From patchwork Sun Jul 31 15:52:23 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: haoxin X-Patchwork-Id: 12933127 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 D1D20C00140 for ; Sun, 31 Jul 2022 15:52:36 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id 156A68E0002; Sun, 31 Jul 2022 11:52:36 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id 108968E0001; Sun, 31 Jul 2022 11:52:36 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id F369A8E0002; Sun, 31 Jul 2022 11:52:35 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from relay.hostedemail.com (smtprelay0016.hostedemail.com [216.40.44.16]) by kanga.kvack.org (Postfix) with ESMTP id E1C298E0001 for ; Sun, 31 Jul 2022 11:52:35 -0400 (EDT) Received: from smtpin21.hostedemail.com (a10.router.float.18 [10.200.18.1]) by unirelay03.hostedemail.com (Postfix) with ESMTP id AFE04A071B for ; Sun, 31 Jul 2022 15:52:35 +0000 (UTC) X-FDA: 79747837470.21.A890D4C Received: from out30-132.freemail.mail.aliyun.com (out30-132.freemail.mail.aliyun.com [115.124.30.132]) by imf27.hostedemail.com (Postfix) with ESMTP id 4143F40054 for ; Sun, 31 Jul 2022 15:52:33 +0000 (UTC) X-Alimail-AntiSpam: AC=PASS;BC=-1|-1;BR=01201311R711e4;CH=green;DM=||false|;DS=||;FP=0|-1|-1|-1|0|-1|-1|-1;HT=ay29a033018046051;MF=xhao@linux.alibaba.com;NM=1;PH=DS;RN=7;SR=0;TI=SMTPD_---0VKwpIWE_1659282745; Received: from localhost.localdomain(mailfrom:xhao@linux.alibaba.com fp:SMTPD_---0VKwpIWE_1659282745) by smtp.aliyun-inc.com; Sun, 31 Jul 2022 23:52:26 +0800 From: Xin Hao To: willy@infradead.org Cc: akpm@linux-foundation.org, adobriyan@gmail.com, keescook@chromium.org, xhao@linux.alibaba.com, linux-kernel@vger.kernel.org, linux-mm@kvack.org Subject: [RFC V2 PATCH] mm: add last level page table numa info to /proc/pid/numa_pgtable Date: Sun, 31 Jul 2022 23:52:23 +0800 Message-Id: <20220731155223.60238-1-xhao@linux.alibaba.com> X-Mailer: git-send-email 2.31.0 MIME-Version: 1.0 ARC-Authentication-Results: i=1; imf27.hostedemail.com; dkim=none; spf=pass (imf27.hostedemail.com: domain of xhao@linux.alibaba.com designates 115.124.30.132 as permitted sender) smtp.mailfrom=xhao@linux.alibaba.com; dmarc=pass (policy=none) header.from=alibaba.com ARC-Seal: i=1; s=arc-20220608; d=hostedemail.com; t=1659282755; a=rsa-sha256; cv=none; b=GzEh6IjQ9EozDw7GjvQDd0ikp4V/1Mzi1AMlobyl5AmsPqwmcv37Hu5Jphjql4TscJ3w/b 2PWqC63ucawR6BTFNZSU6LrIjiyd1bo/mtvvhuQt9dLE8eRb/AwzgeFQSnyxcggsXP/vyz bUBONH6NtD9azoTjf31ClJZOGMFSfpc= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=hostedemail.com; s=arc-20220608; t=1659282755; h=from:from:sender:reply-to:subject:subject:date:date: message-id:message-id:to:to:cc:cc:mime-version:mime-version: content-type:content-transfer-encoding:content-transfer-encoding: in-reply-to:references; bh=PPD3x23j3oJhqIX20kwm3s3Bd9kA19f/iyhYJAaKhFk=; b=xJQtjZBPev88YozEyiaRvmYE2tQuFFkZNEZhOWAvKc6AVvL+dah4yZGSPv9qziZxEYUhPw kiheDcRDNrgaPyVH3y8NP9RSGx8UGw195ZpDrEnZwCpK9Jz0+/BENSxQ0sCNhQBOn3aOXP lLQSOew6fIhb1bCnTigWDKHvzKXwj04= X-Rspamd-Queue-Id: 4143F40054 Authentication-Results: imf27.hostedemail.com; dkim=none; spf=pass (imf27.hostedemail.com: domain of xhao@linux.alibaba.com designates 115.124.30.132 as permitted sender) smtp.mailfrom=xhao@linux.alibaba.com; dmarc=pass (policy=none) header.from=alibaba.com X-Rspamd-Server: rspam01 X-Rspam-User: X-Stat-Signature: usgg5kqr8to5q7q1bz517i84wcquxhca X-HE-Tag: 1659282753-258693 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: In many data center servers, the shared memory architectures is Non-Uniform Memory Access (NUMA), remote numa node data access often brings a high latency problem, but what we are easy to ignore is that the page table remote numa access, It can also leads to a performance degradation. So there add a new interface in /proc, This will help developers to get more info about performance issues if they are caused by cross-NUMA. V1 -> V2 1, As Matthew Wilcox advise, Simplify the codes. 2, Do some codes format fix. V1: https://lore.kernel.org/linux-mm/YuVqdcY8Ibib2LJa@casper.infradead.org/T/ Signed-off-by: Xin Hao --- fs/proc/base.c | 2 ++ fs/proc/internal.h | 1 + fs/proc/task_mmu.c | 87 ++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 90 insertions(+) -- 2.31.0 diff --git a/fs/proc/base.c b/fs/proc/base.c index 8dfa36a99c74..487e82dd3275 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -3224,6 +3224,7 @@ static const struct pid_entry tgid_base_stuff[] = { REG("maps", S_IRUGO, proc_pid_maps_operations), #ifdef CONFIG_NUMA REG("numa_maps", S_IRUGO, proc_pid_numa_maps_operations), + REG("numa_pgtable", S_IRUGO, proc_pid_numa_pgtable_operations), #endif REG("mem", S_IRUSR|S_IWUSR, proc_mem_operations), LNK("cwd", proc_cwd_link), @@ -3571,6 +3572,7 @@ static const struct pid_entry tid_base_stuff[] = { #endif #ifdef CONFIG_NUMA REG("numa_maps", S_IRUGO, proc_pid_numa_maps_operations), + REG("numa_pgtable", S_IRUGO, proc_pid_numa_pgtable_operations), #endif REG("mem", S_IRUSR|S_IWUSR, proc_mem_operations), LNK("cwd", proc_cwd_link), diff --git a/fs/proc/internal.h b/fs/proc/internal.h index 06a80f78433d..e7ed9ef097b6 100644 --- a/fs/proc/internal.h +++ b/fs/proc/internal.h @@ -296,6 +296,7 @@ struct mm_struct *proc_mem_open(struct inode *inode, unsigned int mode); extern const struct file_operations proc_pid_maps_operations; extern const struct file_operations proc_pid_numa_maps_operations; +extern const struct file_operations proc_pid_numa_pgtable_operations; extern const struct file_operations proc_pid_smaps_operations; extern const struct file_operations proc_pid_smaps_rollup_operations; extern const struct file_operations proc_clear_refs_operations; diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index 2d04e3470d4c..967e46f523a0 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -1999,4 +1999,91 @@ const struct file_operations proc_pid_numa_maps_operations = { .release = proc_map_release, }; +struct pgtable_numa_private { + struct proc_maps_private proc_maps; + unsigned long node[MAX_NUMNODES]; +}; + +static int gather_pgtable_numa_stats(pmd_t *pmd, unsigned long addr, + unsigned long end, struct mm_walk *walk) +{ + struct pgtable_numa_private *priv = walk->private; + struct page *page; + int nid; + + if (pmd_huge(*pmd)) { + page = virt_to_page(pmd); + } else { + page = pmd_page(*pmd); + } + + nid = page_to_nid(page); + priv->node[nid]++; + + return 0; +} + +static const struct mm_walk_ops show_numa_pgtable_ops = { + .pmd_entry = gather_pgtable_numa_stats, +}; + +/* + * Display the page talbe allocated per node via /proc. + */ +static int show_numa_pgtable(struct seq_file *m, void *v) +{ + struct pgtable_numa_private *numa_priv = m->private; + struct vm_area_struct *vma = v; + struct mm_struct *mm = vma->vm_mm; + struct file *file = vma->vm_file; + int nid; + + if (!mm) + return 0; + + memset(numa_priv->node, 0, MAX_NUMNODES); + + seq_printf(m, "%08lx ", vma->vm_start); + + if (file) { + seq_puts(m, " file="); + seq_file_path(m, file, "\n\t= "); + } else if (vma->vm_start <= mm->brk && vma->vm_end >= mm->start_brk) { + seq_puts(m, " heap"); + } else if (is_stack(vma)) { + seq_puts(m, " stack"); + } + + /* mmap_lock is held by m_start */ + walk_page_vma(vma, &show_numa_pgtable_ops, numa_priv); + + for_each_node_state(nid, N_MEMORY) { + if (numa_priv->node[nid]) + seq_printf(m, " N%d=%lu", nid, numa_priv->node[nid]); + } + seq_putc(m, '\n'); + + return 0; +} + +static const struct seq_operations proc_pid_numa_pgtable_op = { + .start = m_start, + .next = m_next, + .stop = m_stop, + .show = show_numa_pgtable, +}; + +static int pid_numa_pgtable_open(struct inode *inode, struct file *file) +{ + return proc_maps_open(inode, file, &proc_pid_numa_pgtable_op, + sizeof(struct pgtable_numa_private)); +} + +const struct file_operations proc_pid_numa_pgtable_operations = { + .open = pid_numa_pgtable_open, + .read = seq_read, + .llseek = seq_lseek, + .release = proc_map_release, +}; + #endif /* CONFIG_NUMA */