ATI Drivers report bogus dot_clock to xvidtune

[Alex Deucher]

On 3/28/06, Roland Scheidegger <rscheidegger_lists at hispeed.ch> wrote:
> Greg Stark wrote:
> > When I ran xvidtune It reports the dot clock is 1976225.70Mhz. Of course
> > that's wrong, my graphics card is not capable of driving a 1.9 petahertz
> > signal...
> >
> > I suspect this is the root cause of the problems I have and another user has
> > reported with using mythtv with an ATI card. Since mythtv attempts to sync to
> > the vertical refresh it gets *very* confused trying to match this refresh
> > rate.
> >
> > I'm not sure if this is a display bug in xvidtune or a bug in the driver so I
> > wrote a little test program to print the output from XF86VidModeGetModeLine:
> >
> > dot_clock: 1976225695

I also don't see how the dotclock would be of any use for vblank, but
I suspect you are seeing a limitation of mergedfb.  Mergedfb is
basically a hack to treat 2 crtcs as a single screen.  since you have
two crtcs with different timings there's no way to know which dotclock
or refresh rate, etc. is relevant (since they both are). Hence you end
up with some weirdness for certain fields.  Disable mergedfb if you
want relevant numbers for those fields.

Alex

> This looks like a driver bug. I have no idea though what driver you're
> talking about, there is no such thing as a ati driver (well there is
> technically, a driver wrapper which is called ati, but it doesn't do
> anyhting itself). So is this mach64? rage128? radeon? fglrx (take it to
> ati in that case)?
> If that's radeon are you using mergedfb? It works just fine for me
> without mergedfb.
> That said, I don't know why mythtv would need the pixel clock to sync to
> vblank, I'd think it's a pretty useless number for that.
>
> Roland