kernel.git/drivers/infiniband/core/cm.c, branch linux-4.8.y

IB/cm: Mark stale CM id's whenever the mad agent was unregistered

2016-11-26T08:56:58Z

commit 9db0ff53cb9b43ed75bacd42a89c1a0ab048b2b0 upstream. When there is a CM id object that has port assigned to it, it means that the cm-id asked for the specific port that it should go by it, but if that port was removed (hot-unplug event) the cm-id was not updated. In order to fix that the port keeps a list of all the cm-id's that are planning to go by it, whenever the port is removed it marks all of them as invalid. This commit fixes a kernel panic which happens when running traffic between guests and we force reboot a guest mid traffic, it triggers a kernel panic: Call Trace: [] ? panic+0xa7/0x16f [] ? oops_end+0xe4/0x100 [] ? no_context+0xfb/0x260 [] ? del_timer_sync+0x22/0x30 [] ? __bad_area_nosemaphore+0x125/0x1e0 [] ? process_timeout+0x0/0x10 [] ? bad_area_nosemaphore+0x13/0x20 [] ? __do_page_fault+0x31f/0x480 [] ? default_wake_function+0x0/0x20 [] ? free_msg+0x55/0x70 [mlx5_core] [] ? cmd_exec+0x124/0x840 [mlx5_core] [] ? find_busiest_group+0x244/0x9f0 [] ? do_page_fault+0x3e/0xa0 [] ? page_fault+0x25/0x30 [] ? cm_alloc_msg+0x35/0xc0 [ib_cm] [] ? ib_send_cm_dreq+0xb1/0x1e0 [ib_cm] [] ? cm_destroy_id+0x176/0x320 [ib_cm] [] ? ib_destroy_cm_id+0x10/0x20 [ib_cm] [] ? ipoib_cm_free_rx_reap_list+0xa7/0x110 [ib_ipoib] [] ? ipoib_cm_rx_reap+0x0/0x20 [ib_ipoib] [] ? ipoib_cm_rx_reap+0x15/0x20 [ib_ipoib] [] ? worker_thread+0x170/0x2a0 [] ? autoremove_wake_function+0x0/0x40 [] ? worker_thread+0x0/0x2a0 [] ? kthread+0x96/0xa0 [] ? child_rip+0xa/0x20 [] ? kthread+0x0/0xa0 [] ? child_rip+0x0/0x20 Fixes: a977049dacde ("[PATCH] IB: Add the kernel CM implementation") Signed-off-by: Mark Bloch Signed-off-by: Erez Shitrit Reviewed-by: Maor Gottlieb Signed-off-by: Leon Romanovsky Signed-off-by: Doug Ledford Signed-off-by: Greg Kroah-Hartman

IB/cm: Fix a recently introduced locking bug

2016-06-06T22:55:53Z

ib_cm_notify() can be called from interrupt context. Hence do not reenable interrupts unconditionally in cm_establish(). This patch avoids that lockdep reports the following warning: WARNING: CPU: 0 PID: 23317 at kernel/locking/lockdep.c:2624 trace _hardirqs_on_caller+0x112/0x1b0 DEBUG_LOCKS_WARN_ON(current->hardirq_context) Call Trace: [] dump_stack+0x67/0x92 [] __warn+0xc1/0xe0 [] warn_slowpath_fmt+0x4a/0x50 [] trace_hardirqs_on_caller+0x112/0x1b0 [] trace_hardirqs_on+0xd/0x10 [] _raw_spin_unlock_irq+0x27/0x40 [] ib_cm_notify+0x25c/0x290 [ib_cm] [] srpt_qp_event+0xa1/0xf0 [ib_srpt] [] mlx4_ib_qp_event+0x67/0xd0 [mlx4_ib] [] mlx4_qp_event+0x5a/0xc0 [mlx4_core] [] mlx4_eq_int+0x3d8/0xcf0 [mlx4_core] [] mlx4_msi_x_interrupt+0xc/0x20 [mlx4_core] [] handle_irq_event_percpu+0x64/0x100 [] handle_irq_event+0x34/0x60 [] handle_edge_irq+0x6a/0x150 [] handle_irq+0x15/0x20 [] do_IRQ+0x5c/0x110 [] common_interrupt+0x89/0x89 [] blk_run_queue_async+0x37/0x40 [] rq_completed+0x43/0x70 [dm_mod] [] dm_softirq_done+0x176/0x280 [dm_mod] [] blk_done_softirq+0x52/0x90 [] __do_softirq+0x10f/0x230 [] irq_exit+0xa8/0xb0 [] smp_trace_call_function_single_interrupt+0x2e/0x30 [] smp_call_function_single_interrupt+0x9/0x10 [] call_function_single_interrupt+0x89/0x90 Fixes: commit be4b499323bf (IB/cm: Do not queue work to a device that's going away) Signed-off-by: Bart Van Assche Cc: Erez Shitrit Cc: Sean Hefty Cc: Nikolay Borisov Cc: stable # v4.2+ Acked-by: Erez Shitrit Signed-off-by: Doug Ledford

IB/core: Use hop-limit from IP stack for RoCE

2016-01-19T20:26:56Z

Previously, IPV6_DEFAULT_HOPLIMIT was used as the hop limit value for RoCE. Fixing that by taking ip4_dst_hoplimit and ip6_dst_hoplimit as hop limit values. Signed-off-by: Matan Barak Signed-off-by: Doug Ledford

IB/cm: Fix a recently introduced deadlock

2016-01-19T20:26:55Z

ib_send_cm_drep() calls cm_enter_timewait() while holding a spinlock that can be locked from inside an interrupt handler. Hence do not enable interrupts inside cm_enter_timewait() if called with interrupts disabled. This patch fixes e.g. the following deadlock: Acked-by: Erez Shitrit ================================= [ INFO: inconsistent lock state ] 4.4.0-rc7+ #1 Tainted: G E --------------------------------- inconsistent {HARDIRQ-ON-W} -> {IN-HARDIRQ-W} usage. swapper/8/0 [HC1[1]:SC0[0]:HE0:SE1] takes: (&(&cm_id_priv->lock)->rlock){?.+...}, at: [] cm_establish+0x 74/0x1b0 [ib_cm] {HARDIRQ-ON-W} state was registered at: [] mark_held_locks+0x71/0x90 [] trace_hardirqs_on_caller+0xa7/0x1c0 [] trace_hardirqs_on+0xd/0x10 [] _raw_spin_unlock_irq+0x2b/0x40 [] cm_enter_timewait+0xae/0x100 [ib_cm] [] ib_send_cm_drep+0xb6/0x190 [ib_cm] [] srp_cm_handler+0x128/0x1a0 [ib_srp] [] cm_process_work+0x20/0xf0 [ib_cm] [] cm_dreq_handler+0x135/0x2c0 [ib_cm] [] cm_work_handler+0x75/0xd0 [ib_cm] [] process_one_work+0x1bd/0x460 [] worker_thread+0x118/0x420 [] kthread+0xe4/0x100 [] ret_from_fork+0x3f/0x70 irq event stamp: 1672286 hardirqs last enabled at (1672283): [] poll_idle+0x10/0x80 hardirqs last disabled at (1672284): [] common_interrupt+0x84/0x89 softirqs last enabled at (1672286): [] _local_bh_enable+0x1c/0x50 softirqs last disabled at (1672285): [] irq_enter+0x47/0x70 other info that might help us debug this: Possible unsafe locking scenario: CPU0 ---- lock(&(&cm_id_priv->lock)->rlock); lock(&(&cm_id_priv->lock)->rlock); *** DEADLOCK *** no locks held by swapper/8/0. stack backtrace: CPU: 8 PID: 0 Comm: swapper/8 Tainted: G E 4.4.0-rc7+ #1 Hardware name: Dell Inc. PowerEdge R430/03XKDV, BIOS 1.0.2 11/17/2014 ffff88045af5e950 ffff88046e503a88 ffffffff81251c1b 0000000000000007 0000000000000006 0000000000000003 ffff88045af5ddc0 ffff88046e503ad8 ffffffff810a32f4 0000000000000000 0000000000000000 0000000000000001 Call Trace: [] dump_stack+0x4f/0x74 [] print_usage_bug+0x184/0x190 [] mark_lock_irq+0xf2/0x290 [] mark_lock+0x115/0x1b0 [] mark_irqflags+0x15c/0x170 [] __lock_acquire+0x1ef/0x560 [] lock_acquire+0x62/0x80 [] _raw_spin_lock_irqsave+0x43/0x60 [] cm_establish+0x74/0x1b0 [ib_cm] [] ib_cm_notify+0x31/0x100 [ib_cm] [] srpt_qp_event+0x54/0xd0 [ib_srpt] [] mlx4_ib_qp_event+0x72/0xc0 [mlx4_ib] [] mlx4_qp_event+0x69/0xd0 [mlx4_core] [] mlx4_eq_int+0x51e/0xd50 [mlx4_core] [] mlx4_msi_x_interrupt+0xf/0x20 [mlx4_core] [] handle_irq_event_percpu+0x40/0x110 [] handle_irq_event+0x3f/0x70 [] handle_edge_irq+0x79/0x120 [] handle_irq+0x5d/0x130 [] do_IRQ+0x6d/0x130 [] common_interrupt+0x89/0x89 [] cpuidle_enter_state+0xcf/0x200 [] cpuidle_enter+0x12/0x20 [] call_cpuidle+0x36/0x60 [] cpuidle_idle_call+0x63/0x110 [] cpu_idle_loop+0xfa/0x130 [] cpu_startup_entry+0xe/0x10 [] start_secondary+0x83/0x90 Fixes: commit be4b499323bf ("IB/cm: Do not queue work to a device that's going away") Signed-off-by: Bart Van Assche Cc: Erez Shitrit Cc: stable Signed-off-by: Doug Ledford

IB/mad: pass ib_mad_send_buf explicitly to the recv_handler

2016-01-19T20:25:36Z

Stop abusing wr_id and just pass the parameter explicitly. Signed-off-by: Christoph Hellwig Reviewed-by: Hal Rosenstock Reviewed-by: Ira Weiny Reviewed-by: Sagi Grimberg Signed-off-by: Doug Ledford

IB/core: Validate route when we init ah

2015-12-23T15:35:12Z

In order to make sure API users don't try to use SGIDs which don't conform to the routing table, validate the route before searching the RoCE GID table. Signed-off-by: Matan Barak Signed-off-by: Doug Ledford

IB/cm: Use the source GID index type

2015-12-23T15:35:10Z

Previosuly, cm and cma modules supported only IB and RoCE v1 GID type. In order to support multiple GID types, the gid_type is passed to cm_init_av_by_path and stored in the path record. The rdma cm client would use a default GID type that will be saved in rdma_id_private. Signed-off-by: Matan Barak Signed-off-by: Doug Ledford

IB/core: Add gid_type to gid attribute

2015-12-23T15:35:10Z

In order to support multiple GID types, we need to store the gid_type with each GID. This is also aligned with the RoCE v2 annex "RoCEv2 PORT GID table entries shall have a "GID type" attribute that denotes the L3 Address type". The currently supported GID is IB_GID_TYPE_IB which is also RoCE v1 GID type. This implies that gid_type should be added to roce_gid_table meta-data. Signed-off-by: Matan Barak Signed-off-by: Doug Ledford

IB/core: Avoid calling ib_query_device

2015-12-22T19:39:00Z

Use the cached copy of the attributes present on the device, except for the case of a query originating from user-space, where we have to invoke the driver query_device entry, so they can fill in their udata. Signed-off-by: Or Gerlitz Signed-off-by: Doug Ledford

IB/cm: Remove the usage of smac and vid of qp_attr and cm_av

2015-10-22T03:48:18Z

The cm and cma don't need to explicitly handle vlan and smac, as they are resolved from the GID index now. Removing this portion of code. Signed-off-by: Matan Barak Reviewed-By: Devesh Sharma Signed-off-by: Doug Ledford