Hello,
My static analysis tool reports a possible ABBCCA deadlock in the lpfc
driver in Linux 5.10:
lpfc_els_flush_cmd()
spin_lock(&pring->ring_lock); --> line 8124 (Lock A)
lpfc_sli_cancel_iocbs()
lpfc_sli_release_iocbq()
spin_lock_irqsave(&phba->hbalock, iflags); --> line 1384 (Lock B)
lpfc_sli_abort_taskmgmt()
spin_lock_irqsave(&phba->hbalock, iflags); --> line 11812 (Lock B)
spin_lock(&lpfc_cmd->buf_lock); --> line 11830 (Lock C)
lpfc_abort_handler()
spin_lock(&lpfc_cmd->buf_lock); --> line 4760 (Lock C)
spin_lock(&pring_s4->ring_lock); --> line 4777 (Lock A)
When lpfc_els_flush_cmd(), lpfc_sli_abort_taskmgmt() and
lpfc_abort_handler() are concurrently executed, the deadlock can occur.
I am not quite sure whether this possible deadlock is real and how to
fix it if it is real.
Any feedback would be appreciated, thanks :)
Reported-by: TOTE Robot <oslab@xxxxxxxxxxxxxxx>
Best wishes,
Jia-Ju Bai