You are not logged in.
After an update (I can't tell which one), filezilla stopped working.
Downgrading filezilla as well as libfilezilla doesn't help. I don't know which version was the last one that functioned, but I know
I used filezilla as late as last december, so I downgraded to filezilla-3.29.0-1 and libfilezilla-0.11.1-1.-i686.pkg.tar.xz, and here is what happens
Reading locale option from ~/.config/filezilla/filezilla.xml
wxD-Bus: Signal from /org/freedesktop/DBus, member NameAcquired
wxD-Bus: Reply with serial 2
wxD-Bus: Signal: Error: The name org.gnome.SessionManager was not provided by any .service files
Illegal instruction (core dumped)
Any idea on how to solve this?
Offline
Another thing: filezilla crashes only after the graphic environment is used. That is, if I issue commands like
filezilla --version
or
filezilla --help
then I don't get "Illegal instruction". Does this help?
Offline
As of today, filezilla (now version 3.38.1-1) doesn't crash at the startup, it more or less works but it still crashes with
Illegal instruction (core dumped)
when I open site manager. Any solution?
Offline
I tried on Qemu with an i486 CPU, flags: fpu vme pse cpuid tsc_known_freq x2apic hypervisor cpuid_fault
Mmh, I have a slightly different behaviour: --version and --help work.
Normal startup results in SIGILL:
Dec 01 09:43:34 arch32-stable kernel: traps: filezilla[443] trap invalid opcode ip:b136f5bd sp:bfa0de90 error:0 in librsvg-2.so.2.44.8[b115d000+32a000]
So, I suspect some rust optimization in librsvg (compilation flags) which produces some instructions, which fail on your CPU.
The lowest CPU rust is generating code for is currently i586 (Pentium). This might involve some instructions which are not
available on old AMDs.
On the long term there is nothing we can do about it, but bootstrapping rust on i486..
I emulated on Pentium, Penium II and Pentium III and all of them result in a running filezilla.
Your flags indicate you have some sort of early AMD:
Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca
cmov pat pse36 mmx fxsr sse syscall mmxext 3dnowext 3dnow cpuid 3dnowprefetch vm
mcall
I started to notice, that security fixes and optimizations start to be quite Intel-specific breaking on AMD.
On a side note: my AMD-K6 stopped working with SIGILL in glibc, so optimizations start to creep in there too. I had to
blacklist glibc. This can not be your problem, otherwise you couldn't even boot the machine.
Offline
Hmm, yes I guess mitigations to raw metal hacks like spectre are liable to manifest as instruction sequences written out by compilers in the main. I guess I'd need to start digging in the patches to even start to figure out what they actually did though, because I would have mainly though the mitigations would consist of not doing specific things in a row rather than writing specfic new opcode that are too new, and they're not advertising their patches in plain prose for fairly obvious reasons.
Architecture: pentium4, Testing repos: Yes, Hardware: EeePC 901+2GB RAM+OS half on the SD card.
Offline