From: Sascha Hauer <s.hauer@pengutronix.de>
To: Steffen Trumtrar <s.trumtrar@pengutronix.de>
Cc: barebox@lists.infradead.org
Subject: Re: [PATCH 1/5] drivers: add simple hw_random implementation
Date: Mon, 29 Feb 2016 08:06:43 +0100 [thread overview]
Message-ID: <20160229070643.GD9224@pengutronix.de> (raw)
In-Reply-To: <1456488287-23404-2-git-send-email-s.trumtrar@pengutronix.de>
On Fri, Feb 26, 2016 at 01:04:43PM +0100, Steffen Trumtrar wrote:
> Add a simple hw_random implementation based on code from
> Linux v4.5-rc5.
>
> All the entropypool initialization stuff is left out and
> the obsolete data_read/data_present calls are omitted.
>
> Signed-off-by: Steffen Trumtrar <s.trumtrar@pengutronix.de>
> ---
> +#include <common.h>
> +#include <linux/hw_random.h>
> +
> +static LIST_HEAD(hwrngs);
> +
> +#define RNG_BUFFER_SIZE 32
> +
> +int hwrng_get_data(struct hwrng *rng, u8 *buffer, size_t size, int wait)
> +{
> + if (rng->read)
> + return rng->read(rng, buffer, size, wait);
When rng->read is NULL you already fail in the registration, no need to
check here.
> +
> + return 0;
> +}
> +
> +static int hwrng_init(struct hwrng *rng)
> +{
> + int ret = 0;
> +
> + if (rng->init)
> + ret = rng->init(rng);
> +
> + list_add_tail(&rng->list, &hwrngs);
When rng->init fails you probably don't want to add it to the list.
> +
> + return ret;
> +}
> +
> +static ssize_t rng_dev_read(struct cdev *cdev, void *buf, size_t size,
> + loff_t offset, unsigned long flags)
> +{
> + struct hwrng *rng = container_of(cdev, struct hwrng, cdev);
> + ssize_t cur = 0;
> + int err = 0;
> + int len;
> + size_t count = size;
> +
> + memset(buf, 0, size);
> +
> + while (count) {
You can initialize a variable with the minimum of count and
RNG_BUFFER_SIZE here,
> + len = hwrng_get_data(rng, rng->buf, RNG_BUFFER_SIZE, true);
Use it here instead of RNG_BUFFER_SIZE
> + if (len < 0) {
> + err = len;
> + break;
> + }
> +
> + if (len > count)
> + len = count;
and drop this check.
> +
> + memcpy(buf + cur, rng->buf, len);
> +
> + count -= len;
> + cur += len;
> + }
> +
> + return cur ? : err;
When you get an error in the second loop iteration then cur is true and
you return it instead of the error. Is that what you want?
> +}
> +
> +
> +static struct file_operations rng_chrdev_ops = {
> + .read = rng_dev_read,
> + .lseek = dev_lseek_default,
> +};
> +
> +static int hwrng_register_cdev(struct hwrng *rng)
> +{
> + rng->cdev.name = "hwrng";
> + rng->cdev.flags = DEVFS_IS_CHARACTER_DEV;
> + rng->cdev.ops = &rng_chrdev_ops;
> + rng->cdev.dev = rng->dev;
> +
> + return devfs_create(&rng->cdev);
> +}
> +
> +struct hwrng *hwrng_get_first(void)
> +{
> + if (!list_empty(&hwrngs))
> + return list_first_entry(&hwrngs, struct hwrng, list);
> + else
> + return ERR_PTR(-ENODEV);
Please use positive logic, it's usually easier to read.
> +}
> +
> +int hwrng_register(struct device_d *dev, struct hwrng *rng)
> +{
> + int err;
> +
> + if (rng->name == NULL || rng->read == NULL)
> + return -EINVAL;
> +
> + rng->buf = xzalloc(RNG_BUFFER_SIZE);
Please make sure you do not loose memory in infrastructure code.
> +
> + err = hwrng_init(rng);
> + if (err)
> + return err;
> +
> + rng->dev = dev;
> +
> + err = hwrng_register_cdev(rng);
> +
> + return err;
> +}
> + of the GNU General Public License, incorporated herein by reference.
> +
> + */
> +
> +#ifndef LINUX_HWRANDOM_H_
> +#define LINUX_HWRANDOM_H_
> +
> +#include <linux/list.h>
> +
> +/**
> + * struct hwrng - Hardware Random Number Generator driver
> + * @name: Unique RNG name.
> + * @init: Initialization callback (can be NULL).
> + * @read: New API. drivers can fill up to max bytes of data
> + * into the buffer. The buffer is aligned for any type.
> + * @priv: Private data, for use by the RNG driver.
> + */
> +struct hwrng {
> + const char *name;
> + int (*init)(struct hwrng *rng);
> + int (*read)(struct hwrng *rng, void *data, size_t max, bool wait);
> + unsigned long priv;
This is unused, no?
Sascha
--
Pengutronix e.K. | |
Industrial Linux Solutions | http://www.pengutronix.de/ |
Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0 |
Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 |
_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox
next prev parent reply other threads:[~2016-02-29 7:07 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-02-26 12:04 [PATCH 0/5] HWRNG: add support for HW Random Number Generators Steffen Trumtrar
2016-02-26 12:04 ` [PATCH 1/5] drivers: add simple hw_random implementation Steffen Trumtrar
2016-02-29 7:06 ` Sascha Hauer [this message]
2016-02-26 12:04 ` [PATCH 2/5] lib: random: get_random_bytes from HWRNG if present Steffen Trumtrar
2016-02-26 20:28 ` Jason Cooper
2016-02-29 13:45 ` Steffen Trumtrar
2016-02-26 12:04 ` [PATCH 3/5] ARM: imx25: clk: add rngb clock Steffen Trumtrar
2016-02-26 12:04 ` [PATCH 4/5] ARM: i.MX25: dtsi: add rng node Steffen Trumtrar
2016-02-26 12:04 ` [PATCH 5/5] hw_random: add driver for Freescale RNGC Steffen Trumtrar
2016-02-29 7:33 ` Sascha Hauer
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=20160229070643.GD9224@pengutronix.de \
--to=s.hauer@pengutronix.de \
--cc=barebox@lists.infradead.org \
--cc=s.trumtrar@pengutronix.de \
/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