Re: [PATCH v6] ftrace: disable preemption when recursion locked

From: 王贇
Date: Tue Oct 26 2021 - 22:25:02 EST


Hi, Steven, Miroslav

Should have fixed the comments about bit value, besides, add
a warn in trace_clear_recursion() to make sure the bit < 0
abusing case will get notified.

Please let me know if there are any other issues :-)

Regards,
Michael Wang

On 2021/10/27 上午10:11, 王贇 wrote:
> As the documentation explained, ftrace_test_recursion_trylock()
> and ftrace_test_recursion_unlock() were supposed to disable and
> enable preemption properly, however currently this work is done
> outside of the function, which could be missing by mistake.
>
> And since the internal using of trace_test_and_set_recursion()
> and trace_clear_recursion() also require preemption disabled, we
> can just merge the logical.
>
> This patch will make sure the preemption has been disabled when
> trace_test_and_set_recursion() return bit >= 0, and
> trace_clear_recursion() will enable the preemption if previously
> enabled.
>
> CC: Petr Mladek <pmladek@xxxxxxxx>
> CC: Steven Rostedt <rostedt@xxxxxxxxxxx>
> CC: Miroslav Benes <mbenes@xxxxxxx>
> Reported-by: Abaci <abaci@xxxxxxxxxxxxxxxxx>
> Suggested-by: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
> Signed-off-by: Michael Wang <yun.wang@xxxxxxxxxxxxxxxxx>
> ---
> arch/csky/kernel/probes/ftrace.c | 2 --
> arch/parisc/kernel/ftrace.c | 2 --
> arch/powerpc/kernel/kprobes-ftrace.c | 2 --
> arch/riscv/kernel/probes/ftrace.c | 2 --
> arch/x86/kernel/kprobes/ftrace.c | 2 --
> include/linux/trace_recursion.h | 13 ++++++++++++-
> kernel/livepatch/patch.c | 13 +++++++------
> kernel/trace/ftrace.c | 15 +++++----------
> kernel/trace/trace_functions.c | 5 -----
> 9 files changed, 24 insertions(+), 32 deletions(-)
>
> diff --git a/arch/csky/kernel/probes/ftrace.c b/arch/csky/kernel/probes/ftrace.c
> index b388228..834cffc 100644
> --- a/arch/csky/kernel/probes/ftrace.c
> +++ b/arch/csky/kernel/probes/ftrace.c
> @@ -17,7 +17,6 @@ void kprobe_ftrace_handler(unsigned long ip, unsigned long parent_ip,
> return;
>
> regs = ftrace_get_regs(fregs);
> - preempt_disable_notrace();
> p = get_kprobe((kprobe_opcode_t *)ip);
> if (!p) {
> p = get_kprobe((kprobe_opcode_t *)(ip - MCOUNT_INSN_SIZE));
> @@ -57,7 +56,6 @@ void kprobe_ftrace_handler(unsigned long ip, unsigned long parent_ip,
> __this_cpu_write(current_kprobe, NULL);
> }
> out:
> - preempt_enable_notrace();
> ftrace_test_recursion_unlock(bit);
> }
> NOKPROBE_SYMBOL(kprobe_ftrace_handler);
> diff --git a/arch/parisc/kernel/ftrace.c b/arch/parisc/kernel/ftrace.c
> index 7d14242..90c4345 100644
> --- a/arch/parisc/kernel/ftrace.c
> +++ b/arch/parisc/kernel/ftrace.c
> @@ -210,7 +210,6 @@ void kprobe_ftrace_handler(unsigned long ip, unsigned long parent_ip,
> return;
>
> regs = ftrace_get_regs(fregs);
> - preempt_disable_notrace();
> p = get_kprobe((kprobe_opcode_t *)ip);
> if (unlikely(!p) || kprobe_disabled(p))
> goto out;
> @@ -239,7 +238,6 @@ void kprobe_ftrace_handler(unsigned long ip, unsigned long parent_ip,
> }
> __this_cpu_write(current_kprobe, NULL);
> out:
> - preempt_enable_notrace();
> ftrace_test_recursion_unlock(bit);
> }
> NOKPROBE_SYMBOL(kprobe_ftrace_handler);
> diff --git a/arch/powerpc/kernel/kprobes-ftrace.c b/arch/powerpc/kernel/kprobes-ftrace.c
> index 7154d58..072ebe7 100644
> --- a/arch/powerpc/kernel/kprobes-ftrace.c
> +++ b/arch/powerpc/kernel/kprobes-ftrace.c
> @@ -26,7 +26,6 @@ void kprobe_ftrace_handler(unsigned long nip, unsigned long parent_nip,
> return;
>
> regs = ftrace_get_regs(fregs);
> - preempt_disable_notrace();
> p = get_kprobe((kprobe_opcode_t *)nip);
> if (unlikely(!p) || kprobe_disabled(p))
> goto out;
> @@ -61,7 +60,6 @@ void kprobe_ftrace_handler(unsigned long nip, unsigned long parent_nip,
> __this_cpu_write(current_kprobe, NULL);
> }
> out:
> - preempt_enable_notrace();
> ftrace_test_recursion_unlock(bit);
> }
> NOKPROBE_SYMBOL(kprobe_ftrace_handler);
> diff --git a/arch/riscv/kernel/probes/ftrace.c b/arch/riscv/kernel/probes/ftrace.c
> index aab85a8..7142ec4 100644
> --- a/arch/riscv/kernel/probes/ftrace.c
> +++ b/arch/riscv/kernel/probes/ftrace.c
> @@ -15,7 +15,6 @@ void kprobe_ftrace_handler(unsigned long ip, unsigned long parent_ip,
> if (bit < 0)
> return;
>
> - preempt_disable_notrace();
> p = get_kprobe((kprobe_opcode_t *)ip);
> if (unlikely(!p) || kprobe_disabled(p))
> goto out;
> @@ -52,7 +51,6 @@ void kprobe_ftrace_handler(unsigned long ip, unsigned long parent_ip,
> __this_cpu_write(current_kprobe, NULL);
> }
> out:
> - preempt_enable_notrace();
> ftrace_test_recursion_unlock(bit);
> }
> NOKPROBE_SYMBOL(kprobe_ftrace_handler);
> diff --git a/arch/x86/kernel/kprobes/ftrace.c b/arch/x86/kernel/kprobes/ftrace.c
> index 596de2f..dd2ec14 100644
> --- a/arch/x86/kernel/kprobes/ftrace.c
> +++ b/arch/x86/kernel/kprobes/ftrace.c
> @@ -25,7 +25,6 @@ void kprobe_ftrace_handler(unsigned long ip, unsigned long parent_ip,
> if (bit < 0)
> return;
>
> - preempt_disable_notrace();
> p = get_kprobe((kprobe_opcode_t *)ip);
> if (unlikely(!p) || kprobe_disabled(p))
> goto out;
> @@ -59,7 +58,6 @@ void kprobe_ftrace_handler(unsigned long ip, unsigned long parent_ip,
> __this_cpu_write(current_kprobe, NULL);
> }
> out:
> - preempt_enable_notrace();
> ftrace_test_recursion_unlock(bit);
> }
> NOKPROBE_SYMBOL(kprobe_ftrace_handler);
> diff --git a/include/linux/trace_recursion.h b/include/linux/trace_recursion.h
> index abe1a50..64c03ee 100644
> --- a/include/linux/trace_recursion.h
> +++ b/include/linux/trace_recursion.h
> @@ -135,6 +135,9 @@ static __always_inline int trace_get_context_bit(void)
> # define do_ftrace_record_recursion(ip, pip) do { } while (0)
> #endif
>
> +/*
> + * Preemption is promised to be disabled when return bit >= 0.
> + */
> static __always_inline int trace_test_and_set_recursion(unsigned long ip, unsigned long pip,
> int start)
> {
> @@ -162,11 +165,19 @@ static __always_inline int trace_test_and_set_recursion(unsigned long ip, unsign
> current->trace_recursion = val;
> barrier();
>
> + preempt_disable_notrace();
> +
> return bit;
> }
>
> +/*
> + * Preemption will be enabled (if it was previously enabled).
> + */
> static __always_inline void trace_clear_recursion(int bit)
> {
> + WARN_ON_ONCE(bit < 0);
> +
> + preempt_enable_notrace();
> barrier();
> trace_recursion_clear(bit);
> }
> @@ -178,7 +189,7 @@ static __always_inline void trace_clear_recursion(int bit)
> * tracing recursed in the same context (normal vs interrupt),
> *
> * Returns: -1 if a recursion happened.
> - * >= 0 if no recursion
> + * >= 0 if no recursion.
> */
> static __always_inline int ftrace_test_recursion_trylock(unsigned long ip,
> unsigned long parent_ip)
> diff --git a/kernel/livepatch/patch.c b/kernel/livepatch/patch.c
> index e8029ae..b8d75fb 100644
> --- a/kernel/livepatch/patch.c
> +++ b/kernel/livepatch/patch.c
> @@ -49,14 +49,16 @@ static void notrace klp_ftrace_handler(unsigned long ip,
>
> ops = container_of(fops, struct klp_ops, fops);
>
> + /*
> + *
> + * The ftrace_test_recursion_trylock() will disable preemption,
> + * which is required for the variant of synchronize_rcu() that is
> + * used to allow patching functions where RCU is not watching.
> + * See klp_synchronize_transition() for more details.
> + */
> bit = ftrace_test_recursion_trylock(ip, parent_ip);
> if (WARN_ON_ONCE(bit < 0))
> return;
> - /*
> - * A variant of synchronize_rcu() is used to allow patching functions
> - * where RCU is not watching, see klp_synchronize_transition().
> - */
> - preempt_disable_notrace();
>
> func = list_first_or_null_rcu(&ops->func_stack, struct klp_func,
> stack_node);
> @@ -120,7 +122,6 @@ static void notrace klp_ftrace_handler(unsigned long ip,
> klp_arch_set_pc(fregs, (unsigned long)func->new_func);
>
> unlock:
> - preempt_enable_notrace();
> ftrace_test_recursion_unlock(bit);
> }
>
> diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
> index b7be1df..7392bc7 100644
> --- a/kernel/trace/ftrace.c
> +++ b/kernel/trace/ftrace.c
> @@ -7198,16 +7198,15 @@ void ftrace_reset_array_ops(struct trace_array *tr)
> struct ftrace_ops *op;
> int bit;
>
> + /*
> + * The ftrace_test_and_set_recursion() will disable preemption,
> + * which is required since some of the ops may be dynamically
> + * allocated, they must be freed after a synchronize_rcu().
> + */
> bit = trace_test_and_set_recursion(ip, parent_ip, TRACE_LIST_START);
> if (bit < 0)
> return;
>
> - /*
> - * Some of the ops may be dynamically allocated,
> - * they must be freed after a synchronize_rcu().
> - */
> - preempt_disable_notrace();
> -
> do_for_each_ftrace_op(op, ftrace_ops_list) {
> /* Stub functions don't need to be called nor tested */
> if (op->flags & FTRACE_OPS_FL_STUB)
> @@ -7231,7 +7230,6 @@ void ftrace_reset_array_ops(struct trace_array *tr)
> }
> } while_for_each_ftrace_op(op);
> out:
> - preempt_enable_notrace();
> trace_clear_recursion(bit);
> }
>
> @@ -7279,12 +7277,9 @@ static void ftrace_ops_assist_func(unsigned long ip, unsigned long parent_ip,
> if (bit < 0)
> return;
>
> - preempt_disable_notrace();
> -
> if (!(op->flags & FTRACE_OPS_FL_RCU) || rcu_is_watching())
> op->func(ip, parent_ip, op, fregs);
>
> - preempt_enable_notrace();
> trace_clear_recursion(bit);
> }
> NOKPROBE_SYMBOL(ftrace_ops_assist_func);
> diff --git a/kernel/trace/trace_functions.c b/kernel/trace/trace_functions.c
> index 1f0e63f..9f1bfbe 100644
> --- a/kernel/trace/trace_functions.c
> +++ b/kernel/trace/trace_functions.c
> @@ -186,7 +186,6 @@ static void function_trace_start(struct trace_array *tr)
> return;
>
> trace_ctx = tracing_gen_ctx();
> - preempt_disable_notrace();
>
> cpu = smp_processor_id();
> data = per_cpu_ptr(tr->array_buffer.data, cpu);
> @@ -194,7 +193,6 @@ static void function_trace_start(struct trace_array *tr)
> trace_function(tr, ip, parent_ip, trace_ctx);
>
> ftrace_test_recursion_unlock(bit);
> - preempt_enable_notrace();
> }
>
> #ifdef CONFIG_UNWINDER_ORC
> @@ -298,8 +296,6 @@ static inline void process_repeats(struct trace_array *tr,
> if (bit < 0)
> return;
>
> - preempt_disable_notrace();
> -
> cpu = smp_processor_id();
> data = per_cpu_ptr(tr->array_buffer.data, cpu);
> if (atomic_read(&data->disabled))
> @@ -324,7 +320,6 @@ static inline void process_repeats(struct trace_array *tr,
>
> out:
> ftrace_test_recursion_unlock(bit);
> - preempt_enable_notrace();
> }
>
> static void
>