linux-rdma.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [v2,for-rc] RDMA/qedr: qedr crash while running rdma-tool
@ 2021-10-23 16:45 Alok Prasad
  2021-10-24 10:46 ` Kamal Heib
  0 siblings, 1 reply; 4+ messages in thread
From: Alok Prasad @ 2021-10-23 16:45 UTC (permalink / raw)
  To: jgg, dledford
  Cc: michal.kalderon, ariel.elior, linux-rdma, smalin, aelior, palok,
	alok.prasad7, Michal Kalderon

This patch fixes crash caused by querying qp.
Also corrects the state of gsi qp.

Below call trace is generated while using iproute2 utility
"rdma res show -dd qp" on rdma interface.
==========================================================================
[  302.569794] BUG: kernel NULL pointer dereference, address: 0000000000000034
..
[  302.570378] Hardware name: Dell Inc. PowerEdge R720/0M1GCR, BIOS 1.2.6 05/10/2012
[  302.570500] RIP: 0010:qed_rdma_query_qp+0x33/0x1a0 [qed]
[  302.570861] RSP: 0018:ffffba560a08f580 EFLAGS: 00010206
[  302.570979] RAX: 0000000200000000 RBX: ffffba560a08f5b8 RCX: 0000000000000000
[  302.571100] RDX: ffffba560a08f5b8 RSI: 0000000000000000 RDI: ffff9807ee458090
[  302.571221] RBP: ffffba560a08f5a0 R08: 0000000000000000 R09: ffff9807890e7048
[  302.571342] R10: ffffba560a08f658 R11: 0000000000000000 R12: 0000000000000000
[  302.571462] R13: ffff9807ee458090 R14: ffff9807f0afb000 R15: ffffba560a08f7ec
[  302.571583] FS:  00007fbbf8bfe740(0000) GS:ffff980aafa00000(0000) knlGS:0000000000000000
[  302.571729] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  302.571847] CR2: 0000000000000034 CR3: 00000001720ba001 CR4: 00000000000606f0
[  302.571968] Call Trace:
[  302.572083]  qedr_query_qp+0x82/0x360 [qedr]
[  302.572211]  ib_query_qp+0x34/0x40 [ib_core]
[  302.572361]  ? ib_query_qp+0x34/0x40 [ib_core]
[  302.572503]  fill_res_qp_entry_query.isra.26+0x47/0x1d0 [ib_core]
[  302.572670]  ? __nla_put+0x20/0x30
[  302.572788]  ? nla_put+0x33/0x40
[  302.572901]  fill_res_qp_entry+0xe3/0x120 [ib_core]
[  302.573058]  res_get_common_dumpit+0x3f8/0x5d0 [ib_core]
[  302.573213]  ? fill_res_cm_id_entry+0x1f0/0x1f0 [ib_core]
[  302.573377]  nldev_res_get_qp_dumpit+0x1a/0x20 [ib_core]
[  302.573529]  netlink_dump+0x156/0x2f0
[  302.573648]  __netlink_dump_start+0x1ab/0x260
[  302.573765]  rdma_nl_rcv+0x1de/0x330 [ib_core]
[  302.573918]  ? nldev_res_get_cm_id_dumpit+0x20/0x20 [ib_core]
[  302.574074]  netlink_unicast+0x1b8/0x270
[  302.574191]  netlink_sendmsg+0x33e/0x470
[  302.574307]  sock_sendmsg+0x63/0x70
[  302.574421]  __sys_sendto+0x13f/0x180
[  302.574536]  ? setup_sgl.isra.12+0x70/0xc0
[  302.574655]  __x64_sys_sendto+0x28/0x30
[  302.574769]  do_syscall_64+0x3a/0xb0
[  302.574884]  entry_SYSCALL_64_after_hwframe+0x44/0xae
==========================================================================
Signed-off-by: Ariel Elior <aelior@marvell.com>
Signed-off-by: Michal Kalderon <mkalderon@marvell.com>
Signed-off-by: Shai Malin <smalin@marvell.com>
Signed-off-by: Alok Prasad <palok@marvell.com>
---
v2 (from [1]):
	- Change description.
	- Corrected enum type.
[1] https://patchwork.kernel.org/project/linux-rdma/patch/20210821074339.16614-1-palok@marvell.com/
---
 drivers/infiniband/hw/qedr/verbs.c | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)

diff --git a/drivers/infiniband/hw/qedr/verbs.c b/drivers/infiniband/hw/qedr/verbs.c
index dcb3653db72d..85baa4f730df 100644
--- a/drivers/infiniband/hw/qedr/verbs.c
+++ b/drivers/infiniband/hw/qedr/verbs.c
@@ -2744,15 +2744,20 @@ int qedr_query_qp(struct ib_qp *ibqp,
 	int rc = 0;
 
 	memset(&params, 0, sizeof(params));
+	memset(qp_attr, 0, sizeof(*qp_attr));
+	memset(qp_init_attr, 0, sizeof(*qp_init_attr));
 
-	rc = dev->ops->rdma_query_qp(dev->rdma_ctx, qp->qed_qp, &params);
+	if (qp->qed_qp)
+		rc = dev->ops->rdma_query_qp(dev->rdma_ctx,
+					     qp->qed_qp, &params);
 	if (rc)
 		goto err;
 
-	memset(qp_attr, 0, sizeof(*qp_attr));
-	memset(qp_init_attr, 0, sizeof(*qp_init_attr));
+	if (qp->qp_type == IB_QPT_GSI)
+		qp_attr->qp_state = qedr_get_ibqp_state(QED_ROCE_QP_STATE_RTS);
+	else
+		qp_attr->qp_state = qedr_get_ibqp_state(params.state);
 
-	qp_attr->qp_state = qedr_get_ibqp_state(params.state);
 	qp_attr->cur_qp_state = qedr_get_ibqp_state(params.state);
 	qp_attr->path_mtu = ib_mtu_int_to_enum(params.mtu);
 	qp_attr->path_mig_state = IB_MIG_MIGRATED;
-- 
2.17.1


^ permalink raw reply related	[flat|nested] 4+ messages in thread

* Re: [v2,for-rc] RDMA/qedr: qedr crash while running rdma-tool
  2021-10-23 16:45 [v2,for-rc] RDMA/qedr: qedr crash while running rdma-tool Alok Prasad
@ 2021-10-24 10:46 ` Kamal Heib
  2021-10-27 17:00   ` Jason Gunthorpe
  0 siblings, 1 reply; 4+ messages in thread
From: Kamal Heib @ 2021-10-24 10:46 UTC (permalink / raw)
  To: Alok Prasad
  Cc: michal.kalderon, ariel.elior, linux-rdma, smalin, aelior,
	alok.prasad7, Michal Kalderon, dledford, jgg



On 10/23/21 19:45, Alok Prasad wrote:
> This patch fixes crash caused by querying qp.
> Also corrects the state of gsi qp.
> 
> Below call trace is generated while using iproute2 utility
> "rdma res show -dd qp" on rdma interface.
> ==========================================================================
> [  302.569794] BUG: kernel NULL pointer dereference, address: 0000000000000034
> ..
> [  302.570378] Hardware name: Dell Inc. PowerEdge R720/0M1GCR, BIOS 1.2.6 05/10/2012
> [  302.570500] RIP: 0010:qed_rdma_query_qp+0x33/0x1a0 [qed]
> [  302.570861] RSP: 0018:ffffba560a08f580 EFLAGS: 00010206
> [  302.570979] RAX: 0000000200000000 RBX: ffffba560a08f5b8 RCX: 0000000000000000
> [  302.571100] RDX: ffffba560a08f5b8 RSI: 0000000000000000 RDI: ffff9807ee458090
> [  302.571221] RBP: ffffba560a08f5a0 R08: 0000000000000000 R09: ffff9807890e7048
> [  302.571342] R10: ffffba560a08f658 R11: 0000000000000000 R12: 0000000000000000
> [  302.571462] R13: ffff9807ee458090 R14: ffff9807f0afb000 R15: ffffba560a08f7ec
> [  302.571583] FS:  00007fbbf8bfe740(0000) GS:ffff980aafa00000(0000) knlGS:0000000000000000
> [  302.571729] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [  302.571847] CR2: 0000000000000034 CR3: 00000001720ba001 CR4: 00000000000606f0
> [  302.571968] Call Trace:
> [  302.572083]  qedr_query_qp+0x82/0x360 [qedr]
> [  302.572211]  ib_query_qp+0x34/0x40 [ib_core]
> [  302.572361]  ? ib_query_qp+0x34/0x40 [ib_core]
> [  302.572503]  fill_res_qp_entry_query.isra.26+0x47/0x1d0 [ib_core]
> [  302.572670]  ? __nla_put+0x20/0x30
> [  302.572788]  ? nla_put+0x33/0x40
> [  302.572901]  fill_res_qp_entry+0xe3/0x120 [ib_core]
> [  302.573058]  res_get_common_dumpit+0x3f8/0x5d0 [ib_core]
> [  302.573213]  ? fill_res_cm_id_entry+0x1f0/0x1f0 [ib_core]
> [  302.573377]  nldev_res_get_qp_dumpit+0x1a/0x20 [ib_core]
> [  302.573529]  netlink_dump+0x156/0x2f0
> [  302.573648]  __netlink_dump_start+0x1ab/0x260
> [  302.573765]  rdma_nl_rcv+0x1de/0x330 [ib_core]
> [  302.573918]  ? nldev_res_get_cm_id_dumpit+0x20/0x20 [ib_core]
> [  302.574074]  netlink_unicast+0x1b8/0x270
> [  302.574191]  netlink_sendmsg+0x33e/0x470
> [  302.574307]  sock_sendmsg+0x63/0x70
> [  302.574421]  __sys_sendto+0x13f/0x180
> [  302.574536]  ? setup_sgl.isra.12+0x70/0xc0
> [  302.574655]  __x64_sys_sendto+0x28/0x30
> [  302.574769]  do_syscall_64+0x3a/0xb0
> [  302.574884]  entry_SYSCALL_64_after_hwframe+0x44/0xae
> ==========================================================================
> Signed-off-by: Ariel Elior <aelior@marvell.com>
> Signed-off-by: Michal Kalderon <mkalderon@marvell.com>
> Signed-off-by: Shai Malin <smalin@marvell.com>
> Signed-off-by: Alok Prasad <palok@marvell.com>
> ---
> v2 (from [1]):
> 	- Change description.
> 	- Corrected enum type.
> [1] https://patchwork.kernel.org/project/linux-rdma/patch/20210821074339.16614-1-palok@marvell.com/
> ---
>   drivers/infiniband/hw/qedr/verbs.c | 13 +++++++++----
>   1 file changed, 9 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/infiniband/hw/qedr/verbs.c b/drivers/infiniband/hw/qedr/verbs.c
> index dcb3653db72d..85baa4f730df 100644
> --- a/drivers/infiniband/hw/qedr/verbs.c
> +++ b/drivers/infiniband/hw/qedr/verbs.c
> @@ -2744,15 +2744,20 @@ int qedr_query_qp(struct ib_qp *ibqp,
>   	int rc = 0;
>   
>   	memset(&params, 0, sizeof(params));
> +	memset(qp_attr, 0, sizeof(*qp_attr));
> +	memset(qp_init_attr, 0, sizeof(*qp_init_attr));
>   
> -	rc = dev->ops->rdma_query_qp(dev->rdma_ctx, qp->qed_qp, &params);
> +	if (qp->qed_qp)

I suggest to use "if (qp->qp_type != IB_QPT_GSI)" to match the handling 
of GSI QPs in the QEDR driver.

Thanks,
Kamal

> +		rc = dev->ops->rdma_query_qp(dev->rdma_ctx,
> +					     qp->qed_qp, &params);
>   	if (rc)
>   		goto err;
>   
> -	memset(qp_attr, 0, sizeof(*qp_attr));
> -	memset(qp_init_attr, 0, sizeof(*qp_init_attr));
> +	if (qp->qp_type == IB_QPT_GSI)
> +		qp_attr->qp_state = qedr_get_ibqp_state(QED_ROCE_QP_STATE_RTS);
> +	else
> +		qp_attr->qp_state = qedr_get_ibqp_state(params.state);
>   
> -	qp_attr->qp_state = qedr_get_ibqp_state(params.state);
>   	qp_attr->cur_qp_state = qedr_get_ibqp_state(params.state);
>   	qp_attr->path_mtu = ib_mtu_int_to_enum(params.mtu);
>   	qp_attr->path_mig_state = IB_MIG_MIGRATED;
> 


^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [v2,for-rc] RDMA/qedr: qedr crash while running rdma-tool
  2021-10-24 10:46 ` Kamal Heib
@ 2021-10-27 17:00   ` Jason Gunthorpe
  2021-10-27 18:45     ` [EXT] " Alok Prasad
  0 siblings, 1 reply; 4+ messages in thread
From: Jason Gunthorpe @ 2021-10-27 17:00 UTC (permalink / raw)
  To: Kamal Heib
  Cc: Alok Prasad, michal.kalderon, ariel.elior, linux-rdma, smalin,
	aelior, alok.prasad7, Michal Kalderon, dledford

On Sun, Oct 24, 2021 at 01:46:03PM +0300, Kamal Heib wrote:

> > diff --git a/drivers/infiniband/hw/qedr/verbs.c b/drivers/infiniband/hw/qedr/verbs.c
> > index dcb3653db72d..85baa4f730df 100644
> > +++ b/drivers/infiniband/hw/qedr/verbs.c
> > @@ -2744,15 +2744,20 @@ int qedr_query_qp(struct ib_qp *ibqp,
> >   	int rc = 0;
> >   	memset(&params, 0, sizeof(params));
> > +	memset(qp_attr, 0, sizeof(*qp_attr));
> > +	memset(qp_init_attr, 0, sizeof(*qp_init_attr));
> > -	rc = dev->ops->rdma_query_qp(dev->rdma_ctx, qp->qed_qp, &params);
> > +	if (qp->qed_qp)
> 
> I suggest to use "if (qp->qp_type != IB_QPT_GSI)" to match the handling of
> GSI QPs in the QEDR driver.

Alok? Time is closing to get this in before the merge window.

Jason

^ permalink raw reply	[flat|nested] 4+ messages in thread

* RE: [EXT] Re: [v2,for-rc] RDMA/qedr: qedr crash while running rdma-tool
  2021-10-27 17:00   ` Jason Gunthorpe
@ 2021-10-27 18:45     ` Alok Prasad
  0 siblings, 0 replies; 4+ messages in thread
From: Alok Prasad @ 2021-10-27 18:45 UTC (permalink / raw)
  To: Jason Gunthorpe, Kamal Heib
  Cc: Michal Kalderon, Ariel Elior, linux-rdma, Shai Malin,
	Ariel Elior, alok.prasad7, Michal Kalderon, dledford


> -----Original Message-----
> From: Jason Gunthorpe <jgg@nvidia.com>
> Sent: 27 October 2021 22:31
> To: Kamal Heib <kheib@redhat.com>
> Cc: Alok Prasad <palok@marvell.com>; Michal Kalderon <mkalderon@marvell.com>; Ariel Elior
> <aelior@marvell.com>; linux-rdma@vger.kernel.org; Shai Malin <smalin@marvell.com>; Ariel
> Elior <aelior@marvell.com>; alok.prasad7@gmail.com; Michal Kalderon
> <mkalderon@marvell.com>; dledford@redhat.com
> Subject: [EXT] Re: [v2,for-rc] RDMA/qedr: qedr crash while running rdma-tool
> 
> External Email
> 
> ----------------------------------------------------------------------
> On Sun, Oct 24, 2021 at 01:46:03PM +0300, Kamal Heib wrote:
> 
> > > diff --git a/drivers/infiniband/hw/qedr/verbs.c b/drivers/infiniband/hw/qedr/verbs.c
> > > index dcb3653db72d..85baa4f730df 100644
> > > +++ b/drivers/infiniband/hw/qedr/verbs.c
> > > @@ -2744,15 +2744,20 @@ int qedr_query_qp(struct ib_qp *ibqp,
> > >   	int rc = 0;
> > >   	memset(&params, 0, sizeof(params));
> > > +	memset(qp_attr, 0, sizeof(*qp_attr));
> > > +	memset(qp_init_attr, 0, sizeof(*qp_init_attr));
> > > -	rc = dev->ops->rdma_query_qp(dev->rdma_ctx, qp->qed_qp, &params);
> > > +	if (qp->qed_qp)
> >
> > I suggest to use "if (qp->qp_type != IB_QPT_GSI)" to match the handling of
> > GSI QPs in the QEDR driver.
> 
> Alok? Time is closing to get this in before the merge window.
> 
> Jason

Done now with v3, Sorry for delaying on this.

- Alok

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2021-10-27 18:46 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-10-23 16:45 [v2,for-rc] RDMA/qedr: qedr crash while running rdma-tool Alok Prasad
2021-10-24 10:46 ` Kamal Heib
2021-10-27 17:00   ` Jason Gunthorpe
2021-10-27 18:45     ` [EXT] " Alok Prasad

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).