[ANN] cudaMiner & ccMiner CUDA based mining applications [Windows/Linux/MacOSX]

[cbuchner1]

dont know what i did but all of a sudden it compiled. when i run it however it is saying
[2014-01-08 19:29:35] Unable to query CUDA driver version! Is an nVidia driver installed?

now i do have an nvidia driver installed. what am i missing?

EDIT
just to show its installed, i ran nvcc --version and this is what i got;

nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2013 NVIDIA Corporation
Built on Wed_Jul_17_18:36:13_PDT_2013
Cuda compilation tools, release 5.5, V5.5.0

so it is 100% installed but cudaminer can not find it :s

you're confusing driver and compiler.

You need a working nvidia driver and the nVidia kernel module loaded.

The nvidia-331 driver (if installed via apt-get) should already be working. Alternatively the driver from the toolkit can be installed, but that requires an installation without a running X-Server.

[bigjme]

http://s27.postimg.org/wbcy0799v/Untitled.jpg

what exactly am i missing? the drivers i am using are from xedgers

[CaptainBeck]

I cant seem to get the wallet to sync for some reason, ive added nodes and nothing seems to happen. I knowits not my pc setting because all my other wallets connect.

[cbuchner1]

http://s27.postimg.org/wbcy0799v/Untitled.jpg

what exactly am i missing? the drivers i am using are from xedgers

that's a really good question. Have you tried running the display through the nVidia card, too?

Christian

[bigjme]

the display is run through the nvidia as i dont have onboard graphics....

[bathrobehero]

higher N factor means lower rewards for mining blocks and a higher memory requirement every time it increases. eventually it will get so high that gpus wont have enough memory to even hash once so the hashrate will drop very quickly. this is why cpu mining will carry on as desktop memory is cheap and can be replaced

Thanks for clearing that out! Though I have another question.
What if we have a scrypt-jane cryptocurrency with:
"Nfactor" : 6,
"N" : 128,
instead of YaCoin's 14/32768? What I mean is, can I finetune cudaMiner to hash faster on lower N factor cryptocurrencies, as it should be using less VRAM for the same hashrate or I'm completely off here?

[Lussekatten]

wow... suddenly everybody is talking about Yacoin and not LTC.... Anyhow...downloaded new drivers for my gtx 760. now my card runs smoothly 284Kh/s

[bigjme]

ok so my entire system has just died. so i have lost everything i did to get compiling working, my compiled version, and everything else that was on it. installing the newest 12.04 as we speak, hope to god it works. although i will probably end up spending another 6 hours trying to get cudaminer to compile

[cbuchner1]

What if we have a scrypt-jane cryptocurrency with:
"Nfactor" : 6,
"N" : 128,
instead of YaCoin's 14/32768? What I mean is, can I finetune cudaMiner to hash faster on lower N factor cryptocurrencies, as it should be using less VRAM for the same hashrate or I'm completely off here?

yes, lower NFactor uses very little RAM. Not sure how efficient it would be though.

[bigjme]

cbuchner1 can you help me quickly. im running the cudaminer compile again. and right now i am getting an error of

Code:

collect2: ld returned 1 exit status
make[2]: *** [cudaminer] Error 1
make[2]: Leaving directory `/home/jamie/.local/share/Trash/files/CudaMiner-master'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/home/jamie/.local/share/Trash/files/CudaMiner-master'
make: *** [all] Error 2

i think this is the last part that is an issue so some quick help would be great.

[cbuchner1]

i think this is the last part that is an issue so some quick help would be great.

Code:

collect2: ld returned 1 exit status

the linker has a problem, which must be listed somewhere in the lines above. maybe not finding the CUDA runtime libraries to link against?

[bigjme]

i just ran it again cbuchner1 so i could capture the command where it gives the error. here is the bit of the command that i can copy which might help

Code:

./legacy_kernel.cu(269): Warning: Cannot tell what pointer points to, assuming global memory space
./legacy_kernel.cu(270): Warning: Cannot tell what pointer points to, assuming global memory space
nvcc -g -O2 -Xptxas "-abi=no -v" -arch=sm_20 --maxrregcount=63 -I./compat/jansson -o fermi_kernel.o -c fermi_kernel.cu
ptxas warning : 'option -abi=no' might get deprecated in future
ptxas info    : 272 bytes gmem, 8192 bytes cmem[2], 16 bytes cmem[14]
ptxas info    : Compiling entry function '_Z29fermi_scrypt_core_kernelB_texILi0ELi1EEvPjj' for 'sm_20'
ptxas info    : Used 63 registers, 44 bytes cmem[0], 20 bytes lmem
ptxas info    : Compiling entry function '_Z25fermi_scrypt_core_kernelBILi0EEvPjj' for 'sm_20'
ptxas info    : Used 63 registers, 44 bytes cmem[0], 24 bytes lmem
ptxas info    : Compiling entry function '_Z25fermi_scrypt_core_kernelAILi0EEvPjj' for 'sm_20'
ptxas info    : Used 63 registers, 44 bytes cmem[0], 4 bytes lmem
ptxas info    : Compiling entry function '_Z29fermi_scrypt_core_kernelB_texILi0ELi2EEvPjj' for 'sm_20'
ptxas info    : Used 63 registers, 44 bytes cmem[0], 28 bytes lmem
ptxas info    : Compiling entry function '_Z29fermi_scrypt_core_kernelB_texILi1ELi1EEvPjj' for 'sm_20'
ptxas info    : Used 63 registers, 44 bytes cmem[0], 20 bytes lmem
ptxas info    : Compiling entry function '_Z25fermi_scrypt_core_kernelBILi1EEvPjj' for 'sm_20'
ptxas info    : Used 63 registers, 44 bytes cmem[0], 20 bytes lmem
ptxas info    : Compiling entry function '_Z25fermi_scrypt_core_kernelAILi1EEvPjj' for 'sm_20'
ptxas info    : Used 63 registers, 44 bytes cmem[0], 4 bytes lmem
ptxas info    : Compiling entry function '_Z29fermi_scrypt_core_kernelB_texILi1ELi2EEvPjj' for 'sm_20'
ptxas info    : Used 63 registers, 44 bytes cmem[0], 36 bytes lmem
nvcc -g -O2 -Xptxas "-abi=no -v" -arch=sm_30 --maxrregcount=32 -I./compat/jansson -o kepler_kernel.o -c kepler_kernel.cu
ptxas warning : 'option -abi=no' might get deprecated in future
ptxas info    : 272 bytes gmem, 8196 bytes cmem[3]
ptxas info    : Compiling entry function '_Z26kepler_scrypt_core_kernelBILi0ELi1EEvPjii' for 'sm_30'
ptxas info    : Used 30 registers, 340 bytes cmem[0]
ptxas info    : Compiling entry function '_Z26kepler_scrypt_core_kernelBILi0ELi0EEvPjii' for 'sm_30'
ptxas info    : Used 32 registers, 336 bytes cmem[0], 4 bytes lmem
ptxas info    : Compiling entry function '_Z26kepler_scrypt_core_kernelAILi0EEvPKjii' for 'sm_30'
ptxas info    : Used 30 registers, 336 bytes cmem[0]
ptxas info    : Compiling entry function '_Z26kepler_scrypt_core_kernelBILi0ELi2EEvPjii' for 'sm_30'
ptxas info    : Used 29 registers, 340 bytes cmem[0]
ptxas info    : Compiling entry function '_Z26kepler_scrypt_core_kernelBILi1ELi1EEvPjii' for 'sm_30'
ptxas info    : Used 31 registers, 340 bytes cmem[0]
ptxas info    : Compiling entry function '_Z26kepler_scrypt_core_kernelBILi1ELi0EEvPjii' for 'sm_30'
ptxas info    : Used 32 registers, 336 bytes cmem[0], 4 bytes lmem
ptxas info    : Compiling entry function '_Z26kepler_scrypt_core_kernelAILi1EEvPKjii' for 'sm_30'
ptxas info    : Used 32 registers, 336 bytes cmem[0]
ptxas info    : Compiling entry function '_Z26kepler_scrypt_core_kernelBILi1ELi2EEvPjii' for 'sm_30'
ptxas info    : Used 30 registers, 340 bytes cmem[0]
nvcc -g -O2 -Xptxas "-abi=no -v" -arch=sm_30 --maxrregcount=32 -I./compat/jansson -o test_kernel.o -c test_kernel.cu
ptxas warning : 'option -abi=no' might get deprecated in future
ptxas info    : 272 bytes gmem
ptxas info    : Compiling entry function '_Z24test_scrypt_core_kernelAPKjPj' for 'sm_30'
ptxas info    : Used 27 registers, 336 bytes cmem[0]
ptxas info    : Compiling entry function '_Z24test_scrypt_core_kernelBPjPKj' for 'sm_30'
ptxas info    : Used 32 registers, 336 bytes cmem[0]
nvcc -g -O2 -Xptxas "-abi=no -v" -arch=sm_35 --maxrregcount=32 -I./compat/jansson -o titan_kernel.o -c titan_kernel.cu
ptxas warning : 'option -abi=no' might get deprecated in future
ptxas info    : 272 bytes gmem, 4 bytes cmem[3]
ptxas info    : Compiling entry function '_Z25titan_scrypt_core_kernelBILi0EEvPjPKj' for 'sm_35'
ptxas info    : Used 32 registers, 336 bytes cmem[0]
ptxas info    : Compiling entry function '_Z25titan_scrypt_core_kernelAILi0EEvPKjPj' for 'sm_35'
ptxas info    : Used 27 registers, 336 bytes cmem[0]
ptxas info    : Compiling entry function '_Z25titan_scrypt_core_kernelBILi1EEvPjPKj' for 'sm_35'
ptxas info    : Used 31 registers, 336 bytes cmem[0]
ptxas info    : Compiling entry function '_Z25titan_scrypt_core_kernelAILi1EEvPKjPj' for 'sm_35'
ptxas info    : Used 27 registers, 336 bytes cmem[0]
g++  -g -O2 -pthread -L/usr/local/cuda/lib64  -o cudaminer cudaminer-cpu-miner.o cudaminer-util.o cudaminer-sha2.o cudaminer-scrypt.o salsa_kernel.o spinlock_kernel.o legacy_kernel.o fermi_kernel.o kepler_kernel.o test_kernel.o titan_kernel.o -L/usr/lib/x86_64-linux-gnu -lcurl -Wl,-Bsymbolic-functions -Wl,-z,relro compat/jansson/libjansson.a -lpthread  -lcudart -fopenmp  
cudaminer-cpu-miner.o: In function `miner_thread':
/home/jamie/Downloads/CudaMiner-master/cpu-miner.c:824: undefined reference to `scanhash_scrypt_jane'
collect2: ld returned 1 exit status
make[2]: *** [cudaminer] Error 1
make[2]: Leaving directory `/home/jamie/Downloads/CudaMiner-master'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/home/jamie/Downloads/CudaMiner-master'
make: *** [all] Error 2

the main error part seems to be

Code:

ptxas info    : Used 27 registers, 336 bytes cmem[0]
g++  -g -O2 -pthread -L/usr/local/cuda/lib64  -o cudaminer cudaminer-cpu-miner.o cudaminer-util.o cudaminer-sha2.o cudaminer-scrypt.o salsa_kernel.o spinlock_kernel.o legacy_kernel.o fermi_kernel.o kepler_kernel.o test_kernel.o titan_kernel.o -L/usr/lib/x86_64-linux-gnu -lcurl -Wl,-Bsymbolic-functions -Wl,-z,relro compat/jansson/libjansson.a -lpthread  -lcudart -fopenmp  
cudaminer-cpu-miner.o: In function `miner_thread':
/home/jamie/Downloads/CudaMiner-master/cpu-miner.c:824: undefined reference to `scanhash_scrypt_jane'
collect2: ld returned 1 exit status

so is this a g++ issue or something? as if has an undefined reference. i have tried running the make with g++ and gcc 4.4 and get the same error

[cbuchner1]

i just ran it again cbuchner1 so i could capture the command where it gives the error. here is the bit of the command that i can copy which might help

Code:

cudaminer-cpu-miner.o: In function `miner_thread':
/home/jamie/Downloads/CudaMiner-master/cpu-miner.c:824: undefined reference to `scanhash_scrypt_jane'
collect2: ld returned 1 exit status

so is this a g++ issue or something? as if has an undefined reference. i have tried running the make with g++ and gcc 4.4 and get the same error

this is caused by not running ./autogen.sh before ./configure

[bigjme]

i 100% ran both. in that order

ok so i just called ./autogen.sh again, i was just running the file before.
i am getting a response of
"./autogen.sh: line 1: aclocal: command not found


EDIT!!!!!!!!!!!
someone shoot me, i didnt have automake installed :p
compiles now haha

[cbuchner1]

i 100% ran both. in that order

ok so i just called ./autogen.sh again, i was just running the file before.
i am getting a response of
"./autogen.sh: line 1: aclocal: command not found

I guess you need to install the autoconf and automake packages via apt-get then

[bathrobehero]

yes, lower NFactor uses very little RAM. Not sure how efficient it would be though.

I think it's fairly efficient:


and considering that I have 3 cudaminer instances running on a single GTX 660, by fairly efficient I mean amazing! And yes, this is scrypt-jane.

If I would have a decent CPU, I could get more, my poor Athlon II X2 270 is getting stomped by cudaMiner, even though it's at 4.2 Ghz.

[cbuchner1]

yes, lower NFactor uses very little RAM. Not sure how efficient it would be though.

I think it's fairly efficient
and considering that I have 3 cudaminer instances running on a single GTX 660, by fairly efficient I mean amazing! And yes, this is scrypt-jane.

If I would have a decent CPU, I could get more, my poor Athlon II X2 270 is getting stomped by cudaMiner, even though it's at 4.2 Ghz.

ah yes. The Kekkac hashing on the CPU is singlethreaded, and non SSE optimized yet.

[cbuchner1]

Can someone make a short list of other scrypt-jane currencies please? I only heard about QQCoin so far.

[bigjme]

ok got it all working now. so i need to create the launch file now.
so i created a file called, cudaminer.sh with run permission. put the following code inside of it

Code:

./cudaminer --algo=scrypt-jane -H 0 -i 0 -d 0 -l T16x1 -o http://127.0.0.1:3339 -u user -p pass -D

run it and it doesnt show anything. but if i type it directly into the terminal it loads up and shows. and i am telling this to run in terminal when i click it. any ideas?

[bathrobehero]

ah yes. The Kekkac hashing on the CPU is singlethreaded, and non SSE optimized yet.

It's stressing the CPU in a weird way, I haven't really seen anything like it before. If I run 3-4 instances (or 2 with -i 0) I get massive mouse lag and eventually a BSOD about a timed out CPU interrupt while the GPU is almost idling.