Jump to content
  • Advertisement

hiya83

Member
  • Content Count

    22
  • Joined

  • Last visited

Community Reputation

130 Neutral

About hiya83

  • Rank
    Member

Personal Information

  • Interests
    Programming

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

  1. Hey, I recently upgraded to NSight 5.4.0.17240 (Visual Studio) and opengl debugging just crashes NSight. Anyone has similar experience. I would post on Nvidia DevForums, but they have not answer to any of my previous questions before... Thanks
  2. Vulkan exposes concept of command queues which are abstraction of different engines on the GPU. On the PC, a copy/transfer queue is basically a controller for the DMA engine that you can control independent of the graphics/compute engine. Was wondering if it works the same on mobile. That looks like what I am looking for, I'll give it a shot. Is there something equivalent for iOS? Thanks!
  3. (Posted this in graphics forum too, which was perhaps the wrong forum for it) Hey, I was wondering if on mobile development (Android mainly but iOS as well if you know of it), if there is a GPUView equivalent for whole system debugging so we can figure out if the CPU/GPU are being pipelined efficiently, if there are bubbles, etc. Also slightly tangent question, but do mobile GPU's have a DMA engine exposed as a dedicated Transfer Queue for Vulkan? Thanks!
  4. Hey, I was wondering if on mobile development (Android mainly but iOS as well if you know of it), if there is a GPUView equivalent for whole system debugging so we can figure out if the CPU/GPU are being pipelined efficiently, if there are bubbles, etc. Also slightly tangent question, but do mobile GPU's have a DMA engine exposed as a dedicated Transfer Queue for Vulkan?
  5. hiya83

    Dx12 multinode API's

    A I think I answered my own question, it's for linked node mode which is only applicable when the GPU's are identical (https://developer.nvidia.com/explicit-multi-gpu-programming-directx-12)
  6. Hi, I apologize for creating multiple questions here for dx12 since information is sparse so far. Anyway in some Dx12 api's, there's a node mask for multiple node support, for e.g. D3D12_COMMAND_QUEUE_DESC for CreateCommandQueue & CreateCommandList itself has a node mask. However, these API's are all called on the D3D12Device itself, and when creating the D12Device (D3D12CreateDevice), there isn't an option for it to be a multi-adapter device; you just pass in a single adapter. So what is the point of the node masks in the other API calls? Is there some other use case I am not aware of yet? Thanks!
  7. You're right. I just tried PIX as well and it has the same errors. The same happens with NSight. So it seems if you want to go multi-gpu work on D12 right now, there's no debugger AFAIK (VGD, PIX, NSight) which works at the moment
  8. Well I tried it with vs 2017 community version and it has the same errors.
  9. Hi, Does anyone have suggestions on multi-gpu (cross adapter) debugging in vs2015 for dx12? It seems this is not supported (error codes returned when gpu debugging inside VS) in vs2015. Any idea if this is supported in vs2017? Thanks
  10. hiya83

    Dx12 - timestamp queries

    Thanks for both your responses!
  11. Hi all, Any idea what the mechanism is for timing commands in the GPU command buffer in DX12 now? In DX11 it was BeginDisjoint, End's, EndDisjoint, GetData's. But in DX12, I see I can place EndQuery with a timestamp type, and I also see GetTimestampFrequency to get the frequency, but how do I get the tick value for each of the timestamp EndQuery's? I see ResolveQueryData, but that places the data in a GPU buffer, am I supposed to call that and then copy the data over? If so what is the format of that data? Unfortunately, I have yet to see a sample for timestamp queries yet. Thanks!
  12. hiya83

    Vulkan/D3D11/D3D12 Interop

    Hi, sorry for the much delayed response. Reason I asked is we are now on D11 and for the next update, whether D12 or Vulkan would be a better choice. The primary platform is windows so we don't want that compromised, but we want to support android/iOS as well..
  13. Hi, I tried searching for this but either I failed or couldn't find anything. I know there's D11/D12 interop and there are extensions for GL/D11 (though not very efficient). I was wondering if there's any Vulkan/D11 or Vulkan/D12 interop? Thanks!
  14. Sorry for very late response, but what I meant was this: I have a resource with both uav & srv views, lets say uav0 & srv0, and I call things in following order: OMSetRenderTargetsAndUnorderedAccessViews (rtv bound and uav0 bound to u1) DrawIndexed CSSetUnorderedAccessViews (some other uavs bound to u0 and u1) <------ this does not override the u1 binding above CSSetShaderResources (srv0 bound to t0) <------- this fails since it complains resource already bound as uav Dispatch Hence it seems uav slots for dispatches and draws are not the same? And you're right debug layer caught this, but I would have thought the CSSetUnorderedAccessViews call would have overridden u1 slot and essentially unbind the previous, guess not?..
  15. Ok i hate my life... apparently draw call bindings are different from dispatch bindings?? So the buffer when bound as UAV is written from a draw call pixel shader (i.e. bound using OMSetRenderTargetsAndUnorderedAccessViews to slot u1 since rtv defaults to u0). Then in the following dispatch that uses it as a SRV, I am calling CSSetUnorderedAccessViews with 2 other uav's, thinking that'll overwrite the ones from the previous draw call. Apparently not! I have to call OMSetRenderTargetsAndUnorderedAccessViews again with nulls before the dispatch otherwise it complains of double bind as srv/uav... I guess draw call and dispatch bind slots are separate??
  • Advertisement
×

Important Information

By using GameDev.net, you agree to our community Guidelines, Terms of Use, and Privacy Policy.

We are the game development community.

Whether you are an indie, hobbyist, AAA developer, or just trying to learn, GameDev.net is the place for you to learn, share, and connect with the games industry. Learn more About Us or sign up!

Sign me up!