Pages:
Author

Topic: Wolf's XMR GPU Miner - page 16. (Read 117592 times)

full member
Activity: 138
Merit: 100
September 16, 2016, 03:48:51 PM
#92
If I can ask for some assistance, Im a total noob to compiling, just started taking a C++ course this year. I have compiled a few things in ubuntu successfully but from the repository am i able to build this in windows? I tried using the VS command prompt with nmake and ran into gcc being unrecognized. I tried using cygwin, got a little way through, but that failed as well. Looking at the makefile it says -I..../sgminer-lin64/libcflags/. do i need to have those folders to compile this. I might be in over my head trying to compile this stuff.
Thanks for your time wolf for looking into our issues with the miner Smiley .

$ make
gcc -D_POSIX_SOURCE -D_GNU_SOURCE -O2 -s -I/home/wolf/miners/sgminer-builds/sgminer-lin64/include/ -L/home/wolf/miners/sgminer-builds/sgminer-lin64/lib -c -std=c11 crypto/aesb.c -o crypto/aesb.o
gcc -D_POSIX_SOURCE -D_GNU_SOURCE -O2 -s -I/home/wolf/miners/sgminer-builds/sgminer-lin64/include/ -L/home/wolf/miners/sgminer-builds/sgminer-lin64/lib -c -std=c11 crypto/aesb-x86-impl.c -o crypto/aesb-x86-impl.o
gcc -D_POSIX_SOURCE -D_GNU_SOURCE -O2 -s -I/home/wolf/miners/sgminer-builds/sgminer-lin64/include/ -L/home/wolf/miners/sgminer-builds/sgminer-lin64/lib -c -std=c11 crypto/c_blake256.c -o crypto/c_blake256.o
gcc -D_POSIX_SOURCE -D_GNU_SOURCE -O2 -s -I/home/wolf/miners/sgminer-builds/sgminer-lin64/include/ -L/home/wolf/miners/sgminer-builds/sgminer-lin64/lib -c -std=c11 crypto/c_groestl.c -o crypto/c_groestl.o
gcc -D_POSIX_SOURCE -D_GNU_SOURCE -O2 -s -I/home/wolf/miners/sgminer-builds/sgminer-lin64/include/ -L/home/wolf/miners/sgminer-builds/sgminer-lin64/lib -c -std=c11 crypto/c_keccak.c -o crypto/c_keccak.o
gcc -D_POSIX_SOURCE -D_GNU_SOURCE -O2 -s -I/home/wolf/miners/sgminer-builds/sgminer-lin64/include/ -L/home/wolf/miners/sgminer-builds/sgminer-lin64/lib -c -std=c11 crypto/c_jh.c -o crypto/c_jh.o
gcc -D_POSIX_SOURCE -D_GNU_SOURCE -O2 -s -I/home/wolf/miners/sgminer-builds/sgminer-lin64/include/ -L/home/wolf/miners/sgminer-builds/sgminer-lin64/lib -c -std=c11 crypto/c_skein.c -o crypto/c_skein.o
gcc -D_POSIX_SOURCE -D_GNU_SOURCE -O2 -s -I/home/wolf/miners/sgminer-builds/sgminer-lin64/include/ -L/home/wolf/miners/sgminer-builds/sgminer-lin64/lib -c -std=c11 crypto/oaes_lib.c -o crypto/oaes_lib.o
gcc -D_POSIX_SOURCE -D_GNU_SOURCE -O2 -s -I/home/wolf/miners/sgminer-builds/sgminer-lin64/include/ -L/home/wolf/miners/sgminer-builds/sgminer-lin64/lib -c -std=c11 -maes cryptonight.c -o cryptonight.o
cryptonight.c: In function ‘cryptonight_hash_aesni’:
cryptonight.c:383:18: warning: initialization from incompatible pointer type
  uint64_t *dst = &ctx->long_state[c[0] & 0x1FFFF0];
                  ^
cryptonight.c:426:10: warning: passing argument 1 of ‘keccakf’ from incompatible pointer type
  keccakf(&ctx->state.hs, 24);
          ^
In file included from cryptonight.c:13:0:
crypto/c_keccak.h:22:6: note: expected ‘uint64_t *’ but argument is of type ‘union hash_state *’
 void keccakf(uint64_t st[25], int norounds);
      ^
cryptonight.c: In function ‘cryptonight_ctx’:
cryptonight.c:435:2: warning: implicit declaration of function ‘mmap’ [-Wimplicit-function-declaration]
  ret = mmap(0, sizeof(*ret), PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS|MAP_HUGETLB|MAP_POPULATE, 0, 0);
  ^
cryptonight.c:435:30: error: ‘PROT_READ’ undeclared (first use in this function)
  ret = mmap(0, sizeof(*ret), PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS|MAP_HUGETLB|MAP_POPULATE, 0, 0);
                              ^
cryptonight.c:435:30: note: each undeclared identifier is reported only once for each function it appears in
cryptonight.c:435:40: error: ‘PROT_WRITE’ undeclared (first use in this function)
  ret = mmap(0, sizeof(*ret), PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS|MAP_HUGETLB|MAP_POPULATE, 0, 0);
                                        ^
cryptonight.c:435:52: error: ‘MAP_PRIVATE’ undeclared (first use in this function)
  ret = mmap(0, sizeof(*ret), PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS|MAP_HUGETLB|MAP_POPULATE, 0, 0);
                                                    ^
cryptonight.c:435:64: error: ‘MAP_ANONYMOUS’ undeclared (first use in this function)
  ret = mmap(0, sizeof(*ret), PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS|MAP_HUGETLB|MAP_POPULATE, 0, 0);
                                                                ^
cryptonight.c:435:78: error: ‘MAP_HUGETLB’ undeclared (first use in this function)
  ret = mmap(0, sizeof(*ret), PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS|MAP_HUGETLB|MAP_POPULATE, 0, 0);
                                                                              ^
cryptonight.c:435:90: error: ‘MAP_POPULATE’ undeclared (first use in this function)
  ret = mmap(0, sizeof(*ret), PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS|MAP_HUGETLB|MAP_POPULATE, 0, 0);
                                                                                          ^
cryptonight.c:436:13: error: ‘MAP_FAILED’ undeclared (first use in this function)
  if (ret == MAP_FAILED)
             ^
cryptonight.c:439:3: warning: implicit declaration of function ‘madvise’ [-Wimplicit-function-declaration]
   madvise(ret, sizeof(*ret), MADV_RANDOM|MADV_WILLNEED|MADV_HUGEPAGE);
   ^
cryptonight.c:439:30: error: ‘MADV_RANDOM’ undeclared (first use in this function)
   madvise(ret, sizeof(*ret), MADV_RANDOM|MADV_WILLNEED|MADV_HUGEPAGE);
                              ^
cryptonight.c:439:42: error: ‘MADV_WILLNEED’ undeclared (first use in this function)
   madvise(ret, sizeof(*ret), MADV_RANDOM|MADV_WILLNEED|MADV_HUGEPAGE);
                                          ^
cryptonight.c:439:56: error: ‘MADV_HUGEPAGE’ undeclared (first use in this function)
   madvise(ret, sizeof(*ret), MADV_RANDOM|MADV_WILLNEED|MADV_HUGEPAGE);
                                                        ^
cryptonight.c:441:4: warning: implicit declaration of function ‘mlock’ [-Wimplicit-function-declaration]
    mlock(ret, sizeof(*ret));
    ^
Makefile:9: recipe for target 'all' failed
make: *** [all] Error 1
member
Activity: 81
Merit: 1002
It was only the wind.
September 14, 2016, 09:10:19 PM
#91
full member
Activity: 238
Merit: 100
September 16, 2016, 03:36:11 PM
#91
Oh, Arch...that's a problem  Sad
I can't locate latest AMD driver version 16.9.1 for Arch.
full member
Activity: 238
Merit: 100
September 16, 2016, 03:10:05 PM
#90
member
Activity: 81
Merit: 1002
It was only the wind.
September 14, 2016, 08:12:15 PM
#89
full member
Activity: 238
Merit: 100
September 16, 2016, 12:10:06 PM
#89
legendary
Activity: 3108
Merit: 3199
September 16, 2016, 11:00:33 AM
#88
I give them a try when i at home!!!
newbie
Activity: 50
Merit: 0
September 16, 2016, 09:37:44 AM
#87
There was a bug where the miner would not reconnect to the pool - fixed in git.

Would you mind releasing fixed windows version?
Just in case...
member
Activity: 81
Merit: 1002
It was only the wind.
September 14, 2016, 07:16:30 PM
#86
Can someone who has it working state their driver version? I'll see if I can fix the issue on the problematic ones.
legendary
Activity: 1176
Merit: 1015
September 16, 2016, 06:27:14 AM
#86
I have a faster kernel set, too - my goal here was to equal Claymore, not exceed it.

Oops!

mATX with two pcie slots from 2008:



In my case reason for those low difficulty shares was too much undervolting.

That's pretty low - what clocks?

Well I am happy, it is faster than claymore's. 3.5kH from four hawaii ain't bad...

Too low voltage doesn't crash miner or drivers, you just start getting low difficulty shares and high cpu load.

full member
Activity: 238
Merit: 100
September 16, 2016, 04:15:25 AM
#85
member
Activity: 81
Merit: 1002
It was only the wind.
September 14, 2016, 04:41:35 PM
#84
Tired older drivers 16.20 and still getting the same error that says I should never see it??

If you're going to try different drivers - clean out the old ones first with DDU.
newbie
Activity: 38
Merit: 0
September 16, 2016, 02:04:40 AM
#84

The pool only shows an estimate when you find a share - if you go a while without finding one, the pool will say 0. There was a bug where the miner would not reconnect to the pool - fixed in git.

what am I doing wrong, rejecti go to the pool there is no display





{

 "Algorithms":
[{
      "name": "CryptoNight",
      "devices":[{
            "index": 0,
                                "rawintensity": 916,
            "worksize": 8
                },
           {
            "index": 1,
                                "rawintensity": 924,
            "worksize": 8
                },
           {
            "index": 2,
                                "rawintensity": 924,
            "worksize": 8
                }
          ],

      "pools":[{
            "url": "stratum+tcp://xmr.crypto-pool.fr:3333",
            "user": "43buVZbkWfxhUFRsUPyDQTC",
            "pass": "1"
         }]
}]
}
newbie
Activity: 38
Merit: 0
September 15, 2016, 11:54:19 PM
#83
why miner does not display the speed on the bullet





member
Activity: 81
Merit: 1002
It was only the wind.
September 14, 2016, 03:44:10 PM
#82
Great miner.

With a few MSI Gaming X RX 480 @ 1304 on a research rig (Win10), I get identical speeds to Claymore with some RI/WS tinkering: https://i.imgur.com/3RutBrm.png.

One change request: a quiet mode (analagous to -q) whereby only successful shares and total hashrate is reported would be extremely welcome.

Thanks Smiley


That's exactly what I was going for with this CryptoNight kernel set.
full member
Activity: 504
Merit: 100
Bitgesell (BGL) Decentralized Cryptocurrency!
September 15, 2016, 11:35:24 PM
#82
how much can earn with this gpu miner program ? im interested to this program
newbie
Activity: 5
Merit: 0
September 15, 2016, 11:28:48 PM
#81
I have 6 x Gigabyte G1 RX480 4G overclocked in stock version, that 1290/1750 in core/mem.
Running with 16.9.1 Hotfix.
And Wolf showing ~3470 H/s in totally. So that ~577 H/s per card.
In totally, that at same H/s when compare with Claymore.
But the temperature is seem 30% low than Claymore, so i think the energy cost also lower than.

Problem:
1. At first time, it showing that all of my 6 card working in console screen, but when time past (about 5~6 hours ) there's only showing that 5 card is working (GPU ID 1 is disappearing)
But seem the totally hashrate not changed. I don't know why. Does the machine need to be restart?

2. People always talking about 600~700 H/s in RX480 cards.
I also try many times with wattman but seem its not works.
Why mine so low. If any ideal please let me know!

In wolf, I run with below config that seem product the best performance.

{
   "Algorithms":
   [
      {
         "name": "CryptoNight",
         "devices":
         [
            {
               "index": 0,
               "corefreq": 1290,
          "memfreq": 1750,
               "fanspeed": 60,
               "powertune": 20,
               "threads": 1,
               "rawintensity": 1024,
               "worksize": 8
            },
             ...
         ],
         "pools": ...

https://s9.postimg.io/6g56y75tr/20160916_072124.jpg
legendary
Activity: 1176
Merit: 1015
September 15, 2016, 09:11:15 PM
#80
I have a faster kernel set, too - my goal here was to equal Claymore, not exceed it.

Oops!

mATX with two pcie slots from 2008:



In my case reason for those low difficulty shares was too much undervolting.
newbie
Activity: 50
Merit: 0
September 15, 2016, 08:59:03 AM
#79
As I have stated above, I get around 590 H/s per RX 480 4GB stock card.

However, I have encountered an issue.

This miner refuses to start if an NVIDIA card is used as primary video driver. My current setup is an NVIDIA 1080 on the main 16 slot and then multiple rx 480s on risers from the other slots. Due to work purposes, this cannot be changed.

I have tried multiple changes in the config file.

Any hints?


i like the summary at the top, you use k-miner for Win64 ? can you provide link to download that miner?
or is it something you have to put in the xmr.conf of wolf's miner?

i would like to have more details,
thanks
member
Activity: 81
Merit: 1002
It was only the wind.
September 14, 2016, 12:53:44 PM
#78
low difficulty share, all rejected
I confirm this.

2X ATI Radeon R9 270X with latest 13.9.1 drivers.
My config:
Code:
{
"Algorithms":
[
{
"name": "CryptoNight",
"devices":
[
{
"index": 0,
"corefreq": 1250,
"memfreq": 1250,
"fanspeed": 90,
"powertune": 20,
"threads": 1,
"rawintensity": 511,
"worksize": 8
},
                     {
"index": 1,
"corefreq": 1250,
"memfreq": 1250,
"fanspeed": 90,
"powertune": 20,
"threads": 1,
"rawintensity": 511,
"worksize": 8
}

]
With new Wolf's XMR GPU Miner version 0.4 I only get "Low difficulty shares, rejected"
With same config and drivers on version 0.3 I have no errors!

Oh, I forgot to say - system is Win7 Ultimate x64

Any suggestions?

Try updating drivers - new ones work better with this version.
Oh, excuse me for my mistake - I posted wrong version - I already have latest version 16.9.1 of AMD drivers!

Oooh, I might see the issue - your rawintensity is NOT a multiple of your worksize. Try fixing that - if it persists, then open an issue on GitHub, please.
Pages:
Jump to: