Vulnerabilities

To sort exploits

Decryption of any GEN3 PUP

• Discovered by flatz.

• A bug in the handlers of PUP decryption allows any PS4 on FW 1.62 GEN3 or below to decrypt any GEN3 PUP (retail, testkit, devkit, beta) with a version above 1.00 (post-prototype).

• SM code doesn't reset state after SMI checks failure, so to decrypt arbitrary PUP, you need to ignore mailbox error after PupDecryptHeader cmd (1).

• Fixed around 1.70

Decryption of any usermode SELF from FW 1.00 to 3.70

• Sony reused keys from FW 1.00 to 3.70 on usermode modules. As a result, any usermode module from those FWs can be decrypted on a PS4 running FW between 1.00 and 3.70.

• Fixed in 4.00 with the introduction of new keyset.

.strtab/.symtab kernel table of symbols kept on very low FWs

• Sony used to have two tables of symbols on very low versions: .strtab/.symtab and .dynstr/.dynsym (.strtab/.symtab had all symbols, .dynstr/.dynsym had ~75% of them).

• Seen in 1.01 kernel. Patched in 1.03.

.dynstr/.dynsym kernel table of symbols kept on low FWs

• After Sony removed .strtab/.symtab, they still kept the .dynstr/.dynsym one.

• Patched in 2.50

IDPS leak in sceSblAuthMgrDriveData on low retail FWs

• Discovered by flatz.

• Dump IDPS from 2 EID blocks from kernel: sceSblAuthMgrDriveData(0, in_buf, 0x160, out_buf, 0xA4, 1). Pass 0x160 bytes at 0x90C00 from sflash0s1.crypt into `in_buf` and dump `out_buf`.

• It is possible because someone from sony forgot to encrypt output and that is how it was patched later.

• Patched in 3.00 retail. Works on any TestKit/DevKit FW.

Partial SAMU KeyRings bruteforce by missing HMAC length check in secure kernel

• Discovered by flatz.

• PS4 Crypto Coprocessor (CCP) interface in secure kernel has a bug that allows to dump (or better saying, bruteforce) key rings from SAMU.

That is how AES/HMAC keys from PFS, portability keys, VTRM keys, etc can be retrieved. A crypto flaw was in ability to issue HMAC operation with key length < 16, for example, by setting it to 1 you can bruteforce key bytes one by one by comparing HMAC result with HMAC result with known partial key.

• This trick may work on other crypto hardware as well if it does not restrict key lengths. Amazingly, Intel Secure Key Storage (SKS) of CSME subsystem also has a bug allowing to brute-force any key slot, but the issue exists at hardware level - insecure design of the keys distribution to crypto engines (AES, SHA, RC4). Intel didn't recognize the bug arguing that to access SKS the CSME privileged arbitrary code execution is required, but SKS is exactly designed to protect the ROM generated keys from CSME firmware...

Related:

• https://twitter.com/qlutoo/status/1027691272369262594

• https://yifan.lu/2017/02/19/psvimgtools-decrypt-vita-backups/

• https://www.lolhax.org/2019/01/02/extracting-keys-f00d-crumbs-raccoon-exploit/

• Exploit PoC for PS4 FW 7.55

• Patched since a FW between 7.55 (unpatched) and 9.00 (patched).

Crashdumps encryption using symmetrical key and same key across FW

• see FoF article

• The keys never changed between 1.01 and 3.15 FWs. Then between 3.50 and 4.07 FWs they changed the keys many times but still used symmetrical key.

• Patched on FW 4.50 by using asymmetrical key. Tested between 1.01 and 4.07 FWs.

Hardware Exploits

PCIe man-in-the-middle attack

• First done on 1.01 by failoverflow on PS4 launch !
• Detailed at 33c3: 33c3 slides by Marcan
• Permits kernel and usermode dumping

Syscon glitching

It is possible to glitch the Syscon debug interface to allow access and dump keys. It was originally done by an anonymous member of fail0verflow.

Aeolia and Belize (Southbridge) SCA/DPA

Side Channel Analysis (SCA) with Differential Power Analysis (DPA) on Aeolia and Belize (PS4 Southbridge revisions) has been shown to be able to recover key material. Since Sony never used private/public key pairs, it is possible to exploit this and gain complete control over the Southbridge. You can attack the main FreeBSD kernel from here.

Nearly same methods are working on recent PS4 Pro motherboard NVB-003 that has Belize Southbridge (CXD90046GG).

Contrarly to Aeolia, Belize has ROM readout protection and clears stack which makes it more secure.

Old notes:

This is a hack to gain unsigned code execution on the Southbridge for all motherboard/console revisions. You might be able to glitch the EMC bootrom in order to bypass further signature checks and break the chain of trust. This hack might involve slowing down the Syscon clock. Timing the glitch based on SPI read accesses then either doing a power glitch or clock glitch to skip signature check. If the glitch fails, then we simply reset. This can be done with a very cheap CPLD/FPGA. Most Xbox 360 glitching modchips used a Xilinx Coolrunner because it is cheap and easy to use (board can cost as low as $5).

Related:

• fail0verflow's writeup
• fail0verflow's tweet
• Playstation 4 Rest Mode DEMO REcon Brussels 2018 by Volodymyr Pikhur
• Slides of REcon Brussels 2018 by Volodymyr Pikhur
• jogolden's writeup

Usermode Exploits (Game Savedata)

PS1 games savedata exploits

See PS1 savedata exploits on PS1 Dev Wiki.

Official PS Classic games (warning: some may be remastered, to check) on PS4/PS5 available on PS Store:

• Ape Escape - First-time NA re-release on PS4/PS5
• Hot Shots Golf - First-time NA re-release on PS4/PS5
• I.Q. Intelligence Qube - First-time NA re-release on PS4/PS5
• Jumping Flash! - Free PS4/PS5 version for PS3 PSone Classics owners
• Mr. Driller - PlayStation Plus Premium subscription only (?)
• Oddworld: Abe’s Oddysee - Free PS4/PS5 version for PS3 PSone Classics owners
• Resident Evil: Director’s Cut - PlayStation Plus Premium subscription only
• Syphon Filter - Free PS4/PS5 version for PS3 PSone Classics owners
• Tekken 2 - PlayStation Plus Premium subscription only
• Toy Story 2: Buzz Lightyear To The Rescue! - Free PS4/PS5 version for PS3 PSone Classics owners
• Wild Arms - Free PS4/PS5 version for PS3 PSone Classics owners
• Worms World Party - First-time NA re-release on PS4/PS5
• Worms Armageddon - First-time NA re-release on PS4/PS5

• https://www.playstation.com/en-us/editorial/iconic-must-play-titles-on-playstation-plus-classics-catalog/
• https://www.reddit.com/r/PS3/comments/1cscsb2/all_ps1pspps3_games_that_have_crossbuy_or_free/
• https://www.reddit.com/r/PlayStationPlus/comments/vfg39t/is_there_a_clear_list_of_which_ps1_classic_ps4ps5/

"I have bought some of them on the PS3/Vita and the ones I could claim on the PS4/PS5 were Tekken 2 (which previously was not redeemable), all Syphon Filter games, all Wild Arms games, Legend of Dragoon, Ridge Racer Type 4 and Jumping Flash. Resident Evil Director's Cut is NOT redeemable. The rule of thumb is: if you can buy it on PS4/PS5 - and not only claim it through plus premium/deluxe sub, like Resident Evil 1 - they are redeemable from a previous purchase on PS3/PSP/Vita."

PS2 games savedata exploits

See PS2 savedata exploits on PS2 Dev Wiki.

Official PS2onPS4 games sold on the PS Store (as of September, 2024):

• ADK DAMASHII™ UP0576-CUSA03783_00-SLPS259060000001 https://image.api.playstation.com/cdn/UP0576/CUSA03783_00/BpMMUC8q1MRAsL9iWDh6vbW844hq3JXK.png
• Arc the Lad: Twilight of the Spirits
• Art of Fighting Anthology
• Ape Escape 2
• Bully (Canis Canem Edit)
• Dark Chronicle (Dark Cloud 2) UP9000-CUSA02037_00-SCUS972130000001 https://image.api.playstation.com/cdn/UP9000/CUSA02037_00/hIKSKqBMerypNW49TCECATZSBBUcSBph.png
• Dark Cloud
• Destroy All Humans!
• Destroy All Humans! 2
• Eternal Ring UP1022-CUSA04654_00-SLUS200150000001 https://image.api.playstation.com/cdn/UP1022/CUSA04654_00/DRIS0z7mtNMYZPchoqLnKlhJqyNvM8mZ.png
• FantaVision
• Fatal Fury Battle Archives Vol. 2
• Forbidden Siren
• Fu'un Super Combo UP0576-CUSA03784_00-SLPS257810000001 https://image.api.playstation.com/cdn/UP0576/CUSA03784_00/QWsetumZLYupFHsOIkoGbKYpySGBdtlp.png
• Ghosthunter (English, Japanese) UP9000-PPSA21974_00-SLUS209930000000 https://image.api.playstation.com/vulcan/ap/rnd/202406/0519/64b26d812ffbfec1acffeaa7d3a61effb12400fff2d95935.png
• GTA III
• GTA Vice City
• GTA San Andreas
• Harvest Moon: Save the Homeland
• Harvest Moon: A Wonderful Life Special Edition
• Hot Shots® Tennis (Everybody's Tennis) UP9000-CUSA02193_00-SCUS976100000001 https://image.api.playstation.com/cdn/UP9000/CUSA02193_00/FrJXexHruy7pjB6bCgDidXRbakNfNJJc.png
• Indigo Prophecy
• Jak and Daxter: The Precursor Legacy
• Jak II: Renegade
• Jak 3
• Jak X: Combat Racing
• Kinetica UP9000-CUSA01725_00-SCUS971320000001 https://image.api.playstation.com/cdn/UP9000/CUSA01725_00/EKH34FKOEt3dTXLCiccuawdS8iGIqGLF.png
• Manhunt
• Max Payne
• Metal Slug Anthology
• Okage: Shadow King UP9000-CUSA02199_00-SCUS971290000001, requires PS4 FW version 3.15, although it was compiled with PS4 SDK version 3.008.000, latest patch requires PS4 FW 4.05
• PaRappa the Rapper 2
• Primal
• Psychonauts
• Puzzle Quest: Challenge of the Warlords
• Red Dead Revolver
• Red Faction
• Red Faction II
• Resident Evil Code: Veronica X
• Rise of the Kasai
• Rogue Galaxy
• Samurai Shodown VI
• Sly Raccoon (2002), Sly Cooper and the Thievius Racoonus UP9000-CUSA47431_00-SCUS971980000000 requires PS4 FW ?11.00? (update requires PS4 11.508.000)
• Star Ocean Till The End Of Time
• Star Wars Bounty Hunter
• Star Wars Racer Revenge
• STAR WARS: Jedi Starfighter UP1082-CUSA03473_00-SLUS202930000001 https://image.api.playstation.com/cdn/UP1082/CUSA03473_00/PGRyqtcRKUoAsP4bJAhcoziTwL8940k1.png

EP1006-CUSA03494_00-SLES503710000001 https://image.api.playstation.com/cdn/EP1006/CUSA03494_00/9MsXVY5UULzSHB5BTreuKhwep3KZwvQP.png

• STAR WARS The Clone Wars UP1082-PPSA21985_00-SLUS205100000000 https://image.api.playstation.com/vulcan/ap/rnd/202404/2320/798b83df229613be009a6a9a191606a04846b32eab781c14.png
• The King of Fighters Collection: The Orochi Saga
• The King of Fighters '98 Ultimate Match
• The King of Fighters 2000
• The Mark of Kri
• The Warriors
• Tomb Raider: Legend UP8489-PPSA22453_00-SLUS212030000000 https://image.api.playstation.com/vulcan/ap/rnd/202405/0816/1d9bea712b88097f61b829fac5e96f956fb67225be456f36.png
• Twisted Metal: Black
• War of the Monsters
• Wild Arms 3

• See [1].

Official PS2onPS4 games sold on Bluray Discs:

• ADK DAMASHII™ UP0576-CUSA03783_00-SLPS259060000001 https://image.api.playstation.com/cdn/UP0576/CUSA03783_00/BpMMUC8q1MRAsL9iWDh6vbW844hq3JXK.png
• Art of Fighting Anthology (by Limited Run #375) UP0576-CUSA03754_00-SLUS214870000001 https://image.api.playstation.com/cdn/UP0576/CUSA03754_00/Hf5lUn48Ds3UDNp8NNjdzv7f1BZWGaai.png
• Destroy All Humans! (2005) (PS2 Classic by Limited Run #370, not to be confused with the remake EP4389-CUSA14910_00-DAH1REMAKEEU0000) UP4389-CUSA05232_00-SLUS209450000001 https://image.api.playstation.com/cdn/UP4389/CUSA05232_00/XrgVkqoR5rvZk4tAGi2j7OFfHpAZWKUu.png
• Fatal Fury Battle Archives Vol. 2 (by Limited Run #371) UP0576-CUSA03750_00-SLUS217230000001 https://image.api.playstation.com/cdn/UP0576/CUSA03750_00/gFCLAhlGZwvFkra1p2sozwIZ5SH1OyZO.png
• Fu'un Super Combo UP0576-CUSA03784_00-SLPS257810000001 https://image.api.playstation.com/cdn/UP0576/CUSA03784_00/QWsetumZLYupFHsOIkoGbKYpySGBdtlp.png
• Indigo Prophecy™ (aka Fahrenheit 2005, by Limited Run #331) UP1642-CUSA04798_00-SLUS211960000001 https://image.api.playstation.com/cdn/UP1642/CUSA04798_00/WJFDq83f1tcZ0E2PkEa1rXOba8laaZUV.png
• Jak and Daxter: The Precursor Legacy™ UP9000-CUSA02522_00-SCUS971240000001 https://image.api.playstation.com/cdn/UP9000/CUSA02522_00/o9zJoXqpd4lzarjIbvvZLFjYGLsLvqCp.png
• Jak X Combat Racing™® UP9000-CUSA07842
• Jak II UP9000-CUSA07840
• Jak 3 UP9000-CUSA07841
• METAL SLUG ANTHOLOGY™ (US version by Limited Run #364) UP0576-CUSA03749_00-SLUS215500000001 https://image.api.playstation.com/cdn/UP0576/CUSA03749_00/ImHDRENlttkdiXlm3K8ejNVgLURd3uTw.png
• METAL SLUG ANTHOLOGY™ (EU version by SNK) EP0576-CUSA04156_00-SLES546770000001 https://image.api.playstation.com/cdn/EP0576/CUSA04156_00/NN7npbsEvxIRGI8lBVhm9I5BwFzdGlOK.png
• Psychonauts UP2154-CUSA03881
• Red Faction (by Limited Run #281) UP4389-CUSA06402_00-SLUS200730000001 https://image.api.playstation.com/cdn/UP4389/CUSA06402_00/T07Bf136claKzP3SHF30QLa2xMAFjSpP.png
• Samurai Shodown VI (by Limited Run #329) UP0576-CUSA03787_00-SLUS216290000001 or EP0576-CUSA04158_00-SLES552920000001 https://image.api.playstation.com/cdn/UP0576/CUSA03787_00/CuLRRdOYvdge0IW9LL9Vewj44RCc6OAU.png https://image.api.playstation.com/cdn/EP0576/CUSA04158_00/7SrtqugKMJixAcbprEE0ExGUOHlhL0F7.png
• STAR WARS™ BOUNTY HUNTER™ (US version) UP1082-CUSA03472_00-SLUS204200000001
• STAR WARS™ BOUNTY HUNTER™ (EU version) EP1006-CUSA03493_00-SLES508310000001
• Star Wars Racer Revenge UP1082-CUSA03474, requires PS4 FW version ?3.15, although it was compiled with SDK version 3.008.000?
• The King of Fighters '98 Ultimate Match (by Limited Run #344) UP0576-CUSA03751_00-SLUS218160000001 https://image.api.playstation.com/cdn/UP0576/CUSA03751_00/bp4LfKIjcVTMfKP3O4LrDJHWzY6vZDar.png
• The King of Fighters 2000 (by Limited Run #386) UP0576-CUSA03748_00-SLUS208340000001 https://image.api.playstation.com/cdn/UP0576/CUSA03748_00/tvXJmFqa9zkXAAKCij20B3spadkqGuka.png
• The King of Fighters™ Collection: the Orochi Saga (by Limited Run #393) UP0576-CUSA03753_00-SLUS215540000001 https://image.api.playstation.com/cdn/UP0576/CUSA03753_00/E3gFtUUjCu2WDBSIGeXMV40sfF4uHzZi.png

These PS2onPS4 games can be bought online directly via Limited Run Games for brand new or for example on Ebay for second hand or like new.

PSP games savedata exploits

See PSP savedata exploits on PSP Dev Wiki.

• https://wololo.net/2012/09/01/when-the-psp-and-the-vita-show-their-battle-scars/
• https://wololo.net/talk/viewtopic.php?f=52&t=11183&start=10#p143779
• https://www.playstation.com/en-us/editorial/iconic-must-play-titles-on-playstation-plus-classics-catalog/

PS4/PS5 PS2emu sandbox escape (mast1c0re)

Advantages of the PS4/PS5 PS2emu sandbox escape exploit over most WebKit exploits:

• Bigger kernel attack surface (more usermode privileges) versus WebKit very restricted and becoming more and more with firmware revisions. For example, the PS2emu process uses libkernel_sys, which supports nmount and so mounting of system partitions, whilst neither libkernel_web nor regular libkernel do.
• 100% reliable versus WebKit exploits becoming less and less stable with firmware revisions
• Firmware agnostic (ROP-less code execution) versus almost one WebKit revision every three firmware update

Credits

• CTurt for discovering these vulnerabilities in September 2021.
• CTurt for public disclosure on twitter (2022-09-14).
• flatz, balika011, theflow0, chicken(s), PlayStation for helping CTurt.
• McCaulay for sharing publicly his implementation in February 2023.

Analysis

• Writeup part 1 by CTurt (2022-09-14)
• Writeup part 3 by McCaulay (2023-02-17)
• Writeup part 2 by CTurt (2023-04-02)

Bug Description

After getting code execution in a PS2onPS4 game using a savedata exploit, it is possible to exploit the PS2 emulator to get x86-64 usermode ROP execution. It is then possible, without a kernel exploit, to load another PS2 game in the emulator with a compatibility rate based on the PS2 emulator configuration.

Exploit Implementation

• mast1c0re implementation by McCaulay (2023-02-18)

Patched

No as of PS4 FW 11.50 and PS5 FW 8.00. Using the PS2onPS4 game Okage Shadow King, the exploit should work starting from PS4 FW 3.15 and PS5 FW 1.00.

PS4/PS5 game savedata LUA exploit

• Used by Flatz on 2023-07-27 in his Hypervisor exploit.

• Used by Flatz on 2024-09-14 in his implementation of the umtx UaF kernel exploit.

• Some PS4 (or maybe PS5) games, in disc version (probably also available in PS Store version but potentially patched), can be exploited as they use some LUA interpreter, by crafting an evil save data.

• Possible vulnerable games: Pay Day 2, Mafia III, God of War (which one?).

• Not patched as of PS4 FW ?12.00? and PS5 FW 7.61.

Usermode Exploits (BD-J)

Advantages of most BD-J exploits over most WebKit exploits:

• Bigger kernel attack surface (more usermode privileges) versus WebKit very restricted and becoming more and more with firmware revisions. For example, the BD-J process uses libkernel_sys, which supports nmount and so mount of system partitions, whilst neither libkernel_web nor regular libkernel do.
• 100% reliable versus WebKit exploits becoming less and less stable with firmware revisions
• Firmware agnostic (ROP-less code execution) versus almost one WebKit revision every three firmware update
• JIT enabled allowing to write a kernel exploit in C versus writing in assembly and JavaScript since around FW 2.00

FW <= 10.71 - BD-JB2 - Path traversal sandbox escape by TheFloW

Credits

• TheFloW for the exploits finding (before 2023-09-11), ethical disclose to SCE (2023-09-22) and public disclosure (2023-10-25)
• Previous BD-JB contributors

Analysis

• Removed tweet of BD-JB2 logs on a 7.61 PS5 by TheFloW (2023-09-11)

Bug Description

Basing on BD-JB1 exploit files, in /bdmv/bdjo.xml changing bdjo/applicationManagementTable/baseDirectory to a path of the form `file:///app0/cdc/lib/../../../disc/BDMV/JAR/00000.jar` allows loading a JAR Java executable file.

Exploit Implementation

• PoC by TheFloW (2023-10-25)

Patched

No as of PS4 FW 10.71 (maybe patched on PS4 FW 11.00). Yes on PS5 FW 8.00.

FW <= 9.00 - BD-JB - Five vulnerabilities chained by TheFloW

Credits

• CTurt for FreeDVDBoot exploit on PS2 and the idea to hack BD-J on PS3 and PS4 on twitter (2020-06-27)
• TheFloW for finding these vulnerabilities (around 2021-10-24) and disclosing them publicly on hackerone and hardwear.io (2022-06-10)
• Sleirsgoevy for writing the first public implementation (2022-06-16)
• psxdev, sleirsgoevy and John Törnblom for the public implementations

Analysis

• TheFloW's PS5 kernel exploit announcement (2021-11-07)
• Official vulnerability report by TheFloW (2022-06-10)
• Slides presented at hardwear.io by TheFloW (2022-06-10)

Bug Description

TO ADD DESCRIPTION OF EACH ONE OF THE 5 BUGS:

• #1 com.sony.gemstack.org.dvb.user.UserPreferenceManagerImpl userprefs hijack leading to classes instantiation under privileged context (affecting ?PS3?, PS4, PS5)
• #2 com.oracle.security.Service leading to privileged constructor call (affecting ?PS3?, PS4, not PS5)
• #3 com.sony.gemstack.org.dvb.io.ixc.IxcProxy leading to privileged method call (affecting ?PS3?, PS4, PS5)
• #4 JIT compiler hack leading to usermode arbitrary RW and arbitrary usermode code execution (affecting ?PS3?, PS4, not PS5)
• #5 UDF buffer overflow kernel exploit (affecting ?PS3?, PS4, PS5)

This exploit chain alone does not allow one to run pirated games on PS4 or PS5 as there is not enough RAM allowed in the BD-J process and there are other constraints.

Exploit Implementation

• Implementation of BD-J usermode code execution on PS4 using bugs #1, #2, #3 and #4 by TheFloW (2021-10-24)
• Vuln #1 com.sony.gemstack.org.dvb.user.UserPreferenceManagerImpl implementation by TheFloW
• Vuln #2 com.oracle.security.Service and #3 com.sony.gemstack.org.dvb.io.ixc.IxcProxy chained together by TheFloW
• Vuln #4 JIT compiler hack implementation by TheFloW
• Implementation of BD-J usermode code execution on PS4 using bugs #2, #3 and #4 by sleirsgoevy (2022-06-16)
• Note that no implementation of the UDF kernel exploit has ever been done even by TheFloW, only a kernel panic PoC.

Patched

No as of PS4 FW 9.00 and PS5 FW 4.03. At least partially patched on PS4 FW 9.50 and PS5 FW 5.00.

On PS4 FW 9.03 and PS5 FW ?4.50?, the bug #5 (UDF) has been patched.

Usermode Exploits (WebKit)

WebKit sources

WebKit sources Currently archived up to version 10.01. Useful for developers that can't access PlayStation URLs and also for when Sony inevitably stops hosting the sources in the future.

FW ?10.00?-11.52 - Unknown heap and string overflow (no CVE) leading to crash

Credits

• Debty for PoC public disclose (2024-08-29)

Analysis

• PoC and analysis by Debty (2024-08-29)

Bug Description

• TODO

Implementation description by Debty:
String exploit is not actually an exploit but just a memory exhauster. It is not actually viable so instead there is a feature called "latest iteration".

Exploit Implementation

• PoC by Debty (2024-08-29)

Patched

Yes on PS4 FW 12.00 and PS5 FW 10.00.

Tested working on PS4 FWs 10.00-11.52 and PS5 FWs 6.00-9.60.

FW 10.00-11.02 - JSC DFG Abstract Intepreter clobberWorld Type Confusion (no CVE) leading to arbitrary RW

Credits

• ENKI for public disclose and analysis (2024-06-03)

Analysis

• Analysis by ENKI (2024-06-03)
• Patch commit #1 (2023-07-31)
• Patch commit #2 (2023-05-01)
• Inverting Your Assumptions: A Guide to JIT Comparisons by Jasiel Spelman (2018-04-12)

Bug Description

• TODO

Exploit Implementation

• PoC by ENKI (2024-06-03)

Patched

Yes on PS4 FW 11.50 and PS5 FW 9.00.

Tested working on PS4 FWs 10.00-11.02 and PS5 FWs 6.00-8.60. PS4 FWs <= ?9.60? and PS5 FWs <= ?5.50? are invulnerable.

FW 6.00-9.60 - FrameLoader::loadInSameDocument() UaF (CVE-2022-22620) leading to arbitrary RW

Credits

• Sergei Glazunov, Google Project Zero, for reporting the bug in 2013-01 and answering Maddie Stone's questions in 2022 (2013)
• Maddie Stone, Google Project Zero, for sharing a write-up describing this vulnerability (2022-06-14)
• Abc (anonymous person) for making an OOM PoC for webkit-gtk, PS4 and PS5 (2023-10-03) then making an arbitrary RW PoC (PSFree) for webkit-gtk, PS4 6.00-9.60 and PS5 1.00-5.50 (2023-10-24)
• CelesteBlue for testing and porting abc' PSFree to PS4 6.00-9.60 and PS5 1.00-5.50 (2023-11-04)

Analysis

• WebKit bug-reintroducing commit by Darin Adler reviewed by Alex Christensen (2016-12-31)
• WebKit fix talk by Yusuke Suzuki reviewed by Mark Lam (2022-01-24)
• WebKit fix commit by Yusuke Suzuki reviewed by Mark Lam (2022-01-25)
• Short writeup by Maddie Stone (2022-06-14)
• Detailed writeup by Maddie Stone (2022-06-14)

Bug Description

The History API allows access to (and modification of) a stack of the pages visited in the current frame, and these page states are stored as a SerializedScriptValue. The History API exposes a getter for state, and a method replaceState() which allows overwriting the "most recent" history entry.

The bug is that FrameLoader::loadInSameDocument() takes the state as an argument (stateObject), but does not increase its reference count. Only a HistoryItem object holds a reference to the stateObject. loadInSameDocument() can trigger a callback into user JavaScript through the onblur event. The user's callback can call replaceState() to replace the HistoryItem's state with a new object, therefore dropping the only reference to the stateObject. When the callback returns, loadInSameDocument() will still use this free'd object in its call to statePopped(), leading to the use-after-free.

When loadInSameDocument() is called it changes the focus to the element its scrolling to. If we set the focus on a different element prior to loadInSameDocument()'s execution, the blur event will be fired on that element. Then we can free the stateObject by calling replaceState() in the onblur event handler.

The bug is triggered by history.back() with the target state whose URL contains a hash. Here's a Proof-of-Concept that will crash:

input = document.body.appendChild(document.createElement('input'));

foo = document.body.appendChild(document.createElement('a'));
foo.id = 'foo';

function pop(event) {
    alert('you get a crash after you close this alert');
    event.state; // use the freed SerializedScriptValue
    alert('WebKit version not vulnerable');
}

addEventListener('popstate', pop);

history.pushState('state1', '', location + '#foo'); // URL with a hash
history.pushState('state2', '');

setTimeout(() => {
    input.focus();
    input.onblur = () => {
        history.replaceState('state3', '')
    };
    setTimeout(() => {
        history.back(); // trigger loadInSameDocument()
    }, 1000);
}, 1000);

The user may then trigger a double free and escalate it into an arbitrary read primitive via spraying WTF::StringImpls like in the buildBubbleTree() UaF exploit. The read primitive is used to create the addrof() primitive and is used to save addresses of buffers that will be used to modify a SerializedScriptValue. After freeing the StringImpl (triple free), SerializedScriptValues are sprayed via the postMessage() JavaScript function until one is allocated using the previously freed memory.

The method used to modify the fields of the StringImpl for arbitrary reads can be used can also be used to modify the SerializedScriptValue. Appropriate fields can modified to have deserialization create a JSC::JSArrayBufferView whose m_vector field will point to another JSArrayBufferView, which will be called the worker. The user can modify the worker's fields for arbitrary read/write. Deserialization is done via msg.data where msg is the MessageEvent from postMessage().

A way to know if the system is vulnerable is the appearance of the input HTML element in the PoC page. If the HTML input field stays focused (blue outline) after the second timeout, then the vulnerability is not present. Note that Maddie Stone's PoC will never trigger any sort of crash on release builds as it was meant for builds with memory sanitation that can detect UaFs.

Exploit Implementation

• Simple PoC for ASAN webkit-gtk by Maddie Stone in Maddie Stone's writeups
• Information leak PoC for webkit-gtk by springsec
• OOM PoC for PS4 and PS5 by abc on ps4-dev discord (to mirror)
• Arbitrary RW PoC (PSFree) for PS4 6.00-9.60 and PS5 1.00-5.50 by abc on ps4-dev discord (to mirror)

Patched

Yes on PS4 FW 10.00 and PS5 FW 6.00.

The patch changes the stateObject argument to loadInSameDocument from a raw pointer, SerializedScriptValue*, to a reference-counted pointer, RefPtr<SerializedScriptValue>, so that loadInSameDocument now increments the reference count on the object.

Tested working on PS4 FWs 6.00-9.60 and PS5 FWs 1.00-5.50. PS4 FWs <= 5.56 are invulnerable as the HTML input field stays focused (blue outline) after second timeout whilst it should not if the console were exploitable.

FW 9.00-9.04 - WebCore::CSSFontFaceSet vulnerabilities leading to arbitrary RW

There are many FontFaceSet vulnerabilities. Explore [2].

Credits

• Myles C. Maxfield (litherum), Apple, for adding the vulnerability in WebKit (2016-02-22) then fixing and so disclosing the vulnerability (2021-08-26)
• Maddie Stone, Google Project Zero, for sharing a write-up describing this vulnerability (2021-10-13)
• PS Test discord server community for testing PoCs of many WebKit vulnerabilities on their PS4s (2021-10-13)
• sleirsgoevy for making the first exploit PoC for Safari (2021-10-24) and the first exploit PoC for PS4 FW 9.00-9.04 and PS5 FW 3.00-4.50 (2021-10-27)

Analysis

• WebKit bug-introducing commit by Myles C. Maxfield - r256659 (2016-02-22)
• WebKit fix commit by Myles C. Maxfield - r281648 (2021-08-26)
• WebKit commit adding FontFace tests (2021-09-01)
• WebKit Bugzilla - Bug 229848 - FontFaceSet.has() needs to react to style changes by Myles C. Maxfield (2021-09-02)
• web-platform-tests - Add a test for FontFaceSet.has() #30322 (2021-09-03)
• Mozilla Bugzilla - New wpt failures in /css/css-font-loading/fontfaceset-has.html (with TypeError `fonts.keys() is not iterable`, due to FontFaceSetIterator not behaving like an Iterable) (2021-09-03)
• Mozilla Bugzilla - FontFaceSet's iterators skip items when previous items are removed by Myles C. Maxfield (2021-09-09)
• WebKit Bugzilla - Bug 230119 - FontFaceSet's iterators skip items when previous items are removed by Myles C. Maxfield (2021-09-09)
• WebKit fix commit by Myles C. Maxfield merged by Russell Epstein (2021-09-09)
• Part of vulnerable code
• (archive) Write-up and PoC by Maddie Stone (2021-10-13). Maddie Stone's vulnerability is not CVE-2021-30858 but was guessed to be by Maddie Stone. See write-up edit commit. Warning: Maddie Stone's vulnerability was wrongly classified as a use-after-free by Maddie Stone according to sleirsgoevy.
• Vulnerability description by Wololo (2021-10-14)

Bug Description

Description in WebKit fix commit by Myles C. Maxfield:

After r256659, asking for a failed CSSFontFace's families() returns nullopt. It's possible to add a failed font to a CSSFontFaceSet (of course). When we do that, we recognize the font is failed and do not update our internal data structures, because there's no need to - we cannot do anything useful with a failed font. If you _then_ try to remove the font from the CSSFontFace, we do not call families(), but instead just pull out the raw m_families member, and look in our internal data structures for it, but we do not find it, because it was never added.

Description in Maddie Stone's write-up:

The vulnerability is a use-after-free due to an unchecked end() iterator. There was an assert statement: ASSERT(iterator != m_facesLookupTable.end());, but ASSERTs do not do anything in release builds. Therefore, even if iterator == m_facesLookupTable.end() in the release build, nothing would happen and iterator would still be used. In FontFaceSet a FontFace is not added to the faces lookup table in addToFacesLookupTable if the font has already been deemed to be invalid. However, removeFromFacesLookupTable would still attempt to remove the font, leading to the use-after-free. The patch changes the ASSERT to an if clause. The function will return if iterator == m_facesLookupTable.end(), since the item it wishes to remove is not found in the table.

Description by sleirsgoevy:

On PS4 FWs 9.00-9.04 the constructor returns with an exception, but to C++ code that ignores it. That is how an invalid font is created in the first place. On earlier PS4 FWs the exception is propagated to JavaScript.

Exploit Implementation

• (archive) First exploit PoC for Safari by sleirsgoevy (2021-10-24)
• First exploit PoC for PS4 FW 9.00-9.04 and PS5 FW 3.00-4.50 by sleirsgoevy (2021-10-27)
• Implementation for PS4 FW 9.00 with exFAT kernel exploit in pOOBs4 by ChendoChap (2022-01-17)

Patched

Yes on PS4 FW 9.50 and No as of PS5 FW 4.50.

Might have been introduced in PS4 FW 3.50 and before PS5 FW 1.00 according to dates (need to check). However the vulnerability cannot be exploited in some conditions depending on how WebKit was compiled. For example, on PS4 FWs 7.55-8.53 and PS5 FWs <= 2.00, the FontFaceSet constructor returns with an exception that is propagated to JavaScript, preventing exploitation this way.

Tested working on PS4 FWs 9.00-9.04 and PS5 FWs 3.00-4.50. Untested: PS5 FWs 2.10-2.50, 4.51.

FW 6.00-7.55 - WebCore::ValidationMessage::buildBubbleTree() UaF leading to arbitrary RW

Credits

• Quentin Meffre (@0xdagger) and Mehdi Talbi (@abu_y0ussef) who are Security Researcher at Synacktiv for fuzzing WebKit, finding a way to exploit the vulnerability on PS4, presenting it on Black Hat Europe 2020 ([3]) and sharing the code (2020-12-10)
• sleirsgoevy for porting (although with low success rate) to PS4 FWs 7.00-7.02

Analysis

• WebKit bad fix commit (2019-05-28)
• WebKit good fix commit (2020-09-11)
• Write-up by Quentin Meffre (@0xdagger) and Mehdi Talbi (@abu_y0ussef) (2020-12-10)
• Presentation slides by by Quentin Meffre (@0xdagger) and Mehdi Talbi (@abu_y0ussef) (2020-12-10)

Bug Description

• The method buildBubbleTree makes a call to update the layout during which all user registered JS handlers are executed. If the ValidationMessage is destroyed in a JS callback, this could lead to a Use-After-Free situation when we get back to buildBubbleTree code.

• ValidationMessage::buildBubbleTree is doing layout which can run a script detaching the owner form element, and this ValidationMessage object can be destroyed.

After private disclose by Synacktiv ethical hackers, the vulnerability was fixed in WebKit on September 11st 2020. SIE updated to the patched WebKit with firmware 8.00 released on October 14st 2020.

Exploit Implementation

• Initial 6.xx implementation by Quentin Meffre (@0xdagger) and Mehdi Talbi (@abu_y0ussef) (2020-11-12
• 7.02 implementation with code execution by sleirsgoevy (2020-12-15)
• 6.72-7.02 Kernel exploit using this WebKit exploit by sleirsgoevy (2020-12-16)
• 7.00-7.02 Kernel exploit using this WebKit exploit by ChendoChap (2020-12-16)

Patched

Yes in 8.00 FW. Tested working on FWs 6.00-7.55, not working on FWs <= 5.56. HTML textarea guessed addresses for FWs 6.70-7.55 are known but not for FWs 6.00-6.51 so an attacker needs to make tests to determine these addresses on FWs 6.00-6.51.

FW 6.00-6.72 - bad_hoist Type Confusion exploit (CVE-2018-4386) leading to arbirary RW

Credits

• Lokihardt (from Google Project Zer0) for the exploit PoC (Sep 13, 2018)
• Fire30 for turning the vulnerability into exploit for PS4 (Dec 30, 2019)
• sleirsgoevy for attempting to stabilize the PS4 exploit with a new implementation (Feb 23, 2020)

Analysis

• https://bugs.chromium.org/p/project-zero/issues/detail?id=1665
• https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2018-4386
• https://packetstormsecurity.com/files/155871/Sony-Playstation-4-Webkit-Code-Execution.html
• https://twitter.com/Fire30_/status/1211775229116211200

Bug Description

WebKit: JSC: BytecodeGenerator::hoistSloppyModeFunctionIfNecessary does not invalidate the ForInContext object.

It is possible to craft Javascript in such a way that allows for an object to be passed as the property variable directly as a string to the op_get_direct_pname handler without being properly validated.

This is a Type Confusion exploit.

Exploit Implementation

• Initial implementation by Fire30
• Implementation with code execution by sleirsgoevy

Patched

Yes in 7.00 FW

---

FW 4.50-6.72 - DOMWindow::open heap UaF (CVE-2021-30849) leading to crash

Credits

• Sergei Glazunov (from Google Project Zer0) for the exploit PoC (Jul 1, 2021)

Analysis

• https://bugs.chromium.org/p/project-zero/issues/detail?id=2204

Patched

Yes in 7.00 FW. Vulnerable on PS4 FWs 4.50-6.72. Not vulnerable on FWs <= 4.07. Not vulnerable on FWs >=7.00 according to manual tests but need to check WebKit sources.

---

FW 4.50-6.20 - JSArray::shiftCountWithArrayStorage() OOB RW (CVE-2018-4441) leading to arbitrary RW

Credits

• Lokihardt (from Google Project Zer0) for the exploit PoC (Oct 3, 2018)
• Specter for the rewriting for PS4 (Mar 8, 2019)
• St4rk for helping Specter

Analysis

• Bug report by Lokihardt (Oct 3, 2018)
• WebKit fix commit (Oct 15, 2018)
• Announce of incoming write-up by rkmylo and buherator/stratan/@5tratan, Meligra Team (Feb 25, 2019)
• Write-up mirrored by Nytro (Feb 27, 2019)

Bug Description

We would take the fast path for JSArray::shiftCountWithArrayStorage when the array hasHoles(). However, the code for this was wrong. It would incorrectly update ArrayStorage::m_numValuesInVector.

Exploit Implementation

• PS4 6.20 WebKit exploit by Specter

Patched

Yes in 6.50 FW. It does not work on <= 4.07 FW PS4 according to tests as the exploit fails at step "Triggering memory corruption".

---

FW 6.00-6.20 - JSC::arrayProtoPrivateFuncConcatMemcpy() Information Leak (CVE-2018-4358) ?leading to ASLR defeat?

Credits

• bkth, niklasb and saelo (from phoenhex Team) for the exploit PoC in Safari (Sep 26, 2018)
• Vultra for discovering that the exploit worked on PS4 FW 6.00 (Dec 10, 2018)

Analysis

• Related: https://bugs.chromium.org/p/project-zero/issues/detail?id=1032
• Related: https://doar-e.github.io/blog/2018/07/14/cve-2017-2446-or-jscjsglobalobjectishavingabadtime/
• https://github.com/WebKit/webkit/commit/b68b373dcbfbc68682ceeca8292c5c0051472071
• https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2018-4358
• https://phoenhex.re/2018-09-26/safari-array-concat

Exploit Implementation

• PoC for iOS
• PoC for PS4

Patched

Yes in 6.50 FW

Tested

Works on 6.00-6.20. Not working on PS4 FWs <= 5.56 because JSC (JavaScriptCore) was too old.

---

FW 4.50-5.56 - JSGlobalObject::haveABadTime() Type Confusion (CVE-2017-7005) leading to arbitrary RW

Credits

• Lokihardt (from Google Project Zer0) for the exploit PoC (Mar 20, 2017)
• ALEXZZZ9 for the first PS4 implementation (on 5.01), and at same time for burning the exploit (Feb 20, 2018)
• qwertyoruiop for rewriting and porting to 5.05 and 5.50

Analysis

Project Zer0 Bug Description

Bug Description

When JSGlobalObject::haveABadTime() is called with arrays of a different JSGlobalObject type, type confusion can occur, leading to memory corruption.

Exploit Implementation

• [4]
• PS4 5.50 (WebKit Only)
• [5]

Patched

Yes in 6.00 FW

---

FW ?.??-4.05-5.56 - Document::adoptNode() UaF (CVE-2017-2468) leading to crash

Credits

• Lokihardt (from Google Project Zer0) for the exploit PoC (Jan 23, 2017)
• CelesteBlue for testing on PS4 and PS Vita (May 9, 2020)

Analysis

• exploit report by Lokihardt (Jan 23, 2017)
• Mitre report
• exploitDB report
• Mirror of report and PoC
• Similar bug (CVE-2015-6770) with similar PoC (Oct 8, 2015)

Exploit Implementation

• PoC that just crashes on PS4 FW 4.05-5.56 and PS Vita FW 3.60

Patched

Yes in 6.00 FW. Vulnerable at least on PS4 FWs 4.05-5.56 and PS Vita FW 3.60.

---

FW 4.50-5.56 - WebCore::HTMLFrameElementBase::marginHeight() Heap UaF (CVE-2016-1859) leading to arbitrary RW

Credits

• Liang Chen, wushi of KeenLab, Tencent working with Trend Micro's Zero Day Initiative for discovering this vulnerability (2016-03-16)

Analysis

• NVD description of CVE-2016-1859 (May 5, 2016)
• ZDI advisory for CVE-2016-1859 (May 20, 2016)
• Writeup by Matt Molinyawe, Abdul-Aziz Hariri, and Jasiel Spelman (Trend Micro) (August 1, 2016)
• Writeup by Arayz (Wang Ao) (March 31, 2017)
• On the Analysis of Byte-Granularity Heap Randomization (October 24, 2019)

Bug Description

The specific flaw exists within the handling of GraphicsContext objects. By manipulating a document's elements an attacker can force this object in memory to be reused after it has been freed. An attacker can leverage this vulnerability to execute code under the context of the current process.

CVE-2016-1859 is a use-after-free vulnerability that existed in the Safari web browser. A GraphicsContext object is used in the setPlatformTextDrawingMode function after it has been freed. The successful triggering of the use-after-free vulnerability itself does not allow the attacker to directly change the control flow or disclose arbitrary memory contents. However, the use-after-free yields an arbitrary-memory-write primitive by hijacking a destination pointer that will be used for the memcpy function. Once the arbitrary-memory-write primitive is achieved, the attacker sprays the heap with string objects to achieve the arbitrary-memory-read primitive. Relying on the pointer width heap alignment, the attacker can accurately predict the exact address of one of the string objects among the heap spray and pinpoint the address of member variable. At this point, the attacker can overwrite the length member variable of a string object and partially disclose the out-of-bound heap area exceeding the buffer address of the string. The partial disclosure of the heap memory allows the attacker to extend the information leak step-by-step and ultimately allows full chaining of ROP, which leads to arbitrary code execution.

Exploit Implementation

• PoC publicly available
• No full exploit publicly available but exploitation description is detailed

Patched

Yes in 6.00 FW. Vulnerable on PS4 FWs 4.50-5.56.

---

FW 4.50-5.01 - Element::setAttributeNodeNS() UaF leading to arbitrary RW

Credits

• Lokihardt (from Google Project Zer0) for the exploit PoC (Mar 15, 2017)
• qwertyoruiop for the PS4 exploit (October 2017)
• Specter for the writeup (May 27, 2018)

Analysis

• exploit report by Lokihardt (Mar 15, 2017)
• First test on PS4 by WildCard (Oct 16, 2017)
• Specter's setAttributeNodeNS Exploit Writeup

Bug Description

By forcing setAttributeInternal() to call setAttributeNodeNS() twice, an attribute node reference will be added twice to the list. When one is free()'d, the second attribute still contains a duplicate stale reference, leading to a use-after-free (UAF) scenario.

Exploit Implementation

PS4 5.05 WebKit + Kernel Exploit

Patched

Yes in 5.03 FW.

---

FW 3.15-4.07 - Stack Uninitialized Read UaF leading to arbitrary RW

Credits

• qwertyoruiop for the exploit
• Specter for the writeup

Analysis

• Specter's 4.0x WebKit Exploit Writeup

Bug Description

Via a specially crafted valueOf() function of an arguments.length() function, non-zero indexes of the stack-allocated array are not initialized, leading to a stack uninitialized read. This can be abused to store a reference that can later be re-obtained post-GC (garbage collection) yielding a use-after-free() (UAF) situation.

Exploit Implementation

• Specter's commented and usable version
• mirrorred expl.js from qwertyoruiop

Patched

Yes in 4.50 FW

Tested

Works on 3.15-4.07. Not working on <= 3.11.

---

FW 3.15-3.70 - JSArray::sortCompactedVector() Heap UaF leading to arbitrary RW

Credits

• xyz for the original exploit on PSVita (HENkaku)
• Fire30 for porting to PS4
• Specter for improved PS4 playground

Analysis

• PSVita 3.60 HENKaku WebKit exploit writeup

Bug Description

When attempting to update a vector via sortCompactedVector() - data is written based on a pointer, though the pointer is not re-updated nor nulled. When this memory in free()'d, the reference is maintained and thus memory corruption can occur.

Exploit Implementation

• PSVita 3.60 WebKit exploit by xyz
• PS4 playground 3.15-3.70 by Fire30
• Improved PS4 playground 3.15-3.70 by Specter

Patched

Yes in 4.0?0? FW

Tested

Works on 3.15-3.70. Not working on <= 3.11. Maybe working on 4.00.

---

FW <= 3.50 - WebCore::TimerBase::heapPopMin() Heap UaF leading to crash

Analysis

• WebKit fix commit (17-05-2016)
• Summary of Critical and Exploitable iOS Vulnerabilities in 2016 by Min (Spark) Zheng, Cererdlong, Eakerqiu @ Team OverSky

Bug Description

"As of firmware version 3.55 a patch has been included to prevent a use-after-free segmentation fault from being exploited. This could have led to a ROP chain and code execution. It would have been cool if someone would have done some real research on it..." qwertyoruiop

Exploit Implementation

• Article about qwertyoruiop's tests (20-05-2016)
• Article about initial PoC for PS4 (21-05-2016)
• Initial PoC for PS4 (21-05-2016)
• iOS 9.3.2 WebKit RCE via heapPopMin (07-2016)
• qwertyoruiop's tweet (22-07-2016)
• mirror of iOS 9.3.2 WebKit RCE via heapPopMin

Patched

Yes in 3.55 FW

Tested

Works on 3.15, 3.50 FW. Maybe working on 3.51 FW.

---

FW <= ?2.50? - JavaScript OnLoad Handler Remote Code Execution Vulnerability (CVE-2005-1790) leading to crash or lag

Credits

• Benjamin Tobias Franz for the vulnerability discovery (2005-11-21)
• Stuart Pearson for the Proof of Concept on Microsoft Internet Explorer
• Sam Sharps for the Metasploit port (2012-01)
• Jeerum for disclosing that the vulnerability affects PS4 <=2.50 (2014-10-31).

Analysis

• Metasploit file by Sam Sharps (2012-01)
• PoC by wicar.org (before 2012-11-10)
• PoC for PS4 by Jeerum (2014-10-31)
• PS4 4.55 test of 1js by Jeerum

Bug Description

This bug is triggered when the browser handles a JavaScript 'onLoad' handler in conjunction with an improperly initialized 'window()' JavaScript function. This exploit results in a call to an address lower than the heap. The javascript prompt() places our shellcode near where the call operand points to. We call prompt() multiple times in separate iframes to place our return address. We hide the prompts in a popup window behind the main window. We spray the heap a second time with our shellcode and point the return address to the heap. I use a fairly high address to make this exploit more reliable. Microsoft Internet Explorer will crash when the exploit completes. Also, please note that Microsoft Internet Explorer must allow popups in order to continue exploitation.

Patched

Maybe

Tested

• Working on 1.76-2.50 FW: crash. 3.00-5.50 error CE-36329-3. 4.55 lag in background TV application (for example Netflix application).

---

FW <= 2.03 - WebCore::CSSSelector Heap Overflow (CVE-2014-1303) leading to arbitrary RW

Credits

• KeenTeam for finding and documenting the bug
• Liang Chen from KeenTeam for the writeups
• xyz for porting to PSVita FWs 3.30-3.36
• Fire30 for porting to PS4
• dreadlyei (unknown person, credited by Fire30)

Analysis

• BlackHat EU 2014 'WebKit Everywhere - Secure Or Not?' slides
• BlackHat EU 2014 'WebKit Everywhere - Secure Or Not?' PDF
• Attacking WebKit Applications by exploiting memory corruption bugs by Liang Chen

Bug Description

By forcing addRule() to be called on a CSS Selector via window.getMatchedCSSRules(), a 1-bit OOB write can be achieved and leveraged to corrupt heap memory.

Exploit Implementation

• ROP PoC for PS4 FW 2.03 by Fire30
• wololo article
• WebKit exploit for 3.30-3.36 FW PSVita by xyz: used in vitasploit
• PoC for Linux by RKX1209

Patched

Yes in 2.50 FW

Tested

• Working on 2.00-2.03 FW. Might work on 2.04 (99% sure as 2.04 PUP is about same size as 2.03 PUP).
• Working on AppleWebKit/537.73
• Maybe not working on FW < 2.00.

---

FW <= 2.03-? - WebCore::ImageInputType::attach Heap UaF (CVE-2013-2857) leading to ROP execution

Credits

• Chromium bugs reporters
• JumpCallPop, jam1garner, hedgeberg for inital exploit on Wii U
• yellows8 for ROP on Wii U
• orboditilt for increasing stability on Wii U
• zoogie for porting Wii U exploit to New3DS
• CelesteBlue for testing on PS4 FW 2.03

Analysis

• Chromium bug report with a PoC

Bug Description

Use-after-free with input type image. Error event was fired synchronously blowing away the input element from underneath.

Exploiting this vulnerability on PS4 is not good because:

• This vulnerability does not provide arbitrary RW without code execution, hence ROP chain (at least to stack pivot to JiT code) must be made with a memory dump or decrypted modules for this FW gotten using another vulnerability.
• There is usermode ASLR since about FW 1.70 so ROP chain gadgets must be relocated at runtime. This means another vulnerability allowing usermode arbitrary read is needed.
• As usually an arbitrary read vulnerability also gives arbitrary write, and as arbitrary RW leads to usermode code execution (by hijacking JS pointers in virtual table), this UaF is not needed at all.
• Even if we get ROP chain to work on PS4 with this UaF vulnerability, there is no evidence that a return to JavaScript from ROP chain is doable, making this exploit less convenient than arbitrary RW exploits method of getting code execution then returning to usermode by restoring vtable.

Exploit Implementation

• Initial Wii U implementation
• Wii U stabler implementation (last update May 22, 2018)
• Wii U tabler implementation (last update Jan 13, 2019)
• Wii U stabler implementation by Hiperhazz (last update May 26, 2020)
• New3DS implementation by zoogie (last update Aug 9, 2020)

Patched

Yes in ? FW

Tested

• Working on 2.03 FW. Might work on 2.04 (99% sure as 2.04 PUP is about same size as 2.03 PUP).

---

FW <= 1.76 - JSArray::sort() Heap Overflow (CVE-2012-3748, PSA 2013-0903-1) leading to arbitrary RW

Credits

• Vitaliy Toropov for the exploit on Mac OS X Safari (September 4, 2013)
• nas and Proxima for the first PS4 POC on 1.76 PS4 (Oct. 23, 2014)
• sony for patching the exploit in FW 2.00 (Oct 27, 2014)
• CTurt for the rewriting (PS4 1.76 PlayGround) and implementation with his 1.76 kexploit (December 6, 2015) [6]

Analysis

• CVE-2012-3748
• PSA 2013-0903-1
• Exploit PoC by Vitaliy Toropov

Bug Description

By forcing the compare function to reduce the size of the array, trailing items will be written out of bounds (OOB write), leading to heap memory corruption.

Exploit Implementation

• first POC for 1.76 PS4 by nas and Proxima
• mirror
• live test
• livetest2
• ROP2
• PS4 playground 1.76 by CTurt
• PSVita 2.00-3.20 WebKit exploit

Patched

Yes in 2.00 FW

Tested

• Working on 1.00-1.76 FW, AppleWebKit/531.3-536.26
• Might work on FW 0.930.020.

Usermode securities

Usermode ASLR

• Very old firmwares (<= 1.05) do not have ASLR enabled, but it was introduced sometime before firmware 1.70. "Address Space Layout Randomization" (ASLR) is a security technique which causes the base addresses of modules to be different every time you start the PS4.
• To defeat usermode ASLR on FWs >=1.70, we can use the module imports table to find other modules address once we know SceWebkit2 address.

Module imports table cleaned before execution

• Between 1.76 and 4.05, Sony did that to prevent WebKit exploiters from defeating usermode ASLR easily.
• Now we have to dump entire usermode sandboxed memory, and by studying it we can defeat ASLR:

1. Chose a function (ex: __stack_chk_fail) imported from libkernel.sprx by libSceWebkit2.sprx

2. Read pointer contained at the address where the call is done

3. Substract to this pointer the offset of the function (ex: __stack_chk_fail) in LibKernel module

4. This result is LibKernel base address. This method works for any imported module.

For FW >= 6.00, for web applications, libkernel.sprx has been replaced by libkernel_web.sprx and libSceWebKit2 by libSceNKWebKit.sprx. libkernel.sprx is still used by other applications.

DEP / NX

• "Data Execution Prevention" / "No eXecute" is enabled on all firmwares. It prevents allocating memory as both RW and RX at same time (RWX) so preventing us from writing shellcode to usermode memory then executing it.
• 2 ways to bypass this security: JiT vulnerability (FW <= 1.76) or ROP (all FWs).

JiT removed from webbrowser

• On FW <= 1.76, you could map RWX memory from ROP by abusing the JiT functionality and the sys_jitshm_create and sys_jitshm_alias system calls. This however was fixed after 1.76, as WebKit has been split into two processes. One handles javascript compilation and the other handles other web page elements like image rendering and DOM. The second process will request JiT memory upon hitting JavaScript via IPC (Inter-Process Communication). Since we no longer have access to the process responsible for JiT, we can no longer (at least currently), map RWX memory for proper code execution unless the kernel is patched.
• Checking the source code at ps4-oss, starting as early as FW 6.00, ENABLE_JIT=OFF for -DPORT=PlayStation4. It means that JIT functionality is completely removed from WebKit and there is no JIT coprocess that is allowed to request RWX memory to even attack. Even if there are JIT bugs that can lead us to request RWX memory in other platforms, we can't on the PS4 as there is no longer any JIT process. Unchecked all source codes, JIT process could have been removed earlier than 6.00. All exploits must use ROP.
• Workaround is to use ROP.

Syscalls removed

• See the PS4 Syscalls list.

Syscall 0 disabled i.e Error Kernel: The application directly issues a syscall instruction (24)

• Between 2.00 and 2.57, SCE has removed system call 0, so we can no longer call any syscall we want by specifying the call number in the rax register.
• Doing so now crashes the app and gives error CE-34878-0, SCE_KERNEL_ABORT_REASON_SYSTEM_ILLEGAL_FUNCTION_CALL, with the message "Kernel: The application directly issues a syscall instruction (24)".
• We now have to use wrappers provided to us from the libkernel / libkernel_web / libkernel_sys modules to access system calls.

bpf_write function stripped out of the kernel

• On 4.70, bpfwrite() was stripped out of the kernel entirely to patch kernel vulnerability exploited in 4.55 kexploit.

bpf_open function blocked for unprivileged processes

• On 5.50, opening BPF has been blocked for unprivileged processes such as WebKit and other apps/games. It's still present in the sandbox, however attempting to open it will fail and yield EPERM. This aims blocking BPF kernel exploits especially qwertyoruiop's BPF double free UAF.

bpf_ioctl function blocked or removed

• On FW 5.50+, opening BPF is still possible in less sandboxed apps like TestKit/DevKits fSELFs. But this is useless because ioctl does not work.

Device access blocked/removed from webbrowser

• Around 6.50-6.70, device access got blocked or removed. Now you can no longer access devices from the web browser.

Pointer poisoning in WebKit on 6.xx firmwares

• For select types implemented by WebKit (such as JSC::JSFunction), certain pointer fields are XOR'ed by a cryptographic key generated at runtime. The key is generated once every process launch, one must recover it to unpoison the pointers.

Kernel Exploits

FW <= 11.00 - Remote vulnerabilities in spp (yielding kernel ASLR defeat) (CVE-2006-4304 and no-CVE)

Credits

• 2006-08-23 Martin Husemann, Pavel Cahyna for discovering the first spp bug (CVE-2006-4304) on FreeBSD 4.11-6.1.
• 2023-09-22 TheFloW for discovering that PS4 and PS5 are vulnerable to CVE-2006-4304, discovering second spp bug, and chaining them together.
• 2024-01-27 anonymous for reporting publicly CVE-2006-4304 as working on PS4 and PS5. See [7] and [8].
• 2024-03 iMrDJAi for porting CVE-2006-4304 to PS4 and PS5.
• 2024-04-25 TheFloW for disclosing his HackerOne report including the second spp bug description.
• 2024-04-30 TheFloW for releasing his exploit code for PS4 9.00 and 11.00.

Analysis

• FreeBSD Security Advisory for CVE-2006-4304 (2006-08-23)
• HackerOne report about Remote vulnerabilities in spp by TheFloW (2023-09-22)

Bug Description

A malicious PPPoE server can cause denial-of-service or remote code execution in kernel context on the PS4/PS5. It does not require any usermode code execution to be triggered. There are two vulnerabilities that can be chained together to cause remote kernel Denial of Service, kernel ASLR defeat or kernel code execution : Heap buffer overwrite and overread in sppp_lcp_RCR and sppp_ipcp_RCR (CVE-2006-4304) and Integer underflow in sppp_pap_input leading to heap-buffer overread (no-CVE).

The PS4/PS5 must be connected using an ethernet cable to a device able to trigger PPPoE requests and analyze the responses.

Exploit Implementation

• CVE-2006-4304 PoC for FreeBSD9 by iMrDJAi (2024-04-07)
• CVE-2006-4304 PoC for PS4 and PS5 by iMrDJAi (2024-04-07)
• spp exploit for PS4 9.00 and 11.00 by TheFloW (2024-04-30)

Patched

Yes in 11.02 FW

---

FW <= 9.00 - PPPoE driver remote buffer overflow (CVE-2022-29867)

Credits

• 2021-09-24 m00nbsd for finding the vulnerability
• 2022-05-04 martin of NetBSD for fixing the vulnerability publicly in NetBSD 8 and 9
• 2022-05-11 m00nbsd for disclosing the vulnerability publicly on HackerOne

Analysis

• HackerOne report by m00nbsd (2021-09-24)
• NetBSD Security Advisory about CVE-2022-29867 (2022-05-04)

Bug Description

The PlayStation 4 has a kernel PPPoE driver, that originates from NetBSD. This driver has a kernel heap overflow vulnerability, that an attacker can remotely trigger over the LAN, with the ability to control both the contents that are overflown and their sizes.

Extract of NetBSD 8.3 changelog:

sys/net/if_pppoe.c			1.179

	pppoe(4): fix CVE-2022-29867 - discovery phase local network
	mbuf corruption.
	[martin, ticket #1740]

Do not allocate mbuf clusters when the caller (eroneously) asks
for more than MCLBYTES size, instead fail the allocation.

When we have received multiple PADO offer packets in the discovery
phase, do not combine tags from different packets. We are supposed
to pick one PADO packet and continue session establishment with that.

The second bug could cause code to trigger the first and create
invalid response packets and also overwrite data outside of
the allocated mbuf cluster.

Fixes CVE-2022-29867.

Diff after fix commit in NetBSD 8:

--- src/sys/net/if_pppoe.c	2020/02/13 19:37:39	1.125.6.10
+++ src/sys/net/if_pppoe.c	2022/05/04 15:36:35	1.125.6.11
@@ -1,4 +1,4 @@
-/* $NetBSD: if_pppoe.c,v 1.125.6.10 2020/02/13 19:37:39 martin Exp $ */
+/* $NetBSD: if_pppoe.c,v 1.125.6.11 2022/05/04 15:36:35 sborrill Exp $ */
 
 /*-
  * Copyright (c) 2002, 2008 The NetBSD Foundation, Inc.
@@ -30,7 +30,7 @@
  */
 
 #include <sys/cdefs.h>
-__KERNEL_RCSID(0, "$NetBSD: if_pppoe.c,v 1.125.6.10 2020/02/13 19:37:39 martin Exp $");
+__KERNEL_RCSID(0, "$NetBSD: if_pppoe.c,v 1.125.6.11 2022/05/04 15:36:35 sborrill Exp $");
 
 #ifdef _KERNEL_OPT
 #include "pppoe.h"
@@ -871,6 +871,10 @@ breakbreak:;
 			}
 			sc->sc_ac_cookie_len = ac_cookie_len;
 			memcpy(sc->sc_ac_cookie, ac_cookie, ac_cookie_len);
+		} else if (sc->sc_ac_cookie) {
+			free(sc->sc_ac_cookie, M_DEVBUF);
+			sc->sc_ac_cookie = NULL;
+			sc->sc_ac_cookie_len = 0;
 		}
 		if (relay_sid) {
 			if (sc->sc_relay_sid)
@@ -886,6 +890,10 @@ breakbreak:;
 			}
 			sc->sc_relay_sid_len = relay_sid_len;
 			memcpy(sc->sc_relay_sid, relay_sid, relay_sid_len);
+		} else if (sc->sc_relay_sid) {
+			free(sc->sc_relay_sid, M_DEVBUF);
+			sc->sc_relay_sid = NULL;
+			sc->sc_relay_sid_len = 0;
 		}
 		memcpy(&sc->sc_dest, eh->ether_shost, sizeof sc->sc_dest);
 		callout_stop(&sc->sc_timeout);
@@ -1313,6 +1321,9 @@ pppoe_get_mbuf(size_t len)
 {
 	struct mbuf *m;
 
+	if (len + sizeof(struct ether_header) > MCLBYTES)
+		return NULL;
+
 	MGETHDR(m, M_DONTWAIT, MT_DATA);
 	if (m == NULL)
 		return NULL;

Exploit Implementation

• PoC (poc.c) by m00nbsd not disclosed publicly

Patched

Yes in 9.03 FW according to Specter by diffing PS4 9.00 and 9.03 kernels

---

FW <= 9.00 - exFAT driver heap-based buffer overflow

Credits

• 2021-09-15 TheFloW for finding the vulnerability
• 2021-12-02 zecoxao for advicing to exploit the vulnerability after diffing PS4 9.00 and 9.03 kernels
• 2021-12-13 ChendoChap, Znullptr, Specter for PS4 9.00 kernel exploit implementation release

Analysis

• Vulnerability adviced by zecoxao for exploitation (2021-12-02)
• TheFloW's report on HackerOne (2021-09-15), disclosed on 2022-09-21

Bug Description

The PS4 kernel exFAT driver has a heap-based buffer overflow vulnerability that can be triggered by inserting a malicious USB storage device in PS4 in addition to having usermode code execution. Exploitation requires to flash a crafted exFAT image to a common USB storage device.

Exploit Implementation

• ChendoChap's implementation for PS4 9.00 (2021-12-13)

Patched

Yes in PS4 9.03 FW and PS5 4.50 FW

---

FW <= 7.55 - IP6_EXTHDR_CHECK Double Free (CVE-2020-9892)

Credits

• 2019-09-15 tuexen for finding the FreeBSD vulnerability [9]
• 2020-07-24 TheFloW for finding CVE-2020-9892 in XNU
• 2020-07-26 TheFloW for porting CVE-2020-9892 to PS4
• 2020-07-27 TheFloW for publishing publicly a PoC leading to code execution on XNU. [10]
• 2021-01-12 TheFloW for disclosing publicly the PS4 vulnerability. [11]
• 2021-01-20 sleirsgoevy for making a first working exploit for FreeBSD 9 [12]
• 2021-03-03 sleirsgoevy for making a second working exploit for FreeBSD 9 [13]
• 2021-03-12 sleirsgoevy for making the first public usable exploit for PS4 7.50-7.55 (https://twitter.com/sleirsgoevy/status/1370481212813348865)

Analysis

• Fix handling of Hop-by-Hop options over the loopback interface commits review (2019-09-15 to 2020-05-07)
• Apple iOS 13.6 and iPadOS 13.6 Security Update (2020-07-24)
• Apple macOS Catalina 10.15.6 Security Update (2020-07-24)
• TheFloW's report of the exploit with undisclosed PS4 and FreeBSD 9 PoCs
• TheFloW's writeup and PoC for XNU (2020-07-26)
• Vulnerability adviced by TheFloW for exploitation (2018-02-05)

Bug Description

Memory corruption can be achieved by sending fragmented IPv6 packets to loopback interface due to poor and inconsistent use of IP6_EXTHDR_CHECK.

The macro IP6_EXTHDR_CHECK can free the mbuf if the packet is sent to loopback interface. This fact is not considered in dest6_input(), frag6_input() and more. For example in dest6_input(), the double pointer is not updated.

Hence, when parsing next headers, the mbuf can be free'd once again, leading to a double free which behaves like a use-after-free when we allocate mbuf's again.

Normally, this path would not be triggerable, because sending to loopback interface requires SOCK_RAW root privileges. However, for some reason on the PS4 SOCK_RAW sockets can be opened in Webkit process! Moreover, CelesteBlue confirmed that SOCK_RAW sockets can also be opened in PS4 Kit fSELF.

According to TheFloW, the reliability of the FreeBSD 9 PoC is very high, around 80%, whereas the PS4 PoC's is not very high, he guesses around 20%.

Exploit Implementation

• TheFloW's writeup and PoC for XNU (2020-07-27)
• sleirsgoevy's first exploit PoC for FreeBSD 9 (2021-01-20)
• Demonstration video of sleirsgoevy's first exploit PoC for FreeBSD 9 (2021-01-20)
• Specter's kernel panic PoC for PS4 Web browser (2021-01-15)
• CelesteBlue's kernel panic PoC for PS4 Kit fSELF (2021-01-15)
• WiP exploit code by Specter and tihmstar (2021-02-24)
• sleirsgoevy's second exploit PoC for FreeBSD 9 (2021-03-03)
• Demonstration video of sleirsgoevy's second exploit PoC for FreeBSD 9 (2021-03-03)
• sleirsgoevy's implementation for PS4 7.5x (2021-03-12)
• Specter's implementation for FreeBSD 9 (2021-03-24)

Patched

Yes in 8.00 FW

---

FW <= 7.02 - IPV6_2292PKTOPTIONS UaF (yielding arbitrary kernel R/W) (CVE-2020-7457)

Credits

• 2018-08-18 up to 2020-07-06 Fire30 for finding and keeping the vulnerability as a private 0day for it not to be patched by SIE. [14]
• 2020-07-06 TheFloW for publishing publicly a PoC leading to code execution on FreeBSD. [15]
• sleirsgoevy and ChendoChap for porting the PoC to PS4 and chaining it with the 6.72 and 7.02 WebKit exploits.
• SIE for not patching this vulnerability on PS5 even when patched on PS4.
• TheFlow for announcing that PS5 kernel was exploited: TheFloW's PS5 kernel exploit announcement (2021-11-07) and later that it was that same vulnerability that was present in PS5 FW 3.00-4.51.

Analysis

• TheFloW's hackerone report of the PS4 kernel exploit with a FreeBSD 9-12 PoC
• FreeBSD Security Advisory FreeBSD-SA-20:20.ipv6
• FreeBSD patch for FreeBSD-SA-20:20.ipv6
• TheFloW's hackerone report of the PS5 kernel exploit

Bug Description

Due to missing locks in option IPV6_2292PKTOPTIONS of setsockopt, it is possible to race and free the struct ip6_pktopts buffer, while it is being handled by ip6_setpktopt. This structure contains pointers (ip6po_pktinfo) that can be hijacked to obtain arbitrary kernel R/W primitives. As a consequence, it is easy to have kernel code execution. This vulnerability is reachable from WebKit sandbox and is available in the latest FW, that is 7.02.

Another description: There is a race and use-after-free vulnerability in the FreeBSD kernel IPv6 socket handling. A missing synchronization lock in the `IPV6_2292PKTOPTIONS` option handling in `setsockopt` permits racing `ip6_setpktopt` access to a freed `ip6_pktopts` struct. This exploit overwrites the `ip6po_pktinfo` pointer of a `ip6_pktopts` struct in freed memory to achieve arbitrary kernel read/write.

Exploit Implementation

• TheFloW's PoC for FreeBSD 9 and 12
• PS4 6.72-7.02 WebKit + Kernel Exploit implementation by sleirsgoevy
• PS4 5.05-7.02 WebKit + Kernel Exploit implementation by ChendoChap

Patched

Yes in PS4 7.50 FW and in PS5 5.00 or 5.02 FW. Not working in PS5 FWs <= 2.50.

---

FW <= 5.07 - BPF Race Condition (Yielding Double Free())

Analysis

Specter's Writeup of the 5.05 BPF Race Condition

Bug Description

Due to improper locking, two threads can enter the BPF SETWF ioctl command handler. While the bug is similar to that of 4.55, the method of attack is slightly different. Since write() was removed for BPF in 4.70, instead of triggering a use-after-free with write() - SETWF is ran in parallel via threading. Eventually, both calls will copy the same pointer to the stack, leading to both threads free()'ing the same pointer, poisoning the freelist. This can later be leveraged via heap spraying to corrupt heap memory to obtain arbitrary code execution in supervisor mode (ring0).

Exploit Implementation

• PS4 5.05 WebKit + Kernel Exploit
• PS4 5.05 WebKit + Kernel Exploit Source

Patched

Yes in 5.50 FW

---

FW <= 4.55 - BPF Race Condition (Yielding UaF)

Analysis

Specter's Writeup of the 4.55 BPF Race Condition

Bug Description

Due to improper locking, two threads can enter the BPF ioctl command handlers for setting a new write filter (SETWF) and setting a filter (SETIF). Both threads will reference the same pointer. In specially crafted situations, one thread could free() this pointer while the other thread executes it as a filter post-validation. This allows an unprivileged user to obtain an out-of-bounds (OOB) write on the stack, leading to arbitrary code execution in supervisor mode (ring0).

Exploit Implementation

PS4 4.55 WebKit + Kernel Exploit
PS4 4.55 WebKit + Kernel Exploit Source

Patched

Yes in 4.70 FW

---

FW <= 6.00 ?6.02? - sys_getcontext Information Leak (kASLR defeat) (CVE-2018-17155)

Analysis

• https://www.cvedetails.com/cve/CVE-2018-17155/
• coming soon by CelesteBlue

Bug Description

System call 421 or sys_getcontext() initializes the structure pointed at by ucp to the currently active context. The vulnerability is, some areas of memory copied out are not initialized, and thus the function leaks memory at certain spots. This vector was patched in 6.20, as now before the buffer is used it is initialized to 0 via bzero().

Exploit Implementation

• QuickHEN by CelesteBlue (v2 not released yet)
• KitHEN by CelesteBlue (not released yet)

Patched

Yes somewhere between 6.00 and 6.20 FW

---

FW <= 4.07 - sys_thr_get_ucontext Information Leak (kASLR defeat)

Analysis

Specter's Writeup

Bug Description

System call 634 or sys_thr_get_ucontext() allows to obtain information on a given thread. The vulnerability is, some areas of memory copied out are not initialized, and thus the function leaks memory at certain spots. This vector was patched in 4.50, as now before the buffer is used it is initialized to 0 via bzero().

Exploit Implementation

PS4 4.05 WebKit + Kernel Exploit

Patched

Yes in 4.50 FW

---

FW <= 4.05 - NamedObj Type Confusion (Yielding UaF)

Credits

• Chaitlin Tech for having been the first to show they had pwned PS4 FW 4.01 at Geekpwn convention. (2016-10-24)

official video, tweet 1, tweet 2, tweet 3 (2016-10-25)

• fail0verflow for the first writeup (2017-10-19)
• Specter for rewriting the exploit using a different object, and releasing it publicly (2017-12-27)

Analysis

• fail0verflow's writeup on the 1.01-4.05 namedobj kernel exploit (2017-10-19)
• Specter's first writeup (2017-10-20)
• Specter's writeup on his 4.05 implementation (2017-12-28)

Bug Description

Type confusion in the namedobj system once exploited can lead to an arbitrary free() allowing an attacker to craft a use-after-free() (UAF) situation to corrupt kernel memory. This can be leveraged to eventually obtain an arbitrary code execution primitive in supervisor mode (ring0).

Exploit Implementation

PS4 4.05 WebKit + Kernel Exploit

Patched

Yes in 4.06 FW

Tested

Works on FWs 4.00-4.05. On <= 3.70 FW we have not found a way to leak the target object, but it might be doable as Fail0verflow did it on 1.01.

---

FW <= 1.76 - dlclose Kernel Heap Overflow

Credits

• Discovered by CTurt.
• Privately implemented thanks to qwertyoruiop.
• CTurt published a writeup.
• The exploit was publicly implemented by kR105 and on another side by Zer0xFF and Bigboss (psxdev).

Analysis

Analysis of sys_dynlib_prepare_dlclose PS4 kernel heap overflow (by CTurt with the help of qwertyoruiop)

Bug Description

Integer overflow in the sys_dynlib_prepare_dlclose() system call can lead to a heap overflow causing memory corruption, allowing an attacker to obtain code execution in supervisor mode (ring0).

Exploit Implementation

Public release by kR105

Patched

Yes in 2.00 FW

---

FW <= 1.76 - BadIRET (CVE-2014-9322, CVE-2015-5675)

Credits

• Andy Lutomirski for CVE-2014-9322 (2014-11-22)
• Konstantin Belousov, Andrew Lutomirski for CVE-2015-5675 (2015-07-08)
• Adam Zabrocki (pi3) for asking CTurt to test CVE-2015-5675 on PS4 (2015-08-21) [16], [17]
• Volodymyr Pikhur for exploiting FreeBSD and PS4 in private (2015-09-24) [18]
• CTurt for porting the exploit from FreeBSD 9 to PS4 (2015-12-06) [19]

Analysis

• Fix commit for Linux CVE-2014-9322 by Andy Lutomirski
• NVD Bug Description for CVE-2014-9322
• 2014-9322 analysis by Rafal Wojtczuk
• CVE-2014-9322 analysis by Adam Zabrocki (pi3)

• CVE-2015-5675 PoC for FreeBSD by Konstantin Belousov, Andrew Lutomirski (2015-07-08)
• FreeBSD security advisory by Konstantin Belousov, Andrew Lutomirski (2015-08-25)

• Hacking the PS4, part 3 - Kernel exploitation (by CTurt) (2015-12-17

Bug Description

Faults associated with the stack segment (SS) register are not handled properly, allowing unprivileged users to trigger an IRET instruction that accesses a GS Base from usermode memory, providing an attacker with a method of privilege escalation.

Exploit Implementation

• CVE-2014-9322 kernel panic PoC for Linux by Emeric Nasi (2015-03-04)
• CVE-2014-9322 exploit code for Linux by Ren Kimura (@RKX1209) (2017-07-19)
• CVE-2014-9322 exploit code for Linux by Ren Kimura (@RKX1209) (2017-07-24)
• CVE-2015-5675 PoC for FreeBSD by Konstantin Belousov, Andrew Lutomirski (2015-07-08)

Patched

Yes in 2.00 FW

FW ??? - setlogin Information Leak (CVE-2014-8476)

Warning: this has not been tested on PS4.

Credits

• Mateusz Guzik for finding the vulnerability
• Volodymyr Pikhur for advising to use this vulnerability at his Playstation 4 Rest Mode DEMO in REcon Brussels 2018
• Francisco Falcon for making a PoC on FreeBSD 8.4

Analysis

• CVE details
• FreeBSD Security Advisories

Bug Description

The setlogin function in FreeBSD 8.4 through 10.1-RC4 does not initialize the buffer used to store the login name, which allows local users to obtain sensitive information from kernel memory via a call to getlogin, which returns the entire buffer.

When setlogin(2) is called while setting up a new login session, the login name is copied into an uninitialized stack buffer, which is then copied into a buffer of the same size in the session structure. The getlogin(2) system call returns the entire buffer rather than just the portion occupied by the login name associated with the session.

An unprivileged user can access this memory by calling getlogin(2) and reading beyond the terminating NUL character of the resulting string. Up to 16 (FreeBSD 8) or 32 (FreeBSD 9 and 10) bytes of kernel memory may be leaked in this manner for each invocation of setlogin(2).

This memory may contain sensitive information, such as portions of the file cache or terminal buffers, which an attacker might leverage to obtain elevated privileges.

Exploit Implementation

• PoC for FreeBSD 8.4 by Francisco Falcon

Patched

?

Kernel securities

Kernel ASLR

Since 3.50 FW, ASLR (Address Space Layout Randomization) has been enabled in PS4 kernel. This means that to properly exploit the kernel to escalate privileges, an information disclosure vulnerability will most likely be needed to defeat ASLR and locate the kernel in memory.

Kernel SMAP

PS4 APU does not support SMEP ("Supervisor Mode Execution Prevention") so there is no way it supports SMAP ("Supervisor Mode Access Prevention"). However, in PS4 5.0x FW and above, a sort of SMAP was added to the kernel to prevent exploiters from pivoting the kernel stack pointer (RSP) to usermode memory: attempting to do so would crash the system. Sony probably added checks into the scheduler to check the stack pointer (RSP) against usermode addresses when running in kernel context. A new exploitation strategy is needed to run kernel ROP chains because an exploiters now needs to get his kernel ROP chain inside kernel memory to be executed.

SMAP bypass method: JOP

To bypass PS4 SMAP, qwertyoruiop decided in his 5.05 PS4 kernel exploit to go with the method he used on the Apple iPhone 7 - essentially using JOP to push a bunch of stack frames onto the kernel stack, and memcpy()'ing the kernel ROP chain into RSP. qwertyoruiop explained: "JOP seems to work, but exploit is not reliable enough to repeat it multiple times implementing logic in-between (like on the FW 4.55 kernel bug where every primitive would re-exploit the bug). Using pure JOP logic would be long because of the need to find good instructions gadgets, and would vary a lot from a FW version to another. The strategy chosen is thread-safe and calling-convention aware, but most importantly pivot-less. We use JOP to implement a simple loop based on deref&branch logic. Every iteration runs a function prolog followed by a branch. This pushes lots of stack frames on stack, padding RSP. When loop is done, prepare call to memcpy with RDI = RSP, RSI = controlled pointer, RDX = (size of pushed stack frame * number of iterations - 1). We overwrite all fake frames but one with ROP. Memcpy will return into our first gadget, kickstarting the chain. At tail end of chain we just return into matching function epilog to resume clean execution by popping the one untouched frame. RSP never pivoted so PS4 successfully runs the kernel ROP chain."

• Detailed annotation of the PS4 5.05 kernel exploit by Specter
• Zero2Ring0 Slides by qwertyoruiop (Archive)
• From ROP to JOP article by Marco Ramilli

SMAP bypass method: cli/sti

A PS4 SMAP bypass has been showed by sleirsgoevy in his 6.72 PS4 kernel exploit implementation. It consists in wrapping the main kernel ROP with cli/sti pair, which would prevent it from being preempted. This way the thread's CPU core will not run other kernel code during kernel ROP execution, and other cores have no way of detecting the stack pivot, so the mitigation is defeated.

• cli/sti SMAP bypass in 6.72 PS4 kernel exploit

SMAP bypass method: CVE-2021-29628

A SMAP bypass has been found by m00nbsd while working on FreeBSD 12. It is named CVE-2021-29628 and affects FreeBSD 12.2 and later (til it was patched). It does not work on PS4 because PS4 kernel is based on FreeBSD 9 which did not contain the vulnerability and because PS4 SMAP does not come from FreeBSD but is custom from Sony. It used to work on PS5 before it was disclosed and patched on PS5 FW 2.30 or later according to dates.

• CVE-2021-29628 (FreeBSD SMAP bypass) by m00nbsd
• CVE-2021-29628 (PS5 SMAP bypass) by m00nbsd

CR0.WP protection

At least since firmware 6.51 Sony instrumented all instructions that write to the CR0 register with checks for attempts to clear CR0.WP (Write Protect), which is necessary for patching the kernel. This is what it looks like in 6.51 kernel:

  a1b79:       0f 22 c0                mov    cr0,rax
  a1b7c:       48 a9 00 00 01 00       test   rax,0x10000
  a1b82:       75 02                   jne    a1b86 <-- skip the next instruction if CR0.WP is not cleared
  a1b84:       0f 0b                   ud2    <-- #UD exception, causes a kernel panic
  a1b86:       c3                      ret 

Note that the check is after the write, to prevent a ROP gadget from pointing straight at the mov and skipping the verification.

Bypasses (in chronological order):

• Use an "unintended" mov to cr0 in the middle of another instruction (e.g. instruction "call $+0x220f1c" (e8 17 0f 22 00) contains an unintended "mov cr0, rax" (0f 22 00))
• Use kernel write to give your process JIT permissions, allocate JIT memory, and put entirely custom code there (avoids the problem altogether, as it is specific to ROP)
• Since the IDT is writable on FreeBSD and PS4, it is possible to overwrite an exception handler without clearing CR0.WP first. One can overwrite the handler of #UD with a gadget of their choice (a stack pivot, or a "add rsp, ... ; ret", or whatever else), and the UD2 instruction in the mitigation code will happily jump to it instead of the real handler, with CR0.WP cleared.