From: David Hildenbrand <david@redhat.com>
To: Yafang Shao <laoar.shao@gmail.com>
Cc: akpm@linux-foundation.org, ziy@nvidia.com,
baolin.wang@linux.alibaba.com, lorenzo.stoakes@oracle.com,
Liam.Howlett@oracle.com, npache@redhat.com, ryan.roberts@arm.com,
dev.jain@arm.com, hannes@cmpxchg.org, usamaarif642@gmail.com,
gutierrez.asier@huawei-partners.com, willy@infradead.org,
ast@kernel.org, daniel@iogearbox.net, andrii@kernel.org,
bpf@vger.kernel.org, linux-mm@kvack.org
Subject: Re: [RFC PATCH v2 0/5] mm, bpf: BPF based THP adjustment
Date: Wed, 28 May 2025 22:32:39 +0200 [thread overview]
Message-ID: <4e0e2874-1dfb-428f-9bff-0a6e174399b7@redhat.com> (raw)
In-Reply-To: <CALOAHbCRc=t9o7HGqxAHpgzKmt4xBYjwQ6zGWZXm2E-zu1SjHQ@mail.gmail.com>
>>
>> I mean, with THPs, there are no guarantees, ever :(
>>
>>> If that's the case, we should revisit RFC v1 [0],
>>> where we proposed rejecting THP allocations in certain scenarios for
>>> specific tasks.
>>
>> Hooking into actual page allocation during page faults (e.g., THP size,
>> khugepaged collapse decisions) is IMHO a much better application of ebpf
>> than setting a THP mode per process (or MM ... ) using epbf.
>>
>> So yes, you could drive the system in "always" mode and decide to not
>> allocate THPs during page faults / khugepaged for specific processes.
>>
>> IMHO that also does not contradict the VM_HUGEPAGE / VM_NOHUGEPAGE
>> default setting proposal: VM_HUGEPAGE could feed into the epbf program
>> as yet another parameter to make a decision.
>
> That seems like a viable solution. Thank you for your help.
Good! And the required ebpf hooks would probably come in handy to write
more advanced policies / allocation logic than just "give it THPs" vs.
"don't give it THPs".
--
Cheers,
David / dhildenb
next prev parent reply other threads:[~2025-05-28 20:32 UTC|newest]
Thread overview: 52+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-05-20 6:04 Yafang Shao
2025-05-20 6:04 ` [RFC PATCH v2 1/5] mm: thp: Add a new mode "bpf" Yafang Shao
2025-05-20 6:05 ` [RFC PATCH v2 2/5] mm: thp: Add hook for BPF based THP adjustment Yafang Shao
2025-05-20 6:05 ` [RFC PATCH v2 3/5] mm: thp: add struct ops " Yafang Shao
2025-05-20 6:05 ` [RFC PATCH v2 4/5] bpf: Add get_current_comm to bpf_base_func_proto Yafang Shao
2025-05-20 23:32 ` Andrii Nakryiko
2025-05-20 6:05 ` [RFC PATCH v2 5/5] selftests/bpf: Add selftest for THP adjustment Yafang Shao
2025-05-20 6:52 ` [RFC PATCH v2 0/5] mm, bpf: BPF based " Nico Pache
2025-05-20 7:25 ` Yafang Shao
2025-05-20 13:10 ` Matthew Wilcox
2025-05-20 14:08 ` Yafang Shao
2025-05-20 14:22 ` Lorenzo Stoakes
2025-05-20 14:32 ` Usama Arif
2025-05-20 14:35 ` Lorenzo Stoakes
2025-05-20 14:42 ` Matthew Wilcox
2025-05-20 14:56 ` David Hildenbrand
2025-05-21 4:28 ` Yafang Shao
2025-05-20 14:46 ` Usama Arif
2025-05-20 15:00 ` David Hildenbrand
2025-05-20 9:43 ` David Hildenbrand
2025-05-20 9:49 ` Lorenzo Stoakes
2025-05-20 12:06 ` Yafang Shao
2025-05-20 13:45 ` Lorenzo Stoakes
2025-05-20 15:54 ` David Hildenbrand
2025-05-21 4:02 ` Yafang Shao
2025-05-21 3:52 ` Yafang Shao
2025-05-20 11:59 ` Yafang Shao
2025-05-25 3:01 ` Yafang Shao
2025-05-26 7:41 ` Gutierrez Asier
2025-05-26 9:37 ` Yafang Shao
2025-05-26 8:14 ` David Hildenbrand
2025-05-26 9:37 ` Yafang Shao
2025-05-26 10:49 ` David Hildenbrand
2025-05-26 14:53 ` Liam R. Howlett
2025-05-26 15:54 ` Liam R. Howlett
2025-05-26 16:51 ` David Hildenbrand
2025-05-26 17:07 ` Liam R. Howlett
2025-05-26 17:12 ` David Hildenbrand
2025-05-26 20:30 ` Gutierrez Asier
2025-05-26 20:37 ` David Hildenbrand
2025-05-27 5:46 ` Yafang Shao
2025-05-27 7:57 ` David Hildenbrand
2025-05-27 8:13 ` Yafang Shao
2025-05-27 8:30 ` David Hildenbrand
2025-05-27 8:40 ` Yafang Shao
2025-05-27 9:27 ` David Hildenbrand
2025-05-27 9:43 ` Yafang Shao
2025-05-27 12:19 ` David Hildenbrand
2025-05-28 2:04 ` Yafang Shao
2025-05-28 20:32 ` David Hildenbrand [this message]
2025-05-26 14:32 ` Zi Yan
2025-05-27 5:53 ` Yafang Shao
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=4e0e2874-1dfb-428f-9bff-0a6e174399b7@redhat.com \
--to=david@redhat.com \
--cc=Liam.Howlett@oracle.com \
--cc=akpm@linux-foundation.org \
--cc=andrii@kernel.org \
--cc=ast@kernel.org \
--cc=baolin.wang@linux.alibaba.com \
--cc=bpf@vger.kernel.org \
--cc=daniel@iogearbox.net \
--cc=dev.jain@arm.com \
--cc=gutierrez.asier@huawei-partners.com \
--cc=hannes@cmpxchg.org \
--cc=laoar.shao@gmail.com \
--cc=linux-mm@kvack.org \
--cc=lorenzo.stoakes@oracle.com \
--cc=npache@redhat.com \
--cc=ryan.roberts@arm.com \
--cc=usamaarif642@gmail.com \
--cc=willy@infradead.org \
--cc=ziy@nvidia.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