In the Linux kernel, the following vulnerability has been resolved:
scsi: qla2xxx: Clamp MSI-X derived queue counts to avoid truncation
ha->msix_count is u16, but ha->max_req_queues, ha->max_rsp_queues and
ha->max_qpairs are u8. Deriving the queue count as
"ha->max_req_queues = ha->msix_count - 1" therefore truncates: a board
(or a misconfigured/malicious hot-plugged device) advertising 257 MSI-X
vectors yields msix_count - 1 == 256, which truncates to 0. An MSI-X
count of 1 zeroes it as well, and in target mode the subsequent
"ha->max_req_queues--" then underflows 0 to 255.
When the count is 0, qla2x00_alloc_queues() calls
kzalloc_objs(struct req_que *, 0), which returns ZERO_SIZE_PTR. That is
not NULL, so the allocation check passes and the following
"ha->req_q_map[0] = req" dereferences ZERO_SIZE_PTR, corrupting memory
or crashing the kernel.
Add qla_calc_queue_count() to clamp the derived value into
[1, QLA_MAX_QUEUES - 1] so it always fits in u8 and is never zero, and
use it at all three derivation sites (qla25xx_iospace_config(),
qla83xx_iospace_config() and qla24xx_enable_msix()). Also guard the
target-mode decrement so it cannot reintroduce a zero (which would in
turn underflow max_qpairs).
CVSS Vector: CVSS:3.1/AV:L/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
CVSS Score: 8.4
AV:L - The flaw is in qla2xxx PCI probe (qla2x00_probe_one → iospace_config/qla24xx_enable_msix → qla2x00_alloc_queues). QLogic FC HBAs are local PCI devices in storage servers, blades, and SAN hosts; a board or compromised adapter advertising a truncating MSI-X count triggers it at bind. Not a network/FC-frame path; local PCI is higher severity than physical hotplug.
AC:L - The attacker-controlled PCI MSI-X vector count (config word or pci_msix_vec_count) is stored in u8 max_req_queues. Advertising 1 or 257+ vectors deterministically truncates that count to 0 and underflows max_qpairs to 255; no race or uninfluenced layout is required.
PR:N - Probe reads the MSI-X count from the adapter and autoloads qla2xxx on matching QLogic PCI IDs with no capability or credential check. A malicious, misconfigured, or firmware-compromised HBA needs no Linux account or init-namespace privilege.
UI:N - qla2x00_probe_one runs automatically on PCI enumeration at boot or hotplug bind; no victim must mount a filesystem, open a device node, or otherwise interact.
S:U - The ZERO_SIZE_PTR store and any resulting oops or privilege escalation remain in the host kernel qla2xxx probe path, the same security authority. This is not a VM escape or IOMMU/DMA boundary bypass.
C:H - qla2x00_alloc_queues() writes request/response queue pointers through ZERO_SIZE_PTR and can desynchronize u8 queue maps versus MSI-X arrays (0 versus underflowed 255 qpairs). That kernel memory corruption can be leveraged for information disclosure.
I:H - The wild write of kernel pointers through ZERO_SIZE_PTR, plus the u8 underflow that sizes queue-pair maps to 255 while the request map is a zero-size sentinel, is an out-of-bounds write exploitable for corruption and control-flow hijacking.
A:H - Dereferencing ZERO_SIZE_PTR is defined to take a distinct access fault, so a malicious MSI-X count causes a kernel oops or panic during probe.
| Attack Vector |
Local |
Scope |
Unchanged |
| Attack Complexity |
Low |
Confidentiality Impact |
High |
| Privileges Required |
None |
Integrity Impact |
High |
| User Interaction |
None |
Availability Impact |
High |
AV:L - The flaw is in qla2xxx PCI probe (qla2x00_probe_one → iospace_config/qla24xx_enable_msix → qla2x00_alloc_queues). QLogic FC HBAs are local PCI devices in storage servers, blades, and SAN hosts; a board or compromised adapter advertising a truncating MSI-X count triggers it at bind. Not a network/FC-frame path; local PCI is higher severity than physical hotplug.
AC:L - The attacker-controlled PCI MSI-X vector count (config word or pci_msix_vec_count) is stored in u8 max_req_queues. Advertising 1 or 257+ vectors deterministically truncates that count to 0 and underflows max_qpairs to 255; no race or uninfluenced layout is required.
PR:N - Probe reads the MSI-X count from the adapter and autoloads qla2xxx on matching QLogic PCI IDs with no capability or credential check. A malicious, misconfigured, or firmware-compromised HBA needs no Linux account or init-namespace privilege.
UI:N - qla2x00_probe_one runs automatically on PCI enumeration at boot or hotplug bind; no victim must mount a filesystem, open a device node, or otherwise interact.
S:U - The ZERO_SIZE_PTR store and any resulting oops or privilege escalation remain in the host kernel qla2xxx probe path, the same security authority. This is not a VM escape or IOMMU/DMA boundary bypass.
C:H - qla2x00_alloc_queues() writes request/response queue pointers through ZERO_SIZE_PTR and can desynchronize u8 queue maps versus MSI-X arrays (0 versus underflowed 255 qpairs). That kernel memory corruption can be leveraged for information disclosure.
I:H - The wild write of kernel pointers through ZERO_SIZE_PTR, plus the u8 underflow that sizes queue-pair maps to 255 while the request map is a zero-size sentinel, is an out-of-bounds write exploitable for corruption and control-flow hijacking.
A:H - Dereferencing ZERO_SIZE_PTR is defined to take a distinct access fault, so a malicious MSI-X count causes a kernel oops or panic during probe.
CVSS 3.1