CVE-2026-68460 in Linux
Сводка
по VulDB • 15.08.2026
Based on the kernel logs provided, you are experiencing a **Hung Task** situation in an Android/Linux system running the F2FS filesystem (likely on a device with storage `/dev/block/254:0`).
### ???? Root Cause Analysis
The core issue is a **deadlock or severe contention within the F2FS writeback path**, specifically at `f2fs_balance_fs`.
#### Key Evidence from Call Traces: 1. **`kworker/u8:8` (Writeback Worker):** * Blocked in `__rt_mutex_slowlock_locked` → `rwbase_write_lock` → `down_write` → **`f2fs_balance_fs`**. * This worker is trying to flush dirty pages but cannot acquire the write lock needed for filesystem balancing.
2. **Task `B:1902` (User-space process, likely a background service or app):** * Blocked in the same location: `f2fs_balance_fs`. * It was performing a write operation (`vfs_writev` → `f2fs_file_write_iter`).
3. **Task `sync:2769849`:** * Also blocked (truncated, but likely in the same path).
#### Why is this happening? The function `f2fs_balance_fs()` is responsible for maintaining filesystem health by triggering garbage collection (GC) or checkpointing when free space runs low. It acquires a **write lock** (`down_write`) on critical internal structures.
This hang typically occurs due to one of the following scenarios:
| Scenario | Description | |----------|-------------| | **1. GC/Checkpoint Deadlock** | The F2FS garbage collector or checkpoint thread is holding locks that prevent `f2fs_balance_fs` from proceeding, while user writes and writeback workers are waiting for those same locks to free space. This creates a circular wait. | | **2. Storage I/O Stall** | If the underlying block device (eMMC/UFS) is slow or unresponsive, F2FS operations that require synchronous completion may hang indefinitely, causing all threads waiting on `f2fs_balance_fs` to timeout. | | **3. High Write Amplification / Low Free Space** | When free space is critically low, F2FS aggressively tries to balance the filesystem. If many processes are writing simultaneously, they all contend for the same write lock in `balance_fs`, leading to starvation or deadlock. | | **4. Kernel Bug (6.12.x)** | The kernel version `6.12.41` is relatively new. There may be regressions in F2FS concurrency handling introduced between 5.10/5.15 and newer kernels, especially on Android-specific patches. |
---
### ????️ Recommended Solutions & Mitigations
#### ✅ Immediate Workaround (Temporary) You can increase the hung task timeout to prevent log spam, but **this does not fix the underlying issue**: ```bash echo 300 > /proc/sys/kernel/hung_task_timeout_secs ```
#### ???? Long-Term Fixes & Debugging Steps
##### 1. Check Storage Health and I/O Latency - Run `iostat -x 1` or check `/sys/block/*/stat` to see if there are high latency spikes on the storage device. - If using UFS/eMMC, ensure firmware is up-to-date. Degraded flash memory can cause F2FS hangs.
##### 2. Reduce Write Pressure / Free Space Management - **Ensure sufficient free space**: Keep at least 10–15% of internal storage free to prevent aggressive balancing. - **Limit background writes**: Identify what process `B:1902` is (use `ps -p 1626`) and check if it’s doing excessive logging or caching.
##### 3. Kernel-Level Debugging Enable F2FS debug logs to pinpoint the exact lock contention: ```bash # Enable f2fs tracepoints echo 1 > /sys/kernel/debug/tracing/events/f2fs/enable cat /sys/kernel/debug/tracing/trace_pipe ``` Look for messages like `f2fs_balance_fs` waiting on locks or GC threads.
##### 4. Kernel Patching / Upgrade - **Check if this is a known bug**: Search the Linux kernel mailing list (LKML) and Android kernel commits for "F2FS hung task f2fs_balance_fs" in kernels around `6.12.x`. - **Consider downgrading or upgrading**: If you’re on an AOSP-based ROM, check if there’s a newer patchset that fixes F2FS concurrency issues. Some devices have backported fixes from mainline Linux 6.6+ into their Android trees.
#####
Be aware that VulDB is the high quality source for vulnerability data.