Message boards : Getting started : State: Waiting for memory
Author | Message |
---|---|
jari Send message Joined: 23 Jun 22 Posts: 2 Credit: 4,716,974 RAC: 4,477 |
The GPU task does not seem to run well and stopped with state: "Waiting for memory". Also it seems the GPU is crashing somehow as seen in the below nvidia-smi output: Thu Jun 23 20:04:03 2022 +-----------------------------------------------------------------------------+ | NVIDIA-SMI 510.73.05 Driver Version: 510.73.05 CUDA Version: 11.6 | |-------------------------------+----------------------+----------------------+ | GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC | | Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. | | | | MIG M. | |===============================+======================+======================| | 0 NVIDIA GeForce ... Off | 00000000:02:00.0 Off | N/A | | N/A 60C P8 N/A / N/A | 4MiB / 4096MiB | 0% Default | | | | N/A | +-------------------------------+----------------------+----------------------+ +-----------------------------------------------------------------------------+ | Processes: | | GPU GI CI PID Type Process name GPU Memory | | ID ID Usage | |=============================================================================| | 0 N/A N/A 11294 G /usr/lib/xorg/Xorg 4MiB | +-----------------------------------------------------------------------------+ Thu Jun 23 20:04:08 2022 +-----------------------------------------------------------------------------+ | NVIDIA-SMI 510.73.05 Driver Version: 510.73.05 CUDA Version: 11.6 | |-------------------------------+----------------------+----------------------+ | GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC | | Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. | | | | MIG M. | |===============================+======================+======================| | 0 NVIDIA GeForce ... Off | 00000000:02:00.0 Off | N/A | | N/A 62C P0 N/A / N/A | 35MiB / 4096MiB | 0% Default | | | | N/A | +-------------------------------+----------------------+----------------------+ +-----------------------------------------------------------------------------+ | Processes: | | GPU GI CI PID Type Process name GPU Memory | | ID ID Usage | |=============================================================================| | 0 N/A N/A 11294 G /usr/lib/xorg/Xorg 4MiB | | 0 N/A N/A 208003 C ...le/amicable_OpenCL_v_3_02 29MiB | +-----------------------------------------------------------------------------+ GPU 00000000:02:00.0: Detected Critical Xid Error The CPU work units seem to be running normally, but GPU work load is not passing. System info: CPU: Info: Quad Core model: Intel Core i7-8565U bits: 64 type: MT MCP L2 cache: 8 MiB Speed: 2200 MHz min/max: 400/4600 MHz Core speeds (MHz): 1: 2200 2: 2200 3: 2200 4: 2200 5: 2200 6: 2200 7: 2200 8: 2200 Graphics: Device-1: Intel WhiskeyLake-U GT2 [UHD Graphics 620] driver: i915 v: kernel Device-2: NVIDIA GP108M [GeForce MX250] driver: nvidia v: 515.48.07 Display: x11 server: X.Org 1.20.11 driver: loaded: modesetting,nvidia unloaded: fbdev,nouveau,vesa resolution: 1: 1920x1080~60Hz 2: 1920x1080~60Hz OpenGL: renderer: Mesa Intel UHD Graphics 620 (WHL GT2) v: 4.6 Mesa 20.3.5 I think the GPU is possibly overheating and kicked out form the bus, but not sure why this happens. |
jari Send message Joined: 23 Jun 22 Posts: 2 Credit: 4,716,974 RAC: 4,477 |
Sun Jun 26 10:54:50 2022 | Amicable Numbers | Computation for task amicable_10_21_4584_1655960702.514059_799_1 finished Sun Jun 26 10:54:50 2022 | Amicable Numbers | Starting task amicable_10_21_4584_1655960702.514059_800_0 Sun Jun 26 10:54:52 2022 | Amicable Numbers | Started upload of amicable_10_21_4584_1655960702.514059_799_1_r1275436014_0 Sun Jun 26 10:54:54 2022 | Amicable Numbers | Finished upload of amicable_10_21_4584_1655960702.514059_799_1_r1275436014_0 Sun Jun 26 10:56:31 2022 | Amicable Numbers | Aborting task amicable_10_21_4584_1655960702.514059_800_0: working set size > client RAM limit: 7866.50MB > 7866.34MB Sun Jun 26 10:56:32 2022 | Amicable Numbers | Computation for task amicable_10_21_4584_1655960702.514059_800_0 finished Sun Jun 26 10:56:32 2022 | Amicable Numbers | Starting task amicable_10_21_16704_1655526602.848309_766_2 Sun Jun 26 10:56:34 2022 | Amicable Numbers | Started upload of amicable_10_21_4584_1655960702.514059_800_0_r1962825675_0 Sun Jun 26 10:56:36 2022 | Amicable Numbers | Finished upload of amicable_10_21_4584_1655960702.514059_800_0_r1962825675_0 Sun Jun 26 10:58:22 2022 | Amicable Numbers | Aborting task amicable_10_21_16704_1655526602.848309_766_2: working set size > client RAM limit: 7973.66MB > 7866.34MB Sun Jun 26 10:58:23 2022 | Amicable Numbers | Computation for task amicable_10_21_16704_1655526602.848309_766_2 finished Sun Jun 26 10:58:23 2022 | Amicable Numbers | Starting task amicable_10_21_4584_1655960702.514059_868_1 The GPU tasks are aborting like this... It looks like this is caused by low memory that I should increase the Options->Memory to 90% (from earlier 50%). Then it worked for about 3-5 minutes until the GPU overheated and system was forced to shutdown. So it looks not possible to get the GPU tasks working in this machine. |
Sascha Harper-Noorthoorn van der Kruyff Send message Joined: 16 Jun 22 Posts: 8 Credit: 1,045,938 RAC: 0 |
Jari? Mine says all the time (well ok, not "all" the time, but "a lot") 'waiting for memory', and then you just go do something else, and it starts to work again. I would not worry about it, if I were you (just suppose that was a bad thing... then how com I 'discovered' already 6 amicable pairs?) |
Sascha Harper-Noorthoorn van der Kruyff Send message Joined: 16 Jun 22 Posts: 8 Credit: 1,045,938 RAC: 0 |
Jari... I find that there are always two or three different Amicable Numbers-projects going on in my Boinc manager, what helps me, and maybe you too, is to check up on their progress every so often, and sometimes switch to a project that 'moves forward' instead of 'sticking with the project' that you were busy with but "got stuck" cuz it 'waits for memory' or 'the CPU is busy' or whatever it says |
Message boards : Getting started : State: Waiting for memory
©2024 Sergei Chernykh