Re: Newport graphics - testers & input needed
Frank Wille <[email protected]>
| Newsgroups | gmane.os.netbsd.ports.sgimips |
|---|---|
| Organization | Privates Internet Ostwestfalen/Lippe e.V. |
| Message-ID | <[email protected]> |
Michael wrote:
> there's a weirdness in the way the newport's VC2 chip generates the
> cursor sprite - depending on chip revision, board revision and FSM
> knows what else different values have to be added to the X coordinate.
> The existing logic failed on my R5k Indy with 24bit Newport so I
> adapted it.
Same here. I already reported the problem, and a possible fix to it, a long
time ago: http://www.netbsd.org/cgi-bin/query-pr-single.pl?number=34759
> So, if you have an Indy with 8bit or 24bit newport please post your
> board / chip revisions and the appropriate X offset for your hardware.
> I'll start with mine:
>
> 24bit Newport, board rev. 1, CMAP rev. 5, XMAP rev. 5, VC2 rev. 0,
> offset 29
I have a different board rev., otherwise exactly the same (24bit Newport):
newport0 at gio0: SGI NG1 (board revision 5, cmap revision 5, xmap revision 5, vc2 revision 0), depth 8
And I also need offset 29!
> besides that, I fixed a few bugs in the newport driver:
> - - downwards scrolling now works
> - - scrolls and fills no longer cover an extra line of character cells
> So, text editors are now usable in the console.
Great! Works fine for me.
I saw that you applied exactly the same patch to gio/newport.c which I
suggested in my PR #34759! :)
--
_ Frank Wille ([email protected])
_ // http://sun.hasenbraten.de/~frank/
\X/ Phx @ #AmigaGer