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

[-ck]

I played a little around an found out, that the nonce does'nt have to change the endian on arm.

Is it normal that a Generated target looks like this:

Generated target 0000000000000000000000000000000000000000000000ffff00020d7800780d

on a working stratum machine, it look like this:

[2012-10-13 00:01:34] Generated target 0000000000000000000000000000000000000000000000000000ffff00000000

No it is not normal. The last 8 hex characters should always be zero. The working stratum example is what a normal diff 1 target should look like.

after the call  swab256(target, rtarget); the last bytes of target are destroyed.

But rtarget looks good. 

Very weird. swab256 is a pretty simple function swapping the 8x4byte array members (which means 8 hex values) of rtarget to target and then byteswapping the values of each 4 byte value. It looks like something is stomping on the last few bytes on your arm build though. The endian issue could be coincidence and not the real problem.

I changed the default size of target and rtarget like this:

Code:

static void set_work_target(struct work *work, int diff)
{
        unsigned char rtarget[36], target[36];

Now i got targets like expected..
[2012-10-13 01:39:50] Generated target 0000000000000000000000000000000000000000000000000000ffff00000000

But still the same problem that i get rejected on stratum

 [2012-10-13 01:40:01] RECVD: {"error": [-2, "lowdifficulty", null], "id": 68, "result": false}
 [2012-10-13 01:40:01] PROOF OF WORK RESULT: false (booooo)
 [2012-10-13 01:40:01] Rejected 4e8ec61a Diff 3/1 MM 6 pool 0 (lowdifficulty)

Hmm come to think of it, arm has an issue with needing to allocate memory aligned to a certain size. Presumably that's now a problem on the share submission side now that you've sort of fixed the target generation.

[Krak]

Dynamic intensity is still messed up in 2.8.3. Now it's going up to 15.

I'm loving the new difficulty stats. It's cool getting to see what each share is.

[-ck]

Dynamic intensity is still messed up in 2.8.3. Now it's going up to 15.

Eh? I tested it on windows and linux...

[Krak]

Dynamic intensity is still messed up in 2.8.3. Now it's going up to 15.

Eh? I tested it on windows and linux...

Hmm I dunno then. It took about a minute or two for it to start happening after I started up 2.8.3; it was happening instantly in 2.8.2.

[-ck]

I'm loving the new difficulty stats. It's cool getting to see what each share is.

Yes it makes it far more interesting just watching hashes go by... though of course it's a complete time waster. But if you see anything 3M or bigger, that would be exciting cause it would be a block solve 

[kano]

However
If you see something with 3M and it isn't a BLOCK! - then it is time to weep
(coz it has to actually be above 3054627.5... at the moment)

[Frizz23]

@ckolivas: Will cgminer support BFL ASICs? Are you already working on this?

It will, along with ASICs from other manufacturers. I am not directly working on the driver for it yet because they have not released any specifications yet.

On the other hand, there have been numerous changes made to cgminer in preparation for those ASICs. This is precisely why cgminer now supports the stratum protocol, has massive rolltime support and can display devices up to exahash speed.

Did Josh/BFL already contact you?

From https://forums.butterflylabs.com/showthread.php/125-Can-you-send-a-%28beta%29-ASIC-device-to-ckolivas-%28developer-cgminer%29?p=1667&viewfull=1#post1667
Q: Would it be possible that you send one of your (beta) ASIC devices to ckolivas (the developer of cgminer) so that he can start supporting it?
A: Yes, as soon as we have something available, we will be sending him one.

[dietwice]

2.8.3 on Win7 x64 and XP x32 crashes eventually.

Last time it crashed on 2 different XP rigs at the same time so I think it could be related to some network events processing.

Sorry, no logs

[ocminer]

Is there a possibility to stop a ICA/ZTX worker/thread ?

I'd like to "Disable" ICA2 i.e. and let the others running.. like with GPU's

[NLA]

Is it possible to point particular GPU's in a single instance of cgminer to different pools, rather than having to do so with multiple cgminer instances? For instance, say I want to launch just one cgminer instance, and point one GPU explicitly at Eclipse, one at Bitminter, and one at Slush, all with different login credentials?

[crazyates]

Is it possible to point particular GPU's in a single instance of cgminer to different pools, rather than having to do so with multiple cgminer instances? For instance, say I want to launch just one cgminer instance, and point one GPU explicitly at Eclipse, one at Bitminter, and one at Slush, all with different login credentials?

It's been asked before. Than answer is no, you have to use multiple instances.

[-ck]

@ckolivas: Will cgminer support BFL ASICs? Are you already working on this?

It will, along with ASICs from other manufacturers. I am not directly working on the driver for it yet because they have not released any specifications yet.

On the other hand, there have been numerous changes made to cgminer in preparation for those ASICs. This is precisely why cgminer now supports the stratum protocol, has massive rolltime support and can display devices up to exahash speed.

Did Josh/BFL already contact you?

From https://forums.butterflylabs.com/showthread.php/125-Can-you-send-a-%28beta%29-ASIC-device-to-ckolivas-%28developer-cgminer%29?p=1667&viewfull=1#post1667
Q: Would it be possible that you send one of your (beta) ASIC devices to ckolivas (the developer of cgminer) so that he can start supporting it?
A: Yes, as soon as we have something available, we will be sending him one.

I originally approached Sonny from BFL a long time ago about this and he said they should be able to send me hardware, though I have not been contacted directly by BFL since then, so this is good news.

[-ck]

Is there a possibility to stop a ICA/ZTX worker/thread ?

I'd like to "Disable" ICA2 i.e. and let the others running.. like with GPU's

No.

[-ck]

2.8.3 on Win7 x64 and XP x32 crashes eventually.

Last time it crashed on 2 different XP rigs at the same time so I think it could be related to some network events processing.

Sorry, no logs

Hmm I wonder why. Is this while mining with stratum? Curious to see what happens in your rigs when you disconnect the network cable or something like that. Alas I don't have anything but a laptop that has windows which doesn't mine any useful amount and it doesn't behave like this.

[-ck]

And regarding my absence last month and the family tragedy I mentioned, for those who expressed their well wishes...

http://ck-hack.blogspot.com/2012/10/the-vicissitudes-of-life.html

[AxelMi]

Now i got CGminer 2.8.3 with Stratum running on my ARM Hardware.

i changed the following lines.

Code:

1981c1978
<       char rhash[36];
---
>       char rhash[33];

4596c4594
<       unsigned char hash1[36];
---
>       unsigned char hash1[33];

4608c4606
<       unsigned char rtarget[36], target[36];
---
>       unsigned char rtarget[33], target[33];

4634,4635c4631,4632
<       unsigned char *coinbase, merkle_root[36], merkle_sha[68], *merkle_hash;
<       char header[260], hash1[132], *nonce2;
---
>       unsigned char *coinbase, merkle_root[33], merkle_sha[65], *merkle_hash;
>       char header[257], hash1[129], *nonce2;
4664c4661
<               unsigned char merkle_bin[36];
---
>               unsigned char merkle_bin[33];

on my patched version are the lines different, because i have implemented my own driver.


My GPU Miners with 2.8.3 and running stratum died also ~6hours ago. (20 Machines running all with the win32 build)

[-ck]

Now i got CGminer 2.8.3 with Stratum running on my ARM Hardware.

on my patched version are the lines different, because i have implemented my own driver.

Well done, thanks. These are trivial enough to become part of the cgminer codebase. It does seem ARM wants things aligned to 4 byte boundaries.

Any chance your "own driver" is something other people would like and could become part of cgminer?

My GPU Miners with 2.8.3 and running stratum died also ~6hours ago. (20 Machines running all with the win32 build)

Not so good. I don't know what's plaguing windows builds now.

[AxelMi]

My Driver is only usefull for my self build Hardware.
so i think it doesn't make sense.

But if you wanna have it. you can get it.

[-ck]

My Driver is only usefull for my self build Hardware.
so i think it doesn't make sense.

But if you wanna have it. you can get it.

If it's one of a kind, it's okay there is no point merging the code. Thanks though.

[amigaman]

I was using 2.7.6 on my dual 7970 (Club 3D and Sapphire with AMD cooling) rigs for as long as it was out now, with 1125MHz core & 1000 mem, no voltage change, power setting at -20%, all Win7 32, 1250Mh per rig.
I have noticed all 2.8 builds have the problem of declaring the GPUs sick with that clock rate.
They work with the stock 925 and a little bit more, but above 1000 i get sick fast.
I have not modded the cards in any way, just plugged them in, installed latest AMD driver and started CGM.
It (or the driver?) itself set the clock up to 1125, the maximum, more is not allowed by the bios.
I have to manually reduce engine clock to below 1000 to have it run a few hours, but then it inevitably dies with GPU declared sick, and a windows pops up "CGMiner.exe has stopped working".
What is going on here?

I better stick to 2.7.6 and use the proxy, that does work for days now.