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

perl 5.8.1 issue on 10.3.9 server [ semi solved ]
by Dale Walsh 29 Nov '05

29 Nov '05

site_archiver(a)lists.apple.com Delivered-To: Darwin-dev(a)lists.apple.com Mta-interface: amavisd-new-2.3.3 (20050822) at daleenterprise.com -- Dale _______________________________________________ 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… Working with an associate we noticed that XCode was also experiencing some build issues and as it turned out, we have some software installed that is affecting the environment is some way so I snagged an old B&W G3-400 to use as a build platform. I wasn't able to nail down the specific software at fault however, after a fresh install on the snagged machine I can report that darwinbuild appears to be operating without issues however. While I build the Perl software I'm experiencing an issue that involves setrgid and setruid not being implemented. Seems that this isn't being supported for some reason however I'm not experiencing any build problems. As a side test, I tried building the Perl for 10.4 under 10.3.9 and it builds without issue but also experiences the same problem with setrgid and setruid. Going a step further, I substituted 5.8.6 with 5.8.7, tar'd it back up and this built without problems but also experiences the setrgid and setruid problem. One thing I did notice when I upped the perl to 5.8.7 was that it warned about two patches which seemed to already be applied but this was the only issue that was reported and no build errors were experienced. I made a minor change to the patch file and this resolved the warning about a previously applied patch. I then installed OSXS 10.4, upgraded to 10.4.3 and tried to build the perl software and it seems that it also suffers the same problem with setrgid and setruid. I've checked unistd.h in both the 7W98 and 8C47 and the entries exist for these functions so I'm not sure what's causing the problem. Since my general build issues have been resolved I'm not sure if this is the appropriate list or I should be posting to darwin-userlevel. At this point I'm just happy that I can build the software however the setrgid and setruid issue are causing me problems when the software is implemented. The only other question I have is this, it seems that when building under 10.4 the binaries are being built for ppc and i386, is there a way to reduce the build to ppc only ??? This email sent to site_archiver(a)lists.apple.com

1 0

Re: glibtool question
by Peter O'Gorman 28 Nov '05

28 Nov '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com User-agent: Mozilla Thunderbird 1.0.2 (Macintosh/20050317) -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 plumber Idraulico wrote: |> plumber Idraulico wrote: |> |> glibtool: compile: unable to infer tagged configuration |> |> glibtool: compile: specify a tag with `--tag' |> | when i do the same with --enable-shared=no --enable-static=yes |> | it works well and the lib archive is built correctly |> | |> | does anybody have a solution? ( not re-glibtoolize all the open- |> source |> | world :) ) |> |> Is this the glibtool in /usr/bin ? And the automake in /usr/bin? If not, |> which versions of the tools are you asking about? | | | all version : | the answer is : libtool it's not made for this | i would have like to simplify my work I just committed changes to gnu libtool to allow the -isysroot flag through to gcc unmolested. However I did do some work on older libtool-1.5.x versions to get it working with fat builds. All of the testing of these older builds was done using a fat toolchain and fat libraries in / though so they don't actually work with the tools shipping to developers now. sigh. If you put the flags into the CC or CXX compiler env var then it may work. CC="cc -arch ppc -arch i386 -isysroot=/path" CXX="c++ -arch ppc -arch i386 -isysroot=/path" CPP="cc -E" CXXCPP="g++ -E" Note that you need an automake newer than Apple ships. Latest is 1.9.6, I think 1.8.2 or so started adding the --tag argument for glibtool. | 1 - is there an importance that stored objects have been built with the | pic or non-pic option ? Are you talking about the pic option to gnu libtool? If so the pic option only adds -fno-common and since you are building a single_module library that flag is not strictly required. If you were building a multi_module library and the code had common symbols then the link would fail on the "non-pic" objects. If you try with real non-pic objects, then link will probably fail: imac:~/foo peter$ cat foo.c #include <stdio.h> int foo() {printf("Hello\n"); return 1;} imac:~/foo peter$ cc -mdynamic-no-pic -c -o foo_m.o foo.c imac:~/foo peter$ cc -fno-pic -c -o foo_nopic.o foo.c imac:~/foo peter$ cc -c -o foo_pic.o foo.c imac:~/foo peter$ cc -dynamiclib -o libfoo_m.dylib foo_m.o ld: foo_m.o has local relocation entries in non-writable section (__TEXT,__text) /usr/bin/libtool: internal link edit command failed imac:~/foo peter$ cc -dynamiclib -o libfoo_nopic.dylib foo_nopic.o ld: foo_nopic.o has external relocation entries in non-writable section (__TEXT,__text) for symbols: _printf$LDBL128 ld: foo_nopic.o has local relocation entries in non-writable section (__TEXT,__text) /usr/bin/libtool: internal link edit command failed imac:~/foo peter$ cc -dynamiclib -o libfoo_pic.dylib foo_pic.o imac:~/foo peter$ | | 2 - is it normal that the lib archive has a bigger size than the shared | archive ( +15% ) ? If you create a really small library with just one or two objects and almost no code then the static archive will be smaller :) Peter -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (Darwin) iQCVAwUBQ4sUb7iDAg3OZTLPAQIR2QP8CgeRHFfCFmlo2Bn11VmWaNs0kIDtyH/2 bsWeB55QhtzXzvvPGPss82GCa7WUYaR35hHlx07aX+WNvQeerlJ3ztDpepUDBFgD ChUrhBKHhB/j90OiB9WjPli4ivmdmf1T5ZpAprY5AOhfIcsKm0IFGVugSTyLB1ic Z/5lGPfkQJk= =dyXG -----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… This email sent to site_archiver(a)lists.apple.com

1 0

Re : Re: glibtool question
by plumber Idraulico 27 Nov '05

27 Nov '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com if i do something like this : gcc \ -dynamiclib \ -undefined suppress \ -flat_namespace \ -single_module \ -all_load ./libMy.a \ -install_name /usr/local/lib/libMy.1.dylib \ -compatibility_version 1 -current_version 1.3 -o libMy.1.0.dylib \ -lz because the "normal" way is gcc -dynamiclib *.pic.o \ -install_name /usr/local/lib/libMy.1.dylib \ -compatibility_version 1 -current_version 1.3 -o libMy.1.0.dylib \ -lz and ar rc libMy.a *.non-pic.o GCC 2.95.2 gcc -c -fPIC myfile.c gcc -shared -o libmyfile.so -fPIC myfile.o gcc3 gcc2 gcc4 ? and -fPIC this has an importance under Linux but for a mach-o archive ? this case can be found ? prelink: /usr/bin/mybin: Cannot prelink against non-PIC shared library /usr/lib/libMy.dylib gcc -dynamiclib *.pic.o \ -install_name /usr/local/lib/ libMyOther.1.dylib \ -compatibility_version 1 -current_version 1.3 -o libMyOther.1.0.dylib \ -lz ./libMy.a Best Regards, {Plum} _______________________________________________ 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 have an another question about pic or non pic objects stored inside a library archive 1 - is there an importance that stored objects have been built with the pic or non-pic option ? 2 - is it normal that the lib archive has a bigger size than the shared archive ( +15% ) ? When building a shared library, gcc will compile additional code to be included in the library. That additional code must also be compiled with the proper PIC option. http://developer.apple.com/releasenotes/DeveloperTools/GCC4.html 3 - is there an importance that stored objects have been built with the pic option in a library archive ? This email sent to site_archiver(a)lists.apple.com

1 0

Re: glibtool question
by plumber Idraulico 27 Nov '05

27 Nov '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com all version : the answer is : libtool it's not made for this i would have like to simplify my work yes but gnu-darwin project source this is not the purpose thanks if i do something like this : gcc \ -dynamiclib \ -undefined suppress \ -flat_namespace \ -single_module \ -all_load ./libMy.a \ -install_name /usr/local/lib/libMy.1.dylib \ -compatibility_version 1 -current_version 1.3 -o libMy.1.0.dylib \ -lz because the "normal" way is gcc -dynamiclib *.pic.o \ -install_name /usr/local/lib/libMy.1.dylib \ -compatibility_version 1 -current_version 1.3 -o libMy.1.0.dylib \ -lz and ar rc libMy.a *.non-pic.o Best Regards, {Plum} _______________________________________________ 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… plumber Idraulico wrote: |> glibtool: compile: unable to infer tagged configuration |> glibtool: compile: specify a tag with `--tag' | when i do the same with --enable-shared=no --enable-static=yes | it works well and the lib archive is built correctly | | does anybody have a solution? ( not re-glibtoolize all the open- source | world :) ) Is this the glibtool in /usr/bin ? And the automake in /usr/bin? If not, which versions of the tools are you asking about? It looks as though you are trying to build a darwin project? If that is the case then I'd suggest darwinbuild. Peter darwinbuild i have an another question about pic or non pic objects stored inside a library archive 1 - is there an importance that stored objects have been built with the pic or non-pic option ? 2 - is it normal that the lib archive has a bigger size than the shared archive ( +15% ) ? This email sent to site_archiver(a)lists.apple.com

1 0

Re: shm_open + mmap?
by Matt Watson 25 Nov '05

25 Nov '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On Nov 24, 2005, at 6:35 PM, Brad Parker wrote: You must first increase the length of the file by one means or another. shm_open creates a 0-length file. Are you calling ftruncate() with the size you are passing in to mmap()? This seems to work: #include <stdio.h> #include <sys/file.h> #include <sys/mman.h> static const char *shm_filename = "/tmp/flarm"; #define MAX_BITMAP_OFFSET 8192 int main(void) { int ret, shm_fd, mprot, mflags, ps; size_t size; void *shm_fb; ret = shm_unlink(shm_filename); printf("shm_unlink ret %d\n", ret); shm_fd = shm_open(shm_filename, O_RDWR | O_CREAT, 0666); if (shm_fd < 0) { perror("shm_open"); return 1; } size = MAX_BITMAP_OFFSET*4; mprot = PROT_READ|PROT_WRITE; mflags = MAP_FILE|MAP_SHARED; ps = getpagesize(); printf("pagesize %d\n", ps); size = ((size+ps-1)/ps) * ps; if (ftruncate(shm_fd, size) < 0) { perror("ftruncate"); return 1; } printf("shm_fd %d, size %d\n", shm_fd, size); shm_fb = mmap(NULL, size, mprot, mflags, shm_fd, (off_t)0); if (shm_fb == (void *)-1) { fprintf(stderr, "mmap of %s failed\n", shm_filename); perror("mmap"); return 1; } printf("shm_fb = %p\n", shm_fb); return 0; } _______________________________________________ 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… Solra Bizna wrote: I tried that using ftruncate and it's didn't help. This email sent to site_archiver(a)lists.apple.com

1 0

Re: shm_open + mmap?
by Brad Parker 25 Nov '05

25 Nov '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Organization: Heeltoe Consulting Solra Bizna wrote: >You must first increase the length of the file by one means or >another. shm_open creates a 0-length file. I tried that using ftruncate and it's didn't help. -brad _______________________________________________ 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: shm_open + mmap?
by Solra Bizna 25 Nov '05

25 Nov '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:sender:to:subject:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; b=oaivKdDUm+D92IsfTBX/ZQA3IMihOyUdCESLaByt4TPfAmWi5RVZFXKLv1Muw4Mio3ukhwy4taafKDfabh75+e0ODx0zuXfhq3aUtMXAHIeHuucRy1yNm2/R/QGJO3yItNnjGQBTrHmIVGZL7Os4C6IYHE92ytSYT4Xr0Kbb6+A= You must first increase the length of the file by one means or another. shm_open creates a 0-length file. -:sigma.SB _______________________________________________ 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

shm_open + mmap?
by Brad Parker 25 Nov '05

25 Nov '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Organization: Heeltoe Consulting Can anyone explain why this doesn't work on darwin? I want to create shared memory space which two processes can access. The shm_open works but the mmap fails. I can't figure out why (but have not looked at the sources yet) I've tried all manner of mprot and mflags. I did find the mmap would work if I used "open" instead of "shm_open" *and* the file size was >= the mmap size. The code below always fails at the mmap. I thought shm_open + mmap would work on os 10.4... int shared_fb_init(void) { int ret, shm_fd, mprot, mflags, ps; size_t size; ret = shm_unlink(shm_filename); printf("shm_unlink ret %d\n", ret); shm_fd = shm_open(shm_filename, O_RDWR | O_CREAT, 0666); if (shm_fd < 0) { perror(shm_filename); return -1; } size = MAX_BITMAP_OFFSET*4; mprot = PROT_READ|PROT_WRITE; mflags = MAP_FILE|MAP_SHARED; ps = getpagesize(); printf("pagesize %d\n", ps); size = ((size+ps-1)/ps) * ps; printf("shm_fd %d, size %d\n", shm_fd, size); shm_fb = mmap(NULL, size, mprot, mflags, shm_fd, (off_t)0); if (shm_fb == (void *)-1) { fprintf(stderr, "mmap of %s failed\n", shm_filename); perror(shm_filename); return -1; } ... _______________________________________________ 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: 10.4, MH_BUNDLE, and DYLD_IMAGE_ADDED dyld_events
by Nathan Herring 24 Nov '05

24 Nov '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Thread-index: AcXwxnM2IbldLdN2SIWx8k1pb/3U9gAR8Zjw Thread-topic: 10.4, MH_BUNDLE, and DYLD_IMAGE_ADDED dyld_events Hmm. There's a comment in the top of http://darwinsource.opendarwin.org/10.4.3/dyld-44.2/src/dyld_debug.c: // The following API's are deprecated. // In Mac OS X 10.4 we only do a minimal implementation of these API's // to keep from breaking existing clients (Omni and MS Crash Reporters). // // This minmal implementation only allows inspection of one process and // only reveals the current images in that process (no notification of // later image changes). It assumes both processes use the same dyld // and dyld has not slid in either process. Well, it cannot assume that dyld has not slid, because it always slides bundles. Unfortunately for us, in Office 2004, we're largely CFM, and our only Mach-O code is in, yes, you guessed it, a MH_BUNDLE. Our crash reporter generates these broken logs in 10.4, and if the crash was in our Mach-O code, we have no idea where it was, and they all get bucketed identically in the "unknown" bucket. Thanks for the information. It appears that you (Omni) and I (MS) are the consumers of this deprecated API. -nh -----Original Message----- From: Timothy Wood [mailto:khavi@mac.com] Sent: Wednesday, November 23, 2005 11:08 PM To: Nathan Herring Cc: darwin-dev(a)lists.apple.com Subject: Re: 10.4, MH_BUNDLE, and DYLD_IMAGE_ADDED dyld_events On Nov 23, 2005, at 9:14 PM, Nathan Herring wrote: > For MH_DYLD libraries, vmaddr is set to the base address and > vmaddr_slide is set to 0. For MH_BUNDLE libraries (under 10.3 and > earlier), vmaddr is 0, and vmaddr_slide is set to the base address. > However, in 10.4, vmaddr_slide is set to 0, and we don't have any idea > where the bundle is living in memory. > > We're acquiring this information by calling > _dyld_debug_subscribe_to_events and handling DYLD_IMAGE_ADDED events, > looking in arg[0].vmaddr_slide. This does not seem to be working any > longer. You might be running into Radar 4149320: dyld_debug API shim computes incorrect slide. Note that in 10.4 dyld was massively rewritten. As part of this, the dyld_debug API was deprecated and the shim was added. The way things work now is that dyld has some global state (search for "dyld_all_image_infos" in the dyld source) for listing the loaded modules and notifies listeners by calling a empty function ("notification" element of the aforementioned global). The listeners are expected to have set a breakpoint on this function; it isn't clear to me how to get multiple listeners to work since I've never gotten forwarding of Mach exceptions working right. If you are running in the same process, you can probably just access the globals, though we take the approach of not trusting the process at this point (since, hey, it crashed) and we fork another process to collect the crash. This has its own set of compatibility issues (see the mysterious setgid flag on the real gdb executable in Xcode 2.2 and higher...) At any rate, here is the text of my Radar: ----- In dyld-43, dyld_debug.c, the function slideForHeader is busted. In particular, this breaks for any file that did get slid (MH_BUNDLE or MH_DYLIB most frequently). The two errors that I've spotted so far are: 1) const struct segment_command *sgp = (const struct segment_command *)((char*)mh + sizeof(mh)); This should have "sizeof(*mh)". Otherwise "sgp" starts out 4 bytes into the struct mach_header instead of after it. Obviously this causes badness, but it is hidden in my cases since the cmdsize computed from this bogus pointer is zero (so the 'for' loop doesn't wander off and crash noisily). 2) return (uintptr_t)mh - (uintptr_t)sgp->vmaddr; This is using "mh", a local pointer obtained from malloc (i.e., meaningless garbage), instead of "otherAddressHeader", the target task's pointer. In my 10.4-ification of our OmniMachO framework (used in OmniObjectMeter and OmniCrashCatcher), I've copied over something like this function and fixed it up and I'm starting to get correct slide values. ----- This Radar is still marked 'Open', so apparently it didn't make it into 10.4.3. Note that even though I was able to produce a version of the shim function that worked, we don't use this approach any more when running on 10.4 (we look for the dyld_all_image_infos symbol in / usr/lib/dyld and if so, we use the global table). I hope this helps! -- Timothy J. Wood The Omni Group _______________________________________________ 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: glibtool question
by Peter O'Gorman 24 Nov '05

24 Nov '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com User-agent: Mozilla Thunderbird 1.0.2 (Macintosh/20050317) -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 plumber Idraulico wrote: |> glibtool: compile: unable to infer tagged configuration |> glibtool: compile: specify a tag with `--tag' | when i do the same with --enable-shared=no --enable-static=yes | it works well and the lib archive is built correctly | | does anybody have a solution? ( not re-glibtoolize all the open- source | world :) ) Is this the glibtool in /usr/bin ? And the automake in /usr/bin? If not, which versions of the tools are you asking about? It looks as though you are trying to build a darwin project? If that is the case then I'd suggest darwinbuild. Peter -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (Darwin) iQCVAwUBQ4W21riDAg3OZTLPAQLxTwQArIS5STb+9DYZ5ABtDy4LH/FPZBCIhS2Y fu7F7U1qcESRxwvbBqk7uWio6v8cEdmesq2NBA+fxOfn6J/h3MMzYFW/+IC2tTa3 hVnHn75zr696YOdGdC1noYigdeLSTiC5ImOxL5NU6Y4Lc+OED72ZLTN52O7+QpNR VLkyAAk4PZo= =TKPU -----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… This email sent to site_archiver(a)lists.apple.com

1 0