From: Michal Hocko Subject: Re: [PATCH v4 3/4] mm/page_owner: Print memcg information Date: Mon, 7 Feb 2022 18:20:04 +0100 Message-ID: References: <20220131192308.608837-5-longman@redhat.com> <20220202203036.744010-4-longman@redhat.com> <3f042edb-3769-afea-17a7-899578cd5c69@redhat.com> Mime-Version: 1.0 Return-path: DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=suse.com; s=susede1; t=1644254409; h=from:from:reply-to:date:date:message-id:message-id:to:to:cc:cc: mime-version:mime-version:content-type:content-type: in-reply-to:in-reply-to:references:references; bh=WgeOunKZPWr1u7CmKBkIn3aI6IvIjxAaiaRIxKsTWU4=; b=oJXyY0lg033cJKb3vjdcvI1hPZc0gH3NGf+1IsuEJ2ysu4j/U4GQMpA/uH+PHT9ObqH4c3 /ifMCTOMsZOAoJluh2kAlCYa+QIjE9Jlssbw6yrsZiFNwdqn5cHqm8RNt6f4GoH3hPvEjk GOdslupORGgeUQ8/DjTm1kwISSSRba0= Content-Disposition: inline In-Reply-To: <3f042edb-3769-afea-17a7-899578cd5c69@redhat.com> List-ID: Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Waiman Long Cc: Johannes Weiner , Vladimir Davydov , Andrew Morton , Petr Mladek , Steven Rostedt , Sergey Senozhatsky , Andy Shevchenko , Rasmus Villemoes , linux-kernel@vger.kernel.org, cgroups@vger.kernel.org, linux-mm@kvack.org, Ira Weiny , Mike Rapoport , David Rientjes , Roman Gushchin , Rafael Aquini On Thu 03-02-22 14:03:58, Waiman Long wrote: > On 2/3/22 07:46, Michal Hocko wrote: > > On Wed 02-02-22 15:30:35, Waiman Long wrote: > > [...] > > > +#ifdef CONFIG_MEMCG > > > + unsigned long memcg_data; > > > + struct mem_cgroup *memcg; > > > + bool online; > > > + char name[80]; > > > + > > > + rcu_read_lock(); > > > + memcg_data = READ_ONCE(page->memcg_data); > > > + if (!memcg_data) > > > + goto out_unlock; > > > + > > > + if (memcg_data & MEMCG_DATA_OBJCGS) > > > + ret += scnprintf(kbuf + ret, count - ret, > > > + "Slab cache page\n"); > > > + > > > + memcg = page_memcg_check(page); > > > + if (!memcg) > > > + goto out_unlock; > > > + > > > + online = (memcg->css.flags & CSS_ONLINE); > > > + cgroup_name(memcg->css.cgroup, name, sizeof(name)); > > Is there any specific reason to use another buffer allocated on the > > stack? Also 80B seems too short to cover NAME_MAX. > > > > Nothing else jumped at me. > > I suppose we can print directly into kbuf with cgroup_name(), but using a > separate buffer is easier to read and understand. 79 characters should be > enough for most cgroup names. Some auto-generated names with some kind of > embedded uuids may be longer than that, but the random sequence of hex > digits that may be missing do not convey much information for identification > purpose. We can always increase the buffer length later if it turns out to > be an issue. Cutting a name short sounds like a source of confusion and there doesn't seem to be any good reason for that. -- Michal Hocko SUSE Labs