linux-kernel - Re: [PATCH] x86/traps: Weaken context tracking entry assertions

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20150821132438.GB3362@lerouge>
Date:	Fri, 21 Aug 2015 15:24:39 +0200
From:	Frederic Weisbecker <fweisbec@...il.com>
To:	Andy Lutomirski <luto@...nel.org>
Cc:	x86@...nel.org, Sasha Levin <sasha.levin@...cle.com>,
	Brian Gerst <brgerst@...il.com>,
	Denys Vlasenko <dvlasenk@...hat.com>,
	linux-kernel@...r.kernel.org, Oleg Nesterov <oleg@...hat.com>,
	Borislav Petkov <bp@...en8.de>
Subject: Re: [PATCH] x86/traps: Weaken context tracking entry assertions

On Thu, Aug 20, 2015 at 10:03:21PM -0700, Andy Lutomirski wrote:
> We were asserting that we were all the way in CONTEXT_KERNEL when
> exception handlers were called.  While having this be true is, I
> think, a nice goal (or maybe a variant in which we assert that we're
> in CONTEXT_KERNEL or some new IRQ context), we're not quite there.
> 
> In particular, if an IRQ interrupts the SYSCALL prologue and the IRQ
> handler in turn causes an exception, the exception entry will be
> called in RCU IRQ mode but with CONTEXT_USER.
> 
> This is okay (nothing goes wrong), but until we fix up the SYSCALL
> prologue, we need to avoid warning.

We can avoid interrupts before the context tracking call but we'll
never be able to remove all possibility for exceptions. I don't think
we can assume that without making context tracking more fragile.

> 
> Signed-off-by: Andy Lutomirski <luto@...nel.org>

ACK!

Thanks!

We can indeed definetly trigger an exception in the kernel entry code
(syscall, exception, irq) before the user_exit() call and that
would break the checks. We can fix that later with context tracking
calls on exception entry code. I still think an exception slow path
based on static keys is the best way to go there.

> ---
>  arch/x86/kernel/traps.c | 18 +++++++++---------
>  1 file changed, 9 insertions(+), 9 deletions(-)
> 
> diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c
> index 86a82eafb96f..45e8d9891fa3 100644
> --- a/arch/x86/kernel/traps.c
> +++ b/arch/x86/kernel/traps.c
> @@ -112,7 +112,7 @@ static inline void preempt_conditional_cli(struct pt_regs *regs)
>  void ist_enter(struct pt_regs *regs)
>  {
>  	if (user_mode(regs)) {
> -		CT_WARN_ON(ct_state() != CONTEXT_KERNEL);
> +		rcu_lockdep_assert(rcu_is_watching(), "entry code didn't wake RCU");
>  	} else {
>  		/*
>  		 * We might have interrupted pretty much anything.  In
> @@ -282,7 +282,7 @@ static void do_error_trap(struct pt_regs *regs, long error_code, char *str,
>  {
>  	siginfo_t info;
>  
> -	CT_WARN_ON(ct_state() != CONTEXT_KERNEL);
> +	rcu_lockdep_assert(rcu_is_watching(), "entry code didn't wake RCU");
>  
>  	if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) !=
>  			NOTIFY_STOP) {
> @@ -364,7 +364,7 @@ dotraplinkage void do_bounds(struct pt_regs *regs, long error_code)
>  	const struct bndcsr *bndcsr;
>  	siginfo_t *info;
>  
> -	CT_WARN_ON(ct_state() != CONTEXT_KERNEL);
> +	rcu_lockdep_assert(rcu_is_watching(), "entry code didn't wake RCU");
>  	if (notify_die(DIE_TRAP, "bounds", regs, error_code,
>  			X86_TRAP_BR, SIGSEGV) == NOTIFY_STOP)
>  		return;
> @@ -442,7 +442,7 @@ do_general_protection(struct pt_regs *regs, long error_code)
>  {
>  	struct task_struct *tsk;
>  
> -	CT_WARN_ON(ct_state() != CONTEXT_KERNEL);
> +	rcu_lockdep_assert(rcu_is_watching(), "entry code didn't wake RCU");
>  	conditional_sti(regs);
>  
>  	if (v8086_mode(regs)) {
> @@ -496,7 +496,7 @@ dotraplinkage void notrace do_int3(struct pt_regs *regs, long error_code)
>  		return;
>  
>  	ist_enter(regs);
> -	CT_WARN_ON(ct_state() != CONTEXT_KERNEL);
> +	rcu_lockdep_assert(rcu_is_watching(), "entry code didn't wake RCU");
>  #ifdef CONFIG_KGDB_LOW_LEVEL_TRAP
>  	if (kgdb_ll_trap(DIE_INT3, "int3", regs, error_code, X86_TRAP_BP,
>  				SIGTRAP) == NOTIFY_STOP)
> @@ -729,14 +729,14 @@ static void math_error(struct pt_regs *regs, int error_code, int trapnr)
>  
>  dotraplinkage void do_coprocessor_error(struct pt_regs *regs, long error_code)
>  {
> -	CT_WARN_ON(ct_state() != CONTEXT_KERNEL);
> +	rcu_lockdep_assert(rcu_is_watching(), "entry code didn't wake RCU");
>  	math_error(regs, error_code, X86_TRAP_MF);
>  }
>  
>  dotraplinkage void
>  do_simd_coprocessor_error(struct pt_regs *regs, long error_code)
>  {
> -	CT_WARN_ON(ct_state() != CONTEXT_KERNEL);
> +	rcu_lockdep_assert(rcu_is_watching(), "entry code didn't wake RCU");
>  	math_error(regs, error_code, X86_TRAP_XF);
>  }
>  
> @@ -749,7 +749,7 @@ do_spurious_interrupt_bug(struct pt_regs *regs, long error_code)
>  dotraplinkage void
>  do_device_not_available(struct pt_regs *regs, long error_code)
>  {
> -	CT_WARN_ON(ct_state() != CONTEXT_KERNEL);
> +	rcu_lockdep_assert(rcu_is_watching(), "entry code didn't wake RCU");
>  	BUG_ON(use_eager_fpu());
>  
>  #ifdef CONFIG_MATH_EMULATION
> @@ -775,7 +775,7 @@ dotraplinkage void do_iret_error(struct pt_regs *regs, long error_code)
>  {
>  	siginfo_t info;
>  
> -	CT_WARN_ON(ct_state() != CONTEXT_KERNEL);
> +	rcu_lockdep_assert(rcu_is_watching(), "entry code didn't wake RCU");
>  	local_irq_enable();
>  
>  	info.si_signo = SIGILL;
> -- 
> 2.4.3
> 
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/