http://www.opensource.apple.com/darwinsource/10.4.9.x86
http://www.opensource.apple.com/darwinsource/10.4.9.ppc

http://digg.com/apple/Darwin_Source_Code_R...4_9_x86_and_PPC
Digg if you can.

This means new kernels!

yay!

Got it and having a look, hopefully tonight we will have a 10.4.9 SSE3 Speedstep kernel

Awesome!

Have kernel booting on my PC now, adding autofsb from daemones when i can be bothered next up is efi patch and then speedstep

hopefully someone will fix tsc problems on amd dualcore

is that the Core syncing issue? where you stutter and stuff?

yep also. but all works good on nd 8.8.1 so i guess it's easy to fix on 8.9.1 too

nd was semthex's latest sse3 kernels , i dont have source for that, did it work on mifki/daemones's kernel?

hmm never used both, just made in semthexland. i'll try when i'll back home later

Kiko: this might be an odd question, but Dell BIOS hates OS X when it comes to using both cores. Is this something that can be fixed by making a special kernel for us Dell BIOS users? Or is it impossible?

woops. lord_muad_dib, sorry i just realised mifki is intel only , so no point. gburnham, thats a timing issue as well, i dont know how to fix it, or even where to start. But it should be impossible

EDIT: Try daemones's kernel anyway, im confused now, as his sources have the 0xfb interrupt and some cpuid code

Ok, 10.4.9 Intel SSE3/Speedstep is functional, thanks to daemonES, mifki and paulicat for the patches.

SSE2 (later tonight, have latest version now) and amd coming up later, get kernel and throttle kext on moofspeak in a couple of hours (once i tweak it a bit more)

EDIT: Sleep Works fine for me

It works here as well

Speed Step works fine...Although sleep does not

Does anyone here know how to re-poll/re-connect USB devices via the terminal, because if so , then tha would make alot more hacks sleep and wake perfect (instead of having to unplug the usb mouse to make it work again )

I know how.
You can contact me on IRC.

LE: not applicable to kernel, kernel functions needed instead

Send me a PM, as im going off to bed i am tired, i'll do AMD tommorow, sse2 needs more work, i have the emulator embedded but have a small booting problem.
Apart from that, the Intel one on the irc channel (#kernel on irc.moofspeak.net) is working pretty much perfect, just have to tweak a cpuid setting on that and it should be final or close to it, then i incorporate the amd and vmware patches. Fix SSE2 then release the source.

Kiko, what does the efi patch?

efi patch is only to allow GUI, efi to rom basically, and to make efi_init(); return early.

10.4.9 SSE3 available for download wherever dev's are found. SSE2 in the works.

All working here

Click to view attachment

For those who want to test speedstep sse3 10.4.9 its on moofspeak, AMD coming tonite hopefully with sse2

all working fast with no stuttering at all!

Hi, kiko
Have you able to recomplie the cputhrottle.kext with no debug output?

or even had you have time to see if manual setting the multiple states is possible? (like we discussed long time ago?)

I made a copy of the kext with no debug output, i'll upload it here, and i havent found custom states yet, but i know it can be done. im a bit busy on trying to get sse2 to work (as im not a nice person, i dont get the code ).

Just got the latest kernel, and someone doesn't like you too much Kiko:

Click to view attachment

HAHAHHHAHA, man that is lame lol, netkas must have done it, how can i tell? he cant speak ze engleski

I second that, is there anyway possible for us Dell Bios users to get two cores working rather one ?

Well, does the wake from sleep method fix your cores?

SSE2 where the developers are

lucky bastards with source

Amazing rollout speed! Thanks all and please keep 'em come'in.

This is just hilarious. Kiko, you don't have the SSE2 emulation dump. And even if you did, you wouldn't know what to do with it. There is a reason that you aren't on the trusted list...



Hope you have fun with all this David

The whole idea that a trusted list of people only have access to source code modifications shows just how damaged the OSX community is. While Semthex did a lot of good for the advancement of OSX he also did an equal amount of damage fragmenting the community. How about just get all the code changes public and let anyone work on the code. Move back to the concept of OPEN SOURCE.

Getting this back on track, the Semthex SSE3 emulation was taken from Maxxuss's binary kernel and converted back into source. There is no reason the same can't be done again by Kiko if he has the capability. Good luck to him in his endevour to do so. He certainly doesn't deserve being insulted by other people for what he is giving freely back to the community.

Maxxuss' SSE2 emu was only for three instructions. Semthex and another dev worked on it and it is now only missing a small handful of instructions.

If you don't recall, the source and a diff was released back in December. All that's changed from then to now are trivial. 64-bit was reimplemented from stock Apple source and tweaked ever so slightly. That's it.

I know Rufus was the one who found out that the missing instruction was causing a problem with iTunes. I am not aware of how many instructions they needed to add. But I do know they extracted the SSE3 emulator from Maxxuss's binary kernel and then converted it back to source and expanded on it.

The source code released was only for the SSE3 kernel. SSE2 kernel source was not released which meant the SSE3 emulator was still unavailable for most developers (unless you were close friends to Semthex).

The other big change was the addition of speedstep by Paulicat. Of course since he didn't have access to SSE2 kernel source he couldn't do the changes for that kernel and Semthex's recent attempt did not work correctly. Now that he has retired there is a lot of uncertainty over who will continue working on it.

And now bikedude made another sse2 speedstep kernel, and i dont think that works either, the diff released for semthex in december isnt easy to be found, i havent seen a copy in ages, and when anyone asks for it, they get treated rudely or just shrugged off, so much for ASPL compliant . Anyway i do have the dump, but i dont know what to do with it yet, and gloating over the fact that you do and still cant make a working kernel isnt very smart. Anyway, you wouldnt have known what to do with it if semthex hadnt told you, anyway, heres something funny about bikedude i found edited . I better watch it, he might sue me


Anyway, at least we are on first name basis's now edited , i dont think Christian would be very happy about all this ke-fuffle that is happening nowadays.

haha, you catch me.
Пидар ты овсталийский.
next time i'll do better


and about sources and apsl. It's where I'll beat you
to ask for sources, you need to find developer...but...who did that...officialy - nobody knows
so, nobody you can ask for sources
ask your ass

So you are gloating about keeping the source for yourself? Well now i think we all know what sort of person you are, I'm done with this thread and these petty arguements, i'll let the rest of the community make up thier minds.
This thread is about 10.4.9 source being released, and it, like the semthex retiring thread has gone way off topic, imo it should be closed

any updates on the progress of sse2 yet? or is it still a work in progress?

Its a work in progress for me, the people who have the source have already made a sse2 kernel

i wonder if they've used some of the aTV OS or got some hints for their sse2 kernel. anyway it's always nice to see community based on a hacked OS thriving and that the open source scene is alive and well...and by alive and well i mean dead. (perhaps I just alienated myself?) goodluck with your work kiko and everyone else working on a sse2 kernel

Kiko, do not try to look like a good boy.
you are liar.

netkas, why these attacks. Im not trying to look like anything, i'll let people judge me for what they think i am, if they think im a liar (for whatever reasons) then they are entitled to their opinion.

p.s: learn how to actually call people names before you do, its not "you are liar" its "you are a liar"
p.p.s: w00t 1800 posts
p.p.p.s: sse2 is still in progress

Learn russLearn russian, only then try to learn me english
It's only my SECOND language.
Пидар ты, ебать тебя долго, ебать к верху жопой© Джей и молчаливый боб.

For all these people bitching about not having the source to SSE2 emulation......THERE IS NO F***ING SOURCE....the sse2 emulation was all done in assembly. And all instructions except for MWAIT and MONITOR are FULLY emulated. and guess what, MWAIT and MONITOR are useless (no one uses em). So the SSE2 emulation written by Rufus (and Semthex) is complete, and its not available as a simple little c file you can plug into your project. TIt is a pure binary hack/patch.

And if anyone is like, well, who are you SaberSHO to know all this.....I was Rufus and Semthex main beta tester for the SSE2 kernel. I still have 20+ incremental beta kernels that go from having 2 instructions completed to all of them completed. I spent hours and days on IRC with rufus testing out every single little change...so please...i know what im talking about.

Again, to reiterate...there is no source! so quit bitching about it.

it doesnt have to be C++ to be source x-D. source can be a binary blob like it is now. thats enough source

P.S: ASM is also source

to be honest I don't care who has the source or if it exists at all, I would just like an improved SSE2 kernel. If it came from kiko first, fine or from semthex, that's alright too. Do you think the majority of people on this forum really truly care that their kernel is a genuine semthex kernel or that they are just glad to have a working kernel and system? I don't mean to sound ungrateful because I'm not, I appreciate all the work done by everyone here but this is all getting very juvenile.

Well, im sure Kiko will be releasing a SSE2 patch very soon, since he + co-conspirators decided that writing one themseleves was way harder then hacking a IRC server and just stealing other peoples hard work. And people wonder why no one wanted to work with Kiko, well, there you are, he is a thief. Even if you werent happy that the source was being kept from some people, do you really think that stealing it and causing havoc on the IRC network was going to fix anything? Anyway, everything is equal now, so can this childish {censored} stop now? please? for the love of all that is good about osx86?

Ah, so someone else sees semthex as a major contributor to the segragation of the x86 project and the lack of access to the source if your not on the special "trusted list".

This aspect of the x86 project should have been available to anyone who is interested in building their own kernel (especially since semthex is too stupid to get the DSTROOT right not to mention all that retarded printing/logging/debugging information).

I've managed to obtain the semthex source and the old partial DaemonES source and have worked with another user to apply the diff and make minor changes to the 10.4.9 source to get an acceptable result.

The kernel has been tested with Intel and AMD and seems to work without any issues however it does report the CPU as unknown in the "About this Mac" window (the semthex kernel reports it this way as well)

I have been after DaemonES since November 2006 to get his 10.4.8 cpu identification changes, he stated that he'd give to me but has since been ignoring my requests.

I have everythig down to the php scripts that semthex uses to perform the sse3 emulation bin patching and have no reasons not to pass it along to anyone who wants it and I don't believe in the semthex dog and pony show or the special "trusted list" so availability is based on your desire to obtain the source and not someones belief that you might deserve it.




On another note, it's very possible to modify firmware on an intel motherboard to boot from efi and swap out some efi files so you can natively boot and install from an unmodified install DVD (boots in about 35-45 seconds).

A user on this forum was given a preview during a video chat so he can confirm that the boot occured and that the machine in question was not an Apple machine.

Now that I have successfully been able to modify one particular motherboard I'll be moving on to other motherboards in my possession as time permits and eventually I'll have a collection of update BIOSes I'll be putting on a CD and releasing for anyone interested in experimenting with a modified firmware rather than experimenting with modified installations.

I'm only focusing on Intel manufactured motherboards and have no intentions of performing any work on other brands since I don't own anything but Intel so don't ask me if I plan on making one for your brand.


I've included a utility that will detect your CPU and provide some detialed information so no need to install windows and run cpuid-z as there is now a native Mac tool to perform this task.

--BuildSmart