Mirrors/citra-nightly
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

video_core, core: Move pixel download to its own thread

Browse source 
This uses the mailbox model to move pixel downloading to its own thread, eliminating Nvidia's warnings and (possibly) making use of GPU copy engine.

To achieve this, we created a new mailbox type that is different from the presentation mailbox in that it never discards a rendered frame.

Also, I tweaked the projection matrix thing so that it can just draw the frame upside down instead of having the CPU flip it.
This commit is contained in:
zhupengfei 2020-01-28 21:57:30 +08:00
parent 5b54a99f96
commit 06a0d86e9c
No known key found for this signature in database
GPG key ID: DD129E108BD09378
7 changed files with 324 additions and 169 deletions
Download patch file Download diff file Expand all files Collapse all files

12
src/core/dumping/backend.cpp
View file

@ -8,17 +8,7 @@
namespace VideoDumper {
VideoFrame::VideoFrame(std::size_t width_, std::size_t height_, u8* data_)
: width(width_), height(height_), stride(width * 4), data(width * height * 4) {
// While copying, rotate the image to put the pixels in correct order
// (As OpenGL returns pixel data starting from the lowest position)
for (std::size_t i = 0; i < height; i++) {
for (std::size_t j = 0; j < width; j++) {
for (std::size_t k = 0; k < 4; k++) {
data[i * stride + j * 4 + k] = data_[(height - i - 1) * stride + j * 4 + k];
}
}
}
}
: width(width_), height(height_), stride(width * 4), data(data_, data_ + width * height * 4) {}
Backend::~Backend() = default;
NullBackend::~NullBackend() = default;