linux-arm-kernel.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
From: lars@metafoo.de (Lars-Peter Clausen)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH v5 2/2] iio: twl6030-gpadc: TWL6030, TWL6032 GPADC driver
Date: Thu, 18 Jul 2013 11:48:10 +0200	[thread overview]
Message-ID: <51E7B9DA.9040803@metafoo.de> (raw)
In-Reply-To: <CAPb7_muAzO-jRqgcm89fcKT-hTLq38Aj2p0WAWjGjN9DoQUwPg@mail.gmail.com>

On 07/18/2013 10:36 AM, Oleksandr Kozaruk wrote:
> Hello Lars,
> 
> On Wed, Jul 17, 2013 at 9:04 PM, Lars-Peter Clausen <lars@metafoo.de> wrote:
>>> +static int twl6032_calibration(struct twl6030_gpadc_data *gpadc)
>>> +{
>>> +     int chn, d1 = 0, d2 = 0, temp;
>>> +     u8 trim_regs[17];
>>> +     int ret;
>>> +
>>> +     ret = twl_i2c_read(TWL6030_MODULE_ID2, trim_regs + 1,
>>> +                     TWL6030_GPADC_TRIM1, 16);
>>> +     if (ret < 0) {
>>> +             dev_err(gpadc->dev, "calibration failed\n");
>>> +             return ret;
>>> +     }
>>> +
>>> +     /*
>>> +      * Loop to calculate the value needed for returning voltages from
>>> +      * GPADC not values.
>>> +      *
>>> +      * gain is calculated to 3 decimal places fixed point.
>>> +      */
>>> +     for (chn = 0; chn < TWL6032_GPADC_MAX_CHANNELS; chn++) {
>>> +
>>> +             switch (chn) {
>>> +             case 0:
>>> +             case 1:
>>> +             case 2:
>>> +             case 3:
>>> +             case 4:
>>> +             case 5:
>>> +             case 6:
>>> +             case 11:
>>> +             case 12:
>>> +             case 13:
>>> +             case 14:
>>> +                     /* D1 */
>>> +                     d1 = (trim_regs[3] & 0x1F) << 2;
>>> +                     d1 |= (trim_regs[1] & 0x06) >> 1;
>>> +                     if (trim_regs[1] & 0x01)
>>> +                             d1 = -d1;
>>> +
>>> +                     /* D2 */
>>> +                     d2 = (trim_regs[4] & 0x3F) << 2;
>>> +                     d2 |= (trim_regs[2] & 0x06) >> 1;
>>> +                     if (trim_regs[2] & 0x01)
>>> +                             d2 = -d2;
>>> +                     break;
>>> +             case 8:
>>> +                     /* D1 */
>>> +                     temp = (trim_regs[3] & 0x1F) << 2;
>>> +                     temp |= (trim_regs[1] & 0x06) >> 1;
>>> +                     if (trim_regs[1] & 0x01)
>>> +                             temp = -temp;
>>> +
>>> +                     d1 = (trim_regs[8] & 0x18) << 1;
>>> +                     d1 |= (trim_regs[7] & 0x1E) >> 1;
>>> +                     if (trim_regs[7] & 0x01)
>>> +                             d1 = -d1;
>>> +
>>> +                     d1 += temp;
>>> +
>>> +                     /* D2 */
>>> +                     temp = (trim_regs[4] & 0x3F) << 2;
>>> +                     temp |= (trim_regs[2] & 0x06) >> 1;
>>> +                     if (trim_regs[2] & 0x01)
>>> +                             temp = -temp;
>>> +
>>> +                     d2 = (trim_regs[10] & 0x1F) << 2;
>>> +                     d2 |= (trim_regs[8] & 0x06) >> 1;
>>> +                     if (trim_regs[8] & 0x01)
>>> +                             d2 = -d2;
>>> +
>>> +                     d2 += temp;
>>> +                     break;
>>> +             case 9:
>>> +                     /* D1 */
>>> +                     temp = (trim_regs[3] & 0x1F) << 2;
>>> +                     temp |= (trim_regs[1] & 0x06) >> 1;
>>> +                     if (trim_regs[1] & 0x01)
>>> +                             temp = -temp;
>>> +
>>> +                     d1 = (trim_regs[14] & 0x18) << 1;
>>> +                     d1 |= (trim_regs[12] & 0x1E) >> 1;
>>> +                     if (trim_regs[12] & 0x01)
>>> +                             d1 = -d1;
>>> +
>>> +                     d1 += temp;
>>> +
>>> +                     /* D2 */
>>> +                     temp = (trim_regs[4] & 0x3F) << 2;
>>> +                     temp |= (trim_regs[2] & 0x06) >> 1;
>>> +                     if (trim_regs[2] & 0x01)
>>> +                             temp = -temp;
>>> +
>>> +                     d2 = (trim_regs[16] & 0x1F) << 2;
>>> +                     d2 |= (trim_regs[14] & 0x06) >> 1;
>>> +                     if (trim_regs[14] & 0x01)
>>> +                             d2 = -d2;
>>> +
>>> +                     d2 += temp;
>>> +             case 10:
>>> +                     /* D1 */
>>> +                     d1 = (trim_regs[11] & 0x0F) << 3;
>>> +                     d1 |= (trim_regs[9] & 0x0E) >> 1;
>>> +                     if (trim_regs[9] & 0x01)
>>> +                             d1 = -d1;
>>> +
>>> +                     /* D2 */
>>> +                     d2 = (trim_regs[15] & 0x0F) << 3;
>>> +                     d2 |= (trim_regs[13] & 0x0E) >> 1;
>>> +                     if (trim_regs[13] & 0x01)
>>> +                             d2 = -d2;
>>> +                     break;
>>> +             case 7:
>>> +             case 18:
>>> +                     /* D1 */
>>> +                     temp = (trim_regs[3] & 0x1F) << 2;
>>> +                     temp |= (trim_regs[1] & 0x06) >> 1;
>>> +                     if (trim_regs[1] & 0x01)
>>> +                             temp = -temp;
>>> +
>>> +                     d1 = (trim_regs[5] & 0x7E) >> 1;
>>> +                     if (trim_regs[5] & 0x01)
>>> +                             d1 = -d1;
>>> +.
>>> +                     d1 += temp;
>>> +
>>> +                     /* D2 */
>>> +                     temp = (trim_regs[4] & 0x3F) << 2;
>>> +                     temp |= (trim_regs[2] & 0x06) >> 1;
>>> +                     if (trim_regs[2] & 0x01)
>>> +                             temp = -temp;
>>> +
>>> +                     d2 = (trim_regs[6] & 0xFE) >> 1;
>>> +                     if (trim_regs[6] & 0x01)
>>> +                             d2 = -d2;
>>> +
>>> +                     d2 += temp;
>>> +                     break;
>>
>>
>> There is quite a bit of copy paste in here. Putting the bit swizziling into
>> a helper function should also improbe legibility.
> 
> Trimming bits are scattered randomly between registers and in
> different bit positions.
> I don't see any commonalities, that could be generalized. as Graeme
> said about it "it's a bane of my life".

The pattern is pretty regular though, how about:


int twl6032_get_trim_value(u8 *trim_regs, unsigned int reg0,
	unsigned int reg1, unsigned int mask0, unsigned int mask1,
	unsigned int shift0)
{
	int val;

	val = (trim_regs[reg0] & mask0) << shift0;
	val |= (trim_regs[reg1] & mask1) >> 1;
	if (trim_regs[reg1] & 0x01)
		val = -val;

	return val;
}

and then

	d1 = twl6032_get_trim_value(trim_regs, 3, 1, 0x1f, 0x06, 2);
	d2 = twl6032_get_trim_value(trim_regs, 4, 2, 0x3f, 0x06, 2);

and so on.

  reply	other threads:[~2013-07-18  9:48 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-07-17 11:12 [PATCH v5 0/2] TWL6030, TWL6032 GPADC driver Oleksandr Kozaruk
2013-07-17 11:12 ` [PATCH v5 1/2] ARM: dts: twl: Add GPADC data to device tree Oleksandr Kozaruk
2013-07-17 14:33   ` Sergei Shtylyov
2013-07-17 14:48     ` Lars-Peter Clausen
2013-07-17 11:12 ` [PATCH v5 2/2] iio: twl6030-gpadc: TWL6030, TWL6032 GPADC driver Oleksandr Kozaruk
2013-07-17 18:04   ` Lars-Peter Clausen
2013-07-18  8:36     ` Oleksandr Kozaruk
2013-07-18  9:48       ` Lars-Peter Clausen [this message]
2013-07-19  8:35         ` Oleksandr Kozaruk

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=51E7B9DA.9040803@metafoo.de \
    --to=lars@metafoo.de \
    --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).