Subject: [PATCH] powerpc: fix /proc/self/stack

For the current task, the kernel stack would only tell the last time the
process was rescheduled, if ever. Use the current stack pointer for the
current task.

This is also consistent with some other architectures.

Signed-off-by: Thadeu Lima de Souza Cascardo <[email protected]>
---
arch/powerpc/kernel/stacktrace.c | 7 ++++++-
1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/arch/powerpc/kernel/stacktrace.c b/arch/powerpc/kernel/stacktrace.c
index 6671195..2446066 100644
--- a/arch/powerpc/kernel/stacktrace.c
+++ b/arch/powerpc/kernel/stacktrace.c
@@ -59,7 +59,12 @@ EXPORT_SYMBOL_GPL(save_stack_trace);

void save_stack_trace_tsk(struct task_struct *tsk, struct stack_trace *trace)
{
- save_context_stack(trace, tsk->thread.ksp, tsk, 0);
+ unsigned long sp = tsk->thread.ksp;
+
+ if (tsk == current)
+ sp = current_stack_pointer();
+
+ save_context_stack(trace, sp, tsk, 0);
}
EXPORT_SYMBOL_GPL(save_stack_trace_tsk);

--
2.9.3


2017-03-22 11:14:13

by Michael Ellerman

[permalink] [raw]
Subject: Re: [PATCH] powerpc: fix /proc/self/stack

Thadeu Lima de Souza Cascardo <[email protected]> writes:

> For the current task, the kernel stack would only tell the last time the
> process was rescheduled, if ever. Use the current stack pointer for the
> current task.

You say "fix" in the subject, but is it a bug, or just an enhancement?

> This is also consistent with some other architectures.

Such as .. arm64 and x86 (though it's buried in the unwind code).

> diff --git a/arch/powerpc/kernel/stacktrace.c b/arch/powerpc/kernel/stacktrace.c
> index 6671195..2446066 100644
> --- a/arch/powerpc/kernel/stacktrace.c
> +++ b/arch/powerpc/kernel/stacktrace.c
> @@ -59,7 +59,12 @@ EXPORT_SYMBOL_GPL(save_stack_trace);
>
> void save_stack_trace_tsk(struct task_struct *tsk, struct stack_trace *trace)
> {
> - save_context_stack(trace, tsk->thread.ksp, tsk, 0);
> + unsigned long sp = tsk->thread.ksp;
> +
> + if (tsk == current)
> + sp = current_stack_pointer();
else
sp = tsk->thread.ksp;

Would be clearer IMHO.

> +
> + save_context_stack(trace, sp, tsk, 0);
> }
> EXPORT_SYMBOL_GPL(save_stack_trace_tsk);
>
> --
> 2.9.3


cheers

Subject: [PATCH v2] powerpc: make /proc/self/stack always print the current stack

For the current task, the kernel stack would only tell the last time the
process was rescheduled, if ever. Use the current stack pointer for the
current task.

Otherwise, every once in a while, the stacktrace printed when reading
/proc/self/stack would look like the process is running in userspace,
while it's not, which some may consider as a bug.

This is also consistent with some other architectures, like x86 and arm,
at least.

Signed-off-by: Thadeu Lima de Souza Cascardo <[email protected]>
---
arch/powerpc/kernel/stacktrace.c | 9 ++++++++-
1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/arch/powerpc/kernel/stacktrace.c b/arch/powerpc/kernel/stacktrace.c
index 6671195..d534ed9 100644
--- a/arch/powerpc/kernel/stacktrace.c
+++ b/arch/powerpc/kernel/stacktrace.c
@@ -59,7 +59,14 @@ EXPORT_SYMBOL_GPL(save_stack_trace);

void save_stack_trace_tsk(struct task_struct *tsk, struct stack_trace *trace)
{
- save_context_stack(trace, tsk->thread.ksp, tsk, 0);
+ unsigned long sp;
+
+ if (tsk == current)
+ sp = current_stack_pointer();
+ else
+ sp = tsk->thread.ksp;
+
+ save_context_stack(trace, sp, tsk, 0);
}
EXPORT_SYMBOL_GPL(save_stack_trace_tsk);

--
2.9.3

2017-03-31 12:34:03

by Michael Ellerman

[permalink] [raw]
Subject: Re: [v2] powerpc: make /proc/self/stack always print the current stack

On Mon, 2017-03-27 at 19:32:33 UTC, Thadeu Lima de Souza Cascardo wrote:
> For the current task, the kernel stack would only tell the last time the
> process was rescheduled, if ever. Use the current stack pointer for the
> current task.
>
> Otherwise, every once in a while, the stacktrace printed when reading
> /proc/self/stack would look like the process is running in userspace,
> while it's not, which some may consider as a bug.
>
> This is also consistent with some other architectures, like x86 and arm,
> at least.
>
> Signed-off-by: Thadeu Lima de Souza Cascardo <[email protected]>

Applied to powerpc next, thanks.

https://git.kernel.org/powerpc/c/4f9b514b765a3057341f3236c94877

cheers