One of my computers (6884721) has a Quadro 410 GPU installed. I was happily crunching along and then noticed that at one point the GPU wasn't showing up in the list as it had been. I rebooted and checked again and still nothing, while in the event log I get the line in the title with the error code quoted.

I thought perhaps a driver change was needed, but to no avail - I've tried several combinations and it just won't work. Any ideas?

---

Please post the Boinc startup from the Event log, the first 30 lines will do.

Claggy

3/4/2014 9:45:36 AM | | Starting BOINC client version 7.2.42 for windows_x86_64
3/4/2014 9:45:36 AM | | log flags: file_xfer, sched_ops, task
3/4/2014 9:45:36 AM | | Libraries: libcurl/7.25.0 OpenSSL/1.0.1 zlib/1.2.6
3/4/2014 9:45:36 AM | | Data directory: C:\ProgramData\BOINC
3/4/2014 9:45:36 AM | | Running under account nurse
3/4/2014 9:45:36 AM | | GPU detection failed. error code -1073740940
3/4/2014 9:45:36 AM | | No usable GPUs found
3/4/2014 9:45:36 AM | | app version refers to missing GPU type NVIDIA
3/4/2014 9:45:36 AM | SETI@home | Application uses missing NVIDIA GPU
3/4/2014 9:45:36 AM | | app version refers to missing GPU type NVIDIA
3/4/2014 9:45:36 AM | SETI@home | Application uses missing NVIDIA GPU
3/4/2014 9:45:36 AM | | app version refers to missing GPU type NVIDIA
3/4/2014 9:45:36 AM | SETI@home | Application uses missing NVIDIA GPU
3/4/2014 9:45:36 AM | | app version refers to missing GPU type NVIDIA
3/4/2014 9:45:36 AM | SETI@home | Application uses missing NVIDIA GPU
3/4/2014 9:45:36 AM | | app version refers to missing GPU type NVIDIA
3/4/2014 9:45:36 AM | SETI@home | Application uses missing NVIDIA GPU
3/4/2014 9:45:36 AM | | app version refers to missing GPU type NVIDIA
3/4/2014 9:45:36 AM | SETI@home | Application uses missing NVIDIA GPU
3/4/2014 9:45:36 AM | | app version refers to missing GPU type NVIDIA
3/4/2014 9:45:36 AM | SETI@home | Application uses missing NVIDIA GPU
3/4/2014 9:45:36 AM | SETI@home | Missing coprocessor for task 28no13ab.25501.23380.438086664206.12.98_1
3/4/2014 9:45:36 AM | SETI@home | Missing coprocessor for task 28no13ab.25501.23380.438086664206.12.110_1
3/4/2014 9:45:36 AM | SETI@home | Missing coprocessor for task 28no13ab.25501.23380.438086664206.12.128_1
3/4/2014 9:45:36 AM | SETI@home | Missing coprocessor for task 28no13ab.25501.23380.438086664206.12.92_1
3/4/2014 9:45:36 AM | | Host name: NURSE1
3/4/2014 9:45:36 AM | | Processor: 8 GenuineIntel Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz [Family 6 Model 58 Stepping 9]
3/4/2014 9:45:36 AM | | Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss htt tm pni ssse3 cx16 sse4_1 sse4_2 popcnt aes syscall nx lm vmx smx tm2 pbe
3/4/2014 9:45:36 AM | | OS: Microsoft Windows 7: Professional x64 Edition, Service Pack 1, (06.01.7601.00)
3/4/2014 9:45:36 AM | | Memory: 7.96 GB physical, 15.91 GB virtual
3/4/2014 9:45:36 AM | | Disk: 232.11 GB total, 167.65 GB free
3/4/2014 9:45:36 AM | | Local time is UTC -5 hours
3/4/2014 9:45:36 AM | | Version change (7.2.33 -> 7.2.42)
3/4/2014 9:45:36 AM | SETI@home | URL http://setiathome.berkeley.edu/; Computer ID 6884721; resource share 100
3/4/2014 9:45:36 AM | World Community Grid | URL http://www.worldcommunitygrid.org/; Computer ID 2276963; resource share 100
3/4/2014 9:45:36 AM | World Community Grid | General prefs: from World Community Grid (last modified 18-Jun-2013 15:18:16)
3/4/2014 9:45:36 AM | World Community Grid | Computer location: home
3/4/2014 9:45:36 AM | World Community Grid | General prefs: no separate prefs for home; using your defaults
3/4/2014 9:45:36 AM | | Reading preferences override file
3/4/2014 9:45:36 AM | | Preferences:
3/4/2014 9:45:36 AM | | max memory usage when active: 4073.25MB
3/4/2014 9:45:36 AM | | max memory usage when idle: 7331.84MB
3/4/2014 9:45:36 AM | | max disk usage: 100.00GB
3/4/2014 9:45:36 AM | | max CPUs used: 6
3/4/2014 9:45:36 AM | | suspend work if non-BOINC CPU load exceeds 15%
3/4/2014 9:45:36 AM | | (to change preferences, visit a project web site or select Preferences in the Manager)
3/4/2014 9:45:36 AM | SETI@home | [error] no project URL in task state file
3/4/2014 9:45:36 AM | | Not using a proxy
3/4/2014 9:45:40 AM | | Running CPU benchmarks
3/4/2014 9:45:40 AM | | Suspending computation - CPU benchmarks in progress

---

Have you rebooted it since changing drivers? and what driver versions have you tried?

Claggy

I've rebooted several times to no avail.. I've been through drivers 331.65 (where I noticed problems and updated) then 332.21 and 332.50 most currently.

---

Is that one the main administrator account, or a user account? Is there any user switching going on?

Claggy

This is the main administrator account, yeah - and no, there is no user switching going on. Just for funzies I tried logging in from my network administrator account and it doesn't change - BOINC still runs, everything looks good, but it doesn't detect the GPU. The Nvidia stuff shows the proper card to be installed and it runs as it should - I'm not sure what to make of this..

---

Try what CUDA-Z have to say:
http://cuda-z.sourceforge.net/

---

 


- ALF - "Find out what you don't do well ..... then don't do it!" :)
 

I have no experience with QUADRO GPUs, but I don't see anything related to the installation of the drivers saying an upgrade or a clean install has been performed every time? A clean install might help? Or perhaps try out the card in another pc? Does the card work properly in applications it's meant to be used for in the first place?

---

To boldly crunch ...

Sometimes simply reseating the card can jigger loose some oxidation.

---

"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.

3/4/2014 9:45:36 AM | | GPU detection failed. error code -1073740940

This error is a different one than Error Code 1073740940. Error code -1073740940 (0xc0000374) means that a heap corruption occurred within the program. It can point to a problem with the compiler used (as far as I know, BOINC 7.2.41 was compiled with VS2005), or with the hardware itself.

If several other driver installs and BOINC restarts don't fix it, and a test with a newer beta BOINC (7.3.10 is at the time of typing the latest available) doesn't bring it back either, then it's hardware.

---

Sorry for the lengthy delay in responding..

I ordered a different video card and it failed too, with the same error code. I pulled the video card and started the machine with the onboard INTEL video and it popped right up as a GPU in BOINC.. I'm not sure what to think about this issue now..

---

I'm suddenly receiving the same type of Error. All I did was restart BOINC, Nothing Else. For some reason BOINC 7.2.33 OSX cannot find either of my two cards that were working fine a couple minutes before.

Mon Mar 17 11:50:25 2014 |  | Starting BOINC client version 7.2.33 for x86_64-apple-darwin
Mon Mar 17 11:50:25 2014 |  | log flags: file_xfer, sched_ops, task
Mon Mar 17 11:50:25 2014 |  | Libraries: libcurl/7.26.0 OpenSSL/1.0.1e zlib/1.2.5 c-ares/1.9.1
Mon Mar 17 11:50:25 2014 |  | Data directory: /Volumes/Mov1/BOINC/Maverick/BOINC Data
Mon Mar 17 11:50:25 2014 |  | GPU detection failed. error code 49408
Mon Mar 17 11:50:25 2014 |  | No usable GPUs found
Mon Mar 17 11:50:25 2014 | SETI@home | Found app_info.xml; using anonymous platform
Mon Mar 17 11:50:25 2014 |  | app version refers to missing GPU type ATI
Mon Mar 17 11:50:25 2014 | SETI@home Beta Test | Found app_info.xml; using anonymous platform
Mon Mar 17 11:50:25 2014 |  | app version refers to missing GPU type ATI
Mon Mar 17 11:50:25 2014 |  | app version refers to missing GPU type ATI
Mon Mar 17 11:50:25 2014 | SETI@home Beta Test | Missing coprocessor for task ap_25jn13ac_B0_P1_00384_20140308_21884.wu_2
Mon Mar 17 11:50:25 2014 | SETI@home | Missing coprocessor for task ap_04se13aa_B3_P0_00242_20140315_27902.wu_1...

WTH?

Hmmmmm....
Still works perfectly fine in Mountain Lion.

Mon Mar 17 12:15:13 2014 |  | Starting BOINC client version 7.2.33 for x86_64-apple-darwin
Mon Mar 17 12:15:13 2014 |  | log flags: file_xfer, sched_ops, task
Mon Mar 17 12:15:13 2014 |  | Libraries: libcurl/7.26.0 OpenSSL/1.0.1e zlib/1.2.5 c-ares/1.9.1
Mon Mar 17 12:15:13 2014 |  | Data directory: /Volumes/Mov1/BOINC/ML/BOINC Data
Mon Mar 17 12:15:13 2014 |  | OpenCL: AMD/ATI GPU 0: ATI Radeon Barts PRO Prototype (driver version 1.0, device version OpenCL 1.1, 1024MB, 1024MB available, 868 GFLOPS peak)
Mon Mar 17 12:15:13 2014 |  | OpenCL: AMD/ATI GPU 1: ATI Radeon Barts XT Prototype (driver version 1.0, device version OpenCL 1.1, 1024MB, 1024MB available, 1008 GFLOPS peak)
Mon Mar 17 12:15:13 2014 |  | OpenCL CPU: Intel(R) Xeon(R) CPU           E5462  @ 2.80GHz (OpenCL driver vendor: Apple, driver version 1.1, device version OpenCL 1.2)
Mon Mar 17 12:15:13 2014 | SETI@home | Found app_info.xml; using anonymous platform
Mon Mar 17 12:15:13 2014 |  | Processor: 4 GenuineIntel Intel(R) Xeon(R) CPU E5462 @ 2.80GHz [x86 Family 6 Model 23 Stepping 6]
Mon Mar 17 12:15:13 2014 |  | OS: Mac OS X 10.8.5 (Darwin 12.5.0)

Let's see. I Updated the nVidia Web Driver Last night in Mavericks. Some time later a New CUDA Driver popped up and I installed that, but since it didn't ask for anything other than CUDA Apps to be restarted, nothing was restarted. The first time I restart BOINC after installing the CUDA Driver, B00M, BOINC can't find my AMD cards...

Now When I restart Mavericks I get a popup about the CUDA Driver needing to be Updated but there isn't an Update Available....

Is this a SICK nVidia Joke? Or What?

Is that the 8800GT ? There has been a number of issues with them getting compute Cap 1.0 to work under various changing driver scenarios, with OpenCL in particular on Linux and Mac. IIRC the driver readme's have (or should have) some more detail about that.

---

"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.

Currently I don't have a nVidia card installed, see the last post. The nVidia Web driver was installed many months ago when I was testing the GTS250 with the Mac CUDA App. The only cards installed at present are 2 ATI cards, which nVidia has just killed...as far as BOINC is concerned. Everything else works fine.

How do I uninstall the Mac CUDA driver? There isn't an option in the CUDA Control Panel. There is an Option to uninstall the WEB driver in the Web Driver Control Panel. I've already selected to revert back to the OSX Driver but that doesn't seem to correct the Problem.

Currently I don't have a nVidia card installed, see the last post. The nVidia Web driver was installed many months ago when I was testing the GTS250 with the Mac CUDA App. The only cards installed at present are 2 ATI cards, which nVidia has just killed...as far as BOINC is concerned. Everything else works fine.

How do I uninstall the Mac CUDA driver? There isn't an option in the CUDA Control Panel. There is an Option to uninstall the WEB driver in the Web Driver Control Panel. I've already selected to revert back to the OSX Driver but that doesn't seem to correct the Problem.

Out of my depth there I'm afraid. The most likely scenario indeed would be the nvidia driver broke whatever the Mac equivalent of the OpenCL interconnect layer is, possibly having it disabled/knobbled somehow because of issues.

Does reinstalling/updating the AMD portion restore the OpenCL connectivity for those?

---

"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.

I guess if I ran the 10.9.2 Combined Update that should overwrite any changes to the AMD driver. I have no idea if that will correct the problem. In Mavericks, the CUDA Control Panel has RED around the installed driver and a note saying it has to be Updated. In Mountain Lion, which hasn't been 'Updated', the Cuda Control Panel just says "CUDA 5.5.47 Driver update is available" and there isn't any RED around the installed driver...

I guess if I ran the 10.9.2 Combined Update that should overwrite any changes to the AMD driver. I have no idea if that will correct the problem. In Mavericks, the CUDA Control Panel has RED around the installed driver and a note saying it has to be Updated. In Mountain Lion, which hasn't been 'Updated', the Cuda Control Panel just says "CUDA 5.5.47 Driver update is available" and there isn't any RED around the installed driver...

Yeah, that's where I'm at my limit. I know the Win7+ Windoze multiple driver stack setup pretty well, having seen all the problems with Vista and before. I'm not familiar at all with the Mac Driver arrangement to know the precise cause, nor if they went fully to multiple driver stack support or some (proprietary) mixture.

I do know firstly that there have been various specific issues with Compute cap 1.0, and Linux and Mac 32 Bit OpenCL (Both having moved mostly to 64 Bit).

Being Mavericks I assume the move to total 64 bit native is involved in the equation, but don't know how that might impact the OpenCL detection by the Boinc client. I suspect some of this may possibly be connected to older devices being 32 bit hardware only, but don't quote me on that. It's just a guess.

Might want to see if some other OpenCL dependant applciations, other than the Boinc client, pick up the AMDs properly. It wouldn't be the first time some OS level driver infrastructure change breaks given Boinc version detection.

---

"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.

Well, none of that works. Combined Update, BOINC 7.2.42, Still No GPUs.
LuxMark seems to work just Fine.

Mon Mar 17 15:02:21 2014 |  | Starting BOINC client version 7.2.42 for x86_64-apple-darwin
Mon Mar 17 15:02:21 2014 |  | log flags: file_xfer, sched_ops, task
Mon Mar 17 15:02:21 2014 |  | Libraries: libcurl/7.26.0 OpenSSL/1.0.1e zlib/1.2.5 c-ares/1.9.1
Mon Mar 17 15:02:21 2014 |  | Data directory: /Volumes/Mov1/BOINC/Maverick/BOINC Data
Mon Mar 17 15:02:21 2014 |  | GPU detection failed. error code 49408
Mon Mar 17 15:02:21 2014 |  | No usable GPUs found
Mon Mar 17 15:02:21 2014 | SETI@home | Found app_info.xml; using anonymous platform
Mon Mar 17 15:02:21 2014 |  | app version refers to missing GPU type ATI
Mon Mar 17 15:02:21 2014 | SETI@home Beta Test | Found app_info.xml; using anonymous platform
Mon Mar 17 15:02:21 2014 |  | app version refers to missing GPU type ATI
Mon Mar 17 15:02:21 2014 |  | app version refers to missing GPU type ATI
Mon Mar 17 15:02:21 2014 | SETI@home Beta Test | Missing coprocessor for task ap_25jn13ac_B0_P1_00384_20140308_21884.wu_2
Mon Mar 17 15:02:21 2014 |  | app version refers to missing GPU type ATI
Mon Mar 17 15:02:21 2014 | SETI@home | Missing coprocessor for task ap_04se13aa_B3_P0_00242_20140315_27902.wu_1
Mon Mar 17 15:02:21 2014 |  | OS: Mac OS X 10.9.2 (Darwin 13.1.0)
Mon Mar 17 15:02:21 2014 |  | Version change (7.2.33 -> 7.2.42)...

Now What?

nVidia! You got some 'splaining to do.