If you're not the creator, why are you redistributing it? It has to be publicly available at that point.
What do you mean "why"? To scam dumb people for their money, of course. |
|
|
|
I usually see these on dodgy sites - THIS SIMPLE TRICK MADE HIM $10,000 A WEEK AND BANKS ARE FURIOUS!!!
Yeah, same ol' same ol'. There are templates for these pseudo-websites, you just fill in/change a few words and you got "a bestseller". There are thousands and thousands websites identical to this scheme, all lies and bullshit, only to make you think "what if?" and pull your wallet. |
|
|
|
The guy who created the book is probably 30.
I'll tell you how I made $10k in a month (I'm 15.)
I opened up BurpSuite. I read up on programming languages and I found a few exploits on a few sites.
Within a few days, I made $10,000 USD from bug bounties.
You don't need a fucking book. It will just tell you to find a niche & make a site about it.
Utter bullshit.
Cool story. |
|
|
|
|
Imho the only way a 15yo makes $800 a week is by scamming people. Needs more spanking, less free time, more homework and more home chores. That would cure the kid.
|
|
|
|
Seriously, I've asked nicely first but then he blocked me on Skype and started to ignore my emails...
I don't think I'm the only one to whom this happened so I want to warn people about it.
He promises people the account are legit and that you can change the password but it's a total lie.
The account are hacked and that's why they don't last more than a few days cause he sell those to a lot of people and then those people start to add, delete existing users and then the actual owner of the account notice it and change the password.
It happened to me every time for the Netflix and Hulu account he sold me...
I understand that but if you want other people to listen you can't just run around like maniac and scream "he is a scammer" because nobody will take you seriously, they will think you are just mad and acting out of emotions. What you just wrote is a perfectly understandable and decently written description of the problem. Few bold words of a normal fontsize will be much more effective than a long post full of superbig red font. Just my 2 cents. |
|
|
|
|
I bought one .edu, the seller delivered and the e-mail works. In particular I was looking for an .edu e-mail that would allow me to obtain an AWS Educate starter account. I tried some other .edu e-mails before and none of them worked for this purpose (Amazon somehow didn't see them as valid .edu e-mails). This e-mail from Zoomer does work! Most people probably don't care about some AWS Educate starter account but the fact that it passed Amazon's verification might indicate it will work elsewhere as well.
|
|
|
|
 This is how he treats his customers... To anyone considering buying from this guy, think twice. Well, I don't know if he is a scammer or not but you can either keep your cool and asking for replacements or lose your cool and start this juvenile megabig-font crusade with urgency as if there was no tomorrow. But you can't do the latter and expect that the former will still work, if I was the seller and you behaved this way I would cut you off as well. If you buy something in the store, it doesn't work, you go back to claim the replacement, they maybe don't react the way you expected, and you start acting out and offending people then they will simply call the security (and possibly even the police) and kick you out of there (and possibly file a criminal complaint against you for a verbal attack, threatening, whatever), regardless of the fact that you were originally in your right. Then, after you sort out your newly acquired criminal fame, you can file a rightful civil lawsuit against the store but then if you say one single bad word in front of the judge he will have you arrested for contempt of court, reject your claim and fine you in such an amount that you will regret not keeping your mouth shut. All the while your original claim about a non-working product bought in the story was justified. |
|
|
|
|
I don't want to start a flame war or anything, but which one of those three is in your opinion the best in the terms of profitability for ETH mining and why?
A side question: does anyone know about any ETH proxy that works with Nicehash?
|
|
|
|
That's what I was referring to when I wrote "compiler produced AVX2". AVX(2) provides SIMD instructions an it's unlikely something
like a printf would use it. A memcpy wouldn't use it because of the overhead of loading/storing the data to/from the ymm regs.
It's only useful for vector arith, and apparently the compiler isn't smart enough to convert conventionally coded array processing
loops to AVX2. I'm not even sure *if* the compiler can optimize in this fashion, the existance of so much hand coded AVX2 suggests
otherwise.
I wasn't talking about what is or isn't likely but what is theoretically possible. With -O flags and -march=native the compiler can use all instruction sets it has to its disposal and if and how it decides to translate a higher language into a machine code is only up to it, so if it feels like it it will use AVX2 for printf and that's that. Since we're playing semantic games would you care to explain your concerns with my use of the term cross compiling?
IMO cross compiling can mean any compilation not done on the target machine and not executable on the build machine.
Of course you can have any opinion you want, that's your freedom, but that doesn't change the fact that cross compiling is when a different host/target is desired. Furthermore, your own personal definition doesn't fit this narrative either because we were talking about compiling on avx2 cpu and producing sse code. In such a case the code is executable on the build machine so not even by your very own definition would it be cross compiling. And my comment about you maybe using a core2 was based on the symptoms you decribed and that some server CPUs can
be optimized for efficiency by removing/disabling unneeded features like floating point, AES or AVX.
No, core2 is not and never was a server cpu and I was very clearly talking about real servers in the data centers, not some home desktop computers that you call "servers". I started the whole discussion with complaining about Nicehash not working on my servers anymore and later I confirmed someone else's suggestion that it may be because I am using the same BTC address on all my machines, there are tens of them and each has a different IP. That in itself basically rules out any possible doubt because I could hardly have tens of computers in my living room and even then no ordinary household ISP would give me tens of public IPs. They don't even give a single public IP per household around here anymore by default, you have to ask for it and pay for it. And even furthermore, if I was to have tens of mining "servers" (i.e. desktops) in my home, full of power-hungry GPUs, I wouldn't be able to power them all because an ordinary household has a 20A fuse, that would translate into some 3500 Watts if the power supplies had 80% efficiency. You can maybe power 5 relatively decent mining machines with that but definitely not tens. Now I don't want to sound harsh but this is enough. I very much support your efforts and try to help you and am eager to discuss any technical topic but I will not play this game, especially not at this level. I don't actually mind word games but here you have no real argument in any of what you just said and yet it seems as if you were trying to kind of get back at me regardless and that is just pointless and for you humiliating and for both of us a waste of time. I wouldn't dare to say a word about anything related to your AVX/AVX2 optimizing but when I do say something I try to pay attention on my wording so if you want to look for my errors you would have to try harder but if you really did that you would only confirm a very immature personality. I merely, totally innocently, rightfully and truthfully pointed out, in the brackets even, that cross compiling is something else than what you said and that triggered you and turned you into a rogue word warrior trying to reclaim his supremacy? Come on, kids do that, not adults. |
|
|
|
There are two copies of scrypt-jane, one copy used by scrypt-jane itself and the other used by
argon2. Only the argon2 version is optimized and was taken from the argon2 branch of multi. The other copy was taken
from the windows branch and used by scrypt-jane. At some point I intend to integrate the optimized version for use
by scrypt-jane algo, but it's not a very popular algo. Maybe you could try compiling the argon2 branch of multi to see
if it also has the same error.
You are absolutely right. If I build the argon2 branch with the default build.sh (LTO and the other flags disabled) it compiles, but if I do my uncommenting I get the same error: /tmp/ccAcmNi5.ltrans14.ltrans.o: In function `scrypt_ROMix_sse2.lto_priv.317':
/root/z/cpuminer-multi/ar2/sj/scrypt-jane-romix-template.h:89: undefined reference to `scrypt_ChunkMix_sse2'
/root/z/cpuminer-multi/ar2/sj/scrypt-jane-romix-template.h:89: undefined reference to `scrypt_ChunkMix_sse2'
/root/z/cpuminer-multi/ar2/sj/scrypt-jane-romix-template.h:89: undefined reference to `scrypt_ChunkMix_sse2'
/root/z/cpuminer-multi/ar2/sj/scrypt-jane-romix-template.h:89: undefined reference to `scrypt_ChunkMix_sse2'
/root/z/cpuminer-multi/ar2/sj/scrypt-jane-romix-template.h:89: undefined reference to `scrypt_ChunkMix_sse2'
/tmp/ccAcmNi5.ltrans14.ltrans.o:/root/z/cpuminer-multi/ar2/sj/scrypt-jane-romix-template.h:89: more undefined references to `scrypt_ChunkMix_sse2' follow
/tmp/ccAcmNi5.ltrans14.ltrans.o: In function `scrypt_ROMix_ssse3.lto_priv.316':
/root/z/cpuminer-multi/ar2/sj/scrypt-jane-romix-template.h:89: undefined reference to `scrypt_ChunkMix_ssse3'
/root/z/cpuminer-multi/ar2/sj/scrypt-jane-romix-template.h:89: undefined reference to `scrypt_ChunkMix_ssse3'
/root/z/cpuminer-multi/ar2/sj/scrypt-jane-romix-template.h:89: undefined reference to `scrypt_ChunkMix_ssse3'
/root/z/cpuminer-multi/ar2/sj/scrypt-jane-romix-template.h:89: undefined reference to `scrypt_ChunkMix_ssse3'
/root/z/cpuminer-multi/ar2/sj/scrypt-jane-romix-template.h:89: undefined reference to `scrypt_ChunkMix_ssse3'
/tmp/ccAcmNi5.ltrans14.ltrans.o:/root/z/cpuminer-multi/ar2/sj/scrypt-jane-romix-template.h:89: more undefined references to `scrypt_ChunkMix_ssse3' follow
/tmp/ccAcmNi5.ltrans14.ltrans.o: In function `scrypt_ROMix_avx.lto_priv.315':
/root/z/cpuminer-multi/ar2/sj/scrypt-jane-romix-template.h:89: undefined reference to `scrypt_ChunkMix_avx'
/root/z/cpuminer-multi/ar2/sj/scrypt-jane-romix-template.h:89: undefined reference to `scrypt_ChunkMix_avx'
/root/z/cpuminer-multi/ar2/sj/scrypt-jane-romix-template.h:89: undefined reference to `scrypt_ChunkMix_avx'
/root/z/cpuminer-multi/ar2/sj/scrypt-jane-romix-template.h:89: undefined reference to `scrypt_ChunkMix_avx'
/root/z/cpuminer-multi/ar2/sj/scrypt-jane-romix-template.h:89: undefined reference to `scrypt_ChunkMix_avx'
/tmp/ccAcmNi5.ltrans14.ltrans.o:/root/z/cpuminer-multi/ar2/sj/scrypt-jane-romix-template.h:89: more undefined references to `scrypt_ChunkMix_avx' follow
/tmp/ccAcmNi5.ltrans14.ltrans.o: In function `scrypt_ROMix_xop.lto_priv.314':
/root/z/cpuminer-multi/ar2/sj/scrypt-jane-romix-template.h:89: undefined reference to `scrypt_ChunkMix_xop'
/root/z/cpuminer-multi/ar2/sj/scrypt-jane-romix-template.h:89: undefined reference to `scrypt_ChunkMix_xop'
/root/z/cpuminer-multi/ar2/sj/scrypt-jane-romix-template.h:89: undefined reference to `scrypt_ChunkMix_xop'
/root/z/cpuminer-multi/ar2/sj/scrypt-jane-romix-template.h:89: undefined reference to `scrypt_ChunkMix_xop'
/root/z/cpuminer-multi/ar2/sj/scrypt-jane-romix-template.h:89: undefined reference to `scrypt_ChunkMix_xop'
/tmp/ccAcmNi5.ltrans14.ltrans.o:/root/z/cpuminer-multi/ar2/sj/scrypt-jane-romix-template.h:89: more undefined references to `scrypt_ChunkMix_xop' follow
/tmp/ccAcmNi5.ltrans14.ltrans.o: In function `scrypt_ROMix_avx2.lto_priv.313':
/root/z/cpuminer-multi/ar2/sj/scrypt-jane-romix-template.h:89: undefined reference to `scrypt_ChunkMix_avx2'
/root/z/cpuminer-multi/ar2/sj/scrypt-jane-romix-template.h:89: undefined reference to `scrypt_ChunkMix_avx2'
/root/z/cpuminer-multi/ar2/sj/scrypt-jane-romix-template.h:89: undefined reference to `scrypt_ChunkMix_avx2'
/root/z/cpuminer-multi/ar2/sj/scrypt-jane-romix-template.h:89: undefined reference to `scrypt_ChunkMix_avx2'
/root/z/cpuminer-multi/ar2/sj/scrypt-jane-romix-template.h:89: undefined reference to `scrypt_ChunkMix_avx2'
/tmp/ccAcmNi5.ltrans14.ltrans.o:/root/z/cpuminer-multi/ar2/sj/scrypt-jane-romix-template.h:89: more undefined references to `scrypt_ChunkMix_avx2' follow
collect2: error: ld returned 1 exit status
Makefile:881: recipe for target 'cpuminer' failed
|
|
|
|
The GRS macros have been a pain for me since day 1. I had to make changes so they could be included in
multiple algos. It looks like LTO likes them as much as I do. The problem is they're faster than the SPH versions.
Do the errors occur only with -flto or do they also occur with gcc 5.4.0 with my build.sh options? This code is only
compiled for SSE2 builds, I may have to drop support for it or degrade performance by using SPH functions in the future.
With your build.sh, without messing with LTO, I haven't seen any problem. But that was on avx/avx2 cpus. On non-avx cpu I compiled only today and that was straight with LTO. So now I did just that - downloaded 3.4.0 and ran the untouched build.sh. No error. But look at the remarkable speed difference when I compare it with 3.4.0 built with tpruvot's uncommented (LTO enabled) build.sh: root@xxx:~/z/cpuminer-opt-3.4.0# ./cpuminer -a lyra2re --benchmark
********** cpuminer-multi 1.2-dev ***********
A CPU miner with multi algo support and optimized for CPUs
with AES_NI and AVX extensions.
BTC donation address: 12tdvfF7KmAsihBXQXynT6E6th2c2pByTT
Forked from TPruvot's cpuminer-multi with credits
to Lucas Jones, elmad, palmd, djm34, pooler, ig0tik3d,
Wolf0, Jeff Garzik and Optiminer.
CPU: Intel(R) Xeon(R) CPU X5570 @ 2.93GHz
CPU features: SSE2
SW built on Aug 5 2016 with GCC 5.4.0
SW features: SSE2
Algo features: SSE2 AES AVX AVX2
AES not available, starting mining with SSE2 optimizations...
[2016-08-05 19:03:55] 16 miner threads started, using 'lyra2re' algorithm.
[2016-08-05 19:03:56] Total: 983.04 kH, 872.93 kH/s
[2016-08-05 19:04:00] Total: 3218.22 kH, 927.88 kH/s
[2016-08-05 19:04:06] Total: 4515.37 kH, 896.32 kH/s
[2016-08-05 19:04:10] Total: 4143.24 kH, 907.80 kH/s
root@xxx:~/cpuminer-opt-sse# ./cpuminer -a lyra2re --benchmark
********** cpuminer-multi 1.2-dev ***********
A CPU miner with multi algo support and optimized for CPUs
with AES_NI and AVX extensions.
BTC donation address: 12tdvfF7KmAsihBXQXynT6E6th2c2pByTT
Forked from TPruvot's cpuminer-multi with credits
to Lucas Jones, elmad, palmd, djm34, pooler, ig0tik3d,
Wolf0, Jeff Garzik and Optiminer.
CPU: Intel(R) Xeon(R) CPU X5570 @ 2.93GHz
CPU features: SSE2
SW built on Aug 5 2016 with GCC 5.4.0
SW features: SSE2
Algo features: SSE2 AES AVX AVX2
AES not available, starting mining with SSE2 optimizations...
[2016-08-05 19:04:33] 16 miner threads started, using 'lyra2re' algorithm.
[2016-08-05 19:04:34] Total: 786.43 kH, 753.31 kH/s
[2016-08-05 19:04:38] Total: 3006.12 kH, 993.53 kH/s
[2016-08-05 19:04:43] Total: 4691.35 kH, 988.85 kH/s
[2016-08-05 19:04:48] Total: 4694.50 kH, 994.89 kH/s
[2016-08-05 19:04:53] Total: 4881.65 kH, 998.29 kH/s
it's pity you are not using a versioning system where you would have your improvements on top of tpruvot's. If you used github you could have had automatic linux/windows building done on travis-ci after every commit, including an automatic publishing of the binaries on github.
I can't figure out how to get my code into github. It must be simple but I haven't figured it out. I have an account and played with it using the tutorial but I'm stuck. I also want to make a gradual transition so I don't get bogged down trying to figure out github when I'm trying to focus on the miner. The first phase is to continue development offline and upload releases. I think this should be of help: https://help.github.com/articles/adding-an-existing-project-to-github-using-the-command-line/. Or, to make you feel better, you can always do what I did when I didn't know how to do this same thing: I created an empty repo on github, then git clone it locally, then copied all the files into the directory, then git add it and then just git push it back to github  If I remember correctly when you try to push it the first time it will setup your github credentials, if you don't want it to ask you for your password every single time you can use git config credential.helper store and it will save it and remember. As for travis, it requires .travis.yml file to be present. This file already exists and it's quite self-explanatory if you look into it. Travis for public opensource projects is free and if you hook it up it will then automatically build after every commit you make. Those builds, for linux, are excellent in that they independently confirm your code is compilable (which you already knew anyway, but still..) but I wouldn't try to gather and publish the binaries, though. I wouldn't publish binaries for linux at all, you would just get into troubles and people are already used to compile themselves. But you can also setup building with win64 target using mingw and those I would publish so that you (the people) don't rely on some unknown third party. I don't want to spread paranoia or anything but you never know what did they do to the code, if they added something.. You can also get an .edu email for $2 from here: https://bitcointalk.org/index.php?topic=1321638.0 which will (among others) allow you to get a github student pack for free, which in turn will (among others) give you access to travis-ci.com (as opposite to travis-ci.org) which is for private building (normally it would cost money, this way it would be free) and that you could use as a sandbox maybe.. But if you are serious about github there is one crucial decision you will have to make, not necessarily now. Whether you want to keep working on top of years old tpruvot's code and keep ignoring all changes and possible improvements he has done ever since you took his code and he will keep doing in the future, or if you actually extract all your changes from the old code you took, fork his repo and put them on top of it. That way you will stay in sync with him and will keep your changes being reapplied after every his commit. This would require quite a lot of work, the part where you extract your changes, it's just about making a diff, fast-forward his code to the recent and reapplying the diff. Maybe someone young and enthusiastic could do that? Unfortunately I am old and tired so I am out of the game. Theoretically yes if there exists any earlier executed code that contained compiler produced AVX2 instructions from regular source.
That isn't likely since the capabilities check is done ealy in main.
Let's not forget that we ask gcc to compile and optimize (all those -O2 -O3 -Ofast) for the cpu it's being run on. So regardless whether you actually include any explicit AVX/AVX2 assembler in the code, even a simple printf("hi"); may produce AVX2 instruction(s) if the compiler feels like it. That's the whole point of the compiler compiling for the given cpu (-march=native) - it's allowed to use all the capabilities (and thus instruction sets) of the cpu. EDIT: oh, Fuzzbawls was faster. Btw, Fuzzbawls, you wouldn't get by "ok" with strictly using the github website, you would commit a suicide. Editing file directly via the github website is possible but it's so painful, don't even try it. Let alone everytime you click on "save", of every file you changed, it creates a new commit, so then you can't even review your changes easily. Horrible, terrible. But for non-commandline people many IDEs have git integrated so you don't have to use git but you really would very hardly survive by using just the github website. |
|
|
|
Is this a method you created?
If no, what gives you the right to sell it?
You could ask the same question 99% of the sellers on this forum. |
|
|
|
I just want to make sure I understand the problem definition
- multi is faster with -flto
- multi without -flto is slower than identically compiled opt
- multi with -flto is faster than pre-avx2 compiled without -flto
- opt fails to compile with gcc 5.4.0 with -flto
- -flto compiles with gcc 4.8.4 with no effect in performance.
The significant points are:
- flto is faster with gcc 5.4.0
- code that compiles with -flto using gcc 4.8.4 fails to compile using gcc 5.4.0.
Yes I think you got all those points right. Well, we would have to deal with 5.4.0 sooner or later anyway, it's not going anywhere. The code that fails to compile is pretty ugly. It uses asm function pointers to select targets at compile time.
I've never seen anything like this so it will take a while to understand what is going on. It looks like the code is
self contained and the error doesn't seem to be related to missing libraries.
Ugly or ingenious, crazy anyway. It took me a while to figure where exactly was the error coming from I just looked at the current tpruvot's and this culprit file scrypt-jane-romix-template.h is a bit different now. You know, it's pity you are not using a versioning system where you would have your improvements on top of tpruvot's. If you used github you could have had automatic linux/windows building done on travis-ci after every commit, including an automatic publishing of the binaries on github. |
|
|
|
Excellent work. The easiest way to block the compile error is to comment out the source dir for argon2 and remove the registration
call for argon2 in algo-gate-api.c:register_algo_gate. You can easilly remove any algo this way.
You have demonstrated that LTO improves performance with the new compiler but has some incompatibilities with the existing
argon2 code. I will investigate argon2 to try to solve it.
I am glad I could help. When I was compiling on Xeon X5570 which is neither avx/avx2 I got also this error at the final link (the compiler and the source code was the same, just a different cpu): /tmp/ccmS1O9H.ltrans19.ltrans.o: In function `grsoQ1024ASM':
<artificial>:(.text+0xa530): undefined reference to `grsoT0'
<artificial>:(.text+0xa538): undefined reference to `grsoT1'
<artificial>:(.text+0xa54a): undefined reference to `grsoT0'
<artificial>:(.text+0xa552): undefined reference to `grsoT1'
<artificial>:(.text+0xa564): undefined reference to `grsoT2'
<artificial>:(.text+0xa56c): undefined reference to `grsoT3'
<artificial>:(.text+0xa57e): undefined reference to `grsoT2'
<artificial>:(.text+0xa586): undefined reference to `grsoT3'
<artificial>:(.text+0xa598): undefined reference to `grsoT4'
<artificial>:(.text+0xa5a0): undefined reference to `grsoT5'
<artificial>:(.text+0xa5b2): undefined reference to `grsoT4'
<artificial>:(.text+0xa5ba): undefined reference to `grsoT5'
<artificial>:(.text+0xa5d0): undefined reference to `grsoT6'
<artificial>:(.text+0xa5d8): undefined reference to `grsoT7'
<artificial>:(.text+0xa5ea): undefined reference to `grsoT6'
<artificial>:(.text+0xa5f2): undefined reference to `grsoT7'
<artificial>:(.text+0xa600): undefined reference to `grsoT0'
<artificial>:(.text+0xa608): undefined reference to `grsoT1'
<artificial>:(.text+0xa61a): undefined reference to `grsoT0'
<artificial>:(.text+0xa622): undefined reference to `grsoT1'
<artificial>:(.text+0xa634): undefined reference to `grsoT2'
<artificial>:(.text+0xa63c): undefined reference to `grsoT3'
<artificial>:(.text+0xa64e): undefined reference to `grsoT2'
<artificial>:(.text+0xa656): undefined reference to `grsoT3'
<artificial>:(.text+0xa668): undefined reference to `grsoT4'
<artificial>:(.text+0xa670): undefined reference to `grsoT5'
<artificial>:(.text+0xa682): undefined reference to `grsoT4'
<artificial>:(.text+0xa68a): undefined reference to `grsoT5'
<artificial>:(.text+0xa6a0): undefined reference to `grsoT6'
<artificial>:(.text+0xa6a8): undefined reference to `grsoT7'
<artificial>:(.text+0xa6ba): undefined reference to `grsoT6'
<artificial>:(.text+0xa6c2): undefined reference to `grsoT7'
<artificial>:(.text+0xa730): undefined reference to `grsoT0'
<artificial>:(.text+0xa738): undefined reference to `grsoT1'
<artificial>:(.text+0xa74a): undefined reference to `grsoT0'
<artificial>:(.text+0xa752): undefined reference to `grsoT1'
<artificial>:(.text+0xa764): undefined reference to `grsoT2'
<artificial>:(.text+0xa76c): undefined reference to `grsoT3'
<artificial>:(.text+0xa77e): undefined reference to `grsoT2'
<artificial>:(.text+0xa786): undefined reference to `grsoT3'
<artificial>:(.text+0xa798): undefined reference to `grsoT4'
<artificial>:(.text+0xa7a0): undefined reference to `grsoT5'
<artificial>:(.text+0xa7b2): undefined reference to `grsoT4'
<artificial>:(.text+0xa7ba): undefined reference to `grsoT5'
<artificial>:(.text+0xa7d0): undefined reference to `grsoT6'
<artificial>:(.text+0xa7d8): undefined reference to `grsoT7'
<artificial>:(.text+0xa7ea): undefined reference to `grsoT6'
<artificial>:(.text+0xa7f2): undefined reference to `grsoT7'
<artificial>:(.text+0xa800): undefined reference to `grsoT0'
<artificial>:(.text+0xa808): undefined reference to `grsoT1'
<artificial>:(.text+0xa81a): undefined reference to `grsoT0'
<artificial>:(.text+0xa822): undefined reference to `grsoT1'
<artificial>:(.text+0xa834): undefined reference to `grsoT2'
<artificial>:(.text+0xa83c): undefined reference to `grsoT3'
<artificial>:(.text+0xa84e): undefined reference to `grsoT2'
<artificial>:(.text+0xa856): undefined reference to `grsoT3'
<artificial>:(.text+0xa868): undefined reference to `grsoT4'
<artificial>:(.text+0xa870): undefined reference to `grsoT5'
<artificial>:(.text+0xa882): undefined reference to `grsoT4'
<artificial>:(.text+0xa88a): undefined reference to `grsoT5'
<artificial>:(.text+0xa8a0): undefined reference to `grsoT6'
<artificial>:(.text+0xa8a8): undefined reference to `grsoT7'
<artificial>:(.text+0xa8ba): undefined reference to `grsoT6'
<artificial>:(.text+0xa8c2): undefined reference to `grsoT7'
<artificial>:(.text+0xa930): undefined reference to `grsoT0'
<artificial>:(.text+0xa938): undefined reference to `grsoT1'
<artificial>:(.text+0xa94a): undefined reference to `grsoT0'
<artificial>:(.text+0xa952): undefined reference to `grsoT1'
<artificial>:(.text+0xa964): undefined reference to `grsoT2'
<artificial>:(.text+0xa96c): undefined reference to `grsoT3'
<artificial>:(.text+0xa97e): undefined reference to `grsoT2'
<artificial>:(.text+0xa986): undefined reference to `grsoT3'
<artificial>:(.text+0xa998): undefined reference to `grsoT4'
<artificial>:(.text+0xa9a0): undefined reference to `grsoT5'
<artificial>:(.text+0xa9b2): undefined reference to `grsoT4'
<artificial>:(.text+0xa9ba): undefined reference to `grsoT5'
<artificial>:(.text+0xa9d0): undefined reference to `grsoT6'
<artificial>:(.text+0xa9d8): undefined reference to `grsoT7'
<artificial>:(.text+0xa9ea): undefined reference to `grsoT6'
<artificial>:(.text+0xa9f2): undefined reference to `grsoT7'
<artificial>:(.text+0xaa00): undefined reference to `grsoT0'
<artificial>:(.text+0xaa08): undefined reference to `grsoT1'
<artificial>:(.text+0xaa1a): undefined reference to `grsoT0'
<artificial>:(.text+0xaa22): undefined reference to `grsoT1'
<artificial>:(.text+0xaa34): undefined reference to `grsoT2'
<artificial>:(.text+0xaa3c): undefined reference to `grsoT3'
<artificial>:(.text+0xaa4e): undefined reference to `grsoT2'
<artificial>:(.text+0xaa56): undefined reference to `grsoT3'
<artificial>:(.text+0xaa68): undefined reference to `grsoT4'
<artificial>:(.text+0xaa70): undefined reference to `grsoT5'
<artificial>:(.text+0xaa82): undefined reference to `grsoT4'
<artificial>:(.text+0xaa8a): undefined reference to `grsoT5'
<artificial>:(.text+0xaaa0): undefined reference to `grsoT6'
<artificial>:(.text+0xaaa8): undefined reference to `grsoT7'
<artificial>:(.text+0xaaba): undefined reference to `grsoT6'
<artificial>:(.text+0xaac2): undefined reference to `grsoT7'
<artificial>:(.text+0xab30): undefined reference to `grsoT0'
<artificial>:(.text+0xab38): undefined reference to `grsoT1'
<artificial>:(.text+0xab4a): undefined reference to `grsoT0'
<artificial>:(.text+0xab52): undefined reference to `grsoT1'
<artificial>:(.text+0xab64): undefined reference to `grsoT2'
<artificial>:(.text+0xab6c): undefined reference to `grsoT3'
<artificial>:(.text+0xab7e): undefined reference to `grsoT2'
<artificial>:(.text+0xab86): undefined reference to `grsoT3'
<artificial>:(.text+0xab98): undefined reference to `grsoT4'
<artificial>:(.text+0xaba0): undefined reference to `grsoT5'
<artificial>:(.text+0xabb2): undefined reference to `grsoT4'
<artificial>:(.text+0xabba): undefined reference to `grsoT5'
<artificial>:(.text+0xabd0): undefined reference to `grsoT6'
<artificial>:(.text+0xabd8): undefined reference to `grsoT7'
<artificial>:(.text+0xabea): undefined reference to `grsoT6'
<artificial>:(.text+0xabf2): undefined reference to `grsoT7'
<artificial>:(.text+0xac00): undefined reference to `grsoT0'
<artificial>:(.text+0xac08): undefined reference to `grsoT1'
<artificial>:(.text+0xac1a): undefined reference to `grsoT0'
<artificial>:(.text+0xac22): undefined reference to `grsoT1'
<artificial>:(.text+0xac34): undefined reference to `grsoT2'
<artificial>:(.text+0xac3c): undefined reference to `grsoT3'
<artificial>:(.text+0xac4e): undefined reference to `grsoT2'
<artificial>:(.text+0xac56): undefined reference to `grsoT3'
<artificial>:(.text+0xac68): undefined reference to `grsoT4'
<artificial>:(.text+0xac70): undefined reference to `grsoT5'
<artificial>:(.text+0xac82): undefined reference to `grsoT4'
<artificial>:(.text+0xac8a): undefined reference to `grsoT5'
<artificial>:(.text+0xaca0): undefined reference to `grsoT6'
<artificial>:(.text+0xaca8): undefined reference to `grsoT7'
<artificial>:(.text+0xacba): undefined reference to `grsoT6'
<artificial>:(.text+0xacc2): undefined reference to `grsoT7'
/tmp/ccmS1O9H.ltrans19.ltrans.o: In function `grsoP1024ASM':
<artificial>:(.text+0xadbb): undefined reference to `grsoT0'
<artificial>:(.text+0xadc3): undefined reference to `grsoT1'
<artificial>:(.text+0xadd5): undefined reference to `grsoT0'
<artificial>:(.text+0xaddd): undefined reference to `grsoT1'
<artificial>:(.text+0xadef): undefined reference to `grsoT2'
<artificial>:(.text+0xadf7): undefined reference to `grsoT3'
<artificial>:(.text+0xae09): undefined reference to `grsoT2'
<artificial>:(.text+0xae11): undefined reference to `grsoT3'
<artificial>:(.text+0xae23): undefined reference to `grsoT4'
<artificial>:(.text+0xae2b): undefined reference to `grsoT5'
<artificial>:(.text+0xae3d): undefined reference to `grsoT4'
<artificial>:(.text+0xae45): undefined reference to `grsoT5'
<artificial>:(.text+0xae57): undefined reference to `grsoT6'
<artificial>:(.text+0xae5f): undefined reference to `grsoT7'
<artificial>:(.text+0xae6d): undefined reference to `grsoT6'
<artificial>:(.text+0xae75): undefined reference to `grsoT7'
<artificial>:(.text+0xae9b): undefined reference to `grsoT0'
<artificial>:(.text+0xaea3): undefined reference to `grsoT1'
<artificial>:(.text+0xaeb5): undefined reference to `grsoT0'
<artificial>:(.text+0xaebd): undefined reference to `grsoT1'
<artificial>:(.text+0xaecf): undefined reference to `grsoT2'
<artificial>:(.text+0xaed7): undefined reference to `grsoT3'
<artificial>:(.text+0xaee9): undefined reference to `grsoT2'
<artificial>:(.text+0xaef1): undefined reference to `grsoT3'
<artificial>:(.text+0xaf03): undefined reference to `grsoT4'
<artificial>:(.text+0xaf0b): undefined reference to `grsoT5'
<artificial>:(.text+0xaf1d): undefined reference to `grsoT4'
<artificial>:(.text+0xaf25): undefined reference to `grsoT5'
<artificial>:(.text+0xaf37): undefined reference to `grsoT6'
<artificial>:(.text+0xaf3f): undefined reference to `grsoT7'
<artificial>:(.text+0xaf4d): undefined reference to `grsoT6'
<artificial>:(.text+0xaf55): undefined reference to `grsoT7'
<artificial>:(.text+0xaf7b): undefined reference to `grsoT0'
<artificial>:(.text+0xaf83): undefined reference to `grsoT1'
<artificial>:(.text+0xaf95): undefined reference to `grsoT0'
<artificial>:(.text+0xaf9d): undefined reference to `grsoT1'
<artificial>:(.text+0xafaf): undefined reference to `grsoT2'
<artificial>:(.text+0xafb7): undefined reference to `grsoT3'
<artificial>:(.text+0xafc9): undefined reference to `grsoT2'
<artificial>:(.text+0xafd1): undefined reference to `grsoT3'
<artificial>:(.text+0xafe3): undefined reference to `grsoT4'
<artificial>:(.text+0xafeb): undefined reference to `grsoT5'
<artificial>:(.text+0xaffd): undefined reference to `grsoT4'
<artificial>:(.text+0xb005): undefined reference to `grsoT5'
<artificial>:(.text+0xb017): undefined reference to `grsoT6'
<artificial>:(.text+0xb01f): undefined reference to `grsoT7'
<artificial>:(.text+0xb02d): undefined reference to `grsoT6'
<artificial>:(.text+0xb035): undefined reference to `grsoT7'
<artificial>:(.text+0xb061): undefined reference to `grsoT0'
<artificial>:(.text+0xb069): undefined reference to `grsoT1'
<artificial>:(.text+0xb07b): undefined reference to `grsoT0'
<artificial>:(.text+0xb083): undefined reference to `grsoT1'
<artificial>:(.text+0xb095): undefined reference to `grsoT2'
<artificial>:(.text+0xb09d): undefined reference to `grsoT3'
<artificial>:(.text+0xb0af): undefined reference to `grsoT2'
<artificial>:(.text+0xb0b7): undefined reference to `grsoT3'
<artificial>:(.text+0xb0c9): undefined reference to `grsoT4'
<artificial>:(.text+0xb0d1): undefined reference to `grsoT5'
<artificial>:(.text+0xb0e3): undefined reference to `grsoT4'
<artificial>:(.text+0xb0eb): undefined reference to `grsoT5'
<artificial>:(.text+0xb0fd): undefined reference to `grsoT6'
<artificial>:(.text+0xb105): undefined reference to `grsoT7'
<artificial>:(.text+0xb113): undefined reference to `grsoT6'
<artificial>:(.text+0xb11b): undefined reference to `grsoT7'
<artificial>:(.text+0xb146): undefined reference to `grsoT0'
<artificial>:(.text+0xb14e): undefined reference to `grsoT1'
<artificial>:(.text+0xb160): undefined reference to `grsoT0'
<artificial>:(.text+0xb168): undefined reference to `grsoT1'
<artificial>:(.text+0xb17a): undefined reference to `grsoT2'
<artificial>:(.text+0xb182): undefined reference to `grsoT3'
<artificial>:(.text+0xb194): undefined reference to `grsoT2'
<artificial>:(.text+0xb19c): undefined reference to `grsoT3'
<artificial>:(.text+0xb1ae): undefined reference to `grsoT4'
<artificial>:(.text+0xb1b6): undefined reference to `grsoT5'
<artificial>:(.text+0xb1c8): undefined reference to `grsoT4'
<artificial>:(.text+0xb1d0): undefined reference to `grsoT5'
<artificial>:(.text+0xb1e2): undefined reference to `grsoT6'
<artificial>:(.text+0xb1ea): undefined reference to `grsoT7'
<artificial>:(.text+0xb1f8): undefined reference to `grsoT6'
<artificial>:(.text+0xb200): undefined reference to `grsoT7'
<artificial>:(.text+0xb22c): undefined reference to `grsoT0'
<artificial>:(.text+0xb234): undefined reference to `grsoT1'
<artificial>:(.text+0xb246): undefined reference to `grsoT0'
<artificial>:(.text+0xb24e): undefined reference to `grsoT1'
<artificial>:(.text+0xb260): undefined reference to `grsoT2'
<artificial>:(.text+0xb268): undefined reference to `grsoT3'
<artificial>:(.text+0xb27a): undefined reference to `grsoT2'
<artificial>:(.text+0xb282): undefined reference to `grsoT3'
<artificial>:(.text+0xb294): undefined reference to `grsoT4'
<artificial>:(.text+0xb29c): undefined reference to `grsoT5'
<artificial>:(.text+0xb2ae): undefined reference to `grsoT4'
<artificial>:(.text+0xb2b6): undefined reference to `grsoT5'
<artificial>:(.text+0xb2c8): undefined reference to `grsoT6'
<artificial>:(.text+0xb2d0): undefined reference to `grsoT7'
<artificial>:(.text+0xb2de): undefined reference to `grsoT6'
<artificial>:(.text+0xb2e6): undefined reference to `grsoT7'
<artificial>:(.text+0xb311): undefined reference to `grsoT0'
<artificial>:(.text+0xb319): undefined reference to `grsoT1'
<artificial>:(.text+0xb32b): undefined reference to `grsoT0'
<artificial>:(.text+0xb333): undefined reference to `grsoT1'
<artificial>:(.text+0xb345): undefined reference to `grsoT2'
<artificial>:(.text+0xb34d): undefined reference to `grsoT3'
<artificial>:(.text+0xb35f): undefined reference to `grsoT2'
<artificial>:(.text+0xb367): undefined reference to `grsoT3'
<artificial>:(.text+0xb379): undefined reference to `grsoT4'
<artificial>:(.text+0xb381): undefined reference to `grsoT5'
<artificial>:(.text+0xb393): undefined reference to `grsoT4'
<artificial>:(.text+0xb39b): undefined reference to `grsoT5'
<artificial>:(.text+0xb3ad): undefined reference to `grsoT6'
<artificial>:(.text+0xb3b5): undefined reference to `grsoT7'
<artificial>:(.text+0xb3c3): undefined reference to `grsoT6'
<artificial>:(.text+0xb3cb): undefined reference to `grsoT7'
<artificial>:(.text+0xb3d9): undefined reference to `grsoT0'
<artificial>:(.text+0xb3e1): undefined reference to `grsoT1'
<artificial>:(.text+0xb3f3): undefined reference to `grsoT0'
<artificial>:(.text+0xb3fb): undefined reference to `grsoT1'
<artificial>:(.text+0xb40d): undefined reference to `grsoT2'
<artificial>:(.text+0xb415): undefined reference to `grsoT3'
<artificial>:(.text+0xb427): undefined reference to `grsoT2'
<artificial>:(.text+0xb42f): undefined reference to `grsoT3'
<artificial>:(.text+0xb441): undefined reference to `grsoT4'
<artificial>:(.text+0xb449): undefined reference to `grsoT5'
<artificial>:(.text+0xb45b): undefined reference to `grsoT4'
<artificial>:(.text+0xb463): undefined reference to `grsoT5'
<artificial>:(.text+0xb475): undefined reference to `grsoT6'
<artificial>:(.text+0xb47d): undefined reference to `grsoT7'
<artificial>:(.text+0xb48b): undefined reference to `grsoT6'
<artificial>:(.text+0xb493): undefined reference to `grsoT7'
collect2: error: ld returned 1 exit status
Makefile:1292: recipe for target 'cpuminer' failed
make[2]: *** [cpuminer] Error 1
make[2]: Leaving directory '/root/cpuminer-opt-sse'
Makefile:3320: recipe for target 'all-recursive' failed
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory '/root/cpuminer-opt-sse'
Makefile:658: recipe for target 'all' failed
make: *** [all] Error 2
I solved it by deleting almost all the assembler from algo/groestl/sse2/grso-asm.c. CPU architecture selection is made at compile time. If you do a native compile on a CPU that supports AVX2 you can not run it
on a CPU with only AVX. If you want to cross compile you must specify the arch of the target CPU, and produce seperate executables
for each desired architecture.
Of course you are perfectly right (although cross-compiling is when a different host/target is desired, not just a different cpu type), I just got a bit confused because I saw some remarks "choose runtime" in the source code so I thought maybe there is some decision at runtime. So I have 3 binaries now for sse/avx/avx2. I made three directories and launch the corresponding binary like this: grep -q avx /proc/cpuinfo && feat="avx" || feat="sse" && grep -q avx2 /proc/cpuinfo && feat="avx2"
/somewhere/cpuminer-opt-$feat/cpuminer ... SSE and AVX versions are on par with tpruvot's speed-wise but when I compile them with tpruvot's uncommented build.sh they seem a tiny bit faster but anyway they have those align flags so I stay with it. My logic for AVX2 isn't fully implemented yet in the capablilities checks, had it been it would have displayed a message warning of the impending crash, then crashed. This is what you should see when implemented: CPU features: SSE2 AES AVX
SW built on Aug 5 2016 with GCC 5.4.0
SW features: SSE2 AES AVX AVX2
Algo features: SSE2 AES AVX AVX2
[color=red]Unsupported CPU or SW configuration, miner will likely crash![/color]
Illegal instruction (core dumped)
It's theoretically possibly that the binary would crash even before this print if gcc decided to use some of those "illegal" instructions in the code preceding the print. |
|
|
|
Maybe he is just being proactive?  Or he plays the wishful thinking / self-fulfilling prophecy tactics.. It's like all those people that tell you something and end with "thanks". They didn't ask you to do anything, you didn't do anything for them, yet they end with "thanks". What do they thank you for? Go figure. |
|
|
|
Success! I did this very ugly hack, joblo please don't get a heart attack: --- scrypt-jane-romix-template.h.orig 2016-02-05 22:05:38.000000000 +0000
+++ scrypt-jane-romix-template.h 2016-08-05 00:37:48.949684265 +0000
@@ -86,9 +86,9 @@
for (i = 0; i < /*N - 1*/511; i++, block += chunkWords) {
/* 3: V_i = X */
/* 4: X = H(X) */
- SCRYPT_CHUNKMIX_FN(block + chunkWords, block, NULL, /*r*/1);
+// SCRYPT_CHUNKMIX_FN(block + chunkWords, block, NULL, /*r*/1);
}
- SCRYPT_CHUNKMIX_FN(X, block, NULL, 1);
+// SCRYPT_CHUNKMIX_FN(X, block, NULL, 1);
/* 6: for i = 0 to N - 1 do */
for (i = 0; i < /*N*/512; i += 2) {
@@ -96,13 +96,13 @@
j = X[chunkWords - SCRYPT_BLOCK_WORDS] & /*(N - 1)*/511;
/* 8: X = H(Y ^ V_j) */
- SCRYPT_CHUNKMIX_FN(Y, X, scrypt_item(V, j, chunkWords), 1);
+// SCRYPT_CHUNKMIX_FN(Y, X, scrypt_item(V, j, chunkWords), 1);
/* 7: j = Integerify(Y) % N */
j = Y[chunkWords - SCRYPT_BLOCK_WORDS] & /*(N - 1)*/511;
/* 8: X = H(Y ^ V_j) */
- SCRYPT_CHUNKMIX_FN(X, Y, scrypt_item(V, j, chunkWords), 1);
+// SCRYPT_CHUNKMIX_FN(X, Y, scrypt_item(V, j, chunkWords), 1);
}
/* 10: B' = X */
And now it does compile with -flto and here is the result: CPU: Intel(R) Xeon(R) CPU E5-2676 v3 @ 2.40GHz
CPU features: SSE2 AES AVX AVX2
SW built on Aug 5 2016 with GCC 5.4.0
SW features: SSE2 AES AVX AVX2
Algo features: SSE2 AES AVX AVX2
Start mining with SSE2 AES AVX AVX2
[2016-08-05 00:58:03] 4 miner threads started, using 'lyra2re' algorithm.
[2016-08-05 00:58:04] CPU #0: 65.54 kH, 84.17 kH/s
[2016-08-05 00:58:04] CPU #1: 65.54 kH, 84.25 kH/s
[2016-08-05 00:58:04] CPU #3: 65.54 kH, 84.23 kH/s
[2016-08-05 00:58:04] Total: 196.61 kH, 252.64 kH/s
[2016-08-05 00:58:04] CPU #2: 65.54 kH, 83.86 kH/s
[2016-08-05 00:58:08] CPU #2: 335.45 kH, 84.02 kH/s
[2016-08-05 00:58:08] CPU #1: 336.99 kH, 84.25 kH/s
[2016-08-05 00:58:08] CPU #3: 336.92 kH, 84.24 kH/s
[2016-08-05 00:58:08] Total: 1074.89 kH, 336.68 kH/s
[2016-08-05 00:58:08] CPU #0: 336.67 kH, 84.04 kH/s
[2016-08-05 00:58:13] CPU #2: 420.12 kH, 84.16 kH/s
[2016-08-05 00:58:13] CPU #1: 421.26 kH, 84.35 kH/s
[2016-08-05 00:58:13] CPU #0: 420.18 kH, 84.19 kH/s
[2016-08-05 00:58:13] CPU #3: 421.18 kH, 84.34 kH/s
[2016-08-05 00:58:13] Total: 1682.74 kH, 337.04 kH/s
[2016-08-05 00:58:18] CPU #2: 420.78 kH, 84.16 kH/s
[2016-08-05 00:58:18] CPU #1: 421.77 kH, 84.31 kH/s
[2016-08-05 00:58:18] CPU #0: 420.97 kH, 84.19 kH/s
[2016-08-05 00:58:18] CPU #3: 421.69 kH, 84.26 kH/s
[2016-08-05 00:58:18] Total: 1685.21 kH, 336.92 kH/s
[2016-08-05 00:58:23] CPU #1: 421.54 kH, 84.37 kH/s
[2016-08-05 00:58:23] CPU #3: 421.31 kH, 84.32 kH/s
[2016-08-05 00:58:23] CPU #2: 420.81 kH, 83.99 kH/s
[2016-08-05 00:58:23] Total: 1684.63 kH, 336.87 kH/s
[2016-08-05 00:58:23] CPU #0: 420.93 kH, 84.01 kH/s
[2016-08-05 00:58:28] CPU #2: 419.96 kH, 84.10 kH/s
[2016-08-05 00:58:28] CPU #0: 420.07 kH, 84.10 kH/s
[2016-08-05 00:58:28] CPU #1: 421.87 kH, 84.17 kH/s
[2016-08-05 00:58:28] CPU #3: 421.58 kH, 84.09 kH/s
[2016-08-05 00:58:28] Total: 1683.49 kH, 336.46 kH/s
So using -flto gives another 2.75% speed increase. That's 7.7% speed increase in total over tpruvot. Now this is with -flto and -fuse-linker-plugin: CPU: Intel(R) Xeon(R) CPU E5-2676 v3 @ 2.40GHz
CPU features: SSE2 AES AVX AVX2
SW built on Aug 5 2016 with GCC 5.4.0
SW features: SSE2 AES AVX AVX2
Algo features: SSE2 AES AVX AVX2
Start mining with SSE2 AES AVX AVX2
[2016-08-05 00:55:15] 4 miner threads started, using 'lyra2re' algorithm.
[2016-08-05 00:55:16] CPU #0: 65.54 kH, 84.75 kH/s
[2016-08-05 00:55:16] CPU #1: 65.54 kH, 84.78 kH/s
[2016-08-05 00:55:16] CPU #2: 65.54 kH, 84.56 kH/s
[2016-08-05 00:55:16] CPU #3: 65.54 kH, 84.44 kH/s
[2016-08-05 00:55:16] Total: 262.14 kH, 338.53 kH/s
[2016-08-05 00:55:20] CPU #3: 337.77 kH, 84.06 kH/s
[2016-08-05 00:55:20] Total: 534.38 kH, 338.15 kH/s
[2016-08-05 00:55:20] CPU #2: 338.22 kH, 84.01 kH/s
[2016-08-05 00:55:20] CPU #1: 339.13 kH, 84.09 kH/s
[2016-08-05 00:55:20] CPU #0: 338.98 kH, 84.02 kH/s
[2016-08-05 00:55:25] CPU #0: 420.11 kH, 84.71 kH/s
[2016-08-05 00:55:25] CPU #2: 420.03 kH, 84.49 kH/s
[2016-08-05 00:55:25] CPU #3: 420.31 kH, 84.05 kH/s
[2016-08-05 00:55:25] Total: 1599.59 kH, 337.33 kH/s
[2016-08-05 00:55:25] CPU #1: 420.43 kH, 84.07 kH/s
[2016-08-05 00:55:30] CPU #3: 420.25 kH, 83.97 kH/s
[2016-08-05 00:55:30] Total: 1680.82 kH, 337.24 kH/s
[2016-08-05 00:55:30] CPU #2: 422.44 kH, 83.97 kH/s
[2016-08-05 00:55:30] CPU #0: 423.54 kH, 83.98 kH/s
[2016-08-05 00:55:30] CPU #1: 420.36 kH, 83.97 kH/s
[2016-08-05 00:55:35] CPU #0: 419.88 kH, 84.64 kH/s
[2016-08-05 00:55:35] CPU #2: 419.84 kH, 84.39 kH/s
[2016-08-05 00:55:35] CPU #3: 419.85 kH, 84.00 kH/s
[2016-08-05 00:55:35] Total: 1679.93 kH, 337.00 kH/s
[2016-08-05 00:55:35] CPU #1: 419.85 kH, 84.02 kH/s
[2016-08-05 00:55:40] CPU #0: 423.20 kH, 84.42 kH/s
[2016-08-05 00:55:40] CPU #3: 420.02 kH, 84.32 kH/s
[2016-08-05 00:55:40] Total: 1682.91 kH, 337.15 kH/s
Basically the same speed. Now what if I actually call tpruvot's build.sh, exactly the one I showed in my previous post: CPU: Intel(R) Xeon(R) CPU E5-2676 v3 @ 2.40GHz
CPU features: SSE2 AES AVX AVX2
SW built on Aug 5 2016 with GCC 5.4.0
SW features: SSE2 AES AVX AVX2
Algo features: SSE2 AES AVX AVX2
Start mining with SSE2 AES AVX AVX2
[2016-08-05 01:10:02] 4 miner threads started, using 'lyra2re' algorithm.
[2016-08-05 01:10:03] CPU #0: 65.54 kH, 84.11 kH/s
[2016-08-05 01:10:03] CPU #1: 65.54 kH, 83.93 kH/s
[2016-08-05 01:10:03] CPU #2: 65.54 kH, 83.86 kH/s
[2016-08-05 01:10:03] CPU #3: 65.54 kH, 83.96 kH/s
[2016-08-05 01:10:03] Total: 262.14 kH, 335.86 kH/s
[2016-08-05 01:10:07] CPU #1: 335.71 kH, 84.00 kH/s
[2016-08-05 01:10:07] CPU #2: 335.44 kH, 83.92 kH/s
[2016-08-05 01:10:07] CPU #3: 335.85 kH, 83.99 kH/s
[2016-08-05 01:10:07] Total: 1072.54 kH, 336.02 kH/s
[2016-08-05 01:10:07] CPU #0: 336.45 kH, 83.93 kH/s
[2016-08-05 01:10:12] CPU #1: 420.00 kH, 84.00 kH/s
[2016-08-05 01:10:12] CPU #2: 419.62 kH, 83.92 kH/s
[2016-08-05 01:10:12] CPU #3: 419.93 kH, 83.99 kH/s
[2016-08-05 01:10:12] Total: 1596.00 kH, 335.82 kH/s
[2016-08-05 01:10:12] CPU #0: 419.64 kH, 83.91 kH/s
[2016-08-05 01:10:17] CPU #1: 419.98 kH, 84.05 kH/s
[2016-08-05 01:10:17] CPU #2: 419.58 kH, 83.98 kH/s
[2016-08-05 01:10:17] CPU #3: 419.93 kH, 84.03 kH/s
[2016-08-05 01:10:17] Total: 1679.12 kH, 335.98 kH/s
[2016-08-05 01:10:17] CPU #0: 419.53 kH, 83.99 kH/s
[2016-08-05 01:10:22] CPU #2: 419.92 kH, 84.04 kH/s
[2016-08-05 01:10:22] CPU #1: 420.25 kH, 84.04 kH/s
[2016-08-05 01:10:22] CPU #0: 419.93 kH, 84.04 kH/s
[2016-08-05 01:10:22] CPU #3: 420.18 kH, 84.02 kH/s
[2016-08-05 01:10:22] Total: 1680.28 kH, 336.14 kH/s
Still the same (maximum) speed. So I will be using joblo's cpuminer with tpruvot's (uncommented) build.sh because that build.sh has all those other flags (including -falign-*) which may or may not matter, so just to be safe.. EDIT: when I took the avx2 binary and tried to run it on a avx cpu I got this: CPU: Intel(R) Xeon(R) CPU E5-2670 v2 @ 2.50GHz
CPU features: SSE2 AES AVX
SW built on Aug 5 2016 with GCC 5.4.0
SW features: SSE2 AES AVX AVX2
Algo features: SSE2 AES AVX AVX2
Start mining with SSE2 AES AVX
Illegal instruction (core dumped)
But wasn't the whole idea that all the cpu features will be compiled in and what particular feature shall be used will be determined at the runtime? It's not a big deal, I just recompiled it and I will have two versions (avx and avx2) and run the one that's appropriate to the cpu. Just I thought I would report this. |
|
|
|
joblo, that's an excellent improvement! Now you are definitely faster than tpruvot, at least by 4.8%. I made a better repeatable benchmark of tpruvot's, the numbers are directly in the build.sh: #!/bin/bash
if [ "$OS" = "Windows_NT" ]; then
./mingw64.sh
exit 0
fi
# Linux build
make clean || echo clean
rm -f config.status
./autogen.sh || echo done
# Ubuntu 10.04 (gcc 4.4)
extracflags="-O3 -march=native -w -D_REENTRANT -funroll-loops -fvariable-expansion-in-unroller -fmerge-all-constants -fbranch-target-load-optimize2 -fsched2-use-superblocks -falign-loops=16 -falign-functions=16 -falign-jumps=16 -falign-labels=16"
# Debian 7.7 / Ubuntu 14.04 (gcc 4.7+)
extracflags="$extracflags -Ofast -fuse-linker-plugin -ftree-loop-if-convert-stores"
if [ ! "0" = `cat /proc/cpuinfo | grep -c avx` ]; then
# march native doesn't always works, ex. some Pentium Gxxx (no avx)
extracflags="$extracflags -march=native"
fi
# Intel(R) Xeon(R) CPU E5-2676 v3 @ 2.40GHz 4 threads (d2.xlarge)
#309-310
#CFLAGS="-O3 $extracflags -flto -march=native -DUSE_ASM -pg" ./configure --with-crypto --with-curl
#311-312
CFLAGS="-O3 $extracflags -flto -march=native -DUSE_ASM -pg" CXXFLAGS="-std=gnu++11" ./configure --with-crypto --with-curl
#281
#CFLAGS="-O3 $extracflags -flto -march=native -DUSE_ASM -pg" CXXFLAGS="$CFLAGS" ./configure --with-crypto --with-curl
#280
#CFLAGS="-O3 $extracflags -flto -march=native -DUSE_ASM -pg" CXXFLAGS="$CFLAGS -std=gnu++11" ./configure --with-crypto --with-curl
#269
#CFLAGS="-O3 $extracflags -march=native -DUSE_ASM -pg" ./configure --with-crypto --with-curl
#264
#CFLAGS="-O3 $extracflags -march=native -DUSE_ASM -pg" CXXFLAGS="-std=gnu++11" ./configure --with-crypto --with-curl
#242
#CFLAGS="-O3 $extracflags -march=native -DUSE_ASM -pg" CXXFLAGS="$CFLAGS" ./configure --with-crypto --with-curl
#245
#CFLAGS="-O3 $extracflags -march=native -DUSE_ASM -pg" CXXFLAGS="$CFLAGS -std=gnu++11" ./configure --with-crypto --with-curl
make -j $(grep processor /proc/cpuinfo | wc -l)
strip -s cpuminer
So with him I get 312 at best on this particular machine and that config of flags is basically the default if you uncomment everything so I didn't make him any faster, I just proved he can be much slower if wrong flags are used. Now with yours, without any change, untouched cpuminer-opt-3.4.0.tar.gz, I get this: root@xxx:~/cpuminer-opt# ./cpuminer -a lyra2re --benchmark
CPU: Intel(R) Xeon(R) CPU E5-2676 v3 @ 2.40GHz
CPU features: SSE2 AES AVX AVX2
SW built on Aug 4 2016 with GCC 5.4.0
SW features: SSE2 AES AVX AVX2
Algo features: SSE2 AES AVX AVX2
Start mining with SSE2 AES AVX AVX2
[2016-08-04 23:24:25] 4 miner threads started, using 'lyra2re' algorithm.
[2016-08-04 23:24:26] CPU #1: 65.54 kH, 82.16 kH/s
[2016-08-04 23:24:26] CPU #0: 65.54 kH, 81.67 kH/s
[2016-08-04 23:24:26] CPU #3: 65.54 kH, 81.84 kH/s
[2016-08-04 23:24:26] Total: 196.61 kH, 245.67 kH/s
[2016-08-04 23:24:26] CPU #2: 65.54 kH, 81.68 kH/s
[2016-08-04 23:24:30] CPU #0: 326.68 kH, 81.80 kH/s
[2016-08-04 23:24:30] CPU #3: 327.37 kH, 82.00 kH/s
[2016-08-04 23:24:30] Total: 785.13 kH, 327.64 kH/s
[2016-08-04 23:24:30] CPU #2: 326.73 kH, 81.75 kH/s
[2016-08-04 23:24:30] CPU #1: 328.64 kH, 82.01 kH/s
[2016-08-04 23:24:35] CPU #0: 409.02 kH, 81.78 kH/s
[2016-08-04 23:24:35] CPU #3: 409.99 kH, 81.93 kH/s
[2016-08-04 23:24:35] Total: 1474.38 kH, 327.46 kH/s
[2016-08-04 23:24:35] CPU #2: 408.76 kH, 81.73 kH/s
[2016-08-04 23:24:35] CPU #1: 410.04 kH, 81.94 kH/s
[2016-08-04 23:24:40] CPU #0: 408.89 kH, 81.78 kH/s
[2016-08-04 23:24:40] CPU #3: 409.63 kH, 81.94 kH/s
[2016-08-04 23:24:40] Total: 1637.32 kH, 327.39 kH/s
But when I add -flto I get the following error at the final link: g++ -O3 -march=native -w -flto -std=gnu++11 -Lyes/lib -Lyes/lib -o cpuminer cpuminer-cpu-miner.o cpuminer-util.o cpuminer-uint256.o cpuminer-api.o cpuminer-sysinfos.o cpuminer-algo-gate-api.o algo/groestl/cpuminer-sph_groestl.o algo/skein/cpuminer-sph_skein.o algo/bmw/cpuminer-sph_bmw.o algo/shavite/cpuminer-sph_shavite.o algo/shavite/cpuminer-shavite.o algo/echo/cpuminer-sph_echo.o algo/blake/cpuminer-sph_blake.o algo/heavy/cpuminer-sph_hefty1.o algo/blake/cpuminer-mod_blakecoin.o algo/luffa/cpuminer-sph_luffa.o algo/cubehash/cpuminer-sph_cubehash.o algo/simd/cpuminer-sph_simd.o algo/hamsi/cpuminer-sph_hamsi.o algo/fugue/cpuminer-sph_fugue.o algo/gost/cpuminer-sph_gost.o algo/jh/cpuminer-sph_jh.o algo/keccak/cpuminer-sph_keccak.o algo/keccak/cpuminer-keccak.o algo/sha3/cpuminer-sph_sha2.o algo/sha3/cpuminer-sph_sha2big.o algo/shabal/cpuminer-sph_shabal.o algo/whirlpool/cpuminer-sph_whirlpool.o crypto/cpuminer-blake2s.o crypto/cpuminer-oaes_lib.o crypto/cpuminer-c_keccak.o crypto/cpuminer-c_groestl.o crypto/cpuminer-c_blake256.o crypto/cpuminer-c_jh.o crypto/cpuminer-c_skein.o crypto/cpuminer-hash.o crypto/cpuminer-aesb.o crypto/cpuminer-magimath.o algo/argon2/cpuminer-argon2a.o algo/argon2/ar2/cpuminer-argon2.o algo/argon2/ar2/cpuminer-opt.o algo/argon2/ar2/cpuminer-cores.o algo/argon2/ar2/cpuminer-ar2-scrypt-jane.o algo/argon2/ar2/cpuminer-blake2b.o algo/cpuminer-axiom.o algo/blake/cpuminer-blake.o algo/blake/cpuminer-blake2.o algo/blake/cpuminer-blakecoin.o algo/blake/cpuminer-decred.o algo/blake/cpuminer-pentablake.o algo/bmw/cpuminer-bmw256.o algo/cubehash/sse2/cpuminer-cubehash_sse2.o algo/cryptonight/cpuminer-cryptolight.o algo/cryptonight/cpuminer-cryptonight-common.o algo/cryptonight/cpuminer-cryptonight-aesni.o algo/cryptonight/cpuminer-cryptonight.o algo/cpuminer-drop.o algo/echo/aes_ni/cpuminer-hash.o algo/cpuminer-fresh.o algo/groestl/cpuminer-groestl.o algo/groestl/cpuminer-myr-groestl.o algo/groestl/sse2/cpuminer-grso.o algo/groestl/sse2/cpuminer-grso-asm.o algo/groestl/aes_ni/cpuminer-hash-groestl.o algo/groestl/aes_ni/cpuminer-hash-groestl256.o algo/haval/cpuminer-haval.o algo/heavy/cpuminer-heavy.o algo/heavy/cpuminer-bastion.o algo/cpuminer-hmq1725.o algo/hodl/cpuminer-hodl.o algo/hodl/cpuminer-hodl-gate.o algo/hodl/cpuminer-hodl_arith_uint256.o algo/hodl/cpuminer-hodl_uint256.o algo/hodl/cpuminer-hash.o algo/hodl/cpuminer-hmac_sha512.o algo/hodl/cpuminer-sha256.o algo/hodl/cpuminer-sha512.o algo/hodl/cpuminer-utilstrencodings.o algo/hodl/cpuminer-hodl-wolf.o algo/hodl/cpuminer-aes.o algo/hodl/cpuminer-sha512_avx.o algo/hodl/cpuminer-sha512_avx2.o algo/cpuminer-lbry.o algo/luffa/cpuminer-luffa.o algo/luffa/sse2/cpuminer-luffa_for_sse2.o algo/lyra2/cpuminer-lyra2.o algo/lyra2/cpuminer-sponge.o algo/lyra2/cpuminer-lyra2rev2.o algo/lyra2/cpuminer-lyra2re.o algo/keccak/sse2/cpuminer-keccak.o algo/cpuminer-m7m.o algo/cpuminer-neoscrypt.o algo/cpuminer-nist5.o algo/cpuminer-pluck.o algo/quark/cpuminer-quark.o algo/qubit/cpuminer-qubit.o algo/ripemd/cpuminer-sph_ripemd.o algo/cpuminer-scrypt.o algo/scryptjane/cpuminer-scrypt-jane.o algo/sha2/cpuminer-sha2.o algo/simd/sse2/cpuminer-nist.o algo/simd/sse2/cpuminer-vector.o algo/skein/cpuminer-skein.o algo/skein/cpuminer-skein2.o algo/cpuminer-s3.o algo/tiger/cpuminer-sph_tiger.o algo/whirlpool/cpuminer-whirlpool.o algo/whirlpool/cpuminer-whirlpoolx.o algo/x11/cpuminer-x11.o algo/x11/cpuminer-x11evo.o algo/x11/cpuminer-x11gost.o algo/x11/cpuminer-c11.o algo/x13/cpuminer-x13.o algo/x14/cpuminer-x14.o algo/x15/cpuminer-x15.o algo/x17/cpuminer-x17.o algo/yescrypt/cpuminer-yescrypt.o algo/yescrypt/cpuminer-yescrypt-common.o algo/yescrypt/cpuminer-sha256_Y.o algo/yescrypt/cpuminer-yescrypt-simd.o algo/cpuminer-zr5.o asm/cpuminer-neoscrypt_asm.o asm/cpuminer-sha2-x64.o asm/cpuminer-scrypt-x64.o asm/cpuminer-aesb-x64.o -lcurl -lz -ljansson -lpthread -lssl -lcrypto -lgmp
/tmp/ccVXbbn8.ltrans6.ltrans.o: In function `scrypt_ROMix_avx2':
<artificial>:(.text+0x9712): undefined reference to `scrypt_ChunkMix_avx2'
<artificial>:(.text+0x9729): undefined reference to `scrypt_ChunkMix_avx2'
<artificial>:(.text+0x9760): undefined reference to `scrypt_ChunkMix_avx2'
<artificial>:(.text+0x9785): undefined reference to `scrypt_ChunkMix_avx2'
/tmp/ccVXbbn8.ltrans6.ltrans.o: In function `scrypt_ROMix_xop':
<artificial>:(.text+0x99f2): undefined reference to `scrypt_ChunkMix_xop'
<artificial>:(.text+0x9a09): undefined reference to `scrypt_ChunkMix_xop'
<artificial>:(.text+0x9a40): undefined reference to `scrypt_ChunkMix_xop'
<artificial>:(.text+0x9a65): undefined reference to `scrypt_ChunkMix_xop'
/tmp/ccVXbbn8.ltrans6.ltrans.o: In function `scrypt_ROMix_avx':
<artificial>:(.text+0x9cd2): undefined reference to `scrypt_ChunkMix_avx'
<artificial>:(.text+0x9ce9): undefined reference to `scrypt_ChunkMix_avx'
<artificial>:(.text+0x9d20): undefined reference to `scrypt_ChunkMix_avx'
<artificial>:(.text+0x9d45): undefined reference to `scrypt_ChunkMix_avx'
/tmp/ccVXbbn8.ltrans6.ltrans.o: In function `scrypt_ROMix_ssse3':
<artificial>:(.text+0x9fb2): undefined reference to `scrypt_ChunkMix_ssse3'
<artificial>:(.text+0x9fc9): undefined reference to `scrypt_ChunkMix_ssse3'
<artificial>:(.text+0xa000): undefined reference to `scrypt_ChunkMix_ssse3'
<artificial>:(.text+0xa025): undefined reference to `scrypt_ChunkMix_ssse3'
/tmp/ccVXbbn8.ltrans6.ltrans.o: In function `scrypt_ROMix_sse2':
<artificial>:(.text+0xa292): undefined reference to `scrypt_ChunkMix_sse2'
<artificial>:(.text+0xa2a9): undefined reference to `scrypt_ChunkMix_sse2'
<artificial>:(.text+0xa2e0): undefined reference to `scrypt_ChunkMix_sse2'
<artificial>:(.text+0xa305): undefined reference to `scrypt_ChunkMix_sse2'
collect2: error: ld returned 1 exit status
Makefile:1333: recipe for target 'cpuminer' failed
make[2]: *** [cpuminer] Error 1
make[2]: Leaving directory '/root/cpuminer-opt'
Makefile:3453: recipe for target 'all-recursive' failed
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory '/root/cpuminer-opt'
Makefile:670: recipe for target 'all' failed
make: *** [all] Error 2
If you are unsure how to fix it could you at least guide me how to disable the whole scrypt (optimization) because I am really anxious to see what -flto will do. |
|
|
|
I said that I already know this method, it's nothing new. I will try to work on it for you guys and post the results
I think it would be fair if you try this method (as you promised) for 2 hours (as the seller promised should be enough) and tell us how much you managed to make within that time frame. |
|
|
|
Hi experts, I'm trying to build in my openSUSE linux. But I've got an error:
./build.sh
make: /lib64/libc.so.6: version `GLIBC_2.14' not found (required by make)
make: /lib64/libc.so.6: version `GLIBC_2.17' not found (required by make)
strip: 'cpuminer': No such file
Regards
This problem has nothing to do with cpuminer, your building environment is messed up. Glibc is the very core linux library, if make can't find the version it likes then there is something wrong with either. But if glibc was messed up the system would hardly even boot properly. Maybe try to update? Thanks for the reply. I think my linux has old version of GLIBC_x.x. Version information: /bin/sh: libdl.so.2 (GLIBC_2.2.5) => /lib64/libdl.so.2 libc.so.6 (GLIBC_2.4) => /lib64/libc.so.6 libc.so.6 (GLIBC_2.  => /lib64/libc.so.6 libc.so.6 (GLIBC_2.3) => /lib64/libc.so.6 libc.so.6 (GLIBC_2.11) => /lib64/libc.so.6 libc.so.6 (GLIBC_2.3.4) => /lib64/libc.so.6 libc.so.6 (GLIBC_2.2.5) => /lib64/libc.so.6 /lib64/libreadline.so.5: libc.so.6 (GLIBC_2.4) => /lib64/libc.so.6 libc.so.6 (GLIBC_2.3) => /lib64/libc.so.6 libc.so.6 (GLIBC_2.3.4) => /lib64/libc.so.6 libc.so.6 (GLIBC_2.11) => /lib64/libc.so.6 libc.so.6 (GLIBC_2.2.5) => /lib64/libc.so.6 Now you are looking what versions of glibc these two binaries (/bin/sh and /lib64/libreadline.so.5) require. To find out what version of glibc you actually have just run the library: type /lib64/libc.so.6 and press enter. Presumably it will be equal or higher than what /bin/sh requires and lower than what 'make' requires. But that would mean that you didn't install 'make' a standard way via a package system but somehow sideway. What did you do to your suse?!? |
|
|
|
|
Show Posts
