March 2006 - Darwin-kernel - lists.apple.com

Mapping logical to physical memory addresses
by Tony Scaminaci 05 Mar '06

05 Mar '06

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Thanks for any help, Tony _______________________________________________ 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.… Is there a system command or kernel call that returns the physical address corresponding to the logical address allocated by malloc? In my code, I first do a malloc for all of the free physical ram followed by mlock. Because I'm only allocating from the free physical ram pool, all allocated memory resides in ram (which it must for a memory test application). However, the start address returned by malloc is a local address which I can't figure out how to map to the physical DIMM. If I run two instances of the program at the same time and allocate half the free ram to each instance, both instances have the same local start address. How does one figure out the physical address corresponding to these local addresses returned by malloc? Also, I'm under the impression that the memory manager doesn't run in single-user mode. Does that mean that these local addresses are really physical addresses in single-user mode? This email sent to site_archiver(a)lists.apple.com

1 0

Re: "gdb -arch" problem
by Garth Cummings 03 Mar '06

03 Mar '06

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: "gdb -arch" problem
by Terry Lambert 03 Mar '06

03 Mar '06

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com On Mar 3, 2006, at 11:13 AM, Garth Cummings wrote: Hi Pavan, On Mar 3, 2006, at 3:37 AM, Pavan Chebbi wrote: I was able to do all the steps correctly till I type gdb -arch ppc /mach_kernel The current (Xcode 2.2.1) version of gdb is: Do not use Xcode < 2.2 to develop Intel applications. -- 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.… It now says that "-arch" is unrecognized option. What version of the Xcode tools do you have installed? You need at least Xcode 2.1 installed to have a version of gdb that supports the -arch option. $ gdb -arch ppc /mach_kernel GNU gdb 6.1-20040303 (Apple version gdb-434) (Wed Nov 2 17:28:16 GMT 2005) This email sent to site_archiver(a)lists.apple.com

1 0

Re: "gdb -arch" problem
by Garth Cummings 03 Mar '06

03 Mar '06

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

"gdb -arch" problem
by Pavan Chebbi 03 Mar '06

03 Mar '06

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Domainkey-signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com; h=Message-ID:Received:Date:From:Subject:To:MIME-Version:Content-Type:Content-Transfer-Encoding; b=xL1Us4XAgAfgBxYL442LJEVqmey/eb+LNYx4zFv4iVQzmPG1y3SakboyFxLNiXqeHYmYO/CeZlcFw0FV3tg6HqvoLRFiMSOAgxpgY8m3P8uUSTzZuVocN8ytjymrWyMSAugUYGOgFQIHr1wgjpU28M+fz2ZuUiwRM4WuNWwBdIk= ; _______________________________________________ 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

IGNORE: IOEventSource subclass won't unload?
by Brian Doyle 03 Mar '06

03 Mar '06

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Thanks! Brian On Mar 3, 2006, at 1:51 AM, Brian Doyle wrote: I'm perplexed :( Anyone have any ideas on why this might be happening? Thank you, Brian _______________________________________________ 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.… Hmm, it's late and I'm punchy I guess. I missed that one last little itty-bitty sentence in the IOEventSource docs that says I need to manage the retention of the object myself. Sorry. Apparently it seems adding the object to the workloop and immediately releasing it is bad form too. Would anyone care to tell me why I'm don't have to explicitly release IOCommandGate and IOTimerEventSource objects (i.e. the kext will unload if I don't), but with my custom object I do have to explicitly release it? I've created a custom event source subclassed from IOEventSource that as near as I can tell is correct. It is very simple and follows fairly closely the layout of IOCommandGate. When I call attachEventSource() it attaches, the IOWorkLoop calls checkForWork () and dispatches to the Action handler correctly, there's nothing special about initialization, it calls super::init() and behaves appropriately. After instantiation I attach it to the workloop (retain count==2) and then I release it and have verified that the retain count is 1 (i.e. retained only by the workloop). So it appears all is well and good, until I try to unload my kext, at which point it refuses saying that there is still an instance of my event source. I've attaching several other event sources to the same workloop (an IOCommandGate and an IOTimerEventSource) and I release them after they attach sucessfully (with a verifiable retain count of 1). The code is identical yet these unload just fine (presumably when the workloop is released). Commenting out just the add-it-to-the-workloop part allows everything to unload properly (just allocating and releasing my event source without ever attaching it to the workloop allows the kext to unload fine). So it just seems to cause a problem when it is attached into the workloop. _______________________________________________ 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/lists% 40balance-software.com This email sent to lists(a)balance-software.com This email sent to site_archiver(a)lists.apple.com

1 0

IOEventSource subclass won't unload?
by Brian Doyle 03 Mar '06

03 Mar '06

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com I'm perplexed :( Anyone have any ideas on why this might be happening? Thank you, Brian _______________________________________________ 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've created a custom event source subclassed from IOEventSource that as near as I can tell is correct. It is very simple and follows fairly closely the layout of IOCommandGate. When I call attachEventSource() it attaches, the IOWorkLoop calls checkForWork() and dispatches to the Action handler correctly, there's nothing special about initialization, it calls super::init() and behaves appropriately. After instantiation I attach it to the workloop (retain count==2) and then I release it and have verified that the retain count is 1 (i.e. retained only by the workloop). So it appears all is well and good, until I try to unload my kext, at which point it refuses saying that there is still an instance of my event source. I've attaching several other event sources to the same workloop (an IOCommandGate and an IOTimerEventSource) and I release them after they attach sucessfully (with a verifiable retain count of 1). The code is identical yet these unload just fine (presumably when the workloop is released). Commenting out just the add-it-to-the-workloop part allows everything to unload properly (just allocating and releasing my event source without ever attaching it to the workloop allows the kext to unload fine). So it just seems to cause a problem when it is attached into the workloop. This email sent to site_archiver(a)lists.apple.com

1 0

Re: LOCKLEAF is not used in Darwin8?
by Terry Lambert 03 Mar '06

03 Mar '06

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Regards ----------------------------------------- ----------------- SUPPORTS LOCKLEAF VFS_TBLFSNODELOCK VFS_TBLTHREADSAFE FS must honor ----------------- ----------------- ----------------- NO NO NO YES NO YES NO YES NO YES YES MAYBE ----------------- ----------------- ----------------- -- 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 Mar 1, 2006, at 8:57 PM, Lex Dejavu wrote: I've read throught bsd/vfs and find it's not checked by any function, although it's set somewhere. It's a hint to the FS to do something (or not) based on the FS's locking model. Locks are now generally asserted on the vp at the VFS layer, rather than below the VFS layer on the inode. We don't internally use it because for the FS's where we would have used it, we support thread reentrancy. For third party FS's that don't support thread reentrancy (i.e. neither VFS_TBLTHREADSAFE nor VFS_TBLFSNODELOCK is set), the locking is per fs, per entry into the VFS, unless that FS advertises that is supports node locking as it's model, which means the FS will take care of locking, and the VFS doesn't need to guard reentrancy on its behalf). If you support locking via inode/cnode/whatevernode locking, you are expected to honor LOCKLEAF, e.g.: The "MAYBE" is based on whether you require both vnode lock an inode lock held spearately or not. If you do, then you must ensure the lock ordering is the same each time you take a lock, and that everything works the same. I do not recommend this model at all. Primarily, this is to allow people to more easily port legacy code, but in general, your best bet is to support VFS_TBLTHREADSAFE and thread reentrancy, and let the VFS layer take care of managing the locking on a per vp reentrancy basis, instead of trying to do this yourself, or shave off some additional concurrency by going pure VFS_TBLFSNODELOCK, which is very hard unless you have a full event/ action graph model for your FS implementation. This email sent to site_archiver(a)lists.apple.com

1 0

Re: kernel sysctl
by Brian Bergstrand 02 Mar '06

02 Mar '06

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Mar 2, 2006, at 3:28 PM, Shaun Wexler wrote: Brian Bergstrand <http://www.bergstrand.org/brian/> PGP Key ID: 0xB6C7B6A2 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.1 (Darwin) iD8DBQFEB2baedHYW7bHtqIRAm+JAKD7Du73Spzv6VI0D82h2ZRO8giWVwCg9+a+ 1IXzsxAuYL6Dxi4XojX36kM= =T7Wm -----END PGP SIGNATURE----- _______________________________________________ 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.… >From a kext, how can I retrieve the sysctl "kern.osrevision" using valid KPI? It doesn't seem to be recognized by sysctlbyname()... You can't. Only certain "hw" sub-values are supported via the sysctlbyname() KPI. However, the version_* vars are exported by the LibKern kpi. Those should give you the data you need. $ nm -m /System/Library/Extensions/System.kext/PlugIns/Libkern.kext/ Libkern | grep version_ (undefined) external _version_major (undefined) external _version_minor (undefined) external _version_prerelease_level (undefined) external _version_revision (undefined) external _version_stage (undefined) external _version_variant This email sent to site_archiver(a)lists.apple.com

1 0

kernel sysctl
by Shaun Wexler 02 Mar '06

02 Mar '06

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.… >From a kext, how can I retrieve the sysctl "kern.osrevision" using valid KPI? It doesn't seem to be recognized by sysctlbyname()... -- Shaun Wexler MacFOH http://www.macfoh.com "No problem is so formidable that you can't walk away from it." - Charles Schulz This email sent to site_archiver(a)lists.apple.com

1 0