Re: [PATCH v2 3/5] drm/ttm: Introduce ttm_bo_shrink_kswap_fragmented()

["Christian König" <christian.koenig@amd.com>]

On 4/30/26 08:23, Matthew Brost wrote:
> On Tue, Apr 28, 2026 at 12:07:07PM +0200, Andi Shyti wrote:
>> Hi Matt,
>>
>> ...
>>
>>> +/**
>>> + * ttm_bo_shrink_kswap_fragmented() - Whether in kswap and memory appears
>>> + * fragmented
>>> + * @nid: current node being shrunk
>>> + *
>>> + * Return: true if in kswap and memory appears fragmented, false is not.
>>> + */
>>> +bool ttm_bo_shrink_kswap_fragmented(int nid)
>>
>> This name is a bit to strong and misleading, in my opinion.
>>
> 
> s/ttm_bo_shrink_kswap_fragmented/ttm_bo_shrink_kswap_maybe_fragmented
> 

Maybe ttm_bo_partial_shrink_kswap() or something into that direction?

I don't really have any better idea either.

Regards,
Christian.

> ?
> 
>>> +{
>>> +	enum zone_type zone_type;
>>> +
>>> +	if (!current_is_kswapd())
>>> +		return false;
>>> +
>>> +	if (!numa_valid_node(nid))
>>> +		return false;
>>> +
>>> +#if IS_ENABLED(CONFIG_ZONE_DMA32)
>>> +	zone_type = ZONE_DMA32;
>>> +#else
>>> +	zone_type = ZONE_NORMAL;
>>> +#endif
>>> +
>>> +	for (; zone_type <= ZONE_NORMAL; ++zone_type) {
>>
>> just start with "for (zone_type = 0; ... )" and you don't need
>> the ifdefs above.
>>
> 
> I don't think that is correct. Zero could mean ZONE_DMA which is not
> relavent here.
> 
> For example on BMG + raptor lake system, it clear ZONE_DMA is used.
> 
> cat /proc/buddyinfo
> Node 0, zone      DMA      0      1      0      0      0      0      0      0      1      1      3
> Node 0, zone    DMA32     20     10     11     14     15     15     11     13     13      8    232
> Node 0, zone   Normal    876   1782      6      5      4      4      2      6      4      2   7107
> 
> Matt 
> 
>> Andi
>>
>>> +		struct zone *zone = &NODE_DATA(nid)->node_zones[zone_type];