MMU handling page-fault in embedded RTOS system

Hi all,

I am new to this forum.
anyone here has experience working with an MMU which handles page-fault
in RTOS embedded system?
Is there a contradiction between page-fault handling (a case where not
all the code is resident in RAM/ROM) and RTOS?

I would really appritiate any help on this.

MajorSoul@gmail.com wrote:
>anyone here has experience working with an MMU which handles page-fault
>in RTOS embedded system?
>Is there a contradiction between page-fault handling (a case where not
>all the code is resident in RAM/ROM) and RTOS?

The obvious "contradiction" is the possible insertion of an indefinite
time delay at the point the page fault needs to be handled.
Many systems will allow you to lock certain pages in memory, to avoid
that from happening.

Ok, I understand what you are saying.
What do u mean certain memory ? Dont you mean all memory? if certain
amount of memory will not reside in RAM will still have this
contradiction you mentioned, no ?

MajorSoul@gmail.com wrote:
> 
> I am new to this forum.
> anyone here has experience working with an MMU which handles
> page-fault in RTOS embedded system?
> Is there a contradiction between page-fault handling (a case where
> not all the code is resident in RAM/ROM) and RTOS?

I don't have any help for you, but want to correct your
misapprehension.  This is not a forum.  It is the usenet news
system, and google is only a rather poor interface to that system. 
Please read the following links.

-- 
 Some informative links:
   news:news.announce.newusers
   http://www.geocities.com/nnqweb/
   http://www.catb.org/~esr/faqs/smart-questions.html
   http://www.caliburn.nl/topposting.html
   http://www.netmeister.org/news/learn2quote.html

MajorSoul@gmail.com wrote:
>Roberto Waltman wrote:
>> MajorSoul@gmail.com wrote:
>> 
>>>anyone here has experience working with an MMU which handles page-fault
>>>in RTOS embedded system?
>>>Is there a contradiction between page-fault handling (a case where not
>>>all the code is resident in RAM/ROM) and RTOS?
>> 
>> The obvious "contradiction" is the possible insertion of an indefinite
>> time delay at the point the page fault needs to be handled.
>> Many systems will allow you to lock certain pages in memory, to avoid
>> that from happening.
 >
> Ok, I understand what you are saying.
> What do u mean certain memory ? Dont you mean all memory? 

Please quote the relevant portion of the message that you are replying 
to.  The above context was restored.  Newsgroup etiquette is to avoid 
IM-style abbreviations -- use proper spelling, capitalization, sentence 
structure, and punctuation.

Roberto said "certain pages", not "certain memory".  He meant that 
time-critical portions of code would not be overwritten by the MMU, so 
that it could be executed without the delay of of swapping it into 
memory.  "Certain pages" meant the the selected memory pages that 
contains the time-critical code.  Locking these into memory means not 
allowing them to be overwritten.

-- 
Thad