All of lore.kernel.org
 help / color / mirror / Atom feed
From: Richard Purdie <richard.purdie@linuxfoundation.org>
To: Piotr Lewicki <piotr.lewicki@elfin.de>,
	"Burton, Ross" <ross.burton@intel.com>
Cc: OE-core <openembedded-core@lists.openembedded.org>
Subject: Re: recipe using environment variable does not refresh / rebuild
Date: Thu, 01 Feb 2018 15:09:51 +0000	[thread overview]
Message-ID: <1517497791.3090.50.camel@linuxfoundation.org> (raw)
In-Reply-To: <a7dc8b5e-b277-27c7-b879-4405c1bb04ac@elfin.de>

On Thu, 2018-02-01 at 16:00 +0100, Piotr Lewicki wrote:
> Hi Ross,
> Good idea, but I'm using gitlab to perform autobuilds and I want to
> set the variable there (the "ADD_UBOOT_TO_UPDATE" variable) so that
> it's propagated to the image recipe.
> I'm want to differentiate the image with and without u-boot only by
> setting this one environment variable and without additional mangling
> with local.conf.
> 
> If that's not possible- I'll use your idea with setting it in
> local.conf, but before that I'm trying to understand- why my recipe
> is not rebuilt?

It isn't rebuilt as bitbake isn't psychic!

Bitbake doesn't understand what anonymous python fragments do, so it
can't know that ADD_UBOOT_TO_UPDATE is being changed.

I have a suspicion that it might rebuild, *if* it knew it had to
reparse the recipe since it probably does know which tasks use that
variable. It doesn't know you're poking into BB_ORIGENV which you
really shouldn't be doing.

To make this work as you'd want, you could add ADD_UBOOT_TO_UPDATE to
BB_PRESERVE_ENV. That might be enough to cause it to re-parse each time
you change the value and in turn, if the tasks change based on that,
they might rebuild. I'd at least start there (and stop poking into
BB_ORIGENV).

Cheers,

Richard




      reply	other threads:[~2018-02-01 15:09 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-02-01 14:46 recipe using environment variable does not refresh / rebuild Piotr Lewicki
2018-02-01 14:50 ` Burton, Ross
2018-02-01 15:00   ` Piotr Lewicki
2018-02-01 15:09     ` Richard Purdie [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1517497791.3090.50.camel@linuxfoundation.org \
    --to=richard.purdie@linuxfoundation.org \
    --cc=openembedded-core@lists.openembedded.org \
    --cc=piotr.lewicki@elfin.de \
    --cc=ross.burton@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
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.