Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Tue, 18 Feb 2020 12:53:12 -0800
From:   Roman Gushchin <guro@fb.com>
To:     Wen Yang <wenyang@linux.alibaba.com>
CC:     Andrew Morton <akpm@linux-foundation.org>,
        Christoph Lameter <cl@linux.com>,
        Pekka Enberg <penberg@kernel.org>,
        David Rientjes <rientjes@google.com>,
        Joonsoo Kim <iamjoonsoo.kim@lge.com>,
        Xunlei Pang <xlpang@linux.alibaba.com>, <linux-mm@kvack.org>,
        <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH] mm/slub: Detach node lock from counting free objects
Message-ID: <20200218205312.GA3156@carbon>
References: <20200201031502.92218-1-wenyang@linux.alibaba.com>
 <20200212145247.bf89431272038de53dd9d975@linux-foundation.org>
 <b42f7daa-4aea-1cf8-5bbb-2cd5d48b4e9a@linux.alibaba.com>
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <b42f7daa-4aea-1cf8-5bbb-2cd5d48b4e9a@linux.alibaba.com>
MIME-Version: 1.0
Received-SPF: None (protection.outlook.com: fb.com does not designate
 permitted sender hosts)
X-MS-Exchange-CrossTenant-Network-Message-Id: 94ae379e-6254-4b21-2ce1-08d7b4b49463
X-MS-Exchange-CrossTenant-OriginalArrivalTime: 18 Feb 2020 20:53:17.6043
 (UTC)
X-MS-Exchange-CrossTenant-FromEntityHeader: Hosted
X-MS-Exchange-CrossTenant-Id: 8ae927fe-1255-47a7-a2af-5f3a069daaa2
X-MS-Exchange-CrossTenant-MailboxType: HOSTED
X-MS-Exchange-CrossTenant-UserPrincipalName: q3x1E+oxa5QGnRA4QHu6DYz7O0+fM9oZLLLOSJQIieMpes7oVets7rAavD8UiRm+
X-MS-Exchange-Transport-CrossTenantHeadersStamped: BYAPR15MB3189
X-OriginatorOrg: fb.com
X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10434:6.0.138,18.0.572
 definitions=2020-02-18_06:2020-02-18,2020-02-18 signatures=0
X-Proofpoint-Spam-Details: rule=fb_default_notspam policy=fb_default score=0 bulkscore=0
 mlxlogscore=999 lowpriorityscore=0 priorityscore=1501 malwarescore=0
 adultscore=0 clxscore=1011 spamscore=0 impostorscore=0 suspectscore=5
 mlxscore=0 phishscore=0 classifier=spam adjust=0 reason=mlx scancount=1
 engine=8.12.0-2001150001 definitions=main-2002180137
X-FB-Internal: deliver
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On Sun, Feb 16, 2020 at 12:15:54PM +0800, Wen Yang wrote:
> 
> 
> On 2020/2/13 6:52 上午, Andrew Morton wrote:
> > On Sat,  1 Feb 2020 11:15:02 +0800 Wen Yang <wenyang@linux.alibaba.com> wrote:
> > 
> > > The lock, protecting the node partial list, is taken when couting the free
> > > objects resident in that list. It introduces locking contention when the
> > > page(s) is moved between CPU and node partial lists in allocation path
> > > on another CPU. So reading "/proc/slabinfo" can possibily block the slab
> > > allocation on another CPU for a while, 200ms in extreme cases. If the
> > > slab object is to carry network packet, targeting the far-end disk array,
> > > it causes block IO jitter issue.
> > > 
> > > This fixes the block IO jitter issue by caching the total inuse objects in
> > > the node in advance. The value is retrieved without taking the node partial
> > > list lock on reading "/proc/slabinfo".
> > > 
> > > ...
> > > 
> > > @@ -1768,7 +1774,9 @@ static void free_slab(struct kmem_cache *s, struct page *page)
> > >   static void discard_slab(struct kmem_cache *s, struct page *page)
> > >   {
> > > -	dec_slabs_node(s, page_to_nid(page), page->objects);
> > > +	int inuse = page->objects;
> > > +
> > > +	dec_slabs_node(s, page_to_nid(page), page->objects, inuse);
> > 
> > Is this right?  dec_slabs_node(..., page->objects, page->objects)?
> > 
> > If no, we could simply pass the page* to inc_slabs_node/dec_slabs_node
> > and save a function argument.
> > 
> > If yes then why?
> > 
> 
> Thanks for your comments.
> We are happy to improve this patch based on your suggestions.
> 
> 
> When the user reads /proc/slabinfo, in order to obtain the active_objs
> information, the kernel traverses all slabs and executes the following code
> snippet:
> static unsigned long count_partial(struct kmem_cache_node *n,
>                                         int (*get_count)(struct page *))
> {
>         unsigned long flags;
>         unsigned long x = 0;
>         struct page *page;
> 
>         spin_lock_irqsave(&n->list_lock, flags);
>         list_for_each_entry(page, &n->partial, slab_list)
>                 x += get_count(page);
>         spin_unlock_irqrestore(&n->list_lock, flags);
>         return x;
> }
> 
> It may cause performance issues.
> 
> Christoph suggested "you could cache the value in the userspace application?
> Why is this value read continually?", But reading the /proc/slabinfo is
> initiated by the user program. As a cloud provider, we cannot control user
> behavior. If a user program inadvertently executes cat /proc/slabinfo, it
> may affect other user programs.
> 
> As Christoph said: "The count is not needed for any operations. Just for the
> slabinfo output. The value has no operational value for the allocator
> itself. So why use extra logic to track it in potentially performance
> critical paths?"
> 
> In this way, could we show the approximate value of active_objs in the
> /proc/slabinfo?
> 
> Based on the following information:
> In the discard_slab() function, page->inuse is equal to page->total_objects;
> In the allocate_slab() function, page->inuse is also equal to
> page->total_objects (with one exception: for kmem_cache_node, page-> inuse
> equals 1);
> page->inuse will only change continuously when the obj is constantly
> allocated or released. (This should be the performance critical path
> emphasized by Christoph)
> 
> When users query the global slabinfo information, we may use total_objects
> to approximate active_objs.

Well, from one point of view, it makes no sense, because the ratio between
these two numbers is very meaningful: it's the slab utilization rate.

On the other side, with enabled per-cpu partial lists active_objs has
nothing to do with the reality anyway, so I agree with you, calling
count_partial() is almost useless.

That said, I wonder if the right thing to do is something like the patch below?

Thanks!

Roman

--

diff --git a/mm/slub.c b/mm/slub.c
index 1d644143f93e..ba0505e75ecc 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -2411,14 +2411,16 @@ static inline unsigned long node_nr_objs(struct kmem_cache_node *n)
 static unsigned long count_partial(struct kmem_cache_node *n,
                                        int (*get_count)(struct page *))
 {
-       unsigned long flags;
        unsigned long x = 0;
+#ifdef CONFIG_SLUB_CPU_PARTIAL
+       unsigned long flags;
        struct page *page;
 
        spin_lock_irqsave(&n->list_lock, flags);
        list_for_each_entry(page, &n->partial, slab_list)
                x += get_count(page);
        spin_unlock_irqrestore(&n->list_lock, flags);
+#endif
        return x;
 }
 #endif /* CONFIG_SLUB_DEBUG || CONFIG_SYSFS */