Sign in to follow this  

OpenGL Multi-threaded PBO Question

Recommended Posts

Grumple    177

I'm working on an OpenGL video player that uses Media Foundation as my video frame source.  I start one video frame capture thread per video source, with potentially multiple videos playing at once.  Currently I dump incoming frames to a local buffer on the sub thread, then my main render thread runs through my video capture objects, locking access to the buffer, and calling glTexSubImage2D() to upload to texture.  It renders each texture as a quad for display.


I'm intending to switch to using Pixel Buffer Objects.  What I would LIKE to do, is have twice as many PBO's as video sources, so on any given frame I can be drawing from the previous frames' upload target, while starting an upload to the previous frames render source.   To speed things up even more, I was intending to map all upload target PBO's, give a mapped memory pointer to each video capture sub thread to write frames directly to as they arrive, then unmap and upload on my next main-thread render pass.  


However, it seems as though you can only have one PBO bound/mapped at a time.  So, this means I cant leave mapped ptrs to multiple PBO's to multiple video capture threads at a time between my render.   Is this correct?


As far as I can tell, I have one of two reasonable options:

I can forget having my video capture threads write directly to mapped PBO's.  They can cache frame pixels in a standard local memory buffer, then my my thread can do two loops in its render function;  One to render the previous frames PBO uploads, and one to start uploading  newly cached video frame data to the PBO's that are not being rendered from.   It seems this would still benefit from asynchronous texture uploads, but would also still be bottlenecking.  For example, I'd still be wastefully caching video frames to RAM in sub threads before trying to upload them to PBO/texture memory, and my main thread would be responsible to do the memcpy into the mapped PBO memory.


The second method I though of, was to create two large PBO's that each contain space for one frame of all video sources.  This way my main thread could bind the current render source PBO, loop calling glTexSubImage2D for each texture with appropriate offsets into the PBO and render these textures.  Then it could map the current 'back buffer' PBO, and pass pointers to the correct frame offsets to the video capture threads for direct writes as they receive frame data.  On the next pass I could un-map the 'back buffer' pbo, and repeat previous steps to upload via glTexSubImage2D, then map the previous frames render PBO as the sub thread frame destination, etc, etc.


Sorry for making this long winded but I want to get some general advice/input with a good description of what I'm up to.  Is my general understanding correct?  What are thoughts on trade offs between sub threads being able to perform the memcpy into mapped PBO memory vs having to use a single large PBO to get around only being able to have one mapped with multiple data source threads?



Edited by Grumple

Share this post

Link to post
Share on other sites
BornToCode    1185

I did a similar system like that in the past. What I did is that in my background thread it just decode the frame and store it into an buffer. then send an event to my main thread letting it know it has a buffer available. Then that frame get send down to the main thread along with movie handle which contains an array of PBO. then that frame get inserted into one of the PBO. When it comes time to render I figure which frame the movie is on then grab the correct PBO and apply that to the image and draw. Once the frame changed, I marked that PBO index as invalid and when the thread sends the next event that PBO will be used. So to summarize you have two threads one that is doing Decoding and the main thread which uploads the decoding data to the PBO. then on your render function you just draw the correct pbo.

Edited by BornToCode

Share this post

Link to post
Share on other sites
Grumple    177

Hi BornToCode,  


I thought i had replied to this but apparently it didnt commit for some reason.  Thanks for the insight into your approach.  I ended up doing the 'two large PBOs' method described above, and it worked reasonably well, but I ran into a bottleneck mapping a PBO for writes that eliminated most of my gains.  Here is a new post I made regarding my updated implementation and current glMapBuffer woes:


I'm about to try a circular buffer of PBO's now to (hopefully) eliminate my stall, and will post my results in case it can help anyone else.

Share this post

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

Sign in to follow this  

  • Similar Content

    • By Zaphyk
      I am developing my engine using the OpenGL 3.3 compatibility profile. It runs as expected on my NVIDIA card and on my Intel Card however when I tried it on an AMD setup it ran 3 times worse than on the other setups. Could this be a AMD driver thing or is this probably a problem with my OGL code? Could a different code standard create such bad performance?
    • By Kjell Andersson
      I'm trying to get some legacy OpenGL code to run with a shader pipeline,
      The legacy code uses glVertexPointer(), glColorPointer(), glNormalPointer() and glTexCoordPointer() to supply the vertex information.
      I know that it should be using setVertexAttribPointer() etc to clearly define the layout but that is not an option right now since the legacy code can't be modified to that extent.
      I've got a version 330 vertex shader to somewhat work:
      #version 330 uniform mat4 osg_ModelViewProjectionMatrix; uniform mat4 osg_ModelViewMatrix; layout(location = 0) in vec4 Vertex; layout(location = 2) in vec4 Normal; // Velocity layout(location = 3) in vec3 TexCoord; // TODO: is this the right layout location? out VertexData { vec4 color; vec3 velocity; float size; } VertexOut; void main(void) { vec4 p0 = Vertex; vec4 p1 = Vertex + vec4(Normal.x, Normal.y, Normal.z, 0.0f); vec3 velocity = (osg_ModelViewProjectionMatrix * p1 - osg_ModelViewProjectionMatrix * p0).xyz; VertexOut.velocity = velocity; VertexOut.size = TexCoord.y; gl_Position = osg_ModelViewMatrix * Vertex; } What works is the Vertex and Normal information that the legacy C++ OpenGL code seem to provide in layout location 0 and 2. This is fine.
      What I'm not getting to work is the TexCoord information that is supplied by a glTexCoordPointer() call in C++.
      What layout location is the old standard pipeline using for glTexCoordPointer()? Or is this undefined?
      Side note: I'm trying to get an OpenSceneGraph 3.4.0 particle system to use custom vertex, geometry and fragment shaders for rendering the particles.
    • By markshaw001
      Hi i am new to this forum  i wanted to ask for help from all of you i want to generate real time terrain using a 32 bit heightmap i am good at c++ and have started learning Opengl as i am very interested in making landscapes in opengl i have looked around the internet for help about this topic but i am not getting the hang of the concepts and what they are doing can some here suggests me some good resources for making terrain engine please for example like tutorials,books etc so that i can understand the whole concept of terrain generation.
    • By KarimIO
      Hey guys. I'm trying to get my application to work on my Nvidia GTX 970 desktop. It currently works on my Intel HD 3000 laptop, but on the desktop, every bind textures specifically from framebuffers, I get half a second of lag. This is done 4 times as I have three RGBA textures and one depth 32F buffer. I tried to use debugging software for the first time - RenderDoc only shows SwapBuffers() and no OGL calls, while Nvidia Nsight crashes upon execution, so neither are helpful. Without binding it runs regularly. This does not happen with non-framebuffer binds.
      GLFramebuffer::GLFramebuffer(FramebufferCreateInfo createInfo) { glGenFramebuffers(1, &fbo); glBindFramebuffer(GL_FRAMEBUFFER, fbo); textures = new GLuint[createInfo.numColorTargets]; glGenTextures(createInfo.numColorTargets, textures); GLenum *DrawBuffers = new GLenum[createInfo.numColorTargets]; for (uint32_t i = 0; i < createInfo.numColorTargets; i++) { glBindTexture(GL_TEXTURE_2D, textures[i]); GLint internalFormat; GLenum format; TranslateFormats(createInfo.colorFormats[i], format, internalFormat); // returns GL_RGBA and GL_RGBA glTexImage2D(GL_TEXTURE_2D, 0, internalFormat, createInfo.width, createInfo.height, 0, format, GL_FLOAT, 0); glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_NEAREST); glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_NEAREST); DrawBuffers[i] = GL_COLOR_ATTACHMENT0 + i; glBindTexture(GL_TEXTURE_2D, 0); glFramebufferTexture(GL_FRAMEBUFFER, GL_COLOR_ATTACHMENT0 + i, textures[i], 0); } if (createInfo.depthFormat != FORMAT_DEPTH_NONE) { GLenum depthFormat; switch (createInfo.depthFormat) { case FORMAT_DEPTH_16: depthFormat = GL_DEPTH_COMPONENT16; break; case FORMAT_DEPTH_24: depthFormat = GL_DEPTH_COMPONENT24; break; case FORMAT_DEPTH_32: depthFormat = GL_DEPTH_COMPONENT32; break; case FORMAT_DEPTH_24_STENCIL_8: depthFormat = GL_DEPTH24_STENCIL8; break; case FORMAT_DEPTH_32_STENCIL_8: depthFormat = GL_DEPTH32F_STENCIL8; break; } glGenTextures(1, &depthrenderbuffer); glBindTexture(GL_TEXTURE_2D, depthrenderbuffer); glTexImage2D(GL_TEXTURE_2D, 0, depthFormat, createInfo.width, createInfo.height, 0, GL_DEPTH_COMPONENT, GL_FLOAT, 0); glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_NEAREST); glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_NEAREST); glBindTexture(GL_TEXTURE_2D, 0); glFramebufferTexture(GL_FRAMEBUFFER, GL_DEPTH_ATTACHMENT, depthrenderbuffer, 0); } if (createInfo.numColorTargets > 0) glDrawBuffers(createInfo.numColorTargets, DrawBuffers); else glDrawBuffer(GL_NONE); if (glCheckFramebufferStatus(GL_FRAMEBUFFER) != GL_FRAMEBUFFER_COMPLETE) std::cout << "Framebuffer Incomplete\n"; glBindFramebuffer(GL_FRAMEBUFFER, 0); width = createInfo.width; height = createInfo.height; } // ... // FBO Creation FramebufferCreateInfo gbufferCI; gbufferCI.colorFormats =; gbufferCI.depthFormat = FORMAT_DEPTH_32; gbufferCI.numColorTargets = gbufferCFs.size(); gbufferCI.width = engine.settings.resolutionX; gbufferCI.height = engine.settings.resolutionY; gbufferCI.renderPass = nullptr; gbuffer = graphicsWrapper->CreateFramebuffer(gbufferCI); // Bind glBindFramebuffer(GL_DRAW_FRAMEBUFFER, fbo); // Draw here... // Bind to textures glActiveTexture(GL_TEXTURE0); glBindTexture(GL_TEXTURE_2D, textures[0]); glActiveTexture(GL_TEXTURE1); glBindTexture(GL_TEXTURE_2D, textures[1]); glActiveTexture(GL_TEXTURE2); glBindTexture(GL_TEXTURE_2D, textures[2]); glActiveTexture(GL_TEXTURE3); glBindTexture(GL_TEXTURE_2D, depthrenderbuffer); Here is an extract of my code. I can't think of anything else to include. I've really been butting my head into a wall trying to think of a reason but I can think of none and all my research yields nothing. Thanks in advance!
    • By Adrianensis
      Hi everyone, I've shared my 2D Game Engine source code. It's the result of 4 years working on it (and I still continue improving features ) and I want to share with the community. You can see some videos on youtube and some demo gifs on my twitter account.
      This Engine has been developed as End-of-Degree Project and it is coded in Javascript, WebGL and GLSL. The engine is written from scratch.
      This is not a professional engine but it's for learning purposes, so anyone can review the code an learn basis about graphics, physics or game engine architecture. Source code on this GitHub repository.
      I'm available for a good conversation about Game Engine / Graphics Programming
  • Popular Now