Prime95 version 25.7

[WraithX]

OK, now that P-1 is done on the number I was assigned earlier, I am seeing the following error:

[Main thread Oct 7 19:22] Mersenne number primality test program version 25.7
[Main thread Oct 7 19:27] Starting worker.
<...snip original successful server comms...>
[Comm thread Oct 7 19:27] Done communicating with server.
[Comm thread Oct 8 19:27] Updating computer information on the server
[Comm thread Oct 8 19:27] URL: http://v5.mersenne.org/v5server/?v=0...071d34af06&wg=
[Comm thread Oct 8 19:27] Unexpected CURL library error: couldn't connect to host
[Comm thread Oct 8 19:27] Unexpected CURL library error: couldn't connect to host
[Comm thread Oct 8 19:27] Visit http://mersenneforum.org for help.
[Comm thread Oct 8 19:27] Will try contacting server again in 70 minutes.
[Comm thread Oct 8 20:37] Updating computer information on the server
[Comm thread Oct 8 20:37] URL: http://v5.mersenne.org/v5server/?v=0...071d34af06&wg=
[Comm thread Oct 8 20:37] Unexpected CURL library error: couldn't connect to host
[Comm thread Oct 8 20:37] Unexpected CURL library error: couldn't connect to host
[Comm thread Oct 8 20:37] Visit http://mersenneforum.org for help.
[Comm thread Oct 8 20:37] Will try contacting server again in 70 minutes.
[Comm thread Oct 8 21:47] Updating computer information on the server
[Comm thread Oct 8 21:47] URL: http://v5.mersenne.org/v5server/?v=0...071d34af06&wg=
[Comm thread Oct 8 21:47] Unexpected CURL library error: couldn't connect to host
[Comm thread Oct 8 21:47] Unexpected CURL library error: couldn't connect to host
[Comm thread Oct 8 21:47] Visit http://mersenneforum.org for help.
[Comm thread Oct 8 21:47] Will try contacting server again in 70 minutes.


Is any additional info needed to help diagnose this problem?

[jrk]

Originally Posted by WraithX

[Comm thread Oct 8 19:27] Updating computer information on the server
[Comm thread Oct 8 19:27] URL: http://v5.mersenne.org/v5server/?v=0...071d34af06&wg=
[Comm thread Oct 8 19:27] Unexpected CURL library error: couldn't connect to host
[Comm thread Oct 8 19:27] Unexpected CURL library error: couldn't connect to host
[Comm thread Oct 8 19:27] Visit http://mersenneforum.org for help.
[Comm thread Oct 8 19:27] Will try contacting server again in 70 minutes.

See:

Originally Posted by Uncwilly

It seems that there is an outage on the V5 server. No response from my browser or Prime95.

[tmorrow]

Minor items

1. Help | Help Topics produces "failed to launch help".
2. Edit | Copy ... does nothing

I have discovered an asymmetry in how priorities are handled. This may be expected behaviour but I thought I'd report it just in case. Long winded explanation follows.

I have a quad core system running XP. I have configured 4 workers all doing Trial Factoring, CPU Affinity set to "Smart Assignment", default priority 1. Prime95 25.7 is running as a service from bootup.

Periodically I run other factoring programs like (msieve -p) or (ecm -n), each launched with their low/idle priority switch. This results in an even load on whichever core is used, both the respective Prime95 worker and the factoring program getting an even slice. Upon completion of the factoring job, the cpu is returned back to the worker. This arrangement works well.

However, if I stop and restart Prime95 while these factor programs are running, I find that the 4 workers are assigned to the 4 cores but those workers assigned to cores where the factoring programs are running do no work. Restarting the factoring programs while Prime95 is running restores the even sharing.

[James Heinrich]

Originally Posted by tmorrow

2. Edit | Copy ... does nothing

Works for me -- it copies the current child window's contents to the clipboard (same as Ctrl+C).

[tmorrow]

Originally Posted by James Heinrich

Works for me -- it copies the current child window's contents to the clipboard (same as Ctrl+C).

Ok, I didn't know how it worked. I've tried it and it works fine. It seems that newlines are converted into 2 spaces so all the output is on one large line, is that expected?

[James Heinrich]

Originally Posted by tmorrow

It seems that newlines are converted into 2 spaces so all the output is on one large line, is that expected?

That's not quite it -- the text is copied with a UNIX-style newline (LF -- 0x0A), which if you paste into most Windows-based text editors (which expect newlines to be represented by (CRLF -- 0x0D0A)) is not recognized as a valid line-separator character and is therefore treated as whitespace.

But yes, it would be more convenient if George could tweak it so that the text is copied with CRLF linebreaks, please? :)

[Prime95]

Originally Posted by James Heinrich

it would be more convenient if George could tweak it so that the text is copied with CRLF linebreaks, please? :)

Will do. Next build will also change the menu choice to "Copy Window".

[tmorrow]

I'm using the V4 server. In the last 24 hours I've been receiving several errors:

[Comm thread Oct 10 07:39] Unexpected CURL library error: couldn't connect to host

Is this a server-side issue?

[Uncwilly]

Originally Posted by James Heinrich

• If one thread is idle because it has run out of work to do that doesn't require a lot of RAM, see if it can borrow or steal a worktodo entry from another thread. For example, if thread #1 is doing P-1 work, but it's done all the stage1 work it can for now but needs to wait until more memory is available, it should be able to look at the queued work for other threads and steal (or borrow) low memory work (TF, P-1 stage 1, LL, etc) to do while it waits for more memory to be available.
• I'd like to see all worktodo automagically assigned to available CPUs (unless the user has explicitly requested that certain cores only do certain work). There's no reason for core[1] to be idle when workdoto[2] has work waiting.
  An extension to the above: If one (or more) cores are idle, and another core is working on something that can reasonably exploit more multithreading, respawn the other work with the appropriate number of threads to fully utilize the CPU. I'm currently doing a hack version of this: I have a dual-core CPU running TF-LMH on one core and LL on the other. Problem is that every now and then the server goes down and I can't get any work for the TF thread to do. Normally that would mean one core idle for hours (or days), but I have the LL set to run in 2 threads. Perhaps it's not ideal when TF has work (I'm not sure the of exact inefficiency ratio) but overall my throughput is definitely higher due to capturing all the otherwise-idle time on the TF core.

If the v5 server stays incommunicado, I will do that on some of my machines. On others I may raise the TF bit level on expos 2 or 3 notches. Also, I will then start taking data in hand and start making more high usage entries.

[James Heinrich]

Originally Posted by tmorrow

this a server-side issue?

Yes (server appears to have been down for around 24 hours now).

[Prime95]

Has anyone tried 25.7 on a laptop? Does it stop properly when on battery? Does it resume properly? Is hibernation ever a problem?