From: James Smart jsmart2021@gmail.com
[ Upstream commit e1364711359f3ced054bda9920477c8bf93b74c5 ]
In devloss timer handler and in backend calls to terminate remote port I/O, there is logic to walk through all active IOCBs and validate them to potentially trigger an abort request. This logic is causing illegal memory accesses which leads to a crash. Abort IOCBs, which may be on the list, do not have an associated lpfc_io_buf struct. The driver is trying to map an lpfc_io_buf struct on the IOCB and which results in a bogus address thus the issue.
Fix by skipping over ABORT IOCBs (CLOSE IOCBs are ABORTS that don't send ABTS) in the IOCB scan logic.
Link: https://lore.kernel.org/r/20210421234433.102079-1-jsmart2021@gmail.com Co-developed-by: Justin Tee justin.tee@broadcom.com Signed-off-by: Justin Tee justin.tee@broadcom.com Signed-off-by: James Smart jsmart2021@gmail.com Signed-off-by: Martin K. Petersen martin.petersen@oracle.com Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/scsi/lpfc/lpfc_sli.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-)
diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c index 95caad764fb7..bc5b5fb0b967 100644 --- a/drivers/scsi/lpfc/lpfc_sli.c +++ b/drivers/scsi/lpfc/lpfc_sli.c @@ -11803,13 +11803,20 @@ lpfc_sli_validate_fcp_iocb(struct lpfc_iocbq *iocbq, struct lpfc_vport *vport, lpfc_ctx_cmd ctx_cmd) { struct lpfc_io_buf *lpfc_cmd; + IOCB_t *icmd = NULL; int rc = 1;
if (iocbq->vport != vport) return rc;
- if (!(iocbq->iocb_flag & LPFC_IO_FCP) || - !(iocbq->iocb_flag & LPFC_IO_ON_TXCMPLQ)) + if (!(iocbq->iocb_flag & LPFC_IO_FCP) || + !(iocbq->iocb_flag & LPFC_IO_ON_TXCMPLQ) || + iocbq->iocb_flag & LPFC_DRIVER_ABORTED) + return rc; + + icmd = &iocbq->iocb; + if (icmd->ulpCommand == CMD_ABORT_XRI_CN || + icmd->ulpCommand == CMD_CLOSE_XRI_CN) return rc;
lpfc_cmd = container_of(iocbq, struct lpfc_io_buf, cur_iocbq);