Wikipedia

Address space layout randomization

(Redirected from Address Space Layout Randomization)

Address space layout randomization (ASLR) is a computer security technique involved in preventing exploitation of memory corruption vulnerabilities.[1] In order to prevent an attacker from reliably redirecting code execution to, for example, a particular exploited function in memory, ASLR randomly arranges the address space positions of key data areas of a process, including the base of the executable and the positions of the stack, heap and libraries.

History

edit

The Linux PaX project first coined the term "ASLR", and published the first design and implementation of ASLR in July 2001 as a patch for the Linux kernel. It is seen as a complete implementation, providing a patch for kernel stack randomization since October 2002.[2]

The first mainstream operating system to support ASLR by default was OpenBSD version 3.4 in 2003,[3][4] followed by Linux in 2005.

Benefits

edit

Address space randomization hinders some types of security attacks by making it more difficult for an attacker to predict target addresses. For example, attackers trying to execute return-to-libc attacks must locate the code to be executed, while other attackers trying to execute shellcode injected on the stack have to find the stack first. In both cases, the system makes related memory-addresses unpredictable from the attackers' point of view. These values have to be guessed, and a mistaken guess is not usually recoverable due to the application crashing.

Effectiveness

edit

Address space layout randomization is based upon the low chance of an attacker guessing the locations of randomly placed areas. Security is increased by increasing the search space. Thus, address space randomization is more effective when more entropy is present in the random offsets. Entropy is increased by either raising the amount of virtual memory area space over which the randomization occurs or reducing the period over which the randomization occurs. The period is typically implemented as small as possible, so most systems must increase VMA space randomization.

To defeat the randomization, attackers must successfully guess the positions of all areas they wish to attack. For data areas such as stack and heap, where custom code or useful data can be loaded, more than one state can be attacked by using NOP slides for code or repeated copies of data. This allows an attack to succeed if the area is randomized to one of a handful of values. In contrast, code areas such as library base and main executable need to be discovered exactly. Often these areas are mixed, for example stack frames are injected onto the stack and a library is returned into.

The following variables can be declared:

  •   (entropy bits of stack top)
  •   (entropy bits of mmap() base)
  •   (entropy bits of main executable base)
  •   (entropy bits of heap base)
  •   (attacked bits per attempt of stack entropy)
  •   (attacked bits per attempt of mmap() base entropy)
  •   (attacked bits per attempt of main executable entropy)
  •   (attacked bits per attempt of heap base entropy)
  •   (attempts made)
  •   (total amount of entropy:  )

To calculate the probability of an attacker succeeding, a number of attempts α carried out without being interrupted by a signature-based IPS, law enforcement, or other factor must be assumed; in the case of brute forcing, the daemon cannot be restarted. The number of relevant bits and how many are being attacked in each attempt must also be calculated, leaving however many bits the attacker has to defeat.

The following formulas represent the probability of success for a given set of α attempts on N bits of entropy.

  •   (isolated guessing; address space is re-randomized after each attempt)
  •   (systematic brute forcing on copies of the program with the same address space)

In many systems,   can be in the thousands or millions. On 32-bit systems, a typical amount of entropy N is 8 bits.[5] For 2004 computer speeds, Shacham and co-workers state "... 16 bits of address randomization can be defeated by a brute force attack within minutes."[6] (The authors' statement depends on the ability to attack the same application multiple times without any delay. Proper implementations of ASLR, like that included in grsecurity, provide several methods to make such brute force attacks infeasible. One method involves preventing an executable from executing for a configurable amount of time if it has crashed a certain number of times.) On modern 64-bit systems, these numbers typically reach the millions at least.[citation needed]

Android,[7][non-primary source needed] and possibly other systems,[which?] implement Library Load Order Randomization, a form of ASLR which randomizes the order in which libraries are loaded. This supplies very little entropy. An approximation of the number of bits of entropy supplied per needed library appears below; this does not yet account for varied library sizes, so the actual entropy gained is really somewhat higher. Attackers usually need only one library; the math is more complex with multiple libraries, and shown below as well. The case of an attacker using only one library is a simplification of the more complex formula for  .

  • l (number of libraries loaded)
  • β (number of libraries used by the attacker)
  •  

These values tend to be low even for large values of l, most importantly since attackers typically can use only the C standard library and thus one can often assume that  . However, even for a small number of libraries there are a few bits of entropy gained here; it is thus potentially interesting to combine library load order randomization with VMA address randomization to gain a few extra bits of entropy. These extra bits of entropy will not apply to other mmap() segments, only libraries.

Reducing entropy

edit

Attackers may make use of several methods to reduce the entropy present in a randomized address space, ranging from simple information leaks to attacking multiple bits of entropy per attack (such as by heap spraying). There is little that can be done about this.

It is possible to leak information about memory layout using format string vulnerabilities. Format string functions such as printf use a variable argument list to do their job; format specifiers describe what the argument list looks like. Because of the way arguments are typically passed, each format specifier moves closer to the top of the stack frame. Eventually, the return pointer and stack frame pointer can be extracted, revealing the address of a vulnerable library and the address of a known stack frame; this can eliminate library and stack randomization as an obstacle to an attacker.

One can also decrease entropy in the stack or heap. The stack typically must be aligned to 16 bytes, and so this is the smallest possible randomization interval; while the heap must be page-aligned, typically 4096 bytes. When attempting an attack, it is possible to align duplicate attacks with these intervals; a NOP slide may be used with shellcode injection, and the string '/bin/sh' can be replaced with '////////bin/sh' for an arbitrary number of slashes when attempting to return to system. The number of bits removed is exactly   for n intervals attacked.

Such decreases are limited due to the amount of data in the stack or heap. The stack, for example, is typically limited to 8 MB[8] and grows to much less; this allows for at most 19 bits, although a more conservative estimate would be around 8–10 bits corresponding to 4–16 KB[8] of stack stuffing. The heap on the other hand is limited by the behavior of the memory allocator; in the case of glibc, allocations above 128 KB are created using mmap, limiting attackers to 5 bits of reduction. This is also a limiting factor when brute forcing; although the number of attacks to perform can be reduced, the size of the attacks is increased enough that the behavior could in some circumstances become apparent to intrusion detection systems.

Limitations

edit

ASLR-protected addresses can be leaked by various side channels, removing mitigation utility. Recent attacks have used information leaked by the CPU branch target predictor buffer (BTB) or memory management unit (MMU) walking page tables. It is not clear if this class of ASLR attack can be mitigated. If they cannot, the benefit of ASLR is reduced or eliminated.

Empirical analysis

edit

In August 2024 a paper [9] was published with an empirical analysis of major desktop platforms, including Linux, macOS, and Windows, by examining the variability in the placement of memory objects across various processes, threads, and system restarts. The results show that while some systems as of 2024, like Linux distributions, provide robust randomization, others, like Windows and macOS, often fail to adequately randomize key areas like executable code and libraries. Moreover, they found a significant entropy reduction in the entropy of libraries after the Linux 5.18 version and identify correlation paths that an attacker could leverage to reduce exploitation complexity significantly.

Implementations

edit

Several mainstream, general-purpose operating systems implement ASLR.

Android

edit

Android 4.0 Ice Cream Sandwich provides address space layout randomization (ASLR) to help protect system and third-party applications from exploits due to memory-management issues. Position-independent executable support was added in Android 4.1.[10] Android 5.0 dropped non-PIE support and requires all dynamically linked binaries to be position independent.[11][12] Library load ordering randomization was accepted into the Android open-source project on 26 October 2015,[7][non-primary source needed] and was included in the Android 7.0 release.

DragonFly BSD

edit

DragonFly BSD has an implementation of ASLR based upon OpenBSD's model, added in 2010.[13] It is off by default, and can be enabled by setting the sysctl vm.randomize_mmap to 1.

FreeBSD

edit

Support for ASLR appeared in FreeBSD 13.0.[14][15] It is enabled by default since 13.2.[16]

iOS (iPhone, iPod touch, iPad)

edit

Apple introduced ASLR in iOS 4.3 (released March 2011).[17]

KASLR was introduced in iOS 6.[18] The randomized kernel base is 0x01000000 + ((1+0xRR) * 0x00200000), where 0xRR is a random byte from SHA1 (random data) generated by iBoot (the 2nd-stage iOS Boot Loader).[19]

Linux

edit

The Linux kernel enabled a weak form of ASLR by default since the kernel version 2.6.12, released in June 2005.[20] The PaX and Exec Shield patchsets to the Linux kernel provide more complete implementations. The Exec Shield patch for Linux supplies 19 bits of stack entropy on a period of 16 bytes, and 8 bits of mmap base randomization on a period of 1 page of 4096 bytes. This places the stack base in an area 8 MB wide containing 524,288 possible positions, and the mmap base in an area 1 MB wide containing 256 possible positions.

ASLR can be disabled for a specific process by changing its execution domain, using personality(2).[21] A number of sysctl options control the behavior of mainline ASLR. For example, kernel.randomize_va_space controls what to randomize; the strongest option is 2. vm.mmap_rnd_bits controls how many bits to randomize for mmap.[22]

Position-independent executable (PIE) implements a random base address for the main executable binary and has been in place since April 18, 2004. It provides the same address randomness to the main executable as being used for the shared libraries. The PIE feature cannot be used together with the prelink feature for the same executable. The prelink tool implements randomization at prelink time rather than runtime, because by design prelink aims to handle relocating libraries before the dynamic linker has to, which allows the relocation to occur once for many runs of the program. As a result, real address space randomization would defeat the purpose of prelinking.

In 2014, Marco-Gisbert and Ripoll disclosed offset2lib technique that weakens Linux ASLR for PIE executables. Linux kernels load PIE executables right after their libraries; as a result, there is a fixed offset between the executable and the library functions. If an attacker finds a way to find the address of a function in the executable, the library addresses are also known. They demonstrated an attack that finds the address in fewer than 400 tries. They proposed a new randomize_va_space=3 option to randomize the placement of the executable relative to the library,[5] but it is yet to be incorporated into the upstream as of 2024.[23]

The Linux kernel 5.18 released May 2022 reduced the effectiveness of both 32-bit and 64-bit implementations. Linux filesystems call thp_get_unmapped_area to respond to a file-backed mmap. With a change in 5.18, files greater than 2 MiB are made to return 2 MiB-aligned addresses, so they can be potentially backed by huge pages. (Previously, the increased alignment only applied to Direct Access (DAX) mappings.) In the meantime, the C library (libc) has, over time, grown in size to exceed this 2 MiB threshold, so instead of being aligned to a (typically) 4 KiB page boundary as before, these libraries are now 2 MiB-aligned: a loss of 9 bits of entropy. For 32-bit Linux, many distributions show no randomization at all in the placement of the libc. For 64-bit Linux, the 28 bits of entropy is reduced to 19 bits. In response, Ubuntu has increased its mmap_rnd_bits setting.[24] Martin Doucha added a Linux Test Project testcase to detect this issue.[25]

Kernel address space layout randomization

edit

Kernel address space layout randomization (KASLR) enables address space randomization for the Linux kernel image by randomizing where the kernel code is placed at boot time.[26] KASLR was merged into the Linux kernel mainline in kernel version 3.14, released on 30 March 2014.[27] When compiled in, it can be disabled at boot time by specifying nokaslr as one of the kernel's boot parameters.[28]

There are several side-channel attacks in x86 processors that could leak kernel addresses.[29][30] In late 2017, kernel page-table isolation (KPTI aka KAISER) was developed to defeat these attacks.[31][32] However, this method cannot protect against side-channel attacks utilizing collisions in branch predictor structures.[33]

As of 2021, finer grained kernel address space layout randomization (or function granular KASLR, FGKASLR) is a planned extension of KASLR to randomize down to the function level.[34]

Microsoft Windows

edit

Microsoft's Windows Vista (released January 2007) and later have ASLR enabled only for executables and dynamic link libraries that are specifically linked to be ASLR-enabled.[35] For compatibility, it is not enabled by default for other applications. Typically, only older software is incompatible and ASLR can be fully enabled by editing a registry entry HKLM\SYSTEM\CurrentControlSet\Control\Session Manager\Memory Management\MoveImages,[36] or by installing Microsoft's Enhanced Mitigation Experience Toolkit.

The locations of the heap, stack, Process Environment Block, and Thread Environment Block are also randomized. A security whitepaper from Symantec noted that ASLR in 32-bit Windows Vista may not be as robust as expected, and Microsoft has acknowledged a weakness in its implementation.[37]

Host-based intrusion prevention systems such as WehnTrust[38] and Ozone[39] also offer ASLR for Windows XP and Windows Server 2003 operating systems. WehnTrust is open-source.[40] Complete details of Ozone's implementation are not available.[41]

It was noted in February 2012[42] that ASLR on 32-bit Windows systems prior to Windows 8 can have its effectiveness reduced in low memory situations. A similar effect also had been achieved on Linux in the same research. The test code caused the Mac OS X 10.7.3 system to kernel panic, so it was left unclear about its ASLR behavior in this scenario.

NetBSD

edit

Support for ASLR in userland appeared in NetBSD 5.0 (released April 2009),[43] and was enabled by default in NetBSD-current in April 2016.[44]

Kernel ASLR support on amd64 was added in NetBSD-current in October 2017, making NetBSD the first BSD system to support KASLR.[45]

OpenBSD

edit

In 2003, OpenBSD became the first mainstream operating system to support a strong form of ASLR and to activate it by default.[3] OpenBSD completed its ASLR support in 2008 when it added support for PIE binaries.[46] OpenBSD 4.4's malloc(3) was designed to improve security by taking advantage of ASLR and gap page features implemented as part of OpenBSD's mmap system call, and to detect use-after-free bugs.[47] Released in 2013, OpenBSD 5.3 was the first mainstream operating system to enable position-independent executables by default on multiple hardware platforms, and OpenBSD 5.7 activated position-independent static binaries (Static-PIE) by default.[46]

macOS

edit

In Mac OS X Leopard 10.5 (released October 2007), Apple introduced randomization for system libraries.[48]

In Mac OS X Lion 10.7 (released July 2011), Apple expanded their implementation to cover all applications, stating "address space layout randomization (ASLR) has been improved for all applications. It is now available for 32-bit apps (as are heap memory protections), making 64-bit and 32-bit applications more resistant to attack."[49]

As of OS X Mountain Lion 10.8 (released July 2012) and later, the entire system including the kernel as well as kexts and zones are randomly relocated during system boot.[50]

Solaris

edit

ASLR has been introduced in Solaris beginning with Solaris 11.1 (released October 2012). ASLR in Solaris 11.1 can be set system-wide, per zone, or on a per-binary basis.[51]

Exploitation

edit

A side-channel attack utilizing branch target buffer was demonstrated to bypass ASLR protection.[33] In 2017, an attack named "ASLR⊕Cache" was demonstrated which could defeat ASLR in a web browser using JavaScript.[52]

See also

edit

References

edit
  1. ^ Marco-Gisbert, Hector; Ripoll Ripoll, Ismael (2019-07-22). "Address Space Layout Randomization Next Generation". Applied Sciences. 9 (14): 2928. doi:10.3390/app9142928. hdl:10251/144813. ISSN 2076-3417.
  2. ^ Brad Spengler (October 2003). "PaX: The Guaranteed End of Arbitrary Code Execution" (PDF). grsecurity.net. Slides 22 through 35. Archived (PDF) from the original on 2020-10-25. Retrieved 2015-08-20.
  3. ^ a b Theo De Raadt (2005). "Exploit Mitigation Techniques (updated to include random malloc and mmap) at OpenCON 2005". Archived from the original on 2012-07-16. Retrieved 2009-08-26.
  4. ^ "OpenBSD Innovations". The OpenBSD project. Archived from the original on 2016-09-09. Retrieved 2016-09-12.
  5. ^ a b Marco-Gisbert, Hector; Ripoll, Ismael (2014-11-20). "On the Effectiveness of Full-ASLR on 64-bit Linux" (PDF). Archived from the original (PDF) on 2015-05-08. Retrieved 2016-03-29.
  6. ^ Shacham, H.; Page, M.; Pfaff, B.; Goh, E.J.; Modadugu, N.; Boneh, D (2004). On the Effectiveness of Address-Space Randomization. 11th ACM conference on Computer and communications security. pp. 298–307.
  7. ^ a b "Implement Library Load Order Randomization". Archived from the original on 2023-08-11. Retrieved 2017-06-26.
  8. ^ a b Transistorized memory, such as RAM, ROM, flash and cache sizes as well as file sizes are specified using binary meanings for K (10241), M (10242), G (10243), etc.
  9. ^ Binosi, Lorenzo; Barzasi, Gregorio; Carminati, Michele; Zanero, Stefano; Polino, Mario (2024). "The Illusion of Randomness: An Empirical Analysis of Address Space Layout Randomization Implementations". arXiv:2408.15107 [cs.CR].
  10. ^ "Android Security". Android Developers. Archived from the original on 2011-10-12. Retrieved 2012-07-07.
  11. ^ "oss-security". Archived from the original on 2015-10-05. Retrieved 2015-10-04.
  12. ^ "Revert "Reenable support for non-PIE executables"". Archived from the original on 2023-08-11. Retrieved 2017-06-26.
  13. ^ mmap - add mmap offset randomization Archived 2014-02-01 at the Wayback Machine, DragonFly Gitweb, 25 November 2010.
  14. ^ "Implement Address Space Layout Randomization (ASLR)". Archived from the original on 2019-05-07. Retrieved 2019-02-10.
  15. ^ "ASLR - FreeBSD Wiki". Archived from the original on 2021-05-17. Retrieved 2021-05-17.
  16. ^ "FreeBSD 13.2-RELEASE Release Notes". Archived from the original on 2023-04-11. Retrieved 2023-04-11.
  17. ^ Pwn2Own day 2: iPhone, BlackBerry beaten; Chrome, Firefox no-shows Archived 2012-05-02 at the Wayback Machine, Ars Technica, 11 March 2011
  18. ^ Stefan Esser (2013-03-07). "iOS 6 Exploitation 280 Days Later". Slide 19, "iOS 6 introduces KASLR". Archived from the original on 2019-05-07. Retrieved 2018-04-25.
  19. ^ Tarjei Mandt. "Attacking the iOS Kernel: A Look at 'evasi0n'" (PDF). Archived (PDF) from the original on 2020-12-13. Retrieved 2023-07-23.
  20. ^ Dang, Alan; Miller, Charlie (2009-03-25). "The NX Bit And ASLR". Tom's Hardware. Archived from the original on 2023-08-11. Retrieved 2010-03-20.
  21. ^ personality(2) – Linux Programmer's Manual – System Calls
  22. ^
  23. ^ "[PATCH] ASLRv3: randomize_va_space=3 preventing offset2lib attack". lore.kernel.org.
  24. ^ Miller, Justin (2024-01-08). "ASLRn't: How memory alignment broke library ASLR". zolutal's blog. Retrieved 2024-01-13.
  25. ^ "[LTP] [PATCH 2/2] Add test for ASLRn't bug - Martin Doucha". lore.kernel.org.
  26. ^ Jake Edge (2013-10-09). "Kernel address space layout randomization". LWN.net. Archived from the original on 2014-04-04. Retrieved 2014-04-02.
  27. ^ "Linux kernel 3.14, Section 1.7. Kernel address space randomization". kernelnewbies.org. 2014-03-30. Archived from the original on 2021-01-15. Retrieved 2014-04-02.
  28. ^ "kernel/git/torvalds/linux.git: x86, kaslr: Return location from decompress_kernel (Linux kernel source tree)". kernel.org. 2013-10-13. Archived from the original on 2023-08-11. Retrieved 2014-04-02.
  29. ^ KASLR is Dead: Long Live KASLR (PDF). Engineering Secure Software and Systems 2017. 2017-06-24.
  30. ^ Jang, Yeongjin; Lee, Sangho; Kim, Taesoo (2016). "Breaking Kernel Address Space Layout Randomization with Intel TSX" (PDF). Proceedings of the 2016 ACM SIGSAC Conference on Computer and Communications Security. CCS '16. New York: Association for Computing Machinery. pp. 380–392. doi:10.1145/2976749.2978321. ISBN 9781450341394. S2CID 6293725. Archived (PDF) from the original on 2020-09-21. Retrieved 2017-12-29.
  31. ^ Corbet, Jonathan (2017-12-20). "The current state of kernel page-table isolation". Linux Weekly News. Archived from the original on 2018-01-04. Retrieved 2018-01-04.
  32. ^ Corbet, Jonathan (2017-11-15). "KAISER: hiding the kernel from user space". Linux Weekly News. Archived from the original on 2020-12-08. Retrieved 2017-12-29.
  33. ^ a b Evtyushkin, Dmitry; Ponomarev, Dmitry; Abu-Ghazaleh, Nael (2016). Jump over ASLR: Attacking branch predictors to bypass ASLR (PDF). 2016 49th Annual IEEE/ACM International Symposium on Microarchitecture (MICRO). pp. 1–13. doi:10.1109/MICRO.2016.7783743. ISBN 978-1-5090-3508-3. S2CID 3801142.
  34. ^ "Linux 5.16 Has Early Preparations For Supporting FGKASLR - Phoronix". www.phoronix.com. Archived from the original on 2021-11-10. Retrieved 2021-11-10.
  35. ^ "Windows ISV Software Security Defenses". Msdn.microsoft.com. 2010-12-06. Archived from the original on 2012-04-18. Retrieved 2012-04-10.
  36. ^ Windows Internals: Including Windows Server 2008 and Windows Vista, Fifth Edition (PRO-Developer) ISBN 978-0-7356-2530-3
  37. ^ Ollie Whitehouse (February 2007). "An Analysis of Address Space Layout Randomization on Windows Vista" (PDF). Archived from the original (PDF) on 2019-07-15. Retrieved 2009-01-18.
  38. ^ "WehnTrust". Codeplex.com. Archived from the original on 2009-12-25. Retrieved 2012-04-10.
  39. ^ "Security Architects' Ozone". Security Architects. Archived from the original on 2016-03-04. Retrieved 2012-04-10.
  40. ^ "WehnTrust source code". Archived from the original on 2013-11-28. Retrieved 2013-11-15.
  41. ^ "Address-Space Randomization for Windows Systems" (PDF). Archived (PDF) from the original on 2010-08-05. Retrieved 2012-04-10.
  42. ^ Ollie (2012-03-02). "Research, Develop, Assess, Consult & Educate | Recx: A Partial Technique Against ASLR – Multiple O/Ss". Recxltd.blogspot.co.uk. Archived from the original on 2013-03-23. Retrieved 2012-04-10.
  43. ^ "Announcing NetBSD 5.0". Archived from the original on 2016-04-21. Retrieved 2016-04-25.
  44. ^ Christos Zoulas (2016). "PIE binaries and ASLR are on in the default build for amd64". Archived from the original on 2016-04-22. Retrieved 2016-04-25.
  45. ^ "Kernel ASLR on amd64". 2017. Archived from the original on 2017-10-16. Retrieved 2017-10-16.
  46. ^ a b Kurt Miller (2008). "OpenBSD's Position Independent Executable (PIE) Implementation". Archived from the original on 2011-06-12. Retrieved 2011-07-22.
  47. ^ "libc/stdlib/malloc.c". BSD Cross Reference, OpenBSD src/lib/. Archived from the original on 2014-12-26. Retrieved 2016-09-12.
  48. ^ "Mac OS X – Security – Keeps safe from viruses and malware". Apple. Archived from the original on 2011-05-25. Retrieved 2012-04-10.
  49. ^ "Security". Apple Inc. Archived from the original on 2011-06-06. Retrieved 2011-06-06.
  50. ^ "OS X Mountain Lion Core Technologies Overview" (PDF). June 2012. Archived (PDF) from the original on 2012-07-10. Retrieved 2012-07-25.
  51. ^ Controlling Access to Machine Resources Archived 2013-06-20 at the Wayback Machine, Oracle Information Library, 26 October 2012.
  52. ^ AnC Archived 2017-03-16 at the Wayback Machine VUSec, 2017
edit
Retrieved from "https://en.wiki.x.io/w/index.php?title=Address_space_layout_randomization&oldid=1255415884"