unable to run pintos on ubuntu QEMU with ssh because it "Could not initialize SDL (No available video device)"

Question

I have set to run pintos on my own ubuntu desktop 14.04. For now it runs perfectly on the desktop, as shown in the image below.

Using this ubuntu as a server as well, I wish I could run QEMU via ssh as well. I know that it is possible to run bochs via ssh, but is it possible to do so with QEMU as well?

I assume there must be a way to run qemu as a command line interface (like bochs).

In fact, here is the script for running qemu in pintos script. I assume that pintos should run with $vga eq 'none' on ssh, but it seems not.

# Runs QEMU.
sub run_qemu {
    print "warning: qemu doesn't support --terminal\n"
      if $vga eq 'terminal';
    print "warning: qemu doesn't support jitter\n"
      if defined $jitter;
    my (@cmd) = ('qemu');
    for my $iface (0...3) {
    my ($option) = ('-hda', '-hdb', '-hdc', '-hdd')[$iface];
    push (@cmd, $option, $disks_by_iface[$iface]{FILE_NAME})
      if defined $disks_by_iface[$iface]{FILE_NAME};
    }
    push (@cmd, '-m', $mem);
    push (@cmd, '-net', 'none');
    push (@cmd, '-nographic') if $vga eq 'none';
    push (@cmd, '-serial', 'stdio') if $serial && $vga ne 'none';
    push (@cmd, '-S') if $debug eq 'monitor';
    push (@cmd, '-s', '-S') if $debug eq 'gdb';
    push (@cmd, '-monitor', 'null') if $vga eq 'none' && $debug eq 'none';
    run_command (@cmd);
}

Any help will be appreciated.

Answer 1

I found a solution to my own problem. You just need to give -nographic option no matter what $vga variable is. $vga variable can be one of 'none', 'terminal', 'window'. So you could add push (@cmd, '-nographic') if $vga eq 'window'; to pintos perl script.

I still wonder how they determine if I am running on a terminal, and why they get it wrong. Please let me know if you have any idea.