[ANNOUNCE] 3.4.9-rt17

Dear RT Folks,

I'm pleased to announce the 3.4.9-rt17 stable release.


This release is just an update to the new stable 3.4.9 version
and no RT specific changes have been made.


You can get this release via the git tree at:

  git://git.kernel.org/pub/scm/linux/kernel/git/rt/linux-stable-rt.git

  Head SHA1: 7050e0bb86b8500446cc7379cd9bcbeff57ecd87


Or to build 3.4.9-rt17 directly, the following patches should be applied:

  http://www.kernel.org/pub/linux/kernel/v3.x/linux-3.4.tar.xz

  http://www.kernel.org/pub/linux/kernel/v3.x/patch-3.4.9.xz

  http://www.kernel.org/pub/linux/kernel/projects/rt/3.4/patch-3.4.9-rt17.patch.xz



Enjoy,

-- Steve

Dear Linux-Realtimers

I just had the time to test this realtime release. Due to time constraints 
this is the first test on the 3.4 rt series. The HW Platform i used for testing 
is a ARM PCM043 i.mx35 with some tweaks to decrease latency. One of them is 
using an interrupt to drive our system timer and not using hr timers.
But these tweaks are essentially the same between 3.0.30-rt50 and the now 
tested 3.4.9-rt17. The only thing which is not applied on 3.4.9-rt17 is 
switching the cpu to idle (which gives cold caches and bad latency), since the 
old location was removed an i didn't found the new location but pm is switch 
of in the 3.4 kernelconfig as attached.

The performance of the 3.4.9-rt17 is much much worse than the 3.0.30-rt50 
kernel. While the 3.0 kernel was able to keep up with a 2400khz interrupt 
source in usermode the 3.4 kernel drops out by a wide margin (6ms+?). I even 
thought on the first moment that i didn't enable full realtime...

So if any of you have some comments why the latency performance degraded this 
far or if you have a hint for me where to find the arch_idle function which has 
been located in arch/arm/plat-mxc/include/mach/system.h so the config between 
3.0 and 3.4 is as identical as it can be.

Attached is the kernel config for reference.

Best regards
Tim

PS: please ignore the legalese below this line, its added automatically by the 
mail gateway :
Hottinger Baldwin Messtechnik GmbH, Im Tiefen See 45, 64293 Darmstadt, Germany | www.hbm.com 

Registered as GmbH (German limited liability corporation) in the commercial register at the local court of Darmstadt, HRB 1147  
Company domiciled in Darmstadt | CEO: Andreas Huellhorst | Chairman of the board: James Charles Webster

Als Gesellschaft mit beschraenkter Haftung eingetragen im Handelsregister des Amtsgerichts Darmstadt unter HRB 1147 
Sitz der Gesellschaft: Darmstadt | Geschaeftsfuehrung: Andreas Huellhorst | Aufsichtsratsvorsitzender: James Charles Webster

The information in this email is confidential. It is intended solely for the addressee. If you are not the intended recipient, please let me know and delete this email.

Die in dieser E-Mail enthaltene Information ist vertraulich und lediglich fur den Empfaenger bestimmt. Sollten Sie nicht der eigentliche Empfaenger sein, informieren Sie mich bitte kurz und loeschen diese E-Mail.

Hi Rt-Folks

I just found out that the name of the rt irq of the fec device changed so it 
had the same rtprio as my measurement interrupt. With the rt prios fixed the 
performance seems to be roughly like the 3.0 kernel. If will give a further 
update if there are any findings.

Best regards
Tim

Hottinger Baldwin Messtechnik GmbH, Im Tiefen See 45, 64293 Darmstadt, Germany | www.hbm.com 

Registered as GmbH (German limited liability corporation) in the commercial register at the local court of Darmstadt, HRB 1147  
Company domiciled in Darmstadt | CEO: Andreas Huellhorst | Chairman of the board: James Charles Webster

Als Gesellschaft mit beschraenkter Haftung eingetragen im Handelsregister des Amtsgerichts Darmstadt unter HRB 1147 
Sitz der Gesellschaft: Darmstadt | Geschaeftsfuehrung: Andreas Huellhorst | Aufsichtsratsvorsitzender: James Charles Webster

The information in this email is confidential. It is intended solely for the addressee. If you are not the intended recipient, please let me know and delete this email.

Die in dieser E-Mail enthaltene Information ist vertraulich und lediglich fur den Empfaenger bestimmt. Sollten Sie nicht der eigentliche Empfaenger sein, informieren Sie mich bitte kurz und loeschen diese E-Mail.

Hello,

When applying the rt patches, the OMAP SPI driver referenced below 
doesn't work well. It's working well without the rt patches.

http://git.kernel.org/?p=linux/kernel/git/tmlind/linux-omap.git;a=blob;f=drivers/spi/omap2_mcspi.c;h=969cdd2fe124602d1e65c4218190ed3e7d78ccf6;hb=02f8c6aee8df3cdc935e9bdd4f2d020306035dbe

More precisely, I'm getting a lot of "TXS timed out" and "RXS timed out" 
on line 485/495.

Q1. Has anyone made work OMAP SPI driver with rt patches?

Q2. How can I debug such problem? I have converted all the spin_lock_irq 
to raw_spin_lock_* without being sure if it's a good idea. But in any 
case, the result is the same.

Q3. More generically, while considering a new driver, what needs to be 
watched and patched from the rt point of view to play nicely with the 
real-time architecture? All the mutex? Anything in the worker function? 
I know that I won't become a RT expert overnight, but I'm trying to 
understand.

Many thanks in advance for any help or pointer,

Grégoire

--
To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Hi 

After fixing the realtime prios which got shuffled due to other irq thread names
I just tested the realtime latency for interrupts on the i.mx35. With our own 
systemtimer which does not get into the way of  the measurement interrupt this 
kernel has 30µs worst time (caveat: short measurement period) interrupt 
latency and usermode can keep up with a 2400kHz interrupt rate. While the 
usermode performance seems to be largely unchanged (but i haven't tested very 
thoroughly) the interrupt latency went down about 10µs coming from the 3.0-rt 
series. Its really nice to see that kind of improvement. 

Thanks
Tim

Ignore the legalese please:


Hottinger Baldwin Messtechnik GmbH, Im Tiefen See 45, 64293 Darmstadt, Germany | www.hbm.com 

Registered as GmbH (German limited liability corporation) in the commercial register at the local court of Darmstadt, HRB 1147  
Company domiciled in Darmstadt | CEO: Andreas Huellhorst | Chairman of the board: James Charles Webster

Als Gesellschaft mit beschraenkter Haftung eingetragen im Handelsregister des Amtsgerichts Darmstadt unter HRB 1147 
Sitz der Gesellschaft: Darmstadt | Geschaeftsfuehrung: Andreas Huellhorst | Aufsichtsratsvorsitzender: James Charles Webster

The information in this email is confidential. It is intended solely for the addressee. If you are not the intended recipient, please let me know and delete this email.

Die in dieser E-Mail enthaltene Information ist vertraulich und lediglich für den Empfaenger bestimmt. Sollten Sie nicht der eigentliche Empfaenger sein, informieren Sie mich bitte kurz und loeschen diese E-Mail.