Hi Simon, On 2024-09-24 09:16, Simon McVittie wrote: > In a recent upload this failed on armhf, and only armhf: the actual > tests appear to have all succeeded, but then xvfb-run exited with status > 1 anyway. I am able to reproduce this on the armhf porterbox amdahl. > > A simplified reproducer is: > > xvfb-run -e /dev/stderr -a -s "-noreset" true > > which should exit 0
I tried reproducing this bug in a armhf schroot running in a bookworm VM, and couldn't. (sid-armhf-sbuild)ema@localhost:/tmp$ xvfb-run -e /dev/stderr -a -s "-noreset" true && echo OK _XSERVTransmkdir: Owner of /tmp/.X11-unix should be set to root OK No segfault, and the _XSERVTransmkdir warning goes away by running the command as root and then again as unprivileged user. The system is running the same kernel as amdahl: (sid-armhf-sbuild)ema@localhost:/tmp$ uname -a Linux localhost 6.1.0-25-arm64 #1 SMP Debian 6.1.106-3 (2024-08-26) armv8l GNU/Linux It has the same CPU flags (inside the schroot): (sid-armhf-sbuild)ema@localhost:/tmp$ lscpu | grep Flags Flags: half thumb fastmult vfp edsp neon vfpv3 tls vfpv4 idiva idivt lpae evtstrm aes pmull sha1 sha2 crc32 As well as outside the schroot: ema@localhost:~$ lscpu | grep Flags Flags: fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm lrcpc dcpop asimddp And it has libunwind8 1.7.2-1 installed: (sid-armhf-sbuild)ema@localhost:/tmp$ dpkg -l \*libunwind8\* | grep ^ii ii libunwind8:armhf 1.7.2-1 armhf library to determine the call-chain of a program - runtime I double-checked that the problem can still be reproduced on amdahl, and indeed it can. There must be some other significant difference between the two environments, I'm going to dig further. ema
