integrity: Linux Integrity Module(LIM)
Subsystems (9)
RFC PATCH 4/5 integrity: Linux Integrity Module(LIM)
This is a request for comments for a redesign of the integrity patches. The new version addresses a number of issues, including
- generalizing the measurement API beyond just inode measurements.
- separation of the measurement into distinct collection, appraisal, and commitment phases, for greater flexibility.
- redesigning the interactions with LSM modules (Selinux and Smack) to be acceptable to those groups.
Extended Verification Module(EVM) and the Integrity Measurement Architecture(IMA) were originally implemented as an LSM module. Based on discussions on the LSM mailing list, a decision was made that the LSM hooks should only be used to enforce mandatory access control decisions and a new set of hooks should be defined specifically for integrity.
EVM/IMA was limited to verifying and measuring a file's (i.e. an inode) integrity and the metadata associated with it. Current research is looking into other types of integrity measurements. (i.e. "Linux kernel integrity measurement using contextual inspection", by Peter A. Loscocco, Perry W. Wilson, J. Aaron Pendergrass, C. Durward McDonell,
http://doi.acm.org/10.1145/1314354.1314362) As a result, a requirement of the new integrity framework is support for different types of integrity measurements.
Our work has shown that requests for integrity verification/measurement needs to be based on knowledge of the filesystem, requiring the system to either be labeled with integrity data or depend on the existent LSM security labels. The previous set of integrity patches modified the LSM modules to be integrity context aware, meaning that the LSM modules made integrity data/metadata verification and measurement API calls based on an understanding of the LSM security labels. Both of the LSM maintainers felt that the changes were too intrusive and that integrity enforcement should be made by the integrity provider, not the LSM module.
To address these concerns, Stephen Smalley suggested using the security_audit_rule_match(), renamed to security_filter_rule_match(), to define LSM specific integrity measurement policy rules, in lieu of modifying the LSM modules. In the current set of patches, the integrity API calls can be made either by the integrity provider (i.e. IMA), based on an LSM specific integrity policy, or by an integrity context aware LSM module.
This patch provides an integrity framework(api and hooks), placement of the integrity hooks in the appropriate places in the fs directory, and a dummy provider for an integrity service. Collecting, appraising, and storing of inode and other types of integrity data is supported. Multiple integrity templates, which implement the integrity API, may register themselves. For now, only a single integrity provider can register itself for the integrity hooks. (Support for multiple providers registering themselves for the integrity hooks would require some form of stacking.)
The ten integrity hooks are:
inode_setxattr, inode_post_setxattr, inode_alloc_integrity, inode_free_integrity, inode_init_integrity, file_free_integrity, d_instantiate, bprm_check_integrity, file_mmap, inode_permission
The five integrity API calls provided are:
integrity_must_measure, integrity_collect_measurement, integrity_appraise_measurement, integrity_store_measurement, and integrity_display_template.
The type of integrity data being collected, appraised, stored, or displayed is template dependent.
(Details on the calls and their exact arguments are in linux/integrity.h, included in the patch.)
---
include/linux/integrity.h new file +261 -0
@@ -0,0 +1,261 @@ +/* +* integrity.h +* +* Copyright (C) 2005,2006,2007,2008 IBM Corporation +* Author: Mimi Zohar <[email protected]> +* +* This program is free software; you can redistribute it and/or modify +* it under the terms of the GNU General Public License as published by +* the Free Software Foundation, version 2 of the License. +*/ + +#ifndef _LINUX_INTEGRITY_H +#define _LINUX_INTEGRITY_H + +#include <linux/fs.h> +#include <linux/audit.h> + +#ifdef CONFIG_INTEGRITY +void integrity_audit_msg (int audit_msgno,struct inode *inode, +const unsigned char *fname,char *op, +char *cause,int result); + +enum lim_hooks {INODE_PERMISSION =1 ,FILE_MMAP ,BPRM_CHECK }; +extern int integrity_measure_policy (struct inode *inode,enum lim_hooks func, +int mask); +extern int integrity_measure_rule_add (char *,char *,char *,char *); +extern void integrity_measure_policy_init (void ); +extern void integrity_measure_policy_complete (void ); + +/* +* Integrity API calls: +* +* @collect_measurement: +* Collect template specific measurement data. +* @data contains template specific data used for collecting the +* measurement. +* Return 0 if operation was successful. +* +* @appraise_measurement: +* Appraise the integrity of the template specific measurement data. +* @data contains template specific data used for appraising the +* measurement. +* Return 0 if operation was successful. +* +* @store_measurement: +* Store the template specific data. +* @data contains template specific data used for storing the +* measurement. +* +* @must_measure: +* Measurement decision based on an integrity policy. +* @data contains template specific data used for making policy +* decision. +* Return 0 if operation was successful. +* +* @display_template: +* Display template specific data. +* +*/ + +enum integrity_show_type {INTEGRITY_SHOW_BINARY ,INTEGRITY_SHOW_ASCII }; + +struct template_operations { +int (*collect_measurement)(void *); +int (*appraise_measurement)(void *); +void (*store_measurement)(void *); +int (*must_measure)(void *); +void (*display_template)(struct seq_file *m,void *, +enum integrity_show_type ); +}; +extern int register_template (char *template_name, +struct template_operations *ops); +extern int unregister_template (char *template_name); +extern int integrity_find_template (char *,struct template_operations **ops); + +/* +* Integrity hooks: +* +* @bprm_check_integrity: +* This hook mediates the point when a search for a binary handler will +* begin. At this point, the OS protects against an executable file, +* already open for write, from being executed; and an executable file +* already open for execute, from being modified. So we can be certain +* that any measurements(collect, appraise, store) done here are of +* the file being executed. +* @bprm contains the linux_binprm structure. +* Return 0 if the hook is successful and permission is granted. +* +* @inode_setxattr: +* Check permission before permitting an integrity extended attribute +* to be set. +* @value identified by @name for @dentry. +* +* @inode_post_setxattr: +* Update inode integrity xattr after successful setxattr operation. +* identified by @name for @dentry. +* +* @inode_alloc_integrity: +* Allocate and attach an integrity structure to @inode->i_integrity. The +* i_integrity field is initialized to NULL when the inode structure is +* allocated. +* @inode contains the inode structure. +* Return 0 if operation was successful. +* +* @inode_free_integrity: +* @inode contains the inode structure. +* Deallocate the inode integrity structure and set @inode->i_integrity to +* NULL. +* +* @inode_init_integrity: +* Create inode integrity xattr for a new inode based on the new security +* xattr information. +* @inode contains the inode structure of the newly created inode. +* @dir contains the inode structure of the parent directory. +* @name contains the security xattr name suffix. +* @value contains the security attribute value. +* @len contains the length of the security attribute value. +* +* @inode_permission: +* This hook is called by the existing Linux permission function, when +* a file is opened (as well as many other operations). At this point, +* measurements of files open for read(collect, appraise, store) can +* be made. +* @inode contains the inode structure to check. +* @mask contains the permission mask. +* @nd contains the nameidata (may be NULL). +* +* @file_free_integrity: +* Update the integrity xattr value as necessary. +* *file contains the file structure being closed. +* +* @file_mmap : +* Measurement(collect, appraise, store) of files mmaped for EXEC, +* could be measured at this point. +* @file contains the file structure for file to map (may be NULL). +* @reqprot contains the protection requested by the application. +* @prot contains the protection that will be applied by the kernel. +* @flags contains the operational flags. +* Return 0 if permission is granted. +* +* @d_instantiate: +* Initialize the integrity structure of an inode for a dentry. +* @dentry to complete. +* @inode to attach to this dentry. +* +*/ + +struct integrity_operations { +int (*bprm_check_integrity) (struct linux_binprm *bprm); +int (*inode_setxattr) (struct dentry *dentry,const char *name, +const void *value,size_t size,int flags); +void (*inode_post_setxattr) (struct dentry *dentry,const char *name); +int (*inode_alloc_integrity) (struct inode *inode); +void (*inode_free_integrity) (struct inode *inode); +void (*inode_init_integrity) (struct inode *inode,struct inode *dir, +char **name,void **value,size_t *len); +int (*inode_permission) (struct inode *inode,int mask, +struct nameidata *nd); +void (*file_free_integrity) (struct file *file); +int (*file_mmap) (struct file *file, +unsigned long reqprot,unsigned long prot, +unsigned long flags,unsigned long addr, +unsigned long addr_only); +void (*d_instantiate) (struct dentry *dentry,struct inode *inode); +}; +extern int register_integrity (struct integrity_operations *ops); +extern int unregister_integrity (struct integrity_operations *ops); + +/* global variables */ +extern struct integrity_operations *integrity_ops; + + +int integrity_collect_measurement (char *template_name,void *data); +int integrity_appraise_measurement (char *template_name,void *data); +int integrity_must_measure (char *template_name,void *data); +void integrity_store_measurement (char *template_name,void *data); + +int integrity_bprm_check (struct linux_binprm *bprm); +int integrity_inode_setxattr (struct dentry *dentry,const char *name, +const void *value,size_t size,int flags); +void integrity_inode_post_setxattr (struct dentry *dentry,const char *name); +int integrity_inode_alloc (struct inode *inode); +void integrity_inode_free (struct inode *inode); +void integrity_inode_init_integrity (struct inode *inode,struct inode *dir, +char **name,void **value,size_t *len); +int integrity_inode_permission (struct inode *inode,int mask, +struct nameidata *nd); +void integrity_file_free (struct file *file); +int integrity_file_mmap (struct file *file, +unsigned long reqprot,unsigned long prot, +unsigned long flags,unsigned long addr, +unsigned long addr_only); +void integrity_d_instantiate (struct dentry *dentry,struct inode *inode); + +#else + +static inline int integrity_bprm_check (struct linux_binprm *bprm) +{ +return 0 ; +} + +static inline int integrity_inode_setxattr (struct dentry *dentry, +const char *name,const void *value, +size_t size,int flags) +{ +return 0 ; +} + +static inline void integrity_inode_post_setxattr (struct dentry *dentry, +const char *name) +{ } + +static inline int integrity_inode_alloc (struct inode *inode) +{ +return 0 ; +} + +static inline void integrity_inode_free (struct inode *inode) +{ } + +static inline int integrity_inode_init_integrity (struct inode *inode, +struct inode *dir, +char **name, +void **value, +size_t *len) +{ +return -EOPNOTSUPP ; +} + +static inline int integrity_inode_permission (struct inode *inode,int mask, +struct nameidata *nd) +{ +return 0 ; +} + +static inline int integrity_file_permission (struct file *file,int mask) +{ +return 0 ; +} + +static inline void integrity_file_free (struct file *file) +{ +return ; +} + +static inline int integrity_file_mmap (struct file *file, +unsigned long reqprot,unsigned long prot, +unsigned long flags,unsigned long addr, +unsigned long addr_only) +{ +return 0 ; +} + +static inline void integrity_d_instantiate (struct dentry *dentry, +struct inode *inode) +{ +return ; +} + +#endif +#endif
security/integrity/integrity.c new file +259 -0
259 lines not on this page · raw message
security/integrity/integrity_dummy.c new file +112 -0
112 lines not on this page · raw message
security/integrity/integrity_dummy.h new file +13 -0
@@ -0,0 +1,13 @@ +/* +* integrity_dummy.h +* +* Copyright (C) 2005,2006,2007 IBM Corporation +* Author: Mimi Zohar <[email protected]> +* +* This program is free software; you can redistribute it and/or modify +* it under the terms of the GNU General Public License as published by +* the Free Software Foundation, version 2 of the License. +*/ + +extern struct integrity_operations dummy_integrity_ops; +extern void integrity_fixup_ops (struct integrity_operations *ops);
fs/dcache.c +6 -0
@@ -28,6 +28,7 @@#include <linux/file.h> #include <asm/uaccess.h> #include <linux/security.h> +#include <linux/integrity.h> #include <linux/seqlock.h> #include <linux/swap.h> #include <linux/bootmem.h>
@@ -972,6 +973,7 @@ void d_instantiate(struct dentry *entry, entry->d_inode = inode;fsnotify_d_instantiate (entry, inode);spin_unlock (&dcache_lock); +integrity_d_instantiate (entry, inode);security_d_instantiate (entry, inode); }
@@ -1036,6 +1038,7 @@ struct dentry *d_instantiate_unique(struspin_unlock (&dcache_lock);if (!result) { +integrity_d_instantiate (entry, inode);security_d_instantiate (entry, inode);return NULL ; }
@@ -1173,6 +1176,7 @@ struct dentry *d_splice_alias(struct inoBUG_ON (!(new->d_flags &DCACHE_DISCONNECTED ));fsnotify_d_instantiate (new, inode);spin_unlock (&dcache_lock); +integrity_d_instantiate (new, inode);security_d_instantiate (new, inode);d_rehash (dentry);d_move (new, dentry);
@@ -1183,6 +1187,7 @@ struct dentry *d_splice_alias(struct ino dentry->d_inode = inode;fsnotify_d_instantiate (dentry, inode);spin_unlock (&dcache_lock); +integrity_d_instantiate (dentry, inode);security_d_instantiate (dentry, inode);d_rehash (dentry); }
@@ -1733,6 +1738,7 @@ found:spin_unlock (&dcache_lock); out_nolock:if (actual == dentry) { +integrity_d_instantiate (dentry, inode);security_d_instantiate (dentry, inode);return NULL ; }
fs/ext3/xattr_security.c +8 -0
@@ -9,6 +9,7 @@#include <linux/ext3_jbd.h> #include <linux/ext3_fs.h> #include <linux/security.h> +#include <linux/integrity.h> #include "xattr.h" static size_t
@@ -57,12 +58,19 @@ ext3_init_security(handle_t *handle, str err =security_inode_init_security (inode, dir, &name, &value, &len);if (err) { +/* Even if creation of the security xattr fails, must +* indicate this is a new inode. */ +integrity_inode_init_integrity (inode, dir,NULL ,NULL ,NULL );if (err == -EOPNOTSUPP )return 0 ;return err; } err =ext3_xattr_set_handle (handle, inode,EXT3_XATTR_INDEX_SECURITY , name, value, len,0 ); + +integrity_inode_init_integrity (inode, dir, &name, &value, &len); + err =ext3_xattr_set_handle (handle, inode,EXT3_XATTR_INDEX_SECURITY , + name, value, len,0 );kfree (name);kfree (value);return err;
fs/file_table.c +3 -0
@@ -13,6 +13,7 @@#include <linux/module.h> #include <linux/fs.h> #include <linux/security.h> +#include <linux/integrity.h> #include <linux/eventpoll.h> #include <linux/rcupdate.h> #include <linux/mount.h>
@@ -272,6 +273,7 @@ void __fput(struct file *file)if (file->f_op && file->f_op->release) file->f_op->release (inode, file);security_file_free (file); +integrity_file_free (file);if (unlikely (S_ISCHR (inode->i_mode) && inode->i_cdev !=NULL ))cdev_put (inode->i_cdev);fops_put (file->f_op);
@@ -343,6 +345,7 @@ void put_filp(struct file *file) {if (atomic_dec_and_test (&file->f_count)) {security_file_free (file); +integrity_file_free (file);file_kill (file);file_free (file); }
fs/inode.c +10 -0
@@ -17,6 +17,7 @@#include <linux/hash.h> #include <linux/swap.h> #include <linux/security.h> +#include <linux/integrity.h> #include <linux/pagemap.h> #include <linux/cdev.h> #include <linux/bootmem.h>
@@ -160,6 +161,14 @@ static struct inode *alloc_inode(structinit_rwsem (&inode->i_alloc_sem);lockdep_set_class (&inode->i_alloc_sem, &sb->s_type->i_alloc_sem_key); +if (integrity_inode_alloc (inode)) { +if (inode->i_sb->s_op->destroy_inode) + inode->i_sb->s_op->destroy_inode (inode); +else +kmem_cache_free (inode_cachep, (inode)); +return NULL ; + } + mapping->a_ops = &empty_aops; mapping->host = inode; mapping->flags =0 ;
@@ -190,6 +199,7 @@ void destroy_inode(struct inode *inode) {BUG_ON (inode_has_buffers (inode));security_inode_free (inode); +integrity_inode_free (inode);if (inode->i_sb->s_op->destroy_inode) inode->i_sb->s_op->destroy_inode (inode);else
fs/xattr.c +9 -0
@@ -14,6 +14,7 @@#include <linux/mount.h> #include <linux/namei.h> #include <linux/security.h> +#include <linux/integrity.h> #include <linux/syscalls.h> #include <linux/module.h> #include <linux/fsnotify.h>
@@ -81,11 +82,17 @@ vfs_setxattr(struct dentry *dentry, cons error =security_inode_setxattr (dentry, name, value, size, flags);if (error) goto out; + + error =integrity_inode_setxattr (dentry, name, value, size, flags); +if (error) + goto out; + error = -EOPNOTSUPP ;if (inode->i_op->setxattr) { error = inode->i_op->setxattr (dentry, name, value, size, flags);if (!error) {fsnotify_xattr (dentry); +integrity_inode_post_setxattr (dentry, name);security_inode_post_setxattr (dentry, name, value, size, flags); }
@@ -205,6 +212,8 @@ vfs_removexattr(struct dentry *dentry, cmutex_lock (&inode->i_mutex); error = inode->i_op->removexattr (dentry, name); +if (!error) +integrity_inode_post_setxattr (dentry, name);mutex_unlock (&inode->i_mutex);if (!error)
include/linux/fs.h +3 -0
@@ -653,6 +653,9 @@ struct inode {#ifdef CONFIG_SECURITY void *i_security;#endif +#ifdef CONFIG_INTEGRITY +void *i_integrity; +#endif void *i_private;/* fs or device private pointer */ };
include/linux/audit.h +8 -0
@@ -123,6 +123,11 @@#define AUDIT_LAST_KERN_ANOM_MSG 1799#define AUDIT_ANOM_PROMISCUOUS 1700/* Device changed promiscuous mode */ #define AUDIT_ANOM_ABEND 1701/* Process ended abnormally */ +#define AUDIT_INTEGRITY_DATA 1800/* Data integrity verification */ +#define AUDIT_INTEGRITY_METADATA 1801/* Metadata integrity verification */ +#define AUDIT_INTEGRITY_STATUS 1802/* Integrity enable status */ +#define AUDIT_INTEGRITY_HASH 1803/* Integrity HASH type */ +#define AUDIT_INTEGRITY_PCR 1804/* PCR invalidation msgs */ #define AUDIT_KERNEL 2000/* Asynchronous audit record. NOT A REQUEST. */
@@ -441,6 +446,8 @@ extern int audit_set_loginuid(struct ta#define audit_get_loginuid (t) ((t)->loginuid)#define audit_get_sessionid (t) ((t)->sessionid)extern void audit_log_task_context (struct audit_buffer *ab); +extern void audit_log_inode_context (struct audit_buffer *ab, +struct inode *inode);extern int __audit_ipc_obj (struct kern_ipc_perm *ipcp);extern int __audit_ipc_set_perm (unsigned long qbytes,uid_t uid,gid_t gid,mode_t mode);extern int audit_bprm (struct linux_binprm *bprm);
@@ -521,6 +528,7 @@ extern int audit_signals;#define audit_get_loginuid (t) (-1)#define audit_get_sessionid (t) (-1)#define audit_log_task_context (b) do { ; } while (0) +#define audit_log_inode_context (b, a) do { ; } while (0)#define audit_ipc_obj (i) ({ 0; })#define audit_ipc_set_perm (q,u,g,m) ({ 0; })#define audit_bprm (p) ({ 0; })
security/integrity/integrity_audit.c new file +73 -0
73 lines not on this page · raw message
Documentation/kernel-parameters.txt +5 -0
@@ -821,6 +821,11 @@ and is between 256 and 4096 characters. inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver Format: <irq> + integrity_audit= [INTEGRITY] + Format: { "0" | "1" } + 0 -- disable integrity auditing messages. + 1 -- enable integrity auditing messages. (Default) + inttest= [IA64] iommu= [x86]
security/integrity/Makefile new file +7 -0
@@ -0,0 +1,7 @@ +# +# Makefile for the kernel integrity code +# + +# Object file lists +obj-$(CONFIG_INTEGRITY) += integrity.o integrity_dummy.o \ + integrity_audit.o integrity_policy.o
security/integrity/Kconfig new file +24 -0
@@ -0,0 +1,24 @@ +# +# Integrity configuration +# + +menu "Integrity options" + +config INTEGRITY + bool "Enable different integrity models" + help + This allows you to choose different integrity modules to be + configured into your kernel. + + If you are unsure how to answer this question, answer N. + +config INTEGRITY_AUDIT + bool "Integrity audit boot parameter" + depends on INTEGRITY + default y + help + This option adds a kernel parameter 'integrity_audit', which + allows integrity auditing to be disabled at boot. If this + option is selected, integrity auditing can be disabled with + 'integrity_audit=0' on the kernel command line. +endmenu
security/Kconfig +2 -0
@@ -4,6 +4,8 @@ menu "Security options" +source security/integrity/Kconfig + config KEYS bool "Enable access key retention support" help
fs/exec.c +4 -0
@@ -47,6 +47,7 @@#include <linux/ptrace.h> #include <linux/mount.h> #include <linux/security.h> +#include <linux/integrity.h> #include <linux/syscalls.h> #include <linux/rmap.h> #include <linux/tsacct_kern.h>
@@ -1192,6 +1193,9 @@ int search_binary_handler(struct linux_b retval =security_bprm_check (bprm);if (retval)return retval; + retval =integrity_bprm_check (bprm); +if (retval) +return retval;/* kernel module loader fixup */ /* so we don't try to load run modprobe in kernel space. */
fs/namei.c +10 -2
@@ -24,6 +24,7 @@#include <linux/fsnotify.h> #include <linux/personality.h> #include <linux/security.h> +#include <linux/integrity.h> #include <linux/syscalls.h> #include <linux/mount.h> #include <linux/audit.h>
@@ -286,7 +287,10 @@ int permission(struct inode *inode, intif (retval)return retval; -return security_inode_permission (inode, mask, nd); + retval =security_inode_permission (inode, mask, nd); +if (retval) +return retval; +return integrity_inode_permission (inode, mask, nd); }/**
@@ -462,6 +466,7 @@ static struct dentry * cached_lookup(strstatic int exec_permission_lite (struct inode *inode,struct nameidata *nd) { +int retval;umode_t mode = inode->i_mode;if (inode->i_op && inode->i_op->permission)
@@ -486,7 +491,10 @@ static int exec_permission_lite(struct ireturn -EACCES ; ok: -return security_inode_permission (inode,MAY_EXEC , nd); + retval =security_inode_permission (inode,MAY_EXEC , nd); +if (retval) +return retval; +return integrity_inode_permission (inode,MAY_EXEC , nd); }/*
mm/mmap.c +4 -0
@@ -20,6 +20,7 @@#include <linux/fs.h> #include <linux/personality.h> #include <linux/security.h> +#include <linux/integrity.h> #include <linux/hugetlb.h> #include <linux/profile.h> #include <linux/module.h>
@@ -1036,6 +1037,9 @@ unsigned long do_mmap_pgoff(struct file error =security_file_mmap (file, reqprot, prot, flags, addr,0 );if (error)return error; + error =integrity_file_mmap (file, reqprot, prot, flags, addr,0 ); +if (error) +return error;return mmap_region (file, addr, len, flags, vm_flags, pgoff, accountable);
security/integrity/integrity_policy.c new file +204 -0
204 lines not on this page · raw message
security/Makefile +4 -0
@@ -19,3 +19,7 @@ obj-$(CONFIG_SECURITY_SMACK) += commonc obj-$(CONFIG_SECURITY_CAPABILITIES) += commoncap.o capability.o obj-$(CONFIG_SECURITY_ROOTPLUG) += commoncap.o root_plug.o obj-$(CONFIG_CGROUP_DEVICE) += device_cgroup.o + +# Object integrity file lists +subdir-$(CONFIG_INTEGRITY) += integrity +obj-$(CONFIG_INTEGRITY) += integrity/built-in.o
security/integrity/integrity.c line 2
Quoted 8 lines · On Fri, 23 May 2008 11:05:33 -0400 Mimi Zohar wrote:
---
Index:linux-2.6.26-rc3-git2/security/integrity/integrity.c
===================================================================security/integrity/integrity.c new file +2
-0@@ -0,0 +1,259 @@ +/** +* register_template
Needs short function description.
Quoted 29 lines
security/integrity/integrity.c new file +29
-0+* @template_name: a pointer to a string containing the template name. +* @template_ops: a pointer to the template functions +* +* Register a set of functions to collect, appraise, store, and display +* a template measurement, and a means to decide whether to do them. +* Unlike integrity modules, any number of templates may be registered. +*/ +int register_template (char *template_name, +struct template_operations *template_ops) +{ +int template_len; +struct template_list_entry *entry; + +if (!template_initialized++) { +INIT_LIST_HEAD (&integrity_templates); +mutex_init (&integrity_templates_mutex); + } + + entry =kmalloc (sizeof (*entry),GFP_ATOMIC ); +INIT_LIST_HEAD (&entry->template); + + template_len =strlen (template_name); +if (template_len >TEMPLATE_NAME_LEN_MAX ) + template_len =TEMPLATE_NAME_LEN_MAX ; +memcpy (entry->template_name, template_name, template_len); + entry->template_name[template_len] ='\0' ; + entry->template_ops = template_ops; + +mutex_lock (&integrity_templates_mutex);
security/integrity/integrity.c line 102
Quoted 10 lines
security/integrity/integrity.c new file +10
-0+list_add_rcu (&entry->template, &integrity_templates); +mutex_unlock (&integrity_templates_mutex); +synchronize_rcu (); + +return 0 ; +} +EXPORT_SYMBOL_GPL (register_template); + +/** +* unregister_template
ditto.
Quoted 16 lines
security/integrity/integrity.c new file +16
-0+* @template_name: a pointer to a string containing the template name. +* +* Unregister the template functions +*/ +int unregister_template (char *template_name) +{ +struct template_list_entry *entry; + +mutex_lock (&integrity_templates_mutex); +list_for_each_entry (entry, &integrity_templates, template) { +if (strncmp (entry->template_name, template_name, +strlen (entry->template_name)) ==0 ) { +list_del_rcu (&entry->template); +mutex_unlock (&integrity_templates_mutex); +synchronize_rcu (); +kfree (entry);
security/integrity/integrity.c line 128
Quoted 10 lines
security/integrity/integrity.c new file +10
-0+return 0 ; + } + } +mutex_unlock (&integrity_templates_mutex); +return -EINVAL ; +} +EXPORT_SYMBOL_GPL (unregister_template); + +/** +* integrity_find_template
ditto.
Quoted 15 lines
security/integrity/integrity.c new file +15
-0+* @template_name: a pointer to a string containing the template name. +* @template_ops: a pointer to the template functions +* +* Find the template functions based on the template name. +*/ +int integrity_find_template (char *template_name, +struct template_operations **template_ops) +{ +struct template_list_entry *entry; + +rcu_read_lock (); +list_for_each_entry_rcu (entry, &integrity_templates, template) { +if (strncmp (entry->template_name, template_name, +strlen (entry->template_name)) ==0 ) { + *template_ops = entry->template_ops;
security/integrity/integrity.c line 153
Quoted 10 lines
security/integrity/integrity.c new file +10
-0+rcu_read_unlock (); +return 0 ; + } + } +rcu_read_unlock (); +return 1 ; +} +EXPORT_SYMBOL_GPL (integrity_find_template); + +/* Integrity template API */
Please add kernel-doc for exported symbols (like this one and more below).
Quoted 112 lines
security/integrity/integrity.c new file +106
-0+int integrity_collect_measurement (char *template_name,void *data) +{ +struct template_operations *template_ops; + +if (integrity_find_template (template_name, &template_ops) ==0 ) + template_ops->collect_measurement (data); +return 0 ; +} +EXPORT_SYMBOL_GPL (integrity_collect_measurement); + +int integrity_appraise_measurement (char *template_name,void *data) +{ +struct template_operations *template_ops; + +if (integrity_find_template (template_name, &template_ops) ==0 ) +return template_ops->appraise_measurement (data); +return 1 ; +} +EXPORT_SYMBOL_GPL (integrity_appraise_measurement); + +void integrity_store_measurement (char *template_name,void *data) +{ +struct template_operations *template_ops; + +if (integrity_find_template (template_name, &template_ops) ==0 ) + template_ops->store_measurement (data); +return ; +} +EXPORT_SYMBOL_GPL (integrity_store_measurement); + +int integrity_must_measure (char *template_name,void *data) +{ +struct template_operations *template_ops; + +if (integrity_find_template (template_name, &template_ops) ==0 ) +return template_ops->must_measure (data); +return 1 ; +} +EXPORT_SYMBOL_GPL (integrity_must_measure); + +/* Integrity Hooks */ +int integrity_bprm_check (struct linux_binprm *bprm) +{ +return integrity_ops->bprm_check_integrity (bprm); +} + +int integrity_inode_setxattr (struct dentry *dentry,const char *name, +const void *value,size_t size,int flags) +{ +if (unlikely (IS_PRIVATE (dentry->d_inode))) +return 0 ; +return integrity_ops->inode_setxattr (dentry, name, value, size, flags); +} + +void integrity_inode_post_setxattr (struct dentry *dentry,const char *name) +{ +if (unlikely (IS_PRIVATE (dentry->d_inode))) +return ; + integrity_ops->inode_post_setxattr (dentry, name); +} + +int integrity_inode_alloc (struct inode *inode) +{ +return integrity_ops->inode_alloc_integrity (inode); +} + +void integrity_inode_free (struct inode *inode) +{ + integrity_ops->inode_free_integrity (inode); +} + +int integrity_inode_permission (struct inode *inode,int mask, +struct nameidata *nd) +{ +return integrity_ops->inode_permission (inode, mask, nd); +} + +void integrity_inode_init_integrity (struct inode *inode,struct inode *dir, +char **name,void **value,size_t *len) +{ +if (unlikely (IS_PRIVATE (inode))) +return ; + integrity_ops->inode_init_integrity (inode, dir, name, value, len); +return ; +} +EXPORT_SYMBOL (integrity_inode_init_integrity); + +void integrity_file_free (struct file *file) +{ + integrity_ops->file_free_integrity (file); +} + +int integrity_file_mmap (struct file *file,unsigned long reqprot, +unsigned long prot,unsigned long flags, +unsigned long addr,unsigned long addr_only) +{ +return integrity_ops->file_mmap (file, reqprot, prot, flags, addr, + addr_only); +} + +void integrity_d_instantiate (struct dentry *dentry,struct inode *inode) +{ +if (unlikely (inode &&IS_PRIVATE (inode))) +return ; + integrity_ops->d_instantiate (dentry, inode); +}Index:
linux-2.6.26-rc3-git2/Documentation/kernel-parameters.txt
===================================================================
Documentation/kernel-parameters.txt and is between 256 and 4096 characters. · line 824
Quoted 4 lines
Documentation/kernel-parameters.txt +1
-0@@ -821,6 +821,11 @@ and is between 256 and 4096 characters. inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver Format: <irq> + integrity_audit= [INTEGRITY]
Where is the INTEGRITY tag defined?
Quoted 13 lines
Documentation/kernel-parameters.txt +4
-0+ Format: { "0" | "1" } + 0 -- disable integrity auditing messages. + 1 -- enable integrity auditing messages. (Default) + inttest= [IA64] iommu= [x86]Index:
linux-2.6.26-rc3-git2/security/integrity/integrity_policy.c
===================================================================
security/integrity/integrity_policy.c line 7
Quoted 7 lines
security/integrity/integrity_policy.c new file +7
-0@@ -0,0 +1,204 @@ + +/** +* integrity_measure_rules - determine whether an inode matches the given rule. +* @rule - a pointer to a rule +* @inode - a pointer to an inode +* @func - LIM hook identifier +* @mask - requested action (MAY_READ | MAY_WRITE | MAY_APPEND | MAY_EXEC)
kernel-doc notation for parameters is (e.g.)
* @rule: a pointer to a rule
Quoted 30 lines
security/integrity/integrity_policy.c new file +30
-0+* +* Returns 1 on rule match, 0 on failure. +*/ +static int integrity_measure_rules (struct integrity_measure_rule_entry *rule, +struct inode *inode,enum lim_hooks func, +int mask) +{ +int result =1 ; + +if (result && (rule->func !=0 )) { +if (rule->func != func) + result =0 ; + } +if (result && (rule->mask !=0 )) { +if (rule->mask != mask) + result =0 ; + } +if (result && rule->lsm_subj_rule) { +struct task_struct *tsk = current; +u32 sid; + +security_task_getsecid (tsk, &sid); + result =security_filter_rule_match (sid,AUDIT_SUBJ_USER , +AUDIT_EQUAL , + rule->lsm_subj_rule,NULL ); + } +if (result && rule->lsm_obj_rule) { +u32 osid; + +security_inode_getsecid (inode, &osid);
security/integrity/integrity_policy.c line 88
Quoted 10 lines
security/integrity/integrity_policy.c new file +10
-0+ result =security_filter_rule_match (osid,AUDIT_OBJ_USER , +AUDIT_EQUAL , + rule->lsm_obj_rule,NULL ); + } +return result; +} + +/** +* integrity_measure_policy - base measure decision on: subj, obj, LIM hook, +* and mask
One line for short description.
* @inode: pointer to an inode
Quoted 18 lines
security/integrity/integrity_policy.c new file +18
-0+* @inode - pointer to an inode +* @func - LIM hook identifier +* @mask - requested action (MAY_READ | MAY_WRITE | MAY_APPEND | MAY_EXEC) +* +* Returns 1 on rule match, 0 on failure. +*/ +int integrity_measure_policy (struct inode *inode,enum lim_hooks func,int mask) +{ +struct integrity_measure_rule_entry *entry; +int rc =0 ; + +rcu_read_lock (); +list_for_each_entry_rcu (entry, integrity_measure, list) { + rc =integrity_measure_rules (entry, inode, func, mask); +if (rc) { +rcu_read_unlock (); +return rc; + }
security/integrity/integrity_policy.c line 149
Quoted 10 lines
security/integrity/integrity_policy.c new file +10
-0+ } +rcu_read_unlock (); +return rc; +} + + + +/** +* integrity_measure_rule_add - add integrity measure rules +* @subj - pointer to an LSM subject value
* @subj: etc etc etc
Quoted 45 lines
security/integrity/integrity_policy.c new file +45
-0+* @obj - pointer to an LSM object value +* @func - LIM hook identifier +* @mask - requested action (MAY_READ | MAY_WRITE | MAY_APPEND | MAY_EXEC) +* +* Returns 0 on success, an error code on failure. +*/ +int integrity_measure_rule_add (char *subj,char *obj,char *func,char *mask) +{ +struct integrity_measure_rule_entry *entry; +int result =0 ; + +/* Prevent installed policy from changing */ +if (integrity_measure != &measure_default_rules) { +integrity_audit_msg (AUDIT_INTEGRITY_STATUS ,NULL , +NULL ,"policy_update" ,"already exists" ,1 ); +return -EACCES ; + } + + entry =kzalloc (sizeof (*entry),GFP_ATOMIC ); +INIT_LIST_HEAD (&entry->list); +if (!result && subj) + result =security_filter_rule_init (AUDIT_SUBJ_USER ,AUDIT_EQUAL , + subj, &entry->lsm_subj_rule); +if (!result && obj) + result =security_filter_rule_init (AUDIT_OBJ_USER ,AUDIT_EQUAL , + obj, &entry->lsm_obj_rule); +if (!result && func) { +if (strcmp (func,"INODE_PERMISSION" ) ==0 ) + entry->func =INODE_PERMISSION ; +else if (strcmp (func,"FILE_MMAP" ) ==0 ) + entry->func =FILE_MMAP ; +else if (strcmp (func,"BPRM_CHECK" ) ==0 ) + entry->func =BPRM_CHECK ; +else + result = -EINVAL ; + } +if (!result && mask) { +if (strcmp (mask,"MAY_EXEC" ) ==0 ) + entry->mask =MAY_EXEC ; +else if (strcmp (mask,"MAY_WRITE" ) ==0 ) + entry->mask =MAY_WRITE ; +else if (strcmp (mask,"MAY_READ" ) ==0 ) + entry->mask =MAY_READ ; +else if (strcmp (mask,"MAY_APPEND" ) ==0 ) + entry->mask =MAY_APPEND ;
security/integrity/integrity_policy.c line 204
Quoted 10 lines
security/integrity/integrity_policy.c new file +10
-0+else + result = -EINVAL ; + } +if (!result) { +mutex_lock (&integrity_measure_mutex); +list_add_tail (&entry->list, &measure_policy_rules); +mutex_unlock (&integrity_measure_mutex); + } +return result; +}
---
~Randy
This is a request for comments for a redesign of the integrity patches. The new version addresses a number of issues, including
- generalizing the measurement API beyond just inode measurements.
- separation of the measurement into distinct collection, appraisal, and commitment phases, for greater flexibility.
- redesigning the interactions with LSM modules (Selinux and Smack) to be acceptable to those groups.
Extended Verification Module(EVM) and the Integrity Measurement Architecture(IMA) were originally implemented as an LSM module. Based on discussions on the LSM mailing list, a decision was made that the LSM hooks should only be used to enforce mandatory access control decisions and a new set of hooks should be defined specifically for integrity.
EVM/IMA was limited to verifying and measuring a file's (i.e. an inode) integrity and the metadata associated with it. Current research is looking into other types of integrity measurements. (i.e. "Linux kernel integrity measurement using contextual inspection", by Peter A. Loscocco, Perry W. Wilson, J. Aaron Pendergrass, C. Durward McDonell,
http://doi.acm.org/10.1145/1314354.1314362) As a result, a requirement of the new integrity framework is support for different types of integrity measurements.
Our work has shown that requests for integrity verification/measurement needs to be based on knowledge of the filesystem, requiring the system to either be labeled with integrity data or depend on the existent LSM security labels. The previous set of integrity patches modified the LSM modules to be integrity context aware, meaning that the LSM modules made integrity data/metadata verification and measurement API calls based on an understanding of the LSM security labels. Both of the LSM maintainers felt that the changes were too intrusive and that integrity enforcement should be made by the integrity provider, not the LSM module.
To address these concerns, Stephen Smalley suggested using the security_audit_rule_match(), renamed to security_filter_rule_match(), to define LSM specific integrity measurement policy rules, in lieu of modifying the LSM modules. In the current set of patches, the integrity API calls can be made either by the integrity provider (i.e. IMA), based on an LSM specific integrity policy, or by an integrity context aware LSM module.
This patch provides an integrity framework(api and hooks), placement of the integrity hooks in the appropriate places in the fs directory, and a dummy provider for an integrity service. Collecting, appraising, and storing of inode and other types of integrity data is supported. Multiple integrity templates, which implement the integrity API, may register themselves. For now, only a single integrity provider can register itself for the integrity hooks. (Support for multiple providers registering themselves for the integrity hooks would require some form of stacking.)
The ten integrity hooks are:
inode_setxattr, inode_post_setxattr, inode_alloc_integrity, inode_free_integrity, inode_init_integrity, file_free_integrity, d_instantiate, bprm_check_integrity, file_mmap, inode_permission
The five integrity API calls provided are:
integrity_must_measure, integrity_collect_measurement, integrity_appraise_measurement, integrity_store_measurement, and integrity_display_template.
The type of integrity data being collected, appraised, stored, or displayed is template dependent.
(Details on the calls and their exact arguments are in linux/integrity.h, included in the patch.)
---
include/linux/integrity.h new file +261 -0
@@ -0,0 +1,261 @@ +/* +* integrity.h +* +* Copyright (C) 2005,2006,2007,2008 IBM Corporation +* Author: Mimi Zohar <[email protected]> +* +* This program is free software; you can redistribute it and/or modify +* it under the terms of the GNU General Public License as published by +* the Free Software Foundation, version 2 of the License. +*/ + +#ifndef _LINUX_INTEGRITY_H +#define _LINUX_INTEGRITY_H + +#include <linux/fs.h> +#include <linux/audit.h> + +#ifdef CONFIG_INTEGRITY +void integrity_audit_msg (int audit_msgno,struct inode *inode, +const unsigned char *fname,char *op, +char *cause,int result); + +enum lim_hooks {INODE_PERMISSION =1 ,FILE_MMAP ,BPRM_CHECK }; +extern int integrity_measure_policy (struct inode *inode,enum lim_hooks func, +int mask); +extern int integrity_measure_rule_add (char *,char *,char *,char *); +extern void integrity_measure_policy_init (void ); +extern void integrity_measure_policy_complete (void ); + +/* +* Integrity API calls: +* +* @collect_measurement: +* Collect template specific measurement data. +* @data contains template specific data used for collecting the +* measurement. +* Return 0 if operation was successful. +* +* @appraise_measurement: +* Appraise the integrity of the template specific measurement data. +* @data contains template specific data used for appraising the +* measurement. +* Return 0 if operation was successful. +* +* @store_measurement: +* Store the template specific data. +* @data contains template specific data used for storing the +* measurement. +* +* @must_measure: +* Measurement decision based on an integrity policy. +* @data contains template specific data used for making policy +* decision. +* Return 0 if operation was successful. +* +* @display_template: +* Display template specific data. +* +*/ + +enum integrity_show_type {INTEGRITY_SHOW_BINARY ,INTEGRITY_SHOW_ASCII }; + +struct template_operations { +int (*collect_measurement)(void *); +int (*appraise_measurement)(void *); +void (*store_measurement)(void *); +int (*must_measure)(void *); +void (*display_template)(struct seq_file *m,void *, +enum integrity_show_type ); +}; +extern int register_template (char *template_name, +struct template_operations *ops); +extern int unregister_template (char *template_name); +extern int integrity_find_template (char *,struct template_operations **ops); + +/* +* Integrity hooks: +* +* @bprm_check_integrity: +* This hook mediates the point when a search for a binary handler will +* begin. At this point, the OS protects against an executable file, +* already open for write, from being executed; and an executable file +* already open for execute, from being modified. So we can be certain +* that any measurements(collect, appraise, store) done here are of +* the file being executed. +* @bprm contains the linux_binprm structure. +* Return 0 if the hook is successful and permission is granted. +* +* @inode_setxattr: +* Check permission before permitting an integrity extended attribute +* to be set. +* @value identified by @name for @dentry. +* +* @inode_post_setxattr: +* Update inode integrity xattr after successful setxattr operation. +* identified by @name for @dentry. +* +* @inode_alloc_integrity: +* Allocate and attach an integrity structure to @inode->i_integrity. The +* i_integrity field is initialized to NULL when the inode structure is +* allocated. +* @inode contains the inode structure. +* Return 0 if operation was successful. +* +* @inode_free_integrity: +* @inode contains the inode structure. +* Deallocate the inode integrity structure and set @inode->i_integrity to +* NULL. +* +* @inode_init_integrity: +* Create inode integrity xattr for a new inode based on the new security +* xattr information. +* @inode contains the inode structure of the newly created inode. +* @dir contains the inode structure of the parent directory. +* @name contains the security xattr name suffix. +* @value contains the security attribute value. +* @len contains the length of the security attribute value. +* +* @inode_permission: +* This hook is called by the existing Linux permission function, when +* a file is opened (as well as many other operations). At this point, +* measurements of files open for read(collect, appraise, store) can +* be made. +* @inode contains the inode structure to check. +* @mask contains the permission mask. +* @nd contains the nameidata (may be NULL). +* +* @file_free_integrity: +* Update the integrity xattr value as necessary. +* *file contains the file structure being closed. +* +* @file_mmap : +* Measurement(collect, appraise, store) of files mmaped for EXEC, +* could be measured at this point. +* @file contains the file structure for file to map (may be NULL). +* @reqprot contains the protection requested by the application. +* @prot contains the protection that will be applied by the kernel. +* @flags contains the operational flags. +* Return 0 if permission is granted. +* +* @d_instantiate: +* Initialize the integrity structure of an inode for a dentry. +* @dentry to complete. +* @inode to attach to this dentry. +* +*/ + +struct integrity_operations { +int (*bprm_check_integrity) (struct linux_binprm *bprm); +int (*inode_setxattr) (struct dentry *dentry,const char *name, +const void *value,size_t size,int flags); +void (*inode_post_setxattr) (struct dentry *dentry,const char *name); +int (*inode_alloc_integrity) (struct inode *inode); +void (*inode_free_integrity) (struct inode *inode); +void (*inode_init_integrity) (struct inode *inode,struct inode *dir, +char **name,void **value,size_t *len); +int (*inode_permission) (struct inode *inode,int mask, +struct nameidata *nd); +void (*file_free_integrity) (struct file *file); +int (*file_mmap) (struct file *file, +unsigned long reqprot,unsigned long prot, +unsigned long flags,unsigned long addr, +unsigned long addr_only); +void (*d_instantiate) (struct dentry *dentry,struct inode *inode); +}; +extern int register_integrity (struct integrity_operations *ops); +extern int unregister_integrity (struct integrity_operations *ops); + +/* global variables */ +extern struct integrity_operations *integrity_ops; + + +int integrity_collect_measurement (char *template_name,void *data); +int integrity_appraise_measurement (char *template_name,void *data); +int integrity_must_measure (char *template_name,void *data); +void integrity_store_measurement (char *template_name,void *data); + +int integrity_bprm_check (struct linux_binprm *bprm); +int integrity_inode_setxattr (struct dentry *dentry,const char *name, +const void *value,size_t size,int flags); +void integrity_inode_post_setxattr (struct dentry *dentry,const char *name); +int integrity_inode_alloc (struct inode *inode); +void integrity_inode_free (struct inode *inode); +void integrity_inode_init_integrity (struct inode *inode,struct inode *dir, +char **name,void **value,size_t *len); +int integrity_inode_permission (struct inode *inode,int mask, +struct nameidata *nd); +void integrity_file_free (struct file *file); +int integrity_file_mmap (struct file *file, +unsigned long reqprot,unsigned long prot, +unsigned long flags,unsigned long addr, +unsigned long addr_only); +void integrity_d_instantiate (struct dentry *dentry,struct inode *inode); + +#else + +static inline int integrity_bprm_check (struct linux_binprm *bprm) +{ +return 0 ; +} + +static inline int integrity_inode_setxattr (struct dentry *dentry, +const char *name,const void *value, +size_t size,int flags) +{ +return 0 ; +} + +static inline void integrity_inode_post_setxattr (struct dentry *dentry, +const char *name) +{ } + +static inline int integrity_inode_alloc (struct inode *inode) +{ +return 0 ; +} + +static inline void integrity_inode_free (struct inode *inode) +{ } + +static inline int integrity_inode_init_integrity (struct inode *inode, +struct inode *dir, +char **name, +void **value, +size_t *len) +{ +return -EOPNOTSUPP ; +} + +static inline int integrity_inode_permission (struct inode *inode,int mask, +struct nameidata *nd) +{ +return 0 ; +} + +static inline int integrity_file_permission (struct file *file,int mask) +{ +return 0 ; +} + +static inline void integrity_file_free (struct file *file) +{ +return ; +} + +static inline int integrity_file_mmap (struct file *file, +unsigned long reqprot,unsigned long prot, +unsigned long flags,unsigned long addr, +unsigned long addr_only) +{ +return 0 ; +} + +static inline void integrity_d_instantiate (struct dentry *dentry, +struct inode *inode) +{ +return ; +} + +#endif +#endif
security/integrity/integrity.c new file +307 -0
307 lines not on this page · raw message
security/integrity/integrity_dummy.c new file +112 -0
112 lines not on this page · raw message
security/integrity/integrity_dummy.h new file +13 -0
@@ -0,0 +1,13 @@ +/* +* integrity_dummy.h +* +* Copyright (C) 2005,2006,2007 IBM Corporation +* Author: Mimi Zohar <[email protected]> +* +* This program is free software; you can redistribute it and/or modify +* it under the terms of the GNU General Public License as published by +* the Free Software Foundation, version 2 of the License. +*/ + +extern struct integrity_operations dummy_integrity_ops; +extern void integrity_fixup_ops (struct integrity_operations *ops);
fs/dcache.c +6 -0
@@ -28,6 +28,7 @@#include <linux/file.h> #include <asm/uaccess.h> #include <linux/security.h> +#include <linux/integrity.h> #include <linux/seqlock.h> #include <linux/swap.h> #include <linux/bootmem.h>
@@ -972,6 +973,7 @@ void d_instantiate(struct dentry *entry, entry->d_inode = inode;fsnotify_d_instantiate (entry, inode);spin_unlock (&dcache_lock); +integrity_d_instantiate (entry, inode);security_d_instantiate (entry, inode); }
@@ -1036,6 +1038,7 @@ struct dentry *d_instantiate_unique(struspin_unlock (&dcache_lock);if (!result) { +integrity_d_instantiate (entry, inode);security_d_instantiate (entry, inode);return NULL ; }
@@ -1173,6 +1176,7 @@ struct dentry *d_splice_alias(struct inoBUG_ON (!(new->d_flags &DCACHE_DISCONNECTED ));fsnotify_d_instantiate (new, inode);spin_unlock (&dcache_lock); +integrity_d_instantiate (new, inode);security_d_instantiate (new, inode);d_rehash (dentry);d_move (new, dentry);
@@ -1183,6 +1187,7 @@ struct dentry *d_splice_alias(struct ino dentry->d_inode = inode;fsnotify_d_instantiate (dentry, inode);spin_unlock (&dcache_lock); +integrity_d_instantiate (dentry, inode);security_d_instantiate (dentry, inode);d_rehash (dentry); }
@@ -1733,6 +1738,7 @@ found:spin_unlock (&dcache_lock); out_nolock:if (actual == dentry) { +integrity_d_instantiate (dentry, inode);security_d_instantiate (dentry, inode);return NULL ; }
fs/ext3/xattr_security.c +8 -0
@@ -9,6 +9,7 @@#include <linux/ext3_jbd.h> #include <linux/ext3_fs.h> #include <linux/security.h> +#include <linux/integrity.h> #include "xattr.h" static size_t
@@ -57,12 +58,19 @@ ext3_init_security(handle_t *handle, str err =security_inode_init_security (inode, dir, &name, &value, &len);if (err) { +/* Even if creation of the security xattr fails, must +* indicate this is a new inode. */ +integrity_inode_init_integrity (inode, dir,NULL ,NULL ,NULL );if (err == -EOPNOTSUPP )return 0 ;return err; } err =ext3_xattr_set_handle (handle, inode,EXT3_XATTR_INDEX_SECURITY , name, value, len,0 ); + +integrity_inode_init_integrity (inode, dir, &name, &value, &len); + err =ext3_xattr_set_handle (handle, inode,EXT3_XATTR_INDEX_SECURITY , + name, value, len,0 );kfree (name);kfree (value);return err;
fs/file_table.c +3 -0
@@ -13,6 +13,7 @@#include <linux/module.h> #include <linux/fs.h> #include <linux/security.h> +#include <linux/integrity.h> #include <linux/eventpoll.h> #include <linux/rcupdate.h> #include <linux/mount.h>
@@ -272,6 +273,7 @@ void __fput(struct file *file)if (file->f_op && file->f_op->release) file->f_op->release (inode, file);security_file_free (file); +integrity_file_free (file);if (unlikely (S_ISCHR (inode->i_mode) && inode->i_cdev !=NULL ))cdev_put (inode->i_cdev);fops_put (file->f_op);
@@ -343,6 +345,7 @@ void put_filp(struct file *file) {if (atomic_dec_and_test (&file->f_count)) {security_file_free (file); +integrity_file_free (file);file_kill (file);file_free (file); }
fs/inode.c +10 -0
@@ -17,6 +17,7 @@#include <linux/hash.h> #include <linux/swap.h> #include <linux/security.h> +#include <linux/integrity.h> #include <linux/pagemap.h> #include <linux/cdev.h> #include <linux/bootmem.h>
@@ -160,6 +161,14 @@ static struct inode *alloc_inode(structinit_rwsem (&inode->i_alloc_sem);lockdep_set_class (&inode->i_alloc_sem, &sb->s_type->i_alloc_sem_key); +if (integrity_inode_alloc (inode)) { +if (inode->i_sb->s_op->destroy_inode) + inode->i_sb->s_op->destroy_inode (inode); +else +kmem_cache_free (inode_cachep, (inode)); +return NULL ; + } + mapping->a_ops = &empty_aops; mapping->host = inode; mapping->flags =0 ;
@@ -190,6 +199,7 @@ void destroy_inode(struct inode *inode) {BUG_ON (inode_has_buffers (inode));security_inode_free (inode); +integrity_inode_free (inode);if (inode->i_sb->s_op->destroy_inode) inode->i_sb->s_op->destroy_inode (inode);else
fs/xattr.c +9 -0
@@ -14,6 +14,7 @@#include <linux/mount.h> #include <linux/namei.h> #include <linux/security.h> +#include <linux/integrity.h> #include <linux/syscalls.h> #include <linux/module.h> #include <linux/fsnotify.h>
@@ -81,11 +82,17 @@ vfs_setxattr(struct dentry *dentry, cons error =security_inode_setxattr (dentry, name, value, size, flags);if (error) goto out; + + error =integrity_inode_setxattr (dentry, name, value, size, flags); +if (error) + goto out; + error = -EOPNOTSUPP ;if (inode->i_op->setxattr) { error = inode->i_op->setxattr (dentry, name, value, size, flags);if (!error) {fsnotify_xattr (dentry); +integrity_inode_post_setxattr (dentry, name);security_inode_post_setxattr (dentry, name, value, size, flags); }
@@ -205,6 +212,8 @@ vfs_removexattr(struct dentry *dentry, cmutex_lock (&inode->i_mutex); error = inode->i_op->removexattr (dentry, name); +if (!error) +integrity_inode_post_setxattr (dentry, name);mutex_unlock (&inode->i_mutex);if (!error)
include/linux/fs.h +3 -0
@@ -653,6 +653,9 @@ struct inode {#ifdef CONFIG_SECURITY void *i_security;#endif +#ifdef CONFIG_INTEGRITY +void *i_integrity; +#endif void *i_private;/* fs or device private pointer */ };
include/linux/audit.h +8 -0
@@ -123,6 +123,11 @@#define AUDIT_LAST_KERN_ANOM_MSG 1799#define AUDIT_ANOM_PROMISCUOUS 1700/* Device changed promiscuous mode */ #define AUDIT_ANOM_ABEND 1701/* Process ended abnormally */ +#define AUDIT_INTEGRITY_DATA 1800/* Data integrity verification */ +#define AUDIT_INTEGRITY_METADATA 1801/* Metadata integrity verification */ +#define AUDIT_INTEGRITY_STATUS 1802/* Integrity enable status */ +#define AUDIT_INTEGRITY_HASH 1803/* Integrity HASH type */ +#define AUDIT_INTEGRITY_PCR 1804/* PCR invalidation msgs */ #define AUDIT_KERNEL 2000/* Asynchronous audit record. NOT A REQUEST. */
@@ -441,6 +446,8 @@ extern int audit_set_loginuid(struct ta#define audit_get_loginuid (t) ((t)->loginuid)#define audit_get_sessionid (t) ((t)->sessionid)extern void audit_log_task_context (struct audit_buffer *ab); +extern void audit_log_inode_context (struct audit_buffer *ab, +struct inode *inode);extern int __audit_ipc_obj (struct kern_ipc_perm *ipcp);extern int __audit_ipc_set_perm (unsigned long qbytes,uid_t uid,gid_t gid,mode_t mode);extern int audit_bprm (struct linux_binprm *bprm);
@@ -521,6 +528,7 @@ extern int audit_signals;#define audit_get_loginuid (t) (-1)#define audit_get_sessionid (t) (-1)#define audit_log_task_context (b) do { ; } while (0) +#define audit_log_inode_context (b, a) do { ; } while (0)#define audit_ipc_obj (i) ({ 0; })#define audit_ipc_set_perm (q,u,g,m) ({ 0; })#define audit_bprm (p) ({ 0; })
security/integrity/integrity_audit.c new file +73 -0
73 lines not on this page · raw message
Documentation/kernel-parameters.txt +6 -0
@@ -44,6 +44,7 @@ parameter is applicable: FB The frame buffer device is enabled. HW Appropriate hardware is enabled. IA-64 IA-64 architecture is enabled. + INTEGRITY Integrity support is enabled. IOSCHED More than one I/O scheduler is enabled. IP_PNP IP DHCP, BOOTP, or RARP is enabled. ISAPNP ISA PnP code is enabled.
@@ -821,6 +822,11 @@ and is between 256 and 4096 characters. inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver Format: <irq> + integrity_audit= [INTEGRITY] + Format: { "0" | "1" } + 0 -- disable integrity auditing messages. + 1 -- enable integrity auditing messages. (Default) + inttest= [IA64] iommu= [x86]
security/integrity/Makefile new file +7 -0
@@ -0,0 +1,7 @@ +# +# Makefile for the kernel integrity code +# + +# Object file lists +obj-$(CONFIG_INTEGRITY) += integrity.o integrity_dummy.o \ + integrity_audit.o integrity_policy.o
security/integrity/Kconfig new file +24 -0
@@ -0,0 +1,24 @@ +# +# Integrity configuration +# + +menu "Integrity options" + +config INTEGRITY + bool "Enable different integrity models" + help + This allows you to choose different integrity modules to be + configured into your kernel. + + If you are unsure how to answer this question, answer N. + +config INTEGRITY_AUDIT + bool "Integrity audit boot parameter" + depends on INTEGRITY + default y + help + This option adds a kernel parameter 'integrity_audit', which + allows integrity auditing to be disabled at boot. If this + option is selected, integrity auditing can be disabled with + 'integrity_audit=0' on the kernel command line. +endmenu
security/Kconfig +2 -0
@@ -4,6 +4,8 @@ menu "Security options" +source security/integrity/Kconfig + config KEYS bool "Enable access key retention support" help
fs/exec.c +4 -0
@@ -47,6 +47,7 @@#include <linux/ptrace.h> #include <linux/mount.h> #include <linux/security.h> +#include <linux/integrity.h> #include <linux/syscalls.h> #include <linux/rmap.h> #include <linux/tsacct_kern.h>
@@ -1192,6 +1193,9 @@ int search_binary_handler(struct linux_b retval =security_bprm_check (bprm);if (retval)return retval; + retval =integrity_bprm_check (bprm); +if (retval) +return retval;/* kernel module loader fixup */ /* so we don't try to load run modprobe in kernel space. */
fs/namei.c +10 -2
@@ -24,6 +24,7 @@#include <linux/fsnotify.h> #include <linux/personality.h> #include <linux/security.h> +#include <linux/integrity.h> #include <linux/syscalls.h> #include <linux/mount.h> #include <linux/audit.h>
@@ -286,7 +287,10 @@ int permission(struct inode *inode, intif (retval)return retval; -return security_inode_permission (inode, mask, nd); + retval =security_inode_permission (inode, mask, nd); +if (retval) +return retval; +return integrity_inode_permission (inode, mask, nd); }/**
@@ -462,6 +466,7 @@ static struct dentry * cached_lookup(strstatic int exec_permission_lite (struct inode *inode,struct nameidata *nd) { +int retval;umode_t mode = inode->i_mode;if (inode->i_op && inode->i_op->permission)
@@ -486,7 +491,10 @@ static int exec_permission_lite(struct ireturn -EACCES ; ok: -return security_inode_permission (inode,MAY_EXEC , nd); + retval =security_inode_permission (inode,MAY_EXEC , nd); +if (retval) +return retval; +return integrity_inode_permission (inode,MAY_EXEC , nd); }/*
mm/mmap.c +4 -0
@@ -20,6 +20,7 @@#include <linux/fs.h> #include <linux/personality.h> #include <linux/security.h> +#include <linux/integrity.h> #include <linux/hugetlb.h> #include <linux/profile.h> #include <linux/module.h>
@@ -1036,6 +1037,9 @@ unsigned long do_mmap_pgoff(struct file error =security_file_mmap (file, reqprot, prot, flags, addr,0 );if (error)return error; + error =integrity_file_mmap (file, reqprot, prot, flags, addr,0 ); +if (error) +return error;return mmap_region (file, addr, len, flags, vm_flags, pgoff, accountable);
security/integrity/integrity_policy.c new file +204 -0
204 lines not on this page · raw message
security/Makefile +4 -0
@@ -19,3 +19,7 @@ obj-$(CONFIG_SECURITY_SMACK) += commonc obj-$(CONFIG_SECURITY_CAPABILITIES) += commoncap.o capability.o obj-$(CONFIG_SECURITY_ROOTPLUG) += commoncap.o root_plug.o obj-$(CONFIG_CGROUP_DEVICE) += device_cgroup.o + +# Object integrity file lists +subdir-$(CONFIG_INTEGRITY) += integrity +obj-$(CONFIG_INTEGRITY) += integrity/built-in.o
Andrew Morton
maintainerQuoted 23 lines · On Fri, 23 May 2008 11:05:33 -0400 Mimi Zohar <[email protected]> wrote:
This is a request for comments for a redesign of the integrity patches.
...
security/integrity/integrity.c new file +19
-0+int register_template (char *template_name, +struct template_operations *template_ops) +{ +int template_len; +struct template_list_entry *entry; + +if (!template_initialized++) { +INIT_LIST_HEAD (&integrity_templates); +mutex_init (&integrity_templates_mutex); + } + + entry =kmalloc (sizeof (*entry),GFP_ATOMIC ); +INIT_LIST_HEAD (&entry->template); + + template_len =strlen (template_name); +if (template_len >TEMPLATE_NAME_LEN_MAX ) + template_len =TEMPLATE_NAME_LEN_MAX ; +memcpy (entry->template_name, template_name, template_len); + entry->template_name[template_len] ='\0' ;
security/integrity/integrity.c line 99
Quoted 10 lines
security/integrity/integrity.c new file +10
-0+ entry->template_ops = template_ops; + +mutex_lock (&integrity_templates_mutex); +list_add_rcu (&entry->template, &integrity_templates); +mutex_unlock (&integrity_templates_mutex); +synchronize_rcu (); + +return 0 ; +} +EXPORT_SYMBOL_GPL (register_template);
Should be integrity_register_template()?
Quoted 15 lines
security/integrity/integrity.c new file +15
-0+/** +* unregister_template +* @template_name: a pointer to a string containing the template name. +* +* Unregister the template functions +*/ +int unregister_template (char *template_name) +{ +struct template_list_entry *entry; + +mutex_lock (&integrity_templates_mutex); +list_for_each_entry (entry, &integrity_templates, template) { +if (strncmp (entry->template_name, template_name, +strlen (entry->template_name)) ==0 ) { +list_del_rcu (&entry->template);
security/integrity/integrity.c line 125
Quoted 10 lines
security/integrity/integrity.c new file +10
-0+mutex_unlock (&integrity_templates_mutex); +synchronize_rcu (); +kfree (entry); +return 0 ; + } + } +mutex_unlock (&integrity_templates_mutex); +return -EINVAL ; +} +EXPORT_SYMBOL_GPL (unregister_template);
Similarly.
Quoted 15 lines
security/integrity/integrity.c new file +15
-0+/** +* integrity_find_template +* @template_name: a pointer to a string containing the template name. +* @template_ops: a pointer to the template functions +* +* Find the template functions based on the template name. +*/ +int integrity_find_template (char *template_name, +struct template_operations **template_ops) +{ +struct template_list_entry *entry; + +rcu_read_lock (); +list_for_each_entry_rcu (entry, &integrity_templates, template) { + if (strncmp(entry->template_name, template_name,
security/integrity/integrity.c line 151
Quoted 10 lines
security/integrity/integrity.c new file +10
-0+ strlen(entry->template_name)) == 0) { + *template_ops = entry->template_ops; +rcu_read_unlock (); +return 0 ; + } + } +rcu_read_unlock (); +return 1 ; +} +EXPORT_SYMBOL_GPL (integrity_find_template);
Strange locking. We take the rcu_read_lock then locate a pointer to an object then drop the lock, returning that pointer while doing nothing to ensure the stability of the returned object?
security/integrity/integrity_dummy.c line 97
Quoted 11 lines
...
security/integrity/integrity_dummy.c new file +8
-0+#define set_to_dummy_if_null (ops, function) \ + do { \ + if (!ops->function) { \ + ops->function = dummy_##function; \ + printk(KERN_INFO "Had to override the " #function \ + " integrity operation with the dummy one.\n");\ + } \ + } while (0)
hrm.
Probably the message should identify where it came from? "integrity:
had to override ..."
Quoted 28 lines
...
fs/dcache.c +3
-0@@ -1036,6 +1038,7 @@ struct dentry *d_instantiate_unique(struspin_unlock (&dcache_lock);if (!result) { +integrity_d_instantiate (entry, inode);security_d_instantiate (entry, inode);return NULL ; }@@ -1173,6 +1176,7 @@ struct dentry *d_splice_alias(struct inoBUG_ON (!(new->d_flags &DCACHE_DISCONNECTED ));fsnotify_d_instantiate (new, inode);spin_unlock (&dcache_lock); +integrity_d_instantiate (new, inode);security_d_instantiate (new, inode);d_rehash (dentry);d_move (new, dentry);@@ -1183,6 +1187,7 @@ struct dentry *d_splice_alias(struct ino dentry->d_inode = inode;fsnotify_d_instantiate (dentry, inode);spin_unlock (&dcache_lock); +integrity_d_instantiate (dentry, inode);security_d_instantiate (dentry, inode);d_rehash (dentry); }
fs/dcache.c found: · line 1744
Quoted 7 lines
fs/dcache.c +1
-0@@ -1733,6 +1738,7 @@ found:spin_unlock (&dcache_lock); out_nolock:if (actual == dentry) { +integrity_d_instantiate (dentry, inode);security_d_instantiate (dentry, inode);return NULL ; }
I'm trying to find a non-trivial ->d_instantiate() implementation to see how much overhead is being added to these performance-critical codepaths, but afaict this patchset doesn't add one?
Quoted 19 lines
Index:
linux-2.6.26-rc3-git2/fs/ext3/xattr_security.c
===================================================================fs/ext3/xattr_security.c +4
-0@@ -9,6 +9,7 @@#include <linux/ext3_jbd.h> #include <linux/ext3_fs.h> #include <linux/security.h> +#include <linux/integrity.h> #include "xattr.h" static size_t@@ -57,12 +58,19 @@ ext3_init_security(handle_t *handle, str err =security_inode_init_security (inode, dir, &name, &value, &len);if (err) { +/* Even if creation of the security xattr fails, must +* indicate this is a new inode. */ +integrity_inode_init_integrity (inode, dir,NULL ,NULL ,NULL );
fs/ext3/xattr_security.c ext3_init_security(handle_t *handle, str · line 73
Quoted 10 lines
fs/ext3/xattr_security.c +4
-0if (err == -EOPNOTSUPP )return 0 ;return err; } err =ext3_xattr_set_handle (handle, inode,EXT3_XATTR_INDEX_SECURITY , name, value, len,0 ); + +integrity_inode_init_integrity (inode, dir, &name, &value, &len); + err =ext3_xattr_set_handle (handle, inode,EXT3_XATTR_INDEX_SECURITY , + name, value, len,0 );
Can we put the integrity_inode_init_integrity() call into security_inode_init_security() thus avoiding having to change lots of filesystems?
Quoted 15 lines
fs/ext3/xattr_security.c +1
-0kfree (name);kfree (value);return err;@@ -17,6 +17,7 @@#include <linux/hash.h> #include <linux/swap.h> #include <linux/security.h> +#include <linux/integrity.h> #include <linux/pagemap.h> #include <linux/cdev.h> #include <linux/bootmem.h>
fs/ext3/xattr_security.c static struct inode *alloc_inode(struct · line 170
Quoted 10 lines
fs/ext3/xattr_security.c +7
-0@@ -160,6 +161,14 @@ static struct inode *alloc_inode(structinit_rwsem (&inode->i_alloc_sem);lockdep_set_class (&inode->i_alloc_sem, &sb->s_type->i_alloc_sem_key); +if (integrity_inode_alloc (inode)) { +if (inode->i_sb->s_op->destroy_inode) + inode->i_sb->s_op->destroy_inode (inode); +else +kmem_cache_free (inode_cachep, (inode)); +return NULL ; + }
This code is uncommented and integrity_inode_alloc() also is uncommented. People will want to know what's going on, please.
Again, where do we go to see how much overhead is being added to these codepaths?
Quoted 17 lines
fs/ext3/xattr_security.c +1
-0mapping->a_ops = &empty_aops; mapping->host = inode; mapping->flags =0 ;@@ -190,6 +199,7 @@ void destroy_inode(struct inode *inode) {BUG_ON (inode_has_buffers (inode));security_inode_free (inode); +integrity_inode_free (inode);if (inode->i_sb->s_op->destroy_inode) inode->i_sb->s_op->destroy_inode (inode);else
include/linux/fs.h struct inode { · line 658
Quoted 6 lines
include/linux/fs.h +3
-0@@ -653,6 +653,9 @@ struct inode {#ifdef CONFIG_SECURITY void *i_security;#endif +#ifdef CONFIG_INTEGRITY +void *i_integrity; +#endif
Does this _have_ to be a void*? Something typesafe would be far preferable.
include/linux/fs.h extern int audit_signals; · line 531
Quoted 10 lines
include/linux/fs.h +1
-0void *i_private;/* fs or device private pointer */ };@@ -521,6 +528,7 @@ extern int audit_signals;#define audit_get_loginuid (t) (-1)#define audit_get_sessionid (t) (-1)#define audit_log_task_context (b) do { ; } while (0) +#define audit_log_inode_context (b, a) do { ; } while (0)
static inline C functions are preferable.
The ";" inside the {} is unneeded.
Quoted 22 lines
include/linux/fs.h +0
-0#define audit_ipc_obj (i) ({ 0; })#define audit_ipc_set_perm (q,u,g,m) ({ 0; })#define audit_bprm (p) ({ 0; })Index:
linux-2.6.26-rc3-git2/security/integrity/integrity_audit.c
===================================================================security/integrity/integrity_audit.c new file +14
-0@@ -0,0 +1,73 @@ +/* +* Copyright (C) 2008 IBM Corporation +* Author: Mimi Zohar <[email protected]> +* +* This program is free software; you can redistribute it and/or modify +* it under the terms of the GNU General Public License as published by +* the Free Software Foundation, version 2 of the License. +* +* File: integrity_audit.c +* Audit calls for the integrity subsystem +*/ + +#include <linux/audit.h> +#include <linux/fs.h>
security/integrity/integrity_audit.c line 24
Quoted 10 lines
security/integrity/integrity_audit.c new file +10
-0+#include <linux/integrity.h> + +#ifdef CONFIG_INTEGRITY_AUDIT +static int integrity_audit =1 ; + +static int __init integrity_audit_setup (char *str) +{ +char *op; + + integrity_audit =simple_strtol (str,NULL ,0 );
This will treat "42foo" as valid input. strict_strtoul() fixes that.
security/integrity/integrity_audit.c line 33
Quoted 9 lines
security/integrity/integrity_audit.c new file +9
-0+ op = integrity_audit ?"integrity_audit_enabled" : +"integrity_audit_not_enabled" ; +integrity_audit_msg (AUDIT_INTEGRITY_STATUS ,NULL ,NULL ,NULL , op,0 ); + return 1; +} + +__setup ("integrity_audit=" , integrity_audit_setup); +#else +static int integrity_audit =1 ;
This could be outside the ifdef. If it is correct?
security/integrity/integrity_audit.c line 42
Quoted 9 lines
security/integrity/integrity_audit.c new file +9
-0+#endif + +void integrity_audit_msg (int audit_msgno,struct inode *inode, +const unsigned char *fname,char *op, +char *cause,int result) +{ +struct audit_buffer *ab; +if (!integrity_audit && result ==1 ) +return ;
A newline between end-of-declarations and start-of-code is conventional.
Quoted 28 lines
security/integrity/integrity_audit.c new file +28
-0+ ab =audit_log_start (current->audit_context,GFP_ATOMIC , audit_msgno); +audit_log_format (ab,"integrity: pid=%d uid=%u auid=%u" , + current->pid, current->uid, +audit_get_loginuid (current)); +audit_log_task_context (ab); +switch (audit_msgno) { +case AUDIT_INTEGRITY_DATA : +case AUDIT_INTEGRITY_METADATA : +case AUDIT_INTEGRITY_PCR : +audit_log_format (ab," op=%s cause=%s" , op, cause); +break ; +case AUDIT_INTEGRITY_HASH : +audit_log_format (ab," op=%s hash=%s" , op, cause); +break ; +case AUDIT_INTEGRITY_STATUS : +default : +audit_log_format (ab," op=%s" , op); + } +audit_log_format (ab," comm=" ); +audit_log_untrustedstring (ab, current->comm); +if (fname) { +audit_log_format (ab," name=" ); +audit_log_untrustedstring (ab, fname); + } +if (inode) +audit_log_format (ab," dev=%s ino=%lu" , + inode->i_sb->s_id, inode->i_ino); +audit_log_format (ab," res=%d" , result);
security/integrity/integrity_audit.c
Quoted 10 lines
security/integrity/integrity_audit.c new file +7
-0+audit_log_end (ab); +} +static struct integrity_measure_rule_entry default_rules[] = { + {{NULL ,NULL },NULL ,NULL ,FILE_MMAP ,MAY_EXEC }, + {{NULL ,NULL },NULL ,NULL ,BPRM_CHECK ,MAY_EXEC }, + {{NULL ,NULL },NULL ,NULL ,INODE_PERMISSION ,MAY_READ }, +};
Can we use the
.field = value,
format here please? That will permit the omission of all the NULLs.
Quoted 5 lines
security/integrity/integrity_audit.c new file +5
-0+static struct list_head measure_default_rules; +static struct list_head measure_policy_rules; +static struct list_head *integrity_measure; + +static DEFINE_MUTEX (integrity_measure_mutex);
security/integrity/integrity_audit.c
Quoted 10 lines
security/integrity/integrity_audit.c new file +10
-0+ +/** +* integrity_measure_rules - determine whether an inode matches the given rule. +* @rule - a pointer to a rule +* @inode - a pointer to an inode +* @func - LIM hook identifier +* @mask - requested action (MAY_READ | MAY_WRITE | MAY_APPEND | MAY_EXEC) +* +* Returns 1 on rule match, 0 on failure. +*/
What a strange name. integrity_match_rules()?
Could return a bool type, if you like that sort of thing.
security/integrity/integrity_audit.c
Quoted 7 lines
security/integrity/integrity_audit.c new file +7
-0+static int integrity_measure_rules (struct integrity_measure_rule_entry *rule, +struct inode *inode,enum lim_hooks func, +int mask) +{ +int result =1 ; + +if (result && (rule->func !=0 )) {
The test of the known-to-be-non-zero `result' is a bit weird, btu I guess it makes sense in context, and the compiler should elide it OK.
Quoted 16 lines
security/integrity/integrity_audit.c new file +16
-0+if (rule->func != func) + result =0 ; + } +if (result && (rule->mask !=0 )) { +if (rule->mask != mask) + result =0 ; + } +if (result && rule->lsm_subj_rule) { +struct task_struct *tsk = current; +u32 sid; + +security_task_getsecid (tsk, &sid); + result =security_filter_rule_match (sid,AUDIT_SUBJ_USER , +AUDIT_EQUAL , + rule->lsm_subj_rule,NULL ); + }
security/integrity/integrity_audit.c
Quoted 10 lines
security/integrity/integrity_audit.c new file +10
-0+if (result && rule->lsm_obj_rule) { +u32 osid; + +security_inode_getsecid (inode, &osid); + result =security_filter_rule_match (osid,AUDIT_OBJ_USER , +AUDIT_EQUAL , + rule->lsm_obj_rule,NULL ); + } +return result; +}
However the shole function could be simplified and sped up (depending on how smart the compiler is) via:
if (rule->func && rule->func != func) return 0; if (rule->mask && rule->mask != mask) return 0; ... return 1; }
or similar.
Quoted 15 lines
security/integrity/integrity_audit.c new file +15
-0+/** +* integrity_measure_policy - base measure decision on: subj, obj, LIM hook, +* and mask +* @inode - pointer to an inode +* @func - LIM hook identifier +* @mask - requested action (MAY_READ | MAY_WRITE | MAY_APPEND | MAY_EXEC) +* +* Returns 1 on rule match, 0 on failure. +*/ +int integrity_measure_policy (struct inode *inode,enum lim_hooks func,int mask) +{ +struct integrity_measure_rule_entry *entry; +int rc =0 ; + +rcu_read_lock ();
security/integrity/integrity_audit.c
Quoted 10 lines
security/integrity/integrity_audit.c new file +10
-0+list_for_each_entry_rcu (entry, integrity_measure, list) { + rc =integrity_measure_rules (entry, inode, func, mask); +if (rc) { +rcu_read_unlock (); +return rc; + } + } +rcu_read_unlock (); +return rc; +}
"measure"? Or "match"?
security/integrity/integrity_audit.c
Quoted 12 lines
security/integrity/integrity_audit.c new file +12
-0+/** +* integrity_measure_policy_init - initialize the default and policy rules. +*/ +void integrity_measure_policy_init (void ) +{ +int i; + +INIT_LIST_HEAD (&measure_default_rules); +for (i =0 ; i <ARRAY_SIZE (default_rules); i++) +list_add (&default_rules[i].list, &measure_default_rules); + integrity_measure = &measure_default_rules; +mutex_init (&integrity_measure_mutex);
The mutex_init() is unneeded.
Quoted 11 lines
security/integrity/integrity_audit.c new file +11
-0+INIT_LIST_HEAD (&measure_policy_rules); +} + +/** +* integrity_measure_policy_complete - wait to replace default_rules with +* a complete policy ruleset. +*/ +void integrity_measure_policy_complete (void ) +{ +char *op ="policy_update" ; +char *cause ="already exists" ;
security/integrity/integrity_audit.c
Quoted 10 lines
security/integrity/integrity_audit.c new file +10
-0+int result =1 ; + +if (integrity_measure == &measure_default_rules) { + integrity_measure = &measure_policy_rules; + cause ="complete" ; + result =0 ; + } +integrity_audit_msg (AUDIT_INTEGRITY_STATUS ,NULL , +NULL , op, cause, result); +}
Does it actually "wait"?
The name again doesn't seem to match the behaviour. "foo_complete" would mean "tell listeners that foo has completed". What you have here is a "foo_wait".
Quoted 12 lines
security/integrity/integrity_audit.c new file +12
-0+/** +* integrity_measure_rule_add - add integrity measure rules +* @subj - pointer to an LSM subject value +* @obj - pointer to an LSM object value +* @func - LIM hook identifier +* @mask - requested action (MAY_READ | MAY_WRITE | MAY_APPEND | MAY_EXEC) +* +* Returns 0 on success, an error code on failure. +*/ +int integrity_measure_rule_add (char *subj,char *obj,char *func,char *mask) +{ +struct integrity_measure_rule_entry *entry;
security/integrity/integrity_audit.c
Quoted 10 lines
security/integrity/integrity_audit.c new file +10
-0+int result =0 ; + +/* Prevent installed policy from changing */ +if (integrity_measure != &measure_default_rules) { +integrity_audit_msg (AUDIT_INTEGRITY_STATUS ,NULL , +NULL ,"policy_update" ,"already exists" ,1 ); +return -EACCES ; + } + + entry =kzalloc (sizeof (*entry),GFP_ATOMIC );
GFP_ATOMIC is unreliable. GFP_KERNEL is much much preferable, and I suspect that it can be used here?
Quoted 38 lines
security/integrity/integrity_audit.c new file +36
-0+INIT_LIST_HEAD (&entry->list); +if (!result && subj) + result =security_filter_rule_init (AUDIT_SUBJ_USER ,AUDIT_EQUAL , + subj, &entry->lsm_subj_rule); +if (!result && obj) + result =security_filter_rule_init (AUDIT_OBJ_USER ,AUDIT_EQUAL , + obj, &entry->lsm_obj_rule); +if (!result && func) { +if (strcmp (func,"INODE_PERMISSION" ) ==0 ) + entry->func =INODE_PERMISSION ; +else if (strcmp (func,"FILE_MMAP" ) ==0 ) + entry->func =FILE_MMAP ; +else if (strcmp (func,"BPRM_CHECK" ) ==0 ) + entry->func =BPRM_CHECK ; +else + result = -EINVAL ; + } +if (!result && mask) { +if (strcmp (mask,"MAY_EXEC" ) ==0 ) + entry->mask =MAY_EXEC ; +else if (strcmp (mask,"MAY_WRITE" ) ==0 ) + entry->mask =MAY_WRITE ; +else if (strcmp (mask,"MAY_READ" ) ==0 ) + entry->mask =MAY_READ ; +else if (strcmp (mask,"MAY_APPEND" ) ==0 ) + entry->mask =MAY_APPEND ; +else + result = -EINVAL ; + } +if (!result) { +mutex_lock (&integrity_measure_mutex); +list_add_tail (&entry->list, &measure_policy_rules); +mutex_unlock (&integrity_measure_mutex); + } +return result; +}
Quoted 29 lines · Andrew Morton
Quoted 27 lines · On Fri, 23 May 2008 11:05:33 -0400 Mimi Zohar <[email protected]> wrote:
This is a request for comments for a redesign of the integrity patches.
...
security/integrity/integrity.c new file +23
-0+int register_template (char *template_name, +struct template_operations *template_ops) +{ +int template_len; +struct template_list_entry *entry; + +if (!template_initialized++) { +INIT_LIST_HEAD (&integrity_templates); +mutex_init (&integrity_templates_mutex); + } + + entry =kmalloc (sizeof (*entry),GFP_ATOMIC ); +INIT_LIST_HEAD (&entry->template); + + template_len =strlen (template_name); +if (template_len >TEMPLATE_NAME_LEN_MAX ) + template_len =TEMPLATE_NAME_LEN_MAX ; +memcpy (entry->template_name, template_name, template_len); + entry->template_name[template_len] ='\0' ; + entry->template_ops = template_ops; + +mutex_lock (&integrity_templates_mutex); +list_add_rcu (&entry->template, &integrity_templates);
Quoted 8 lines
security/integrity/integrity.c line 99
Quoted 6 lines
security/integrity/integrity.c new file +6
-0+mutex_unlock (&integrity_templates_mutex); +synchronize_rcu (); + +return 0 ; +} +EXPORT_SYMBOL_GPL (register_template);Should be
integrity_register_template()?
yes, definitely.
Quoted 19 lines
Quoted 19 lines
security/integrity/integrity.c new file +19
-0+/** +* unregister_template +* @template_name: a pointer to a string containing the template name. +* +* Unregister the template functions +*/ +int unregister_template (char *template_name) +{ +struct template_list_entry *entry; + +mutex_lock (&integrity_templates_mutex); +list_for_each_entry (entry, &integrity_templates, template) { +if (strncmp (entry->template_name, template_name, +strlen (entry->template_name)) ==0 ) { +list_del_rcu (&entry->template); +mutex_unlock (&integrity_templates_mutex); +synchronize_rcu (); +kfree (entry); +return 0 ;
Quoted 8 lines
security/integrity/integrity.c line 125
Quoted 6 lines
security/integrity/integrity.c new file +6
-0+ } + } +mutex_unlock (&integrity_templates_mutex); +return -EINVAL ; +} +EXPORT_SYMBOL_GPL (unregister_template);Similarly.
Ok
Quoted 21 lines
Quoted 21 lines
security/integrity/integrity.c new file +21
-0+/** +* integrity_find_template +* @template_name: a pointer to a string containing the template name. +* @template_ops: a pointer to the template functions +* +* Find the template functions based on the template name. +*/ +int integrity_find_template (char *template_name, +struct template_operations **template_ops) +{ +struct template_list_entry *entry; + +rcu_read_lock (); +list_for_each_entry_rcu (entry, &integrity_templates, template) { +if (strncmp (entry->template_name, template_name, +strlen (entry->template_name)) ==0 ) { + *template_ops = entry->template_ops; +rcu_read_unlock (); +return 0 ; + } + }
Quoted 8 lines
security/integrity/integrity.c line 151
Quoted 4 lines
security/integrity/integrity.c new file +4
-0+rcu_read_unlock (); +return 1 ; +} +EXPORT_SYMBOL_GPL (integrity_find_template);Strange locking. We take the
rcu_read_lockthen locate a pointer to an object then drop the lock, returning that pointer while doing nothing to ensure the stability of the returned object?
yep, I messed up the locking here. Will fix it.
Quoted 14 lines
security/integrity/integrity_dummy.c line 97
Quoted 11 lines
...
security/integrity/integrity_dummy.c new file +8
-0+#define set_to_dummy_if_null (ops, function) \ + do { \ + if (!ops->function) { \ + ops->function = dummy_##function; \ + printk(KERN_INFO "Had to override the " #function \ + " integrity operation with the dummy one.\n");\ + } \ + } while (0)hrm.
Quoted 2 lines
Probably the message should identify where it came from? "integrity:
had to override ..."
Ok.
Quoted 31 lines
Quoted 31 lines
...
fs/dcache.c +3
-0@@ -1036,6 +1038,7 @@ struct dentry *d_instantiate_unique(struspin_unlock (&dcache_lock);if (!result) { +integrity_d_instantiate (entry, inode);security_d_instantiate (entry, inode);return NULL ; }@@ -1173,6 +1176,7 @@ struct dentry *d_splice_alias(struct inoBUG_ON (!(new->d_flags &DCACHE_DISCONNECTED ));fsnotify_d_instantiate (new, inode);spin_unlock (&dcache_lock); +integrity_d_instantiate (new, inode);security_d_instantiate (new, inode);d_rehash (dentry);d_move (new, dentry);@@ -1183,6 +1187,7 @@ struct dentry *d_splice_alias(struct ino dentry->d_inode = inode;fsnotify_d_instantiate (dentry, inode);spin_unlock (&dcache_lock); +integrity_d_instantiate (dentry, inode);security_d_instantiate (dentry, inode);d_rehash (dentry); }@@ -1733,6 +1738,7 @@ found:spin_unlock (&dcache_lock); out_nolock:if (actual == dentry) {
Quoted 8 lines
fs/dcache.c found: · line 1744
Quoted 4 lines
fs/dcache.c +1
-0+integrity_d_instantiate (dentry, inode);security_d_instantiate (dentry, inode);return NULL ; }I'm trying to find a non-trivial ->d_instantiate() implementation to see how much overhead is being added to these performance-critical codepaths, but afaict this patchset doesn't add one?
The current IMA, as a stand alone integrity provider, does not use integrity_d_instantiate. It was used by EVM to verify the integrity of the hash and store the results in i_integrity.
(Refer to 2.6.22-rc1-mm1-broken-out.tar.bz2)
We could defer this discussion by removing it for now, as it is not being used by the IMA.
Quoted 25 lines
Quoted 25 lines
Index:
linux-2.6.26-rc3-git2/fs/ext3/xattr_security.c
===================================================================fs/ext3/xattr_security.c +4
-0@@ -9,6 +9,7 @@#include <linux/ext3_jbd.h> #include <linux/ext3_fs.h> #include <linux/security.h> +#include <linux/integrity.h> #include "xattr.h" static size_t@@ -57,12 +58,19 @@ ext3_init_security(handle_t *handle, str err =security_inode_init_security (inode, dir, &name, &value, &len);if (err) { +/* Even if creation of the security xattr fails, must +* indicate this is a new inode. */ +integrity_inode_init_integrity (inode, dir,NULL ,NULL ,NULL );if (err == -EOPNOTSUPP )return 0 ;return err; } err =ext3_xattr_set_handle (handle, inode,EXT3_XATTR_INDEX_SECURITY , name, value, len,0 );
Quoted 8 lines
fs/ext3/xattr_security.c ext3_init_security(handle_t *handle, str · line 73
Quoted 4 lines
fs/ext3/xattr_security.c +4
-0+ +integrity_inode_init_integrity (inode, dir, &name, &value, &len); + err =ext3_xattr_set_handle (handle, inode,EXT3_XATTR_INDEX_SECURITY , + name, value, len,0 );Can we put the
integrity_inode_init_integrity()call intosecurity_inode_init_security()thus avoiding having to change lots of filesystems?
That is fine by me, not sure how others feel.
Quoted 18 lines
Quoted 18 lines
fs/ext3/xattr_security.c +1
-0kfree (name);kfree (value);return err;@@ -17,6 +17,7 @@#include <linux/hash.h> #include <linux/swap.h> #include <linux/security.h> +#include <linux/integrity.h> #include <linux/pagemap.h> #include <linux/cdev.h> #include <linux/bootmem.h> @@ -160,6 +161,14 @@ static struct inode *alloc_inode(structinit_rwsem (&inode->i_alloc_sem);lockdep_set_class (&inode->i_alloc_sem, &sb->s_type->i_alloc_sem_key);
Quoted 10 lines
fs/ext3/xattr_security.c static struct inode *alloc_inode(struct · line 170
Quoted 7 lines
fs/ext3/xattr_security.c +7
-0+if (integrity_inode_alloc (inode)) { +if (inode->i_sb->s_op->destroy_inode) + inode->i_sb->s_op->destroy_inode (inode); +else +kmem_cache_free (inode_cachep, (inode)); +return NULL ; + }This code is uncommented and
integrity_inode_alloc()also is uncommented. People will want to know what's going on, please.
integrity_inode_alloc() is the wrapper which calls inode_alloc_integrity(). It is commented in integrity.h. I'll comment the code here as well.
Quoted 2 lines
Again, where do we go to see how much overhead is being added to these codepaths?
The IMA patch(5/5) instantiates the inode_alloc_integrity hook.
.inode_alloc_integrity = ima_inode_alloc_integrity,
Quoted 14 lines
Quoted 14 lines
fs/ext3/xattr_security.c +1
-0mapping->a_ops = &empty_aops; mapping->host = inode; mapping->flags =0 ;@@ -190,6 +199,7 @@ void destroy_inode(struct inode *inode) {BUG_ON (inode_has_buffers (inode));security_inode_free (inode); +integrity_inode_free (inode);if (inode->i_sb->s_op->destroy_inode) inode->i_sb->s_op->destroy_inode (inode);else
Quoted 12 lines
include/linux/fs.h struct inode { · line 658
Quoted 9 lines
include/linux/fs.h +3
-0@@ -653,6 +653,9 @@ struct inode {#ifdef CONFIG_SECURITY void *i_security;#endif +#ifdef CONFIG_INTEGRITY +void *i_integrity; +#endif Does this _have_ to be a void*? Something typesafe would be far preferable.
Different integrity providers will define it differently, using their specific data structure.
Quoted 13 lines
include/linux/fs.h extern int audit_signals; · line 531
Quoted 10 lines
include/linux/fs.h +1
-0void *i_private;/* fs or device private pointer */ };@@ -521,6 +528,7 @@ extern int audit_signals;#define audit_get_loginuid (t) (-1)#define audit_get_sessionid (t) (-1)#define audit_log_task_context (b) do { ; } while (0) +#define audit_log_inode_context (b, a) do { ; } while (0)static inline C functions are preferable.
Quoted 1 line
The ";" inside the {} is unneeded.
Ok
Quoted 26 lines
Quoted 26 lines
include/linux/fs.h +0
-0#define audit_ipc_obj (i) ({ 0; })#define audit_ipc_set_perm (q,u,g,m) ({ 0; })#define audit_bprm (p) ({ 0; })Index:
linux-2.6.26-rc3-git2/security/integrity/integrity_audit.c
===================================================================security/integrity/integrity_audit.c new file +18
-0@@ -0,0 +1,73 @@ +/* +* Copyright (C) 2008 IBM Corporation +* Author: Mimi Zohar <[email protected]> +* +* This program is free software; you can redistribute it and/or modify +* it under the terms of the GNU General Public License as published by +* the Free Software Foundation, version 2 of the License. +* +* File: integrity_audit.c +* Audit calls for the integrity subsystem +*/ + +#include <linux/audit.h> +#include <linux/fs.h> +#include <linux/integrity.h> + +#ifdef CONFIG_INTEGRITY_AUDIT +static int integrity_audit =1 ;
Quoted 8 lines
security/integrity/integrity_audit.c line 24
Quoted 6 lines
security/integrity/integrity_audit.c new file +6
-0+ +static int __init integrity_audit_setup (char *str) +{ +char *op; + + integrity_audit =simple_strtol (str,NULL ,0 );This will treat "42foo" as valid input.
strict_strtoul()fixes that.
Thank you for pointing this out.
Quoted 11 lines
security/integrity/integrity_audit.c line 33
Quoted 9 lines
security/integrity/integrity_audit.c new file +9
-0+ op = integrity_audit ?"integrity_audit_enabled" : +"integrity_audit_not_enabled" ; +integrity_audit_msg (AUDIT_INTEGRITY_STATUS ,NULL ,NULL ,NULL , op,0 ); + return 1; +} + +__setup ("integrity_audit=" , integrity_audit_setup); +#else +static int integrity_audit =1 ;This could be outside the ifdef. If it is correct?
Yes, will fix it.
Quoted 11 lines
security/integrity/integrity_audit.c line 42
Quoted 9 lines
security/integrity/integrity_audit.c new file +9
-0+#endif + +void integrity_audit_msg (int audit_msgno,struct inode *inode, +const unsigned char *fname,char *op, +char *cause,int result) +{ +struct audit_buffer *ab; +if (!integrity_audit && result ==1 ) +return ;A newline between end-of-declarations and start-of-code is conventional.
Of course, don't know how I missed it.
Quoted 41 lines
Quoted 28 lines
security/integrity/integrity_audit.c new file +28
-0+ ab =audit_log_start (current->audit_context,GFP_ATOMIC , audit_msgno); +audit_log_format (ab,"integrity: pid=%d uid=%u auid=%u" , + current->pid, current->uid, +audit_get_loginuid (current)); +audit_log_task_context (ab); +switch (audit_msgno) { +case AUDIT_INTEGRITY_DATA : +case AUDIT_INTEGRITY_METADATA : +case AUDIT_INTEGRITY_PCR : +audit_log_format (ab," op=%s cause=%s" , op, cause); +break ; +case AUDIT_INTEGRITY_HASH : +audit_log_format (ab," op=%s hash=%s" , op, cause); +break ; +case AUDIT_INTEGRITY_STATUS : +default : +audit_log_format (ab," op=%s" , op); + } +audit_log_format (ab," comm=" ); +audit_log_untrustedstring (ab, current->comm); +if (fname) { +audit_log_format (ab," name=" ); +audit_log_untrustedstring (ab, fname); + } +if (inode) +audit_log_format (ab," dev=%s ino=%lu" , + inode->i_sb->s_id, inode->i_ino); +audit_log_format (ab," res=%d" , result);security/integrity/integrity_audit.c
Quoted 10 lines
security/integrity/integrity_audit.c new file +7
-0+audit_log_end (ab); +} +static struct integrity_measure_rule_entry default_rules[] = { + {{NULL ,NULL },NULL ,NULL ,FILE_MMAP ,MAY_EXEC }, + {{NULL ,NULL },NULL ,NULL ,BPRM_CHECK ,MAY_EXEC }, + {{NULL ,NULL },NULL ,NULL ,INODE_PERMISSION ,MAY_READ }, +};Can we use the
Quoted 3 lines
.field = value,
format here please? That will permit the omission of all the NULLs.
np
Quoted 9 lines
Quoted 9 lines
security/integrity/integrity_audit.c new file +9
-0+static struct list_head measure_default_rules; +static struct list_head measure_policy_rules; +static struct list_head *integrity_measure; + +static DEFINE_MUTEX (integrity_measure_mutex); + +/** +* integrity_measure_rules - determine whether an inode matches the given rule. +* @rule - a pointer to a rule
Quoted 8 lines
Quoted 6 lines
+ * @inode - a pointer to an inode
+ * @func - LIM hook identifier
+ * @mask - requested action (MAY_READ|MAY_WRITE|MAY_APPEND|MAY_EXEC)
+ *
+ * Returns 1 on rule match, 0 on failure.
+ */What a strange name.
integrity_match_rules()
The match decision, here, is deciding whether or not to measure a file.
The functions below are called: integrity_measure_policy_init(), integrity_measure_policy_complete(), integrity_measure_policy_add(). To parallel the rest of the names, then the function here would be integrity_measure_rules_match(). Or I could remove 'measure' from all of the function names, as you're suggesting.
choice 2: integrity_policy_init, integrity_policy_add, integrity_policy_complete/end, integrity_policy_match, integrity_rules_match
choice 3: integrity_init_policy, integrity_add_policy, integrity_complete/end_policy, integrity_match_policy, integrity_match_rules
Quoted 51 lines
Could return a bool type, if you like that sort of thing.
security/integrity/integrity_audit.c
Quoted 7 lines
security/integrity/integrity_audit.c new file +7
-0+static int integrity_measure_rules (struct integrity_measure_rule_entry *rule, +struct inode *inode,enum lim_hooks func, +int mask) +{ +int result =1 ; + +if (result && (rule->func !=0 )) {The test of the known-to-be-non-zero `result' is a bit weird, btu I guess it makes sense in context, and the compiler should elide it OK.
Quoted 16 lines
security/integrity/integrity_audit.c new file +16
-0+if (rule->func != func) + result =0 ; + } +if (result && (rule->mask !=0 )) { +if (rule->mask != mask) + result =0 ; + } +if (result && rule->lsm_subj_rule) { +struct task_struct *tsk = current; +u32 sid; + +security_task_getsecid (tsk, &sid); + result =security_filter_rule_match (sid,AUDIT_SUBJ_USER , +AUDIT_EQUAL , + rule->lsm_subj_rule,NULL ); + }security/integrity/integrity_audit.c
Quoted 10 lines
security/integrity/integrity_audit.c new file +10
-0+if (result && rule->lsm_obj_rule) { +u32 osid; + +security_inode_getsecid (inode, &osid); + result =security_filter_rule_match (osid,AUDIT_OBJ_USER , +AUDIT_EQUAL , + rule->lsm_obj_rule,NULL ); + } +return result; +}However the shole function could be simplified and sped up (depending on how smart the compiler is) via:
if (rule->func && rule->func != func) return 0; if (rule->mask && rule->mask != mask) return 0; ... return 1; }
Quoted 1 line
or similar.
yes, of course. thanks!
Quoted 19 lines
Quoted 19 lines
security/integrity/integrity_audit.c new file +19
-0+/** +* integrity_measure_policy - base measure decision on: subj, obj, LIM hook, +* and mask +* @inode - pointer to an inode +* @func - LIM hook identifier +* @mask - requested action (MAY_READ | MAY_WRITE | MAY_APPEND | MAY_EXEC) +* +* Returns 1 on rule match, 0 on failure. +*/ +int integrity_measure_policy (struct inode *inode,enum lim_hooks func,int mask) +{ +struct integrity_measure_rule_entry *entry; +int rc =0 ; + +rcu_read_lock (); +list_for_each_entry_rcu (entry, integrity_measure, list) { + rc =integrity_measure_rules (entry, inode, func, mask); +if (rc) { +rcu_read_unlock ();
Quoted 8 lines
security/integrity/integrity_audit.c
Quoted 6 lines
security/integrity/integrity_audit.c new file +6
-0+return rc; + } + } +rcu_read_unlock (); +return rc; +}"measure"? Or "match"?
discussion above.
Quoted 6 lines
Quoted 6 lines
security/integrity/integrity_audit.c new file +6
-0+/** +* integrity_measure_policy_init - initialize the default and policy rules. +*/ +void integrity_measure_policy_init (void ) +{ +int i;
Quoted 8 lines
security/integrity/integrity_audit.c
Quoted 6 lines
security/integrity/integrity_audit.c new file +6
-0+ +INIT_LIST_HEAD (&measure_default_rules); +for (i =0 ; i <ARRAY_SIZE (default_rules); i++) +list_add (&default_rules[i].list, &measure_default_rules); + integrity_measure = &measure_default_rules; +mutex_init (&integrity_measure_mutex);The
mutex_init()is unneeded.
Ok
Quoted 15 lines
Quoted 15 lines
security/integrity/integrity_audit.c new file +15
-0+INIT_LIST_HEAD (&measure_policy_rules); +} + +/** +* integrity_measure_policy_complete - wait to replace default_rules with +* a complete policy ruleset. +*/ +void integrity_measure_policy_complete (void ) +{ +char *op ="policy_update" ; +char *cause ="already exists" ; +int result =1 ; + +if (integrity_measure == &measure_default_rules) { + integrity_measure = &measure_policy_rules;
Quoted 8 lines
security/integrity/integrity_audit.c
Quoted 6 lines
security/integrity/integrity_audit.c new file +6
-0+ cause ="complete" ; + result =0 ; + } +integrity_audit_msg (AUDIT_INTEGRITY_STATUS ,NULL , +NULL , op, cause, result); +}Does it actually "wait"?
No, it doesn't wait. The function is called when the file closes, meaning that all of the rules have been added. Refer to ima_release_policy() in the IMA patch (5/5). Only after this function is called, are the new set of rules used.
Quoted 3 lines
The name again doesn't seem to match the behaviour. "
foo_complete" would mean "tell listeners that foo has completed". What you have here is a "foo_wait".
Juxtaposed to the function below called integrity_measure_policy_add(), how about integrity_measure_policy_end()?
Quoted 17 lines
Quoted 17 lines
security/integrity/integrity_audit.c new file +17
-0+/** +* integrity_measure_rule_add - add integrity measure rules +* @subj - pointer to an LSM subject value +* @obj - pointer to an LSM object value +* @func - LIM hook identifier +* @mask - requested action (MAY_READ | MAY_WRITE | MAY_APPEND | MAY_EXEC) +* +* Returns 0 on success, an error code on failure. +*/ +int integrity_measure_rule_add (char *subj,char *obj,char *func,char *mask) +{ +struct integrity_measure_rule_entry *entry; +int result =0 ; + +/* Prevent installed policy from changing */ +if (integrity_measure != &measure_default_rules) { + integrity_audit_msg(AUDIT_INTEGRITY_STATUS, NULL,
Quoted 8 lines
security/integrity/integrity_audit.c
Quoted 5 lines
security/integrity/integrity_audit.c new file +5
-0+ NULL, "policy_update", "already exists", 1); +return -EACCES ; + } + + entry =kzalloc (sizeof (*entry),GFP_ATOMIC );
GFP_ATOMICis unreliable.GFP_KERNELis much much preferable, and I suspect that it can be used here?
Probably. Will look into it.
Quoted 38 lines
Quoted 38 lines
security/integrity/integrity_audit.c new file +36
-0+INIT_LIST_HEAD (&entry->list); +if (!result && subj) + result =security_filter_rule_init (AUDIT_SUBJ_USER ,AUDIT_EQUAL , + subj, &entry->lsm_subj_rule); +if (!result && obj) + result =security_filter_rule_init (AUDIT_OBJ_USER ,AUDIT_EQUAL , + obj, &entry->lsm_obj_rule); +if (!result && func) { +if (strcmp (func,"INODE_PERMISSION" ) ==0 ) + entry->func =INODE_PERMISSION ; +else if (strcmp (func,"FILE_MMAP" ) ==0 ) + entry->func =FILE_MMAP ; +else if (strcmp (func,"BPRM_CHECK" ) ==0 ) + entry->func =BPRM_CHECK ; +else + result = -EINVAL ; + } +if (!result && mask) { +if (strcmp (mask,"MAY_EXEC" ) ==0 ) + entry->mask =MAY_EXEC ; +else if (strcmp (mask,"MAY_WRITE" ) ==0 ) + entry->mask =MAY_WRITE ; +else if (strcmp (mask,"MAY_READ" ) ==0 ) + entry->mask =MAY_READ ; +else if (strcmp (mask,"MAY_APPEND" ) ==0 ) + entry->mask =MAY_APPEND ; +else + result = -EINVAL ; + } +if (!result) { +mutex_lock (&integrity_measure_mutex); +list_add_tail (&entry->list, &measure_policy_rules); +mutex_unlock (&integrity_measure_mutex); + } +return result; +}
James Morris
maintainerQuoted 5 lines · Mimi Zohar
Quoted 3 lines
Can we put the
integrity_inode_init_integrity()call intosecurity_inode_init_security()thus avoiding having to change lots of filesystems?That is fine by me, not sure how others feel.
I agree it makes sense to do this.
Signature · 2 lines
James Morris <[email protected]>