gatos-devel Mailing List for GATOS

Hello Rune,

Thanks for the update.
But it doesn't work...
As always... Nothing on TV... And black screen.
Killing X does not help.
<Alt><Ctrl><Del> boots.

http://alon.barlev.googlepages.com/xorg.conf
http://alon.barlev.googlepages.com/Xorg.0.log

I have:
01:00.0 0300: 1002:4e50 VGA compatible controller: ATI Technologies
Inc RV350 [Mobility Radeon 9600 M10]

I really hope you can figure it out!!!

Best Regards,
Alon Bar-Lev.

> On 5/28/06, Rune Petersen <ru...@me...> wrote:
> > Alon Bar-Lev wrote:
> > > On 5/25/06, Rune Petersen <ru...@me...> wrote:
> > >> tv_output does not work with R300+ based cards.
> > >>
> > >> I do have a possible fix if someone is willing to test it.
> > >
> > > I am!!!!
> >
> > since I don't know which version you're using you can make your own..
> >
> > in the file theater_out.c in function computeRestarts() add the
> > following just before "return hChanged;"
> >
> >      save->timing_cntl =
> >          (save->timing_cntl & 0x00ffffff) |
> >          ((0x72 * 640 / constPtr->horResolution)
> >          << 24);
> >
> >
> >
> > Good luck,
> > Rune Petersen
> >
>
>
>

Molle Bestefich wrote:
> Rune Petersen wrote:
>> >> Did it work with the old/stable patch?
>> >
>> > It did, except for missing sync on the TV.
>>
>> In other words TV-out didn't work either.
> 
> It was able to drive the output, so I'd say it worked considerably better.

not really...

> 
>> > My theory is that my particular card has to be driven at 50Hz, mainly
>> > based on the fact that I've seen the Windows driver drive the main
>> > display at 50Hz..  But let's not discuss that :-).
>>
>> You can always give it a try.
> 
> Modifying xorg.conf to drive everything at 50Hz enough, or should I
> patch something?

No patch needed..
But you are likely going to need to set the ForceMinDotClock option.


Rune Petersen

Rune Petersen wrote:
> >> Did it work with the old/stable patch?
> >
> > It did, except for missing sync on the TV.
>
> In other words TV-out didn't work either.

It was able to drive the output, so I'd say it worked considerably better.

> > My theory is that my particular card has to be driven at 50Hz, mainly
> > based on the fact that I've seen the Windows driver drive the main
> > display at 50Hz..  But let's not discuss that :-).
>
> You can always give it a try.

Modifying xorg.conf to drive everything at 50Hz enough, or should I
patch something?

> The old/"stable" patch does not contain any improvements for cards with
> RT100 for TV-out.
> Try the old/stable patch with this setup, and send me the log.

Ok, will do.

I can capture beautiful video with Avview on my ATI Radeon 7500 All-in-Wonder. 
But when I try to load these files into a Video Editor, there are problems. I 
have tried to edit using MainActor on Linux (Debian), and also using Ulead 
Video Studio 8 on a Windows 2000 system. The Ulead-Windows combination won't 
load the files at all. MainActor loads, but it takes minutes just to load a 
small file, and then each click operation takes more minutes. Perhaps  
theoretically this should be considered a problem with the Video Editors, but 
with those Vid Eds I can edit vid caps produced with the Ati-Tv software on 
Windows, which I don't particularly like. I think there must be some setup 
parameters I can use with Avview to get it to produce video that I can edit
with these tools. Anybody have any ideas?
- per...@sb...

Alon Bar-Lev wrote:
> On 5/25/06, Rune Petersen <ru...@me...> wrote:
>> tv_output does not work with R300+ based cards.
>>
>> I do have a possible fix if someone is willing to test it.
> 
> I am!!!!

since I don't know which version you're using you can make your own..

in the file theater_out.c in function computeRestarts() add the 
following just before "return hChanged;"

     save->timing_cntl =
         (save->timing_cntl & 0x00ffffff) |
         ((0x72 * 640 / constPtr->horResolution)
         << 24);



Good luck,
Rune Petersen

Molle Bestefich wrote:
> Rune Petersen wrote:
>> > By convention of the ever-so-awkward and cryptic conventions used in
>> > xorg.conf, the X option to force two heads on even when not detected
>> > seems to be "MergedFB" "on".
>>
>> Only single head works...
> 
> Ok, thought so.
> 
>> > So, so far it still seems that the new patch broke tvout with my
>> > particular hardware :-/.
>>
>> Did it work with the old/stable patch?
> 
> It did, except for missing sync on the TV.

In other words TV-out didn't work either.
> 
> My theory is that my particular card has to be driven at 50Hz, mainly
> based on the fact that I've seen the Windows driver drive the main
> display at 50Hz..  But let's not discuss that :-).

You can always give it a try.
> 
>> if not, go back to stable.
> 
> Aye?
> Stable patch or no patch at all?

The old/"stable" patch does not contain any improvements for cards with 
RT100 for TV-out.
Try the old/stable patch with this setup, and send me the log.


Rune Petersen

El Viernes, 26 de Mayo de 2006 00:00, Alberto Hernando escribi=F3:
> I know how to
> compile xorg 6.9, but not 7.0 (not yet). So if you can provide a patch
> against 6.9, I can test it soon and quickly.


Hi again.

I've just compiled x.org 7.1. This new version was just released. The ati=20
driver is version 6.6.0, and there are tvout patches only for 6.5.x. Well,=
=20
I'm going to wait. When the patch is updated against 6.6.0, if it includes=
=20
the fix for igp9100, I'll test it.

Thx again for your efforts, Rune.

	Alberto

Rune Petersen wrote:
> > By convention of the ever-so-awkward and cryptic conventions used in
> > xorg.conf, the X option to force two heads on even when not detected
> > seems to be "MergedFB" "on".
>
> Only single head works...

Ok, thought so.

> > So, so far it still seems that the new patch broke tvout with my
> > particular hardware :-/.
>
> Did it work with the old/stable patch?

It did, except for missing sync on the TV.

My theory is that my particular card has to be driven at 50Hz, mainly
based on the fact that I've seen the Windows driver drive the main
display at 50Hz..  But let's not discuss that :-).

> if not, go back to stable.

Aye?
Stable patch or no patch at all?

> > Mostly I was confused with the way it mentioned a single monitor
> > section, when people using Radeon and TV-out most often has two
> > monitors attached.  If the tvout hack is only intended to work with
> > the main monitor disabled and xorg.conf switched for regular / tvout
> > mode, then that should probably be mentioned.  Otherwise, you should
> > probably mention how to configure xorg.conf for using the Radeon
> > driver with two monitors and the tvout hack.  IMHO.
>
> Two monitor is not working unless if want clone-mode,
> which is set up just like single monitor mode.

Ok.  My point was that the above is not clear from the documentation
on your site.

> Please change your "device" section to look like this:
[snip]
> And send me the log.

Ok.

Molle Bestefich wrote:
...
> This particular graphics card often fails to find this particular TV
> using it's connection-detection hardware.  Back when I was using
> Windows, I could sometimes restart the machine and then it would
> detect the TV, but the safest method was always just checking the
> 'force TV output' option in the Catalyst driver, which I suspect was
> made for this exact purpose.

The good old days.

> 
> By convention of the ever-so-awkward and cryptic conventions used in
> xorg.conf, the X option to force two heads on even when not detected
> seems to be "MergedFB" "on".

Only single head works...

> 
> Adding "MergedFB" "on" to the minimal configuration, I get this in the log:

This just happens to kill TV-out.

> 
> Still no TV output, though.
> 
> (Probably "MergedFB" "on" is effectless anyway, since both the old and
> the new patch could only drive the TV when turning DVI output off, but
> it was worth a shot.)

> 
> So, so far it still seems that the new patch broke tvout with my
> particular hardware :-/.

Did it work with the old/stable patch? if not, go back to stable.

> Mostly I was confused with the way it mentioned a single monitor
> section, when people using Radeon and TV-out most often has two
> monitors attached.  If the tvout hack is only intended to work with
> the main monitor disabled and xorg.conf switched for regular / tvout
> mode, then that should probably be mentioned.  Otherwise, you should
> probably mention how to configure xorg.conf for using the Radeon
> driver with two monitors and the tvout hack.  IMHO.

Two monitor is not working unless if want clone-mode, which is set up 
just like single monitor mode.



Please change your "device" section to look like this:

Section "Device"
	Identifier  "Card0"
	Driver      "ati"
	VendorName  "ATI Technologies Inc"
	BoardName   "Radeon R200 QL [Radeon 8500 LE]"
	BusID       "PCI:1:0:0"
	Option      "RenderAccel" "on"
	Option      "MonitorLayout" "AUTO, NONE"
	Option      "TVOutput" "PAL"
         Option      "IgnoreEDID" "on"
EndSection

And send me the log.

Rune Petersen

Please send me your Xorg.conf. I'll answer this properly then I have 
some time...


Rune Petersen

Molle Bestefich wrote:
> Using my previous almost-working-except-TV-is-out-of-sync config:
>> Config0
>> |-->Screen0
>> |      |-->Card0
>> |      |-->Monitor0 (flatscreen)
>> |      |-->Monitor1 (tv)
>> |-->Mouse0
>> |-->Keyboard0
> 
> But applying the new patch, plus the changes you suggested:
>> Option       "IgnoreEDID" "true"
>> VertRefresh  60
>> # "MonitorLayout" ...
>> # Option "CRT2Position" ...
>> # Option "MetaModes" ...
> 
> The machine now consistently does a hard lock-up on 'startx'.
> There's still just blank (means no sync) output on the TV.
> 
> Pressing <NUMLOCK> doesn't switch on/off the numlock light, and
> <CTRL><ALT><DEL> and <CTRL><BACKSPACE> has no effect.  But it does
> seem that the X server completes it's startup, and that I can get it
> to change resolution with <CTRL><ALT>(<+>/<->) (judging by the "BIOS
> hotkeys disabled" at the end of Xorg.0.log).
> 
> ===== tail /var/log/Xorg.0.log =====
> (II) Mouse0: ps2EnableDataReporting: succeeded
> (II) RADEON(0): BIOS HotKeys Disabled
> (II) RADEON(0): BIOS HotKeys Disabled
> (II) RADEON(0): BIOS HotKeys Disabled
> (II) RADEON(0): BIOS HotKeys Disabled
> (II) RADEON(0): BIOS HotKeys Disabled
> (II) RADEON(0): BIOS HotKeys Disabled
> (II) RADEON(0): [drm] removed 1 reserved context for kernel
> (II) RADEON(0): [drm] unmapping 8192 bytes of SAREA 0xe1d77000 at 
> 0xb7933000
> FreeFontPath: FPE "/usr/share/fonts/misc/" refcount is 2, should be 1; 
> fixing.
> ====================================
> 
>> A golden rule:
>> I everything fails return to "hello world".
> 
> Okay, I've removed the flatscreen and minimized the config.
> Now X stopped locking up.
> 
> Still no output on S-VHS, though.  Only on the flatscreen.
> Switching resolutions just causes the flatscreen to switch resolution.
> 
> Maybe it has something to do with this line in Xorg.0.log:
> ====================================
> (WW) RADEON(0): Failed to detect secondary monitor, MergedFB/Clone mode 
> disabled
> ====================================
> 
> This particular graphics card often fails to find this particular TV
> using it's connection-detection hardware.  Back when I was using
> Windows, I could sometimes restart the machine and then it would
> detect the TV, but the safest method was always just checking the
> 'force TV output' option in the Catalyst driver, which I suspect was
> made for this exact purpose.
> 
> By convention of the ever-so-awkward and cryptic conventions used in
> xorg.conf, the X option to force two heads on even when not detected
> seems to be "MergedFB" "on".
> 
> Adding "MergedFB" "on" to the minimal configuration, I get this in the log:
> 
> ====================================
> (II) RADEON(0): Primary:
> Monitor   -- TMDS
> Connector -- DVI-I
> DAC Type  -- TVDAC/ExtDAC
> TMDS Type -- Internal
> DDC Type  -- DVI_DDC
> (II) RADEON(0): Secondary:
> Monitor   -- NONE
> Connector -- VGA
> DAC Type  -- Primary
> TMDS Type -- NONE
> DDC Type  -- VGA_DDC
> (II) RADEON(0): PLL parameters: rf=2700 rd=12 min=20000 max=35000; 
> xclk=27500
> (WW) RADEON(0): Failed to detect secondary monitor, MergedFB/Clone mode 
> disabled
> (II) RADEON(0): MergedFB mode forced on.
> (II) RADEON(0): Validating modes on Primary head ---------
> (II) RADEON(0): Television: Using hsync range of 30.00-40.00 kHz
> (II) RADEON(0): Television: Using vrefresh value of 60.00 Hz
> (II) RADEON(0): Validating CRTC2 modes for MergedFB ------------
> (II) RADEON(0): CRT2 Monitor: Using hsync range of 30.00-40.00 kHz
> (II) RADEON(0): CRT2 Monitor: Using vrefresh value of 60.00 Hz
> (II) RADEON(0): Total of 2 CRTC2 modes found for MergedFB------------
> (II) RADEON(0): Modes for CRT1: ********************
> (--) RADEON(0): Virtual size is 800x600 (pitch 896)
> (**) RADEON(0): *Default mode "800x600": 40.0 MHz, 37.9 kHz, 60.3 Hz
> (II) RADEON(0): Modeline "800x600"   40.00  800 840 968 1056  600 601
> 605 628 +hsync +vsync
> (**) RADEON(0): *Default mode "640x480": 25.2 MHz, 31.5 kHz, 60.0 Hz
> (II) RADEON(0): Modeline "640x480"   25.20  640 656 752 800  480 490
> 492 525 -hsync -vsync
> (II) RADEON(0): Modes for CRT2: ********************
> (--) RADEON(0): Virtual size is 800x600 (pitch 896)
> (**) RADEON(0): *Default mode "800x600": 40.0 MHz, 37.9 kHz, 60.3 Hz
> (II) RADEON(0): Modeline "800x600"   40.00  800 840 968 1056  600 601
> 605 628 +hsync +vsync
> (**) RADEON(0): *Default mode "640x480": 25.2 MHz, 31.5 kHz, 60.0 Hz
> (II) RADEON(0): Modeline "640x480"   25.20  640 656 752 800  480 490
> 492 525 -hsync -vsync
> (II) RADEON(0): Generating MergedFB mode list
> (II) RADEON(0): No MetaModes given, linking largest common modes by default
> (II) RADEON(0): Merged "800x600" (800x600) and "800x600" (800x600) to
> 800x600 (Clone)
> ====================================
> 
> Still no TV output, though.
> 
> (Probably "MergedFB" "on" is effectless anyway, since both the old and
> the new patch could only drive the TV when turning DVI output off, but
> it was worth a shot.)
> 
> So, so far it still seems that the new patch broke tvout with my
> particular hardware :-/.
> 
> Not sure what to try next?
> 
>> In your honor I have updated the page with a little more detail.
> 
> Hey, thanks!
> 
>> If you need more, tell me what you need.
> 
> Mostly I was confused with the way it mentioned a single monitor
> section, when people using Radeon and TV-out most often has two
> monitors attached.  If the tvout hack is only intended to work with
> the main monitor disabled and xorg.conf switched for regular / tvout
> mode, then that should probably be mentioned.  Otherwise, you should
> probably mention how to configure xorg.conf for using the Radeon
> driver with two monitors and the tvout hack.  IMHO.
> 

Using my previous almost-working-except-TV-is-out-of-sync config:
> Config0
> |-->Screen0
> |      |-->Card0
> |      |-->Monitor0 (flatscreen)
> |      |-->Monitor1 (tv)
> |-->Mouse0
> |-->Keyboard0

But applying the new patch, plus the changes you suggested:
> Option       "IgnoreEDID" "true"
> VertRefresh  60
> # "MonitorLayout" ...
> # Option "CRT2Position" ...
> # Option "MetaModes" ...

The machine now consistently does a hard lock-up on 'startx'.
There's still just blank (means no sync) output on the TV.

Pressing <NUMLOCK> doesn't switch on/off the numlock light, and
<CTRL><ALT><DEL> and <CTRL><BACKSPACE> has no effect.  But it does
seem that the X server completes it's startup, and that I can get it
to change resolution with <CTRL><ALT>(<+>/<->) (judging by the "BIOS
hotkeys disabled" at the end of Xorg.0.log).

=3D=3D=3D=3D=3D tail /var/log/Xorg.0.log =3D=3D=3D=3D=3D
(II) Mouse0: ps2EnableDataReporting: succeeded
(II) RADEON(0): BIOS HotKeys Disabled
(II) RADEON(0): BIOS HotKeys Disabled
(II) RADEON(0): BIOS HotKeys Disabled
(II) RADEON(0): BIOS HotKeys Disabled
(II) RADEON(0): BIOS HotKeys Disabled
(II) RADEON(0): BIOS HotKeys Disabled
(II) RADEON(0): [drm] removed 1 reserved context for kernel
(II) RADEON(0): [drm] unmapping 8192 bytes of SAREA 0xe1d77000 at 0xb793300=
0
FreeFontPath: FPE "/usr/share/fonts/misc/" refcount is 2, should be 1; fixi=
ng.
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D

> A golden rule:
> I everything fails return to "hello world".

Okay, I've removed the flatscreen and minimized the config.
Now X stopped locking up.

Still no output on S-VHS, though.  Only on the flatscreen.
Switching resolutions just causes the flatscreen to switch resolution.

Maybe it has something to do with this line in Xorg.0.log:
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
(WW) RADEON(0): Failed to detect secondary monitor, MergedFB/Clone mode dis=
abled
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D

This particular graphics card often fails to find this particular TV
using it's connection-detection hardware.  Back when I was using
Windows, I could sometimes restart the machine and then it would
detect the TV, but the safest method was always just checking the
'force TV output' option in the Catalyst driver, which I suspect was
made for this exact purpose.

By convention of the ever-so-awkward and cryptic conventions used in
xorg.conf, the X option to force two heads on even when not detected
seems to be "MergedFB" "on".

Adding "MergedFB" "on" to the minimal configuration, I get this in the log:

=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
(II) RADEON(0): Primary:
 Monitor   -- TMDS
 Connector -- DVI-I
 DAC Type  -- TVDAC/ExtDAC
 TMDS Type -- Internal
 DDC Type  -- DVI_DDC
(II) RADEON(0): Secondary:
 Monitor   -- NONE
 Connector -- VGA
 DAC Type  -- Primary
 TMDS Type -- NONE
 DDC Type  -- VGA_DDC
(II) RADEON(0): PLL parameters: rf=3D2700 rd=3D12 min=3D20000 max=3D35000; =
xclk=3D27500
(WW) RADEON(0): Failed to detect secondary monitor, MergedFB/Clone mode dis=
abled
(II) RADEON(0): MergedFB mode forced on.
(II) RADEON(0): Validating modes on Primary head ---------
(II) RADEON(0): Television: Using hsync range of 30.00-40.00 kHz
(II) RADEON(0): Television: Using vrefresh value of 60.00 Hz
(II) RADEON(0): Validating CRTC2 modes for MergedFB ------------
(II) RADEON(0): CRT2 Monitor: Using hsync range of 30.00-40.00 kHz
(II) RADEON(0): CRT2 Monitor: Using vrefresh value of 60.00 Hz
(II) RADEON(0): Total of 2 CRTC2 modes found for MergedFB------------
(II) RADEON(0): Modes for CRT1: ********************
(--) RADEON(0): Virtual size is 800x600 (pitch 896)
(**) RADEON(0): *Default mode "800x600": 40.0 MHz, 37.9 kHz, 60.3 Hz
(II) RADEON(0): Modeline "800x600"   40.00  800 840 968 1056  600 601
605 628 +hsync +vsync
(**) RADEON(0): *Default mode "640x480": 25.2 MHz, 31.5 kHz, 60.0 Hz
(II) RADEON(0): Modeline "640x480"   25.20  640 656 752 800  480 490
492 525 -hsync -vsync
(II) RADEON(0): Modes for CRT2: ********************
(--) RADEON(0): Virtual size is 800x600 (pitch 896)
(**) RADEON(0): *Default mode "800x600": 40.0 MHz, 37.9 kHz, 60.3 Hz
(II) RADEON(0): Modeline "800x600"   40.00  800 840 968 1056  600 601
605 628 +hsync +vsync
(**) RADEON(0): *Default mode "640x480": 25.2 MHz, 31.5 kHz, 60.0 Hz
(II) RADEON(0): Modeline "640x480"   25.20  640 656 752 800  480 490
492 525 -hsync -vsync
(II) RADEON(0): Generating MergedFB mode list
(II) RADEON(0): No MetaModes given, linking largest common modes by default
(II) RADEON(0): Merged "800x600" (800x600) and "800x600" (800x600) to
800x600 (Clone)
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D

Still no TV output, though.

(Probably "MergedFB" "on" is effectless anyway, since both the old and
the new patch could only drive the TV when turning DVI output off, but
it was worth a shot.)

So, so far it still seems that the new patch broke tvout with my
particular hardware :-/.

Not sure what to try next?

> In your honor I have updated the page with a little more detail.

Hey, thanks!

> If you need more, tell me what you need.

Mostly I was confused with the way it mentioned a single monitor
section, when people using Radeon and TV-out most often has two
monitors attached.  If the tvout hack is only intended to work with
the main monitor disabled and xorg.conf switched for regular / tvout
mode, then that should probably be mentioned.  Otherwise, you should
probably mention how to configure xorg.conf for using the Radeon
driver with two monitors and the tvout hack.  IMHO.

T24gNS8yNS8wNiwgUnVuZSBQZXRlcnNlbiA8cnVuZUBtZWdhaHVydHMuZGs+IHdyb3RlOgo+IHR2
X291dHB1dCBkb2VzIG5vdCB3b3JrIHdpdGggUjMwMCsgYmFzZWQgY2FyZHMuCj4KPiBJIGRvIGhh
dmUgYSBwb3NzaWJsZSBmaXggaWYgc29tZW9uZSBpcyB3aWxsaW5nIHRvIHRlc3QgaXQuCgpJIGFt
ISEhIQo=

El Jueves, 25 de Mayo de 2006 22:43, Rune Petersen escribi=F3:
> No they are not, 9100 is R200 and 9100igp is RS300.
>
> A good resource to find that family your card is:
> http://www.rojakpot.com/showarticle.aspx?artno=3D88&pgno=3D0
>
> tv_output does not work with R300+ based cards.
>
> I do have a possible fix if someone is willing to test it.


I am!

I can test whatever you think can help. Just bear in mind that I know how t=
o=20
compile xorg 6.9, but not 7.0 (not yet). So if you can provide a patch=20
against 6.9, I can test it soon and quickly.

	Alberto

Alberto Hernando wrote:
> El Jueves, 25 de Mayo de 2006 21:08, Rune Petersen escribi=F3:
>>>>> (II) RADEON(0): Detected Rage Theatre revision 00000003
>>>> One of the few that have TV-out via RT100
>>> Hey, great :-).
>>>
>>> I'll report back with my results.
>=20
>=20
> Hi.
>=20
> I still haven't tried this, because I need to learn how to compile xorg=
 7=20
> first. But I'm a bit lost after the discussion. I think that there shou=
ld be=20
> some changes, and in some configurations this should work. Right? Also,=
 I see=20
> in the web that radeon 9100 works. Is radeon igp9100 (my chipset) the s=
ame?=20

No they are not, 9100 is R200 and 9100igp is RS300.

A good resource to find that family your card is:
http://www.rojakpot.com/showarticle.aspx?artno=3D88&pgno=3D0

tv_output does not work with R300+ based cards.

I do have a possible fix if someone is willing to test it.


Rune Petersen

El Jueves, 25 de Mayo de 2006 21:08, Rune Petersen escribi=F3:
> >> > (II) RADEON(0): Detected Rage Theatre revision 00000003
> >>
> >> One of the few that have TV-out via RT100
> >
> > Hey, great :-).
> >
> > I'll report back with my results.


Hi.

I still haven't tried this, because I need to learn how to compile xorg 7=20
first. But I'm a bit lost after the discussion. I think that there should b=
e=20
some changes, and in some configurations this should work. Right? Also, I s=
ee=20
in the web that radeon 9100 works. Is radeon igp9100 (my chipset) the same?=
=20
I'll search in my logs (still xorg 6.9) what version of Rage Theatre I have=
,=20
if any. If someone thinks that I can have a good chance to make it work,=20
perhaps I'll try sooner than later.

	Alberto

Rune Petersen wrote:

> I hope you now know its a hack =)

A hack that works is much better than clean code that doesn't do
what you need it to do!  ;-)

Gary

Molle Bestefich wrote:
> Rune Petersen wrote:
>> > (**) RADEON(0): Option "MonitorLayout" "TMDS,TV"
>> > (**) RADEON(0): Option "CRT2Position" "Clone"
>> > (**) RADEON(0): Option "MetaModes"
>> "1280x1024,800x600;1280x1024,640x480;800x600,800x600;1280x1024;800x600;640x480" 
>>
>>
>> Did you want to fail?
> 
> No, I've tried about a million different configurations, and this one
> happened to work well, so that's why it has stuck.

A golden rule:
I everything fails return to "hello world".

> 
>> Anyone of these options are capable of breaking TV-out (maybe not 
>> MetaModes)
> 
> Pretty lame driver, if half the TV-out options you can give it will
> automagically break TV-out completely.

I know, but there are more important problems than this (and its a hack).

> 
>> start simple:
>> http://megahurts.dk/rune/tv_output.html
> 
> The instructions on your page are extremely vague, but I'll try.

In your honor I have updated the page with a little more detail.
If you need more, tell me what you need.
> 
>> > (II) RADEON(0): Valid Mode from established timing table: 640x480
>> EDID modes from LCD monitors are useless for TV-out disable it.
> 
> I'm using a flatscreen too, it's very useful for that.

I know...

> I'll disable EDID (and risk breaking the flatscreen if I configure
> things wrong), but this should really be fixed in the driver so that
> disabling EDID is not necessary.

Its a hack, If you can get me the specs from ATI I'll look in to it.

> 
>> Change this in your Xorg.conf
>>
>> "Device" section:
>>     Option     "IgnoreEDID" "true"
>>
>> "Monitor" section:
>>          HorizSync    30.0 - 40.0
>>          VertRefresh  60
> 
> Not sure which monitor you mean, I currently have:
> 
> Config0
> |-->Screen0
> |      |-->Card0
> |      |-->Monitor0 (flatscreen)
> |      |-->Monitor1 (tv)
> |-->Mouse0
> |-->Keyboard0
> 
> I'm not sure X nor the Radeon driver picks this up correctly, but I
> believe that's how the documentation I read at the time wanted me to
> do.

The docs are useless because this is a hack.... please delete Monitor1.

> 
>> > (II) RADEON(0): Detected Rage Theatre revision 00000003
>>
>> One of the few that have TV-out via RT100
> 
> Hey, great :-).
> 
> I'll report back with my results.

good luck,


Rune Petersen


P.S
I hope you now know its a hack =)

Rune Petersen wrote:
> > (**) RADEON(0): Option "MonitorLayout" "TMDS,TV"
> > (**) RADEON(0): Option "CRT2Position" "Clone"
> > (**) RADEON(0): Option "MetaModes"
> "1280x1024,800x600;1280x1024,640x480;800x600,800x600;1280x1024;800x600;64=
0x480"
>
> Did you want to fail?

No, I've tried about a million different configurations, and this one
happened to work well, so that's why it has stuck.

> Anyone of these options are capable of breaking TV-out (maybe not MetaMod=
es)

Pretty lame driver, if half the TV-out options you can give it will
automagically break TV-out completely.

> start simple:
> http://megahurts.dk/rune/tv_output.html

The instructions on your page are extremely vague, but I'll try.

> > (II) RADEON(0): Valid Mode from established timing table: 640x480
> EDID modes from LCD monitors are useless for TV-out disable it.

I'm using a flatscreen too, it's very useful for that.
I'll disable EDID (and risk breaking the flatscreen if I configure
things wrong), but this should really be fixed in the driver so that
disabling EDID is not necessary.

> Change this in your Xorg.conf
>
> "Device" section:
> =09Option     "IgnoreEDID" "true"
>
> "Monitor" section:
>          HorizSync    30.0 - 40.0
>          VertRefresh  60

Not sure which monitor you mean, I currently have:

Config0
|-->Screen0
|      |-->Card0
|      |-->Monitor0 (flatscreen)
|      |-->Monitor1 (tv)
|-->Mouse0
|-->Keyboard0

I'm not sure X nor the Radeon driver picks this up correctly, but I
believe that's how the documentation I read at the time wanted me to
do.

> > (II) RADEON(0): Detected Rage Theatre revision 00000003
>
> One of the few that have TV-out via RT100

Hey, great :-).

I'll report back with my results.

Molle Bestefich wrote:
> Rune Petersen wrote:
>> Molle Bestefich wrote:
>> > New patch:
>> > ========
>> > All resolutions generate output on DVI-D.
>> > No output is ever produced on the TV.
>> >
>> > Does this mean anything to you?
>> > Would you like to see xorg.conf?
>>
>> I would like to see the log....
>> You should still loose output on DVI.
> 
> Attached.
> Off-list since gatos-devel usually drops attachments anyway, but feel
> free to keep the conversation on list :-).


> (**) RADEON(0): Option "MonitorLayout" "TMDS,TV"
> (**) RADEON(0): Option "CRT2Position" "Clone"
> (**) RADEON(0): Option "MetaModes" "1280x1024,800x600;1280x1024,640x480;800x600,800x600;1280x1024;800x600;640x480"

Did you want to fail?
Anyone of these options are capable of breaking TV-out (maybe not MetaModes)

start simple:
http://megahurts.dk/rune/tv_output.html

> (II) RADEON(0): Panel infos found from DDC detailed: 1280x1024
....
> (II) RADEON(0): Valid Mode from established timing table: 640x480
EDID modes from LCD monitors are useless for TV-out disable it.

Change this in your Xorg.conf

"Device" section:
	Option     "IgnoreEDID" "true"

"Monitor" section:
         HorizSync    30.0 - 40.0
         VertRefresh  60

> (II) Loading sub module "theater_out"
> (II) LoadModule: "theater_out"
> (II) Loading /usr/lib/xorg/modules/multimedia/theater_out_drv.so
> (II) Module theater_out: vendor="X.Org Foundation"
> 	compiled for 7.0.0, module version = 1.0.0
> 	ABI class: X.Org Video Driver, version 0.8
> (II) RADEON(0): No response from device 0 on VIP bus
> (II) RADEON(0): Device 1 on VIP bus ids as 0x4d541002
> (II) RADEON(0): No response from device 2 on VIP bus
> (II) RADEON(0): No response from device 3 on VIP bus
> (II) RADEON(0): Detected Rage Theatre as device 1 on VIP bus
> (II) RADEON(0): Detected Rage Theatre revision 00000003

One of the few that have TV-out via RT100


Rune Petersen

My system is Debian unstable. I am trying to compile the driver for use the
tv_out of my ati radeon 9000.
I tried it with debian testing wich has Xorg 6.9 and it works fine, bat now
in debian unstable wich uses Xorg 7.0 dont work.

When I execute the configure command ends showme the following:

checking whether to include TV Out support... yes
checking for NONE/share/X11/sgml/defs.ent... no
checking for linuxdoc... no
checking for ps2pdf... /usr/bin/ps2pdf
checking Whether to build documentation... no
checking Whether to build pdf documentation... yes
configure: creating ./config.status
config.status: creating Makefile
config.status: creating src/Makefile
config.status: creating man/Makefile
config.status: creating config.h
config.status: executing depfiles commands


 ... and If I execute make ....

make  all-recursive
make[1]: se ingresa al directorio `/home/fran/xserver-xorg-video-ati-6.5.8.=
0
'
Making all in src
make[2]: se ingresa al directorio `/home/fran/xserver-xorg-video-ati-6.5.8.=
0
/src'
if /bin/sh ../libtool --mode=3Dcompile gcc -DHAVE_CONFIG_H -I. -I. -I..
-DXFree86Server -DIN_MODULE -DXFree86Module -DXFree86LOADER
-I/usr/include/xorg   -I/usr/include/drm -I/usr/include/X11/dri   -g -O2 -M=
T
ati.lo -MD -MP -MF ".deps/ati.Tpo" \
          -c -o ati.lo `test -f 'ati.c' || echo './'`ati.c; \
        then mv -f ".deps/ati.Tpo" ".deps/ati.Plo"; \
        else rm -f ".deps/ati.Tpo"; exit 1; \
        fi
mkdir .libs
 gcc -DHAVE_CONFIG_H -I. -I. -I.. -DXFree86Server -DIN_MODULE
-DXFree86Module -DXFree86LOADER -I/usr/include/xorg -I/usr/include/drm
-I/usr/include/X11/dri -g -O2 -MT ati.lo -MD -MP -MF .deps/ati.Tpo -c ati.c
-fPIC -DPIC -o .libs/ati.o
if /bin/sh ../libtool --mode=3Dcompile gcc -DHAVE_CONFIG_H -I. -I. -I..
-DXFree86Server -DIN_MODULE -DXFree86Module -DXFree86LOADER
-I/usr/include/xorg   -I/usr/include/drm -I/usr/include/X11/dri   -g -O2 -M=
T
atiadapter.lo -MD -MP -MF ".deps/atiadapter.Tpo" \
          -c -o atiadapter.lo `test -f 'atiadapter.c' || echo
'./'`atiadapter.c; \
        then mv -f ".deps/atiadapter.Tpo" ".deps/atiadapter.Plo"; \
        else rm -f ".deps/atiadapter.Tpo"; exit 1; \
        fi
 gcc -DHAVE_CONFIG_H -I. -I. -I.. -DXFree86Server -DIN_MODULE
-DXFree86Module -DXFree86LOADER -I/usr/include/xorg -I/usr/include/drm
-I/usr/include/X11/dri -g -O2 -MT atiadapter.lo -MD -MP -MF
.deps/atiadapter.Tpo -c atiadapter.c  -fPIC -DPIC -o .libs/atiadapter.o
if /bin/sh ../libtool --mode=3Dcompile gcc -DHAVE_CONFIG_H -I. -I. -I..
-DXFree86Server -DIN_MODULE -DXFree86Module -DXFree86LOADER
-I/usr/include/xorg   -I/usr/include/drm -I/usr/include/X11/dri   -g -O2 -M=
T
atibus.lo -MD -MP -MF ".deps/atibus.Tpo" \
          -c -o atibus.lo `test -f 'atibus.c' || echo './'`atibus.c; \
        then mv -f ".deps/atibus.Tpo" ".deps/atibus.Plo"; \
        else rm -f ".deps/atibus.Tpo"; exit 1; \
        fi
 gcc -DHAVE_CONFIG_H -I. -I. -I.. -DXFree86Server -DIN_MODULE
-DXFree86Module -DXFree86LOADER -I/usr/include/xorg -I/usr/include/drm
-I/usr/include/X11/dri -g -O2 -MT atibus.lo -MD -MP -MF .deps/atibus.Tpo -c
atibus.c  -fPIC -DPIC -o .libs/atibus.o
In file included from atistruct.h:41,
                 from atibus.c:33:
atidripriv.h:40:23: error: GL/glxint.h: No such file or directory
atidripriv.h:41:26: error: GL/glxtokens.h: No such file or directory
In file included from atistruct.h:41,
                 from atibus.c:33:
atidripriv.h:55: error: syntax error before '__GLXvisualConfig'
In file included from atibus.c:33:
atistruct.h:481: error: syntax error before '__GLXvisualConfig'
atistruct.h:481: warning: no semicolon at end of struct or union
atistruct.h:508: error: syntax error before '}' token
atistruct.h:508: warning: data definition has no type or storage class
atibus.c: In function 'ATIClaimResources':
atibus.c:76: error: dereferencing pointer to incomplete type
atibus.c:76: error: dereferencing pointer to incomplete type
atibus.c:85: error: dereferencing pointer to incomplete type
atibus.c:86: error: dereferencing pointer to incomplete type
atibus.c:87: error: dereferencing pointer to incomplete type
atibus.c:88: error: dereferencing pointer to incomplete type
atibus.c:90: error: dereferencing pointer to incomplete type
atibus.c:92: error: dereferencing pointer to incomplete type
atibus.c:96: error: dereferencing pointer to incomplete type
atibus.c:97: error: dereferencing pointer to incomplete type
atibus.c:102: error: dereferencing pointer to incomplete type
atibus.c:104: error: dereferencing pointer to incomplete type
atibus.c:109: error: dereferencing pointer to incomplete type
atibus.c:113: error: dereferencing pointer to incomplete type
atibus.c:115: error: dereferencing pointer to incomplete type
atibus.c:116: error: dereferencing pointer to incomplete type
atibus.c:119: error: dereferencing pointer to incomplete type
atibus.c:120: error: dereferencing pointer to incomplete type
atibus.c:122: error: dereferencing pointer to incomplete type
atibus.c:126: error: dereferencing pointer to incomplete type
atibus.c:129: error: dereferencing pointer to incomplete type
atibus.c:145: error: dereferencing pointer to incomplete type
atibus.c:149: error: dereferencing pointer to incomplete type
atibus.c: In function 'ATIClaimBusSlot':
atibus.c:174: error: dereferencing pointer to incomplete type
atibus.c:177: error: dereferencing pointer to incomplete type
atibus.c:181: error: dereferencing pointer to incomplete type
atibus.c:183: error: dereferencing pointer to incomplete type
atibus.c:186: error: dereferencing pointer to incomplete type
make[2]: *** [atibus.lo] Error 1
make[2]: se sale del directorio `/home/fran/xserver-xorg-video-ati-6.5.8.0
/src'
make[1]: *** [all-recursive] Error 1
make[1]: se sale del directorio `/home/fran/xserver-xorg-video-ati-6.5.8.0'
make: *** [all] Error 2


what I am doing wrong?

Thanks

Hey folks,

thanks to some help I've managed to get avview to get
no sound, and no video. As opposed to errors left and
right.

I'm still in need of assistance, or some direction as
to where to go next.

I'm using:
   SuSE 9.3
   Xorg 6.9
   The microcode file was found and initialized

If there's anything else I can add to this (log files,
and such) let me know and I'll be more than happy to
furnish the information.

Hope to get this going soon.

Thanks.

~Mr. Anderson

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 

Molle Bestefich wrote:
> Rune Petersen wrote:
>> In theater_out.c replace TV_6DAC_CNTL_NHOLD with TV_DAC_CNTL_NHOLD.
>> I've updated the patch...
> 
> Using the change you mention above, didn't re-download the patch.
> 
> Old patch:
> ========
> All resolutions generate output on DVI-D, except 800x600.
> Using 800x600, DVI-D output is disabled and TV output is enabled.
> TV shows nothing but flicker, though.
> 
> My guess:
> Because the TV is driven at 60 instead of 50 Hz.
nope...
> 
> The flat screen monitor attached to DVI-D probably reports that only
> 60Hz input is accepted over DVI - it says in the manual that the
> recommended frequency is 60Hz.
> I think the X ATI driver incorrectly always drives the DAC at 60Hz,
> even when it's using the same DAC to drive TV output, which should be
> at 50Hz.
I can find many reasons to run the DAC at 60Hz (It doesn't drive the TV 
output AFAIK)
ATI's own driver does this.
> 
> Also, it would be very nice if the X ATI driver could somehow use the
> other DAC on the card to drive the TV output, so the flatscreen
> doesn't go blank.  I've had this work before, but it might have been
> with VGA and TV, not DVI-D and TV - not entirely sure now that I think
> about it.
Yes it would... but there are more pressing matters...
> 
> New patch:
> ========
> All resolutions generate output on DVI-D.
> No output is ever produced on the TV.
> 
> Does this mean anything to you?
> Would you like to see xorg.conf?

I would like to see the log....
You should still loose output on DVI.

Rune Petersen wrote:
> In theater_out.c replace TV_6DAC_CNTL_NHOLD with TV_DAC_CNTL_NHOLD.
> I've updated the patch...

Using the change you mention above, didn't re-download the patch.

Old patch:
=3D=3D=3D=3D=3D=3D=3D=3D
All resolutions generate output on DVI-D, except 800x600.
Using 800x600, DVI-D output is disabled and TV output is enabled.
TV shows nothing but flicker, though.

My guess:
Because the TV is driven at 60 instead of 50 Hz.

The flat screen monitor attached to DVI-D probably reports that only
60Hz input is accepted over DVI - it says in the manual that the
recommended frequency is 60Hz.
I think the X ATI driver incorrectly always drives the DAC at 60Hz,
even when it's using the same DAC to drive TV output, which should be
at 50Hz.

Also, it would be very nice if the X ATI driver could somehow use the
other DAC on the card to drive the TV output, so the flatscreen
doesn't go blank.  I've had this work before, but it might have been
with VGA and TV, not DVI-D and TV - not entirely sure now that I think
about it.

New patch:
=3D=3D=3D=3D=3D=3D=3D=3D
All resolutions generate output on DVI-D.
No output is ever produced on the TV.

Does this mean anything to you?
Would you like to see xorg.conf?

Hi,
 I have a graphics card " ALL IN WONDER / RAGE 128 / 32MB " and I would l=
ike to
know if the version of xorg 7.0 contains gatos natively.
 If not, how to make to use gatos and xorg 7.0?
( it's for make a GENTOO PVR )
 Thanks for your answer.
 Best regards.

Spencer wrote:
> Is the version of avview from Fedora core?
No, this is a fresh build from the source.