lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <nsqdetgfr7p5klex77alaeoosgi5o5phhnri5kcn735wsn7e5o@phuellsm23nd>
Date: Sun, 29 Sep 2024 15:31:14 -0400
From: Alex Lanzano <lanzano.alex@...il.com>
To: Jonathan Cameron <jic23@...nel.org>
Cc: Lars-Peter Clausen <lars@...afoo.de>, 
	linux-kernel-mentees@...ts.linuxfoundation.org, skhan@...uxfoundation.org, linux-kernel@...r.kernel.org, 
	linux-iio@...r.kernel.org
Subject: Re: [PATCH] iio: imu: bmi270: Add spi driver for bmi270 imu

Thanks for the review!

On Sun, Sep 29, 2024 at 03:54:37PM GMT, Jonathan Cameron wrote:
> On Fri, 27 Sep 2024 14:37:10 -0400
> Alex Lanzano <lanzano.alex@...il.com> wrote:
> 
> > Implement SPI driver for the Bosch BMI270 6-axis IMU. Provide raw read
> > write access to acceleration and angle velocity measurements via the SPI
> > interface on the device.
> > 
> > Signed-off-by: Alex Lanzano <lanzano.alex@...il.com>
> 
> A few minor things inline but looks good in general.
> 
> Jonathan
> 
> > diff --git a/drivers/iio/imu/bmi270/bmi270.h b/drivers/iio/imu/bmi270/bmi270.h
> > index 608b29ea58a3..8950e6234203 100644
> > --- a/drivers/iio/imu/bmi270/bmi270.h
> > +++ b/drivers/iio/imu/bmi270/bmi270.h
> > @@ -4,11 +4,13 @@
> >  #define BMI270_H_
> >  
> >  #include <linux/regmap.h>
> > +#include <linux/iio/iio.h>
> >  
> >  struct device;
> >  struct bmi270_data {
> >  	struct device *dev;
> >  	struct regmap *regmap;
> > +	__le16 sample __aligned(IIO_DMA_MINALIGN);
> 
> For the read path you are bouncing anyway, so the DMA_MINALIGN is only needed
> for anything the write direction.  Make the suggested change below and that
> will bounce as well so that you don't need this.
> 

Understood! Will remove in v2 and use spi_write_then_read

> >  };
> >  
> >  extern const struct regmap_config bmi270_regmap_config;
> > diff --git a/drivers/iio/imu/bmi270/bmi270_core.c b/drivers/iio/imu/bmi270/bmi270_core.c
> > index 8e45343d6472..4decdad791d9 100644
> > --- a/drivers/iio/imu/bmi270/bmi270_core.c
> > +++ b/drivers/iio/imu/bmi270/bmi270_core.c
> > @@ -66,16 +66,9 @@ enum bmi270_scan {
> >  	BMI270_SCAN_GYRO_Z,
> >  };
> >  
> > -const struct regmap_config bmi270_regmap_config = {
> > -	.reg_bits = 8,
> > -	.val_bits = 8,
> > -};
> > -EXPORT_SYMBOL_NS_GPL(bmi270_regmap_config, IIO_BMI270);
> > -
> >  static int bmi270_get_data(struct bmi270_data *bmi270_device,
> >  			   int chan_type, int axis, int *val)
> >  {
> > -	__le16 sample;
> >  	int reg;
> >  	int ret;
> >  
> > @@ -90,11 +83,13 @@ static int bmi270_get_data(struct bmi270_data *bmi270_device,
> >  		return -EINVAL;
> >  	}
> >  
> > -	ret = regmap_bulk_read(bmi270_device->regmap, reg, &sample, sizeof(sample));
> > +	ret = regmap_bulk_read(bmi270_device->regmap, reg,
> > +			       &bmi270_device->sample,
> > +			       sizeof(bmi270_device->sample));
> >  	if (ret)
> >  		return ret;
> >  
> > -	*val = sign_extend32(le16_to_cpu(sample), 15);
> > +	*val = sign_extend32(le16_to_cpu(bmi270_device->sample), 15);
> >  
> >  	return 0;
> >  }
> > diff --git a/drivers/iio/imu/bmi270/bmi270_i2c.c b/drivers/iio/imu/bmi270/bmi270_i2c.c
> > index f70dee2d8a64..ce8279ae90cd 100644
> > --- a/drivers/iio/imu/bmi270/bmi270_i2c.c
> > +++ b/drivers/iio/imu/bmi270/bmi270_i2c.c
> > @@ -9,12 +9,17 @@
> >  
> >  #include "bmi270.h"
> >  
> > +const struct regmap_config bmi270_i2c_regmap_config = {
> static const
> 
> (same for spi one)
> 

Will fix in v2!

> > +	.reg_bits = 8,
> > +	.val_bits = 8,
> > +};
> > +
> >  static int bmi270_i2c_probe(struct i2c_client *client)
> >  {
> >  	struct regmap *regmap;
> >  	struct device *dev = &client->dev;
> >  
> > -	regmap = devm_regmap_init_i2c(client, &bmi270_regmap_config);
> > +	regmap = devm_regmap_init_i2c(client, &bmi270_i2c_regmap_config);
> >  	if (IS_ERR(regmap))
> >  		return dev_err_probe(dev, PTR_ERR(regmap),
> >  				     "Failed to init i2c regmap");
> > diff --git a/drivers/iio/imu/bmi270/bmi270_spi.c b/drivers/iio/imu/bmi270/bmi270_spi.c
> > new file mode 100644
> > index 000000000000..906b9b852a09
> > --- /dev/null
> > +++ b/drivers/iio/imu/bmi270/bmi270_spi.c
> > @@ -0,0 +1,89 @@
> > +// SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
> > +
> > +#include <linux/module.h>
> > +#include <linux/spi/spi.h>
> > +#include <linux/iio/iio.h>
> > +#include <linux/module.h>
> > +#include <linux/mod_devicetable.h>
> > +#include <linux/regmap.h>
> Alphabetical order preferred.
> 

Will fix in v2

> > +
> > +#include "bmi270.h"
> > +
> > +/*
> > + * The following two functions are taken from the BMI323 spi driver code.
> > + * In section 6.4 of the BMI270 data it specifies that after a read
> > + * operation the first data byte from the device is a dummy byte
> > + */
> > +static int bmi270_regmap_spi_read(void *context, const void *reg_buf,
> > +				  size_t reg_size, void *val_buf,
> > +				  size_t val_size)
> > +{
> > +	struct spi_device *spi = context;
> 
> I'd be tempted to rename the input parameter context to spi and then
> parse it directly to the spi_write_then_read()
> 

Will do! I named it context since that's what the function pointer uses

> > +
> > +	return spi_write_then_read(spi, reg_buf, reg_size, val_buf, val_size);
> > +}
> > +
> > +static int bmi270_regmap_spi_write(void *context, const void *data,
> > +				   size_t count)
> > +{
> > +	struct spi_device *spi = context;
> > +	u8 *data_buff = (u8 *)data;
> > +
> > +	/*
> > +	 * Remove the extra pad byte since its only needed for the read
> > +	 * operation
> > +	 */
> > +	data_buff[1] = data_buff[0];
> > +	return spi_write(spi, data_buff + 1, count - 1);
> That needs a DMA safe buffer (unlike write_then_read which always
> bounces).  I'd avoid that complexity by using spi_write_then_read
> here as well but set the read to 0 length and pass NULL for the buffer.
> That function is intended to be used like this as it special cases 0
> length for either write or read buffers.
> 

Understood! Will fix in v2.

> > +}
> 

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ