Messages in this thread | | | Date | Fri, 8 May 2015 09:58:06 +0200 | From | Ingo Molnar <> | Subject | Re: [PATCH 3/3] context_tracking,x86: remove extraneous irq disable & enable from context tracking on syscall entry |
| |
* Rik van Riel <riel@redhat.com> wrote:
> On 05/07/2015 08:29 AM, Ingo Molnar wrote: > > > > * Frederic Weisbecker <fweisbec@gmail.com> wrote: > > > >>>> We cannot take the lock_trace(task) from irq context, and we > >>>> probably do not need to anyway, since we do not care about a > >>>> precise stack trace for the task. > >>> > >>> So one worry with this and similar approaches of statistically > >>> detecting user mode would be the fact that on the way out to > >>> user-space we don't really destroy the previous call trace - we > >>> just pop off the stack (non-destructively), restore RIPs and are > >>> gone. > >>> > >>> We'll need that percpu flag I suspect. > >> > >> Note we have the context tracking state which tells where the > >> current task is: user/system/guest. > > > > Yes, but that overhead is what I'm suggesting we get rid of, I thought > > Rik was trying to find a mechanism that would be independent of that? > > One thing at a time :) > > I am working on the timer sampling stuff, which should be easy to > adapt to a different user/system/guest/irq/softirq/... tracking > thing, if somebody else comes up with a more efficient way to do > that.
So if you make the timer sampling use a percpu variable, and set that variable from the existing callbacks, then we could do this gradually: first the timer sampling uses the flag, then RCU could use it, and finally we could push it out to minimal assembly code.
But it's important to start out with a percpu flag to track this all.
Thanks,
Ingo
| |