All of lore.kernel.org
 help / color / mirror / Atom feed
* Removing do_compile task and ordering problem
@ 2015-05-21 10:47 Marcin Krzemiński
  2015-05-21 12:55 ` Paul Eggleton
  0 siblings, 1 reply; 4+ messages in thread
From: Marcin Krzemiński @ 2015-05-21 10:47 UTC (permalink / raw)
  To: yocto

[-- Attachment #1: Type: text/plain, Size: 524 bytes --]

Hi,

I am writing recipe that inherits from *native*.
I removed do_compile task using: *deltask do_compile*
When i was ruing bitbake my-recipe all works fine, but when I added recipe
to *EXTRA_IMAGEDEPENDS *and run* bitbake core-image-minimal *tasks were
reordered in some strange way that task do_install was performed before
do_fetch.
When I hanged *deltask do_compile *to *do_compile[noexec] = "1" * all went
back to normal.
I found *deltask *in manual so I do not know, is it a bug or not?

Regards
Marcin

[-- Attachment #2: Type: text/html, Size: 694 bytes --]

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: Removing do_compile task and ordering problem
  2015-05-21 10:47 Removing do_compile task and ordering problem Marcin Krzemiński
@ 2015-05-21 12:55 ` Paul Eggleton
  2015-05-21 14:20   ` mar.krzeminski
  2015-05-21 14:29   ` Richard Purdie
  0 siblings, 2 replies; 4+ messages in thread
From: Paul Eggleton @ 2015-05-21 12:55 UTC (permalink / raw)
  To: Marcin Krzemiński; +Cc: yocto

Hi Marcin,

On Thursday 21 May 2015 12:47:05 Marcin Krzemiński wrote:
> I am writing recipe that inherits from *native*.
> I removed do_compile task using: *deltask do_compile*
> When i was ruing bitbake my-recipe all works fine, but when I added recipe
> to *EXTRA_IMAGEDEPENDS *and run* bitbake core-image-minimal *tasks were
> reordered in some strange way that task do_install was performed before
> do_fetch.
> When I hanged *deltask do_compile *to *do_compile[noexec] = "1" * all went
> back to normal.
> I found *deltask *in manual so I do not know, is it a bug or not?

I suspect this is because all deltask does is delete the task, it does not 
reconnect the dependency chain such that tasks that depended on the deleted 
task would depend on the tasks that the deleted task depended upon, so what 
you get in this case is do_install no longer having any dependencies. To be 
honest the current behaviour seems reasonable to me, though it is not 
immediately obvious and ought to be described in the manual. I think in your 
situation you really do want to set the noexec flag rather than using deltask.

Cheers,
Paul

-- 

Paul Eggleton
Intel Open Source Technology Centre


^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: Removing do_compile task and ordering problem
  2015-05-21 12:55 ` Paul Eggleton
@ 2015-05-21 14:20   ` mar.krzeminski
  2015-05-21 14:29   ` Richard Purdie
  1 sibling, 0 replies; 4+ messages in thread
From: mar.krzeminski @ 2015-05-21 14:20 UTC (permalink / raw)
  To: Paul Eggleton; +Cc: yocto

[-- Attachment #1: Type: text/plain, Size: 1410 bytes --]

After reading manual I assumed that /deltask/ will reorder the dependences.
If this is expected behaviour it is fine for me :)
I use /noexec/ flag then.

Regards,
Marcin


W dniu 21.05.2015 o 14:55, Paul Eggleton pisze:
> Hi Marcin,
>
> On Thursday 21 May 2015 12:47:05 Marcin Krzemiński wrote:
>> I am writing recipe that inherits from *native*.
>> I removed do_compile task using: *deltask do_compile*
>> When i was ruing bitbake my-recipe all works fine, but when I added recipe
>> to *EXTRA_IMAGEDEPENDS *and run* bitbake core-image-minimal *tasks were
>> reordered in some strange way that task do_install was performed before
>> do_fetch.
>> When I hanged *deltask do_compile *to *do_compile[noexec] = "1" * all went
>> back to normal.
>> I found *deltask *in manual so I do not know, is it a bug or not?
> I suspect this is because all deltask does is delete the task, it does not
> reconnect the dependency chain such that tasks that depended on the deleted
> task would depend on the tasks that the deleted task depended upon, so what
> you get in this case is do_install no longer having any dependencies. To be
> honest the current behaviour seems reasonable to me, though it is not
> immediately obvious and ought to be described in the manual. I think in your
> situation you really do want to set the noexec flag rather than using deltask.
>
> Cheers,
> Paul
>


[-- Attachment #2: Type: text/html, Size: 1936 bytes --]

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: Removing do_compile task and ordering problem
  2015-05-21 12:55 ` Paul Eggleton
  2015-05-21 14:20   ` mar.krzeminski
@ 2015-05-21 14:29   ` Richard Purdie
  1 sibling, 0 replies; 4+ messages in thread
From: Richard Purdie @ 2015-05-21 14:29 UTC (permalink / raw)
  To: Paul Eggleton; +Cc: yocto

On Thu, 2015-05-21 at 13:55 +0100, Paul Eggleton wrote:
> Hi Marcin,
> 
> On Thursday 21 May 2015 12:47:05 Marcin Krzemiński wrote:
> > I am writing recipe that inherits from *native*.
> > I removed do_compile task using: *deltask do_compile*
> > When i was ruing bitbake my-recipe all works fine, but when I added recipe
> > to *EXTRA_IMAGEDEPENDS *and run* bitbake core-image-minimal *tasks were
> > reordered in some strange way that task do_install was performed before
> > do_fetch.
> > When I hanged *deltask do_compile *to *do_compile[noexec] = "1" * all went
> > back to normal.
> > I found *deltask *in manual so I do not know, is it a bug or not?
> 
> I suspect this is because all deltask does is delete the task, it does not 
> reconnect the dependency chain such that tasks that depended on the deleted 
> task would depend on the tasks that the deleted task depended upon, so what 
> you get in this case is do_install no longer having any dependencies. To be 
> honest the current behaviour seems reasonable to me, though it is not 
> immediately obvious and ought to be described in the manual. I think in your 
> situation you really do want to set the noexec flag rather than using deltask.

Correct, this is basically the difference between deltask and noexec.
noexec has the behaviour you want in this case and we should clarify in
the manual.

Cheers,

Richard





^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2015-05-21 14:30 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2015-05-21 10:47 Removing do_compile task and ordering problem Marcin Krzemiński
2015-05-21 12:55 ` Paul Eggleton
2015-05-21 14:20   ` mar.krzeminski
2015-05-21 14:29   ` Richard Purdie

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.