Re: [rtl] Soft Real Time
On Sat, 24 Jun 2000 [EMAIL PROTECTED] wrote: On Sun, Jun 25, 2000 at 12:04:36AM +0200, Herman Bruyninckx wrote: We have a control problem in which we need a simple controller to run at preferably 30 000 Hz... So, keeping everything in cache is really needed... Do you really need SMP _PowerPC_ ? If not we have something to beta test on SMP x86. We don't need PowerPC (that part of the thread did not apply to my reply). I have a dual Pentium III 700Mhz, and I wanted to investigate how it compares to DSP performance, when everything can be kept in cache. -- [EMAIL PROTECTED] (Ph.D.)Fax: +32-(0)16-32 29 87 Dept. Mechanical Eng., Div. PMA, Katholieke Universiteit Leuven, Belgium -- [rtl] --- To unsubscribe: echo "unsubscribe rtl" | mail [EMAIL PROTECTED] OR echo "unsubscribe rtl Your_email" | mail [EMAIL PROTECTED] --- For more information on Real-Time Linux see: http://www.rtlinux.org/rtlinux/
Re: [rtl] Soft Real Time
On Sun, Jun 25, 2000 at 09:04:26AM -0700, al lykken wrote: [synergy ppc boards] I have not spent a lot of time getting RTLinux working because I was told that it was not ready. I will certainly re-visit the issue after all the comments. We had a joint project with Synergy to make RTL run on the Synergy boards. Synergy later decided to focus their attention on the hardware and we got distracted with other projects, so we left it less than complete. However, PPC RTLinux worked first on UP Synergy boards. - Victor Yodaiken FSMLabs: www.fsmlabs.com www.rtlinux.com FSMLabs is a servicemark and a service of VJY Associates L.L.C, New Mexico. -- [rtl] --- To unsubscribe: echo "unsubscribe rtl" | mail [EMAIL PROTECTED] OR echo "unsubscribe rtl Your_email" | mail [EMAIL PROTECTED] --- For more information on Real-Time Linux see: http://www.rtlinux.org/rtlinux/
Re: [rtl] Soft Real Time
We have an experimental RTL V3 with an option to turn off Linux on one PC. Measured latencies are very low -- but this is early code. If you are interested in testing please send email to me. On Sun, Jun 25, 2000 at 08:42:57PM +0200, Bernhard Kuhn wrote: Herman Bruyninckx wrote: I have a dual Pentium III 700Mhz, and I wanted to investigate how it compares to DSP performance, when everything can be kept in cache. Keeping things in cache is only half of the way ... the bigger problem is to bother around with latencies caused by the pci-architecture ... with modern chip-sets, it´s hard to tell what exactly is going on ... several things have to be taken into account: 1. CPU to Host-Bridge latency: you will have to disable caching and write combinig in case your I/O card is memory mapped. Even then, you have latencies when sharing the CPU-bus with another processor. 2. Host-Bridge to PCI-Bus latency: you might disable the read/write fifo (usual depth: 64) of the Host-Bridge. 3. you realy should disable PCI-Burst operations, which can by up to 64 cycles. Otherwise having, for example, five PCI-device, the arbitration could go up to 10 µs in this stage. 4. Latencys caused be the I/O-Card itself If a worst case latency of about 20 µs are just fine for your application, then stay with RTL and standard settings. Otherwise it´s going deep into details: One solution could be the mentioned idea with the second OS on the second CPU. Another method would be to disallow any kind of linux-kernel activities on the second processor. Some times ago, i took a look into the code of the kernel scheduler ... it should be feasabel to keep away user-space processes and kernel threads from the second processor by modifiying the code a little bit. Linux-Interrupts can be directed the the first CPU by simply reprograming the I/O-APIC, as far as i got it. So the second CPU should completly belongs to your RTL-application, that even could fit into the L1-Cache of the CPU. If this didn´t scared you then go reading on: Getting rid of the PCI-latencies is a little bit more difficult: You could use the three-wire APIC protocoll to attach a special I/O-Card directly onto the CPU. The maximum latency/jitter is less then one microsecond in this case, but then you have to bother around with a 100 MHz serial line, simulation a local APIC ... Just a dream ... Bernhard -- [rtl] --- To unsubscribe: echo "unsubscribe rtl" | mail [EMAIL PROTECTED] OR echo "unsubscribe rtl Your_email" | mail [EMAIL PROTECTED] --- For more information on Real-Time Linux see: http://www.rtlinux.org/rtlinux/ -- - Victor Yodaiken FSMLabs: www.fsmlabs.com www.rtlinux.com FSMLabs is a servicemark and a service of VJY Associates L.L.C, New Mexico. -- [rtl] --- To unsubscribe: echo "unsubscribe rtl" | mail [EMAIL PROTECTED] OR echo "unsubscribe rtl Your_email" | mail [EMAIL PROTECTED] --- For more information on Real-Time Linux see: http://www.rtlinux.org/rtlinux/
Re: [rtl] Soft Real Time
[EMAIL PROTECTED] wrote: We have an experimental RTL V3 with an option to turn off Linux on one PC. Measured latencies are very low -- but this is early code. If you are interested in testing please send email to me. Very interessting! I´d like to have a look at it, because i intend to modify the kernel-scheduler in the following way (if i ever have enough time to do so :-) ) o user-spaces processes, kernel-threads and linux-interrupts should be fixable to specific CPUs. o Additionaly to the usual time-sharing "nice-value", a priority is assigned to user-space processes, resulting in somewhat i would call a "hierarchical priority encoded preemptiv time sharing scheduler" :-) This surely wouldn´t apply to machine control systems, but is intended for multimedia (happy fragging!) ... i know that there are several other projects out there, that care especialy about the second point (red linux, qlinux), but AFAIK, nobody cares about the first one ... BTW: how was you talk in San Jose? Bernhard -- [rtl] --- To unsubscribe: echo "unsubscribe rtl" | mail [EMAIL PROTECTED] OR echo "unsubscribe rtl Your_email" | mail [EMAIL PROTECTED] --- For more information on Real-Time Linux see: http://www.rtlinux.org/rtlinux/
Re: [rtl] Soft Real Time
On Sun, Jun 25, 2000 at 10:49:11PM +0200, Bernhard Kuhn wrote: [EMAIL PROTECTED] wrote: We have an experimental RTL V3 with an option to turn off Linux on one PC. Measured latencies are very low -- but this is early code. If you are interested in testing please send email to me. Very interessting! I´d like to have a look at it, because i intend to modify the kernel-scheduler in the following way (if i ever have enough time to do so :-) ) o user-spaces processes, kernel-threads and linux-interrupts should be fixable to specific CPUs. Ingo did Linux interrupts already for 2.3 x86. The others should be easy. BTW: We are under enormous pressure from potential funders who argue that we need to abandon open source entirely to keep Lineo from simply riding us down. I expect RTAI in the future will have done PPC ports before RTLinux and etc. I'm attempting to manouver around both this problem and funder demands. We will see. o Additionaly to the usual time-sharing "nice-value", a priority is assigned to user-space processes, resulting in somewhat i would call a "hierarchical priority encoded preemptiv time sharing scheduler" :-) I have a solution to this, but [see above] This surely wouldn´t apply to machine control systems, but is intended for multimedia (happy fragging!) ... i know that there are several other projects out there, that care especialy about the second point (red linux, qlinux), but AFAIK, nobody cares about the first one ... RedLinux contains RTLinux 0.9 code, I don't know the others. But [see above] BTW: how was you talk in San Jose? Packed. Very interesting. -- - Victor Yodaiken FSMLabs: www.fsmlabs.com www.rtlinux.com FSMLabs is a servicemark and a service of VJY Associates L.L.C, New Mexico. -- [rtl] --- To unsubscribe: echo "unsubscribe rtl" | mail [EMAIL PROTECTED] OR echo "unsubscribe rtl Your_email" | mail [EMAIL PROTECTED] --- For more information on Real-Time Linux see: http://www.rtlinux.org/rtlinux/
Re: [rtl] Soft Real Time
[EMAIL PROTECTED] wrote: BTW: We are under enormous pressure from potential funders who argue that we need to abandon open source entirely to keep Lineo from simply riding us down. I expect RTAI in the future will have done PPC ports before RTLinux and etc. I'm attempting to manouver around both this problem and funder demands. We will see. hmm ... I am allways wondering why customers prefer wasting their money for licences instead paying for good services ... that´s that basic Idea with linux and with mechanisms like the web: any good line of code can be reproduced million times without any costs ... the good lines will sum up over time and that´s it! The earlier the pointy haired bosses get used to this, the earlier they will understand what´s going on: if any feature is realy interessting for any application, then how long will it take until somebody will reprogram the hidden code because he needs it or just because he is courious or doesn´t know how to spend his time in another way? :-) If the code is too special and only applies for a few customers, then nobody will care wether the code is available or not ... You know that all this rt-stuff is no rocket technology and can be performed by a small group of people (or even by a single person) within a few month ... That´s why Linux scares QNX and Lynx (Ah, sorry LynuxWorks): They are just jumping onto the open source train like many other companies to *protect* their invests: If somebody has published his code long befor you, then he´s the boss. It´s more like in the scientific world: publish or parish! Coming back to Lineo: although it´s a really strange company to me, they also recogniced the need to publish their stuff ... To sum it up: sell services, not products! Ok, this is a little bit more sophisticated, but as long as you can´t delete any line of open source code, time can´t be turned back! just my 0.02 euro ... Bernhard -- [rtl] --- To unsubscribe: echo "unsubscribe rtl" | mail [EMAIL PROTECTED] OR echo "unsubscribe rtl Your_email" | mail [EMAIL PROTECTED] --- For more information on Real-Time Linux see: http://www.rtlinux.org/rtlinux/
