Bitcoin Forum
December 08, 2016, 02:27:17 AM *
News: To be able to use the next phase of the beta forum software, please ensure that your email address is correct/functional.
 
   Home   Help Search Donate Login Register  
Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 [20] 21 22 23 24 25 26 27 28 29 30 31 32 33 »
  Print  
Author Topic: BTCMiner - Open Source Bitcoin Miner for ZTEX FPGA Boards, 215 MH/s on LX150  (Read 153792 times)
ztex
Donator
Sr. Member
*
Offline Offline

Activity: 367

ZTEX FPGA Boards


View Profile WWW
May 07, 2012, 10:20:41 AM
 #381

Quote
The effect can be reduced by reducing the number of devices per thread (parameter -n). But the only way to solv this is to choose a faster pool server.

Does reducing the number of devices also help if you're seeing overflows?

Usually yes. This depends on the amount of CPU cores, the size of the cluster and the OS. It's probably not wise to run 200 threads on a single core Atom.


Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction. Advertise here.
1481164037
Hero Member
*
Offline Offline

Posts: 1481164037

View Profile Personal Message (Offline)

Ignore
1481164037
Reply with quote  #2

1481164037
Report to moderator
1481164037
Hero Member
*
Offline Offline

Posts: 1481164037

View Profile Personal Message (Offline)

Ignore
1481164037
Reply with quote  #2

1481164037
Report to moderator
1481164037
Hero Member
*
Offline Offline

Posts: 1481164037

View Profile Personal Message (Offline)

Ignore
1481164037
Reply with quote  #2

1481164037
Report to moderator
CA Coins
Donator
Sr. Member
*
Offline Offline

Activity: 306


View Profile
May 07, 2012, 10:30:46 AM
 #382

Quote
The effect can be reduced by reducing the number of devices per thread (parameter -n). But the only way to solv this is to choose a faster pool server.

Does reducing the number of devices also help if you're seeing overflows?

Usually yes. This depends on the amount of CPU cores, the size of the cluster and the OS. It's probably not wise to run 200 threads on a single core Atom.



Thanks, I will give it a try.  Also, not sure if anybody has already mentioned, the block log option is listed as -lb on the help instead of -bl

EDIT:  Just terminated the process remotely with control-c.  When trying to restart, got:
Code:
ztex_ufm1_15y1-2012-L5-A1: New device: bitfile=ztex_ufm1_15y1   f_default=200.00MHz  f_max=240.00MHz  HpC=1.0H
Warning: Error uploading bitstream: FPGA configuration failed: DONE pin does not go high, possible USB transfer errors (INIT_B_HIST=222): Retrying it ...
Warning: Error uploading bitstream: FPGA configuration failed: DONE pin does not go high, possible USB transfer errors (INIT_B_HIST=222): Retrying it ...
Warning: High speed FPGA configuration failed, trying low speed mode:Error uploading bitstream: FPGA configuration failed: DONE pin does not go high, possible USB transfer errors (INIT_B_HIST=222): Trying low speed mode
Warning: Error uploading bitstream: FPGA configuration failed: DONE pin does not go high (size=4220313 ,  0 bytes got lost;  checksum=3 , should be 3;  INIT_B_HIST=222): Retrying it ...
Warning: Error uploading bitstream: FPGA configuration failed: DONE pin does not go high (size=4220313 ,  0 bytes got lost;  checksum=3 , should be 3;  INIT_B_HIST=222): Retrying it ...
I have noticed this before with the quads that went away with a power-cycle.  any thoughts?
BR0KK
Hero Member
*****
Offline Offline

Activity: 742



View Profile
May 07, 2012, 05:42:56 PM
 #383

USB Problem .... I have that too some times.


CA Coins
Donator
Sr. Member
*
Offline Offline

Activity: 306


View Profile
May 07, 2012, 08:43:25 PM
 #384

Any way to fix this remotely without powering off/on the board?

EDIT:  Did a rescan ~10hrs after the initial failed attempt and the board loaded the firmware without problems.  Maybe it needed to wait until the old firmware is unloaded after a control-c?
ztex
Donator
Sr. Member
*
Offline Offline

Activity: 367

ZTEX FPGA Boards


View Profile WWW
May 08, 2012, 08:00:57 AM
 #385

Warning: Error uploading bitstream: FPGA configuration failed: DONE pin does not go high (size=4220313 ,  0 bytes got lost;  checksum=3 , should be 3;  INIT_B_HIST=222): Retrying

That are no USB errors (checksum is correct and nothing gets lost).

Usually that are power problems, i.e. if the load step after configuration cannot be handled (also see my last email). It is also possible that the core voltage regulator gets to hot. If you use the coolers delivered with the board the heat sinks should orientated (can be rotated by 90°) so that the fans blow at the regulators.

Just a note: Stop the cluster using the 'q' command. This puts the boards in suspend mode. (But this is not the cause of you problem)

CA Coins
Donator
Sr. Member
*
Offline Offline

Activity: 306


View Profile
May 08, 2012, 09:03:24 AM
 #386

Warning: Error uploading bitstream: FPGA configuration failed: DONE pin does not go high (size=4220313 ,  0 bytes got lost;  checksum=3 , should be 3;  INIT_B_HIST=222): Retrying

That are no USB errors (checksum is correct and nothing gets lost).

Usually that are power problems, i.e. if the load step after configuration cannot be handled (also see my last email). It is also possible that the core voltage regulator gets to hot. If you use the coolers delivered with the board the heat sinks should orientated (can be rotated by 90°) so that the fans blow at the regulators.

Just a note: Stop the cluster using the 'q' command. This puts the boards in suspend mode. (But this is not the cause of you problem)

It's weird that it was running fine and only happened after I stopped the cluster then tried to restart it.  I have noticed it before with the q command so I stopped using it.  While running, I haven't noticed this problem. 

In terms of the power issue, thanks for the recommendation.  I re-routed the power cables and the quads have been running smoothly.  I'll send you the details/logs via email.
gr0bi42
Full Member
***
Offline Offline

Activity: 158


View Profile WWW
May 09, 2012, 08:09:48 AM
 #387

Quote
The effect can be reduced by reducing the number of devices per thread (parameter -n). But the only way to solv this is to choose a faster pool server.

Does reducing the number of devices also help if you're seeing overflows?

Usually yes. This depends on the amount of CPU cores, the size of the cluster and the OS. It's probably not wise to run 200 threads on a single core Atom.



Thanks, I will give it a try.  Also, not sure if anybody has already mentioned, the block log option is listed as -lb on the help instead of -bl

EDIT:  Just terminated the process remotely with control-c.  When trying to restart, got:
Code:
ztex_ufm1_15y1-2012-L5-A1: New device: bitfile=ztex_ufm1_15y1   f_default=200.00MHz  f_max=240.00MHz  HpC=1.0H
Warning: Error uploading bitstream: FPGA configuration failed: DONE pin does not go high, possible USB transfer errors (INIT_B_HIST=222): Retrying it ...
Warning: Error uploading bitstream: FPGA configuration failed: DONE pin does not go high, possible USB transfer errors (INIT_B_HIST=222): Retrying it ...
Warning: High speed FPGA configuration failed, trying low speed mode:Error uploading bitstream: FPGA configuration failed: DONE pin does not go high, possible USB transfer errors (INIT_B_HIST=222): Trying low speed mode
Warning: Error uploading bitstream: FPGA configuration failed: DONE pin does not go high (size=4220313 ,  0 bytes got lost;  checksum=3 , should be 3;  INIT_B_HIST=222): Retrying it ...
Warning: Error uploading bitstream: FPGA configuration failed: DONE pin does not go high (size=4220313 ,  0 bytes got lost;  checksum=3 , should be 3;  INIT_B_HIST=222): Retrying it ...
I have noticed this before with the quads that went away with a power-cycle.  any thoughts?

+1. Got the same problem several times. Boards are mining without problems. After stop of BTCMiner or reboot sometimes I got this "FPGA configuration failed" errors. Last time yesterday. I disconnected the boards from power for about 3 secs (power off, wait until fan stops, power on). Same error again. Once again power off, waiting around 30 secs, power on => success. Boards are mining fine since then.

Donations are welcome: 1Btf3BqUegfe5iFdWsgfBf1Ew3YsAvsrLT
catfish
Sr. Member
****
Offline Offline

Activity: 270


teh giant catfesh


View Profile
May 09, 2012, 09:46:07 PM
 #388

Hey Stefan

Firstly, many thanks for the swift delivery of the boards, and the excellent packaging. Everything works as expected.

Whilst I've got 10 of the boards running now (5 with Zalman passive heatsinks, 5 with supplied Xilence HSF combos, and I admit I may have used the 'wrong' thermal compound, but they run cooooool), I'm finding that my cheap-and-cheerful 4-port USB2 hubs (bought in bulk, since I knew I'd be needing 10 of them) are unpredictable as hell.

Connect 4 boards to one hub, and the other (plus the hub) to another, and only 3 or 4 (seemingly randomly selected) boards will be visible in scan (-i) mode. Connecting two boards to each 4-port hub, each 4-port hub plugged into the computer, and the 5th board directly connected to the computer, sometimes works.

My theory is that these are poor quality USB hubs, and coupled with the 1-into-5 barrel plug splitter I use (its cable gets stupidly hot, so I am going to have to buy individual barrel plugs and wire them up myself to PCIe power adapters using your recommendation), I may be overloading the USB bus ground.

Is this likely? Or would overloading the USB ground simply 'let the smoke out of the wires'? Whilst my second test pentagon (I'm building my FPGA arrays in sets of five, in order to use 60W ATX power cables per set) has stupidly hot daisy-chain power cables, it hasn't failed yet.... though I'll be re-building it tomorrow. Incidentally, the 2.1/5.5 barrel splitter that gets really hot is a local Maplin part, not your Ztex part, which seems to stay cool. Must be better quality cable.

Once I've done a proper job on powering the boards - using individual barrel plugs fed from PCIe power cables - do you think the USB unreliability will go away?


If not, can you recommend a Ztex-approved USB hub that can be daisy-chained out to the claimed scalability (127 devices on the bus)? Currently, I need a separate computer for every 5 Ztex boards - not due to power supply issues (though power splitting may be the cause), but because simply chaining multiple hubs together to give lots of cascading USB ports doesn't always work.

I expect that this is due to the cheap rubbish I bought from Amazon, but it'd be good to know that this is the case. To connect 25 of your FPGAs to one host computer, do all the USB hubs need to have separate power supplies? If not, can you recommend a suitable known-good USB hub model? I've spent too much on FPGAs to lose it all - especially if due to a £1.50 USB hub with soldering done by a Chinese bloke at the end of a 16 hour shift, shorting out inside the USB hub.

Equally, if using over 3 chained USB hubs requires each hub to be powered, then further expansion will require loads of power sockets just for the USB hub PSUs. Surely in this case, one may as well go without powered hubs and use individual 60W switching-mode PSUs... if one has to have loads of power sockets available anyway? With the whole benefit of using ATX PSUs being clean, protected 12V DC at plenty of current for Ztex FPGA boards, it'd make no sense to require tens of individual PSUs for the necessary USB hubs (which are either 4-port or 7-port according the the USB2 spec, IIRC).


Right now, my systems are hashing away like crazy, but are definitely not as happy as they could be. The power daisy-chain cable is FAR too hot for my comfort, and the USB cables are also hotter than I'd expect.

This is almost certainly due to my cheap initial build - can you recommend a USB hub (preferably one with lots of ports) that you *know* works well with your boards, if powered by ATX PSUs?

all the best
catfish

...so I give in to the rhythm, the click click clack
I'm too wasted to fight back...


BTC: 1A7HvdGGDie3P5nDpiskG8JxXT33Yu6Gct
Turbor
Legendary
*
Offline Offline

Activity: 1008


BitMinter


View Profile WWW
May 10, 2012, 05:26:29 PM
 #389

Ask him about the USB hubs ! Grin https://bitcointalk.org/index.php?topic=40058.msg891599#msg891599


ztex
Donator
Sr. Member
*
Offline Offline

Activity: 367

ZTEX FPGA Boards


View Profile WWW
May 11, 2012, 06:33:34 AM
 #390

My theory is that these are poor quality USB hubs, and coupled with the 1-into-5 barrel plug splitter I use (its cable gets stupidly hot, so I am going to have to buy individual barrel plugs and wire them up myself to PCIe power adapters using your recommendation), I may be overloading the USB bus ground.

Is this likely? Or would overloading the USB ground simply 'let the smoke out of the wires'? Whilst my second test pentagon (I'm building my FPGA arrays in sets of five, in order to use 60W ATX power cables per set) has stupidly hot daisy-chain power cables, it hasn't failed yet.... though I'll be re-building it tomorrow. Incidentally, the 2.1/5.5 barrel splitter that gets really hot is a local Maplin part, not your Ztex part, which seems to stay cool. Must be better quality cable.

The power cables must not get warm or even hot.

For powering your amount of boards I recommend  this method: http://wiki.ztex.de/doku.php?id=en:ztex_boards:ztex_fpga_boards:cluster_power_supplies#usb-fpga_module_115x_and_115dlarge_clusters .

Most critical are the GND lines. As higher their resistance as more current flows through the USB shielding. Therefore I do not recommend daisy-chaining the power cables for larger amounts of boards.

catfish
Sr. Member
****
Offline Offline

Activity: 270


teh giant catfesh


View Profile
May 11, 2012, 07:56:42 PM
 #391

My theory is that these are poor quality USB hubs, and coupled with the 1-into-5 barrel plug splitter I use (its cable gets stupidly hot, so I am going to have to buy individual barrel plugs and wire them up myself to PCIe power adapters using your recommendation), I may be overloading the USB bus ground.

Is this likely? Or would overloading the USB ground simply 'let the smoke out of the wires'? Whilst my second test pentagon (I'm building my FPGA arrays in sets of five, in order to use 60W ATX power cables per set) has stupidly hot daisy-chain power cables, it hasn't failed yet.... though I'll be re-building it tomorrow. Incidentally, the 2.1/5.5 barrel splitter that gets really hot is a local Maplin part, not your Ztex part, which seems to stay cool. Must be better quality cable.

The power cables must not get warm or even hot.

For powering your amount of boards I recommend  this method: http://wiki.ztex.de/doku.php?id=en:ztex_boards:ztex_fpga_boards:cluster_power_supplies#usb-fpga_module_115x_and_115dlarge_clusters .

Most critical are the GND lines. As higher their resistance as more current flows through the USB shielding. Therefore I do not recommend daisy-chaining the power cables for larger amounts of boards.

Excellent, thanks Stefan.

Though I must comment that your boards are *VERY* resilient - I've not been very well and haven't actually got any of the jobs done, yet 10 of the 1.15x boards, running with cheap USB hubs, hot-to-touch daisy-chain power cables (I cut one open to check the wire gauge... it's like ethernet cable), in my cellar office with ambient temperature of 38˚C (not a typo, that's 38˚C ) - all 10 have been working perfectly and the slowest board of the lot still manages 200 MHz and megahash/sec.

Absolutely blinding quality. Once I can acquire enough 2.1/5.5 barrel plugs and solder them up to 18AWG wires, I've got a spare 850W ATX PSU which should be able to power them all.

Oh yeah - and all 10 of my currently-running Ztex FPGAs (I have 25 to build) - are being controlled on Mac OS X Snow Leopard. So any potential Ztex buyers who run Apple Macs.... give me a shout and I'll send you 32-bit or 64-bit libraries for the libusb. Java works OK with the Apple runtime. I had to hack Stefan's makefiles but... needs must. Can't be arsed with building another Linux box, and I won't have Windows on my network.

Seriously - for everyone else here - my setup is pretty much the worst stress-test that Stefan's boards could possibly have. Ambient temp is above working temp, 5 of the boards use passive heatsinks with limited airflow, and the other has out-of-spec resistance on the power lines. If Stefan's kit was duff quality then I'd have broken 10 boards by now, but I've had them running for a week without ANY problems. Each 5-board cluster gives over 1 gigahash/sec. Even at these temperatures.

Nice one. Impressed.

...so I give in to the rhythm, the click click clack
I'm too wasted to fight back...


BTC: 1A7HvdGGDie3P5nDpiskG8JxXT33Yu6Gct
nelisky
Legendary
*
Offline Offline

Activity: 1554


View Profile
May 13, 2012, 04:49:20 PM
 #392


Oh yeah - and all 10 of my currently-running Ztex FPGAs (I have 25 to build) - are being controlled on Mac OS X Snow Leopard. So any potential Ztex buyers who run Apple Macs.... give me a shout and I'll send you 32-bit or 64-bit libraries for the libusb. Java works OK with the Apple runtime. I had to hack Stefan's makefiles but... needs must. Can't be arsed with building another Linux box, and I won't have Windows on my network.


I would be very interested in both the libraries and the recipe to build them myself. Given my ztex boards are running off a tp-link router, I would really like to avoid spinning out my windows or linux VM to update firmwares, and getting java in said router is not something I'm looking forward to do.
BR0KK
Hero Member
*****
Offline Offline

Activity: 742



View Profile
May 13, 2012, 11:58:39 PM
 #393

Im a mac user too and would love to get them running by default under mac ... Then i can get rid of my paralells VM Smiley

CA Coins
Donator
Sr. Member
*
Offline Offline

Activity: 306


View Profile
May 14, 2012, 01:05:27 AM
 #394

Hey Catfish, agree on the build quality of these boards. Would also love to get binaries for running them on OS X. Thanks.
ztex
Donator
Sr. Member
*
Offline Offline

Activity: 367

ZTEX FPGA Boards


View Profile WWW
May 14, 2012, 08:26:05 AM
 #395

Hey Catfish, agree on the build quality of these boards. Would also love to get binaries for running them on OS X. Thanks.

See http://wiki.ztex.de/doku.php?id=en:software:porting&#macos_port

BR0KK
Hero Member
*****
Offline Offline

Activity: 742



View Profile
May 14, 2012, 03:33:55 PM
 #396

Nice to know but where exactly am i supposed to copy that file to?  Working directory <--Where do i find that?

A complete link to a folder would be nice Smiley

ztex
Donator
Sr. Member
*
Offline Offline

Activity: 367

ZTEX FPGA Boards


View Profile WWW
May 15, 2012, 08:44:15 AM
 #397

Nice to know but where exactly am i supposed to copy that file to?  Working directory <--Where do i find that?

A complete link to a folder would be nice Smiley

The working directory is the directory from where you start an application. It can be any directory.

BR0KK
Hero Member
*****
Offline Offline

Activity: 742



View Profile
May 15, 2012, 08:59:02 PM
 #398

Thank u ill try it Smiley

€: with my small knowledge of OSX (and linux) i can't get it to mine with osx.... Ill try it again but it would be nice if you could add OSX support some day in the future.


€1:
It kinda runs now. I copied the file to /library/Java/Extensions

in Terminal i can start BTCMiner now,. but it won't recognize any of my boards ....

if i run it with -i it displays "no devices found"




€2:
i installed libusb but can not talk my devices to mine with btcminer natively. Still no devices found.

ztex
Donator
Sr. Member
*
Offline Offline

Activity: 367

ZTEX FPGA Boards


View Profile WWW
May 23, 2012, 08:10:18 AM
 #399

Thank u ill try it Smiley

€: with my small knowledge of OSX (and linux) i can't get it to mine with osx.... Ill try it again but it would be nice if you could add OSX support some day in the future.


€1:
It kinda runs now. I copied the file to /library/Java/Extensions

in Terminal i can start BTCMiner now,. but it won't recognize any of my boards ....

if i run it with -i it displays "no devices found"




€2:
i installed libusb but can not talk my devices to mine with btcminer natively. Still no devices found.

Looks like USB problems. I'm not familiar with Mac, but a wrong / non-existent libusb should be reported. Are there maybe permission constraints?

BR0KK
Hero Member
*****
Offline Offline

Activity: 742



View Profile
May 23, 2012, 10:00:24 PM
 #400

i have no clue Smiley

Maybe catfish can help me out


€: Saw the great work from eldertyrell today. Will there be a tricone bitstream for Ztex boards?

Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 [20] 21 22 23 24 25 26 27 28 29 30 31 32 33 »
  Print  
 
Jump to:  

Sponsored by , a Bitcoin-accepting VPN.
Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!