Which sequence are you running it on, what OS, what architecture?
This has been plaguing me for some time now, and i was wondering if there is anything that can be done about it. sr2sieve will crash and an error box will pop up at pmin=12732376276526713 and up. Nothing in any of the text files tells me what is going on. I'll try running the client verbosely to see if anything else comes up. I've just been running with the -s switch.
When in doubt, kick it until it works.
Which sequence are you running it on, what OS, what architecture?
Win XP, Amd X2 and as for sequence i'm not sure what you mean by that.. everything is gen/8 from looking at -vv
[Edit] Looks like it runs for about 1 to 10G then errors... *grumble* [/Edit]
Last edited by KWSN_Dagger; 03-02-2009 at 03:29 PM. Reason: run length
When in doubt, kick it until it works.
There are a few pieces of information that will help:
1. What is the address (memory location) of the crash? This should be given somewhere in the popup window.
2. Is the crash repeatable? If you rename the checkpoint file so that it starts from the same place each time, does each crash occur at the same p? If it is not repeatable, does the computer pass a Prime95 torture test?
3. Does the 32-bit executable have the same problem?
edit: I assumed you were using the 64-bit executable. If not, can you try running with the command-line switches -Bsse2/16 -Gsse2/16 --ladder=sse2/16 and see if there is any change in the address of the crash.
Updated some non related files and reboots, seems to be working fine now, as it hasn't crashed since then. I'd love to run 64 bit apps on my comp, but having a 32 bit OS doesn't help. Thanks for all the replies.
When in doubt, kick it until it works.
There was an array overrun bug affecting sr2sieve versions 1.8.6 to 1.8.8 which could possibly cause a crash while sieving, I think 1.8.8 was the latest version when this thread was started. This bug was fixed in version 1.8.9.