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

Re: Nonstandard serial speeds
by Steve Checkoway 10 Nov '09

10 Nov '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On Nov 9, 2009, at 9:16 PM, Godfrey van der Linden wrote: -- Steve Checkoway _______________________________________________ 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… You may, if you are very, very lucky and the low level driver has implemented it, use the IOSSIOSPEED ioctl the argument is a uint32_t * to the desired baudrate. You can find the header at <IOKit/serial/ioss.h> Thanks for the suggestion. Unfortunately, this particular driver doesn't seem to implement it. I'll have to do something like filtering data in the microcontroller and only send the data of interest. Oh well. This email sent to site_archiver(a)lists.apple.com

1 0

Re: ubc and removable devices
by Mike Mackovitch 10 Nov '09

10 Nov '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com User-agent: Mutt/1.5.20 (2009-06-14) On Tue, Nov 10, 2009 at 08:03:17AM -0800, Mike Mackovitch wrote: > On Mon, Nov 09, 2009 at 08:32:08PM +0000, Joel Reymont wrote: > > Do I correctly understand that the UBC pages responsible for a > > removable device get evicted when said device is ejected? > > vnodes have a "struct ubc_info" that associate them with the > corresponding VM objects. When a file system is unmounted all > the vnodes are disassociated from the file system (vclean()) > which includes cleaning up the UBC reference (ubc_destroy_named()). > > So, yes. Technically speaking: the vnodes/UBC get cleaned up on the file system unmount... and not the device ejection. So, an unmount/mount would suffice. The *device* doesn't need to be physically ejected/removed/reattached. --macko _______________________________________________ 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: ubc and removable devices
by Mike Mackovitch 10 Nov '09

10 Nov '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com User-agent: Mutt/1.5.20 (2009-06-14) On Mon, Nov 09, 2009 at 08:32:08PM +0000, Joel Reymont wrote: > Do I correctly understand that the UBC pages responsible for a > removable device get evicted when said device is ejected? vnodes have a "struct ubc_info" that associate them with the corresponding VM objects. When a file system is unmounted all the vnodes are disassociated from the file system (vclean()) which includes cleaning up the UBC reference (ubc_destroy_named()). So, yes. HTH --macko _______________________________________________ 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: ubc and removable devices
by Joel Reymont 10 Nov '09

10 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:subject:mime-version :content-type:from:in-reply-to:date:cc:content-transfer-encoding :message-id:references:to:x-mailer; bh=eWzinMXrE+8TLXtivLSGixmtKRa4mSQ7jnvo5DYV0GU=; b=vLh1LcieWYeVnWz27mg7bVMQGhSHt+gYtW+gIpTlL5h4jXjHb6feqS+KdaU3UUtuAE +BoeYww3QKa6R26bleWWm67iqjVslVKX9oLPV8MbuTTv4cDA3n9DRi7W11jtA7rto3HR s6HXy3Na9mFO30KjH7oL5LpA8DSSeMpRYl+DY= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=subject:mime-version:content-type:from:in-reply-to:date:cc :content-transfer-encoding:message-id:references:to:x-mailer; b=mmg6WTjwBmq30uM/nAlMTCGOVgUE/MorpU/FzhrAkOi7dMjdOj5AYxIBdK4W8EYJtl ciDNYVjVhiKuwH6u+Gif8PDiAis8q9d3yC6qjprFYL1c615EJP0ZgJThcfd7mxy+Q0Jn odEqwAGwH5GhGyL3FRrKHVl9pifolCTsSYRdg= On Nov 10, 2009, at 7:53 AM, Stefan Haller wrote: Yes but it doesn't evict mmap-ed dynamic libraries. You need to allocate 2x physical memory for it to happen. I have 4Gb of physical memory so allocating twice that and then touching every page (below) takes over 4 minutes. It's much faster to place the app on a removable device and then eject the device to purge the caches of the app data. I haven't looked for this behavior in the kernel source code but it seems to be consistent and logical when I think about it. Another alternative is rebooting but that takes about a minute and requires me to restore all the apps I have running. --- #!/usr/bin/env python import mmap SIZE = 8589934592L # 8Gb map = mmap.mmap(-1, SIZE) map[::4096] = '\x01' * (SIZE // 4096) --- #include <sys/mman.h> #include <fcntl.h> #include <stdio.h> static long SIZE = 8589934592L; for (i = 0; i < SIZE; i += 4096) result[i] = 1; return 0; } return 1; } --- firefox for android! http://wagerlabs.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… Did you try the "purge" command-line tool? int main(int argc, char**argv) { char *result = mmap(0, SIZE, PROT_READ|PROT_WRITE, MAP_ANON| MAP_PRIVATE, -1, 0); if (result != MAP_FAILED) { unsigned long i; This email sent to site_archiver(a)lists.apple.com

1 0

Re: ubc and removable devices
by lists＠haller-berlin.de 10 Nov '09

10 Nov '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com User-agent: MacSOUP/2.8.2 (Mac OS X version 10.6.1 (x86)) Joel Reymont <joelr1(a)gmail.com> wrote: > Ejecting a device to empty the UBC surely beats rebooting or waiting > to allocate twice the physical memory, 4min+ for 8gb in my case. Did you try the "purge" command-line tool? -Stefan -- Stefan Haller Berlin, Germany http://www.haller-berlin.de/ _______________________________________________ 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: Nonstandard serial speeds
by Godfrey van der Linden 10 Nov '09

10 Nov '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com You can find the header at <IOKit/serial/ioss.h> Godfrey On 2009-11-10, at 9:31 AM, Steve Checkoway wrote: -- Steve Checkoway _______________________________________________ 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/gvdl%40mac.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… You may, if you are very, very lucky and the low level driver has implemented it, use the IOSSIOSPEED ioctl the argument is a uint32_t * to the desired baudrate. I have a microcontroller that seems perfectly willing to speak serial at 1Mb/s (which is a handy division of the clock rate). On my Mac, I cannot seem to get tcsetattr() to accept that nonstandard speed. Is there any way to achieve that? This email sent to gvdl(a)mac.com This email sent to site_archiver(a)lists.apple.com

1 0

Nonstandard serial speeds
by Steve Checkoway 09 Nov '09

09 Nov '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com -- Steve Checkoway _______________________________________________ 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 microcontroller that seems perfectly willing to speak serial at 1Mb/s (which is a handy division of the clock rate). On my Mac, I cannot seem to get tcsetattr() to accept that nonstandard speed. Is there any way to achieve that? This email sent to site_archiver(a)lists.apple.com

1 0

ubc and removable devices
by Joel Reymont 09 Nov '09

09 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=NuNxQdIs5cxv1jyEnXLpcATrI2/14lyVyHa/OEqn3Xs=; b=oLjq5yMBKnxBQJtBGmqVdgPXo/YFfoDMmVEZj6tKOaL3YMtcrmmwAkS0JtZwqZPQXM MTjNKoRP+3kLfUcmrsQSbWJGE8SKWCmtRxCcIG5P73a3A/YNsLqRCU0t6FrNYq7iLPZ0 63HgIjA4EAvc9HAR8mq1XS5NIVmiXmWJ+PLA8= 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=opNsAV54o8is5D0plbwDrTEDDcb/w2RlIfei4jcCe3iNyArHiLd6JMcStWx+/GUECb ZPZrrbpSWpsyijYEsw4Px6SBljpRlRbvEKM1ffvMuYEjMQc5jbUkHXPMHM9u0Qy42+jk S0HSjC2G8MqOOukhrXL6H06Ijw7J1SUUV5LjI= Thanks, Joel --- firefox for android! http://wagerlabs.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… Do I correctly understand that the UBC pages responsible for a removable device get evicted when said device is ejected? Ejecting a device to empty the UBC surely beats rebooting or waiting to allocate twice the physical memory, 4min+ for 8gb in my case. This email sent to site_archiver(a)lists.apple.com

1 0

Re: vfs, ubc and pre-fetching on snow leopard
by Joel Reymont 09 Nov '09

09 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:subject:mime-version :content-type:from:in-reply-to:date:cc:content-transfer-encoding :message-id:references:to:x-mailer; bh=yRHbmj9Ya0sfmXiDGj5R3T4pgwmdDSs92Vg66paiDa4=; b=U3UMNq2m7Xkzlg5mmuVSVs7S9Bo7LIZicprtqESJI5hyu66q97EOp0hQJK/5rSkzW9 +nbkH90IpYF0OnctdWECV9C6eDiL1ioo1W9/81c5D4ZQn81PLo9Ziqw8EoJiA3ezjv9s JJpUGplGJThq9GK94W41dyLYZO94m9o5ya3C8= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=subject:mime-version:content-type:from:in-reply-to:date:cc :content-transfer-encoding:message-id:references:to:x-mailer; b=RDoSxiuDJzoTsyR/WzisxToNK1IcfRrsSZ+VGzJCW4tknkGLLNYvb3UZV5gQqtAoFO dsECA2sS0JfO9JDmsISbo+Wm0JRnUBCP+gvBmdqFRIzmfVv1jl0b8aX/dT0SwU6Sq7J8 +n7CzM0z+CJCw0cENpowVd+eL+O+HUJWxpick= On Nov 9, 2009, at 4:55 AM, Eric Gouriou wrote: Once I see a different stack trace, e.g I know that initialization code in the libraries is being run. Here's iteration #1 (73 pages) __LINKEDIT, __symtab = 2 pages, 5016 pages in __LINKEDIT, __symtab = 2 pages, 5018 pages in __DATA, __const = 4 pages, 4850 pages in __DATA, __const = 4 pages, 4854 pages in __DATA, __const = 4 pages, 4858 pages in __DATA, __const = 4 pages, 4862 pages in __DATA, __const = 4 pages, 4866 pages in __DATA, __const = 6 pages, 4956 pages in __DATA, __const = 6 pages, 4962 pages in __DATA, __const = 6 pages, 4968 pages in __DATA, __const = 6 pages, 4974 pages in __DATA, __const = 6 pages, 4980 pages in __DATA, __objc_classrefs = 6 pages, 4986 pages in __DATA, __data = 6 pages, 4992 pages in __DATA, __data = 7 pages, 4998 pages in iteration #2 (84 pages) __LINKEDIT, __symtab = 2 pages, 5018 pages in __DATA, __const = 6 pages, 4962 pages in __DATA, __const = 6 pages, 4968 pages in __DATA, __const = 6 pages, 4974 pages in __DATA, __const = 6 pages, 4980 pages in __DATA, __objc_classrefs = 6 pages, 4986 pages in __DATA, __data = 6 pages, 4992 pages in __DATA, __data = 7 pages, 4998 pages in __DATA, __data = 7 pages, 5005 pages in __LINKEDIT, __symtab = 1 pages, 5024 pages in __LINKEDIT, __symtab = 2 pages, 5025 pages in __LINKEDIT, __symtab = 4 pages, 5027 pages in __LINKEDIT, __symtab = 25 pages, 5031 pages in and iteration #3 (54 pages) __LINKEDIT, __symtab = 2 pages, 5016 pages in __LINKEDIT, __symtab = 2 pages, 5018 pages in __DATA, __const = 4 pages, 4850 pages in __DATA, __const = 4 pages, 4854 pages in __DATA, __const = 4 pages, 4858 pages in __DATA, __const = 4 pages, 4862 pages in __DATA, __const = 4 pages, 4866 pages in __DATA, __const = 4 pages, 4870 pages in __DATA, __const = 4 pages, 4874 pages in __DATA, __const = 4 pages, 4878 pages in __DATA, __const = 4 pages, 4882 pages in __DATA, __const = 4 pages, 4886 pages in __DATA, __const = 5 pages, 4916 pages in __DATA, __const = 5 pages, 4951 pages in --- firefox for android! http://wagerlabs.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… Is Firefox/Minefield still single-threaded by the time your measurements are captured ? If not this could also lead to run-to-run variations. I'm only looking at the differences in the dyld stage, usually the first 13-20 page-ins. Do you see that order being stable across all runs ? Not really. I'll investigate the Firefox code but I would also like to figure out the variations in the dyld portion of the page-ins, before library initializers run. Following are 3 complete samples of dyld-related page-ins. I determined that these are the product of pre-initialization dyld work by looking at the user-space stack trace, e.g. this is before initializers dyld`ImageLoaderMachOCompressed::rebase (ImageLoader::LinkContext const&)+0x63 dyld`ImageLoader::recursiveRebase (ImageLoader::LinkContext const&)+0x67 dyld`ImageLoader::recursiveRebase (ImageLoader::LinkContext const&)+0x45 dyld`ImageLoader::link(ImageLoader::LinkContext const&, bool, bool, ImageLoader::RPathChain const&)+0x89 dyld`dyld::link(ImageLoader*, bool, ImageLoader::RPathChain const&)+0x76 dyld`dyld::_main(macho_header const*, unsigned long, int, char const**, char const**, char const**)+0xb06 dyld`dyldbootstrap::start(macho_header const*, int, char const**, long)+0x31f dyld`_dyld_start+0x2a firefox-bin`0x100000000 libobjc.A.dylib`_mapStrHash+0xb libobjc.A.dylib`NXMapGet+0x2f libobjc.A.dylib`addNamedClass+0x26 libobjc.A.dylib`_read_images+0x247 libobjc.A.dylib`map_images_nolock+0x4f2 libobjc.A.dylib`map_images+0x72 dyld`dyld::notifyBatchPartial(dyld_image_states, bool, char const* (*)(dyld_image_states, unsigned int, dyld_image_info const*))+0x2fb dyld`dyld::registerImageStateBatchChangeHandler (dyld_image_states, char const* (*)(dyld_image_states, unsigned int, dyld_image_info const*))+0x1ba libSystem.B.dylib`dyld_register_image_state_change_handler+0x58 libobjc.A.dylib`_objc_init+0x3a dyld`ImageLoaderMachO::doModInitFunctions (ImageLoader::LinkContext const&)+0xe4 dyld`ImageLoader::recursiveInitialization (ImageLoader::LinkContext const&, unsigned int)+0xec dyld`ImageLoader::recursiveInitialization (ImageLoader::LinkContext const&, unsigned int)+0x9d dyld`ImageLoader::recursiveInitialization (ImageLoader::LinkContext const&, unsigned int)+0x9d dyld`ImageLoader::recursiveInitialization (ImageLoader::LinkContext const&, unsigned int)+0x9d dyld`ImageLoader::recursiveInitialization (ImageLoader::LinkContext const&, unsigned int)+0x9d dyld`ImageLoader::recursiveInitialization (ImageLoader::LinkContext const&, unsigned int)+0x9d dyld`ImageLoader::recursiveInitialization (ImageLoader::LinkContext const&, unsigned int)+0x9d dyld`ImageLoader::recursiveInitialization (ImageLoader::LinkContext const&, unsigned int)+0x9d dyld`ImageLoader::runInitializers (ImageLoader::LinkContext const&)+0x3a I would expect dyld to always be doing the same thing with the same library but either this is not the case or the kernel is reacting completely different each time. This email sent to site_archiver(a)lists.apple.com

1 0

RE: Building Objc4 on Windows
by Matthieu Cormier 09 Nov '09

09 Nov '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Thread-index: AcphQFA2TYilarMSRkqV3jvuTGLP+w== Thread-topic: RE: Building Objc4 on Windows _______________________________________________ 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