I should add, since a couple of people have emailed directly about
possible bandwidth problems, that I can reach pretty much full utilisation
of the 115kbaud bandwidth. For instance I can send 1 byte and receive 8
bytes back (which works out at 99kbaud with 1 start and 2 stop bits) at
1kHz, so the delay seems (to me and I don't know much about this) to be
specifically in switching between read and write, or more
likely in recognizing that data has arrived (ie the interrupt). Any
suggestions welcome...
Cheers,
Richard.
On Fri, 9 Mar 2001, Richard Reeve wrote:
> Hi,
>
> I have been trying to do some 2 way communication using the serial port,
> and although I can get communications to run a lot faster than they did
> under linux, there still seems to be a limit as to how fast I can switch
> between read and write using rt_com. Under linux the latency was about
> 10ms, so I could only communicate at about 50Hz (send, 10ms switch,
> receive, 10ms switch, ignoring the time for signal), using rt_com, I don't
> seem to be able to beat 1kHz (thus about .5ms for switch and send, .5ms
> switch and receive, making a bit over .25ms latency).
>
> Is this normal? I would like to be able to communicate up to 2kHz ideally,
> as this is closer to the theoretical baud rate limit (115200 baud) given
> the amount of data I'm transmitting (1 byte send, 4 receive), but I
> haven't been able to speed it up, despite playing around with the rt_com
> code itself. Does anyone have any ideas?
>
> Cheers,
>
> Richard Reeve
> University of Stirling
> Stirling
> UK
>
> -- [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/
>
-- [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/
