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

[joblo]

Tossed away everything I had done with cryptonight and started form scratch.
It took 2 hours to get my first share 180 KH/s.

V3.0.4 coming right up.

Done



Edit: Thanks Wolf0

Edit: I may have overstated the hash rate, more like 170.

I think the addition of cryptonight broke the core2 compile. I've disabled the download
link for now.

Edit: v3.0.5

https://drive.google.com/file/d/0B0lVSGQYLJIZTXdKVHNpdGRuTW8/view?usp=sharing

[joblo]

I am about to drop support for x86_64 target for two reasons. There is only one generation
of Intel 64 bit CPU it applies to no one has expressed an interest in it.

The x86_64 kernels will live on in cpuminer-multi and I will maintain the x86_64 kernels
ifthey are the best available for a particular algo.. Wil like be dropped in he next release.

[pallas]

Thanks Joblo, I'll test release 3.0.5 later today :-)

[joblo]

Thanks Joblo, I'll test release 3.0.5 later today :-)

I'm noticing cryptonight is about 10% slower than Wolf0's build. Let me know if you can
confirm it. I only made two significant changes that shouldn't have had such an impact.

Edit: never mind I found it. Will wait a while before a new release in case something else
pops up.

[joblo]

I found my hash!

X11 back up to peak, it has been down a bit for as couple of releases but it's back.
I also restored cryptonight performance to the same level as Wolf0.

v3.0.6 should be the best release yet, the most algos, the highest performance
and the widest HW support.

It's baking in the oven.

[joblo]

I found my hash!

X11 back up to peak, it has been down a bit for as couple of releases but it's back.
I also restored cryptonight performance to the same level as Wolf0.

v3.0.6 should be the best release yet, the most algos, the highest performance
and the widest HW support.

It's baking in the oven.

This out turned out well. I'm running out of excused to delay windows support.

https://drive.google.com/file/d/0B0lVSGQYLJIZSmFXUnZrdDFkTjg/view?usp=sharing

Edit: It has been observed that better performance is achieved when using the CPU name
instead of native for the arch argument. It doesn't seem to work for more recent CPU,
the only accepte dvalue is corei7-avx, rather than haswell, sandybridge etc.
On older CPUs such as the core2 it may make a difference. YMMV

[joblo]

I found my hash!

X11 back up to peak, it has been down a bit for as couple of releases but it's back.
I also restored cryptonight performance to the same level as Wolf0.

v3.0.6 should be the best release yet, the most algos, the highest performance
and the widest HW support.

It's baking in the oven.

This out turned out well. I'm running out of excused to delay windows support.

https://drive.google.com/file/d/0B0lVSGQYLJIZSmFXUnZrdDFkTjg/view?usp=sharing

Edit: It has been observed that better performance is achieved when using the CPU name
instead of native for the arch argument. It doesn't seem to work for more recent CPU,
the only accepte dvalue is corei7-avx, rather than haswell, sandybridge etc.
On older CPUs such as the core2 it may make a difference. YMMV

Just when I thought there was no more optimising to do I found 4% more in quark. I'll
see if it works on other algos.

[sp_]

Why is c11 slower than x11?

[joblo]

Why is c11 slower than x11?

I haven't been able to integrate the groestl AES_NI  optimizations yet. I'm having
that problem with many algos that use groestl. Only x11 and quark are working.
It could do wonders on some other algos especially groestl itself. If I can get it
working it coould be a 100% boost.

4% boost in quark coming due to implementation of fast reinit_groestl. Maybe it
will work in ccminer too. It' was simple, just clone init_groestl and remove the constant
initializations. It speeds up the init every time groestl is run. Just make sure to do
a full init the first time.

Edit: only worked for quark because quark runs twice in the chain. Only need to do a
reinit before the second run. Full init works but is slower. No init is even faster but
never finds blocks.

Some of my improvements have come from optimizing the ctx init, avoiding
doing it for nothing. I haven't looked a ccminer but there may be opportunities there.
If it works for you don't forget where you got the idea.

[sp_]

4% boost in quark coming due to implementation of fast reinit_groestl. Maybe it
will work in ccminer too. It' was simple, just clone init_groestl and remove the constant
initializations. It speeds up the init every time groestl is run. Just make sure to do
a full init the first time.

ccminer only to the init pass once per algo.  Until it finds a solution for all the hashfunctions

Edit: only worked for quark because quark runs twice in the chain. Only need to do a
reinit before the second run. Full init works but is slower. No init is even faster but
never finds blocks.

ccminer only run groestl once.

Why don't you share your code on github so I can add 100%  

[joblo]

4% boost in quark coming due to implementation of fast reinit_groestl. Maybe it
will work in ccminer too. It' was simple, just clone init_groestl and remove the constant
initializations. It speeds up the init every time groestl is run. Just make sure to do
a full init the first time.

ccminer only to the init pass once per algo.  Until it finds a solution for all the hashfunctions

Edit: only worked for quark because quark runs twice in the chain. Only need to do a
reinit before the second run. Full init works but is slower. No init is even faster but
never finds blocks.

ccminer only run groestl once.

In cpuminer it only appears once in the code but it runs twice.

Why don't you share your code on github so I can add 100%  
[/quote]

You can get it from google drive, help yourself. I do intend to get on github when
things settle down. There are lots of algos that need AES_NI implementations,
so far only groestl and echo. A couple more have SSE2 and the rest are dead slow.

Wolf's cryptonight is nice. CPU runs cool and performance is higher than the other algos
when compared with ccminer. For mining XMR a CPU is better.

[joblo]

Progress update.

I found 4% more hash in quark and I've tested some of the more obscure algos so another
3.0 update is coming before 3.1. I'll take anorther day to look for more low hanging fruit
and to a full suite of testing before releasing. I want this to be super stable.

Then I will start on windows, I promise.

V 3.0.7 almost ready.

Edit

I was checking some stats while testing and here is how much has been gained since
the project forked.

quark + 27%
qubit  + 36
x13    + 92
x15    + 76

It's come a long way.

[zTheWolfz]

a little speedup by better compiling:

Intel(R) Core(TM) i5 CPU         760  @ 2.80GHz
AES_NI: No.
SSE2: No, start mining without optimizations...
[....]
[2016-01-26 13:39:37] accepted: 7/7 (100.00%), 224.92 kH/s yes!

about 5-10% more by using this commandline (to be adapted to your own cpu):

./configure CFLAGS="-march=nehalem -Ofast -DNO_AES_NI" CXXFLAGS=$CFLAGS --with-crypto --with-curl

and remember to "make clean"

Think any of this would help me to get a completion compile on the problem AMD system I posted about a few pages back?

https://bitcointalk.org/index.php?topic=1326803.msg13657053#msg13657053

the cpu doesn't look to have AES: did you try compiling with -DNO_AES_NI?

I just tried that and still get errors. maybe someone could help me on this one I'm a real noob with this stuff.

./configure CFLAGS="-march=??? -Ofast" CXXFLAGS=$CFLAGS --with-crypto --with-curl ??

Code:

Processors Information
-------------------------------------------------------------------------

Processor 1			ID = 0
	Number of cores		8 (max 8)
	Number of threads	8 (max 8)
	Name			AMD Opteron 6124
	Codename		Magny-Cours
	Specification		AMD Opteron(tm) Processor 6124 HE
	Package 		Socket G34 (1974)
	CPUID			F.9.1
	Extended CPUID		10.9
	Core Stepping		HY-D1
	Technology		45 nm
	TDP Limit		74.4 Watts
	Core Speed		1039.8 MHz
	Multiplier x Bus Speed	4.0 x 259.9 MHz
	HT Link speed		2859.4 MHz
	Stock frequency		1800 MHz
	Instructions sets	MMX (+), 3DNow! (+), SSE, SSE2, SSE3, SSE4A, x86-64, AMD-V
	L1 Data cache		8 x 64 KBytes, 2-way set associative, 64-byte line size
	L1 Instruction cache	8 x 64 KBytes, 2-way set associative, 64-byte line size
	L2 cache		8 x 512 KBytes, 16-way set associative, 64-byte line size
	L3 cache		10 MBytes, 96-way set associative, 64-byte line size
	FID/VID Control		yes
	FID range		4.0x - 9.0x
	Max VID			1.050 V
	# of P-States		5
	P-State			FID 0x2 - VID 0x2F - IDD 6 (9.00x - 0.962 V)
	P-State			FID 0x0 - VID 0x31 - IDD 6 (8.00x - 0.937 V)
	P-State			FID 0x10A - VID 0x34 - IDD 5 (6.50x - 0.900 V)
	P-State			FID 0x106 - VID 0x35 - IDD 4 (5.50x - 0.887 V)
	P-State			FID 0x100 - VID 0x37 - IDD 4 (4.00x - 0.862 V)


	Package Type		0x3
	Model			25
	String 1		0x0
	String 2		0x0
	Page			0x0
	CmpCap			8
	ApicIdCoreSize		16
	TDC Limit		48 Amps
	Boosted P-States	0
	Max non-turbo ratio	9.00x
	Max turbo ratio		9.00x
	Max CPU COF		18
	Core Performance Boost	no
	P-State			0, FID 0x2 - VID 0x2F (9.00x - 0.962 V)
	P-State			1, FID 0x0 - VID 0x31 (8.00x - 0.937 V)
	P-State			2, FID 0x10A - VID 0x34 (6.50x - 0.900 V)
	P-State			3, FID 0x106 - VID 0x35 (5.50x - 0.887 V)
	P-State			4, FID 0x100 - VID 0x37 (4.00x - 0.862 V)
	Attached device		PCI device at bus 0, device 24, function 0
	Attached device		PCI device at bus 0, device 24, function 1
	Attached device		PCI device at bus 0, device 24, function 2
	Attached device		PCI device at bus 0, device 24, function 3
	Attached device		PCI device at bus 0, device 24, function 4
	TSC			2339.8 MHz

	Temperature 0		29°C (84°F) [0xE8] (Core #0)
	Power 0			27.60 W (Package)

First post your console session with your command line and error messages. *
And stick with the instructions in the READme.md file.

You've got an AMD Opteron 64 so try to find a match
https://gcc.gnu.org/onlinedocs/gcc-4.9.2/gcc/i386-and-x86-64-Options.html

* I can be prickly, so be careful with your questions, This is your first warning.
    I'm not trying to be mean, it just happens, so don't take it personally. Well
    formulated questions with supporting data and a demonstration you did
    your best to solve it will always get the best response (and not just from me).

Getting closer it does start but hangs at checking the CPU.

Code:

h8qg6@h8qg6-VirtualBox:~/Downloads/cpuminer-opt-3.0.3$ ./cpuminer -a x11 -o stratum+tcp://x11.usa.nicehash.com:3336 -u 18tvS3deKZK5q4eTtPRWYeEMWmGmuErHgz.H8QG6 -p d=0.01 -t 1

         **********  cpuminer-opt  v3.0.3  *********** 
     BTC donation address: 12tdvfF7KmAsihBXQXynT6E6th2c2pByTT
     Forked from TPruvot's cpuminer-multi-1.2pre with credits
     to Lucas Jones, elmad, palmd, djm34, pooler, ig0tik3d,
     Wolf0 and Jeff Garzik.

Only quark and qubit support SSE at this time. Whenever mining
other algos with an SSE2 capable CPU the generic x86_64 miner
will be used.

Checking CPU capatibility...

Ok now I have something towork with. I have a workaround that requires some code changes.
I will explain the changes in detail so you shouldn't have any problems.

First backup the original.

Edit file cpu-miner.c and make the folllwing changes to block the actual check and hard code the
result.

Replace any calls to has_aesni(), has_sse2 with true
There might be a couple of each.

ex:
change "cpu_aesni = has_aesni();" to "cpu_aesni = true;"

You can also set each one to false if the feature isn't supported but try true for both to start.

remove the call to "check_cpu_capability()" on line 3157. Just delete the line.

Save, compile and run. Let me know how it goes.

Edit:It seems your CPU does not have AES_NI to you will likely have to set cpu_aesni to false
in cpu-miner.c and use -DNO_AES_NI.

Got it working some what just hasn't shown any hashrate yet.

[zTheWolfz]

May have a better solution for these problems. Order 4x of these tonight AMD Opteron 6276 to replace the 4x 6124 HE.
http://www.cpu-world.com/CPUs/Bulldozer/AMD-Opteron%206276.html
 

[pallas]

May have a better solution for these problems. Order 4x of these tonight AMD Opteron 6276 to replace the 4x 6124 HE.
http://www.cpu-world.com/CPUs/Bulldozer/AMD-Opteron%206276.html
 

64 cores with aes is a good start ;-)
I wonder how they perform in cryptonight.

[joblo]

Got it working some what just hasn't shown any hashrate yet.

Was the CPU working or just idling? Did you try other pools or algos?

I've streamlined the check in v3.0.7. The check for SSE2 wasn't working and
with the plan to drop seperate generic x86_64 target the SSE check isn't
needed anymore.

Startimg in 3.0.7 it will display seperatety whether the CPU and build
support AES_NI and select the appropriate target. The startup display
will also be directly linked to the target selection, previously there were
two seperate checks.

[joblo]

Download 3.0.7 now. More stable than 3.0.6.

https://drive.google.com/file/d/0B0lVSGQYLJIZWjJpXzAtemRaUWc/view?usp=sharing

[zTheWolfz]

May have a better solution for these problems. Order 4x of these tonight AMD Opteron 6276 to replace the 4x 6124 HE.
http://www.cpu-world.com/CPUs/Bulldozer/AMD-Opteron%206276.html
 

64 cores with aes is a good start ;-)
I wonder how they perform in cryptonight.

Should be here Wednesday or Thursday then I have to break it down and replace. Maybe some test before the weekend.
Doing my research now to see if there is anything extra I may need to do like Update or change bios.
I have the [H] overclocking bios installed atm.

[zTheWolfz]

Got it working some what just hasn't shown any hashrate yet.

Was the CPU working or just idling? Did you try other pools or algos?

I've streamlined the check in v3.0.7. The check for SSE2 wasn't working and
with the plan to drop seperate generic x86_64 target the SSE check isn't
needed anymore.

Startimg in 3.0.7 it will display seperatety whether the CPU and build
support AES_NI and select the appropriate target. The startup display
will also be directly linked to the target selection, previously there were
two seperate checks.

Was a very low load on the CPU's no matter how many cores used. If a big amount was used the OS seemed to lock up.
But I think that may have more to do the the VM setup than miner software, but can't say for sure either way.
Yes I tried other algos, but not different pools yet. I'll give your v3.0.7 a go a later on today & post back the results.

[joblo]

It didn't take long to hit a brick wall with windows.

static inline void transform(cubehashParm *sp )

Expected '(' to follow 'inline'

WTF?

 I guess that settles it.