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 0ea6bdf5fd7c8d..75312603aa1f73 100644
--- a/drivers/gpu/drm/radeon/evergreen.c
+++ b/drivers/gpu/drm/radeon/evergreen.c
@@ -3728,6 +3728,12 @@ 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);
@@ -3876,10 +3882,6 @@ 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 0a32d89e627add..d7968b8c92e199 100644
--- a/drivers/gpu/drm/radeon/ni.c
+++ b/drivers/gpu/drm/radeon/ni.c
@@ -1711,6 +1711,12 @@ 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);
@@ -1857,10 +1863,6 @@ 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 4973bff37fec8a..d0314ecbd7c18e 100644
--- a/drivers/gpu/drm/radeon/r100.c
+++ b/drivers/gpu/drm/radeon/r100.c
@@ -3869,6 +3869,12 @@ 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 */
@@ -4024,9 +4030,6 @@ 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 c60350e6872dde..b9b776f1e5822c 100644
--- a/drivers/gpu/drm/radeon/r300.c
+++ b/drivers/gpu/drm/radeon/r300.c
@@ -1382,6 +1382,12 @@ 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 */
@@ -1516,9 +1522,6 @@ 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 6fce2eb4dd16b3..4e796ecf9ea477 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 */
@@ -411,10 +417,6 @@ 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 f795a4e092cb72..e1aece73b370c1 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 */
@@ -297,9 +303,6 @@ 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 0740db3fcd2268..aa3066a173a7ce 100644
--- a/drivers/gpu/drm/radeon/r600.c
+++ b/drivers/gpu/drm/radeon/r600.c
@@ -2940,6 +2940,12 @@ static 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);
@@ -3094,10 +3100,6 @@ 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 73051ce3121ee4..233a9b9fa1f7a4 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 */
@@ -535,9 +541,6 @@ 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 46fa1b07c56023..670b555d2ca229 100644
--- a/drivers/gpu/drm/radeon/rs600.c
+++ b/drivers/gpu/drm/radeon/rs600.c
@@ -923,6 +923,12 @@ 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 */
@@ -1047,9 +1053,6 @@ 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 5706d2ac75ab1c..fad6633dce3963 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 */
@@ -753,9 +759,6 @@ 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 ffcba730c57cb3..21c7d7b26e5554 100644
--- a/drivers/gpu/drm/radeon/rv515.c
+++ b/drivers/gpu/drm/radeon/rv515.c
@@ -532,6 +532,12 @@ 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 */
@@ -662,9 +668,6 @@ 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 d63fe1d0f53f9a..4181053cc94faa 100644
--- a/drivers/gpu/drm/radeon/rv770.c
+++ b/drivers/gpu/drm/radeon/rv770.c
@@ -1041,6 +1041,12 @@ 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);
@@ -1180,10 +1186,6 @@ 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 aefefd5000ee54..a964a967959f9c 100644
--- a/drivers/gpu/drm/radeon/si.c
+++ b/drivers/gpu/drm/radeon/si.c
@@ -4374,6 +4374,12 @@ 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);
@@ -4534,10 +4540,6 @@ 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);