[PATCH] powerpc/security: Use a mutex for interrupt exit code patching

Nicholas Piggin npiggin at gmail.com
Thu Oct 28 14:03:24 AEDT 2021


Excerpts from Russell Currey's message of October 27, 2021 5:24 pm:
> The mitigation-patching.sh script in the powerpc selftests toggles
> all mitigations on and off simultaneously, revealing that rfi_flush
> and stf_barrier cannot safely operate at the same time due to races
> in updating the static key.
> 
> On some systems, the static key code throws a warning and the kernel
> remains functional.  On others, the kernel will hang or crash.
> 
> Fix this by slapping on a mutex.
> 
> Fixes: 13799748b957 ("powerpc/64: use interrupt restart table to speed up return from interrupt")
> Signed-off-by: Russell Currey <ruscur at russell.cc>

Doh,

Acked-by: Nicholas Piggin <npiggin at gmail.com>

> ---
>  arch/powerpc/lib/feature-fixups.c | 9 +++++++++
>  1 file changed, 9 insertions(+)
> 
> diff --git a/arch/powerpc/lib/feature-fixups.c b/arch/powerpc/lib/feature-fixups.c
> index cda17bee5afe..9956277fbd88 100644
> --- a/arch/powerpc/lib/feature-fixups.c
> +++ b/arch/powerpc/lib/feature-fixups.c
> @@ -228,6 +228,7 @@ static void do_stf_exit_barrier_fixups(enum stf_barrier_type types)
>  
>  static bool stf_exit_reentrant = false;
>  static bool rfi_exit_reentrant = false;
> +static DEFINE_MUTEX(exit_flush_lock);
>  
>  static int __do_stf_barrier_fixups(void *data)
>  {
> @@ -253,6 +254,9 @@ void do_stf_barrier_fixups(enum stf_barrier_type types)
>  	 * low level interrupt exit code before patching. After the patching,
>  	 * if allowed, then flip the branch to allow fast exits.
>  	 */
> +
> +	// Prevent static key update races with do_rfi_flush_fixups()
> +	mutex_lock(&exit_flush_lock);
>  	static_branch_enable(&interrupt_exit_not_reentrant);
>  
>  	stop_machine(__do_stf_barrier_fixups, &types, NULL);
> @@ -264,6 +268,7 @@ void do_stf_barrier_fixups(enum stf_barrier_type types)
>  
>  	if (stf_exit_reentrant && rfi_exit_reentrant)
>  		static_branch_disable(&interrupt_exit_not_reentrant);
> +	mutex_unlock(&exit_flush_lock);
>  }
>  
>  void do_uaccess_flush_fixups(enum l1d_flush_type types)
> @@ -486,6 +491,9 @@ void do_rfi_flush_fixups(enum l1d_flush_type types)
>  	 * without stop_machine, so this could be achieved with a broadcast
>  	 * IPI instead, but this matches the stf sequence.
>  	 */
> +
> +	// Prevent static key update races with do_stf_barrier_fixups()
> +	mutex_lock(&exit_flush_lock);
>  	static_branch_enable(&interrupt_exit_not_reentrant);
>  
>  	stop_machine(__do_rfi_flush_fixups, &types, NULL);
> @@ -497,6 +505,7 @@ void do_rfi_flush_fixups(enum l1d_flush_type types)
>  
>  	if (stf_exit_reentrant && rfi_exit_reentrant)
>  		static_branch_disable(&interrupt_exit_not_reentrant);
> +	mutex_unlock(&exit_flush_lock);
>  }
>  
>  void do_barrier_nospec_fixups_range(bool enable, void *fixup_start, void *fixup_end)
> -- 
> 2.33.1
> 
> 


More information about the Linuxppc-dev mailing list