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

[-ck]

Hi!

Testing version 2.6.4 on Ubuntu 12.04, found a (minor) bug, but I think it could be an older one. Yesterday I configured my system to use the onboard GPU (3300) for display and the PCIe card (5770) for mining only. Unfortunately, ADL and CL GPU mapping does not seem to work.

Ok, fixed it by myself. There is definitely a bug. When finding ADL devices, one loop mixed up CL and ADL devices. Here is the diff which fixes it:

Code:

hko@cave:~/src/cgminer/cgminer$ git diff
diff --git a/adl.c b/adl.c
index 69c06de..c60f08c 100644
--- a/adl.c
+++ b/adl.c
@@ -339,16 +339,16 @@ void init_adl(int nDevs)
                }
        }
 
-       for (gpu = 0; gpu < devices; gpu++) {
+       for (gpu = 0; gpu < nDevs; gpu++) {
                struct gpu_adl *ga;
                int iAdapterIndex;
                int lpAdapterID;
                ADLODPerformanceLevels *lpOdPerformanceLevels;
                int lev;
-
-               i = vadapters[gpu].id;
+               int adlGpu = gpus[gpu].virtual_adl;
+               i = vadapters[adlGpu].id;
                iAdapterIndex = lpInfo[i].iAdapterIndex;
-               gpus[gpu].virtual_gpu = vadapters[gpu].virtual_gpu;
+               gpus[gpu].virtual_gpu = vadapters[adlGpu].virtual_gpu;
 
                /* Get unique identifier of the adapter, 0 means not AMD */
                result = ADL_Adapter_ID_Get(iAdapterIndex, &lpAdapterID);
@@ -358,11 +358,11 @@ void init_adl(int nDevs)
                }
 
                if (gpus[gpu].deven == DEV_DISABLED) {
-                       gpus[i].gpu_engine =
-                       gpus[i].gpu_memclock =
-                       gpus[i].gpu_vddc =
-                       gpus[i].gpu_fan =
-                       gpus[i].gpu_powertune = 0;
+                       gpus[gpu].gpu_engine =
+                       gpus[gpu].gpu_memclock =
+                       gpus[gpu].gpu_vddc =
+                       gpus[gpu].gpu_fan =
+                       gpus[gpu].gpu_powertune = 0;
                        continue;
                }
 

and a debug log file fromfixed code:

Code:

[2012-08-11 16:01:32] Started cgminer 2.6.4
 [2012-08-11 16:01:32] Loaded configuration file /home/xxxxxxx/.cgminer/cgminer.conf
 [2012-08-11 16:01:32] CL Platform 0 vendor: Advanced Micro Devices, Inc.
 [2012-08-11 16:01:32] CL Platform 0 name: AMD Accelerated Parallel Processing
 [2012-08-11 16:01:32] CL Platform 0 version: OpenCL 1.2 AMD-APP (937.2)
 [2012-08-11 16:01:32] Platform 0 devices: 1
 [2012-08-11 16:01:32] 	0	Juniper
 [2012-08-11 16:01:32] GPU 0 iAdapterIndex 0 strUDID 296:38420:4098:33613:4163 iBusNumber 1 iDeviceNumber 
5 iFunctionNumber 0 iVendorID 4098 strAdapterName  ATI Radeon HD 3300 Graphics 
 [2012-08-11 16:01:32] GPU 1 iAdapterIndex 2 strUDID 512:26808:4098:9539:4098 iBusNumber 2 iDeviceNumber 0
 iFunctionNumber 0 iVendorID 4098 strAdapterName  ATI Radeon HD 5700 Series 
 [2012-08-11 16:01:32] ADL found more devices than opencl!
 [2012-08-11 16:01:32] There is possibly at least one GPU that doesn't support OpenCL
 [2012-08-11 16:01:32] Use the gpu map feature to reliably map OpenCL to ADL
 [2012-08-11 16:01:32] Mapping OpenCL device 0 to ADL device 1
 [2012-08-11 16:01:32] WARNING: Number of OpenCL and ADL devices did not match!
 [2012-08-11 16:01:32] Hardware monitoring may NOT match up with devices!
 [2012-08-11 16:01:32] GPU 0 ATI Radeon HD 5700 Series hardware monitoring enabled
 [2012-08-11 16:01:32] Setting GPU 0 engine clock to 945
 [2012-08-11 16:01:32] Setting GPU 0 memory clock to 180
 [2012-08-11 16:01:32] Setting GPU 0 voltage to 1.125
 [2012-08-11 16:01:32] Popping work to work thread
 [2012-08-11 16:01:32] Probing for an alive pool
 [2012-08-11 16:01:32] Popping work to stage thread

---
Nite69
WalletId: 17gNvfoD2FDqTfESUxNEmTukGbGVAiJhXp

Much appreciated, thanks!

[-ck]

For those having issues with bamt continually restarting the latest cgminer, I've tracked it down to the mother file in bamt in /opt/bamt/mother . As a temporary fix, you can copy this hacked one to replace it:
http://ck.kolivas.org/apps/cgminer/temp/mother

[Nite69]

Hi!

Testing version 2.6.4 on Ubuntu 12.04, found a (minor) bug, but I think it could be an older one. Yesterday I configured my system to use the onboard GPU (3300) for display and the PCIe card (5770) for mining only. Unfortunately, ADL and CL GPU mapping does not seem to work.

Ok, fixed it by myself. There is definitely a bug. When finding ADL devices, one loop mixed up CL and ADL devices. Here is the diff which fixes it:

Code:

hko@cave:~/src/cgminer/cgminer$ git diff
diff --git a/adl.c b/adl.c
index 69c06de..c60f08c 100644
--- a/adl.c
+++ b/adl.c
@@ -339,16 +339,16 @@ void init_adl(int nDevs)
                }
        }
 
-       for (gpu = 0; gpu < devices; gpu++) {
+       for (gpu = 0; gpu < nDevs; gpu++) {
                struct gpu_adl *ga;
                int iAdapterIndex;
                int lpAdapterID;
                ADLODPerformanceLevels *lpOdPerformanceLevels;
                int lev;
-
-               i = vadapters[gpu].id;
+               int adlGpu = gpus[gpu].virtual_adl;
+               i = vadapters[adlGpu].id;
                iAdapterIndex = lpInfo[i].iAdapterIndex;
-               gpus[gpu].virtual_gpu = vadapters[gpu].virtual_gpu;
+               gpus[gpu].virtual_gpu = vadapters[adlGpu].virtual_gpu;
 
                /* Get unique identifier of the adapter, 0 means not AMD */
                result = ADL_Adapter_ID_Get(iAdapterIndex, &lpAdapterID);
@@ -358,11 +358,11 @@ void init_adl(int nDevs)
                }
 
                if (gpus[gpu].deven == DEV_DISABLED) {
-                       gpus[i].gpu_engine =
-                       gpus[i].gpu_memclock =
-                       gpus[i].gpu_vddc =
-                       gpus[i].gpu_fan =
-                       gpus[i].gpu_powertune = 0;
+                       gpus[gpu].gpu_engine =
+                       gpus[gpu].gpu_memclock =
+                       gpus[gpu].gpu_vddc =
+                       gpus[gpu].gpu_fan =
+                       gpus[gpu].gpu_powertune = 0;
                        continue;
                }
 

and a debug log file fromfixed code:

Code:

[2012-08-11 16:01:32] Started cgminer 2.6.4
 [2012-08-11 16:01:32] Loaded configuration file /home/xxxxxxx/.cgminer/cgminer.conf
 [2012-08-11 16:01:32] CL Platform 0 vendor: Advanced Micro Devices, Inc.
 [2012-08-11 16:01:32] CL Platform 0 name: AMD Accelerated Parallel Processing
 [2012-08-11 16:01:32] CL Platform 0 version: OpenCL 1.2 AMD-APP (937.2)
 [2012-08-11 16:01:32] Platform 0 devices: 1
 [2012-08-11 16:01:32] 	0	Juniper
 [2012-08-11 16:01:32] GPU 0 iAdapterIndex 0 strUDID 296:38420:4098:33613:4163 iBusNumber 1 iDeviceNumber 
5 iFunctionNumber 0 iVendorID 4098 strAdapterName  ATI Radeon HD 3300 Graphics 
 [2012-08-11 16:01:32] GPU 1 iAdapterIndex 2 strUDID 512:26808:4098:9539:4098 iBusNumber 2 iDeviceNumber 0
 iFunctionNumber 0 iVendorID 4098 strAdapterName  ATI Radeon HD 5700 Series 
 [2012-08-11 16:01:32] ADL found more devices than opencl!
 [2012-08-11 16:01:32] There is possibly at least one GPU that doesn't support OpenCL
 [2012-08-11 16:01:32] Use the gpu map feature to reliably map OpenCL to ADL
 [2012-08-11 16:01:32] Mapping OpenCL device 0 to ADL device 1
 [2012-08-11 16:01:32] WARNING: Number of OpenCL and ADL devices did not match!
 [2012-08-11 16:01:32] Hardware monitoring may NOT match up with devices!
 [2012-08-11 16:01:32] GPU 0 ATI Radeon HD 5700 Series hardware monitoring enabled
 [2012-08-11 16:01:32] Setting GPU 0 engine clock to 945
 [2012-08-11 16:01:32] Setting GPU 0 memory clock to 180
 [2012-08-11 16:01:32] Setting GPU 0 voltage to 1.125
 [2012-08-11 16:01:32] Popping work to work thread
 [2012-08-11 16:01:32] Probing for an alive pool
 [2012-08-11 16:01:32] Popping work to stage thread

---
Nite69
WalletId: 17gNvfoD2FDqTfESUxNEmTukGbGVAiJhXp

Much appreciated, thanks!

Please check the functionality of the patch, I'm not very familiar with the code as whole. Feel free to use, copy modify etc the pathc.. of course, if you find a better way to fix it, please do.

[420]

I suggest another thread for CGMINER BFLT compatibility

[CrazyGuy]

Code:

So I am trying to get 4x5970 working under Linux.

Aticonfig shows 8 GPUs but cgminer bombs out.
[code]
* 0. 17:00.0 ATI Radeon HD 5900 Series
  1. 16:00.0 ATI Radeon HD 5900 Series
  2. 13:00.0 ATI Radeon HD 5900 Series
  3. 12:00.0 ATI Radeon HD 5900 Series
  4. 0b:00.0 ATI Radeon HD 5900 Series
  5. 0a:00.0 ATI Radeon HD 5900 Series
  6. 07:00.0 ATI Radeon HD 5900 Series
  7. 06:00.0 ATI Radeon HD 5900 Series

* - Default adapter

cgminer -n

Code:

X Error of failed request:  BadValue (integer parameter out of range for operation)
  Major opcode of failed request:  140 (ATIFGLEXTENSION)
  Minor opcode of failed request:  67 ()
  Value in failed request:  0x17
  Serial number of failed request:  21
  Current serial number in output stream:  21

cgminer -D -T (w/ config file) has same error.

Unplugging PCIe power cables for one 5970 (6 active GPUs) works.[/code]

Has this been resolved? I'm having the same issue. Works fine with 6 gpus but not 8. I had 8 gpus running fine under linuxcoin for 6 months with phoenix miner. I've decided to upgrade to xubuntu 12.04 and I'd prefer not to go back to phoenix miner.

[gnar1ta$]

My temps have gone over the overheat threshold but didn't give me any kind of message.

What is cgminer's behavior supposed to be? Notifying or merely scaling back clock/voltages?

Maybe that only controls the auto fan speeds to the desired temperature but there's a temp cutoff i think i already reached and it did nothing as well or I wans't there to see anything done it was hashing at just over 90 degrees

i thought the fan speed would JUST be tied to the temperature GOAL, making overheat temperature a useless stat so far unless it IS only for setting AUTO GPU CLOCK which I don't use that feature

You need to use auto-gpu for temp cutoff to work, it will show OFF for any device that hits the value (95 default). If you don't want your clocks to throttle set a static range like 900-900. It's in the readme by the way 

[420]

My temps have gone over the overheat threshold but didn't give me any kind of message.

What is cgminer's behavior supposed to be? Notifying or merely scaling back clock/voltages?

Maybe that only controls the auto fan speeds to the desired temperature but there's a temp cutoff i think i already reached and it did nothing as well or I wans't there to see anything done it was hashing at just over 90 degrees

i thought the fan speed would JUST be tied to the temperature GOAL, making overheat temperature a useless stat so far unless it IS only for setting AUTO GPU CLOCK which I don't use that feature

You need to use auto-gpu for temp cutoff to work, it will show OFF for any device that hits the value (95 default). If you don't want your clocks to throttle set a static range like 900-900. It's in the readme by the way  

i doesn't seem like the 95C throttling lowered my mhash very much, but if they keep running that hot may be bad, should set 90C maybe

okay radeon drivers 12.6 and CGMINER isn't setting my vddc voltage, anyone have suggestions? (used driver sweeper before installing this driver, and no current running third party overclock programs, i have to run them to set over voltage)

it also appears like 9.5 is an intensity that works

[bitlane]

FOR SCRYPT MINING:
I thought I should pop in here and comment.....

CGMiner 2.6.4 is ROCK SOLID. No Hicupps, No Crashes...nothing.

I am currently using it on all 3 recent generations of ATI Cards (5/6/7xxx) all using CAT 12.6/SDK 2.6 and everything just seems to WORK.

I have also found that the AUTOTUNE feature works great. I only specify --shaders, -I (Intensity) and -g (threads) and the performance is better than if I mess with it.

This is my HD 7970 BAT file for LTC Mining....and as I said, it doesn't require any screwing around. Autotune is great.

Code:

cgminer ^
-o http://litecoinpool.org:9332 -u bitlane -p password ^
-o http://192.168.0.50:9332 -u bitlane -p password ^
--failover-only --api-listen --api-network --api-port 4028 --api-allow W:192.168.0/24 --auto-fan --temp-target 69 ^
--scrypt --shaders 2048 -g 4 -I 13

As my BAT file strings were getting unmanageable and LONG, I have resorted to using the CARET (^/shift-6) to continue long lines and avoid side-scrolling.
(I still don't like using config files).
I would recommend it for anyone still using BAT files in Windows and wants to keep everything straight/easy to read etc.

Thanks again for all your hard work Con.

[-ck]

FOR SCRYPT MINING:
I thought I should pop in here and comment.....

CGMiner 2.6.4 is ROCK SOLID. No Hicupps, No Crashes...nothing.

I am currently using it on all 3 recent generations of ATI Cards (5/6/7xxx) all using CAT 12.6/SDK 2.6 and everything just seems to WORK.

I have also found that the AUTOTUNE feature works great. I only specify --shaders, -I (Intensity) and -g (threads) and the performance is better than if I mess with it.

Thanks again for all your hard work Con.

Awesome. Thanks a lot for the feedback bitlane 

[easyJoe]

2.6.4 still shows 2.6.3 in the display.

One of my binaries? Which one?

http://ck.kolivas.org/apps/cgminer/cgminer-2.6.4-x86_64-built.tar.bz2

Thanks. It looks like the code was up to date, but I built it before pulling the version number change on the laptop, sorry about that. Reuploaded packages.

Just download it. Still shows 2.6.3.

[Inaba]

2.6.4 crashing occasionally.  Latest one:

Code:

***glibcdetected***/opt/miners/cgminer/cgminer:double free or corruption(!prev):0x0a03dbe8***
=======Backtrace:=========
/lib/i686/cmov/libc.so.6(+0x6b381)[0xb7680381]
/lib/i686/cmov/libc.so.6(+0x6cbd8)[0xb7681bd8]
/lib/i686/cmov/libc.so.6(cfree+0x6d)[0xb7684cbd]
/usr/lib/torsocks/libtorsocks.so(+0x2930)[0xb783e930]
/usr/lib/torsocks/libtorsocks.so(connect+0x560)[0xb7840b50]
/usr/lib/libcurl.so.4(+0x3630a)[0xb77f230a]
/usr/lib/libcurl.so.4(Curl_connecthost+0x11a)[0xb77f282a]
/usr/lib/libcurl.so.4(Curl_connect+0x1c7)[0xb77de357]
/usr/lib/libcurl.so.4(+0x3090d)[0xb77ec90d]
/usr/lib/libcurl.so.4(curl_easy_perform+0x43)[0xb77eda13]
/opt/miners/cgminer/cgminer[0x8058f2d]
/opt/miners/cgminer/cgminer[0x8053c30]
/lib/i686/cmov/libpthread.so.0(+0x5955)[0xb781c955]
/lib/i686/cmov/libc.so.6(clone+0x5e)[0xb76e15ee]

[silverbox]

2.6.4 crashing occasionally. 

Ditto.  I switched back to 2.5.0. 

[ancow]

2.6.4 crashing occasionally.  Latest one:

Code:

***glibcdetected***/opt/miners/cgminer/cgminer:double free or corruption(!prev):0x0a03dbe8***
=======Backtrace:=========
/lib/i686/cmov/libc.so.6(+0x6b381)[0xb7680381]
/lib/i686/cmov/libc.so.6(+0x6cbd8)[0xb7681bd8]
/lib/i686/cmov/libc.so.6(cfree+0x6d)[0xb7684cbd]
/usr/lib/torsocks/libtorsocks.so(+0x2930)[0xb783e930]
/usr/lib/torsocks/libtorsocks.so(connect+0x560)[0xb7840b50]
/usr/lib/libcurl.so.4(+0x3630a)[0xb77f230a]
/usr/lib/libcurl.so.4(Curl_connecthost+0x11a)[0xb77f282a]
/usr/lib/libcurl.so.4(Curl_connect+0x1c7)[0xb77de357]
/usr/lib/libcurl.so.4(+0x3090d)[0xb77ec90d]
/usr/lib/libcurl.so.4(curl_easy_perform+0x43)[0xb77eda13]
/opt/miners/cgminer/cgminer[0x8058f2d]
/opt/miners/cgminer/cgminer[0x8053c30]
/lib/i686/cmov/libpthread.so.0(+0x5955)[0xb781c955]
/lib/i686/cmov/libc.so.6(clone+0x5e)[0xb76e15ee]

From that backtrace, it would seem the problem is something to do with TOR - do you have to use it? Could you try without and see whether the crash goes away?

[DiabloD3]

Neither of these two examples have the same, copied solution. They were both solved independently in similar ways as they are both soving the same problem. It's no suprise the timing is also similar as these issue are being found at the same time.

I don't know why you're defending his obvious theft. They do have the same copied solution, even if Kano made a poor effort to obfuscate it; while the first one might have only had one or two ways it could be fixed, this last one has numerous possible solutions (the most obvious being significantly better than the one I ended up taking except for making merging from cgminer harder if cgminer hadn't adopted it too); yet in both cases Kano used the exact same solution - only trivial/non-substantial changes to the syntax/names were made.

Neither of these bugs were in fact found in cgminer: the curlring issue was one I experienced myself personally, spent hours debugging, and finally identified a solution for; the only way Kano would have even known it existed was by reading BFGMiner's commit log; the JSON escaping issue only really affect BFGMiner because its compiled-in prefix on Windows uses backslashes.

Look, if you're going to keep this shit up, I and con BOTH will DMCA your existence itself (yes, we will send a DMCA to your mother's womb, the one you presumably slithered out of) if you even think about misrepresenting the copyright or the license of code contained within cgminer.

[DiabloD3]

Luke, let me remind you: I'm a mod. Reporting my own comment to me is not an effective way of protesting my relationship with your mother. I'm not sure why you're complaining, its purely business.

[crazyates]

Luke, let me remind you: I'm a mod. Reporting my own comment to me is not an effective way of protesting my relationship with your mother. I'm not sure why you're complaining, its purely business.

LOL

[CrazyGuy]

Is it possible to disable bitforce devices in the same way you can disable GPU devices? I like to run two instances of cgminer so my gpus and singles are seperated by worker. I am able to get around things by running cgminer with gpus disabled first then runing cgminer with nothing disabled. My second instance shows all of the singles as off, which is fine, but I would rather them be removed from the list.

[kano]

Is it possible to disable bitforce devices in the same way you can disable GPU devices? I like to run two instances of cgminer so my gpus and singles are seperated by worker. I am able to get around things by running cgminer with gpus disabled first then runing cgminer with nothing disabled. My second instance shows all of the singles as off, which is fine, but I would rather them be removed from the list.

-S noauto

[ralree]

Working great, 7970 with flashed Ghz edition vbios at 1220/700, Windows 8, on air cooling (fan @ 100%).  700+Mh/s.  Love it.  Going to watercool to get to 800+ this week.

[CrazyGuy]

Is it possible to disable bitforce devices in the same way you can disable GPU devices? I like to run two instances of cgminer so my gpus and singles are seperated by worker. I am able to get around things by running cgminer with gpus disabled first then runing cgminer with nothing disabled. My second instance shows all of the singles as off, which is fine, but I would rather them be removed from the list.

-S noauto

Ah, yes, thanks!