Re: [PATCH v5 v5 4/6] mm/sparse-vmemmap: Fix DAX vmemmap accounting with optimization

[Muchun Song]

> On Apr 23, 2026, at 20:32, Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> wrote:
> 
> On Thu, 23 Apr 2026 20:04:21 +0800 Muchun Song <muchun.song@xxxxxxxxx> wrote:
> 
>> Ah, looks like I need to update the patch to fix this.
> 
> OK.  Please have a think about which patches we should backport into
> -stable.  Possibly all 6!  And the timing of this backport.

In theory, I think all of these require backporting, and each carries a certain
level of potential impact.

> 
> Ideally, split the series up into one series for 7.1-rcX+stable, another
> series for 7.2.rc1.

Given that these issues have existed for a long time and the trigger
frequency is very low, I suggest we take a more conservative timing.

We can prioritize getting the series into 7.2-rc1 first. Once they have
'baked' in the mainline for a while without issues, we can then proceed
with the backport to -stable (7.1.x). This avoids rushing risky changes
into stable branches.

Thanks,
Muchun