0.5s lag after binding texture on Nvidia

Started by
11 comments, last by TheChubu 6 years, 7 months ago

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 = gbufferCFs.data();
	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!

Advertisement

Sorry guys, but bump. I still can't find an answer, nor anyone with this issue.

Try profiling it on GPU View and seeing if there's anything in that data to explain what's going on.

8 minutes ago, Hodgman said:

Try profiling it on GPU View and seeing if there's anything in that data to explain what's going on.

Okay I'll try figure it out tomorrow, as I've never used it before and it looks super confusing.

I'm pretty new to GPU View too, so I'm not really sure what to look for :D:| 

I looked through the capture that you PM'ed me, hoping to maybe find that some NVidia thread was busy while your game was stalled, or the GPU was busy with some kind of DMA command or something... but all I can understand from this is that the GPU is idling a lot, and your game's main thread is extremely busy :( 
Son5om2.png

This is what a well-performing capture should look like though -- notice the HW queue and the game's device context are constantly full of queued up work.
XF8d4Vl.png

 

Have you tried adding manual timing code to your game, to try and locate exactly which functions are blocking the CPU? You say that if you disable some code, the performance issue is gone... but try timing different bits of code to see if you can find where the time is going.

Like I said the only issue comes from glbindtexture of a framebuffer texture. Regular textures work fine. Bitting works fine so I know it's not an issue of populating the framebuffer asynchronously. I used breakpoints to figure out the timing and what causes the issue.  Keep in mind vsync is on so that'd why there's not much work to be done. The scene is a simple crytek sponza with no lighting yet (it's enabled on my Intel but I disabled it for now) so there's not many commands.  I haven't multithreaded anything yet so it shouldn't matter if it's idle. This happens no matter how many times I restart so not an issue of Nvidia working on something else. The rest of the frame takes 12ms due to vsync. Also I had all this and so much more running before I improved the rendering architecture (I wrote my own parser and exporter for faster loading, redesigned the rendering wrappers to make vulkan work better, and I made everything draw based on shader and material first rather than object) 

 

Edit: Oh and thank you so much for helping me so far! 

Are you checking for glErrors? Or better, using arb_debug_output or khr_debug_output?

"I AM ZE EMPRAH OPENGL 3.3 THE CORE, I DEMAND FROM THEE ZE SHADERZ AND MATRIXEZ"

My journals: dustArtemis ECS framework and Making a Terrain Generator

1 hour ago, KarimIO said:

I used breakpoints to figure out the timing and what causes the issue.

Which GL function calls contain these massive stalls?

9 hours ago, TheChubu said:

Are you checking for glErrors? Or better, using arb_debug_output or khr_debug_output?

Yeah, the debug output. I only get info and one low warning which is just giving me buffer sizes I think. The latter shows up every frame. 

8 hours ago, Hodgman said:

Which GL function calls contain these massive stalls?

Like I said, glbindtexture but only when used with a framebuffer texture. I've checked the creation of it a hundred times over and don't think there's any problems. 

@Hodgman @TheChubu Sorry but buuump. Any ideas, guys?

This topic is closed to new replies.

Advertisement