DPDK patches and discussions
 help / color / mirror / Atom feed
* [dpdk-dev] Are there considerations about resource recycling by the process, not by the OS?
@ 2014-09-05  3:07 Zhangkun (K)
  2014-09-05  3:17 ` Masaru Oki
  0 siblings, 1 reply; 5+ messages in thread
From: Zhangkun (K) @ 2014-09-05  3:07 UTC (permalink / raw)
  To: dev

Hi,
   I develop the program with the use of dpdk and the process is core dump when the program is started.
The process is crash in case of initing memory failed and calling rte_panic function.
 I search all rte_panic function in the dpdk, and find it so many. Therefore, in case the pr ocess happen fail,
it will abort. The resource allocated before is not released by the process, but by the OS system.
So I have a question: Are there considerations about resource recycling by the process, not by the OS ?

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [dpdk-dev] Are there considerations about resource recycling by the process, not by the OS?
  2014-09-05  3:07 [dpdk-dev] Are there considerations about resource recycling by the process, not by the OS? Zhangkun (K)
@ 2014-09-05  3:17 ` Masaru Oki
  2014-09-05  3:39   ` Zhangkun (K)
  0 siblings, 1 reply; 5+ messages in thread
From: Masaru Oki @ 2014-09-05  3:17 UTC (permalink / raw)
  To: Zhangkun (K); +Cc: dev

Hi,

you may try
 $ gdb your_program
 gdb> b rte_panic
 gdb> run
 stopped at rte_panic
 gdb> bt

sorry, it is not answer for your question.
by the way, do you run the program as root?



2014-09-05 12:07 GMT+09:00 Zhangkun (K) <zhang.zhangkun@huawei.com>:

> Hi,
>    I develop the program with the use of dpdk and the process is core dump
> when the program is started.
> The process is crash in case of initing memory failed and calling
> rte_panic function.
>  I search all rte_panic function in the dpdk, and find it so many.
> Therefore, in case the pr ocess happen fail,
> it will abort. The resource allocated before is not released by the
> process, but by the OS system.
> So I have a question: Are there considerations about resource recycling by
> the process, not by the OS ?
>

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [dpdk-dev] Are there considerations about resource recycling by the process, not by the OS?
  2014-09-05  3:17 ` Masaru Oki
@ 2014-09-05  3:39   ` Zhangkun (K)
  2014-09-05  8:18     ` Richardson, Bruce
  2014-09-05  9:32     ` Tetsuya.Mukawa
  0 siblings, 2 replies; 5+ messages in thread
From: Zhangkun (K) @ 2014-09-05  3:39 UTC (permalink / raw)
  To: Masaru Oki; +Cc: dev

Hi,
    Yes, the program is running as root.
    I want to know if or not the DPDK Community considers about resource recycling.

From: Masaru Oki [mailto:m-oki@stratosphere.co.jp]
Date: 2014年9月5日 11:18
To: Zhangkun (K)
Cc: dev@dpdk.org
Subject: Re: [dpdk-dev] Are there considerations about resource recycling by the process, not by the OS?

Hi,

you may try
 $ gdb your_program
 gdb> b rte_panic
 gdb> run
 stopped at rte_panic
 gdb> bt

sorry, it is not answer for your question.
by the way, do you run the program as root?


2014-09-05 12:07 GMT+09:00 Zhangkun (K) <zhang.zhangkun@huawei.com<mailto:zhang.zhangkun@huawei.com>>:
Hi,
   I develop the program with the use of dpdk and the process is core dump when the program is started.
The process is crash in case of initing memory failed and calling rte_panic function.
 I search all rte_panic function in the dpdk, and find it so many. Therefore, in case the pr ocess happen fail,
it will abort. The resource allocated before is not released by the process, but by the OS system.
So I have a question: Are there considerations about resource recycling by the process, not by the OS ?


^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [dpdk-dev] Are there considerations about resource recycling by the process, not by the OS?
  2014-09-05  3:39   ` Zhangkun (K)
@ 2014-09-05  8:18     ` Richardson, Bruce
  2014-09-05  9:32     ` Tetsuya.Mukawa
  1 sibling, 0 replies; 5+ messages in thread
From: Richardson, Bruce @ 2014-09-05  8:18 UTC (permalink / raw)
  To: Zhangkun (K), Masaru Oki; +Cc: dev

Can you perhaps send on the log and the traceback from the panic call. From the error message we may be able to tell you why your process is failing.

/Bruce

> -----Original Message-----
> From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Zhangkun (K)
> Sent: Friday, September 05, 2014 4:40 AM
> To: Masaru Oki
> Cc: dev@dpdk.org
> Subject: Re: [dpdk-dev] Are there considerations about resource recycling by the
> process, not by the OS?
> 
> Hi,
>     Yes, the program is running as root.
>     I want to know if or not the DPDK Community considers about resource
> recycling.
> 
> From: Masaru Oki [mailto:m-oki@stratosphere.co.jp]
> Date: 2014年9月5日 11:18
> To: Zhangkun (K)
> Cc: dev@dpdk.org
> Subject: Re: [dpdk-dev] Are there considerations about resource recycling by the
> process, not by the OS?
> 
> Hi,
> 
> you may try
>  $ gdb your_program
>  gdb> b rte_panic
>  gdb> run
>  stopped at rte_panic
>  gdb> bt
> 
> sorry, it is not answer for your question.
> by the way, do you run the program as root?
> 
> 
> 2014-09-05 12:07 GMT+09:00 Zhangkun (K)
> <zhang.zhangkun@huawei.com<mailto:zhang.zhangkun@huawei.com>>:
> Hi,
>    I develop the program with the use of dpdk and the process is core dump when
> the program is started.
> The process is crash in case of initing memory failed and calling rte_panic
> function.
>  I search all rte_panic function in the dpdk, and find it so many. Therefore, in
> case the pr ocess happen fail,
> it will abort. The resource allocated before is not released by the process, but by
> the OS system.
> So I have a question: Are there considerations about resource recycling by the
> process, not by the OS ?


^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [dpdk-dev] Are there considerations about resource recycling by the process, not by the OS?
  2014-09-05  3:39   ` Zhangkun (K)
  2014-09-05  8:18     ` Richardson, Bruce
@ 2014-09-05  9:32     ` Tetsuya.Mukawa
  1 sibling, 0 replies; 5+ messages in thread
From: Tetsuya.Mukawa @ 2014-09-05  9:32 UTC (permalink / raw)
  To: Zhangkun (K); +Cc: dev

Hi Zhangkun,

Do you talk about resource management of DPDK when a process is finished?
If so, I guess DPDK doesn't care so much about freeing resources once
allocated by the process.
For example, there is no function to free a mempool.

Thanks,
Tetsuya

(2014/09/05 12:39), Zhangkun (K) wrote:
> Hi,
>     Yes, the program is running as root.
>     I want to know if or not the DPDK Community considers about resource recycling.
>
> From: Masaru Oki [mailto:m-oki@stratosphere.co.jp]
> Date: 2014年9月5日 11:18
> To: Zhangkun (K)
> Cc: dev@dpdk.org
> Subject: Re: [dpdk-dev] Are there considerations about resource recycling by the process, not by the OS?
>
> Hi,
>
> you may try
>  $ gdb your_program
>  gdb> b rte_panic
>  gdb> run
>  stopped at rte_panic
>  gdb> bt
>
> sorry, it is not answer for your question.
> by the way, do you run the program as root?
>
>
> 2014-09-05 12:07 GMT+09:00 Zhangkun (K) <zhang.zhangkun@huawei.com<mailto:zhang.zhangkun@huawei.com>>:
> Hi,
>    I develop the program with the use of dpdk and the process is core dump when the program is started.
> The process is crash in case of initing memory failed and calling rte_panic function.
>  I search all rte_panic function in the dpdk, and find it so many. Therefore, in case the pr ocess happen fail,
> it will abort. The resource allocated before is not released by the process, but by the OS system.
> So I have a question: Are there considerations about resource recycling by the process, not by the OS ?
>

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2014-09-05  9:27 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2014-09-05  3:07 [dpdk-dev] Are there considerations about resource recycling by the process, not by the OS? Zhangkun (K)
2014-09-05  3:17 ` Masaru Oki
2014-09-05  3:39   ` Zhangkun (K)
2014-09-05  8:18     ` Richardson, Bruce
2014-09-05  9:32     ` Tetsuya.Mukawa

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).