Here's the patch to make it work with OpenCL 1.1 (and therefore Nvidia cards). Replace function OpenCL::WriteBufferPattern in file AppOpenCL.cpp with the following code: void OpenCL::WriteBufferPattern(uint device_num, string buffername, size_t data_length, void* pattern, size_t pattern_length)
{
_clState& GPUstate = GPUstates[device_num];
if (GPUstate.buffers[buffername] == NULL)
cout << "Buffer " << buffername << " not found on GPU #" << device_num << endl;
#ifdef CL_VERSION_1_2
cl_int status = clEnqueueFillBuffer(GPUstate.commandQueue, GPUstate.buffers[buffername], pattern, pattern_length, 0, data_length, 0, NULL, NULL);
#else
uint8_t buffer[data_length];
for(uint16_t i=0; i<(data_length / pattern_length);i++)
memcpy((&buffer[i*pattern_length]), pattern, pattern_length);
cl_int status = clEnqueueWriteBuffer(GPUstate.commandQueue, GPUstate.buffers[buffername], CL_TRUE, 0, data_length, buffer, 0, NULL, NULL);
#endif
if (globalconfs.coin.config.GetValue<bool>("opencldebug"))
cout << "Write buffer pattern " << buffername << ", " << pattern_length << " bytes. Status: " << status << endl;
}
This runs for me, but I am getting 0 fermats/s, 0 gandalfs/s.
0 TOTAL
most likely because my card it too old and I had to set worksize 64 in primecoin.conf. If you have a newer Nvidia card (with a "compute capability version" >= 2.0 according to http://en.wikipedia.org/wiki/CUDA#Supported_GPUs ), try to set worksize 512 and see what this gives you. Hi, I got it to compile, but I modifided the line to due to compile errors. |
|
|
|
|
Any word on getting this to work on NVIDIA cards? From what I understand it's because the nvidia cards don't support opencl 1.2 (yet?). Any potential workarounds on windows or linux?
|
|
|
|
|
Have there been any improved pool miners recently?
Is v7.1 still the latest? And more importantly has it been ported to linux? I've been mining using erkmos' build, but the last update was on the 20th. Is there a thread that I am missing?
|
|
|
|
I'm getting "The procedure entry point clEnqueueFillBuffer could not be located in the dynamic link library" and crash. Win8 x64 & GTX 560
You need to upgrade your amd/nvidia drivers to fix that one, but it will probably still crash after that. Same here on Win8 x64 Nvidia C2075. The driver was updated in July and should support openCL 1.1 You assume clEnqueueFillBuffer works in OCL 1.1 It seems clEnqueueFillBuffer indeed requires OpenCL 1.2. I could do without it but it's kind of a pain in the ass. Nvidia doesn't have drivers that support OpenCL 1.2? Ahhhh so this is why it crashes on OSX 10.8, which ships with OpenCL 1.1 and there is no newer driver available. According to Wikipedia, when OS X 10.9 Mavericks ships this Fall, OpenCL 1.2 and OpenGL 4.1 will be built in. So people with Nvidia cards are just SOL until 10.9 comes out? |
|
|
|
where can I download this gpu miner  If you want money you dont get money. Nowhere. No one will give you for free, because we paid for it. But you are smart guy.  Hahahaha, so from reading this thread, most people can't get it to compile, or run, or when they do it crashes almost immediately. The performance is said to be roughly equal the the performance of a CPU. Yet you are telling people to pay you for something you didn't even make. This is really really fucking funny. You should get a scammer tag for this kind of behaviour. It's GNU/GPL world, dude. RedHat and many other linux distributors dooes almost same thing. They sell they sell other people's achievements for big money. lol, says the guy selling the miner for 0.49 BTC in his sig. |
|
|
|
Hi, I am running jhprimeminer the precompiled linux version by bengtåke or clintar (not trough wine!). I am running it on my VPS Ubuntu 13.04 x64 linux server and I noticed a big difference in pps when: - I run it foreground = average 1000-2000 pps - or when is it running the background = average 3000-4000 pps So if I leave my ssh session open and the jhprimeminer is running in foreground I get double speed. But I can't have it open all the time... I also tried to run it as a daemon (daemonize), and also used renice commend to se the priority to -20 (maximum). With no significant improvement. Does anybody know why is this happening and if it is possible to run it at full speed even if it's in the background? I don't understand. From what you say, it runs faster in the background. I'm running a build from erkmos natively on ubunutu. I've only run it in the foreground, but if running in the back ground makes it faster, i'll do that. That said, I can't think of any reason that running in the background would be that much faster. Assuming the printing to the console is negligible. |
|
|
|
I got the same error on OSX.
Try to install libgmp3-dev I use MacPorts as my package manager and I have the 'gmp' (@5.0.5) installed already. Try adding -L/opt/local/lib to your LIBPATHS and -I/opt/local/include to INCLUDEPATHS variables in Makefile. P.S. You should try Homebrew its much better than macports. So I just updated to the newest version (since this morning) and it compiled on the first try with some warnings. |
|
|
|
I got the same error on OSX.
Try to install libgmp3-dev I use MacPorts as my package manager and I have the 'gmp' (@5.0.5) installed already. |
|
|
|
Anyone compiled it on smos yet? I've installed openssl cpulimit libgpm-dev but it's still missing something... make
g++ -c -mtune=native -O3 -Wall -Wextra -I/usr/local/include -I/usr/include src/primecoinMiner/bn2.cpp -o src/primecoinMiner/bn2.o
In file included from src/primecoinMiner/global.h:57,
from src/primecoinMiner/bn2.cpp:1:
src/primecoinMiner/prime.h:18:19: error: gmpxx.h: No such file or directory
In file included from src/primecoinMiner/bn2.cpp:1:
src/primecoinMiner/global.h:65:17: error: gmp.h: No such file or directory
In file included from src/primecoinMiner/global.h:57,
from src/primecoinMiner/bn2.cpp:1:
src/primecoinMiner/prime.h:29: error: ‘mpz_class’ does not name a type
src/primecoinMiner/prime.h:30: error: ‘mpz_class’ does not name a type
src/primecoinMiner/prime.h:31: error: ‘mpz_class’ does not name a type
src/primecoinMiner/prime.h:32: error: ‘mpz_class’ does not name a type
src/primecoinMiner/prime.h:33: error: ‘mpz_class’ does not name a type
src/primecoinMiner/prime.h:49: error: ‘mpz_class’ has not been declared
src/primecoinMiner/prime.h:53: error: variable or field ‘PrimorialAt’ declared void
src/primecoinMiner/prime.h:53: error: ‘mpz_class’ was not declared in this scope
src/primecoinMiner/prime.h:53: error: ‘bn’ was not declared in this scope
src/primecoinMiner/prime.h:53: error: ‘mpz_class’ was not declared in this scope
src/primecoinMiner/prime.h:53: error: ‘mpzPrimorial’ was not declared in this scope
In file included from src/primecoinMiner/global.h:57,
from src/primecoinMiner/bn2.cpp:1:
src/primecoinMiner/prime.h:63: error: ISO C++ forbids declaration of ‘mpz_class’ with no type
src/primecoinMiner/prime.h:63: error: expected ‘,’ or ‘...’ before ‘&’ token
src/primecoinMiner/prime.h:64: error: ISO C++ forbids declaration of ‘mpz_class’ with no type
src/primecoinMiner/prime.h:64: error: expected ‘,’ or ‘...’ before ‘&’ token
src/primecoinMiner/prime.h:241: error: ‘mpz_class’ does not name a type
src/primecoinMiner/prime.h:242: error: ‘mpz_class’ does not name a type
src/primecoinMiner/prime.h:315: error: ‘mpz_class’ has not been declared
src/primecoinMiner/prime.h:315: error: ‘mpz_class’ has not been declared
src/primecoinMiner/prime.h:348: error: ‘mpz_class’ has not been declared
src/primecoinMiner/prime.h:348: error: ‘mpz_class’ has not been declared
src/primecoinMiner/prime.h: In constructor ‘CSieveOfEratosthenes::CSieveOfEratosthenes(unsigned int, unsigned int, int&, int&)’:
src/primecoinMiner/prime.h:321: error: ‘class CSieveOfEratosthenes’ has no member named ‘mpzHash’
src/primecoinMiner/prime.h:322: error: ‘class CSieveOfEratosthenes’ has no member named ‘mpzFixedMultiplier’
src/primecoinMiner/prime.h: In member function ‘void CSieveOfEratosthenes::Init(unsigned int, unsigned int, int&, int&, unsigned int)’:
src/primecoinMiner/prime.h:352: error: ‘class CSieveOfEratosthenes’ has no member named ‘mpzHash’
src/primecoinMiner/prime.h:353: error: ‘class CSieveOfEratosthenes’ has no member named ‘mpzFixedMultiplier’
src/primecoinMiner/prime.h: At global scope:
src/primecoinMiner/prime.h:458: error: variable or field ‘mpz_set_uint256’ declared void
src/primecoinMiner/prime.h:458: error: ‘mpz_t’ was not declared in this scope
src/primecoinMiner/prime.h:458: error: expected primary-expression before ‘&’ token
src/primecoinMiner/prime.h:458: error: ‘u’ was not declared in this scope
In file included from src/primecoinMiner/bn2.cpp:1:
src/primecoinMiner/global.h:214: error: ‘mpz_class’ does not name a type
src/primecoinMiner/bn2.cpp:116: warning: unused parameter ‘a’
src/primecoinMiner/bn2.cpp:195: warning: unused parameter ‘ctx’
make: *** [src/primecoinMiner/bn2.o] Error 1
I got the same error on OSX. |
|
|
|
|
Keep on keeping on, I guess.
|
|
|
|
Actually, I'm having the same problem with v6 as I previously had with v5, where it runs fine for a while, but then drops down to about 40% load. Also, my Val/h is much lower then before, previously it was around 17-18 with v4 and with v6 after an hour or so it's hovering around 5. Any idea what is going on mumus?
Make sure you aren't throttling down because of heat issues. I don't think that is a problem either, the 8 cores are all within 57 to 66 C. I don't think that should be a problem. |
|
|
|
The miner seems to be running, but it keeps throwing an error:
err:ntdll:RtlpWaitForCriticalSection section 0x140061040 "?" wait timed out in thread 002f, blocked by 0032, retrying (60 sec)
Any idea why?
Edit: It doesn't appear to matter too much seeing as I just found a block for the pool.
I have no idea but I will try to find out what can cause this. This is happening in v6 running through wine on my linux box (Ubuntu 13.04). This is the same machine that were the load is dropping. I am working around that problem by killing and restarting the miner every half hour, BTW). I've tried to test also on a 12 core Linux (Ubuntu 12.10) machine using wine and I didn't got so far. It's crashing all the time so I'm also restarting using a script. The problem that I have only 1Gb RAM + 1Gb swap, and it seems that it is killed because of low memory. Under windows I can't see that has any memory leak and usually it uses about 30Mb (depending on the parameters). More investigation is needed. the problem that I'm less familiar to work under Linux especially when it comes to coding. I can fairly safely rule out memory issues. I have 48 GB in that machine with around 32 GB free. Is it possible that it is related to storage in the L1 cache? I am also using the computer for other work stuff (it's my work desktop). I no almost nothing about L1, so this probably sounds dumb. |
|
|
|
The miner seems to be running, but it keeps throwing an error:
err:ntdll:RtlpWaitForCriticalSection section 0x140061040 "?" wait timed out in thread 002f, blocked by 0032, retrying (60 sec)
Any idea why?
Edit: It doesn't appear to matter too much seeing as I just found a block for the pool.
I have no idea but I will try to find out what can cause this. This is happening in v6 running through wine on my linux box (Ubuntu 13.04). This is the same machine that were the load is dropping. I am working around that problem by killing and restarting the miner every half hour, BTW). |
|
|
|
We know the diff is not linear.
But it seems when we go to a new integer, the difficulty is lower.
Whit x.95 is more difficult than x+1. (8,95 more difficult than 9)
Is this right?
It should not be but it probably is. In my understanding, a block should be valid if the length of its chain is either greater than the current difficulty, or its chain length is equal to the integer part of the current difficulty and the fractional bits part (I don't know exactly how that is determined) is greater than the fractional part of the difficulty. So at difficulty 9.5, every block with a chain length of 10 or more should be accepted, and approximately half of the length 9 blocks should be accepted. However, my intuition is that the difficulty graph with its sharp increases whenever an integral difficulty value has been reached implies that the fractional bits are being considered even when the chain length is greater than the difficulty. Maybe someone could look at the source to see how it's actually done (I'm at work right now and can't do it, and when I last looked at the source I found this fractional part business really confusing and skipped it.) Onkel Paul From the design paper: http://ppcoin.org/static/primecoin-paper.pdfLet k be the prime chain length. The prime chain is p0, p1, …, pk-1. Let r be the Fermat
test remainder of the next number in chain pk. Now pk/r is used to measure the
difficulty of the chain. Even though the distribution of r/pk is not strictly uniform, but
experiments have shown that the difficulty adjustment behavior is reasonably good in
practice. The prime chain length is then computed with a fractional length part:
d = k + (pk-r)/pk
Note if pk passes probable primality tests then it should be considered as a chain of
higher integral length.
A continuous length target adjustment is employed with similar features to the difficulty
adjustment in ppcoin (King 2012). Length target is stepped up or down through integral
boundaries during length target adjustment, at fixed step-up/step-down threshold of
255/256 <-> 1.
|
|
|
|
|
The miner seems to be running, but it keeps throwing an error:
err:ntdll:RtlpWaitForCriticalSection section 0x140061040 "?" wait timed out in thread 002f, blocked by 0032, retrying (60 sec)
Any idea why?
Edit: It doesn't appear to matter too much seeing as I just found a block for the pool.
|
|
|
|
Forget about it guys, this miner was never gonna happen. Obviously the people who started and promoted this thread had no idea what they were getting into.
The reaper guy's miner is probably the only GPU miner we will EVER be seeing.
So it would seem, sadly. I hope once mrtlt's is open sourced we can get some real community development going for the good of the coin. Hopefully by that point, I will have finish the crunch at work and I can try to dive in as well. |
|
|
|
|
ignored. Whew, much better.
|
|
|
|
Actually, I'm having the same problem with v6 as I previously had with v5, where it runs fine for a while, but then drops down to about 40% load. Also, my Val/h is much lower then before, previously it was around 17-18 with v4 and with v6 after an hour or so it's hovering around 5. Any idea what is going on mumus?
I've tried to find out but I had no problem running on my machines. I will try again on one of my new Linux machine. So I realized that I was running an older version of wine on my linux box (1.4.0) so I'm in the process of updating, I'll let you know if it fixes it. @Sterling, It's hard for me to tell if my xpm/hr is constant, because I have three machines that are workers for the same account, so I can't really tell what is coming from where on ypool. So it turns out I had a bunch of pending updates (~200), so I updated them and thus far it seems that my Val/h is back around where it was. I'll let you know if the CPU load drops again. So it keeps dropping the load. It's at about 25% now. I can't figure out why, I have plenty of free memory (36 GB) and there is no noticeable IO or network load. Any ideas? |
|
|
|
Actually, I'm having the same problem with v6 as I previously had with v5, where it runs fine for a while, but then drops down to about 40% load. Also, my Val/h is much lower then before, previously it was around 17-18 with v4 and with v6 after an hour or so it's hovering around 5. Any idea what is going on mumus?
I've tried to find out but I had no problem running on my machines. I will try again on one of my new Linux machine. So I realized that I was running an older version of wine on my linux box (1.4.0) so I'm in the process of updating, I'll let you know if it fixes it. @Sterling, It's hard for me to tell if my xpm/hr is constant, because I have three machines that are workers for the same account, so I can't really tell what is coming from where on ypool. So it turns out I had a bunch of pending updates (~200), so I updated them and thus far it seems that my Val/h is back around where it was. I'll let you know if the CPU load drops again. |
|
|
|
Actually, I'm having the same problem with v6 as I previously had with v5, where it runs fine for a while, but then drops down to about 40% load. Also, my Val/h is much lower then before, previously it was around 17-18 with v4 and with v6 after an hour or so it's hovering around 5. Any idea what is going on mumus?
I've tried to find out but I had no problem running on my machines. I will try again on one of my new Linux machine. So I realized that I was running an older version of wine on my linux box (1.4.0) so I'm in the process of updating, I'll let you know if it fixes it. @Sterling, It's hard for me to tell if my xpm/hr is constant, because I have three machines that are workers for the same account, so I can't really tell what is coming from where on ypool. |
|
|
|
|
Show Posts
