Re: [PATCH RFC 1/5] sched,numa: build table of node hop distance

2014-10-14 Thread Rik van Riel
On 10/14/2014 02:47 AM, Peter Zijlstra wrote: On Sun, Oct 12, 2014 at 09:28:04AM -0400, Rik van Riel wrote: On 10/12/2014 09:17 AM, Peter Zijlstra wrote: On Wed, Oct 08, 2014 at 03:37:26PM -0400, r...@redhat.com wrote: + sched_domains_numa_hops = kzalloc(sizeof(int) * nr_node_ids *

Re: [PATCH RFC 1/5] sched,numa: build table of node hop distance

2014-10-14 Thread Peter Zijlstra
On Sun, Oct 12, 2014 at 09:28:04AM -0400, Rik van Riel wrote: > On 10/12/2014 09:17 AM, Peter Zijlstra wrote: > >On Wed, Oct 08, 2014 at 03:37:26PM -0400, r...@redhat.com wrote: > >>+ sched_domains_numa_hops = kzalloc(sizeof(int) * nr_node_ids * > >>nr_node_ids, GFP_KERNEL); > >>+ if

Re: [PATCH RFC 1/5] sched,numa: build table of node hop distance

2014-10-14 Thread Peter Zijlstra
On Sun, Oct 12, 2014 at 09:28:04AM -0400, Rik van Riel wrote: On 10/12/2014 09:17 AM, Peter Zijlstra wrote: On Wed, Oct 08, 2014 at 03:37:26PM -0400, r...@redhat.com wrote: + sched_domains_numa_hops = kzalloc(sizeof(int) * nr_node_ids * nr_node_ids, GFP_KERNEL); + if

Re: [PATCH RFC 1/5] sched,numa: build table of node hop distance

2014-10-14 Thread Rik van Riel
On 10/14/2014 02:47 AM, Peter Zijlstra wrote: On Sun, Oct 12, 2014 at 09:28:04AM -0400, Rik van Riel wrote: On 10/12/2014 09:17 AM, Peter Zijlstra wrote: On Wed, Oct 08, 2014 at 03:37:26PM -0400, r...@redhat.com wrote: + sched_domains_numa_hops = kzalloc(sizeof(int) * nr_node_ids *

Re: [PATCH RFC 1/5] sched,numa: build table of node hop distance

2014-10-12 Thread Rik van Riel
On 10/12/2014 09:17 AM, Peter Zijlstra wrote: On Wed, Oct 08, 2014 at 03:37:26PM -0400, r...@redhat.com wrote: + sched_domains_numa_hops = kzalloc(sizeof(int) * nr_node_ids * nr_node_ids, GFP_KERNEL); + if (!sched_domains_numa_hops) + return; That's potentially a

Re: [PATCH RFC 1/5] sched,numa: build table of node hop distance

2014-10-12 Thread Peter Zijlstra
On Wed, Oct 08, 2014 at 03:37:26PM -0400, r...@redhat.com wrote: > + sched_domains_numa_hops = kzalloc(sizeof(int) * nr_node_ids * > nr_node_ids, GFP_KERNEL); > + if (!sched_domains_numa_hops) > + return; That's potentially a _BIG_ table (1M for a 512 node system). The

Re: [PATCH RFC 1/5] sched,numa: build table of node hop distance

2014-10-12 Thread Peter Zijlstra
On Wed, Oct 08, 2014 at 03:37:26PM -0400, r...@redhat.com wrote: + sched_domains_numa_hops = kzalloc(sizeof(int) * nr_node_ids * nr_node_ids, GFP_KERNEL); + if (!sched_domains_numa_hops) + return; That's potentially a _BIG_ table (1M for a 512 node system). The

Re: [PATCH RFC 1/5] sched,numa: build table of node hop distance

2014-10-12 Thread Rik van Riel
On 10/12/2014 09:17 AM, Peter Zijlstra wrote: On Wed, Oct 08, 2014 at 03:37:26PM -0400, r...@redhat.com wrote: + sched_domains_numa_hops = kzalloc(sizeof(int) * nr_node_ids * nr_node_ids, GFP_KERNEL); + if (!sched_domains_numa_hops) + return; That's potentially a

[PATCH RFC 1/5] sched,numa: build table of node hop distance

2014-10-08 Thread riel
From: Rik van Riel In order to more efficiently figure out where to place workloads that span multiple NUMA nodes, it makes sense to estimate how many hops away nodes are from each other. Also add some comments to sched_init_numa. Signed-off-by: Rik van Riel Suggested-by: Peter Zijlstra ---

[PATCH RFC 1/5] sched,numa: build table of node hop distance

2014-10-08 Thread riel
From: Rik van Riel r...@redhat.com In order to more efficiently figure out where to place workloads that span multiple NUMA nodes, it makes sense to estimate how many hops away nodes are from each other. Also add some comments to sched_init_numa. Signed-off-by: Rik van Riel r...@redhat.com