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@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[ luis: backported to 3.16: adjusted context ]
Signed-off-by: Luis Henriques <luis.henriques@canonical.com>
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);
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 */
}
#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
(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