Bug 42852

Your chip requires KMS to work properly.

Em Sáb, 2011-11-12 às 16:21 +0000, bugzilla-daemon@freedesktop.org
escreveu:

> https://bugs.freedesktop.org/show_bug.cgi?id=42852
> 
> --- Comment #1 from Alex Deucher <agd5f@yahoo.com> 2011-11-12 08:21:46 PST ---
> Your chip requires KMS to work properly.
> 


FreeBSD does not have KMS.. 
On another notebook Radeon XPRESS 200M 5955 
it works ok...

As I said in the bug report, the chip works very good Only after power
on,
and the driver "ati" or "radeon" is not aware of KMS either...

Seems to me that the screen memory is not align propper with
the registers on the chipset, a propper align situation  occurs only
after power on,
or after about 10 resets.. 

A propper reset or initial program in the chip (using the ati, or radeon
driver)
would resolve the problem...

Can someone please point me to a documentation of the radeon 6250 that
explains how to reset the chip to power on state???

Thanks for any help,

Sergio

(In reply to comment #2)
> Em Sáb, 2011-11-12 às 16:21 +0000, bugzilla-daemon@freedesktop.org
> escreveu:

> FreeBSD does not have KMS.. 
> On another notebook Radeon XPRESS 200M 5955 
> it works ok...
> 

That's a completely different chip that is supported properly with both UMS and KMS.

> As I said in the bug report, the chip works very good Only after power
> on,
> and the driver "ati" or "radeon" is not aware of KMS either...
> 

There is very limited UMS support for your chip, but it's not nearly as robust as the KMS support and there are no plans at the moment to do any more work on UMS.

> Seems to me that the screen memory is not align propper with
> the registers on the chipset, a propper align situation  occurs only
> after power on,
> or after about 10 resets.. 
> 
> A propper reset or initial program in the chip (using the ati, or radeon
> driver)
> would resolve the problem...
> 
> Can someone please point me to a documentation of the radeon 6250 that
> explains how to reset the chip to power on state???

You can take a look at the KMS code and port the relevant changes to UMS.
http://cgit.freedesktop.org/~airlied/linux/tree/drivers/gpu/drm/radeon?h=drm-fixes
There isn't just a simple function to reset the chip, you need to reprogram the memory controller, the display hardware, etc.
As your problem appears to be display related, I would take a look at the code in atombios_encoders.c and atombios_crtc.c in KMS and compare it to the code in atombios_output.c and atombios_crtc.c in the UMS tree.  Your chip is in the PALM family and has a DCE4.1 based display block.

> You can take a look at the KMS code and port the relevant changes to UMS.
> http://cgit.freedesktop.org/~airlied/linux/tree/drivers/gpu/drm/radeon?h=drm-fixes
> There isn't just a simple function to reset the chip, you need to reprogram the
> memory controller, the display hardware, etc.
> As your problem appears to be display related, I would take a look at the code
> in atombios_encoders.c and atombios_crtc.c in KMS and compare it to the code in
> atombios_output.c and atombios_crtc.c in the UMS tree.  Your chip is in the
> PALM family and has a DCE4.1 based display block.
> 

Thank you very much for the help..

Another work around is to reprogram gdm to not reset the X server
at the next login... 

As the Xserver works ok after power on, I need to use it only once..
of course I will not be able to suspend the notebook.

Sergio

*** Bug 46314 has been marked as a duplicate of this bug. ***

If I understand this correctly, there is no plan to support FreeBSD for the Radeon 6310 HD then?

(In reply to comment #6)
> If I understand this correctly, there is no plan to support FreeBSD for the Radeon 6310 HD then?

No more than is currently supported unless someone steps up to improve/maintain the UMS code for newer asics.