[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#1050501: rubberband: buildd for armel is too slow for 'Stretcher' test to complete within 5 minutes



Source: rubberband
Version: 3.3.0+dfsg-1
Severity: minor

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

I just noticed that rubberband is not migrating to testing due to a
failing build on armel. I saw it was already tried 8 times, with the
same result each time: failure.

Comparing the build log of armhf and armel, I noticed the following:

armhf:
dh_auto_test -- --timeout-multiplier 10
	cd obj-arm-linux-gnueabihf && LC_ALL=C.UTF-8 MESON_TESTTHREADS=8 meson test --timeout-multiplier 10
ninja: Entering directory `/<<PKGBUILDDIR>>/obj-arm-linux-gnueabihf'
ninja: no work to do.
1/8 Allocators        OK               0.05s
2/8 VectorOps         OK               0.04s
3/8 VectorOpsComplex  OK               0.03s
4/8 StretchCalculator OK               0.02s
5/8 SignalBits        OK               0.03s
6/8 FFT               OK               0.08s
7/8 Resampler         OK               1.31s
8/8 Stretcher         OK              23.88s

armel:
dh_auto_test -- --timeout-multiplier 10
	cd obj-arm-linux-gnueabi && LC_ALL=C.UTF-8 MESON_TESTTHREADS=4 meson test --timeout-multiplier 10
ninja: Entering directory `/<<PKGBUILDDIR>>/obj-arm-linux-gnueabi'
ninja: no work to do.
1/8 Allocators        OK               0.09s
2/8 VectorOps         OK               0.05s
3/8 VectorOpsComplex  OK               0.06s
4/8 StretchCalculator OK               0.04s
5/8 SignalBits        OK               0.10s
6/8 FFT               OK               0.23s
7/8 Resampler         OK              21.59s
8/8 Stretcher         TIMEOUT        300.03s   killed by signal 15 SIGTERM

The Resampler test takes 16.48 TIMES as much time on armel compared to
armhf. If we apply that same factor to the Stretcher test, the expected
duration is 393 seconds, so it's no wonder that if the timeout is set to
300 seconds, it would thus time out.

I saw in `debian/rules` (line 20) `--timeout-multiplier 10`, so perhaps
that needs to be increased (for armel)? 

HTH,
  Diederik

- -- System Information:
Debian Release: trixie/sid
  APT prefers unstable-debug
  APT policy: (500, 'unstable-debug'), (500, 'unstable'), (500, 'testing'), (101, 'experimental')
Architecture: amd64 (x86_64)

Kernel: Linux 6.5.0-0-amd64 (SMP w/16 CPU threads; PREEMPT)
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8), LANGUAGE=en_US:en
Shell: /bin/sh linked to /usr/bin/dash
Init: systemd (via /run/systemd/system)
LSM: AppArmor: enabled

-----BEGIN PGP SIGNATURE-----

iHUEARYIAB0WIQT1sUPBYsyGmi4usy/XblvOeH7bbgUCZOiQhgAKCRDXblvOeH7b
blUrAP99xag/FTbDbFCjIbObG6Hl/5k1qrwNaWS0YTGzRwOkKQD/e//HQNSEN6o6
A5jOwkGltfTKRygpRFjCD1UfJp+m1gg=
=KDNl
-----END PGP SIGNATURE-----


Reply to: