Messages in this thread |  | | Date | Wed, 15 Nov 2000 10:38:15 +0100 | From | Ralf Baechle <> | Subject | Re: [PATCH] show_task() and thread_saved_pc() fix for x86 |
| |
On Tue, Nov 14, 2000 at 10:19:32AM +0100, Jean Wolter wrote:
> > > OTOH, the value is used only by Alt-SysRq-T, so... Hell knows. > > > > No, it's also used by 'ps -l'. See wchan. > > ps -l uses get_wchan() (an architecture specific function from > arch/*/kernel/process.c) to get the return address from > schedule(). And now thread_saved_pc() seems to do the same (at least > on x86). Is there any reason to have two architecture specific > functions doing the same or do I miss something? > > Jean > > PS: Architectures other then x86 use thread_saved_pc() to implement > get_wchan(). If the debug output of Alt-SysRq-T is supposed to show > the waiting channel we should use get_wchan() instead of thread_saved_pc().
Probably historic reasons, it's been that way as long as I can think back. Yet the use of thread_saved_pc() in kernel/sched.c should imho be considered a buglet and be replaced by get_wchan to get more meaningful debugging information.
Ralf - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org Please read the FAQ at http://www.tux.org/lkml/
|  |