Author

Topic: OFFICIAL CGMINER mining software thread for linux/win/osx/mips/arm/r-pi 4.11.0 - page 790. (Read 5806015 times)

newbie
Activity: 23
Merit: 0
Thanks. Again, it's not a temperature problem!
Card is well cooled and staying around ~75°C.
Mmh. Is the temperature measured on the GPU die, or is there an external temperature sensor on the heat sink?

If the latter ⇒ bad heat transfer paste could still be the reason.
full member
Activity: 174
Merit: 100
Thanks. Again, it's not a temperature problem!
Card is well cooled and staying around ~75°C.

I already tried different intensities (3,5,7,8,9 and dynamic) and also lowering the GPU clock. Still same behaviour.
btw.. tried Catalyst 11.6, 11.7 and 11.8 too
full member
Activity: 164
Merit: 100
I agree but this is not the case here..

Till version 1.5.5 I reach constant 421MHash with my HD6970 at a temperature of ~75°C.
Since version 1.5.6 and with all recent versions the problem occurs. I think it may have something to do with the new kernels used..


The problem might be that the newer kernels (1.5.6+) pushes the card harder.

           Pushes the card harder -> generating more heat -> GPU freq. starts to oscillate.


So try lowering the intensity and see what happens. It might solve your problems....


/GoK

I remembered this came up in an earlier post:

Decreasing intensity can help as well. It's the combination of the latest phatk kernel and the high intensity in cgminer that pushes cards to their limit. You may find the reduction in hash rate to be insignificant but the stability to increase. Try -I 7 instead of 8.
full member
Activity: 174
Merit: 100
I agree but this is not the case here..

Till version 1.5.5 I reach constant 421MHash with my HD6970 at a temperature of ~75°C.
Since version 1.5.6 and with all recent versions the problem occurs. I think it may have something to do with the new kernels used..
full member
Activity: 174
Merit: 100
1.6.1 finally fixes this zero byte problem..good work

Performance problem with HD6970 still exists. GPU clock is jumping up&down every second, resulting in approx 80MHash less output.
full member
Activity: 168
Merit: 100
Live long and prosper. \\//,
I found some temporarily fix to reduce the cpu consumption in windows...
use this tool http://www.adsciengineering.com/StartAffinity/
to set the affinity of cgminer to one core. Use quotes on your cgminer cmdline and the id you would set in the taskmanager.


Or is there a way to actually get rid of it yet?
It only uses a total amount of CPU that amounts to 100% of one CPU core due to the strange way the ATI driver is written, and binding the application to one core just means all the work will go to that one core; it does not actually decrease CPU usage.
Yeah but 100% of one core is less than ~70% of 4 cores. (I'm sure i can say that since the cpu fan runs much slower then)
~70% of 4 = 3 GPU
(1 core with 100% / 1 GPU)
-ck
legendary
Activity: 4088
Merit: 1631
Ruu \o/
I found some temporarily fix to reduce the cpu consumption in windows...
use this tool http://www.adsciengineering.com/StartAffinity/
to set the affinity of cgminer to one core. Use quotes on your cgminer cmdline and the id you would set in the taskmanager.


Or is there a way to actually get rid of it yet?
It only uses a total amount of CPU that amounts to 100% of one CPU core due to the strange way the ATI driver is written, and binding the application to one core just means all the work will go to that one core; it does not actually decrease CPU usage.
Yeah but 100% of one core is less than ~70% of 4 cores. (I'm sure i can say that since the cpu fan runs much slower then)
Okay that is much crappier than I'd even imagined it would be on windows.
-ck
legendary
Activity: 4088
Merit: 1631
Ruu \o/
New release: Version 1.6.1, Links in first post. Smiley

Executive summary of user visible changes:
Hopefully fixed the zero sized binary bug \o/
Hopefully fixed the (old ncurses) resize crash
Hopefully fixed the memory leak in cpu mining and the inappropriately high rejects
Tidier build and smarter config and new --disable-opencl option
Less false positives for "unable to provide work fast enough"
New --shares option (mine X shares and exit)
New --enable-cpu and --disable-gpu options
Version info


Full changelog:
- Copy cgminer path, not cat it.
- Switching between redrawing windows does not fix the crash with old
libncurses, so redraw both windows, but only when the window size hasn't
changed.
- Reinstate minimum 1 extra in queue to make it extremely unlikely to ever have
0 staged work items and any idle time.
- Return -1 if no input is detected from the menu to prevent it being
interpreted as a 0.
- Make pthread, libcurl and libcurses library checks mandatory or fail.
- Add a --disable-opencl configure option to make it possible to override
detection of opencl and build without GPU mining support.
- Confusion over the variable name for number of devices was passing a bogus
value which likely was causing the zero sized binary issue.
- cgminer no longer supports default url user and pass so remove them.
- Don't show value of intensity since it's dynamic by default.
- Add options to explicitly enable CPU mining or disable GPU mining.
- Convert the opt queue into a minimum number of work items to have queued
instead of an extra number to decrease risk of getting idle devices without
increasing risk of higher rejects.
- Statify tv_sort.
- Check for SSE2 before trying to build 32 bit SSE2 assembly version. Prevents
build failure when yasm is installed but -msse2 is not specified.
- Add some defines to configure.ac to enable exporting of values and packaging,
and clean up output.
- Give convenient summary at end of ./configure.
- Display version information and add --version command line option, and make
sure we flush stdout.
- Enable curses after the mining threads are set up so that failure messages
won't be lost in the curses interface.
- Disable curses after inputting a pool if we requested no curses interface.
- Add an option to break out after successfully mining a number of accepted
shares.
- Exit with a failed return code if we did not reach opt_shares.
- The cpu mining work data can get modified before we copy it if we submit it
async, and the sync submission is not truly sync anyway, so just submit it sync.


---
If you're wondering why I include the entire build in the linux binary tarball, it's so you can do 'make install' to install it into a system location.  Wink
legendary
Activity: 1666
Merit: 1057
Marketing manager - GO MP
I found some temporarily fix to reduce the cpu consumption in windows...
use this tool http://www.adsciengineering.com/StartAffinity/
to set the affinity of cgminer to one core. Use quotes on your cgminer cmdline and the id you would set in the taskmanager.


Or is there a way to actually get rid of it yet?
It only uses a total amount of CPU that amounts to 100% of one CPU core due to the strange way the ATI driver is written, and binding the application to one core just means all the work will go to that one core; it does not actually decrease CPU usage.
Yeah but 100% of one core is less than ~70% of 4 cores. (I'm sure i can say that since the cpu fan runs much slower then)
full member
Activity: 186
Merit: 100
Nice commits yesterday, thanks! Really polishing it up.
-ck
legendary
Activity: 4088
Merit: 1631
Ruu \o/
I never had any compile errors form cgminer so this is the first one:

Code:
.../usr/lib/gcc/i686-pc-linux-gnu/4.4.5/include/xmmintrin.h:32:3: error: #error "SSE instruction set not enabled"

That's the new 32 bit assembly cpu mining code which is only being half compiled. A workaround for the moment is to add -msse2 to your CFLAGS till I can figure out how to fix it.
-ck
legendary
Activity: 4088
Merit: 1631
Ruu \o/
I found some temporarily fix to reduce the cpu consumption in windows...
use this tool http://www.adsciengineering.com/StartAffinity/
to set the affinity of cgminer to one core. Use quotes on your cgminer cmdline and the id you would set in the taskmanager.


Or is there a way to actually get rid of it yet?
It only uses a total amount of CPU that amounts to 100% of one CPU core due to the strange way the ATI driver is written, and binding the application to one core just means all the work will go to that one core; it does not actually decrease CPU usage.
legendary
Activity: 910
Merit: 1000
Quality Printing Services by Federal Reserve Bank
I never had any compile errors form cgminer so this is the first one:

Code:
In file included from sha256_sse2_i386.c:21:
...
mv -f .deps/cgminer-sha256_sse4_amd64.Tpo .deps/cgminer-sha256_sse4_amd64.Po
gcc -DHAVE_CONFIG_H -I. -pthread -fno-strict-aliasing -I./compat/jansson -I./lib -I./lib  -DHAS_YASM -g -O2 -MT cgminer-sha256_sse2_i386.o -MD -MP -MF .deps/cgminer-sha256_sse2_i386.Tpo -c -o cgminer-sha256_sse2_i386.o `test -f 'sha256_sse2_i386.c' || echo './'`sha256_sse2_i386.c
/usr/lib/gcc/i686-pc-linux-gnu/4.4.5/include/xmmintrin.h:32:3: error: #error "SSE instruction set not enabled"
sha256_sse2_i386.c:25: error: expected ‘)’ before ‘*’ token
sha256_sse2_i386.c:50: error: expected ‘=’, ‘,’, ‘;’, ‘asm’ or ‘__attribute__’ before ‘g_4sha256_k’
sha256_sse2_i386.c:51: error: expected ‘=’, ‘,’, ‘;’, ‘asm’ or ‘__attribute__’ before ‘sha256_consts_m128i’
sha256_sse2_i386.c: In function ‘scanhash_sse2_32’:
sha256_sse2_i386.c:62: error: ‘__m128i’ undeclared (first use in this function)
sha256_sse2_i386.c:62: error: (Each undeclared identifier is reported only once
sha256_sse2_i386.c:62: error: for each function it appears in.)
sha256_sse2_i386.c:62: error: expected ‘;’ before ‘m_4w’
sha256_sse2_i386.c:63: error: expected ‘;’ before ‘m_4hash’
sha256_sse2_i386.c:64: error: expected ‘;’ before ‘m_4hash1’
sha256_sse2_i386.c:65: error: expected ‘;’ before ‘offset’
sha256_sse2_i386.c:72: error: expected specifier-qualifier-list before ‘__m128i’
sha256_sse2_i386.c:80: error: ‘m_4hash’ undeclared (first use in this function)
sha256_sse2_i386.c:84: error: ‘m_4w’ undeclared (first use in this function)
sha256_sse2_i386.c:87: error: ‘m_4hash1’ undeclared (first use in this function)
sha256_sse2_i386.c:90: error: ‘sha256_consts_m128i’ undeclared (first use in this function)
sha256_sse2_i386.c:92: error: ‘offset’ undeclared (first use in this function)
sha256_sse2_i386.c:105: error: ‘union ’ has no member named ‘m’
sha256_sse2_i386.c:106: error: ‘union ’ has no member named ‘i’
sha256_sse2_i386.c:114: error: ‘union ’ has no member named ‘m’
sha256_sse2_i386.c:115: error: ‘union ’ has no member named ‘i’
make[2]: *** [cgminer-sha256_sse2_i386.o] Error 1
make[2]: Leaving directory `/blah/cgminer-1.6.0'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/blah/cgminer-1.6.0'
make: *** [all] Error 2


legendary
Activity: 1666
Merit: 1057
Marketing manager - GO MP
I found some temporarily fix to reduce the cpu consumption in windows...
use this tool http://www.adsciengineering.com/StartAffinity/
to set the affinity of cgminer to one core. Use quotes on your cgminer cmdline and the id you would set in the taskmanager.


Or is there a way to actually get rid of it yet?
full member
Activity: 373
Merit: 100
P.S. I really want to kill AMD, upgraded to latest Debian testing and the 10.7 or 10.8 or whatever they are consume 100% CPU. Argh...
I know what you mean. You can get the old package versions from here: http://snapshot.debian.org/package/fglrx-driver/1:11-6-3/

I really haven't noticed a difference in either gaming or mining performance, and this way the CPU usage is minimised, so I see no reason not to downgrade.
-ck
legendary
Activity: 4088
Merit: 1631
Ruu \o/
Another couple:

1) So what does more exactly longpoll do and how to enable it with cgminer? I was under impression that it should be enabled by default, unless I specify --no-longpoll, but somehow it doesn't work for me on Slush pool. Shall I add LP to the pool URL?

2) So what's the best kernel for Radeon 5xxx ATM, shall I add -k phatk for optimum performance?

P.S. I really want to kill AMD, upgraded to latest Debian testing and the 10.7 or 10.8 or whatever they are consume 100% CPU. Argh...
I believe slush doesn't have longpoll. cgminer detects it by default and tries to enable it. Luckily cgminer is very good at minimising rejects even without longpoll.

Best kernel is likely phatk, and I'm pretty sure it will use that anyway by default.
newbie
Activity: 59
Merit: 0
Another couple:

1) So what does more exactly longpoll do and how to enable it with cgminer? I was under impression that it should be enabled by default, unless I specify --no-longpoll, but somehow it doesn't work for me on Slush pool. Shall I add LP to the pool URL?

2) So what's the best kernel for Radeon 5xxx ATM, shall I add -k phatk for optimum performance?

P.S. I really want to kill AMD, upgraded to latest Debian testing and the 10.7 or 10.8 or whatever they are consume 100% CPU. Argh...
newbie
Activity: 59
Merit: 0
Couple of observations:

1) Since cgminer is unusable with older curses, maybe make a compile-time switch to disable them completely? Especially now that there is a monitor option etc.
2) ./configure doesn't fail if no curses development files were detected, but compilation fails subsequently.
-ck
legendary
Activity: 4088
Merit: 1631
Ruu \o/
Is ask rate and scan time the same thing?
There is no official terminology for any of this stuff, but I'm assuming what ask-rate is, I guess the answer is yes.
member
Activity: 84
Merit: 10
Is ask rate and scan time the same thing?
Jump to: