Re: [darktable-user] open CL 2.0.6

2016-09-17 Thread Ulrich Pegelow 

Hi,

short update from my side. Looks like I have found a way to restore the 
original OpenCL performance of NVIDIA devices with recent driver versions.


Currently we have some other issues with the OpenCL codepath in master 
which prevents me from working there. If this gets sorted out soon, I 
will apply the changes there. If problems there persist longer I'll make 
a dedicated patch into the darktable-2.0.x branch latest tomorrow.


Ulrich

Am 13.09.2016 um 06:12 schrieb I. Ivanov:

Hi Guys,

Did somebody notice slowing down since DT 2.0.6? I am on ubuntu 16.04 64
bit. 8Gig RAM. I noticed somewhat slower performance of DT. Experimented
with turning off Open CL and it looks like the speed improved. My
security patches are up to date as released by ubuntu.

Is it only me?

Regards,

B



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] open CL 2.0.6

2016-09-16 Thread I. Ivanov 

I am attaching my logs

this is darktable 2.0.6
copyright (c) 2009-2016 johannes hanika
darktable-...@lists.darktable.org

compile options:
  bit depth is 64 bit
  normal build
  OpenMP support enabled
  OpenCL support enabled
  Lua support enabled, API version 3.0.0
  Colord support enabled
  gPhoto2 support enabled
  GraphicsMagick support enabled

CPU~Quad core Intel Core i7-2630QM (-HT-MCP-) speed/max~800/2900 MHz 
Kernel~4.4.0-36-generic x86_64 Up~3 days Mem~3038.2/7877.1MB 
HDD~500.1GB(73.3% used) Procs~289 Client~Shell inxi~2.2.35


Graphics:  Card-1: Intel 2nd Generation Core Processor Family Integrated 
Graphics Controller

   Card-2: NVIDIA GF108M [GeForce GT 525M]
   Display Server: X.Org 1.18.3 driver: nvidia Resolution: 
1920x1080@60.00hz, 1366x768@60.06hz
   GLX Renderer: GeForce GT 525M/PCIe/SSE2 GLX Version: 4.5.0 
NVIDIA 361.42


https://drive.google.com/open?id=0B-ibE69DzumKSXh2VmtyQUJRV2c

clocked results for exporting 50 images.

open cl on pinned true - 10 min 30s
open cl on pinned false - 9 min 13s
open cl off - 8 min 15s

For me - it appears that *open cl off* is the fastest. I have no 
explanation why I perceived that pinned=true is faster. It certainly 
"looked" faster to me when in dartkable mode. But the numbers are above. 
I "think" what happen is that I increased the complexity how many 
modules I activate and this somehow convinced me that DT slowed down. In 
fact - what did happen is - my images became more complex and it simply 
takes more time for DT to deal with them.


Hope this info is useful.

Thank you,

B





On 2016-09-16 01:08 PM, Michael Below wrote:

Hi,

another example from me. As far as I can see, pinning has a slightly
worse performance than the default.

My system:
CPU~Quad core AMD Phenom II X4 810 (-MCP-) speed~2600 MHz (max)
Kernel~4.6.0-1-amd64 x86_64 Up~3:45 Mem~2389.4/5956.0MB
HDD~3250.7GB(17.9% used) Procs~300 Client~Shell inxi~2.3.1

I think it would improve my use case most if the "atrous" module would
run on GPU. There seems to be some issue with tile size that makes the
equalizer module take e.g. 13 seconds on some images.

Cheers
Michael


Am Fr 16 Sep 2016 07:37:45 CEST
schrieb Ulrich Pegelow :


Thanks for sharing. Yours is a good example of an OpenCL system that
is not limited by host<->device memory transfers. In a typical export
job your system spends about 30% of its time in memory transfer, the
rest is pure computing. That's a very good situation in which pinned
memory does not give advantages - maybe even slow down a bit.

Others have systems which are purely limited by memory transfer. We
have reports of insane cases where over 95% of the OpenCL pixelpipe
is used by memory transfers. Those are the ones where
opencl_use_pinned_memory makes a real difference.

Ulrich

Am 15.09.2016 um 22:11 schrieb KOVÁCS István:

Hi,

Core2-Duo E6550 @ 2.33GHz +Nvidia GeForce GTX 650 / 2 GB, driver
361.42, OpenCL 1.2 CUDA, darktable 2.0.6 from PPA.
With pinned memory, performance is slightly (about 10%?) worse.
There are lines like
[opencl_profiling] spent  0,3774 seconds in [Map Buffer]
that are only seen in the 'pinned' log.
One notable difference after exporting 114 photos:
pinned = false gives
[opencl_summary_statistics] device 'GeForce GTX 650': 8960 out of
8960 events were successful and 0 events lost

pinned = true gives
[opencl_summary_statistics] device 'GeForce GTX 650': 9933 out of
9933 events were successful and 0 events lost

as one of the last lines in the output.
My opencl-related darktablerc entries:
opencl=TRUE
opencl_async_pixelpipe=false
opencl_avoid_atomics=false
opencl_checksum=2684983341
opencl_device_priority=*/!0,*/*/*
opencl_library=
opencl_memory_headroom=300
opencl_memory_requirement=768
opencl_micro_nap=1000
opencl_number_event_handles=25
opencl_omit_whitebalance=
opencl_size_roundup=16
opencl_synch_cache=false
opencl_use_cpu_devices=false
opencl_use_pinned_memory=false

The logs are at:
http://tech.kovacs-telekes.org/files/darktable-opencl-pinned-memory/

Thanks,
Kofa
  


darktable user mailing list
to unsubscribe send a mail to
darktable-user+unsubscr...@lists.darktable.org








darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] open CL 2.0.6

2016-09-16 Thread I. Ivanov 

Guys,

what commands do you issue in order to provide Ulrich with log files and 
system info?


I am sorry  trying to learn and don't know them  trying to help 
somehow...


Regards,

B


On 2016-09-16 06:16 AM, Patrick Shanahan wrote:

* Chester  [09-16-16 08:56]:

Well, I have go back to old Nvidia driver:

Graphics:  Card: NVIDIA GF106 [GeForce GTS 450]
Display Server: X.Org 1.18.3 drivers: nvidia (unloaded:
fbdev,vesa,nouveau)
Resolution: 1680x1050@59.88hz, 1920x1200@59.95hz
GLX Renderer: GeForce GTS 450/PCIe/SSE2
GLX Version: 4.4.0 NVIDIA 340.96

andd now Darktable works as always :-)

Thanks


El 15/09/16 a las 11:16, Ulrich Pegelow escribió:

Thanks. That's a small advantage for opencl_use_pinned_memory=TRUE.

Still even with the flag set to TRUE we loose quite some time in the
host_memory->device_memory step. I expect that I can change our code to
get some further improvements there in the next days.

Ulrich

Am 15.09.2016 um 16:10 schrieb Chester:

ULrich: here are my config and the two files:

darktable --version
this is darktable 2.0.6
copyright (c) 2009-2016 johannes hanika
darktable-...@lists.darktable.org

compile options:
   bit depth is 64 bit
   normal build
   OpenMP support enabled
   OpenCL support enabled
   Lua support enabled, API version 3.0.0
   Colord support enabled
   gPhoto2 support enabled
   GraphicsMagick support enabled

inxi
CPU~Quad core Intel Core i7 930 (-HT-MCP-) speed/max~1600/2801 MHz
Kernel~4.4.0-36-generic x86_64 Up~1:50 Mem~2072.5/20069.5MB
HDD~3768.8GB(3.6% used) Procs~268 Client~Shell inxi~2.2.35

inxi -G
Graphics:  Card: NVIDIA GF106 [GeForce GTS 450]
Display Server: X.Org 1.18.3 drivers: nvidia (unloaded:
fbdev,vesa,nouveau)
Resolution: 1680x1050@59.88hz, 1920x1200@59.95hz
GLX Renderer: GeForce GTS 450/PCIe/SSE2
GLX Version: 4.5.0 NVIDIA 361.42




darktable user mailing list
to unsubscribe send a mail to
darktable-user+unsubscr...@lists.darktable.org



--
Carlos Arigós
https://www.flickr.com/photos/chesterhimes/sets


darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org


Graphics:  Card: NVIDIA GF106 [GeForce GTS 450]
Display Server: X.Org 1.18.4 drivers: nvidia (unloaded:
fbdev,nv,vesa,nouveau)
Resolution: 1920x1080@60.00hz, 1920x1080@60.00hz
GLX Renderer: GeForce GTS 450/PCIe/SSE2
GLX Version: 4.5.0 NVIDIA 370.28

  



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] open CL 2.0.6

2016-09-16 Thread Patrick Shanahan 
* Chester  [09-16-16 08:56]:
> Well, I have go back to old Nvidia driver:
> 
> Graphics:  Card: NVIDIA GF106 [GeForce GTS 450]
>Display Server: X.Org 1.18.3 drivers: nvidia (unloaded:
> fbdev,vesa,nouveau)
>Resolution: 1680x1050@59.88hz, 1920x1200@59.95hz
>GLX Renderer: GeForce GTS 450/PCIe/SSE2
>GLX Version: 4.4.0 NVIDIA 340.96
> 
> andd now Darktable works as always :-)
> 
> Thanks
> 
> 
> El 15/09/16 a las 11:16, Ulrich Pegelow escribió:
> > Thanks. That's a small advantage for opencl_use_pinned_memory=TRUE.
> > 
> > Still even with the flag set to TRUE we loose quite some time in the
> > host_memory->device_memory step. I expect that I can change our code to
> > get some further improvements there in the next days.
> > 
> > Ulrich
> > 
> > Am 15.09.2016 um 16:10 schrieb Chester:
> > > ULrich: here are my config and the two files:
> > > 
> > > darktable --version
> > > this is darktable 2.0.6
> > > copyright (c) 2009-2016 johannes hanika
> > > darktable-...@lists.darktable.org
> > > 
> > > compile options:
> > >   bit depth is 64 bit
> > >   normal build
> > >   OpenMP support enabled
> > >   OpenCL support enabled
> > >   Lua support enabled, API version 3.0.0
> > >   Colord support enabled
> > >   gPhoto2 support enabled
> > >   GraphicsMagick support enabled
> > > 
> > > inxi
> > > CPU~Quad core Intel Core i7 930 (-HT-MCP-) speed/max~1600/2801 MHz
> > > Kernel~4.4.0-36-generic x86_64 Up~1:50 Mem~2072.5/20069.5MB
> > > HDD~3768.8GB(3.6% used) Procs~268 Client~Shell inxi~2.2.35
> > > 
> > > inxi -G
> > > Graphics:  Card: NVIDIA GF106 [GeForce GTS 450]
> > >Display Server: X.Org 1.18.3 drivers: nvidia (unloaded:
> > > fbdev,vesa,nouveau)
> > >Resolution: 1680x1050@59.88hz, 1920x1200@59.95hz
> > >GLX Renderer: GeForce GTS 450/PCIe/SSE2
> > >GLX Version: 4.5.0 NVIDIA 361.42
> > > 
> > 
> > 
> > 
> > darktable user mailing list
> > to unsubscribe send a mail to
> > darktable-user+unsubscr...@lists.darktable.org
> > 
> > 
> 
> -- 
> Carlos Arigós
> https://www.flickr.com/photos/chesterhimes/sets
> 
> 
> darktable user mailing list
> to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org


Graphics:  Card: NVIDIA GF106 [GeForce GTS 450]
   Display Server: X.Org 1.18.4 drivers: nvidia (unloaded:
   fbdev,nv,vesa,nouveau)
   Resolution: 1920x1080@60.00hz, 1920x1080@60.00hz
   GLX Renderer: GeForce GTS 450/PCIe/SSE2
   GLX Version: 4.5.0 NVIDIA 370.28

 
-- 
(paka)Patrick Shanahan   Plainfield, Indiana, USA  @ptilopteri
http://en.opensuse.orgopenSUSE Community Memberfacebook/ptilopteri
http://wahoo.no-ip.orgPhoto Album: http://wahoo.no-ip.org/gallery2
Registered Linux User #207535@ http://linuxcounter.net

darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] open CL 2.0.6

2016-09-16 Thread Patrick Shanahan 
* Chester  [09-16-16 08:56]:
> ULrich: here are my config and the two files:
> 
> darktable --version
> this is darktable 2.0.6
> copyright (c) 2009-2016 johannes hanika
> darktable-...@lists.darktable.org
> 
> compile options:
>   bit depth is 64 bit
>   normal build
>   OpenMP support enabled
>   OpenCL support enabled
>   Lua support enabled, API version 3.0.0
>   Colord support enabled
>   gPhoto2 support enabled
>   GraphicsMagick support enabled
> 
> inxi
> CPU~Quad core Intel Core i7 930 (-HT-MCP-) speed/max~1600/2801 MHz
> Kernel~4.4.0-36-generic x86_64 Up~1:50 Mem~2072.5/20069.5MB
> HDD~3768.8GB(3.6% used) Procs~268 Client~Shell inxi~2.2.35
> 
> inxi -G
> Graphics:  Card: NVIDIA GF106 [GeForce GTS 450]
>Display Server: X.Org 1.18.3 drivers: nvidia (unloaded:
> fbdev,vesa,nouveau)
>Resolution: 1680x1050@59.88hz, 1920x1200@59.95hz
>GLX Renderer: GeForce GTS 450/PCIe/SSE2
>GLX Version: 4.5.0 NVIDIA 361.42

 [lotsa trim]

Chester, I have same card, gts450, and opencl worked quite well for me
maybe 2 years ago around dt 1.4, but my system is very much more
responsive when opencl is disabled.  And speed is important to me as I
work large volumes of files during soccer season.

-- 
(paka)Patrick Shanahan   Plainfield, Indiana, USA  @ptilopteri
http://en.opensuse.orgopenSUSE Community Memberfacebook/ptilopteri
http://wahoo.no-ip.orgPhoto Album: http://wahoo.no-ip.org/gallery2
Registered Linux User #207535@ http://linuxcounter.net

darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] open CL 2.0.6

2016-09-15 Thread KOVÁCS István 
Hi,

Core2-Duo E6550 @ 2.33GHz +Nvidia GeForce GTX 650 / 2 GB, driver
361.42, OpenCL 1.2 CUDA, darktable 2.0.6 from PPA.
With pinned memory, performance is slightly (about 10%?) worse.
There are lines like
[opencl_profiling] spent  0,3774 seconds in [Map Buffer]
that are only seen in the 'pinned' log.
One notable difference after exporting 114 photos:
pinned = false gives
[opencl_summary_statistics] device 'GeForce GTX 650': 8960 out of 8960
events were successful and 0 events lost

pinned = true gives
[opencl_summary_statistics] device 'GeForce GTX 650': 9933 out of 9933
events were successful and 0 events lost

as one of the last lines in the output.
My opencl-related darktablerc entries:
opencl=TRUE
opencl_async_pixelpipe=false
opencl_avoid_atomics=false
opencl_checksum=2684983341
opencl_device_priority=*/!0,*/*/*
opencl_library=
opencl_memory_headroom=300
opencl_memory_requirement=768
opencl_micro_nap=1000
opencl_number_event_handles=25
opencl_omit_whitebalance=
opencl_size_roundup=16
opencl_synch_cache=false
opencl_use_cpu_devices=false
opencl_use_pinned_memory=false

The logs are at:
http://tech.kovacs-telekes.org/files/darktable-opencl-pinned-memory/

Thanks,
Kofa

On 15 September 2016 at 16:16, Ulrich Pegelow
 wrote:
> Thanks. That's a small advantage for opencl_use_pinned_memory=TRUE.
>
> Still even with the flag set to TRUE we loose quite some time in the
> host_memory->device_memory step. I expect that I can change our code to get
> some further improvements there in the next days.
>
> Ulrich
>
>
> Am 15.09.2016 um 16:10 schrieb Chester:
>>
>> ULrich: here are my config and the two files:
>>
>> darktable --version
>> this is darktable 2.0.6
>> copyright (c) 2009-2016 johannes hanika
>> darktable-...@lists.darktable.org
>>
>> compile options:
>>   bit depth is 64 bit
>>   normal build
>>   OpenMP support enabled
>>   OpenCL support enabled
>>   Lua support enabled, API version 3.0.0
>>   Colord support enabled
>>   gPhoto2 support enabled
>>   GraphicsMagick support enabled
>>
>> inxi
>> CPU~Quad core Intel Core i7 930 (-HT-MCP-) speed/max~1600/2801 MHz
>> Kernel~4.4.0-36-generic x86_64 Up~1:50 Mem~2072.5/20069.5MB
>> HDD~3768.8GB(3.6% used) Procs~268 Client~Shell inxi~2.2.35
>>
>> inxi -G
>> Graphics:  Card: NVIDIA GF106 [GeForce GTS 450]
>>Display Server: X.Org 1.18.3 drivers: nvidia (unloaded:
>> fbdev,vesa,nouveau)
>>Resolution: 1680x1050@59.88hz, 1920x1200@59.95hz
>>GLX Renderer: GeForce GTS 450/PCIe/SSE2
>>GLX Version: 4.5.0 NVIDIA 361.42
>>
>
> 
> darktable user mailing list
> to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org
>

darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] open CL 2.0.6

2016-09-15 Thread Ulrich Pegelow 

Am 15.09.2016 um 09:31 schrieb Tobias Ellinghaus:

With a speed difference like that, couldn't we run a small benchmark at init
time (we already compare the speed to the CPU) and set the flag accordingly at
runtime?


Probably we should.

Ulrich

darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] open CL 2.0.6

2016-09-15 Thread Ulrich Pegelow 
Well, in your case I see the differences as only marginal - the time 
spent in the OpenCL pixelpipe differs only by 2% between the two setting 
(in favor of TRUE). Not sure if differences persist if you would repeat 
profiling several times to get out any fluctuations.


So it seems that your combination of GPU and driver does not profit from 
the opencl_use_pinned_memory flag. But in your case it would not harm 
either to change the default to TRUE.


To others: I am interested to see if there are systems where 
opencl_use_pinned_memory=TRUE gives a heavy negative impact on performance.


Ulrich

Am 15.09.2016 um 06:00 schrieb Jack Bowling:

On 09/14/2016 09:56 AM, Ulrich Pegelow wrote:

Well, there obviously is an issue with OpenCL and NVIDIA. However, a
quick check reveals that this is not related to 2.0.6 versus 2.0.5.

In fact it seems that NVIDIA did some changes to their drivers in the
way they handle memory transfers over the IDE interface.

There is a quick fix for that in darktable. You can switch config
variable opencl_use_pinned_memory to TRUE (can be found in darktablerc).
At least here on my this makes a difference of up to a factor of 30
(oldish GeForce GTS 450 and 367.35 driver).



Setting pinned_memory=true leads to slower render times on my box. Here
is system info on my fully updated Ubuntu 16.04 box:

$ darktable --version
this is darktable 2.0.6
copyright (c) 2009-2016 johannes hanika
darktable-...@lists.darktable.org

compile options:
  bit depth is 64 bit
  normal build
  OpenMP support enabled
  OpenCL support enabled
  Lua support enabled, API version 3.0.0
  Colord support enabled
  gPhoto2 support enabled
  GraphicsMagick support enabled

$ inxi
CPU~Octa core AMD FX-8300 Eight-Core (-MCP-) speed/max~1400/3300 MHz
Kernel~4.4.0-36-generic x86_64 Up~8 days Mem~2495.3/32090.4MB
HDD~23734.6GB(33.4% used) Procs~340 Client~Shell inxi~2.2.35

$ inxi -G
Graphics:  Card: NVIDIA GK107 [GeForce GT 740]
   Display Server: X.Org 1.18.3 drivers: nvidia (unloaded:
fbdev,vesa,nouveau)
   Resolution: 2560x1440@59.95hz
   GLX Renderer: GeForce GT 740/PCIe/SSE2
   GLX Version: 4.5.0 NVIDIA 361.42

Here is the relevant paste from my darktable config:

opencl=TRUE
opencl_async_pixelpipe=false
opencl_avoid_atomics=false
opencl_checksum=4188966525
opencl_device_priority=*/!0,*/*/*
opencl_library=
opencl_memory_headroom=1000
opencl_memory_requirement=768
opencl_micro_nap=1000
opencl_number_event_handles=25
opencl_omit_whitebalance=
opencl_size_roundup=16
opencl_synch_cache=false
opencl_use_cpu_devices=false
opencl_use_pinned_memory=false

Note the high headroom necessary to prevent atrous dumping to CPU.

Attached are two text files of "darktable -d opencl -d perf" output, one
with pinned_memory=true and one with pinned_memory=false.

Jack




darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] open CL 2.0.6

2016-09-14 Thread Colin Adams 
Yes, it's working now. I guess the crash must have been a coincidence.
Slower if anything, but there isn't much in it. I can't be sure.
Running with NVIDIA 370.28 driver.

On Wed, 14 Sep 2016 at 18:37 Ulrich Pegelow 
wrote:

> Do I understand correctly that you can run with the flag set to TRUE?
> What are your findings in terms of speed improvements (if any)?
>
> Am 14.09.2016 um 19:34 schrieb Colin Adams:
> > No.
> > Doesn't happen anymore.
> >
> > On Wed, 14 Sep 2016 at 18:26 Ulrich Pegelow  > > wrote:
> >
> > Any backtrace?
> >
> > Am 14.09.2016 um 19:12 schrieb Colin Adams:
> > > It causes darktable 2.0.5 (Fedora) to crash. Switching back to
> false
> > > cures the problem. So please don't change.
>
>
> 
> darktable user mailing list
> to unsubscribe send a mail to
> darktable-user+unsubscr...@lists.darktable.org
>
>


darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] open CL 2.0.6

2016-09-14 Thread Ulrich Pegelow 

Any backtrace?

Am 14.09.2016 um 19:12 schrieb Colin Adams:

It causes darktable 2.0.5 (Fedora) to crash. Switching back to false
cures the problem. So please don't change.

On Wed, 14 Sep 2016 at 17:56 Ulrich Pegelow > wrote:

Well, there obviously is an issue with OpenCL and NVIDIA. However, a
quick check reveals that this is not related to 2.0.6 versus 2.0.5.

In fact it seems that NVIDIA did some changes to their drivers in the
way they handle memory transfers over the IDE interface.

There is a quick fix for that in darktable. You can switch config
variable opencl_use_pinned_memory to TRUE (can be found in darktablerc).
At least here on my this makes a difference of up to a factor of 30
(oldish GeForce GTS 450 and 367.35 driver).

Background: that switch controls the way of memory transfer between host
and OpenCL device, namely the use of pre-pinned memory. When the flag
was introduced it did only make some improvements on AMD/ATI devices,
while at that time NVIDIA devices would show no or a slight negative
effects. Therefore the flag is set to FALSE by default. It seems that
newer NVIDIA drivers get extremely slow if the default non-pinned memory
transfer method is used.

If my findings are confirmed we will change the default setting of that
flag for new installations. Users of existing installations will need to
change the config flag manually.

Please check and report back.

Ulrich




darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] open CL 2.0.6

2016-09-14 Thread Ulrich Pegelow 
Well, there obviously is an issue with OpenCL and NVIDIA. However, a 
quick check reveals that this is not related to 2.0.6 versus 2.0.5.


In fact it seems that NVIDIA did some changes to their drivers in the 
way they handle memory transfers over the IDE interface.


There is a quick fix for that in darktable. You can switch config 
variable opencl_use_pinned_memory to TRUE (can be found in darktablerc). 
At least here on my this makes a difference of up to a factor of 30 
(oldish GeForce GTS 450 and 367.35 driver).


Background: that switch controls the way of memory transfer between host 
and OpenCL device, namely the use of pre-pinned memory. When the flag 
was introduced it did only make some improvements on AMD/ATI devices, 
while at that time NVIDIA devices would show no or a slight negative 
effects. Therefore the flag is set to FALSE by default. It seems that 
newer NVIDIA drivers get extremely slow if the default non-pinned memory 
transfer method is used.


If my findings are confirmed we will change the default setting of that 
flag for new installations. Users of existing installations will need to 
change the config flag manually.


Please check and report back.

Ulrich

Am 14.09.2016 um 00:47 schrieb I. Ivanov:

I find it strange...

When I upgraded from 14.04 to 16.04 - DT was at version 2.0.5 and nvidia
361. I actually experienced speed "gain" - did not clock it but it was
very noticeable. I worked in this state for several weeks - all happy,
no changes in settings to DT.

2.0.6 was installed on 2016-09-06. I noticed it works but did not test
any further. Then I didn't use the computer till 11th. Installed the
following updates

2016-09-11 19:56:16 status installed gnome-menus:amd64 3.13.3-6ubuntu3.1
2016-09-11 19:56:16 status installed desktop-file-utils:amd64 0.22-1ubuntu5
2016-09-11 19:56:17 status installed mime-support:all 3.59ubuntu1
2016-09-11 19:56:17 status installed bamfdaemon:amd64
0.5.3~bzr0+16.04.20160701-0ubuntu1
2016-09-11 19:56:17 status installed man-db:amd64 2.7.5-1
2016-09-11 19:56:18 status installed libc-bin:amd64 2.23-0ubuntu3
2016-09-11 19:56:18 status installed dbus:amd64 1.10.6-1ubuntu3
2016-09-11 19:56:18 status installed gconf2:amd64 3.2.6-3ubuntu6
2016-09-11 19:56:18 status installed hicolor-icon-theme:all 0.15-0ubuntu1
2016-09-11 19:56:18 status installed libglib2.0-0:i386
2.48.1-1~ubuntu16.04.1
2016-09-11 19:56:18 status installed libglib2.0-0:amd64
2.48.1-1~ubuntu16.04.1
2016-09-11 19:56:18 status installed sgml-base:all 1.26+nmu4ubuntu1
2016-09-11 19:56:19 status installed google-chrome-stable:amd64
53.0.2785.101-1
2016-09-11 19:56:19 status installed libp11-kit0:amd64
0.23.2-5~ubuntu16.04.1
2016-09-11 19:56:19 status installed libp11-kit0:i386 0.23.2-5~ubuntu16.04.1
2016-09-11 19:56:19 status installed p11-kit-modules:amd64
0.23.2-5~ubuntu16.04.1
2016-09-11 19:56:19 status installed libaccountsservice0:amd64
0.6.40-2ubuntu11.2
2016-09-11 19:56:19 status installed accountsservice:amd64
0.6.40-2ubuntu11.2
2016-09-11 19:56:19 status installed file-roller:amd64 3.16.5-0ubuntu1.2
2016-09-11 19:56:19 status installed gnome-font-viewer:amd64 3.16.2-1ubuntu1
2016-09-11 19:56:19 status installed libappstream-glib8:amd64
0.5.13-1ubuntu3
2016-09-11 19:56:19 status installed libimlib2:amd64 1.4.7-1ubuntu0.1
2016-09-11 19:56:19 status installed metacity-common:all 1:3.18.7-0ubuntu0.1
2016-09-11 19:56:19 status installed libmetacity-private3a:amd64
1:3.18.7-0ubuntu0.1
2016-09-11 19:56:19 status installed libnm-gtk-common:all
1.2.0-0ubuntu0.16.04.4
2016-09-11 19:56:19 status installed libnm-gtk0:amd64 1.2.0-0ubuntu0.16.04.4
2016-09-11 19:56:19 status installed libnma-common:all
1.2.0-0ubuntu0.16.04.4
2016-09-11 19:56:19 status installed libnma0:amd64 1.2.0-0ubuntu0.16.04.4
2016-09-11 19:56:19 status installed network-manager-gnome:amd64
1.2.0-0ubuntu0.16.04.4
2016-09-11 19:56:21 status installed snapd:amd64 2.14.2~16.04
2016-09-11 19:56:21 status installed p11-kit:amd64 0.23.2-5~ubuntu16.04.1
2016-09-11 19:56:21 status installed libc-bin:amd64 2.23-0ubuntu3

and noticed drop in performance - mainly when using darktable (not much
light table). Took a chance to turn off open CL and the performance
improved. After reading the thread

https://www.mail-archive.com/darktable-dev@lists.darktable.org/msg01176.html

Tried to compare with export of a single image.
without open CL - 23s for about 20 MB RAW.
Same image - no change
with open CL - 41s

The OS and the images are stored on SSD so the networking does not come
into play.
I can work without open CL - it is not a deal breaker but the behavior
is surprising.

Regards,
B



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] open CL 2.0.6

2016-09-13 Thread I. Ivanov 

I find it strange...

When I upgraded from 14.04 to 16.04 - DT was at version 2.0.5 and nvidia 
361. I actually experienced speed "gain" - did not clock it but it was 
very noticeable. I worked in this state for several weeks - all happy, 
no changes in settings to DT.


2.0.6 was installed on 2016-09-06. I noticed it works but did not test 
any further. Then I didn't use the computer till 11th. Installed the 
following updates


2016-09-11 19:56:16 status installed gnome-menus:amd64 3.13.3-6ubuntu3.1
2016-09-11 19:56:16 status installed desktop-file-utils:amd64 0.22-1ubuntu5
2016-09-11 19:56:17 status installed mime-support:all 3.59ubuntu1
2016-09-11 19:56:17 status installed bamfdaemon:amd64 
0.5.3~bzr0+16.04.20160701-0ubuntu1

2016-09-11 19:56:17 status installed man-db:amd64 2.7.5-1
2016-09-11 19:56:18 status installed libc-bin:amd64 2.23-0ubuntu3
2016-09-11 19:56:18 status installed dbus:amd64 1.10.6-1ubuntu3
2016-09-11 19:56:18 status installed gconf2:amd64 3.2.6-3ubuntu6
2016-09-11 19:56:18 status installed hicolor-icon-theme:all 0.15-0ubuntu1
2016-09-11 19:56:18 status installed libglib2.0-0:i386 
2.48.1-1~ubuntu16.04.1
2016-09-11 19:56:18 status installed libglib2.0-0:amd64 
2.48.1-1~ubuntu16.04.1

2016-09-11 19:56:18 status installed sgml-base:all 1.26+nmu4ubuntu1
2016-09-11 19:56:19 status installed google-chrome-stable:amd64 
53.0.2785.101-1
2016-09-11 19:56:19 status installed libp11-kit0:amd64 
0.23.2-5~ubuntu16.04.1

2016-09-11 19:56:19 status installed libp11-kit0:i386 0.23.2-5~ubuntu16.04.1
2016-09-11 19:56:19 status installed p11-kit-modules:amd64 
0.23.2-5~ubuntu16.04.1
2016-09-11 19:56:19 status installed libaccountsservice0:amd64 
0.6.40-2ubuntu11.2
2016-09-11 19:56:19 status installed accountsservice:amd64 
0.6.40-2ubuntu11.2

2016-09-11 19:56:19 status installed file-roller:amd64 3.16.5-0ubuntu1.2
2016-09-11 19:56:19 status installed gnome-font-viewer:amd64 3.16.2-1ubuntu1
2016-09-11 19:56:19 status installed libappstream-glib8:amd64 
0.5.13-1ubuntu3

2016-09-11 19:56:19 status installed libimlib2:amd64 1.4.7-1ubuntu0.1
2016-09-11 19:56:19 status installed metacity-common:all 1:3.18.7-0ubuntu0.1
2016-09-11 19:56:19 status installed libmetacity-private3a:amd64 
1:3.18.7-0ubuntu0.1
2016-09-11 19:56:19 status installed libnm-gtk-common:all 
1.2.0-0ubuntu0.16.04.4

2016-09-11 19:56:19 status installed libnm-gtk0:amd64 1.2.0-0ubuntu0.16.04.4
2016-09-11 19:56:19 status installed libnma-common:all 
1.2.0-0ubuntu0.16.04.4

2016-09-11 19:56:19 status installed libnma0:amd64 1.2.0-0ubuntu0.16.04.4
2016-09-11 19:56:19 status installed network-manager-gnome:amd64 
1.2.0-0ubuntu0.16.04.4

2016-09-11 19:56:21 status installed snapd:amd64 2.14.2~16.04
2016-09-11 19:56:21 status installed p11-kit:amd64 0.23.2-5~ubuntu16.04.1
2016-09-11 19:56:21 status installed libc-bin:amd64 2.23-0ubuntu3

and noticed drop in performance - mainly when using darktable (not much 
light table). Took a chance to turn off open CL and the performance 
improved. After reading the thread


https://www.mail-archive.com/darktable-dev@lists.darktable.org/msg01176.html

Tried to compare with export of a single image.
without open CL - 23s for about 20 MB RAW.
Same image - no change
with open CL - 41s

The OS and the images are stored on SSD so the networking does not come 
into play.
I can work without open CL - it is not a deal breaker but the behavior 
is surprising.


Regards,
B


On 2016-09-13 03:04 AM, Jonathan Niehof wrote:

There was a thread on the -dev list with someone else having slowness
on Ubuntu 16.04 with OpenCL on nVidia and it looked like a driver
issue. They switched from version 361 of the nVidia driver to 340 and
saw an improvement, might be worth trying.

Thread here: 
https://www.mail-archive.com/darktable-dev@lists.darktable.org/msg01176.html

darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org





darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] open CL 2.0.6

2016-09-12 Thread darktable 
On Mon, 12 Sep 2016 21:12:42 -0700
"I. Ivanov"  wrote:

>Did somebody notice slowing down since DT 2.0.6? I am on ubuntu
>16.04 64 bit. 8Gig RAM. I noticed somewhat slower performance of DT.
>Experimented with turning off Open CL and it looks like the speed
>improved. My security patches are up to date as released by ubuntu.

What kind of hardware? A 'real' video card (nvideia, radeon) or a
'fake' (intel or built in the cpu?

They did some changes to the opencl to turn off if it is a 'fake'
video card.

> • OpenCL: properly discard CPU-based OpenCL devices. 

-- 
sknahT

vyS

darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org