1kHz timer behavior on Cyclone V

[Janet Schneider]

Hi everyone,


I ran more tests, and removed prints from the ISR unless they went out of bounds for what we are expecting.  The results were quite good - I did not see a latency greater than one microsecond on an hour run with a 1ms timer.  Thanks for the pointers!


Janet

________________________________
From: users <users-bounces at rtems.org> on behalf of Thomas Doerfler (nt) <Thomas.Doerfler at imd-systems.de>
Sent: Thursday, December 8, 2016 10:58:45 PM
To: users at rtems.org
Subject: Re: 1kHz timer behavior on Cyclone V

Hi,

it's even worse: since the transfer of one character (8 bits) requires a
leading startbit and a trailing stopbit (= total of 10 bits), you can
only print 11.52 characters per 1ms.

Am 09.12.2016 um 07:48 schrieb Sebastian Huber:
> Hello Janet,
>
> a printk() inside an ISR is quite bad. Lets assume it uses a serial
> line with 115200 bits per second, then you can print 14.4 characters
> per 1ms.
>


--
IMD Ingenieurbuero fuer Microcomputertechnik
Thomas Doerfler           Herbststrasse 8
D-82178 Puchheim          Germany
email:    Thomas.Doerfler at imd-systems.de
PGP public key available on request

_______________________________________________
users mailing list
users at rtems.org
https://urldefense.proofpoint.com/v2/url?u=http-3A__lists.rtems.org_mailman_listinfo_users&d=DgICAg&c=5VD0RTtNlTh3ycd41b3MUw&r=ipHQnEo1XnvLdE9wWxwn7A&m=gGtMRvkv_KU_QKlX4Ok-OJFsK2YEYe87e5XZRSdxx1M&s=TmBU5SG2_DqB5X_BRiO02uesLXme42UeKriGSBwt2kI&e=
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rtems.org/pipermail/users/attachments/20161214/786334f2/attachment-0002.html>