| 1 |
On 08/15 02:32, Corentin “Nado” Pazdera wrote: |
| 2 |
> August 15, 2018 2:59 PM, tuxic@××××××.de wrote: |
| 3 |
> |
| 4 |
> > Yes I did reboot the sustem. In my initial mail I mentioned a tool |
| 5 |
> > called CUDA-Z and Blender, which both reports a missing CUDA device. |
| 6 |
> |
| 7 |
> Ok, so you do not have a specific error which might have been thrown by the module? |
| 8 |
> Other ideas, check dev-util/nvidia-cuda-toolkit version and double check nvidia/nvidia_uvm with modinfo to ensure they are installed and loaded correctly with the right version? |
| 9 |
> Could you also run /opt/cuda/extras/demo_suite/deviceQuery (from nvidia-cuda-toolkit) and show its output? |
| 10 |
> |
| 11 |
> My installation works, so at least we know their version is not completely broken... |
| 12 |
> Driver version: 396.51 |
| 13 |
> Cuda version: 9.2.88 |
| 14 |
> |
| 15 |
> -- |
| 16 |
> Corentin “Nado” Pazdera |
| 17 |
> |
| 18 |
|
| 19 |
I compiled the new version of the driver again and rebooted the |
| 20 |
system. |
| 21 |
|
| 22 |
# dmesg | grep -i nvidia: |
| 23 |
|
| 24 |
[ 11.375631] nvidia_drm: module license 'MIT' taints kernel. |
| 25 |
[ 12.313260] nvidia-nvlink: Nvlink Core is being initialized, major device number 246 |
| 26 |
[ 12.313586] nvidia 0000:07:00.0: vgaarb: changed VGA decodes: olddecodes=io+mem,decodes=none:owns=io+mem |
| 27 |
[ 12.313691] nvidia 0000:02:00.0: enabling device (0000 -> 0003) |
| 28 |
[ 12.313737] nvidia 0000:02:00.0: vgaarb: changed VGA decodes: olddecodes=io+mem,decodes=none:owns=none |
| 29 |
[ 12.313826] NVRM: loading NVIDIA UNIX x86_64 Kernel Module 396.51 Tue Jul 31 10:43:06 PDT 2018 (using threaded interrupts) |
| 30 |
[ 12.491106] input: HDA NVidia HDMI as /devices/pci0000:00/0000:00:0b.0/0000:02:00.1/sound/card2/input9 |
| 31 |
[ 12.492291] input: HDA NVidia HDMI as /devices/pci0000:00/0000:00:0b.0/0000:02:00.1/sound/card2/input10 |
| 32 |
[ 12.493772] input: HDA NVidia HDMI as /devices/pci0000:00/0000:00:02.0/0000:07:00.1/sound/card1/input11 |
| 33 |
[ 12.494605] input: HDA NVidia HDMI as /devices/pci0000:00/0000:00:02.0/0000:07:00.1/sound/card1/input12 |
| 34 |
[ 13.963644] caller _nv001112rm+0xe3/0x1d0 [nvidia] mapping multiple BARs |
| 35 |
[ 34.236553] caller _nv001112rm+0xe3/0x1d0 [nvidia] mapping multiple BARs |
| 36 |
[ 34.516495] nvidia-modeset: Loading NVIDIA Kernel Mode Setting Driver for UNIX platforms 396.51 Tue Jul 31 14:52:09 PDT 2018 |
| 37 |
|
| 38 |
# modprobe -a nvidia-uvm |
| 39 |
|
| 40 |
# dmesg | grep uvm |
| 41 |
|
| 42 |
[ 209.441956] nvidia-uvm: Loaded the UVM driver in 8 mode, major device number 245 |
| 43 |
|
| 44 |
|
| 45 |
# /opt/cuda/extras/demo_suite/deviceQuery |
| 46 |
/opt/cuda/extras/demo_suite/deviceQuery Starting... |
| 47 |
|
| 48 |
CUDA Device Query (Runtime API) version (CUDART static linking) |
| 49 |
|
| 50 |
cudaGetDeviceCount returned 30 |
| 51 |
-> unknown error |
| 52 |
Result = FAIL |
| 53 |
[1] 5086 exit 1 /opt/cuda/extras/demo_suite/deviceQuery |
| 54 |
|
| 55 |
CUDA-Z shows also "no CUDA device" |
| 56 |
|
| 57 |
# modinfo nvidia-uvm |
| 58 |
filename: /lib/modules/4.18.0-RT/video/nvidia-uvm.ko |
| 59 |
supported: external |
| 60 |
license: MIT |
| 61 |
depends: nvidia |
| 62 |
name: nvidia_uvm |
| 63 |
vermagic: 4.18.0-RT SMP preempt mod_unload |
| 64 |
parm: uvm_perf_prefetch_enable:uint |
| 65 |
parm: uvm_perf_prefetch_threshold:uint |
| 66 |
parm: uvm_perf_prefetch_min_faults:uint |
| 67 |
parm: uvm_perf_thrashing_enable:uint |
| 68 |
parm: uvm_perf_thrashing_threshold:uint |
| 69 |
parm: uvm_perf_thrashing_pin_threshold:uint |
| 70 |
parm: uvm_perf_thrashing_lapse_usec:uint |
| 71 |
parm: uvm_perf_thrashing_nap_usec:uint |
| 72 |
parm: uvm_perf_thrashing_epoch_msec:uint |
| 73 |
parm: uvm_perf_thrashing_max_resets:uint |
| 74 |
parm: uvm_perf_thrashing_pin_msec:uint |
| 75 |
parm: uvm_perf_map_remote_on_native_atomics_fault:uint |
| 76 |
parm: uvm_hmm:Enable (1) or disable (0) HMM mode. Default: 0. Ignored if CONFIG_HMM is not set, or if NEXT settings conflict with HMM. (int) |
| 77 |
parm: uvm_global_oversubscription:Enable (1) or disable (0) global oversubscription support. (int) |
| 78 |
parm: uvm_leak_checker:Enable uvm memory leak checking. 0 = disabled, 1 = count total bytes allocated and freed, 2 = per-allocation origin tracking. (int) |
| 79 |
parm: uvm_force_prefetch_fault_support:uint |
| 80 |
parm: uvm_debug_enable_push_desc:Enable push description tracking (int) |
| 81 |
parm: uvm_page_table_location:Set the location for UVM-allocated page tables. Choices are: vid, sys. (charp) |
| 82 |
parm: uvm_perf_access_counter_mimc_migration_enable:Whether MIMC access counters will trigger migrations.Valid values: <= -1 (default policy), 0 (off), >= 1 (on) (int) |
| 83 |
parm: uvm_perf_access_counter_momc_migration_enable:Whether MOMC access counters will trigger migrations.Valid values: <= -1 (default policy), 0 (off), >= 1 (on) (int) |
| 84 |
parm: uvm_perf_access_counter_batch_count:uint |
| 85 |
parm: uvm_perf_access_counter_granularity:Size of the physical memory region tracked by each counter. Valid values asof Volta: 64k, 2m, 16m, 16g (charp) |
| 86 |
parm: uvm_perf_access_counter_threshold:Number of remote accesses on a region required to trigger a notification.Valid values: [1, 65535] (uint) |
| 87 |
parm: uvm_perf_reenable_prefetch_faults_lapse_msec:uint |
| 88 |
parm: uvm_perf_fault_batch_count:uint |
| 89 |
parm: uvm_perf_fault_replay_policy:uint |
| 90 |
parm: uvm_perf_fault_replay_update_put_ratio:uint |
| 91 |
parm: uvm_perf_fault_max_batches_per_service:uint |
| 92 |
parm: uvm_perf_fault_max_throttle_per_service:uint |
| 93 |
parm: uvm_perf_fault_coalesce:uint |
| 94 |
parm: uvm_fault_force_sysmem:Force (1) using sysmem storage for pages that faulted. Default: 0. (int) |
| 95 |
parm: uvm_perf_map_remote_on_eviction:int |
| 96 |
parm: uvm_channel_num_gpfifo_entries:uint |
| 97 |
parm: uvm_channel_gpfifo_loc:charp |
| 98 |
parm: uvm_channel_gpput_loc:charp |
| 99 |
parm: uvm_channel_pushbuffer_loc:charp |
| 100 |
parm: uvm_enable_debug_procfs:Enable debug procfs entries in /proc/driver/nvidia-uvm (int) |
| 101 |
parm: uvm8_ats_mode:Override the default ATS (Address Translation Services) UVM mode by disabling (0) or enabling (1) (int) |
| 102 |
parm: uvm_driver_mode:Set the uvm kernel driver mode. Choices include: 8 (charp) |
| 103 |
parm: uvm_debug_prints:Enable uvm debug prints. (int) |
| 104 |
parm: uvm_enable_builtin_tests:Enable the UVM built-in tests. (This is a security risk) (int) |
| 105 |
|
| 106 |
|
| 107 |
# ls -l /lib/modules/4.18.0-RT/video/nvidia-uvm.ko |
| 108 |
-rw-r--r-- 1 root root 1405808 Aug 15 16:49 /lib/modules/4.18.0-RT/video/nvidia-uvm.ko |
| 109 |
(just installed minytes before) |
| 110 |
|
| 111 |
# uname -a |
| 112 |
Linux solfire 4.18.0-RT #1 SMP PREEMPT Mon Aug 13 05:15:26 CEST 2018 x86_64 AMD Phenom(tm) II X6 1090T Processor AuthenticAMD GNU/Linux |
| 113 |
(the kernel version matches) |
| 114 |
|
| 115 |
# eix nvidia-cuda-toolkit |
| 116 |
|
| 117 |
[I] dev-util/nvidia-cuda-toolkit |
| 118 |
Available versions: [M](~)6.5.14(0/6.5.14) [M](~)6.5.19-r1(0/6.5.19) [M](~)7.5.18-r2(0/7.5.18) [M](~)8.0.44(0/8.0.44) [M](~)8.0.61(0/8.0.61) (~)9.0.176(0/9.0.176) (~)9.1.85(0/9.1.85) (~)9.2.88(0/9.2.88) {debugger doc eclipse profiler} |
| 119 |
Installed versions: 9.2.88(0/9.2.88)(06:31:32 PM 08/14/2018)(-debugger -doc -eclipse -profiler) |
| 120 |
Homepage: https://developer.nvidia.com/cuda-zone |
| 121 |
Description: NVIDIA CUDA Toolkit (compiler and friends) |
| 122 |
|
| 123 |
|
| 124 |
|
| 125 |
It becomes even more weird... |
Archives