19. True random number generator (RNG)

19.1 Introduction

The RNG is a true random number generator that continuously provides 32-bit entropy samples, based on an analog noise source. It can be used by the application as a live entropy source to build a NIST compliant deterministic random bit generator (DRBG).

The RNG true random number generator has been tested using NIST statistical test suite SP800-22 rev1a (April 2010).

19.2 RNG main features

19.3 RNG functional description

19.3.1 RNG block diagram

Figure 116 shows the RNG block diagram.

RNG block diagram showing internal components like AHB interface, Banked Registers (RNG_CR, RNG_DR, RNG_SR), Fault detection Clock checker, Data shift reg, Post-processing logic (8-bit LFSR x2), Sampling & Normalization (x 2), and Analog noise sources (source 1, source 2). It also shows external signals: 32-bit AHB Bus, rng_it, rng_hclk, and rng_clk. The diagram is divided into AHB clock domain and RNG clock domain. MSV42096V2 is noted at the bottom right.

Figure 116. RNG block diagram

MSV42096V2

RNG block diagram showing internal components like AHB interface, Banked Registers (RNG_CR, RNG_DR, RNG_SR), Fault detection Clock checker, Data shift reg, Post-processing logic (8-bit LFSR x2), Sampling & Normalization (x 2), and Analog noise sources (source 1, source 2). It also shows external signals: 32-bit AHB Bus, rng_it, rng_hclk, and rng_clk. The diagram is divided into AHB clock domain and RNG clock domain. MSV42096V2 is noted at the bottom right.

19.3.2 RNG internal signals

Table 121 describes a list of useful-to-know internal signals available at the RNG level, not at the STM32 product level (on pads).

Table 121. RNG internal input/output signals

Signal nameSignal typeDescription
rng_itDigital outputRNG global interrupt request
rng_hclkDigital inputAHB clock
rng_clkDigital inputRNG dedicated clock, asynchronous to rng_hclk

19.3.3 Random number generation

The true random number generator (RNG) delivers truly random data through its AHB interface at deterministic intervals. Within its boundary the RNG implements the entropy source model pictured on Figure 117 , and provides three main functions to the application:

Figure 117. Entropy source model

Figure 117. Entropy source model diagram showing the internal components of the RNG. The diagram shows a stack of blocks: Noise Source (with a dice icon), Digitization, Post-processing (optional), and Conditioning (optional). The Noise Source, Digitization, and Post-processing blocks are grouped by a dashed box labeled 'Entropy source'. Arrows show data flow from Noise Source up through Digitization and Post-processing to Conditioning. An arrow labeled 'Raw data' points from Post-processing to Conditioning. An arrow points from Conditioning to 'Output'. Another arrow points from Conditioning to 'Health tests', which in turn points to 'Error message'. A separate 'Output (raw data or digitized noise source)' is shown with an arrow pointing to the Digitization block. The diagram is labeled MSv42095V1.
Figure 117. Entropy source model diagram showing the internal components of the RNG. The diagram shows a stack of blocks: Noise Source (with a dice icon), Digitization, Post-processing (optional), and Conditioning (optional). The Noise Source, Digitization, and Post-processing blocks are grouped by a dashed box labeled 'Entropy source'. Arrows show data flow from Noise Source up through Digitization and Post-processing to Conditioning. An arrow labeled 'Raw data' points from Post-processing to Conditioning. An arrow points from Conditioning to 'Output'. Another arrow points from Conditioning to 'Health tests', which in turn points to 'Error message'. A separate 'Output (raw data or digitized noise source)' is shown with an arrow pointing to the Digitization block. The diagram is labeled MSv42095V1.

The main components of the RNG are:

The components pictured above are detailed hereafter.

Noise source

The noise source is the component that contains the non-deterministic, entropy-providing activity that is ultimately responsible for the uncertainty associated with the bitstring output by the entropy source. It is composed of:

This noise source sampling is independent to the AHB interface clock frequency (rng_hclk).

Note: In Section 19.6 the recommended RNG clock frequencies are given.

Post processing

The sample values obtained from a true random noise source consist of 2-bit bitstrings. Because this noise source output is biased, the RNG implements a post-processing component that reduces that bias to a tolerable level.

The RNG post-processing consists of two stages, applied to each noise source bits:

This component is clocked by the RNG clock.

The times required between two random number generations, and between the RNG initialization and availability of first sample are described in Section 19.5 .

Output buffer

The RNG_DR data output register can store up to two 16-bit words, which have been output from the post-processing component (LFSR). In order to read back 32-bit random samples, it is required to wait 42 RNG clock cycles.

Whenever a random number is available through the RNG_DR register, the DRDY flag changes from 0 to 1. This flag remains high until the output buffer becomes empty after reading one word from the RNG_DR register.

Note: When interrupts are enabled an interrupt is generated when this data ready flag transitions from 0 to 1. Interrupt is then cleared automatically by the RNG as explained above.

Health checks

This component ensures that the entire entropy source (with its noise source) starts then operates as expected, obtaining assurance that failures are caught quickly and with a high probability and reliability.

The RNG implements the following health check features.

  1. 1. Continuous health tests, running indefinitely on the output of the noise source
    • – Repetition count test, flagging an error when:
      1. a) One of the noise source has provided more than 64 consecutive bits at a constant value (“0” or “1”)
      2. b) One of the noise sources has delivered more than 32 consecutive occurrence of two bits patterns (“01” or “10”)
  2. 2. Vendor specific continuous test
    • – Real-time “too slow” sampling clock detector, flagging an error when one RNG clock cycle is smaller than AHB clock cycle divided by 16.

The CECS and SECS status bits in the RNG_SR register indicate when an error condition is detected, as detailed in Section 19.3.7 .

Note: An interrupt can be generated when an error is detected.

19.3.4 RNG initialization

When a hardware reset occurs the following chain of events occurs:

  1. 1. The analog noise source is enabled, and the logic starts sampling the analog output after four RNG clock cycles, filling LFSR shift register and associated 16-bit post-processing shift register.
  2. 2. The output buffer is refilled automatically according to the RNG usage.

The associated initialization time can be found in Section 19.5 .

19.3.5 RNG operation

Normal operation

To run the RNG using interrupts, the following steps are recommended:

  1. 1. Enable the interrupts by setting the IE bit in the RNG_CR register. At the same time, enable the RNG by setting the bit RNGEN=1.
  2. 2. An interrupt is now generated when a random number is ready or when an error occurs. Therefore, at each interrupt, check that:
    • – No error occurred. The SEIS and CEIS bits must be set to 0 in the RNG_SR register.
    • – A random number is ready. The DRDY bit must be set to 1 in the RNG_SR register.
    • – If the above two conditions are true the content of the RNG_DR register can be read.

To run the RNG in polling mode following steps are recommended:

  1. 1. Enable the random number generation by setting the RNGEN bit to “1” in the RNG_CR register.
  2. 2. Read the RNG_SR register and check that:
    • – No error occurred (the SEIS and CEIS bits must be set to 0)
    • – A random number is ready (the DRDY bit must be set to 1)
  3. 3. If above conditions are true read the content of the RNG_DR register.

Note: When data is not ready (DRDY = 0) RNG_DR returns zero. It is recommended to always verify that RNG_DR is different from zero. Because when it is the case a seed error occurred between RNG_SR polling and RND_DR output reading (rare event).

Low-power operation

If the power consumption is a concern to the application, low-power strategies can be used, as described in Section 19.3.8 .

Software post-processing

If a NIST approved DRBG with 128 bits of security strength is required an approved random generator software must be built around the RNG true random number generator.

19.3.6 RNG clocking

The RNG runs on two different clocks: the AHB bus clock and a dedicated RNG clock.

The AHB clock is used to clock the AHB banked registers and the post-processing component. The RNG clock is used for noise source sampling. Recommended clock configurations are detailed in Section 19.6: RNG entropy source validation .

Note: When the CED bit in the RNG_CR register is set to 0, the RNG clock frequency must be higher than the AHB clock frequency divided by 16, otherwise the clock checker always flags a clock error (CECS = 1 in the RNG_SR register).

See Section 19.3.1: RNG block diagram for details (AHB and RNG clock domains).

19.3.7 Error management

In parallel to random number generation a health check block verifies the correct noise source behavior and the frequency of the RNG source clock as detailed in this section. Associated error state is also described.

Clock error detection

When the clock error detection is enabled (CED = 0) and if the RNG clock frequency is too low, the RNG sets to 1 both the CEIS and CECS bits to indicate that a clock error occurred. In this case, the application must check that the RNG clock is configured correctly (see Section 19.3.6: RNG clocking ) and then it must clear the CEIS bit interrupt flag. The CECS bit is automatically cleared when the clocking condition is normal.

Note: The clock error has no impact on generated random numbers that is the application can still read the RNG_DR register.

CEIS is set only when CECS is set to 1 by RNG.

Noise source error detection

When a noise source (or seed) error occurs, the RNG stops generating random numbers and sets to 1 both SEIS and SECS bits to indicate that a seed error occurred. If a value is available in the RNG_DR register, it must not be used as it may not have enough entropy.

In order to fully recover from a seed error application must clear the SEIS bit by writing it to “0”, then clear and set the RNGEN bit to reinitialize and restart the RNG.

19.3.8 RNG low-power use

If power consumption is a concern, the RNG can be disabled as soon as the DRDY bit is set to “1” by setting the RNGEN bit to “0” in the RNG_CR register. The 32-bit random value stored in the RNG_DR register is still available. If a new random is needed the application needs to re-enable the RNG and wait for 42+4 RNG clock cycles.

When disabling the RNG the user deactivates all the analog seed generators, whose power consumption is given in the datasheet electrical characteristics section.

19.4 RNG interrupts

In the RNG an interrupt can be produced on the following events:

Dedicated interrupt enable control bits are available as shown in Table 122 .

Table 122. RNG interrupt requests

Interrupt acronymInterrupt eventEvent flagEnable control bitInterrupt clear method
RNGData ready flagDRDYIENone (automatic)
Seed error flagSEISIEWrite 0 to SEIS.
Clock error flagCEISIEWrite 0 to CEIS

The user can enable or disable the above interrupt sources individually by changing the mask bits or the general interrupt control bit IE in the RNG_CR register. The status of the individual interrupt sources can be read from the RNG_SR register.

Note: Interrupts are generated only when RNG is enabled.

19.5 RNG processing time

The RNG can produce one 32-bit random numbers every 42 RNG clock cycles.

After enabling or re-enabling the RNG using the RNGEN bit, it takes 46 RNG clock cycles before random data are available.

19.6 RNG entropy source validation

19.6.1 Introduction

In order to assess the amount of entropy available from the RNG, STMicroelectronics has tested the peripheral using NIST SP800-22 rev1a statistical tests. The results can be provided on demand or the customer can reproduce the tests.

For more information on running this NIST statistical test suite, refer to STM32 microcontrollers random number generation validation using NIST statistical test suite application note (AN4230), available on STMicroelectronics website.

19.6.2 Validation conditions

STMicroelectronics has tested the RNG true random number generator in the following conditions:

Table 123. RNG configurations

ConfigurationRNG_CR bitsLoop number (N)RNG_HTCR registerRNG_NSCR register
NISTC bitRNG_CONFIG1 [5:0]CLKDIV [3:0]RNG_CONFIG2 [2:0] (1)RNG_CONFIG3 [3:0] (2)CED bit
ARefer to NIST compliant RNG configuration table in AN4230 available from www.st.com . This application note also indicates if this configuration is part of an existing NIST SP800-90B Entropy Certificate listed on https://csrc.nist.gov/projects/cryptographic-module-validation-program .
B10x180x0 (3)0x00x0010x0000 NA (4)default
C00x0F0x00xD02default
  1. 1. 0x1 value is recommended instead of 0x0 for RNG_CONFIG2[2:0], when RNG power consumption is critical. See the end of Section 19.3.8 for details.
  2. 2. RNG_CONFIG3[1:0] defines the loop number N: 0x0 corresponds to N=1, 0x1 to N=2, 0x2 to N=3, 0x3 to N=4
  3. 3. The noise source sampling must be NA or less. Hence, if the RNG clock is different from NA, this value of CLKDIV must be adapted. See the CLKDIV bitfield description in Section 19.7.1 for details.
  4. 4. This value can be fixed in the RNG driver (it does not depend upon the STM32 product).

19.6.3 Data collection

In order to run statistical tests, it is required to collect samples from the entropy source at the raw data level as well as at the output of the entropy source. For details on data collection and the running of statistical test suites refer to AN4230 “ STM32 microcontrollers random number generation validation using NIST statistical test suite ” available from www.st.com .

In bypass mode the bits [31:30] of the fourth word are always stuck at 0. Hence the continuous capture of samples is started from the fifth word.

19.7 RNG registers

The RNG is associated with a control register, a data register and a status register.

19.7.1 RNG control register (RNG_CR)

Address offset: 0x000

Reset value: 0x0000 0000

31302928272625242322212019181716
Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.
1514131211109876543210
Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.CEDRes.IERNGENRes.Res.
rwrwrw

Bits 31:6 Reserved, must be kept at reset value.

Bit 5 CED : Clock error detection

0: Clock error detection enabled

1: Clock error detection is disabled

The clock error detection cannot be enabled nor disabled on-the-fly when the RNG is enabled, that is to enable or disable CED, the RNG must be disabled.

Bit 4 Reserved, must be kept at reset value.

Bit 3 IE : Interrupt enable

0: RNG interrupt is disabled

1: RNG interrupt is enabled. An interrupt is pending as soon as the DRDY, SEIS, or CEIS is set in the RNG_SR register.

Bit 2 RNGEN : True random number generator enable

0: True random number generator is disabled. Analog noise sources are powered off and logic clocked by the RNG clock is gated.

1: True random number generator is enabled.

Bits 1:0 Reserved, must be kept at reset value.

19.7.2 RNG status register (RNG_SR)

Address offset: 0x004

Reset value: 0x0000 0000

31302928272625242322212019181716
Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.
1514131211109876543210
Res.Res.Res.Res.Res.Res.Res.Res.Res.SEISCEISRes.Res.SECSCECSDRDY
rc_w0rc_w0rrr

Bits 31:7 Reserved, must be kept at reset value.

Bit 6 SEIS : Seed error interrupt status

This bit is set at the same time as SECS. It is cleared by writing 0. Writing 1 has no effect.

0: No faulty sequence detected

1: At least one faulty sequence is detected. See SECS bit description for details.

An interrupt is pending if IE = 1 in the RNG_CR register.

Bit 5 CEIS : Clock error interrupt status

This bit is set at the same time as CECS. It is cleared by writing 0. Writing 1 has no effect.

0: The RNG clock is correct ( \( f_{\text{RNGCLK}} > f_{\text{HCLK}}/16 \) )

1: The RNG is detected too slow ( \( f_{\text{RNGCLK}} < f_{\text{HCLK}}/16 \) )

An interrupt is pending if IE = 1 in the RNG_CR register.

Bits 4:3 Reserved, must be kept at reset value.

Bit 2 SECS : Seed error current status

0: No faulty sequence has currently been detected. If the SEIS bit is set, this means that a faulty sequence was detected and the situation has been recovered.

1: One of the noise sources has provided more than 64 consecutive bits at a constant value ("0" or "1"), or more than 32 consecutive occurrence of two bits patterns ("01" or "10")

Bit 1 CECS : Clock error current status

0: The RNG clock is correct ( \( f_{RNGCLK} > f_{HCLK}/16 \) ). If the CEIS bit is set, this means that a slow clock was detected and the situation has been recovered.

1: The RNG clock is too slow ( \( f_{RNGCLK} < f_{HCLK}/16 \) ).

Note: CECS bit is valid only if the CED bit in the RNG_CR register is set to 0.

Bit 0 DRDY : Data ready

0: The RNG_DR register is not yet valid, no random data is available.

1: The RNG_DR register contains valid random data.

Once the RNG_DR register has been read, this bit returns to 0 until a new random value is generated.

If IE=1 in the RNG_CR register, an interrupt is generated when DRDY = 1.

19.7.3 RNG data register (RNG_DR)

Address offset: 0x008

Reset value: 0x0000 0000

The RNG_DR register is a read-only register that delivers a 32-bit random value when read. After being read, this register delivers a new random value after 42 periods of RNG clock if the output FIFO is empty.

The content of this register is valid when the DRDY = 1 and the value is not 0x0, even if RNGEN = 0.

31302928272625242322212019181716
RNDATA[31:16]
rrrrrrrrrrrrrrrr
1514131211109876543210
RNDATA[15:0]
rrrrrrrrrrrrrrrr

Bits 31:0 RNDATA[31:0] : Random data

32-bit random data, which are valid when DRDY = 1. When DRDY = 0, the RNDATA value is zero.

When DRDY is set, it is recommended to always verify that RNG_DR is different from zero.

The zero value means that a seed error occurred between RNG_SR polling and RND_DR output reading (a rare event).

19.7.4 RNG register map

Table 124. RNG register map and reset map

OffsetRegister name313029282726252423222120191817161514131211109876543210
0x000RNG_CRRes.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.CEDRes.IERNGENRes.Res.
Reset value000000
0x004RNG_SRRes.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.Res.SEISCEISRes.Res.Res.SECSCECSDRDY
Reset value00000000
0x008RNG_DRRNDATA[31:0]
Reset value00000000000000000000000000000000

Refer to Section 2.2: Memory organization for the register boundary addresses.