January 2009 - Darwin-dev - lists.apple.com

Re: Adding partition to expanded drive
by Andy Taylor 20 Jan '09

20 Jan '09

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=kyw21Rw3TP5uFggL0etIhUlpXyIcZKBXQBimHI7oNJ8=; b=msm72bRq1n0Y4A/E0xJ4n5lzHGWFF/HOuGEHh/i8fsTivifT22VOePMI3L93F6CNm1 GKmLVPJf/snTlQZrC2+3WSPVpDeBgf3/4/oLzvX8YrVRkHuZEpntTe/DPHU7DbMZsquJ 2mx6onwLMLK7ObqhW+UpXyw3S/KETIRvroYeQ= 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=BfjT5eB0ERz1whWFcjLG8RP5bzXi1I0S6X89RGBrfKSq1X9ocr2lZotu0CJPWp8LZ3 N+NK5AbhH64Zhv9WJKrLgCU68hTRtm5bYD0IH0A4BLKL+SdHtBAbrQq25ABxjJBIsus/ dd+MmszK7qAU5N1gSkEvUfV+heXliDHON9whA= _______________________________________________ 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_i386_exception_state, cr2 and faultvaddr
by Joel Reymont 19 Jan '09

19 Jan '09

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:cc:message-id:from:to :in-reply-to:content-type:content-transfer-encoding:mime-version :subject:date:references:x-mailer; bh=CkIFN6ON5txQXm1a5t3uCl2xoqmWlkIz5XTF6svUJNQ=; b=UfFO6HvJkCh7P1Jje77h/8PldDMhfwkoLmap5oVfptbirYJdvRYtau5oCwENYKF+VC PckQLCtK9VMYNuwZ3X7Ysomlfc30qXO9xUsl/c/m4HZgwG25uzdJHOmQToRuqLcyWTCU vO6C7NuOgV7ShQh+EuT1Rd5VFWaOj9VDxQ6v8= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=cc:message-id:from:to:in-reply-to:content-type :content-transfer-encoding:mime-version:subject:date:references :x-mailer; b=w1AaSd9MhOj8TVz+IDPupdIHvwHjbAIFjGtwM/+Gl/cg8iml3gXo6AyDCnXMteabgm 6G/aX67q8p014FkQ0rzgk0PG0gooPE5bwuoe7CpNVB+YE6q7zq8hSzFb0SEwJncuwOgV uB3e+1BioTMLrrRJu94RV1oMKRNjywA0T+cn4= Terry, On Jan 19, 2009, at 10:35 PM, Terry Lambert wrote: What is GP2? Is this the EIP in the saved registers or the faultvaddr? Do you mean getcontext (man ucontext)? According to the sigaction manpage, the handler has this interface void handler(int, siginfo_t *info, ucontext_t *uap); so the user context (thread state?) is given. What am I missing? Thanks, Joel --- http://twitter.com/wagerlabs _______________________________________________ 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 GP2 does not get the faulting address on i385 if it is in the 64 bit hole. Otherwise it is the faulting address. I need to get hold of the return EIP, I think, i.e. the address I will be going back to. Don't expect siginfo to not change. Unless you are a debugger, and rev every time week rev the os, expect your code to break. A call to getthreadstate is a better bet. This email sent to site_archiver(a)lists.apple.com

1 0

Re: __darwin_i386_exception_state, cr2 and faultvaddr
by Terry Lambert 19 Jan '09

19 Jan '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com -- Terry On Jan 19, 2009, at 2:16 PM, Joel Reymont <joelr1(a)gmail.com> wrote: What do these stand for and how are they used? The signal number is obvious but the return EIP is not. Thanks in advance, Joel --- --- http://twitter.com/wagerlabs _______________________________________________ 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/tlambert%40apple.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 GP2 does not get the faulting address on i385 if it is in the 64 bit hole. Otherwise it is the faulting address. Don't expect siginfo to not change. Unless you are a debugger, and rev every time week rev the os, expect your code to break. A call to getthreadstate is a better bet. I've been looking at /usr/include/mach/i386/_structs.h and xnu/bsd/ dev/i386/unix_signal.c (sendsig) for quite a while but I don't understand the purpose of faultvaddr and user_cr2. I have a hunch that this is the trap address but will it match EIP in saved registers? I need to grab the signal number and the return EIP in my signal handler. #if __DARWIN_UNIX03 #define _STRUCT_X86_EXCEPTION_STATE32 struct __darwin_i386_exception_state _STRUCT_X86_EXCEPTION_STATE32 { unsigned int __trapno; unsigned int __err; unsigned int __faultvaddr; }; #else /* !__DARWIN_UNIX03 */ This email sent to tlambert(a)apple.com This email sent to site_archiver(a)lists.apple.com

1 0

__darwin_i386_exception_state, cr2 and faultvaddr
by Joel Reymont 19 Jan '09

19 Jan '09

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:from:to :content-type:content-transfer-encoding:mime-version:subject:date :x-mailer; bh=qgx8MDb5Ca/rBrMPIrikoVLZi9LwpM/CdKtKQ4YBm9o=; b=NcIR8O4w+F7TjT/blKYXtj2U+M1OMza/N5/jGRUXPnRIonlbH+eYgPpPooyE+q4Uwj L13ejI8pzpD/v3J4hLkZ8OjYdFlBO1ksrPL70M3Lg3KXKPFa4dou4FdeW4oF60dUpPFe BOveVo2bJloNGErtmC0lKYidqtaxINQCaQzlA= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:from:to:content-type:content-transfer-encoding :mime-version:subject:date:x-mailer; b=SVcFLiKQU1Uilom4jvpk+fmUek7dngm3/p8wWGUXaGfAQENxFX5J+tQlLjhCc+VAjK 5bSTBHtcSlRIs7veIEAD/eAmy4vlUGfTLjqcbTgTPeJw+elnS7oOJJI8l9cBLOSe1cTo 9aQYY1FDRetSTGBZj6crUHxdc4Y5YHBukxQVM= What do these stand for and how are they used? The signal number is obvious but the return EIP is not. Thanks in advance, Joel --- --- http://twitter.com/wagerlabs _______________________________________________ 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've been looking at /usr/include/mach/i386/_structs.h and xnu/bsd/dev/ i386/unix_signal.c (sendsig) for quite a while but I don't understand the purpose of faultvaddr and user_cr2. I have a hunch that this is the trap address but will it match EIP in saved registers? I need to grab the signal number and the return EIP in my signal handler. #if __DARWIN_UNIX03 #define _STRUCT_X86_EXCEPTION_STATE32 struct __darwin_i386_exception_state _STRUCT_X86_EXCEPTION_STATE32 { unsigned int __trapno; unsigned int __err; unsigned int __faultvaddr; }; #else /* !__DARWIN_UNIX03 */ This email sent to site_archiver(a)lists.apple.com

1 0

Re: Why is virtual memory so high? (crash)
by Shawn Erickson 19 Jan '09

19 Jan '09

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:mime-version:received:in-reply-to:references :date:message-id:subject:from:to:cc:content-type :content-transfer-encoding; bh=In1cgRbDCHLlyDYkPdGEWgBJqlfjyKu9FuLvV+tp3xQ=; b=p5TheZeruPlyVOSYXJZ5eqgh+cyNeBBOuqyV8QfwlPySPUc9F/f+yaG03Fmuwb7hsT r7t4m/8aRoFEjrHl/CU4zGaKHRUprfl6D91Q4XCPce3EiUr4wq/T9vYcZ4tBIMJ6NnVj xnvqYfuycVcuuiDCq5w0N5JWC7VvdkQp142LU= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:content-transfer-encoding; b=dt/NqLcnKJ7fvks3WN3YWRThjG/pqJg3+F2S/R9rOBAdEYZF5lkC4GVM35IZjlgFwY z831GjCaNtWKZtJLOdQrDTidh+A/KruLFjbSvFvlpdNezDXqyMYgBFF2Wy6uBdeIK5mH hhSe1dnu1SiFR3CUL0jbqkdgOcFsMkHyXjYM0= On Mon, Jan 19, 2009 at 1:18 AM, Mario Emmenlauer <mario(a)emmenlauer.de> wrote: > I have a large C/C++ project with thousands of lines of code and > many thirdparty-libraries. When it startes up, it reserves around 1.8GB > of virtual memory (40MB real memory), and I don't know why. Shared library mappings, contiguous reserved allocation blocks used by malloc, etc. (and who knows what the third-party libraries do). > Its a 32bit application that should use all real memory possible (4GB), > but it crashes after reserving only 2.5GB real mem when the virtual mem > reaches 4GB. The crash is logical, but why does it start with 1.8GB > vmem initially? On ppc it reserves only 350MB vmem. Different platforms can have different behaviors and libraries loaded. So I am not to surprise the results are different between Intel and PowerPC. > Are there better/other tools I could try to inspect where the additional vmem is reserved? vmmap (http://developer.apple.com/documentation/Darwin/Reference/ManPages/man1/vmm…) -Shawn _______________________________________________ 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

Problems retrieving settings for proxy with authentication
by Cristescu Alexandru 19 Jan '09

19 Jan '09

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:X-Mailer:Date:From:Subject:To:MIME-Version:Content-Type:Message-ID; b=zLHnyjFgm+Lh1NRit1IrLuVu2QIZmjDfKARSVJKoBvrtrfp1Wskw1huICnzOqtD0D6DuJ3Lb2c4TGbBhjhMcJ2mWGMKFOD5wBB0lYgdaUtV7iGh4GUgXBpSoKiM87XStdKevAcWnSRSZjF8FnDgUH6p4bmPTZb5F2QD9sdaDhj0=; _______________________________________________ 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: Adding partition to expanded drive
by Stephen Hoffman 19 Jan '09

19 Jan '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Hard to say. The EFI/UEFI spec is ill-considered in this area, IMNSHO. So, is this a bug in the implementation of GPT, or am I doing something wrong? I had several conversations with the folks that were working on the spec (was implementing GPT on another platform) and there were various cases where the GPT design falls apart. Disk devices that present volumes that can expand entirely on-line are one such area; I haven't looked to see if the specs have gotten around to recognizing that volumes are increasingly virtual now. The whole recovery design seems odd. If the primary GPT (the blocks after the MBR) are stomped on, what are the chance that the rest of the volume structures are valid; that the rest of the core volume structures within the partition(s) haven't also been stomped on? (And what happens when you have three GPT structures during a transition - the front GPT, old back GPT, new back GPT - what happens when you get rolling with an expansion and fail; which GPT do you believe, and how do you clean up?) And if you're going to do a backup of a critical structure, why at the back of the disk and not at a known block-offset pattern (look for a backup every 1001 blocks into the disk, for instance) within the disk? There have also been versions of EFI firmware on some platforms that stomped on disks, too; where a non-GPT volume structure is created on a disk (one that wasn't completely erased) and the console finds the backup GPT and helpfully stomps on the front of the disk without any notification to the user. Where the console corrupts a disk. My vote would be to get rid of the backup GPT entirely. Allow it, but don't require it. Or use a block offset from within the disk. But don't ever let the console get "silently helpful" here. Whoever did the GPT support for this particular case would have to address this. But it's just not as simple as it looks here - or at least it wasn't. _______________________________________________ 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: cache clearing from keeping code and data in the same segment
by Joel Reymont 19 Jan '09

19 Jan '09

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:cc:message-id:from:to :in-reply-to:content-type:content-transfer-encoding:mime-version :subject:date:references:x-mailer; bh=sWRfeZs+AcfhTZOUTOZ2Uupxu0uc2fZCZbzHiKNR67c=; b=oKo//0le8XL5MiUHzfee47ZHspuVkZprZ2W4eGjd8f2ajTgGCBE+H3pnCTG3b7X8et Kceyf0gKsEBsUVx/9tF2+/MXV7mY/LodtA4nEbzrmTBsII9XYIDhZQkmQeQIHepNKZRV nfrIJSJyKeBWmuMWNW9tFMorhqEiNl0TMdo6A= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=cc:message-id:from:to:in-reply-to:content-type :content-transfer-encoding:mime-version:subject:date:references :x-mailer; b=XQWT8vNoNa8Zd0UKv8cNpBlf7WlYHoPcfZK4ESYFHHcDmlPf4xZPk4DVOPO8HUZNwb vMpwsgR3QdikhKUW33HUPq7wNnxXpLCwFy1qWoeTQoGF4HwpHIknOhGFyD93A2Q+RhdA FS5R+PCDqRX9qqxLr7V1YQcQ2EhDKXCyDpRGQ= On Jan 19, 2009, at 4:44 PM, Terry Lambert wrote: --- http://twitter.com/wagerlabs _______________________________________________ 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… Also I will reiterate that if you target the iPhone like you've claimed to be doing, the mappings for writable and executable are mutually exclusive. Yes, thank you Terry. The code generated for the iPhone will NOT be self-modifying or interpreted. After that, ARM binary code is ARM binary code is ARM binary code. This email sent to site_archiver(a)lists.apple.com

1 0

Re: cache clearing from keeping code and data in the same segment
by Terry Lambert 19 Jan '09

19 Jan '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com -- Terry On Jan 19, 2009, at 4:34 AM, Joel Reymont <joelr1(a)gmail.com> wrote: On Jan 19, 2009, at 12:25 PM, Jonas Maebe wrote: --- http://twitter.com/wagerlabs _______________________________________________ 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/tlambert%40apple.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… It is a function of the CPU. I pointed out that data mappings are separate when data and code are butted up against each other at page boundaries when you were asking about mach-o file layout earlier. This is pretty clearly spelled out in the blue and white architecture books; no yellow, orange, or red books necessary. Also I will reiterate that if you target the iPhone like you've claimed to be doing, the mappings for writable and executable are mutually exclusive. Keeping code and read/write data close to each other (putting them in the same segment may qualify) may result in the icache getting invalidated on data writes (to deal with self-modifying code), [...] Such flushing, if any, would performed by the cpu though, the OS has no influence over that. So you are saying that this is a function of the CPU as opposed to the OSX itself? I am dealing with self-modifying code (Forth), although I'm willing to trade certain things for performance. This email sent to tlambert(a)apple.com This email sent to site_archiver(a)lists.apple.com

1 0

Re: cache clearing from keeping code and data in the same segment
by Jonas Maebe 19 Jan '09

19 Jan '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On 19 Jan 2009, at 13:34, Joel Reymont wrote: On Jan 19, 2009, at 12:25 PM, Jonas Maebe wrote: Yes. Jonas _______________________________________________ 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… Keeping code and read/write data close to each other (putting them in the same segment may qualify) may result in the icache getting invalidated on data writes (to deal with self-modifying code), [...] Such flushing, if any, would performed by the cpu though, the OS has no influence over that. So you are saying that this is a function of the CPU as opposed to the OSX itself? I am dealing with self-modifying code (Forth), although I'm willing to trade certain things for performance. Self-modifying code is in many cases one of the most effective ways to kill performance on modern cpus. This email sent to site_archiver(a)lists.apple.com

1 0