August 2006 - Darwin-dev - lists.apple.com

Re: Noob Darwin Question
by Markus Hitter 10 Aug '06

10 Aug '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Am 10.08.2006 um 21:14 schrieb TjL: I have an old PC (Windows 98 generation) which is currently running FreeBSD. Would it be possible to download and compile the sources for Darwin and run it on that machine? Is there an FAQ somewhere which explains how to install this on a generic computer? For the older releases, see OpenDarwin. Markus - - - - - - - - - - - - - - - - - - - Dipl. Ing. Markus Hitter http://www.jump-ing.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… You can use the OpenDarwin 7.2.3 release, corresponding to Mac OS X 10.3.2. This is the last distributed release compiled outside Apple. Then there's a Darwin 8.0.1 release available from Apple Darwin, requiring SSE2 extensions in your processor. One can get pretty far with compiling an own full OS, see release.opendarwin.org. Nobody picked up the stuff to make a distribution CD, though. Regarding the "re-opended" i386 sources at Mac OS Forge I've yet to hear something. The i386 part of the kernel sources changed drastically with the advent of i386-Macs. The PPC kernel was still buildable for Intel, but the result wasn't compatible with the kernel coming with Mac OS X-Intel. Mac OS Forge is pretty much in the beginnings right now, so you have to couple at least four things: Get the build environment (HowTo's, darwinbuild) from OpenDarwin. Get the sources from Apple Darwin. Get patches from release.opendarwin.org as a lot of packages won't build out of the box. Get the new kernel (still named "xnu") from Mac OS Forge. This email sent to site_archiver(a)lists.apple.com

1 0

SSH issue on MacMini (Intel)
by Jeffrey Shawn Klotz 10 Aug '06

10 Aug '06

site_archiver(a)lists.apple.com Delivered-To: Darwin-dev(a)lists.apple.com Organization: Cyber Tech Now User-agent: Mozilla Thunderbird 1.0.7 (Macintosh/20050923) _______________________________________________ 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: Maximum number of syscalls per thread?
by Michael Smith 10 Aug '06

10 Aug '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On Aug 10, 2006, at 12:02 PM, darwin-dev-request(a)lists.apple.com wrote: Just a quick question! I'm running an app which is used for testing i/o. Right now, I'm running a thread at an elevated priority which does a lot of pread() calls. The app works fine but is a little slow & right now, I'm fine-tuning it. One issue, though; when I fire up BigTop, find my process and look at the system call throughput, I notice that it appears to be capped at 9000 calls/sec. I'm guessing this is being throttled by a kernel parameter. Your guess (with regards to a limit on the syscall rate) is incorrect. I have observed a single-process syscall rate over 650,000/sec on an intel Mini, so your 9,000 is barely ticking over. The device I'm hitting is more than capable of beating that, so I'm definitely not saturating the driver. Thoughts / hints / ridicule welcome :-) This sentence is logically invalid. The *device* may well be capable of more than 9,000 IOTs/sec, but you haven't established anything at all regarding the driver here. Have you actually examined the interaction between driver and device to see what sort of utilisation you're seeing? Without more details of your test and driver architecture it's hard to offer you more suggestions. You are aware that the clustering layer and buffer cache will both affect the relationship between system calls and I/Os? Have you exercised your driver to see what it's maximal work rate is? Which system is your device in? How is it connected? Whose driver, and how is that driver architected? = Mike _______________________________________________ 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

Noob Darwin Question
by TjL 10 Aug '06

10 Aug '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Domainkey-signature: a=rsa-sha1; q=dns; c=nofws; s=beta; d=gmail.com; h=received:message-id:date:from:to:subject:mime-version:content-type:content-transfer-encoding:content-disposition; b=lWI0yCWBq+yabyKSTM+/7aom+TcYuIMFyZ4SL1ONiDmfrztBrPeCWHNTHu/ywWsaOU2O2tm18aeNg1nb50af+fhBqk8cx+ythJ9z666AmCRr3rEqedAkcEG5MToJ58TMS4YjglNHcwMoCgCnE9CxYwW/zY0K7QyNOjMwsmy4yL0= I read about the Intel core being opened or re-opened or whatever and it rekindled my interest in the project. I have an old PC (Windows 98 generation) which is currently running FreeBSD. Would it be possible to download and compile the sources for Darwin and run it on that machine? Is there an FAQ somewhere which explains how to install this on a generic computer? _______________________________________________ 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: Intel XNU bug report
by almisr 10 Aug '06

10 Aug '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Yep, it was an abuse language ( for the first part ) as I answered Eric I did not think that would create such a disorder :) Yep , you did quite a bit of the early integration of ACPI into FreeBSD, Both of these were undertaken by great folks at Apple then thanks a lot to great Apple folks :) but that does not change my background-comment if such a stupid code was present I believe as nobody would have missed it ( great Apple folks included ) and OSX would not work ... it was just a little "irronic" Best Regards , {|} _______________________________________________ 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 not certain which Mike you're referring to, but if you mean me, I should point out that while I was involved with the effort to bring ACPI to FreeBSD, I have been no more than a spectator with regards to both its integration into Mac OS X and the 64-bit support work. Both of these were undertaken by great folks at Apple, and it would be grossly dishonest for me to accept any sort of credit for their hard work. = Mike ACPI is large API, and requires a large work, one does not contribute like that ... I do not have the practice ( on other projects ) to see conclusions also fast and alien related BUGS. This email sent to site_archiver(a)lists.apple.com

1 0

Re: Intel XNU bug report
by Michael Smith 10 Aug '06

10 Aug '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On Aug 9, 2006, at 12:02 PM, darwin-dev-request(a)lists.apple.com wrote: Mike has implemented ACPI 2 standard in IA-64 for Darwin, Mike and Mitsuru and Nate are the ACPI-project on freeBSD = Mike _______________________________________________ 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… so i think if there were such "stupid-code" Mike couldn't miss it I'm not certain which Mike you're referring to, but if you mean me, I should point out that while I was involved with the effort to bring ACPI to FreeBSD, I have been no more than a spectator with regards to both its integration into Mac OS X and the 64-bit support work. Both of these were undertaken by great folks at Apple, and it would be grossly dishonest for me to accept any sort of credit for their hard work. This email sent to site_archiver(a)lists.apple.com

1 0

[quasi-OT] Amit Singh in SF
by Graham J Lee 10 Aug '06

10 Aug '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Hi, Cheers, -- Graham Lee "Oxford University's UNIX Expert" Quick, run before they find the truth http://iamleeg.blogspot.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… just observed in the Apple Store SF that Amit Singh (Mac OS X Internals) is talking there[*] tomorrow at 12:00, should be interesting to any fellow WWDC attendees. Graham. [*]You wonder whether anyone from Retail knows about the rest of Apple; "oh yes, I'll be in SF anyway that week" "really? Coming here on holiday are you?" This email sent to site_archiver(a)lists.apple.com

1 0

Maximum number of syscalls per thread?
by Alison Cassidy 09 Aug '06

09 Aug '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Hi all, Best regards, Allie _______________________________________________ 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… Just a quick question! I'm running an app which is used for testing i/o. Right now, I'm running a thread at an elevated priority which does a lot of pread() calls. The app works fine but is a little slow & right now, I'm fine-tuning it. One issue, though; when I fire up BigTop, find my process and look at the system call throughput, I notice that it appears to be capped at 9000 calls/sec. I'm guessing this is being throttled by a kernel parameter. Is there any way this can be modified? I can understand why there is an upper limit on something like that, but I don't mind if my particular app/thread dominates the machine. The device I'm hitting is more than capable of beating that, so I'm definitely not saturating the driver. Thoughts / hints / ridicule welcome :-) This email sent to site_archiver(a)lists.apple.com

1 0

Re: Intel XNU bug report
by Kevin Van Vechten 09 Aug '06

09 Aug '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On Aug 9, 2006, at 1:02 PM, Graham J Lee wrote: On 9 Aug 2006, at 12:53, Chad Leigh wrote: - Kevin _______________________________________________ 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… Would be nice if Darwin/Apple etc used a name that is already in use by others so a to not offer more confusion. how about EM64T, or "washing machine controller chip with a few extensions"? Seriously, AMD invented this stuff and they call it x86-64, so I think that's a reasonable decision and already in use. i38664 would have been wrong. NetBSD, RedHat/Fedora/Centos use x86_64. The kernel and developer tools have standardized on "x86_64" for the name of the Mach-O architecture. Using x86_64 is probably the most straightforward way to describe 64-bit on Intel as supported by Mac OS X. This email sent to site_archiver(a)lists.apple.com

1 0

Re: Intel XNU bug report
by Graham J Lee 09 Aug '06

09 Aug '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On 9 Aug 2006, at 12:53, Chad Leigh wrote: On Aug 8, 2006, at 9:09 PM, Eric Albert wrote: On Aug 8, 2006, at 1:55 PM, almisr wrote: IA-32-64 would be appropriately unconfusifying, then ;-) -- Graham Lee "Oxford University's UNIX Expert" Quick, run before they find the truth http://iamleeg.blogspot.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… Mike has implemented ACPI 2 standard in IA-64 for Darwin, By the way, IA-64 is the name Intel uses for Itanium, which Darwin does not support. We've been using "Intel 64-bit" or "x86-64" for our latest architecture. Sun I think uses x64 which is kind of a nice shorthand for x86-64. A lot of people like FreeBSD use amd64 since AMD fathered the particular 64bit architecture which intel copied. Would be nice if Darwin/Apple etc used a name that is already in use by others so a to not offer more confusion. how about EM64T, or "washing machine controller chip with a few extensions"? Seriously, AMD invented this stuff and they call it x86-64, so I think that's a reasonable decision and already in use. i38664 would have been wrong. NetBSD, RedHat/Fedora/Centos use x86_64. But IA-64 is Itanium in most people's mind, that I agree with. This email sent to site_archiver(a)lists.apple.com

1 0