(OLD) BFGMiner: modular FPGA/GPU, GBT, Stratum, RPC, Avalon/Lnx/OpnWrt/PPA/W64

[farfie]

Looks like CGMiner broke other things in 2.5.0, killing performance on BFL Singles and declaring ModMiners sick. If you use either of these FPGAs or encounter other problems (please report them!), feel free to use 2.4.4 until I get them resolved:

• Windows: zip (sig + PGP key)
• Source code: git | tbz2 | zip

I have many singles running at all kinds of different firmwares, and they seem to be operating at the proper hashrate/U they always have, regardless of pool, or miner. Whether it be CG, BFG, 2.4.4, or 2.5.0. Are you certain something is amiss?

[1714176724]

1714176724

[Luke-Jr]

Looks like CGMiner broke other things in 2.5.0, killing performance on BFL Singles and declaring ModMiners sick. If you use either of these FPGAs or encounter other problems (please report them!), feel free to use 2.4.4 until I get them resolved:

• Windows: zip (sig + PGP key)
• Source code: git | tbz2 | zip

I have many singles running at all kinds of different firmwares, and they seem to be operating at the proper hashrate/U they always have, regardless of pool, or miner. Whether it be CG, BFG, 2.4.4, or 2.5.0. Are you certain something is amiss?

Until I figure out exactly what's going on, I can't be certain. So far I've just observed that my hashrate is much worse with 2.5.0 than it was with 2.4.4.

[farfie]

Until I figure out exactly what's going on, I can't be certain. So far I've just observed that my hashrate is much worse with 2.5.0 than it was with 2.4.4.

http://puu.sh/HglI

I will post a comparison image in about 20 hours (as that's about how long it will take to have similar shares submitted) of BFGminer. So you (well, we) can take a look at the differences over a ~24 hour period. Unless you're not interested, in which I won't bother

[bitpop]

howd you go b&w?

[crazyates]

howd you go b&w?

I think it's a Logmein session, and the lowest setting doesn't transmit color.

[farfie]

I think it's a Logmein session, and the lowest setting doesn't transmit color.

Yes it is. It goes gray-scale when it's not the active window on the controlling computer.

[farfie]

And here's with BFG 2.4.4 across both machines, within a reasonably similar amount of operating time in comparison to the screenshot prior.
http://puu.sh/HnVn

[Luke-Jr]

NEW VERSION 2.5.1, JULY 13 2012

Human readable changelog:

• Restore BitFORCE driver performance
• Don't declare devices SICK if they're just initialising (this is mainly for ModMiner during bitstream upload)
• Don't count hardware errors during initial ModMiner clock speed detection
• BitFORCE FPGA autodetection on Windows
• Added a new average hashrate based on accepted shares (Utility, but in hashes-per-second terms)
• Include average wait time and "kernel" name (either "full work" or "nonce range") to RPC API for BitFORCE driver
• Fixed compile issues on BSD build
• Show progress of ModMiner bitstream upload
• Improved hashrate measurement (some hashes weren't counted during longpoll)
• Various fixes to RPC-based miner.php
• Fix RPC API on Big Endian rigs (such as routers)

Full changelog

• Replace CPU Algo in header with runtime
• Bugfix: Calculate diff-1 utility to fix utility-hashrate on pools with diff!=1
• Add utility hashrate to curses display
• Show units in kh, Gh, Th, etc as needed to use at most 3 integer digits
• Use FTD2XX.DLL on Windows to autodetect BitFORCE SHA256 devices
• bitforce_get_result returns -1 on error now.
• Check return value of read in BFgets
• Bugfix: modminer: Count hashes done before work restart
• Bugfix: modminer: Adapt "get nonce" error condition to new scanhash=>-1 error API
• Bugfix: Make our Windows nanosleep/sleep replacements standards-compliant (which fixes nmsleep) and include compat.h for bitforce (for sleep)
• miner.php fix rig # when miners fail
• Fix whitespace mangling.
• bitforce: Use "full work" vs "nonce range" for kernel name
• Abbrv. correction
• Remove superfluous ave_wait
• Put kname change for broken nonce-range back in
• Add average wait time to api stats
• Revert "Merge branch 'ave_time' of https://github.com/pshep/cgminer.git"
• Add average return time to api stats
• Missed one nonce-range disabling.
• Remove bitforce_thread_init The delay thing does nothing useful... when long poll comes around, all threads restart at the same time anyway.
• Change timeouts to time-vals for accuracy.
• More BFL tweaks. Add delay between closing and reopening port. Remove buffer clear in re-init Add kernel type (mini-rig or single)
• Revert "Change BFL driver thread initialising to a constant 100ms delay between devices instead of a random arrangement."
• Only try to shut down work cleanly if we've successfully connected and started mining.
• Fix spelling.
• modminer: Firmware returns 0xffffff00 immediately if we set clockspeed too high
• Bugfix: modminer: Actually count good shares
• Bugfix: Adapt OpenCL scanhash errors to driver API change (errors are now -1, not 0)
• Remove bitforce_thread_init The delay thing does nothing useful... when long poll comes around, all threads restart at the same time anyway.
• fix API support for big endian machines
• Bugfix: Use const struct device_api* for mt_disable
• modminer: Show progress of bitstream upload
• Bugfix: Don't declare devices SICK if they're just busy initializing
• Bugfix: Calculate nsec in nmsleep correctly
• miner.php allow rig names in number buttons
• Change BFL driver thread initialising to a constant 100ms delay between devices instead of a random arrangement.
• Spelling typo.
• Time opencl work from start of queueing a kernel till it's flushed when calculating dynamic intensity.
• Modify te scanhash API to use an int64_t and return -1 on error, allowing zero to be a valid return value.
• Check for work restart after the hashmeter is invoked for we lose the hashes otherwise contributed in the count.
• Remove disabled: label from mining thread function, using a separate mt_disable function.
• Style changes.
• Cope with signals interrupting the nanosleep of nmsleep.
• Use standard cfsetispeed/cfsetospeed to set baud rate on *nix
• miner.php split() flagged deprecated in PHP 5.3.0
• Bugfix: Use nmsleep instead of restart_wait, so we always wait the full time

[Luke-Jr]

FWIW, some statistics after nearly 3½ days (BFGMiner 2.5.1):

Code:

ICA 0:                | 379.2/379.0/370.8Mh/s | A:25995 R:177 HW:  0 U: 5.18/m
BFL 0:  66.3C         | 830.3/830.3/818.7Mh/s | A:57391 R:895 HW:  0 U:11.44/m
MMQ 0:                | 444.0/553.6/521.1Mh/s | A:36533 R:237 HW:135 U: 7.28/m

Note the MMQ 0 only has 3 FPGAs, not 4.

[streblo]

I am having problems with running 1 BFL single with BFGMiner. I auto-tuned in Windows and ran BitMinter successfully, so the hardware is working fine.

Code:

alan@alan-Vostro-1500:~/bc$ git clone git://github.com/luke-jr/bfgminer
Cloning into 'bfgminer'...
remote: Counting objects: 9683, done.
remote: Compressing objects: 100% (3324/3324), done.
remote: Total 9683 (delta 6579), reused 9435 (delta 6348)
Receiving objects: 100% (9683/9683), 5.44 MiB | 102 KiB/s, done.
Resolving deltas: 100% (6579/6579), done.

alan@alan-Vostro-1500:~/bc$ cd bfgminer/

alan@alan-Vostro-1500:~/bc/bfgminer$ ./autogen.sh 
configure.ac:52: installing `./compile'
configure.ac:17: installing `./config.guess'
configure.ac:17: installing `./config.sub'
configure.ac:22: installing `./install-sh'
configure.ac:22: installing `./missing'
ccan/Makefile.am: installing `./depcomp'

alan@alan-Vostro-1500:~/bc/bfgminer$ ./configure --disable-opencl --enable-bitforce
...
------------------------------------------------------------------------
bfgminer 2.5.1
------------------------------------------------------------------------


Configuration Options Summary:

  curses.TUI...........: FOUND: ncurses

  OpenCL...............: Disabled
    ADL................: SDK NOT found, GPU monitoring support DISABLED
  BitForce.FPGAs.......: Enabled
  Icarus.FPGAs.........: Disabled
  ModMiner.FPGAs.......: Disabled
  Ztex.FPGAs...........: Disabled
  libudev.detection....: yes

  CPU Mining...........: Disabled

Compilation............: make (or gmake)
  CPPFLAGS.............: 
  CFLAGS...............: -g -O2
  LDFLAGS..............:  -lpthread 
  LDADD................:  -lcurl   compat/jansson/libjansson.a -lpthread    -lm -ludev 

Installation...........: make install (as root if needed, with 'su' or 'sudo')
  prefix...............: /usr/local

alan@alan-Vostro-1500:~/bc/bfgminer$ make
...

alan@alan-Vostro-1500:~/bc/bfgminer$ ls /dev/ttyUSB*
/dev/ttyUSB0

alan@alan-Vostro-1500:~/bc/bfgminer$ ls /dev/serial/by-id/
usb-Butterfly_Labs_Inc._BitFORCE_SHA256-if00-port0

alan@alan-Vostro-1500:~/bc/bfgminer$ ./bfgminer 
All devices disabled, cannot mine!

alan@alan-Vostro-1500:~/bc/bfgminer$ ./bfgminer -S /dev/ttyUSB0 
All devices disabled, cannot mine!

Plx halp 

[Luke-Jr]

I am having problems with running 1 BFL single with BFGMiner. I auto-tuned in Windows and ran BitMinter successfully, so the hardware is working fine.

Code:

alan@alan-Vostro-1500:~/bc$ git clone git://github.com/luke-jr/bfgminer
Cloning into 'bfgminer'...
remote: Counting objects: 9683, done.
remote: Compressing objects: 100% (3324/3324), done.
remote: Total 9683 (delta 6579), reused 9435 (delta 6348)
Receiving objects: 100% (9683/9683), 5.44 MiB | 102 KiB/s, done.
Resolving deltas: 100% (6579/6579), done.

alan@alan-Vostro-1500:~/bc$ cd bfgminer/

alan@alan-Vostro-1500:~/bc/bfgminer$ ./autogen.sh 
configure.ac:52: installing `./compile'
configure.ac:17: installing `./config.guess'
configure.ac:17: installing `./config.sub'
configure.ac:22: installing `./install-sh'
configure.ac:22: installing `./missing'
ccan/Makefile.am: installing `./depcomp'

alan@alan-Vostro-1500:~/bc/bfgminer$ ./configure --disable-opencl --enable-bitforce
...
------------------------------------------------------------------------
bfgminer 2.5.1
------------------------------------------------------------------------


Configuration Options Summary:

  curses.TUI...........: FOUND: ncurses

  OpenCL...............: Disabled
    ADL................: SDK NOT found, GPU monitoring support DISABLED
  BitForce.FPGAs.......: Enabled
  Icarus.FPGAs.........: Disabled
  ModMiner.FPGAs.......: Disabled
  Ztex.FPGAs...........: Disabled
  libudev.detection....: yes

  CPU Mining...........: Disabled

Compilation............: make (or gmake)
  CPPFLAGS.............: 
  CFLAGS...............: -g -O2
  LDFLAGS..............:  -lpthread 
  LDADD................:  -lcurl   compat/jansson/libjansson.a -lpthread    -lm -ludev 

Installation...........: make install (as root if needed, with 'su' or 'sudo')
  prefix...............: /usr/local

alan@alan-Vostro-1500:~/bc/bfgminer$ make
...

alan@alan-Vostro-1500:~/bc/bfgminer$ ls /dev/ttyUSB*
/dev/ttyUSB0

alan@alan-Vostro-1500:~/bc/bfgminer$ ls /dev/serial/by-id/
usb-Butterfly_Labs_Inc._BitFORCE_SHA256-if00-port0

alan@alan-Vostro-1500:~/bc/bfgminer$ ./bfgminer 
All devices disabled, cannot mine!

alan@alan-Vostro-1500:~/bc/bfgminer$ ./bfgminer -S /dev/ttyUSB0 
All devices disabled, cannot mine!

Plx halp 

Can you try with --debug and post that output?

[streblo]

Can you try with --debug and post that output?

I would, but I just fixed the problem:  running "sudo usermod -a -G dialout USERNAME" fixed it. I don't know what this command does, but ostensibly it gives my user account access to this computers virtual serial ports? Thanks for the great software, pool, and overall support for bitcoin, Luke-Jr.

[Luke-Jr]

running "sudo usermod -a -G dialout USERNAME" fixed it. I don't know what this command does, but ostensibly it gives my user account access to this computers virtual serial ports?

Ah, yes. I should add that one to a README once I figure out what all the various distros group it under...

[Luke-Jr]

Just thought I'd point out Console2 for Windows users who don't like the limitations of Microsoft's command window.

[cdhowie]

After extracting the Windows binaries, AVG detects "PSW.KeyLogger.AVU" in pdcurses.dll.  Since the functionality of pdcurses is to read keystrokes, this doesn't surprise me terribly, but it is a bit disconcerting.  Can this be resolved somehow?

[Luke-Jr]

After extracting the Windows binaries, AVG detects "PSW.KeyLogger.AVU" in pdcurses.dll.  Since the functionality of pdcurses is to read keystrokes, this doesn't surprise me terribly, but it is a bit disconcerting.  Can this be resolved somehow?

Report it to your AVG as a false positive?

[cdhowie]

After extracting the Windows binaries, AVG detects "PSW.KeyLogger.AVU" in pdcurses.dll.  Since the functionality of pdcurses is to read keystrokes, this doesn't surprise me terribly, but it is a bit disconcerting.  Can this be resolved somehow?

Report it to your AVG as a false positive?

I have sent it in already... though in my experience they usually don't do much without contact from someone involved in the production or compilation of the software.  :/

[cdhowie]

I have sent it in already... though in my experience they usually don't do much without contact from someone involved in the production or compilation of the software.  :/

AVG Research Lab has analyzed the file(s) you have sent from your AVG Virus Vault. Below you can find the results for each file. The final verdict on the file is either a correct detection or a false positive detection.

Further information about the verdicts are available at our website:
http://www.avg.com/faq-1184

"c:\Users\Chris Howie\Desktop\bfgminer-2.5.1-win32\pdcurses.dll" - detection is correct

Yeah, that's more or less what I expected.

[rjk]

I have sent it in already... though in my experience they usually don't do much without contact from someone involved in the production or compilation of the software.  :/

AVG Research Lab has analyzed the file(s) you have sent from your AVG Virus Vault. Below you can find the results for each file. The final verdict on the file is either a correct detection or a false positive detection.

Further information about the verdicts are available at our website:
http://www.avg.com/faq-1184

"c:\Users\Chris Howie\Desktop\bfgminer-2.5.1-win32\pdcurses.dll" - detection is correct

Yeah, that's more or less what I expected.

Wonder if you can zip up the source code for just pdcurses.dll and send them in together.

[cdhowie]

Wonder if you can zip up the source code for just pdcurses.dll and send them in together.

Their GUI only seems to allow for detected files to be submitted.  It doesn't detect the original ZIP file that I downloaded as containing anything malicious.

Maybe if I added the DLL with no compression...