From: David Hildenbrand <david@redhat.com>
To: Oscar Salvador <osalvador@suse.de>
Cc: Mike Kravetz <mike.kravetz@oracle.com>,
Muchun Song <songmuchun@bytedance.com>,
Michal Hocko <mhocko@kernel.org>,
linux-mm@kvack.org, linux-kernel@vger.kernel.org
Subject: Re: [RFC PATCH 2/2] mm,page_alloc: Make alloc_contig_range handle free hugetlb pages
Date: Wed, 10 Feb 2021 15:36:41 +0100 [thread overview]
Message-ID: <6e999708-2edb-c25f-4aee-217d2f3cc037@redhat.com> (raw)
In-Reply-To: <20210210142424.GC3636@localhost.localdomain>
On 10.02.21 15:24, Oscar Salvador wrote:
> On Wed, Feb 10, 2021 at 09:23:59AM +0100, David Hildenbrand wrote:
>> On 08.02.21 11:38, Oscar Salvador wrote:
>>> Free hugetlb pages are trickier to handle as to in order to guarantee
>>> no userspace appplication disruption, we need to replace the
>>> current free hugepage with a new one.
>>>
>>> In order to do that, a new function called alloc_and_dissolve_huge_page
>>> in introduced.
>>> This function will first try to get a new fresh hugetlb page, and if it
>>> succeeds, it will dissolve the old one.
>>>
>>
>> Thanks for looking into this! Can we move this patch to #1 in the series? It
>> is the easier case.
>>
>> I also wonder if we should at least try on the memory unplug path to keep
>> nr_pages by at least trying to allocate at new one if required, and printing
>> a warning if that fails (after all, we're messing with something configured
>> by the admin - "nr_pages"). Note that gigantic pages are special (below).
>
> So, do you mean to allocate a new fresh hugepage in case we have a free
> hugetlb page within the range we are trying to offline? That makes some
> sense I guess.
>
> I can have a look at that, and make hotplug code use the new
> alloc_and_dissolve().
Yes, with the difference that hotplug code most probably wants to
continue even if allocation failed (printing a warning) - mimix existing
behavior. For alloc_contig, I'd say, fail if we cannot "relocate free
huge pages that are still required to no modify nr_pages".
alloc_and_dissolve() should only allocate a page if really required
(e.g., not sure if we could skip allocation in some cases - like with
surplus pages, needs some investigation), such that the admin-configured
nr_pages stays unchanged.
--
Thanks,
David / dhildenb
next prev parent reply other threads:[~2021-02-10 14:36 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-02-08 10:38 [RFC PATCH 0/2] Make alloc_contig_range handle Hugetlb pages Oscar Salvador
2021-02-08 10:38 ` [RFC PATCH 1/2] mm,page_alloc: Make alloc_contig_range handle in-use hugetlb pages Oscar Salvador
2021-02-10 8:56 ` David Hildenbrand
2021-02-10 14:09 ` Oscar Salvador
2021-02-10 14:11 ` David Hildenbrand
2021-02-10 14:14 ` Oscar Salvador
2021-02-10 14:22 ` David Hildenbrand
2021-02-11 0:56 ` Mike Kravetz
2021-02-11 10:40 ` David Hildenbrand
2021-02-08 10:38 ` [RFC PATCH 2/2] mm,page_alloc: Make alloc_contig_range handle free " Oscar Salvador
2021-02-10 8:23 ` David Hildenbrand
2021-02-10 14:24 ` Oscar Salvador
2021-02-10 14:36 ` David Hildenbrand [this message]
2021-02-25 21:43 ` Mike Kravetz
2021-02-25 22:15 ` David Hildenbrand
2021-02-11 1:16 ` Mike Kravetz
2021-02-11 21:38 ` Oscar Salvador
2021-02-08 10:39 ` [RFC PATCH 0/2] Make alloc_contig_range handle Hugetlb pages 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=6e999708-2edb-c25f-4aee-217d2f3cc037@redhat.com \
--to=david@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mhocko@kernel.org \
--cc=mike.kravetz@oracle.com \
--cc=osalvador@suse.de \
--cc=songmuchun@bytedance.com \
/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).