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

Problem when compiling Xprt of X11R7.1
by Peter Dyballa 20 Oct '06

20 Oct '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Hello! and many more like In the end making of Xprt fails: collect2: ld returned 1 exit status make[3]: *** [Xprt] Error 1 The actual collect2 invocation is: -- Greetings Pete _______________________________________________ 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… When on Mac OS X 10.4.8 Xprt is built by GCC 4.0.1 from dynamic and static libraries I get one message /sw/lib/odcctools/bin/ld: multiple definitions of symbol _serverGeneration /usr/X11R7/lib/libXfont.dylib(miscutil.o) definition of _serverGeneration ../../dix/.libs/libdix.a(globals.o) definition of _serverGeneration in section (__DATA,__data) /sw/lib/odcctools/bin/ld: multiple definitions of symbol _Xalloc /usr/X11R7/lib/libXfont.dylib(miscutil.o) definition of _Xalloc ../../os/.libs/libos.a(utils.o) definition of _Xalloc in section (__TEXT,__text) /sw/lib/odcctools/bin/ld: multiple definitions of symbol _Xcalloc /usr/X11R7/lib/libXfont.dylib(miscutil.o) definition of _Xcalloc ../../os/.libs/libos.a(utils.o) definition of _Xcalloc in section (__TEXT,__text) What is/can be the cause? Warnings like these above, for both sections, happen also for other targets (Xvfb, Xnest, Xdmx), but these get made. In making libdix.a warnings are emitted: ar cru .libs/libdix.a .libs/atom.o .libs/colormap.o .libs/ cursor.o .libs/devices.o .libs/dispatch.o .libs/dixfonts.o .libs/ dixutils.o .libs/events.o .libs/extension.o .libs/ffs.o .libs/ gc.o .libs/globals.o .libs/glyphcurs.o .libs/grabs.o .libs/ initatoms.o .libs/main.o .libs/pixmap.o .libs/privates.o .libs/ property.o .libs/resource.o .libs/swaprep.o .libs/swapreq.o .libs/ tables.o .libs/window.o .libs/strcasecmp.o /sw/lib/odcctools/bin/ranlib: file: .libs/libdix.a(strcasecmp.o) has no symbols ranlib .libs/libdix.a ranlib: file: .libs/libdix.a(strcasecmp.o) has no symbols creating libdix.la which also happens for some other (static) libraries. CFLAGS used are: -pipe -fPIC -O0, in LDFLAGS -dead_strip is used; G4/7450, MACOSX_DEPLOYMENT_TARGET is set to 10.4. Thread model: posix gcc version 4.0.1 (Apple Computer, Inc. build 5363) /usr/libexec/gcc/powerpc-apple-darwin8/4.0.1/collect2 -dynamic -arch ppc7450 -dead_strip -macosx_version_min 10.4 - weak_reference_mismatches non-weak -o Xprt -lcrt1.o -L/sw/lib/ freetype219/lib -L/usr/local/lib -L/sw/lib/ncurses -L/sw/lib -L/Users/ pete/Quellen/X11R7.1/xserver -L/usr/X11R7/lib -L/usr/lib/gcc/powerpc- apple-darwin8/4.0.1 -L/usr/lib/gcc/powerpc-apple-darwin8/4.0.1 -L/usr/ lib/gcc/powerpc-apple-darwin8/4.0.1/../../.. Xprt-attributes.o Xprt- AttrValid.o Xprt-ddxInit.o Xprt-Init.o Xprt-mediaSizes.o Xprt-Oid.o Xprt-spooler.o Xprt-Util.o Xprt-miinitext-wrapper.o Xprt-dpmsstubs- wrapper.o Xprt-fbcmap.o /usr/X11R7/lib/libX11.dylib /usr/X11R7/lib/ libxcb-xlib.dylib /usr/X11R7/lib/libxcb.dylib /usr/X11R7/lib/ libXfont.dylib /sw/lib/freetype219/lib/libfreetype.dylib /usr/X11R7/ lib/libfontenc.dylib -lz /usr/X11R7/lib/libXdmcp.dylib /usr/X11R7/lib/ libXau.dylib ../../dix/.libs/libdix.a ../../xkb/.libs/libxkb.a ../../ xkb/.libs/libxkbstubs.a ../../dbe/.libs/libdbe.a ../../XTrap/.libs/ libxtrap.a ../../record/.libs/librecord.a ../../composite/.libs/ libcomposite.a ../../randr/.libs/librandr.a ../../xfixes/.libs/ libxfixes.a ../../damageext/.libs/libdamageext.a ../../Xi/.libs/ libXi.a ../../miext/damage/.libs/libdamage.a ../../miext/cw/.libs/ libcw.a ../../os/.libs/libcwrapper.a ../../os/.libs/libos.a ps/.libs/ libps.a raster/.libs/libraster.a pcl/.libs/libpcl.a pcl-mono/.libs/ libpcl.a ../../fb/.libs/libfb.a ../../render/.libs/librender.a ../../ mi/.libs/libmi.a ../../Xext/.libs/libXext.a -lm -lgcc_s.10.4 -lgcc - lSystemStubs -lSystem Some day we may discover how to make magnets that can point in any direction. This email sent to site_archiver(a)lists.apple.com

1 0

Re: BSD sockets calls on OSX Tiger and thread safety
by Vincent Lubet 18 Oct '06

18 Oct '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Vincent On Oct 18, 2006, at 8:46 AM, Dave Zarzycki wrote: The APIs are thread safe. davez On Oct 18, 2006, at 8:38 AM, jmzorko(a)mac.com wrote: Hello, all ... Regards, John Falling You - exploring the beauty of voice and sound http://www.fallingyou.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… To illustrate what Dave said, if you call send() on the same socket file descriptor from 2 threads you have no guarantee on what data will be sent first. If ordering is important your program, need to provide your own synchronization mechanism. Sending from 2 threads on 2 distinct file descriptors is not a problem (except if the 2 file descriptors are logically related for a specific application but that's rather rare...) But like all thread safe APIs, one can still use them in an unsafe manner. How thread-safe is OSXs' BSD sockets implementation? Can I call select(), recv(), send(), socket(), setsockopt(), etc. from multiple threads simultaneously without locking, or are there things I need to be aware of? This email sent to site_archiver(a)lists.apple.com

1 0

Re: BSD sockets calls on OSX Tiger and thread safety
by Dave Zarzycki 18 Oct '06

18 Oct '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com The APIs are thread safe. davez On Oct 18, 2006, at 8:38 AM, jmzorko(a)mac.com wrote: Hello, all ... Regards, John Falling You - exploring the beauty of voice and sound http://www.fallingyou.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/zarzycki%40apple.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… But like all thread safe APIs, one can still use them in an unsafe manner. How thread-safe is OSXs' BSD sockets implementation? Can I call select(), recv(), send(), socket(), setsockopt(), etc. from multiple threads simultaneously without locking, or are there things I need to be aware of? This email sent to zarzycki(a)apple.com This email sent to site_archiver(a)lists.apple.com

1 0

BSD sockets calls on OSX Tiger and thread safety
by jmzorko＠mac.com 18 Oct '06

18 Oct '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Hello, all ... Regards, John Falling You - exploring the beauty of voice and sound http://www.fallingyou.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… How thread-safe is OSXs' BSD sockets implementation? Can I call select(), recv(), send(), socket(), setsockopt(), etc. from multiple threads simultaneously without locking, or are there things I need to be aware of? This email sent to site_archiver(a)lists.apple.com

1 0

Re: Re: pthread_cancel() during read...
by mchenry2＠rossumtech.com 18 Oct '06

18 Oct '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com User-agent: Web-Based Email 4.8.6 Hi All, On May 30, Per Mildner offered up a small program (included below) which seemed to indicated that (even with _APPLE_C_SOURCE defined) that a 'read' call was not pthread_cancel(able). The thread ended with Matt Watson of apple being able to reproduce the problem, but no further info... I'm wondering if there is any additional info on this problem. In particular, I'm wondering if I can expect different behavior in jaguar. If so, I'll be satisfied with my very crude work-around. For what its worth, I verified that Per's program still fails to work properly on my system... Darwin R29Tiger-4.local 8.7.0 Darwin Kernel Version 8.7.0: Fri May 26 15:20:53 PDT 2006; root:xnu-792.6.76.obj~1/RELEASE_PPC Power Macintosh powerpc compiled via "gcc -D_APPLE_C_SOURCE foo.c" Thanks much, -mike Below is a copy of Per's original program/email... #include <assert.h> #include <pthread.h> #include <stdio.h> #include <stdlib.h> #include <unistd.h> static int exit_status = EXIT_FAILURE; void * func(void *arg) { char buf; exit_status = EXIT_SUCCESS; while (1) { ssize_t res; fprintf(stderr, "calling read..");fflush(stderr); res = read(STDIN_FILENO, &buf, 1); fprintf(stderr, ".called read()==%ld\n", (long)res);fflush(stderr); } return NULL; } int main(void) { pthread_t t; void *retval; assert(!pthread_create(&t, NULL, func, NULL)); sleep(1); assert(!pthread_cancel(t)); assert(!pthread_join(t, &retval)); assert(PTHREAD_CANCELED == retval); return exit_status; } # The first <CR> below is for passing the line to shell (all this is done in bash from within xemacs) wormwood2:~/sicstus/sicstus4wormwood2 perm$ gcc -D_APPLE_C_SOURCE foo.c && ./a.out <CR> calling read.. HANGS UNTIL an additional <CR> entered wormwood2:~/sicstus/sicstus4wormwood2 perm$ gcc -D_APPLE_C_SOURCE foo.c && ./a.out <CR> <CR> calling read...called read()==1 calling read.. HANGS HERE until an additional <CR> entered _______________________________________________ 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: Need help getting CoreFoundation Lite to work on Win32
by Pete Gontier 18 Oct '06

18 Oct '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Thread-index: AcbyWlmMl9ZbYl5NEduD3QARJG7C6A== Thread-topic: Need help getting CoreFoundation Lite to work on Win32 User-agent: Microsoft-Entourage/11.2.3.060209 _______________________________________________ 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

Need help getting CoreFoundation Lite to work on Win32
by ADAM FRANKOWSKI 17 Oct '06

17 Oct '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=s1024; d=rogers.com; h=Message-ID:Received:Date:From:Subject:To:MIME-Version:Content-Type:Content-Transfer-Encoding; b=DRcLx1nMoVlGGKO9WZksjO/CusrQV7Sq/sQS8Wv9EkuaJeFYBgmFFUsOhJMWiTDpCWEfFzoJwpYV/S6QfVaDUweUgrkxousUSOZaQnihPI4JKxCuIbO75/7YWoYC0iHAfUI47fFYuTMnomcCqOXOtlW8bnysTDFitTGO/uJvBpY= ; Hello, Has anyone been able to get CFLite (CF-386.27) to compile on Windows? I would appreciate any information that could help me get it work. Thanks in advance, adam _______________________________________________ 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

Basic questions on building from darwin source
by Rick Mann 16 Oct '06

16 Oct '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com --- ================================== twinaero:~/darwin/lsof-8 rmann$ make echo "Creating pseudo-copy of sources in the build directory..."; Creating pseudo-copy of sources in the build directory... Configuring lsof... Unknown Darwin release: 8.8.1 Assuming Darwin 1.4 ----------------------------------------------------------------- It's necessary to check out some XNU kernel header files from the open source CVS repository. See this URL for more information: http://www.opensource.apple.com/tools/cvs/docs.html You must first specify your registered Darwin user name in order to access the open source repository. The companion Darwin user name password must be stored in ~/.cvspass. You must also specify the open source branch from which the headers should be obtained. What is your registered Darwin user name? me(a)me.com -------------------------------------------------------------------- Now you must specify the open source branch of the CVS repository from which the header files will be checked out. These are some likely open source branches: Darwin 1.2.1: xnu-3-1 Darwin 1.3.1: xnu-4-2 Darwin 1.4.1: xnu-9-1 Mac OS X 10.0 (4K78): Apple-123-5 Mac OS X 10.0.4 (4Q12): Apple-124-13 Mac OS X 10.1 (5G64): Apple-201 Mac OS X 10.1.1 (5M28): Apple-201-5 Mac OS X 10.1.2 (5P48): Apple-201-14 Please specify a branch to use. (It needn't be one of the suggested ones.) What branch? _______________________________________________ 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… How do I build parts of the Darwin open source? Are there useful instructions anywhere? Is there any easy way to check out ALL of the source (everything necessary to build x86) for a particular version, rather than checking out each module separately? I'm finding Apple's open-source stuff to be very frustrating to use. I want to develop some functionality like lsof, so I got the source and started looking through it. It's very convoluted, and supports many many platforms, so determining how it does the one thing I want to do on Darwin is a pain. I thought it would be better if I could build it in XCode and step through it in the debugger, but of course it doesn't have an XCode project. So I thought I'd just build it with make, but that wants to check out more stuff. That's where I ran into problems. First it tells me to look here: http://www.opensource.apple.com/tools/ cvs/docs.html, but that just takes me to http://developer.apple.com/ opensource/index.html, which is very generic. The CVS link on that page doesn't help much either. The problem is that the build script asks me what branch I want to use, suggests a few. I want the latest, but I don't know what the branch names are. I tried a few of the suggested names, but none that I tried worked. FWIW, I've included a transcript of the make process so far. Thanks for any help! This email sent to site_archiver(a)lists.apple.com

1 0

Re: transfer dev environment between Windows and OS X?
by j o a r 16 Oct '06

16 Oct '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On 16 okt 2006, at 20.18, The Spaminator wrote: j o a r _______________________________________________ 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 Will resource forks cause problems in the windows environment? theoriticaly, they are just another file, and as long as they are stored with their corresponding data fork, copying back and forth should not have any negative effects right? What type of resource forks are you interested in preserving? I think that most SCM systems ignore (as in strip and discard) resource forks. This is usually not a problem, since source files, project files and other developer resources typically don't use resource forks. smime.p7s

1 0

Re: transfer dev environment between Windows and OS X?
by The Spaminator 16 Oct '06

16 Oct '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com User-agent: Thunderbird 1.5.0.7 (Windows/20060909) This is not strictly a dev question, but I thought the answer might be found here. I have a Macintosh development environment, and our source code repository is Windows based. Will resource forks cause problems in the windows environment? theoriticaly, they are just another file, and as long as they are stored with their corresponding data fork, copying back and forth should not have any negative effects right? These files will never be modified outside of a Macintosh environment, they are transferred to a windows machine and stored there only. I just want to make sure that I got it right. I have copied an environment to windows, and then copied it back (extra careful to make sure the resource forks were also copied), and i could build fine. Can I now assume that storing on windows is ok as long as the resource forks are also stored? are there any other gotchas? _______________________________________________ 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