 riperiperi
|
bf77d1cab9
GPU: Pass SpanOrArray for Texture SetData to avoid copy (#3745)
|
há 3 anos atrás |
 gdkchan
|
923089a298
Fast path for Inline-to-Memory texture data transfers (#3610)
|
há 3 anos atrás |
|
gdkchan
|
2232e4ae87
Vulkan backend (#2518)
|
há 3 anos atrás |
|
gdkchan
|
37b6e081da
Fix DMA linear texture copy fast path (#3496)
|
há 3 anos atrás |
 Logan Stromberg
|
6eb85e846f
Reduce some unnecessary allocations in DMA handler (#2886)
|
há 3 anos atrás |
|
gdkchan
|
79408b68c3
De-tile GOB when DMA copying from block linear to pitch kind memory regions (#3207)
|
há 4 anos atrás |
|
gdkchan
|
7bfb5f79b8
When copying linear textures, DMA should ignore region X/Y (#3121)
|
há 4 anos atrás |
|
riperiperi
|
c52158b733
Add timestamp to 16-byte/4-word semaphore releases. (#3049)
|
há 4 anos atrás |
|
riperiperi
|
cda659955c
Texture Sync, incompatible overlap handling, data flush improvements. (#2971)
|
há 4 anos atrás |
|
gdkchan
|
a87f7f2029
Fix DMA copy fast path line size when xCount < stride (#2942)
|
há 4 anos atrás |
|
riperiperi
|
521a07e612
Add support for releasing a semaphore to DmaClass (#2926)
|
há 4 anos atrás |
|
gdkchan
|
ac4ec1a015
Account for negative strides on DMA copy (#2623)
|
há 4 anos atrás |
|
gdkchan
|
ff5df5d8a1
Support non-contiguous copies on I2M and DMA engines (#2473)
|
há 4 anos atrás |
|
gdkchan
|
40b21cc3c4
Separate GPU engines (part 2/2) (#2440)
|
há 4 anos atrás |
|
gdkchan
|
8b44eb1c98
Separate GPU engines and make state follow official docs (part 1/2) (#2422)
|
há 4 anos atrás |
