Adrian Andréas Posted January 3, 2013 Share Posted January 3, 2013 Hello, I'm Brazilian and I'm watching from the beginning of the development of ML on a PC AMD. I would love to try and post the bugs. But I do not know where to start. I can install SL and then go straight to ML? Sorry for English .. And if I'm messing up let me know! you guys are the best.... Link to comment Share on other sites More sharing options...
instant idiot Posted January 3, 2013 Share Posted January 3, 2013 Hello, I'm Brazilian and I'm watching from the beginning of the development of ML on a PC AMD. I would love to try and post the bugs. But I do not know where to start. I can install SL and then go straight to ML? Sorry for English .. And if I'm messing up let me know! you guys are the best.... Hi Adrian, welcome! In short, you'll need to have access to a working Hackintosh or a real Mac, download myHack and use it to create a 10.8 installer, replace corecrypto.kext with this one, and finally try booting with the kernels from here (mach_sse42_only seems to be working best at the moment). 1 Link to comment Share on other sites More sharing options...
Andy Vandijck Posted January 3, 2013 Share Posted January 3, 2013 I'll make a new kernel later. This one will be based more on Sinetek's patches (with a few additions of my own) I hope to fix that failed issue. This one won't boot faster than Sinetek's kernel (not adding driver load patch, I have a hunch there) Post when I've built it... Link to comment Share on other sites More sharing options...
AkimoA Posted January 3, 2013 Share Posted January 3, 2013 here a pic from a pretty good working 10.8.2 http://cl.ly/image/3D070F1d102Q see notes in picture here some GPU related stuff from Console -> http://cl.ly/1d2z3L3T461b Link to comment Share on other sites More sharing options...
theconnactic Posted January 3, 2013 Author Share Posted January 3, 2013 Finish testing all kernels. The only one that works for my Athlon II X2 250 is Andy's latest and greatest, and in safe mode only, otherwise i'll have severe graphics artifacts and no GUI at all. The curiosity is that, in safe mode, i can see in system report my graphics card properly displayed as a G210 both in PCI Cards/Graphics and Displays section. The driver is show as loaded. That indicates the kernel can load the nVidia kext, but some missing instruction (ssse3?) is making it to malfunction. The mach_sse42_only sometimes also boot in safe mode, but frequently panics. Best regards Link to comment Share on other sites More sharing options...
Deltac0 Posted January 3, 2013 Share Posted January 3, 2013 Hi! I finally made it... Installed 10.8.2 on my AMD! Okay, now the info: myHack 10.8.2 Sinetek's & AnV's kernels Patched CoreCrypto.kext AppleDecrypt.kext Removed Dont Steal Mac OS X.kext Phenom II x4 965 ATI Radeon HD5870 ASUS M4A785TD-V EVO 8GB DDR3 1333Mhz I got it booting to desktop with both kernels. AnV's kernel (with -nosse3bit) was much slower than Sinetek's (I used mach_sse42_only). AnV's kernel also got pretty bad GUI problems, for example: The whole menubar @ top was green / red gradient (sorry I didn't take a pic, it was hilarious ). Also got the FIPS error with AnV. Sinetek's kernel booted pretty fast, straight to the desktop. Still some GUI problems (mostly looks like running OS X on a GPU from 90's). It's just like some basic GFX lag. I'm ATM on my SL installation, but I'll boot back to ML (installing Lnx2Mac's ethernet) and I'll report how does it recognize the CPU / RAM / etc. Link to comment Share on other sites More sharing options...
wastez Posted January 3, 2013 Share Posted January 3, 2013 My system is working best with the mach_sse42_only. Also have graphic fragment problems. Link to comment Share on other sites More sharing options...
SS01 Posted January 3, 2013 Share Posted January 3, 2013 10.8.2 with sinetek kernel: -v npci=0x3000 System boots, major graphics issues. System unusable. 10.8.2 with AnV kernel: -v npci=0x3000 -nossse3bit System boots, minor graphics issues, after about 10 minutes of use they usually solve themselves. System usable Screenshots later If the next release could help with the GFX issues, it'd be appreciated - they're annoying haha. EDIT: oops, got the kernels switched. Sineteks has the better graphics. A fix would still be appreciated :cheeky: Link to comment Share on other sites More sharing options...
gedna Posted January 3, 2013 Share Posted January 3, 2013 hello all, iv been reading this thread for some time now, keen to try it myself, though some wonders, amd 9600 quite old, will it work? also atiata.kext thank you Link to comment Share on other sites More sharing options...
Deltac0 Posted January 3, 2013 Share Posted January 3, 2013 hello all, iv been reading this thread for some time now, keen to try it myself, though some wonders, amd 9600 quite old, will it work? also atiata.kext thank you If it works with Lion, it should work with ML too. However, you're probably going to get the same graphics problems as all of us. EDIT: I'll try to take a video about all these funny graphics bugs now EDIT2: Damn no batteries @ camera...... So I recommend you all to test yourselves! It's not really that hard... EDIT3: Sinetek's kernel seems to be pretty stable, 1h runtime now. The graphics problems seem to go away with time. Link to comment Share on other sites More sharing options...
Andy Vandijck Posted January 3, 2013 Share Posted January 3, 2013 I made the new kernel. This one does not directly load the drivers and has some of the old code removed. Also no codesig override (may fix failing FIPS). Enough talk, this needs testing Attached are the kernel and the diff It's the birthday edition (today is my birthday lol -> 03/01/1985 ) Test it and let me know how it works. The flags are the same as the other one. -nossse3bit = don't set SSSE3 supported bit (probably needed) -nosse42bit = don't set SSE4.2 supported bit (probably not needed) -nodecryptor = don't use built-in decryptor but use decryption kext Please test and report back SinetekAnV_test1.diff.zip amd_kernel_mini.zip Link to comment Share on other sites More sharing options...
SS01 Posted January 3, 2013 Share Posted January 3, 2013 Happy birthday AnV! I'll test the new kernel now. Meanwhile, I've found that setting apps to minimize with the "Scale" effect greatly reduces graphics problems. Link to comment Share on other sites More sharing options...
Shaneee Posted January 3, 2013 Share Posted January 3, 2013 I made the new kernel. This one does not directly load the drivers and has some of the old code removed. Also no codesig override (may fix failing FIPS). Enough talk, this needs testing Attached are the kernel and the diff It's the birthday edition (today is my birthday lol -> 03/01/1985 ) Test it and let me know how it works. The flags are the same as the other one. -nossse3bit = don't set SSSE3 supported bit (probably needed) -nosse42bit = don't set SSE4.2 supported bit (probably not needed) -nodecryptor = don't use built-in decryptor but use decryption kext Please test and report back First of all Happy Birthday mate xD Kernel boots using -nossse3bit as it did before but still a no go with my nVidia GT 430 FIPS integrity check still shows it Fails. Link to comment Share on other sites More sharing options...
AkimoA Posted January 3, 2013 Share Posted January 3, 2013 happy birthday m8 all the best some testing Fips Fail still there ;( like before after it succeeded booted with -nodecrypt and -nossse3bit GUI Window server gets Nuts :wink2: http://cl.ly/image/1G3V3I3s3k0Q and the minute gap macx swapon Link to comment Share on other sites More sharing options...
Andy Vandijck Posted January 3, 2013 Share Posted January 3, 2013 No FIPS fail with -nodecryptor? Link to comment Share on other sites More sharing options...
Deltac0 Posted January 3, 2013 Share Posted January 3, 2013 Fips fail, gfx issues still present (not as much as with the previous kernel). Btw, my Safari can't display letter "a". I hope this post will show them right. Can't post from Opera , it gets SIGBUS signal and shuts down instantly, same with Skype. Both work with Sinetek's kernel (well, Skype crashes but that's something else, and it isn't instant crash). Some pics xD : Btw, one of the hardest posts I've ever written on any forum. 2 Link to comment Share on other sites More sharing options...
SS01 Posted January 3, 2013 Share Posted January 3, 2013 Tested with -nossse3bit. FIPS fail. macx_swapon hangs for a minute or two. Bad graphics fractals. List of apps which are unusable because of graphics: (On sinetek's kernel) Safari Firefox Chrome half-works, a few brown screens here and there. As far as browsers go, Opera is your best bet, no issues at all. I'll keep updating this as I go on. Link to comment Share on other sites More sharing options...
Shaneee Posted January 3, 2013 Share Posted January 3, 2013 FIPS Integrity Check Fails with -nodecryptor on yes. I have your Decyptor kext installed aswell. Still get black screen on which I can type then it disappears which is weird.. Not there in safe boot or ig Graphics kexts removed. Link to comment Share on other sites More sharing options...
theconnactic Posted January 3, 2013 Author Share Posted January 3, 2013 Same as akimoa and Shane here. With Athlons, it appears to be worse than with any other model of CPU: with both your kernels, Andy, and with the one from Sinetek which work (mach_sse42_only), i can only boot in safe mode. Seems that the issue lies on the opcode emulator. Somehow the graphics kexts load but behave strangely. Best regards! Link to comment Share on other sites More sharing options...
Deltac0 Posted January 3, 2013 Share Posted January 3, 2013 FIPS Integrity Check Fails with -nodecryptor on yes. I have your Decyptor kext installed aswell. Still get black screen on which I can type then it disappears which is weird.. Not there in safe boot or ig Graphics kexts removed. Try to reset BIOS or something, it worked for me... No idea what I actually did... And I think Akimoa got it working too! EDIT: Oh, Connactic was faster... EDIT2: Maybe this helps... When applications crash, the report shows always libsystem_kernel.dylib and libdispatch.dylib. I could try patching these with amd_insn_patcher... ? Btw, I get crash always when I go to System Prefs, open the "Sound" panel and change my volume from there. EDIT3: Oh... Exception Type: EXC_BAD_ACCESS (SIGSEGV) Full crash report: Process: System Preferences [321] Path: /Applications/System Preferences.app/Contents/MacOS/System Preferences Identifier: com.apple.systempreferences Version: 12.0 (12.0) Build Info: SystemPrefsApp-232001000000000~1 Code Type: X86-64 (Native) Parent Process: launchd [128] User ID: 501 Date/Time: 2013-01-03 19:56:46.124 +0200 OS Version: Mac OS X 10.8.2 (12C60) Report Version: 10 Interval Since Last Report: 7465 sec Crashes Since Last Report: 63 Per-App Interval Since Last Report: 396 sec Per-App Crashes Since Last Report: 14 Anonymous UUID: A854201A-3C5E-B7F2-68FA-F879F011A898 Crashed Thread: 0 Dispatch queue: com.apple.main-thread Exception Type: EXC_BAD_ACCESS (SIGSEGV) Exception Codes: KERN_INVALID_ADDRESS at 0x000000000000008c VM Regions Near 0x8c: --> __TEXT 000000010a535000-000000010a559000 [ 144K] r-x/rwx SM=COW /Applications/System Preferences.app/Contents/MacOS/System Preferences Application Specific Information: com.apple.preference.sound v.6.0 (Sound) objc[321]: GC: forcing GC OFF because OBJC_DISABLE_GC is set Thread 0 Crashed:: Dispatch queue: com.apple.main-thread 0 com.apple.AppKit 0x00007fff8bf6a82f audioCompleted1WithRelease + 42 1 libdispatch.dylib 0x00007fff860550b6 _dispatch_client_callout + 8 2 libdispatch.dylib 0x00007fff8605a0c8 _dispatch_main_queue_callback_4CF + 275 3 com.apple.CoreFoundation 0x00007fff8e7430fe __CFRunLoopRun + 1614 4 com.apple.CoreFoundation 0x00007fff8e7426b2 CFRunLoopRunSpecific + 290 5 com.apple.HIToolbox 0x00007fff855a30a4 RunCurrentEventLoopInMode + 209 6 com.apple.HIToolbox 0x00007fff855a2e42 ReceiveNextEventCommon + 356 7 com.apple.HIToolbox 0x00007fff855a2cd3 BlockUntilNextEventMatchingListInMode + 62 8 com.apple.AppKit 0x00007fff8bab3613 _DPSNextEvent + 685 9 com.apple.AppKit 0x00007fff8bab2ed2 -[NSApplication nextEventMatchingMask:untilDate:inMode:dequeue:] + 128 10 com.apple.AppKit 0x00007fff8baaa283 -[NSApplication run] + 517 11 com.apple.AppKit 0x00007fff8ba4ecb6 NSApplicationMain + 869 12 libdyld.dylib 0x00007fff870db7e1 start + 1 Thread 1:: Dispatch queue: com.apple.libdispatch-manager 0 libsystem_kernel.dylib 0x00007fff8aafad16 kevent + 10 1 libdispatch.dylib 0x00007fff86057dea _dispatch_mgr_invoke + 883 2 libdispatch.dylib 0x00007fff860579ee _dispatch_mgr_thread + 54 Thread 2: 0 libsystem_kernel.dylib 0x00007fff8aafa6d6 __workq_kernreturn + 10 1 libsystem_c.dylib 0x00007fff890bdeec _pthread_workq_return + 25 2 libsystem_c.dylib 0x00007fff890bdcb3 _pthread_wqthread + 412 3 libsystem_c.dylib 0x00007fff890a8171 start_wqthread + 13 Thread 3: 0 libsystem_kernel.dylib 0x00007fff8aafa6d6 __workq_kernreturn + 10 1 libsystem_c.dylib 0x00007fff890bdeec _pthread_workq_return + 25 2 libsystem_c.dylib 0x00007fff890bdcb3 _pthread_wqthread + 412 3 libsystem_c.dylib 0x00007fff890a8171 start_wqthread + 13 Thread 4: 0 libsystem_kernel.dylib 0x00007fff8aafa6d6 __workq_kernreturn + 10 1 libsystem_c.dylib 0x00007fff890bdeec _pthread_workq_return + 25 2 libsystem_c.dylib 0x00007fff890bdcb3 _pthread_wqthread + 412 3 libsystem_c.dylib 0x00007fff890a8171 start_wqthread + 13 Thread 5: 0 libsystem_kernel.dylib 0x00007fff8aafa6d6 __workq_kernreturn + 10 1 libsystem_c.dylib 0x00007fff890bdeec _pthread_workq_return + 25 2 libsystem_c.dylib 0x00007fff890bdcb3 _pthread_wqthread + 412 3 libsystem_c.dylib 0x00007fff890a8171 start_wqthread + 13 Thread 6: 0 libsystem_kernel.dylib 0x00007fff8aafa6d6 __workq_kernreturn + 10 1 libsystem_c.dylib 0x00007fff890bdeec _pthread_workq_return + 25 2 libsystem_c.dylib 0x00007fff890bdcb3 _pthread_wqthread + 412 3 libsystem_c.dylib 0x00007fff890a8171 start_wqthread + 13 Thread 7: 0 libsystem_kernel.dylib 0x00007fff8aafa6d6 __workq_kernreturn + 10 1 libsystem_c.dylib 0x00007fff890bdeec _pthread_workq_return + 25 2 libsystem_c.dylib 0x00007fff890bdcb3 _pthread_wqthread + 412 3 libsystem_c.dylib 0x00007fff890a8171 start_wqthread + 13 Thread 0 crashed with X86 Thread State (64-bit): rax: 0x000000000000008c rbx: 0x0000000000000000 rcx: 0x00007fbb44012060 rdx: 0x00000000a1a1a1a1 rdi: 0x00007fbb44012058 rsi: 0x00007fbb44012000 rbp: 0x00007fff556c9620 rsp: 0x00007fff556c9600 r8: 0x0000000000000009 r9: 0x000000006bec1582 r10: 0x00007fbb45e5c390 r11: 0x000000002ed0d62b r12: 0x00007fbb45e4fb50 r13: 0x0000000000000000 r14: 0x00007fbb44012058 r15: 0x00007fff73ebd880 rip: 0x00007fff8bf6a82f rfl: 0x0000000000010206 cr2: 0x000000000000008c Logical CPU: 3 Link to comment Share on other sites More sharing options...
theconnactic Posted January 3, 2013 Author Share Posted January 3, 2013 Since i only can boot in safe mode, sound's not an issue here. In fact, sound it's not, period. Link to comment Share on other sites More sharing options...
Andy Vandijck Posted January 3, 2013 Share Posted January 3, 2013 Guess I'll have to use Sinetek diff with my bit set code only Also his last opemu (02/01/2012). If that works we'll drop my patch and his is the way to go... Link to comment Share on other sites More sharing options...
Deltac0 Posted January 3, 2013 Share Posted January 3, 2013 Since i only can boot in safe mode, sound's not an issue here. In fact, sound it's not, period. Yeah, same here... No sound (VIA VT1708S xD), but even without sound the thingy crashes... Link to comment Share on other sites More sharing options...
Andy Vandijck Posted January 3, 2013 Share Posted January 3, 2013 I'm compiling Sinetek's diff right now with only the 2 boot options -nossse3bit -nosse42bit I hope this one works (apart from that it's Sinetek his patch) Post when done... Here it is... Use the -nossse3bit boot flag to boot. NOTE: This has no built-in kernel decryptor... use my AppleDecrypt.kext or FakeSMC Please test it and report back A.S.A.P please sinetek_patch_new.diff.zip sinetek_kernel.zip 1 Link to comment Share on other sites More sharing options...
AkimoA Posted January 3, 2013 Share Posted January 3, 2013 Fips Fail still Present or lets say ported over rest behaves pretty much the same noticed also the last build didnt panic at shutdown but the mini before did Link to comment Share on other sites More sharing options...
Recommended Posts