On 1/6/16 2:32 AM, Khalid Aziz wrote:
> On 01/05/2016 02:40 AM, Nitin Gupta wrote:
>> Orabug: 22495713
>>
>> Currently, NUMA node distance matrix is initialized only
>> when a machine descriptor (MD) exists. However, sun4u
>> machines (e.g. Sun Blade 2500) do not have an MD and thus
>> distance values were left uninitialized. The initialization
>> is now moved such that it happens on both sun4u and sun4v.
>>
>> Signed-off-by: Nitin Gupta <nitin.m.gupta@xxxxxxxxxx>
>> Tested-by: Mikael Pettersson <mikpelinux@xxxxxxxxx>
>> ---
>> arch/sparc/mm/init_64.c | 15 ++++++++-------
>> 1 file changed, 8 insertions(+), 7 deletions(-)
>>
>> diff --git a/arch/sparc/mm/init_64.c b/arch/sparc/mm/init_64.c
>> index 3025bd5..ff63db5 100644
>> --- a/arch/sparc/mm/init_64.c
>> +++ b/arch/sparc/mm/init_64.c
>> @@ -1267,13 +1267,6 @@ static int __init numa_parse_mdesc(void)
>> int i, j, err, count;
>> u64 node;
>>
>> - /* Some sane defaults for numa latency values */
>> - for (i = 0; i < MAX_NUMNODES; i++) {
>> - for (j = 0; j < MAX_NUMNODES; j++)
>> - numa_latency[i][j] = (i == j) ?
>> - LOCAL_DISTANCE : REMOTE_DISTANCE;
>> - }
>> -
>> node = mdesc_node_by_name(md, MDESC_NODE_NULL, "latency-groups");
>> if (node == MDESC_NODE_NULL) {
>> mdesc_release(md);
>> @@ -1374,6 +1367,14 @@ static int __init bootmem_init_numa(void)
>> numadbg("bootmem_init_numa()\n");
>>
>> if (numa_enabled) {
>> + int i, j;
>> + /* Some sane defaults for numa latency values */
>> + for (i = 0; i < MAX_NUMNODES; i++) {
>> + for (j = 0; j < MAX_NUMNODES; j++)
>> + numa_latency[i][j] = (i == j) ?
>> + LOCAL_DISTANCE : REMOTE_DISTANCE;
>> + }
>> +
>> if (tlb_type == hypervisor)
>> err = numa_parse_mdesc();
>> else
>>
>
> If one were to boot up with "numa=off" kernel option, numa_latency array wouldn't be initialized by this code which will cause node_distance() to return 0 for any two nodes. Would it make more sense to do this initialization outside of "if (numa_enabled)" or is a value of 0, instead of LOCAL_DISTANCE when numa is not enabled, for node_distance() not a problem?
>
I see that in places in page_alloc, numa_distance() is used based just on
compile time CONFIG_NUMA. So, yes the initialization should be moved outside
of dynamic config 'numa_enabled'. I will correct this in next patch v2.
Thanks,
Nitin
--
To unsubscribe from this list: send the line "unsubscribe sparclinux" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
