CVE WatchAug 4, 2026

CVE-2024-26960

In the Linux kernel, the following vulnerability has been resolved: mm: swap: fix race between free_swap_and_cache() and swapoff() There w

CVSS

7.0

High

EPSS

0.2%

KEV

—

Exploit Today

0-100

Published: May 1, 2024 · Last modified: Aug 4, 2026 · CWE-362

EPSS · 30d

0.2%EPSS · 30 days0.2%

2026-07-092026-08-05

Technical description

In the Linux kernel, the following vulnerability has been resolved: mm: swap: fix race between free_swap_and_cache() and swapoff() There was previously a theoretical window where swapoff() could run and teardown a swap_info_struct while a call to free_swap_and_cache() was running in another thread. This could cause, amongst other bad possibilities, swap_page_trans_huge_swapped() (called by free_swap_and_cache()) to access the freed memory for swap_map. This is a theoretical problem and I haven't been able to provoke it from a test case. But there has been agreement based on code review that this is possible (see link below). Fix it by using get_swap_device()/put_swap_device(), which will stall swapoff(). There was an extra check in _swap_info_get() to confirm that the swap entry was not free. This isn't present in get_swap_device() because it doesn't make sense in general due to the race between getting the reference and swapoff. So I've added an equivalent check directly in free_swap_and_cache(). Details of how to provoke one possible issue (thanks to David Hildenbrand for deriving this): --8<----- __swap_entry_free() might be the last user and result in "count == SWAP_HAS_CACHE". swapoff->try_to_unuse() will stop as soon as soon as si->inuse_pages==0. So the question is: could someone reclaim the folio and turn si->inuse_pages==0, before we completed swap_page_trans_huge_swapped(). Imagine the following: 2 MiB folio in the swapcache. Only 2 subpages are still references by swap entries. Process 1 still references subpage 0 via swap entry. Process 2 still references subpage 1 via swap entry. Process 1 quits. Calls free_swap_and_cache(). -> count == SWAP_HAS_CACHE [then, preempted in the hypervisor etc.] Process 2 quits. Calls free_swap_and_cache(). -> count == SWAP_HAS_CACHE Process 2 goes ahead, passes swap_page_trans_huge_swapped(), and calls __try_to_reclaim_swap(). __try_to_reclaim_swap()->folio_free_swap()->delete_from_swap_cache()-> put_swap_folio()->free_swap_slot()->swapcache_free_entries()-> swap_entry_free()->swap_range_free()-> ... WRITE_ONCE(si->inuse_pages, si->inuse_pages - nr_entries); What stops swapoff to succeed after process 2 reclaimed the swap cache but before process1 finished its call to swap_page_trans_huge_swapped()? --8<-----

Official references

Related CVEs

CVECVSSEPSSKEVRExploitTitleMod.

CVE-2026-706407.0 HIG

—

———llama.cpp builds b1886 through b7445 contain a race condition use-after-free vulnerability in the LLaMA-Android JNI wrapper where bench_1model() and free_1context() lack synchronization, allowing Thread A to operate on freed memory while Thread B concurrently frees the llama_context. Attackers can exploit this by performing heap spray with attacker-controlled data containing a fake vtable to hijack the vtable pointer at offset +0x30, causing llama_batch_allocr::clear() to dereference arbitrary memory and achieve remote code execution.13h CVE-2026-436318.1 HIG

—

———llama.cpp builds b7492 through the latest b9060 contains a use-after-free vulnerability in the vocab pointer of llama-server when the --sleep-idle-seconds feature is enabled, allowing unauthenticated remote attackers to execute arbitrary code. Attackers can trigger the vulnerability by sending requests to affected endpoints while the server transitions to sleep mode, causing concurrent worker threads to dereference a freed vocab pointer that can be reclaimed with attacker-controlled data to achieve remote code execution.13h CVE-2026-191397.4 HIG

—

———Race in CredentialProvider in Google Chrome on Windows prior to 151.0.7922.109 allowed a local attacker to perform OS-level privilege escalation via a malicious file. (Chromium security severity: High)6h CVE-2026-481545.9 MED

15.9%

——5GoRest is a Golang starter kit built with the Gin framework for prototyping and developing RESTful APIs. In versions prior to 1.12.2 nMemorySecret2FA contains a race condition due to an unsynchronized package-level map used to store 2FA secrets. Multiple HTTP handlers in handler/login.go and handler/twoFA.go read from and write to this map concurrently, and because Go's runtime treats unsynchronized concurrent map access as an unrecoverable fatal error, an attacker can repeatedly trigger this condition to crash the process on demand. This results in high, repeatable availability impact with no confidentiality or integrity consequences. This issue has been fixed in version 1.12.2.2d CVE-2026-476206.5 MED

12.8%

——4NVIDIA Dynamo for Linux contains a vulnerability where an attacker could cause a race condition in the LoRA manager singleton initialization. A successful exploit of this vulnerability might lead to data tampering and denial of service.19h CVE-2025-15630—

8.7%

——3A race condition exists in the cloud-based Omada device adoption process when an attacker may be able to interact with the adoption workflow before a legitimate device completes registration, resulting in provisioning information being delivered to an attacker. Successful exploitation may allow disclosure of provisioning information intended for a legitimate device.19h