CVE-2026-64003

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> scsi: core: Run queues for all non-SDEV_DEL devices from scsi_run_host_queues<br /> <br /> While a SCSI host is in a recovery state, scsi_mq_requeue_cmd() will not<br /> set the requeue list for a requeued command to be kicked in the future.<br /> The expectation is a call to scsi_run_host_queues() will kick all SCSI<br /> devices once the recovery state is cleared.<br /> <br /> However, scsi_run_host_queues() uses shost_for_each_device() which uses<br /> scsi_device_get() and so will ignore devices in a partially removed<br /> state like SDEV_CANCEL. But these devices may also have requeued<br /> requests, leaving their requests stuck from not being kicked and causing<br /> the removal process of the device to hang.<br /> <br /> scsi_run_host_queues() needs to run against more devices than the macro<br /> shost_for_each_device() allows. Instead of using the too limiting<br /> scsi_device_get() state checks, only ignore devices in SDEV_DEL state or<br /> when unable to acquire a reference. Attempt to run the queues for all<br /> other devices when scsi_run_host_queues() is called.