From patchwork Thu May 12 07:03:47 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: CGEL X-Patchwork-Id: 12847222 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 1EB3EC433FE for ; Thu, 12 May 2022 07:04:02 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id 89B9E6B0075; Thu, 12 May 2022 03:04:02 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id 84A146B0078; Thu, 12 May 2022 03:04:02 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 6C5228D0001; Thu, 12 May 2022 03:04:02 -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 5B47C6B0075 for ; Thu, 12 May 2022 03:04:02 -0400 (EDT) Received: from smtpin23.hostedemail.com (a10.router.float.18 [10.200.18.1]) by unirelay07.hostedemail.com (Postfix) with ESMTP id 2EB4220AED for ; Thu, 12 May 2022 07:04:02 +0000 (UTC) X-FDA: 79456201524.23.4B2F49D Received: from mail-pj1-f54.google.com (mail-pj1-f54.google.com [209.85.216.54]) by imf09.hostedemail.com (Postfix) with ESMTP id 7BD6314009A for ; Thu, 12 May 2022 07:03:53 +0000 (UTC) Received: by mail-pj1-f54.google.com with SMTP id l11-20020a17090a49cb00b001d923a9ca99so4090035pjm.1 for ; Thu, 12 May 2022 00:04:01 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20210112; h=from:to:cc:subject:date:message-id:in-reply-to:references :mime-version:content-transfer-encoding; bh=YaKkDgE/meZO4FIXQmv4E0z5fb3+Tw8MQ7HlMv6R3Uw=; b=D+qcVV3vBYHAHMkb5IgxSk4d3mGwU9VrnKBDKgTI5G6PcFDs252lNzghbMO3yzxsbe 3DiuGR5X8lE/KNw1+7gAGgHgKZ80zb2JFlp44P9u8J4XEOa0ZAeyi6zRTSWC79tRTyeL FEgqLuheH6nPzKpjFEwvxq5hGmqgdLHef/Sa+lDplsQ/VNT+V7CpYzWHMjTPdqrytNVH CALzURkuV0vNWu8Gz1ZW+tCaEaZcRQOzPeOvHsnDINaEIwSF6BjZzF6q+fgt3gBzbPEk rIskNjXqAuzAJ05Z79Izzv82378M2yqoh3vp6LpRNLWaH69SG5GlwjA/PNYKjn7P+TP0 Qhow== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references:mime-version:content-transfer-encoding; bh=YaKkDgE/meZO4FIXQmv4E0z5fb3+Tw8MQ7HlMv6R3Uw=; b=GDYAGX6QgcYtHiAJcwPU3i31wXoB2Z6ekpqwVMYWAmdQUVC5dthCnc7cy+GA8Zwzli mzqnGRnwWFYXv/KZXtqbekq0JX4bt1ZmoawpUtmEz63XCc6beFDahr5lGOMUWHePDbW6 +L2fpjC7cUKWVXIXVdb4lNO91M+k9cjiEc5Zl7FsLJt2tfag3/1UB8ytYl5+pBCXFI6l BxSlPysaFNlcVNWu8JBPa4ETxBS7GpBixz+4uEehivDkhPb/+D4w1OPSypPGXelD5G+o oKWNMDzlavaS68pA5YWej3MMaCoQVY6fXRcGjDFZqIxBKsz6ezJ8Vs/bH4j05Qn97fDv UO7Q== X-Gm-Message-State: AOAM533Qporu4TZ01o2rS6WX3hLjbISaqUxG/0uxspLf9W2xG5qEvSrr hD+6OW/svGWUIdsmAa3a93Q= X-Google-Smtp-Source: ABdhPJxZPGWCtuQRHvZcbRylfZCmCp6Zorucdirkh9XB4hpZZE51CVQZ3DlYsnaL40sQ+wZBUiuC3w== X-Received: by 2002:a17:902:c94f:b0:15e:b542:3ef6 with SMTP id i15-20020a170902c94f00b0015eb5423ef6mr29122883pla.55.1652339040473; Thu, 12 May 2022 00:04:00 -0700 (PDT) Received: from localhost.localdomain ([193.203.214.57]) by smtp.gmail.com with ESMTPSA id c7-20020a655a87000000b003db6f4a96c4sm567993pgt.32.2022.05.12.00.03.57 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Thu, 12 May 2022 00:03:59 -0700 (PDT) From: cgel.zte@gmail.com X-Google-Original-From: xu.xin16@zte.com.cn To: ammarfaizi2@gnuweeb.org, akpm@linux-foundation.org Cc: cgel.zte@gmail.com, linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org, ran.xiaokai@zte.com.cn, wang.yong12@zte.com.cn, willy@infradead.org, xu.xin16@zte.com.cn, yang.yang29@zte.com.cn, zhang.yunkai@zte.com.cn Subject: [PATCH v7] mm/ksm: introduce ksm_force for each process Date: Thu, 12 May 2022 07:03:47 +0000 Message-Id: <20220512070347.1628163-1-xu.xin16@zte.com.cn> X-Mailer: git-send-email 2.25.1 In-Reply-To: References: MIME-Version: 1.0 X-Stat-Signature: gqc8zz9q66qy1gznri8yh4nt1hu75jqg X-Rspamd-Server: rspam07 X-Rspamd-Queue-Id: 7BD6314009A X-Rspam-User: Authentication-Results: imf09.hostedemail.com; dkim=pass header.d=gmail.com header.s=20210112 header.b=D+qcVV3v; dmarc=pass (policy=none) header.from=gmail.com; spf=pass (imf09.hostedemail.com: domain of cgel.zte@gmail.com designates 209.85.216.54 as permitted sender) smtp.mailfrom=cgel.zte@gmail.com X-HE-Tag: 1652339033-845026 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: xu xin To use KSM, we have to explicitly call madvise() in application code, which means installed apps on OS needs to be uninstall and source code needs to be modified. It is inconvenient. In order to change this situation, We add a new proc file ksm_force under /proc// to support turning on/off KSM scanning of a process's mm dynamically. If ksm_force is set to 1, force all anonymous and 'qualified' VMAs of this mm to be involved in KSM scanning without explicitly calling madvise to mark VMA as MADV_MERGEABLE. But It is effective only when the klob of /sys/kernel/mm/ksm/run is set as 1. If ksm_force is set to 0, cancel the feature of ksm_force of this process and unmerge those merged pages belonging to VMAs which is not madvised as MADV_MERGEABLE of this process, but leave MADV_MERGEABLE areas merged. Signed-off-by: xu xin Reviewed-by: Yang Yang Reviewed-by: Ran Xiaokai Reviewed-by: wangyong Reviewed-by: Yunkai Zhang Suggested-by: Matthew Wilcox Suggested-by: Ammar Faizi --- v7: - remove over-zeroing in ksm_force_write() and using strncpy_from_user instead of copy_from_user. v6: - modify the way of "return" - remove unnecessary words in Documentation/admin-guide/mm/ksm.rst - add additional notes to "set 0 to ksm_force" in Documentation/../ksm.rst and Documentation/../proc.rst v5: - fix typos in Documentation/filesystem/proc.rst v4: - fix typos in commit log - add interface descriptions under Documentation/ v3: - fix compile error of mm/ksm.c v2: - fix a spelling error in commit log. - remove a redundant condition check in ksm_force_write(). --- Documentation/admin-guide/mm/ksm.rst | 19 +++++- Documentation/filesystems/proc.rst | 17 +++++ fs/proc/base.c | 97 ++++++++++++++++++++++++++++ include/linux/mm_types.h | 9 +++ mm/ksm.c | 32 ++++++++- 5 files changed, 171 insertions(+), 3 deletions(-) diff --git a/Documentation/admin-guide/mm/ksm.rst b/Documentation/admin-guide/mm/ksm.rst index b244f0202a03..8cabc2504005 100644 --- a/Documentation/admin-guide/mm/ksm.rst +++ b/Documentation/admin-guide/mm/ksm.rst @@ -32,7 +32,7 @@ are swapped back in: ksmd must rediscover their identity and merge again). Controlling KSM with madvise ============================ -KSM only operates on those areas of address space which an application +KSM can operates on those areas of address space which an application has advised to be likely candidates for merging, by using the madvise(2) system call:: @@ -70,6 +70,23 @@ Applications should be considerate in their use of MADV_MERGEABLE, restricting its use to areas likely to benefit. KSM's scans may use a lot of processing power: some installations will disable KSM for that reason. +Controlling KSM with procfs +=========================== + +KSM can also operate on anonymous areas of address space of those processes's +knob ``/proc//ksm_force`` is on, even if app codes doesn't call madvise() +explicitly to advise specific areas as MADV_MERGEABLE. + +You can set ksm_force to 1 to force all anonymous and qualified VMAs of +this process to be involved in KSM scanning. + e.g. ``echo 1 > /proc//ksm_force`` + +You can also set ksm_force to 0 to cancel that force feature of this process +and unmerge those merged pages which belongs to those VMAs not marked as +MADV_MERGEABLE of this process. But that still leave those pages belonging to +VMAs marked as MADV_MERGEABLE merged (fallback to the default state). + e.g. ``echo 0 > /proc//ksm_force`` + .. _ksm_sysfs: KSM daemon sysfs interface diff --git a/Documentation/filesystems/proc.rst b/Documentation/filesystems/proc.rst index 061744c436d9..8890b8b457a4 100644 --- a/Documentation/filesystems/proc.rst +++ b/Documentation/filesystems/proc.rst @@ -47,6 +47,7 @@ fixes/update part 1.1 Stefani Seibold June 9 2009 3.10 /proc//timerslack_ns - Task timerslack value 3.11 /proc//patch_state - Livepatch patch operation state 3.12 /proc//arch_status - Task architecture specific information + 3.13 /proc//ksm_force - Setting of mandatory involvement in KSM 4 Configuring procfs 4.1 Mount options @@ -2176,6 +2177,22 @@ AVX512_elapsed_ms the task is unlikely an AVX512 user, but depends on the workload and the scheduling scenario, it also could be a false negative mentioned above. +3.13 /proc//ksm_force - Setting of mandatory involvement in KSM +----------------------------------------------------------------------- +When CONFIG_KSM is enabled, this file can be used to specify if this +process's anonymous memory can be involved in KSM scanning without app codes +explicitly calling madvise to mark memory address as MADV_MERGEABLE. + +If writing 1 to this file, the kernel will force all anonymous and qualified +memory to be involved in KSM scanning without explicitly calling madvise to +mark memory address as MADV_MERGEABLE. But that is effective only when the +klob of '/sys/kernel/mm/ksm/run' is set as 1. + +If writing 0 to this file, the mandatory KSM feature of this process's will +be cancelled and unmerge those merged pages which belongs to those areas not +marked as MADV_MERGEABLE of this process, but leave those pages belonging to +areas marked as MADV_MERGEABLE merged (fallback to the default state). + Chapter 4: Configuring procfs ============================= diff --git a/fs/proc/base.c b/fs/proc/base.c index 8dfa36a99c74..99ab0e8cdcbc 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -96,6 +96,7 @@ #include #include #include +#include #include #include "internal.h" #include "fd.h" @@ -3168,6 +3169,100 @@ static int proc_pid_ksm_merging_pages(struct seq_file *m, struct pid_namespace * return 0; } + +static ssize_t ksm_force_read(struct file *file, char __user *buf, size_t count, + loff_t *ppos) +{ + struct task_struct *task; + struct mm_struct *mm; + char buffer[PROC_NUMBUF]; + ssize_t len; + int ret; + + task = get_proc_task(file_inode(file)); + if (!task) + return -ESRCH; + + mm = get_task_mm(task); + ret = 0; + if (mm) { + len = snprintf(buffer, sizeof(buffer), "%d\n", mm->ksm_force); + ret = simple_read_from_buffer(buf, count, ppos, buffer, len); + mmput(mm); + } + + return ret; +} + +static ssize_t ksm_force_write(struct file *file, const char __user *buf, + size_t count, loff_t *ppos) +{ + struct task_struct *task; + struct mm_struct *mm; + char buffer[PROC_NUMBUF]; + int force; + int err = 0; + int str_len; + + if (count > sizeof(buffer) - 1) { + count = sizeof(buffer) - 1; + } + + str_len = strncpy_from_user(buffer, buf, count); + if (str_len < 0) + return -EFAULT; + buffer[str_len] = '\0'; + + err = kstrtoint(strstrip(buffer), 0, &force); + if (err) + return err; + + if (force != 0 && force != 1) + return -EINVAL; + + task = get_proc_task(file_inode(file)); + if (!task) + return -ESRCH; + + mm = get_task_mm(task); + if (!mm) + goto out_put_task; + + if (mm->ksm_force != force) { + if (mmap_write_lock_killable(mm)) { + err = -EINTR; + goto out_mmput; + } + + if (force == 0) + mm->ksm_force = force; + else { + /* + * Force anonymous pages of this mm to be involved in KSM merging + * without explicitly calling madvise. + */ + if (!test_bit(MMF_VM_MERGEABLE, &mm->flags)) + err = __ksm_enter(mm); + if (!err) + mm->ksm_force = force; + } + + mmap_write_unlock(mm); + } + +out_mmput: + mmput(mm); +out_put_task: + put_task_struct(task); + + return err < 0 ? err : count; +} + +static const struct file_operations proc_pid_ksm_force_operations = { + .read = ksm_force_read, + .write = ksm_force_write, + .llseek = generic_file_llseek, +}; #endif /* CONFIG_KSM */ #ifdef CONFIG_STACKLEAK_METRICS @@ -3303,6 +3398,7 @@ static const struct pid_entry tgid_base_stuff[] = { #endif #ifdef CONFIG_KSM ONE("ksm_merging_pages", S_IRUSR, proc_pid_ksm_merging_pages), + REG("ksm_force", S_IRUSR|S_IWUSR, proc_pid_ksm_force_operations), #endif }; @@ -3639,6 +3735,7 @@ static const struct pid_entry tid_base_stuff[] = { #endif #ifdef CONFIG_KSM ONE("ksm_merging_pages", S_IRUSR, proc_pid_ksm_merging_pages), + REG("ksm_force", S_IRUSR|S_IWUSR, proc_pid_ksm_force_operations), #endif }; diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index b34ff2cdbc4f..1b1592c2f5cf 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -661,6 +661,15 @@ struct mm_struct { * merging. */ unsigned long ksm_merging_pages; + /* + * If true, force anonymous pages of this mm to be involved in KSM + * merging without explicitly calling madvise. It is effctive only + * when the klob of '/sys/kernel/mm/ksm/run' is set as 1. If false, + * cancel the feature of ksm_force of this process and unmerge + * those merged pages which is not madvised as MERGEABLE of this + * process, but leave MERGEABLE areas merged. + */ + bool ksm_force; #endif } __randomize_layout; diff --git a/mm/ksm.c b/mm/ksm.c index 38360285497a..c9f672dcc72e 100644 --- a/mm/ksm.c +++ b/mm/ksm.c @@ -334,6 +334,34 @@ static void __init ksm_slab_free(void) mm_slot_cache = NULL; } +/* Check if vma is qualified for ksmd scanning */ +static bool ksm_vma_check(struct vm_area_struct *vma) +{ + unsigned long vm_flags = vma->vm_flags; + + if (!(vma->vm_flags & VM_MERGEABLE) && !(vma->vm_mm->ksm_force)) + return false; + + if (vm_flags & (VM_SHARED | VM_MAYSHARE | + VM_PFNMAP | VM_IO | VM_DONTEXPAND | + VM_HUGETLB | VM_MIXEDMAP)) + return false; /* just ignore this vma*/ + + if (vma_is_dax(vma)) + return false; + +#ifdef VM_SAO + if (vm_flags & VM_SAO) + return false; +#endif +#ifdef VM_SPARC_ADI + if (vm_flags & VM_SPARC_ADI) + return false; +#endif + + return true; +} + static __always_inline bool is_stable_node_chain(struct stable_node *chain) { return chain->rmap_hlist_len == STABLE_NODE_CHAIN; @@ -523,7 +551,7 @@ static struct vm_area_struct *find_mergeable_vma(struct mm_struct *mm, if (ksm_test_exit(mm)) return NULL; vma = vma_lookup(mm, addr); - if (!vma || !(vma->vm_flags & VM_MERGEABLE) || !vma->anon_vma) + if (!vma || !ksm_vma_check(vma) || !vma->anon_vma) return NULL; return vma; } @@ -2297,7 +2325,7 @@ static struct rmap_item *scan_get_next_rmap_item(struct page **page) vma = find_vma(mm, ksm_scan.address); for (; vma; vma = vma->vm_next) { - if (!(vma->vm_flags & VM_MERGEABLE)) + if (!ksm_vma_check(vma)) continue; if (ksm_scan.address < vma->vm_start) ksm_scan.address = vma->vm_start;