kernel.git/drivers/infiniband/ulp/srp, branch linux-3.12.y

IB/srp: Fix deadlock between host removal and multipathd

2014-09-17T20:54:14Z

commit bcc05910359183b431da92713e98eed478edf83a upstream. If scsi_remove_host() is invoked after a SCSI device has been blocked, if the fast_io_fail_tmo or dev_loss_tmo work gets scheduled on the workqueue executing srp_remove_work() and if an I/O request is scheduled after the SCSI device had been blocked by e.g. multipathd then the following deadlock can occur: kworker/6:1 D ffff880831f3c460 0 195 2 0x00000000 Call Trace: [] schedule+0x29/0x70 [] schedule_timeout+0x10f/0x2a0 [] msleep+0x2f/0x40 [] __blk_drain_queue+0x4e/0x180 [] blk_cleanup_queue+0x225/0x230 [] __scsi_remove_device+0x62/0xe0 [scsi_mod] [] scsi_forget_host+0x6f/0x80 [scsi_mod] [] scsi_remove_host+0x7a/0x130 [scsi_mod] [] srp_remove_work+0x95/0x180 [ib_srp] [] process_one_work+0x1ea/0x6c0 [] worker_thread+0x11b/0x3a0 [] kthread+0xed/0x110 [] ret_from_fork+0x7c/0xb0 multipathd D ffff880096acc460 0 5340 1 0x00000000 Call Trace: [] schedule+0x29/0x70 [] schedule_timeout+0x10f/0x2a0 [] io_schedule_timeout+0x9b/0xf0 [] wait_for_completion_io_timeout+0xdc/0x110 [] blk_execute_rq+0x9b/0x100 [] sg_io+0x1a5/0x450 [] scsi_cmd_ioctl+0x2a1/0x430 [] scsi_cmd_blk_ioctl+0x42/0x50 [] sd_ioctl+0xbe/0x140 [sd_mod] [] blkdev_ioctl+0x234/0x840 [] block_ioctl+0x41/0x50 [] do_vfs_ioctl+0x300/0x520 [] SyS_ioctl+0x41/0x80 [] tracesys+0xd0/0xd5 Fix this by scheduling removal work on another workqueue than the transport layer timers. Signed-off-by: Bart Van Assche Reviewed-by: Sagi Grimberg Reviewed-by: David Dillow Cc: Sebastian Parschauer Signed-off-by: Roland Dreier Signed-off-by: Jiri Slaby

IB/srp: Fix a sporadic crash triggered by cable pulling

2014-07-17T11:43:17Z

commit 024ca90151f5e4296d30f72c13ff9a075e23c9ec upstream. Avoid that the loops that iterate over the request ring can encounter a pointer to a SCSI command in req->scmnd that is no longer associated with that request. If the function srp_unmap_data() is invoked twice for a SCSI command that is not in flight then that would cause ib_fmr_pool_unmap() to be invoked with an invalid pointer as argument, resulting in a kernel oops. Reported-by: Sagi Grimberg Reference: http://thread.gmane.org/gmane.linux.drivers.rdma/19068/focus=19069 Signed-off-by: Bart Van Assche Reviewed-by: Sagi Grimberg Signed-off-by: Roland Dreier Signed-off-by: Jiri Slaby

IB/srp: Report receive errors correctly

2013-12-04T19:05:24Z

commit cd4e38542a5c2cab94e5410fb17c1cc004a60792 upstream. The IB spec does not guarantee that the opcode is available in error completions. Hence do not rely on it. See also commit 948d1e889e5b ("IB/srp: Introduce srp_handle_qp_err()"). Signed-off-by: Bart Van Assche Signed-off-by: Roland Dreier Signed-off-by: Greg Kroah-Hartman

IB/srp: Avoid offlining operational SCSI devices

2013-12-04T19:05:24Z

commit 99b6697a50c2acbe3ca2772d359fc9a28835dc84 upstream. If SCSI commands are submitted with a SCSI request timeout that is lower than the the IB RC timeout, it can happen that the SCSI error handler has already started device recovery before transport layer error handling starts. So it can happen that the SCSI error handler tries to abort a SCSI command after it has been reset by srp_rport_reconnect(). Tell the SCSI error handler that such commands have finished and that it is not necessary to continue its recovery strategy for commands that have been reset by srp_rport_reconnect(). Signed-off-by: Bart Van Assche Signed-off-by: Roland Dreier Signed-off-by: Greg Kroah-Hartman

IB/srp: Remove target from list before freeing Scsi_Host structure

2013-12-04T19:05:24Z

commit 65d7dd2f3479ef5aec1d9ddd1481cb7851c11af6 upstream. Remove an SRP target from the SRP target list before invoking the last scsi_host_put() call. This change is necessary because that last put frees the memory that holds the srp_target_port structure. This patch prevents the following kernel oops: RIP: 0010:[] __lock_acquire+0x500/0x1570 Call Trace: [] lock_acquire+0xa4/0x120 [] _spin_lock+0x36/0x70 [] srp_remove_work+0xef/0x180 [ib_srp] [] worker_thread+0x21c/0x3d0 [] kthread+0x96/0xa0 [] child_rip+0xa/0x20 Signed-off-by: Vu Pham Signed-off-by: Greg Kroah-Hartman [ bvanassche - Modified path description and CC'ed stable. ] Signed-off-by: Bart Van Assche Signed-off-by: Roland Dreier

IB/srp: Let srp_abort() return FAST_IO_FAIL if TL offline

2013-07-11T23:43:48Z

If the transport layer is offline it is more appropriate to let srp_abort() return FAST_IO_FAIL instead of SUCCESS. Reported-by: Sebastian Riemer Acked-by: David Dillow Signed-off-by: Bart Van Assche Signed-off-by: Roland Dreier

IB/srp: Bump driver version and release date

2013-07-01T17:42:02Z

Signed-off-by: Vu Pham Signed-off-by: Bart Van Assche Signed-off-by: Roland Dreier

IB/srp: Make HCA completion vector configurable

2013-07-01T17:40:55Z

Several InfiniBand HCAs allow configuring the completion vector per CQ. This allows spreading the workload created by IB completion interrupts over multiple MSI-X vectors and hence over multiple CPU cores. In other words, configuring the completion vector properly not only allows reducing latency on an initiator connected to multiple SRP targets but also allows improving throughput. Signed-off-by: Bart Van Assche Acked-by: David Dillow Signed-off-by: Roland Dreier

IB/srp: Maintain a single connection per I_T nexus

2013-07-01T17:40:07Z

An SRP target is required to maintain a single connection between initiator and target. This means that if the 'add_target' attribute is used to create a second connection to a target, the first connection will be logged out and that the SCSI error handler will kick in. The SCSI error handler will cause the SRP initiator to reconnect, which will cause I/O over the second connection to fail. Avoid such ping-pong behavior by disabling relogins. If reconnecting manually is necessary, that is possible by deleting and recreating an rport via sysfs. Signed-off-by: Bart Van Assche Signed-off-by: Sebastian Riemer Acked-by: David Dillow Signed-off-by: Roland Dreier

IB/srp: Fail I/O fast if target offline

2013-07-01T17:37:13Z

If reconnecting failed we know that no command completion will be received anymore. Hence let the SCSI error handler fail such commands immediately. Signed-off-by: Bart Van Assche Acked-by: David Dillow Signed-off-by: Roland Dreier