g the HRTIMER softirq to wake it up.
That was with ksoftirqd running as SCHED_OTHER, which is what it
defaulted to. The priority inversion between the userspace processes
is gone after making ktimersoftd the same priority as the
highest-priority application.
Tested-by: Brian Silverman
Sebastian had some questions about this patch when I first sent it to rt-users.
On Wed, May 25, 2016 at 12:54 PM, Sebastian Andrzej Siewior
wrote:
> On 05/25/2016 08:00 PM, Brian Silverman wrote:
>>> Why can't the application drop the RT priority before its exit? Wouldn't
&
Without this, a realtime process which has called mlockall exiting
causes large latencies for other realtime processes at the same or
lower priorities. This seems like a fairly common use case too, because
realtime processes generally want their memory locked into RAM.
Signed-off-by: Brian
Commit-ID: 746db9443ea57fd9c059f62c4bfbf41cf224fe13
Gitweb: http://git.kernel.org/tip/746db9443ea57fd9c059f62c4bfbf41cf224fe13
Author: Brian Silverman
AuthorDate: Wed, 18 Feb 2015 16:23:56 -0800
Committer: Ingo Molnar
CommitDate: Mon, 23 Mar 2015 10:47:55 +0100
sched: Fix
On Mon, Mar 9, 2015 at 12:08 PM, Sebastian Andrzej Siewior
wrote:
> * Brian Silverman | 2015-03-05 00:16:20 [-0500]:
>
>>Beforehand, 000 is spending most of its time in interrupts, but bash
>>is doing something related to memory management on it in between.
>>
On Mon, Mar 9, 2015 at 1:34 PM, Sebastian Andrzej Siewior
wrote:
> From what I can tell not beeing a sched guy is that the patch looks
> reasonable since the timeout gets only set to zero on enqueue_task_rt().
> Is there something special you do to trigger this?
I posted some test code with two
28 prev_prio=28 prev_state=D ==>
next_comm=swapper/1 next_pid=0 next_prio=120
...
-0 [001] d...1.. 6854.629319:
softirq_check_pending_idle <-tick_nohz_idle_enter
My tracing_off() call is in softirq_check_pending_idle, so that's it.
Thanks,
Brian Silverman
--
To unsubscri
) == 0);
assert(pthread_create(&rt, NULL, realtime, NULL) == 0);
assert(pthread_join(nrt, NULL) == 0);
assert(pthread_join(rt, NULL) == 0);
return 0;
}
On Wed, Feb 18, 2015 at 7:23 PM, wrote:
> From: Brian Silverman
>
> When non-realtime tasks get priority-inheritance boosted to a re
Commit-ID: 30a6b8031fe14031ab27c1fa3483cb9780e7f63c
Gitweb: http://git.kernel.org/tip/30a6b8031fe14031ab27c1fa3483cb9780e7f63c
Author: Brian Silverman
AuthorDate: Sat, 25 Oct 2014 20:20:37 -0400
Committer: Thomas Gleixner
CommitDate: Sun, 26 Oct 2014 16:16:18 +0100
futex: Fix a race
(when they get created at
all) are usually a mess, to the point of crashing tools used to examine
them.
Signed-off-by: Brian Silverman
---
kernel/futex.c | 44 +++-
1 file changed, 27 insertions(+), 17 deletions(-)
diff --git a/kernel/futex.c b/kernel/fu
On Sat, 25 Oct 2014, Thomas Gleixner wrote:
> > pi_state_free and exit_pi_state_list both clean up futex_pi_state's.
> > exit_pi_state_list takes the hb lock first, and most callers of
> > pi_state_free do too. requeue_pi didn't, which causes lots of problems.
>
> "causes lots of problems" is not
-1, 0));
int i;
for (i = 0; i < NUMBER_TESTERS; ++i) {
if (fork() == 0) {
while (1) {
RunTest(&shared_memory[i]);
}
}
}
}
On Thu, Oct 23, 2014 at 3:22 PM, Brian Silverman wrote:
> pi_state_free and exit_pi_state_list bot
int of crashing tools used to examine
them.
Signed-off-by: Brian Silverman
---
I am not subscribed to the list so please CC me on any responses.
kernel/futex.c | 32 +---
1 file changed, 21 insertions(+), 11 deletions(-)
diff --git a/kernel/futex.c b/kernel/futex.
13 matches
Mail list logo
