- 20 Jun, 2017 31 commits
-
-
James Smart authored
When running nvme detach-ns /dev/nvme0n1 -n 1 command, the nvmet lpfc driver crashes with this stack dump: kernel BUG at /root/NVME/lpfc_8.4/lpfc_sli.c:1393! invalid opcode: 0000 [#1] SMP Workqueue: nvmet-fc-cpu0 nvmet_fc_do_work_on_cpu [nvmet_fc] lpfc_sli4_issue_wqe+0x357/0x440 [lpfc] lpfc_nvmet_xmt_fcp_abort+0x36b/0x5c0 [lpfc] nvmet_fc_abort_op+0x30/0x50 [nvmet_fc] nvmet_fc_do_work_on_cpu+0xd9/0x130 [nvmet_fc] process_one_work+0x14e/0x410 worker_thread+0x116/0x490 kthread+0xc7/0xe0 ret_from_fork+0x3f/0x70 Crash is due to an uninitialized iocbq->vport pointer. Explicitly set the iocbq->vport field to phba->pport in lpfc_nvmet_sol_fcp_issue_abort as it does all abort iocbq initialization in the routine. Using phba->pport is ok because target does not support NPIV instances. Signed-off-by:
Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by:
James Smart <james.smart@broadcom.com> Reviewed-by:
Hannes Reinecke <hare@suse.com> Signed-off-by:
Martin K. Petersen <martin.petersen@oracle.com>
-
James Smart authored
During every reset, IOCBs are allocated. So, at one point, number of allocated IOCBs reaches maximum limit and lpfc_sli_next_iotag fails. Allocate IOCBs only during initialization. Reuse them after every reset instead of allocating new set of IOCBs. Signed-off-by:
-
James Smart authored
OS crashes after the completion of firmware download. Failure in posting SCSI SGL buffers because number of SGL buffers is less than total count. Some of the pending IOs are not completed by driver. SGL buffers for these IOs are not added back to the list. Pending IOs are not completed because lpfc_wq_list list is initialized before completion of pending IOs. Postpone lpfc_wq_list reinitialization by moving lpfc_sli4_queue_destroy() after lpfc_hba_down_post(). Signed-off-by:
-
James Smart authored
In a server with an 8G adapter and a 32G adapter, running NVME and FCP, the server would crash with the following stack. RIP: 0010: ... lpfc_nvme_register_port+0x38/0x420 [lpfc] lpfc_nlp_state_cleanup+0x154/0x4f0 [lpfc] lpfc_nlp_set_state+0x9d/0x1a0 [lpfc] lpfc_cmpl_prli_prli_issue+0x35f/0x440 [lpfc] lpfc_disc_state_machine+0x78/0x1c0 [lpfc] lpfc_cmpl_els_prli+0x17c/0x1f0 [lpfc] lpfc_sli_sp_handle_rspiocb+0x39b/0x6b0 [lpfc] lpfc_sli_handle_slow_ring_event_s3+0x134/0x2d0 [lpfc] lpfc_work_done+0x8ac/0x13b0 [lpfc] lpfc_do_work+0xf1/0x1b0 [lpfc] Crash, on the 8G adapter, is due to a vport which does not have a nvme local port structure. It's not supposed to have one. NVME is not supported on the 8G adapter, so the NVME PRLI, which started this flow shouldn't have been sent in the first place. Correct discovery engine to recognize when on an SLI3 rport, which doesn't support SLI3, if the rport supports only NVME, don't send a NVME PRLI. Instead, as no FC4 will be used, a LOGO is sent. If rport is FCP and NVME, only execute the SCSI PRLI. Signed-off-by:
-
James Smart authored
Since unsol rcv ISR and command cmpl ISR both access/lock this list, separate get/put lists will reduce contention. Replaced struct list_head lpfc_nvmet_ctx_list; with struct list_head lpfc_nvmet_ctx_get_list; struct list_head lpfc_nvmet_ctx_put_list; and all correpsonding locks and counters. Signed-off-by:
-
James Smart authored
Removed unnecessary bzero of context area. Due to size of sg list, added a substantial delay and played havoc on cpu caches. Signed-off-by:
-
James Smart authored
Vport creation fails for SLI-3 adapters. Mailbox submission fails because mailbox interrupt is disabled. Mailbox interrupt is disabled during port reset. Do reset only for physical port. Signed-off-by:
-
James Smart authored
First line of nvme_info output is not consistent. There is an Extra colon in the format. First line of output will contain one of the following strings: NVME Initiator Enabled NVME Target Enabled NVME Disabled Signed-off-by:
-
James Smart authored
There is a null pointer dereference that can happen in the FOF interrupt handler. The driver was not setting up cq->assoc_qp_for sli4_hba->oas_cq. Initialize cq->assoc_qp before accessing it. Signed-off-by:
-
Xiang Chen authored
There is a change for abort dev for v3 hw: add registers to configure unaborted iptt for a device, and then inform this to logic. Signed-off-by:
Xiang Chen <chenxiang66@hisilicon.com> Signed-off-by:
John Garry <john.garry@huawei.com> Signed-off-by:
-
Xiang Chen authored
Add code to fill the interface of phy_hard_reset, phy_get_max_linkrate, and phy enable/disable. Signed-off-by:
-
Xiang Chen authored
Add code for interface get_wideport_bitmap. Signed-off-by:
-
Xiang Chen authored
Add code to prepare internal abort command. Signed-off-by:
-
Xiang Chen authored
Add code to itct setup and free for v3 hw. Signed-off-by:
-
Xiang Chen authored
Add code to prepare ATA frame for v3 hw Signed-off-by:
-
Xiang Chen authored
Add code to prepare SMP frame. Signed-off-by:
-
Xiang Chen authored
Add code to prepare SSP frame and deliver it to hardware. Signed-off-by:
-
Xiang Chen authored
Add v3 cq interrupt handler slot_complete_v3_hw(). Note: The slot error handling needs to be further refined in the future to examine all fields in the error record, and handle appropriately, instead of current solution - just report SAS_OPEN_REJECT. Signed-off-by:
-
Xiang Chen authored
Add code to initialise interrupts and add some interrupt handlers. Also add function hisi_sas_v3_destroy_irqs() to clean-up irqs upon module unloading. Signed-off-by:
-
Xiang Chen authored
Add code to configure PHYs for v3 hw. Signed-off-by:
-
Xiang Chen authored
Add code to initialise v3 hardware. Signed-off-by:
-
John Garry authored
Add the code to initialise the controller which is based on pci device in hisi_sas_v3_hw.c The core controller routines are still in hisi_sas_main.c; some common initialisation functions are also exported from hisi_sas_main.c For pci-based controller, the device properties, like phy count and sas address are read from the firmware, same as platform device-based controller. Signed-off-by:
-
John Garry authored
Add skeleton driver for v3 hw in hisi_sas_v3_hw.c File hisi_sas_v3_hw.c will serve 2 purposes: - probing and initialisation of the controller based on pci device - hw layer for v3-based controllers The controller design is quite similar to v2 hw in hip07. However key differences include: -All v2 hw bugs are fixed (hopefully), so workarounds are not required -support for device deregistration -some interrupt modifications -configurable max device support Signed-off-by:
-
John Garry authored
Move the functionality to retrieve the fw info into a dedicated device type-agnostic function, hisi_sas_get_fw_info(). The reasoning is that this function will be required for future pci-based platforms. Also add some debug logs for failure. Signed-off-by:
-
John Garry authored
Since hip08 SAS controller is based on pci device, add hisi_hba.pci_dev for hip08 (will be v3), and also rename hisi_hba.pdev to .platform_dev for clarity. In addition, for common code which wants to reference the controller device struct, add hisi_hba.dev, and change the common code to use it. Signed-off-by:
-
Xiang Chen authored
Relocate get_ncq_tag_v2_hw() to a common location, as future hw versions will require it. Also rename with "hisi_sas_" prefix for consistency. Signed-off-by:
-
Xiang Chen authored
Relocate get_ata_protocol() to a common location, as future hw versions will require it. Also rename with "hisi_sas_" prefix for consistency. Signed-off-by:
-
Xiang Chen authored
Relocate get_ata_protocol() to a common location, as future hw versions will require it. Also rename with "hisi_sas_" prefix for consistency. Signed-off-by:
-
Xiang Chen authored
Currently hisi_hba.lock is locked to deliver and receive a command to/from any hw queue. This causes much contention at high data-rates. To boost performance, lock on a per queue basis for sending and receiving commands to/from hw. Certain critical regions still need to be locked in the delivery and completion stages with hisi_hba.lock. New element hisi_sas_device.dq is added to store the delivery queue for a device, so it does not need to be needlessly re-calculated for every task. Signed-off-by:
-
John Garry authored
Currently hisi_sas_device.device_id is a u64. This can create a problem in selecting the queue for a device, in that this code does a 64b division on device id. For some 32b systems, 64b division is slow and the lib reference must be explicitly included. The device id does not need to be 64b in size, so, as a solution, just make as an int. Also, struct hisi_sas_device elements are re-ordered to improve packing efficiency. Signed-off-by:
-
Xiang Chen authored
We need to check for timeout before task status, or the task will be mistook as completed internal abort command. Also add protection for sas_task.task_state_flags in hisi_sas_tmf_timedout(). Signed-off-by:
-
- 13 Jun, 2017 9 commits
-
-
James Smart authored
Set lpfc driver revision to 11.4.0.0 Signed-off-by:
-
James Smart authored
Administrator intervention is currently required to get good numbers when switching from running latency tests to IOPS tests. The configured interrupt coalescing values will greatly effect the results of these tests. Currently, the driver has a single coalescing value set by values of the module attribute. This patch changes the driver to support auto-configuration of the coalescing value based on the total number of outstanding IOs and average number of CQEs processed per interrupt for an EQ. Values are checked every 5 seconds. The driver defaults to the automatic selection. Automatic selection can be disabled by the new lpfc_auto_imax module_parameter. Older hardware can only change interrupt coalescing by mailbox command. Newer hardware supports change via a register. The patch support both. Signed-off-by:
-
James Smart authored
Addressed the following reported defects: ** CID 1411552: Control flow issues (MISSING_BREAK) /drivers/scsi/lpfc/lpfc_sli.c: 13259 in lpfc_sli4_nvmet_handle_rcqe() ** CID 1411553: Memory - illegal accesses (OVERRUN) /drivers/scsi/lpfc/lpfc_sli.c: 16218 in lpfc_fc_frame_check() ** CID 1411553: Memory - illegal accesses (OVERRUN) Overrunning array "lpfc_rctl_names" of 202 8-byte elements at element index 244 (byte offset 1952) using index "fc_hdr->fh_r_ctl" (which evaluates to 244). ** CID 1411554: Null pointer dereferences (REVERSE_INULL) /drivers/scsi/lpfc/lpfc_nvmet.c: 2131 in lpfc_nvmet_unsol_fcp_abort_cmp() ** CID 1411555: Memory - illegal accesses (UNINIT) /drivers/scsi/lpfc/lpfc_nvmet.c: 180 in lpfc_nvmet_ctxbuf_post() Signed-off-by:
-
James Smart authored
vports cannot login to target. For vports, lpfc_nodelist is allocated for targets only on completion of GFF_ID command. Driver checks if lpfc_nodelist exists for target before sending GFF_ID. So, GFF_ID and PLOGI are not sent. As mentioned by the comment in lpfc_prep_node_fc4type() routine, do not send GFF_ID only if this NPortID is previously identified as FCP target. Send GFF_ID if it is a newly identified remote port from GID_FT response. Signed-off-by:
-
James Smart authored
The nvmet driver was rejecting the initiator's PRLI because its reg_rpi for the PLOGI was still outstanding. The initiator would resend the PRLI without delay and get the same answer. The PRLI retries would exhaust causing the nvme initiator to set the nvmet ndlp to UNMAPPED. The driver's lpfc_els_retry handler did not have a policy for an LS_RJT with explanation CMD_IN_PROGRESS for PRLI or NVME_PRLI. This caused the delay to remain at 0 but retry set 1. Fix: When the ELS response is LS_RJT, TPC and the command was PRLI or NVME_PRLI, just set the delay to 1000 mS to get a 1 second delay on the PRLI retry. This was enough to allow the REG_RPI to complete at the target. Signed-off-by:
-
James Smart authored
Kernel panic when log_verbose is set to 0xffffffff phba->pport is dereferenced before it is initialized Fix: Do not dereference phba->pport if it is NULL Signed-off-by:
-
James Smart authored
System panic with general protection fault during driver load The driver uses a static array sli4_hba.handler_name to store the irq handler names. If the io_channel_irqs exceeds the pre-allocated size (32+1), then the driver will overwrite other fields of sli4_hba. Fix: Dynamically allocate handler_name. Signed-off-by:
-
James Smart authored
Null pointer dereference when BFS VM is powered off The driver incorrectly uses sli3_ring on SLI-4 adapters Use the correct ring structure based on sli_rev Signed-off-by:
Raphael Silva <raphasil@linux.vnet.ibm.com> Signed-off-by:
-
James Smart authored
On hbacmd reset failure, observing wrong string "nline" in kernel log. On failure, non negative value (1) is returned from sysfs store routine. It is interpreted as count by kernel and store routine is called again with the remaining characters as input. Fix: Return negative error code (-EIO) in case of failure. Signed-off-by:
-
