IB/cm: Fix a recently introduced locking bug
PATCH 1/1 IB/cm: Fix a recently introduced locking bug
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: <IRQ> [<ffffffff812bd0e5>] dump_stack+0x67/0x92 [<ffffffff81056f21>] __warn+0xc1/0xe0 [<ffffffff81056f8a>] warn_slowpath_fmt+0x4a/0x50 [<ffffffff810a5932>] trace_hardirqs_on_caller+0x112/0x1b0 [<ffffffff810a59dd>] trace_hardirqs_on+0xd/0x10 [<ffffffff815992c7>] _raw_spin_unlock_irq+0x27/0x40 [<ffffffffa0382e9c>] ib_cm_notify+0x25c/0x290 [ib_cm] [<ffffffffa068fbc1>] srpt_qp_event+0xa1/0xf0 [ib_srpt] [<ffffffffa04efb97>] mlx4_ib_qp_event+0x67/0xd0 [mlx4_ib] [<ffffffffa034ec0a>] mlx4_qp_event+0x5a/0xc0 [mlx4_core] [<ffffffffa03365f8>] mlx4_eq_int+0x3d8/0xcf0 [mlx4_core] [<ffffffffa0336f9c>] mlx4_msi_x_interrupt+0xc/0x20 [mlx4_core] [<ffffffff810b0914>] handle_irq_event_percpu+0x64/0x100 [<ffffffff810b09e4>] handle_irq_event+0x34/0x60 [<ffffffff810b3a6a>] handle_edge_irq+0x6a/0x150 [<ffffffff8101ad05>] handle_irq+0x15/0x20 [<ffffffff8101a66c>] do_IRQ+0x5c/0x110 [<ffffffff8159a2c9>] common_interrupt+0x89/0x89 [<ffffffff81297a17>] blk_run_queue_async+0x37/0x40 [<ffffffffa0163e53>] rq_completed+0x43/0x70 [dm_mod] [<ffffffffa0164896>] dm_softirq_done+0x176/0x280 [dm_mod] [<ffffffff812a26c2>] blk_done_softirq+0x52/0x90 [<ffffffff8105bc1f>] __do_softirq+0x10f/0x230 [<ffffffff8105bec8>] irq_exit+0xa8/0xb0 [<ffffffff8103653e>] smp_trace_call_function_single_interrupt+0x2e/0x30 [<ffffffff81036549>] smp_call_function_single_interrupt+0x9/0x10 [<ffffffff8159a959>] call_function_single_interrupt+0x89/0x90 <EOI>
--- drivers/infiniband/core/cm.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/infiniband/core/cm.c +2 -2
@@ -3452,14 +3452,14 @@ static int cm_establish(struct ib_cm_id *cm_id) work->cm_event.event =IB_CM_USER_ESTABLISHED ;/* Check if the device started its remove_one */ -spin_lock_irq (&cm.lock); +spin_lock_irqsave (&cm.lock, flags);if (!cm_dev->going_down) {queue_delayed_work (cm.wq, &work->work,0 ); }else {kfree (work); ret = -ENODEV ; } -spin_unlock_irq (&cm.lock); +spin_unlock_irqrestore (&cm.lock, flags); out:return ret;
Signature · 6 lines
2.7.3 -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
A message that is not in the archive (sent off-list or not yet received): [email protected]
Hello Bart,
Is this supposed to fix the issue I reported earlier on the li st?
http://marc.info/?l=linux-rdma&m=145890571132495&w=2
Quoted 69 lines · Bart Van Assche
ib_cm_notify()can be called from interrupt context. Hence do not reenable interrupts unconditionally incm_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: <IRQ> [<ffffffff812bd0e5>] dump_stack+0x67/0x92 [<ffffffff81056f21>] __warn+0xc1/0xe0 [<ffffffff81056f8a>] warn_slowpath_fmt+0x4a/0x50 [<ffffffff810a5932>] trace_hardirqs_on_caller+0x112/0x1b0 [<ffffffff810a59dd>] trace_hardirqs_on+0xd/0x10 [<ffffffff815992c7>] _raw_spin_unlock_irq+0x27/0x40 [<ffffffffa0382e9c>] ib_cm_notify+0x25c/0x290 [ib_cm] [<ffffffffa068fbc1>] srpt_qp_event+0xa1/0xf0 [ib_srpt] [<ffffffffa04efb97>] mlx4_ib_qp_event+0x67/0xd0 [mlx4_ib] [<ffffffffa034ec0a>] mlx4_qp_event+0x5a/0xc0 [mlx4_core] [<ffffffffa03365f8>] mlx4_eq_int+0x3d8/0xcf0 [mlx4_core] [<ffffffffa0336f9c>] mlx4_msi_x_interrupt+0xc/0x20 [mlx4_core] [<ffffffff810b0914>] handle_irq_event_percpu+0x64/0x100 [<ffffffff810b09e4>] handle_irq_event+0x34/0x60 [<ffffffff810b3a6a>] handle_edge_irq+0x6a/0x150 [<ffffffff8101ad05>] handle_irq+0x15/0x20 [<ffffffff8101a66c>] do_IRQ+0x5c/0x110 [<ffffffff8159a2c9>] common_interrupt+0x89/0x89 [<ffffffff81297a17>] blk_run_queue_async+0x37/0x40 [<ffffffffa0163e53>] rq_completed+0x43/0x70 [dm_mod] [<ffffffffa0164896>] dm_softirq_done+0x176/0x280 [dm_mod] [<ffffffff812a26c2>] blk_done_softirq+0x52/0x90 [<ffffffff8105bc1f>] __do_softirq+0x10f/0x230 [<ffffffff8105bec8>] irq_exit+0xa8/0xb0 [<ffffffff8103653e>] smp_trace_call_function_single_interrupt+0x2e/0x30 [<ffffffff81036549>] smp_call_function_single_interrupt+0x9/0x10 [<ffffffff8159a959>] call_function_single_interrupt+0x89/0x90 <EOI>Fixes commit be4b499323bf (IB/cm: Do not queue work to a device that's going away)Signed-off-by Bart Van Assche <[email protected]>Cc Erez Shitrit <erezsh-VPRAkNaXOzVWk0Htik3J/[email protected]>Cc Sean Hefty <[email protected]>Cc Nikolay Borisov <[email protected]>Cc stable <[email protected]> # v4.2+--- drivers/infiniband/core/cm.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)drivers/infiniband/core/cm.c +2
-3@@ -3452,14 +3452,14 @@ static int cm_establish(struct ib_cm_id *cm_id) work->cm_event.event =IB_CM_USER_ESTABLISHED ;/* Check if the device started its remove_one */ -spin_lock_irq (&cm.lock); +spin_lock_irqsave (&cm.lock, flags);if (!cm_dev->going_down) {queue_delayed_work (cm.wq, &work->work,0 ); }else {kfree (work); ret = -ENODEV ; } -spin_unlock_irq (&cm.lock); +spin_unlock_irqrestore (&cm.lock, flags); out:return ret; --
Quoted 2 lines
2.7.3
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
A message that is not in the archive (sent off-list or not yet received): CAJFSNy5g9OLwNYwOfbXHUyX-sJQUvbJ=r_rmozKnefCKYADttA-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org
Quoted 2 lines · Nikolay Borisov
Is this supposed to fix the issue I reported earlier on the list?
http://marc.info/?l=linux-rdma&m=145890571132495&w=2
Hello Nikolay,
Although I have not yet analyzed your report in depth, I doubt this will fix what you reported earlier today.
Bart.
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
A message that is not in the archive (sent off-list or not yet received): [email protected]
The reason for asking was due to me being CCed and I thought it was relevant, Sorry for the noise :)
Quoted 6 lines · Bart Van Assche
Quoted 3 lines · Nikolay Borisov
Is this supposed to fix the issue I reported earlier on the list?
http://marc.info/?l=linux-rdma&m=145890571132495&w=2
Quoted 6 lines
Hello Nikolay,
Although I have not yet analyzed your report in depth, I doubt this will fix what you reported earlier today.
Bart.
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
Erez Shitrit
Acked-byQuoted 37 lines · Bart Van Assche
ib_cm_notify()can be called from interrupt context. Hence do not reenable interrupts unconditionally incm_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: <IRQ> [<ffffffff812bd0e5>] dump_stack+0x67/0x92 [<ffffffff81056f21>] __warn+0xc1/0xe0 [<ffffffff81056f8a>] warn_slowpath_fmt+0x4a/0x50 [<ffffffff810a5932>] trace_hardirqs_on_caller+0x112/0x1b0 [<ffffffff810a59dd>] trace_hardirqs_on+0xd/0x10 [<ffffffff815992c7>] _raw_spin_unlock_irq+0x27/0x40 [<ffffffffa0382e9c>] ib_cm_notify+0x25c/0x290 [ib_cm] [<ffffffffa068fbc1>] srpt_qp_event+0xa1/0xf0 [ib_srpt] [<ffffffffa04efb97>] mlx4_ib_qp_event+0x67/0xd0 [mlx4_ib] [<ffffffffa034ec0a>] mlx4_qp_event+0x5a/0xc0 [mlx4_core] [<ffffffffa03365f8>] mlx4_eq_int+0x3d8/0xcf0 [mlx4_core] [<ffffffffa0336f9c>] mlx4_msi_x_interrupt+0xc/0x20 [mlx4_core] [<ffffffff810b0914>] handle_irq_event_percpu+0x64/0x100 [<ffffffff810b09e4>] handle_irq_event+0x34/0x60 [<ffffffff810b3a6a>] handle_edge_irq+0x6a/0x150 [<ffffffff8101ad05>] handle_irq+0x15/0x20 [<ffffffff8101a66c>] do_IRQ+0x5c/0x110 [<ffffffff8159a2c9>] common_interrupt+0x89/0x89 [<ffffffff81297a17>] blk_run_queue_async+0x37/0x40 [<ffffffffa0163e53>] rq_completed+0x43/0x70 [dm_mod] [<ffffffffa0164896>] dm_softirq_done+0x176/0x280 [dm_mod] [<ffffffff812a26c2>] blk_done_softirq+0x52/0x90 [<ffffffff8105bc1f>] __do_softirq+0x10f/0x230 [<ffffffff8105bec8>] irq_exit+0xa8/0xb0 [<ffffffff8103653e>] smp_trace_call_function_single_interrupt+0x2e/0x30 [<ffffffff81036549>] smp_call_function_single_interrupt+0x9/0x10 [<ffffffff8159a959>] call_function_single_interrupt+0x89/0x90 <EOI>
Quoted 2 lines
Fixes commit be4b499323bf (IB/cm: Do not queue work to a device that's going away)Signed-off-by Bart Van Assche <[email protected]>
Quoted 32 lines
Cc Erez Shitrit <erezsh-VPRAkNaXOzVWk0Htik3J/[email protected]>Cc Sean Hefty <[email protected]>Cc Nikolay Borisov <[email protected]>Cc stable <[email protected]> # v4.2+--- drivers/infiniband/core/cm.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)drivers/infiniband/core/cm.c +2
-3@@ -3452,14 +3452,14 @@ static int cm_establish(struct ib_cm_id *cm_id) work->cm_event.event =IB_CM_USER_ESTABLISHED ;/* Check if the device started its remove_one */ -spin_lock_irq (&cm.lock); +spin_lock_irqsave (&cm.lock, flags);if (!cm_dev->going_down) {queue_delayed_work (cm.wq, &work->work,0 ); }else {kfree (work); ret = -ENODEV ; } -spin_unlock_irq (&cm.lock); +spin_unlock_irqrestore (&cm.lock, flags); out:return ret; --2.7.3
Quoted 4 lines
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
A message that is not in the archive (sent off-list or not yet received): CAAk-MO9BbZ_cpT2vQofG3DLCXfV1CAtqm1je1KE1u1Y3Gg8Pdg-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org
Quoted 9 lines · Erez Shitrit
Quoted 5 lines · Bart Van Assche
ib_cm_notify()can be called from interrupt context. Hence do not reenable interrupts unconditionally incm_establish().[ ... ]Fixes commit be4b499323bf (IB/cm: Do not queue work to a device that's going away)Signed-off-by Bart Van Assche <[email protected]>
Hello Doug,
Do you think it will be possible to send this patch to Linus before kernel v4.7 is released?
Thanks,
Bart.
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html