From: Hou Tao <houtao@huaweicloud.com>
To: linux-kernel@vger.kernel.org
Cc: linux-pci@vger.kernel.org, linux-mm@kvack.org,
linux-nvme@lists.infradead.org,
Bjorn Helgaas <bhelgaas@google.com>,
Logan Gunthorpe <logang@deltatee.com>,
Alistair Popple <apopple@nvidia.com>,
Leon Romanovsky <leonro@nvidia.com>,
Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
Tejun Heo <tj@kernel.org>,
"Rafael J . Wysocki" <rafael@kernel.org>,
Danilo Krummrich <dakr@kernel.org>,
Andrew Morton <akpm@linux-foundation.org>,
David Hildenbrand <david@kernel.org>,
Lorenzo Stoakes <lorenzo.stoakes@oracle.com>,
Keith Busch <kbusch@kernel.org>, Jens Axboe <axboe@kernel.dk>,
Christoph Hellwig <hch@lst.de>, Sagi Grimberg <sagi@grimberg.me>,
houtao1@huawei.com
Subject: [PATCH 00/13] Enable compound page for p2pdma memory
Date: Sat, 20 Dec 2025 12:04:33 +0800 [thread overview]
Message-ID: <20251220040446.274991-1-houtao@huaweicloud.com> (raw)
From: Hou Tao <houtao1@huawei.com>
Hi,
device-dax has already supported compound page. It not only reduces the
cost of struct page significantly, it also improve the performance of
get_user_pages when 2MB or 1GB page size is used. We are experimenting
to use p2p dma to directly transfer the content of NVMe SSD into NPU.
The size of NPU HBM is 32GB or larger and there are at most 8 NPUs in
the host. When using the base page, the memory overhead is about 4GB for
128GB HBM, and the mapping of 32GB HBM into userspace takes about 0.8
second. Considering ZONE_DEVICE memory type has already supported the
compound page, enabling the compound page support for p2pdma memory as
well. After applying the patch set, when using the 1GB page, the memory
overhead is about 2MB and the mmap costs about 0.04 ms.
The main difference between the compound page support of device-dax and
p2pdma is that p2pdma inserts the page into user vma during mmap instead
of page fault. The main reason is simplicity. The patch set is
structured as shown below:
Patch #1~#2: tiny bug fixes for p2pdma
Patch #3~#5: add callbacks support in kernfs and sysfs, include
pagesize, may_split and get_unmapped_area. These callbacks are necessary
for the support of compound page when mmaping sysfs binary file.
Patch #6~#7: create compound page for p2pdma memory in the kernel.
Patch #8~#10: support the mapping of compound page in userspace.
Patch #11~#12: support the compound page for NVMe CMB.
Patch #13: enable the support for compound page for p2pdma memory.
Please see individual patches for more details. Comments and
suggestions are always welcome.
Hou Tao (13):
PCI/P2PDMA: Release the per-cpu ref of pgmap when vm_insert_page()
fails
PCI/P2PDMA: Fix the warning condition in p2pmem_alloc_mmap()
kernfs: add support for get_unmapped_area callback
kernfs: add support for may_split and pagesize callbacks
sysfs: support get_unmapped_area callback for binary file
PCI/P2PDMA: add align parameter for pci_p2pdma_add_resource()
PCI/P2PDMA: create compound page for aligned p2pdma memory
mm/huge_memory: add helpers to insert huge page during mmap
PCI/P2PDMA: support get_unmapped_area to return aligned vaddr
PCI/P2PDMA: support compound page in p2pmem_alloc_mmap()
PCI/P2PDMA: add helper pci_p2pdma_max_pagemap_align()
nvme-pci: introduce cmb_devmap_align module parameter
PCI/P2PDMA: enable compound page support for p2pdma memory
drivers/accel/habanalabs/common/hldio.c | 3 +-
drivers/nvme/host/pci.c | 10 +-
drivers/pci/p2pdma.c | 140 ++++++++++++++++++++++--
fs/kernfs/file.c | 79 +++++++++++++
fs/sysfs/file.c | 15 +++
include/linux/huge_mm.h | 4 +
include/linux/kernfs.h | 3 +
include/linux/pci-p2pdma.h | 30 ++++-
include/linux/sysfs.h | 4 +
mm/huge_memory.c | 66 +++++++++++
10 files changed, 339 insertions(+), 15 deletions(-)
--
2.29.2
next reply other threads:[~2025-12-20 4:16 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-12-20 4:04 Hou Tao [this message]
2025-12-20 4:04 ` [PATCH 01/13] PCI/P2PDMA: Release the per-cpu ref of pgmap when vm_insert_page() fails Hou Tao
2025-12-22 16:49 ` Logan Gunthorpe
2025-12-20 4:04 ` [PATCH 02/13] PCI/P2PDMA: Fix the warning condition in p2pmem_alloc_mmap() Hou Tao
2025-12-22 16:50 ` Logan Gunthorpe
2025-12-20 4:04 ` [PATCH 03/13] kernfs: add support for get_unmapped_area callback Hou Tao
2025-12-20 15:43 ` kernel test robot
2025-12-20 15:57 ` kernel test robot
2025-12-20 4:04 ` [PATCH 04/13] kernfs: add support for may_split and pagesize callbacks Hou Tao
2025-12-20 4:04 ` [PATCH 05/13] sysfs: support get_unmapped_area callback for binary file Hou Tao
2025-12-20 4:04 ` [PATCH 06/13] PCI/P2PDMA: add align parameter for pci_p2pdma_add_resource() Hou Tao
2025-12-20 4:04 ` [PATCH 07/13] PCI/P2PDMA: create compound page for aligned p2pdma memory Hou Tao
2025-12-20 4:04 ` [PATCH 08/13] mm/huge_memory: add helpers to insert huge page during mmap Hou Tao
2025-12-20 4:04 ` [PATCH 09/13] PCI/P2PDMA: support get_unmapped_area to return aligned vaddr Hou Tao
2025-12-20 4:04 ` [PATCH 10/13] PCI/P2PDMA: support compound page in p2pmem_alloc_mmap() Hou Tao
2025-12-22 17:04 ` Logan Gunthorpe
2025-12-24 2:20 ` Hou Tao
2025-12-20 4:04 ` [PATCH 11/13] PCI/P2PDMA: add helper pci_p2pdma_max_pagemap_align() Hou Tao
2025-12-20 4:04 ` [PATCH 12/13] nvme-pci: introduce cmb_devmap_align module parameter Hou Tao
2025-12-20 22:22 ` kernel test robot
2025-12-20 4:04 ` [PATCH 13/13] PCI/P2PDMA: enable compound page support for p2pdma memory Hou Tao
2025-12-22 17:10 ` Logan Gunthorpe
2025-12-21 12:19 ` [PATCH 00/13] Enable compound page " Leon Romanovsky
[not found] ` <416b2575-f5e7-7faf-9e7c-6e9df170bf1a@huaweicloud.com>
2025-12-24 1:37 ` Hou Tao
2025-12-24 9:22 ` Leon Romanovsky
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=20251220040446.274991-1-houtao@huaweicloud.com \
--to=houtao@huaweicloud.com \
--cc=akpm@linux-foundation.org \
--cc=apopple@nvidia.com \
--cc=axboe@kernel.dk \
--cc=bhelgaas@google.com \
--cc=dakr@kernel.org \
--cc=david@kernel.org \
--cc=gregkh@linuxfoundation.org \
--cc=hch@lst.de \
--cc=houtao1@huawei.com \
--cc=kbusch@kernel.org \
--cc=leonro@nvidia.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linux-nvme@lists.infradead.org \
--cc=linux-pci@vger.kernel.org \
--cc=logang@deltatee.com \
--cc=lorenzo.stoakes@oracle.com \
--cc=rafael@kernel.org \
--cc=sagi@grimberg.me \
--cc=tj@kernel.org \
/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