[dpdk-dev] [RFC 03/11] malloc: index heaps using heap ID rather than NUMA node

[Burakov, Anatoly]

On 13-Jul-18 5:05 PM, Alejandro Lucero wrote:
>> -       /* get pointer to global configuration */
>> -       mcfg = rte_eal_get_configuration()->mem_config;
>> -
>> -       for (i = 0; i < RTE_MAX_NUMA_NODES; i++) {
>> -               if ((socket != SOCKET_ID_ANY) && (socket != i))
>> +       for (idx = 0; idx < rte_socket_count(); idx++) {
>> +               int cur_socket = rte_socket_id_by_idx(idx);
>> +               if ((socket != SOCKET_ID_ANY) && (socket != cur_socket))
>>                          continue;
>>
>> -               malloc_heap_get_stats(&mcfg->malloc_heaps[i], &stats);
>> -               if (stats.greatest_free_size > len) {
>> -                       len = stats.greatest_free_size;
>> -                       *s = i;
>> +               size_t cur_len = heap_max_free_elem(idx, align);
>> +               if (cur_len > len) {
>> +                       len = cur_len;
>> +                       *s = cur_socket;
>>                  }
>>          }
>>
>> -       if (len < MALLOC_ELEM_OVERHEAD + align)
>> -               return 0;
>> -
>> -       return len - MALLOC_ELEM_OVERHEAD - align;
>> +       return len;
>>
> 
> Is it worth to set *s to some safe value if no space at all?

No, the value of *s is set externally anyway, and is not used of return 
value is 0.

Thanks for other comments, will fix when the next iteration comes.


-- 
Thanks,
Anatoly