I know. That’s what the comments in the code say.  The question is *why*.   

You can see why the dependency on the model number is so horrible. Why inflict 
that on userspace?

Regards,
Alan

Alan Altmark
Senior z/VM Software Engineer
IBM zSystems
Endicott, NY         USA

> On Nov 15, 2022, at 1:59 AM, Sven Schnelle <[email protected]> wrote:
> 
> Alan Altmark <[email protected]> writes:
> 
>> Sven asked about this on an internal IBM discussion channel and I
>> mentioned that there are two issues that I can see in the Linux 3270
>> support:
>> 
>> 1.  Linux is definitely doing it wrong on z/VM.  Linux infers the
>> dimensions of the screen from a model number.  (Bzzzzzzt!)  If you
>> have a standard-size screen for model 2, 3, 4, or 5, then the model
>> number from DIAG 0x210 will be correct.  Otherwise it will reflect
>> model 2.  I'm not sure why CP does that or why Linux is using the
>> model number at all.  Even at the turn of the century it was
>> irrelevant.
> 
> The model number is only fetched to present it via the TUB* ioctls to
> userspace. The linux kernel itself only uses the row/column values.

----------------------------------------------------------------------
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO LINUX-390 or visit
http://www2.marist.edu/htbin/wlvindex?LINUX-390

Reply via email to