All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Yann E. MORIN" <yann.morin.1998@free.fr>
To: Leo <info.leo@widler.de>
Cc: buildroot@buildroot.org
Subject: Re: [Buildroot] [PATCH 1/1] package/attr: add --enable-multilib to fix build error for BL808 based boards
Date: Mon, 27 Nov 2023 17:10:32 +0100	[thread overview]
Message-ID: <20231127161032.GQ3177259@scaer> (raw)
In-Reply-To: <0A422087-89C4-4C88-8263-E05CC5BA1FB8@widler.de>

Leo, All,

On 2023-11-27 08:07 +0100, Leo spake thusly:
> defconfig can be found here: [1]https://github.com/openbouffalo/buildroot_bouffalo/blob/main/configs/pine64_ox64_full_defconfig
> The link to the tool chain is also in there.
> Sure it would be better to get to the root of the problem. My solution was just a quick and dirty fix.

Thanks for the links.

I tried to reproduce locally, after following the instructions in the
buildroot_bouffalo readme file, but attr does build properly here with
the aforementioned defconfig.

I also tried to reproduce in a container using the reference image by
Buildroot (./utils/docker-run) and it also succeeds.

I also tried to run with the ubuntu:22.04 that is used in the github
worflows for buildroot_bouffalo, and (after installing the necessary
dependencies as visible in .github/workflows/buildroot.yml), it also
succeeds.

I also tried without using ccache, and it also succeeded.

Thus, it comforts my opinion that using --enable-multilib when
configuring attr, is not the proper solution.

However, I do see that your have workflows failing with that issue,
indeed, but of all the tests I did, none could reproduce it.

I see that you are using a cached ccache for your workflows; I wonder if
somehow that cache did get corrupted. Can you try and run a workflow
without ccache?

Regards,
Yann E. MORIN.

> Regards,
> Leonard Widler
> 
> Am 27. November 2023 07:48:13 MEZ schrieb "Yann E. MORIN" <yann.morin.1998@free.fr>:
> 
>  Leo, All,
> 
>  On 2023-11-26 23:17 +0100, Leo spake thusly:
> 
>  I'll change my name, thanks for letting me know
> 
>  The failure is on none of the auto builders. I have uploaded the config.log here: [1]https://pastebin.com/ZzP6pNvx
> 
> 
>  Can you also provide the Buildroot defconfig, please?
> 
>  (Also, I see you are running as root; this is not required to run
>  Buildroot, and is even dangerous and not really supported; please run as
>  a non-root user.)
> 
> 
>  And to fix this issue I added the --enable-multilib option which resolves the problem.
> 
> 
>  The issue looks like a misconfiguration somewhere:
> 
>      .../crti.o: can't link double-float modules with soft-float modules
>      ...//ld: failed to merge target specific data of file .../crti.o
> 
>  So, we want to understand the problem rather than paper-over it.
> 
>  Where did you get your toolchain (Xuantie-900 linux-5.10.4 glibc gcc
>  Toolchain V2.6.1 B-20220906)? Is it public, so we can try and reproduce
>  the issue?
> 
>  Regards,
>  Yann E. MORIN.
> 
> 
>  Regards,
>  Leonard Widler
> 
>  Am 26. November 2023 22:56:44 MEZ schrieb "Yann E. MORIN" <yann.morin.1998@free.fr>:
> 
>   leow149, All,
> 
>   Thank you for your contribution! Here is a quick review.
> 
>   First, we'll need you to use your real name as the author of the patch;
>   we do not accept annonymous (od pseduonymou) contributions.
> 
>   On 2023-11-26 22:43 +0100, leow149 spake thusly:
> 
>   This commit fixes a build error which specifically happens for Bouffallo Labs BL808 based boards, but might also happen on other RiscV64 Boards.
> 
> 
>   Please wrap your commit log to ~72 chars.
> 
>   Waht is the error that you see? Could you see if there is a similar
>   build failure on out autobulders:
> 
>   [2]http://autobuild.buildroot.org/?static=1&reason=attr-%
> 
> 
>   The last build error for attr was more than a year ago now, so it is
>   relatively surprising to see that attr does not build. So we'll need a
>   bit more context.
> 
> 
>   Signed-off-by: leow149 <info.leo@widler.de>
> 
>   --------------------------------------------------------------------------------------------------------------------------------
> 
>    package/attr/attr.mk | 5 +++--
>    1 file changed, 3 insertions(+), 2 deletions(-)
> 
>   diff --git a/package/attr/attr.mk b/package/attr/attr.mk
>   index 0327497e1e..89f7d21f0c 100644
>   --- a/package/attr/attr.mk
>   +++ b/package/attr/attr.mk
>   @@ -13,7 +13,8 @@ ATTR_CPE_ID_VENDOR = attr_project
> 
>    ATTR_INSTALL_STAGING = YES
> 
>   -ATTR_CONF_OPTS = --disable-nls
>   -
>   +ATTR_CONF_OPTS = \
>   +       --disable-nls \
>   +       --enable-multilib
> 
> 
>   Buildroot does not generate multilib toolchains, and when using an
>   external toolchain that is multilib, the proper multilib diretory is
>   copies over to the sysroot.
> 
>   So, --eanble-multilib looksa like the wrong approach, at least without
>   more explanations.
> 
>   Can you elaborate on the issue you encounter, and what makes you think
>   --enable-multilib is the correct solution?
> 
>   Thanks!
> 
>   Regards,
>   Yann E. MORIN.
> 
> 
>    $(eval $(autotools-package))
>    $(eval $(host-autotools-package))
>   --
>   2.30.2
> 
>   --------------------------------------------------------------------------------------------------------------------------------
> 
>   buildroot mailing list
>   buildroot@buildroot.org
>   [3]https://lists.buildroot.org/mailman/listinfo/buildroot
> 
>  Links:
>  1.
>  [2]https://pastebin.com/ZzP6pNvx
>  2.
>  [3]http://autobuild.buildroot.org/?static=1&reason=attr-%
> 
>  3.
>  [4]https://lists.buildroot.org/mailman/listinfo/buildroot
> 
> 
> 
> 
>   --------------------------------------------------------------------------------------------------------------------------------
> 
>  buildroot mailing list
>  buildroot@buildroot.org
>  [5]https://lists.buildroot.org/mailman/listinfo/buildroot
> 
> Links:
> 1. https://github.com/openbouffalo/buildroot_bouffalo/blob/main/configs/pine64_ox64_full_defconfig
> 2. https://pastebin.com/ZzP6pNvx
> 3. http://autobuild.buildroot.org/?static=1&reason=attr-%
> 4. https://lists.buildroot.org/mailman/listinfo/buildroot
> 5. https://lists.buildroot.org/mailman/listinfo/buildroot

> _______________________________________________
> buildroot mailing list
> buildroot@buildroot.org
> https://lists.buildroot.org/mailman/listinfo/buildroot


-- 
.-----------------.--------------------.------------------.--------------------.
|  Yann E. MORIN  | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: |
| +33 662 376 056 | Software  Designer | \ / CAMPAIGN     |  ___               |
| +33 561 099 427 `------------.-------:  X  AGAINST      |  \e/  There is no  |
| http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL    |   v   conspiracy.  |
'------------------------------^-------^------------------^--------------------'
_______________________________________________
buildroot mailing list
buildroot@buildroot.org
https://lists.buildroot.org/mailman/listinfo/buildroot

      reply	other threads:[~2023-11-27 16:10 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-11-26 21:43 [Buildroot] [PATCH 1/1] package/attr: add --enable-multilib to fix build error for BL808 based boards leow149
2023-11-26 21:56 ` Yann E. MORIN
2023-11-26 22:17   ` Leo
2023-11-27  6:48     ` Yann E. MORIN
2023-11-27  7:07       ` Leo
2023-11-27 16:10         ` Yann E. MORIN [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=20231127161032.GQ3177259@scaer \
    --to=yann.morin.1998@free.fr \
    --cc=buildroot@buildroot.org \
    --cc=info.leo@widler.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 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.