Re: [PATCH v2 00/14] mm: Asynchronous + multithreaded memmap init for ZONE_DEVICE

[Dan Williams]

On Tue, Jul 17, 2018 at 8:50 AM Michal Hocko <mhocko@xxxxxxxxxx> wrote:
>
> On Tue 17-07-18 10:46:39, Pavel Tatashin wrote:
> > > > Hi Dan,
> > > >
> > > > I am worried that this work adds another way to multi-thread struct
> > > > page initialization without re-use of already existing method. The
> > > > code is already a mess, and leads to bugs [1] because of the number of
> > > > different memory layouts, architecture specific quirks, and different
> > > > struct page initialization methods.
> > >
> > > Yes, the lamentations about the complexity of the memory hotplug code
> > > are known. I didn't think this set made it irretrievably worse, but
> > > I'm biased and otherwise certainly want to build consensus with other
> > > mem-hotplug folks.
> > >
> > > >
> > > > So, when DEFERRED_STRUCT_PAGE_INIT is used we initialize struct pages
> > > > on demand until page_alloc_init_late() is called, and at that time we
> > > > initialize all the rest of struct pages by calling:
> > > >
> > > > page_alloc_init_late()
> > > >   deferred_init_memmap() (a thread per node)
> > > >     deferred_init_pages()
> > > >        __init_single_page()
> > > >
> > > > This is because memmap_init_zone() is not multi-threaded. However,
> > > > this work makes memmap_init_zone() multi-threaded. So, I think we
> > > > should really be either be using deferred_init_memmap() here, or teach
> > > > DEFERRED_STRUCT_PAGE_INIT to use new multi-threaded memmap_init_zone()
> > > > but not both.
> > >
> > > I agree it would be good to look at unifying the 2 async
> > > initialization approaches, however they have distinct constraints. All
> > > of the ZONE_DEVICE memmap initialization work happens as a hotplug
> > > event where the deferred_init_memmap() threads have already been torn
> > > down. For the memory capacities where it takes minutes to initialize
> > > the memmap it is painful to incur a global flush of all initialization
> > > work. So, I think that a move to rework deferred_init_memmap() in
> > > terms of memmap_init_async() is warranted because memmap_init_async()
> > > avoids a global sync and supports the hotplug case.
> > >
> > > Unfortunately, the work to unite these 2 mechanisms is going to be
> > > 4.20 material, at least for me, since I'm taking an extended leave,
> > > and there is little time for me to get this in shape for 4.19. I
> > > wouldn't be opposed to someone judiciously stealing from this set and
> > > taking a shot at the integration, I likely will not get back to this
> > > until September.
> >
> > Hi Dan,
> >
> > I do not want to hold your work, so if Michal or Andrew are OK with
> > the general approach of teaching    memmap_init_zone() to be async
> > without re-using deferred_init_memmap() or without changing
> > deferred_init_memmap() to use the new memmap_init_async() I will
> > review your patches.
>
> Well, I would rather have a sane code base than rush anything in. I do
> agree with Pavel that we the number of async methods we have right now
> is really disturbing. Applying yet another one will put additional
> maintenance burden on whoever comes next.

I thought we only had the one async implementation presently, this
makes it sound like we have more than one? Did I miss the other(s)?

> Is there any reason that this work has to target the next merge window?
> The changelog is not really specific about that.

Same reason as any other change in this space, hardware availability
continues to increase. These patches are a direct response to end user
reports of unacceptable init latency with current kernels.

> There no numbers or
> anything that would make this sound as a high priority stuff.

>From the end of the cover letter:

"With this change an 8 socket system was observed to initialize pmem
namespaces in ~4 seconds whereas it was previously taking ~4 minutes."

My plan if this is merged would be to come back and refactor it with
the deferred_init_memmap() implementation, my plan if this is not
merged would be to come back and refactor it with the
deferred_init_memmap() implementation.

In practical terms, 0day has noticed a couple minor build fixes are needed:
https://lists.01.org/pipermail/kbuild-all/2018-July/050229.html
https://lists.01.org/pipermail/kbuild-all/2018-July/050231.html

...and I'm going to be offline until September. I thought it best to
post this before I go, and I'm open to someone else picking up this
work to get in shape for merging per community feedback.