High · published September 15, 2025
In the Linux kernel, the following vulnerability has been resolved:
io_uring/af_unix: defer registered files gc to io_uring release
Instead of putting io_uring's registered files in unix_gc() we want it
to be done by io_uring itself. The trick here is to consider io_uring
registered files for cycle detection but not actually putting them down.
Because io_uring can't register other ring instances, this will remove
all refs to the ring file triggering the ->release path and clean up
with io_ring_ctx_free().
[axboe: add kerneldoc comment to skb, fold in skb leak fix]
| Product | Versions | Fixed in |
|---|---|---|
| linux/linux_kernel | ≥ 5.1, < 5.4.220 | 5.4.220 |
| linux/linux_kernel | ≥ 5.5, < 5.10.150 | 5.10.150 |
| linux/linux_kernel | ≥ 5.11, < 5.15.75 | 5.15.75 |
| linux/linux_kernel | ≥ 5.16, < 5.19.17 | 5.19.17 |
| linux/linux_kernel | ≥ 6.0, < 6.0.3 | 6.0.3 |