;-) I only use the latency numbers to give me an idea of if I can use that hardware for a particular build -- meaning that if I am driving a 1880's era steam locomotive at 10MPH then I can get away with crappy latency, but if I am a Ducati 250 Road Racer wit ha top speed of 170MPH, then I am going to sweat the sand in the road 8-)
I would love to see the perfect timing assumption in the code broken to reflect the real world. Also a plug for the NURB -- you can do the offset evaluation as realtime as opposed to expected timing... EBo -- On Jul 11 2013 6:07 AM, Michael Haberler wrote: > my message probably should have read: > > "I am not aware of an RT-PREEMPT kernel for the BB, but I think it is > possible to build one now based on the patch below and it would be > great if somebody actually built one and reported figures" > > "hard rt" meaning Xenomai? those we have - on the order of 40-50uS > > -- > > now if somebody went around and outlined an RTAPI patch to pass the > actual invocation time to RT threads as discussed in Wichita, we > could > go ahead and timebase-correct some of the code which relies on the > "assumed time equals actual time" assumption, like motion and pid > comps, then we likely could put the latency discussion to rest for > the > the better part (I admit it is fun to compare latency though, it's a > bit like 'my motorbike is tuned grandiously better than your > motorbike' - nevermind the 10mpH speed limit ;) > > - Michael > > > > > Am 11.07.2013 um 13:44 schrieb EBo <[email protected]>: > >> Does anyone have latency tests for the BB with RT-PREEMPT and >> possibly >> the hard RT? >> >> Thanks and best regards, >> >> EBo -- >> >> On Jul 11 2013 1:46 AM, Michael Haberler wrote: >>> heads up for beagleBone users and folks dabbling with 3.8-based >>> kernels (most likely for ARM platforms): >>> >>> this patch seems to fit the current 3.8-based BeagleBone kernels, >>> meaning it should be possible to create an RT-PREEMPT kernel for >>> the >>> BB >>> >>> - Michael >>> >>> >>> Anfang der weitergeleiteten Nachricht: >>> >>>> Von: Sebastian Andrzej Siewior <[email protected]> >>>> Betreff: [ANNOUNCE] 3.8.13-rt14 >>>> Datum: 11. Juli 2013 09:38:58 MESZ >>>> An: linux-rt-users <[email protected]> >>>> Kopie: LKML <[email protected]>, Thomas Gleixner >>>> <[email protected]>, [email protected], John Kacur >>>> <[email protected]> >>>> >>>> Dear RT Folks, >>>> >>>> I'm pleased to announce the 3.8.13-rt14 release. >>>> >>>> changes since v3.8.13-rt13: >>>> - added a patch from Uwe Kleine-König to fix the UP breakage >>>> introduced by >>>> the recent s/sbin spinlocks/raw_spin_lock/ change. >>>> - added a fix for a livelock of workqueue vs ata-piix. Found by >>>> Carsten >>>> Emde, patch proposed by Thomas Gleixner. >>>> >>>> The delta patch against v3.8.13-rt13 is appended below and can be >>>> found here: >>>> >>>> >>>> >>>> https://www.kernel.org/pub/linux/kernel/projects/rt/3.8/incr/patch-3.8.13-rt13-rt14.patch.xz >>>> >>>> The RT patch against 3.8.13 can be found here: >>>> >>>> >>>> >>>> https://www.kernel.org/pub/linux/kernel/projects/rt/3.8/patch-3.8.13-rt14.patch.xz >>>> >>>> The split quilt queue is available at: >>>> >>>> >>>> >>>> https://www.kernel.org/pub/linux/kernel/projects/rt/3.8/patches-3.8.13-rt14.tar.xz >>>> >>>> Sebastian >>>> >>>> diff --git a/include/linux/list_bl.h b/include/linux/list_bl.h >>>> index ddfd46a..becd7a6 100644 >>>> --- a/include/linux/list_bl.h >>>> +++ b/include/linux/list_bl.h >>>> @@ -131,8 +131,10 @@ static inline void hlist_bl_lock(struct >>>> hlist_bl_head *b) >>>> bit_spin_lock(0, (unsigned long *)b); >>>> #else >>>> raw_spin_lock(&b->lock); >>>> +#if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_SPINLOCK) >>>> __set_bit(0, (unsigned long *)b); >>>> #endif >>>> +#endif >>>> } >>>> >>>> static inline void hlist_bl_unlock(struct hlist_bl_head *b) >>>> @@ -140,7 +142,9 @@ static inline void hlist_bl_unlock(struct >>>> hlist_bl_head *b) >>>> #ifndef CONFIG_PREEMPT_RT_BASE >>>> __bit_spin_unlock(0, (unsigned long *)b); >>>> #else >>>> +#if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_SPINLOCK) >>>> __clear_bit(0, (unsigned long *)b); >>>> +#endif >>>> raw_spin_unlock(&b->lock); >>>> #endif >>>> } >>>> diff --git a/kernel/workqueue.c b/kernel/workqueue.c >>>> index 0d49ddf..99855b3 100644 >>>> --- a/kernel/workqueue.c >>>> +++ b/kernel/workqueue.c >>>> @@ -42,6 +42,7 @@ >>>> #include <linux/lockdep.h> >>>> #include <linux/idr.h> >>>> #include <linux/locallock.h> >>>> +#include <linux/delay.h> >>>> >>>> #include "workqueue_sched.h" >>>> >>>> @@ -1157,7 +1158,7 @@ static int try_to_grab_pending(struct >>>> work_struct *work, bool is_dwork, >>>> local_unlock_irqrestore(pendingb_lock, *flags); >>>> if (work_is_canceling(work)) >>>> return -ENOENT; >>>> - cpu_relax(); >>>> + cpu_chill(); >>>> return -EAGAIN; >>>> } >>>> >>>> diff --git a/localversion-rt b/localversion-rt >>>> index 9f7d0bd..08b3e75 100644 >>>> --- a/localversion-rt >>>> +++ b/localversion-rt >>>> @@ -1 +1 @@ >>>> --rt13 >>>> +-rt14 >>>> -- >>>> To unsubscribe from this list: send the line "unsubscribe >>>> linux-rt-users" in >>>> the body of a message to [email protected] >>>> More majordomo info at http://vger.kernel.org/majordomo-info.html >>> >>> >>> >>> >>> ------------------------------------------------------------------------------ >>> See everything from the browser to the database with AppDynamics >>> Get end-to-end visibility with application monitoring from >>> AppDynamics >>> Isolate bottlenecks and diagnose root cause in seconds. >>> Start your free trial of AppDynamics Pro today! >>> >>> >>> http://pubads.g.doubleclick.net/gampad/clk?id=48808831&iu=/4140/ostg.clktrk >>> _______________________________________________ >>> Emc-developers mailing list >>> [email protected] >>> https://lists.sourceforge.net/lists/listinfo/emc-developers >> >> >> >> ------------------------------------------------------------------------------ >> See everything from the browser to the database with AppDynamics >> Get end-to-end visibility with application monitoring from >> AppDynamics >> Isolate bottlenecks and diagnose root cause in seconds. >> Start your free trial of AppDynamics Pro today! >> >> http://pubads.g.doubleclick.net/gampad/clk?id=48808831&iu=/4140/ostg.clktrk >> _______________________________________________ >> Emc-developers mailing list >> [email protected] >> https://lists.sourceforge.net/lists/listinfo/emc-developers > > > > ------------------------------------------------------------------------------ > See everything from the browser to the database with AppDynamics > Get end-to-end visibility with application monitoring from > AppDynamics > Isolate bottlenecks and diagnose root cause in seconds. > Start your free trial of AppDynamics Pro today! > > http://pubads.g.doubleclick.net/gampad/clk?id=48808831&iu=/4140/ostg.clktrk > _______________________________________________ > Emc-developers mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/emc-developers ------------------------------------------------------------------------------ See everything from the browser to the database with AppDynamics Get end-to-end visibility with application monitoring from AppDynamics Isolate bottlenecks and diagnose root cause in seconds. Start your free trial of AppDynamics Pro today! http://pubads.g.doubleclick.net/gampad/clk?id=48808831&iu=/4140/ostg.clktrk _______________________________________________ Emc-developers mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-developers
