Openwall GNU/*/Linux - a small security-enhanced Linux distro for servers
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Thu, 22 Aug 2013 13:33:34 +0400
From: Solar Designer <solar@...nwall.com>
To: john-dev@...ts.openwall.com
Subject: Re: Parallella: Litecoin mining

Rafael,

On Wed, Aug 21, 2013 at 09:46:18PM +0100, Rafael Waldo Delgado Doblas wrote:
> Here is the terminal:
> 
>  [2013-08-21 20:43:02] Started cgminer 3.3.1
>  [2013-08-21 20:43:03] Probing for an alive pool
>  [2013-08-21 20:43:05] Network diff set to 56.9M
>  [2013-08-21 20:43:05] Long-polling activated for
> http://ltc.kattare.com:9332/LP
> (5s):655.0 (avg):1.031Kh/s | A:0  R:0  HW:0  WU:0.0/m
> 
> Did you compile the epiphany code with the script build.sh?

You're right, I did not.  I thought configure and make would take care
of that, especially given that your changes to configure looked like
they're specific to Parallella (that is, they're not acceptable upstream).

I ran build.sh now, and cgminer works like you've shown above for about
one minute, then segfaults (I ran it twice so far, same result):

solar@...aro-ubuntu-desktop:~/LordRafa/cgminer-build$ sudo -E LD_LIBRARY_PATH=${ELIBS} EPIPHANY_HDF=${EHDF} bash
root@...aro-ubuntu-desktop:~/LordRafa/cgminer-build# ./cgminer -o http://ltc.kattare.com:9332 --scrypt -u lordrafa.1 -p 123456 -T
 [2013-08-22 09:25:47] Started cgminer 3.3.1
 [2013-08-22 09:25:48] Probing for an alive pool
 [2013-08-22 09:25:49] Network diff set to 56.9M
 [2013-08-22 09:25:49] Long-polling activated for http://ltc.kattare.com:9332/LP
Segmentation fault1.031Kh/s | A:0  R:0  HW:0  WU:0.0/m
root@...aro-ubuntu-desktop:~/LordRafa/cgminer-build#

I had to explicitly switch to root to see the segfault message.  When
running via the run.sh script, cgminer appears to terminate on its own
(the segfault message is somehow suppressed).

Do you see the same problem?

My guess is that the segfault might be happening whenever the first
share is found... although 1 minute to find 1 share may be too little
time at this low speed.

Anyhow, you need to test with long enough runs that the pool's accepted
share count would be non-zero (and rejected zero).

Alexander

Powered by blists - more mailing lists

Your e-mail address:

Powered by Openwall GNU/*/Linux - Powered by OpenVZ