After setting the environment variables correctly, clinfo and
vector_addition codes are working fine. But the error with "rt -z" is the
same.
Do I need to set any other configuration paths?
How does rt detect opencl?
Regards
Shailesh
Shailesh Tripathi
B.Tech. Part-IV
Electronics Engineering
IIT-BHU (Varanasi)
On Thu, Mar 30, 2017 at 9:45 PM, Vasco Alexandre da Silva Costa <
vasco.co...@gmail.com> wrote:
> If OpenCL isn't detected on your system, the OpenCL backend won't be used.
>
> Did you check if OpenCL is properly installed in your system with 'clinfo'?
>
> On Thu, Mar 30, 2017 at 3:48 PM, Shailesh Tripathi <
> shailesh.tripathi.ec...@itbhu.ac.in> wrote:
>
>> Hello,
>> Wait, I think there is some confusion. The argument for OpenCL
>> compilation is "rt -z1" or "rt -z".
>> I think it is the second one.
>> Since "*rcc"* has not been ported yet, I am just creating "*sph*" and
>> using "rt -z" command:
>> When I try this, it says
>>
>>
>>
>>> *Planning to run with 8 processors*
>>>
>>> */usr/brlcad/dev-7.26.1/bin/rt -M -z -u*
>>>
>>> *opendb model;*
>>> *tree /home/**path_to_build**/mick1.g shapes.r; *
>>>
>>
>>> *rt: rt_dirbuild(model) failure *
>>>
>> *Raytrace failed.*
>>>
>>
>> Is there any example (with the commands) to run opencl raytracing?
>>
>> Regards
>> Shailesh
>>
>>
>> Shailesh Tripathi
>> B.Tech. Part-IV
>> Electronics Engineering
>> IIT-BHU (Varanasi)
>>
>> On Thu, Mar 30, 2017 at 7:49 PM, Shailesh Tripathi <
>> shailesh.tripathi.ec...@itbhu.ac.in> wrote:
>>
>>> Hello,
>>> Sorry for the late reply. As I said I got busy with my exams.
>>>
>>>> You should see a consistent difference in performance, but this would
>>>> require compiling and installing both with and without OpenCL support to
>>>> make sure the same compilation options are being used.
>>>>
>>> I don't exactly understand by what you meant by compiling in OpenCL.
>>> Right now, I am just following the raytrace tutorial. The command are as
>>> follows (n order):
>>>
>>> i
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> *n sph.s sph 4 4 4 3in rcc.s rcc 4 4 0 0 0 4 3draw sph.s rcc.sr
>>>> shapes.r u sph.s u rcc.smater shapes.r plastic 0 255 0 0B shapes.rrt*
>>>>
>>> This gives the raytrace output in the frame buffer
>>> But in the output log there is nothing like "PREP" as you said.
>>> Also for openCL raytracing, I am simply using "rt -z1" in place of the
>>> "rt" command. But still the output log is exact same with no signs of
>>> OpenCL compilation.
>>>
>>> Please tell me how and what exactly to do for OpenCL compilation.
>>>
>>> Regards
>>> Shailesh
>>>
>>>
>>> Shailesh Tripathi
>>> B.Tech. Part-IV
>>> Electronics Engineering
>>> IIT-BHU (Varanasi)
>>>
>>> On Tue, Mar 28, 2017 at 9:24 PM, Vasco Alexandre da Silva Costa <
>>> vasco.co...@gmail.com> wrote:
>>>
>>>> On Tue, Mar 28, 2017 at 2:57 AM, Christopher Sean Morrison <
>>>> brl...@mac.com> wrote:
>>>>
>>>>>
>>>>> > On Mar 27, 2017, at 8:01 PM, Shailesh Tripathi <
>>>>> shailesh.tripathi.ec...@itbhu.ac.in> wrote:
>>>>> >
>>>>> > Finally, now all the errors are resolved. deviceQuery' is working
>>>>> fine with CUDA which hints that OpenCL is also setup.
>>>>> > Now, as I told the 'rt' command and 'rt and 'rt -z1' gives
>>>>> quantitatively equal results.
>>>>> > How can I make sure that 'rt -z1' is making use of GPU?
>>>>>
>>>>> You should see a consistent difference in performance, but this would
>>>>> require compiling and installing both with and without OpenCL support to
>>>>> make sure the same compilation options are being used.
>>>>>
>>>>> Another tell-tale sign is that the OpenCL path does not yet handle
>>>>> booleans. If you render the havoc sample model, does it look like the
>>>>> rendering here http://brlcad.org/wiki/User:Vasco.costa/GSoC15/logs ?
>>>>>
>>>>> Last but not least, it should say "Compiling OpenCL programs...” and
>>>>> “OCLPREP” instead of “PREP” in the output.
>>>>>
>>>>
>>>> Yes, the easiest way to know for sure is to look at the console log
>>>> output.
>>>>
>>>> --
>>>> Vasco Alexandre da Silva Costa
>>>> PhD in Computer Engineering (Computer Graphics)
>>>> Instituto Superior Técnico/University of Lisbon, Portugal
>>>>
>>>> ------------------------------------------------------------
>>>> ------------------
>>>> Check out the vibrant tech community on one of the world's most
>>>> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
>>>> _______________________________________________
>>>> BRL-CAD Developer mailing list
>>>> brlcad-devel@lists.sourceforge.net
>>>> https://lists.sourceforge.net/lists/listinfo/brlcad-devel
>>>>
>>>>
>>>
>>
>> ------------------------------------------------------------
>> ------------------
>> Check out the vibrant tech community on one of the world's most
>> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
>> _______________________________________________
>> BRL-CAD Developer mailing list
>> brlcad-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/brlcad-devel
>>
>>
>
>
> --
> Vasco Alexandre da Silva Costa
> PhD in Computer Engineering (Computer Graphics)
> Instituto Superior Técnico/University of Lisbon, Portugal
>
> ------------------------------------------------------------
> ------------------
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> _______________________________________________
> BRL-CAD Developer mailing list
> brlcad-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/brlcad-devel
>
>
------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
BRL-CAD Developer mailing list
brlcad-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/brlcad-devel
