nanogui: Problem running demos when using Framebuffer option
Subject:
Re: Problem running demos when using Framebuffer option
From:
Jordan Crouse ####@####.####
Date:
6 Dec 2000 16:28:44 -0000
Message-Id: <3A2E6A03.F5A39948@censoft.com>
> Anyone out there who actually managed to get
> Microwindows running using the Framebuffer option? :)
Actually, the funny thing is that Microwindows runs *better* in
framebuffer mode. Greg Haerr does all of his development on a
framebuffer machine, and
I currently have three different processors on my desk, all running
Microwindows in framebuffer mode. To tell you the honest truth, I would
be less surprised if you were having trouble running MW in X11 mode.
I have seen the symptoms you are describing sometimes where I have had
an infinite loop or a segfault on the MIPS VR4122 machine I am working
on. I had a situation where Nano-X was getting caught in a race
condition before the graphics had a chance to initalize completely, so
all I saw was some partial drawing. I also saw it on the Ipaq when I
made a stupid keyboard driver error.
What demo are you trying to run?
Jordan
Supriyo Chatterjea wrote:
>
> Hi all,
>
> I was really hoping I wouldn't be having to send this
> mail with all the help that you ppl have
> given...alas...the problem just refuses to go!!!! :)
>
> Like I said earlier on...my Framebuffer has been set
> up properly (I can see the penguin during booting up).
> And I'm not getting any error messages when I try to
> run any of the Microwindows demos.
>
> But the problem is (and this is THE only one right
> now!) everytime I try to run a certain demo, the upper
> portion of the screen just goes blank or starts
> flickering in a weird manner. At times the mouse dies
> as well...but Tasnim's trick:
>
> # /etc/init.d/gpm stop
> # gpm -R -t ms
>
> takes care of the mouse issue.
>
> I've tried the whole procedure on another computer as
> well (since I was thiking that probably my nVidia RIVA
> TNT2 M64 is not compatible with VESA 2.0). This second
> computer has an ATI 3D Rage Pro card...I used the
> usual vesab drivers however....and got the same
> results. I'm gonna try the atyfb driver tomorrow to
> see if that works. Any idea if that might be the
> problem?
>
> One thing that might be useful to know....when I ran
> "landmine" on the ATI 3D Rage Pro card using 0301, the
> whole screen became very dim!!
>
> I also tried Tasnim's suggestion of using config.fb
> instead of config....but that didn't help either! And
> his other suggestion about using 789 instead of 0301
> didn't work as well...somehow it doesn't accept 789 as
> a valid VESA mode.
>
> Anyone out there who actually managed to get
> Microwindows running using the Framebuffer option? :)
>
> Could someone please tell me how I might be able to
> solve this problem? I'm getting kinda desperate! :)
>
> Regards,
> Supriyo
>
> --- Alex Holden ####@####.#### wrote: > On
> Tue, 5 Dec 2000, Tasnim Ahmed wrote:
> > > # gpm -R -t imps2
> >
> > Just to clarify this, you need to run gpm in
> > repeater mode by starting it
> > with the -R flag. This causes it to write mouse data
> > in mouse-systems
> > format (regardless of the kind of mouse) to a fifo
> > at /dev/gpmdata whilst
> > in graphics mode. There is a problem with this in
> > that the mouse-systems
> > format doesn't include any wheel data, so in order
> > to use the wheel under
> > X, you need to use the native X mouse driver in
> > IMPS/2 mode. But if you're
> > already running gpm in repeater mode, it will also
> > try to read /dev/psaux
> > when in graphics mode (at the same time as X) and
> > the mouse data will be
> > corrupted. So if you use gpm in repeater mode and
> > tell X to use
> > /dev/gpmdata you can't use the wheel, and if you use
> > gpm in non repeater
> > mode and tell X to use /dev/psaux you can't use
> > Nano-X in framebuffer mode
> > without restarting gpm. Long term, we should really
> > write our own mouse
> > drivers- the gpm code is a mess anyway, plus a
> > typical embedded system
> > wouldn't use it so it's a waste of resources
> > (memory, flash or disk
> > space, and processor time). We'll also need to do it
> > to get wheel support.
> >
> > --
> > ------- Alex Holden -------
> > http://www.linuxhacker.org/
> > http://www.robogeeks.org/
> >
>
> __________________________________________________
> Do You Yahoo!?
> Yahoo! Mail - Free email you can access from anywhere!
> http://mail.yahoo.com.sg/
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: ####@####.####
> For additional commands, e-mail: ####@####.####
