webrevs for macosx changes to jdk7u-osx (original) (raw)

Phil Race philip.race at oracle.com
Mon Nov 21 15:28:20 PST 2011

I understand that this isn't 7ux but ultimately this will go to the 7ux forest via an hg push, somewhere along the line before that happens all this needs to be undone. In fact it needs to be undone before we merge the macos port and definitely before we create shippable bits.

It would be wiser to base the client portion off the code we are going to use (the os x port). For all I know (right now) a lot of the same changes will surface in a diff against that and then we can clean up the OS X port so a clean copy goes into 7ux for all the changes to date.

-phil.

On 11/21/2011 2:53 PM, Paul Hohensee wrote:

Note that Mike isn't proposing to push this into the baseline, rather he's proposing to push it into the jdk7u-osx development forest, which is logically a child of jdk7u-dev. We'd like to get an X11-based build going so everyone except the client group can get work done in the jdk7u-osx forest. The hotspot group has already pushed a recent version of hs23 into it.

Paul On 11/21/11 5:41 PM, Phil Race wrote: Hmm, I just realised I didn't read properly and I now see that "The AWT/client code is however taken from the BSD forest."

In other words, these aren't from the macos x port as I'd supposed. That explains a lot and this is maybe even worse than I'd first realised. If these changes are based on the bsd port, I can't be sure without checking if the same files in the same places are even touched in the current macos x port. And likely not in the same way .. so the "client folks" might not even notice these vestiges. And there's enough to do without finding and un-doing or re-doing these .. It is the shared code changes that concern me most as its not like we'll just "rm" that like you could rm the src/bsd directory .. Although I also see a whole bunch of new src/solaris files (inc. many for sound) What's that about ? I'm really (really) uncomfortable with pushing this into mainline. My strong recommendation is to reconsider this whole thing as I believe its going to pollute mainline big time. I think its a big mistake. -phil. On 11/21/2011 2:08 PM, Michael McMahon wrote: Phil,

Thanks for looking at this. The client code in this webrev was intended just as a temporary version to get a basic system up and running. I was thinking that the client folks would replace it all en-masse when they are ready. Hence, the fact that some recent changes might be missing, or otherwise not of ideal quality, might not be such a serious problem. In any case, I'll look at all of these points and make the changes. Thanks, Michael. On 21/11/11 18:53, Phil Race wrote: Michael,

Thanks for preparing this so we can see what the delta is. I have a few comments from a very quick skim over this This includes a whole bunch of "isBSD" checks and the like which I don't think belong in mainline. There also seems to be a bit of schizophrenia around what System.getProperty("os.name"); should return. In one place I see "Mac OS X" http://cr.openjdk.java.net/~michaelm/7113349/1/jdk/modified/src/share/classes/sun/print/PSPrinterJob.java.sdiff.html here I see "Darwin" http://cr.openjdk.java.net/~michaelm/7113349/1/jdk/modified/src/share/classes/sun/font/FontUtilities.java.sdiff.html

I'm also a bit concerned by what looks like its going to remove some only recently added code here :- http://cr.openjdk.java.net/~michaelm/7113349/1/jdk/modified/src/share/classes/sun/awt/FontConfiguration.java.sdiff.html And there "not so clean" comments and changes in shared 2D code where some editing is justified before adding to mainline, and perhaps even examination as to whether its the right change http://cr.openjdk.java.net/~michaelm/7113349/1/jdk/modified/src/share/classes/sun/print/RasterPrinterJob.java.sdiff.html The changes here http://cr.openjdk.java.net/~michaelm/7113349/1/jdk/modified/src/share/native/sun/font/AccelGlyphCache.c.sdiff.html maybe would be better to be just "#include <stdlib.h>" on all platforms. You'd obviously want to make sure it all builds properly on the other platforms but I think at least the Solaris and Linux builds should be OK. In fact such a change was just made in JDK 8 for this case :- http://cr.openjdk.java.net/~michaelm/7113349/1/jdk/modified/src/share/native/sun/awt/medialib/mlibtypes.h.sdiff.html

So I don't think these changes should be pushed into mainline wholesale as is. It needs some review by area owners and perhaps changes beyond those I noticed. -phil. On 11/21/2011 7:46 AM, Michael McMahon wrote: Hi, The following webrevs are an initial set of changes taken from the macosx-port forest to be applied to the jd7u-osx forest at: http://hg.openjdk.java.net/jdk7u/jdk7u-osx/ This will eventually be merged into the main jdk7u-dev forest. Hotspot has already integrated into this forest. This changeset includes the changes (in the jdk and corba) repositories that are needed to build and run openjdk on MacOSX. The AWT/client code is however taken from the BSD forest. So, the native Apple AWT/client code will be integrated later. By default, the VM will run in headless mode. To run with the X windows AWT, set the environment variable AWTTOOLKIT to XToolkit. This is a work in progress and much remains to be done. But, it does build and run. All comments welcome. Thanks, Michael. JDK repo ===== Modified files ------------------ http://cr.openjdk.java.net/~michaelm/7113349/1/jdk/modified/ New files ------------ http://cr.openjdk.java.net/~michaelm/7113349/1/jdk/new/ Corba repo ======= http://cr.openjdk.java.net/~michaelm/7113349/1/corba/webrev/

More information about the macosx-port-dev mailing list