July 2011 - Darwin-kernel - lists.apple.com

Re: vnode_put(...): iocount < 1 revisited
by Ken Hornstein 28 Jul '11

28 Jul '11

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com >Thanks for your comments. I think it all makes sense. I took some time to >correlate your comments with the code. I agree that I needed the >vnode_recycle(). In fact, vnop_remove() in my driver behaves the same way. >I am still unsure about how the last iocount gets dropped. If there is no >open() again, the vnode never gets recycled? I will keep looking at the >code. If the vnode is in this situation (doesn't really exist except in local namecache), then the iocount acquired in cache_lookup() will be dropped in vn_auth_open() (see the vnode_put() at vfs_vnops.c:454, at least in my copy of vfs_vnops.c). If the open succeeds, then eventually the iocount will be dropped in open1() (vfs_syscalls.c). Of course if the open _does_ succeed then the vnode will still have a usecount. It occurs to me that you should actually also purge that vnode from the namecache as well when you call vnode_recycle() (even though when the vnode is recycled it will do that); the vnode might have a usecount and it can't be recycled until both the usecount and the iocount reach zero. If that happened then you'd still be stuck in that loop until the other owners of that vnode were done with it. --Ken _______________________________________________ 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 debug deadlocks in KEXT?
by Ken Hornstein 28 Jul '11

28 Jul '11

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com >> First, if you don't have it, you should get the Kernel Debug Kit >> corresponding to the kernel version you're using. For kernel >> versions up to 10.6.2, you can get them here: >> >> https://developer.apple.com/hardwaredrivers/download/kerneldebugkits.html > >That page is no longer being maintained. Kernel Debug Kits now live on the >Apple Developer Downloads site <https://connect.apple.com> under Hardware >& Drivers. Hate to be a whiner ... but didn't I say that in the rest of that message? And while we're on the subject .... if you search for "kernel debug kit" (even directly on Apple's web site) you get that unmaintained page. Of course it's Apple's prerogative to put the kernel debug kit whereever they want, but when a search leads to a web page at Apple that doesn't actually contain what you need, it just makes kernel development even harder than it already is :-/ If someone could put a link on that page saying "for Kernel Debug Kits for newer versions of MacOS X you need to go to here", that would be greatly appreciated. --Ken _______________________________________________ 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: vnode_put(...): iocount < 1 revisited
by mogambo 28 Jul '11

28 Jul '11

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=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=SUhUFjoPFlwzRrYcoXFELW9hE3cNAuSnYAKN7bF3cx4=; b=Mpr2Kx7cClpavJJCLN7UkFOstRCs3xbEmARLcc+26WcN3+YW2I8YLQ50y40kYcYQEe MkFucnVYWXinOpKlE/N7KgdqFRNQeK0/2Cqx1Vq6KD2fAD9WwrKyTEwTLIZ9ODKpzA26 PhhUUND6ONF9FjBCOb+4Huf8oG2yHa4mvSrlM= _______________________________________________ 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: Lion panic debugging
by Kevin Brock 27 Jul '11

27 Jul '11

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: How to debug deadlocks in KEXT?
by Garth Cummings 27 Jul '11

27 Jul '11

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: Lion panic debugging
by Garth Cummings 27 Jul '11

27 Jul '11

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: Lion panic debugging
by Antoine Missout 27 Jul '11

27 Jul '11

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: Lion panic debugging
by Brian Bechtel 27 Jul '11

27 Jul '11

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=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:content-transfer-encoding; bh=oCc94szWAvmiy+cgbj/gkHCW+6Quf0f2MFNhmG6KT9M=; b=CXHwDhJZ7oFB2xFOFAaPTQ7zTugyFp4ApunFUCnrLjT7Zm0L1pnuxbLrNMQDurUZ4Z ehSnYhQDJmn6itsLtzTv0MSODx4QuvjJdEkUl3+465YG44z8krwrdC3t1imQWqyhWYCB DkICEIyLIfhS+Me92YWwnY0rQqovPLjfMOso0= On Wed, Jul 27, 2011 at 12:23 PM, Kevin Brock <apple(a)kevin.com> wrote: > On Jul 27, 2011, at 11:48 AM, Shantonu Sen wrote: > > There's not an address in the backtrace which resolves to anything at all. > > Please be more specific. Addresses in kexts wouldn't be in the static kernel > text, obviously. > > Obviously. > I'm not talking about kext addresses not resolving. There are always frames > in the backtrace which correspond to OS X panic handling code, or to the OS > functions that called the kext, or to the call that actually crashed. The > crash occurred on 11A511, and in the BT below, with the 11A511 > KernelDebugKit, none of the addresses as shown in the BT resolves to a known > symbol. I believe you aren't specifying "-arch i386" on your gdb command line. When I do so, it works for me. Downloading the kernel_debug_kit_10.7_11a511.dmg and mounting it, you get /Volumes/KernelDebugKit. I then did the following: $ gdb -arch i386 /Volumes/KernelDebugKit/mach_kernel GNU gdb 6.3.50-20050815 (Apple version gdb-1705) (Fri Jul 1 10:50:06 UTC 2011) Copyright 2004 Free Software Foundation, Inc. GDB is free software, covered by the GNU General Public License, and you are welcome to change it and/or distribute copies of it under certain conditions. Type "show copying" to see the conditions. There is absolutely no warranty for GDB. Type "show warranty" for details. This GDB was configured as "x86_64-apple-darwin"... (gdb) source /Volumes/KernelDebugKit/kgmacros Loading Kernel GDB Macros package. Type "help kgm" for more info. (gdb) x/i 0x22032e 0x22032e <panic+510>: mov 0x85fab4,%eax (gdb) x/i 0x24480e 0x24480e <zalloc_canblock+1198>: mov %edi,%esi (gdb) x/i 0x245069 0x245069 <zalloc+25>: add $0x8,%esp (gdb) x/i 0x27e164 0x27e164 <_vm_map_entry_create+30>: mov %eax,%ebx (gdb) x/i 0x23837c 0x23837c <task_create_internal+60>: jmp 0x2383c9 <task_create_internal+137> (gdb) x/i 0x5645bf 0x5645bf <fork_create_child+47>: test %eax,%eax (gdb) x/i 0x5646e9 0x5646e9 <cloneproc+73>: test %eax,%eax (gdb) x/i 0x564b06 0x564b06 <fork1+662>: test %eax,%eax (gdb) x/i 0x564c4d 0x564c4d <fork+45>: test %eax,%eax (gdb) x/i 0x5f015a 0x5f015a <unix_syscall64+554>: cmpl $0x0,0x114(%esi) (gdb) x/i 0x2e3977 0x2e3977 <lo64_unix_scall+23>: mov %esi,%esp (gdb) However, in panics such as these, the panic message "zalloc: \"non-kernel map entries\" tells you that something is over allocating memory. The non-kernel map entries zone was the zone where the overallocation was first noticed, but it almost never is the zone where the overallocation is happening. Later in the panic log, you'll have a list of zones with more than 1 million allocations, and a backtrace of the suggested leaking code (a new feature in Lion.) That backtrace should be examined, rather than the backtrace of the panic. In this kind of panic, the panic backtrace just gives you information about the code path which noticed that we ran out of allocation space, rather than the code which causes the leak. Good luck. _______________________________________________ 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: Lion panic debugging
by Andreas Fink 27 Jul '11

27 Jul '11

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: Lion panic debugging
by Kevin Brock 27 Jul '11

27 Jul '11

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