[oe] (no subject)

Mark Asselstine mark.asselstine at windriver.com
Tue Jul 31 14:30:02 UTC 2018


On Tuesday, July 31, 2018 7:54:22 AM EDT William Delacre wrote:
> Le ven. 27 juil. 2018 à 18:40, S. Lockwood-Childs <sjl at vctlabs.com> a
> 
> écrit :
> > It depends on the role of the particular dependency:
> > 
> > * tool you need to run during build             <-- native dependency
> > * library you need to link against during build <-- target dependency
> 
> Thanks for the precision :-)
> 
> My question may seem a bit stupid but how do you know which is which ?
> Let’s say I want to write a recipe to install any specific software on my
> target, how do I choose the dependencies ?
> I have to refer to the software Makefile/CMakeList ?

It is not always easy to see via code inspection. I would suggest keep a 
second build for a different arch than your build host, possibly ARM. Start by 
using the non-native DEPENDS and only if there is an issue with the ARM build 
look at using -native for a DEPENDS.

To be honest I haven't had to worry about this too much as it is rare to run 
into issues.

Mark

> 
> William
> 
> > --SJLC
> > 
> > On Fri, Jul 27, 2018 at 06:25:27PM +0200, William Delacre wrote:
> > > Okay, it makes sense!
> > > 
> > > My question may seem stupid but then why don’t we always use ‘-native’
> > 
> > for
> > 
> > > the dependencies ?
> > > 
> > > Is it because sometimes we know that the behavior of ‘foo’ and
> > 
> > ‘foo-native’
> > 
> > > would be the same, independently of the architecture ?
> > > 
> > > William
> > > 
> > > Le ven. 27 juil. 2018 à 18:16, Mark Asselstine <
> > > 
> > > mark.asselstine at windriver.com> a écrit :
> > > > On Friday, July 27, 2018 12:04:37 PM EDT William Delacre wrote:
> > > > > Thanks a lot for the answer, it worked perfectly !
> > > > > 
> > > > > There is just one point I don’t really understand : why do we need
> > > > > to
> > > > 
> > > > add a
> > > > 
> > > > > dependency to ‘python-incremental-NATIVE’ ?
> > > > > 
> > > > > What would happen if I just add to the python-twisted recipe :
> > > > > 
> > > > > DEPENDS += “${PYTHON_PN}-incremental“
> > > > 
> > > > This would assume that python is completely arch independent, which it
> > 
> > is
> > 
> > > > not.
> > > > Assume for example python-incremental had some C extensions, using
> > 
> > DEPENDS
> > 
> > > > would build them for the target arch yet possibly attempt to run them
> > 
> > on
> > 
> > > > your
> > > > build host. -native takes care of this.
> > > > 
> > > > Some may say this is a bit of a hack since the difference in ARCH
> > 
> > *might*
> > 
> > > > equate to a difference in behavior. But I have dealt with many python
> > > > recipes
> > > > and have yet to run into any scenario where this potential issue
> > 
> > exists.
> > 
> > > > At least that is what I can recall at the moment. Unfortunately I have
> > > > forgotten some of the subtleties that might be in play.
> > > > 
> > > > Mark
> > > > 
> > > > > William
> > > > > 
> > > > > Le ven. 27 juil. 2018 à 17:03, Mark Asselstine <
> > > > > 
> > > > > mark.asselstine at windriver.com> a écrit :
> > > > > > On Thu, Jul 26, 2018 at 3:40 PM, Mark Asselstine
> > > > > > 
> > > > > > <mark.asselstine at windriver.com> wrote:
> > > > > > > On Mon, Jul 23, 2018 at 8:28 AM, William Delacre
> > > > > > > 
> > > > > > > <williamdelacre at gmail.com> wrote:
> > > > > > >> Hi ! :-)
> > > > > > >> 
> > > > > > >> I’m facing issues when trying to run ‘bitbake python-twisted’
> > > > > > 
> > > > > > (do_compile
> > > > > > 
> > > > > > >> error)
> > > > > > >> 
> > > > > > >> Log says :
> > > > 
> > > > > > >> “Download error on https://pypi.python.org/simple/incremental/:
> > > > [Errno
> > > > 
> > > > > > -3]
> > > > > > 
> > > > > > >> Temporary failure in name resolution
> > > > > > 
> > > > > > With the additional info you sent me along with a second look at
> > 
> > the
> > 
> > > > > > error I can fairly confidently say I know what the issue is.
> > > > > > 
> > > > > > Python, like some other languages (ruby, go...) will attempt to
> > > > > > download dependencies/requirements if they are not present. This
> > > > > > is
> > > > > > done "outside" of the bitbake system and thus will ignore
> > > > > > BB_NO_NETWORK and is generally a bad thing. It isn't quite host
> > > > > > contamination but it is very similar. What we do with our python
> > > > > > recipes in bitbake is review a python package's source setup.py,
> > > > > > requirements.txt...files in order to determine what other python
> > > > > > packages are needed, add them as RDEPENDS or similar in order to
> > 
> > avoid
> > 
> > > > > > the python attempting to complete a download such as this.
> > > > > > 
> > > > > > In this case python-twisted 'requires' python-incremental but the
> > > > > > Rocko version of the python-twisted recipe lacks the needed
> > 
> > assignment
> > 
> > > > > > to ensure this is available in the recipe sysroot and thus the
> > > > > > attempted download.
> > > > > > 
> > > > > > Cherry pick commit b08b570c8624303acbb6bb064a7bda466a679df4
> > > > > > [python-twisted: avoid downloading build dependencies] to Rocko
> > > > > > and
> > > > > > you will be good to go. You should also follow up this thread with
> > 
> > a
> > 
> > > > > > request to have this commit put onto Rocko to avoid others hitting
> > 
> > the
> > 
> > > > > > same issue.
> > > > > > 
> > > > > > Mark
> > > > > > 
> > > > > > >> ............
> > > > > > >> 
> > > > > > >> distutils.errors.DistutilsError : Could not find suitable
> > > > 
> > > > distribution
> > > > 
> > > > > > for
> > > > > > 
> > > > > > >> Requirement.parse(‘incremental>=16.10.1’)”
> > > > > > >> 
> > > > > > >> First of all I’m not connected to internet, so I’ve
> > 
> > pre-downloaded
> > 
> > > > all
> > > > 
> > > > > > the
> > > > > > 
> > > > > > >> required packages (twisted-17.9.0.tar.bz2 &
> > > > 
> > > > incremental-17.5.0.tar.gz
> > > > 
> > > > > > are
> > > > > > 
> > > > > > >> present in my build/downloads folder)
> > > > > > >> 
> > > > > > >> It’s the first time I have some “downloading issues” and I’m
> > 
> > stuck
> > 
> > > > on
> > > > 
> > > > > > >> it
> > > > > > >> for a few hours now..
> > > > > > >> 
> > > > > > >> Any tips for me ?
> > > > > > > 
> > > > > > > Check all aspects of what you have manually downloaded and
> > 
> > populated
> > 
> > > > > > > in your downloads. My guess is that a checksum or some other
> > 
> > detail
> > 
> > > > > > > isn't matching and it is rejecting what you have and thus
> > 
> > attempting
> > 
> > > > > > > to re-download the file.
> > > > > > > 
> > > > > > > Mark
> > > > > > > 
> > > > > > >> Thanks in advance !
> > > > > > >> --
> > > > > > >> _______________________________________________
> > > > > > >> Openembedded-devel mailing list
> > > > > > >> Openembedded-devel at lists.openembedded.org
> > 
> > http://lists.openembedded.org/mailman/listinfo/openembedded-devel
> > 
> > > --
> > > _______________________________________________
> > > Openembedded-devel mailing list
> > > Openembedded-devel at lists.openembedded.org
> > > http://lists.openembedded.org/mailman/listinfo/openembedded-devel







More information about the Openembedded-devel mailing list