Internet151
|
|
February 28, 2012, 11:44:26 AM |
|
Found a small bug, when cgminer passes 200k accepted shares in Windows 7 the GPU temp and fan speed columns go completely blank and are also unable to be accessed via the API (which I noticed vis ANUBIS).
Ha, +1, I also noticed this, didn't link it with the 200k shares though... ;-) It's more likely to be simply that the ADL library gave up and now says it's not working. Not really helpful, but you can check the API config command: It will say (ADL=Y,ADL in use=N) if all cards are showing blank for the GPU/Fan/Temp values. If any cards are still working (unexpected) then it will still say (ADL in use=Y) Also, if you are using ANUBIS - does it have a log of the stats over time? You should be able to tell exactly what and where if it logs all the stats (I don't know how much it logs) ANUBIS doesn't log anything, and all the cards are still working and hashing away. It's just a bug that affects the display of those values and makes those specific values unretrievable via the API.
|
|
|
|
kano
Legendary
Offline
Activity: 4592
Merit: 1851
Linux since 1997 RedHat 4
|
|
February 28, 2012, 11:58:51 AM |
|
Found a small bug, when cgminer passes 200k accepted shares in Windows 7 the GPU temp and fan speed columns go completely blank and are also unable to be accessed via the API (which I noticed vis ANUBIS).
Ha, +1, I also noticed this, didn't link it with the 200k shares though... ;-) It's more likely to be simply that the ADL library gave up and now says it's not working. Not really helpful, but you can check the API config command: It will say (ADL=Y,ADL in use=N) if all cards are showing blank for the GPU/Fan/Temp values. If any cards are still working (unexpected) then it will still say (ADL in use=Y) Also, if you are using ANUBIS - does it have a log of the stats over time? You should be able to tell exactly what and where if it logs all the stats (I don't know how much it logs) ANUBIS doesn't log anything, and all the cards are still working and hashing away. It's just a bug that affects the display of those values and makes those specific values unretrievable via the API. That's what I said The ADL library that AMD wrote and cgminer uses is what controls GPU/Fan/Temp. The ADL library is known to just give up and die sometimes. cgminer keeps going but can no longer access that information ... that's why it's blank (like if you forget "export DISPLAY=:0" in linux) You will find that anything in cgminer that requires ADL now no longer works. That really means: if cgminer was given --auto-gpu or --auto-fan (or both) it is also time to stop and restart cgminer since it is no longer monitoring your fan and/or gpu.
|
|
|
|
Internet151
|
|
February 28, 2012, 01:20:12 PM |
|
Found a small bug, when cgminer passes 200k accepted shares in Windows 7 the GPU temp and fan speed columns go completely blank and are also unable to be accessed via the API (which I noticed vis ANUBIS).
Ha, +1, I also noticed this, didn't link it with the 200k shares though... ;-) It's more likely to be simply that the ADL library gave up and now says it's not working. Not really helpful, but you can check the API config command: It will say (ADL=Y,ADL in use=N) if all cards are showing blank for the GPU/Fan/Temp values. If any cards are still working (unexpected) then it will still say (ADL in use=Y) Also, if you are using ANUBIS - does it have a log of the stats over time? You should be able to tell exactly what and where if it logs all the stats (I don't know how much it logs) ANUBIS doesn't log anything, and all the cards are still working and hashing away. It's just a bug that affects the display of those values and makes those specific values unretrievable via the API. That's what I said The ADL library that AMD wrote and cgminer uses is what controls GPU/Fan/Temp. The ADL library is known to just give up and die sometimes. cgminer keeps going but can no longer access that information ... that's why it's blank (like if you forget "export DISPLAY=:0" in linux) You will find that anything in cgminer that requires ADL now no longer works. That really means: if cgminer was given --auto-gpu or --auto-fan (or both) it is also time to stop and restart cgminer since it is no longer monitoring your fan and/or gpu. The ADL library isn't just randomly dying here. It happens at 200k accepted shares (this was easy for me to recognize since I have a 32 GH/s mining farm). I'm just pointing out my observation so hopefully it gets fixed.
|
|
|
|
DeathAndTaxes
Donator
Legendary
Offline
Activity: 1218
Merit: 1079
Gerald Davis
|
|
February 28, 2012, 01:33:24 PM |
|
The ADL library isn't just randomly dying here. It happens at 200k accepted shares (this was easy for me to recognize since I have a 32 GH/s mining farm). I'm just pointing out my observation so hopefully it gets fixed.
This time. Has it died @ 200K previously too? Does it die at 200K each time? If it is a repeatable and verifiable bug that is more useful than just it dying once @ 200K (which may not have any significance).
|
|
|
|
jamesg
VIP
Legendary
Offline
Activity: 1358
Merit: 1000
AKA: gigavps
|
|
February 28, 2012, 01:56:32 PM |
|
Kano & Conman, I have my BFL single running on BAMT with cgminer but it would seem that the API call "devs" does not report back about the BFL single. Here is the output from the "devs" call. STATUS=S ,Code=9 ,Msg=2 GPU(s) ,Description=cgminer 2.2.7 |GPU=0,Enabled=Y,Status=Alive,Temperature=73.50,Fan Speed=2355,Fan Percent=48,GPU Clock=895,Memory Clock=1000,GPU Voltage=1.100,GPU Activity=99,Powertune=0,MHS av=364.64,MHS 5s=368.09,Accepted=81,Rejected=0,Hardware Errors=0,Utility=4.87,Intensity=9,Last Share Pool=0,Last Share Time=1330436872 |GPU=1,Enabled=Y,Status=Alive,Temperature=73.50,Fan Speed=1892,Fan Percent=42,GPU Clock=895,Memory Clock=1000,GPU Voltage=1.100,GPU Activity=99,Powertune=0,MHS av=363.33,MHS 5s=367.82,Accepted=73,Rejected=0,Hardware Errors=0,Utility=4.38,Intensity=9,Last Share Pool=0,Last Share Time=1330436857
So the devs call is missing all BFL info.
|
|
|
|
Internet151
|
|
February 28, 2012, 02:17:33 PM |
|
The ADL library isn't just randomly dying here. It happens at 200k accepted shares (this was easy for me to recognize since I have a 32 GH/s mining farm). I'm just pointing out my observation so hopefully it gets fixed.
This time. Has it died @ 200K previously too? Does it die at 200K each time? If it is a repeatable and verifiable bug that is more useful than just it dying once @ 200K (which may not have any significance). Hrm, I looked back over my 24 rigs and I'm afraid I was wrong. It doesn't always happen 200k but the bug is occurring on all of them within hours of each other after running for about a week straight with 4 GPU's per rig. It will take some time but i'll try to identify the exact trigger.
|
|
|
|
SiegeBreaker
Newbie
Offline
Activity: 26
Merit: 0
|
|
February 28, 2012, 03:15:13 PM |
|
|
|
|
|
bulanula
|
|
February 28, 2012, 05:23:28 PM |
|
Yeah. I am looking forward to these tasty results !
|
|
|
|
P_Shep
Legendary
Offline
Activity: 1795
Merit: 1208
This is not OK.
|
|
February 28, 2012, 05:59:14 PM |
|
Kano & Conman, I have my BFL single running on BAMT with cgminer but it would seem that the API call "devs" does not report back about the BFL single. Here is the output from the "devs" call. STATUS=S ,Code=9 ,Msg=2 GPU(s) ,Description=cgminer 2.2.7 |GPU=0,Enabled=Y,Status=Alive,Temperature=73.50,Fan Speed=2355,Fan Percent=48,GPU Clock=895,Memory Clock=1000,GPU Voltage=1.100,GPU Activity=99,Powertune=0,MHS av=364.64,MHS 5s=368.09,Accepted=81,Rejected=0,Hardware Errors=0,Utility=4.87,Intensity=9,Last Share Pool=0,Last Share Time=1330436872 |GPU=1,Enabled=Y,Status=Alive,Temperature=73.50,Fan Speed=1892,Fan Percent=42,GPU Clock=895,Memory Clock=1000,GPU Voltage=1.100,GPU Activity=99,Powertune=0,MHS av=363.33,MHS 5s=367.82,Accepted=73,Rejected=0,Hardware Errors=0,Utility=4.38,Intensity=9,Last Share Pool=0,Last Share Time=1330436857
So the devs call is missing all BFL info. Yeah, cgminer is missing all FPGA related API stuff
|
|
|
|
kano
Legendary
Offline
Activity: 4592
Merit: 1851
Linux since 1997 RedHat 4
|
|
February 29, 2012, 02:47:06 AM |
|
Kano & Conman,
I have my BFL single running on BAMT with cgminer but it would seem that the API call "devs" does not report back about the BFL single. Here is the output from the "devs" call.
...
So the devs call is missing all BFL info.
Yeah, cgminer is missing all FPGA related API stuff Feel free to add to the bounty https://bitcointalk.org/index.php?topic=66699.0
|
|
|
|
kano
Legendary
Offline
Activity: 4592
Merit: 1851
Linux since 1997 RedHat 4
|
|
February 29, 2012, 10:54:49 AM |
|
|
|
|
|
bitlane
Internet detective
Sr. Member
Offline
Activity: 462
Merit: 250
I heart thebaron
|
|
February 29, 2012, 12:03:31 PM Last edit: February 29, 2012, 12:13:42 PM by bitlane |
|
First off, this is going to sound extremely STUPID because I know neither how to code or the proper questions to ask.....lol
'Efficiency' (E) Can this value be queried using RPC ? or does this value have to be calculated locally using 'Q' and 'A' (say, for a monitoring Script)....?
ie. I can retreive the ACCEPTED value per miner (A) value using ($value['Accepted'])......the REJECTED value per miner using ($value['Rejected'])
I have tried throwing in 'Efficiency' as a tag and can't seem to get a value per miner for it. A little help ?
Again, I apologize for soiling the sanctity of coding....and for being so stupid...lol
|
|
|
|
kano
Legendary
Offline
Activity: 4592
Merit: 1851
Linux since 1997 RedHat 4
|
|
February 29, 2012, 01:19:55 PM |
|
Yeah it doesn't supply numbers that can be calculated directly - that would be a waste.
It's Accepted / Getworks ( * 100)
However, remember that the number is not very useful - it is just how many shares you found per getwork. RollNTime allows that to be up to 500% with cgminer.
|
|
|
|
DeathAndTaxes
Donator
Legendary
Offline
Activity: 1218
Merit: 1079
Gerald Davis
|
|
February 29, 2012, 01:37:25 PM Last edit: February 29, 2012, 03:16:04 PM by DeathAndTaxes |
|
Yeah it doesn't supply numbers that can be calculated directly - that would be a waste.
It's Accepted / Getworks ( * 100)
However, remember that the number is not very useful - it is just how many shares you found per getwork. RollNTime allows that to be up to 500% with cgminer.
And pools with higher difficulty shares (like p2pool) or frequent LP (like merge mining alt-chain with much lower difficulty) can have an efficiency that is much lower even when everything is optimal. Really efficiency only matters as a pool op. Deepbit collectively finds ~700 shares per second. That is a lot of getworks. Even a modest improvement in efficiency reduces load but even that is marginal because peak load will still happen at an LP even with 500% efficiency. I think a stat like the following would be a more useful metric (and prevent continually obsession w/ efficiency): Effectiveness = accepted / worked Effectiveness = accepted / (accepted + rejected + stale share + hardware failure) Having a high effectiveness (no it isn't a stat in cgminer) is more important than a high efficiency.
|
|
|
|
kano
Legendary
Offline
Activity: 4592
Merit: 1851
Linux since 1997 RedHat 4
|
|
February 29, 2012, 02:20:11 PM |
|
Is there a way to get an email notification when GPU goes into SICK or DEAD state?
If not, can the APIs be used to query stats from cgminer. If possible, we could write cgminer-watchdog to monitor the critical stats and notify you via email or sms.
That's what I'm writing at the moment ... ... but it will allow you to check many different things ...
|
|
|
|
niooron
|
|
February 29, 2012, 05:06:06 PM |
|
Is there a way to set the OpenCL platform per card? Cgminer keeps crashing with a mixed 5xxx and 6xxx rig.
|
|
|
|
sharky112065
|
|
February 29, 2012, 05:31:12 PM |
|
Is there a way to set the OpenCL platform per card? Cgminer keeps crashing with a mixed 5xxx and 6xxx rig.
I'm not sure as to why it is crashing on you. I have a rig that has 3 5870's and 1 6970, and it runs just fine. Cgminer also automatically generates .bin's for each type of card (Caymen and Cypress). Anyway, right out of the freaking manual (README)... Q: I have multiple SDKs installed, can I choose which one it uses? A: Run cgminer with the -n option and it will list all the platforms currently installed. Then you can tell cgminer which platform to use with --gpu-platform. --kernel|-k <arg> Override kernel to use (diablo, poclbm, phatk or diakgcn) - one value or comma separated
|
Donations welcome: 12KaKtrK52iQjPdtsJq7fJ7smC32tXWbWr
|
|
|
sharky112065
|
|
February 29, 2012, 05:34:59 PM |
|
Cool. Let me know if you need any help making it build/work on Windows.
Thanks.
Anyone can now build Cgminer on Windows. http://pastebin.com/3pzivj32I spent a lot of time documenting and testing the documentation step by step on both Windows 7 and Windows XP. Shouldn't take more than 30 min to set up the build environment.
|
Donations welcome: 12KaKtrK52iQjPdtsJq7fJ7smC32tXWbWr
|
|
|
niooron
|
|
February 29, 2012, 06:25:01 PM |
|
Is there a way to set the OpenCL platform per card? Cgminer keeps crashing with a mixed 5xxx and 6xxx rig.
I'm not sure as to why it is crashing on you. I have a rig that has 3 5870's and 1 6970, and it runs just fine. Cgminer also automatically generates .bin's for each type of card (Caymen and Cypress). Anyway, right out of the freaking manual (README)... Q: I have multiple SDKs installed, can I choose which one it uses? A: Run cgminer with the -n option and it will list all the platforms currently installed. Then you can tell cgminer which platform to use with --gpu-platform. --kernel|-k <arg> Override kernel to use (diablo, poclbm, phatk or diakgcn) - one value or comma separated I'm trying to use the 2.1 SDK for the 5xxx and 2.6 for the 6xxx. Everything works fine using the 2.6 sdk, but it obviously isnt efficient for the 5xxx series. Phoenix 2.0 lets me use the setup I want, but it lacks lots of features. I also tried making cgminer use bins compiled for 2.1 while using the 2.6 sdk but it just recompiles them again. The ideal would be making --gpu-platform a per device option, something like --gpu-platform 0,0,1
|
|
|
|
P_Shep
Legendary
Offline
Activity: 1795
Merit: 1208
This is not OK.
|
|
February 29, 2012, 08:06:02 PM |
|
I'm trying to compile cgminer for windows with CPU mining enabled, but I can't get it to link when I specify -msse2. It seems that sha256_xmm.o isn't being included. I'm getting the following error: CCLD cgminer.exe cgminer-sha256_sse2_i386.o:sha256_sse2_i386.c:(.text+0x155): undefined reference to `@CalcSha256_x86@12' cgminer-sha256_sse2_i386.o:sha256_sse2_i386.c:(.text+0x16a): undefined reference to `@CalcSha256_x86@12' collect2: ld returned 1 exit status
without -msse2 it compiles fine, but of course, without the SSE2 optimised code. Any help to kick this into action???
|
|
|
|
|