Re: [hylafax-users] Linux and RTOS

[Lee Howard <faxguy@xxxxxxxxxxxxxxxx>]

Michael Hallager wrote:
> > Are you having a problem for which you believe that you need a "fancy"
> > serial controller?
> >
> > Truth told there is a lot that has changed.  Serial communications were
> > important 10 years ago.  External serial modems work better in stress
> > testing on Linux 2.2 than they do in Linux 2.6.
> >
> > However, the reality is that you're not likely to notice any difference
> > unless you're pushing the envelope.
> >     
>
> Define (In broad terms) what "pushing the envelope" is please Lee.

In Class 2.0/2.1 with ECM you can, in theory and in rare real-world 
practice, have a full ECM block (64KB) go from the DCE to the DTE all at 
once (instead of in small 256-byte nibbles, as would usually be the 
case).  So if your DTE-DCE speed is 115200 bps then you have a much 
greater chance of a buffer overflow if serial controller's flow control 
is dysfunctional than if it were 2400 bps.  However, you cannot 
typically operate fax modems with a DTE-DCE speed of 2400 bps.  Usually 
it needs to be at 19200 bps for non-V.34 scenarios and 57600 bps for 
V.34-scenarios.  So there is some risk there.

"Pushing the envelope" is when you exaggerate the conditions that cause 
risk to be something well-beyond what one would expect in an average 
situation.

Thanks,

Lee.


____________________ HylaFAX(tm) Users Mailing List _______________________
 To subscribe/unsubscribe, click http://lists.hylafax.org/cgi-bin/lsg2.cgi
On UNIX: mail -s unsubscribe hylafax-users-request@xxxxxxxxxxx < /dev/null
 *To learn about commercial HylaFAX(tm) support, mail sales@xxxxxxxxx*