RDRAND (for "read random"; known as Intel Secure Key Technology, previously known as Bull Mountain) is an instruction for returning random numbers from an Intel on-chip hardware random number generator which has been seeded by an on-chip entropy source.
RDRAND is available in Ivy Bridge processors[a] and is part of the Intel 64 and IA-32 instruction set architectures. AMD added support for the instruction in June 2015.
The random number generator is compliant with security and cryptographic standards such as NIST SP 800-90A, FIPS 140-2, and ANSI X9.82. Intel also requested Cryptography Research Inc. to review the random number generator in 2012, which resulted in the paper Analysis of Intel's Ivy Bridge Digital Random Number Generator.
RDSEED is similar to
RDRAND and provides lower-level access to the entropy-generating hardware. The
RDSEED generator and processor instruction
rdseed are available with Intel Broadwell CPUs and AMD Zen CPUs.
CPUID instruction can be used to check whether the central processing unit (CPU) supports the
RDRAND instruction on both AMD and Intel CPUs. If supported, bit 30 of the ECX register is set after calling CPUID standard function
01H. AMD processors are checked for the feature using the same test.
RDSEED availability can be checked on Intel CPUs in a similar manner. If
RDSEED is supported, the bit 18 of the EBX register is set after calling CPUID standard function
The opcode for
0x0F 0xC7, followed by a ModRM byte that specifies the destination register and optionally combined with a REX prefix in 64-bit mode.
Intel Secure Key is Intel's name for both the
RDRAND instruction and the underlying random number generator (RNG) hardware implementation, which was codenamed "Bull Mountain" during development. Intel calls their RNG a "digital random number generator" or DRNG. The generator takes pairs of 256-bit raw entropy samples generated by the hardware entropy source and applies them to an Advanced Encryption Standard (AES) (in CBC-MAC mode) conditioner which reduces them to a single 256-bit conditioned entropy sample. A deterministic random-bit generator called CTR_DRBG defined in NIST SP 800-90A is seeded by the output from the conditioner, providing cryptographically secure random numbers to applications requesting them via the
RDRAND instruction. The hardware will issue a maximum of 511 128-bit samples before changing the seed value. Using the
RDSEED operation provides access to the conditioned 256-bit samples from the AES-CBC-MAC.
RDSEED instruction was added to Intel Secure Key for seeding another pseudorandom number generator, available in Broadwell CPUs. The entropy source for the
RDSEED instruction runs asynchronously on a self-timed circuit and uses thermal noise within the silicon to output a random stream of bits at the rate of 3 GHz, slower than the effective 6.4 Gbit/s obtainable from
RDRAND (both rates are shared between all cores and threads). The
RDSEED instruction is intended for seeding a software PRNG of arbitrary width, whereas the
RDRAND is intended for applications that merely require high-quality random numbers. If cryptographic security is not required, a software PRNG such as Xorshift is usually faster.
On an Intel Core i7-7700K, 4500 MHz (45 x 100 MHz) processor (Kaby Lake-S microarchitecture), a single
RDSEED instruction takes 110ns or 463 clock cycles, regardless of the operand size (16/32/64 bits). This number of clock cycles applies to all processors with Skylake or Kaby Lake microarchitecture. On the Silvermont microarchitecture processors, each of the instructions take around 1472 clock cycles, regardless of the operand size; and on Ivy Bridge processors
RDRAND takes up to 117 clock cycles.
On an AMD Ryzen CPU, each of the instructions takes around 1200 clock cycles for 16-bit or 32-bit operand, and around 2500 clock cycles for a 64-bit operand.
An astrophysical Monte Carlo simulator examined the time to generate 107 64-bit random numbers using
RDRAND on a quad-core Intel i7-3740 QM processor. They found that a C implementation of
RDRAND ran about 2× slower than the default random number generator in C, and about 20× slower than the Mersenne Twister. Although a Python module of
RDRAND has been constructed, it was found to be 20× slower than the default random number generator in Python.
A microcode update released by Intel in June 2020, designed to mitigate the CrossTalk vulnerability (see the security issues section below), negatively impacts the performance of
RDSEED due to additional security controls. On processors with the mitigations applied, each affected instruction incurs additional latency and simultaneous execution of
RDSEED across cores is effectively serialised. Intel introduced a mechanism to relax these security checks, thus reducing the performance impact in most scenarios, but Intel processors do not apply this security relaxation by default.
Visual C++ 2015 provides intrinsic wrapper support for the
RDSEED functions. GCC 4.6+ and Clang 3.2+ provide intrinsic functions for
RDRAND when -mrdrnd is specified in the flags, also setting __RDRND__ to allow conditional compilation. Newer versions additionally provide
immintrin.h to wrap these built-ins into functions compatible with version 12.1+ of Intel's C Compiler. These functions write random data to the location pointed to by their parameter, and return 1 on success.
It is an option to generate cryptographically-secure random numbers using
RDSEED in OpenSSL, to help secure communications.
The first[dubious ] scientific application of
RDRAND can be found in astrophysics. Radio observations of low-mass stars and brown dwarfs have revealed that a number of them emit bursts of radio waves. These radio waves are caused by magnetic reconnection, the same process that causes solar flares on the Sun.
RDRAND was used to generate large quantities of random numbers for a Monte Carlo simulator, to model physical properties of the brown dwarfs and the effects of the instruments that observe them. They found that about 5% of brown dwarfs are sufficiently magnetic to emit strong radio bursts. They also evaluated the performance of the
RDRAND instruction in C and Python compared to other random number generators.
In September 2013, in response to a New York Times article revealing the NSA's effort to weaken encryption, Theodore Ts'o publicly posted concerning the use of
/dev/random in the Linux kernel:
I am so glad I resisted pressure from Intel engineers to let
/dev/randomrely only on the
RDRANDinstruction. To quote from the [New York Times article]: 'By this year, the Sigint Enabling Project had found ways inside some of the encryption chips that scramble information for businesses and governments, either by working with chipmakers to insert back doors...' Relying solely on the hardware random number generator which is using an implementation sealed inside a chip which is impossible to audit is a BAD idea.
Linus Torvalds dismissed concerns about the use of
RDRAND in the Linux kernel, and pointed out that it is not used as the only source of entropy for
/dev/random, but rather used to improve the entropy by combining the values received from
RDRAND with other sources of randomness. However, Taylor Hornby of Defuse Security demonstrated that the Linux random number generator could become insecure if a backdoor is introduced into the
RDRAND instruction that specifically targets the code using it. Hornby's proof-of-concept implementation works on an unmodified Linux kernel prior to version 3.13. The issue was fixed in the Linux kernel in 2013.
Developers changed the FreeBSD kernel away from using
RDRAND and VIA PadLock directly with the comment "For FreeBSD 10, we are going to backtrack and remove
RDRAND and Padlock backends and feed them into Yarrow instead of delivering their output directly to /dev/random. It will still be possible to access hardware random number generators, that is,
RDRAND, Padlock etc., directly by inline assembly or by using OpenSSL from userland, if required, but we cannot trust them any more." FreeBSD /dev/random uses Fortuna and RDRAND started from FreeBSD 11.
On June 9th 2020, researchers from Vrije Universiteit Amsterdam published a side-channel attack named CrossTalk (CVE-2020-0543) that affected
RDRAND on a number of Intel processors. They discovered that outputs from the hardware digital random number generator (DRNG) were stored in a staging buffer that was shared across all cores. The vulnerability allowed malicious code running on an affected processor to read
RDSEED instruction results from a victim application running on another core of that same processor, including applications running inside Intel SGX enclaves. The researchers developed a proof of concept exploit which extracted a complete ECDSA key from an SGX enclave running on a separate CPU core, after only one signature operation. The vulnerability affects scenarios where untrusted code runs alongside trusted code on the same processor, such as in a shared hosting environment.
Intel refers to the CrossTalk vulnerability as Special Register Buffer Data Sampling (SRBDS). In response to the research, Intel released microcode updates to mitigate the issue. The updated microcode ensures that off-core accesses are delayed until sensitive operations - specifically the
EGETKEY instructions - are completed and the staging buffer has been overwritten. The SRBDS attack also affects other instructions, such as those that read MSRs, but Intel did not apply additional security protections to them due to performance concerns and the reduced need for confidentiality of those instructions' results. A wide range of Intel processors released between 2012 and 2019 were affected, including desktop, mobile, and server processors. The mitigations themselves resulted in negative performance impacts when using the affected instructions, particularly when executed in parallel by multi-threaded applications, due to increased latency introduced by the security checks and the effective serialisation of affected instructions across cores. Intel introduced an opt-out option, configurable via the
IA32_MCU_OPT_CTRL MSR on each logical processor, which improves performance by disabling the additional security checks for instructions executing outside of an SGX enclave.
- In some Ivy Bridge versions, due to a bug, the RDRAND instruction causes an Illegal Instruction exception.
- "What is Intel® Secure Key Technology?". Intel. Retrieved 2020-09-23.
- Hofemeier, Gael (2011-06-22). "Find out about Intel's new RDRAND Instruction". Intel Developer Zone Blogs. Retrieved 30 December 2013.
- "Intel Digital Random Number Generator (DRNG): Software Implementation Guide, Revision 1.1" (PDF). Intel Corporation. 2012-08-07. Retrieved 2012-11-25.
- Desktop 3rd Generation Intel Core Processor Family, Specification Update (PDF). Intel Corporation. January 2013.
- "AMD64 Architecture Programmer's Manual Volume 3: General-Purpose and System Instructions" (PDF). AMD Developer Guides, Manuals & ISA Documents. June 2015. Retrieved 16 October 2015.
- Barker, Elaine; Kelsey, John (January 2012). "Recommendation for Random Number Generation Using Deterministic Random Bit Generators" (PDF). National Institute of Standards and Technology. Retrieved September 16, 2013. Cite journal requires
- Hamburg, Mike; Kocher, Paul; Marson, Mark (2012-03-12). "Analysis of Intel's Ivy Bridge Digital Random Number Generator" (PDF). Cryptography Research, Inc. Archived from the original (PDF) on 2014-12-30. Retrieved 2015-08-21.
- Hofemeier, Gael (2012-07-26). "Introduction to Intel AES-NI and Intel SecureKey Instructions". Intel Developer Zone. Intel. Retrieved 2015-10-24.
- "AMD Starts Linux Enablement On Next-Gen "Zen" Architecture - Phoronix". www.phoronix.com. Retrieved 2015-10-25.
- "Volume 1, Section 7.3.17, 'Random Number Generator Instruction'" (PDF). Intel® 64 and IA-32 Architectures Software Developer’s Manual Combined Volumes: 1, 2A, 2B, 2C, 3A, 3B and 3C. Intel Corporation. June 2013. p. 177. Retrieved 24 June 2013.
All Intel processors that support the RDRAND instruction indicate the availability of the RDRAND instruction via reporting CPUID.01H:ECX.RDRAND[bit 30] = 1
- "AMD64 Architecture Programmer's Manual Volume 3: General-Purpose and System Instructions" (PDF). AMD. June 2015. p. 278. Retrieved 15 October 2015.
Support for the RDRAND instruction is optional. On processors that support the instruction, CPUID Fn0000_0001_ECX[RDRAND] = 1
"Volume 1, Section 7.3.17, 'Random Number Generator Instruction'" (PDF). Intel® 64 and IA-32 Architectures Software Developer’s Manual Combined Volumes: 1, 2A, 2B, 2C, 3A, 3B and 3C. Intel Corporation. June 2013. p. 177. Retrieved 25 October 2015.
All Intel processors that support the RDSEED instruction indicate the availability of the RDSEED instruction via reporting CPUID.(EAX=07H, ECX=0H):EBX.RDSEED[bit 18] = 1
- "Intel® Digital Random Number Generator (DRNG) Software Implementation Guide | Intel® Developer Zone". Software.intel.com. Retrieved 2014-01-30.
- Taylor, Greg; Cox, George (September 2011). "Behind Intel's New Random-Number Generator". IEEE Spectrum.
- John Mechalas (November 2012). "The Difference Between RDRAND and RDSEED". software.intel.com. Intel Corporation. Retrieved 1 January 2014.
- Mechalas, John. "Intel Digital Random Number Generator (DRNG) Software Implementation Guide, Section 3.2.1 Entropy Source (ES)". Intel Software. Intel. Retrieved 18 February 2015.
- https://software.intel.com/en-us/articles/intel-digital-random-number-generator-drng-software-implementation-guide says 800 megabytes, which is 6.4 gigabits, per second
- The simplest 64-bit implementation of Xorshift has 3 XORs and 3 shifts; if these are executed in a tight loop on 4 cores at 2GHz, the throughput is 80 Gb/sec. In practice it will be less due to load/store overheads etc, but is still likely to exceed the 6.4 Gb/sec of
RDRAND. On the other hand, the quality of
RDRAND's numbers should be higher than that of a software PRNG like Xorshift.
- Route, Matthew (August 10, 2017). "Radio-flaring Ultracool Dwarf Population Synthesis". The Astrophysical Journal. 845: 66. arXiv:1707.02212. doi:10.3847/1538-4357/aa7ede.
- "Special Register Buffer Data Sampling". Intel. Retrieved 26 December 2020.
- "x86 intrinsics list". docs.microsoft.com. 2020-02-28. Retrieved 2020-05-07.
- "X86 Built-in Functions - Using the GNU Compiler Collection (GCC)".
- "Intel® C++ Compiler 19.1 Developer Guide and Reference". 2019-12-23.
- Perlroth, Nicole; Larson, Jeff; Shane, Scott (September 5, 2013). "N.S.A. Able to Foil Basic Safeguards of Privacy on Web". The New York Times. Retrieved November 15, 2017.
- Ts'o, Theodore (September 6, 2013). "I am so glad I resisted pressure from Intel engineers to let /dev/random rely..." Archived from the original on 2018-06-11.
- Richard Chirgwin (2013-12-09). "FreeBSD abandoning hardware randomness". The Register.
- Gavin Clarke (10 September 2013). "Torvalds shoots down call to yank 'backdoored' Intel RDRAND in Linux crypto". theregister.co.uk. Retrieved 12 March 2014.
- Taylor Hornby (6 December 2013). "RDRAND backdoor proof of concept is working! Stock kernel (3.8.13), only the RDRAND instruction is modified". Retrieved 9 April 2015.
- Taylor Hornby [@DefuseSec] (10 September 2013). "I wrote a short dialogue explaining why Linux's use of RDRAND is problematic. http://pastebin.com/A07q3nL3 /cc @kaepora @voodooKobra" (Tweet). Retrieved 11 January 2016 – via Twitter.
- Daniel J. Bernstein; Tanja Lange (16 May 2014). "Randomness generation" (PDF). Retrieved 9 April 2015.
- Hornby, Taylor (2017-05-09). "You want to keep RDRAND enabled. What I did just showed that in an older version of the kernel RDRAND could potentially control the output". @DefuseSec. Retrieved 2019-10-30.
- "FreeBSD Quarterly Status Report". Freebsd.org. Retrieved 2014-01-30.
- "random(4)". www.freebsd.org. Retrieved 2020-09-25.
- Ragab, Hany; Milburn, Alyssa; Razavi, Kaveh; Bos, Herbert; Giuffrida, Cristiano. "CrossTalk: Speculative Data Leaks Across Cores Are Real" (PDF). Systems and Network Security Group, Vrije Universiteit Amsterdam (VUSec). Retrieved 26 December 2020.
- "VUSec RIDL cpuid_leak PoC, modified to leak rdrand output". GitHub. Retrieved 26 December 2020.
- "Processors Affected: Special Register Buffer Data Sampling". Intel Developer Zone. Retrieved 26 December 2020.