CVE-2026-63811
In the Linux kernel, the following vulnerability has been resolved:
f2fs: read COW data with the original inode during atomic write
When updating an atomic-write file, f2fs_write_begin() may read the previously written data back from the COW inode: prepare_atomic_write_begin() locates the block in the COW inode and sets use_cow, and the read bio is then built with the COW inode:
and f2fs_grab_read_bio() decides whether to schedule fs-layer decryption (STEP_DECRYPT) for the bio based on that inode via fscrypt_inode_uses_fs_layer_crypto().
However, the folio being filled belongs to the original inode (folio->mapping->host == inode), and the data stored in the COW block was encrypted (or left as plaintext) using the original inode's context, not the COW inode's -- see f2fs_encrypt_one_page(), which keys off fio->page->mapping->host. fscrypt_decrypt_pagecache_blocks() likewise operates on folio->mapping->host.
Read full descriptionShow less
The COW inode is created as a tmpfile in the parent directory and inherits its encryption policy from there. With test_dummy_encryption the newly created COW inode gets the dummy policy and becomes encrypted, while a pre-existing regular file -- created before the policy applied, e.g. already present in the on-disk image -- stays unencrypted. The read path then sets STEP_DECRYPT based on the encrypted COW inode and calls fscrypt_decrypt_pagecache_blocks() on a folio whose host (the unencrypted original inode) has a NULL ->i_crypt_info, dereferencing it:
The COW inode is only needed to locate the on-disk block, and that block address is already resolved into @blkaddr by prepare_atomic_write_begin() via __find_data_block(cow_inode, ...); f2fs_submit_page_read() then reads from that physical @blkaddr directly, so the inode argument only selects the post-read crypto context, not which block is fetched. Reading with @inode therefore returns the same (latest, not-yet-committed) COW data, while making both the fs-layer decryption decision and the inline crypto path use the correct (original inode's) key.
With the COW inode no longer used at the read site, the use_cow flag has no remaining consumer; drop it from f2fs_write_begin() and prepare_atomic_write_begin().
Technical details traces, logs and code from the original report
f2fs_submit_page_read(use_cow ? F2FS_I(inode)->cow_inode : inode, ...); Oops: general protection fault, probably for non-canonical address ... KASAN: null-ptr-deref in range [0x0000000000000008-0x000000000000000f] RIP: 0010:fscrypt_decrypt_pagecache_blocks+0xa0/0x310 Workqueue: f2fs_post_read_wq f2fs_post_read_work Call Trace: fscrypt_decrypt_bio+0x1eb/0x340 f2fs_post_read_work+0xba/0x140 process_one_work+0x91c/0x1a40 worker_thread+0x677/0xe90 kthread+0x2bc/0x3a0
CVSS
NVD hasn't assigned a CVSS score to this CVE (common since the April 2026 policy change).
Exploitation probability (EPSS)
- Probability of exploitation in the next 30 days: 0.11%
- Percentile among all scored CVEs: 1
- Score date: 10/6/2026
EPSS (Exploit Prediction Scoring System, FIRST) estimates how likely a vulnerability is to be exploited in the wild within 30 days. It complements CVSS (impact) and CISA KEV (confirmed exploitation).
Affected technologies (1)
⚠ AI-inferred from the description — NVD hasn't analyzed this CVE yet, these aren't verified CPEs.
References
Raw JSON (NVD)
Show
{
"id": "CVE-2026-63811",
"cveTags": [],
"metrics": {},
"affected": [
{
"source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67",
"affectedData": [
{
"repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git",
"vendor": "Linux",
"product": "Linux",
"versions": [
{
"status": "affected",
"version": "591fc34e1f98b0d7eef4aa3440bfdff3c5a1cadd",
"lessThan": "a92332f32a8d31a7eee47b1dc1d751cb3319908f",
"versionType": "git"
},
{
"status": "affected",
"version": "591fc34e1f98b0d7eef4aa3440bfdff3c5a1cadd",
"lessThan": "a41075acde0124d2f8a5f563068a5d63e8ffd57b",
"versionType": "git"
}
],
"programFiles": [
"fs/f2fs/data.c"
],
"defaultStatus": "unaffected"
},
{
"repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git",
"vendor": "Linux",
"product": "Linux",
"versions": [
{
"status": "affected",
"version": "6.4"
},
{
"status": "unaffected",
"version": "0",
"lessThan": "6.4",
"versionType": "semver"
},
{
"status": "unaffected",
"version": "7.1.3",
"versionType": "semver",
"lessThanOrEqual": "7.1.*"
},
{
"status": "unaffected",
"version": "7.2",
"versionType": "original_commit_for_fix",
"lessThanOrEqual": "*"
}
],
"programFiles": [
"fs/f2fs/data.c"
],
"defaultStatus": "affected"
}
]
}
],
"published": "2026-07-19T12:16:54.023",
"references": [
{
"url": "https://git.kernel.org/stable/c/a41075acde0124d2f8a5f563068a5d63e8ffd57b",
"source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67"
},
{
"url": "https://git.kernel.org/stable/c/a92332f32a8d31a7eee47b1dc1d751cb3319908f",
"source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67"
}
],
"vulnStatus": "Awaiting Analysis",
"descriptions": [
{
"lang": "en",
"value": "In the Linux kernel, the following vulnerability has been resolved:\n\nf2fs: read COW data with the original inode during atomic write\n\nWhen updating an atomic-write file, f2fs_write_begin() may read the\npreviously written data back from the COW inode:\nprepare_atomic_write_begin() locates the block in the COW inode and sets\nuse_cow, and the read bio is then built with the COW inode:\n\n\tf2fs_submit_page_read(use_cow ? F2FS_I(inode)->cow_inode : inode,\n\t\t\t ...);\n\nand f2fs_grab_read_bio() decides whether to schedule fs-layer decryption\n(STEP_DECRYPT) for the bio based on that inode via\nfscrypt_inode_uses_fs_layer_crypto().\n\nHowever, the folio being filled belongs to the original inode\n(folio->mapping->host == inode), and the data stored in the COW block was\nencrypted (or left as plaintext) using the original inode's context, not\nthe COW inode's -- see f2fs_encrypt_one_page(), which keys off\nfio->page->mapping->host. fscrypt_decrypt_pagecache_blocks() likewise\noperates on folio->mapping->host.\n\nThe COW inode is created as a tmpfile in the parent directory and inherits\nits encryption policy from there. With test_dummy_encryption the newly\ncreated COW inode gets the dummy policy and becomes encrypted, while a\npre-existing regular file -- created before the policy applied, e.g.\nalready present in the on-disk image -- stays unencrypted. The read\npath then sets STEP_DECRYPT based on the encrypted COW inode and calls\nfscrypt_decrypt_pagecache_blocks() on a folio whose host (the unencrypted\noriginal inode) has a NULL ->i_crypt_info, dereferencing it:\n\n Oops: general protection fault, probably for non-canonical address ...\n KASAN: null-ptr-deref in range [0x0000000000000008-0x000000000000000f]\n RIP: 0010:fscrypt_decrypt_pagecache_blocks+0xa0/0x310\n Workqueue: f2fs_post_read_wq f2fs_post_read_work\n Call Trace:\n fscrypt_decrypt_bio+0x1eb/0x340\n f2fs_post_read_work+0xba/0x140\n process_one_work+0x91c/0x1a40\n worker_thread+0x677/0xe90\n kthread+0x2bc/0x3a0\n\nThe COW inode is only needed to locate the on-disk block, and that block\naddress is already resolved into @blkaddr by prepare_atomic_write_begin()\nvia __find_data_block(cow_inode, ...); f2fs_submit_page_read() then reads\nfrom that physical @blkaddr directly, so the inode argument only selects\nthe post-read crypto context, not which block is fetched. Reading with\n@inode therefore returns the same (latest, not-yet-committed) COW data,\nwhile making both the fs-layer decryption decision and the inline crypto\npath use the correct (original inode's) key.\n\nWith the COW inode no longer used at the read site, the use_cow flag has no\nremaining consumer; drop it from f2fs_write_begin() and\nprepare_atomic_write_begin()."
}
],
"lastModified": "2026-08-17T05:17:18.187",
"sourceIdentifier": "416baaa9-dc9f-4396-8d5f-8c081fb06d67"
}