{"dataType":"CVE_RECORD","dataVersion":"5.2","cveMetadata":{"cveId":"CVE-2023-53718","assignerOrgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","state":"PUBLISHED","assignerShortName":"Linux","dateReserved":"2025-10-22T13:21:37.347Z","datePublished":"2025-10-22T13:23:50.809Z","dateUpdated":"2026-08-05T09:15:41.495Z"},"containers":{"cna":{"providerMetadata":{"orgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","shortName":"Linux","dateUpdated":"2026-08-05T09:15:41.495Z"},"descriptions":[{"lang":"en","value":"In the Linux kernel, the following vulnerability has been resolved:\n\nring-buffer: Do not swap cpu_buffer during resize process\n\nWhen ring_buffer_swap_cpu was called during resize process,\nthe cpu buffer was swapped in the middle, resulting in incorrect state.\nContinuing to run in the wrong state will result in oops.\n\nThis issue can be easily reproduced using the following two scripts:\n/tmp # cat test1.sh\n//#! /bin/sh\nfor i in `seq 0 100000`\ndo\n         echo 2000 > /sys/kernel/debug/tracing/buffer_size_kb\n         sleep 0.5\n         echo 5000 > /sys/kernel/debug/tracing/buffer_size_kb\n         sleep 0.5\ndone\n/tmp # cat test2.sh\n//#! /bin/sh\nfor i in `seq 0 100000`\ndo\n        echo irqsoff > /sys/kernel/debug/tracing/current_tracer\n        sleep 1\n        echo nop > /sys/kernel/debug/tracing/current_tracer\n        sleep 1\ndone\n/tmp # ./test1.sh &\n/tmp # ./test2.sh &\n\nA typical oops log is as follows, sometimes with other different oops logs.\n\n[  231.711293] WARNING: CPU: 0 PID: 9 at kernel/trace/ring_buffer.c:2026 rb_update_pages+0x378/0x3f8\n[  231.713375] Modules linked in:\n[  231.714735] CPU: 0 PID: 9 Comm: kworker/0:1 Tainted: G        W          6.5.0-rc1-00276-g20edcec23f92 #15\n[  231.716750] Hardware name: linux,dummy-virt (DT)\n[  231.718152] Workqueue: events update_pages_handler\n[  231.719714] pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)\n[  231.721171] pc : rb_update_pages+0x378/0x3f8\n[  231.722212] lr : rb_update_pages+0x25c/0x3f8\n[  231.723248] sp : ffff800082b9bd50\n[  231.724169] x29: ffff800082b9bd50 x28: ffff8000825f7000 x27: 0000000000000000\n[  231.726102] x26: 0000000000000001 x25: fffffffffffff010 x24: 0000000000000ff0\n[  231.728122] x23: ffff0000c3a0b600 x22: ffff0000c3a0b5c0 x21: fffffffffffffe0a\n[  231.730203] x20: ffff0000c3a0b600 x19: ffff0000c0102400 x18: 0000000000000000\n[  231.732329] x17: 0000000000000000 x16: 0000000000000000 x15: 0000ffffe7aa8510\n[  231.734212] x14: 0000000000000000 x13: 0000000000000000 x12: 0000000000000002\n[  231.736291] x11: ffff8000826998a8 x10: ffff800082b9baf0 x9 : ffff800081137558\n[  231.738195] x8 : fffffc00030e82c8 x7 : 0000000000000000 x6 : 0000000000000001\n[  231.740192] x5 : ffff0000ffbafe00 x4 : 0000000000000000 x3 : 0000000000000000\n[  231.742118] x2 : 00000000000006aa x1 : 0000000000000001 x0 : ffff0000c0007208\n[  231.744196] Call trace:\n[  231.744892]  rb_update_pages+0x378/0x3f8\n[  231.745893]  update_pages_handler+0x1c/0x38\n[  231.746893]  process_one_work+0x1f0/0x468\n[  231.747852]  worker_thread+0x54/0x410\n[  231.748737]  kthread+0x124/0x138\n[  231.749549]  ret_from_fork+0x10/0x20\n[  231.750434] ---[ end trace 0000000000000000 ]---\n[  233.720486] Unable to handle kernel NULL pointer dereference at virtual address 0000000000000000\n[  233.721696] Mem abort info:\n[  233.721935]   ESR = 0x0000000096000004\n[  233.722283]   EC = 0x25: DABT (current EL), IL = 32 bits\n[  233.722596]   SET = 0, FnV = 0\n[  233.722805]   EA = 0, S1PTW = 0\n[  233.723026]   FSC = 0x04: level 0 translation fault\n[  233.723458] Data abort info:\n[  233.723734]   ISV = 0, ISS = 0x00000004, ISS2 = 0x00000000\n[  233.724176]   CM = 0, WnR = 0, TnD = 0, TagAccess = 0\n[  233.724589]   GCS = 0, Overlay = 0, DirtyBit = 0, Xs = 0\n[  233.725075] user pgtable: 4k pages, 48-bit VAs, pgdp=0000000104943000\n[  233.725592] [0000000000000000] pgd=0000000000000000, p4d=0000000000000000\n[  233.726231] Internal error: Oops: 0000000096000004 [#1] PREEMPT SMP\n[  233.726720] Modules linked in:\n[  233.727007] CPU: 0 PID: 9 Comm: kworker/0:1 Tainted: G        W          6.5.0-rc1-00276-g20edcec23f92 #15\n[  233.727777] Hardware name: linux,dummy-virt (DT)\n[  233.728225] Workqueue: events update_pages_handler\n[  233.728655] pstate: 200000c5 (nzCv daIF -PAN -UAO -TCO -DIT -SSBS BTYPE=--)\n[  233.729054] pc : rb_update_pages+0x1a8/0x3f8\n[  233.729334] lr : rb_update_pages+0x154/0x3f8\n[  233.729592] sp : ffff800082b9bd50\n[  233.729792] x29: ffff800082b9bd50 x28: ffff8000825f7000 x27: 00000000\n---truncated---"}],"metrics":[{"cvssV3_1":{"version":"3.1","vectorString":"CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H","baseScore":7.8,"baseSeverity":"HIGH"},"scenarios":[{"lang":"en","value":"AV:L - Both sides of the race are reached only through local tracefs/debugfs file writes (`buffer_size_kb` and `current_tracer`/latency-tracer activity); there is no remote or adjacent-network path into `ring_buffer_resize()` or `ring_buffer_swap_cpu()`.\nAC:L - The attacker drives both racers itself — a loop writing `buffer_size_kb` while the enabled `irqsoff`/`wakeup` tracer swaps buffers on latency events the attacker generates with its own syscalls (and can force repeatedly via `echo 0 > tracing_max_latency`); the upstream reproducer is two trivial shell loops that oops the kernel within minutes.\nPR:L - No `capable()`/`CAP_SYS_ADMIN` check guards either path — access is pure tracefs DAC, and the `gid=` mount option plus Android/perfetto, ChromeOS and embedded tracing setups routinely grant an unprivileged tracing user write access to `buffer_size_kb` and `current_tracer`.\nUI:N - The attacker performs all writes and generates the latency events itself; no victim action or cooperation is required.\nS:U - The corruption is confined to kernel ring-buffer state within the same security authority; no VM, IOMMU, or sandbox boundary is crossed.\nC:H - The stale `update_done` completion leads to `rb_remove_pages(cpu_buffer, 0)` freeing every page of a live ring buffer while the buffer still references them, so subsequent reads of `trace`/`trace_pipe` return the contents of freed-and-reallocated kernel pages — an arbitrary kernel-memory disclosure primitive.\nI:H - The same use-after-free lets active tracing writers keep recording partially attacker-controlled data (e.g. via `trace_marker`) into freed pages, and the concurrent `rb_insert_pages()`/`__rb_allocate_pages()` races corrupt the buffer_page lists and double-free pages — a heap write primitive suitable for control-flow hijacking.\nA:H - The commit demonstrates a `RB_WARN_ON` in `rb_update_pages()` followed by a kernel NULL-pointer-dereference oops, and mass-freeing a live ring buffer's pages reliably panics the machine."}]}],"affected":[{"product":"Linux","vendor":"Linux","defaultStatus":"unaffected","repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","programFiles":["kernel/trace/ring_buffer.c","kernel/trace/trace.c"],"versions":[{"version":"83f40318dab00e3298a1f6d0b12ac025e84e478d","lessThan":"66a3b2a121386702663065d5c9e5a33c03d3f4a2","status":"affected","versionType":"git"},{"version":"83f40318dab00e3298a1f6d0b12ac025e84e478d","lessThan":"49b830d75f03d5dd41146d10e4d3e2a8211c4b94","status":"affected","versionType":"git"},{"version":"83f40318dab00e3298a1f6d0b12ac025e84e478d","lessThan":"128c06a34cfe55212632533a706b050d54552741","status":"affected","versionType":"git"},{"version":"83f40318dab00e3298a1f6d0b12ac025e84e478d","lessThan":"02e52d7daaa3f0f48819f198092cf4871065bbf7","status":"affected","versionType":"git"},{"version":"83f40318dab00e3298a1f6d0b12ac025e84e478d","lessThan":"8a96c0288d0737ad77882024974c075345c72011","status":"affected","versionType":"git"}]},{"product":"Linux","vendor":"Linux","defaultStatus":"affected","repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","programFiles":["kernel/trace/ring_buffer.c","kernel/trace/trace.c"],"versions":[{"version":"3.5","status":"affected"},{"version":"0","lessThan":"3.5","status":"unaffected","versionType":"semver"},{"version":"5.10.192","lessThanOrEqual":"5.10.*","status":"unaffected","versionType":"semver"},{"version":"5.15.128","lessThanOrEqual":"5.15.*","status":"unaffected","versionType":"semver"},{"version":"6.1.47","lessThanOrEqual":"6.1.*","status":"unaffected","versionType":"semver"},{"version":"6.4.12","lessThanOrEqual":"6.4.*","status":"unaffected","versionType":"semver"},{"version":"6.5","lessThanOrEqual":"*","status":"unaffected","versionType":"original_commit_for_fix"}]}],"cpeApplicability":[{"nodes":[{"operator":"OR","negate":false,"cpeMatch":[{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"3.5","versionEndExcluding":"5.10.192"},{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"3.5","versionEndExcluding":"5.15.128"},{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"3.5","versionEndExcluding":"6.1.47"},{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"3.5","versionEndExcluding":"6.4.12"},{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"3.5","versionEndExcluding":"6.5"}]}]}],"references":[{"url":"https://git.kernel.org/stable/c/66a3b2a121386702663065d5c9e5a33c03d3f4a2"},{"url":"https://git.kernel.org/stable/c/49b830d75f03d5dd41146d10e4d3e2a8211c4b94"},{"url":"https://git.kernel.org/stable/c/128c06a34cfe55212632533a706b050d54552741"},{"url":"https://git.kernel.org/stable/c/02e52d7daaa3f0f48819f198092cf4871065bbf7"},{"url":"https://git.kernel.org/stable/c/8a96c0288d0737ad77882024974c075345c72011"}],"title":"ring-buffer: Do not swap cpu_buffer during resize process","x_generator":{"engine":"bippy-1.2.0"}}}}