From: Christoph Hellwig <hch@lst.de>
To: Ralph Campbell <rcampbell@nvidia.com>
Cc: Christoph Hellwig <hch@lst.de>,
linux-mm@kvack.org, nouveau@lists.freedesktop.org,
linux-kselftest@vger.kernel.org, linux-kernel@vger.kernel.org,
Jerome Glisse <jglisse@redhat.com>,
John Hubbard <jhubbard@nvidia.com>,
Alistair Popple <apopple@nvidia.com>,
Jason Gunthorpe <jgg@nvidia.com>,
Bharata B Rao <bharata@linux.ibm.com>, Zi Yan <ziy@nvidia.com>,
"Kirill A . Shutemov" <kirill.shutemov@linux.intel.com>,
Yang Shi <yang.shi@linux.alibaba.com>,
Ben Skeggs <bskeggs@redhat.com>, Shuah Khan <shuah@kernel.org>,
Andrew Morton <akpm@linux-foundation.org>,
Logan Gunthorpe <logang@deltatee.com>,
Dan Williams <dan.j.williams@intel.com>,
linux-nvdimm@lists.01.org, linux-fsdevel@vger.kernel.org
Subject: Re: [PATCH v3 3/6] mm: support THP migration to device private memory
Date: Wed, 2 Dec 2020 11:14:26 +0100 [thread overview]
Message-ID: <20201202101426.GC7597@lst.de> (raw)
In-Reply-To: <bbf1f0df-85f3-5887-050e-beb2aad750f2@nvidia.com>
[adding a few of the usual suspects]
On Wed, Nov 11, 2020 at 03:38:42PM -0800, Ralph Campbell wrote:
> There are 4 types of ZONE_DEVICE struct pages:
> MEMORY_DEVICE_PRIVATE, MEMORY_DEVICE_FS_DAX, MEMORY_DEVICE_GENERIC, and
> MEMORY_DEVICE_PCI_P2PDMA.
>
> Currently, memremap_pages() allocates struct pages for a physical address range
> with a page_ref_count(page) of one and increments the pgmap->ref per CPU
> reference count by the number of pages created since each ZONE_DEVICE struct
> page has a pointer to the pgmap.
>
> The struct pages are not freed until memunmap_pages() is called which
> calls put_page() which calls put_dev_pagemap() which releases a reference to
> pgmap->ref. memunmap_pages() blocks waiting for pgmap->ref reference count
> to be zero. As far as I can tell, the put_page() in memunmap_pages() has to
> be the *last* put_page() (see MEMORY_DEVICE_PCI_P2PDMA).
> My RFC [1] breaks this put_page() -> put_dev_pagemap() connection so that
> the struct page reference count can go to zero and back to non-zero without
> changing the pgmap->ref reference count.
>
> Q1: Is that safe? Is there some code that depends on put_page() dropping
> the pgmap->ref reference count as part of memunmap_pages()?
> My testing of [1] seems OK but I'm sure there are lots of cases I didn't test.
It should be safe, but the audit you've done is important to make sure
we do not miss anything important.
> MEMORY_DEVICE_PCI_P2PDMA:
> Struct pages are created in pci_p2pdma_add_resource() and represent device
> memory accessible by PCIe bar address space. Memory is allocated with
> pci_alloc_p2pmem() based on a byte length but the gen_pool_alloc_owner()
> call will allocate memory in a minimum of PAGE_SIZE units.
> Reference counting is +1 per *allocation* on the pgmap->ref reference count.
> Note that this is not +1 per page which is what put_page() expects. So
> currently, a get_page()/put_page() works OK because the page reference count
> only goes 1->2 and 2->1. If it went to zero, the pgmap->ref reference count
> would be incorrect if the allocation size was greater than one page.
>
> I see pci_alloc_p2pmem() is called by nvme_alloc_sq_cmds() and
> pci_p2pmem_alloc_sgl() to create a command queue and a struct scatterlist *.
> Looks like sg_page(sg) returns the ZONE_DEVICE struct page of the scatterlist.
> There are a huge number of places sg_page() is called so it is hard to tell
> whether or not get_page()/put_page() is ever called on MEMORY_DEVICE_PCI_P2PDMA
> pages.
Nothing should call get_page/put_page on them, as they are not treated
as refcountable memory. More importantly nothing is allowed to keep
a reference longer than the time of the I/O.
> pci_p2pmem_virt_to_bus() will return the physical address and I guess
> pfn_to_page(physaddr >> PAGE_SHIFT) could return the struct page.
>
> Since there is a clear allocation/free, pci_alloc_p2pmem() can probably be
> modified to increment/decrement the MEMORY_DEVICE_PCI_P2PDMA struct page
> reference count. Or maybe just leave it at one like it is now.
And yes, doing that is probably a sensible safe guard.
> MEMORY_DEVICE_FS_DAX:
> Struct pages are created in pmem_attach_disk() and virtio_fs_setup_dax() with
> an initial reference count of one.
> The problem I see is that there are 3 states that are important:
> a) memory is free and not allocated to any file (page_ref_count() == 0).
> b) memory is allocated to a file and in the page cache (page_ref_count() == 1).
> c) some gup() or I/O has a reference even after calling unmap_mapping_pages()
> (page_ref_count() > 1). ext4_break_layouts() basically waits until the
> page_ref_count() == 1 with put_page() calling wake_up_var(&page->_refcount)
> to wake up ext4_break_layouts().
> The current code doesn't seem to distinguish (a) and (b). If we want to use
> the 0->1 reference count to signal (c), then the page cache would have hold
> entries with a page_ref_count() == 0 which doesn't match the general page cache
I think the sensible model here is to grab a reference when it is
added to the page cache. That is exactly how normal system memory pages
work.
next prev parent reply other threads:[~2020-12-02 10:14 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-11-06 0:51 [PATCH v3 0/6] mm/hmm/nouveau: add THP migration to migrate_vma_* Ralph Campbell
2020-11-06 0:51 ` [PATCH v3 1/6] mm/thp: add prep_transhuge_device_private_page() Ralph Campbell
2020-11-06 7:55 ` Christoph Hellwig
2020-11-06 20:56 ` Ralph Campbell
2020-11-06 12:14 ` Matthew Wilcox
2020-11-06 20:34 ` Ralph Campbell
2020-11-06 0:51 ` [PATCH v3 2/6] mm/migrate: move migrate_vma_collect_skip() Ralph Campbell
2020-11-06 7:56 ` Christoph Hellwig
2020-11-06 7:57 ` Christoph Hellwig
2020-11-06 0:51 ` [PATCH v3 3/6] mm: support THP migration to device private memory Ralph Campbell
2020-11-06 8:03 ` Christoph Hellwig
2020-11-06 21:26 ` Ralph Campbell
2020-11-09 9:14 ` Christoph Hellwig
2020-11-09 21:34 ` Ralph Campbell
2020-11-11 23:38 ` Ralph Campbell
2020-11-20 20:01 ` Jason Gunthorpe
2020-12-02 10:08 ` Christoph Hellwig
2020-12-05 8:22 ` Roger Pau Monné
2020-12-02 10:14 ` Christoph Hellwig [this message]
2020-12-02 18:01 ` Logan Gunthorpe
2020-11-06 0:51 ` [PATCH v3 4/6] mm/thp: add THP allocation helper Ralph Campbell
2020-11-06 8:01 ` Christoph Hellwig
2020-11-06 21:09 ` Ralph Campbell
2020-11-06 0:51 ` [PATCH v3 5/6] mm/hmm/test: add self tests for THP migration Ralph Campbell
2020-11-06 0:51 ` [PATCH v3 6/6] nouveau: support THP migration to private memory Ralph Campbell
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=20201202101426.GC7597@lst.de \
--to=hch@lst.de \
--cc=akpm@linux-foundation.org \
--cc=apopple@nvidia.com \
--cc=bharata@linux.ibm.com \
--cc=bskeggs@redhat.com \
--cc=dan.j.williams@intel.com \
--cc=jgg@nvidia.com \
--cc=jglisse@redhat.com \
--cc=jhubbard@nvidia.com \
--cc=kirill.shutemov@linux.intel.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-kselftest@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linux-nvdimm@lists.01.org \
--cc=logang@deltatee.com \
--cc=nouveau@lists.freedesktop.org \
--cc=rcampbell@nvidia.com \
--cc=shuah@kernel.org \
--cc=yang.shi@linux.alibaba.com \
--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;
as well as URLs for NNTP newsgroup(s).