Topic: OFFICIAL CGMINER mining software thread for linux/win/osx/mips/arm/r-pi 4.11.0 - page 166. (Read 5805670 times)

i have API issues with cgwatcher too. I think the least version [or even 2 last ones] are crushed with API ussage.

No. Why do you ask?

I'm trying to use JSON API with the following command sent over RPC:
{"command":"pools"}

I'm getting .."status":"E","When":11/28/2013 07:18:45","Code":14","Msg":"Invalid command","Description":"cgminer 3.8.3"...

API works however with regular text request/response and the same call is working fine for cgminer 3.8.1

Did something changed? or is it a bug?

You donated?

Yes, that's why I wrote it -as pure mercenary work- and maintained it for a year and gracefully offered to hand it off, even handholding various people on how to fork the existing code. But it's just a burden on the current code and has nothing in common with bitcoin mining on asics and now earns me nothing but headaches. Plus I don't believe in altcoins of any sort so it's hypocritical of me to maintain it.

Wasn't there a rather big donation to get scrypt support in cgminer in the first place?

Maybe you want to reconsider that given the latest boom in price, although i don't know the reasons that you stopped in the first place.

Yep I'll change it in the next few days

We're not as scary as we may appear. If you push code via github to cgminer that doesn't affect other drivers, we will probably give you some basic changes we'd prefer before merging the code with cgminer, but are happy to include code with the master cgminer code if you do it that way and are happy to support it. Ideally we prefer to get the hardware ourselves so we can extend/develop/bugfix/support the code for the hardware ourselves until the hardware is no longer relevant, but if not, I'm happy to merge code from you while you personally support the code unless it interferes with other drivers or the code stops being supported as bugs appear (which they always do).

MrTeal and I have developed mining hardware called Chili using BFL chips and it uses the BLF driver.  The driver interface / protocol defines a command (ZTX) that allows cgminer to request three voltages from the hardware.  The 3 values are typically  3.3V, 1.0V, and 12V.  cgminer displays the 3.3V value on the screen but in my opinion this is the least interesting voltage value.  I would prefer it to show the (nominal) 1.0V value as this directly impacts hashing speed and power.  On other BFL hardware, this value is fixed at the factory, but on a Chili, it is configurable by software allowing automatic overclocking.  Thus the interest in displaying this value.

Please consider this reply a formal request to change the displayed value from the 3.3V value to the 1.0V value.  If you have enough screen real estate, it should show 3 digits to the right of the decimal point.  Ex: 1.057V  Thank you.

BTW - We have added a new field to the ZCX command to distinguish our hardware apart from BFL hardware:

Thanks for that. No you were getting the timer overruns previously as well. I saw them on your successful logging when you put it up so even when it's "fine" it's not entirely fine, it's just that it sort of recovers.

Next!
http://ck.kolivas.org/apps/cgminer/temp/cgminer-w00t.exe

EDIT: And one greater than 9000:
http://ck.kolivas.org/apps/cgminer/temp/cgminer-9001.exe

OK, tested!  Aside [again] The previous run of 3.5.1 lasted 25 hours without problems, although I do remember seeing a few AMU LEDs come on yesterday for a couple of seconds and I noticed some messages which were something like the following (completely from memory, as no logging enabled, can't remember the exact mS timings):
  [2013-11-xx xx:xx:xx] TIMEOUT AMUx something took xxx mS, was xxx mS
  [2013-11-xx xx:xx:xx] TIMEOUT AMUy something took xxx mS, was xxx mS
The above disturbances didn't cause any problems, so it may just be normal network problems, but I haven't noticed them before, so perhaps it's important, perhaps not.

Anyway, the test of cgminer-ffs ran for almost 2 hours, then a zombie appeared.  I turned on debug from the display and let run for a further 5 minutes, then stopped it.

Logfile is here:
  https://dl.dropboxusercontent.com/u/44240170/logfile-ffs.txt

And.... back to 3.5.1 for the evening!

yeah, i searched for it after my post

are vga still supported?

and why i can't copy past anymore with cgminer?

Thanks very much for that. Believe it or not I am (very slowly) making progress.

Here's the next test please:
http://ck.kolivas.org/apps/cgminer/temp/cgminer-ffs.exe

DA is the value in 1diff of all shares accepted.
If you submit 10 50diff shares and all are accepted then DA would be 500

I've no idea what you mean those % numbers are.

Have I asked in the wrong thread? :-)
Please advise if so...

Interesting thought. It's easy enough to raise the number of file descriptors allowed in unix, but the default on windows should be much higher. Either way we should be removing the semaphores when the devices get zombied so that's at least one fix worth adding, thanks.

EDIT: Checked the code and the semaphores are being removed so that's not it.

This from 3.8.3 may give a clue:
 [2013-11-26 11:54:02] SEM: Icarus USB open failed '/tmp/cgminer-usb-1-58' err (24) Too many open files

It revives the zombies until it runs out of file descriptors on unix.  Maybe some similar resource exhaustion is happening on Windows?

Summary of runtime statistics:

 [2013-11-26 11:54:14] Started at [2013-11-23 11:22:25]
 [2013-11-26 11:54:14] Pool: http://api.polmine.pl:8347
 [2013-11-26 11:54:14] Runtime: 72 hrs : 31 mins : 49 secs
 [2013-11-26 11:54:14] Average hashrate: 1298.2 Megahash/s
 [2013-11-26 11:54:14] Solved blocks: 0
 [2013-11-26 11:54:14] Best share difficulty: 204K
 [2013-11-26 11:54:14] Share submissions: 47169
 [2013-11-26 11:54:14] Accepted shares: 46323
 [2013-11-26 11:54:14] Rejected shares: 846
 [2013-11-26 11:54:14] Accepted difficulty shares: 76307
 [2013-11-26 11:54:14] Rejected difficulty shares: 919
 [2013-11-26 11:54:14] Reject ratio: 1.8%
 [2013-11-26 11:54:14] Hardware errors: 837
 [2013-11-26 11:54:14] Utility (accepted shares / min): 10.64/min               
 [2013-11-26 11:54:14] Work Utility (diff1 shares solved / min): 13.76/min

 [2013-11-26 11:54:14] Stale submissions discarded due to new blocks: 0         
 [2013-11-26 11:54:14] Unable to get work from server occasions: 0               
 [2013-11-26 11:54:14] Work items generated locally: 130566
 [2013-11-26 11:54:14] Submitting work remotely delay occasions: 0               
 [2013-11-26 11:54:14] New blocks detected on network: 508