Re: [PATCH] ARM: ftrace: Trace function entry before updating index

From: Steven Rostedt
Date: Wed Jul 18 2012 - 13:38:50 EST


On Mon, 2012-07-16 at 19:03 -0700, Colin Cross wrote:
> Commit 722b3c74695377d11d18a52f3da08114d37f3f37 modified x86 ftrace to
> avoid tracing all functions called from irqs when function graph was
> used with a filter. Port the same fix to ARM.
>
> Cc: Steven Rostedt <rostedt@xxxxxxxxxxx>

Acked-by: Steven Rostedt <rostedt@xxxxxxxxxxx>

Someone on the ARM side needs to pull this in.

I'll look into fixing this for PPC.

Thanks!

-- Steve

> Signed-off-by: Colin Cross <ccross@xxxxxxxxxxx>
> ---
>
> It looks like the same issue affects blackfin, microblaze, mips,
> parisc, powerpc, s390, sh, and sparc, but I don't have patches to
> fix those.
>
> arch/arm/kernel/ftrace.c | 17 +++++++++--------
> 1 files changed, 9 insertions(+), 8 deletions(-)
>
> diff --git a/arch/arm/kernel/ftrace.c b/arch/arm/kernel/ftrace.c
> index df0bf0c..34e5664 100644
> --- a/arch/arm/kernel/ftrace.c
> +++ b/arch/arm/kernel/ftrace.c
> @@ -179,19 +179,20 @@ void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr,
> old = *parent;
> *parent = return_hooker;
>
> - err = ftrace_push_return_trace(old, self_addr, &trace.depth,
> - frame_pointer);
> - if (err == -EBUSY) {
> - *parent = old;
> - return;
> - }
> -
> trace.func = self_addr;
> + trace.depth = current->curr_ret_stack + 1;
>
> /* Only trace if the calling function expects to */
> if (!ftrace_graph_entry(&trace)) {
> - current->curr_ret_stack--;
> *parent = old;
> + return;
> + }
> +
> + err = ftrace_push_return_trace(old, self_addr, &trace.depth,
> + frame_pointer);
> + if (err == -EBUSY) {
> + *parent = old;
> + return;
> }
> }
>


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/