Prevent NULL pointer dereference in case when radeon_ring_fini() did it's job.
Reading of r100_cp_ring_info and radeon_ring_gfx debugfs entries will lead to a KP if ring buffer was deallocated, e.g. on failed ring test. Seen on PA-RISC machine having "radeon: ring test failed (scratch(0x8504)=0xCAFEDEAD)" issue.
Signed-off-by: Alex Ivanov gnidorah@p0n4ik.tk --- diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c index 2417571..413cdd1 100644 --- a/drivers/gpu/drm/radeon/r100.c +++ b/drivers/gpu/drm/radeon/r100.c @@ -2933,10 +2933,11 @@ static int r100_debugfs_cp_ring_info(struct seq_file *m, void *data) seq_printf(m, "CP_RB_RPTR 0x%08x\n", rdp); seq_printf(m, "%u free dwords in ring\n", ring->ring_free_dw); seq_printf(m, "%u dwords in ring\n", count); - for (j = 0; j <= count; j++) { - i = (rdp + j) & ring->ptr_mask; - seq_printf(m, "r[%04d]=0x%08x\n", i, ring->ring[i]); - } + if (ring->ready) + for (j = 0; j <= count; j++) { + i = (rdp + j) & ring->ptr_mask; + seq_printf(m, "r[%04d]=0x%08x\n", i, ring->ring[i]); + } return 0; }
diff --git a/drivers/gpu/drm/radeon/radeon_ring.c b/drivers/gpu/drm/radeon/radeon_ring.c index 46a25f0..a890756 100644 --- a/drivers/gpu/drm/radeon/radeon_ring.c +++ b/drivers/gpu/drm/radeon/radeon_ring.c @@ -839,10 +839,11 @@ static int radeon_debugfs_ring_info(struct seq_file *m, void *data) * packet that is the root issue */ i = (ring->rptr + ring->ptr_mask + 1 - 32) & ring->ptr_mask; - for (j = 0; j <= (count + 32); j++) { - seq_printf(m, "r[%5d]=0x%08x\n", i, ring->ring[i]); - i = (i + 1) & ring->ptr_mask; - } + if (ring->ready) + for (j = 0; j <= (count + 32); j++) { + seq_printf(m, "r[%5d]=0x%08x\n", i, ring->ring[i]); + i = (i + 1) & ring->ptr_mask; + } return 0; }
On Fri, Sep 20, 2013 at 9:36 AM, Alex Ivanov gnidorah@p0n4ik.tk wrote:
Prevent NULL pointer dereference in case when radeon_ring_fini() did it's job.
Reading of r100_cp_ring_info and radeon_ring_gfx debugfs entries will lead to a KP if ring buffer was deallocated, e.g. on failed ring test. Seen on PA-RISC machine having "radeon: ring test failed (scratch(0x8504)=0xCAFEDEAD)" issue.
Signed-off-by: Alex Ivanov gnidorah@p0n4ik.tk
Applied. thanks!
Alex
diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c index 2417571..413cdd1 100644 --- a/drivers/gpu/drm/radeon/r100.c +++ b/drivers/gpu/drm/radeon/r100.c @@ -2933,10 +2933,11 @@ static int r100_debugfs_cp_ring_info(struct seq_file *m, void *data) seq_printf(m, "CP_RB_RPTR 0x%08x\n", rdp); seq_printf(m, "%u free dwords in ring\n", ring->ring_free_dw); seq_printf(m, "%u dwords in ring\n", count);
for (j = 0; j <= count; j++) {i = (rdp + j) & ring->ptr_mask;seq_printf(m, "r[%04d]=0x%08x\n", i, ring->ring[i]);}
if (ring->ready)for (j = 0; j <= count; j++) {i = (rdp + j) & ring->ptr_mask;seq_printf(m, "r[%04d]=0x%08x\n", i, ring->ring[i]);} return 0;}
diff --git a/drivers/gpu/drm/radeon/radeon_ring.c b/drivers/gpu/drm/radeon/radeon_ring.c index 46a25f0..a890756 100644 --- a/drivers/gpu/drm/radeon/radeon_ring.c +++ b/drivers/gpu/drm/radeon/radeon_ring.c @@ -839,10 +839,11 @@ static int radeon_debugfs_ring_info(struct seq_file *m, void *data) * packet that is the root issue */ i = (ring->rptr + ring->ptr_mask + 1 - 32) & ring->ptr_mask;
for (j = 0; j <= (count + 32); j++) {seq_printf(m, "r[%5d]=0x%08x\n", i, ring->ring[i]);i = (i + 1) & ring->ptr_mask;}
if (ring->ready)for (j = 0; j <= (count + 32); j++) {seq_printf(m, "r[%5d]=0x%08x\n", i, ring->ring[i]);i = (i + 1) & ring->ptr_mask;} return 0;}
20.09.2013, в 22:33, Alex Deucher alexdeucher@gmail.com написал(а):
On Fri, Sep 20, 2013 at 9:36 AM, Alex Ivanov gnidorah@p0n4ik.tk wrote:
Prevent NULL pointer dereference in case when radeon_ring_fini() did it's job.
Reading of r100_cp_ring_info and radeon_ring_gfx debugfs entries will lead to a KP if ring buffer was deallocated, e.g. on failed ring test. Seen on PA-RISC machine having "radeon: ring test failed (scratch(0x8504)=0xCAFEDEAD)" issue.
Signed-off-by: Alex Ivanov gnidorah@p0n4ik.tk
Applied. thanks!
Alex
One more. Thank you!
Signed-off-by: Alex Ivanov gnidorah@p0n4ik.tk --- diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c index e29faa7..e6d1897 100644 --- a/drivers/gpu/drm/radeon/radeon_device.c +++ b/drivers/gpu/drm/radeon/radeon_device.c @@ -1319,6 +1319,11 @@ int radeon_device_init(struct radeon_device *rdev, if (r) return r; } + /* If ring buffer or PCI GART got uninitialized, we should't run tests */ + if (!rdev->accel_working) { + DRM_INFO("radeon: acceleration disabled, skipping tests and benchmark.\n"); + return 0; + } if ((radeon_testing & 1)) { radeon_test_moves(rdev); }
P.S.: don't run means don't allow to run, by either feeding radeon.test=X or radeon.benchmark=1
22.09.2013, в 5:18, Alex Ivanov gnidorah@p0n4ik.tk написал(а):
20.09.2013, в 22:33, Alex Deucher alexdeucher@gmail.com написал(а):
On Fri, Sep 20, 2013 at 9:36 AM, Alex Ivanov gnidorah@p0n4ik.tk wrote:
Prevent NULL pointer dereference in case when radeon_ring_fini() did it's job.
Reading of r100_cp_ring_info and radeon_ring_gfx debugfs entries will lead to a KP if ring buffer was deallocated, e.g. on failed ring test. Seen on PA-RISC machine having "radeon: ring test failed (scratch(0x8504)=0xCAFEDEAD)" issue.
Signed-off-by: Alex Ivanov gnidorah@p0n4ik.tk
Applied. thanks!
Alex
One more. Thank you!
Signed-off-by: Alex Ivanov gnidorah@p0n4ik.tk
diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c index e29faa7..e6d1897 100644 --- a/drivers/gpu/drm/radeon/radeon_device.c +++ b/drivers/gpu/drm/radeon/radeon_device.c @@ -1319,6 +1319,11 @@ int radeon_device_init(struct radeon_device *rdev, if (r) return r; }
/* If ring buffer or PCI GART got uninitialized, we should't run tests */if (!rdev->accel_working) {DRM_INFO("radeon: acceleration disabled, skipping tests and benchmark.\n");return 0;} if ((radeon_testing & 1)) { radeon_test_moves(rdev); }
dri-devel mailing list dri-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/dri-devel
On Mon, Sep 23, 2013 at 7:59 AM, Alex Ivanov gnidorah@p0n4ik.tk wrote:
P.S.: don't run means don't allow to run, by either feeding radeon.test=X or radeon.benchmark=1
Good catch. I applied a slightly different variant of the same idea.
Thanks,
Alex
22.09.2013, в 5:18, Alex Ivanov gnidorah@p0n4ik.tk написал(а):
20.09.2013, в 22:33, Alex Deucher alexdeucher@gmail.com написал(а):
On Fri, Sep 20, 2013 at 9:36 AM, Alex Ivanov gnidorah@p0n4ik.tk wrote:
Prevent NULL pointer dereference in case when radeon_ring_fini() did it's job.
Reading of r100_cp_ring_info and radeon_ring_gfx debugfs entries will lead to a KP if ring buffer was deallocated, e.g. on failed ring test. Seen on PA-RISC machine having "radeon: ring test failed (scratch(0x8504)=0xCAFEDEAD)" issue.
Signed-off-by: Alex Ivanov gnidorah@p0n4ik.tk
Applied. thanks!
Alex
One more. Thank you!
Signed-off-by: Alex Ivanov gnidorah@p0n4ik.tk
diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c index e29faa7..e6d1897 100644 --- a/drivers/gpu/drm/radeon/radeon_device.c +++ b/drivers/gpu/drm/radeon/radeon_device.c @@ -1319,6 +1319,11 @@ int radeon_device_init(struct radeon_device *rdev, if (r) return r; }
/* If ring buffer or PCI GART got uninitialized, we should't run tests */if (!rdev->accel_working) {DRM_INFO("radeon: acceleration disabled, skipping tests and benchmark.\n");return 0;} if ((radeon_testing & 1)) { radeon_test_moves(rdev); }
dri-devel mailing list dri-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/dri-devel
dri-devel mailing list dri-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/dri-devel
dri-devel@lists.freedesktop.org
-
Alex Deucher -
Alex Ivanov