Topic: Bounty for Cracking Bitmain S9 BMminer - page 3. (Read 38682 times)
Waiting for an update that will display board and chip temps, then will try again.
Yes, thanks for testing. Encouraging to know that restoring to the default configuration is possible!
I ended up having to flash the bitmain firmware update. I wasn't able to see the GH/S(avg) but it's working fine now.
Yes, thanks for testing. Encouraging to know that restoring to the default configuration is possible!
Tried on a B1 14T miner at 625 frequency. Miner worked and was up and running. As fubly stated I wasn't getting a good hash read out. Using awesome miner I was able to determine hash rate was about the same over a 30 minute test. Problem I ran into was. I was only able to see chip temp for one board, it read 114 which I felt was way to hot. My other B1's are running high 80's. I restored my BMminer and things are back to normal. I think Fubly is heading in the right direction, just a few bugs need to be worked out. I think the goal here if I'm not mistaken is doing away with BMminer, but also achieving a better hash rate or lower frequency.
thank you for testing,
looks like everyone afraid to use it on S9 without check experience !
Tried on a B1 14T miner at 625 frequency. Miner worked and was up and running. As fubly stated I wasn't getting a good hash read out. Using awesome miner I was able to determine hash rate was about the same over a 30 minute test. Problem I ran into was. I was only able to see chip temp for one board, it read 114 which I felt was way to hot. My other B1's are running high 80's. I restored my BMminer and things are back to normal. I think Fubly is heading in the right direction, just a few bugs need to be worked out. I think the goal here if I'm not mistaken is doing away with BMminer, but also achieving a better hash rate or lower frequency.
Anybody try this yet?
Problem -upload bmminer- reported by Sierra8561
https://github.com/Zwilla/bmminer-cgminer492/issues/1
is fixed.
Problem: I was using an additional command of copy, but this was only available on developer machine.
Problem -Flash back- reported by NotFuzzyWarm
https://github.com/Zwilla/bmminer-cgminer492/issues/2
is fixed
Problem: I forgot to "cd" into the folder where the other files are
API: Problem API-Stats reported by NotFuzzyWarm
https://github.com/Zwilla/bmminer-cgminer492/issues/3
working on atm. I reported it here:
https://bitcointalksearch.org/topic/m.16193700
The software is running, but the stats page shows not all values atm.
https://github.com/Zwilla/bmminer-cgminer492/issues/1
is fixed.
Problem: I was using an additional command of copy, but this was only available on developer machine.
Problem -Flash back- reported by NotFuzzyWarm
https://github.com/Zwilla/bmminer-cgminer492/issues/2
is fixed
Problem: I forgot to "cd" into the folder where the other files are
API: Problem API-Stats reported by NotFuzzyWarm
https://github.com/Zwilla/bmminer-cgminer492/issues/3
working on atm. I reported it here:
https://bitcointalksearch.org/topic/m.16193700
The software is running, but the stats page shows not all values atm.
Just tried it and got this.. cp: invalid option -- 'n' BusyBox v1.21.1 (2015-05-15 12:25:19 CST) multi-call binary. Usage: cp [OPTIONS] SOURCE... DEST Copy SOURCE(s) to DEST -a Same as -dpR -R,-r Recurse -d,-P Preserve symlinks (default if -R) -L Follow all symlinks -H Follow symlinks on command line -p Preserve file attributes if possible -f Overwrite -i Prompt before overwrite -l,-s Create (sym)links
Go Back
Have you watched his video yet?Go Back
Got slow 'net connection here at work so I haven't. Speaking of which... back to work work work
Yes watched it twice. I'll check it again and make sure I didn't miss a step
Just tried it and got this.. cp: invalid option -- 'n' BusyBox v1.21.1 (2015-05-15 12:25:19 CST) multi-call binary. Usage: cp [OPTIONS] SOURCE... DEST Copy SOURCE(s) to DEST -a Same as -dpR -R,-r Recurse -d,-P Preserve symlinks (default if -R) -L Follow all symlinks -H Follow symlinks on command line -p Preserve file attributes if possible -f Overwrite -i Prompt before overwrite -l,-s Create (sym)links
Go Back
Have you watched his video yet?Go Back
Got slow 'net connection here at work so I haven't. Speaking of which... back to work work work
Just tried it and got this.. cp: invalid option -- 'n' BusyBox v1.21.1 (2015-05-15 12:25:19 CST) multi-call binary. Usage: cp [OPTIONS] SOURCE... DEST Copy SOURCE(s) to DEST -a Same as -dpR -R,-r Recurse -d,-P Preserve symlinks (default if -R) -L Follow all symlinks -H Follow symlinks on command line -p Preserve file attributes if possible -f Overwrite -i Prompt before overwrite -l,-s Create (sym)links
Go Back
Go Back
anyone tried this yet?
I will be trying it tonight after work. I have 2 s9's at home, a batch-1 and a batch-3. Of course the 1st step will be to generate an archive file of their current configurations. Ya know, as a just in case... 
The b1 will be the victim. Currently slightly under clocked @ 612MHz I believe maybe 606. Stock speed supposed to be 650 but chip temps too high there. If things look good sometime over the weekend the b3 will follow.
@fubly re:
Quote
Is it ok if we start a new thread about cgminer 4.9.2 for Antminer S9. I think we all wrote here a lot of BS and as fast I can forgot this as better?
Download this: https://github.com/Zwilla/bmminer-cgminer492/blob/master/update-script/AntminerS9-Update-Bmminer.tar
Open your miners frontpage http://antminer
Click on System
Click on Upgrade
Goto the menu / line "Restore backup:"
Select the file "AntminerS9-Update-Bmminer.tar"
Click on the button "upload Archive"
Wait 25 seconds
Check MinerStatus Page
TODO: some little fixing at the api
Thank you very much, this should make things painless Download this: https://github.com/Zwilla/bmminer-cgminer492/blob/master/update-script/AntminerS9-Update-Bmminer.tar
Open your miners frontpage http://antminer
Click on System
Click on Upgrade
Goto the menu / line "Restore backup:"
Select the file "AntminerS9-Update-Bmminer.tar"
Click on the button "upload Archive"
Wait 25 seconds
Check MinerStatus Page
TODO: some little fixing at the api
Und ja, I think it is time for you to make a new thread on this to give your work here a fresh start.
Tip: for your new thread, immediately add 1 or 2 replies to it with just the text:
By just editing those later you can add any important update info right at the top of the thread.
anyone tried this yet?
Making now some tweaks to the frontpage and update script, to backup all. So that you can download it.
Finish.
Now the script save your original files, you can download them direct from antminer`s frontpage.
How it works? Watch this video now!
http://www.youtube.com/watch?v=d-eDsjqGTDo
Dear Community,
I hope you have all now, that waiting and donating for.
I need a break!
Have a nice weekend and I wish you all a lucky block Friday!
p.s. donation alway welcome: 16L45ub3SpHZ6dYbqseip7Fv4BEJSj9xEo
Finish.
Now the script save your original files, you can download them direct from antminer`s frontpage.
How it works? Watch this video now!
http://www.youtube.com/watch?v=d-eDsjqGTDo
Dear Community,
I hope you have all now, that waiting and donating for.
I need a break!
Have a nice weekend and I wish you all a lucky block Friday!
p.s. donation alway welcome: 16L45ub3SpHZ6dYbqseip7Fv4BEJSj9xEo
I was on a point to revert all my work to yesterday 9am, but now I fixed a memory leak and the new version which is at git now,
compiles perfect (with some errors) and works well.
I am confused. How can you say "compiles perfect", but still have some errors. Is that not the very definition of not compiling perfectly? What compile time errors are you getting?compiles perfect (with some errors) and works well.
This:
Code:
CC cgminer-driver-btm-c5.o
driver-btm-c5.c:55:0: warning: "offsetof" redefined [enabled by default]
#define offsetof(TYPE, MEMBER) ((size_t) &((TYPE *)0)->MEMBER)
^
In file included from ./lib/string.h:34:0,
from uthash.h:29,
from miner.h:37,
from driver-btm-c5.c:35:
/usr/lib/gcc/arm-angstrom-linux-gnueabi/4.8.3/include/stddef.h:413:0: note: this is the location of the previous definition
#define offsetof(TYPE, MEMBER) __builtin_offsetof (TYPE, MEMBER)
^
CC cgminer-sha2_c5.o
CCLD cgminer
driver-btm-c5.c:55:0: warning: "offsetof" redefined [enabled by default]
#define offsetof(TYPE, MEMBER) ((size_t) &((TYPE *)0)->MEMBER)
^
In file included from ./lib/string.h:34:0,
from uthash.h:29,
from miner.h:37,
from driver-btm-c5.c:35:
/usr/lib/gcc/arm-angstrom-linux-gnueabi/4.8.3/include/stddef.h:413:0: note: this is the location of the previous definition
#define offsetof(TYPE, MEMBER) __builtin_offsetof (TYPE, MEMBER)
^
CC cgminer-sha2_c5.o
CCLD cgminer
How to upgrade!
https://www.youtube.com/watch?v=h8geGuODpUE
the Git that the software is free to use under own responsibility since it cant be proven do to conflict of interest.
NO REBOOT NEEDED!
After uploading, the page will change to index.html
After some seconds you will see, that cgminer 4.9.2 is running.
If you want to see what the upgrade script will do, read this file!
https://github.com/Zwilla/bmminer-cgminer492/blob/master/update-script/restoreConfig.sh
Is it ok if we start a new thread about cgminer 4.9.2 for Antminer S9. I think we all wrote here a lot of BS and as fast I can forgot this as better?
TODO: some little fixing at the api
https://www.youtube.com/watch?v=h8geGuODpUE
the Git that the software is free to use under own responsibility since it cant be proven do to conflict of interest.
Code:
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
NO REBOOT NEEDED!
After uploading, the page will change to index.html
After some seconds you will see, that cgminer 4.9.2 is running.
If you want to see what the upgrade script will do, read this file!
https://github.com/Zwilla/bmminer-cgminer492/blob/master/update-script/restoreConfig.sh
Is it ok if we start a new thread about cgminer 4.9.2 for Antminer S9. I think we all wrote here a lot of BS and as fast I can forgot this as better?
- Download this: https://github.com/Zwilla/bmminer-cgminer492/blob/master/update-script/AntminerS9-Update-Bmminer.tar
- Open your miners frontpage http://antminer
- Click on System
- Click on Upgrade
- Goto the menu / line "Restore backup:"
- Select the file "AntminerS9-Update-Bmminer.tar"
- Click on the button "upload Archive"
- Wait 25 seconds
- Check MinerStatus Page
TODO: some little fixing at the api
Am I correct in thinking that this is a straight swap - bmminer for a cgminer fork - & that the strange/closed source bitmain driver code has not been altered or incorporated into the cgminer fork?
Anyone?....
https://github.com/bitmaintech/bmminer/blob/master/driver-bitmain.c
no this is cgminer driver, not linux
i can't find any linux driver structures in this git maybe they published it elsewhere or not at all
Am I correct in thinking that this is a straight swap - bmminer for a cgminer fork - & that the strange/closed source bitmain driver code has not been altered or incorporated into the cgminer fork?
Anyone?....
I was on a point to revert all my work to yesterday 9am, but now I fixed a memory leak and the new version which is at git now,
compiles perfect (with some errors) and works well.
I am confused. How can you say "compiles perfect", but still have some errors. Is that not the very definition of not compiling perfectly? What compile time errors are you getting? compiles perfect (with some errors) and works well.
last 18 hours I had many problems by fixing the compile errors at the S9 driver.
I was on a point to revert all my work to yesterday 9am, but now I fixed a memory leak and the new version which is at git now,
compiles perfect (with some errors) and works well.
The main problem was this parameter "dev->frequency" this won't set the frequency well, and the code breaks and breaks and breaks.
This var is only used to get this: "config_parameter.frequency"
I think you all can see that one of the hashboard produced more hardware errors than the other one, I think if we can change it by a new option
--bitmain-freq = this is default on every Antminer S9
--bitmain-freq1
--bitmain-freq2
--bitmain-freq3
I will try it now at first with manual values and if it works enable this option.
By this way you can handle the options.
If you not edit your bmminer.conf the software will use the default for all boards
If you set the option the software will use it.
I was on a point to revert all my work to yesterday 9am, but now I fixed a memory leak and the new version which is at git now,
compiles perfect (with some errors) and works well.
The main problem was this parameter "dev->frequency" this won't set the frequency well, and the code breaks and breaks and breaks.
This var is only used to get this: "config_parameter.frequency"
Code:
dev->freq[i] = frequency; // can we set now different frequency, if we split this?
I think you all can see that one of the hashboard produced more hardware errors than the other one, I think if we can change it by a new option
--bitmain-freq = this is default on every Antminer S9
--bitmain-freq1
--bitmain-freq2
--bitmain-freq3
I will try it now at first with manual values and if it works enable this option.
By this way you can handle the options.
If you not edit your bmminer.conf the software will use the default for all boards
If you set the option the software will use it.
How can it be proofed, that what was build is absolute the same as the compiled git?
Jump to: