>Date: Tue, 4 May 1999 12:14:41 -0700 (PDT)
>From: Bill Studenmund <skippy@macro.Stanford.EDU>
>
>As I understand this patch, it's a patch to get X working on a display X
>doesn't currently work on. Also, the failure mode is that X won't work,
>not that trying the X server will crash the machine. :-)

The old Xcfbpmax supported this hardware.  NetBSD/pmax will not be
shipping Xcfbpmax with 1.4.  For those of us with MFB hardware, this
is a "show stopper".  It's a loss of functionality, not missing new
functionality.

With all the lip service being paid to ideas like "all ports are
created equal" and "if it's in the tree, we support it equally on all
platforms" on the current-users list, I don't see how this bug can be
considered anything other than a show-stopper.  You can't convince me
that a failure of Xfree86 to run on, for example, an S3 video card,
wouldn't be considered critical enough to be pulled up to 1.4.

I worked very hard to get a patch out for this as quickly as possible,
so it could be in 1.4.  It's very frustrating to be told that it's
"too late".

When I started using NetBSD, there was a prominent banner on the main
page of www.netbsd.org reading: "We will ship no software before its
time."  I took that as a commitment to ship quality software without
glaring bugs in supposedly "supported" systems.  Perhaps there should
now be a new banner reading, "We will ship anything we have available
in order to make our releases fit the planned schedule, as long as the
i386 port works reasonably well."

This patch doesn't affect any hardware other than the turbochannel
mfb.  It's a very localized change.  Without the change, the hardware
doesn't work.  With the change, it works.  I can't think of any reason
why this patch would need any serious release engineering done on it.
It's a no-brainer.

--
entropy -- it's not just a good idea, it's the second law.