[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <alpine.DEB.1.00.0802051813250.19033@chino.kir.corp.google.com>
Date: Tue, 5 Feb 2008 18:17:50 -0800 (PST)
From: David Rientjes <rientjes@...gle.com>
To: Lee Schermerhorn <Lee.Schermerhorn@...com>
cc: KOSAKI Motohiro <kosaki.motohiro@...fujitsu.com>,
linux-kernel@...r.kernel.org,
Andrew Morton <akpm@...ux-foundation.org>,
Christoph Lameter <clameter@....com>,
Paul Jackson <pj@....com>, Mel Gorman <mel@....ul.ie>,
torvalds@...ux-foundation.org, Eric Whitney <eric.whitney@...com>
Subject: Re: [2.6.24 regression][BUGFIX] numactl --interleave=all doesn't
works on memoryless node.
On Tue, 5 Feb 2008, Lee Schermerhorn wrote:
> Index: Linux/mm/mempolicy.c
> ===================================================================
> --- Linux.orig/mm/mempolicy.c 2008-02-05 11:25:17.000000000 -0500
> +++ Linux/mm/mempolicy.c 2008-02-05 16:03:11.000000000 -0500
> @@ -131,7 +131,7 @@ static int mpol_check_policy(int mode, n
> return -EINVAL;
> break;
> }
> - return nodes_subset(*nodes, node_states[N_HIGH_MEMORY]) ? 0 : -EINVAL;
> + return 0;
> }
>
> /* Generate a custom zonelist for the BIND policy. */
This change will be necessary when the nodemask passed from the syscall is
saved in the struct mempolicy as the intent of the application as well.
> @@ -188,8 +188,6 @@ static struct mempolicy *mpol_new(int mo
> switch (mode) {
> case MPOL_INTERLEAVE:
> policy->v.nodes = *nodes;
> - nodes_and(policy->v.nodes, policy->v.nodes,
> - node_states[N_HIGH_MEMORY]);
> if (nodes_weight(policy->v.nodes) == 0) {
> kmem_cache_free(policy_cache, policy);
> return ERR_PTR(-EINVAL);
> @@ -426,9 +424,13 @@ static int contextualize_policy(int mode
> if (!nodes)
> return 0;
>
> + /*
> + * Restrict the nodes to the allowed nodes in the cpuset.
> + * This is guaranteed to be a subset of nodes with memory.
> + */
> cpuset_update_task_memory_state();
> - if (!cpuset_nodes_subset_current_mems_allowed(*nodes))
> - return -EINVAL;
> + nodes_and(*nodes, *nodes, cpuset_current_mems_allowed);
> +
> return mpol_check_policy(mode, nodes);
> }
>
I would defer the intersection until later because contextualize_policy()
is called before mpol_new() so we have no struct mempolicy to save the
intent in. It doesn't matter for the sake of this change, I know, but you
could move this intersection to mpol_new() and give us an opportunity to
store the user's nodemask in the mempolicy with a one-line change and get
the same desired result.
You can now remove cpuset_nodes_subset_current_mems_allowed() from
linux/cpuset.h.
> @@ -797,7 +799,7 @@ static long do_mbind(unsigned long start
> if (end == start)
> return 0;
>
> - if (mpol_check_policy(mode, nmask))
> + if (contextualize_policy(mode, nmask))
> return -EINVAL;
>
> new = mpol_new(mode, nmask);
> @@ -915,10 +917,6 @@ asmlinkage long sys_mbind(unsigned long
> err = get_nodes(&nodes, nmask, maxnode);
> if (err)
> return err;
> -#ifdef CONFIG_CPUSETS
> - /* Restrict the nodes to the allowed nodes in the cpuset */
> - nodes_and(nodes, nodes, current->mems_allowed);
> -#endif
> return do_mbind(start, len, mode, &nodes, flags);
> }
>
Looks good, thanks for doing this.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists