STM32 hw entropy #2611

adustm · 2016-09-02T15:20:01Z

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 · 2016-09-05T08:09:25Z

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 · 2016-09-06T15:27:53Z

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

sg- · 2016-09-09T21:04:29Z

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

adustm · 2016-09-12T12:36:33Z

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 · 2016-09-27T09:40:19Z

Bump @sg- @0xc0170

0xc0170 · 2016-09-27T09:45:59Z

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 · 2016-09-27T09:46:48Z

hal/targets/hal/TARGET_STM/TARGET_STM32L0/entropy_hardware_poll.c

+
+    /* Get Random byte */
+    for( uint32_t i = 0; i < len; i++ ){
+        rng_get_byte( output + i );


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 · 2016-09-27T12:24:00Z

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 · 2016-09-27T12:31:17Z

retest uvisor

mazimkhan · 2016-09-27T12:36:25Z

retest uvisor

sg- · 2016-09-28T03:12:06Z

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

adustm · 2016-09-29T12:54:31Z

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 · 2016-09-29T13:35:46Z

@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- · 2016-09-30T19:25:12Z

@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- · 2016-10-03T16:15:34Z

@adustm bump.

adustm · 2016-10-04T12:47:51Z

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 · 2016-10-04T15:48:04Z

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

adustm · 2016-10-06T12:09:32Z

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

0xc0170 · 2016-10-06T15:22:19Z

Thanks for handling that first number .

LGTM

bridadan · 2016-10-06T16:36:10Z

/morph test

mbed-bot · 2016-10-06T18:09:03Z

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 · 2016-10-06T18:13:23Z

@0xc0170 Good to remove needs: works label?

sg- · 2016-10-07T16:40:29Z

@adustm target.json conflict to resolve :)

update code accordingly

adustm · 2016-10-10T08:16:35Z

Hello @sg-, it's done.

mazimkhan · 2016-10-10T09:14:31Z

targets/targets.json

-        "device_has": ["ANALOGIN", "ANALOGOUT", "I2C", "I2CSLAVE", "INTERRUPTIN", "LOWPOWERTIMER", "PORTIN", "PORTINOUT", "PORTOUT", "PWMOUT", "RTC", "SERIAL", "SERIAL_FC", "SERIAL_ASYNCH", "SLEEP", "SPI", "SPISLAVE", "STDIO_MESSAGES"],
-        "release_versions": ["2", "5"],
+        "device_has": ["ANALOGIN", "ANALOGOUT", "I2C", "I2CSLAVE", "INTERRUPTIN", "LOWPOWERTIMER", "PORTIN", "PORTINOUT", "PORTOUT", "PWMOUT", "RTC", "SERIAL", "SERIAL_FC", "SERIAL_ASYNCH", "SLEEP", "SPI", "SPISLAVE", "STDIO_MESSAGES", "TRNG"],
+        "release_versions": ["2", "5"]


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

sg- · 2016-10-10T15:23:35Z

/morph test-nightly

mbed-bot · 2016-10-10T22:29:28Z

Result: FAILURE

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

/morph test-nightly

bridadan · 2016-10-10T22:50:22Z

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

sg- added the needs: CI label Sep 9, 2016

sg- added needs: work and removed needs: CI labels Sep 16, 2016

0xc0170 reviewed Sep 27, 2016

View reviewed changes

adustm force-pushed the STM32_entropy branch from dd6aafd to 8b647ec Compare September 27, 2016 12:22

adustm force-pushed the STM32_entropy branch from 8b647ec to f3888f9 Compare September 29, 2016 12:50

0xc0170 added needs: review and removed needs: work labels Sep 29, 2016

sg- added needs: work and removed needs: review labels Oct 5, 2016

adustm force-pushed the STM32_entropy branch from 38ff0a4 to e8dceac Compare October 6, 2016 12:08

sg- added needs: CI and removed needs: work labels Oct 7, 2016

adustm added 6 commits October 10, 2016 10:09

Enable TLS entropy in targets.json for STM32L4 and STM32L0 products

4f1f272

Add random generator for stm32L0 + stm32L4

24832ca

Add Periphclock enable for L476 products

5069c73

Change MBEDTLS_ENTROPY_ALT to device has TRNG

e189bc7

update code accordingly

rework after directory moves

bda200e

Add objects.h for L073

f140559

adustm force-pushed the STM32_entropy branch from e8dceac to f140559 Compare October 10, 2016 08:15

mazimkhan reviewed Oct 10, 2016

View reviewed changes

0xc0170 added the needs: work label Oct 10, 2016

andreaslarssonublox mentioned this pull request Oct 10, 2016

Enabled TRNG functionality for UBLOX_EVK_ODIN_W2 target. #2970

Merged

Fix typo

7bb6117

sg- removed the needs: work label Oct 10, 2016

sg- added ready for merge and removed needs: CI labels Oct 10, 2016

sg- merged commit f9ee683 into ARMmbed:master Oct 12, 2016

sg- added release-version: 5.2.0-rc3 and removed ready for merge labels Oct 12, 2016

adustm deleted the STM32_entropy branch October 13, 2016 07:38

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

STM32 hw entropy #2611

STM32 hw entropy #2611

adustm commented Sep 2, 2016

adustm commented Sep 5, 2016

bridadan commented Sep 6, 2016

sg- commented Sep 9, 2016

adustm commented Sep 12, 2016

screamerbg commented Sep 27, 2016

0xc0170 commented Sep 27, 2016

0xc0170 Sep 27, 2016 •

edited

Loading

adustm commented Sep 27, 2016

mazimkhan commented Sep 27, 2016

mazimkhan commented Sep 27, 2016

sg- commented Sep 28, 2016

adustm commented Sep 29, 2016

0xc0170 commented Sep 29, 2016

sg- commented Sep 30, 2016

sg- commented Oct 3, 2016

adustm commented Oct 4, 2016

adustm commented Oct 4, 2016

adustm commented Oct 6, 2016

0xc0170 commented Oct 6, 2016

bridadan commented Oct 6, 2016

mbed-bot commented Oct 6, 2016

bridadan commented Oct 6, 2016

sg- commented Oct 7, 2016

adustm commented Oct 10, 2016

mazimkhan Oct 10, 2016

sg- commented Oct 10, 2016

mbed-bot commented Oct 10, 2016

bridadan commented Oct 10, 2016

STM32 hw entropy #2611

STM32 hw entropy #2611

Conversation

adustm commented Sep 2, 2016

adustm commented Sep 5, 2016

bridadan commented Sep 6, 2016

sg- commented Sep 9, 2016

adustm commented Sep 12, 2016

screamerbg commented Sep 27, 2016

0xc0170 commented Sep 27, 2016

0xc0170 Sep 27, 2016 • edited Loading

Choose a reason for hiding this comment

adustm commented Sep 27, 2016

mazimkhan commented Sep 27, 2016

mazimkhan commented Sep 27, 2016

sg- commented Sep 28, 2016

adustm commented Sep 29, 2016

0xc0170 commented Sep 29, 2016

sg- commented Sep 30, 2016

sg- commented Oct 3, 2016

adustm commented Oct 4, 2016

adustm commented Oct 4, 2016

adustm commented Oct 6, 2016

0xc0170 commented Oct 6, 2016

bridadan commented Oct 6, 2016

mbed-bot commented Oct 6, 2016

Result: SUCCESS

Output

bridadan commented Oct 6, 2016

sg- commented Oct 7, 2016

adustm commented Oct 10, 2016

mazimkhan Oct 10, 2016

Choose a reason for hiding this comment

sg- commented Oct 10, 2016

mbed-bot commented Oct 10, 2016

Result: FAILURE

bridadan commented Oct 10, 2016

0xc0170 Sep 27, 2016 •

edited

Loading