* [PATCH v2 0/3] [media] atmel-isi: Add DT support for Atmel ISI driver
@ 2014-03-25 10:41 Josh Wu
2014-03-25 10:41 ` [PATCH v2 1/3] [media] atmel-isi: add v4l2 async probe support Josh Wu
` (2 more replies)
0 siblings, 3 replies; 13+ messages in thread
From: Josh Wu @ 2014-03-25 10:41 UTC (permalink / raw)
To: g.liakhovetski
Cc: linux-media, m.chehab, nicolas.ferre, linux-arm-kernel,
laurent.pinchart, josh.wu
This patch series add DT support for atmel ISI driver. It can support the
common v4l2 DT interfaces.
v1 --> v2:
modified the device tree binding document to remove the a optional property.
Josh Wu (3):
[media] atmel-isi: add v4l2 async probe support
[media] atmel-isi: convert the pdata from pointer to structure
[media] atmel-isi: add primary DT support
.../devicetree/bindings/media/atmel-isi.txt | 50 +++++++++++++++++
drivers/media/platform/soc_camera/atmel-isi.c | 56 +++++++++++++++-----
include/media/atmel-isi.h | 4 ++
3 files changed, 98 insertions(+), 12 deletions(-)
create mode 100644 Documentation/devicetree/bindings/media/atmel-isi.txt
--
1.7.9.5
^ permalink raw reply [flat|nested] 13+ messages in thread
* [PATCH v2 1/3] [media] atmel-isi: add v4l2 async probe support
2014-03-25 10:41 [PATCH v2 0/3] [media] atmel-isi: Add DT support for Atmel ISI driver Josh Wu
@ 2014-03-25 10:41 ` Josh Wu
2014-03-25 10:41 ` [PATCH v2 2/3] [media] atmel-isi: convert the pdata from pointer to structure Josh Wu
2014-03-25 10:45 ` [PATCH v2 3/3] [media] atmel-isi: add primary DT support Josh Wu
2 siblings, 0 replies; 13+ messages in thread
From: Josh Wu @ 2014-03-25 10:41 UTC (permalink / raw)
To: g.liakhovetski
Cc: linux-media, m.chehab, nicolas.ferre, linux-arm-kernel,
laurent.pinchart, josh.wu
Signed-off-by: Josh Wu <josh.wu@atmel.com>
---
v1 --> v2:
no change.
drivers/media/platform/soc_camera/atmel-isi.c | 5 +++++
include/media/atmel-isi.h | 4 ++++
2 files changed, 9 insertions(+)
diff --git a/drivers/media/platform/soc_camera/atmel-isi.c b/drivers/media/platform/soc_camera/atmel-isi.c
index a1a95fe..9d977c5 100644
--- a/drivers/media/platform/soc_camera/atmel-isi.c
+++ b/drivers/media/platform/soc_camera/atmel-isi.c
@@ -989,6 +989,11 @@ static int atmel_isi_probe(struct platform_device *pdev)
soc_host->v4l2_dev.dev = &pdev->dev;
soc_host->nr = pdev->id;
+ if (isi->pdata.asd_sizes) {
+ soc_host->asd = isi->pdata.asd;
+ soc_host->asd_sizes = isi->pdata.asd_sizes;
+ }
+
ret = soc_camera_host_register(soc_host);
if (ret) {
dev_err(&pdev->dev, "Unable to register soc camera host\n");
diff --git a/include/media/atmel-isi.h b/include/media/atmel-isi.h
index 2b02347..c2e5703 100644
--- a/include/media/atmel-isi.h
+++ b/include/media/atmel-isi.h
@@ -106,6 +106,8 @@
#define ISI_DATAWIDTH_8 0x01
#define ISI_DATAWIDTH_10 0x02
+struct v4l2_async_subdev;
+
struct isi_platform_data {
u8 has_emb_sync;
u8 emb_crc_sync;
@@ -118,6 +120,8 @@ struct isi_platform_data {
u32 frate;
/* Using for ISI_MCK */
u32 mck_hz;
+ struct v4l2_async_subdev **asd; /* Flat array, arranged in groups */
+ int *asd_sizes; /* 0-terminated array of asd group sizes */
};
#endif /* __ATMEL_ISI_H__ */
--
1.7.9.5
^ permalink raw reply related [flat|nested] 13+ messages in thread
* [PATCH v2 2/3] [media] atmel-isi: convert the pdata from pointer to structure
2014-03-25 10:41 [PATCH v2 0/3] [media] atmel-isi: Add DT support for Atmel ISI driver Josh Wu
2014-03-25 10:41 ` [PATCH v2 1/3] [media] atmel-isi: add v4l2 async probe support Josh Wu
@ 2014-03-25 10:41 ` Josh Wu
2014-05-18 20:59 ` Guennadi Liakhovetski
2014-03-25 10:45 ` [PATCH v2 3/3] [media] atmel-isi: add primary DT support Josh Wu
2 siblings, 1 reply; 13+ messages in thread
From: Josh Wu @ 2014-03-25 10:41 UTC (permalink / raw)
To: g.liakhovetski
Cc: linux-media, m.chehab, nicolas.ferre, linux-arm-kernel,
laurent.pinchart, josh.wu
Now the platform data is initialized by allocation of isi
structure. In the future, we use pdata to store the dt parameters.
Signed-off-by: Josh Wu <josh.wu@atmel.com>
---
v1 --> v2:
no change.
drivers/media/platform/soc_camera/atmel-isi.c | 22 +++++++++++-----------
1 file changed, 11 insertions(+), 11 deletions(-)
diff --git a/drivers/media/platform/soc_camera/atmel-isi.c b/drivers/media/platform/soc_camera/atmel-isi.c
index 9d977c5..f4add0a 100644
--- a/drivers/media/platform/soc_camera/atmel-isi.c
+++ b/drivers/media/platform/soc_camera/atmel-isi.c
@@ -84,7 +84,7 @@ struct atmel_isi {
struct clk *mck;
unsigned int irq;
- struct isi_platform_data *pdata;
+ struct isi_platform_data pdata;
u16 width_flags; /* max 12 bits */
struct list_head video_buffer_list;
@@ -350,7 +350,7 @@ static void start_dma(struct atmel_isi *isi, struct frame_buffer *buffer)
cfg1 &= ~ISI_CFG1_FRATE_DIV_MASK;
/* Enable linked list */
- cfg1 |= isi->pdata->frate | ISI_CFG1_DISCR;
+ cfg1 |= isi->pdata.frate | ISI_CFG1_DISCR;
/* Enable codec path and ISI */
ctrl = ISI_CTRL_CDC | ISI_CTRL_EN;
@@ -797,7 +797,7 @@ static int isi_camera_set_bus_param(struct soc_camera_device *icd)
/* Make choises, based on platform preferences */
if ((common_flags & V4L2_MBUS_HSYNC_ACTIVE_HIGH) &&
(common_flags & V4L2_MBUS_HSYNC_ACTIVE_LOW)) {
- if (isi->pdata->hsync_act_low)
+ if (isi->pdata.hsync_act_low)
common_flags &= ~V4L2_MBUS_HSYNC_ACTIVE_HIGH;
else
common_flags &= ~V4L2_MBUS_HSYNC_ACTIVE_LOW;
@@ -805,7 +805,7 @@ static int isi_camera_set_bus_param(struct soc_camera_device *icd)
if ((common_flags & V4L2_MBUS_VSYNC_ACTIVE_HIGH) &&
(common_flags & V4L2_MBUS_VSYNC_ACTIVE_LOW)) {
- if (isi->pdata->vsync_act_low)
+ if (isi->pdata.vsync_act_low)
common_flags &= ~V4L2_MBUS_VSYNC_ACTIVE_HIGH;
else
common_flags &= ~V4L2_MBUS_VSYNC_ACTIVE_LOW;
@@ -813,7 +813,7 @@ static int isi_camera_set_bus_param(struct soc_camera_device *icd)
if ((common_flags & V4L2_MBUS_PCLK_SAMPLE_RISING) &&
(common_flags & V4L2_MBUS_PCLK_SAMPLE_FALLING)) {
- if (isi->pdata->pclk_act_falling)
+ if (isi->pdata.pclk_act_falling)
common_flags &= ~V4L2_MBUS_PCLK_SAMPLE_RISING;
else
common_flags &= ~V4L2_MBUS_PCLK_SAMPLE_FALLING;
@@ -835,9 +835,9 @@ static int isi_camera_set_bus_param(struct soc_camera_device *icd)
if (common_flags & V4L2_MBUS_PCLK_SAMPLE_FALLING)
cfg1 |= ISI_CFG1_PIXCLK_POL_ACTIVE_FALLING;
- if (isi->pdata->has_emb_sync)
+ if (isi->pdata.has_emb_sync)
cfg1 |= ISI_CFG1_EMB_SYNC;
- if (isi->pdata->full_mode)
+ if (isi->pdata.full_mode)
cfg1 |= ISI_CFG1_FULL_MODE;
isi_writel(isi, ISI_CTRL, ISI_CTRL_DIS);
@@ -912,7 +912,7 @@ static int atmel_isi_probe(struct platform_device *pdev)
if (IS_ERR(isi->pclk))
return PTR_ERR(isi->pclk);
- isi->pdata = pdata;
+ memcpy(&isi->pdata, pdata, sizeof(struct isi_platform_data));
isi->active = NULL;
spin_lock_init(&isi->lock);
INIT_LIST_HEAD(&isi->video_buffer_list);
@@ -928,7 +928,7 @@ static int atmel_isi_probe(struct platform_device *pdev)
/* Set ISI_MCK's frequency, it should be faster than pixel
* clock.
*/
- ret = clk_set_rate(isi->mck, pdata->mck_hz);
+ ret = clk_set_rate(isi->mck, isi->pdata.mck_hz);
if (ret < 0)
return ret;
}
@@ -962,9 +962,9 @@ static int atmel_isi_probe(struct platform_device *pdev)
goto err_ioremap;
}
- if (pdata->data_width_flags & ISI_DATAWIDTH_8)
+ if (isi->pdata.data_width_flags & ISI_DATAWIDTH_8)
isi->width_flags = 1 << 7;
- if (pdata->data_width_flags & ISI_DATAWIDTH_10)
+ if (isi->pdata.data_width_flags & ISI_DATAWIDTH_10)
isi->width_flags |= 1 << 9;
isi_writel(isi, ISI_CTRL, ISI_CTRL_DIS);
--
1.7.9.5
^ permalink raw reply related [flat|nested] 13+ messages in thread
* [PATCH v2 3/3] [media] atmel-isi: add primary DT support
2014-03-25 10:41 [PATCH v2 0/3] [media] atmel-isi: Add DT support for Atmel ISI driver Josh Wu
2014-03-25 10:41 ` [PATCH v2 1/3] [media] atmel-isi: add v4l2 async probe support Josh Wu
2014-03-25 10:41 ` [PATCH v2 2/3] [media] atmel-isi: convert the pdata from pointer to structure Josh Wu
@ 2014-03-25 10:45 ` Josh Wu
2014-03-26 16:12 ` Laurent Pinchart
` (2 more replies)
2 siblings, 3 replies; 13+ messages in thread
From: Josh Wu @ 2014-03-25 10:45 UTC (permalink / raw)
To: g.liakhovetski
Cc: linux-media, m.chehab, nicolas.ferre, linux-arm-kernel,
grant.likely, galak, rob, mark.rutland, robh+dt, ijc+devicetree,
pawel.moll, devicetree, laurent.pinchart, Josh Wu
This patch add the DT support for Atmel ISI driver.
It use the same v4l2 DT interface that defined in video-interfaces.txt.
Signed-off-by: Josh Wu <josh.wu@atmel.com>
Cc: devicetree@vger.kernel.org
---
v1 --> v2:
refine the binding document.
add port node description.
removed the optional property.
.../devicetree/bindings/media/atmel-isi.txt | 50 ++++++++++++++++++++
drivers/media/platform/soc_camera/atmel-isi.c | 31 +++++++++++-
2 files changed, 79 insertions(+), 2 deletions(-)
create mode 100644 Documentation/devicetree/bindings/media/atmel-isi.txt
diff --git a/Documentation/devicetree/bindings/media/atmel-isi.txt b/Documentation/devicetree/bindings/media/atmel-isi.txt
new file mode 100644
index 0000000..11c98ee
--- /dev/null
+++ b/Documentation/devicetree/bindings/media/atmel-isi.txt
@@ -0,0 +1,50 @@
+Atmel Image Sensor Interface (ISI) SoC Camera Subsystem
+----------------------------------------------
+
+Required properties:
+- compatible: must be "atmel,at91sam9g45-isi"
+- reg: physical base address and length of the registers set for the device;
+- interrupts: should contain IRQ line for the ISI;
+- clocks: list of clock specifiers, corresponding to entries in
+ the clock-names property;
+- clock-names: must contain "isi_clk", which is the isi peripherial clock.
+
+ISI supports a single port node with parallel bus. It should contain one
+'port' child node with child 'endpoint' node. Please refer to the bindings
+defined in Documentation/devicetree/bindings/media/video-interfaces.txt.
+
+Example:
+ isi: isi@f0034000 {
+ compatible = "atmel,at91sam9g45-isi";
+ reg = <0xf0034000 0x4000>;
+ interrupts = <37 IRQ_TYPE_LEVEL_HIGH 5>;
+
+ clocks = <&isi_clk>;
+ clock-names = "isi_clk";
+
+ pinctrl-names = "default";
+ pinctrl-0 = <&pinctrl_isi>;
+
+ port {
+ #address-cells = <1>;
+ #size-cells = <0>;
+
+ isi_0: endpoint {
+ remote-endpoint = <&ov2640_0>;
+ };
+ };
+ };
+
+ i2c1: i2c@f0018000 {
+ ov2640: camera@0x30 {
+ compatible = "omnivision,ov2640";
+ reg = <0x30>;
+
+ port {
+ ov2640_0: endpoint {
+ remote-endpoint = <&isi_0>;
+ bus-width = <8>;
+ };
+ };
+ };
+ };
diff --git a/drivers/media/platform/soc_camera/atmel-isi.c b/drivers/media/platform/soc_camera/atmel-isi.c
index f4add0a..d6a1f7b 100644
--- a/drivers/media/platform/soc_camera/atmel-isi.c
+++ b/drivers/media/platform/soc_camera/atmel-isi.c
@@ -19,6 +19,7 @@
#include <linux/interrupt.h>
#include <linux/kernel.h>
#include <linux/module.h>
+#include <linux/of.h>
#include <linux/platform_device.h>
#include <linux/slab.h>
@@ -33,6 +34,7 @@
#define VID_LIMIT_BYTES (16 * 1024 * 1024)
#define MIN_FRAME_RATE 15
#define FRAME_INTERVAL_MILLI_SEC (1000 / MIN_FRAME_RATE)
+#define ISI_DEFAULT_MCLK_FREQ 25000000
/* Frame buffer descriptor */
struct fbd {
@@ -885,6 +887,20 @@ static int atmel_isi_remove(struct platform_device *pdev)
return 0;
}
+static int atmel_isi_probe_dt(struct atmel_isi *isi,
+ struct platform_device *pdev)
+{
+ struct device_node *node = pdev->dev.of_node;
+
+ /* Default settings for ISI */
+ isi->pdata.full_mode = 1;
+ isi->pdata.mck_hz = ISI_DEFAULT_MCLK_FREQ;
+ isi->pdata.frate = ISI_CFG1_FRATE_CAPTURE_ALL;
+ isi->pdata.data_width_flags = ISI_DATAWIDTH_8 | ISI_DATAWIDTH_10;
+
+ return 0;
+}
+
static int atmel_isi_probe(struct platform_device *pdev)
{
unsigned int irq;
@@ -896,7 +912,7 @@ static int atmel_isi_probe(struct platform_device *pdev)
struct isi_platform_data *pdata;
pdata = dev->platform_data;
- if (!pdata || !pdata->data_width_flags) {
+ if ((!pdata || !pdata->data_width_flags) && !pdev->dev.of_node) {
dev_err(&pdev->dev,
"No config available for Atmel ISI\n");
return -EINVAL;
@@ -912,7 +928,11 @@ static int atmel_isi_probe(struct platform_device *pdev)
if (IS_ERR(isi->pclk))
return PTR_ERR(isi->pclk);
- memcpy(&isi->pdata, pdata, sizeof(struct isi_platform_data));
+ if (pdata)
+ memcpy(&isi->pdata, pdata, sizeof(struct isi_platform_data));
+ else /* dt probe */
+ atmel_isi_probe_dt(isi, pdev);
+
isi->active = NULL;
spin_lock_init(&isi->lock);
INIT_LIST_HEAD(&isi->video_buffer_list);
@@ -1014,11 +1034,18 @@ err_alloc_ctx:
return ret;
}
+static const struct of_device_id atmel_isi_of_match[] = {
+ { .compatible = "atmel,at91sam9g45-isi" },
+ { }
+};
+MODULE_DEVICE_TABLE(of, atmel_isi_of_match);
+
static struct platform_driver atmel_isi_driver = {
.remove = atmel_isi_remove,
.driver = {
.name = "atmel_isi",
.owner = THIS_MODULE,
+ .of_match_table = atmel_isi_of_match,
},
};
--
1.7.9.5
^ permalink raw reply related [flat|nested] 13+ messages in thread
* Re: [PATCH v2 3/3] [media] atmel-isi: add primary DT support
2014-03-25 10:45 ` [PATCH v2 3/3] [media] atmel-isi: add primary DT support Josh Wu
@ 2014-03-26 16:12 ` Laurent Pinchart
2014-03-30 21:20 ` Guennadi Liakhovetski
2014-05-18 21:51 ` Guennadi Liakhovetski
2 siblings, 0 replies; 13+ messages in thread
From: Laurent Pinchart @ 2014-03-26 16:12 UTC (permalink / raw)
To: Josh Wu
Cc: g.liakhovetski, linux-media, m.chehab, nicolas.ferre,
linux-arm-kernel, grant.likely, galak, rob, mark.rutland,
robh+dt, ijc+devicetree, pawel.moll, devicetree
Hi Josh,
Thank you for the patch.
On Tuesday 25 March 2014 18:45:20 Josh Wu wrote:
> This patch add the DT support for Atmel ISI driver.
> It use the same v4l2 DT interface that defined in video-interfaces.txt.
>
> Signed-off-by: Josh Wu <josh.wu@atmel.com>
> Cc: devicetree@vger.kernel.org
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
> ---
> v1 --> v2:
> refine the binding document.
> add port node description.
> removed the optional property.
>
> .../devicetree/bindings/media/atmel-isi.txt | 50 +++++++++++++++++
> drivers/media/platform/soc_camera/atmel-isi.c | 31 +++++++++++-
> 2 files changed, 79 insertions(+), 2 deletions(-)
> create mode 100644 Documentation/devicetree/bindings/media/atmel-isi.txt
>
> diff --git a/Documentation/devicetree/bindings/media/atmel-isi.txt
> b/Documentation/devicetree/bindings/media/atmel-isi.txt new file mode
> 100644
> index 0000000..11c98ee
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/media/atmel-isi.txt
> @@ -0,0 +1,50 @@
> +Atmel Image Sensor Interface (ISI) SoC Camera Subsystem
> +----------------------------------------------
> +
> +Required properties:
> +- compatible: must be "atmel,at91sam9g45-isi"
> +- reg: physical base address and length of the registers set for the
> device;
> +- interrupts: should contain IRQ line for the ISI;
> +- clocks: list of clock specifiers, corresponding to entries in
> + the clock-names property;
> +- clock-names: must contain "isi_clk", which is the isi peripherial clock.
> +
> +ISI supports a single port node with parallel bus. It should contain one
> +'port' child node with child 'endpoint' node. Please refer to the bindings
> +defined in Documentation/devicetree/bindings/media/video-interfaces.txt.
> +
> +Example:
> + isi: isi@f0034000 {
> + compatible = "atmel,at91sam9g45-isi";
> + reg = <0xf0034000 0x4000>;
> + interrupts = <37 IRQ_TYPE_LEVEL_HIGH 5>;
> +
> + clocks = <&isi_clk>;
> + clock-names = "isi_clk";
> +
> + pinctrl-names = "default";
> + pinctrl-0 = <&pinctrl_isi>;
> +
> + port {
> + #address-cells = <1>;
> + #size-cells = <0>;
> +
> + isi_0: endpoint {
> + remote-endpoint = <&ov2640_0>;
> + };
> + };
> + };
> +
> + i2c1: i2c@f0018000 {
> + ov2640: camera@0x30 {
> + compatible = "omnivision,ov2640";
> + reg = <0x30>;
> +
> + port {
> + ov2640_0: endpoint {
> + remote-endpoint = <&isi_0>;
> + bus-width = <8>;
> + };
> + };
> + };
> + };
> diff --git a/drivers/media/platform/soc_camera/atmel-isi.c
> b/drivers/media/platform/soc_camera/atmel-isi.c index f4add0a..d6a1f7b
> 100644
> --- a/drivers/media/platform/soc_camera/atmel-isi.c
> +++ b/drivers/media/platform/soc_camera/atmel-isi.c
> @@ -19,6 +19,7 @@
> #include <linux/interrupt.h>
> #include <linux/kernel.h>
> #include <linux/module.h>
> +#include <linux/of.h>
> #include <linux/platform_device.h>
> #include <linux/slab.h>
>
> @@ -33,6 +34,7 @@
> #define VID_LIMIT_BYTES (16 * 1024 * 1024)
> #define MIN_FRAME_RATE 15
> #define FRAME_INTERVAL_MILLI_SEC (1000 / MIN_FRAME_RATE)
> +#define ISI_DEFAULT_MCLK_FREQ 25000000
>
> /* Frame buffer descriptor */
> struct fbd {
> @@ -885,6 +887,20 @@ static int atmel_isi_remove(struct platform_device
> *pdev) return 0;
> }
>
> +static int atmel_isi_probe_dt(struct atmel_isi *isi,
> + struct platform_device *pdev)
> +{
> + struct device_node *node = pdev->dev.of_node;
> +
> + /* Default settings for ISI */
> + isi->pdata.full_mode = 1;
> + isi->pdata.mck_hz = ISI_DEFAULT_MCLK_FREQ;
> + isi->pdata.frate = ISI_CFG1_FRATE_CAPTURE_ALL;
> + isi->pdata.data_width_flags = ISI_DATAWIDTH_8 | ISI_DATAWIDTH_10;
> +
> + return 0;
> +}
> +
> static int atmel_isi_probe(struct platform_device *pdev)
> {
> unsigned int irq;
> @@ -896,7 +912,7 @@ static int atmel_isi_probe(struct platform_device *pdev)
> struct isi_platform_data *pdata;
>
> pdata = dev->platform_data;
> - if (!pdata || !pdata->data_width_flags) {
> + if ((!pdata || !pdata->data_width_flags) && !pdev->dev.of_node) {
> dev_err(&pdev->dev,
> "No config available for Atmel ISI\n");
> return -EINVAL;
> @@ -912,7 +928,11 @@ static int atmel_isi_probe(struct platform_device
> *pdev) if (IS_ERR(isi->pclk))
> return PTR_ERR(isi->pclk);
>
> - memcpy(&isi->pdata, pdata, sizeof(struct isi_platform_data));
> + if (pdata)
> + memcpy(&isi->pdata, pdata, sizeof(struct isi_platform_data));
> + else /* dt probe */
> + atmel_isi_probe_dt(isi, pdev);
> +
> isi->active = NULL;
> spin_lock_init(&isi->lock);
> INIT_LIST_HEAD(&isi->video_buffer_list);
> @@ -1014,11 +1034,18 @@ err_alloc_ctx:
> return ret;
> }
>
> +static const struct of_device_id atmel_isi_of_match[] = {
> + { .compatible = "atmel,at91sam9g45-isi" },
> + { }
> +};
> +MODULE_DEVICE_TABLE(of, atmel_isi_of_match);
> +
> static struct platform_driver atmel_isi_driver = {
> .remove = atmel_isi_remove,
> .driver = {
> .name = "atmel_isi",
> .owner = THIS_MODULE,
> + .of_match_table = atmel_isi_of_match,
> },
> };
--
Regards,
Laurent Pinchart
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [PATCH v2 3/3] [media] atmel-isi: add primary DT support
2014-03-25 10:45 ` [PATCH v2 3/3] [media] atmel-isi: add primary DT support Josh Wu
2014-03-26 16:12 ` Laurent Pinchart
@ 2014-03-30 21:20 ` Guennadi Liakhovetski
2014-03-31 9:05 ` Josh Wu
2014-05-18 21:51 ` Guennadi Liakhovetski
2 siblings, 1 reply; 13+ messages in thread
From: Guennadi Liakhovetski @ 2014-03-30 21:20 UTC (permalink / raw)
To: Josh Wu
Cc: Linux Media Mailing List, m.chehab, nicolas.ferre,
linux-arm-kernel, grant.likely, galak, rob, mark.rutland,
robh+dt, ijc+devicetree, pawel.moll, devicetree,
Laurent Pinchart, Ben Dooks
Hi Josh,
Please correct me if I'm wrong, but I don't see how this is going to work
without the central part - building asynchronous V4L2 data structures from
the DT, something that your earlier patch "media: soc-camera: OF cameras"
was doing, but which you stopped developing after a discussion with Ben
(added to Cc).
Thanks
Guennadi
On Tue, 25 Mar 2014, Josh Wu wrote:
> This patch add the DT support for Atmel ISI driver.
> It use the same v4l2 DT interface that defined in video-interfaces.txt.
>
> Signed-off-by: Josh Wu <josh.wu@atmel.com>
> Cc: devicetree@vger.kernel.org
> ---
> v1 --> v2:
> refine the binding document.
> add port node description.
> removed the optional property.
>
> .../devicetree/bindings/media/atmel-isi.txt | 50 ++++++++++++++++++++
> drivers/media/platform/soc_camera/atmel-isi.c | 31 +++++++++++-
> 2 files changed, 79 insertions(+), 2 deletions(-)
> create mode 100644 Documentation/devicetree/bindings/media/atmel-isi.txt
>
> diff --git a/Documentation/devicetree/bindings/media/atmel-isi.txt b/Documentation/devicetree/bindings/media/atmel-isi.txt
> new file mode 100644
> index 0000000..11c98ee
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/media/atmel-isi.txt
> @@ -0,0 +1,50 @@
> +Atmel Image Sensor Interface (ISI) SoC Camera Subsystem
> +----------------------------------------------
> +
> +Required properties:
> +- compatible: must be "atmel,at91sam9g45-isi"
> +- reg: physical base address and length of the registers set for the device;
> +- interrupts: should contain IRQ line for the ISI;
> +- clocks: list of clock specifiers, corresponding to entries in
> + the clock-names property;
> +- clock-names: must contain "isi_clk", which is the isi peripherial clock.
> +
> +ISI supports a single port node with parallel bus. It should contain one
> +'port' child node with child 'endpoint' node. Please refer to the bindings
> +defined in Documentation/devicetree/bindings/media/video-interfaces.txt.
> +
> +Example:
> + isi: isi@f0034000 {
> + compatible = "atmel,at91sam9g45-isi";
> + reg = <0xf0034000 0x4000>;
> + interrupts = <37 IRQ_TYPE_LEVEL_HIGH 5>;
> +
> + clocks = <&isi_clk>;
> + clock-names = "isi_clk";
> +
> + pinctrl-names = "default";
> + pinctrl-0 = <&pinctrl_isi>;
> +
> + port {
> + #address-cells = <1>;
> + #size-cells = <0>;
> +
> + isi_0: endpoint {
> + remote-endpoint = <&ov2640_0>;
> + };
> + };
> + };
> +
> + i2c1: i2c@f0018000 {
> + ov2640: camera@0x30 {
> + compatible = "omnivision,ov2640";
> + reg = <0x30>;
> +
> + port {
> + ov2640_0: endpoint {
> + remote-endpoint = <&isi_0>;
> + bus-width = <8>;
> + };
> + };
> + };
> + };
> diff --git a/drivers/media/platform/soc_camera/atmel-isi.c b/drivers/media/platform/soc_camera/atmel-isi.c
> index f4add0a..d6a1f7b 100644
> --- a/drivers/media/platform/soc_camera/atmel-isi.c
> +++ b/drivers/media/platform/soc_camera/atmel-isi.c
> @@ -19,6 +19,7 @@
> #include <linux/interrupt.h>
> #include <linux/kernel.h>
> #include <linux/module.h>
> +#include <linux/of.h>
> #include <linux/platform_device.h>
> #include <linux/slab.h>
>
> @@ -33,6 +34,7 @@
> #define VID_LIMIT_BYTES (16 * 1024 * 1024)
> #define MIN_FRAME_RATE 15
> #define FRAME_INTERVAL_MILLI_SEC (1000 / MIN_FRAME_RATE)
> +#define ISI_DEFAULT_MCLK_FREQ 25000000
>
> /* Frame buffer descriptor */
> struct fbd {
> @@ -885,6 +887,20 @@ static int atmel_isi_remove(struct platform_device *pdev)
> return 0;
> }
>
> +static int atmel_isi_probe_dt(struct atmel_isi *isi,
> + struct platform_device *pdev)
> +{
> + struct device_node *node = pdev->dev.of_node;
> +
> + /* Default settings for ISI */
> + isi->pdata.full_mode = 1;
> + isi->pdata.mck_hz = ISI_DEFAULT_MCLK_FREQ;
> + isi->pdata.frate = ISI_CFG1_FRATE_CAPTURE_ALL;
> + isi->pdata.data_width_flags = ISI_DATAWIDTH_8 | ISI_DATAWIDTH_10;
> +
> + return 0;
> +}
> +
> static int atmel_isi_probe(struct platform_device *pdev)
> {
> unsigned int irq;
> @@ -896,7 +912,7 @@ static int atmel_isi_probe(struct platform_device *pdev)
> struct isi_platform_data *pdata;
>
> pdata = dev->platform_data;
> - if (!pdata || !pdata->data_width_flags) {
> + if ((!pdata || !pdata->data_width_flags) && !pdev->dev.of_node) {
> dev_err(&pdev->dev,
> "No config available for Atmel ISI\n");
> return -EINVAL;
> @@ -912,7 +928,11 @@ static int atmel_isi_probe(struct platform_device *pdev)
> if (IS_ERR(isi->pclk))
> return PTR_ERR(isi->pclk);
>
> - memcpy(&isi->pdata, pdata, sizeof(struct isi_platform_data));
> + if (pdata)
> + memcpy(&isi->pdata, pdata, sizeof(struct isi_platform_data));
> + else /* dt probe */
> + atmel_isi_probe_dt(isi, pdev);
> +
> isi->active = NULL;
> spin_lock_init(&isi->lock);
> INIT_LIST_HEAD(&isi->video_buffer_list);
> @@ -1014,11 +1034,18 @@ err_alloc_ctx:
> return ret;
> }
>
> +static const struct of_device_id atmel_isi_of_match[] = {
> + { .compatible = "atmel,at91sam9g45-isi" },
> + { }
> +};
> +MODULE_DEVICE_TABLE(of, atmel_isi_of_match);
> +
> static struct platform_driver atmel_isi_driver = {
> .remove = atmel_isi_remove,
> .driver = {
> .name = "atmel_isi",
> .owner = THIS_MODULE,
> + .of_match_table = atmel_isi_of_match,
> },
> };
>
> --
> 1.7.9.5
>
---
Guennadi Liakhovetski, Ph.D.
Freelance Open-Source Software Developer
http://www.open-technology.de/
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [PATCH v2 3/3] [media] atmel-isi: add primary DT support
2014-03-30 21:20 ` Guennadi Liakhovetski
@ 2014-03-31 9:05 ` Josh Wu
2014-07-17 11:00 ` Laurent Pinchart
0 siblings, 1 reply; 13+ messages in thread
From: Josh Wu @ 2014-03-31 9:05 UTC (permalink / raw)
To: Guennadi Liakhovetski
Cc: Linux Media Mailing List, m.chehab, nicolas.ferre,
linux-arm-kernel, grant.likely, galak, rob, mark.rutland,
robh+dt, ijc+devicetree, pawel.moll, devicetree,
Laurent Pinchart, Ben Dooks
Dear Guennadi
On 3/31/2014 5:20 AM, Guennadi Liakhovetski wrote:
> Hi Josh,
>
> Please correct me if I'm wrong, but I don't see how this is going to work
> without the central part - building asynchronous V4L2 data structures from
> the DT, something that your earlier patch
Here you mean Bryan Wu not me, right? ;-)
Bryan write the patch "[v2] media: soc-camera: OF cameras" in:
https://patchwork.linuxtv.org/patch/22288/.
And I saw Ben Dooks already sent out his patch to support soc-camera OF
now (https://patchwork.linuxtv.org/patch/23304/) which is simpler than
Bryan's.
> "media: soc-camera: OF cameras"
> was doing, but which you stopped developing after a discussion with Ben
> (added to Cc).
And yes, atmel-isi dt patch should not work without above SoC-Camera of
support patch.
But as the atmel-isi dt binding document and port node can be finalized.
So I think this patch is ready for the mainline.
BTW: I will test Ben's patch with atmel-isi.
thanks and best regards,
Josh Wu
>
> Thanks
> Guennadi
>
> On Tue, 25 Mar 2014, Josh Wu wrote:
>
>> This patch add the DT support for Atmel ISI driver.
>> It use the same v4l2 DT interface that defined in video-interfaces.txt.
>>
>> Signed-off-by: Josh Wu <josh.wu@atmel.com>
>> Cc: devicetree@vger.kernel.org
>> ---
>> v1 --> v2:
>> refine the binding document.
>> add port node description.
>> removed the optional property.
>>
>> .../devicetree/bindings/media/atmel-isi.txt | 50 ++++++++++++++++++++
>> drivers/media/platform/soc_camera/atmel-isi.c | 31 +++++++++++-
>> 2 files changed, 79 insertions(+), 2 deletions(-)
>> create mode 100644 Documentation/devicetree/bindings/media/atmel-isi.txt
>>
>> diff --git a/Documentation/devicetree/bindings/media/atmel-isi.txt b/Documentation/devicetree/bindings/media/atmel-isi.txt
>> new file mode 100644
>> index 0000000..11c98ee
>> --- /dev/null
>> +++ b/Documentation/devicetree/bindings/media/atmel-isi.txt
>> @@ -0,0 +1,50 @@
>> +Atmel Image Sensor Interface (ISI) SoC Camera Subsystem
>> +----------------------------------------------
>> +
>> +Required properties:
>> +- compatible: must be "atmel,at91sam9g45-isi"
>> +- reg: physical base address and length of the registers set for the device;
>> +- interrupts: should contain IRQ line for the ISI;
>> +- clocks: list of clock specifiers, corresponding to entries in
>> + the clock-names property;
>> +- clock-names: must contain "isi_clk", which is the isi peripherial clock.
>> +
>> +ISI supports a single port node with parallel bus. It should contain one
>> +'port' child node with child 'endpoint' node. Please refer to the bindings
>> +defined in Documentation/devicetree/bindings/media/video-interfaces.txt.
>> +
>> +Example:
>> + isi: isi@f0034000 {
>> + compatible = "atmel,at91sam9g45-isi";
>> + reg = <0xf0034000 0x4000>;
>> + interrupts = <37 IRQ_TYPE_LEVEL_HIGH 5>;
>> +
>> + clocks = <&isi_clk>;
>> + clock-names = "isi_clk";
>> +
>> + pinctrl-names = "default";
>> + pinctrl-0 = <&pinctrl_isi>;
>> +
>> + port {
>> + #address-cells = <1>;
>> + #size-cells = <0>;
>> +
>> + isi_0: endpoint {
>> + remote-endpoint = <&ov2640_0>;
>> + };
>> + };
>> + };
>> +
>> + i2c1: i2c@f0018000 {
>> + ov2640: camera@0x30 {
>> + compatible = "omnivision,ov2640";
>> + reg = <0x30>;
>> +
>> + port {
>> + ov2640_0: endpoint {
>> + remote-endpoint = <&isi_0>;
>> + bus-width = <8>;
>> + };
>> + };
>> + };
>> + };
>> diff --git a/drivers/media/platform/soc_camera/atmel-isi.c b/drivers/media/platform/soc_camera/atmel-isi.c
>> index f4add0a..d6a1f7b 100644
>> --- a/drivers/media/platform/soc_camera/atmel-isi.c
>> +++ b/drivers/media/platform/soc_camera/atmel-isi.c
>> @@ -19,6 +19,7 @@
>> #include <linux/interrupt.h>
>> #include <linux/kernel.h>
>> #include <linux/module.h>
>> +#include <linux/of.h>
>> #include <linux/platform_device.h>
>> #include <linux/slab.h>
>>
>> @@ -33,6 +34,7 @@
>> #define VID_LIMIT_BYTES (16 * 1024 * 1024)
>> #define MIN_FRAME_RATE 15
>> #define FRAME_INTERVAL_MILLI_SEC (1000 / MIN_FRAME_RATE)
>> +#define ISI_DEFAULT_MCLK_FREQ 25000000
>>
>> /* Frame buffer descriptor */
>> struct fbd {
>> @@ -885,6 +887,20 @@ static int atmel_isi_remove(struct platform_device *pdev)
>> return 0;
>> }
>>
>> +static int atmel_isi_probe_dt(struct atmel_isi *isi,
>> + struct platform_device *pdev)
>> +{
>> + struct device_node *node = pdev->dev.of_node;
>> +
>> + /* Default settings for ISI */
>> + isi->pdata.full_mode = 1;
>> + isi->pdata.mck_hz = ISI_DEFAULT_MCLK_FREQ;
>> + isi->pdata.frate = ISI_CFG1_FRATE_CAPTURE_ALL;
>> + isi->pdata.data_width_flags = ISI_DATAWIDTH_8 | ISI_DATAWIDTH_10;
>> +
>> + return 0;
>> +}
>> +
>> static int atmel_isi_probe(struct platform_device *pdev)
>> {
>> unsigned int irq;
>> @@ -896,7 +912,7 @@ static int atmel_isi_probe(struct platform_device *pdev)
>> struct isi_platform_data *pdata;
>>
>> pdata = dev->platform_data;
>> - if (!pdata || !pdata->data_width_flags) {
>> + if ((!pdata || !pdata->data_width_flags) && !pdev->dev.of_node) {
>> dev_err(&pdev->dev,
>> "No config available for Atmel ISI\n");
>> return -EINVAL;
>> @@ -912,7 +928,11 @@ static int atmel_isi_probe(struct platform_device *pdev)
>> if (IS_ERR(isi->pclk))
>> return PTR_ERR(isi->pclk);
>>
>> - memcpy(&isi->pdata, pdata, sizeof(struct isi_platform_data));
>> + if (pdata)
>> + memcpy(&isi->pdata, pdata, sizeof(struct isi_platform_data));
>> + else /* dt probe */
>> + atmel_isi_probe_dt(isi, pdev);
>> +
>> isi->active = NULL;
>> spin_lock_init(&isi->lock);
>> INIT_LIST_HEAD(&isi->video_buffer_list);
>> @@ -1014,11 +1034,18 @@ err_alloc_ctx:
>> return ret;
>> }
>>
>> +static const struct of_device_id atmel_isi_of_match[] = {
>> + { .compatible = "atmel,at91sam9g45-isi" },
>> + { }
>> +};
>> +MODULE_DEVICE_TABLE(of, atmel_isi_of_match);
>> +
>> static struct platform_driver atmel_isi_driver = {
>> .remove = atmel_isi_remove,
>> .driver = {
>> .name = "atmel_isi",
>> .owner = THIS_MODULE,
>> + .of_match_table = atmel_isi_of_match,
>> },
>> };
>>
>> --
>> 1.7.9.5
>>
> ---
> Guennadi Liakhovetski, Ph.D.
> Freelance Open-Source Software Developer
> http://www.open-technology.de/
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [PATCH v2 2/3] [media] atmel-isi: convert the pdata from pointer to structure
2014-03-25 10:41 ` [PATCH v2 2/3] [media] atmel-isi: convert the pdata from pointer to structure Josh Wu
@ 2014-05-18 20:59 ` Guennadi Liakhovetski
2014-05-23 3:15 ` Josh Wu
0 siblings, 1 reply; 13+ messages in thread
From: Guennadi Liakhovetski @ 2014-05-18 20:59 UTC (permalink / raw)
To: Josh Wu
Cc: linux-media, m.chehab, nicolas.ferre, linux-arm-kernel, laurent.pinchart
Hi Josh,
I'm still waiting for an update of Ben's patches to then also apply yours,
but I decided to have a look at yours now to see if I find anything, that
might be worth changing. A small note to this one below.
On Tue, 25 Mar 2014, Josh Wu wrote:
> Now the platform data is initialized by allocation of isi
> structure. In the future, we use pdata to store the dt parameters.
>
> Signed-off-by: Josh Wu <josh.wu@atmel.com>
> ---
> v1 --> v2:
> no change.
>
> drivers/media/platform/soc_camera/atmel-isi.c | 22 +++++++++++-----------
> 1 file changed, 11 insertions(+), 11 deletions(-)
>
> diff --git a/drivers/media/platform/soc_camera/atmel-isi.c b/drivers/media/platform/soc_camera/atmel-isi.c
> index 9d977c5..f4add0a 100644
> --- a/drivers/media/platform/soc_camera/atmel-isi.c
> +++ b/drivers/media/platform/soc_camera/atmel-isi.c
[snip]
> @@ -912,7 +912,7 @@ static int atmel_isi_probe(struct platform_device *pdev)
> if (IS_ERR(isi->pclk))
> return PTR_ERR(isi->pclk);
>
> - isi->pdata = pdata;
> + memcpy(&isi->pdata, pdata, sizeof(struct isi_platform_data));
I think it'd be better to use
+ memcpy(&isi->pdata, pdata, sizeof(isi->pdata));
This way if the type of the pdata changes at any time in the future this
line will not have to be changed. If you don't mind I can make this change
myself, so you don't have to make a new version just for this.
Thanks
Guennadi
---
Guennadi Liakhovetski, Ph.D.
Freelance Open-Source Software Developer
http://www.open-technology.de/
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [PATCH v2 3/3] [media] atmel-isi: add primary DT support
2014-03-25 10:45 ` [PATCH v2 3/3] [media] atmel-isi: add primary DT support Josh Wu
2014-03-26 16:12 ` Laurent Pinchart
2014-03-30 21:20 ` Guennadi Liakhovetski
@ 2014-05-18 21:51 ` Guennadi Liakhovetski
2014-05-23 3:12 ` Josh Wu
2 siblings, 1 reply; 13+ messages in thread
From: Guennadi Liakhovetski @ 2014-05-18 21:51 UTC (permalink / raw)
To: Josh Wu
Cc: linux-media, m.chehab, nicolas.ferre, linux-arm-kernel,
grant.likely, galak, rob, mark.rutland, robh+dt, ijc+devicetree,
pawel.moll, devicetree, laurent.pinchart
On Tue, 25 Mar 2014, Josh Wu wrote:
> This patch add the DT support for Atmel ISI driver.
> It use the same v4l2 DT interface that defined in video-interfaces.txt.
>
> Signed-off-by: Josh Wu <josh.wu@atmel.com>
> Cc: devicetree@vger.kernel.org
> ---
> v1 --> v2:
> refine the binding document.
> add port node description.
> removed the optional property.
>
> .../devicetree/bindings/media/atmel-isi.txt | 50 ++++++++++++++++++++
> drivers/media/platform/soc_camera/atmel-isi.c | 31 +++++++++++-
> 2 files changed, 79 insertions(+), 2 deletions(-)
> create mode 100644 Documentation/devicetree/bindings/media/atmel-isi.txt
[snip]
> diff --git a/drivers/media/platform/soc_camera/atmel-isi.c b/drivers/media/platform/soc_camera/atmel-isi.c
> index f4add0a..d6a1f7b 100644
> --- a/drivers/media/platform/soc_camera/atmel-isi.c
> +++ b/drivers/media/platform/soc_camera/atmel-isi.c
[snip]
> @@ -885,6 +887,20 @@ static int atmel_isi_remove(struct platform_device *pdev)
> return 0;
> }
>
> +static int atmel_isi_probe_dt(struct atmel_isi *isi,
> + struct platform_device *pdev)
> +{
> + struct device_node *node = pdev->dev.of_node;
> +
> + /* Default settings for ISI */
> + isi->pdata.full_mode = 1;
> + isi->pdata.mck_hz = ISI_DEFAULT_MCLK_FREQ;
> + isi->pdata.frate = ISI_CFG1_FRATE_CAPTURE_ALL;
The above flags eventually should probably partially be added as new
driver-specific DT properties, partially derived from DT clock bindings.
But I'm ok to have them fixed like this in the initial version.
> + isi->pdata.data_width_flags = ISI_DATAWIDTH_8 | ISI_DATAWIDTH_10;
Whereas these flags, I think, should already now be derived from the
bus-width standard property? v4l2_of_parse_parallel_bus() will extract
them for you and I just asked Ben to add a call to
v4l2_of_parse_endpoint() to his patch. Consequently you'll have to
rearrange bus-width interpretation in isi_camera_try_bus_param() a bit and
use OF parsing results there directly if available? Or maybe you find a
better way. It would certainly be better to extend your probing code and
just use OF results to initialise isi->width_flags, but that might be
impossible, because OF parsing would be performed inside
soc_camera_host_register() and your isi_camera_try_bus_param() can also be
called immediately from it if all required I2C devices are already
available? If your I2C subdevice drivers defer probing until the host has
probed, then you could initialise .width_flags after
soc_camera_host_register(), but you cannot rely on that.
Thanks
Guennadi
---
Guennadi Liakhovetski, Ph.D.
Freelance Open-Source Software Developer
http://www.open-technology.de/
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [PATCH v2 3/3] [media] atmel-isi: add primary DT support
2014-05-18 21:51 ` Guennadi Liakhovetski
@ 2014-05-23 3:12 ` Josh Wu
0 siblings, 0 replies; 13+ messages in thread
From: Josh Wu @ 2014-05-23 3:12 UTC (permalink / raw)
To: Guennadi Liakhovetski
Cc: linux-media, m.chehab, nicolas.ferre, linux-arm-kernel,
grant.likely, galak, rob, mark.rutland, robh+dt, ijc+devicetree,
pawel.moll, devicetree, laurent.pinchart
Hi, Guennadi
On 5/19/2014 5:51 AM, Guennadi Liakhovetski wrote:
> On Tue, 25 Mar 2014, Josh Wu wrote:
>
>> This patch add the DT support for Atmel ISI driver.
>> It use the same v4l2 DT interface that defined in video-interfaces.txt.
>>
>> Signed-off-by: Josh Wu <josh.wu@atmel.com>
>> Cc: devicetree@vger.kernel.org
>> ---
>> v1 --> v2:
>> refine the binding document.
>> add port node description.
>> removed the optional property.
>>
>> .../devicetree/bindings/media/atmel-isi.txt | 50 ++++++++++++++++++++
>> drivers/media/platform/soc_camera/atmel-isi.c | 31 +++++++++++-
>> 2 files changed, 79 insertions(+), 2 deletions(-)
>> create mode 100644 Documentation/devicetree/bindings/media/atmel-isi.txt
> [snip]
>
>> diff --git a/drivers/media/platform/soc_camera/atmel-isi.c b/drivers/media/platform/soc_camera/atmel-isi.c
>> index f4add0a..d6a1f7b 100644
>> --- a/drivers/media/platform/soc_camera/atmel-isi.c
>> +++ b/drivers/media/platform/soc_camera/atmel-isi.c
> [snip]
>
>> @@ -885,6 +887,20 @@ static int atmel_isi_remove(struct platform_device *pdev)
>> return 0;
>> }
>>
>> +static int atmel_isi_probe_dt(struct atmel_isi *isi,
>> + struct platform_device *pdev)
>> +{
>> + struct device_node *node = pdev->dev.of_node;
>> +
>> + /* Default settings for ISI */
>> + isi->pdata.full_mode = 1;
>> + isi->pdata.mck_hz = ISI_DEFAULT_MCLK_FREQ;
>> + isi->pdata.frate = ISI_CFG1_FRATE_CAPTURE_ALL;
> The above flags eventually should probably partially be added as new
> driver-specific DT properties, partially derived from DT clock bindings.
> But I'm ok to have them fixed like this in the initial version.
>
>> + isi->pdata.data_width_flags = ISI_DATAWIDTH_8 | ISI_DATAWIDTH_10;
> Whereas these flags, I think, should already now be derived from the
> bus-width standard property?
yes. I agree.
> v4l2_of_parse_parallel_bus() will extract
> them for you and I just asked Ben to add a call to
> v4l2_of_parse_endpoint() to his patch.
Is it better to call v4l2_of_parse_endpoint() in the atmel-isi driver? I
think the dt parsing stuff should be done by host driver and sensor
driver itself. No need to call v4l2_of_parse_endpoint() in soc-camera.c.
> Consequently you'll have to
> rearrange bus-width interpretation in isi_camera_try_bus_param() a bit and
> use OF parsing results there directly if available? Or maybe you find a
> better way. It would certainly be better to extend your probing code and
> just use OF results to initialise isi->width_flags, but that might be
> impossible, because OF parsing would be performed inside
> soc_camera_host_register() and your isi_camera_try_bus_param() can also be
> called immediately from it if all required I2C devices are already
> available?
I am little bit confuse here. I don't see any issue in above case. Since
atmel_isi_probe_dt() will always be called earlier then
soc_camera_host_register().
That means when soc_camera_host_register() called
isi_camera_try_bus_param(), the isi->width_flags are already initialized
in a valid value by atmel_isi_probe_dt().
Am I missing anything here?
> If your I2C subdevice drivers defer probing until the host has
> probed, then you could initialise .width_flags after
> soc_camera_host_register(), but you cannot rely on that.
I tested these two cases without any issue:
1. In dtb, the i2c sensor dt node probe earlier than atmel-isi dt node.
i2c sensor will do a defer probe here as mclk is not found until
atmel-isi driver probed and call soc_camera_host_register().
2. In dtb, the atmel-isi dt node is probed earlier than i2c sensor.
Best Regards,
Josh Wu
>
> Thanks
> Guennadi
> ---
> Guennadi Liakhovetski, Ph.D.
> Freelance Open-Source Software Developer
> http://www.open-technology.de/
> --
> To unsubscribe from this list: send the line "unsubscribe devicetree" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [PATCH v2 2/3] [media] atmel-isi: convert the pdata from pointer to structure
2014-05-18 20:59 ` Guennadi Liakhovetski
@ 2014-05-23 3:15 ` Josh Wu
0 siblings, 0 replies; 13+ messages in thread
From: Josh Wu @ 2014-05-23 3:15 UTC (permalink / raw)
To: Guennadi Liakhovetski
Cc: linux-media, m.chehab, nicolas.ferre, linux-arm-kernel, laurent.pinchart
Hi, Guennadi
On 5/19/2014 4:59 AM, Guennadi Liakhovetski wrote:
> Hi Josh,
>
> I'm still waiting for an update of Ben's patches to then also apply yours,
> but I decided to have a look at yours now to see if I find anything, that
> might be worth changing. A small note to this one below.
>
> On Tue, 25 Mar 2014, Josh Wu wrote:
>
>> Now the platform data is initialized by allocation of isi
>> structure. In the future, we use pdata to store the dt parameters.
>>
>> Signed-off-by: Josh Wu <josh.wu@atmel.com>
>> ---
>> v1 --> v2:
>> no change.
>>
>> drivers/media/platform/soc_camera/atmel-isi.c | 22 +++++++++++-----------
>> 1 file changed, 11 insertions(+), 11 deletions(-)
>>
>> diff --git a/drivers/media/platform/soc_camera/atmel-isi.c b/drivers/media/platform/soc_camera/atmel-isi.c
>> index 9d977c5..f4add0a 100644
>> --- a/drivers/media/platform/soc_camera/atmel-isi.c
>> +++ b/drivers/media/platform/soc_camera/atmel-isi.c
> [snip]
>
>> @@ -912,7 +912,7 @@ static int atmel_isi_probe(struct platform_device *pdev)
>> if (IS_ERR(isi->pclk))
>> return PTR_ERR(isi->pclk);
>>
>> - isi->pdata = pdata;
>> + memcpy(&isi->pdata, pdata, sizeof(struct isi_platform_data));
> I think it'd be better to use
>
> + memcpy(&isi->pdata, pdata, sizeof(isi->pdata));
>
> This way if the type of the pdata changes at any time in the future this
> line will not have to be changed. If you don't mind I can make this change
> myself, so you don't have to make a new version just for this.
Thanks for pointing it out. I think I will sent out a new version of
patch (include bus-width parsing) then I will included with this fix.
Best Regards,
Josh Wu
>
> Thanks
> Guennadi
> ---
> Guennadi Liakhovetski, Ph.D.
> Freelance Open-Source Software Developer
> http://www.open-technology.de/
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [PATCH v2 3/3] [media] atmel-isi: add primary DT support
2014-03-31 9:05 ` Josh Wu
@ 2014-07-17 11:00 ` Laurent Pinchart
2014-07-18 3:15 ` Josh Wu
0 siblings, 1 reply; 13+ messages in thread
From: Laurent Pinchart @ 2014-07-17 11:00 UTC (permalink / raw)
To: Josh Wu
Cc: Guennadi Liakhovetski, Linux Media Mailing List, m.chehab,
nicolas.ferre, linux-arm-kernel, grant.likely, galak, rob,
mark.rutland, robh+dt, ijc+devicetree, pawel.moll, devicetree,
Ben Dooks
Hi Josh,
What's the status of this patch set ? Do you plan to rebase and resubmit it ?
On Monday 31 March 2014 17:05:13 Josh Wu wrote:
> Dear Guennadi
>
> On 3/31/2014 5:20 AM, Guennadi Liakhovetski wrote:
> > Hi Josh,
> >
> > Please correct me if I'm wrong, but I don't see how this is going to work
> > without the central part - building asynchronous V4L2 data structures from
> > the DT, something that your earlier patch
>
> Here you mean Bryan Wu not me, right? ;-)
> Bryan write the patch "[v2] media: soc-camera: OF cameras" in:
> https://patchwork.linuxtv.org/patch/22288/.
> And I saw Ben Dooks already sent out his patch to support soc-camera OF
> now (https://patchwork.linuxtv.org/patch/23304/) which is simpler than
> Bryan's.
>
> > "media: soc-camera: OF cameras"
> > was doing, but which you stopped developing after a discussion with Ben
> > (added to Cc).
>
> And yes, atmel-isi dt patch should not work without above SoC-Camera of
> support patch.
> But as the atmel-isi dt binding document and port node can be finalized.
> So I think this patch is ready for the mainline.
>
> BTW: I will test Ben's patch with atmel-isi.
--
Regards,
Laurent Pinchart
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [PATCH v2 3/3] [media] atmel-isi: add primary DT support
2014-07-17 11:00 ` Laurent Pinchart
@ 2014-07-18 3:15 ` Josh Wu
0 siblings, 0 replies; 13+ messages in thread
From: Josh Wu @ 2014-07-18 3:15 UTC (permalink / raw)
To: Laurent Pinchart
Cc: Guennadi Liakhovetski, Linux Media Mailing List, m.chehab,
nicolas.ferre, linux-arm-kernel, grant.likely, galak, rob,
mark.rutland, robh+dt, ijc+devicetree, pawel.moll, devicetree,
Ben Dooks
Dear Laurent
On 7/17/2014 7:00 PM, Laurent Pinchart wrote:
> Hi Josh,
>
> What's the status of this patch set ? Do you plan to rebase and resubmit it ?
Thanks for the reminding.
yes, I will rebase it and resubmit the new version for this patch set
with the data bus width support.
Thanks.
Best Regards,
Josh Wu
>
> On Monday 31 March 2014 17:05:13 Josh Wu wrote:
>> Dear Guennadi
>>
>> On 3/31/2014 5:20 AM, Guennadi Liakhovetski wrote:
>>> Hi Josh,
>>>
>>> Please correct me if I'm wrong, but I don't see how this is going to work
>>> without the central part - building asynchronous V4L2 data structures from
>>> the DT, something that your earlier patch
>> Here you mean Bryan Wu not me, right? ;-)
>> Bryan write the patch "[v2] media: soc-camera: OF cameras" in:
>> https://patchwork.linuxtv.org/patch/22288/.
>> And I saw Ben Dooks already sent out his patch to support soc-camera OF
>> now (https://patchwork.linuxtv.org/patch/23304/) which is simpler than
>> Bryan's.
>>
>>> "media: soc-camera: OF cameras"
>>> was doing, but which you stopped developing after a discussion with Ben
>>> (added to Cc).
>> And yes, atmel-isi dt patch should not work without above SoC-Camera of
>> support patch.
>> But as the atmel-isi dt binding document and port node can be finalized.
>> So I think this patch is ready for the mainline.
>>
>> BTW: I will test Ben's patch with atmel-isi.
^ permalink raw reply [flat|nested] 13+ messages in thread
end of thread, other threads:[~2014-07-18 3:16 UTC | newest]
Thread overview: 13+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2014-03-25 10:41 [PATCH v2 0/3] [media] atmel-isi: Add DT support for Atmel ISI driver Josh Wu
2014-03-25 10:41 ` [PATCH v2 1/3] [media] atmel-isi: add v4l2 async probe support Josh Wu
2014-03-25 10:41 ` [PATCH v2 2/3] [media] atmel-isi: convert the pdata from pointer to structure Josh Wu
2014-05-18 20:59 ` Guennadi Liakhovetski
2014-05-23 3:15 ` Josh Wu
2014-03-25 10:45 ` [PATCH v2 3/3] [media] atmel-isi: add primary DT support Josh Wu
2014-03-26 16:12 ` Laurent Pinchart
2014-03-30 21:20 ` Guennadi Liakhovetski
2014-03-31 9:05 ` Josh Wu
2014-07-17 11:00 ` Laurent Pinchart
2014-07-18 3:15 ` Josh Wu
2014-05-18 21:51 ` Guennadi Liakhovetski
2014-05-23 3:12 ` Josh Wu
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).