linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <greg@kroah.com>
To: Johannes Poehlmann <johannes.poehlmann@izt-labs.de>
Cc: linux-kernel@vger.kernel.org, Evgeniy Polyakov <zbr@ioremap.net>
Subject: Re: [PATCH v3 1/4] w1: ds1wm: fix and simplify register access
Date: Tue, 18 Jul 2017 16:15:03 +0200	[thread overview]
Message-ID: <20170718141502.GA18857@kroah.com> (raw)
In-Reply-To: <1500377213-1117-2-git-send-email-johannes.poehlmann@izt-labs.de>

On Tue, Jul 18, 2017 at 01:26:50PM +0200, Johannes Poehlmann wrote:
> o Replace incorrect register offsett calculation by
> direct configuration of bus_shift in mfd-cell.
> Indirect definition of address-shift by resource size
> was unobvious and should have used a binary log.
> o Make endian clean, make HW-endianness configurable.
> o Use ioread*, iowrite* instead of __raw_readb,__raw_writeb
> to also use memory-barriers when accessing HW-registers.
> We do not want reordering to happen here.

Can you format the above to make it a bit more easier to read?

And why do all of this in one single patch?  Shouldn't this be multiple
ones?

> 
> Signed-off-by: Johannes Poehlmann <johannes.poehlmann@izt-labs.de>
> Acked-by: Evgeniy Polyakov <zbr@ioremap.net>
> ---
>  drivers/w1/masters/ds1wm.c | 84 ++++++++++++++++++++++++++++++++++++++++++----
>  include/linux/mfd/ds1wm.h  |  9 +++++
>  2 files changed, 87 insertions(+), 6 deletions(-)
> 
> diff --git a/drivers/w1/masters/ds1wm.c b/drivers/w1/masters/ds1wm.c
> index fd2e9da..6bba2fe 100644
> --- a/drivers/w1/masters/ds1wm.c
> +++ b/drivers/w1/masters/ds1wm.c
> @@ -95,7 +95,8 @@ static struct {
>  
>  struct ds1wm_data {
>  	void     __iomem *map;
> -	int      bus_shift; /* # of shifts to calc register offsets */
> +	unsigned int      bus_shift; /* # of shifts to calc register offsets */
> +	int      is_hw_big_endian;

bool?



>  	struct platform_device *pdev;
>  	const struct mfd_cell   *cell;
>  	int      irq;
> @@ -115,12 +116,66 @@ struct ds1wm_data {
>  static inline void ds1wm_write_register(struct ds1wm_data *ds1wm_data, u32 reg,
>  					u8 val)
>  {
> -	__raw_writeb(val, ds1wm_data->map + (reg << ds1wm_data->bus_shift));
> +	if (ds1wm_data->is_hw_big_endian) {
> +		switch (ds1wm_data->bus_shift) {
> +		case 0:
> +			iowrite8(val, ds1wm_data->map + (reg << 0));
> +			break;
> +		case 1:
> +			iowrite16be((u16)val, ds1wm_data->map+(reg<<1));
> +			break;
> +		case 2:
> +			iowrite32be((u32)val, ds1wm_data->map+(reg<<2));
> +			break;
> +		}
> +	} else {
> +		switch (ds1wm_data->bus_shift) {
> +		case 0:
> +			iowrite8(val, ds1wm_data->map + (reg << 0));
> +			break;
> +		case 1:
> +			iowrite16((u16) val, ds1wm_data->map+(reg << 1));
> +			break;
> +		case 2:
> +			iowrite32((u32) val, ds1wm_data->map+(reg << 2));
> +			break;
> +		}
> +	}
>  }
>  
>  static inline u8 ds1wm_read_register(struct ds1wm_data *ds1wm_data, u32 reg)
>  {
> -	return __raw_readb(ds1wm_data->map + (reg << ds1wm_data->bus_shift));
> +
> +	u32 val = 0;
> +
> +	if (ds1wm_data->is_hw_big_endian) {
> +		switch (ds1wm_data->bus_shift) {
> +		case 0:
> +			val = ioread8(ds1wm_data->map + (reg << 0));
> +			break;
> +		case 1:
> +			val = ioread16be(ds1wm_data->map + (reg << 1));
> +			break;
> +		case 2:
> +			val = ioread32be(ds1wm_data->map + (reg << 2));
> +			break;
> +		}
> +	} else {
> +		switch (ds1wm_data->bus_shift) {
> +		case 0:
> +			val = ioread8(ds1wm_data->map + (reg << 0));
> +			break;
> +		case 1:
> +			val = ioread16(ds1wm_data->map + (reg << 1));
> +			break;
> +		case 2:
> +			val = ioread32(ds1wm_data->map + (reg << 2));
> +			break;
> +		}
> +	}
> +	dev_dbg(&ds1wm_data->pdev->dev,
> +		"ds1wm_read_register reg: %d, 32 bit val:%x\n", reg, val);
> +	return (u8) val;
>  }
>  
>  
> @@ -473,9 +528,6 @@ static int ds1wm_probe(struct platform_device *pdev)
>  	if (!ds1wm_data->map)
>  		return -ENOMEM;
>  
> -	/* calculate bus shift from mem resource */
> -	ds1wm_data->bus_shift = resource_size(res) >> 3;
> -
>  	ds1wm_data->pdev = pdev;
>  	ds1wm_data->cell = mfd_get_cell(pdev);
>  	if (!ds1wm_data->cell)
> @@ -484,6 +536,26 @@ static int ds1wm_probe(struct platform_device *pdev)
>  	if (!plat)
>  		return -ENODEV;
>  
> +	/* how many bits to shift register number to get register offset */
> +	if (plat->bus_shift > 2) {
> +		dev_err(&ds1wm_data->pdev->dev,
> +			"illegal bus shift %d, not written",
> +			ds1wm_data->bus_shift);
> +		return -EINVAL;
> +	}
> +
> +	ds1wm_data->bus_shift = plat->bus_shift;
> +	/* make sure resource has space for 8 registers */
> +	if ((8 << ds1wm_data->bus_shift) > resource_size(res)) {
> +		dev_err(&ds1wm_data->pdev->dev,
> +			"memory resource size %d to small, should be %d\n",
> +			(int) resource_size(res),
> +			8 << ds1wm_data->bus_shift);
> +		return -EINVAL;
> +	}
> +
> +	ds1wm_data->is_hw_big_endian = plat->is_hw_big_endian;
> +
>  	res = platform_get_resource(pdev, IORESOURCE_IRQ, 0);
>  	if (!res)
>  		return -ENXIO;
> diff --git a/include/linux/mfd/ds1wm.h b/include/linux/mfd/ds1wm.h
> index 38a372a..4efd626 100644
> --- a/include/linux/mfd/ds1wm.h
> +++ b/include/linux/mfd/ds1wm.h
> @@ -3,6 +3,7 @@
>  struct ds1wm_driver_data {
>  	int active_high;
>  	int clock_rate;
> +
>  	/* in milliseconds, the amount of time to */
>  	/* sleep following a reset pulse. Zero    */
>  	/* should work if your bus devices recover*/
> @@ -10,4 +11,12 @@ struct ds1wm_driver_data {
>  	/* ds1wm implements the precise timings of*/
>  	/* a reset pulse/presence detect sequence.*/
>  	unsigned int reset_recover_delay;
> +
> +	/* Say 1 here for big endian Hardware */
> +	/* (only relevant with bus-shift > 0 */
> +	int is_hw_big_endian;

bool?

And who is setting this?  What is the initial value?

> +
> +	/* left shift of register number to get register address offsett */
> +	/* only 0,1,2 allowed for 8,16 or 32 bit bus width respectively */
> +	unsigned int bus_shift;

u8?

Please use multi-line comments in the correct format, checkpatch.pl
should have complained about these, right?

thanks,

greg k-h

  reply	other threads:[~2017-07-18 14:15 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-06-19 11:38 [PATCH V1] one wire ds1wm patch Johannes Pöhlmann
2017-06-19 14:55 ` Evgeniy Polyalkov
2017-06-19 15:24   ` Greg Kroah-Hartman
2017-06-19 17:38 ` kbuild test robot
     [not found] ` <1498214835-11186-1-git-send-email-johannes.poehlmann@izt-labs.de>
2017-06-29 13:32   ` [PATCH v2 0/4] w1: ds1wm: register access patch Evgeniy Polyakov
     [not found] ` <1498214835-11186-4-git-send-email-johannes.poehlmann@izt-labs.de>
2017-07-17 14:55   ` [PATCH v2 3/4] w1: ds1wm: silence interrupts on HW before claiming the interrupt Greg Kroah-Hartman
     [not found] ` <1498214835-11186-5-git-send-email-johannes.poehlmann@izt-labs.de>
2017-07-17 14:55   ` [PATCH v2 4/4] w1: ds1wm: add messages to make incorporation in mfd-drivers easier Greg Kroah-Hartman
     [not found] ` <1498214835-11186-2-git-send-email-johannes.poehlmann@izt-labs.de>
2017-07-17 14:55   ` [PATCH v2 1/4] w1: ds1wm: fix and simplify register access Greg Kroah-Hartman
2017-07-18 11:26 ` [PATCH v3 0/4] w1: ds1wm: register access patch Johannes Poehlmann
2017-07-18 11:26 ` [PATCH v3 1/4] w1: ds1wm: fix and simplify register access Johannes Poehlmann
2017-07-18 14:15   ` Greg Kroah-Hartman [this message]
2017-07-25 11:27     ` [PATCH v4 0/5] w1: ds1wm: register access patch Johannes Poehlmann
2017-08-27  7:19       ` Evgeniy Polyakov
2017-07-25 11:27     ` [PATCH v4 1/5] w1: ds1wm: fix register offset (bus shift) calculation Johannes Poehlmann
2017-07-25 11:27     ` [PATCH v4 2/5] w1: ds1wm: make endian clean and use standard io memory accessors Johannes Poehlmann
2017-07-25 11:27     ` [PATCH v4 3/5] w1: ds1wm: add level interrupt modes Johannes Poehlmann
2017-07-25 11:27     ` [PATCH v4 4/5] w1: ds1wm: silence interrupts on HW before claiming the interrupt Johannes Poehlmann
2017-07-25 11:27     ` [PATCH v4 5/5] w1: ds1wm: add messages to make incorporation in mfd-drivers easier Johannes Poehlmann
2017-07-18 11:26 ` [PATCH v3 2/4] w1: ds1wm: add level interrupt modes Johannes Poehlmann
2017-07-18 14:15   ` Greg Kroah-Hartman
2017-07-18 11:26 ` [PATCH v3 3/4] w1: ds1wm: silence interrupts on HW before claiming the interrupt Johannes Poehlmann
2017-07-18 14:15   ` Greg Kroah-Hartman
2017-07-18 11:26 ` [PATCH v3 4/4] w1: ds1wm: add messages to make incorporation in mfd-drivers easier Johannes Poehlmann
2017-07-18 14:16   ` Greg Kroah-Hartman

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=20170718141502.GA18857@kroah.com \
    --to=greg@kroah.com \
    --cc=johannes.poehlmann@izt-labs.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=zbr@ioremap.net \
    /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).