[PATCH 1/2] ASoC: Ux500: Fixup use of clocks

[PATCH 1/2] ASoC: Ux500: Fixup use of clocks

[Ulf Hansson]

From: Ulf Hansson <ulf.hansson@linaro.org>

Make sure clocks are being prepared and unprepared as well
as enabled and disabled.

Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
---
 sound/soc/ux500/ux500_msp_dai.c |   12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

diff --git a/sound/soc/ux500/ux500_msp_dai.c b/sound/soc/ux500/ux500_msp_dai.c
index be94bf9..e11187f 100644
--- a/sound/soc/ux500/ux500_msp_dai.c
+++ b/sound/soc/ux500/ux500_msp_dai.c
@@ -398,11 +398,13 @@ static int ux500_msp_dai_startup(struct snd_pcm_substream *substream,
 		return ret;
 	}
 
-	/* Enable clock */
+	/* Prepare and enable clock */
 	dev_dbg(dai->dev, "%s: Enabling MSP-clock.\n", __func__);
-	clk_enable(drvdata->clk);
+	ret = clk_prepare_enable(drvdata->clk);
+	if (ret)
+		regulator_disable(drvdata->reg_vape);
 
-	return 0;
+	return ret;
 }
 
 static void ux500_msp_dai_shutdown(struct snd_pcm_substream *substream,
@@ -428,8 +430,8 @@ static void ux500_msp_dai_shutdown(struct snd_pcm_substream *substream,
 			__func__, dai->id, snd_pcm_stream_str(substream));
 	}
 
-	/* Disable clock */
-	clk_disable(drvdata->clk);
+	/* Disable and unprepare clock */
+	clk_disable_unprepare(drvdata->clk);
 
 	/* Disable regulator */
 	ret = regulator_disable(drvdata->reg_vape);
-- 
1.7.10

[PATCH 2/2] ASoC: Ux500: Control apb clock

[Ulf Hansson]

From: Ulf Hansson <ulf.hansson@linaro.org>

When switching to common clock driver for ux500 this clock needs to
be handled as well. Before this clock was internally managed by the
clock driver itself.

Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
---
 sound/soc/ux500/ux500_msp_dai.c |   38 ++++++++++++++++++++++++++++++++------
 sound/soc/ux500/ux500_msp_dai.h |    1 +
 2 files changed, 33 insertions(+), 6 deletions(-)

diff --git a/sound/soc/ux500/ux500_msp_dai.c b/sound/soc/ux500/ux500_msp_dai.c
index e11187f..74bb3c0 100644
--- a/sound/soc/ux500/ux500_msp_dai.c
+++ b/sound/soc/ux500/ux500_msp_dai.c
@@ -398,13 +398,28 @@ static int ux500_msp_dai_startup(struct snd_pcm_substream *substream,
 		return ret;
 	}
 
-	/* Prepare and enable clock */
-	dev_dbg(dai->dev, "%s: Enabling MSP-clock.\n", __func__);
+	/* Prepare and enable clocks */
+	dev_dbg(dai->dev, "%s: Enabling MSP-clocks.\n", __func__);
+	ret = clk_prepare_enable(drvdata->pclk);
+	if (ret) {
+		dev_err(drvdata->msp->dev,
+			"%s: Failed to prepare/enable pclk!\n", __func__);
+		goto err_pclk;
+	}
+
 	ret = clk_prepare_enable(drvdata->clk);
-	if (ret)
-		regulator_disable(drvdata->reg_vape);
+	if (ret) {
+		dev_err(drvdata->msp->dev,
+			"%s: Failed to prepare/enable clk!\n", __func__);
+		goto err_clk;
+	}
 
 	return ret;
+err_clk:
+	clk_disable_unprepare(drvdata->pclk);
+err_pclk:
+	regulator_disable(drvdata->reg_vape);
+	return ret;
 }
 
 static void ux500_msp_dai_shutdown(struct snd_pcm_substream *substream,
@@ -430,8 +445,9 @@ static void ux500_msp_dai_shutdown(struct snd_pcm_substream *substream,
 			__func__, dai->id, snd_pcm_stream_str(substream));
 	}
 
-	/* Disable and unprepare clock */
+	/* Disable and unprepare clocks */
 	clk_disable_unprepare(drvdata->clk);
+	clk_disable_unprepare(drvdata->pclk);
 
 	/* Disable regulator */
 	ret = regulator_disable(drvdata->reg_vape);
@@ -782,6 +798,14 @@ static int __devinit ux500_msp_drv_probe(struct platform_device *pdev)
 	}
 	prcmu_qos_add_requirement(PRCMU_QOS_APE_OPP, (char *)pdev->name, 50);
 
+	drvdata->pclk = clk_get(&pdev->dev, "apb_pclk");
+	if (IS_ERR(drvdata->pclk)) {
+		ret = (int)PTR_ERR(drvdata->pclk);
+		dev_err(&pdev->dev, "%s: ERROR: clk_get of pclk failed (%d)!\n",
+			__func__, ret);
+		goto err_pclk;
+	}
+
 	drvdata->clk = clk_get(&pdev->dev, NULL);
 	if (IS_ERR(drvdata->clk)) {
 		ret = (int)PTR_ERR(drvdata->clk);
@@ -812,8 +836,9 @@ static int __devinit ux500_msp_drv_probe(struct platform_device *pdev)
 
 err_init_msp:
 	clk_put(drvdata->clk);
-
 err_clk:
+	clk_put(drvdata->pclk);
+err_pclk:
 	devm_regulator_put(drvdata->reg_vape);
 
 	return ret;
@@ -829,6 +854,7 @@ static int __devexit ux500_msp_drv_remove(struct platform_device *pdev)
 	prcmu_qos_remove_requirement(PRCMU_QOS_APE_OPP, "ux500_msp_i2s");
 
 	clk_put(drvdata->clk);
+	clk_put(drvdata->pclk);
 
 	ux500_msp_i2s_cleanup_msp(pdev, drvdata->msp);
 
diff --git a/sound/soc/ux500/ux500_msp_dai.h b/sound/soc/ux500/ux500_msp_dai.h
index 98202a3..9c778d9 100644
--- a/sound/soc/ux500/ux500_msp_dai.h
+++ b/sound/soc/ux500/ux500_msp_dai.h
@@ -69,6 +69,7 @@ struct ux500_msp_i2s_drvdata {
 	/* Clocks */
 	unsigned int master_clk;
 	struct clk *clk;
+	struct clk *pclk;
 
 	/* Regulators */
 	int vape_opp_constraint;
-- 
1.7.10

[PATCH 2/2] ASoC: Ux500: Control apb clock

[Linus Walleij]

On Mon, Oct 22, 2012 at 2:32 PM, Ulf Hansson <ulf.hansson@stericsson.com> wrote:

> From: Ulf Hansson <ulf.hansson@linaro.org>
>
> When switching to common clock driver for ux500 this clock needs to
> be handled as well. Before this clock was internally managed by the
> clock driver itself.
>
> Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>

Looks correct, Acked-by: Linus Walleij <linus.walleij@linaro.org>

Yours,
Linus Walleij

[PATCH 2/2] ASoC: Ux500: Control apb clock

[Mark Brown]

On Mon, Oct 22, 2012 at 02:32:05PM +0200, Ulf Hansson wrote:
> From: Ulf Hansson <ulf.hansson@linaro.org>
> 
> When switching to common clock driver for ux500 this clock needs to
> be handled as well. Before this clock was internally managed by the
> clock driver itself.

Applied but why isn't this a deficiency in the common clock
implementation for the platform?
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20121022/fe94e73f/attachment-0001.sig>

[PATCH 2/2] ASoC: Ux500: Control apb clock

[Ulf Hansson]

On 22 October 2012 15:32, Mark Brown
<broonie@opensource.wolfsonmicro.com> wrote:
> On Mon, Oct 22, 2012 at 02:32:05PM +0200, Ulf Hansson wrote:
>> From: Ulf Hansson <ulf.hansson@linaro.org>
>>
>> When switching to common clock driver for ux500 this clock needs to
>> be handled as well. Before this clock was internally managed by the
>> clock driver itself.
>
> Applied but why isn't this a deficiency in the common clock
> implementation for the platform?

Good question. :-) Patches have just been sent to Mike Turquette clk
tree for adding the clock lookups.
First, I thought I might wanted to group them all in a series but
decided that splitting them are probably easier to handle.

Kind regards
Ulf Hansson

[PATCH 1/2] ASoC: Ux500: Fixup use of clocks

[Linus Walleij]

On Mon, Oct 22, 2012 at 2:32 PM, Ulf Hansson <ulf.hansson@stericsson.com> wrote:

> From: Ulf Hansson <ulf.hansson@linaro.org>
>
> Make sure clocks are being prepared and unprepared as well
> as enabled and disabled.
>
> Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>

Acked-by: Linus Walleij <linus.walleij@linaro.org>

Thanks,
Linus Walleij

[PATCH 1/2] ASoC: Ux500: Fixup use of clocks

[Mark Brown]

On Mon, Oct 22, 2012 at 02:32:04PM +0200, Ulf Hansson wrote:
> From: Ulf Hansson <ulf.hansson@linaro.org>
> 
> Make sure clocks are being prepared and unprepared as well
> as enabled and disabled.

Applied, thanks.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20121022/e8e52d11/attachment.sig>

[PATCH 1/2] ASoC: Ux500: Fixup use of clocks

[Ulf Hansson]

On 22 October 2012 15:30, Mark Brown
<broonie@opensource.wolfsonmicro.com> wrote:
> On Mon, Oct 22, 2012 at 02:32:04PM +0200, Ulf Hansson wrote:
>> From: Ulf Hansson <ulf.hansson@linaro.org>
>>
>> Make sure clocks are being prepared and unprepared as well
>> as enabled and disabled.
>
> Applied, thanks.

I can not find this patch on any "next-tree" yet. Same goes for:
"[PATCH 2/2] ASoC: Ux500: Control apb clock".

Maybe I should be more patient, but thought it make sense to send a
ping on this. :-)

Kind regards
Ulf Hansson

[PATCH 1/2] ASoC: Ux500: Fixup use of clocks

[Mark Brown]

On Mon, Nov 05, 2012 at 03:27:08PM +0100, Ulf Hansson wrote:

> I can not find this patch on any "next-tree" yet. Same goes for:
> "[PATCH 2/2] ASoC: Ux500: Control apb clock".

> Maybe I should be more patient, but thought it make sense to send a
> ping on this. :-)

As I've had to tell you before pings are pointless, don't do this - to
repeat, if you think something has been missed resend it.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20121106/d3358676/attachment.sig>