[Rtai] fifo behaviour

[Bernhard Pfund]

Just to anticipate Paolo's answer: Create a standalone example showing 
the problem, otherwise nobody will be able to help you.

Creating such examples often helps to regain an overview of your code 
and things can clear up, so eventually you'll be able to solve your 
issues yourself.

Developing low-level code such as kernel modules is complex enough and 
RTAI is not something you can master in two weeks. So be patient, take 
small steps and try to contribute your insights to the community. Be 
specific regarding your questions and kind in the way you ask them...

My 0.02$

Bernhard


Venkatesh Rao wrote:
> we are using fifos for communication between a task running in kernel 
> module and a user space program. the kernel module writes 3 double 
> variables and one unsigned integer to the fifo every 40 microseconds.  
> all these variables are part of a structure. on executing the program 
> samples in a particular region are corrupted for all the variables. 
> whenever we run the program, the error occurs in the same region and 
> thereafter the samples get their correct values. this is puzzling.
> 
> when we executed the loop writing to the fifo a fixed number of times 
> say 500 times, then different no. of values were obtained by the user 
> program. for instance if double var1, var2 were written 500 times to the 
> fifo, the user program obtained different no. of samples
> 
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> Rtai mailing list
> Rtai at rtai.org
> https://mail.rtai.org/cgi-bin/mailman/listinfo/rtai