Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Ryzen 7950x lacks feature sse2 #5122

Closed
ahorek opened this issue Mar 2, 2023 · 6 comments · Fixed by #5123 or #5397
Closed

Ryzen 7950x lacks feature sse2 #5122

ahorek opened this issue Mar 2, 2023 · 6 comments · Fixed by #5123 or #5397
Labels
C: Client C: Server P: Minor
Milestone
Client/Manager 8.0.0

Comments

@ahorek
Copy link
Contributor

ahorek commented Mar 2, 2023 •
edited by AenBleidd
Loading

Describe the bug
Einstein@Home tasks are being rejected with an error
[version] CPU [' family 25 model 97 stepping 2 '] lacks feature ' sse2 '

28.02.2023 17:45:16 | | 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
as you can see the list is truncated because the internal buffer isn't large enough, but I'm not sure it's the reason for the original error...

cat /proc/cpuinfo

processor       : 0
vendor_id       : AuthenticAMD
cpu family      : 25
model           : 97
model name      : AMD Ryzen 9 7950X 16-Core Processor
stepping        : 2
microcode       : 0xa601203
cpu MHz         : 4499.798
cache size      : 1024 KB
physical id     : 0
siblings        : 32
core id         : 0
cpu cores       : 16
apicid          : 0
initial apicid  : 0
fpu             : yes
fpu_exception   : yes
cpuid level     : 16
wp              : yes
flags           : 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 umip pku ospke avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg avx512_vpopcntdq rdpid overflow_recov succor smca fsrm flush_l1d
bugs            : sysret_ss_attrs spectre_v1 spectre_v2 spec_store_bypass
bogomips        : 8999.57
TLB size        : 3584 4K pages
clflush size    : 64
cache_alignment : 64
address sizes   : 48 bits physical, 48 bits virtual
power management: ts ttp tm hwpstate cpb eff_freq_ro [13] [14]

I manually tested cpuid feature check and it also says sse2 IS supported.

Steps To Reproduce
add a project Einstein@Home and enable Multi-Directional Gravitational Wave search on O3 (CPU)

Expected behavior
it doesn't accept work due to a missing feature sse2 which is supported on this CPU (no VMs or hacks are involved). Other boinc projects work just fine.

System Information

  • OS: Linux desktop-b49ep3v.easy.local 6.2.0-1275.native #1 SMP Wed Feb 22 18:59:06 PST 2023 x86_64 GNU/Linux
  • BOINC Version: 7.20.5

Additional context
see related code
https://github.com/BOINC/boinc/blob/master/lib/hostinfo.h#L53
https://github.com/BOINC/boinc/blob/master/client/hostinfo_unix.cpp#L530
https://github.com/BOINC/boinc/blob/master/client/hostinfo_unix.cpp#L683

and the original report https://einsteinathome.org/cs/content/ryzen-7950x-lacks-feature-sse2

let me know if there's more info / logs or tests I can provide. Thanks!
@AenBleidd
Copy link
Member

Message [version] CPU [' family 25 model 97 stepping 2 '] lacks feature ' sse2 ' is different from the message that could be sent from the vanilla code: https://github.com/BOINC/boinc/blob/master/sched/plan_class_spec.cpp#L278
That makes me think that E@H runs customized version of this utility, thus it can't be verified and/or fixed from our side

@davidpanderson
Copy link
Contributor

The vanilla code has a 1024-char buffer for CPU features, and I think it's being exceeded here.
I'll change it to std::string.

@davidpanderson davidpanderson mentioned this issue Mar 2, 2023
Merged
@ahorek
Copy link
Contributor Author

ahorek commented Mar 3, 2023

thanks @davidpanderson for looking into it, I've tried your branch, but it didn't help:

2023-03-03 01:27:35.7167 [PID=2377505]   Request: [USER#xxxxx] [HOST#13128375] [IP xxx.xxx.xxx.226] client 7.23.0
2023-03-03 01:27:35.7336 [PID=2377505] [debug]   have_master:0 have_working: 1 have_db: 1
2023-03-03 01:27:35.7336 [PID=2377505] [debug]   using working prefs
2023-03-03 01:27:35.7336 [PID=2377505] [debug]   have db 1; dbmod 1665796207.000000; global mod 1665796207.000000
2023-03-03 01:27:35.7336 [PID=2377505] [debug]   sending db prefs in reply
2023-03-03 01:27:35.7337 [PID=2377505]    [send] effective_ncpus 32 max_jobs_on_host_cpu 999999 max_jobs_on_host 999999
2023-03-03 01:27:35.7337 [PID=2377505]    [send] effective_ngpus 0 max_jobs_on_host_gpu 999999
2023-03-03 01:27:35.7337 [PID=2377505]    [send] Not using matchmaker scheduling; Not using EDF sim
2023-03-03 01:27:35.7337 [PID=2377505]    [send] CPU: req 1382400.00 sec, 32.00 instances; est delay 0.00
2023-03-03 01:27:35.7337 [PID=2377505]    [send] work_req_seconds: 1382400.00 secs
2023-03-03 01:27:35.7337 [PID=2377505]    [send] available disk 9.99 GB, work_buf_min 17280
2023-03-03 01:27:35.7338 [PID=2377505]    [send] active_frac 1.000000 on_frac 0.999772 DCF 1.000000
2023-03-03 01:27:35.7345 [PID=2377505]    [mixed] sending non-locality work first (0.9112)
2023-03-03 01:27:35.7731 [PID=2377505]    [send] [HOST#13128375] will accept beta work.  Scanning for beta work.
2023-03-03 01:27:35.7824 [PID=2377505]    [mixed] sending locality work second
2023-03-03 01:27:35.7875 [PID=2377505]    [send] send_old_work() no feasible result older than 336.0 hours
2023-03-03 01:27:35.7933 [PID=2377505]    [send] send_old_work() no feasible result younger than 216.4 hours and older than 168.0 hours
2023-03-03 01:27:35.8139 [PID=2377505]    [version] Checking plan class 'GW-SSE2'
2023-03-03 01:27:35.8149 [PID=2377505]    [version] reading plan classes from file '/BOINC/projects/EinsteinAtHome/plan_class_spec.xml'
2023-03-03 01:27:35.8149 [PID=2377505]    [version] CPU [' family 25 model 97 stepping 2 '] lacks feature ' sse2 '
2023-03-03 01:27:35.8150 [PID=2377505]    [version] no app version available: APP#59 (einstein_O3MD1) PLATFORM#7 (x86_64-pc-linux-gnu) min_version 0
2023-03-03 01:27:35.8150 [PID=2377505]    [version] no app version available: APP#59 (einstein_O3MD1) PLATFORM#1 (i686-pc-linux-gnu) min_version 0
2023-03-03 01:27:35.8173 [PID=2377505] [debug]   [HOST#13128375] MSG(high) No work sent
2023-03-03 01:27:35.8173 [PID=2377505] [debug]   [HOST#13128375] MSG(high) see scheduler log messages on https://einsteinathome.org/host/13128375/log
2023-03-03 01:27:35.8173 [PID=2377505]    Sending reply to [HOST#13128375]: 0 results, delay req 60.00
2023-03-03 01:27:35.8173 [PID=2377505]    Scheduler ran 0.108 seconds

03.03.2023 2:26:03 |  | Starting BOINC client version 7.23.0 for x86_64-pc-linux-gnu
03.03.2023 2:26:03 |  | This a development version of BOINC and may not function properly
03.03.2023 2:26:03 |  | log flags: file_xfer, sched_ops, task
03.03.2023 2:26:03 |  | Libraries: libcurl/7.87.0 OpenSSL/3.0.8 zlib/1.2.13 zstd/1.5.4 nghttp2/1.52.0
03.03.2023 2:26:03 |  | Data directory: /home/ahorek/boi/boinc/packages/generic/sea/BOINC
03.03.2023 2:26:03 |  | No usable GPUs found
03.03.2023 2:26:03 |  | libc:  version 2.37
03.03.2023 2:26:03 |  | Host name: desktop-b49ep3v.easy.local
03.03.2023 2:26:03 |  | Processor: 32 AuthenticAMD AMD Ryzen 9 7950X 16-Core Processor [Family 25 Model 97 Stepping 2]
03.03.2023 2:26:03 |  | 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
03.03.2023 2:26:03 |  | OS: Linux Clear Linux OS: Clear Linux OS [6.2.0-1275.native|libc 2.37]
03.03.2023 2:26:03 |  | Memory: 30.50 GB physical, 64.00 MB virtual
03.03.2023 2:26:03 |  | Disk: 457.24 GB total, 404.86 GB free
03.03.2023 2:26:03 |  | Local time is UTC +1 hours
03.03.2023 2:26:03 |  | Config: GUI RPC allowed from any host
03.03.2023 2:26:03 |  | No general preferences found - using defaults
03.03.2023 2:26:03 |  | Preferences:
03.03.2023 2:26:03 |  | -  When computer is in use
03.03.2023 2:26:03 |  | -     'In use' means mouse/keyboard input in last 3.0 minutes
03.03.2023 2:26:03 |  | -     don't use GPU
03.03.2023 2:26:03 |  | -     Use at most 100% of the CPU time
03.03.2023 2:26:03 |  | -     suspend if non-BOINC CPU load exceeds 25%
03.03.2023 2:26:03 |  | -     max memory usage: 15.25 GB
03.03.2023 2:26:03 |  | -  When computer is not in use
03.03.2023 2:26:03 |  | -     max CPUs used: 32
03.03.2023 2:26:03 |  | -     Use at most 100% of the CPU time
03.03.2023 2:26:03 |  | -     suspend if non-BOINC CPU load exceeds 50%
03.03.2023 2:26:03 |  | -     max memory usage: 27.45 GB
03.03.2023 2:26:03 |  | -     Suspend if no input in last 60.000000 minutes
03.03.2023 2:26:03 |  | -  Suspend if running on batteries
03.03.2023 2:26:03 |  | -  Store at least 0.10 days of work
03.03.2023 2:26:03 |  | -  Store up to an additional 0.50 days of work
03.03.2023 2:26:03 |  | -  max disk usage: 404.76 GB
03.03.2023 2:26:03 |  | -  (to change preferences, visit a project web site or select Preferences in the Manager)
03.03.2023 2:26:03 |  | Setting up project and slot directories
03.03.2023 2:26:03 |  | Checking active tasks
03.03.2023 2:26:03 |  | Setting up GUI RPC socket
03.03.2023 2:26:03 |  | Checking presence of 0 project files
03.03.2023 2:26:03 |  | This computer is not attached to any projects
03.03.2023 2:26:09 |  | Fetching configuration file from https://einstein.phys.uwm.edu/get_project_config.php
03.03.2023 2:26:23 | Einstein@Home | Fetching scheduler list
03.03.2023 2:26:25 | Einstein@Home | Master file download succeeded
03.03.2023 2:26:30 | Einstein@Home | Sending scheduler request: Project initialization.
03.03.2023 2:26:30 | Einstein@Home | Requesting new tasks for CPU
03.03.2023 2:26:34 | Einstein@Home | Scheduler request completed: got 0 new tasks
...
03.03.2023 2:26:36 | Einstein@Home | Started download of einstein_icon.png
03.03.2023 2:26:36 | Einstein@Home | Started download of Android.jpg
03.03.2023 2:26:36 | Einstein@Home | Started download of Arecibo_full.jpg
03.03.2023 2:26:36 | Einstein@Home | Started download of Arecibo_platform.jpg
03.03.2023 2:26:36 | Einstein@Home | Started download of Fermi_grsky.jpg
03.03.2023 2:26:36 | Einstein@Home | Started download of Fermi_satellite.jpg
03.03.2023 2:26:36 | Einstein@Home | Started download of GW_BBH1.jpg
03.03.2023 2:26:36 | Einstein@Home | Started download of GW_BBH2.jpg
03.03.2023 2:26:36 | Einstein@Home | Started download of LIGO_Hanford.jpg
03.03.2023 2:26:36 | Einstein@Home | Started download of LIGO_Livingston.jpg
03.03.2023 2:26:36 | Einstein@Home | Started download of LIGO_laser.jpg
03.03.2023 2:26:36 | Einstein@Home | Started download of LIGO_optics.jpg
03.03.2023 2:26:36 | Einstein@Home | Started download of LIGO_schematic.jpg
03.03.2023 2:26:36 | Einstein@Home | Started download of LIGO_seisisol.jpg
03.03.2023 2:26:36 | Einstein@Home | Started download of LIGO_vacuum.jpg
03.03.2023 2:26:36 | Einstein@Home | Started download of Parkes_full.jpg
03.03.2023 2:26:36 | Einstein@Home | Started download of Pulsars_J2007.jpg
03.03.2023 2:26:36 | Einstein@Home | Started download of Pulsars_Manhattan.jpg
03.03.2023 2:26:36 | Einstein@Home | Started download of Pulsars_crab_opt.jpg
03.03.2023 2:26:36 | Einstein@Home | Started download of Pulsars_crab_xr.jpg
03.03.2023 2:26:36 | Einstein@Home | Started download of Pulsars_schem1.jpg
03.03.2023 2:26:36 | Einstein@Home | Started download of Pulsars_schem2.jpg
03.03.2023 2:26:36 | Einstein@Home | Started download of Pulsars_vela.jpg
03.03.2023 2:26:37 | Einstein@Home | Finished download of einstein_icon.png (0 bytes)
03.03.2023 2:26:37 | Einstein@Home | Finished download of Android.jpg (0 bytes)
03.03.2023 2:26:37 | Einstein@Home | Finished download of Arecibo_full.jpg (0 bytes)
03.03.2023 2:26:37 | Einstein@Home | Finished download of Arecibo_platform.jpg (0 bytes)
03.03.2023 2:26:37 | Einstein@Home | Finished download of Fermi_grsky.jpg (0 bytes)
03.03.2023 2:26:37 | Einstein@Home | Finished download of Fermi_satellite.jpg (0 bytes)
03.03.2023 2:26:37 | Einstein@Home | Finished download of GW_BBH1.jpg (0 bytes)
03.03.2023 2:26:37 | Einstein@Home | Finished download of GW_BBH2.jpg (0 bytes)
03.03.2023 2:26:37 | Einstein@Home | Finished download of LIGO_Hanford.jpg (0 bytes)
03.03.2023 2:26:37 | Einstein@Home | Finished download of LIGO_Livingston.jpg (0 bytes)
03.03.2023 2:26:37 | Einstein@Home | Finished download of LIGO_laser.jpg (0 bytes)
03.03.2023 2:26:37 | Einstein@Home | Finished download of LIGO_optics.jpg (0 bytes)
03.03.2023 2:26:37 | Einstein@Home | Finished download of LIGO_schematic.jpg (0 bytes)
03.03.2023 2:26:37 | Einstein@Home | Finished download of LIGO_seisisol.jpg (0 bytes)
03.03.2023 2:26:37 | Einstein@Home | Finished download of LIGO_vacuum.jpg (0 bytes)
03.03.2023 2:26:37 | Einstein@Home | Finished download of Parkes_full.jpg (0 bytes)
03.03.2023 2:26:37 | Einstein@Home | Finished download of Pulsars_J2007.jpg (0 bytes)
03.03.2023 2:26:38 | Einstein@Home | Finished download of Pulsars_Manhattan.jpg (0 bytes)
03.03.2023 2:26:38 | Einstein@Home | Finished download of Pulsars_crab_opt.jpg (0 bytes)
03.03.2023 2:26:38 | Einstein@Home | Finished download of Pulsars_crab_xr.jpg (0 bytes)
03.03.2023 2:26:38 | Einstein@Home | Finished download of Pulsars_schem1.jpg (0 bytes)
03.03.2023 2:26:38 | Einstein@Home | Finished download of Pulsars_schem2.jpg (0 bytes)
03.03.2023 2:26:38 | Einstein@Home | Finished download of Pulsars_vela.jpg (0 bytes)
03.03.2023 2:27:34 | Einstein@Home | Sending scheduler request: To fetch work.
03.03.2023 2:27:34 | Einstein@Home | Requesting new tasks for CPU
03.03.2023 2:27:36 | Einstein@Home | Scheduler request completed: got 0 new tasks
03.03.2023 2:27:36 | Einstein@Home | No work sent
03.03.2023 2:27:36 | Einstein@Home | see scheduler log messages on https://einsteinathome.org/host/13128375/log

@davidpanderson
Copy link
Contributor

It will work only when Einstein@home makes this change in their server code.

@AenBleidd AenBleidd added this to the Client/Manager 7.22.1 milestone Mar 4, 2023
@bema-aei
Copy link
Contributor

bema-aei commented Oct 16, 2023 •
edited
Loading

In the current version of the code (master branch), p_features size of the host record is still fixed 1024. That doesn't look right (and probably won't help):

    char p_features[1024];

https://github.com/BOINC/boinc/blob/master/db/boinc_db_types.h#L369

@AenBleidd
Copy link
Member

@davidpanderson, I believe this is a valid point. Could you please fix that as well? Thank you in advance

@AenBleidd AenBleidd reopened this Oct 16, 2023
@AenBleidd AenBleidd linked a pull request Oct 16, 2023 that will close this issue
complete #5123: client and server: increase p_features size from 1024 to 8192 #5397
Merged
@AenBleidd AenBleidd moved this from Backlog to Review in progress in Client/Manager Oct 16, 2023
@github-project-automation github-project-automation bot moved this from Review in progress to Done in Client/Manager Oct 16, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
C: Client C: Server P: Minor
Projects
Client/Manager
Archived in project
Milestone
Client/Manager 8.0.0
4 participants
@davidpanderson @AenBleidd @ahorek @bema-aei