What a surprise...
Created attachment 129780 [details, diff] 1.0.0.16-kernel-2.6.23.patch And here's the patch. Still untested but I guess it should work.
Created attachment 129819 [details, diff] 1.0.0.16-kernel-2.6.23.patch Small correction: set_dumpable is a GPL-only symbol now. But I've tested it now and my patch works...
vmware-any-any-update113 contains the fixes according to: http://www.vmware.com/community/thread.jspa?messageID=704962򬇂 http://www.vmware.com/community/thread.jspa?messageID=732061򲮝
Thanks dev-zero, When I get a chance, hopefully this evening, I'll try swapping over from the vmware-supplied sources for Workstation 6 to the vmware-any-any-update modules. That'll make things a bit easier to maintain, I just wasn't sure they were ready for showtime. I'll make the change in the overlay, and then if you'd be able to test it out for me, that'd be very helpful (I don't have any boxes running 2.6.23-rc*)... 5:) I'll let you know once I've made the change...
Ok, vmware-modules-1.0.0.16-r1 is now in the tree and unmasked, and is currently built using vmware-any-any-update113, so I'm gonna mark this bug as closed. Thanks for you patience... 5;)
Reopening as this is broken in the stable tree.. I'm now working towards 2.6.23 stabling in 2 weeks time, so this stuff should ideally be fixed in stable if possible.
Hiya Daniel, I'm not certain exactly what you mean. Currently all of the vmware-modules ebuilds are now building using vmware-any-any-update113, which should include support for 2.6.23. That means all the currently stable vmware-modules ebuilds support 2.6.23. The only ebuild that *isn't* based on vmware-any-any-update113 is vmware-modules-1.0.0.17, which isn't marked as stable. So, have you had a specific instance of the current ebuilds not building on 2.6.23, or were you asking for this to be turned into a stabilization bug for vmware-modules-1.0.0.16-r1, or is it just the rather unusual way we've set up vmware-modules to work that caused the confusion? 5:)
Comment #5 suggested that vmware-modules-1.0.0.16-r1 was the first fixed version, and that version isn't marked stable. I have no idea what vmware-any-any is but if you are sure this is fixed in the stable tree then feel free to close the bug.
Hi, sorry, we do indeed have a very misleading build system. There's a shared set of sources (call vmware-any-any-update) for all vmware products, that's updated by a third party (who happens to also work for vmware). When these aren't available, the sources from the binary package are used, but they're usually old and aren't updated until the next product version bump. So, what happens is each product gets a version number (workstation-6 is currently 17) and there's an ebuild for the product based on it's original source. Then, when possible, a -r1 release is made once the vmware-any-any-update code (third party code) has been made compatible with the most recent kernel. The third party source location is bumped in the vmware.eclass, which then affects all packages, including stable ones. Since bumping stable packages without testing is a bad idea, the changes are always made in the overlay first, and then committed to the main tree after a period of testing. Sadly, we haven't found a better way of working this system. After all that it does mean that the stable packages will compile against 2.6.23 without problem. A newer update for 2.6.24 is now testing in the vmware overlay. I'll re-close the bug now, sorry for the confusion...
