Bug#1093564: libsoup3: FTBFS on riscv64: server-test fails

[Aurelien Jarno]

Source: libsoup3
Version: 3.6.4-1
Severity: serious
Tags: ftbfs patch
Justification: fails to build from source (but built successfully in the past)
X-Debbugs-Cc: debian-riscv at lists.debian.org
User: debian-riscv at lists.debian.org
Usertags: riscv64

Dear maintainer,

libsoup3 fails to build from source on riscv64. It seems to be a mix of
flaky tests like on other architectures and a riscv64 specific way of
failing:

| Summary of Failures:
| 
| 25/38 server-test              ERROR            60.96s   killed by signal 11 SIGSEGV
| 
| Ok:                 37  
| Expected Fail:      0   
| Fail:               1   
| Unexpected Pass:    0   
| Skipped:            0   
| Timeout:            0   
| dh_auto_test: error: cd obj-riscv64-linux-gnu && DEB_PYTHON_INSTALL_LAYOUT=deb LC_ALL=C.UTF-8 MESON_TESTTHREADS=4 meson test --verbose --timeout-multiplier 5 returned exit code 1
| make[1]: *** [debian/rules:39: override_dh_auto_test-arch] Error 25
| make[1]: Leaving directory '/build/reproducible-path/libsoup3-3.6.4'
| make: *** [debian/rules:25: binary-arch] Error 2
| dpkg-buildpackage: error: debian/rules binary-arch subprocess returned exit status 2
| --------------------------------------------------------------------------------
| Build finished at 2025-01-19T19:51:21Z


The full build log is available there:
https://buildd.debian.org/status/fetch.php?pkg=libsoup3&arch=riscv64&ver=3.6.4-1&stamp=1737316290&raw=0

Although reported as a SIGSEGV, itIt appears that the test timeout is
too short for the riscv64 build daemons.  Increasing the timeout
multiplier from 5 to 6, like in the patch below, allows the testsuite to
pass.


--- libsoup3-3.6.4/debian/rules
+++ libsoup3-3.6.4/debian/rules
@@ -36,7 +36,7 @@
 # instead, when we have a better idea of how long they take on various
 # architectures
 override_dh_auto_test-arch:
-	dh_auto_test -- --timeout-multiplier 5
+	dh_auto_test -- --timeout-multiplier 6
 
 # debhelper >= 13.4 makes all of /usr/libexec executable, which is not
 # quite right for installed-tests


Regards
Aurelien