STM32 hw entropy

[adustm]

Hello @0xc0170 and @screamerbg
This Pull Request is the completion of the HW entropy for STM32 platforms.
Every ST platform having a TrueRandomGenerator in the microcontroler is now available.

I performed the test bellow on NUCLEO_L476RG, NUCLEO_L073RZ, NUCLEO_L432KC and DISCO_L476VG
with IAR, ARM and GCC_ARM toolchains

#include "test_env.h"
#include "entropy_poll.h"

#define BUFF_SIZE       500
unsigned char myrndbuff[BUFF_SIZE]={0};
size_t *olen;

int main() {
    GREENTEA_SETUP(20, "default_auto");
    printf("MBED: st_rng_entropy\r\n");
    int ret = 0, i=0;
    bool result = false;
    size_t mysize=0;
    olen = &mysize;
    ret = mbedtls_hardware_poll( NULL, myrndbuff, BUFF_SIZE, olen );
    if (ret == 0 ) {
        printf("tls_ret_ok\n");
        printf("%i\n",(int)*olen);
        if (*olen == BUFF_SIZE) {
            printf("size ok\n");
            result = true;
            for (i=0;i<BUFF_SIZE;i++){unsigned char ch=myrndbuff[i];printf("%u\n",ch);}
        } else printf("size ko\n");
    } else printf("tls_ret_ko\n");
    printf("Test end\n");

  GREENTEA_TESTSUITE_RESULT(result);
}

[adustm]

Hello,
Something went wrong in the automatic tests. Could you let me know what it is ? I cannot see the results.
Kind regards
Armelle

[bridadan]

@mazimkhan Looks like there was a failure with the uvisor tests, any idea what's going on here?

[sg-]

@adustm I'd like to hold on this until the tests get merged. Should be soon. #2603

[adustm]

Hello @-sg,
I passed the tls test of #2603, as requested by @screamerbg
I hope you will be able to merge both soon.
Cheers
Armelle

[screamerbg]

Bump @sg- @0xc0170

[0xc0170]

Bump @sg- @0xc0170

We have been working on adding TRNG HAL last days, the PR should be merged today, and this PR should be updated to use it (it require changes in mbedtls ). I can do that manually to bring this up to date if required.

we will keep you updated

[0xc0170]

One thing I noticed while reading reference manual for some STM32F4 RNG peripheral, once RNG is started, the first byte should be discarded. this is not done for any STM as I have seen so far.

[adustm]

Hello, It's been 2 weeks and I missed the target release 126...
Could you please merge this PR (as you already did with the previous one #2253 ) ? It is the exact same code for some other platforms

Kind regards
Armelle

[mazimkhan]

retest uvisor

[mazimkhan]

retest uvisor

[sg-]

@adustm This needs to be updated to use TRGN in the device_has field per #2765

[adustm]

Hello,
I have updated this PR wrt to the changes from @0xc0170 (use device TRNG instead of the macro MBEDTLS_ENTROPY_ALT ) from #2765 .
I have passed again the test from the mbedtls self test used in #2603
I think this PR is now ready. Let me know if you agree.
Kind regards
Armelle

[0xc0170]

@adustm Thank you ! Got a question. in trng_init, shouldn't you read one byte to discard it (when you enable RNG peripheral)? I raised that question above, I recall it was specified in STM32F4 reference manual.

Looks fine to me otherwise.

@sg- Please review

[sg-]

@adustm Thank you ! Got a question. in trng_init, shouldn't you read one byte to discard it (when you enable RNG peripheral)? I raised that question above, I recall it was specified in STM32F4 reference manual.

@adustm bump.

[sg-]

@adustm bump.

[adustm]

Hello @sg-
I confirm the specification point. I have made some internal discussions, as I wanted this code to be part of the ST HAL init function, but it needs to be done by me.
I'll modify this PR and add the modification for every platforms.
Kind regards

[adustm]

... files have change of location... I'll update as soon as I can... I'll let you know
cheers

[adustm]

Hello @sg-
I think it is ready now .
Let me know !
Kind regards
Armelle

[0xc0170]

Thanks for handling that first number .

LGTM

[bridadan]

/morph test

[mbed-bot]

Result: SUCCESS

Your command has finished executing! Here's what you wrote!

/morph test

Output

mbed Build Number: 1067

All builds and test passed!

[bridadan]

@0xc0170 Good to remove needs: works label?

[sg-]

@adustm target.json conflict to resolve :)

[adustm]

Hello @sg-, it's done.

[mazimkhan]

Please put , after "release_versions": ["2", "5"]. It breaks the builds.

[sg-]

/morph test-nightly

[mbed-bot]

Result: FAILURE

Your command has finished executing! Here's what you wrote!

/morph test-nightly

[bridadan]

Same deal, all failing tests are currently in the nightly 👍 LGTM