November 2003 - Darwin-kernel

Operating Systems Research Engineers
by George Lindner 10 Nov '03

10 Nov '03

Operating Systems Researchers, The Most Prestigious Communications Lab in Silicon Valley, CA currently has tremendous employment opportunities available. They are currently looking for individuals to join their Operating Systems Group. Please review the description and if you are interested in exploring these opportunities further please forward me a copy of your resume in Microsoft Word format and provide me with a convenient time to follow up with you or if you may know of any individuals who may be in the market for a similar opportunity please feel free to forward them my contact information. My firm does offer referral bonuses for any assistance in locating qualified individuals for our client's needs that are hired. I would much rather work with an individual with industry experience and knowledge then rely on a job posting and I thought that with you may have industry connections or associates that may be exploring opportunities at this time. The opportunities below are immediate needs so your timely response is necessary and would be greatly appreciated. I would like to thank you in advance for any assistance you may be able to provide me with and I look forward to speaking with you or any referral you may have soon. 1. Operating system Researcher (Scheduling and Process Management) The Communications Laboratory is searching for a researcher with expertise in operating system process management and scheduling. The qualified individual should have a PhD in computer science or engineering, should have at least 3 years experience as a researcher in industry. Experience with aspects of mobile computing and concepts such as operating system extensibility an added plus. The individual will part of a larger operating systems team and will lead research in the area of scheduling, process management, and network communications. Experience with Linux an added plus. 2. Operating System Researcher (Kernel Design and Implementation) The Communications Laboratory is searching for a researcher with expertise in operating system kernels. The qualified individual should have a PhD in computer science or engineering, should have at least 3 years experience as a researcher in industry. Experience with aspects of mobile computing and concepts such as operating system extensibility an added plus. The individual will part of a larger operating systems team and will lead research in the area of kernel design. Experience with Linux an added plus. 3. Operating System Researcher (File Systems and Storage Management) The Communications Laboratory is searching for a researcher with expertise in operating system with particular focus on file systems and storage management. The qualified individual should have a PhD in computer science or engineering, should have at least 3 years experience as a researcher in industry. Experience with aspects of mobile computing and concepts such as operating system extensibility an added plus. The individual will part of a larger operating systems team and will lead research in the area of file systems and storage management. Experience with Linux an added plus. George Lindner Valoure Staffing Specialists Sr. Recruitment Coordinator Direct: 818.784.8302 Main: 818.784.8635 Fax: 818.784.9011 glindner(a)valoure.com <http://www.valoure.com/> www.valoure.com ************************************************************************ Please Note The information in this E-mail message is legally privileged and confidential information intended only for the use of the individual(s) named above. If you, the reader of this message, are not the intended recipient, you are hereby notified that you should not further disseminate, distribute, or forward this E-mail message. If you have received this E-mail in error, please notify the sender. Thank you ************************************************************************ _______________________________________________ darwin-kernel mailing list | darwin-kernel(a)lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel Do not post admin requests to the list. They will be ignored.

1 0

Re: panicd & DB_KERN_DUMP_ON_PANIC
by Quinn 10 Nov '03

10 Nov '03

At 12:46 -0500 29/10/03, Andrew Gallatin wrote: When reading the Panther sources, I noticed that Panther has finally grown the ability to do a crashdump (hurray!!). However, there appears to be a daemon (panicd?) that needs to be run in order to collect the crashdump on a remote node. I can't seem to find this. A search on the ADC site turns up nothing. Can somebody point me at it, please? AFAIK this isn't public yet. I'm working with the engineer responsible for this and we're planning on producing a package for developers Real Soon Now. When that happens depends on our respective schedules. Sorry about the delay. S+E -- Quinn "The Eskimo!" <http://www.apple.com/developer/> Apple Developer Technical Support * Networking, Communications, Hardware _______________________________________________ darwin-kernel mailing list | darwin-kernel(a)lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel Do not post admin requests to the list. They will be ignored.

1 0

Re: String Question
by Quinn 10 Nov '03

10 Nov '03

At 1:33 -0500 10/11/03, Steve Milner wrote: The other night I was perusing through some of the kernel source when I noticed strcat and strcpy is used quite a bit in kernel. I also notices string.h didn't have an definition for strl*. Why are strl* functions not in the kernel but are in the provided C library? The kernel does not implement the standard C library. Rather, specific standard C library routines are implemented for use in kernel. This duplication tends to occur on a case-by-case basis, as routines are needed. strlcpy/cat are relatively new routines, and so this hasn't happened for them yet. Apple is in the process of nailing down the binary compatibility story for the BSD parts of the kernel (the Kernel Programming Interfaces effort we introduced at WWDC 2003). Feedback from developers about missing routines is an important part of that effort. I'm sure you could make an argument that access to these routines is even more important for kernel code than for user space code. You should feel free to file a bug to request them to be included in the KPI. <http://developer.apple.com/bugreporter/> S+E -- Quinn "The Eskimo!" <http://www.apple.com/developer/> Apple Developer Technical Support * Networking, Communications, Hardware _______________________________________________ darwin-kernel mailing list | darwin-kernel(a)lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel Do not post admin requests to the list. They will be ignored.

1 0

String Question
by Steve Milner 10 Nov '03

10 Nov '03

Good day all, The other night I was perusing through some of the kernel source when I noticed strcat and strcpy is used quite a bit in kernel. I also notices string.h didn't have an definition for strl*. Why are strl* functions not in the kernel but are in the provided C library? Cheers, Steve _______________________________________________ darwin-kernel mailing list | darwin-kernel(a)lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel Do not post admin requests to the list. They will be ignored.

1 0

Re: HELP! vn_open and kernel panic in Panther
by Justin Walker 09 Nov '03

09 Nov '03

On Nov 9, 2003, at 12:22, Hao Li wrote: Hi all! I've a little app with its own kernel extension that logs some data back to the app in the user space. The app creates a temp file, and the kext open it and write data back. I'm not sure if it's the best way to do that, but it works well in Jaguar. Your instincts are correct. It's not the best way to do this, and you've found one of them. Using in-the-kernel interfaces to do "user mode" operations is a bad idea in general. But now I am fixing a version for Panther, it crashes at vn_open: user space app does: f = open(path, O_CREAT|O_RDWR|O_FSYNC, S_IRUSR|S_IWUSR); if ( f == -1 ) return nil; close(f); the kext is then called with the "path" to do the following: NDINIT(&nd, LOOKUP, FOLLOW, UIO_SYSSPACE, path, p); if ( (error = vn_open(&nd, FWRITE | O_FSYNC, 0)) != 0 ) ... And it paniced at vn_open. Anyone knows why? Since you have gone to the trouble of building a kext to communicate with your app, have the app pull the data from the kext, rather than have the kext write to the file; file I/O is the job of user-space apps. Regards, Justin -- Justin C. Walker, Curmudgeon-At-Large * Institute for General Semantics | Some people have a mental | horizon of radius zero, and | call it their point of view. | -- David Hilbert *--------------------------------------*-------------------------------* _______________________________________________ darwin-kernel mailing list | darwin-kernel(a)lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel Do not post admin requests to the list. They will be ignored.

1 0

HELP! vn_open and kernel panic in Panther
by Hao Li 09 Nov '03

09 Nov '03

Hi all! I've a little app with its own kernel extension that logs some data back to the app in the user space. The app creates a temp file, and the kext open it and write data back. I'm not sure if it's the best way to do that, but it works well in Jaguar. But now I am fixing a version for Panther, it crashes at vn_open: user space app does: f = open(path, O_CREAT|O_RDWR|O_FSYNC, S_IRUSR|S_IWUSR); if ( f == -1 ) return nil; close(f); the kext is then called with the "path" to do the following: NDINIT(&nd, LOOKUP, FOLLOW, UIO_SYSSPACE, path, p); if ( (error = vn_open(&nd, FWRITE | O_FSYNC, 0)) != 0 ) ... And it paniced at vn_open. Anyone knows why? Best Regards Hao Li _________________________________________________________________ Help STOP SPAM with the new MSN 8 and get 2 months FREE* http://join.msn.com/?page=features/junkmail _______________________________________________ darwin-kernel mailing list | darwin-kernel(a)lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel Do not post admin requests to the list. They will be ignored.

1 0

Re: EXC_BAD_ACCESS instead of SIGSEGV?
by Jim Magee 06 Nov '03

06 Nov '03

On Nov 6, 2003, at 7:59 AM, Simon Gray wrote: I'm working with the GNU libsigsegv library (http://sourceforge.net/projects/libsigsegv/) as I'm doing some investigative work on page-mapping in Mac OSX 10.3 and I've run into a problem: The basic configure script tests signal handling via the code below but no longer works in 10.3 as a GDB trace shows that an EXC_BAD_ACCESS is generated instead of SIGSEGV in the function "crasher()", and it doesn't seem possible to install a signal handler for EXC_BAD_ACCESS (unsurprisingly, as its not a BSD signal). Inexperienced as I am with Mach, I'm really not sure the best approach to resolving this issue - any pointers? All faults like these are delivered as Mach exceptions first. BSD has installed a Mach exception handler that converts the Mach exception into a signal. GDB installs its Mach exception handler BEFORE BSD's in the chain of exception handlers to try for a watched task/process. So, GDB sees and reports on Mach exceptions - not signals. That's why EXC_BAD_ACCESS is confusing to you. If you want to provide a signal handler for these, you need to install handlers for SIGSEGV and SIGBUS. SIGSEGV applies when address is completely invalid (nothing mapped there). SIGBUS applies for failures when something is mapped there - but you can't access it (permissions, force unmounts of a mapped file, etc...). --Jim PS: Because of where GDB sits in the exception path, you will still see its handler kick in BEFORE the process' signal handler. You need to continue past that spot in GDB to see if the process will handle it itself. This also sometimes results in a "premature crashreport" as the crash reporter mechanism also sits in front of the "last resort" BSD signal delivery exception handler. We plan to address this in the future by allowing exception handlers between BSD's signal delivery handler and Mach's default (kill the task) behavior and moving crashreporter's exception handler to sit there. We're still working with the GDB folks to decide where they want their handler to sit (both places have some value in a gdb environment). _______________________________________________ darwin-kernel mailing list | darwin-kernel(a)lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel Do not post admin requests to the list. They will be ignored.

1 0

EXC_BAD_ACCESS instead of SIGSEGV?
by Simon Gray 06 Nov '03

06 Nov '03

Good morning all, I wonder whether someone could point me in the right direction for this problem. I'm working with the GNU libsigsegv library (http://sourceforge.net/projects/libsigsegv/) as I'm doing some investigative work on page-mapping in Mac OSX 10.3 and I've run into a problem: The basic configure script tests signal handling via the code below but no longer works in 10.3 as a GDB trace shows that an EXC_BAD_ACCESS is generated instead of SIGSEGV in the function "crasher()", and it doesn't seem possible to install a signal handler for EXC_BAD_ACCESS (unsurprisingly, as its not a BSD signal). Inexperienced as I am with Mach, I'm really not sure the best approach to resolving this issue - any pointers? best regards Simon Gray PS I would have attached the code below but it goes over the 8K limit :( PPS i can just fit the primary code in below... file: conftest.c ------------------------------------------------------------------------ ------------------------------------------------------------------------ ----- #define PACKAGE_NAME "" #define PACKAGE_TARNAME "" #define PACKAGE_VERSION "" #define PACKAGE_STRING "" #define PACKAGE_BUGREPORT "" #define PACKAGE "libsigsegv" #define VERSION "2.1" #define STDC_HEADERS 1 #define HAVE_SYS_TYPES_H 1 #define HAVE_SYS_STAT_H 1 #define HAVE_STDLIB_H 1 #define HAVE_STRING_H 1 #define HAVE_MEMORY_H 1 #define HAVE_STRINGS_H 1 #define HAVE_INTTYPES_H 1 #define HAVE_STDINT_H 1 #define HAVE_UNISTD_H 1 #define HAVE_DLFCN_H 1 #define HAVE_SYS_SIGNAL_H 1 #define CFG_SIGNALS "signals-macos.h" #define HAVE_UNISTD_H 1 #define HAVE_GETPAGESIZE 1 #define HAVE_SYSCONF_PAGESIZE 1 #define HAVE_MMAP_ANON 1 /* end confdefs.h. */ #include <stdlib.h> #include <signal.h> #if HAVE_SYS_SIGNAL_H # include <sys/signal.h> #endif #include "./src/fault-macos-powerpc.c" #include <sys/types.h> #include <sys/mman.h> #if HAVE_MMAP_DEVZERO # include <fcntl.h> # ifndef MAP_FILE # define MAP_FILE 0 # endif #endif #ifndef PROT_NONE # define PROT_NONE 0 #endif #if HAVE_MMAP_ANON # define zero_fd -1 # define map_flags MAP_ANON | MAP_PRIVATE #elif HAVE_MMAP_ANONYMOUS # define zero_fd -1 # define map_flags MAP_ANONYMOUS | MAP_PRIVATE #elif HAVE_MMAP_DEVZERO static int zero_fd; # define map_flags MAP_FILE | MAP_PRIVATE #endif unsigned long page; int handler_called = 0; void sigsegv_handler (int sig, int code, struct sigcontext *scp) { void *fault_address = (void *) (get_fault_addr (scp)); handler_called++; if (handler_called == 10) exit (4); if (fault_address != (void*)(page + 0x678)) exit (3); if (mprotect ((void *) page, 0x10000, PROT_READ | PROT_WRITE) < 0) exit (2); } void crasher (unsigned long p) { *(int *) (p + 0x678) = 42; } int main () { void *p; struct sigaction action; /* Preparations. */ #if !HAVE_MMAP_ANON && !HAVE_MMAP_ANONYMOUS && HAVE_MMAP_DEVZERO zero_fd = open ("/dev/zero", O_RDONLY, 0644); #endif /* Setup some mmaped memory. */ #ifdef __hpux /* HP-UX 10 mmap() often fails when given a hint. So give the OS complete freedom about the address range. */ p = mmap ((void *) 0, 0x10000, PROT_READ | PROT_WRITE, map_flags, zero_fd, 0); #else p = mmap ((void *) 0x12340000, 0x10000, PROT_READ | PROT_WRITE, map_flags, zero_fd, 0); #endif if (p == (void *)(-1)) exit (2); page = (unsigned long) p; /* Make it read-only. */ if (mprotect ((void *) page, 0x10000, PROT_READ) < 0) exit (2); /* Install the SIGSEGV handler. */ sigemptyset(&action.sa_mask); action.sa_handler = (void (*) (int)) &sigsegv_handler; action.sa_flags = 0; sigaction (EXC_BAD_ACCESS, &action, (struct sigaction *) NULL); /* SRG - a quick check but too much to hope for ;-) */ sigaction (SIGSEGV, &action, (struct sigaction *) NULL); sigaction (SIGBUS, &action, (struct sigaction *) NULL); /* The first write access should invoke the handler and then complete. */ crasher (page); /* The second write access should not invoke the handler. */ crasher (page); /* Check that the handler was called only once. */ if (handler_called != 1) exit (1); /* Test passed! */ return 0; } end file-------------------------------------------------------------------- ------------------------------------------------------------------------ -------- _______________________________________________ darwin-kernel mailing list | darwin-kernel(a)lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel Do not post admin requests to the list. They will be ignored.

1 0

EXC_BAD_ACCESS instead of SIGSEGV?
by Simon Gray 06 Nov '03

06 Nov '03

Good morning all, I wonder whether someone could point me in the right direction for this problem. I'm working with the GNU libsigsegv library (http://sourceforge.net/projects/libsigsegv/) as I'm doing some investigative work on page-mapping in Mac OSX 10.3 and I've run into a problem: The basic configure script tests signal handling via the code below but no longer works in 10.3 as a GDB trace shows that an EXC_BAD_ACCESS is generated instead of SIGSEGV in the function "crasher()", and it doesn't seem possible to install a signal handler for EXC_BAD_ACCESS (unsurprisingly, as its not a BSD signal). Inexperienced as I am with Mach, I'm really not sure the best approach to resolving this issue - any pointers? best regards Simon Gray PS I would have attached the code below but it goes over the 8K limit :( _______________________________________________ darwin-kernel mailing list | darwin-kernel(a)lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel Do not post admin requests to the list. They will be ignored.

1 0

Re: Newbie question about building Darwin
by Felix Kronlage 05 Nov '03

05 Nov '03

On Tue, Nov 04, 2003 at 10:18:41PM -0800, Jay O'Conor wrote: > Everything progresses reasonably well until near the end of the build. > One of the last steps that make wants to execute is 'kextsymboltool'. > Unfortunately, this tools isn't built by any of the steps in the kernel > build procedure in the manual, so the build aborts with an error: Shantonu just posted a mail a few days ago, that he provides a tarball with the toolchain for building xnu. (if you would have taken a look at the archives, you would have found that out yourself.) See: http://www.opendarwin.org/~ssen/ You should be able to compile kext_tools with 'xcodebuild' from the commandline, btw. -fkr -- gpg-fingerprint: 076E 1E87 3E05 1C7F B1A0 8A48 0D31 9BD3 D9AC 74D0 Felix Kronlage | FKR-RIPE | http://www.hazardous.org/fkr/ fkr(a){grummel.net|opendarwin.org} | http://www.opendarwin.org/~fkr/ [demime 0.98b removed an attachment of type application/pgp-signature] _______________________________________________ darwin-kernel mailing list | darwin-kernel(a)lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel Do not post admin requests to the list. They will be ignored.

1 0