net: thunderbolt: Tear down DMA paths before stopping the rings
Summary
| CVE | CVE-2026-74691 |
|---|
| State | PUBLISHED |
|---|
| Assigner | Linux |
|---|
| Source Priority | CVE Program / NVD first with legacy fallback |
|---|
| Published | 2026-08-22 16:16:43 UTC |
|---|
| Updated | 2026-08-22 16:16:43 UTC |
|---|
| Description | In the Linux kernel, the following vulnerability has been resolved:
net: thunderbolt: Tear down DMA paths before stopping the rings
tbnet_tear_down() stops both rings and frees their frame buffers before
calling tb_xdomain_disable_paths(). tb_ring_stop() zeroes the ring's
descriptor base and tbnet_free_buffers() unmaps and frees the pages the
frames sit in, so by the time __tb_path_deactivate_hop() polls the hop's
'pending' bit, anything still in flight has nowhere to drain to.
The teardown sequence has been in this order since the driver was added.
The setup path has not: commit ff7cd07f3064 ("net: thunderbolt: Enable
DMA paths only after rings are enabled") moved the path enable to the end
of tbnet_connected_work() and documented why:
/* Both logins successful so enable the rings, high-speed DMA
* paths and start the network device queue.
*
* Note we enable the DMA paths last to make sure we have primed
* the Rx ring before any incoming packets are allowed to
* arrive.
*/
Teardown was never updated to match, so the rings and the paths now come
down in the same order they go up instead of in reverse.
On an ASMedia ASM4242 host router the 'pending' bit then never clears:
every teardown burns the full 500 ms timeout and
__tb_path_deactivate_hop() returns -ETIMEDOUT. Raising the timeout to
5 s does not help, so the hop is not slow to drain, it never drains
at all.
The failure is invisible above the thunderbolt core.
__tb_path_deactivate_hops() is void and only calls tb_port_warn();
tb_path_deactivate(), tb_tunnel_deactivate() and
__tb_disconnect_xdomain_paths() are void as well, and
tb_disconnect_xdomain_paths() ends in an unconditional "return 0". So
tb_xdomain_disable_paths() reports success and the netdev_warn() below
it never fires. Repeated teardowns eventually take the XDomain control
channel down, after which the peer node is gone and only a power cycle
brings the controller back.
Deactivating the paths first fixes it. Measured with kretprobes on a
stock v6.17 tree with no other patches applied, on a link that was up
and had just carried traffic:
before: __tb_path_deactivate_hop() returns 0 for the first hop, then
-ETIMEDOUT for the second 500335 us later
after: 0 for both, 525 us apart
Alternating the two orderings ABBA over three load levels, four
teardowns per arm: every teardown failed before the change (21 of 21
that ran), none failed after (0 of 24). The before arms ran short
because the link died partway through. The same split shows up when
the interface is enslaved to a bond instead of just brought down, which
is how I ran into this in the first place. Throughput and latency after
the change are unchanged.
Hosts whose routers drain the hop despite the stale descriptor base see
no functional difference, since the paths end up deactivated either way. |
|---|
Vendor Declared Affected Products
| Source | Vendor | Product | Version | Platforms |
|---|
| CNA |
Linux |
Linux |
affected e69b6c02b4c3b8d03be7136f90dd9551ad5a5a5e 7cce39109206bc5497e0953806563644b88bfc44 git |
Not specified |
| CNA |
Linux |
Linux |
affected e69b6c02b4c3b8d03be7136f90dd9551ad5a5a5e 0da9a6d27155ad072dd76db8cd637feead99a0e0 git |
Not specified |
| CNA |
Linux |
Linux |
affected e69b6c02b4c3b8d03be7136f90dd9551ad5a5a5e b5a21615f627c48dafaa6ef82a34a5b97a4352aa git |
Not specified |
| CNA |
Linux |
Linux |
affected e69b6c02b4c3b8d03be7136f90dd9551ad5a5a5e 103a9b663ac1cacb8465aeff18f84a247154a562 git |
Not specified |
| CNA |
Linux |
Linux |
affected e69b6c02b4c3b8d03be7136f90dd9551ad5a5a5e 4dd71cb0d23d40cb58fe4261c7bd183dca66caa0 git |
Not specified |
| CNA |
Linux |
Linux |
affected e69b6c02b4c3b8d03be7136f90dd9551ad5a5a5e 9a482b2b117e5fa656b6d24fc01799e8ac2d4368 git |
Not specified |
| CNA |
Linux |
Linux |
affected e69b6c02b4c3b8d03be7136f90dd9551ad5a5a5e 68bf02b6b4ad3f748c6db71fd77b6c0402d252f4 git |
Not specified |
| CNA |
Linux |
Linux |
affected 4.15 |
Not specified |
| CNA |
Linux |
Linux |
unaffected 4.15 semver |
Not specified |
| CNA |
Linux |
Linux |
unaffected 5.15.216 5.15.* semver |
Not specified |
| CNA |
Linux |
Linux |
unaffected 6.1.183 6.1.* semver |
Not specified |
| CNA |
Linux |
Linux |
unaffected 6.6.152 6.6.* semver |
Not specified |
| CNA |
Linux |
Linux |
unaffected 6.12.104 6.12.* semver |
Not specified |
| CNA |
Linux |
Linux |
unaffected 6.18.45 6.18.* semver |
Not specified |
| CNA |
Linux |
Linux |
unaffected 7.1.9 7.1.* semver |
Not specified |
| CNA |
Linux |
Linux |
unaffected 7.2 * original_commit_for_fix |
Not specified |
References
| Reference | Source | Link | Tags |
|---|
| git.kernel.org/stable/c/4dd71cb0d23d40cb58fe4261c7bd183dca66caa0 |
416baaa9-dc9f-4396-8d5f-8c081fb06d67 |
git.kernel.org |
|
| git.kernel.org/stable/c/7cce39109206bc5497e0953806563644b88bfc44 |
416baaa9-dc9f-4396-8d5f-8c081fb06d67 |
git.kernel.org |
|
| git.kernel.org/stable/c/9a482b2b117e5fa656b6d24fc01799e8ac2d4368 |
416baaa9-dc9f-4396-8d5f-8c081fb06d67 |
git.kernel.org |
|
| git.kernel.org/stable/c/68bf02b6b4ad3f748c6db71fd77b6c0402d252f4 |
416baaa9-dc9f-4396-8d5f-8c081fb06d67 |
git.kernel.org |
|
| git.kernel.org/stable/c/0da9a6d27155ad072dd76db8cd637feead99a0e0 |
416baaa9-dc9f-4396-8d5f-8c081fb06d67 |
git.kernel.org |
|
| git.kernel.org/stable/c/b5a21615f627c48dafaa6ef82a34a5b97a4352aa |
416baaa9-dc9f-4396-8d5f-8c081fb06d67 |
git.kernel.org |
|
| git.kernel.org/stable/c/103a9b663ac1cacb8465aeff18f84a247154a562 |
416baaa9-dc9f-4396-8d5f-8c081fb06d67 |
git.kernel.org |
|
| CVE Program record |
CVE.ORG |
www.cve.org |
canonical |
| NVD vulnerability detail |
NVD |
nvd.nist.gov |
canonical, analysis |
No vendor comments have been submitted for this CVE.
There are currently no legacy QID mappings associated with this CVE.