On Thu, Sep 08, 2016 at 09:30:53PM -0400, Rik van Riel wrote:
> Commit 4d9424669946 ("mm: convert p[te|md]_mknonnuma and remaining
> page table manipulations") changed NUMA balancing from _PAGE_NUMA
> to using PROT_NONE, and was quickly found to introduce a regression
> with NUMA grouping.
>
> It was followed up by these changesets:
>
> 53da3bc2ba9e ("mm: fix up numa read-only thread grouping logic")
> bea66fbd11af ("mm: numa: group related processes based on VMA flags instead
> of page table flags")
> b191f9b106ea ("mm: numa: preserve PTE write permissions across a NUMA hinting
> fault")
>
> The first of those two changesets try alternate approaches to NUMA
> grouping, which apparently do not work as well as looking at the PTE
> write permissions.
>
> The latter patch preserves the PTE write permissions across a NUMA
> protection fault. However, it forgets to revert the condition for
> whether or not to group tasks together back to what it was before
> 3.19, even though the information is now preserved in the page tables
> once again.
>
> This patch brings the NUMA grouping heuristic back to what it was
> before changeset 4d9424669946, which the changelogs of subsequent
> changesets suggest worked best.
>
> We have all the information again. We should probably use it.
>
Patch looks ok other than the comment above the second hunk being out of
date. Out of curiousity, what workload benefitted from this? I saw a mix
of marginal results when I ran this on a 2-socket and 4-socket box.
--
Mel Gorman
SUSE Labs
