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

[loshia]

Hello Con,

I might be wrong but i found a possible memleak in

void roll_work(struct work *work)...

More info

   if (work->ntime)
      work->ntime = offset_ntime(work->ntime, 1);
should be something like

   if (work->ntime) {
      char *memleak = work->ntime;
      work->ntime = offset_ntime(work->ntime, 1);
      free(memleak);
  }

as long as offset_ntime returns new pointer and old work->ntime pointer is lost and never freed on free_work call

PS: consider it as confirmed and fixed My tplinks are a lot mare happy with the above fix. Everything is working fine and most important memory is not leaking

[1638188138]

1638188138

[1638188138]

1638188138

[1638188138]

1638188138

[1638188138]

1638188138

[-ck]

Hello Con,

I might be wrong but i found a possible memleak in

void roll_work(struct work *work)...

More info

   if (work->ntime)
      work->ntime = offset_ntime(work->ntime, 1);
should be something like

   if (work->ntime) {
      char *memleak = work->ntime;
      work->ntime = offset_ntime(work->ntime, 1);
      free(memleak);
  }

as long as offset_ntime returns new pointer and old work->ntime pointer is lost and never freed on free_work call

PS: consider it as confirmed and fixed My tplinks are a lot mare happy with the above fix. Everything is working fine and most important memory is not leaking

Good catch, looks valid, thanks!

EDIT: Fixed in git, along with an opportunity for a further optimisation.

[loshia]

Hello Con,

I might be wrong but i found a possible memleak in

void roll_work(struct work *work)...

More info

   if (work->ntime)
      work->ntime = offset_ntime(work->ntime, 1);
should be something like

   if (work->ntime) {
      char *memleak = work->ntime;
      work->ntime = offset_ntime(work->ntime, 1);
      free(memleak);
  }

as long as offset_ntime returns new pointer and old work->ntime pointer is lost and never freed on free_work call

PS: consider it as confirmed and fixed My tplinks are a lot mare happy with the above fix. Everything is working fine and most important memory is not leaking

Good catch, looks valid, thanks!

EDIT: Fixed in git, along with an opportunity for a further optimisation.

thanks!!!!

[os2sam]

EDIT: Fixed in git, along with an opportunity for a further optimisation.

Any optimizations to help Block Erupters? Still not quite stable for me, 4.x that is.

[-ck]

EDIT: Fixed in git, along with an opportunity for a further optimisation.

Any optimizations to help Block Erupters? Still not quite stable for me, 4.x that is.

Oh dear, if they're not good by now, they never will be 

4.0.x had regressions but there should be no reason 4.1 is any less stable with erupters than any of the best 3.x versions.

[loshia]

EDIT: Fixed in git, along with an opportunity for a further optimisation.

Any optimizations to help Block Erupters? Still not quite stable for me, 4.x that is.

Oh dear, if they're not good by now, they never will be 

4.0.x had regressions but there should be no reason 4.1 is any less stable with erupters than any of the best 3.x versions.

By the way commit d1c5bfa5041c4e7b9bfa1be45a4d3bbe1cd45c25 is a killer - reduces cpu a lot.
Thank you VERY MUCH!!!

[os2sam]

EDIT: Fixed in git, along with an opportunity for a further optimisation.

Any optimizations to help Block Erupters? Still not quite stable for me, 4.x that is.

Oh dear, if they're not good by now, they never will be 

4.0.x had regressions but there should be no reason 4.1 is any less stable with erupters than any of the best 3.x versions.

Well, I think I may have used an inaccurate phrase in "not quite stable".

It seems to run and doesn't crash.  But it does turn off several erupters on both machines I use, erupters that run just fine on 3.8.4 and 3.11.0.

[techman05]

I only have 5 eroupters  and they seem to go forever. I haven't had a constant green light in a while and I've done each update with precomplied versions. I think the only time mine go visibly green is probably when a new block is found.

On another note I'll miss that green light when I finaly upgrade them out. Its the only way to know when things crap out without checking the cgminer program. Also not if someone does put a green light of death on their next device "Put a dimmer control"

[Buchi-88]

The new CGMiner also support the new Antminer U2?

I mean yes, only the Heatsink is new?

regards

Try it and let us know
We don't have any U2's but I think they are the same with just a better heat sink?

Hello,

today i obtained my new Antminer U1 batch, so i see they are HW Revision ANT USB V1.2 so the same Boardas the U2 but with the small Heatsink.

And they run 1A at Cgminer 4.1.0 with 2 GH/s so thanks

regards

[techman05]

Thats good you got 1.2 boards but I do I search on ebay and people still have pictures of the u1 advertising as U2's so potentially room for scams.

Have you tried to push it past 2GH? I'd like to think if the U2s start at 2GH maybe you can get 2.5 or a lil bit more since they're supposed to have the updated resistors.


Pics for the lookers
https://bitcointalk.org/index.php?topic=490559.msg5526949#msg5526949

[deagel]

hi,
how can i update the cgminer version from a antminer s1? Normally the software is deployed with a new firmware.

thanks

hi,
any feedback?

[deagel]

hi,
how can i update the cgminer version from a antminer s1? Normally the software is deployed with a new firmware.

thanks

Yeah I can build binaries and ipk files now.
I'm not 100% sure of building a complete firmware file, that will come a little later.

The luci web code that displays the stats doesn't work with the correct API code, so I'll see if I can sort that out today and hack a binary that works with both the luci code and of course includes the correct API output for anyone who accesses it.

hi,
thats sounds good

[kano]

hi,
how can i update the cgminer version from a antminer s1? Normally the software is deployed with a new firmware.

thanks

Yeah I can build binaries and ipk files now.
I'm not 100% sure of building a complete firmware file, that will come a little later.

The luci web code that displays the stats doesn't work with the correct API code, so I'll see if I can sort that out today and hack a binary that works with both the luci code and of course includes the correct API output for anyone who accesses it.

hi,
thats sounds good

Yeah, but I've not got that sorted out yet.
(My first attempt was a guess and it didn't solve luci not displaying the first two sections)
I'll get it soon ...

[PatMan]

hi,
how can i update the cgminer version from a antminer s1? Normally the software is deployed with a new firmware.

thanks

Yeah I can build binaries and ipk files now.
I'm not 100% sure of building a complete firmware file, that will come a little later.

The luci web code that displays the stats doesn't work with the correct API code, so I'll see if I can sort that out today and hack a binary that works with both the luci code and of course includes the correct API output for anyone who accesses it.

hi,
thats sounds good

Yeah, but I've not got that sorted out yet.
(My first attempt was a guess and it didn't solve luci not displaying the first two sections)
I'll get it soon ...

Good man kano.

A bit off topic, but have you or con played with the new 14.04 LTS (beta) release yet? I've had it on one of my laptops for a week or so & it seems to run sweet. Was wondering if you guys have looked at it yet & if cgminer will work with it "as is".

Thanks & Peace.

[freebit13]

Yeah, but I've not got that sorted out yet.
(My first attempt was a guess and it didn't solve luci not displaying the first two sections)
I'll get it soon ...

Thank you Kano... Looking forward to it! Token of appreciation sent to you and CK for all the work you guys put in.

[Buddy]

I'm using the cgminer with gridseed support, I have one main issue which is really annoying can you tell me if it's fixed and which version it's fixed on. The problem is that I have to unplug the USB for my gridseed before cgminer starts, once cgminer has started only then I can plug the USB in otherwise the cgminer just freezes or crashes. If I'm accessing the rig remotely and I need to restart the windows computer then I'm stuck as some one physically needs to be at the rig.

Thank you guys

[-ck]

I'm using the cgminer with gridseed support, I have one main issue which is really annoying can you tell me if it's fixed and which version it's fixed on. The problem is that I have to unplug the USB for my gridseed before cgminer starts, once cgminer has started only then I can plug the USB in otherwise the cgminer just freezes or crashes. If I'm accessing the rig remotely and I need to restart the windows computer then I'm stuck as some one physically needs to be at the rig.

Thank you guys

Official cgminer contains no gridseed support so you have to seek support from gridseed or whoever is maintaining the cgminer fork with gridseed support.

[RowanX]

In CGMiner, you can disable output by pressing [d] (display), then s (silent) ... but how to you re-enable output? After switching to silent mode I can't seem to get back into the display options. Just wondering how its possible without having to quit and restart the software.

[-ck]

In CGMiner, you can disable output by pressing [d] (display), then s (silent) ... but how to you re-enable output? After switching to silent mode I can't seem to get back into the display options. Just wondering how its possible without having to quit and restart the software.

You cannot, it's a permanent shutdown of the display.

[kano]

...
A bit off topic, but have you or con played with the new 14.04 LTS (beta) release yet? I've had it on one of my laptops for a week or so & it seems to run sweet. Was wondering if you guys have looked at it yet & if cgminer will work with it "as is".

Thanks & Peace.

Not gone near 14.04 yet.
I tried 13.10 on a new i7 4771 the other week and it wouldn't boot fully
(Been sitting idle for 2 weeks now)
Maybe I'll have to try that then to get it working ...