[LOCKED] cpuminer-opt v3.12.3, open source optimized multi-algo CPU miner

[hmage]

[root@CC ~]# rpm -qa|grep gcc
libgcc-4.4.7-18.el6.x86_64
gcc-c++-4.4.7-18.el6.x86_64
gcc-4.4.7-18.el6.x86_64
libgcc-4.4.7-18.el6.i686

Yes, you need to upgrade to newer gcc. Or ask joblo to support a compiler that's 9 years old.

[zhq123456]

[root@CC ~]# rpm -qa|grep gcc
libgcc-4.4.7-18.el6.x86_64
gcc-c++-4.4.7-18.el6.x86_64
gcc-4.4.7-18.el6.x86_64
libgcc-4.4.7-18.el6.i686

Yes, you need to upgrade to newer gcc. Or ask joblo to support a compiler that's 9 years old.

Is 4.4

[hmage]

[root@CC ~]# rpm -qa|grep gcc
libgcc-4.4.7-18.el6.x86_64
gcc-c++-4.4.7-18.el6.x86_64
gcc-4.4.7-18.el6.x86_64
libgcc-4.4.7-18.el6.i686

Yes, you need to upgrade to newer gcc. Or ask joblo to support a compiler that's 9 years old.

Is 4.4

You need 4.5 or newer.

[joblo]

You can try moving the pragma directives outside the function to see if that works.
If it's trivial to support gcc4.4 I'll make the change in the next release.

[zhq123456]

[root@CC ~]# rpm -qa|grep gcc
libgcc-4.4.7-18.el6.x86_64
gcc-c++-4.4.7-18.el6.x86_64
gcc-4.4.7-18.el6.x86_64
libgcc-4.4.7-18.el6.i686

Yes, you need to upgrade to newer gcc. Or ask joblo to support a compiler that's 9 years old.

Is 4.4

You need 4.5 or newer.

Are installed GCC4.8, starting long wait... , thank you for your answer

[zhq123456]

[root@CC ~]# rpm -qa|grep gcc
libgcc-4.4.7-18.el6.x86_64
gcc-c++-4.4.7-18.el6.x86_64
gcc-4.4.7-18.el6.x86_64
libgcc-4.4.7-18.el6.i686

Yes, you need to upgrade to newer gcc. Or ask joblo to support a compiler that's 9 years old.

Is 4.4

You need 4.5 or newer.

Are installed GCC4.8, starting long wait... , thank you for your answer

make[2]: *** [algo/argon2/ar2/cpuminer-ar2-scrypt-jane.o] Error 1
make[2]: Leaving directory `/usr/cpuminer-opt'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/usr/cpuminer-opt'
make: *** [all] Error 2

[root@CC cpuminer-opt]# gcc -v
Using built-in specs.
COLLECT_GCC=gcc
COLLECT_LTO_WRAPPER=/usr/local/gcc/libexec/gcc/x86_64-unknown-linux-gnu/4.8.2/lto-wrapper
Target: x86_64-unknown-linux-gnu
Configured with: ../configure --prefix=/usr/local/gcc --enable-threads=posix --disable-checking --enable-languages=c,c++ --disable-multilib
Thread model: posix
gcc version 4.8.2 (GCC)
[root@CC cpuminer-opt]# 

[hmage]

[root@CC ~]# rpm -qa|grep gcc
libgcc-4.4.7-18.el6.x86_64
gcc-c++-4.4.7-18.el6.x86_64
gcc-4.4.7-18.el6.x86_64
libgcc-4.4.7-18.el6.i686

Yes, you need to upgrade to newer gcc. Or ask joblo to support a compiler that's 9 years old.

Is 4.4

You need 4.5 or newer.

Are installed GCC4.8, starting long wait... , thank you for your answer

make[2]: *** [algo/argon2/ar2/cpuminer-ar2-scrypt-jane.o] Error 1
make[2]: Leaving directory `/usr/cpuminer-opt'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/usr/cpuminer-opt'
make: *** [all] Error 2

[root@CC cpuminer-opt]# gcc -v
Using built-in specs.
COLLECT_GCC=gcc
COLLECT_LTO_WRAPPER=/usr/local/gcc/libexec/gcc/x86_64-unknown-linux-gnu/4.8.2/lto-wrapper
Target: x86_64-unknown-linux-gnu
Configured with: ../configure --prefix=/usr/local/gcc --enable-threads=posix --disable-checking --enable-languages=c,c++ --disable-multilib
Thread model: posix
gcc version 4.8.2 (GCC)
[root@CC cpuminer-opt]# 

This is different error, and you didn't include it here, it's higher. You've pasted make output saying that there was an error above the pasted text.

[zhq123456]

[root@CC cpuminer-opt]# ./autogen.sh
[root@CC cpuminer-opt]# CFLAGS="-O3 -march=native -Wall" CXXFLAGS="$CFLAGS -std=gnu++11" ./configure --with-curl
checking build system type... x86_64-unknown-linux-gnu
checking host system type... x86_64-unknown-linux-gnu
checking target system type... x86_64-unknown-linux-gnu
checking for a BSD-compatible install... /usr/bin/install -c
checking whether build environment is sane... yes
checking for a thread-safe mkdir -p... /bin/mkdir -p
checking for gawk... gawk
checking whether make sets $(MAKE)... yes
checking whether to enable maintainer-specific portions of Makefiles... no
checking for style of include used by make... GNU
checking for gcc... gcc
checking for C compiler default output file name... a.out
checking whether the C compiler works... yes
checking whether we are cross compiling... no
checking for suffix of executables...
checking for suffix of object files... o
checking whether we are using the GNU C compiler... yes
checking whether gcc accepts -g... yes
checking for gcc option to accept ISO C89... none needed
checking dependency style of gcc... gcc3
checking for gcc option to accept ISO C99... -std=gnu99
checking how to run the C preprocessor... gcc -std=gnu99 -E
checking for grep that handles long lines and -e... /bin/grep
checking for egrep... /bin/grep -E
checking whether gcc -std=gnu99 needs -traditional... no
checking whether gcc -std=gnu99 and cc understand -c and -o together... yes
checking dependency style of gcc -std=gnu99... gcc3
checking for ranlib... ranlib
checking for g++... g++
checking whether we are using the GNU C++ compiler... yes
checking whether g++ accepts -g... yes
checking dependency style of g++... gcc3
checking for ANSI C header files... yes
checking for sys/types.h... yes
checking for sys/stat.h... yes
checking for stdlib.h... yes
checking for string.h... yes
checking for memory.h... yes
checking for strings.h... yes
checking for inttypes.h... yes
checking for stdint.h... yes
checking for unistd.h... yes
checking sys/endian.h usability... no
checking sys/endian.h presence... no
checking for sys/endian.h... no
checking sys/param.h usability... yes
checking sys/param.h presence... yes
checking for sys/param.h... yes
checking syslog.h usability... yes
checking syslog.h presence... yes
checking for syslog.h... yes
checking for sys/sysctl.h... yes
checking whether be32dec is declared... no
checking whether le32dec is declared... no
checking whether be32enc is declared... no
checking whether le32enc is declared... no
checking whether le16dec is declared... no
checking whether le16enc is declared... no
checking for working alloca.h... yes
checking for alloca... yes
checking for getopt_long... yes
checking whether we can compile AVX code... yes
checking whether we can compile XOP code... yes
checking whether we can compile AVX2 code... no
configure: WARNING: The assembler does not support the AVX2 instruction set.
checking for json_loads in -ljansson... no
checking for pthread_create in -lpthread... yes
checking whether __uint128_t is supported... yes
configure: creating ./config.status
config.status: creating Makefile
config.status: creating compat/Makefile
config.status: creating compat/jansson/Makefile
config.status: creating cpuminer-config.h
config.status: cpuminer-config.h is unchanged
config.status: executing depfiles commands
[root@CC cpuminer-opt]# make
make  all-recursive
make[1]: Entering directory `/usr/cpuminer-opt'
Making all in compat
make[2]: Entering directory `/usr/cpuminer-opt/compat'
Making all in jansson
make[3]: Entering directory `/usr/cpuminer-opt/compat/jansson'
make[3]: Nothing to be done for `all'.
make[3]: Leaving directory `/usr/cpuminer-opt/compat/jansson'
make[3]: Entering directory `/usr/cpuminer-opt/compat'
make[3]: Nothing to be done for `all-am'.
make[3]: Leaving directory `/usr/cpuminer-opt/compat'
make[2]: Leaving directory `/usr/cpuminer-opt/compat'
make[2]: Entering directory `/usr/cpuminer-opt'
gcc -std=gnu99 -DHAVE_CONFIG_H -I.  -Iyes/include -fno-strict-aliasing -I./compat/jansson -I. -Iyes/include -Wno-pointer-sign -Wno-pointer-to-int-cast   -O3 -march=native -Wall  -Iyes/include -MT algo/argon2/ar2/cpuminer-ar2-scrypt-jane.o -MD -MP -MF algo/argon2/ar2/.deps/cpuminer-ar2-scrypt-jane.Tpo -c -o algo/argon2/ar2/cpuminer-ar2-scrypt-jane.o `test -f 'algo/argon2/ar2/ar2-scrypt-jane.c' || echo './'`algo/argon2/ar2/ar2-scrypt-jane.c
In file included from algo/argon2/ar2/sj/scrypt-jane-salsa64.h:12:0,
                 from algo/argon2/ar2/sj/scrypt-jane-romix.h:2,
                 from algo/argon2/ar2/ar2-scrypt-jane.c:19:
algo/argon2/ar2/sj/scrypt-jane-romix-basic.h:8:1: warning: ‘scrypt_romix_nop’ defined but not used [-Wunused-function]
 scrypt_romix_nop(scrypt_mix_word_t *blocks, size_t nblocks) {
 ^
/tmp/cc5W6W1g.s: Assembler messages:
/tmp/cc5W6W1g.s:21: Error: ambiguous operand size or operands invalid for `vpxor'
/tmp/cc5W6W1g.s:22: Error: ambiguous operand size or operands invalid for `vpxor'
/tmp/cc5W6W1g.s:23: Error: ambiguous operand size or operands invalid for `vpxor'
/tmp/cc5W6W1g.s:24: Error: ambiguous operand size or operands invalid for `vpxor'
/tmp/cc5W6W1g.s:30: Error: ambiguous operand size or operands invalid for `vpxor'
/tmp/cc5W6W1g.s:31: Error: ambiguous operand size or operands invalid for `vpxor'
/tmp/cc5W6W1g.s:32: Error: ambiguous operand size or operands invalid for `vpxor'
/tmp/cc5W6W1g.s:33: Error: ambiguous operand size or operands invalid for `vpxor'
/tmp/cc5W6W1g.s:37: Error: ambiguous operand size or operands invalid for `vpxor'
/tmp/cc5W6W1g.s:38: Error: ambiguous operand size or operands invalid for `vpxor'
/tmp/cc5W6W1g.s:39: Error: ambiguous operand size or operands invalid for `vpxor'
/tmp/cc5W6W1g.s:40: Error: ambiguous operand size or operands invalid for `vpxor'
/tmp/cc5W6W1g.s:48: Error: ambiguous operand size or operands invalid for `vpaddq'
/tmp/cc5W6W1g.s:49: Error: ambiguous operand size or operands invalid for `vpshufd'
/tmp/cc5W6W1g.s:50: Error: ambiguous operand size or operands invalid for `vpxor'
/tmp/cc5W6W1g.s:51: Error: ambiguous operand size or operands invalid for `vpaddq'
/tmp/cc5W6W1g.s:52: Error: ambiguous operand size or operands invalid for `vpsrlq'
/tmp/cc5W6W1g.s:53: Error: ambiguous operand size or operands invalid for `vpxor'
/tmp/cc5W6W1g.s:54: Error: ambiguous operand size or operands invalid for `vpsllq'
/tmp/cc5W6W1g.s:55: Error: ambiguous operand size or operands invalid for `vpxor'
/tmp/cc5W6W1g.s:56: Error: ambiguous operand size or operands invalid for `vpaddq'
/tmp/cc5W6W1g.s:57: Error: ambiguous operand size or operands invalid for `vpsrlq'
/tmp/cc5W6W1g.s:58: Error: ambiguous operand size or operands invalid for `vpxor'
/tmp/cc5W6W1g.s:59: Error: ambiguous operand size or operands invalid for `vpsllq'
/tmp/cc5W6W1g.s:60: Error: ambiguous operand size or operands invalid for `vpxor'
/tmp/cc5W6W1g.s:61: Error: ambiguous operand size or operands invalid for `vpaddq'
/tmp/cc5W6W1g.s:62: Error: ambiguous operand size or operands invalid for `vpshufd'
/tmp/cc5W6W1g.s:63: Error: no such instruction: `vpermq ymm1,ymm1,0x39'
/tmp/cc5W6W1g.s:64: Error: no such instruction: `vpermq ymm10,ymm2,0x4e'
/tmp/cc5W6W1g.s:65: Error: ambiguous operand size or operands invalid for `vpxor'
/tmp/cc5W6W1g.s:66: Error: no such instruction: `vpermq ymm3,ymm3,0x93'
/tmp/cc5W6W1g.s:67: Error: ambiguous operand size or operands invalid for `vpaddq'
/tmp/cc5W6W1g.s:68: Error: ambiguous operand size or operands invalid for `vpshufd'
/tmp/cc5W6W1g.s:69: Error: ambiguous operand size or operands invalid for `vpxor'
/tmp/cc5W6W1g.s:70: Error: ambiguous operand size or operands invalid for `vpaddq'
/tmp/cc5W6W1g.s:71: Error: ambiguous operand size or operands invalid for `vpsrlq'
/tmp/cc5W6W1g.s:72: Error: ambiguous operand size or operands invalid for `vpxor'
/tmp/cc5W6W1g.s:73: Error: ambiguous operand size or operands invalid for `vpsllq'
/tmp/cc5W6W1g.s:74: Error: ambiguous operand size or operands invalid for `vpxor'
/tmp/cc5W6W1g.s:75: Error: ambiguous operand size or operands invalid for `vpaddq'
/tmp/cc5W6W1g.s:76: Error: ambiguous operand size or operands invalid for `vpsrlq'
/tmp/cc5W6W1g.s:77: Error: ambiguous operand size or operands invalid for `vpxor'
/tmp/cc5W6W1g.s:78: Error: ambiguous operand size or operands invalid for `vpsllq'
/tmp/cc5W6W1g.s:79: Error: no such instruction: `vpermq ymm1,ymm1,0x93'
/tmp/cc5W6W1g.s:80: Error: ambiguous operand size or operands invalid for `vpxor'
/tmp/cc5W6W1g.s:81: Error: no such instruction: `vpermq ymm2,ymm10,0x4e'
/tmp/cc5W6W1g.s:82: Error: ambiguous operand size or operands invalid for `vpaddq'
/tmp/cc5W6W1g.s:83: Error: ambiguous operand size or operands invalid for `vpshufd'
/tmp/cc5W6W1g.s:84: Error: no such instruction: `vpermq ymm3,ymm3,0x39'
/tmp/cc5W6W1g.s:85: Error: ambiguous operand size or operands invalid for `vpxor'
/tmp/cc5W6W1g.s:90: Error: ambiguous operand size or operands invalid for `vpaddq'
/tmp/cc5W6W1g.s:91: Error: ambiguous operand size or operands invalid for `vpaddq'
/tmp/cc5W6W1g.s:92: Error: ambiguous operand size or operands invalid for `vpaddq'
/tmp/cc5W6W1g.s:93: Error: ambiguous operand size or operands invalid for `vpaddq'
make[2]: *** [algo/argon2/ar2/cpuminer-ar2-scrypt-jane.o] Error 1
make[2]: Leaving directory `/usr/cpuminer-opt'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/usr/cpuminer-opt'
make: *** [all] Error 2
[root@CC cpuminer-opt]# gcc -v
Using built-in specs.
COLLECT_GCC=gcc
COLLECT_LTO_WRAPPER=/usr/local/gcc/libexec/gcc/x86_64-unknown-linux-gnu/4.8.2/lto-wrapper
Target: x86_64-unknown-linux-gnu
Configured with: ../configure --prefix=/usr/local/gcc --enable-threads=posix --disable-checking --enable-languages=c,c++ --disable-multilib
Thread model: posix
gcc version 4.8.2 (GCC)

[hmage]

/tmp/cc5W6W1g.s:63: Error: no such instruction: `vpermq ymm1,ymm1,0x39'

GNU assembler is too old as well, unfortunately. It doesn't know what 'vpermq' is -- that's an AVX2 instruction.

Same goes for other instructions it doesn't know about.

GNU assembler is part of binutils. Try installing newer version of binutils.

I know for sure that binutils 2.25 supports it, but some earlier versions likely will too.

[joblo]

It's not so trivial anymore.

It might be worthwhile to focus the problem a little more. I suggest trying to compile with a lesser arch
like core2 to see if you can, then try to other arches until you identify the one that fails.

The problem seems to be with the compile environment but it could also be your CPU. If it's an AMD it
adds another dimension because their implementation of some features is incompatible with Intel.
The RELEASE_NOTES have some tips.

[hmage]

It's not so trivial anymore.

It might be worthwhile to focus the problem a little more. I suggest trying to compile with a lesser arch
like core2 to see if you can, then try to other arches until you identify the one that fails.

The problem seems to be with the compile environment but it could also be your CPU. If it's an AMD it
adds another dimension because their implementation of some features is incompatible with Intel.
The RELEASE_NOTES have some tips.

Won't help, he needs to update binutils. Even if he uses -march=core2 it will still emit AVX2 instructions and old GNU assembler will break the compilation -- the file responsible is algo/argon2/ar2/sj/scrypt-jane-mix_salsa64-avx2.h, it will emit that instruction no matter what -march compiler option is provided.

You can easily replicate that problem yourself with installing gcc 4.8 and binutils 2.20 (which doesn't support avx2) in a chroot, container or virtual machine.

[joblo]

It's not so trivial anymore.

It might be worthwhile to focus the problem a little more. I suggest trying to compile with a lesser arch
like core2 to see if you can, then try to other arches until you identify the one that fails.

The problem seems to be with the compile environment but it could also be your CPU. If it's an AMD it
adds another dimension because their implementation of some features is incompatible with Intel.
The RELEASE_NOTES have some tips.

Won't help, he needs to update binutils. Even if he uses -march=core2 it will still emit AVX2 instructions and old GNU assembler will break the compilation -- the file responsible is algo/argon2/ar2/sj/scrypt-jane-mix_salsa64-avx2.h, it will emit that instruction no matter what -march compiler option is provided.

You can easily replicate that problem yourself with installing gcc 4.8 and binutils 2.20 (which doesn't support avx2) in a chroot, container or virtual machine.

Interesting. It seems you're right but only because the compiler was upgraded, otherwise the AVX2 code would have been skipped:

Code:

        #if (COMPILER_GCC >= 40700)
                #define X86_64ASM_AVX2

I agree the way forward would be to upgrade all of gcc's dependencies: rpm -q --requires gcc

[zhq123456]

Could you tell me a question, if I use centos6.3 successfully compiled, will generate a minerd file, and then I in other versions of the same machine on centos (without relying on) can run normally?

[virasog]

I'm taking a look at this now but I'm trying to figure out exactly what you have tried to do that is different than the other CPU miners out there looks like you might be trying to go into using it for some different algorithms and I'm not sure if that's even possible or worth the time that you are putting into it, or the time that others would put into using it.  We spend a lot of time beating the dead horses here and that is not getting anyone anywhere.

[dinamx]

I'm taking a look at this now but I'm trying to figure out exactly what you have tried to do that is different than the other CPU miners out there looks like you might be trying to go into using it for some different algorithms and I'm not sure if that's even possible or worth the time that you are putting into it, or the time that others would put into using it.  We spend a lot of time beating the dead horses here and that is not getting anyone anywhere.

well if you wonder what is different in his optimized version
the hash rate increases 20-30% at least in my case
for lot of algos, even for the new one tribus

That in itself is a triumph imho and for sure has greatly aided small miners who can only cpu mine or having a bunch of idle cpus
(even if not so profitable, maybe because people wish to join crypto party too? )

[joblo]

@joblo
Hey, I have a Ryzen that I'd like to test HW SHA on. But from what I understand you have to compile the sources yourself which is a bit beyond me. Any plans of adding windows binaries with this function compiled in?
Thanks.

I have not found a way to compile on Windows with SHA. I have tried 3 different mingw environments without success.
One user was successful but didn't share his procedure.

[Elder III]

AMD just announced Ryzen PRO with "Built-in AES 128-bit Encryption engine" so will that have any impact on mining?

I found the announcement, it would have been nice if you had posted the link.

https://www.amd.com/en/ryzen-pro

It's all marketing speak so I have no idea what it means technically. AES_NI is already a 128 bit encryption
and decryption engine so I have no idea what improvements are implied, maybe a faster implementation.

Another implied improvement is the mention of low-latency cache. Intel cache performance is currently better
than AMD so maybe the're catching up. That combined with the larger-than-Intel cache size could further give
an edge to AMD in the CPU market.

With Ryzen they have pulled ahead of Intel with threads, price, cache size, HW SHA. Between Threadripper
(4 channel DDR4, more threads, even bigger cache) and whatever improvements come in the Ryzen Pro there's
not much left where Intel has an edge.

It's nice to see some real competition in the CPU market. It will be interesting to see how Intel will respond.

On a bit of a tangent...

I am still curious about AVX2 performance on Ryzen. I have read that their implementation isn't native but I haven't
seen any performance comparisons. If anyone has a Ryzen I would be interested in a performance comparison of
AVX vs AVX2 on algos that have AVX2 optimizations.

If you can list the algos you would like tested I can try it out on both Ryzen R7 1700X and R5 1600 CPUs. Possibly on Threadripper too towards the end of summer if I talk myself into getting one.

[joblo]

I am still curious about AVX2 performance on Ryzen. I have read that their implementation isn't native but I haven't
seen any performance comparisons. If anyone has a Ryzen I would be interested in a performance comparison of
AVX vs AVX2 on algos that have AVX2 optimizations.

If you can list the algos you would like tested I can try it out on both Ryzen R7 1700X and R5 1600 CPUs. Possibly on Threadripper too towards the end of summer if I talk myself into getting one.

Lyra2v2, Lyra2z and Deep have the most AVX2 and should see the biggest improvement. All the X algos, qubit, timetravel
and a few others also have varying amounts of AVX2 code.

[Sanglotslongs2]

Hello, with a Intel CPU i7 3820 what is the best .exe files to use on windows ?
cpuminer-aes-avx.exe
cpuminer-aes-avx2.exe
cpuminer-aes-sse42.exe
cpuminer-sse2.exe
cpuminer-sse42.exe

?

[Elder III]

I am still curious about AVX2 performance on Ryzen. I have read that their implementation isn't native but I haven't
seen any performance comparisons. If anyone has a Ryzen I would be interested in a performance comparison of
AVX vs AVX2 on algos that have AVX2 optimizations.

If you can list the algos you would like tested I can try it out on both Ryzen R7 1700X and R5 1600 CPUs. Possibly on Threadripper too towards the end of summer if I talk myself into getting one.

Lyra2v2, Lyra2z and Deep have the most AVX2 and should see the biggest improvement. All the X algos, qubit, timetravel
and a few others also have varying amounts of AVX2 code.

Preliminary results:

Ryzen 1700X @ stock settings with 8 threads set in the batch file.

Lyra2v2 = 1001 Kh/s on avx2
             = 1016 Kh/s on avx
Timetravel = 799 Kh/s on avx2
                = 843 Kh/s on avx
X11          = 686 Kh/s on avx2
                = 711 Kh/s on avx
X17          = 294 Kh/s on avx2
                = 294 Kh/s on avx

X17 was the same for both options, but I was surprised to see avx beating avx2 for the other algorithms tested. My understanding was that it would likely be the other way around....