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

Apache Perl CGI environment question
by John Michael Zorko 07 Feb '05

07 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

1 0

Re: detecting logout
by Pete Gontier 07 Feb '05

07 Feb '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com User-agent: Microsoft-Entourage/10.1.6.040913.0 After much tearing of hair and gnashing of teeth, it looks like the solution for me involves calling SessionGetInfo (to determine whether the current session is potentially a GUI session) and CGSessionCopyCurrentDictionary (to determine whether the current session is actually a GUI session). My code assumes that any potential GUI session which stops being an actual GUI session has been "logged out" of the console. Whenever the console user changes and the current session is not an actual GUI session, I quit. This may not be perfect, so feedback is welcome. (Happily, I don't have to worry about SSH logins [and doing so would thus be undesirable] because that would not be useful for this particular porgram.) circa 2/1/05 11:57 AM, "Pete Gontier" <pete(a)m-audio.com> wrote: > This may be a Mac OS X question as opposed to a Darwin question. > Nevertheless, this is the only relevant place I know of to ask this > question. > > I am developing a per-user bootstrap daemon. This program used to be a > startup item, but for obscure reasons not relevant to the problem at hand, > it became necessary to morph it into a per-user bootstrap daemon (and not, > it should be noted, a global bootstrap daemon). The key difference between > the two forms of program for purposes of this discussion is that a startup > item launches once at startup and a per-user bootstrap daemon launches > whenever a user logs in. This means that there may be many instances of a > single per-user bootstrap daemon, which is fine. The trick is knowing when > each instance should quit. When the user logs out, per-user bootstrap > daemons are not automagically killed, which I suppose may be a feature to > someone somewhere. For me, it presents a problem, because there is no reason > for an instance of my program to hang around after the user logs out. The > trick is knowing when that is. > > Apple's QA1133 seems to be the available wisdom on knowing when users login > and logout. > > http://developer.apple.com/qa/qa2001/qa1133.html > > Unfortunately, this QA seems to be out of date. It seems to describe the > pre-Panther, pre-Fast User Switching universe. QA1133 tells us we are to > believe the user is logging out when the user name is NULL, but, under > Panther, if user 1 logs in, user 2 logs in, and then user 2 logs out, at no > time will the console user be NULL. Even when the user is re-authenticating > to switch back to user 1, the console user is "loginwindow" as opposed to > NULL. (The user may be NULL after the *last* user logs out, but I haven't > tested this.) > > I could enable the copious debug logging done by the distributed > notification server and maybe pick out some choice undocumented strings as > they fly around in the system and hope to interpret their meaning correctly, > but that seems unlikely to be future-proof. > > If I must resort to higher-level APIs, I'd prefer Carbon-oriented solutions > rather than Cocoa-oriented solutions because the app in question is written > in C++, but I'll settle in a pinch for anything that works cleanly without > pulling me up into being a client of the window server. > > Any ideas? > > -- > > Pete Gontier > http://www.m-audio.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/pete%40m-audio.com > > This email sent to pete(a)m-audio.com -- Pete Gontier http://www.m-audio.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: Catching signal errors
by Kevin Harris 07 Feb '05

07 Feb '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com User-agent: Mozilla Thunderbird 1.0 (X11/20041206) Matt Watson wrote: write(1, sig_message, sizeof(sig_message)); exit(1); } -Kevin- _______________________________________________ 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… void fatalSigHandler(int sig, siginfo_t* info, void* context) { // We are very limited in what we are allowed to do in a signal handler. // Memory should not be allocated, and most libc functions should not be used. static char sig_message[] = "Aborting: Fatal signal occurred!\n"; Never call exit() from a signal handler. You want _Exit() instead. exit() will call atexit() handlers which will are likely not signal-safe. And you probably want to use fd 2 (stderr), instead of fd 1 in the write() call. Good suggestions. That's especially important when using dlopen or NSLinkModule on libraries. They could cause further signals or even signal deadlocks when they get unloaded. That said: if you're not going to add any more info in your signal handler, or do any cleanup like unlink() files, it may just be best to let it crash and generate a crash log. I assumed that he'd replace the handler with something else. One of the handlers I have for a project calls execv() to restart the daemon in case of a crash. That is after writing errors to a log and closing open file handles/resetting signal masks (as they are both inherited by the new process), and doing some other general cleanup of the daemon's lock/state files. This email sent to site_archiver(a)lists.apple.com

1 0

Re: Catching signal errors
by Matt Watson 07 Feb '05

07 Feb '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com write(1, sig_message, sizeof(sig_message)); exit(1); } matt. _______________________________________________ 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… void fatalSigHandler(int sig, siginfo_t* info, void* context) { // We are very limited in what we are allowed to do in a signal handler. // Memory should not be allocated, and most libc functions should not be used. static char sig_message[] = "Aborting: Fatal signal occurred!\n"; Never call exit() from a signal handler. You want _Exit() instead. exit() will call atexit() handlers which will are likely not signal-safe. And you probably want to use fd 2 (stderr), instead of fd 1 in the write() call. That said: if you're not going to add any more info in your signal handler, or do any cleanup like unlink() files, it may just be best to let it crash and generate a crash log. This email sent to site_archiver(a)lists.apple.com

1 0

Re: Catching signal errors
by Kevin Harris 07 Feb '05

07 Feb '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com User-agent: Mozilla Thunderbird 1.0 (X11/20041206) Rick Steele wrote: Forgive the ignorance but, if one has a function like this within an application: void SumFunction(){ How about this snippet: -------------------------------- #include <signal.h> #include <stdlib.h> #include <unistd.h> typedef void (*full_sighandler_t)(int,siginfo_t*,void*); void installSignalHandler(int sig, full_sighandler_t handler) { struct sigaction temp; sigaction(sig, 0, &temp); // Set the SA_SIGINFO flag so that sa_sigaction may be safely used. temp.sa_flags = SA_SIGINFO; temp.sa_sigaction = handler; sigemptyset(&temp.sa_mask); sigaction(sig, &temp, NULL); } write(1, sig_message, sizeof(sig_message)); exit(1); } int main(int argc, char** argv) { installSignalHandler(SIGSEGV, &fatalSigHandler); installSignalHandler(SIGBUS, &fatalSigHandler); int* foo = NULL; *foo = 0; exit(0); } -------------------------------- You should read the man pages for signal and sigaction. -Kevin- _______________________________________________ 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… Ptr pPtr = -1L; //create something that will obviously crash if deferenced char pSomethingElse; pSomethingElse = (*pPtr); // Causing the error } how would some one catch the signal that would inevitable occur so the app might be able to close down gracefully? Try 0 (or NULL) -- it should cause a SIGSEGV or SIGBUS (depending on the platform), whereas -1L could (potentially) be a valid chunk of memory. void fatalSigHandler(int sig, siginfo_t* info, void* context) { // We are very limited in what we are allowed to do in a signal handler. // Memory should not be allocated, and most libc functions should not be used. static char sig_message[] = "Aborting: Fatal signal occurred!\n"; This email sent to site_archiver(a)lists.apple.com

1 0

ApplePanels and backlight values
by Benjamin Herrenschmidt 07 Feb '05

07 Feb '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Hi ! ApplePanelA builds a ramp of panel values based on a list of panels (min/mid/max values) contained in it's property list. On most panels, these values go up to 0xff for the max backlight value. I've traced the ATI driver and it seems that it will blast the value passed by the control call directly into the appropriate LVDS control register (doesn't apply it's own scaling, only eventually inverts). My question is, is it safe to actually uses values up to 0xff ? It seems that the ApplePanelA::doIntegerSet() method is passing the value as-is to the driver (which is writing it as-is to the hw). So we end up using actual backlight value up to the HW maximum of 0xff... However, when I played with that back in the OS 9 days, I noticed the system was never writing the full 0xff to the LVDS modulation in the chip (usually max'ed it out around 0xd9). Unless I've missed something, the plist for that same panel (family 8) are all ending up with 0xff for max. I just want to make sure I won't damage any HW .... Regards, 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/site_archiver%40lists.app… This email sent to site_archiver(a)lists.apple.com

1 0

Catching signal errors
by Rick Steele 06 Feb '05

06 Feb '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Forgive the ignorance but, if one has a function like this within an application: void SumFunction(){ also does anyone know how to determine the size of a malloc'd object? ie: void *pObj = malloc(sizeof(long)); ... long pSizeOfPointer = GetPtrSize(pObj); -- Best Wishes Rick Steele _______________________________________________ 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… Ptr pPtr = -1L; //create something that will obviously crash if deferenced char pSomethingElse; pSomethingElse = (*pPtr); // Causing the error } how would some one catch the signal that would inevitable occur so the app might be able to close down gracefully? This email sent to site_archiver(a)lists.apple.com

1 0

Re: signal and dyld
by Markus Hitter 06 Feb '05

06 Feb '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Am 04.02.2005 um 20:08 schrieb Dair Grant: I'm using Xcode 1.5, gcc 3.3, with MACOSX_DEPLOYMENT_TARGET set to 10.2 and dead-stripping turned on. Cheers, Markus - - - - - - - - - - - - - - - - - - - Dipl. Ing. Markus Hitter http://www.jump-ing.de/ _______________________________________________ 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… If my brain isn't too tired already, I remember some issues in the combination of C++ and dead code stripping. You should avoid it unless you have Xcode 2.0, IIRC. See Xcode users list. This email sent to site_archiver(a)lists.apple.com

1 0

signal and dyld
by Dair Grant 04 Feb '05

04 Feb '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Hi, I have an app, built with Xcode, which includes iostream and attempts to register a signal handler. This combination appears to prevent launching on 10.2.x, and I'm trying to find a workaround. I'm using Xcode 1.5, gcc 3.3, with MACOSX_DEPLOYMENT_TARGET set to 10.2 and dead-stripping turned on. With this combination, iostream ends up leaving several references to wchar_t functions (e.g., towupper) in the final binary. Even though I never call these functions, enabling dead-stripping does not remove them. When I include a call to signal() to install my handler, I halt with: #0 0x8fe01220 in halt #1 0x8fe10654 in link_in_need_modules #2 0x8fe129c4 in _dyld_bind_fully_image_containing_address #3 0x900052b4 in _dyld_bind_fully_image_containing_address #4 0x9000520c in sigaction__ #5 0x9000f220 in signal__ #6 0x00008b30 in main I also get some undefined references to the console: undefined reference to _towupper expected to be defined in /usr/lib/libSystem.B.dylib If I remove the call to signal(), the app launches normally and no warnings about undefined references are emitted. The problem is presumably this "bind fully" step, where these unresolved symbols are looked up and not found. sigaction__ takes a flag indicating if it should perform this step or not (for my purposes, it would be ideal if it didn't; the signal handler I'm trying to install will always be there), but I can't see a public API to install a handler that allows control over this. The closest option seems to be _sigvec_nobind, but as this seems to be a private API I'd rather not use it unless I have to. If I have to assume that dead-stripping isn't going to remove everything, is there a solution to this that would allow an app which references (but never calls) missing functions to install a signal handler? Either some way to install a signal handler which doesn't fully bind the image, or some way to link the app that will cause this binding step to accept that they're missing in the same way they're accepted when the app is bound when launched without a signal handler? dair _______________________________________________ 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

Serial devices and cups printing
by Martin Bestmann 04 Feb '05

04 Feb '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Hi, Thanks, Martin _______________________________________________ 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'm working on a driver for a USB to serial device. I based my work on the AppleCDC sample code. Everything works great so far. Now I'm trying to print to a serial printer using the cups & gimp printing framework. I'm select my serial port in the Advanced Device List of the Printer Setup Utility. For some strange reason the device URI is fax://dev/myserial. When doing the same on a machine with a build in serial port I get serial://dev/modem. When I manually change the URI to serial I get a permission denied error when the port gets be opened. Here are my questions: 1) What do I need to do to get my device to identified as a serial instead of a fax device. What kind of call is the cups driver making to the device to determine this? 2) Any idea why I can not simply change the fax:// URI to serial://? This email sent to site_archiver(a)lists.apple.com

1 0