OFFICIAL CGMINER mining software thread for linux/win/osx/mips/arm/r-pi 4.11.1

[Endeavour79]

Thanks!

[1714494607]

1714494607

[1714494607]

1714494607

[1714494607]

1714494607

[dostortugas]

Long story short! hallelujah!

The "thread restarting" works finally! Even with only one thread becoming sick the miner restarts properly!

I tested every git update against it to keep ck posted and finally - 5 mins ago I could confirm a restart!
Having a buggy card is at least good for testing!

Code:

13:24 <@chem|st> restarted itself
13:25 <@ckolivas> and working fine?
13:25 <@ckolivas> normal rate and both threads??
13:25 <@chem|st> both threads but one is slower than the other
13:25 <@chem|st> ok balances now
13:25 <@ckolivas> sweet mother of god \o/
13:25 <@chem|st> \o/
13:25 <@ckolivas> halleluyah
13:26 <@chem|st> that was a very long progress to get this shit working fine!
13:26 <@chem|st> good work buddy!
13:26 <@ckolivas> thanks man
13:26  * ckolivas weeps
13:26 <@ckolivas> tears of joy

Please consider some juice for the hard work, sweat and depression ck had with it: 148KkS2vgVi4VzUi4JcKzM2PMaMVPi3nnq

[Ghostofkobra]

Hi,
I noticed that i now get a lot of longpoll errors from slush's pool (for some reason) but BTCGuild seems to be working fine.
However i noticed that it wasnt obvious which pool it was that was giving me the error in a multipool setup so i would appreciate if you incorporated this small change in the next release.

diff old_main.c main.c
3795c3795
<                                       "longpoll failed, sleeping for 30s");
---
>                                       "longpoll failed for %s, sleeping for 30s",lp_url);

It just turns this:
   longpoll failed, sleeping for 30s
into this:
   Longpoll failed for http://api2.bitcoin.cz:8332/LP, sleeping for 30s


Best
/GoK

[ovidiusoft]

Do 'SICK' GPU's ever get restarted? Sometimes I get 'Thread 3 idle for more than 60 seconds, GPU 1 declared SICK!' and mining stops on that GPU. If I restart it manually, work resumes just fine, so I guess it cgminer should try to restart the GPU automagically? I waited 3 minutes and the GPU was not restarted, maybe there is a bigger timeout?

Edit: When it gets marked as DEAD, it does get restarted immediately.

[Ali]

Thanks for the cpu-only client.

[-ck]

Hi,
I noticed that i now get a lot of longpoll errors from slush's pool (for some reason) but BTCGuild seems to be working fine.
However i noticed that it wasnt obvious which pool it was that was giving me the error in a multipool setup so i would appreciate if you incorporated this small change in the next release.

diff old_main.c main.c
3795c3795
<                                       "longpoll failed, sleeping for 30s");
---
>                                       "longpoll failed for %s, sleeping for 30s",lp_url);

It just turns this:
   longpoll failed, sleeping for 30s
into this:
   Longpoll failed for http://api2.bitcoin.cz:8332/LP, sleeping for 30s


Best
/GoK

Good idea. Though I might just put the pool number rather than the whole URL because the URL can be quite long leading to a very long log message. ?

[-ck]

Do 'SICK' GPU's ever get restarted? Sometimes I get 'Thread 3 idle for more than 60 seconds, GPU 1 declared SICK!' and mining stops on that GPU. If I restart it manually, work resumes just fine, so I guess it cgminer should try to restart the GPU automagically? I waited 3 minutes and the GPU was not restarted, maybe there is a bigger timeout?

Edit: When it gets marked as DEAD, it does get restarted immediately.

Yes indeed, that's what I've spent weeks working on to get right. To restart it if it can only. The 5 minutes was a nominal choice based on the theory that if it gets sick, giving it a few minutes to cool down will make it more likely to respond. Note that if it gets sick a LOT, you may benefit from decreasing the intensity and sacrificing a few MH just to get a more stable throughput (eg -I 7 instead of .

[tomek74]

./configure works fine, but
make for cgminer 1.5.5 fails now with the following error on Mac mini Lion 10.7:

Code:

ld: warning: ignoring file x86_64/libx8664.a, file was built for archive which is not the architecture being linked (x86_64)
Undefined symbols for architecture x86_64:
  "_CalcSha256_x64", referenced from:
      _scanhash_sse2_64 in cgminer-sha256_sse2_amd64.o
  "_CalcSha256_x64_sse4", referenced from:
      _scanhash_sse4_64 in cgminer-sha256_sse4_amd64.o
ld: symbol(s) not found for architecture x86_64
collect2: ld returned 1 exit status
make[2]: *** [cgminer] Error 1

[tomek74]

./configure works fine, but
make for cgminer 1.5.5 fails now with the following error on Mac mini Lion 10.7:

Code:

ld: warning: ignoring file x86_64/libx8664.a, file was built for archive which is not the architecture being linked (x86_64)
Undefined symbols for architecture x86_64:
  "_CalcSha256_x64", referenced from:
      _scanhash_sse2_64 in cgminer-sha256_sse2_amd64.o
  "_CalcSha256_x64_sse4", referenced from:
      _scanhash_sse4_64 in cgminer-sha256_sse4_amd64.o
ld: symbol(s) not found for architecture x86_64
collect2: ld returned 1 exit status
make[2]: *** [cgminer] Error 1

After uninstalling yasm (with sudo port uninstall yasm) it compiles and runs fine.

[tomek74]

Oh, it is not really working for me. I am still having difficulties with cgminer on Mac mini. It shows 4124 Megahash/s which is not realistic on my Mac mini. (DiableMiner shows me 48350 khash/sec)
During 17 minutest there were still 0 accepted shares.

Code:

localhost:cgminer thomas$ ./cgminer -o http://api2.bitcoin.cz:8332 -u name.name-p password -I 8
[2011-08-17 01:47:41] 
Summary of runtime statistics:
                    
[2011-08-17 01:47:41] Started at [2011-08-17 01:30:37]                    
[2011-08-17 01:47:41] Runtime: 0 hrs : 17 mins : 3 secs                    
[2011-08-17 01:47:41] Average hashrate: 4124.6 Megahash/s                    
[2011-08-17 01:47:41] Queued work requests: 11733                    
[2011-08-17 01:47:41] Share submissions: 0                    
[2011-08-17 01:47:41] Accepted shares: 0                    
[2011-08-17 01:47:41] Rejected shares: 0                    
[2011-08-17 01:47:41] Hardware errors: 0                    
[2011-08-17 01:47:41] Efficiency (accepted / queued): 0%                    
[2011-08-17 01:47:41] Utility (accepted shares / min): 0.00/min
                    
[2011-08-17 01:47:41] Discarded work due to new blocks: 0                    
[2011-08-17 01:47:41] Stale submissions discarded due to new blocks: 0                    
[2011-08-17 01:47:41] Unable to get work from server occasions: 0                    
[2011-08-17 01:47:41] Work items generated locally: 0                    
[2011-08-17 01:47:41] Submitting work remotely delay occasions: 0                    
[2011-08-17 01:47:41] New blocks detected on network: 19
                    
[2011-08-17 01:47:41] Summary of per device statistics:
                    
[2011-08-17 01:47:41]  GPU 0: [4394.3 / 4124.6 Mh/s] [Q:11732  A:0  R:0  HW:0  E:0%  U:0.00/m]    

[-ck]

Oh, it is not really working for me. I am still having difficulties with cgminer on Mac mini. It shows 4124 Megahash/s which is not realistic on my Mac mini. (DiableMiner shows me 48350 khash/sec)
During 17 minutest there were still 0 accepted shares.

Code:

localhost:cgminer thomas$ ./cgminer -o http://api2.bitcoin.cz:8332 -u name.name-p password -I 8
[2011-08-17 01:47:41] 
Summary of runtime statistics:
                    
[2011-08-17 01:47:41] Started at [2011-08-17 01:30:37]                    
[2011-08-17 01:47:41] Runtime: 0 hrs : 17 mins : 3 secs                    
[2011-08-17 01:47:41] Average hashrate: 4124.6 Megahash/s                    
[2011-08-17 01:47:41] Queued work requests: 11733                    
[2011-08-17 01:47:41] Share submissions: 0                    
[2011-08-17 01:47:41] Accepted shares: 0                    
[2011-08-17 01:47:41] Rejected shares: 0                    
[2011-08-17 01:47:41] Hardware errors: 0                    
[2011-08-17 01:47:41] Efficiency (accepted / queued): 0%                    
[2011-08-17 01:47:41] Utility (accepted shares / min): 0.00/min
                    
[2011-08-17 01:47:41] Discarded work due to new blocks: 0                    
[2011-08-17 01:47:41] Stale submissions discarded due to new blocks: 0                    
[2011-08-17 01:47:41] Unable to get work from server occasions: 0                    
[2011-08-17 01:47:41] Work items generated locally: 0                    
[2011-08-17 01:47:41] Submitting work remotely delay occasions: 0                    
[2011-08-17 01:47:41] New blocks detected on network: 19
                    
[2011-08-17 01:47:41] Summary of per device statistics:
                    
[2011-08-17 01:47:41]  GPU 0: [4394.3 / 4124.6 Mh/s] [Q:11732  A:0  R:0  HW:0  E:0%  U:0.00/m]    

19 new blocks in 17 minutes? That doesn't even sound like bitcoin...

[Tx2000]

Apologize if it was already addressed or noted somewhere but I cannot seem to find it.  Is it possible to set Intensity per GPU?  If not, would that be something difficult to add?  I like being able to manage each GPU individually and in my case, one of my GPUs isn't always a dedicated miner while the other is.

[-ck]

Apologize if it was already addressed or noted somewhere but I cannot seem to find it.  Is it possible to set Intensity per GPU?  If not, would that be something difficult to add?  I like being able to manage each GPU individually and in my case, one of my GPUs isn't always a dedicated miner while the other is.

Unfortunately there is not a way to do that at the moment. In the readme you'll see it recommends running 2 instances of cgminer selecting which GPUs to use. Use one instance in dynamic mode and the other with a set intensity.

[ovidiusoft]

I might have detected a memory leak in 1.5.5. After ~8 hours of work, cgminer got oom-killed. I restarted the process and I'm taking ps snapshots every 10 minutes to see if I'm right. I'll be back .

[ovidiusoft]

Yes indeed, that's what I've spent weeks working on to get right. To restart it if it can only. The 5 minutes was a nominal choice based on the theory that if it gets sick, giving it a few minutes to cool down will make it more likely to respond. Note that if it gets sick a LOT, you may benefit from decreasing the intensity and sacrificing a few MH just to get a more stable throughput (eg -I 7 instead of .

Only happened twice in 4-5 hours, so I think I'm ok. It's 5 minutes then... I need patience... or a command line switch?

Edit: confirmed, it restarts after 5 minutes. An option would be nice - my GPUs cool down way faster than that, and I think that's the case with most miners, most of them have extra fans and cooling.

[-ck]

New phatk and poclbm kernels in git tree for those who've been waiting for them.

Hopefully new release just around the corner with that

[-ck]

Too good to pass up a new release even with only 2 real changes.

Version 1.5.6 (you all know where by now - links in first post).

Executive summary: New faster phatk kernel (2.2) courtesy of phateus porting it (thanks!) and with my own custom modifications on top. Fixed multiple pool longpoll management.

[d3m0n1q_733rz]

Too good to pass up a new release even with only 2 real changes.

Version 1.5.6 (you all know where by now - links in first post).

Executive summary: New faster phatk kernel (2.2) courtesy of phateus porting it (thanks!) and with my own custom modifications on top. Fixed multiple pool longpoll management.

From all of us... *Hugs*   

[d3m0n1q_733rz]

I've noticed something strange about the Windows version of cgminer.  With only GPU mining enabled, there are 6 instances of the program running; 5 of which are using 1,216K of memory and 0% CPU.  Is there a purpose for these threads or is your miner having accidental child processes?

[Ghostofkobra]

Hi,
diff old_main.c main.c
3795c3795
<                                       "longpoll failed, sleeping for 30s");
---
>                                       "longpoll failed for %s, sleeping for 30s",lp_url);
Best
/GoK

Good idea. Though I might just put the pool number rather than the whole URL because the URL can be quite long leading to a very long log message. ?

That works just fine,

Thanx
/GoK