2.6.36-rc2 regression: Wrong screen resolution (1024x768 instead of 1680x1050)

2.6.36-rc2 regression: Wrong screen resolution (1024x768 instead of 1680x1050)

[Tino Keitel]

Hi,

booting 2.6.36-rc2-git4 (commit
d4348c678977c7093438bbbf2067c49396ae941b) results in a screen
resolution of 1024x768 instead of 1680x1050.  It works fine with
2.6.35.2.

The framebuffer console uses the correct resolution, but only uses a
part of the screen for text output. Xorg output looks distorted.

Xorg log is attached.

$ xrandr 
Screen 0: minimum 320 x 200, current 1024 x 768, maximum 4096 x 4096
VGA1 disconnected (normal left inverted right x axis y axis)
DVI1 connected 1024x768+0+0 (normal left inverted right x axis y axis)
454mm x 284mm
   1680x1050      59.9 +
   1280x1024      60.0  
   1280x960       60.0  
   1024x768       60.0* 
   800x600        60.3  
   640x480        60.0  
   720x400        70.1  
TV1 connected 1024x768+0+0 (normal left inverted right x axis y axis)
0mm x 0mm
   848x480        30.0 +
   640x480        30.0 +
   1024x768       30.0* 
   800x600        30.0  

This is a Mac mini Core 2 Duo with Intel i945 graphics.

Regards,
Tino

Re: 2.6.36-rc2 regression: Wrong screen resolution (1024x768 instead of 1680x1050)

[Tino Keitel]

[-- Attachment #1: Type: text/plain, Size: 138 bytes --]

On Thu, Aug 26, 2010 at 09:30:39 +0200, Tino Keitel wrote:

[...]

> Xorg log is attached.

It wasn't. Not it is attached.

Regards,
Tino

[-- Attachment #2: Xorg.0.log.gz --]
[-- Type: application/octet-stream, Size: 5468 bytes --]

Re: 2.6.36-rc2 regression: Wrong screen resolution (1024x768 instead of 1680x1050)

[Tino Keitel]

[-- Attachment #1: Type: text/plain, Size: 138 bytes --]

On Thu, Aug 26, 2010 at 09:30:39 +0200, Tino Keitel wrote:

[...]

> Xorg log is attached.

It wasn't. Not it is attached.

Regards,
Tino

[-- Attachment #2: Xorg.0.log.gz --]
[-- Type: application/octet-stream, Size: 5468 bytes --]

[-- Attachment #3: Type: text/plain, Size: 159 bytes --]

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: 2.6.36-rc2 regression: Wrong screen resolution (1024x768 instead of 1680x1050)

[Tino Keitel]

On Thu, Aug 26, 2010 at 09:30:39 +0200, Tino Keitel wrote:
> Hi,
> 
> booting 2.6.36-rc2-git4 (commit
> d4348c678977c7093438bbbf2067c49396ae941b) results in a screen
> resolution of 1024x768 instead of 1680x1050.  It works fine with
> 2.6.35.2.
> 
> The framebuffer console uses the correct resolution, but only uses a
> part of the screen for text output. Xorg output looks distorted.
> 
> Xorg log is attached.
> 
> $ xrandr 
> Screen 0: minimum 320 x 200, current 1024 x 768, maximum 4096 x 4096
> VGA1 disconnected (normal left inverted right x axis y axis)
> DVI1 connected 1024x768+0+0 (normal left inverted right x axis y axis)
> 454mm x 284mm
>    1680x1050      59.9 +
>    1280x1024      60.0  
>    1280x960       60.0  
>    1024x768       60.0* 
>    800x600        60.3  
>    640x480        60.0  
>    720x400        70.1  
> TV1 connected 1024x768+0+0 (normal left inverted right x axis y axis)
> 0mm x 0mm
>    848x480        30.0 +
>    640x480        30.0 +
>    1024x768       30.0* 
>    800x600        30.0  

There is no TV connected, and it was correctly reported as
"disconnected" with 2.6.35.

The Mac mini has some special EDID handling, as the EDID data comes
from the analog EDID channel IIRC, if that helps.

Regards,
Tino

Re: 2.6.36-rc2 regression: Wrong screen resolution (1024x768 instead of 1680x1050)

[Tino Keitel]

On Thu, Aug 26, 2010 at 09:30:39 +0200, Tino Keitel wrote:
> Hi,
> 
> booting 2.6.36-rc2-git4 (commit
> d4348c678977c7093438bbbf2067c49396ae941b) results in a screen
> resolution of 1024x768 instead of 1680x1050.  It works fine with
> 2.6.35.2.
> 
> The framebuffer console uses the correct resolution, but only uses a
> part of the screen for text output. Xorg output looks distorted.
> 
> Xorg log is attached.
> 
> $ xrandr 
> Screen 0: minimum 320 x 200, current 1024 x 768, maximum 4096 x 4096
> VGA1 disconnected (normal left inverted right x axis y axis)
> DVI1 connected 1024x768+0+0 (normal left inverted right x axis y axis)
> 454mm x 284mm
>    1680x1050      59.9 +
>    1280x1024      60.0  
>    1280x960       60.0  
>    1024x768       60.0* 
>    800x600        60.3  
>    640x480        60.0  
>    720x400        70.1  
> TV1 connected 1024x768+0+0 (normal left inverted right x axis y axis)
> 0mm x 0mm
>    848x480        30.0 +
>    640x480        30.0 +
>    1024x768       30.0* 
>    800x600        30.0  

There is no TV connected, and it was correctly reported as
"disconnected" with 2.6.35.

The Mac mini has some special EDID handling, as the EDID data comes
from the analog EDID channel IIRC, if that helps.

Regards,
Tino

Re: 2.6.36-rc2 regression: Wrong screen resolution (1024x768 instead of 1680x1050)

[Maciej Rutecki]

On czwartek, 26 sierpnia 2010 o 09:30:39 Tino Keitel wrote:
> Hi,
> 
> booting 2.6.36-rc2-git4 (commit
> d4348c678977c7093438bbbf2067c49396ae941b) results in a screen
> resolution of 1024x768 instead of 1680x1050.  It works fine with
> 2.6.35.2.
> 
> The framebuffer console uses the correct resolution, but only uses a
> part of the screen for text output. Xorg output looks distorted.
> 
> Xorg log is attached.
> 
> $ xrandr
> Screen 0: minimum 320 x 200, current 1024 x 768, maximum 4096 x 4096
> VGA1 disconnected (normal left inverted right x axis y axis)
> DVI1 connected 1024x768+0+0 (normal left inverted right x axis y axis)
> 454mm x 284mm
>    1680x1050      59.9 +
>    1280x1024      60.0
>    1280x960       60.0
>    1024x768       60.0*
>    800x600        60.3
>    640x480        60.0
>    720x400        70.1
> TV1 connected 1024x768+0+0 (normal left inverted right x axis y axis)
> 0mm x 0mm
>    848x480        30.0 +
>    640x480        30.0 +
>    1024x768       30.0*
>    800x600        30.0
> 
> This is a Mac mini Core 2 Duo with Intel i945 graphics.
> 

Seems be similar to:
https://bugzilla.kernel.org/show_bug.cgi?id=17301

Regards
-- 
Maciej Rutecki
http://www.maciek.unixy.pl

Re: 2.6.36-rc2 regression: Wrong screen resolution (1024x768 instead of 1680x1050)

[Tino Keitel]

[-- Attachment #1: Type: text/plain, Size: 645 bytes --]

On Sun, Aug 29, 2010 at 15:19:20 +0200, Maciej Rutecki wrote:
> > TV1 connected 1024x768+0+0 (normal left inverted right x axis y axis)
> > 0mm x 0mm
> >    848x480        30.0 +
> >    640x480        30.0 +
> >    1024x768       30.0*
> >    800x600        30.0
> > 
> > This is a Mac mini Core 2 Duo with Intel i945 graphics.
> > 
> 
> Seems be similar to:
> https://bugzilla.kernel.org/show_bug.cgi?id=17301

Maybe. I don't know if this is connected, but after disabling TV out
using the attached patch, the screen has no signal after suspend to
RAM. However, the patch helped to get back the native LCD resolution
after boot.

Regards,
Tino

[-- Attachment #2: disable_tv_out.diff --]
[-- Type: text/x-diff, Size: 1185 bytes --]

diff --git a/drivers/gpu/drm/i915/intel_overlay.c b/drivers/gpu/drm/i915/intel_overlay.c
index 4f00390..1d306a4 100644
--- a/drivers/gpu/drm/i915/intel_overlay.c
+++ b/drivers/gpu/drm/i915/intel_overlay.c
@@ -25,6 +25,8 @@
  *
  * Derived from Xorg ddx, xf86-video-intel, src/i830_video.c
  */
+
+#include <linux/seq_file.h>
 #include "drmP.h"
 #include "drm.h"
 #include "i915_drm.h"
diff --git a/drivers/gpu/drm/i915/intel_tv.c b/drivers/gpu/drm/i915/intel_tv.c
index d2029ef..f71e488 100644
--- a/drivers/gpu/drm/i915/intel_tv.c
+++ b/drivers/gpu/drm/i915/intel_tv.c
@@ -1279,6 +1279,8 @@ intel_tv_detect_type (struct intel_tv *intel_tv)
 	 *  1 0 X svideo
 	 *  0 0 0 Component
 	 */
+	type = -1;
+	goto out;
 	if ((tv_dac & TVDAC_SENSE_MASK) == (TVDAC_B_SENSE | TVDAC_C_SENSE)) {
 		DRM_DEBUG_KMS("Detected Composite TV connection\n");
 		type = DRM_MODE_CONNECTOR_Composite;
@@ -1292,7 +1294,7 @@ intel_tv_detect_type (struct intel_tv *intel_tv)
 		DRM_DEBUG_KMS("No TV connection detected\n");
 		type = -1;
 	}
-
+out:
 	/* Restore interrupt config */
 	spin_lock_irqsave(&dev_priv->user_irq_lock, irqflags);
 	i915_enable_pipestat(dev_priv, 0, PIPE_HOTPLUG_INTERRUPT_ENABLE |

Re: 2.6.36-rc2 regression: Wrong screen resolution (1024x768 instead of 1680x1050)

[Tino Keitel]

[-- Attachment #1: Type: text/plain, Size: 645 bytes --]

On Sun, Aug 29, 2010 at 15:19:20 +0200, Maciej Rutecki wrote:
> > TV1 connected 1024x768+0+0 (normal left inverted right x axis y axis)
> > 0mm x 0mm
> >    848x480        30.0 +
> >    640x480        30.0 +
> >    1024x768       30.0*
> >    800x600        30.0
> > 
> > This is a Mac mini Core 2 Duo with Intel i945 graphics.
> > 
> 
> Seems be similar to:
> https://bugzilla.kernel.org/show_bug.cgi?id=17301

Maybe. I don't know if this is connected, but after disabling TV out
using the attached patch, the screen has no signal after suspend to
RAM. However, the patch helped to get back the native LCD resolution
after boot.

Regards,
Tino

[-- Attachment #2: disable_tv_out.diff --]
[-- Type: text/x-diff, Size: 1185 bytes --]

diff --git a/drivers/gpu/drm/i915/intel_overlay.c b/drivers/gpu/drm/i915/intel_overlay.c
index 4f00390..1d306a4 100644
--- a/drivers/gpu/drm/i915/intel_overlay.c
+++ b/drivers/gpu/drm/i915/intel_overlay.c
@@ -25,6 +25,8 @@
  *
  * Derived from Xorg ddx, xf86-video-intel, src/i830_video.c
  */
+
+#include <linux/seq_file.h>
 #include "drmP.h"
 #include "drm.h"
 #include "i915_drm.h"
diff --git a/drivers/gpu/drm/i915/intel_tv.c b/drivers/gpu/drm/i915/intel_tv.c
index d2029ef..f71e488 100644
--- a/drivers/gpu/drm/i915/intel_tv.c
+++ b/drivers/gpu/drm/i915/intel_tv.c
@@ -1279,6 +1279,8 @@ intel_tv_detect_type (struct intel_tv *intel_tv)
 	 *  1 0 X svideo
 	 *  0 0 0 Component
 	 */
+	type = -1;
+	goto out;
 	if ((tv_dac & TVDAC_SENSE_MASK) == (TVDAC_B_SENSE | TVDAC_C_SENSE)) {
 		DRM_DEBUG_KMS("Detected Composite TV connection\n");
 		type = DRM_MODE_CONNECTOR_Composite;
@@ -1292,7 +1294,7 @@ intel_tv_detect_type (struct intel_tv *intel_tv)
 		DRM_DEBUG_KMS("No TV connection detected\n");
 		type = -1;
 	}
-
+out:
 	/* Restore interrupt config */
 	spin_lock_irqsave(&dev_priv->user_irq_lock, irqflags);
 	i915_enable_pipestat(dev_priv, 0, PIPE_HOTPLUG_INTERRUPT_ENABLE |

[-- Attachment #3: Type: text/plain, Size: 159 bytes --]

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx