Author

Topic: OFFICIAL CGMINER mining software thread for linux/win/osx/mips/arm/r-pi 4.11.0 - page 682. (Read 5806015 times)

hero member
Activity: 535
Merit: 500
when using anything above intensity 9 on my 5970 w/ 11.12 drivers. The cpu usage is pegged... 9 and below cpu usage is 0%
Any reason why its higher on I 10+? thanks
This is the infamous cpu bug.
hero member
Activity: 896
Merit: 1000
Buy this account on March-2019. New Owner here!!
when using anything above intensity 9 on my 5970 w/ 11.12 drivers. The cpu usage is pegged... 9 and below cpu usage is 0%
Any reason why its higher on I 10+? thanks

just curious what mhash your getting with your 5970 using 11.12 ? I messed around with several different configurations over the last couple of days for mine and the best I came up with is 11.9 w/ cgminer 2.20 - I get a nice 820 mhash - woohoo!

btw: ckvolias do you know why Anubis web front end stopped working with the 2.2.0 upgrade?

thanks!
full member
Activity: 210
Merit: 100
when using anything above intensity 9 on my 5970 w/ 11.12 drivers. The cpu usage is pegged... 9 and below cpu usage is 0%
Any reason why its higher on I 10+? thanks
You are only supposed to use 10+ with the new 79xx cards. I'm positive Con mentioned that somewhere around here.
legendary
Activity: 2450
Merit: 1002
when using anything above intensity 9 on my 5970 w/ 11.12 drivers. The cpu usage is pegged... 9 and below cpu usage is 0%
Any reason why its higher on I 10+? thanks
donator
Activity: 919
Merit: 1000
Any system changes needed with Linux to move from 2.1.1 to 2.2.0?

I'm observing massive problems with the latest updates.

I kept the previous working binary as reference, but was not able to isolate the problem. First obvious thing is ADL fails to initialize:
Code:
[2012-01-29 19:30:32] Started cgminer 2.2.0
[2012-01-29 19:30:32] CL Platform vendor: Advanced Micro Devices, Inc.
[2012-01-29 19:30:32] CL Platform name: AMD Accelerated Parallel Processing
[2012-01-29 19:30:32] CL Platform version: OpenCL 1.1 AMD-APP-SDK-v2.4 (595.10)
[2012-01-29 19:30:32] ADL Initialisation Error!

The remaining initialization sequence looks ok, at least no obvious differences to what 2.1.1 was reporting:
Code:
[2012-01-29 19:30:35] Init GPU thread 0 GPU 0 virtual GPU 0
[2012-01-29 19:30:35] CL Platform vendor: Advanced Micro Devices, Inc.
[2012-01-29 19:30:35] CL Platform name: AMD Accelerated Parallel Processing
[2012-01-29 19:30:35] CL Platform version: OpenCL 1.1 AMD-APP-SDK-v2.4 (595.10)
[2012-01-29 19:30:35] List of devices:
[2012-01-29 19:30:35]   0       Cayman
[2012-01-29 19:30:35]   1       Cayman
[2012-01-29 19:30:35]   2       Cayman
[2012-01-29 19:30:35] Selected 0: Cayman
[2012-01-29 19:30:35] Preferred vector width reported 4
[2012-01-29 19:30:35] Max work group size reported 256
[2012-01-29 19:30:35] Long-polling activated for http://pit.deepbit.net:8332/listenChannel
[2012-01-29 19:30:35] Loaded binary image phatk110817Caymanbitalignv2w128long4.bin
[2012-01-29 19:30:35] Initialising kernel phatk110817.cl with BFI_INT, 2 vectors and worksize 128
[2012-01-29 19:30:35] initCl() finished. Found Cayman
[2012-01-29 19:30:35] Pushing ping to thread 1

...but the hash rates achieved are at about 30%
Code:
[2012-01-29 20:18:56] GPU0 (5s):131.0 (avg):123.5 Mh/s | A:53 R:0 HW:0 U:1.77/m I: 9
[2012-01-29 20:18:56] GPU1 (5s):103.2 (avg):121.8 Mh/s | A:55 R:0 HW:0 U:1.84/m I: 9
[2012-01-29 20:18:56] GPU2 (5s):95.6 (avg):121.2 Mh/s | A:43 R:0 HW:0 U:1.44/m I: 9


of what 2.1.1 does:
Code:
[2012-01-29 20:22:51] GPU0 (5s):363.3 (avg):370.0 Mh/s | A:5 R:0 HW:0 U:4.58/m | I: 9
[2012-01-29 20:22:51] GPU1 (5s):360.9 (avg):365.9 Mh/s | A:2 R:0 HW:0 U:3.61/m | I: 9
[2012-01-29 20:22:51] GPU2 (5s):372.4 (avg):350.6 Mh/s | A:4 R:0 HW:0 U:4.35/m | I: 9

As said, using the same APP and ADL version, same kernel. I also tried with deleting previous kernel binaries with the same result. I am using the current git HEAD for 2.2.0 (82af288e69).

Any ideas?
hero member
Activity: 535
Merit: 500
Ok, I'm running 2.1 SDK now, but most of gpu stats (like temp, clocks, voltage, etc)  isn't showing in gpu menu.  Huh However yesterday with 2.1.2 and 2.5 it worked fine. Using  2.2.0 now.

 UPD: oops, looks, like missing "[C]hange settings" option in 2.2.0.

That happens when cgminer doesn't detect ADL support.  For me it happens when using remote desktop.   The change in driver version is likely coincidental.  cgminer makes changes to cards and gets "stats" via ADL.

Are you accessing the rig remotely?  If not try a reboot and use cgminer before anything else.  I have seen AMD ADL driver? service? crash and that causes those options to "disapear".
No, I don't use remote desktop. ADL is working. Ckolivas has figured out the problem and made a temp fix. Read the thread a bit further.
newbie
Activity: 78
Merit: 0
hi

2 questions:

1) Any chance version 2.2.0 might be slower on single or dual (crosfire) 5830 ?

2) Any chance older cgminer versions restored not only fans, but clocks (or at least memory clock) of each card when pressing "q" ? i lower my memory to 300, but today i noticed it remains at 300 and never goes up. (Geeks3D GPU Caps Viewer detects max memory clock as 300 also).

EDIT: i just noticed that my linux box still running whatever drivers i installed on it when i built it, and SDK 2.4 still produces 300+ mhash with cgminer, meanwhile, my linux pcs are down to 280 or so. I did some reading and found out that the newer SDKs (2.6+) had that effect (starting from drivers 11.12). Even after fully removing 11.12 and installing 11.10, or even manually changing opencl (and 2 other DLLs) like its explained on these forums, i am unable to get back the old performance Sad

EDIT2:

After carefully reading your notes, i found out that the importat thing was to generate the (i guess kernel) binaries with the correct SDK. Since i had the ones from the correctly working SDK on CGMINER older versions folder, im fine now XD 2.2.0 is still slower than 2.1.1 tho.
donator
Activity: 1218
Merit: 1080
Gerald Davis
Ok, I'm running 2.1 SDK now, but most of gpu stats (like temp, clocks, voltage, etc)  isn't showing in gpu menu.  Huh However yesterday with 2.1.2 and 2.5 it worked fine. Using  2.2.0 now.

 UPD: oops, looks, like missing "[C]hange settings" option in 2.2.0.

That happens when cgminer doesn't detect ADL support.  For me it happens when using remote desktop.   The change in driver version is likely coincidental.  cgminer makes changes to cards and gets "stats" via ADL.

Are you accessing the rig remotely?  If not try a reboot and use cgminer before anything else.  I have seen AMD ADL driver? service? crash and that causes those options to "disapear".
legendary
Activity: 1876
Merit: 1000
I've got an issue that I hope someone can offer some help with.
I am unable to restart a GPU manually after it has been declared SICK. After changing the core voltage/clock to something (presumably) more stable, I can't restart the GPU.
If I try to enable it I am told that I must restart it first. If I try to restart it cgminer says "Thread x no longer exists":

Code:
Thread 0: 199.3 Mh/s Enabled ALIVE
Thread 2: 202.2 Mh/s Enabled ALIVE

GPU 1: 0.0 / 183.1 Mh/s | A:11360  R:85  HW:0  U:2.50/m  I:6
22.0 C  F: 32 9.532824E-130: 157 MHz  M: 300 Mhz  V: 1.000V  A: 0% P: 0%
Last initialised: [2012-01-29 09:52:15]
Intensity: 6
Thread 1: 0.0 Mh/s Disabled DEAD reported in [2012-01-29 13:55:01]
Thread 3: 0.0 Mh/s Disabled DEAD reported in [2012-01-29 13:55:02]

[E]nable [D]isable [I]ntensity [R]estart GPU [C]hange settings
Or press any other key to continue
Select GPU to attempt to restart:
1
Attempting to restart threads of GPU 1
GPU 0: 401.5 / 401.0 Mh/s | A:25054  R:154  HW:0  U:5.52/m  I:6[2012-01-29 15:31:20] Thread 1 no longer exists
[2012-01-29 15:31:20] Thread 3 no longer exists
63.5 C  F: 61% (2872 RPM)  E: 875 MHz  M: 250 Mhz  V: 1.072V  A: 99% P: 0%
Last initialised: [2012-01-26 11:51:18]
Intensity: 6
Thread 0: 199.4 Mh/s Enabled ALIVE
Thread 2: 202.4 Mh/s Enabled ALIVE

GPU 1 is the sick GPU. GPU 0 is fine.

If I restart cgminer, GPU 0 starts up without problems.

count your blessings...  once I have a dead one, i have to reboot to get it back.
legendary
Activity: 980
Merit: 1008
I've got an issue that I hope someone can offer some help with.
I am unable to restart a GPU manually after it has been declared SICK. After changing the core voltage/clock to something (presumably) more stable, I can't restart the GPU.
If I try to enable it I am told that I must restart it first. If I try to restart it cgminer says "Thread x no longer exists":

Code:
Thread 0: 199.3 Mh/s Enabled ALIVE
Thread 2: 202.2 Mh/s Enabled ALIVE

GPU 1: 0.0 / 183.1 Mh/s | A:11360  R:85  HW:0  U:2.50/m  I:6
22.0 C  F: 32 9.532824E-130: 157 MHz  M: 300 Mhz  V: 1.000V  A: 0% P: 0%
Last initialised: [2012-01-29 09:52:15]
Intensity: 6
Thread 1: 0.0 Mh/s Disabled DEAD reported in [2012-01-29 13:55:01]
Thread 3: 0.0 Mh/s Disabled DEAD reported in [2012-01-29 13:55:02]

[E]nable [D]isable [I]ntensity [R]estart GPU [C]hange settings
Or press any other key to continue
Select GPU to attempt to restart:
1
Attempting to restart threads of GPU 1
GPU 0: 401.5 / 401.0 Mh/s | A:25054  R:154  HW:0  U:5.52/m  I:6[2012-01-29 15:31:20] Thread 1 no longer exists
[2012-01-29 15:31:20] Thread 3 no longer exists
63.5 C  F: 61% (2872 RPM)  E: 875 MHz  M: 250 Mhz  V: 1.072V  A: 99% P: 0%
Last initialised: [2012-01-26 11:51:18]
Intensity: 6
Thread 0: 199.4 Mh/s Enabled ALIVE
Thread 2: 202.4 Mh/s Enabled ALIVE

GPU 1 is the sick GPU. GPU 0 is fine.

If I restart cgminer, GPU 0 starts up without problems.
hero member
Activity: 535
Merit: 500
Do I need to enable engine clock autotune for "temp-cutoff" activation? Or it's active by default?
You need to enable --auto-gpu .However if you dont give it a range for engine clock speeds, it won't adjust any clock speeds, but it will still cut off the device if it hits the --temp-cutoff threshold.
Great, thanks!
legendary
Activity: 2576
Merit: 1186
6,Reporting temperature/safety measurements can allows cgminer to shut off the FPGA if it gets too hot.

6 needs additional hardware, but we notice that the normal error rate is less than 0.1%, if the error rate goes high, that means something goes wrong.
But by then, it's too late (and the FPGA destroyed), right?

is there any reference docs about "noncerange extension" feature?  Huh
https://en.bitcoin.it/wiki/Getwork#noncerange

It occurs to me that I'm not really aware of how FPGAs handle SHA256 internally; if it doesn't use the normal uint32-based algorithm, it might not be practical to implement.
-ck
legendary
Activity: 4088
Merit: 1631
Ruu \o/
Do I need to enable engine clock autotune for "temp-cutoff" activation? Or it's active by default?
You need to enable --auto-gpu .However if you dont give it a range for engine clock speeds, it won't adjust any clock speeds, but it will still cut off the device if it hits the --temp-cutoff threshold.
hero member
Activity: 535
Merit: 500
 Do I need to enable engine clock autotune for "temp-cutoff" activation? Or it's active by default?
-ck
legendary
Activity: 4088
Merit: 1631
Ruu \o/
Latest git tends to die a miserable death, Con.
The previous build i reported testing some 12 hours ago works just fine.
Has been fixed. Dodgy cleanups.
hero member
Activity: 535
Merit: 500
But why 2.1.2 don't have such issue then?
It stopped after it found the first device. 2.2.0 finds all the devices and then reorders them correctly according to their PCI bus position. As I said you were lucky it wasn't giving you temps and fanspeed for the 4250 instead because it happened to put the 4250 second in the driver order even though you physically have it first in the slot - You can see it has a higher adapterindex in that list despite having a lower busnumber.

As for a solution to this problem that is correct as opposed to "lucky"... I'm thinking, but I have no ideas apart from explicitly telling cgminer at this stage.
This won't be the correct fix, but this exe should work around the problem for you for now (put into 2.2.0 directory):
http://ck.kolivas.org/apps/cgminer/temp/cgminer.exe
Yeah, it's working, even with 2.1 sdk. Thanks!
full member
Activity: 210
Merit: 100
...Can we get a re-post of those facepalms please?
Ask and you shall receive.



Code:
root@haerdalis::/home/jake# /opt/bcm/cgminer-testing/cgminer                    [2012-01-29 13:48:56] Started cgminer 2.2.0
WARNING: All config files need .conf: /etc/modprobe.d/00local, it will be ignored in a future release.
FATAL: Module nvidia not found.
*** glibc detected *** /opt/bcm/cgminer-testing: realloc(): invalid old size: 0xbfa81ce8 ***
======= Backtrace: =========
/lib/i686/cmov/libc.so.6(+0x6b381)[0xb75b7381]
/lib/i686/cmov/libc.so.6(+0x70aad)[0xb75bcaad]
/lib/i686/cmov/libc.so.6(realloc+0xdd)[0xb75bcd8d]
/opt/bcm/cgminer-testing[0x80590ad]
/usr/lib/libcurl-gnutls.so.4(Curl_client_write+0x198)[0xb770e6e8]
/usr/lib/libcurl-gnutls.so.4(Curl_readwrite+0x12c0)[0xb7721b00]
/usr/lib/libcurl-gnutls.so.4(+0x29f1b)[0xb7721f1b]
/usr/lib/libcurl-gnutls.so.4(curl_easy_perform+0x43)[0xb7722d93]
/opt/bcm/cgminer-testing[0x805a1db]
/opt/bcm/cgminer-testing[0x80509d2]
/opt/bcm/cgminer-testing[0x80566a8]
/lib/i686/cmov/libc.so.6(__libc_start_main+0xe6)[0xb7562ca6]
/opt/bcm/cgminer-testing[0x804ad01]
======= Memory map: ========
...
Latest git tends to die a miserable death, Con.
The previous build i reported testing some 12 hours ago works just fine.
-ck
legendary
Activity: 4088
Merit: 1631
Ruu \o/
But why 2.1.2 don't have such issue then?
It stopped after it found the first device. 2.2.0 finds all the devices and then reorders them correctly according to their PCI bus position. As I said you were lucky it wasn't giving you temps and fanspeed for the 4250 instead because it happened to put the 4250 second in the driver order even though you physically have it first in the slot - You can see it has a higher adapterindex in that list despite having a lower busnumber.

As for a solution to this problem that is correct as opposed to "lucky"... I'm thinking, but I have no ideas apart from explicitly telling cgminer at this stage.
This won't be the correct fix, but this exe should work around the problem for you for now (put into 2.2.0 directory):
http://ck.kolivas.org/apps/cgminer/temp/cgminer.exe
hero member
Activity: 807
Merit: 500
Hmm thanks for the link, but that doesn't even appear in my code anywhere. Presumably that's ufa they're talking about. Meanwhile cgminer mines fine on my nvidia laptop so  Huh
I guess I'm a bit late to be responding now regardless, but yes that code was for (and posted by) ufasoft.  I quoted what I did because the person who they were responding to said cgminer stopped working in their environment with the same nvidia driver version leading me to believe it could possibly be a similar cause, so I thought I'd bring it up just in case.
-ck
legendary
Activity: 4088
Merit: 1631
Ruu \o/
Question: Can your other device (4250) mine?
I guess you don't need to answer that, I checked.

So this is the problem. One device that can mine and one that can't. There's already nothing to correlate openCL devices with devices that appear in the ATI display library apart from "how many there are" and "in what order they appear". Put a non-opencl device in there (like the 4250) and it's a mess... GAH. Nothing short of some kind of massive whitelist will fix this. If it worked previously you were lucky it wasn't monitoring the 1st card instead.

EDIT: This recurring problem of having no way of correlating opencl and ATI devices is getting seriously long in the tooth. Can we get a re-post of those facepalms please?
But why 2.1.2 don't have such issue then?
It stopped after it found the first device. 2.2.0 finds all the devices and then reorders them correctly according to their PCI bus position. As I said you were lucky it wasn't giving you temps and fanspeed for the 4250 instead because it happened to put the 4250 second in the driver order even though you physically have it first in the slot - You can see it has a higher adapterindex in that list despite having a lower busnumber.

As for a solution to this problem that is correct as opposed to "lucky"... I'm thinking, but I have no ideas apart from explicitly telling cgminer at this stage.
Jump to: