kswapd0: page allocation failure: order:0, mode:0x820(GFP_ATOMIC), nodemask=(null),cpuset=/,mems_allowed=0 (Kernel v6.5.9, 32bit ppc)

[Vlastimil Babka (SUSE)]

On 6/4/24 8:01 PM, Yosry Ahmed wrote:
> On Tue, Jun 4, 2024 at 10:54 AM Yu Zhao <yuzhao at google.com> wrote:
>> There was a lot of user memory in the DMA zone. So at a point the
>> highmem zone was full and allocation fallback happened.
>>
>> The problem with zone fallback is that recent allocations go into
>> lower zones, meaning they are further back on the LRU list. This
>> applies to both user memory and zsmalloc memory -- the latter has a
>> writeback LRU. On top of this, neither the zswap shrinker nor the
>> zsmalloc shrinker (compaction) is zone aware. So page reclaim might
>> have trouble hitting the right target zone.
> 
> I see what you mean. In this case, yeah I think the internal
> fragmentation in the zsmalloc pools may be the reason behind the
> problem.
> 
> How many CPUs does this machine have? I am wondering if 32 can be an
> overkill for small machines, perhaps the number of pools should be
> max(nr_cpus, 32)?
> 
> Alternatively, the number of pools should scale with the memory size
> in some way, such that we only increase fragmentation when it's
> tolerable.

Sounds like a good idea to me, maybe a combination of both. No point in
trying to scale if there's no benefit and only downside of more memory
consumption.