1)
Message boards :
Number crunching :
Tour de Primes 2015
(Message 83645)
Posted 3009 days ago by vmc
Nice :) Never thought I may win this with just one PC. Good portion of luck though :)
CG Scott and Randall.
|
2)
Message boards :
Project Staging Area :
Kernel error on 32bit linux
(Message 27334)
Posted 4600 days ago by vmc
Oh, the error is alwayes after n hit. so when the client want to post a prime, it'll fault due to 'Kernel Fatal'
|
3)
Message boards :
Project Staging Area :
Kernel error on 32bit linux
(Message 27230)
Posted 4602 days ago by vmc
It's a VPS/VM where it's impossible to update the kernel (on my side, dunno what the hoster is waiting for)
|
4)
Message boards :
Project Staging Area :
Kernel error on 32bit linux
(Message 27218)
Posted 4603 days ago by vmc
It's prpclient-3.3.6beta-linux_32
Oh, I've to say, that this happened once in about 1000 WUs.
But I also got 'missing lresults.txt" or stuff like this,
the app thought it was an ctrl-c, but no.
I've restarted the app and it's working without any problems :/
Dunno what 'd happend there.
|
5)
Message boards :
Project Staging Area :
Kernel error on 32bit linux
(Message 27212)
Posted 4603 days ago by vmc
Hi,
today I've got a FATAL: Kernel too old (with a 2.6.9, yeah old)
I've not straced it yet, but maybe someone has an idea what's the problem here.
I don't think that the llr app has to use some new kernel feauters .. maybe just a wrong 'asm {int 0x80}' call (as I've got an invlid (assembler) instruction before)
Ideas?
|
6)
Message boards :
Project Staging Area :
PRNet Discussion (Old)
(Message 20879)
Posted 4856 days ago by vmc
You might want to add some error checking.
The POSIX thread-calls always set errno ...
|
7)
Message boards :
Project Staging Area :
PRNet Discussion (Old)
(Message 20872)
Posted 4857 days ago by vmc
I've seen this bug and fixed it for my 32bit version (Since the prp-zipbal does NOT include any prpclient for 32bit, although it's posted to do so ... !) .
Also I don't see a coherence between priority and threads. But not yet checked the code.
|
8)
Message boards :
Project Staging Area :
PRNet Discussion (Old)
(Message 20870)
Posted 4857 days ago by vmc
I'll take a look at the code today evening :)
Running on GDB without triggering the error is not a solution :/
|
9)
Message boards :
Project Staging Area :
PRNet Discussion (Old)
(Message 20865)
Posted 4857 days ago by vmc
Hi,
got some ugly assertion-error today with 3.1.5 on a Gentoo 64-bit system.
PRPNet 3.1.5 PPSE
Starting Proth prime test of 9465*2^700596+1
Using FFT length 64K, a = 11
9465*2^700596+1 is not prime. Proth RES64: 388EFD5FFBA4AC62 Time : 1650.680 sec.
prpclient: tpp.c:63: __pthread_tpp_change_priority: Assertion `new_prio == -1 || (new_prio >= __sched_fifo_min_prio && new_prio <= __sched_fifo_max_prio)' failed.
Aborted
After restarting prpclient, it computed 9465*2^700596+1 again and ended without an error.
Just wanted to post it, maybe it helps the developer :)
|
