Re: [dpdk-users] Integration from Dpdk18.05 to Dpdk19.11 - rte_timer_subsystem_init(void)

[Carrillo, Erik G]

Hi,

Response in-line:

> -Original Message-
> From: David Marchand 
> Sent: Monday, January 18, 2021 2:42 AM
> To: Li, Jiu (NSB - CN/Hangzhou) ; Robert Sanford
> ; Carrillo, Erik G 
> Cc: users@dpdk.org; Dong, Shaojie (NSB - CN/Hangzhou)
> ; Ye, Hua (NSB - CN/Hangzhou)
> ; Alasaarela, Risto (Nokia - FI/Oulu)
> ; Jarvelaid, Rain (Nokia - FI/Oulu)
> ; Wu, Jianyue (NSB - CN/Hangzhou)
> ; Pan, Jianyong (NSB - CN/Hangzhou)
> 
> Subject: Re: [dpdk-users] Integration from Dpdk18.05 to Dpdk19.11 -
> rte_timer_subsystem_init(void)
> 
> On Sun, Jan 17, 2021 at 11:04 PM Li, Jiu (NSB - CN/Hangzhou)  sbell.com> wrote:
> >
> > Hello!  Dpdk experts,
> >
> > On Dpdk 18.05,  void rte_timer_subsystem_init(void)
> > On Dpdk 19.11,  intrte_timer_subsystem_init(void) implementation
> changed, which will return 0, -EALREADY or -ENOMEM;
> >
> > There is still have  dpdk "process" mode (instead of pdkd thread mode)
> deployment in my side.
> > Can I have a question?
> >
> > If rte_timer_subsystem_init() is called one time is enough?
> > After rte_timer_subsystem_init() called with return 0 by one process , then
> other processes are able to use "rte timer" service without issue, right?
> >

rte_timer_subsystem_init() needs to be called in each process that wishes to 
create and manage timers. 

Note that secondary processes need to be using distinct lcores, as described 
here:
http://doc.dpdk.org/guides-19.11/prog_guide/multi_proc_support.html#multi-process-limitations

"All DPDK processes running as a single application and using shared memory 
must have distinct coremask/corelist arguments. It is not possible to have a 
primary and secondary instance, or two secondary instances, using any of the 
same logical cores. Attempting to do so can cause corruption of memory pool 
caches, among other issues."

Regards,
Erik

> 
> Copied timer library maintainers.
> 
> 
> --
> David Marchand

Re: [dpdk-users] Integration from Dpdk18.05 to Dpdk19.11 - rte_timer_subsystem_init(void)

[David Marchand]

On Sun, Jan 17, 2021 at 11:04 PM Li, Jiu (NSB - CN/Hangzhou)
 wrote:
>
> Hello!  Dpdk experts,
>
> On Dpdk 18.05,  void rte_timer_subsystem_init(void)
> On Dpdk 19.11,  intrte_timer_subsystem_init(void) implementation changed, 
> which will return 0, -EALREADY or -ENOMEM;
>
> There is still have  dpdk "process" mode (instead of pdkd thread mode) 
> deployment in my side.
> Can I have a question?
>
> If rte_timer_subsystem_init() is called one time is enough?
> After rte_timer_subsystem_init() called with return 0 by one process , then 
> other processes are able to use "rte timer" service without issue, right?
>

Copied timer library maintainers.


-- 
David Marchand

[dpdk-users] Integration from Dpdk18.05 to Dpdk19.11 - rte_timer_subsystem_init(void)

[Li, Jiu (NSB - CN/Hangzhou)]

Hello!  Dpdk experts,

On Dpdk 18.05,  void rte_timer_subsystem_init(void)
On Dpdk 19.11,  intrte_timer_subsystem_init(void) implementation changed, 
which will return 0, -EALREADY or -ENOMEM;

There is still have  dpdk "process" mode (instead of pdkd thread mode) 
deployment in my side.
Can I have a question?

If rte_timer_subsystem_init() is called one time is enough?
After rte_timer_subsystem_init() called with return 0 by one process , then 
other processes are able to use "rte timer" service without issue, right?

Thanks in advance!

Best Regards.
  U-Plane
(Clark),  Jiu LI