Solved Xorg and/or Intel regression, or just me

Until now, with FreeBSD 10.x (both .0 and .1), I'd been able to install x11-servers/xorg-server without hal. In addition, the problem I'd had with Intel HD4000 cards, that one couldn't quit out of X or even switch to a virtual terminal, was gone.

Yesterday, did a reinstall on a laptop with one of these cards. Firstly, when I did X in my usual way, the mouse (trackpad actually) didn't work. This surprised me enough to reinstall once or twice (nothing valuable, nothing important to do yesterday, and I figured let me get it right), and it's definitely the case on this machine at least. I suspect most people now install the server from packages, and its version requires sysutils/hal so I don't know if others will even notice that.

Secondly, and more annoyingly -- with the new install (and again, I tested this a couple of times yesterday, but still don't know if it's a Just Me(TM) issue), once again I'm unable to quit X or move to a virtual terminal. I'm wondering if anyone else has seen this, or is it just me. (I have another laptop with an Intel card that I'm afraid to update right now.)
 
In FreeBSD 10.X, switching back to the console with a KMS driver requires that vt(4) be enabled in /boot/loader.conf:
Code:
kern.vty=vt

Those who install binary packages get the default options, which includes the HAL option enabled. When the X server is updated to 1.14 in ports, that option will be disabled by default. Until then, build from ports.
 
Sorry, I wasn't clear, I am aware that the package includes HAL.

When I did the original fresh install, I built from ports (disabling HAL). I tried this a few times, including two separate installs, thinking that I might have missed something, and in all cases the mouse wouldn't work. (The keyboard did.)
 
The trackpad was the less important thing, actually. It was, until now, working out of the box without HAL as long as I compiled the server without HAL support. That, however, is little more than inconvenience. If I have to use HAL, it actually saves me some time, because then I will install it from packages. I'm more upset about the Intel issue, because not being able to easily exit and restart X and go to VT's is more of a regression--this makes me think it might be a Just Me issue, as I would think that such a thing would have been commented on by many already. The HAL thing -- I don't know what the majority do, but it's quite likely that most folks are used to using it, and, especially now that the package installs the new version, have never bothered with seeing what happens if its compiled without it.

That doesn't mean I don't appreciate your, errm, input (horrible pun but couldn't resist), but as for that aspect of it, it's more an idle curiosity question, is this a Just Me thing. The Intel problem is the one that I find more bothersome.
 
Well, to follow up, I just did an installation in VirtualBox. In that case, compiling and unchecking HAL support worked. Everything was fine. So, I'm not quite sure what went wrong with the other installation. The obvious guess is that I forgot to uncheck HAL support, but as I've said, I tried a few times. Still, this seems to be one of those things -- maybe the laptop has flaky hardware.
 
Okay, and on a further follow-up, after it seemed alright on the VM, I did another install on my other Intel based laptop and all was fine. So, I'm guessing it's some kind of problem specific to the laptop where I'm having the issue. Sorry for the waste of space.

Also, trying once more with the original laptop in question I got the mouse to work but the Intel card still doesn't. This is despite now having the same version of card and driver in both machines--well, at least to the point where they're marked as HD4000 in /var/log/Xorg.0.log
At any rate, as far as I can tell, with the wide sampling of two machines, any problems seem to be limited to one.
 
Back
Top