From: John Hubbard <jhubbard@nvidia.com>
To: Joao Martins <joao.m.martins@oracle.com>, <linux-mm@kvack.org>
Cc: <linux-kernel@vger.kernel.org>, <linux-rdma@vger.kernel.org>,
Andrew Morton <akpm@linux-foundation.org>,
Jason Gunthorpe <jgg@ziepe.ca>,
Doug Ledford <dledford@redhat.com>,
Matthew Wilcox <willy@infradead.org>
Subject: Re: [PATCH 1/4] mm/gup: add compound page list iterator
Date: Wed, 3 Feb 2021 15:00:01 -0800 [thread overview]
Message-ID: <955dbe68-7302-a8bc-f0b5-e9032d7f190e@nvidia.com> (raw)
In-Reply-To: <20210203220025.8568-2-joao.m.martins@oracle.com>
On 2/3/21 2:00 PM, Joao Martins wrote:
> Add an helper that iterates over head pages in a list of pages. It
> essentially counts the tails until the next page to process has a
> different head that the current. This is going to be used by
> unpin_user_pages() family of functions, to batch the head page refcount
> updates once for all passed consecutive tail pages.
>
> Suggested-by: Jason Gunthorpe <jgg@nvidia.com>
> Signed-off-by: Joao Martins <joao.m.martins@oracle.com>
> ---
> mm/gup.c | 29 +++++++++++++++++++++++++++++
> 1 file changed, 29 insertions(+)
>
> diff --git a/mm/gup.c b/mm/gup.c
> index d68bcb482b11..4f88dcef39f2 100644
> --- a/mm/gup.c
> +++ b/mm/gup.c
> @@ -215,6 +215,35 @@ void unpin_user_page(struct page *page)
> }
> EXPORT_SYMBOL(unpin_user_page);
>
> +static inline unsigned int count_ntails(struct page **pages, unsigned long npages)
Silly naming nit: could we please name this function count_pagetails()? count_ntails
is a bit redundant, plus slightly less clear.
> +{
> + struct page *head = compound_head(pages[0]);
> + unsigned int ntails;
> +
> + for (ntails = 1; ntails < npages; ntails++) {
> + if (compound_head(pages[ntails]) != head)
> + break;
> + }
> +
> + return ntails;
> +}
> +
> +static inline void compound_next(unsigned long i, unsigned long npages,
> + struct page **list, struct page **head,
> + unsigned int *ntails)
> +{
> + if (i >= npages)
> + return;
> +
> + *ntails = count_ntails(list + i, npages - i);
> + *head = compound_head(list[i]);
> +}
> +
> +#define for_each_compound_head(i, list, npages, head, ntails) \
When using macros, which are dangerous in general, you have to worry about
things like name collisions. I really dislike that C has forced this unsafe
pattern upon us, but of course we are stuck with it, for iterator helpers.
Given that we're stuck, you should probably use names such as __i, __list, etc,
in the the above #define. Otherwise you could stomp on existing variables.
> + for (i = 0, compound_next(i, npages, list, &head, &ntails); \
> + i < npages; i += ntails, \
> + compound_next(i, npages, list, &head, &ntails))
> +
> /**
> * unpin_user_pages_dirty_lock() - release and optionally dirty gup-pinned pages
> * @pages: array of pages to be maybe marked dirty, and definitely released.
>
thanks,
--
John Hubbard
NVIDIA
next prev parent reply other threads:[~2021-02-03 23:00 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-02-03 22:00 [PATCH 0/4] mm/gup: page unpining improvements Joao Martins
2021-02-03 22:00 ` [PATCH 1/4] mm/gup: add compound page list iterator Joao Martins
2021-02-03 23:00 ` John Hubbard [this message]
2021-02-04 11:27 ` Joao Martins
2021-02-04 16:09 ` Joao Martins
2021-02-04 19:53 ` Jason Gunthorpe
2021-02-04 23:37 ` John Hubbard
2021-02-03 22:00 ` [PATCH 2/4] mm/gup: decrement head page once for group of subpages Joao Martins
2021-02-03 23:28 ` John Hubbard
2021-02-04 11:27 ` Joao Martins
2021-02-03 22:00 ` [PATCH 3/4] mm/gup: add a range variant of unpin_user_pages_dirty_lock() Joao Martins
2021-02-03 23:37 ` John Hubbard
2021-02-04 11:35 ` Joao Martins
2021-02-04 16:30 ` Joao Martins
2021-02-04 0:11 ` John Hubbard
2021-02-04 11:47 ` Joao Martins
2021-02-03 22:00 ` [PATCH 4/4] RDMA/umem: batch page unpin in __ib_mem_release() Joao Martins
2021-02-04 0:15 ` John Hubbard
2021-02-04 12:29 ` Joao Martins
2021-02-04 20:00 ` Jason Gunthorpe
2021-02-05 17:00 ` Joao Martins
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=955dbe68-7302-a8bc-f0b5-e9032d7f190e@nvidia.com \
--to=jhubbard@nvidia.com \
--cc=akpm@linux-foundation.org \
--cc=dledford@redhat.com \
--cc=jgg@ziepe.ca \
--cc=joao.m.martins@oracle.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linux-rdma@vger.kernel.org \
--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