diff mbox series

[v4,10/13] mm/mempolicy: VMA allocation for many preferred

Message ID 1615952410-36895-11-git-send-email-feng.tang@intel.com (mailing list archive)
State New, archived
Headers show
Series Introduced multi-preference mempolicy | expand

Commit Message

Feng Tang March 17, 2021, 3:40 a.m. UTC 
From: Ben Widawsky <ben.widawsky@intel.com>

This patch implements MPOL_PREFERRED_MANY for alloc_pages_vma(). Like
alloc_pages_current(), alloc_pages_vma() needs to support policy based
decisions if they've been configured via mbind(2).

The temporary "hack" of treating MPOL_PREFERRED and MPOL_PREFERRED_MANY
can now be removed with this, too.

All the actual machinery to make this work was part of
("mm/mempolicy: Create a page allocator for policy")

Link: https://lore.kernel.org/r/20200630212517.308045-11-ben.widawsky@intel.com
Signed-off-by: Ben Widawsky <ben.widawsky@intel.com>
Signed-off-by: Feng Tang <feng.tang@intel.com>
---
 mm/mempolicy.c | 29 +++++++++++++++++++++--------
 1 file changed, 21 insertions(+), 8 deletions(-)

Comments

Michal Hocko April 14, 2021, 1:14 p.m. UTC | #1 
On Wed 17-03-21 11:40:07, Feng Tang wrote:
[...]
> @@ -2301,10 +2300,26 @@ alloc_pages_vma(gfp_t gfp, int order, struct vm_area_struct *vma,
>  		 * does not allow the current node in its nodemask, we allocate
>  		 * the standard way.
>  		 */
> -		if ((pol->mode == MPOL_PREFERRED ||
> -		     pol->mode == MPOL_PREFERRED_MANY) &&
> -		    !(pol->flags & MPOL_F_LOCAL))
> +		if (pol->mode == MPOL_PREFERRED || !(pol->flags & MPOL_F_LOCAL)) {
>  			hpage_node = first_node(pol->nodes);
> +		} else if (pol->mode == MPOL_PREFERRED_MANY) {
> +			struct zoneref *z;
> +
> +			/*
> +			 * In this policy, with direct reclaim, the normal
> +			 * policy based allocation will do the right thing - try
> +			 * twice using the preferred nodes first, and all nodes
> +			 * second.
> +			 */
> +			if (gfp & __GFP_DIRECT_RECLAIM) {
> +				page = alloc_pages_policy(pol, gfp, order, NUMA_NO_NODE);
> +				goto out;
> +			}
> +
> +			z = first_zones_zonelist(node_zonelist(numa_node_id(), GFP_HIGHUSER),
> +						 gfp_zone(GFP_HIGHUSER), &pol->nodes);
> +			hpage_node = zone_to_nid(z->zone);
> +		}
>  
>  		nmask = policy_nodemask(gfp, pol);
>  		if (!nmask || node_isset(hpage_node, *nmask)) {
> @@ -2330,9 +2345,7 @@ alloc_pages_vma(gfp_t gfp, int order, struct vm_area_struct *vma,
>  		}
>  	}
>  
> -	nmask = policy_nodemask(gfp, pol);
> -	preferred_nid = policy_node(gfp, pol, node);
> -	page = __alloc_pages_nodemask(gfp, order, preferred_nid, nmask);
> +	page = alloc_pages_policy(pol, gfp, order, NUMA_NO_NODE);
>  	mpol_cond_put(pol);
>  out:
>  	return page;

OK, it took me a while to grasp this but the code is a mess I have to
say. Not that it was an act of beauty before but this just makes it much
harder to follow. And alloc_pages_policy doesn't really help I have to
say. I would have expected that a dedicated alloc_pages_preferred and a
general fallback to __alloc_pages_nodemask would have been much easier
to follow.
diff mbox series

Patch

diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index a92efe7..8fe76a7 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -2273,8 +2273,6 @@  alloc_pages_vma(gfp_t gfp, int order, struct vm_area_struct *vma,
 {
 	struct mempolicy *pol;
 	struct page *page;
-	int preferred_nid;
-	nodemask_t *nmask;
 
 	pol = get_vma_policy(vma, addr);
 
@@ -2288,6 +2286,7 @@  alloc_pages_vma(gfp_t gfp, int order, struct vm_area_struct *vma,
 	}
 
 	if (unlikely(IS_ENABLED(CONFIG_TRANSPARENT_HUGEPAGE) && hugepage)) {
+		nodemask_t *nmask;
 		int hpage_node = node;
 
 		/*
@@ -2301,10 +2300,26 @@  alloc_pages_vma(gfp_t gfp, int order, struct vm_area_struct *vma,
 		 * does not allow the current node in its nodemask, we allocate
 		 * the standard way.
 		 */
-		if ((pol->mode == MPOL_PREFERRED ||
-		     pol->mode == MPOL_PREFERRED_MANY) &&
-		    !(pol->flags & MPOL_F_LOCAL))
+		if (pol->mode == MPOL_PREFERRED || !(pol->flags & MPOL_F_LOCAL)) {
 			hpage_node = first_node(pol->nodes);
+		} else if (pol->mode == MPOL_PREFERRED_MANY) {
+			struct zoneref *z;
+
+			/*
+			 * In this policy, with direct reclaim, the normal
+			 * policy based allocation will do the right thing - try
+			 * twice using the preferred nodes first, and all nodes
+			 * second.
+			 */
+			if (gfp & __GFP_DIRECT_RECLAIM) {
+				page = alloc_pages_policy(pol, gfp, order, NUMA_NO_NODE);
+				goto out;
+			}
+
+			z = first_zones_zonelist(node_zonelist(numa_node_id(), GFP_HIGHUSER),
+						 gfp_zone(GFP_HIGHUSER), &pol->nodes);
+			hpage_node = zone_to_nid(z->zone);
+		}
 
 		nmask = policy_nodemask(gfp, pol);
 		if (!nmask || node_isset(hpage_node, *nmask)) {
@@ -2330,9 +2345,7 @@  alloc_pages_vma(gfp_t gfp, int order, struct vm_area_struct *vma,
 		}
 	}
 
-	nmask = policy_nodemask(gfp, pol);
-	preferred_nid = policy_node(gfp, pol, node);
-	page = __alloc_pages_nodemask(gfp, order, preferred_nid, nmask);
+	page = alloc_pages_policy(pol, gfp, order, NUMA_NO_NODE);
 	mpol_cond_put(pol);
 out:
 	return page;