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

[Diapolo]

Not sure what's going on, but when I attempt to start cgminer, my display driver immediately crashes, and all I get from cgminer, is this:

Code:

[2012-02-26 09:29:29] LONGPOLL requested work restart, waiting on fresh work
[2012-02-26 09:29:29] Failed to tq_push in queue_request
[2012-02-26 09:29:39] LONGPOLL requested work restart, waiting on fresh work
[2012-02-26 09:29:39] Failed to tq_push in queue_request

A search on "tq_push in queue_request" turned up nothing useful.

When I exit cgminer, I get this:

Code:

[2012-02-26 09:45:59]
Summary of runtime statistics:

[2012-02-26 09:45:59] Started at [2012-02-26 09:12:24]
[2012-02-26 09:45:59] Pool: http://127.0.0.1:9332
[2012-02-26 09:45:59] Runtime: 0 hrs : 0 mins : 0 secs
[2012-02-26 09:45:59] Average hashrate: 672.8 Megahash/s
[2012-02-26 09:45:59] Solved blocks: 0
[2012-02-26 09:45:59] Queued work requests: 2254
[2012-02-26 09:45:59] Share submissions: 0
[2012-02-26 09:45:59] Accepted shares: 0
[2012-02-26 09:45:59] Rejected shares: 0
[2012-02-26 09:45:59] Hardware errors: 0
[2012-02-26 09:45:59] Efficiency (accepted / queued): 0%
[2012-02-26 09:45:59] Utility (accepted shares / min): 0.00/min

[2012-02-26 09:45:59] Discarded work due to new blocks: 204
[2012-02-26 09:45:59] Stale submissions discarded due to new blocks: 0
[2012-02-26 09:45:59] Unable to get work from server occasions: 95
[2012-02-26 09:45:59] Work items generated locally: 24762
[2012-02-26 09:45:59] Submitting work remotely delay occasions: 0
[2012-02-26 09:45:59] New blocks detected on network: 6

The things that concern me, is that the runtime at 0 hours, 0 mins, 0 secs, and the hashrate of 672.8.  If, after 33 minutes, the runtime is 0, how is there /any/ hashrate at all?  How is is 672, when pcolbm reports only 26.  Is cgminer really 25x faster?

Any suggestions on what to look at?

Asus M5A99X MB
AMD FX-4100
8GB RAM
Nvidia 8800GT, driver version 285.62
Windows 7 x64 SP1

I'm still using an old video card because it did what I needed it to do when I upgraded everything else.  I found Bitcoin just a couple weeks later.  I may upgrade to something else, I may not... haven't decided yet.

Thanks,

-Prayer

You could simply try another kernel, as there are 4 of them . Have you installed the latest NVIDIA drivers?

Add one of this switches to your command line:

-k diablo
-k diakgcn
-k poclbm
-k phatk

Dia

[1713509038]

1713509038

[rjk]

I would love to see an expanded version of this that goes down to 100 (or lower) memclock using latest version of cgminer, latest drivers, and SDK 2.1.

Here is some tests

You sir, are full of awesome. Will we be able to see the other test results soon?
Do you have a 5870 to test with?

[spiccioli]

...
So either my CPU is too weak to handle 5 GPUs or cpu usage can be lower, but it does not depend on catalyst/amd sdk.

BTW I've tested catalyst 11.12 (11.11 gives problems installing) and 12.1 on xubuntu 11.10 64bit this morning with more or less same levels of CPU usage.

spiccioli

While it's running at 60% CPU ...
cat /proc/cpuinfo

kano,

here it is, cgminer alone around 40%

Code:

processor	: 0
vendor_id	: AuthenticAMD
cpu family	: 16
model		: 6
model name	: AMD Sempron(tm) 145 Processor
stepping	: 3
cpu MHz		: 800.000
cache size	: 1024 KB
fpu		: yes
fpu_exception	: yes
cpuid level	: 5
wp		: yes
flags		: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm 3dnowext 3dnow constant_tsc up rep_good nopl nonstop_tsc extd_apicid pni monitor cx16 popcnt lahf_lm svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt nodeid_msr npt lbrv svm_lock nrip_save
bogomips	: 5623.63
TLB size	: 1024 4K pages
clflush size	: 64
cache_alignment	: 64
address sizes	: 48 bits physical, 48 bits virtual
power management: ts ttp tm stc 100mhzsteps hwpstate

Note it says CPU MHz 800.00 so I think I've being chasing the wrong problem for two days, if CPU is at 2.8Ghz cgminer uses around 10% of CPU

Thanks for your help!

spiccioli.

[Prayer]

You could simply try another kernel, as there are 4 of them . Have you installed the latest NVIDIA drivers?

Add one of this switches to your command line:

-k diablo
-k diakgcn
-k poclbm
-k phatk

Dia

Thanks for the tip.  Drivers are up to date, and I tried all 4 kernels.  On all 4 kernels, the display driver still crashes immediately.

Code:

cgminer -o http://localhost:9332 -O user:pass -d 0 -k (kernel)

If I run the poclbm executable from the guiminer package, it seems to work just fine.  GPU is pegged and things appear to be running at ~25M.  Guess I'll just do this for now and wait a new release of cgminer.

Is there anything I can do to actually help figure out what's happening?

Thanks,

-Prayer

[DeathAndTaxes]

You could simply try another kernel, as there are 4 of them . Have you installed the latest NVIDIA drivers?

Add one of this switches to your command line:

-k diablo
-k diakgcn
-k poclbm
-k phatk

Dia

Thanks for the tip.  Drivers are up to date, and I tried all 4 kernels.  On all 4 kernels, the display driver still crashes immediately.

Code:

cgminer -o http://localhost:9332 -O user:pass -d 0 -k (kernel)

If I run the poclbm executable from the guiminer package, it seems to work just fine.  GPU is pegged and things appear to be running at ~25M.  Guess I'll just do this for now and wait a new release of cgminer.

Is there anything I can do to actually help figure out what's happening?

Thanks,

-Prayer

Lets start simple.  
Delete (or move) any bin files in cgminer folder.
Return all cards to stock.
Don't run any third party utilities (Trixx, AB, Catalyst Control Center).  If they are set to run at boot change that temporarily.
Do a reboot try running cgminer with code above.  Does it still crash?  Is so well then you stumped me.
If not then you have a starting point.

The only two things I can think of is
a) cgminer is very efficient.  It pushes a card harder than other miners at a given clock thus you can't always run same clock.  What matters is MH/s (or more directly shares/min) not clock so cgminer can still be the best miner
b) you got some other junk utility like afterburner which is causing a conflict.

[Diapolo]

You could simply try another kernel, as there are 4 of them . Have you installed the latest NVIDIA drivers?

Add one of this switches to your command line:

-k diablo
-k diakgcn
-k poclbm
-k phatk

Dia

Thanks for the tip.  Drivers are up to date, and I tried all 4 kernels.  On all 4 kernels, the display driver still crashes immediately.

Code:

cgminer -o http://localhost:9332 -O user:pass -d 0 -k (kernel)

If I run the poclbm executable from the guiminer package, it seems to work just fine.  GPU is pegged and things appear to be running at ~25M.  Guess I'll just do this for now and wait a new release of cgminer.

Is there anything I can do to actually help figure out what's happening?

Thanks,

-Prayer

Lets start simple.  
Delete (or move) any bin files in cgminer folder.
Return all cards to stock.
Don't run any third party utilities (Trixx, AB, Catalyst Control Center).  If they are set to run at boot change that temporarily.
Do a reboot try running cgminer with code above.  Does it still crash?  Is so well then you stumped me.
If not then you have a starting point.

The only two things I can think of is
a) cgminer is very efficient.  It pushes a card harder than other miners at a given clock thus you can't always run same clock.  What matters is MH/s (or more directly shares/min) not clock so cgminer can still be the best miner
b) you got some other junk utility like afterburner which is causing a conflict.

It would be interesting to know, if any .bin files are generated at all ... perhaps it's a compilation issue with the OpenCL compiler.
You could try to add -v 1 -w 64 to force no vectors and worksize of 64 (perhaps CGMINER autodetection messes something up), I guess this should be save on NVIDIA, but I have really no mining experience with their GPUs.

Dia

[spiccioli]

...
So either my CPU is too weak to handle 5 GPUs or cpu usage can be lower, but it does not depend on catalyst/amd sdk.

BTW I've tested catalyst 11.12 (11.11 gives problems installing) and 12.1 on xubuntu 11.10 64bit this morning with more or less same levels of CPU usage.

spiccioli

While it's running at 60% CPU ...
cat /proc/cpuinfo

Note it says CPU MHz 800.00 so I think I've being chasing the wrong problem for two days, if CPU is at 2.8Ghz cgminer uses around 10% of CPU

Thanks for your help!

spiccioli.

kano,

intensity has an even bigger influence than CPU speed, at intensity 'd', same 800 MHz CPU speed, cgminer uses around 50% CPU time

Code:

top - 23:00:10 up  3:39,  2 users,  load average: 0.06, 0.03, 0.05
Tasks: 145 total,   1 running, 144 sleeping,   0 stopped,   0 zombie
Cpu(s): 10.2%us,  8.5%sy,  0.9%ni, 78.2%id,  2.1%wa,  0.0%hi,  0.1%si,  0.0%st
Mem:   4055036k total,  1491940k used,  2563096k free,    52424k buffers
Swap:  4187132k total,        0k used,  4187132k free,   418096k cached

  PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
17249 user      20   0  515m 134m  37m S 48.9  3.4  11:41.81 cgminer
 1307 user      20   0  175m 106m 3596 S  3.9  2.7   5:34.95 python
18999 user      20   0 19348 1292  924 R  2.0  0.0   0:00.01 top
    1 root      20   0 24000 2104 1292 S  0.0  0.1   0:01.17 init
    2 root      20   0     0    0    0 S  0.0  0.0   0:00.00 kthreadd
    3 root      20   0     0    0    0 S  0.0  0.0   0:00.85 ksoftirqd/0
    5 root      20   0     0    0    0 S  0.0  0.0   0:00.22 kworker/u:0
    6 root      RT   0     0    0    0 S  0.0  0.0   0:00.00 migration/0

spiccioli

[SiegeBreaker]

I would love to see an expanded version of this that goes down to 100 (or lower) memclock using latest version of cgminer, latest drivers, and SDK 2.1.

Here is some tests
https://i.imgur.com/TPRiol.png

You sir, are full of awesome. Will we be able to see the other test results soon?
Do you have a 5870 to test with?

@tenzor Were you running this on linux? And are you willing to share the script you used to generate these results?

[kano]

...
kano,

intensity has an even bigger influence than CPU speed, at intensity 'd', same 800 MHz CPU speed, cgminer uses around 50% CPU time

Code:

...

spiccioli

Actually - logically as the intensity drops, the CPU must increase.

When the intensity goes down, it means the work is broken up into smaller pieces.
More pieces means more CPU processing and transferring for the same amount of hashes.

I run 2x6950 on a 3.07GHz i3 CPU running at 1.2GHz and -I 9 and get ~2% CPU (hmm 2 seems low - I would have guessed 5 ...)
(that % number means of a single core in linux coz if you e.g. have 2 cores - total is CPU 200%)
So it's not totally surprising getting high CPU use with 5 GPU's and low power CPU and lower intensity.

[Diapolo]

I would love to see an expanded version of this that goes down to 100 (or lower) memclock using latest version of cgminer, latest drivers, and SDK 2.1.

Here is some tests

Wow, now that are very nice graphs ... it would be so cool, if you could add the diakgcn kernel into that (-k diakgcn). Not because I think it is faster, but it would be great to know how it performs in comparison to the other available kernels.

Dia

[Hawkix]

Can cgminer set GPU/MEM frequencies out of BIOS ranges? For example, something like MSI Afterburner in Windows can do?

[kano]

Can cgminer set GPU/MEM frequencies out of BIOS ranges? For example, something like MSI Afterburner in Windows can do?

cgminer uses ADL (AMD Display Library, written by ... AMD)
If ADL can't do it, cgminer can't do it.
The expected answer to your question is 'no'

[Diapolo]

Can cgminer set GPU/MEM frequencies out of BIOS ranges? For example, something like MSI Afterburner in Windows can do?

The answer is no, it just can set what the ADL / the driver allows it to, which is more limited, what AfterBurner can offer.

Dia

[-ck]

Can cgminer set GPU/MEM frequencies out of BIOS ranges? For example, something like MSI Afterburner in Windows can do?

The answer is no, it just can set what the ADL / the driver allows it to, which is more limited, what AfterBurner can offer.

Dia

The answer is yes, it will allow you to send commands outside the bios range but the GPU can happily ignore them. It can do more than the bios range, but less than windows specific tools that bypass the driver and poke it directly.

[Diapolo]

Can cgminer set GPU/MEM frequencies out of BIOS ranges? For example, something like MSI Afterburner in Windows can do?

The answer is no, it just can set what the ADL / the driver allows it to, which is more limited, what AfterBurner can offer.

Dia

The answer is yes, it will allow you to send commands outside the bios range but the GPU can happily ignore them. It can do more than the bios range, but less than windows specific tools that bypass the driver and poke it directly.

He asked 2 questions and I answered the 2nd one :-P ...

Dia

[cuz0882]

Some of my 5970's are not accepting any commands from cgminer. I had to use afterburner on about half my rigs. This is to be expected?

[-ck]

Some of my 5970's are not accepting any commands from cgminer. I had to use afterburner on about half my rigs. This is to be expected?

That would be precisely what 3 posts in a row just said.

[spiccioli]

...
kano,

intensity has an even bigger influence than CPU speed, at intensity 'd', same 800 MHz CPU speed, cgminer uses around 50% CPU time

Code:

...

spiccioli

Actually - logically as the intensity drops, the CPU must increase.

When the intensity goes down, it means the work is broken up into smaller pieces.
More pieces means more CPU processing and transferring for the same amount of hashes.

I run 2x6950 on a 3.07GHz i3 CPU running at 1.2GHz and -I 9 and get ~2% CPU (hmm 2 seems low - I would have guessed 5 ...)
(that % number means of a single core in linux coz if you e.g. have 2 cores - total is CPU 200%)
So it's not totally surprising getting high CPU use with 5 GPU's and low power CPU and lower intensity.

kano,

with lower intensity occupation of GPU is lower, so I was thinking that the thread serving a GPU with lower intensity had to be sleeping a lot more, while it is, instead, sleeping a lot less.

Code:

73.0 C  F: 24% (1335 RPM)  E: 800 MHz  M: 150 Mhz  V: 1.000V  A: 80% P: 0%
Last initialised: [2012-02-26 23:06:14]
Intensity: Dynamic (only one thread in use)
Thread 1: 299.8 Mh/s Enabled ALIVE

How many more hashes (or how longer does it keep counting) are counted by a GPU when intensity is increased by one?

spiccioli.

[kano]

...
kano,

intensity has an even bigger influence than CPU speed, at intensity 'd', same 800 MHz CPU speed, cgminer uses around 50% CPU time

Code:

...

spiccioli

Actually - logically as the intensity drops, the CPU must increase.

When the intensity goes down, it means the work is broken up into smaller pieces.
More pieces means more CPU processing and transferring for the same amount of hashes.

I run 2x6950 on a 3.07GHz i3 CPU running at 1.2GHz and -I 9 and get ~2% CPU (hmm 2 seems low - I would have guessed 5 ...)
(that % number means of a single core in linux coz if you e.g. have 2 cores - total is CPU 200%)
So it's not totally surprising getting high CPU use with 5 GPU's and low power CPU and lower intensity.

kano,

with lower intensity occupation of GPU is lower, so I was thinking that the thread serving a GPU with lower intensity had to be sleeping a lot more, while it is, instead, sleeping a lot less.

Code:

73.0 C  F: 24% (1335 RPM)  E: 800 MHz  M: 150 Mhz  V: 1.000V  A: 80% P: 0%
Last initialised: [2012-02-26 23:06:14]
Intensity: Dynamic (only one thread in use)
Thread 1: 299.8 Mh/s Enabled ALIVE

How many more hashes (or how longer does it keep counting) are counted by a GPU when intensity is increased by one?

spiccioli.

Yes the CPU is sleeping a lot less with lower intensity as I said here:

When the intensity goes down, it means the work is broken up into smaller pieces.
More pieces means more CPU processing and transferring for the same amount of hashes.

I'll explain again I guess:
A work request is 2^32 hashes (a full nonce range)
It maybe could be done in 1 one long go or 2^32 tiny goes.
It is actually, of course, done somewhere in the middle of that.

kano goes read the code to try get the numbers ...

OK it looks like (though I could be wrong) that with intensity 9 it divides the work into pieces of (2^(15+9) * vectors) hashes
So about 16 million hashes with -v 1 (there are other limitations it does to this number but it looks like that's maybe correct for me with -I 9)

This figure doesn't really matter if it's wrong, but how I've described how it works is correct.
Anyway assuming it is (2^(15+9)) or 16 million - it takes about 52ms for a single 6950 GPU to do that.
Thus every 52ms (if I had only one 6950) the CPU needs to setup and feed data into the GPU and then wait about 52ms to get the answer back, then setup the next 16 million hashes and feed them into the GPU ... over and again ... until it's processed the whole 2^32 hashes

If instead the intensity was 8, each setup would contain half the number of hashes (2^(15+8)) or 8 million, or approx 26ms so the CPU would have to setup work twice as often to complete the whole 2^32 nonce range - thus the CPU usage would increase since it's doing twice as many setups for the same amount of work - to do the complete 2^32 hashes

[Vbs]

Dia, have you though on changing (for example, in VECTORS2),

Code:

	V[7] ^= 0x136032edU;
	bool result = V[7].x & V[7].y;
	if (!result) {
		if (!V[7].x)
			output[FOUND] = output[NFLAG & nonce.x] = nonce.x;
		if (!V[7].y)
			output[FOUND] = output[NFLAG & nonce.y] = nonce.y;
	}

to

Code:

	uint result = V[7].x == 0x136032edU ? nonce.x:0u;
	result = V[7].y == 0x136032edU ? nonce.y:result;
	if (result)
		output[FOUND] = output[NFLAG & result] = result;

This should give a small boost.

Btw, the probability that both nonce.x and nonce.y are correct at the same time is 1/(2^64). This means, that on a 400MH/s card, this will probably happen once in every ~1463 years!