RPC Miners (CPU/4way/CUDA/OpenCL) - page 11.

dbitcoin

hero member

Activity: 742

Merit: 500

BTCDig - mining pool

Problem with linux compilation.

Ubuntu 10.04, gcc 4.4.3
cmake buid configs without problems, all dependencies installed.

> make rpcminer

Code:

In file included from /src/bitcoin-rpc-miner/src/rpcminer/../minercommon/minerheaders.h:58,
from /src/bitcoin-rpc-miner/src/rpcminer/rpcminerclient.h:26,
from /src/bitcoin-rpc-miner/src/rpcminer/rpcminerclient.cpp:19:
/src/bitcoin-rpc-miner/src/rpcminer/../minercommon/../serialize.h: In function ‘unsigned int GetSizeOfCompactSize(uint64)’:
/src/bitcoin-rpc-miner/src/rpcminer/../minercommon/../serialize.h:167: error: ‘USHRT_MAX’ was not declared in this scope
/src/bitcoin-rpc-miner/src/rpcminer/../minercommon/../serialize.h:168: error: ‘UINT_MAX’ was not declared in this scope
/src/bitcoin-rpc-miner/src/rpcminer/../minercommon/../serialize.h: In function ‘void WriteCompactSize(Stream&, uint64)’:
/src/bitcoin-rpc-miner/src/rpcminer/../minercommon/../serialize.h:180: error: ‘USHRT_MAX’ was not declared in this scope
/src/bitcoin-rpc-miner/src/rpcminer/../minercommon/../serialize.h:187: error: ‘UINT_MAX’ was not declared in this scope
....

zholer

newbie

Activity: 4

Merit: 0

I first tried to build this on my ubuntu 10.04 system, but it failed with the same errors as flags8192. Experimenting with installing different version of wxWidgets did not help.
I finally git cloned from git://github.com/doublec/bitcoin-pool which seems to use pudinpop's code. This finally compiled on my machine as bitcoinr binary, but I can't figure out how to pass username as a parameter to it. Nothing in the readme file lets me do this.

Tolsi

full member

Activity: 171

Merit: 100

Here is 0 khash/s on opencl version with ati hd6870 again.

puddinpop

member

Activity: 103

Merit: 17

Updated source to bitcoin SVN rev 241. Downloads available in the first post.

WSDN

sr. member

Activity: 493

Merit: 250

IDENA.IO - Proof-Of-Person Blockchain

Thanks friend continue working u are exellent! thanks again!

puddinpop

member

Activity: 103

Merit: 17

Quote from: WSDN on February 13, 2011, 04:38:06 PM

hi @puddinpop i have the next problem :$ i have a new xp64 and downloading again yours miners with the old version of minners i not have any problem, but with new have this screen.

I reinstall my windows and deleted all my files for this i need to download again the miners if you have the old versions please put a link =)
thanks in advance

You need to install the Microsoft Visual C++ 2005 Redistributable Package (x86)

WSDN

sr. member

Activity: 493

Merit: 250

IDENA.IO - Proof-Of-Person Blockchain

hi @puddinpop i have the next problem :$ i have a new xp64 and downloading again yours miners with the old version of minners i not have any problem, but with new have this screen.

I reinstall my windows and deleted all my files for this i need to download again the miners if you have the old versions please put a link =)
thanks in advance

flags8192

newbie

Activity: 4

Merit: 0

like compile, but when you start error

./rpcminer-cuda -url=http://mining.bitcoin.cz:8332/ -user=xxxxxxx.xxxxxxxxx -password=xxxxxxxxx -statsurl=http://mining.bitcoin.cz/stats/json/
Client will start 1 miner threads
Work will be refreshed every 4000 ms
1 CUDA GPU devices found
Setting CUDA device to first device found
Loading module bitcoinminercuda_11.cubin
Unable to load CUDA module

what is Problem?

sonihr

sr. member

Activity: 454

Merit: 250

THE TIME HAS COME...

problem is the exactly the same as Epicenter. I have ATI radeon 4350 windows 7 64bit tried sdk 2.1 2.2 2.3...

nlai

newbie

Activity: 11

Merit: 0

What's the difference between the 4way and the crypto?

I'm using an Intel i7

Tolsi

full member

Activity: 171

Merit: 100

I have ATI HD6870, ATI Stream 2.3 SDK, Win7sp1 x64 and I have a same problem as epicenter. Then I replaced bitcoinmineropencl.cl from r191 (bitcoin-remote-20101215-win32bin.zip) (newer are don't work, 0 khash/s), now I see:

Code:

>rpcminer-opencl.exe -url=http://mining.bitcoin.cz:8332 -user=*** -password=*** -statsurl=http://mining.bitcoin.cz/stats/json/
Client will start 1 miner threads
Work will be refreshed every 4000 ms
1 OpenCL platforms found
1 OpenCL GPU devices found on platform 0
Setting OpenCL device to first device found
Available extensions : cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_3d_image_writes cl_khr_byte_addressable_store cl_khr_gl_sharing cl_amd_device_attribute_query cl_amd_printf cl_amd_media_ops cl_amd_popcnt cl_khr_d3d10_sharing
Defining AMDMEDIAOPS
Create context rval=0
Target = ffffffffffffffffffffffffffffffffffffffffffffffffffffffff00000000
Create command queue rval=0
Creating program with source
Building program with options -D AMDMEDIAOPS
Build program rval=0
build STATUS:
build LOG:C:\Users\...\AppData\Local\Temp\OCL7B85.tmp.cl(21): warning: type qualifier
   specified more than once
  __constant const uint f1=0xFF000000;
   ^

C:\Users\...\AppData\Local\Temp\OCL7B85.tmp.cl(22): warning: type qualifier
   specified more than once
  __constant const uint f2=0x00FF0000;
   ^

C:\Users\...\AppData\Local\Temp\OCL7B85.tmp.cl(23): warning: type qualifier
   specified more than once
  __constant const uint f3=0x0000FF00;
   ^

C:\Users\...\AppData\Local\Temp\OCL7B85.tmp.cl(24): warning: type qualifier
   specified more than once
  __constant const uint f4=0x000000FF;
   ^

C:\Users\...\AppData\Local\Temp\OCL7B85.tmp.cl(25): warning: type qualifier
   specified more than once
  __constant const uint f5=0xFFFFFFFF;
   ^

┌М♠°Ї√☻►┴i♥
Create kernel rval=0
Done allocating OpenCL resources for (16,16)
Finding best configuration step end (16,16) 4ms prev best=9223372036854775807ms

Done allocating OpenCL resources for (16,32)
Finding best configuration step end (16,32) 1ms prev best=9223372036854775807ms

Done allocating OpenCL resources for (16,64)
Finding best configuration step end (16,64) 0ms prev best=9223372036854775807ms

Done allocating OpenCL resources for (16,128)
Finding best configuration step end (16,128) 1ms prev best=9223372036854775807m
s
Done allocating OpenCL resources for (16,256)
Finding best configuration step end (16,256) 1ms prev best=9223372036854775807m
s
Done allocating OpenCL resources for (32,16)
Finding best configuration step end (32,16) 1ms prev best=9223372036854775807ms

Done allocating OpenCL resources for (32,32)
Finding best configuration step end (32,32) 1ms prev best=9223372036854775807ms

Done allocating OpenCL resources for (32,64)
Finding best configuration step end (32,64) 2ms prev best=9223372036854775807ms

Done allocating OpenCL resources for (32,128)
Finding best configuration step end (32,128) 0ms prev best=9223372036854775807m
s
Done allocating OpenCL resources for (32,256)
Finding best configuration step end (32,256) 1ms prev best=9223372036854775807m
s
Done allocating OpenCL resources for (64,16)
Finding best configuration step end (64,16) 1ms prev best=9223372036854775807ms

Done allocating OpenCL resources for (64,32)
Finding best configuration step end (64,32) 1ms prev best=9223372036854775807ms

Done allocating OpenCL resources for (64,64)
Finding best configuration step end (64,64) 1ms prev best=9223372036854775807ms

Done allocating OpenCL resources for (64,128)
Finding best configuration step end (64,128) 0ms prev best=9223372036854775807m
s
Done allocating OpenCL resources for (64,256)
Finding best configuration step end (64,256) 1ms prev best=9223372036854775807m
s
Done allocating OpenCL resources for (128,16)
Finding best configuration step end (128,16) 1ms prev best=9223372036854775807m
s
Done allocating OpenCL resources for (128,32)
Finding best configuration step end (128,32) 0ms prev best=9223372036854775807m
s
Done allocating OpenCL resources for (128,64)
Finding best configuration step end (128,64) 1ms prev best=9223372036854775807m
s
Done allocating OpenCL resources for (128,128)
Finding best configuration step end (128,128) 0ms prev best=9223372036854775807
ms
Done allocating OpenCL resources for (128,256)
Finding best configuration step end (128,256) 1ms prev best=9223372036854775807
ms
Done allocating OpenCL resources for (16,16)
37238 khash/s
49181 khash/s
48631 khash/s
Server Stats : 661 active workers at an estimated 0 khash/s
0 shares in current round, which started
....

I left it overnight and none blocks aren't resolved. In the morning I saw the same: 0 shares in current round, which started. In Bitcoin Pooled Mining I see Score 0.0000.
CPU miner works fine.
Today I will try to change bitcoinmineropencl.cl, but promise nothing, I do not know c.

And in http://pastebin.com/AvymGnMJ i see:

Model Sp Clk W Mhash/s Mhash/W
6870 1120 900 151 232.47 1.540

but i have max 60000 khash/s and 60% cpu load, 18% gpu load (not full) Sad

screenshot

ps with python OpenCL bitcoin miner i have ~230Mhash/s

epicenter

newbie

Activity: 10

Merit: 0

I have an ATI Radeon HD 5800
I tried running the cpu miner and it worked, but opencl wouldn't run.
So got the stream 2.3k SDK and now opencl can run, BUT
I cant get any work done. Same command line parameters in the cpu miner gets work done (~4500 hash/s)

Quote

C:\bitcoin-rpcminer-20110128-win32bin>rpcminer-opencl -url=http://mining.bitcoin.cz:8332 -user=XXX -password=XXX -gpu=0
Client will start 1 miner threads
Work will be refreshed every 4000 ms
1 OpenCL platforms found
1 OpenCL GPU devices found on platform 0
Setting OpenCL device to device 0
Available extensions : cl_amd_fp64 cl_khr_global_int32_base_atomics cl_khr_globa
l_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_exte
nded_atomics cl_khr_3d_image_writes cl_khr_byte_addressable_store cl_khr_gl_shar
ing cl_amd_device_attribute_query cl_amd_printf cl_amd_media_ops cl_amd_popcnt c
l_khr_d3d10_sharing
Defining AMDMEDIAOPS
Create context rval=0
Target = ffffffffffffffffffffffffffffffffffffffffffffffffffffffff00000000
Create command queue rval=0
Creating program with source
Building program with options -D AMDMEDIAOPS
0 khash/s
0 khash/s
0 khash/s
0 khash/s
0 khash/s
0 khash/s
0 khash/s
0 khash/s
0 khash/s
0 khash/s
0 khash/s

Compilation terminated.

flags8192

newbie

Activity: 4

Merit: 0

Please help to compile, it fails, an error:

~/bitcoin-remote-rpc-20110128-src$ make
[ 1%] Building NVCC (Device) object cmake-bitcoin/./bitcoin_generated_bitcoinminercuda.cu.o
Scanning dependencies of target bitcoin
[ 2%] Building CXX object cmake-bitcoin/CMakeFiles/bitcoin.dir/__/src/db.cpp.o
In file included from /usr/local/include/wx-2.8/wx/wx.h:15,
from /home/flags8192/bitcoin-remote-rpc-20110128-src/src/headers.h:33,
from /home/flags8192/bitcoin-remote-rpc-20110128-src/src/db.cpp:5:
/usr/local/include/wx-2.8/wx/defs.h:42: error: #error "No Target! You should use wx-config program for compilation flags!"
In file included from /usr/local/include/wx-2.8/wx/cursor.h:24,
from /usr/local/include/wx-2.8/wx/event.h:22,

ps: ubuntu 10.10

t3h

newbie

Activity: 19

Merit: 0

The CUDA miner is working nicely on my GTX470. 80000kHash/sec stock, 105000kHash/sec overclocked to 775mhz (works fine on stock vCore). I can push ~115k out of it by overclocking further, but it gets unstable unless I raise the vCore, then my 500w PSU doesn't like it

Should have gotten into this earlier, it takes ages to mine anything now :|

Just one feature request: when you overclock too far, the display driver crashes and reboots. Can you detect this and reload the CUDA part? It keeps going with a stupidly large number of kHashes/sec (I think it's just silently failing), and never finds anything, so you come back and find that you haven't been mining for a few hours.

FooDSt4mP

full member

Activity: 182

Merit: 100

Ok, I did a make rpcminer, and it builds the kernel and everything fine, but fails on linking (undefined reference to `cuInit'). I have added the cuda lib directory to the ld config file and ran ldconfig as root. I also tried setting the LD_LIBRARY_PATH with no success. Perhaps there's a -lcuda missing somewhere, but I've never worked with CMake, so I have no idea where look. Any ideas?

puddinpop

member

Activity: 103

Merit: 17

Quote from: FooDSt4mP on February 06, 2011, 04:47:26 PM

First off, thanks for the great work. When I saw this, I was really excited to start using my GPU to mine. Unfortunately, you are using cutil, which is not an official part of the CUDA SDK, and not in the linux version. My compile fails with fatal error: cutil_inline.h: No such file or directory, and I confirmed that cutil_inline.h not anywhere in the SDK directory. This is confirmed by a moderator at http://forums.nvidia.com/index.php?showtopic=192055 .

Quote

cutil is something used only for SDK samples to make code shorter and is not suitable for use in a real application. It is completely unsupported and not included with the CUDA toolkit
(don't use it, don't steal it from the SDK, just leave it alone and do error checking the right way) .

So my question is, how widespread is the usage of cutil in the code? If it's not too bad, I'll try to put together a patch, but this is the first time I've messed with CUDA. If it's not something I can handle, I'll happily donate a few bitcoins to someone (as soon as I can mine them, my cpu hasn't turned up anything yet) for helping me get this running.

The last release does not need cutil when compiling the CUDA RPC miner, and likewise the CUDA daemon or GUI miners, since they all share the same miner code. Only the remote miners still need it. If you are using CMake to create makefiles, it is likely you didn't turn off building of the remote miners.

FooDSt4mP

full member

Activity: 182

Merit: 100

First off, thanks for the great work. When I saw this, I was really excited to start using my GPU to mine. Unfortunately, you are using cutil, which is not an official part of the CUDA SDK, and not in the linux version. My compile fails with fatal error: cutil_inline.h: No such file or directory, and I confirmed that cutil_inline.h not anywhere in the SDK directory. This is confirmed by a moderator at http://forums.nvidia.com/index.php?showtopic=192055 .

Quote

cutil is something used only for SDK samples to make code shorter and is not suitable for use in a real application. It is completely unsupported and not included with the CUDA toolkit
(don't use it, don't steal it from the SDK, just leave it alone and do error checking the right way) .

So my question is, how widespread is the usage of cutil in the code? If it's not too bad, I'll try to put together a patch, but this is the first time I've messed with CUDA. If it's not something I can handle, I'll happily donate a few bitcoins to someone (as soon as I can mine them, my cpu hasn't turned up anything yet) for helping me get this running.

BitCoinPurse

newbie

Activity: 34

Merit: 0

Quote from: ledskof on February 05, 2011, 07:37:24 PM

Quote from: DarkMatter on February 05, 2011, 02:19:56 PM

Not a W7 issue, using it right now (64bit) and with nvidia 8800gt (cuda) and no khash drops when in background.

I thought he meant that the ATI Streaming SDK v 2.1 didn't take on his Windows 7 box. I had the same problem with 2.1 and 2.2, which is why I'm running 2.3.

That's was precisely it. I'm stuck with the 2.3 SDK, though I've tried for days to downgrade... I'm not in the mood to slick my OS...

ledskof

newbie

Activity: 42

Merit: 0

Quote from: DarkMatter on February 05, 2011, 02:19:56 PM

Not a W7 issue, using it right now (64bit) and with nvidia 8800gt (cuda) and no khash drops when in background.

I thought he meant that the ATI Streaming SDK v 2.1 didn't take on his Windows 7 box. I had the same problem with 2.1 and 2.2, which is why I'm running 2.3.

DarkMatter

member

Activity: 67

Merit: 10

Stop trying to steal my account, thanks.

Quote from: BitCoinPurse on February 04, 2011, 06:53:10 PM

Quote from: puddinpop on February 04, 2011, 03:16:06 PM

Quote from: BitCoinPurse on January 29, 2011, 06:36:38 PM

I doesn't seem to be "building the kernel"...

Did you get it to work with the older kernel?

No, I haven't been able to isolate the issue. I tried to downgrade to the ATI SDK 2.1, but it didn't take on my crappy W7 box.

Also, I'm way out of my league tweaking .cl files. Grin

I'm happy with my 50Khash for now...

Not a W7 issue, using it right now (64bit) and with nvidia 8800gt (cuda) and no khash drops when in background.

Topic: RPC Miners (CPU/4way/CUDA/OpenCL) - page 11. (Read 548522 times)