March 2005 - Darwin-dev - lists.apple.com

Re: Standard way to communicate between NKEs and user space?
by Josh Graessley 03 Mar '05

03 Mar '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com -josh On Mar 2, 2005, at 11:27 PM, OL&L Lists wrote: Thanks, _______________________________________________ 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 Kernel control is one common way. It uses sockets in user space to send data to a kernel extension. The kernel extension uses an interface defined in sys/kern_control.h to exchange information with a process running in user space. I am sure this question has been asked many times but can someone point me to the docs that explain the standard way on OS X to enable two-way communication between a NKE KEXT and a user-space app? Michael _______________________________________________ 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/jgraessley% 40apple.com This email sent to jgraessley(a)apple.com smime.p7s

1 0

Standard way to communicate between NKEs and user space?
by OL&L Lists 02 Mar '05

02 Mar '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Thanks, Michael _______________________________________________ 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 sure this question has been asked many times but can someone point me to the docs that explain the standard way on OS X to enable two-way communication between a NKE KEXT and a user-space app? This email sent to site_archiver(a)lists.apple.com

1 0

Re: Virtual memory leaks...
by David Leimbach 02 Mar '05

02 Mar '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Domainkey-signature: a=rsa-sha1; q=dns; c=nofws; s=beta; d=gmail.com; h=received:message-id:date:from:reply-to:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:references; b=J967HJ8f1fncxT0OtqWHvWonu+AMWnOJifRvIi5uikVtU+d3Q55d2vFUICB1lMPmsw41x1UtzKpbzsssjl4SCPMvhSk9qnm69Kity4jlACi/Ca2bDeP7Rlu+kr2ywFJya409mnt2gICdrYCJ/cOqORkxrHxGIj/15elH4SjGPM8= I've done a little vm logging in my time too. I used mach_override [http://rentzsch.com/mach_override/] To catch calls to vm_allocate and vm_deallocate. These are the system calls that actually work behind the scenes of Mac OS X's malloc/free implementation mach_override will let you "hook into" all calls to vm_allocate/vm_deallocate and from there you can call the original function and add any sort of logging before and after the real function runs. The Mac OS X malloc implementation does a lot of "pooling" in that not all malloc calls are going to call vm_allocate and not all free calls are going to call vm_deallocate. If you do some experimentation you'll probably see this for yourself. Also the source is available to libSystem and its constituent parts. Dave On Wed, 02 Mar 2005 12:10:49 -0800, Michael Burbidge <mburbidg(a)adobe.com> wrote: > We have a very large application that uses a couple of dozen libraries > from various sources. Our application is leaking virtual memory. > Periodically, there is a 1 Mb bump in virtual memory, which is never > reclaimed. Over time, several days, the accumulative affects of this > leak are that we run out of virtual memory. We've tried all the > standard tools such as leaks, and MallocDebug, but our application is > too large for these. They either crash or slow our application down to > the point that it is useless. > > I'm wondering if there is someway for me to get my hooks into the > system allocator. If so, everytime I see a 1 Mb bump in virtual memory > I could log the stack to a file. The stack may lead me to the leak. > > Any suggestions would be appreciated. > > Michael- > > _______________________________________________ > 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/leimy2k%40gmail.com > > This email sent to leimy2k(a)gmail.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

Virtual memory leaks...
by Michael Burbidge 02 Mar '05

02 Mar '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Any suggestions would be appreciated. Michael- _______________________________________________ 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… We have a very large application that uses a couple of dozen libraries from various sources. Our application is leaking virtual memory. Periodically, there is a 1 Mb bump in virtual memory, which is never reclaimed. Over time, several days, the accumulative affects of this leak are that we run out of virtual memory. We've tried all the standard tools such as leaks, and MallocDebug, but our application is too large for these. They either crash or slow our application down to the point that it is useless. I'm wondering if there is someway for me to get my hooks into the system allocator. If so, everytime I see a 1 Mb bump in virtual memory I could log the stack to a file. The stack may lead me to the leak. This email sent to site_archiver(a)lists.apple.com

1 0

Re: __attribute__((constructor)) in dylib does not run with DYLD_INSERT_LIBRARIES
by Andreas Tobler 01 Mar '05

01 Mar '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com User-agent: Mozilla Thunderbird 1.0 (Macintosh/20041206) Peter O'Gorman wrote: | So if you build libhello with -bundle instead of -dynamiclib it works | somehow. But this is a no go for me. | I'm not aware of any workaround for panther. There is no workaround for getting libraries inserted with DYLD_INSERT_LIBRARIES to have their static constructors called on panther that I know of. Nor is there any standard that I am aware of which requires this functionality. Well, I can wait, or I _have_ to wait :) Andreas _______________________________________________ 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… Why? Well, the sample I showed was with dlopen, yes. But I need it in general with dylibs. I tried to build libgcj with -bundle but this conflicts with the whole gcc tree. So I really need -dynamiclib. Since you are dlopening the library, and panther uses dlcompat, and dlcompat looks for a symbol named "_init", you can use a hack. Inserting a "void _init() { ; }" into the .c file which has your static constructors should cause them to be called. They must appear in the same translation unit as the static constructor, or you must link your library using the - -single_module flag if you want dyld to call them. If I were you I'd check for different dlopen and dyld behavior on tiger... My need is more or less covered in tiger. I expect some issues but I can't test that often. This email sent to site_archiver(a)lists.apple.com

1 0

Re: Getting entire Darwin source tree via CVS
by Shantonu Sen 01 Mar '05

01 Mar '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com <http://darwinsource.opendarwin.org/tarballs/apsl/xnu-517.11.1.tar.gz> Google for various howtos on this topic. Shantonu On Mar 1, 2005, at 9:21 PM, Peter Seebach wrote: Hmm. What's the equivalent of just the complete kernel tree, then? I'd actually like to be able to run custom kernels... _______________________________________________ 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… In message <C6CD08AE-D7E1-4EA4-9EFB-71F81820A929(a)opendarwin.org>, Shantonu Sen writes: You should think of Darwin like you do Linux, not Net/FreeBSD. There is no single CVS repository that will build all the user and kernel parts of a given Linux distribution (unless you build that CVS repo yourself with cvs imports) -s _______________________________________________ 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/ssen% 40opendarwin.org This email sent to ssen(a)opendarwin.org This email sent to site_archiver(a)lists.apple.com

1 0

Re: Getting entire Darwin source tree via CVS
by seebs＠plethora.net 01 Mar '05

01 Mar '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com In message <C6CD08AE-D7E1-4EA4-9EFB-71F81820A929(a)opendarwin.org>, Shantonu Sen writes: >You should think of Darwin like you do Linux, not Net/FreeBSD. There is >no single CVS repository that will build all the user and kernel parts >of a given Linux distribution (unless you build that CVS repo yourself >with cvs imports) Hmm. What's the equivalent of just the complete kernel tree, then? I'd actually like to be able to run custom kernels... -s _______________________________________________ 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: Getting entire Darwin source tree via CVS
by Shantonu Sen 01 Mar '05

01 Mar '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Shantonu On Mar 1, 2005, at 9:08 PM, Peter Seebach wrote: _______________________________________________ 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 should think of Darwin like you do Linux, not Net/FreeBSD. There is no single CVS repository that will build all the user and kernel parts of a given Linux distribution (unless you build that CVS repo yourself with cvs imports) In message <b80e0dc33e3a5b0e5d578ae5db496e97(a)opendarwin.org>, Kevin Van Vechten writes: Putting the sources in CVS isn't going to make a "build tree" alone. OpenDarwin's mirror of Apple sources offers anonymous rsync access. See <http://darwinsource.opendarwin.org/> for details. Also on that page are the Darwin Build Scripts, which assist compilation of a large number of Darwin projects. I guess I've been spoiled by NetBSD, where a checkout of src gives you a complete buildable source tree with all the files in the right places. :) -s _______________________________________________ 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/ssen% 40opendarwin.org This email sent to ssen(a)opendarwin.org This email sent to site_archiver(a)lists.apple.com

1 0

Re: Getting entire Darwin source tree via CVS
by seebs＠plethora.net 01 Mar '05

01 Mar '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com In message <b80e0dc33e3a5b0e5d578ae5db496e97(a)opendarwin.org>, Kevin Van Vechten writes: >Putting the sources in CVS isn't going to make a "build tree" alone. >OpenDarwin's mirror of Apple sources offers anonymous rsync access. >See <http://darwinsource.opendarwin.org/> for details. Also on that >page are the Darwin Build Scripts, which assist compilation of a large >number of Darwin projects. I guess I've been spoiled by NetBSD, where a checkout of src gives you a complete buildable source tree with all the files in the right places. :) -s _______________________________________________ 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: Getting entire Darwin source tree via CVS
by Kevin Van Vechten 01 Mar '05

01 Mar '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com - Kevin On Mar 1, 2005, at 6:15 PM, Peter Seebach wrote: Is there a way to download the entire Darwin CVS source tree in one shot? I connected to Apple's Darwin CVS repository, but it lists only a few certain modules as being available. Is there a way to download the entire source tree via CVS? _______________________________________________ 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… Putting the sources in CVS isn't going to make a "build tree" alone. OpenDarwin's mirror of Apple sources offers anonymous rsync access. See <http://darwinsource.opendarwin.org/> for details. Also on that page are the Darwin Build Scripts, which assist compilation of a large number of Darwin projects. In message <a06210205be4acef47420(a)[10.250.3.49]>, OL&L Lists writes: There doesn't appear to be, and if you get all the packages from the page, they're not really a build tree. I would love a way to check out a complete tree for a given version automatically; unfortunately, the OpenDarwin people are quite out of date, and Apple doesn't offer a complete CVS tree. This email sent to site_archiver(a)lists.apple.com

1 0