[PATCH v2 3/4] dmaengine: sun6i: Add support for 34-bit physical addresses

[Samuel Holland]

Recent Allwinner SoCs support >4 GiB of DRAM, so those variants of the
DMA engine support >32 bit physical addresses. This is accomplished by
placing the high bits in the "para" word in the DMA descriptor.

DMA descriptors themselves can be located at >32 bit addresses by
putting the high bits in the LSBs of the descriptor address register,
taking advantage of the required DMA descriptor alignment. However,
support for this is not really necessary, so we can avoid the
complication by allocating them from the DMA_32 zone.

Acked-by: Maxime Ripard <maxime@xxxxxxxxxx>
Signed-off-by: Samuel Holland <samuel@xxxxxxxxxxxx>
---

Changes in v2:
 - Fix `checkpatch.pl --strict` style issues (missing spaces)

 drivers/dma/sun6i-dma.c | 39 ++++++++++++++++++++++++++++++++++-----
 1 file changed, 34 insertions(+), 5 deletions(-)

diff --git a/drivers/dma/sun6i-dma.c b/drivers/dma/sun6i-dma.c
index a9334f969b28..bd5958185ed1 100644
--- a/drivers/dma/sun6i-dma.c
+++ b/drivers/dma/sun6i-dma.c
@@ -90,6 +90,14 @@
 
 #define DMA_CHAN_CUR_PARA	0x1c
 
+/*
+ * LLI address mangling
+ *
+ * The LLI link physical address is also mangled, but we avoid dealing
+ * with that by allocating LLIs from the DMA32 zone.
+ */
+#define SET_SRC_HIGH_ADDR(x)		((((x) >> 32) & 0x3U) << 16)
+#define SET_DST_HIGH_ADDR(x)		((((x) >> 32) & 0x3U) << 18)
 
 /*
  * Various hardware related defines
@@ -132,6 +140,7 @@ struct sun6i_dma_config {
 	u32 dst_burst_lengths;
 	u32 src_addr_widths;
 	u32 dst_addr_widths;
+	bool has_high_addr;
 	bool has_mbus_clk;
 };
 
@@ -223,6 +232,12 @@ to_sun6i_desc(struct dma_async_tx_descriptor *tx)
 	return container_of(tx, struct sun6i_desc, vd.tx);
 }
 
+static inline bool sun6i_dma_has_high_addr(struct sun6i_dma_dev *sdev)
+{
+	return IS_ENABLED(CONFIG_ARCH_DMA_ADDR_T_64BIT) &&
+		sdev->cfg->has_high_addr;
+}
+
 static inline void sun6i_dma_dump_com_regs(struct sun6i_dma_dev *sdev)
 {
 	dev_dbg(sdev->slave.dev, "Common register:\n"
@@ -645,7 +660,7 @@ static struct dma_async_tx_descriptor *sun6i_dma_prep_dma_memcpy(
 	if (!txd)
 		return NULL;
 
-	v_lli = dma_pool_alloc(sdev->pool, GFP_NOWAIT, &p_lli);
+	v_lli = dma_pool_alloc(sdev->pool, GFP_DMA32 | GFP_NOWAIT, &p_lli);
 	if (!v_lli) {
 		dev_err(sdev->slave.dev, "Failed to alloc lli memory\n");
 		goto err_txd_free;
@@ -655,6 +670,9 @@ static struct dma_async_tx_descriptor *sun6i_dma_prep_dma_memcpy(
 	v_lli->dst = dest;
 	v_lli->len = len;
 	v_lli->para = NORMAL_WAIT;
+	if (sun6i_dma_has_high_addr(sdev))
+		v_lli->para |= SET_SRC_HIGH_ADDR(src) |
+			       SET_DST_HIGH_ADDR(dest);
 
 	burst = convert_burst(8);
 	width = convert_buswidth(DMA_SLAVE_BUSWIDTH_4_BYTES);
@@ -705,7 +723,7 @@ static struct dma_async_tx_descriptor *sun6i_dma_prep_slave_sg(
 		return NULL;
 
 	for_each_sg(sgl, sg, sg_len, i) {
-		v_lli = dma_pool_alloc(sdev->pool, GFP_NOWAIT, &p_lli);
+		v_lli = dma_pool_alloc(sdev->pool, GFP_DMA32 | GFP_NOWAIT, &p_lli);
 		if (!v_lli)
 			goto err_lli_free;
 
@@ -715,6 +733,9 @@ static struct dma_async_tx_descriptor *sun6i_dma_prep_slave_sg(
 		if (dir == DMA_MEM_TO_DEV) {
 			v_lli->src = sg_dma_address(sg);
 			v_lli->dst = sconfig->dst_addr;
+			if (sun6i_dma_has_high_addr(sdev))
+				v_lli->para |= SET_SRC_HIGH_ADDR(sg_dma_address(sg)) |
+					       SET_DST_HIGH_ADDR(sconfig->dst_addr);
 			v_lli->cfg = lli_cfg;
 			sdev->cfg->set_drq(&v_lli->cfg, DRQ_SDRAM, vchan->port);
 			sdev->cfg->set_mode(&v_lli->cfg, LINEAR_MODE, IO_MODE);
@@ -728,6 +749,9 @@ static struct dma_async_tx_descriptor *sun6i_dma_prep_slave_sg(
 		} else {
 			v_lli->src = sconfig->src_addr;
 			v_lli->dst = sg_dma_address(sg);
+			if (sun6i_dma_has_high_addr(sdev))
+				v_lli->para |= SET_SRC_HIGH_ADDR(sconfig->src_addr) |
+					       SET_DST_HIGH_ADDR(sg_dma_address(sg));
 			v_lli->cfg = lli_cfg;
 			sdev->cfg->set_drq(&v_lli->cfg, vchan->port, DRQ_SDRAM);
 			sdev->cfg->set_mode(&v_lli->cfg, IO_MODE, LINEAR_MODE);
@@ -786,7 +810,7 @@ static struct dma_async_tx_descriptor *sun6i_dma_prep_dma_cyclic(
 		return NULL;
 
 	for (i = 0; i < periods; i++) {
-		v_lli = dma_pool_alloc(sdev->pool, GFP_NOWAIT, &p_lli);
+		v_lli = dma_pool_alloc(sdev->pool, GFP_DMA32 | GFP_NOWAIT, &p_lli);
 		if (!v_lli) {
 			dev_err(sdev->slave.dev, "Failed to alloc lli memory\n");
 			goto err_lli_free;
@@ -798,12 +822,18 @@ static struct dma_async_tx_descriptor *sun6i_dma_prep_dma_cyclic(
 		if (dir == DMA_MEM_TO_DEV) {
 			v_lli->src = buf_addr + period_len * i;
 			v_lli->dst = sconfig->dst_addr;
+			if (sun6i_dma_has_high_addr(sdev))
+				v_lli->para |= SET_SRC_HIGH_ADDR(buf_addr + period_len * i) |
+					       SET_DST_HIGH_ADDR(sconfig->dst_addr);
 			v_lli->cfg = lli_cfg;
 			sdev->cfg->set_drq(&v_lli->cfg, DRQ_SDRAM, vchan->port);
 			sdev->cfg->set_mode(&v_lli->cfg, LINEAR_MODE, IO_MODE);
 		} else {
 			v_lli->src = sconfig->src_addr;
 			v_lli->dst = buf_addr + period_len * i;
+			if (sun6i_dma_has_high_addr(sdev))
+				v_lli->para |= SET_SRC_HIGH_ADDR(sconfig->src_addr) |
+					       SET_DST_HIGH_ADDR(buf_addr + period_len * i);
 			v_lli->cfg = lli_cfg;
 			sdev->cfg->set_drq(&v_lli->cfg, vchan->port, DRQ_SDRAM);
 			sdev->cfg->set_mode(&v_lli->cfg, IO_MODE, LINEAR_MODE);
@@ -1174,8 +1204,6 @@ static struct sun6i_dma_config sun50i_a64_dma_cfg = {
 };
 
 /*
- * TODO: Add support for more than 4g physical addressing.
- *
  * The A100 binding uses the number of dma channels from the
  * device tree node.
  */
@@ -1194,6 +1222,7 @@ static struct sun6i_dma_config sun50i_a100_dma_cfg = {
 			     BIT(DMA_SLAVE_BUSWIDTH_2_BYTES) |
 			     BIT(DMA_SLAVE_BUSWIDTH_4_BYTES) |
 			     BIT(DMA_SLAVE_BUSWIDTH_8_BYTES),
+	.has_high_addr = true,
 	.has_mbus_clk = true,
 };
 
-- 
2.35.1