May 2008 - Darwin-dev - lists.apple.com

Re: Diskarbitration callback
by Axel Luttgens 08 May '08

08 May '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Le 8 mai 08 à 10:21, Andy Green a écrit : Cheers, Andy 8<---- struct Nothing { bool dummy; }; static void AndygDiskArbCallback( DADiskRef disk, DADissenterRef dissenter, void * context) { printf("If I ever see this, my problems are over\n"); Nothing * n = (Nothing *)context; n->dummy = true; // I want to give a semaphore here to unlock the main thread } int main(int argc, char *argv[]) { const char * cStr = "disk0s2"; Nothing * n = new Nothing; n->dummy = false; DASessionRef diskArbSession = DASessionCreate(kCFAllocatorDefault); assert(diskArbSession); DADiskRef disk = DADiskCreateFromBSDName( kCFAllocatorDefault, diskArbSession, cStr ); assert(disk); printf("confirming %s == %s??\n", DADiskGetBSDName(disk), cStr); DADiskRef whole = DADiskCopyWholeDisk( disk ); if ( whole ) { printf("wholedisk == %s\n", DADiskGetBSDName(whole)); DADiskUnmount(whole, kDADiskUnmountOptionWhole | kDADiskUnmountOptionForce, AndygDiskArbCallback, n); CFRelease(whole); } // I'll do this too, once I get the callback from Unmount... // DADiskEject( disk, kDADiskEjectOptionDefault, AndygDiskArbCallback, NULL ); CFRelease(disk); CFRelease(diskArbSession); printf("AT THE END OF IT ALL, n=%d\n", n->dummy); delete(n); } Hello Andy, HTH, Axel #include <stdio.h> #include <DiskArbitration/DiskArbitration.h> typedef struct { int dummy; } Nothing; int main(int argc, char *argv[]) { const char * cStr; static Nothing * n; DASessionRef diskArbSession; DADiskRef disk; DADiskRef whole; SInt32 rc; cStr = "disk1s1"; n = (Nothing *) malloc(sizeof(Nothing)); n->dummy = -1; diskArbSession = NULL; disk = NULL; whole = NULL; rc = -1; free(n); 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… I've written some code which uses the diskArbitration framework to unmount (and eject) a disk. It works, in that the disk gets unmounted, but my callback isn't being called, even though the operation is successful. Any chance of spotting the mistake in this code fragment? :- printf("------Wait..."); sleep(1); // instead of sleep, I want to wait on a semaphore here, until the unmount is complete. // unfortunately the callback isn't being called! You'll find hereafter an amended version of your code. For legibility purposes, I kept it very verbose and as as close as possible to your own version. But please check and double-check! I wrote it very quickly and am not very well versed myself into those matters. ;-) void AndygDiskArbCallback( DADiskRef disk, DADissenterRef dissenter, void * context); void AndygDiskArbCallback( DADiskRef disk, DADissenterRef dissenter, void * context) { printf("If I ever see this, my problems are over\n"); if (dissenter == NULL) ((Nothing *)context)->dummy = 0; else ((Nothing *)context)->dummy = DADissenterGetStatus(dissenter) ? DADissenterGetStatus(dissenter) : -1; // I want to give a semaphore here to unlock the main thread CFRunLoopStop(CFRunLoopGetCurrent()); } diskArbSession = DASessionCreate(kCFAllocatorDefault); if (diskArbSession != NULL) { DASessionScheduleWithRunLoop(diskArbSession, CFRunLoopGetCurrent(), CFSTR("drvHostBaseDA")); disk = DADiskCreateFromBSDName(kCFAllocatorDefault, diskArbSession, cStr); if (disk != NULL) { printf("Got disk bearing BSD name '%s' starting with string '%s'. \n", DADiskGetBSDName(disk), cStr); whole = DADiskCopyWholeDisk(disk); if (whole != NULL) { printf("Going to unmount whole disk with BSD name '%s'.\n", DADiskGetBSDName(whole)); DADiskUnmount(whole, kDADiskUnmountOptionWhole | kDADiskUnmountOptionForce, AndygDiskArbCallback, n); rc = CFRunLoopRunInMode(CFSTR("drvHostBaseDA"), 10.0, TRUE); if (rc == kCFRunLoopRunHandledSource) { printf("Something happened.\n"); if (n->dummy == 0) { printf("Unmount has been successful.\n"); // I'll do this too, once I get the callback from Unmount... // DADiskEject( disk, kDADiskEjectOptionDefault, AndygDiskArbCallback, NULL ); } else { printf("Unmount failed with code '%d'.\n", n->dummy); } } CFRelease(whole); } CFRelease(diskArbSession); } DASessionUnscheduleFromRunLoop(diskArbSession, CFRunLoopGetCurrent(), CFSTR("drvHostBaseDA")); CFRelease(disk); } This email sent to site_archiver(a)lists.apple.com

1 0

Re: Diskarbitration callback
by Philip Aker 08 May '08

08 May '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On 08-05-08, at 01:21, Andy Green wrote: 8<---- struct Nothing { bool dummy; }; static void AndygDiskArbCallback( DADiskRef disk, DADissenterRef dissenter, void * context) { printf("If I ever see this, my problems are over\n"); Nothing * n = (Nothing *)context; n->dummy = true; // I want to give a semaphore here to unlock the main thread } int main(int argc, char *argv[]) { const char * cStr = "disk0s2"; Nothing * n = new Nothing; n->dummy = false; DASessionRef diskArbSession = DASessionCreate(kCFAllocatorDefault); assert(diskArbSession); DADiskRef disk = DADiskCreateFromBSDName( kCFAllocatorDefault, diskArbSession, cStr ); assert(disk); printf("confirming %s == %s??\n", DADiskGetBSDName(disk), cStr); DADiskRef whole = DADiskCopyWholeDisk( disk ); if ( whole ) { printf("wholedisk == %s\n", DADiskGetBSDName(whole)); DADiskUnmount(whole, kDADiskUnmountOptionWhole | kDADiskUnmountOptionForce, AndygDiskArbCallback, n); CFRelease(whole); } // I'll do this too, once I get the callback from Unmount... // DADiskEject( disk, kDADiskEjectOptionDefault, AndygDiskArbCallback, NULL ); CFRelease(disk); CFRelease(diskArbSession); printf("AT THE END OF IT ALL, n=%d\n", n->dummy); delete(n); } Philip Aker echo astwta@lvpc.dslh@nl | tr a-z@. p-za-o.@ Sent from my SE/30 _______________________________________________ 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've written some code which uses the diskArbitration framework to unmount (and eject) a disk. It works, in that the disk gets unmounted, but my callback isn't being called, even though the operation is successful. Any chance of spotting the mistake in this code fragment? :- I think you'll have to schedule the session with a run loop to get any callback actions. printf("------Wait..."); sleep(1); // instead of sleep, I want to wait on a semaphore here, until the unmount is complete. // unfortunately the callback isn't being called! This email sent to site_archiver(a)lists.apple.com

1 0

FW: Diskarbitration callback
by Andy Green 08 May '08

08 May '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Thread-index: AciwRO8cKHSTxPPcQry4X3WFMZFGAwAn5HWA Thread-topic: Diskarbitration callback I've written some code which uses the diskArbitration framework to unmount (and eject) a disk. It works, in that the disk gets unmounted, but my callback isn't being called, even though the operation is successful. Any chance of spotting the mistake in this code fragment? :- Cheers, Andy 8<---- struct Nothing { bool dummy; }; static void AndygDiskArbCallback( DADiskRef disk, DADissenterRef dissenter, void * context) { printf("If I ever see this, my problems are over\n"); Nothing * n = (Nothing *)context; n->dummy = true; // I want to give a semaphore here to unlock the main thread } int main(int argc, char *argv[]) { const char * cStr = "disk0s2"; Nothing * n = new Nothing; n->dummy = false; DASessionRef diskArbSession = DASessionCreate(kCFAllocatorDefault); assert(diskArbSession); DADiskRef disk = DADiskCreateFromBSDName( kCFAllocatorDefault, diskArbSession, cStr ); assert(disk); printf("confirming %s == %s??\n", DADiskGetBSDName(disk), cStr); DADiskRef whole = DADiskCopyWholeDisk( disk ); if ( whole ) { printf("wholedisk == %s\n", DADiskGetBSDName(whole)); DADiskUnmount(whole, kDADiskUnmountOptionWhole | kDADiskUnmountOptionForce, AndygDiskArbCallback, n); CFRelease(whole); } printf("------Wait..."); sleep(1); // instead of sleep, I want to wait on a semaphore here, until the unmount is complete. // unfortunately the callback isn't being called! // I'll do this too, once I get the callback from Unmount... // DADiskEject( disk, kDADiskEjectOptionDefault, AndygDiskArbCallback, NULL ); CFRelease(disk); CFRelease(diskArbSession); printf("AT THE END OF IT ALL, n=%d\n", n->dummy); delete(n); } _______________________________________________ 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

Diagnosing a sporadically non-responsive Mac?
by Andre-John Mas 08 May '08

08 May '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Hi, Andre _______________________________________________ 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 a Mac that I use that has started to sporadically not respond. What I mean by that is that it can be running for a day or so and then quit accepting network requests or will accept a mouse click to launch an application on the dock, but the application will bounce for ever. Is there anything I could install that could help me diagnose the issue? Maybe something that will get triggered on a CPU spike or something of the sort. This email sent to site_archiver(a)lists.apple.com

1 0

Re: Getting a file path of dylib
by Jason Watts 07 May '08

07 May '08

site_archiver(a)lists.apple.com Delivered-To: Darwin-dev(a)lists.apple.com Domainkey-signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=sbcglobal.net; h=Received:X-YMail-OSG:X-Yahoo-Newman-Property:Message-Id:From:To:In-Reply-To:Content-Type:Content-Transfer-Encoding:Mime-Version:Subject:Date:References:X-Mailer; b=pFMBr3KpRYGOTxSxwvNCXUam16uFOWnV9Ni/P9ReuMM+jTV5Zm32VpafnrKV5ASjleLpvhdjz8ziw5/bGGQv1Wl0gsaF04UE42KsiFXSizeayX4PVnoUpZIqtKCDZZEXKmeVV513HUTOGThSrJHSKnH3ZVLTGnJg8A1mzXPrBqY= ; I think dladdr () will do what you want (in dlfcn.h). See man dladdr. ¿.) Jason Watts On May 7, 2008, at 11:35 AM, David Alter wrote: thanks for the help -dave _______________________________________________ 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/jjwatts%40sbcglobal.net _______________________________________________ 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… s there a way to get the file path of a dylib from within the dylib. I do NOT want the path of the executable that loads it I want the path of the dylib its self. This email sent to jjwatts(a)sbcglobal.net This email sent to site_archiver(a)lists.apple.com

1 0

Getting a file path of dylib
by David Alter 07 May '08

07 May '08

site_archiver(a)lists.apple.com Delivered-To: Darwin-dev(a)lists.apple.com thanks for the help -dave _______________________________________________ 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… s there a way to get the file path of a dylib from within the dylib. I do NOT want the path of the executable that loads it I want the path of the dylib its self. This email sent to site_archiver(a)lists.apple.com

1 0

Re: Trying again: Question concerning Tiger->Leopard changes for unformatted BSD disk device.
by Chris Sarcone 06 May '08

06 May '08

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: Trying again: Question concerning Tiger->Leopard changes for unformatted BSD disk device.
by kkikuchi＠rite-solutions.com 06 May '08

06 May '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Importance: Normal I am writing a program to talk to a USB device that looks like an unformatted BSD filesystem (eg., /dev/rdisk?) -- by the way, that all works (ioreg looks as exptected, I think). On Tiger 10.4.11, my (c++ command-line) applicattion can open the disk (open(dev_path,(O_RDWR|O_EXLOCK)), and read and write to it. However, when run on Leopard (10.5, 10.5.2), even as sudo, the application fails to open the disk with a "Permission denied" error. The /dev/rdisk? shows up on Leopard as read-only (on Tiger it's read/write for the owner, read-only for group) and the Disk Utility says that the unformatted disk is "Locked" and is not writable. The 'ioreg -l' display (IOMedia class) says the device is 'Writable' ( = Yes) in Tiger and is not writable in Leopard. Thanks for any help anyone can give to resolve this situation. Kyle Kikuchi P.S.: Also, it seems that I have to use the character (rdisk) device rather than the block device -- the application doesn't work on Tiger when I use the block device. _______________________________________________ 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: mach_absolute_time() vs. sleep() [solved]
by Terry Lambert 06 May '08

06 May '08

site_archiver(a)lists.apple.com Delivered-To: Darwin-dev(a)lists.apple.com Conversion to lvalue types happens before the operation is completed. -- 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 May 1, 2008, at 1:55 PM, Steve Checkoway wrote: So let's consider the three cases above: 1. info.denom / 1e9 * info.numer has units of mt * ns / 1e9 = mt * s. 2. 1e9 * info.numer / info.denom has units of 1e9 * ns / mt = as / mt (that is attoseconds per mach time or 10^-18 seconds per mach time) 3. 1e-9 * info.numer / info.denom has units of 1e-9 * ns / mt = s/mt. Your original (i.e., number 3) was correct. I'm not sure exactly what Terry was talking about when he was saying you were making mistakes. The following code is a complete working example. The numbers I get are pretty close to being the number of seconds asleep. The order of the operators in C is such that the multiply happens before the divide. If that happens, and your time base happens to be close (the closer to exactly 1GHz, the more likely this is to be an issue), then you've effectively reduce the value below the resolution of the significance of the digits in the mantissa. When that happens, rounding gets to be very important, and tiny rounding issues get magnified. If I had been assured this wasn't homework, I would have pointed this out. Moving the 1e-9 multiply to the next line ensures the order of the divide and the "multiply" (divide by 1e9) avoid losing significant digits. Amusingly, I've seen this problem before internally; usually it's someone trying to second-guess the clock; I say amusingly because all our clock timers are done on the basis of real intervals, not lbolt values, so in something like a sleep(6);, we don't have the +/- 1 second rounding error you might see on other platforms. This email sent to site_archiver(a)lists.apple.com

1 0

Re: C++ std::locale's ctor only accepts "C" and "POSIX"
by Lutz Helwing 05 May '08

05 May '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com User-agent: Thunderbird 2.0.0.9 (Macintosh/20071031) I made a small mistake when writing my initial message. I forgot to print the e.what() message when the std::locale ctor throws an exception. I've updated the code in the link to do that. Here's the new output on my OS X machine; as expected, it just says that the locale name is invalid: === BEGING TERMINAL OUTPUT === Trying c locales... Succeeded in calling setlocale(LC_ALL, "C") Succeeded in calling setlocale(LC_ALL, "POSIX") Succeeded in calling setlocale(LC_ALL, "en_US") Succeeded in calling setlocale(LC_ALL, "en_US.UTF-8") Succeeded in calling setlocale(LC_ALL, "pl_PL") Succeeded in calling setlocale(LC_ALL, "pl_PL.ISO8859-2") Trying C++ locales... Succeeded ctor for std::locale with "C" Succeeded ctor for std::locale with "POSIX" Failed ctor for std::locale with "en_US": locale::facet::_S_create_c_locale name not valid Failed ctor for std::locale with "en_US.UTF-8": locale::facet::_S_create_c_locale name not valid Failed ctor for std::locale with "pl_PL": locale::facet::_S_create_c_locale name not valid Failed ctor for std::locale with "pl_PL.ISO8859-2": locale::facet::_S_create_c_locale name not valid === END TERMINAL OUTPUT === -Patrick On Mar 8, 2008, at 9:57 PM, Patrick Rutkowski wrote: I'm currently studying how locales work in C/C++, I'm using OS X machine and a Linux machine for testing. I wrote the following test program to make sure that I can successfully initialize a few locales on both my systems: http://www.rutski89.com:31597/upload/test.cpp Running this on my linux machine I get exactly the output I expected, all success: === BEGING TERMINAL OUTPUT === Trying c locales... Succeeded in calling setlocale(LC_ALL, "C") Succeeded in calling setlocale(LC_ALL, "POSIX") Succeeded in calling setlocale(LC_ALL, "en_US") Succeeded in calling setlocale(LC_ALL, "en_US.UTF-8") Succeeded in calling setlocale(LC_ALL, "pl_PL") Succeeded in calling setlocale(LC_ALL, "pl_PL.ISO8859-2") Trying C++ locales... Succeeded ctor for std::locale with "C" Succeeded ctor for std::locale with "POSIX" Succeeded ctor for std::locale with "en_US" Succeeded ctor for std::locale with "en_US.UTF-8" Succeeded ctor for std::locale with "pl_PL" Succeeded ctor for std::locale with "pl_PL.ISO8859-2" === END TERMINAL OUTPUT === However, when running on my OS X machine with the stock g++ (running on 10.5.1) I get the following: === BEGING TERMINAL OUTPUT === Trying c locales... Succeeded in calling setlocale(LC_ALL, "C") Succeeded in calling setlocale(LC_ALL, "POSIX") Succeeded in calling setlocale(LC_ALL, "en_US") Succeeded in calling setlocale(LC_ALL, "en_US.UTF-8") Succeeded in calling setlocale(LC_ALL, "pl_PL") Succeeded in calling setlocale(LC_ALL, "pl_PL.ISO8859-2") Trying C++ locales... Succeeded ctor for std::locale with "C" Succeeded ctor for std::locale with "POSIX" Failed ctor for std::locale with "en_US" Failed ctor for std::locale with "en_US.UTF-8" Failed ctor for std::locale with "pl_PL" Failed ctor for std::locale with "pl_PL.ISO8859-2" === END TERMINAL OUTPUT === It seems that the system is only able to create C++ locales for "C" and "POSIX", which is puzzling, and also very unsatisfactory. It would be unhappy, but at least settling, if I found a page that said something like "C++ locales on OS X only support "C" and "POSIX", tough luck!". However, such a page is nowhere to be found, and in fact I have found evidence to the contrary. Searching "site:apple.com std::locale" in google brings up various pages where people describe issues with the usage of std::locale. However, these pages pertain to issues with the usage of the locale after it has already been created, which implies that the people who made the posts were actually able to create the locale to begin with; so why am I not able to create them? I've been banging my head against the wall for a while now; any help would be much appreciated -Patrick _______________________________________________ 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 I have exactly the same problem and it's a rather fatal one in my case. Am i assuming right that i can't use UTF-8 in my application? I'm using a part of the boost library. This is most likely the place where the exception is thrown. It works when i set the default locale of my shell to "C" or "POSIX" but nothing else... I'm relatively unexperienced in working with locales so i don't have the slightest idea Since i can't find very much about this exception i wonder if anyone has solved this problem. smime.p7s

1 0