linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Muchun Song <songmuchun@bytedance.com>
To: Michal Hocko <mhocko@suse.com>
Cc: "Jonathan Corbet" <corbet@lwn.net>,
	"Mike Kravetz" <mike.kravetz@oracle.com>,
	"Thomas Gleixner" <tglx@linutronix.de>,
	"Ingo Molnar" <mingo@redhat.com>,
	bp@alien8.de, x86@kernel.org, hpa@zytor.com,
	dave.hansen@linux.intel.com, luto@kernel.org,
	"Peter Zijlstra" <peterz@infradead.org>,
	"Alexander Viro" <viro@zeniv.linux.org.uk>,
	"Andrew Morton" <akpm@linux-foundation.org>,
	paulmck@kernel.org, mchehab+huawei@kernel.org,
	pawan.kumar.gupta@linux.intel.com,
	"Randy Dunlap" <rdunlap@infradead.org>,
	oneukum@suse.com, anshuman.khandual@arm.com, jroedel@suse.de,
	"Mina Almasry" <almasrymina@google.com>,
	"David Rientjes" <rientjes@google.com>,
	"Matthew Wilcox" <willy@infradead.org>,
	"Oscar Salvador" <osalvador@suse.de>,
	"Song Bao Hua (Barry Song)" <song.bao.hua@hisilicon.com>,
	"David Hildenbrand" <david@redhat.com>,
	"HORIGUCHI NAOYA(堀口 直也)" <naoya.horiguchi@nec.com>,
	"Joao Martins" <joao.m.martins@oracle.com>,
	"Xiongchun duan" <duanxiongchun@bytedance.com>,
	linux-doc@vger.kernel.org, LKML <linux-kernel@vger.kernel.org>,
	"Linux Memory Management List" <linux-mm@kvack.org>,
	linux-fsdevel <linux-fsdevel@vger.kernel.org>,
	"Miaohe Lin" <linmiaohe@huawei.com>,
	"Chen Huang" <chenhuang5@huawei.com>,
	"Bodeddula Balasubramaniam" <bodeddub@amazon.com>
Subject: Re: [External] Re: [PATCH v18 9/9] mm: hugetlb: optimize the code with the help of the compiler
Date: Thu, 11 Mar 2021 15:33:20 +0800	[thread overview]
Message-ID: <CAMZfGtV1Fp1RiQ64c9RrMmZ+=EwjGRHjwL8Wx3Q0YRWbbKF6xg@mail.gmail.com> (raw)
In-Reply-To: <YEjoozshsvKeMAAu@dhcp22.suse.cz>

On Wed, Mar 10, 2021 at 11:41 PM Michal Hocko <mhocko@suse.com> wrote:
>
> On Mon 08-03-21 18:28:07, Muchun Song wrote:
> > When the "struct page size" crosses page boundaries we cannot
> > make use of this feature. Let free_vmemmap_pages_per_hpage()
> > return zero if that is the case, most of the functions can be
> > optimized away.
>
> I am confused. Don't you check for this in early_hugetlb_free_vmemmap_param already?

Right.

> Why do we need any runtime checks?

If the size of the struct page is not power of 2, compiler can think
is_hugetlb_free_vmemmap_enabled() always return false. So
the code snippet of this user can be optimized away.

E.g.

if (is_hugetlb_free_vmemmap_enabled())
        /* do something */

The compiler can drop "/* do something */" directly, because
it knows is_hugetlb_free_vmemmap_enabled() always returns
false.

Thanks.

>
> > Signed-off-by: Muchun Song <songmuchun@bytedance.com>
> > Reviewed-by: Miaohe Lin <linmiaohe@huawei.com>
> > Reviewed-by: Oscar Salvador <osalvador@suse.de>
> > Tested-by: Chen Huang <chenhuang5@huawei.com>
> > Tested-by: Bodeddula Balasubramaniam <bodeddub@amazon.com>
> > ---
> >  include/linux/hugetlb.h | 3 ++-
> >  mm/hugetlb_vmemmap.c    | 7 +++++++
> >  mm/hugetlb_vmemmap.h    | 6 ++++++
> >  3 files changed, 15 insertions(+), 1 deletion(-)
> >
> > diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h
> > index c70421e26189..333dd0479fc2 100644
> > --- a/include/linux/hugetlb.h
> > +++ b/include/linux/hugetlb.h
> > @@ -880,7 +880,8 @@ extern bool hugetlb_free_vmemmap_enabled;
> >
> >  static inline bool is_hugetlb_free_vmemmap_enabled(void)
> >  {
> > -     return hugetlb_free_vmemmap_enabled;
> > +     return hugetlb_free_vmemmap_enabled &&
> > +            is_power_of_2(sizeof(struct page));
> >  }
> >  #else
> >  static inline bool is_hugetlb_free_vmemmap_enabled(void)
> > diff --git a/mm/hugetlb_vmemmap.c b/mm/hugetlb_vmemmap.c
> > index 33e42678abe3..1ba1ef45c48c 100644
> > --- a/mm/hugetlb_vmemmap.c
> > +++ b/mm/hugetlb_vmemmap.c
> > @@ -265,6 +265,13 @@ void __init hugetlb_vmemmap_init(struct hstate *h)
> >       BUILD_BUG_ON(__NR_USED_SUBPAGE >=
> >                    RESERVE_VMEMMAP_SIZE / sizeof(struct page));
> >
> > +     /*
> > +      * The compiler can help us to optimize this function to null
> > +      * when the size of the struct page is not power of 2.
> > +      */
> > +     if (!is_power_of_2(sizeof(struct page)))
> > +             return;
> > +
> >       if (!hugetlb_free_vmemmap_enabled)
> >               return;
> >
> > diff --git a/mm/hugetlb_vmemmap.h b/mm/hugetlb_vmemmap.h
> > index cb2bef8f9e73..29aaaf7b741e 100644
> > --- a/mm/hugetlb_vmemmap.h
> > +++ b/mm/hugetlb_vmemmap.h
> > @@ -21,6 +21,12 @@ void hugetlb_vmemmap_init(struct hstate *h);
> >   */
> >  static inline unsigned int free_vmemmap_pages_per_hpage(struct hstate *h)
> >  {
> > +     /*
> > +      * This check aims to let the compiler help us optimize the code as
> > +      * much as possible.
> > +      */
> > +     if (!is_power_of_2(sizeof(struct page)))
> > +             return 0;
> >       return h->nr_free_vmemmap_pages;
> >  }
> >  #else
> > --
> > 2.11.0
> >
>
> --
> Michal Hocko
> SUSE Labs


  reply	other threads:[~2021-03-11  7:34 UTC|newest]

Thread overview: 52+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-03-08 10:27 [PATCH v18 0/9] Free some vmemmap pages of HugeTLB page Muchun Song
2021-03-08 10:27 ` [PATCH v18 1/9] mm: memory_hotplug: factor out bootmem core functions to bootmem_info.c Muchun Song
2021-03-10 14:14   ` Michal Hocko
2021-03-11  2:58     ` [External] " Muchun Song
2021-03-11  8:45       ` Muchun Song
2021-03-11  8:53         ` Michal Hocko
2021-03-11  9:05           ` Muchun Song
2021-03-08 10:28 ` [PATCH v18 2/9] mm: hugetlb: introduce a new config HUGETLB_PAGE_FREE_VMEMMAP Muchun Song
2021-03-08 10:28 ` [PATCH v18 3/9] mm: hugetlb: free the vmemmap pages associated with each HugeTLB page Muchun Song
2021-03-10 14:32   ` Michal Hocko
2021-03-11  3:35     ` [External] " Muchun Song
2021-03-08 10:28 ` [PATCH v18 4/9] mm: hugetlb: alloc " Muchun Song
2021-03-10 14:21   ` Oscar Salvador
2021-03-11  4:13     ` [External] " Muchun Song
2021-03-10 15:19   ` Michal Hocko
2021-03-10 18:56     ` Mike Kravetz
2021-03-10 21:11       ` Michal Hocko
2021-03-10 21:49         ` Paul E. McKenney
2021-03-10 22:10           ` Mike Kravetz
2021-03-10 23:28             ` Paul E. McKenney
2021-03-11  8:40               ` Michal Hocko
2021-03-11 12:17                 ` Michal Hocko
2021-03-11 17:59                   ` Mike Kravetz
2021-03-11 22:53                     ` Mike Kravetz
2021-03-12  8:15                       ` Michal Hocko
2021-03-12 17:50                         ` Mike Kravetz
2021-03-11  4:26     ` [External] " Muchun Song
2021-03-11  8:46       ` Michal Hocko
2021-03-11  8:49         ` Muchun Song
2021-03-08 10:28 ` [PATCH v18 5/9] mm: hugetlb: set the PageHWPoison to the raw error page Muchun Song
2021-03-10 15:27   ` Michal Hocko
2021-03-11  6:34     ` [External] " Muchun Song
2021-03-11  8:50       ` Michal Hocko
2021-03-11  9:13         ` Muchun Song
2021-03-08 10:28 ` [PATCH v18 6/9] mm: hugetlb: add a kernel parameter hugetlb_free_vmemmap Muchun Song
2021-03-10 15:37   ` Michal Hocko
2021-03-10 17:15     ` Randy Dunlap
2021-03-11  6:36       ` [External] " Muchun Song
2021-03-11  6:36     ` Muchun Song
2021-03-08 10:28 ` [PATCH v18 7/9] mm: hugetlb: introduce nr_free_vmemmap_pages in the struct hstate Muchun Song
2021-03-08 10:28 ` [PATCH v18 8/9] mm: hugetlb: gather discrete indexes of tail page Muchun Song
2021-03-10 15:39   ` Michal Hocko
2021-03-08 10:28 ` [PATCH v18 9/9] mm: hugetlb: optimize the code with the help of the compiler Muchun Song
2021-03-10 15:41   ` Michal Hocko
2021-03-11  7:33     ` Muchun Song [this message]
2021-03-11  8:55       ` [External] " Michal Hocko
2021-03-11  9:08         ` Muchun Song
2021-03-11  9:39           ` Michal Hocko
2021-03-11 10:00             ` Muchun Song
2021-03-11 12:16               ` Michal Hocko
2021-03-11 13:00                 ` Muchun Song
2021-03-11 13:45                 ` Oscar Salvador

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='CAMZfGtV1Fp1RiQ64c9RrMmZ+=EwjGRHjwL8Wx3Q0YRWbbKF6xg@mail.gmail.com' \
    --to=songmuchun@bytedance.com \
    --cc=akpm@linux-foundation.org \
    --cc=almasrymina@google.com \
    --cc=anshuman.khandual@arm.com \
    --cc=bodeddub@amazon.com \
    --cc=bp@alien8.de \
    --cc=chenhuang5@huawei.com \
    --cc=corbet@lwn.net \
    --cc=dave.hansen@linux.intel.com \
    --cc=david@redhat.com \
    --cc=duanxiongchun@bytedance.com \
    --cc=hpa@zytor.com \
    --cc=joao.m.martins@oracle.com \
    --cc=jroedel@suse.de \
    --cc=linmiaohe@huawei.com \
    --cc=linux-doc@vger.kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=luto@kernel.org \
    --cc=mchehab+huawei@kernel.org \
    --cc=mhocko@suse.com \
    --cc=mike.kravetz@oracle.com \
    --cc=mingo@redhat.com \
    --cc=naoya.horiguchi@nec.com \
    --cc=oneukum@suse.com \
    --cc=osalvador@suse.de \
    --cc=paulmck@kernel.org \
    --cc=pawan.kumar.gupta@linux.intel.com \
    --cc=peterz@infradead.org \
    --cc=rdunlap@infradead.org \
    --cc=rientjes@google.com \
    --cc=song.bao.hua@hisilicon.com \
    --cc=tglx@linutronix.de \
    --cc=viro@zeniv.linux.org.uk \
    --cc=willy@infradead.org \
    --cc=x86@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;
as well as URLs for NNTP newsgroup(s).