- 27 Oct, 2011 1 commit
-
-
Nicholas Bellinger authored
This patch changes target core to also check for -ENOMEM from fabric callbacks to signal QUEUE_FULL status, instead of just -EAGAIN in order to catch a larger set of fabric failure cases that want to trigger QUEUE_FULL logic. This includes the callbacks for ->write_pending(), ->queue_data_in() and ->queue_status(). It also makes transport_generic_write_pending() return zero upon QUEUE_FULL, and removes two unnecessary -EAGAIN checks to catch write pending QUEUE_FULL cases from transport_generic_new_cmd() failures in transport_handle_cdb_direct() and transport_processing_thread():TRANSPORT_NEW_CMD_MAP state. Reported-by:
Bart Van Assche <bvanassche@acm.org> Cc: Bart Van Assche <bvanassche@acm.org> Cc: Christoph Hellwig <hch@lst.de> Cc: Roland Dreier <roland@purestorage.com> Signed-off-by:
Nicholas A. Bellinger <nab@linux-iscsi.org>
-
- 26 Oct, 2011 2 commits
-
-
Nicholas Bellinger authored
This patch addresses an issue with buggy userspace code sending I/O via scsi-generic that does not explictly clear their associated read buffers. It adds an explict memset of the first SGL entry within tcm_loop_new_cmd_map() for SCF_SCSI_CONTROL_SG_IO_CDB payloads that are currently guaranteed to be a single SGL by target-core code. This issue is a side effect of the v3.1-rc1 merge to remove the extra memcpy between certain control CDB types using a contigious + cleared buffer in target-core, and performing a memcpy into the SGL list within tcm_loop. It was originally mainfesting itself by udev + scsi_id + scsi-generic not properly setting up the expected /dev/disk/by-id/ symlinks because the INQUIRY payload was containing extra bogus data preventing the proper NAA IEEE WWN from being parsed by userspace. Cc: Christoph Hellwig <hch@lst.de> Cc: Andy Grover <agrover@redhat.com> Cc: stable@kernel.org Signed-off-by: -
Nicholas Bellinger authored
This patch fixes the following compile warning in target_core_cdb.c in recent linux-next code due to the new use of EXPORT_SYMBOL() for target_get_task_cdb(). drivers/target/target_core_cdb.c:1316: warning: data definition has no type or storage class drivers/target/target_core_cdb.c:1316: warning: type defaults to ‘int’ in declaration of ‘EXPORT_SYMBOL’ drivers/target/target_core_cdb.c:1316: warning: parameter names (without types) in function declaration Signed-off-by:
-
- 24 Oct, 2011 37 commits
-
-
Nicholas Bellinger authored
This patch removes the legacy usage of se_task->task_timer and associated infrastructure that originally was used as a way to help manage buggy backend SCSI LLDs that in certain cases would never return back an outstanding task. This includes the removal of target_complete_timeout_work(), timeout logic from transport_complete_task(), transport_task_timeout_handler(), transport_start_task_timer(), the per device task_timeout configfs attribute, and all task_timeout associated structure members and defines in target_core_base.h This is being removed in preparation to make transport_complete_task() run in lock-less mode. Cc: Christoph Hellwig <hch@lst.de> Signed-off-by: -
Nicholas Bellinger authored
This patch converts target-core to use se_cmd->t_transport_sent instead of a duplicated se_cmd->transport_sent member in a handful of locations. It also updates iscsi_target to properly use ->t_transport_sent instead of it's own iscsi_cmd_t->transport_sent value that was not being assigned. Reported-by:
Christoph Hellwig <hch@lst.de> Cc: Christoph Hellwig <hch@lst.de> Signed-off-by:
-
Christoph Hellwig authored
If we only have a single task per command (which at least in my testing is the by far most common case) we do not have to allocate a new per-task S/G list but can reuse the one from the command. (nab: Fix BIDI handling in transport_free_dev_tasks) Signed-off-by:
-
Nicholas Bellinger authored
This patch fixes a bug for BIDI handling in transport_generic_new_cmd() where cmd->t_task_cdbs_left and Co. where not taking into account the extra task count generated during the first call to transport_allocate_data_tasks(). Cc: Christoph Hellwig <hch@lst.de> Cc: Boaz Harrosh <bharrosh@panasas.com> Signed-off-by: -
Christoph Hellwig authored
There were only two callers, and one of them always wants the call to transport_allocate_data_tasks anyway. Also drop the constant lba argument to transport_allocate_data_tasks and move the variables inside it into the minimum required scope. Signed-off-by:
-
Christoph Hellwig authored
These are two fairly small functions, and merging them gives a much more readable control flow, and opportunities for more useful comments. It also moves all code related to resources allocation closer together and allows to remove a forward declaration for transport_allocate_tasks. Signed-off-by:
-
Christoph Hellwig authored
This field is never used given that BIDI handling happens at the command and not the task level. Remove it and the dead code in pscsi that tries to work on it. It also prevents pSCSI passthrough for the two currently enabled BIDI commands now that task->task_sg_bidi support has been removed. Signed-off-by:
-
Nicholas Bellinger authored
Remove the now unnecessary extra call to transport_subsystem_check_init() in target_core_register_fabric(), and also merge transport_subsystem_reqmods() directly into transport_subsystem_check_init(). Reported-by:
Christoph Hellwig <hch@infradead.org> Signed-off-by:
-
Christoph Hellwig authored
Instead of abusing the target processing thread for offloading I/O completion in the backends to user context add a new workqueue. This means completions can be processed as fast as available CPU time allows it, including in parallel with other completions and more importantly I/O submission or QUEUE FULL retries. This should give much better performance especially on loaded systems. As a fallout we can merge all the completed states into a single one. On the downside this change complicates lun reset handling a bit by requiring us to cancel a work item only for those states that have it initialized. The alternative would be to either always initialize the work item to a dummy handler, or always use the same handler and do a switch on the state. The long term solution will be a flag that says that the command has an initialized work item, but that's only going to be useful once we have more users. Signed-off-by:
-
Christoph Hellwig authored
Signed-off-by:
-
Christoph Hellwig authored
We never check for this state, and it makes testing for a completed state much harder given that it overrides the existing state. Also remove the unused deferred_t_state which is related to it. Signed-off-by:
-
Christoph Hellwig authored
We never queue an command with this state, and only set it in a completely bogus place in tcm_fc. Signed-off-by:
-
Christoph Hellwig authored
We only need to decrement dev->depth_left if failing a command from __transport_execute_tasks. Instead of doing it first thing in transport_generic_request_failure and requiring a pseudo-flag argument for it just opencode the decrement in the two callers (which should be factored into a single one anyway) Signed-off-by:
-
Christoph Hellwig authored
Currently we stop the timers for all tasks in a command fairly late during I/O completion, which is fairly pointless and requires all kinds of safety checks. Instead delete pending timers early on in transport_complete_task, thus ensuring no new timers firest after that. We take t_state_lock a bit later in that function thus making sure currenly running timers are out of the criticial section. To be completely sure the timer has finished we also add another del_timer_sync call when freeing the task. This also allows removing TF_TIMER_RUNNING as it would be equivalent to TF_ACTIVE now. Signed-off-by:
-
Christoph Hellwig authored
TF_TIMER_STOP is useless as it only helps to mitigate a tiny race during deleting the timer. But given that we have cleared TF_ACTIVE at this point we already have another mitigation a few lines down the function. Signed-off-by:
-
Christoph Hellwig authored
Signed-off-by:
-
Christoph Hellwig authored
list_for_each_entry_safe only protects against deletions from the list, but not against any concurrent modifications. Given that we drop t_state_lock inside the loop it is not safe in transport_free_dev_tasks. Instead of use a local dispose_list that we move all tasks that are to be deleted to. This is safe because we never do list_emptry checks on t_list to check if a command is on the list anywhere. Signed-off-by:
-
Christoph Hellwig authored
Change one remaining user of transport_cmd_check_stop(cmd, 2, 0) to the transport_cmd_check_stop_to_fabric wrapper. Signed-off-by:
-
Christoph Hellwig authored
We always operated on the same queue, so move finding it into the function, just like we do for all other helpers operating on it. Signed-off-by:
-
Christoph Hellwig authored
Add a new boolean at_head parameter to transport_add_cmd_to_queue and thus obsolete the SCF_EMULATE_QUEUE_FULL flag. Signed-off-by:
-
Christoph Hellwig authored
Remove the need for the transport_qf_callback callback by making sure we have specific states with specific handlers for the two queue full cases. Signed-off-by:
-
Christoph Hellwig authored
Remove the dpo_emulated, fua_write_emulated, fua_read_emulated and write_cache_emulated methods, and replace them with a simple bitfields in se_subsystem_api in those cases where they ever returned one. Signed-off-by:
-
Christoph Hellwig authored
Do not block the submitting thread when handling a SYNCHRONIZE CACHE command, but implement it asynchronously by sending the FLUSH command ourself and calling transport_complete_sync_cache from the completion handler. Signed-off-by:
-
Nicholas Bellinger authored
This patch fixes a bug with the handling of REPORT TARGET PORT GROUPS containing a smaller allocation length than the payload requires causing memory writes beyond the end of the buffer. This patch checks for the minimum 4 byte length for the response payload length, and also checks upon each loop of T10_ALUA(su_dev)->tg_pt_gps_list to ensure the Target port group and Target port descriptor list is able to fit into the remaining allocation length. If the response payload exceeds the allocation length length, then rd_len is still increments to indicate to the initiator that the payload has been truncated. Reported-by:
Roland Dreier <roland@purestorage.com> Cc: stable@kernel.org Signed-off-by:
Nicholas Bellinger <nab@risingtidesystems.com>
-
Christoph Hellwig authored
Add a switch statement implementing the CDB LBA/len update directly in target_get_task_cdb and remove the old ->transport_split_cdb callback and all its implementations. Signed-off-by:
-
Christoph Hellwig authored
Instead of calling out to the backends from the core to get a per-task CDB and then modify it for the LBA/len pair used for this CDB provide a helper that writes the adjusted CDB into a provided buffer and call this method from ->do_task in pscsi. Signed-off-by:
-
Christoph Hellwig authored
The most commonly used file, iblock and rd backends have no use for a per-task CDB and thus don't need a method to copy it into their otherwise unused CDB fields. Signed-off-by:
-
Christoph Hellwig authored
Rearrange the fields in se_task to avoid holes. Also increase the flags field to 16 bits as we have the space for it, and this makes adding new flags safer. Signed-off-by:
-
Christoph Hellwig authored
This is a squashed version of the following unnecessary se_task structure member removal patches: target: remove the task_execute_queue field in se_task Instead of using a separate flag we can simply do list_emptry checks on t_execute_list if we make sure to always use list_del_init to remove a task from the list. Also factor some duplicate code into a new __transport_remove_task_from_execute_queue helper. target: remove the read-only task_no field in se_task The task_no field never was initialized and only used in debug printks, so kill it. target: remove the task_padded_sg field in se_task This field is only check in one place and not actually needed there. Rationale: - transport_do_task_sg_chain asserts that we have task_sg_chaining set early on - we only make use of the sg_prev_nents field we calculate based on it if there is another sg list that gets chained onto this one, which never happens for the last (or only) task. Signed-off-by: -
Christoph Hellwig authored
Replace various atomic_t variables that were mostly under t_state_lock with new flags in task_flags. Note that the execution error path didn't take t_state_lock before, so add it there. Signed-off-by:
-
Christoph Hellwig authored
This is a squashed version of the following se_task cleanup patches: target: remove the unused task_state_flags field in se_task target: remove the unused se_obj_ptr field in se_task target: remove the se_dev field in se_task Signed-off-by: -
Christoph Hellwig authored
This is a squashed version of the following target_core_base.h cleanup patches: target: remove the unused SHUTDOWN_SIGS defintion target: remove unused se_mem leftovers target: remove the unused map_func_t typedef target: move TRANSPORT_IOV_DATA_BUFFER to the iscsi-specific code Signed-off-by: -
Nicholas Bellinger authored
This patch fixes a bug with tcm_loop where performing a scsi_host rescan was causing an oops due to a received scsi_cmnd->device->id value not matching a previously configured tcm_loop_tpg entry in tcm_loop_hba->tl_hba_tpgs[] obtained from within tcm_loop_queuecommand() code. This fix adds an explict check for tcm_loop_tpg->tl_hba in order to ensure tcm_loop_make_naa_tpg() has already been invoked to initialize a given tcm_loop_tpg entry, and also adds an explict clear of tcm_loop_tpg->tl_hba from within the tcm_loop_drop_naa_tpg() release path. This bug was manifesting itself with the following OOPs: [176289.430909] BUG: unable to handle kernel NULL pointer dereference at 0000000000000090 [176289.431337] IP: [<ffffffffa0395617>] transport_processing_thread+0x1e3/0x794 [target_core_mod] [176289.431399] PGD 22e9b067 PUD 23375067 PMD 0 [176289.431399] Oops: 0000 [#1] SMP [176289.431815] CPU 1 [176289.431815] Modules linked in: tcm_loop target_core_stgt target_core_pscsi target_core_file target_core_iblock target_core_mod crc32c ib_cm ib_sa ib_mad ib_core qla2xxx scsi_tgt configfs fcoe libfcoe libfc scsi_transport_fc ipv6 iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi sr_mod cdrom sd_mod ata_piix libata e1000 mptspi mptscsih mptbase [last unloaded: target_core_mod] [176289.431815] [176289.431815] Pid: 12339, comm: LIO_iblock Tainted: G W 3.1.0-rc8+ [176289.431815] RIP: 0010:[<ffffffffa0395617>] [<ffffffffa0395617>] transport_processing_thread+0x1e3/0x794 [target_core_mod] [176289.431815] RSP: 0018:ffff880023bfbe10 EFLAGS: 00010283 [176289.431815] RAX: 0000000000000000 RBX: ffff88002d600040 RCX: ffff88002d600108 [176289.431815] RDX: ffff88000c9e50bc RSI: 0000000000000246 RDI: 0000000000000246 [176289.431815] RBP: ffff880023bfbee0 R08: ffff88002d600108 R09: 0000000000000000 [176289.431815] R10: ffff88002fc8cc80 R11: ffffffff81671b60 R12: ffff88002d600108 [176289.431815] R13: ffff88000c9e4f38 R14: ffff88000c9e50b8 R15: 0000000000000000 [176289.431815] FS: 0000000000000000(0000) GS:ffff88002fc80000(0000) knlGS:0000000000000000 [176289.431815] CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b [176289.431815] CR2: 0000000000000090 CR3: 000000002a33f000 CR4: 00000000000006e0 [176289.431815] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [176289.431815] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 [176289.431815] Process LIO_iblock (pid: 12339, threadinfo ffff880023bfa000, task ffff88002a2e0000) [176289.431815] Stack: [176289.431815] 0000000000011280 0000000000000246 ffff88002a2e0000 ffff880023a58900 [176289.431815] ffff880023bfbed0 ffff880023bfa000 ffff880023bfa000 ffff88000c9e50d0 [176289.431815] ffff88000c9e50c0 ffff88000c9e50bc ffff880023bfa000 ffff880023bfbfd8 [176289.431815] Call Trace: [176289.431815] [<ffffffff81056657>] ? wake_up_bit+0x25/0x25 [176289.431815] [<ffffffffa0395434>] ? transport_handle_cdb_direct+0x92/0x92 [target_core_mod] [176289.431815] [<ffffffff8105619a>] kthread+0x7d/0x85 [176289.431815] [<ffffffff813cbcb4>] kernel_thread_helper+0x4/0x10 [176289.431815] [<ffffffff8105611d>] ? kthread_worker_fn+0x16d/0x16d [176289.431815] [<ffffffff813cbcb0>] ? gs_change+0x13/0x13 [176289.431815] Code: 67 05 00 00 41 8b 84 24 4c ff ff ff ff c8 83 f8 11 0f 87 f0 04 00 00 89 c0 ff 24 c5 b0 c6 39 a0 0f 0b eb fe 48 8b 83 d8 00 00 00 [176289.431815] RIP [<ffffffffa0395617>] transport_processing_thread+0x1e3/0x794 [target_core_mod] [176289.431815] RSP <ffff880023bfbe10> [176289.431815] CR2: 0000000000000090 [176295.041004] ---[ end trace 85dc6865b23b8f3e ]--- Reported-by:
Paolo Bonzini <pbonzini@redhat.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by:
-
Nicholas Bellinger authored
This patch removes the legacy device active I/O shutdown code that was originally called from transport_processing_thread() context during shutdown including transport_processing_shutdown() and transport_release_all_cmds(). This is due to the fact that in modern configfs control plane code by the time shutdown of an se_device instance in transport_processing_thread() is allowed to occur via: rmdir /sys/kernel/config/target/core/$HBA/$DEV all active I/O will already have been ceased while removing active configfs fabric Port/LUN symlinks. Eg: the removal of an active se_device is protected by inter-module VFS references from active Port/LUN symlinks. Two WARN_ON() checks have been added in their place before exiting transport_processing_thread() to watch out for any leaked descriptors. Cc: Christoph Hellwig <hch@lst.de> Signed-off-by: -
Nicholas Bellinger authored
This patch merges transport_cmd_finish_abort_tmr() logic into a single transport_cmd_finish_abort() function by adding a cmd->se_tmr_req check around transport_lun_remove_cmd(), and updates the single caller within core_tmr_drain_tmr_list(). Reported-by:
-
Nicholas Bellinger authored
This patch removes a number of SCF_SE_LUN_CMD flag abuses within iscsi-target code to determine when iscsit_release_cmd() or transport_generic_free_cmd() should be called while releasing an individual iscsi_cmd descriptor. In the place of SCF_SE_LUN_CMD checks, this patch converts existing code to use a new iscsit_free_cmd() that inspects iscsi_cmd->iscsi_opcode types to determine which of the above functions should be invoked. It also removes the now unnecessary special case checking in iscsit_release_commands_from_conn(). (hch: Use iscsit_free_cmd instead of open-coded alternative) Reported-by:
-
Nicholas Bellinger authored
This patch converts se_cmd->transport_wait_for_tasks(se_cmd, 1) usage to use transport_generic_free_cmd() directly in target-core and iscsi-target fabric usage. The includes: *) Removal of the optional transport_generic_free_cmd() call from within transport_generic_wait_for_tasks() *) Usage of existing SCF_SUPPORTED_SAM_OPCODE to determine when transport_generic_wait_for_tasks() processing may occur instead of checking se_cmd->transport_wait_for_tasks() *) Move transport_generic_wait_for_tasks() call ahead of core_dec_lacl_count() and transport_lun_remove_cmd() in transport_generic_free_cmd() to follow existing logic for iscsi-target w/ se_cmd->transport_wait_for_tasks(se_cmd, 1) *) Removal of se_cmd->transport_wait_for_tasks() function pointer *) Rename transport_generic_wait_for_tasks() -> transport_wait_for_tasks(), and add docbook comment. *) Add EXPORT_SYMBOL for transport_wait_for_tasks() For the case in iscsi_target_erl2.c:iscsit_prepare_cmds_for_realligance() where se_cmd->transport_wait_for_tasks(se_cmd, 0) is called, this patch adds a direct call to transport_wait_for_tasks(). (hch: Fix transport_generic_free_cmd() usage in iscsit_release_commands_from_conn) (nab: Add patch: Ensure that TMRs hit wait_for_tasks logic during release) Reported-by:
-
