From: David Hildenbrand <david@redhat.com>
To: Michal Hocko <mhocko@suse.com>
Cc: Oscar Salvador <osalvador@suse.de>,
Andrew Morton <akpm@linux-foundation.org>,
Anshuman Khandual <anshuman.khandual@arm.com>,
Vlastimil Babka <vbabka@suse.cz>,
Pavel Tatashin <pasha.tatashin@soleen.com>,
linux-mm@kvack.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH v5 1/5] mm,memory_hotplug: Allocate memmap from the added memory range
Date: Wed, 24 Mar 2021 15:05:54 +0100 [thread overview]
Message-ID: <0e58dce3-d68f-f8cb-936e-c7ae49c1eeaa@redhat.com> (raw)
In-Reply-To: <YFtBTeBBCPnBesQ5@dhcp22.suse.cz>
On 24.03.21 14:40, Michal Hocko wrote:
> On Wed 24-03-21 14:13:31, David Hildenbrand wrote:
>> On 24.03.21 13:37, Michal Hocko wrote:
>>> On Wed 24-03-21 13:23:47, David Hildenbrand wrote:
>>>> On 24.03.21 13:10, Michal Hocko wrote:
>>>>> On Wed 24-03-21 13:03:29, Michal Hocko wrote:
>>>>>> On Wed 24-03-21 11:12:59, Oscar Salvador wrote:
>>>>> [...]
>>>>>
>>>>> an additional remark
>>>>>
>>>>>>> - online_pages()->move_pfn_range_to_zone(): Accounts for node/zone's spanned pages
>>>>>>> - online_pages()->zone->present_pages += nr_pages;
>>>>>
>>>>> I am pretty sure you shouldn't account vmmemmap pages to the target zone
>>>>> in some cases - e.g. vmemmap cannot be part of the movable zone, can it?
>>>>> So this would be yet another special casing. This patch has got it wrong
>>>>> unless I have missed some special casing.
>>>>>
>>>>
>>>> It's a bit unfortunate that we have to discuss the very basic design
>>>> decisions again.
>>>
>>> It would be great to have those basic design decisions layed out in the
>>> changelog.
>>>
>>>> @Oscar, maybe you can share the links where we discussed all this and add
>>>> some of it to the patch description.
>>>>
>>>> I think what we have right here is good enough for an initial version, from
>>>> where on we can improve things without having to modify calling code.
>>>
>>> I have to say I really dislike vmemmap proliferation into
>>> {on,off}lining. It just doesn't belong there from a layering POV. All
>>> this code should care about is to hand over pages to the allocator and
>>> make them visible.
>>
>> Well, someone has to initialize the vmemmap of the vmemmap pages ( which is
>> itself :) ),
>
> Yeah, and I would expect this to be done when the vmemmap space is
> reserved. This is at the hotadd time and we do not know the zone but
> that shouldn't really matter because their zone can be quite arbitrary
> kernel zone. As mentioned previously I do not think associating those
> with zone movable is a good idea as they are fundamentally not movable.
I don't think that's an issue. Just another special case to keep things
simple. (and not completely fragment zones, mess with zone shrinking etc.)
> It is likely that the zone doesn't really matter for these pages anyway
> and the only think we do care about is that they are not poisoned and
> there is at least something but again it would be much better to have a
> single place where all those details are done (including accounting)
> rather than trying to wrap head around different pfns when onlining
> pages and grow potential and suble bugs there.
Exactly, as you said, the zone doesn't really matter - thus, this patch
just handles it as simple as possible: keep them in the same zone as the
hole memory block. No fragmented zones. no special casing. simple.
Details are actually pretty much all at a single place when
onlining/offlining().
>
>> and as the vemmap does not span complete sections things can
>> get very weird as we can only set whole sections online (there was more to
>> that, I think it's buried in previous discussions).
>
> Yes the section can only online as whole. This is an important "detail"
> and it would deserve some more clarification in the changelog as well.
Indeed.
> You have invested quite some energy into code consolidation and checks
> to make sure that hotplugged code doesn't have holes and this work bends
> those rules. vmemmap is effectivelly a hole in a memblock/section. I
> think we should re-evaluate some of those constrains rather than try to
> work them around somehow.
It's an offset in the beginning, so it's a special case. And the
question is if there is a real benefit in handling it differently, for
example, messing with online sections, messing with zones .. I am not
convinced that the added complexity gives us a real benefit. But I shall
be taught otherwise.
BTW: I once thought about having
online_memory_block(block)/offline_memory_block(block) as separate
functions instead of having pretty generic (error prone?)
online_pages()/offline_pages(). Then, these details would just go in
there and memory blocks + online/offline logic would simply be
self-contained.
--
Thanks,
David / dhildenb
next prev parent reply other threads:[~2021-03-24 14:06 UTC|newest]
Thread overview: 63+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-03-19 9:26 [PATCH v5 0/5] Allocate memmap from hotadded memory (per device) Oscar Salvador
2021-03-19 9:26 ` [PATCH v5 1/5] mm,memory_hotplug: Allocate memmap from the added memory range Oscar Salvador
2021-03-19 10:20 ` David Hildenbrand
2021-03-19 10:31 ` Oscar Salvador
2021-03-19 12:04 ` David Hildenbrand
2021-03-23 10:11 ` Michal Hocko
2021-03-24 10:12 ` Oscar Salvador
2021-03-24 12:03 ` Michal Hocko
2021-03-24 12:10 ` Michal Hocko
2021-03-24 12:23 ` David Hildenbrand
2021-03-24 12:37 ` Michal Hocko
2021-03-24 13:13 ` David Hildenbrand
2021-03-24 13:40 ` Michal Hocko
2021-03-24 14:05 ` David Hildenbrand [this message]
2021-03-24 13:27 ` Oscar Salvador
2021-03-24 14:42 ` Michal Hocko
2021-03-24 14:52 ` David Hildenbrand
2021-03-24 16:04 ` Michal Hocko
2021-03-24 19:16 ` David Hildenbrand
2021-03-25 8:07 ` Oscar Salvador
2021-03-25 9:17 ` Michal Hocko
2021-03-25 10:55 ` Oscar Salvador
2021-03-25 11:08 ` David Hildenbrand
2021-03-25 11:23 ` Oscar Salvador
2021-03-25 12:35 ` Michal Hocko
2021-03-25 12:40 ` David Hildenbrand
2021-03-25 14:08 ` Michal Hocko
2021-03-25 14:09 ` David Hildenbrand
2021-03-25 14:34 ` Michal Hocko
2021-03-25 14:46 ` David Hildenbrand
2021-03-25 15:12 ` Michal Hocko
2021-03-25 15:19 ` David Hildenbrand
2021-03-25 15:35 ` Michal Hocko
2021-03-25 15:40 ` David Hildenbrand
2021-03-25 16:07 ` Michal Hocko
2021-03-25 16:20 ` David Hildenbrand
2021-03-25 16:36 ` Michal Hocko
2021-03-25 16:47 ` Michal Hocko
2021-03-25 16:55 ` David Hildenbrand
2021-03-25 22:06 ` Oscar Salvador
2021-03-26 8:35 ` Michal Hocko
2021-03-26 8:52 ` David Hildenbrand
2021-03-26 8:57 ` Oscar Salvador
2021-03-26 12:15 ` Oscar Salvador
2021-03-26 13:36 ` David Hildenbrand
2021-03-26 14:38 ` Michal Hocko
2021-03-26 14:53 ` David Hildenbrand
2021-03-26 15:31 ` Michal Hocko
2021-03-26 16:03 ` David Hildenbrand
2021-03-26 8:55 ` Oscar Salvador
2021-03-26 9:11 ` Michal Hocko
2021-03-25 18:08 ` David Hildenbrand
2021-03-25 12:26 ` Michal Hocko
2021-03-25 14:02 ` Oscar Salvador
2021-03-25 14:40 ` Michal Hocko
2021-03-19 9:26 ` [PATCH v5 2/5] acpi,memhotplug: Enable MHP_MEMMAP_ON_MEMORY when supported Oscar Salvador
2021-03-23 10:40 ` Michal Hocko
2021-03-19 9:26 ` [PATCH v5 3/5] mm,memory_hotplug: Add kernel boot option to enable memmap_on_memory Oscar Salvador
2021-03-23 10:47 ` Michal Hocko
2021-03-24 8:45 ` Oscar Salvador
2021-03-24 9:02 ` Michal Hocko
2021-03-19 9:26 ` [PATCH v5 4/5] x86/Kconfig: Introduce ARCH_MHP_MEMMAP_ON_MEMORY_ENABLE Oscar Salvador
2021-03-19 9:26 ` [PATCH v5 5/5] arm64/Kconfig: " Oscar Salvador
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=0e58dce3-d68f-f8cb-936e-c7ae49c1eeaa@redhat.com \
--to=david@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=anshuman.khandual@arm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mhocko@suse.com \
--cc=osalvador@suse.de \
--cc=pasha.tatashin@soleen.com \
--cc=vbabka@suse.cz \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).