[radeonhd] 0x94CC:0x1545:0x3650: <Radeonhd RV610 HD2400Pro>

[Benjamin Herrenschmidt]

On Wed, 2008-08-13 at 03:52 -0700, Tai Tri Nguyen wrote:
> Hi all,
> 
> Thanks for your helps. That seems an interesting issue. 
> 
> Ben,
>  I'm working on a powerpc sequoia using AMCC 440EPX CPU.

Ok, I don't think I have one of these, but I have an SPe and a 405EX so
that should be similar. In any case, it will take a little bit of time
before I can get to these r6xx endian issues so I suggest you look at
what you get with an r5xx so we can properly separate the issues caused
by the lack of surface swapping from whatever other pixman issues we may
have.

Cheers,
Ben.

> Best regards,
> 
> Tai Nguyen
> 
> -----Original Message-----
> From: Benjamin Herrenschmidt [mailto:benh at kernel.crashing.org] 
> Sent: Wednesday, August 13, 2008 5:38 PM
> To: Dave Airlie
> Cc: Tai Tri Nguyen; Alex Deucher; xorg-driver-ati at lists.x.org; Tuan
> Thanh Phan
> Subject: Re: [radeonhd] 0x94CC:0x1545:0x3650: <Radeonhd RV610 HD2400Pro>
> 
> On Wed, 2008-08-13 at 20:07 +1000, Dave Airlie wrote:
> > On Wed, Aug 13, 2008 at 7:54 PM, Tai Tri Nguyen <ttnguyen at amcc.com>
> wrote:
> > > Hi Alex,
> > >
> > > The driver I'm using is from git, the latest version of 
> > > xf86-video-ati  >>git clone 
> > > git://anongit.freedesktop.org/git/xorg/driver/xf86-video-ati
> > >
> > > I have just downloaded and compiled again but the issue still exist.
> > > Video can display in 24 bits depth color but perform wrong color.
> > > Do you have any idea for the issue?
> > >
> > 
> > I think this is a pixman bug, but I'm not sure,
> > 
> > Ben you have any luck tracking it down further?
> > 
> > Actually R600 doesn't have surfaces like previous cards so actually we
> 
> > need to program some bits of r600.
> 
> Hi !
> 
> There are two different things at play here:
> 
>  - There may be a problem with pixman, I've had report of color issues
> with some apps / themes and I've observed some myself with the fedora
> default theme, on cards like r500 where we pretty much know the surface
> swapping works.
> 
>  - The r600 has no surfaces, at least not the way they used to work, so
> it's expected that colors will be wrong on that one.
> 
> For the later, I don't know yet what can be done, ie. I don't know what
> they replaced surfaces with and how we can deal with endianness on it,
> I'll have to catch up with you and Alex, look at available docs, etc...
> 
> At this stage, it's hard to say which of these two problems is affecting
> Tai Tri. I would suggest he tries an R500 card such as X1650 or similar,
> this is what we use on Bimini and appears to work reasonably well, at
> least we have working surface swappers. I'll have to track down the
> pixman issue (if that's one, btw, I could use your help here Dave in
> isolating what X rendering ops are actually going wrong).
> 
> For R6xx, I can't tell yet what we'll do. I have a 2400HD too so I
> should hopefully be able to work on that as soon as I find some spare
> time.
> 
> Tai: What machine is this ? I can try to do a similar setup, I have a
> whole range of AMCC eval boards here, I was planning to give the X1650 a
> go some time next week or so in fact...
> 
> Cheers,
> Ben.
> --------------------------------------------------------
> 
> CONFIDENTIALITY NOTICE: This e-mail message, including any attachments, is for the sole use of the intended recipient(s) and contains information that is confidential and proprietary to Applied Micro Circuits Corporation or its subsidiaries. It is to be used solely for the purpose of furthering the parties' business relationship. All unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message.