig out Johannes patch in the morning and clean it up and send it along as well. Thanks, Josef -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxx. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>
