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

Re: Kernel Debug Kit for 10.5.2 build 9C7010 ?
by Jeshua Lacock 27 Mar '08

27 Mar '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On Mar 27, 2008, at 5:41 PM, David Elliott wrote: Hey Rob, On Mar 27, 2008, at 4:24 PM, Garth Cummings wrote: Hello, Regards, Jeshua Lacock, Owner <http://OpenOSX.com> phone: 877.240.1364 _______________________________________________ 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 Mar 27, 2008, at 7:36 PM, Garth Cummings wrote: it's unusual for a software update to change the installed build number like this I meant "security update"...software updates change the build number as a matter of course. I believe this is due to the Time Machine & Airport update, not the security update. True; the Time Machine & Airport update brings the Darwin Kernel Version to 9.2.2 This email sent to site_archiver(a)lists.apple.com

1 0

Re: Kernel Debug Kit for 10.5.2 build 9C7010 ?
by David Elliott 27 Mar '08

27 Mar '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Hey Rob, On Mar 27, 2008, at 4:24 PM, Garth Cummings wrote: -Dave _______________________________________________ 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 Mar 27, 2008, at 7:36 PM, Garth Cummings wrote: it's unusual for a software update to change the installed build number like this I meant "security update"...software updates change the build number as a matter of course. I believe this is due to the Time Machine & Airport update, not the security update. This email sent to site_archiver(a)lists.apple.com

1 0

Re: Kernel Debug Kit for 10.5.2 build 9C7010 ?
by Garth Cummings 27 Mar '08

27 Mar '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Hey Rob, On Mar 27, 2008, at 4:24 PM, Garth Cummings wrote: --gc ____________________________________________________________________ Garth Cummings gcummings(a)apple.com Sr. Software Engineer Apple Developer Technical Support <http://developer.apple.com/technicalsupport> WWDC 2008 • June 9-13 • Moscone West • San Francisco, CA <http://developer.apple.com/wwdc/> _______________________________________________ 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's unusual for a software update to change the installed build number like this I meant "security update"...software updates change the build number as a matter of course. This email sent to site_archiver(a)lists.apple.com

1 0

Re: Kernel Debug Kit for 10.5.2 build 9C7010 ?
by Garth Cummings 27 Mar '08

27 Mar '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: high accuracy timing options?
by Aaron Turner 27 Mar '08

27 Mar '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=beta; h=domainkey-signature:received:received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; bh=3KmzDoT6vwy2YUKWcKl1Dw4gnH4LzfQdEQ0IXn32uqU=; b=fejxA9ZrYRRijXYIbKk2NllG0FDdpeyAuzEHqd+474imI7Hd+DkhH2+EepbtB9aBPhnCVvFKj/OlNd9DlM9tdxI+T4G5FNFeiwk9hGnQrRrMiHZx9jHBZy307y8zAo9Q+VXXf0RXMEr8SnTPbEOfpH9wYukKejsTEgLwQd1ze0s= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=beta; h=message-id:date:from:to:subject:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; b=ckGTwBsW/DhYjngUy1p/LitmhGhDyWBYynsrRic5nCItGArzu9bq/1T1NyV2RREYkbMfXZUaqRhSzrsEwau1YgJOrEG5HvQGAUo5URdaA/lDovxeNh9+pzvYRnzPf4JGIej13/q4Uma4FN133e9qgx7tssXsV2R5G3Nadu21Xr8= On Thu, Mar 27, 2008 at 2:31 PM, Jonas Maebe <jonas.maebe(a)elis.ugent.be> wrote: > > On 27 Mar 2008, at 22:25, Andrew Gallatin wrote: > > One thing to keep in mind about MacOSX is that the sleep > > is *really* accurate compared to Linux, BSD and Solaris. > > > > See http://www.dragonflybsd.org/docs/nanosleep/index.shtml > > And also that this is mainly because for very small delays, Darwin > lets the "waiting" thread simply burn cpu cycles until the delay is > over (at least when I looked into this a couple of years ago). The > result is that very small delays are (were) quite inefficient from a > multi-processing perspective on Darwin (with the mentioned upshot that > they're pretty accurate). My current code uses nanosleep(). Works great up to 125k pps (less then 1% error). @ 130k, it goes way off (~10% error). -- Aaron Turner http://synfin.net/ http://tcpreplay.synfin.net/ - Pcap editing & replay tools for Unix They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety. -- Benjamin Franklin _______________________________________________ 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

Kernel Debug Kit for 10.5.2 build 9C7010 ?
by Rob Cas 27 Mar '08

27 Mar '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=beta; h=domainkey-signature:received:received:message-id:date:from:to:subject:mime-version:content-type; bh=Hk8H+lNA0zxF1FU+PYN3YYWiGYJO3gyxzO0hzuIpV2E=; b=NlNjpTmplLw8iSZMxTVExzNViD+cZsfXN7CMvKVZINLnjKhJSm2avTEyGtlW9qVLcN1ipALyaD3D4iZlVqRAZbeFk/p4R5TpxfObzTIedpUgqQPOcbNS7wEEW1rwosjR/I/OAnkCOIcl3YIiWUAgMtJivfDe3QgWVSC2wdrHiWc= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=beta; h=message-id:date:from:to:subject:mime-version:content-type; b=ZkjJcfxWux9VorByaHVC3K6jDEwOcdFVcxJpsKkAGbe8WDmHUtwzDd6o0NVMywDcDmvC8c/dhxEVUauq3zzxWLClJAj2FR37dzGStll47QBHysw/Fw/szS2qe+GtfsFGDh/F5rhGQKHQA3F2xiERP7E90EAwjSxFfe7ZrfCuBNw= _______________________________________________ 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: high accuracy timing options?
by Jonas Maebe 27 Mar '08

27 Mar '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com One thing to keep in mind about MacOSX is that the sleep is *really* accurate compared to Linux, BSD and Solaris. 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… On 27 Mar 2008, at 22:25, Andrew Gallatin wrote: See http://www.dragonflybsd.org/docs/nanosleep/index.shtml And also that this is mainly because for very small delays, Darwin lets the "waiting" thread simply burn cpu cycles until the delay is over (at least when I looked into this a couple of years ago). The result is that very small delays are (were) quite inefficient from a multi-processing perspective on Darwin (with the mentioned upshot that they're pretty accurate). This email sent to site_archiver(a)lists.apple.com

1 0

Re: high accuracy timing options?
by Andrew Gallatin 27 Mar '08

27 Mar '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Aaron Turner writes: > > First since everyone keeps asking: I'm trying to introduce artificial > delays in my code in order to accurately send packets at a user > specified rate- generally in the range of a few packets/sec all the > way to 160,000+p/s which is about where you start running into the > hardware limits. Code is currently pretty portable: Linux, *BSD, > Solaris, OS X, Windows/Cygwin and even MIPS based hardware running on > the international space station or so I've been told. > > Considering the basic overhead of processing the packets, sending > them, etc you start seeing delay times well under 10ms which is where > traditional sleep methods start becoming unreliable. One thing to keep in mind about MacOSX is that the sleep is *really* accurate compared to Linux, BSD and Solaris. See http://www.dragonflybsd.org/docs/nanosleep/index.shtml Drew _______________________________________________ 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: high accuracy timing options?
by Aaron Turner 27 Mar '08

27 Mar '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=beta; h=domainkey-signature:received:received:message-id:date:from:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; bh=lTkbCq6q6FIx0MabcvIXKUhTL2vZEK3Hoc+seOSKttg=; b=OES0g/60ZevsaznZWgLosZVyzN+HeNYs/i8ueR5LQ8utCh4gxZ2TCazn+74NV0GvRPK3jp3zvTiYMYbgK6eW4+AwsWuFxRNdkrWoMSKYmoLmee4/j7csKSJuYy/7qznCZqo4Ab9C7VKqbHn5N+0WAdZd89R1kK+OaoCmitYSYzU= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=beta; h=message-id:date:from:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; b=J88SsrJX4HGGyPP3wThbQ2HMJ2wJijK+jiFk9YTRK1OUkO8Qr14AmYK1kkdvSIPM1+2VWuUpp3vPJfXrDjYhrQqn5yYIxh+TRjLJ5kcP+T3c9vD5pipqLa5j6fi01rKOyL3d4NiDHVy0fdzq/hAdWDAtoOaz8mZxrslRQj4Z8+Q= On Thu, Mar 27, 2008 at 1:42 PM, Steve Sisak <sgs-lists(a)codewell.com> wrote: > At 3:14 PM -0400 3/27/08, Steve Sisak wrote: > >Note this code is written in Eudora, not tested. > > Here's a working version with numbers for a MacBook Pro 2.16GHz: > > Still not sure what you're trying to do, but this is the same as your > code minus "back", which can't happen -- hope this tells you > something. > > Since UpTime() just returns the processor timebase register, I'm > pretty sure it can't fail to increment. > > Note that the absolute value of fiveSeconds is going to vary based on > processor architecture and speed. > > -------- [snip] Wow, come back to my desk and i've got an inbox full of replies... thanks guys. First since everyone keeps asking: I'm trying to introduce artificial delays in my code in order to accurately send packets at a user specified rate- generally in the range of a few packets/sec all the way to 160,000+p/s which is about where you start running into the hardware limits. Code is currently pretty portable: Linux, *BSD, Solaris, OS X, Windows/Cygwin and even MIPS based hardware running on the international space station or so I've been told. Considering the basic overhead of processing the packets, sending them, etc you start seeing delay times well under 10ms which is where traditional sleep methods start becoming unreliable. Anyways, based on Steve's email below, I'll definitely have to look more closely at the AbsoluteTime based methods. Not sure if it's hooked into the HPET or RDTSC or what, but being able to reliably convert to current time is a big win (which is one of the big problems of the RDTSC). Thanks, Aaron _______________________________________________ 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: high accuracy timing options?
by Steve Sisak 27 Mar '08

27 Mar '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Here's a working version with numbers for a MacBook Pro 2.16GHz: -------- #include <CoreServices/CoreServices.h> #include <sys/time.h> #define NonZero(x) ((x).hi | (x).lo) int main(int argc, char *argv[]) { u_int32_t count = 0, fail = 0, back = 0, i; AbsoluteTime sleep_until, now, last, time_left; AbsoluteTime fiveSeconds = DurationToAbsolute(5000 * kDurationMillisecond); } while (NonZero(time_left)); printf("Ran UpTime() %lu times in 5 seconds.\n", count); printf("%f UpTime()/usec\n", (float)count / 5000 / 1000); printf("Failed to increment %lu times\n", fail); printf("Back in time %lu times\n", back); printf("-------------\n"); count = 0; fail = 0; back = 0; } exit(0); } Ran UpTime() 54272359 times in 5 seconds. 10.854471 UpTime()/usec Failed to increment 0 times Back in time 0 times ------------- Ran UpTime() 54571267 times in 5 seconds. 10.914254 UpTime()/usec Failed to increment 0 times Back in time 0 times ------------- Ran UpTime() 55075625 times in 5 seconds. 11.015125 UpTime()/usec Failed to increment 0 times Back in time 0 times ------------- Ran UpTime() 55023597 times in 5 seconds. 11.004719 UpTime()/usec Failed to increment 0 times Back in time 0 times ------------- Ran UpTime() 55205105 times in 5 seconds. 11.041020 UpTime()/usec Failed to increment 0 times Back in time 0 times ------------- TimerTest has exited with status 0. _______________________________________________ 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… At 3:14 PM -0400 3/27/08, Steve Sisak wrote: Note this code is written in Eudora, not tested. Still not sure what you're trying to do, but this is the same as your code minus "back", which can't happen -- hope this tells you something. Since UpTime() just returns the processor timebase register, I'm pretty sure it can't fail to increment. Note that the absolute value of fiveSeconds is going to vary based on processor architecture and speed. for (i = 0; i < 5; i ++) { time_left = fiveSeconds; // make non-zero for fail case sleep_until = AddAbsoluteToAbsolute(UpTime(), fiveSeconds); last = UpTime(); do { count++; now = UpTime(); if ((now.lo == last.lo) && (now.hi == last.hi)) { fail ++; } else { time_left = SubAbsoluteFromAbsolute(sleep_until, now); } last = now; This email sent to site_archiver(a)lists.apple.com

1 0