032e4c4ca3
gl_rasterizer: Skip zero addr/sized regions on flush/invalidate.
2019-03-16 22:03:19 -04:00
2392e146b0
Merge pull request #2244 from bunnei/gpu-mem-refactor
...
video_core: Refactor to use MemoryManager interface for all memory access.
2019-03-16 21:59:45 -04:00
10118c71e0
memory: Simplify rasterizer cache operations.
2019-03-16 00:41:08 -04:00
574e89d924
video_core: Refactor to use MemoryManager interface for all memory access.
...
# Conflicts:
# src/video_core/engines/kepler_memory.cpp
# src/video_core/engines/maxwell_3d.cpp
# src/video_core/morton.cpp
# src/video_core/morton.h
# src/video_core/renderer_opengl/gl_global_cache.cpp
# src/video_core/renderer_opengl/gl_global_cache.h
# src/video_core/renderer_opengl/gl_rasterizer_cache.cpp
2019-03-16 00:38:48 -04:00
2eaf6c41a4
gpu: Use host address for caching instead of guest address.
2019-03-14 22:34:42 -04:00
84d3cdf7d7
Merge pull request #2233 from ReinUsesLisp/morton-cleanup
...
video_core/morton: Miscellaneous changes
2019-03-14 21:23:12 -04:00
6788ebffc8
Merge pull request #2229 from ReinUsesLisp/vk-sampler-cache
...
vk_sampler_cache: Implement a sampler cache
2019-03-14 21:22:34 -04:00
8bd17aa044
Merge pull request #2216 from ReinUsesLisp/rasterizer-system
...
gl_rasterizer: Use system instance passed from argument
2019-03-14 16:37:05 -04:00
4e6c667586
Merge pull request #2227 from lioncash/override
...
renderer_opengl/gl_global_cache: Add missing override specifiers
2019-03-13 17:05:49 -04:00
ffe2e50458
video_core/morton: Use enum to describe MortonCopyPixels128 mode
2019-03-13 16:35:21 -03:00
6ed6129b4f
video_core/morton: Remove unused parameter in MortonSwizzle
2019-03-13 16:35:10 -03:00
9030a8259f
video_core/morton: Remove clang-format off when it's not needed
2019-03-13 16:16:45 -03:00
fdf76a25ab
video_core/morton: Remove unused functions
2019-03-13 16:15:54 -03:00
a63295a872
video_core/texture: Fix up sampler lod bias
2019-03-13 00:45:54 -03:00
a3734d7e31
vk_sampler_cache: Use operator== instead of memcmp
...
Co-Authored-By: ReinUsesLisp <reinuseslisp@airmail.cc>
2019-03-12 21:05:36 -03:00
aa59d77c3b
vk_sampler_cache: Implement a sampler cache
2019-03-12 20:20:57 -03:00
8ebeb9ade2
video_core/texture: Add a raw representation of TSCEntry
2019-03-12 16:56:29 -03:00
2ad44a453f
Merge pull request #2215 from ReinUsesLisp/samplers
...
gl_rasterizer: Encapsulate sampler queries into methods
2019-03-12 13:10:53 -04:00
3350c0a779
renderer_opengl/gl_global_cache: Replace indexing for assignment with insert_or_assign
...
The previous code had some minor issues with it, really not a big deal,
but amending it is basically 'free', so I figured, "why not?".
With the standard container maps, when:
map[key] = thing;
is done, this can cause potentially undesirable behavior in certain
scenarios. In particular, if there's no value associated with the key,
then the map constructs a default initialized instance of the value
type.
In this case, since it's a std::shared_ptr (as a type alias) that is
the value type, this will construct a std::shared_pointer, and then
assign over it (with objects that are quite large, or actively heap
allocate this can be extremely undesirable).
We also make the function take the region by value, as we can avoid a
copy (and by extension with std::shared_ptr, a copy causes an atomic
reference count increment), in certain scenarios when ownership isn't a
concern (i.e. when ReserveGlobalRegion is called with an rvalue
reference, then no copy at all occurs). So, it's more-or-less a "free"
gain without many downsides.
2019-03-11 12:20:35 -04:00
1070c020db
renderer_opengl/gl_global_cache: Append missing override specifiers
...
Two of the functions here are overridden functions, so we can append
these specifiers to make it explicit.
2019-03-11 12:02:30 -04:00
a6c048920e
gl_rasterizer: Use system instance passed from argument
2019-03-11 03:17:21 -03:00
633ce92908
Merge pull request #2147 from ReinUsesLisp/texture-clean
...
shader_ir: Remove "extras" from the MetaTexture
2019-03-10 17:28:36 -04:00
4a84921b31
Merge pull request #2143 from ReinUsesLisp/texview
...
gl_rasterizer_cache: Create texture views for array discrepancies
2019-03-10 17:27:49 -04:00
a0be7b3b92
gl_rasterizer: Encapsulate sampler queries into methods
2019-03-09 04:35:57 -03:00
6ee0ba64c8
gl_rasterizer: Minor logger changes
2019-03-09 03:34:49 -03:00
160fc63c72
Merge pull request #2209 from lioncash/reorder
...
video_core/gpu_thread: Silence a -Wreorder warning
2019-03-08 12:04:26 -05:00
78c803b4f3
Merge pull request #2208 from lioncash/gpu
...
video_core/gpu: Make GPU's destructor virtual
2019-03-08 12:03:58 -05:00
1143923cdd
Merge pull request #2191 from ReinUsesLisp/maxwell-to-vk
...
maxwell_to_vk: Initial implementation
2019-03-08 11:51:08 -05:00
e7ac5a6adf
dma_pusher: Store command_list_header by copy
...
Instead of holding a reference that will get invalidated by
dma_pushbuffer.pop(), hold it as a copy. This doesn't have any
performance cost since CommandListHeader is 8 bytes long.
2019-03-08 04:06:54 -03:00
c2d4c8b95e
video_core/gpu_thread: Remove unimplemented WaitForIdle function prototype
...
This function didn't have a definition, so we can remove it to prevent
accidentally attempting to use it.
2019-03-07 16:08:52 -05:00
48a461a629
video_core/gpu_thread: Amend constructor initializer list order
...
Moves the data members to satisfy the order they're declared as in the
constructor initializer list.
Silences a -Wreorder warning.
2019-03-07 16:05:49 -05:00
24e2e601d5
video_core/gpu: Make GPU's destructor virtual
...
Because of the recent separation of GPU functionality into sync/async
variants, we need to mark the destructor virtual to provide proper
destruction behavior, given we use the base class within the System
class.
Prior to this, it was undefined behavior whether or not the destructor
in the derived classes would ever execute.
2019-03-07 15:59:45 -05:00
4f352833a5
Merge pull request #2055 from bunnei/gpu-thread
...
Asynchronous GPU command processing
2019-03-07 10:41:53 -05:00
076c76f4e4
Merge pull request #2149 from ReinUsesLisp/decoders-style
...
gl_rasterizer_cache: Move format conversion functions to their own file
2019-03-06 21:56:20 -05:00
84ad81ee67
gpu_thread: Fix deadlock with threading idle state check.
2019-03-06 21:48:57 -05:00
63aa08acbe
gpu_thread: (HACK) Ignore flush on FlushAndInvalidateRegion.
2019-03-06 21:48:57 -05:00
3f1b4fb23a
gpu: Always flush.
2019-03-06 21:48:57 -05:00
aaa373585c
gpu: Refactor a/synchronous implementations into their own classes.
2019-03-06 21:48:57 -05:00
7b574f406b
gpu: Move command processing to another thread.
2019-03-06 21:48:57 -05:00
d2ff93c319
Merge pull request #2190 from lioncash/ogl-global
...
core: Remove the global telemetry accessor function
2019-03-06 21:41:53 -05:00
ac51d048a9
gpu: Refactor command and swap buffers interface for asynch.
2019-03-06 21:09:09 -05:00
4483089d70
gpu: Refactor to take RendererBase instead of RasterizerInterface.
2019-03-06 21:09:09 -05:00
22f105c06d
Merge pull request #2203 from lioncash/engines-include
...
video_core/engines: Remove unnecessary includes
2019-03-06 10:51:27 -05:00
f9ee0dc7ee
video_core/engines: Remove unnecessary includes
...
Removes a few unnecessary dependencies on core-related machinery, such
as the core.h and memory.h, which reduces the amount of rebuilding
necessary if those files change.
This also uncovered some indirect dependencies within other source
files. This also fixes those.
2019-03-05 20:35:32 -05:00
42085ff110
video_core/surface: Remove obsolete TODO in PixelFormatFromRenderTargetFormat()
...
This isn't needed anymore, according to Hexagon
2019-03-05 10:15:06 -05:00
07e13d6728
Merge pull request #2165 from ReinUsesLisp/unbind-tex
...
gl_rasterizer: Unbind textures but don't apply the gl_state
2019-03-04 13:51:59 -05:00
90febaf717
video_core/renderer_opengl: Replace direct usage of global system object accessors
...
We already pass a reference to the system object to the constructor of the renderer,
so we can just use that instead of using the global accessor functions.
2019-03-04 10:24:09 -05:00
1f6571b3de
maxwell_to_vk: Initial implementation
2019-03-04 04:06:05 -03:00
a461e266ea
Merge pull request #2183 from ReinUsesLisp/vk-buffer-cache-clang
...
vk_buffer_cache: Fix clang-format
2019-03-02 14:43:15 -05:00
3c39b39bbc
Merge pull request #2182 from bunnei/my-wasted-friday
...
fuck git for ruining my day, I will learn but I will not forgive
2019-03-02 00:57:15 -05:00
8e84e81e74
vk_buffer_cache: Fix clang-format
2019-03-02 02:16:45 -03:00
ab70c2583d
fuck git for ruining my day, I will learn but I will not forgive
2019-03-02 00:01:34 -05:00
35c105a108
vk_buffer_cache: Implement a buffer cache
...
This buffer cache is just like OpenGL's buffer cache with some minor
style changes. It uses VKStreamBuffer.
2019-03-01 17:33:36 -03:00
e85066dac7
gl_rasterizer: Remove texture unbinding after dispatching a draw call
...
Unbinding was required when OpenGL delete operations didn't unbind a
resource if it was bound. This is no longer needed and can be removed.
2019-02-28 00:17:50 -03:00
bb3ab7d66c
gl_state: Fixup multibind bug
2019-02-28 00:17:03 -03:00
1b13859af8
Merge pull request #2152 from ReinUsesLisp/vk-stream-buffer
...
vk_stream_buffer: Implement a stream buffer
2019-02-27 21:19:15 -05:00
1f5d6a8fed
Merge pull request #2121 from FernandoS27/texception2
...
Improve the Accuracy of the Rasterizer Cache through a Texception Pass
2019-02-27 21:17:55 -05:00
66f4fd4c81
Merge pull request #2172 from lioncash/reorder
...
gl_rasterizer/vk_memory_manager: Silence -Wreorder warnings
2019-02-27 21:14:20 -05:00
7ea097e5c2
Devirtualize Register/Unregister and use a wrapper instead.
2019-02-27 21:58:50 -04:00
5a9204dbd7
Corrections and redesign.
2019-02-27 21:58:49 -04:00
d6b9b51606
Fix linux compile error.
2019-02-27 21:58:48 -04:00
e64fa4d2ea
Remove NotifyFrameBuffer as we are doing a texception pass every drawcall.
2019-02-27 21:58:47 -04:00
3558c88442
Remove certain optimizations that caused texception to fail in certain scenarios.
2019-02-27 21:58:45 -04:00
e9d84ef22c
Bug fixes and formatting
2019-02-27 21:58:44 -04:00
5bc82d124c
rasterizer_cache_gl: Implement Texception Pass
2019-02-27 21:58:43 -04:00
8932001610
rasterizer_cache_gl: Implement Partial Reinterpretation of Surfaces.
2019-02-27 21:58:40 -04:00
44ea2810e4
rasterizer_cache: mark reinterpreted surfaces and add ability to reload marked surfaces on next use.
2019-02-27 21:58:39 -04:00
d583fc1e97
rasterizer_cache_gl: Notify on framebuffer change
2019-02-27 21:58:37 -04:00
45b6d2d349
rasterizer_cache: Expose FlushObject to Child classes and allow redefining of Register and Unregister
2019-02-27 21:57:33 -04:00
f15e2dd881
Merge pull request #2163 from ReinUsesLisp/bitset-dirty
...
maxwell_3d: Use std::bitset to manage dirty flags
2019-02-27 20:50:08 -05:00
27ddbeb01c
gl_rasterizer_cache: Create texture views for array discrepancies
...
When a texture is sampled in a shader with a different array mode than
the cached state, create a texture view and bind that to the shader
instead.
2019-02-27 14:41:06 -03:00
66e023fba2
Merge pull request #2167 from lioncash/namespace
...
common: Move Quaternion, Rectangle, Vec2, Vec3, and Vec4 into the Common namespace
2019-02-27 11:19:53 -05:00
16ea93c11e
vk_memory_manager: Reorder constructor initializer list in terms of member declaration order
...
Reorders members in the order that they would actually be initialized
in. Silences a -Wreorder warning.
2019-02-27 11:08:19 -05:00
a6a783b3dc
gl_rasterizer: Reorder constructor initializer list in terms of member declaration order
...
Orders the members in the order they would actually be initialized in.
Silences a -Wreorder warning.
2019-02-27 11:08:19 -05:00
e7eff72e83
gl_shader_disk_cache: Remove #pragma once from cpp file
...
This is only necessary in headers. Silences a warning with clang.
2019-02-27 11:02:49 -05:00
b9238edd0d
common/math_util: Move contents into the Common namespace
...
These types are within the common library, so they should be within the
Common namespace.
2019-02-27 03:38:39 -05:00
0ad3c031f4
gl_rasterizer_cache: Move format conversion to its own file
2019-02-26 20:08:27 -03:00
0ccd490fcd
decoders: Minor style changes
2019-02-26 20:08:27 -03:00
d91e35a50a
renderer_opengl: Update pixel format tracking
2019-02-26 03:47:16 -03:00
5219edd715
maxwell_3d: Use std::bitset to manage dirty flags
2019-02-26 03:01:48 -03:00
730eb1dad7
vk_stream_buffer: Remove copy code path
2019-02-26 02:09:43 -03:00
5ca63d0675
shader/decode: Remove extras from MetaTexture
2019-02-26 00:11:30 -03:00
48e6f77c03
shader/decode: Split memory and texture instructions decoding
2019-02-26 00:11:30 -03:00
c1b2e35625
shader/track: Resolve variable shadowing warnings
2019-02-25 09:10:59 -05:00
c07987dfab
Merge pull request #2118 from FernandoS27/ipa-improve
...
shader_decompiler: Improve Accuracy of Attribute Interpolation.
2019-02-24 23:04:22 -05:00
c4243c07cc
Merge pull request #2119 from FernandoS27/fix-copy
...
rasterizer_cache_gl: Only do fast layered copy on the same format.
2019-02-24 23:03:52 -05:00
90c780e6f3
Merge pull request #2139 from degasus/dma_pusher
...
video_core/dma_pusher: The full list of headers at once.
2019-02-24 04:15:49 -05:00
33a0597603
vk_stream_buffer: Implement a stream buffer
...
This manages two kinds of streaming buffers: one for unified memory
models and one for dedicated GPUs. The first one skips the copy from the
staging buffer to the real buffer, since it creates an unified buffer.
This implementation waits for all fences to finish their operation
before "invalidating". This is suboptimal since it should allocate
another buffer or start searching from the beginning. There is room for
improvement here.
This could also handle AMD's "pinned" memory (a heap with 256 MiB) that
seems to be designed for buffer streaming.
2019-02-24 04:27:51 -03:00
281a8bf259
vk_resource_manager: Minor VKFenceWatch changes
2019-02-24 04:19:04 -03:00
f7090bacc5
Merge pull request #2146 from ReinUsesLisp/vulkan-scheduler
...
vk_scheduler: Implement a scheduler
2019-02-23 23:32:43 -05:00
d062991643
Merge pull request #2150 from ReinUsesLisp/fixup-layer-swizzle
...
gl_rasterizer_cache: Fixup parameter order in layered swizzle
2019-02-23 23:31:34 -05:00
92050c4d86
vk_memory_manager: Fixup commit interval allocation
...
VKMemoryCommitImpl was using as the end of its interval "begin + end".
That ended up wasting memory.
2019-02-24 01:04:41 -03:00
abef11a540
gl_rasterizer_cache: Fixup parameter order in layered swizzle
2019-02-23 23:27:30 -03:00
f546fb35ed
vk_scheduler: Implement a scheduler
...
The scheduler abstracts command buffer and fence management with an
interface that's able to do OpenGL-like operations on Vulkan command
buffers.
It returns by value a command buffer and fence that have to be used for
subsequent operations until Flush or Finish is executed, after that the
current execution context (the pair of command buffers and fences) gets
invalidated a new one must be fetched. Thankfully validation layers will
quickly detect if this is skipped throwing an error due to modifications
to a sent command buffer.
2019-02-22 01:33:32 -03:00
94b27bb8a5
Merge pull request #2138 from ReinUsesLisp/vulkan-memory-manager
...
vk_memory_manager: Implement memory manager
2019-02-21 22:26:54 -05:00
9539c4203b
Merge pull request #2125 from ReinUsesLisp/fixup-glstate
...
gl_state: Synchronize gl_state even when state is disabled
2019-02-20 21:47:46 -05:00
ae437320c8
Merge pull request #2130 from lioncash/system_engine
...
video_core: Remove usages of System::GetInstance() within the engines
2019-02-20 21:24:56 -05:00
6dd40976d0
video_core/dma_pusher: Simplyfy Step() logic.
...
As fetching command list headers and and the list of command headers is a fixed 1:1 relation now, they can be implemented within a single call.
This cleans up the Step() logic quite a bit.
2019-02-19 10:28:42 +01:00
717394c980
video_core/dma_pusher: The full list of headers at once.
...
Fetching every u32 from memory leads to a big overhead. So let's fetch all of them as a block if possible.
This reduces the Memory::* calls by the dma_pusher by a factor of 10.
2019-02-19 09:58:38 +01:00
b675c97cdd
vk_memory_manager: Implement memory manager
...
A memory manager object handles the memory allocations for a device. It
allocates chunks of Vulkan memory objects and then suballocates.
2019-02-19 03:42:28 -03:00
bunnei
ReinUsesLisp
Mat M
Fernando Sahmkow
Markus Wick