From: Vlastimil Babka <vbabka@suse.cz>
To: John Hubbard <jhubbard@nvidia.com>, Michal Hocko <mhocko@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>,
Christoph Hellwig <hch@infradead.org>,
Ira Weiny <ira.weiny@intel.com>, Jan Kara <jack@suse.cz>,
Jason Gunthorpe <jgg@ziepe.ca>,
Jerome Glisse <jglisse@redhat.com>,
LKML <linux-kernel@vger.kernel.org>,
linux-mm@kvack.org, linux-fsdevel@vger.kernel.org,
Dan Williams <dan.j.williams@intel.com>,
Daniel Black <daniel@linux.ibm.com>,
Matthew Wilcox <willy@infradead.org>,
Mike Kravetz <mike.kravetz@oracle.com>
Subject: Re: [PATCH 1/3] mm/mlock.c: convert put_page() to put_user_page*()
Date: Fri, 9 Aug 2019 10:12:48 +0200 [thread overview]
Message-ID: <420a5039-a79c-3872-38ea-807cedca3b8a@suse.cz> (raw)
In-Reply-To: <d1ecb0d4-ea6a-637d-7029-687b950b783f@nvidia.com>
On 8/9/19 12:59 AM, John Hubbard wrote:
>>> That's true. However, I'm not sure munlocking is where the
>>> put_user_page() machinery is intended to be used anyway? These are
>>> short-term pins for struct page manipulation, not e.g. dirtying of page
>>> contents. Reading commit fc1d8e7cca2d I don't think this case falls
>>> within the reasoning there. Perhaps not all GUP users should be
>>> converted to the planned separate GUP tracking, and instead we should
>>> have a GUP/follow_page_mask() variant that keeps using get_page/put_page?
>>>
>>
>> Interesting. So far, the approach has been to get all the gup callers to
>> release via put_user_page(), but if we add in Jan's and Ira's vaddr_pin_pages()
>> wrapper, then maybe we could leave some sites unconverted.
>>
>> However, in order to do so, we would have to change things so that we have
>> one set of APIs (gup) that do *not* increment a pin count, and another set
>> (vaddr_pin_pages) that do.
>>
>> Is that where we want to go...?
>>
We already have a FOLL_LONGTERM flag, isn't that somehow related? And if
it's not exactly the same thing, perhaps a new gup flag to distinguish
which kind of pinning to use?
> Oh, and meanwhile, I'm leaning toward a cheap fix: just use gup_fast() instead
> of get_page(), and also fix the releasing code. So this incremental patch, on
> top of the existing one, should do it:
>
...
> @@ -411,7 +409,13 @@ static unsigned long __munlock_pagevec_fill(struct pagevec *pvec,
> if (PageTransCompound(page))
> break;
>
> - get_page(page);
> + /*
> + * Use get_user_pages_fast(), instead of get_page() so that the
> + * releasing code can unconditionally call put_user_page().
> + */
> + ret = get_user_pages_fast(start, 1, 0, &page);
Um the whole reason of __munlock_pagevec_fill() was to avoid the full
page walk cost, which made a 14% difference, see 7a8010cd3627 ("mm:
munlock: manual pte walk in fast path instead of follow_page_mask()")
Replacing simple get_page() with page walk to satisfy API requirements
seems rather suboptimal to me.
> + if (ret != 1)
> + break;
> /*
> * Increase the address that will be returned *before* the
> * eventual break due to pvec becoming full by adding the page
>
>
> thanks,
>
next prev parent reply other threads:[~2019-08-09 8:12 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-08-05 22:20 [PATCH 0/3] mm/: 3 more put_user_page() conversions john.hubbard
2019-08-05 22:20 ` [PATCH 1/3] mm/mlock.c: convert put_page() to put_user_page*() john.hubbard
2019-08-07 11:01 ` Michal Hocko
2019-08-07 23:32 ` John Hubbard
2019-08-08 6:21 ` Michal Hocko
2019-08-08 11:09 ` Vlastimil Babka
2019-08-08 19:20 ` John Hubbard
2019-08-08 22:59 ` John Hubbard
2019-08-08 23:41 ` Ira Weiny
2019-08-08 23:57 ` John Hubbard
2019-08-09 18:22 ` Weiny, Ira
2019-08-09 8:12 ` Vlastimil Babka [this message]
2019-08-09 8:23 ` Michal Hocko
2019-08-09 9:05 ` John Hubbard
2019-08-09 9:16 ` Michal Hocko
2019-08-09 13:58 ` Jan Kara
2019-08-09 17:52 ` Michal Hocko
2019-08-09 18:14 ` Weiny, Ira
2019-08-09 18:36 ` John Hubbard
2019-08-05 22:20 ` [PATCH 2/3] mm/mempolicy.c: " john.hubbard
2019-08-05 22:20 ` [PATCH 3/3] mm/ksm: " john.hubbard
2019-08-06 21:59 ` [PATCH 0/3] mm/: 3 more put_user_page() conversions Andrew Morton
2019-08-06 22:05 ` John Hubbard
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=420a5039-a79c-3872-38ea-807cedca3b8a@suse.cz \
--to=vbabka@suse.cz \
--cc=akpm@linux-foundation.org \
--cc=dan.j.williams@intel.com \
--cc=daniel@linux.ibm.com \
--cc=hch@infradead.org \
--cc=ira.weiny@intel.com \
--cc=jack@suse.cz \
--cc=jgg@ziepe.ca \
--cc=jglisse@redhat.com \
--cc=jhubbard@nvidia.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mhocko@kernel.org \
--cc=mike.kravetz@oracle.com \
--cc=willy@infradead.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;
as well as URLs for NNTP newsgroup(s).