>> Are the specifications for a winmodem a secret? If not, it
>> seems to me that a Linux winmodem driver would be a Simple Matter Of
>> Programming (tm). I could imagine timing needs that might require one
>> to extend Linux interrupt handling or scheduling a tiny bit, but that
>> is about the only special need that comes to mind. Maybe the issue
>> is that you have to implement some huge modem protocol stack? If more
>> of the modem must be implemented in software, I wonder if that would
>> enable one to write interesting extensions, such as encodings optimized
>> for low latency for interactive video games.
>
> Unfortunatelly there is no standards for winmodem :-(( Only API-level
> standards required for windows-driver... This means also that winmodems
> are incompatible on hardware level :-((
It's not quite that bad. With just two drivers, Linux could support most
modems. Hardware vendors mostly buy chips and drivers from Rockwell,
and they don't like writing drivers any more than we do.
Some modem types:
1. Modems based on Rockwell's RPI chipset. These modems accept some
standard AT commands for identification. The OS only needs to supply
compression and error correction. (these are common)
2. U.S. Robotics WinModems, well-known for their lack of a normal UART.
I'm fairly sure these are like RPI modems. (these are common)
3. IBM's MWave, maybe also used/developed by AT&T. The OS must swap
DSP code in/out of the device as needed, but other than that it
seems to be a normal modem. (these are found on IBM hardware)
4. Apple one made a piece of junk (the geomodem?) that wasn't a modem.
Unlike the others I've listed, it did not MODulate and DEModulate.
Microsoft calls these "software modems", which is accurate. We'd
need something like the soundmodem driver for it. (uncommon)
About that patent: the Linux drivers can just refuse to negotiate
compression. It's useless for transferring gzipped files anyway and
it adds latency.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu