Currently, we just assume that it will stick around by virtue of the submitter's reference, but later patches will allow the syscall to return early and we can't rely on that reference at that point. While I'm not aware of any reports of it, Xiubo pointed out that this may fix a use-after-free. If the wait for a reply times out or is canceled via signal, and then the reply comes in after the syscall returns, the client can end up trying to access r_parent without a reference. Take an extra reference to the inode when setting r_parent and release it when releasing the request.
Showing the changes to fs (1 of 1 file). Show all
Unified · Split
| @@ -708,8 +708,10 @@ void ceph_mdsc_release_request(struct kref *kref) · 707 unmodified lines | ||
| 708 | 708 | /* avoid calling iput_final() in mds dispatch threads */ |
| 709 | 709 | ceph_async_iput(req->r_inode); |
| 710 | 710 | } |
| 711 | − if (req->r_parent) | |
| 711 | + if (req->r_parent) { | |
| 712 | 712 | ceph_put_cap_refs(ceph_inode(req->r_parent), CEPH_CAP_PIN); |
| 713 | + ceph_async_iput(req->r_parent); | |
| 714 | + } | |
| 713 | 715 | ceph_async_iput(req->r_target_inode); |
| 714 | 716 | if (req->r_dentry) |
| 715 | 717 | dput(req->r_dentry); |
| @@ -2676,8 +2678,10 @@ int ceph_mdsc_submit_request(struct ceph_mds_client *mdsc, struct inode *dir, · 1960 unmodified lines | ||
| 2676 | 2678 | /* take CAP_PIN refs for r_inode, r_parent, r_old_dentry */ |
| 2677 | 2679 | if (req->r_inode) |
| 2678 | 2680 | ceph_get_cap_refs(ceph_inode(req->r_inode), CEPH_CAP_PIN); |
| 2679 | − if (req->r_parent) | |
| 2681 | + if (req->r_parent) { | |
| 2680 | 2682 | ceph_get_cap_refs(ceph_inode(req->r_parent), CEPH_CAP_PIN); |
| 2683 | + ihold(req->r_parent); | |
| 2684 | + } | |
| 2681 | 2685 | if (req->r_old_dentry_dir) |
| 2682 | 2686 | ceph_get_cap_refs(ceph_inode(req->r_old_dentry_dir), |
| 2683 | 2687 | CEPH_CAP_PIN); |