[ANN] cpuminer-multi v1.3.5 (Linux + Windows VStudio/MinGW64) GPL Open Source

[joblo]

Yikes. Two identical systems, different results, and it crashes in a standard library.

Either the application is passing a bad pointer to curl or curl is corrupt. The easiest thing to try is to reinstall the curl
packages. You could also confirm the same compile options were used. If the entire process from source to compile
to run time environment to input data are identical it gets more complicated and hardware needs to be considered. Memtest maybe.

Edit: I assume you already tried getting a fresh copy of the source.

gdb can be used to get a backtrace to find out where the bad pointer is coming from.

Edit: Cryptoninght at Nicehash needs a protocol tweak but it shouldn't cause as crash. Try another pool to rule that out.

[1713449346]

1713449346

[1713449346]

1713449346

[1713449346]

1713449346

[1713449346]

1713449346

[Hotmetal]

your problem seems related to libcurl, so try to rebuild it yourself with the /usr/local prefix, maybe an uncommon cpu ? atom ?

I prefer using the package management to do things.. Maintaining source builds gets messy.

Manufacturer: Dell Inc.
Product Name: PowerEdge R620
model name      : Intel(R) Xeon(R) CPU E5-2620 0 @ 2.00GHz

I have about 20 blades, all of them with stock Ubuntu 16.04 and all of them have the same behavior.

[Hotmetal]

Yikes. Two identical systems, different results, and it crashes in a standard library.

Either the application is passing a bad pointer to curl or curl is corrupt. The easiest thing to try is to reinstall the curl
packages. You could also confirm the same compile options were used. If the entire process from source to compile
to run time environment to input data are identical it gets more complicated and hardware needs to be considered. Memtest maybe.

Edit: I assume you already tried getting a fresh copy of the source.

gdb can be used to get a backtrace to find out where the bad pointer is coming from.

Edit: Cryptoninght at Nicehash needs a protocol tweak but it shouldn't cause as crash. Try another pool to rule that out.

root@apollo:~/44/cpuminer-multi# gdb ./cpuminer
GNU gdb (Ubuntu 7.11.1-0ubuntu1~16.04) 7.11.1
Copyright (C) 2016 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
and "show warranty" for details.
This GDB was configured as "x86_64-linux-gnu".
Type "show configuration" for configuration details.
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>.
Find the GDB manual and other documentation resources online at:
<http://www.gnu.org/software/gdb/documentation/>.
For help, type "help".
Type "apropos word" to search for commands related to "word"...
Reading symbols from ./cpuminer...(no debugging symbols found)...done.
(gdb) r
Starting program: /root/44/cpuminer-multi/cpuminer
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
** cpuminer-multi 1.3 by tpruvot@github **
BTC donation address: 1FhDPLPpw18X4srecguG3MxJYe4a1JsZnd (tpruvot)

[New Thread 0x2aaab0a2f700 (LWP 11060)]
[New Thread 0x2aaab0c33700 (LWP 11061)]
[2016-12-10 06:08:07] Starting Stratum on stratum+tcp://yiimp.ccminer.org:4252
[New Thread 0x2aaab0e37700 (LWP 11062)]

Thread 3 "cpuminer" received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x2aaab0c33700 (LWP 11061)]
0x00002aaaab0c2b80 in ?? () from /usr/lib/x86_64-linux-gnu/libcurl.so.4
(gdb) bt
#0  0x00002aaaab0c2b80 in ?? () from /usr/lib/x86_64-linux-gnu/libcurl.so.4
#1  0x00002aaaab0c414a in curl_maprintf () from /usr/lib/x86_64-linux-gnu/libcurl.so.4
#2  0x00002aaaab0e9096 in ?? () from /usr/lib/x86_64-linux-gnu/libcurl.so.4
#3  0x00002aaaab0c1164 in ?? () from /usr/lib/x86_64-linux-gnu/libcurl.so.4
#4  0x00002aaaab0d2493 in ?? () from /usr/lib/x86_64-linux-gnu/libcurl.so.4
#5  0x00002aaaab0d2eed in curl_multi_perform () from /usr/lib/x86_64-linux-gnu/libcurl.so.4
#6  0x00002aaaab0c994b in curl_easy_perform () from /usr/lib/x86_64-linux-gnu/libcurl.so.4
#7  0x0000000000434998 in ?? ()
#8  0x0000000000405974 in ?? ()
#9  0x00002aaaab96370a in start_thread (arg=0x2aaab0c33700) at pthread_create.c:333
#10 0x00002aaaaadd982d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:109

So to confirm what you thought, it's not specific to cryptonight but rather with the interaction of the libcurl library.
What is weird is i have a lot of blades and all of them are not behaving nicely.

[Hotmetal]

Just for fun, i went ahead and tested a slightly older release:

https://github.com/tpruvot/cpuminer-multi/archive/v1.0.9-multi.tar.gz

root@apollo:~/44/1/cpuminer-multi-1.0.9-multi# ./cpuminer  -a cryptonight  -p cami@hack.co.za  -o stratum+tcp://cryptonight.eu.nicehash.com:3355  -u 12d7SjLcyEwLCnZhPeQhMnME3dDF9CnSuY -t 12 -q
** cpuminer-multi 1.0.9 by Tanguy Pruvot (tpruvot@github) **
 based on Lucas Jones fork of pooler cpuminer 2.4

BTC donation address: 1FhDPLPpw18X4srecguG3MxJYe4a1JsZnd

[2016-12-10 06:26:32] Using JSON-RPC 2.0
[2016-12-10 06:26:32] CPU Supports AES-NI: YES
[2016-12-10 06:26:32] Starting Stratum on stratum+tcp://cryptonight.eu.nicehash.com:3355
[2016-12-10 06:26:32] 12 miner threads started, using 'cryptonight' algorithm.
[2016-12-10 06:26:33] Pool set diff to 20000
[2016-12-10 06:26:36] thread 11: 66 hashes, 16.70 H/s
[2016-12-10 06:26:36] thread 3: 66 hashes, 14.91 H/s
[2016-12-10 06:26:36] thread 7: 66 hashes, 14.40 H/s
[2016-12-10 06:26:36] thread 5: 66 hashes, 14.28 H/s
[2016-12-10 06:26:37] thread 8: 66 hashes, 14.01 H/s
[2016-12-10 06:26:37] thread 10: 66 hashes, 12.79 H/s
[2016-12-10 06:26:38] thread 4: 66 hashes, 11.61 H/s
[2016-12-10 06:26:38] thread 6: 66 hashes, 11.35 H/s
[2016-12-10 06:26:38] thread 1: 66 hashes, 11.26 H/s
[2016-12-10 06:26:38] thread 0: 66 hashes, 11.23 H/s
[2016-12-10 06:26:38] thread 9: 66 hashes, 10.65 H/s
[2016-12-10 06:26:38] thread 2: 66 hashes, 10.21 H/s

Hmm!

[Hotmetal]

I then tried v1.1:

root@apollo:~/44/1/cpuminer-multi-1.1-multi# gdb ./cpuminer
GNU gdb (Ubuntu 7.11.1-0ubuntu1~16.04) 7.11.1
Copyright (C) 2016 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
and "show warranty" for details.
This GDB was configured as "x86_64-linux-gnu".
Type "show configuration" for configuration details.
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>.
Find the GDB manual and other documentation resources online at:
<http://www.gnu.org/software/gdb/documentation/>.
For help, type "help".
Type "apropos word" to search for commands related to "word"...
Reading symbols from ./cpuminer...(no debugging symbols found)...done.
(gdb) r
Starting program: /root/44/1/cpuminer-multi-1.1-multi/cpuminer
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
** cpuminer-multi 1.1 by Tanguy Pruvot (tpruvot@github) **
BTC donation address: 1FhDPLPpw18X4srecguG3MxJYe4a1JsZnd

[New Thread 0x2aaab0a2e700 (LWP 19010)]
[New Thread 0x2aaab0c31700 (LWP 19011)]
[2016-12-10 06:43:18] Starting Stratum on stratum+tcp://mine.xpool.ca:1131
[New Thread 0x2aaab0e34700 (LWP 19012)]

Thread 3 "cpuminer" received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x2aaab0c31700 (LWP 19011)]
_IO_vfscanf_internal (s=s@entry=0x2aaab0a30620, format=format@entry=0x2aaaab0f308b "%15[^:]:%[^\n]",
    argptr=argptr@entry=0x2aaab0a30748, errp=errp@entry=0x0) at vfscanf.c:279
279     vfscanf.c: No such file or directory.
(gdb) bt
#0  _IO_vfscanf_internal (s=s@entry=0x2aaab0a30620, format=format@entry=0x2aaaab0f308b "%15[^:]:%[^\n]",
    argptr=argptr@entry=0x2aaab0a30748, errp=errp@entry=0x0) at vfscanf.c:279
#1  0x00002aaaaad3ebcc in __GI___isoc99_vsscanf (string=0x2aaab8001320 "http://mine.xpool.ca:1131",
    format=0x2aaaab0f308b "%15[^:]:%[^\n]", args=args@entry=0x2aaab0a30748) at isoc99_vsscanf.c:43
#2  0x00002aaaaad3eb57 in __isoc99_sscanf (s=<optimized out>, format=<optimized out>) at isoc99_sscanf.c:31
#3  0x00002aaaab0be7db in ?? () from /usr/lib/x86_64-linux-gnu/libcurl.so.4
#4  0x00002aaaab0d2493 in ?? () from /usr/lib/x86_64-linux-gnu/libcurl.so.4
#5  0x00002aaaab0d2eed in curl_multi_perform () from /usr/lib/x86_64-linux-gnu/libcurl.so.4
#6  0x00002aaaab0c994b in curl_easy_perform () from /usr/lib/x86_64-linux-gnu/libcurl.so.4
#7  0x0000000000468e08 in ?? ()
#8  0x000000000046a073 in ?? ()
#9  0x00002aaaab96370a in start_thread (arg=0x2aaab0c31700) at pthread_create.c:333

[joblo]

That's bizarre.

No one has reported this with cpuminer-opt even though the stratum code was taken from multi-1.2.
Could you give it a try as another data point?

The interface to curl is the call to curl_easy perform. It would be useful to dump the curl parameter where this
is called to compare what works with what doesn't.

A hardware fault seem very unlikely since the problem occurs on all your blades. However, if you did one full
install and cloned the rest an install glitch would have been propagated to all blades. You could reinstall the OS
on one system to test this.

[zakuti]

hello

i try to build it on visual studio 2013 but i get next error

error LNK1181: cannot open input file 'libcurl.x86.lib' cpuminer

can someone help me ?

thank you.

[Epsylon3]

32bit dependencies are not prebuilt, they are huge and polute the repo

[zakuti]

but i have 64 bit operation system

[zakuti]

i think that compilied but miner does not start mining

please look picture

http://www.picz.ge/img/s4/1612/13/7/7804c9a7d51f.png
what problem is it ?

[Epsylon3]

patience, you are closer sorry i've no easy answer for all your weird problems. All i can say is it can work for all these cases :p

superlux, try another mingw, msys2 (cygwin) is not proper to do "native windows binaries"

[joblo]

For cpuminer-opt I bundle a bunch of DLLs with the EXE for portability. Maybe that works for
cpuminer-multi also.

[cdavid53]

Hello

I would like to use several pools.
is it possible ? and if yes, how?

thank you in advance

[integrale]

Hello Guy´s 
First of all im new here but not on Mining , you all the Dev done a great job with the Miner´s .

may i ask some of the Dev to answer me some questions about TIMETRAVEL , especially  witch OS ( seems only Linux ) is to use if im right , how i set the algo in the command line ,   have searched few Days  cant find any usefull information  ,

THX in Advance

[Epsylon3]

nothing special about it, just like any other algos... fyi my "linux" branch is the dev one.. windows one is more for releases (so you can compile the linux branch on windows too)

[Epsylon3]

made the 1.3.1 release with timetravel, see first post

[SkycityVictorians]

made the 1.3.1 release with timetravel, see first post

Nice job, thanks

Friendly Regards
- SkycityVictorians

[integrale]

@Epsylon3

thank you  for the great job ,

i7 2600 @ 244 k/h
Core 2 Duo E6600   sse2 doesnt work in my case
Athlon 64 x 2          sse2 doesnt too
core 2 Duo E6700   sse2 doesnt too

all Machines running win 7 64bit



if you are so kind PM me your Mac adress for some Donation

[Epsylon3]

its in the first post

what doesnt work ? ah right, didnt made it for very old cpus there are limits to the compatibility... built with --march=corei7

your core2 has 12 years now joblo (cpuminer-opt) will do one for you with (more than) my improvement..

[integrale]

Oh Man thats good news ,  you bewared my PC^s  from a long flight downwards the Balcony   lol



You Guy´s are Great