OLD: BFGMiner 3.10.0: modular ASIC+FPGA, GBT+Strtm, RPC, Mac/Lnx/W64, AntU1, DRB

[Zich]

My only guess is you built it on a vfat filesystem (why???)

 , I am new to linux. I just get my pi yesterday. I use win32diskimager to write wheezy image to sd card without formating the disk.

[1715228866]

1715228866

[1715228866]

1715228866

[Luke-Jr]

Have you considered including your favorite build of hidapi in you git/configure?

No, I consider it a bug to embed libraries in software.
The only reason I still do it with libblkmaker is that no other software is using it yet.

I ask because Arch does not have a package for it yet and I prefer you build than one I can come up with (not that my builds are bad and don't work, just because you know the ins and outs of it better than I).

I have never used Arch, let alone made a package for it.
Perhaps whoever is maintaining the BFGMiner packages would be willing to do this.

Fair enough, I would like to contend that hidapi  is in the same boat (as you described) as libblkmaker though.  I will defend your choice to include those libraries because they make a better bfgminer.  Why one and not the other?

Perhaps not many, but there are other software using hidapi.

[Zich]

run as root sudo bfgminer.

Gonna have some problems with your Nano Fury's on wheezy and PI.

If you don't ... let me know how you got them working.

Ok, now can run after reinstall all.
And the result is same with yours.

Code:

 [2013-11-17 13:19:32] Loaded configuration file /root/.bfgminer/bfgminer.conf
 [2013-11-17 13:19:33] Probing for an alive pool
 [2013-11-17 13:19:33] Pool 0 stratum+tcp://stratum.btcguild.com:3333 alive
 [2013-11-17 13:19:34] Network difficulty changed to 511M ( 3.66Ph/s)
 [2013-11-17 13:19:34] mcp2210_get_gpio_input: Failed to get current GPIO input values
 [2013-11-17 13:19:34] NFY 0: checkport failed
 [2013-11-17 13:19:34] NFY 0 failure

Any luck with older version?
Ok, my fault. I plug it directly to Pi. Waiting for my Dlink 

What should i do if i want to add different version? Let the 3.6.0 folder intact and add new folder?

[RicRock]

Ok, my fault. I plug it directly to Pi. Waiting for my Dlink 

What should i do if i want to add different version? Let the 3.6.0 folder intact and add new folder?

Def needs a powered hub.

No luck on older versions for me.

I'm not good with git, not sure how you can move up and down in versions easily while maintaining your current version.

[Zich]

Def needs a powered hub.

No luck on older versions for me.

I'm not good with git, not sure how you can move up and down in versions easily while maintaining your current version.

Actually i am still in 3.6.0. Not sure how to add other version 

[Digi128pci]

I got past that but i use USB powered hub ...

But it fails further ...

Code:

[M]anage devices [P]ool management [S]ettings [D]isplay options  [H]elp [Q]uit
 Connected to eu-stratum-lb489kj.btcguild.com diff 2 with stratum as user digi128pci_avalon
 Block: ...7ee030e7 #269461  Diff:511M ( 3.66Ph/s)  Started: [22:51:21]
 ST:2  F:0  NB:1  AS:0  BW:[163/ 31 B/s]  E:0.87  I:14.61uBTC/hr  BS:4
 0            | 202.4/263.0/178.1Mh/s | A:1 R:0+3( 75%) HW:8/ 33%
--------------------------------------------------------------------------------
 NFY 0:       | OFF  /272.7/184.7Mh/s | A:1 R:0+3( 75%) HW:8/ 33%
--------------------------------------------------------------------------------
 [2013-11-13 22:51:23] NFY 0: Previous nonce mismatch (4th try), recalibrating
 [2013-11-13 22:51:25] Accepted 3fa4ee98 NFY 0  Diff 4/2
 [2013-11-13 22:51:25] Stratum from pool 0 requested work update
 [2013-11-13 22:51:25] Pool 0 is hiding block contents from us
 [2013-11-13 22:51:25] Reconnect requested from pool 0 to eu-stratum-lb489kj.btcguild.com:3333
 [2013-11-13 22:51:26] Stratum from pool 0 requested work update
 [2013-11-13 22:51:32] mcp2210_spi_transfer: Failed to issue SPI transfer
 [2013-11-13 22:51:34] mcp2210_spi_transfer: Failed to continue SPI transfer (59 bytes remaining)
 [2013-11-13 22:51:34] NFY 0 failure, attempting to reinitialize
 [2013-11-13 22:51:34] mcp2210_set_cfg_spi: Error setting current SPI config (248)
 [2013-11-13 22:51:34] mcp2210_set_cfg_spi: Error setting current SPI config (248)
 [2013-11-13 22:51:34] NFY 0 failure, disabling!
 [2013-11-13 22:51:35] NFY 0: bitfury_init_oldbuf: Giving up after 4 tries
 [2013-11-13 22:51:35] NFY 0: bitfury_init_oldbuf: Giving up after 4 tries
 [2013-11-13 22:51:36] NFY 0: bitfury_init_oldbuf: Giving up after 4 tries
 [2013-11-13 22:51:37] NFY 0: bitfury_init_oldbuf: Giving up after 4 tries

[djaychela]

I've been running 3.4.0 on SuSE linux (compiled from the git source) for a few weeks now, initially using Block Erupters, but last week I got a Red Fury, which works fine, aside from having LOTs of hardware errors (59% or so).

Just compiled 3.6.0 from source, get a segmentation fault when I try to run it - any ideas, given that 3.4.0 compiles and runs without issue?

Also, anyone got a link to a clear howto on getting a previous version from git?  I've done a fair bit of looking and found nothing that works, or that is helpful.  Downloading the zip/tar.gz of the source of 3.5.2 doesn't work as it won't compile from the files that github gives you; I've only had luck by cloning the repo which gives you only the current version, which obviously isn't any good to me...  I know there are good reasons for using git but it doesn't seem at all friendly to anyone who wants to download and compile a specific version.

[jedimstr]

I've been running 3.4.0 on SuSE linux (compiled from the git source) for a few weeks now, initially using Block Erupters, but last week I got a Red Fury, which works fine, aside from having LOTs of hardware errors (59% or so).

Just compiled 3.6.0 from source, get a segmentation fault when I try to run it - any ideas, given that 3.4.0 compiles and runs without issue?

Also, anyone got a link to a clear howto on getting a previous version from git?  I've done a fair bit of looking and found nothing that works, or that is helpful.  Downloading the zip/tar.gz of the source of 3.5.2 doesn't work as it won't compile from the files that github gives you; I've only had luck by cloning the repo which gives you only the current version, which obviously isn't any good to me...  I know there are good reasons for using git but it doesn't seem at all friendly to anyone who wants to download and compile a specific version.

Much easier than you think.


git checkout [branchname]

https://www.atlassian.com/git/tutorial/git-branches#!checkout

[curtiskam]

I've been using bfgminer for quite a while now and I love it.  One thing I've never been able to find is a current interface guide.  I understand most of the 3.6 interface, but could someone break down this line for me?

ST:28 F:2 NB:98 AS:0 BW:[137/90 B/s] E:26.81 I: 441uBTC/hr BS:65k

Thanks in advance: Curtis

[xyzzy099]

I've been using bfgminer for quite a while now and I love it.  One thing I've never been able to find is a current interface guide.  I understand most of the 3.6 interface, but could someone break down this line for me?

ST:28 F:2 NB:98 AS:0 BW:[137/90 B/s] E:26.81 I: 441uBTC/hr BS:65k

Thanks in advance: Curtis

From the BFGMiner README:

The BFGMiner status line shows:
 ST:1  F:0  NB:1  AS:0  BW:[ 75/241 B/s]  E:2.42  I:12.99mBTC/hr  BS:2.71k

ST is STaged work items (ready to use).
F  is network Failure occasions (server down or slow to provide work)
NB is New Blocks detected on the network
AS is Active Submissions (shares in the process of submitting)
BW is BandWidth usage on the network (received/sent)
E  is Efficiency defined as number of shares accepted (multiplied by their
          difficulty) per 2 KB of bandwidth
I  is expected Income, calculated by actual shares submitted in 100% PPS value
          (assumes Bitcoin, does not account for altcoin conversions!)
BS is the all time Best Share difficulty you've found

[djaychela]

I've been running 3.4.0 on SuSE linux (compiled from the git source) for a few weeks now, initially using Block Erupters, but last week I got a Red Fury, which works fine, aside from having LOTs of hardware errors (59% or so).

Just compiled 3.6.0 from source, get a segmentation fault when I try to run it - any ideas, given that 3.4.0 compiles and runs without issue?

Also, anyone got a link to a clear howto on getting a previous version from git?  I've done a fair bit of looking and found nothing that works, or that is helpful.  Downloading the zip/tar.gz of the source of 3.5.2 doesn't work as it won't compile from the files that github gives you; I've only had luck by cloning the repo which gives you only the current version, which obviously isn't any good to me...  I know there are good reasons for using git but it doesn't seem at all friendly to anyone who wants to download and compile a specific version.

Much easier than you think.


git checkout [branchname]

https://www.atlassian.com/git/tutorial/git-branches#!checkout

I'm clearly an idiot, as I've got nowhere with that.  I can't see how you can specify a previous version from that; running checkout only gives me M libblkmaker.  As I say, I can see how git is clearly useful for someone who is well versed in it and for version management on a project you are contributing to, but it's as clear as mud how I can get the source for 3.5.2 from it; to me, anyway, so I must be stupid.

[jedimstr]

I've been running 3.4.0 on SuSE linux (compiled from the git source) for a few weeks now, initially using Block Erupters, but last week I got a Red Fury, which works fine, aside from having LOTs of hardware errors (59% or so).

Just compiled 3.6.0 from source, get a segmentation fault when I try to run it - any ideas, given that 3.4.0 compiles and runs without issue?

Also, anyone got a link to a clear howto on getting a previous version from git?  I've done a fair bit of looking and found nothing that works, or that is helpful.  Downloading the zip/tar.gz of the source of 3.5.2 doesn't work as it won't compile from the files that github gives you; I've only had luck by cloning the repo which gives you only the current version, which obviously isn't any good to me...  I know there are good reasons for using git but it doesn't seem at all friendly to anyone who wants to download and compile a specific version.

Much easier than you think.


git checkout [branchname]

https://www.atlassian.com/git/tutorial/git-branches#!checkout

I'm clearly an idiot, as I've got nowhere with that.  I can't see how you can specify a previous version from that; running checkout only gives me M libblkmaker.  As I say, I can see how git is clearly useful for someone who is well versed in it and for version management on a project you are contributing to, but it's as clear as mud how I can get the source for 3.5.2 from it; to me, anyway, so I must be stupid.

Should be something along the lines of:

git checkout bfgminer-3.5.x

Do this in the directory you have bfgminer cloned. This will switch the local workspace to that branch.

You can also jump to a specific Tag (version). Instructions here: http://stackoverflow.com/questions/791959/how-to-use-git-to-download-a-particular-tag

[xyzzy099]

I am using the version 3.4.0 BFGMiner that is distributed as part of Bertmod 0.3 on my KnCMiner Saturn.  I added the lines:

Code:

"coinbase-addr": <my payout address>,
"coinbase-sig": <My block sig>

to the end of the config file, because I wanted to solo mine on my local bitcoin-qt.

It doesn't generate any errors, but the hashrate goes down from ~285 GH/s to more like 10 GH/s  

If I remove the above two lines, it goes back to normal.

Anybody have any idea what is going wrong here?

[jedimstr]

I am using the version 3.4.0 BFGMiner that is distributed as part of Bertmod 0.3 on my KnCMiner Saturn.  I added the lines:

Code:

"coinbase-addr": <my payout address>,
"coinbase-sig": <My block sig>

to the end of the config file, because I wanted to solo mine on my local bitcoin-qt.

It doesn't generate any errors, but the hashrate goes down from ~285 GH/s to more like 10 GH/s 

If I remove the above two lines, it goes back to normal.

Anybody have any idea what is going wrong here?

How long has your bitcoin-qt been running. Is the Blockchain fully up-to-date or is it still updating?

[Luke-Jr]

I am using the version 3.4.0 BFGMiner that is distributed as part of Bertmod 0.3 on my KnCMiner Saturn.  I added the lines:

Code:

"coinbase-addr": <my payout address>,
"coinbase-sig": <My block sig>

to the end of the config file, because I wanted to solo mine on my local bitcoin-qt.

It doesn't generate any errors, but the hashrate goes down from ~285 GH/s to more like 10 GH/s  

If I remove the above two lines, it goes back to normal.

Anybody have any idea what is going wrong here?

Is it using 100% CPU, or less? I wonder if the BBB can't keep up with generating work via GBT...

[xyzzy099]

I am using the version 3.4.0 BFGMiner that is distributed as part of Bertmod 0.3 on my KnCMiner Saturn.  I added the lines:

Code:

"coinbase-addr": <my payout address>,
"coinbase-sig": <My block sig>

to the end of the config file, because I wanted to solo mine on my local bitcoin-qt.

It doesn't generate any errors, but the hashrate goes down from ~285 GH/s to more like 10 GH/s  

If I remove the above two lines, it goes back to normal.

Anybody have any idea what is going wrong here?

Is it using 100% CPU, or less? I wonder if the BBB can't keep up with generating work via GBT...

Yeah, looks like that's it - always over 95% CPU when GBT is enabled, ~22% otherwise.

Bummer.

[Luke-Jr]

I am using the version 3.4.0 BFGMiner that is distributed as part of Bertmod 0.3 on my KnCMiner Saturn.  I added the lines:

Code:

"coinbase-addr": <my payout address>,
"coinbase-sig": <My block sig>

to the end of the config file, because I wanted to solo mine on my local bitcoin-qt.

It doesn't generate any errors, but the hashrate goes down from ~285 GH/s to more like 10 GH/s  

If I remove the above two lines, it goes back to normal.

Anybody have any idea what is going wrong here?

Is it using 100% CPU, or less? I wonder if the BBB can't keep up with generating work via GBT...

Yeah, looks like that's it - always over 95% CPU when GBT is enabled, ~22% otherwise.

Bummer.

GBT is entirely unoptimised right now, so there's a possibility of improving it.
In theory, I should be able to reduce its CPU load to approximately the same as stratum's.
Just need to find the time...

[xyzzy099]

I am using the version 3.4.0 BFGMiner that is distributed as part of Bertmod 0.3 on my KnCMiner Saturn.  I added the lines:

Code:

"coinbase-addr": <my payout address>,
"coinbase-sig": <My block sig>

to the end of the config file, because I wanted to solo mine on my local bitcoin-qt.

It doesn't generate any errors, but the hashrate goes down from ~285 GH/s to more like 10 GH/s  

If I remove the above two lines, it goes back to normal.

Anybody have any idea what is going wrong here?

Is it using 100% CPU, or less? I wonder if the BBB can't keep up with generating work via GBT...

Yeah, looks like that's it - always over 95% CPU when GBT is enabled, ~22% otherwise.

Bummer.

GBT is entirely unoptimised right now, so there's a possibility of improving it.
In theory, I should be able to reduce its CPU load to approximately the same as stratum's.
Just need to find the time...

Yeah, I imagine your plate is pretty full.  I like the idea of GBT in general- it enforces some honesty on pool operators (not that any of them are dishonest, to my knowledge ), and it is currently the ONLY real way to solo mine with ASIC h/w without all the complexity of installing and configuring stratum pool software - but it does seem to be pretty heavy-weight at the moment.  I notice that even Eligius will auto-switch you to stratum if you try to use GBT, unless you explicitly disable stratum.

I hope you find some time to look into it soon.

[Luke-Jr]

I like the idea of GBT in general- it enforces some honesty on pool operators (not that any of them are dishonest, to my knowledge )

Even if you assume the pool operator is perfectly honest, blind pooling makes the servers an excellent target for someone to break into...
If you can double-spend enough, it might even be profitable to organise a hold-up of all the big pool operators and force them to let you access the servers so you can do a 10-deep reorg or so.
Fully implemented GBT will make this kind of attack impractical, thus giving pool operators more personal security too.

[Zich]

I got past that but i use USB powered hub ...

But it fails further ...

Code:

[M]anage devices [P]ool management [S]ettings [D]isplay options  [H]elp [Q]uit
 Connected to eu-stratum-lb489kj.btcguild.com diff 2 with stratum as user digi128pci_avalon
 Block: ...7ee030e7 #269461  Diff:511M ( 3.66Ph/s)  Started: [22:51:21]
 ST:2  F:0  NB:1  AS:0  BW:[163/ 31 B/s]  E:0.87  I:14.61uBTC/hr  BS:4
 0            | 202.4/263.0/178.1Mh/s | A:1 R:0+3( 75%) HW:8/ 33%
--------------------------------------------------------------------------------
 NFY 0:       | OFF  /272.7/184.7Mh/s | A:1 R:0+3( 75%) HW:8/ 33%
--------------------------------------------------------------------------------
 [2013-11-13 22:51:23] NFY 0: Previous nonce mismatch (4th try), recalibrating
 [2013-11-13 22:51:25] Accepted 3fa4ee98 NFY 0  Diff 4/2
 [2013-11-13 22:51:25] Stratum from pool 0 requested work update
 [2013-11-13 22:51:25] Pool 0 is hiding block contents from us
 [2013-11-13 22:51:25] Reconnect requested from pool 0 to eu-stratum-lb489kj.btcguild.com:3333
 [2013-11-13 22:51:26] Stratum from pool 0 requested work update
 [2013-11-13 22:51:32] mcp2210_spi_transfer: Failed to issue SPI transfer
 [2013-11-13 22:51:34] mcp2210_spi_transfer: Failed to continue SPI transfer (59 bytes remaining)
 [2013-11-13 22:51:34] NFY 0 failure, attempting to reinitialize
 [2013-11-13 22:51:34] mcp2210_set_cfg_spi: Error setting current SPI config (248)
 [2013-11-13 22:51:34] mcp2210_set_cfg_spi: Error setting current SPI config (248)
 [2013-11-13 22:51:34] NFY 0 failure, disabling!
 [2013-11-13 22:51:35] NFY 0: bitfury_init_oldbuf: Giving up after 4 tries
 [2013-11-13 22:51:35] NFY 0: bitfury_init_oldbuf: Giving up after 4 tries
 [2013-11-13 22:51:36] NFY 0: bitfury_init_oldbuf: Giving up after 4 tries
 [2013-11-13 22:51:37] NFY 0: bitfury_init_oldbuf: Giving up after 4 tries

I got similar result after using usb hub