January 2009 - Darwin-kernel

Re: Question regarding vnodes and memory mapped files
by Michael Smith 21 Jan '09

21 Jan '09

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.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.… This email sent to site_archiver(a)lists.apple.com

1 0

Re: Question regarding vnodes and memory mapped files
by Quinn 21 Jan '09

21 Jan '09

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com 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 14:51 +0530 21/1/09, Narayan Subramanian wrote: My question here is: Do we not call close() on the executable files descriptor after we mmap() them into the process address space (inside dyld)? I am assuming that the vnode corresponding to the file descriptor should get freed after the file descriptor gets closed (right?). (I mean isn't that the only way we can move around an (on disk) executable file that being run by another process?). If that is true, how is it that we are still able to trace the vnode that represents the file that was mmap()ed, after its descriptor was close()d? Who caches this vnode, how and where? If this vnode is still 'alive' how is one able to physically move the executable file (mapped by the process) on disk? The file descriptor may or may not have been closed; that's irrelevant because the vnode pager is holding a reference to the vnode. The fact that the vnode has active references does not prevent you from moving the image within the volume or, for that matter, deleting the image off the volume. The vnode pager goes through VFS, and the VFS plug-ins are expected to deal with this sort of thing. This email sent to site_archiver(a)lists.apple.com

1 0

Question regarding vnodes and memory mapped files
by Narayan Subramanian 21 Jan '09

21 Jan '09

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:date:message-id:subject :from:to:content-type; bh=C53vx89mexek3mFNoXMHNFkj61Voa0N9NEnI17NiQaQ=; b=F3w9EDhO+PUErjr/lQcPv2BXhTmzaDNfP1lUGimHJMI5X97goPWLrCqRWDEUwXvncL InL/uhyym6Qy5cR8tEYQLNOpsjp7NhHwZCUdzVV7N7ssmHY2giyjS3IBOkMUPQuF14qq xYT1+dai9/mp1FN7+EjMtIvoVY2LFvEzKmU44= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:date:message-id:subject:from:to:content-type; b=ZKHVYNZzTppIRAufdQDrc0ni7n+T2O8u1AtGecBTg+QHC9ah4elo2fdOANCx4cNbhg GK66Ft702PKCN9M5wDkpFIzOWLlkhPm1iVCsaaeguXQdWoAW/GCJUYI7qRItKdSarkas Okr1ZEemxrIs4r+cQJRs7DxDoK5gVHC8wLg6k= _______________________________________________ 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: KAUTH_FILEOP_EXEC & KAUTH_VNODE_EXECUTE freeze
by Terry Lambert 21 Jan '09

21 Jan '09

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.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.… This email sent to site_archiver(a)lists.apple.com

1 0

Re: KAUTH_FILEOP_EXEC & KAUTH_VNODE_EXECUTE freeze
by Liviu Andron 21 Jan '09

21 Jan '09

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:in-reply-to:references :date:message-id:subject:from:to:content-type; bh=9EzUkh2FTRj5G5uNDSQyCbQcXl6BU1MPDHVgpkIXuHs=; b=Az6g4uPZ/7CRUu6zTQZF/zrPbc9xykPgpZZoeJ0FmR08AawTwKng5bd1sTPpsDBffg kXnL9uP+ranuVGt21syOX/T8HmuEY0yVd5GgMSGlW9SaVbA0hNObi8OkPoBbCl288p+c cN7KQRmOJVw4aeFHMruhKxVTIAkrvpTOcfxPs= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :content-type; b=mcyyABwpvBhNfuEkJbwJ+20lYHyNaF+9QJ4fTEFoUfzdPPPYXewUIcUUNrmgHRK1zn 7UIEDBeUyO1s78hNkAK8ljVyutaFF67anMccaJKa+UD8WPRIdcBu8ic9ARpW0wPXG2Hc 9gzUy+J3jGkmLiOW6ctrTxqPehEPyZvK+ZjVk= -- Terry On Jan 19, 2009, at 2:17 AM, Quinn <email@hidden> wrote: Please file these as bugs against the sample code. <http://developer.apple.com/bugreporter/> _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (email@hidden) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/email@hidden _______________________________________________ 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 Warning: ERESTART is intended to be propagated up to the user/kernel boundary to cause the system call to be redriven from user space buy the system call stub. We intentionally use this to unwind state that we can't otherwise unwind. If you intercept this at the kernel level you will potentially break things related to locking. At 5:21 +0200 17/1/09, Liviu Andron wrote: 2) msleep related question: "Callers of this routine must be prepared for premature return, and check that the reason for sleeping has gone away" In the sample above, I can check if the reason for return is the user space answer, but how can I see if it is the timeout exit ? Is EWOULDBLOCK ok ? Yes. Also be prepared for EINTR and ERESTART, which both indicate that the thread was woken up in order to process a signal. 3) KauthORama sample has 2 problems - debug configuration panic on the line: "assert (*actionStrPtr != NULL)" - it doesn't know about KAUTH_FILEOP_DELETE 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 (email@hidden) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/email@hidden This email sent to email@hidden This email sent to email@hidden

1 0

Re: KAUTH_FILEOP_EXEC & KAUTH_VNODE_EXECUTE freeze
by Liviu Andron 20 Jan '09

20 Jan '09

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:in-reply-to:references :date:message-id:subject:from:to:content-type; bh=E24BuVLG24nSK1nAdeR0+iFQ1yjTq6lWc9e7mZNwgcA=; b=JkqZ+7sKCXyLxHdnEXZ6N9bJVU6ILai/xYPh3tT+s0ml+iqA6MJFBYhwY/8nkDhQvt BkC2iY4gr+SphFT4chTn+sXVVdi4jmebMuje7NKAAlsFc7NnqABIVV5HKzHuXlwZXM0J BXqRUL5Q1QpSxUyTJ/wKFKjs03wzT/hx280S4= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :content-type; b=DyH+6vZTuKDPf+98+vkzEAVoJyiFKdKMBbe+jKKhnxxcXCEE+X/ZVKHOYzH5Sb+zws QV6UX0XRjHZ/RPk5RW7/gOJVEXwz4L7WesxW1dGlaLl9CO994FWNJWsfm1MD4QNTkXHq pL/KNfQKrDH+k9pr6UHTDXFNShkFSv0+9EgU0= _______________________________________________ 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: KAUTH_FILEOP_EXEC & KAUTH_VNODE_EXECUTE freeze
by Terry Lambert 19 Jan '09

19 Jan '09

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com -- Terry On Jan 19, 2009, at 2:17 AM, Quinn <eskimo1(a)apple.com> wrote: Please file these as bugs against the sample code. <http://developer.apple.com/bugreporter/> _______________________________________________ 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.… Warning: ERESTART is intended to be propagated up to the user/kernel boundary to cause the system call to be redriven from user space buy the system call stub. We intentionally use this to unwind state that we can't otherwise unwind. If you intercept this at the kernel level you will potentially break things related to locking. At 5:21 +0200 17/1/09, Liviu Andron wrote: 2) msleep related question: "Callers of this routine must be prepared for premature return, and check that the reason for sleeping has gone away" In the sample above, I can check if the reason for return is the user space answer, but how can I see if it is the timeout exit ? Is EWOULDBLOCK ok ? Yes. Also be prepared for EINTR and ERESTART, which both indicate that the thread was woken up in order to process a signal. 3) KauthORama sample has 2 problems - debug configuration panic on the line: "assert (*actionStrPtr != NULL)" - it doesn't know about KAUTH_FILEOP_DELETE 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/tlambert%40apple.com This email sent to tlambert(a)apple.com This email sent to site_archiver(a)lists.apple.com

1 0

Re: Re: How to contitnule this kind of kernel debug -----Does large memory affects?
by mm w 19 Jan '09

19 Jan '09

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:in-reply-to:references :date:message-id:subject:from:to:cc:content-type :content-transfer-encoding; bh=zk8UeFGW8DzI02teUDrxzNQBKSw+ipJolDm9ZwYXhBA=; b=NH4pXP4Gg2vEbPEaY1AUhyvcadt1rhtY9W3mT9oWGSEKdbN7Cj+qP4OhfaDZ/EyJiM 2I/VWjj/E8VfSHYVwaiuJVlFl1U70x43XrOyb5XmouhHuQcJ5uWAILVrX9f4EBGpWugR iS+PPf5jIv3JrghUXDyyCT00nmlm0izfXjO/o= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:content-transfer-encoding; b=etfEa75s5FYpnfgVVhle+M2n5omAmB9ze+63nfpHjEBzfdbwXpJCYS4I1AepRcCwsd oBhwijZkD0WrBfv0V8mGic9+Oi1r2cPaYQFSlaxI3g5VMNlPSBhBGJLDkTtHkWn+Ncqq HjWnMiPzX/HuYkk7u5v+XwYr065R/rbjhvmNc= this is the same for 32-bit Windows this is not new it's just logic, the current state of macosx (leopard) is to be able to handle 64 bit instruction in a 32-bit mode Cheers! 2009/1/19 searockcliff <searockcliff(a)163.com>: > Do you mean this is a bug of the kernel of Mac OS X? > Do you have some links? > > 在2009-01-19，"Michael Smith" <drivers(a)mu.org> 写道： >> >>On Jan 18, 2009, at 7:28 PM, searockcliff wrote: >> >>> I find the line of code near the panic is memcpy(). >>> But the parameters are fine. >>> >>> One important thing: >>> The kernel panic only occurs on Mac Pro with memory larger than 2GB, >>> like 3GB, 4GB,9GB and so on. >>> If I reduce the memory used to 2GB by boot-args, the kernel panic >>> disappears. >>> >>> So my question is: >>> Can the Mac OS X kernel NOT handle large memory (>2GB) well? >> >>Ask yourself this simple question: >> >>Of all the folks writing code for MacOS X, are you *really* the only >>one that has discovered this mysterious flaw? >> >>> Or the memcpy() function fails sometime on large memory? >> >>See above. >> >> = Mike >> > > > ________________________________ > 《大话西游外传》贺岁新作，送豪宅、送你5000元压岁钱 > _______________________________________________ > 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/openspecies%40gmail.com > > This email sent to openspecies(a)gmail.com > -- -mmw _______________________________________________ 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: How to contitnule this kind of kernel debug -----Does large memory affects?
by William Kucharski 19 Jan '09

19 Jan '09

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com On Jan 18, 2009, at 8:28 PM, searockcliff wrote: I find the line of code near the panic is memcpy(). But the parameters are fine. So my question is: Can the Mac OS X kernel NOT handle large memory (>2GB) well? Or the memcpy() function fails sometime on large memory? It sounds suspiciously like you're having 32-bit/64-bit address issues. William Kucharski _______________________________________________ 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.… One important thing: The kernel panic only occurs on Mac Pro with memory larger than 2GB, like 3GB, 4GB,9GB and so on. If I reduce the memory used to 2GB by boot-args, the kernel panic disappears. What are the addresses you are copying between, and how much are you trying to copy at one time? This email sent to site_archiver(a)lists.apple.com

1 0

Re: KAUTH_FILEOP_EXEC & KAUTH_VNODE_EXECUTE freeze
by Quinn 19 Jan '09

19 Jan '09

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Please file these as bugs against the sample code. <http://developer.apple.com/bugreporter/> 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 5:21 +0200 17/1/09, Liviu Andron wrote: 2) msleep related question: "Callers of this routine must be prepared for premature return, and check that the reason for sleeping has gone away" In the sample above, I can check if the reason for return is the user space answer, but how can I see if it is the timeout exit ? Is EWOULDBLOCK ok ? Yes. Also be prepared for EINTR and ERESTART, which both indicate that the thread was woken up in order to process a signal. 3) KauthORama sample has 2 problems - debug configuration panic on the line: "assert(*actionStrPtr != NULL)" - it doesn't know about KAUTH_FILEOP_DELETE This email sent to site_archiver(a)lists.apple.com

1 0