Re: [PATCH 0/4] mm: misc cleanups from __GFP_UNMAPPED series

[Vlastimil Babka (SUSE)]

On 5/18/26 22:58, Brendan Jackman wrote:
> On Mon, 18 May 2026 at 13:48, Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> wrote:
>>
>> On Mon, 18 May 2026 20:20:50 +0000 Brendan Jackman <jackmanb@xxxxxxxxxx> wrote:
>>
>> > On Mon May 18, 2026 at 6:15 PM UTC, Andrew Morton wrote:
>> > > On Wed, 13 May 2026 12:35:12 +0000 Brendan Jackman <jackmanb@xxxxxxxxxx> wrote:
>> > >
>> > >> In v2 of the __GFP_UNMAPPED series [0], we realised that some of the patches
>> > >> could potentially be merged as independent cleanups.
>> > >>
>> > >> These are all independent of one another, if you think some are useful
>> > >> cleanups and others are pointless churn, it should be fine to just pick
>> > >> whatever subset you prefer.
>> > >>
>> > >> No functional change intended.
>> > >
>> > > Thanks, Brendan.  I added this to mm.git's mm-new branch for testing,
>> > > along with a note that [4/4] may be getting an update arising from
>> > > Vlastimil's feedback.
>> >
>> > Cool thanks, (also I got the "added to mm-new branch notification" and
>> > realised why I missed it last time, broken mail filter).
>> >
>> > The 4/4 update is already up, in reply to the thread.
>>
>> Is that "mm/page_alloc: tidy up pindex helpers"?
>>
>> For some reason I didn't receive that email, directly or via any
>> mailing list.  I do have the similar "mm/page_alloc: remove ifdefs from
>> pindex helpers" from May 13.

Found it, seems to be threaded well with In-reply-to: too, maybe it's the
different subject?

https://lore.kernel.org/all/20260517230556.590677-1-jackmanb@xxxxxxxxxx/

Should be fine to replace the current "mm/page_alloc: remove ifdefs from
pindex helpers" in mm.git with that.

> Ooh, that's unfortunate... I used git-send-email directl to send that,
> which is not how I work most of the time, so I suspect something went
> wrong with that particular setup. Will keep an eye on that.
> 
>> Oh well, please check tomorrow's mm.git and send along whatever is
>> needed?
> 
> Ack, thanks again.