mersenneforum.org  

Go Back   mersenneforum.org > Great Internet Mersenne Prime Search > Software

Reply
 
Thread Tools
Old 2022-08-31, 20:19   #672
joblack
 
joblack's Avatar
 
Oct 2008
n00bville

13408 Posts
Default

Quote:
Originally Posted by SuikaPredator View Post
The Prime95.exe just crashes without any hint box, is there something wrong with my settings?
Code:
parameters: ECM2=N/A,1,2,1277,-1,2400000000,240000000000,1
...

Yes, I had the same problem. Check out if there is a (cumulative) Windows update. After installing the Windows update and a reboot, it worked again for me.

Last fiddled with by S485122 on 2022-08-31 at 21:12 Reason: no need to quote the rest
joblack is offline   Reply With Quote
Old 2022-09-01, 17:22   #673
SuikaPredator
 
Aug 2022
China

25 Posts
Default

Quote:
Originally Posted by Prime95 View Post
What CPU are you using?

As an aside, ECM on M1277 requires a lot of homework. Multithreading will do no good. You should run 6 workers each using one core. Second, prime95 is the wrong tool for ECM stage 2. There is a complicated method to use GMP-ECM for stage 2. Version 30.9 hopes to rectify the stage 2 problem but it isn't ready yet.
CPU=Intel(R) Core(TM) i7-10710U CPU @ 1.10GHz

Thanks for the suggestions.
SuikaPredator is offline   Reply With Quote
Old 2022-09-01, 17:24   #674
SuikaPredator
 
Aug 2022
China

408 Posts
Default

Quote:
Originally Posted by joblack View Post
Yes, I had the same problem. Check out if there is a (cumulative) Windows update. After installing the Windows update and a reboot, it worked again for me.
I tried to install the windows updates and rebooted, but the problem still occurs afterwards.
SuikaPredator is offline   Reply With Quote
Old 2022-09-04, 09:37   #675
kruoli
 
kruoli's Avatar
 
"Oliver"
Sep 2017
Porta Westfalica, DE

131910 Posts
Default

When having a custom affinity block in local.txt on Linux, everything there seems to get mangled with:
Code:
[Worker #1]
Affinity=16,18,20,22,24,26,28,30
Code:
[Work thread Sep 4 11:33] Setting affinity to run worker on logical CPUs 16 (zero-based)
[Work thread Sep 4 11:33] Optimal P-1 factoring of M62353261 using up to 61440MB of memory.
[Work thread Sep 4 11:33] Assuming no factors below 2^75 and 2 primality tests saved if a factor is found.
[Work thread Sep 4 11:33] Optimal bounds are B1=770000, B2=427945000
[Work thread Sep 4 11:33] Chance of finding a factor is an estimated 6.41%
[Work thread Sep 4 11:33]
[Work thread Sep 4 11:33] Using FMA3 FFT length 3360K, Pass1=896, Pass2=3840, clm=1, 8 threads
[Work thread Sep 4 11:33] Setting affinity to run helper thread 7 on logical CPUs 30 (zero-based)
[Work thread Sep 4 11:33] Setting affinity to run helper thread 6 on logical CPUs 28 (zero-based)
[Work thread Sep 4 11:33] Setting affinity to run helper thread 5 on logical CPUs 26 (zero-based)
[Work thread Sep 4 11:33] Setting affinity to run helper thread 4 on logical CPUs 24 (zero-based)
[Work thread Sep 4 11:33] Setting affinity to run helper thread 3 on logical CPUs 22 (zero-based)
[Work thread Sep 4 11:33] Setting affinity to run helper thread 2 on logical CPUs 20 (zero-based)
[Work thread Sep 4 11:33] Setting affinity to run helper thread 1 on logical CPUs 18 (zero-based)
So far that's expected. But in htop, it is shown:
Code:
    1[||                                              1.3%]   9[|||||||||||||||||||||||||||||||||||||||||||||||99.3%]    17[                                                0.0%]  25[                                                0.0%]
    2[|||                                             5.3%]  10[|||||||||||||||||||||||||||||||||||||||||||||||98.6%]    18[                                                0.0%]  26[                                                0.0%]
    3[                                                0.0%]  11[|||||||||||||||||||||||||||||||||||||||||||||||98.6%]    19[                                                0.0%]  27[                                                0.0%]
    4[                                                0.0%]  12[|||||||||||||||||||||||||||||||||||||||||||||||98.0%]    20[                                                0.0%]  28[                                                0.0%]
    5[                                                0.0%]  13[|||||||||||||||||||||||||||||||||||||||||||||||98.0%]    21[                                                0.0%]  29[                                                0.0%]
    6[|                                               0.7%]  14[|||||||||||||||||||||||||||||||||||||||||||||||98.0%]    22[                                                0.0%]  30[                                                0.0%]
    7[                                                0.0%]  15[|||||||||||||||||||||||||||||||||||||||||||||||98.7%]    23[                                                0.0%]  31[                                                0.0%]
    8[                                                0.0%]  16[|||||||||||||||||||||||||||||||||||||||||||||||98.0%]    24[                                                0.0%]  32[                                                0.0%]
Stating 8,9,10,11,... as Affinity will also give mangled results...

Edit: Yes, I know that htop is not 0 based. But this should not lessen my point.

Last fiddled with by kruoli on 2022-09-04 at 09:42 Reason: Clarification.
kruoli is online now   Reply With Quote
Old 2022-09-04, 10:37   #676
Prime95
P90 years forever!
 
Prime95's Avatar
 
Aug 2002
Yeehaw, FL

1FD716 Posts
Default

Prime95 knows nothing about htop's (or Linux's) logical CPU numbering.

Prime95 relies on hwloc's numbering. Run a benchmark (and abort it). The hwloc topology is written to results.bench.txt. What does that look like?
Prime95 is offline   Reply With Quote
Old 2022-09-04, 12:35   #677
kruoli
 
kruoli's Avatar
 
"Oliver"
Sep 2017
Porta Westfalica, DE

1,319 Posts
Default

Unfortunately, I had to abort it, I would have liked to perform the benchmark as usual, but it got stuck:
Code:
[Work thread Sep 4 14:23] Benchmarking multiple workers to measure the impact of memory bandwidth
[Work thread Sep 4 14:23] Timing 3360K FFT, 16 cores, 1 worker.  Average times:  1.12 ms.  Total throughput: 895.28 iter/sec.
[Work thread Sep 4 14:23] Timing 3360K FFT, 16 cores, 2 workers.  Average times:  1.20, 22.45 ms.  Total throughput: 874.65 iter/sec.
[Work thread Sep 4 14:24] Timing 3360K FFT, 16 cores, 4 workers.  Average times:  2.68,  4.74,  2.97, 29.97 ms.  Total throughput: 954.01 iter/sec.
[Work thread Sep 4 14:24] Timing 3360K FFT, 16 cores, 8 workers.  Average times: 22.19, 21.57, 21.59, 13.88, 20.02, 19.42,  7.88, 47.09 ms.  Total throughput: 459.34 iter/sec.
[Work thread Sep 4 14:25] Timing 3360K FFT, 16 cores, 16 workers.
The times do look really weird. The system is idle otherwise. Ctrl+C gives ^C[Sep 4 14:28] Stopping all worker windows., but never progresses, so I have to Ctrl+Z and then kill it.

Meanwhile, the file (hwloc) looks like the numbering of htop output (only 0 based now):
Code:
AMD Ryzen 9 5950X 16-Core Processor
CPU speed: 2236.10 MHz, 15/1 performance/efficiency cores
CPU features: 3DNow! Prefetch, SSE, SSE2, SSE4, AVX, AVX2, FMA
L1 cache size: 16x32 KB, L2 cache size: 16x512 KB, L3 cache size: 2x32 MB
L1 cache line size: 64 bytes, L2 cache line size: 64 bytes
Machine topology as determined by hwloc library:
 Machine#0 (total=65780764KB, DMIProductName="To Be Filled By O.E.M.", DMIProductVersion="To Be Filled By O.E.M.", DMIBoardVendor=ASRock, DMIBoardName="B550 Extreme4", DMIBoardVersion=, DMIBoardAssetTag=, DMIChassisVendor="To Be Filled By O.E.M.", DMIChassisType=3, DMIChassisVersion="To Be Filled By O.E.M.", DMIChassisAssetTag="To Be Filled By O.E.M.", DMIBIOSVendor="American Megatrends International, LLC.", DMIBIOSVersion=P2.00, DMIBIOSDate=04/21/2021, DMISysVendor="To Be Filled By O.E.M.", Backend=Linux, LinuxCgroup=/, OSName=Linux, OSRelease=5.10.0-14-amd64, OSVersion="#1 SMP Debian 5.10.113-1 (2022-04-29)", HostName=chia-debian, Architecture=x86_64, hwlocVersion=2.6.0, ProcessName=mprime)
  Package#0 (total=65780764KB, CPUVendor=AuthenticAMD, CPUFamilyNumber=25, CPUModelNumber=33, CPUModel="AMD Ryzen 9 5950X 16-Core Processor            ", CPUStepping=0)
    L3#0 (size=32768KB, linesize=64, ways=16, Inclusive=0)
      L2#0 (size=512KB, linesize=64, ways=8, Inclusive=1)
        L1d#0 (size=32KB, linesize=64, ways=8, Inclusive=0)
          Core#0 (cpuset: 0x00010001)
            PU#0 (cpuset: 0x00000001)
            PU#16 (cpuset: 0x00010000)
      L2#1 (size=512KB, linesize=64, ways=8, Inclusive=1)
        L1d#1 (size=32KB, linesize=64, ways=8, Inclusive=0)
          Core#1 (cpuset: 0x00020002)
            PU#1 (cpuset: 0x00000002)
            PU#17 (cpuset: 0x00020000)
      L2#2 (size=512KB, linesize=64, ways=8, Inclusive=1)
        L1d#2 (size=32KB, linesize=64, ways=8, Inclusive=0)
          Core#2 (cpuset: 0x00040004)
            PU#2 (cpuset: 0x00000004)
            PU#18 (cpuset: 0x00040000)
      L2#3 (size=512KB, linesize=64, ways=8, Inclusive=1)
        L1d#3 (size=32KB, linesize=64, ways=8, Inclusive=0)
          Core#3 (cpuset: 0x00080008)
            PU#3 (cpuset: 0x00000008)
            PU#19 (cpuset: 0x00080000)
      L2#4 (size=512KB, linesize=64, ways=8, Inclusive=1)
etc.
So mprime does not take the numbers from hwloc, either.

Edit, it says in the above:
15/1 performance/efficiency cores. What?!

Last fiddled with by kruoli on 2022-09-04 at 12:42 Reason: Addition.
kruoli is online now   Reply With Quote
Old 2022-09-04, 13:42   #678
Prime95
P90 years forever!
 
Prime95's Avatar
 
Aug 2002
Yeehaw, FL

3×11×13×19 Posts
Default

Please set "KindVerbosity=1" in prime.txt and get the hwloc topology again. Thanks.

Then in local.txt try "Core0Ranking=1" and "Core15Ranking=1"

Last fiddled with by Prime95 on 2022-09-04 at 13:45
Prime95 is offline   Reply With Quote
Old 2022-09-04, 14:45   #679
kruoli
 
kruoli's Avatar
 
"Oliver"
Sep 2017
Porta Westfalica, DE

131910 Posts
Default

Quote:
Originally Posted by Prime95 View Post
Please set "KindVerbosity=1" in prime.txt and get the hwloc topology again. Thanks.
Code:
[Sun Sep  4 16:37:47 2022]
Hwloc 15 kinds
Hwloc kind 0: Efficiency: 0, Cpuset: 0x20002000
Hwloc kind 1: Efficiency: 1, Cpuset: 0x10001000
Hwloc kind 2: Efficiency: 2, Cpuset: 0x02000200
Hwloc kind 3: Efficiency: 3, Cpuset: 0x04000400
Hwloc kind 4: Efficiency: 4, Cpuset: 0x40004000
Hwloc kind 5: Efficiency: 5, Cpuset: 0x80008000
Hwloc kind 6: Efficiency: 6, Cpuset: 0x01000100
Hwloc kind 7: Efficiency: 7, Cpuset: 0x08000800
Hwloc kind 8: Efficiency: 8, Cpuset: 0x00080008
Hwloc kind 9: Efficiency: 9, Cpuset: 0x00020002
Hwloc kind 10: Efficiency: 10, Cpuset: 0x00800080
Hwloc kind 11: Efficiency: 11, Cpuset: 0x00200020
Hwloc kind 12: Efficiency: 12, Cpuset: 0x00100010
Hwloc kind 13: Efficiency: 13, Cpuset: 0x00040004
Hwloc kind 14: Efficiency: 14, Cpuset: 0x00410041
Hwloc 15 kinds
Hwloc kind 0: Efficiency: 0, Cpuset: 0x20002000
Hwloc kind 1: Efficiency: 1, Cpuset: 0x10001000
Hwloc kind 2: Efficiency: 2, Cpuset: 0x02000200
Hwloc kind 3: Efficiency: 3, Cpuset: 0x04000400
Hwloc kind 4: Efficiency: 4, Cpuset: 0x40004000
Hwloc kind 5: Efficiency: 5, Cpuset: 0x80008000
Hwloc kind 6: Efficiency: 6, Cpuset: 0x01000100
Hwloc kind 7: Efficiency: 7, Cpuset: 0x08000800
Hwloc kind 8: Efficiency: 8, Cpuset: 0x00080008
Hwloc kind 9: Efficiency: 9, Cpuset: 0x00020002
Hwloc kind 10: Efficiency: 10, Cpuset: 0x00800080
Hwloc kind 11: Efficiency: 11, Cpuset: 0x00200020
Hwloc kind 12: Efficiency: 12, Cpuset: 0x00100010
Hwloc kind 13: Efficiency: 13, Cpuset: 0x00040004
Hwloc kind 14: Efficiency: 14, Cpuset: 0x00410041
Compare your results to other computers at http://www.mersenne.org/report_benchmarks
AMD Ryzen 9 5950X 16-Core Processor
CPU speed: 2193.89 MHz, 15/1 performance/efficiency cores
CPU features: 3DNow! Prefetch, SSE, SSE2, SSE4, AVX, AVX2, FMA
L1 cache size: 16x32 KB, L2 cache size: 16x512 KB, L3 cache size: 2x32 MB
L1 cache line size: 64 bytes, L2 cache line size: 64 bytes
Machine topology as determined by hwloc library:
 Machine#0 (total=65780764KB, DMIProductName="To Be Filled By O.E.M.", DMIProductVersion="To Be Filled By O.E.M.", DMIBoardVendor=ASRock, DMIBoardName="B550 Extreme4", DMIBoardVersion=, DMIBoardAssetTag=, DMIChassisVendor="To Be Filled By O.E.M.", DMIChassisType=3, DMIChassisVersion="To Be Filled By O.E.M.", DMIChassisAssetTag="To Be Filled By O.E.M.", DMIBIOSVendor="American Megatrends International, LLC.", DMIBIOSVersion=P2.00, DMIBIOSDate=04/21/2021, DMISysVendor="To Be Filled By O.E.M.", Backend=Linux, LinuxCgroup=/, OSName=Linux, OSRelease=5.10.0-14-amd64, OSVersion="#1 SMP Debian 5.10.113-1 (2022-04-29)", HostName=chia-debian, Architecture=x86_64, hwlocVersion=2.6.0, ProcessName=mprime)
  Package#0 (total=65780764KB, CPUVendor=AuthenticAMD, CPUFamilyNumber=25, CPUModelNumber=33, CPUModel="AMD Ryzen 9 5950X 16-Core Processor            ", CPUStepping=0)
    L3#0 (size=32768KB, linesize=64, ways=16, Inclusive=0)
      L2#0 (size=512KB, linesize=64, ways=8, Inclusive=1)
        L1d#0 (size=32KB, linesize=64, ways=8, Inclusive=0)
          Core#0 (cpuset: 0x00010001)
            PU#0 (cpuset: 0x00000001)
            PU#16 (cpuset: 0x00010000)
      L2#1 (size=512KB, linesize=64, ways=8, Inclusive=1)
        L1d#1 (size=32KB, linesize=64, ways=8, Inclusive=0)
          Core#1 (cpuset: 0x00020002)
            PU#1 (cpuset: 0x00000002)
            PU#17 (cpuset: 0x00020000)
      L2#2 (size=512KB, linesize=64, ways=8, Inclusive=1)
        L1d#2 (size=32KB, linesize=64, ways=8, Inclusive=0)
          Core#2 (cpuset: 0x00040004)
            PU#2 (cpuset: 0x00000004)
            PU#18 (cpuset: 0x00040000)
      L2#3 (size=512KB, linesize=64, ways=8, Inclusive=1)
        L1d#3 (size=32KB, linesize=64, ways=8, Inclusive=0)
          Core#3 (cpuset: 0x00080008)
            PU#3 (cpuset: 0x00000008)
            PU#19 (cpuset: 0x00080000)
      L2#4 (size=512KB, linesize=64, ways=8, Inclusive=1)
etc.
Quote:
Originally Posted by Prime95 View Post
Then in local.txt try "Core0Ranking=1" and "Core15Ranking=1"
The benchmark is still getting stuck...
Code:
[Work thread Sep 4 16:41] Worker starting
[Work thread Sep 4 16:41] Your timings will be written to the results.bench.txt file.
[Work thread Sep 4 16:41] Compare your results to other computers at http://www.mersenne.org/report_benchmarks
[Work thread Sep 4 16:41] Benchmarking multiple workers to measure the impact of memory bandwidth
[Work thread Sep 4 16:41] Timing 3360K FFT, 16 cores, 1 worker.  Average times:  1.11 ms.  Total throughput: 904.39 iter/sec.
[Work thread Sep 4 16:41] Timing 3360K FFT, 16 cores, 2 workers.  Average times:  1.24, 21.79 ms.  Total throughput: 854.31 iter/sec.
[Work thread Sep 4 16:42] Timing 3360K FFT, 16 cores, 4 workers.  Average times:  2.66,  4.82,  2.93, 29.83 ms.  Total throughput: 959.19 iter/sec.
[Work thread Sep 4 16:42] Timing 3360K FFT, 16 cores, 8 workers.  Average times: 20.12, 20.58, 20.08, 20.47, 18.84, 18.83,  6.42, 43.78 ms.  Total throughput: 481.62 iter/sec.
[Work thread Sep 4 16:42] Timing 3360K FFT, 16 cores, 16 workers.
Bench file:
Code:
[Sun Sep  4 16:41:18 2022]
Hwloc 15 kinds
Hwloc kind 0: Efficiency: 0, Cpuset: 0x20002000
Hwloc kind 1: Efficiency: 1, Cpuset: 0x10001000
Hwloc kind 2: Efficiency: 2, Cpuset: 0x02000200
Hwloc kind 3: Efficiency: 3, Cpuset: 0x04000400
Hwloc kind 4: Efficiency: 4, Cpuset: 0x40004000
Hwloc kind 5: Efficiency: 5, Cpuset: 0x80008000
Hwloc kind 6: Efficiency: 6, Cpuset: 0x01000100
Hwloc kind 7: Efficiency: 7, Cpuset: 0x08000800
Hwloc kind 8: Efficiency: 8, Cpuset: 0x00080008
Hwloc kind 9: Efficiency: 9, Cpuset: 0x00020002
Hwloc kind 10: Efficiency: 10, Cpuset: 0x00800080
Hwloc kind 11: Efficiency: 11, Cpuset: 0x00200020
Hwloc kind 12: Efficiency: 12, Cpuset: 0x00100010
Hwloc kind 13: Efficiency: 13, Cpuset: 0x00040004
Hwloc kind 14: Efficiency: 14, Cpuset: 0x00410041
Hwloc 15 kinds
Hwloc kind 0: Efficiency: 0, Cpuset: 0x20002000
Hwloc kind 1: Efficiency: 1, Cpuset: 0x10001000
Hwloc kind 2: Efficiency: 2, Cpuset: 0x02000200
Hwloc kind 3: Efficiency: 3, Cpuset: 0x04000400
Hwloc kind 4: Efficiency: 4, Cpuset: 0x40004000
Hwloc kind 5: Efficiency: 5, Cpuset: 0x80008000
Hwloc kind 6: Efficiency: 6, Cpuset: 0x01000100
Hwloc kind 7: Efficiency: 7, Cpuset: 0x08000800
Hwloc kind 8: Efficiency: 8, Cpuset: 0x00080008
Hwloc kind 9: Efficiency: 9, Cpuset: 0x00020002
Hwloc kind 10: Efficiency: 10, Cpuset: 0x00800080
Hwloc kind 11: Efficiency: 11, Cpuset: 0x00200020
Hwloc kind 12: Efficiency: 12, Cpuset: 0x00100010
Hwloc kind 13: Efficiency: 13, Cpuset: 0x00040004
Hwloc kind 14: Efficiency: 14, Cpuset: 0x00410041
Compare your results to other computers at http://www.mersenne.org/report_benchmarks
AMD Ryzen 9 5950X 16-Core Processor
CPU speed: 2195.93 MHz, 15/1 performance/efficiency cores
CPU features: 3DNow! Prefetch, SSE, SSE2, SSE4, AVX, AVX2, FMA
L1 cache size: 16x32 KB, L2 cache size: 16x512 KB, L3 cache size: 2x32 MB
L1 cache line size: 64 bytes, L2 cache line size: 64 bytes
Machine topology as determined by hwloc library:
 Machine#0 (total=65780764KB, DMIProductName="To Be Filled By O.E.M.", DMIProductVersion="To Be Filled By O.E.M.", DMIBoardVendor=ASRock, DMIBoardName="B550 Extreme4", DMIBoardVersion=, DMIBoardAssetTag=, DMIChassisVendor="To Be Filled By O.E.M.", DMIChassisType=3, DMIChassisVersion="To Be Filled By O.E.M.", DMIChassisAssetTag="To Be Filled By O.E.M.", DMIBIOSVendor="American Megatrends International, LLC.", DMIBIOSVersion=P2.00, DMIBIOSDate=04/21/2021, DMISysVendor="To Be Filled By O.E.M.", Backend=Linux, LinuxCgroup=/, OSName=Linux, OSRelease=5.10.0-14-amd64, OSVersion="#1 SMP Debian 5.10.113-1 (2022-04-29)", HostName=chia-debian, Architecture=x86_64, hwlocVersion=2.6.0, ProcessName=mprime)
  Package#0 (total=65780764KB, CPUVendor=AuthenticAMD, CPUFamilyNumber=25, CPUModelNumber=33, CPUModel="AMD Ryzen 9 5950X 16-Core Processor            ", CPUStepping=0)
    L3#0 (size=32768KB, linesize=64, ways=16, Inclusive=0)
      L2#0 (size=512KB, linesize=64, ways=8, Inclusive=1)
        L1d#0 (size=32KB, linesize=64, ways=8, Inclusive=0)
          Core#0 (cpuset: 0x00010001)
            PU#0 (cpuset: 0x00000001)
            PU#16 (cpuset: 0x00010000)
      L2#1 (size=512KB, linesize=64, ways=8, Inclusive=1)
        L1d#1 (size=32KB, linesize=64, ways=8, Inclusive=0)
          Core#1 (cpuset: 0x00020002)
            PU#1 (cpuset: 0x00000002)
            PU#17 (cpuset: 0x00020000)
      L2#2 (size=512KB, linesize=64, ways=8, Inclusive=1)
        L1d#2 (size=32KB, linesize=64, ways=8, Inclusive=0)
          Core#2 (cpuset: 0x00040004)
            PU#2 (cpuset: 0x00000004)
            PU#18 (cpuset: 0x00040000)
      L2#3 (size=512KB, linesize=64, ways=8, Inclusive=1)
        L1d#3 (size=32KB, linesize=64, ways=8, Inclusive=0)
          Core#3 (cpuset: 0x00080008)
            PU#3 (cpuset: 0x00000008)
            PU#19 (cpuset: 0x00080000)
      L2#4 (size=512KB, linesize=64, ways=8, Inclusive=1)
        L1d#4 (size=32KB, linesize=64, ways=8, Inclusive=0)
          Core#4 (cpuset: 0x00100010)
            PU#4 (cpuset: 0x00000010)
            PU#20 (cpuset: 0x00100000)
      L2#5 (size=512KB, linesize=64, ways=8, Inclusive=1)
        L1d#5 (size=32KB, linesize=64, ways=8, Inclusive=0)
          Core#5 (cpuset: 0x00200020)
            PU#5 (cpuset: 0x00000020)
            PU#21 (cpuset: 0x00200000)
      L2#6 (size=512KB, linesize=64, ways=8, Inclusive=1)
        L1d#6 (size=32KB, linesize=64, ways=8, Inclusive=0)
          Core#6 (cpuset: 0x00400040)
            PU#6 (cpuset: 0x00000040)
            PU#22 (cpuset: 0x00400000)
      L2#7 (size=512KB, linesize=64, ways=8, Inclusive=1)
        L1d#7 (size=32KB, linesize=64, ways=8, Inclusive=0)
          Core#7 (cpuset: 0x00800080)
            PU#7 (cpuset: 0x00000080)
            PU#23 (cpuset: 0x00800000)
    L3#1 (size=32768KB, linesize=64, ways=16, Inclusive=0)
      L2#8 (size=512KB, linesize=64, ways=8, Inclusive=1)
        L1d#8 (size=32KB, linesize=64, ways=8, Inclusive=0)
          Core#8 (cpuset: 0x01000100)
            PU#8 (cpuset: 0x00000100)
            PU#24 (cpuset: 0x01000000)
      L2#9 (size=512KB, linesize=64, ways=8, Inclusive=1)
        L1d#9 (size=32KB, linesize=64, ways=8, Inclusive=0)
          Core#9 (cpuset: 0x02000200)
            PU#9 (cpuset: 0x00000200)
            PU#25 (cpuset: 0x02000000)
      L2#10 (size=512KB, linesize=64, ways=8, Inclusive=1)
        L1d#10 (size=32KB, linesize=64, ways=8, Inclusive=0)
          Core#10 (cpuset: 0x04000400)
            PU#10 (cpuset: 0x00000400)
            PU#26 (cpuset: 0x04000000)
      L2#11 (size=512KB, linesize=64, ways=8, Inclusive=1)
        L1d#11 (size=32KB, linesize=64, ways=8, Inclusive=0)
          Core#11 (cpuset: 0x08000800)
            PU#11 (cpuset: 0x00000800)
            PU#27 (cpuset: 0x08000000)
      L2#12 (size=512KB, linesize=64, ways=8, Inclusive=1)
        L1d#12 (size=32KB, linesize=64, ways=8, Inclusive=0)
          Core#12 (cpuset: 0x10001000)
            PU#12 (cpuset: 0x00001000)
            PU#28 (cpuset: 0x10000000)
      L2#13 (size=512KB, linesize=64, ways=8, Inclusive=1)
        L1d#13 (size=32KB, linesize=64, ways=8, Inclusive=0)
          Core#13 (cpuset: 0x20002000)
            PU#13 (cpuset: 0x00002000)
            PU#29 (cpuset: 0x20000000)
      L2#14 (size=512KB, linesize=64, ways=8, Inclusive=1)
        L1d#14 (size=32KB, linesize=64, ways=8, Inclusive=0)
          Core#14 (cpuset: 0x40004000)
            PU#14 (cpuset: 0x00004000)
            PU#30 (cpuset: 0x40000000)
      L2#15 (size=512KB, linesize=64, ways=8, Inclusive=1)
        L1d#15 (size=32KB, linesize=64, ways=8, Inclusive=0)
          Core#15 (cpuset: 0x80008000)
            PU#15 (cpuset: 0x00008000)
            PU#31 (cpuset: 0x80000000)
kruoli is online now   Reply With Quote
Old 2022-09-04, 16:03   #680
Prime95
P90 years forever!
 
Prime95's Avatar
 
Aug 2002
Yeehaw, FL

11111110101112 Posts
Default

Quote:
Originally Posted by Prime95 View Post
Then in local.txt try "Core0Ranking=1" and "Core15Ranking=1"
Try setting all 16 rankings to 1 in local.txt as well as:

NumThreadingNodes=1
NumComputeThreadingNodes=1

if that fails try version 30.6
Prime95 is offline   Reply With Quote
Old 2022-09-04, 16:09   #681
kruoli
 
kruoli's Avatar
 
"Oliver"
Sep 2017
Porta Westfalica, DE

1,319 Posts
Default

Thank you, now it works fine.
Code:
[Work thread Sep 4 18:06] Timing 3360K FFT, 16 cores, 1 worker.  Average times:  1.13 ms.  Total throughput: 881.82 iter/sec.
[Work thread Sep 4 18:07] Timing 3360K FFT, 16 cores, 2 workers.  Average times:  1.91,  1.78 ms.  Total throughput: 1084.77 iter/sec.
[Work thread Sep 4 18:07] Timing 3360K FFT, 16 cores, 4 workers.  Average times:  3.81,  8.42,  3.75,  8.32 ms.  Total throughput: 768.21 iter/sec.
[Work thread Sep 4 18:07] Timing 3360K FFT, 16 cores, 8 workers.  Average times: 19.69, 20.47, 19.89, 19.79, 20.06, 19.52, 19.50, 19.90 ms.  Total throughput: 403.06 iter/sec.
[Work thread Sep 4 18:07] Timing 3360K FFT, 16 cores, 16 workers.  Average times: 45.31, 45.74, 46.31, 45.78, 45.80, 46.49, 46.35, 46.28, 45.49, 46.07, 45.56, 45.99, 45.78, 46.66, 45.91, 45.91 ms.  Total throughput: 348.12 iter/sec.
Although the bench file stilll displays the things weirdly:
Code:
[Sun Sep  4 18:06:27 2022]
Hwloc 15 kinds
Hwloc kind 0: Efficiency: 0, Cpuset: 0x20002000
Hwloc kind 1: Efficiency: 1, Cpuset: 0x10001000
Hwloc kind 2: Efficiency: 2, Cpuset: 0x02000200
Hwloc kind 3: Efficiency: 3, Cpuset: 0x04000400
Hwloc kind 4: Efficiency: 4, Cpuset: 0x40004000
Hwloc kind 5: Efficiency: 5, Cpuset: 0x80008000
Hwloc kind 6: Efficiency: 6, Cpuset: 0x01000100
Hwloc kind 7: Efficiency: 7, Cpuset: 0x08000800
Hwloc kind 8: Efficiency: 8, Cpuset: 0x00080008
Hwloc kind 9: Efficiency: 9, Cpuset: 0x00020002
Hwloc kind 10: Efficiency: 10, Cpuset: 0x00800080
Hwloc kind 11: Efficiency: 11, Cpuset: 0x00200020
Hwloc kind 12: Efficiency: 12, Cpuset: 0x00100010
Hwloc kind 13: Efficiency: 13, Cpuset: 0x00040004
Hwloc kind 14: Efficiency: 14, Cpuset: 0x00410041
Hwloc 15 kinds
Hwloc kind 0: Efficiency: 0, Cpuset: 0x20002000
Hwloc kind 1: Efficiency: 1, Cpuset: 0x10001000
Hwloc kind 2: Efficiency: 2, Cpuset: 0x02000200
Hwloc kind 3: Efficiency: 3, Cpuset: 0x04000400
Hwloc kind 4: Efficiency: 4, Cpuset: 0x40004000
Hwloc kind 5: Efficiency: 5, Cpuset: 0x80008000
Hwloc kind 6: Efficiency: 6, Cpuset: 0x01000100
Hwloc kind 7: Efficiency: 7, Cpuset: 0x08000800
Hwloc kind 8: Efficiency: 8, Cpuset: 0x00080008
Hwloc kind 9: Efficiency: 9, Cpuset: 0x00020002
Hwloc kind 10: Efficiency: 10, Cpuset: 0x00800080
Hwloc kind 11: Efficiency: 11, Cpuset: 0x00200020
Hwloc kind 12: Efficiency: 12, Cpuset: 0x00100010
Hwloc kind 13: Efficiency: 13, Cpuset: 0x00040004
Hwloc kind 14: Efficiency: 14, Cpuset: 0x00410041
kruoli is online now   Reply With Quote
Old 2022-09-04, 16:24   #682
Prime95
P90 years forever!
 
Prime95's Avatar
 
Aug 2002
Yeehaw, FL

3×11×13×19 Posts
Default

I may try linking with hwloc 2.8.0 or 2.7.1 for you to test. All those local.txt entries basically overrode the information returned by hwloc. Prime95 though has a bug -- it should not have hung.
Prime95 is offline   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
Do not post your results here! kar_bon Prime Wiki 40 2022-04-03 19:05
what should I post ? science_man_88 science_man_88 24 2018-10-19 23:00
Where to post job ad? xilman Linux 2 2010-12-15 16:39
Moderated Post kar_bon Forum Feedback 3 2010-09-28 08:01
Something that I just had to post/buy dave_0273 Lounge 1 2005-02-27 18:36

All times are UTC. The time now is 16:55.


Sat Jan 28 16:55:34 UTC 2023 up 163 days, 14:24, 0 users, load averages: 0.78, 0.84, 0.91

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2023, Jelsoft Enterprises Ltd.

This forum has received and complied with 0 (zero) government requests for information.

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.
A copy of the license is included in the FAQ.

≠ ± ∓ ÷ × · − √ ‰ ⊗ ⊕ ⊖ ⊘ ⊙ ≤ ≥ ≦ ≧ ≨ ≩ ≺ ≻ ≼ ≽ ⊏ ⊐ ⊑ ⊒ ² ³ °
∠ ∟ ° ≅ ~ ‖ ⟂ ⫛
≡ ≜ ≈ ∝ ∞ ≪ ≫ ⌊⌋ ⌈⌉ ∘ ∏ ∐ ∑ ∧ ∨ ∩ ∪ ⨀ ⊕ ⊗ 𝖕 𝖖 𝖗 ⊲ ⊳
∅ ∖ ∁ ↦ ↣ ∩ ∪ ⊆ ⊂ ⊄ ⊊ ⊇ ⊃ ⊅ ⊋ ⊖ ∈ ∉ ∋ ∌ ℕ ℤ ℚ ℝ ℂ ℵ ℶ ℷ ℸ 𝓟
¬ ∨ ∧ ⊕ → ← ⇒ ⇐ ⇔ ∀ ∃ ∄ ∴ ∵ ⊤ ⊥ ⊢ ⊨ ⫤ ⊣ … ⋯ ⋮ ⋰ ⋱
∫ ∬ ∭ ∮ ∯ ∰ ∇ ∆ δ ∂ ℱ ℒ ℓ
𝛢𝛼 𝛣𝛽 𝛤𝛾 𝛥𝛿 𝛦𝜀𝜖 𝛧𝜁 𝛨𝜂 𝛩𝜃𝜗 𝛪𝜄 𝛫𝜅 𝛬𝜆 𝛭𝜇 𝛮𝜈 𝛯𝜉 𝛰𝜊 𝛱𝜋 𝛲𝜌 𝛴𝜎𝜍 𝛵𝜏 𝛶𝜐 𝛷𝜙𝜑 𝛸𝜒 𝛹𝜓 𝛺𝜔