Alan Altmark <[EMAIL PROTECTED]> writes:
> Nah.  Instead of VCTC, use z/VM 4.2 and guest LANs with virtual
> HiperSockets.

Did I miss something?  The "zSeries HiperSockets" Redpaper
said that HiperSockets are QDIO devices, and that they use
the binary-only "qeth" device driver.  If not, Nick's still
right in suggesting a split between kernel-release-sensitive
(i.e. source) and IBM-proprietary (i.e. binary) portions.

>                WDNYADD.  (We don't need yet another device driver).

In the general case, I agree.  In the case at hand, it would seem
otherwise.

> I don't mean to say that the driver's design won't change (I
> don't know),
> but keep in mind that such a change isn't free, and that
> changing it is a
> business decision, not a technical one.

Sure.  And hopefully IBM's business partners, SuSE and Red Hat at
least, will remind IBM of why it would be good for business, both
theirs and IBM's.

Ross Patterson
Computer Associates

Reply via email to