July 2007 - Darwin-dev - lists.apple.com

Re: dup2() Problem
by Terry Lambert 07 Jul '07

07 Jul '07

site_archiver(a)lists.apple.com Delivered-To: Darwin-dev(a)lists.apple.com "testfile" (relative to the current directory). and flags: O_CREAT | O_TRUNC | O_WRONLY and mode: 0666 #include <fcntl.h> /* open */ #include <unistd.h> /* dup2 */ #include <stdio.h> /* perror, fprintf */ #include <stdlib.h> /* exit */ int main(int ac, char *av[]) { int fd; if (dup2(fd, 1) < 0) { perror("FAIL: dup2"); exit(2); } fprintf(stderr, "PASS\n"); exit(0); } ...I run it, and it prints "PASS" (try this code yourself). if (new < 0 || new >= p->p_rlimit[RLIMIT_NOFILE].rlim_cur || new >= maxfiles) { fp_drop(p, old, fp, 1); proc_fdunlock(p); return (EBADF); } if ((ofp = fdp->fd_ofiles[old]) == NULL || (fdp->fd_ofileflags[old] & UF_RESERVED)) { _fdrelse(fdp, new); return (EBADF); } -- Thanks, -- Terry _______________________________________________ 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 Jul 6, 2007, at 3:39 PM, Norm Green wrote: Thanks Mark. Yes the code below is short snipet. I ran the program under ktrace and see that the system() call causes a call to dup2(), which is failing with errno EBADF: 5738 sh CALL open(0x303f50,0x601,0x1b6) 5738 sh NAMI "testfile" 5738 sh RET open 5 5738 sh CALL dup2(0x5,0x1) 5738 sh RET dup2 -1 errno 9 Bad file descriptor 5738 sh CALL write(0x2,0x1800400,0x28) 5738 sh GIO fd 2 wrote 40 bytes "/bin/sh: line 1: 1: Bad file descriptor " 5738 sh RET write 40/0x28 5738 sh CALL exit(0x1) This seems to indicate that all is well with the new file I opened since it returned fd 5. So does that mean something's wrong with stdout?have put traces in to check the value of fileno(stdout) and it's always 1. This isn't code, it's a system call trace, so we can't compile it up and see it fail for ourselves. Nevertheless, we can gain the insight that you are using file: So we know that you haven't dropped yourself into an owner exclusion or exclusion group situation, and you haven't picked a mode incompatible with being able to dup2 it (i.e. a dup2 is morally equivalent to an open; so for example, if you had specified an open falg of O_EXCL, you would be attempting to open the file in violation of the exclusive use warranty made by the system to the other decriptor). So I write a small regression test that results in an identical ktrace (except it doesn't fail, of course): if ((fd = open("testfile", O_CREAT | O_TRUNC | O_WRONLY, 0666)) == -1) { perror("FAIL: open"); exit(1); } So we look in the publically published Tiger sources (which you can also look at, if you are motivated) in the dup2 implementation in kern_descrip.c: Translation: If I exceed the maximum number of open files according to my resource limit, return EBADF. You might be doing this, but you didn't priint the output of getrlimit() for RLIMIT_NOFILE and a list of the other fd's you have open, so we haven't given us enough information to know. flags = fdp->fd_ofileflags[new]; if ((flags & (UF_RESERVED | UF_CLOSING)) == UF_RESERVED) { fp_drop(p, old, fp, 1); proc_fdunlock(p); return (EBADF); } Translation: If the fd 1 is already in the process of being opened, but has hung in this process (e.g. an NFS server went away out from under us, etc.), then return EBADF. You might be doing this, but since you didn't fstat fd 1 and report the output and we don't have enough information to know. Translation: If the thing I'm dup'ing is not open, or is in the process of being opened, but has hung (e.g. it's a socket that has not yet finished connecting, etc.), then return EBADF. You might be doing this, but since you didn't give us information about your filesystem topogy (which MUST be odd: we don't ship by default with a "/usr/tmp" directory, yet your open in the previous posting succeeded), and you didn't fstat fd 5 immediately following the open, we don't have enough information to know. So without source code that I too can make fail, and without answers to the other questions, I'd have to say this is "pilot error" on your part. I am one of the primary maintainers of the particular kernel code in question, and could resolve this question in an instant, if you had provided a failing C source code example, and tell you the specific pilot error involved. If it wasn't pilot error, and was in fact a MacOS X specific problem, I could actually fix it on our end. But only IF there was a bug report to which I could attach the code changes in order to demonstrate that I wasn't just adding risk of breaking someone depending on the current behaviour, and there was actually a customer impact involved in not fixing the code. As it is, you will need to answer the specific requests for information above for an accurate diagnosis. This email sent to site_archiver(a)lists.apple.com

1 0

RE: dup2() Problem
by Norm Green 06 Jul '07

06 Jul '07

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: Problem with system()
by Norm Green 06 Jul '07

06 Jul '07

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: Code Obfuscation
by Thomas Inskip 06 Jul '07

06 Jul '07

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Dkim-signature: a=rsa-sha1; c=relaxed/relaxed; d=gmail.com; s=beta; h=domainkey-signature:received:received:in-reply-to:references:mime-version:content-type:message-id:cc:content-transfer-encoding:from:subject:date:to:x-mailer; b=TL9D61rtxY9sSFf3EwRmVmIxs2KnKPShmOCenA6icvVy5QvxlmR89UIFRCKU5gVi8u8fXJINuCOVG3PYiJSHTzFw5CSMSFpvjZamwGqTQe4EPmrefulk7wXTbYCT+xyzZB/VgofxrhP+Q9OgeWX1f2J3GgmxzMCP0sGLYaglcQA= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=beta; h=received:in-reply-to:references:mime-version:content-type:message-id:cc:content-transfer-encoding:from:subject:date:to:x-mailer; b=bnjiJfYLe24jGvYVJblqVswCDXcnNACngvMkY/VBJsuQGR4oEvSxL7e/eupBkeQzcjcQd8kVuwH5XJDAcjYT2ojBA8OPEZPHlcOVaTg8ZQpCUZ+97UJiNhLZLEZBlP5H2ot+5qrPbGUNXXH+KZQEeJbivMcmdh3IlBlLEBee2/o= Thank you. On Jul 6, 2007, at 2:32 PM, Michael Smith wrote: This may be slightly off topic, but does anyone know of any object code obfuscators which will work with mach-o binaries (PPC and x86)? I've looked around but haven't had much luck. The only product even remotely related I've been able to find is a C/C++ source code obfuscator from stunnix (http://stunnix.com/prod/cxxo/ overview.shtml), which I assume won't much good in terms obfuscating the generated object code. = Mike _______________________________________________ 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… Very helpful. It looks like it will fit the bill perfectly. I know none of these solutions is perfect, but the purpose here is more to discourage reverse engineering than to make it impossible. On Jul 3, 2007, at 12:04 PM, darwin-dev-request(a)lists.apple.com wrote: Not quite what you were asking about, but the Cloakware Transcoder product will probably get you the results you're looking for. This email sent to site_archiver(a)lists.apple.com

1 0

Re: Problem with system()
by Terry Lambert 06 Jul '07

06 Jul '07

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com int rc = system("date >/usr/tmp/tmp.kPCvVa"); If fails with rc == 256 and errno == 0 and prints this to stderr: sh: line 1: 1: Bad file descriptor the shell handles the redirection and creates the output file so its not really an indicator that your path is ok. its a bad idea to just call 'date' and not give the full path since some user program called date (and someone somwhere has probably done this) would interfere with proper behavior. The system() library routine operates by doing a: execl( "/bin/sh", "-c", "date >/usr/tmp/tmp.kPCvVa", NULL); _exit(127); /* not reached normally */ -- Terry _______________________________________________ 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 Jul 6, 2007, at 12:08 PM, chas williams - CONTRACTOR wrote: In message <75D481373A76D311A5790090274F2B5C09289B39(a)mailsrv1.gemstone.com >,Nor m Green writes: I know the path is OK because the file /usr/tmp/tmp.kPCvVa *IS* created, but it has a size of zero. For this particular case, you are most likely correct, and it's the raw "date" that's not being correctly interpreted in his code. Since he is getting back what he is getting back (256 - btw, the correct way to decode this is using the macros used by waitpid() - see the man page), then it's most likely a bad fd. I expect that either "date" is not in the default shell path, or he is invoking this as root (in which case, security dictates that his path be cleaned out on him by the shell program), or his default path is bogus, or he has built and replaced /bin/sh with his own code, or linked it to another sell program, other than the MacOS X default, which is a specifically modified "bash". Probably setting the "date" to "/bin/date" instead would fix this: The bad fd the shell is complaining about is likely the fd in the fread() of the popen() of the date command, built into the shell, and the difference is whether or not stack pages contain "bad garbage" or "good garbage" in the shell itself (as Justin Walker pointed out might be the case). Note that if he is running this program as root, and there are local accounts, it would be rather trivial to get root on the machine as an ordinary unprivileged user account using this program as is... This email sent to site_archiver(a)lists.apple.com

1 0

Getting hibernation system (MacBook 10.4.6( - IOHibernatePoller?) to read from *different* set of files - can that even work?
by Jim Witte 06 Jul '07

06 Jul '07

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Hi, Jim _______________________________________________ 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 *completely* new to Darwin programming, and this idea probably won't work at all: a couple of times, the battery ran down and the book went into (due to some bug) a "permanent sleep" mode - indicator light was on, USB ad FW power lines were on but nothing would wake the computer. So, not knowing about the hibernate feature (I should have just pulled the battery), I did a hard reset (power button) instead. But I made copies of both the sleepimage and swapfile0 files, but not the app_profiles folder (except in one case). Is there any way to rewrite the part of Darwin that specifies the path to the image files, so I might be able to restore the computer to a previous state (or change the path before putting the computer to sleep)? Or at least restore it enough to pull off some web addresses from minimized windows in Camino..), or is there other information (processor registers, etc) that are stored somewhere else that are needed) As I said, probably impossible.. This email sent to site_archiver(a)lists.apple.com

1 0

Re: sysctlbyname( "hw.availcpu" ) fails
by Terry Lambert 06 Jul '07

06 Jul '07

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On Jun 29, 2007, at 1:33 AM, Steve Checkoway wrote: Note that I'm not the original poster, but yes there is. :) -- Terry _______________________________________________ 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 Jul 6, 2007, at 4:35 AM, Daniel Vollmer wrote: On 29.06.2007, at 19:27, Eric Schlegel wrote: That's more time than I'd care to take on start up. I appreciate the suggestions though. I've been using Gestalt and figured I should maybe move to some newer interface. Is Gestalt failing to give you the info you need in some way? MenuMeters (which displays your CPU type as well as the number of processors — so only cosmetic in a sense) for example is terribly incorrect on Intel Macs. NXGetLocalArchInfo() always returns "Intel 80486" and Gestalt gives "i586" (Pentium) for every CPU type. I've fixed a version of MenuMeters for my personal use to use "system_profiler" for now, as I have not found a programmatic way to obtain the information (number of physical processors, not core count (as "2 x Intel Core 2 Duo" is incorrect!) as well as the cpuid string). Please file a bug against NXGetLocalArchInfo() and another one against Gestalt. In general, we class our CPUs by their capability bits, not their CPUID information, and anyone not using capability bits is going to be unhappy. This email sent to site_archiver(a)lists.apple.com

1 0

Re: dup2() Problem
by Mark Rhoads 06 Jul '07

06 Jul '07

site_archiver(a)lists.apple.com Delivered-To: Darwin-dev(a)lists.apple.com Organization: WiredRoads User-agent: Thunderbird 2.0.0.0 (Macintosh/20070326) Hello, Hi, After the fork I do: int logfd = open(logName, O_WRONLY|O_APPEND|O_CREAT, 0644); dup2(logfd, STDOUT_FILENO); I presume that the above code snippet is in the child process, e.g.: if ( 0 == fork() ) { ... } Given the EBADF errno, I'd guess that either logfd or STDOUT_FILENO does not contain or refer to an expected value for an fd. As well as following other advice given in other posts, you might check whether some bit of custom kit in your 'old as dirt' and 'works fine elsewhere' code is (re)defining STDOUT_FILENO, or even 'dup2' or 'fork' to something unexpected. Also, check that some bug or side-effect is not changing the value of logName or what it references after your 'fork', and thus causing open() to intermittently fail (even though you might have successfully tested it before). If your code snippet happens to be in the parent process space instead, then it is also possible that the child process of a vfork() could be mucking things up for you. This bit of advice is worth repeating from other posts: Build and run a small test case. Best Regards --Mark Rhoads _______________________________________________ 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… Norm Green wrote: I'm having a problem on Darwin i386 using dup2() in a child process. This is after a fork() call but before execve() call. The open() call succeeds but the dup2() call always fails with errno=9, which suggests one of the args is not an open fd. I'm sure I'm not closing either in my code. This email sent to site_archiver(a)lists.apple.com

1 0

Re: Problem with system()
by chas williams - CONTRACTOR 06 Jul '07

06 Jul '07

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com In message <75D481373A76D311A5790090274F2B5C09289B39(a)mailsrv1.gemstone.com>,Nor m Green writes: > int rc = system("date >/usr/tmp/tmp.kPCvVa"); > >If fails with rc == 256 and errno == 0 and prints this to stderr: > >sh: line 1: 1: Bad file descriptor > >I know the path is OK because the file /usr/tmp/tmp.kPCvVa *IS* created, but >it has a size of zero. the shell handles the redirection and creates the output file so its not really an indicator that your path is ok. its a bad idea to just call 'date' and not give the full path since some user program called date (and someone somwhere has probably done this) would interfere with proper behavior. _______________________________________________ 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: Problem with system()
by Justin C. Walker 06 Jul '07

06 Jul '07

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On Jul 6, 2007, at 11:44 , Norm Green wrote: See my comment about good vs. bad garbage. The probability is high that it's your code, not Mac OS X, that is at fault. Justin -- Justin C. Walker, Curmudgeon-At-Large, Director Institute for the Enhancement of the Director's Income -------- The path of least resistance: it's not just for electricity any more. -------- _______________________________________________ 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 are not having a good time with file descriptors, are you? Right, but only on Darwin ;-) The code in question runs fine on Solaris, HPUX, AIX and Linux. That may be, but how can I determine what's wrong? I think Terry addressed this yesterday. You flatter us: we really are not able to divine the content of your code, and unwind the myriad things that might be going wrong. Give us something to go on, and someone here may be able to help. This email sent to site_archiver(a)lists.apple.com

1 0