BZLZHH 28c3cfc1d6 [Fix] (DirectVulkan, MG_State): make a shader-written storage buffer readable on Magma ...

Reading a buffer a compute shader wrote gave zeros: the frontend shadow that MapBuffer
resolves against is only maintained by uploads, and Magma had no path back. Every
KHR-GL40.texture_gather case ends by dispatching a compute shader into an SSBO and
comparing the mapped result, so 66 of 75 failed on it.

Magma needs no readback: EnsureGpuResidentStorage - the same host-visible coherent
adoption the transform feedback capture already uses - makes the shadow BE the memory
the shader writes, so binding a buffer as a shader storage buffer now adopts it. What
coherence does not give is ordering: the writes are visible once they have happened,
and the CPU was reading before the dispatch had retired. The readback op therefore
submits the recorded work and waits.

That exposed a mistake in the frontend flag this rides on: MarkGpuWritten skipped
GPU-resident buffers, reasoning there was no shadow to refresh. True, but the wait is
still needed - "reconcile with the GPU write" is not always "copy it back", and which
of the two it is belongs to the backend. The flag now only says a write is outstanding;
DirectGLES's readback still skips its persistent-mapped buffers when copying.

KHR-GL40.texture_gather on Magma: 66 failures -> 19 (the rest are rectangle textures,
mipmap completeness and tessellation, all still to do). Espryt stays at 75/75.

2026-08-04 11:55:17 -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

[Feat] (DirectVulkan): GPU transform feedback capture via VK_EXT_transform_feedback

2026-07-31 17:31:41 -04:00

ProgramFactory.h

[Feat] (DirectVulkan): GPU transform feedback capture via VK_EXT_transform_feedback

2026-07-31 17:31:41 -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

[Fix] (DirectVulkan, MG_State): make a shader-written storage buffer readable on Magma

2026-08-04 11:55:17 -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, MG_State): make a shader-written storage buffer readable on Magma

2026-08-04 11:55:17 -04:00

VkBufferManager.h

[Feat] (DirectVulkan): GPU transform feedback capture via VK_EXT_transform_feedback

2026-07-31 17:31:41 -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

[Fix] (DirectVulkan): multisample resolve blits; per-buffer blit skip; RGBA-widened renderbuffers

2026-08-01 02:46:16 -04:00

VkRenderPassManager.h

[Fix] (DirectVulkan): raw sRGB attachment writes while FRAMEBUFFER_SRGB is off

2026-08-01 01:15:57 -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

[Fix] (DirectVulkan): submit pending work before growing a texture's mip chain

2026-08-02 13:09:48 -04:00

VkTextureManager.h

[Fix] (DirectVulkan): submit pending work before growing a texture's mip chain

2026-08-02 13:09:48 -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

[Fix] (DirectVulkan, MG_State): make a shader-written storage buffer readable on Magma

2026-08-04 11:55:17 -04:00

VulkanRenderer.h

[Fix] (DirectVulkan, MG_State): make a shader-written storage buffer readable on Magma

2026-08-04 11:55:17 -04:00