Message boards :
Number crunching :
GPU Detection Failed, Error Code 1073740940
Message board moderation
Previous · 1 · 2
Author | Message |
---|---|
Batter Up Send message Joined: 5 May 99 Posts: 1946 Credit: 24,860,347 RAC: 0 |
I know nothing but why are you blaming NVIDIA for BOINC not finding an ATI card? |
TBar Send message Joined: 22 May 99 Posts: 5204 Credit: 840,779,836 RAC: 2,768 |
Probably Because it worked Just Fine BEFORE I installed the CUDA *UpDate* that now has a Big RED *UpDate Required* in the CUDA Control Panel. There is also a note that says; CUDA Driver update required It Also says "No Newer CUDA Driver Available" Now When I restart Mavericks I get a popup about the CUDA Driver needing to be Updated but there isn't an Update Available.... The CUDA Driver Needs to be Nuked. :-/ |
rob smith Send message Joined: 7 Mar 03 Posts: 22158 Credit: 416,307,556 RAC: 380 |
Did you get your update from MickySoft's site or from Nvidia/Geforce - the former are known to mess things up quite well when updating Nvidia GPU drivers. (Tee-shirt obtained and not wanted) Bob Smith Member of Seti PIPPS (Pluto is a Planet Protest Society) Somewhere in the (un)known Universe? |
TBar Send message Joined: 22 May 99 Posts: 5204 Credit: 840,779,836 RAC: 2,768 |
Did you get your update from MickySoft's site or from Nvidia/Geforce - the former are known to mess things up quite well when updating Nvidia GPU drivers. (Tee-shirt obtained and not wanted) As mentioned Previously, The *Update* was preformed by the Offical nVidia Control Panel. There are 2 Mac nVidia Control Panels, one for the Web Driver, the other for the CUDA Driver. I'm off to nVidia to see how to remove this faulty CUDA Driver. |
jason_gee Send message Joined: 24 Nov 06 Posts: 7489 Credit: 91,093,184 RAC: 0 |
LuxMark seems to work just Fine. Hmmm, If Luxmark detects & uses OpenCL devices fine, that seems to imply one or another change has run afoul of something in the Boinc client detection code (probably some fault there). So if OpenCL is working (on the AMDs), and Boinc client isn't detecting right, I'd recommend filing a bug report there (as opposed to nVidia). It's not the first time dicey detection has been found in Boinc client code. In the past, something similar has happened on Windows Boinc client, with Cuda detection (as opposed to openCL). In that case it was restricted to Optimus enabled machines, with the Boinc client setting a parameter improperly. For Most systems it appeared to work fine, but as soon as the improper parameter came into play with optimus sleeping devices, no devices were detected [or now I recall, incorrect 'driver version unknown' in detection logs.]. Probably something like that's going on IMO. [Edit:] looking at boinc client gpu_openCL.cpp detection to see if it matches recommended practices. "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. |
TBar Send message Joined: 22 May 99 Posts: 5204 Credit: 840,779,836 RAC: 2,768 |
IT WORKS!!!! Live chat, they told me to install CUDA 5.5.25, http://www.nvidia.com/object/macosx-cuda-5.5.25-driver.html Mon Mar 17 16:19:38 2014 | | Starting BOINC client version 7.2.42 for x86_64-apple-darwin Mon Mar 17 16:19:38 2014 | | log flags: file_xfer, sched_ops, task Mon Mar 17 16:19:38 2014 | | Libraries: libcurl/7.26.0 OpenSSL/1.0.1e zlib/1.2.5 c-ares/1.9.1 Mon Mar 17 16:19:38 2014 | | Data directory: /Volumes/Mov1/BOINC/Maverick/BOINC Data Mon Mar 17 16:19:38 2014 | | OpenCL: AMD/ATI GPU 0: ATI Radeon Barts XT Prototype (driver version 1.2 (Jan 28 2014 20:11:47), device version OpenCL 1.2, 1024MB, 1024MB available, 1008 GFLOPS peak) Mon Mar 17 16:19:38 2014 | | OpenCL: AMD/ATI GPU 1: ATI Radeon Barts PRO Prototype (driver version 1.2 (Jan 28 2014 20:11:47), device version OpenCL 1.2, 1024MB, 1024MB available, 868 GFLOPS peak) Mon Mar 17 16:19:38 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 16:19:38 2014 | SETI@home | Found app_info.xml; using anonymous platform Mon Mar 17 16:19:38 2014 | SETI@home Beta Test | Found app_info.xml; using anonymous platform Mon Mar 17 16:19:38 2014 | | Host name: TomsMacPro.local Mon Mar 17 16:19:38 2014 | | Processor: 4 GenuineIntel Intel(R) Xeon(R) CPU E5462 @ 2.80GHz [x86 Family 6 Model 23 Stepping 6] Mon Mar 17 16:19:38 2014 | | OS: Mac OS X 10.9.2 (Darwin 13.1.0) Mon Mar 17 16:19:38 2014 | | Local time is UTC -4 hours Mon Mar 17 16:19:38 2014 | | Config: use all coprocessors Mon Mar 17 16:19:38 2014 | SETI@home | URL http://setiathome.berkeley.edu/; Computer ID 6796479; resource share 200 Mon Mar 17 16:19:38 2014 | SETI@home Beta Test | URL http://setiweb.ssl.berkeley.edu/beta/; Computer ID 63959; resource share 8... They suggested I don't run the Updater without a nVidia card installed....however, it seems to have worked with 5.5.25 ??? Now when the APs disappear for over a week, like they are probably going to do, I can swap out the 6850 & GTS250 and see if it works again. Jason, do you have a Updated Mac CUDA App yet? :-) Hey, what happened to the uploads? It's broken AGAIN? |
jason_gee Send message Joined: 24 Nov 06 Posts: 7489 Credit: 91,093,184 RAC: 0 |
Congrats :) the Boinc client OpenCL detection code looks a little spaghettified, so I'm glad I don't have to keep digging in there (yet). Jason, do you have a Updated Mac CUDA App yet? With x41zc, I know Edward's been fighting Cuda library issues over on CA, so I'm currently tossing around my options to make me of more help there (Namely acquisition of an NV equipped MAC). With some other things out of the way I should hopefully be able to test my remote Mac build environment soon (kind friend's Mac) ) starting with basic detection right through test pieces then app build if working. In general development my end, some business associates (non-boinc-seti related) have read me the riot act, demanding that x42 be a complete 'proper' refactor/re-engineering effort, and cross platform (WIN/Linux/MAC/Arm-Cuda-readyish). That's slowed things quite a bit, particularly with the amount of wrapping of Boinc libraries I'll have to do to meet standards... But, practicalities ironed out, it does all mean the first 3 platforms should get x42phase1 tests at roughly the same time, and ARM-Cuda support relatively straightforward when it's available. "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. |
Jord Send message Joined: 9 Jun 99 Posts: 15184 Credit: 4,362,181 RAC: 3 |
Sorry for the lengthy delay in responding.. So, now that the others have had their say in this thread, back to the OP. If a different videocard in the same PCIe slot doesn't work either, it sounds as if you have a problem with the PCIe slot. Do you have more than one slot, so you can test another one, even if it's at a slower speed? Edit: Did you use the same power connectors between the videocards? You may want to check that they're still in good nick as well, and else try another connector (with molex -> PCIe-6pin converter, if necessary) or PSU. |
TBar Send message Joined: 22 May 99 Posts: 5204 Credit: 840,779,836 RAC: 2,768 |
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 running the Driver Uninstaller App and completely remove the Driver. Then Restart and Reinstall the Driver. Display Driver Uninstaller (DDU) / Cleaner made for Display Drivers |
Claggy Send message Joined: 5 Jul 99 Posts: 4654 Credit: 47,537,079 RAC: 4 |
I was thinking more along the lines of a problem with the OS installation, maybe if the OP has a spare drive he could try a fresh installation and see if the GPU detection works on that. Claggy |
Claggy Send message Joined: 5 Jul 99 Posts: 4654 Credit: 47,537,079 RAC: 4 |
In general development my end, some business associates (non-boinc-seti related) have read me the riot act, demanding that x42 be a complete 'proper' refactor/re-engineering effort, and cross platform (WIN/Linux/MAC/Arm-Cuda-readyish). That's slowed things quite a bit, particularly with the amount of wrapping of Boinc libraries I'll have to do to meet standards... I thought that was rather precocious of those business associates, considering you're an (as afar as I know) unpaid volunteer developer, I would hope there was some payment, or at the very least some hardware, Which brings us to this post at Einstein: http://einstein.phys.uwm.edu/forum_thread.php?id=9686&nowrap=true#130145 Maybe whowever those business associates are, they'll supply you with one of those Jetson TK1 Development Kits, ;) Claggy |
jason_gee Send message Joined: 24 Nov 06 Posts: 7489 Credit: 91,093,184 RAC: 0 |
In general development my end, some business associates (non-boinc-seti related) have read me the riot act, demanding that x42 be a complete 'proper' refactor/re-engineering effort, and cross platform (WIN/Linux/MAC/Arm-Cuda-readyish). That's slowed things quite a bit, particularly with the amount of wrapping of Boinc libraries I'll have to do to meet standards... LoL, I thought so too. Periodic contract work that pays the bills, and no not for nVidia, so I do what they say (If I want more work ;) ). Still haven't received my T-shirt for helping NV with Cuda 6 development, so I'm not counting on any hardware. Those prototype ARM+Kepler GPU prototyping boards were sent out, I believe, to test Cuda + ARM toolchain functionality in advance of Maxwell, (750ti being the architectural test of the GPU portion) though that's full of surmising. "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. |
©2024 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.