GHSA-6RVQ-R5R4-HCV8
Vulnerability from github – Published: 2026-09-17 18:31 – Updated: 2026-09-17 18:31In the Linux kernel, the following vulnerability has been resolved:
net: page_pool: fix UAF in __page_pool_release_netmem_dma on xa_cmpxchg race
This bug was discovered while testing the hns3 driver under channel
reconfiguration (ethtool -L / ethtool -G) with iperf3 traffic on
arm64. The race is intermittently triggered when page_pool_destroy()
runs page_pool_scrub() concurrently with page return via
page_pool_put_netmem() on a different CPU. A WARN in
page_pool_clear_pp_info() surfaced the dangling DMA index bits left
by the cmpxchg loser, which led to the investigation.
page_pool_scrub() iterates pool->dma_mapped via xa_for_each() with no page ref held. __page_pool_release_netmem_dma() currently reads and writes netmem fields (dma_addr, DMA index bits in pp_magic) after xa_cmpxchg() returns. The unref path calls put_page() unconditionally regardless of the cmpxchg outcome; when it loses the cmpxchg, it still frees the page before the scrub winner finishes these netmem accesses, so scrub touches a freed page -- a Use-After-Free.
Fix this by splitting the DMA release into two functions:
-
__page_pool_unmap_netmem_dma() caches dma_addr before xa_cmpxchg(), does the cmpxchg to remove the DMA mapping, and calls dma_unmap on the cached address. It never touches netmem fields after the cmpxchg, making it safe for the scrub path which holds no page ref.
-
__page_pool_release_netmem_dma() wraps the above and additionally clears dma_addr and DMA index bits in netmem fields. This is safe only when the caller holds a page ref, so it is used by the return path (page_pool_return_netmem).
The scrub path calls __page_pool_unmap_netmem_dma() directly; the return path calls __page_pool_release_netmem_dma().
{
"affected": [],
"aliases": [
"CVE-2026-90201"
],
"database_specific": {
"cwe_ids": [],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2026-09-17T17:17:15Z",
"severity": null
},
"details": "In the Linux kernel, the following vulnerability has been resolved:\n\nnet: page_pool: fix UAF in __page_pool_release_netmem_dma on xa_cmpxchg race\n\nThis bug was discovered while testing the hns3 driver under channel\nreconfiguration (`ethtool -L` / `ethtool -G`) with iperf3 traffic on\narm64. The race is intermittently triggered when page_pool_destroy()\nruns page_pool_scrub() concurrently with page return via\npage_pool_put_netmem() on a different CPU. A WARN in\npage_pool_clear_pp_info() surfaced the dangling DMA index bits left\nby the cmpxchg loser, which led to the investigation.\n\npage_pool_scrub() iterates pool-\u003edma_mapped via xa_for_each() with no\npage ref held. __page_pool_release_netmem_dma() currently reads and\nwrites netmem fields (dma_addr, DMA index bits in pp_magic) after\nxa_cmpxchg() returns. The unref path calls put_page() unconditionally\nregardless of the cmpxchg outcome; when it loses the cmpxchg, it still\nfrees the page before the scrub winner finishes these netmem accesses,\nso scrub touches a freed page -- a Use-After-Free.\n\nFix this by splitting the DMA release into two functions:\n\n1. __page_pool_unmap_netmem_dma() caches dma_addr before xa_cmpxchg(),\n does the cmpxchg to remove the DMA mapping, and calls dma_unmap on\n the cached address. It never touches netmem fields after the cmpxchg,\n making it safe for the scrub path which holds no page ref.\n\n2. __page_pool_release_netmem_dma() wraps the above and additionally\n clears dma_addr and DMA index bits in netmem fields. This is safe\n only when the caller holds a page ref, so it is used by the return\n path (page_pool_return_netmem).\n\nThe scrub path calls __page_pool_unmap_netmem_dma() directly; the return\npath calls __page_pool_release_netmem_dma().",
"id": "GHSA-6rvq-r5r4-hcv8",
"modified": "2026-09-17T18:31:55Z",
"published": "2026-09-17T18:31:55Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-90201"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/24ef02f934eeb48830cff6b739abc3c62b1d107b"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/424a9fc4876cc7f28e9cb0aa8d92e920d726e350"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/9b65b0253ad5a66f73efee9a79a21a1e2b57cf59"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/bbfef303f980c1c078b8fa142e10a0e2fbcdd247"
}
],
"schema_version": "1.4.0",
"severity": []
}
Sightings
| Author | Source | Type | Date | Other |
|---|
Nomenclature
- Seen: The vulnerability was mentioned, discussed, or observed by the user.
- Confirmed: The vulnerability has been validated from an analyst's perspective.
- Published Proof of Concept: A public proof of concept is available for this vulnerability.
- Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
- Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
- Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
- Not confirmed: The user expressed doubt about the validity of the vulnerability.
- Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.
Browse all ATT&CK techniques and the vulnerabilities related to each.
Related by attack behaviour
Vulnerabilities whose description is nearest to this one in the vector space of the CIRCL/vulnerability-attack-technique-biencoder model. This is a similarity search over the bi-encoder space (plain cosine), not a classification, and it has no measured accuracy.