[PATCH RESEND] dma: tegra: add const to of_device_id.data's structure

The member "data" of structure of_device_id is changed to const
pointer. Hence initializing this with const type data.
This will fix the compilation warning like
drivers/dma/tegra20-apb-dma.c:1217:9: warning: assignment discards 'const' qualifier from
pointer target type [enabled by default]

This warning came after the change
---------
commit 98d7bbb9929bcc14e11ac8a55245a4f2dc174e17
Author: Uwe Kleine-König [off-list ref]
    of: add const to struct *of_device_id.data
---------

Signed-off-by: Laxman Dewangan <redacted>
Acked-by: Stephen Warren <redacted>
Acked-by: Vinod Koul <vinod.koul-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>

---
Resending patch to Uwe and adding ack from Stephen and Vinod.
Fix typo in description.

 drivers/dma/tegra20-apb-dma.c |    8 ++++----
 1 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/dma/tegra20-apb-dma.c b/drivers/dma/tegra20-apb-dma.c
index 24acd71..09a15ff 100644
--- a/drivers/dma/tegra20-apb-dma.c
+++ b/drivers/dma/tegra20-apb-dma.c

@@ -201,7 +201,7 @@ struct tegra_dma {
 	struct clk			*dma_clk;
 	spinlock_t			global_lock;
 	void __iomem			*base_addr;
-	struct tegra_dma_chip_data	*chip_data;
+	const struct tegra_dma_chip_data	*chip_data;
 
 	/* Some register need to be cache before suspend */
 	u32				reg_gen;

@@ -1173,14 +1173,14 @@ static void tegra_dma_free_chan_resources(struct dma_chan *dc)
 }
 
 /* Tegra20 specific DMA controller information */
-static struct tegra_dma_chip_data tegra20_dma_chip_data = {
+static const struct tegra_dma_chip_data tegra20_dma_chip_data = {
 	.nr_channels		= 16,
 	.max_dma_count		= 1024UL * 64,
 };
 
 #if defined(CONFIG_OF)
 /* Tegra30 specific DMA controller information */
-static struct tegra_dma_chip_data tegra30_dma_chip_data = {
+static const struct tegra_dma_chip_data tegra30_dma_chip_data = {
 	.nr_channels		= 32,
 	.max_dma_count		= 1024UL * 64,
 };

@@ -1204,7 +1204,7 @@ static int __devinit tegra_dma_probe(struct platform_device *pdev)
 	struct tegra_dma *tdma;
 	int ret;
 	int i;
-	struct tegra_dma_chip_data *cdata = NULL;
+	const struct tegra_dma_chip_data *cdata = NULL;
 
 	if (pdev->dev.of_node) {
 		const struct of_device_id *match;

-- 
1.7.1.1

Hello,

On Wed, Aug 29, 2012 at 10:47:27AM +0530, Laxman Dewangan wrote:

The member "data" of structure of_device_id is changed to const
pointer. Hence initializing this with const type data.
This will fix the compilation warning like
drivers/dma/tegra20-apb-dma.c:1217:9: warning: assignment discards 'const' qualifier from
pointer target type [enabled by default]

This warning came after the change
---------
commit 98d7bbb9929bcc14e11ac8a55245a4f2dc174e17
Author: Uwe Kleine-König [off-list ref]
    of: add const to struct *of_device_id.data
---------

Signed-off-by: Laxman Dewangan <ldewangan@nvidia.com>
Acked-by: Stephen Warren <redacted>
Acked-by: Vinod Koul <redacted>

I took the patch into my series now and split it in two to prevent
warnings in the middle of the series. I assumed the two acks for both
patches and send them in a moment in reply to this mail.

You can find the updated series

	git://git.pengutronix.de/git/ukl/linux.git ofdeviceiddata

Arnd or Olof, can you please update the corresponding branch arm-soc
from there?

Best regards and thanks
Uwe

From: Laxman Dewangan <ldewangan@nvidia.com>

This prepares *of_device_id.data becoming const. Without this change
the following warning would occur:

	drivers/dma/tegra20-apb-dma.c: In function 'tegra_dma_probe':
	drivers/dma/tegra20-apb-dma.c:1210:9: warning: assignment discards 'const' qualifier from pointer target type [enabled by default]

[ukl: split Laxman's patch to prevent warnings in the middle of the
series. Original patch was sent with msgid
1346217447-19730-1-git-send-email-ldewangan@nvidia.com]

Signed-off-by: Laxman Dewangan <ldewangan@nvidia.com>
Acked-by: Stephen Warren <redacted>
Acked-by: Vinod Koul <redacted>
Signed-off-by: Uwe Kleine-König <redacted>
---
 drivers/dma/tegra20-apb-dma.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/dma/tegra20-apb-dma.c b/drivers/dma/tegra20-apb-dma.c
index d52dbc6..29a39b8 100644
--- a/drivers/dma/tegra20-apb-dma.c
+++ b/drivers/dma/tegra20-apb-dma.c

@@ -201,7 +201,7 @@ struct tegra_dma {
 	struct clk			*dma_clk;
 	spinlock_t			global_lock;
 	void __iomem			*base_addr;
-	struct tegra_dma_chip_data	*chip_data;
+	const struct tegra_dma_chip_data *chip_data;
 
 	/* Some register need to be cache before suspend */
 	u32				reg_gen;

@@ -1197,7 +1197,7 @@ static int __devinit tegra_dma_probe(struct platform_device *pdev)
 	struct tegra_dma *tdma;
 	int ret;
 	int i;
-	struct tegra_dma_chip_data *cdata = NULL;
+	const struct tegra_dma_chip_data *cdata = NULL;
 
 	if (pdev->dev.of_node) {
 		const struct of_device_id *match;

-- 
1.7.10.4

From: Laxman Dewangan <redacted>

Since of_device_id.data is declared as a pointer to const data a few
more consts can be added in this driver.

[ukl: split Laxman's patch to prevent warnings in the middle of the
series. Original patch was sent with msgid
1346217447-19730-1-git-send-email-ldewangan-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org]

Signed-off-by: Laxman Dewangan <redacted>
Acked-by: Stephen Warren <redacted>
Acked-by: Vinod Koul <vinod.koul-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>
Signed-off-by: Uwe Kleine-König <redacted>
---
 drivers/dma/tegra20-apb-dma.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/dma/tegra20-apb-dma.c b/drivers/dma/tegra20-apb-dma.c
index 29a39b8..56adcfc 100644
--- a/drivers/dma/tegra20-apb-dma.c
+++ b/drivers/dma/tegra20-apb-dma.c

@@ -1166,14 +1166,14 @@ static void tegra_dma_free_chan_resources(struct dma_chan *dc)
 }
 
 /* Tegra20 specific DMA controller information */
-static struct tegra_dma_chip_data tegra20_dma_chip_data = {
+static const struct tegra_dma_chip_data tegra20_dma_chip_data = {
 	.nr_channels		= 16,
 	.max_dma_count		= 1024UL * 64,
 };
 
 #if defined(CONFIG_OF)
 /* Tegra30 specific DMA controller information */
-static struct tegra_dma_chip_data tegra30_dma_chip_data = {
+static const struct tegra_dma_chip_data tegra30_dma_chip_data = {
 	.nr_channels		= 32,
 	.max_dma_count		= 1024UL * 64,
 };

-- 
1.7.10.4

On 08/29/2012 02:51 AM, Uwe Kleine-König wrote:

From: Laxman Dewangan <redacted>

This prepares *of_device_id.data becoming const. Without this change
the following warning would occur:

	drivers/dma/tegra20-apb-dma.c: In function 'tegra_dma_probe':
	drivers/dma/tegra20-apb-dma.c:1210:9: warning: assignment discards 'const' qualifier from pointer target type [enabled by default]

[ukl: split Laxman's patch to prevent warnings in the middle of the
series. Original patch was sent with msgid
1346217447-19730-1-git-send-email-ldewangan-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org]

Signed-off-by: Laxman Dewangan <redacted>
Acked-by: Stephen Warren <redacted>
Acked-by: Vinod Koul <vinod.koul-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>
Signed-off-by: Uwe Kleine-König <redacted>

Uwe,

I don't see these patches in linux-next, so compilation is still broken.
Can you please apply them? Thanks.

On Mon, Sep 10, 2012 at 12:21:09PM -0600, Stephen Warren wrote:

On 08/29/2012 02:51 AM, Uwe Kleine-König wrote:

quoted

From: Laxman Dewangan <ldewangan@nvidia.com>

This prepares *of_device_id.data becoming const. Without this change
the following warning would occur:

	drivers/dma/tegra20-apb-dma.c: In function 'tegra_dma_probe':
	drivers/dma/tegra20-apb-dma.c:1210:9: warning: assignment discards 'const' qualifier from pointer target type [enabled by default]

[ukl: split Laxman's patch to prevent warnings in the middle of the
series. Original patch was sent with msgid
1346217447-19730-1-git-send-email-ldewangan@nvidia.com]

Signed-off-by: Laxman Dewangan <ldewangan@nvidia.com>
Acked-by: Stephen Warren <redacted>
Acked-by: Vinod Koul <redacted>
Signed-off-by: Uwe Kleine-König <redacted>

Uwe,

I don't see these patches in linux-next, so compilation is still broken.
Can you please apply them? Thanks.

I have, but my new tree didn't make it into arm-soc. Arnd/Olof, can you
please refetch it?

Thanks
Uwe

-- 
Pengutronix e.K.                           | Uwe Kleine-König            |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |