CVE-2026-23148: nvmet: fix race in nvmet_bio_done() leading to NULL pointer dereference
In the Linux kernel, the following vulnerability has been resolved:
nvmet: fix race in nvmet_bio_done() leading to NULL pointer dereference
There is a race condition in nvmet_bio_done() that can cause a NULL
pointer dereference in blk_cgroup_bio_start():
1. nvmet_bio_done() is called when a bio completes
2. nvmet_req_complete() is called, which invokes req->ops->queue_response(req)
3. The queue_response callback can re-queue and re-submit the same request
4. The re-submission reuses the same inline_bio from nvmet_req
5. Meanwhile, nvmet_req_bio_put() (called after nvmet_req_complete)
invokes bio_uninit() for inline_bio, which sets bio->bi_blkg to NULL
6. The re-submitted bio enters submit_bio_noacct_nocheck()
7. blk_cgroup_bio_start() dereferences bio->bi_blkg, causing a crash:
BUG: kernel NULL pointer dereference, address: 0000000000000028
#PF: supervisor read access in kernel mode
RIP: 0010:blk_cgroup_bio_start+0x10/0xd0
Call Trace:
submit_bio_noacct_nocheck+0x44/0x250
nvmet_bdev_execute_rw+0x254/0x370 [nvmet]
process_one_work+0x193/0x3c0
worker_thread+0x281/0x3a0
Fix this by reordering nvmet_bio_done() to call nvmet_req_bio_put()
BEFORE nvmet_req_complete(). This ensures the bio is cleaned up before
the request can be re-submitted, preventing the race condition.
Security readout for executives and security teams
Plain-English summary
A race condition in the Linux NVMe target subsystem can crash the kernel, disrupting storage services and dependent workloads. The supplied CVSS 3.1 score is 7.5 because the documented impact is availability loss, with no stated confidentiality or integrity impact.
Executive priority
Treat this as a high-priority availability issue for organizations using Linux NVMe targets. Patch internet-reachable, shared-storage, or business-critical target hosts first. Systems not using nvmet are less likely to be exposed, but version and configuration validation is necessary before lowering priority.
Technical view
In nvmet_bio_done(), response handling could requeue and resubmit a request before its inline bio was cleaned up. Subsequent cleanup set bi_blkg to NULL while the reused bio was entering submission, causing blk_cgroup_bio_start() to dereference NULL. The fix moves nvmet_req_bio_put() before nvmet_req_complete().
Likely exposure
Exposure is most likely on Linux systems operating the NVMe target subsystem, particularly where requests use the affected inline-bio lifecycle. The supplied version data is difficult to interpret reliably, so assess deployed kernel builds against the cited stable commits rather than relying only on version strings.
Exploitation context
The supplied vector describes network reachability, low complexity, no privileges, and no user interaction, with availability impact only. However, the bundle provides no exploit demonstration or operational attack details. The CVE is not listed as KEV, and the supplied evidence does not establish active exploitation.
Researcher notes
The failure is a use-order race involving request completion, callback-driven resubmission, inline_bio reuse, and later bio_uninit() cleanup. The supplied record identifies three stable-kernel commits, but its affected-version fields appear ambiguous. Confirm ancestry or vendor backport status before declaring a specific build vulnerable or fixed.
Mitigation direction
Apply a vendor-supported kernel release containing the cited nvmet fix.
Prioritize exposed or business-critical NVMe target hosts.
Check Linux distribution advisories for backported fixes and exact package versions.
Restrict NVMe target access to trusted initiators and networks where operationally feasible.
Validation and detection
Inventory systems running or configured for the Linux NVMe target subsystem.
Record kernel package versions, build identifiers, and distribution advisory status.
Confirm the deployed kernel includes an applicable cited stable fix.
Review kernel logs for nvmet-related NULL dereferences and unexpected storage-host crashes.
After updating, verify NVMe target workloads and failover behavior in controlled testing.
Generated from the cited source records. This long-tail analysis has not been individually reviewed by a named human.
Potential ATT&CK relevance
Conservative CVE-to-ATT&CK context
These mappings and lookup hints may be relevant to the vulnerability behavior, CWE, affected product, or exposure path. Glexia-inferred context is not an official MITRE, ATT&CK, CWE, or CVE Program mapping.
ATT&CK lookup starting points
Use these exact CWE pages and searches to review the Glexia ATT&CK library from this CVE's weakness and description context.
cve · low confidence lookup
CVE-2026-23148 mapping review
Open the CVE-to-ATT&CK bridge for reviewed, inferred, or future official mappings tied to this CVE.
These fields come from the CVE record and ADP containers, not from Glexia's Take. They preserve time-varying source decisions such as CISA SSVC, KEV status, CVSS metrics, and provider references.
1CVSS vectors
3Timeline events
0ADP providers
4Source links
CVSS vector scores
1 official score
We collect every scored CVSS vector available in the official CNA and ADP containers. When more than one version is present, the table keeps the source vectors side by side instead of collapsing them into the highest score.