* [bug] rdma-core/librdmacm/cma.c rdma_create_qp_ex:add id->cq refer to qp_attr->cq, when qp_attr->cq is not NULL, and id->cq is NULL
@ 2024-08-27 6:20 靳文宾
2024-08-28 0:51 ` Zhijian Li (Fujitsu)
0 siblings, 1 reply; 5+ messages in thread
From: 靳文宾 @ 2024-08-27 6:20 UTC (permalink / raw)
To: linux-rdma
I think the following modification is more correct,
if there have something i dont know, pls tell me, thx
diff --git a/librdmacm/cma.c b/librdmacm/cma.c
index 7b924bd0d..9e71ba858 100644
--- a/librdmacm/cma.c
+++ b/librdmacm/cma.c
@@ -1654,10 +1654,20 @@ int rdma_create_qp_ex(struct rdma_cm_id *id,
if (ret)
return ret;
- if (!attr->send_cq)
+ if (!attr->send_cq) {
attr->send_cq = id->send_cq;
- if (!attr->recv_cq)
+ } else {
+ if (!id->recv_cq)
+ id->recv_cq = attr->recv_cq;
+ }
+ if (!attr->recv_cq) {
attr->recv_cq = id->recv_cq;
+ } else {
+ if (!id->recv_cq)
+ id->recv_cq = attr->recv_cq;
+ }
+
+
if (id->srq && !attr->srq)
attr->srq = id->srq;
qp = ibv_create_qp_ex(id->verbs, attr);
^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [bug] rdma-core/librdmacm/cma.c rdma_create_qp_ex:add id->cq refer to qp_attr->cq, when qp_attr->cq is not NULL, and id->cq is NULL
2024-08-27 6:20 [bug] rdma-core/librdmacm/cma.c rdma_create_qp_ex:add id->cq refer to qp_attr->cq, when qp_attr->cq is not NULL, and id->cq is NULL 靳文宾
@ 2024-08-28 0:51 ` Zhijian Li (Fujitsu)
2024-08-29 9:08 ` Leon Romanovsky
2024-08-29 12:04 ` [bug] rdma-core/librdmacm/cma.c rdma_create_qp_exï¼add " Zhu Yanjun
0 siblings, 2 replies; 5+ messages in thread
From: Zhijian Li (Fujitsu) @ 2024-08-28 0:51 UTC (permalink / raw)
To: 靳文宾, linux-rdma@vger.kernel.org
Not related to your modification,
In general,
- Please report a bug with more details, such as how to reproduce, your expectations and what's the impact, etc...
- Please follow the patch submit instruction[1] or a PR if you want to submit a patch to the communities.
[1] https://docs.kernel.org/process/submitting-patches.html
On 27/08/2024 14:20, 靳文宾 wrote:
>
>
>
>
>
> I think the following modification is more correct,
> if there have something i dont know, pls tell me, thx
>
>
>
>
>
> diff --git a/librdmacm/cma.c b/librdmacm/cma.c
> index 7b924bd0d..9e71ba858 100644
> --- a/librdmacm/cma.c
> +++ b/librdmacm/cma.c
> @@ -1654,10 +1654,20 @@ int rdma_create_qp_ex(struct rdma_cm_id *id,
> if (ret)
> return ret;
>
>
> - if (!attr->send_cq)
> + if (!attr->send_cq) {
> attr->send_cq = id->send_cq;
> - if (!attr->recv_cq)
> + } else {
> + if (!id->recv_cq)
> + id->recv_cq = attr->recv_cq;
> + }
> + if (!attr->recv_cq) {
> attr->recv_cq = id->recv_cq;
> + } else {
> + if (!id->recv_cq)
> + id->recv_cq = attr->recv_cq;
> + }
> +
> +
> if (id->srq && !attr->srq)
> attr->srq = id->srq;
> qp = ibv_create_qp_ex(id->verbs, attr);
>
>
>
>
>
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [bug] rdma-core/librdmacm/cma.c rdma_create_qp_ex:add id->cq refer to qp_attr->cq, when qp_attr->cq is not NULL, and id->cq is NULL
2024-08-28 0:51 ` Zhijian Li (Fujitsu)
@ 2024-08-29 9:08 ` Leon Romanovsky
2024-08-29 12:04 ` [bug] rdma-core/librdmacm/cma.c rdma_create_qp_exï¼add " Zhu Yanjun
1 sibling, 0 replies; 5+ messages in thread
From: Leon Romanovsky @ 2024-08-29 9:08 UTC (permalink / raw)
To: Zhijian Li (Fujitsu); +Cc: 靳文宾, linux-rdma@vger.kernel.org
On Wed, Aug 28, 2024 at 12:51:30AM +0000, Zhijian Li (Fujitsu) wrote:
> Not related to your modification,
>
> In general,
> - Please report a bug with more details, such as how to reproduce, your expectations and what's the impact, etc...
> - Please follow the patch submit instruction[1] or a PR if you want to submit a patch to the communities.
>
> [1] https://docs.kernel.org/process/submitting-patches.html
+1
Thanks
>
>
> On 27/08/2024 14:20, 靳文宾 wrote:
> >
> >
> >
> >
> >
> > I think the following modification is more correct,
> > if there have something i dont know, pls tell me, thx
> >
> >
> >
> >
> >
> > diff --git a/librdmacm/cma.c b/librdmacm/cma.c
> > index 7b924bd0d..9e71ba858 100644
> > --- a/librdmacm/cma.c
> > +++ b/librdmacm/cma.c
> > @@ -1654,10 +1654,20 @@ int rdma_create_qp_ex(struct rdma_cm_id *id,
> > if (ret)
> > return ret;
> >
> >
> > - if (!attr->send_cq)
> > + if (!attr->send_cq) {
> > attr->send_cq = id->send_cq;
> > - if (!attr->recv_cq)
> > + } else {
> > + if (!id->recv_cq)
> > + id->recv_cq = attr->recv_cq;
> > + }
> > + if (!attr->recv_cq) {
> > attr->recv_cq = id->recv_cq;
> > + } else {
> > + if (!id->recv_cq)
> > + id->recv_cq = attr->recv_cq;
> > + }
> > +
> > +
> > if (id->srq && !attr->srq)
> > attr->srq = id->srq;
> > qp = ibv_create_qp_ex(id->verbs, attr);
> >
> >
> >
> >
> >
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [bug] rdma-core/librdmacm/cma.c rdma_create_qp_exï¼add id->cq refer to qp_attr->cq, when qp_attr->cq is not NULL, and id->cq is NULL
2024-08-28 0:51 ` Zhijian Li (Fujitsu)
2024-08-29 9:08 ` Leon Romanovsky
@ 2024-08-29 12:04 ` Zhu Yanjun
2024-09-02 7:00 ` Leon Romanovsky
1 sibling, 1 reply; 5+ messages in thread
From: Zhu Yanjun @ 2024-08-29 12:04 UTC (permalink / raw)
To: Zhijian Li (Fujitsu), 靳文宾,
linux-rdma@vger.kernel.org
在 2024/8/28 8:51, Zhijian Li (Fujitsu) 写道:
> Not related to your modification,
>
> In general,
> - Please report a bug with more details, such as how to reproduce, your expectations and what's the impact, etc...
> - Please follow the patch submit instruction[1] or a PR if you want to submit a patch to the communities.
>
> [1] https://docs.kernel.org/process/submitting-patches.html
It seems a rdma-core problem. Not sure if it is reasonable to use the
above link.
Zhu Yanjun
>
>
> On 27/08/2024 14:20, 靳文宾 wrote:
>>
>>
>>
>>
>>
>> I think the following modification is more correct,
>> if there have something i dont know, pls tell me, thx
>>
>>
>>
>>
>>
>> diff --git a/librdmacm/cma.c b/librdmacm/cma.c
>> index 7b924bd0d..9e71ba858 100644
>> --- a/librdmacm/cma.c
>> +++ b/librdmacm/cma.c
>> @@ -1654,10 +1654,20 @@ int rdma_create_qp_ex(struct rdma_cm_id *id,
>> if (ret)
>> return ret;
>>
>>
>> - if (!attr->send_cq)
>> + if (!attr->send_cq) {
>> attr->send_cq = id->send_cq;
>> - if (!attr->recv_cq)
>> + } else {
>> + if (!id->recv_cq)
>> + id->recv_cq = attr->recv_cq;
>> + }
>> + if (!attr->recv_cq) {
>> attr->recv_cq = id->recv_cq;
>> + } else {
>> + if (!id->recv_cq)
>> + id->recv_cq = attr->recv_cq;
>> + }
>> +
>> +
>> if (id->srq && !attr->srq)
>> attr->srq = id->srq;
>> qp = ibv_create_qp_ex(id->verbs, attr);
>>
>>
>>
>>
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [bug] rdma-core/librdmacm/cma.c rdma_create_qp_exï¼add id->cq refer to qp_attr->cq, when qp_attr->cq is not NULL, and id->cq is NULL
2024-08-29 12:04 ` [bug] rdma-core/librdmacm/cma.c rdma_create_qp_exï¼add " Zhu Yanjun
@ 2024-09-02 7:00 ` Leon Romanovsky
0 siblings, 0 replies; 5+ messages in thread
From: Leon Romanovsky @ 2024-09-02 7:00 UTC (permalink / raw)
To: Zhu Yanjun
Cc: Zhijian Li (Fujitsu), 靳文宾,
linux-rdma@vger.kernel.org
On Thu, Aug 29, 2024 at 08:04:55PM +0800, Zhu Yanjun wrote:
> 在 2024/8/28 8:51, Zhijian Li (Fujitsu) 写道:
> > Not related to your modification,
> >
> > In general,
> > - Please report a bug with more details, such as how to reproduce, your expectations and what's the impact, etc...
> > - Please follow the patch submit instruction[1] or a PR if you want to submit a patch to the communities.
> >
> > [1] https://docs.kernel.org/process/submitting-patches.html
>
> It seems a rdma-core problem. Not sure if it is reasonable to use the above
> link.
It is, https://github.com/linux-rdma/rdma-core/blob/master/Documentation/contributing.md
# Contributing to rdma-core
rdma-core is a userspace project for a Linux kernel interface and follows many
of the same expectations as contributing to the Linux kernel:
- One change per patch
Carefully describe your change in the commit message and break up work into
appropriate reviewable commits.
Refer to [Linux Kernel Submitting Patches](https://github.com/torvalds/linux/blob/master/Documentation/process/submitting-patches.rst)
for general information.
- Developer Certificate of Origin 'Signed-off-by' lines
Include a Signed-off-by line to indicate your submission is suitable
licensed and you have the legal authority to make this submission
and accept the [DCO](#developers-certificate-of-origin-11)
- Broadly follow the [Linux Kernel coding style](https://github.com/torvalds/linux/blob/master/Documentation/process/coding-style.rst)
As in the Linux Kernel, commits that are fixing bugs should be marked with a
Fixes: line to help backporting.
....
>
> Zhu Yanjun
>
> >
> >
> > On 27/08/2024 14:20, 靳文宾 wrote:
> > >
> > >
> > >
> > >
> > >
> > > I think the following modification is more correct,
> > > if there have something i dont know, pls tell me, thx
> > >
> > >
> > >
> > >
> > >
> > > diff --git a/librdmacm/cma.c b/librdmacm/cma.c
> > > index 7b924bd0d..9e71ba858 100644
> > > --- a/librdmacm/cma.c
> > > +++ b/librdmacm/cma.c
> > > @@ -1654,10 +1654,20 @@ int rdma_create_qp_ex(struct rdma_cm_id *id,
> > > if (ret)
> > > return ret;
> > >
> > >
> > > - if (!attr->send_cq)
> > > + if (!attr->send_cq) {
> > > attr->send_cq = id->send_cq;
> > > - if (!attr->recv_cq)
> > > + } else {
> > > + if (!id->recv_cq)
> > > + id->recv_cq = attr->recv_cq;
> > > + }
> > > + if (!attr->recv_cq) {
> > > attr->recv_cq = id->recv_cq;
> > > + } else {
> > > + if (!id->recv_cq)
> > > + id->recv_cq = attr->recv_cq;
> > > + }
> > > +
> > > +
> > > if (id->srq && !attr->srq)
> > > attr->srq = id->srq;
> > > qp = ibv_create_qp_ex(id->verbs, attr);
> > >
> > >
> > >
> > >
>
>
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2024-09-02 7:00 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2024-08-27 6:20 [bug] rdma-core/librdmacm/cma.c rdma_create_qp_ex:add id->cq refer to qp_attr->cq, when qp_attr->cq is not NULL, and id->cq is NULL 靳文宾
2024-08-28 0:51 ` Zhijian Li (Fujitsu)
2024-08-29 9:08 ` Leon Romanovsky
2024-08-29 12:04 ` [bug] rdma-core/librdmacm/cma.c rdma_create_qp_exï¼add " Zhu Yanjun
2024-09-02 7:00 ` Leon Romanovsky
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox