From: "Zi Yan" <ziy@nvidia.com>
To: "Barry Song" <21cnbao@gmail.com>, "Juan Yescas" <jyescas@google.com>
Cc: <linux-mm@kvack.org>, <muchun.song@linux.dev>, <rppt@kernel.org>,
<david@redhat.com>, <osalvador@suse.de>,
<akpm@linux-foundation.org>, <lorenzo.stoakes@oracle.com>,
"Jann Horn" <jannh@google.com>, <Liam.Howlett@oracle.com>,
<minchan@kernel.org>, <jaewon31.kim@samsung.com>,
<charante@codeaurora.org>,
"Suren Baghdasaryan" <surenb@google.com>,
"Kalesh Singh" <kaleshsingh@google.com>,
"T.J. Mercier" <tjmercier@google.com>,
"Isaac Manjarres" <isaacmanjarres@google.com>,
<iamjoonsoo.kim@lge.com>, <quic_charante@quicinc.com>
Subject: Re: mm: CMA reservations require 32MiB alignment in 16KiB page size kernels instead of 8MiB in 4KiB page size kernel.
Date: Sun, 19 Jan 2025 19:39:32 -0500 [thread overview]
Message-ID: <D76HCOXO3CMW.8VC1GOASB0WX@nvidia.com> (raw)
In-Reply-To: <CAGsJ_4xL6C07Okt0-5hm-RSBhp9RZxDCnctMxrCZ5z4DD41+kQ@mail.gmail.com>
On Sun Jan 19, 2025 at 6:55 PM EST, Barry Song wrote:
<snip>
>> >>>
>> >>>
>> >>> However, with this workaround, we can't use transparent huge pages.
>> >>>
>> >>> Is the CMA_MIN_ALIGNMENT_BYTES requirement alignment only to support huge pages?
>> No. CMA_MIN_ALIGNMENT_BYTES is limited by CMA_MIN_ALIGNMENT_PAGES, which
>> is equal to pageblock size. Enabling THP just bumps the pageblock size.
>
> Currently, THP might be mTHP, which can have a significantly smaller
> size than 32MB. For
> example, on arm64 systems with a 16KiB page size, a 2MB CONT-PTE mTHP
> is possible.
> Additionally, mTHP relies on the CONFIG_TRANSPARENT_HUGEPAGE configuration.
>
> I wonder if it's possible to enable CONFIG_TRANSPARENT_HUGEPAGE
> without necessarily
> using 32MiB THP. If we use other sizes, such as 64KiB, perhaps a large
> pageblock size wouldn't
> be necessary?
I think this should work by reducing MAX_PAGE_ORDER like Juan did for
the experiment. But MAX_PAGE_ORDER is a macro right now, Kconfig needs
to be changed and kernel needs to be recompiled. Not sure if it is OK
for Juan's use case.
--
Best Regards,
Yan, Zi
next prev parent reply other threads:[~2025-01-20 0:39 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-01-17 22:51 Juan Yescas
2025-01-17 22:52 ` Juan Yescas
2025-01-17 23:00 ` Juan Yescas
2025-01-17 23:19 ` Zi Yan
2025-01-19 23:55 ` Barry Song
2025-01-20 0:39 ` Zi Yan [this message]
2025-01-20 8:14 ` David Hildenbrand
2025-01-20 15:29 ` Zi Yan
2025-01-20 17:59 ` David Hildenbrand
2025-01-22 2:08 ` Juan Yescas
2025-01-22 2:24 ` Zi Yan
2025-01-22 4:06 ` Juan Yescas
2025-01-22 6:52 ` Barry Song
2025-01-22 8:04 ` David Hildenbrand
2025-01-22 8:11 ` David Hildenbrand
2025-01-22 12:49 ` Zi Yan
2025-01-22 13:58 ` David Hildenbrand
2025-01-20 0:17 ` Barry Song
2025-01-20 0:26 ` Zi Yan
2025-01-20 0:38 ` Barry Song
2025-01-20 0:45 ` Zi Yan
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=D76HCOXO3CMW.8VC1GOASB0WX@nvidia.com \
--to=ziy@nvidia.com \
--cc=21cnbao@gmail.com \
--cc=Liam.Howlett@oracle.com \
--cc=akpm@linux-foundation.org \
--cc=charante@codeaurora.org \
--cc=david@redhat.com \
--cc=iamjoonsoo.kim@lge.com \
--cc=isaacmanjarres@google.com \
--cc=jaewon31.kim@samsung.com \
--cc=jannh@google.com \
--cc=jyescas@google.com \
--cc=kaleshsingh@google.com \
--cc=linux-mm@kvack.org \
--cc=lorenzo.stoakes@oracle.com \
--cc=minchan@kernel.org \
--cc=muchun.song@linux.dev \
--cc=osalvador@suse.de \
--cc=quic_charante@quicinc.com \
--cc=rppt@kernel.org \
--cc=surenb@google.com \
--cc=tjmercier@google.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