CVE-2025-68230
Unknown Unknown - Not Provided
BaseFortify

Publication date: 2025-12-16

Last updated on: 2025-12-18

Assigner: kernel.org

Description
In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu: fix gpu page fault after hibernation on PF passthrough On PF passthrough environment, after hibernate and then resume, coralgemm will cause gpu page fault. Mode1 reset happens during hibernate, but partition mode is not restored on resume, register mmCP_HYP_XCP_CTL and mmCP_PSP_XCP_CTL is not right after resume. When CP access the MQD BO, wrong stride size is used, this will cause out of bound access on the MQD BO, resulting page fault. The fix is to ensure gfx_v9_4_3_switch_compute_partition() is called when resume from a hibernation. KFD resume is called separately during a reset recovery or resume from suspend sequence. Hence it's not required to be called as part of partition switch. (cherry picked from commit 5d1b32cfe4a676fe552416cb5ae847b215463a1a)
CVSS Scores
EPSS Scores
Probability:
Percentile:
Meta Information
Published
2025-12-16
Last Modified
2025-12-18
Generated
2026-05-07
AI Q&A
2025-12-16
EPSS Evaluated
2026-05-05
NVD
Affected Vendors & Products
Showing 1 associated CPE
Vendor Product Version / Range
linux linux_kernel *
Helpful Resources
Exploitability
CWE
CWE Icon
KEV
KEV Icon
CWE ID Description
CWE-UNKNOWN
Attack-Flow Graph
AI Powered Q&A
Can you explain this vulnerability to me?

This vulnerability occurs in the Linux kernel's AMD GPU driver (drm/amdgpu) in a PF passthrough environment. After the system hibernates and then resumes, a GPU page fault happens because the partition mode is not properly restored. Specifically, certain registers (mmCP_HYP_XCP_CTL and mmCP_PSP_XCP_CTL) are incorrect after resume, causing the compute partition to use a wrong stride size when accessing the MQD buffer object. This out-of-bound access leads to a GPU page fault. The fix involves ensuring that the function gfx_v9_4_3_switch_compute_partition() is called upon resume from hibernation to restore the partition mode correctly.


How can this vulnerability impact me? :

This vulnerability can cause GPU page faults after resuming from hibernation in environments using PF passthrough. This may lead to system instability, crashes, or degraded GPU performance, especially in systems relying on AMD GPUs with this driver. It could disrupt workflows that depend on GPU compute tasks and potentially cause data loss or interruption of services that utilize GPU acceleration.


What immediate steps should I take to mitigate this vulnerability?

To mitigate this vulnerability, ensure that your Linux kernel is updated with the fix that calls gfx_v9_4_3_switch_compute_partition() when resuming from hibernation. This will prevent the GPU page fault caused by incorrect partition mode restoration after hibernation. Applying the relevant kernel patch or updating to a kernel version that includes this fix is recommended.


Ask Our AI Assistant
Need more information? Ask your question to get an AI reply (Powered by our expertise)
0/70
EPSS Chart