Re: [patch] Real-Time Preemption, -VP-2.6.9-rc4-mm1-U0

[Ingo Molnar]

* Robert Wisniewski <bob@xxxxxxxxxxxxxx> wrote:

> Ingo Molnar writes:
>  > 
>  > * Robert Wisniewski <bob@xxxxxxxxxxxxxx> wrote:
>  > 
>  > >  > > cmpxchg (basically: try reserve; if fail retry; else write), with
>  > >  > > per-cpu buffers.
>  > >  > 
>  > >  > this still does not solve all problems related to irq entries: if the
>  > >  > IRQ interrups the tracing code after a 'successful reserve' but before
>  > >  > the 'else write' point, and the trace is printed/saved from an
>  > >  > interrupt, then there will be an incomplete entry in the trace.
>  > > 
>  > > That is incorrect.  The system behavior needed to generate an
>  > > incomplete entry is far more complicated and unlikely than what you
>  > > describe.
>  > 
>  > ah, but i'm talking about actual first-hand experience, not supposition. 
>  > It happens quite easily with latency traces (which are saved/printed
>  > from IRQ entries) and it can be very annoying to analyze. My first
>  > tracers tried to do things without the IRQ flag, so i've seen both
>  > methods.
> 
> This means that other code you've written has this happen, it doesn't mean
> the fundamental model is broken.  Also, if what you claim is true and there
> really is this contention, then it both means that 1) there are many many
> other higher priority tasks in the system than the one you are trying to
> trace, and 2) it's questionable whether you want to use locks.

_interrupts_. The latency tracer does traces like:

 00000002 0.022ms (+0.000ms): mark_page_accessed (zap_pte_range)
 00000002 0.022ms (+0.000ms): page_remove_rmap (zap_pte_range)
 00000002 0.022ms (+0.000ms): free_page_and_swap_cache (zap_pte_range)
 00000002 0.022ms (+0.001ms): put_page (zap_pte_range)
 00010002 0.023ms (+0.000ms): do_IRQ (zap_pte_range)
 00010002 0.023ms (+0.000ms): do_IRQ (<00000000>)
 00010003 0.024ms (+0.004ms): mask_and_ack_8259A (do_IRQ)
 00010003 0.029ms (+0.000ms): redirect_hardirq (do_IRQ)
 00010000 0.029ms (+0.000ms): handle_IRQ_event (do_IRQ)

and i just pointed out why i didnt use relayfs.

	Ingo
-
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/