GHSA-RMW9-5H46-94MM
Vulnerability from github – Published: 2026-06-09 15:32 – Updated: 2026-10-21 15:40In the Linux kernel, the following vulnerability has been resolved:
RDMA/rxe: Fix iova-to-va conversion for MR page sizes != PAGE_SIZE
The current implementation incorrectly handles memory regions (MRs) with page sizes different from the system PAGE_SIZE. The core issue is that rxe_set_page() is called with mr->page_size step increments, but the page_list stores individual struct page pointers, each representing PAGE_SIZE of memory.
ib_sg_to_page() has ensured that when i>=1 either a) SG[i-1].dma_end and SG[i].dma_addr are contiguous or b) SG[i-1].dma_end and SG[i].dma_addr are mr->page_size aligned.
This leads to incorrect iova-to-va conversion in scenarios:
1) page_size < PAGE_SIZE (e.g., MR: 4K, system: 64K): ibmr->iova = 0x181800 sg[0]: dma_addr=0x181800, len=0x800 sg[1]: dma_addr=0x173000, len=0x1000
Access iova = 0x181800 + 0x810 = 0x182010 Expected VA: 0x173010 (second SG, offset 0x10) Before fix: - index = (0x182010 >> 12) - (0x181800 >> 12) = 1 - page_offset = 0x182010 & 0xFFF = 0x10 - xarray[1] stores system page base 0x170000 - Resulting VA: 0x170000 + 0x10 = 0x170010 (wrong)
2) page_size > PAGE_SIZE (e.g., MR: 64K, system: 4K): ibmr->iova = 0x18f800 sg[0]: dma_addr=0x18f800, len=0x800 sg[1]: dma_addr=0x170000, len=0x1000
Access iova = 0x18f800 + 0x810 = 0x190010 Expected VA: 0x170010 (second SG, offset 0x10) Before fix: - index = (0x190010 >> 16) - (0x18f800 >> 16) = 1 - page_offset = 0x190010 & 0xFFFF = 0x10 - xarray[1] stores system page for dma_addr 0x170000 - Resulting VA: system page of 0x170000 + 0x10 = 0x170010 (wrong)
Yi Zhang reported a kernel panic[1] years ago related to this defect.
Solution: 1. Replace xarray with pre-allocated rxe_mr_page array for sequential indexing (all MR page indices are contiguous) 2. Each rxe_mr_page stores both struct page* and offset within the system page 3. Handle MR page_size != PAGE_SIZE relationships: - page_size > PAGE_SIZE: Split MR pages into multiple system pages - page_size <= PAGE_SIZE: Store offset within system page 4. Add boundary checks and compatibility validation
This ensures correct iova-to-va conversion regardless of MR page size and system PAGE_SIZE relationship, while improving performance through array-based sequential access.
Tests on 4K and 64K PAGE_SIZE hosts: - rdma-core/pytests $ ./build/bin/run_tests.py --dev eth0_rxe - blktest: $ TIMEOUT=30 QUICK_RUN=1 USE_RXE=1 NVMET_TRTYPES=rdma ./check nvme srp rnbd
[1] https://lore.kernel.org/all/CAHj4cs9XRqE25jyVw9rj9YugffLn5+f=1znaBEnu1usLOciD+g@mail.gmail.com/T/
{
"affected": [],
"aliases": [
"CVE-2026-46325"
],
"database_specific": {
"cwe_ids": [
"CWE-190"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2026-06-09T14:16:42Z",
"severity": "CRITICAL"
},
"details": "In the Linux kernel, the following vulnerability has been resolved:\n\nRDMA/rxe: Fix iova-to-va conversion for MR page sizes != PAGE_SIZE\n\nThe current implementation incorrectly handles memory regions (MRs) with\npage sizes different from the system PAGE_SIZE. The core issue is that\nrxe_set_page() is called with mr-\u003epage_size step increments, but the\npage_list stores individual struct page pointers, each representing\nPAGE_SIZE of memory.\n\nib_sg_to_page() has ensured that when i\u003e=1 either\na) SG[i-1].dma_end and SG[i].dma_addr are contiguous\nor\nb) SG[i-1].dma_end and SG[i].dma_addr are mr-\u003epage_size aligned.\n\nThis leads to incorrect iova-to-va conversion in scenarios:\n\n1) page_size \u003c PAGE_SIZE (e.g., MR: 4K, system: 64K):\n ibmr-\u003eiova = 0x181800\n sg[0]: dma_addr=0x181800, len=0x800\n sg[1]: dma_addr=0x173000, len=0x1000\n\n Access iova = 0x181800 + 0x810 = 0x182010\n Expected VA: 0x173010 (second SG, offset 0x10)\n Before fix:\n - index = (0x182010 \u003e\u003e 12) - (0x181800 \u003e\u003e 12) = 1\n - page_offset = 0x182010 \u0026 0xFFF = 0x10\n - xarray[1] stores system page base 0x170000\n - Resulting VA: 0x170000 + 0x10 = 0x170010 (wrong)\n\n2) page_size \u003e PAGE_SIZE (e.g., MR: 64K, system: 4K):\n ibmr-\u003eiova = 0x18f800\n sg[0]: dma_addr=0x18f800, len=0x800\n sg[1]: dma_addr=0x170000, len=0x1000\n\n Access iova = 0x18f800 + 0x810 = 0x190010\n Expected VA: 0x170010 (second SG, offset 0x10)\n Before fix:\n - index = (0x190010 \u003e\u003e 16) - (0x18f800 \u003e\u003e 16) = 1\n - page_offset = 0x190010 \u0026 0xFFFF = 0x10\n - xarray[1] stores system page for dma_addr 0x170000\n - Resulting VA: system page of 0x170000 + 0x10 = 0x170010 (wrong)\n\nYi Zhang reported a kernel panic[1] years ago related to this defect.\n\nSolution:\n1. Replace xarray with pre-allocated rxe_mr_page array for sequential\n indexing (all MR page indices are contiguous)\n2. Each rxe_mr_page stores both struct page* and offset within the\n system page\n3. Handle MR page_size != PAGE_SIZE relationships:\n - page_size \u003e PAGE_SIZE: Split MR pages into multiple system pages\n - page_size \u003c= PAGE_SIZE: Store offset within system page\n4. Add boundary checks and compatibility validation\n\nThis ensures correct iova-to-va conversion regardless of MR page size\nand system PAGE_SIZE relationship, while improving performance through\narray-based sequential access.\n\nTests on 4K and 64K PAGE_SIZE hosts:\n- rdma-core/pytests\n $ ./build/bin/run_tests.py --dev eth0_rxe\n- blktest:\n $ TIMEOUT=30 QUICK_RUN=1 USE_RXE=1 NVMET_TRTYPES=rdma ./check nvme srp rnbd\n\n[1] https://lore.kernel.org/all/CAHj4cs9XRqE25jyVw9rj9YugffLn5+f=1znaBEnu1usLOciD+g@mail.gmail.com/T/",
"id": "GHSA-rmw9-5h46-94mm",
"modified": "2026-10-21T15:40:33Z",
"published": "2026-06-09T15:32:18Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-46325"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/12985e5915a0b8354796efadaaeb201eed115377"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/2a585a08a43c568ed6e9f27a97fd3d95106b9e68"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/409c2c5508f3d30627bea576f8676de523cb906e"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/836f6c13c9674027793f720be3f15ecd2b90b6ca"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/e327021bd73b134c18167e2cd2a52eb5597fcba6"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H",
"type": "CVSS_V3"
}
]
}
Drag nodes to arrange them. Double-click a related vulnerability to open it, or a source to visit its link. The Physics button on the left tunes the layout and the simulation; the legend at the bottom left hides or shows nodes by kind and by sighting type.
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.
Vulnerability timeline
Sightings and recorded dates, grouped by day · UTC
Loading timeline…
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.