tc 3.1 on vbox 3.2.8

[jhancock]

When I boot tc 3.1 on the latest virtual box 3.2.8, windows xp host, the mouse movement in tiny core have two problems:
1 - the movement is not very smooth.  much less so than running just about any other recent linux distro on virtual box.
2 - the mouse seems to lose its boundaries.  For examples, I find the mous not able to move further down to click a button.

Is there something additional I can add to tc to get better integration?

thanks, Jon

[maro]

Jon, I'm using VBox a fair bit and what you describe reminds of of some behaviour I noticed quite some time ago (maybe last year ??). I have therefore today upgraded from v3.2.4 to v3.2.8 (also on XP host) just to see if things have regressed. Unfortuantely I can't confirm your findings (yet), albeit I've only done some limited testing.

One thing that might be different between us is that I'm using a "wrapper" called Portable-VirtualBox. In my view that leaves VBox a bit more flexible, e.g. when it comes to upgrading. But it might be that there is no tangible benefit and I'm just doing it because I've got so used to this modus operandi.

From memory (and I could be wrong and just confuse this with some other issues I once had with VBox) resizing the VBox window or changing from windowed mode to fullscreen and back to windowed mode (e.g. via "Right-CTRL F", then release the two keys, followed again by "Right-CTRL F") might have brought things back into line. I'm also not sure that the problem was limited to TinyCore, IIRC there were other guests (e.g. Slitaz) equally effected. Maybe the type of X server plays a role. With what I'm doing I get most of the time away with using Xvesa, but I can't remember if Xorg was involved when things were "going funny".

I guess the upshot is that you might need to carefully monitor under which conditions (e.g. which X server, windowed or fullscreen mode, ...) the "fun" happens. Maybe this way it can be better identified and a bug raised (presumably against VBox). I'm just not convinced that TC is the culprit here.

[jur]

I have noticed something like that. I found that the mouse had a different scale factor inside compared to outside, so that it would lose its boundary reference. I was able to work around this problem by pulling the mouse cursor to the opposite side and beyond compared to the side that was giving a problem. I had to do this all the time, quite limiting.

[curaga]

I've seen references to this with other Xvesa-using distros. Complain to Oracle, if vbox thinks it's a good thing to allow window resizing when the guest is not running the vbox driver it's clearly wrong behavior.

Disclaimer: I don't use virtualbox. The above is just my guess.

[jhancock]

Would the vbox guest additions help with the mouse problems?  Or should I start by using Xorg instead of Xvesa?

I've done a little research into the effort to compile/install vbox guest additions and it looks like a lot of work.  Does anyone have this available so I can simply try to see if it fixes the mouse behavior?  Would it be possible to have the guest additions available as a package?

thanks, Jon

[^thehatsrule^]

The screen resizing addition requires Xorg afaik.  I think there is a package for 2.x but some have been working on it for 3.x

[eschvoca]

Hi,

I'm pretty sure this is an issue with vbox 3.2.8.  I've been using that release on Windows and Linux hosts with many different Linux guests (RHEL4, TCL 3.1, Slitaz, etc) and they all do the same thing.  No solution yet.  I was running vbox 3.0.x and things were fine. 

The problem is the mouse stops before the edge of the window as it thinks it is at the edge and then it may jump to the outside edge of the window (on the guest).  This seems to be best described as an error in mouse scaling between the VM and the host....very annoying.

Thanks.