Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
From:   Peter Xu <peterx@redhat.com>
To:     linux-mm@kvack.org, linux-kernel@vger.kernel.org
Cc:     Jan Kara <jack@suse.cz>, John Hubbard <jhubbard@nvidia.com>,
        peterx@redhat.com, Linus Torvalds <torvalds@linux-foundation.org>,
        Michal Hocko <mhocko@suse.com>,
        Kirill Tkhai <ktkhai@virtuozzo.com>,
        Kirill Shutemov <kirill@shutemov.name>,
        Oleg Nesterov <oleg@redhat.com>,
        Andrew Morton <akpm@linux-foundation.org>,
        Jann Horn <jannh@google.com>,
        Andrea Arcangeli <aarcange@redhat.com>,
        Jason Gunthorpe <jgg@nvidia.com>,
        Matthew Wilcox <willy@infradead.org>,
        Hugh Dickins <hughd@google.com>
Subject: [PATCH v2 2/3] mm: gup: allow FOLL_PIN to scale in SMP
Date:   Fri,  7 May 2021 11:05:52 -0400
Message-Id: <20210507150553.208763-3-peterx@redhat.com>
In-Reply-To: <20210507150553.208763-1-peterx@redhat.com>
References: <20210507150553.208763-1-peterx@redhat.com>
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Precedence: bulk

From: Andrea Arcangeli <aarcange@redhat.com>

has_pinned cannot be written by each pin-fast or it won't scale in
SMP. This isn't "false sharing" strictly speaking (it's more like
"true non-sharing"), but it creates the same SMP scalability
bottleneck of "false sharing".

To verify the improvement, below test is done on 40 cpus host with Intel(R)
Xeon(R) CPU E5-2630 v4 @ 2.20GHz (must be with CONFIG_GUP_TEST=y):

  $ sudo chrt -f 1 ./gup_test -a  -m 512 -j 40

Where we can get (average value for 40 threads):

  Old kernel: 477729.97 (+- 3.79%)
  New kernel:  89144.65 (+-11.76%)

On a similar condition with 256 cpus, this commits increases the SMP
scalability of pin_user_pages_fast() executed by different threads of the same
process by more than 4000%.

Signed-off-by: Andrea Arcangeli <aarcange@redhat.com>
[peterx: rewrite commit message, add parentheses against "(A & B)"]
Reviewed-by: John Hubbard <jhubbard@nvidia.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
---
 mm/gup.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/mm/gup.c b/mm/gup.c
index 63a079e361a3d..9933bc5c2eff2 100644
--- a/mm/gup.c
+++ b/mm/gup.c
@@ -1292,7 +1292,7 @@ static __always_inline long __get_user_pages_locked(struct mm_struct *mm,
 		BUG_ON(*locked != 1);
 	}
 
-	if (flags & FOLL_PIN)
+	if ((flags & FOLL_PIN) && !atomic_read(&mm->has_pinned))
 		atomic_set(&mm->has_pinned, 1);
 
 	/*
@@ -2617,7 +2617,7 @@ static int internal_get_user_pages_fast(unsigned long start,
 				       FOLL_FAST_ONLY)))
 		return -EINVAL;
 
-	if (gup_flags & FOLL_PIN)
+	if ((gup_flags & FOLL_PIN) && !atomic_read(&current->mm->has_pinned))
 		atomic_set(&current->mm->has_pinned, 1);
 
 	if (!(gup_flags & FOLL_FAST_ONLY))
-- 
2.31.1