OFFICIAL CGMINER mining software thread for linux/win/osx/mips/arm/r-pi 4.11.1

[Askit2]

I'm having a bad time compiling the 3.5.0 cgminer.

I start with a git pull. I finally got that working so I can do everything from the command line.
Actually git clone (con's git address). Pretty sure I copied that too.

well not everything as copying the libusb folder that I have already gotten working I haven't done from the prompt.

Then I have tried both ways
./autogen.sh followed by the configure to link the working lubusb.
and
./autogen.sh --enable- (I put in everything but am too lazy to type it all out.)

./autogen.sh by itself says something about not devices and completes but I am guessing it isn't successfully (am assuming the opencl thing made this work before) but configure completes seemingly correctly and says mining is enabled on all the devices.

./autogen.sh with enables runs and says mining is enabled on all the devices that I enabled and actually all that seem to be in the main distro. I just copied kano's.

make clean completes without errors

make completes without errors

Ok I noticed under ./autogen.sh that a ../ltmain.sh is required and not found. I wonder If I can use one from the last version or why that one didn't clone.
if I copy one as soon as I run the autogen.sh it is deleted or removed in some way before the check for it shows up. I have tried copying it in after starting ./autogen.sh. Maybe this time it won't disappear.

./cgminer says something like starting cgminer 3.5.0Segmentation fault

I have always made my own since I got my Raspberry pi. The version 3.4.3 I made still runs. So it isn't like I am screwed. I did everything the same way the first time, and a few other times. I tried Kano's way the rest of the times.

[-ck]

I'm having a bad time compiling the 3.5.0 cgminer.

I start with a git pull. I finally got that working so I can do everything from the command line.
Actually git clone (con's git address). Pretty sure I copied that too.

well not everything as copying the libusb folder that I have already gotten working I haven't done from the prompt.

Then I have tried both ways
./autogen.sh followed by the configure to link the working lubusb.
and
./autogen.sh --enable- (I put in everything but am too lazy to type it all out.)

There is no way to link in your own libusb any more. You have to use the statically included libusb.

[-ck]

Hello,

i have the same problem with Timeout messages for all BE.

AMU1: TIMEOUT GetResults took 1531ms but was 100ms
AMU0: TIMEOUT GetResults took 1531ms but was 100ms
AMU3: TIMEOUT GetResults took 1531ms but was 100ms
AMU2: TIMEOUT GetResults took 1531ms but was 100ms

Timeout variates from few 100ms to 2000ms or more. 
Till nowi used cgminer 3.3.0 and there was no problem, when i tried 3.4.0 and now currently 3.5.0 version i got this messagess.
I run cgminer in Windows XP SP2.
Zadig drivers are installed ( otherwise i couldn't run cgminer 3.3.0 ).
USB HUB has 3.5A max output, so there is no under power problem.
I also rebooted XP before i started cgminer 3.5.0.

A number of things potentially going on. We have problems with some usb implementations on certain OSs and the libusb library we use. 

How long does it normally take before you see these errors and do all of them have it at the same time?

XP is a bit limited for USB. Try starting with just 2 BEs and increase it by one at a time (waiting long enough for the errors to normally appear) till you see the errors reappear.
If possible, try windows7 or ideally a linux to see if they go away.

FYI I randomly see this error on my win7 machines.  I can usually cause it by initiating other things on the PC.  The PC isn't resource starved, but I definitely noticed the errors pop up when I kick something off, like say windows update.

What a disgrace, we've been fighting with this issue for as long as I can remember and we went to all that effort to include the least worst version of libusb statically yet it still doesn't go away. <insert frustration tag>

Here try a build with statically linked version of the latest libusbx instead please since you have a way of reproducing the issue:
http://ck.kolivas.org/apps/cgminer/temp/cgminer-nogpu.exe
http://ck.kolivas.org/apps/cgminer/temp/cgminer.exe

[Askit2]

I'm having a bad time compiling the 3.5.0 cgminer.

I start with a git pull. I finally got that working so I can do everything from the command line.
Actually git clone (con's git address). Pretty sure I copied that too.

well not everything as copying the libusb folder that I have already gotten working I haven't done from the prompt.

Then I have tried both ways
./autogen.sh followed by the configure to link the working lubusb.
and
./autogen.sh --enable- (I put in everything but am too lazy to type it all out.)

There is no way to link in your own libusb any more. You have to use the statically included libusb.

I figured that much out. I tried it using the older directions first. I later retried with Kano's code that he posts for his other versions.

The issue seems to be a missing file at least so far.

Ok I noticed under ./autogen.sh that a ../ltmain.sh is required and not found. I wonder If I can use one from the last version or why that one didn't clone.
If I copy one as soon as I run the autogen.sh it is deleted or removed in some way before the check for it shows up. I have tried copying it in after starting ./autogen.sh. Maybe this time it won't disappear.

Never mind my impatience has gotten the better of me. It deletes, checks, then copies one in. I didn't wait long enough.

[-ck]

I'm having a bad time compiling the 3.5.0 cgminer.

I start with a git pull. I finally got that working so I can do everything from the command line.
Actually git clone (con's git address). Pretty sure I copied that too.

well not everything as copying the libusb folder that I have already gotten working I haven't done from the prompt.

Then I have tried both ways
./autogen.sh followed by the configure to link the working lubusb.
and
./autogen.sh --enable- (I put in everything but am too lazy to type it all out.)

There is no way to link in your own libusb any more. You have to use the statically included libusb.

I figured that much out. I tried it using the older directions first. I later retried with Kano's code that he posts for his other versions.

The issue seems to be a missing file at least so far.

Ok I noticed under ./autogen.sh that a ../ltmain.sh is required and not found. I wonder If I can use one from the last version or why that one didn't clone.
If I copy one as soon as I run the autogen.sh it is deleted or removed in some way before the check for it shows up. I have tried copying it in after starting ./autogen.sh. Maybe this time it won't disappear.

ltmain.sh is supposed to be generated by the autotools themselves and is deleted every time.

[Beastlymac]

Initial results for 3.5.0 on Raspberry Pi with Bitfury BF1 and BFL gear are underwhelming. Running overnight I woke up to the BF1s running anywhere from 0 to about 1.2 Gh/s. I restarted cgminer and things seem to pick back up. Will report the status when I get home from work

The BF1 draws power entirely from USB. How are you connecting it to an RPi ?

It is running on a powered USB hub (albeit a cheap one) with an external 4A supply running five BF1s, it ran stable fine from Ubuntu and BFGminer for a few days so the hub shouldn't be a problem

Edit - Pi is running new Raspbian release

Seems to be something with raspbian wheezy. I have had it up stable for around 10 hours now on minepeon.

[Askit2]

I'm having a bad time compiling the 3.5.0 cgminer.

I start with a git pull. I finally got that working so I can do everything from the command line.
Actually git clone (con's git address). Pretty sure I copied that too.

well not everything as copying the libusb folder that I have already gotten working I haven't done from the prompt.

Then I have tried both ways
./autogen.sh followed by the configure to link the working lubusb.
and
./autogen.sh --enable- (I put in everything but am too lazy to type it all out.)

There is no way to link in your own libusb any more. You have to use the statically included libusb.

I figured that much out. I tried it using the older directions first. I later retried with Kano's code that he posts for his other versions.

The issue seems to be a missing file at least so far.

Ok I noticed under ./autogen.sh that a ../ltmain.sh is required and not found. I wonder If I can use one from the last version or why that one didn't clone.
If I copy one as soon as I run the autogen.sh it is deleted or removed in some way before the check for it shows up. I have tried copying it in after starting ./autogen.sh. Maybe this time it won't disappear.

ltmain.sh is supposed to be generated by the autotools themselves and is deleted every time.

./autogen.sh fails with a can not find makefile.in.

[-ck]

Initial results for 3.5.0 on Raspberry Pi with Bitfury BF1 and BFL gear are underwhelming. Running overnight I woke up to the BF1s running anywhere from 0 to about 1.2 Gh/s. I restarted cgminer and things seem to pick back up. Will report the status when I get home from work

The BF1 draws power entirely from USB. How are you connecting it to an RPi ?

It is running on a powered USB hub (albeit a cheap one) with an external 4A supply running five BF1s, it ran stable fine from Ubuntu and BFGminer for a few days so the hub shouldn't be a problem

Edit - Pi is running new Raspbian release

Seems to be something with raspbian wheezy. I have had it up stable for around 10 hours now on minepeon.

In that case I wonder if it's the older libudev in raspbian? We continue to get shafted by external libraries...

[-ck]

./autogen.sh fails with a can not find makefile.in.

Maybe you're missing some packages on your distro? Check you have autoconf automake and libtool at the very least

[Askit2]

./autogen.sh fails with a can not find makefile.in.

Maybe you're missing some packages on your distro? Check you have autoconf automake and libtool at the very least

already installed says newest version when I try again.

[-ck]

./autogen.sh fails with a can not find makefile.in.

Maybe you're missing some packages on your distro? Check you have autoconf automake and libtool at the very least

already installed says newest version when I try again.

I just git cloned on arch RPi and it built fine so not sure what's going on... Heck it even mines fine on the bitfury plugged straight into it.

[MineForeman.com]

./autogen.sh fails with a can not find makefile.in.

Maybe you're missing some packages on your distro? Check you have autoconf automake and libtool at the very least

already installed says newest version when I try again.

Make sure with arch you;-

pacman -Syu base-devel

That will give you everything you need to compile cgminer.

[mdude77]

What a disgrace, we've been fighting with this issue for as long as I can remember and we went to all that effort to include the least worst version of libusb statically yet it still doesn't go away. <insert frustration tag>

Here try a build with statically linked version of the latest libusbx instead please since you have a way of reproducing the issue:
http://ck.kolivas.org/apps/cgminer/temp/cgminer-nogpu.exe
http://ck.kolivas.org/apps/cgminer/temp/cgminer.exe

Code:

 cgminer version 3.5.0 - Started: [2013-09-30 20:32:53]
--------------------------------------------------------------------------------
 (5s):11.47G (avg):9.541Gh/s | A:92  R:0  HW:6  WU:135.8/m
 ST: 2  SS: 0  NB: 1  LW: 211  GF: 0  RF: 0
 Connected to xxxxxxx diff 8 with stratum as user xxxx
 Block: 0003f366a3179664...  Diff:149M  Started: [20:32:53]  Best share: 70
--------------------------------------------------------------------------------
 [P]ool management [S]ettings [D]isplay options [Q]uit
 AMU  0:                | 242.6M/287.2Mh/s | A:16 R:0 HW:0 WU:  4.9/m
 AMU  1:                | 262.7M/293.0Mh/s | A: 4 R:0 HW:0 WU:  9.8/m
 AMU  2:                | 238.1M/247.4Mh/s | A: 7 R:0 HW:0 WU:  8.2/m
 AMU  3:                | 247.0M/348.7Mh/s | A: 0 R:0 HW:0 WU:  1.6/m
 AMU  4:                | 239.3M/268.3Mh/s | A: 9 R:0 HW:0 WU:  4.9/m
 AMU  5:                | 241.7M/292.8Mh/s | A: 0 R:0 HW:0 WU:  4.9/m
 AMU  6:                | 266.3M/325.9Mh/s | A: 9 R:0 HW:0 WU:  6.5/m
 AMU  7:                | 215.9M/246.0Mh/s | A: 0 R:0 HW:0 WU:  3.3/m
 AMU  8:                | 250.0M/346.2Mh/s | A: 1 R:0 HW:0 WU:  6.5/m
 AMU  9:                | 263.2M/318.2Mh/s | A: 1 R:0 HW:0 WU:  4.9/m
 AMU 10:                | 271.2M/327.2Mh/s | A: 2 R:0 HW:0 WU:  8.2/m
 AMU 11:                | 236.3M/316.1Mh/s | A: 8 R:0 HW:0 WU:  3.3/m
 AMU 12:                | 197.2M/239.0Mh/s | A: 0 R:0 HW:0 WU:  3.3/m
 AMU 13:                | 250.6M/351.1Mh/s | A: 8 R:0 HW:0 WU:  3.3/m
 AMU 14:                | 214.9M/311.1Mh/s | A: 0 R:0 HW:0 WU:  0.0/m
 AMU 15:                | 231.1M/275.7Mh/s | A: 1 R:0 HW:0 WU:  4.9/m
 AMU 16:                | 235.6M/311.5Mh/s | A: 1 R:0 HW:0 WU:  1.6/m
 AMU 17:                | 225.6M/260.5Mh/s | A: 1 R:0 HW:0 WU:  3.3/m
 AMU 18:                | 201.0M/242.8Mh/s | A: 0 R:0 HW:0 WU:  3.3/m
 AMU 19:                | 260.1M/332.6Mh/s | A: 1 R:0 HW:1 WU:  3.3/m
 AMU 20:                | 309.1M/427.4Mh/s | A: 1 R:0 HW:1 WU:  1.6/m
 AMU 21:                | 219.4M/290.3Mh/s | A: 1 R:0 HW:1 WU:  3.3/m
 AMU 22:                | 267.6M/338.2Mh/s | A: 0 R:0 HW:1 WU:  6.5/m
 AMU 23:                | 245.6M/331.1Mh/s | A: 0 R:0 HW:0 WU:  4.9/m
 AMU 24:                | 240.0M/336.9Mh/s | A: 1 R:0 HW:1 WU:  6.5/m
 AMU 25:                | 270.0M/346.5Mh/s | A: 1 R:0 HW:0 WU:  6.5/m
 AMU 26:                | 215.1M/311.4Mh/s | A: 0 R:0 HW:0 WU:  0.0/m
 AMU 27:                | 215.1M/311.3Mh/s | A: 0 R:0 HW:1 WU:  0.0/m
 AMU 28:                | 256.5M/239.1Mh/s | A: 0 R:0 HW:0 WU:  9.8/m
 AMU 29:                | 243.9M/317.1Mh/s | A: 1 R:0 HW:0 WU:  3.3/m
 AMU 30:                | 230.1M/302.0Mh/s | A:17 R:0 HW:0 WU:  6.5/m
 AMU 31:                | 248.6M/268.3Mh/s | A: 1 R:0 HW:0 WU:  6.5/m
 AMU 32:                | 215.3M/311.6Mh/s | A: 0 R:0 HW:0 WU:  0.0/m
--------------------------------------------------------------------------------

 [2013-09-30 20:33:10] AMU25: TIMEOUT GetResults took 892ms but was 100ms
 [2013-09-30 20:33:10] AMU19: TIMEOUT GetResults took 892ms but was 100ms
 [2013-09-30 20:33:10] AMU17: TIMEOUT GetResults took 892ms but was 100ms
 [2013-09-30 20:33:10] AMU24: TIMEOUT GetResults took 893ms but was 100ms
 [2013-09-30 20:33:10] AMU13: TIMEOUT GetResults took 893ms but was 100ms
 [2013-09-30 20:33:10] AMU1: TIMEOUT GetResults took 893ms but was 100ms
 [2013-09-30 20:33:10] AMU0: TIMEOUT GetResults took 893ms but was 100ms
 [2013-09-30 20:33:10] AMU16: TIMEOUT GetResults took 702ms but was 100ms
 [2013-09-30 20:33:10] AMU4: TIMEOUT GetResults took 894ms but was 100ms
 [2013-09-30 20:33:10] AMU26: TIMEOUT GetResults took 894ms but was 100ms
 [2013-09-30 20:33:10] AMU10: TIMEOUT GetResults took 834ms but was 100ms
 [2013-09-30 20:33:10] AMU27: TIMEOUT GetResults took 903ms but was 100ms
 [2013-09-30 20:33:16] Accepted 0d96b852 Diff 18/8 AMU 6 pool 0
 [2013-09-30 20:33:25] Accepted 0c337ca2 Diff 20/8 AMU 0 pool 0
 [2013-09-30 20:33:26] Accepted 17860599 Diff 10/8 AMU 30 pool 0
 [2013-09-30 20:33:28] Accepted 0c127d24 Diff 21/8 AMU 11 pool 0

Sorry, I still get it.  I can reproduce it on demand by firing up my cgminer instance for LTC mining.

M

[-ck]

What a disgrace, we've been fighting with this issue for as long as I can remember and we went to all that effort to include the least worst version of libusb statically yet it still doesn't go away. <insert frustration tag>

Here try a build with statically linked version of the latest libusbx instead please since you have a way of reproducing the issue:
http://ck.kolivas.org/apps/cgminer/temp/cgminer-nogpu.exe
http://ck.kolivas.org/apps/cgminer/temp/cgminer.exe

Sorry, I still get it.  I can reproduce it on demand by firing up my cgminer instance for LTC mining.

Thanks for trying. It was worth a try anyway but we think we had the least worst library in there already. At this rate I may end up having to write my own USB library. Rest assured we shall continue to bash our heads against brick walls on this one.

[techman05]

when I get a latency warning I always thought it was a crappy share doing it. I've got/had 3.4.3 running in rotate mode since the pool I keep wanting to use forces a restart of all mining. When that happens my room glows like a scooby doo villan and though tje show is fine I WANT TO SLEEP. I then force switch with the switch pool option and let it do its work again.
I'm running windows 7 pc with only 6 eroupters so far.

[lano1106]

There is no way to link in your own libusb any more. You have to use the statically included libusb.

Well, technically, it is not totally exact. To be more accurate, there is no more official way to link with your own libs. However, I did just that by hacking a couple of lines in the configure script.

Unless Con resort to cryptography and obfuscation, there willl always be a way :-)

If I weren't sucking with automake and autoconf script syntax, I would offer you new configure switches ie:

--with-libusb=path_to_pc_file       Use system libusb libs

I have seen a couple of projects doing it like this. ie: Provide a safe version, let the flexibility to users willing to use their own version.

Firefox come to my mind.

[Askit2]

./autogen.sh fails with a can not find makefile.in.

Maybe you're missing some packages on your distro? Check you have autoconf automake and libtool at the very least

already installed says newest version when I try again.

I just git cloned on arch RPi and it built fine so not sure what's going on... Heck it even mines fine on the bitfury plugged straight into it.

Apparently I had a id10T error. It should have worked fine. I had run it so many times since I had used cflags blah blah blah ./autogen.sh blah blah.... I forgot to add the cflags on the start..... My first way was exactly how I had been running it to use the other libusb. copy the git, ./autogen.sh, CFLAGS configure, make clean, make. When I had issues I added to ./autogen.sh but forgot the cflags. So things didn't work. As I should have expected. I am still unsure about if the seg fault will come back but it seems able to build without me messing it up.

[-ck]

What a disgrace, we've been fighting with this issue for as long as I can remember and we went to all that effort to include the least worst version of libusb statically yet it still doesn't go away. <insert frustration tag>

Here try a build with statically linked version of the latest libusbx instead please since you have a way of reproducing the issue:
http://ck.kolivas.org/apps/cgminer/temp/cgminer-nogpu.exe
http://ck.kolivas.org/apps/cgminer/temp/cgminer.exe

Sorry, I still get it.  I can reproduce it on demand by firing up my cgminer instance for LTC mining.

Thanks for trying. It was worth a try anyway but we think we had the least worst library in there already. At this rate I may end up having to write my own USB library. Rest assured we shall continue to bash our heads against brick walls on this one.

Out of sheer desperation, I have rewritten the usb transfer functions to use asynchronous communications and implemented our own timeout mechanism within the cgminer code to cancel transfers as an experiment instead of relying on libusb. Can you please try redownloading one of those binaries listed above again to see if it makes any difference?

[zamot]

Hello,

i have the same problem with Timeout messages for all BE.

AMU1: TIMEOUT GetResults took 1531ms but was 100ms
AMU0: TIMEOUT GetResults took 1531ms but was 100ms
AMU3: TIMEOUT GetResults took 1531ms but was 100ms
AMU2: TIMEOUT GetResults took 1531ms but was 100ms

Timeout variates from few 100ms to 2000ms or more. 
Till nowi used cgminer 3.3.0 and there was no problem, when i tried 3.4.0 and now currently 3.5.0 version i got this messagess.
I run cgminer in Windows XP SP2.
Zadig drivers are installed ( otherwise i couldn't run cgminer 3.3.0 ).
USB HUB has 3.5A max output, so there is no under power problem.
I also rebooted XP before i started cgminer 3.5.0.

A number of things potentially going on. We have problems with some usb implementations on certain OSs and the libusb library we use. 

How long does it normally take before you see these errors and do all of them have it at the same time?

XP is a bit limited for USB. Try starting with just 2 BEs and increase it by one at a time (waiting long enough for the errors to normally appear) till you see the errors reappear.
If possible, try windows7 or ideally a linux to see if they go away.

This timeout messages appears randomly time based, but around 10 - 15 seconds after starting of cgminer. I tried with one BE and then i added one per one other BE's, but timeout appears with one attached BE also.
I will try with cgminer that you posted link with next post.
Unfortunaly i can not try directly on Windows 7, because i do not have all privileges to install something into.
the next ry ist with linux, but i must first install it.

[CrazyGuy]

CentOS release 6.4 (Final) x64 - cgminer 3.5.0

segmentation fault trying to when trying to limit to 1 BAS device "--usb BAS:1" , no issues with "--usb :1"

Core was generated by `/home/crazyguy/cgminer/cgminer --usb BAS:1'.
Program terminated with signal 11, Segmentation fault.
#0  0x0000003dfb084f1d in __strcasecmp_l_sse2 () from /lib64/libc.so.6