From: Baolin Wang <baolin.wang@linux.alibaba.com>
To: David Hildenbrand <david@redhat.com>,
Ryan Roberts <ryan.roberts@arm.com>,
akpm@linux-foundation.org, hughd@google.com
Cc: willy@infradead.org, ioworker0@gmail.com,
wangkefeng.wang@huawei.com, ying.huang@intel.com,
21cnbao@gmail.com, shy828301@gmail.com, ziy@nvidia.com,
linux-mm@kvack.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH 2/8] mm: memory: extend finish_fault() to support large folio
Date: Wed, 8 May 2024 17:06:28 +0800 [thread overview]
Message-ID: <043a4f2d-e08d-4cea-a73d-819586509b12@linux.alibaba.com> (raw)
In-Reply-To: <5564e708-4a9f-4010-806e-4c5a7a5d2ebe@redhat.com>
On 2024/5/8 15:15, David Hildenbrand wrote:
> On 08.05.24 05:44, Baolin Wang wrote:
>>
>>
>> On 2024/5/7 18:37, Ryan Roberts wrote:
>>> On 06/05/2024 09:46, Baolin Wang wrote:
>>>> Add large folio mapping establishment support for finish_fault() as
>>>> a preparation,
>>>> to support multi-size THP allocation of anonymous shmem pages in the
>>>> following
>>>> patches.
>>>>
>>>> Signed-off-by: Baolin Wang <baolin.wang@linux.alibaba.com>
>>>> ---
>>>> mm/memory.c | 43 +++++++++++++++++++++++++++++++++----------
>>>> 1 file changed, 33 insertions(+), 10 deletions(-)
>>>>
>>>> diff --git a/mm/memory.c b/mm/memory.c
>>>> index eea6e4984eae..936377220b77 100644
>>>> --- a/mm/memory.c
>>>> +++ b/mm/memory.c
>>>> @@ -4747,9 +4747,12 @@ vm_fault_t finish_fault(struct vm_fault *vmf)
>>>> {
>>>> struct vm_area_struct *vma = vmf->vma;
>>>> struct page *page;
>>>> + struct folio *folio;
>>>> vm_fault_t ret;
>>>> bool is_cow = (vmf->flags & FAULT_FLAG_WRITE) &&
>>>> !(vma->vm_flags & VM_SHARED);
>>>> + int type, nr_pages, i;
>>>> + unsigned long addr = vmf->address;
>>>> /* Did we COW the page? */
>>>> if (is_cow)
>>>> @@ -4780,24 +4783,44 @@ vm_fault_t finish_fault(struct vm_fault *vmf)
>>>> return VM_FAULT_OOM;
>>>> }
>>>> + folio = page_folio(page);
>>>> + nr_pages = folio_nr_pages(folio);
>>>> +
>>>> + if (unlikely(userfaultfd_armed(vma))) {
>>>> + nr_pages = 1;
>>>> + } else if (nr_pages > 1) {
>>>> + unsigned long start = ALIGN_DOWN(vmf->address, nr_pages *
>>>> PAGE_SIZE);
>>>> + unsigned long end = start + nr_pages * PAGE_SIZE;
>>>> +
>>>> + /* In case the folio size in page cache beyond the VMA
>>>> limits. */
>>>> + addr = max(start, vma->vm_start);
>>>> + nr_pages = (min(end, vma->vm_end) - addr) >> PAGE_SHIFT;
>>>> +
>>>> + page = folio_page(folio, (addr - start) >> PAGE_SHIFT);
>>>
>>> I still don't really follow the logic in this else if block. Isn't it
>>> possible
>>> that finish_fault() gets called with a page from a folio that isn't
>>> aligned with
>>> vmf->address?
>>>
>>> For example, let's say we have a file who's size is 64K and which is
>>> cached in a
>>> single large folio in the page cache. But the file is mapped into a
>>> process at
>>> VA 16K to 80K. Let's say we fault on the first page (VA=16K). You
>>> will calculate
>>
>> For shmem, this doesn't happen because the VA is aligned with the
>> hugepage size in the shmem_get_unmapped_area() function. See patch 7.
>
> Does that cover mremap() and MAP_FIXED as well.
Good point. Thanks for pointing this out.
> We should try doing this as cleanly as possible, to prepare for the
> future / corner cases.
Sure. Let me re-think about the algorithm.
next prev parent reply other threads:[~2024-05-08 9:06 UTC|newest]
Thread overview: 52+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-05-06 8:46 [PATCH 0/8] add mTHP support for anonymous shmem Baolin Wang
2024-05-06 8:46 ` [PATCH 1/8] mm: move highest_order() and next_order() out of the THP config Baolin Wang
2024-05-07 10:21 ` Ryan Roberts
2024-05-08 2:13 ` Baolin Wang
2024-05-08 9:06 ` Ryan Roberts
2024-05-08 9:40 ` Baolin Wang
2024-05-06 8:46 ` [PATCH 2/8] mm: memory: extend finish_fault() to support large folio Baolin Wang
2024-05-07 10:37 ` Ryan Roberts
2024-05-08 3:44 ` Baolin Wang
2024-05-08 7:15 ` David Hildenbrand
2024-05-08 9:06 ` Baolin Wang [this message]
2024-05-08 8:53 ` Ryan Roberts
2024-05-08 9:31 ` Baolin Wang
2024-05-08 10:47 ` Ryan Roberts
2024-05-09 1:10 ` Baolin Wang
2024-05-06 8:46 ` [PATCH 3/8] mm: shmem: add an 'order' parameter for shmem_alloc_hugefolio() Baolin Wang
2024-05-06 8:46 ` [PATCH 4/8] mm: shmem: add THP validation for PMD-mapped THP related statistics Baolin Wang
2024-05-06 8:46 ` [PATCH 5/8] mm: shmem: add multi-size THP sysfs interface for anonymous shmem Baolin Wang
2024-05-07 10:52 ` Ryan Roberts
2024-05-08 4:45 ` Baolin Wang
2024-05-08 7:08 ` David Hildenbrand
2024-05-08 7:12 ` David Hildenbrand
2024-05-08 9:02 ` Ryan Roberts
2024-05-08 9:56 ` Baolin Wang
2024-05-08 10:48 ` Ryan Roberts
2024-05-08 12:02 ` David Hildenbrand
2024-05-08 12:10 ` David Hildenbrand
2024-05-08 12:43 ` Ryan Roberts
2024-05-08 12:44 ` Ryan Roberts
2024-05-08 12:45 ` David Hildenbrand
2024-05-08 12:54 ` Ryan Roberts
2024-05-08 13:07 ` David Hildenbrand
2024-05-08 13:44 ` Ryan Roberts
2024-05-06 8:46 ` [PATCH 6/8] mm: shmem: add mTHP support " Baolin Wang
2024-05-07 10:46 ` kernel test robot
2024-05-08 6:03 ` Baolin Wang
2024-05-06 8:46 ` [PATCH 7/8] mm: shmem: add mTHP size alignment in shmem_get_unmapped_area Baolin Wang
2024-05-06 8:46 ` [PATCH 8/8] mm: shmem: add mTHP counters for anonymous shmem Baolin Wang
2024-05-06 10:54 ` [PATCH 0/8] add mTHP support " Lance Yang
2024-05-07 1:47 ` Baolin Wang
2024-05-07 6:50 ` Lance Yang
2024-05-07 10:20 ` Ryan Roberts
2024-05-08 5:45 ` Baolin Wang
[not found] ` <CGME20240508113934eucas1p13a3972f3f9955365f40155e084a7c7d5@eucas1p1.samsung.com>
2024-05-08 11:39 ` Daniel Gomez
2024-05-08 11:58 ` David Hildenbrand
2024-05-08 14:28 ` Daniel Gomez
2024-05-08 17:03 ` David Hildenbrand
2024-05-09 19:18 ` Daniel Gomez
2024-05-09 3:08 ` Baolin Wang
2024-05-08 19:23 ` Luis Chamberlain
2024-05-09 17:48 ` David Hildenbrand
2024-05-10 18:53 ` Luis Chamberlain
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=043a4f2d-e08d-4cea-a73d-819586509b12@linux.alibaba.com \
--to=baolin.wang@linux.alibaba.com \
--cc=21cnbao@gmail.com \
--cc=akpm@linux-foundation.org \
--cc=david@redhat.com \
--cc=hughd@google.com \
--cc=ioworker0@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=ryan.roberts@arm.com \
--cc=shy828301@gmail.com \
--cc=wangkefeng.wang@huawei.com \
--cc=willy@infradead.org \
--cc=ying.huang@intel.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 read-only IMAP folder(s) and NNTP newsgroup(s).