From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-qk0-f198.google.com (mail-qk0-f198.google.com [209.85.220.198]) by kanga.kvack.org (Postfix) with ESMTP id 8E1CC6B0253 for ; Sat, 13 Aug 2016 13:34:35 -0400 (EDT) Received: by mail-qk0-f198.google.com with SMTP id j6so36629215qkc.3 for ; Sat, 13 Aug 2016 10:34:35 -0700 (PDT) Received: from mx1.redhat.com (mx1.redhat.com. [209.132.183.28]) by mx.google.com with ESMTPS id u65si7143730qkc.130.2016.08.13.10.34.34 for (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Sat, 13 Aug 2016 10:34:34 -0700 (PDT) Date: Sat, 13 Aug 2016 13:34:29 -0400 (EDT) From: Mikulas Patocka Subject: Re: [dm-devel] [RFC PATCH 2/2] mm, mempool: do not throttle PF_LESS_THROTTLE tasks In-Reply-To: <20160812123242.GH3639@dhcp22.suse.cz> Message-ID: References: <20160722091558.GF794@dhcp22.suse.cz> <878twt5i1j.fsf@notabene.neil.brown.name> <20160725083247.GD9401@dhcp22.suse.cz> <87lh0n4ufs.fsf@notabene.neil.brown.name> <20160727182411.GE21859@dhcp22.suse.cz> <87eg6e4vhc.fsf@notabene.neil.brown.name> <20160728071711.GB31860@dhcp22.suse.cz> <20160803143419.GC1490@dhcp22.suse.cz> <20160812123242.GH3639@dhcp22.suse.cz> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-linux-mm@kvack.org List-ID: To: Michal Hocko Cc: Mel Gorman , NeilBrown , Tetsuo Handa , LKML , linux-mm@kvack.org, "dm-devel@redhat.com David Rientjes" , Ondrej Kozina , Andrew Morton On Fri, 12 Aug 2016, Michal Hocko wrote: > On Thu 04-08-16 14:49:41, Mikulas Patocka wrote: > > > On Wed, 3 Aug 2016, Michal Hocko wrote: > > > > > But the device congestion is not the only condition required for the > > > throttling. The pgdat has also be marked congested which means that the > > > LRU page scanner bumped into dirty/writeback/pg_reclaim pages at the > > > tail of the LRU. That should only happen if we are rotating LRUs too > > > quickly. AFAIU the reclaim shouldn't allow free ticket scanning in that > > > situation. > > > > The obvious problem here is that mempool allocations should sleep in > > mempool_alloc() on &pool->wait (until someone returns some entries into > > the mempool), they should not sleep inside the page allocator. > > I agree that mempool_alloc should _primarily_ sleep on their own > throttling mechanism. I am not questioning that. I am just saying that > the page allocator has its own throttling which it relies on and that > cannot be just ignored because that might have other undesirable side > effects. So if the right approach is really to never throttle certain > requests then we have to bail out from a congested nodes/zones as soon > as the congestion is detected. > > Now, I would like to see that something like that is _really_ necessary. Currently, it is not a problem - device mapper reports the device as congested only if the underlying physical disks are congested. But once we change it so that device mapper reports congested state on its own (when it has too many bios in progress), this starts being a problem. I would add PF_NO_THROTTLE or __GFP_NO_THROTTLE to mempool_alloc. Or - we can prevent the memory reclaim from throttling if we see both __GFP_NOMEMALLOC and __GFP_NORETRY - that would be sufficient to detect mempool_alloc usage and it wouldn't hurt other __GFP_NORETRY users. Mikulas > I believe that we should simply start with easier part and get rid of > throttle_vm_writeout because that seems like a left over from the past. > If that turns out unsatisfactory and we have clear picture when the > throttling is harmful/suboptimal then we can move on with a more complex > solution. Does this sound like a way forward? > > -- > Michal Hocko > SUSE Labs -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org