[PATCH] mtd: otp: Put factory OTP/NVRAM into the entropy pool

[Jason A. Donenfeld]

On Fri, Jun 02, 2023 at 11:23:18PM +0200, Linus Walleij wrote:
> The factory OTP, if supported, contains factory-programmed
> information such as typically the serial number or production
> week for the chip.
> 
> As this is device-unique information, submit it into the
> system entropy pool.
> 
> This does not count as improvement of the entropy as such
> but in practice it makes it a bit more random to mix in these
> numbers.
> 
> Cc: Jason A. Donenfeld <Jason at zx2c4.com>
> Cc: Michael Walle <michael at walle.cc>
> Signed-off-by: Linus Walleij <linus.walleij at linaro.org>

Acked-by: Jason A. Donenfeld <Jason at zx2c4.com>

Great idea!

Just one question below. Feel free to disregard if it's silly.

> ---
> This is similar to the patch I made to add MMC/SD-card serial
> numbers to randomness, just with raw NOR flash.
> ---
>  drivers/mtd/mtdcore.c | 20 ++++++++++++++++++++
>  1 file changed, 20 insertions(+)
> 
> diff --git a/drivers/mtd/mtdcore.c b/drivers/mtd/mtdcore.c
> index 60670b2f70b9..efc8cf76dc60 100644
> --- a/drivers/mtd/mtdcore.c
> +++ b/drivers/mtd/mtdcore.c
> @@ -23,6 +23,7 @@
>  #include <linux/idr.h>
>  #include <linux/backing-dev.h>
>  #include <linux/gfp.h>
> +#include <linux/random.h>
>  #include <linux/slab.h>
>  #include <linux/reboot.h>
>  #include <linux/leds.h>
> @@ -966,6 +967,25 @@ static int mtd_otp_nvmem_add(struct mtd_info *mtd)
>  		}
>  
>  		if (size > 0) {
> +			/*
> +			 * The factory OTP contains thing such as a unique serial
> +			 * number and is small, so let's read it out and put it
> +			 * into the entropy pool.
> +			 */
> +			void *otp;
> +
> +			otp = kmalloc(size, GFP_KERNEL);
> +			if (!otp)
> +				return -ENOMEM;
> +			err = mtd_nvmem_fact_otp_reg_read(mtd, 0, otp, size);
> +			if (err < 0) {
> +				kfree(otp);
> +				return err;
> +			}
> +			if (err == size)
> +				add_device_randomness(otp, size);

What if instead you just did `add_device_randomness(otp, err)`, without
the conditional checking that `err == size`? You check for the actual
error case above (`err < 0`), and so in the case that
mtd_nvmem_fact_otp_reg_read() returns less than you thought it should,
it can't hurt to still add it to the rng.

Jason