From patchwork Tue Sep 24 17:15:18 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Johannes Weiner X-Patchwork-Id: 11159271 Return-Path: Received: from mail.kernel.org (pdx-korg-mail-1.web.codeaurora.org [172.30.200.123]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id C2F1B1599 for ; Tue, 24 Sep 2019 17:15:23 +0000 (UTC) Received: from kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by mail.kernel.org (Postfix) with ESMTP id 7C0852146E for ; Tue, 24 Sep 2019 17:15:23 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=fail reason="signature verification failed" (2048-bit key) header.d=cmpxchg-org.20150623.gappssmtp.com header.i=@cmpxchg-org.20150623.gappssmtp.com header.b="qgN8j5wh" DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 7C0852146E Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=cmpxchg.org Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=owner-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix) id 67CFC6B0008; Tue, 24 Sep 2019 13:15:22 -0400 (EDT) Delivered-To: linux-mm-outgoing@kvack.org Received: by kanga.kvack.org (Postfix, from userid 40) id 5DD6D6B000C; Tue, 24 Sep 2019 13:15:22 -0400 (EDT) X-Original-To: int-list-linux-mm@kvack.org X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 4A55B6B000D; Tue, 24 Sep 2019 13:15:22 -0400 (EDT) X-Original-To: linux-mm@kvack.org X-Delivered-To: linux-mm@kvack.org Received: from forelay.hostedemail.com (smtprelay0051.hostedemail.com [216.40.44.51]) by kanga.kvack.org (Postfix) with ESMTP id 1E1726B0008 for ; Tue, 24 Sep 2019 13:15:22 -0400 (EDT) Received: from smtpin14.hostedemail.com (10.5.19.251.rfc1918.com [10.5.19.251]) by forelay04.hostedemail.com (Postfix) with SMTP id BE5EA81E1 for ; Tue, 24 Sep 2019 17:15:21 +0000 (UTC) X-FDA: 75970465242.14.trip63_2eb5d1ad95a23 X-Spam-Summary: 2,0,0,27f9039e6bd3fd04,d41d8cd98f00b204,hannes@cmpxchg.org,:akpm@linux-foundation.org:josef@toxicpanda.com::linux-kernel@vger.kernel.org,RULES_HIT:2:41:69:355:379:541:800:960:966:973:988:989:1260:1311:1314:1345:1437:1515:1535:1605:1606:1730:1747:1777:1792:2194:2196:2199:2200:2393:2553:2559:2562:2689:3138:3139:3140:3141:3142:3308:3865:3866:3867:3868:3870:3871:3872:3874:4118:4250:4321:4385:5007:6261:6653:7576:8957:10004:10226:11026:11473:11658:11914:12043:12114:12291:12296:12297:12438:12517:12519:12555:12679:12683:12895:12986:13161:13229:13894:14096:14394:21080:21212:21324:21444:21451:21627:30045:30054:30064:30070:30090,0,RBL:209.85.222.196:@cmpxchg.org:.lbl8.mailshell.net-62.14.0.100 66.201.201.201,CacheIP:none,Bayesian:0.5,0.5,0.5,Netcheck:none,DomainCache:0,MSF:not bulk,SPF:fp,MSBL:0,DNSBL:neutral,Custom_rules:0:0:0,LFtime:25,LUA_SUMMARY:none X-HE-Tag: trip63_2eb5d1ad95a23 X-Filterd-Recvd-Size: 7806 Received: from mail-qk1-f196.google.com (mail-qk1-f196.google.com [209.85.222.196]) by imf31.hostedemail.com (Postfix) with ESMTP for ; Tue, 24 Sep 2019 17:15:20 +0000 (UTC) Received: by mail-qk1-f196.google.com with SMTP id h126so2527110qke.10 for ; Tue, 24 Sep 2019 10:15:20 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=cmpxchg-org.20150623.gappssmtp.com; s=20150623; h=from:to:cc:subject:date:message-id:mime-version :content-transfer-encoding; bh=mVvT+zKf/9SBbC4bQyrI1w+/sdRkQQsfH03mgMh0xIQ=; b=qgN8j5wh4ZY3/4P7ncwh8L4fitk036GmpCfIAE3FYJHqfWCclrEDPw5gwn6VBp1D+9 VLis0Fc34AJXorm/PAha5x2FeqoOxq2k3nydriTtsNPhlRdI2gAJcGxzPGNoCvCM6lxG ufwjNum6f5uRG0vcaKPgGQANbM33Kkz+zh7ukB1XJxml5l/1lyJcn+zml8S5wEtXA5Xr jbPwhYEiyEOzjRHXoPW41IwAnl/DPdAT9dGs5iA1oIeLzYq+z3NS7tp+uoWIhuMgl1eY frsYXFC23zuIqXA52BrS3oiRc2NUB13mVRwW8UBkEVHQPxLnn6FYhfis6EUfu8Ra+FrF +yHg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:cc:subject:date:message-id:mime-version :content-transfer-encoding; bh=mVvT+zKf/9SBbC4bQyrI1w+/sdRkQQsfH03mgMh0xIQ=; b=QHNkEMXxo+OHa9rxfhp/ATUv3bOJLPDUvnrDwlInXl68FYswK/Ux5Zzg4nA+QzAZgy 990uTZIlYtsAvpEf2m24ZwGLhNrGfrRecLgx88tIreIJ0Lop6twJxc4bJU3xJqRg4LFL ly/s4J0sf3AYH/ivWb8cMGGrlFXTD66U6oCZ0rfMgGZZT7Se2L6R1DGI/t3+yBmEwXvT GQ8XeLmXc7lDaTYSmwfm71MbHWPta+kNWOyze0gfiOdjr3BAilHoMFDoSGa02EdeHlvH j/3aWnXhDD1x+uQPky2zlENhIsKG/Ryh4UO7M2h29kzCZH8KowWQlHSsDE/Q04xDoy5n uqEg== X-Gm-Message-State: APjAAAX9AuaOF8eClu4W3XlmX4g8N/gPpa28irtneSYqyHILKh/r8AnO HjJVpeg4gv3EYt2cjmL9UlU2lg== X-Google-Smtp-Source: APXvYqwt7+cEOOnoNq+IeVwbFeMvXPiYOy2qk8m4i5VPe0Mjp6mV1JJrbC2SX0NZDlpOzE+QuU7bMw== X-Received: by 2002:a37:a00d:: with SMTP id j13mr3679827qke.2.1569345320080; Tue, 24 Sep 2019 10:15:20 -0700 (PDT) Received: from localhost ([2620:10d:c091:500::a51]) by smtp.gmail.com with ESMTPSA id g31sm1995914qte.78.2019.09.24.10.15.19 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Tue, 24 Sep 2019 10:15:19 -0700 (PDT) From: Johannes Weiner To: Andrew Morton Cc: Josef Bacik , linux-mm@kvack.org, linux-kernel@vger.kernel.org Subject: [PATCH] mm: drop mmap_sem before calling balance_dirty_pages() in write fault Date: Tue, 24 Sep 2019 13:15:18 -0400 Message-Id: <20190924171518.26682-1-hannes@cmpxchg.org> X-Mailer: git-send-email 2.23.0 MIME-Version: 1.0 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: From: Johannes Weiner One of our services is observing hanging ps/top/etc under heavy write IO, and the task states show this is an mmap_sem priority inversion: A write fault is holding the mmap_sem in read-mode and waiting for (heavily cgroup-limited) IO in balance_dirty_pages(): [<0>] balance_dirty_pages+0x724/0x905 [<0>] balance_dirty_pages_ratelimited+0x254/0x390 [<0>] fault_dirty_shared_page.isra.96+0x4a/0x90 [<0>] do_wp_page+0x33e/0x400 [<0>] __handle_mm_fault+0x6f0/0xfa0 [<0>] handle_mm_fault+0xe4/0x200 [<0>] __do_page_fault+0x22b/0x4a0 [<0>] page_fault+0x45/0x50 [<0>] 0xffffffffffffffff Somebody tries to change the address space, contending for the mmap_sem in write-mode: [<0>] call_rwsem_down_write_failed_killable+0x13/0x20 [<0>] do_mprotect_pkey+0xa8/0x330 [<0>] SyS_mprotect+0xf/0x20 [<0>] do_syscall_64+0x5b/0x100 [<0>] entry_SYSCALL_64_after_hwframe+0x3d/0xa2 [<0>] 0xffffffffffffffff The waiting writer locks out all subsequent readers to avoid lock starvation, and several threads can be seen hanging like this: [<0>] call_rwsem_down_read_failed+0x14/0x30 [<0>] proc_pid_cmdline_read+0xa0/0x480 [<0>] __vfs_read+0x23/0x140 [<0>] vfs_read+0x87/0x130 [<0>] SyS_read+0x42/0x90 [<0>] do_syscall_64+0x5b/0x100 [<0>] entry_SYSCALL_64_after_hwframe+0x3d/0xa2 [<0>] 0xffffffffffffffff To fix this, do what we do for cache read faults already: drop the mmap_sem before calling into anything IO bound, in this case the balance_dirty_pages() function, and return VM_FAULT_RETRY. Signed-off-by: Johannes Weiner Signed-off-by: Johannes Weiner Reviewed-by: Matthew Wilcox (Oracle) Acked-by: Kirill A. Shutemov Signed-off-by: Kirill A. Shutemov Reported-by: syzbot+03ee87124ee05af991bd@syzkaller.appspotmail.com Reviewed-by: Matthew Wilcox (Oracle) Acked-by: Johannes Weiner Signed-off-by: Kirill A. Shutemov Reported-by: syzbot+03ee87124ee05af991bd@syzkaller.appspotmail.com Acked-by: Johannes Weiner Reviewed-by: Matthew Wilcox (Oracle) --- mm/memory.c | 53 ++++++++++++++++++++++++++++++++++++++++------------- 1 file changed, 40 insertions(+), 13 deletions(-) diff --git a/mm/memory.c b/mm/memory.c index 2e796372927f..da5eb1d67447 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -2221,12 +2221,14 @@ static vm_fault_t do_page_mkwrite(struct vm_fault *vmf) * * The function expects the page to be locked and unlocks it. */ -static void fault_dirty_shared_page(struct vm_area_struct *vma, - struct page *page) +static int fault_dirty_shared_page(struct vm_fault *vmf) { + struct vm_area_struct *vma = vmf->vma; struct address_space *mapping; + struct page *page = vmf->page; bool dirtied; bool page_mkwrite = vma->vm_ops && vma->vm_ops->page_mkwrite; + int ret = 0; dirtied = set_page_dirty(page); VM_BUG_ON_PAGE(PageAnon(page), page); @@ -2239,16 +2241,36 @@ static void fault_dirty_shared_page(struct vm_area_struct *vma, mapping = page_rmapping(page); unlock_page(page); + if (!page_mkwrite) + file_update_time(vma->vm_file); + + /* + * Throttle page dirtying rate down to writeback speed. + * + * mapping may be NULL here because some device drivers do not + * set page.mapping but still dirty their pages + * + * Drop the mmap_sem before waiting on IO, if we can. The file + * is pinning the mapping, as per above. + */ if ((dirtied || page_mkwrite) && mapping) { - /* - * Some device drivers do not set page.mapping - * but still dirty their pages - */ + struct file *fpin = NULL; + + if ((vmf->flags & + (FAULT_FLAG_ALLOW_RETRY | FAULT_FLAG_RETRY_NOWAIT)) == + FAULT_FLAG_ALLOW_RETRY) { + fpin = get_file(vma->vm_file); + up_read(&vma->vm_mm->mmap_sem); + ret = VM_FAULT_RETRY; + } + balance_dirty_pages_ratelimited(mapping); + + if (fpin) + fput(fpin); } - if (!page_mkwrite) - file_update_time(vma->vm_file); + return ret; } /* @@ -2491,6 +2513,7 @@ static vm_fault_t wp_page_shared(struct vm_fault *vmf) __releases(vmf->ptl) { struct vm_area_struct *vma = vmf->vma; + int ret = VM_FAULT_WRITE; get_page(vmf->page); @@ -2514,10 +2537,10 @@ static vm_fault_t wp_page_shared(struct vm_fault *vmf) wp_page_reuse(vmf); lock_page(vmf->page); } - fault_dirty_shared_page(vma, vmf->page); + ret |= fault_dirty_shared_page(vmf); put_page(vmf->page); - return VM_FAULT_WRITE; + return ret; } /* @@ -3561,7 +3584,7 @@ static vm_fault_t do_shared_fault(struct vm_fault *vmf) return ret; } - fault_dirty_shared_page(vma, vmf->page); + ret |= fault_dirty_shared_page(vmf); return ret; } @@ -3576,7 +3599,6 @@ static vm_fault_t do_shared_fault(struct vm_fault *vmf) static vm_fault_t do_fault(struct vm_fault *vmf) { struct vm_area_struct *vma = vmf->vma; - struct mm_struct *vm_mm = vma->vm_mm; vm_fault_t ret; /* @@ -3617,7 +3639,12 @@ static vm_fault_t do_fault(struct vm_fault *vmf) /* preallocated pagetable is unused: free it */ if (vmf->prealloc_pte) { - pte_free(vm_mm, vmf->prealloc_pte); + /* + * XXX: Accessing vma->vm_mm now is not safe. The page + * fault handler may have dropped the mmap_sem a long + * time ago. Only s390 derefs that parameter. + */ + pte_free(vma->vm_mm, vmf->prealloc_pte); vmf->prealloc_pte = NULL; } return ret;