Re: [ANNOUNCE] v4.19.50-cip3-rt2


Zoran
 

Your log is corrupted, see?
You are correct, I should write that these are preliminary WTH/ad-hoc
results (I need to fix/rerun these in the correct setup, I know).

It is usec, not nsec, AFAICT.
Again, correct. I wrote nsec instead usec, having in mind INTEL
technology per asm instruction. INTEL illusions/delusions in my brain.
I am trying to catch up on armv7, A8 and A9, since these gadgets
started ruling the embedded industry.

Thank you,
Zoran
_______

On Fri, Jul 26, 2019 at 11:05 PM Pavel Machek <pavel@denx.de> wrote:

On Wed 2019-07-24 18:06:46, Zoran S wrote:
(limitations): I could NOT connect with ETH to target (something is
wrongly set with my local.conf for YOCTO rootfs)...

Here is my approximate recipe for rootfs I am using for the test
(initramfs) for your inspection (it is thud, local.conf moved to
warrior):
https://github.com/ZoranStojsavljevic/bbb-yocto/blob/master/bbb-releases/bbb-warrior/local.conf

More or/and less, here are results after approximately 63 minutes of
running rt-tests:

root@beaglebone:~# uname -a
Linux beaglebone 4.19.13-cip1-rt1 #1 PREEMPT RT Tue Jul 23 12:18:18
GMT 2019 armv7l GNU/Linux
_______
policy: fifo: loadavg: 291.28 287.63 267.78 21/486 30014
policy: fifo: loadavg: 186.97 215.90 243.41 5/458 22788
policy: fifo: loadavg: 248.16 243.95 244.63 194/486 8800
policy: fifo: loadavg: 283.55 240.50 243.74 9/318 2383
T: 0 (13903) P:80 I:1000 C:3942593 Min: 19 Act: 52 Avg: 45 Max: 93
T: 0 (13903) P:80 I:1000 C:3785664 Min: 19 Act: 44 Avg: 45 Max: 935
T: 0 (13903) P:80 I:1000 C:3716107 Min: 19 Act: 60 Avg: 45 Max: 936
policy: fifo: loadavg: 222.11 176.72 90.66 5/213 21249
T: 0 (13903) P:80 I:1000 C:2213895 Min: 19 Act: 38 Avg: 46 Max: 905
T: 0 (13903) P:80 I:1000 C: 181403 Min: 21 Act: 49 Avg: 45 Max: 90
T: 0 (13903) P:80 I:1000 C: 181320 Min: 21 Act: 29 Avg: 45 Max: 90
Your log is corrupted, see?

If it was real, it would be

Max: 905
Max: 90

But you have

Max: 905
Max: 90

Notice the whitespace.

MIN 19 ns, MAX 936 ns (3x over 900 ns)... I would be really concerned
about MAX over > 900 ns!
It is usec, not nsec, AFAICT.

Best regards,
Pavel

--
DENX Software Engineering GmbH, Managing Director: Wolfgang Denk
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany

Join cip-dev@lists.cip-project.org to automatically receive all group messages.