Topic: OFFICIAL CGMINER mining software thread for linux/win/osx/mips/arm/r-pi 4.11.0 - page 516. (Read 5806088 times)

New release - 2.7.0, August 18th 2012

Normally a minor version (2.6->2.7) update brings with it instability, but in fact this release includes some very heavily tested code that was a long time in the making and because of its magnitude and impact it warranted a version update.


Human readable changelog:

The main change in this version is a complete rewrite of the getwork requesting mechanism. I've been slowly hacking away at it for some time, but finally gave up in disgust and have rewritten it almost entirely. Previously mining threads would occasionally throw out a request for more work, some arbitrary test would be done on whether more work should be requested, and it handed off the message to another thread which spawned another thread and that then sent the request and ... anyway the mechanism was so asynchronous that the arse didn't know what its head was doing by the time it was deciding on what to do for work. Worse yet it was hard to find the right place to reuse work and so it was never reused to its utmost potential. This is mostly my fault for gradually hacking on more and more asynchronous threaded components to cgminer and the demands for getting work have been increasing sharply of late with new hardware. The rewrite involves scheduling a new request based on the rate the old work items get used up, and is much better at predicting when it needs to leak work to backup pools and less likely to throw a "pool is not providing work fast enough" message. Overall you should now see much more Local Work (LW), the efficiency will be higher on pools that support rolltime, less work will be discarded, any magnitude rig will be kept solidly busy - note this MAY mean your overclocks will become that much more stressed if you have set clocks very aggressively. Thanks to numerous people who tested this on IRC during its development phase. For many of you, you'll be wondering what the fuss is about cause it will just appear as business as usual.

New pool strategy: Balance.
--balance           Change multipool strategy from failover to even share balance
This is to differentiate itself from the existing pool strategy, Load balance:
--load-balance      Change multipool strategy from failover to efficiency based balance

With the change to queueing and more roll work being possible than ever before, the imbalance between pools that support rolltime and those that don't will now be extreme in load balance strategy. To offset that, and since the number of people using load balance has been increasing, the new strategy was added to try and give roughly the same number of shares to each pool. This required some code to estimate a rolling average work completion rate that was not dependent on difficulty, and would cope with dips and peaks as pools are enabled/disabled/fail. Otherwise you could end up mining 100% on solo since you would rarely be submitting only possible block solutions, and not shares.

New statistic: Work Utility. With higher difficulty share supporting pools in the testing phase, it is going to be hard to monitor overall work performance based on successful share submission. To counter this, work utility is a value based on the amount of difficulty 1 shares solved, whether they're accepted or rejected by the pool. This value will always be higher than the current "utility". (Note that difficulty 1 share counting on scrypt is not supported since the work is compared to the target on the GPU itself, but the total shares solved will be displayed).

Other minor bugfixes.


Full changelog:

- Introduce a new statistic, Work Utility, which is the number of difficulty 1
shares solved per minute. This is useful for measuring a relative rate of work
that is independent of reject rate and target difficulty.
- Implement a new pool strategy, BALANCE, which monitors work performed per pool
as a rolling average every 10 minutes to try and distribute work evenly over all
the pools. Do this by monitoring diff1 solutions to allow different difficulty
target pools to be treated equally, along with solo mining. Update the
documentation to describe this strategy and more accurately describe the
load-balance one.
- Getwork fail was not being detected. Remove a vast amount of unused variables
and functions used in the old queue request mechanism and redefine the getfail
testing.
- Don't try to start devices that don't support scrypt when scrypt mining.
- 0 is a valid return value for read so only break out if read returns -1.
- Consider us lagging only once our queue is almost full and no staged work.
- Simplify the enough work algorithm dramatically.
- Only queue from backup pools once we have nothing staged.
- Don't keep queueing work indefinitely if we're in opt failover mode.
- Make sure we don't opt out of queueing more work if all the queued work is
from one pool.
- Set lagging flag if we're on the last of our staged items.
- Reinstate clone on grabbing work.
- Grab clones from hashlist wherever possible first.
- Cull all the early queue requests since we request every time work is popped
now.
- Keep track of staged rollable work item counts to speed up clone_available.
- Make expiry on should_roll to 2/3 time instead of share duration since some
hardware will have very fast share times.
- Do the cheaper comparison first.
- Check that we'll get 1 shares' worth of work time by rolling before saying we
should roll the work.
- Simplify all those total_secs usages by initialising it to 1 second.
- Overlap queued decrementing with staged incrementing.
- Artificially set the pool lagging flag on pool switch in failover only mode as
well.
- Artificially set the pool lagging flag on work restart to avoid messages about
slow pools after every longpoll.
- Factor in opt_queue value into enough work queued or staged.
- Roll work whenever we can on getwork.
- Queue requests for getwork regardless and test whether we should send for a
getwork from the getwork thread itself.
- Get rid of age_work().
- 0 is a valid return value for read so only break out if read returns -1.
- Offset libusb reads/writes by length written as well in ztex.
- Cope with timeouts and partial reads in ztex code.
- fpga serial I/O extra debug (disabled by default)

2.6.4 is woking good with 15 BFL Singles a week!

Does anyone have experience with CGminer and configuration on Dropbox without dropbox windowsclient?

Yes, and it is going to get MUCH worse on the next release as I clean up the queueing to make it as efficient as possible.

I have been able to get a better balance between 2 pools by using the rotate option. Get alot closer to having even shares on each pool. With the load balance at the times I had the biggest differences between pools, I would sometimes get as much as 2/3 at one pool and 1/3 at the other. Does that sound like a correct ratio that the rollntime could cause?

This all being to attempt to lower variance. At wat cost is my trying to split work evenly between 2 pools effecting efficiency versus just using the failover feature?

I've no idea what creates that log ...

Well... I started having this issue: (Not bitching, just providing feedback)


I have two Pi, both configured identical except for hostname and IP. Both have two BFL Singles attached directly. I switched back to 2.5.0 on one of them and it has not crashed since, while on the other running 2.6.5 it has after 6-7 hours with the Comms error as shown above. I have also swapped which Pi was running 2.6.5 and this issue seems to follow it. (after 6+ hours) I have also swapped the BFL's around, it appears to be random as to which fails. Open for suggestions or let me know if there is a way I can help with configuring a log file or something that.

This happens after switching user:

I doesnt actually crash and exit, the interface just stops responding. Also it stops submitting shares, I use the 50btc.com pool and when the miner hangs the drop in mhash/s is seen quickly in the pools API. And GPU-Z shows the gpus continue running the kernels.

still no luck, get the APPCRASH again

Make your msdos window bigger by default.

For some reason when I try to run cgminer with 2 GPUs and 4 BFLs it crashes. I have to run 2 separate instances, one with -S noauto for GPUs and other with all the BFLs. I can run 2 GPUs and 2 BFLs on same cgminer instance, but not all 6 at once. It happens on all versions of cgminer. Any idea why this is?

Not sure if this helps at all, but here is the windows log:
Problem signature:
  Problem Event Name:   APPCRASH
  Application Name:   cgminer.exe
  Application Version:   0.0.0.0
  Application Timestamp:   502b065b
  Fault Module Name:   cgminer.exe
  Fault Module Version:   0.0.0.0
  Fault Module Timestamp:   502b065b
  Exception Code:   c0000005
  Exception Offset:   0000a74d
  OS Version:   6.1.7601.2.1.0.256.4
  Locale ID:   1033
  Additional Information 1:   0a9e
  Additional Information 2:   0a9e372d3b4ad19135b953a78882e789
  Additional Information 3:   0a9e
  Additional Information 4:   0a9e372d3b4ad19135b953a78882e789

Not all GPUs support fanspeed monitoring by RPM; only the set percentage. You can see that from status line at the top. cgminer shouldn't be trying to show you the rpm below as it's completely invalid.

Quick question: Can someone tell me why the fan output on my 2nd card looks all wonky? Top card is a Gigabyte 7970, and bottom card is a Gigabyte 7770.

I have found a bug that is reproducible. It seems cgminer always declares 1 ztex board dead after a while (1h or so). Interestingly enough this does not happen on my other rigs that are x86 based, and have also GPU's connected to them. Very weird indeed. Reproducible also in BFGMiner.

Also, if 1 board is declared dead and is kept connected to the computer, no more boards will be decalred dead. If I disconnect the dead board, soon another one will get sick and dead.

System setup:
CGMiner 2.6.4 running on Raspberry Pi
14-18 ztex 1.15x

Have only one GPU 6770, same thing there.
On LTC mining I have I=11, I change it to dynamic, it jumps to 9 for a second or two then falls to -10 and stays there.

CGMINER crashes when using Switch User on Windows. This is the only miner that does this, all other miners continue happily mining in the background.
When the crash happens, the mining kernel continues to run on the gpus but no shares are submitted, and the cgminer interface just hangs and doesnt respond.

I tried all versions back to 2.0.4, they all crash when switching user.

My machine:
Windows 7 x64
Catalyst 12.6, APP 2.7
Radeon 7950 and 5850

I'm not worried about it, especially with the explanation's of the probable cause.
Thanks,
Sam

Yeah, we had this discussion before, I'll be damned if I can find it again though.  Worst search function ever.  The per device settings should be okay, I just moved my config file to the newer version and didn't touch anything else.  Let me try it with a clean config, be right back.

Edit: Seems to be working as expected with no config file.  Now I have to find out what is borked up in my old config file.  
       At least I didn't go "OMG, I'm geeting less Mh/s with new version, you broke something!!!11!!"  

Edit#2:  Okay, I spoke to soon, after running a few minutes, the display GPU pegs down to -10 still.  It doesn't happen right after starting.

Edit#3:  Okay, further "testing" shows that it it happening when my display GPU is set to dynamic, and the other GPU is set to a fixed intensity.  Also seeing a massive jump in CPU usage when it happens, using 50% CPU, one whole core when it goes to -10 I.  Going to let it run for a bit with both GPUs set to dynamic and see if it happens.

Edit#4:  After running with both GPUs set to dynamic, now the GPU with no monitor connected has pegged to -10 I.  Going to have to use the older version I have for the time being.  :/