no cpu work for linux?
8 Jul 2023 17:57:06 UTC
Topic 229773
(moderation:
Must be out.
snip of log for this machine-
-----
[HOST#13137814] MSG(high) No work sent [HOST#13137814] MSG(high) No work is available for Binary Radio Pulsar Search (Arecibo,GBT) [HOST#13137814] MSG(high) No work is available for Binary Radio Pulsar Search (Arecibo,GBT,long) [HOST#13137814] MSG(high) No work is available for All-Sky Gravitational Wave search on O3 [HOST#13137814] MSG(high) No work is available for Multi-Directional Gravitational Wave search on O3 (CPU) [HOST#13137814] MSG(high) see scheduler log messages on https://einsteinathome.org/host/13137814/log [PID=4147543] Sending reply to [HOST#13137814]: 0 results, delay req 60.00 2023-07-08 17:09:24.4178 [PID=4147543] Scheduler ran 0.059 seconds
-----
I can't always tell how much or how little work is really available from the server status pages. I guess they are pretty up to date.
I put Deb. 12 on this machine trying to get an amdgpu 7900XT up on the newer kernel and firmware - such as is available but still was having trouble with that gpu. So I disabled gpu work for now. Set boinc for use no coprocessors or gpu - for now.
I was running the BRP7 cpu tasks but they are out.
more on system:
OS - Linux Debian Debian GNU/Linux 12 (bookworm) [6.1.0-10-amd64|libc 2.36]
CPU - AuthenticAMD AMD Ryzen 7 7800X3D 8-Core Processor [Family 25 Model 97 Stepping 2]
Still running GRP #5 on one intel box here on Deb. 12 with no gpu. ---------------------
------------
Application Gamma-ray pulsar search #5 1.08 (FGRPSSE)
Name LATeah1090F_1032.0_2040660_0.0
------------
jobs running on another box. Deb 12 on intel cpu.
The server status page show 'tasks to send' @ 4288 for FGRP5 -- 11 am local GMT -7 here -- but over on the left 'project daemons' side of the page it shows 'FGRP5 work generator on einstein 4 NOT RUNNING' but I just got work on this box @ 1:34 am local this morning.
I dumped my app_config file for the box in question to make sure it wasn't being too selective or getting in the way.
Yes it is. No work for BRP4,
)
Yes it is. No work for BRP4, BRP4G and O3MD1 at the moment. See server status.
Tnx, Scrooge. I thought
)
Tnx, Scrooge. I thought that's what I was seeing. After 3x checking my setup and configs, I figured I'd ask another one to confirm. I've overlooked so many silly things in the past.
I'm going to have to wait a bit for the firmware to catch up before I get this gpu running on this setup anyhow.
Cheers!
Seems like there is still a
)
Seems like there is still a little work left for linux CPU - hsgamma_FGRP5. Status page shows ~4613 w/u available this morning 7/9. One of my intel boxes has been receiving them, but this machine will not fetch any..
Also over on the left side it shows 'einstein 4 - FGRP5 work generator' as Not Running. But maybe the 4613 are still in the queue for outbound. Otherwise I don't know how this other box would still be receiving them. ?
Can't get the AMD - Ryzen box to fetch anything.
Mike wrote: Seems like there
)
Setup a zero resource share project to get gpu tasks in the mean time, lots of projects have gpu tasks for Linux
Hi Mike,the server log
)
Hi Mike,
the server log linked to also contains this:
I think this is something that has been explained here. For getting FGRP5 work I have successfully tried the suggested workaround, which is invoking the stock science app through the "anonymous platform", i.e. an app_info.xml.
walton748 wrote: Hi
)
this is the way.
_________________________________________________________________________
Wow, thanks for that
)
Wow, thanks for that Walton748. I followed that thread and was looking for that missing sse message, but I didn't read that server log carefully enough and missed it. Interestingly, my boinc event.log for startup and running showed no problem or error - it reported the feature as available. Go figure ? Maybe it's the 7.20.5 version of client. I'll re-read the full thread for the work-around. There wasn't enough buffer space somewhere to store the full 'feature' list string for the newer cpus I think I recall. Maybe just because the feature reports as being present doesn't mean it's necessarily available. ?
Here's a snip of that event.log - yeah the feature list is getting pretty wild.
-----------
Tue 11 Jul 2023 07:49:44 PM PDT | | Processor: 8 AuthenticAMD AMD Ryzen 7 7800X3D 8-Core Processor [Family 25 Model 97 Stepping 2]
Tue 11 Jul 2023 07:49:44 PM PDT | | Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good amd_lbr_v2 nopl nonstop_tsc cpuid extd_apicid aperfmperf rapl pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_llc mwaitx cpb cat_l3 cdp_l3 hw_pstate ssbd mba perfmon_v2 ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 erms invpcid cqm rdt_a avx512f avx512dq rdseed adx smap avx512ifma clflushopt clwb avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local avx512_bf16 clzero irperf xsaveerptr rdpru wbnoinvd cppc arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif x2avic v_spec_ctrl avx512vbmi u
-----------
that's pretty extreme. maybe I blew the c&p. Thanks Walton and Ian & Steve and everybody.