[PATCH 3.16.y-ckt 006/135] drm/radeon: fix VM flush on cayman/aruba (v3)

[Luis Henriques]

3.16.7-ckt6 -stable review patch.  If anyone has any objections, please let me know.

------------------

From: Alex Deucher <alexander.deucher at amd.com>

commit cbfc35b90f3b4853d1eb9fcb82e99531d6a1c629 upstream.

We need to wait for the GPUVM flush to complete.  There
was some confusion as to how this mechanism was supposed
to work.  The operation is not atomic.  For GPU initiated
invalidations you need to read back a VM register to
introduce enough latency for the update to complete.

v2: drop gart changes
v3: just read back rather than polling

Reviewed-by: Christian König <christian.koenig at amd.com>
Signed-off-by: Alex Deucher <alexander.deucher at amd.com>
[ luis: backported to 3.16: adjusted context ]
Signed-off-by: Luis Henriques <luis.henriques at canonical.com>
---
 drivers/gpu/drm/radeon/ni.c     | 10 ++++++++++
 drivers/gpu/drm/radeon/ni_dma.c |  6 ++++++
 drivers/gpu/drm/radeon/nid.h    | 24 ++++++++++++++++++++++++
 3 files changed, 40 insertions(+)

diff --git a/drivers/gpu/drm/radeon/ni.c b/drivers/gpu/drm/radeon/ni.c
index 5a33ca681867..699bd1083e1f 100644
--- a/drivers/gpu/drm/radeon/ni.c
+++ b/drivers/gpu/drm/radeon/ni.c
@@ -2514,6 +2514,16 @@ void cayman_vm_flush(struct radeon_device *rdev, int ridx, struct radeon_vm *vm)
 	radeon_ring_write(ring, PACKET0(VM_INVALIDATE_REQUEST, 0));
 	radeon_ring_write(ring, 1 << vm->id);
 
+	/* wait for the invalidate to complete */
+	radeon_ring_write(ring, PACKET3(PACKET3_WAIT_REG_MEM, 5));
+	radeon_ring_write(ring, (WAIT_REG_MEM_FUNCTION(0) |  /* always */
+				 WAIT_REG_MEM_ENGINE(0))); /* me */
+	radeon_ring_write(ring, VM_INVALIDATE_REQUEST >> 2);
+	radeon_ring_write(ring, 0);
+	radeon_ring_write(ring, 0); /* ref */
+	radeon_ring_write(ring, 0); /* mask */
+	radeon_ring_write(ring, 0x20); /* poll interval */
+
 	/* sync PFP to ME, otherwise we might get invalid PFP reads */
 	radeon_ring_write(ring, PACKET3(PACKET3_PFP_SYNC_ME, 0));
 	radeon_ring_write(ring, 0x0);
diff --git a/drivers/gpu/drm/radeon/ni_dma.c b/drivers/gpu/drm/radeon/ni_dma.c
index 6927db4d8db7..5c8240ecee3a 100644
--- a/drivers/gpu/drm/radeon/ni_dma.c
+++ b/drivers/gpu/drm/radeon/ni_dma.c
@@ -398,5 +398,11 @@ void cayman_dma_vm_flush(struct radeon_device *rdev, int ridx, struct radeon_vm
 	radeon_ring_write(ring, DMA_PACKET(DMA_PACKET_SRBM_WRITE, 0, 0, 0));
 	radeon_ring_write(ring, (0xf << 16) | (VM_INVALIDATE_REQUEST >> 2));
 	radeon_ring_write(ring, 1 << vm->id);
+
+	/* wait for invalidate to complete */
+	radeon_ring_write(ring, DMA_SRBM_READ_PACKET);
+	radeon_ring_write(ring, (0xff << 20) | (VM_INVALIDATE_REQUEST >> 2));
+	radeon_ring_write(ring, 0); /* mask */
+	radeon_ring_write(ring, 0); /* value */
 }
 
diff --git a/drivers/gpu/drm/radeon/nid.h b/drivers/gpu/drm/radeon/nid.h
index 2e12e4d69253..ad7125486894 100644
--- a/drivers/gpu/drm/radeon/nid.h
+++ b/drivers/gpu/drm/radeon/nid.h
@@ -1133,6 +1133,23 @@
 #define	PACKET3_MEM_SEMAPHORE				0x39
 #define	PACKET3_MPEG_INDEX				0x3A
 #define	PACKET3_WAIT_REG_MEM				0x3C
+#define		WAIT_REG_MEM_FUNCTION(x)                ((x) << 0)
+                /* 0 - always
+		 * 1 - <
+		 * 2 - <=
+		 * 3 - ==
+		 * 4 - !=
+		 * 5 - >=
+		 * 6 - >
+		 */
+#define		WAIT_REG_MEM_MEM_SPACE(x)               ((x) << 4)
+                /* 0 - reg
+		 * 1 - mem
+		 */
+#define		WAIT_REG_MEM_ENGINE(x)                  ((x) << 8)
+                /* 0 - me
+		 * 1 - pfp
+		 */
 #define	PACKET3_MEM_WRITE				0x3D
 #define	PACKET3_PFP_SYNC_ME				0x42
 #define	PACKET3_SURFACE_SYNC				0x43
@@ -1272,6 +1289,13 @@
 					 (1 << 21) |			\
 					 (((n) & 0xFFFFF) << 0))
 
+#define DMA_SRBM_POLL_PACKET		((9 << 28) |			\
+					 (1 << 27) |			\
+					 (1 << 26))
+
+#define DMA_SRBM_READ_PACKET		((9 << 28) |			\
+					 (1 << 27))
+
 /* async DMA Packet types */
 #define	DMA_PACKET_WRITE				  0x2
 #define	DMA_PACKET_COPY					  0x3
-- 
2.1.4