WARNING Do not upgrade to 375/376.xx drivers (for xx<48)

[Glennmas]

19:53:12:WARNING:WU00:FS00:Changed SMP threads from 7 to 6 this can cause some work units to fail
19:53:12:WU00:FS00:Running FahCore: "C:\Program Files (x86)\FAHClient/FAHCoreWrapper.exe" C:/Users/Gebruiker/AppData/Roaming/FAHClient/cores/web.stanford.edu/~pande/Win32/AMD64/Core_a4.fah/FahCore_a4.exe -dir 00 -suffix 01 -version 704 -lifeline 13164 -checkpoint 15 -np 6
19:53:12:WU00:FS00:Started FahCore on PID 10964
19:53:12:WU00:FS00:Core PID:6480
19:53:12:WU00:FS00:FahCore 0xa4 started
19:53:12:WU00:FS00:0xa4:
19:53:12:WU00:FS00:0xa4:*------------------------------*
19:53:12:WU00:FS00:0xa4:Folding@Home Gromacs GB Core
19:53:12:WU00:FS00:0xa4:Version 2.27 (Dec. 15, 2010)
19:53:12:WU00:FS00:0xa4:
19:53:12:WU00:FS00:0xa4:Preparing to commence simulation
19:53:12:WU00:FS00:0xa4:- Looking at optimizations...
19:53:12:WU00:FS00:0xa4:- Files status OK
19:53:12:WU00:FS00:0xa4:- Expanded 825691 -> 1403132 (decompressed 169.9 percent)
19:53:12:WU00:FS00:0xa4:Called DecompressByteArray: compressed_data_size=825691 data_size=1403132, decompressed_data_size=1403132 diff=0
19:53:12:WU00:FS00:0xa4:- Digital signature verified
19:53:12:WU00:FS00:0xa4:
19:53:12:WU00:FS00:0xa4:Project: 9036 (Run 446, Clone 5, Gen 158)
19:53:12:WU00:FS00:0xa4:
19:53:12:WU00:FS00:0xa4:Assembly optimizations on if available.
19:53:12:WU00:FS00:0xa4:Entering M.D.
19:53:18:WU00:FS00:0xa4:Mapping NT from 6 to 6
19:53:18:WU00:FS00:0xa4:Completed 0 out of 250000 steps (0%)
19:54:19:WU00:FS00:0xa4:Completed 2500 out of 250000 steps (1%)
this are the loggs you would need it gues

[Glennmas]

managed to get it working again but their are 3wus appearing right now 1 for gpu 1for cpu and 1 for nothing and its wu 0 and keeps downloading but doesnt matter im up and running again

[bruce]

Sure, it could be driver related ... but not necessarily. Post the first few pages of FAH's log.

Did you do a clean install?

[SombraGuerrero]

From the looks of the news on Geforce Experience, it looks like 70 is the next planned Game Ready release. I've never used Nvidia's hotfixes before this point. I would assume they get merged back into the main line. I'd bet Nvidia is still holding out hope that OpenMM will fix the underlying bug though.

[foldy]

I think the hotfix driver only means it was not tested widely and there could be issues arise in some games as side effect.
So after some longer tests and fixes it gets into next release. But for folding we can use the hotfix driver now.

There is also no difference in the installer package compared to a beta or WHQL release driver.
So you just can install it as update over any previous driver version.

[Glennmas]

i did

[silverpulser]

The 376.48 Hot Fix driver seems to work for me. I am using a Windows 10, 64 bit Pro, release build 14393.576 machine with a GTX 750Ti GPU. Currently running a Core 21 WU.

[SombraGuerrero]

There's someone back on the EVGA forums reporting clFlush and max retry limit exceptions on core_18 work units using a 1080 with the hotfix driver. Has anyone else seen problems with core_18 with 376.48?

[ComputerGenie]

There's someone back on the EVGA forums reporting clFlush and max retry limit exceptions on core_18 work units using a 1080 with the hotfix driver. Has anyone else seen problems with core_18 with 376.48?

I dug through my logs and project 9677 ran on mine without a hitch; however, it's the only 18 that's been loaded, so I'm not sure a data-point of 1 counts one way or the other.

[SombraGuerrero]

Well I have been having no issues with core_18 either as I've watched a couple core_18 WUS process through, so it must be an environmental thing. This person did mention that it was only a problem on his 1080 and not his other 980. I think if there's one thing I've learned from all the discussion around this bug, it's definitely that the architecture in the Geforce 10 series is substantially different from model to model and line to line, especially when compared to previous generations.

[red454]

I did the hotfix yesterday and my Titan X is running the 3rd WU on core 21 with no issues so far... 556207 PPD currently folding for Team 71836 (Roche Diagnostics)

[frederikhk]

I ran multiple WU's with the hotfix driver and this is the only one i've had fail.

11:34:15:WU01:FS01:0x21:ERROR:exception: Error downloading array energyBuffer: clEnqueueReadBuffer (-5)
11:34:15:WU01:FS01:0x21:Saving result file logfile_01.txt
11:34:15:WU01:FS01:0x21:Saving result file log.txt
11:34:15:WU01:FS01:0x21:Folding@home Core Shutdown: BAD_WORK_UNIT
11:34:17:WARNING:WU01:FS01:FahCore returned: BAD_WORK_UNIT (114 = 0x72)
11:34:18:WU01:FS01:Sending unit results: id:01 state:SEND error:FAULTY project:11403 run:6 clone:3 gen:230 core:0x21 unit:0x0000013c8ca304f255ed4f6627d97d11
11:34:18:WU01:FS01:Uploading 3.35KiB to 140.163.4.242
11:34:18:WU01:FS01:Connecting to 140.163.4.242:8080
11:34:18:WU01:FS01:Upload complete
11:34:18:WU01:FS01:Server responded WORK_ACK (400)


I do not know if this is a WU related problem or a driver problem.

[Joe_H]

Welcome to the folding support forum frederikhk.

The WU in question has since been successfully folded by someone else:

Hi frederikhk (team 111065),
Your WU (P11403 R6 C3 G230) was added to the stats database on 2016-12-26 04:08:18 for 7308.9 points of credit.

Hi ******** (team ******),
Your WU (P11403 R6 C3 G230) was added to the stats database on 2016-12-26 11:07:39 for 82346.9 points of credit.

You received partial credit for the WU.

Code: Select all

11:34:15:WU01:FS01:0x21:ERROR:exception: Error downloading array energyBuffer: clEnqueueReadBuffer (-5)

This is the same error reported when other WU's have failed using the pre-hotfix version 375 and 376 drivers. Possibly the workaround code in the hotfix driver was not used or did not account for all issues that could come up during WU processing.

[frederikhk]

I suspected the same thing. Thank you!

[n_w95482]

I had the same issue with the hotfix driver on my laptop (GTX 960M). Most WUs were fine, but occasionally it would throw that clEnqueueReadBuffer error (from what I could see, all were on core 21). The GPU would also randomly drop to 405 MHz and stay there until after a reboot. I went back to 372.70 and it has been cranking away happily ever since.