67f881e714
glsl: Fix floating point compare ops
...
Logic for ordered/unordered ops was wrong.
2021-07-22 21:51:36 -04:00
bd24fa9713
glsl: Query GL Device for FP16 extension support
2021-07-22 21:51:36 -04:00
3482df1176
glsl: Simply FP storage atomics
2021-07-22 21:51:36 -04:00
9cc1b8a873
glsl: F16x2 storage atomics
2021-07-22 21:51:36 -04:00
11ba190462
glsl: Revert ssbo aliasing. Storage Atomics impl
2021-07-22 21:51:36 -04:00
e99d01ff53
glsl: implement phi nodes
2021-07-22 21:51:36 -04:00
3d9ecbe998
glsl: Wip storage atomic ops
2021-07-22 21:51:36 -04:00
df793fc049
glsl: Implement FCMP
2021-07-22 21:51:36 -04:00
cdde730219
glsl: Add a more robust fp formatter
2021-07-22 21:51:36 -04:00
ac7b0ebcb7
glsl: More FP fixes
2021-07-22 21:51:36 -04:00
3064bde415
glsl: FP function fixes
2021-07-22 21:51:36 -04:00
65c6f73e43
glsl: More FP instructions/fixes
2021-07-22 21:51:36 -04:00
5e9095ef22
glsl: Add many FP32/64 instructions
2021-07-22 21:51:36 -04:00
53667ddd4e
glsl: Fixup build issues
2021-07-22 21:51:36 -04:00
ef7bd53f18
glsl: Implement more Integer ops
2021-07-22 21:51:36 -04:00
266a3d60e3
glsl: Implement BF*
2021-07-22 21:51:36 -04:00
0f40b0e61c
glsl: Implement a few Integer instructions
2021-07-22 21:51:36 -04:00
fb75d122a2
glsl: Use std::string_view for Emit function args.
2021-07-22 21:51:35 -04:00
115c162b9a
glsl: Pass IR::Inst& to Emit functions
2021-07-22 21:51:35 -04:00
78f5eb90d7
glsl: INeg and IAdd negate tests
2021-07-22 21:51:35 -04:00
e221baccdd
glsl: Reusable typed variables. IADD32
2021-07-22 21:51:35 -04:00
faf4cd72c5
glsl: Fix program linking and cbuf
2021-07-22 21:51:35 -04:00
64337f004d
glsl: Fix "reg" allocing
...
based on glasm with some tweaks
2021-07-22 21:51:35 -04:00
eaff1030de
glsl: Initial backend
2021-07-22 21:51:35 -04:00
3d822faea1
spirv: Reduce log severity of mismatching denorm rules
2021-07-22 21:51:35 -04:00
7ac55c2a75
shader: Fix loop safety to SSA pass
2021-07-22 21:51:35 -04:00
8fb2048934
vk_rasterizer: Exit render passes on fragment barriers
2021-07-22 21:51:35 -04:00
dbf7cb9f90
vk_graphics_pipeline: Fix path with no VK_EXT_extended_dynamic_state
2021-07-22 21:51:35 -04:00
94e751f415
buffer_cache: Invalidate fast buffers on compute
2021-07-22 21:51:35 -04:00
61cd7dd301
shader: Add logging
2021-07-22 21:51:35 -04:00
373f75d944
shader: Add shader loop safety check settings
...
Also add a setting for enable Nsight Aftermath.
2021-07-22 21:51:35 -04:00
487057b8d2
shader: Comment why the array component is not read in TMML
2021-07-22 21:51:35 -04:00
ba3bdf1d41
vulkan_device: Enable VK_EXT_vertex_input_dynamic_state
2021-07-22 21:51:35 -04:00
41cca8b8ad
vk_pipeline_cache: Skip cached pipelines with different dynamic state
2021-07-22 21:51:35 -04:00
5445799260
main: Fix Open Transferable Shader Cache context item
...
Opens the new shader cache directory location for the specified title, if it exists.
2021-07-22 21:51:35 -04:00
3c125d4134
tmml: Remove index component from coords vec
...
The lod query functions exposed by the rendering API's do not make use of the texturearray layer indexing.
2021-07-22 21:51:35 -04:00
ea038d6653
vulkan: Add VK_EXT_vertex_input_dynamic_state support
...
Reduces the number of total pipelines generated on Vulkan.
Tested on Super Smash Bros. Ultimate.
2021-07-22 21:51:35 -04:00
cb78a1b494
shader: Reorder shader cache directories
2021-07-22 21:51:35 -04:00
3025b2f605
vk_rasterizer: Implement first index
2021-07-22 21:51:35 -04:00
d554778311
vulkan: Use VK_EXT_provoking_vertex when available
2021-07-22 21:51:35 -04:00
d52bacf6f0
spirv/convert: Catch more signed operations oversights
...
The sign bit on integers of size < 32 was not properly preserved in casts
2021-07-22 21:51:35 -04:00
8554a644df
spirv/convert: Catch more broken signed operations on Nvidia OpenGL
...
BitCast U32 to S32 before converting to float on drivers with broken
signed operations.
2021-07-22 21:51:35 -04:00
cd8427367e
gl_buffer_cache: Use unorm internal formats for snorm texture buffer views
...
Fixes black textures in UE4 games
2021-07-22 21:51:35 -04:00
5befc0bf87
shader_environment: Fix local memory size calculations
2021-07-22 21:51:35 -04:00
60a96c49e5
buffer_cache: Fix copy based uniform bindings tracking
2021-07-22 21:51:35 -04:00
15bdd27cac
shader_environment: Add shader_local_memory_crs_size to local memory size
...
Fixes DOOM 2016 missing local memory
2021-07-22 21:51:35 -04:00
7eaa74ad23
gl_texture_cache: Create image storage views
...
Fixes SULD.D tests.
2021-07-22 21:51:35 -04:00
b1ed64ac18
gl_shader_util: Move shader utility code to a separate file
2021-07-22 21:51:35 -04:00
12fe7210d2
gl_shader_cache: Store workers in shader cache object
2021-07-22 21:51:35 -04:00
cffd4716c5
vk_pipeline_cache,shader_notify: Add shader notifications
2021-07-22 21:51:35 -04:00
48aad8dc05
vk_pipeline_cache: Add asynchronous shaders
2021-07-22 21:51:35 -04:00
2a0aeaa3d2
vk_rasterizer: Flush work on clear and dispatches
2021-07-22 21:51:34 -04:00
c736b9ffab
DMA: Restrict optimised path for BlockToLinear further.
2021-07-22 21:51:34 -04:00
f45f7b5c2a
vk_swapchain: Handle outdated swapchains
...
Fixes pixelated presentation on Intel devices.
2021-07-22 21:51:34 -04:00
562af30181
shader: Fix VertexA Shaders.
2021-07-22 21:51:34 -04:00
ec9a78885e
shader: Add 2D and 3D variants to SUATOM and SURED
...
Used by Claybook.
2021-07-22 21:51:34 -04:00
b02c78b276
vk_buffer_cache: Handle null texture buffers
...
Fixes a crash on Age of Calamity cutscenes.
2021-07-22 21:51:34 -04:00
8f099af6a8
nsight_aftermath_tracker: Fix SPIR-V module writes
2021-07-22 21:51:34 -04:00
8c954fcaee
vk_pipeline_cache: Set support_derivative_control to true
2021-07-22 21:51:34 -04:00
4f8b68fb04
shader: Avoid CPU side undefined behavior on I2F
2021-07-22 21:51:34 -04:00
79f2fe1a39
glasm: Use ARB_derivative_control conditionally
2021-07-22 21:51:34 -04:00
4a2361a1e2
buffer_cache: Reduce uniform buffer size from shader usage
...
Increases performance significantly on certain titles.
2021-07-22 21:51:34 -04:00
e57ee3b7fd
transform_feedback: Read buffer stride from index instead of layout
2021-07-22 21:51:34 -04:00
46bd362d0d
fixed_pipeline_state: Use regular for loop instead of ranges for perf
...
MSVC generates better code for it.
2021-07-22 21:51:34 -04:00
d26271b014
vk_swapchain: Avoid recreating the swapchain on each frame
...
Recreate only when requested (or sRGB is changed) instead of tracking
the frontend's size. That size is still used as a hint.
2021-07-22 21:51:34 -04:00
22f0c4f002
emit_glasm_context_get_set: Remove unused variable
2021-07-22 21:51:34 -04:00
5539b13c5a
shader,glasm: Implement legacy texcoord loads
2021-07-22 21:51:34 -04:00
cf9f88e5a7
glasm: Implement legacy varyings
2021-07-22 21:51:34 -04:00
ac0f5d2ab6
shader: Track legacy varyings
2021-07-22 21:51:34 -04:00
05d41fa9b7
shader: Add support for "negative" and unaligned offsets
...
"Negative" offsets don't exist. They are shown as such due to a bug in
nvdisasm.
Unaligned offsets have been proved to read the aligned offset. For
example, when reading an U32, if the offset is 6, the offset read will
be 4.
2021-07-22 21:51:34 -04:00
5d170de0b5
shader: Implement ISCADD32I
2021-07-22 21:51:34 -04:00
adc43297c5
spirv: Fix output generics with components
2021-07-22 21:51:34 -04:00
1148a4eac7
vulkan: Conditionally use shaderInt16
...
Add support for Polaris AMD devices.
2021-07-22 21:51:34 -04:00
77372443c3
vulkan: Enable depth bounds and use it conditionally
...
Intel devices pre-Xe don't support this.
2021-07-22 21:51:34 -04:00
c44b16124f
vk_buffer_cache: Add transform feedback usage to buffers
2021-07-22 21:51:34 -04:00
916ca74324
opengl: Declare fragment outputs even if they are not used
...
Fixes Ori and the Blind Forest's menu on GLASM. For some reason
(probably high level optimizations) it is not sanitized on SPIR-V for
OpenGL. Vulkan is unaffected by this change.
2021-07-22 21:51:34 -04:00
a7e9756671
buffer_cache: Mark uniform buffers as dirty if any enable bit changes
2021-07-22 21:51:34 -04:00
329dea217d
shader: Always initialize up reference in structure control flow
...
Fixes ubsan issue.
2021-07-22 21:51:34 -04:00
99f2c31b64
vulkan_device: Enable float64 and int64 conditionally
...
Add Intel Xe support.
2021-07-22 21:51:34 -04:00
d093522fac
shader: Fix ImageWrite indexing
2021-07-22 21:51:34 -04:00
d738ad4d0b
spirv: Fix image and image buffer descriptor index usage
2021-07-22 21:51:34 -04:00
eb8464cb3d
glasm: Fix immediate texture coordinate
2021-07-22 21:51:34 -04:00
457dda69cc
shader: Clang-format secondary textures
2021-07-22 21:51:34 -04:00
627161c38e
shader: Fix secondary textures
2021-07-22 21:51:34 -04:00
dd39b87b0c
shader: Adhere to disk shader cache setting
2021-07-22 21:51:34 -04:00
b659212dbd
shader: Fix TMML queries
2021-07-22 21:51:34 -04:00
fbf5cdcba0
shader: Fix FSwizzleAdd folding when going through phi nodes
2021-07-22 21:51:34 -04:00
871c9f1ced
shader/exception: Fix compilation errors on gcc
2021-07-22 21:51:34 -04:00
b6c087496b
glasm: Reduce reg allocation leaks from an exception to a log
2021-07-22 21:51:34 -04:00
56d4a9ebde
texture_cache: Reduce invalid image/sampler error severity
2021-07-22 21:51:34 -04:00
b7764c3a79
shader: Handle host exceptions
2021-07-22 21:51:34 -04:00
83db7abae6
glasm: Use integer lod for TXQ
2021-07-22 21:51:33 -04:00
3b595fe8b2
glasm: Prepare XFB from state instead of global registers
2021-07-22 21:51:33 -04:00
e240a62017
glasm: Fix global memory fallbacks
2021-07-22 21:51:33 -04:00
8f3043c3cf
Revert "glasm: Skip phi moves on undefined instructions"
...
Causes regressions on Bowser's Fury.
2021-07-22 21:51:33 -04:00
2aa30353b7
glasm: Remove unintentional '\n' on Undef32
2021-07-22 21:51:33 -04:00
adb591a757
glasm: Use storage buffers instead of global memory when possible
2021-07-22 21:51:33 -04:00
f58f79c85d
glasm: Implement Y direction
2021-07-22 21:51:33 -04:00
586c785366
glasm: Skip phi moves on undefined instructions
2021-07-22 21:51:33 -04:00
b9c8814ea9
glasm: Implement undef instructions
2021-07-22 21:51:33 -04:00
8763cc1ff7
glasm: Fix global memory callbacks
2021-07-22 21:51:33 -04:00
a41b2ed391
gl_shader_cache: Add disk shader cache
2021-07-22 21:51:33 -04:00
a49532c8eb
video_core,shader: Clang-format fixes
2021-07-22 21:51:33 -04:00
eacf18cce9
gl_shader_cache: Rename Program abstractions into Pipeline
2021-07-22 21:51:33 -04:00
48aafe0961
glasm: Release phi node registers after they are no longer needed
2021-07-22 21:51:33 -04:00
77ee733c3a
glasm: Remove unintentionally committed fmt::prints
2021-07-22 21:51:33 -04:00
70c9281fbf
glasm: Fix INeg32 on negative immediates
2021-07-22 21:51:33 -04:00
75fd0079db
glasm: Remove unnecessary value types
2021-07-22 21:51:33 -04:00
379b305b4b
glasm: Throw when there are register leaks
2021-07-22 21:51:33 -04:00
ca05a13c62
glasm: Catch more register leaks
...
Add support for null registers. These are used when an instruction has
no usages.
This comes handy when an instruction is only used for its CC value, with
the caveat of having to invalidate all pseudo-instructions before
defining the instruction itself in the register allocator. This commits
changes this.
Workaround a bug on Nvidia's condition codes conditional execution using
branches.
2021-07-22 21:51:33 -04:00
9fbfe7d676
glasm: Fix usage counting on phi nodes
2021-07-22 21:51:33 -04:00
4017928213
gl_shader_cache: Do not flip tessellation on OpenGL
2021-07-22 21:51:33 -04:00
80884e3270
gl_graphics_program: Fix texture buffer bindings
2021-07-22 21:51:33 -04:00
c721767bcc
glasm: Implement global memory fallbacks
2021-07-22 21:51:33 -04:00
0794273870
glasm: Implement int64 add and subtract
2021-07-22 21:51:33 -04:00
7fdf0d7d33
emit_glasm_context_get_set: Remove unused variable
2021-07-22 21:51:33 -04:00
e30d4fa976
glasm: Implement indirect attribute loads
2021-07-22 21:51:33 -04:00
c8414e686f
glasm: Implement image atomics
2021-07-22 21:51:33 -04:00
3a7ca6a7db
glasm: Reorder unreachable image atomic insts
...
Reorder them to the bottom of the file for readability.
2021-07-22 21:51:33 -04:00
e565eb361a
glasm: Implement gl_Layer stores
2021-07-22 21:51:33 -04:00
89e341d56a
glasm: Implement SampleId
2021-07-22 21:51:33 -04:00
77d8c44b68
glasm: Implement IsHelperInvocation
2021-07-22 21:51:33 -04:00
ddf601919f
glasm: Fix EmitVertex's optimization
2021-07-22 21:51:33 -04:00
1bccb43cbe
gl_shader_cache: Conditionally use viewport mask
2021-07-22 21:51:33 -04:00
c31521512f
gl_shader_cache,glasm: Conditionally use typeless image reads extension
2021-07-22 21:51:33 -04:00
df406246d9
gl_shader_cache: Improve GLASM error print logic
2021-07-22 21:51:33 -04:00
84feabac88
glasm: Implement forced early Z
2021-07-22 21:51:33 -04:00
6bc54e12a0
glasm: Set transform feedback state
2021-07-22 21:51:33 -04:00
69b910e9e7
video_core: Abstract transform feedback translation utility
2021-07-22 21:51:33 -04:00
7dadb2bef3
glasm: Simplify patch reads
2021-07-22 21:51:33 -04:00
b382f57b28
glasm: Fix output patch reads
...
With this, Luigi's Mansion's sand renders properly.
2021-07-22 21:51:33 -04:00
c07cc9d6a5
gl_shader_cache: Pass shader runtime information
2021-07-22 21:51:33 -04:00
9e7b6622c2
shader: Split profile and runtime information in separate structs
2021-07-22 21:51:33 -04:00
eb15667905
emit_glasm_context_get_and_set.cpp: Add missing semicolons
2021-07-22 21:51:33 -04:00
781a87175c
glasm: Fix patch attribute declarations
2021-07-22 21:51:33 -04:00
36d040da70
glasm: Implement FSWZADD
2021-07-22 21:51:33 -04:00
3da7b98d37
glasm: Implement PrimitiveId attribute read
2021-07-22 21:51:33 -04:00
394b96a2fe
glasm: Implement clip distance stores
2021-07-22 21:51:32 -04:00
a5d978e91e
glasm: Fix tessellation input attributes
2021-07-22 21:51:32 -04:00
0d7d85c81e
glasm: Add missing semicolon on tesscoord reading
2021-07-22 21:51:32 -04:00
48d4e26326
glasm: Fix tessellation headers
2021-07-22 21:51:32 -04:00
9ec2303ad6
glasm: Add tessellation shader declarations
2021-07-22 21:51:32 -04:00
2913ca811e
glasm: Implement TessellationEvaluationPoint
2021-07-22 21:51:32 -04:00
54decced92
gl_shader_manager: Zero initialize current assembly programs
2021-07-22 21:51:32 -04:00
c0e4074721
gl_shader_manager: Remove unintentionally committed #pragma
2021-07-22 21:51:32 -04:00
a569ac418e
glasm: Implement patch memory
2021-07-22 21:51:32 -04:00
164b8c1ec5
glasm: Fix InvocationId declaration
2021-07-22 21:51:32 -04:00
d5db96386d
glasm: Implement InvocationId
2021-07-22 21:51:32 -04:00
679e7146a7
glasm: Optimize EmitVertex into EMIT
2021-07-22 21:51:32 -04:00
79929be833
glasm: Implement geometry shader attribute reads
2021-07-22 21:51:32 -04:00
83cef0426b
glasm: Properly declare attributes on geometry programs
2021-07-22 21:51:32 -04:00
fad139a3e6
glasm: Declare geometry program headers
2021-07-22 21:51:32 -04:00
690b1841e6
renderer_opengl: State track compute assembly programs
2021-07-22 21:51:32 -04:00
c5ca4fe451
renderer_opengl: State track assembly programs
2021-07-22 21:51:32 -04:00
0a54291c9c
glasm: Fix potential aliasing bug on cube array samples
2021-07-22 21:51:32 -04:00
8fdb00a2b5
glasm: Implement ImageWrite
2021-07-22 21:51:32 -04:00
dadd192b30
glasm: Implement ImageRead
2021-07-22 21:51:32 -04:00
3d0ffc6ad0
glasm: Implement EmitVertex and EndPrimitive
2021-07-22 21:51:32 -04:00
f79cbbf814
glasm: Implement ImageGradient
2021-07-22 21:51:32 -04:00
291f220be3
glasm: Implement 64-bit shifts
2021-07-22 21:51:32 -04:00
d957b3a8fe
glasm: Implement barriers
2021-07-22 21:51:32 -04:00
b60b3fa113
glasm: Fix compute stage name
2021-07-22 21:51:32 -04:00
96962c1d3c
glasm: Fix phi instruction types
2021-07-22 21:51:32 -04:00
91a3c2c1c0
glasm: Implement PREC on relevant instructions
2021-07-22 21:51:32 -04:00
accad56ee7
glasm: Implement stores to gl_ViewportIndex
2021-07-22 21:51:32 -04:00
2494dbe183
glasm: Implement gl_PointSize stores
2021-07-22 21:51:32 -04:00
9415c435fc
glasm: Implement gl_PointCoord
2021-07-22 21:51:32 -04:00
12dcb9fcc2
glasm: Implement ImageQueryLod
2021-07-22 21:51:32 -04:00
4a22942f45
glasm: Implement ImageFetch
2021-07-22 21:51:32 -04:00
3777592ada
glasm: Implement IADD.CC
2021-07-22 21:51:32 -04:00
98ed8ff103
glasm: Implement BFE.CC
2021-07-22 21:51:32 -04:00
2e0d56da7e
glasm: Implement SelectU1
2021-07-22 21:51:32 -04:00
85fc7e584e
HACK: Bind stages before and after bindings
...
Works around a bug where program parameters are only applied to the
current stage, and this one wasn't bound at the moment.
Affects all SSBO usages on GLASM.
2021-07-22 21:51:32 -04:00
e8ed904805
glasm: Implement gl_WorkGroupID
2021-07-22 21:51:32 -04:00
0a42277a4f
glasm: Implement TXQ and improve texture info reads
2021-07-22 21:51:32 -04:00
c560bf99c2
glasm: Implement gl_FrongFacing attribute
2021-07-22 21:51:32 -04:00
8b7d5912d6
glasm: Support textures used in more than one stage
2021-07-22 21:51:32 -04:00
3d3ed53511
glasm: Implement textureGather instructions
2021-07-22 21:51:32 -04:00
0fa421f82f
glasm: Implement gl_FragDepth and gl_SampleMask stores
2021-07-22 21:51:32 -04:00
1ee7f8b943
glasm: Do not alias ConditionRef for now
...
Immediate condition refs where not handled correctly. Just move the
value for now.
2021-07-22 21:51:32 -04:00
9bb3e008c9
shader: Read branch conditions from an instruction
...
Fixes the identity removal pass.
2021-07-22 21:51:32 -04:00
4bad415bca
glasm: Implement InstanceId and VertexId
2021-07-22 21:51:31 -04:00
afcb140185
glasm: Add missing return value on move assignment
2021-07-22 21:51:31 -04:00
fb3ba62b3a
glasm: Fix aliased bitcasts ref counting
2021-07-22 21:51:31 -04:00
f1b334b9f9
glasm: Remove unintentional comma on vector insert
2021-07-22 21:51:31 -04:00
ec6fc5fe78
glasm: Implement TEX and TEXS instructions
...
Remove lod clamp from texture instructions with lod, as this is not
needed (nor supported).
2021-07-22 21:51:31 -04:00
c42a6143a5
glasm: Add support for non-2D texture samples
2021-07-22 21:51:31 -04:00
bee9fb0563
glasm: Reorder unreachable image instructions to the bottom
2021-07-22 21:51:31 -04:00
e6b4d461d2
glasm: Add support for texture offsets
2021-07-22 21:51:31 -04:00
bf2949df10
glasm: Improve texture sampling instructions
2021-07-22 21:51:31 -04:00
db2f0f4108
emit_glasm: Enable ARB_draw_buffers when needed
2021-07-22 21:51:31 -04:00
3c06293e20
emit_glasm: Add support for reading position attributes
2021-07-22 21:51:31 -04:00
f7a2340205
shader_recompiler: GCC fixes
...
Fixes members of unnamed union not being accessible, and one function
without a declaration.
2021-07-22 21:51:31 -04:00
d4f9c798d6
glasm: Implement rest of shared mem
2021-07-22 21:51:31 -04:00
258f2dec1b
opengl: Initial (broken) support to GLASM shaders
2021-07-22 21:51:31 -04:00
776ab3ea12
shader: Use a non-trivial dummy to construct ASL node union
2021-07-22 21:51:31 -04:00
38e7b8c805
emit_spirv: Jump to loop body with local variable
...
Silence unused variable warning
2021-07-22 21:51:31 -04:00
464f13fe0b
glasm: Implement derivative instructions on GLASM
2021-07-22 21:51:31 -04:00
9fb2ea08e8
glasm: Initial (broken) implementation of TEX on GLASM
2021-07-22 21:51:31 -04:00
1f3446b47e
glasm: Implement some graphics instructions on GLASM
2021-07-22 21:51:31 -04:00
31d402ee74
glasm: Add Void type to GLASM values
2021-07-22 21:51:31 -04:00
3764750339
glasm: Add graphics specific shader declarations to GLASM
2021-07-22 21:51:31 -04:00
057dee4856
glasm: Implement local memory for glasm
2021-07-22 21:51:31 -04:00
ab5dbe7c29
emit_spirv: Add missing block in case
2021-07-22 21:51:31 -04:00
bf5e48ffe4
glasm: Initial implementation of phi nodes on GLASM
2021-07-22 21:51:31 -04:00
0f88fb5d72
glasm: Write result to scalar on integer comparison instructions
2021-07-22 21:51:31 -04:00
d4385c34e3
glasm: Declare NV_shader_thread_group when needed
2021-07-22 21:51:31 -04:00
568d813eea
vk_update_descriptor: Properly initialize payload on the update descriptor queue
2021-07-22 21:51:31 -04:00
d54d7de40e
glasm: Rework control flow introducing a syntax list
...
This commit regresses VertexA shaders, their transformation pass has to
be adapted to the new control flow.
2021-07-22 21:51:31 -04:00
7ff5851608
glasm: Implement Storage atomics
...
StorageAtomicExchangeU64 is failing test seemingly due to failure storing 64-bit
result into the register
2021-07-22 21:51:31 -04:00
8c81a20ace
glasm: Ensure reg alloc order across compilers on GLASM
...
Use a struct constructor to serialize register allocation arguments to
ensure registers are allocated in the same order regardless of the
compiler used.
The A and B functions can be called in any order when passed as
arguments to "foo":
foo(A(), B())
But the order is guaranteed for curly-braced constructor calls in
classes:
Foo{A(), B()}
Use this to get consistent behavior.
2021-07-22 21:51:31 -04:00
c917290497
glasm: Enable unintentionally disabled register aliasing on GLASM
2021-07-22 21:51:31 -04:00
70fbede213
glasm: Review all GLASM insts to be aware of register aliasing
2021-07-22 21:51:31 -04:00
c4fd6b55bc
glasm: Implement shuffle and vote instructions on GLASM
2021-07-22 21:51:31 -04:00
decda4a2c7
glasm: Add MUFU instructions to GLASM
2021-07-22 21:51:31 -04:00
5b18a12df2
glasm: Implement IAbs64 and INeg64 on GLASM
2021-07-22 21:51:31 -04:00
3b6a632237
shader: Add floating-point rounding to I2F
2021-07-22 21:51:31 -04:00
3f00a2ad3f
glasm: Properly clamp Fp64 on GLASM
2021-07-22 21:51:31 -04:00
deda89372f
glasm: Fix register allocation when moving immediate on GLASM
2021-07-22 21:51:31 -04:00
0839e46736
glasm: Implement SelectU64 on GLASM
2021-07-22 21:51:31 -04:00
6237300e36
glasm: Fix clamps so the min value has priority on NAN on GLASM
2021-07-22 21:51:31 -04:00
8eb72ff0dc
glasm: Fix moving U64 immediates to registers in GLASM
2021-07-22 21:51:31 -04:00
80813b1d14
glasm: Implement storage atomic ops
2021-07-22 21:51:31 -04:00
ad61b47f80
glasm: Add conversion instructions to GLASM
2021-07-22 21:51:31 -04:00
7703d65f23
glasm: Add fp min/max insts and fix store for fp64 on GLASM
2021-07-22 21:51:31 -04:00
43a448d98d
glasm: Add logical instructions on GLASM
2021-07-22 21:51:31 -04:00
99352741af
glasm: Remove duplicated Fp64 pack instructions on GLASM
2021-07-22 21:51:30 -04:00
45ef62d3ba
glasm: Remove unnecesary new white space on Clamp GLASM
2021-07-22 21:51:30 -04:00
b4953e79ee
glasm: Add floating-point comparisons on GLASM
2021-07-22 21:51:30 -04:00
6705f56029
emit_glasm: Implement more integer alu ops
2021-07-22 21:51:30 -04:00
3e10709091
glasm: Reimplement bitwise ops and BFI/BFE
2021-07-22 21:51:30 -04:00
4502595bc2
glasm: Initial GLASM fp64 support
2021-07-22 21:51:30 -04:00
9f851e3832
glasm: Implement GLASM fp16 packing and move bitwise insns
2021-07-22 21:51:30 -04:00
4de65fbff4
glasm: Remove unused functions left from rebase
2021-07-22 21:51:30 -04:00
6358b0d0c1
glasm: Specify namespace when using FormatTo
2021-07-22 21:51:30 -04:00
939dab7120
glasm: Implement more GLASM composite instructions
2021-07-22 21:51:30 -04:00
01e18581b9
vk_pipeline_cache: Enable int8 and int16 types on Vulkan
2021-07-22 21:51:30 -04:00
1c9307969c
glasm: Make GLASM aware of types
2021-07-22 21:51:30 -04:00
934d300246
glasm: Use CMP.S for Select32
...
also fixes ADD and SUB to use U modifier
2021-07-22 21:51:30 -04:00
68cc445b8e
glasm: Implement more logical ops
2021-07-22 21:51:30 -04:00
941c6dc740
glasm: Implement BFI, BFE
...
Along with implementations of common instructions along the way
2021-07-22 21:51:30 -04:00
3e841f6441
glasm: Use BitField instead of C bitfields
2021-07-22 21:51:30 -04:00
2b04b4d27f
glasm: Remove unused argument in identity instructions on GLASM
2021-07-22 21:51:30 -04:00
dc02cb92e4
gl_rasterizer: Flush L2 caches before glFlush on GLASM
2021-07-22 21:51:30 -04:00
2c81ad8311
glasm: Initial GLASM compute implementation for testing
2021-07-22 21:51:30 -04:00
6fd190d1ae
glasm: Implement basic GLASM instructions
2021-07-22 21:51:30 -04:00
c1ba685d9c
glasm: Changes to GLASM register allocator and emit context
2021-07-22 21:51:30 -04:00
36f1586267
vk_scheduler: Use locks instead of SPSC a queue
...
This tries to fix a data race where we'd wait forever for the GPU.
2021-07-22 21:51:30 -04:00
56c47951c5
vk_query_cache: Wait before reading queries
2021-07-22 21:51:30 -04:00
a515036604
vk_master_semaphore: Use fetch_add to increase master semaphore tick
2021-07-22 21:51:30 -04:00
b10cf64c48
glasm: Add GLASM backend infrastructure
2021-07-22 21:51:30 -04:00
09dc23f971
shader: ISET.X implementation
2021-07-22 21:51:30 -04:00
bfa47539f6
gl_shader_cache: Remove code unintentionally committed
2021-07-22 21:51:30 -04:00
b725db8709
shader: Fixup SPIR-V emit header namespaces
2021-07-22 21:51:30 -04:00
bed090807a
Move SPIR-V emission functions to their own header
2021-07-22 21:51:30 -04:00
ee61ec2c39
shader: Optimize NVN Fallthrough
2021-07-22 21:51:30 -04:00
153a77efee
shader: Stub SR_AFFINITY
2021-07-22 21:51:30 -04:00
7ecc6de56a
shader: Implement Int32 SUATOM/SURED
2021-07-22 21:51:30 -04:00
d621e96d0d
shader: Initial OpenGL implementation
2021-07-22 21:51:30 -04:00
850b08a16c
spirv: Be aware of NAN unaware drivers
2021-07-22 21:51:30 -04:00
fde47152d9
spirv: Add SSBO read fallbacks when no aliasing is available
2021-07-22 21:51:29 -04:00
fd913bceaf
spirv: Add OpKill fallback to demote
2021-07-22 21:51:29 -04:00
d2a0f9d7ad
spirv: Do not enable ShaderLayer
...
This is enabled by an extension instead of the capability.
2021-07-22 21:51:29 -04:00
2b434b74af
spirv: Enable DemoteToHelperInvocationEXT only when supported
2021-07-22 21:51:29 -04:00
cfd873275d
spirv: Use OriginLowerLeft when requested
2021-07-22 21:51:29 -04:00
bafe9e35a9
spirv: Only add image operands mask when needed
2021-07-22 21:51:29 -04:00
d2e811db2e
spirv: Workaround image unsigned offset bug
...
Workaround bug on Nvidia's OpenGL SPIR-V compiler when using unsigned
texture offsets.
2021-07-22 21:51:29 -04:00
4ead714910
spirv: Add int8 and int16 capabilities only when supported
2021-07-22 21:51:29 -04:00
33bebc3412
spirv: Add integer clamping workarounds
...
Workaround more bugs on Nvidia's OpenGL SPIR-V compiler.
2021-07-22 21:51:29 -04:00
7b03b97118
spirv: Implement int8 and int16 conversion fallbacks
2021-07-22 21:51:29 -04:00
48a17298d7
spirv: Support OpenGL uniform buffers and change bindings
2021-07-22 21:51:29 -04:00
d5d6778ba5
spirv: Desambiguate descriptor names
...
Worksaround a bug on Nvidia's OpenGL SPIR-V compiler where names are
used for name matching.
2021-07-22 21:51:29 -04:00
a46d91b1ef
shader: Add OpenGL shader profile options
2021-07-22 21:51:29 -04:00
028f0033bd
shader: Remove shader util
2021-07-22 21:51:29 -04:00
c49d56c931
shader: Address feedback
2021-07-22 21:51:29 -04:00
b541f5e5e3
shader: Implement VertexA stage
2021-07-22 21:51:29 -04:00
da936d6ad8
shader: Implement delegation of Exit to dispatcher on CFG
2021-07-22 21:51:29 -04:00
f4b82b8dd7
vk_graphics_pipeline: Fix texture buffer descriptors
2021-07-22 21:51:29 -04:00
fb14820c86
shader: Fix IADD3.CC
2021-07-22 21:51:29 -04:00
53acdda772
vk_scheduler: Allow command submission on worker thread
...
This changes how Scheduler::Flush works. It queues the current command
buffer to be sent to the GPU but does not do it immediately. The Vulkan
worker thread takes care of that. Users will have to use
Scheduler::Flush + Scheduler::WaitWorker to get the previous behavior.
Scheduler::Finish is unchanged.
To avoid waiting on work never queued, Scheduler::Wait sends the current
command buffer if that's what the caller wants to wait.
2021-07-22 21:51:29 -04:00
c5425b38c1
vk_compute_pass: Fix -Wshadow warning
2021-07-22 21:51:29 -04:00
025b20f96a
shader: Move pipeline cache logic to separate files
...
Move code to separate files to be able to reuse it from OpenGL. This
greatly simplifies the pipeline cache logic on Vulkan.
Transform feedback state is not yet abstracted and it's still
intrusively stored inside vk_pipeline_cache. It will be moved when
needed on OpenGL.
2021-07-22 21:51:29 -04:00
ac8835659e
vulkan: Defer descriptor set work to the Vulkan thread
...
Move descriptor lookup and update code to a separate thread. Delaying
this removes work from the main GPU thread and allows creating
descriptor layouts on another thread. This reduces a bit the workload
of the main thread when new pipelines are encountered.
2021-07-22 21:51:29 -04:00
2f3c3dfc10
vulkan: Rework descriptor allocation algorithm
...
Create multiple descriptor pools on demand. There are some degrees of
freedom what is considered a compatible pool to avoid wasting large
pools on small descriptors.
2021-07-22 21:51:29 -04:00
5ed871398b
vk_graphics_pipeline: Generate specialized pipeline config functions and improve code
2021-07-22 21:51:29 -04:00
f4ace63957
shader: Accelerate pipeline transitions and use dirty flags for shaders
2021-07-22 21:51:29 -04:00
20e86fd615
shader: Fix BFE s32 undefined check
...
Our unit tests were hitting this exception.
2021-07-22 21:51:29 -04:00
8fda599a31
vk_compute_pipeline: Fix index comparison oversight on compute texture buffers
2021-07-22 21:51:29 -04:00
50eb03382e
shader: Fix error checking in bitfieldExtract and implement bitfieldInsert folding
2021-07-22 21:51:29 -04:00
0c0ee9d897
vulkan_device: Require shaderClipDistance and shaderCullDistance features
2021-07-22 21:51:29 -04:00
5b1b06f11e
vk_graphics_pipeline: Guard against non-tessellation pipelines using patches
2021-07-22 21:51:29 -04:00
57464e3b72
shader: Fix storage type when reading patches on tess control
2021-07-22 21:51:29 -04:00
d2b54c6e42
shader: Fix VMNMX selector B
2021-07-22 21:51:29 -04:00
2dc86372c7
shader: Fix bugs and build issues on GCC
2021-07-22 21:51:29 -04:00
7a1f296cda
shader: Fix render targets with null attachments
2021-07-22 21:51:29 -04:00
155be4a8d3
shader: Increase the maximum number of storage buffers
...
Compute shaders spill uniform buffers on storage buffers, increasing the
expected number.
2021-07-22 21:51:29 -04:00
fe25f42403
shader: Remove identity removal pass for better build times
2021-07-22 21:51:29 -04:00
0c7230a606
shader: Add more strict validation the pass
2021-07-22 21:51:29 -04:00
25949b864c
shader: Fix forward referencing identity instructions when inserting phi
2021-07-22 21:51:29 -04:00
92a01984e6
shader: Remove invalidated blocks in dead code elimination pass
2021-07-22 21:51:29 -04:00
ameerj
ReinUsesLisp
Rodrigo Locatti
lat9nq
FernandoS27