May 2007 - Darwin-kernel - lists.apple.com

Re: System.framework/PrivateHeaders
by Terry Lambert 22 May '07

22 May '07

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com You have to build your own xnu to get them. Because of this, they are considered SPI (System Private Interfaces). -- Terry _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/site_archiver%40lists.… On May 22, 2007, at 11:47 AM, Thomas Inskip wrote: I am trying to build fs_usage on MacOS X 10.4.9, but it won't compile because some types, such as kbufinfo_t, kd_threadmap, and kd_buf. The included makefile sets the header search paths to / System/Library/Frameworks/System.framework/PrivateHeaders and / System/Library/Frameworks/System.framework/PrivateHeaders/bsd, which are not available with the standard Xcode distribution. Surprisingly the compiler does no complain about not being able to find any header files, so it must be finding the wrong ones elsewhere, presumably. Could someone please tell me where I can get System.framework/ PrivateHeaders? I have found past threads to this regard in the archives, but unfortunately no solutions. These header files describe the contract between the kernel and the code that you promise you will recompile each time you recompile the kernel, so that the data interfaces described by the structures and other data objects in those headers will match up between them. If the reason you are doing this is to use the code as the basis of your own tool, the normally suggested route is to write code to talk to the system supplied tool instead. This isolates your code from changes which could otherwise cause it to break, if you were to use the interfaces directly, yourself. This email sent to site_archiver(a)lists.apple.com

1 0

System.framework/PrivateHeaders
by Thomas Inskip 22 May '07

22 May '07

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Dkim-signature: a=rsa-sha1; c=relaxed/relaxed; d=gmail.com; s=beta; h=domainkey-signature:received:received:mime-version:content-transfer-encoding:message-id:content-type:to:from:subject:date:x-mailer; b=tJ7Lh1OcyDK7suZwYyvhOhFHdGftlyFzDikKaEPSNwPPmtw/RPuJ/qqm/plF0BQAVSOOXkZSs86LrPvIYwJeRL5oKCrvtSuGOcnQzZEflVYqcxKO/Tu/kFGkmyuRqCovucPofa5BxfriwzoS4E53N/DrkPNIp3WJFLMbRKSUcpY= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=beta; h=received:mime-version:content-transfer-encoding:message-id:content-type:to:from:subject:date:x-mailer; b=rG4MgjlbGl/ma5JP2LWypfrXZlwdVznTsuSDEEowzGK8rDMQi1OVFjO7DLaNBDy8n0hfep2Fc45HNl5FpQHCW4dIJ2rT/mbyZtz0LIHjxl0FnYKNG50i3YlMQR8WwIHSKAiD5q7abTTxOCVG+I9Qrsb11hJtTv858SgcDkIOCDE= Thanks in advance. _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/site_archiver%40lists.… I am trying to build fs_usage on MacOS X 10.4.9, but it won't compile because some types, such as kbufinfo_t, kd_threadmap, and kd_buf. The included makefile sets the header search paths to /System/Library/ Frameworks/System.framework/PrivateHeaders and /System/Library/ Frameworks/System.framework/PrivateHeaders/bsd, which are not available with the standard Xcode distribution. Surprisingly the compiler does no complain about not being able to find any header files, so it must be finding the wrong ones elsewhere, presumably. Could someone please tell me where I can get System.framework/ PrivateHeaders? I have found past threads to this regard in the archives, but unfortunately no solutions. This email sent to site_archiver(a)lists.apple.com

1 0

(no subject)
by Todd Heberlein 22 May '07

22 May '07

site_archiver(a)lists.apple.com Delivered-To: Darwin-kernel(a)lists.apple.com Here is the BSM audit record from a PowerPC showing the execve: Thanks, Todd _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/site_archiver%40lists.… I apologize if this is a bonehead question, but why is the Mac doing an execve() on "/dev/null"? header,123,1,execve(2),0,Mon May 21 09:30:23 2007, + 431 msec path,/dev/null path,/dev/null attribute,20666,root,wheel,46807492,0,50331650 subject,heberlei,heberlei,staff,heberlei,staff, 452,384,50331650,0.0.0.0 return,success,0 trailer,123 This email sent to site_archiver(a)lists.apple.com

1 0

Re: all connect(2) calls are failures?
by Todd Heberlein 18 May '07

18 May '07

site_archiver(a)lists.apple.com Delivered-To: Darwin-kernel(a)lists.apple.com On May 18, 2007, at 2:30 AM, Josh Graessley wrote: Thanks! Todd _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/site_archiver%40lists.… Most apps use non-blocking connects. These will start the connect and return immediately with an error indicating that the operation is in progress. This isn't a failure, it's just part of the API. This email sent to site_archiver(a)lists.apple.com

1 0

Re: all connect(2) calls are failures?
by Josh Graessley 18 May '07

18 May '07

site_archiver(a)lists.apple.com Delivered-To: Darwin-kernel(a)lists.apple.com -josh (from afar) A typical audit record is: The audit flags are set for "all". Thanks, Todd _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/jgraessley%40apple.com _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/site_archiver%40lists.… Most apps use non-blocking connects. These will start the connect and return immediately with an error indicating that the operation is in progress. This isn't a failure, it's just part of the API. On May 17, 2007, at 4:13 AM, Todd Heberlein <todd_heberlein(a)mac.com> wrote: I am looking through Apple's BSM audit logs (Mac OS X 10.4.9 on a G5), and all records for the connect(2) system call to an IP address are reported as failures. There are no examples of successful connection requests even though I know they took place. header,88,1,connect(2),0,Tue May 15 09:44:38 2007, + 731 msec argument,1,0x10,fd socket-inet,2,80,17.250.248.77 subject,heberlei,heberlei,staff,heberlei,staff, 384,196,50331650,0.0.0.0 return,failure : Operation now in progress,4294967295 trailer,88 Is this a bug in Apple's kernel (i.e., inadvertently reporting a successful system call as a failure), or did Apple choose *not* to report successful network connections? This email sent to jgraessley(a)apple.com This email sent to site_archiver(a)lists.apple.com

1 0

boot failure xnu-792.17.14 DEBUG_PPC
by Arifumi Matsumoto 18 May '07

18 May '07

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Hi, I compiled xnu-792.17.14 for both RELEASE_PPC and DEBUG_PPC. My PBG4'15 successfully booted with RELEASE_PPC mach_kernel. However, with DEBUG_PPC kernel, it tells me to reboot several seconds after the apple logo. This is my first time to build a custom kernel, (though I didn't change any configurations for both kernels) so I might miss some additional steps for using DEBUG_PPC kernel. Anybody experienced the same trouble ? I'm glad if I can have any information about DEBUG kernel. Best regards, -- a(a)arifumi.net _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/site_archiver%40lists.… This email sent to site_archiver(a)lists.apple.com

1 0

Re: Regarding setuid Problem on Mac OS X 10.2
by Quinn 18 May '07

18 May '07

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com i.e setuid(uid) API is not working . Its return -1. The same code is working fine on High end but I got the problem on Mac OS X 10.2 series. S+E -- Quinn "The Eskimo!" <http://www.apple.com/developer/> Apple Developer Relations, Developer Technical Support, Core OS/Hardware _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/site_archiver%40lists.… At 17:16 +0530 17/5/07, subodh singh wrote: I got one problem on Macintosh Mac OS X 10.2.4, You should not support 10.2.4. Even if you want to support 10.2.x, you should only support the /last/ version of that train (10.2.8). Doing anything else is a waste of your time and, in the case of these older OS versions, is doing your users a disservice (the software updates contain vital changes, and by supporting anything other than the latest version you're encouraging folks not to update). So please advise and also suggest me some alternative way to achieve the same. When setuid fails, what's the value of errno? That should give you some idea as to what's going wrong. However, there's really only one common reason for setuid to fail: you don't have privileges. My guess that this isn't a problem with the setuid call, but with the infrastructure that you're using to run privileged code. Change your code to print the results from <x-man-page://2/getuid> and <x-man-page://2/geteuid> and see what you get. This email sent to site_archiver(a)lists.apple.com

1 0

Regarding setuid Problem on Mac OS X 10.2
by subodh singh 17 May '07

17 May '07

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Dkim-signature: a=rsa-sha1; c=relaxed/relaxed; d=gmail.com; s=beta; h=domainkey-signature:received:received:message-id:date:from:to:subject:mime-version:content-type:content-transfer-encoding:content-disposition; b=C6zztDD150VKCDlVmt3fAHyM28wOieUqDNu/6XNCS5lJb0fwzUHPjfyvfjvTAzpPno1fGIfvuz0DsxQ0aFLu0zXKwlc/ev+wF7m0fujon0x5emUILu+l6E17RHaaLF0Zaa1SBgvJpGb95IJXfgK+DHZgs7fhuVEv4/tBpJMg/aA= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=beta; h=received:message-id:date:from:to:subject:mime-version:content-type:content-transfer-encoding:content-disposition; b=iMvntqx91kkahYGfBGANgiLD1jeD/AgR1vPi1gmKqddCWpgits2h5Au3UEQ8BVkwdwYfaXSmr+PdGnq7labqkIq3o5PeZdGZC9f9XZVp17NZB3luA0M/68DC6RGW4eQvH0c7G7NeTzK6/75pw5c377TybZtk98f+1rI/hyFrkGc= Hello Guys, I got one problem on Macintosh Mac OS X 10.2.4, i.e setuid(uid) API is not working . Its return -1. The same code is working fine on High end but I got the problem on Mac OS X 10.2 series. So please advise and also suggest me some alternative way to achieve the same. Thanks & regards Subodh Singh _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/site_archiver%40lists.… This email sent to site_archiver(a)lists.apple.com

1 0

all connect(2) calls are failures?
by Todd Heberlein 17 May '07

17 May '07

site_archiver(a)lists.apple.com Delivered-To: Darwin-kernel(a)lists.apple.com A typical audit record is: The audit flags are set for "all". Thanks, Todd _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/site_archiver%40lists.… I am looking through Apple's BSM audit logs (Mac OS X 10.4.9 on a G5), and all records for the connect(2) system call to an IP address are reported as failures. There are no examples of successful connection requests even though I know they took place. header,88,1,connect(2),0,Tue May 15 09:44:38 2007, + 731 msec argument,1,0x10,fd socket-inet,2,80,17.250.248.77 subject,heberlei,heberlei,staff,heberlei,staff, 384,196,50331650,0.0.0.0 return,failure : Operation now in progress,4294967295 trailer,88 Is this a bug in Apple's kernel (i.e., inadvertently reporting a successful system call as a failure), or did Apple choose *not* to report successful network connections? This email sent to site_archiver(a)lists.apple.com

1 0

Re: devfs_make_node() bug in 9.0.0d8
by Quinn 16 May '07

16 May '07

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com <http://developer.apple.com/bugreporter/> Share and Enjoy [1] -- Quinn "The Eskimo!" <http://www.apple.com/developer/> Apple Developer Relations, Developer Technical Support, Core OS/Hardware [1] But be careful about the "share" part (-: _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/site_archiver%40lists.… At 15:28 +0800 16/5/07, Nick Blievers wrote: I noticed something while porting our driver to leopard today. Thanks for taking the time to test on Leopard. However, please don't post your results to a public mailing list. Leopard is still pre-release software and, as such, is covered by the NDA you agreed to when you joined Apple Developer Connection. Also, regardless of whether the software is released or pre-release, the best way to report problems is to file a bug using the page below. A whole bunch of Apple kernel engineers read this mailing list. I'm sure we'd all be happier if they spent their time reading questions and replying, rather than routing bugs between the mailing list and our bug database. This email sent to site_archiver(a)lists.apple.com

1 0