swung0x48 37111ae992 [Perf] (DirectVulkan): snapshot-gated consecutive-draw fast path skips SetupDraw re-resolution

2026-07-30 09:40:54 -04:00

..

BufferArena.cpp

[Feat] (MG_Backend/DirectVulkan): Buffer arena, buffer slice

2026-03-19 17:23:52 +08:00

BufferArena.h

[Feat] (MG_Backend/DirectVulkan): Buffer arena, buffer slice

2026-03-19 17:23:52 +08:00

BufferSlice.h

[Feat] (MG_Backend/DirectVulkan): Buffer arena, buffer slice

2026-03-19 17:23:52 +08:00

FrameContext.cpp

[Feat] (DirectVulkan): pre-pass command stream for reorderable out-of-pass work

2026-07-30 02:43:13 -04:00

FrameContext.h

[Feat] (DirectVulkan): pre-pass command stream for reorderable out-of-pass work

2026-07-30 02:43:13 -04:00

PipelineFactory.cpp

[Fix] (DirectVulkan): harden the leak-fix round after adversarial review - pipeline memo now drops at every command-buffer boundary (a flush-loop-memoized pipeline could age out and be destroyed while its submission was in flight), mid-frame drains no longer rewind the arena or advance the cache-aging clocks in presenting apps (gated to every 8th drain since the last Present, so readback/fence-heavy frames neither churn conversions nor shrink the 1024-boundary retire window), render-pass eviction notifies the pipeline cache once per sweep batch instead of once per dying pass, descriptor pools use FREE_DESCRIPTOR_SET_BIT so a destroyed layout's cached sets are freed back and credited instead of abandoning pool slots (the live-layout age sweep that could orphan slots is removed - layout destruction is the sole purge path), and renderbuffer respecify parks the old backing for aged destruction instead of destroying it while possibly in flight

2026-07-27 22:51:26 -04:00

PipelineFactory.h

[Fix] (DirectVulkan): harden the leak-fix round after adversarial review - pipeline memo now drops at every command-buffer boundary (a flush-loop-memoized pipeline could age out and be destroyed while its submission was in flight), mid-frame drains no longer rewind the arena or advance the cache-aging clocks in presenting apps (gated to every 8th drain since the last Present, so readback/fence-heavy frames neither churn conversions nor shrink the 1024-boundary retire window), render-pass eviction notifies the pipeline cache once per sweep batch instead of once per dying pass, descriptor pools use FREE_DESCRIPTOR_SET_BIT so a destroyed layout's cached sets are freed back and credited instead of abandoning pool slots (the live-layout age sweep that could orphan slots is removed - layout destruction is the sole purge path), and renderbuffer respecify parks the old backing for aged destruction instead of destroying it while possibly in flight

2026-07-27 22:51:26 -04:00

ProgramFactory.cpp

[Fix] (DirectVulkan): rewrite implicit-LOD fragment samples to explicit LOD 0 when every bound sampler is pinned to a single mip level - Adreno 650 (driver 512.502) reads outside a full-screen colour render target's allocation on its implicit-LOD sampling path and faults the GPU, which killed MC 26.2 on its own blit shader (texture(InSampler, texCoord)) between frames 344-421 on every run; this is the same driver defect the default-framebuffer blit shader already works around with textureLod, but an application's shader cannot be edited, so ForceExplicitLod0SamplePass converts OpImageSample*ImplicitLod to the explicit form at the SPIR-V level under a new CompileOptionBit that is only requested when the rewrite provably cannot move a texel (every sampler binding on a single-level view, no anisotropy, and either a LOD clamp that already pins lambda at 0 or min and mag filters that agree - an explicit LOD 0 always takes the magnification side of the min/mag decision); a single-level view now also clamps its sampler to mipmapMode NEAREST with maxLod min(maxLod, 0.25) rather than 0, since collapsing the clamp would make every fragment magnify and quietly retire the min filter; and the program's backend hash memo grows from one slot to four so a program resolved under two compile-flag sets in the same frame stops re-hashing every stage's SPIR-V once per draw

2026-07-29 03:26:57 -04:00

ProgramFactory.h

[Fix] (DirectVulkan): rewrite implicit-LOD fragment samples to explicit LOD 0 when every bound sampler is pinned to a single mip level - Adreno 650 (driver 512.502) reads outside a full-screen colour render target's allocation on its implicit-LOD sampling path and faults the GPU, which killed MC 26.2 on its own blit shader (texture(InSampler, texCoord)) between frames 344-421 on every run; this is the same driver defect the default-framebuffer blit shader already works around with textureLod, but an application's shader cannot be edited, so ForceExplicitLod0SamplePass converts OpImageSample*ImplicitLod to the explicit form at the SPIR-V level under a new CompileOptionBit that is only requested when the rewrite provably cannot move a texel (every sampler binding on a single-level view, no anisotropy, and either a LOD clamp that already pins lambda at 0 or min and mag filters that agree - an explicit LOD 0 always takes the magnification side of the min/mag decision); a single-level view now also clamps its sampler to mipmapMode NEAREST with maxLod min(maxLod, 0.25) rather than 0, since collapsing the clamp would make every fragment magnify and quietly retire the min filter; and the program's backend hash memo grows from one slot to four so a program resolved under two compile-flag sets in the same frame stops re-hashing every stage's SPIR-V once per draw

2026-07-29 03:26:57 -04:00

SwapchainObject.cpp

[Fix] (DirectVulkan): stop loading and carrying dead default-framebuffer content

2026-07-30 02:40:47 -04:00

SwapchainObject.h

[Fix] (DirectVulkan): stop loading and carrying dead default-framebuffer content

2026-07-30 02:40:47 -04:00

UniformManager.cpp

[Perf] (DirectVulkan): reuse unchanged global-UBO slices and skip identical descriptor binds

2026-07-30 08:18:21 -04:00

UniformManager.h

[Perf] (DirectVulkan): reuse unchanged global-UBO slices and skip identical descriptor binds

2026-07-30 08:18:21 -04:00

VertexInputStateBuilder.cpp

[Feat] (MG_Backend/DirectVulkan): vertex input state factory

2026-02-17 10:24:01 +08:00

VertexInputStateBuilder.h

[Feat] (MG_Backend/DirectVulkan): vertex input state factory

2026-02-17 10:24:01 +08:00

VertexInputStateFactory.cpp

[Perf] (DirectVulkan): drop per-draw weak-ptr locks, re-resolves and rebuilt masks from the sampled-texture and vertex paths

2026-07-30 08:01:07 -04:00

VertexInputStateFactory.h

[Perf] (DirectVulkan): drop per-draw weak-ptr locks, re-resolves and rebuilt masks from the sampled-texture and vertex paths

2026-07-30 08:01:07 -04:00

VkBufferManager.cpp

[Fix] (DirectVulkan): drain frame transients on present-less paths - readback waits, suspended presentation, blocking sync waits and flush completion polls now run Present's per-frame drains (deferred buffer/texture releases, transient arena rewind, descriptor cursors, retired command buffers, conversion caches) whenever every submission is provably complete, so offscreen/minimized workloads stay bounded; never blocks, frames-in-flight overlap untouched

2026-07-27 21:45:21 -04:00

VkBufferManager.h

[Fix] (DirectVulkan): drain frame transients on present-less paths - readback waits, suspended presentation, blocking sync waits and flush completion polls now run Present's per-frame drains (deferred buffer/texture releases, transient arena rewind, descriptor cursors, retired command buffers, conversion caches) whenever every submission is provably complete, so offscreen/minimized workloads stay bounded; never blocks, frames-in-flight overlap untouched

2026-07-27 21:45:21 -04:00

VkBufferObject.cpp

[Fix] (MG_Backend/DirectVulkan, ShaderTranspiler, MG_Test, TraceReplay): make iterationRP retrace pass on 64-lane Vulkan devices with subgroup-width emulation and format-aware readback

2026-07-20 02:36:23 -04:00

VkBufferObject.h

[Fix] (MG_Backend/DirectVulkan, ShaderTranspiler, MG_Test, TraceReplay): make iterationRP retrace pass on 64-lane Vulkan devices with subgroup-width emulation and format-aware readback

2026-07-20 02:36:23 -04:00

VkClearManager.cpp

[Perf] (DirectVulkan): skip pending-clear probes through a lock-free empty check

2026-07-30 07:02:03 -04:00

VkClearManager.h

[Perf] (DirectVulkan): snapshot-gated consecutive-draw fast path skips SetupDraw re-resolution

2026-07-30 09:40:54 -04:00

VkRenderPassManager.cpp

[Perf] (DirectVulkan): drop per-draw weak-ptr locks, re-resolves and rebuilt masks from the sampled-texture and vertex paths

2026-07-30 08:01:07 -04:00

VkRenderPassManager.h

[Perf] (DirectVulkan): snapshot-gated consecutive-draw fast path skips SetupDraw re-resolution

2026-07-30 09:40:54 -04:00

VkSamplerManager.cpp

[Fix] (DirectVulkan): rewrite implicit-LOD fragment samples to explicit LOD 0 when every bound sampler is pinned to a single mip level - Adreno 650 (driver 512.502) reads outside a full-screen colour render target's allocation on its implicit-LOD sampling path and faults the GPU, which killed MC 26.2 on its own blit shader (texture(InSampler, texCoord)) between frames 344-421 on every run; this is the same driver defect the default-framebuffer blit shader already works around with textureLod, but an application's shader cannot be edited, so ForceExplicitLod0SamplePass converts OpImageSample*ImplicitLod to the explicit form at the SPIR-V level under a new CompileOptionBit that is only requested when the rewrite provably cannot move a texel (every sampler binding on a single-level view, no anisotropy, and either a LOD clamp that already pins lambda at 0 or min and mag filters that agree - an explicit LOD 0 always takes the magnification side of the min/mag decision); a single-level view now also clamps its sampler to mipmapMode NEAREST with maxLod min(maxLod, 0.25) rather than 0, since collapsing the clamp would make every fragment magnify and quietly retire the min filter; and the program's backend hash memo grows from one slot to four so a program resolved under two compile-flag sets in the same frame stops re-hashing every stage's SPIR-V once per draw

2026-07-29 03:26:57 -04:00

VkSamplerManager.h

[Fix] (DirectVulkan): rewrite implicit-LOD fragment samples to explicit LOD 0 when every bound sampler is pinned to a single mip level - Adreno 650 (driver 512.502) reads outside a full-screen colour render target's allocation on its implicit-LOD sampling path and faults the GPU, which killed MC 26.2 on its own blit shader (texture(InSampler, texCoord)) between frames 344-421 on every run; this is the same driver defect the default-framebuffer blit shader already works around with textureLod, but an application's shader cannot be edited, so ForceExplicitLod0SamplePass converts OpImageSample*ImplicitLod to the explicit form at the SPIR-V level under a new CompileOptionBit that is only requested when the rewrite provably cannot move a texel (every sampler binding on a single-level view, no anisotropy, and either a LOD clamp that already pins lambda at 0 or min and mag filters that agree - an explicit LOD 0 always takes the magnification side of the min/mag decision); a single-level view now also clamps its sampler to mipmapMode NEAREST with maxLod min(maxLod, 0.25) rather than 0, since collapsing the clamp would make every fragment magnify and quietly retire the min filter; and the program's backend hash memo grows from one slot to four so a program resolved under two compile-flag sets in the same frame stops re-hashing every stage's SPIR-V once per draw

2026-07-29 03:26:57 -04:00

VkTextureManager.cpp

[Perf] (DirectVulkan): memoize sampled-texture resources across draws

2026-07-30 07:02:02 -04:00

VkTextureManager.h

[Perf] (DirectVulkan): snapshot-gated consecutive-draw fast path skips SetupDraw re-resolution

2026-07-30 09:40:54 -04:00

VkTextureSamplerManager.cpp

Merge branch 'dev' into Feat/Backend-Direct-Vulkan

2026-03-08 10:31:44 +08:00

VkTimerQueryManager.cpp

[Feat] (MG_Impl, MG_Backend): GL_ARB_timer_query on both backends

2026-07-10 01:08:38 +00:00

VkTimerQueryManager.h

[Feat] (MG_Impl, MG_Backend): GL_ARB_timer_query on both backends

2026-07-10 01:08:38 +00:00

VulkanRenderer.cpp

[Perf] (DirectVulkan): snapshot-gated consecutive-draw fast path skips SetupDraw re-resolution

2026-07-30 09:40:54 -04:00

VulkanRenderer.h

[Perf] (DirectVulkan): snapshot-gated consecutive-draw fast path skips SetupDraw re-resolution

2026-07-30 09:40:54 -04:00