Re: [PATCH 2/2] tracing: syscall_regfunc() should not skip kernelthreads

From: H. Peter Anvin
Date: Tue Mar 19 2013 - 17:27:50 EST


On 03/19/2013 08:36 AM, Steven Rostedt wrote:
> On Tue, 2013-03-19 at 15:10 +0000, David Howells wrote:
>> Steven Rostedt <rostedt@xxxxxxxxxxx> wrote:
>>
>>> Why? If we remove the tracepoint from the slowpath and use a table swap,
>>> then we wouldn't need to use the slowpath at all.
>>
>> How are you engineering a table swap? Do you patch the system call code to
>> change the immediate address loaded or do you put in a level of indirection?
>
> Patching the call site would probably be the easiest method.
>
> We've gotten pretty good at doing that ;-)
>

Yes, given that the machinery is already there we might as well use it.

-hpa


--
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/