Bitcoin Forum
December 14, 2018, 06:42:50 PM *
News: Latest Bitcoin Core release: 0.17.0 [Torrent].
 
   Home   Help Search Login Register More  
Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 [13] 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 ... 191 »
  Print  
Author Topic: [ANN]: cpuminer-opt v3.8.8.1, open source optimized multi-algo CPU miner  (Read 417449 times)
pallas
Legendary
*
Offline Offline

Activity: 1862
Merit: 1087


Black Belt Developer


View Profile
January 29, 2016, 08:56:55 AM
 #241

It didn't take long to hit a brick wall with windows.

static inline void transform(cubehashParm *sp )

Expected '(' to follow 'inline'

WTF?

 I guess that settles it.

visual studio or mingw?

1544812970
Hero Member
*
Offline Offline

Posts: 1544812970

View Profile Personal Message (Offline)

Ignore
1544812970
Reply with quote  #2

1544812970
Report to moderator
1544812970
Hero Member
*
Offline Offline

Posts: 1544812970

View Profile Personal Message (Offline)

Ignore
1544812970
Reply with quote  #2

1544812970
Report to moderator
1544812970
Hero Member
*
Offline Offline

Posts: 1544812970

View Profile Personal Message (Offline)

Ignore
1544812970
Reply with quote  #2

1544812970
Report to moderator
100% New Software
PC, Mac, Android, & HTML5 Clients
Krill Rakeback
Low Rake
Bitcoin Poker 3.0
Bad Beat Jackpot
SwC Poker Relaunch
PLAY NOW
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction. Advertise here.
1544812970
Hero Member
*
Offline Offline

Posts: 1544812970

View Profile Personal Message (Offline)

Ignore
1544812970
Reply with quote  #2

1544812970
Report to moderator
joblo
Legendary
*
Offline Offline

Activity: 1148
Merit: 1050


View Profile
January 29, 2016, 09:50:25 AM
 #242

It didn't take long to hit a brick wall with windows.

static inline void transform(cubehashParm *sp )

Expected '(' to follow 'inline'

WTF?

 I guess that settles it.

visual studio or mingw?

VS. It worked for me for ccminer so I was hoping if I got the project file right it would work.
I was very careful with it matching the file list side by side with the makefile. There were
windows versions of all the miners I harvested from but I didn't compile any of them
myself.

I'll try mingw again, this time in a VM

cpuminer-opt developer, https://bitcointalk.org/index.php?topic=1326803.0
BTC: 12tdvfF7KmAsihBXQXynT6E6th2c2pByTT,
ETH: 0x72122edabcae9d3f57eab0729305a425f6fef6d0
djm34
Legendary
*
Offline Offline

Activity: 1274
Merit: 1040


View Profile WWW
January 29, 2016, 10:00:15 AM
 #243

It didn't take long to hit a brick wall with windows.

static inline void transform(cubehashParm *sp )

Expected '(' to follow 'inline'

WTF?

 I guess that settles it.

visual studio or mingw?

VS. It worked for me for ccminer so I was hoping if I got the project file right it would work.
I was very careful with it matching the file list side by side with the makefile. There were
windows versions of all the miners I harvested from but I didn't compile any of them
myself.

I'll try mingw again, this time in a VM

Epsylon3 version of cpuminer compiles well with visual studio

djm34 facebook page
BTC: 1NENYmxwZGHsKFmyjTc5WferTn5VTFb7Ze
Pledge for neoscrypt ccminer to that address: 16UoC4DmTz2pvhFvcfTQrzkPTrXkWijzXw
joblo
Legendary
*
Offline Offline

Activity: 1148
Merit: 1050


View Profile
January 29, 2016, 10:12:54 AM
 #244

It didn't take long to hit a brick wall with windows.

static inline void transform(cubehashParm *sp )

Expected '(' to follow 'inline'

WTF?

 I guess that settles it.

visual studio or mingw?

VS. It worked for me for ccminer so I was hoping if I got the project file right it would work.
I was very careful with it matching the file list side by side with the makefile. There were
windows versions of all the miners I harvested from but I didn't compile any of them
myself.

I'll try mingw again, this time in a VM

Epsylon3 version of cpuminer compiles well with visual studio

I'll try compiling darkcoin minerd and cp3u. If they don't both compile then I know I have a bigger
problem.

cpuminer-opt developer, https://bitcointalk.org/index.php?topic=1326803.0
BTC: 12tdvfF7KmAsihBXQXynT6E6th2c2pByTT,
ETH: 0x72122edabcae9d3f57eab0729305a425f6fef6d0
zTheWolfz
Full Member
***
Offline Offline

Activity: 232
Merit: 150



View Profile
January 29, 2016, 01:22:55 PM
 #245

It didn't take long to hit a brick wall with windows.

static inline void transform(cubehashParm *sp )

Expected '(' to follow 'inline'

WTF?

 I guess that settles it.

visual studio or mingw?

VS. It worked for me for ccminer so I was hoping if I got the project file right it would work.
I was very careful with it matching the file list side by side with the makefile. There were
windows versions of all the miners I harvested from but I didn't compile any of them
myself.

I'll try mingw again, this time in a VM

Epsylon3 version of cpuminer compiles well with visual studio

I'll try compiling darkcoin minerd and cp3u. If they don't both compile then I know I have a bigger
problem.

Not sure if there is a un-compiled version or Linux version of cpuminer_x64_SSE2, but the .exe of this
will mine x11 & a few other coins in Windows7-x64. No AES support, none of the cpuminer_x64_avx flavors
worked on any of my rigs, unsupported hardware? Using the cpuminer_x64_sse2 on the AMD rig ATM mining x11.
AMD rig has Server R2012 R2 as a OS. 42 day trail left on that. lol

Edit: On another note I have a old AMD FX duel CPU motherboard & hardware setting collecting dust if I can fine the time
from RL I'll setup a test bed with that "IF" I can find a spare PSU to use with it & a spot somewhere to place it. lol
joblo
Legendary
*
Offline Offline

Activity: 1148
Merit: 1050


View Profile
January 29, 2016, 04:30:56 PM
 #246


Not sure if there is a un-compiled version or Linux version of cpuminer_x64_SSE2, but the .exe of this
will mine x11 & a few other coins in Windows7-x64. No AES support, none of the cpuminer_x64_avx flavors
worked on any of my rigs, unsupported hardware? Using the cpuminer_x64_sse2 on the AMD rig ATM mining x11.
AMD rig has Server R2012 R2 as a OS. 42 day trail left on that. lol

Edit: On another note I have a old AMD FX duel CPU motherboard & hardware setting collecting dust if I can fine the time
from RL I'll setup a test bed with that "IF" I can find a spare PSU to use with it & a spot somewhere to place it. lol

I don't follow your first paragraph. Are you talking about another fork? What is cpuminer_x64_sse2?

cpuminer-opt developer, https://bitcointalk.org/index.php?topic=1326803.0
BTC: 12tdvfF7KmAsihBXQXynT6E6th2c2pByTT,
ETH: 0x72122edabcae9d3f57eab0729305a425f6fef6d0
zTheWolfz
Full Member
***
Offline Offline

Activity: 232
Merit: 150



View Profile
January 29, 2016, 04:51:15 PM
 #247


Not sure if there is a un-compiled version or Linux version of cpuminer_x64_SSE2, but the .exe of this
will mine x11 & a few other coins in Windows7-x64. No AES support, none of the cpuminer_x64_avx flavors
worked on any of my rigs, unsupported hardware? Using the cpuminer_x64_sse2 on the AMD rig ATM mining x11.
AMD rig has Server R2012 R2 as a OS. 42 day trail left on that. lol

Edit: On another note I have a old AMD FX duel CPU motherboard & hardware setting collecting dust if I can fine the time
from RL I'll setup a test bed with that "IF" I can find a spare PSU to use with it & a spot somewhere to place it. lol

I don't follow your first paragraph. Are you talking about another fork? What is cpuminer_x64_sse2?
Its found in the \NiceHashMiner_v1.2.2.2\bin folder after unpacked

runline: cpuminer_x64_SSE2 -a x11 -o stratum+tcp://x11.usa.nicehash.com:3336 -u 18tvS3deKZK5q4eTtPRWYeEMWmGmuErHgz.H8QG6 -p d=0.01 -t 32



Edit: Apparently is seems to be cpuminer-muti 1.1 renamed.? My guess is Nichhash software conf line makes it use sse2 when ran with it,
but with no options for any of the GPU algos to be mined on the CPU. Since I don't see sse2 being used in my run or maybe sse2 is used but just not printed as in use.

 
joblo
Legendary
*
Offline Offline

Activity: 1148
Merit: 1050


View Profile
January 29, 2016, 07:00:06 PM
 #248


Not sure if there is a un-compiled version or Linux version of cpuminer_x64_SSE2, but the .exe of this
will mine x11 & a few other coins in Windows7-x64. No AES support, none of the cpuminer_x64_avx flavors
worked on any of my rigs, unsupported hardware? Using the cpuminer_x64_sse2 on the AMD rig ATM mining x11.
AMD rig has Server R2012 R2 as a OS. 42 day trail left on that. lol

Edit: On another note I have a old AMD FX duel CPU motherboard & hardware setting collecting dust if I can fine the time
from RL I'll setup a test bed with that "IF" I can find a spare PSU to use with it & a spot somewhere to place it. lol

I don't follow your first paragraph. Are you talking about another fork? What is cpuminer_x64_sse2?
Its found in the \NiceHashMiner_v1.2.2.2\bin folder after unpacked

runline: cpuminer_x64_SSE2 -a x11 -o stratum+tcp://x11.usa.nicehash.com:3336 -u 18tvS3deKZK5q4eTtPRWYeEMWmGmuErHgz.H8QG6 -p d=0.01 -t 32



Edit: Apparently is seems to be cpuminer-muti 1.1 renamed.? My guess is Nichhash software conf line makes it use sse2 when ran with it,
but with no options for any of the GPU algos to be mined on the CPU. Since I don't see sse2 being used in my run or maybe sse2 is used but just not printed as in use.

 

OK, that's not mine, you'll have to talk to nicehash about that.

cpuminer-opt developer, https://bitcointalk.org/index.php?topic=1326803.0
BTC: 12tdvfF7KmAsihBXQXynT6E6th2c2pByTT,
ETH: 0x72122edabcae9d3f57eab0729305a425f6fef6d0
zTheWolfz
Full Member
***
Offline Offline

Activity: 232
Merit: 150



View Profile
January 29, 2016, 08:13:44 PM
 #249

Wasn't asking for help with it was just bringing it to you attention in case you didn't know about it.
It and cpumimer-muti both run GPU algos with my limited testing with each on Intel & AMD CPU's in windows no VM or 2nd OS needed.

I did compile 3.07 on 2 Intel rigs & the AMD rig, still no luck with the AMD, only seeing a 20% load max no matter the core amount used.
Intel on the other hand runs about the same as 3.02 that was in use, 3.0.7 from a quick look was just a tad lower on hashrate than 3.0.2.
But that was just the start up numbers, long term that could change, will have to check it later once its been working a while.
joblo
Legendary
*
Offline Offline

Activity: 1148
Merit: 1050


View Profile
January 29, 2016, 09:22:41 PM
 #250

Wasn't asking for help with it was just bringing it to you attention in case you didn't know about it.
It and cpumimer-muti both run GPU algos with my limited testing with each on Intel & AMD CPU's in windows no VM or 2nd OS needed.

I did compile 3.07 on 2 Intel rigs & the AMD rig, still no luck with the AMD, only seeing a 20% load max no matter the core amount used.
Intel on the other hand runs about the same as 3.02 that was in use, 3.0.7 from a quick look was just a tad lower on hashrate than 3.0.2.
But that was just the start up numbers, long term that could change, will have to check it later once its been working a while.

I'm confused. Your first paragraph says you aren't asking for help and the AMD rig is working.
But the second one is saying you are having a problems with 3.0.7 and the AMD rig isn't working.

Is the problem with cpuminer-opt or the rig?

Thanks for the info abnout cpumier-multi. It seems to claim SSE2 optimizations so how does it
compare with cpuminer-opt? Maybe they did a better job.

You're probably right about 3.0.2 performance. I'm reviewing the performance of all releases because
I think some of my tweaks actually hurt performance.

Thanks for the extensive testing.

cpuminer-opt developer, https://bitcointalk.org/index.php?topic=1326803.0
BTC: 12tdvfF7KmAsihBXQXynT6E6th2c2pByTT,
ETH: 0x72122edabcae9d3f57eab0729305a425f6fef6d0
zTheWolfz
Full Member
***
Offline Offline

Activity: 232
Merit: 150



View Profile
January 29, 2016, 10:24:15 PM
 #251

Wasn't asking for help with it was just bringing it to you attention in case you didn't know about it.
It and cpumimer-muti both run GPU algos with my limited testing with each on Intel & AMD CPU's in windows no VM or 2nd OS needed.

I did compile 3.07 on 2 Intel rigs & the AMD rig, still no luck with the AMD, only seeing a 20% load max no matter the core amount used.
Intel on the other hand runs about the same as 3.02 that was in use, 3.0.7 from a quick look was just a tad lower on hashrate than 3.0.2.
But that was just the start up numbers, long term that could change, will have to check it later once its been working a while.

I'm confused. Your first paragraph says you aren't asking for help and the AMD rig is working.
But the second one is saying you are having a problems with 3.0.7 and the AMD rig isn't working.

Is the problem with cpuminer-opt or the rig?

Thanks for the info abnout cpumier-multi or . It seems to claim SSE2 optimizations so how does it
compare with cpuminer-opt? Maybe they did a better job.

You're probably right about 3.0.2 performance. I'm reviewing the performance of all releases because
I think some of my tweaks actually hurt performance.

Thanks for the extensive testing.

AMD is working with cpuminer-muti or cpuminer_x64_SSE2 in windows, in VM your versions doesn't work for the AMD rig.
Yours loads now and appears to be working but at a low load rate "20% or less" and I never see any shares completed with the amount of time I've let it run.
Over 2 hours nothing.
joblo
Legendary
*
Offline Offline

Activity: 1148
Merit: 1050


View Profile
January 30, 2016, 01:19:36 AM
 #252

Wasn't asking for help with it was just bringing it to you attention in case you didn't know about it.
It and cpumimer-muti both run GPU algos with my limited testing with each on Intel & AMD CPU's in windows no VM or 2nd OS needed.

I did compile 3.07 on 2 Intel rigs & the AMD rig, still no luck with the AMD, only seeing a 20% load max no matter the core amount used.
Intel on the other hand runs about the same as 3.02 that was in use, 3.0.7 from a quick look was just a tad lower on hashrate than 3.0.2.
But that was just the start up numbers, long term that could change, will have to check it later once its been working a while.

I'm confused. Your first paragraph says you aren't asking for help and the AMD rig is working.
But the second one is saying you are having a problems with 3.0.7 and the AMD rig isn't working.

Is the problem with cpuminer-opt or the rig?

Thanks for the info abnout cpumier-multi or . It seems to claim SSE2 optimizations so how does it
compare with cpuminer-opt? Maybe they did a better job.

You're probably right about 3.0.2 performance. I'm reviewing the performance of all releases because
I think some of my tweaks actually hurt performance.

Thanks for the extensive testing.

AMD is working with cpuminer-muti or cpuminer_x64_SSE2 in windows, in VM your versions doesn't work for the AMD rig.
Yours loads now and appears to be working but at a low load rate "20% or less" and I never see any shares completed with the amount of time I've let it run.
Over 2 hours nothing.

OK so there is a problem with cpuminer-opt in a Linux VM on your AMD rig but the other miners work ok
natively in Windows on the same machine? Is this correct?

How does the performance compare between cpuminer-multi and cpuminer_x64_SSE2. It has SSE2 in
the name but does it have the performance? It should be significantly faster than than multi.

Edit and another thing you could try the other miners in the VM?This eliminates another variable. Thanks.
When did it break for cpuminer-opt?

cpuminer-opt developer, https://bitcointalk.org/index.php?topic=1326803.0
BTC: 12tdvfF7KmAsihBXQXynT6E6th2c2pByTT,
ETH: 0x72122edabcae9d3f57eab0729305a425f6fef6d0
zTheWolfz
Full Member
***
Offline Offline

Activity: 232
Merit: 150



View Profile
January 30, 2016, 01:53:04 AM
 #253

Wasn't asking for help with it was just bringing it to you attention in case you didn't know about it.
It and cpumimer-muti both run GPU algos with my limited testing with each on Intel & AMD CPU's in windows no VM or 2nd OS needed.

I did compile 3.07 on 2 Intel rigs & the AMD rig, still no luck with the AMD, only seeing a 20% load max no matter the core amount used.
Intel on the other hand runs about the same as 3.02 that was in use, 3.0.7 from a quick look was just a tad lower on hashrate than 3.0.2.
But that was just the start up numbers, long term that could change, will have to check it later once its been working a while.

I'm confused. Your first paragraph says you aren't asking for help and the AMD rig is working.
But the second one is saying you are having a problems with 3.0.7 and the AMD rig isn't working.

Is the problem with cpuminer-opt or the rig?

Thanks for the info abnout cpumier-multi or . It seems to claim SSE2 optimizations so how does it
compare with cpuminer-opt? Maybe they did a better job.

You're probably right about 3.0.2 performance. I'm reviewing the performance of all releases because
I think some of my tweaks actually hurt performance.

Thanks for the extensive testing.

AMD is working with cpuminer-muti or cpuminer_x64_SSE2 in windows, in VM your versions doesn't work for the AMD rig.
Yours loads now and appears to be working but at a low load rate "20% or less" and I never see any shares completed with the amount of time I've let it run.
Over 2 hours nothing.

OK so there is a problem with cpuminer-opt in a Linux VM on your AMD rig but the other miners work ok
natively in Windows on the same machine? Is this correct?

How does the performance compare between cpuminer-multi and cpuminer_x64_SSE2. It has SSE2 in
the name but does it have the performance? It should be significantly faster than than multi.

Edit and another thing you could try the other miners in the VM?This eliminates another variable. Thanks.
When did it break for cpuminer-opt?
#1 yes

#2 Haven't had much of a change to take a close look at those numbers yet.
RL is being a pain atm.

#3 When I get the time I'll be more then glad to, I did try or tried to compile cpuminer-multi  with no luck in the VM so far.
No source for cpuminer_x64_SSE2 to compile that I'm aware of.

#4 hasn't never worked for the AMD rig, but by this time next week with any luck I'll have my new CPU's up and running so it will no longer matter.
I will try to get you as much info as I can if you'd like that before the change over, but looks like I maybe the only one running this type of setup.
So that may not help anyone, but I willing to see what I can dig up.
joblo
Legendary
*
Offline Offline

Activity: 1148
Merit: 1050


View Profile
January 30, 2016, 06:45:49 AM
 #254

The following may be of interest to devs of cpuminer, ccminer and maybe others.

Update:

More architectural changes made. Most declarations moved from cpu-miner.c
to miner.h so no files will need to include it. The system I am famiiar with could prevent
being embedded adn it was the default. The owner had to explicitly allow  inclusion
and to whom. Nice feature.

I still have to migrate full_test and get_opt to use the gate. I will have to split up the options
into common options and algo specific options. cpuminer-c will handler the common options
and the algos will handle the algo-specific options that apply to them.

I will also look into enhancing the registration process to have the algo and its custom
options added to the argument list and help file. These are the only remaining holes in the gate,
to my knowledge.

The gate is almost complete. It's called a gate because it is the only path between cpu-miner.c
and the algos. A very popular technique where I come from.

A developpers preview should be avilable in a couple of days.

I am hopeful the rearchitecting will help solve some of the other nagging issues I have been having,
like flipping between undefined and multi defined.

Previous post:
--------
I've done a lot of rebuilding under the hood because  miner_thread had become
polluted with too much algo specific code. Well, that is all gone.
Folllowing is a summary of the changes. I will make a preview available before release.

- restructured the file system to be more modular with respect to algos

- removed bloat added to support multiple CPU targets.

- removed all algo specific code from miner.h and cpu-miner.c, and eventually util.c
   and anywhere else I find it, it all lives now in the algo's source file.

- adding algos doesn't require any changes to base code, a registry interface file requires adding one entry.

- redesigned the scanhash engine to use function pointers instead of switch/case.

- removed many long switch/case statements.

- created a registry where algos register their functions with cpu-miner.

- registry functions include scanhash, hash, init_ctx, get_max64, and get_custom_opts with plans
  to add full_test and any others I find.

- enabled initialization of algo context outside of miner_thread loop.

- changes do not increase performance directly but enables more efficient management of contexts
  which helps certain algos, specificaly quark.

cpuminer-opt developer, https://bitcointalk.org/index.php?topic=1326803.0
BTC: 12tdvfF7KmAsihBXQXynT6E6th2c2pByTT,
ETH: 0x72122edabcae9d3f57eab0729305a425f6fef6d0
pallas
Legendary
*
Offline Offline

Activity: 1862
Merit: 1087


Black Belt Developer


View Profile
January 30, 2016, 09:30:16 AM
 #255

Nice work! I like clean and modular code!

joblo
Legendary
*
Offline Offline

Activity: 1148
Merit: 1050


View Profile
January 30, 2016, 02:28:40 PM
 #256

Nice work! I like clean and modular code!

Thanks, I haven't yet found a way to run multiple entry functions upon program load..
The system I'm familiar with could declare any file with a main function to have it be run
automaticaaly when the program is loaded. Essentially it functions similarly to a OS init
system but for an application. If I can do that in c/c++ I could plug the only remaining
crack in the wall between algo code and base code. If I can't do it  from within the
algo file I might have to write an init system.

cpuminer-opt developer, https://bitcointalk.org/index.php?topic=1326803.0
BTC: 12tdvfF7KmAsihBXQXynT6E6th2c2pByTT,
ETH: 0x72122edabcae9d3f57eab0729305a425f6fef6d0
joblo
Legendary
*
Offline Offline

Activity: 1148
Merit: 1050


View Profile
January 30, 2016, 03:06:32 PM
 #257

Wasn't asking for help with it was just bringing it to you attention in case you didn't know about it.
It and cpumimer-muti both run GPU algos with my limited testing with each on Intel & AMD CPU's in windows no VM or 2nd OS needed.

I did compile 3.07 on 2 Intel rigs & the AMD rig, still no luck with the AMD, only seeing a 20% load max no matter the core amount used.
Intel on the other hand runs about the same as 3.02 that was in use, 3.0.7 from a quick look was just a tad lower on hashrate than 3.0.2.
But that was just the start up numbers, long term that could change, will have to check it later once its been working a while.

I'm confused. Your first paragraph says you aren't asking for help and the AMD rig is working.
But the second one is saying you are having a problems with 3.0.7 and the AMD rig isn't working.

Is the problem with cpuminer-opt or the rig?

Thanks for the info abnout cpumier-multi or . It seems to claim SSE2 optimizations so how does it
compare with cpuminer-opt? Maybe they did a better job.

You're probably right about 3.0.2 performance. I'm reviewing the performance of all releases because

I think some of my tweaks actually hurt performance.

Thanks for the extensive testing.

AMD is working with cpuminer-muti or cpuminer_x64_SSE2 in windows, in VM your versions doesn't work for the AMD rig.
Yours loads now and appears to be working but at a low load rate "20% or less" and I never see any shares completed with the amount of time I've let it run.
Over 2 hours nothing.

OK so there is a problem with cpuminer-opt in a Linux VM on your AMD rig but the other miners work ok
natively in Windows on the same machine? Is this correct?

How does the performance compare between cpuminer-multi and cpuminer_x64_SSE2. It has SSE2 in
the name but does it have the performance? It should be significantly faster than than multi.

Edit and another thing you could try the other miners in the VM?This eliminates another variable. Thanks.
When did it break for cpuminer-opt?
#1 yes

#2 Haven't had much of a change to take a close look at those numbers yet.
RL is being a pain atm.

#3 When I get the time I'll be more then glad to, I did try or tried to compile cpuminer-multi  with no luck in the VM so far.
No source for cpuminer_x64_SSE2 to compile that I'm aware of.

#4 hasn't never worked for the AMD rig, but by this time next week with any luck I'll have my new CPU's up and running so it will no longer matter.
I will try to get you as much info as I can if you'd like that before the change over, but looks like I maybe the only one running this type of setup.
So that may not help anyone, but I willing to see what I can dig up.

No pressure, you can test whver you pleae or not at all. I'm very appreciative of everything you've done so far.

It appears the VM is the issue. I don't know why it woulg connect to stratum and get new work but never
start hashing. At least I know where to start looking.

I can test cpuminer_x64_SSE2 myself, it was just a curiosity.

Many thanks.

cpuminer-opt developer, https://bitcointalk.org/index.php?topic=1326803.0
BTC: 12tdvfF7KmAsihBXQXynT6E6th2c2pByTT,
ETH: 0x72122edabcae9d3f57eab0729305a425f6fef6d0
joblo
Legendary
*
Offline Offline

Activity: 1148
Merit: 1050


View Profile
January 30, 2016, 04:17:20 PM
 #258

How to use algo gate, for devs.

algo gate is the new miner_thread engine for running an algo in the correct environment.

It makes adding new algos easy. Here's how:

1. define the usual functions in the algo's source file. scanhash and hash functions are mandatory

2. Define a registration function

Code:
bool register_x11_algo( algo_gate_t* gate )
{
  gate->init_ctx               = &init_x11_aes_ctx;
  gate->scanhash            = &scanhash_x11_aes;
  gate->hash                    = &x11_hash_aes;
// gate->get_custom_opt = &x11_get_custom_opt;
// gate->get_max64         = &x11_get_max64;
  return true;
};

3. add "#include "algo-gate-api.h" to the algo's source file.

4. add a new entry to initialize_algo_gate in algo-gate-api.c to register your algo

Code:
     case ALGO_X11:
        register_x11_algo( &algo_gate );
        break;

5. Compile and run. When cpuminer starts it will use the algo registry
    without needing to know anything about the algo except it's name.

That's all there is to it. Any miner using the Garzick/Pooler front end can implement this.
The only pain is retrofitting all the existing algos.


More to come



cpuminer-opt developer, https://bitcointalk.org/index.php?topic=1326803.0
BTC: 12tdvfF7KmAsihBXQXynT6E6th2c2pByTT,
ETH: 0x72122edabcae9d3f57eab0729305a425f6fef6d0
zTheWolfz
Full Member
***
Offline Offline

Activity: 232
Merit: 150



View Profile
January 30, 2016, 05:55:55 PM
 #259

Great work.

Well looks like I will be doing that upgrade sooner than I thought. Seller shipped my stuff
usps 2 day priority showing it should be here today in a matter of hour actually, Postman runs early on Sat.   Cool
zTheWolfz
Full Member
***
Offline Offline

Activity: 232
Merit: 150



View Profile
January 30, 2016, 07:39:10 PM
 #260

Okay here is were I am: Installed Ubuntu on a small 64GB SSD I had laying around.
Everything looks to have gone fine with the compile of v3.0.7. But

I'm getting this below once mining is started.

Illegal instruction (core dumped)
then the miner dies...
Code:
thewolf@thewolf-H8QG6:~/Downloads/cpuminer-opt-3.0.7$ ./cpuminer -a x11 -o stratum+tcp://x11.usa.nicehash.com:3336 -u 18tvS3deKZK5q4eTtPRWYeEMWmGmuErHgz.H8QG6 -p x

         **********  cpuminer-opt  v3.0.7  ***********
     BTC donation address: 12tdvfF7KmAsihBXQXynT6E6th2c2pByTT
     Forked from TPruvot's cpuminer-multi-1.2pre with credits
     to Lucas Jones, elmad, palmd, djm34, pooler, ig0tik3d,
     Wolf0 and Jeff Garzik.

Checking CPU capatibility...
        AMD Opteron(tm) Processor 6124 HE
CPU & SW Build Supports AES_NI: NO
[2016-01-30 13:34:37] Starting Stratum on stratum+tcp://x11.usa.nicehash.com:3336
[2016-01-30 13:34:37] 32 miner threads started, using 'x11' algorithm.
[2016-01-30 13:34:37] Stratum difficulty set to 0.04
[2016-01-30 13:34:37] x11.usa.nicehash.com:3336 x11 block 662269
Illegal instruction (core dumped)
thewolf@thewolf-H8QG6:~/Downloads/cpuminer-opt-3.0.7$

On zpool.ca  & xpool.ca I get  the below.

Code:

thewolf@thewolf-H8QG6:~/Downloads/cpuminer-opt-3.0.7$ ./cpuminer -o stratum+tcp://mine.zpool.ca:3533 -u 18tvS3deKZK5q4eTtPRWYeEMWmGmuErHgz -p x -t 16

         **********  cpuminer-opt  v3.0.7  ***********
     BTC donation address: 12tdvfF7KmAsihBXQXynT6E6th2c2pByTT
     Forked from TPruvot's cpuminer-multi-1.2pre with credits
     to Lucas Jones, elmad, palmd, djm34, pooler, ig0tik3d,
     Wolf0 and Jeff Garzik.

Checking CPU capatibility...
        AMD Opteron(tm) Processor 6124 HE
CPU & SW Build Supports AES_NI: NO
[2016-01-30 13:54:17] Starting Stratum on stratum+tcp://mine.zpool.ca:3533
[2016-01-30 13:54:17] 16 miner threads started, using 'scrypt' algorithm.
[2016-01-30 13:54:18] Stratum difficulty set to 0.016
[2016-01-30 13:54:18] mine.zpool.ca:3533 scrypt block 577802
[2016-01-30 13:54:18] CPU #8: 4.50 kH/s
[2016-01-30 13:54:18] CPU #0: 4.49 kH/s
[2016-01-30 13:54:18] CPU #14: 4.46 kH/s
[2016-01-30 13:54:18] accepted: 0/1 (0.00%), 13.44 kH/s nooooo
[2016-01-30 13:54:18] reject reason: Invalid share
[2016-01-30 13:54:18] CPU #0: 4.42 kH/s
[2016-01-30 13:54:18] CPU #11: 4.50 kH/s
[2016-01-30 13:54:18] CPU #1: 4.49 kH/s
[2016-01-30 13:54:19] CPU #3: 4.50 kH/s
[2016-01-30 13:54:19] CPU #4: 4.50 kH/s
[2016-01-30 13:54:19] CPU #2: 4.24 kH/s
[2016-01-30 13:54:19] CPU #6: 4.50 kH/s
[2016-01-30 13:54:19] CPU #5: 4.33 kH/s
[2016-01-30 13:54:19] CPU #7: 4.47 kH/s
[2016-01-30 13:54:19] CPU #9: 4.50 kH/s
[2016-01-30 13:54:19] CPU #5: 4.42 kH/s
[2016-01-30 13:54:18] reject reason: Invalid share
[2016-01-30 13:54:18] CPU #0: 4.42 kH/s
[2016-01-30 13:54:18] CPU #11: 4.50 kH/s
[2016-01-30 13:54:18] CPU #1: 4.49 kH/s
[2016-01-30 13:54:19] CPU #3: 4.50 kH/s
[2016-01-30 13:54:19] CPU #4: 4.50 kH/s
[2016-01-30 13:54:19] CPU #2: 4.24 kH/s
[2016-01-30 13:54:19] CPU #6: 4.50 kH/s
[2016-01-30 13:54:19] CPU #5: 4.33 kH/s
[2016-01-30 13:54:19] CPU #7: 4.47 kH/s
[2016-01-30 13:54:19] CPU #9: 4.50 kH/s
[2016-01-30 13:54:19] CPU #5: 4.42 kH/s
[2016-01-30 13:54:18] CPU #14: 4.46 kH/s
[2016-01-30 13:54:18] accepted: 0/1 (0.00%), 13.44 kH/s nooooo
[2016-01-30 13:54:18] reject reason: Invalid share
[2016-01-30 13:54:18] CPU #0: 4.42 kH/s
[2016-01-30 13:54:18] CPU #11: 4.50 kH/s
[2016-01-30 13:54:18] CPU #1: 4.49 kH/s
[2016-01-30 13:54:19] CPU #3: 4.50 kH/s
[2016-01-30 13:54:19] CPU #4: 4.50 kH/s
[2016-01-30 13:54:19] CPU #2: 4.24 kH/s
[2016-01-30 13:54:19] CPU #6: 4.50 kH/s
[2016-01-30 13:54:19] CPU #5: 4.33 kH/s
[2016-01-30 13:54:19] CPU #7: 4.47 kH/s
[2016-01-30 13:54:19] CPU #9: 4.50 kH/s
[2016-01-30 13:54:19] CPU #5: 4.42 kH/s
[2016-01-30 13:54:18] CPU #11: 4.50 kH/s
[2016-01-30 13:54:18] CPU #1: 4.49 kH/s
[2016-01-30 13:54:19] CPU #3: 4.50 kH/s
[2016-01-30 13:54:19] CPU #4: 4.50 kH/s
[2016-01-30 13:54:19] CPU #2: 4.24 kH/s
[2016-01-30 13:54:19] CPU #6: 4.50 kH/s
[2016-01-30 13:54:19] CPU #5: 4.33 kH/s
[2016-01-30 13:54:19] CPU #7: 4.47 kH/s
[2016-01-30 13:54:19] CPU #9: 4.50 kH/s
[2016-01-30 13:54:19] CPU #5: 4.42 kH/s
[2016-01-30 13:54:19] CPU #15: 4.50 kH/s
[2016-01-30 13:54:19] accepted: 0/2 (0.00%), 71.50 kH/s nooooo
[2016-01-30 13:54:19] reject reason: Invalid share
[2016-01-30 13:54:19] accepted: 0/3 (0.00%), 71.50 kH/s nooooo
[2016-01-30 13:54:19] reject reason: Invalid share
[2016-01-30 13:54:19] CPU #9: 4.50 kH/s
[2016-01-30 13:54:19] accepted: 0/4 (0.00%), 71.51 kH/s nooooo
Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 [13] 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 ... 191 »
  Print  
 
Jump to:  

Sponsored by , a Bitcoin-accepting VPN.
Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!