Linux-libre 4.3.2-gnu

7 files changed, 486 insertions, 265 deletions

diff --git a/drivers/media/platform/coda/Makefile b/drivers/media/platform/coda/Makefile
index 834e504bf..9342ac57b 100644
--- a/drivers/media/platform/coda/Makefile
+++ b/drivers/media/platform/coda/Makefile
@@ -1,5 +1,5 @@
 ccflags-y += -I$(src)
 
-coda-objs := coda-common.o coda-bit.o coda-h264.o coda-jpeg.o
+coda-objs := coda-common.o coda-bit.o coda-gdi.o coda-h264.o coda-jpeg.o
 
 obj-$(CONFIG_VIDEO_CODA) += coda.o
diff --git a/drivers/media/platform/coda/coda-bit.c b/drivers/media/platform/coda/coda-bit.c
index 109797bb8..fd7819d89 100644
--- a/drivers/media/platform/coda/coda-bit.c
+++ b/drivers/media/platform/coda/coda-bit.c
@@ -226,8 +226,12 @@ void coda_fill_bitstream(struct coda_ctx *ctx, bool streaming)
 {
 	struct vb2_buffer *src_buf;
 	struct coda_buffer_meta *meta;
+	unsigned long flags;
 	u32 start;
 
+	if (ctx->bit_stream_param & CODA_BIT_STREAM_END_FLAG)
+		return;
+
 	while (v4l2_m2m_num_src_bufs_ready(ctx->fh.m2m_ctx) > 0) {
 		/*
 		 * Only queue a single JPEG into the bitstream buffer, except
@@ -252,6 +256,13 @@ void coda_fill_bitstream(struct coda_ctx *ctx, bool streaming)
 			continue;
 		}
 
+		/* Dump empty buffers */
+		if (!vb2_get_plane_payload(src_buf, 0)) {
+			src_buf = v4l2_m2m_src_buf_remove(ctx->fh.m2m_ctx);
+			v4l2_m2m_buf_done(src_buf, VB2_BUF_STATE_DONE);
+			continue;
+		}
+
 		/* Buffer start position */
 		start = ctx->bitstream_fifo.kfifo.in &
 			ctx->bitstream_fifo.kfifo.mask;
@@ -271,8 +282,13 @@ void coda_fill_bitstream(struct coda_ctx *ctx, bool streaming)
 				meta->start = start;
 				meta->end = ctx->bitstream_fifo.kfifo.in &
 					    ctx->bitstream_fifo.kfifo.mask;
+				spin_lock_irqsave(&ctx->buffer_meta_lock,
+						  flags);
 				list_add_tail(&meta->list,
 					      &ctx->buffer_meta_list);
+				ctx->num_metas++;
+				spin_unlock_irqrestore(&ctx->buffer_meta_lock,
+						       flags);
 
 				trace_coda_bit_queue(ctx, src_buf, meta);
 			}
@@ -331,7 +347,6 @@ static int coda_alloc_framebuffers(struct coda_ctx *ctx,
 {
 	struct coda_dev *dev = ctx->dev;
 	int width, height;
-	dma_addr_t paddr;
 	int ysize;
 	int ret;
 	int i;
@@ -351,7 +366,10 @@ static int coda_alloc_framebuffers(struct coda_ctx *ctx,
 		size_t size;
 		char *name;
 
-		size = ysize + ysize / 2;
+		if (ctx->tiled_map_type == GDI_TILED_FRAME_MB_RASTER_MAP)
+			size = round_up(ysize, 4096) + ysize / 2;
+		else
+			size = ysize + ysize / 2;
 		if (ctx->codec->src_fourcc == V4L2_PIX_FMT_H264 &&
 		    dev->devtype->product != CODA_DX6)
 			size += ysize / 4;
@@ -367,11 +385,23 @@ static int coda_alloc_framebuffers(struct coda_ctx *ctx,
 
 	/* Register frame buffers in the parameter buffer */
 	for (i = 0; i < ctx->num_internal_frames; i++) {
-		paddr = ctx->internal_frames[i].paddr;
+		u32 y, cb, cr;
+
 		/* Start addresses of Y, Cb, Cr planes */
-		coda_parabuf_write(ctx, i * 3 + 0, paddr);
-		coda_parabuf_write(ctx, i * 3 + 1, paddr + ysize);
-		coda_parabuf_write(ctx, i * 3 + 2, paddr + ysize + ysize / 4);
+		y = ctx->internal_frames[i].paddr;
+		cb = y + ysize;
+		cr = y + ysize + ysize/4;
+		if (ctx->tiled_map_type == GDI_TILED_FRAME_MB_RASTER_MAP) {
+			cb = round_up(cb, 4096);
+			cr = 0;
+			/* Packed 20-bit MSB of base addresses */
+			/* YYYYYCCC, CCyyyyyc, cccc.... */
+			y = (y & 0xfffff000) | cb >> 20;
+			cb = (cb & 0x000ff000) << 12;
+		}
+		coda_parabuf_write(ctx, i * 3 + 0, y);
+		coda_parabuf_write(ctx, i * 3 + 1, cb);
+		coda_parabuf_write(ctx, i * 3 + 2, cr);
 
 		/* mvcol buffer for h.264 */
 		if (ctx->codec->src_fourcc == V4L2_PIX_FMT_H264 &&
@@ -384,7 +414,7 @@ static int coda_alloc_framebuffers(struct coda_ctx *ctx,
 	/* mvcol buffer for mpeg4 */
 	if ((dev->devtype->product != CODA_DX6) &&
 	    (ctx->codec->src_fourcc == V4L2_PIX_FMT_MPEG4))
-		coda_parabuf_write(ctx, 97, ctx->internal_frames[i].paddr +
+		coda_parabuf_write(ctx, 97, ctx->internal_frames[0].paddr +
 					    ysize + ysize/4 + ysize/4);
 
 	return 0;
@@ -712,6 +742,32 @@ err_clk_per:
 	return ret;
 }
 
+static void coda9_set_frame_cache(struct coda_ctx *ctx, u32 fourcc)
+{
+	u32 cache_size, cache_config;
+
+	if (ctx->tiled_map_type == GDI_LINEAR_FRAME_MAP) {
+		/* Luma 2x0 page, 2x6 cache, chroma 2x0 page, 2x4 cache size */
+		cache_size = 0x20262024;
+		cache_config = 2 << CODA9_CACHE_PAGEMERGE_OFFSET;
+	} else {
+		/* Luma 0x2 page, 4x4 cache, chroma 0x2 page, 4x3 cache size */
+		cache_size = 0x02440243;
+		cache_config = 1 << CODA9_CACHE_PAGEMERGE_OFFSET;
+	}
+	coda_write(ctx->dev, cache_size, CODA9_CMD_SET_FRAME_CACHE_SIZE);
+	if (fourcc == V4L2_PIX_FMT_NV12) {
+		cache_config |= 32 << CODA9_CACHE_LUMA_BUFFER_SIZE_OFFSET |
+				16 << CODA9_CACHE_CR_BUFFER_SIZE_OFFSET |
+				0 << CODA9_CACHE_CB_BUFFER_SIZE_OFFSET;
+	} else {
+		cache_config |= 32 << CODA9_CACHE_LUMA_BUFFER_SIZE_OFFSET |
+				8 << CODA9_CACHE_CR_BUFFER_SIZE_OFFSET |
+				8 << CODA9_CACHE_CB_BUFFER_SIZE_OFFSET;
+	}
+	coda_write(ctx->dev, cache_config, CODA9_CMD_SET_FRAME_CACHE_CONFIG);
+}
+
 /*
  * Encoder context operations
  */
@@ -789,9 +845,12 @@ static int coda_start_encoding(struct coda_ctx *ctx)
 		break;
 	}
 
-	ctx->frame_mem_ctrl &= ~CODA_FRAME_CHROMA_INTERLEAVE;
+	ctx->frame_mem_ctrl &= ~(CODA_FRAME_CHROMA_INTERLEAVE | (0x3 << 9) |
+				 CODA9_FRAME_TILED2LINEAR);
 	if (q_data_src->fourcc == V4L2_PIX_FMT_NV12)
 		ctx->frame_mem_ctrl |= CODA_FRAME_CHROMA_INTERLEAVE;
+	if (ctx->tiled_map_type == GDI_TILED_FRAME_MB_RASTER_MAP)
+		ctx->frame_mem_ctrl |= (0x3 << 9) | CODA9_FRAME_TILED2LINEAR;
 	coda_write(dev, ctx->frame_mem_ctrl, CODA_REG_BIT_FRAME_MEM_CTRL);
 
 	if (dev->devtype->product == CODA_DX6) {
@@ -913,6 +972,9 @@ static int coda_start_encoding(struct coda_ctx *ctx)
 		value = (ctx->params.bitrate & CODA_RATECONTROL_BITRATE_MASK)
 			<< CODA_RATECONTROL_BITRATE_OFFSET;
 		value |=  1 & CODA_RATECONTROL_ENABLE_MASK;
+		value |= (ctx->params.vbv_delay &
+			  CODA_RATECONTROL_INITIALDELAY_MASK)
+			 << CODA_RATECONTROL_INITIALDELAY_OFFSET;
 		if (dev->devtype->product == CODA_960)
 			value |= BIT(31); /* disable autoskip */
 	} else {
@@ -920,7 +982,7 @@ static int coda_start_encoding(struct coda_ctx *ctx)
 	}
 	coda_write(dev, value, CODA_CMD_ENC_SEQ_RC_PARA);
 
-	coda_write(dev, 0, CODA_CMD_ENC_SEQ_RC_BUF_SIZE);
+	coda_write(dev, ctx->params.vbv_size, CODA_CMD_ENC_SEQ_RC_BUF_SIZE);
 	coda_write(dev, ctx->params.intra_refresh,
 		   CODA_CMD_ENC_SEQ_INTRA_REFRESH);
 
@@ -996,6 +1058,7 @@ static int coda_start_encoding(struct coda_ctx *ctx)
 		ret = -EFAULT;
 		goto out;
 	}
+	ctx->initialized = 1;
 
 	if (dst_fourcc != V4L2_PIX_FMT_JPEG) {
 		if (dev->devtype->product == CODA_960)
@@ -1036,6 +1099,8 @@ static int coda_start_encoding(struct coda_ctx *ctx)
 			coda_write(dev, ctx->iram_info.buf_btp_use,
 					CODA9_CMD_SET_FRAME_AXI_BTP_ADDR);
 
+			coda9_set_frame_cache(ctx, q_data_src->fourcc);
+
 			/* FIXME */
 			coda_write(dev, ctx->internal_frames[2].paddr,
 				   CODA9_CMD_SET_FRAME_SUBSAMP_A);
@@ -1326,6 +1391,9 @@ static void coda_seq_end_work(struct work_struct *work)
 	mutex_lock(&ctx->buffer_mutex);
 	mutex_lock(&dev->coda_mutex);
 
+	if (ctx->initialized == 0)
+		goto out;
+
 	v4l2_dbg(1, coda_debug, &dev->v4l2_dev,
 		 "%d: %s: sent command 'SEQ_END' to coda\n", ctx->idx,
 		 __func__);
@@ -1334,11 +1402,22 @@ static void coda_seq_end_work(struct work_struct *work)
 			 "CODA_COMMAND_SEQ_END failed\n");
 	}
 
+	/*
+	 * FIXME: Sometimes h.264 encoding fails with 8-byte sequences missing
+	 * from the output stream after the h.264 decoder has run. Resetting the
+	 * hardware after the decoder has finished seems to help.
+	 */
+	if (dev->devtype->product == CODA_960)
+		coda_hw_reset(ctx);
+
 	kfifo_init(&ctx->bitstream_fifo,
 		ctx->bitstream.vaddr, ctx->bitstream.size);
 
 	coda_free_framebuffers(ctx);
 
+	ctx->initialized = 0;
+
+out:
 	mutex_unlock(&dev->coda_mutex);
 	mutex_unlock(&ctx->buffer_mutex);
 }
@@ -1448,9 +1527,12 @@ static int __coda_start_decoding(struct coda_ctx *ctx)
 	/* Update coda bitstream read and write pointers from kfifo */
 	coda_kfifo_sync_to_device_full(ctx);
 
-	ctx->frame_mem_ctrl &= ~CODA_FRAME_CHROMA_INTERLEAVE;
+	ctx->frame_mem_ctrl &= ~(CODA_FRAME_CHROMA_INTERLEAVE | (0x3 << 9) |
+				 CODA9_FRAME_TILED2LINEAR);
 	if (dst_fourcc == V4L2_PIX_FMT_NV12)
 		ctx->frame_mem_ctrl |= CODA_FRAME_CHROMA_INTERLEAVE;
+	if (ctx->tiled_map_type == GDI_TILED_FRAME_MB_RASTER_MAP)
+		ctx->frame_mem_ctrl |= (0x3 << 9) | CODA9_FRAME_TILED2LINEAR;
 	coda_write(dev, ctx->frame_mem_ctrl, CODA_REG_BIT_FRAME_MEM_CTRL);
 
 	ctx->display_idx = -1;
@@ -1496,6 +1578,7 @@ static int __coda_start_decoding(struct coda_ctx *ctx)
 		coda_write(dev, 0, CODA_REG_BIT_BIT_STREAM_PARAM);
 		return -ETIMEDOUT;
 	}
+	ctx->initialized = 1;
 
 	/* Update kfifo out pointer from coda bitstream read pointer */
 	coda_kfifo_sync_from_device(ctx);
@@ -1578,30 +1661,13 @@ static int __coda_start_decoding(struct coda_ctx *ctx)
 				CODA7_CMD_SET_FRAME_AXI_DBKC_ADDR);
 		coda_write(dev, ctx->iram_info.buf_ovl_use,
 				CODA7_CMD_SET_FRAME_AXI_OVL_ADDR);
-		if (dev->devtype->product == CODA_960)
+		if (dev->devtype->product == CODA_960) {
 			coda_write(dev, ctx->iram_info.buf_btp_use,
 					CODA9_CMD_SET_FRAME_AXI_BTP_ADDR);
-	}
 
-	if (dev->devtype->product == CODA_960) {
-		int cbb_size, crb_size;
-
-		coda_write(dev, -1, CODA9_CMD_SET_FRAME_DELAY);
-		/* Luma 2x0 page, 2x6 cache, chroma 2x0 page, 2x4 cache size */
-		coda_write(dev, 0x20262024, CODA9_CMD_SET_FRAME_CACHE_SIZE);
-
-		if (dst_fourcc == V4L2_PIX_FMT_NV12) {
-			cbb_size = 0;
-			crb_size = 16;
-		} else {
-			cbb_size = 8;
-			crb_size = 8;
+			coda_write(dev, -1, CODA9_CMD_SET_FRAME_DELAY);
+			coda9_set_frame_cache(ctx, dst_fourcc);
 		}
-		coda_write(dev, 2 << CODA9_CACHE_PAGEMERGE_OFFSET |
-				32 << CODA9_CACHE_LUMA_BUFFER_SIZE_OFFSET |
-				cbb_size << CODA9_CACHE_CB_BUFFER_SIZE_OFFSET |
-				crb_size << CODA9_CACHE_CR_BUFFER_SIZE_OFFSET,
-				CODA9_CMD_SET_FRAME_CACHE_CONFIG);
 	}
 
 	if (src_fourcc == V4L2_PIX_FMT_H264) {
@@ -1654,6 +1720,7 @@ static int coda_prepare_decode(struct coda_ctx *ctx)
 	struct coda_dev *dev = ctx->dev;
 	struct coda_q_data *q_data_dst;
 	struct coda_buffer_meta *meta;
+	unsigned long flags;
 	u32 reg_addr, reg_stride;
 
 	dst_buf = v4l2_m2m_next_dst_buf(ctx->fh.m2m_ctx);
@@ -1732,6 +1799,7 @@ static int coda_prepare_decode(struct coda_ctx *ctx)
 		coda_write(dev, ctx->iram_info.axi_sram_use,
 				CODA7_REG_BIT_AXI_SRAM_USE);
 
+	spin_lock_irqsave(&ctx->buffer_meta_lock, flags);
 	meta = list_first_entry_or_null(&ctx->buffer_meta_list,
 					struct coda_buffer_meta, list);
 
@@ -1751,6 +1819,7 @@ static int coda_prepare_decode(struct coda_ctx *ctx)
 			kfifo_in(&ctx->bitstream_fifo, buf, pad);
 		}
 	}
+	spin_unlock_irqrestore(&ctx->buffer_meta_lock, flags);
 
 	coda_kfifo_sync_to_device_full(ctx);
 
@@ -1772,6 +1841,7 @@ static void coda_finish_decode(struct coda_ctx *ctx)
 	struct vb2_buffer *dst_buf;
 	struct coda_buffer_meta *meta;
 	unsigned long payload;
+	unsigned long flags;
 	int width, height;
 	int decoded_idx;
 	int display_idx;
@@ -1897,12 +1967,21 @@ static void coda_finish_decode(struct coda_ctx *ctx)
 	} else {
 		val = coda_read(dev, CODA_RET_DEC_PIC_FRAME_NUM) - 1;
 		val -= ctx->sequence_offset;
-		mutex_lock(&ctx->bitstream_mutex);
+		spin_lock_irqsave(&ctx->buffer_meta_lock, flags);
 		if (!list_empty(&ctx->buffer_meta_list)) {
 			meta = list_first_entry(&ctx->buffer_meta_list,
 					      struct coda_buffer_meta, list);
 			list_del(&meta->list);
-			if (val != (meta->sequence & 0xffff)) {
+			ctx->num_metas--;
+			spin_unlock_irqrestore(&ctx->buffer_meta_lock, flags);
+			/*
+			 * Clamp counters to 16 bits for comparison, as the HW
+			 * counter rolls over at this point for h.264. This
+			 * may be different for other formats, but using 16 bits
+			 * should be enough to detect most errors and saves us
+			 * from doing different things based on the format.
+			 */
+			if ((val & 0xffff) != (meta->sequence & 0xffff)) {
 				v4l2_err(&dev->v4l2_dev,
 					 "sequence number mismatch (%d(%d) != %d)\n",
 					 val, ctx->sequence_offset,
@@ -1911,13 +1990,13 @@ static void coda_finish_decode(struct coda_ctx *ctx)
 			ctx->frame_metas[decoded_idx] = *meta;
 			kfree(meta);
 		} else {
+			spin_unlock_irqrestore(&ctx->buffer_meta_lock, flags);
 			v4l2_err(&dev->v4l2_dev, "empty timestamp list!\n");
 			memset(&ctx->frame_metas[decoded_idx], 0,
 			       sizeof(struct coda_buffer_meta));
 			ctx->frame_metas[decoded_idx].sequence = val;
 			ctx->sequence_offset++;
 		}
-		mutex_unlock(&ctx->bitstream_mutex);
 
 		trace_coda_dec_pic_done(ctx, &ctx->frame_metas[decoded_idx]);
 
@@ -1960,7 +2039,7 @@ static void coda_finish_decode(struct coda_ctx *ctx)
 		dst_buf->v4l2_buf.timecode = meta->timecode;
 		dst_buf->v4l2_buf.timestamp = meta->timestamp;
 
-		trace_coda_dec_rot_done(ctx, meta, dst_buf);
+		trace_coda_dec_rot_done(ctx, dst_buf, meta);
 
 		switch (q_data_dst->fourcc) {
 		case V4L2_PIX_FMT_YUV420:
diff --git a/drivers/media/platform/coda/coda-common.c b/drivers/media/platform/coda/coda-common.c
index 6e41b0ce8..c48ef89f9 100644
--- a/drivers/media/platform/coda/coda-common.c
+++ b/drivers/media/platform/coda/coda-common.c
@@ -15,6 +15,7 @@
 #include <linux/debugfs.h>
 #include <linux/delay.h>
 #include <linux/firmware.h>
+#include <linux/gcd.h>
 #include <linux/genalloc.h>
 #include <linux/interrupt.h>
 #include <linux/io.h>
@@ -61,10 +62,9 @@ int coda_debug;
 module_param(coda_debug, int, 0644);
 MODULE_PARM_DESC(coda_debug, "Debug level (0-2)");
 
-struct coda_fmt {
-	char *name;
-	u32 fourcc;
-};
+static int disable_tiling;
+module_param(disable_tiling, int, 0644);
+MODULE_PARM_DESC(disable_tiling, "Disable tiled frame buffers");
 
 void coda_write(struct coda_dev *dev, u32 data, u32 reg)
 {
@@ -90,17 +90,17 @@ void coda_write_base(struct coda_ctx *ctx, struct coda_q_data *q_data,
 	u32 base_cb, base_cr;
 
 	switch (q_data->fourcc) {
-	case V4L2_PIX_FMT_YVU420:
-		/* Switch Cb and Cr for YVU420 format */
-		base_cr = base_y + q_data->bytesperline * q_data->height;
-		base_cb = base_cr + q_data->bytesperline * q_data->height / 4;
-		break;
-	case V4L2_PIX_FMT_YUV420:
 	case V4L2_PIX_FMT_NV12:
+	case V4L2_PIX_FMT_YUV420:
 	default:
 		base_cb = base_y + q_data->bytesperline * q_data->height;
 		base_cr = base_cb + q_data->bytesperline * q_data->height / 4;
 		break;
+	case V4L2_PIX_FMT_YVU420:
+		/* Switch Cb and Cr for YVU420 format */
+		base_cr = base_y + q_data->bytesperline * q_data->height;
+		base_cb = base_cr + q_data->bytesperline * q_data->height / 4;
+		break;
 	case V4L2_PIX_FMT_YUV422P:
 		base_cb = base_y + q_data->bytesperline * q_data->height;
 		base_cr = base_cb + q_data->bytesperline * q_data->height / 2;
@@ -111,40 +111,6 @@ void coda_write_base(struct coda_ctx *ctx, struct coda_q_data *q_data,
 	coda_write(ctx->dev, base_cr, reg_y + 8);
 }
 
-/*
- * Array of all formats supported by any version of Coda:
- */
-static const struct coda_fmt coda_formats[] = {
-	{
-		.name = "YUV 4:2:0 Planar, YCbCr",
-		.fourcc = V4L2_PIX_FMT_YUV420,
-	},
-	{
-		.name = "YUV 4:2:0 Planar, YCrCb",
-		.fourcc = V4L2_PIX_FMT_YVU420,
-	},
-	{
-		.name = "YUV 4:2:0 Partial interleaved Y/CbCr",
-		.fourcc = V4L2_PIX_FMT_NV12,
-	},
-	{
-		.name = "YUV 4:2:2 Planar, YCbCr",
-		.fourcc = V4L2_PIX_FMT_YUV422P,
-	},
-	{
-		.name = "H264 Encoded Stream",
-		.fourcc = V4L2_PIX_FMT_H264,
-	},
-	{
-		.name = "MPEG4 Encoded Stream",
-		.fourcc = V4L2_PIX_FMT_MPEG4,
-	},
-	{
-		.name = "JPEG Encoded Images",
-		.fourcc = V4L2_PIX_FMT_JPEG,
-	},
-};
-
 #define CODA_CODEC(mode, src_fourcc, dst_fourcc, max_w, max_h) \
 	{ mode, src_fourcc, dst_fourcc, max_w, max_h }
 
@@ -190,9 +156,9 @@ static const struct coda_video_device coda_bit_encoder = {
 	.type = CODA_INST_ENCODER,
 	.ops = &coda_bit_encode_ops,
 	.src_formats = {
+		V4L2_PIX_FMT_NV12,
 		V4L2_PIX_FMT_YUV420,
 		V4L2_PIX_FMT_YVU420,
-		V4L2_PIX_FMT_NV12,
 	},
 	.dst_formats = {
 		V4L2_PIX_FMT_H264,
@@ -205,9 +171,9 @@ static const struct coda_video_device coda_bit_jpeg_encoder = {
 	.type = CODA_INST_ENCODER,
 	.ops = &coda_bit_encode_ops,
 	.src_formats = {
+		V4L2_PIX_FMT_NV12,
 		V4L2_PIX_FMT_YUV420,
 		V4L2_PIX_FMT_YVU420,
-		V4L2_PIX_FMT_NV12,
 		V4L2_PIX_FMT_YUV422P,
 	},
 	.dst_formats = {
@@ -224,9 +190,9 @@ static const struct coda_video_device coda_bit_decoder = {
 		V4L2_PIX_FMT_MPEG4,
 	},
 	.dst_formats = {
+		V4L2_PIX_FMT_NV12,
 		V4L2_PIX_FMT_YUV420,
 		V4L2_PIX_FMT_YVU420,
-		V4L2_PIX_FMT_NV12,
 	},
 };
 
@@ -238,9 +204,9 @@ static const struct coda_video_device coda_bit_jpeg_decoder = {
 		V4L2_PIX_FMT_JPEG,
 	},
 	.dst_formats = {
+		V4L2_PIX_FMT_NV12,
 		V4L2_PIX_FMT_YUV420,
 		V4L2_PIX_FMT_YVU420,
-		V4L2_PIX_FMT_NV12,
 		V4L2_PIX_FMT_YUV422P,
 	},
 };
@@ -261,38 +227,21 @@ static const struct coda_video_device *coda9_video_devices[] = {
 	&coda_bit_decoder,
 };
 
-static bool coda_format_is_yuv(u32 fourcc)
+/*
+ * Normalize all supported YUV 4:2:0 formats to the value used in the codec
+ * tables.
+ */
+static u32 coda_format_normalize_yuv(u32 fourcc)
 {
 	switch (fourcc) {
+	case V4L2_PIX_FMT_NV12:
 	case V4L2_PIX_FMT_YUV420:
 	case V4L2_PIX_FMT_YVU420:
-	case V4L2_PIX_FMT_NV12:
 	case V4L2_PIX_FMT_YUV422P:
-		return true;
+		return V4L2_PIX_FMT_YUV420;
 	default:
-		return false;
-	}
-}
-
-static const char *coda_format_name(u32 fourcc)
-{
-	int i;
-
-	for (i = 0; i < ARRAY_SIZE(coda_formats); i++) {
-		if (coda_formats[i].fourcc == fourcc)
-			return coda_formats[i].name;
+		return fourcc;
 	}
-
-	return NULL;
-}
-
-/*
- * Normalize all supported YUV 4:2:0 formats to the value used in the codec
- * tables.
- */
-static u32 coda_format_normalize_yuv(u32 fourcc)
-{
-	return coda_format_is_yuv(fourcc) ? V4L2_PIX_FMT_YUV420 : fourcc;
 }
 
 static const struct coda_codec *coda_find_codec(struct coda_dev *dev,
@@ -396,7 +345,6 @@ static int coda_enum_fmt(struct file *file, void *priv,
 	struct video_device *vdev = video_devdata(file);
 	const struct coda_video_device *cvd = to_coda_video_device(vdev);
 	const u32 *formats;
-	const char *name;
 
 	if (f->type == V4L2_BUF_TYPE_VIDEO_OUTPUT)
 		formats = cvd->src_formats;
@@ -408,11 +356,7 @@ static int coda_enum_fmt(struct file *file, void *priv,
 	if (f->index >= CODA_MAX_FORMATS || formats[f->index] == 0)
 		return -EINVAL;
 
-	name = coda_format_name(formats[f->index]);
-	strlcpy(f->description, name, sizeof(f->description));
 	f->pixelformat = formats[f->index];
-	if (!coda_format_is_yuv(formats[f->index]))
-		f->flags |= V4L2_FMT_FLAG_COMPRESSED;
 
 	return 0;
 }
@@ -504,9 +448,9 @@ static int coda_try_fmt(struct coda_ctx *ctx, const struct coda_codec *codec,
 			      S_ALIGN);
 
 	switch (f->fmt.pix.pixelformat) {
+	case V4L2_PIX_FMT_NV12:
 	case V4L2_PIX_FMT_YUV420:
 	case V4L2_PIX_FMT_YVU420:
-	case V4L2_PIX_FMT_NV12:
 		/*
 		 * Frame stride must be at least multiple of 8,
 		 * but multiple of 16 for h.264 or JPEG 4:2:x
@@ -645,6 +589,22 @@ static int coda_s_fmt(struct coda_ctx *ctx, struct v4l2_format *f)
 	q_data->rect.width = f->fmt.pix.width;
 	q_data->rect.height = f->fmt.pix.height;
 
+	switch (f->fmt.pix.pixelformat) {
+	case V4L2_PIX_FMT_NV12:
+		if (f->type == V4L2_BUF_TYPE_VIDEO_OUTPUT) {
+			ctx->tiled_map_type = GDI_TILED_FRAME_MB_RASTER_MAP;
+			if (!disable_tiling)
+				break;
+		}
+		/* else fall through */
+	case V4L2_PIX_FMT_YUV420:
+	case V4L2_PIX_FMT_YVU420:
+		ctx->tiled_map_type = GDI_LINEAR_FRAME_MAP;
+		break;
+	default:
+		break;
+	}
+
 	v4l2_dbg(1, coda_debug, &ctx->dev->v4l2_dev,
 		"Setting format for type %d, wxh: %dx%d, fmt: %d\n",
 		f->type, q_data->width, q_data->height, q_data->fourcc);
@@ -831,6 +791,104 @@ static int coda_decoder_cmd(struct file *file, void *fh,
 	return 0;
 }
 
+static int coda_g_parm(struct file *file, void *fh, struct v4l2_streamparm *a)
+{
+	struct coda_ctx *ctx = fh_to_ctx(fh);
+	struct v4l2_fract *tpf;
+
+	if (a->type != V4L2_BUF_TYPE_VIDEO_OUTPUT)
+		return -EINVAL;
+
+	a->parm.output.capability = V4L2_CAP_TIMEPERFRAME;
+	tpf = &a->parm.output.timeperframe;
+	tpf->denominator = ctx->params.framerate & CODA_FRATE_RES_MASK;
+	tpf->numerator = 1 + (ctx->params.framerate >>
+			      CODA_FRATE_DIV_OFFSET);
+
+	return 0;
+}
+
+/*
+ * Approximate timeperframe v4l2_fract with values that can be written
+ * into the 16-bit CODA_FRATE_DIV and CODA_FRATE_RES fields.
+ */
+static void coda_approximate_timeperframe(struct v4l2_fract *timeperframe)
+{
+	struct v4l2_fract s = *timeperframe;
+	struct v4l2_fract f0;
+	struct v4l2_fract f1 = { 1, 0 };
+	struct v4l2_fract f2 = { 0, 1 };
+	unsigned int i, div, s_denominator;
+
+	/* Lower bound is 1/65535 */
+	if (s.numerator == 0 || s.denominator / s.numerator > 65535) {
+		timeperframe->numerator = 1;
+		timeperframe->denominator = 65535;
+		return;
+	}
+
+	/* Upper bound is 65536/1, map everything above to infinity */
+	if (s.denominator == 0 || s.numerator / s.denominator > 65536) {
+		timeperframe->numerator = 1;
+		timeperframe->denominator = 0;
+		return;
+	}
+
+	/* Reduce fraction to lowest terms */
+	div = gcd(s.numerator, s.denominator);
+	if (div > 1) {
+		s.numerator /= div;
+		s.denominator /= div;
+	}
+
+	if (s.numerator <= 65536 && s.denominator < 65536) {
+		*timeperframe = s;
+		return;
+	}
+
+	/* Find successive convergents from continued fraction expansion */
+	while (f2.numerator <= 65536 && f2.denominator < 65536) {
+		f0 = f1;
+		f1 = f2;
+
+		/* Stop when f2 exactly equals timeperframe */
+		if (s.numerator == 0)
+			break;
+
+		i = s.denominator / s.numerator;
+
+		f2.numerator = f0.numerator + i * f1.numerator;
+		f2.denominator = f0.denominator + i * f2.denominator;
+
+		s_denominator = s.numerator;
+		s.numerator = s.denominator % s.numerator;
+		s.denominator = s_denominator;
+	}
+
+	*timeperframe = f1;
+}
+
+static uint32_t coda_timeperframe_to_frate(struct v4l2_fract *timeperframe)
+{
+	return ((timeperframe->numerator - 1) << CODA_FRATE_DIV_OFFSET) |
+		timeperframe->denominator;
+}
+
+static int coda_s_parm(struct file *file, void *fh, struct v4l2_streamparm *a)
+{
+	struct coda_ctx *ctx = fh_to_ctx(fh);
+	struct v4l2_fract *tpf;
+
+	if (a->type != V4L2_BUF_TYPE_VIDEO_OUTPUT)
+		return -EINVAL;
+
+	tpf = &a->parm.output.timeperframe;
+	coda_approximate_timeperframe(tpf);
+	ctx->params.framerate = coda_timeperframe_to_frate(tpf);
+
+	return 0;
+}
+
 static int coda_subscribe_event(struct v4l2_fh *fh,
 				const struct v4l2_event_subscription *sub)
 {
@@ -871,31 +929,13 @@ static const struct v4l2_ioctl_ops coda_ioctl_ops = {
 	.vidioc_try_decoder_cmd	= coda_try_decoder_cmd,
 	.vidioc_decoder_cmd	= coda_decoder_cmd,
 
+	.vidioc_g_parm		= coda_g_parm,
+	.vidioc_s_parm		= coda_s_parm,
+
 	.vidioc_subscribe_event = coda_subscribe_event,
 	.vidioc_unsubscribe_event = v4l2_event_unsubscribe,
 };
 
-void coda_set_gdi_regs(struct coda_ctx *ctx)
-{
-	struct gdi_tiled_map *tiled_map = &ctx->tiled_map;
-	struct coda_dev *dev = ctx->dev;
-	int i;
-
-	for (i = 0; i < 16; i++)
-		coda_write(dev, tiled_map->xy2ca_map[i],
-				CODA9_GDI_XY2_CAS_0 + 4 * i);
-	for (i = 0; i < 4; i++)
-		coda_write(dev, tiled_map->xy2ba_map[i],
-				CODA9_GDI_XY2_BA_0 + 4 * i);
-	for (i = 0; i < 16; i++)
-		coda_write(dev, tiled_map->xy2ra_map[i],
-				CODA9_GDI_XY2_RAS_0 + 4 * i);
-	coda_write(dev, tiled_map->xy2rbc_config, CODA9_GDI_XY2_RBC_CONFIG);
-	for (i = 0; i < 32; i++)
-		coda_write(dev, tiled_map->rbc2axi_map[i],
-				CODA9_GDI_RBC2_AXI_0 + 4 * i);
-}
-
 /*
  * Mem-to-mem operations.
  */
@@ -949,14 +989,14 @@ static void coda_pic_run_work(struct work_struct *work)
 static int coda_job_ready(void *m2m_priv)
 {
 	struct coda_ctx *ctx = m2m_priv;
+	int src_bufs = v4l2_m2m_num_src_bufs_ready(ctx->fh.m2m_ctx);
 
 	/*
 	 * For both 'P' and 'key' frame cases 1 picture
 	 * and 1 frame are needed. In the decoder case,
 	 * the compressed frame can be in the bitstream.
 	 */
-	if (!v4l2_m2m_num_src_bufs_ready(ctx->fh.m2m_ctx) &&
-	    ctx->inst_type != CODA_INST_DECODER) {
+	if (!src_bufs && ctx->inst_type != CODA_INST_DECODER) {
 		v4l2_dbg(1, coda_debug, &ctx->dev->v4l2_dev,
 			 "not ready: not enough video buffers.\n");
 		return 0;
@@ -969,27 +1009,17 @@ static int coda_job_ready(void *m2m_priv)
 	}
 
 	if (ctx->inst_type == CODA_INST_DECODER && ctx->use_bit) {
-		struct list_head *meta;
-		bool stream_end;
-		int num_metas;
-		int src_bufs;
+		bool stream_end = ctx->bit_stream_param &
+				  CODA_BIT_STREAM_END_FLAG;
+		int num_metas = ctx->num_metas;
 
-		if (ctx->hold && !v4l2_m2m_num_src_bufs_ready(ctx->fh.m2m_ctx)) {
+		if (ctx->hold && !src_bufs) {
 			v4l2_dbg(1, coda_debug, &ctx->dev->v4l2_dev,
 				 "%d: not ready: on hold for more buffers.\n",
 				 ctx->idx);
 			return 0;
 		}
 
-		stream_end = ctx->bit_stream_param &
-			     CODA_BIT_STREAM_END_FLAG;
-
-		num_metas = 0;
-		list_for_each(meta, &ctx->buffer_meta_list)
-			num_metas++;
-
-		src_bufs = v4l2_m2m_num_src_bufs_ready(ctx->fh.m2m_ctx);
-
 		if (!stream_end && (num_metas + src_bufs) < 2) {
 			v4l2_dbg(1, coda_debug, &ctx->dev->v4l2_dev,
 				 "%d: not ready: need 2 buffers available (%d, %d)\n",
@@ -998,8 +1028,8 @@ static int coda_job_ready(void *m2m_priv)
 		}
 
 
-		if (!v4l2_m2m_num_src_bufs_ready(ctx->fh.m2m_ctx) &&
-		    !stream_end && (coda_get_bitstream_payload(ctx) < 512)) {
+		if (!src_bufs && !stream_end &&
+		    (coda_get_bitstream_payload(ctx) < 512)) {
 			v4l2_dbg(1, coda_debug, &ctx->dev->v4l2_dev,
 				 "%d: not ready: not enough bitstream data (%d).\n",
 				 ctx->idx, coda_get_bitstream_payload(ctx));
@@ -1015,6 +1045,7 @@ static int coda_job_ready(void *m2m_priv)
 
 	v4l2_dbg(1, coda_debug, &ctx->dev->v4l2_dev,
 			"job ready\n");
+
 	return 1;
 }
 
@@ -1052,32 +1083,6 @@ static const struct v4l2_m2m_ops coda_m2m_ops = {
 	.unlock		= coda_unlock,
 };
 
-static void coda_set_tiled_map_type(struct coda_ctx *ctx, int tiled_map_type)
-{
-	struct gdi_tiled_map *tiled_map = &ctx->tiled_map;
-	int luma_map, chro_map, i;
-
-	memset(tiled_map, 0, sizeof(*tiled_map));
-
-	luma_map = 64;
-	chro_map = 64;
-	tiled_map->map_type = tiled_map_type;
-	for (i = 0; i < 16; i++)
-		tiled_map->xy2ca_map[i] = luma_map << 8 | chro_map;
-	for (i = 0; i < 4; i++)
-		tiled_map->xy2ba_map[i] = luma_map << 8 | chro_map;
-	for (i = 0; i < 16; i++)
-		tiled_map->xy2ra_map[i] = luma_map << 8 | chro_map;
-
-	if (tiled_map_type == GDI_LINEAR_FRAME_MAP) {
-		tiled_map->xy2rbc_config = 0;
-	} else {
-		dev_err(&ctx->dev->plat_dev->dev, "invalid map type: %d\n",
-			tiled_map_type);
-		return;
-	}
-}
-
 static void set_default_params(struct coda_ctx *ctx)
 {
 	unsigned int max_w, max_h, usize, csize;
@@ -1094,8 +1099,8 @@ static void set_default_params(struct coda_ctx *ctx)
 	ctx->params.framerate = 30;
 
 	/* Default formats for output and input queues */
-	ctx->q_data[V4L2_M2M_SRC].fourcc = ctx->codec->src_fourcc;
-	ctx->q_data[V4L2_M2M_DST].fourcc = ctx->codec->dst_fourcc;
+	ctx->q_data[V4L2_M2M_SRC].fourcc = ctx->cvd->src_formats[0];
+	ctx->q_data[V4L2_M2M_DST].fourcc = ctx->cvd->dst_formats[0];
 	ctx->q_data[V4L2_M2M_SRC].width = max_w;
 	ctx->q_data[V4L2_M2M_SRC].height = max_h;
 	ctx->q_data[V4L2_M2M_DST].width = max_w;
@@ -1116,8 +1121,11 @@ static void set_default_params(struct coda_ctx *ctx)
 	ctx->q_data[V4L2_M2M_DST].rect.width = max_w;
 	ctx->q_data[V4L2_M2M_DST].rect.height = max_h;
 
-	if (ctx->dev->devtype->product == CODA_960)
-		coda_set_tiled_map_type(ctx, GDI_LINEAR_FRAME_MAP);
+	/*
+	 * Since the RBC2AXI logic only supports a single chroma plane,
+	 * macroblock tiling only works for to NV12 pixel format.
+	 */
+	ctx->tiled_map_type = GDI_LINEAR_FRAME_MAP;
 }
 
 /*
@@ -1244,9 +1252,7 @@ static int coda_start_streaming(struct vb2_queue *q, unsigned int count)
 
 	q_data_src = get_q_data(ctx, V4L2_BUF_TYPE_VIDEO_OUTPUT);
 	if (q->type == V4L2_BUF_TYPE_VIDEO_OUTPUT) {
-		if (q_data_src->fourcc == V4L2_PIX_FMT_H264 ||
-		    (q_data_src->fourcc == V4L2_PIX_FMT_JPEG &&
-		     ctx->dev->devtype->product == CODA_7541)) {
+		if (ctx->inst_type == CODA_INST_DECODER && ctx->use_bit) {
 			/* copy the buffers that were queued before streamon */
 			mutex_lock(&ctx->bitstream_mutex);
 			coda_fill_bitstream(ctx, false);
@@ -1315,7 +1321,6 @@ static int coda_start_streaming(struct vb2_queue *q, unsigned int count)
 			goto err;
 	}
 
-	ctx->initialized = 1;
 	return ret;
 
 err:
@@ -1334,6 +1339,7 @@ static void coda_stop_streaming(struct vb2_queue *q)
 	struct coda_ctx *ctx = vb2_get_drv_priv(q);
 	struct coda_dev *dev = ctx->dev;
 	struct vb2_buffer *buf;
+	unsigned long flags;
 	bool stop;
 
 	stop = ctx->streamon_out && ctx->streamon_cap;
@@ -1368,20 +1374,23 @@ static void coda_stop_streaming(struct vb2_queue *q)
 			queue_work(dev->workqueue, &ctx->seq_end_work);
 			flush_work(&ctx->seq_end_work);
 		}
-		mutex_lock(&ctx->bitstream_mutex);
+		spin_lock_irqsave(&ctx->buffer_meta_lock, flags);
 		while (!list_empty(&ctx->buffer_meta_list)) {
 			meta = list_first_entry(&ctx->buffer_meta_list,
 						struct coda_buffer_meta, list);
 			list_del(&meta->list);
 			kfree(meta);
 		}
-		mutex_unlock(&ctx->bitstream_mutex);
+		ctx->num_metas = 0;
+		spin_unlock_irqrestore(&ctx->buffer_meta_lock, flags);
 		kfifo_init(&ctx->bitstream_fifo,
 			ctx->bitstream.vaddr, ctx->bitstream.size);
-		ctx->initialized = 0;
 		ctx->runcounter = 0;
 		ctx->aborting = 0;
 	}
+
+	if (!ctx->streamon_out && !ctx->streamon_cap)
+		ctx->bit_stream_param &= ~CODA_BIT_STREAM_END_FLAG;
 }
 
 static const struct vb2_ops coda_qops = {
@@ -1469,6 +1478,12 @@ static int coda_s_ctrl(struct v4l2_ctrl *ctrl)
 	case V4L2_CID_JPEG_RESTART_INTERVAL:
 		ctx->params.jpeg_restart_interval = ctrl->val;
 		break;
+	case V4L2_CID_MPEG_VIDEO_VBV_DELAY:
+		ctx->params.vbv_delay = ctrl->val;
+		break;
+	case V4L2_CID_MPEG_VIDEO_VBV_SIZE:
+		ctx->params.vbv_size = min(ctrl->val * 8192, 0x7fffffff);
+		break;
 	default:
 		v4l2_dbg(1, coda_debug, &ctx->dev->v4l2_dev,
 			"Invalid control, id=%d, val=%d\n",
@@ -1528,6 +1543,14 @@ static void coda_encode_ctrls(struct coda_ctx *ctx)
 	v4l2_ctrl_new_std(&ctx->ctrls, &coda_ctrl_ops,
 		V4L2_CID_MPEG_VIDEO_CYCLIC_INTRA_REFRESH_MB, 0,
 		1920 * 1088 / 256, 1, 0);
+	v4l2_ctrl_new_std(&ctx->ctrls, &coda_ctrl_ops,
+		V4L2_CID_MPEG_VIDEO_VBV_DELAY, 0, 0x7fff, 1, 0);
+	/*
+	 * The maximum VBV size value is 0x7fffffff bits,
+	 * one bit less than 262144 KiB
+	 */
+	v4l2_ctrl_new_std(&ctx->ctrls, &coda_ctrl_ops,
+		V4L2_CID_MPEG_VIDEO_VBV_SIZE, 0, 262144, 1, 0);
 }
 
 static void coda_jpeg_encode_ctrls(struct coda_ctx *ctx)
@@ -1726,6 +1749,7 @@ static int coda_open(struct file *file)
 	mutex_init(&ctx->bitstream_mutex);
 	mutex_init(&ctx->buffer_mutex);
 	INIT_LIST_HEAD(&ctx->buffer_meta_list);
+	spin_lock_init(&ctx->buffer_meta_lock);
 
 	coda_lock(ctx);
 	list_add(&ctx->list, &dev->instances);
@@ -1769,7 +1793,7 @@ static int coda_release(struct file *file)
 	v4l2_m2m_ctx_release(ctx->fh.m2m_ctx);
 
 	/* In case the instance was not running, we still need to call SEQ_END */
-	if (ctx->initialized && ctx->ops->seq_end_work) {
+	if (ctx->ops->seq_end_work) {
 		queue_work(dev->workqueue, &ctx->seq_end_work);
 		flush_work(&ctx->seq_end_work);
 	}
@@ -2157,7 +2181,7 @@ static int coda_probe(struct platform_device *pdev)
 	/* Get IRAM pool from device tree or platform data */
 	pool = of_gen_pool_get(np, "iram", 0);
 	if (!pool && pdata)
-		pool = gen_pool_get(pdata->iram_dev);
+		pool = gen_pool_get(pdata->iram_dev, NULL);
 	if (!pool) {
 		dev_err(&pdev->dev, "iram pool not available\n");
 		return -ENOMEM;
diff --git a/drivers/media/platform/coda/coda-gdi.c b/drivers/media/platform/coda/coda-gdi.c
new file mode 100644
index 000000000..aaa7afc68
--- /dev/null
+++ b/drivers/media/platform/coda/coda-gdi.c
@@ -0,0 +1,150 @@
+/*
+ * Coda multi-standard codec IP
+ *
+ * Copyright (C) 2014 Philipp Zabel, Pengutronix
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <linux/bitops.h>
+#include "coda.h"
+
+#define XY2_INVERT	BIT(7)
+#define XY2_ZERO	BIT(6)
+#define XY2_TB_XOR	BIT(5)
+#define XY2_XYSEL	BIT(4)
+#define XY2_Y		(1 << 4)
+#define XY2_X		(0 << 4)
+
+#define XY2(luma_sel, luma_bit, chroma_sel, chroma_bit) \
+	(((XY2_##luma_sel) | (luma_bit)) << 8 | \
+	 (XY2_##chroma_sel) | (chroma_bit))
+
+static const u16 xy2ca_zero_map[16] = {
+	XY2(ZERO, 0, ZERO, 0),
+	XY2(ZERO, 0, ZERO, 0),
+	XY2(ZERO, 0, ZERO, 0),
+	XY2(ZERO, 0, ZERO, 0),
+	XY2(ZERO, 0, ZERO, 0),
+	XY2(ZERO, 0, ZERO, 0),
+	XY2(ZERO, 0, ZERO, 0),
+	XY2(ZERO, 0, ZERO, 0),
+	XY2(ZERO, 0, ZERO, 0),
+	XY2(ZERO, 0, ZERO, 0),
+	XY2(ZERO, 0, ZERO, 0),
+	XY2(ZERO, 0, ZERO, 0),
+	XY2(ZERO, 0, ZERO, 0),
+	XY2(ZERO, 0, ZERO, 0),
+	XY2(ZERO, 0, ZERO, 0),
+	XY2(ZERO, 0, ZERO, 0),
+};
+
+static const u16 xy2ca_tiled_map[16] = {
+	XY2(Y,    0, Y,    0),
+	XY2(Y,    1, Y,    1),
+	XY2(Y,    2, Y,    2),
+	XY2(Y,    3, X,    3),
+	XY2(X,    3, ZERO, 0),
+	XY2(ZERO, 0, ZERO, 0),
+	XY2(ZERO, 0, ZERO, 0),
+	XY2(ZERO, 0, ZERO, 0),
+	XY2(ZERO, 0, ZERO, 0),
+	XY2(ZERO, 0, ZERO, 0),
+	XY2(ZERO, 0, ZERO, 0),
+	XY2(ZERO, 0, ZERO, 0),
+	XY2(ZERO, 0, ZERO, 0),
+	XY2(ZERO, 0, ZERO, 0),
+	XY2(ZERO, 0, ZERO, 0),
+	XY2(ZERO, 0, ZERO, 0),
+};
+
+/*
+ * RA[15:0], CA[15:8] are hardwired to contain the 24-bit macroblock
+ * start offset (macroblock size is 16x16 for luma, 16x8 for chroma).
+ * Bits CA[4:0] are set using XY2CA above. BA[3:0] seems to be unused.
+ */
+
+#define RBC_CA		(0 << 4)
+#define RBC_BA		(1 << 4)
+#define RBC_RA		(2 << 4)
+#define RBC_ZERO	(3 << 4)
+
+#define RBC(luma_sel, luma_bit, chroma_sel, chroma_bit) \
+	(((RBC_##luma_sel) | (luma_bit)) << 6 | \
+	 (RBC_##chroma_sel) | (chroma_bit))
+
+static const u16 rbc2axi_tiled_map[32] = {
+	RBC(ZERO, 0, ZERO, 0),
+	RBC(ZERO, 0, ZERO, 0),
+	RBC(ZERO, 0, ZERO, 0),
+	RBC(CA,   0, CA,   0),
+	RBC(CA,   1, CA,   1),
+	RBC(CA,   2, CA,   2),
+	RBC(CA,   3, CA,   3),
+	RBC(CA,   4, CA,   8),
+	RBC(CA,   8, CA,   9),
+	RBC(CA,   9, CA,  10),
+	RBC(CA,  10, CA,  11),
+	RBC(CA,  11, CA,  12),
+	RBC(CA,  12, CA,  13),
+	RBC(CA,  13, CA,  14),
+	RBC(CA,  14, CA,  15),
+	RBC(CA,  15, RA,   0),
+	RBC(RA,   0, RA,   1),
+	RBC(RA,   1, RA,   2),
+	RBC(RA,   2, RA,   3),
+	RBC(RA,   3, RA,   4),
+	RBC(RA,   4, RA,   5),
+	RBC(RA,   5, RA,   6),
+	RBC(RA,   6, RA,   7),
+	RBC(RA,   7, RA,   8),
+	RBC(RA,   8, RA,   9),
+	RBC(RA,   9, RA,  10),
+	RBC(RA,  10, RA,  11),
+	RBC(RA,  11, RA,  12),
+	RBC(RA,  12, RA,  13),
+	RBC(RA,  13, RA,  14),
+	RBC(RA,  14, RA,  15),
+	RBC(RA,  15, ZERO, 0),
+};
+
+void coda_set_gdi_regs(struct coda_ctx *ctx)
+{
+	struct coda_dev *dev = ctx->dev;
+	const u16 *xy2ca_map;
+	u32 xy2rbc_config;
+	int i;
+
+	switch (ctx->tiled_map_type) {
+	case GDI_LINEAR_FRAME_MAP:
+	default:
+		xy2ca_map = xy2ca_zero_map;
+		xy2rbc_config = 0;
+		break;
+	case GDI_TILED_FRAME_MB_RASTER_MAP:
+		xy2ca_map = xy2ca_tiled_map;
+		xy2rbc_config = CODA9_XY2RBC_TILED_MAP |
+				CODA9_XY2RBC_CA_INC_HOR |
+				(16 - 1) << 12 | (8 - 1) << 4;
+		break;
+	}
+
+	for (i = 0; i < 16; i++)
+		coda_write(dev, xy2ca_map[i],
+				CODA9_GDI_XY2_CAS_0 + 4 * i);
+	for (i = 0; i < 4; i++)
+		coda_write(dev, XY2(ZERO, 0, ZERO, 0),
+				CODA9_GDI_XY2_BA_0 + 4 * i);
+	for (i = 0; i < 16; i++)
+		coda_write(dev, XY2(ZERO, 0, ZERO, 0),
+				CODA9_GDI_XY2_RAS_0 + 4 * i);
+	coda_write(dev, xy2rbc_config, CODA9_GDI_XY2_RBC_CONFIG);
+	if (xy2rbc_config) {
+		for (i = 0; i < 32; i++)
+			coda_write(dev, rbc2axi_tiled_map[i],
+					CODA9_GDI_RBC2_AXI_0 + 4 * i);
+	}
+}
diff --git a/drivers/media/platform/coda/coda.h b/drivers/media/platform/coda/coda.h
index 8e0af221b..59b2af9c7 100644
--- a/drivers/media/platform/coda/coda.h
+++ b/drivers/media/platform/coda/coda.h
@@ -128,6 +128,8 @@ struct coda_params {
 	enum v4l2_mpeg_video_multi_slice_mode slice_mode;
 	u32			framerate;
 	u16			bitrate;
+	u16			vbv_delay;
+	u32			vbv_size;
 	u32			slice_max_bits;
 	u32			slice_max_mb;
 };
@@ -165,15 +167,8 @@ struct coda_iram_info {
 	phys_addr_t	next_paddr;
 };
 
-struct gdi_tiled_map {
-	int xy2ca_map[16];
-	int xy2ba_map[16];
-	int xy2ra_map[16];
-	int rbc2axi_map[32];
-	int xy2rbc_config;
-	int map_type;
 #define GDI_LINEAR_FRAME_MAP 0
-};
+#define GDI_TILED_FRAME_MB_RASTER_MAP 1
 
 struct coda_ctx;
 
@@ -227,12 +222,14 @@ struct coda_ctx {
 	struct coda_buffer_meta		frame_metas[CODA_MAX_FRAMEBUFFERS];
 	u32				frame_errors[CODA_MAX_FRAMEBUFFERS];
 	struct list_head		buffer_meta_list;
+	spinlock_t			buffer_meta_lock;
+	int				num_metas;
 	struct coda_aux_buf		workbuf;
 	int				num_internal_frames;
 	int				idx;
 	int				reg_idx;
 	struct coda_iram_info		iram_info;
-	struct gdi_tiled_map		tiled_map;
+	int				tiled_map_type;
 	u32				bit_stream_param;
 	u32				frm_dis_flg;
 	u32				frame_mem_ctrl;
diff --git a/drivers/media/platform/coda/coda_regs.h b/drivers/media/platform/coda/coda_regs.h
index 7d0262411..3490602fa 100644
--- a/drivers/media/platform/coda/coda_regs.h
+++ b/drivers/media/platform/coda/coda_regs.h
@@ -51,6 +51,7 @@
 #define		CODA7_STREAM_SEL_64BITS_ENDIAN	(1 << 1)
 #define		CODA_STREAM_ENDIAN_SELECT	(1 << 0)
 #define CODA_REG_BIT_FRAME_MEM_CTRL		0x110
+#define		CODA9_FRAME_TILED2LINEAR	(1 << 11)
 #define		CODA_FRAME_CHROMA_INTERLEAVE	(1 << 2)
 #define		CODA_IMAGE_ENDIAN_SELECT	(1 << 0)
 #define CODA_REG_BIT_BIT_STREAM_PARAM		0x114
@@ -263,6 +264,10 @@
 #define		CODADX6_PICHEIGHT_MASK				0x3ff
 #define		CODA7_PICHEIGHT_MASK				0xffff
 #define CODA_CMD_ENC_SEQ_SRC_F_RATE				0x194
+#define		CODA_FRATE_RES_OFFSET				0
+#define		CODA_FRATE_RES_MASK				0xffff
+#define		CODA_FRATE_DIV_OFFSET				16
+#define		CODA_FRATE_DIV_MASK				0xffff
 #define CODA_CMD_ENC_SEQ_MP4_PARA				0x198
 #define		CODA_MP4PARAM_VERID_OFFSET			6
 #define		CODA_MP4PARAM_VERID_MASK			0x01
@@ -448,7 +453,12 @@
 #define CODA9_GDI_XY2_RAS_F			(CODA9_GDMA_BASE + 0x88c)
 
 #define CODA9_GDI_XY2_RBC_CONFIG		(CODA9_GDMA_BASE + 0x890)
+#define		CODA9_XY2RBC_SEPARATE_MAP		BIT(19)
+#define		CODA9_XY2RBC_TOP_BOT_SPLIT		BIT(18)
+#define		CODA9_XY2RBC_TILED_MAP			BIT(17)
+#define		CODA9_XY2RBC_CA_INC_HOR			BIT(16)
 #define CODA9_GDI_RBC2_AXI_0			(CODA9_GDMA_BASE + 0x8a0)
 #define CODA9_GDI_RBC2_AXI_1F			(CODA9_GDMA_BASE + 0x91c)
+#define	CODA9_GDI_TILEDBUF_BASE			(CODA9_GDMA_BASE + 0x920)
 
 #endif
diff --git a/drivers/media/platform/coda/trace.h b/drivers/media/platform/coda/trace.h
index 781bf7286..d9099a0f7 100644
--- a/drivers/media/platform/coda/trace.h
+++ b/drivers/media/platform/coda/trace.h
@@ -48,7 +48,7 @@ TRACE_EVENT(coda_bit_done,
 	TP_printk("minor = %d, ctx = %d", __entry->minor, __entry->ctx)
 );
 
-TRACE_EVENT(coda_enc_pic_run,
+DECLARE_EVENT_CLASS(coda_buf_class,
 	TP_PROTO(struct coda_ctx *ctx, struct vb2_buffer *buf),
 
 	TP_ARGS(ctx, buf),
@@ -69,28 +69,17 @@ TRACE_EVENT(coda_enc_pic_run,
 		  __entry->minor, __entry->index, __entry->ctx)
 );
 
-TRACE_EVENT(coda_enc_pic_done,
+DEFINE_EVENT(coda_buf_class, coda_enc_pic_run,
 	TP_PROTO(struct coda_ctx *ctx, struct vb2_buffer *buf),
+	TP_ARGS(ctx, buf)
+);
 
-	TP_ARGS(ctx, buf),
-
-	TP_STRUCT__entry(
-		__field(int, minor)
-		__field(int, index)
-		__field(int, ctx)
-	),
-
-	TP_fast_assign(
-		__entry->minor = ctx->fh.vdev->minor;
-		__entry->index = buf->v4l2_buf.index;
-		__entry->ctx = ctx->idx;
-	),
-
-	TP_printk("minor = %d, index = %d, ctx = %d",
-		  __entry->minor, __entry->index, __entry->ctx)
+DEFINE_EVENT(coda_buf_class, coda_enc_pic_done,
+	TP_PROTO(struct coda_ctx *ctx, struct vb2_buffer *buf),
+	TP_ARGS(ctx, buf)
 );
 
-TRACE_EVENT(coda_bit_queue,
+DECLARE_EVENT_CLASS(coda_buf_meta_class,
 	TP_PROTO(struct coda_ctx *ctx, struct vb2_buffer *buf,
 		 struct coda_buffer_meta *meta),
 
@@ -117,7 +106,13 @@ TRACE_EVENT(coda_bit_queue,
 		  __entry->ctx)
 );
 
-TRACE_EVENT(coda_dec_pic_run,
+DEFINE_EVENT(coda_buf_meta_class, coda_bit_queue,
+	TP_PROTO(struct coda_ctx *ctx, struct vb2_buffer *buf,
+		 struct coda_buffer_meta *meta),
+	TP_ARGS(ctx, buf, meta)
+);
+
+DECLARE_EVENT_CLASS(coda_meta_class,
 	TP_PROTO(struct coda_ctx *ctx, struct coda_buffer_meta *meta),
 
 	TP_ARGS(ctx, meta),
@@ -140,54 +135,20 @@ TRACE_EVENT(coda_dec_pic_run,
 		  __entry->minor, __entry->start, __entry->end, __entry->ctx)
 );
 
-TRACE_EVENT(coda_dec_pic_done,
+DEFINE_EVENT(coda_meta_class, coda_dec_pic_run,
 	TP_PROTO(struct coda_ctx *ctx, struct coda_buffer_meta *meta),
-
-	TP_ARGS(ctx, meta),
-
-	TP_STRUCT__entry(
-		__field(int, minor)
-		__field(int, start)
-		__field(int, end)
-		__field(int, ctx)
-	),
-
-	TP_fast_assign(
-		__entry->minor = ctx->fh.vdev->minor;
-		__entry->start = meta->start;
-		__entry->end = meta->end;
-		__entry->ctx = ctx->idx;
-	),
-
-	TP_printk("minor = %d, start = 0x%x, end = 0x%x, ctx = %d",
-		  __entry->minor, __entry->start, __entry->end, __entry->ctx)
+	TP_ARGS(ctx, meta)
 );
 
-TRACE_EVENT(coda_dec_rot_done,
-	TP_PROTO(struct coda_ctx *ctx, struct coda_buffer_meta *meta,
-		 struct vb2_buffer *buf),
-
-	TP_ARGS(ctx, meta, buf),
-
-	TP_STRUCT__entry(
-		__field(int, minor)
-		__field(int, start)
-		__field(int, end)
-		__field(int, index)
-		__field(int, ctx)
-	),
-
-	TP_fast_assign(
-		__entry->minor = ctx->fh.vdev->minor;
-		__entry->start = meta->start;
-		__entry->end = meta->end;
-		__entry->index = buf->v4l2_buf.index;
-		__entry->ctx = ctx->idx;
-	),
+DEFINE_EVENT(coda_meta_class, coda_dec_pic_done,
+	TP_PROTO(struct coda_ctx *ctx, struct coda_buffer_meta *meta),
+	TP_ARGS(ctx, meta)
+);
 
-	TP_printk("minor = %d, start = 0x%x, end = 0x%x, index = %d, ctx = %d",
-		  __entry->minor, __entry->start, __entry->end, __entry->index,
-		  __entry->ctx)
+DEFINE_EVENT(coda_buf_meta_class, coda_dec_rot_done,
+	TP_PROTO(struct coda_ctx *ctx, struct vb2_buffer *buf,
+		 struct coda_buffer_meta *meta),
+	TP_ARGS(ctx, buf, meta)
 );
 
 #endif /* __CODA_TRACE_H__ */