From: Henning Schild <henning.schild@siemens.com>
To: Jan Kiszka <jan.kiszka@siemens.com>
Cc: "Maxim Yu. Osipov" <mosipov@ilbers.de>,
isar-users <isar-users@googlegroups.com>
Subject: Re: SDK feedback
Date: Wed, 12 Sep 2018 13:02:38 +0200 [thread overview]
Message-ID: <20180912130238.1d94e306@md1pvb1c.ad001.siemens.net> (raw)
In-Reply-To: <99c2c7b3-5cfd-659f-1bcd-5ce8a5b932ef@siemens.com>
Am Wed, 12 Sep 2018 12:15:40 +0200
schrieb Jan Kiszka <jan.kiszka@siemens.com>:
> On 12.09.18 09:42, Maxim Yu. Osipov wrote:
> > Hi Jan,
> >
> > On 09/11/2018 08:57 PM, Jan Kiszka wrote:
> >> On 11.09.18 19:55, Jan Kiszka wrote:
> >>> Hi Maksim,
> >>>
> >>> finally, I actually used the SDK on an ARM target, and it worked
> >>> nicely! One question and one feature request, though:
> >>>
> >>> I did not do the chroot dance and rather looked for a way to use
> >>> the cross compiler directly. And it /seems/ to work, at least for
> >>> the trivial case:
> >>>
> >>> export LD_LIBRARY_PATH=/path-to-sdk/usr/lib/x86_64-linux-gnu/
> >>> sdk/usr/bin/arm-linux-gnueabihf-gcc --sysroot /path-to-sdk \
> >>> -o hello hello.c
> >>>
> >>> Does this make sense, at least for the classic "I need a
> >>> crosscompiler and the corresponding target libs" case? When will
> >>> it start to break?
> >
> > Honestly I always played with chroot and never tried
> > LD_LIBRARY_PATH approach in more complicated case, so I can't say
> > which problems may pop up.
>
> Henning, maybe once worth a try on some of the recent examples we now
> have (just for testing, they are normally built as part of a custom
> Isar layer).
People that use that layer found their own solution. They just take the
buildchroot as a sysroot for the cross-compiler of their distro. No
need to have a debian cross compiler or chroot. That has its
limitations, but is also really simple and seems to work well
enough ... so far.
Henning
> >> Forgot to state: my host was a SUSE 42.2.
> >
> > Nice to hear that it works under SUSE too
> > (I only tried sdk on Fedora box).
If you chrooted, you used Debian ;).
Henning
> >>
> >>>
> >>>
> >>> And then I noticed that the deployed SDK image could be a bit
> >>> slimmer, specifically by purging /var/cache/apt prior to
> >>> packaging. Too lazy now to write my own patch. :)
> >
> > Got it ;) (I will prepare the patch).
>
> TIA!
>
> BTW, if that direct call model works, we may shrink the SDK archive
> for that purpose even further (as there is no need to have a complete
> chroot anymore).
>
> Jan
>
prev parent reply other threads:[~2018-09-12 11:02 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-09-11 17:55 Jan Kiszka
2018-09-11 17:57 ` Jan Kiszka
2018-09-12 7:42 ` Maxim Yu. Osipov
2018-09-12 10:15 ` Jan Kiszka
2018-09-12 11:02 ` Henning Schild [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=20180912130238.1d94e306@md1pvb1c.ad001.siemens.net \
--to=henning.schild@siemens.com \
--cc=isar-users@googlegroups.com \
--cc=jan.kiszka@siemens.com \
--cc=mosipov@ilbers.de \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox