From mboxrd@z Thu Jan 1 00:00:00 1970 From: Michal Hocko Subject: Re: [rfc patch-rt] radix-tree: Partially disable memcg accounting in radix_tree_node_alloc() Date: Fri, 6 Jan 2017 09:55:48 +0100 Message-ID: <20170106085548.GD5556@dhcp22.suse.cz> References: <20161223163213.szj43nv7rnelkbty@linutronix.de> <1483690403.11478.8.camel@gmx.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Sebastian Andrzej Siewior , Thomas Gleixner , LKML , linux-rt-users , Steven Rostedt To: Mike Galbraith Return-path: Content-Disposition: inline In-Reply-To: <1483690403.11478.8.camel@gmx.de> Sender: linux-kernel-owner@vger.kernel.org List-Id: linux-rt-users.vger.kernel.org On Fri 06-01-17 09:13:23, Mike Galbraith wrote: > radix-tree: Partially disable memcg accounting in radix_tree_node_alloc() > > Having no preload, which turns accounting off for non-rt kernels, trying to > allocate coming from shmem_fault() when memcg is full sends us scurrying off > to pagefault_out_of_memory(), with dramatic (usually terminal) consequences. > LTP's madvise06 testcase triggers this quite well, and per gitk, the below > was the beginning of RT memcg woes. > > 58e698af4c63 radix-tree: account radix_tree_node to memory cgroup > > Turn memcg accounting off for RT in the problematic path. I am really wondering why this is RT specific and the non RT kernels doesn't have any problem. > Signed-off-by: Mike Galbraith > Cc: stable-rt@vger.kernel.org # +v4.6-rt > --- > lib/radix-tree.c | 7 +++++++ > 1 file changed, 7 insertions(+) > > --- a/lib/radix-tree.c > +++ b/lib/radix-tree.c > @@ -303,6 +303,13 @@ radix_tree_node_alloc(struct radix_tree_ > if (!gfpflags_allow_blocking(gfp_mask) && !in_interrupt()) { > struct radix_tree_preload *rtp; > > +#if defined(CONFIG_PREEMPT_RT_FULL) && defined(CONFIG_MEMCG) > + /* > + * Arriving here from shmem_fault() and meeting a full memcg > + * will send us to pagefault_out_of_memory(), and a dead box. > + */ > + gfp_mask &= ~__GFP_ACCOUNT; > +#endif > /* > * Even if the caller has preloaded, try to allocate from the > * cache first for the new node to get accounted to the memory -- Michal Hocko SUSE Labs