On 08/03/2018 05:02 AM, Andy Shevchenko wrote:
> On Thu, Aug 2, 2018 at 10:58 PM, Tony Battersby <tonyb@xxxxxxxxxxxxxxx> wrote:
>> dma_pool_alloc() scales poorly when allocating a large number of pages
>> because it does a linear scan of all previously-allocated pages before
>> allocating a new one. Improve its scalability by maintaining a separate
>> list of pages that have free blocks ready to (re)allocate. In big O
>> notation, this improves the algorithm from O(n^2) to O(n).
>> struct dma_pool { /* the pool */
>> +#define POOL_FULL_IDX 0
>> +#define POOL_AVAIL_IDX 1
>> +#define POOL_N_LISTS 2
>> + struct list_head page_list[POOL_N_LISTS];
> To be consistent with naming scheme and common practice I would rather
> name the last one as
>
> POOL_MAX_IDX 2
OK.
>
>> + INIT_LIST_HEAD(&retval->page_list[0]);
>> + INIT_LIST_HEAD(&retval->page_list[1]);
> You introduced defines and don't use them.
>
Just a matter of style. In this context, it only matters that both
index 0 and 1 get initialized, not which index corresponds to which
list. But I suppose using the defines would improve keyword search, so
I'll change it.
