- 28 Mar, 2017 7 commits
-
-
Jitendra Bhivare authored
Currently, ASYNC PDU default queue size is set to max connections. This leaves only one buffer per connection for any ASYNC PDUs from targets. Double the size of the default queue. Signed-off-by:
Jitendra Bhivare <jitendra.bhivare@broadcom.com> Reviewed-by:
Tomas Henzl <thenzl@redhat.com> Reviewed-by:
Chris Leech <cleech@redhat.com> Signed-off-by:
Martin K. Petersen <martin.petersen@oracle.com>
-
Jitendra Bhivare authored
scsi_flush_work flushes workqueue for the Scsi_Host. In iSCSI offload enabled host, this would wait for all other sessions under the host. Use flush_work for the session being removed instead. Signed-off-by:
-
Jitendra Bhivare authored
spin_unlock_bh back_lock is used in beiscsi_eh_device_reset instead of spin_lock. Signed-off-by:
-
Jitendra Bhivare authored
CID needs to be freed even when invalidate or upload connection fails. Attempt to close connection 3 times before freeing CID. Set cleanup_type to INVALIDATE instead of force TCP_RST. This unnecessarily is terminating connection with reset instead of gracefully closing it. Set save_cfg to 0 - session not to be saved on flash. Add delay and process CQ before uploading connection. Signed-off-by:
-
Jitendra Bhivare authored
scsi host12: BS_1377 : mgmt_invalidate_connection Failed for cid=256 BUG: unable to handle kernel NULL pointer dereference at 0000000000000008 IP: [<ffffffff81332ebf>] __list_add+0xf/0xc0 PGD 0 Oops: 0000 [#1] SMP Modules linked in: ... CPU: 9 PID: 1542 Comm: iscsid Tainted: G ------------ T 3.10.0-514.el7.x86_64 #1 Hardware name: HP ProLiant DL360 Gen9/ProLiant DL360 Gen9, BIOS P89 09/12/2016 task: ffff88076f310fb0 ti: ffff88076bba8000 task.ti: ffff88076bba8000 RIP: 0010:[<ffffffff81332ebf>] [<ffffffff81332ebf>] __list_add+0xf/0xc0 RSP: 0018:ffff88076bbab8e8 EFLAGS: 00010046 RAX: 0000000000000246 RBX: ffff88076bbab990 RCX: 0000000000000000 RDX: 0000000000000000 RSI: ffff880468badf58 RDI: ffff88076bbab990 RBP: ffff88076bbab900 R08: 0000000000000246 R09: 00000000000020de R10: 0000000000000000 R11: ffff88076bbab5be R12: 0000000000000000 R13: ffff880468badf58 R14: 000000000001adb0 R15: ffff88076f310fb0 FS: 00007f377124a880(0000) GS:ffff88046fa40000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000000008 CR3: 0000000771318000 CR4: 00000000001407e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 Stack: ffff88076bbab990 ffff880468badf50 0000000000000001 ffff88076bbab938 ffffffff810b128b 0000000000000246 00000000cf9b7040 ffff880468bac7a0 0000000000000000 ffff880468bac7a0 ffff88076bbab9d0 ffffffffa05a6ea3 Call Trace: [<ffffffff810b128b>] prepare_to_wait+0x7b/0x90 [<ffffffffa05a6ea3>] beiscsi_mccq_compl_wait+0x153/0x330 [be2iscsi] [<ffffffff810b1600>] ? wake_up_atomic_t+0x30/0x30 [<ffffffffa05981b1>] beiscsi_ep_disconnect+0x91/0x2d0 [be2iscsi] [<ffffffffa0202ffa>] iscsi_if_ep_disconnect.isra.14+0x5a/0x70 [scsi_transport_iscsi] [<ffffffffa02042fb>] iscsi_if_recv_msg+0x113b/0x14a0 [scsi_transport_iscsi] [<ffffffff811dffd8>] ? __kmalloc_node_track_caller+0x58/0x290 [<ffffffffa02046ee>] iscsi_if_rx+0x8e/0x1f0 [scsi_transport_iscsi] [<ffffffff815a351d>] netlink_unicast+0xed/0x1b0 [<ffffffff815a38fe>] netlink_sendmsg+0x31e/0x690 [<ffffffff815a03e4>] ? netlink_rcv_wake+0x44/0x60 [<ffffffff815a19e3>] ? netlink_recvmsg+0x1e3/0x450 beiscsi_mccq_compl_wait gets called even when MCC tag allocation failed for mgmt_invalidate_connection. mcc_wait is not initialized for tag 0 so causes crash in prepare_to_wait. Signed-off-by:
-
Tomohiro Kusumi authored
These aren't really falling through to anywhere meaningful. Signed-off-by:
Tomohiro Kusumi <tkusumi@tuxera.com> Reviewed-by:
Subhash Jadavani <subhashj@codeaurora.org> Signed-off-by:
-
Dan Carpenter authored
We don't call the remove() function unless probe() succeeds so "oud" can't be NULL here. Plus, if it were NULL, we dereference it on the next line so it would crash anyway. [mkp: applied by hand] Signed-off-by:
Dan Carpenter <dan.carpenter@oracle.com> Acked-by:
Boaz Harrosh <ooo@electrozaur.com> Signed-off-by:
-
- 23 Mar, 2017 33 commits
-
-
Brian King authored
Bump driver version Signed-off-by:
Brian King <brking@linux.vnet.ibm.com> Reviewed-by:
Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by:
-
Brian King authored
This patch fixes a hang that can occur in ATA EH with ipr. With ipr's usage of libata, commands should never end up on ap->eh_done_q. The timeout function we use for ipr, even for SATA devices, is scsi_times_out, so ATA_QCFLAG_EH_SCHEDULED never gets set for ipr and EH is driven completely by ipr and SCSI. The SCSI EH thread ends up calling ipr's eh_device_reset_handler, which then calls ata_std_error_handler. This ends up calling ipr_sata_reset, which issues a reset to the device. This should result in all pending commands getting failed back and having ata_qc_complete called for them, which should end up clearing ATA_QCFLAG_FAILED as qc->flags gets zeroed in ata_qc_free. This ensures that when we end up in ata_eh_finish, we don't do anything more with the command. On adapters that only support a single interrupt and when running with two MSI-X vectors or less, the adapter firmware guarantees that responses to all outstanding commands are sent back prior to sending the response to the SATA reset command. On newer adapters supporting multiple HRRQs, however, this can no longer be guaranteed, since the command responses and reset response may be processed on different HRRQs. If ipr returns from ipr_sata_reset before the outstanding command was returned, this sends us down the path of __ata_eh_qc_complete which then moves the associated scsi_cmd from the work_q in scsi_eh_bus_device_reset to ap->eh_done_q, which then will sit there forever and we will be wedged. This patch fixes this up by ensuring that any outstanding commands are flushed before returning from eh_device_reset_handler for a SATA device. Reported-by:
David Jeffery <djeffery@redhat.com> Signed-off-by:
-
Brian King authored
This patch closes up some potential race conditions observed in the error handling paths in ipr while debugging an issue resulting in a hang with SATA error handling. These patches ensure we are holding the correct lock when adding and removing commands from the free and pending queues in some error scenarios. Signed-off-by:
-
Brian King authored
This fixes a race condition in the error handlomg paths of ipr. While a command is outstanding to the adapter, it is placed on a pending queue for the hrrq it is associated with, while holding the HRRQ lock. When a command is completed, it is removed from the pending queue, under HRRQ lock, and placed on a local list. This list is then iterated through without any locks and each command's done function is invoked, inside of which, the command gets returned to the free list while grabbing the HRRQ lock. This fixes two race conditions when commands have been removed from the pending list but have not yet been added to the free list. Both of these changes fix race conditions that could result in returning success from eh_abort_handler and then later calling scsi_done for the same request. The first race condition is in ipr_cancel_op. It looks through each pending queue to see if the command to be aborted is still outstanding or not. Rather than looking on the pending queue, reverse the logic to check to look for commands that are NOT on the free queue. The second race condition can occur when in ipr_wait_for_ops where we are waiting for responses for commands we've aborted. Signed-off-by:
-
Brian King authored
Removes some code in __ipr_eh_dev_reset which was modifying the ipr_cmd done function. This should have already been setup at command allocation time and if its since been changed, it means we are in the ipr_erp* functions and need to wait for them to complete and don't want to override that here. Signed-off-by:
-
Brian King authored
Following a command abort or device reset, ipr's EH handlers wait for the commands getting aborted to get sent back from the adapter prior to returning from the EH handler. This fixes up some cases where the completion handler was not getting called, which would have resulted in the EH thread waiting until it timed out, greatly extending EH time. Signed-off-by:
-
Xiaofei Tan authored
Add helper function is_sata_phy_v2_hw() to judge whether the attached device is SATA disk for a root PHY. Signed-off-by:
Xiaofei Tan <tanxiaofei@huawei.com> Signed-off-by:
John Garry <john.garry@huawei.com> Signed-off-by:
-
Xiang Chen authored
When SMP IO is sent, sas_protocol_ata couldn't judge whether the disk is SATA or SAS disk. So use dev_is_sata to identify SATA or SAS disk. Signed-off-by:
Xiang Chen <chenxiang66@hisilicon.com> Signed-off-by:
-
John Garry authored
Unless we actually get some sort of failure in hisi_sas_lu_reset(), don't print a message. Signed-off-by:
-
Xiang Chen authored
When an SMP task timeouts, it will call lldd_abort_task to release the associated slot, and then will release the sas_task. Currently in lldd_abort_task, if we fail to internally abort IO, then the slot of SMP IO is not released, but sas_task will still be later released, so the slot's sas_task is NULL, which will cause NULL pointer when hisi_sas_slot_task_free happens later. To resolve, check the return value of internal abort, and release the slot if it failed. Signed-off-by:
-
John Garry authored
Add function for upper-layer to reset controller when all else fails. Signed-off-by:
-
John Garry authored
For consistency, remove the "hisi_sas_" prefix. Signed-off-by:
-
Xiaofei Tan authored
Handle the situation that PHY UP and DOWN irq happen simultaneously. There is no mechanism of SoC HW to ensure this situation will never happen. So, we add this handle just in case. Signed-off-by:
-
John Garry authored
This patch includes: (1) Disable transport layer retry (2) Support CQ time and count interrupt coal (3) fix link FIFO full issue Signed-off-by:
Zhao Nenglong <zhaonenglong@hisilicon.com> Signed-off-by:
-
Xiang Chen authored
There are some rules to decide which error code has the high priority when errors happen together: (1) Error phase of CQ decides the error happens on RX or TX; (2) For TX error, when DMA/TRANS TX error happen simultaneously, the priority of DMA TX error is higher than TRANS TX error, so for the priority of TX error: DW2 (DMA TX part) > DW0; (3) For RX error, when TRANS/DMA/SIPC RX error happen simultaneously, the priority of TRANS RX error is higher than DMA and SIPC RX error, and we should also keep the rules (the priority of DW3 > DW2), so for the priority of RX error: DW1 > DW3 > DW2(SIPC RX part); (4) There are also a priority we should keep in the same error type. So, modify slot error code to handle this. In addition to this, some some error codes are modified according to recommendation from SoC designer. Signed-off-by: -
John Garry authored
When a TMF or internal abort times-out, do not free slot. We expect this to be done upon later escalated error handling. Signed-off-by:
-
John Garry authored
Some more locking needs to be added/modified for when read-modify-writing sas_task.task_state_flags. Note: since we can attempt to grab this lock in interrupt context we should use irq variant of spin_lock. Signed-off-by: -
Xiang Chen authored
After hardreset, we clear up IOs of remote disks, so we need to free those slots in LLDD. Signed-off-by:
-
John Garry authored
Check in slot_complete_v2_hw() for whether a task has already been completed by upper layer. Signed-off-by:
-
John Garry authored
When issuing an LU reset for a SATA target, issue an internal abort and a hard reset. Signed-off-by:
-
John Garry authored
Currently an internal abort is executed regardless of the result of the TMF. We should also check the result of the internal abort to see if we should free the slot. So change the status code STAT_IO_COMPLETE to TMF_RESP_FUNC_SUCC, meaning the slot has been successfully aborted. Signed-off-by:
-
Xiang Chen authored
For error codes which need abort-and-retry, simulate IO timeout and let SCSI+ATA layers process those errors. Previously for SSP, we should try to abort the IO in the LLDD and then pass back to upper layer, but sometimes this would also error. So Instead of adding special error handling for this scenario in the LLDD, allow the upper layer to handle completely. No performance hit is seen by taking this approach. Signed-off-by:
-
John Garry authored
We currently do a hard reset for a link reset. Change this to do a link reset only. Signed-off-by:
Xiang Chen <chenxiang@hisilicon.com> Signed-off-by:
-
John Garry authored
When sas_port is NULL, then return SAS_PHY_DOWN. In addition, when the sas_dev is gone then explicitly return SAS_PHY_DOWN. Signed-off-by:
-
John Garry authored
Currently when a root PHY is deformed from a asd_sas_port we try to release the slots in the LLDD, and fail. Regardless, it is not right to release this early. This patch removes the deformed function. As it was before, port deformation is still done in hisi_sas_phy_down(). It would be nice to actually remove the hisi_sas_port_{de}formed() pair, however we cannot as we need to know the asd_sas_port index libsas has associated with an asd_sas_phy. The hw does actually generate a port id for a PHY, but this seems to a random number, so ignored for this purpose. This patch also changes the code to link slots to the hisi_sas_device, and not hisi_sas_port. Signed-off-by: -
Xiang Chen authored
Add softreset to clear IO after internal abort device for SATA disk. The SATA error handling for the controller is based on device internal abort and softreset function. The controller does not support internal abort for single IO, so we need to execute internal abort for device. Signed-off-by:
-
John Garry authored
Relocate the PHY init code from LLDD hw init path to hisi_sas_scan_start(). Signed-off-by:
-
Xiang Chen authored
There are some scenarios that we need to warm-reset to reset registers of SAS controller. During reset we disable interrupts/DQs/PHYs, and after reset we re-init the hardware and rescan the topology to see if anything changed. Signed-off-by:
-
John Garry authored
Introduce function to get hisi_sas_port from asd_sas_port. Signed-off-by:
-
Satish Kharat authored
This is a bug introduced when they moved the fip subcodes to central place. Was sending FIP_SC_VL_NOTE in fip.fip_subcode for VLAN request in fnic_fcoe_send_vlan_req. Change is to use FIP_SC_VL_REQ instead. Signed-off-by:
Satish Kharat <satishkh@cisco.com> Signed-off-by:
Sesidhar Baddela <sebaddel@cisco.com> Signed-off-by:
-
Satish Kharat authored
The IO and Abort latency counter counts the time taken to complete the IO and abort command into broad buckets. This is not intended for performance measurement, just a debug statistic. current_max_io_time tries to keep track of the maximum time an IO has taken to complete if it is > 30sec. Signed-off-by:
-
Satish Kharat authored
Just a simple counter of number of check conditions encountered on that host. Signed-off-by:
-
Satish Kharat authored
If SCSI-ML has already issued abort on a command i.e FNIC_IOREQ_ABTS_PENDING is set and we get a IO completion, avoid this being flagged as out-of-order completion by setting the FNIC_IO_DONE flag in fnic_fcpio_icmnd_cmpl_handler Signed-off-by:
-
