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

Re: rand() and srand() broken?
by Jason Foreman 25 Nov '09

25 Nov '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On Nov 25, 2009, at 11:26 AM, Derek Gaston wrote: > This piece of code should generate 100 random numbers between 0 and 1... I am expecting that they be fairly spaced out as well... What basis do you have for this expectation? Is it spelled out in any specification or standard somewhere? You're seeding the RNG with a monotonically increasing seed, and getting what looks like monotonically increasing starting values for your random sequences. The fact that you get what you expect on Linux seems to be purely coincidental. > This is actually indicative of what we actually do in one of our applications (I know it doesn't usually make much sense to reseed all the time... but you're just going to have to trust me that it's necessary in our case ;-) I think you're going to have to explain why you feel reseeding is necessary. You probably shouldn't be doing it. Jason _______________________________________________ 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 smime.p7s

1 0

Re: rand() and srand() broken?
by Greg Shenaut 25 Nov '09

25 Nov '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On Nov 25, 2009, at 9:26 AM, Derek Gaston wrote: #include <stdlib.h> #include <iostream> int main() { for(unsigned int i=0; i<100; i++) { srand(i+1); double value = rand()/(RAND_MAX + 1.0); std::cout<<value<<std::endl; } } Here is the output I get on Snow Leopard: 7.82637e-06 1.56527e-05 2.34791e-05 3.13055e-05 3.91318e-05 4.69582e-05 5.47846e-05 6.2611e-05 7.04373e-05 7.82637e-05 8.60901e-05 9.39164e-05 0.000101743 0.000109569 0.000117396 0.000125222 0.000133048 0.000140875 ..... Running the _exact_ same code on Linux I get: 0.840188 0.700976 0.56138 0.916458 0.274746 0.135439 0.486904 0.352761 0.206965 0.565811 0.926345 0.7856 0.632643 ..... Thanks for any help! Derek _______________________________________________ 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/gkshenaut%40ucdavis.edu _______________________________________________ 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… Hello everyone... I've been struggling with rand and srand on OSX 10.6 and 10.5. It seems to me that something is terribly broken. Here is a short program I wrote to convince myself of that: Why are you reseeding the rng in each loop? This will never be random. Try moving the srand call to before the loop and seeding it only once. This piece of code should generate 100 random numbers between 0 and 1... I am expecting that they be fairly spaced out as well... This is actually indicative of what we actually do in one of our applications (I know it doesn't usually make much sense to reseed all the time... but you're just going to have to trust me that it's necessary in our case ;-) Note how _not_ random they are! This is actually what we are seeing in our application! Notice how they are nice and random... and well spread out through the 0 to 1 range. Am I doing something wrong?? I've tested this on several Linux distros and it always works fine.... what is different about rand and srand on OSX? This email sent to gkshenaut(a)ucdavis.edu This email sent to site_archiver(a)lists.apple.com

1 0

Re: rand() and srand() broken?
by Alastair Houghton 25 Nov '09

25 Nov '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Dkim-signature: v=1; a=rsa-sha1; c=relaxed; d=alastairs-place.net; h= subject:mime-version:content-type:from:in-reply-to:date:cc: content-transfer-encoding:message-id:references:to; q=dns/txt; s= aug07; bh=nyq3AiEg0i9kzH1ECdh+x3KGTUA=; b=CmR9alopGCT/04UjwQgA7J kvPcEgh60T2owuQ7duil42zbf6D9BYm2HFn7oMm0gcdOI3rOSdmSUhxYbqhBmO31 bgTUlMyGd/Gzj2GIk2yUl/YGdvwZgmZ2tvCE1pySWZl+HkVwj1RHdjR4QTiqFkL/ Kgr4lNMqsNBd1a9mID3oo= On 25 Nov 2009, at 17:26, Derek Gaston wrote: > This is actually indicative of what we actually do in one of our applications (I know it doesn't usually make much sense to reseed all the time... but you're just going to have to trust me that it's necessary in our case ;-) It's wrong. Whether you think it's necessary or not, it's just plain wrong. Pseudo-random number generators usually don't guarantee good output if you re-seed them all the time. Perhaps you should explain the problem that you think requires reseeding every time and someone can help you to design your algorithm better? You might also want to try the random() function rather than the rand() function, since it's usually a better generator. I haven't looked to see whether that's true on OS X, but it normally has a better range and often uses a better algorithm behind the scenes too. Kind regards, Alastair. -- http://alastairs-place.net _______________________________________________ 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: [APPL:DarwinDev] rand() and srand() broken?
by Jay A. Kreibich 25 Nov '09

25 Nov '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com User-agent: Mutt/1.5.18 (2008-05-17) On Wed, Nov 25, 2009 at 10:26:51AM -0700, Derek Gaston scratched on the wall: > Hello everyone... I've been struggling with rand and srand on OSX 10.6 and 10.5. It seems to me that something is terribly broken. Here is a short program I wrote to convince myself of that: > > #include <stdlib.h> > #include <iostream> > > int main() > { > for(unsigned int i=0; i<100; i++) > { > srand(i+1); > double value = rand()/(RAND_MAX + 1.0); > std::cout<<value<<std::endl; > } > } > > This piece of code should generate 100 random numbers between > 0 and 1... I am expecting that they be fairly spaced out as well... I wouldn't trust it to do that. A good PRNG generates *sequences* that are well spaced and seemingly random. You don't have sequences. > This is actually indicative of what we actually do in one of our > applications (I know it doesn't usually make much sense to > reseed all the time... but you're just going to have to trust > me that it's necessary in our case ;-) Not only does it not make sense, as this code shows, it is a very bad idea. You're setting the seed over and over with very similar numbers (in the scope of a full unsigned integer) and getting similar first values. Not exactly a surprise. Since this isn't a sequence, there is little reason to assume they'd be fairly spaced out. You need a sequence off a single seed for that. > Am I doing something wrong?? Yeah, stop re-seeding the PRNG. Do it once on startup and never again. Or, find a much better source of semi-random seeds. Clearly the Linux implementation does a better job at producing first-value numbers that cover the number domain a bit better. But that doesn't discount the fact you're using the PRNG incorrectly. -j -- Jay A. Kreibich < J A Y @ K R E I B I.C H > "Our opponent is an alien starship packed with atomic bombs. We have a protractor." "I'll go home and see if I can scrounge up a ruler and a piece of string." --from Anathem by Neal Stephenson _______________________________________________ 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: rand() and srand() broken?
by Andrew Myrick 25 Nov '09

25 Nov '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com I can't speak to the results you're seeing (perhaps it's worth filing a bug?), but have you considered seeding from a high quality entropy source with sranddev()? I modified your test program to call it inside and outside of the loop, and got what appeared to be reasonable results in both cases. -Andrew On Nov 25, 2009, at 9:26 AM, Derek Gaston wrote: > Hello everyone... I've been struggling with rand and srand on OSX 10.6 and 10.5. It seems to me that something is terribly broken. Here is a short program I wrote to convince myself of that: > > #include <stdlib.h> > #include <iostream> > > int main() > { > for(unsigned int i=0; i<100; i++) > { > srand(i+1); > double value = rand()/(RAND_MAX + 1.0); > std::cout<<value<<std::endl; > } > } > > This piece of code should generate 100 random numbers between 0 and 1... I am expecting that they be fairly spaced out as well... > > This is actually indicative of what we actually do in one of our applications (I know it doesn't usually make much sense to reseed all the time... but you're just going to have to trust me that it's necessary in our case ;-) > > Here is the output I get on Snow Leopard: > > 7.82637e-06 > 1.56527e-05 > 2.34791e-05 > 3.13055e-05 > 3.91318e-05 > 4.69582e-05 > 5.47846e-05 > 6.2611e-05 > 7.04373e-05 > 7.82637e-05 > 8.60901e-05 > 9.39164e-05 > 0.000101743 > 0.000109569 > 0.000117396 > 0.000125222 > 0.000133048 > 0.000140875 > ..... > > Note how _not_ random they are! This is actually what we are seeing in our application! > > Running the _exact_ same code on Linux I get: > > 0.840188 > 0.700976 > 0.56138 > 0.916458 > 0.274746 > 0.135439 > 0.486904 > 0.352761 > 0.206965 > 0.565811 > 0.926345 > 0.7856 > 0.632643 > ..... > > Notice how they are nice and random... and well spread out through the 0 to 1 range. > > Am I doing something wrong?? I've tested this on several Linux distros and it always works fine.... what is different about rand and srand on OSX? > > Thanks for any help! > > Derek > _______________________________________________ > 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/amyrick%40apple.com > > This email sent to amyrick(a)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

rand() and srand() broken?
by Derek Gaston 25 Nov '09

25 Nov '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:from:content-type :content-transfer-encoding:subject:date:message-id:to:mime-version :x-mailer; bh=N0jhSLA7Y92/PdZwTnEljnFq00e38w1GaMvpluKbh/o=; b=f5QxEc/EfYj1Jbl/vT3OEouzVEH4jEhC5Lr829gK9G57VZcIcY/9k4tsoeA86tF2jl IdbqggTQmANSNhNIR4kKWmBjFcgkF8BnMkDEpXhE8wEN634AHuijorDNMU9O9HkCXF+f MMLpawEz7o6E9h7Fhj4sY7reIoo2YyoYFFb7g= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=from:content-type:content-transfer-encoding:subject:date:message-id :to:mime-version:x-mailer; b=Phg5OvYpfQnllggwXiB/CnNtb2O2G1VOnrPAseSKZW8xpszw0zVvdG2vlDCf3Qt0zZ 3sXBsNOGd2HW3pyvvfLAn0jDh1vhEzDELvoEHzR4ZWw6k8CVatQEi4f9N2xA9z6T2s5z 5BFuhk18fNGQ5w8BatO9r/M9Bjg3ZFRG0daxE= Hello everyone... I've been struggling with rand and srand on OSX 10.6 and 10.5. It seems to me that something is terribly broken. Here is a short program I wrote to convince myself of that: #include <stdlib.h> #include <iostream> int main() { for(unsigned int i=0; i<100; i++) { srand(i+1); double value = rand()/(RAND_MAX + 1.0); std::cout<<value<<std::endl; } } This piece of code should generate 100 random numbers between 0 and 1... I am expecting that they be fairly spaced out as well... This is actually indicative of what we actually do in one of our applications (I know it doesn't usually make much sense to reseed all the time... but you're just going to have to trust me that it's necessary in our case ;-) Here is the output I get on Snow Leopard: 7.82637e-06 1.56527e-05 2.34791e-05 3.13055e-05 3.91318e-05 4.69582e-05 5.47846e-05 6.2611e-05 7.04373e-05 7.82637e-05 8.60901e-05 9.39164e-05 0.000101743 0.000109569 0.000117396 0.000125222 0.000133048 0.000140875 ..... Note how _not_ random they are! This is actually what we are seeing in our application! Running the _exact_ same code on Linux I get: 0.840188 0.700976 0.56138 0.916458 0.274746 0.135439 0.486904 0.352761 0.206965 0.565811 0.926345 0.7856 0.632643 ..... Notice how they are nice and random... and well spread out through the 0 to 1 range. Am I doing something wrong?? I've tested this on several Linux distros and it always works fine.... what is different about rand and srand on OSX? Thanks for any help! Derek _______________________________________________ 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: Open Source Releases
by William Siegrist 24 Nov '09

24 Nov '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On Nov 13, 2009, at 4:58 PM, William Siegrist wrote: > The Darwin Team is proud to announce the availability of the following Open Source releases: > > Mac OS X 10.6.2 > CF-550.13 (for Mac OS X 10.6.2) has now been posted. http://opensource.apple.com/source/CF/CF-550.13/ -- William Siegrist Darwin Team _______________________________________________ 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: Open Source Releases
by Fabien Costantini 22 Nov '09

22 Nov '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=yahoo.fr; s=s1024; t=1258874781; bh=9LcpS+XZaiWxU5a3OcozSPGpfXC+I+C3/DDV08mgycY=; h=Message-ID:X-YMail-OSG:Received:X-Mailer:References:Date:From:Subject:To:In-Reply-To:MIME-Version:Content-Type:Content-Transfer-Encoding; b=RPWqYT+CLHXtZ0vx+s+7cydUcAakk2g6MVoB010JFZrOgYLlK7NGXfaQakFcTIT4WAVqtXNdho60gDsvESHP5pRXQZdIi+uXqtXEmjxx0dNN51gKRq7rwziiwJQn6NeBTudqs18IJs/tYLx9IFODYHlVEyAF8993HA6oRjPCg+0= Domainkey-signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.fr; h=Message-ID:X-YMail-OSG:Received:X-Mailer:References:Date:From:Subject:To:In-Reply-To:MIME-Version:Content-Type:Content-Transfer-Encoding; b=Y0qLH8Mk7+rSjYFOpF/F4fvxTt8nQuJ7dIuTjsHaINq6AbEYMc049x/gchs/qksBC65/Kp6JAc25NdyNFZcozk+xeq15w7ApAYiinB51xqvvgyhonAZ9ZsR89PpLfwpyJdVZ40IQc8TmLgbXInjlOjrzXCDEXmLrgdAGaqDpOTk=; Found an answer to my question: I finally rebuilt the sources manually (without darwinbuild) and was able to generate my debug kernel properly. I just had to export my debug and target variables to my environment then ran: make exporthdrs all make all and it worked, as explained in the apple darwin documentation. ----- Original Message ---- I tcompiled the latest xnu 1228.15.4 (for build 10C540) release, and it gave me a fat (i386,x86_64) mach_kernel which boots the system, but extremely slowly (~10 min to boot on my intel macbook pro with snow leopard 10.6.2). I used the latest svn darwinbuild tool and the latest 10.6.2 open source xnu release. Am I missing something ? Is there a recommended procedure to build this kernel properly ? TIA, Fabien ________________________________ * Subject: Open Source Releases * From: William Siegrist <email@hidden> * Date: Fri, 13 Nov 2009 16:58:34 -0800 ________________________________ The Darwin Team is proud to announce the availability of the following Open Source releases: Mac OS X 10.6.2 Developer Tools 3.2.1 They are available in the usual location, at <http://opensource.apple.com/>. You will also notice pages for iPhone 3.1.1 and 3.1.2, but those releases had no new open source code to post. Thank you. William Siegrist Darwin Team __________________________________________________________________ Yahoo! Canada Toolbar: Search from anywhere on the web, and bookmark your favourite sites. Download it now http://ca.toolbar.yahoo.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/fab672000%40yahoo.fr This email sent to fab672000(a)yahoo.fr __________________________________________________________________ The new Internet Explorer® 8 - Faster, safer, easier. Optimized for Yahoo! Get it Now for Free! at http://downloads.yahoo.com/ca/internetexplorer/ _______________________________________________ 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: kevents, waitpid() with WNOHANG race
by Dave Keck 22 Nov '09

22 Nov '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:sender:received:in-reply-to :references:from:date:x-google-sender-auth:message-id:subject:to:cc :content-type; bh=//F7RHQlwoAmYU9kYH4oCTOVwsgz4zTIW1lr5Z/lhgQ=; b=D4S4hEFvMSfIf5hVLJG5cNg/SDJemwJLeNit8yEpIbdvROh+cU17VwgIGr+WWzOfSE N+3QTl+Jo7EZcTgCNLjlMiZoALAfGFbR675k8u2ul0WZzplHpzxrNz4vHFzcPJ4dOlIj ZpxUuw15zJGXxEpDVSKQIkOWA4pmELWONQmlo= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:from:date :x-google-sender-auth:message-id:subject:to:cc:content-type; b=uPnp9fQLRqzgQVpdC/DCyKCj3o7g4e8HwW2ReUXdGJDLnDtcmS2xJqGJrXW9h95+X1 oD8f0RfTThtOIS0/EDcaURfsmzjfrIhTctZSOZPhpPtkZUjx8fUmw2PuBGVqdFBTJX3X nEs7qEwlOVHkGm8ASMRgRtt9S29GPNY27KB1w= > Could you file a bug on this and attach a test case? rdar://7416090 And for anyone else, here's my test case with relevant comments: http://themha.com/kevent_waitpid_race.c _______________________________________________ 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: Open Source Releases
by Fabien Costantini 22 Nov '09

22 Nov '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=yahoo.fr; s=s1024; t=1258871643; bh=MGhy7dS/vQgRVz48nb5My5E7LjaWHTI90TGDJ9oz4lI=; h=Message-ID:X-YMail-OSG:Received:X-Mailer:Date:From:Subject:To:MIME-Version:Content-Type; b=vd7r7vO2q24rZAkE6ikrzYdY9j6Ky9X6hMdhWYB0X5nvI6TFULLT2Oy3j+Jk3r08+AY8kcMthwLzVL5wg82GiIvY88FchuImoEqrkffxryUoo8SZzMx9mLwAujXWXIuCmZtiCJQt+H95P81ICeeBDtp9Iz32NcMpaEg4IOAPnjM= Domainkey-signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.fr; h=Message-ID:X-YMail-OSG:Received:X-Mailer:Date:From:Subject:To:MIME-Version:Content-Type; b=PeLNpM/kJg1MrwZMoSe1ojoAZCtlTQqU8gZ/anPb2yP809SXr4zox7pqmyZdGlUHo3nqBTF6nc+a6n8XxrlZ+MIqpLGo3dkPZuJY4x404hHdQNf0/QuoDK5Zx+d8fSTJAlJ9wPL7Dlxm003fJ6dORRDp0SE8aF2IXXBiQ2UC1Uc=; I tcompiled the latest xnu 1228.15.4 (for build 10C540) release, and it gave me a fat (i386,x86_64) mach_kernel which boots the system, but extremely slowly (~10 min to boot on my intel macbook pro with snow leopard 10.6.2). I used the latest svn darwinbuild tool and the latest 10.6.2 open source xnu release. Am I missing something ? Is there a recommended procedure to build this kernel properly ? TIA, Fabien ________________________________ * Subject: Open Source Releases * From: William Siegrist <email@hidden> * Date: Fri, 13 Nov 2009 16:58:34 -0800 ________________________________ The Darwin Team is proud to announce the availability of the following Open Source releases: Mac OS X 10.6.2 Developer Tools 3.2.1 They are available in the usual location, at <http://opensource.apple.com/>. You will also notice pages for iPhone 3.1.1 and 3.1.2, but those releases had no new open source code to post. Thank you. William Siegrist Darwin Team __________________________________________________________________ Yahoo! Canada Toolbar: Search from anywhere on the web, and bookmark your favourite sites. Download it now http://ca.toolbar.yahoo.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