October 2004 - Darwin-dev - lists.apple.com

Re: problems compiling Libsystem
by Kevin Van Vechten 03 Oct '04

03 Oct '04

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… This email sent to site_archiver(a)lists.apple.com

1 0

problems compiling Libsystem
by FarVibes 03 Oct '04

03 Oct '04

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… This email sent to site_archiver(a)lists.apple.com

1 0

Re: mmap modified data not to be swapped out but saved?
by Ben Dougall 03 Oct '04

03 Oct '04

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On Saturday, October 2, 2004, at 10:58 pm, Dylan McNamee wrote: A simple experiment should validate this hypothesis: try it, and drive some paging! ok thanks, ben. _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… Ben Dougall asked: with mmap once you've read in some of a file and made some modifications is there anyway to avoid those modifications being swapped out to the backing store? what would be preferable is in the event of the system needing more memory, rather than swapping out those modifications, save them directly to the original file -- is that what happens now, by default maybe, or is there anyway of doing that? While I haven't examined Darwin's code lately, I am pretty familiar with the Mach's VM code from years past, and my memory says that what you want to have happen is supported. (At pageout, "dirty" pages are written back to the file backing the mapped region, which is a file-backed Memory Object). Revisiting the mmap man page reminds me that this might only be the case if the region is mmapped including the MAP_SHARED flag. right, it'd certainly make sense if that is an option as otherwise, if the modified data is definitely for saving, it could easily get saved to the drive, then loaded again, unnecessarily, then saved to the file. rather than just straight to the file. This email sent to site_archiver(a)lists.apple.com

1 0

Re: quick Open Directory question
by Jason Townsend 03 Oct '04

03 Oct '04

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com From: Finlay Dobbie <finlay(a)dobbie.net> For dsOpenDirProxy, you're supposed to provide a tDataNode and two tDataBuffers for authentication purposes. How are you meant to allocate those without a dir ref? Or am I going to have to create a throw away local dir reference to allocate those structures, open the remote directory proxy, then close the local dir reference? -Jason _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… If you look at the sources to the DirectoryService framework, you'll notice that the dir ref is ignored for the various allocation functions. So you can simply pass in zero for that when calling dsDataBufferAllocate() for example. http://www.opensource.apple.com/darwinsource/10.3.5/ DirectoryService-257.1/APIFramework/DirServicesUtils.cpp This email sent to site_archiver(a)lists.apple.com

1 0

Re: mmap modified data not to be swapped out but saved?
by Dylan McNamee 02 Oct '04

02 Oct '04

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com with mmap once you've read in some of a file and made some modifications is there anyway to avoid those modifications being swapped out to the backing store? what would be preferable is in the event of the system needing more memory, rather than swapping out those modifications, save them directly to the original file -- is that what happens now, by default maybe, or is there anyway of doing that? A simple experiment should validate this hypothesis: try it, and drive some paging! Have fun! Dylan McNamee _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… Ben Dougall asked: While I haven't examined Darwin's code lately, I am pretty familiar with the Mach's VM code from years past, and my memory says that what you want to have happen is supported. (At pageout, "dirty" pages are written back to the file backing the mapped region, which is a file-backed Memory Object). Revisiting the mmap man page reminds me that this might only be the case if the region is mmapped including the MAP_SHARED flag. This email sent to site_archiver(a)lists.apple.com

1 0

mmap modified data not to be swapped out but saved?
by Ben Dougall 02 Oct '04

02 Oct '04

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com thanks, ben. _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… with mmap once you've read in some of a file and made some modifications is there anyway to avoid those modifications being swapped out to the backing store? what would be preferable is in the event of the system needing more memory, rather than swapping out those modifications, save them directly to the original file -- is that what happens now, by default maybe, or is there anyway of doing that? This email sent to site_archiver(a)lists.apple.com

1 0

Re: gcc3.4 plus Apple's gcc3.3
by Bill Northcott 01 Oct '04

01 Oct '04

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Enjoy Bill Northcott From: jmzorko(a)mac.com Subject: gcc3.4 plus Apple's gcc3.3 I am involved in porting a C++ class framework we use at work to OSX / Darwin PPC so we can target OSX as well as Win32 and Linux. This code demands gcc3.4, which shouldn't be an issue (gcc3.4 "make bootstrap" seemed to work just fine on my 1.33GHz 12" PowerBook), but I would like to, if possible, use gcc-select to choose between Apple's gcc3.3 and FSFs gcc3.4. Has anyone else done this? I've not researched how I would approach doing this yet (this posting is part of that research <smile>). _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… You can easily build and install Apple versions of gcc-3.4 and gcc-4.0 (nee 3.5). Just go to the FSF gcc site at gcc.gnu.org. Go to the cvs section to get code. What you need is all on the apple-ppc-branch cvs tag. If you just use apple-ppc-branch tag you will get the latest state of the branch which may not build. A better approach is to look at the web cvs. Down the bottom of the pages within the compiler code you will find a pop-up which lists all the tags. I think apple-gcc-17xx tags below about 1720 are gcc-3.4 and above that are gcc-4.0 (3.5). Use the latest of those tags. You will find the most recent Xcode compilers identify themselves with these same tag numbers: [PBG4-BN:JAGS/JAGS-0.65/build] billn% gcc-3.4 -v Reading specs from /usr/lib/gcc/powerpc-apple-darwin/3.4/specs gcc version 3.4 20031014 (Apple Computer, Inc. build 1716) Once you have the code just follow the instructions in README.APPLE to build and install. This will not damage existing gcc-2.95, gcc-3.2 or gcc-3.3 installations. You can change the default compiler using gcc_select as documented in the developer tools. To build these recent compilers, you must have the latest Xcode-1.5. With a little futzing on the latest versions yoou can also build other languages and handy extras like libffi and the GNU Objective-C runtime. In my usage, I have found these Apple compilers essentially identical to their FSF counterparts. They work perfectly with Xcode and all the other Apple tools. This email sent to site_archiver(a)lists.apple.com

1 0

Re: gcc3.4 plus Apple's gcc3.3
by jmzorko＠mac.com 01 Oct '04

01 Oct '04

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Stan, It's just a shell script that you could hack up, but I don't recommend you try - FSF 3.4 is missing dozens of features essential to building Mac apps, such as framework path searching, alignment pragmas, and similar kinds of bits. Some of those features are needed by system headers and could cause interesting (heh) runtime problems. Ultra portable pure ISO/K&R C code, such as what GCC itself is written in (for reasons that should be obvious), will be OK, but just one Cocoa or Carbon call can mess you up. So I recommend you install gcc-3.4 under its own name only, and have all of your 3.4-using code ask for 3.4 explicitly by name. Regards, John Falling You - exploring the beauty of voice and sound New album, "Touch", available now http://www.magnatune.com/artists/falling_you _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… I am involved in porting a C++ class framework we use at work to OSX / Darwin PPC so we can target OSX as well as Win32 and Linux. This code demands gcc3.4, which shouldn't be an issue (gcc3.4 "make bootstrap" seemed to work just fine on my 1.33GHz 12" PowerBook), but I would like to, if possible, use gcc-select to choose between Apple's gcc3.3 and FSFs gcc3.4. Has anyone else done this? I've not researched how I would approach doing this yet (this posting is part of that research <smile>). Will do. I won't be making any Cocoa / Carbon calls ... at least not at this stage. Perhaps when Tiger comes out, it will sport a super-cool Apple version of gcc3.4. Of course, by that time, our team leader slash compiler jockey slash FSF-just-released-a-new-version-of-gcc-3-minutes-ago-why-the-hell- aren't-you-using-it-yet? will demand that we all use gcc4.0.x.y.z :-) This email sent to site_archiver(a)lists.apple.com

1 0