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

[cablepair]

Managed to get my 447 Mhash/s with my 5870s !

Will post updated stats and details soon. Testing for stability for now.

Hopefully some can appreciate my work and give me some love over at

1Ahbp2y64JFSQu1D6N5rkhw4PtaY9oXtE3

Thanks !

as I said earlier I would gladly give a donation for detailed benchmarking information, specifically what SDKs and which addon flags like vectors and work size, if all you did was push the core clock way up to 990 or something that is not going to be stable for long.
 
waiting to see what you got...

[QuantumFoam]

I just want to say in regards to 5870

I consider myself a bit of an expert when it comes to mining with theses cards as I have been doing so for almost a year now and have about 20 of them, they are excellent mining cards very efficient and stable , I have a 5870 that's been mining 11 months straight without any Problems no joke

In any case in regards to drivers and speed I have tested them all fairly extensively and ccc 11.9 with the Sdk 2.5.xx it comes with gives 430-445 mhash depending on the card with 950/180 clocks and the latest cgminer with stock kernel and 9 intensity

If any can prove me wrong (or right) I would love to see it, and would gladly give donation to anyone that provides a detailed benchmark

I'm running 5870s in my mining machine and gave your 180 memory clock setting a try. I found the difference between 300 and 180 to be about a 5mh loss per card @ 180, and negligible difference in power consumption as measured by kill a watt. It could be that you're running some other settings in cgminer that I'm not? I have worksize set to 256, otherwise all defaults. Engine clocks are 950. This is v2.2.1 on a xubuntu 11.04 machine.

[kano]

I just want to say in regards to 5870

I consider myself a bit of an expert when it comes to mining with theses cards as I have been doing so for almost a year now and have about 20 of them, they are excellent mining cards very efficient and stable , I have a 5870 that's been mining 11 months straight without any Problems no joke

In any case in regards to drivers and speed I have tested them all fairly extensively and ccc 11.9 with the Sdk 2.5.xx it comes with gives 430-445 mhash depending on the card with 950/180 clocks and the latest cgminer with stock kernel and 9 intensity

If any can prove me wrong (or right) I would love to see it, and would gladly give donation to anyone that provides a detailed benchmark

I'm running 5870s in my mining machine and gave your 180 memory clock setting a try. I found the difference between 300 and 180 to be about a 5mh loss per card @ 180, and negligible difference in power consumption as measured by kill a watt. It could be that you're running some other settings in cgminer that I'm not? I have worksize set to 256, otherwise all defaults. Engine clocks are 950. This is v2.2.1 on a xubuntu 11.04 machine.

There are more options to it that just cgminer version and OS version, that can make a difference.
The two of you would also have to compare the output of cgminer -n
(and also be sure that you generated the *.bin with the same OpenCL versions - at the time you generated it)
The name of the *.bin will tell you some of the options you are using also.

[cablepair]

I just want to say in regards to 5870

I consider myself a bit of an expert when it comes to mining with theses cards as I have been doing so for almost a year now and have about 20 of them, they are excellent mining cards very efficient and stable , I have a 5870 that's been mining 11 months straight without any Problems no joke

In any case in regards to drivers and speed I have tested them all fairly extensively and ccc 11.9 with the Sdk 2.5.xx it comes with gives 430-445 mhash depending on the card with 950/180 clocks and the latest cgminer with stock kernel and 9 intensity

If any can prove me wrong (or right) I would love to see it, and would gladly give donation to anyone that provides a detailed benchmark

I'm running 5870s in my mining machine and gave your 180 memory clock setting a try. I found the difference between 300 and 180 to be about a 5mh loss per card @ 180, and negligible difference in power consumption as measured by kill a watt. It could be that you're running some other settings in cgminer that I'm not? I have worksize set to 256, otherwise all defaults. Engine clocks are 950. This is v2.2.1 on a xubuntu 11.04 machine.

what SDK are you using and what speeds are you getting with 950/300 ? Also I am using default worksize and vectors, I have tested and never seen any increase only decrease and stay the same by changing what is default for worksize and vectors with cgminer

let me show you a snap shot from my anubis console of one of the 4 rigs I have with 5870s each.



this rig has 3 x 2 gb models and 2 x 1 gb model thats why your seeing some variance in the fan speeds and voltages
yes the amount of electricity and heat difference between 300/180 may be negligible however IMO when your running 20 gpus every bit counts, even if you save 1 kwhr a day it adds up over time and I see no speed increase what so ever by changing mem clocks from 180 to 300

[kano]

Oh yeah that reminds me
Anyone who takes notice of that last column (Intensity) and ever changes them to different values per GPU - I posted a fix (ckolivas pull request) for that yesterday since it was always showing only the first GPU value for all GPUs
I'd guess no one ever changes it like that so no one noticed it (me neither of course)
(I was testing the new modified miner.php that's in there as well to change different values using the recent options that were added when I found that one)

[QuantumFoam]

what SDK are you using and what speeds are you getting with 950/300 ?

v2.4 sdk. That's probably the issue, thought I had 2.1 installed. I'll see if I can remove 2.4 and install 2.1, hopefully that will make a difference. Currently getting ~430mh at 950/300 setting, that drops to 425 when going to 950/180.

[-ck]

what SDK are you using and what speeds are you getting with 950/300 ?

v2.4 sdk. That's probably the issue, thought I had 2.1 installed. I'll see if I can remove 2.4 and install 2.1, hopefully that will make a difference. Currently getting ~430mh at 950/300 setting, that drops to 425 when going to 950/180.

Allegedly 2.4 and 2.5 do need more memory speed for some strange reason.

[QuantumFoam]

Allegedly 2.4 and 2.5 do need more memory speed for some strange reason.

BTW, any ideas where I'd start looking to see why cgminer stops detecting fan rpm speeds/gpu temps after it has been previously run? Only way to get it to detect them again is to reboot the machine.

[-ck]

Allegedly 2.4 and 2.5 do need more memory speed for some strange reason.

BTW, any ideas where I'd start looking to see why cgminer stops detecting fan rpm speeds/gpu temps after it has been previously run? Only way to get it to detect them again is to reboot the machine.

It's at a driver level that this is failing. I have no idea why it would change its mind to not support it on future runs. When it fails, if you start it with debugging enabled you will get what the ADL error is, but that still won't give an answer as to how to fix it. Something like "ADL Initialisation Error!" any error with adl in it.

[QuantumFoam]

[2012-02-04 22:53:02] Pushing ping to longpoll thread
[2012-02-04 22:53:02] ADL Initialisation Error!
[2012-02-04 22:53:02] Pushing ping to thread 0
[2012-02-04 22:53:02] Init GPU thread 0
[2012-02-04 22:53:02] List of devices:
[2012-02-04 22:53:02]   0       Cypress
[2012-02-04 22:53:02]   1       Cypress
[2012-02-04 22:53:02]   2       Cypress
[2012-02-04 22:53:02]   3       Juniper
[2012-02-04 22:53:02]   4       Cypress
[2012-02-04 22:53:02] Selected 0: Cypress
[2012-02-04 22:53:02] Preferred vector width reported 4
[2012-02-04 22:53:02] Max work group size reported 256


Only place it has a line with ADL.

[-ck]

[2012-02-04 22:53:02] Pushing ping to longpoll thread
[2012-02-04 22:53:02] ADL Initialisation Error!
[2012-02-04 22:53:02] Pushing ping to thread 0
[2012-02-04 22:53:02] Init GPU thread 0
[2012-02-04 22:53:02] List of devices:
[2012-02-04 22:53:02]   0       Cypress
[2012-02-04 22:53:02]   1       Cypress
[2012-02-04 22:53:02]   2       Cypress
[2012-02-04 22:53:02]   3       Juniper
[2012-02-04 22:53:02]   4       Cypress
[2012-02-04 22:53:02] Selected 0: Cypress
[2012-02-04 22:53:02] Preferred vector width reported 4
[2012-02-04 22:53:02] Max work group size reported 256


Only place it has a line with ADL.

Code:

	result = ADL_Main_Control_Create (ADL_Main_Memory_Alloc, 1);
	if (result != ADL_OK) {
		applog(LOG_INFO, "ADL Initialisation Error! Error %d!", result);
		return ;
	}

The newer version should also show what the error is, but the main initialisation is failing. What version of cgminer are you running?

It will be one of the following.

Code:

#define 	ADL_OK_WAIT   4
 	All OK, but need to wait.
#define 	ADL_OK_RESTART   3
 	All OK, but need restart.
#define 	ADL_OK_MODE_CHANGE   2
 	All OK but need mode change.
#define 	ADL_OK_WARNING   1
 	All OK, but with warning.
#define 	ADL_OK   0
 	ADL function completed successfully.
#define 	ADL_ERR   -1
 	Generic Error. Most likely one or more of the Escape calls to the driver failed!
#define 	ADL_ERR_NOT_INIT   -2
 	ADL not initialized.
#define 	ADL_ERR_INVALID_PARAM   -3
 	One of the parameter passed is invalid.
#define 	ADL_ERR_INVALID_PARAM_SIZE   -4
 	One of the parameter size is invalid.
#define 	ADL_ERR_INVALID_ADL_IDX   -5
 	Invalid ADL index passed.
#define 	ADL_ERR_INVALID_CONTROLLER_IDX   -6
 	Invalid controller index passed.
#define 	ADL_ERR_INVALID_DIPLAY_IDX   -7
 	Invalid display index passed.
#define 	ADL_ERR_NOT_SUPPORTED   -8
 	Function not supported by the driver.
#define 	ADL_ERR_NULL_POINTER   -9
 	Null Pointer error.
#define 	ADL_ERR_DISABLED_ADAPTER   -10
 	Call can't be made due to disabled adapter.
#define 	ADL_ERR_INVALID_CALLBACK   -11
 	Invalid Callback.
#define 	ADL_ERR_RESOURCE_CONFLICT   -12
 	Display Resource conflict.

[QuantumFoam]

Oops, I might have tried that with 2.1.2 instead of 2.2.1. I have both on the machine currently. I'll try rerunning with 2.2.1

EDIT: Ok I am an idiot, this was caused by my failure to type
export DISPLAY=:0

in the shell each time before I ran it. I mistakenly thought that command applied to the entire login session. Hope that info helps anyone else running into the same problem. Carry on.

[gnar1ta$]

Using 2.2.1 on a mixed card rig

Code:

[2012-02-05 00:01:17] CL Platform vendor: Advanced Micro Devices, Inc.
[2012-02-05 00:01:17] CL Platform name: AMD Accelerated Parallel Processing
[2012-02-05 00:01:17] CL Platform version: OpenCL 1.1 AMD-APP-SDK-v2.4 (595.10)
[2012-02-05 00:01:17] GPU 0 AMD Radeon HD 6800 Series hardware monitoring enabled
[2012-02-05 00:01:17] GPU 1 ATI Radeon HD 5900 Series hardware monitoring enabled
[2012-02-05 00:01:17] GPU 2 ATI Radeon HD 5900 Series hardware monitoring enabled
[2012-02-05 00:01:17] Failed to ADL_Overdrive5_FanSpeedInfo_Get
[2012-02-05 00:01:17] GPU 3 AMD Radeon HD 6800 Series hardware monitoring enabled
[2012-02-05 00:01:17] Dual GPUs detected: 2 and 1
[2012-02-05 00:01:17] 4 GPU devices detected

And the second core of the 5970 keeps showing OFF after 12-18 hrs running with this error

Code:

[2012-02-04 03:37:42] Thread 2 idle for more than 60 seconds, GPU 2 declared SICK!
[2012-02-04 03:37:42] Attempting to restart GPU
[2012-02-04 03:37:42] Thread 4 still exists, killing it off
[2012-02-04 03:37:42] Thread 5 still exists, killing it off
[2012-02-04 03:37:42] Thread 4 restarted
[2012-02-04 03:37:43] Thread 5 restarted
[2012-02-04 03:37:44] Thread 4 being disabled
[2012-02-04 03:37:44] Thread 5 being disabled

I first thought it was clocks, but it has been running for 5 months at the same clocks.  I lowered them a few times anyway but still the same error.  If I go back to 2.1.2 it runs fine, 24 + hours now. 

[-ck]

Oops, I might have tried that with 2.1.2 instead of 2.2.1. I have both on the machine currently. I'll try rerunning with 2.2.1

EDIT: Ok I am an idiot, this was caused by my failure to type
export DISPLAY=:0

in the shell each time before I ran it. I mistakenly thought that command applied to the entire login session. Hope that info helps anyone else running into the same problem. Carry on.

You know I was actually about to say that's what usually causes  it, but I doubted it would happen between runs. Seems I was wrong 

[-ck]

Using 2.2.1 on a mixed card rig

Code:

[2012-02-05 00:01:17] CL Platform vendor: Advanced Micro Devices, Inc.
[2012-02-05 00:01:17] CL Platform name: AMD Accelerated Parallel Processing
[2012-02-05 00:01:17] CL Platform version: OpenCL 1.1 AMD-APP-SDK-v2.4 (595.10)
[2012-02-05 00:01:17] GPU 0 AMD Radeon HD 6800 Series hardware monitoring enabled
[2012-02-05 00:01:17] GPU 1 ATI Radeon HD 5900 Series hardware monitoring enabled
[2012-02-05 00:01:17] GPU 2 ATI Radeon HD 5900 Series hardware monitoring enabled
[2012-02-05 00:01:17] Failed to ADL_Overdrive5_FanSpeedInfo_Get
[2012-02-05 00:01:17] GPU 3 AMD Radeon HD 6800 Series hardware monitoring enabled
[2012-02-05 00:01:17] Dual GPUs detected: 2 and 1
[2012-02-05 00:01:17] 4 GPU devices detected

And the second core of the 5970 keeps showing OFF after 12-18 hrs running with this error

Code:

[2012-02-04 03:37:42] Thread 2 idle for more than 60 seconds, GPU 2 declared SICK!
[2012-02-04 03:37:42] Attempting to restart GPU
[2012-02-04 03:37:42] Thread 4 still exists, killing it off
[2012-02-04 03:37:42] Thread 5 still exists, killing it off
[2012-02-04 03:37:42] Thread 4 restarted
[2012-02-04 03:37:43] Thread 5 restarted
[2012-02-04 03:37:44] Thread 4 being disabled
[2012-02-04 03:37:44] Thread 5 being disabled

I first thought it was clocks, but it has been running for 5 months at the same clocks.  I lowered them a few times anyway but still the same error.  If I go back to 2.1.2 it runs fine, 24 + hours now. 

Well 2.2.1 has the dual GPU linkage going on which 2.1.2 does not. Bear in mind the auto fanspeed control now looks at the temps of both GPUs, but overall this should run things cooler rather than hotter so unless you have some specific setup, or maybe have autofan off on one of the devices or different clock settings... Also, have you checked it's actually detecting the right card in the right device and linking the right devices?

[gnar1ta$]

Well 2.2.1 has the dual GPU linkage going on which 2.1.2 does not. Bear in mind the auto fanspeed control now looks at the temps of both GPUs, but overall this should run things cooler rather than hotter so unless you have some specific setup, or maybe have autofan off on one of the devices or different clock settings... Also, have you checked it's actually detecting the right card in the right device and linking the right devices?

I use these flags

Code:

-I 8 --gpu-engine 950,850,830,950 --gpu-memclock 850,180,180,850  --auto-fan --temp-target 75 --temp-overheat 82

How do I check it's detecting right?  It lists in the same order as 2.1.2. --gpu-reorder worth checking?

[-ck]

I use these flags

Code:

-I 8 --gpu-engine 950,850,830,950 --gpu-memclock 850,180,180,850  --auto-fan --temp-target 75 --temp-overheat 82

How do I check it's detecting right?  It lists in the same order as 2.1.2. --gpu-reorder worth checking?

I can't see anything wrong with that. Without --gpu-reorder it uses the same order as 2.1.2 would have detected. I guess you'd know pretty quickly if it was setting the wrong speed on the wrong device and...

[2012-02-05 00:01:17] GPU 0 AMD Radeon HD 6800 Series hardware monitoring enabled
[2012-02-05 00:01:17] GPU 1 ATI Radeon HD 5900 Series hardware monitoring enabled
[2012-02-05 00:01:17] GPU 2 ATI Radeon HD 5900 Series hardware monitoring enabled
[2012-02-05 00:01:17] GPU 3 AMD Radeon HD 6800 Series hardware monitoring enabled

looks pretty convincing. So I'm at a loss for why it should be any worse.

Note also that you are having it being disabled and reading OFF. There is only one place in the code where cgminer does this itself - when it hits the thermal cutoff limit. Now it is possible there is some code convolution issue going on somehow that makes the fan not rise when one of the GPUs is being restarted. Note that your bug report shows thread 2 being idle (which would be GPU 1) and then it proceeds to disable threads 4 and 5 (being GPU 2)... Interesting... I'll audit the code, but perhaps try it without auto-fan on, setting what you know to be a safe static fan speed and see if the problem persists.

[mmortal03]

Also, on a related note, maybe your processor can get 8MH/s and the 8MH/s extra isn't even coming from the video card.  If that is the case, cutting the CPU frequency in half would lower it to a 4 MH/s gain at a still much more expensive MH/W (although realistically, even if it is the GPU getting that, running the CPU 100% at half frequency will probably still draw enough power to make it a net loss).

Are you theorizing that my CPU is explicitly doing some of the work, as if I was CPU mining, or do you mean something more nuanced than that? I didn't think my CPU would be used for explicit mining unless I had it specifically enabled to do so. I'm carrying out the U experiments that you guys suggested as we speak, btw.

[gnar1ta$]

That rig runs fairly cool so I'll try overnight without auto-fan.  Here is the rest of the errors from 2.2.1 sessions - I just lowered clock speeds for the different sessions.

Code:

[2012-02-02 18:22:19] Thread 1 idle for more than 60 seconds, GPU 1 declared SICK!
[2012-02-02 18:22:19] Attempting to restart GPU
[2012-02-02 18:22:19] Thread 2 still exists, killing it off
[2012-02-02 18:22:19] Thread 3 still exists, killing it off
[2012-02-02 18:22:19] Thread 2 restarted
[2012-02-02 18:22:20] Thread 3 restarted
[2012-02-02 18:22:21] Accepted 00000000.40372a1e.d75ae9ba GPU 0 thread 0 pool 0
[2012-02-02 18:22:21] Thread 2 being disabled
[2012-02-02 18:22:21] Thread 3 being disabled

[2012-02-02 17:58:17] Thread 3 idle for more than 60 seconds, GPU 3 declared SICK!
[2012-02-02 17:58:17] Attempting to restart GPU
[2012-02-02 17:58:17] Thread 6 still exists, killing it off
[2012-02-02 17:58:17] Thread 7 still exists, killing it off
[2012-02-02 17:58:18] Thread 6 restarted
[2012-02-02 17:58:19] Thread 7 restarted

[2012-02-03 21:19:13] Thread 4 still exists, killing it off
[2012-02-03 21:19:13] Thread 5 still exists, killing it off
[2012-02-03 21:19:14] Thread 4 restarted
[2012-02-03 21:19:14] Accepted 00000000.0cd03974.6e1c40e3 GPU 0 thread 1 pool 0
[2012-02-03 21:19:14] Thread 5 restarted
[2012-02-03 21:19:15] Thread 4 being disabled
[2012-02-03 21:19:16] Thread 5 being disabled

[2012-02-03 20:47:14] Thread 2 idle for more than 60 seconds, GPU 2 declared SICK!
[2012-02-03 20:47:14] Attempting to restart GPU
[2012-02-03 20:47:14] Thread 4 still exists, killing it off
[2012-02-03 20:47:14] Thread 5 still exists, killing it off
[2012-02-03 20:47:14] Thread 4 restarted
[2012-02-03 20:47:15] Accepted 00000000.b422fb13.d18b69ee GPU 1 thread 3 pool 0
[2012-02-03 20:47:15] Thread 5 restarted
[2012-02-03 20:47:15] Accepted 00000000.8db73352.22b376e6 GPU 0 thread 1 pool 0
[2012-02-03 20:47:16] Thread 4 being disabled
[2012-02-03 20:47:16] Thread 5 being disabled

[2012-02-04 17:29:18] Thread 2 idle for more than 60 seconds, GPU 2 declared SICK!
[2012-02-04 17:29:18] Attempting to restart GPU
[2012-02-04 17:29:18] Thread 4 still exists, killing it off
[2012-02-04 17:29:18] Thread 5 still exists, killing it off
[2012-02-04 17:29:19] Thread 4 restarted
[2012-02-04 17:29:19] Thread 5 restarted
[2012-02-04 17:29:20] Thread 4 being disabled
[2012-02-04 17:29:20] Thread 5 being disabled

[-ck]

The pool where donations were going was hacked and I'm considering moving all my shares to p2pool as well now. I have grave concerns about centralising work to pools and increasingly see p2pool - or something like it - as the solution for bitcoin's future strength, going back to its decentralised nature as its strength. This means I won't realistically have a way of accepting small hashrate contributions donations with --donation that I can reasonably support. So after much angst I have decided that I will be deprecating the donations feature in upcoming versions and go back to the previous donation model of as-and-when you feel like it.

I thank those who have used the --donation feature greatly till now. It averaged around 400Mh/s over that time and at least kept me "mining" while my own mining rig was dead for over a month.

I recommend people disable --donation now and restart their miners for I don't know what will happen to hashes going to the pool during this instability (it is going offline for 24+ hours likely).