Bug 44898

another short bug report, for this problem is here :
https://bugs.launchpad.net/lpbugreporter/+bug/752542/comments/466

Created attachment 55781 [details] [review]
Litttle patch by keithp (on irc) for one first test

Comment on attachment 55781 [details] [review]
Litttle patch by keithp (on irc) for one first test

This test, have not solved the problèrem, keithp tell me to test the second little patch

Created attachment 55782 [details] [review]
second little patch from keithp

No more result with this one...

Does this still happen in drm-intel-next?  An eDP fix from Daniel seems to have fixed a lot of our "blank screen" issues.

I try this kernel version : http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.4-rc3-precise/  with no more result.

Today i try to test :
git clone git://git.kernel.org/pub/scm/linux/kernel/git/keithp/linux.git
cd linux
git checkout -t origin/drm-intel-fixes
complil' 

and boot whith this one...

Pleas use this the git tree at http://cgit.freedesktop.org/~danvet/drm-intel/ and try the drm-intel-next-queued branch.

Created attachment 60214 [details]
new log

log with rm.debug=0x06 from this kernel  git://people.freedesktop.org/~danvet/drm-intel 
branch drm-intel-next-queued

a new fresh install whith ubuntu precise 12.04 on the same machine

uname -m : i686
uname -r : 3.4.0-rc3+

with this kernel :

git clone git://people.freedesktop.org/~danvet/drm-intel
cd drm-intel
git checkout -t origin/drm-intel-next-queued
cp /boot/config-`uname -r` .config && sudo make oldconfig && sudo make menuconfig
make-kpkg --initrd --revision=002 kernel_image kernel_headers modules_image
cd ..
dpkg -i *
reboot

i try to boot whith nomodeset, and drm.debug=0x06  but i have a black screen...

xserver-xorg-core                         2:1.11.4-0ubuntu6
linux-image-3.4.0-rc3+                    002
libgl1-mesa-dri                           8.0.1-0ubuntu5
libgl1-mesa-glx                           8.0.1-0ubuntu5
libglu1-mesa                              8.0.1-0ubuntu5
libdrm-intel1                             2.4.30-1ubuntu1

i put le Xorg.0.log-NEW in file flag "new log"

thanks for help...

Created attachment 60223 [details]
dmesg-17-04-2012

dmesg whith kernel  command line: BOOT_IMAGE=/boot/vmlinuz-3.4.0-rc3+ root=UUID=cc1afab4-2d6d-46a6-8556-1b5c2f1cfaa3 ro quiet splash drm.debug=0x06 vt.handoff=7

Created attachment 60224 [details]
intel_reg_dumper_17-04-2012

Going out on a limb and associating these two bugs.

Can one of you get the output from /sys/kernel/debug/dri/0/i915_gem_interrupt?  You may need to capture it a few times to get the output I'm looking for, specifically what the "South Display Interrupt identity" field is when the hotplug occurs.  I'm guessing bit 23, 22, or 21 is stuck on, but it could be one of the others.

After you've gathered that, you could try this patch, which ought to shut up your hotplug interrupts at least:

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 967b92e..9b6d8ac 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -2106,6 +2106,8 @@ static int ironlake_irq_postinstall(struct drm_device *dev
                                SDE_AUX_MASK);
        }
 
+       hotplug_mask = 0;
+
        dev_priv->pch_irq_mask = ~hotplug_mask;
 
        I915_WRITE(SDEIIR, I915_READ(SDEIIR));

*** This bug has been marked as a duplicate of bug 42278 ***

Created attachment 60285 [details]
dmesg-19-04-2012

After this patch

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 967b92e..9b6d8ac 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -2106,6 +2106,8 @@ static int ironlake_irq_postinstall(struct drm_device
*dev
                                SDE_AUX_MASK);
        }

+       hotplug_mask = 0;
+
        dev_priv->pch_irq_mask = ~hotplug_mask;

        I915_WRITE(SDEIIR, I915_READ(SDEIIR));

Created attachment 60286 [details]
Xorg.log-19-04-2012

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 967b92e..9b6d8ac 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -2106,6 +2106,8 @@ static int ironlake_irq_postinstall(struct drm_device
*dev
                                SDE_AUX_MASK);
        }

+       hotplug_mask = 0;
+
        dev_priv->pch_irq_mask = ~hotplug_mask;

        I915_WRITE(SDEIIR, I915_READ(SDEIIR));

Created attachment 60287 [details]
intel_reg_dumper-19-04-2012

after this patch

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 967b92e..9b6d8ac 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -2106,6 +2106,8 @@ static int ironlake_irq_postinstall(struct drm_device
*dev
                                SDE_AUX_MASK);
        }

+       hotplug_mask = 0;
+
        dev_priv->pch_irq_mask = ~hotplug_mask;

        I915_WRITE(SDEIIR, I915_READ(SDEIIR));

Whith this patch its much better :

- i can see the desktop in 1920*1080 (whith no option : not nomodeset in kernel boot)

but...

the scren turn black every 1 or two second..., and some times "in an all black mode" and do not retrun

On IRC, ickle_ and jbarnes  tell me to put :
drm_kms_helper.poll=0 in the kernel boot line.

After boot with that option, its better, the screen seams to be OK


but...

the screen flash in "in an all black mode" sometime :
- on some startup i do not see the desktop at all
- on some startup i see the desktop, can use it, but some minutes after, it turn in "all black mode"... 
- sometimes i can use it with no problem...

It work like Bug 42278 :  without patch on an ubuntu kernel : 3.3.0-030300-generic-pae

But its the same behavior than before :
sometime black screnn randomly

I try to "stress" by playing video HD, but its pretty mellow , if you have a better method...

Since output detection seems to be the cause of the black screen, I'd expect you to see it whenever some app polls the output configuration.  The media player or other desktop apps may do that periodically (every GNOME app used to do it at app startup time).