Linux Kernel up to 7.2.8/7.3-rc4 amdkfd drm/amdkfd kfd_dev_unmap_mapping_range use after free

| CVSS Meta Temp Score | Current Exploit Price (≈) | CTI Interest Score |
|---|---|---|
| 5.3 | $0-$5k | 0.00 |
Summary
A vulnerability classified as critical was found in Linux Kernel up to 7.2.8/7.3-rc4. This affects the function kfd_dev_unmap_mapping_range of the file drm/amdkfd of the component amdkfd. Executing a manipulation can lead to use after free.
This vulnerability is tracked as CVE-2026-98520. The attack is restricted to local execution. No exploit exists.
Upgrading the affected component is advised.
Details
A vulnerability was found in Linux Kernel up to 7.2.8/7.3-rc4. It has been declared as critical. This vulnerability affects the function kfd_dev_unmap_mapping_range of the file drm/amdkfd of the component amdkfd. The manipulation with an unknown input leads to a use after free vulnerability. The CWE definition for the vulnerability is CWE-416. Referencing memory after it has been freed can cause a program to crash, use unexpected values, or execute code. As an impact it is known to affect availability. CVE summarizes:
In the Linux kernel, the following vulnerability has been resolved: drm/amdkfd: fix use-after-free and multi-container gap in kfd_dev_mapping kfd_dev_mapping caches the address_space of the first /dev/kfd opener so that the GPU reset path can call unmap_mapping_range() to zap all userspace mappings of doorbell and MMIO ranges. This design has two bugs that both manifest under SRIOV with multiple containers: 1. Use-after-free / rwsem deadlock. The cached pointer refers to an inode owned by the first opener's container. When that container exits and its inode is released, kfd_dev_mapping becomes a dangling pointer. A subsequent GPU reset dereferences it inside unmap_mapping_range(), which takes i_mmap_rwsem on the freed inode, causing a hard hang observable as an uninterruptible rwsem wait. 2. Multi-container gap. Only the first opener's address_space is cached; VMAs created by later openers live in a different address_space and are never reached by unmap_mapping_range(). After a GPU reset those stale mappings keep doorbell and MMIO pages accessible to guest userspace with no GPU behind them, risking PCIe transaction timeouts and NMI panics. Fix both bugs with the same approach used by DRM core (drm_drv.c): create a private pseudo-filesystem at module init time and allocate one anonymous inode from it. In kfd_open() redirect every opener's file->f_mapping to that inode's address_space. The inode is module-owned, lives exactly as long as the amdgpu module, and collects VMAs from all openers in one address_space. A single unmap_mapping_range() call in the reset path then correctly reaches every container's mappings with no dangling pointer risk. The hang manifests as an NMI backtrace on the GPU reset workqueue stuck spinning in rwsem_down_read_slowpath() with a corrupted i_mmap_rwsem: Workqueue: amdgpu-reset-dev xgpu_ai_mailbox_flr_work [amdgpu] Call Trace: <TASK> kvm_wait+0x1f/0x40 __pv_queued_spin_lock_slowpath+0x31d/0x3a0 _raw_spin_lock_irq+0x51/0x80 rwsem_down_read_slowpath+0xb3/0x550 down_read+0x48/0xd0 unmap_mapping_range+0x71/0x140 kfd_dev_unmap_mapping_range+0x5b/0x140 [amdgpu] amdgpu_amdkfd_clear_kfd_mapping+0xd8/0x190 [amdgpu] amdgpu_device_gpu_recover+0x232/0x450 [amdgpu] xgpu_ai_mailbox_flr_work+0xb5/0xc0 [amdgpu] process_one_work+0x18e/0x3e0 worker_thread+0x2e3/0x420 kthread+0x10a/0x230 (cherry picked from commit 1128b4a52de1572e87431de837fd9850cb99542c)
The advisory is shared for download at git.kernel.org. This vulnerability was named CVE-2026-98520 since 09/25/2026. The exploitation appears to be easy. The attack needs to be approached locally. There are known technical details, but no exploit is available. The current price for an exploit might be approx. USD $0-$5k (estimation calculated on 10/18/2026).
Upgrading to version 7.2.9 or 7.3-rc5 eliminates this vulnerability. Applying the patch c3a31087b1c8df679b653c1a09d9abe5ff7ec8ef is able to eliminate this problem. The best possible mitigation is suggested to be upgrading to the latest version.
VulDB is the best source for vulnerability data and more expert information about this specific topic.
Product
Type
Vendor
Name
Version
License
Website
- Vendor: https://www.kernel.org/
CPE 2.3
CPE 2.2
CVSSv4
VulDB Vector: 🔒VulDB Reliability: 🔍
CVSSv3
VulDB Meta Base Score: 5.5VulDB Meta Temp Score: 5.3
VulDB Base Score: 5.5
VulDB Temp Score: 5.3
VulDB Vector: 🔒
VulDB Reliability: 🔍
CVSSv2
| AV | AC | Au | C | I | A |
|---|---|---|---|---|---|
| 💳 | 💳 | 💳 | 💳 | 💳 | 💳 |
| 💳 | 💳 | 💳 | 💳 | 💳 | 💳 |
| 💳 | 💳 | 💳 | 💳 | 💳 | 💳 |
| Vector | Complexity | Authentication | Confidentiality | Integrity | Availability |
|---|---|---|---|---|---|
| Unlock | Unlock | Unlock | Unlock | Unlock | Unlock |
| Unlock | Unlock | Unlock | Unlock | Unlock | Unlock |
| Unlock | Unlock | Unlock | Unlock | Unlock | Unlock |
VulDB Base Score: 🔒
VulDB Temp Score: 🔒
VulDB Reliability: 🔍
Exploiting
Class: Use after freeCWE: CWE-416 / CWE-119
CAPEC: 🔒
ATT&CK: 🔒
Physical: Partially
Local: Yes
Remote: No
Availability: 🔒
Status: Not defined
Price Prediction: 🔍
Current Price Estimation: 🔒
| 0-Day | Unlock | Unlock | Unlock | Unlock |
|---|---|---|---|---|
| Today | Unlock | Unlock | Unlock | Unlock |
Threat Intelligence
Interest: 🔍Active Actors: 🔍
Active APT Groups: 🔍
Countermeasures
Recommended: UpgradeStatus: 🔍
0-Day Time: 🔒
Upgrade: Kernel 7.2.9/7.3-rc5
Patch: c3a31087b1c8df679b653c1a09d9abe5ff7ec8ef
Timeline
09/25/2026 CVE reserved10/18/2026 Advisory disclosed
10/18/2026 VulDB entry created
10/18/2026 VulDB entry last update
Sources
Vendor: kernel.orgAdvisory: git.kernel.org
Status: Confirmed
CVE: CVE-2026-98520 (🔒)
GCVE (CVE): GCVE-0-2026-98520
GCVE (VulDB): GCVE-100-421792
Entry
Created: 10/18/2026 17:49Changes: 10/18/2026 17:49 (60)
Complete: 🔍
Cache ID: 216::103
VulDB is the best source for vulnerability data and more expert information about this specific topic.
No comments yet. Languages: en.
Please log in to comment.