From mboxrd@z Thu Jan 1 00:00:00 1970 From: Johannes Weiner Subject: Re: [PATCH 03/14] net: tcp_memcontrol: properly detect ancestor socket pressure Date: Sat, 14 Nov 2015 10:15:18 -0500 Message-ID: <20151114151518.GB28175@cmpxchg.org> References: <1447371693-25143-1-git-send-email-hannes@cmpxchg.org> <1447371693-25143-4-git-send-email-hannes@cmpxchg.org> <20151114124552.GI31308@esperanza> Mime-Version: 1.0 Return-path: Content-Disposition: inline In-Reply-To: <20151114124552.GI31308@esperanza> Sender: linux-kernel-owner@vger.kernel.org List-ID: Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Vladimir Davydov Cc: David Miller , Andrew Morton , Tejun Heo , Michal Hocko , netdev@vger.kernel.org, linux-mm@kvack.org, cgroups@vger.kernel.org, linux-kernel@vger.kernel.org, kernel-team@fb.com On Sat, Nov 14, 2015 at 03:45:52PM +0300, Vladimir Davydov wrote: > On Thu, Nov 12, 2015 at 06:41:22PM -0500, Johannes Weiner wrote: > > When charging socket memory, the code currently checks only the local > > page counter for excess to determine whether the memcg is under socket > > pressure. But even if the local counter is fine, one of the ancestors > > could have breached its limit, which should also force this child to > > enter socket pressure. This currently doesn't happen. > > > > Fix this by using page_counter_try_charge() first. If that fails, it > > means that either the local counter or one of the ancestors are in > > excess of their limit, and the child should enter socket pressure. > > > > Signed-off-by: Johannes Weiner > > Reviewed-by: Vladimir Davydov Thanks Vladimir! > For the record: it was broken by commit 3e32cb2e0a12 ("mm: memcontrol: > lockless page counters"). I didn't realize that. Fixes: 3e32cb2e0a12 ("mm: memcontrol: lockless page counters")