May 2006 - Darwin-dev - lists.apple.com

Re: Experiencing build issues with some libraries
by Dale Walsh 18 May '06

18 May '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Mta-interface: amavisd-new-2.3.3 (2005-08-22) + Maia Mailguard 1.1.0 at daleenterprise.com -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On May 18, 2006, at 03:27 , Martin Costabel wrote: Dale Walsh wrote: [] Yes, I examined the files and it covers both architectures. -- Martin - -- Dale -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2.2 (Darwin) iD8DBQFEbC7YiD9DTPch4RQRAkE/AJ4vHpaS3AV1k5xd9JZy8cJpAU5ZxACgiuAc 9D1K1XwvcAI8pRMnwV3Vy6w= =MR6Y -----END PGP SIGNATURE----- _______________________________________________ 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… dylib: Library not loaded: /usr/X11R6/lib/libreetype.6.dylib Reference from: /private/var/tmp/graphics/graphics-1.obj/src/ caretaker Reason: no suitable image found. Did find: /usr/X11R6/lib/libfreetype.6.dylib: can't map What does "file" say about that dylib? Does it correspond to your OS arch? /usr/X11R6/lib/libfreetype.6.3.dylib: Mach-O fat file with 2 architectures /usr/X11R6/lib/libfreetype.6.3.dylib (for architecture ppc): Mach-O dynamically linked shared library stub ppc /usr/X11R6/lib/libfreetype.6.3.dylib (for architecture i386): Mach- O dynamically linked shared library stub i386 /usr/X11R6/lib/libexpat.0.4.dylib: Mach-O fat file with 2 architectures /usr/X11R6/lib/libexpat.0.4.dylib (for architecture ppc): Mach-O dynamically linked shared library stub ppc /usr/X11R6/lib/libexpat.0.4.dylib (for architecture i386): Mach-O dynamically linked shared library stub i386 This email sent to site_archiver(a)lists.apple.com

1 0

Re: Experiencing build issues with some libraries
by Martin Costabel 18 May '06

18 May '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com User-agent: Thunderbird 1.5.0.2 (Macintosh/20060308) Dale Walsh wrote: [] dylib: Library not loaded: /usr/X11R6/lib/libreetype.6.dylib Reference from: /private/var/tmp/graphics/graphics-1.obj/src/caretaker Reason: no suitable image found. Did find: /usr/X11R6/lib/libfreetype.6.dylib: can't map What does "file" say about that dylib? Does it correspond to your OS arch? -- Martin _______________________________________________ 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: MacCentral Post
by Corey O'Connor 18 May '06

18 May '06

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:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; b=SUFgdnjulVB/BZzypmdqnQ5/MjC8KlxQAAaGaBxJVGzuUDXtgBZrAq+wbS+Ks1WTRdsgF5WoHxHkr0QJnlLvjTphBc5zeu1xkV4S8t2Wdjkc6NS16npHARv7A6zbJz5JuP9jSimvqgjM1TibWBevJLYRpNitFopGMCDRYnGImbE= I'm working (with others here at ERTOS) on an experimental port of Darwin kernel to the L4 microkernel. It's all open source, see here: http://www.ertos.nicta.com.au/software/darbat/ Our latest achievements include booting on the EFI release hardware (x86 imac and mac mini), and hopefully we'll have loading of binary drivers working Real Soon Now. We're working on another source release. Looks like a great project! Though, at this rate it looks like this L4 kernel has a better chance of becoming the opensource kernel for X86 Mac OS X than the Apple Mach kernel! ;-) I think I'll start studying L4... -- -Corey O'Connor _______________________________________________ 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… On 5/17/06, Charles Gray <cgray.darwin(a)cse.unsw.edu.au> wrote: This email sent to site_archiver(a)lists.apple.com

1 0

Re: intercepting system calls works on tiger(10.4.5) but not on panther(10.3.9)
by Puthali H.B 18 May '06

18 May '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com ok, i'm not very much clear on this, so please pardon, if its very obvious. Actually I don't use RTLD_NEXT, but rather get a handle for libSystem.dylib using dlopen and use that handle in dlsym. And if I use the resulting function handle from dlsym, shouldn't it call the function from libSystem, rather than my own? as it does in tiger? and will try again with syscall, hopefully it will work. >>> Peter O'Gorman <peter(a)pogma.com> 5/18/2006 10:38:38 am >>> On Wed, 2006-05-17 at 22:50 -0600, Puthali H.B wrote: > > But on panther, I face two kinds of problems, depending on whether I > use dlopen or syscall to call the underlying system connect. > > If I use dlopen, followed by dlsym, then it loops in the over-ridden > connect, which is very weird. Not really, on panther dlsym(RTLD_NEXT,"connect") is always going to give you the first connect function it finds when force flat namespace is in effect. So you get into an infinite loop, exhaust stack and crash. This is fixed in tiger. > > where as, if I use syscall, then it throws up the following- Again, works for me: peter$ cat connect.c #include <stdio.h> #include <sys/syscall.h> int connect(int s, const struct sockaddr *name, int namelen) { fprintf(stderr,"Connect Called\n"); return syscall(SYS_connect,s,name,namelen); } peter$ cc -dynamiclib -o test_connect.dylib connect.c peter$ env DYLD_INSERT_LIBRARIES=./test_connect.dylib \ DYLD_FORCE_FLAT_NAMESPACE=1 ftp ftp.pogma.com Connect Called Connected to ftp.pogma.com. I think you probably want to avoid dlsym RTLD_NEXT et.al. on panther and just use the syscalls. Peter _______________________________________________ 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: Experiencing build issues with some libraries
by Dale Walsh 18 May '06

18 May '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Mta-interface: amavisd-new-2.3.3 (2005-08-22) + Maia Mailguard 1.1.0 at daleenterprise.com -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 - -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Kevin Van Vetchen, any ideas how to resolve this issue?? The lack of a buildable XFree86 or X11 project doesn't help much either. - - -- Dale On May 16, 2006, at 08:58 , Dale Walsh wrote: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Is there a way to fix this? - -- Dale -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2.2 (Darwin) This email sent to dale(a)daleenterprise.com - -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2.2 (Darwin) iD8DBQFEa/QsiD9DTPch4RQRArdAAJ92kAsMLa1lW8F+bUL9KuuyKi1FfgCgqFCU BWB6KMsjINCRZRlzW7jtsN0= =RIIB - -----END PGP SIGNATURE----- -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2.2 (Darwin) iD8DBQFEbAIsiD9DTPch4RQRAmlaAKCLmXbWxz6ZACs6l9MAqsYKC9aA4ACcCbma SH6wndOvyar68h7RVO75oFU= =0wZj -----END PGP SIGNATURE----- _______________________________________________ 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… Since the files are provided for the build root I would expect them to be function yet any time I try linking to these files I get the "can't map" problem which makes building impossible. I'm trying to build a project that is dependent on some librarires in /usr/X11R6/lib. When ever a library in this directory is linked during the build phase it fails. Here is the output. dylib: Library not loaded: /usr/X11R6/lib/libreetype.6.dylib Reference from: /private/var/tmp/graphics/graphics-1.obj/src/ caretaker Reason: no suitable image found. Did find: /usr/X11R6/lib/libfreetype.6.dylib: can't map I also get the same for /usr/X11R6/lib/libexpat.0.dylib and other libraries which are required by the project. iD8DBQFEacyOiD9DTPch4RQRAr/GAKC81Bbhkl7gFrxYjBfhG9IwujIo6wCgoL4k t1ZKgy9XE3vV9n007bV/PHU= =sCNk -----END PGP SIGNATURE----- _______________________________________________ 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/dale% 40daleenterprise.com This email sent to site_archiver(a)lists.apple.com

1 0

Re: intercepting system calls works on tiger(10.4.5) but not on panther(10.3.9)
by Peter O'Gorman 18 May '06

18 May '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On Wed, 2006-05-17 at 22:50 -0600, Puthali H.B wrote: > > But on panther, I face two kinds of problems, depending on whether I > use dlopen or syscall to call the underlying system connect. > > If I use dlopen, followed by dlsym, then it loops in the over-ridden > connect, which is very weird. Not really, on panther dlsym(RTLD_NEXT,"connect") is always going to give you the first connect function it finds when force flat namespace is in effect. So you get into an infinite loop, exhaust stack and crash. This is fixed in tiger. > > where as, if I use syscall, then it throws up the following- Again, works for me: peter$ cat connect.c #include <stdio.h> #include <sys/syscall.h> int connect(int s, const struct sockaddr *name, int namelen) { fprintf(stderr,"Connect Called\n"); return syscall(SYS_connect,s,name,namelen); } peter$ cc -dynamiclib -o test_connect.dylib connect.c peter$ env DYLD_INSERT_LIBRARIES=./test_connect.dylib \ DYLD_FORCE_FLAT_NAMESPACE=1 ftp ftp.pogma.com Connect Called Connected to ftp.pogma.com. I think you probably want to avoid dlsym RTLD_NEXT et.al. on panther and just use the syscalls. Peter _______________________________________________ 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 signature.asc

1 0

Re: intercepting system calls works on tiger(10.4.5) but not on panther(10.3.9)
by Puthali H.B 18 May '06

18 May '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com I essentially do the same thing. Only different thing is, I call the system connect call from my connect. I'm attaching the source file, which I compile it as gcc -dynamiclib -o test_connect.dylib -fPIC test_connect.c Then in the terminal, I do - export DYLD_INSERT_LIBRARIES=test_connect.dylib export DYLD_FORCE_FLAT_NAMESPACE=YES and then I run - ftp ftp.gnu.org On tiger, the control first comes into my over-ridden connect, and then the system connect call gets called without any hitch, and FTP goes through. But on panther, I face two kinds of problems, depending on whether I use dlopen or syscall to call the underlying system connect. If I use dlopen, followed by dlsym, then it loops in the over-ridden connect, which is very weird. where as, if I use syscall, then it throws up the following- dyld: FTP multiple definitions of symbol _connect test_connect.dylib(ccyRQhrn.o) definition of _connect /usr/lib/libSystem.B.dylib(connect.So) definition of _connect Trace/BPT trap Any idea on how I can over-come this? - Puthali >>> Peter O'Gorman <peter(a)pogma.com> 5/17/2006 7:48:00 pm >>> On Tue, 2006-05-16 at 23:01 -0600, Puthali H.B wrote: > Actually I've a dynamic library which over-rides system calls, namely > connect,recv,send,sendto,accept... > and when I preload this on Tiger, using DYLD_INSERT_LIBRARIES AND > DYLD_FORCE_FLAT_NAMESPACE, everything works fine. > > When I pre-loaded this library in terminal and ran FTP got the > following error- > > dyld: FTP multiple definitions of symbol _connect > test_connect.dylib(ccyRQhrn.o) definition of _connect > /usr/lib/libSystem.B.dylib(connect.So) definition of _connect > Trace/BPT trap Works for me on panther (10.3.9): peter$ cc -dynamiclib -o test_connect.dylib connect.c peter$ env DYLD_INSERT_LIBRARIES=./test_connect.dylib DYLD_FORCE_FLAT_NAMESPACE=1 ftp ftp.pogma.com Connect Called ftp: connect: Unknown error: 0 ftp> quit peter$ cat connect.c #include <stdio.h> int connect() { fprintf(stderr,"Connect Called\n"); return 1; } I assume that you're doing something different? Peter _______________________________________________ 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 test_connect.c

1 0

Re: MacCentral Post
by Shawn Erickson 18 May '06

18 May '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On May 17, 2006, at 7:57 AM, Andrew Gallatin wrote: Perhaps people here may want to comment. Admittedly, I'm a lurker more then a kernel hacker, but I thought this was addressed weeks/ months ago with the recent release of the source of the kernel for Intel. The Intel kernel sources have never been released. Or at least there is no xnu in the 10.4.{4,5,6}.x86 download pages at http://www.opensource.apple.com/darwinsource/ -Shawn _______________________________________________ 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… Lance Westerhoff writes: It is a shame the kernel is now officially closed. "now officially closed" ... you care to show me an Apple statement to that fact? This email sent to site_archiver(a)lists.apple.com

1 0

Re: MacCentral Post
by Charles Gray 18 May '06

18 May '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com User-agent: Debian Thunderbird 1.0.7 (X11/20051017) On 17/05/06, Marco Thoma <thomasoft(a)ggs.ch> wrote: thank you. but what then is the benefit that darwin is open source? what do all the open source darwin devs then change?! Speak for yourself, thanks :) http://www.ertos.nicta.com.au/software/darbat/ Chuck _______________________________________________ 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… Finlay Dobbie wrote: There are no "open source darwin devs". I'm working (with others here at ERTOS) on an experimental port of Darwin kernel to the L4 microkernel. It's all open source, see here: Our latest achievements include booting on the EFI release hardware (x86 imac and mac mini), and hopefully we'll have loading of binary drivers working Real Soon Now. We're working on another source release. So yeah, there are most certainly "open source darwin devs", and we're not just tinkering with a few compiler arguments. If Apple do decide to stop releasing their kernel source, this will be a major headache for us. There doesn't seem to be anything but speculation in this macworld article, however. This email sent to site_archiver(a)lists.apple.com

1 0

Re: intercepting system calls works on tiger(10.4.5) but not on panther(10.3.9)
by Kevin Van Vechten 18 May '06

18 May '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com - Kevin On May 17, 2006, at 7:18 AM, Peter O'Gorman wrote: On Tue, 2006-05-16 at 23:01 -0600, Puthali H.B wrote: Actually I've a dynamic library which over-rides system calls, namely connect,recv,send,sendto,accept... and when I preload this on Tiger, using DYLD_INSERT_LIBRARIES AND DYLD_FORCE_FLAT_NAMESPACE, everything works fine. When I pre-loaded this library in terminal and ran FTP got the following error- dyld: FTP multiple definitions of symbol _connect test_connect.dylib(ccyRQhrn.o) definition of _connect /usr/lib/libSystem.B.dylib(connect.So) definition of _connect Trace/BPT trap Works for me on panther (10.3.9): peter$ cc -dynamiclib -o test_connect.dylib connect.c peter$ env DYLD_INSERT_LIBRARIES=./test_connect.dylib DYLD_FORCE_FLAT_NAMESPACE=1 ftp ftp.pogma.com Connect Called ftp: connect: Unknown error: 0 ftp> quit peter$ cat connect.c #include <stdio.h> int connect() { fprintf(stderr,"Connect Called\n"); return 1; } I assume that you're doing something different? _______________________________________________ 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've used this approach for intercepting syscalls successfully (on 10.3.9), you can see what I've done here: <http://cvs.opendarwin.org/cgi-bin/cvsweb.cgi/projects/darwinbuild/ darwintrace/> Peter _______________________________________________ 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/kevin% 40opendarwin.org This email sent to kevin(a)opendarwin.org This email sent to site_archiver(a)lists.apple.com

1 0