June 2006 - Darwin-dev - lists.apple.com

i386_signal_cpu
by Roberto Cruz 15 Jun '06

15 Jun '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Any success to use this function ? This will solve a lot of issues for me. Regards Cryptonome _______________________________________________ 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: launch agent questions
by Dave Zarzycki 15 Jun '06

15 Jun '06

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: DirectoryServices-258: dsDataListInsertAfter() logic bug
by Brian Bergstrand 15 Jun '06

15 Jun '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Jun 15, 2006, at 2:45 PM, Brian Bergstrand wrote: This was fixed in one of the Sec updates. Sorry about that. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.3 (Darwin) iD8DBQFEkc0ZedHYW7bHtqIRAkF4AJwMB/JRuzjUHgcP6oIBP0dlXeRwSwCg8oj/ ON6C7wQJ6mmY5yDNQ7wzrng= =Xiux -----END PGP SIGNATURE----- _______________________________________________ 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… DS-255 will return an error if the allocation succeeds. So, is there any workaround other than duplicating dsAllocListNodeFromBuffPriv() (and its support functions) in my plugin. I will file a bug (but obviously it's a known one as it was fixed in 10.4 - but has not been fixed in any 10.3.9 Sec Update). However, I need a workable solution in the present. Note to self: Always make sure test machines have the latest updates applied. This email sent to site_archiver(a)lists.apple.com

1 0

Re: darwinbuild for 10.4.6 PPC
by Kevin Van Vechten 15 Jun '06

15 Jun '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com - Kevin On Jun 13, 2006, at 11:25 AM, Peter Lovell wrote: Regards.....Peter On Jun 13, 2006, at 2:20 PM, Anton Kuzmin wrote: Anton _______________________________________________ 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… The missing 8I127.plist is an oversight. I'll try to get it posted soon. That said, xnu is pretty self-contained, so it should work fine to initialize with an earlier Tiger build and then build xnu-729.6.70 within that BuildRoot. Thanks. I was trying to do it using darwinbuild, which makes the whole build-process simple. But the latest plist I see is for 10.4.5, not 10.4.6. Yep, did it yesterday, just follow http://developer.apple.com/documentation/Darwin/Conceptual/ KernelProgramming/build/chapter_18_section_3.html It's pretty much accurate, except that you may have to copy some headers into appropriate places by hand to build kextsymboltool. Peter Lovell wrote: Hi all, has anyone been successful building darwin xnu-792.6.70 (a.k.a. 10.4.6) for PPC ? There doesn't seem to be the necessary plist file 8I127.plist for this version, or at least I was unable to find it. Has anyone else built it? Note - this is for the PPC version, not for Intel. Thanks.....Peter _______________________________________________ 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/kevin% 40opendarwin.org This email sent to kevin(a)opendarwin.org This email sent to site_archiver(a)lists.apple.com

1 0

DirectoryServices-258: dsDataListInsertAfter() logic bug
by Brian Bergstrand 15 Jun '06

15 Jun '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 There's a logic bug in dsDataListInsertAfter() on 10.3.9: DirServicesUtils.cpp:1099 As opposed to DirectoryServices-352.1 (10.4.x) Thanks. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.3 (Darwin) iD8DBQFEkbjRedHYW7bHtqIRAuf4AJ4h48XZ0zTIxDHWgEFP85K09ZxLsgCfV8hP BZrtLDyJBEUUGwoxWjTcAiw= =UrRX -----END PGP SIGNATURE----- _______________________________________________ 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… pNewNode = ::dsAllocListNodeFromBuffPriv( inDataNode->fBufferData, inDataNode->fBufferLength ); if ( pNewNode != nil ) DirServicesUtils.cpp:1161 pNewNode = ::dsAllocListNodeFromBuffPriv( inDataNode->fBufferData, inDataNode->fBufferLength ); if ( pNewNode == nil ) DS-255 will return an error if the allocation succeeds. So, is there any workaround other than duplicating dsAllocListNodeFromBuffPriv() (and its support functions) in my plugin. I will file a bug (but obviously it's a known one as it was fixed in 10.4 - but has not been fixed in any 10.3.9 Sec Update). However, I need a workable solution in the present. This email sent to site_archiver(a)lists.apple.com

1 0

Re: /dev/cu.* vs /dev/tty.* ?
by Justin C. Walker 15 Jun '06

15 Jun '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On Jun 15, 2006, at 12:26 , Andre-John Mas wrote: Hi, I don't know that this helps you much :-} Cheers, Justin -- Justin C. Walker, Curmudgeon-At-Large Institute for the Absorption of Federal Funds -------- Some people have a mental horizon of radius zero, and call it their point of view. -- David Hilbert -------- _______________________________________________ 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 am using the Java Communications API, implemented at rxtx.org and whe I call the API it lists both the /dev/cu and /dev/tty devices. I am not sure that both need to be listed, especially if this is presented to the user. Some people feel both should be listed, others like myself don't think so. I hoping someone who is more familiar with the inner workings can tell me what the difference is between the two, and more specifically which one a serial communication tool should use. Typically, the tty* and cu* are different versions of the same "physical" thing: the tty's are for incoming (to the system) calls; the cu's are for outgoing (from the system) calls ("cu" == "call unix"; don't ask :-}). You can match them with their major/minor pairs, noting that the cu devices have odd minors, while the tty devices have even minors ('OR'ing a low-order bits gets you the "cu" for a corresponding "tty"). This email sent to site_archiver(a)lists.apple.com

1 0

/dev/cu.* vs /dev/tty.* ?
by Andre-John Mas 15 Jun '06

15 Jun '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Hi, I am using the Java Communications API, implemented at rxtx.org and whe I call the API it lists both the /dev/cu and /dev/tty devices. I am not sure that both need to be listed, especially if this is presented to the user. Some people feel both should be listed, others like myself don't think so. I hoping someone who is more familiar with the inner workings can tell me what the difference is between the two, and more specifically which one a serial communication tool should use. 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… This email sent to site_archiver(a)lists.apple.com

1 0

ATAIdentify: was Re: Darwin-dev Digest, Vol 3, Issue 158
by Peter Cassidy 15 Jun '06

15 Jun '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com I'd like to retrieve the system volume's hard disk serial number. For drives which indeed support such a serial number it can be read using the ioreg tool. Although I'd prefer to read the hard disk serial number using an API (e.g. IOKit) instead of using a dedicated tool, I'm more concerned about the different paths I seem to have to go through the dictionaries to get different drives' serial numbers when the drives are attached by different busses (see ioreg output below). >From ATASMARTLib.h ; Regards, Pete C _______________________________________________ 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 15 Jun 2006, at 20:02, darwin-dev-request(a)lists.apple.com wrote; It's part of the SMART API. I used it when hacking SMARTMonTools, some years back. Since (I think) MacOS X.3, the API to do an ATAIdentify has been in there. You just need to create a plug-in interface to kIOATASMARTUserClientTypeID, query the plug-in to get a handle to the interface. Then send the Identify command. /*! @function GetATAIdentifyData @abstract Reads the 512-byte data provided by the drive in response to the ATA IDENTIFY DEVICE command. @discussion Reads the 512-byte data provided by the drive in response to the ATA IDENTIFY DEVICE command. See section 8.15 of ATA/ATAPI-6. The data placed in buffer is guaranteed to be in native endian form on return. (i.e. it will be byte swapped on big endian platforms, so the caller need not do anything) @param interface A valid IOATASMARTInterface**. @param buffer A valid buffer. @param inSize The number of bytes to place in the buffer. @param outSize The number of bytes placed in the buffer. Can be NULL if the information is not required by the caller. @return An IOReturn result code. If inSize is greater than 512 or less than 1, kIOReturnBadArgument is returned. */ IOReturn ( *GetATAIdentifyData ) ( void * interface, void * buffer, UInt32 inSize, UInt32 * outSize ); I have some working dummy-code which does just this. Let me know if you want it. It's ancient but I built it on 10.4.6/ppc and it's fine. Here's a chunk of the Ident, including the s/n; GetATAIdentifyData myErr = 0 0x00000000 - 04 5A 3F FF C8 37 00 10 00 00 00 00 00 3F 00 00 .Z?.. 7.......?.. 0x00000010 - 00 00 00 00 20 20 20 20 20 20 4D 52 47 34 32 36 .... MRG426 0x00000020 - 4B 34 48 35 42 4A 47 48 00 03 3D 98 00 04 4D 4F K4H5BJGH..=...MO 0x00000030 - 34 41 41 44 37 41 48 69 74 61 63 68 69 20 49 43 4AAD7AHitachi IC 0x00000040 - 32 35 4E 30 38 30 41 54 4D 52 30 34 2D 30 20 20 25N080ATMR04-0 This email sent to site_archiver(a)lists.apple.com

1 0

Re: cpufrequency
by Howard Oakley 15 Jun '06

15 Jun '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Thread-index: AcaQoSh8Zz+QsPyUEdq1MgAKlZmxFg== Thread-topic: cpufrequency User-agent: Microsoft-Entourage/11.2.4.060510 On 15/6/06 11:54, Dirk Schelfhout wrote: > You can use google to find more. But 2 quick pastes. OK - I am taking this off-list now, to chew the fat over the literature. If anyone other than Dirk wants to be cced into our discussion, email me and I'll add you. I don't intend posting a concluding summary to the list unless we come up with anything earth-shattering. Howard. Dr Howard Oakley The Works columnist for MacUser magazine (UK) http://www.macuser.co.uk/ http://www.howardoakley.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: Retrieving hard disk serial number
by Chris Zubrzycki 15 Jun '06

15 Jun '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Jun 15, 2006, at 11:10 AM, Dirk Stegemann wrote: Hello, (originally posted at carbon-dev) - -chris zubrzycki - - -- PGP public key: http://homepage.mac.com/beren/publickey.txt ID: 0xA2ABC070 Fingerprint: 26B0 BA6B A409 FA83 42B3 1688 FBF9 8232 A2AB C070 ======================================================== -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.3 (Darwin) Comment: Please sign reply-http://www.gnupg.org iEYEARECAAYFAkSRfdAACgkQ+/mCMqKrwHAWmACeNZFpnjcAjkLMTbYtjxh6V6F3 ZMoAoLMMJa5gqDscFVZR87v8GDs6lzYj =M1Cq -----END PGP SIGNATURE----- _______________________________________________ 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'd like to retrieve the system volume's hard disk serial number. For drives which indeed support such a serial number it can be read using the ioreg tool. Although I'd prefer to read the hard disk serial number using an API (e.g. IOKit) instead of using a dedicated tool, I'm more concerned about the different paths I seem to have to go through the dictionaries to get different drives' serial numbers when the drives are attached by different busses (see ioreg output below). Also, the *keys* used to query the dictionary for the S/N differ, e.g. "device serial" vs. "Serial Number". Apple's System Profiler gets them all, but maybe it also uses different code to read the S/N from different devices? Is there a recommended / unified way to retrieve a hard drive's serial number? At least, are there known pitfalls? Thanks for pointers, links, sample code snippets, comments, or whatever you might consider useful! Try looking at smartmontools for some examples on how to query different drives. Unix _IS_ user friendly... It's just selective about who its friends are. This email sent to site_archiver(a)lists.apple.com

1 0