From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751331AbdAPMCo (ORCPT ); Mon, 16 Jan 2017 07:02:44 -0500 Received: from mx2.suse.de ([195.135.220.15]:40865 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751054AbdAPMCm (ORCPT ); Mon, 16 Jan 2017 07:02:42 -0500 Date: Mon, 16 Jan 2017 13:02:36 +0100 From: Michal Hocko To: Tim Chen Cc: Andrew Morton , Ying Huang , dave.hansen@intel.com, ak@linux.intel.com, aaron.lu@intel.com, linux-mm@kvack.org, linux-kernel@vger.kernel.org, Hugh Dickins , Shaohua Li , Minchan Kim , Rik van Riel , Andrea Arcangeli , "Kirill A . Shutemov" , Vladimir Davydov , Johannes Weiner , Hillf Danton , Christian Borntraeger , Jonathan Corbet Subject: Re: [PATCH v5 0/9] mm/swap: Regular page swap optimizations Message-ID: <20170116120236.GG13641@dhcp22.suse.cz> References: MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.6.0 (2016-04-01) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hi, I am seeing a lot of preempt unsafe warnings with the current mmotm and I assume that this patchset has introduced the issue. I haven't checked more closely but get_swap_page didn't use this_cpu_ptr before "mm/swap: add cache for swap slots allocation" [ 57.812314] BUG: using smp_processor_id() in preemptible [00000000] code: kswapd0/527 [ 57.814360] caller is debug_smp_processor_id+0x17/0x19 [ 57.815237] CPU: 1 PID: 527 Comm: kswapd0 Tainted: G W 4.9.0-mmotm-00135-g4e9a9895ebef #1042 [ 57.816019] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.1-1 04/01/2014 [ 57.816019] ffffc900001939c0 ffffffff81329c60 0000000000000001 ffffffff81a0ce06 [ 57.816019] ffffc900001939f0 ffffffff81343c2a 00000000000137a0 ffffea0000dfd2a0 [ 57.816019] ffff88003c49a700 ffffc90000193b10 ffffc90000193a00 ffffffff81343c53 [ 57.816019] Call Trace: [ 57.816019] [] dump_stack+0x68/0x92 [ 57.816019] [] check_preemption_disabled+0xce/0xe0 [ 57.816019] [] debug_smp_processor_id+0x17/0x19 [ 57.816019] [] get_swap_page+0x19/0x183 [ 57.816019] [] shmem_writepage+0xce/0x38c [ 57.816019] [] shrink_page_list+0x81f/0xdbf [ 57.816019] [] shrink_inactive_list+0x2ab/0x594 [ 57.816019] [] shrink_node_memcg+0x4c7/0x673 [ 57.816019] [] shrink_node+0xc4/0x282 [ 57.816019] [] ? shrink_node+0xc4/0x282 [ 57.816019] [] kswapd+0x656/0x834 [ 57.816019] [] ? mem_cgroup_shrink_node+0x2e1/0x2e1 [ 57.816019] [] ? call_usermodehelper_exec_async+0x124/0x12d [ 57.816019] [] kthread+0xf9/0x101 [ 57.816019] [] ? _raw_spin_unlock_irq+0x2c/0x4a [ 57.816019] [] ? kthread_park+0x5a/0x5a [ 57.816019] [] ? umh_complete+0x25/0x25 [ 57.816019] [] ret_from_fork+0x27/0x40 I thought a simple diff --git a/mm/swap_slots.c b/mm/swap_slots.c index 8cf941e09941..732194de58a4 100644 --- a/mm/swap_slots.c +++ b/mm/swap_slots.c @@ -303,7 +303,7 @@ swp_entry_t get_swap_page(void) swp_entry_t entry, *pentry; struct swap_slots_cache *cache; - cache = this_cpu_ptr(&swp_slots); + cache = &get_cpu_var(swp_slots); entry.val = 0; if (check_cache_active()) { @@ -322,11 +322,13 @@ swp_entry_t get_swap_page(void) } mutex_unlock(&cache->alloc_lock); if (entry.val) - return entry; + goto out; } get_swap_pages(1, &entry); +out: + put_cpu_var(swp_slots); return entry; } would be a way to go but the function takes a sleeping lock so disabling the preemption is not a way forward. So this is either preempt safe for some reason - which should be IMHO documented in a comment - and raw_cpu_ptr can be used or this needs a deeper thought. -- Michal Hocko SUSE Labs