View previous topic :: View next topic |
Author |
Message |
hanj Veteran
Joined: 19 Aug 2003 Posts: 1490
|
Posted: Tue Mar 24, 2020 1:59 pm Post subject: |
|
|
Anyone have a process of getting the 32bit version installed on older gentoo systems? I have a pile of x86 machines doing file storage, etc. Would love to get these installed. I see 32bit linux .deb+.rpm on their download page for 7.5.1
Thanks!
hanji _________________ Server Admin Blog - Uno-Code.com |
|
Back to top |
|
|
Decibels Veteran
Joined: 16 Aug 2002 Posts: 1623 Location: U.S.A.
|
Posted: Tue Mar 24, 2020 2:57 pm Post subject: |
|
|
axl wrote: | Do you have opencl flag enabled? Could emerge clinfo and try to run that to try to diagnose opencl. |
Well finally got cuda working and more info out of clinfo, but still won't fold on the GPU.
Have latest:
nvidia-cuda-sdk-10.1.243 with cuda and opencl useflags
nvidia-cuda-toolkit-10.2.89-r1
nvidia-drivers-440.59 had to do the libglvnd upgrading stuff.
Then got this after rebooting:
Quote: | ./deviceQuery
./deviceQuery Starting...
CUDA Device Query (Runtime API) version (CUDART static linking)
Detected 1 CUDA Capable device(s)
Device 0: "GeForce GTX 1060 3GB"
CUDA Driver Version / Runtime Version 10.2 / 10.2
CUDA Capability Major/Minor version number: 6.1
Total amount of global memory: 3014 MBytes (3160735744 bytes)
( 9) Multiprocessors, (128) CUDA Cores/MP: 1152 CUDA Cores
GPU Max Clock rate: 1772 MHz (1.77 GHz)
Memory Clock rate: 4004 Mhz
Memory Bus Width: 192-bit
L2 Cache Size: 1572864 bytes
Maximum Texture Dimension Size (x,y,z) 1D=(131072), 2D=(131072, 65536), 3D=(16384, 16384, 16384)
Maximum Layered 1D Texture Size, (num) layers 1D=(32768), 2048 layers
Maximum Layered 2D Texture Size, (num) layers 2D=(32768, 32768), 2048 layers
Total amount of constant memory: 65536 bytes
Total amount of shared memory per block: 49152 bytes
Total number of registers available per block: 65536
Warp size: 32
Maximum number of threads per multiprocessor: 2048
Maximum number of threads per block: 1024
Max dimension size of a thread block (x,y,z): (1024, 1024, 64)
Max dimension size of a grid size (x,y,z): (2147483647, 65535, 65535)
Maximum memory pitch: 2147483647 bytes
Texture alignment: 512 bytes
Concurrent copy and kernel execution: Yes with 2 copy engine(s)
Run time limit on kernels: Yes
Integrated GPU sharing Host Memory: No
Support host page-locked memory mapping: Yes
Alignment requirement for Surfaces: Yes
Device has ECC support: Disabled
Device supports Unified Addressing (UVA): Yes
Device supports Compute Preemption: Yes
Supports Cooperative Kernel Launch: Yes
Supports MultiDevice Co-op Kernel Launch: Yes
Device PCI Domain ID / Bus ID / location ID: 0 / 1 / 0
Compute Mode:
< Default (multiple host threads can use ::cudaSetDevice() with device simultaneously) >
deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 10.2, CUDA Runtime Version = 10.2, NumDevs = 1, Device0 = GeForce GTX 1060 3GB
Result = PASS
|
Quote: | clinfo
Number of platforms 2
Platform Name NVIDIA CUDA
Platform Vendor NVIDIA Corporation
Platform Version OpenCL 1.2 CUDA 10.2.131
Platform Profile FULL_PROFILE
Platform Extensions cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_fp64 cl_khr_byte_addressable_store cl_khr_icd cl_khr_gl_sharing cl_nv_compiler_options cl_nv_device_attribute_query cl_nv_pragma_unroll cl_nv_copy_opts cl_nv_create_buffer cl_khr_int64_base_atomics cl_khr_int64_extended_atomics
Platform Extensions function suffix NV
Platform Name Clover
Platform Vendor Mesa
Platform Version OpenCL 1.1 Mesa 19.3.5
Platform Profile FULL_PROFILE
Platform Extensions cl_khr_icd
Platform Extensions function suffix MESA
Platform Name NVIDIA CUDA
Number of devices 1
Device Name GeForce GTX 1060 3GB
Device Vendor NVIDIA Corporation
Device Vendor ID 0x10de
Device Version OpenCL 1.2 CUDA
Driver Version 440.59
Device OpenCL C Version OpenCL C 1.2
Device Type GPU
Device Topology (NV) PCI-E, 01:00.0
Device Profile FULL_PROFILE
Device Available Yes
Compiler Available Yes
Linker Available Yes
Max compute units 9
Max clock frequency 1771MHz
Compute Capability (NV) 6.1
Device Partition (core)
Max number of sub-devices 1
Supported partition types None
Supported affinity domains (n/a)
Max work item dimensions 3
Max work item sizes 1024x1024x64
Max work group size 1024
Preferred work group size multiple 32
Warp size (NV) 32
Preferred / native vector sizes
char 1 / 1
short 1 / 1
int 1 / 1
long 1 / 1
half 0 / 0 (n/a)
float 1 / 1
double 1 / 1 (cl_khr_fp64)
Half-precision Floating-point support (n/a)
Single-precision Floating-point support (core)
Denormals Yes
Infinity and NANs Yes
Round to nearest Yes
Round to zero Yes
Round to infinity Yes
IEEE754-2008 fused multiply-add Yes
Support is emulated in software No
Correctly-rounded divide and sqrt operations Yes
Double-precision Floating-point support (cl_khr_fp64)
Denormals Yes
Infinity and NANs Yes
Round to nearest Yes
Round to zero Yes
Round to infinity Yes
IEEE754-2008 fused multiply-add Yes
Support is emulated in software No
Address bits 64, Little-Endian
Global memory size 3160735744 (2.944GiB)
Error Correction support No
Max memory allocation 790183936 (753.6MiB)
Unified memory for Host and Device No
Integrated memory (NV) No
Minimum alignment for any data type 128 bytes
Alignment of base address 4096 bits (512 bytes)
Global Memory cache type Read/Write
Global Memory cache size 442368 (432KiB)
Global Memory cache line size 128 bytes
Image support Yes
Max number of samplers per kernel 32
Max size for 1D images from buffer 268435456 pixels
Max 1D or 2D image array size 2048 images
Max 2D image size 16384x32768 pixels
Max 3D image size 16384x16384x16384 pixels
Max number of read image args 256
Max number of write image args 16
Local memory type Local
Local memory size 49152 (48KiB)
Registers per block (NV) 65536
Max number of constant args 9
Max constant buffer size 65536 (64KiB)
Max size of kernel argument 4352 (4.25KiB)
Queue properties
Out-of-order execution Yes
Profiling Yes
Prefer user sync for interop No
Profiling timer resolution 1000ns
Execution capabilities
Run OpenCL kernels Yes
Run native kernels No
Kernel execution timeout (NV) Yes
Concurrent copy and kernel execution (NV) Yes
Number of async copy engines 2
printf() buffer size 1048576 (1024KiB)
Built-in kernels (n/a)
Device Extensions cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_fp64 cl_khr_byte_addressable_store cl_khr_icd cl_khr_gl_sharing cl_nv_compiler_options cl_nv_device_attribute_query cl_nv_pragma_unroll cl_nv_copy_opts cl_nv_create_buffer cl_khr_int64_base_atomics cl_khr_int64_extended_atomics
Platform Name Clover
Number of devices 0
NULL platform behavior
clGetPlatformInfo(NULL, CL_PLATFORM_NAME, ...) No platform
clGetDeviceIDs(NULL, CL_DEVICE_TYPE_ALL, ...) Success [NV]
clCreateContext(NULL, ...) [default] Success [NV]
clCreateContext(NULL, ...) [other] `��&V
clCreateContextFromType(NULL, CL_DEVICE_TYPE_DEFAULT) No platform
clCreateContextFromType(NULL, CL_DEVICE_TYPE_CPU) No devices found in platform
clCreateContextFromType(NULL, CL_DEVICE_TYPE_GPU) No platform
clCreateContextFromType(NULL, CL_DEVICE_TYPE_ACCELERATOR) No devices found in platform
clCreateContextFromType(NULL, CL_DEVICE_TYPE_CUSTOM) Invalid device type for platform
clCreateContextFromType(NULL, CL_DEVICE_TYPE_ALL) No platform
|
Still get this in log.txt though:
Quote: | [93m14:33:43:WARNING:WU01:FS01:Failed to get assignment from '18.218.241.186:80': No WUs available for this configuration[0m
[91m14:33:43:ERROR:WU01:FS01:Exception: Could not get an assignment[0m |
hariskar
Quote: | Maybe try with value 0? |
FAH won't start then. Says does but doesn't. Setting to 1 or -1 works, but still get the above that No WUs available for this configuration.
But if change to -1 it assigns it and still get this, just different slot number:
Quote: | [93m14:56:03:WARNING:WU01:FS4294967295:Failed to get assignment from '18.218.241.186:80': No WUs available for this configuration[0m
[91m14:56:03:ERROR:WU01:FS4294967295:Exception: Could not get an assignment[0m |
_________________ Support bacteria – they’re the only culture some people have.”
– Steven Wright |
|
Back to top |
|
|
axl Veteran
Joined: 11 Oct 2002 Posts: 1144 Location: Romania
|
Posted: Tue Mar 24, 2020 3:02 pm Post subject: |
|
|
Decibels wrote: | Still get this in log.txt though:
Quote: | [93m14:33:43:WARNING:WU01:FS01:Failed to get assignment from '18.218.241.186:80': No WUs available for this configuration[0m
[91m14:33:43:ERROR:WU01:FS01:Exception: Could not get an assignment[0m |
|
You need to watch FOH when it starts if it says anything about finding opencl. The fact that you don't get a WorkUnit immediately is quite common. But the logs at start should say if GPU has been initialized correctly. |
|
Back to top |
|
|
Decibels Veteran
Joined: 16 Aug 2002 Posts: 1623 Location: U.S.A.
|
Posted: Tue Mar 24, 2020 3:26 pm Post subject: |
|
|
axl wrote: | Decibels wrote: | Still get this in log.txt though:
Quote: | [93m14:33:43:WARNING:WU01:FS01:Failed to get assignment from '18.218.241.186:80': No WUs available for this configuration[0m
[91m14:33:43:ERROR:WU01:FS01:Exception: Could not get an assignment[0m |
|
You need to watch FOH when it starts if it says anything about finding opencl. The fact that you don't get a WorkUnit immediately is quite common. But the logs at start should say if GPU has been initialized correctly. |
I have been. It says this:
Quote: | 15:19:58:Trying to access database...
15:19:58:Successfully acquired database lock
15:19:58:Enabled folding slot 00: PAUSED cpu:3 (by user)
15:19:58:Enabled folding slot -1: PAUSED gpu:0:GP106 [GeForce GTX 1060 3GB] 3935 (by user)
15:19:59:8:127.0.0.1:New Web connection |
But in system section says this:
Quote: | 15:19:58:******************************* System ********************************
15:19:58: CPU: Intel(R) Core(TM) i5-3470 CPU @ 3.20GHz
15:19:58: CPU ID: GenuineIntel Family 6 Model 58 Stepping 9
15:19:58: CPUs: 4
15:19:58: Memory: 7.76GiB
15:19:58: Free Memory: 4.38GiB
15:19:58: Threads: POSIX_THREADS
15:19:58: OS Version: 4.19
15:19:58: Has Battery: false
15:19:58: On Battery: false
15:19:58: UTC Offset: -5
15:19:58: PID: 7466
15:19:58: CWD: /opt/foldingathome
15:19:58: OS: Linux 4.19.66-gentoo x86_64
15:19:58: OS Arch: AMD64
15:19:58: GPUs: 1
15:19:58: GPU 0: Bus:1 Slot:0 Func:0 NVIDIA:7 GP106 [GeForce GTX 1060 3GB] 3935
15:19:58:CUDA Device 0: Platform:0 Device:0 Bus:1 Slot:0 Compute:6.1 Driver:10.2
15:19:58: OpenCL: Not detected: clGetDeviceIDs() returned -1
15:19:58:*********************************************************************** |
Think just need to give up and let the CPU run. _________________ Support bacteria – they’re the only culture some people have.”
– Steven Wright |
|
Back to top |
|
|
axl Veteran
Joined: 11 Oct 2002 Posts: 1144 Location: Romania
|
Posted: Tue Mar 24, 2020 3:28 pm Post subject: |
|
|
eselect opencl set to nvidia. also maybe use -opencl for mesa. You don't need mesa's version of opencl. I think you now have 2 opencl implementations. |
|
Back to top |
|
|
Decibels Veteran
Joined: 16 Aug 2002 Posts: 1623 Location: U.S.A.
|
Posted: Tue Mar 24, 2020 4:13 pm Post subject: |
|
|
axl wrote: | eselect opencl set to nvidia. also maybe use -opencl for mesa. You don't need mesa's version of opencl. I think you now have 2 opencl implementations. |
You are correct. Trying desperately to get opencl working and not sure that CUDA on nvidia implemented it. At least that is my understanding. I had it in mesa as useflag. So took it out. Restarted and got this:
(oh and had eselect opencl as nvidia)
Quote: | 15:57:16:******************************* System ********************************
15:57:16: CPU: Intel(R) Core(TM) i5-3470 CPU @ 3.20GHz
15:57:16: CPU ID: GenuineIntel Family 6 Model 58 Stepping 9
15:57:16: CPUs: 4
15:57:16: Memory: 7.76GiB
15:57:16: Free Memory: 7.57GiB
15:57:16: Threads: POSIX_THREADS
15:57:16: OS Version: 4.19
15:57:16: Has Battery: false
15:57:16: On Battery: false
15:57:16: UTC Offset: -5
15:57:16: PID: 3471
15:57:16: CWD: /opt/foldingathome
15:57:16: OS: Linux 4.19.66-gentoo x86_64
15:57:16: OS Arch: AMD64
15:57:16: GPUs: 1
15:57:16: GPU 0: Bus:1 Slot:0 Func:0 NVIDIA:7 GP106 [GeForce GTX 1060 3GB] 3935
15:57:16: CUDA Device 0: Platform:0 Device:0 Bus:1 Slot:0 Compute:6.1 Driver:10.2
15:57:16:OpenCL Device 0: Platform:0 Device:0 Bus:1 Slot:0 Compute:1.2 Driver:440.59 |
Config:
Quote: | 16:06:28:<config>
16:06:28: <!-- Folding Slot Configuration -->
16:06:28: <client-type v='advanced'/>
16:06:28: <opencl-index v='-1'/>
16:06:28:
16:06:28: <!-- Slot Control -->
16:06:28: <power v='FULL'/>
16:06:28:
16:06:28: <!-- User Information -->
16:06:28: <passkey v='********************************'/>
16:06:28: <team v='*****'/>
16:06:28: <user v='*****'/>
16:06:28:
16:06:28: <!-- Folding Slots -->
16:06:28: <slot id='0' type='CPU'/>
16:06:28: <slot id='18446744073709551615' type='GPU'/>
16:06:28:</config> |
Tried 0 and -1 and think 1 in opencl-index and still nogo. _________________ Support bacteria – they’re the only culture some people have.”
– Steven Wright |
|
Back to top |
|
|
axl Veteran
Joined: 11 Oct 2002 Posts: 1144 Location: Romania
|
Posted: Tue Mar 24, 2020 4:16 pm Post subject: |
|
|
don't worry. at one point you will pick a working unit and start crunching data. there are tones of contributors. they didn't expect this many and I guess they have a bottle neck there. at least the start up procedure is indicating that you setup things correctly. it found your drivers correctly and it doesn't show it as paused anymore. |
|
Back to top |
|
|
Decibels Veteran
Joined: 16 Aug 2002 Posts: 1623 Location: U.S.A.
|
Posted: Tue Mar 24, 2020 4:26 pm Post subject: |
|
|
I think it was just paused cause it wouldn't take the new config numbers trying unless stopped it and started it.
Ya, see a lot of people having problems either with GPU's not doing the crunching. So hopefully it works itself out.
Thanks for helping me with getting the CUDA and OpenCL working! _________________ Support bacteria – they’re the only culture some people have.”
– Steven Wright |
|
Back to top |
|
|
Decibels Veteran
Joined: 16 Aug 2002 Posts: 1623 Location: U.S.A.
|
Posted: Tue Mar 24, 2020 4:34 pm Post subject: |
|
|
It's working now. Probably from all the emerging and useflag changes,.... Jeessh.
Then finally changed the opencl-index to 0 from -1 and now using the GPU.
Hurahhhaa. Thanks a lot. Science Thanks You Also.
Well maybe not. Cause the log.txt is still showing -1 for opencl-index.
Got it going full speed now, so will test it later and let it run for awhile. _________________ Support bacteria – they’re the only culture some people have.”
– Steven Wright
Last edited by Decibels on Tue Mar 24, 2020 4:37 pm; edited 1 time in total |
|
Back to top |
|
|
axl Veteran
Joined: 11 Oct 2002 Posts: 1144 Location: Romania
|
Posted: Tue Mar 24, 2020 4:36 pm Post subject: |
|
|
I went through the same issues. I'm glad I could help. Let's hope others will find this helpful as well. |
|
Back to top |
|
|
Decibels Veteran
Joined: 16 Aug 2002 Posts: 1623 Location: U.S.A.
|
Posted: Tue Mar 24, 2020 10:24 pm Post subject: |
|
|
I gather that this: No WUs available for this configuration
doesn't always mean it isn't setup correctly. Cause now the CPU is doing it.
I think they are slammed and also have different files for different types of CPU's and GPU's????
So it just doesn't have any Work Units available for the configuration at this IP Address. Go check else where. _________________ Support bacteria – they’re the only culture some people have.”
– Steven Wright |
|
Back to top |
|
|
Ant P. Watchman
Joined: 18 Apr 2009 Posts: 6920
|
Posted: Tue Mar 24, 2020 10:57 pm Post subject: |
|
|
The project work servers have completely run dry. There's more coming soon, they've finished analysing SARS-NCoV-2 itself and the next project coming in the next few days is to test possible treatments for it, so all of this is making a real difference. In the meantime they're still issuing occasional WUs for the usual causes like cancer.
Head of the project has been posting regular status updates about this on social media: https://nitter.net/drGregBowman |
|
Back to top |
|
|
GD Tux's lil' helper
Joined: 11 May 2005 Posts: 138 Location: Greece
|
Posted: Thu Mar 26, 2020 11:37 pm Post subject: |
|
|
Hello all,
Does anyone know if F@H client will work with Clover? According to clinfo I've got a working (?) OpenCL 1.1 implementation with mesa 20.0.2, llvm-9.0.1... |
|
Back to top |
|
|
axl Veteran
Joined: 11 Oct 2002 Posts: 1144 Location: Romania
|
Posted: Thu Mar 26, 2020 11:47 pm Post subject: |
|
|
opencl is opencl. it's not important which implementation you have. just don't have 2 conflicting ones. |
|
Back to top |
|
|
GD Tux's lil' helper
Joined: 11 May 2005 Posts: 138 Location: Greece
|
Posted: Fri Mar 27, 2020 1:54 am Post subject: |
|
|
axl wrote: | opencl is opencl. it's not important which implementation you have. just don't have 2 conflicting ones. |
Right. Unfortunately I'm getting the following:
Code: | 23:40:57:WU02:FS01:0x22:ERROR:126: Bad platformId size.
23:40:57:WU02:FS01:0x22:Saving result file ../logfile_01.txt
23:40:57:WU02:FS01:0x22:Saving result file science.log
23:40:57:WU02:FS01:0x22:Folding@home Core Shutdown: BAD_WORK_UNIT
23:40:58:WARNING:WU02:FS01:FahCore returned: BAD_WORK_UNIT (114 = 0x72) |
And according to posts in other forums, this might happen with problematic OpenCL implementations. Is OpenCL 1.1 OK or does F@H require compliance >=1.2? |
|
Back to top |
|
|
Ant P. Watchman
Joined: 18 Apr 2009 Posts: 6920
|
Posted: Fri Mar 27, 2020 4:22 am Post subject: |
|
|
GD wrote: | And according to posts in other forums, this might happen with problematic OpenCL implementations. Is OpenCL 1.1 OK or does F@H require compliance >=1.2? |
I've never known Clover to work with anything other than trivial demos. You really need 1.2 to use just about anything in portage with opencl support, and ROCm 3.1.0 is completely fubar so amdgpu-pro-opencl is the only way at present.
If you're using kernel 5.5.* or 5.6_rc*, you also need these bugfixes or all hell will break loose when it actually tries to use OpenCL: https://patchwork.kernel.org/project/dri-devel/list/?series=261285 |
|
Back to top |
|
|
linux_os2 Apprentice
Joined: 29 Aug 2018 Posts: 223 Location: Zedelgem Belgium
|
Posted: Fri Mar 27, 2020 8:54 pm Post subject: |
|
|
SOLVED
I have a 1080 ti.
Following error running foldingathome:
CUDA: Not detected: cuInit() returned 100
OpenCL: Not detected: clGetPlatformIDs() returned -1001
Code: |
*********************** Log Started 2020-03-27T19:59:03Z ***********************
19:59:03:************************* Folding@home Client *************************
19:59:03: Website: https://foldingathome.org/
19:59:03: Copyright: (c) 2009-2018 foldingathome.org
19:59:03: Author: Joseph Coffland <joseph@cauldrondevelopment.com>
19:59:03: Args: -v
19:59:03: Config: /opt/foldingathome/config.xml
19:59:03:******************************** Build ********************************
19:59:03: Version: 7.5.1
19:59:03: Date: May 12 2018
19:59:03: Time: 22:51:07
19:59:03: Repository: Git
19:59:03: Revision: 4705bf53c635f88b8fe85af7675557e15d491ff0
19:59:03: Branch: master
19:59:03: Compiler: GNU 4.4.7 20120313 (Red Hat 4.4.7-18)
19:59:03: Options: -std=gnu++98 -O3 -funroll-loops
19:59:03: Platform: linux2 4.14.0-3-amd64
19:59:03: Bits: 64
19:59:03: Mode: Release
19:59:03:******************************* System ********************************
19:59:03: CPU: Intel(R) Xeon(R) CPU E5-2640 v4 @ 2.40GHz
19:59:03: CPU ID: GenuineIntel Family 6 Model 79 Stepping 1
19:59:03: CPUs: 40
19:59:03: Memory: 126.05GiB
19:59:03:Free Memory: 116.26GiB
19:59:03: Threads: POSIX_THREADS
19:59:03: OS Version: 4.19
19:59:03:Has Battery: false
19:59:03: On Battery: false
19:59:03: UTC Offset: 1
19:59:03: PID: 39217
19:59:03: CWD: /opt/foldingathome
19:59:03: OS: Linux 4.19.57-gentoo x86_64
19:59:03: OS Arch: AMD64
19:59:03: GPUs: 1
19:59:03: GPU 0: Bus:2 Slot:0 Func:0 NVIDIA:8 GP102 [GeForce GTX 1080 Ti] 11380
19:59:03: CUDA: Not detected: cuInit() returned 100
19:59:03: OpenCL: Not detected: clGetPlatformIDs() returned -1001
19:59:03:***********************************************************************
|
DeviceQuery:
Code: |
marco@localhost /opt/cuda/sdk/bin/x86_64/linux/release $ ./deviceQuery
./deviceQuery Starting...
CUDA Device Query (Runtime API) version (CUDART static linking)
Detected 1 CUDA Capable device(s)
Device 0: "GeForce GTX 1080 Ti"
CUDA Driver Version / Runtime Version 10.2 / 10.2
CUDA Capability Major/Minor version number: 6.1
Total amount of global memory: 11176 MBytes (11719409664 bytes)
(28) Multiprocessors, (128) CUDA Cores/MP: 3584 CUDA Cores
GPU Max Clock rate: 1582 MHz (1.58 GHz)
Memory Clock rate: 5505 Mhz
Memory Bus Width: 352-bit
L2 Cache Size: 2883584 bytes
Maximum Texture Dimension Size (x,y,z) 1D=(131072), 2D=(131072, 65536), 3D=(16384, 16384, 16384)
Maximum Layered 1D Texture Size, (num) layers 1D=(32768), 2048 layers
Maximum Layered 2D Texture Size, (num) layers 2D=(32768, 32768), 2048 layers
Total amount of constant memory: 65536 bytes
Total amount of shared memory per block: 49152 bytes
Total number of registers available per block: 65536
Warp size: 32
Maximum number of threads per multiprocessor: 2048
Maximum number of threads per block: 1024
Max dimension size of a thread block (x,y,z): (1024, 1024, 64)
Max dimension size of a grid size (x,y,z): (2147483647, 65535, 65535)
Maximum memory pitch: 2147483647 bytes
Texture alignment: 512 bytes
Concurrent copy and kernel execution: Yes with 2 copy engine(s)
Run time limit on kernels: Yes
Integrated GPU sharing Host Memory: No
Support host page-locked memory mapping: Yes
Alignment requirement for Surfaces: Yes
Device has ECC support: Disabled
Device supports Unified Addressing (UVA): Yes
Device supports Compute Preemption: Yes
Supports Cooperative Kernel Launch: Yes
Supports MultiDevice Co-op Kernel Launch: Yes
Device PCI Domain ID / Bus ID / location ID: 0 / 2 / 0
Compute Mode:
< Default (multiple host threads can use ::cudaSetDevice() with device simultaneously) >
deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 10.2, CUDA Runtime Version = 10.2, NumDevs = 1
Result = PASS
|
glxgears on 4K screen fullscreen
Code: |
marco@localhost /opt/cuda/sdk/bin/x86_64/linux/release $ glxgears -fullscreen
Running synchronized to the vertical refresh. The framerate should be
approximately the same as the monitor refresh rate.
299 frames in 5.0 seconds = 59.753 FPS
300 frames in 5.0 seconds = 59.938 FPS
300 frames in 5.0 seconds = 59.934 FPS
299 frames in 5.0 seconds = 59.759 FPS
300 frames in 5.0 seconds = 59.946 FPS
300 frames in 5.0 seconds = 59.947 FPS
300 frames in 5.0 seconds = 59.949 FPS
marco@localhost /opt/cuda/sdk/bin/x86_64/linux/release $ glxgears -info
Running synchronized to the vertical refresh. The framerate should be
approximately the same as the monitor refresh rate.
GL_RENDERER = GeForce GTX 1080 Ti/PCIe/SSE2
GL_VERSION = 4.6.0 NVIDIA 440.59
GL_VENDOR = NVIDIA Corporation
|
clinfo
Code: |
marco@localhost /opt/cuda/sdk/bin/x86_64/linux/release $ clinfo
Number of platforms 1
Platform Name NVIDIA CUDA
Platform Vendor NVIDIA Corporation
Platform Version OpenCL 1.2 CUDA 10.2.131
Platform Profile FULL_PROFILE
Platform Extensions cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_fp64 cl_khr_byte_addressable_store cl_khr_icd cl_khr_gl_sharing cl_nv_compiler_options cl_nv_device_attribute_query cl_nv_pragma_unroll cl_nv_copy_opts cl_nv_create_buffer cl_khr_int64_base_atomics cl_khr_int64_extended_atomics
Platform Extensions function suffix NV
Platform Name NVIDIA CUDA
Number of devices 1
Device Name GeForce GTX 1080 Ti
Device Vendor NVIDIA Corporation
Device Vendor ID 0x10de
Device Version OpenCL 1.2 CUDA
Driver Version 440.59
Device OpenCL C Version OpenCL C 1.2
Device Type GPU
Device Topology (NV) PCI-E, 02:00.0
Device Profile FULL_PROFILE
Device Available Yes
Compiler Available Yes
Linker Available Yes
Max compute units 28
Max clock frequency 1582MHz
Compute Capability (NV) 6.1
Device Partition (core)
Max number of sub-devices 1
Supported partition types None
Supported affinity domains (n/a)
Max work item dimensions 3
Max work item sizes 1024x1024x64
Max work group size 1024
Preferred work group size multiple 32
Warp size (NV) 32
Preferred / native vector sizes
char 1 / 1
short 1 / 1
int 1 / 1
long 1 / 1
half 0 / 0 (n/a)
float 1 / 1
double 1 / 1 (cl_khr_fp64)
Half-precision Floating-point support (n/a)
Single-precision Floating-point support (core)
Denormals Yes
Infinity and NANs Yes
Round to nearest Yes
Round to zero Yes
Round to infinity Yes
IEEE754-2008 fused multiply-add Yes
Support is emulated in software No
Correctly-rounded divide and sqrt operations Yes
Double-precision Floating-point support (cl_khr_fp64)
Denormals Yes
Infinity and NANs Yes
Round to nearest Yes
Round to zero Yes
Round to infinity Yes
IEEE754-2008 fused multiply-add Yes
Support is emulated in software No
Address bits 64, Little-Endian
Global memory size 11719409664 (10.91GiB)
Error Correction support No
Max memory allocation 2929852416 (2.729GiB)
Unified memory for Host and Device No
Integrated memory (NV) No
Minimum alignment for any data type 128 bytes
Alignment of base address 4096 bits (512 bytes)
Global Memory cache type Read/Write
Global Memory cache size 1376256 (1.312MiB)
Global Memory cache line size 128 bytes
Image support Yes
Max number of samplers per kernel 32
Max size for 1D images from buffer 268435456 pixels
Max 1D or 2D image array size 2048 images
Max 2D image size 16384x32768 pixels
Max 3D image size 16384x16384x16384 pixels
Max number of read image args 256
Max number of write image args 16
Local memory type Local
Local memory size 49152 (48KiB)
Registers per block (NV) 65536
Max number of constant args 9
Max constant buffer size 65536 (64KiB)
Max size of kernel argument 4352 (4.25KiB)
Queue properties
Out-of-order execution Yes
Profiling Yes
Prefer user sync for interop No
Profiling timer resolution 1000ns
Execution capabilities
Run OpenCL kernels Yes
Run native kernels No
Kernel execution timeout (NV) Yes
Concurrent copy and kernel execution (NV) Yes
Number of async copy engines 2
printf() buffer size 1048576 (1024KiB)
Built-in kernels (n/a)
Device Extensions cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_fp64 cl_khr_byte_addressable_store cl_khr_icd cl_khr_gl_sharing cl_nv_compiler_options cl_nv_device_attribute_query cl_nv_pragma_unroll cl_nv_copy_opts cl_nv_create_buffer cl_khr_int64_base_atomics cl_khr_int64_extended_atomics
NULL platform behavior
clGetPlatformInfo(NULL, CL_PLATFORM_NAME, ...) No platform
clGetDeviceIDs(NULL, CL_DEVICE_TYPE_ALL, ...) Success [NV]
clCreateContext(NULL, ...) [default] Success [NV]
clCreateContextFromType(NULL, CL_DEVICE_TYPE_DEFAULT) No platform
clCreateContextFromType(NULL, CL_DEVICE_TYPE_CPU) No devices found in platform
clCreateContextFromType(NULL, CL_DEVICE_TYPE_GPU) No platform
clCreateContextFromType(NULL, CL_DEVICE_TYPE_ACCELERATOR) No devices found in platform
clCreateContextFromType(NULL, CL_DEVICE_TYPE_CUSTOM) Invalid device type for platform
clCreateContextFromType(NULL, CL_DEVICE_TYPE_ALL) No platform
|
OK SOLVED
user foldingathome must be in the video group. |
|
Back to top |
|
|
axl Veteran
Joined: 11 Oct 2002 Posts: 1144 Location: Romania
|
Posted: Sat Mar 28, 2020 6:56 pm Post subject: |
|
|
A bit offtopic. I don't exactly know where to check this. So it's clear mesa's opencl implementation conflicts with nvidia's implementation. And you should have only one, if you use nvidia drivers. But I think the nouveau/mesa one also _might_ work for some things. I'm not sure it will work for FAH, but I seem to recall trying johntheripper with nouveau/opencl from mesa and I seem to remember it worked.
Mesa, according to wikipedia, has only implemented opencl through gallium for amd. 1.0. 1.1 and 1.2. Not sure of the status of nvidia cards, and not sure the status of nouveau. Unfortunately I am using monitors which make use of digital ports, and there was a regression a few years back, and it never got fixed. So... nvidia-drivers. Still, not the worse driver of all time. I seem to recall hating fglrx for many years. |
|
Back to top |
|
|
Massimo B. Veteran
Joined: 09 Feb 2005 Posts: 1773 Location: PB, Germany
|
Posted: Fri Apr 03, 2020 8:13 am Post subject: |
|
|
I'm new and started folding now Can I just join the team by adding the team ID?
I'm folding with <power v='MEDIUM'/> and <idle v='false'/>. Setting idle to true almost never runs a job as my machine is never idle.
What is the preferred way to use less ressources while I use my machine? I thought about renice 10, is there some way to do that via configuration? Would priority=low do something similar?
What difference does OpenCL do? I installed dev-util/intel-ocl-sdk and it seems to be working and detecting all details according to clinfo. I have an Intel-only system with i7-4790, no GPU. _________________ HP ZBook Power 15.6" G8 i7-11800H|HP EliteDesk 800G1 i7-4790|HP Compaq Pro 6300 i7-3770 |
|
Back to top |
|
|
Ant P. Watchman
Joined: 18 Apr 2009 Posts: 6920
|
Posted: Fri Apr 03, 2020 8:07 pm Post subject: |
|
|
Massimo B. wrote: | What is the preferred way to use less ressources while I use my machine? |
It's nice 19 by default. If that's too much then you can use schedtool -B to reduce its priority further, or tell cpufreq to ignore it:
Code: | echo 1 >| /sys/devices/system/cpu/cpufreq/ondemand/ignore_nice_load |
Quote: | What difference does OpenCL do? |
Nothing, unless you add a slot[type='GPU'] to config.xml. Expect much higher temperatures. |
|
Back to top |
|
|
thunderrd n00b
Joined: 20 Aug 2010 Posts: 59
|
Posted: Sat Apr 25, 2020 4:51 am Post subject: |
|
|
Is the cuda toolkit needed by foldingathome, and if it is, why is it not pulled in by emerge? |
|
Back to top |
|
|
|
|
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
|
|