From: grant.likely@secretlab.ca (Grant Likely)
To: linux-arm-kernel@lists.infradead.org
Subject: [RFC 1/2] ARM:Tegra: Device Tree Support: Initialize the audio card from the device tree.
Date: Fri, 27 May 2011 15:05:08 -0600 [thread overview]
Message-ID: <20110527210508.GK6645@ponder.secretlab.ca> (raw)
In-Reply-To: <20110527205706.21000.34832.stgit@riker>
On Fri, May 27, 2011 at 01:57:13PM -0700, John Bonesio wrote:
> This patch makes it so the top level audio card is initialized from the device
> tree. This is just the first step getting the audio complex of devices
> iniialized from device tree nodes.
>
> Signed-off-by: John Bonesio<bones@secretlab.ca>
> ---
>
> arch/arm/boot/dts/tegra-harmony.dts | 4 +++
> arch/arm/mach-tegra/board-dt.c | 3 ++
> sound/soc/tegra/harmony.c | 45 ++++++++++++++++++++++++++++++++---
> 3 files changed, 48 insertions(+), 4 deletions(-)
>
> diff --git a/arch/arm/boot/dts/tegra-harmony.dts b/arch/arm/boot/dts/tegra-harmony.dts
> index 05521a5..217a7f0 100644
> --- a/arch/arm/boot/dts/tegra-harmony.dts
> +++ b/arch/arm/boot/dts/tegra-harmony.dts
> @@ -53,6 +53,10 @@
> clock-frequency = <400000>;
> };
>
> + harmony_audio: audio_card {
> + compatible = "nvidia,harmony-audio";
> + };
> +
By convention, '_' is discouraged in the device tree names. Following
the generic names recommended practice, this node should be named
simply 'sound'.
You can also squash these two patches together I think; it will make
for a simpler patch overall. Doing so would get rid of the temporary
code that you need to resort to here.
> serial at 70006300 {
> status = "ok";
> clock-frequency = < 216000000 >;
> diff --git a/arch/arm/mach-tegra/board-dt.c b/arch/arm/mach-tegra/board-dt.c
> index c498e84..a569ad9 100644
> --- a/arch/arm/mach-tegra/board-dt.c
> +++ b/arch/arm/mach-tegra/board-dt.c
> @@ -56,6 +56,9 @@ static struct platform_device *tegra250_devices[] __initdata = {
> &tegra_i2c_device2,
> &tegra_i2c_device3,
> &tegra_i2c_device4,
> + &tegra_i2s_device1,
> + &tegra_das_device,
> + &tegra_pcm_device,
> };
>
> static __initdata struct tegra_clk_init_table tegra_dt_clk_init_table[] = {
> diff --git a/sound/soc/tegra/harmony.c b/sound/soc/tegra/harmony.c
> index f225087..faeec14 100644
> --- a/sound/soc/tegra/harmony.c
> +++ b/sound/soc/tegra/harmony.c
> @@ -52,6 +52,18 @@
>
> #define DRV_NAME "tegra-snd-harmony"
>
> +/* temporary - info will go into device tree */
> +#define TEGRA_GPIO_PW2 178
> +#define TEGRA_GPIO_PX0 184
> +#define TEGRA_GPIO_PX1 185
> +
> +#define HARMONY_GPIO_WM8903(_x_) (TEGRA_NR_GPIOS + (_x_))
> +#define TEGRA_GPIO_SPKR_EN HARMONY_GPIO_WM8903(2)
> +#define TEGRA_GPIO_HP_DET TEGRA_GPIO_PW2
> +#define TEGRA_GPIO_INT_MIC_EN TEGRA_GPIO_PX0
> +#define TEGRA_GPIO_EXT_MIC_EN TEGRA_GPIO_PX1
> +/* end temporary */
> +
> #define GPIO_SPKR_EN BIT(0)
> #define GPIO_INT_MIC_EN BIT(1)
> #define GPIO_EXT_MIC_EN BIT(2)
> @@ -287,6 +299,14 @@ static struct snd_soc_card snd_soc_harmony = {
> .num_links = 1,
> };
>
> +/* temporary - put this into the device tree */
> +static struct harmony_audio_platform_data harmony_audio_pdata = {
> + .gpio_spkr_en = TEGRA_GPIO_SPKR_EN,
> + .gpio_hp_det = TEGRA_GPIO_HP_DET,
> + .gpio_int_mic_en = TEGRA_GPIO_INT_MIC_EN,
> + .gpio_ext_mic_en = TEGRA_GPIO_EXT_MIC_EN,
> +};
> +
Yes, you should not need a static structure here. All the data is
dynamic.
> static __devinit int tegra_snd_harmony_probe(struct platform_device *pdev)
> {
> struct snd_soc_card *card = &snd_soc_harmony;
> @@ -307,10 +327,15 @@ static __devinit int tegra_snd_harmony_probe(struct platform_device *pdev)
> }
> #endif
>
> - pdata = pdev->dev.platform_data;
> - if (!pdata) {
> - dev_err(&pdev->dev, "no platform data supplied\n");
> - return -EINVAL;
> + if (pdev->dev.of_node) {
> + pdev->dev.platform_data = &harmony_audio_pdata; /* temporary */
> + pdata = pdev->dev.platform_data;
This is actually a bit dangerous, or at least a bad example. In the
dt case, the platform_data structure should be allocated per-instance.
> + } else {
> + pdata = pdev->dev.platform_data;
> + if (!pdata) {
> + dev_err(&pdev->dev, "no platform data supplied\n");
> + return -EINVAL;
> + }
> }
>
> harmony = kzalloc(sizeof(struct tegra_harmony), GFP_KERNEL);
> @@ -374,11 +399,23 @@ static int __devexit tegra_snd_harmony_remove(struct platform_device *pdev)
> return 0;
> }
>
> +#if defined(CONFIG_OF)
> +/* Match table for of_platform binding */
> +static const struct of_device_id harmony_of_match[] __devinitconst = {
> + { .compatible = "nvidia,harmony-audio", },
> + {},
> +};
> +MODULE_DEVICE_TABLE(of, harmony_of_match);
> +#endif
> +
> static struct platform_driver tegra_snd_harmony_driver = {
> .driver = {
> .name = DRV_NAME,
> .owner = THIS_MODULE,
> .pm = &snd_soc_pm_ops,
> +#if defined(CONFIG_OF)
> + .of_match_table = harmony_of_match,
> +#endif
The #ifdef() is no longer necessary here if you add an #else clause
defining harmony_of_match to NULL above.
In general, it is a good start.
g.
next prev parent reply other threads:[~2011-05-27 21:05 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-05-27 20:56 [RFC 0/2] ARM: Tegra: Device Tree: Audio John Bonesio
[not found] ` <20110527205721.21000.78599.stgit@riker>
2011-05-27 21:06 ` [RFC 2/2] ARM:Tegra: Device Tree Support: Initialize audio card gpio's from the device tree Grant Likely
2011-05-28 1:24 ` Mark Brown
2011-05-30 3:11 ` Olof Johansson
2011-05-30 3:38 ` Mark Brown
2011-05-30 6:11 ` Grant Likely
2011-05-30 6:18 ` Mitch Bradley
2011-05-30 6:22 ` Grant Likely
2011-05-30 7:01 ` Mark Brown
2011-05-30 16:22 ` Grant Likely
2011-05-30 18:54 ` Segher Boessenkool
2011-05-30 19:20 ` Grant Likely
2011-05-30 20:53 ` Mitch Bradley
2011-05-31 17:55 ` Stephen Warren
2011-05-31 18:42 ` Mitch Bradley
2011-06-01 15:59 ` Stephen Warren
2011-06-01 16:18 ` Mark Brown
2011-06-02 15:40 ` Grant Likely
2011-06-01 21:32 ` Mitch Bradley
2011-06-03 21:24 ` Stephen Warren
2011-06-04 0:25 ` Mitch Bradley
2011-06-02 14:59 ` Grant Likely
2011-06-02 15:40 ` Grant Likely
2011-06-28 21:39 ` Grant Likely
2011-05-30 23:27 ` Benjamin Herrenschmidt
2011-05-30 23:49 ` Olof Johansson
2011-05-31 0:58 ` Segher Boessenkool
2011-05-31 10:24 ` Mark Brown
2011-05-30 7:10 ` Mark Brown
2011-05-30 23:26 ` Benjamin Herrenschmidt
2011-05-31 10:03 ` Mark Brown
[not found] ` <20110527205706.21000.34832.stgit@riker>
2011-05-27 21:05 ` Grant Likely [this message]
2011-05-28 1:28 ` [RFC 1/2] ARM:Tegra: Device Tree Support: Initialize the audio card " Mark Brown
2011-06-01 7:07 ` Barry Song
2011-06-01 16:47 ` Grant Likely
2011-06-02 9:07 ` Barry Song
2011-06-02 16:04 ` Grant Likely
2011-06-02 16:21 ` Barry Song
2011-06-02 21:43 ` Russell King - ARM Linux
2011-06-03 2:32 ` Barry Song
2011-06-03 6:20 ` Russell King - ARM Linux
2011-06-02 21:36 ` Russell King - ARM Linux
2011-06-03 1:19 ` Barry Song
2011-06-07 3:44 ` Barry Song
2011-06-14 15:42 ` Grant Likely
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20110527210508.GK6645@ponder.secretlab.ca \
--to=grant.likely@secretlab.ca \
--cc=linux-arm-kernel@lists.infradead.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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).