mersenneforum.org  

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

Reply
 
Thread Tools
Old 2018-10-20, 05:05   #1
Prime95
P90 years forever!
 
Prime95's Avatar
 
Aug 2002
Yeehaw, FL

22×1,873 Posts
Default SkylakeX teasers (aka prime95 29.5)

If you own a Skylake-X machine (or a Xeon supporting AVX-512) and you are feeling brave then Prime95 version 29.5 build 2 supporting AVX-512 is available.

This is not heavily tested but ought to work. Doing a double-check or Gerbicz PRP might be wise. Let me know if you find bugs or it is not an improvement over 29.4.

Download link:

____________original d/l link removed_______________


Linux, Windows 64-bit: EDIT: new download links in post #163

Last fiddled with by Prime95 on 2019-01-21 at 04:07
Prime95 is offline   Reply With Quote
Old 2018-10-20, 07:52   #2
mackerel
 
mackerel's Avatar
 
Feb 2016
UK

419 Posts
Default

I'd like to run benchmarks with it once Windows version is available.
mackerel is offline   Reply With Quote
Old 2018-10-20, 08:39   #3
ET_
Banned
 
ET_'s Avatar
 
"Luigi"
Aug 2002
Team Italia

2×29×83 Posts
Default

I'd like to run benchmarks with it once a 7820X is available to me.
ET_ is online now   Reply With Quote
Old 2018-10-20, 11:46   #4
ET_
Banned
 
ET_'s Avatar
 
"Luigi"
Aug 2002
Team Italia

2×29×83 Posts
Default

Quote:
Originally Posted by Prime95 View Post
If you own a Skylake-X machine (or a Xeon supporting AVX-512) and you are feeling brave then Prime95 version 29.5 build 2 supporting AVX-512 is available.

This is not heavily tested but ought to work. Doing a double-check or Gerbicz PRP might be wise. Let me know if you find bugs or it is not an improvement over 29.4.

Download link:
Linux 64-bit: ftp://mersenne.org/gimps/p95v295b2.linux64.tar.gz

If there is enough interest, I'll make a Windows build.
Is it safe stopping a run with v29.4 and restarting it with 29.5, or the savefile is not compatible?

EDIT: Yess, it is. Restarted after a Jacobi test. Running a test on AWS.

Last fiddled with by ET_ on 2018-10-20 at 12:27
ET_ is online now   Reply With Quote
Old 2018-10-20, 14:25   #5
Chuck
 
Chuck's Avatar
 
May 2011
Orange Park, FL

2×443 Posts
Default

I'd like a Windows version to try.
Chuck is offline   Reply With Quote
Old 2018-10-20, 14:59   #6
GP2
 
GP2's Avatar
 
Sep 2003

2·5·7·37 Posts
Default

It crashes in the benchmark:

Code:
Benchmark type (0 = Throughput, 1 = FFT timings, 2 = Trial factoring) (0):

FFTs to benchmark
Minimum FFT size (in K) (2048):
Maximum FFT size (in K) (8192):
Benchmark with round-off checking enabled (N):
Benchmark all-complex FFTs (for LLR,PFGW,PRP users) (N):
Limit FFT sizes (mimic older benchmarking code) (N):

CPU cores to benchmark
Benchmark hyperthreading (Y):

Throughput benchmark options
Benchmark all FFT implementations to find best one for your machine (N):
Time to run each benchmark (in seconds) (15):

Accept the answers above? (Y):

....

[Work thread Oct 20 14:43] Timing 4800K FFT, 1 core, 1 worker.  Average times: 21.78 ms.  Total throughput: 45.92 iter/sec.
[Work thread Oct 20 14:43] Timing 4800K FFT, 1 core hyperthreaded, 1 worker.  Average times: 26.67 ms.  Total throughput: 37.50 iter/sec.
free(): invalid pointer
Aborted
Second run crashed in a different place:

Code:
[Work thread Oct 20 14:55] Timing 3000K FFT, 1 core, 1 worker.  Average times: 11.69 ms.  Total throughput: 85.56 iter/sec.
[Work thread Oct 20 14:56] Timing 3000K FFT, 1 core hyperthreaded, 1 worker.  Average times: 13.79 ms.  Total throughput: 72.49 iter/sec.
free(): invalid pointer
Aborted

Code:
model name      : Intel(R) Xeon(R) Platinum 8124M CPU @ 3.00GHz
stepping        : 4
microcode       : 0x2000043
cpu MHz         : 3411.502
cache size      : 25344 KB
GP2 is offline   Reply With Quote
Old 2018-10-20, 19:14   #7
ATH
Einyen
 
ATH's Avatar
 
Dec 2003
Denmark

3×5×11×19 Posts
Default

More crashes in the Throughput benchmark on EC2 c5d.large.

0: Throughput
1K - 8192K
roundoff: No
All-complex: No
Limit FFT sizes: No
hyperthreading: Y
all FFT: N
15 sec

After 12K FFT if cannot find next size, no matter if doing 1K-8192K or 10K-20K:
Quote:
*** Error in `./mprime': free(): invalid next size (normal): 0x00007f92200505e0 ***
At 4800K and 5760K it fails several times when benchmarking from lower sizes, but not when starting from those sizes going up.

Failed once at 46080K but not 2nd time.

Code:
Compare your results to other computers at http://www.mersenne.org/report_benchmarks
Intel(R) Xeon(R)
CPU speed: 2999.97 MHz, with hyperthreading
CPU features: Prefetchw, SSE, SSE2, SSE4, AVX, AVX2, FMA, AVX512F
L1 cache size: 32 KB
L2 cache size: 256 KB, L3 cache size: 25344 KB
L1 cache line size: 64 bytes
L2 cache line size: 64 bytes
TLBS: 64
Machine topology as determined by hwloc library:
 Machine#0 (local=3803772KB, total=3803772KB, DMIProductName=c5d.large, DMIProductVersion=, DMIBoardVendor="Amazon EC2", DMIBoardName=, DMIBoardVersion=, DMIBoardAssetTag=i-083f59d8339403d6c, DMIChassisVendor="Amazon EC2", DMIChassisType=1, DMIChassisVersion=, DMIChassisAssetTag="Amazon EC2", DMIBIOSVendor="Amazon EC2", DMIBIOSVersion=1.0, DMIBIOSDate=10/16/2017, DMISysVendor="Amazon EC2", Backend=Linux, OSName=Linux, OSRelease=4.9.76-3.78.amzn1.x86_64, OSVersion="#1 SMP Fri Jan 12 19:51:35 UTC 2018", HostName=ip-172-31-35-125, Architecture=x86_64, hwlocVersion=1.11.10, ProcessName=mprime)
  Package#0 (CPUVendor=GenuineIntel, CPUFamilyNumber=6, CPUModelNumber=85, CPUModel="Intel(R) Xeon(R) Platinum 8124M CPU @ 3.00GHz", CPUStepping=4)
    L3 (size=25344KB, linesize=64, ways=11, Inclusive=0)
      L2 (size=1024KB, linesize=64, ways=16, Inclusive=0)
        L1d (size=32KB, linesize=64, ways=8, Inclusive=0)
          Core#0 (cpuset: 0x00000003)
            PU#0 (cpuset: 0x00000001)
            PU#1 (cpuset: 0x00000002)





[Work thread Oct 20 16:49] Timing 12K FFT, 1 core, 1 worker.  Average times:  0.02 ms.  Total throughput: 40185.95 iter/sec.
[Work thread Oct 20 16:49] Timing 12K FFT, 1 core hyperthreaded, 1 worker.  Average times:  0.04 ms.  Total throughput: 23843.36 iter/sec.
*** Error in `./mprime': free(): invalid next size (normal): 0x00007f92200505e0 ***
======= Backtrace: =========
/lib64/libc.so.6(+0x81499)[0x7f9228ae2499]
./mprime[0x45ecf6]
./mprime[0x4409fc]
./mprime[0x440f9e]
./mprime[0x444bb0]
./mprime[0x444cf2]
./mprime[0x448d54]
./mprime[0x47cbca]
/lib64/libpthread.so.0(+0x7de5)[0x7f92294b7de5]
/lib64/libc.so.6(clone+0x6d)[0x7f9228b5fbad]
======= Memory map: ========
00400000-026a0000 r-xp 00000000 00:15 5878764499047460622                /mnt-efs/z/p95v295b2/mprime
0289f000-028a1000 r-xp 0229f000 00:15 5878764499047460622                /mnt-efs/z/p95v295b2/mprime
028a1000-028dc000 rwxp 022a1000 00:15 5878764499047460622                /mnt-efs/z/p95v295b2/mprime
028dc000-02903000 rwxp 00000000 00:00 0
02ffa000-0301b000 rwxp 00000000 00:00 0                                  [heap]
7f9218000000-7f921804b000 rwxp 00000000 00:00 0
7f921804b000-7f921c000000 ---p 00000000 00:00 0
7f921c000000-7f921c021000 rwxp 00000000 00:00 0
7f921c021000-7f9220000000 ---p 00000000 00:00 0
7f9220000000-7f9220076000 rwxp 00000000 00:00 0
7f9220076000-7f9224000000 ---p 00000000 00:00 0
7f9226847000-7f922685d000 r-xp 00000000 103:02 2617                      /lib64/libgcc_s-7-20170915.so.1
7f922685d000-7f9226a5c000 ---p 00016000 103:02 2617                      /lib64/libgcc_s-7-20170915.so.1
7f9226a5c000-7f9226a5d000 rwxp 00015000 103:02 2617                      /lib64/libgcc_s-7-20170915.so.1
7f9226a5d000-7f9226a5e000 ---p 00000000 00:00 0
7f9226a5e000-7f922725e000 rwxp 00000000 00:00 0
7f922725e000-7f922725f000 ---p 00000000 00:00 0
7f922725f000-7f9227a5f000 rwxp 00000000 00:00 0
7f9227a5f000-7f9227a60000 ---p 00000000 00:00 0
7f9227a60000-7f9228260000 rwxp 00000000 00:00 0
7f9228260000-7f9228261000 ---p 00000000 00:00 0
7f9228261000-7f9228a61000 rwxp 00000000 00:00 0
7f9228a61000-7f9228c24000 r-xp 00000000 103:02 2690                      /lib64/libc-2.17.so
7f9228c24000-7f9228e23000 ---p 001c3000 103:02 2690                      /lib64/libc-2.17.so
7f9228e23000-7f9228e27000 r-xp 001c2000 103:02 2690                      /lib64/libc-2.17.so
7f9228e27000-7f9228e29000 rwxp 001c6000 103:02 2690                      /lib64/libc-2.17.so
7f9228e29000-7f9228e2e000 rwxp 00000000 00:00 0
7f9228e2e000-7f9228ea3000 r-xp 00000000 103:02 3438                      /usr/lib64/libgmp.so.10.2.0
7f9228ea3000-7f92290a2000 ---p 00075000 103:02 3438                      /usr/lib64/libgmp.so.10.2.0
7f92290a2000-7f92290a4000 rwxp 00074000 103:02 3438                      /usr/lib64/libgmp.so.10.2.0
7f92290a4000-7f92290a6000 r-xp 00000000 103:02 2696                      /lib64/libdl-2.17.so
7f92290a6000-7f92292a6000 ---p 00002000 103:02 2696                      /lib64/libdl-2.17.so
7f92292a6000-7f92292a7000 r-xp 00002000 103:02 2696                      /lib64/libdl-2.17.so
7f92292a7000-7f92292a8000 rwxp 00003000 103:02 2696                      /lib64/libdl-2.17.so
7f92292a8000-7f92292af000 r-xp 00000000 103:02 2720                      /lib64/librt-2.17.so
7f92292af000-7f92294ae000 ---p 00007000 103:02 2720                      /lib64/librt-2.17.so
7f92294ae000-7f92294af000 r-xp 00006000 103:02 2720                      /lib64/librt-2.17.so
7f92294af000-7f92294b0000 rwxp 00007000 103:02 2720                      /lib64/librt-2.17.so
7f92294b0000-7f92294c7000 r-xp 00000000 103:02 2716                      /lib64/libpthread-2.17.so
7f92294c7000-7f92296c6000 ---p 00017000 103:02 2716                      /lib64/libpthread-2.17.so
7f92296c6000-7f92296c7000 r-xp 00016000 103:02 2716                      /lib64/libpthread-2.17.so
7f92296c7000-7f92296c8000 rwxp 00017000 103:02 2716                      /lib64/libpthread-2.17.so
7f92296c8000-7f92296cc000 rwxp 00000000 00:00 0
7f92296cc000-7f92297cd000 r-xp 00000000 103:02 2698                      /lib64/libm-2.17.so
7f92297cd000-7f92299cc000 ---p 00101000 103:02 2698                      /lib64/libm-2.17.so
7f92299cc000-7f92299cd000 r-xp 00100000 103:02 2698                      /lib64/libm-2.17.so
7f92299cd000-7f92299ce000 rwxp 00101000 103:02 2698                      /lib64/libm-2.17.so
7f92299ce000-7f92299f0000 r-xp 00000000 103:02 2683                      /lib64/ld-2.17.so
7f9229be4000-7f9229be9000 rwxp 00000000 00:00 0
7f9229bec000-7f9229bef000 rwxp 00000000 00:00 0
7f9229bef000-7f9229bf0000 r-xp 00021000 103:02 2683                      /lib64/ld-2.17.so
7f9229bf0000-7f9229bf1000 rwxp 00022000 103:02 2683                      /lib64/ld-2.17.so
7f9229bf1000-7f9229bf2000 rwxp 00000000 00:00 0
7ffce22f0000-7ffce2311000 rwxp 00000000 00:00 0                          [stack]
7ffce23d6000-7ffce23d8000 r--p 00000000 00:00 0                          [vvar]
7ffce23d8000-7ffce23da000 r-xp 00000000 00:00 0                          [vdso]
ffffffffff600000-ffffffffff601000 r-xp 00000000 00:00 0                  [vsyscall]






[Work thread Oct 20 17:51] Timing 4800K FFT, 1 core, 1 worker.  Average times: 18.15 ms.  Total throughput: 55.10 iter/sec.
[Work thread Oct 20 17:51] Timing 4800K FFT, 1 core hyperthreaded, 1 worker.  Average times: 23.30 ms.  Total throughput: 42.91 iter/sec.
*** Error in `./mprime': free(): invalid pointer: 0x00007f619843cbc0 ***
======= Backtrace: =========
/lib64/libc.so.6(+0x81499)[0x7f61a0776499]
./mprime[0x4582e8]
./mprime[0x45ed65]
./mprime[0x440a3a]
./mprime[0x440f9e]
./mprime[0x444bb0]
./mprime[0x444cf2]
./mprime[0x448d54]
./mprime[0x47cbca]
/lib64/libpthread.so.0(+0x7de5)[0x7f61a114bde5]
/lib64/libc.so.6(clone+0x6d)[0x7f61a07f3bad]
======= Memory map: ========
00400000-026a0000 r-xp 00000000 00:15 5878764499047460622                /mnt-efs/z/p95v295b2/mprime
0289f000-028a1000 r-xp 0229f000 00:15 5878764499047460622                /mnt-efs/z/p95v295b2/mprime
028a1000-028dc000 rwxp 022a1000 00:15 5878764499047460622                /mnt-efs/z/p95v295b2/mprime
028dc000-02903000 rwxp 00000000 00:00 0
03e2d000-03e4e000 rwxp 00000000 00:00 0                                  [heap]
7f618fdea000-7f618fe00000 r-xp 00000000 103:02 2617                      /lib64/libgcc_s-7-20170915.so.1
7f618fe00000-7f618ffff000 ---p 00016000 103:02 2617                      /lib64/libgcc_s-7-20170915.so.1
7f618ffff000-7f6190000000 rwxp 00015000 103:02 2617                      /lib64/libgcc_s-7-20170915.so.1
7f6190000000-7f6192743000 rwxp 00000000 00:00 0
7f6192743000-7f6194000000 ---p 00000000 00:00 0
7f6194000000-7f6194021000 rwxp 00000000 00:00 0
7f6194021000-7f6198000000 ---p 00000000 00:00 0
7f6198000000-7f619896e000 rwxp 00000000 00:00 0
7f619896e000-7f619c000000 ---p 00000000 00:00 0
7f619c158000-7f619e6f1000 rwxp 00000000 00:00 0
7f619e6f1000-7f619e6f2000 ---p 00000000 00:00 0
7f619e6f2000-7f619eef2000 rwxp 00000000 00:00 0
7f619eef2000-7f619eef3000 ---p 00000000 00:00 0
7f619eef3000-7f619f6f3000 rwxp 00000000 00:00 0
7f619f6f3000-7f619f6f4000 ---p 00000000 00:00 0
7f619f6f4000-7f619fef4000 rwxp 00000000 00:00 0
7f619fef4000-7f619fef5000 ---p 00000000 00:00 0
7f619fef5000-7f61a06f5000 rwxp 00000000 00:00 0
7f61a06f5000-7f61a08b8000 r-xp 00000000 103:02 2690                      /lib64/libc-2.17.so
7f61a08b8000-7f61a0ab7000 ---p 001c3000 103:02 2690                      /lib64/libc-2.17.so
7f61a0ab7000-7f61a0abb000 r-xp 001c2000 103:02 2690                      /lib64/libc-2.17.so
7f61a0abb000-7f61a0abd000 rwxp 001c6000 103:02 2690                      /lib64/libc-2.17.so
7f61a0abd000-7f61a0ac2000 rwxp 00000000 00:00 0
7f61a0ac2000-7f61a0b37000 r-xp 00000000 103:02 3438                      /usr/lib64/libgmp.so.10.2.0
7f61a0b37000-7f61a0d36000 ---p 00075000 103:02 3438                      /usr/lib64/libgmp.so.10.2.0
7f61a0d36000-7f61a0d38000 rwxp 00074000 103:02 3438                      /usr/lib64/libgmp.so.10.2.0
7f61a0d38000-7f61a0d3a000 r-xp 00000000 103:02 2696                      /lib64/libdl-2.17.so
7f61a0d3a000-7f61a0f3a000 ---p 00002000 103:02 2696                      /lib64/libdl-2.17.so
7f61a0f3a000-7f61a0f3b000 r-xp 00002000 103:02 2696                      /lib64/libdl-2.17.so
7f61a0f3b000-7f61a0f3c000 rwxp 00003000 103:02 2696                      /lib64/libdl-2.17.so
7f61a0f3c000-7f61a0f43000 r-xp 00000000 103:02 2720                      /lib64/librt-2.17.so
7f61a0f43000-7f61a1142000 ---p 00007000 103:02 2720                      /lib64/librt-2.17.so
7f61a1142000-7f61a1143000 r-xp 00006000 103:02 2720                      /lib64/librt-2.17.so
7f61a1143000-7f61a1144000 rwxp 00007000 103:02 2720                      /lib64/librt-2.17.so
7f61a1144000-7f61a115b000 r-xp 00000000 103:02 2716                      /lib64/libpthread-2.17.so
7f61a115b000-7f61a135a000 ---p 00017000 103:02 2716                      /lib64/libpthread-2.17.so
7f61a135a000-7f61a135b000 r-xp 00016000 103:02 2716                      /lib64/libpthread-2.17.so
7f61a135b000-7f61a135c000 rwxp 00017000 103:02 2716                      /lib64/libpthread-2.17.so
7f61a135c000-7f61a1360000 rwxp 00000000 00:00 0
7f61a1360000-7f61a1461000 r-xp 00000000 103:02 2698                      /lib64/libm-2.17.so
7f61a1461000-7f61a1660000 ---p 00101000 103:02 2698                      /lib64/libm-2.17.so
7f61a1660000-7f61a1661000 r-xp 00100000 103:02 2698                      /lib64/libm-2.17.so
7f61a1661000-7f61a1662000 rwxp 00101000 103:02 2698                      /lib64/libm-2.17.so
7f61a1662000-7f61a1684000 r-xp 00000000 103:02 2683                      /lib64/ld-2.17.so
7f61a1878000-7f61a187d000 rwxp 00000000 00:00 0
7f61a1880000-7f61a1883000 rwxp 00000000 00:00 0
7f61a1883000-7f61a1884000 r-xp 00021000 103:02 2683                      /lib64/ld-2.17.so
7f61a1884000-7f61a1885000 rwxp 00022000 103:02 2683                      /lib64/ld-2.17.so
7f61a1885000-7f61a1886000 rwxp 00000000 00:00 0
7ffd15a2b000-7ffd15a4c000 rwxp 00000000 00:00 0                          [stack]
7ffd15ad3000-7ffd15ad5000 r--p 00000000 00:00 0                          [vvar]
7ffd15ad5000-7ffd15ad7000 r-xp 00000000 00:00 0                          [vdso]
ffffffffff600000-ffffffffff601000 r-xp 00000000 00:00 0                  [vsyscall]





[Work thread Oct 20 18:17] Timing 5760K FFT, 1 core, 1 worker.  Average times: 22.84 ms.  Total throughput: 43.78 iter/sec.
[Work thread Oct 20 18:17] Timing 5760K FFT, 1 core hyperthreaded, 1 worker.  Average times: 29.25 ms.  Total throughput: 34.19 iter/sec.
*** Error in `./mprime': free(): invalid pointer: 0x00007f0f304febc0 ***
======= Backtrace: =========
/lib64/libc.so.6(+0x81499)[0x7f0f37d4f499]
./mprime[0x4582e8]
./mprime[0x45ed65]
./mprime[0x440a3a]
./mprime[0x440f9e]
./mprime[0x444bb0]
./mprime[0x444cf2]
./mprime[0x448d54]
./mprime[0x47cbca]
/lib64/libpthread.so.0(+0x7de5)[0x7f0f38724de5]
/lib64/libc.so.6(clone+0x6d)[0x7f0f37dccbad]
======= Memory map: ========
00400000-026a0000 r-xp 00000000 00:15 5878764499047460622                /mnt-efs/z/p95v295b2/mprime
0289f000-028a1000 r-xp 0229f000 00:15 5878764499047460622                /mnt-efs/z/p95v295b2/mprime
028a1000-028dc000 rwxp 022a1000 00:15 5878764499047460622                /mnt-efs/z/p95v295b2/mprime
028dc000-02903000 rwxp 00000000 00:00 0
03431000-03452000 rwxp 00000000 00:00 0                                  [heap]
7f0f20000000-7f0f20021000 rwxp 00000000 00:00 0
7f0f20021000-7f0f24000000 ---p 00000000 00:00 0
7f0f28000000-7f0f28b4c000 rwxp 00000000 00:00 0
7f0f28b4c000-7f0f2c000000 ---p 00000000 00:00 0
7f0f2d2e7000-7f0f30000000 rwxp 00000000 00:00 0
7f0f30000000-7f0f305c4000 rwxp 00000000 00:00 0
7f0f305c4000-7f0f34000000 ---p 00000000 00:00 0
7f0f358b7000-7f0f358b8000 ---p 00000000 00:00 0
7f0f358b8000-7f0f360b8000 rwxp 00000000 00:00 0
7f0f362b5000-7f0f362cb000 r-xp 00000000 103:02 2617                      /lib64/libgcc_s-7-20170915.so.1
7f0f362cb000-7f0f364ca000 ---p 00016000 103:02 2617                      /lib64/libgcc_s-7-20170915.so.1
7f0f364ca000-7f0f364cb000 rwxp 00015000 103:02 2617                      /lib64/libgcc_s-7-20170915.so.1
7f0f364cb000-7f0f364cc000 ---p 00000000 00:00 0
7f0f364cc000-7f0f36ccc000 rwxp 00000000 00:00 0
7f0f36ccc000-7f0f36ccd000 ---p 00000000 00:00 0
7f0f36ccd000-7f0f374cd000 rwxp 00000000 00:00 0
7f0f374cd000-7f0f374ce000 ---p 00000000 00:00 0
7f0f374ce000-7f0f37cce000 rwxp 00000000 00:00 0
7f0f37cce000-7f0f37e91000 r-xp 00000000 103:02 2690                      /lib64/libc-2.17.so
7f0f37e91000-7f0f38090000 ---p 001c3000 103:02 2690                      /lib64/libc-2.17.so
7f0f38090000-7f0f38094000 r-xp 001c2000 103:02 2690                      /lib64/libc-2.17.so
7f0f38094000-7f0f38096000 rwxp 001c6000 103:02 2690                      /lib64/libc-2.17.so
7f0f38096000-7f0f3809b000 rwxp 00000000 00:00 0
7f0f3809b000-7f0f38110000 r-xp 00000000 103:02 3438                      /usr/lib64/libgmp.so.10.2.0
7f0f38110000-7f0f3830f000 ---p 00075000 103:02 3438                      /usr/lib64/libgmp.so.10.2.0
7f0f3830f000-7f0f38311000 rwxp 00074000 103:02 3438                      /usr/lib64/libgmp.so.10.2.0
7f0f38311000-7f0f38313000 r-xp 00000000 103:02 2696                      /lib64/libdl-2.17.so
7f0f38313000-7f0f38513000 ---p 00002000 103:02 2696                      /lib64/libdl-2.17.so
7f0f38513000-7f0f38514000 r-xp 00002000 103:02 2696                      /lib64/libdl-2.17.so
7f0f38514000-7f0f38515000 rwxp 00003000 103:02 2696                      /lib64/libdl-2.17.so
7f0f38515000-7f0f3851c000 r-xp 00000000 103:02 2720                      /lib64/librt-2.17.so
7f0f3851c000-7f0f3871b000 ---p 00007000 103:02 2720                      /lib64/librt-2.17.so
7f0f3871b000-7f0f3871c000 r-xp 00006000 103:02 2720                      /lib64/librt-2.17.so
7f0f3871c000-7f0f3871d000 rwxp 00007000 103:02 2720                      /lib64/librt-2.17.so
7f0f3871d000-7f0f38734000 r-xp 00000000 103:02 2716                      /lib64/libpthread-2.17.so
7f0f38734000-7f0f38933000 ---p 00017000 103:02 2716                      /lib64/libpthread-2.17.so
7f0f38933000-7f0f38934000 r-xp 00016000 103:02 2716                      /lib64/libpthread-2.17.so
7f0f38934000-7f0f38935000 rwxp 00017000 103:02 2716                      /lib64/libpthread-2.17.so
7f0f38935000-7f0f38939000 rwxp 00000000 00:00 0
7f0f38939000-7f0f38a3a000 r-xp 00000000 103:02 2698                      /lib64/libm-2.17.so
7f0f38a3a000-7f0f38c39000 ---p 00101000 103:02 2698                      /lib64/libm-2.17.so
7f0f38c39000-7f0f38c3a000 r-xp 00100000 103:02 2698                      /lib64/libm-2.17.so
7f0f38c3a000-7f0f38c3b000 rwxp 00101000 103:02 2698                      /lib64/libm-2.17.so
7f0f38c3b000-7f0f38c5d000 r-xp 00000000 103:02 2683                      /lib64/ld-2.17.so
7f0f38e51000-7f0f38e56000 rwxp 00000000 00:00 0
7f0f38e59000-7f0f38e5c000 rwxp 00000000 00:00 0
7f0f38e5c000-7f0f38e5d000 r-xp 00021000 103:02 2683                      /lib64/ld-2.17.so
7f0f38e5d000-7f0f38e5e000 rwxp 00022000 103:02 2683                      /lib64/ld-2.17.so
7f0f38e5e000-7f0f38e5f000 rwxp 00000000 00:00 0
7ffc599f3000-7ffc59a14000 rwxp 00000000 00:00 0                          [stack]
7ffc59a63000-7ffc59a65000 r--p 00000000 00:00 0                          [vvar]
7ffc59a65000-7ffc59a67000 r-xp 00000000 00:00 0                          [vdso]
ffffffffff600000-ffffffffff601000 r-xp 00000000 00:00 0                  [vsyscall]







[Work thread Oct 20 18:56] Timing 46080K FFT, 1 core, 1 worker.  Average times: 222.34 ms.  Total throughput:  4.50 iter/sec.
[Work thread Oct 20 18:56] Timing 46080K FFT, 1 core hyperthreaded, 1 worker.  Average times: 288.53 ms.  Total throughput:  3.47 iter/sec.
*** Error in `./mprime': free(): invalid pointer: 0x00007f71224ca3c0 ***
======= Backtrace: =========
/lib64/libc.so.6(+0x81499)[0x7f71264a3499]
./mprime[0x4582e8]
./mprime[0x45ed65]
./mprime[0x440a3a]
./mprime[0x440f9e]
./mprime[0x444bb0]
./mprime[0x444cf2]
./mprime[0x448d54]
./mprime[0x47cbca]
/lib64/libpthread.so.0(+0x7de5)[0x7f7126e78de5]
/lib64/libc.so.6(clone+0x6d)[0x7f7126520bad]
======= Memory map: ========
00400000-026a0000 r-xp 00000000 00:15 5878764499047460622                /mnt-efs/z/p95v295b2/mprime
0289f000-028a1000 r-xp 0229f000 00:15 5878764499047460622                /mnt-efs/z/p95v295b2/mprime
028a1000-028dc000 rwxp 022a1000 00:15 5878764499047460622                /mnt-efs/z/p95v295b2/mprime
028dc000-02903000 rwxp 00000000 00:00 0
02c50000-02c71000 rwxp 00000000 00:00 0                                  [heap]
7f70f9277000-7f7110000000 rwxp 00000000 00:00 0
7f7110000000-7f7110021000 rwxp 00000000 00:00 0
7f7110021000-7f7114000000 ---p 00000000 00:00 0
7f7118000000-7f711a50c000 rwxp 00000000 00:00 0
7f711a50c000-7f711c000000 ---p 00000000 00:00 0
7f711cae6000-7f711cae7000 ---p 00000000 00:00 0
7f711cae7000-7f711d2e7000 rwxp 00000000 00:00 0
7f7120000000-7f7122ab2000 rwxp 00000000 00:00 0
7f7122ab2000-7f7124000000 ---p 00000000 00:00 0
7f7124a09000-7f7124a1f000 r-xp 00000000 103:02 2617                      /lib64/libgcc_s-7-20170915.so.1
7f7124a1f000-7f7124c1e000 ---p 00016000 103:02 2617                      /lib64/libgcc_s-7-20170915.so.1
7f7124c1e000-7f7124c1f000 rwxp 00015000 103:02 2617                      /lib64/libgcc_s-7-20170915.so.1
7f7124c1f000-7f7124c20000 ---p 00000000 00:00 0
7f7124c20000-7f7125420000 rwxp 00000000 00:00 0
7f7125420000-7f7125421000 ---p 00000000 00:00 0
7f7125421000-7f7125c21000 rwxp 00000000 00:00 0
7f7125c21000-7f7125c22000 ---p 00000000 00:00 0
7f7125c22000-7f7126422000 rwxp 00000000 00:00 0
7f7126422000-7f71265e5000 r-xp 00000000 103:02 2690                      /lib64/libc-2.17.so
7f71265e5000-7f71267e4000 ---p 001c3000 103:02 2690                      /lib64/libc-2.17.so
7f71267e4000-7f71267e8000 r-xp 001c2000 103:02 2690                      /lib64/libc-2.17.so
7f71267e8000-7f71267ea000 rwxp 001c6000 103:02 2690                      /lib64/libc-2.17.so
7f71267ea000-7f71267ef000 rwxp 00000000 00:00 0
7f71267ef000-7f7126864000 r-xp 00000000 103:02 3438                      /usr/lib64/libgmp.so.10.2.0
7f7126864000-7f7126a63000 ---p 00075000 103:02 3438                      /usr/lib64/libgmp.so.10.2.0
7f7126a63000-7f7126a65000 rwxp 00074000 103:02 3438                      /usr/lib64/libgmp.so.10.2.0
7f7126a65000-7f7126a67000 r-xp 00000000 103:02 2696                      /lib64/libdl-2.17.so
7f7126a67000-7f7126c67000 ---p 00002000 103:02 2696                      /lib64/libdl-2.17.so
7f7126c67000-7f7126c68000 r-xp 00002000 103:02 2696                      /lib64/libdl-2.17.so
7f7126c68000-7f7126c69000 rwxp 00003000 103:02 2696                      /lib64/libdl-2.17.so
7f7126c69000-7f7126c70000 r-xp 00000000 103:02 2720                      /lib64/librt-2.17.so
7f7126c70000-7f7126e6f000 ---p 00007000 103:02 2720                      /lib64/librt-2.17.so
7f7126e6f000-7f7126e70000 r-xp 00006000 103:02 2720                      /lib64/librt-2.17.so
7f7126e70000-7f7126e71000 rwxp 00007000 103:02 2720                      /lib64/librt-2.17.so
7f7126e71000-7f7126e88000 r-xp 00000000 103:02 2716                      /lib64/libpthread-2.17.so
7f7126e88000-7f7127087000 ---p 00017000 103:02 2716                      /lib64/libpthread-2.17.so
7f7127087000-7f7127088000 r-xp 00016000 103:02 2716                      /lib64/libpthread-2.17.so
7f7127088000-7f7127089000 rwxp 00017000 103:02 2716                      /lib64/libpthread-2.17.so
7f7127089000-7f712708d000 rwxp 00000000 00:00 0
7f712708d000-7f712718e000 r-xp 00000000 103:02 2698                      /lib64/libm-2.17.so
7f712718e000-7f712738d000 ---p 00101000 103:02 2698                      /lib64/libm-2.17.so
7f712738d000-7f712738e000 r-xp 00100000 103:02 2698                      /lib64/libm-2.17.so
7f712738e000-7f712738f000 rwxp 00101000 103:02 2698                      /lib64/libm-2.17.so
7f712738f000-7f71273b1000 r-xp 00000000 103:02 2683                      /lib64/ld-2.17.so
7f71275a5000-7f71275aa000 rwxp 00000000 00:00 0
7f71275ad000-7f71275b0000 rwxp 00000000 00:00 0
7f71275b0000-7f71275b1000 r-xp 00021000 103:02 2683                      /lib64/ld-2.17.so
7f71275b1000-7f71275b2000 rwxp 00022000 103:02 2683                      /lib64/ld-2.17.so
7f71275b2000-7f71275b3000 rwxp 00000000 00:00 0
7ffdd4a9a000-7ffdd4abb000 rwxp 00000000 00:00 0                          [stack]
7ffdd4adc000-7ffdd4ade000 r--p 00000000 00:00 0                          [vvar]
7ffdd4ade000-7ffdd4ae0000 r-xp 00000000 00:00 0                          [vdso]
ffffffffff600000-ffffffffff601000 r-xp 00000000 00:00 0                  [vsyscall]

Last fiddled with by ATH on 2018-10-20 at 19:15
ATH is offline   Reply With Quote
Old 2018-10-20, 19:36   #8
fivemack
(loop (#_fork))
 
fivemack's Avatar
 
Feb 2006
Cambridge, England

24·3·7·19 Posts
Default

I tried running the benchmark on my i9-7940X and got very confusing behaviour : 'mpstat -P ALL 1' showed that nothing was scheduled on cores 14-27 (the other hyper threads of 0-13) when the test was running 28 workers, but 'top' indicated that it was using more than 1400% CPU.
fivemack is offline   Reply With Quote
Old 2018-10-20, 21:15   #9
Prime95
P90 years forever!
 
Prime95's Avatar
 
Aug 2002
Yeehaw, FL

22×1,873 Posts
Default

Grr. a memory corruption problem.

I did run a full benchmark, but with an earlier debug version that allocated more memory for sin/cos tables than was necessary.

I'll work on a fix and post a Linux & Windows binary.

@Fivemack: I'll check that the machine I built on has hwloc 2.0 installed.
Prime95 is offline   Reply With Quote
Old 2018-10-21, 01:25   #10
Prime95
P90 years forever!
 
Prime95's Avatar
 
Aug 2002
Yeehaw, FL

22·1,873 Posts
Default

The length 12800 and 120K FFTs were allocating too little memory. Any FFT with a large pass 1 size was zeroing several hundred bytes too many (which may or may not cause an issue based on how the allocated memory block was aligned to a 4KB boundary).

My SkyLakeX had hwloc 1.11 installed -- hopefully the newer hwloc will solve fivemack's issues.

Running a few more tests, then I'll build new executables.
Prime95 is offline   Reply With Quote
Old 2018-10-21, 04:02   #11
Prime95
P90 years forever!
 
Prime95's Avatar
 
Aug 2002
Yeehaw, FL

22·1,873 Posts
Default

29.5 build 3. Hopefully I didn't do anything stupid building these.

Linux 64-bit: https://www.dropbox.com/s/x08exhqopj...64.tar.gz?dl=0
Windows 64-bit (totally untested): https://www.dropbox.com/s/hq6bwc4tgu...win64.zip?dl=0
Prime95 is offline   Reply With Quote
Reply

Thread Tools


All times are UTC. The time now is 19:50.

Mon May 17 19:50:16 UTC 2021 up 39 days, 14:31, 0 users, load averages: 3.37, 2.75, 2.56

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, 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.