Jump to content


  • Content Count

  • Joined

  • Last visited

About Drag0nFly

  • Rank
    InsanelyMac Protégé
  1. Drag0nFly

    VoodooHDA with fix for random KP

    Made a typo in my last post–the audio codec for the machine which produces static hiss when using 24 or 32 bits per sample is ALC662, not ALC862.
  2. @Zenith432–sadly this USB3 driver does not work as well as your VoodooHDA branch (2.8.2d6) on my two systems (one Sandy Bridge Shuttle SH67H7, the other an Ivy Bridge Shuttle SH61V). Both systems are running 10.7.5. When loading the kext, it does not register peripherals plugged into the USB3 ports, nor log anything to the syslog (unsure if it should). I used the Caldigit driver on 10.7.4, but as we all know 10.7.5 breaks it. Some debugging output (happy to provide more, as I'd certainly like USB3 to be functioning on these boxes)– $ ./xhcdump running: Found a device of class GenericUSBXHCI: IOService:/AppleACPIPlatformExpert/PCI0@0/AppleACPIPCI/RP03@1C,2/IOPCI2PCIBridge/PXSX@0/GenericUSBXHCI USBCmd RS Y HCRST N INTE Y HSEE N LHCRST N CSS N CRS N EWE Y EU3S N USBSts HCH N HSE N EINT Y PCD N SSS N RSS N SRE N CNR N HCE N DNCtrl 0xffff CRCr CRR N Config 32 MFIndex 9380 Last Time Sync xHC 0 milliseconds <-> CPU 0 nanoseconds # Configured Endpoints 0 # Interrupts: Total 0, Serviced 0, Inactive 0, Offline 0 Port 1 PortSC CCS N PED N OCA N PR N PLS RxDetect PP Y Speed Unknown PIC Off LWS N CSC N PEC N WRC N OCC N PRC N PLC N CEC N CAS N WCE Y WDE Y WOE Y DR N WPR N PortPmsc U1 0 U2 0 FLA N PortLi LEC 0 Port 2 PortSC CCS N PED N OCA N PR N PLS RxDetect PP Y Speed Unknown PIC Off LWS N CSC N PEC N WRC N OCC N PRC N PLC N CEC N CAS N WCE Y WDE Y WOE Y DR N WPR N PortPmsc U1 0 U2 0 FLA N PortLi LEC 0 Port 3 PortSC CCS N PED N OCA N PR N PLS RxDetect PP Y Speed Unknown PIC Off LWS N CSC N PEC N WRC N OCC N PRC N PLC N CEC N CAS N WCE Y WDE Y WOE Y DR N WPR N PortPmsc L1S Invalid RWE N HIRD 50 us L1Slot 0 HLE N TestMode Disabled Port 4 PortSC CCS N PED N OCA N PR N PLS RxDetect PP Y Speed Unknown PIC Off LWS N CSC N PEC N WRC N OCC N PRC N PLC N CEC N CAS N WCE Y WDE Y WOE Y DR N WPR N PortPmsc L1S Invalid RWE N HIRD 50 us L1Slot 0 HLE N TestMode Disabled Interrupter 0 iman IP Y imod I 40000 ns C 0 ns erstsz 1 erdp DESI 0 EHB Y $ ./xhcdump caps: Found a device of class GenericUSBXHCI: IOService:/AppleACPIPlatformExpert/PCI0@0/AppleACPIPCI/RP03@1C,2/IOPCI2PCIBridge/PXSX@0/GenericUSBXHCI Kext Version 1.2.4 Vendor 0x1b21, Device 0x1042, Revision 0 CapLength 32 HCIVersion 0x96 MaxSlots 32, MaxIntrs 8, Rsvd(1) 0, MaxPorts 4 IST 1 microframes, ERST Max 32768, Rsvd(2) 0x17, SPR N, Max Scratchpad Bufs 0 U1 Device Exit Latency 0, Rsvd(3) 0, U2 Device Exit Latency 0 AC64 N, BNC N, CSZ N, PPC N, PIND N, LHRC N, LTC N, NSS Y, Rsvd(4) 0x1, MaxPSASize 65536 xHC Extended Cap ID 1, Specific 0 Legacy CTLSTS 0xc0010000 xHC Extended Cap ID 2, Specific 0x300 Supported Protocol Name 0x20425355 PortOffset 1 PortCount 2 ProtocolDefined 0 xHC Extended Cap ID 2, Specific 0x200 Supported Protocol Name 0x20425355 PortOffset 3 PortCount 2 ProtocolDefined 0x1 DBOff 0x1800 RTSOff 0x1000 PageSize 4096 Using MSI
  3. Drag0nFly

    VoodooHDA with fix for random KP

    A follow-up to my previous comment regarding multichannel 5.1 (bitstream/encoded) audio: this now works perfectly on both systems I am currently using (with ALC888 & ALC862); so the information I provided earlier was incorrect. Your driver is also stable as hell–I have yet (after probably 40 reboots and a system running ~50 days) experienced any crashes for which the VoodooHDA driver was the culprit. *knock wood* I've also been running the latest XBMC (12.2) using your driver and have so far not experienced any issues at all (using DD5.1 / DTS5.1 & PCM output), audio sync is perfect & no audio dropouts. The only real problem I've encountered with the driver appears on my Ivy Bridge system, which has the ALC862 chip: static is heard if using 24-bit or 32-bit sampling width. The audio is being heard faintly in the background, and the static disappears if switching to 16-bit in Audio MIDI Setup. Upon booting the system it defaults to 32 bit, which made this issue immediately apparent. (I haven't found any way of setting this permanently to something else using Audio MIDI Setup, btw.) Under Linux (ALSA) 24-bit audio output appears to be fine. Unsure if this is a trivial fix or not, but since it's the only real fault I've found with it I thought I'd let you know. Again, big kudos for a VoodooHDA branch which is miles ahead of what is being provided elsewhere.
  4. Drag0nFly

    VoodooHDA with fix for random KP

    @Zenith432 - just wanted to let you know that your audio driver is working perfectly. I've now tested it on two systems; one using the ALC888 codec (Shuttle SH67H7) and an old Core2 Duo system (Dell Dimension 8200) using the Intel 82801H on-board audio. ALC888: The S/PDIF port is working (currently outputting 96kHz / 24-bit, stereo) and the volume levels using the analog audio ports are correct. Sweet. The only thing I noticed - and this is the same with all VoodooHDA drivers I've previously tested - is that the 5.1 output is missing with the optical/toslink. HDMI reports 8 channels; but I haven't been able to test this yet. Nevertheless, I am plenty happy to be able to use the optical for 2-ch stereo, and with a high sampling rate. (Actually, the prior VoodooHDA versions were reporting 4 channels for S/PDIF, but in actuality had no working output at all.) 82801H (vendor ID: 8086, device ID: 284b): This chip has not been working on _any_ of the recent VoodooHDA's previously tested. I've had to use an ancient 0.2.1 version with it in order to produce anything other than crackling audio output. So major kudos! Glad someone who knows what they are doing is maintaining the source code for these drivers. Now if only there was a permanent solution for the HD3000 ring stall issue ("apparent hang in main graphics engine") I'd be a happy Hackintosh camper. My last crash actually corrupted the main hfs+ partition, not recoverable with Disk Util.
  5. Drag0nFly

    VoodooHDA with fix for random KP

    Ok, thanks - the 2.8.2d6 download works now. And yeah, I noticed they had tagged the bzipped tar-file file as "encrypted" or "damaged". Gotta love these lame file-sharing sites. (who actually pays for this service anyway?) I seldom get KPs on my Shuttle SH67H7 system with VoodooHDA, but they do occur. Hopefully the volume levels and S/PDIF port will be working with this recent driver (unlike the 0.7.x revisions which all had sound distortion/clipping issues with this codec.) I must have tried 10 different versions before finding the (older) 0.2.62 revision which worked the best...
  6. Drag0nFly

    VoodooHDA with fix for random KP

    Glad to see that work is being done on the VoodooHDA driver. I'm still running v0.2.62 with my ALC888 chip, as that is the only version (so far) in which the volume levels are not distorted and – more importantly – the S/PDIF port is working with. However, when I attempt to download the most recent 0.8.2 build mediafire throws an error. Why not include the driver as an attachment to the forum post for members to download directly? Archive Download Blocked The file you attempted to download is an archive that is damaged or possibly encrypted. MediaFire does not support unlimited downloads of broken or encrypted archives and the limit for this file has been reached. This file may only be distributed from a premium account. We have informed the owner that sharing of this file has been limited and how they can resolve this issue. Still have questions, or think we've made a mistake? Please contact support for further assistance.
  7. Just upgraded to 10.7.5 and noticed that the location of the connector information in AppleIntelSNBGraphicsFB apparently has changed. modigliani's binary now creates the following - 0000000: cffa edfe 0700 0001 0300 0000 ............ 000000c: 0b00 0000 0600 0000 7804 0000 ........x... 0000018: 8500 0000 0000 0000 1900 0000 ............ 0000024: 3801 0000 5f5f 5445 5854 0000 8...__TEXT.. 0000030: 0000 0000 0000 0000 0000 0000 ............ Instead of 0000000: 0102 0200 1007 0000 1007 0000 ............ 000000c: 0304 0000 0004 0000 0900 0000 ............ 0000018: 0205 0000 0004 0000 0700 0000 ............ 0000024: 0000 0000 0000 0000 0000 0000 ............ 0000030: 0000 0000 0000 0000 0000 0000 ............ I guess this is a simple enough fix unless the whole HD3000 driver was rewritten for 10.7.5 (If so it is probably using a similar driver as for 10.8.) Just wanted to throw this out there as modigliani's patch was instrumental in fixing the ring stall issue on my system. Would be nice to be able to implement this on 10.7.5 as well. EDIT: it looks like this may at least partly be caused by 'D _PlatformInformationList' now being 'd _PlatformInformationList' (lowercase d). Gotta love these nonsensical changes. /System/Library/Extensions/AppleIntelSNBGraphicsFB.kext/Contents/MacOS $ nm -arch x86_64 AppleIntelSNBGraphicsFB | grep 'PlatformInformationList' 000000000002a620 d _PlatformInformationList 000000000000bffd t __GLOBAL__D_PlatformInformationList 000000000000bfec t __GLOBAL__I_PlatformInformationList
  8. Drag0nFly

    OS X 10.7 Lion on Shuttle SH67H3

    This is good news. But what did you replace the NullCPUPowerManagement kext with? I've tried to get rid of that one before, but the system hasn't been able to boot without it. Also, the IntelCPUMonitor kexts which others have been using instead did not load on the Shuttle. Haven't tried removing it on 10.7.3 or 10.7.4 though, but doubt if there is any difference. As for sound I haven't been able to get AppleHDA working due to lack of a working DSDT, but I have however been using VoodooHDA (0.2.62), which is for the most part working fine. Unlike other (and–newer) VoodooHDA releases it has working S/PDIF output and correct volume levels (i.e, without distortion). Earlier it tended to kernel panic the system on boot, but it has subsided *knock wood* lately; either because of the OS X updates or (more likely) a BIOS update.
  9. @Manwe150, @frankpc – thanks for replying. What confused me was the two HDMI entries in the output (0205 & 0406); how do you know which one to use apart from trial-and-error? I do have the hex file generated (using modgliali's binary), and I guess I'll use xxd to modify it before reinjecting. I also have the OS X Devtools installed, but need to upgrade to 10.7.3/.4 in order for it to work again (after the HD3000 memory debacle which prompted a reinstallation). Seems Apple has disabled it due to version issues (although I somehow doubt there are many changes affecting this between 10.7.2 & 10.7.3)
  10. They both work at the same time, yes – for spanning desktops and mirroring.
  11. @Manwe150 – do you mind reposting those instructions? I realize changing this is probably trivial, but I am still in the dark as to what to change (and how; i.e, which hex editor? Do you simply remove the lines referencing unwanted outputs–LVDS for instance in my case? My connector info is in post #161.) Since I do have all my outputs working (only a DVI-I and a HDMI port), I am hoping this modification will eliminate the "apparent hang in main graphics engine" / ring buffer / stampWait: Overflowed checking for stamp 0x1a103820 on Main ring: called from waiting for blt buffer (3D blt) errors with the HD3000. It has been rumored to fix it, although I have not been able to find concrete evidence of it. btw. – I've also tried RevoGirl's kext edits which aimed to reduce the amount of memory available to the HD3000 (from 512mb to 288mb). However, those changes caused serious graphical glitches and the system had to be re-installed.
  12. Anyone still reading this thread? My system crashes weekly -- sometimes daily due to this bug. It happens regardless of heavy gfx activity (i.e, an OpenGL screensaver) or just doing stuff in ‘‘Finder’’ and even Terminal.app. The ring-buffer overflow bug appears to be trivial to fix for anyone who knows what they are doing (i.e, not me). I simply need the connector information in the HD3000 tables to correctly reflect one DVI port and one HDMI port; which is what is present on this Shuttle SH67H7 system.
  13. Ok, I got the HDMI port working with SMproductid set to "MacBookPro8,1" in /Extra/smbios.plist But I do need some help to get the tables fixed in the HD3000 driver.
  14. @Modigliani - thanks for this very useful tool. I am trying to sort out the "apparent hang in main graphics engine" bug on my Shuttle SH67H7 (running 10.7.3), and would also like to get the built-in HDMI port working. The system has one DVI-D & one HDMI port (which does send output but it's garbled) I am am a bit unsure how to modify the table (also: how to you accomplish this with “vim”?) The output references LVDS (which does not exist) and has two HDMI entries (0205 & 0406) Since it is probably a trivial question for someone more familiar with this, I am wondering what changes I should make to have the right config for my Shuttle system (using DVI as the main output, with mirroring on the HDMI port) Below the information your executable produces when being run on the Shuttle SH67H7 – $ ./modgliani-intelHD3000 60+0 records in 60+0 records out 60 bytes transferred in 0.000130 secs (461758 bytes/sec) 0000000: 0102 0400 1007 0000 1007 0000 ............ 000000c: LVDS_0503 0000 0200 0000 3000 0000 ........0... 0000018: HDMI_0205 0000 0004 0000 0700 0000 ............ 0000024: DVI_0304 0000 0004 0000 0900 0000 ............ 0000030: HDMI_0406 0000 0004 0000 0900 0000 ............ Conectors total: 0400 The table file: table-SH67H7.bin.txt
  15. Drag0nFly


    An update -- for my system at least (which is using the ALC888 codec on a H67 chipset, a Shuttle SH67H7), VoodooHDA 0.2.62 does enable the S/PDIF digital optical (TOSlink) output. I basically tried all the other (incl. the more recent) VoodooHDA versions and none of them came close to getting the digital audio port actually producing sound. VoodooHDA 2.7.x also has a way too high output level for this audio chipset (on the analog outs), causing distortions when playing dynamic material such as movies in VLC or MPlayerX. 0.2.62 did not have that issue, and the volume on the players can actually be set at 100% (not the system volume), without causing distortions. On the other VoodooHDA's the sound started to distort at around ~50%! So it, to me at least, it appears that support for certain well-known codecs has suffered during recent VodooHDA development. Obviously it would be nice to have this support re-instated in the 'current' versions (along with proper 5.1 support over S/PDIF).