[PATCH v3 2/3] crypto: atmel-sha204a - fix truncated 32-byte blocking read

[Thorsten Blum]

Hi Lothar,

On Wed, Apr 22, 2026 at 09:09:35PM +0000, Lothar Rubusch wrote:
> The ATSHA204A returns a 35-byte packet consisting of a 1-byte count,
> 32 bytes of entropy, and a 2-byte CRC. The current blocking read
> implementation was incorrectly copying data starting from the
> count byte, leading to offset data and truncated entropy.
> 
> Additionally, the chip requires significant execution time to
> generate random numbers, going by the datasheet. Reading the I2C bus
> too early results in the chip NACK-ing or returning a partial buffer
> followed by zeros.
> 
> Verification:
> Tests before showed repeadetly reading only 8 bytes of entropy:
> $ head -c 32 /dev/hwrng | hexdump -C
> 00000000  02 28 85 b3 47 40 f2 ee  00 00 00 00 00 00 00 00  |.(..G at ..........|
> 00000010  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
> 00000020
> 
> After this patch applied, the result will be as follows:
> $ head -c 32 /dev/hwrng | hexdump -C
> 00000000  5a fc 3f 13 14 68 fe 06  68 0a bd 04 83 6e 09 69  |Z.?..h..h....n.i|
> 00000010  75 ff cf 87 10 84 3b c9  c1 df ae eb 45 53 4c c3  |u.....;.....ESL.|
> 00000020
> 
> Fix these issues by:
> Increase cmd.msecs to 30ms to provide sufficient execution time. Then
> set cmd.rxsize to RANDOM_RSP_SIZE (35 bytes) to capture the entire
> hardware response. Eventually, correct the memcpy() offset to index 1 of
> the data buffer to skip the count byte and retrieve exactly 32 bytes of
> entropy.
> 
> Fixes: da001fb651b0 ("crypto: atmel-i2c - add support for SHA204A random number generator")
> Signed-off-by: Lothar Rubusch <l.rubusch at gmail.com>

Thank you for your patches. I tested patch 2/3 on real hardware and it
fixes rngtest for me. However, I have a few comments below.

> ---
>  drivers/crypto/atmel-sha204a.c | 9 +++++++--
>  1 file changed, 7 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/crypto/atmel-sha204a.c b/drivers/crypto/atmel-sha204a.c
> index 19720bdd446d..f7dc00d0f4cd 100644
> --- a/drivers/crypto/atmel-sha204a.c
> +++ b/drivers/crypto/atmel-sha204a.c
> @@ -19,6 +19,9 @@
>  #include <linux/workqueue.h>
>  #include "atmel-i2c.h"
>  
> +#define ATMEL_RNG_BLOCK_SIZE 32
> +#define ATMEL_RNG_EXEC_TIME 30
> +
>  static void atmel_sha204a_rng_done(struct atmel_i2c_work_data *work_data,
>  				   void *areq, int status)
>  {
> @@ -91,13 +94,15 @@ static int atmel_sha204a_rng_read(struct hwrng *rng, void *data, size_t max,
>  	i2c_priv = container_of(rng, struct atmel_i2c_client_priv, hwrng);
>  
>  	atmel_i2c_init_random_cmd(&cmd);
> +	cmd.msecs = ATMEL_RNG_EXEC_TIME;
> +	cmd.rxsize = RANDOM_RSP_SIZE;

atmel_i2c_init_random_cmd() already sets cmd.rxsize to RANDOM_RSP_SIZE.

Changing cmd.msecs does not appear to be strictly necessary for the fix.
The only difference I observe is that rngtest runs faster with the new
value.

Here are my test results without changing cmd.msecs:

$ sudo head -c 300000 /dev/hwrng | rngtest -c 100
[...]
rngtest: starting FIPS tests...
rngtest: bits received from input: 2000032
rngtest: FIPS 140-2 successes: 100
rngtest: FIPS 140-2 failures: 0
[...]
rngtest: input channel speed: (min=1.118; avg=3.746; max=6510416.667)Kibits/s
rngtest: FIPS tests speed: (min=26.272; avg=27.471; max=32.829)Mibits/s
rngtest: Program run time: 538942640 microseconds

and with 'cmd.msecs = ATMEL_RNG_EXEC_TIME':

$ sudo head -c 300000 /dev/hwrng | rngtest -c 100
[...]
rngtest: starting FIPS tests...
rngtest: bits received from input: 2000032
rngtest: FIPS 140-2 successes: 100
rngtest: FIPS 140-2 failures: 0
[...]
rngtest: input channel speed: (min=1.584; avg=5.295; max=6510416.667)Kibits/s
rngtest: FIPS tests speed: (min=26.602; avg=27.321; max=28.257)Mibits/s
rngtest: Program run time: 381309284 microseconds

>  	ret = atmel_i2c_send_receive(i2c_priv->client, &cmd);
>  	if (ret)
>  		return ret;
>  
> -	max = min(sizeof(cmd.data), max);
> -	memcpy(data, cmd.data, max);
> +	max = min_t(size_t, ATMEL_RNG_BLOCK_SIZE, max);
> +	memcpy(data, &cmd.data[1], max);

This works and fixes rngtest for me on real hardware.

min_t() is not strictly necessary here, since the types are compatible
and min() is sufficient.

I agree with Ard that patch 2/3 should be a separate patch for easier
stable backporting.

Thanks,
Thorsten