CVE-2021-47143

What it actually does

Missing error handling after a failed device_add() call leaves a smcd_dev entry in smcd_dev_list; the subsequent kfree creates a dangling list node and corrupts the linked list.

Why it matters to the business

Systems may experience kernel crashes, memory corruption, or unpredictable behavior during SMC device initialization failures, leading to unplanned downtime and potential data loss on affected kernels.

How an attack unfolds

An attacker with local access triggers repeated SMC device registration attempts that fail at device_add(), leaving stale list entries. Later operations that walk or free the list dereference invalid memory, enabling denial-of-service or, with additional primitives, kernel code execution.

How to fix it

Upgrade the affected kernel packages to a version that contains the upstream fix for the list-cleanup logic. On AlmaLinux 8 systems apply the module versions listed for rust-toolset, cargo, clippy, rls, rust, rust-analysis, rust-debugger-common, rust-doc, rust-gdb, rust-lldb, rust-src, rust-std-static, rust-std-static-wasm32-unknown-unknown, and rustfmt (all fixed in the 1.54.0-3.module_el8.5.0+2599+d655d86c or equivalent 1.54.0-1 builds). Reboot after the update.

How to confirm the fix worked

Confirm the running kernel or package versions match or exceed the fixed module releases; inspect /proc/version or rpm -q output for the corrected build strings and verify the smc module no longer contains the vulnerable list-add path.

Typical attack path

1. Obtain local access on a system using the SMC protocol stack
2. Force repeated device_add() failures for smcd devices
3. Trigger list traversal or cleanup paths that encounter the dangling entry
4. Achieve memory corruption or kernel panic

How to detect it

Review kernel logs for SMC list-related warnings or use-after-free oops around device registration failures; no public exploit artifacts are known.

What your security team should do

• Prioritize patching systems that expose the SMC stack or run workloads that register many virtual devices
• Apply the listed AlmaLinux module updates within standard change windows
• Monitor for kernel oops or list-corruption traces in dmesg after deployment