Editing User:Rishabhsuthar32/GSoC20/Report

From BRL-CAD

Warning: You are not logged in. Your IP address will be publicly visible if you make any edits. If you log in or create an account, your edits will be attributed to your username, along with other benefits.

The edit can be undone. Please check the comparison below to verify that this is what you want to do, and then save the changes below to finish undoing the edit.
Latest revision Your text
Line 1: Line 1:
= Project Report =
+
= Development Logs =
  
 
== Project Plan ==
 
== Project Plan ==
BRL-CAD has one of the oldest and fastest parallel ray tracing implementations around but it doesn't currently leverage the GPU very much. With implicit geometry and constructive solid geometry (CSG) Boolean operations, it also has a very different approach to ray tracing that has its own set of academic challenges.
 
 
The project here was to introduce a GPGPU pipeline into BRL-CAD using OpenCL for existing primitives and parallelize them for faster computation compared to the non-GPGPU implementation.
 
 
Some of the primitives that were planned to be converted in this 2020 season of Google Summer of Code are polyhedron with an arbitrary number (ARBN), FASTGEN4 CLINE, hyperboloids of one sheet (HYP), etc.
 
  
  
 
== Patches Submitted ==
 
== Patches Submitted ==
 
Following patches include modifications for better functioning and resolution of bugs encountered while rendering via OpenCL in MGED terminal.
 
* OpenCL_CMAKE: https://sourceforge.net/p/brlcad/patches/549/
 
* OpenCL rendering bug: https://sourceforge.net/p/brlcad/patches/551/
 
 
Following are the patches submitted for OpenCL versions of some of the primitives.
 
 
* CLINE: https://sourceforge.net/p/brlcad/patches/541/
 
* CLINE: https://sourceforge.net/p/brlcad/patches/541/
 
* ARBN: https://sourceforge.net/p/brlcad/patches/543/
 
* ARBN: https://sourceforge.net/p/brlcad/patches/543/
Line 22: Line 11:
 
* METABALL: https://sourceforge.net/p/brlcad/patches/548/
 
* METABALL: https://sourceforge.net/p/brlcad/patches/548/
 
* HYP: https://sourceforge.net/p/brlcad/patches/553/
 
* HYP: https://sourceforge.net/p/brlcad/patches/553/
 +
 +
* OpenCL_CMAKE: https://sourceforge.net/p/brlcad/patches/549/
 +
* OpenCL rendering bug: https://sourceforge.net/p/brlcad/patches/551/
  
  
Line 54: Line 46:
  
 
== To Do List ==
 
== To Do List ==
 
* '''Memory Allocation''' : One of the major challenges I faced was finding the OpenCL counterpart of bu_malloc which works fine while passing the variables from .c to .cl file. While I tried with malloc as is used normally in C programs, it apparently didn't work. The renderings were simply not occurring in the OpenCL version. Hence, the HYP and SUPERELL primitives, which didn't use the bu_malloc command, rendered easily while others didn't. The patches submitted above have no compilation or computation error. If the memory allocation thing works, they'd render just as fine.
 
 
* '''Using Printf statement''': Another issue where I got stuck was being unable to use printf statement in OpenCL files. The strategy to debug opencl kernels using printf is to enable printf command by passing enumerations stated below in an array of context properties to the function, clCreateContext, which creates opencl context as follows:
 
<pre>
 
CL_PRINTF_CALLBACK_ARM    0x40B0
 
CL_PRINTF_BUFFERSIZE_ARM  0x40B1
 
cl_context_properties context_properties[] =
 
{
 
CL_PRINTF_CALLBACK_ARM, (cl_context_properties)printf_callback,
 
CL_PRINTF_BUFFERSIZE_ARM, (cl_context_properties)0x100000,
 
CL_CONTEXT_PLATFORM, (cl_context_properties)platform,
 
0
 
};
 
</pre>
 
This enables the device side printf built in function for OpenCL C versions prior to 1.2. It also extends the cl_context_properties enumeration to allow a user defined printf callback and/or printf buffer size. OpenCL kernel code then has access to:
 
  #pragma OPENCL EXTENSION cl_arm_printf : enable
 
This pragma must be enabled for a kernel to have access to the printf built in function in OpenCL C versions prior to 1.2.
 
 
For the purpose of debugging in mged I used bu_log instead of printf by passing following array of context properties
 
<pre>
 
cl_context_properties context_properties[] =
 
{
 
CL_PRINTF_CALLBACK_ARM, (cl_context_properties)bu_log,
 
CL_PRINTF_BUFFERSIZE_ARM, (cl_context_properties)0x100000,
 
CL_CONTEXT_PLATFORM, (cl_context_properties)platform,
 
0
 
};
 
</pre>
 
 
But unfortunately this doesn't work and reports error that context cannot be created. This needs further investigation. The print statement would also be useful for debugging the memory allocation issue, as to where and why it is failing.
 
 
== References ==
 
* '''Project Plan''': https://brlcad.org/wiki/User:Rishabhsuthar32/GSoC20/Project
 
* '''Development Logs''': https://brlcad.org/wiki/User:Rishabhsuthar32/GSoC20/Log
 

Please note that all contributions to BRL-CAD may be edited, altered, or removed by other contributors. If you do not want your writing to be edited mercilessly, then do not submit it here.
You are also promising us that you wrote this yourself, or copied it from a public domain or similar free resource (see BRL-CAD:Copyrights for details). Do not submit copyrighted work without permission!

To edit this page, please answer the question that appears below (more info):

Cancel Editing help (opens in new window)