Received-SPF: pass (google.com: domain of linux-kernel+bounces-96165-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.199.223 as permitted sender) client-ip=147.75.199.223;
Date: Thu, 7 Mar 2024 12:11:13 -0800
From: Andrew Morton <akpm@linux-foundation.org>
To: rulinhuang <rulin.huang@intel.com>
Cc: urezki@gmail.com, bhe@redhat.com, colin.king@intel.com,
 hch@infradead.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org,
 lstoakes@gmail.com, tianyou.li@intel.com, tim.c.chen@intel.com,
 wangyang.guo@intel.com, zhiguo.zhou@intel.com
Subject: Re: [PATCH v8] mm/vmalloc: Eliminated the lock contention from
 twice to once
Message-Id: <20240307121113.16cc480515195f1fe945bbac@linux-foundation.org>
In-Reply-To: <20240307021440.64967-1-rulin.huang@intel.com>
References: <20240307021440.64967-1-rulin.huang@intel.com>
Precedence: bulk
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit

On Wed,  6 Mar 2024 21:14:40 -0500 rulinhuang <rulin.huang@intel.com> wrote:

> When allocating a new memory area where the mapping address range is
> known, it is observed that the vmap_node->busy.lock is acquired twice.
> 
> The first acquisition occurs in the alloc_vmap_area() function when
> inserting the vm area into the vm mapping red-black tree. The second
> acquisition occurs in the setup_vmalloc_vm() function when updating the
> properties of the vm, such as flags and address, etc.
> 
> Combine these two operations together in alloc_vmap_area(), which
> improves scalability when the vmap_node->busy.lock is contended.
> By doing so, the need to acquire the lock twice can also be eliminated
> to once.
> 
> With the above change, tested on intel sapphire rapids
> platform(224 vcpu), a 4% performance improvement is
> gained on stress-ng/pthread(https://github.com/ColinIanKing/stress-ng),
> which is the stress test of thread creations.
> 

Thanks.  We're late in -rc7 so I'll queue this up for merging in the
next merge window.