libsigsegv fails to build on emulated aarch64 [core-updates]

Hi guix

I noticed a new build failure on ci.guix.gnu.org:

https://ci.guix.gnu.org/build/660357/log/raw.

Relevant log output:

make[2]: Leaving directory '/tmp/guix-build-libsigsegv-2.13.drv-0/libsigsegv-2.13'

make[1]: Leaving directory '/tmp/guix-build-libsigsegv-2.13.drv-0/libsigsegv-2.13'

phase `build' succeeded after 44.7 seconds

starting phase `check'

yes

checking for working strerror function... Making check in src

make[1]: Entering directory '/tmp/guix-build-libsigsegv-2.13.drv-0/libsigsegv-2.13/src'

make[1]: Nothing to be done for 'check'.

make[1]: Leaving directory '/tmp/guix-build-libsigsegv-2.13.drv-0/libsigsegv-2.13/src'

Making check in tests

make[1]: Entering directory '/tmp/guix-build-libsigsegv-2.13.drv-0/libsigsegv-2.13/tests'

make check-TESTS

make[2]: Entering directory '/tmp/guix-build-libsigsegv-2.13.drv-0/libsigsegv-2.13/tests'

time.h

checking for struct tm.tm_zone... yes

checking for fake locale system (OpenBSD)... make[3]: Entering directory '/tmp/guix-build-libsigsegv-2.13.drv-0/libsigsegv-2.13/tests'

yes

checking for struct tm.tm_gmtoff... FAIL: stackoverflow2

PASS: sigsegv2

../build-aux/test-driver: line 109: 6663 Segmentation fault "$@" > $log_file 2>&1

PASS: sigsegv3

PASS: sigsegv1

FAIL: stackoverflow1

yes

checking for nlink_t... (cached) yes

checking whether unlink honors trailing slashes... yes

checking for O_CLOEXEC... no

checking for Solaris 11.4 locale system... no

checking for getlocalename_l... ============================================================================

Testsuite summary for libsigsegv 2.13

============================================================================

# TOTAL: 5

# PASS: 3

# SKIP: 0

# XFAIL: 0

# FAIL: 2

# XPASS: 0

# ERROR: 0

============================================================================

See tests/test-suite.log

============================================================================

make[3]: *** [Makefile:729: test-suite.log] Error 1

make[3]: Leaving directory '/tmp/guix-build-libsigsegv-2.13.drv-0/libsigsegv-2.13/tests'

make[2]: *** [Makefile:837: check-TESTS] Error 2

make[2]: Leaving directory '/tmp/guix-build-libsigsegv-2.13.drv-0/libsigsegv-2.13/tests'

make[1]: *** [Makefile:966: check-am] Error 2

make[1]: Leaving directory '/tmp/guix-build-libsigsegv-2.13.drv-0/libsigsegv-2.13/tests'

make: *** [Makefile:432: check-recursive] Error 1

Test suite failed, dumping logs.

--- ./tests/test-suite.log --------------------------------------------------

===========================================

libsigsegv 2.13: tests/test-suite.log

===========================================

# TOTAL: 5

# PASS: 3

# SKIP: 0

# XFAIL: 0

# FAIL: 2

# XPASS: 0

# ERROR: 0

.. contents:: :depth: 2

FAIL: stackoverflow1

====================

qemu: uncaught target signal 11 (Segmentation fault) - core dumped

FAIL stackoverflow1 (exit status: 139)

FAIL: stackoverflow2

====================

Starting recursion pass 1.

Stack overflow 1 missed.

FAIL stackoverflow2 (exit status: 1)

error: in phase 'check': uncaught exception:

%exception #<&invoke-error program: "make" arguments: ("check" "-j" "16") exit-status: 2 term-signal: #f stop-signal: #f>

phase `check' failed after 3.5 seconds

command "make" "check" "-j" "16" failed with status 2

builder for `/gnu/store/ww5bf6xz13wxjs1sjvjc2kmwq5mrjdj5-libsigsegv-2.13.drv' failed with exit code 1

@ build-failed /gnu/store/ww5bf6xz13wxjs1sjvjc2kmwq5mrjdj5-libsigsegv-2.13.drv - 1 builder for `/gnu/store/ww5bf6xz13wxjs1sjvjc2kmwq5mrjdj5-libsigsegv-2.13.drv' failed with exit code 1

cannot build derivation `/gnu/store/5pfnq2666wp2gg1h6yl0c92din8n24wc-gawk-5.1.0.drv': 1 dependencies couldn't be built

Greetings,

Maxime.

Hi,

Maxime Devos <maximedevos@telenet.be> skribis:

For now I worked around it by offloading this to a “real” machine

(overdrive1), where it builds fine. I wonder if there’s much we can do

regarding QEMU’s behavior here.

Ludo’.

Ludovic Courtès schreef op zo 11-07-2021 om 00:19 [+0200]:

Maybe detect if QEMU is used, and if so, don't run the test suite?

Not really a ‘clean’ solution though, w.r.t. reproducibility,

and I wouldn't know how to detect this.

If this is a bug in QEMU, then ideally that would be fixed in QEMU,

but I wouldn't know where to look.

Greetings,

Maxime.

Hello,

Maxime Devos <maximedevos@telenet.be> skribis:

Yeah, I’d rather avoid that.

It could be that someone else on the intertubes stumbled upon that

issue, that’d be great. It could be that libsigsegv plays tricks that

don’t fare well with QEMU’s expectations, as in

https://bugzilla.redhat.com/show_bug.cgi?id=1493304#c5. We should ask

on bug-libsigsegv@gnu.org.

Thanks,

Ludo’.

Hi all,

Ludovic Courtès <ludo@gnu.org> writes:

(I just realized I never actually replied to this!)

Configuring with "--disable-stackvma" seems to fix this. Doing this

makes libsigsegv use a different heuristic for determining if a SIGSEGV

was a stack overflow. I don't think it should impact functionality.

Perhaps just apply that to aarch64 until there's a proper fix?

This is probably a QEMU bug... I will try to report this to upstream

QEMU when I can, as I can't find my notes on this right now.

--

Sarah

On Wed, Sep 29, 2021 at 07:40:20PM -0700, Sarah Morgensen wrote:

I came across this on x86_64 when using our qemu-binfmt service when

building for powerpc-linux too, and I'm pretty sure powerpc64le-linux

and armhf-linux also. I haven't tried going the other direction, from

aarch64-linux and emulating x86_64/i686 to see if it happens there too.

--

Efraim Flashner <efraim@flashner.co.il> ????? ?????

GPG key = A28B F40C 3E55 1372 662D 14F7 41AA E7DC CA3D 8351

Confidentiality cannot be guaranteed on emails sent or received unencrypted

Hi!

Sarah Morgensen <iskarian@mgsn.dev> skribis:

Sounds like a good interim measure, for aarch64 and powerpc at least as

Efraim writes.

If you can come up with a patch, we could apply it in the upcoming

rebuild.

Thanks,

Ludo’.