Received-SPF: pass (google.com: domain of linux-kernel+bounces-120589-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45e3:2400::1 as permitted sender) client-ip=2604:1380:45e3:2400::1;
From: Peng Zhang <zhangpeng362@huawei.com>
To: <linux-mm@kvack.org>, <linux-kernel@vger.kernel.org>
CC: <akpm@linux-foundation.org>, <peterx@redhat.com>, <david@redhat.com>,
	<axelrasmussen@google.com>, <surenb@google.com>, <lokeshgidra@google.com>,
	<Liam.Howlett@oracle.com>, <wangkefeng.wang@huawei.com>,
	<sunnanyong@huawei.com>
Subject: [PATCH] userfaultfd: early return in dup_userfaultfd()
Date: Wed, 27 Mar 2024 17:08:35 +0800
Message-ID: <20240327090835.3232629-1-zhangpeng362@huawei.com>
Precedence: bulk
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Content-Type: text/plain

From: ZhangPeng <zhangpeng362@huawei.com>

When vma->vm_userfaultfd_ctx.ctx is NULL, vma->vm_flags should have
cleared __VM_UFFD_FLAGS. Therefore, there is no need to down_write or
clear the flag, which will affect fork performance. Fix this by
returning early if octx is NULL in dup_userfaultfd().

By applying this patch we can get a 1.3% performance improvement for
lmbench fork_prot. Results are as follows:
                   base      early return
Process fork+exit: 419.1106  413.4804

Signed-off-by: ZhangPeng <zhangpeng362@huawei.com>
---
 fs/userfaultfd.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/fs/userfaultfd.c b/fs/userfaultfd.c
index 60dcfafdc11a..3e6ddda6f159 100644
--- a/fs/userfaultfd.c
+++ b/fs/userfaultfd.c
@@ -657,7 +657,10 @@ int dup_userfaultfd(struct vm_area_struct *vma, struct list_head *fcs)
 	struct userfaultfd_fork_ctx *fctx;
 
 	octx = vma->vm_userfaultfd_ctx.ctx;
-	if (!octx || !(octx->features & UFFD_FEATURE_EVENT_FORK)) {
+	if (!octx)
+		return 0;
+
+	if (!(octx->features & UFFD_FEATURE_EVENT_FORK)) {
 		vma_start_write(vma);
 		vma->vm_userfaultfd_ctx = NULL_VM_UFFD_CTX;
 		userfaultfd_set_vm_flags(vma, vma->vm_flags & ~__VM_UFFD_FLAGS);
-- 
2.25.1