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

Finding the architectures of a binary
by Iceberg-Dev 10 Nov '08

10 Nov '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:mime-version :content-transfer-encoding:message-id:content-type:to:from:subject :date:x-mailer; bh=eI/nHdJ9nj20/JlANz7fGZLBsc/URLW91s8EVeCFYMM=; b=DUSEnTBftT8WyUqrbg1txZmbSkhjTioAcG/SuoIMZ4keCn2+Vq5EwbBzcKfbZbrSwh 6/DugffL6NTw0573uHVv0kstFSUonwVK8zJpd7F7jQn7W8cpfzLs1EJ5A8nh8yI35Ehj mUHsrAqV/qdxzRnyVnTGEmOMdAjsdc6OcEIhs= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:content-transfer-encoding:message-id:content-type:to :from:subject:date:x-mailer; b=KhfO/Y5Y8Kc0h72QWLnluP1ByHTZM+q0wU48SPWisE0SadLcxV79aNtPVkod1Anf9f rBIH2GACdZSvr7gGHDOXdYCSMkTC7E1t9kOzj0gapcJ/4lxQ2QehJUTg/czy7PI7OUKY ot3Kve7dA6OWCHehYDIo4QD/gw7CWoNdG+3yU= What I've found so far: What I don't know/understand: _______________________________________________ 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'm trying to find all the architectures available in a binary using code. - lipo tool can be used to do this but, apparently, its source code is not available in Darwin. - There is a new method in NSBundle in Mac OS X 10.5 and later to find the architectures of a bundle. This won't work with tools. - There can be at least 2 valid binary "formats": peff and mach-o (32- bit 64-bit). - It is apparently possible to get information about a mach-o binary using the dyld APIs and checking the mach_header (32 and 64) of the different images. - I'm not sure to understand how you can inspect a binary without loading it in memory. It looks as if the dyld API work only with the running binary (no parameter for instance to enumerate the number of images with _dyld_image_count();). - There doesn't seem to be an API for peff binaries. Is it safe to assume that the magic cookie is always at the beginning of the file? This email sent to site_archiver(a)lists.apple.com

1 0

Re: Disk Partition Offset?
by Kevin Elliott 09 Nov '08

09 Nov '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Is there a way to determine the sector offset on the disk of a volume (or partition) without actually parsing the partition table? -Kevin Elliott kelliott(a)mac.com http://web.mac.com/kelliott/ _______________________________________________ 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 Nov 7, 2008, at 9:07 PM, Duane Murphy wrote: I don't believe so. In the end, IOKit (specifically the IOPartitionScheme classes) is the final authority on this and the information is not simply exported from IOPartitionScheme drivers. Most (all?) of the PartitionScheme's hide free space gaps by not publishing IOMedia objects for them. And nothing in the overall design requires that IOMedia objects be created in the same order as they exist on disk. Obviously such a driver would be unusual... but that's not much to go on when it comes to mass storage! This email sent to site_archiver(a)lists.apple.com

1 0

Disk Partition Offset?
by Duane Murphy 08 Nov '08

08 Nov '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com DiskArbitration returns lots of useful information about disks and disk partitions. Very nice. statfs has other really useful information about the the file system on the disk. But I haven't found how to determine the sector offset of a partition/ volume on the disk? Is there a way to determine the sector offset on the disk of a volume (or partition) without actually parsing the partition table? Thanks for the ideas, ...Duane _______________________________________________ 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: Getting to the Symbol/String table
by Jean-Daniel Dupas 07 Nov '08

07 Nov '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Le 7 nov. 08 à 13:25, Rohitash a écrit : Hi, I want to be able to find a symbol given an address . I guess using atos (man atos for details) is not an option ? _______________________________________________ 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… For that, I would like to know how do I go to the begining of the symbol/string table present in the __LINKEDIT section. I can parse through the headers and reach to the LC_SYMTAB segment and the __LINKEDIT segment . From here on , how can I use the below values ( including the slide amount ) to get the base of the symbol table and the string table ( so that I can scan through the lists to find my symbol ) : symtab_command.symoff - An integer containing the byte offset from the start of the file to the location of the symbol table entries. symtab_command.stroff - An integer containing the byte offset from the start of the image to the location of the string table. My logic goes as below : 1 . Get the count of the images _dyld_image_count and loop through it and get the emage headers through _dyld_get_image_header() . 2 . Check for symtab/linkedit segments and store their addressess accordingly. A couple of more questions : 1. I wanted to know the actual reasoning for the slide amount which I get through _dyld_get_image_vmaddr_slide() . 2. dlsym() complains of _start/start as not a valid symbol even though I can see it from the o/p of the nm command. Thanks. This email sent to site_archiver(a)lists.apple.com

1 0

Getting to the Symbol/String table
by Rohitash 07 Nov '08

07 Nov '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Organization: Oracle Corporation User-agent: Thunderbird 2.0.0.17 (Windows/20080914) Thanks. -- Disclaimer : The statements and opinions expressed here are my own and do not necessarily represent those of Oracle Corporation. _______________________________________________ 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… Hi, I want to be able to find a symbol given an address . For that, I would like to know how do I go to the begining of the symbol/string table present in the __LINKEDIT section. I can parse through the headers and reach to the LC_SYMTAB segment and the __LINKEDIT segment . From here on , how can I use the below values ( including the slide amount ) to get the base of the symbol table and the string table ( so that I can scan through the lists to find my symbol ) : symtab_command.symoff - An integer containing the byte offset from the start of the file to the location of the symbol table entries. symtab_command.stroff - An integer containing the byte offset from the start of the image to the location of the string table. My logic goes as below : 1 . Get the count of the images _dyld_image_count and loop through it and get the emage headers through _dyld_get_image_header() . 2 . Check for symtab/linkedit segments and store their addressess accordingly. A couple of more questions : 1. I wanted to know the actual reasoning for the slide amount which I get through _dyld_get_image_vmaddr_slide() . 2. dlsym() complains of _start/start as not a valid symbol even though I can see it from the o/p of the nm command. This email sent to site_archiver(a)lists.apple.com

1 0

Weird problems with getutxent_wtmp()
by Greg Rumple 05 Nov '08

05 Nov '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

Two Machine Debugging Crashes in Ethernet Driver
by Duane Murphy 05 Nov '08

05 Nov '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com When I am two machine debugging over ethernet, I keep getting crashes in the ethernet driver com.apple.iokit.AppleYukon2, etc. on the target (the machine being debugged). Sometimes debugging can continue for several minutes; other times it crashes in a few seconds. The crashes usually happen when I am **not** doing any commands from gdb. I've been doing two machine debugging for a while, but I've not had to do debugging for several months. So this is the first time with Leopard 10.5.5. I've debugged with Leopard before, just not the most recent version. Is this something with this version of the OS (both sides are the same)? Is it this version of gdb (but the crash is on the target machine)? Am I forgetting to turn something off? The kernel panic reads: "simple lock deadlock detection: ..." AppleYokon2 is the first kext, then IOPCIFamily, IONetworkingFamily. The kext I am debugging has nothing to do with networking. Thanks for the help, ...Duane ps I can't debug with FireWire, because the gdb system is on an external FireWire driver (it's just easier that way). _______________________________________________ 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: Re: mstackrealign, update_prebinding and the misaligned_stack_error crashes
by Cristescu Alexandru 04 Nov '08

04 Nov '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=yahoo.com; h=X-YMail-OSG:Received:Date:From:Subject:To:In-Reply-To:MIME-Version:Content-Type:Content-Transfer-Encoding:Message-ID; b=4RAb7aEFub4/DFYM4YpwVaerju99tQ+pysOCbYAbTo6VP5BXXGxmQ16XGAolMseovQgmU/f0PqR6KzYL0jHhEllEPEdjcxYqc03EPVb2XBCcNdoPj3Ae7eOqYPrEKKOWnagq3M2ISSYyj1l+l04mLRvG/0Ib+YKsFSA8z8UbOdk=; _______________________________________________ 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 fix_prebinding.sh.zip

1 0

Re: Handling of events on serial port
by Terry Lambert 03 Nov '08

03 Nov '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: Handling of events on serial port
by Rakesh Singhal 03 Nov '08

03 Nov '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:message-id:date:from:to :subject:cc:in-reply-to:mime-version:content-type:references; bh=72kG5QIrltP12hIFOsOQlC4UDCVx7pjn0drKb6u/j6w=; b=cD/uKZA5B94sI6YUp50c3n2CiGhlgv/BARDtBciG6SlCxx7rpm0YWrCd7IRcwZhZa+ 5WUrGaxwZjywN/pnOuZvsrmLJjCLDCUKEpWfrSJuUcATbQk6BY4vcjQLqcMZiJDdxdf2 ay2mkTQSnYc434Ocit3vCeFFxLbgwFUquvYGA= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:date:from:to:subject:cc:in-reply-to:mime-version :content-type:references; b=ZvbMuRfkgo69I+dw/Xno9/LZ8qm0Fdv/fDtGvz1ziWBYdnamyDMeOIVggUYTk0kEys Eut80t/ZdJO42XMWcnjJKxDuNW8TFETpcBoP+mAE34LpszJ1rSBdOQ5DY+FJVm3QtJdF OpJJtrpSSQ0IEAlC7UDrNqRz2CGzmjVAgtJDM= _______________________________________________ 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