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

Re: vfs, ubc and pre-fetching on snow leopard
by Eric Gouriou 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-sha1; c=relaxed; d=pobox.com; h=subject :mime-version:content-type:from:in-reply-to:date:cc :content-transfer-encoding:message-id:references:to; s=sasl; bh= aYwelKxzwPS6UW0Li8ZxpxfHKLE=; b=bJD4hODxGqDoUPzqEk/RTYWfPuuiraIH eZxQ2CdCFGxq0ICya3rpR1O/g4ImFnIItNzLBRPI32H4xr/Yp5eAPR57zc7n3OSM p/iQrEVH9SLDvnWGx4aUQYiIef3U7E/D2iPc7BJPe2olyTxbXQwHNxDkithX+PKw eMfKs11qBjo= Domainkey-signature: a=rsa-sha1; c=nofws; d=pobox.com; h=subject :mime-version:content-type:from:in-reply-to:date:cc :content-transfer-encoding:message-id:references:to; q=dns; s= sasl; b=Zm0gST1qszHp5vU6O2aXUqlH/PKpzwDHAyvOQ3hi1PMhJ1CuJJpi9JS7 0dl6Zfc+or4WuzR3BlkZgTjqNkmBX+i+G2uSGsjGGNwvK5Uhi5yvXJYVYPKCTHw4 4bOiUWGFIm4aNgBu/eDYNR6BW62GJ4VVNPcDShdJ9tqiQGAMBss= On Nov 8, 2009, at 11:31 AM, Joel Reymont wrote: > I reboot and run Firefox using the same arguments but get completely different page-in patterns every time. Why does this happen and how does this work? > > How does the VFS determine the pages to fetch and the number of pages? I'll let a kernel expert reply. There is some speculative read-ahead, the amount of which was increased in SnowLeopard. It is dependent on the system resources available and is likely to keep changing from release to release. When mapping libraries dyld issues asynchronous read requests for pages likely to be useful soonish. This includes LINKEDIT (needed for the dynamic linking process) and some parts of TEXT and DATA. This might cause some run-to-run variation between the on-demand reads and those speculative reads. 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 trying to optimize disk access during cold startup of Firefox and I thought I'd reorder the symbols in the dylib by time. I determined that the function call chain during startup is A, B, C, D, E, F. I assumed that by ordering the symbols this way I'll get sequential disk access which doesn't seem to be the case. That's the recommended symbol ordering strategy. Do you see that order being stable across all runs ? > I monitor page-ins using a simple dtrace script [1]. I then translate these page-ins to actual segments and sections of the XUL dynamic library using the output from 'size -m -l XUL'. > > One run may give me (head -10 ...) > > __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 > > whereas another run > > __LINKEDIT, __symtab = 2 pages, 5016 pages in > __DATA, __const = 5 pages, 4921 pages in > __DATA, __const = 6 pages, 4956 pages in > __DATA, __data = 7 pages, 4998 pages in > __TEXT, __const = 40 pages, 256 pages in > __TEXT, __cstring = 44 pages, 307 pages in > __TEXT, __text = 73 pages, 2390 pages in > > Where in the kernel is the logic that's applied here and what's the proper way to optimize my disk access? Ordering by first access is your best bet. Eric > > Thanks, Joel > > [1] xul-pageins.d > > fbt::vnode_pagein:entry > { > self->v_name = stringof(((vnode_t)arg0)->v_name); > } > > fbt::vnode_pagein:entry > /self->v_name == "XUL"/ > { > printf("vnode_pagein: %s, offset: %u, size: %u\n", > self->v_name, arg3, arg4); > } > > --- > 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/eric.gouriou%40pobox.com > > This email sent to eric.gouriou(a)pobox.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

vfs, ubc and pre-fetching on snow leopard
by Joel Reymont 08 Nov '09

08 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=3uyKho2x5AsRzFwQlWE1EOxTBoBb5S2FcYUjjYifC80=; b=n/VEP4497ykOT61VWKcrGybJb1Y717D5g0/RrJEtdRrYV83ATX0zB5xXb/xf9gq6R/ Nusl2xGjWmFbd6jzmiEWA9xaqCG8o7xUBtcwjJfzgS8/az6SZehzGVFxFu6dzdXKuCM4 +tFFuC5KL7wZddAyDBz6MtOfV2ByerMbXMyuQ= 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=TIAoeJcFIvT1DCIebp+ypPm0wrNsJDeCUG20ksqv+GdsxlvX7mD+cgYvgtCHmkKJms /Or2RnrrCvUKE4Jvey5+ShZJ/rYDxKKex47pp/3vNTbwj3GKVXhfqqKv4wn5p27ZAPWi NIuuW49f5xShYWaEYU6fRTpU3rPqPLXPGCfRg= How does the VFS determine the pages to fetch and the number of pages? One run may give me (head -10 ...) __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 whereas another run __LINKEDIT, __symtab = 2 pages, 5016 pages in __DATA, __const = 5 pages, 4921 pages in __DATA, __const = 6 pages, 4956 pages in __DATA, __data = 7 pages, 4998 pages in __TEXT, __const = 40 pages, 256 pages in __TEXT, __cstring = 44 pages, 307 pages in __TEXT, __text = 73 pages, 2390 pages in Thanks, Joel [1] xul-pageins.d fbt::vnode_pagein:entry { self->v_name = stringof(((vnode_t)arg0)->v_name); } fbt::vnode_pagein:entry /self->v_name == "XUL"/ { printf("vnode_pagein: %s, offset: %u, size: %u\n", self->v_name, arg3, arg4); } --- 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… I reboot and run Firefox using the same arguments but get completely different page-in patterns every time. Why does this happen and how does this work? I'm trying to optimize disk access during cold startup of Firefox and I thought I'd reorder the symbols in the dylib by time. I determined that the function call chain during startup is A, B, C, D, E, F. I assumed that by ordering the symbols this way I'll get sequential disk access which doesn't seem to be the case. I monitor page-ins using a simple dtrace script [1]. I then translate these page-ins to actual segments and sections of the XUL dynamic library using the output from 'size -m -l XUL'. Where in the kernel is the logic that's applied here and what's the proper way to optimize my disk access? This email sent to site_archiver(a)lists.apple.com

1 0

Re: Difference between com.apple.FinderInfo (getxattr()) and ATTR_CMN_FNDRINFO (getattrlist())
by Chris Suter 08 Nov '09

08 Nov '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Hi Rohan, On Sun, Nov 8, 2009 at 11:30 PM, rohan a <info1686(a)gmail.com> wrote: > The getattrlist() API returns Finder Flags through ATTR_CMN_FNDRINFO > while the getxattr() returns them through the Extended attribute > com.apple.FinderInfo. > Any flags set using /Developer/Tools/GetFileInfo except for L (Locked) > flag shows up in com.apple.FinderInfo. The L flag shows up as uchg in > chflags > > Though the two look identical are there any differences? Are there any > flags returned by only one of the 2 API's. Is anyone a subset of the > other ? While doing a backup of files, is it ok to use either of them > ( ATTR_CMN_FNDRINFO or com.apple.FinderInfo ) or should I use both of > them ? They're almost certainly identical. To be sure, you can check the sources: <http://opensource.apple.com/> Regards, Chris _______________________________________________ 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

Difference between com.apple.FinderInfo (getxattr()) and ATTR_CMN_FNDRINFO (getattrlist())
by rohan a 08 Nov '09

08 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:received:date:message-id:subject :from:to:content-type; bh=RPHyhlFt0xiJpFCq4v7GYyqmNrjTHaq00l0Jog40QXc=; b=vdNbBuxQPL9T/Lyf8OpAeHhRRhn/s7sUEoyAPKFJBWj87N61uHXuLfDj9gXjbogMI1 Io4P/FxWk4v379e5UtrSJA6klnf75Euupgb33dMGs/Gqk+3UaER6F3OkwtoL4Bm/6oD+ Uy2fGAbr5WoBUu0tw42Qwb6+8Of6BKIl4hLgk= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:date:message-id:subject:from:to:content-type; b=cYcvEeUCi2fi5KQBJottgre6Dgn+D1opA6mUKB09twE7N84kUOunEgt0/cfGUX6HTa 0QpxsC3dM9MGIBhVXryDBwvLP+/pcx9RIFE77s9vSxGfpppuxl46rivhzjlXtKTIJXPW YFN5GDOBAXNrVw2oIqYEo2FjyrjYjr/fNmf04= Hi, The getattrlist() API returns Finder Flags through ATTR_CMN_FNDRINFO while the getxattr() returns them through the Extended attribute com.apple.FinderInfo. Any flags set using /Developer/Tools/GetFileInfo except for L (Locked) flag shows up in com.apple.FinderInfo. The L flag shows up as uchg in chflags Though the two look identical are there any differences? Are there any flags returned by only one of the 2 API's. Is anyone a subset of the other ? While doing a backup of files, is it ok to use either of them ( ATTR_CMN_FNDRINFO or com.apple.FinderInfo ) or should I use both of them ? Thanks _______________________________________________ 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

Browser based application and crash reporter
by Arnab Ganguly 08 Nov '09

08 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:received:date:message-id:subject :from:to:content-type; bh=I3NRpwvS63x1ixaT6kx/YYyijH2Nf7cKEiV0GHidVts=; b=OP7VhRH4GMNCcOA3m2K/pCDC8qbRmbWuMASTc4xMpLiwo7nfsllNSVLylS2sg2Un6Q so26ZbY5JQKMuIfRSSgMk8PkQ9wdTF9Q7aFmUA32G7oAau0fkaJxh7aZDNgnrKnKl5EA Gz1aAKg7lKG1xxqyNYGFy8+2xmBKHQyoYZ1I4= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:date:message-id:subject:from:to:content-type; b=WgmLHKmv87NqCjJLNDCAZsymORJF5ACexthOvdim9iA3X6FnDruWiNbGE74k2n5ANw 70/4ZOnK9qCHZCjZ4CQ/RlyIEQMPzgkuf0l1ejz7DlS1FhNVU5aSeYkceIXln8MMtAIu tTi88IIVpTTQMpRHA1RHOeZKQ6ZEBzZtPvTkM= _______________________________________________ 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: where are my code page-ins?
by Joel Reymont 05 Nov '09

05 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:content-type:mime-version :subject:from:in-reply-to:date:content-transfer-encoding:message-id :references:to:x-mailer; bh=3yKhC2Nm+ee9QxGNxb9y+EaHA3LJ4JEGI06tdBHIBfQ=; b=MCPxTpvUP4KbkSTxGS0+1J+0moI1hoG1CuWHdvtsZn58ZPW+yXGiwGrhKGChLDyWJ0 NEPNTqJ5t3Yq4SHsKZCbqUNmjd4R6TKiID5fiknGj17abGWe7HDKbpXdiR4rjqwqxzZc LV9/lFhDtPEcYBebRJH6tckDpJxRWuQpyYOzw= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=content-type:mime-version:subject:from:in-reply-to:date :content-transfer-encoding:message-id:references:to:x-mailer; b=Hp8tKefhtWc8rtwhm8TteGgv6EKliKokBAeDobze3qeHjleC3BnAG0nPZh8BRjgMgX FQ1F5LmfS0b/H6O84Uzj7LPhFmA6V4MluVoKISwSUtEeWt/sOE33IbRU+XWyeV/AwQBR shryPQFCNQIZdEJeCRZAflmuCdz8KubB4L0xE= My code page-ins are missing due to a stupid error in my analysis code. I apologize for the noise! Thanks, Joel On Nov 5, 2009, at 8:34 PM, Joel Reymont wrote: I run the script like this 'dtrace -Zqws page-fault.d'. --- 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… I'm tracing page faults in a Firefox library (XUL) with a simple DTrace script that looks for vnode page-ins [1]. This is on Snow Leopard 10.6.1, 64 bits. I clear the cache beforehand by allocating 8Gb of memory (twice physical 4Gb) and touching every page [0]. The script produces a list of page faults [2] which I try to map to segments using the output from 'size -m -l XUL' [3]. Trouble is, I don't see code page ins (__TEXT, __text) at all! When do my code pages get paged in and how can I track this? This email sent to site_archiver(a)lists.apple.com

1 0

where are my code page-ins?
by Joel Reymont 05 Nov '09

05 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=zI0Zr130pcGK0kxsCEAac8QIair1w7nfxHv2Xdmf7Lg=; b=A6cvOi9vg65G9ul5T2w//CtZjoKCi4j1WMSBisRohEZtLtcgKDbdWS9TxDJE4NQJ5Y KhE9nwr6pIrdBdYCA4pa1C5ElZN/L/K7WZz30iUK/ye6tYIo/bGLuo7cUpBOG3mMMcQU CY9gJfNcEjSagu3x25IdxeHH1VZN210HNAbow= 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=tvcJBQeVmsamjoj3bbj4CyXEI/8Xe06gKtyft/P1LnmqNBpXutZyWuRl7x3BAzJGy4 0viVMHks5DlXo0XNTdyWdGPapom1sA24okuOx5pCoTfNBxZJE5+dNbUIsQOqZDw3rjni 5E0bfxcHVsms3RNskTFU1scov6OG7juhOzuPk= I run the script like this 'dtrace -Zqws page-fault.d'. When do my code pages get paged in and how can I track this? Thanks, Joel --- [0] purge-cache.c #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; } [1] page-fault.d fbt::vnode_pagein:entry { self->v_name = stringof(((vnode_t)arg0)->v_name); } fbt::vnode_pagein:entry /self->v_name == "XUL"/ { printf("vnode_pagein: %s, offset: %u, size: %u\n", self->v_name, arg3, arg4); } [2] Output from page-fault.d vnode_pagein: XUL, offset: 19890176, size: 16384 vnode_pagein: XUL, offset: 19906560, size: 16384 vnode_pagein: XUL, offset: 19922944, size: 16384 vnode_pagein: XUL, offset: 19939328, size: 16384 vnode_pagein: XUL, offset: 19955712, size: 16384 vnode_pagein: XUL, offset: 20135936, size: 20480 vnode_pagein: XUL, offset: 20156416, size: 24576 vnode_pagein: XUL, offset: 20303872, size: 24576 vnode_pagein: XUL, offset: 20328448, size: 28672 vnode_pagein: XUL, offset: 20357120, size: 28672 vnode_pagein: XUL, offset: 20459520, size: 561152 vnode_pagein: XUL, offset: 20426752, size: 4096 vnode_pagein: XUL, offset: 20430848, size: 4096 vnode_pagein: XUL, offset: 20434944, size: 4096 vnode_pagein: XUL, offset: 20439040, size: 8192 vnode_pagein: XUL, offset: 19853312, size: 4096 vnode_pagein: XUL, offset: 20447232, size: 4096 vnode_pagein: XUL, offset: 20451328, size: 8192 vnode_pagein: XUL, offset: 14548992, size: 1048576 vnode_pagein: XUL, offset: 20385792, size: 8192 vnode_pagein: XUL, offset: 12963840, size: 1048576 vnode_pagein: XUL, offset: 8192, size: 1040384 vnode_pagein: XUL, offset: 11943936, size: 1019904 vnode_pagein: XUL, offset: 11333632, size: 610304 vnode_pagein: XUL, offset: 9469952, size: 1048576 vnode_pagein: XUL, offset: 1060864, size: 1048576 vnode_pagein: XUL, offset: 3645440, size: 1048576 vnode_pagein: XUL, offset: 4694016, size: 819200 vnode_pagein: XUL, offset: 2658304, size: 987136 vnode_pagein: XUL, offset: 5701632, size: 1048576 vnode_pagein: XUL, offset: 7528448, size: 1048576 vnode_pagein: XUL, offset: 20393984, size: 8192 vnode_pagein: XUL, offset: 20402176, size: 8192 vnode_pagein: XUL, offset: 19722240, size: 16384 vnode_pagein: XUL, offset: 19738624, size: 16384 vnode_pagein: XUL, offset: 19755008, size: 16384 vnode_pagein: XUL, offset: 19771392, size: 16384 vnode_pagein: XUL, offset: 19787776, size: 16384 vnode_pagein: XUL, offset: 19804160, size: 16384 vnode_pagein: XUL, offset: 19820544, size: 16384 vnode_pagein: XUL, offset: 19836928, size: 16384 vnode_pagein: XUL, offset: 19857408, size: 16384 vnode_pagein: XUL, offset: 19873792, size: 16384 vnode_pagein: XUL, offset: 20410368, size: 16384 vnode_pagein: XUL, offset: 19972096, size: 20480 vnode_pagein: XUL, offset: 19992576, size: 20480 vnode_pagein: XUL, offset: 20013056, size: 20480 vnode_pagein: XUL, offset: 20033536, size: 20480 vnode_pagein: XUL, offset: 20054016, size: 20480 vnode_pagein: XUL, offset: 20074496, size: 20480 vnode_pagein: XUL, offset: 20094976, size: 20480 vnode_pagein: XUL, offset: 20115456, size: 20480 vnode_pagein: XUL, offset: 20180992, size: 24576 vnode_pagein: XUL, offset: 20205568, size: 24576 vnode_pagein: XUL, offset: 20230144, size: 24576 vnode_pagein: XUL, offset: 20254720, size: 24576 vnode_pagein: XUL, offset: 20279296, size: 24576 vnode_pagein: XUL, offset: 14012416, size: 536576 vnode_pagein: XUL, offset: 10518528, size: 815104 vnode_pagein: XUL, offset: 8921088, size: 548864 vnode_pagein: XUL, offset: 2109440, size: 532480 vnode_pagein: XUL, offset: 5513216, size: 188416 vnode_pagein: XUL, offset: 6934528, size: 593920 vnode_pagein: XUL, offset: 6750208, size: 184320 vnode_pagein: XUL, offset: 2641920, size: 16384 vnode_pagein: XUL, offset: 1048576, size: 12288 [3] Segments and sections according to 'size -m -l XUL' Segment __TEXT: 18673664 (vmaddr 0x0 fileoff 0) Section __text: 13477360 (addr 0x1470 offset 5232) Section __symbol_stub1: 9234 (addr 0xcdba60 offset 13482592) Section __stub_helper: 27718 (addr 0xcdde72 offset 13491826) Section __const: 1119212 (addr 0xce4ac0 offset 13519552) Section __cstring: 436667 (addr 0xdf5eb0 offset 14638768) Section __dof_mozilla: 2316 (addr 0xe60870 offset 15075440) Section __gcc_except_tab: 72218 (addr 0xe6117c offset 15077756) Section __unwind_info: 224100 (addr 0xe72bbc offset 15150012) Section __eh_frame: 3299552 (addr 0xea9720 offset 15374112) total 18668377 Segment __DATA: 1814528 (vmaddr 0x11cf000 fileoff 18673664) Section __nl_symbol_ptr: 1136 (addr 0x11cf000 offset 18673664) Section __la_symbol_ptr: 12312 (addr 0x11cf470 offset 18674800) Section __dyld: 16 (addr 0x11d2488 offset 18687112) Section __mod_init_func: 1504 (addr 0x11d2498 offset 18687128) Section __const: 1567536 (addr 0x11d2a80 offset 18688640) Section __cfstring: 3232 (addr 0x13515b0 offset 20256176) Section __objc_data: 1760 (addr 0x1352250 offset 20259408) Section __objc_msgrefs: 10208 (addr 0x1352930 offset 20261168) Section __objc_selrefs: 368 (addr 0x1355110 offset 20271376) Section __objc_classrefs: 1528 (addr 0x1355280 offset 20271744) Section __objc_superrefs: 160 (addr 0x1355878 offset 20273272) Section __objc_const: 12912 (addr 0x1355918 offset 20273432) Section __objc_classlist: 176 (addr 0x1358b88 offset 20286344) Section __objc_catlist: 56 (addr 0x1358c38 offset 20286520) Section __objc_protolist: 24 (addr 0x1358c70 offset 20286576) Section __objc_imageinfo: 8 (addr 0x1358c88 offset 20286600) Section __data: 106632 (addr 0x1358ca0 offset 20286624) Section __bss: 61412 (addr 0x1372d40 offset 0) Section __common: 29384 (addr 0x1381d40 offset 0) total 1810364 Segment __LINKEDIT: 19382272 (vmaddr 0x138a000 fileoff 20393984) total 39870464 [4] Mapping of page-ins to segments segment: __TEXT, size: 18673664, offset: 0 segment: __DATA, size: 1814528, offset: 18673664 segment: __LINKEDIT, size: 19382272, offset: 20393984 __DATA, __cfstring = 5 pages __DATA, __cfstring = 5 pages __DATA, __cfstring = 5 pages __DATA, __cfstring = 5 pages __DATA, __cfstring = 6 pages __DATA, __cfstring = 6 pages __DATA, __cfstring = 6 pages __DATA, __cfstring = 6 pages __DATA, __cfstring = 6 pages __DATA, __objc_classlist = 6 pages __LINKEDIT, __symtab = 6 pages __LINKEDIT, __symtab = 7 pages __LINKEDIT, __symtab = 7 pages __LINKEDIT, __symtab = 2 pages __LINKEDIT, __symtab = 2 pages __DATA, __cfstring = 4 pages __DATA, __cfstring = 4 pages __DATA, __cfstring = 4 pages __DATA, __cfstring = 4 pages __DATA, __cfstring = 4 pages __DATA, __cfstring = 4 pages __DATA, __cfstring = 4 pages __DATA, __cfstring = 4 pages __DATA, __cfstring = 4 pages __DATA, __cfstring = 4 pages __DATA, __cfstring = 4 pages __DATA, __cfstring = 4 pages __DATA, __cfstring = 4 pages __DATA, __cfstring = 4 pages __DATA, __cfstring = 4 pages __LINKEDIT, __symtab = 4 pages __DATA, __cfstring = 5 pages __DATA, __cfstring = 5 pages __DATA, __cfstring = 5 pages __DATA, __cfstring = 5 pages __DATA, __cfstring = 5 pages __LINKEDIT, __symtab = 8 pages __TEXT, __cstring = 256 pages __LINKEDIT, __symtab = 128 pages __LINKEDIT, __symtab = 1 pages __LINKEDIT, __symtab = 1 pages __LINKEDIT, __symtab = 1 pages __LINKEDIT, __symtab = 2 pages __DATA, __cfstring = 1 pages __LINKEDIT, __symtab = 1 pages __LINKEDIT, __symtab = 2 pages __LINKEDIT, __symtab = 2 pages __TEXT, __symbol_stub1 = 256 pages __LINKEDIT, __symtab = 9 pages __TEXT, __symbol_stub1 = 254 pages __TEXT, __cstring = 131 pages __TEXT, __symbol_stub1 = 249 pages __TEXT, __symbol_stub1 = 256 pages __TEXT, __symbol_stub1 = 134 pages __TEXT, __symbol_stub1 = 130 pages __TEXT, __symbol_stub1 = 149 pages __TEXT, __symbol_stub1 = 256 pages __TEXT, __symbol_stub1 = 256 pages __TEXT, __symbol_stub1 = 199 pages __TEXT, __symbol_stub1 = 200 pages __TEXT, __symbol_stub1 = 241 pages __TEXT, __symbol_stub1 = 256 pages __TEXT, __symbol_stub1 = 256 pages __TEXT, __symbol_stub1 = 46 pages __TEXT, __symbol_stub1 = 145 pages __TEXT, __symbol_stub1 = 45 pages __TEXT, __symbol_stub1 = 4 pages __TEXT, __symbol_stub1 = 3 pages --- 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… I'm tracing page faults in a Firefox library (XUL) with a simple DTrace script that looks for vnode page-ins [1]. This is on Snow Leopard 10.6.1, 64 bits. I clear the cache beforehand by allocating 8Gb of memory (twice physical 4Gb) and touching every page [0]. The script produces a list of page faults [2] which I try to map to segments using the output from 'size -m -l XUL' [3]. Trouble is, I don't see code page ins (__TEXT, __text) at all! 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: file offset of __DATA, __bss and __common
by Joel Reymont 05 Nov '09

05 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=hMuc4ELC0yl/IWRNZPXzYvg3r3Y7mvqilT5zuOlCr90=; b=F/4yri4omBEOGDMEZCzd4515vqdJQB/t4j8DmP4OvrMgAdBN8c8jsHTvTlGurYP01G a7lKEq9fjevnd+ewRhEpFcoYGxB14eKAd2qOypqfOQhdBWvyN3XH2KDZ3GNbKQQZ57mD ygphaiqlFiXiqpiejqfsr96Ci47EhllKBLotg= 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=uxUr8moIAQc8jvKIImaaGA6GLxkUOJCa2tGvcfb6zVir15wfqv154FGLppFFZO+Yqw T8wenhUTC7kHv3UgKDf9emtI7zOJK/5YmTxxG1jN1Zec8E/qmR0whG2UZNzAMFlc70ay zD0RceiUdaB6s5OJN7L0aWM1Exl3cJqvBfIgw= On Nov 5, 2009, at 4:40 PM, Shantonu Sen wrote: --- 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… Yes, they do answer your question. Did you follow both pieces of advise? You would have seen the S_ZEROFILL section attribute with otool -lv, which you then could have looked up in the Mach-O file format reference. Mea culpa. 'Uninitialized static variables' was staring me in the face but I totally missed it! This email sent to site_archiver(a)lists.apple.com

1 0

Re: file offset of __DATA, __bss and __common
by Shantonu Sen 05 Nov '09

05 Nov '09

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: file offset of __DATA, __bss and __common
by Joel Reymont 05 Nov '09

05 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=/8Lmd/t1W5wsNag2MclFTWNM0zHiHbh78JeIwh+tsKk=; b=tR+Fx7T/5oO8pb4L2g+7l0J/O1QxMHdYrf14sUu9wLKrZD9FakSZ+7HdWwEOZqM3A1 9oEmAWKSc5m+amOGzsrGw+RjBvFNoeB9BTu9eMRvp0aktpOYBoNu9Djy7SxdG6P+Jknt KCYowiv1UgcGd1RDDTy/YlKBPq/xELfuCuXcM= 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=Ury26wGguiUweCYcKQi8S9grgur1zvAhMd29tA57e0vOrZDsdk2qn/kGMFUeEgYnNZ bLqYyANiBKedV784VHqcj2SvCmhS4mC5vi/z5c6h+Z9ECNN+xFH7VzMwSRH23Z4caSvP gkGWdKSiE8AtWyixFHbg2VqhHSGOtkl8DXjWc= Empirically, it looks like __DATA, __data = __bss + __common. Thanks, Joel P.S. Here's a pointer of my own (wrote this a while ago): http://tinyco.de/2009/01/26/creating-mac-binaries-on-any-platform.html On Nov 5, 2009, at 4:21 PM, Shantonu Sen wrote: Shantonu Sen ssen(a)apple.com Sent from my Mac Pro On Nov 5, 2009, at 7:57 AM, Rustam Muginov wrote: --- 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… While I much appreciate the pointers so far, they don't answer my question. <http://developer.apple.com/mac/library/documentation/DeveloperTools/Concept… > Had you tried "otool -lV"? This email sent to site_archiver(a)lists.apple.com

1 0