lpfc 8.3.37: Provide support for FCoE protocol dual-chute (ULP) operation
Hello James Smart,
The patch 962bc51b04b2: "[SCSI] lpfc 8.3.37: Provide support for FCoE protocol dual-chute (ULP) operation" from Jan 3, 2013, leads to the following warning:
"drivers/scsi/lpfc/lpfc_sli.c:12818 lpfc_wq_create()
warn: 0x800 is larger than 8 bits"
[ This is from a not ready for release Smatch check. ]
drivers/scsi/lpfc/lpfc_sli.c
12818 if (phba->sli4_hba.fw_func_mode & LPFC_DUA_MODE)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
fw_func_mode is unsigned char.LPFC_DUA_MODE is 0x800.
This condition is always false because 0xff & 0x800 is zero.
12819 bf_set(lpfc_mbx_wq_create_dua, &wq_create->u.request, 1);
12820
12821 rc = lpfc_sli_issue_mbox(phba, mbox, MBX_POLL);
12822 /* The IOCTL status is embedded in the mailbox subheader. */
12823 shdr_status = bf_get(lpfc_mbox_hdr_status, &shdr->response);
12824 shdr_add_status = bf_get(lpfc_mbox_hdr_add_status, &shdr->response);
12825 if (shdr_status || shdr_add_status || rc) {
12826 lpfc_printf_log(phba, KERN_ERR, LOG_INIT,
12827 "2503 WQ_CREATE mailbox failed with "
12828 "status x%x add_status x%x, mbx status x%x\n",
12829 shdr_status, shdr_add_status, rc);
12830 status = -ENXIO;
12831 goto out;
12832 }
12833 wq->queue_id = bf_get(lpfc_mbx_wq_create_q_id, &wq_create->u.response);
12834 if (wq->queue_id == 0xFFFF) {
12835 status = -ENXIO;
12836 goto out;
12837 }
12838 if (phba->sli4_hba.fw_func_mode & LPFC_DUA_MODE) {
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^Same.
12839 wq->db_format = bf_get(lpfc_mbx_wq_create_db_format,
12840 &wq_create->u.response);
12841 if ((wq->db_format != LPFC_DB_LIST_FORMAT) &&
regards,
dan carpenter
Dan,
Thanks for pointing this out. I'm addressing it as the first patch in 8.3.38, which I'll be pushing shortly.
-- james s
Quoted 48 lines · Dan Carpenter
Hello James Smart,
The patch
962bc51b04b2: "[SCSI] lpfc 8.3.37: Provide support for FCoE protocol dual-chute (ULP) operation" from Jan 3, 2013, leads to the following warning:
"drivers/scsi/lpfc/lpfc_sli.c:12818lpfc_wq_create()warn: 0x800 is larger than 8 bits"[ This is from a not ready for release Smatch check. ]
drivers/scsi/lpfc/lpfc_sli.c
12818 if (phba->sli4_hba.fw_func_mode&LPFC_DUA_MODE)^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
fw_func_modeis unsigned char.LPFC_DUA_MODEis0x800.
This condition is always false because0xff&0x800is zero.12819
bf_set(lpfc_mbx_wq_create_dua, &wq_create->u.request, 1);
1282012821 rc = lpfc_sli_issue_mbox(phba, mbox, MBX_POLL); 12822 /* The IOCTL status is embedded in the mailbox subheader. */ 12823 shdr_status = bf_get(lpfc_mbox_hdr_status, &shdr->response); 12824 shdr_add_status = bf_get(lpfc_mbox_hdr_add_status, &shdr->response); 12825 if (shdr_status || shdr_add_status || rc) { 12826 lpfc_printf_log(phba, KERN_ERR, LOG_INIT, 12827 "2503 WQ_CREATE mailbox failed with " 12828 "status x%x add_status x%x, mbx status x%x\n", 12829 shdr_status, shdr_add_status, rc); 12830 status = -ENXIO; 12831 goto out; 12832 } 12833 wq->queue_id = bf_get(lpfc_mbx_wq_create_q_id, &wq_create->u.response); 12834 if (wq->queue_id == 0xFFFF) { 12835 status = -ENXIO; 12836 goto out; 12837 } 12838 if (phba->sli4_hba.fw_func_mode & LPFC_DUA_MODE) { ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^Same.
12839
wq->db_format= bf_get(lpfc_mbx_wq_create_db_format,
12840&wq_create->u.response);
12841 if ((wq->db_format!=LPFC_DB_LIST_FORMAT) &®ards,
dan carpenter