* [PATCH] iwcm: don't hold the irq disabled lock on iw_rem_ref
@ 2019-09-03 19:22 Sagi Grimberg
2019-09-03 19:33 ` Mark Bloch
0 siblings, 1 reply; 4+ messages in thread
From: Sagi Grimberg @ 2019-09-03 19:22 UTC (permalink / raw)
To: linux-rdma; +Cc: Jason Gunthorpe
This may be the final put on a qp and result in freeing
resourcesand should not be done with interrupts disabled.
Produce the following warning:
--
[ 317.026048] WARNING: CPU: 1 PID: 443 at kernel/smp.c:425 smp_call_function_many+0xa0/0x260
[ 317.026131] Call Trace:
[ 317.026159] ? load_new_mm_cr3+0xe0/0xe0
[ 317.026161] on_each_cpu+0x28/0x50
[ 317.026183] __purge_vmap_area_lazy+0x72/0x150
[ 317.026200] free_vmap_area_noflush+0x7a/0x90
[ 317.026202] remove_vm_area+0x6f/0x80
[ 317.026203] __vunmap+0x71/0x210
[ 317.026211] siw_free_qp+0x8d/0x130 [siw]
[ 317.026217] destroy_cm_id+0xc3/0x200 [iw_cm]
[ 317.026222] rdma_destroy_id+0x224/0x2b0 [rdma_cm]
[ 317.026226] nvme_rdma_reset_ctrl_work+0x2c/0x70 [nvme_rdma]
[ 317.026235] process_one_work+0x1f4/0x3e0
[ 317.026249] worker_thread+0x221/0x3e0
[ 317.026252] ? process_one_work+0x3e0/0x3e0
[ 317.026256] kthread+0x117/0x130
[ 317.026264] ? kthread_create_worker_on_cpu+0x70/0x70
[ 317.026275] ret_from_fork+0x35/0x40
--
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
---
drivers/infiniband/core/iwcm.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/infiniband/core/iwcm.c b/drivers/infiniband/core/iwcm.c
index 72141c5b7c95..94566271dbff 100644
--- a/drivers/infiniband/core/iwcm.c
+++ b/drivers/infiniband/core/iwcm.c
@@ -427,7 +427,9 @@ static void destroy_cm_id(struct iw_cm_id *cm_id)
break;
}
if (cm_id_priv->qp) {
+ spin_unlock_irqrestore(&cm_id_priv->lock, flags);
cm_id_priv->id.device->ops.iw_rem_ref(cm_id_priv->qp);
+ spin_lock_irqsave(&cm_id_priv->lock, flags);
cm_id_priv->qp = NULL;
}
spin_unlock_irqrestore(&cm_id_priv->lock, flags);
--
2.17.1
^ permalink raw reply related [flat|nested] 4+ messages in thread
* Re: [PATCH] iwcm: don't hold the irq disabled lock on iw_rem_ref
2019-09-03 19:22 [PATCH] iwcm: don't hold the irq disabled lock on iw_rem_ref Sagi Grimberg
@ 2019-09-03 19:33 ` Mark Bloch
2019-09-04 1:07 ` Sagi Grimberg
2019-09-04 6:03 ` Jason Gunthorpe
0 siblings, 2 replies; 4+ messages in thread
From: Mark Bloch @ 2019-09-03 19:33 UTC (permalink / raw)
To: Sagi Grimberg, linux-rdma; +Cc: Jason Gunthorpe
On 9/3/19 12:22 PM, Sagi Grimberg wrote:
> This may be the final put on a qp and result in freeing
> resourcesand should not be done with interrupts disabled.
>
> Produce the following warning:
> --
> [ 317.026048] WARNING: CPU: 1 PID: 443 at kernel/smp.c:425 smp_call_function_many+0xa0/0x260
> [ 317.026131] Call Trace:
> [ 317.026159] ? load_new_mm_cr3+0xe0/0xe0
> [ 317.026161] on_each_cpu+0x28/0x50
> [ 317.026183] __purge_vmap_area_lazy+0x72/0x150
> [ 317.026200] free_vmap_area_noflush+0x7a/0x90
> [ 317.026202] remove_vm_area+0x6f/0x80
> [ 317.026203] __vunmap+0x71/0x210
> [ 317.026211] siw_free_qp+0x8d/0x130 [siw]
> [ 317.026217] destroy_cm_id+0xc3/0x200 [iw_cm]
> [ 317.026222] rdma_destroy_id+0x224/0x2b0 [rdma_cm]
> [ 317.026226] nvme_rdma_reset_ctrl_work+0x2c/0x70 [nvme_rdma]
> [ 317.026235] process_one_work+0x1f4/0x3e0
> [ 317.026249] worker_thread+0x221/0x3e0
> [ 317.026252] ? process_one_work+0x3e0/0x3e0
> [ 317.026256] kthread+0x117/0x130
> [ 317.026264] ? kthread_create_worker_on_cpu+0x70/0x70
> [ 317.026275] ret_from_fork+0x35/0x40
> --
>
> Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
> ---
> drivers/infiniband/core/iwcm.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/drivers/infiniband/core/iwcm.c b/drivers/infiniband/core/iwcm.c
> index 72141c5b7c95..94566271dbff 100644
> --- a/drivers/infiniband/core/iwcm.c
> +++ b/drivers/infiniband/core/iwcm.c
> @@ -427,7 +427,9 @@ static void destroy_cm_id(struct iw_cm_id *cm_id)
> break;
> }
> if (cm_id_priv->qp) {
> + spin_unlock_irqrestore(&cm_id_priv->lock, flags);
> cm_id_priv->id.device->ops.iw_rem_ref(cm_id_priv->qp);
> + spin_lock_irqsave(&cm_id_priv->lock, flags);
> cm_id_priv->qp = NULL;
Shouldn't you first do cm_id_priv->qp = NULL and only then
unlock and destroy the qp?
Mark
> }
> spin_unlock_irqrestore(&cm_id_priv->lock, flags);
>
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH] iwcm: don't hold the irq disabled lock on iw_rem_ref
2019-09-03 19:33 ` Mark Bloch
@ 2019-09-04 1:07 ` Sagi Grimberg
2019-09-04 6:03 ` Jason Gunthorpe
1 sibling, 0 replies; 4+ messages in thread
From: Sagi Grimberg @ 2019-09-04 1:07 UTC (permalink / raw)
To: Mark Bloch, linux-rdma; +Cc: Jason Gunthorpe
>> This may be the final put on a qp and result in freeing
>> resourcesand should not be done with interrupts disabled.
>>
>> Produce the following warning:
>> --
>> [ 317.026048] WARNING: CPU: 1 PID: 443 at kernel/smp.c:425 smp_call_function_many+0xa0/0x260
>> [ 317.026131] Call Trace:
>> [ 317.026159] ? load_new_mm_cr3+0xe0/0xe0
>> [ 317.026161] on_each_cpu+0x28/0x50
>> [ 317.026183] __purge_vmap_area_lazy+0x72/0x150
>> [ 317.026200] free_vmap_area_noflush+0x7a/0x90
>> [ 317.026202] remove_vm_area+0x6f/0x80
>> [ 317.026203] __vunmap+0x71/0x210
>> [ 317.026211] siw_free_qp+0x8d/0x130 [siw]
>> [ 317.026217] destroy_cm_id+0xc3/0x200 [iw_cm]
>> [ 317.026222] rdma_destroy_id+0x224/0x2b0 [rdma_cm]
>> [ 317.026226] nvme_rdma_reset_ctrl_work+0x2c/0x70 [nvme_rdma]
>> [ 317.026235] process_one_work+0x1f4/0x3e0
>> [ 317.026249] worker_thread+0x221/0x3e0
>> [ 317.026252] ? process_one_work+0x3e0/0x3e0
>> [ 317.026256] kthread+0x117/0x130
>> [ 317.026264] ? kthread_create_worker_on_cpu+0x70/0x70
>> [ 317.026275] ret_from_fork+0x35/0x40
>> --
>>
>> Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
>> ---
>> drivers/infiniband/core/iwcm.c | 2 ++
>> 1 file changed, 2 insertions(+)
>>
>> diff --git a/drivers/infiniband/core/iwcm.c b/drivers/infiniband/core/iwcm.c
>> index 72141c5b7c95..94566271dbff 100644
>> --- a/drivers/infiniband/core/iwcm.c
>> +++ b/drivers/infiniband/core/iwcm.c
>> @@ -427,7 +427,9 @@ static void destroy_cm_id(struct iw_cm_id *cm_id)
>> break;
>> }
>> if (cm_id_priv->qp) {
>> + spin_unlock_irqrestore(&cm_id_priv->lock, flags);
>> cm_id_priv->id.device->ops.iw_rem_ref(cm_id_priv->qp);
>> + spin_lock_irqsave(&cm_id_priv->lock, flags);
>> cm_id_priv->qp = NULL;
>
> Shouldn't you first do cm_id_priv->qp = NULL and only then
> unlock and destroy the qp?
Probably...
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH] iwcm: don't hold the irq disabled lock on iw_rem_ref
2019-09-03 19:33 ` Mark Bloch
2019-09-04 1:07 ` Sagi Grimberg
@ 2019-09-04 6:03 ` Jason Gunthorpe
1 sibling, 0 replies; 4+ messages in thread
From: Jason Gunthorpe @ 2019-09-04 6:03 UTC (permalink / raw)
To: Mark Bloch; +Cc: Sagi Grimberg, linux-rdma
On Tue, Sep 03, 2019 at 07:33:08PM +0000, Mark Bloch wrote:
>
>
> On 9/3/19 12:22 PM, Sagi Grimberg wrote:
> > This may be the final put on a qp and result in freeing
> > resourcesand should not be done with interrupts disabled.
> >
> > Produce the following warning:
>
> Shouldn't you first do cm_id_priv->qp = NULL and only then
> unlock and destroy the qp?
>
> Mark
> > }
> > spin_unlock_irqrestore(&cm_id_priv->lock, flags);
Would avoid the sketchy unlock too.. But does it work?
Jason
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2019-09-04 6:03 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-09-03 19:22 [PATCH] iwcm: don't hold the irq disabled lock on iw_rem_ref Sagi Grimberg
2019-09-03 19:33 ` Mark Bloch
2019-09-04 1:07 ` Sagi Grimberg
2019-09-04 6:03 ` Jason Gunthorpe
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).