media: ti-vpe: vpe: Fix Motion Vector vpdma stride

commit 3dc2046c ("[media] media: ti-vpe: vpe: allow use of user
specified stride") and commit da4414ea ("[media] media: ti-vpe: vpdma:
add support for user specified stride") resulted in the Motion Vector
stride to be the same as the image stride.

This caused memory corruption in the output image as mentioned in
commit 00db9699 ("[media] media: ti-vpe: vpe: Fix line stride
for output motion vector").

Fixes: 3dc2046c ("[media] media: ti-vpe: vpe: allow use of user specified stride")
Fixes: da4414ea ("[media] media: ti-vpe: vpdma: add support for user specified stride")
Signed-off-by: Benoit Parrot <bparrot@ti.com>
Acked-by: Nikhil Devshatwar <nikhil.nd@ti.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>

drivers/media/platform/ti-vpe/vpe.c

@@ -1013,11 +1013,14 @@ static void add_out_dtd(struct vpe_ctx *ctx, int port)
 	dma_addr_t dma_addr;
 	u32 flags = 0;
 	u32 offset = 0;
+	u32 stride;
 
 	if (port == VPE_PORT_MV_OUT) {
 		vpdma_fmt = &vpdma_misc_fmts[VPDMA_DATA_FMT_MV];
 		dma_addr = ctx->mv_buf_dma[mv_buf_selector];
 		q_data = &ctx->q_data[Q_DATA_SRC];
+		stride = ALIGN((q_data->width * vpdma_fmt->depth) >> 3,
+			       VPDMA_STRIDE_ALIGN);
 	} else {
 		/* to incorporate interleaved formats */
 		int plane = fmt->coplanar ? p_data->vb_part : 0;
@@ -1044,6 +1047,7 @@ static void add_out_dtd(struct vpe_ctx *ctx, int port)
 		}
 		/* Apply the offset */
 		dma_addr += offset;
+		stride = q_data->bytesperline[VPE_LUMA];
 	}
 
 	if (q_data->flags & Q_DATA_FRAME_1D)
@@ -1055,7 +1059,7 @@ static void add_out_dtd(struct vpe_ctx *ctx, int port)
 			   MAX_W, MAX_H);
 
 	vpdma_add_out_dtd(&ctx->desc_list, q_data->width,
-			  q_data->bytesperline[VPE_LUMA], &q_data->c_rect,
+			  stride, &q_data->c_rect,
 			  vpdma_fmt, dma_addr, MAX_OUT_WIDTH_REG1,
 			  MAX_OUT_HEIGHT_REG1, p_data->channel, flags);
 }
@@ -1074,10 +1078,13 @@ static void add_in_dtd(struct vpe_ctx *ctx, int port)
 	dma_addr_t dma_addr;
 	u32 flags = 0;
 	u32 offset = 0;
+	u32 stride;
 
 	if (port == VPE_PORT_MV_IN) {
 		vpdma_fmt = &vpdma_misc_fmts[VPDMA_DATA_FMT_MV];
 		dma_addr = ctx->mv_buf_dma[mv_buf_selector];
+		stride = ALIGN((q_data->width * vpdma_fmt->depth) >> 3,
+			       VPDMA_STRIDE_ALIGN);
 	} else {
 		/* to incorporate interleaved formats */
 		int plane = fmt->coplanar ? p_data->vb_part : 0;
@@ -1104,6 +1111,7 @@ static void add_in_dtd(struct vpe_ctx *ctx, int port)
 		}
 		/* Apply the offset */
 		dma_addr += offset;
+		stride = q_data->bytesperline[VPE_LUMA];
 
 		if (q_data->flags & Q_DATA_INTERLACED_SEQ_TB) {
 			/*
@@ -1139,10 +1147,10 @@ static void add_in_dtd(struct vpe_ctx *ctx, int port)
 	if (p_data->vb_part && fmt->fourcc == V4L2_PIX_FMT_NV12)
 		frame_height /= 2;
 
-	vpdma_add_in_dtd(&ctx->desc_list, q_data->width,
-			 q_data->bytesperline[VPE_LUMA], &q_data->c_rect,
-		vpdma_fmt, dma_addr, p_data->channel, field, flags, frame_width,
-		frame_height, 0, 0);
+	vpdma_add_in_dtd(&ctx->desc_list, q_data->width, stride,
+			 &q_data->c_rect, vpdma_fmt, dma_addr,
+			 p_data->channel, field, flags, frame_width,
+			 frame_height, 0, 0);
 }
 
 /*