Mac Pro 1,1 panics after CPU upgrade

Discussion in 'Mac Pro' started by monokakata, Feb 27, 2012.

  1. monokakata macrumors 68000

    monokakata

    Joined:
    May 8, 2008
    Location:
    Hilo, Hawai'i
    #1
    I've gotten a couple of panics after upgrading my 1,1 quad to an 8-core (5355s).

    Lion 10.7.3, updated via the combo updater before the upgrade, and then again after. I did a clean Lion install on this machine last November. There seem to be some old drivers hanging around from the Snow Leopard installation, but they've been there for months without causing trouble.

    In both cases the reference was to the AppleIntelCPUPowermanagement item.

    Before the upgrade, the MP would regularly fail to wake from sleep -- or I should say, would appear to wake from sleep (LED steady) but would not put out any video, so all I could do was shut down via the power switch. This would happen on maybe 20-25% of the times it went to sleep.

    Dell U2711 connected via miniDSP-->DSP cable.

    I could never find any indication in the console logs about these wake failures.

    I have reset SMC (several times), reset PRAM, repaired permissions, etc.

    The most recent panic happened after I'd set Energy Saver to "never."

    What I find out there via Google had to do with hackintosh problems, not to standard Apple machines.

    Is is possible that the kext is specific to the old dual-core chips and is barfing in some way because it sees the new quad-cores?

    I don't mind reinstalling Lion, if that seems like a good idea.

    Any ideas?



    Interval Since Last Panic Report: 65 sec
    Panics Since Last Report: 1
    Anonymous UUID: 2A513DD7-4BFC-410A-AB25-C0579B8A1FEE

    Mon Feb 27 09:49:06 2012
    panic(cpu 0 caller 0x1c348d2): "pmLock: waited too long, held by 0@0"@/SourceCache/AppleIntelCPUPowerManagement/AppleIntelCPUPowerManagement-167.3/pmYonahProcessor.c:845
    Backtrace (CPU 0), Frame : Return Address (4 potential args on stack)
    0x8a05bcc8 : 0x22032e (0x6ad91c 0x8a05bce8 0x229f00 0x0)
    0x8a05bcf8 : 0x1c348d2 (0x1c486d6 0x0 0x0 0x2d2711)
    0x8a05bd38 : 0x1c2c7f4 (0x81d180 0x1 0x128f5000 0x0)
    0x8a05bdd8 : 0x1c2ea99 (0xffffffff 0x7fffffff 0x0 0x0)
    0x8a05be68 : 0x2cfb83 (0xffffffff 0x7fffffff 0x10 0x282)
    0x8a05be88 : 0x22d335 (0x867840 0x2316892a 0x1c 0x868620)
    0x8a05bec8 : 0x22e110 (0x137257f0 0x867840 0x0 0x0)
    0x8a05bf18 : 0x22f612 (0x1372580c 0x0 0x8a05bf58 0x240b4e)
    0x8a05bf78 : 0x22f740 (0x23e300 0x84ad60 0x0 0x2c62b6)
    0x8a05bf98 : 0x23e4e1 (0x23e300 0x84ad60 0x0 0x0)
    0x8a05bfc8 : 0x2c6d0c (0x84ad60 0x0 0x10 0x1343d190)
    Kernel Extensions in backtrace:
    com.apple.driver.AppleIntelCPUPowerManagement(167.3)[E7A581F5-68AD-40A7-9044-10DBAE790332]@0x1c2a000->0x1c54fff

    BSD process name corresponding to current thread: kernel_task

    Mac OS version:
    11D50

    Kernel version:
    Darwin Kernel Version 11.3.0: Thu Jan 12 18:48:32 PST 2012; root:xnu-1699.24.23~1/RELEASE_I386
    Kernel UUID: 8986BDEF-9D4E-32D2-8566-2196FD2D35B2
    System model name: MacPro1,1 (Mac-F4208DC8)

    System uptime in nanoseconds: 172532579102
    last loaded kext at 39914839893: com.apple.driver.AppleBluetoothMultitouch 66.6 (addr 0x1e2d000, size 49152)
    last unloaded kext at 105473644333: com.roxio.BluRaySupport 1.1.6 (addr 0xb47000, size 40960)
    loaded kexts:
    com.bresink.driver.BRESINKx86Monitoring 8.0
    com.paceap.kext.pacesupport.snowleopard 5.7.2
    com.digidesign.iokit.DigiDal 8.0.4f2
    com.apple.driver.AppleBluetoothMultitouch 66.6
    com.apple.filesystems.autofs 3.0
    com.apple.driver.AppleHDA 2.1.7f9
    com.apple.driver.AppleUpstreamUserClient 3.5.9
    com.apple.driver.AppleMCCSControl 1.0.26
    com.apple.driver.AudioAUUC 1.59
    com.apple.kext.ATIFramebuffer 7.1.8
    com.apple.iokit.IOUserEthernet 1.0.0d1
    com.apple.driver.AppleMCEDriver 1.1.9
    com.apple.Dont_Steal_Mac_OS_X 7.0.0
    com.apple.driver.AudioIPCDriver 1.2.2
    com.apple.driver.ACPI_SMC_PlatformPlugin 4.7.5d4
    com.apple.ATIRadeonX3000 7.1.8
    com.apple.driver.AppleLPC 1.5.3
    com.apple.iokit.SCSITaskUserClient 3.0.3
    com.apple.AppleFSCompression.AppleFSCompressionTypeDataless 1.0.0d1
    com.apple.AppleFSCompression.AppleFSCompressionTypeZlib 1.0.0d1
    com.apple.BootCache 33
    com.apple.iokit.IOAHCIBlockStorage 2.0.1
    com.apple.driver.AppleFWOHCI 4.8.9
    com.apple.driver.AppleAHCIPort 2.2.0
    com.apple.driver.AppleIntelPIIXATA 2.5.1
    com.apple.driver.AppleIntel8254XEthernet 2.1.3b1
    com.apple.driver.AppleUSBHub 4.5.0
    com.apple.driver.AppleEFINVRAM 1.5.0
    com.apple.driver.AppleUSBEHCI 4.5.8
    com.apple.driver.AppleUSBUHCI 4.4.5
    com.apple.driver.AppleACPIButtons 1.4
    com.apple.driver.AppleRTC 1.4
    com.apple.driver.AppleHPET 1.6
    com.apple.driver.AppleSMBIOS 1.7
    com.apple.driver.AppleACPIEC 1.4
    com.apple.driver.AppleAPIC 1.5
    com.apple.driver.AppleIntelCPUPowerManagementClient 167.3.0
    com.apple.nke.applicationfirewall 3.2.30
    com.apple.security.quarantine 1.1
    com.apple.driver.AppleIntelCPUPowerManagement 167.3.0
    com.apple.driver.AppleMultitouchDriver 220.62.1
    com.apple.driver.IOBluetoothHIDDriver 4.0.3f12
    com.apple.kext.triggers 1.0
    com.apple.driver.DspFuncLib 2.1.7f9
    com.apple.iokit.IOFireWireIP 2.2.4
    com.apple.driver.AppleUSBAudio 2.8.2f5
    com.apple.iokit.IOSurface 80.0
    com.apple.iokit.IOBluetoothSer
    Model: MacPro1,1, BootROM MP11.005C.B08, 8 processors, 2.66 GHz, 12 GB, SMC 1.7f10
    Graphics: ATI Radeon HD 5770, ATI Radeon HD 5770, PCIe, 1024 MB
    Memory Module: DIMM Riser A/DIMM 1, 2 GB, DDR2 FB-DIMM, 667 MHz, 0x02FE, 0x4542453231464434414A46542D36452D4520
    Memory Module: DIMM Riser A/DIMM 2, 2 GB, DDR2 FB-DIMM, 667 MHz, 0x02FE, 0x4542453231464434414A46542D36452D4520
    Memory Module: DIMM Riser B/DIMM 1, 1 GB, DDR2 FB-DIMM, 667 MHz, 0x0000, 0x000000000000000000000000000000000000
    Memory Module: DIMM Riser B/DIMM 2, 1 GB, DDR2 FB-DIMM, 667 MHz, 0x0000, 0x000000000000000000000000000000000000
    Memory Module: DIMM Riser A/DIMM 3, 2 GB, DDR2 FB-DIMM, 667 MHz, 0x02FE, 0x4542453231464434414A46542D36452D4520
    Memory Module: DIMM Riser A/DIMM 4, 2 GB, DDR2 FB-DIMM, 667 MHz, 0x02FE, 0x4542453231464434414A46542D36452D4520
    Memory Module: DIMM Riser B/DIMM 3, 1 GB, DDR2 FB-DIMM, 667 MHz, 0x0000, 0x000000000000000000000000000000000000
    Memory Module: DIMM Riser B/DIMM 4, 1 GB, DDR2 FB-DIMM, 667 MHz, 0x0000, 0x000000000000000000000000000000000000
    Bluetooth: Version 4.0.3f12, 2 service, 18 devices, 2 incoming serial ports
    Network Service: Ethernet 2, Ethernet, en1
    PCI Card: ATI Radeon HD 5770, sppci_displaycontroller, Slot-1
    PCI Card: ATI Radeon HD 5770, ATY,HoolockParent, Slot-1
    Serial ATA Device: ST31000528AS, 1 TB
    Serial ATA Device: Hitachi HDS723020BLA642, 2 TB
    Serial ATA Device: Hitachi HDS723020BLA642, 2 TB
    Serial ATA Device: ST32000542AS, 2 TB
    Parallel ATA Device: TSSTcorp CDDVDW SH-S222L
    Parallel ATA Device: PIONEER DVD-RW DVR-112D
    USB Device: Keyboard Hub, apple_vendor_id, 0x1006, 0xfd400000 / 3
    USB Device: Apple Keyboard, apple_vendor_id, 0x0220, 0xfd420000 / 7
    USB Device: hub_device, 0x0424 (SMSC), 0x2514, 0xfd300000 / 2
    USB Device: hub_device, 0x0424 (SMSC), 0x2640, 0xfd310000 / 4
    USB Device: miscellaneous_device, 0x046d (Logitech Inc.), 0x0821, 0xfd313000 / 6
    USB Device: Ultra Fast Media Reader, 0x0424 (SMSC), 0x4063, 0xfd311000 / 5
    USB Device: Bluetooth USB Host Controller, apple_vendor_id, 0x8206, 0x5d200000 / 3
    USB Device: Fast Track, 0x0763 (M-Audio), 0x2024, 0x5d100000 / 2
    FireWire Device: built-in_hub, 800mbit_speed
     
  2. mulo macrumors 68020

    mulo

    Joined:
    Aug 22, 2010
    Location:
    Behind you
    #2
    its unable to alter the power given to the CPU (I think)
    and yes, the driver (kext) is probably specific to that cpu, so reinstalling is probably a good idea.

    you can just make a small partition and install to that and see if that runs fine, if it does then reinstall the whole thing :)

    and use code tags for your panic log

    Code:
    tada!
     
  3. monokakata thread starter macrumors 68000

    monokakata

    Joined:
    May 8, 2008
    Location:
    Hilo, Hawai'i
    #3
    I suppose I should know what you mean by "code tags," but even with the "tada!" example, I don't.
     
  4. James Craner, Feb 27, 2012
    Last edited: Feb 27, 2012

    James Craner macrumors 68000

    James Craner

    Joined:
    Sep 13, 2002
    Location:
    Bristol, UK
    #4
    Out of interest why? Not even sure I understand the reason why it's important even if it is code, which it is not - its just a log. Not trying to be difficult just wondering why.


    Edit - Don't worry I think I can see the use of code tags now - makes it easier to see the post text, rather than 'code'
     
  5. James Craner macrumors 68000

    James Craner

    Joined:
    Sep 13, 2002
    Location:
    Bristol, UK
    #5
    Its one of the post format options - see screenshot below.

    MacRumors Forums - Reply to Topic.png
     
  6. 666sheep macrumors 68040

    666sheep

    Joined:
    Dec 7, 2009
    Location:
    Poland
    #6
    OP - upgrade your MP firmware to 2.1, it should solve your issue.
    See http://forum.netkas.org/index.php/topic,1094.0.html (register to see attachments).

    @James Craner: posting long text in code tags gives scrolled text box instead of extremely long single post.
     
  7. James Craner, Feb 27, 2012
    Last edited: Feb 27, 2012

    James Craner macrumors 68000

    James Craner

    Joined:
    Sep 13, 2002
    Location:
    Bristol, UK
    #7
    Thank you, Learnt something tonight :)
     
  8. monokakata thread starter macrumors 68000

    monokakata

    Joined:
    May 8, 2008
    Location:
    Hilo, Hawai'i
    #8
    Thanks. I'll do that right away. I'd already found that thread but hadn't registered and thus couldn't get the firmware file.

    And thanks to others about the display tips. I was scratching my bald head because as an old Fortran programmer I'm used to "!" for comments but didn't see what it had to do with posting a log.

    Always something new to learn, but stopping learning = being dead.
     
  9. monokakata thread starter macrumors 68000

    monokakata

    Joined:
    May 8, 2008
    Location:
    Hilo, Hawai'i
    #9
    It gets weirder.

    I've done the Netkas firmware upgrade (to 2,1).

    Today I've been running Compressor (4.0.2) and when it reaches the end of a job (sometimes a job that takes 2 hours) the Mac Pro immediately stops and reboots itself.

    The Compressor log shows

    2/27/12 5:10:22.310 PM Compressor: kCGErrorIllegalArgument: CGSInvalidateGroupedWindowShadow

    and, strangely, that Please Sleep (which is disabled) tried to start, but couldn't.

    The first time, the restart was successful.

    The second time, no chime.

    So I used the trick I discovered yesterday -- pull out one of the memory boards slightly, and reseat it. Then it will boot.

    I don't know what to make of this, especially how to interpret the memory card trick. Earlier today I used some contact cleaner and canned air to clean out the card edges and the memory slots. However, the card-pulling trick worked before I did the cleaning (which I did because I thought that during the CPU upgrade some crap might have gotten on the contacts).

    I'd be thinking I had a memory problem, except the memory's been in there and untouched for about a month, and I've never seen these problems. And Compressor completed its tasks appropriately before the CPU upgrade.

    Ideas welcome.

    I'm going to do a fresh Lion install, but I don't have high hopes for that.
     
  10. ashman70 macrumors 6502a

    Joined:
    Dec 20, 2010
    #10
    Ok so if we go back to before you upgraded your CPU's, did you ever have any of these problems?

    I can't imagine how simply upgrading the CPU's in your MP has caused all of this, other then perhaps when you reassembled your MP after the processor upgrade, something got damaged, perhaps a memory daughter card or a module, I don't know.

    Are you using iStat Pro on your MP so you can look at fan speeds and temps? Its a widget you can get for free, very useful.
     
  11. monokakata, Feb 27, 2012
    Last edited: Feb 27, 2012

    monokakata thread starter macrumors 68000

    monokakata

    Joined:
    May 8, 2008
    Location:
    Hilo, Hawai'i
    #11
    I agree that the fact of an upgrade isn't likely to have caused all this by itself.

    iStat reports nothing unusual. CPU temps are 36 and 38 at the moment, and rose into the high 40s when Compressor was cranking away. Fan speeds are low (as always).

    I don't discount the possibility that I did some damage somewhere, but I've been inside lots of computers for lots of years and so all I can say is that it's possible but unlikely.

    What do you think about the memory card business? It's as if when the machine gets into a bad state, it fails to see any memory at all (or thinks it's all bad) and thus won't proceed . . . but the act of breaking and re-establishing the connection between one memory card and the motherboard is enough to wipe that bad finding out.

    And the boot failure only happens after a panic (that is, an official panic that's logged) or the Compressor weirdness, or maybe one or two other times.

    I don't know. I've run out of ideas. Obviously I wouldn't be replying this quickly if I'd started a Lion reinstall, but I suppose I'll do it later tonight.

    Edit/addition -- yesterday, when it wouldn't boot (one of the several times) I grabbed my Snow Leopard drive and replaced the Lion drive with it. Instant boot, with no memory card fiddling. I haven't seen any signs that my Lion system disk is having trouble, but nevertheless when it wouldn't boot with Lion, it would boot into SL.

    And to answer your question -- no, I had none of these problems before. I had a few different problems, but not these. I have no memory of every having had a panic with this box. The worst it would do was to spin up its fans upon waking, and sometimes refuse to wake from sleep. It doesn't spin up its fans anymore.
     
  12. ashman70 macrumors 6502a

    Joined:
    Dec 20, 2010
    #12
    Very strange. I mean I have had some experience with machines in the past like yours that have run (servers mostly) for years without issue, and then you do something minor or major to them and they are never the same, instabilities all over the place, and there is no rhyme or reason to it.

    I think what you need to do is try and establish a baseline, which might be hard to do if this is your day to day computer. But if you can, either with the SL or Lion drive, use it for a few days and note all the irregularities and instabilities, hang, unresponsiveness, etc, make nots, to see if you can nail it down. Do you have any original discs for the machine you can run hardware diagnostics with?
     
  13. monokakata thread starter macrumors 68000

    monokakata

    Joined:
    May 8, 2008
    Location:
    Hilo, Hawai'i
    #13
    That's probably a good idea. I'll start keeping my own log.

    But indeed it is my main machine, and just now I'm under pressure to complete editing a 35 minute video before one of the performers returns to Japan (mid-March).

    I can work around the Compressor thing by not having anything else critical going when it's about to complete.

    I don't have the old disks. I work at two sites, and the 1,1 came out here last year when I got a 2010 at my other place. I see various mem tests out there on the web and I'll try one of them, just in case the 8 gigs I bought recently off of eBay are not up to standard.

    And I'll do the Lion reinstall. And do some more cleaning, just in case something snuck in.

    Magical thinking, maybe.
     
  14. mulo macrumors 68020

    mulo

    Joined:
    Aug 22, 2010
    Location:
    Behind you
    #14
    and it puts it into a scrollable box for itself if there is enough of it, which there is, so it doesn't take up as much space in the post :)
     
  15. drrbc macrumors newbie

    Joined:
    Oct 29, 2009
    #15
    Two (!) 5770's?

    Stopping the power to the CPU, but not an on die interrupt.... Exceeding case TDP?

    First, set your fans to 2,000 minimum- it may get your encode done today.

    Then use Hardware Monitor to watch temps and current draw, esp. the "CPU A Proximity", "Power Supply Location" (1 and 2), "Expansion Slots", and the "PCIe Slot X" current draws.

    Pulling a riser card induces an SMC reset IIRC, but I've also seen RAM sticks pass mem tests only to have my problems disappear when the suspected stick was replaced with new RAM.

    My thoughts... does Compressor use OpenCL? If so, the 5770 ('s- plural!) may be overheating an area that causes OS X to issue a TDP interrupt, but it's NOT the Xeon's doing it. It's just that the new proc's addition sent the TDP(case) over the 300 watt rating.

    Try bumping up the fan minimum and see what happens.

    Warmest Regs,
    Robert

    3.3Ghz MacPro1,1
    2X BSEL/VID'd Intel Confidential X5350's
    Koolance and EK waterblocks to GeoThermal loop cooling
    12GB 667 FB-DIMMs
    2.5TB (total) twin internal striped RAID arrays
    .45 Longslide (laser sighting)
    ATI 5770 to 2X 30in. Apple Cinema Displays
    Phased plasma rifle (in the 40 watt range)
    Internal Pioneer BDR-205
     
  16. monokakata, Feb 28, 2012
    Last edited: Feb 28, 2012

    monokakata thread starter macrumors 68000

    monokakata

    Joined:
    May 8, 2008
    Location:
    Hilo, Hawai'i
    #16
    Thanks, Robert.

    No, there's only one 5770. I saw that too and wondered what the hell it was -- had no idea. There's a lot of legacy hardware that shows up in the report -- my legacy, that is. I did have a USB Bluetooth dongle, but it's been gone for a couple of months (replaced by the actual card).

    This AM it wouldn't stay up for more than 5 or 6 minutes without going down.

    I started a Lion install from a USB stick and it got well into it, and crashed again. Restarted, and restarted the install.

    I'd be thinking about overheating too, but every time I go to the machine all the parts are cool.

    Ah, such fun I'm having. The Lion install just crashed again. Well.

    Added:

    Might as well just close this off. It booted happily into SL from another disk -- came up fine. I brought it up and took it down a few times, reset SMC and PRAM, no problems. So I was thinking that maybe there was an elusive disk problem on the Lion disk. But just now, it crashed in SL.

    I think it's hosed. And for the first time, I'm in that awkward position I see others in . . . I can't do without a Mac Pro, so I'll have to get a new one perhaps just before the new models hit. I'll look for a refurb. I do remember what I was told back in the eighties, when I was looking for my first computer (which turned out to be a PDP-11) and was fussing about what might be coming out: "It's always the wrong time to buy a computer, so just buy it."

    In the meantime, I have enough FW800 enclosures (including a dual one) that I can pop out all my drives and limp along with my old MBP, until I replace the MP.
     
  17. monokakata thread starter macrumors 68000

    monokakata

    Joined:
    May 8, 2008
    Location:
    Hilo, Hawai'i
    #17
    I put the old Xeons back in, ran the same procedures that unfailingly crashed it with the 5355s in it, and they ran perfectly.

    Then I put the 5355s back in again, and no boot.

    Put the old ones in once more, and I'm back in business at my old speed, but at least I'm running.

    A bad CPU was very low on my list of possibilties, but I think the results of my swapping derby show that that's what the problem was.
     
  18. ashman70 macrumors 6502a

    Joined:
    Dec 20, 2010
    #18
    What are the specs on the quad core CPU's you took out, what stepping are they? Are they a matched pair?
     
  19. monokakata thread starter macrumors 68000

    monokakata

    Joined:
    May 8, 2008
    Location:
    Hilo, Hawai'i
    #19
    They were stepping-matched (the SL9YM version).

    And I have to say that although I don't want to name names, the person I bought them from is a member of our community here and when I emailed him saying at least one of them was bad, he immediately refunded what I'd paid him.

    Talk about a stand-up guy!

    He got them from somebody else, never used them, and had no way to know whether they were OK or not. I'll return the chips and he's going to take it up with the people he got them from.
     
  20. ashman70 macrumors 6502a

    Joined:
    Dec 20, 2010
    #20
    Well at least your one take away from this is that you can now probably swap the CPU's in a Mac Pro blindfolded with only one hand, lol.
     

Share This Page