Posts by Icecold

1) Message boards : Number crunching : Windows GPU errors RTX 3080 (Message 6891)
Posted 31 Oct 2020 by Icecold

It is not. This is the results of running it:

mfaktc v0.21 (64bit built) Compiletime options THREADS_PER_BLOCK 256 SIEVE_SIZE_LIMIT 32kiB SIEVE_SIZE 193154bits SIEVE_SPLIT 250 MORE_CLASSES enabled Runtime options SievePrimes 25000 SievePrimesAdjust 1 SievePrimesMin 5000 SievePrimesMax 100000 NumStreams 3 CPUStreams 3 GridSize 3 GPU Sieving enabled GPUSievePrimes 82486 GPUSieveSize 64Mi bits GPUSieveProcessSize 16Ki bits Checkpoints enabled CheckpointDelay 300s WorkFileAddDelay disabled Stages enabled StopAfterFactor class PrintMode full V5UserID (none) ComputerID (none) AllowSleep no TimeStampInResults no CUDA version info binary compiled for CUDA 10.0 CUDA runtime version 10.0 CUDA driver version 11.10 CUDA device info name GeForce RTX 3080 compute capability 8.6 max threads per block 1024 max shared memory per MP 102400 byte number of multiprocessors 68 clock rate (CUDA cores) 1725MHz memory clock rate: 9501MHz memory bus width: 320 bit Automatic parameters threads per grid 557056 GPUSievePrimes (adjusted) 82486 GPUsieve minimum exponent 1055144 running a simple selftest... ERROR: cudaGetLastError() returned 48: no kernel image is available for execution on the device

That was with the standard executable that comes with BOINC, but it does the same thing on the one I compiled.

2) Message boards : Number crunching : Windows GPU errors RTX 3080 (Message 6889)
Posted 31 Oct 2020 by Icecold

On the default BOINC exe:

mfaktc v0.21 (64bit built) Compiletime options THREADS_PER_BLOCK 256 SIEVE_SIZE_LIMIT 32kiB SIEVE_SIZE 193154bits SIEVE_SPLIT 250 MORE_CLASSES enabled Runtime options SievePrimes 25000 SievePrimesAdjust 1 SievePrimesMin 5000 SievePrimesMax 100000 NumStreams 3 CPUStreams 3 GridSize 3 GPU Sieving enabled GPUSievePrimes 82486 GPUSieveSize 64Mi bits GPUSieveProcessSize 16Ki bits Checkpoints enabled CheckpointDelay 300s WorkFileAddDelay disabled Stages enabled StopAfterFactor class PrintMode full V5UserID (none) ComputerID (none) AllowSleep no TimeStampInResults no CUDA version info binary compiled for CUDA 10.0 CUDA runtime version 10.0 CUDA driver version 11.10 CUDA device info name GeForce RTX 3080 compute capability 8.6 max threads per block 1024 max shared memory per MP 102400 byte number of multiprocessors 68 clock rate (CUDA cores) 1725MHz memory clock rate: 9501MHz memory bus width: 320 bit Automatic parameters threads per grid 557056 GPUSievePrimes (adjusted) 82486 GPUsieve minimum exponent 1055144 ########## testcase 1/2867 ########## Starting trial factoring M50804297 from 2^67 to 2^68 (0.59 GHz-days) Using GPU kernel "75bit_mul32_gs" Date Time | class Pct | time ETA | GHz-d/day Sieve Wait Oct 31 14:16 | 3387 0.1% | 0.001 n.a. | n.a. 82485 n.a.% ERROR: cudaGetLastError() returned 48: no kernel image is available for execution on the device

3) Message boards : Number crunching : Windows GPU errors RTX 3080 (Message 6887)
Posted 31 Oct 2020 by Icecold

With the exe I compiled:

mfaktc v0.21 (64bit built) Compiletime options THREADS_PER_BLOCK 256 SIEVE_SIZE_LIMIT 32kiB SIEVE_SIZE 193154bits SIEVE_SPLIT 250 MORE_CLASSES enabled Runtime options SievePrimes 25000 SievePrimesAdjust 1 SievePrimesMin 5000 SievePrimesMax 100000 NumStreams 3 CPUStreams 3 GridSize 3 GPU Sieving enabled GPUSievePrimes 82486 GPUSieveSize 64Mi bits GPUSieveProcessSize 16Ki bits Checkpoints enabled CheckpointDelay 30s WorkFileAddDelay 600s Stages enabled StopAfterFactor bitlevel PrintMode full V5UserID (none) ComputerID (none) AllowSleep no TimeStampInResults no CUDA version info binary compiled for CUDA 11.10 CUDA runtime version 11.10 CUDA driver version 11.10 CUDA device info name GeForce RTX 3080 compute capability 8.6 max threads per block 1024 max shared memory per MP 102400 byte number of multiprocessors 68 clock rate (CUDA cores) 1725MHz memory clock rate: 9501MHz memory bus width: 320 bit Automatic parameters threads per grid 557056 GPUSievePrimes (adjusted) 82486 GPUsieve minimum exponent 1055144 ########## testcase 1/2867 ########## Starting trial factoring M50804297 from 2^67 to 2^68 (0.59 GHz-days) Using GPU kernel "75bit_mul32_gs" Date Time | class Pct | time ETA | GHz-d/day Sieve Wait Oct 31 14:11 | 3387 0.1% | 0.001 n.a. | n.a. 82485 n.a.% ERROR: cudaGetLastError() returned 98: invalid device function

mfaktc v0.21 (64bit built) Compiletime options THREADS_PER_BLOCK 256 SIEVE_SIZE_LIMIT 32kiB SIEVE_SIZE 193154bits SIEVE_SPLIT 250 MORE_CLASSES enabled Runtime options SievePrimes 25000 SievePrimesAdjust 1 SievePrimesMin 5000 SievePrimesMax 100000 NumStreams 3 CPUStreams 3 GridSize 3 GPU Sieving enabled GPUSievePrimes 82486 GPUSieveSize 64Mi bits GPUSieveProcessSize 16Ki bits Checkpoints enabled CheckpointDelay 30s WorkFileAddDelay 600s Stages enabled StopAfterFactor bitlevel PrintMode full V5UserID (none) ComputerID (none) AllowSleep no TimeStampInResults no CUDA version info binary compiled for CUDA 11.10 CUDA runtime version 11.10 CUDA driver version 11.10 CUDA device info name GeForce RTX 3080 compute capability 8.6 max threads per block 1024 max shared memory per MP 102400 byte number of multiprocessors 68 clock rate (CUDA cores) 1725MHz memory clock rate: 9501MHz memory bus width: 320 bit Automatic parameters threads per grid 557056 GPUSievePrimes (adjusted) 82486 GPUsieve minimum exponent 1055144 running a simple selftest... ERROR: cudaGetLastError() returned 98: invalid device function

4) Message boards : Number crunching : Windows GPU errors RTX 3080 (Message 6883)
Posted 31 Oct 2020 by Icecold

I appreciate the reply! I don't think there is a 10.x driver version that works with RTX 3080. I tried compiling from source and specifying 11.10 but it still errors - this task shows the following:

CUDA version info
binary compiled for CUDA 11.10
CUDA runtime version 11.10
CUDA driver version 11.10

5) Message boards : Number crunching : Windows GPU errors RTX 3080 (Message 6881)
Posted 31 Oct 2020 by Icecold

I was working under the thought process that the RTX 3080 may not work correctly with CUDA 10.1. I compiled mfakt from source specifying CUDA 11.10 but am still getting errors. This task is worth looking at http://srbase.my-firewall.org/sr5/result.php?resultid=22792002

6) Message boards : Number crunching : Windows GPU errors RTX 3080 (Message 6880)
Posted 31 Oct 2020 by Icecold

I just got an RTX 3080. I'm running Windows 10 and I installed the 10.1 CUDA Toolkit. SRBase errors after 3 seconds of running. These are the tasks - http://srbase.my-firewall.org/sr5/results.php?hostid=210547

The error seems to be ERROR: cudaGetLastError() returned 48: no kernel image is available for execution on the device. I've never ran SRBase on this machine before(and normally run Linux), but Primegrid PPS Sieve which uses CUDA seems to run fine. Any ideas?