Skip to contentSkip to conversation

iio:gyro:mpu3050 Treat otp value as a __le64 and use FIELD_GET() to break up

PATCH Superseded Jonathan Cameron 3 participants 5 messages Nov 28, 2020 – Nov 29, 2020

Subsystems (2)

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore

Jonathan Cameron

author maintainer
#0 lore raw

PATCH 1/1 iio:gyro:mpu3050 Treat otp value as a __le64 and use FIELD_GET() to break up

From: Jonathan Cameron <[email protected]>

Inspired by Andy Shevchenko's proposal to use get_unaligned_leXX().

The whole one time programable memory is treated as a single 64bit little endian value. Thus we can avoid a lot of messy handling of fields overlapping byte boundaries by just loading and manipulating it as an __le64 converted to a u64. That lets us just use FIELD_GET() and GENMASK() to extract the values desired.

Note only build tested.

Cc Andy Shevchenko <[email protected]>
Cc Linus Walleij <[email protected]>
---
 drivers/iio/gyro/mpu3050-core.c | 28 ++++++++++++++++------------
 1 file changed, 16 insertions(+), 12 deletions(-)
drivers/iio/gyro/mpu3050-core.c +16 -12
index 00e58060968c..6380fc1bdee3 100644
@@ -13,6 +13,7 @@
  * TODO: add support for setting up the low pass 3dB frequency.
  */
 
+#include <linux/bitfield.h>
 #include <linux/bitops.h>
 #include <linux/delay.h>
 #include <linux/err.h>
@@ -784,7 +785,8 @@ static int mpu3050_read_mem(struct mpu3050 *mpu3050,
 static int mpu3050_hw_init(struct mpu3050 *mpu3050)
 {
 	int ret;
-	u8 otp[8];
+	__le64 otp_le;
+	u64 otp;
 
 	/* Reset */
 	ret = regmap_update_bits(mpu3050->map,
@@ -815,29 +817,31 @@ static int mpu3050_hw_init(struct mpu3050 *mpu3050)
 				MPU3050_MEM_USER_BANK |
 				MPU3050_MEM_OTP_BANK_0),
 			       0,
-			       sizeof(otp),
-			       otp);
+			       sizeof(otp_le),
+			       (u8 *)&otp_le);
 	if (ret)
 		return ret;
 
 	/* This is device-unique data so it goes into the entropy pool */
-	add_device_randomness(otp, sizeof(otp));
+	add_device_randomness(&otp_le, sizeof(otp_le));
+
+	otp = le64_to_cpu(otp_le);
 
 	dev_info(mpu3050->dev,
-		 "die ID: %04X, wafer ID: %02X, A lot ID: %04X, "
-		 "W lot ID: %03X, WP ID: %01X, rev ID: %02X\n",
+		 "die ID: %04lX, wafer ID: %02lX, A lot ID: %04lX, "
+		 "W lot ID: %03lX, WP ID: %01lX, rev ID: %02lX\n",
 		 /* Die ID, bits 0-12 */
-		 (otp[1] << 8 | otp[0]) & 0x1fff,
+		 FIELD_GET(GENMASK(12, 0), otp),
 		 /* Wafer ID, bits 13-17 */
-		 ((otp[2] << 8 | otp[1]) & 0x03e0) >> 5,
+		 FIELD_GET(GENMASK(17, 13), otp),
 		 /* A lot ID, bits 18-33 */
-		 ((otp[4] << 16 | otp[3] << 8 | otp[2]) & 0x3fffc) >> 2,
+		 FIELD_GET(GENMASK(33, 18), otp),
 		 /* W lot ID, bits 34-45 */
-		 ((otp[5] << 8 | otp[4]) & 0x3ffc) >> 2,
+		 FIELD_GET(GENMASK(45, 34), otp),
 		 /* WP ID, bits 47-49 */
-		 ((otp[6] << 8 | otp[5]) & 0x0380) >> 7,
+		 FIELD_GET(GENMASK(49, 47), otp),
 		 /* rev ID, bits 50-55 */
-		 otp[6] >> 2);
+		 FIELD_GET(GENMASK(55, 50), otp));
 
 	return 0;
 }
Signature · 1 line
2.29.2
#1 lore raw
Quoted 10 lines · Jonathan Cameron

From: Jonathan Cameron <[email protected]>

Inspired by Andy Shevchenko's proposal to use get_unaligned_leXX().

The whole one time programable memory is treated as a single 64bit little endian value. Thus we can avoid a lot of messy handling of fields overlapping byte boundaries by just loading and manipulating it as an __le64 converted to a u64. That lets us just use FIELD_GET() and GENMASK() to extract the values desired.

Quoted 5 lines

Note only build tested.

Cc Andy Shevchenko <[email protected]>
Cc Linus Walleij <[email protected]>

Are there any specific prerequisites? linux-next?

When I apply this and try to compile for an ARMv7 target I get a lot of noise and an error:

In file included from <command-line>:0:0:
../drivers/iio/gyro/mpu3050-core.c: In function ‘mpu3050_hw_init’:
../include/linux/bits.h:36:11: warning: right shift count is negative [-Wshift-count-negative]
(~UL(0) >> (BITS_PER_LONG - 1 - (h))))

           ^
./../include/linux/compiler_types.h:299:9: note: in definition of
macro ‘__compiletime_assert’
   if (!(condition))     \
         ^~~~~~~~~

./../include/linux/compiler_types.h:319:2: note: in expansion of macro ‘_compiletime_assert’
_compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__) ^~~~~~~~~~~~~~~~~~~
../include/linux/build_bug.h:39:37: note: in expansion of macro ‘compiletime_assert’

 #define BUILD_BUG_ON_MSG(cond, msg) compiletime_assert(!(cond), msg)
                                     ^~~~~~~~~~~~~~~~~~

../include/linux/bitfield.h:46:3: note: in expansion of macro ‘BUILD_BUG_ON_MSG’
BUILD_BUG_ON_MSG(!__builtin_constant_p(_mask), \
^~~~~~~~~~~~~~~~
../include/linux/bitfield.h:108:3: note: in expansion of macro ‘__BF_FIELD_CHECK’
__BF_FIELD_CHECK(_mask, _reg, 0U, "FIELD_GET: "); \
^~~~~~~~~~~~~~~~
../include/linux/dev_printk.h:118:33: note: in expansion of macro ‘FIELD_GET’
_dev_info(dev, dev_fmt(fmt), ##__VA_ARGS__)

                                 ^~~~~~~~~~~
../include/linux/bits.h:38:31: note: in expansion of macro ‘__GENMASK’
  (GENMASK_INPUT_CHECK(h, l) + __GENMASK(h, l))
                               ^~~~~~~~~
../include/linux/dev_printk.h:118:33: note: in expansion of macro ‘GENMASK’
  _dev_info(dev, dev_fmt(fmt), ##__VA_ARGS__)
                                 ^~~~~~~~~~~

../drivers/iio/gyro/mpu3050-core.c:830:2: note: in expansion of macro ‘dev_info’
dev_info(mpu3050->dev,
^~~~~~~~
../include/linux/bits.h:36:11: warning: right shift count is negative [-Wshift-count-negative]
(~UL(0) >> (BITS_PER_LONG - 1 - (h))))

           ^
./../include/linux/compiler_types.h:299:9: note: in definition of
macro ‘__compiletime_assert’
   if (!(condition))     \
         ^~~~~~~~~

./../include/linux/compiler_types.h:319:2: note: in expansion of macro ‘_compiletime_assert’
_compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__) ^~~~~~~~~~~~~~~~~~~
../include/linux/build_bug.h:39:37: note: in expansion of macro ‘compiletime_assert’

 #define BUILD_BUG_ON_MSG(cond, msg) compiletime_assert(!(cond), msg)
                                     ^~~~~~~~~~~~~~~~~~

../include/linux/bitfield.h:48:3: note: in expansion of macro ‘BUILD_BUG_ON_MSG’
BUILD_BUG_ON_MSG((_mask) == 0, _pfx "mask is zero"); \
^~~~~~~~~~~~~~~~
../include/linux/bitfield.h:108:3: note: in expansion of macro ‘__BF_FIELD_CHECK’
__BF_FIELD_CHECK(_mask, _reg, 0U, "FIELD_GET: "); \
^~~~~~~~~~~~~~~~
../include/linux/dev_printk.h:118:33: note: in expansion of macro ‘FIELD_GET’
_dev_info(dev, dev_fmt(fmt), ##__VA_ARGS__)

                                 ^~~~~~~~~~~

(...)

Yours,
Linus Walleij

#2 lore raw
Quoted 20 lines · Linus Walleij
Quoted 10 lines · Jonathan Cameron

From: Jonathan Cameron <[email protected]>

Inspired by Andy Shevchenko's proposal to use get_unaligned_leXX().

The whole one time programable memory is treated as a single 64bit little endian value. Thus we can avoid a lot of messy handling of fields overlapping byte boundaries by just loading and manipulating it as an __le64 converted to a u64. That lets us just use FIELD_GET() and GENMASK() to extract the values desired.

Quoted 5 lines

Note only build tested.

Cc Andy Shevchenko <[email protected]>
Cc Linus Walleij <[email protected]>

Are there any specific prerequisites? linux-next?

Quoted 8 lines

When I apply this and try to compile for an ARMv7 target I get a lot of noise and an error:

In file included from <command-line>:0:0:
../drivers/iio/gyro/mpu3050-core.c: In function ‘mpu3050_hw_init’:
../include/linux/bits.h:36:11: warning: right shift count is negative [-Wshift-count-negative]
(~UL(0) >> (BITS_PER_LONG - 1 - (h))))

I think GENMASK_ULL() has to be used.

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#3 lore raw
Quoted 24 lines · Andy Shevchenko
Quoted 23 lines · Linus Walleij
Quoted 10 lines · Jonathan Cameron

From: Jonathan Cameron <[email protected]>

Inspired by Andy Shevchenko's proposal to use get_unaligned_leXX().

The whole one time programable memory is treated as a single 64bit little endian value. Thus we can avoid a lot of messy handling of fields overlapping byte boundaries by just loading and manipulating it as an __le64 converted to a u64. That lets us just use FIELD_GET() and GENMASK() to extract the values desired.

Quoted 5 lines

Note only build tested.

Cc Andy Shevchenko <[email protected]>
Cc Linus Walleij <[email protected]>

Are there any specific prerequisites? linux-next?

When I apply this and try to compile for an ARMv7 target I get a lot of noise and an error:

Quoted 7 lines
Quoted 5 lines

In file included from <command-line>:0:0:
../drivers/iio/gyro/mpu3050-core.c: In function ‘mpu3050_hw_init’:
../include/linux/bits.h:36:11: warning: right shift count is negative [-Wshift-count-negative]
(~UL(0) >> (BITS_PER_LONG - 1 - (h))))

I think GENMASK_ULL() has to be used.

Oh indeed. It works as long as one just try to test-compile it on a 64bit machine of course :D

Yours,
Linus Walleij

author maintainer
#4 lore raw
Quoted 32 lines · On Sun, 29 Nov 2020 02:11:01 +0100 Linus Walleij <[email protected]>…
Quoted 30 lines · Andy Shevchenko
Quoted 28 lines · Linus Walleij
Quoted 10 lines · Jonathan Cameron

From: Jonathan Cameron <[email protected]>

Inspired by Andy Shevchenko's proposal to use get_unaligned_leXX().

The whole one time programable memory is treated as a single 64bit little endian value. Thus we can avoid a lot of messy handling of fields overlapping byte boundaries by just loading and manipulating it as an __le64 converted to a u64. That lets us just use FIELD_GET() and GENMASK() to extract the values desired.

Quoted 5 lines

Note only build tested.

Cc Andy Shevchenko <[email protected]>
Cc Linus Walleij <[email protected]>

Are there any specific prerequisites? linux-next?

When I apply this and try to compile for an ARMv7 target I get a lot of noise and an error:

In file included from <command-line>:0:0:
../drivers/iio/gyro/mpu3050-core.c: In function ‘mpu3050_hw_init’:
../include/linux/bits.h:36:11: warning: right shift count is negative [-Wshift-count-negative]
(~UL(0) >> (BITS_PER_LONG - 1 - (h))))

Quoted 4 lines
Quoted 1 line

I think GENMASK_ULL() has to be used.

Oh indeed. It works as long as one just try to test-compile it on a 64bit machine of course :D

Doh. That's me being lazy :(

Will flip them all over to GENMASK_ULL and do a 32 bit build test. I checked that the FIELD_GET would be fine, but forgot to look at GENMASK.

oops.

Jonathan

Quoted 3 lines

Yours,
Linus Walleij