Since the release of X11R7.0, a new input driver "vmmouse" was added to the upstream repository, and a release made. This is a mouse driver for use under VMware, which should be added to FC5 for completeness. I've created an initial package based from the "xorg-x11-drv-mouse" package, and imported it into CVS. Here is what remains to be done: 1) It needs to be added to beehive. 2) Build it in dist-fc5 3) Add it as a driver to xorg-x11-drivers, for the 3 arches it builds for. 4) Notify jeremy in case our config tools need to be updated to autodetect VMware's USB IDs or whatever, and automatically configure X to use vmmouse under VMware installations. (Not sure if that's the case or not, but if it is, it would be convenient to VMware users I suppose.) CC'ing katzj and notting
Why does vmware need a separate mouse driver????
(In reply to comment #1) > Why does vmware need a separate mouse driver???? I'm told that it is to avoid synchronization problems due to host mouse acceleration plus guest OS mouse acceleration, and that it makes it easy to implement smart mouse grabbing. ie: When the edge of the screen is reached in the guest OS, the mouse can be released, and it will seemlessly move away in the right spot on the host. Similar things are done inside qemu for example. Anyway, VMware Inc. has requested this be added to the OS, and it is a standard Xorg driver now, so there's no good reason for us to not add it. If we don't add it now, we'll get a RHEL request later to add it, and we'll be adding a driver to RHEL which had no Fedora testing.
As per: https://intranet.corp.redhat.com/ic/intranet/PackageListProcess.html notting: Can you rubber stamp this, so it can get into test3 ASAP? jeremy: Does this require any installer, etc. changes? I believe it is desireable if the vmmouse driver gets assigned for all VMware installations. I'm not sure how to detect that currently though, however that is something that could be done at a later date, as it works without it still. jkeating: Can you add 'xorg-x11-drv-vmmouse' to beehive, etc. TIA guys...
Yes, there will be real installer changes needed if we want to associate the different mouse driver if you're on VMware. And that's going to require some digging to figure out how you can tell that you need to, etc. I don't know that this is really viable for FC5. Test2 (the nominal feature freeze) was a month ago.
(In reply to comment #4) > Yes, there will be real installer changes needed if we want to associate the > different mouse driver if you're on VMware. And that's going to require some > digging to figure out how you can tell that you need to, etc. I assumed that might be necessary, however the driver is still useful being in the OS, even if it isn't configured by default. > I don't know that this is really viable for FC5. Test2 (the nominal > feature freeze) was a month ago. I realize that. A month ago the driver didn't exist to make a package for. Anyway, including it is pretty much 0 risk to us, and gets some end user testing, even if it has to be manually configured. I believe VMware's installation might even configure it itself if it's present, but I'm not sure. I don't see any reason to not include it really, but I also don't feel strongly to fight for its inclusion if there is strong opposition either. If we had monolithic X, I'd have just put the driver source into the monolith and rebuilt it, and nobody would know/care or have needed to approved it really. Currently I'm just wasting cycles caught up in needless "process" for a low risk driver that doesn't harm anything. Please either approve it, or move this bug to FC6Target so I can get immediate closure. Thanks in advance.
Package approvals for FC5: Tech lead (notting): refused Why does vmware need a separate mouse driver???? (comment #1) Installer (jeremy): refused Probably requires installer changes to fully integrate. It is past the FC5 package inclusion deadline. RELENG (jkeating): refused Didn't get necessary prerequisite approvals.
Defering driver inclusion until FC6.
Already included for a while now....