On 29 Oct 2020, at 20:28, Andrew Morton wrote:

> On Thu, 29 Oct 2020 17:31:28 -0400 Zi Yan <ziy@nvidia.com> wrote:
>
>>>
>>> Shall you add Fixes tag to commit
>>> 1da2f328fa643bd72197dfed0c655148af31e4eb? And may cc stable.
>>
>> Sure.
>>
>> Fixes: 1da2f328fa64 (“mm,thp,compaction,cma: allow THP migration for CMA allocations”)
>>
>> stable cc'ed.
>
> A think a cc:stable really requires a description of the end-user
> visible effects of the bug.  Could you please provide that?

Sure.

For example, in a system with 16GB memory and an 8GB CMA region reserved by hugetlb_cma,
if we first allocate 10GB THPs and mlock them (so some THPs are allocated in the CMA
region and mlocked), reserving 6 1GB hugetlb pages via
/sys/kernel/mm/hugepages/hugepages-1048576kB/nr_hugepages will get stuck (looping
in too_many_isolated function) until we kill either task. With the patch applied,
oom will kill the application with 10GB THPs and let hugetlb page reservation finish.

—
Best Regards,
Yan Zi