silRTOS Port on ARM9

>ROB wrote:
>> Few month back, i download silRTOS from
>> http://silcomm.com/silRTOS_V2.2_MIPS32_Port.zip
>> and ran silRTOS on MIPS but have many question/doubt:

>
>The author of silRTOS apparently wants people to ask him by email for 
>downloads.  Weird though that might be, I think it is appropriate to 
>respect that wish and not post direct download links.
>

Sorry, apologize for keeping link here. Please don't use link
http://silcomm.com/silRTOS_V2.2_MIPS32_Port.zip 
which were mentioned to download silRTOS. Please drop mail to silcomm.com
for any download or infomation. 

Thanks David for valuable tips !! 

>You have the source code, and you have access to information from the 
>author.  Look at the code, ask the author, measure the running system. 
>There are very few people, as far as I know, who actually use silRTOS - 
>the website has no information, so the curious developers are very 
>unlikely to give it a second thought.
>
>> 1. Is it safe for critical/safety operations?
>
>Ask the author.  If he says yes, he's either telling the truth - or he 
>is a dangerous amateur in this field.  More likely, he'll say he doesn't

>know of any bugs or problems, but cannot give you any sort of guarantee 
>or certification - if you choose to use it in critical systems, it's up 
>to *you* to be sure it is safe (of course, that's true of any system, 
>even if it is certified).
>
>> 2. Interrupt latency ?
>
>Check the code, ask the author, measure the code.
>
>> 3. How good is the code?
>
>Look at and see - that's why you've got the code.
>
>> 4. Why should i choose this comapre to vxWorks, which is most popular?
>
>vwWorks is probably the "most popular" in certain tightly-specified 
>niches, but since we don't know what you are working on, no one can tell

>you if it is an appropriate choice.  There are vast numbers of embedded 
>RTOS's and non-RT OS's available - the most popular and best choices 
>depend on the requirements.  It's *your* job to figure this out, not 
>ours, and not the author of silRTOS.
>
>> 5. Can source code tell about quality?
>> 
>
>Yes, obviously.  But the source code won't tell you everything - I 
>presume there is plenty of documentation for silRTOS as well (or else 
>why are you bothering to look at it at all?).
>
>> If i ask these question to vendor, they will always praise their RTOS
so
>> wanted to hear from embedded experts.
>> 
>
>It's free software (at least in price - as I say, I don't know the 
>license).  What incentive would the author have to give trumped up 
>claims about the software?  Given the website, or lack of it, it seems 
>he is not too bothered about whether anyone uses the software or not. 
>So try asking him - you've absolutely nothing to lose.  Certainly you 
>won't find many experts in silRTOS here - few people are interested in 
>trying out some secret software available only be email.  Of course, if 
>you can persuade the author to make a website (even just a SourceForge 
>project), provide information about silRTOS, or join comp.arch.embedded,

>then he'd see a lot more interested - we're always happy to learn about 
>new RTOS's.
>
>In the meantime, if you are looking for a safety-critical RTOS, I'd drop

>silRTOS.  Even if it is solid software, the low usage means it has had 
>little real-world testing compared to alternatives.  Look at one of the 
>big commercial systems like vxWorks if the features and price are 
>appropriate, or look at www.freertos.org as an alternative.  They have a

>full-featured open source RTOS, a low-price version of the same system 
>with less license obligations, and a certified safety-critical version. 
>  They also have documentation, a large user base, and an author who 
>will happily answer your questions here in comp.arch.embedded.
>
>

In message <483e5ab1$0$23817$8404b019@news.wineasy.se>, David Brown 
<david@westcontrol.removethisbit.com> writes
>
>> 1. Is it safe for critical/safety operations?
>
>Ask the author.  If he says yes, he's either telling the truth - or he 
>is a dangerous amateur in this field.  More likely, he'll say he 
>doesn't know of any bugs or problems, but cannot give you any sort of 
>guarantee or certification - if you choose to use it in critical 
>systems, it's up to *you* to be sure it is safe (of course, that's true 
>of any system, even if it is certified).

The author will have the certification and all the supporting 
documentation.


>> 5. Can source code tell about quality?

 From the supporting documentation and who did the certification

-- 
\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\
\/\/\/\/\ Chris Hills  Staffs  England     /\/\/\/\/
\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/

Chris H wrote:
> In message <483e5ab1$0$23817$8404b019@news.wineasy.se>, David Brown 
> <david@westcontrol.removethisbit.com> writes
>>
>>> 1. Is it safe for critical/safety operations?
>>
>> Ask the author.  If he says yes, he's either telling the truth - or he 
>> is a dangerous amateur in this field.  More likely, he'll say he 
>> doesn't know of any bugs or problems, but cannot give you any sort of 
>> guarantee or certification - if you choose to use it in critical 
>> systems, it's up to *you* to be sure it is safe (of course, that's 
>> true of any system, even if it is certified).
> 
> The author will have the certification and all the supporting 
> documentation.
> 

Certification and related documents certainly make it much easier to 
ensure that the RTOS is safe and works properly - but you still have to 
do appropriate checking and testing yourself.  If you build an plane 
control system using third-party certified software, and the plane 
crashes, are you going to tell the inquiry that you didn't bother 
testing the software because it was certified safe?  Clearly you have a 
lot more work to do if there are no certifications or other assurances, 
but you have the responsibility in any case.

> 
>>> 5. Can source code tell about quality?
> 
>  From the supporting documentation and who did the certification
>