CCminer(SP-MOD) Modded GPU kernels.

[djm34]

These endless discussions. Got to give it to you guys. I would have given up a long time ago 

I try to put some discussions, otherwise it would look like some endless call for donation 

[1714753136]

1714753136

[1714753136]

1714753136

[1714753136]

1714753136

[bensam1231]

It isn't new nor secret that Wolf0 has been working for quite some time on a new miners (he might be able to make more accurate statement if he passes by), things like that are more often discussed on #ccminer.

I always respect the work of other people so if someone include a gpl licensing, and I happen to use that piece of code, then I keep it as I found it with the licensing. (also as you don't seem to know, even though I repeat it all the time, I don't release private really often... so 98.5% of the time, when I release something it is directly on github wirh proper licensing and the other time, the code is even given with the binaries).

And again it isn't about what other may or may not understand  (which I don't really care) it is just the way it should be done.
I want people to deal with my code the proper way, so don't expect me to deal with other people code differently.

Alright, so it's the way 'you' do things...

As I've mentioned before, a miner fee is the best way of supporting developers long term while they work on improving their miner. It scales to both small and large miners. So if you don't care about picking apart the rats nest of licensing most people don't care about, then a miner fee works fine. All the GPL code is available to everyone already anyway, so people aren't losing out on that GPL code if it's not freely available. They're losing out on the kernel optimizations, which really have nothing to do with the GPL code, hence why we talked about something like a blackbox miner.

If you're really worried you could post the rest of the miner without the kernels next to it so people have it readily available and can see what it's built off of (not that anyone cares).

Two versions, one without the kernel and one with the kernel that isn't opensource. I'm sure they'll understand your spirit of keeping things alive. ^^


Ethereum is still hot guys, you should be mining, Nvidia or not.

[sp_]

My git is still active. Just submitted a 0.2% increase in lyra2v2.

yeah I saw you commited something you removed... still the same activity as usual... half of the change which happens are just to correct some shit you put it in at the previous round...  

It isn't release 78, actually it is more likely 39 (in terms of real progress). 
Seriously can't you just check what you are doing before committing anything 

Stop trolling and finish your pledge miner. 

[sp_]

so basically you expect other dev to work on the opensource so to make your private miner faster
pallas says thank you to you 

I will study your new optimized Neoscrypt code and learn.
You should study my code as well. We use different ways to reach our goal. I do small compiler/assembly optimalizations while you are reinventing the algorithm. You should buy my private miner and analyze it. I will give you a discount of 0.05 BTC. 

[djm34]

so basically you expect other dev to work on the opensource so to make your private miner faster
pallas says thank you to you 

I will study your new optimized Neoscrypt code and learn.
You should study my code as well. We use different ways to reach our goal. I do small compiler/assembly optimalizations while you are reinventing the algorithm. You should buy my private miner and analyze it. I will give you a discount of 0.05 BTC. 

that's your problem, you need to study my code to get something out, while if you knew what you were doing it is the original code would have to read to make it easier for the gpu.
Basically you don't get any real upgrade because you don't look at what the code does, you are just trying random stuff expecting it will work based on what and other devs wrote... well good luck with that

[sp_]

so basically you expect other dev to work on the opensource so to make your private miner faster
pallas says thank you to you 

I will study your new optimized Neoscrypt code and learn.
You should study my code as well. We use different ways to reach our goal. I do small compiler/assembly optimalizations while you are reinventing the algorithm. You should buy my private miner and analyze it. I will give you a discount of 0.05 BTC. 

that's your problem, you need to study my code to get something out, while if you knew what you were doing it is the original code would have to read to make it easier for the gpu.
Basically you don't get any real upgrade because you don't look at what the code does, you are just trying random stuff expecting it will work based on what and other devs wrote... well good luck with that

I must be the luckiest man in the universe then. My private kernals are the fastest in the world.

[djm34]

so basically you expect other dev to work on the opensource so to make your private miner faster
pallas says thank you to you 

I will study your new optimized Neoscrypt code and learn.
You should study my code as well. We use different ways to reach our goal. I do small compiler/assembly optimalizations while you are reinventing the algorithm. You should buy my private miner and analyze it. I will give you a discount of 0.05 BTC. 

that's your problem, you need to study my code to get something out, while if you knew what you were doing it is the original code would have to read to make it easier for the gpu.
Basically you don't get any real upgrade because you don't look at what the code does, you are just trying random stuff expecting it will work based on what and other devs wrote... well good luck with that

I must be the luckiest man in the universe then. My private kernals are the fastest in the world.

I feel like discussing with a tv commercial right now... where's the fucking remote ?

[sp_]

so basically you expect other dev to work on the opensource so to make your private miner faster
pallas says thank you to you  

I will study your new optimized Neoscrypt code and learn.
You should study my code as well. We use different ways to reach our goal. I do small compiler/assembly optimalizations while you are reinventing the algorithm. You should buy my private miner and analyze it. I will give you a discount of 0.05 BTC.  

that's your problem, you need to study my code to get something out, while if you knew what you were doing it is the original code would have to read to make it easier for the gpu.
Basically you don't get any real upgrade because you don't look at what the code does, you are just trying random stuff expecting it will work based on what and other devs wrote... well good luck with that

I must be the luckiest man in the universe then. My private kernals are the fastest in the world.

I feel like discussing with a tv commercial right now... where's the fucking remote ?

Saturday night. You should go and DJ. That is your profession isn't it?

[Eliovp]

Ethereum to the moon. that's what i'm seeing!

Damn i'm so happy that i own AMD cards 

[hashbrown9000]

Ethereum to the moon. that's what i'm seeing!

Nope. Probably just bob and his minions p'n'd-ing

[Eliovp]

Ethereum to the moon. that's what i'm seeing!

Nope. Probably just bob and his minions p'n'd-ing

Nope? i just ROI'ed my ROI from my ROI on top of my ROI 

[sp_]

This thread has 550 000 Views. This is real Human programming....

[djm34]

so basically you expect other dev to work on the opensource so to make your private miner faster
pallas says thank you to you  

I will study your new optimized Neoscrypt code and learn.
You should study my code as well. We use different ways to reach our goal. I do small compiler/assembly optimalizations while you are reinventing the algorithm. You should buy my private miner and analyze it. I will give you a discount of 0.05 BTC.  

that's your problem, you need to study my code to get something out, while if you knew what you were doing it is the original code would have to read to make it easier for the gpu.
Basically you don't get any real upgrade because you don't look at what the code does, you are just trying random stuff expecting it will work based on what and other devs wrote... well good luck with that

I must be the luckiest man in the universe then. My private kernals are the fastest in the world.

I feel like discussing with a tv commercial right now... where's the fucking remote ?

Saturday night. You should go and DJ. That is your profession isn't it?

lol what make you think that ? The vibe I put on your thread  
is SP your profession ?  

[sp_]

so basically you expect other dev to work on the opensource so to make your private miner faster
pallas says thank you to you  

I will study your new optimized Neoscrypt code and learn.
You should study my code as well. We use different ways to reach our goal. I do small compiler/assembly optimalizations while you are reinventing the algorithm. You should buy my private miner and analyze it. I will give you a discount of 0.05 BTC.  

that's your problem, you need to study my code to get something out, while if you knew what you were doing it is the original code would have to read to make it easier for the gpu.
Basically you don't get any real upgrade because you don't look at what the code does, you are just trying random stuff expecting it will work based on what and other devs wrote... well good luck with that

I must be the luckiest man in the universe then. My private kernals are the fastest in the world.

I feel like discussing with a tv commercial right now... where's the fucking remote ?

Saturday night. You should go and DJ. That is your profession isn't it?

lol what make you think that ? The vibe I put on your thread  
is SP your profession ?  

I am a DJ. You are not

[djm34]

so basically you expect other dev to work on the opensource so to make your private miner faster
pallas says thank you to you  

I will study your new optimized Neoscrypt code and learn.
You should study my code as well. We use different ways to reach our goal. I do small compiler/assembly optimalizations while you are reinventing the algorithm. You should buy my private miner and analyze it. I will give you a discount of 0.05 BTC.  

that's your problem, you need to study my code to get something out, while if you knew what you were doing it is the original code would have to read to make it easier for the gpu.
Basically you don't get any real upgrade because you don't look at what the code does, you are just trying random stuff expecting it will work based on what and other devs wrote... well good luck with that

I must be the luckiest man in the universe then. My private kernals are the fastest in the world.

I feel like discussing with a tv commercial right now... where's the fucking remote ?

Saturday night. You should go and DJ. That is your profession isn't it?

lol what make you think that ? The vibe I put on your thread  
is SP your profession ?  

I am a DJ. You are not

never said I was
You are making stuff up then telling me that I am not what you made up good one
(you should stop drugs, it clearly alters you consciousness and your programming skill)

[joblo]

so basically you expect other dev to work on the opensource so to make your private miner faster
pallas says thank you to you 

I will study your new optimized Neoscrypt code and learn.
You should study my code as well. We use different ways to reach our goal. I do small compiler/assembly optimalizations while you are reinventing the algorithm. You should buy my private miner and analyze it. I will give you a discount of 0.05 BTC. 

I'm aware of a couple of  assembly code optimizing techniques I've used before but require an initimate knowledge
of the CPU (GPU in this case) architecture including memory interface, cache organization and execution environment.

I intend to give this a try with cpuminer-opt once I get up to speed on Intel architecture. Would you be interested in
doing it for cuda? I could explain the details.

[djm34]

so basically you expect other dev to work on the opensource so to make your private miner faster
pallas says thank you to you 

I will study your new optimized Neoscrypt code and learn.
You should study my code as well. We use different ways to reach our goal. I do small compiler/assembly optimalizations while you are reinventing the algorithm. You should buy my private miner and analyze it. I will give you a discount of 0.05 BTC. 

I'm aware of a couple of  assembly code optimizing techniques I've used before but require an initimate knowledge
of the CPU (GPU in this case) architecture including memory interface, cache organization and execution environment.

I intend to give this a try with cpuminer-opt once I get up to speed on Intel architecture. Would you be interested in
doing it for cuda? I could explain the details.

yeah because it is well known that we just do random stuff 

[sp_]

Winners don`t use drugs. My crypto kernals are profitable. Yours are not. Finish the plegde please...

[joblo]

so basically you expect other dev to work on the opensource so to make your private miner faster
pallas says thank you to you  

I will study your new optimized Neoscrypt code and learn.
You should study my code as well. We use different ways to reach our goal. I do small compiler/assembly optimalizations while you are reinventing the algorithm. You should buy my private miner and analyze it. I will give you a discount of 0.05 BTC.  

I'm aware of a couple of  assembly code optimizing techniques I've used before but require an initimate knowledge
of the CPU (GPU in this case) architecture including memory interface, cache organization and execution environment.

I intend to give this a try with cpuminer-opt once I get up to speed on Intel architecture. Would you be interested in
doing it for cuda? I could explain the details.

yeah because it is well known that we just do random stuff  

LOL. Trial and error also works sometimes, the key is to figure out exactly why it works.

In my teaser ity's about code scheduling to maximize throughput by avoiding processor stalls
and maximizing superscalar operation.. It's not something that compilers can do easilly
because it requires so much anlysis and detailed understanding of the operation of
a CPU.

You may think I'm joking or boasting but I'm willing to discuss it openly and be subject to humiliation.

You may also wonder why I would do this. If the techniques work I would ask the developpers to
open some of their private code. Also I want to leverage the cuda expertise available to improve
the product for everyone.

I'll post one technique in a while that requires CPU support, but I have to dig out my old processor manuals
for a refresher. This one is something compilers should do since it is documented.

Edit: I've reviewed my manuals and checked out haswell optimization maual online and didn't see
anything at first glance that indicates they have the support to do the following:

allocate load

A special form of the load instruction will cause a line of cache to be allocated without accessing memory to fill it.
This is usefull when allocating mem and you don't care what data is in it. Memory isn't accessed unless the cache line
gets flushed for other reasons. An if the buffer is used for a short time it may never need to access memory.

Example:

uint32_t* p = allocate_load( size )
crunch some data
free_and_invalidate_cache ( &p )

even better lock  the cache line to guarantee it never gets flushed, effectively expanding the
register set.

struct * my_struct  my_regs = alocate_load _and_lock_line( size )
do stuff with my_regs.r1 etc
free_unlock_and_invalidate_cache( &my_regs );

What do you think?

[s7icky]

Im selling popcorn for 0.1 BTC with 500% faster pop.