CVE-2026-46325

Aliases:UBUNTU-CVE-2026-46325DEBIAN-CVE-2026-46325CGA-675j-m83h-2w25CGA-798q-894c-r49mCGA-c8v5-pq86-xv29CGA-j6pm-q5q9-qr6hCGA-pg3x-p843-8cwr
Advisory lineage Upstream: 0 Downstream: 10
Analyzed
Published: 09 Jun 2026, 12:25
Last modified:05 Aug 2026, 12:31

Vulnerability Summary

Overall Risk (default)
high
70/100
CVSS Score
9.8 CRITICAL
v3.1 (cve.org)
EPSS Score
0.35% LOW
0% probability 0.00%
KEV
Not listed
Ransomware
No reports
Public exploits
None found
Dark Web
Not detected

Timeline

09 Jun 2026, 12:25
Published
Vulnerability first disclosed
05 Aug 2026, 12:31
Last Modified
Vulnerability information updated

Description

In 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/

CVSS Metrics

  • v3.1•CRITICAL•Score: 9.8CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H

EPSS Trends

Current EPSS score: 0.35%• Percentile: 28%

Affected Systems

  • chainguard•linux-aws-6.12

    < 0

  • chainguard•linux-azure-6.12

    all

  • chainguard•linux-gcp-6.12

    all

  • chainguard•linux-qemu-6.12

    all

  • chainguard•linux-vmware-6.12

    < 0

  • debian•linux

    all | < 6.18.14-1

  • ubuntu•linux

    < 6.8.0-137.137 | < 6.17.0-40.40

  • ubuntu•linux-allwinner-5.19

    all

  • ubuntu•linux-aws

    < 6.8.0-1062.65 | < 6.17.0-1019.19

  • ubuntu•linux-aws-5.0

    all

  • ubuntu•linux-aws-5.11

    all

  • ubuntu•linux-aws-5.13

    all

  • ubuntu•linux-aws-5.19

    all

  • ubuntu•linux-aws-5.3

    all

  • ubuntu•linux-aws-5.8

    all

  • ubuntu•linux-aws-6.14

    all

  • ubuntu•linux-aws-6.17

    < 6.17.0-1019.19~24.04.1

  • ubuntu•linux-aws-6.2

    all

  • ubuntu•linux-aws-6.5

    all

  • ubuntu•linux-aws-6.8

    all | < 6.8.0-1062.65~22.04.2

  • ubuntu•linux-aws-fips

    < 6.8.0-1062.65+fips1

  • ubuntu•linux-azure

    all | < 6.8.0-1064.72 | all

  • ubuntu•linux-azure-5.11

    all

  • ubuntu•linux-azure-5.13

    all

  • ubuntu•linux-azure-5.19

    all

  • ubuntu•linux-azure-5.3

    all

  • ubuntu•linux-azure-5.8

    all

  • ubuntu•linux-azure-6.11

    all

  • ubuntu•linux-azure-6.14

    all

  • ubuntu•linux-azure-6.17

    < 6.17.0-1021.21~24.04.1

  • ubuntu•linux-azure-6.2

    all

  • ubuntu•linux-azure-6.5

    all

  • ubuntu•linux-azure-6.8

    < 6.8.0-1064.72~22.04.1

  • ubuntu•linux-azure-edge

    all

  • ubuntu•linux-azure-fde

    all | < 6.8.0-1063.70 | all | all

  • ubuntu•linux-azure-fde-5.19

    all

  • ubuntu•linux-azure-fde-6.14

    all

  • ubuntu•linux-azure-fde-6.17

    < 6.17.0-1018.18~24.04.1

  • ubuntu•linux-azure-fde-6.2

    all

  • ubuntu•linux-azure-fde-6.8

    < 6.8.0-1063.70~22.04.1

  • ubuntu•linux-azure-fde-7.0

    all

  • ubuntu•linux-azure-fips

    < 6.8.0-1064.72+fips1

  • ubuntu•linux-azure-nvidia

    all

  • ubuntu•linux-azure-nvidia-6.14

    all

  • ubuntu•linux-bluefield

    all | < 6.8.0-1028.33

  • ubuntu•linux-fips

    < 6.8.0-137.137+fips1

  • ubuntu•linux-gcp

    all | < 6.8.0-1065.73 | < 6.17.0-1020.22

  • ubuntu•linux-gcp-5.11

    all

  • ubuntu•linux-gcp-5.13

    all

  • ubuntu•linux-gcp-5.19

    all

Showing first 50 affected entries in server-rendered view.

References (24)