mersenneforum.org

mersenneforum.org (https://www.mersenneforum.org/index.php)
-   Software (https://www.mersenneforum.org/forumdisplay.php?f=10)
-   -   Prime95 v30.3 (https://www.mersenneforum.org/showthread.php?t=25823)

Prime95 2020-08-10 02:26

Prime95 v30.3
 
Prime95 version 30.3 build 6 is available.

From whatsnew.txt:

[CODE]1) PRP proofs. This allows GIMPS to double-check a PRP test at less than 1% of the cost of a full PRP test!
PRP proofs require lots of temporary disk space. See readme.txt for details.
PRP proofs require uploading a large proof file. See readme.txt for details.
PRP proof verifications require downloading a modest verification file. See readme.txt for details.
2) Proofs automatically uploaded to server in v30.2.
3) First time LL, World-record LL, 100M-digit LL work preference is deprecated.
4) New resource limits menu choice and dialog box. Consult readme.txt before making changes to these settings.
Some options previously in Test/Worker Windows and Options/CPU are moved to the resources dialog box.
5) LL-DC and PRP-DC combined into a single work preference.
6) Warning raised if temporary disk space is less than 1.5GB -- you may not get first time prime tests.
7) Thanks to Mihai Preda, the P-1 probability calculator has been improved. This change results in a
lower optimal B1 value and higher optimal B2 value.
[/CODE]


Download links:
Windows 64-bit: [URL]https://mersenne.org/ftp_root/gimps/p95v303b6.win64.zip[/URL]
Linux 64-bit: [URL]https://mersenne.org/ftp_root/gimps/p95v303b6.linux64.tar.gz[/URL]
FreeBSD11 64-bit: [URL]https://mersenne.org/ftp_root/gimps/p95v303b6.FreeBSD11-64.tar.gz[/URL]
Source: [URL]https://mersenne.org/ftp_root/gimps/p95v303b6.source.zip[/URL]

Windows Service 64-bit: [URL]https://mersenne.org/ftp_root/gimps/p95v303b6.win64.service.zip[/URL]
Linux 32-bit: [URL]https://mersenne.org/ftp_root/gimps/p95v303b6.linux32.tar.gz[/URL]
Windows 32-bit: [URL]https://mersenne.org/ftp_root/gimps/p95v303b6.win32.zip[/URL]
Windows Service 32-bit: [URL]https://mersenne.org/ftp_root/gimps/p95v303b6.win32.service.zip[/URL]

Prime95 2020-08-10 02:27

Placeholder for bugs fixes

1) If a proof is not being generated prime95 erroneously tries to which gives a file not found error. Fixed in 30.3 build 2.
2) An error getting the CERT starting value loops forever trying to get the value from the server. Semi-fixed in 30.3 build 2 -- now tries once an hour.
3) PRP test on a big exponent that would benefit from trial factoring was not getting the needed TF. Fixed in 30.3 build 2 -- code copied from the LL code path.
4) Most PRP cofactor proof files will not upload. Fixed in 30.3 build 2
5) Adding PRP=1,2,exponent,-1 defaulted to "TF needed" and "no P-1 needed". This is inconsistent. The new default will be "no TF needed" and "no P-1 needed". Fixed in 30.3 build 3.
6) Debug level 2 in PrimeNet section of prime.txt outputs tons of binary proof data. Fixed in 30.3 build 3 by outputting <binary> instead.
7) Certification error message "MD5 checksum mismatch" can be misleading. Fixed in 30.3 build 4.
8) If SequentialWorkTodo=0 and you are doing PRP work, then your jobs will mistakenly be interrupted for priority work. Fixed in 30.3 build 4.
9) The setting for maximum CERT CPU percentage not working properly. Fixed in 30.3 build 4.
10) An error in pre-allocating the interim residues file disk space improperly increased the proof power rather than decreased the proof power. Fixed in 30.3 build 6.
11) Wrong hwloc DLL included for win32. Fixed in build 6.
12) Crashes during load on Vista 64-bit and earlier with K32GetModuleFileNameExA not found. Possibly fixed in 30.3 build 6.

Prime95 2020-08-10 02:34

New in v30.3 compared to 30.2
 
The DLL for libgmp is not compatible with the previous one. This was causing issues for some users (finding the wrong version of the DLL). In this release, the DLL has been renamed to libgmp-gw1.dll.

For those with limited disk space, this version can simulate a higher proof power using less disk space but a larger proof file to upload. For example we can simulate a power = 8 proof by doing a power=7 proof on the first half of the PRP test and a second power=7
proof on the second half of a PRP test. This uses half the temporary disk space and twice the upload bandwidth.

LL-DC and PRP-DC are combined into a single work preference.

Error detection/recovery in prime95 is much more robust.

pepi37 2020-08-10 07:23

I try it on sequence 95*10^n-1


[QUOTE][Aug 10 09:20:55] Waiting 5 seconds to stagger worker starts.
[Aug 10 09:21:00] Worker starting
[Aug 10 09:21:00] Setting affinity to run worker on CPU core #4
[Aug 10 09:21:00] Setting affinity to run helper thread 1 on CPU core #5
[Aug 10 09:21:00] Setting affinity to run helper thread 2 on CPU core #6
[Aug 10 09:21:08] Resuming PRP test of 95*10^1809228-1 using FMA3 FFT length 448K, Pass1=448, Pass2=1K, clm=4, 3 threads
[Aug 10 09:21:08] Iteration: 4072948 / 6010131 [67.76%].
[B][Aug 10 09:22:34] Generating proof for 95*10^1809228-1. Proof power = 0, Hash length = 64
[Aug 10 09:22:34] Cannot open PRP proof interim residues file:
[Aug 10 09:22:34] Errno: 2, No such file or directory
[Aug 10 09:22:34] DOSerrno: 3
[Aug 10 09:22:34] Waiting 5 minutes to try proof generation again.
[Aug 10 09:22:34] Waiting five minutes before restarting.[/B][/QUOTE]

pinhodecarlos 2020-08-10 10:01

Do you need a proof for your manual work outside GIMPS boundary?

Prime95 2020-08-10 15:14

Proofs only work for base 2 numbers. I'll work on a fix for the bug you found.

kriesel 2020-08-10 16:47

re prime95/mprime v30.3:[QUOTE=Prime95;553087]
Error detection/recovery in prime95 is much more robust.[/QUOTE]Always good. In what ways was it improved?
(Did GEC get implemented for the CERT runs?)

Prime95 2020-08-10 17:16

[QUOTE=kriesel;553139]re prime95/mprime v30.3:Always good. In what ways was it improved?
(Did GEC get implemented for the CERT runs?)[/QUOTE]

GEC for CERT -- no.

Interim residues are now written with an MD5 checksum.

Entire proof file is read back after being written to compare MD5 checksums.

Numerous cases of sloppy error handling fixed. I code proof-of-concept with sloppy error handling first. Then I go back and clean up.

ATH 2020-08-10 18:54

1 Attachment(s)
Error on trying to upload a proof file for a very small PRP CF in mprime on EC2:

"Error getting number from proof header"

Edit: The file is still not uploaded. Trying to attach it here.


[CODE]
[Comm thread Aug 10 14:32:19] Got assignment 9B32734729207C3C235DD65DBD241B7A: PRP M2344571
[Work thread Aug 10 14:32:19] Resuming.
[Comm thread Aug 10 14:32:19] Sending expected completion date for M2344571: Aug 10 2020
[Work thread Aug 10 14:32:19] Starting Gerbicz error-checking PRP test of M2344571/5123586391565541063572897 using AVX-512 FFT length 120K, Pass1=192, Pass2=640, clm=1
[Work thread Aug 10 14:32:19] Preallocating disk space for the proof interim residues file p2344571.residues
[Work thread Aug 10 14:32:19] PRP proof using power=7 and 64-bit hash size.
[Work thread Aug 10 14:32:19] Proof requires 0.0GB of temporary disk space and uploading a 2MB proof file.
[Comm thread Aug 10 14:32:20] Done communicating with server.
[Work thread Aug 10 14:36:30] Iteration: 1000000 / 2344576 [42.651635%], roundoff: 0.101, ms/iter: 0.249, ETA: 00:05:34
[Work thread Aug 10 14:36:30] Gerbicz error check passed at iteration 1000000.
[Work thread Aug 10 14:36:30] M2344571/5123586391565541063572897 interim PRP residue A3E2916C933308D5 at iteration 1000000
[Work thread Aug 10 14:40:41] Iteration: 2000000 / 2344576 [85.303270%], roundoff: 0.110, ms/iter: 0.249, ETA: 00:01:25
[Work thread Aug 10 14:40:41] Gerbicz error check passed at iteration 2000000.
[Work thread Aug 10 14:40:41] M2344571/5123586391565541063572897 interim PRP residue 5A3CD751E40E3CA4 at iteration 2000000
[Work thread Aug 10 14:42:08] Gerbicz error check passed at iteration 2344569.
[Work thread Aug 10 14:42:13] Generating proof for M2344571. Proof power = 7, Hash length = 64
[Work thread Aug 10 14:42:13] Root hash = 616605A749A0C958D8DF3E72FC32C804BBD2B1288AA709C47FE65853A297850D
[Work thread Aug 10 14:42:13] hash0 = 1ACBF839AA5DF27F
[Work thread Aug 10 14:42:13] hash1 = 782DBDD6964A10A1
[Work thread Aug 10 14:42:13] hash2 = D1EC5C0BEAC72ADA
[Work thread Aug 10 14:42:14] hash3 = 49AB07B48B882843
[Work thread Aug 10 14:42:14] hash4 = 0055006F7821EA7E
[Work thread Aug 10 14:42:15] hash5 = E7ADE8B93091DF2B
[Work thread Aug 10 14:42:17] Proof construction cost 9348 squarings
[Work thread Aug 10 14:42:17] Proof verification will cost 18317 squarings
[Work thread Aug 10 14:42:17] M2344571/5123586391565541063572897 is not prime. Type-5 RES64: CD8AD0D724D8D9__. Wh8: 396B1DA3,1632460,00000000
[Comm thread Aug 10 14:42:17] Sending result to server: UID: athath/ec2-PRPCF, M2344571/5123586391565541063572897 is not prime. Type-5 RES64: CD8AD0D724D8D9__. Wh8: 396B1DA3,1632460,00000000, AID: 9B32734729207C3C235DD65DBD241B7A
[Comm thread Aug 10 14:42:17]



[Comm thread Aug 10 14:54:13] MD5 of p2344571.proof is 015d98c962f822406773acc18ae0933d
[Comm thread Aug 10 14:54:13] Error getting number from proof header
[Comm thread Aug 10 15:59:13] MD5 of p2344571.proof is 015d98c962f822406773acc18ae0933d
[Comm thread Aug 10 15:59:13] Error getting number from proof header



[Comm thread Aug 10 17:04:13] MD5 of p2344571.proof is 015d98c962f822406773acc18ae0933d
[Comm thread Aug 10 17:04:13] Error getting number from proof header
[Comm thread Aug 10 18:09:13] MD5 of p2344571.proof is 015d98c962f822406773acc18ae0933d
[Comm thread Aug 10 18:09:13] Error getting number from proof header


[/CODE]

Prime95 2020-08-10 23:30

[QUOTE=ATH;553152]Error on trying to upload a proof file for a very small PRP CF [/QUOTE]

Will have a fix tonight.

Prime95 2020-08-11 00:59

Please upgrade to 30.3 build 2.

Two of the reported bugs were serious:
1) PRPs without proofs (say, if you upgraded from v29) will fail at the the end of the PRP test by trying to create a proof file.
2) Inability to upload proofs from cofactor PRP tests.

ATH 2020-08-11 02:33

Ok 30.3b2 worked to upload the proof.

Happy5214 2020-08-11 12:42

I had an error when trying to certify a PRP-CF proof (v30.3b2, Ubuntu 20.04):

[code][Worker #1 Aug 11 07:40] Starting certification of M8608507 using FFT length 448K, Pass1=448, Pass2=1K, clm=4
[Comm thread Aug 11 07:40] CURL library error:
[Comm thread Aug 11 07:40] CURL library error:
[Worker #1 Aug 11 07:40] Error getting CERT starting value.
[Worker #1 Aug 11 07:40] Aborting processing of this work unit -- will try again later.
[/code]

Aramis Wyler 2020-08-12 03:31

Running mprime for the first time on a new Ryzen 5 3600.


I used mostly defaults - 2 workers and 3 cores each - with work type 150 (First time Prime checks).


I'm posting because I'm getting an enormous number of potential round off errors on each worker. The build is new, the cpu could be defective, but I haven't seen any errors or heat issues other than these roundoff errors.


[C][Worker #2 Aug 11 23:22] Setting affinity to run helper thread 2 on CPU core #6
[Worker #2 Aug 11 23:22] M110534549 stage 1 is 1.17% complete.
[Worker #2 Aug 11 23:23] Possible roundoff error (0.5), backtracking to last save file.
[Worker #2 Aug 11 23:23] Setting affinity to run helper thread 1 on CPU core #5
[Worker #2 Aug 11 23:23] Using FMA3 FFT length 6M, Pass1=1536, Pass2=4K, clm=1, 3 threads
[Worker #2 Aug 11 23:23] Setting affinity to run helper thread 2 on CPU core #6
[Worker #2 Aug 11 23:23] M110534549 stage 1 is 1.22% complete.
[Worker #1 Aug 11 23:23] M110534311 stage 1 is 1.23% complete. Time: 112.608 sec.
[Worker #1 Aug 11 23:24] Possible roundoff error (0.5), backtracking to last save file.
[Worker #1 Aug 11 23:24] Setting affinity to run helper thread 1 on CPU core #2
[Worker #1 Aug 11 23:24] Using FMA3 FFT length 6M, Pass1=1536, Pass2=4K, clm=1, 3 threads
[Worker #1 Aug 11 23:24] Setting affinity to run helper thread 2 on CPU core #3
[Worker #1 Aug 11 23:24] M110534311 stage 1 is 0.14% complete.
[Worker #2 Aug 11 23:25] Possible roundoff error (0.5), backtracking to last save file.
[Worker #2 Aug 11 23:25] Setting affinity to run helper thread 1 on CPU core #5
[Worker #2 Aug 11 23:25] Setting affinity to run helper thread 2 on CPU core #6
[Worker #2 Aug 11 23:25] Using FMA3 FFT length 6M, Pass1=1536, Pass2=4K, clm=1, 3 threads
[Worker #2 Aug 11 23:25] M110534549 stage 1 is 1.40% complete.
[Worker #2 Aug 11 23:25] Possible roundoff error (0.5), backtracking to last save file.
[Worker #2 Aug 11 23:25] Setting affinity to run helper thread 2 on CPU core #6
[Worker #2 Aug 11 23:25] Setting affinity to run helper thread 1 on CPU core #5
[Worker #2 Aug 11 23:25] Using FMA3 FFT length 6M, Pass1=1536, Pass2=4K, clm=1, 3 threads
[Worker #2 Aug 11 23:25] M110534549 stage 1 is 1.40% complete.
[Worker #1 Aug 11 23:25] M110534311 stage 1 is 0.74% complete. Time: 112.937 sec.
[Worker #1 Aug 11 23:26] Possible roundoff error (0.5), backtracking to last save file.
[Worker #1 Aug 11 23:26] Using FMA3 FFT length 6M, Pass1=1536, Pass2=4K, clm=1, 3 threads
[Worker #1 Aug 11 23:26] Setting affinity to run helper thread 2 on CPU core #3
[Worker #1 Aug 11 23:26] Setting affinity to run helper thread 1 on CPU core #2
[Worker #1 Aug 11 23:26] M110534311 stage 1 is 0.14% complete. [/C]


EDIT: This is with v30.3 build 2 on 64 bit debian.

intelfx 2020-08-12 07:02

[QUOTE=Happy5214;553273]I had an error when trying to certify a PRP-CF proof (v30.3b2, Ubuntu 20.04):

[code][Worker #1 Aug 11 07:40] Starting certification of M8608507 using FFT length 448K, Pass1=448, Pass2=1K, clm=4
[Comm thread Aug 11 07:40] CURL library error:
[Comm thread Aug 11 07:40] CURL library error:
[Worker #1 Aug 11 07:40] Error getting CERT starting value.
[Worker #1 Aug 11 07:40] Aborting processing of this work unit -- will try again later.
[/code][/QUOTE]
Same here.


Two issues:
[LIST=1][*]When I first started mprime today, mprime reported that it got CERT work, but then proceeded to work on previous assignments:
[CODE]
Aug 12 07:28:32 stratofortress.nexus.i.intelfx.name mprime[15264]: [Comm thread Aug 12 07:28] Sending expected completion date for M110701609: Aug 23 2020
Aug 12 07:28:33 stratofortress.nexus.i.intelfx.name mprime[15264]: [Work thread Aug 12 07:28] Running Jacobi error check. [Aug 12 07:28] PrimeNet success code with additional info:
Aug 12 07:28:33 stratofortress.nexus.i.intelfx.name mprime[15264]: [Comm thread Aug 12 07:28] Server assigned CERT work.
Aug 12 07:28:33 stratofortress.nexus.i.intelfx.name mprime[15264]: [Comm thread Aug 12 07:28] Got assignment 005E9C2063038514CF0D0DD5E4DCFCAE: CERT M10447057
Aug 12 07:28:33 stratofortress.nexus.i.intelfx.name mprime[15264]: [Comm thread Aug 12 07:28] Done communicating with server.
Aug 12 07:28:58 stratofortress.nexus.i.intelfx.name mprime[15264]: [Work thread] Passed. Time: 25.646 sec.
Aug 12 07:28:58 stratofortress.nexus.i.intelfx.name mprime[15264]: [Work thread Aug 12 07:28] Resuming primality test of M109983959 using FMA3 FFT length 6M, Pass1=1536, Pass2=4K, clm=1, 16 threads
[/CODE][*]When I manually edited my [c]worktodo.txt[/c] to place the cert work in front of the queue, mprime entered a failure loop:
[code]
Aug 12 07:30:06 stratofortress.nexus.i.intelfx.name mprime[16107]: [Work thread Aug 12 07:30] Starting certification of M10447057 using FMA3 FFT length 560K, Pass1=448, Pass2=1280, clm=1, 16 threads
Aug 12 07:30:06 stratofortress.nexus.i.intelfx.name mprime[16107]: [Comm thread Aug 12 07:30] CURL library error:
Aug 12 07:30:06 stratofortress.nexus.i.intelfx.name mprime[16107]: [Comm thread Aug 12 07:30] CURL library error:
Aug 12 07:30:06 stratofortress.nexus.i.intelfx.name mprime[16107]: [Work thread Aug 12 07:30] Error getting CERT starting value. Will try again later.
Aug 12 07:30:06 stratofortress.nexus.i.intelfx.name mprime[16107]: [Work thread Aug 12 07:30] Aborting processing of this work unit.
[/code](mprime went on to repeat these messages indefinitely)[/LIST]Are those bugs, server problems or misconfigurations on my part?


[B]Edit:[/B] when I subsequently edited [c]worktodo.txt[/c] to put the CERT assignment in the back of the queue and restarted mprime, it still attempted to pick up the CERT assignment, despite it was at the end of the queue (which suggests priority behavior for CERT work). Hence I conclude that (1) is a bug.

Prime95 2020-08-12 10:35

Due to a server issue, Linux clients can neither get the CERT starting value, nor upload proofs. Aaron or I will have a fix today.

Prime95 2020-08-12 10:38

[QUOTE=Aramis Wyler;553383]Running mprime for the first time on a new Ryzen 5 3600.

I used mostly defaults - 2 workers and 3 cores each - with work type 150 (First time Prime checks).

I'm posting because I'm getting an enormous number of potential round off errors on each worker. The build is new, the cpu could be defective, but I haven't seen any errors or heat issues other than these roundoff errors.[/QUOTE]

Hardware issues. Do the standard remedies, try lowering memory frequencies, or CPU speed, or increase voltages. Find a combination that can pass the torture test.

S485122 2020-08-12 19:03

Updated the software to the latest version.

Received a Cert work unit ... to do some certifying of a cofactor for the factored Mersenne number 10482449. The configured work preference is double checking primality testing.

Working with the configuration I will be spared from cofactor certifying (and any other CERT jobs.) When viewing the result of the cert work done : "n/a" one has to go to the status of the number [url]https://www.mersenne.org/report_exponent/?exp_lo=10482449&full=1[/url] to see that the cofactor work has been certified OK (Verified). But it is not clear what the the status of the exponent is : fully factored ? Anyway that type of work (cofactors) is absolutely not something I signed up for. The imposing of that kind of work is at ends with the "let the user decide" philosophy of the project.

There obviously remains a wee bit of tuning to do on PrimeNet.

Jacob

Prime95 2020-08-12 19:39

[QUOTE=S485122;553468]Updated the software to the latest version.

Received a Cert work unit ... to do some certifying of a cofactor for the factored Mersenne number 10482449. The configured work preference is double checking primality testing.

Working with the configuration I will be spared from cofactor certifying (and any other CERT jobs.) When viewing the result of the cert work done : "n/a" one has to go to the status of the number [url]https://www.mersenne.org/report_exponent/?exp_lo=10482449&full=1[/url] to see that the cofactor work has been certified OK (Verified). But it is not clear what the the status of the exponent is : fully factored ? Anyway that type of work (cofactors) is absolutely not something I signed up for. The imposing of that kind of work is at ends with the "let the user decide" philosophy of the project.

There obviously remains a wee bit of tuning to do on PrimeNet.[/QUOTE]

CERT for PRP-CF is trivially quick work. I'm not sure why you found it so distasteful.

CERT for PRP is really a kind of PRP-DC. It is not a separate work preference choice as the server does not have much of that work type to hand out.

I'm glad you figured out how to disable CERT work. Kriesel also disabled CERT work because of the impact on his LL testing -- a Jacobi check to save his LL test and another Jacobi check on resume. I'm thinking the ability to turn off CERT work needs to be more prominent -- perhaps a checkbox at the bottom of the Worker Windows dialog box.

I agree, the server web pages need a lot of work due to proofs.

kriesel 2020-08-12 20:52

To clarify, I set download rate to 0 on prime95 on most of my systems but not all, to a point where I think I'll be doing my fair share. Doing an order of magnitude more CERTs than I do primality tests was a small drag on throughput/efficiency and made my testing throughput unpredictable. There were others that were interested in doing more CERTs than they were being assigned. So throttling my CERT throughput down considerably from the initial disparity created a win-win. And I am appreciative of those who are doing CERTs on my PRP or PRPDC in 120M-200M. These runs are to possibly detect any issues with fft length cutoffs etc, well ahead of the wavefront. ([URL]https://www.mersenneforum.org/showpost.php?p=501181&postcount=6;[/URL] similar lower priority effort with LL/LLDC at [URL]https://www.mersenneforum.org/showpost.php?p=501178&postcount=4[/URL])

GIMPS is going through a complicated transition currently, and more rapidly it seems than originally projected. Software bugs are being identified and dealt with, in server and client code. Good bug reports, and patience, are recommended.

It will take a long time to get the bulk of the clients updated. Early adopters of prime95/mprime v30.x are bearing the brunt of CERT for both mprime/prime95 and gpuowl production. (Either curtisc or Ben Delo updating a fraction of their fleet would help a lot. But like for everyone in this all-volunteer project, their kit, their call. And if they had started already, we wouldn't know without doing some checking.)

ATH 2020-08-12 22:22

[QUOTE=Prime95;553471]CERT for PRP-CF is trivially quick work. I'm not sure why you found it so distasteful.[/QUOTE]

CERT for PRP-CF for a 10.48M exponent took 18 sec on 8 cores, so 2.5 min tops if running it on a single core, and maybe 3-5 minutes at most if you have a very old cpu and running it on 1 core.

Xyzzy 2020-08-12 23:03

Is there a worktype option to select proof work?

Uncwilly 2020-08-12 23:24

[QUOTE=Xyzzy;553483]Is there a worktype option to select proof work?[/QUOTE]Select DC or PRP-CF-DC and it is a freebie that you get. There are not enough being generated to keep the Cert runners flush with work. If you want to generate the proofs, just do first time checks with the latest version.

intelfx 2020-08-13 04:11

Experienced the priority issue again:


[code]
Aug 13 07:06:17 stratofortress.nexus.i.intelfx.name mprime[69751]: [Work thread Aug 13 07:06] Running Jacobi error check. [Aug 13 07:06] Sending expected completion date for M110701609: Aug 24 2020
Aug 13 07:06:18 stratofortress.nexus.i.intelfx.name mprime[69751]: [Comm thread Aug 13 07:06] PrimeNet success code with additional info:
Aug 13 07:06:18 stratofortress.nexus.i.intelfx.name mprime[69751]: [Comm thread Aug 13 07:06] Server assigned CERT work.
Aug 13 07:06:18 stratofortress.nexus.i.intelfx.name mprime[69751]: [Comm thread Aug 13 07:06] Got assignment 55203D43C36D845B45BCF307EAB00E67: CERT M10445371
Aug 13 07:06:18 stratofortress.nexus.i.intelfx.name mprime[69751]: [Comm thread Aug 13 07:06] Done communicating with server.
Aug 13 07:06:41 stratofortress.nexus.i.intelfx.name mprime[69751]: [Work thread] Passed. Time: 24.635 sec.
Aug 13 07:06:41 stratofortress.nexus.i.intelfx.name mprime[69751]: [Work thread Aug 13 07:06] Resuming primality test of M109983959 using FMA3 FFT length 6M, Pass1=1536, Pass2=4K, clm=1, 16 threads
Aug 13 07:06:41 stratofortress.nexus.i.intelfx.name mprime[69751]: [Work thread Aug 13 07:06] Iteration: 62704259 / 109983959 [57.01%].
Aug 13 07:06:58 stratofortress.nexus.i.intelfx.name mprime[69751]: [Work thread Aug 13 07:06] Iteration: 62710000 / 109983959 [57.01%], ms/iter: 2.909, ETA: 38:11:38
Aug 13 07:07:22 stratofortress.nexus.i.intelfx.name systemd[1]: Stopping mprime.service...
Aug 13 07:07:22 stratofortress.nexus.i.intelfx.name mprime[69751]: [Main thread Aug 13 07:07] Stopping all worker windows.
Aug 13 07:07:22 stratofortress.nexus.i.intelfx.name mprime[69751]: [Work thread Aug 13 07:07] Stopping primality test of M109983959 at iteration 62718308 [57.02%]
Aug 13 07:07:22 stratofortress.nexus.i.intelfx.name mprime[69751]: [Work thread Aug 13 07:07] Worker stopped.
Aug 13 07:07:22 stratofortress.nexus.i.intelfx.name mprime[69751]: [Main thread Aug 13 07:07] Execution halted.
Aug 13 07:07:22 stratofortress.nexus.i.intelfx.name mprime[69751]: [Main thread Aug 13 07:07] Choose Test/Continue to restart.
Aug 13 07:07:22 stratofortress.nexus.i.intelfx.name systemd[1]: mprime.service: Succeeded.
Aug 13 07:07:22 stratofortress.nexus.i.intelfx.name systemd[1]: Stopped mprime.service.
Aug 13 07:07:22 stratofortress.nexus.i.intelfx.name systemd[1]: mprime.service: Consumed 10min 46.312s CPU time.

Aug 13 07:07:22 stratofortress.nexus.i.intelfx.name systemd[1]: Started mprime.service.
Aug 13 07:07:22 stratofortress.nexus.i.intelfx.name mprime[70335]: [Main thread Aug 13 07:07] Mersenne number primality test program version 30.3
Aug 13 07:07:22 stratofortress.nexus.i.intelfx.name mprime[70335]: [Main thread Aug 13 07:07] Optimizing for CPU architecture: AMD Zen, L2 cache size: 16x512 KB, L3 cache size: 4x16 MB
Aug 13 07:07:22 stratofortress.nexus.i.intelfx.name mprime[70335]: [Main thread Aug 13 07:07] Starting worker.
Aug 13 07:07:23 stratofortress.nexus.i.intelfx.name mprime[70335]: [Work thread Aug 13 07:07] Worker starting
Aug 13 07:07:23 stratofortress.nexus.i.intelfx.name mprime[70335]: [Work thread Aug 13 07:07] Setting affinity to run worker on CPU core #1
Aug 13 07:07:23 stratofortress.nexus.i.intelfx.name mprime[70335]: [Work thread Aug 13 07:07] Setting affinity to run helper thread 1 on CPU core #2
Aug 13 07:07:23 stratofortress.nexus.i.intelfx.name mprime[70335]: [Work thread Aug 13 07:07] Setting affinity to run helper thread 3 on CPU core #4
Aug 13 07:07:23 stratofortress.nexus.i.intelfx.name mprime[70335]: [Work thread Aug 13 07:07] Setting affinity to run helper thread 4 on CPU core #5
Aug 13 07:07:23 stratofortress.nexus.i.intelfx.name mprime[70335]: [Work thread Aug 13 07:07] Setting affinity to run helper thread 5 on CPU core #6
Aug 13 07:07:23 stratofortress.nexus.i.intelfx.name mprime[70335]: [Work thread Aug 13 07:07] Setting affinity to run helper thread 6 on CPU core #7
Aug 13 07:07:23 stratofortress.nexus.i.intelfx.name mprime[70335]: [Work thread Aug 13 07:07] Setting affinity to run helper thread 7 on CPU core #8
Aug 13 07:07:23 stratofortress.nexus.i.intelfx.name mprime[70335]: [Work thread Aug 13 07:07] Setting affinity to run helper thread 8 on CPU core #9
Aug 13 07:07:23 stratofortress.nexus.i.intelfx.name mprime[70335]: [Work thread Aug 13 07:07] Setting affinity to run helper thread 9 on CPU core #10
Aug 13 07:07:23 stratofortress.nexus.i.intelfx.name mprime[70335]: [Work thread Aug 13 07:07] Setting affinity to run helper thread 10 on CPU core #11
Aug 13 07:07:23 stratofortress.nexus.i.intelfx.name mprime[70335]: [Work thread Aug 13 07:07] Setting affinity to run helper thread 11 on CPU core #12
Aug 13 07:07:23 stratofortress.nexus.i.intelfx.name mprime[70335]: [Work thread Aug 13 07:07] Setting affinity to run helper thread 12 on CPU core #13
Aug 13 07:07:23 stratofortress.nexus.i.intelfx.name mprime[70335]: [Work thread Aug 13 07:07] Setting affinity to run helper thread 13 on CPU core #14
Aug 13 07:07:23 stratofortress.nexus.i.intelfx.name mprime[70335]: [Work thread Aug 13 07:07] Setting affinity to run helper thread 14 on CPU core #15
Aug 13 07:07:23 stratofortress.nexus.i.intelfx.name mprime[70335]: [Work thread Aug 13 07:07] Setting affinity to run helper thread 15 on CPU core #16
Aug 13 07:07:23 stratofortress.nexus.i.intelfx.name mprime[70335]: [Work thread Aug 13 07:07] Setting affinity to run helper thread 2 on CPU core #3
Aug 13 07:07:23 stratofortress.nexus.i.intelfx.name mprime[70335]: [Work thread Aug 13 07:07] Starting certification of M10445371 using FMA3 FFT length 560K, Pass1=448, Pass2=1280, clm=1, 16 threads
[/code]
That is: mprime does not attempt to work on the newly received CERT assignment unless manually restarted (at which point the CERT assignment(s) are performed first regardless of their position in worktodo.txt, which suggests priority behavior.)


Would like to know if this is a bug or intended behavior.

Prime95 2020-08-13 04:59

[QUOTE=intelfx;553495]
That is: mprime does not attempt to work on the newly received CERT assignment unless manually restarted (at which point the CERT assignment(s) are performed first regardless of their position in worktodo.txt, which suggests priority behavior.)

Would like to know if this is a bug or intended behavior.[/QUOTE]

I believe mprime checks for priority work once an hour.

kriesel 2020-08-13 12:49

[CODE][Tue Aug 11 15:57:12 2020]
Unexpected error during upload: {"error_status":401,"error_message":"Unauthorized","error_description":""}[/CODE]It would be good if it told in that case, what exponent's proof file it was trying to upload at the time, such as by giving the file name: "Unexpected error during upload of file nnnnnnnn-l.proof:". (Some of us have prime95 instances with multiple workers. Some of us may be using prime95 to upload proofs from gpuowl.)

Also, I do not see any subsequent log entry of successful upload, in prime.log, results.txt, or results.json.txt. I think this was for a gpuowl proof file for [M]138000061[/M].

Uncwilly 2020-08-13 13:52

I just upgraded a machine to v30.3 (from 29) and it started to do TF on some of the PRP-CF's that were queued up.

Prime95 2020-08-13 15:33

[QUOTE=Uncwilly;553541]I just upgraded a machine to v30.3 (from 29) and it started to do TF on some of the PRP-CF's that were queued up.[/QUOTE]

Odd. What did the worktodo.txt lines look like?

In the meantime, put "SkipTrialFactoring=1" in prime.txt

Uncwilly 2020-08-13 15:50

They look like this now:
[CODE]PRP=AID,1,2,10445909,-1,64,0,"1158749967781922153929"
PRP=AID,1,2,10447607,-1,64,0,"709324882075208391407"
PRP=AID,1,2,3360769,-1,99,0,3,5,"7420577953,1976985451084487,121365941392574291033"
PRP=AID,1,2,2902771,-1,59,0,"17643545420524016638023623"
PRP=AID,1,2,10459489,-1,64,0,"1082825031643242155311"
PRP=AID,1,2,4713857,-1,99,0,3,5,"141415711,2441360721372359,121781663060087391769"
PRP=AID,1,2,5300761,-1,99,0,3,5,"22263196201,121673322749022149191"
PRP=AID,1,2,5453839,-1,99,0,3,5,"261784273,121728904184791573927"
PRP=AID,1,2,5706013,-1,99,0,"19480328383,296438787377,5191111836037529,12979380910552501759,121716143484835542329"
PRP=AID,1,2,5943127,-1,99,0,"7630067053284433,121645664538289521281"
[/CODE]
Before hand I think some did not have the 99 (or other number) in place. It did not run those that have the 99. But it did do the exponents that had other numbers up to the normal P95 limit. It also seemed to have dropped the assignments where it found factors.

Yeah for beta!!!

And BTW, except Windows "Defender", the corporate antivirus and Symantec/Norton Insight did not squawk about the files from mersenne.org like they did from dropbox.

storm5510 2020-08-13 15:53

[QUOTE=ATH;553481]CERT for PRP-CF for a 10.48M exponent took 18 sec on 8 cores, so 2.5 min tops if running it on a single core, and maybe 3-5 minutes at most if you have a very old cpu and running it on 1 core.[/QUOTE]

A person will not always get a certification for a cofactor if they are running the same. I ran a certify yesterday, 166-million and change, on my i5 which took 5 hours to complete. I run [U]only[/U] cofactor tests. I do not have the CPU horsepower many others here have. I have no qualms about running them

About proof file names. I have noticed most are prefixed with "pA" while a few others are just "p." Both carry the [I].proof[/I] extension. There must be a difference. When [I]Prime95[/I] sends a certification with "pA" in front, a text line appears indicating the transfer was successful. This does not happen with the latter. They also assume the same file title as the backup files.

Since George straightened me out, I have had no problems.


[U]Edit:[/U] Correction. Prime95 does give details about the sending of proof files with the "p" prefix.

kriesel 2020-08-13 18:09

[QUOTE=storm5510;553555]I ran a certify yesterday, 166-million and change, on my i5 which took 5 hours to complete.[/QUOTE]([M]166000013[/M] Thank you!)

Prime95 2020-08-13 21:02

[QUOTE=Uncwilly;553554]They look like this now:
[CODE]PRP=AID,1,2,10445909,-1,64,0,"1158749967781922153929"
PRP=AID,1,2,10447607,-1,64,0,"709324882075208391407"
PRP=AID,1,2,3360769,-1,99,0,3,5,"7420577953,1976985451084487,121365941392574291033"
PRP=AID,1,2,2902771,-1,59,0,"17643545420524016638023623"
PRP=AID,1,2,10459489,-1,64,0,"1082825031643242155311"
PRP=AID,1,2,4713857,-1,99,0,3,5,"141415711,2441360721372359,121781663060087391769"
PRP=AID,1,2,5300761,-1,99,0,3,5,"22263196201,121673322749022149191"
PRP=AID,1,2,5453839,-1,99,0,3,5,"261784273,121728904184791573927"
PRP=AID,1,2,5706013,-1,99,0,"19480328383,296438787377,5191111836037529,12979380910552501759,121716143484835542329"
PRP=AID,1,2,5943127,-1,99,0,"7630067053284433,121645664538289521281"
[/CODE][/QUOTE]

How did you get these assignments? From prime95 contacting PrimeNet, or manually, or copied from [url]http://mersenne.org/workload [/url]?

For PRP-CF the how far factored field should always be 99. Certainly it should be at least the TJAOI minimum of 65 or 66.

Uncwilly 2020-08-13 21:10

[QUOTE=Prime95;553579]How did you get these assignments? From prime95 contacting PrimeNet, or manually, or copied from [url]http://mersenne.org/workload?[/url][/QUOTE]Those came from through Prime95 contacting PrimeNet in the last few days. (I was away from the machine at the time. That was before I upgraded the machine to v30 at all. I looked at some where my machine found factors and TJAOI found some in the last few hours, so, my assignment was outdated.

Prime95 2020-08-13 21:14

[QUOTE=storm5510;553555]A person will not always get a certification for a cofactor if they are running the same. I ran a certify yesterday, 166-million and change, on my i5 which took 5 hours to complete. I run [U]only[/U] cofactor tests. I do not have the CPU horsepower many others here have. I have no qualms about running them.[/QUOTE]


I've worried about how to assign CERT exponents. There are three "levels": PRP-CF exponents at roughly 10,000,000. Wavefront at about 100,000,000, and EFF wannabees at 333,000,000.

1) It does not seem right to assign a 333,000,000 CERT to a slow computer doing PRP-CF, ECM, or some such.
2) I don't mind doing CERTs for PRP-CF, but at least one user does.
3) Interrupting LL is more costly than interrupting PRP. In some respects, the CERT exponent ought to be large enough to justify the expense.

I had planned on using the "Certification work limit (% of CPU time)" in resource limits to control assignments, but this may not be the most intuitive (or easily located) solution

Perhaps the current work preference can also play a role in CERT assignments.

Just thinking out loud here. Right now it is a "free for all" on CERT assignments. I'll come up with something sooner or later.

Mark Rose 2020-08-13 21:27

Why not make it an opt-in/out checkbox, like getting low assignments?

storm5510 2020-08-14 00:01

As I understand it, certification files are sent on timed intervals. If a person is actually running, then this is fine. What if a person stops the process? I had to stop earlier today because a nasty cluster of thunderstorms moved through my area. I [U]do not[/U] run anything with lightning dancing around.

I restarted four hours later. There was a certification file on one machine which did not get sent before I stopped. GUI programs, like [I]Prime95[/I], are "event driven." They require a user action to affect a change in behavior. From my antique programming days, I know GUI applications are basically a collection of functions and sub-programs.

What I would like to suggest is a "manual trigger" to send certification files so they are not left waiting for hours. This might possibly be done with [I]Stop[/I] in the [I]Test[/I] drop-down menu. Call the routine which initiates the transfer. It would remain transparent to the user, as it is now. I have never looked at the code, and would not know what I was looking at if I did. Just an idea...

ATH 2020-08-14 00:10

[QUOTE=Uncwilly;553541]I just upgraded a machine to v30.3 (from 29) and it started to do TF on some of the PRP-CF's that were queued up.[/QUOTE]

My Prime95 also started doing TF that was not needed (I only use Prime95 right now to grab AID for my gpuowl runs).

[CODE]
[Wed Aug 12 01:20:23 2020 - ver 30.3]
Registering assignment: PRP M92136887
Assignment registered as: FE5F87105785D76E778F6C6CDE89DCF0
Sending expected completion date for M92136887: Aug 19 2020
PrimeNet success code with additional info:
Server assigned CERT work.
Got assignment 57179D7B4FA512973588C2FFDD5B9E2E: CERT M5820589
Updating computer information on the server
Sending expected completion date for M92136887: Aug 19 2020
Sending expected completion date for M5820589: Aug 12 2020
Sending result to server: UID: athath/5960X, M92136887 no factor to 2^63, Wh4: 744F6529, AID: FE5F87105785D76E778F6C6CDE89DCF0

PrimeNet success code with additional info:
Result was not needed. TF on M92136887, sf: 77, ef: 63
CPU credit is 0.0000 GHz-days.
Sending result to server: UID: athath/5960X, M92136887 no factor from 2^63 to 2^64, Wh4: 744F6529, AID: FE5F87105785D76E778F6C6CDE89DCF0

PrimeNet success code with additional info:
Result was not needed. TF on M92136887, sf: 63, ef: 64
CPU credit is 0.0213 GHz-days.
Sending result to server: UID: athath/5960X, M92136887 no factor from 2^64 to 2^65, Wh4: 744F6529, AID: FE5F87105785D76E778F6C6CDE89DCF0

PrimeNet success code with additional info:
Result was not needed. TF on M92136887, sf: 64, ef: 65
CPU credit is 0.0406 GHz-days.






[Thu Aug 13 01:20:38 2020 - ver 30.3]
Registering assignment: PRP M91718971
Assignment registered as: 8169F09B0212A9E8FFA80778D097E5E7
Sending expected completion date for M91718971: Aug 20 2020
PrimeNet error 7: Invalid parameter
parameter p int: Invalid int value/precision '-4817'
[Thu Aug 13 18:44:51 2020 - ver 30.3]
Updating computer information on the server
Sending result to server: UID: athath/5960X, M91718971 no factor to 2^63, Wh4: 704F625D, AID: 8169F09B0212A9E8FFA80778D097E5E7

PrimeNet success code with additional info:
Result was not needed. TF on M91718971, sf: 77, ef: 63
CPU credit is 0.0000 GHz-days.
Sending result to server: UID: athath/5960X, M91718971 no factor from 2^63 to 2^64, Wh4: 704F625D, AID: 8169F09B0212A9E8FFA80778D097E5E7

PrimeNet success code with additional info:
Result was not needed. TF on M91718971, sf: 63, ef: 64
CPU credit is 0.0214 GHz-days.
Sending result to server: UID: athath/5960X, M91718971 no factor from 2^64 to 2^65, Wh4: 704F625D, AID: 8169F09B0212A9E8FFA80778D097E5E7

PrimeNet success code with additional info:
Result was not needed. TF on M91718971, sf: 64, ef: 65
CPU credit is 0.0407 GHz-days.






[Fri Aug 14 01:20:23 2020 - ver 30.3]
Registering assignment: PRP M91774321
PrimeNet error 40: No assignment
ra: already assigned, exponent: 91774321, A: 1, b: 2, c: -1
Registering assignment: PRP M94125289
Assignment registered as: 854FD102A2C848D4A9E04509E42C756A
Sending expected completion date for M94125289: Aug 28 2020
Sending result to server: UID: athath/5960X, M91774321 no factor to 2^63, Wh4: 70E962B9

PrimeNet error 40: No assignment
TF result for M91774321 was not needed
Sending result to server: UID: athath/5960X, M91774321 no factor from 2^63 to 2^64, Wh4: 70E962B9

PrimeNet error 40: No assignment
TF result for M91774321 was not needed
[/CODE]

Xyzzy 2020-08-14 00:28

If there was a cert worktype option we are certain enough people would opt-in to take care of them.

:mike:

kriesel 2020-08-14 00:51

[QUOTE=storm5510;553590]What I would like to suggest is a "manual trigger" to send certification files so they are not left waiting for hours..[/QUOTE]
I suggest another check box at Advanced, Manual Communication, labeled something like "Send proof file(s) now"

Uncwilly 2020-08-14 00:59

[QUOTE=storm5510;553590]I restarted four hours later. There was a certification file on one machine which did not get sent before I stopped. GUI programs, like [I]Prime95[/I], are "event driven." They require a user action to affect a change in behavior. From my antique programming days, I know GUI applications are basically a collection of functions and sub-programs.[/QUOTE]Wait and see. I would wager a doughnut that it will take care of things by itself. That is what .spl files are for.
:doh!:

Prime95 2020-08-14 01:20

[QUOTE=ATH;553591]My Prime95 also started doing TF that was not needed (I only use Prime95 right now to grab AID for my gpuowl runs).[/QUOTE]

What did you put in worktodo.txt?

If it was "PRP=1,2,exp,-1" then prime 95 assumes it has been TF'ed to 2^0 and 0 tests will be saved by P-1 (that is, it needs TF and does not need P-1. That's inconsistent and I'll change it).

If it was "PRP=1,2,exp,-1,how_far_factored,0" then I have no clue what is going on.

There is some other strange stuff happening. I get different error responses from the server than you do.

Prime95 2020-08-14 01:42

[QUOTE=Uncwilly;553583]Those came from through Prime95 contacting PrimeNet in the last few days. (I was away from the machine at the time. That was before I upgraded the machine to v30 at all. I looked at some where my machine found factors and TJAOI found some in the last few hours, so, my assignment was outdated.[/QUOTE]

I looked at the Primnet code and tried getting a few assignments of my own. I don't see how anything other than 99 gets in the worktodo file. Can you run with "Debug=1" in the Primenet section of prime.txt? Then email/post prime.log if it happens again.

garo 2020-08-14 08:02

[QUOTE=Mark Rose;553586]Why not make it an opt-in/out checkbox, like getting low assignments?[/QUOTE]

Agreed. I have some boxes that are really slow or on only a few hours each week. These would happily do the CERT work, But they are not good for anything other than TF otherwise. Maybe P-1 but the memory usage from P-1 has an impact on a 2GB machine.

S485122 2020-08-14 09:43

It seems that at least the cofactor certification results are not written to results.txt, only to results.json.txt.

I have two wishes :
- one should be easy to fulfil : a time stamp for each result written to results.txt and results.json.txt, even if a few seconds apart ;
- the second is the possibility to have dates in another format in those files (for instance the format used on this site : yyyyy-mm-dd hh:mm:ss). This could give problems since it introduces one more variant among the many formats to be parsed by the manual result page.

Jacob

axn 2020-08-14 09:55

[QUOTE=S485122;553623]a time stamp for each result written to results.txt and results.json.txt, even if they are a few seconds apart ;[/QUOTE]

All json entries have a timestamp field in it (UTC). This is also a standard format, independent of any locale-specific settings.

S485122 2020-08-14 10:38

[QUOTE=axn;553624]All json entries have a timestamp field in it (UTC). This is also a standard format, independent of any locale-specific settings.[/QUOTE]Indeed, I was referring to the time stamp between square brackets line before each result. I didn't take the time to check if the time appeared somewhere else in a different format. I just looked at a result.jason.txt file with results coming a few seconds apart, and indeed even if there is only one time line for different results, each has is own timestamp. This means that my wishes where already fulfilled with the advent of the jason formatted result files. Sorry for the distraction [noparse]:-([/noparse].

Jacob

storm5510 2020-08-14 12:56

[QUOTE=Uncwilly;553598]Wait and see. I would wager a doughnut that it will take care of things by itself. That is what .spl files are for.
:doh!:[/QUOTE]

There is no [I].spl[/I] file. Yes, it probably will take care of itself. I was concerned about an elongated period of time between the sending of the result and the upload of the certification file. If the period of time in between is not all that important, then I will not concern myself with it any further.

Uncwilly 2020-08-14 13:55

1 Attachment(s)
[QUOTE=Prime95;553601]Can you run with "Debug=1" in the Primenet section of prime.txt? Then email/post prime.log if it happens again.[/QUOTE]
It was already set. I trimmed the file to include about a week before until a few hours after it went through the TF.

ATH 2020-08-14 14:26

[QUOTE=Prime95;553600]If it was "PRP=1,2,exp,-1" then prime 95 assumes it has been TF'ed to 2^0 and 0 tests will be saved by P-1 (that is, it needs TF and does not need P-1. That's inconsistent and I'll change it).[/QUOTE]

Yes I always just use "PRP=1,2,exp,-1" to just register the exponent and get an AID. Version 29.8 does not start doing TF, something has changed?

Uncwilly 2020-08-14 14:30

[QUOTE=ATH;553650]Yes I always just use "PRP=1,2,exp,-1" to just register the exponent and get an AID. Version 29.8 does not start doing TF, something has changed?[/QUOTE]I am guessing that the "priority work" code path is seeing something odd. I think that some of my assignments didn't get the 99 that was expected, because they were cofactor assignments.

kriesel 2020-08-14 14:35

[QUOTE=ATH;553650]Yes I always just use "PRP=1,2,exp,-1" to just register the exponent and get an AID. Version 29.8 does not start doing TF, something has changed?[/QUOTE]The workaround seems apparent; start using "PRP=1,2,exp,-1,78,0" or some such.

Uncwilly 2020-08-14 14:40

[QUOTE=kriesel;553654]The workaround seems apparent; start using "PRP=1,2,exp,-1,78,0" or some such.[/QUOTE]
The problem should be fixed, not catered too. We are beta testing. This is a bug. ATH is a big boy and can deal with some bonus TF. My bonus TF did not take to long. But, users should not have to fix the assignments that PrimeNet hands out.

ATH 2020-08-14 15:30

[QUOTE=kriesel;553654]The workaround seems apparent; start using "PRP=1,2,exp,-1,78,0" or some such.[/QUOTE]

Sure I will, that is no big deal. I'm just curious if this is intended or a bug, since it is a change in behavior.


Somehow it is the "correct" behavior to start trial factoring if no current limit is provided, but with all the GPU TF that has been done in the last many years it is not really the most efficient behavior anymore to assume no TF has been done.

Uncwilly 2020-08-14 15:54

Suggestion:

On a machine where there are multiple workers, allow the user to designate which worker gets to do the cert runs. If I have 4 cores running DC's and 2 running PRP-CF, I might want to let the tiny CF work yield to the Certs, vs the LL-DCs.


Separate issue: with the LL-DC and the PRP-DC work type combined, my machine that has been knocking out LL-DC's down in the Cat 0-1 area (although it and another one got Cat 4 and then Cat 3 work, I don't know why) has now picked up some PRP-DC work. I want to help with the LL-DC trailing edge.

Prime95 2020-08-14 16:20

[QUOTE=ATH;553650]Yes I always just use "PRP=1,2,exp,-1" to just register the exponent and get an AID. Version 29.8 does not start doing TF, something has changed?[/QUOTE]

Thanks for the log file. It clearly shows the problem is with prime95 sometimes losing the "has been factored to 2^99" information. The server is always sending the correct information.

For now, if you are doing PRPCF work add SkipTrialFactoring=1 to prime.txt.

The change from 29.8 is that PRP will now do any need TF before running the PRP test. This code is necessary for some 100Mdigit assignments.

Prime95 2020-08-14 20:20

@ath, @uncwilly: I'm trying to reproduce the problem. So far unsuccessfully. If you notice any of the "99"s in worktodo.txt changing let me know -- especially if it correlates with some other activity like a result reported or CERT assignment or whatever.

From uncwilly's description it looks like the corruption of worktodo.txt occurs in 29.8 also (he reported immediate TF work upon upgrading to 30.3). I've stared at the prime95 code and cannot see where it could be happening.

kriesel 2020-08-14 21:24

[QUOTE=Uncwilly;553655]The problem [B]should be fixed, not catered too[/B]. We are beta testing. This is a bug. ATH is a big boy and can deal with some bonus TF. My bonus TF did not take to long. But, users should not have to fix the assignments that PrimeNet hands out.[/QUOTE]The issue I responded on is not as I understood it with what PrimeNet hands out, but with what PrimeNet makes of what a user manually enters in worktodo.txt, taking the user's input literally, as computers have been known to do for, well, their entire history.
Period instead of a comma, one iteration of a for loop instead of 3, space mission lost, as one example. "For i=1.3" vs. "For i=1,3"

Fixed or catered to is a false dichotomy. George has accepted it as an issue. And subsequent to my post to which you responded as quoted above, it appears to have been a behavior present in 29.8 also, so present for about a year. (Longer?) Unfortunately we only have one George Woltman, and there's a lot to do. And a workaround is useful in the meantime.
It is not best practice to make manual cpu assignments without checking tf level first. Nor to operate with incomplete worktodo records, as that makes the system more fragile.
It's not cost me any cpu tf cycles, although I do a lot of insert PRP=1,2,exp,-1,tf,0 or DoubleCheck=exp,tf,0, to get an assignment/AID. Give the code what it needs, works well.
There are a lot of different cases.
PrimeNet issues the whole worktodo line is one; tries to make sense of what the user entered is another, more complex one.
Client reads incomplete worktodo line, and has worktype PRP, k 1, base 2, exponent exp, c -1, but nothing for tf so 0, tests-saved not entered so 0. No N/A, so client asks server for an assignment matching that during a manual communication.
Maybe the server is too trusting of the client?
Or the client doesn't accept a new tf level returned by the server with the AID, assuming the server goes and looks up the exponent in the database for that case?
George's recent message sounds like the latter. [URL]https://www.mersenneforum.org/showpost.php?p=553662&postcount=55[/URL]
PRP-CF is moot here since I don't run that, but some people do, and a fix will come eventually.

Ensigm 2020-08-14 22:26

Default P-1/ECM stage 2 memory limit
 
Default P-1/ECM stage 2 memory limit is 0.000000 GB for a fresh install, which will result in stage 2 being deferred (to see this, try testing M1277). Maybe keep the original 8MB as default?

Uncwilly 2020-08-14 22:43

[QUOTE=kriesel;553692]The issue I responded on is not as I understood it with what PrimeNet hands out, but with what PrimeNet makes of what a user manually enters in worktodo.txt, [/quote]ATH and I had what appears to be the same issue when upgrading from 29.8 to 30.3 (not 30.1 or 30.2) I had not manually entered those lines in the worktodo.txt. When I submitted the log, I included about a week's worth of data before the oops. This is when some or all of the assignments in question were handed out.

[QUOTE]And subsequent to my post to which you responded as quoted above, it appears to have been a behavior present in 29.8 also, so present for about a year. (Longer?)[/QUOTE]I never saw any TF work done on a PRP assignment while using 29.8. As noted elsewhere the code for PRP assignments changed to do needed TF. What is useful for the 100M digit range is causing issues with the PRP-CF. We are bug reporting in this thread (as we should be doing). Leaving things in the state that shows the anomalous behaviour can give the data to find the source. To bypass it gives us a less complete data set.

[QUOTE]It is not best practice to make manual cpu assignments without checking tf level first. Nor to operate with incomplete worktodo records, as that makes the system more fragile.
It's not cost me any cpu tf cycles, although I do a lot of insert PRP=1,2,exp,-1,tf,0 or DoubleCheck=exp,tf,0, to get an assignment/AID. Give the code what it needs, works well.[/QUOTE]Explain to me what I did that was ill advised. I let Prime95 fetch assignments from PrimeNet all normal, no self assignments.

[quote]PRP-CF is moot here [FONT="Arial Black"]since I don't run that[/FONT], but some people do, and a fix will come eventually.[/QUOTE]It is not moot. It is an issue. It may be a symptom of some other issue that is a really bad.

Uncwilly 2020-08-14 22:46

[QUOTE=Ensigm;553701]Default P-1/ECM stage 2 memory limit is 0.000000 GB for a fresh install, which will result in stage 2 being deferred (to see this, try testing M1277). Maybe keep the original 8MB as default?[/QUOTE]
8MB is currently a waste for P-1 or ECM. Unless you can devote ~100MB don't bother with ECM or P-1. 8MB is legacy from the dark ages. LL and PRP use way more than that now.

Ensigm 2020-08-14 22:57

[QUOTE=Uncwilly;553704]8MB is currently a waste for P-1 or ECM. Unless you can devote ~100MB don't bother with ECM or P-1. 8MB is legacy from the dark ages. LL and PRP use way more than that now.[/QUOTE]


Then there should be a minimum memory for having P-1/ECM as a work preference (manual testing is fine). Otherwise someone will receive the work unit, complete stage 1, and be stranded there.

Ensigm 2020-08-14 23:05

Anyways, 0 MB should not be the default. I downloaded v30 on a testing machine and ran ECM for several hours, and was surprised that no work had been completed. It took me a while to find out that they were all stuck before stage two.

Prime95 2020-08-14 23:36

[QUOTE=Ensigm;553701]Default P-1/ECM stage 2 memory limit is 0.000000 GB for a fresh install, which will result in stage 2 being deferred (to see this, try testing M1277). Maybe keep the original 8MB as default?[/QUOTE]

I'm increasing the default to 256MB. Perhaps it should be higher (like 512MB or 1/8th of RAM, whichever is less).

storm5510 2020-08-14 23:41

[QUOTE=Uncwilly;553704]8MB is currently a waste for P-1 or ECM. Unless you can devote ~100MB don't bother with ECM or P-1. 8MB is legacy from the dark ages. LL and PRP use way more than that now.[/QUOTE]

8 MB is not necessarily a waste on ECM's if you want to run small exponents on an old machine. I do it. The memory allocation determines the size of the exponents assigned. 512 MB is the long-standing minimum for P-1. I usually use 1024 (1 GB) or 1536 (1.5 GB). A few times to 3072 (3 GB). Never higher.

chalsall 2020-08-14 23:44

[QUOTE=Prime95;553711]Perhaps it should be higher (like 512MB or 1/8th of RAM, whichever is less).[/QUOTE]

"What do you need?" - Steven Wright

You already have P-1 specialists, so how much RAM is optimal (but not excessive without explicit authorization) for other work?

pinhodecarlos 2020-08-15 07:44

Think I've joined the PRP-DC-CF party...1 core two threads.

De Wandelaar 2020-08-15 08:11

1 Attachment(s)
Strange characters appear on the communication panel on 1 computer during a certification (not seen on other computers), see attach :

pinhodecarlos 2020-08-15 11:04

Whereabouts can I see the CERTs have been uploaded?

ATH 2020-08-15 12:10

In the Comm thread: "Proof file successfully uploaded"

pinhodecarlos 2020-08-15 12:30

[QUOTE=ATH;553778]In the Comm thread: "Proof file successfully uploaded"[/QUOTE]

Only shown one hour later, thought it would upload the proof straightaway. Thank you.

Awesomeotts 2020-08-15 12:31

[QUOTE=pinhodecarlos;553741]Think I've joined the PRP-DC-CF party...1 core two threads.[/QUOTE]

Same

storm5510 2020-08-15 16:04

[QUOTE=pinhodecarlos;553782]Only shown one hour later, thought it would upload the proof straightaway. Thank you.[/QUOTE]

Usually within 30 minutes, if [I]Prime95[/I] is still running. If not, then the next run. There is a way to see a bit more information. In [I]prime.txt[/I], find the [Primenet] section and set [I]Debug=1[/I]. The added information will appear in the Communication Thread window.

pinhodecarlos 2020-08-15 16:21

How do I run type CERT jobs?


edit: can we edit [I]prime.txt [/I]whilst the client is running?

Prime95 2020-08-15 16:39

[QUOTE=De Wandelaar;553746]Strange characters appear on the communication panel on 1 computer during a certification (not seen on other computers), see attach :[/QUOTE]

In prime.txt, set Debug to less than 2 in the [PrimeNet] section. I suspect you have it set to 2 which is dumping a lot of binary data.

Uncwilly 2020-08-15 16:41

[QUOTE=pinhodecarlos;553797]How do I run type CERT jobs?[/QUOTE]
Sit back and wait for them to roll in.

storm5510 2020-08-15 16:47

[QUOTE=pinhodecarlos;553797]How do I run type CERT jobs?[/QUOTE]

Exclusively? I do not know if this is possible.

[U]Someone correct me if I am wrong[/U]: Certification work is assigned based on a percentage of your CPU time you wish to give it. All of this is in the [I]Options[/I] menu, then [I]Resource Limits[/I]. At the bottom of the first dialog box, the amount of cert data you wish to download per day, in MB. Advanced button, "Certification work limit (% of CPU time:)" Unless you have changed them, both of these settings should be at their default values.

[B]Edit:[/B]

[QUOTE=Uncwilly]Sit back and wait for them to roll in.[/QUOTE]

As he says.

De Wandelaar 2020-08-15 16:53

[QUOTE=Prime95;553798]In prime.txt, set Debug to less than 2 in the [PrimeNet] section. I suspect you have it set to 2 which is dumping a lot of binary data.[/QUOTE]

Indeed, Debug was equal to 2. Now it has been set to 1.
Thanks for your answer !

pinhodecarlos 2020-08-15 16:55

So on readme.txt you have ProofPower examples from 7 to 10, but on undoc.txt you state that ProofPower can go from 5 to 12., which one prevails? Is the former only to show examples?


edit: Upload runs fine from UK buy my upload speed is way above the UK average. Good work!

storm5510 2020-08-15 17:37

There is one thing about the running of certifications I do not understand, the download limit. This is in megabytes. I am running a certification now. The only thing I see is the assignment line in [I]worktodo.txt[/I]. I guess I was expecting to see something larger downloaded. :confused:

firejuggler 2020-08-15 17:44

[QUOTE=pinhodecarlos;553803]So on readme.txt you have ProofPower examples from 7 to 10, but on undoc.txt you state that ProofPower can go from 5 to 12., which one prevails? Is the former only to show examples?


edit: Upload runs fine from UK buy my upload speed is way above the UK average. Good work![/QUOTE]


From what i've seen, the 'usual' power for 87M cert is a ProofPower of 8. Whith an additionnal spare space of 10 G, I work with a PP ( spare the jokes) of 9. Each adittionnal PP double the temp file space needed, for a marginal speed increase.

Prime95 2020-08-15 19:00

[QUOTE=pinhodecarlos;553803]So on readme.txt you have ProofPower examples from 7 to 10, but on undoc.txt you state that ProofPower can go from 5 to 12., which one prevails? Is the former only to show examples?[/QUOTE]

The readme contains a subset of possible proof powers (the ones I'm trying to encourage).

The actual proof power is determined by the available temp disk space (up to the maximum proof power that makes sense).

Going into more detail than is necessary, if you don't give prime95 much temp disk space it will run a 7-by-2, 6-by-2, or 5-by-3 proof. A 7-by-2 is a power=7 proof on the first half and again on the second half of the PRP test. This simulates a power=8 proof using half the disk space and twice the bandwidth.

[QUOTE=storm5510;553806]There is one thing about the running of certifications I do not understand, the download limit. This is in megabytes. I am running a certification now. The only thing I see is the assignment line in [I]worktodo.txt[/I]. I guess I was expecting to see something larger downloaded. :confused:[/QUOTE]

When the CERT starts it downloads a 13MB starting value for wavefront exponents. If you set your daily download limit to 4MB, you will get no more than one wavefront CERT assignment every 3 days.

[QUOTE=firejuggler;553807]From what i've seen, the 'usual' power for 87M cert is a ProofPower of 8. Whith an additionnal spare space of 10 G, I work with a PP ( spare the jokes) of 9. Each adittionnal PP double the temp file space needed, for a marginal speed increase.[/QUOTE]

Correct, except that power=9 is the maximum. Power 10 becomes a possibility at 108,000,000.

pinhodecarlos 2020-08-15 19:25

Thank you for the above details.
Have a fast landline connection but a slower PC, will have to run the PRP DC CF or PRP CF only. Is there a way to only accept CERT work done by front wave? Have the bandwidth but not the CPU power for first time PRP.

Prime95 2020-08-15 19:52

[QUOTE=pinhodecarlos;553815]Thank you for the above details.
Have a fast landline connection but a slower PC, will have to run the PRP DC CF or PRP CF only. Is there a way to only accept CERT work done by front wave? Have the bandwidth but not the CPU power for first time PRP.[/QUOTE]

There is no way at present to control the type of CERT work you get. If a particular type of CERT work is objectionable, simply set download MB limit to zero and you won't get any. CERT firepower is not a problem for GIMPS.

P.S. In the last 24 hours we certified well over 100 wavefront proofs, probably closer to 150. All done by a relatively small number of beta testers.

Prime95 2020-08-15 23:33

I've given up on ATH & Uncwilly's disappearing "99,0" for PRP-CF assignments. I tried to reproduce in 30.3 and 29.8 with an upgrade to 30.3. No luck. Code review -- no luck. Whatever the problem is/was, its subtle.

storm5510 2020-08-15 23:40

[QUOTE=Prime95;553811]...When the CERT starts it downloads a 13MB starting value for wavefront exponents. If you set your daily download limit to 4MB, you will get no more than one wavefront CERT assignment every 3 days...[/QUOTE]

I think there may be a verbiage confusion. From a user's perspective, a "download" would come from Primenet to the them. "Upload" would be from them to Primenet. This is why the math threw me until I saw the example above. By changing that particular setting, I change how much I am sending to Primenet. This makes far more sense, and explains the bandwidth settings.

Thank you for your reply.

Uncwilly 2020-08-16 02:10

The problem only occurred on 1 of the 2 machines that I upgraded. And it was all in the Worker 2 queue.

kriesel 2020-08-17 17:11

"insufficient ram to ever run stage 2."
 
I installed prime95 v30.3b2 on a new build with old components in an even older case I had gutted. Gave it some work to do while the Celeron1840 was not busy feeding gpus. Was surprised to see a 6GB system yield a stage-1-only P-1 run for M122582657, then the error message in the title, since I've run P-1 on 2GB elsewhere.
There was an m122582657 file remaining, which I moved to a large memory system for stage 2, which indicated it could do 480 relative primes in one pass.

I then investigated further on the 6GB system.
It turns out that in prime95, v30.3b2, fresh install on a newly built system,
Options, Resource Limits, Advanced,
Daytime and Nighttime P-1/ECM memory limits had defaulted to [B]0[/B] GB.
Other systems had had earlier versions of prime95 upgraded to v30.3b2 and probably already had enlarged P-1 memory limit values in prime.txt.

kruoli 2020-08-17 19:12

Feature request:
What do you think about an option for saving GMP-ECM stage 1 ECM files instead of a line to results.txt, e.g. in the form [c]exponent_B1_sigma.ecm[/c]. That way, it would be a lot easier to automate starting stage 2.
Another question: Could the ECM hook also be applied to P-1, please?

Ensigm 2020-08-17 21:20

[QUOTE=kriesel;554027]I installed prime95 v30.3b2 on a new build with old components in an even older case I had gutted. Gave it some work to do while the Celeron1840 was not busy feeding gpus. Was surprised to see a 6GB system yield a stage-1-only P-1 run for M122582657, then the error message in the title, since I've run P-1 on 2GB elsewhere.
There was an m122582657 file remaining, which I moved to a large memory system for stage 2, which indicated it could do 480 relative primes in one pass.

I then investigated further on the 6GB system.
It turns out that in prime95, v30.3b2, fresh install on a newly built system,
Options, Resource Limits, Advanced,
Daytime and Nighttime P-1/ECM memory limits had defaulted to [B]0[/B] GB.
Other systems had had earlier versions of prime95 upgraded to v30.3b2 and probably already had enlarged P-1 memory limit values in prime.txt.[/QUOTE]


Yes, this is the issue I mentioned in [URL="https://www.mersenneforum.org/showpost.php?p=553701&postcount=58"]#58[/URL]. Hopefully it will be fixed soon in later versions, as the default will be raised to 256MB ([URL="https://www.mersenneforum.org/showpost.php?p=553711&postcount=63"]#63[/URL]).

kriesel 2020-08-18 03:57

PMinus1 unsupported work type?
 
[CODE][Main thread Aug 17 06:07] Mersenne number primality test program version 30.3
[Main thread Aug 17 06:07] Optimizing for CPU architecture: Core i3/i5/i7, L2 cache size: 2x256 KB, L3 cache size: 2 MB
[Main thread Aug 17 06:07] Starting worker.
[Main thread Aug 17 11:45] Stopping all worker windows.
[Main thread Aug 17 11:45] Execution halted.
[Main thread Aug 17 11:45] Choose Test/Continue to restart.
[Main thread Aug 17 11:45] Starting worker.
[Comm thread Aug 17 22:10] Updating computer information on the server
[Comm thread Aug 17 22:10] Exchanging program options with server
[Comm thread Aug 17 22:10] [B]Registering assignment: P-1 M120996011[/B]
[Comm thread Aug 17 22:10] [B]PrimeNet error 44: Invalid assignment type[/B]
[Comm thread Aug 17 22:10] [B]ra: unsupported assignment work type: 3[/B][/CODE]That was the result of Advanced, P-1, and filling in bounds B1, B2, exponent, selecting Ok,
then Advanced, Manual communication, Send new completion dates to server, Ok,
in prime95 V30.3b2. Worktodo line created by that is Pminus1=1,2,120996011,-1,600000,16000000

keisentraut 2020-08-18 19:14

I just tested this and feel honored to have certified [URL="https://www.mersenne.org/report_exponent/?exp_lo=98282071&full=1"]M98282071[/URL] from George Woltman :smile: One minor comment, not sure if it was mentioned before: For now, [U]mprime seems to ask for certification work on every startup, even if it has queued enough other stuff.[/U] I had to move the CERT work manually to the top, otherwise it would not have been completed for a few more weeks until my other test finishes.

Prime95 2020-08-18 19:40

[QUOTE=keisentraut;554169]I just tested this and feel honored to have certified [URL="https://www.mersenne.org/report_exponent/?exp_lo=98282071&full=1"]M98282071[/URL] from George Woltman :smile: One minor comment, not sure if it was mentioned before: For now, [U]mprime seems to ask for certification work on every startup, even if it has queued enough other stuff.[/U] I had to move the CERT work manually to the top, otherwise it would not have been completed for a few more weeks until my other test finishes.[/QUOTE]

Thanks for the CERT of my PRP Run.

You do not need to re-arrange your worktodo.txt file. Cert work will get done quickly regardless of where it appears in the file.

kruoli 2020-08-19 09:09

The sum-input-output-error-checking got removed? Because of GEC? Or is it rather needless these days, since flaky machines should not do LL-DC, but rather PRP with GEC?

Viliam Furik 2020-08-19 11:59

Self-certification
 
It happened to me twice already, that I got assigned CERT work for my manual results. Is it OK, or should it be fixed? I am not sure if the results could be counterfeited...

firejuggler 2020-08-19 12:42

I think that manual testing is "outside" work, not officially "sanctionned" therefore, escape the rules.

kriesel 2020-08-19 13:42

[QUOTE=Viliam Furik;554231]It happened to me twice already, that I got assigned CERT work for my manual results. Is it OK, or should it be fixed? I am not sure if the results could be counterfeited...[/QUOTE]Not an issue, manual or otherwise. [url]https://www.mersenneforum.org/showpost.php?p=551951&postcount=181[/url]

storm5510 2020-08-19 14:00

[QUOTE=firejuggler;554233]I think that manual testing is "outside" work, not officially "sanctionned" therefore, escape the rules.[/QUOTE]

Only if an attempt is made to submit such work without having a valid assignment ID.

The test options in the Advanced menu may be a holdover from another time and not given much attention now. Sometimes, as a programmer, it is simply better to leave something of this type in a complex program rather than try to remove it.


All times are UTC. The time now is 16:28.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2022, Jelsoft Enterprises Ltd.