dmaengine: sun6i: Fixes

PATCH Superseded Jean-Francois Moine 1 participant 3 messages Mar 18, 2016

Subsystems (2)

Arm/Allwinner sunXi SoC support dmaengine

lore

Share & tools

lore

Jean-Francois Moine

author

2016-03-18 08:52 UTC #0 lore raw

v3 cover dmaengine: sun6i: Fixes

This patch series replaces part of the previous series 'dmaengine: sun6i: Fixes and upgrade for audio transfers'. It contains only fixes for a normal use of the DMA driver.

Changes since v2
- Acked-by Maxime Ripard added
- remove the audio requirements
Changes since v1 (comments from Vinod Koul and Sergei Shtylyov - thanks)
- typo fixes
- change some comments
- better handling of burst and bus width
- remove useless code in the cyclic capability patch

Jean-Francois Moine (2):
dmaengine: sun6i: Fix the access of the IRQ register
dmaengine: sun6i: Fix impossible settings of burst and bus width

 drivers/dma/sun6i-dma.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

Signature · 1 line

2.7.4

Jean-Francois Moine

author

2016-03-18 08:48 UTC #1 lore raw

PATCH v3 1/2 dmaengine: sun6i: Fix the access of the IRQ register

The IRQ register number is computed, but this number was not used and the register was the one indexed by the channel index instead. Then, only the first DMA channel was working.

Acked-by Maxime Ripard <[email protected]>

Signed-off-by Jean-Francois Moine <[email protected]>

---
 drivers/dma/sun6i-dma.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

drivers/dma/sun6i-dma.c +2 -2

index 2db12e4..2ec320d 100644
@@ -381,9 +381,9 @@ static int sun6i_dma_start_desc(struct sun6i_vchan *vchan)
 	irq_reg = pchan->idx / DMA_IRQ_CHAN_NR;
 	irq_offset = pchan->idx % DMA_IRQ_CHAN_NR;
 
-	irq_val = readl(sdev->base + DMA_IRQ_EN(irq_offset));
+	irq_val = readl(sdev->base + DMA_IRQ_EN(irq_reg));
 	irq_val |= DMA_IRQ_QUEUE << (irq_offset * DMA_IRQ_CHAN_WIDTH);
-	writel(irq_val, sdev->base + DMA_IRQ_EN(irq_offset));
+	writel(irq_val, sdev->base + DMA_IRQ_EN(irq_reg));
 
 	writel(pchan->desc->p_lli, pchan->base + DMA_CHAN_LLI_ADDR);
 	writel(DMA_CHAN_ENABLE_START, pchan->base + DMA_CHAN_ENABLE);

Signature · 1 line

2.7.4

Jean-Francois Moine

author

2016-03-18 08:50 UTC #2 lore raw

PATCH v3 2/2 dmaengine: sun6i: Fix impossible settings of burst and bus width

In reply to Jean-Francois Moine

In the commit 1f9cd915b64bb95f (dmaengine: sun6i: Fix memcpy operation), the signed values returned by convert_burst() and convert_buswidth() were stored in an unsigned value.
Then, these values were considered as errors when non null.

As a result, DMA transfers were rejected when the burst or buswidth had values different from 1, as 8 for the burst or 4 for the bus width.

Acked-by Maxime Ripard <[email protected]>

Signed-off-by Jean-Francois Moine <[email protected]>

---
 drivers/dma/sun6i-dma.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

drivers/dma/sun6i-dma.c +5 -5

index 2ec320d..3579ee7 100644
@@ -281,25 +281,25 @@ static inline int sun6i_dma_cfg_lli(struct sun6i_dma_lli *lli,
 				    dma_addr_t dst, u32 len,
 				    struct dma_slave_config *config)
 {
-	u8 src_width, dst_width, src_burst, dst_burst;
+	s8 src_width, dst_width, src_burst, dst_burst;
 
 	if (!config)
 		return -EINVAL;
 
 	src_burst = convert_burst(config->src_maxburst);
-	if (src_burst)
+	if (src_burst < 0)
 		return src_burst;
 
 	dst_burst = convert_burst(config->dst_maxburst);
-	if (dst_burst)
+	if (dst_burst < 0)
 		return dst_burst;
 
 	src_width = convert_buswidth(config->src_addr_width);
-	if (src_width)
+	if (src_width < 0)
 		return src_width;
 
 	dst_width = convert_buswidth(config->dst_addr_width);
-	if (dst_width)
+	if (dst_width < 0)
 		return dst_width;
 
 	lli->cfg = DMA_CHAN_CFG_SRC_BURST(src_burst) |

Signature · 1 line

2.7.4