GNU bug report logs - #66249
Building for armv7 fails

Previous Next

To add a comment to this bug, you must first unarchive it, by sending
a message to control AT debbugs.gnu.org, with unarchive 66249 in the body.
You can then email your comments to 66249 AT debbugs.gnu.org in the normal way.

Toggle the display of automated, internal messages from the tracker.

Message #5 received at submit <at> debbugs.gnu.org (full text, mbox):

From: Sébastien Lerique <sl <at> eauchat.org>
To: bug-guix <at> gnu.org
Subject: Building for armv7 fails
Date: Thu, 28 Sep 2023 10:19:57 +0200

Dear Guix!

I'm packaging ripgrep in an ARM VM, which needs to build Rust itself,
and fails with the following end of the log file (with the terminal
colours not represented):

--8<---------------cut here---------------start------------->8---
root <at> vm-remarkable2:~# guix pack -R -S /emacsbin=bin \
                         emacs-no-x fd ripgrep sqlite \
                         htop fish \
                         aspell aspell-dict-en aspell-dict-fr

[...]
ESC[32m(43/173) BUILDING typenum v1.12.0ESC[0m
> /tmp/guix-build-rust-1.54.0.drv-0/mrustc/bin/mrustc rustc-1.54.0-src/vendor/typenum/src/lib.rs -o output/rustc-build/libtypenum-1_12_0.rlib --crate-name typenum --crate-type rlib -C emit-depfile=output/rustc-build/libtypenum-1_12_0.rlib.d --crate-tag 1_12_0 --cfg debug_assertions -O -L output -L output/rustc-build
/tmp/guix-build-rust-1.54.0.drv-0/mrustc/output/rustc-build/build_typenum-1_12_0/consts.rs:2113: error:0:Duplicate definition of name 'U2' in namespace scope (::""::generated::consts) crate::generated::consts::U2/*t:TypeAlias[::""::generated::consts::U2]*/, and crate::generated::consts::U2/*t:TypeAlias[::""::generated::consts::U2]*/
ESC[31mProcess was terminated with signal 6
ESC[0mFAILING COMMAND:  /tmp/guix-build-rust-1.54.0.drv-0/mrustc/bin/mrustc rustc-1.54.0-src/vendor/typenum/src/lib.rs -o output/rustc-build/libtypenum-1_12_0.rlib --crate-name typenum --crate-type rlib -C emit-depfile=output/rustc-build/libtypenum-1_12_0.rlib.d --crate-tag 1_12_0 --cfg debug_assertions -O -L output -L output/rustc-build
BUILD FAILED
make: *** [minicargo.mk:228: output/rustc] Error 1
error: in phase 'build': uncaught exception:
%exception #<&invoke-error program: "make" arguments: ("-f" "minicargo.mk" "output/rustc" "RUSTC_TARGET=armv7-unknown-linux-gnueabihf" "RUSTC_VERSION=1.54.0" "MRUSTC_TARGET_VER=1.54" "CFLAGS=-std=gnu11" "OUTDIR_SUF=") exit-status: 2 term-signal: #f stop-signal: #f>
phase `build' failed after 10098.8 seconds
command "make" "-f" "minicargo.mk" "output/rustc" "RUSTC_TARGET=armv7-unknown-linux-gnueabihf" "RUSTC_VERSION=1.54.0" "MRUSTC_TARGET_VER=1.54" "CFLAGS=-std=gnu11" "OUTDIR_SUF=" failed with status 2
--8<---------------cut here---------------end--------------->8---

Here's the current guix version:

--8<---------------cut here---------------start------------->8---
root <at> vm-remarkable2:~# guix describe
Generation 2	Sep 26 2023 12:59:04	(current)
  guix b2887da
    repository URL: https://git.savannah.gnu.org/git/guix.git
    branch: master
    commit: b2887da40c6cb9d541a9df67aa726694f64d8943
--8<---------------cut here---------------end--------------->8---

And if more needed:

--8<---------------cut here---------------start------------->8---
root <at> vm-remarkable2:~# uname -a
Linux vm-remarkable2 6.1.0-12-armmp-lpae #1 SMP Debian 6.1.52-1 (2023-09-07) armv7l GNU/Linux
--8<---------------cut here---------------end--------------->8---


I have no idea what is triggering this, would anybody have hints to
debug?

Thanks for anything!
And best,
Sébastien

Message #8 received at 66249 <at> debbugs.gnu.org (full text, mbox):

From: Sébastien Lerique <sl <at> eauchat.org>
To: 66249 <at> debbugs.gnu.org
Subject: Re: bug#66249: Acknowledgement (Building for armv7 fails)
Date: Thu, 28 Sep 2023 10:36:38 +0200

Actually the command is:
--8<---------------cut here---------------start------------->8---
root <at> vm-remarkable2:~# guix pack -R -S /emacsbin=bin \
                         emacs-no-x fd ripgrep sqlite \
                         htop fish \
                         aspell aspell-dict-en aspell-dict-fr
[... fails ...]
--8<---------------cut here---------------end--------------->8---

Then the log cited in my previous message is
`/var/log/guix/drvs/sc/r2cc0imq25c45r8amvjsa4wbbxwhwc-rust-1.54.0.drv.gz`

Best,
S.

Message #11 received at 66249 <at> debbugs.gnu.org (full text, mbox):

From: Sébastien Lerique <sl <at> eauchat.org>
To: 66249 <at> debbugs.gnu.org
Subject: Re: bug#66249: Acknowledgement (Building for armv7 fails)
Date: Thu, 28 Sep 2023 13:14:53 +0200

retitle 66249 Building rust-1.54 fails on armv7

Message #14 received at 66249 <at> debbugs.gnu.org (full text, mbox):

From: Maxim Cournoyer <maxim.cournoyer <at> gmail.com>
To: Sébastien Lerique <sl <at> eauchat.org>
Cc: 66249 <at> debbugs.gnu.org
Subject: Re: bug#66249: Building for armv7 fails
Date: Tue, 03 Oct 2023 22:14:47 -0400

Hello,

Sébastien Lerique <sl <at> eauchat.org> writes:

> retitle 66249 Building rust-1.54 fails on armv7

As you've found out, this is something that has to do with building rust
on that architecture, more precisely building 'mrustc', which is used in
Guix to boostrap rust.  The upstream issue would be this one I think:
<https://github.com/thepowersgang/mrustc/issues/78>.  It's about GCC
requiring more memory than 32 bit can address but the author mentioned a
2 GiB or something limit, I'm not sure what that is attributable to.

If you want to solve this I'd recommend you get in touch with the mrustc
author!  There is a #mrustc channel on Libera.chat IRC, as far as I
recall.

-- 
Thanks,
Maxim

Message #17 received at 66249 <at> debbugs.gnu.org (full text, mbox):

From: Greg Hogan <code <at> greghogan.com>
To: Maxim Cournoyer <maxim.cournoyer <at> gmail.com>
Cc: Sébastien Lerique <sl <at> eauchat.org>, 66249 <at> debbugs.gnu.org
Subject: Re: bug#66249: Building for armv7 fails
Date: Thu, 31 Oct 2024 14:16:23 -0400

On Tue, Oct 3, 2023 at 10:16 PM Maxim Cournoyer
<maxim.cournoyer <at> gmail.com> wrote:
>
> Hello,
>
> Sébastien Lerique <sl <at> eauchat.org> writes:
>
> > retitle 66249 Building rust-1.54 fails on armv7
>
> As you've found out, this is something that has to do with building rust
> on that architecture, more precisely building 'mrustc', which is used in
> Guix to boostrap rust.  The upstream issue would be this one I think:
> <https://github.com/thepowersgang/mrustc/issues/78>.  It's about GCC
> requiring more memory than 32 bit can address but the author mentioned a
> 2 GiB or something limit, I'm not sure what that is attributable to.
>
> If you want to solve this I'd recommend you get in touch with the mrustc
> author!  There is a #mrustc channel on Libera.chat IRC, as far as I
> recall.

Is there anything for Guix to do here or can the issue be closed?

Message #20 received at 66249 <at> debbugs.gnu.org (full text, mbox):

From: Sébastien Lerique <sl <at> eauchat.org>
To: Greg Hogan <code <at> greghogan.com>
Cc: 66249 <at> debbugs.gnu.org, Maxim Cournoyer <maxim.cournoyer <at> gmail.com>
Subject: Re: bug#66249: Building for armv7 fails
Date: Thu, 31 Oct 2024 23:13:09 +0100

On 31 Oct 2024 at 14:16, Greg Hogan <code <at> greghogan.com> wrote:
> Is there anything for Guix to do here or can the issue be closed?
>

I'd say this can be closed as unsolved, at least until some progress is
found in https://github.com/thepowersgang/mrustc/issues/78 .

Thanks for the follow-up, and best
-- 
Sébastien

Message #25 received at 66249 <at> debbugs.gnu.org (full text, mbox):

From: Maxim Cournoyer <maxim.cournoyer <at> gmail.com>
To: Sébastien Lerique <sl <at> eauchat.org>
Cc: 66249 <at> debbugs.gnu.org, Greg Hogan <code <at> greghogan.com>
Subject: Re: bug#66249: Building for armv7 fails
Date: Fri, 01 Nov 2024 14:25:52 +0900

Hi,

Sébastien Lerique <sl <at> eauchat.org> writes:

> On 31 Oct 2024 at 14:16, Greg Hogan <code <at> greghogan.com> wrote:
>> Is there anything for Guix to do here or can the issue be closed?
>>
>
> I'd say this can be closed as unsolved, at least until some progress is
> found in https://github.com/thepowersgang/mrustc/issues/78 .
>
> Thanks for the follow-up, and best

Marked as 'notfixed' via Debbugs (if that does what I think it does),
and closed.

-- 
Thanks,
Maxim

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.