public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Philipp Zabel <philipp.zabel@gmail.com>
To: LKML <linux-kernel@vger.kernel.org>
Cc: Samuel Ortiz <sameo@openedhand.com>
Subject: Re: [PATCH] mfd: let asic3 use mem resource instead of bus_shift
Date: Sun, 06 Jul 2008 17:05:54 +0200	[thread overview]
Message-ID: <1215356754.1672.1.camel@localhost.localdomain> (raw)
In-Reply-To: <1215289200.23398.40.camel@localhost.localdomain>

Am Samstag, den 05.07.2008, 22:20 +0200 schrieb Philipp Zabel:
> The bus_shift parameter in platform_data is not needed
> as we can tell the driver with the IOMEM_RESOURCE whether
> the ASIC is located on a 16bit or 32bit memory bus.
> 
> bus_shift as a configuration parameter is confusing at best
> because sometimes it means "shift register number left by n bits",
> sometimes it means "shift 32bit register address right by n bits"
> (see asic->bus_shift = 2 - pdata->bus_shift ...).
> 
> The htc-egpio driver uses a more descriptive bus_width parameter,
> but for drivers where the register map size is fixed, we don't even
> need this. With this patch, the asic3 driver will calculate its
> bus_shift from the memory resource size (ASIC3_MAP_SIZE_16BIT or
> ASIC3_MAP_SIZE_32BIT).
> 
> Signed-off-by: Philipp Zabel <philipp.zabel@gmail.com>
> ---
>  drivers/mfd/asic3.c       |   11 +++++------
>  include/linux/mfd/asic3.h |    3 ++-
>  2 files changed, 7 insertions(+), 7 deletions(-)
> 
> diff --git a/drivers/mfd/asic3.c b/drivers/mfd/asic3.c
> index f2fb233..3b870e7 100644
> --- a/drivers/mfd/asic3.c
> +++ b/drivers/mfd/asic3.c
> @@ -313,6 +313,7 @@ static int __init asic3_irq_probe(struct platform_device *pdev)
>  	struct asic3 *asic = platform_get_drvdata(pdev);
>  	unsigned long clksel = 0;
>  	unsigned int irq, irq_base;
> +	int map_size;
>  
>  	asic->irq_nr = platform_get_irq(pdev, 0);
>  	if (asic->irq_nr < 0)
> @@ -551,8 +552,8 @@ static int __init asic3_probe(struct platform_device *pdev)
>  		goto out_free;
>  	}
>  
> -
> -	asic->mapping = ioremap(mem->start, PAGE_SIZE);
> +	map_size = mem->end - mem->start + 1;
> +	asic->mapping = ioremap(mem->start, map_size);
>  	if (!asic->mapping) {
>  		ret = -ENOMEM;
>  		dev_err(asic->dev, "Couldn't ioremap\n");
> @@ -561,10 +562,8 @@ static int __init asic3_probe(struct platform_device *pdev)
>  
>  	asic->irq_base = pdata->irq_base;
>  
> -	if (pdata && pdata->bus_shift)
> -		asic->bus_shift = 2 - pdata->bus_shift;
> -	else
> -		asic->bus_shift = 0;
> +	/* calculate bus shift from mem resource */
> +	asic->bus_shift = 2 - (map_size >> 12);
>  
>  	clksel = 0;
>  	asic3_write_register(asic, ASIC3_OFFSET(CLOCK, SEL), clksel);
> diff --git a/include/linux/mfd/asic3.h b/include/linux/mfd/asic3.h
> index 8f8c46c..2de526a 100644
> --- a/include/linux/mfd/asic3.h
> +++ b/include/linux/mfd/asic3.h
> @@ -498,6 +498,7 @@ struct asic3_platform_data {
>  #define ASIC3_SDIO_CTRL_LEDCtrl              0x7C
>  #define ASIC3_SDIO_CTRL_SoftwareReset        0x1C0
>  
> -#define ASIC3_MAP_SIZE	     		     0x2000
> +#define ASIC3_MAP_SIZE_32BIT	0x2000
> +#define ASIC3_MAP_SIZE_16BIT	0x1000
>  
>  #endif /* __ASIC3_H__ */

Second try. I forgot to remove bus_shift struct asic3_platform_data
before.

regards
Philipp

---

Subject: [PATCH] mfd: let asic3 use mem resource instead of bus_shift

The bus_shift parameter in platform_data is not needed
as we can tell the driver with the IOMEM_RESOURCE whether
the ASIC is located on a 16bit or 32bit memory bus.

The htc-egpio driver uses a more descriptive bus_width parameter,
but for drivers where the register map size fixed, we don't even
need this.
bus_shift as a configuration parameter is confusing at best
because sometimes it means "shift register number left by n bits",
sometimes it means "shift 32bit register address right by n bits"
(see asic->bus_shift = 2 - pdata->bus_shift ...).

Signed-off-by: Philipp Zabel <philipp.zabel@gmail.com>
---
 drivers/mfd/asic3.c       |   11 +++++------
 include/linux/mfd/asic3.h |    5 ++---
 2 files changed, 7 insertions(+), 9 deletions(-)

diff --git a/drivers/mfd/asic3.c b/drivers/mfd/asic3.c
index f2fb233..3b870e7 100644
--- a/drivers/mfd/asic3.c
+++ b/drivers/mfd/asic3.c
@@ -313,6 +313,7 @@ static int __init asic3_irq_probe(struct platform_device *pdev)
 	struct asic3 *asic = platform_get_drvdata(pdev);
 	unsigned long clksel = 0;
 	unsigned int irq, irq_base;
+	int map_size;
 
 	asic->irq_nr = platform_get_irq(pdev, 0);
 	if (asic->irq_nr < 0)
@@ -551,8 +552,8 @@ static int __init asic3_probe(struct platform_device *pdev)
 		goto out_free;
 	}
 
-
-	asic->mapping = ioremap(mem->start, PAGE_SIZE);
+	map_size = mem->end - mem->start + 1;
+	asic->mapping = ioremap(mem->start, map_size);
 	if (!asic->mapping) {
 		ret = -ENOMEM;
 		dev_err(asic->dev, "Couldn't ioremap\n");
@@ -561,10 +562,8 @@ static int __init asic3_probe(struct platform_device *pdev)
 
 	asic->irq_base = pdata->irq_base;
 
-	if (pdata && pdata->bus_shift)
-		asic->bus_shift = 2 - pdata->bus_shift;
-	else
-		asic->bus_shift = 0;
+	/* calculate bus shift from mem resource */
+	asic->bus_shift = 2 - (map_size >> 12);
 
 	clksel = 0;
 	asic3_write_register(asic, ASIC3_OFFSET(CLOCK, SEL), clksel);
diff --git a/include/linux/mfd/asic3.h b/include/linux/mfd/asic3.h
index 8f8c46c..322cd6d 100644
--- a/include/linux/mfd/asic3.h
+++ b/include/linux/mfd/asic3.h
@@ -20,8 +20,6 @@ struct asic3_platform_data {
 	u16 *gpio_config;
 	unsigned int gpio_config_num;
 
-	unsigned int bus_shift;
-
 	unsigned int irq_base;
 
 	unsigned int gpio_base;
@@ -498,6 +496,7 @@ struct asic3_platform_data {
 #define ASIC3_SDIO_CTRL_LEDCtrl              0x7C
 #define ASIC3_SDIO_CTRL_SoftwareReset        0x1C0
 
-#define ASIC3_MAP_SIZE	     		     0x2000
+#define ASIC3_MAP_SIZE_32BIT	0x2000
+#define ASIC3_MAP_SIZE_16BIT	0x1000
 
 #endif /* __ASIC3_H__ */
-- 
1.5.6



  reply	other threads:[~2008-07-06 15:06 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-07-05 20:20 [PATCH] mfd: let asic3 use mem resource instead of bus_shift Philipp Zabel
2008-07-06 15:05 ` Philipp Zabel [this message]
2008-07-10  0:30   ` Samuel Ortiz

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=1215356754.1672.1.camel@localhost.localdomain \
    --to=philipp.zabel@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=sameo@openedhand.com \
    /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