radeon: Fix system hang issue when using KMS with older cards

commit e49f3959a96dc279860af7e86e6dbcfda50580a5 upstream.

The current radeon driver initialization routines, when using KMS, are written
so that the IRQ installation routine is called before initializing the WB buffer
and the CP rings. With some ASICs, though, the IRQ routine tries to access the
GFX_INDEX ring causing a call to RREG32 with the value of -1 in
radeon_fence_read. This, in turn causes the system to completely hang with some
cards, requiring a hard reset.

A call stack that can cause such a hang looks like this (using rv515 ASIC for the
example here):
 * rv515_init (rv515.c)
 * radeon_irq_kms_init (radeon_irq_kms.c)
 * drm_irq_install (drm_irq.c)
 * radeon_driver_irq_preinstall_kms (radeon_irq_kms.c)
 * rs600_irq_process (rs600.c)
 * radeon_fence_process - due to SW interrupt (radeon_fence.c)
 * radeon_fence_read (radeon_fence.c)
 * hang due to RREG32(-1)

The patch moves the IRQ installation to the card startup routine, after the ring
has been initialized, but before the IRQ has been set. This fixes the issue, but
requires a check to see if the IRQ is already installed, as is the case in the
system resume codepath.
I have tested the patch on three machines using the rv515, the rv770 and the
evergreen ASIC. They worked without issues.

This seems to be a known issue and has been reported on several bug tracking
sites by various distributions (see links below). Most of reports recommend
booting the system with KMS disabled and then enabling KMS by reloading the
radeon module. For some reason, this was indeed a usable workaround, however,
UMS is now deprecated and disabled by default.

Bug reports:
https://bugzilla.redhat.com/show_bug.cgi?id=845745
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/561789
https://bbs.archlinux.org/viewtopic.php?id=156964

Signed-off-by: Adis Hamzić <adis@hamzadis.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

13 files changed, 78 insertions, 45 deletions

diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c
index 6a8776eae1ae..300099d31187 100644
--- a/drivers/gpu/drm/radeon/evergreen.c
+++ b/drivers/gpu/drm/radeon/evergreen.c
@@ -3258,6 +3258,12 @@ static int evergreen_startup(struct radeon_device *rdev)
 	}
 
 	/* Enable IRQ */
+	if (!rdev->irq.installed) {
+		r = radeon_irq_kms_init(rdev);
+		if (r)
+			return r;
+	}
+
 	r = r600_irq_init(rdev);
 	if (r) {
 		DRM_ERROR("radeon: IH init failed (%d).\n", r);
@@ -3409,10 +3415,6 @@ int evergreen_init(struct radeon_device *rdev)
 	if (r)
 		return r;
 
-	r = radeon_irq_kms_init(rdev);
-	if (r)
-		return r;
-
 	rdev->ring[RADEON_RING_TYPE_GFX_INDEX].ring_obj = NULL;
 	r600_ring_init(rdev, &rdev->ring[RADEON_RING_TYPE_GFX_INDEX], 1024 * 1024);
 
diff --git a/drivers/gpu/drm/radeon/ni.c b/drivers/gpu/drm/radeon/ni.c
index d706da8986b4..f5387b3e699f 100644
--- a/drivers/gpu/drm/radeon/ni.c
+++ b/drivers/gpu/drm/radeon/ni.c
@@ -1614,6 +1614,12 @@ static int cayman_startup(struct radeon_device *rdev)
 	}
 
 	/* Enable IRQ */
+	if (!rdev->irq.installed) {
+		r = radeon_irq_kms_init(rdev);
+		if (r)
+			return r;
+	}
+
 	r = r600_irq_init(rdev);
 	if (r) {
 		DRM_ERROR("radeon: IH init failed (%d).\n", r);
@@ -1744,10 +1750,6 @@ int cayman_init(struct radeon_device *rdev)
 	if (r)
 		return r;
 
-	r = radeon_irq_kms_init(rdev);
-	if (r)
-		return r;
-
 	ring->ring_obj = NULL;
 	r600_ring_init(rdev, ring, 1024 * 1024);
 
diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c
index fe33d35dae8c..40ed0e5b95a3 100644
--- a/drivers/gpu/drm/radeon/r100.c
+++ b/drivers/gpu/drm/radeon/r100.c
@@ -3952,6 +3952,12 @@ static int r100_startup(struct radeon_device *rdev)
 	}
 
 	/* Enable IRQ */
+	if (!rdev->irq.installed) {
+		r = radeon_irq_kms_init(rdev);
+		if (r)
+			return r;
+	}
+
 	r100_irq_set(rdev);
 	rdev->config.r100.hdp_cntl = RREG32(RADEON_HOST_PATH_CNTL);
 	/* 1M ring buffer */
@@ -4113,9 +4119,6 @@ int r100_init(struct radeon_device *rdev)
 	r = radeon_fence_driver_init(rdev);
 	if (r)
 		return r;
-	r = radeon_irq_kms_init(rdev);
-	if (r)
-		return r;
 	/* Memory manager */
 	r = radeon_bo_init(rdev);
 	if (r)
diff --git a/drivers/gpu/drm/radeon/r300.c b/drivers/gpu/drm/radeon/r300.c
index fa14383f9ca0..63b53aa9536f 100644
--- a/drivers/gpu/drm/radeon/r300.c
+++ b/drivers/gpu/drm/radeon/r300.c
@@ -1405,6 +1405,12 @@ static int r300_startup(struct radeon_device *rdev)
 	}
 
 	/* Enable IRQ */
+	if (!rdev->irq.installed) {
+		r = radeon_irq_kms_init(rdev);
+		if (r)
+			return r;
+	}
+
 	r100_irq_set(rdev);
 	rdev->config.r300.hdp_cntl = RREG32(RADEON_HOST_PATH_CNTL);
 	/* 1M ring buffer */
@@ -1545,9 +1551,6 @@ int r300_init(struct radeon_device *rdev)
 	r = radeon_fence_driver_init(rdev);
 	if (r)
 		return r;
-	r = radeon_irq_kms_init(rdev);
-	if (r)
-		return r;
 	/* Memory manager */
 	r = radeon_bo_init(rdev);
 	if (r)
diff --git a/drivers/gpu/drm/radeon/r420.c b/drivers/gpu/drm/radeon/r420.c
index f3fcaacfea01..1a9347fdd15e 100644
--- a/drivers/gpu/drm/radeon/r420.c
+++ b/drivers/gpu/drm/radeon/r420.c
@@ -265,6 +265,12 @@ static int r420_startup(struct radeon_device *rdev)
 	}
 
 	/* Enable IRQ */
+	if (!rdev->irq.installed) {
+		r = radeon_irq_kms_init(rdev);
+		if (r)
+			return r;
+	}
+
 	r100_irq_set(rdev);
 	rdev->config.r300.hdp_cntl = RREG32(RADEON_HOST_PATH_CNTL);
 	/* 1M ring buffer */
@@ -417,10 +423,6 @@ int r420_init(struct radeon_device *rdev)
 	if (r) {
 		return r;
 	}
-	r = radeon_irq_kms_init(rdev);
-	if (r) {
-		return r;
-	}
 	/* Memory manager */
 	r = radeon_bo_init(rdev);
 	if (r) {
diff --git a/drivers/gpu/drm/radeon/r520.c b/drivers/gpu/drm/radeon/r520.c
index ebcc15b03c9f..57be78461ab5 100644
--- a/drivers/gpu/drm/radeon/r520.c
+++ b/drivers/gpu/drm/radeon/r520.c
@@ -194,6 +194,12 @@ static int r520_startup(struct radeon_device *rdev)
 	}
 
 	/* Enable IRQ */
+	if (!rdev->irq.installed) {
+		r = radeon_irq_kms_init(rdev);
+		if (r)
+			return r;
+	}
+
 	rs600_irq_set(rdev);
 	rdev->config.r300.hdp_cntl = RREG32(RADEON_HOST_PATH_CNTL);
 	/* 1M ring buffer */
@@ -301,9 +307,6 @@ int r520_init(struct radeon_device *rdev)
 	r = radeon_fence_driver_init(rdev);
 	if (r)
 		return r;
-	r = radeon_irq_kms_init(rdev);
-	if (r)
-		return r;
 	/* Memory manager */
 	r = radeon_bo_init(rdev);
 	if (r)
diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c
index b1ff9ccd2d62..8c403d946acc 100644
--- a/drivers/gpu/drm/radeon/r600.c
+++ b/drivers/gpu/drm/radeon/r600.c
@@ -2470,6 +2470,12 @@ int r600_startup(struct radeon_device *rdev)
 	}
 
 	/* Enable IRQ */
+	if (!rdev->irq.installed) {
+		r = radeon_irq_kms_init(rdev);
+		if (r)
+			return r;
+	}
+
 	r = r600_irq_init(rdev);
 	if (r) {
 		DRM_ERROR("radeon: IH init failed (%d).\n", r);
@@ -2624,10 +2630,6 @@ int r600_init(struct radeon_device *rdev)
 	if (r)
 		return r;
 
-	r = radeon_irq_kms_init(rdev);
-	if (r)
-		return r;
-
 	rdev->ring[RADEON_RING_TYPE_GFX_INDEX].ring_obj = NULL;
 	r600_ring_init(rdev, &rdev->ring[RADEON_RING_TYPE_GFX_INDEX], 1024 * 1024);
 
diff --git a/drivers/gpu/drm/radeon/rs400.c b/drivers/gpu/drm/radeon/rs400.c
index 4cf381b3a6d8..5099bd32e195 100644
--- a/drivers/gpu/drm/radeon/rs400.c
+++ b/drivers/gpu/drm/radeon/rs400.c
@@ -417,6 +417,12 @@ static int rs400_startup(struct radeon_device *rdev)
 	}
 
 	/* Enable IRQ */
+	if (!rdev->irq.installed) {
+		r = radeon_irq_kms_init(rdev);
+		if (r)
+			return r;
+	}
+
 	r100_irq_set(rdev);
 	rdev->config.r300.hdp_cntl = RREG32(RADEON_HOST_PATH_CNTL);
 	/* 1M ring buffer */
@@ -541,9 +547,6 @@ int rs400_init(struct radeon_device *rdev)
 	r = radeon_fence_driver_init(rdev);
 	if (r)
 		return r;
-	r = radeon_irq_kms_init(rdev);
-	if (r)
-		return r;
 	/* Memory manager */
 	r = radeon_bo_init(rdev);
 	if (r)
diff --git a/drivers/gpu/drm/radeon/rs600.c b/drivers/gpu/drm/radeon/rs600.c
index d25cf869d08d..52480010156d 100644
--- a/drivers/gpu/drm/radeon/rs600.c
+++ b/drivers/gpu/drm/radeon/rs600.c
@@ -864,6 +864,12 @@ static int rs600_startup(struct radeon_device *rdev)
 	}
 
 	/* Enable IRQ */
+	if (!rdev->irq.installed) {
+		r = radeon_irq_kms_init(rdev);
+		if (r)
+			return r;
+	}
+
 	rs600_irq_set(rdev);
 	rdev->config.r300.hdp_cntl = RREG32(RADEON_HOST_PATH_CNTL);
 	/* 1M ring buffer */
@@ -994,9 +1000,6 @@ int rs600_init(struct radeon_device *rdev)
 	r = radeon_fence_driver_init(rdev);
 	if (r)
 		return r;
-	r = radeon_irq_kms_init(rdev);
-	if (r)
-		return r;
 	/* Memory manager */
 	r = radeon_bo_init(rdev);
 	if (r)
diff --git a/drivers/gpu/drm/radeon/rs690.c b/drivers/gpu/drm/radeon/rs690.c
index f2c3b9d75f18..c46900c2c1aa 100644
--- a/drivers/gpu/drm/radeon/rs690.c
+++ b/drivers/gpu/drm/radeon/rs690.c
@@ -628,6 +628,12 @@ static int rs690_startup(struct radeon_device *rdev)
 	}
 
 	/* Enable IRQ */
+	if (!rdev->irq.installed) {
+		r = radeon_irq_kms_init(rdev);
+		if (r)
+			return r;
+	}
+
 	rs600_irq_set(rdev);
 	rdev->config.r300.hdp_cntl = RREG32(RADEON_HOST_PATH_CNTL);
 	/* 1M ring buffer */
@@ -759,9 +765,6 @@ int rs690_init(struct radeon_device *rdev)
 	r = radeon_fence_driver_init(rdev);
 	if (r)
 		return r;
-	r = radeon_irq_kms_init(rdev);
-	if (r)
-		return r;
 	/* Memory manager */
 	r = radeon_bo_init(rdev);
 	if (r)
diff --git a/drivers/gpu/drm/radeon/rv515.c b/drivers/gpu/drm/radeon/rv515.c
index 43af363ead37..0532bbee9f95 100644
--- a/drivers/gpu/drm/radeon/rv515.c
+++ b/drivers/gpu/drm/radeon/rv515.c
@@ -386,6 +386,12 @@ static int rv515_startup(struct radeon_device *rdev)
 	}
 
 	/* Enable IRQ */
+	if (!rdev->irq.installed) {
+		r = radeon_irq_kms_init(rdev);
+		if (r)
+			return r;
+	}
+
 	rs600_irq_set(rdev);
 	rdev->config.r300.hdp_cntl = RREG32(RADEON_HOST_PATH_CNTL);
 	/* 1M ring buffer */
@@ -520,9 +526,6 @@ int rv515_init(struct radeon_device *rdev)
 	r = radeon_fence_driver_init(rdev);
 	if (r)
 		return r;
-	r = radeon_irq_kms_init(rdev);
-	if (r)
-		return r;
 	/* Memory manager */
 	r = radeon_bo_init(rdev);
 	if (r)
diff --git a/drivers/gpu/drm/radeon/rv770.c b/drivers/gpu/drm/radeon/rv770.c
index 591040b94660..4a3937fa2dc7 100644
--- a/drivers/gpu/drm/radeon/rv770.c
+++ b/drivers/gpu/drm/radeon/rv770.c
@@ -1099,6 +1099,12 @@ static int rv770_startup(struct radeon_device *rdev)
 	}
 
 	/* Enable IRQ */
+	if (!rdev->irq.installed) {
+		r = radeon_irq_kms_init(rdev);
+		if (r)
+			return r;
+	}
+
 	r = r600_irq_init(rdev);
 	if (r) {
 		DRM_ERROR("radeon: IH init failed (%d).\n", r);
@@ -1237,10 +1243,6 @@ int rv770_init(struct radeon_device *rdev)
 	if (r)
 		return r;
 
-	r = radeon_irq_kms_init(rdev);
-	if (r)
-		return r;
-
 	rdev->ring[RADEON_RING_TYPE_GFX_INDEX].ring_obj = NULL;
 	r600_ring_init(rdev, &rdev->ring[RADEON_RING_TYPE_GFX_INDEX], 1024 * 1024);
 
diff --git a/drivers/gpu/drm/radeon/si.c b/drivers/gpu/drm/radeon/si.c
index 2dbd585c81f1..e22b460a4f0c 100644
--- a/drivers/gpu/drm/radeon/si.c
+++ b/drivers/gpu/drm/radeon/si.c
@@ -3876,6 +3876,12 @@ static int si_startup(struct radeon_device *rdev)
 	}
 
 	/* Enable IRQ */
+	if (!rdev->irq.installed) {
+		r = radeon_irq_kms_init(rdev);
+		if (r)
+			return r;
+	}
+
 	r = si_irq_init(rdev);
 	if (r) {
 		DRM_ERROR("radeon: IH init failed (%d).\n", r);
@@ -4044,10 +4050,6 @@ int si_init(struct radeon_device *rdev)
 	if (r)
 		return r;
 
-	r = radeon_irq_kms_init(rdev);
-	if (r)
-		return r;
-
 	ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 	ring->ring_obj = NULL;
 	r600_ring_init(rdev, ring, 1024 * 1024);