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

Re: getcontext() -> Bus error?
by Terry Lambert 14 Feb '08

14 Feb '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com -- Terry On Feb 14, 2008, at 10:17 AM, Edd Dawson <lists(a)mr-edd.co.uk> wrote: Is that bad? Does the storage type of the ucontext_t matter? Thanks, Edd _______________________________________________ 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… When you call getcontext/makecontext/setcontext, you are hacking the contents of your context. This includes register and other state, including the stack register. This makes sense, since these functions are typically used to implement user space threads schedulers. Then you go and depend on the contents of a value which is either stack relative, or which has been loaded into a volatile register, and expect the code to "do the right thing", even though you are saying something like "load a bunch of stack relative data off an offset of -256 off the stack" (or whatever), or you are saying "load a bunch of data relative the the address contained in register such-and-such I just overwrote". Either of these can now refer to totally different data, but the compiler didn't know about that as a barrier, so it uses the stale contents not knowing the contents of what it is using was changed out from under it. Either way you look at it, this stuff is going to break under those conditions. Minimally, you should be using a chunk of heap or global data, the address of which is contained in a variable that has been declared "volatile" so that (a) it will be passed to the functions in a volatile register, and (b) after the function returns, it doesn't try to pop data from the stack back into the register, particularly because your new stack doesn't have the saved data on it anyway, and even if it did, the data it wants to pop back only exists on the old stack, which you are no longer referencing. I think on other systems you are simply lucking out due to internal implementation, or your compiler is doing different optimizations by default. Using functions like these correctly will generally take a pretty fundamental understanding of the machine architecture, the assembly code that will be emitted by your compiler, and the relevant standards governing the functions behaviour. Terry Lambert wrote: This declaration is on the main program stack: ucontext_t ctx; it's not allocated on the heap (e.g. via malloc), and it's not global. This has worked for me on other systems so I'm interested to know about any darwin-specific (or general) coding faux-pas in this area. This email sent to site_archiver(a)lists.apple.com

1 0

Re: getcontext() -> Bus error?
by Edd Dawson 14 Feb '08

14 Feb '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com User-agent: Thunderbird 2.0.0.9 (Windows/20071031) This declaration is on the main program stack: ucontext_t ctx; Is that bad? Does the storage type of the ucontext_t matter? Thanks, Edd _______________________________________________ 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… Terry Lambert wrote: it's not allocated on the heap (e.g. via malloc), and it's not global. This has worked for me on other systems so I'm interested to know about any darwin-specific (or general) coding faux-pas in this area. This email sent to site_archiver(a)lists.apple.com

1 0

Re: Darwin 9.2 Sources
by Rosyna 14 Feb '08

14 Feb '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On Feb 13, 2008, at 2:17 PM, William Siegrist wrote: --- Sincerely, Rosyna Keller Technical Support/Carbon troll/Always needs a hug Unsanity: Unsane Tools for Insanely Great People It's either this, or imagining Phil Schiller in a thong. _______________________________________________ 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… It looks as though all traces of CFNetwork have been removed from the 10.5.x sources directories. We are already aware of the missing CF-476.10 project. We are working on those still. If any other project is missing, please email me. This email sent to site_archiver(a)lists.apple.com

1 0

Re: getcontext() -> Bus error?
by Terry Lambert 14 Feb '08

14 Feb '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com This declaration is on the main program stack: ucontext_t ctx; it's not allocated on the heap (e.g. via malloc), and it's not global. -- Terry Sorry? You lost me! Which stack variable? Kind regards, Edd _______________________________________________ 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 Feb 12, 2008, at 10:15 AM, Edd Dawson wrote: Terry Lambert wrote: Given your use of a stack variable, it makes all sorts if sense. 8-). This isn't an entirely pointless follow-up; if you can shed some light on why the code failed on Darwin but not on Linux, I'm interested to know more. All I thought I needed to do before calling makecontext() was ensure that I had a stack allocated and uc_link set appropriately? On Feb 11, 2008, at 11:10 AM, Edd Dawson <lists(a)mr-edd.co.uk> wrote: Edd Dawson wrote: Kevin Van Vechten wrote: On Feb 3, 2008, at 4:43 PM, Edd Dawson wrote: I hope this is my final question on the matter. The following code is failing on Leopard. The assert() in main() is triggered. The vector's size appears to be 3?! #define _XOPEN_SOURCE #include <ucontext.h> #include <signal.h> #include <cassert> #include <cstdio> #include <cstdlib> #include <vector> struct context { context(ucontext_t ctx, void (*func)()) : ctx_(ctx), func_(func), stack_(MINSIGSTKSZ) { ctx_.uc_stack.ss_sp = &stack_.front(); ctx_.uc_stack.ss_size = stack_.size(); ctx_.uc_link = 0; makecontext(&ctx_, func_, 0); } context(ucontext_t ctx) : ctx_(ctx), func_(0), stack_() { } static void change(context &from, context &to) { swapcontext(&from.ctx_, &to.ctx_); } ucontext_t ctx_; void (*func_)(); std::vector<char> stack_; }; std::vector<context *> all; void die(const char *msg) { perror(msg); exit(666); } void hello() { puts("hello"); assert(all.size() == 2); context::change(*all[1], *all[0]); } int main() { ucontext_t ctx; if (getcontext(&ctx) == -1) die("getcontext failed"); all.push_back(new context(ctx)); if (getcontext(&ctx) == -1) die("getcontext failed"); all.push_back(new context(ctx, &hello)); assert(all.size() == 2); context::change(*all[0], *all[1]); return 0; } This all works fine when I move getcontext() in to the same function as the associated makecontext(), which makes sense from a certain point of view. This email sent to site_archiver(a)lists.apple.com

1 0

Scheduling class for a process
by Stan Morris 13 Feb '08

13 Feb '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:mime-version:content-type; bh=KdZkKwi4kGt6RpW212JRjlDOnZNwUUC67zqu7IUq4Io=; b=Pl0RC8CSlOYri9Iy7ABadoVdlBOq2t9kQsSdxyidoD4Z+urVNyQeidbYR8ndBjIWeDeURamltOOeZgT20C91IqPtk0FN1Kleg90Nrn4r14Afv3lft7jF4iL9TsUc4qxp4MSi3DmTRvQHD7OFYVsh1TYRpRU8/draADkeEa7cY24= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:date:from:to:subject:mime-version:content-type; b=GA2LY0wIyeccz8Boe9zO0G9oaicWwFc1CpKnEGb2ZGw2KFEHlkM7WlIxl+3RT5RkQqPuDXlGg0cclwcFPp/FHdlmcsB2MBDswd6I3G/ewARYBakVGF9C74nP6qsoP44q+ZrTLQ0K8322v/bM+47n9KbJm5jnV5GN+QvNNmzNXjA= _______________________________________________ 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: Darwin 9.2 Sources
by Joe Ranieri 13 Feb '08

13 Feb '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com http://www.opensource.apple.com/darwinsource/10.5.2/ Joe Ranieri Lead Magician, Alacatia Labs, Inc. http://alacatia.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… The Darwin team is pleased to announce the availability of the source code for Darwin 9.2. These sources correspond to Mac OS X 10.5.2. The sources are available at: We are already aware of the missing CF-476.10 project. We are working on those still. If any other project is missing, please email me. Thanks for getting these up so quickly. The sources for dyld seem to be missing though... This email sent to site_archiver(a)lists.apple.com

1 0

Darwin 9.2 Sources
by William Siegrist 13 Feb '08

13 Feb '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com http://www.opensource.apple.com/darwinsource/10.5.2/ Thank you. -Bill ---- William Siegrist Software Support Engineer Mac OS Forge http://macosforge.org/ wsiegrist(a)apple.com 408 862 7337 _______________________________________________ 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 The Darwin team is pleased to announce the availability of the source code for Darwin 9.2. These sources correspond to Mac OS X 10.5.2. The sources are available at: We are already aware of the missing CF-476.10 project. We are working on those still. If any other project is missing, please email me. smime.p7s

1 0

Re: Trouble updating a complex software solution
by Graham J Lee 13 Feb '08

13 Feb '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Hi Sorin, On 11 Feb 2008, at 16:41, Sorin Negoita wrote: Hello, Cheers, Graham. _______________________________________________ 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 have a software solution formed of 3 kexts, 3 launchd daemons, 3 agents and 1 IPC mechanism used for communication between daemons and agents. I can't find a reliable way of updating this solution. Sophos Anti-Virus for Mac has a similar arrangement of components with interdependent interfaces. The approach we use for updating is to deliver updates as a metapackage, and use the installer postflight script to ensure that everything is restarted in proper sequence. This email sent to site_archiver(a)lists.apple.com

1 0

Re: what can i get from io_registry_entry_t for a hard disk
by Chris Sarcone 13 Feb '08

13 Feb '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

what can i get from io_registry_entry_t for a hard disk
by Nick Rogers 13 Feb '08

13 Feb '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