Skip to contentSkip to conversation

drm/i915/gen4: Fix interrupt setup ordering

PATCH In review Adam Jackson 2 participants 3 messages May 27, 2010 – Jun 4, 2010

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 drm/i915/gen4: Fix interrupt setup ordering

Unmask, then enable interrupts, then enable interrupt sources; matches PCH ordering. The old way (sources, enable, unmask) gives a window during which interrupt conditions would appear in ISR but would never reach IIR and thus never raise an IRQ. Since interrupts only trigger on rising edges in ISR, this would lead to conditions where (for example) output hotplugging would never fire an interrupt because it was already stuck on in ISR.

Also, since we know IIR and PIPExSTAT have been cleared during irq_preinstall, don't clear them again during irq_postinstall, nothing good can come of that.

---
 drivers/gpu/drm/i915/i915_irq.c |   50 ++++++++++++++++++---------------------
 1 files changed, 23 insertions(+), 27 deletions(-)
drivers/gpu/drm/i915/i915_irq.c +23 -27
index a7e4b1f..f9d2cc0 100644
@@ -1402,29 +1402,10 @@ int i915_driver_irq_postinstall(struct drm_device *dev)
 	dev_priv->pipestat[1] = 0;
 
 	if (I915_HAS_HOTPLUG(dev)) {
-		u32 hotplug_en = I915_READ(PORT_HOTPLUG_EN);
-
-		/* Note HDMI and DP share bits */
-		if (dev_priv->hotplug_supported_mask & HDMIB_HOTPLUG_INT_STATUS)
-			hotplug_en |= HDMIB_HOTPLUG_INT_EN;
-		if (dev_priv->hotplug_supported_mask & HDMIC_HOTPLUG_INT_STATUS)
-			hotplug_en |= HDMIC_HOTPLUG_INT_EN;
-		if (dev_priv->hotplug_supported_mask & HDMID_HOTPLUG_INT_STATUS)
-			hotplug_en |= HDMID_HOTPLUG_INT_EN;
-		if (dev_priv->hotplug_supported_mask & SDVOC_HOTPLUG_INT_STATUS)
-			hotplug_en |= SDVOC_HOTPLUG_INT_EN;
-		if (dev_priv->hotplug_supported_mask & SDVOB_HOTPLUG_INT_STATUS)
-			hotplug_en |= SDVOB_HOTPLUG_INT_EN;
-		if (dev_priv->hotplug_supported_mask & CRT_HOTPLUG_INT_STATUS)
-			hotplug_en |= CRT_HOTPLUG_INT_EN;
-		/* Ignore TV since it's buggy */
-
-		I915_WRITE(PORT_HOTPLUG_EN, hotplug_en);
-
 		/* Enable in IER... */
 		enable_mask |= I915_DISPLAY_PORT_INTERRUPT;
 		/* and unmask in IMR */
-		i915_enable_irq(dev_priv, I915_DISPLAY_PORT_INTERRUPT);
+		dev_priv->irq_mask_reg &= ~I915_DISPLAY_PORT_INTERRUPT;
 	}
 
 	/*
@@ -1442,16 +1423,31 @@ int i915_driver_irq_postinstall(struct drm_device *dev)
 	}
 	I915_WRITE(EMR, error_mask);
 
-	/* Disable pipe interrupt enables, clear pending pipe status */
-	I915_WRITE(PIPEASTAT, I915_READ(PIPEASTAT) & 0x8000ffff);
-	I915_WRITE(PIPEBSTAT, I915_READ(PIPEBSTAT) & 0x8000ffff);
-	/* Clear pending interrupt status */
-	I915_WRITE(IIR, I915_READ(IIR));
-
-	I915_WRITE(IER, enable_mask);
 	I915_WRITE(IMR, dev_priv->irq_mask_reg);
+	I915_WRITE(IER, enable_mask);
 	(void) I915_READ(IER);
 
+	if (I915_HAS_HOTPLUG(dev)) {
+		u32 hotplug_en = I915_READ(PORT_HOTPLUG_EN);
+
+		/* Note HDMI and DP share bits */
+		if (dev_priv->hotplug_supported_mask & HDMIB_HOTPLUG_INT_STATUS)
+			hotplug_en |= HDMIB_HOTPLUG_INT_EN;
+		if (dev_priv->hotplug_supported_mask & HDMIC_HOTPLUG_INT_STATUS)
+			hotplug_en |= HDMIC_HOTPLUG_INT_EN;
+		if (dev_priv->hotplug_supported_mask & HDMID_HOTPLUG_INT_STATUS)
+			hotplug_en |= HDMID_HOTPLUG_INT_EN;
+		if (dev_priv->hotplug_supported_mask & SDVOC_HOTPLUG_INT_STATUS)
+			hotplug_en |= SDVOC_HOTPLUG_INT_EN;
+		if (dev_priv->hotplug_supported_mask & SDVOB_HOTPLUG_INT_STATUS)
+			hotplug_en |= SDVOB_HOTPLUG_INT_EN;
+		if (dev_priv->hotplug_supported_mask & CRT_HOTPLUG_INT_STATUS)
+			hotplug_en |= CRT_HOTPLUG_INT_EN;
+		/* Ignore TV since it's buggy */
+
+		I915_WRITE(PORT_HOTPLUG_EN, hotplug_en);
+	}
+
 	opregion_enable_asle(dev);
 
 	return 0;
Signature · 1 line
1.7.0.1
#1 lore raw
Quoted 8 lines · Adam Jackson

Unmask, then enable interrupts, then enable interrupt sources; matches PCH ordering. The old way (sources, enable, unmask) gives a window during which interrupt conditions would appear in ISR but would never reach IIR and thus never raise an IRQ. Since interrupts only trigger on rising edges in ISR, this would lead to conditions where (for example) output hotplugging would never fire an interrupt because it was already stuck on in ISR.

Quoted 5 lines

Also, since we know IIR and PIPExSTAT have been cleared during irq_preinstall, don't clear them again during irq_postinstall, nothing good can come of that.

Just to follow up, I've tested this (combined with the already-merged VGA paranoia patch) on GM45. Output hotplug still works and I'm no longer seeing stuck bits in ISR.

- ajax

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

#2 lore raw
Quoted 13 lines · Adam Jackson
Quoted 12 lines · Adam Jackson

Unmask, then enable interrupts, then enable interrupt sources; matches PCH ordering. The old way (sources, enable, unmask) gives a window during which interrupt conditions would appear in ISR but would never reach IIR and thus never raise an IRQ. Since interrupts only trigger on rising edges in ISR, this would lead to conditions where (for example) output hotplugging would never fire an interrupt because it was already stuck on in ISR.

Also, since we know IIR and PIPExSTAT have been cleared during irq_preinstall, don't clear them again during irq_postinstall, nothing good can come of that.

Quoted 5 lines
Quoted 1 line

Just to follow up, I've tested this (combined with the already-merged VGA paranoia patch) on GM45. Output hotplug still works and I'm no longer seeing stuck bits in ISR.

Thanks for the followup. It really helps me to know what people have tested on, and whether they were fixing a bug due to code inspection or due to actually solving a problem they were hitting on hardware.

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx