linux-kernel - Re: [PATCH v4 3/3] mm/mempolicy: introduce MPOL_WEIGHTED

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date: Wed, 31 Jan 2024 00:12:24 -0500
From: Gregory Price <gregory.price@...verge.com>
To: Gregory Price <gourry.memverge@...il.com>
Cc: linux-mm@...ck.org, linux-kernel@...r.kernel.org,
	linux-doc@...r.kernel.org, linux-fsdevel@...r.kernel.org,
	linux-api@...r.kernel.org, corbet@....net,
	akpm@...ux-foundation.org, honggyu.kim@...com, rakie.kim@...com,
	hyeongtak.ji@...com, mhocko@...nel.org, ying.huang@...el.com,
	vtavarespetr@...ron.com, jgroves@...ron.com,
	ravis.opensrc@...ron.com, sthanneeru@...ron.com,
	emirakhur@...ron.com, Hasan.Maruf@....com, seungjun.ha@...sung.com,
	hannes@...xchg.org, dan.j.williams@...el.com,
	Srinivasulu Thanneeru <sthanneeru.opensrc@...ron.com>
Subject: Re: [PATCH v4 3/3] mm/mempolicy: introduce MPOL_WEIGHTED_INTERLEAVE
 for weighted interleaving

On Tue, Jan 30, 2024 at 01:20:46PM -0500, Gregory Price wrote:
> +	/* Continue allocating from most recent node and adjust the nr_pages */
> +	node = me->il_prev;
> +	weight = me->il_weight;
> +	if (weight && node_isset(node, nodes)) {
> +		node_pages = min(rem_pages, weight);
> +		nr_allocated = __alloc_pages_bulk(gfp, node, NULL, node_pages,
> +						  NULL, page_array);
> +		page_array += nr_allocated;
> +		total_allocated += nr_allocated;
> +		/* if that's all the pages, no need to interleave */
> +		if (rem_pages < weight) {
> +			/* stay on current node, adjust il_weight */
> +			me->il_weight -= rem_pages;
> +			return total_allocated;
> +		} else if (rem_pages == weight) {
> +			/* move to next node / weight */
> +			me->il_prev = next_node_in(node, nodes);
> +			me->il_weight = get_il_weight(next_node);

Sigh, I managed to miss a small update that killed next_node in favor of
operating directly on il_prev. Can you squash this fix into the patch?
Otherwise I can submit a separate patch.

~Gregory


diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index 7cd92f4ec0d7..2c1aef8eab70 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -2382,7 +2382,7 @@ static unsigned long alloc_pages_bulk_array_weighted_interleave(gfp_t gfp,
        unsigned int weight_total = 0;
        unsigned long rem_pages = nr_pages;
        nodemask_t nodes;
-       int nnodes, node, next_node;
+       int nnodes, node;
        int resume_node = MAX_NUMNODES - 1;
        u8 resume_weight = 0;
        int prev_node;
@@ -2412,7 +2412,7 @@ static unsigned long alloc_pages_bulk_array_weighted_interleave(gfp_t gfp,
                } else if (rem_pages == weight) {
                        /* move to next node / weight */
                        me->il_prev = next_node_in(node, nodes);
-                       me->il_weight = get_il_weight(next_node);
+                       me->il_weight = get_il_weight(me->il_prev);
                        return total_allocated;
                }
                /* Otherwise we adjust remaining pages, continue from there */