16" MBP GPU Panics

[gazwas]

So why does yours need a logic board if its the integrated GPU in the intel CPU causing the issue?

 

[StolenPhone]

Because the Intel GPU is on the logic board?

See the logic board here: https://d3nevzfk7ii3be.cloudfront.net/igi/eUDq5E4tFcvmNHFT.huge

Its the big yellow box on the center upper left. Thats the intel CPU with integrated Intel GPU 630.

Maybe its a bad CPU/GPU, so it need replacement? Maybe its a bad solder joint? But the fact the chip is on the logic board is SOP for Apple to just swap it out.

 

[gazwas]

Because the Intel GPU is on the logic board?

See the logic board here: https://d3nevzfk7ii3be.cloudfront.net/igi/eUDq5E4tFcvmNHFT.huge

Its the big yellow box on the center upper left. Thats the intel CPU with integrated Intel GPU 630.

OH ok. I thought they would just replace the CPU but it looks from your pic its all one piece/soldered.

 

[StolenPhone]

Yep. BTW...I had it wrong... its the one on the right, not the left. That big one I said was actually the AMD. The Intel is to the right of it.

 

[gazwas]

My BTO 16” MBP has just left China so are you saying all the 9th Gen CPU’s are faulty and should be worried or is this a very isolated batch of CPU’s.

 

[visualinventor]

My BTO 16” MBP has just left China so are you saying all the 9th Gen CPU’s are faulty and should be worried or is this a very isolated batch of CPU’s.

The issue seems to be in the custom built to order MacBooks and the intel GPU. I used my machine successfully for 2.5 months before I had any issues. Since they’ve started getting reports and broken machines I have to believe that they’ve figured out a fix. So your results may vary.
[automerge]1583024432[/automerge]

So hats off to visualinventor in the Apple forums. He was able to reproduce this faithfully by running Geekbench 5. He found that its actually the Intel GPU causing this by running the metal tests against the Metal api.

Thanks for the shout out @StolenPhone. Glad it was helpful.
I’m curious if your logs had that same signalHardwareError?

 

[Richardvs]

The issue seems to be in the custom built to order MacBooks and the intel GPU. I used my machine successfully for 2.5 months before I had any issues. Since they’ve started getting reports and broken machines I have to believe that they’ve figured out a fix. So your results may vary.
[automerge]1583024432[/automerge]

That is not true, i have a base build i9, 16GB 1TB, 5500M 4GB with freezes, kernel panics. And with the geekbench test i got the same errors with Intel GPU. Also low geekbench score. I have 3 days left to return....

 

[matram]

There are several reports of issues around switching between iGPU and dGPU an also when restarting from sleep. These I have experienced on my 16”, these issues seems to be SW related and has improved a bit with the latest Catalina release.

But this seems to be more about crashes during use of some apps, or?

 

[gazwas]

So Apple have not found a fix for this issue and are still sending out machines with faulty components? As you have a base spec do you know the manufacture date of your machine, maybe an 2019 build machine before the issue became more widespread and the problem discovered.

 

[Richardvs]

So Apple have not found a fix for this issue and are still sending out machines with faulty components? As you have a base spec do you know the manufacture date of your machine, maybe an 2019 build machine before the issue became more widespread and the problem discovered.

Checked the manufacturing date and now i see it’s manufactured at 2019-11-11. Battery is from 2019-10-08... Got a old one from the store..... (Not Apple Store)

 

[matram]

Graphics switching issues and recovery from sleep appears to be software drivers related to the new Navi GPU. Some improvements were made in Catalina 10.15.3.

 

[gazwas]

Graphics switching issues and recovery from sleep appears to be software drivers related to the new Navi GPU. Some improvements were made in Catalina 10.15.3.

However, this appears to be hardware failure according to replies above as they are having logic board (CPU) replacements.

 

[Mark_EL]

Hmmmm I tried the Geekbench 5 metal test the way Visualinventor described it. I actually did 3 runs. The first run, it gave a Metal score of 2230. Second time it gave 2228 I believe. But on the third run, the computer completely crashed and shut down...

Now I am not that technical, but when I restarted it, the machine gave this crash report:

panic(cpu 0 caller 0xffffff7fa3d34232): "Graphics firmware encountered an exception (eip: 0xddf5, fault #: 0xd)\n"@skl/sched3/IGGuC.cpp:864

Backtrace (CPU 0), Frame : Return Address

0xffffff83ca093b70 : 0xffffff801f53bb1b

0xffffff83ca093bc0 : 0xffffff801f6733e5

0xffffff83ca093c00 : 0xffffff801f664e5e

0xffffff83ca093c50 : 0xffffff801f4e2a40

0xffffff83ca093c70 : 0xffffff801f53b207

0xffffff83ca093d70 : 0xffffff801f53b5eb

0xffffff83ca093dc0 : 0xffffff801fcd24f9

0xffffff83ca093e30 : 0xffffff7fa3d34232

0xffffff83ca093e80 : 0xffffff7fa3cacd24

0xffffff83ca093ee0 : 0xffffff801fc40ccd

0xffffff83ca093f30 : 0xffffff801fc3f612

0xffffff83ca093f70 : 0xffffff801fc3eb9c

0xffffff83ca093fa0 : 0xffffff801f4e213e

Kernel Extensions in backtrace:

com.apple.driver.AppleIntelKBLGraphics(14.0.3)[1B1F5C29-179F-374D-BACF-919D3A83FBBD]@0xffffff7fa3cab000->0xffffff7fa3d9dfff

dependency: com.apple.iokit.IOPCIFamily(2.9)[1286D5E5-A6A1-3C44-A244-04C068903DB2]@0xffffff7f9ff10000

dependency: com.apple.iokit.IOSurface(269.6)[640503DC-6033-3EF6-85F4-F48282EF48DD]@0xffffff7fa1a7f000

dependency: com.apple.iokit.IOGraphicsFamily(569.3)[7B5CA52B-0BD1-34AB-BFC9-7D5C3A0DBE7C]@0xffffff7fa319e000

dependency: com.apple.iokit.IOAcceleratorFamily2(438.2.8)[9176E894-E24A-36BA-AFDA-0DB11F3F512A]@0xffffff7fa31ef000



BSD process name corresponding to current thread: kernel_task

Boot args: chunklist-security-epoch=0 -chunklist-no-rev2-dev



Mac OS version:

19C57



Kernel version:

Darwin Kernel Version 19.2.0: Sat Nov 9 03:47:04 PST 2019; root:xnu-6153.61.1~20/RELEASE_X86_64

Kernel UUID: C3E7E405-C692-356B-88D3-C30041FD1E72

Kernel slide: 0x000000001f200000

Kernel text base: 0xffffff801f400000

__HIB text base: 0xffffff801f300000

System model name: MacBookPro16,1 (Mac-E1008331FDC96864)

System shutdown begun: NO



System uptime in nanoseconds: 3518976357371

last loaded kext at 346973728989: @filesystems.afpfs 11.2 (addr 0xffffff7fa68e1000, size 360448)

loaded kexts:

com.getdropbox.dropbox.kext 1.10.3

com.intel.driver.EnergyDriver 3.7.0

@filesystems.afpfs 11.2

@nke.asp-tcp 8.1

@filesystems.smbfs 3.4

@kext.AMDRadeonX6000 3.0.4

@kext.AMDRadeonServiceManager 3.0.4

>!AGraphicsDevicePolicy 4.5.21

@AGDCPluginDisplayMetrics 4.5.21

>!AHV 1

|IOUserEthernet 1.0.1

|IO!BSerialManager 7.0.2f4

>pmtelemetry 1

>!AUpstreamUserClient 3.6.8

>AGPM 111.4.1

>!APlatformEnabler 2.7.0d0

>X86PlatformShim 1.0.0

@Dont_Steal_Mac_OS_X 7.0.0

>AGDCBacklightControl 4.5.21

>!A!IKBLGraphics 14.0.3

>!ABacklight 180.1

>ACPI_SMC_PlatformPlugin 1.0.0

>!AThunderboltIP 3.1.3

>BridgeAudioCommunication 6.60

>!ATopCaseHIDEventDriver 3420.1

>!AHIDALSService 1

>!AFIVRDriver 4.1.0

>!AMCCSControl 1.13

>!AMuxControl2 4.5.21

>!ABridgeAudio!C 6.60

>!AGFXHDA 100.1.422

>!A!IPCHPMC 2.0.1

>!A!ICFLGraphicsFramebuffer 14.0.3

>!A!ISlowAdaptiveClocking 4.0.0

>!AAVEBridge 6.1

@fileutil 20.036.15

@filesystems.autofs 3.0

>BCMWLANFirmware4355.Hashstore 1

>BCMWLANFirmware4364.Hashstore 1

>BCMWLANFirmware4377.Hashstore 1

>!ABCMWLANBusInterfacePCIe 1

>!AFileSystemDriver 3.0.1

@filesystems.hfs.kext 522.0.9

@BootCache 40

@!AFSCompression.!AFSCompressionTypeDataless 1.0.0d1

@!AFSCompression.!AFSCompressionTypeZlib 1.0.0

>!AVirtIO 1.0

@filesystems.apfs 1412.61.1

@private.KextAudit 1.0

>!ASmartBatteryManager 161.0.0

>!AACPIButtons 6.1

>!ASMBIOS 2.1

>!AACPIEC 6.1

>!AAPIC 1.7

$!AImage4 1

@nke.applicationfirewall 303

$TMSafetyNet 8

@!ASystemPolicy 2.0.0

|EndpointSecurity 1

$SecureRemotePassword 1.0

@kext.AMDRadeonX6100HWLibs 1.0

@kext.AMDRadeonX6000HWServices 3.0.4

|IOAVB!F 800.17

>!ASSE 1.0

@!AGPUWrangler 4.5.21

@kext.AMDRadeonX6000Framebuffer 3.0.4

>!ABacklightExpert 1.1.0

>IOPlatformPluginLegacy 1.0.0

>!AHDA!C 283.15

|IOHDA!F 283.15

>!AHIDKeyboard 209

>!AActuatorDriver 3420.2

|IO!BHost!CUARTTransport 7.0.2f4

|IO!BHost!CTransport 7.0.2f4

>!A!ILpssUARTv1 3.0.60

>!A!ILpssUARTCommon 3.0.60

>!AOnboardSerial 1.0

>!AHS!BDriver 3420.1

>IO!BHIDDriver 7.0.2f4

>!AMultitouchDriver 3420.2

>!AInputDeviceSupport 3420.4

@kext.AMDSupport 3.0.4

>!ASMBus!C 1.0.18d1

>!AGraphicsControl 4.5.21

>X86PlatformPlugin 1.0.0

>!AThunderboltEDMSink 4.2.2

>!AThunderboltDPOutAdapter 6.2.4

|IONDRVSupport 569.3

>!ASMBusPCI 1.0.14d1

>IOPlatformPlugin!F 6.0.0d8

@!AGraphicsDeviceControl 4.5.21

|IOAccelerator!F2 438.2.8

|IOGraphics!F 569.3

|IOSlowAdaptiveClocking!F 1.0.0

@plugin.IOgPTPPlugin 800.14

|IOEthernetAVB!C 1.1.0

@kext.triggers 1.0

>usb.IOUSBHostHIDDevice 1.2

>usb.cdc.ecm 5.0.0

>usb.cdc.ncm 5.0.0

>usb.cdc 5.0.0

>usb.networking 5.0.0

>usb.!UHostCompositeDevice 1.2

>!ABCMWLANCore 1.0.0

>mDNSOffloadUserClient 1.0.1b8

>IOImageLoader 1.0.0

|IO80211!FV2 1200.12.2b1

>corecapture 1.0.4

|IOSkywalk!F 1

|IOSurface 269.6

@filesystems.hfs.encodings.kext 1

|IOAudio!F 300.2

@vecLib.kext 1.2.0

|IOSerial!F 11

>!AXsanScheme 3

>usb.!UVHCIBCE 1.2

>usb.!UVHCI 1.2

>usb.!UVHCICommonBCE 1.0

>usb.!UVHCICommon 1.0

>!AEffaceableNOR 1.0

|IOBufferCopy!C 1.1.0

|IOBufferCopyEngine!F 1

|IONVMe!F 2.1.0

>!AThunderboltPCIDownAdapter 2.5.2

>!AThunderboltDPInAdapter 6.2.4

>!AThunderboltDPAdapter!F 6.2.4

>!AHPM 3.4.4

>!A!ILpssI2C!C 3.0.60

>!A!ILpssDmac 3.0.60

>!A!ILpssI2C 3.0.60

>!AThunderboltNHI 5.8.1

|IOThunderbolt!F 7.4.7

>usb.!UXHCIPCI 1.2

>usb.!UXHCI 1.2

>usb.!UHostPacketFilter 1.0

|IOUSB!F 900.4.2

>!AEFINVRAM 2.1

>!AEFIRuntime 2.1

>!ASMCRTC 1.0

|IOSMBus!F 1.1

|IOHID!F 2.0.0

$quarantine 4

$sandbox 300.0

@Kext.!AMatch 1.0.0d1

>!AKeyStore 2

>!UTDM 489.60.3

|IOSCSIBlockCommandsDevice 422.0.2

>!ACredentialManager 1.0

>KernelRelayHost 1

>!ASEPManager 1.0.1

>IOSlaveProcessor 1

>!AFDEKeyStore 28.30

>!AEffaceable!S 1.0

>!AMobileFileIntegrity 1.0.5

@kext.CoreTrust 1

|CoreAnalytics!F 1

|IOTimeSync!F 800.14

|IONetworking!F 3.4

>DiskImages 493.0.0

|IO!B!F 7.0.2f4

|IO!BPacketLogger 7.0.2f4

|IOUSBMass!SDriver 157.40.7

|IOSCSIArchitectureModel!F 422.0.2

|IO!S!F 2.1

|IOUSBHost!F 1.2

>usb.!UCommon 1.0

>!UHostMergeProperties 1.2

>!ABusPower!C 1.0

|IOReport!F 47

>!AACPIPlatform 6.1

>!ASMC 3.1.9

>watchdog 1

|IOPCI!F 2.9

|IOACPI!F 1.4

@kec.pthread 1

@kec.corecrypto 1.0

@kec.Libm 1
[automerge]1583059467[/automerge]
There is one thing I don't understand though... Why would this be a hardware failure and not a software thing? Because when I run the same test with OpenCL, all is fine. I would think if this was hardware related, it would always give problems, no matter what drivers are used, would it not?

Like I said, I am not all that tech savvy and I don't even really know what the difference is between OpenCL and Metal but still....

 

[gazwas]

@Mark_EL The simple fact it crashed doing a seemingly simple benchmark test that Mac users have done for years without problem is rather worrying.

@visualinventor Is there any chance this is a Geekbench 5 bug?

Have you ran Geekbench 4 and got the same errors?

 

[Mark_EL]

@Mark_EL The simple fact it crashed doing a seemingly simple benchmark test that Mac users have done for years without problem is rather worrying.

@visualinventor Is there any chance this is a Geekbench 5 bug?

Have you ran Geekbench 4 and got the same errors?

While I agree on that, I think it could be a software or firmware version that makes the system crash (?).

I have not tried Geekbench 4, but perhaps will...


Edit; As I only have the trial version of Geekbench 4, I am not able to perform a Metal test on version 4...

 

[Squuiid]

Fans are likely not spinning fast enough to cool the hardware.
The same thing happens with the Aquantia 10G chip in the 2018 Mac mini.
Download Macsfancontrol, set fans to full speed and re-run tests.
It’ll be interesting, and pretty insightful, to hear people’s results.

Mac Pro users have used this utility for years and years. It’s free too.

Macs Fan Control - control fans on Apple computers, also on Windows via Boot Camp

Control fans on Apple computers, also on Windows via Boot Camp. Monitoring of fan speed and temperature sensors. Solve fan noise and overheating problems, iMac HDD replacement

www.crystalidea.com

 

[gazwas]

Fans are likely not spinning fast enough to cool the hardware.
The same thing happens with the Aquantia 10G chip in the 2018 Mac mini.
Download Macsfancontrol, set fans to full speed and re-run tests.
It’ll be interesting, and pretty insightful, to hear people’s results.

Mac Pro users have used this utility for years and years. It’s free too.

Macs Fan Control - control fans on Apple computers, also on Windows via Boot Camp

Control fans on Apple computers, also on Windows via Boot Camp. Monitoring of fan speed and temperature sensors. Solve fan noise and overheating problems, iMac HDD replacement

www.crystalidea.com

Are you suggesting that running these Geekbench tests in an effort to diagnose/test their MBP’s they are in fact over heating the CPU and permanently damaging the computer?

EDIT: And why is this not a thing with 2019 15” MBP’s if they share the same CPU but inferior cooling?

Have any 2019 15” MBP users reported any issues with the Intel GPU?

 

[Dedicated ESC key]

I’m on my third custom 16”. I ran the above tests and also get all the errors.
I have taken screenshots. Can someone tell me what I do now? How do I describe this problem to apple?
I have spent so long trouble shooting with them already.
Thanks

 

[gazwas]

I’m on my third custom 16”. I ran the above tests and also get all the errors.
I have taken screenshots. Can someone tell me what I do now? How do I describe this problem to apple?
I have spent so long trouble shooting with them already.
Thanks

Someone need to run these tests on Geekbench 4 to rule out software bugs and see if they get similar errors.

 

[Camarillo Brillo]

Why did I even read this??

Yeah I get the errors too but I’m not convinced that means I need a new logic board. My computer seems to be working fine with regards to the intel gpu. I think every single 16” will get this same error in the Geekbench Metal test. I doubt that means they all have faulty logic boards

 

[StolenPhone]

I can't confirm the slowness issues, but if it GPU panics, according to Apple it *IS* a hardware problem and confirmed to me by their executive team. If its firmware related, I would think its stupid for Apple to be shipping out brand new laptops. Any metal API against my Intel chip results in a full GPU panic with the Powerplay failed to resume error. Thats a serious issue. If it doesn't panic when others have the problem, then that may be something completely different. This thread was about the panic. Luckily the Geekbench software was able to determine it is indeed the Intel GPU using metal drivers that forces the panic.
[automerge]1583082584[/automerge]

Someone need to run these tests on Geekbench 4 to rule out software bugs and see if they get similar errors.

If it GPU pnics, it is not. I can run on the Intel GPU and fire up Photoshop/Illustrator/Game, etc and BAM... GPU panic. I can't comment on slowness... but the panic is hardware as confirmed by Apple.

 

[Mark_EL]

But how can you run the Intel GPU and fire up Photoshop for example? I mean, as soon as I launch Photoshop, my MBP automatically will switch to to the dedicated GPU. And therefore, IRLC this issue has not really effected me in any way. I had 2 or 3 kernel panics when I just got this machine, while it was connected to my dock. (waking up after sleep I believe). I have not had any other panics since. Well, except for the one today when I was testing with Geekbench 5.

 

[gazwas]

If it GPU pnics, it is not. I can run on the Intel GPU and fire up Photoshop/Illustrator/Game, etc and BAM... GPU panic. I can't comment on slowness... but the panic is hardware as confirmed by Apple.

Is this not worth a try to maybe rule it out though, might save a lot of hysteria?

I just ran the Geekbench 5 compute benchmark on my 2012 iMac and console through up lots of error too. It could be software incompatibility?

 

[StolenPhone]

But how can you run the Intel GPU and fire up Photoshop for example? I mean, as soon as I launch Photoshop, my MBP automatically will switch to to the dedicated GPU. And therefore, IRLC this issue has not really effected me in any way. I had 2 or 3 kernel panics when I just got this machine, while it was connected to my dock. (waking up after sleep I believe). I have not had any other panics since. Well, except for the one today when I was testing with Geekbench 5.

When you run Photoshop, the software apparently uses Intel (if already in Intel mode) initially before it decides to switch to the AMD. This is caused by the API calls and obviously the metal API calls. This happens before the Apple software decides its a good idea to use the AMD GPU. Then it panics.

If I have the machine set to only use the AMD, I can run Photoshop with no panics and all is well. So its certainly the metal calls on the Intel side.

The panic's may get worse. Mine initially was every several days. It then went to a few times a day. Now its any time I use the Intel with Metal API. Its pretty much immediate. The degradation points heavily towards hardware as its now instantaneous. I have a few friends with the exact same machine with absolutely no panics whatsoever. So its not software or all MBPs would get regular panics.
[automerge]1583087338[/automerge]

Is this not worth a try to maybe rule it out though, might save a lot of hysteria?

I just ran the Geekbench 5 compute benchmark on my 2012 iMac and console through up lots of error too. It could be software incompatibility?

Read my response above. Its ruled out. but I'm not discussing slowness. I'm discussing panics.

 

[Mark_EL]

When you run Photoshop, the software apparently uses Intel (if already in Intel mode) initially before it decides to switch to the AMD. This is caused by the API calls and obviously the metal API calls. This happens before the Apple software decides its a good idea to use the AMD GPU. Then it panics.

If I have the machine set to only use the AMD, I can run Photoshop with no panics and all is well. So its certainly the metal calls on the Intel side.

The panic's may get worse. Mine initially was every several days. It then went to a few times a day. Now its any time I use the Intel with Metal API. Its pretty much immediate. The degradation points heavily towards hardware as its now instantaneous. I have a few friends with the exact same machine with absolutely no panics whatsoever. So its not software or all MBPs would get regular panics.

I see... Sucks 🙁 I guess I have been more lucky then you then, up to now. Because when I launch PS for example while running Intel, it just switches without any hickups. The Metal Geekbench test did not go well for me, and the speed result is definitely not what it should be, compared to others. But I can’t really be bothered about that tbh, since the system will use the dedicated GPU anyway when power is needed. So I guess I am fine as long as it keeps performing like it does now....