From: Herbert Poetzl (herbert_at_13thfloor.at)
Date: Thu 20 Feb 2003 - 17:16:57 GMT
On Thu, Feb 20, 2003 at 04:26:19PM +0000, Jonathan Sambrook wrote:
> I'm looking at the ctx patch for DSVR to see how we could integrate it
> into our operation (see http://www.dsvr.co.uk).
hmm, maybe you could elaborate a little bit
on your plans regarding the "operation"
> We'd like vserver users to be as ignorant of their vserver-ness as
> possible, hence cloaking /proc/self/status.
but what about the other 10 or 20 indices for
a virtual server environment? (e.g. mknod, ethernet
etc ...) or do you give all capabilities?
> But for investigating/debugging/hacking-on-vserver purposes, this
> cloaking should be sysctl-able.
> 
> The ctx sysctl should itself be cloakable too.
hmm, why not use a special capability to give
this kind of access, this would allow you to
enter a context with or without the ability
to get the extended /proc/self/status ...
> All this might not suit all tastes, so whilst the patch defaults to the
> most secure option, the usual:
> 
> 	echo 1 > /proc/sys/kernel/ctx/visible-self-status
> 	echo 1 > /proc/sys/kernel/ctx/visible-sysctrls
> 
> would restore previous behaviour.
> 
> Comments please.
sounds to me like "maybe we could sell a 
virtual server as dedicated machine?"
best,
Herbert
PS: nice page, though, unfortunately not HTML 
or XHTML compliant ...
> Jonathan
> 
> -- 
>                    
>  Jonathan Sambrook 
> Software  Developer 
>  Designer  Servers
> diff -Nur linux-2.4.20-quota-ctx/arch/i386/config.in linux-2.4.20/arch/i386/config.in
> --- linux-2.4.20-quota-ctx/arch/i386/config.in	2003-01-30 14:40:56.000000000 +0000
> +++ linux-2.4.20/arch/i386/config.in	2003-02-20 14:10:41.000000000 +0000
> @@ -286,6 +286,11 @@
>  bool 'System V IPC' CONFIG_SYSVIPC
>  bool 'BSD Process Accounting' CONFIG_BSD_PROCESS_ACCT
>  bool 'Sysctl support' CONFIG_SYSCTL
> +
> +if [ "$CONFIG_SYSCTL" = "y" ]; then
> +	bool 'Sysctl support for Context Patch' CONFIG_CTX_SYSCTLS
> +fi
> +	
>  if [ "$CONFIG_PROC_FS" = "y" ]; then
>     choice 'Kernel core (/proc/kcore) format' \
>  	"ELF		CONFIG_KCORE_ELF	\
> diff -Nur linux-2.4.20-quota-ctx/arch/ppc/config.in linux-2.4.20/arch/ppc/config.in
> --- linux-2.4.20-quota-ctx/arch/ppc/config.in	2003-01-30 14:40:57.000000000 +0000
> +++ linux-2.4.20/arch/ppc/config.in	2003-02-18 13:28:35.000000000 +0000
> @@ -161,6 +161,11 @@
>  
>  bool 'Networking support' CONFIG_NET
>  bool 'Sysctl support' CONFIG_SYSCTL
> +
> +if [ "$CONFIG_SYSCTL" = "y" ]; then
> +	bool 'Sysctl support for Context Patch' CONFIG_CTX_SYSCTLS
> +fi
> +	
>  bool 'System V IPC' CONFIG_SYSVIPC
>  bool 'BSD Process Accounting' CONFIG_BSD_PROCESS_ACCT
>  
> diff -Nur linux-2.4.20-quota-ctx/arch/sparc/config.in linux-2.4.20/arch/sparc/config.in
> --- linux-2.4.20-quota-ctx/arch/sparc/config.in	2003-01-30 14:40:57.000000000 +0000
> +++ linux-2.4.20/arch/sparc/config.in	2003-02-18 13:28:35.000000000 +0000
> @@ -65,6 +65,11 @@
>  bool 'System V IPC' CONFIG_SYSVIPC
>  bool 'BSD Process Accounting' CONFIG_BSD_PROCESS_ACCT
>  bool 'Sysctl support' CONFIG_SYSCTL
> +
> +if [ "$CONFIG_SYSCTL" = "y" ]; then
> +	bool 'Sysctl support for Context Patch' CONFIG_CTX_SYSCTLS
> +fi
> +	
>  if [ "$CONFIG_PROC_FS" = "y" ]; then
>     define_bool CONFIG_KCORE_ELF y
>  fi
> diff -Nur linux-2.4.20-quota-ctx/arch/sparc64/config.in linux-2.4.20/arch/sparc64/config.in
> --- linux-2.4.20-quota-ctx/arch/sparc64/config.in	2003-01-30 14:40:57.000000000 +0000
> +++ linux-2.4.20/arch/sparc64/config.in	2003-02-18 13:28:35.000000000 +0000
> @@ -64,6 +64,11 @@
>  bool 'System V IPC' CONFIG_SYSVIPC
>  bool 'BSD Process Accounting' CONFIG_BSD_PROCESS_ACCT
>  bool 'Sysctl support' CONFIG_SYSCTL
> +
> +if [ "$CONFIG_SYSCTL" = "y" ]; then
> +	bool 'Sysctl support for Context Patch' CONFIG_CTX_SYSCTLS
> +fi
> +	
>  if [ "$CONFIG_PROC_FS" = "y" ]; then
>     define_bool CONFIG_KCORE_ELF y
>  fi
> diff -Nur linux-2.4.20-quota-ctx/Documentation/Configure.help linux-2.4.20/Documentation/Configure.help
> --- linux-2.4.20-quota-ctx/Documentation/Configure.help	2003-01-30 14:40:57.000000000 +0000
> +++ linux-2.4.20/Documentation/Configure.help	2003-02-20 14:10:41.000000000 +0000
> @@ -526,6 +526,13 @@
>    The umem driver has been allocated block major number 116.
>    See Documentation/devices.txt for recommended device naming.
>  
> +Context Patch Sysctls
> +CONFIG_CTX_SYSCTLS
> +	Saying Y here will allow you to configure some of the behaviour of
> +	the Jacques Gelinas Context Patch
> +
> +	See http://www.solucorp.qc.ca/miscprj/s_context.hc
> +
>  Network block device support
>  CONFIG_BLK_DEV_NBD
>    Saying Y here will allow your computer to be a client for network
> diff -Nur linux-2.4.20-quota-ctx/fs/proc/array.c linux-2.4.20/fs/proc/array.c
> --- linux-2.4.20-quota-ctx/fs/proc/array.c	2003-01-30 14:40:56.000000000 +0000
> +++ linux-2.4.20/fs/proc/array.c	2003-02-18 13:28:35.000000000 +0000
> @@ -279,6 +279,9 @@
>  			    cap_t(p->cap_bset));
>  }
>  
> +#ifdef CONFIG_CTX_SYSCTLS
> +int sysctl_ctx_unsafe_self_status = 0; // Default to safe
> +#endif
>  
>  int proc_pid_status(struct task_struct *task, char * buffer)
>  {
> @@ -299,47 +302,54 @@
>  	buffer = task_sig(task, buffer);
>  	buffer = task_cap(task, buffer);
>  #ifdef __NR_new_s_context
> -	if (task->s_info != NULL){
> -		int i;
> -		buffer += sprintf (buffer,"s_context: %d [",task->s_context);
> -		for (i=0; i<NB_S_CONTEXT; i++){
> -			short int ctx = task->s_info->s_context[i];
> -			if (ctx == 0) break;
> -			buffer += sprintf (buffer," %d",ctx);
> +	if(current->s_context == 0 
> +#ifdef CONFIG_CTX_SYSCTLS
> +			|| sysctl_ctx_unsafe_self_status
> +#endif
> +	)
> +	{
> +		if (task->s_info != NULL){
> +			int i;
> +			buffer += sprintf (buffer,"s_context: %d [",task->s_context);
> +			for (i=0; i<NB_S_CONTEXT; i++){
> +				short int ctx = task->s_info->s_context[i];
> +				if (ctx == 0) break;
> +				buffer += sprintf (buffer," %d",ctx);
> +			}
> +			*buffer++ = ']';
> +			*buffer++ = '\n';
> +			buffer += sprintf (buffer,"ctxticks: %d %ld %d\n"
> +				,atomic_read(&task->s_info->ticks),task->counter
> +				,task->s_info->refcount);
> +			buffer += sprintf (buffer,"ctxflags: %d\n"
> +				,task->s_info->flags);
> +			buffer += sprintf (buffer,"initpid: %d\n"
> +				,task->s_info->initpid);
> +		}else{
> +			buffer += sprintf (buffer,"s_context: %d\n",task->s_context);
> +			buffer += sprintf (buffer,"ctxticks: none\n");
> +			buffer += sprintf (buffer,"ctxflags: none\n");
> +			buffer += sprintf (buffer,"initpid: none\n");
>  		}
> -		*buffer++ = ']';
> -		*buffer++ = '\n';
> -		buffer += sprintf (buffer,"ctxticks: %d %ld %d\n"
> -			,atomic_read(&task->s_info->ticks),task->counter
> -			,task->s_info->refcount);
> -		buffer += sprintf (buffer,"ctxflags: %d\n"
> -			,task->s_info->flags);
> -		buffer += sprintf (buffer,"initpid: %d\n"
> -			,task->s_info->initpid);
> -	}else{
> -		buffer += sprintf (buffer,"s_context: %d\n",task->s_context);
> -		buffer += sprintf (buffer,"ctxticks: none\n");
> -		buffer += sprintf (buffer,"ctxflags: none\n");
> -		buffer += sprintf (buffer,"initpid: none\n");
> -	}
> -	if (task->ip_info != NULL){
> -		int i;
> -		buffer += sprintf (buffer,"ipv4root:");
> -		for (i=0; i<task->ip_info->nbipv4; i++){
> -			buffer += sprintf (buffer," %08x"
> -				,task->ip_info->ipv4[i]);
> +		if (task->ip_info != NULL){
> +			int i;
> +			buffer += sprintf (buffer,"ipv4root:");
> +			for (i=0; i<task->ip_info->nbipv4; i++){
> +				buffer += sprintf (buffer," %08x"
> +					,task->ip_info->ipv4[i]);
> +			}
> +			*buffer++ = '\n';
> +			buffer += sprintf (buffer,"ipv4root_bcast: %08x\n"
> +				,task->ip_info->v4_bcast);
> +			buffer += sprintf (buffer,"ipv4root_refcnt: %d\n"
> +				,task->ip_info->refcount);
> +		}else{
> +			buffer += sprintf (buffer,"ipv4root: 0\n");
> +			buffer += sprintf (buffer,"ipv4root_bcast: 0\n");
>  		}
> -		*buffer++ = '\n';
> -		buffer += sprintf (buffer,"ipv4root_bcast: %08x\n"
> -			,task->ip_info->v4_bcast);
> -		buffer += sprintf (buffer,"ipv4root_refcnt: %d\n"
> -			,task->ip_info->refcount);
> -	}else{
> -		buffer += sprintf (buffer,"ipv4root: 0\n");
> -		buffer += sprintf (buffer,"ipv4root_bcast: 0\n");
> +		buffer += sprintf (buffer,"__NR_new_s_context: %d\n",__NR_new_s_context);
> +		buffer += sprintf (buffer,"__NR_set_ipv4root: %d rev2\n",__NR_set_ipv4root);
>  	}
> -	buffer += sprintf (buffer,"__NR_new_s_context: %d\n",__NR_new_s_context);
> -	buffer += sprintf (buffer,"__NR_set_ipv4root: %d rev2\n",__NR_set_ipv4root);
>  #endif
>  #if defined(CONFIG_ARCH_S390)
>  	buffer = task_show_regs(task, buffer);
> diff -Nur linux-2.4.20-quota-ctx/fs/proc/generic.c linux-2.4.20/fs/proc/generic.c
> --- linux-2.4.20-quota-ctx/fs/proc/generic.c	2003-01-30 14:40:56.000000000 +0000
> +++ linux-2.4.20/fs/proc/generic.c	2003-02-20 13:39:23.000000000 +0000
> @@ -249,6 +249,9 @@
>   * Don't create negative dentries here, return -ENOENT by hand
>   * instead.
>   */
> +#if defined CONFIG_CTX_SYSCTLS 
> +extern int sysctl_ctx_unsafe_sysctrls;
> +#endif
>  struct dentry *proc_lookup(struct inode * dir, struct dentry *dentry)
>  {
>  	struct inode *inode;
> @@ -264,6 +267,12 @@
>  				continue;
>  			if (de->namelen != dentry->d_name.len)
>  				continue;
> +#if defined CONFIG_CTX_SYSCTLS 
> +      // Cloak /proc/sys/kernel/ctx outside of root ctx unless explicitly visible
> +      if( de->namelen == 3 && !strcmp("ctx", de->name) && 
> +          current->s_context != 0 && !sysctl_ctx_unsafe_sysctrls )
> +				continue;
> +#endif
>  			if (!memcmp(dentry->d_name.name, de->name, de->namelen)) {
>  				int ino = de->low_ino;
>  				error = -EINVAL;
> @@ -281,6 +290,10 @@
>  	return ERR_PTR(error);
>  }
>  
> +#if defined CONFIG_CTX_SYSCTLS 
> +int sysctl_ctx_unsafe_sysctrls = 0; // Default to safe
> +#endif
> +
>  /*
>   * This returns non-zero if at EOF, so that the /proc
>   * root directory can use this and check if it should
> @@ -331,6 +344,10 @@
>  			}
>  
>  			do {
> +#if defined CONFIG_CTX_SYSCTLS 
> +				// Ctx sysctls only visible to root context (by default)
> +				if( current->s_context == 0 || sysctl_ctx_unsafe_sysctrls || strcmp(de->name, "ctx") )
> +#endif
>  				if (filldir(dirent, de->name, de->namelen, filp->f_pos,
>  					    de->low_ino, de->mode >> 12) < 0)
>  					return 0;
> Binary files linux-2.4.20-quota-ctx/fs/proc/.generic.c.swp and linux-2.4.20/fs/proc/.generic.c.swp differ
> diff -Nur linux-2.4.20-quota-ctx/include/linux/sysctl.h linux-2.4.20/include/linux/sysctl.h
> --- linux-2.4.20-quota-ctx/include/linux/sysctl.h	2003-01-30 14:40:56.000000000 +0000
> +++ linux-2.4.20/include/linux/sysctl.h	2003-02-20 09:53:20.000000000 +0000
> @@ -124,6 +124,7 @@
>  	KERN_CORE_USES_PID=52,		/* int: use core or core.%pid */
>  	KERN_TAINTED=53,	/* int: various kernel tainted flags */
>  	KERN_CADPID=54,		/* int: PID of the process to notify on CAD */
> +	KERN_CTX,                    /* Context Patch */
>  };
> 
> 
> diff -Nur linux-2.4.20-quota-ctx/kernel/sysctl.c linux-2.4.20/kernel/sysctl.c
> --- linux-2.4.20-quota-ctx/kernel/sysctl.c	2003-01-30 14:40:56.000000000 +0000
> +++ linux-2.4.20/kernel/sysctl.c	2003-02-20 09:34:08.000000000 +0000
> @@ -86,6 +86,25 @@
>  extern int sysctl_userprocess_debug;
>  #endif
>  
> +#ifdef CONFIG_CTX_SYSCTLS
> +extern int sysctl_ctx_unsafe_self_status;
> +extern int sysctl_ctx_unsafe_sysctrls;
> +
> +enum
> +{
> +	CTX_SYSCTRL_UNSAFE_SYSCTRLS=1,  // int: non-root-ctx visible ctx sysctls
> +	CTX_SYSCTRL_UNSAFE_SELF_STATUS  // int: non-root-ctx visible ctx info in /proc/self/status
> +};
> +
> +static ctl_table ctx_table[] = {
> +	{CTX_SYSCTRL_UNSAFE_SYSCTRLS,"visible-sysctrls",
> +	 &sysctl_ctx_unsafe_sysctrls,sizeof(int),0644,NULL,&proc_dointvec},
> +	{CTX_SYSCTRL_UNSAFE_SELF_STATUS,"visible-self-status",
> +	 &sysctl_ctx_unsafe_self_status,sizeof(int),0644,NULL,&proc_dointvec},
> +	 {0}
> +};
> +#endif
> +
>  #ifdef CONFIG_PPC32
>  extern unsigned long zero_paged_on, powersave_nap;
>  int proc_dol2crvec(ctl_table *table, int write, struct file *filp,
> @@ -256,6 +275,9 @@
>  	{KERN_S390_USER_DEBUG_LOGGING,"userprocess_debug",
>  	 &sysctl_userprocess_debug,sizeof(int),0644,NULL,&proc_dointvec},
>  #endif
> +#ifdef CONFIG_CTX_SYSCTLS
> +	{KERN_CTX, "ctx", NULL, 0, 0700, ctx_table},
> +#endif
>  	{0}
>  };
>