From nobody Sun Sep 14 07:57:28 2025 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 vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id B4113C27C76 for ; Wed, 25 Jan 2023 15:31:14 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S236087AbjAYPbM (ORCPT ); Wed, 25 Jan 2023 10:31:12 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:53376 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S236079AbjAYPbG (ORCPT ); Wed, 25 Jan 2023 10:31:06 -0500 Received: from us-smtp-delivery-44.mimecast.com (us-smtp-delivery-44.mimecast.com [205.139.111.44]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id C4A3C59B75 for ; Wed, 25 Jan 2023 07:31:01 -0800 (PST) Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-595-ZpQQD_SlPyav-ZCRRyDwfg-1; Wed, 25 Jan 2023 10:29:27 -0500 X-MC-Unique: ZpQQD_SlPyav-ZCRRyDwfg-1 Received: from smtp.corp.redhat.com (int-mx04.intmail.prod.int.rdu2.redhat.com [10.11.54.4]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 2B93E811E9C; Wed, 25 Jan 2023 15:29:27 +0000 (UTC) Received: from comp-core-i7-2640m-0182e6.redhat.com (ovpn-208-16.brq.redhat.com [10.40.208.16]) by smtp.corp.redhat.com (Postfix) with ESMTP id 68BFD2026D4B; Wed, 25 Jan 2023 15:29:25 +0000 (UTC) From: Alexey Gladkov To: LKML , containers@lists.linux.dev, linux-fsdevel@vger.kernel.org Cc: Alexey Dobriyan , Al Viro , Andrew Morton , Christian Brauner , Val Cowan Subject: [RFC PATCH v1 1/6] proc: Fix separator for subset option Date: Wed, 25 Jan 2023 16:28:48 +0100 Message-Id: In-Reply-To: References: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.4 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" Signed-off-by: Alexey Gladkov --- fs/proc/root.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/proc/root.c b/fs/proc/root.c index 3c2ee3eb1138..5f1015b6418d 100644 --- a/fs/proc/root.c +++ b/fs/proc/root.c @@ -91,7 +91,7 @@ static int proc_parse_subset_param(struct fs_context *fc,= char *value) struct proc_fs_context *ctx =3D fc->fs_private; =20 while (value) { - char *ptr =3D strchr(value, ','); + char *ptr =3D strchr(value, '+'); =20 if (ptr !=3D NULL) *ptr++ =3D '\0'; --=20 2.33.6 From nobody Sun Sep 14 07:57:28 2025 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 vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 088FBC54EAA for ; Wed, 25 Jan 2023 15:30:04 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S236064AbjAYPaB (ORCPT ); Wed, 25 Jan 2023 10:30:01 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:51474 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S235873AbjAYP3u (ORCPT ); Wed, 25 Jan 2023 10:29:50 -0500 Received: from us-smtp-delivery-44.mimecast.com (us-smtp-delivery-44.mimecast.com [205.139.111.44]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id AD5C659245 for ; Wed, 25 Jan 2023 07:29:40 -0800 (PST) Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-22-3L7hQhr3NEibyln3yWV5BQ-1; Wed, 25 Jan 2023 10:29:30 -0500 X-MC-Unique: 3L7hQhr3NEibyln3yWV5BQ-1 Received: from smtp.corp.redhat.com (int-mx04.intmail.prod.int.rdu2.redhat.com [10.11.54.4]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 4FF52857A81; Wed, 25 Jan 2023 15:29:29 +0000 (UTC) Received: from comp-core-i7-2640m-0182e6.redhat.com (ovpn-208-16.brq.redhat.com [10.40.208.16]) by smtp.corp.redhat.com (Postfix) with ESMTP id 8003C2026D4B; Wed, 25 Jan 2023 15:29:27 +0000 (UTC) From: Alexey Gladkov To: LKML , containers@lists.linux.dev, linux-fsdevel@vger.kernel.org Cc: Alexey Dobriyan , Al Viro , Andrew Morton , Christian Brauner , Val Cowan Subject: [RFC PATCH v1 2/6] proc: Add allowlist to control access to procfs files Date: Wed, 25 Jan 2023 16:28:49 +0100 Message-Id: In-Reply-To: References: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.4 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" If, after creating a container and mounting procfs, the system configuration may change and new files may appear in procfs. Files including writable root or any other users. In most cases, it is known in advance which files in procfs the user needs in the container. It is much easier to control the list of what you want than to control the list of unwanted files. To do this, subset=3Dallowlist is added to control the visibility of static files in procfs (not process pids). After that, the control file /proc/allowlist appears in the root of the filesystem. This file contains a list of files and directories that will be visible in this vmountpoint. Immediately after mount, this file contains only one name - the name of the file itself. The admin can add names, read this file to get the current state of the allowlist. The file behaves almost like a regular file. Changes are applied when the file is closed. To prevent changes to allowlist, admin should remove its name from the list of allowed files. After this change, the file will disappear. Signed-off-by: Alexey Gladkov --- fs/proc/Kconfig | 10 ++ fs/proc/Makefile | 1 + fs/proc/generic.c | 15 ++- fs/proc/internal.h | 29 +++++ fs/proc/proc_allowlist.c | 221 +++++++++++++++++++++++++++++++++++++++ fs/proc/root.c | 27 ++++- include/linux/proc_fs.h | 8 ++ 7 files changed, 306 insertions(+), 5 deletions(-) create mode 100644 fs/proc/proc_allowlist.c diff --git a/fs/proc/Kconfig b/fs/proc/Kconfig index 32b1116ae137..bfe80b1fd31f 100644 --- a/fs/proc/Kconfig +++ b/fs/proc/Kconfig @@ -108,3 +108,13 @@ config PROC_PID_ARCH_STATUS config PROC_CPU_RESCTRL def_bool n depends on PROC_FS + +config PROC_ALLOW_LIST + bool "/proc/allowlist support" + depends on PROC_FS + default n + help + Provides a way to restrict access to certain files in procfs. Mounting + procfs with subset=3Dallowlist will add the file /proc/allowlist which + contains a list of files and directories that should be accessed. To + prevent the list from being changed, the file itself must be excluded. diff --git a/fs/proc/Makefile b/fs/proc/Makefile index bd08616ed8ba..3c7d3dacbd2f 100644 --- a/fs/proc/Makefile +++ b/fs/proc/Makefile @@ -34,3 +34,4 @@ proc-$(CONFIG_PROC_VMCORE) +=3D vmcore.o proc-$(CONFIG_PRINTK) +=3D kmsg.o proc-$(CONFIG_PROC_PAGE_MONITOR) +=3D page.o proc-$(CONFIG_BOOT_CONFIG) +=3D bootconfig.o +proc-$(CONFIG_PROC_ALLOW_LIST) +=3D proc_allowlist.o diff --git a/fs/proc/generic.c b/fs/proc/generic.c index 587b91d9d998..d4c8589987e7 100644 --- a/fs/proc/generic.c +++ b/fs/proc/generic.c @@ -43,7 +43,7 @@ void pde_free(struct proc_dir_entry *pde) kmem_cache_free(proc_dir_entry_cache, pde); } =20 -static int proc_match(const char *name, struct proc_dir_entry *de, unsigne= d int len) +int proc_match(const char *name, struct proc_dir_entry *de, unsigned int l= en) { if (len < de->namelen) return -1; @@ -251,6 +251,9 @@ struct dentry *proc_lookup_de(struct inode *dir, struct= dentry *dentry, if (de) { pde_get(de); read_unlock(&proc_subdir_lock); + if (!proc_pde_access_allowed(proc_sb_info(dir->i_sb), de)) { + return ERR_PTR(-ENOENT); + } inode =3D proc_get_inode(dir->i_sb, de); if (!inode) return ERR_PTR(-ENOMEM); @@ -266,7 +269,7 @@ struct dentry *proc_lookup(struct inode *dir, struct de= ntry *dentry, { struct proc_fs_info *fs_info =3D proc_sb_info(dir->i_sb); =20 - if (fs_info->pidonly =3D=3D PROC_PIDONLY_ON) + if (fs_info->pidonly =3D=3D PROC_PIDONLY_ON && !proc_has_allowlist(fs_inf= o)) return ERR_PTR(-ENOENT); =20 return proc_lookup_de(dir, dentry, PDE(dir)); @@ -284,6 +287,9 @@ struct dentry *proc_lookup(struct inode *dir, struct de= ntry *dentry, int proc_readdir_de(struct file *file, struct dir_context *ctx, struct proc_dir_entry *de) { + struct inode *inode =3D file_inode(file); + struct proc_fs_info *fs_info =3D proc_sb_info(inode->i_sb); + int i; =20 if (!dir_emit_dots(file, ctx)) @@ -307,7 +313,8 @@ int proc_readdir_de(struct file *file, struct dir_conte= xt *ctx, struct proc_dir_entry *next; pde_get(de); read_unlock(&proc_subdir_lock); - if (!dir_emit(ctx, de->name, de->namelen, + if (proc_pde_access_allowed(fs_info, de) && + !dir_emit(ctx, de->name, de->namelen, de->low_ino, de->mode >> 12)) { pde_put(de); return 0; @@ -327,7 +334,7 @@ int proc_readdir(struct file *file, struct dir_context = *ctx) struct inode *inode =3D file_inode(file); struct proc_fs_info *fs_info =3D proc_sb_info(inode->i_sb); =20 - if (fs_info->pidonly =3D=3D PROC_PIDONLY_ON) + if (fs_info->pidonly =3D=3D PROC_PIDONLY_ON && !proc_has_allowlist(fs_inf= o)) return 1; =20 return proc_readdir_de(file, ctx, PDE(inode)); diff --git a/fs/proc/internal.h b/fs/proc/internal.h index b701d0207edf..999d105f6f96 100644 --- a/fs/proc/internal.h +++ b/fs/proc/internal.h @@ -84,6 +84,16 @@ static inline void pde_make_permanent(struct proc_dir_en= try *pde) pde->flags |=3D PROC_ENTRY_PERMANENT; } =20 +static inline bool pde_is_allowlist(const struct proc_dir_entry *pde) +{ + return pde->flags & PROC_ENTRY_ALLOWLIST; +} + +static inline void pde_make_allowlist(struct proc_dir_entry *pde) +{ + pde->flags |=3D PROC_ENTRY_ALLOWLIST; +} + extern struct kmem_cache *proc_dir_entry_cache; void pde_free(struct proc_dir_entry *pde); =20 @@ -187,6 +197,7 @@ struct proc_dir_entry *proc_create_reg(const char *name= , umode_t mode, struct proc_dir_entry **parent, void *data); struct proc_dir_entry *proc_register(struct proc_dir_entry *dir, struct proc_dir_entry *dp); +extern int proc_match(const char *, struct proc_dir_entry *, unsigned int); extern struct dentry *proc_lookup(struct inode *, struct dentry *, unsigne= d int); struct dentry *proc_lookup_de(struct inode *, struct dentry *, struct proc= _dir_entry *); extern int proc_readdir(struct file *, struct dir_context *); @@ -318,3 +329,21 @@ static inline void pde_force_lookup(struct proc_dir_en= try *pde) /* /proc/net/ entries can be changed under us by setns(CLONE_NEWNET) */ pde->proc_dops =3D &proc_net_dentry_ops; } + +/* + * proc_allowlist.c + */ +#ifdef CONFIG_PROC_ALLOW_LIST +extern bool proc_has_allowlist(struct proc_fs_info *); +extern bool proc_pde_access_allowed(struct proc_fs_info *, struct proc_dir= _entry *); +#else +static inline bool proc_has_allowlist(struct proc_fs_info *fs_info) +{ + return false; +} + +static inline bool proc_pde_access_allowed(struct proc_fs_info *fs_info, s= truct proc_dir_entry *pde) +{ + return true; +} +#endif diff --git a/fs/proc/proc_allowlist.c b/fs/proc/proc_allowlist.c new file mode 100644 index 000000000000..b38e11b04199 --- /dev/null +++ b/fs/proc/proc_allowlist.c @@ -0,0 +1,221 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * linux/fs/proc/proc_allowlist.c + * + * Copyright (C) 2022 + * + * Author: Alexey Gladkov + */ +#include +#include +#include +#include +#include +#include "internal.h" + +#define FILE_SEQFILE(f) ((struct seq_file *)((f)->private_data)) +#define FILE_DATA(f) (FILE_SEQFILE(f)->private) + +bool proc_has_allowlist(struct proc_fs_info *fs_info) +{ + bool ret; + unsigned long flags; + + read_lock_irqsave(&fs_info->allowlist_lock, flags); + ret =3D (fs_info->allowlist =3D=3D NULL); + read_unlock_irqrestore(&fs_info->allowlist_lock, flags); + + return ret; +} + +bool proc_pde_access_allowed(struct proc_fs_info *fs_info, struct proc_dir= _entry *de) +{ + bool ret =3D false; + char *ptr; + unsigned long flags; + + read_lock_irqsave(&fs_info->allowlist_lock, flags); + + if (!fs_info->allowlist) { + read_unlock_irqrestore(&fs_info->allowlist_lock, flags); + + if (!pde_is_allowlist(de)) + ret =3D true; + + return ret; + } + + ptr =3D fs_info->allowlist; + + while (*ptr !=3D '\0') { + struct proc_dir_entry *pde; + char *sep, *end; + size_t len, pathlen; + + if (!(sep =3D strchr(ptr, '\n'))) + pathlen =3D strlen(ptr); + else + pathlen =3D (sep - ptr); + + if (!pathlen) + goto next; + + pde =3D de; + end =3D NULL; + len =3D pathlen; + + while (ptr !=3D end && len > 0) { + end =3D ptr + len - 1; + + while (1) { + if (*end =3D=3D '/') { + end++; + break; + } + if (end =3D=3D ptr) + break; + end--; + } + + if (proc_match(end, pde, ptr + len - end)) + goto next; + + len =3D end - ptr - 1; + pde =3D pde->parent; + } + + ret =3D true; + break; +next: + ptr +=3D pathlen + 1; + } + + read_unlock_irqrestore(&fs_info->allowlist_lock, flags); + + return ret; +} + +static int show_allowlist(struct seq_file *m, void *v) +{ + struct proc_fs_info *fs_info =3D proc_sb_info(m->file->f_inode->i_sb); + char *p =3D fs_info->allowlist; + unsigned long flags; + + read_lock_irqsave(&fs_info->allowlist_lock, flags); + if (p) + seq_puts(m, p); + read_unlock_irqrestore(&fs_info->allowlist_lock, flags); + + return 0; +} + +static int open_allowlist(struct inode *inode, struct file *file) +{ + struct proc_fs_info *fs_info =3D proc_sb_info(inode->i_sb); + int ret; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + // we need this because shrink_dcache_sb() can't drop our own dentry. + if (!proc_pde_access_allowed(fs_info, PDE(inode))) + return -ENOENT; + + // we want a null-terminated string so not all 128K are available. + ret =3D single_open_size(file, show_allowlist, NULL, SZ_128K); + + if (!ret && (file->f_mode & FMODE_WRITE) && + (file->f_flags & O_APPEND) && !(file->f_flags & O_TRUNC)) + show_allowlist(FILE_SEQFILE(file), NULL); + + return ret; +} + +static ssize_t write_allowlist(struct file *file, const char __user *buffe= r, size_t count, loff_t *pos) +{ + struct seq_file *seq_file =3D FILE_SEQFILE(file); + ssize_t ret; + ssize_t n =3D count; + const char *ptr =3D buffer; + + if ((seq_file->count + count) >=3D (seq_file->size - 1)) + return -EFBIG; + + while (n > 0) { + char chunk[SZ_256]; + loff_t chkpos =3D 0; + ssize_t i, len; + + len =3D simple_write_to_buffer(chunk, sizeof(chunk), &chkpos, ptr, n); + if (len < 0) + return len; + + for (i =3D 0; i < len; i++) { + if (!isprint(chunk[i]) && chunk[i] !=3D '\n') + return -EINVAL; + } + + ret =3D seq_write(seq_file, chunk, len); + if (ret < 0) + return -EINVAL; + + ptr +=3D len; + n -=3D len; + } + + if (pos) + *pos +=3D count; + + return count; +} + +static int close_allowlist(struct inode *inode, struct file *file) +{ + struct seq_file *seq_file =3D FILE_SEQFILE(file); + struct proc_fs_info *fs_info =3D proc_sb_info(inode->i_sb); + + if (seq_file->buf && (file->f_mode & FMODE_WRITE)) { + char *buf; + + if (!seq_get_buf(seq_file, &buf)) + return -EIO; + *buf =3D '\0'; + + if (strcmp(seq_file->buf, fs_info->allowlist)) { + unsigned long flags; + + buf =3D kstrndup(seq_file->buf, seq_file->count, GFP_KERNEL_ACCOUNT); + if (!buf) + return -EIO; + + write_lock_irqsave(&fs_info->allowlist_lock, flags); + + shrink_dcache_sb(inode->i_sb); + + kfree(fs_info->allowlist); + fs_info->allowlist =3D buf; + + write_unlock_irqrestore(&fs_info->allowlist_lock, flags); + } + } + + return single_release(inode, file); +} + +static const struct proc_ops proc_allowlist_ops =3D { + .proc_open =3D open_allowlist, + .proc_read =3D seq_read, + .proc_lseek =3D seq_lseek, + .proc_write =3D write_allowlist, + .proc_release =3D close_allowlist, +}; + +static int __init proc_allowlist_init(void) +{ + struct proc_dir_entry *pde; + pde =3D proc_create("allowlist", S_IRUSR | S_IWUSR, NULL, &proc_allowlist= _ops); + pde_make_permanent(pde); + pde_make_allowlist(pde); + return 0; +} +fs_initcall(proc_allowlist_init); diff --git a/fs/proc/root.c b/fs/proc/root.c index 5f1015b6418d..1564f5cd118d 100644 --- a/fs/proc/root.c +++ b/fs/proc/root.c @@ -32,6 +32,7 @@ struct proc_fs_context { enum proc_hidepid hidepid; int gid; enum proc_pidonly pidonly; + enum proc_allowlist allowlist; }; =20 enum proc_param { @@ -99,6 +100,9 @@ static int proc_parse_subset_param(struct fs_context *fc= , char *value) if (*value !=3D '\0') { if (!strcmp(value, "pid")) { ctx->pidonly =3D PROC_PIDONLY_ON; + } else if (IS_ENABLED(CONFIG_PROC_ALLOW_LIST) && + !strcmp(value, "allowlist")) { + ctx->allowlist =3D PROC_ALLOWLIST_ON; } else { return invalf(fc, "proc: unsupported subset option - %s\n", value); } @@ -142,6 +146,18 @@ static int proc_parse_param(struct fs_context *fc, str= uct fs_parameter *param) return 0; } =20 +static char *proc_init_allowlist(void) +{ + char *content =3D kstrdup("allowlist\n", GFP_KERNEL_ACCOUNT); + + if (!content) { + pr_err("proc_init_allowlist: allocation allowlist failed\n"); + return NULL; + } + + return content; +} + static void proc_apply_options(struct proc_fs_info *fs_info, struct fs_context *fc, struct user_namespace *user_ns) @@ -152,8 +168,14 @@ static void proc_apply_options(struct proc_fs_info *fs= _info, fs_info->pid_gid =3D make_kgid(user_ns, ctx->gid); if (ctx->mask & (1 << Opt_hidepid)) fs_info->hide_pid =3D ctx->hidepid; - if (ctx->mask & (1 << Opt_subset)) + if (ctx->mask & (1 << Opt_subset)) { fs_info->pidonly =3D ctx->pidonly; + if (ctx->allowlist =3D=3D PROC_ALLOWLIST_ON) { + fs_info->allowlist =3D proc_init_allowlist(); + } else { + fs_info->allowlist =3D NULL; + } + } } =20 static int proc_fill_super(struct super_block *s, struct fs_context *fc) @@ -167,6 +189,8 @@ static int proc_fill_super(struct super_block *s, struc= t fs_context *fc) if (!fs_info) return -ENOMEM; =20 + rwlock_init(&fs_info->allowlist_lock); + fs_info->pid_ns =3D get_pid_ns(ctx->pid_ns); proc_apply_options(fs_info, fc, current_user_ns()); =20 @@ -271,6 +295,7 @@ static void proc_kill_sb(struct super_block *sb) =20 kill_anon_super(sb); put_pid_ns(fs_info->pid_ns); + kfree(fs_info->allowlist); kfree(fs_info); } =20 diff --git a/include/linux/proc_fs.h b/include/linux/proc_fs.h index 0260f5ea98fe..9105d75aeb18 100644 --- a/include/linux/proc_fs.h +++ b/include/linux/proc_fs.h @@ -24,6 +24,7 @@ enum { #else PROC_ENTRY_PERMANENT =3D 1U << 0, #endif + PROC_ENTRY_ALLOWLIST =3D 1U << 1, }; =20 struct proc_ops { @@ -58,6 +59,11 @@ enum proc_pidonly { PROC_PIDONLY_ON =3D 1, }; =20 +enum proc_allowlist { + PROC_ALLOWLIST_OFF =3D 0, + PROC_ALLOWLIST_ON =3D 1, +}; + struct proc_fs_info { struct pid_namespace *pid_ns; struct dentry *proc_self; /* For /proc/self */ @@ -65,6 +71,8 @@ struct proc_fs_info { kgid_t pid_gid; enum proc_hidepid hide_pid; enum proc_pidonly pidonly; + char *allowlist; + rwlock_t allowlist_lock; }; =20 static inline struct proc_fs_info *proc_sb_info(struct super_block *sb) --=20 2.33.6 From nobody Sun Sep 14 07:57:28 2025 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 vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id E95CAC54E94 for ; Wed, 25 Jan 2023 15:31:18 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S236097AbjAYPbR (ORCPT ); Wed, 25 Jan 2023 10:31:17 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:53374 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S236080AbjAYPbG (ORCPT ); Wed, 25 Jan 2023 10:31:06 -0500 Received: from us-smtp-delivery-44.mimecast.com (us-smtp-delivery-44.mimecast.com [205.139.111.44]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 904405999A for ; Wed, 25 Jan 2023 07:31:01 -0800 (PST) Received: from mimecast-mx02.redhat.com (mx3-rdu2.redhat.com [66.187.233.73]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-341-08M_FhysPKuYZMQaiPJFQg-1; Wed, 25 Jan 2023 10:29:31 -0500 X-MC-Unique: 08M_FhysPKuYZMQaiPJFQg-1 Received: from smtp.corp.redhat.com (int-mx04.intmail.prod.int.rdu2.redhat.com [10.11.54.4]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 69DC72804832; Wed, 25 Jan 2023 15:29:31 +0000 (UTC) Received: from comp-core-i7-2640m-0182e6.redhat.com (ovpn-208-16.brq.redhat.com [10.40.208.16]) by smtp.corp.redhat.com (Postfix) with ESMTP id 9A4D72026D76; Wed, 25 Jan 2023 15:29:29 +0000 (UTC) From: Alexey Gladkov To: LKML , containers@lists.linux.dev, linux-fsdevel@vger.kernel.org Cc: Alexey Dobriyan , Al Viro , Andrew Morton , Christian Brauner , Val Cowan Subject: [RFC PATCH v1 3/6] proc: Check that subset= option has been set Date: Wed, 25 Jan 2023 16:28:50 +0100 Message-Id: <346dd92ea62d8469416e12ab71b67b775eb2494b.1674660533.git.legion@kernel.org> In-Reply-To: References: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.4 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" Refactor subset option. Before this option had only one value - pid. Now another meaning has appeared and therefore their combinations are possible. Signed-off-by: Alexey Gladkov --- fs/proc/generic.c | 4 ++-- fs/proc/inode.c | 16 +++++++++++++--- fs/proc/internal.h | 6 ------ fs/proc/proc_allowlist.c | 22 ++++------------------ fs/proc/root.c | 27 +++++++++++++++++++-------- include/linux/proc_fs.h | 15 +++++---------- 6 files changed, 43 insertions(+), 47 deletions(-) diff --git a/fs/proc/generic.c b/fs/proc/generic.c index d4c8589987e7..71a38b275814 100644 --- a/fs/proc/generic.c +++ b/fs/proc/generic.c @@ -269,7 +269,7 @@ struct dentry *proc_lookup(struct inode *dir, struct de= ntry *dentry, { struct proc_fs_info *fs_info =3D proc_sb_info(dir->i_sb); =20 - if (fs_info->pidonly =3D=3D PROC_PIDONLY_ON && !proc_has_allowlist(fs_inf= o)) + if ((fs_info->subset & PROC_SUBSET_PIDONLY) && !(fs_info->subset & PROC_S= UBSET_ALLOWLIST)) return ERR_PTR(-ENOENT); =20 return proc_lookup_de(dir, dentry, PDE(dir)); @@ -334,7 +334,7 @@ int proc_readdir(struct file *file, struct dir_context = *ctx) struct inode *inode =3D file_inode(file); struct proc_fs_info *fs_info =3D proc_sb_info(inode->i_sb); =20 - if (fs_info->pidonly =3D=3D PROC_PIDONLY_ON && !proc_has_allowlist(fs_inf= o)) + if ((fs_info->subset & PROC_SUBSET_PIDONLY) && !(fs_info->subset & PROC_S= UBSET_ALLOWLIST)) return 1; =20 return proc_readdir_de(file, ctx, PDE(inode)); diff --git a/fs/proc/inode.c b/fs/proc/inode.c index f495fdb39151..4c486237a16b 100644 --- a/fs/proc/inode.c +++ b/fs/proc/inode.c @@ -184,9 +184,19 @@ static int proc_show_options(struct seq_file *seq, str= uct dentry *root) seq_printf(seq, ",gid=3D%u", from_kgid_munged(&init_user_ns, fs_info->pi= d_gid)); if (fs_info->hide_pid !=3D HIDEPID_OFF) seq_printf(seq, ",hidepid=3D%s", hidepid2str(fs_info->hide_pid)); - if (fs_info->pidonly !=3D PROC_PIDONLY_OFF) - seq_printf(seq, ",subset=3Dpid"); - + if (fs_info->subset & PROC_SUBSET_SET) { + bool need_delim =3D false; + seq_printf(seq, ",subset=3D"); + if (fs_info->subset & PROC_SUBSET_PIDONLY) { + seq_printf(seq, "pid"); + need_delim =3D true; + } + if (fs_info->subset & PROC_SUBSET_ALLOWLIST) { + if (need_delim) + seq_printf(seq, "+"); + seq_printf(seq, "allowlist"); + } + } return 0; } =20 diff --git a/fs/proc/internal.h b/fs/proc/internal.h index 999d105f6f96..3e1b1f29b13d 100644 --- a/fs/proc/internal.h +++ b/fs/proc/internal.h @@ -334,14 +334,8 @@ static inline void pde_force_lookup(struct proc_dir_en= try *pde) * proc_allowlist.c */ #ifdef CONFIG_PROC_ALLOW_LIST -extern bool proc_has_allowlist(struct proc_fs_info *); extern bool proc_pde_access_allowed(struct proc_fs_info *, struct proc_dir= _entry *); #else -static inline bool proc_has_allowlist(struct proc_fs_info *fs_info) -{ - return false; -} - static inline bool proc_pde_access_allowed(struct proc_fs_info *fs_info, s= truct proc_dir_entry *pde) { return true; diff --git a/fs/proc/proc_allowlist.c b/fs/proc/proc_allowlist.c index b38e11b04199..2153acb8e467 100644 --- a/fs/proc/proc_allowlist.c +++ b/fs/proc/proc_allowlist.c @@ -16,38 +16,24 @@ #define FILE_SEQFILE(f) ((struct seq_file *)((f)->private_data)) #define FILE_DATA(f) (FILE_SEQFILE(f)->private) =20 -bool proc_has_allowlist(struct proc_fs_info *fs_info) -{ - bool ret; - unsigned long flags; - - read_lock_irqsave(&fs_info->allowlist_lock, flags); - ret =3D (fs_info->allowlist =3D=3D NULL); - read_unlock_irqrestore(&fs_info->allowlist_lock, flags); - - return ret; -} - bool proc_pde_access_allowed(struct proc_fs_info *fs_info, struct proc_dir= _entry *de) { bool ret =3D false; char *ptr; unsigned long flags; =20 - read_lock_irqsave(&fs_info->allowlist_lock, flags); - - if (!fs_info->allowlist) { - read_unlock_irqrestore(&fs_info->allowlist_lock, flags); - + if (!(fs_info->subset & PROC_SUBSET_ALLOWLIST)) { if (!pde_is_allowlist(de)) ret =3D true; =20 return ret; } =20 + read_lock_irqsave(&fs_info->allowlist_lock, flags); + ptr =3D fs_info->allowlist; =20 - while (*ptr !=3D '\0') { + while (ptr && *ptr !=3D '\0') { struct proc_dir_entry *pde; char *sep, *end; size_t len, pathlen; diff --git a/fs/proc/root.c b/fs/proc/root.c index 1564f5cd118d..6e9b125072e5 100644 --- a/fs/proc/root.c +++ b/fs/proc/root.c @@ -31,8 +31,7 @@ struct proc_fs_context { unsigned int mask; enum proc_hidepid hidepid; int gid; - enum proc_pidonly pidonly; - enum proc_allowlist allowlist; + unsigned int subset; }; =20 enum proc_param { @@ -91,6 +90,8 @@ static int proc_parse_subset_param(struct fs_context *fc,= char *value) { struct proc_fs_context *ctx =3D fc->fs_private; =20 + ctx->subset |=3D PROC_SUBSET_SET; + while (value) { char *ptr =3D strchr(value, '+'); =20 @@ -99,10 +100,10 @@ static int proc_parse_subset_param(struct fs_context *= fc, char *value) =20 if (*value !=3D '\0') { if (!strcmp(value, "pid")) { - ctx->pidonly =3D PROC_PIDONLY_ON; + ctx->subset |=3D PROC_SUBSET_PIDONLY; } else if (IS_ENABLED(CONFIG_PROC_ALLOW_LIST) && !strcmp(value, "allowlist")) { - ctx->allowlist =3D PROC_ALLOWLIST_ON; + ctx->subset |=3D PROC_SUBSET_ALLOWLIST; } else { return invalf(fc, "proc: unsupported subset option - %s\n", value); } @@ -169,8 +170,8 @@ static void proc_apply_options(struct proc_fs_info *fs_= info, if (ctx->mask & (1 << Opt_hidepid)) fs_info->hide_pid =3D ctx->hidepid; if (ctx->mask & (1 << Opt_subset)) { - fs_info->pidonly =3D ctx->pidonly; - if (ctx->allowlist =3D=3D PROC_ALLOWLIST_ON) { + fs_info->subset =3D ctx->subset; + if (ctx->subset & PROC_SUBSET_ALLOWLIST) { fs_info->allowlist =3D proc_init_allowlist(); } else { fs_info->allowlist =3D NULL; @@ -346,14 +347,21 @@ static int proc_root_getattr(struct user_namespace *m= nt_userns, =20 static struct dentry *proc_root_lookup(struct inode * dir, struct dentry *= dentry, unsigned int flags) { - if (!proc_pid_lookup(dentry, flags)) - return NULL; + struct proc_fs_info *fs_info =3D proc_sb_info(dir->i_sb); + + if (!(fs_info->subset & PROC_SUBSET_SET) || (fs_info->subset & PROC_SUBSE= T_PIDONLY)) { + if (!proc_pid_lookup(dentry, flags)) + return NULL; + } =20 return proc_lookup(dir, dentry, flags); } =20 static int proc_root_readdir(struct file *file, struct dir_context *ctx) { + struct inode *inode =3D file_inode(file); + struct proc_fs_info *fs_info =3D proc_sb_info(inode->i_sb); + if (ctx->pos < FIRST_PROCESS_ENTRY) { int error =3D proc_readdir(file, ctx); if (unlikely(error <=3D 0)) @@ -361,6 +369,9 @@ static int proc_root_readdir(struct file *file, struct = dir_context *ctx) ctx->pos =3D FIRST_PROCESS_ENTRY; } =20 + if ((fs_info->subset & PROC_SUBSET_SET) && !(fs_info->subset & PROC_SUBSE= T_PIDONLY)) + return 1; + return proc_pid_readdir(file, ctx); } =20 diff --git a/include/linux/proc_fs.h b/include/linux/proc_fs.h index 9105d75aeb18..08d0d0ae6e42 100644 --- a/include/linux/proc_fs.h +++ b/include/linux/proc_fs.h @@ -53,15 +53,10 @@ enum proc_hidepid { HIDEPID_NOT_PTRACEABLE =3D 4, /* Limit pids to only ptraceable pids */ }; =20 -/* definitions for proc mount option pidonly */ -enum proc_pidonly { - PROC_PIDONLY_OFF =3D 0, - PROC_PIDONLY_ON =3D 1, -}; - -enum proc_allowlist { - PROC_ALLOWLIST_OFF =3D 0, - PROC_ALLOWLIST_ON =3D 1, +enum proc_subset { + PROC_SUBSET_SET =3D (1 << 0), + PROC_SUBSET_PIDONLY =3D (1 << 1), + PROC_SUBSET_ALLOWLIST =3D (1 << 2), }; =20 struct proc_fs_info { @@ -70,7 +65,7 @@ struct proc_fs_info { struct dentry *proc_thread_self; /* For /proc/thread-self */ kgid_t pid_gid; enum proc_hidepid hide_pid; - enum proc_pidonly pidonly; + unsigned int subset; char *allowlist; rwlock_t allowlist_lock; }; --=20 2.33.6 From nobody Sun Sep 14 07:57:28 2025 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 vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id E9D6DC54EAA for ; Wed, 25 Jan 2023 15:50:21 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S235442AbjAYPuT (ORCPT ); Wed, 25 Jan 2023 10:50:19 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:33816 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S234845AbjAYPuS (ORCPT ); Wed, 25 Jan 2023 10:50:18 -0500 X-Greylist: delayed 1232 seconds by postgrey-1.37 at lindbergh.monkeyblade.net; Wed, 25 Jan 2023 07:50:17 PST Received: from us-smtp-delivery-44.mimecast.com (unknown [207.211.30.44]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 11FF75867F for ; Wed, 25 Jan 2023 07:50:16 -0800 (PST) Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-619-LG1apAgMNhCnHMZXWZtq6A-1; Wed, 25 Jan 2023 10:29:34 -0500 X-MC-Unique: LG1apAgMNhCnHMZXWZtq6A-1 Received: from smtp.corp.redhat.com (int-mx04.intmail.prod.int.rdu2.redhat.com [10.11.54.4]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 73EAC85C6E2; Wed, 25 Jan 2023 15:29:33 +0000 (UTC) Received: from comp-core-i7-2640m-0182e6.redhat.com (ovpn-208-16.brq.redhat.com [10.40.208.16]) by smtp.corp.redhat.com (Postfix) with ESMTP id B32BE2026D4B; Wed, 25 Jan 2023 15:29:31 +0000 (UTC) From: Alexey Gladkov To: LKML , containers@lists.linux.dev, linux-fsdevel@vger.kernel.org Cc: Alexey Dobriyan , Al Viro , Andrew Morton , Christian Brauner , Val Cowan Subject: [RFC PATCH v1 4/6] proc: Allow to use the allowlist filter in userns Date: Wed, 25 Jan 2023 16:28:51 +0100 Message-Id: <76e8b2d0c0651af6906351b7d43fa2a4d117dc04.1674660533.git.legion@kernel.org> In-Reply-To: References: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.4 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" Signed-off-by: Alexey Gladkov --- fs/proc/proc_allowlist.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/proc/proc_allowlist.c b/fs/proc/proc_allowlist.c index 2153acb8e467..c605f73622bd 100644 --- a/fs/proc/proc_allowlist.c +++ b/fs/proc/proc_allowlist.c @@ -100,7 +100,7 @@ static int open_allowlist(struct inode *inode, struct f= ile *file) struct proc_fs_info *fs_info =3D proc_sb_info(inode->i_sb); int ret; =20 - if (!capable(CAP_SYS_ADMIN)) + if (!ns_capable(current_user_ns(), CAP_SYS_ADMIN)) return -EPERM; =20 // we need this because shrink_dcache_sb() can't drop our own dentry. @@ -199,7 +199,7 @@ static const struct proc_ops proc_allowlist_ops =3D { static int __init proc_allowlist_init(void) { struct proc_dir_entry *pde; - pde =3D proc_create("allowlist", S_IRUSR | S_IWUSR, NULL, &proc_allowlist= _ops); + pde =3D proc_create("allowlist", S_IRUSR | S_IWUSR | S_IRGRP | S_IWGRP | = S_IROTH | S_IWOTH, NULL, &proc_allowlist_ops); pde_make_permanent(pde); pde_make_allowlist(pde); return 0; --=20 2.33.6 From nobody Sun Sep 14 07:57:28 2025 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 vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 1B0DFC54E94 for ; Wed, 25 Jan 2023 15:31:16 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S235536AbjAYPbP (ORCPT ); Wed, 25 Jan 2023 10:31:15 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:53176 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S236078AbjAYPbG (ORCPT ); Wed, 25 Jan 2023 10:31:06 -0500 Received: from us-smtp-delivery-44.mimecast.com (us-smtp-delivery-44.mimecast.com [205.139.111.44]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 7463F301BD for ; Wed, 25 Jan 2023 07:31:01 -0800 (PST) Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-407-s0K0uYCeMX6KY8aP2rrCxQ-1; Wed, 25 Jan 2023 10:29:36 -0500 X-MC-Unique: s0K0uYCeMX6KY8aP2rrCxQ-1 Received: from smtp.corp.redhat.com (int-mx04.intmail.prod.int.rdu2.redhat.com [10.11.54.4]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 8A695811E6E; Wed, 25 Jan 2023 15:29:35 +0000 (UTC) Received: from comp-core-i7-2640m-0182e6.redhat.com (ovpn-208-16.brq.redhat.com [10.40.208.16]) by smtp.corp.redhat.com (Postfix) with ESMTP id C796A2026D4B; Wed, 25 Jan 2023 15:29:33 +0000 (UTC) From: Alexey Gladkov To: LKML , containers@lists.linux.dev, linux-fsdevel@vger.kernel.org Cc: Alexey Dobriyan , Al Viro , Andrew Morton , Christian Brauner , Val Cowan Subject: [RFC PATCH v1 5/6] proc: Validate incoming allowlist Date: Wed, 25 Jan 2023 16:28:52 +0100 Message-Id: <18bb5a8c0c81211cba5b865b4fbb5c2dd6b9e688.1674660533.git.legion@kernel.org> In-Reply-To: References: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.4 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" Signed-off-by: Alexey Gladkov --- fs/proc/internal.h | 10 +++ fs/proc/proc_allowlist.c | 165 ++++++++++++++++++++++++++++++--------- fs/proc/root.c | 22 +----- include/linux/proc_fs.h | 7 +- 4 files changed, 149 insertions(+), 55 deletions(-) diff --git a/fs/proc/internal.h b/fs/proc/internal.h index 3e1b1f29b13d..2ca4e53a4b4b 100644 --- a/fs/proc/internal.h +++ b/fs/proc/internal.h @@ -334,8 +334,18 @@ static inline void pde_force_lookup(struct proc_dir_en= try *pde) * proc_allowlist.c */ #ifdef CONFIG_PROC_ALLOW_LIST +extern int proc_allowlist_append(struct list_head *, const char *, size_t); +extern void proc_allowlist_free(struct list_head *); extern bool proc_pde_access_allowed(struct proc_fs_info *, struct proc_dir= _entry *); #else +static inline int proc_allowlist_append(struct list_head *, const char *, = size_t) +{ + return 0; +} +static inline void proc_allowlist_free(struct list_head *) +{ + return; +} static inline bool proc_pde_access_allowed(struct proc_fs_info *fs_info, s= truct proc_dir_entry *pde) { return true; diff --git a/fs/proc/proc_allowlist.c b/fs/proc/proc_allowlist.c index c605f73622bd..0115015c74f0 100644 --- a/fs/proc/proc_allowlist.c +++ b/fs/proc/proc_allowlist.c @@ -11,16 +11,56 @@ #include #include #include +#include #include "internal.h" =20 #define FILE_SEQFILE(f) ((struct seq_file *)((f)->private_data)) #define FILE_DATA(f) (FILE_SEQFILE(f)->private) =20 +int proc_allowlist_append(struct list_head *allowlist, const char *path, s= ize_t len) +{ + struct allowlist_entry *new; + + if (!len) + return 0; + + new =3D kmalloc(sizeof(*new), GFP_KERNEL_ACCOUNT); + if (!new) + goto nomem; + + new->path =3D kstrndup(path, len, GFP_KERNEL_ACCOUNT); + if (!new->path) + goto nomem; + + INIT_LIST_HEAD(&new->list); + list_add_tail(&new->list, allowlist); + + return 0; +nomem: + if (new) { + kfree(new->path); + kfree(new); + } + return -ENOMEM; +} + +void proc_allowlist_free(struct list_head *allowlist) +{ + struct list_head *el, *next; + struct allowlist_entry *entry; + + list_for_each_safe(el, next, allowlist) { + entry =3D list_entry(el, struct allowlist_entry, list); + kfree(entry->path); + kfree(entry); + } +} + bool proc_pde_access_allowed(struct proc_fs_info *fs_info, struct proc_dir= _entry *de) { bool ret =3D false; - char *ptr; unsigned long flags; + struct list_head *el, *next; =20 if (!(fs_info->subset & PROC_SUBSET_ALLOWLIST)) { if (!pde_is_allowlist(de)) @@ -31,24 +71,13 @@ bool proc_pde_access_allowed(struct proc_fs_info *fs_in= fo, struct proc_dir_entry =20 read_lock_irqsave(&fs_info->allowlist_lock, flags); =20 - ptr =3D fs_info->allowlist; - - while (ptr && *ptr !=3D '\0') { - struct proc_dir_entry *pde; - char *sep, *end; - size_t len, pathlen; + list_for_each_safe(el, next, &fs_info->allowlist) { + struct allowlist_entry *entry =3D list_entry(el, struct allowlist_entry,= list); =20 - if (!(sep =3D strchr(ptr, '\n'))) - pathlen =3D strlen(ptr); - else - pathlen =3D (sep - ptr); - - if (!pathlen) - goto next; - - pde =3D de; - end =3D NULL; - len =3D pathlen; + struct proc_dir_entry *pde =3D de; + char *end =3D NULL; + char *ptr =3D entry->path; + size_t len =3D strlen(entry->path); =20 while (ptr !=3D end && len > 0) { end =3D ptr + len - 1; @@ -72,8 +101,7 @@ bool proc_pde_access_allowed(struct proc_fs_info *fs_inf= o, struct proc_dir_entry =20 ret =3D true; break; -next: - ptr +=3D pathlen + 1; +next: ; } =20 read_unlock_irqrestore(&fs_info->allowlist_lock, flags); @@ -84,12 +112,18 @@ bool proc_pde_access_allowed(struct proc_fs_info *fs_i= nfo, struct proc_dir_entry static int show_allowlist(struct seq_file *m, void *v) { struct proc_fs_info *fs_info =3D proc_sb_info(m->file->f_inode->i_sb); - char *p =3D fs_info->allowlist; unsigned long flags; + struct list_head *el, *next; + struct allowlist_entry *entry; =20 read_lock_irqsave(&fs_info->allowlist_lock, flags); - if (p) - seq_puts(m, p); + + list_for_each_safe(el, next, &fs_info->allowlist) { + entry =3D list_entry(el, struct allowlist_entry, list); + seq_puts(m, entry->path); + seq_puts(m, "\n"); + } + read_unlock_irqrestore(&fs_info->allowlist_lock, flags); =20 return 0; @@ -155,34 +189,93 @@ static ssize_t write_allowlist(struct file *file, con= st char __user *buffer, siz return count; } =20 +static int allowlist_cmp(void *priv, const struct list_head *a, const stru= ct list_head *b) +{ + struct allowlist_entry *ia =3D list_entry(a, struct allowlist_entry, list= ); + struct allowlist_entry *ib =3D list_entry(b, struct allowlist_entry, list= ); + + return strcmp(ia->path, ib->path); +} + +static int recreate_allowlist(struct proc_fs_info *fs_info, const char *bu= f, size_t buflen) +{ + const char *ptr =3D buf; + size_t len =3D buflen; + size_t lineno =3D 1; + int ret =3D 0; + LIST_HEAD(allowlist); + + while (len > 0) { + char *sep; + size_t pathlen; + + if (!(sep =3D memchr(ptr, '\n', len))) + pathlen =3D buflen; + else + pathlen =3D (sep - ptr); + + if (pathlen > 0) { + ret =3D -ENAMETOOLONG; + if (pathlen >=3D PATH_MAX) { + pr_crit("allowlist:%lu: pathname is too long\n", lineno); + goto err; + } + + ret =3D -EINVAL; + if (*ptr =3D=3D '/') { + pr_crit("allowlist:%lu: the name must be relative to the mount point\n= ", lineno); + goto err; + } + if (!isalpha(*ptr)) { + pr_crit("allowlist:%lu: name must start with a letter\n", lineno); + goto err; + } + + proc_allowlist_append(&allowlist, ptr, pathlen); + } + + ptr +=3D pathlen + 1; + len -=3D pathlen + 1; + + lineno++; + } + + proc_allowlist_free(&fs_info->allowlist); + INIT_LIST_HEAD(&fs_info->allowlist); + + if (!list_empty(&allowlist)) { + list_replace(&allowlist, &fs_info->allowlist); + list_sort(NULL, &fs_info->allowlist, allowlist_cmp); + } + + return 0; +err: + proc_allowlist_free(&allowlist); + return ret; +} + static int close_allowlist(struct inode *inode, struct file *file) { struct seq_file *seq_file =3D FILE_SEQFILE(file); struct proc_fs_info *fs_info =3D proc_sb_info(inode->i_sb); =20 if (seq_file->buf && (file->f_mode & FMODE_WRITE)) { + unsigned long flags; char *buf; =20 if (!seq_get_buf(seq_file, &buf)) return -EIO; *buf =3D '\0'; =20 - if (strcmp(seq_file->buf, fs_info->allowlist)) { - unsigned long flags; - - buf =3D kstrndup(seq_file->buf, seq_file->count, GFP_KERNEL_ACCOUNT); - if (!buf) - return -EIO; - - write_lock_irqsave(&fs_info->allowlist_lock, flags); - - shrink_dcache_sb(inode->i_sb); - - kfree(fs_info->allowlist); - fs_info->allowlist =3D buf; + write_lock_irqsave(&fs_info->allowlist_lock, flags); =20 + if (recreate_allowlist(fs_info, seq_file->buf, seq_file->count) < 0) { write_unlock_irqrestore(&fs_info->allowlist_lock, flags); + return -EIO; } + + shrink_dcache_sb(inode->i_sb); + write_unlock_irqrestore(&fs_info->allowlist_lock, flags); } =20 return single_release(inode, file); diff --git a/fs/proc/root.c b/fs/proc/root.c index 6e9b125072e5..18436d70bb12 100644 --- a/fs/proc/root.c +++ b/fs/proc/root.c @@ -147,18 +147,6 @@ static int proc_parse_param(struct fs_context *fc, str= uct fs_parameter *param) return 0; } =20 -static char *proc_init_allowlist(void) -{ - char *content =3D kstrdup("allowlist\n", GFP_KERNEL_ACCOUNT); - - if (!content) { - pr_err("proc_init_allowlist: allocation allowlist failed\n"); - return NULL; - } - - return content; -} - static void proc_apply_options(struct proc_fs_info *fs_info, struct fs_context *fc, struct user_namespace *user_ns) @@ -171,11 +159,8 @@ static void proc_apply_options(struct proc_fs_info *fs= _info, fs_info->hide_pid =3D ctx->hidepid; if (ctx->mask & (1 << Opt_subset)) { fs_info->subset =3D ctx->subset; - if (ctx->subset & PROC_SUBSET_ALLOWLIST) { - fs_info->allowlist =3D proc_init_allowlist(); - } else { - fs_info->allowlist =3D NULL; - } + if (ctx->subset & PROC_SUBSET_ALLOWLIST) + proc_allowlist_append(&fs_info->allowlist, "allowlist", 10); } } =20 @@ -191,6 +176,7 @@ static int proc_fill_super(struct super_block *s, struc= t fs_context *fc) return -ENOMEM; =20 rwlock_init(&fs_info->allowlist_lock); + INIT_LIST_HEAD(&fs_info->allowlist); =20 fs_info->pid_ns =3D get_pid_ns(ctx->pid_ns); proc_apply_options(fs_info, fc, current_user_ns()); @@ -296,7 +282,7 @@ static void proc_kill_sb(struct super_block *sb) =20 kill_anon_super(sb); put_pid_ns(fs_info->pid_ns); - kfree(fs_info->allowlist); + proc_allowlist_free(&fs_info->allowlist); kfree(fs_info); } =20 diff --git a/include/linux/proc_fs.h b/include/linux/proc_fs.h index 08d0d0ae6e42..81c6b4b2ae97 100644 --- a/include/linux/proc_fs.h +++ b/include/linux/proc_fs.h @@ -59,6 +59,11 @@ enum proc_subset { PROC_SUBSET_ALLOWLIST =3D (1 << 2), }; =20 +struct allowlist_entry { + struct list_head list; + char *path; +}; + struct proc_fs_info { struct pid_namespace *pid_ns; struct dentry *proc_self; /* For /proc/self */ @@ -66,8 +71,8 @@ struct proc_fs_info { kgid_t pid_gid; enum proc_hidepid hide_pid; unsigned int subset; - char *allowlist; rwlock_t allowlist_lock; + struct list_head allowlist; }; =20 static inline struct proc_fs_info *proc_sb_info(struct super_block *sb) --=20 2.33.6 From nobody Sun Sep 14 07:57:28 2025 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 vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 9848BC54E94 for ; Wed, 25 Jan 2023 15:30:06 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S236068AbjAYPaE (ORCPT ); Wed, 25 Jan 2023 10:30:04 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:51790 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S236050AbjAYP3u (ORCPT ); Wed, 25 Jan 2023 10:29:50 -0500 Received: from us-smtp-delivery-44.mimecast.com (us-smtp-delivery-44.mimecast.com [205.139.111.44]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 0024E59B63 for ; Wed, 25 Jan 2023 07:29:42 -0800 (PST) Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-554-J2YxgZukPJadicB8_9FLOw-1; Wed, 25 Jan 2023 10:29:38 -0500 X-MC-Unique: J2YxgZukPJadicB8_9FLOw-1 Received: from smtp.corp.redhat.com (int-mx04.intmail.prod.int.rdu2.redhat.com [10.11.54.4]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 97A4E800B30; Wed, 25 Jan 2023 15:29:37 +0000 (UTC) Received: from comp-core-i7-2640m-0182e6.redhat.com (ovpn-208-16.brq.redhat.com [10.40.208.16]) by smtp.corp.redhat.com (Postfix) with ESMTP id D290B2026D4B; Wed, 25 Jan 2023 15:29:35 +0000 (UTC) From: Alexey Gladkov To: LKML , containers@lists.linux.dev, linux-fsdevel@vger.kernel.org Cc: Alexey Dobriyan , Al Viro , Andrew Morton , Christian Brauner , Val Cowan Subject: [RFC PATCH v1 6/6] doc: proc: Add description of subset=allowlist Date: Wed, 25 Jan 2023 16:28:53 +0100 Message-Id: In-Reply-To: References: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.4 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" Signed-off-by: Alexey Gladkov --- Documentation/filesystems/proc.rst | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/Documentation/filesystems/proc.rst b/Documentation/filesystems= /proc.rst index e224b6d5b642..c2598bca8193 100644 --- a/Documentation/filesystems/proc.rst +++ b/Documentation/filesystems/proc.rst @@ -2213,6 +2213,16 @@ information about processes information, just add id= entd to this group. subset=3Dpid hides all top level files and directories in the procfs that are not related to tasks. =20 +subset=3Dallowlist allows you to specify a list of files and directories to +which you want to provide access. If the option is specified, then the +/proc/allowlist will appear at the top level of the filesystem. By default= , this +file contains only its name. The user can add or remove other filenames and +directories. To prohibit editing the allowlist, you need to exclude its na= me +from the list of allowed ones. + +Different subset=3D option arguments can be combined using the plus(+) del= imiter. +For example: subset=3Dpid+allowlist + Chapter 5: Filesystem behavior =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D =20 --=20 2.33.6