From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-3.7 required=3.0 tests=HEADER_FROM_DIFFERENT_DOMAINS, INCLUDES_PATCH,LOTS_OF_MONEY,MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS, URIBL_BLOCKED autolearn=no autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 48A32C2BB55 for ; Thu, 16 Apr 2020 17:22:05 +0000 (UTC) Received: from kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by mail.kernel.org (Postfix) with ESMTP id 00D6C2076D for ; Thu, 16 Apr 2020 17:22:04 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 00D6C2076D Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=canonical.com Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=owner-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix) id 877B88E00D7; Thu, 16 Apr 2020 13:22:04 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id 829268E00BC; Thu, 16 Apr 2020 13:22:04 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 716538E00D7; Thu, 16 Apr 2020 13:22:04 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from forelay.hostedemail.com (smtprelay0227.hostedemail.com [216.40.44.227]) by kanga.kvack.org (Postfix) with ESMTP id 538AA8E00BC for ; Thu, 16 Apr 2020 13:22:04 -0400 (EDT) Received: from smtpin14.hostedemail.com (10.5.19.251.rfc1918.com [10.5.19.251]) by forelay02.hostedemail.com (Postfix) with ESMTP id 1A8C06122 for ; Thu, 16 Apr 2020 17:22:04 +0000 (UTC) X-FDA: 76714386168.14.leg68_6940d1a660234 X-HE-Tag: leg68_6940d1a660234 X-Filterd-Recvd-Size: 13816 Received: from youngberry.canonical.com (youngberry.canonical.com [91.189.89.112]) by imf25.hostedemail.com (Postfix) with ESMTP for ; Thu, 16 Apr 2020 17:22:03 +0000 (UTC) Received: from mail-wm1-f70.google.com ([209.85.128.70]) by youngberry.canonical.com with esmtps (TLS1.2:ECDHE_RSA_AES_128_GCM_SHA256:128) (Exim 4.86_2) (envelope-from ) id 1jP8DB-0001Xd-SC for linux-mm@kvack.org; Thu, 16 Apr 2020 17:22:01 +0000 Received: by mail-wm1-f70.google.com with SMTP id f81so1477153wmf.2 for ; Thu, 16 Apr 2020 10:22:01 -0700 (PDT) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:date:from:to:cc:subject:message-id:references :mime-version:content-disposition:in-reply-to; bh=qh4D9CE/fwCVPs0RPqHBykT7zfOm4kgXeMe1OaweoiA=; b=ZlF8s7ZBrwq1Z7I0kI2OmJ5Rkb8bMT4oL01Zdg77P6w7HDFSRV/xTLEfROeoB3v4u0 DGj+HYoxIXNFUa9GsPpdLtkCL0Gwxw6t/Sg4pqnkQDhtTH6eK3oAzF/2O9SVeglqBMq/ ccKUMzPjn71tea0Ds48EzHZ6TxA19rEY7KBIjsJ+fPVEH0YEk9S4Hw/UYtU8w6RC+3JY mj9LKtUdw5SDU0yX3a9SWIdzQgAuZhXgonGpoMltNwYeaHA8LZxWN2Ynv0q+YSjV217g Vzs+eIXOorEZrwE8vpEYdzhaPeyjI1v+hyJFPgknB6i+W9SrlIrJ5HnvJdDThHC9W9BR IYEw== X-Gm-Message-State: AGi0PuZL9YrmF7+UlHrDYb23T7T4zW8Tr0RidXmHKawAPkrs+1M7Kj1S bwulUNO4mOHKbJ0YCgPhKe/9g5XF2MHnpxuVSDQQRTwa2sCVire5/BSi98KXV5aGkDrFCLgGm/r qoWHjUyix005pG72n2IWSBAXrXK22 X-Received: by 2002:a1c:e242:: with SMTP id z63mr6107090wmg.72.1587057721320; Thu, 16 Apr 2020 10:22:01 -0700 (PDT) X-Google-Smtp-Source: APiQypLaQ3KqX31QNOFqNLjJ3XDnQN7TYeWJ3tcPMPaxnFrrUHF0+tkXbgJYOImqF51XyOxYGhLOow== X-Received: by 2002:a1c:e242:: with SMTP id z63mr6107058wmg.72.1587057720834; Thu, 16 Apr 2020 10:22:00 -0700 (PDT) Received: from localhost (host123-127-dynamic.36-79-r.retail.telecomitalia.it. [79.36.127.123]) by smtp.gmail.com with ESMTPSA id q8sm4333493wmg.22.2020.04.16.10.21.59 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Thu, 16 Apr 2020 10:22:00 -0700 (PDT) Date: Thu, 16 Apr 2020 19:21:58 +0200 From: Andrea Righi To: "Huang, Ying" Cc: Andrew Morton , Minchan Kim , Anchal Agarwal , linux-mm@kvack.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH v2] mm: swap: use fixed-size readahead during swapoff Message-ID: <20200416172158.GA3352@xps-13> References: <20200413111810.GA801367@xps-13> <87a73f7d71.fsf@yhuang-dev.intel.com> <20200413133150.GA810380@xps-13> <87wo6i6efn.fsf@yhuang-dev.intel.com> <20200414130520.GF810380@xps-13> <87v9m1zd83.fsf@yhuang-dev.intel.com> <20200415073239.GG810380@xps-13> <20200415120011.GI810380@xps-13> <871rooz2gs.fsf@yhuang-dev.intel.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <871rooz2gs.fsf@yhuang-dev.intel.com> X-Bogosity: Ham, tests=bogofilter, spamicity=0.000000, version=1.2.4 Sender: owner-linux-mm@kvack.org Precedence: bulk X-Loop: owner-majordomo@kvack.org List-ID: On Thu, Apr 16, 2020 at 08:41:39AM +0800, Huang, Ying wrote: > Andrea Righi writes: > > > On Wed, Apr 15, 2020 at 09:32:47AM +0200, Andrea Righi wrote: > >> On Wed, Apr 15, 2020 at 10:37:00AM +0800, Huang, Ying wrote: > >> > Andrea Righi writes: > >> > > >> > > On Tue, Apr 14, 2020 at 09:31:24AM +0800, Huang, Ying wrote: > >> > >> Andrea Righi writes: > >> > >> > >> > >> > On Mon, Apr 13, 2020 at 09:00:34PM +0800, Huang, Ying wrote: > >> > >> >> Andrea Righi writes: > >> > >> >> > >> > >> >> [snip] > >> > >> >> > >> > >> >> > diff --git a/mm/swap_state.c b/mm/swap_state.c > >> > >> >> > index ebed37bbf7a3..c71abc8df304 100644 > >> > >> >> > --- a/mm/swap_state.c > >> > >> >> > +++ b/mm/swap_state.c > >> > >> >> > @@ -20,6 +20,7 @@ > >> > >> >> > #include > >> > >> >> > #include > >> > >> >> > #include > >> > >> >> > +#include > >> > >> >> > #include > >> > >> >> > > >> > >> >> > #include > >> > >> >> > @@ -507,6 +508,14 @@ static unsigned long swapin_nr_pages(unsigned long offset) > >> > >> >> > max_pages = 1 << READ_ONCE(page_cluster); > >> > >> >> > if (max_pages <= 1) > >> > >> >> > return 1; > >> > >> >> > + /* > >> > >> >> > + * If current task is using too much memory or swapoff is running > >> > >> >> > + * simply use the max readahead size. Since we likely want to load a > >> > >> >> > + * lot of pages back into memory, using a fixed-size max readhaead can > >> > >> >> > + * give better performance in this case. > >> > >> >> > + */ > >> > >> >> > + if (oom_task_origin(current)) > >> > >> >> > + return max_pages; > >> > >> >> > > >> > >> >> > hits = atomic_xchg(&swapin_readahead_hits, 0); > >> > >> >> > pages = __swapin_nr_pages(prev_offset, offset, hits, max_pages, > >> > >> >> > >> > >> >> Thinks this again. If my understanding were correct, the accessing > >> > >> >> pattern during swapoff is sequential, why swap readahead doesn't work? > >> > >> >> If so, can you root cause that firstly? > >> > >> > > >> > >> > Theoretically if the pattern is sequential the current heuristic should > >> > >> > already select a big readahead size, but apparently it's not doing that. > >> > >> > > >> > >> > I'll repeat my tests tracing the readahead size during swapoff to see > >> > >> > exactly what's going on here. > >> > >> > >> > >> I haven't verify it. It may be helpful to call lookup_swap_cache() > >> > >> before swapin_readahead() in unuse_pte_range(). The theory behind it is > >> > >> to update the swap readahead statistics via lookup_swap_cache(). > >> > > > >> > > I did more tests trying to collect some useful information. > >> > > > >> > > In particular I've been focusing at tracing the distribution of the > >> > > values returned by swapin_nr_pages() in different scenarios. > >> > > > >> > > To do so I made swapin_nr_pages() trace-able and I used the following > >> > > bcc command to measure the distrubution of the returned values: > >> > > > >> > > # argdist-bpfcc -c -C 'r::swapin_nr_pages(unsigned long offset):unsigned long:$retval' > >> > > > >> > > I've collected this metric in the following scenarios: > >> > > - 5.6 vanilla > >> > > - 5.6 + lookup_swap_cache() before swapin_readahead() in > >> > > unuse_pte_range() > >> > > - 5.6 + atomic_inc(&swapin_readahead_hits) before swapin_readahead() > >> > > in unuse_pte_range() > >> > > - 5.6 + swapin_readahead_hits=last_readahead_pages (in the atomic way) > >> > > before swapin_readahead() in unuse_pte_range() > >> > > > >> > > Each kernel has been tested both with swappiness=0 and swappiness=60. > >> > > Results are pretty much identical changing the swappiness, so I'm just > >> > > reporting the default case here (w/ swappiness=60). > >> > > > >> > > Result > >> > > ====== > >> > > > >> > > = swapoff performance (elapsed time) = > >> > > > >> > > vanilla 22.09s > >> > > lookup_swap_cache() 23.87s > >> > > hits++ 16.10s > >> > > hits=last_ra_pages 8.81s > >> > > > >> > > = swapin_nr_pages() $retval distribution = > >> > > > >> > > 5.6 vanilla: > >> > > r::swapin_nr_pages(unsigned long offset):unsigned long:$retval > >> > > COUNT EVENT > >> > > 36948 $retval = 8 > >> > > 44151 $retval = 4 > >> > > 49290 $retval = 1 > >> > > 527771 $retval = 2 > >> > > > >> > > 5.6 lookup_swap_cache() before swapin_readahead(): > >> > > r::swapin_nr_pages(unsigned long offset):unsigned long:$retval > >> > > COUNT EVENT > >> > > 13093 $retval = 1 > >> > > 56703 $retval = 8 > >> > > 123067 $retval = 2 > >> > > 366118 $retval = 4 > >> > > > >> > > 5.6 atomic_inc(&swapin_readahead_hits) before swapin_readahead(): > >> > > r::swapin_nr_pages(unsigned long offset):unsigned long:$retval > >> > > COUNT EVENT > >> > > 2589 $retval = 1 > >> > > 8016 $retval = 2 > >> > > 40021 $retval = 8 > >> > > 566038 $retval = 4 > >> > > > >> > > 5.6 swapin_readahead_hits=last_readahead_pages before swapin_readahead(): > >> > > r::swapin_nr_pages(unsigned long offset):unsigned long:$retval > >> > > COUNT EVENT > >> > > 785 $retval = 2 > >> > > 1072 $retval = 1 > >> > > 21844 $retval = 4 > >> > > 644168 $retval = 8 > >> > > > >> > > In the vanilla case, the readahead heuristic seems to choose 2 pages > >> > > most of the time. This is because we are not properly considering the > >> > > hits (hits are always 0 in the swapoff code path) and, as you correctly > >> > > pointed out, we can fix this by calling lookup_swap_cache() in > >> > > unuse_pte_range() before calling swapin_readahead(). > >> > > > >> > > With this change the distribution of the readahead size moves more > >> > > toward 4 pages, but we still have some 2s. That looks good, however it > >> > > doesn't seem to speed up swapoff very much... maybe because calling > >> > > lookup_swap_cache() introduces a small overhead? (still need to > >> > > investigate about this theory). > >> > > > >> > > In the next test I've tried to always increment hits by 1 before calling > >> > > swapin_readahead() in unuse_pte_range(). This is basically cheating, > >> > > because I'm faking the hit ratio, forcing the heuristic to use a larger > >> > > readahead size; in fact, the readahead size moves even more toward 4 > >> > > pages and swapoff performance are a little better now. > >> > > > >> > > Pushing even more the "cheating" I can pretend that the previous > >> > > readahead was all hits (swapin_readahead_hits=last_readahead_pages), so > >> > > I'm forcing the heuristic to move toward the max size and keep using it. > >> > > The result here is pretty much identical to my fixed-size patch, because > >> > > swapin_nr_pages() returns the max readahead size pretty much all the > >> > > time during swapoff (8 pages or, more in general, vm.page-cluster). > >> > > > >> > > Personally I don't like very much forcing the heuristic in this way, > >> > > it'd be nice if it would just work by accounting the proper hit ratio > >> > > (so just by adding lookup_swap_cache() as you correctly suggested), but > >> > > this solution doesn't seem to improve performance in reality. For this > >> > > reason I still think we should consider the swapoff scenario like a > >> > > special one and somehow bypass the readahead heuristic and always return > >> > > the max readahead size. > >> > > > >> > > Looking at the hits of the previous step in the swapoff case just > >> > > doesn't work, because we may have some misses, but they will become hits > >> > > very soon, since we are reading all the swapped out pages back into > >> > > memory. This is why using the max readahead size gives better > >> > > swapoff performance. > >> > > > >> > > What do you think? > >> > > >> > >From your description, it appears that you are using cluster readahead > >> > instead of vma readahead. Can you verify this via, > >> > > >> > # cat /sys/kernel/mm/swap/vma_ra_enabled > >> > >> # cat /sys/kernel/mm/swap/vma_ra_enabled > >> true > >> > >> However, it's still using the cluster readahead because I'm using a > >> swap file and nr_rotate_swap=1, so, according to the following, it's > >> never using the vma readahead: > >> > >> static inline bool swap_use_vma_readahead(void) > >> { > >> return READ_ONCE(enable_vma_readahead) && !atomic_read(&nr_rotate_swap); > >> } > >> > >> I'll investigate more on this, I think there's no reason to prevent the > >> usage of vma readahead if the underlying device is non-rotational. > > > > Few more details about this. > > > > Even if I have vma_ra_enabled (in /sys/kernel/mm/swap/vma_ra_enabled) it > > looks like my virtio_blk device is considered rotational by default: > > > > $ cat /sys/block/vda/queue/rotational > > 1 > > > > Therefore the vma readahead is not used. If I change this to > > "rotational" then the vma readahead is used; this is confirmed by the > > fact that swapin_nr_pages isn't called anymore: > > > > r::swapin_nr_pages(unsigned long offset):unsigned long:$retval > > COUNT EVENT > > 13 $retval = 1 > > 18 $retval = 2 > > 23 $retval = 4 > > 29 $retval = 8 > > > > swapoff time: 16.44s > > > > In terms of swapoff performance vma readahead works better than the > > vanilla cluster readahead (~21s), but the "improved" cluster readahead > > (with lookup_swap_cache() in unuse_pte_range()) still gives slightly > > better results (~12s). > > Thanks for testing. Can you check the return value of > __swapin_nr_pages() to verify whether vma readahead works as expected? This is the vanilla kernel w/ vma readahead enabled: r::__swapin_nr_pages():unsigned int:$retval COUNT EVENT 1286 $retval = 1 6516 $retval = 4 81252 $retval = 8 260815 $retval = 2 swapoff time: real 18.20 And this is the kernel with the lookup_swap_cache() patch and vma readahead enabled: r::__swapin_nr_pages():unsigned int:$retval COUNT EVENT 321 $retval = 2 509 $retval = 1 3017 $retval = 4 124090 $retval = 8 swapoff time: 15.33s So, vma readahead seems to work as expected and swapoff now is faster also with vma readahead enabled. -Andrea