Commit Graph
719 Commits
Author SHA1 Message Date
swung0x48 01116f7b41 [Fix] (DirectVulkan): treat an incomplete sampled texture as unbound in the collect path too, and stop dereferencing a declined texture sync 2026-08-27 12:52:18 -04:00
swung0x48 e1d5bdc4a5 [Fix] (Blend): decline an unsupported dual-source blend instead of throwing through the GL ABI 2026-08-27 10:48:21 -04:00
swung0x48 685d83e3ec [Fix] (Review): bound a texture readback by the image's mip count, refuse cross-format depth/stencil copies, and probe the renderbuffer's real VkFormat 2026-08-27 08:35:05 -04:00
swung0x48 009b140691 [Fix] (Review): clamp integer and floating-point border colours to the sampled format's representable range 2026-08-27 08:35:05 -04:00
swung0x48 8e072bc793 [Fix] (DirectVulkan): answer a texture-image query from the CPU shadow when the texture has no VkImage 2026-08-27 08:35:02 -04:00
swung0x48 88ee75be0e [Fix] (DirectVulkan): resolve renderbuffer VkFormats through the shared texture table and police copy size-compatibility 2026-08-27 08:35:01 -04:00
swung0x48 0dbb4ceba8 [Fix] (DirectVulkan): resolve arbitrary and integer border colours through VK_EXT_custom_border_color, clamped to the sampled format 2026-08-27 08:35:01 -04:00
swung0x48 f7e23d5d83 [Fix] (DirectVulkan): generate mipmaps for cube-map-array and 1D-array targets, and stop shrinking an array's layer count down the chain 2026-08-27 05:37:14 -04:00
swung0x48 de09646d5e [Fix] (DirectVulkan): one shared attachment layer count for the render pass and the clear key, and direct VkResult checks in the render-pass builder 2026-08-27 05:04:44 -04:00
swung0x48 df1bcdba09 [Fix] (DirectVulkan): legal view types for layered 3D/cube attachments, and a real failure channel for the render-pass builder 2026-08-27 04:19:06 -04:00
swung0x48 06744fde7f [Fix] (Getter): close the review findings - six-stage combined uniform blocks, bounded uniform-block bindings, honest vertex-stream count, per-format sample ceilings 2026-08-27 03:54:16 -04:00
swung0x48 f1354dc25e [Fix] (DirectVulkan): hash the sample-shading state into the pipeline memo key 2026-08-27 03:35:56 -04:00
swung0x48 acf86d1fb6 [Feat] (RenderState): implement glMinSampleShading and the GL_SAMPLE_SHADING enable on both backends 2026-08-27 03:35:55 -04:00
Swung0x48 31252cf0da [Fix] (Tessellation): bound the pass-through control-stage cache and stop baking "draw nothing" for levels GL clamps 2026-08-27 03:18:14 -04:00
Swung0x48 eb9e4fdac1 [Fix] (DirectVulkan): report the index the arbitrary-restart rewrite cannot represent instead of silently drawing a different vertex 2026-08-27 03:18:10 -04:00
Swung0x48 90b7a689c5 [Fix] (Backend): resolve primitive restart per draw - never for a non-indexed one, and never on an index the type cannot hold 2026-08-27 03:18:09 -04:00
Swung0x48 d5286e69b6 [Feat] (Tessellation): implement glPatchParameterfv and bake the default levels into both pass-through control stages 2026-08-27 02:11:19 -04:00
Swung0x48 0f2fcbc469 [Fix] (Backend): honour desktop GL_PRIMITIVE_RESTART with an arbitrary index instead of throwing through the C GL ABI 2026-08-27 02:11:18 -04:00
swung0x48 59f7059bf4 [Fix, Test] (DirectVulkan): keep the draw when a program declares an image-backed resource the application left unbound 2026-08-24 05:31:15 -04:00
swung0x48 4446c861be [Fix, Test] (DirectVulkan): keep the draw when a program declares a storage block the application left unbound 2026-08-24 04:57:05 -04:00
swung0x48 6a8bf4c03c [Fix, Test] (DirectVulkan): resolve a lowered atomic-counter block from the atomic-counter binding points 2026-08-22 21:45:13 -04:00
swung0x48 121b99f8c3 [Fix, Test] (DirectVulkan, GLImpl): size a 1D-array texture's mip chain and layered attachment by the axis that carries its layers 2026-08-22 12:52:10 -04:00
swung0x48 e8d79344d6 [Fix] (DirectVulkan): resolve a cube-map-array sampler to the cube-map-array texture target 2026-08-22 12:47:13 -04:00
swung0x48 473d9951b7 [Fix, Test] (DirectVulkan, GLState): apply a texture view level and layer window at every subresource boundary 2026-08-22 11:17:09 -04:00
swung0x48 6162603072 [Feature, Fix, Test] (GLState, GLImpl, DirectVulkan, DirectGLES): implement glTextureView over shared texture storage 2026-08-22 10:41:52 -04:00
swung0x48 d4247db6c3 [Fix, Test] (ShaderTranspiler, GLImpl, ProgramState, DirectVulkan): keep fp64 where the backend consumes it natively 2026-08-22 00:57:06 -04:00
swung0x48 fdbe0b3117 [Fix, Test] (DirectGLES, DirectVulkan, GLImpl, GLState): ask the last link, not the live attach list, what stages a program has 2026-08-21 08:01:48 -04:00
swung0x48 36b9d26b9d [Fix] (DirectVulkan): match the 2_10_10_10 storage image view format to the texture's own 2026-08-21 02:54:47 -04:00
swung0x48 5fa849674e [Fix] (DirectVulkan): gate the GL_DOUBLE vertex narrowing on the same fp64 flag the shader demotion uses 2026-08-21 00:28:03 -04:00
swung0x48 ef66aea73b [Fix, Test] (GLImpl, DirectGLES, DirectVulkan): address a 1D array's copy-image layers on Z, not Y 2026-08-21 00:15:01 -04:00
swung0x48 7aa91e8024 [Fix, Test] (DirectGLES, DirectVulkan): narrow GL_DOUBLE vertex arrays to float32 instead of dropping them 2026-08-20 23:54:49 -04:00
swung0x48 3477d87b50 [Fix] (DirectVulkan): back a 1D array with its layers in arrayLayers, not in the image height 2026-08-20 14:14:21 -04:00
swung0x48 c2a081fa75 [Fix] (GLState, DirectVulkan): bust the texture-sync skip when a re-spec moved only the shape 2026-08-20 14:02:05 -04:00
swung0x48 02c9b8a32d [Fix, Test] (GLImpl, DirectVulkan, MG_IntegrationTest): record glVertexAttribLFormat's state and drop the array at draw 2026-08-20 13:44:44 -04:00
swung0x48 898c39f1de [Fix] (DirectGLES, DirectVulkan): decline a null copy-image endpoint and settle a renderbuffer on its attachment layout 2026-08-20 10:23:03 -04:00
swung0x48 52718ecf84 [Fix, Test] (GLImpl, DirectGLES, DirectVulkan): accept GL_RENDERBUFFER endpoints in glCopyImageSubData 2026-08-20 10:11:38 -04:00
swung0x48 f297af7d2b [Fix] (GLImpl, DirectGLES, DirectVulkan): floor every advertised sample cap and clamp the realised count in the backends 2026-08-20 04:10:45 -04:00
swung0x48 c09045fe59 [Fix, Test] (DirectVulkan, ShaderTranspiler, TraceReplay): repair iterationRP's missing reduction barrier
Program 203 reuses prefixSumCache for a second subgroup reduction before every workgroup invocation has consumed the first result. Add a fingerprint-gated SPIR-V pass that inserts the missing Workgroup acquire-release barrier while preserving native subgroup operations.

Keep the repair opt-in behind MOBILEGL_ITERATIONRP_FIX_BARRIER, cover insertion, pass-through, and idempotence, and enable it together with the existing iterationRP subgroup repairs for the matching Linux and Android CI retraces.
2026-08-19 23:42:17 -04:00
swung0x48 6aed3b08f3 [Fix] (DirectVulkan, GLImpl, MG_State, ShaderTranspiler): second audit round over the remaining memo sites
Six more verified defects from the residual memo/cache mechanisms:

- Program resource cache (DirectVulkan reflection): glShaderStorageBlockBinding
  deliberately does not bump the backend state version, and the SSO pipeline
  composite is unnamed so the by-name in-place patch can never reach its slot -
  the composite kept serving pre-rebind SSBO bindings. The cache now keys on the
  program's block-binding version; a binding-only change re-applies the overrides
  by name instead of re-running spirv-reflect. SetShaderStorageBlockBinding also
  gains the equality bail-out its uniform-block sibling has, so the composite
  mirror's replay stops churning the version every draw.
- LinkProgram's allowVSOnlyPrograms function-static latch never set its own
  initialized flag (dead memo, re-read every call) - and completing it would have
  frozen a per-backend capability across re-initialization. Replaced with a fresh
  per-link read from the null-checked active backend.
- Query object registry: drained at full library teardown (DestroyAllQueryObjects,
  mirroring DestroyAllSyncObjects) - undeleted queries and their backend wrappers
  leaked across Destroy/Initialize cycles, stale ids stayed IsQuery == GL_TRUE in
  the re-initialized library, and a later delete could hand the old backend's
  wrapper to a different backend's DeleteBackendQuery.
- Converted vertex streams and the host-side EBO max-index scan now SyncGpuWrites
  before reading the coherent mapping: XFB/SSBO/image writes are merely recorded
  at that point, so the conversion read pre-write bytes (the restart-index
  rewrite already synced; these two host reads did not).
- Zero-stride converted bindings: both converters rejected stride 0, making the
  factory's documented single-element conversion unreachable and silently
  dropping every draw using such a binding; the stride is substituted with the
  element size for the one-element case.
- DemoteFloat64Pass block relayout: measurement queued into the module eagerly,
  so a mid-struct failure left a half-relaid-out block (compacted offsets before
  the failing member, 64-bit offsets after) while claiming the block was left
  alone. Decoration writes are now collected and committed only when the whole
  block measures successfully.
2026-08-19 16:41:28 -04:00
swung0x48 281467a345 [Fix] (DirectGLES, DirectVulkan, MG_State): close stale-cache, A-B-A and state-leak holes across the memo layers
Audit of every memoization implementation; sixteen verified defects fixed:

DirectGLES backend:
- Broadcast draw-buffer memo: cleared at MakeCurrent/DestroyEGLContext like its
  sibling shadows; its identity+version key is only monotonic within one GLContext,
  so a library teardown + re-init could false-hit on a recycled FBO address.
- Backend texture id re-mint (RecreateBackendTexture) now bumps an attachment
  generation that the SyncCurrentFBO gate and every FBO twin compare, so driver
  FBOs re-attach instead of keeping the deleted texture name; the attachment walk
  re-enters until the generation is quiescent (a walk itself can re-mint).
- Buffer id re-mint (persistent-map adoption, immutable-store retire) now bumps a
  generation the VAO twin sync compares, forcing a full re-emit of the baked
  glVertexAttribPointer / element-array bindings that frontend versions cannot see.
- VAO element-array sync memo: bound-object identity joins the wrapping Uint16
  slot version (same pairing the ResolvedDrawBuffers IBO memo already uses).

DirectVulkan backend:
- EBO slice memo gains the mapped-buffer guard its vertex-binding sibling has: a
  shadow-backed persistent map mutates with no epoch bump, so a hit must decline.
- VkClearManager::MergeClearPayload keeps colorEncoding/colorInt/colorUint with
  the color, so deferred glClearBufferiv/uiv no longer degrade to all-zero float.
- GetOrCreateComputePipeline no longer memoizes a failed creation (same contract
  as PipelineFactory): a transient driver failure was permanently disabling every
  dispatch of that program.
- Explicit-LOD-0 verdict memo keys on the sampling-resolution generation; sampler
  filter/aniso/LOD setters bump only that counter, so the old key served a stale
  verdict (wrong SPIR-V variant) after glTexParameter/glSamplerParameter changes.
- SetupDraw fast path declines instead of re-arming on a moved sampling-resolution
  generation (the snapshot bakes the LOD verdict into its pipeline), and
  recomputes the XfbCapture bit so the first draw after glBeginTransformFeedback
  cannot bind the undecorated variant and silently capture nothing.
- VertexInputStateFactory eviction epoch is drawn from a process-wide source: VAO
  state-pointer memos outlive the factory across renderer recreation, and a fresh
  factory restarting at epoch 1 would dereference a dead factory's entry.
- Cached render passes re-read the live renderbuffer clear payload at begin (the
  clear VALUE is not in the pass hash; the entry's inline snapshot replayed the
  creation-time color and dropped the newly queued one).
- FramebufferObject gains a never-reused lifetime id, keyed into the render-pass
  fast-path memo and the SetupDraw snapshot beside the raw pointer + Uint16
  version pair, which address reuse plus fresh version counts could equal.
- SyncTextureResource's preserved-content image goes through the deferred-release
  ring on both failure paths instead of a synchronous destructor under the GPU.

MG_State frontend:
- Layer-1 compile memo is env-disciplined like layers 2/3: a node computed against
  a dead CompileEnv (e.g. pre-capability fallback limits) no longer answers
  glCompileShader forever once the environment's content changes.
- Pipeline composite cache rebuilds from each stage program's last-link shader
  snapshot (new LinkedShaderRef list + pinned link inputs) instead of the live
  attach list and current compile nodes: post-link glAttachShader/glCompileShader
  must not leak into the composite while the (lifetimeId, linkVersion) signature
  still hits - GL's "as last linked" rule.
2026-08-19 16:41:26 -04:00
swung0x48 c7e36986e7 [Fix] (ShaderTranspiler, DirectVulkan, MG_IntegrationTest): patch both of iterationRP's under-declared subgroup scratch arrays
The previous commit's fingerprint was pinned to one array's incidental
dimensions - workgroup exactly 32x16x1, element exactly vec2, length
exactly 32 - which is the auto-exposure reduction and nothing else. The
pack ships the same idiom twice:

  - auto-exposure:  32x16 (512 invocations), shared vec2 prefixSumCache[32]
  - RTW warp:       1024 invocations,        shared float prefixSumCache[64]

so the warp kept writing 128 subgroups into 64 entries on an 8-lane
device and the retrace stayed bit-identically wrong (ssim 0.027902).

Key the fingerprint on the pack's idiom instead of one array's shape: a
workgroup array of 32-bit floats indexed by gl_SubgroupID, fed by a
subgroup scan, whose declared length is below ceil(invocations / native
width). Three properties keep that a targeted repair rather than a
general array resizer:

  - the index must BE gl_SubgroupID (through OpCopyObject, a signedness
    OpBitcast, or a spill whose every store is that id), so an index
    masked or clamped into range is left alone;
  - the >= 16-lane early-out is retained, so every module on the devices
    the pack was written for passes through byte-identical;
  - growth is certified against maxComputeSharedMemorySize using a
    natural-alignment layout model, and declined outright when a
    declaration cannot be sized, so a patched module can never fail
    pipeline creation where the original would not have.

Verified against the shaders the CI trace actually contains: of the 14
compute modules in the fixture exactly these two change, the other
twelve are byte-identical, and all fourteen pass spirv-val. The
integration scenario grows a second case for the 1024-invocation shape;
both abort with heap corruption when the patch is disabled.
2026-08-19 16:36:06 -04:00
swung0x48 d8576a2ed3 [Fix] (DirectVulkan, ShaderTranspiler, MG_IntegrationTest, SelfTest, TraceReplay): use native subgroups and patch iterationRP's under-declared scratch
iterationRP's Program 203 declares shared vec2 prefixSumCache[32] for a
512-invocation workgroup indexed by gl_SubgroupID; any device narrower
than 16 lanes partitions into more than 32 subgroups and the pack writes
shared memory out of bounds (heap corruption on lavapipe's CPU
rasterizer, ssim 0.028 on the CI retrace). Fix it where the fault lies -
in the fixture - and keep the GL contract sound everywhere else:

- FixIterationRPSubgroupScratchPass: fingerprint-gated SPIR-V pass that
  grows exactly that array to ceil(invocations/width) entries on sub-16-lane devices; every other module passes through byte-identical.
- DeriveNumSubgroupsPass stays default-on for the Adreno topology bug
  and is made spec-sound: pipelines request REQUIRE_FULL_SUBGROUPS
  whenever the workgroup shape makes the flag legal (computeFullSubgroups
  enabled, local_size_x a multiple of the native width, subgroup count
  within maxComputeWorkgroupSubgroups).
- EmulateSubgroupsPass: 32-lane virtual-subgroup lowering kept in-tree
  as a last resort, enabled only by MOBILEGL_MAGMA_EMULATE_SUBGROUP=1 on
  devices with no native subgroup support; fails closed on extended
  subgroup instructions and on modules whose added scratch would exceed
  maxComputeSharedMemorySize.
- IterationRPFirstReductionScenario skips gracefully outside the pack's
  16..256-lane source domain; the new IterationRPScratchFixScenario runs
  the fixture-shaped reduction on any width and asserts the exact
  width-independent total. DriverPost keeps reporting FAIL on
  out-of-domain devices.
- Program203 -> IterationRP rename throughout; the per-trace
  num_subgroups_quirk plumbing is removed from the trace replayer, JNI
  chain, and CI workflows.
2026-08-19 09:48:11 -04:00
swung0x48 2b6c2b561c [Fix, Test] (DirectVulkan, ShaderTranspiler, TraceReplay): derive NumSubgroups behind opt-in quirk 2026-08-18 22:31:57 -04:00
swung0x48andClaude Fable 5 12c94111b5 [Fix, Test] (DirectVulkan, SelfTest, MG_IntegrationTest, TraceReplay): snapshot sampler/image feedback and add Program 203 diagnostics
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-08-18 03:04:09 -04:00
swung0x48 14d3901d30 [Chore, Test] (MG_State, MG_Util, DirectGLES, DirectVulkan): make SPIR-V validation task-local 2026-08-15 22:58:28 -04:00
swung0x48 d4766513e4 [Fix, Test] (DirectVulkan): replay Photon descriptor pressure correctly 2026-08-15 22:35:39 -04:00
swung0x48 8acd885594 [Fix, Test] (DirectVulkan): preserve viewport-index program metadata 2026-08-15 09:40:13 -04:00
swung0x48 10ff5e2b18 [Fix, Test] (DirectGLES, DirectVulkan): advertise indirect draw capabilities accurately
- advertise GL_ARB_draw_indirect when supported
- gate GL_ARB_base_instance on complete non-zero firstInstance semantics
- synchronize Driver POST reporting
- add capability and extension-advertisement regression tests
2026-08-15 06:30:44 -04:00
swung0x48 0deff52a1b [Fix, Test] (DirectVulkan, trace-replay): replay quarter-turn surfaces correctly 2026-08-13 06:45:28 -04:00
swung0x48 42ad62b54c [Fix, Test] (MG_Util, MG_IntegrationTest): advertise the GL 4.3 VIEWPORT_BOUNDS_RANGE floor on a GLES driver that has no such query, instead of a range admitting no origin 2026-08-13 04:44:23 -04:00