* Jan Kiszka | 2015-04-25 09:26:13 [+0200]:
>Acked-by: Jan Kiszka
>
>This way around makes more sense as you changed the patch significantly.
Now I took the proper one.
>Thanks,
>Jan
Sebastian
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message t
On 2015-04-25 09:20, Mike Galbraith wrote:
> On Fri, 2015-04-24 at 11:00 +0200, Jan Kiszka wrote:
>
>> The approach looks good to me, but the commit log deserves a rework now.
>
> Ok, we agree on the approach, and that the changelog wants a bit of
> attention, so either you're gonna rewrite it to
On Fri, 2015-04-24 at 11:00 +0200, Jan Kiszka wrote:
> The approach looks good to me, but the commit log deserves a rework now.
Ok, we agree on the approach, and that the changelog wants a bit of
attention, so either you're gonna rewrite it to suit you, do a pretty
changelog, and I ack, or I take
On Fri, 2015-04-24 at 11:00 +0200, Jan Kiszka wrote:
> The approach looks good to me, but the commit log deserves a rework now.
Yeah. While you're at it, you should change my chop to an ack or a
tested-by too, as it's your patch, I just rearranged a bit.
-Mike
--
To unsubscribe from t
wn fscking "Preformatted" switch, so whitespace should be fine.
>
> Oh, btw, if anyone (else) makes a 4.1-rt, your rt push work will want
> one of those nifty hirq tags lest box make boom due to trying to do that
> not only way late, but with irqs enabled which pisses sched all off.
atever. Box works
and it makes line count shrink...
I downgraded evolution v3.16->v3.12 to restore its ability to read it's
own fscking "Preformatted" switch, so whitespace should be fine.
Oh, btw, if anyone (else) makes a 4.1-rt, your rt push work will want
one of those nifty hi
On Thu, 23 Apr 2015 09:19:26 +0200
Mike Galbraith wrote:
> > CC kernel/irq_work.o
> > In file included from ../include/asm-generic/percpu.h:6:0,
> > from ../arch/x86/include/asm/percpu.h:522,
> > from ../arch/x86/include/asm/current.h:5,
> >
On Thu, 2015-04-23 at 09:12 +0200, Jan Kiszka wrote:
> On 2015-04-23 09:01, Mike Galbraith wrote:
> > On Thu, 2015-04-23 at 08:50 +0200, Jan Kiszka wrote:
> > > On 2015-04-23 08:11, Mike Galbraith wrote:
> > > > @@ -103,6 +98,9 @@ EXPORT_SYMBOL_GPL(irq_work_queue_on);
> > > > /* Enqueue the irq wo
On 2015-04-23 08:58, Mike Galbraith wrote:
> On Thu, 2015-04-23 at 08:29 +0200, Jan Kiszka wrote:
>>
>>> void irq_work_tick(void)
>>> {
>>> -#ifdef CONFIG_PREEMPT_RT_FULL
>>> - irq_work_run_list(this_cpu_ptr(&lazy_list));
>>> -#else
>>> - struct llist_head *raised = &__get_cpu_var(rai
On 2015-04-23 09:01, Mike Galbraith wrote:
> On Thu, 2015-04-23 at 08:50 +0200, Jan Kiszka wrote:
>> On 2015-04-23 08:11, Mike Galbraith wrote:
>>> @@ -103,6 +98,9 @@ EXPORT_SYMBOL_GPL(irq_work_queue_on);
>>> /* Enqueue the irq work @work on the current CPU */
>>> bool irq_work_queue(struct irq_w
On Thu, 2015-04-23 at 08:50 +0200, Jan Kiszka wrote:
> On 2015-04-23 08:11, Mike Galbraith wrote:
> > @@ -103,6 +98,9 @@ EXPORT_SYMBOL_GPL(irq_work_queue_on);
> > /* Enqueue the irq work @work on the current CPU */
> > bool irq_work_queue(struct irq_work *work)
> > {
> > + bool realtime =
ter pick one.
Attaching patch since either evolution fscked up again (it does that),
or someone has managed to turn it into a completely useless piece of
crap... if so, likely the same dipstick who made it save messages such
that you need fromdos to wipe away the shite it smears all over it.
On 2015-04-23 08:11, Mike Galbraith wrote:
> @@ -103,6 +98,9 @@ EXPORT_SYMBOL_GPL(irq_work_queue_on);
> /* Enqueue the irq work @work on the current CPU */
> bool irq_work_queue(struct irq_work *work)
> {
> + bool realtime = IS_ENABLED(CONFIG_PREEMPT_RT_FULL);
> + bool raise = false;
t; dead-box fully repeatable, and now fixed.
>
> Except you kinda forgot to run the raised list. The reformatted
> (which saved two whole lines;) patch below adds that to
> irq_work_tick(), which fixes the livelock both powertop and perf top
> otherwise meet.
>
> Subj
raised list. The reformatted
(which saved two whole lines;) patch below adds that to
irq_work_tick(), which fixes the livelock both powertop and perf top
otherwise meet.
Subject: [PATCH RT 3.18] irq_work: Provide a soft-irq based queue
Date: Thu, 16 Apr 2015 18:28:16 +0200
From: Jan Kiszka
On Thu, 2015-04-16 at 18:28 +0200, Jan Kiszka wrote:
> Instead of turning all irq_work requests into lazy ones on -rt, just
> move their execution from hard into soft-irq context.
>
> This resolves deadlocks of ftrace which will queue work from
> arbitrary
> contexts, including those that have lo
Instead of turning all irq_work requests into lazy ones on -rt, just
move their execution from hard into soft-irq context.
This resolves deadlocks of ftrace which will queue work from arbitrary
contexts, including those that have locks held that are needed for
raising a soft-irq.
Signed-off-by: J
17 matches
Mail list logo
