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 Received: from kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by smtp.lore.kernel.org (Postfix) with ESMTP id 414BCC433F5 for ; Thu, 12 May 2022 20:03:21 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id C5F888D0009; Thu, 12 May 2022 16:03:20 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id C0E568D0006; Thu, 12 May 2022 16:03:20 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id A87E98D0009; Thu, 12 May 2022 16:03:20 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from relay.hostedemail.com (smtprelay0012.hostedemail.com [216.40.44.12]) by kanga.kvack.org (Postfix) with ESMTP id 9AFA08D0006 for ; Thu, 12 May 2022 16:03:20 -0400 (EDT) Received: from smtpin17.hostedemail.com (a10.router.float.18 [10.200.18.1]) by unirelay02.hostedemail.com (Postfix) with ESMTP id 6B5B53192A for ; Thu, 12 May 2022 20:03:20 +0000 (UTC) X-FDA: 79458165360.17.EBC5430 Received: from mail-pj1-f51.google.com (mail-pj1-f51.google.com [209.85.216.51]) by imf04.hostedemail.com (Postfix) with ESMTP id 42687400B8 for ; Thu, 12 May 2022 20:03:09 +0000 (UTC) Received: by mail-pj1-f51.google.com with SMTP id l11-20020a17090a49cb00b001d923a9ca99so5857991pjm.1 for ; Thu, 12 May 2022 13:03:19 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20210112; h=date:from:to:cc:subject:in-reply-to:message-id:references :mime-version; bh=jVrb2+IraqeZ1X6TcLlWAw3K/3OKowkPYsrTxsU4XuM=; b=i3uW/9FN2Ta1j+1wem+qsbvLpIuSBayyv9KAEOIy20fOfxOW8omRBZTtrqxwquOWbi 9ubPiWhdh6+f+LCD+l7y7//9BRVfTjK5PRq/2iL1yeiaxCvMJyGlp6wyanST5vPMjK4O VmlngIJkUrO1fM9/oOAqlpmhWAX+ZGeMeQnsXy/9qMbxjuKLervC3LXUcaWVvaDF2JFf AjXjRMRhTjksfOBzK4mjyHbQcBNwxBxt98fYudNjUWHJ7pavdoMN0NSCUtMePYuXxLaE 78wkJiJ3mnQVa/QdmAx84p1AaIAmSUpAPsgnyn2Wga5E3G9106MCHvIAGoykaCSGW2PE dlFg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:date:from:to:cc:subject:in-reply-to:message-id :references:mime-version; bh=jVrb2+IraqeZ1X6TcLlWAw3K/3OKowkPYsrTxsU4XuM=; b=UIR0lZrOPjyyh9IVn0mEx9KrVvFfDXPHJAYcvC5QtotbjNUeyVkllQMJai/nJhhlF0 2mgjrluks3C/vmhtPEXz6T7Fwz2qxsan8bcys1CLSyc+w7W1Yog/mncfeo+Hi2XeQhjD ITwyfUN4MafNltQR84M2wOEq9DrvAOuw/FXfVvRTLnjp+tfGUqGlH66eWdagyDvit0Se w78p+5OACmsY0QyP78jLVJciCxjdUB0NjyGq723UJIpMvfH4b+nKbRNT9aK7088hNrHC xQ/6/ExlTEaDZXMABPSXgG+Ad4xFiEfBWlP/AuApj8Jq0If5lRoDhEmA2MM7u9ZYVI9B ZGVw== X-Gm-Message-State: AOAM530MwrrQfIFljNOfdHjAvskvY0llG29oNRoVsL6wLh6kEHnr4lFl 5N10x3MZCa+iAJA5D5JCc1adVQ== X-Google-Smtp-Source: ABdhPJzK1+iDzvykokfD90jql3TdPhDzQkBr64U+r+wmXBycuj1BMvp6jyokncW8HmsRW8HMeN7nRA== X-Received: by 2002:a17:90b:1bd1:b0:1dc:7118:cf28 with SMTP id oa17-20020a17090b1bd100b001dc7118cf28mr1188888pjb.3.1652385798730; Thu, 12 May 2022 13:03:18 -0700 (PDT) Received: from [2620:15c:29:204:8122:ccb0:c783:c757] ([2620:15c:29:204:8122:ccb0:c783:c757]) by smtp.gmail.com with ESMTPSA id y14-20020a17090aa40e00b001d287fd3f79sm189524pjp.46.2022.05.12.13.03.17 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Thu, 12 May 2022 13:03:18 -0700 (PDT) Date: Thu, 12 May 2022 13:03:17 -0700 (PDT) From: David Rientjes To: Zach O'Keefe cc: Rongwei Wang , Alex Shi , David Hildenbrand , Matthew Wilcox , Michal Hocko , Pasha Tatashin , Peter Xu , SeongJae Park , Song Liu , Vlastimil Babka , Yang Shi , Zi Yan , linux-mm@kvack.org, Andrea Arcangeli , Andrew Morton , Arnd Bergmann , Axel Rasmussen , Chris Kennelly , Chris Zankel , Helge Deller , Hugh Dickins , Ivan Kokshaysky , "James E.J. Bottomley" , Jens Axboe , "Kirill A. Shutemov" , Matt Turner , Max Filippov , Miaohe Lin , Minchan Kim , Patrick Xia , Pavel Begunkov , Thomas Bogendoerfer , calling@linux.alibaba.com Subject: Re: [PATCH v5 10/13] mm/madvise: add MADV_COLLAPSE to process_madvise() In-Reply-To: Message-ID: <3a53435-5d5a-d6cb-5739-5c444523fc7@google.com> References: <20220504214437.2850685-1-zokeefe@google.com> <20220504214437.2850685-11-zokeefe@google.com> <502a3ced-f3c6-7117-3b24-d80d204d66ee@linux.alibaba.com> MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII X-Rspamd-Server: rspam10 X-Rspamd-Queue-Id: 42687400B8 Authentication-Results: imf04.hostedemail.com; dkim=pass header.d=google.com header.s=20210112 header.b="i3uW/9FN"; spf=pass (imf04.hostedemail.com: domain of rientjes@google.com designates 209.85.216.51 as permitted sender) smtp.mailfrom=rientjes@google.com; dmarc=pass (policy=reject) header.from=google.com X-Rspam-User: X-Stat-Signature: 4uybi593bocxhp8o8zoeibuxazy3gpsf X-HE-Tag: 1652385789-837755 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 Wed, 11 May 2022, Zach O'Keefe wrote: > Hey Rongwei, > > Thanks for taking the time to review! > > On Tue, May 10, 2022 at 5:49 PM Rongwei Wang > wrote: > > > > Hi, Zach > > > > Thanks for your great patchset! > > Recently, We also try to collapse THP in this way, likes performance > > degradation due to using too much hugepages in our scenes. > > Rongwei, could you elaborate on this? I can understand undesired overhead for allocation of a hugepage at the time of fault if there is not enough benefit derived by the hugepages over the long term (like for a database workload), is this the performance degradation you're referring to? Otherwise I'm unfamiliar with performance degradation for using too much hugepages after they have been allocated :) Maybe RSS grows too much and we run into memory pressure? It would be good to know more if you can share details here. > > And there is a doubt about process_madvise(MADV_COLLAPSE) when we test > > this patchset:. It seems that process_madvise(MADV_COLLAPSE) rely on > > madvise(MADV_HUGEPAGE)? If the vma wasn't marked with 'hg', > > process_madvise(MADV_COLLAPSE) will fail to collapse. And if I miss > > something, please let me know. > > > > I tried to have MADV_COLLAPSE follow the same THP eligibility > semantics as khugepaged and at-fault: either THP=always, or > THP=madvise and the vma is marked with MADV_HUGEPAGE, as you point > out. > > If I understand you correctly, the usefulness of > process_madvise(MADV_COLLAPSE) is limited in the case where > THP=madvise and a CAP_SYS_ADMIN user is requesting a collapse of > behalf of another process since they don't have a way to mark the > target memory as eligible (which requires VM_HUGEPAGE). > > If so, I think that's a valid point, and your suggestion below of a > supporting MADV_[NO]HUGEPAGE for process_madvise(2) makes sense. For > the sake of exploring all options, I'll mention that there was also a > previous idea suggested by Yang Shi where MADV_COLLAPSE could also set > VM_HUGEPAGE[1]. > If a user is doing MADV_COLLAPSE on behalf of itself, it seems unnecessary to need to do MADV_HUGEPAGE before that regardless of system-wide settings that it may not control? Same point for a root user doing this on behalf of another user. It could either do MADV_HUGEPAGE and change the behavior that the user perhaps requested behind its back (not desired) or it could temporarily set the system-wide setting to allow THP always before doing the MADV_COLLAPSE (it's root). We can simply allow MADV_COLLAPSE to always collapse in either context regardless of VM_HUGEPAGE, VM_NOHUGEPAGE, or system-wide settings? > Since it's possible supporting MADV_[NO]HUGEPAGE for > process_madivse(2) has applications outside a subsequent > MADV_COLLAPSE, and since I don't see process_madvise(MADV_COLLAPSE) to > be in a hot path, I'd vote in favor of your suggestion and include > process_madvise(MADV_[NO]HUGEPAGE) support in v6 unless others object. > > Thanks again for your review and your suggestion! > Zach > > [1] https://lore.kernel.org/linux-mm/CAHbLzkqLRBd6u3qn=KqpOhRcPZtpGXbTXLUjK1z=4d_dQ06Pvw@mail.gmail.com/ > > > If so, how about introducing process_madvise(MADV_HUGEPAGE) or > > process_madvise(MADV_NOHUGEPAGE)? The former helps to mark the target > > vma with 'hg', and the collapse process can be finished completely with > > the help of other processes. the latter could let some special vma avoid > > collapsing when setting 'THP=always'. > > > > Best regards, > > -wrw > > > > On 5/5/22 5:44 AM, Zach O'Keefe wrote: > > > Allow MADV_COLLAPSE behavior for process_madvise(2) if caller has > > > CAP_SYS_ADMIN or is requesting collapse of it's own memory. > > > > > > Signed-off-by: Zach O'Keefe > > > --- > > > mm/madvise.c | 6 ++++-- > > > 1 file changed, 4 insertions(+), 2 deletions(-) > > > > > > diff --git a/mm/madvise.c b/mm/madvise.c > > > index 638517952bd2..08c11217025a 100644 > > > --- a/mm/madvise.c > > > +++ b/mm/madvise.c > > > @@ -1168,13 +1168,15 @@ madvise_behavior_valid(int behavior) > > > } > > > > > > static bool > > > -process_madvise_behavior_valid(int behavior) > > > +process_madvise_behavior_valid(int behavior, struct task_struct *task) > > > { > > > switch (behavior) { > > > case MADV_COLD: > > > case MADV_PAGEOUT: > > > case MADV_WILLNEED: > > > return true; > > > + case MADV_COLLAPSE: > > > + return task == current || capable(CAP_SYS_ADMIN); > > > default: > > > return false; > > > } > > > @@ -1452,7 +1454,7 @@ SYSCALL_DEFINE5(process_madvise, int, pidfd, const struct iovec __user *, vec, > > > goto free_iov; > > > } > > > > > > - if (!process_madvise_behavior_valid(behavior)) { > > > + if (!process_madvise_behavior_valid(behavior, task)) { > > > ret = -EINVAL; > > > goto release_task; > > > } >