On 4/27/2014 3:37 PM, Chuck Lever wrote:
Why not first create a new id+qp and assign them - and then destroy the old
id+qp?
see SRP related section: ib_srp.x:srp_create_target_ib()
Anyway it is indeed important to guarantee that no xmit flows happens
concurrently to that,
and cleanups are p
On Apr 27, 2014, at 6:12 AM, Sagi Grimberg wrote:
> On 4/24/2014 6:01 PM, Chuck Lever wrote:
>> On Apr 24, 2014, at 3:12 AM, Sagi Grimberg wrote:
>>
>>> On 4/24/2014 2:30 AM, Devesh Sharma wrote:
Hi Chuck
Following is the complete call trace of a typical NFS-RDMA transaction
>
On 4/24/2014 6:01 PM, Chuck Lever wrote:
On Apr 24, 2014, at 3:12 AM, Sagi Grimberg wrote:
On 4/24/2014 2:30 AM, Devesh Sharma wrote:
Hi Chuck
Following is the complete call trace of a typical NFS-RDMA transaction while
mounting a share.
It is unavoidable to stop calling post-send in case i
half Of Chuck Lever
>> Sent: Thursday, April 24, 2014 8:31 PM
>> To: Sagi Grimberg
>> Cc: Devesh Sharma; Linux NFS Mailing List; linux-rdma@vger.kernel.org;
>> Trond Myklebust
>> Subject: Re: [PATCH V1] NFS-RDMA: fix qp pointer validation checks
>>
>>
>> O
rg
> Cc: Devesh Sharma; Linux NFS Mailing List; linux-rdma@vger.kernel.org;
> Trond Myklebust
> Subject: Re: [PATCH V1] NFS-RDMA: fix qp pointer validation checks
>
>
> On Apr 24, 2014, at 3:12 AM, Sagi Grimberg
> wrote:
>
> > On 4/24/2014 2:30 AM, Devesh Sharma wr
On Apr 24, 2014, at 3:12 AM, Sagi Grimberg wrote:
> On 4/24/2014 2:30 AM, Devesh Sharma wrote:
>> Hi Chuck
>>
>> Following is the complete call trace of a typical NFS-RDMA transaction while
>> mounting a share.
>> It is unavoidable to stop calling post-send in case it is not created.
>> There
On 4/24/2014 2:30 AM, Devesh Sharma wrote:
Hi Chuck
Following is the complete call trace of a typical NFS-RDMA transaction while
mounting a share.
It is unavoidable to stop calling post-send in case it is not created.
Therefore, applying checks to the connection state is a must
While registeri
vent rep
88046230ef30 status 0 opcode 7 length 48
--->>>>> New Task Initialised<<<<<<<<<---
Apr 23 20:00:34 neo03-el64 kernel: RPC: new task initialized, procpid 3491
> -----Original Message-
> From: linux-rdma-ow...
> -Original Message-
> From: Chuck Lever [mailto:chuck.le...@oracle.com]
> Sent: Tuesday, April 15, 2014 6:10 AM
> To: Devesh Sharma
> Cc: Linux NFS Mailing List; linux-rdma@vger.kernel.org; Trond Myklebust
> Subject: Re: [PATCH V1] NFS-RDMA: fix qp pointer validation
rnel.org; Trond Myklebust
>> Subject: Re: [PATCH V1] NFS-RDMA: fix qp pointer validation checks
>>
>> Hi Devesh-
>>
>>
>> On Apr 13, 2014, at 12:01 AM, Chuck Lever wrote:
>>
>>>
>>> On Apr 11, 2014, at 7:51 PM, Devesh Sharma
>>
pc]
> > <4> [] ? printk+0x41/0x48 <4> [] ?
> > bit_waitqueue+0x17/0xd0 <4> []
> rpc_execute+0x61/0xa0
> > [sunrpc] <4> [] rpc_run_task+0x75/0x90 [sunrpc] <4>
> > [] rpc_call_sync+0x42/0x70 [sunrpc] <4>
> > [] ? handle_pte_fault+0x487/0xb5
Hi Chuck
> -Original Message-
> From: Chuck Lever [mailto:chuck.le...@oracle.com]
> Sent: Tuesday, April 15, 2014 2:24 AM
> To: Devesh Sharma
> Cc: Linux NFS Mailing List; linux-rdma@vger.kernel.org; Trond Myklebust
> Subject: Re: [PATCH V1] NFS-RDMA: fix qp pointer
Hi Devesh-
On Apr 13, 2014, at 12:01 AM, Chuck Lever wrote:
>
> On Apr 11, 2014, at 7:51 PM, Devesh Sharma wrote:
>
>> Hi Chuck,
>> Yes that is the case, Following is the trace I got.
>>
>> <4>RPC: 355 setting alarm for 6 ms
>> <4>RPC: 355 sync task going to sleep
>> <4>RPC:
> <4> [] vfs_getattr+0x51/0x80
> <4> [] vfs_fstatat+0x60/0x80
> <4> [] vfs_stat+0x1b/0x20
> <4> [] sys_newstat+0x24/0x50
> <4> [] ? audit_syscall_entry+0x1d7/0x200
> <4> [] system_call_fastpath+0x16/0x1b
> <4>Code: 48 89 85 78 ff ff ff
15/0x20 [sunrpc] <4>
> > [] ? rpcauth_lookupcred+0x70/0xc0 [sunrpc] <4>
> > [] nfs_permission+0xa8/0x1e0 [nfs] <4>
> > [] __link_path_walk+0xad/0x1030 <4>
> > [] ? handle_pte_fault+0x487/0xb50 <4>
> > [] ? copy_termios+0x6a/0x80 <4&g
ystem_call_fastpath+0x16/0x1b
> <4>Code: 48 89 85 78 ff ff ff 48 8b 40 08 8b 40 1c 89 45 94 b8 ff ff ff ff f0
> 41 0f c1 85 e0 05 00 00 49 8b 04 24 48 8d 55 d8 48 8b 78 10 <48> 8b 07 ff 90
> b0 01 00 00 85 c0 89 c3 74 09 80 3d 56 8d 05 00
> <1>RIP [] rpcrdma_deregister_frmr
> -Original Message-
> From: Chuck Lever [mailto:chuck.le...@oracle.com]
> Sent: Thursday, April 10, 2014 12:44 PM
> To: Steve Wise
> Cc: Devesh Sharma; Linux NFS Mailing List; linux-rdma@vger.kernel.org; Trond
> Myklebust
> Subject: Re: [PATCH V1] NFS-RDMA: fix
x1b
<4>Code: 48 89 85 78 ff ff ff 48 8b 40 08 8b 40 1c 89 45 94 b8 ff ff ff ff f0
41 0f c1 85 e0 05 00 00 49 8b 04 24 48 8d 55 d8 48 8b 78 10 <48> 8b 07 ff 90 b0
01 00 00 85 c0 89 c3 74 09 80 3d 56 8d 05 00
<1>RIP [] rpcrdma_deregister_frmr_external+0x9c/0xe0
[xprtrdma]
&
On Apr 10, 2014, at 1:42 PM, Devesh Sharma wrote:
>> However it seems to me the new (!ia->ri_id->qp) checks outside the connect
>> logic are unnecessary.
>>
>> Clearly, as you noticed, the ib_post_{send,recv} verbs do not check that
>> their
>> "qp" argument is NULL before dereferencing it.
>>
post_send(). Thus it is correct to DECR_CQCOUNT
>> twice in that case because each WR will trigger a separate completion event.
>>
>>
>>> -Original Message-
>>> From: Chuck Lever [mailto:chuck.le...@oracle.com]
>>> Sent: Thursday, April 10, 2
> -Original Message-
> From: Chuck Lever [mailto:chuck.le...@oracle.com]
> Sent: Thursday, April 10, 2014 5:56 AM
> To: Devesh Sharma
> Cc: Linux NFS Mailing List; linux-rdma@vger.kernel.org; Trond Myklebust
> Subject: Re: [PATCH V1] NFS-RDMA: fix qp pointer validation
Message-
From: Chuck Lever [mailto:chuck.le...@oracle.com]
Sent: Thursday, April 10, 2014 1:57 AM
To: Devesh Sharma
Cc: Linux NFS Mailing List; linux-rdma@vger.kernel.org; Trond Myklebust
Subject: Re: [PATCH V1] NFS-RDMA: fix qp pointer validation checks
On Apr 9, 2014, at 4:22 PM, Trond Myklebust
wr
CQCOUNT twice
in that case because each WR will trigger a separate completion event.
> -Original Message-
> From: Chuck Lever [mailto:chuck.le...@oracle.com]
> Sent: Thursday, April 10, 2014 1:57 AM
> To: Devesh Sharma
> Cc: Linux NFS Mailing List; linux-rdma@vger.kerne
Trond Myklebust
Subject: Re: [PATCH V1] NFS-RDMA: fix qp pointer validation checks
On Apr 9, 2014, at 4:22 PM, Trond Myklebust
wrote:
> Hi Devesh,
>
> This looks a lot better. I still have a couple of small suggestions, though.
>
> On Apr 9, 2014, at 14:40, Devesh Sharma
On Apr 9, 2014, at 4:22 PM, Trond Myklebust
wrote:
> Hi Devesh,
>
> This looks a lot better. I still have a couple of small suggestions, though.
>
> On Apr 9, 2014, at 14:40, Devesh Sharma wrote:
>
>> If the rdma_create_qp fails to create qp due to device firmware being in
>> invalid state
Hi Devesh,
This looks a lot better. I still have a couple of small suggestions, though.
On Apr 9, 2014, at 14:40, Devesh Sharma wrote:
> If the rdma_create_qp fails to create qp due to device firmware being in
> invalid state
> xprtrdma still tries to destroy the non-existant qp and ends up in
If the rdma_create_qp fails to create qp due to device firmware being in
invalid state
xprtrdma still tries to destroy the non-existant qp and ends up in a NULL
pointer reference
crash.
Adding proper checks for vaidating QP pointer avoids this to happen.
Signed-off-by: Devesh Sharma
---
net/su
27 matches
Mail list logo
