public inbox for isar-users@googlegroups.com
 help / color / mirror / Atom feed
* bitbake multiconfig pipeline serialization
@ 2018-02-05 13:38 Alexander Smirnov
  2018-02-05 13:41 ` Jan Kiszka
  0 siblings, 1 reply; 2+ messages in thread
From: Alexander Smirnov @ 2018-02-05 13:38 UTC (permalink / raw)
  To: isar-users

Hi all,

I've noticed some undesired behavior in bitbake using multiconfig - it 
never runs the same recipe:task in parallel for different configurations.

For example:
  - You are going to build 'arm-jessie' and 'x86-jessie'.
  - Task buildchroot:do_rootfs will be serialized, for example 'arm' 
will be completed first and *only* then 'x86' will be started.

I see no technical issues to run them in parallel in Isar, because we 
have completely split multiconfigs in our 'tmp/work' directory.

I've briefly checked bitbake 'git log' and found some commits that could 
somehow related to this serialization:

  - 937acf267fa9e45f538695b2cf8aa83232a96240
  - a9041fc96a14e718c0c1d1676e705343b9e872d3

So if somebody knows more details about this issue, I'll be happy to see 
them here :-)

-- 
With best regards,
Alexander Smirnov

ilbers GmbH
Baierbrunner Str. 28c
D-81379 Munich
+49 (89) 122 67 24-0
http://ilbers.de/
Commercial register Munich, HRB 214197
General manager: Baurzhan Ismagulov

^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: bitbake multiconfig pipeline serialization
  2018-02-05 13:38 bitbake multiconfig pipeline serialization Alexander Smirnov
@ 2018-02-05 13:41 ` Jan Kiszka
  0 siblings, 0 replies; 2+ messages in thread
From: Jan Kiszka @ 2018-02-05 13:41 UTC (permalink / raw)
  To: Alexander Smirnov, isar-users, Larson, Chris

On 2018-02-05 14:38, Alexander Smirnov wrote:
> Hi all,
> 
> I've noticed some undesired behavior in bitbake using multiconfig - it
> never runs the same recipe:task in parallel for different configurations.
> 
> For example:
>  - You are going to build 'arm-jessie' and 'x86-jessie'.
>  - Task buildchroot:do_rootfs will be serialized, for example 'arm' will
> be completed first and *only* then 'x86' will be started.
> 
> I see no technical issues to run them in parallel in Isar, because we
> have completely split multiconfigs in our 'tmp/work' directory.
> 
> I've briefly checked bitbake 'git log' and found some commits that could
> somehow related to this serialization:
> 
>  - 937acf267fa9e45f538695b2cf8aa83232a96240
>  - a9041fc96a14e718c0c1d1676e705343b9e872d3
> 
> So if somebody knows more details about this issue, I'll be happy to see
> them here :-)
> 

Chris, can you enlighten us?

Jan

-- 
Siemens AG, Corporate Technology, CT RDA IOT SES-DE
Corporate Competence Center Embedded Linux

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2018-02-05 13:41 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-02-05 13:38 bitbake multiconfig pipeline serialization Alexander Smirnov
2018-02-05 13:41 ` Jan Kiszka

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox