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

Thread safety question
by Steve Gehrman 18 Feb '05

18 Feb '05

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 BOOL variable_initialized=NO; void* myVariable; NSLock* myLock; // assume lock has been initialized in main - (void)isThisThreadSafe; { if ( !variable_initialized ) { [myLock lock]; if ( !variable_initialized ) myVariable = allocateStuff(); variable_initialized = YES; [myLock unlock]; } return myVariable; } Can anyone confirm this, or offer advice on an alternate solution? Thanks. -steve

1 0

TerminatorX Mac OS X Port
by Jeshua Lacock 16 Feb '05

16 Feb '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Greetings, I did a GDB backtrace and got: Best regards, Jeshua Lacock __________________________ Programmer/Owner Phone: 877.240.1364 http://OpenOSX.com Fax: 415.462.6211 -_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_ _______________________________________________ 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… It seems that I had to make a few changes to get TerminatorX built here on Darwin 7.7.0. With a bit of work, I managed to get it to build cleanly, but, I get a "Bus Error" when running the binaries. Note that I changed "malloc.h" to "stdlib.h", and "soundcard.h" to "CoreAudio.h". Program received signal EXC_BAD_ACCESS, Could not access memory. 0x00091350 in std::list<tX_seqpar*, std::allocator<tX_seqpar*> >::insert(std::_List_iterator<tX_seqpar*, tX_seqpar*&, tX_seqpar**>, tX_seqpar* const&) (this=0x14d7dc, __position={<_List_iterator_base> = {_M_node = 0xbffff950}, <No data fields>}, __x=@0xbffff9b0) at /usr/include/gcc/darwin/3.3/c++/bits/list.tcc:97 97 __tmp->_M_prev = __position._M_node->_M_prev; (gdb) bt #0 0x00091350 in std::list<tX_seqpar*, std::allocator<tX_seqpar*> >::insert(std::_List_iterator<tX_seqpar*, tX_seqpar*&, tX_seqpar**>, tX_seqpar* const&) (this=0x14d7dc, __position={<_List_iterator_base> = {_M_node = 0xbffff950}, <No data fields>}, __x=@0xbffff9b0) at /usr/include/gcc/darwin/3.3/c++/bits/list.tcc:97 #1 0x00000000 in ?? () Cannot access memory at address 0x0 Cannot access memory at address 0x0 #2 0x00091330 in std::list<tX_seqpar*, std::allocator<tX_seqpar*> >::insert(std::_List_iterator<tX_seqpar*, tX_seqpar*&, tX_seqpar**>, tX_seqpar* const&) (this=0x14d7dc, __position={<_List_iterator_base> = {_M_node = 0xbffff9d0}, <No data fields>}, __x=@0xbffff9b0) at /usr/include/gcc/darwin/3.3/c++/ppc-darwin/bits/gthr-default.h:514 #3 0x0001f9cc in tX_seqpar_master_volume::tX_seqpar_master_volume() (this=0x3000220) at tX_seqpar.cc:425 #4 0x00092ff0 in __static_initialization_and_destruction_0(int, int) (__initialize_p=1, __priority=65535) at tX_mastergui.cc:90 #5 0x8fe17990 in __dyld_call_module_initializers_for_objects () #6 0x8fe17458 in __dyld_call_module_initializers () #7 0x8fe14584 in __dyld__dyld_make_delayed_module_initializer_calls () #8 0x00002a1c in _call_mod_init_funcs () at /SourceCache/Csu/Csu-46/crt.c:299 #9 0x000028f8 in _start (argc=1, argv=0xbffffe64, envp=0xbffffe6c) at /SourceCache/Csu/Csu-46/crt.c:217 #10 0x00002858 in start () I looked around a bit and can't figure it out. I am willing to pay you for your time. Any and help is greatly appreciated. This email sent to site_archiver(a)lists.apple.com

1 0

Re: Correction: Serial port arbitration
by Godfrey van der Linden 15 Feb '05

15 Feb '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Godfrey On Feb 15, , at 10:30, Godfrey van der Linden wrote: #include <IOKit/serial/ioss.h> #include <IOKit/serial/IOSerialKeys.h> int fd; int preempt; tryAgain: for (;;) { do { fd = open(<tty.dialindevice>, O_NONBLOCK); if (-1 == fd) if (EBUSY == errno) continue; else exit(1); // Fatal error // clear O_NONBLOCK flag if (-1 == fcntl(fd, F_SETFL, 0)) exit(1); // fatal error if (-1 == ioctl(fd, TIOCEXCL, 0)) if (EBUSY == errno) { close(fd); fd = -1; continue; } else exit(1); // Fatal error } while(false); if (-1 != fd) break; // We must have had an EBUSY error to be here so wait for idle if (kIOReturnSuccess != IORegistryEntrySetProperty(portre, CFSTR(kIOTTYWaitForIdleKey), kCFBooleanTrue) exit(1); // Fatal error }; // At this point you have the port exclusively and non-preemtibly // so now is a good time to program up the modem or do any other // equipment initialisation you desire. preempt = 1;` if (-1 == ioctl(fd, IOSSPREEMPT, &preempt)) // Enable preemption exit(1); // Fatal error // Now the port is in the preemptible - blocking state // time to issue a blocking read or a select as needed // waiting for whatever event you are want. if (-1 == read(fd, buf, sizeof(buf)) // Pseudo code if (EBUSY == errno) { close(fd); // We have been preempted. goto tryAgain; } else exit(1); // probably fatal // We have a connection and it is time to say we are no longer // preemptible preempt = 0; if (-1 == ioctl(fd, IOSSPREEMPT, &preempt)) // Disable preemption exit(1); // Fatal error _______________________________________________ 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/gvdl%40apple.com This email sent to gvdl(a)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… And another correction, the error returned is not EIO but EBUSY when you are preempted, again see below. (I knew I shouldn't have written the code without testing ;-) Dan you are right, I got the logic inverted. The procedure should read (Same caveats as last time). // iterate over I/O registry to find desired node and keep your // registry entry node (See DTS docs for a write up on how to do this) io_registry_entry_t portre = <do the search> // That's it while we are running non-preemtibly/exclusive further // attempts to open the port on either cu.* or tty.* will fail with EBUSY This email sent to site_archiver(a)lists.apple.com

1 0

Correction: Serial port arbitration
by Godfrey van der Linden 15 Feb '05

15 Feb '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com #include <IOKit/serial/ioss.h> #include <IOKit/serial/IOSerialKeys.h> int fd; int preempt; tryAgain: for (;;) { do { fd = open(<tty.dialindevice>, O_NONBLOCK); if (-1 == fd) if (EBUSY == errno) continue; else exit(1); // Fatal error // clear O_NONBLOCK flag if (-1 == fcntl(fd, F_SETFL, 0)) exit(1); // fatal error if (-1 == ioctl(fd, TIOCEXCL, 0)) if (EBUSY == errno) { close(fd); fd = -1; continue; } else exit(1); // Fatal error } while(false); if (-1 != fd) break; // We must have had an EBUSY error to be here so wait for idle if (kIOReturnSuccess != IORegistryEntrySetProperty(portre, CFSTR(kIOTTYWaitForIdleKey), kCFBooleanTrue) exit(1); // Fatal error }; // At this point you have the port exclusively and non-preemtibly // so now is a good time to program up the modem or do any other // equipment initialisation you desire. preempt = 1;` if (-1 == ioctl(fd, IOSSPREEMPT, &preempt)) // Enable preemption exit(1); // Fatal error // Now the port is in the preemptible - blocking state // time to issue a blocking read or a select as needed // waiting for whatever event you are want. if (-1 == read(fd, buf, sizeof(buf)) // Pseudo code if (EIO == errno) { close(fd); // We have been preempted. goto tryAgain; } else exit(1); // probably fatal // We have a connection and it is time to say we are no longer // preemptible preempt = 0; if (-1 == ioctl(fd, IOSSPREEMPT, &preempt)) // Disable preemption exit(1); // Fatal error _______________________________________________ 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… Dan you are right, I got the logic inverted. The procedure should read (Same caveats as last time). // iterate over I/O registry to find desired node and keep your // registry entry node (See DTS docs for a write up on how to do this) io_registry_entry_t portre = <do the search> // That's it while we are running non-preemtibly/exclusive further // attempts to open the port on either cu.* or tty.* will fail with EBUSY This email sent to site_archiver(a)lists.apple.com

1 0

Re: shm tries
by appledisclose＠batchmaker.de 15 Feb '05

15 Feb '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Organization: handmade ! User-agent: MacSOUP/D-2.5 (Mac OS X version 10.3.8) David Osguthorpe <David.Osguthorpe(a)uchsc.edu> wrote: > On Mon, Feb 14, 2005 at 02:39:37PM -0500, John Francini wrote: > > You might have to roll your own... Thanks for all answers, I already had this feeling... -- Hado Hein, Berlin, Fed.Rep. of Germany http://www.batchmaker.de (software authoring, including stage lighting) _______________________________________________ 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: Serial port arbitration (UUCP device locking protocol?)
by Dan Bernstein 15 Feb '05

15 Feb '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:reply-to:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:references; b=G/qNwK4N7MaYXXUwM4n2tGaaqpms1V+ysgGVHkZCqds7a8yBg/lgWGn5coeG9gMF20tYR9c80yf0c6taGMNl/kuOfS/el8olcECU/d8RvjIyUy7K5bi8xBLpRG2jP3fiEh7r7G1kXihKV1uFUb2b/HhjChnGhy1tfW0fBLutFHI= Thanks again Godfrey. Sorry to keep bugging you with this, but I'm still confused. In ioss.h it says that if the preempt flag is true, then the device is pre-emptible, whereas in your code, you set it to 0 (false) right before you say "Now the port is in the preemptible - blocking state". I also tried doing a read there, and it didn't seem to be pre-emptible. On the other hand, when I do a read after setting the preempt flag to 1, then read blocks until another program opens the device, at which point it returns 0. Further read return -1 with errno==EBUSY. Did I get it all wrong? -- Dan On Mon, 14 Feb 2005 10:17:06 -0800, Godfrey van der Linden <gvdl(a)apple.com> wrote: > G'day Dan, > > > On Sat, 12 Feb 2005 11:47:00 -0800, Godfrey van der Linden > > <gvdl(a)apple.com> wrote: > >> // At this point you have the port exclusively and non-preemtibly > >> // so now is a good time to program up the modem or do any other > >> // equipment initialisation you desire. > > > > How do I do that? Can I just write to the dialin device or should I > > open the dialout? > > At this point in the sample code the fd is an exclusive open and > non-preemptible handle on the serial port, you can read/write to it to > talk directly to the modem. > > Godfrey > > _______________________________________________ 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: Problems with compiling xnu-517.11.1
by Kevin Van Vechten 14 Feb '05

14 Feb '05

site_archiver(a)lists.apple.com Delivered-To: Darwin-dev(a)lists.apple.com - Kevin On Feb 13, 2005, at 11:16 PM, Kevin Van Vechten wrote: - Kevin On Feb 13, 2005, at 2:33 PM, Benjamin Herrenschmidt wrote: _______________________________________________ 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… A new tarball been posted both at Apple's site and at OpenDarwin. I verified it builds OK on my system. Ben Herrenschmidt correctly diagnosed the problem as an incomplete version of osfmk/ppc/machine_routines_asm.s. Ben's diagnosis is correct. I'll try to post a fix with the complete version of this file tomorrow. It seems the file osfmk/ppc/machine_routines_asm.s is incomplete, it's bit "cropped" in half some way ... Ben. _______________________________________________ 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

Re: shm tries
by David Osguthorpe 14 Feb '05

14 Feb '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com User-agent: Mutt/1.4.2.1i On Mon, Feb 14, 2005 at 02:39:37PM -0500, John Francini wrote: > Other unix systems include tools such as ipcs and ipcrm that you could > use to remove shared memory segments. > > For whatever reason Mac OS X doesn't include any command line tools to > play with the various System V-style IPC facilities, including shared > memory. > > You might have to roll your own... > or you could just download the ipcs ppc.rpm from opendarwin which gives you both these tools last version I got was ipcs-1.0-0.ppc.rpm David _______________________________________________ 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 tries
by John Francini 14 Feb '05

14 Feb '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com You might have to roll your own... John Francini On Feb 14, 2005, at 5:53 AM, Hado Hein wrote: When I terminate the app in the debugger them shms are staying in ram. http://www.batchmaker.de (software authoring, including stage lighting) _______________________________________________ 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/francini%40mac.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… Other unix systems include tools such as ipcs and ipcrm that you could use to remove shared memory segments. For whatever reason Mac OS X doesn't include any command line tools to play with the various System V-style IPC facilities, including shared memory. I'm doing trial on a software that will allocate a lot of shm segments. At the moment I'm doing just one shmget, but tend to do hundreds of them Is there a way, besides a restart as i'm doing now, that can clear those straw segments cause I haven't called shmdt on them while debugging. -- Hado Hein, Berlin, Fed.Rep. of Germany This email sent to francini(a)mac.com This email sent to site_archiver(a)lists.apple.com

1 0

Re: Serial port arbitration (UUCP device locking protocol?)
by Godfrey van der Linden 14 Feb '05

14 Feb '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com G'day Dan, On Sat, 12 Feb 2005 11:47:00 -0800, Godfrey van der Linden <gvdl(a)apple.com> wrote: // At this point you have the port exclusively and non-preemtibly // so now is a good time to program up the modem or do any other // equipment initialisation you desire. How do I do that? Can I just write to the dialin device or should I open the dialout? Godfrey _______________________________________________ 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… At this point in the sample code the fd is an exclusive open and non-preemptible handle on the serial port, you can read/write to it to talk directly to the modem. This email sent to site_archiver(a)lists.apple.com

1 0