On Fri, Jan 29, 2016 at 10:50:28AM +0530, Pratyush Anand wrote:
> Implement save_stack_trace_regs, so that stacktrace of a kprobe events can
> be obtained.
>
> Without this we see following warning:
> "save_stack_trace_regs() not implemented yet."
> when we execute:
> echo stacktrace > /sys/kernel/debug/tracing/trace_options
> echo "p kfree" >> /sys/kernel/debug/tracing/kprobe_events
> echo 1 > /sys/kernel/debug/tracing/events/kprobes/enable
>
> Reported-by: Chunyu Hu <ch...@redhat.com>
> Signed-off-by: Pratyush Anand <pan...@redhat.com>
> ---
> arch/s390/kernel/stacktrace.c | 13 +++++++++++++
> 1 file changed, 13 insertions(+)
>
> diff --git a/arch/s390/kernel/stacktrace.c b/arch/s390/kernel/stacktrace.c
> index 1785cd82253c..586da400f931 100644
> --- a/arch/s390/kernel/stacktrace.c
> +++ b/arch/s390/kernel/stacktrace.c
> @@ -94,3 +94,16 @@ void save_stack_trace_tsk(struct task_struct *tsk, struct
> stack_trace *trace)
> trace->entries[trace->nr_entries++] = ULONG_MAX;
> }
> EXPORT_SYMBOL_GPL(save_stack_trace_tsk);
> +
> +void save_stack_trace_regs(struct pt_regs *regs, struct stack_trace *trace)
> +{
> + unsigned long sp, low, high;
> +
> + sp = kernel_stack_pointer(regs);
> + low = (unsigned long) task_stack_page(current);
> + high = (unsigned long) task_pt_regs(current);
> + save_context_stack(trace, sp, low, high, 0);
> + if (trace->nr_entries < trace->max_entries)
> + trace->entries[trace->nr_entries++] = ULONG_MAX;
> +}
> +EXPORT_SYMBOL_GPL(save_stack_trace_regs);
While playing around with this, I discovered a couple of bugs in our
stacktrace code. However this patch is also not correct, since it will save
a stacktrace only if being called in process context, but not for interrupt
context.
I will fix this within this patch.
