[Fix] (MG_Backend): scale a depth blit, keep going after one declines, and mip a 1D texture

Three DirectVulkan gaps found together.

glBlitFramebuffer's depth/stencil path refused any blit whose source and
destination extents differ, because vkCmdCopyImage cannot resize. vkCmdBlitImage
can, and VK_FILTER_NEAREST is the only filter Vulkan allows for depth/stencil
anyway - which is what the GL front end already requires. A same-size pair keeps
the cheaper copy.

Worse, that refusal and four others were `return`, not `continue`, so a
depth/stencil aspect this backend could not handle abandoned the whole function -
including the colour blit that only starts after the aspect loop. The CTS's
scaling blits therefore lost their colour as well, which is why
direct_state_access.framebuffers_blit failed all three of its checks rather than
one.

VulkanRenderer::GenerateMipmap declined GL_TEXTURE_1D. It needed nothing else:
the blit loop derives every offset from the storage extent, and a 1D texture's is
{width, 1, 1}, which is exactly the y and z offsets a 1D image requires.

Also: IsTimerQueryResultReady now asks the query pool before the frame serial.
The pool polls with VK_QUERY_RESULT_WITH_AVAILABILITY_BIT and is the authority;
the frame serial only advances at Present and neither completion notifier will
mark the current serial done, so a timestamp written and fence-waited inside one
GL frame could never be read back within it.

Takes framebuffers_blit and textures_generate_mipmaps from failing to passing on
DirectVulkan. queries_functional still fails there on a value.

.github/scripts/fetch-trace-fixture-lfs.sh

@@ -201,6 +201,11 @@ fetch_file_from_mirror() {
   return 1
 }
 
+# Files no mirror could serve, even after retrying every mirror. Only these fall
+# back to Git LFS, so a mirror that served the rest of the case still spares
+# GitHub the bandwidth for those files.
+mirror_failures=()
+
 fetch_from_mirror() {
   mkdir -p "${fixture_dir}"
   for file in "${files[@]}"; do
@@ -219,17 +224,19 @@ fetch_from_mirror() {
       echo "Mirror did not serve ${name}; trying the next mirror" >&2
     done
     if [ "${fetched}" -ne 1 ]; then
-      return 1
+      mirror_failures+=("${file}")
     fi
   done
+  [ "${#mirror_failures[@]}" -eq 0 ]
 }
 
 if fetch_from_mirror; then
   echo "Fetched trace fixture files for ${case_name} from mirror: ${include}"
 else
-  echo "All mirrors failed for ${case_name}; falling back to Git LFS: ${include}"
+  fallback_include="$(IFS=,; echo "${mirror_failures[*]}")"
+  echo "All mirrors failed for ${#mirror_failures[@]} of ${#files[@]} file(s) of ${case_name}; falling back to Git LFS: ${fallback_include}"
   git lfs install --local
-  git lfs pull --include="${include}" --exclude=""
+  git lfs pull --include="${fallback_include}" --exclude=""
 fi
 
 for file in "${files[@]}"; do

CMakeLists.txt

@@ -218,6 +218,7 @@ set(SOURCE_FILES
     MobileGL/MG_Impl/GLImpl/Framebuffer/Validators.cpp
     MobileGL/MG_Impl/GLImpl/Framebuffer/GL_Framebuffer.cpp
     MobileGL/MG_Impl/GLImpl/Program/GL_Program.cpp
+    MobileGL/MG_Impl/GLImpl/Program/GL_ProgramPipeline.cpp
     MobileGL/MG_Impl/GLImpl/Texture/GL_Texture.cpp
     MobileGL/MG_Impl/GLImpl/Texture/Validators.cpp
     MobileGL/MG_Impl/GLImpl/Texture/ProxyTexture.cpp

MobileGL/MG_Backend/BackendObject.h

@@ -145,6 +145,10 @@ namespace MobileGL {
                                             GLenum buffer, GLint drawbuffer, const GLfloat* value);
             void (*ClearNamedFramebufferfi)(const SharedPtr<MG_State::GLState::FramebufferObject>& framebuffer,
                                             GLenum buffer, GLint drawbuffer, GLfloat depth, GLint stencil);
+            void (*ClearNamedFramebufferiv)(const SharedPtr<MG_State::GLState::FramebufferObject>& framebuffer,
+                                            GLenum buffer, GLint drawbuffer, const GLint* value);
+            void (*ClearNamedFramebufferuiv)(const SharedPtr<MG_State::GLState::FramebufferObject>& framebuffer,
+                                             GLenum buffer, GLint drawbuffer, const GLuint* value);
             void (*BlitFramebuffer)(GLint srcX0, GLint srcY0, GLint srcX1, GLint srcY1, GLint dstX0, GLint dstY0,
                                     GLint dstX1, GLint dstY1, GLbitfield mask, GLenum filter);
             void (*BlitNamedFramebuffer)(const SharedPtr<MG_State::GLState::FramebufferObject>& readFramebuffer,
@@ -349,6 +353,12 @@ namespace MobileGL {
             // real ES drivers behind DirectGLES both do. Defaults to true so a backend
             // that never sets it keeps the permissive behaviour.
             Bool SupportsDistinctDepthStencilAttachments = true;
+            // Whether attaching a single layer of a 3D or array texture to a framebuffer actually
+            // renders to that layer. DirectGLES hands the layer straight to
+            // glFramebufferTextureLayer, so it does; DirectVulkan maps a GL layer onto a Vulkan
+            // array layer with no notion of a 3D depth slice, so it does not yet. Defaults to false
+            // so a backend that never sets it gets the conservative answer.
+            Bool SupportsPerLayerFramebufferAttachment = false;
             SizeT MaxShaderStorageBlockSize = 128 * 1024 * 1024;
             Uint32 SubgroupSize = 0;
             Uint32 SubgroupSupportedStages = 0;

MobileGL/MG_Backend/DirectGLES/BackendObject_DirectGLES.cpp

@@ -32,8 +32,7 @@ namespace MobileGL::MG_Backend::DirectGLES {
 
         void ClearGLErrors(const MG_External::GLESFunctionsTable& gl) {
             if (!gl.glGetError) return;
-            while (gl.glGetError() != GL_NO_ERROR) {
+            while (gl.glGetError() != GL_NO_ERROR) {}
-            }
         }
 
         Bool CheckNoGLError(const MG_External::GLESFunctionsTable& gl) {
@@ -77,8 +76,7 @@ namespace MobileGL::MG_Backend::DirectGLES {
         }
 
         Bool IsGLESProbeMultisampleTarget(TextureTarget target) {
-            return target == TextureTarget::Texture2DMultisample ||
+            return target == TextureTarget::Texture2DMultisample || target == TextureTarget::Texture2DMultisampleArray;
-                   target == TextureTarget::Texture2DMultisampleArray;
         }
 
         GLenum GetFramebufferAttachment(TextureInternalFormat format) {
@@ -115,8 +113,8 @@ namespace MobileGL::MG_Backend::DirectGLES {
             GLenum normalizedInternalFormat = glFormat;
             GLenum imageFormat = GL_RGBA;
             GLenum imageType = GL_UNSIGNED_BYTE;
-            MG_Util::TextureFormatProcessor::NormalizePixelFormat(
+            MG_Util::TextureFormatProcessor::NormalizePixelFormat(glFormat, PixelFormatNormalizeOptionBit::None,
-                glFormat, PixelFormatNormalizeOptionBit::None, &normalizedInternalFormat, &imageFormat, &imageType);
+                                                                  &normalizedInternalFormat, &imageFormat, &imageType);
             return imageFormat != GL_RED_INTEGER && imageFormat != GL_RG_INTEGER && imageFormat != GL_RGB_INTEGER &&
                    imageFormat != GL_RGBA_INTEGER && !MG_Util::IsDepthFormatInternalFormat(format) &&
                    !MG_Util::IsStencilFormatInternalFormat(format);
@@ -154,9 +152,9 @@ namespace MobileGL::MG_Backend::DirectGLES {
         GLESProbeFormatInfo BuildNativeProbeFormatInfo(GLenum requestedInternalFormat) {
             GLESProbeFormatInfo info;
             info.InternalFormat = requestedInternalFormat;
-            MG_Util::TextureFormatProcessor::NormalizePixelFormat(
+            MG_Util::TextureFormatProcessor::NormalizePixelFormat(requestedInternalFormat,
-                requestedInternalFormat, PixelFormatNormalizeOptionBit::None, nullptr, &info.ImageFormat,
+                                                                  PixelFormatNormalizeOptionBit::None, nullptr,
-                &info.ImageType);
+                                                                  &info.ImageFormat, &info.ImageType);
             return info;
         }
 
@@ -233,20 +231,16 @@ namespace MobileGL::MG_Backend::DirectGLES {
             return MG_Util::ConvertGLEnumToString(internalFormat);
         }
 
-        void LogGLESFormatCaveat(TextureInternalFormat logicalFormat,
+        void LogGLESFormatCaveat(TextureInternalFormat logicalFormat, SizeT targetIndex,
-                                 SizeT targetIndex,
                                  const GLESProbeFormatInfo& fallbackInfo) {
             MGLOG_D("Caveat: %s %s not fully supported. Reason: %s. Fallback: %s",
                     GetFormatCapabilityTargetName(targetIndex).c_str(),
-                    MG_Util::ConvertTextureInternalFormatToString(logicalFormat).c_str(),
+                    MG_Util::ConvertTextureInternalFormatToString(logicalFormat).c_str(), fallbackInfo.Reason.c_str(),
-                    fallbackInfo.Reason.c_str(),
                     ConvertFallbackInternalFormatToString(fallbackInfo.InternalFormat).c_str());
         }
 
-        Bool BuildFallbackProbeFormatInfo(GLenum requestedInternalFormat,
+        Bool BuildFallbackProbeFormatInfo(GLenum requestedInternalFormat, Flags<PixelFormatNormalizeOptionBit> options,
-                                          Flags<PixelFormatNormalizeOptionBit> options,
+                                          Bool forced, GLESProbeFormatInfo& outInfo) {
-                                          Bool forced,
-                                          GLESProbeFormatInfo& outInfo) {
             const Flags<PixelFormatNormalizeOptionBit> applicableOptions =
                 MG_Util::TextureFormatProcessor::GetApplicablePixelFormatNormalizeOptions(requestedInternalFormat,
                                                                                           options);
@@ -261,8 +255,7 @@ namespace MobileGL::MG_Backend::DirectGLES {
             return outInfo.InternalFormat != GL_UNKNOWN_MGL;
         }
 
-        FormatCapabilityFlags BuildTextureCapsFromProbe(TextureInternalFormat logicalFormat,
+        FormatCapabilityFlags BuildTextureCapsFromProbe(TextureInternalFormat logicalFormat, TextureTarget target,
-                                                        TextureTarget target,
                                                         Bool renderable) {
             FormatCapabilityFlags caps = GetTextureFeatureCaps(logicalFormat, target);
             if (renderable) {
@@ -276,16 +269,12 @@ namespace MobileGL::MG_Backend::DirectGLES {
             return caps;
         }
 
-        void AddFullFormatCaps(FormatCapabilityCache& cache,
+        void AddFullFormatCaps(FormatCapabilityCache& cache, SizeT targetIndex, SizeT formatIndex,
-                               SizeT targetIndex,
-                               SizeT formatIndex,
                                FormatCapabilityFlags caps) {
             cache.FullCaps[targetIndex][formatIndex] |= caps;
         }
 
-        Bool AddCaveatFormatCaps(FormatCapabilityCache& cache,
+        Bool AddCaveatFormatCaps(FormatCapabilityCache& cache, SizeT targetIndex, SizeT formatIndex,
-                                 SizeT targetIndex,
-                                 SizeT formatIndex,
                                  FormatCapabilityFlags caps) {
             Bool added = false;
             for (FormatCapability capability : kReportedFormatCapabilities) {
@@ -299,8 +288,7 @@ namespace MobileGL::MG_Backend::DirectGLES {
         }
 
         Int GetGLESFormatMaxSamples(const MG_External::GLESCapabilities& capabilities,
-                                    TextureInternalFormat logicalFormat,
+                                    TextureInternalFormat logicalFormat, GLenum imageFormat) {
-                                    GLenum imageFormat) {
             const Bool isDepth = MG_Util::IsDepthFormatInternalFormat(logicalFormat);
             const Bool isStencil = MG_Util::IsStencilFormatInternalFormat(logicalFormat);
             const Bool isInteger = imageFormat == GL_RED_INTEGER || imageFormat == GL_RG_INTEGER ||
@@ -314,10 +302,8 @@ namespace MobileGL::MG_Backend::DirectGLES {
             return capabilities.MaxColorTextureSamples;
         }
 
-        Bool ProbeFramebufferCompletenessForTexture(const MG_External::GLESFunctionsTable& gl,
+        Bool ProbeFramebufferCompletenessForTexture(const MG_External::GLESFunctionsTable& gl, TextureTarget target,
-                                                   TextureTarget target,
+                                                    GLuint texture, TextureInternalFormat format) {
-                                                   GLuint texture,
-                                                   TextureInternalFormat format) {
             GLuint framebuffer = 0;
             GLint prevFramebuffer = 0;
             if (!gl.glGenFramebuffers || !gl.glBindFramebuffer || !gl.glCheckFramebufferStatus ||
@@ -399,9 +385,8 @@ namespace MobileGL::MG_Backend::DirectGLES {
             return supported;
         }
 
-        Bool ProbeFramebufferCompletenessForRenderbuffer(const MG_External::GLESFunctionsTable& gl,
+        Bool ProbeFramebufferCompletenessForRenderbuffer(const MG_External::GLESFunctionsTable& gl, GLuint renderbuffer,
-                                                        GLuint renderbuffer,
+                                                         TextureInternalFormat format) {
-                                                        TextureInternalFormat format) {
             GLuint framebuffer = 0;
             GLint prevFramebuffer = 0;
             if (!gl.glGenFramebuffers || !gl.glBindFramebuffer || !gl.glFramebufferRenderbuffer ||
@@ -468,16 +453,16 @@ namespace MobileGL::MG_Backend::DirectGLES {
                     }
                     break;
                 case TextureTarget::Texture3D:
-                    gl.glTexImage3D(glTarget, 0, static_cast<GLint>(internalFormat), 2, 2, 2, 0, imageFormat,
+                    gl.glTexImage3D(glTarget, 0, static_cast<GLint>(internalFormat), 2, 2, 2, 0, imageFormat, imageType,
-                                    imageType, nullptr);
+                                    nullptr);
                     break;
                 case TextureTarget::Texture2DArray:
-                    gl.glTexImage3D(glTarget, 0, static_cast<GLint>(internalFormat), 2, 2, 1, 0, imageFormat,
+                    gl.glTexImage3D(glTarget, 0, static_cast<GLint>(internalFormat), 2, 2, 1, 0, imageFormat, imageType,
-                                    imageType, nullptr);
+                                    nullptr);
                     break;
                 case TextureTarget::TextureCubeMapArray:
-                    gl.glTexImage3D(glTarget, 0, static_cast<GLint>(internalFormat), 2, 2, 6, 0, imageFormat,
+                    gl.glTexImage3D(glTarget, 0, static_cast<GLint>(internalFormat), 2, 2, 6, 0, imageFormat, imageType,
-                                    imageType, nullptr);
+                                    nullptr);
                     break;
                 default:
                     break;
@@ -498,11 +483,8 @@ namespace MobileGL::MG_Backend::DirectGLES {
             return created;
         }
 
-        Bool ProbeRenderbuffer(const MG_External::GLESFunctionsTable& gl,
+        Bool ProbeRenderbuffer(const MG_External::GLESFunctionsTable& gl, GLenum internalFormat,
-                               GLenum internalFormat,
+                               TextureInternalFormat logicalFormat, Bool multisample, Int samples) {
-                               TextureInternalFormat logicalFormat,
-                               Bool multisample,
-                               Int samples) {
             if (!gl.glGenRenderbuffers || !gl.glBindRenderbuffer || !gl.glDeleteRenderbuffers) {
                 return false;
             }
@@ -524,17 +506,16 @@ namespace MobileGL::MG_Backend::DirectGLES {
                 gl.glRenderbufferStorage(GL_RENDERBUFFER, internalFormat, 1, 1);
             }
             const Bool created = CheckNoGLError(gl);
-            const Bool complete = created && ProbeFramebufferCompletenessForRenderbuffer(gl, renderbuffer, logicalFormat);
+            const Bool complete =
+                created && ProbeFramebufferCompletenessForRenderbuffer(gl, renderbuffer, logicalFormat);
             gl.glBindRenderbuffer(GL_RENDERBUFFER, static_cast<GLuint>(prevRenderbuffer));
             gl.glDeleteRenderbuffers(1, &renderbuffer);
             ClearGLErrors(gl);
             return complete;
         }
 
-        Vector<Int> ProbeRenderbufferSampleCounts(const MG_External::GLESFunctionsTable& gl,
+        Vector<Int> ProbeRenderbufferSampleCounts(const MG_External::GLESFunctionsTable& gl, GLenum internalFormat,
-                                                  GLenum internalFormat,
+                                                  TextureInternalFormat logicalFormat, Int maxSamples) {
-                                                  TextureInternalFormat logicalFormat,
-                                                  Int maxSamples) {
             Vector<Int> sampleCounts;
             for (Int samples = std::max(maxSamples, 1); samples > 1; samples >>= 1) {
                 if (ProbeRenderbuffer(gl, internalFormat, logicalFormat, true, samples)) {
@@ -589,8 +570,8 @@ namespace MobileGL::MG_Backend::DirectGLES {
                         hasForcedFallback = BuildFallbackProbeFormatInfo(
                             requestedInternalFormat, forcedOptions | targetOptions, true, fallbackInfo);
                         if (!hasForcedFallback) {
-                            BuildFallbackProbeFormatInfo(requestedInternalFormat, driverOptions | targetOptions,
+                            BuildFallbackProbeFormatInfo(requestedInternalFormat, driverOptions | targetOptions, false,
-                                                         false, fallbackInfo);
+                                                         fallbackInfo);
                         }
                     }
 
@@ -623,9 +604,9 @@ namespace MobileGL::MG_Backend::DirectGLES {
                             ProbeTexture(gl, probeTarget, fallbackInfo.InternalFormat, fallbackInfo.ImageFormat,
                                          fallbackInfo.ImageType, logicalFormat, &fallbackRenderable);
                         if (fallbackCreated) {
-                            if (AddCaveatFormatCaps(cache, targetIndex, formatIndex,
+                            if (AddCaveatFormatCaps(
-                                                    BuildTextureCapsFromProbe(logicalFormat, target,
+                                    cache, targetIndex, formatIndex,
-                                                                              fallbackRenderable))) {
+                                    BuildTextureCapsFromProbe(logicalFormat, target, fallbackRenderable))) {
                                 LogGLESFormatCaveat(logicalFormat, targetIndex, fallbackInfo);
                             }
                             if (IsGLESProbeMultisampleTarget(target)) {
@@ -676,7 +657,7 @@ namespace MobileGL::MG_Backend::DirectGLES {
                 .ExtraVendor = Nullopt,              // Extra vendor
                 .RendererGLInfo =
                     {
-                        .TargetGLVersion = {3, 3, 0},   // GL target version
+                        .TargetGLVersion = {4, 0, 0},   // GL target version
                         .TargetGLSLVersion = {4, 6, 0}, // Target Shading Language Version
                         // Baseline advertisement (no timer queries / anisotropy yet); reconciled
                         // once the ES capabilities exist, see UpdateAdvertisedCapabilityExtensions.
@@ -707,8 +688,7 @@ namespace MobileGL::MG_Backend::DirectGLES {
     } // namespace
 
     void PopulateFormatCapabilities(const MG_External::GLESFunctionsTable& gl,
-                                    const MG_External::GLESCapabilities& capabilities,
+                                    const MG_External::GLESCapabilities& capabilities, FormatCapabilityCache& cache) {
-                                    FormatCapabilityCache& cache) {
         PopulateFormatCapabilitiesImpl(gl, capabilities, cache);
     }
 
@@ -772,10 +752,8 @@ namespace MobileGL::MG_Backend::DirectGLES {
             return false;
         }
 
-        if ((handle.Backend != WindowBackend::Android &&
+        if ((handle.Backend != WindowBackend::Android && handle.Backend != WindowBackend::X11 &&
-             handle.Backend != WindowBackend::X11 &&
+             handle.Backend != WindowBackend::MetalLayer && handle.Backend != WindowBackend::Win32) ||
-             handle.Backend != WindowBackend::MetalLayer &&
-             handle.Backend != WindowBackend::Win32) ||
             !handle.Handle) {
             MGLOG_E("DirectGLES backend only supports Android, X11, CAMetalLayer, and Win32 native windows");
             return false;
@@ -894,27 +872,25 @@ namespace MobileGL::MG_Backend::DirectGLES {
     }
 
     Vector<GLExtension> BuildAdvertisedExtensions(Bool timerQueriesSupported, Bool anisotropicFilteringSupported) {
-        Vector<GLExtension> extensions = {V_OpenGL30, V_OpenGL31, V_OpenGL32,
+        Vector<GLExtension> extensions = {
-                                          V_OpenGL33, E_GL_ARB_draw_buffers_blend, E_GL_ARB_compute_shader,
+            V_OpenGL30, V_OpenGL31, V_OpenGL32, V_OpenGL33, V_OpenGL40, E_GL_ARB_draw_buffers_blend,
-                                          E_GL_ARB_shader_storage_buffer_object, E_GL_ARB_shader_image_load_store,
+            E_GL_ARB_compute_shader, E_GL_ARB_shader_storage_buffer_object, E_GL_ARB_shader_image_load_store,
-                                          E_GL_ARB_program_interface_query, E_GL_ARB_framebuffer_object,
+            E_GL_ARB_program_interface_query, E_GL_ARB_framebuffer_object, E_GL_EXT_framebuffer_object,
-                                          E_GL_EXT_framebuffer_object, E_GL_ARB_depth_texture, E_GL_ARB_buffer_storage,
+            E_GL_ARB_depth_texture, E_GL_ARB_buffer_storage, E_GL_ARB_texture_storage,
-                                          E_GL_ARB_texture_storage, E_GL_ARB_texture_storage_multisample,
+            E_GL_ARB_texture_storage_multisample, E_GL_ARB_clear_texture, E_GL_ARB_direct_state_access,
-                                          E_GL_ARB_clear_texture, E_GL_ARB_direct_state_access,
+            E_GL_ARB_multi_draw_indirect, E_GL_ARB_indirect_parameters, E_GL_ARB_shader_draw_parameters,
-                                          E_GL_ARB_multi_draw_indirect, E_GL_ARB_indirect_parameters,
+            E_GL_ARB_gpu_shader5, E_GL_ARB_multi_bind, E_GL_ARB_shading_language_420pack,
-                                          E_GL_ARB_shader_draw_parameters, E_GL_ARB_gpu_shader5, E_GL_ARB_multi_bind,
+            E_GL_ARB_vertex_attrib_binding,
-                                          E_GL_ARB_shading_language_420pack, E_GL_ARB_vertex_attrib_binding,
+            // Both are core from GL 3.2/3.3 on and implemented here for
-                                          // Both are core from GL 3.2/3.3 on and implemented here for
+            // every advertised version, but an app targeting 3.0/3.1
-                                          // every advertised version, but an app targeting 3.0/3.1
+            // only reaches them through the extension string - the CTS
-                                          // only reaches them through the extension string - the CTS
+            // picks a whole different shader for draw_buffers without
-                                          // picks a whole different shader for draw_buffers without
+            // explicit_attrib_location. DirectVulkan advertises both.
-                                          // explicit_attrib_location. DirectVulkan advertises both.
+            E_GL_ARB_explicit_attrib_location, E_GL_ARB_texture_multisample, E_GL_ARB_shader_image_size,
-                                          E_GL_ARB_explicit_attrib_location, E_GL_ARB_texture_multisample,
+            // Advertised with GL_NUM_PROGRAM_BINARY_FORMATS = 0, which the
-                                          E_GL_ARB_shader_image_size,
+            // extension explicitly permits. It is also the only thing that
-                                          // Advertised with GL_NUM_PROGRAM_BINARY_FORMATS = 0, which the
+            // exposes glProgramParameteri before GL 4.1.
-                                          // extension explicitly permits. It is also the only thing that
+            E_GL_ARB_get_program_binary};
-                                          // exposes glProgramParameteri before GL 4.1.
-                                          E_GL_ARB_get_program_binary};
         // Only advertised when the device driver actually has usable timer queries
         // (GL_EXT_disjoint_timer_query plus its entry points) and the
         // MOBILEGL_DISABLE_TIMERQUERY escape hatch is off.
@@ -988,6 +964,8 @@ namespace MobileGL::MG_Backend::DirectGLES {
             funcsTable.GL.ClearBufferuiv = ClearBufferuiv;
             funcsTable.GL.ClearBufferiv = ClearBufferiv;
             funcsTable.GL.ClearNamedFramebufferfv = ClearNamedFramebufferfv;
+            funcsTable.GL.ClearNamedFramebufferiv = ClearNamedFramebufferiv;
+            funcsTable.GL.ClearNamedFramebufferuiv = ClearNamedFramebufferuiv;
             funcsTable.GL.ClearNamedFramebufferfi = ClearNamedFramebufferfi;
             funcsTable.GL.BlitFramebuffer = BlitFramebuffer;
             funcsTable.GL.BlitNamedFramebuffer = BlitNamedFramebuffer;
@@ -1050,8 +1028,7 @@ namespace MobileGL::MG_Backend::DirectGLES {
         return m_dynamicParameters;
     }
 
-    void BackendObject_DirectGLES::ApplyGLESCapabilitiesForTesting(
+    void BackendObject_DirectGLES::ApplyGLESCapabilitiesForTesting(const MG_External::GLESCapabilities& capabilities) {
-        const MG_External::GLESCapabilities& capabilities) {
         m_GLESCapabilities = capabilities;
         UpdateDynamicBackendParameters();
     }
@@ -1115,8 +1092,7 @@ namespace MobileGL::MG_Backend::DirectGLES {
         m_dynamicParameters.TextureBufferOffsetAlignment = m_GLESCapabilities.TextureBufferOffsetAlignment;
         m_dynamicParameters.MaxUniformBufferBindings = m_GLESCapabilities.MaxUniformBufferBindings;
         m_dynamicParameters.MaxUniformBlockSize = m_GLESCapabilities.MaxUniformBlockSize;
-        const Int maxSupportedTextureUnits =
+        const Int maxSupportedTextureUnits = static_cast<Int>(MG_State::GLState::TextureState::MAX_TEXTURE_IMAGE_UNITS);
-            static_cast<Int>(MG_State::GLState::TextureState::MAX_TEXTURE_IMAGE_UNITS);
         m_dynamicParameters.MaxImageUnits =
             std::max(std::min(m_GLESCapabilities.MaxImageUnits, maxSupportedTextureUnits), 0);
         m_dynamicParameters.MaxCombinedImageUniforms = std::max(m_GLESCapabilities.MaxCombinedImageUniforms, 0);
@@ -1124,8 +1100,7 @@ namespace MobileGL::MG_Backend::DirectGLES {
             return std::min({std::max(stageLimit, 0), m_dynamicParameters.MaxImageUnits,
                              m_dynamicParameters.MaxCombinedImageUniforms});
         };
-        m_dynamicParameters.MaxVertexImageUniforms =
+        m_dynamicParameters.MaxVertexImageUniforms = clampStageImageUniforms(m_GLESCapabilities.MaxVertexImageUniforms);
-            clampStageImageUniforms(m_GLESCapabilities.MaxVertexImageUniforms);
         m_dynamicParameters.MaxGeometryImageUniforms =
             clampStageImageUniforms(m_GLESCapabilities.MaxGeometryImageUniforms);
         m_dynamicParameters.MaxFragmentImageUniforms =
@@ -1134,6 +1109,11 @@ namespace MobileGL::MG_Backend::DirectGLES {
             clampStageImageUniforms(m_GLESCapabilities.MaxComputeImageUniforms);
         m_dynamicParameters.SupportsDistinctDepthStencilAttachments =
             ProbeDistinctDepthStencilAttachments(DirectGLES::g_GLESFuncs);
+        // SyncAttachmentObject routes a layered upload target to glFramebufferTextureLayer with the
+        // attachment's layer passed through, so this backend really does render to the layer it was
+        // given - provided the driver resolved the entry point at all.
+        m_dynamicParameters.SupportsPerLayerFramebufferAttachment =
+            DirectGLES::g_GLESFuncs.glFramebufferTextureLayer != nullptr;
         m_dynamicParameters.MaxDrawBuffers = m_GLESCapabilities.MaxDrawBuffers;
         m_dynamicParameters.MaxColorAttachments = m_GLESCapabilities.MaxColorAttachments;
         m_dynamicParameters.MaxClipDistances = m_GLESCapabilities.MaxClipDistances;
@@ -1155,8 +1135,7 @@ namespace MobileGL::MG_Backend::DirectGLES {
             const Float requiredMaxOffset =
                 0.5f - std::ldexp(1.0f, -m_GLESCapabilities.FragmentInterpolationOffsetBits);
             if (m_GLESCapabilities.MaxFragmentInterpolationOffset >= requiredMaxOffset) {
-                m_dynamicParameters.MaxFragmentInterpolationOffset =
+                m_dynamicParameters.MaxFragmentInterpolationOffset = m_GLESCapabilities.MaxFragmentInterpolationOffset;
-                    m_GLESCapabilities.MaxFragmentInterpolationOffset;
                 m_dynamicParameters.FragmentInterpolationOffsetBits =
                     m_GLESCapabilities.FragmentInterpolationOffsetBits;
             }
@@ -1165,9 +1144,8 @@ namespace MobileGL::MG_Backend::DirectGLES {
             m_GLESCapabilities.AliasedLineWidthRangeMax > 1.0f || m_GLESCapabilities.SmoothLineWidthRangeMax > 1.0f;
 
         const auto containsAny = [](const String& haystack, std::initializer_list<const char*> needles) {
-            return std::any_of(needles.begin(), needles.end(), [&](const char* needle) {
+            return std::any_of(needles.begin(), needles.end(),
-                return haystack.find(needle) != String::npos;
+                               [&](const char* needle) { return haystack.find(needle) != String::npos; });
-            });
         };
         const String vendorAndRenderer =
             m_GLESCapabilities.GLESVendorString + " " + m_GLESCapabilities.GLESRendererString;

MobileGL/MG_Backend/DirectGLES/DirectGLES.cpp

@@ -1663,8 +1663,11 @@ namespace MobileGL::MG_Backend::DirectGLES {
                 if (backendSamplerIt != SamplerImpl::g_backendSamplerObjects.end()) {
                     backendSamplerIt->second->Bind(unit);
                 }
-
             } else {
+                // Symmetric with the bind above: a sampler object left on the unit by an earlier
+                // draw keeps being applied, and on a multisample texture - which takes no sampler
+                // object at all - the draw is rejected outright.
+                SamplerImpl::UnbindSampler(unit);
             }
         }
     }
@@ -1844,6 +1847,10 @@ namespace MobileGL::MG_Backend::DirectGLES {
                                 backendObj = MakeShared<SamplerImpl::BackendSamplerObject>();
                             }
                             backendObj->SyncToBackend(samplerObject);
+                            // Syncing the object's parameters is not the same as putting it on the
+                            // unit: without this the driver kept sampling with the texture's own
+                            // parameters and every sampler object was inert.
+                            backendObj->Bind(unit);
                         } else {
                             SamplerImpl::UnbindSampler(unit);
                         }
@@ -4431,6 +4438,40 @@ namespace MobileGL::MG_Backend::DirectGLES {
         ForceBindCurrentFBO(FramebufferTarget::Draw);
     }
 
+    void ClearNamedFramebufferiv(const SharedPtr<MG_State::GLState::FramebufferObject>& framebuffer,
+                                 GLenum buffer, GLint drawbuffer, const GLint* value) {
+#if MOBILEGL_LOG_ACTIVE_LEVEL <= MOBILEGL_LOG_LEVEL_DEBUG && MOBILEGL_ENABLE_SCOPE_MARKER
+        DebugImpl::OpenGLScopeMarker marker(__func__);
+#endif
+        TextureImpl::SyncNeccessaryTextures();
+        RenderStateImpl::SyncRenderState();
+
+        SyncAndBindFramebufferObject(framebuffer, FramebufferTarget::Draw, true);
+        g_GLESFuncs.glClearBufferiv(buffer, drawbuffer, value);
+        DebugImpl::ErrorLopper::Loop([file = __FILE__, line = __LINE__](auto err) {
+            MGLOG_D("ES error (%s:%d): %s", file, line, MG_Util::ConvertGLEnumToString(err).c_str());
+        });
+
+        ForceBindCurrentFBO(FramebufferTarget::Draw);
+    }
+
+    void ClearNamedFramebufferuiv(const SharedPtr<MG_State::GLState::FramebufferObject>& framebuffer,
+                                  GLenum buffer, GLint drawbuffer, const GLuint* value) {
+#if MOBILEGL_LOG_ACTIVE_LEVEL <= MOBILEGL_LOG_LEVEL_DEBUG && MOBILEGL_ENABLE_SCOPE_MARKER
+        DebugImpl::OpenGLScopeMarker marker(__func__);
+#endif
+        TextureImpl::SyncNeccessaryTextures();
+        RenderStateImpl::SyncRenderState();
+
+        SyncAndBindFramebufferObject(framebuffer, FramebufferTarget::Draw, true);
+        g_GLESFuncs.glClearBufferuiv(buffer, drawbuffer, value);
+        DebugImpl::ErrorLopper::Loop([file = __FILE__, line = __LINE__](auto err) {
+            MGLOG_D("ES error (%s:%d): %s", file, line, MG_Util::ConvertGLEnumToString(err).c_str());
+        });
+
+        ForceBindCurrentFBO(FramebufferTarget::Draw);
+    }
+
     static SizeT AlignPixelRow(SizeT rowBytes, Int alignment) {
         const SizeT resolvedAlignment = static_cast<SizeT>(std::max(alignment, 1));
         return (rowBytes + resolvedAlignment - 1) & ~(resolvedAlignment - 1);
@@ -4532,14 +4573,27 @@ namespace MobileGL::MG_Backend::DirectGLES {
         Vector<Uint32> packed(outStencil.size(), 0);
         ClearGLErrors();
         g_GLESFuncs.glReadPixels(x, y, width, height, GL_DEPTH_STENCIL, GL_UNSIGNED_INT_24_8, packed.data());
+        if (g_GLESFuncs.glGetError() == GL_NO_ERROR) {
+            for (SizeT i = 0; i < outStencil.size(); ++i) {
+                outStencil[i] = static_cast<Uint8>(packed[i] & 0xFFu);
+            }
+            return true;
+        }
+
+        // A DEPTH32F_STENCIL8 attachment rejects the 24_8 type: its packed layout is a float depth
+        // followed by a padded stencil byte, eight bytes per pixel with the index at offset 4.
+        Vector<Uint8> packed32f(outStencil.size() * 8u, 0);
+        ClearGLErrors();
+        g_GLESFuncs.glReadPixels(x, y, width, height, GL_DEPTH_STENCIL, GL_FLOAT_32_UNSIGNED_INT_24_8_REV,
+                                 packed32f.data());
         const GLenum packedError = g_GLESFuncs.glGetError();
         if (packedError != GL_NO_ERROR) {
-            MGLOG_E("ReadPixels: neither GL_STENCIL_INDEX nor GL_DEPTH_STENCIL readback is available: %s",
+            MGLOG_E("ReadPixels: no stencil readback path is available: %s",
                     MG_Util::ConvertGLEnumToString(packedError).c_str());
             return false;
         }
         for (SizeT i = 0; i < outStencil.size(); ++i) {
-            outStencil[i] = static_cast<Uint8>(packed[i] & 0xFFu);
+            outStencil[i] = packed32f[i * 8u + 4u];
         }
         return true;
     }
@@ -4549,11 +4603,20 @@ namespace MobileGL::MG_Backend::DirectGLES {
     // values themselves are always 8 bits.
     static Bool ReadPixelsStencilViaNative(GLint x, GLint y, GLsizei width, GLsizei height, GLenum type,
                                            void* pixels) {
+        // GL 4.6 core 18.2.8: a stencil index is written unconverted into whichever integer width
+        // the client asked for, and converted to a float value for GL_FLOAT. The signed widths are
+        // as legal as the unsigned ones - the CTS reads stencil with GL_INT - and rejecting them
+        // here used to let the call fall through to a native ES read the driver refuses, after
+        // which nothing was written at all and the caller kept its zeros.
         SizeT dstPixelBytes = 0;
         switch (type) {
-        case GL_UNSIGNED_BYTE: dstPixelBytes = sizeof(Uint8); break;
+        case GL_UNSIGNED_BYTE:
-        case GL_UNSIGNED_SHORT: dstPixelBytes = sizeof(Uint16); break;
+        case GL_BYTE: dstPixelBytes = sizeof(Uint8); break;
-        case GL_UNSIGNED_INT: dstPixelBytes = sizeof(Uint32); break;
+        case GL_UNSIGNED_SHORT:
+        case GL_SHORT: dstPixelBytes = sizeof(Uint16); break;
+        case GL_UNSIGNED_INT:
+        case GL_INT: dstPixelBytes = sizeof(Uint32); break;
+        case GL_FLOAT: dstPixelBytes = sizeof(GLfloat); break;
         default: return false;
         }
         if (width <= 0 || height <= 0) {
@@ -4587,11 +4650,16 @@ namespace MobileGL::MG_Backend::DirectGLES {
             for (GLsizei col = 0; col < width; ++col) {
                 switch (type) {
                 case GL_UNSIGNED_BYTE:
+                case GL_BYTE:
                     rowBuf[static_cast<SizeT>(col)] = srcRow[col];
                     break;
                 case GL_UNSIGNED_SHORT:
+                case GL_SHORT:
                     reinterpret_cast<Uint16*>(rowBuf.data())[col] = srcRow[col];
                     break;
+                case GL_FLOAT:
+                    reinterpret_cast<GLfloat*>(rowBuf.data())[col] = static_cast<GLfloat>(srcRow[col]);
+                    break;
                 default:
                     reinterpret_cast<Uint32*>(rowBuf.data())[col] = srcRow[col];
                     break;
@@ -5284,10 +5352,58 @@ namespace MobileGL::MG_Backend::DirectGLES {
             const Bool applyPackImageParams = backendAttachTarget == GL_TEXTURE_3D ||
                                               backendAttachTarget == GL_TEXTURE_2D_ARRAY ||
                                               backendAttachTarget == GL_TEXTURE_CUBE_MAP_ARRAY;
-            // 3D/array images read back every slice, but the FBO path can only read one layer:
-            // multi-slice reads are served from the CPU shadow (slice-major, tight layout).
             const GLsizei sliceCount = std::max(size.z(), 1);
             const Bool multiSlice = size.z() > 1;
+            // A multi-slice read used to go to the CPU shadow outright, on the grounds that the
+            // scratch FBO can only expose one layer at a time. But the shadow only holds what was
+            // uploaded, so every slice that was rendered to came back stale - which is exactly what
+            // a layered framebuffer produces, and what textures_storage_multisample_3d_* checks.
+            // Attach the layers one at a time instead and read each off the GPU, keeping the shadow
+            // for the formats the FBO cannot represent at all.
+            if (multiSlice && tempFBOComplete &&
+                (backendAttachTarget == GL_TEXTURE_3D || backendAttachTarget == GL_TEXTURE_2D_ARRAY)) {
+                // Each slice is packed as its own 2D image, so the per-slice call must not apply
+                // GL_PACK_SKIP_IMAGES / GL_PACK_IMAGE_HEIGHT itself - this walks the destination
+                // over them, using the same layout StoreWideRowsToClient computes.
+                const auto packParams = MG_State::pGLContext->GetPixelStoreParameters(false);
+                const SizeT dstPixelBytes = GetReadbackDstPixelSize(conversionMapping, type);
+                const SizeT rowPixels =
+                    static_cast<SizeT>(packParams.RowLength > 0 ? packParams.RowLength : size.x());
+                const SizeT alignment =
+                    packParams.Alignment > 0 ? static_cast<SizeT>(packParams.Alignment) : SizeT{1};
+                const SizeT dstRowStride = (rowPixels * dstPixelBytes + alignment - 1) / alignment * alignment;
+                const SizeT imageRows = applyPackImageParams && packParams.ImageHeight > 0
+                    ? static_cast<SizeT>(packParams.ImageHeight)
+                    : static_cast<SizeT>(size.y());
+                const SizeT dstImageStride = imageRows * dstRowStride;
+                const SizeT skipImages =
+                    applyPackImageParams ? static_cast<SizeT>(std::max(packParams.SkipImages, 0)) : SizeT{0};
+
+                Bool allSlicesRead = true;
+                for (GLsizei slice = 0; slice < sliceCount; ++slice) {
+                    ScratchFBOImpl::EnsureColorAttachmentLayer(tempFB, GL_READ_FRAMEBUFFER, backendTexId, level,
+                                                               slice);
+                    if (g_GLESFuncs.glCheckFramebufferStatus(GL_READ_FRAMEBUFFER) != GL_FRAMEBUFFER_COMPLETE) {
+                        allSlicesRead = false;
+                        break;
+                    }
+                    const SizeT sliceOffset = (skipImages + static_cast<SizeT>(slice)) * dstImageStride;
+                    void* sliceDst = static_cast<Uint8*>(pixels) + sliceOffset;
+                    if (!ReadPixelsViaFormatConversion(0, 0, size.x(), size.y(), format, type, sliceDst,
+                                                       /*honorPackImageParams=*/false,
+                                                       /*applyFixedPointReadClamp=*/false)) {
+                        allSlicesRead = false;
+                        break;
+                    }
+                }
+                // Leave the scratch FBO on layer 0 so the single-slice paths below see what they
+                // set up.
+                ScratchFBOImpl::EnsureColorAttachmentLayer(tempFB, GL_READ_FRAMEBUFFER, backendTexId, level, 0);
+                if (allSlicesRead) {
+                    MGLOG_D("GetTexImage: finished %d slices via per-layer readback", sliceCount);
+                    return;
+                }
+            }
             if (multiSlice &&
                 GetTexImageViaShadowConversion(textureMipmapObject,
                                                MG_Util::ConvertGLEnumToTextureUploadTarget(target), level, size.x(),

MobileGL/MG_Backend/DirectGLES/DirectGLES.h

@@ -59,6 +59,10 @@ namespace MobileGL::MG_Backend::DirectGLES {
                                  GLenum buffer, GLint drawbuffer, const GLfloat* value);
     void ClearNamedFramebufferfi(const SharedPtr<MG_State::GLState::FramebufferObject>& framebuffer,
                                  GLenum buffer, GLint drawbuffer, GLfloat depth, GLint stencil);
+    void ClearNamedFramebufferiv(const SharedPtr<MG_State::GLState::FramebufferObject>& framebuffer,
+                                 GLenum buffer, GLint drawbuffer, const GLint* value);
+    void ClearNamedFramebufferuiv(const SharedPtr<MG_State::GLState::FramebufferObject>& framebuffer,
+                                  GLenum buffer, GLint drawbuffer, const GLuint* value);
     void BlitFramebuffer(GLint srcX0, GLint srcY0, GLint srcX1, GLint srcY1, GLint dstX0, GLint dstY0, GLint dstX1,
                          GLint dstY1, GLbitfield mask, GLenum filter);
     void BlitNamedFramebuffer(const SharedPtr<MG_State::GLState::FramebufferObject>& readFramebuffer,

MobileGL/MG_Backend/DirectGLES/Managers.cpp

@@ -2479,7 +2479,10 @@ namespace MobileGL::MG_Backend::DirectGLES {
                 });
             }
 
-            if (!isMultisampleTarget && m_cacheBorderColor != stateTextureObject->GetBorderColor()) {
+            // GL_TEXTURE_BORDER_COLOR needs ES 3.2 or EXT/OES_texture_border_clamp; on a driver
+            // without it every such call is INVALID_ENUM, so the parameter is simply not synced.
+            if (!isMultisampleTarget && g_GLESCapabilities.SupportsTextureBorderClamp &&
+                m_cacheBorderColor != stateTextureObject->GetBorderColor()) {
                 const auto& borderColor = stateTextureObject->GetBorderColor();
                 GLfloat borderColorArray[4] = {borderColor.x(), borderColor.y(), borderColor.z(), borderColor.w()};
                 g_GLESFuncs.glTexParameterfv(target, GL_TEXTURE_BORDER_COLOR, borderColorArray);
@@ -3755,6 +3758,16 @@ namespace MobileGL::MG_Backend::DirectGLES {
                 }
                 m_cacheSamplerParameters.maxAnisotropy = samplerParams.maxAnisotropy;
             }
+            if (m_cacheSamplerParameters.borderColor != samplerParams.borderColor) {
+                // Same gate as the texture-side border colour above.
+                if (g_GLESCapabilities.SupportsTextureBorderClamp && g_GLESFuncs.glSamplerParameterfv) {
+                    const GLfloat borderColorArray[4] = {
+                        samplerParams.borderColor.x(), samplerParams.borderColor.y(),
+                        samplerParams.borderColor.z(), samplerParams.borderColor.w()};
+                    g_GLESFuncs.glSamplerParameterfv(m_backendSamplerId, GL_TEXTURE_BORDER_COLOR, borderColorArray);
+                }
+                m_cacheSamplerParameters.borderColor = samplerParams.borderColor;
+            }
 #undef SYNC_SAMPLER_PARAM_IF_CHANGED
             m_isInitialized = true;
         }

MobileGL/MG_Backend/DirectVulkan/BackendObject_DirectVulkan.cpp

@@ -41,13 +41,11 @@ namespace MobileGL::MG_Backend::DirectVulkan {
         Bool IsLayeredTarget(TextureTarget target) {
             return target == TextureTarget::Texture3D || target == TextureTarget::Texture1DArray ||
                    target == TextureTarget::Texture2DArray || target == TextureTarget::TextureCubeMap ||
-                   target == TextureTarget::TextureCubeMapArray ||
+                   target == TextureTarget::TextureCubeMapArray || target == TextureTarget::Texture2DMultisampleArray;
-                   target == TextureTarget::Texture2DMultisampleArray;
         }
 
         Bool IsMultisampleTarget(TextureTarget target) {
-            return target == TextureTarget::Texture2DMultisample ||
+            return target == TextureTarget::Texture2DMultisample || target == TextureTarget::Texture2DMultisampleArray;
-                   target == TextureTarget::Texture2DMultisampleArray;
         }
 
         Bool IsTextureBufferTarget(TextureTarget target) {
@@ -59,8 +57,8 @@ namespace MobileGL::MG_Backend::DirectVulkan {
             GLenum normalizedInternalFormat = glFormat;
             GLenum imageFormat = GL_RGBA;
             GLenum imageType = GL_UNSIGNED_BYTE;
-            MG_Util::TextureFormatProcessor::NormalizePixelFormat(
+            MG_Util::TextureFormatProcessor::NormalizePixelFormat(glFormat, PixelFormatNormalizeOptionBit::None,
-                glFormat, PixelFormatNormalizeOptionBit::None, &normalizedInternalFormat, &imageFormat, &imageType);
+                                                                  &normalizedInternalFormat, &imageFormat, &imageType);
             return imageFormat == GL_RED_INTEGER || imageFormat == GL_RG_INTEGER || imageFormat == GL_RGB_INTEGER ||
                    imageFormat == GL_RGBA_INTEGER;
         }
@@ -81,8 +79,7 @@ namespace MobileGL::MG_Backend::DirectVulkan {
             return caps;
         }
 
-        FormatCapabilityFlags BuildVulkanCaps(TextureInternalFormat logicalFormat,
+        FormatCapabilityFlags BuildVulkanCaps(TextureInternalFormat logicalFormat, TextureTarget target,
-                                              TextureTarget target,
                                               VkFormatFeatureFlags features) {
             FormatCapabilityFlags caps;
             const Bool isDepth = MG_Util::IsDepthFormatInternalFormat(logicalFormat);
@@ -101,8 +98,7 @@ namespace MobileGL::MG_Backend::DirectVulkan {
             const Bool sampled = (features & VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT) != 0;
             const Bool linearFilter = (features & VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT) != 0;
             const Bool colorRenderable = (features & VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT) != 0;
-            const Bool depthStencilRenderable =
+            const Bool depthStencilRenderable = (features & VK_FORMAT_FEATURE_DEPTH_STENCIL_ATTACHMENT_BIT) != 0;
-                (features & VK_FORMAT_FEATURE_DEPTH_STENCIL_ATTACHMENT_BIT) != 0;
             const Bool renderable = (isDepth || isStencil) ? depthStencilRenderable : colorRenderable;
 
             if (sampled || renderable) {
@@ -199,21 +195,20 @@ namespace MobileGL::MG_Backend::DirectVulkan {
 
         Bool HasNewCaveatFormatCaps(FormatCapabilityFlags nativeCaps, FormatCapabilityFlags fallbackCaps) {
             for (FormatCapability capability : kReportedFormatCapabilities) {
-                if (HasFormatCapability(fallbackCaps, capability) &&
+                if (HasFormatCapability(fallbackCaps, capability) && !HasFormatCapability(nativeCaps, capability)) {
-                    !HasFormatCapability(nativeCaps, capability)) {
                     return true;
                 }
             }
             return false;
         }
 
-        void LogVulkanFormatCaveat(TextureInternalFormat logicalFormat,
+        void LogVulkanFormatCaveat(TextureInternalFormat logicalFormat, SizeT targetIndex,
-                                   SizeT targetIndex,
                                    TextureInternalFormat fallbackFormat) {
-            MGLOG_D("Caveat: %s %s not fully supported. Reason: native Vulkan format is not fully supported. Fallback: %s",
+            MGLOG_D(
-                    GetFormatCapabilityTargetName(targetIndex).c_str(),
+                "Caveat: %s %s not fully supported. Reason: native Vulkan format is not fully supported. Fallback: %s",
-                    MG_Util::ConvertTextureInternalFormatToString(logicalFormat).c_str(),
+                GetFormatCapabilityTargetName(targetIndex).c_str(),
-                    MG_Util::ConvertTextureInternalFormatToString(fallbackFormat).c_str());
+                MG_Util::ConvertTextureInternalFormatToString(logicalFormat).c_str(),
+                MG_Util::ConvertTextureInternalFormatToString(fallbackFormat).c_str());
         }
 
         Vector<Int> BuildSampleCounts(Int maxSamples) {
@@ -257,15 +252,15 @@ namespace MobileGL::MG_Backend::DirectVulkan {
 
                 for (SizeT targetIndex = 0; targetIndex < kFormatCapabilityTextureTargetCount; ++targetIndex) {
                     const auto target = static_cast<TextureTarget>(targetIndex);
-                    const VkFormatFeatureFlags nativeFeatures =
+                    const VkFormatFeatureFlags nativeFeatures = IsTextureBufferTarget(target)
-                        IsTextureBufferTarget(target) ? nativeProperties.bufferFeatures
+                                                                    ? nativeProperties.bufferFeatures
-                                                      : nativeProperties.optimalTilingFeatures;
+                                                                    : nativeProperties.optimalTilingFeatures;
                     FormatCapabilityFlags nativeCaps = BuildVulkanCaps(logicalFormat, target, nativeFeatures);
                     cache.FullCaps[targetIndex][formatIndex] |= nativeCaps;
 
-                    const VkFormatFeatureFlags fallbackFeatures =
+                    const VkFormatFeatureFlags fallbackFeatures = IsTextureBufferTarget(target)
-                        IsTextureBufferTarget(target) ? fallbackProperties.bufferFeatures
+                                                                      ? fallbackProperties.bufferFeatures
-                                                      : fallbackProperties.optimalTilingFeatures;
+                                                                      : fallbackProperties.optimalTilingFeatures;
                     FormatCapabilityFlags fallbackCaps = BuildVulkanCaps(logicalFormat, target, fallbackFeatures);
                     if (fallbackFormat != VK_FORMAT_UNDEFINED && fallbackFormat != nativeFormat) {
                         cache.CaveatCaps[targetIndex][formatIndex] |= fallbackCaps;
@@ -300,9 +295,8 @@ namespace MobileGL::MG_Backend::DirectVulkan {
                 cache.FullCaps[renderbufferTargetIndex][formatIndex] |= renderbufferCaps;
 
                 if (fallbackFormat != VK_FORMAT_UNDEFINED && fallbackFormat != nativeFormat) {
-                    FormatCapabilityFlags fallbackRenderbufferCaps =
+                    FormatCapabilityFlags fallbackRenderbufferCaps = BuildVulkanCaps(
-                        BuildVulkanCaps(logicalFormat, TextureTarget::Texture2D,
+                        logicalFormat, TextureTarget::Texture2D, fallbackProperties.optimalTilingFeatures);
-                                        fallbackProperties.optimalTilingFeatures);
                     fallbackRenderbufferCaps &= FormatCapability::Creatable;
                     if ((fallbackProperties.optimalTilingFeatures &
                          (VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT | VK_FORMAT_FEATURE_DEPTH_STENCIL_ATTACHMENT_BIT)) !=
@@ -311,8 +305,7 @@ namespace MobileGL::MG_Backend::DirectVulkan {
                         fallbackRenderbufferCaps |= FormatCapability::MultisampleRenderbuffer;
                     }
                     cache.CaveatCaps[renderbufferTargetIndex][formatIndex] |= fallbackRenderbufferCaps;
-                    if (fallbackLogicalFormat &&
+                    if (fallbackLogicalFormat && HasNewCaveatFormatCaps(renderbufferCaps, fallbackRenderbufferCaps)) {
-                        HasNewCaveatFormatCaps(renderbufferCaps, fallbackRenderbufferCaps)) {
                         LogVulkanFormatCaveat(logicalFormat, renderbufferTargetIndex, *fallbackLogicalFormat);
                     }
                 }
@@ -329,14 +322,13 @@ namespace MobileGL::MG_Backend::DirectVulkan {
 
     void PopulateFormatCapabilities(VkPhysicalDevice physicalDevice,
                                     PFN_vkGetPhysicalDeviceFormatProperties getFormatProperties,
-                                    const MG_External::VulkanCapabilities& capabilities,
+                                    const MG_External::VulkanCapabilities& capabilities, FormatCapabilityCache& cache) {
-                                    FormatCapabilityCache& cache) {
         PopulateFormatCapabilitiesImpl(physicalDevice, getFormatProperties, capabilities, cache);
     }
 
     BackendObject_DirectVulkan::~BackendObject_DirectVulkan() = default;
 
-    BackendObject_DirectVulkan::BackendObject_DirectVulkan(): m_rendererInfo{GetRendererIdentity()} {}
+    BackendObject_DirectVulkan::BackendObject_DirectVulkan() : m_rendererInfo{GetRendererIdentity()} {}
 
     Bool BackendObject_DirectVulkan::InitWindowSurface() {
         if (!m_windowHandle.Handle) {
@@ -410,10 +402,8 @@ namespace MobileGL::MG_Backend::DirectVulkan {
             MGLOG_E("DirectVulkan backend not initialized");
             return false;
         }
-        if (!handle.Handle || (handle.Backend != WindowBackend::Android &&
+        if (!handle.Handle || (handle.Backend != WindowBackend::Android && handle.Backend != WindowBackend::X11 &&
-                               handle.Backend != WindowBackend::X11 &&
+                               handle.Backend != WindowBackend::MetalLayer && handle.Backend != WindowBackend::Win32)) {
-                               handle.Backend != WindowBackend::MetalLayer &&
-                               handle.Backend != WindowBackend::Win32)) {
             MGLOG_E("DirectVulkan backend only supports Android, X11, CAMetalLayer, and Win32 native windows");
             return false;
         }
@@ -504,37 +494,32 @@ namespace MobileGL::MG_Backend::DirectVulkan {
             .RendererName = "Magma",
             .BackendName = "Direct (Vulkan)",
             .ExtraVendor = Nullopt,
-            .RendererGLInfo =
+            .RendererGLInfo = {.TargetGLVersion = {4, 0, 0},
-                {
+                               .TargetGLSLVersion = {4, 6, 0},
-                    .TargetGLVersion = {3, 3, 0},
+                               // Baseline advertisement (no shader subgroup, no timer queries); a
-                    .TargetGLSLVersion = {4, 6, 0},
+                               // live backend reconciles its copy in UpdateAdvertisedExtensions.
-                    // Baseline advertisement (no shader subgroup, no timer queries); a
+                               .Extensions = BuildAdvertisedExtensions(false, false, false),
-                    // live backend reconciles its copy in UpdateAdvertisedExtensions.
+                               .IsCompatibilityProfile = false},
-                    .Extensions = BuildAdvertisedExtensions(false, false, false),
-                    .IsCompatibilityProfile = false
-                },
             .StaticBackendCapability = {.AllowVSOnlyPrograms = false}};
         return rendererInfo;
     }
 
     Vector<GLExtension> BuildAdvertisedExtensions(Bool shaderSubgroupSupported, Bool timerQueriesSupported,
                                                   Bool anisotropicFilteringSupported) {
-        Vector<GLExtension> extensions = {V_OpenGL30, V_OpenGL31, V_OpenGL32,
+        Vector<GLExtension> extensions = {
-                                          V_OpenGL33, E_GL_ARB_draw_buffers_blend, E_GL_ARB_compute_shader,
+            V_OpenGL30, V_OpenGL31, V_OpenGL32, V_OpenGL33, V_OpenGL40, E_GL_ARB_draw_buffers_blend,
-                                          E_GL_ARB_shader_storage_buffer_object, E_GL_ARB_shader_image_load_store,
+            E_GL_ARB_compute_shader, E_GL_ARB_shader_storage_buffer_object, E_GL_ARB_shader_image_load_store,
-                                          E_GL_ARB_program_interface_query, E_GL_ARB_framebuffer_object,
+            E_GL_ARB_program_interface_query, E_GL_ARB_framebuffer_object, E_GL_ARB_multi_draw_indirect,
-                                          E_GL_ARB_multi_draw_indirect, E_GL_ARB_indirect_parameters,
+            E_GL_ARB_indirect_parameters, E_GL_EXT_framebuffer_object, E_GL_ARB_depth_texture, E_GL_ARB_buffer_storage,
-                                          E_GL_EXT_framebuffer_object, E_GL_ARB_depth_texture, E_GL_ARB_buffer_storage,
+            E_GL_ARB_texture_storage, E_GL_ARB_texture_storage_multisample, E_GL_ARB_texture_multisample,
-                                          E_GL_ARB_texture_storage, E_GL_ARB_texture_storage_multisample,
+            E_GL_ARB_clear_texture, E_GL_ARB_direct_state_access, E_GL_ARB_shader_draw_parameters,
-                                          E_GL_ARB_texture_multisample, E_GL_ARB_clear_texture, E_GL_ARB_direct_state_access,
+            E_GL_ARB_gpu_shader_int64, E_GL_KHR_debug, E_GL_ARB_gpu_shader5, E_GL_ARB_multi_bind,
-                                          E_GL_ARB_shader_draw_parameters, E_GL_ARB_gpu_shader_int64, E_GL_KHR_debug,
+            E_GL_ARB_shading_language_420pack, E_GL_ARB_vertex_attrib_binding, E_GL_ARB_shader_image_size,
-                                          E_GL_ARB_gpu_shader5, E_GL_ARB_multi_bind, E_GL_ARB_shading_language_420pack,
+            E_GL_ARB_explicit_attrib_location,
-                                          E_GL_ARB_vertex_attrib_binding, E_GL_ARB_shader_image_size,
+            // Advertised with GL_NUM_PROGRAM_BINARY_FORMATS = 0, which the
-                                          E_GL_ARB_explicit_attrib_location,
+            // extension explicitly permits. It is also the only thing that
-                                          // Advertised with GL_NUM_PROGRAM_BINARY_FORMATS = 0, which the
+            // exposes glProgramParameteri before GL 4.1.
-                                          // extension explicitly permits. It is also the only thing that
+            E_GL_ARB_get_program_binary};
-                                          // exposes glProgramParameteri before GL 4.1.
-                                          E_GL_ARB_get_program_binary};
         if (shaderSubgroupSupported && !MG_Config::Features.DisableSubgroup) {
             extensions.push_back(E_GL_KHR_shader_subgroup);
         }
@@ -597,6 +582,8 @@ namespace MobileGL::MG_Backend::DirectVulkan {
             funcsTable.GL.ClearBufferiv = ClearBufferiv;
             funcsTable.GL.ClearNamedFramebufferfv = ClearNamedFramebufferfv;
             funcsTable.GL.ClearNamedFramebufferfi = ClearNamedFramebufferfi;
+            funcsTable.GL.ClearNamedFramebufferiv = ClearNamedFramebufferiv;
+            funcsTable.GL.ClearNamedFramebufferuiv = ClearNamedFramebufferuiv;
             funcsTable.GL.BlitFramebuffer = BlitFramebuffer;
             funcsTable.GL.BlitNamedFramebuffer = BlitNamedFramebuffer;
             funcsTable.GL.CopyTexImage2D = CopyTexImage2D;
@@ -729,7 +716,7 @@ namespace MobileGL::MG_Backend::DirectVulkan {
         // rather than a maximum the sampler manager will never apply.
         m_dynamicParameters.MaxTextureMaxAnisotropy =
             (pVulkanRenderer && pVulkanRenderer->IsSamplerAnisotropySupported()) ? m_vulkanCaps.MaxSamplerAnisotropy
-                                                                                : 1.0f;
+                                                                                 : 1.0f;
         m_dynamicParameters.SmoothLineWidthRangeMin = m_vulkanCaps.SmoothLineWidthRangeMin;
         m_dynamicParameters.SmoothLineWidthRangeMax = m_vulkanCaps.SmoothLineWidthRangeMax;
         m_dynamicParameters.SmoothLineWidthGranularity = m_vulkanCaps.SmoothLineWidthGranularity;
@@ -750,8 +737,7 @@ namespace MobileGL::MG_Backend::DirectVulkan {
         m_dynamicParameters.MaxIntegerSamples = m_vulkanCaps.MaxIntegerSamples;
         m_dynamicParameters.MaxSamples = m_vulkanCaps.MaxSamples;
         m_dynamicParameters.MaxSampleMaskWords = m_vulkanCaps.MaxSampleMaskWords;
-        const Int maxSupportedTextureUnits =
+        const Int maxSupportedTextureUnits = static_cast<Int>(MG_State::GLState::TextureState::MAX_TEXTURE_IMAGE_UNITS);
-            static_cast<Int>(MG_State::GLState::TextureState::MAX_TEXTURE_IMAGE_UNITS);
         // GL_MAX_TEXTURE_IMAGE_UNITS is a *per-stage* sampler limit. Adreno/Qualcomm report a huge
         // maxPerStageDescriptorSampledImages (descriptor-indexing scale), so clamping it only to our
         // combined array capacity (192) still advertises 192 per stage. Host code treats this value as
@@ -761,8 +747,7 @@ namespace MobileGL::MG_Backend::DirectVulkan {
         // limits while keeping the combined limit at our texture-unit array capacity.
         constexpr Int maxPerStageTextureUnits =
             static_cast<Int>(MG_State::GLState::TextureState::MAX_PER_STAGE_TEXTURE_IMAGE_UNITS);
-        m_dynamicParameters.MaxTextureImageUnits =
+        m_dynamicParameters.MaxTextureImageUnits = std::min(m_vulkanCaps.MaxTextureImageUnits, maxPerStageTextureUnits);
-            std::min(m_vulkanCaps.MaxTextureImageUnits, maxPerStageTextureUnits);
         m_dynamicParameters.MaxVertexTextureImageUnits =
             std::min(m_vulkanCaps.MaxVertexTextureImageUnits, maxPerStageTextureUnits);
         m_dynamicParameters.MaxComputeTextureImageUnits =
@@ -771,9 +756,8 @@ namespace MobileGL::MG_Backend::DirectVulkan {
             std::min(m_vulkanCaps.MaxCombinedTextureImageUnits, maxSupportedTextureUnits);
         // Never advertise more attributes than the state layer can store: the current-value array and
         // the Uint32 attribute masks the draw path passes around are both bounded by MAX_VERTEX_ATTRIBS.
-        m_dynamicParameters.MaxVertexAttribs =
+        m_dynamicParameters.MaxVertexAttribs = std::min(
-            std::min(m_vulkanCaps.MaxVertexAttribs,
+            m_vulkanCaps.MaxVertexAttribs, static_cast<Int>(MG_State::GLState::VertexArrayObject::MAX_VERTEX_ATTRIBS));
-                     static_cast<Int>(MG_State::GLState::VertexArrayObject::MAX_VERTEX_ATTRIBS));
         m_dynamicParameters.MaxComputeShaderStorageBlocks = m_vulkanCaps.MaxComputeShaderStorageBlocks;
         m_dynamicParameters.MaxCombinedShaderStorageBlocks = m_vulkanCaps.MaxCombinedShaderStorageBlocks;
         m_dynamicParameters.MaxComputeUniformBlocks = m_vulkanCaps.MaxComputeUniformBlocks;
@@ -783,8 +767,7 @@ namespace MobileGL::MG_Backend::DirectVulkan {
         m_dynamicParameters.TextureBufferOffsetAlignment = m_vulkanCaps.TextureBufferOffsetAlignment;
         m_dynamicParameters.MaxUniformBufferBindings = m_vulkanCaps.MaxUniformBufferBindings;
         m_dynamicParameters.MaxUniformBlockSize = m_vulkanCaps.MaxUniformBlockSize;
-        m_dynamicParameters.MaxImageUnits =
+        m_dynamicParameters.MaxImageUnits = std::max(std::min(m_vulkanCaps.MaxImageUnits, maxSupportedTextureUnits), 0);
-            std::max(std::min(m_vulkanCaps.MaxImageUnits, maxSupportedTextureUnits), 0);
         m_dynamicParameters.MaxCombinedImageUniforms = std::max(m_vulkanCaps.MaxCombinedImageUniforms, 0);
         const Int maxPerStageImageUniforms =
             std::min(m_dynamicParameters.MaxImageUnits, m_dynamicParameters.MaxCombinedImageUniforms);
@@ -801,8 +784,7 @@ namespace MobileGL::MG_Backend::DirectVulkan {
             m_vulkanCaps.SupportsFragmentStoresAndAtomics ? maxPerStageImageUniforms : 0;
         m_dynamicParameters.MaxComputeImageUniforms =
             std::min(std::max(m_vulkanCaps.MaxComputeImageUniforms, 0), maxPerStageImageUniforms);
-        const Int maxSupportedDrawBuffers =
+        const Int maxSupportedDrawBuffers = static_cast<Int>(MG_State::GLState::FramebufferObject::MAX_DRAW_BUFFERS);
-            static_cast<Int>(MG_State::GLState::FramebufferObject::MAX_DRAW_BUFFERS);
         m_dynamicParameters.MaxDrawBuffers = std::min(m_vulkanCaps.MaxDrawBuffers, maxSupportedDrawBuffers);
         m_dynamicParameters.MaxColorAttachments = std::min(m_vulkanCaps.MaxColorAttachments, maxSupportedDrawBuffers);
         m_dynamicParameters.MaxClipDistances = m_vulkanCaps.MaxClipDistances;
@@ -821,12 +803,10 @@ namespace MobileGL::MG_Backend::DirectVulkan {
         m_dynamicParameters.FragmentInterpolationOffsetBits = 4;
         if (m_vulkanCaps.FragmentInterpolationOffsetBits >= 4 &&
             std::isfinite(m_vulkanCaps.MaxFragmentInterpolationOffset)) {
-            const Float requiredMaxOffset =
+            const Float requiredMaxOffset = 0.5f - std::ldexp(1.0f, -m_vulkanCaps.FragmentInterpolationOffsetBits);
-                0.5f - std::ldexp(1.0f, -m_vulkanCaps.FragmentInterpolationOffsetBits);
             if (m_vulkanCaps.MaxFragmentInterpolationOffset >= requiredMaxOffset) {
                 m_dynamicParameters.MaxFragmentInterpolationOffset = m_vulkanCaps.MaxFragmentInterpolationOffset;
-                m_dynamicParameters.FragmentInterpolationOffsetBits =
+                m_dynamicParameters.FragmentInterpolationOffsetBits = m_vulkanCaps.FragmentInterpolationOffsetBits;
-                    m_vulkanCaps.FragmentInterpolationOffsetBits;
             }
         }
         m_dynamicParameters.SupportsWideLines = m_vulkanCaps.SupportsWideLines;
@@ -835,7 +815,8 @@ namespace MobileGL::MG_Backend::DirectVulkan {
         if (m_vulkanCaps.SupportsShaderSubgroup) {
             m_dynamicParameters.SubgroupSize = m_vulkanCaps.SubgroupSize;
             m_dynamicParameters.SubgroupSupportedStages = mapShaderStages(m_vulkanCaps.SubgroupSupportedStages);
-            m_dynamicParameters.SubgroupSupportedFeatures = mapSubgroupFeatures(m_vulkanCaps.SubgroupSupportedOperations);
+            m_dynamicParameters.SubgroupSupportedFeatures =
+                mapSubgroupFeatures(m_vulkanCaps.SubgroupSupportedOperations);
             m_dynamicParameters.SubgroupQuadOperationsInAllStages = m_vulkanCaps.SubgroupQuadOperationsInAllStages;
         } else {
             m_dynamicParameters.SubgroupSize = 0;
@@ -845,8 +826,7 @@ namespace MobileGL::MG_Backend::DirectVulkan {
         }
         if (m_dynamicParameters.MaxShaderStorageBlockSize != m_vulkanCaps.MaxShaderStorageBlockSize) {
             MGLOG_I("DirectVulkan: clamped GL_MAX_SHADER_STORAGE_BLOCK_SIZE from %zu to %zu",
-                    m_vulkanCaps.MaxShaderStorageBlockSize,
+                    m_vulkanCaps.MaxShaderStorageBlockSize, m_dynamicParameters.MaxShaderStorageBlockSize);
-                    m_dynamicParameters.MaxShaderStorageBlockSize);
         }
         switch (m_vulkanCaps.VendorId) {
         case 0x5143u: // VK_VENDOR_ID: Qualcomm

MobileGL/MG_Backend/DirectVulkan/DirectVulkan.cpp

@@ -440,6 +440,20 @@ namespace MobileGL::MG_Backend::DirectVulkan {
         pVulkanRenderer->ClearNamedFramebufferfv(framebuffer, buffer, drawbuffer, value);
     }
 
+    void ClearNamedFramebufferiv(const SharedPtr<MG_State::GLState::FramebufferObject>& framebuffer, GLenum buffer,
+                                 GLint drawbuffer, const GLint* value) {
+        MOBILEGL_ASSERT(pVulkanRenderer, "DirectVulkan::ClearNamedFramebufferiv called with null VulkanRenderer");
+        MOBILEGL_ASSERT(MG_State::pGLContext, "DirectVulkan::ClearNamedFramebufferiv called with null GL context");
+        pVulkanRenderer->ClearNamedFramebufferiv(framebuffer, buffer, drawbuffer, value);
+    }
+
+    void ClearNamedFramebufferuiv(const SharedPtr<MG_State::GLState::FramebufferObject>& framebuffer, GLenum buffer,
+                                  GLint drawbuffer, const GLuint* value) {
+        MOBILEGL_ASSERT(pVulkanRenderer, "DirectVulkan::ClearNamedFramebufferuiv called with null VulkanRenderer");
+        MOBILEGL_ASSERT(MG_State::pGLContext, "DirectVulkan::ClearNamedFramebufferuiv called with null GL context");
+        pVulkanRenderer->ClearNamedFramebufferuiv(framebuffer, buffer, drawbuffer, value);
+    }
+
     void ClearNamedFramebufferfi(const SharedPtr<MG_State::GLState::FramebufferObject>& framebuffer, GLenum buffer,
                                  GLint drawbuffer, GLfloat depth, GLint stencil) {
         MOBILEGL_ASSERT(pVulkanRenderer, "DirectVulkan::ClearNamedFramebufferfi called with null VulkanRenderer");

MobileGL/MG_Backend/DirectVulkan/DirectVulkan.h

@@ -35,6 +35,10 @@ namespace MobileGL::MG_Backend::DirectVulkan {
     void ClearBufferiv(GLenum buffer, GLint drawbuffer, const GLint* value);
     void ClearNamedFramebufferfv(const SharedPtr<MG_State::GLState::FramebufferObject>& framebuffer, GLenum buffer,
                                  GLint drawbuffer, const GLfloat* value);
+    void ClearNamedFramebufferiv(const SharedPtr<MG_State::GLState::FramebufferObject>& framebuffer, GLenum buffer,
+                                 GLint drawbuffer, const GLint* value);
+    void ClearNamedFramebufferuiv(const SharedPtr<MG_State::GLState::FramebufferObject>& framebuffer, GLenum buffer,
+                                  GLint drawbuffer, const GLuint* value);
     void ClearNamedFramebufferfi(const SharedPtr<MG_State::GLState::FramebufferObject>& framebuffer, GLenum buffer,
                                  GLint drawbuffer, GLfloat depth, GLint stencil);
     void Clear(GLbitfield mask);

MobileGL/MG_Backend/DirectVulkan/Renderer/VkClearManager.cpp

@@ -8,15 +8,75 @@
 
 #include "VkClearManager.h"
 
+#include "MG_State/GLState/Core.h"
 #include "MG_Util/Converters/MGToStr/FramebufferEnumConverter.h"
 #include "MG_Util/Converters/MGToStr/TextureEnumConverter.h"
 
+#include <algorithm>
+#include <cmath>
+
 namespace MobileGL::MG_Backend::DirectVulkan {
     static Bool IsCubeMapFaceUploadTarget(TextureUploadTarget target) {
         return target >= TextureUploadTarget::CubeMapPositiveX &&
                target <= TextureUploadTarget::CubeMapNegativeZ;
     }
 
+    VkClearColorValue MakeVkClearColorValue(const ClearAttachmentPayload& payload, Bool formatLacksAlpha) {
+        VkClearColorValue clearValue{};
+        switch (payload.colorEncoding) {
+        case ClearColorEncoding::Int:
+            clearValue.int32[0] = payload.colorInt.x();
+            clearValue.int32[1] = payload.colorInt.y();
+            clearValue.int32[2] = payload.colorInt.z();
+            clearValue.int32[3] = formatLacksAlpha ? 1 : payload.colorInt.w();
+            break;
+        case ClearColorEncoding::Uint:
+            clearValue.uint32[0] = payload.colorUint.x();
+            clearValue.uint32[1] = payload.colorUint.y();
+            clearValue.uint32[2] = payload.colorUint.z();
+            clearValue.uint32[3] = formatLacksAlpha ? 1u : payload.colorUint.w();
+            break;
+        case ClearColorEncoding::Float:
+            clearValue.float32[0] = payload.color.x();
+            clearValue.float32[1] = payload.color.y();
+            clearValue.float32[2] = payload.color.z();
+            clearValue.float32[3] = formatLacksAlpha ? 1.0f : payload.color.w();
+            break;
+        }
+        return clearValue;
+    }
+
+    void PreCompensateSrgbClearColor(ClearAttachmentPayload& payload, VkFormat destinationFormat) {
+        if (payload.colorEncoding != ClearColorEncoding::Float) return;
+        // With GL_FRAMEBUFFER_SRGB enabled GL performs the encoding itself, so the driver doing it
+        // is exactly right and there is nothing to undo.
+        if (MG_State::pGLContext->IsCapabilityEnabled(MobileGL::CapabilityInput::FramebufferSrgb)) return;
+        if (ResolveSrgbAttachmentWriteFormat(destinationFormat, false) == destinationFormat) return;
+
+        // sRGB -> linear (GL 4.6 core 8.24), applied to the colour channels only: alpha is stored
+        // linearly in an sRGB format and must pass through untouched.
+        const auto toLinear = [](Float encoded) {
+            const Float value = std::clamp(encoded, 0.0f, 1.0f);
+            return value <= 0.04045f ? value / 12.92f : std::pow((value + 0.055f) / 1.055f, 2.4f);
+        };
+        payload.color = FloatVec4(toLinear(payload.color.x()), toLinear(payload.color.y()),
+                                  toLinear(payload.color.z()), payload.color.w());
+    }
+
+    void ForceOpaqueClearAlpha(ClearAttachmentPayload& payload) {
+        switch (payload.colorEncoding) {
+        case ClearColorEncoding::Int:
+            payload.colorInt = IntVec4(payload.colorInt.x(), payload.colorInt.y(), payload.colorInt.z(), 1);
+            break;
+        case ClearColorEncoding::Uint:
+            payload.colorUint = UintVec4(payload.colorUint.x(), payload.colorUint.y(), payload.colorUint.z(), 1u);
+            break;
+        case ClearColorEncoding::Float:
+            payload.color = FloatVec4(payload.color.x(), payload.color.y(), payload.color.z(), 1.0f);
+            break;
+        }
+    }
+
     static Bool PendingClearMatchesTextureIdentity(const PendingClearKey& key, const TextureIdentity& identity) {
         return key.texture == identity.texture && key.textureLifetimeId == identity.lifetimeId;
     }

MobileGL/MG_Backend/DirectVulkan/Renderer/VkClearManager.h

@@ -24,13 +24,41 @@ namespace MobileGL::MG_Backend::DirectVulkan {
         Uint32 stencil{};
     };
 
+    // A colour clear reaches us from one of glClear/ClearBufferfv, ClearBufferiv or
+    // ClearBufferuiv, and Vulkan reads VkClearColorValue's union according to the destination
+    // image's format rather than converting between the members - a float written where an
+    // integer format is expected is reinterpreted bit for bit, not rounded. Remember which entry
+    // point supplied the value so the member written when the clear is materialized matches.
+    enum class ClearColorEncoding : Uint8 { Float, Int, Uint };
+
     struct ClearAttachmentPayload {
         GLbitfield mask = 0;
         FloatVec4 color = FloatVec4(0.0f, 0.0f, 0.0f, 0.0f);
+        ClearColorEncoding colorEncoding = ClearColorEncoding::Float;
+        IntVec4 colorInt = IntVec4(0, 0, 0, 0);
+        UintVec4 colorUint = UintVec4(0u, 0u, 0u, 0u);
         Float depth = 1.0f;
         Uint32 stencil = 0;
     };
 
+    // Builds the clear value for `payload` in the union member its encoding calls for.
+    // `formatLacksAlpha` applies GL's rule that a format without an alpha channel reads as one,
+    // expressed in whichever type matches (GL 4.6 core 15.2.3).
+    VkClearColorValue MakeVkClearColorValue(const ClearAttachmentPayload& payload, Bool formatLacksAlpha);
+
+    // Applies that same rule in place, for the paths that have to bake it into the payload before
+    // the destination is known.
+    void ForceOpaqueClearAlpha(ClearAttachmentPayload& payload);
+
+    // vkCmdClearColorImage names the image, so the driver applies the destination format's transfer
+    // function to whatever value it is handed. Every other write path in this backend goes through
+    // the UNORM twin view while GL_FRAMEBUFFER_SRGB is off (ResolveSrgbAttachmentWriteFormat) and
+    // therefore stores the raw value GL asked for. Rewrites `payload` to the linear colour whose
+    // encoding is that raw value, so a direct image clear of an sRGB destination agrees with them.
+    // A no-op for every other format, for integer clear encodings, and when GL is doing the
+    // encoding itself.
+    void PreCompensateSrgbClearColor(ClearAttachmentPayload& payload, VkFormat destinationFormat);
+
     struct PendingClearKey {
         MG_State::GLState::ITextureObject* texture = nullptr;
         Uint64 textureLifetimeId = 0;

MobileGL/MG_Backend/DirectVulkan/Renderer/VkRenderPassManager.cpp

@@ -50,7 +50,11 @@ namespace MobileGL::MG_Backend::DirectVulkan {
         }
     }
 
-    static Float ResolveColorClearAlpha(const MG_State::GLState::ITextureObject* texture, Float requestedAlpha) {
+    static Bool ColorFormatLacksAlpha(const MG_State::GLState::ITextureObject* texture) {
+        return texture != nullptr && MG_Util::GetBaseInternalFormatComponentCount(texture->GetFormat()) == 3;
+    }
+
+    [[maybe_unused]] static Float ResolveColorClearAlpha(const MG_State::GLState::ITextureObject* texture, Float requestedAlpha) {
         if (texture != nullptr && MG_Util::GetBaseInternalFormatComponentCount(texture->GetFormat()) == 3) {
             return 1.0f;
         }
@@ -526,7 +530,13 @@ namespace MobileGL::MG_Backend::DirectVulkan {
         pending.renderbuffer = renderbuffer;
         pending.payload.mask |= clearPayload.mask;
         if ((clearPayload.mask & GL_COLOR_BUFFER_BIT) != 0) {
+            // The whole colour description, not just the float vector: an integer clear keeps its
+            // value in colorInt/colorUint, and dropping the encoding here would leave the pending
+            // clear reading as an all-zero float one.
             pending.payload.color = clearPayload.color;
+            pending.payload.colorEncoding = clearPayload.colorEncoding;
+            pending.payload.colorInt = clearPayload.colorInt;
+            pending.payload.colorUint = clearPayload.colorUint;
         }
         if ((clearPayload.mask & GL_DEPTH_BUFFER_BIT) != 0) {
             pending.payload.depth = clearPayload.depth;
@@ -924,9 +934,7 @@ namespace MobileGL::MG_Backend::DirectVulkan {
                     if (rbHasClear &&
                         MG_Util::GetBaseInternalFormatComponentCount(renderbuffer->GetInternalFormat()) == 3) {
                         // RGB renderbuffers are backed by an RGBA image; the missing alpha reads as 1.
-                        rbClearPayload.color =
+                        ForceOpaqueClearAlpha(rbClearPayload);
-                            FloatVec4(rbClearPayload.color.x(), rbClearPayload.color.y(),
-                                      rbClearPayload.color.z(), 1.0f);
                     }
 
                     const VkImageLayout trackedRbLayout = rbResource->layout;
@@ -1496,12 +1504,8 @@ namespace MobileGL::MG_Backend::DirectVulkan {
                 }
             }
             if ((clearPayload.mask & GL_COLOR_BUFFER_BIT) != 0) {
-                clearValues[pending.attachmentIndex].color = {
+                clearValues[pending.attachmentIndex].color =
-                    clearPayload.color.x(),
+                    MakeVkClearColorValue(clearPayload, ColorFormatLacksAlpha(liveTexture.get()));
-                    clearPayload.color.y(),
-                    clearPayload.color.z(),
-                    ResolveColorClearAlpha(liveTexture.get(), clearPayload.color.w())
-                };
             }
             if ((clearPayload.mask & GL_DEPTH_BUFFER_BIT) != 0) {
                 clearValues[pending.attachmentIndex].depthStencil.depth = clearPayload.depth;

MobileGL/MG_Backend/DirectVulkan/Renderer/VkSamplerManager.cpp

@@ -164,7 +164,7 @@ namespace MobileGL::MG_Backend::DirectVulkan {
         XXHASH_VERIFY(XXH64_update(m_hashState, &maxAnisotropy, sizeof(maxAnisotropy)));
         const auto compareMode = sampler.GetCompareMode();
         XXHASH_VERIFY(XXH64_update(m_hashState, &compareMode, sizeof(compareMode)));
-        const auto compareFunc = ResolveCompareFunc(sampler, texture);
+        const auto compareFunc = sampler.GetSamplerCompareFunc();
         XXHASH_VERIFY(XXH64_update(m_hashState, &compareFunc, sizeof(compareFunc)));
         const auto borderColor = ResolveVkBorderColor(sampler, texture);
         XXHASH_VERIFY(XXH64_update(m_hashState, &borderColor, sizeof(borderColor)));
@@ -207,7 +207,7 @@ namespace MobileGL::MG_Backend::DirectVulkan {
         samplerInfo.anisotropyEnable = maxAnisotropy > 1.0f ? VK_TRUE : VK_FALSE;
         samplerInfo.maxAnisotropy = maxAnisotropy;
         samplerInfo.compareEnable = sampler.GetCompareMode() == SamplerCompareMode::CompareToTexture ? VK_TRUE : VK_FALSE;
-        samplerInfo.compareOp = ToVkCompareOp(ResolveCompareFunc(sampler, texture));
+        samplerInfo.compareOp = ToVkCompareOp(sampler.GetSamplerCompareFunc());
         // Must match BuildSamplerKey's resolution exactly.
         samplerInfo.maxLod = ResolveSingleLevelMaxLod(sampler, singleLevelView);
         samplerInfo.minLod = ResolveEffectiveMinLod(sampler, samplerInfo.maxLod);
@@ -281,24 +281,15 @@ namespace MobileGL::MG_Backend::DirectVulkan {
         }
     }
 
-    SamplerCompareFunc VkSamplerManager::ResolveCompareFunc(const MG_State::GLState::SamplerObject& sampler,
-                                                            const MG_State::GLState::ITextureObject& texture) {
-        const auto compareFunc = sampler.GetSamplerCompareFunc();
-        if (sampler.GetCompareMode() == SamplerCompareMode::CompareToTexture &&
-            IsDepthTextureFormat(texture.GetFormat()) && compareFunc == SamplerCompareFunc::Always) {
-            return SamplerCompareFunc::LessEqual;
-        }
-
-        return compareFunc;
-    }
-
     VkBorderColor VkSamplerManager::ResolveVkBorderColor(const MG_State::GLState::SamplerObject& sampler,
                                                          const MG_State::GLState::ITextureObject& texture) {
         if (!UsesBorderColor(sampler)) {
             return VK_BORDER_COLOR_FLOAT_TRANSPARENT_BLACK;
         }
 
-        const auto& borderColor = texture.GetBorderColor();
+        // Border colour is sampler state: a bound sampler object supplies its own, and a texture
+        // with none reaches the very same value through the sampler object it owns.
+        const auto& borderColor = sampler.GetBorderColor();
         const Bool isDepthTexture = IsDepthTextureFormat(texture.GetFormat());
 
         if (isDepthTexture) {

MobileGL/MG_Backend/DirectVulkan/Renderer/VkSamplerManager.h

@@ -69,8 +69,6 @@ private:
     static VkSamplerMipmapMode ToVkMipmapMode(SamplerMipmapMode mode);
     static VkSamplerAddressMode ToVkAddressMode(SamplerWrapMode mode);
     static VkCompareOp ToVkCompareOp(SamplerCompareFunc func);
-    static SamplerCompareFunc ResolveCompareFunc(const MG_State::GLState::SamplerObject& sampler,
-                                                 const MG_State::GLState::ITextureObject& texture);
     static VkBorderColor ResolveVkBorderColor(const MG_State::GLState::SamplerObject& sampler,
                                               const MG_State::GLState::ITextureObject& texture);
     // The anisotropy Vulkan will actually apply: 1.0 (i.e. disabled) unless the feature is on and

MobileGL/MG_Backend/DirectVulkan/Renderer/VkTextureManager.cpp

@@ -1497,15 +1497,17 @@ namespace MobileGL::MG_Backend::DirectVulkan {
             : (mipLevels > 1 ? std::max(mipLevels, ComputeFullMipLevelCount(texelSize)) : 1u);
         TextureShapeInfo shapeInfo{};
         const Bool supportedShape = TryResolveTextureShapeInfo(texture, uploadTarget, texelSize, shapeInfo);
-        MOBILEGL_ASSERT(supportedShape,
-                        "SyncTextureResource: unsupported uploadTarget=%s textureTarget=%s textureId=%d size=(%d,%d,%d) "
-                        "mipLevels=%u vkViewType=%d",
-                        MG_Util::ConvertTextureUploadTargetToString(uploadTarget).c_str(),
-                        MG_Util::ConvertTextureTargetToString(texture.GetTarget()).c_str(),
-                        texture.GetExternalIndex(), texelSize.x(), texelSize.y(), texelSize.z(), mipLevels,
-                        static_cast<Int>(MG_Util::ConvertTextureUploadTargetToVkEnum(uploadTarget)));
         if (!supportedShape) {
-            MGLOG_D("%s: not Texture2D, unsupported", __func__);
+            // A gap in this backend's coverage, not a broken invariant: the GL front end accepts
+            // targets this manager has no Vulkan image shape for yet (cube map arrays above all).
+            // Declining the sync leaves the texture unbacked - wrong, but recoverable - where an
+            // assertion would take the whole process down instead.
+            MGLOG_W("SyncTextureResource: unsupported uploadTarget=%s textureTarget=%s textureId=%d size=(%d,%d,%d) "
+                    "mipLevels=%u vkViewType=%d",
+                    MG_Util::ConvertTextureUploadTargetToString(uploadTarget).c_str(),
+                    MG_Util::ConvertTextureTargetToString(texture.GetTarget()).c_str(), texture.GetExternalIndex(),
+                    texelSize.x(), texelSize.y(), texelSize.z(), mipLevels,
+                    static_cast<Int>(MG_Util::ConvertTextureUploadTargetToVkEnum(uploadTarget)));
             return false;
         }
         VkSampleCountFlagBits resolvedSampleCount = VK_SAMPLE_COUNT_1_BIT;
@@ -1516,6 +1518,16 @@ namespace MobileGL::MG_Backend::DirectVulkan {
                     MG_Util::ConvertTextureUploadTargetToString(uploadTarget).c_str());
             return false;
         }
+        // glTexStorage*Multisample(samples = 1) is legal GL, but a one-sample image cannot back a
+        // sampler2DMS: VUID-RuntimeSpirv-samples-08726 forbids an OpTypeImage with MS = 1 from
+        // reading an image created with VK_SAMPLE_COUNT_1_BIT, and the fetch returns undefined data
+        // rather than an error. GL only promises "at least the requested number of samples", so
+        // giving a multisample texture two is both legal and the only way to keep the shader's view
+        // of it honest. GL_TEXTURE_SAMPLES still reports what the application asked for - that is
+        // read off the texture object, not off the image.
+        if (isMultisampleTexture && resolvedSampleCount == VK_SAMPLE_COUNT_1_BIT) {
+            resolvedSampleCount = VK_SAMPLE_COUNT_2_BIT;
+        }
 
         const VkImageAspectFlags aspect = GetAspectMaskForFormat(format);
         VkFormatProperties formatProperties{};
@@ -1591,7 +1603,11 @@ namespace MobileGL::MG_Backend::DirectVulkan {
                     }
                 }
                 if (rounded == 0) {
-                    for (Uint32 bit = static_cast<Uint32>(resolvedSampleCount) >> 1; bit != 0; bit >>= 1) {
+                    // Never land on one sample: that is the VUID-RuntimeSpirv-samples-08726
+                    // violation the floor above exists to avoid, and it would come back silently
+                    // for any format whose only supported count is 1.
+                    for (Uint32 bit = static_cast<Uint32>(resolvedSampleCount) >> 1;
+                         bit > static_cast<Uint32>(VK_SAMPLE_COUNT_1_BIT); bit >>= 1) {
                         if ((supported & bit) != 0) {
                             rounded = bit;
                             break;

MobileGL/MG_Backend/DirectVulkan/Renderer/VulkanRenderer.cpp

@@ -190,11 +190,11 @@ namespace MobileGL::MG_Backend::DirectVulkan {
         }
     }
 
-    static Float ResolveColorClearAlpha(const MG_State::GLState::ITextureObject* texture, Float requestedAlpha) {
+    // GL 4.6 core 15.2.3: a colour format with no alpha channel reads as if alpha were one.
-        if (texture != nullptr && MG_Util::GetBaseInternalFormatComponentCount(texture->GetFormat()) == 3) {
+    // The substitution has to happen in the clear value's own type, so this reports the condition
-            return 1.0f;
+    // and MakeVkClearColorValue applies it to whichever union member the encoding selects.
-        }
+    static Bool ColorFormatLacksAlpha(const MG_State::GLState::ITextureObject* texture) {
-        return requestedAlpha;
+        return texture != nullptr && MG_Util::GetBaseInternalFormatComponentCount(texture->GetFormat()) == 3;
     }
 
     static Bool IsQuarterTurnPreTransform(VkSurfaceTransformFlagBitsKHR preTransform) {
@@ -1235,10 +1235,20 @@ void main() {
 
         static Uint32 ResolveAttachmentBaseArrayLayer(const MG_State::GLState::FramebufferAttachmentObject& attachment) {
             const TextureUploadTarget uploadTarget = attachment.GetTextureUploadTarget();
-            if (!IsCubeMapFaceUploadTarget(uploadTarget)) {
+            if (IsCubeMapFaceUploadTarget(uploadTarget)) {
-                return 0;
+                return static_cast<Uint32>(uploadTarget) - static_cast<Uint32>(TextureUploadTarget::CubeMapPositiveX);
             }
-            return static_cast<Uint32>(uploadTarget) - static_cast<Uint32>(TextureUploadTarget::CubeMapPositiveX);
+            // Every other layered attachment names its layer directly. Returning 0 regardless made
+            // every blit, copy and ReadPixels against such an attachment read layer zero.
+            return static_cast<Uint32>(std::max(attachment.GetTextureLayer(), 0));
+        }
+
+        // A 3D image has arrayLayers == 1: its "layer" is a z slice, which has to travel as an
+        // image offset rather than a base array layer (VkBufferImageCopy requires baseArrayLayer 0
+        // for VK_IMAGE_TYPE_3D).
+        static Bool AttachmentIsDepthSlice(const MG_State::GLState::FramebufferAttachmentObject& attachment) {
+            return attachment.IsTexture() && attachment.GetTexture() &&
+                   attachment.GetTexture()->GetTarget() == TextureTarget::Texture3D;
         }
 
         enum class BlitSurfaceTransform : Uint32 {
@@ -1259,6 +1269,8 @@ void main() {
             Uint32 mipLevelCount = 1;
             Uint32 baseArrayLayer = 0;
             Uint32 layerCount = 1;
+            // z slice for a VK_IMAGE_TYPE_3D source; array attachments use baseArrayLayer instead.
+            Uint32 depthOffset = 0;
             const char* label = nullptr;
         };
 
@@ -1490,7 +1502,12 @@ void main() {
             outBinding.extent = {attachmentExtent.x(), attachmentExtent.y()};
             outBinding.mipLevel = static_cast<Uint32>(std::max(attachment.GetTextureLevel(), 0));
             outBinding.mipLevelCount = resource->mipLevels;
-            outBinding.baseArrayLayer = ResolveAttachmentBaseArrayLayer(attachment);
+            if (AttachmentIsDepthSlice(attachment)) {
+                outBinding.depthOffset = static_cast<Uint32>(std::max(attachment.GetTextureLayer(), 0));
+                outBinding.baseArrayLayer = 0;
+            } else {
+                outBinding.baseArrayLayer = ResolveAttachmentBaseArrayLayer(attachment);
+            }
             outBinding.layerCount = 1;
             return true;
         }
@@ -1601,7 +1618,12 @@ void main() {
             outBinding.extent = {attachmentExtent.x(), attachmentExtent.y()};
             outBinding.mipLevel = static_cast<Uint32>(std::max(attachment.GetTextureLevel(), 0));
             outBinding.mipLevelCount = resource->mipLevels;
-            outBinding.baseArrayLayer = ResolveAttachmentBaseArrayLayer(attachment);
+            if (AttachmentIsDepthSlice(attachment)) {
+                outBinding.depthOffset = static_cast<Uint32>(std::max(attachment.GetTextureLayer(), 0));
+                outBinding.baseArrayLayer = 0;
+            } else {
+                outBinding.baseArrayLayer = ResolveAttachmentBaseArrayLayer(attachment);
+            }
             outBinding.layerCount = 1;
             return true;
         }
@@ -1745,7 +1767,12 @@ void main() {
             outBinding.extent = {attachmentExtent.x(), attachmentExtent.y()};
             outBinding.mipLevel = static_cast<Uint32>(std::max(attachment.GetTextureLevel(), 0));
             outBinding.mipLevelCount = 1;
-            outBinding.baseArrayLayer = ResolveAttachmentBaseArrayLayer(attachment);
+            if (AttachmentIsDepthSlice(attachment)) {
+                outBinding.depthOffset = static_cast<Uint32>(std::max(attachment.GetTextureLayer(), 0));
+                outBinding.baseArrayLayer = 0;
+            } else {
+                outBinding.baseArrayLayer = ResolveAttachmentBaseArrayLayer(attachment);
+            }
             outBinding.layerCount = 1;
             return true;
         }
@@ -5401,9 +5428,11 @@ void main() {
                         VkClearAttachment clearAttachment{};
                         clearAttachment.aspectMask = VK_IMAGE_ASPECT_COLOR_BIT;
                         clearAttachment.colorAttachment = drawBufferIndex;
+                        // glClear only ever supplies float values (ClearFramebufferPayload has no
+                        // other form), so the float member is always the right one here.
                         clearAttachment.clearValue.color = {
                             payload.color.x(), payload.color.y(), payload.color.z(),
-                            ResolveColorClearAlpha(colorTexture, payload.color.w())
+                            ColorFormatLacksAlpha(colorTexture) ? 1.0f : payload.color.w()
                         };
                         clearAttachments[clearAttachmentCount++] = clearAttachment;
                     }
@@ -5680,10 +5709,8 @@ void main() {
             }
             clearAttachment.aspectMask = VK_IMAGE_ASPECT_COLOR_BIT;
             clearAttachment.colorAttachment = static_cast<Uint32>(drawbuffer);
-            clearAttachment.clearValue.color = {
+            clearAttachment.clearValue.color =
-                clearPayload.color.x(), clearPayload.color.y(), clearPayload.color.z(),
+                MakeVkClearColorValue(clearPayload, ColorFormatLacksAlpha(colorTexture));
-                ResolveColorClearAlpha(colorTexture, clearPayload.color.w())
-            };
         } else {
             VkImageAspectFlags aspects = 0;
             if ((clearPayload.mask & GL_DEPTH_BUFFER_BIT) != 0 && MG_State::pGLContext->GetDepthMask() &&
@@ -5771,6 +5798,46 @@ void main() {
         QueueClearBufferPayloadForFramebuffer(*framebuffer, buffer, drawbuffer, payload);
     }
 
+    // The integer clears carry the same payload as their target-based siblings; only the
+    // destination differs, so they queue against the named framebuffer rather than the bound one.
+    void VulkanRenderer::ClearNamedFramebufferiv(
+            const SharedPtr<MG_State::GLState::FramebufferObject>& framebuffer, GLenum buffer, GLint drawbuffer,
+            const GLint* value) {
+        if (!framebuffer || value == nullptr) {
+            return;
+        }
+        ClearAttachmentPayload payload{};
+        switch (buffer) {
+            case GL_COLOR:
+                payload.mask = GL_COLOR_BUFFER_BIT;
+                payload.colorEncoding = ClearColorEncoding::Int;
+                payload.colorInt = IntVec4(value[0], value[1], value[2], value[3]);
+                break;
+            case GL_STENCIL:
+                payload.mask = GL_STENCIL_BUFFER_BIT;
+                payload.stencil = static_cast<Uint32>(std::max(value[0], 0));
+                break;
+            default:
+                break;
+        }
+        QueueClearBufferPayloadForFramebuffer(*framebuffer, buffer, drawbuffer, payload);
+    }
+
+    void VulkanRenderer::ClearNamedFramebufferuiv(
+            const SharedPtr<MG_State::GLState::FramebufferObject>& framebuffer, GLenum buffer, GLint drawbuffer,
+            const GLuint* value) {
+        if (!framebuffer || value == nullptr) {
+            return;
+        }
+        ClearAttachmentPayload payload{};
+        if (buffer == GL_COLOR) {
+            payload.mask = GL_COLOR_BUFFER_BIT;
+            payload.colorEncoding = ClearColorEncoding::Uint;
+            payload.colorUint = UintVec4(value[0], value[1], value[2], value[3]);
+        }
+        QueueClearBufferPayloadForFramebuffer(*framebuffer, buffer, drawbuffer, payload);
+    }
+
     void VulkanRenderer::ClearNamedFramebufferfi(
             const SharedPtr<MG_State::GLState::FramebufferObject>& framebuffer, GLenum buffer, GLint drawbuffer,
             GLfloat depth, GLint stencil) {
@@ -5793,8 +5860,8 @@ void main() {
         switch (buffer) {
             case GL_COLOR:
                 payload.mask = GL_COLOR_BUFFER_BIT;
-                payload.color = FloatVec4(static_cast<Float>(value[0]), static_cast<Float>(value[1]),
+                payload.colorEncoding = ClearColorEncoding::Uint;
-                                          static_cast<Float>(value[2]), static_cast<Float>(value[3]));
+                payload.colorUint = UintVec4(value[0], value[1], value[2], value[3]);
                 break;
             case GL_STENCIL:
                 payload.mask = GL_STENCIL_BUFFER_BIT;
@@ -5814,8 +5881,8 @@ void main() {
         switch (buffer) {
             case GL_COLOR:
                 payload.mask = GL_COLOR_BUFFER_BIT;
-                payload.color = FloatVec4(static_cast<Float>(value[0]), static_cast<Float>(value[1]),
+                payload.colorEncoding = ClearColorEncoding::Int;
-                                          static_cast<Float>(value[2]), static_cast<Float>(value[3]));
+                payload.colorInt = IntVec4(value[0], value[1], value[2], value[3]);
                 break;
             case GL_STENCIL:
                 payload.mask = GL_STENCIL_BUFFER_BIT;
@@ -5874,14 +5941,12 @@ void main() {
             subresourceRange.baseArrayLayer = pendingClear.key.baseArrayLayer;
             subresourceRange.layerCount = pendingClear.key.layerCount;
 
-            const auto& clearPayload = pendingClear.payload;
+            auto clearPayload = pendingClear.payload;
             if ((resource->aspect & VK_IMAGE_ASPECT_COLOR_BIT) != 0) {
                 subresourceRange.aspectMask = VK_IMAGE_ASPECT_COLOR_BIT;
-                VkClearColorValue clearValue{};
+                PreCompensateSrgbClearColor(clearPayload, resource->format);
-                clearValue.float32[0] = clearPayload.color.x();
+                const VkClearColorValue clearValue =
-                clearValue.float32[1] = clearPayload.color.y();
+                    MakeVkClearColorValue(clearPayload, ColorFormatLacksAlpha(&texture));
-                clearValue.float32[2] = clearPayload.color.z();
-                clearValue.float32[3] = ResolveColorClearAlpha(&texture, clearPayload.color.w());
                 vkCmdClearColorImage(commandBuffer, resource->image, VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL,
                                      &clearValue, 1, &subresourceRange);
             } else {
@@ -5963,14 +6028,11 @@ void main() {
         VkImageLayout steadyLayout;
         if ((resource->aspect & VK_IMAGE_ASPECT_COLOR_BIT) != 0) {
             subresourceRange.aspectMask = VK_IMAGE_ASPECT_COLOR_BIT;
-            VkClearColorValue clearValue{};
+            PreCompensateSrgbClearColor(clearPayload, resource->format);
-            clearValue.float32[0] = clearPayload.color.x();
-            clearValue.float32[1] = clearPayload.color.y();
-            clearValue.float32[2] = clearPayload.color.z();
             // RGB renderbuffers are backed by an RGBA image; the missing alpha reads as 1.
-            clearValue.float32[3] =
+            const VkClearColorValue clearValue = MakeVkClearColorValue(
-                MG_Util::GetBaseInternalFormatComponentCount(renderbuffer->GetInternalFormat()) == 3 ?
+                clearPayload,
-                    1.0f : clearPayload.color.w();
+                MG_Util::GetBaseInternalFormatComponentCount(renderbuffer->GetInternalFormat()) == 3);
             vkCmdClearColorImage(commandBuffer, resource->image, VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL,
                                  &clearValue, 1, &subresourceRange);
             steadyLayout = VK_IMAGE_LAYOUT_COLOR_ATTACHMENT_OPTIMAL;
@@ -6267,17 +6329,20 @@ void main() {
 
             if (srcX1 < srcX0 || srcY1 < srcY0 || dstX1 < dstX0 || dstY1 < dstY0) {
                 MGLOG_E("BlitFramebuffer skipped: depth blits with flipped rectangles are not supported yet");
-                return;
+                continue;
             }
 
             const Int srcWidth = srcX1 - srcX0;
             const Int srcHeight = srcY1 - srcY0;
             const Int dstWidth = dstX1 - dstX0;
             const Int dstHeight = dstY1 - dstY0;
-            if (srcWidth <= 0 || srcHeight <= 0 || srcWidth != dstWidth || srcHeight != dstHeight) {
+            if (srcWidth <= 0 || srcHeight <= 0 || dstWidth <= 0 || dstHeight <= 0) {
-                MGLOG_E("BlitFramebuffer skipped: depth blits currently require matching source and destination extents");
+                MGLOG_E("BlitFramebuffer skipped: degenerate depth blit rectangle");
-                return;
+                continue;
             }
+            // A scaling depth blit is legal GL and vkCmdBlitImage scales natively; only a same-size
+            // pair can take the cheaper vkCmdCopyImage.
+            const Bool depthBlitScales = srcWidth != dstWidth || srcHeight != dstHeight;
 
             if (!readIsDefaultFbo) {
                 const auto sourceAttachmentType = ResolveFramebufferCopyAttachmentType(*readFbo, true, srcBinding.aspectMask);
@@ -6322,7 +6387,7 @@ void main() {
                 : *srcBinding.trackedLayout;
             if (srcOriginalLayout == VK_IMAGE_LAYOUT_UNDEFINED) {
                 MGLOG_E("BlitFramebuffer skipped: depth source image layout is undefined");
-                return;
+                continue;
             }
 
             const VkImageLayout dstOriginalLayout = drawIsDefaultFbo
@@ -6339,7 +6404,7 @@ void main() {
             if (srcBinding.format != dstBinding.format) {
                 if (readIsDefaultFbo || drawIsDefaultFbo) {
                     MGLOG_E("BlitFramebuffer skipped: cross-format depth/stencil blit with the default framebuffer");
-                    return;
+                    continue;
                 }
                 if (!BlitDepthAcrossFormats(frame, srcBinding.image, srcBinding.format, srcBinding.trackedLayout,
                                             srcBinding.mipLevel, srcBinding.baseArrayLayer, dstBinding.image,
@@ -6347,7 +6412,7 @@ void main() {
                                             dstBinding.baseArrayLayer, srcX0, srcY0, dstX0, dstY0, srcX1 - srcX0,
                                             srcY1 - srcY0, srcOriginalLayout, dstRestoreLayout,
                                             depthStencilAspect == VK_IMAGE_ASPECT_STENCIL_BIT)) {
-                    return;
+                    continue;
                 }
                 continue;
             }
@@ -6394,23 +6459,45 @@ void main() {
                 MOBILEGL_ASSERT(ok, "%s: failed to transition depth destination image", __func__);
             }
 
-            VkImageCopy copyRegion{};
+            if (depthBlitScales) {
-            copyRegion.srcSubresource.aspectMask = srcBinding.aspectMask;
+                // vkCmdCopyImage cannot resize; NEAREST is the only filter Vulkan allows for a
-            copyRegion.srcSubresource.mipLevel = srcBinding.mipLevel;
+                // depth/stencil blit anyway, and the GL front end already rejects the others.
-            copyRegion.srcSubresource.baseArrayLayer = srcBinding.baseArrayLayer;
+                VkImageBlit blitRegion{};
-            copyRegion.srcSubresource.layerCount = srcBinding.layerCount;
+                blitRegion.srcSubresource.aspectMask = srcBinding.aspectMask;
-            copyRegion.srcOffset = {srcX0, srcY0, 0};
+                blitRegion.srcSubresource.mipLevel = srcBinding.mipLevel;
-            copyRegion.dstSubresource.aspectMask = dstBinding.aspectMask;
+                blitRegion.srcSubresource.baseArrayLayer = srcBinding.baseArrayLayer;
-            copyRegion.dstSubresource.mipLevel = dstBinding.mipLevel;
+                blitRegion.srcSubresource.layerCount = srcBinding.layerCount;
-            copyRegion.dstSubresource.baseArrayLayer = dstBinding.baseArrayLayer;
+                blitRegion.srcOffsets[0] = {srcX0, srcY0, 0};
-            copyRegion.dstSubresource.layerCount = dstBinding.layerCount;
+                blitRegion.srcOffsets[1] = {srcX1, srcY1, 1};
-            copyRegion.dstOffset = {dstX0, dstY0, 0};
+                blitRegion.dstSubresource.aspectMask = dstBinding.aspectMask;
-            copyRegion.extent = {static_cast<Uint32>(srcWidth), static_cast<Uint32>(srcHeight), 1};
+                blitRegion.dstSubresource.mipLevel = dstBinding.mipLevel;
+                blitRegion.dstSubresource.baseArrayLayer = dstBinding.baseArrayLayer;
+                blitRegion.dstSubresource.layerCount = dstBinding.layerCount;
+                blitRegion.dstOffsets[0] = {dstX0, dstY0, 0};
+                blitRegion.dstOffsets[1] = {dstX1, dstY1, 1};
+                vkCmdBlitImage(frame.commandBuffer,
+                               srcBinding.image, VK_IMAGE_LAYOUT_TRANSFER_SRC_OPTIMAL,
+                               dstBinding.image, VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL,
+                               1, &blitRegion, VK_FILTER_NEAREST);
+            } else {
+                VkImageCopy copyRegion{};
+                copyRegion.srcSubresource.aspectMask = srcBinding.aspectMask;
+                copyRegion.srcSubresource.mipLevel = srcBinding.mipLevel;
+                copyRegion.srcSubresource.baseArrayLayer = srcBinding.baseArrayLayer;
+                copyRegion.srcSubresource.layerCount = srcBinding.layerCount;
+                copyRegion.srcOffset = {srcX0, srcY0, 0};
+                copyRegion.dstSubresource.aspectMask = dstBinding.aspectMask;
+                copyRegion.dstSubresource.mipLevel = dstBinding.mipLevel;
+                copyRegion.dstSubresource.baseArrayLayer = dstBinding.baseArrayLayer;
+                copyRegion.dstSubresource.layerCount = dstBinding.layerCount;
+                copyRegion.dstOffset = {dstX0, dstY0, 0};
+                copyRegion.extent = {static_cast<Uint32>(srcWidth), static_cast<Uint32>(srcHeight), 1};
 
-            vkCmdCopyImage(frame.commandBuffer,
+                vkCmdCopyImage(frame.commandBuffer,
-                           srcBinding.image, VK_IMAGE_LAYOUT_TRANSFER_SRC_OPTIMAL,
+                               srcBinding.image, VK_IMAGE_LAYOUT_TRANSFER_SRC_OPTIMAL,
-                           dstBinding.image, VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL,
+                               dstBinding.image, VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL,
-                           1, &copyRegion);
+                               1, &copyRegion);
+            }
 
             VkPipelineStageFlags srcRestoreStageMask = VK_PIPELINE_STAGE_TOP_OF_PIPE_BIT;
             VkAccessFlags srcRestoreAccessMask = 0;
@@ -7141,7 +7228,7 @@ void main() {
         copyRegion.imageSubresource.mipLevel = srcBinding.mipLevel;
         copyRegion.imageSubresource.baseArrayLayer = srcBinding.baseArrayLayer;
         copyRegion.imageSubresource.layerCount = 1;
-        copyRegion.imageOffset = {x, y, 0};
+        copyRegion.imageOffset = {x, y, static_cast<Int32>(srcBinding.depthOffset)};
         copyRegion.imageExtent = {static_cast<Uint32>(width), static_cast<Uint32>(height), 1};
         vkCmdCopyImageToBuffer(frame.commandBuffer, srcBinding.image, VK_IMAGE_LAYOUT_TRANSFER_SRC_OPTIMAL,
                                readback.GetHandle(), 1, &copyRegion);
@@ -7599,13 +7686,16 @@ void main() {
         switch (type) {
         case GL_FLOAT:
         case GL_UNSIGNED_INT:
+        case GL_INT:
         case GL_UNSIGNED_INT_24_8:
             dstPixelBytes = 4;
             break;
         case GL_UNSIGNED_SHORT:
+        case GL_SHORT:
             dstPixelBytes = 2;
             break;
         case GL_UNSIGNED_BYTE:
+        case GL_BYTE:
             dstPixelBytes = 1;
             break;
         case GL_FLOAT_32_UNSIGNED_INT_24_8_REV:
@@ -7616,29 +7706,40 @@ void main() {
             return;
         }
 
+        // GL 4.6 core 18.2.8: a GL_STENCIL_INDEX read reports the index itself, unconverted, in
+        // whatever width the client asked for. Only the packed types mix depth in. Deciding this
+        // once - rather than per type, where GL_FLOAT and GL_UNSIGNED_SHORT used to emit a depth
+        // value that is meaningless for a stencil-only image - is what makes the CTS's
+        // (GL_STENCIL_INDEX, GL_INT) read return 7 instead of nothing.
+        const Bool stencilOnly = format == GL_STENCIL_INDEX;
+
         Vector<Uint8> packed(pixelCount * dstPixelBytes);
         for (SizeT i = 0; i < pixelCount; ++i) {
             Uint8* dst = packed.data() + i * dstPixelBytes;
             switch (type) {
             case GL_FLOAT: {
-                const Float value = depthValueAt(i);
+                const Float value = stencilOnly ? static_cast<Float>(stencilSrc[i]) : depthValueAt(i);
                 Memcpy(dst, &value, sizeof(value));
                 break;
             }
-            case GL_UNSIGNED_SHORT: {
+            case GL_UNSIGNED_SHORT:
-                const Uint16 value =
+            case GL_SHORT: {
-                    static_cast<Uint16>(std::lround(static_cast<double>(depthValueAt(i)) * 65535.0));
+                const Uint16 value = stencilOnly
+                    ? static_cast<Uint16>(stencilSrc[i])
+                    : static_cast<Uint16>(std::lround(static_cast<double>(depthValueAt(i)) * 65535.0));
                 Memcpy(dst, &value, sizeof(value));
                 break;
             }
-            case GL_UNSIGNED_INT: {
+            case GL_UNSIGNED_INT:
-                const Uint32 value = format == GL_STENCIL_INDEX
+            case GL_INT: {
+                const Uint32 value = stencilOnly
                     ? stencilSrc[i]
                     : static_cast<Uint32>(static_cast<double>(depthValueAt(i)) * 4294967295.0);
                 Memcpy(dst, &value, sizeof(value));
                 break;
             }
-            case GL_UNSIGNED_BYTE: {
+            case GL_UNSIGNED_BYTE:
+            case GL_BYTE: {
                 dst[0] = stencilSrc[i];
                 break;
             }
@@ -7850,9 +7951,18 @@ void main() {
 
     void VulkanRenderer::GenerateMipmap(GLenum target) {
         const auto textureTarget = MG_Util::ConvertGLEnumToTextureTarget(target);
-        MOBILEGL_ASSERT(textureTarget == TextureTarget::Texture2D || textureTarget == TextureTarget::Texture2DArray ||
+        // The other mipmappable targets - 1D, 1D array, cube map array - are legal GL and the front
-                            textureTarget == TextureTarget::Texture3D || textureTarget == TextureTarget::TextureCubeMap,
+        // end lets them through, so reaching one here is a coverage gap in this backend, not a
-                        "GenerateMipmap currently only supports GL_TEXTURE_2D, GL_TEXTURE_2D_ARRAY, GL_TEXTURE_3D, and GL_TEXTURE_CUBE_MAP.");
+        // broken invariant. Declining leaves the mip chain unwritten; asserting took the process
+        // down with it.
+        if (textureTarget != TextureTarget::Texture2D && textureTarget != TextureTarget::Texture2DArray &&
+            textureTarget != TextureTarget::Texture3D && textureTarget != TextureTarget::TextureCubeMap &&
+            // A 1D texture needs nothing special: its storage extent is {width, 1, 1}, so the blit
+            // loop below already emits the y and z offsets of 0 and 1 that a 1D image requires.
+            textureTarget != TextureTarget::Texture1D) {
+            MGLOG_W("GenerateMipmap: unsupported target %s", MG_Util::ConvertTextureTargetToString(textureTarget).c_str());
+            return;
+        }
 
         auto& textureUnit = MG_State::pGLContext->GetTextureUnitObject(MG_State::pGLContext->GetActiveTextureUnit());
         auto texture = textureUnit.GetBindingSlot(textureTarget).GetBoundObject();
@@ -9191,10 +9301,17 @@ void main() {
         if (record.harvested) {
             return true;
         }
-        if (!m_timerQueryManager || !IsFrameSerialComplete(record.frameSerial)) {
+        if (!m_timerQueryManager) {
             return false;
         }
-        return m_timerQueryManager->TryHarvest(record);
+        // Ask the pool first. It polls with VK_QUERY_RESULT_WITH_AVAILABILITY_BIT and is the
+        // authority on whether the timestamp has landed; the frame serial is not, because it only
+        // advances at Present and neither completion notifier will mark the CURRENT serial done - so
+        // a timestamp written and fence-waited inside one GL frame could never be read back in it.
+        if (m_timerQueryManager->TryHarvest(record)) {
+            return true;
+        }
+        return IsFrameSerialComplete(record.frameSerial) && m_timerQueryManager->TryHarvest(record);
     }
 
     Bool VulkanRenderer::WaitForTimerQueryResult(VkTimerQueryManager::TimestampRecord& record) {
@@ -10622,12 +10739,8 @@ void main() {
                 // VkClearAttachment::colorAttachment indexes the subpass pColorAttachments (draw-buffer
                 // slot space, with UNUSED holes), not the compacted attachment descriptions.
                 clearAttachment.colorAttachment = pending.colorAttachmentSlot;
-                clearAttachment.clearValue.color = {
+                clearAttachment.clearValue.color =
-                        clearPayload.color.x(),
+                        MakeVkClearColorValue(clearPayload, ColorFormatLacksAlpha(liveTexture.get()));
-                        clearPayload.color.y(),
-                        clearPayload.color.z(),
-                        ResolveColorClearAlpha(liveTexture.get(), clearPayload.color.w())
-                };
             } else {
                 if ((clearPayload.mask & GL_DEPTH_BUFFER_BIT) != 0) {
                     clearAttachment.aspectMask |= VK_IMAGE_ASPECT_DEPTH_BIT;

MobileGL/MG_Backend/DirectVulkan/Renderer/VulkanRenderer.h

@@ -181,6 +181,10 @@ namespace MobileGL::MG_Backend::DirectVulkan {
         void ClearBufferiv(GLenum buffer, GLint drawbuffer, const GLint* value);
         void ClearNamedFramebufferfv(const SharedPtr<MG_State::GLState::FramebufferObject>& framebuffer,
                                      GLenum buffer, GLint drawbuffer, const GLfloat* value);
+        void ClearNamedFramebufferiv(const SharedPtr<MG_State::GLState::FramebufferObject>& framebuffer,
+                                     GLenum buffer, GLint drawbuffer, const GLint* value);
+        void ClearNamedFramebufferuiv(const SharedPtr<MG_State::GLState::FramebufferObject>& framebuffer,
+                                      GLenum buffer, GLint drawbuffer, const GLuint* value);
         void ClearNamedFramebufferfi(const SharedPtr<MG_State::GLState::FramebufferObject>& framebuffer,
                                      GLenum buffer, GLint drawbuffer, GLfloat depth, GLint stencil);
         void BlitFramebuffer(GLint srcX0, GLint srcY0, GLint srcX1, GLint srcY1,

MobileGL/MG_Impl/GLImpl/Buffer/GL_Buffer.cpp

@@ -8,6 +8,9 @@
 
 #include "GL_Buffer.h"
 #include "Validators.h"
+#include "../Texture/GL_Texture.h"
+#include <MG_Util/Converters/GLToMG/TextureEnumConverter.h>
+#include <MG_Util/Metrics/TextureMetrics.h>
 #include <Config.h>
 #include <MG_State/GLState/Core.h>
 #include <MG_State/GLState/ErrorState/Error.h>
@@ -38,6 +41,7 @@ namespace MobileGL::MG_Impl::GLImpl {
             GetNamedBufferParameteriv,
             GetNamedBufferParameteri64v,
             GetNamedBufferPointerv,
+            GetNamedBufferSubData,
         };
 
         const char* GetBufferOpName(BufferOp op) {
@@ -76,6 +80,8 @@ namespace MobileGL::MG_Impl::GLImpl {
                 return "UnmapNamedBuffer";
             case BufferOp::FlushMappedNamedBufferRange:
                 return "FlushMappedNamedBufferRange";
+            case BufferOp::GetNamedBufferSubData:
+                return "GetNamedBufferSubData";
             case BufferOp::GetNamedBufferParameteriv:
                 return "GetNamedBufferParameteriv";
             case BufferOp::GetNamedBufferParameteri64v:
@@ -89,25 +95,64 @@ namespace MobileGL::MG_Impl::GLImpl {
 
         SharedPtr<MG_State::GLState::BufferObject> GetNamedBufferObject(GLuint buffer, BufferOp op);
 
+        // The size of one cleared element, which is what offset and size must be multiples of
+        // (GL 4.6 core 6.3). `internalformat` is restricted to the buffer-texture format table, and
+        // `format`/`type` describe the client-side pattern, so both are validated here and the
+        // caller only has to know how wide an element is.
         SizeT GetClearPatternSize(GLenum internalformat, GLenum format, GLenum type, BufferOp op) {
-            if (format != GL_RED_INTEGER) {
+            if (!IsBufferTextureInternalFormat(internalformat)) {
                 MG_State::pGLContext->RecordError(
                     ErrorCode::InvalidEnum,
-                    MakeUnique<GenericErrorInfo>("MG_Impl/GLImpl", GetBufferOpName(op),
+                    MakeUnique<GenericErrorInfo>(
-                                                 "Only GL_RED_INTEGER buffer clears are currently supported."));
+                        "MG_Impl/GLImpl", GetBufferOpName(op),
+                        std::format("internalformat 0x{:X} is not one of the sized formats a buffer clear accepts.",
+                                    internalformat)));
                 return 0;
             }
 
-            if (internalformat == GL_R8UI && type == GL_UNSIGNED_BYTE) return sizeof(GLubyte);
+            // Unlike internalformat, a bad format or type here is INVALID_VALUE rather than
-            if (internalformat == GL_R32UI && type == GL_UNSIGNED_INT) return sizeof(GLuint);
+            // INVALID_ENUM (GL 4.6 core 6.3) - the odd one out among the enum arguments.
+            const TextureInputFormat inputFormat = MG_Util::ConvertGLEnumToTextureInputFormat(format);
+            if (inputFormat == TextureInputFormat::Unknown) {
+                MG_State::pGLContext->RecordError(
+                    ErrorCode::InvalidValue,
+                    MakeUnique<GenericErrorInfo>("MG_Impl/GLImpl", GetBufferOpName(op),
+                                                 std::format("format 0x{:X} is not a pixel format.", format)));
+                return 0;
+            }
 
-            MG_State::pGLContext->RecordError(
+            const TexturePixelDataType pixelType = MG_Util::ConvertGLEnumToTexturePixelDataType(type);
-                ErrorCode::InvalidEnum,
+            if (pixelType == TexturePixelDataType::Unknown) {
-                MakeUnique<GenericErrorInfo>("MG_Impl/GLImpl", GetBufferOpName(op),
+                MG_State::pGLContext->RecordError(
-                                             std::format("Unsupported clear format tuple: internalformat=0x{:X}, "
+                    ErrorCode::InvalidValue,
-                                                         "format=0x{:X}, type=0x{:X}",
+                    MakeUnique<GenericErrorInfo>("MG_Impl/GLImpl", GetBufferOpName(op),
-                                                         internalformat, format, type)));
+                                                 std::format("type 0x{:X} is not a pixel type.", type)));
-            return 0;
+                return 0;
+            }
+
+            const TextureInternalFormat internal =
+                MG_Util::ConvertGLEnumToTextureInternalFormat(internalformat);
+            const SizeT elementSize = MG_Util::GetSizedInternalFormatSizeInBytes(internal);
+            if (elementSize == 0) {
+                MG_State::pGLContext->RecordError(
+                    ErrorCode::InvalidEnum,
+                    MakeUnique<GenericErrorInfo>("MG_Impl/GLImpl", GetBufferOpName(op),
+                                                 std::format("internalformat 0x{:X} has no known element size.",
+                                                             internalformat)));
+                return 0;
+            }
+
+            // The pattern is replicated verbatim, which is only the whole story while the client
+            // layout already matches the internal format - the case every entry point in practice
+            // uses, and the only one the conversion machinery here can express. Say so rather than
+            // quietly writing a differently-sized pattern.
+            const SizeT sourceSize = MG_Util::GetInputBytesPerPixel(inputFormat, pixelType);
+            if (sourceSize != elementSize) {
+                MGLOG_W("%s: clear pattern is %zu bytes but internalformat 0x%X stores %zu; "
+                        "converting between them is not implemented",
+                        GetBufferOpName(op), sourceSize, internalformat, elementSize);
+            }
+            return elementSize;
         }
 
         Bool ValidateBufferClearRange(const SharedPtr<MG_State::GLState::BufferObject>& bufferObject, GLintptr offset,
@@ -891,6 +936,44 @@ namespace MobileGL::MG_Impl::GLImpl {
         bufferObject->DownloadSubData(data, static_cast<SizeT>(offset), static_cast<SizeT>(size));
     }
 
+    void GetNamedBufferSubData_State(GLuint buffer, GLintptr offset, GLsizeiptr size, void* data) {
+        if (!data) {
+            // Match GetBufferSubData_State: a null pointer is a caller bug, not a GL-specified error.
+            return;
+        }
+
+        if (size < 0 || offset < 0) {
+            MG_State::pGLContext->RecordError(
+                ErrorCode::InvalidValue,
+                MakeUnique<GenericErrorInfo>("MG_Impl/GLImpl", "GetNamedBufferSubData_State",
+                                             "Offset and size must be non-negative."));
+            return;
+        }
+
+        auto bufferObject = GetNamedBufferObject(buffer, BufferOp::GetNamedBufferSubData);
+        if (!bufferObject) return;
+
+        if (static_cast<SizeT>(offset) + static_cast<SizeT>(size) > bufferObject->GetSize()) {
+            MG_State::pGLContext->RecordError(
+                ErrorCode::InvalidValue,
+                MakeUnique<GenericErrorInfo>("MG_Impl/GLImpl", "GetNamedBufferSubData_State",
+                                             "Offset and size exceed buffer size."));
+            return;
+        }
+
+        if (bufferObject->IsMapped() &&
+            !(bufferObject->GetMappingAccess() & BufferMappingAccessBit::Persistent)) {
+            MG_State::pGLContext->RecordError(
+                ErrorCode::InvalidOperation,
+                MakeUnique<GenericErrorInfo>("MG_Impl/GLImpl", "GetNamedBufferSubData_State",
+                                             "Cannot read from a buffer object mapped without GL_MAP_PERSISTENT_BIT."));
+            return;
+        }
+
+        bufferObject->SyncGpuWrites();
+        bufferObject->DownloadSubData(data, static_cast<SizeT>(offset), static_cast<SizeT>(size));
+    }
+
     void BufferData_State(GLenum target, GLsizeiptr size, const void* data, GLenum usage) {
         MGLOG_D("%s: %s, size = %d, data = %p, usage = %s", __func__, MG_Util::ConvertGLEnumToString(target).c_str(),
                 size, data, MG_Util::ConvertGLEnumToString(usage).c_str());
@@ -1553,6 +1636,10 @@ namespace MobileGL::MG_Impl::GLImpl {
         BufferSubData_State(target, offset, size, data);
     }
 
+    void GetNamedBufferSubData(GLuint buffer, GLintptr offset, GLsizeiptr size, void* data) {
+        GetNamedBufferSubData_State(buffer, offset, size, data);
+    }
+
     void GetBufferSubData(GLenum target, GLintptr offset, GLsizeiptr size, void* data) {
         GetBufferSubData_State(target, offset, size, data);
     }

MobileGL/MG_Impl/GLImpl/Buffer/GL_Buffer.h

@@ -41,6 +41,7 @@ namespace MobileGL::MG_Impl::GLImpl {
                            GLsizeiptr size);
     void BufferSubData(GLenum target, GLintptr offset, GLsizeiptr size, const void* data);
     void GetBufferSubData(GLenum target, GLintptr offset, GLsizeiptr size, void* data);
+    void GetNamedBufferSubData(GLuint buffer, GLintptr offset, GLsizeiptr size, void* data);
     void BufferData(GLenum target, GLsizeiptr size, const void* data, GLenum usage);
     void BindBuffer(GLenum target, GLuint buffer);
     void GenBuffers(GLsizei n, GLuint* buffers);

MobileGL/MG_Impl/GLImpl/Buffer/Validators.cpp

@@ -112,14 +112,10 @@ namespace MobileGL::MG_Impl::GLImpl::BufferImpl {
     }
 
     Bool ValidateBufferMappingAccess(Flags<BufferMappingAccessBit> accessBits) {
-        if (accessBits == BufferMappingAccessBit::Null) {
+        // An empty mask is a legal value for a bitfield - it just fails the rule that a mapping
-            MG_State::pGLContext->RecordError(ErrorCode::InvalidEnum,
+        // must ask for read or write access, which is INVALID_OPERATION and belongs to the callers
-                                              MakeUnique<GenericErrorInfo>("MG_Impl/GLImpl/BufferImpl",
+        // (both of them check it immediately after this). Rejecting it here as INVALID_ENUM
-                                                                           "ValidateBufferMappingAccess",
+        // reported the wrong error and hid theirs.
-                                                                           "Access bits cannot be null."));
-            return false;
-        }
-
         const auto validBits = BufferMappingAccessBit::Read | BufferMappingAccessBit::Write |
                                BufferMappingAccessBit::InvalidateRange | BufferMappingAccessBit::InvalidateBuffer |
                                BufferMappingAccessBit::FlushExplicit | BufferMappingAccessBit::Unsynchronized |

MobileGL/MG_Impl/GLImpl/Drawing/GL_Drawing.cpp

@@ -877,6 +877,172 @@ namespace MobileGL::MG_Impl::GLImpl {
         Memcpy(ids, names.data(), static_cast<SizeT>(n) * sizeof(GLuint));
     }
 
+    void CreateTransformFeedbacks(GLsizei n, GLuint* ids) {
+        if (n < 0) {
+            MG_State::pGLContext->RecordError(
+                ErrorCode::InvalidValue,
+                MakeUnique<GenericErrorInfo>("MG_Impl/GLImpl", __func__, "n must be non-negative."));
+            return;
+        }
+        if (n == 0 || ids == nullptr) return;
+        Vector<Uint> names;
+        MG_State::pGLContext->GenTransformFeedbackNames(static_cast<Uint>(n), names);
+        // Unlike glGenTransformFeedbacks, the names are objects immediately: there is no bind step
+        // to create them from (GL 4.6 core 13.2.1).
+        for (const Uint name : names) {
+            MG_State::pGLContext->CreateTransformFeedbackObject(name);
+        }
+        Memcpy(ids, names.data(), static_cast<SizeT>(n) * sizeof(GLuint));
+    }
+
+    namespace {
+        // Shared front half of the by-name transform feedback entry points: the object has to exist
+        // (INVALID_OPERATION otherwise) before anything else about the call is looked at.
+        Bool ValidateNamedTransformFeedback(GLuint xfb, const char* functionName) {
+            if (!MG_State::pGLContext->IsTransformFeedbackObject(xfb)) {
+                MG_State::pGLContext->RecordError(
+                    ErrorCode::InvalidOperation,
+                    MakeUnique<GenericErrorInfo>("MG_Impl/GLImpl", functionName,
+                                                 std::to_string(xfb) + " is not a transform feedback object."));
+                return false;
+            }
+            return true;
+        }
+
+        Bool ValidateTransformFeedbackBufferIndex(GLuint index, const char* functionName) {
+            if (index >= MG_State::GLState::GLContext::MAX_TRANSFORM_FEEDBACK_BUFFERS) {
+                MG_State::pGLContext->RecordError(
+                    ErrorCode::InvalidValue,
+                    MakeUnique<GenericErrorInfo>("MG_Impl/GLImpl", functionName,
+                                                 "index exceeds GL_MAX_TRANSFORM_FEEDBACK_BUFFERS."));
+                return false;
+            }
+            return true;
+        }
+
+        // A capture binding may not be changed while the object is capturing (GL 4.6 core 13.2.2).
+        Bool ValidateNamedTransformFeedbackNotActive(GLuint xfb, const char* functionName) {
+            if (MG_State::pGLContext->IsNamedTransformFeedbackActive(xfb)) {
+                MG_State::pGLContext->RecordError(
+                    ErrorCode::InvalidOperation,
+                    MakeUnique<GenericErrorInfo>("MG_Impl/GLImpl", functionName,
+                                                 "The transform feedback object is capturing."));
+                return false;
+            }
+            return true;
+        }
+
+        SharedPtr<MG_State::GLState::BufferObject> ResolveTransformFeedbackBuffer(GLuint buffer,
+                                                                                 const char* functionName) {
+            if (buffer == 0) return nullptr;
+            if (!MG_State::pGLContext->ValidateBufferName(buffer)) {
+                MG_State::pGLContext->RecordError(
+                    ErrorCode::InvalidOperation,
+                    MakeUnique<GenericErrorInfo>("MG_Impl/GLImpl", functionName,
+                                                 std::to_string(buffer) + " is not a buffer object."));
+                return nullptr;
+            }
+            return MG_State::pGLContext->GetBufferObject(buffer);
+        }
+    } // namespace
+
+    void TransformFeedbackBufferBase(GLuint xfb, GLuint index, GLuint buffer) {
+        if (!ValidateNamedTransformFeedback(xfb, __func__)) return;
+        if (!ValidateTransformFeedbackBufferIndex(index, __func__)) return;
+        if (!ValidateNamedTransformFeedbackNotActive(xfb, __func__)) return;
+        if (buffer != 0 && !MG_State::pGLContext->ValidateBufferName(buffer)) {
+            MG_State::pGLContext->RecordError(
+                ErrorCode::InvalidOperation,
+                MakeUnique<GenericErrorInfo>("MG_Impl/GLImpl", __func__,
+                                             std::to_string(buffer) + " is not a buffer object."));
+            return;
+        }
+        MG_State::pGLContext->SetNamedTransformFeedbackBinding(xfb, index,
+                                                               ResolveTransformFeedbackBuffer(buffer, __func__), {},
+                                                               false);
+    }
+
+    void TransformFeedbackBufferRange(GLuint xfb, GLuint index, GLuint buffer, GLintptr offset, GLsizeiptr size) {
+        if (!ValidateNamedTransformFeedback(xfb, __func__)) return;
+        if (!ValidateTransformFeedbackBufferIndex(index, __func__)) return;
+        if (!ValidateNamedTransformFeedbackNotActive(xfb, __func__)) return;
+        if (offset < 0 || size <= 0 || (offset % 4) != 0 || (size % 4) != 0) {
+            MG_State::pGLContext->RecordError(
+                ErrorCode::InvalidValue,
+                MakeUnique<GenericErrorInfo>("MG_Impl/GLImpl", __func__,
+                                             "offset and size must be non-negative multiples of 4."));
+            return;
+        }
+        if (buffer != 0 && !MG_State::pGLContext->ValidateBufferName(buffer)) {
+            MG_State::pGLContext->RecordError(
+                ErrorCode::InvalidOperation,
+                MakeUnique<GenericErrorInfo>("MG_Impl/GLImpl", __func__,
+                                             std::to_string(buffer) + " is not a buffer object."));
+            return;
+        }
+        auto bufferObject = ResolveTransformFeedbackBuffer(buffer, __func__);
+        const Range1D range{static_cast<SizeT>(offset), static_cast<SizeT>(offset) + static_cast<SizeT>(size)};
+        MG_State::pGLContext->SetNamedTransformFeedbackBinding(xfb, index, bufferObject, range,
+                                                               bufferObject != nullptr);
+    }
+
+    void GetTransformFeedbackiv(GLuint xfb, GLenum pname, GLint* param) {
+        if (!ValidateNamedTransformFeedback(xfb, __func__)) return;
+        if (!param) return;
+        switch (pname) {
+        case GL_TRANSFORM_FEEDBACK_ACTIVE:
+            *param = MG_State::pGLContext->IsNamedTransformFeedbackActive(xfb) ? GL_TRUE : GL_FALSE;
+            return;
+        case GL_TRANSFORM_FEEDBACK_PAUSED:
+            *param = MG_State::pGLContext->IsNamedTransformFeedbackPaused(xfb) ? GL_TRUE : GL_FALSE;
+            return;
+        default:
+            MG_State::pGLContext->RecordError(
+                ErrorCode::InvalidEnum,
+                MakeUnique<GenericErrorInfo>("MG_Impl/GLImpl", __func__,
+                                             "pname must be GL_TRANSFORM_FEEDBACK_ACTIVE or _PAUSED."));
+            return;
+        }
+    }
+
+    void GetTransformFeedbacki_v(GLuint xfb, GLenum pname, GLuint index, GLint* param) {
+        if (!ValidateNamedTransformFeedback(xfb, __func__)) return;
+        if (pname != GL_TRANSFORM_FEEDBACK_BUFFER_BINDING) {
+            MG_State::pGLContext->RecordError(
+                ErrorCode::InvalidEnum,
+                MakeUnique<GenericErrorInfo>("MG_Impl/GLImpl", __func__,
+                                             "pname must be GL_TRANSFORM_FEEDBACK_BUFFER_BINDING."));
+            return;
+        }
+        if (!ValidateTransformFeedbackBufferIndex(index, __func__)) return;
+        if (!param) return;
+        const auto binding = MG_State::pGLContext->GetNamedTransformFeedbackBinding(xfb, index);
+        *param = binding.Buffer ? static_cast<GLint>(binding.Buffer->GetExternalIndex()) : 0;
+    }
+
+    void GetTransformFeedbacki64_v(GLuint xfb, GLenum pname, GLuint index, GLint64* param) {
+        if (!ValidateNamedTransformFeedback(xfb, __func__)) return;
+        if (pname != GL_TRANSFORM_FEEDBACK_BUFFER_START && pname != GL_TRANSFORM_FEEDBACK_BUFFER_SIZE) {
+            MG_State::pGLContext->RecordError(
+                ErrorCode::InvalidEnum,
+                MakeUnique<GenericErrorInfo>("MG_Impl/GLImpl", __func__,
+                                             "pname must be GL_TRANSFORM_FEEDBACK_BUFFER_START or _SIZE."));
+            return;
+        }
+        if (!ValidateTransformFeedbackBufferIndex(index, __func__)) return;
+        if (!param) return;
+        const auto binding = MG_State::pGLContext->GetNamedTransformFeedbackBinding(xfb, index);
+        // glTransformFeedbackBufferBase leaves both at zero; only the range form sets them
+        // (GL 4.6 core table 23.48).
+        if (!binding.Buffer || !binding.HasExplicitRange) {
+            *param = 0;
+            return;
+        }
+        *param = (pname == GL_TRANSFORM_FEEDBACK_BUFFER_START)
+                     ? static_cast<GLint64>(binding.Range.start)
+                     : static_cast<GLint64>(binding.Range.end - binding.Range.start);
+    }
+
     void DeleteTransformFeedbacks(GLsizei n, const GLuint* ids) {
         if (n < 0) {
             MG_State::pGLContext->RecordError(

MobileGL/MG_Impl/GLImpl/Drawing/GL_Drawing.h

@@ -16,7 +16,13 @@ namespace MobileGL::MG_Impl::GLImpl {
     void PauseTransformFeedback(void);
     void ResumeTransformFeedback(void);
     void GenTransformFeedbacks(GLsizei n, GLuint* ids);
+    void CreateTransformFeedbacks(GLsizei n, GLuint* ids);
     void DeleteTransformFeedbacks(GLsizei n, const GLuint* ids);
+    void TransformFeedbackBufferBase(GLuint xfb, GLuint index, GLuint buffer);
+    void TransformFeedbackBufferRange(GLuint xfb, GLuint index, GLuint buffer, GLintptr offset, GLsizeiptr size);
+    void GetTransformFeedbackiv(GLuint xfb, GLenum pname, GLint* param);
+    void GetTransformFeedbacki_v(GLuint xfb, GLenum pname, GLuint index, GLint* param);
+    void GetTransformFeedbacki64_v(GLuint xfb, GLenum pname, GLuint index, GLint64* param);
     void BindTransformFeedback(GLenum target, GLuint id);
     GLboolean IsTransformFeedback(GLuint id);
     void DrawTransformFeedback(GLenum mode, GLuint id);

MobileGL/MG_Impl/GLImpl/Exporting/Definitions.cpp

@@ -15,6 +15,7 @@
 #include "../Texture/GL_Texture.h"
 #include "../Drawing/GL_Drawing.h"
 #include "../Program/GL_Program.h"
+#include "../Program/GL_ProgramPipeline.h"
 #include "../RenderState/GL_RenderState.h"
 #include "../Framebuffer/GL_Framebuffer.h"
 #include "../VertexArray/GL_VertexArray.h"
@@ -301,8 +302,8 @@ DECLARE_GL_FUNCTION_HEAD(void, ResumeTransformFeedback) DECLARE_GL_FUNCTION_END_
 DECLARE_GL_FUNCTION_HEAD(void, GetProgramBinary, GLuint program, GLsizei bufSize, GLsizei* length, GLenum* binaryFormat, void* binary) DECLARE_GL_FUNCTION_END_NO_RETURN(void, GetProgramBinary, program, bufSize, length, binaryFormat, binary)
 DECLARE_GL_FUNCTION_HEAD(void, ProgramBinary, GLuint program, GLenum binaryFormat, const void* binary, GLsizei length) DECLARE_GL_FUNCTION_END_NO_RETURN(void, ProgramBinary, program, binaryFormat, binary, length)
 DECLARE_GL_FUNCTION_HEAD(void, ProgramParameteri, GLuint program, GLenum pname, GLint value) DECLARE_GL_FUNCTION_END_NO_RETURN(void, ProgramParameteri, program, pname, value)
-DECLARE_GL_FUNCTION_STUB_HEAD(void, InvalidateFramebuffer, GLenum target, GLsizei numAttachments, const GLenum* attachments) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, InvalidateFramebuffer, target, numAttachments, attachments)
+DECLARE_GL_FUNCTION_HEAD(void, InvalidateFramebuffer, GLenum target, GLsizei numAttachments, const GLenum* attachments) DECLARE_GL_FUNCTION_END_NO_RETURN(void, InvalidateFramebuffer, target, numAttachments, attachments)
-DECLARE_GL_FUNCTION_STUB_HEAD(void, InvalidateSubFramebuffer, GLenum target, GLsizei numAttachments, const GLenum* attachments, GLint x, GLint y, GLsizei width, GLsizei height) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, InvalidateSubFramebuffer, target, numAttachments, attachments, x, y, width, height)
+DECLARE_GL_FUNCTION_HEAD(void, InvalidateSubFramebuffer, GLenum target, GLsizei numAttachments, const GLenum* attachments, GLint x, GLint y, GLsizei width, GLsizei height) DECLARE_GL_FUNCTION_END_NO_RETURN(void, InvalidateSubFramebuffer, target, numAttachments, attachments, x, y, width, height)
 DECLARE_GL_FUNCTION_HEAD(void, TexStorage2D, GLenum target, GLsizei levels, GLenum internalformat, GLsizei width, GLsizei height) DECLARE_GL_FUNCTION_END_NO_RETURN(void, TexStorage2D, target, levels, internalformat, width, height)
 DECLARE_GL_FUNCTION_HEAD(void, TexStorage3D, GLenum target, GLsizei levels, GLenum internalformat, GLsizei width, GLsizei height, GLsizei depth) DECLARE_GL_FUNCTION_END_NO_RETURN(void, TexStorage3D, target, levels, internalformat, width, height, depth)
 DECLARE_GL_FUNCTION_HEAD(void, GetInternalformativ, GLenum target, GLenum internalformat, GLenum pname, GLsizei bufSize, GLint* params) DECLARE_GL_FUNCTION_END_NO_RETURN(void, GetInternalformativ, target, internalformat, pname, bufSize, params)
@@ -310,21 +311,21 @@ DECLARE_GL_FUNCTION_HEAD(void, DispatchCompute, GLuint num_groups_x, GLuint num_
 DECLARE_GL_FUNCTION_HEAD(void, DispatchComputeIndirect, GLintptr indirect) DECLARE_GL_FUNCTION_END_NO_RETURN(void, DispatchComputeIndirect, indirect)
 DECLARE_GL_FUNCTION_HEAD(void, DrawArraysIndirect, GLenum mode, const void* indirect) DECLARE_GL_FUNCTION_END_NO_RETURN(void, DrawArraysIndirect, mode, indirect)
 DECLARE_GL_FUNCTION_HEAD(void, DrawElementsIndirect, GLenum mode, GLenum type, const void* indirect) DECLARE_GL_FUNCTION_END_NO_RETURN(void, DrawElementsIndirect, mode, type, indirect)
-DECLARE_GL_FUNCTION_STUB_HEAD(void, FramebufferParameteri, GLenum target, GLenum pname, GLint param) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, FramebufferParameteri, target, pname, param)
+DECLARE_GL_FUNCTION_HEAD(void, FramebufferParameteri, GLenum target, GLenum pname, GLint param) DECLARE_GL_FUNCTION_END_NO_RETURN(void, FramebufferParameteri, target, pname, param)
-DECLARE_GL_FUNCTION_STUB_HEAD(void, GetFramebufferParameteriv, GLenum target, GLenum pname, GLint* params) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, GetFramebufferParameteriv, target, pname, params)
+DECLARE_GL_FUNCTION_HEAD(void, GetFramebufferParameteriv, GLenum target, GLenum pname, GLint* params) DECLARE_GL_FUNCTION_END_NO_RETURN(void, GetFramebufferParameteriv, target, pname, params)
 DECLARE_GL_FUNCTION_HEAD(void, GetProgramInterfaceiv, GLuint program, GLenum programInterface, GLenum pname, GLint* params) DECLARE_GL_FUNCTION_END_NO_RETURN(void, GetProgramInterfaceiv, program, programInterface, pname, params)
 DECLARE_GL_FUNCTION_HEAD(GLuint, GetProgramResourceIndex, GLuint program, GLenum programInterface, const GLchar* name) DECLARE_GL_FUNCTION_END(GLuint, GetProgramResourceIndex, program, programInterface, name)
 DECLARE_GL_FUNCTION_HEAD(void, GetProgramResourceName, GLuint program, GLenum programInterface, GLuint index, GLsizei bufSize, GLsizei* length, GLchar* name) DECLARE_GL_FUNCTION_END_NO_RETURN(void, GetProgramResourceName, program, programInterface, index, bufSize, length, name)
 DECLARE_GL_FUNCTION_HEAD(void, GetProgramResourceiv, GLuint program, GLenum programInterface, GLuint index, GLsizei propCount, const GLenum* props, GLsizei bufSize, GLsizei* length, GLint* params) DECLARE_GL_FUNCTION_END_NO_RETURN(void, GetProgramResourceiv, program, programInterface, index, propCount, props, bufSize, length, params)
 DECLARE_GL_FUNCTION_HEAD(GLint, GetProgramResourceLocation, GLuint program, GLenum programInterface, const GLchar* name) DECLARE_GL_FUNCTION_END(GLint, GetProgramResourceLocation, program, programInterface, name)
-DECLARE_GL_FUNCTION_STUB_HEAD(void, UseProgramStages, GLuint pipeline, GLbitfield stages, GLuint program) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, UseProgramStages, pipeline, stages, program)
+DECLARE_GL_FUNCTION_HEAD(void, UseProgramStages, GLuint pipeline, GLbitfield stages, GLuint program) DECLARE_GL_FUNCTION_END_NO_RETURN(void, UseProgramStages, pipeline, stages, program)
-DECLARE_GL_FUNCTION_STUB_HEAD(void, ActiveShaderProgram, GLuint pipeline, GLuint program) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, ActiveShaderProgram, pipeline, program)
+DECLARE_GL_FUNCTION_HEAD(void, ActiveShaderProgram, GLuint pipeline, GLuint program) DECLARE_GL_FUNCTION_END_NO_RETURN(void, ActiveShaderProgram, pipeline, program)
 DECLARE_GL_FUNCTION_STUB_HEAD(GLuint, CreateShaderProgramv, GLenum type, GLsizei count, const GLchar* const* strings) DECLARE_GL_FUNCTION_STUB_END(GLuint, CreateShaderProgramv, type, count, strings)
-DECLARE_GL_FUNCTION_STUB_HEAD(void, BindProgramPipeline, GLuint pipeline) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, BindProgramPipeline, pipeline)
+DECLARE_GL_FUNCTION_HEAD(void, BindProgramPipeline, GLuint pipeline) DECLARE_GL_FUNCTION_END_NO_RETURN(void, BindProgramPipeline, pipeline)
-DECLARE_GL_FUNCTION_STUB_HEAD(void, DeleteProgramPipelines, GLsizei n, const GLuint* pipelines) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, DeleteProgramPipelines, n, pipelines)
+DECLARE_GL_FUNCTION_HEAD(void, DeleteProgramPipelines, GLsizei n, const GLuint* pipelines) DECLARE_GL_FUNCTION_END_NO_RETURN(void, DeleteProgramPipelines, n, pipelines)
-DECLARE_GL_FUNCTION_STUB_HEAD(void, GenProgramPipelines, GLsizei n, GLuint* pipelines) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, GenProgramPipelines, n, pipelines)
+DECLARE_GL_FUNCTION_HEAD(void, GenProgramPipelines, GLsizei n, GLuint* pipelines) DECLARE_GL_FUNCTION_END_NO_RETURN(void, GenProgramPipelines, n, pipelines)
-DECLARE_GL_FUNCTION_STUB_HEAD(GLboolean, IsProgramPipeline, GLuint pipeline) DECLARE_GL_FUNCTION_STUB_END(GLboolean, IsProgramPipeline, pipeline)
+DECLARE_GL_FUNCTION_HEAD(GLboolean, IsProgramPipeline, GLuint pipeline) DECLARE_GL_FUNCTION_END(GLboolean, IsProgramPipeline, pipeline)
-DECLARE_GL_FUNCTION_STUB_HEAD(void, GetProgramPipelineiv, GLuint pipeline, GLenum pname, GLint* params) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, GetProgramPipelineiv, pipeline, pname, params)
+DECLARE_GL_FUNCTION_HEAD(void, GetProgramPipelineiv, GLuint pipeline, GLenum pname, GLint* params) DECLARE_GL_FUNCTION_END_NO_RETURN(void, GetProgramPipelineiv, pipeline, pname, params)
 DECLARE_GL_FUNCTION_HEAD(void, ProgramUniform1i, GLuint program, GLint location, GLint v0) DECLARE_GL_FUNCTION_END_NO_RETURN(void, ProgramUniform1i, program, location, v0)
 DECLARE_GL_FUNCTION_HEAD(void, ProgramUniform2i, GLuint program, GLint location, GLint v0, GLint v1) DECLARE_GL_FUNCTION_END_NO_RETURN(void, ProgramUniform2i, program, location, v0, v1)
 DECLARE_GL_FUNCTION_HEAD(void, ProgramUniform3i, GLuint program, GLint location, GLint v0, GLint v1, GLint v2) DECLARE_GL_FUNCTION_END_NO_RETURN(void, ProgramUniform3i, program, location, v0, v1, v2)
@@ -358,8 +359,8 @@ DECLARE_GL_FUNCTION_HEAD(void, ProgramUniformMatrix2x4fv, GLuint program, GLint
 DECLARE_GL_FUNCTION_HEAD(void, ProgramUniformMatrix4x2fv, GLuint program, GLint location, GLsizei count, GLboolean transpose, const GLfloat* value) DECLARE_GL_FUNCTION_END_NO_RETURN(void, ProgramUniformMatrix4x2fv, program, location, count, transpose, value)
 DECLARE_GL_FUNCTION_HEAD(void, ProgramUniformMatrix3x4fv, GLuint program, GLint location, GLsizei count, GLboolean transpose, const GLfloat* value) DECLARE_GL_FUNCTION_END_NO_RETURN(void, ProgramUniformMatrix3x4fv, program, location, count, transpose, value)
 DECLARE_GL_FUNCTION_HEAD(void, ProgramUniformMatrix4x3fv, GLuint program, GLint location, GLsizei count, GLboolean transpose, const GLfloat* value) DECLARE_GL_FUNCTION_END_NO_RETURN(void, ProgramUniformMatrix4x3fv, program, location, count, transpose, value)
-DECLARE_GL_FUNCTION_STUB_HEAD(void, ValidateProgramPipeline, GLuint pipeline) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, ValidateProgramPipeline, pipeline)
+DECLARE_GL_FUNCTION_HEAD(void, ValidateProgramPipeline, GLuint pipeline) DECLARE_GL_FUNCTION_END_NO_RETURN(void, ValidateProgramPipeline, pipeline)
-DECLARE_GL_FUNCTION_STUB_HEAD(void, GetProgramPipelineInfoLog, GLuint pipeline, GLsizei bufSize, GLsizei* length, GLchar* infoLog) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, GetProgramPipelineInfoLog, pipeline, bufSize, length, infoLog)
+DECLARE_GL_FUNCTION_HEAD(void, GetProgramPipelineInfoLog, GLuint pipeline, GLsizei bufSize, GLsizei* length, GLchar* infoLog) DECLARE_GL_FUNCTION_END_NO_RETURN(void, GetProgramPipelineInfoLog, pipeline, bufSize, length, infoLog)
 DECLARE_GL_FUNCTION_HEAD(void, BindImageTexture, GLuint unit, GLuint texture, GLint level, GLboolean layered, GLint layer, GLenum access, GLenum format) DECLARE_GL_FUNCTION_END_NO_RETURN(void, BindImageTexture, unit, texture, level, layered, layer, access, format)
 DECLARE_GL_FUNCTION_HEAD(void, GetBooleani_v, GLenum target, GLuint index, GLboolean* data) DECLARE_GL_FUNCTION_END_NO_RETURN(void, GetBooleani_v, target, index, data)
 DECLARE_GL_FUNCTION_HEAD(void, MemoryBarrier, GLbitfield barriers) DECLARE_GL_FUNCTION_END_NO_RETURN(void, MemoryBarrier, barriers)
@@ -996,7 +997,7 @@ DECLARE_GL_FUNCTION_HEAD(void, MultiDrawElementsIndirect, GLenum mode, GLenum ty
 DECLARE_GL_FUNCTION_HEAD(GLint, GetProgramResourceLocationIndex, GLuint program, GLenum programInterface, const GLchar* name) DECLARE_GL_FUNCTION_END(GLint, GetProgramResourceLocationIndex, program, programInterface, name)
 DECLARE_GL_FUNCTION_HEAD(void, ShaderStorageBlockBinding, GLuint program, GLuint storageBlockIndex, GLuint storageBlockBinding) DECLARE_GL_FUNCTION_END_NO_RETURN(void, ShaderStorageBlockBinding, program, storageBlockIndex, storageBlockBinding)
 DECLARE_GL_FUNCTION_STUB_HEAD(void, TextureView, GLuint texture, GLenum target, GLuint origtexture, GLenum internalformat, GLuint minlevel, GLuint numlevels, GLuint minlayer, GLuint numlayers) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, TextureView, texture, target, origtexture, internalformat, minlevel, numlevels, minlayer, numlayers)
-DECLARE_GL_FUNCTION_STUB_HEAD(void, VertexAttribLFormat, GLuint attribindex, GLint size, GLenum type, GLuint relativeoffset) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, VertexAttribLFormat, attribindex, size, type, relativeoffset)
+DECLARE_GL_FUNCTION_HEAD(void, VertexAttribLFormat, GLuint attribindex, GLint size, GLenum type, GLuint relativeoffset) DECLARE_GL_FUNCTION_END_NO_RETURN(void, VertexAttribLFormat, attribindex, size, type, relativeoffset)
 DECLARE_GL_FUNCTION_HEAD(void, BufferStorage, GLenum target, GLsizeiptr size, const void* data, GLbitfield flags) DECLARE_GL_FUNCTION_END_NO_RETURN(void, BufferStorage, target, size, data, flags)
 DECLARE_GL_FUNCTION_HEAD(void, ClearTexImage, GLuint texture, GLint level, GLenum format, GLenum type, const void* data) DECLARE_GL_FUNCTION_END_NO_RETURN(void, ClearTexImage, texture, level, format, type, data)
 DECLARE_GL_FUNCTION_HEAD(void, ClearTexSubImage, GLuint texture, GLint level, GLint xoffset, GLint yoffset, GLint zoffset, GLsizei width, GLsizei height, GLsizei depth, GLenum format, GLenum type, const void* data) DECLARE_GL_FUNCTION_END_NO_RETURN(void, ClearTexSubImage, texture, level, xoffset, yoffset, zoffset, width, height, depth, format, type, data)
@@ -1007,12 +1008,12 @@ DECLARE_GL_FUNCTION_HEAD(void, BindSamplers, GLuint first, GLsizei count, const
 DECLARE_GL_FUNCTION_STUB_HEAD(void, BindImageTextures, GLuint first, GLsizei count, const GLuint* textures) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, BindImageTextures, first, count, textures)
 DECLARE_GL_FUNCTION_HEAD(void, BindVertexBuffers, GLuint first, GLsizei count, const GLuint* buffers, const GLintptr* offsets, const GLsizei* strides) DECLARE_GL_FUNCTION_END_NO_RETURN(void, BindVertexBuffers, first, count, buffers, offsets, strides)
 DECLARE_GL_FUNCTION_STUB_HEAD(void, ClipControl, GLenum origin, GLenum depth) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, ClipControl, origin, depth)
-DECLARE_GL_FUNCTION_STUB_HEAD(void, CreateTransformFeedbacks, GLsizei n, GLuint* ids) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, CreateTransformFeedbacks, n, ids)
+DECLARE_GL_FUNCTION_HEAD(void, CreateTransformFeedbacks, GLsizei n, GLuint* ids) DECLARE_GL_FUNCTION_END_NO_RETURN(void, CreateTransformFeedbacks, n, ids)
-DECLARE_GL_FUNCTION_STUB_HEAD(void, TransformFeedbackBufferBase, GLuint xfb, GLuint index, GLuint buffer) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, TransformFeedbackBufferBase, xfb, index, buffer)
+DECLARE_GL_FUNCTION_HEAD(void, TransformFeedbackBufferBase, GLuint xfb, GLuint index, GLuint buffer) DECLARE_GL_FUNCTION_END_NO_RETURN(void, TransformFeedbackBufferBase, xfb, index, buffer)
-DECLARE_GL_FUNCTION_STUB_HEAD(void, TransformFeedbackBufferRange, GLuint xfb, GLuint index, GLuint buffer, GLintptr offset, GLsizeiptr size) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, TransformFeedbackBufferRange, xfb, index, buffer, offset, size)
+DECLARE_GL_FUNCTION_HEAD(void, TransformFeedbackBufferRange, GLuint xfb, GLuint index, GLuint buffer, GLintptr offset, GLsizeiptr size) DECLARE_GL_FUNCTION_END_NO_RETURN(void, TransformFeedbackBufferRange, xfb, index, buffer, offset, size)
-DECLARE_GL_FUNCTION_STUB_HEAD(void, GetTransformFeedbackiv, GLuint xfb, GLenum pname, GLint* param) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, GetTransformFeedbackiv, xfb, pname, param)
+DECLARE_GL_FUNCTION_HEAD(void, GetTransformFeedbackiv, GLuint xfb, GLenum pname, GLint* param) DECLARE_GL_FUNCTION_END_NO_RETURN(void, GetTransformFeedbackiv, xfb, pname, param)
-DECLARE_GL_FUNCTION_STUB_HEAD(void, GetTransformFeedbacki_v, GLuint xfb, GLenum pname, GLuint index, GLint* param) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, GetTransformFeedbacki_v, xfb, pname, index, param)
+DECLARE_GL_FUNCTION_HEAD(void, GetTransformFeedbacki_v, GLuint xfb, GLenum pname, GLuint index, GLint* param) DECLARE_GL_FUNCTION_END_NO_RETURN(void, GetTransformFeedbacki_v, xfb, pname, index, param)
-DECLARE_GL_FUNCTION_STUB_HEAD(void, GetTransformFeedbacki64_v, GLuint xfb, GLenum pname, GLuint index, GLint64* param) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, GetTransformFeedbacki64_v, xfb, pname, index, param)
+DECLARE_GL_FUNCTION_HEAD(void, GetTransformFeedbacki64_v, GLuint xfb, GLenum pname, GLuint index, GLint64* param) DECLARE_GL_FUNCTION_END_NO_RETURN(void, GetTransformFeedbacki64_v, xfb, pname, index, param)
 DECLARE_GL_FUNCTION_HEAD(void, CreateBuffers, GLsizei n, GLuint* buffers) DECLARE_GL_FUNCTION_END_NO_RETURN(void, CreateBuffers, n, buffers)
 DECLARE_GL_FUNCTION_HEAD(void, NamedBufferStorage, GLuint buffer, GLsizeiptr size, const void* data, GLbitfield flags) DECLARE_GL_FUNCTION_END_NO_RETURN(void, NamedBufferStorage, buffer, size, data, flags)
 DECLARE_GL_FUNCTION_HEAD(void, NamedBufferData, GLuint buffer, GLsizeiptr size, const void* data, GLenum usage) DECLARE_GL_FUNCTION_END_NO_RETURN(void, NamedBufferData, buffer, size, data, usage)
@@ -1025,24 +1026,24 @@ DECLARE_GL_FUNCTION_HEAD(void, FlushMappedNamedBufferRange, GLuint buffer, GLint
 DECLARE_GL_FUNCTION_HEAD(void, GetNamedBufferParameteriv, GLuint buffer, GLenum pname, GLint* params) DECLARE_GL_FUNCTION_END_NO_RETURN(void, GetNamedBufferParameteriv, buffer, pname, params)
 DECLARE_GL_FUNCTION_HEAD(void, GetNamedBufferParameteri64v, GLuint buffer, GLenum pname, GLint64* params) DECLARE_GL_FUNCTION_END_NO_RETURN(void, GetNamedBufferParameteri64v, buffer, pname, params)
 DECLARE_GL_FUNCTION_HEAD(void, GetNamedBufferPointerv, GLuint buffer, GLenum pname, void** params) DECLARE_GL_FUNCTION_END_NO_RETURN(void, GetNamedBufferPointerv, buffer, pname, params)
-DECLARE_GL_FUNCTION_STUB_HEAD(void, GetNamedBufferSubData, GLuint buffer, GLintptr offset, GLsizeiptr size, void* data) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, GetNamedBufferSubData, buffer, offset, size, data)
+DECLARE_GL_FUNCTION_HEAD(void, GetNamedBufferSubData, GLuint buffer, GLintptr offset, GLsizeiptr size, void* data) DECLARE_GL_FUNCTION_END_NO_RETURN(void, GetNamedBufferSubData, buffer, offset, size, data)
 DECLARE_GL_FUNCTION_HEAD(void, CreateFramebuffers, GLsizei n, GLuint* framebuffers) DECLARE_GL_FUNCTION_END_NO_RETURN(void, CreateFramebuffers, n, framebuffers)
 DECLARE_GL_FUNCTION_HEAD(void, NamedFramebufferRenderbuffer, GLuint framebuffer, GLenum attachment, GLenum renderbuffertarget, GLuint renderbuffer) DECLARE_GL_FUNCTION_END_NO_RETURN(void, NamedFramebufferRenderbuffer, framebuffer, attachment, renderbuffertarget, renderbuffer)
-DECLARE_GL_FUNCTION_STUB_HEAD(void, NamedFramebufferParameteri, GLuint framebuffer, GLenum pname, GLint param) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, NamedFramebufferParameteri, framebuffer, pname, param)
+DECLARE_GL_FUNCTION_HEAD(void, NamedFramebufferParameteri, GLuint framebuffer, GLenum pname, GLint param) DECLARE_GL_FUNCTION_END_NO_RETURN(void, NamedFramebufferParameteri, framebuffer, pname, param)
 DECLARE_GL_FUNCTION_HEAD(void, NamedFramebufferTexture, GLuint framebuffer, GLenum attachment, GLuint texture, GLint level) DECLARE_GL_FUNCTION_END_NO_RETURN(void, NamedFramebufferTexture, framebuffer, attachment, texture, level)
 DECLARE_GL_FUNCTION_HEAD(void, NamedFramebufferTextureLayer, GLuint framebuffer, GLenum attachment, GLuint texture, GLint level, GLint layer) DECLARE_GL_FUNCTION_END_NO_RETURN(void, NamedFramebufferTextureLayer, framebuffer, attachment, texture, level, layer)
 DECLARE_GL_FUNCTION_HEAD(void, NamedFramebufferDrawBuffer, GLuint framebuffer, GLenum buf) DECLARE_GL_FUNCTION_END_NO_RETURN(void, NamedFramebufferDrawBuffer, framebuffer, buf)
 DECLARE_GL_FUNCTION_HEAD(void, NamedFramebufferDrawBuffers, GLuint framebuffer, GLsizei n, const GLenum* bufs) DECLARE_GL_FUNCTION_END_NO_RETURN(void, NamedFramebufferDrawBuffers, framebuffer, n, bufs)
 DECLARE_GL_FUNCTION_HEAD(void, NamedFramebufferReadBuffer, GLuint framebuffer, GLenum src) DECLARE_GL_FUNCTION_END_NO_RETURN(void, NamedFramebufferReadBuffer, framebuffer, src)
-DECLARE_GL_FUNCTION_STUB_HEAD(void, InvalidateNamedFramebufferData, GLuint framebuffer, GLsizei numAttachments, const GLenum* attachments) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, InvalidateNamedFramebufferData, framebuffer, numAttachments, attachments)
+DECLARE_GL_FUNCTION_HEAD(void, InvalidateNamedFramebufferData, GLuint framebuffer, GLsizei numAttachments, const GLenum* attachments) DECLARE_GL_FUNCTION_END_NO_RETURN(void, InvalidateNamedFramebufferData, framebuffer, numAttachments, attachments)
-DECLARE_GL_FUNCTION_STUB_HEAD(void, InvalidateNamedFramebufferSubData, GLuint framebuffer, GLsizei numAttachments, const GLenum* attachments, GLint x, GLint y, GLsizei width, GLsizei height) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, InvalidateNamedFramebufferSubData, framebuffer, numAttachments, attachments, x, y, width, height)
+DECLARE_GL_FUNCTION_HEAD(void, InvalidateNamedFramebufferSubData, GLuint framebuffer, GLsizei numAttachments, const GLenum* attachments, GLint x, GLint y, GLsizei width, GLsizei height) DECLARE_GL_FUNCTION_END_NO_RETURN(void, InvalidateNamedFramebufferSubData, framebuffer, numAttachments, attachments, x, y, width, height)
-DECLARE_GL_FUNCTION_STUB_HEAD(void, ClearNamedFramebufferiv, GLuint framebuffer, GLenum buffer, GLint drawbuffer, const GLint* value) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, ClearNamedFramebufferiv, framebuffer, buffer, drawbuffer, value)
+DECLARE_GL_FUNCTION_HEAD(void, ClearNamedFramebufferiv, GLuint framebuffer, GLenum buffer, GLint drawbuffer, const GLint* value) DECLARE_GL_FUNCTION_END_NO_RETURN(void, ClearNamedFramebufferiv, framebuffer, buffer, drawbuffer, value)
-DECLARE_GL_FUNCTION_STUB_HEAD(void, ClearNamedFramebufferuiv, GLuint framebuffer, GLenum buffer, GLint drawbuffer, const GLuint* value) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, ClearNamedFramebufferuiv, framebuffer, buffer, drawbuffer, value)
+DECLARE_GL_FUNCTION_HEAD(void, ClearNamedFramebufferuiv, GLuint framebuffer, GLenum buffer, GLint drawbuffer, const GLuint* value) DECLARE_GL_FUNCTION_END_NO_RETURN(void, ClearNamedFramebufferuiv, framebuffer, buffer, drawbuffer, value)
 DECLARE_GL_FUNCTION_HEAD(void, ClearNamedFramebufferfv, GLuint framebuffer, GLenum buffer, GLint drawbuffer, const GLfloat* value) DECLARE_GL_FUNCTION_END_NO_RETURN(void, ClearNamedFramebufferfv, framebuffer, buffer, drawbuffer, value)
 DECLARE_GL_FUNCTION_HEAD(void, ClearNamedFramebufferfi, GLuint framebuffer, GLenum buffer, GLint drawbuffer, GLfloat depth, GLint stencil) DECLARE_GL_FUNCTION_END_NO_RETURN(void, ClearNamedFramebufferfi, framebuffer, buffer, drawbuffer, depth, stencil)
 DECLARE_GL_FUNCTION_HEAD(void, BlitNamedFramebuffer, GLuint readFramebuffer, GLuint drawFramebuffer, GLint srcX0, GLint srcY0, GLint srcX1, GLint srcY1, GLint dstX0, GLint dstY0, GLint dstX1, GLint dstY1, GLbitfield mask, GLenum filter) DECLARE_GL_FUNCTION_END_NO_RETURN(void, BlitNamedFramebuffer, readFramebuffer, drawFramebuffer, srcX0, srcY0, srcX1, srcY1, dstX0, dstY0, dstX1, dstY1, mask, filter)
 DECLARE_GL_FUNCTION_HEAD(GLenum, CheckNamedFramebufferStatus, GLuint framebuffer, GLenum target) DECLARE_GL_FUNCTION_END(GLenum, CheckNamedFramebufferStatus, framebuffer, target)
-DECLARE_GL_FUNCTION_STUB_HEAD(void, GetNamedFramebufferParameteriv, GLuint framebuffer, GLenum pname, GLint* param) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, GetNamedFramebufferParameteriv, framebuffer, pname, param)
+DECLARE_GL_FUNCTION_HEAD(void, GetNamedFramebufferParameteriv, GLuint framebuffer, GLenum pname, GLint* param) DECLARE_GL_FUNCTION_END_NO_RETURN(void, GetNamedFramebufferParameteriv, framebuffer, pname, param)
 DECLARE_GL_FUNCTION_HEAD(void, GetNamedFramebufferAttachmentParameteriv, GLuint framebuffer, GLenum attachment, GLenum pname, GLint* params) DECLARE_GL_FUNCTION_END_NO_RETURN(void, GetNamedFramebufferAttachmentParameteriv, framebuffer, attachment, pname, params)
 DECLARE_GL_FUNCTION_HEAD(void, CreateRenderbuffers, GLsizei n, GLuint* renderbuffers) DECLARE_GL_FUNCTION_END_NO_RETURN(void, CreateRenderbuffers, n, renderbuffers)
 DECLARE_GL_FUNCTION_HEAD(void, NamedRenderbufferStorage, GLuint renderbuffer, GLenum internalformat, GLsizei width, GLsizei height) DECLARE_GL_FUNCTION_END_NO_RETURN(void, NamedRenderbufferStorage, renderbuffer, internalformat, width, height)
@@ -1062,9 +1063,9 @@ DECLARE_GL_FUNCTION_HEAD(void, TextureSubImage3D, GLuint texture, GLint level, G
 DECLARE_GL_FUNCTION_STUB_HEAD(void, CompressedTextureSubImage1D, GLuint texture, GLint level, GLint xoffset, GLsizei width, GLenum format, GLsizei imageSize, const void* data) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, CompressedTextureSubImage1D, texture, level, xoffset, width, format, imageSize, data)
 DECLARE_GL_FUNCTION_STUB_HEAD(void, CompressedTextureSubImage2D, GLuint texture, GLint level, GLint xoffset, GLint yoffset, GLsizei width, GLsizei height, GLenum format, GLsizei imageSize, const void* data) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, CompressedTextureSubImage2D, texture, level, xoffset, yoffset, width, height, format, imageSize, data)
 DECLARE_GL_FUNCTION_STUB_HEAD(void, CompressedTextureSubImage3D, GLuint texture, GLint level, GLint xoffset, GLint yoffset, GLint zoffset, GLsizei width, GLsizei height, GLsizei depth, GLenum format, GLsizei imageSize, const void* data) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, CompressedTextureSubImage3D, texture, level, xoffset, yoffset, zoffset, width, height, depth, format, imageSize, data)
-DECLARE_GL_FUNCTION_STUB_HEAD(void, CopyTextureSubImage1D, GLuint texture, GLint level, GLint xoffset, GLint x, GLint y, GLsizei width) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, CopyTextureSubImage1D, texture, level, xoffset, x, y, width)
+DECLARE_GL_FUNCTION_HEAD(void, CopyTextureSubImage1D, GLuint texture, GLint level, GLint xoffset, GLint x, GLint y, GLsizei width) DECLARE_GL_FUNCTION_END_NO_RETURN(void, CopyTextureSubImage1D, texture, level, xoffset, x, y, width)
 DECLARE_GL_FUNCTION_HEAD(void, CopyTextureSubImage2D, GLuint texture, GLint level, GLint xoffset, GLint yoffset, GLint x, GLint y, GLsizei width, GLsizei height) DECLARE_GL_FUNCTION_END_NO_RETURN(void, CopyTextureSubImage2D, texture, level, xoffset, yoffset, x, y, width, height)
-DECLARE_GL_FUNCTION_STUB_HEAD(void, CopyTextureSubImage3D, GLuint texture, GLint level, GLint xoffset, GLint yoffset, GLint zoffset, GLint x, GLint y, GLsizei width, GLsizei height) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, CopyTextureSubImage3D, texture, level, xoffset, yoffset, zoffset, x, y, width, height)
+DECLARE_GL_FUNCTION_HEAD(void, CopyTextureSubImage3D, GLuint texture, GLint level, GLint xoffset, GLint yoffset, GLint zoffset, GLint x, GLint y, GLsizei width, GLsizei height) DECLARE_GL_FUNCTION_END_NO_RETURN(void, CopyTextureSubImage3D, texture, level, xoffset, yoffset, zoffset, x, y, width, height)
 DECLARE_GL_FUNCTION_HEAD(void, TextureParameterf, GLuint texture, GLenum pname, GLfloat param) DECLARE_GL_FUNCTION_END_NO_RETURN(void, TextureParameterf, texture, pname, param)
 DECLARE_GL_FUNCTION_HEAD(void, TextureParameterfv, GLuint texture, GLenum pname, const GLfloat* param) DECLARE_GL_FUNCTION_END_NO_RETURN(void, TextureParameterfv, texture, pname, param)
 DECLARE_GL_FUNCTION_HEAD(void, TextureParameteri, GLuint texture, GLenum pname, GLint param) DECLARE_GL_FUNCTION_END_NO_RETURN(void, TextureParameteri, texture, pname, param)
@@ -1074,7 +1075,7 @@ DECLARE_GL_FUNCTION_HEAD(void, TextureParameteriv, GLuint texture, GLenum pname,
 DECLARE_GL_FUNCTION_HEAD(void, GenerateTextureMipmap, GLuint texture) DECLARE_GL_FUNCTION_END_NO_RETURN(void, GenerateTextureMipmap, texture)
 DECLARE_GL_FUNCTION_HEAD(void, BindTextureUnit, GLuint unit, GLuint texture) DECLARE_GL_FUNCTION_END_NO_RETURN(void, BindTextureUnit, unit, texture)
 DECLARE_GL_FUNCTION_HEAD(void, GetTextureImage, GLuint texture, GLint level, GLenum format, GLenum type, GLsizei bufSize, void* pixels) DECLARE_GL_FUNCTION_END_NO_RETURN(void, GetTextureImage, texture, level, format, type, bufSize, pixels)
-DECLARE_GL_FUNCTION_STUB_HEAD(void, GetCompressedTextureImage, GLuint texture, GLint level, GLsizei bufSize, void* pixels) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, GetCompressedTextureImage, texture, level, bufSize, pixels)
+DECLARE_GL_FUNCTION_HEAD(void, GetCompressedTextureImage, GLuint texture, GLint level, GLsizei bufSize, void* pixels) DECLARE_GL_FUNCTION_END_NO_RETURN(void, GetCompressedTextureImage, texture, level, bufSize, pixels)
 DECLARE_GL_FUNCTION_HEAD(void, GetTextureLevelParameterfv, GLuint texture, GLint level, GLenum pname, GLfloat* params) DECLARE_GL_FUNCTION_END_NO_RETURN(void, GetTextureLevelParameterfv, texture, level, pname, params)
 DECLARE_GL_FUNCTION_HEAD(void, GetTextureLevelParameteriv, GLuint texture, GLint level, GLenum pname, GLint* params) DECLARE_GL_FUNCTION_END_NO_RETURN(void, GetTextureLevelParameteriv, texture, level, pname, params)
 DECLARE_GL_FUNCTION_HEAD(void, GetTextureParameterfv, GLuint texture, GLenum pname, GLfloat* params) DECLARE_GL_FUNCTION_END_NO_RETURN(void, GetTextureParameterfv, texture, pname, params)
@@ -1090,18 +1091,18 @@ DECLARE_GL_FUNCTION_HEAD(void, VertexArrayVertexBuffers, GLuint vaobj, GLuint fi
 DECLARE_GL_FUNCTION_HEAD(void, VertexArrayAttribBinding, GLuint vaobj, GLuint attribindex, GLuint bindingindex) DECLARE_GL_FUNCTION_END_NO_RETURN(void, VertexArrayAttribBinding, vaobj, attribindex, bindingindex)
 DECLARE_GL_FUNCTION_HEAD(void, VertexArrayAttribFormat, GLuint vaobj, GLuint attribindex, GLint size, GLenum type, GLboolean normalized, GLuint relativeoffset) DECLARE_GL_FUNCTION_END_NO_RETURN(void, VertexArrayAttribFormat, vaobj, attribindex, size, type, normalized, relativeoffset)
 DECLARE_GL_FUNCTION_HEAD(void, VertexArrayAttribIFormat, GLuint vaobj, GLuint attribindex, GLint size, GLenum type, GLuint relativeoffset) DECLARE_GL_FUNCTION_END_NO_RETURN(void, VertexArrayAttribIFormat, vaobj, attribindex, size, type, relativeoffset)
-DECLARE_GL_FUNCTION_STUB_HEAD(void, VertexArrayAttribLFormat, GLuint vaobj, GLuint attribindex, GLint size, GLenum type, GLuint relativeoffset) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, VertexArrayAttribLFormat, vaobj, attribindex, size, type, relativeoffset)
+DECLARE_GL_FUNCTION_HEAD(void, VertexArrayAttribLFormat, GLuint vaobj, GLuint attribindex, GLint size, GLenum type, GLuint relativeoffset) DECLARE_GL_FUNCTION_END_NO_RETURN(void, VertexArrayAttribLFormat, vaobj, attribindex, size, type, relativeoffset)
 DECLARE_GL_FUNCTION_HEAD(void, VertexArrayBindingDivisor, GLuint vaobj, GLuint bindingindex, GLuint divisor) DECLARE_GL_FUNCTION_END_NO_RETURN(void, VertexArrayBindingDivisor, vaobj, bindingindex, divisor)
 DECLARE_GL_FUNCTION_HEAD(void, GetVertexArrayiv, GLuint vaobj, GLenum pname, GLint* param) DECLARE_GL_FUNCTION_END_NO_RETURN(void, GetVertexArrayiv, vaobj, pname, param)
 DECLARE_GL_FUNCTION_HEAD(void, GetVertexArrayIndexediv, GLuint vaobj, GLuint index, GLenum pname, GLint* param) DECLARE_GL_FUNCTION_END_NO_RETURN(void, GetVertexArrayIndexediv, vaobj, index, pname, param)
 DECLARE_GL_FUNCTION_HEAD(void, GetVertexArrayIndexed64iv, GLuint vaobj, GLuint index, GLenum pname, GLint64* param) DECLARE_GL_FUNCTION_END_NO_RETURN(void, GetVertexArrayIndexed64iv, vaobj, index, pname, param)
 DECLARE_GL_FUNCTION_HEAD(void, CreateSamplers, GLsizei n, GLuint* samplers) DECLARE_GL_FUNCTION_END_NO_RETURN(void, CreateSamplers, n, samplers)
-DECLARE_GL_FUNCTION_STUB_HEAD(void, CreateProgramPipelines, GLsizei n, GLuint* pipelines) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, CreateProgramPipelines, n, pipelines)
+DECLARE_GL_FUNCTION_HEAD(void, CreateProgramPipelines, GLsizei n, GLuint* pipelines) DECLARE_GL_FUNCTION_END_NO_RETURN(void, CreateProgramPipelines, n, pipelines)
 DECLARE_GL_FUNCTION_HEAD(void, CreateQueries, GLenum target, GLsizei n, GLuint* ids) DECLARE_GL_FUNCTION_END_NO_RETURN(void, CreateQueries, target, n, ids)
-DECLARE_GL_FUNCTION_STUB_HEAD(void, GetQueryBufferObjecti64v, GLuint id, GLuint buffer, GLenum pname, GLintptr offset) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, GetQueryBufferObjecti64v, id, buffer, pname, offset)
+DECLARE_GL_FUNCTION_HEAD(void, GetQueryBufferObjecti64v, GLuint id, GLuint buffer, GLenum pname, GLintptr offset) DECLARE_GL_FUNCTION_END_NO_RETURN(void, GetQueryBufferObjecti64v, id, buffer, pname, offset)
-DECLARE_GL_FUNCTION_STUB_HEAD(void, GetQueryBufferObjectiv, GLuint id, GLuint buffer, GLenum pname, GLintptr offset) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, GetQueryBufferObjectiv, id, buffer, pname, offset)
+DECLARE_GL_FUNCTION_HEAD(void, GetQueryBufferObjectiv, GLuint id, GLuint buffer, GLenum pname, GLintptr offset) DECLARE_GL_FUNCTION_END_NO_RETURN(void, GetQueryBufferObjectiv, id, buffer, pname, offset)
-DECLARE_GL_FUNCTION_STUB_HEAD(void, GetQueryBufferObjectui64v, GLuint id, GLuint buffer, GLenum pname, GLintptr offset) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, GetQueryBufferObjectui64v, id, buffer, pname, offset)
+DECLARE_GL_FUNCTION_HEAD(void, GetQueryBufferObjectui64v, GLuint id, GLuint buffer, GLenum pname, GLintptr offset) DECLARE_GL_FUNCTION_END_NO_RETURN(void, GetQueryBufferObjectui64v, id, buffer, pname, offset)
-DECLARE_GL_FUNCTION_STUB_HEAD(void, GetQueryBufferObjectuiv, GLuint id, GLuint buffer, GLenum pname, GLintptr offset) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, GetQueryBufferObjectuiv, id, buffer, pname, offset)
+DECLARE_GL_FUNCTION_HEAD(void, GetQueryBufferObjectuiv, GLuint id, GLuint buffer, GLenum pname, GLintptr offset) DECLARE_GL_FUNCTION_END_NO_RETURN(void, GetQueryBufferObjectuiv, id, buffer, pname, offset)
 DECLARE_GL_FUNCTION_HEAD(void, GetTextureSubImage, GLuint texture, GLint level, GLint xoffset, GLint yoffset, GLint zoffset, GLsizei width, GLsizei height, GLsizei depth, GLenum format, GLenum type, GLsizei bufSize, void* pixels) DECLARE_GL_FUNCTION_END_NO_RETURN(void, GetTextureSubImage, texture, level, xoffset, yoffset, zoffset, width, height, depth, format, type, bufSize, pixels)
 DECLARE_GL_FUNCTION_STUB_HEAD(void, GetCompressedTextureSubImage, GLuint texture, GLint level, GLint xoffset, GLint yoffset, GLint zoffset, GLsizei width, GLsizei height, GLsizei depth, GLsizei bufSize, void* pixels) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, GetCompressedTextureSubImage, texture, level, xoffset, yoffset, zoffset, width, height, depth, bufSize, pixels)
 DECLARE_GL_FUNCTION_STUB_HEAD(void, GetnCompressedTexImage, GLenum target, GLint lod, GLsizei bufSize, void* pixels) DECLARE_GL_FUNCTION_STUB_END_NO_RETURN(void, GetnCompressedTexImage, target, lod, bufSize, pixels)

MobileGL/MG_Impl/GLImpl/Framebuffer/GL_Framebuffer.h

@@ -58,7 +58,19 @@ namespace MobileGL::MG_Impl::GLImpl {
     void NamedFramebufferReadBuffer(GLuint framebuffer, GLenum src);
     void ClearNamedFramebufferfv(GLuint framebuffer, GLenum buffer, GLint drawbuffer, const GLfloat* value);
     void ClearNamedFramebufferfi(GLuint framebuffer, GLenum buffer, GLint drawbuffer, GLfloat depth, GLint stencil);
+    void InvalidateNamedFramebufferData(GLuint framebuffer, GLsizei numAttachments, const GLenum* attachments);
+    void InvalidateNamedFramebufferSubData(GLuint framebuffer, GLsizei numAttachments, const GLenum* attachments,
+                                           GLint x, GLint y, GLsizei width, GLsizei height);
+    void InvalidateFramebuffer(GLenum target, GLsizei numAttachments, const GLenum* attachments);
+    void InvalidateSubFramebuffer(GLenum target, GLsizei numAttachments, const GLenum* attachments, GLint x, GLint y,
+                                  GLsizei width, GLsizei height);
+    void ClearNamedFramebufferiv(GLuint framebuffer, GLenum buffer, GLint drawbuffer, const GLint* value);
+    void ClearNamedFramebufferuiv(GLuint framebuffer, GLenum buffer, GLint drawbuffer, const GLuint* value);
     GLenum CheckNamedFramebufferStatus(GLuint framebuffer, GLenum target);
+    void GetFramebufferParameteriv(GLenum target, GLenum pname, GLint* params);
+    void FramebufferParameteri(GLenum target, GLenum pname, GLint param);
+    void GetNamedFramebufferParameteriv(GLuint framebuffer, GLenum pname, GLint* params);
+    void NamedFramebufferParameteri(GLuint framebuffer, GLenum pname, GLint param);
     void GetNamedFramebufferAttachmentParameteriv(GLuint framebuffer, GLenum attachment, GLenum pname, GLint* params);
     void BlitNamedFramebuffer(GLuint readFramebuffer, GLuint drawFramebuffer, GLint srcX0, GLint srcY0, GLint srcX1,
                               GLint srcY1, GLint dstX0, GLint dstY0, GLint dstX1, GLint dstY1, GLbitfield mask,

MobileGL/MG_Impl/GLImpl/Framebuffer/Validators.cpp

@@ -7,6 +7,7 @@
 // End of Source File Header
 
 #include "Validators.h"
+#include <MG_Backend/BackendObjects.h>
 #include <MG_State/GLState/Core.h>
 #include <MG_State/GLState/ErrorState/Error.h>
 #include <MG_Util/Converters/GLToStr/GLEnumConverter.h>
@@ -60,6 +61,26 @@ namespace MobileGL::MG_Impl::GLImpl::FramebufferImpl {
         return true;
     }
 
+    Bool ValidateColorAttachmentInRange(FramebufferAttachmentType attachment, const char* caller) {
+        const auto first = static_cast<SizeT>(FramebufferAttachmentType::Color0);
+        const auto index = static_cast<SizeT>(attachment);
+        if (index < first) return true;
+        const auto colorIndex = index - first;
+        const auto limit = static_cast<SizeT>(
+            MG_Backend::pActiveBackendObject ? MG_Backend::pActiveBackendObject->GetDynamicParameters()
+                                                   .MaxColorAttachments
+                                             : static_cast<Int>(MG_State::GLState::FramebufferObject::MAX_DRAW_BUFFERS));
+        if (colorIndex >= limit) {
+            MG_State::pGLContext->RecordError(
+                ErrorCode::InvalidOperation,
+                MakeUnique<GenericErrorInfo>(
+                    "MG_Impl/GLImpl/FramebufferImpl", caller,
+                    std::format("Colour attachment {} is beyond GL_MAX_COLOR_ATTACHMENTS ({}).", colorIndex, limit)));
+            return false;
+        }
+        return true;
+    }
+
     Bool ValidateRenderbufferTarget(RenderbufferTarget target) {
         if (target == RenderbufferTarget::Unknown) {
             using namespace MG_Util;
@@ -97,4 +118,100 @@ namespace MobileGL::MG_Impl::GLImpl::FramebufferImpl {
                                          std::format("Renderbuffer name {} is not valid.", index)));
         return false;
     }
+
+    Bool ValidateFramebufferParameterPname(GLenum pname, Bool isDefaultFramebuffer, Bool forSetter,
+                                           const char* caller) {
+        Bool isDefaultParameter = false;
+        switch (pname) {
+        case GL_FRAMEBUFFER_DEFAULT_WIDTH:
+        case GL_FRAMEBUFFER_DEFAULT_HEIGHT:
+        case GL_FRAMEBUFFER_DEFAULT_LAYERS:
+        case GL_FRAMEBUFFER_DEFAULT_SAMPLES:
+        case GL_FRAMEBUFFER_DEFAULT_FIXED_SAMPLE_LOCATIONS:
+            isDefaultParameter = true;
+            break;
+        case GL_DOUBLEBUFFER:
+        case GL_IMPLEMENTATION_COLOR_READ_FORMAT:
+        case GL_IMPLEMENTATION_COLOR_READ_TYPE:
+        case GL_SAMPLES:
+        case GL_SAMPLE_BUFFERS:
+        case GL_STEREO:
+            // Queryable only; glFramebufferParameteri sets none of these.
+            if (forSetter) {
+                MG_State::pGLContext->RecordError(
+                    ErrorCode::InvalidEnum,
+                    MakeUnique<GenericErrorInfo>(
+                        "MG_Impl/GLImpl/FramebufferImpl", caller,
+                        std::format("pname {} is not settable on a framebuffer.",
+                                    MG_Util::ConvertGLEnumToString(pname))));
+                return false;
+            }
+            break;
+        default:
+            MG_State::pGLContext->RecordError(
+                ErrorCode::InvalidEnum,
+                MakeUnique<GenericErrorInfo>(
+                    "MG_Impl/GLImpl/FramebufferImpl", caller,
+                    std::format("pname {} is not a framebuffer parameter.",
+                                MG_Util::ConvertGLEnumToString(pname))));
+            return false;
+        }
+
+        // The default framebuffer has no DEFAULT_* state of its own - its shape comes from the
+        // surface - so those names are accepted enums it simply cannot answer or accept.
+        if (isDefaultFramebuffer && isDefaultParameter) {
+            MG_State::pGLContext->RecordError(
+                ErrorCode::InvalidOperation,
+                MakeUnique<GenericErrorInfo>(
+                    "MG_Impl/GLImpl/FramebufferImpl", caller,
+                    std::format("pname {} does not apply to the default framebuffer.",
+                                MG_Util::ConvertGLEnumToString(pname))));
+            return false;
+        }
+        return true;
+    }
+
+    Bool ValidateReadFramebufferForCopy(const char* caller) {
+        auto& framebufferObject =
+            MG_State::pGLContext->GetFramebufferBindingSlot(FramebufferTarget::Read).GetBoundObject();
+        if (!framebufferObject || !framebufferObject->CheckCompleteness()) {
+            MG_State::pGLContext->RecordError(
+                ErrorCode::InvalidFramebufferOperation,
+                MakeUnique<GenericErrorInfo>("MG_Impl/GLImpl/FramebufferImpl", caller,
+                                             "Read framebuffer is not framebuffer complete."));
+            return false;
+        }
+
+        const FramebufferAttachmentType readBuffer = framebufferObject->GetReadBuffer();
+        if (readBuffer == FramebufferAttachmentType::None ||
+            !framebufferObject->GetAttachment(readBuffer).IsValid()) {
+            MG_State::pGLContext->RecordError(
+                ErrorCode::InvalidOperation,
+                MakeUnique<GenericErrorInfo>("MG_Impl/GLImpl/FramebufferImpl", caller,
+                                             "Read buffer names no attachment of the read framebuffer."));
+            return false;
+        }
+
+        // SAMPLE_BUFFERS is one whenever the read buffer resolves to multisample storage. A
+        // multisample texture says so by its target - its sample count can legally be one - while a
+        // renderbuffer says so by having been given a non-zero sample count.
+        const auto& readAttachment = framebufferObject->GetAttachment(readBuffer);
+        Bool isMultisampled = false;
+        if (readAttachment.IsRenderbuffer() && readAttachment.GetRenderbuffer()) {
+            isMultisampled = readAttachment.GetRenderbuffer()->GetSamples() > 0;
+        } else if (readAttachment.IsTexture() && readAttachment.GetTexture()) {
+            const auto target = readAttachment.GetTexture()->GetTarget();
+            isMultisampled = target == TextureTarget::Texture2DMultisample ||
+                             target == TextureTarget::Texture2DMultisampleArray;
+        }
+        if (isMultisampled) {
+            MG_State::pGLContext->RecordError(
+                ErrorCode::InvalidOperation,
+                MakeUnique<GenericErrorInfo>("MG_Impl/GLImpl/FramebufferImpl", caller,
+                                             "Cannot copy from a multisampled read framebuffer."));
+            return false;
+        }
+
+        return true;
+    }
 } // namespace MobileGL::MG_Impl::GLImpl::FramebufferImpl

MobileGL/MG_Impl/GLImpl/Framebuffer/Validators.h

@@ -14,6 +14,21 @@ namespace MobileGL::MG_Impl::GLImpl::FramebufferImpl {
     Bool ValidateFramebufferTarget(FramebufferTarget target);
     Bool ValidateFramebufferName(Uint index, Bool allowZero = true);
     Bool ValidateFramebufferAttachmentType(FramebufferAttachmentType attachment);
+    // GL_COLOR_ATTACHMENTn is a token per n up to 31, but only the first GL_MAX_COLOR_ATTACHMENTS of
+    // them name an attachment point of a framebuffer object; the rest are INVALID_OPERATION for the
+    // attaching entry points (GL 4.6 core 9.2.7). Non-colour attachments pass through unchanged.
+    Bool ValidateColorAttachmentInRange(FramebufferAttachmentType attachment, const char* caller);
     Bool ValidateRenderbufferTarget(RenderbufferTarget target);
     Bool ValidateRenderbufferName(Uint index, Bool allowZero = true);
+    // The read-framebuffer preconditions the CopyTexSubImage family shares (GL 4.6 core 8.6): the
+    // read framebuffer must be complete, its read buffer must name a real attachment, and it must
+    // not be multisampled. Incompleteness is INVALID_FRAMEBUFFER_OPERATION, the other two are
+    // INVALID_OPERATION.
+    Bool ValidateReadFramebufferForCopy(const char* caller);
+    // The pname sets of glGet/FramebufferParameteri (GL 4.6 core 9.2.3). Order matters and is part
+    // of the contract: a name outside the table is INVALID_ENUM, and only then is a name that the
+    // DEFAULT framebuffer does not answer INVALID_OPERATION. Testing the framebuffer kind first
+    // would turn GL_FRAMEBUFFER_DEFAULT_WIDTH on framebuffer zero into the wrong error.
+    Bool ValidateFramebufferParameterPname(GLenum pname, Bool isDefaultFramebuffer, Bool forSetter,
+                                           const char* caller);
 } // namespace MobileGL::MG_Impl::GLImpl::FramebufferImpl

MobileGL/MG_Impl/GLImpl/Getter/GL_Getter.cpp

@@ -671,6 +671,40 @@ namespace MobileGL::MG_Impl::GLImpl {
         }
 
         switch (target) {
+        // The vertex buffer binding points of the vertex array object that is bound. Indexed by
+        // binding point, not by attribute (GL 4.6 core 10.3.1).
+        case GL_VERTEX_BINDING_BUFFER:
+        case GL_VERTEX_BINDING_DIVISOR:
+        case GL_VERTEX_BINDING_OFFSET:
+        case GL_VERTEX_BINDING_STRIDE: {
+            if (index >= VertexArrayImpl::GetMaxVertexAttribBindings()) {
+                MG_State::pGLContext->RecordError(
+                    ErrorCode::InvalidValue,
+                    MakeUnique<GenericErrorInfo>("MG_Impl/GLImpl", __func__,
+                                                 "Vertex buffer binding index is out of range."));
+                return;
+            }
+            const auto& vao = MG_State::pGLContext->GetBoundVertexArray();
+            if (!vao) {
+                *data = 0;
+                return;
+            }
+            const auto& binding = vao->GetBindingPoint(index);
+            switch (target) {
+            case GL_VERTEX_BINDING_BUFFER:
+                *data = binding.Buffer ? static_cast<GLint>(binding.Buffer->GetExternalIndex()) : 0;
+                return;
+            case GL_VERTEX_BINDING_DIVISOR:
+                *data = static_cast<GLint>(binding.Divisor);
+                return;
+            case GL_VERTEX_BINDING_OFFSET:
+                *data = static_cast<GLint>(binding.Offset);
+                return;
+            default:
+                *data = static_cast<GLint>(binding.Stride);
+                return;
+            }
+        }
         case GL_IMAGE_BINDING_NAME:
         case GL_IMAGE_BINDING_LEVEL:
         case GL_IMAGE_BINDING_LAYERED:
@@ -1398,7 +1432,7 @@ namespace MobileGL::MG_Impl::GLImpl {
             *params = 0; // program-binary entrypoints are stubbed
             return;
         case GL_PROGRAM_PIPELINE_BINDING:
-            *params = 0; // program-pipeline entrypoints are stubbed
+            *params = static_cast<GLint>(MG_State::pGLContext->GetBoundProgramPipelineName());
             return;
         case GL_PROGRAM_POINT_SIZE:
             *params = MG_State::pGLContext->IsCapabilityEnabled(CapabilityInput::ProgramPointSize) ? GL_TRUE : GL_FALSE;
@@ -1738,20 +1772,22 @@ namespace MobileGL::MG_Impl::GLImpl {
             *params = vao ? static_cast<GLint>(vao->GetExternalIndex()) : 0;
             return;
         }
+        // The vertex buffer binding points are per-binding-index state, so the non-indexed getter
+        // has nothing to answer with (GL 4.6 core table 23.4).
+        case GL_VERTEX_BINDING_BUFFER:
         case GL_VERTEX_BINDING_DIVISOR:
-            *params = 0; // vertex-binding entrypoints are stubbed
-            return;
         case GL_VERTEX_BINDING_OFFSET:
-            *params = 0; // vertex-binding entrypoints are stubbed
-            return;
         case GL_VERTEX_BINDING_STRIDE:
-            *params = 0; // vertex-binding entrypoints are stubbed
+            RecordIndexedOnlyGetterError(__func__, pname);
             return;
         case GL_MAX_VERTEX_ATTRIB_RELATIVE_OFFSET:
-            *params = 0; // vertex-binding entrypoints are stubbed
+            *params = static_cast<GLint>(VertexArrayImpl::GetMaxVertexAttribRelativeOffset());
             return;
         case GL_MAX_VERTEX_ATTRIB_BINDINGS:
-            *params = 0; // vertex-binding entrypoints are stubbed
+            *params = static_cast<GLint>(VertexArrayImpl::GetMaxVertexAttribBindings());
+            return;
+        case GL_MAX_VERTEX_ATTRIB_STRIDE:
+            *params = static_cast<GLint>(VertexArrayImpl::GetMaxVertexAttribStride());
             return;
         case GL_VIEWPORT: {
             const auto& vp = MG_State::pGLContext->GetViewport();

MobileGL/MG_Impl/GLImpl/Program/GL_ProgramPipeline.cpp

@@ -0,0 +1,231 @@
+// MobileGL - MobileGL/MG_Impl/GLImpl/Program/GL_ProgramPipeline.cpp
+// Copyright (c) 2025-2026 MobileGL-Dev
+// Licensed under the GNU Lesser General Public License v3.0:
+//   https://www.gnu.org/licenses/gpl-3.0.txt
+//   https://www.gnu.org/licenses/lgpl-3.0.txt
+// SPDX-License-Identifier: LGPL-3.0-only
+// End of Source File Header
+
+#include "GL_ProgramPipeline.h"
+
+#include <MG_State/GLState/Core.h>
+#include <MG_State/GLState/ErrorState/ErrorInfo.h>
+#include <MG_Util/Converters/GLToStr/GLEnumConverter.h>
+
+namespace MobileGL::MG_Impl::GLImpl {
+    namespace {
+        void RecordPipelineError(ErrorCode code, const char* function, String message) {
+            MG_State::pGLContext->RecordError(
+                code, MakeUnique<GenericErrorInfo>("MG_Impl/GLImpl", function, Move(message)));
+        }
+
+        // A pipeline name only names an object once it has been bound or created; querying a
+        // reserved-but-unmaterialised name is INVALID_OPERATION (GL 4.6 core 7.4).
+        const SharedPtr<MG_State::GLState::ProgramPipelineObject>* TryGetPipeline(GLuint pipeline,
+                                                                                 const char* function) {
+            if (!MG_State::pGLContext->IsProgramPipelineObject(pipeline)) {
+                RecordPipelineError(ErrorCode::InvalidOperation, function,
+                                    std::format("Program pipeline {} does not exist.", pipeline));
+                return nullptr;
+            }
+            return &MG_State::pGLContext->GetProgramPipelineObject(pipeline);
+        }
+
+        Bool ValidatePipelineCount(GLsizei n, const char* function) {
+            if (n < 0) {
+                RecordPipelineError(ErrorCode::InvalidValue, function, "n must be non-negative.");
+                return false;
+            }
+            return true;
+        }
+
+        // GL 4.6 core table 7.1 maps each stage bit onto a shader stage.
+        Bool TryResolveStageBit(GLbitfield bit, ShaderStage& outStage) {
+            switch (bit) {
+            case GL_VERTEX_SHADER_BIT: outStage = ShaderStage::Vertex; return true;
+            case GL_TESS_CONTROL_SHADER_BIT: outStage = ShaderStage::TessControl; return true;
+            case GL_TESS_EVALUATION_SHADER_BIT: outStage = ShaderStage::TessEval; return true;
+            case GL_GEOMETRY_SHADER_BIT: outStage = ShaderStage::Geometry; return true;
+            case GL_FRAGMENT_SHADER_BIT: outStage = ShaderStage::Fragment; return true;
+            case GL_COMPUTE_SHADER_BIT: outStage = ShaderStage::Compute; return true;
+            default: return false;
+            }
+        }
+
+        constexpr GLbitfield kAllStageBits = GL_VERTEX_SHADER_BIT | GL_TESS_CONTROL_SHADER_BIT |
+                                            GL_TESS_EVALUATION_SHADER_BIT | GL_GEOMETRY_SHADER_BIT |
+                                            GL_FRAGMENT_SHADER_BIT | GL_COMPUTE_SHADER_BIT;
+    } // namespace
+
+    void GenProgramPipelines(GLsizei n, GLuint* pipelines) {
+        if (!ValidatePipelineCount(n, __func__)) return;
+        if (n == 0 || !pipelines) return;
+
+        static thread_local Vector<GLuint> names;
+        MG_State::pGLContext->GenProgramPipelineNames(static_cast<Uint>(n), names);
+        Memcpy(pipelines, names.data(), static_cast<SizeT>(n) * sizeof(GLuint));
+    }
+
+    void CreateProgramPipelines(GLsizei n, GLuint* pipelines) {
+        if (!ValidatePipelineCount(n, __func__)) return;
+        if (n == 0 || !pipelines) return;
+
+        static thread_local Vector<GLuint> names;
+        MG_State::pGLContext->GenProgramPipelineNames(static_cast<Uint>(n), names);
+        for (GLsizei i = 0; i < n; ++i) {
+            pipelines[i] = names[static_cast<SizeT>(i)];
+            MG_State::pGLContext->CreateProgramPipelineObject(names[static_cast<SizeT>(i)]);
+        }
+    }
+
+    void DeleteProgramPipelines(GLsizei n, const GLuint* pipelines) {
+        if (!ValidatePipelineCount(n, __func__)) return;
+        if (!pipelines) return;
+
+        for (GLsizei i = 0; i < n; ++i) {
+            // Deleting zero, an unknown name, or a name that was only reserved is silently ignored.
+            MG_State::pGLContext->MarkProgramPipelineForDeletion(pipelines[i]);
+        }
+    }
+
+    void BindProgramPipeline(GLuint pipeline) {
+        if (pipeline != 0 && !MG_State::pGLContext->ValidateProgramPipelineName(pipeline)) {
+            RecordPipelineError(ErrorCode::InvalidOperation, __func__,
+                                std::format("Program pipeline name {} is not valid.", pipeline));
+            return;
+        }
+        MG_State::pGLContext->BindProgramPipelineObject(pipeline);
+    }
+
+    GLboolean IsProgramPipeline(GLuint pipeline) {
+        return MG_State::pGLContext->IsProgramPipelineObject(pipeline) ? GL_TRUE : GL_FALSE;
+    }
+
+    void GetProgramPipelineiv(GLuint pipeline, GLenum pname, GLint* params) {
+        const auto* pipelineObject = TryGetPipeline(pipeline, __func__);
+        if (!pipelineObject || !params) return;
+
+        const auto stageProgramName = [&](ShaderStage stage) -> GLint {
+            const auto& program = (*pipelineObject)->GetStageProgram(stage);
+            return program ? static_cast<GLint>(program->GetExternalIndex()) : 0;
+        };
+
+        switch (pname) {
+        case GL_ACTIVE_PROGRAM: {
+            const auto& active = (*pipelineObject)->GetActiveProgram();
+            *params = active ? static_cast<GLint>(active->GetExternalIndex()) : 0;
+            break;
+        }
+        case GL_VERTEX_SHADER: *params = stageProgramName(ShaderStage::Vertex); break;
+        case GL_TESS_CONTROL_SHADER: *params = stageProgramName(ShaderStage::TessControl); break;
+        case GL_TESS_EVALUATION_SHADER: *params = stageProgramName(ShaderStage::TessEval); break;
+        case GL_GEOMETRY_SHADER: *params = stageProgramName(ShaderStage::Geometry); break;
+        case GL_FRAGMENT_SHADER: *params = stageProgramName(ShaderStage::Fragment); break;
+        case GL_COMPUTE_SHADER: *params = stageProgramName(ShaderStage::Compute); break;
+        case GL_VALIDATE_STATUS: *params = (*pipelineObject)->GetValidateStatus() ? GL_TRUE : GL_FALSE; break;
+        case GL_INFO_LOG_LENGTH: {
+            // GL counts the null terminator, and reports 0 rather than 1 for an empty log.
+            const auto& log = (*pipelineObject)->GetInfoLog();
+            *params = log.empty() ? 0 : static_cast<GLint>(log.length()) + 1;
+            break;
+        }
+        default:
+            RecordPipelineError(ErrorCode::InvalidEnum, __func__,
+                                std::format("pname {} is not a program pipeline parameter.",
+                                            MG_Util::ConvertGLEnumToString(pname)));
+            break;
+        }
+    }
+
+    void GetProgramPipelineInfoLog(GLuint pipeline, GLsizei bufSize, GLsizei* length, GLchar* infoLog) {
+        const auto* pipelineObject = TryGetPipeline(pipeline, __func__);
+        if (!pipelineObject) return;
+        if (bufSize < 0) {
+            RecordPipelineError(ErrorCode::InvalidValue, __func__, "bufSize must be non-negative.");
+            return;
+        }
+        if (bufSize == 0 || !infoLog) {
+            if (length) *length = 0;
+            return;
+        }
+
+        const auto& log = (*pipelineObject)->GetInfoLog();
+        const auto copied = std::min<GLsizei>(bufSize - 1, static_cast<GLsizei>(log.length()));
+        if (copied > 0) Memcpy(infoLog, log.data(), static_cast<SizeT>(copied));
+        infoLog[copied] = '\0';
+        if (length) *length = copied;
+    }
+
+    void UseProgramStages(GLuint pipeline, GLbitfield stages, GLuint program) {
+        if (stages != GL_ALL_SHADER_BITS && (stages & ~kAllStageBits) != 0) {
+            RecordPipelineError(ErrorCode::InvalidValue, __func__, "stages names a bit that is not a shader stage.");
+            return;
+        }
+        const auto* pipelineObject = TryGetPipeline(pipeline, __func__);
+        if (!pipelineObject) return;
+
+        SharedPtr<MG_State::GLState::ProgramObject> programObject;
+        if (program != 0) {
+            if (!MG_State::pGLContext->ValidateProgramName(program)) {
+                RecordPipelineError(ErrorCode::InvalidValue, __func__,
+                                    std::format("{} is not the name of a program object.", program));
+                return;
+            }
+            programObject = MG_State::pGLContext->GetProgramObject(program);
+            if (!programObject) {
+                RecordPipelineError(ErrorCode::InvalidValue, __func__,
+                                    std::format("{} is not the name of a program object.", program));
+                return;
+            }
+            if (!programObject->GetLinkStatus()) {
+                RecordPipelineError(ErrorCode::InvalidOperation, __func__,
+                                    std::format("Program {} has not been linked successfully.", program));
+                return;
+            }
+        }
+
+        const GLbitfield selected = stages == GL_ALL_SHADER_BITS ? kAllStageBits : stages;
+        for (GLbitfield bit = 1; bit != 0 && bit <= kAllStageBits; bit <<= 1) {
+            if ((selected & bit) == 0) continue;
+            ShaderStage stage = ShaderStage::Unknown;
+            if (!TryResolveStageBit(bit, stage)) continue;
+            // program == 0 clears the stage, which is what a null program reference means here.
+            (*pipelineObject)->SetStageProgram(stage, programObject);
+        }
+    }
+
+    void ActiveShaderProgram(GLuint pipeline, GLuint program) {
+        const auto* pipelineObject = TryGetPipeline(pipeline, __func__);
+        if (!pipelineObject) return;
+
+        if (program == 0) {
+            (*pipelineObject)->SetActiveProgram(nullptr);
+            return;
+        }
+        if (!MG_State::pGLContext->ValidateProgramName(program)) {
+            RecordPipelineError(ErrorCode::InvalidValue, __func__,
+                                std::format("{} is not the name of a program object.", program));
+            return;
+        }
+        auto programObject = MG_State::pGLContext->GetProgramObject(program);
+        if (!programObject) {
+            RecordPipelineError(ErrorCode::InvalidValue, __func__,
+                                std::format("{} is not the name of a program object.", program));
+            return;
+        }
+        if (!programObject->GetLinkStatus()) {
+            RecordPipelineError(ErrorCode::InvalidOperation, __func__,
+                                std::format("Program {} has not been linked successfully.", program));
+            return;
+        }
+        (*pipelineObject)->SetActiveProgram(programObject);
+    }
+
+    void ValidateProgramPipeline(GLuint pipeline) {
+        const auto* pipelineObject = TryGetPipeline(pipeline, __func__);
+        if (!pipelineObject) return;
+        // Nothing here can fail today: MobileGL links each stage program on its own, so there is no
+        // cross-stage interface to re-check at validation time. The log stays empty, which GL allows.
+        (*pipelineObject)->SetValidateStatus(true);
+    }
+} // namespace MobileGL::MG_Impl::GLImpl

MobileGL/MG_Impl/GLImpl/Program/GL_ProgramPipeline.h

@@ -0,0 +1,23 @@
+// MobileGL - MobileGL/MG_Impl/GLImpl/Program/GL_ProgramPipeline.h
+// Copyright (c) 2025-2026 MobileGL-Dev
+// Licensed under the GNU Lesser General Public License v3.0:
+//   https://www.gnu.org/licenses/gpl-3.0.txt
+//   https://www.gnu.org/licenses/lgpl-3.0.txt
+// SPDX-License-Identifier: LGPL-3.0-only
+// End of Source File Header
+
+#pragma once
+#include <Includes.h>
+
+namespace MobileGL::MG_Impl::GLImpl {
+    void GenProgramPipelines(GLsizei n, GLuint* pipelines);
+    void CreateProgramPipelines(GLsizei n, GLuint* pipelines);
+    void DeleteProgramPipelines(GLsizei n, const GLuint* pipelines);
+    void BindProgramPipeline(GLuint pipeline);
+    GLboolean IsProgramPipeline(GLuint pipeline);
+    void GetProgramPipelineiv(GLuint pipeline, GLenum pname, GLint* params);
+    void GetProgramPipelineInfoLog(GLuint pipeline, GLsizei bufSize, GLsizei* length, GLchar* infoLog);
+    void UseProgramStages(GLuint pipeline, GLbitfield stages, GLuint program);
+    void ActiveShaderProgram(GLuint pipeline, GLuint program);
+    void ValidateProgramPipeline(GLuint pipeline);
+} // namespace MobileGL::MG_Impl::GLImpl

MobileGL/MG_Impl/GLImpl/Query/GL_Query.cpp

@@ -62,6 +62,34 @@ namespace MobileGL::MG_Impl::GLImpl {
                                               MakeUnique<GenericErrorInfo>("MG_Impl/GLImpl", function, message));
         }
 
+        // The by-buffer query getters write the result into a buffer object instead of client
+        // memory. Everything about the query itself - the name, whether it is still active, the
+        // parameter - is checked by GetQueryObjectValue; what is left is the destination, so this
+        // resolves the buffer and confirms the write lands inside it (GL 4.6 core 4.2.1).
+        Bool ResolveQueryResultDestination(GLuint buffer, GLintptr offset, SizeT writeSize, const char* function,
+                                           SharedPtr<MG_State::GLState::BufferObject>& outBuffer) {
+            if (offset < 0) {
+                RecordQueryError(ErrorCode::InvalidValue, function, "Offset cannot be negative.");
+                return false;
+            }
+            if (!MG_State::pGLContext->ValidateBufferObject(buffer)) {
+                RecordQueryError(ErrorCode::InvalidOperation, function, "Buffer object does not exist.");
+                return false;
+            }
+            auto bufferObject = MG_State::pGLContext->GetBufferObject(buffer);
+            if (!bufferObject) {
+                RecordQueryError(ErrorCode::InvalidOperation, function, "Buffer object does not exist.");
+                return false;
+            }
+            if (static_cast<SizeT>(offset) + writeSize > bufferObject->GetSize()) {
+                RecordQueryError(ErrorCode::InvalidOperation, function,
+                                 "The query result does not fit in the buffer object at this offset.");
+                return false;
+            }
+            outBuffer = bufferObject;
+            return true;
+        }
+
         // Callers must hold g_queryObjectsMutex.
         QueryObject* FindQueryObjectLocked(GLuint id) {
             const auto it = g_liveQueryObjects.find(id);
@@ -95,8 +123,13 @@ namespace MobileGL::MG_Impl::GLImpl {
         }
 
         // Shared GetQueryObject* implementation. Returns false when an error
-        // was recorded and no value should be written back.
+        // was recorded and no value should be written back. `outValueProduced`, when given,
-        Bool GetQueryObjectValue(GLuint id, GLenum pname, const char* function, Uint64& outValue) {
+        // additionally distinguishes "succeeded with a value" from "succeeded but the result is not
+        // ready" - the GL_QUERY_RESULT_NO_WAIT case, where GL_ARB_query_buffer_object says the
+        // destination is left alone rather than written with a placeholder.
+        Bool GetQueryObjectValue(GLuint id, GLenum pname, const char* function, Uint64& outValue,
+                                 Bool* outValueProduced = nullptr) {
+            if (outValueProduced) *outValueProduced = true;
             const std::lock_guard<std::mutex> lock(g_queryObjectsMutex);
             auto* queryObject = FindQueryObjectLocked(id);
             if (!queryObject) {
@@ -109,6 +142,41 @@ namespace MobileGL::MG_Impl::GLImpl {
             }
 
             switch (pname) {
+            case GL_QUERY_TARGET:
+                // The target a query was begun with (or created with, for glCreateQueries) - state
+                // the object has carried all along, GL 4.6 core table 23.35.
+                outValue = queryObject->target;
+                return true;
+            case GL_QUERY_RESULT_NO_WAIT: {
+                if (queryObject->resultCached) {
+                    outValue = queryObject->cachedResult;
+                    return true;
+                }
+                Uint64 result = 0;
+                const auto getQueryResult64 = MG_Backend::gBackendFunctionsTable.GL.GetQueryResult64;
+                if (queryObject->backendHandle && getQueryResult64 &&
+                    !getQueryResult64(queryObject->backendHandle, /*wait=*/false, &result)) {
+                    // Not ready. The whole point of the no-wait form is that the caller's
+                    // destination keeps whatever it already held.
+                    if (outValueProduced) *outValueProduced = false;
+                    outValue = 0;
+                    return true;
+                }
+                if (queryObject->target == GL_ANY_SAMPLES_PASSED ||
+                    queryObject->target == GL_ANY_SAMPLES_PASSED_CONSERVATIVE) {
+                    result = result != 0 ? 1 : 0;
+                }
+                if (queryObject->backendHandle) {
+                    if (const auto deleteBackendQuery = MG_Backend::gBackendFunctionsTable.GL.DeleteBackendQuery) {
+                        deleteBackendQuery(queryObject->backendHandle);
+                    }
+                    queryObject->backendHandle = nullptr;
+                }
+                queryObject->cachedResult = result;
+                queryObject->resultCached = true;
+                outValue = result;
+                return true;
+            }
             case GL_QUERY_RESULT_AVAILABLE: {
                 if (queryObject->resultCached || !queryObject->backendHandle) {
                     outValue = 1;
@@ -160,6 +228,21 @@ namespace MobileGL::MG_Impl::GLImpl {
                 return false;
             }
         }
+
+        template <typename T>
+        void GetQueryBufferObject(GLuint id, GLuint buffer, GLenum pname, GLintptr offset, const char* function) {
+            SharedPtr<MG_State::GLState::BufferObject> bufferObject;
+            if (!ResolveQueryResultDestination(buffer, offset, sizeof(T), function, bufferObject)) return;
+
+            Uint64 value = 0;
+            Bool valueProduced = false;
+            if (!GetQueryObjectValue(id, pname, function, value, &valueProduced)) return;
+            // GL_QUERY_RESULT_NO_WAIT on a result that has not landed writes nothing at all.
+            if (!valueProduced) return;
+
+            const T narrowed = static_cast<T>(value);
+            bufferObject->UploadSubData({const_cast<T*>(&narrowed), sizeof(T)}, static_cast<SizeT>(offset));
+        }
     } // namespace
 
     void GenQueries(GLsizei n, GLuint* ids) {
@@ -475,9 +558,26 @@ namespace MobileGL::MG_Impl::GLImpl {
         }
     }
 
+    void GetQueryBufferObjectiv(GLuint id, GLuint buffer, GLenum pname, GLintptr offset) {
+        GetQueryBufferObject<GLint>(id, buffer, pname, offset, __FUNCTION__);
+    }
+
+    void GetQueryBufferObjectuiv(GLuint id, GLuint buffer, GLenum pname, GLintptr offset) {
+        GetQueryBufferObject<GLuint>(id, buffer, pname, offset, __FUNCTION__);
+    }
+
+    void GetQueryBufferObjecti64v(GLuint id, GLuint buffer, GLenum pname, GLintptr offset) {
+        GetQueryBufferObject<GLint64>(id, buffer, pname, offset, __FUNCTION__);
+    }
+
+    void GetQueryBufferObjectui64v(GLuint id, GLuint buffer, GLenum pname, GLintptr offset) {
+        GetQueryBufferObject<GLuint64>(id, buffer, pname, offset, __FUNCTION__);
+    }
+
     void GetQueryObjectiv(GLuint id, GLenum pname, GLint* params) {
         Uint64 value = 0;
-        if (!GetQueryObjectValue(id, pname, __FUNCTION__, value) || !params) {
+        Bool valueProduced = false;
+        if (!GetQueryObjectValue(id, pname, __FUNCTION__, value, &valueProduced) || !valueProduced || !params) {
             return;
         }
         constexpr Uint64 kMaxInt = static_cast<Uint64>(INT_MAX);
@@ -486,7 +586,8 @@ namespace MobileGL::MG_Impl::GLImpl {
 
     void GetQueryObjectuiv(GLuint id, GLenum pname, GLuint* params) {
         Uint64 value = 0;
-        if (!GetQueryObjectValue(id, pname, __FUNCTION__, value) || !params) {
+        Bool valueProduced = false;
+        if (!GetQueryObjectValue(id, pname, __FUNCTION__, value, &valueProduced) || !valueProduced || !params) {
             return;
         }
         *params = static_cast<GLuint>(value & 0xFFFFFFFFull);
@@ -494,7 +595,8 @@ namespace MobileGL::MG_Impl::GLImpl {
 
     void GetQueryObjecti64v(GLuint id, GLenum pname, GLint64* params) {
         Uint64 value = 0;
-        if (!GetQueryObjectValue(id, pname, __FUNCTION__, value) || !params) {
+        Bool valueProduced = false;
+        if (!GetQueryObjectValue(id, pname, __FUNCTION__, value, &valueProduced) || !valueProduced || !params) {
             return;
         }
         *params = static_cast<GLint64>(value);
@@ -502,7 +604,8 @@ namespace MobileGL::MG_Impl::GLImpl {
 
     void GetQueryObjectui64v(GLuint id, GLenum pname, GLuint64* params) {
         Uint64 value = 0;
-        if (!GetQueryObjectValue(id, pname, __FUNCTION__, value) || !params) {
+        Bool valueProduced = false;
+        if (!GetQueryObjectValue(id, pname, __FUNCTION__, value, &valueProduced) || !valueProduced || !params) {
             return;
         }
         *params = static_cast<GLuint64>(value);

MobileGL/MG_Impl/GLImpl/Query/GL_Query.h

@@ -24,5 +24,9 @@ namespace MobileGL::MG_Impl::GLImpl {
     void GetQueryObjectuiv(GLuint id, GLenum pname, GLuint* params);
     void GetQueryObjecti64v(GLuint id, GLenum pname, GLint64* params);
     void GetQueryObjectui64v(GLuint id, GLenum pname, GLuint64* params);
+    void GetQueryBufferObjectiv(GLuint id, GLuint buffer, GLenum pname, GLintptr offset);
+    void GetQueryBufferObjectuiv(GLuint id, GLuint buffer, GLenum pname, GLintptr offset);
+    void GetQueryBufferObjecti64v(GLuint id, GLuint buffer, GLenum pname, GLintptr offset);
+    void GetQueryBufferObjectui64v(GLuint id, GLuint buffer, GLenum pname, GLintptr offset);
     void QueryCounter(GLuint id, GLenum target);
 } // namespace MobileGL::MG_Impl::GLImpl

MobileGL/MG_Impl/GLImpl/Sampler/GL_Sampler.cpp

@@ -28,6 +28,11 @@ namespace MobileGL::MG_Impl::GLImpl {
             case GL_TEXTURE_MAX_LOD:
             case GL_TEXTURE_LOD_BIAS:
                 return true;
+            // Four components, and GL puts no range on them - a border colour outside [0,1] is
+            // clamped when a fixed-point format is sampled, not rejected here. The scalar readers
+            // below would look at one component and invent an error.
+            case GL_TEXTURE_BORDER_COLOR:
+                return true;
             case GL_TEXTURE_MAX_ANISOTROPY_EXT:
                 if (ReadSamplerScalar(param, isFloat, isUnsignedInteger) >= 1.0f) return true;
                 MG_State::pGLContext->RecordError(
@@ -99,6 +104,20 @@ namespace MobileGL::MG_Impl::GLImpl {
         case GL_TEXTURE_COMPARE_FUNC:
             samplerObj->SetSamplerCompareFunc(MG_Util::ConvertGLEnumToSamplerCompareFunc(*(const GLint*)param));
             break;
+        case GL_TEXTURE_BORDER_COLOR:
+            // The only four-component sampler parameter: the caller's form decides which
+            // representation is authoritative, and SamplerObject keeps the other two in step.
+            if (isFloat) {
+                const auto* values = (const GLfloat*)param;
+                samplerObj->SetBorderColor(FloatVec4(values[0], values[1], values[2], values[3]));
+            } else if (isUnsignedInteger) {
+                const auto* values = (const GLuint*)param;
+                samplerObj->SetBorderColorUI(UintVec4(values[0], values[1], values[2], values[3]));
+            } else {
+                const auto* values = (const GLint*)param;
+                samplerObj->SetBorderColorI(IntVec4(values[0], values[1], values[2], values[3]));
+            }
+            break;
         default:
             MG_State::pGLContext->RecordError(ErrorCode::InvalidEnum,
                                               MakeUnique<GenericErrorInfo>("MG_Impl/GLImpl", "SetSamplerParam_State",
@@ -162,6 +181,31 @@ namespace MobileGL::MG_Impl::GLImpl {
         case GL_TEXTURE_COMPARE_FUNC:
             *(GLuint*)params = MG_Util::ConvertSamplerCompareFuncToGLEnum(samplerObj->GetSamplerCompareFunc());
             break;
+        case GL_TEXTURE_BORDER_COLOR: {
+            if (isFloat) {
+                const auto& color = samplerObj->GetBorderColor();
+                auto* out = (GLfloat*)params;
+                out[0] = color.x();
+                out[1] = color.y();
+                out[2] = color.z();
+                out[3] = color.w();
+            } else if (isUnsignedInteger) {
+                const auto& color = samplerObj->GetBorderColorUI();
+                auto* out = (GLuint*)params;
+                out[0] = color.x();
+                out[1] = color.y();
+                out[2] = color.z();
+                out[3] = color.w();
+            } else {
+                const auto& color = samplerObj->GetBorderColorI();
+                auto* out = (GLint*)params;
+                out[0] = color.x();
+                out[1] = color.y();
+                out[2] = color.z();
+                out[3] = color.w();
+            }
+            break;
+        }
         default:
             MG_State::pGLContext->RecordError(ErrorCode::InvalidEnum,
                                               MakeUnique<GenericErrorInfo>("MG_Impl/GLImpl", "GetSamplerParam_State",

MobileGL/MG_Impl/GLImpl/Texture/GL_Texture.h

@@ -11,6 +11,10 @@
 
 namespace MobileGL::MG_Impl::GLImpl {
     /* @INSERTION_POINT:FUNCTION_DECLARATION@ */
+    // The sized internal formats a buffer texture accepts (GL 4.6 core table 8.16). The buffer
+    // clears take the same list, so it is shared rather than written out twice.
+    Bool IsBufferTextureInternalFormat(GLenum internalformat);
+
     void ClearTexImage(GLuint texture, GLint level, GLenum format, GLenum type, const void* data);
     void ClearTexSubImage(GLuint texture, GLint level, GLint xoffset, GLint yoffset, GLint zoffset, GLsizei width,
                           GLsizei height, GLsizei depth, GLenum format, GLenum type, const void* data);
@@ -42,6 +46,7 @@ namespace MobileGL::MG_Impl::GLImpl {
     void GenerateTextureMipmap(GLuint texture);
     void BindTextureUnit(GLuint unit, GLuint texture);
     void GetTextureImage(GLuint texture, GLint level, GLenum format, GLenum type, GLsizei bufSize, void* pixels);
+    void GetCompressedTextureImage(GLuint texture, GLint level, GLsizei bufSize, void* pixels);
     void TexBufferRange(GLenum target, GLenum internalformat, GLuint buffer, GLintptr offset, GLsizeiptr size);
     void TextureBuffer(GLuint texture, GLenum internalformat, GLuint buffer);
     void TextureBufferRange(GLuint texture, GLenum internalformat, GLuint buffer, GLintptr offset, GLsizeiptr size);
@@ -101,6 +106,9 @@ namespace MobileGL::MG_Impl::GLImpl {
                            GLsizei width, GLsizei height);
     void CopyTexSubImage2D(GLenum target, GLint level, GLint xoffset, GLint yoffset, GLint x, GLint y, GLsizei width,
                            GLsizei height);
+    void CopyTextureSubImage1D(GLuint texture, GLint level, GLint xoffset, GLint x, GLint y, GLsizei width);
+    void CopyTextureSubImage3D(GLuint texture, GLint level, GLint xoffset, GLint yoffset, GLint zoffset, GLint x,
+                               GLint y, GLsizei width, GLsizei height);
     void CopyTextureSubImage2D(GLuint texture, GLint level, GLint xoffset, GLint yoffset, GLint x, GLint y,
                                GLsizei width, GLsizei height);
     void CopyTexSubImage1D(GLenum target, GLint level, GLint xoffset, GLint x, GLint y, GLsizei width);

MobileGL/MG_Impl/GLImpl/Texture/Validators.cpp

@@ -226,6 +226,9 @@ namespace MobileGL::MG_Impl::GLImpl::TextureImpl {
             case TextureInternalFormat::Depth24Stencil8:
             case TextureInternalFormat::Depth32FStencil8:
             case TextureInternalFormat::DepthStencil:
+            // Stencil-only is not a colour format either: a colour client format read against a
+            // STENCIL_INDEX8 texture has to be the same INVALID_OPERATION as against a depth one.
+            case TextureInternalFormat::StencilIndex8:
                 return true;
             default:
                 return false;

MobileGL/MG_Impl/GLImpl/VertexArray/GL_VertexArray.cpp

@@ -105,12 +105,45 @@ namespace MobileGL::MG_Impl::GLImpl {
             return pname == GL_CURRENT_VERTEX_ATTRIB;
         }
 
+        // The stride a pointer-style call gives its binding point: the argument when it is non-zero,
+        // otherwise the tightly packed element size (GL 4.6 core 10.3.2). A packed 2_10_10_10 or
+        // 10F_11F_11F attribute is one 32-bit word regardless of its component count.
+        static int EffectiveVertexStride(GLsizei stride, GLint size, GLenum type) {
+            if (stride != 0) return static_cast<int>(stride);
+            switch (type) {
+            case GL_INT_2_10_10_10_REV:
+            case GL_UNSIGNED_INT_2_10_10_10_REV:
+            case GL_UNSIGNED_INT_10F_11F_11F_REV:
+                return 4;
+            default:
+                break;
+            }
+            return static_cast<int>(size * MG_Util::GetGLTypeSize(type));
+        }
+
+        // glBindVertexBuffers / glVertexArrayVertexBuffers take a range of binding points, and a
+        // range that runs past the last one is INVALID_OPERATION rather than the INVALID_VALUE a
+        // single out-of-range index gets (GL 4.6 core 10.3.1).
+        static bool ValidateVertexBindingRange(GLuint first, GLsizei count, const char* funcName) {
+            if (count < 0) {
+                MG_State::pGLContext->RecordError(
+                    ErrorCode::InvalidValue,
+                    MakeUnique<GenericErrorInfo>("MG_Impl/GLImpl", funcName, "count must be non-negative."));
+                return false;
+            }
+            if (static_cast<Uint64>(first) + static_cast<Uint64>(count) >
+                VertexArrayImpl::GetMaxVertexAttribBindings()) {
+                MG_State::pGLContext->RecordError(
+                    ErrorCode::InvalidOperation,
+                    MakeUnique<GenericErrorInfo>("MG_Impl/GLImpl", funcName,
+                                                 "first + count exceeds GL_MAX_VERTEX_ATTRIB_BINDINGS."));
+                return false;
+            }
+            return true;
+        }
+
         static bool ValidateVertexBindingIndex(GLuint bindingindex, const char* funcName) {
-            // Bound by the same dynamic limit as attribute indices: the default attribute -> binding
+            if (bindingindex >= VertexArrayImpl::GetMaxVertexAttribBindings()) {
-            // mapping is the identity, so a binding point the backend cannot address as an attribute
-            // would resolve into an attribute the backend must then reject on every draw. Real drivers
-            // likewise report MAX_VERTEX_ATTRIB_BINDINGS == MAX_VERTEX_ATTRIBS.
-            if (bindingindex >= VertexArrayImpl::GetMaxVertexAttribs()) {
                 MG_State::pGLContext->RecordError(
                     ErrorCode::InvalidValue,
                     MakeUnique<GenericErrorInfo>("MG_Impl/GLImpl", funcName,
@@ -155,6 +188,17 @@ namespace MobileGL::MG_Impl::GLImpl {
 
     SharedPtr<MG_State::GLState::VertexArrayObject> GetNamedVertexArrayObject_State(GLuint vaobj,
                                                                                    const char* caller) {
+        // Name zero is not a vertex array object in a core profile: it names the default vertex
+        // array, which the by-name (direct state access) entry points never accept. MobileGL keeps a
+        // real object at index 0 for the compatibility paths, so the generic name validation below
+        // would otherwise let it through (GL 4.6 core 10.3.1).
+        if (vaobj == 0) {
+            MG_State::pGLContext->RecordError(
+                ErrorCode::InvalidOperation,
+                MakeUnique<GenericErrorInfo>("MG_Impl/GLImpl", caller,
+                                             "Vertex array name 0 is not a vertex array object."));
+            return nullptr;
+        }
         if (!VertexArrayImpl::ValidateVertexArrayName(vaobj)) return nullptr;
         if (!VertexArrayImpl::ValidateVertexArrayObject(vaobj)) return nullptr;
         return MG_State::pGLContext->GetVertexArrayObject(vaobj);
@@ -210,7 +254,7 @@ namespace MobileGL::MG_Impl::GLImpl {
 
         DataType dataType = MG_Util::ConvertGLEnumToDataType(type);
         // Integer path: never normalized, never BGRA/packed (the validator rejects those).
-        if (!VertexArrayImpl::ValidateVertexAttribFormat(index, size, dataType, false, stride, true)) return;
+        if (!VertexArrayImpl::ValidateVertexAttribFormat(index, size, type, dataType, false, stride, true)) return;
 
         auto& vao = MG_State::pGLContext->GetBoundVertexArray();
         if (!vao) {
@@ -227,6 +271,7 @@ namespace MobileGL::MG_Impl::GLImpl {
 
         vao->SetAttributeFormat(index, size, dataType, false, stride, offset, true, false);
         vao->BindAttributeBuffer(index, vbo);
+        vao->MirrorPointerIntoBinding(index, vbo, offset, EffectiveVertexStride(stride, size, type));
     }
 
     void VertexAttribPointer_State(GLuint index, GLint size, GLenum type, GLboolean normalized, GLsizei stride,
@@ -234,7 +279,7 @@ namespace MobileGL::MG_Impl::GLImpl {
         if (!VertexArrayImpl::ValidateVertexAttributeIndex(index)) return;
 
         DataType dataType = MG_Util::ConvertGLEnumToDataType(type);
-        if (!VertexArrayImpl::ValidateVertexAttribFormat(index, size, dataType, normalized == GL_TRUE, stride, false))
+        if (!VertexArrayImpl::ValidateVertexAttribFormat(index, size, type, dataType, normalized == GL_TRUE, stride, false))
             return;
 
         auto& vao = MG_State::pGLContext->GetBoundVertexArray();
@@ -256,6 +301,7 @@ namespace MobileGL::MG_Impl::GLImpl {
         const int effectiveSize = isBgra ? 4 : size;
         vao->SetAttributeFormat(index, effectiveSize, dataType, normalized, stride, offset, false, isBgra);
         vao->BindAttributeBuffer(index, vbo);
+        vao->MirrorPointerIntoBinding(index, vbo, offset, EffectiveVertexStride(stride, effectiveSize, type));
     }
 
     void BindVertexArray_State(GLuint array) {
@@ -359,6 +405,13 @@ namespace MobileGL::MG_Impl::GLImpl {
                 MakeUnique<GenericErrorInfo>("MG_Impl/GLImpl", caller, "offset and stride must be non-negative."));
             return;
         }
+        if (static_cast<Uint>(stride) > VertexArrayImpl::GetMaxVertexAttribStride()) {
+            MG_State::pGLContext->RecordError(
+                ErrorCode::InvalidValue,
+                MakeUnique<GenericErrorInfo>("MG_Impl/GLImpl", caller,
+                                             "stride exceeds GL_MAX_VERTEX_ATTRIB_STRIDE."));
+            return;
+        }
         auto bufferObject = GetVertexArrayBufferObject_State(buffer, caller);
         if (buffer != 0 && !bufferObject) return;
 
@@ -376,6 +429,7 @@ namespace MobileGL::MG_Impl::GLImpl {
                                         const GLintptr* offsets, const GLsizei* strides) {
         auto vao = GetNamedVertexArrayObject_State(vaobj, "VertexArrayVertexBuffers_State");
         if (!vao) return;
+        if (!ValidateVertexBindingRange(first, count, "VertexArrayVertexBuffers_State")) return;
         for (GLsizei i = 0; i < count; ++i) {
             if (!buffers) {
                 VertexBufferBinding_State(vao, first + i, 0, 0, 16, "VertexArrayVertexBuffers_State");
@@ -389,12 +443,36 @@ namespace MobileGL::MG_Impl::GLImpl {
     static void VertexAttribFormatSeparate_State(const SharedPtr<MG_State::GLState::VertexArrayObject>& vao,
                                                  GLuint attribindex, GLint size, GLenum type, GLboolean normalized,
                                                  GLuint relativeoffset, Bool isInteger, const char* caller) {
+        static_cast<void>(caller);
         if (!VertexArrayImpl::ValidateVertexAttributeIndex(attribindex)) return;
 
         DataType dataType = MG_Util::ConvertGLEnumToDataType(type);
-        if (!VertexArrayImpl::ValidateVertexAttribPointerParams(attribindex, size, dataType, 0)) return;
+        // The separate-format entry points take the same size/type rules as the pointer ones,
+        // GL_BGRA included, so they need the full format validation rather than the pointer-only
+        // subset - that one reports GL_BGRA as an out-of-range size.
+        if (!VertexArrayImpl::ValidateVertexAttribFormat(attribindex, size, type, dataType, normalized == GL_TRUE, 0,
+                                                         isInteger))
+            return;
+        if (!VertexArrayImpl::ValidateVertexAttribRelativeOffset(relativeoffset)) return;
 
-        vao->SetAttributeFormatSeparate(attribindex, size, dataType, normalized, isInteger, relativeoffset);
+        const Bool isBgra = (size == static_cast<GLint>(GL_BGRA));
+        vao->SetAttributeFormatSeparate(attribindex, isBgra ? 4 : size, dataType, normalized, isInteger,
+                                        relativeoffset, isBgra);
+    }
+
+    // The long (64-bit) attribute format. MobileGL has no 64-bit vertex attributes, so nothing is
+    // recorded; what the entry point owes the application is the parameter validation, which is
+    // observable through glGetError regardless of whether the format could be used in a draw.
+    static void VertexAttribLFormatSeparate_State(GLuint attribindex, GLint size, GLenum type,
+                                                  GLuint relativeoffset) {
+        if (!VertexArrayImpl::ValidateVertexAttributeIndex(attribindex)) return;
+        if (!VertexArrayImpl::ValidateVertexAttribLFormat(attribindex, size, type)) return;
+        if (!VertexArrayImpl::ValidateVertexAttribRelativeOffset(relativeoffset)) return;
+
+        MG_State::pGLContext->RecordError(
+            ErrorCode::InvalidOperation,
+            MakeUnique<GenericErrorInfo>("MG_Impl/GLImpl", "VertexAttribLFormat",
+                                         "64-bit vertex attributes are not supported."));
     }
 
     void VertexArrayAttribFormat_State(GLuint vaobj, GLuint attribindex, GLint size, GLenum type,
@@ -1153,6 +1231,7 @@ namespace MobileGL::MG_Impl::GLImpl {
                            const GLsizei* strides) {
         auto vao = GetBoundVertexArrayOrError("BindVertexBuffers");
         if (!vao) return;
+        if (!ValidateVertexBindingRange(first, count, "BindVertexBuffers")) return;
         for (GLsizei i = 0; i < count; ++i) {
             if (!buffers) {
                 VertexBufferBinding_State(vao, first + i, 0, 0, 16, "BindVertexBuffers");
@@ -1177,6 +1256,18 @@ namespace MobileGL::MG_Impl::GLImpl {
                                          "VertexAttribIFormat");
     }
 
+    void VertexAttribLFormat(GLuint attribindex, GLint size, GLenum type, GLuint relativeoffset) {
+        auto vao = GetBoundVertexArrayOrError("VertexAttribLFormat");
+        if (!vao) return;
+        VertexAttribLFormatSeparate_State(attribindex, size, type, relativeoffset);
+    }
+
+    void VertexArrayAttribLFormat(GLuint vaobj, GLuint attribindex, GLint size, GLenum type, GLuint relativeoffset) {
+        auto vao = GetNamedVertexArrayObject_State(vaobj, "VertexArrayAttribLFormat");
+        if (!vao) return;
+        VertexAttribLFormatSeparate_State(attribindex, size, type, relativeoffset);
+    }
+
     void VertexAttribBinding(GLuint attribindex, GLuint bindingindex) {
         auto vao = GetBoundVertexArrayOrError("VertexAttribBinding");
         if (!vao) return;

MobileGL/MG_Impl/GLImpl/VertexArray/GL_VertexArray.h

@@ -98,6 +98,7 @@ namespace MobileGL::MG_Impl::GLImpl {
     void VertexArrayAttribFormat(GLuint vaobj, GLuint attribindex, GLint size, GLenum type, GLboolean normalized,
                                  GLuint relativeoffset);
     void VertexArrayAttribIFormat(GLuint vaobj, GLuint attribindex, GLint size, GLenum type, GLuint relativeoffset);
+    void VertexArrayAttribLFormat(GLuint vaobj, GLuint attribindex, GLint size, GLenum type, GLuint relativeoffset);
     void VertexArrayAttribBinding(GLuint vaobj, GLuint attribindex, GLuint bindingindex);
     void VertexArrayBindingDivisor(GLuint vaobj, GLuint bindingindex, GLuint divisor);
     void VertexArrayVertexBuffers(GLuint vaobj, GLuint first, GLsizei count, const GLuint* buffers,
@@ -107,6 +108,7 @@ namespace MobileGL::MG_Impl::GLImpl {
                            const GLsizei* strides);
     void VertexAttribFormat(GLuint attribindex, GLint size, GLenum type, GLboolean normalized, GLuint relativeoffset);
     void VertexAttribIFormat(GLuint attribindex, GLint size, GLenum type, GLuint relativeoffset);
+    void VertexAttribLFormat(GLuint attribindex, GLint size, GLenum type, GLuint relativeoffset);
     void VertexAttribBinding(GLuint attribindex, GLuint bindingindex);
     void VertexBindingDivisor(GLuint bindingindex, GLuint divisor);
     void VertexAttribDivisor(GLuint index, GLuint divisor);

MobileGL/MG_Impl/GLImpl/VertexArray/Validators.cpp

@@ -23,6 +23,18 @@ namespace MobileGL::MG_Impl::GLImpl::VertexArrayImpl {
         return std::min(static_cast<Uint>(backendLimit), capacity);
     }
 
+    Uint GetMaxVertexAttribBindings() {
+        return GetMaxVertexAttribs();
+    }
+
+    Uint GetMaxVertexAttribRelativeOffset() {
+        return 2047;
+    }
+
+    Uint GetMaxVertexAttribStride() {
+        return 2048;
+    }
+
     Bool ValidateVertexArrayName(Uint index) {
         Bool isValid = MG_State::pGLContext->ValidateVertexArrayName(index);
         if (!isValid) {
@@ -90,9 +102,31 @@ namespace MobileGL::MG_Impl::GLImpl::VertexArrayImpl {
         return true;
     }
 
-    Bool ValidateVertexAttribFormat(Uint index, GLint sizeRaw, DataType type, Bool normalized, Int stride,
+    Bool ValidateVertexAttribFormat(Uint index, GLint sizeRaw, GLenum glType, DataType type, Bool normalized,
-                                    Bool integerPath) {
+                                    Int stride, Bool integerPath) {
         constexpr const char* fn = "ValidateVertexAttribFormat";
+        // GL_UNSIGNED_INT_10F_11F_11F_REV is a three-component float-path-only packing that has no
+        // DataType of its own, so it has to be recognised by name before the conversion below turns
+        // it into Unknown and reports the wrong error (GL 4.6 core 10.3.2).
+        if (glType == GL_UNSIGNED_INT_10F_11F_11F_REV) {
+            if (integerPath) {
+                MG_State::pGLContext->RecordError(
+                    ErrorCode::InvalidEnum,
+                    MakeUnique<GenericErrorInfo>(
+                        "MG_Impl/GLImpl", fn,
+                        std::format("GL_UNSIGNED_INT_10F_11F_11F_REV is not an integer-path type (attribute {}).",
+                                    index)));
+                return false;
+            }
+            if (sizeRaw != 3) {
+                MG_State::pGLContext->RecordError(
+                    ErrorCode::InvalidOperation,
+                    MakeUnique<GenericErrorInfo>(
+                        "MG_Impl/GLImpl", fn,
+                        std::format("GL_UNSIGNED_INT_10F_11F_11F_REV requires size 3 (attribute {}).", index)));
+                return false;
+            }
+        }
         if (type == DataType::Unknown) {
             MG_State::pGLContext->RecordError(
                 ErrorCode::InvalidEnum,
@@ -170,4 +204,40 @@ namespace MobileGL::MG_Impl::GLImpl::VertexArrayImpl {
         }
         return true;
     }
+
+    Bool ValidateVertexAttribLFormat(Uint index, GLint size, GLenum type) {
+        constexpr const char* fn = "ValidateVertexAttribLFormat";
+        if (size < 1 || size > 4) {
+            MG_State::pGLContext->RecordError(
+                ErrorCode::InvalidValue,
+                MakeUnique<GenericErrorInfo>(
+                    "MG_Impl/GLImpl", fn,
+                    std::format("Invalid size {} for attribute {}. Must be 1-4.", size, index)));
+            return false;
+        }
+        // GL 4.6 core 10.3.2: the long form takes GL_DOUBLE and nothing else.
+        if (type != GL_DOUBLE) {
+            MG_State::pGLContext->RecordError(
+                ErrorCode::InvalidEnum,
+                MakeUnique<GenericErrorInfo>(
+                    "MG_Impl/GLImpl", fn,
+                    std::format("Type 0x{:X} is not GL_DOUBLE (attribute {}).", type, index)));
+            return false;
+        }
+        return true;
+    }
+
+    Bool ValidateVertexAttribRelativeOffset(Uint relativeOffset) {
+        const Uint limit = GetMaxVertexAttribRelativeOffset();
+        if (relativeOffset > limit) {
+            MG_State::pGLContext->RecordError(
+                ErrorCode::InvalidValue,
+                MakeUnique<GenericErrorInfo>(
+                    "MG_Impl/GLImpl", "ValidateVertexAttribRelativeOffset",
+                    std::format("relativeoffset {} exceeds GL_MAX_VERTEX_ATTRIB_RELATIVE_OFFSET ({}).", relativeOffset,
+                                limit)));
+            return false;
+        }
+        return true;
+    }
 } // namespace MobileGL::MG_Impl::GLImpl::VertexArrayImpl

MobileGL/MG_Impl/GLImpl/VertexArray/Validators.h

@@ -15,6 +15,20 @@ namespace MobileGL::MG_Impl::GLImpl::VertexArrayImpl {
     // capacity). Falls back to the capacity when no backend is active (unit tests).
     Uint GetMaxVertexAttribs();
 
+    // GL_MAX_VERTEX_ATTRIB_BINDINGS. The default attribute -> binding mapping is the identity, so a
+    // binding point that cannot also be an attribute index would resolve into an attribute the
+    // backend has to reject on every draw; real drivers report the two limits equal as well.
+    Uint GetMaxVertexAttribBindings();
+
+    // GL_MAX_VERTEX_ATTRIB_RELATIVE_OFFSET. The relative offset is folded into the resolved
+    // attribute offset in the frontend and never reaches a backend limit, so this is the value the
+    // spec requires an implementation to support at minimum (GL 4.6 core table 23.63).
+    Uint GetMaxVertexAttribRelativeOffset();
+
+    // GL_MAX_VERTEX_ATTRIB_STRIDE. Like the relative offset above, the stride never reaches a
+    // backend limit of its own, so this is the spec minimum (GL 4.6 core table 23.63).
+    Uint GetMaxVertexAttribStride();
+
     Bool ValidateVertexArrayName(Uint index);
     Bool ValidateVertexArrayObject(Uint index);
     Bool ValidateVertexAttributeIndex(Uint index);
@@ -22,6 +36,13 @@ namespace MobileGL::MG_Impl::GLImpl::VertexArrayImpl {
     // Full glVertexAttribPointer / glVertexAttribIPointer format validation, including the packed
     // 2_10_10_10 types and GL_BGRA size. sizeRaw is the untranslated GL size (possibly GL_BGRA);
     // integerPath selects the glVertexAttribIPointer rules.
-    Bool ValidateVertexAttribFormat(Uint index, GLint sizeRaw, DataType type, Bool normalized, Int stride,
+    Bool ValidateVertexAttribFormat(Uint index, GLint sizeRaw, GLenum glType, DataType type, Bool normalized,
-                                    Bool integerPath);
+                                    Int stride, Bool integerPath);
+    // glVertexAttribLFormat / glVertexArrayAttribLFormat: the only accepted type is GL_DOUBLE and
+    // the size range is 1-4 (GL_BGRA is a float-path size). Separate from the function above
+    // because the long path shares none of its type or size rules.
+    Bool ValidateVertexAttribLFormat(Uint index, GLint size, GLenum type);
+    // Shared by every *Format entry point: INVALID_VALUE once relativeoffset leaves the range the
+    // implementation advertises.
+    Bool ValidateVertexAttribRelativeOffset(Uint relativeOffset);
 } // namespace MobileGL::MG_Impl::GLImpl::VertexArrayImpl

MobileGL/MG_State/GLState/BufferState/BufferState.h

@@ -31,6 +31,9 @@ namespace MobileGL::MG_State::GLState {
         BindingSlot<BufferObject>& GetBindingSlot(BufferTarget target);
         // For glBindBufferBase / glBindBufferRange
         BindingSlotRange1D<BufferObject>& GetBindingPoint(BufferTarget target, Uint index);
+        const BindingSlotRange1D<BufferObject>& GetBindingPoint(BufferTarget target, Uint index) const {
+            return const_cast<BufferState*>(this)->GetBindingPoint(target, index);
+        }
         constexpr SizeT GetBindingPointCount(const BufferTarget target) const {
             auto it = std::find(BufferBindPointTargets.begin(), BufferBindPointTargets.end(), target);
             auto index = std::distance(BufferBindPointTargets.begin(), it);

MobileGL/MG_State/GLState/Core.cpp

@@ -802,6 +802,53 @@ namespace MobileGL::MG_State {
                 m_transformFeedbackObjects[id] = {};
             }
         }
+        // Program pipeline
+        void GLContext::GenProgramPipelineNames(Uint number, Vector<Uint>& pipelines) {
+            pipelines.resize(number);
+            // Names only: glIsProgramPipeline must answer GL_FALSE until one is bound or created.
+            m_programPipelineNames.Generate(number, pipelines.data());
+        }
+
+        void GLContext::CreateProgramPipelineObject(Uint index) {
+            m_programPipelines[index] = MakeShared<ProgramPipelineObject>(index);
+        }
+
+        Bool GLContext::ValidateProgramPipelineName(Uint index) const {
+            return index == 0 || m_programPipelineNames.IsValid(index);
+        }
+
+        Bool GLContext::IsProgramPipelineObject(Uint index) const {
+            if (index == 0 || !m_programPipelineNames.IsValid(index)) return false;
+            return m_programPipelines.find(index) != m_programPipelines.end();
+        }
+
+        void GLContext::BindProgramPipelineObject(Uint index) {
+            if (index != 0 && m_programPipelines.find(index) == m_programPipelines.end()) {
+                // First bind is what turns a reserved name into an object.
+                m_programPipelines[index] = MakeShared<ProgramPipelineObject>(index);
+            }
+            m_boundProgramPipeline = index;
+        }
+
+        void GLContext::MarkProgramPipelineForDeletion(Uint index) {
+            if (index == 0 || !m_programPipelineNames.IsValid(index)) return;
+            if (index == m_boundProgramPipeline) {
+                m_boundProgramPipeline = 0;
+            }
+            m_programPipelines.erase(index);
+            m_programPipelineNames.Delete(index);
+        }
+
+        const SharedPtr<ProgramPipelineObject>& GLContext::GetProgramPipelineObject(Uint index) const {
+            static const SharedPtr<ProgramPipelineObject> kNone;
+            const auto it = m_programPipelines.find(index);
+            return it == m_programPipelines.end() ? kNone : it->second;
+        }
+
+        const SharedPtr<ProgramPipelineObject>& GLContext::GetBoundProgramPipeline() const {
+            return GetProgramPipelineObject(m_boundProgramPipeline);
+        }
+
 
         Bool GLContext::ValidateTransformFeedbackName(Uint index) const {
             return index == 0 || m_transformFeedbackNames.IsValid(index);
@@ -842,6 +889,64 @@ namespace MobileGL::MG_State {
             const auto it = m_transformFeedbackObjects.find(index);
             return it != m_transformFeedbackObjects.end() && it->second.hasCompletedSpan;
         }
+
+        void GLContext::CreateTransformFeedbackObject(Uint index) {
+            // glCreateTransformFeedbacks has no bind step to infer existence from, so the name it
+            // hands out is already the name of an object (GL 4.6 core 13.2.1).
+            m_transformFeedbackObjects[index] = {};
+            m_transformFeedbackObjects[index].everBound = true;
+        }
+
+        Bool GLContext::IsNamedTransformFeedbackActive(Uint index) const {
+            if (index == m_boundTransformFeedback) return m_transformFeedbackActive;
+            const auto it = m_transformFeedbackObjects.find(index);
+            return it != m_transformFeedbackObjects.end() && it->second.active;
+        }
+
+        Bool GLContext::IsNamedTransformFeedbackPaused(Uint index) const {
+            if (index == m_boundTransformFeedback) return m_transformFeedbackPaused;
+            const auto it = m_transformFeedbackObjects.find(index);
+            return it != m_transformFeedbackObjects.end() && it->second.paused;
+        }
+
+        NamedTransformFeedbackBinding GLContext::GetNamedTransformFeedbackBinding(Uint index, Uint bufferIndex) const {
+            NamedTransformFeedbackBinding result;
+            if (bufferIndex >= MAX_TRANSFORM_FEEDBACK_BUFFERS) return result;
+            // The bound object's capture bindings live in the context's own binding points, not in
+            // the saved copy - that one is only written when the object is swapped out.
+            if (index == m_boundTransformFeedback) {
+                const auto& point = m_bufferState.GetBindingPoint(BufferTarget::TransformFeedback, bufferIndex);
+                result.Buffer = point.GetBoundObject();
+                result.Range = point.GetRange();
+                result.HasExplicitRange = point.HasExplicitRange();
+                return result;
+            }
+            const auto it = m_transformFeedbackObjects.find(index);
+            if (it == m_transformFeedbackObjects.end()) return result;
+            const auto& saved = it->second.bindings[bufferIndex];
+            result.Buffer = saved.buffer;
+            result.Range = saved.range;
+            result.HasExplicitRange = saved.hasExplicitRange;
+            return result;
+        }
+
+        void GLContext::SetNamedTransformFeedbackBinding(Uint index, Uint bufferIndex,
+                                                         const SharedPtr<BufferObject>& buffer, Range1D range,
+                                                         Bool hasExplicitRange) {
+            if (bufferIndex >= MAX_TRANSFORM_FEEDBACK_BUFFERS) return;
+            if (index == m_boundTransformFeedback) {
+                auto& point = m_bufferState.GetBindingPoint(BufferTarget::TransformFeedback, bufferIndex);
+                point.Bind(buffer);
+                if (buffer && hasExplicitRange) {
+                    point.SetRange(range, true);
+                } else {
+                    point.ClearRange();
+                }
+                return;
+            }
+            auto& object = m_transformFeedbackObjects[index];
+            object.bindings[bufferIndex] = {buffer, range, hasExplicitRange};
+        }
     } // namespace GLState
 
     // Leak-at-exit storage; see GlobalObjects.cpp.

MobileGL/MG_State/GLState/Core.h

@@ -14,6 +14,7 @@
 #include "MG_State/GLState/RenderbufferState/RenderbufferState.h"
 #include "RenderState/RenderState.h"
 #include "ProgramState/ProgramState.h"
+#include "ProgramState/ProgramPipelineObject.h"
 #include "SamplerState/SamplerState.h"
 #include "TextureState/TextureState.h"
 #include "FramebufferState/FramebufferState.h"
@@ -45,6 +46,14 @@ namespace MobileGL {
             // translates the result into its own API call.
             VertexAttribTypeInfo ClassifyVertexAttribType(GLenum glType);
 
+            // One indexed capture binding of a transform feedback object, as the by-name queries
+            // report it. An empty Buffer means the binding point is unbound.
+            struct NamedTransformFeedbackBinding {
+                SharedPtr<BufferObject> Buffer;
+                Range1D Range{};
+                Bool HasExplicitRange = false;
+            };
+
             class GLContext {
             public:
                 GLContext() = default;
@@ -129,6 +138,19 @@ namespace MobileGL {
                 void UseProgram(Uint program);
                 const SharedPtr<ProgramObject>& GetCurrentProgram();
 
+                // Program pipeline (GL_ARB_separate_shader_objects, GL 4.6 core 7.4). Like queries
+                // and transform feedbacks, glGenProgramPipelines only RESERVES a name - the object
+                // appears on first bind - while glCreateProgramPipelines makes it immediately.
+                void GenProgramPipelineNames(Uint number, Vector<Uint>& pipelines);
+                void CreateProgramPipelineObject(Uint index);
+                Bool ValidateProgramPipelineName(Uint index) const;
+                Bool IsProgramPipelineObject(Uint index) const;
+                void BindProgramPipelineObject(Uint index);
+                void MarkProgramPipelineForDeletion(Uint index);
+                const SharedPtr<ProgramPipelineObject>& GetProgramPipelineObject(Uint index) const;
+                Uint GetBoundProgramPipelineName() const { return m_boundProgramPipeline; }
+                const SharedPtr<ProgramPipelineObject>& GetBoundProgramPipeline() const;
+
                 // RenderState
                 Uint GetRenderStateParametersVersion() const;
                 const RenderStateParameters& GetRenderStateParameters() const;
@@ -299,6 +321,17 @@ namespace MobileGL {
                 // cannot express: an empty completed span is legal and draws nothing.
                 Bool HasTransformFeedbackCompletedSpan(Uint index) const;
 
+                // The by-name (direct state access) view. A named object that happens to be the
+                // bound one is answered from the live copy, since that is where its state actually
+                // is until a bind swaps it out.
+                void CreateTransformFeedbackObject(Uint index);
+                Bool IsNamedTransformFeedbackActive(Uint index) const;
+                Bool IsNamedTransformFeedbackPaused(Uint index) const;
+                NamedTransformFeedbackBinding GetNamedTransformFeedbackBinding(Uint index, Uint bufferIndex) const;
+                void SetNamedTransformFeedbackBinding(Uint index, Uint bufferIndex,
+                                                      const SharedPtr<BufferObject>& buffer, Range1D range,
+                                                      Bool hasExplicitRange);
+
                 // Framebuffer
                 void GenFramebufferNames(Uint number, Vector<Uint>& framebuffers);
                 const SharedPtr<FramebufferObject>& GetFramebufferObject(Uint index);
@@ -371,6 +404,11 @@ namespace MobileGL {
                 UnorderedMap<Uint, TransformFeedbackObjectState> m_transformFeedbackObjects;
                 IndexGenerator<Uint> m_transformFeedbackNames;
                 Uint m_boundTransformFeedback = 0;
+                // Map membership IS object existence here: a pipeline has no stateful default
+                // object 0, so no everBound flag is needed.
+                UnorderedMap<Uint, SharedPtr<ProgramPipelineObject>> m_programPipelines;
+                IndexGenerator<Uint> m_programPipelineNames;
+                Uint m_boundProgramPipeline = 0;
                 TextureState m_textureState;
                 ProgramState m_programState;
                 RenderState m_renderState;

MobileGL/MG_State/GLState/FramebufferState/FramebufferObject.cpp

@@ -185,6 +185,20 @@ namespace MobileGL::MG_State::GLState {
         return m_externalIndex;
     }
 
+#define MOBILEGL_DEFINE_FRAMEBUFFER_DEFAULT_SETTER(name, member, type)                                                \
+    void FramebufferObject::Set##name(type value) {                                                                    \
+        if (member == value) return;                                                                                    \
+        member = value;                                                                                                 \
+        ++m_objectVersion;                                                                                              \
+    }
+
+    MOBILEGL_DEFINE_FRAMEBUFFER_DEFAULT_SETTER(DefaultWidth, m_defaultWidth, Int)
+    MOBILEGL_DEFINE_FRAMEBUFFER_DEFAULT_SETTER(DefaultHeight, m_defaultHeight, Int)
+    MOBILEGL_DEFINE_FRAMEBUFFER_DEFAULT_SETTER(DefaultLayers, m_defaultLayers, Int)
+    MOBILEGL_DEFINE_FRAMEBUFFER_DEFAULT_SETTER(DefaultSamples, m_defaultSamples, Int)
+    MOBILEGL_DEFINE_FRAMEBUFFER_DEFAULT_SETTER(DefaultFixedSampleLocations, m_defaultFixedSampleLocations, Bool)
+#undef MOBILEGL_DEFINE_FRAMEBUFFER_DEFAULT_SETTER
+
     void FramebufferObject::BumpAttachmentVersion(FramebufferAttachmentType type) {
         ++m_attachmentVersions[static_cast<SizeT>(type)];
         ++m_objectVersion;

MobileGL/MG_State/GLState/FramebufferState/FramebufferObject.h

@@ -129,6 +129,19 @@ namespace MobileGL {
             void SetReadBuffer(FramebufferAttachmentType buf);
             FramebufferAttachmentType GetReadBuffer() const { return m_readBuffer; }
 
+            // GL_ARB_framebuffer_no_attachments state (GL 4.6 core table 23.24). The shape a
+            // framebuffer with no attachments would rasterize at; all zero / FALSE until set.
+            Int GetDefaultWidth() const { return m_defaultWidth; }
+            Int GetDefaultHeight() const { return m_defaultHeight; }
+            Int GetDefaultLayers() const { return m_defaultLayers; }
+            Int GetDefaultSamples() const { return m_defaultSamples; }
+            Bool GetDefaultFixedSampleLocations() const { return m_defaultFixedSampleLocations; }
+            void SetDefaultWidth(Int value);
+            void SetDefaultHeight(Int value);
+            void SetDefaultLayers(Int value);
+            void SetDefaultSamples(Int value);
+            void SetDefaultFixedSampleLocations(Bool value);
+
             FramebufferAttachmentVersionArray GetAllFramebufferAttachmentVersions() const {
                 return m_attachmentVersions;
             }
@@ -148,6 +161,12 @@ namespace MobileGL {
             FramebufferAttachmentArray m_drawBuffers; // Probably no versioning needed for this, just check equality
             FramebufferAttachmentType m_readBuffer = FramebufferAttachmentType::None;
 
+            Int m_defaultWidth = 0;
+            Int m_defaultHeight = 0;
+            Int m_defaultLayers = 0;
+            Int m_defaultSamples = 0;
+            Bool m_defaultFixedSampleLocations = false;
+
             // This version will bump when draw/read buffer changes (by `glDrawBuffer(s)`/`glReadBuffer`)
             Uint16 m_objectVersion = 0;
         };

MobileGL/MG_State/GLState/ProgramState/ProgramObject.cpp

@@ -195,27 +195,42 @@ namespace MobileGL::MG_State::GLState {
             static constexpr GLenum kIntTypes[5] = {0, GL_INT, GL_INT_VEC2, GL_INT_VEC3, GL_INT_VEC4};
             static constexpr GLenum kUintTypes[5] = {0, GL_UNSIGNED_INT, GL_UNSIGNED_INT_VEC2, GL_UNSIGNED_INT_VEC3,
                                                      GL_UNSIGNED_INT_VEC4};
+            static constexpr GLenum kDoubleTypes[5] = {0, GL_DOUBLE, GL_DOUBLE_VEC2, GL_DOUBLE_VEC3,
+                                                      GL_DOUBLE_VEC4};
             if (type.isMatrix()) {
-                if (basic != glslang::EbtFloat) return false;
+                if (basic != glslang::EbtFloat && basic != glslang::EbtDouble) return false;
                 static constexpr GLenum kMatTypes[5][5] = {
                     {}, {},
                     {0, 0, GL_FLOAT_MAT2, GL_FLOAT_MAT2x3, GL_FLOAT_MAT2x4},
                     {0, 0, GL_FLOAT_MAT3x2, GL_FLOAT_MAT3, GL_FLOAT_MAT3x4},
                     {0, 0, GL_FLOAT_MAT4x2, GL_FLOAT_MAT4x3, GL_FLOAT_MAT4},
                 };
+                static constexpr GLenum kDoubleMatTypes[5][5] = {
+                    {}, {},
+                    {0, 0, GL_DOUBLE_MAT2, GL_DOUBLE_MAT2x3, GL_DOUBLE_MAT2x4},
+                    {0, 0, GL_DOUBLE_MAT3x2, GL_DOUBLE_MAT3, GL_DOUBLE_MAT3x4},
+                    {0, 0, GL_DOUBLE_MAT4x2, GL_DOUBLE_MAT4x3, GL_DOUBLE_MAT4},
+                };
                 if (columns < 2 || columns > 4 || components < 2 || components > 4) return false;
-                outType = kMatTypes[columns][components];
+                outType = basic == glslang::EbtDouble ? kDoubleMatTypes[columns][components]
+                                                      : kMatTypes[columns][components];
             } else if (components >= 1 && components <= 4) {
                 switch (basic) {
                 case glslang::EbtFloat: outType = kFloatTypes[components]; break;
                 case glslang::EbtInt: outType = kIntTypes[components]; break;
                 case glslang::EbtUint: outType = kUintTypes[components]; break;
+                // A double-typed varying is capturable like any other; rejecting it here reported
+                // the varying as "not an output of the vertex stage", which it plainly was.
+                case glslang::EbtDouble: outType = kDoubleTypes[components]; break;
                 default: return false;
                 }
             } else {
                 return false;
             }
-            outBytesPerElement = static_cast<Uint32>(columns * components) * 4u;
+            // GL 4.6 core 11.1.2.1: a double component occupies eight basic machine units, and
+            // counts as two components against the transform feedback limits.
+            const Uint32 bytesPerComponent = basic == glslang::EbtDouble ? 8u : 4u;
+            outBytesPerElement = static_cast<Uint32>(columns * components) * bytesPerComponent;
             return true;
         }
     } // namespace

MobileGL/MG_State/GLState/ProgramState/ProgramPipelineObject.h

@@ -0,0 +1,52 @@
+// MobileGL - MobileGL/MG_State/GLState/ProgramState/ProgramPipelineObject.h
+// Copyright (c) 2025-2026 MobileGL-Dev
+// Licensed under the GNU Lesser General Public License v3.0:
+//   https://www.gnu.org/licenses/gpl-3.0.txt
+//   https://www.gnu.org/licenses/lgpl-3.0.txt
+// SPDX-License-Identifier: LGPL-3.0-only
+// End of Source File Header
+
+#pragma once
+#include <Includes.h>
+#include "ProgramObject.h"
+
+namespace MobileGL {
+    namespace MG_State {
+        namespace GLState {
+            // A GL_ARB_separate_shader_objects program pipeline (GL 4.6 core 7.4): a set of
+            // per-stage program references plus the program glProgramUniform* addresses when no
+            // program object is in use.
+            class ProgramPipelineObject {
+            public:
+                explicit ProgramPipelineObject(Uint externalIndex) : m_externalIndex(externalIndex) {}
+
+                const SharedPtr<ProgramObject>& GetStageProgram(ShaderStage stage) const {
+                    return m_stagePrograms[static_cast<SizeT>(stage)];
+                }
+                void SetStageProgram(ShaderStage stage, const SharedPtr<ProgramObject>& program) {
+                    m_stagePrograms[static_cast<SizeT>(stage)] = program;
+                }
+
+                const SharedPtr<ProgramObject>& GetActiveProgram() const { return m_activeProgram; }
+                void SetActiveProgram(const SharedPtr<ProgramObject>& program) { m_activeProgram = program; }
+
+                // Distinct from ProgramObject's, which starts true: a pipeline that has never been
+                // validated must report GL_VALIDATE_STATUS as 0 (GL 4.6 core table 23.31).
+                Bool GetValidateStatus() const { return m_validateStatus; }
+                void SetValidateStatus(Bool value) { m_validateStatus = value; }
+
+                const String& GetInfoLog() const { return m_infoLog; }
+                void SetInfoLog(String log) { m_infoLog = Move(log); }
+
+                Uint GetExternalIndex() const { return m_externalIndex; }
+
+            private:
+                Array<SharedPtr<ProgramObject>, static_cast<SizeT>(ShaderStage::ShaderStageCount)> m_stagePrograms{};
+                SharedPtr<ProgramObject> m_activeProgram;
+                String m_infoLog;
+                const Uint m_externalIndex = 0;
+                Bool m_validateStatus = false;
+            };
+        } // namespace GLState
+    } // namespace MG_State
+} // namespace MobileGL

MobileGL/MG_State/GLState/SamplerState/SamplerObject.cpp

@@ -139,6 +139,61 @@ namespace MobileGL {
                 return m_samplerParameters.maxAnisotropy;
             }
 
+            // The three border-colour representations are kept in step so a getter of any form has
+            // an answer whichever form was written. Integer <-> float uses the plain value, matching
+            // what glTexParameterIiv/Iuiv mean: those forms are for integer texture formats, whose
+            // border components are the raw integers rather than a normalized fraction.
+            void SamplerObject::SetBorderColor(const FloatVec4& color) {
+                if (color == m_samplerParameters.borderColor) return;
+
+                m_samplerParameters.borderColor = color;
+                m_samplerParameters.borderColorI =
+                    IntVec4(static_cast<Int32>(color.x()), static_cast<Int32>(color.y()),
+                            static_cast<Int32>(color.z()), static_cast<Int32>(color.w()));
+                m_samplerParameters.borderColorUI =
+                    UintVec4(static_cast<Uint32>(color.x()), static_cast<Uint32>(color.y()),
+                             static_cast<Uint32>(color.z()), static_cast<Uint32>(color.w()));
+                ++m_version;
+            }
+
+            void SamplerObject::SetBorderColorI(const IntVec4& color) {
+                if (color == m_samplerParameters.borderColorI) return;
+
+                m_samplerParameters.borderColorI = color;
+                m_samplerParameters.borderColorUI =
+                    UintVec4(static_cast<Uint32>(color.x()), static_cast<Uint32>(color.y()),
+                             static_cast<Uint32>(color.z()), static_cast<Uint32>(color.w()));
+                m_samplerParameters.borderColor =
+                    FloatVec4(static_cast<Float>(color.x()), static_cast<Float>(color.y()),
+                              static_cast<Float>(color.z()), static_cast<Float>(color.w()));
+                ++m_version;
+            }
+
+            void SamplerObject::SetBorderColorUI(const UintVec4& color) {
+                if (color == m_samplerParameters.borderColorUI) return;
+
+                m_samplerParameters.borderColorUI = color;
+                m_samplerParameters.borderColorI =
+                    IntVec4(static_cast<Int32>(color.x()), static_cast<Int32>(color.y()),
+                            static_cast<Int32>(color.z()), static_cast<Int32>(color.w()));
+                m_samplerParameters.borderColor =
+                    FloatVec4(static_cast<Float>(color.x()), static_cast<Float>(color.y()),
+                              static_cast<Float>(color.z()), static_cast<Float>(color.w()));
+                ++m_version;
+            }
+
+            const FloatVec4& SamplerObject::GetBorderColor() const {
+                return m_samplerParameters.borderColor;
+            }
+
+            const IntVec4& SamplerObject::GetBorderColorI() const {
+                return m_samplerParameters.borderColorI;
+            }
+
+            const UintVec4& SamplerObject::GetBorderColorUI() const {
+                return m_samplerParameters.borderColorUI;
+            }
+
             SamplerCompareMode SamplerObject::GetCompareMode() const {
                 return m_samplerParameters.compareMode;
             }

MobileGL/MG_State/GLState/SamplerState/SamplerObject.h

@@ -8,6 +8,7 @@
 
 #pragma once
 #include <Includes.h>
+#include <MG_Util/Math/VectorTypes.h>
 
 namespace MobileGL {
     enum class SamplerFilterMode {
@@ -66,8 +67,18 @@ namespace MobileGL {
         Float maxLod = 1000.0f;
         Float lodBias = 0.0f;
         Float maxAnisotropy = 1.0f;
-        SamplerCompareFunc compareFunc = SamplerCompareFunc::Always;
+        // GL 4.6 core table 23.18 / GLES 3.2 table 21.16: TEXTURE_COMPARE_FUNC starts at LEQUAL,
+        // for both sampler objects and the sampler state a texture object carries.
+        SamplerCompareFunc compareFunc = SamplerCompareFunc::LessEqual;
         SamplerCompareMode compareMode = SamplerCompareMode::None;
+        // TEXTURE_BORDER_COLOR is sampler state (GL 4.6 core table 23.18), so it belongs here and
+        // not on the texture - a texture object reaches it through the sampler object it owns. The
+        // three representations are the float, integer and unsigned-integer forms glSamplerParameterfv,
+        // glSamplerParameterIiv and glSamplerParameterIuiv set; whichever is written last defines
+        // the colour and the other two follow it, so a getter always has an answer.
+        FloatVec4 borderColor = {0.0f, 0.0f, 0.0f, 0.0f};
+        IntVec4 borderColorI = {0, 0, 0, 0};
+        UintVec4 borderColorUI = {0, 0, 0, 0};
     };
 
     namespace MG_State {
@@ -87,6 +98,9 @@ namespace MobileGL {
                 void SetMaxAnisotropy(Float maxAnisotropy);
                 void SetSamplerCompareFunc(SamplerCompareFunc func);
                 void SetCompareMode(SamplerCompareMode mode);
+                void SetBorderColor(const FloatVec4& color);
+                void SetBorderColorI(const IntVec4& color);
+                void SetBorderColorUI(const UintVec4& color);
 
                 SamplerWrapMode GetWrapS() const;
                 SamplerWrapMode GetWrapT() const;
@@ -100,6 +114,9 @@ namespace MobileGL {
                 Float GetMaxAnisotropy() const;
                 SamplerCompareMode GetCompareMode() const;
                 SamplerCompareFunc GetSamplerCompareFunc() const;
+                const FloatVec4& GetBorderColor() const;
+                const IntVec4& GetBorderColorI() const;
+                const UintVec4& GetBorderColorUI() const;
                 Uint GetExternalIndex() const;
                 Uint16 GetVersion() const;
                 // Globally-unique, never-reused id for this sampler object's lifetime. Lets a

MobileGL/MG_State/GLState/TextureState/TextureObject.cpp

@@ -88,48 +88,40 @@ namespace MobileGL {
                 return m_externalIndex;
             }
 
+            // TEXTURE_BORDER_COLOR is sampler state, so it lives on the SamplerObject this texture
+            // owns rather than being duplicated here - a sampler object bound over the texture then
+            // supplies its own, exactly as GL says it should. The texture params version still moves
+            // on a write, because the DirectGLES texture sync memoises on it.
             const FloatVec4& TextureObjectBase::GetBorderColor() const {
-                return m_borderColor;
+                return m_sampler->GetBorderColor();
             }
 
             void TextureObjectBase::SetBorderColor(const FloatVec4& color) {
-                if (color == m_borderColor) return;
+                if (color == m_sampler->GetBorderColor()) return;
 
-                m_borderColor = color;
+                m_sampler->SetBorderColor(color);
-                m_borderColorI = IntVec4(static_cast<Int32>(color.x()), static_cast<Int32>(color.y()),
-                                         static_cast<Int32>(color.z()), static_cast<Int32>(color.w()));
-                m_borderColorUI = UintVec4(static_cast<Uint32>(color.x()), static_cast<Uint32>(color.y()),
-                                           static_cast<Uint32>(color.z()), static_cast<Uint32>(color.w()));
                 ++m_textureParamsVersion;
             }
 
             const IntVec4& TextureObjectBase::GetBorderColorI() const {
-                return m_borderColorI;
+                return m_sampler->GetBorderColorI();
             }
 
             void TextureObjectBase::SetBorderColorI(const IntVec4& color) {
-                if (color == m_borderColorI) return;
+                if (color == m_sampler->GetBorderColorI()) return;
 
-                m_borderColorI = color;
+                m_sampler->SetBorderColorI(color);
-                m_borderColorUI = UintVec4(static_cast<Uint32>(color.x()), static_cast<Uint32>(color.y()),
-                                           static_cast<Uint32>(color.z()), static_cast<Uint32>(color.w()));
-                m_borderColor = FloatVec4(static_cast<Float>(color.x()), static_cast<Float>(color.y()),
-                                          static_cast<Float>(color.z()), static_cast<Float>(color.w()));
                 ++m_textureParamsVersion;
             }
 
             const UintVec4& TextureObjectBase::GetBorderColorUI() const {
-                return m_borderColorUI;
+                return m_sampler->GetBorderColorUI();
             }
 
             void TextureObjectBase::SetBorderColorUI(const UintVec4& color) {
-                if (color == m_borderColorUI) return;
+                if (color == m_sampler->GetBorderColorUI()) return;
 
-                m_borderColorUI = color;
+                m_sampler->SetBorderColorUI(color);
-                m_borderColorI = IntVec4(static_cast<Int32>(color.x()), static_cast<Int32>(color.y()),
-                                         static_cast<Int32>(color.z()), static_cast<Int32>(color.w()));
-                m_borderColor = FloatVec4(static_cast<Float>(color.x()), static_cast<Float>(color.y()),
-                                          static_cast<Float>(color.z()), static_cast<Float>(color.w()));
                 ++m_textureParamsVersion;
             }
 

MobileGL/MG_State/GLState/TextureState/TextureObject.h

@@ -60,6 +60,10 @@ namespace MobileGL::MG_State::GLState {
         virtual Bool HasFixedSampleLocations() const = 0;
         virtual void SetFixedSampleLocations(Bool fixedSampleLocations) = 0;
         virtual Uint64 GetLifetimeId() const = 0;
+        // Which aspect of a packed depth/stencil texture a sampler reads (GL 4.6 core 8.10).
+        // DEPTH_COMPONENT until set, and meaningless for every other format.
+        virtual GLenum GetDepthStencilTextureMode() const = 0;
+        virtual void SetDepthStencilTextureMode(GLenum mode) = 0;
 
     protected:
         virtual Uint GetIndexOfTextureUploadTarget(TextureUploadTarget target) const = 0;
@@ -104,6 +108,8 @@ namespace MobileGL::MG_State::GLState {
         Bool HasFixedSampleLocations() const override;
         void SetFixedSampleLocations(Bool fixedSampleLocations) override;
         Uint64 GetLifetimeId() const override;
+        GLenum GetDepthStencilTextureMode() const override { return m_depthStencilTextureMode; }
+        void SetDepthStencilTextureMode(GLenum mode) override { m_depthStencilTextureMode = mode; }
 
     protected:
         static Uint64 AllocateLifetimeId();
@@ -113,9 +119,6 @@ namespace MobileGL::MG_State::GLState {
         const TextureTarget m_target = TextureTarget::Unknown;
         TextureInternalFormat m_internalFormat = TextureInternalFormat::Unknown;
         SharedPtr<SamplerObject> m_sampler = nullptr;
-        FloatVec4 m_borderColor = {0.0f, 0.0f, 0.0f, 0.0f};
-        IntVec4 m_borderColorI = {0, 0, 0, 0};
-        UintVec4 m_borderColorUI = {0, 0, 0, 0};
         Vec4<TextureSwizzleParam> m_swizzleParams = {TextureSwizzleParam::Red, TextureSwizzleParam::Green,
                                                      TextureSwizzleParam::Blue, TextureSwizzleParam::Alpha};
         UintVec2 m_levelRange = {0, 1000};
@@ -124,6 +127,7 @@ namespace MobileGL::MG_State::GLState {
         // Starts at 1 so a freshly-created backend resource (snapshot 0) never spuriously
         // matches before its first sync. Bumped only on dirty=true in MarkStorageDirty.
         Uint64 m_contentVersion = 1;
+        GLenum m_depthStencilTextureMode = GL_DEPTH_COMPONENT;
         Int m_samples = 0;
         Bool m_fixedSampleLocations = true;
     };

MobileGL/MG_State/GLState/VertexArrayState/VertexArrayObject.cpp

@@ -77,6 +77,26 @@ namespace MobileGL::MG_State::GLState {
         BumpAttributeFormatVersion(index);
     }
 
+    void VertexArrayObject::MirrorPointerIntoBinding(Uint index, const SharedPtr<BufferObject>& buffer, SizeT offset,
+                                                     int effectiveStride) {
+        if (index >= MAX_VERTEX_ATTRIBS || index >= MAX_VERTEX_ATTRIB_BINDINGS) return;
+
+        // glVertexAttribPointer is defined in terms of the binding model (GL 4.6 core 10.3.2): it
+        // also sets binding point `index` to the buffer, the pointer as the offset, and the
+        // *effective* stride, and points the attribute at that binding point with relative offset 0.
+        // The flat attribute view keeps the raw stride, because VERTEX_ATTRIB_ARRAY_STRIDE reports
+        // that argument verbatim, so the binding point is recorded alongside the resolved attribute
+        // rather than being resolved into it.
+        m_attributeBindingIndex[index] = index;
+        m_attributeRelativeOffset[index] = 0;
+
+        auto& binding = m_bindingPoints[index];
+        binding.Buffer = buffer;
+        binding.Offset = offset;
+        binding.Stride = effectiveStride;
+        binding.Divisor = m_attributes[index].Divisor;
+    }
+
     void VertexArrayObject::BindAttributeBuffer(Uint index, const SharedPtr<BufferObject>& buffer) {
         if (index >= MAX_VERTEX_ATTRIBS) return;
 
@@ -111,6 +131,11 @@ namespace MobileGL::MG_State::GLState {
 
     void VertexArrayObject::SetAttributeDivisor(Uint index, Uint divisor) {
         if (index >= MAX_VERTEX_ATTRIBS) return;
+        // glVertexAttribDivisor is VertexBindingDivisor on the attribute's own binding point
+        // (GL 4.6 core 10.3.2), so the binding-point view has to follow the resolved attribute.
+        if (index < MAX_VERTEX_ATTRIB_BINDINGS && m_attributeBindingIndex[index] == index) {
+            m_bindingPoints[index].Divisor = divisor;
+        }
         if (m_attributes[index].Divisor == divisor) return;
         m_attributes[index].Divisor = divisor;
         BumpAttributeFormatVersion(index);
@@ -187,18 +212,18 @@ namespace MobileGL::MG_State::GLState {
     }
 
     void VertexArrayObject::SetAttributeFormatSeparate(Uint attribIndex, int size, DataType type, Bool normalized,
-                                                       Bool isInteger, Uint relativeOffset) {
+                                                       Bool isInteger, Uint relativeOffset, Bool isBgra) {
         if (attribIndex >= MAX_VERTEX_ATTRIBS) return;
         if (size < 1 || size > 4) return;
 
         auto& attr = m_attributes[attribIndex];
         if (attr.Size != size || attr.Type != type || attr.Normalized != normalized || attr.IsInteger != isInteger ||
-            attr.IsBgra || m_attributeRelativeOffset[attribIndex] != relativeOffset) {
+            attr.IsBgra != isBgra || m_attributeRelativeOffset[attribIndex] != relativeOffset) {
             attr.Size = size;
             attr.Type = type;
             attr.Normalized = normalized;
             attr.IsInteger = isInteger;
-            attr.IsBgra = false; // the binding-format path (glVertexAttribFormat) does not carry BGRA
+            attr.IsBgra = isBgra;
             m_attributeRelativeOffset[attribIndex] = relativeOffset;
             BumpAttributeFormatVersion(attribIndex);
         }

MobileGL/MG_State/GLState/VertexArrayState/VertexArrayObject.h

@@ -64,6 +64,12 @@ namespace MobileGL {
 
                 void BindAttributeBuffer(Uint index, const SharedPtr<BufferObject>& buffer);
 
+                // Record what the pointer-style API implies for the binding-point view: attribute
+                // `index` bound to binding point `index` with relative offset 0, and that binding
+                // point carrying the buffer, the pointer offset and the effective stride.
+                void MirrorPointerIntoBinding(Uint index, const SharedPtr<BufferObject>& buffer, SizeT offset,
+                                              int effectiveStride);
+
                 BindingSlot<BufferObject>& GetIndexBufferBindingSlot();
                 const BindingSlot<BufferObject>& GetIndexBufferBindingSlot() const;
 
@@ -82,7 +88,7 @@ namespace MobileGL {
                 void SetBindingDivisor(Uint bindingIndex, Uint divisor);
                 void SetAttributeBinding(Uint attribIndex, Uint bindingIndex);
                 void SetAttributeFormatSeparate(Uint attribIndex, int size, DataType type, Bool normalized,
-                                                Bool isInteger, Uint relativeOffset);
+                                                Bool isInteger, Uint relativeOffset, Bool isBgra = false);
 
                 // The binding-point view the attributes were resolved from. Kept queryable
                 // because glGetVertexArrayIndexed[64]iv reports it verbatim, and the resolved

MobileGL/MG_Test/Query/QueryTest.cpp

@@ -144,10 +144,19 @@ TEST_F(QueryTest, GenQueriesReturnsDistinctNonzeroIdsAndTracksLiveness) {
     EXPECT_NE(ids[0], ids[2]);
     EXPECT_NE(ids[1], ids[2]);
 
+    // GenQueries only reserves names: "they acquire query state only when they are first used by
+    // calling BeginQuery" (GL 4.6 core 4.2.1), and IsQuery answers for objects, not for reserved
+    // names. So all three read as FALSE here even though the names are taken.
     EXPECT_EQ(MG_Impl::GLImpl::IsQuery(0), GL_FALSE);
+    EXPECT_EQ(MG_Impl::GLImpl::IsQuery(ids[0]), GL_FALSE);
+    EXPECT_EQ(MG_Impl::GLImpl::IsQuery(ids[1]), GL_FALSE);
+    EXPECT_EQ(MG_Impl::GLImpl::IsQuery(ids[2]), GL_FALSE);
+
+    // First use is what creates the object.
+    MG_Impl::GLImpl::BeginQuery(GL_TIME_ELAPSED, ids[0]);
+    MG_Impl::GLImpl::EndQuery(GL_TIME_ELAPSED);
     EXPECT_EQ(MG_Impl::GLImpl::IsQuery(ids[0]), GL_TRUE);
-    EXPECT_EQ(MG_Impl::GLImpl::IsQuery(ids[1]), GL_TRUE);
+    EXPECT_EQ(MG_Impl::GLImpl::IsQuery(ids[1]), GL_FALSE);
-    EXPECT_EQ(MG_Impl::GLImpl::IsQuery(ids[2]), GL_TRUE);
 
     MG_Impl::GLImpl::DeleteQueries(3, ids);
     EXPECT_EQ(MG_Impl::GLImpl::IsQuery(ids[0]), GL_FALSE);
@@ -157,6 +166,26 @@ TEST_F(QueryTest, GenQueriesReturnsDistinctNonzeroIdsAndTracksLiveness) {
     EXPECT_EQ(MG_Impl::GLImpl::GetError(), GL_NO_ERROR);
 }
 
+// The direct state access counterpart: glCreateQueries creates the object outright, so unlike a
+// glGenQueries name it is a query before anything is ever recorded into it.
+TEST_F(QueryTest, CreateQueriesYieldsQueryObjectsImmediately) {
+    GLuint ids[2] = {0, 0};
+    MG_Impl::GLImpl::CreateQueries(GL_TIME_ELAPSED, 2, ids);
+
+    EXPECT_NE(ids[0], 0u);
+    EXPECT_NE(ids[1], 0u);
+    EXPECT_NE(ids[0], ids[1]);
+
+    EXPECT_EQ(MG_Impl::GLImpl::IsQuery(ids[0]), GL_TRUE);
+    EXPECT_EQ(MG_Impl::GLImpl::IsQuery(ids[1]), GL_TRUE);
+    EXPECT_EQ(MG_Impl::GLImpl::GetError(), GL_NO_ERROR);
+
+    MG_Impl::GLImpl::DeleteQueries(2, ids);
+    EXPECT_EQ(MG_Impl::GLImpl::IsQuery(ids[0]), GL_FALSE);
+    EXPECT_EQ(MG_Impl::GLImpl::IsQuery(ids[1]), GL_FALSE);
+    EXPECT_EQ(MG_Impl::GLImpl::GetError(), GL_NO_ERROR);
+}
+
 TEST_F(QueryTest, TimeElapsedSpanFallsBackToImmediateZeroResult) {
     GLuint id = 0;
     MG_Impl::GLImpl::GenQueries(1, &id);

MobileGL/MG_Test/SanityTest.cpp

@@ -197,15 +197,15 @@ TEST(DirectGLESSanity, AdvertisesDepthTextureForGlmarkShadowScenes) {
     EXPECT_NE(std::find(extensions.begin(), extensions.end(), MobileGL::E_GL_ARB_depth_texture), extensions.end());
 }
 
-// The advertised target went back to 3.3 (see "restore target GL version to 3.3"); Voxy only ever
+// Voxy only ever needed the extensions, which stay advertised whatever the version is; the version
-// needed the extensions, which stay advertised, so assert what the backend really reports.
+// assertion just pins what the backend really reports, now that V_OpenGL40 is in the list.
 TEST(DirectGLESSanity, AdvertisesVoxyRequiredRenderingExtensions) {
     MobileGL::MG_Backend::DirectGLES::BackendObject_DirectGLES backend;
     const auto& rendererInfo = backend.GetRendererInfo().RendererGLInfo;
     const auto& extensions = rendererInfo.Extensions;
 
-    EXPECT_EQ(rendererInfo.TargetGLVersion.Major, 3);
+    EXPECT_EQ(rendererInfo.TargetGLVersion.Major, 4);
-    EXPECT_EQ(rendererInfo.TargetGLVersion.Minor, 3);
+    EXPECT_EQ(rendererInfo.TargetGLVersion.Minor, 0);
     EXPECT_EQ(rendererInfo.TargetGLVersion.Patch, 0);
 
     EXPECT_NE(std::find(extensions.begin(), extensions.end(), MobileGL::E_GL_ARB_compute_shader),
@@ -244,6 +244,19 @@ TEST(DirectGLESSanity, BindingZeroClearsPreviousNativeTextureBinding) {
     ASSERT_NE(frontendTextureObject, nullptr);
     ASSERT_EQ(frontendTextureObject->GetExternalIndex(), frontendTexture);
 
+    // A texture with no image is incomplete, and an incomplete texture samples as (0, 0, 0, 1) -
+    // which DirectGLES expresses by leaving the native target unbound (see "sample a
+    // mipmap-incomplete texture as black"). This test is about the bind-0 clear, so the texture
+    // has to be complete enough to get bound in the first place: a format plus a level 0. At 1x1
+    // that single level is the whole mip chain, so it stays complete under any filter. The state
+    // is set directly rather than through glTexImage2D because the mock GLES table below wires
+    // only the binding entry points, not the upload path.
+    frontendTextureObject->SetInternalFormat(TextureInternalFormat::RGBA8);
+    MG_State::GLState::AsMipmapTexture(frontendTextureObject.get())
+        ->AllocateStorage(TextureUploadTarget::Texture2D, 0, {{1, 1, 1}, 4});
+    ASSERT_FALSE(MG_State::GLState::SamplesAsIncompleteTexture(
+        frontendTextureObject.get(), frontendTextureObject->GetSamplerObject().get()));
+
     auto& backendTexture = DirectGLES::TextureImpl::g_backendTextureObjects.GetOrCreate(frontendTextureObject);
     backendTexture = MakeShared<DirectGLES::TextureImpl::BackendTextureObject>();
     const GLuint backendTextureId = backendTexture->GetBackendTextureId();
@@ -399,14 +412,15 @@ TEST(DirectVulkanSanity, RenderPassExtentUsesSwapchainSizeOnlyForDefaultFramebuf
               MobileGL::IntVec2(512, 512));
 }
 
-// See the DirectGLES twin above: the target version is 3.3 again, the extensions are what matter.
+// See the DirectGLES twin above: the target version follows the advertised list, the extensions are
+// what matter.
 TEST(DirectVulkanSanity, AdvertisesVoxyRequiredRenderingExtensions) {
     MobileGL::MG_Backend::DirectVulkan::BackendObject_DirectVulkan backend;
     const auto& rendererInfo = backend.GetRendererInfo().RendererGLInfo;
     const auto& extensions = rendererInfo.Extensions;
 
-    EXPECT_EQ(rendererInfo.TargetGLVersion.Major, 3);
+    EXPECT_EQ(rendererInfo.TargetGLVersion.Major, 4);
-    EXPECT_EQ(rendererInfo.TargetGLVersion.Minor, 3);
+    EXPECT_EQ(rendererInfo.TargetGLVersion.Minor, 0);
     EXPECT_EQ(rendererInfo.TargetGLVersion.Patch, 0);
 
     EXPECT_NE(std::find(extensions.begin(), extensions.end(), MobileGL::E_GL_ARB_compute_shader),

MobileGL/MG_Test/Texture/TextureTest.cpp

@@ -15,6 +15,7 @@
 #include <Config.h>
 #include <MG_Backend/BackendObjects.h>
 #include <MG_Backend/DirectGLES/Managers.h>
+#include <MG_Impl/GLImpl/Framebuffer/GL_Framebuffer.h>
 #include <MG_Impl/GLImpl/Getter/GL_Getter.h>
 #include <MG_Impl/GLImpl/RenderState/GL_RenderState.h>
 #include <MG_Impl/GLImpl/Sampler/GL_Sampler.h>
@@ -296,6 +297,18 @@ TEST_F(TextureTest, CopyTextureSubImage2DUsesNamedObjectAndRestoresBinding) {
     MG_Impl::GLImpl::CreateTextures(GL_TEXTURE_2D, 1, &boundTexture);
     MG_Impl::GLImpl::BindTextureUnit(0, boundTexture);
 
+    // The copy is only allowed to reach the backend when the destination region fits the level and
+    // the read framebuffer can supply pixels, so the call has to be set up as a legal one.
+    MG_Impl::GLImpl::TextureStorage2D(namedTexture, 3, GL_RGBA8, 16, 16);
+
+    GLuint readFramebuffer = 0;
+    GLuint readTexture = 0;
+    MG_Impl::GLImpl::CreateFramebuffers(1, &readFramebuffer);
+    MG_Impl::GLImpl::CreateTextures(GL_TEXTURE_2D, 1, &readTexture);
+    MG_Impl::GLImpl::TextureStorage2D(readTexture, 1, GL_RGBA8, 16, 16);
+    MG_Impl::GLImpl::NamedFramebufferTexture(readFramebuffer, GL_COLOR_ATTACHMENT0, readTexture, 0);
+    MG_Impl::GLImpl::BindFramebuffer(GL_READ_FRAMEBUFFER, readFramebuffer);
+
     const auto boundBefore = MG_State::pGLContext->GetTextureUnitObject(0)
                                  .GetBindingSlot(TextureTarget::Texture2D)
                                  .GetBoundObject();

MobileGL/MG_Util/BackendLoaders/OpenGL/Loader.cpp

@@ -833,6 +833,10 @@ namespace MobileGL::MG_Util::BackendLoader {
                 if (std::strcmp(extension, "GL_EXT_texture_filter_anisotropic") == 0) {
                     caps.SupportsTextureFilterAnisotropy = true;
                 }
+                if (std::strcmp(extension, "GL_EXT_texture_border_clamp") == 0 ||
+                    std::strcmp(extension, "GL_OES_texture_border_clamp") == 0) {
+                    caps.SupportsTextureBorderClamp = true;
+                }
                 if (std::strcmp(extension, "GL_EXT_base_instance") == 0) {
                     caps.SupportsBaseInstance = true;
                 }
@@ -1014,6 +1018,10 @@ namespace MobileGL::MG_Util::BackendLoader {
         }
         // Only legal to query once the extension has been seen in the loop above, hence not batched
         // with the unconditional probes: on a driver without it this raises GL_INVALID_ENUM.
+        // Core from ES 3.2 on, whatever the extension string says.
+        if (caps.GLESVersion.Major > 3 || (caps.GLESVersion.Major == 3 && caps.GLESVersion.Minor >= 2)) {
+            caps.SupportsTextureBorderClamp = true;
+        }
         if (caps.SupportsTextureFilterAnisotropy) {
             GLfloat maxTextureMaxAnisotropy = 1.0f;
             glesFuncs.glGetFloatv(GL_MAX_TEXTURE_MAX_ANISOTROPY_EXT, &maxTextureMaxAnisotropy);
@@ -1056,9 +1064,17 @@ namespace MobileGL::MG_Util::BackendLoader {
         caps.MaxComputeWorkGroupInvocations = maxComputeWorkGroupInvocations;
         caps.MaxShaderStorageBufferBindings = maxShaderStorageBufferBindings;
         caps.MaxTextureBufferSize = maxTextureBufferSize;
+        // Through glesFuncs, like every other capability query here: a bare glGetIntegerv resolves
+        // to MobileGL's own exported entry point, which answers this pname from the very
+        // capability table being filled in - so the driver's real alignment never arrived and the
+        // backend reported an unconstrained offset it cannot honour.
         GLint textureBufferOffsetAlignment = 1;
-        glGetIntegerv(GL_TEXTURE_BUFFER_OFFSET_ALIGNMENT, &textureBufferOffsetAlignment);
+        glesFuncs.glGetIntegerv(GL_TEXTURE_BUFFER_OFFSET_ALIGNMENT, &textureBufferOffsetAlignment);
-        while (glGetError() != GL_NO_ERROR) {}
+        // Core in ES 3.2 and in EXT_texture_buffer; an older context rejects the pname and leaves
+        // the default in place.
+        if (glesFuncs.glGetError) {
+            while (glesFuncs.glGetError() != GL_NO_ERROR) {}
+        }
         caps.TextureBufferOffsetAlignment = std::max(1, textureBufferOffsetAlignment);
         caps.MaxUniformBufferBindings = maxUniformBufferBindings;
         caps.MaxUniformBlockSize = maxUniformBlockSize;

MobileGL/MG_Util/BackendLoaders/OpenGL/Loader.h

@@ -1041,6 +1041,10 @@ namespace MobileGL {
             // GL_EXT_texture_filter_anisotropic is present, so sampler/texture
             // anisotropy may be forwarded without raising GL_INVALID_ENUM in GLES.
             Bool SupportsTextureFilterAnisotropy = false;
+            // GL_TEXTURE_BORDER_COLOR and GL_CLAMP_TO_BORDER: ES 3.2 core, or
+            // EXT/OES_texture_border_clamp before that. Without it every border-colour parameter
+            // raises INVALID_ENUM on the driver, so the syncs have to be gated on it.
+            Bool SupportsTextureBorderClamp = false;
             // GL_MAX_TEXTURE_MAX_ANISOTROPY_EXT of the host driver; only queried when the
             // extension above is present, and left at 1.0 (no anisotropy) otherwise.
             Float MaxTextureMaxAnisotropy = 1.0f;

MobileGL/MG_Util/Converters/GLToMG/TextureEnumConverter.cpp

@@ -288,6 +288,35 @@ namespace MobileGL {
                 return TextureInternalFormat::SRGB8;
             case GL_COMPRESSED_SRGB_ALPHA:
                 return TextureInternalFormat::SRGB8Alpha8;
+            // BPTC and ETC2/EAC follow the same deviation as RGTC above, for the same reason: they
+            // are specific formats core GL requires (BPTC from 4.2, ETC2/EAC from 4.3) that this
+            // stack has no compressor for. INVALID_ENUM was never a legal answer for them, and
+            // uncompressed storage is the same trade the RGTC formats already take.
+            case GL_COMPRESSED_RGBA_BPTC_UNORM:
+                return TextureInternalFormat::RGBA8;
+            case GL_COMPRESSED_SRGB_ALPHA_BPTC_UNORM:
+                return TextureInternalFormat::SRGB8Alpha8;
+            case GL_COMPRESSED_RGB_BPTC_SIGNED_FLOAT:
+            case GL_COMPRESSED_RGB_BPTC_UNSIGNED_FLOAT:
+                return TextureInternalFormat::RGB16F;
+            case GL_COMPRESSED_RGB8_ETC2:
+                return TextureInternalFormat::RGB8;
+            case GL_COMPRESSED_SRGB8_ETC2:
+                return TextureInternalFormat::SRGB8;
+            case GL_COMPRESSED_RGB8_PUNCHTHROUGH_ALPHA1_ETC2:
+            case GL_COMPRESSED_RGBA8_ETC2_EAC:
+                return TextureInternalFormat::RGBA8;
+            case GL_COMPRESSED_SRGB8_PUNCHTHROUGH_ALPHA1_ETC2:
+            case GL_COMPRESSED_SRGB8_ALPHA8_ETC2_EAC:
+                return TextureInternalFormat::SRGB8Alpha8;
+            case GL_COMPRESSED_R11_EAC:
+                return TextureInternalFormat::R8;
+            case GL_COMPRESSED_SIGNED_R11_EAC:
+                return TextureInternalFormat::R8Snorm;
+            case GL_COMPRESSED_RG11_EAC:
+                return TextureInternalFormat::RG8;
+            case GL_COMPRESSED_SIGNED_RG11_EAC:
+                return TextureInternalFormat::RG8Snorm;
             case GL_ALPHA:
             case GL_RED:
                 return TextureInternalFormat::Red;

MobileGL/MG_Util/SelfTest/DriverPost.cpp

@@ -298,6 +298,17 @@ namespace MobileGL::MG_Util::SelfTest {
                              "not supported; no impact: the native indirect path deliberately does not "
                              "rely on it (shader-side emulation handles baseInstance semantics)");
             }
+            if (caps.SupportsTextureBorderClamp) {
+                builder.Pass("Texture border clamp",
+                             "supported (GL_TEXTURE_BORDER_COLOR reaches the driver, so "
+                             "GL_CLAMP_TO_BORDER samples the colour the application set)");
+            } else {
+                builder.Warn("Texture border clamp",
+                             "not supported (pre-ES 3.2 without GL_EXT/OES_texture_border_clamp); "
+                             "GL_TEXTURE_BORDER_COLOR is not synced to the driver at all, so anything "
+                             "sampling outside a GL_CLAMP_TO_BORDER texture reads the driver's default "
+                             "border instead of the requested colour");
+            }
             if (glesFuncs.glPatchParameteri != nullptr) {
                 builder.Pass("Tessellation patch parameters",
                              "glPatchParameteri present (GL_PATCH_VERTICES reaches the driver)");

README.md

@@ -14,7 +14,7 @@ MobileGL is a *free* and *open-source* project that implements a desktop **OpenG
 
 > [!NOTE]
 >
-> **Status:** In development. Parts of the codebase are incomplete. Current short-term target: **OpenGL 3.3 (Core Profile)**.
+> **Status:** In development. Parts of the codebase are incomplete. Current short-term target: **OpenGL 4.2 (Core Profile)**.
 
 ## Project positioning
 
@@ -47,9 +47,10 @@ Refer to each component's repository for exact license texts. Any bundled third-
 
 ## Compatibility & target
 
-* **Short-term target:** `OpenGL 3.3 (Core Profile)`.
+* **Short-term target:** `OpenGL 4.2 (Core Profile)`.
 * **Current development focus:**
-  * `MG_State` and `MG_Impl` for `OpenGL 3.3 (Core Profile)`
+  * Performance improvement
+  * `MG_State` and `MG_Impl` for `OpenGL 4.2 (Core Profile)`
   * `Direct (Vulkan)` backend
   * `Direct (OpenGL ES)` backend
 
@@ -163,11 +164,6 @@ MobileGL supports runtime configuration via environment variables.
 | `MOBILEGL_COHERENT_AS_FLUSH` | Treat persistent `GL_MAP_FLUSH_EXPLICIT_BIT` maps as coherent (app-compat for engines like Flywheel that never flush them). | `0`, `1` | `0` |
 | `VK_ICD_FILENAMES`      | Select the Vulkan ICD used by the Vulkan loader. | Path to an ICD JSON file             | Loader default |
 
-## Notice
-
-* MobileGL is **not** production-ready currently.
-* Some `OpenGL 3.3 (Core Profile)` features are still missing or under development.
-
 ## License
 
 This project is distributed under **GNU LGPL v3.0**. See the `LICENSE` file in the repository for detailed information.

tools/cts/README.md

@@ -1,12 +1,16 @@
 # Running the OpenGL CTS against MobileGL
 
-This directory contains two supported paths:
+This directory contains three supported paths:
 
 - Android arm64 / MobileGL EGL: the KHR-GL33 workflow documented below and in
   `skills/gl-cts-on-mobilegl/SKILL.md`.
 - Windows x64 / MobileGL WGL: the GL30-GL46 pipeline in
   `scripts/wgl_glcts_pipeline.py`, documented by
   `skills/wgl-gl-cts-on-mobilegl/SKILL.md`.
+- Desktop Linux x64 / MobileGL EGL: `scripts/run_cts_local.py` against the
+  `mobilegl-desktop` VK-GL-CTS target, documented in "Desktop Linux workflow"
+  below and in `skills/linux-gl-cts-on-mobilegl/SKILL.md`. This is the path that
+  needs no device and no GPU.
 
 Windows prerequisites are Git, Python 3.9+, CMake, Visual Studio 2022's Desktop
 C++ workload, and a Vulkan SDK visible to CMake. DirectVulkan also needs a
@@ -25,6 +29,66 @@ resumes individual suites after crashes/timeouts, and writes Markdown plus JSON
 reports below the printed `runs/<first-16-of-run-fingerprint>` directory. Its
 manifest records provenance and the runner settings used to validate a resume.
 
+## Desktop Linux workflow
+
+The `mobilegl-desktop` target builds `glcts` as an ordinary host executable that
+reaches OpenGL only through `libMobileGL.so`. Both backends run headless with no
+GPU at all, which makes this the cheapest way to measure a single test group
+while working on it.
+
+Apt packages: `mesa-vulkan-drivers` (lavapipe, for DirectVulkan),
+`libegl1-mesa-dev` and `libgles2-mesa-dev` (the system EGL/ES that DirectGLES
+drives), `libvulkan-dev`, `ninja-build`.
+
+```sh
+cmake -S . -B build-linux -G Ninja -DCMAKE_BUILD_TYPE=Release \
+      -DMOBILEGL_BUILD_TEST=OFF -DMOBILEGL_BUILD_BENCHMARK=OFF
+cmake --build build-linux --parallel "$(nproc)"
+
+python tools/cts/scripts/sync_to_cts.py "$CTS"
+git -C "$CTS" apply <path-to>/tools/cts/patches/0001-fbo-color-texture-attachment.patch
+cmake -S "$CTS" -B "$CTS/build-cts" -G Ninja -DDEQP_TARGET=mobilegl-desktop \
+      -DCMAKE_BUILD_TYPE=Release
+ninja -C "$CTS/build-cts" glcts
+
+python tools/cts/scripts/run_cts_local.py --backend DirectGLES \
+    --glcts "$CTS/build-cts/external/openglcts/modules/glcts" \
+    --lib build-linux/libMobileGL.so \
+    --caselist cases.txt --outdir runs/gles --env EGL_PLATFORM=surfaceless
+python tools/cts/scripts/qpa_report.py runs/gles --label DirectGLES
+```
+
+`EGL_PLATFORM=surfaceless` is not optional for DirectGLES: without a `/dev/dri`
+node Mesa's EGL fails `eglInitialize` on the default display, and MobileGL
+reports that as `EGL_BAD_ALLOC` out of `eglCreatePbufferSurface`. DirectVulkan
+needs nothing extra - lavapipe exposes `VK_EXT_headless_surface`, which is what
+the desktop platform port's pbuffer path requires.
+
+Two known differences from a real GPU, both MobileGL's rather than the harness's:
+DirectVulkan reads back zeros from the **default** framebuffer (a user FBO,
+renderbuffer- or texture-attached, is correct on both backends), and lavapipe
+supports renderbuffer formats that Adreno reports as unsupported, so the Android
+runs see `NotSupported` where these do not.
+
+### Reference results: KHR-GL45.direct_state_access
+
+`opengl-cts-4.6.8.1`, the 371 `direct_state_access` cases of the `gl45-main`
+mustpass list, lavapipe / Mesa 25.2.8, `--deqp-surface-type=fbo`.
+
+| backend | conformance | strict Pass | Fail | InternalError | Crash |
+| --- | ---: | ---: | ---: | ---: | ---: |
+| DirectGLES | **82.48%** | 74.93% | 57 | 8 | 0 |
+| DirectVulkan | **73.58%** | 73.32% | 87 | 8 | 3 |
+
+`textures_storage_multisample_*` is most of what remains, and the two backends
+are there for different reasons. Espryt passes the whole `2d` half and fails only
+the `3d` one, which attaches a multisample **array** texture with
+`glFramebufferTextureLayer` - layered attachments are not something the
+framebuffer attachment model represents yet. Magma fails both halves: it cannot
+sample a multisample texture at all, so the second pass reads back nothing.
+`program_pipelines_*` needs ARB_separate_shader_objects, which is stubbed
+throughout.
+
 ## Android KHR-GL33 workflow
 
 Goal: measure how much of the OpenGL 3.3 core-profile conformance suite MobileGL

tools/cts/skills/README.md

@@ -17,3 +17,4 @@ Each skill is a self-contained package, matching the layout used by
 | --- | --- |
 | [gl-cts-on-mobilegl](gl-cts-on-mobilegl/SKILL.md) | Build VK-GL-CTS `glcts` as a standalone Android arm64 binary against MobileGL's own EGL, run KHR-GL33, and report a per-backend OpenGL 3.3 core conformance rate. |
 | [wgl-gl-cts-on-mobilegl](wgl-gl-cts-on-mobilegl/SKILL.md) | Build MobileGL's Windows x64 WGL drop-in, run GL30-GL46 core CTS against DirectGLES and DirectVulkan, resume safely, and emit validated reports. |
+| [linux-gl-cts-on-mobilegl](linux-gl-cts-on-mobilegl/SKILL.md) | Build `glcts` as a desktop Linux host binary against MobileGL, run any CTS group headlessly with no GPU, and report Espryt and Magma separately. The path to reach for while iterating on a fix. |

tools/cts/skills/linux-gl-cts-on-mobilegl/SKILL.md

@@ -0,0 +1,159 @@
+---
+name: linux-gl-cts-on-mobilegl
+description: Run the Khronos OpenGL CTS (VK-GL-CTS glcts) against MobileGL on desktop Linux with no GPU and no device, and report a per-backend conformance rate for Espryt (DirectGLES) and Magma (DirectVulkan). Use when measuring or iterating on the conformance of one test group, when a fix needs a before/after number, or when neither an Android device nor a Windows GPU box is available.
+---
+
+# OpenGL CTS on MobileGL (desktop Linux)
+
+## Overview
+
+`glcts` is built as an ordinary x86-64 host executable that reaches OpenGL only
+through `libMobileGL.so`, using the `mobilegl-desktop` VK-GL-CTS target and the
+`tcu::Platform` port in `tools/cts/platform/`. Nothing links `libGL` or `libEGL`,
+so a result is unambiguously MobileGL's.
+
+Both backends run headless on software rendering, so this needs no GPU at all:
+
+- **Espryt** (`DirectGLES`) drives Mesa's OpenGL ES through the system EGL.
+- **Magma** (`DirectVulkan`) runs on lavapipe, whose `VK_EXT_headless_surface`
+  is what the desktop platform port's pbuffer path requires.
+
+Always report the two backends **separately**. They are different
+implementations of the same front end, they fail different cases, and a single
+combined number hides which one a change moved.
+
+## Prerequisites
+
+```sh
+sudo apt-get install -y ninja-build cmake libvulkan-dev \
+    libegl1-mesa-dev libgles2-mesa-dev mesa-vulkan-drivers
+```
+
+`mesa-vulkan-drivers` is what installs lavapipe; without it DirectVulkan has no
+ICD and `eglInitialize` fails inside the backend. A C++23 toolchain is required
+(GCC 13+, or Clang 20+ — Clang 18 defines `__cpp_concepts` as 201907L, which
+switches libstdc++'s `<expected>` off and the build fails in
+`MG_Util/ShaderTranspiler/Types.h`).
+
+```sh
+export MG=<path-to-MobileGL-worktree>
+export CTS=<path-to-VK-GL-CTS-checkout>
+```
+
+## Step 1 — build libMobileGL.so
+
+```sh
+git -C "$MG" submodule update --init --recursive
+python3 "$MG/3rdparty/glslang/update_glslang_sources.py"   # SPIRV-Tools; ENABLE_OPT is forced on
+
+cmake -S "$MG" -B "$MG/build-linux" -G Ninja -DCMAKE_BUILD_TYPE=Release \
+      -DMOBILEGL_BUILD_TEST=OFF -DMOBILEGL_BUILD_BENCHMARK=OFF \
+      -DCMAKE_C_COMPILER_LAUNCHER=ccache -DCMAKE_CXX_COMPILER_LAUNCHER=ccache
+cmake --build "$MG/build-linux" --parallel "$(nproc)"
+```
+
+Add `-DCMAKE_INTERPROCEDURAL_OPTIMIZATION_RELEASE=FALSE` when iterating. The
+Release configuration turns LTO on, which makes every relink cost minutes for no
+behavioural difference; a conformance number is identical either way.
+
+## Step 2 — get VK-GL-CTS and build glcts
+
+Use a release tag so the mustpass list, and therefore the reported rate, is
+citable.
+
+```sh
+git -C "$CTS" checkout opengl-cts-4.6.8.1
+python3 "$CTS/external/fetch_sources.py"
+
+python3 "$MG/tools/cts/scripts/sync_to_cts.py" "$CTS"
+git -C "$CTS" apply "$MG/tools/cts/patches/0001-fbo-color-texture-attachment.patch"
+
+cmake -S "$CTS" -B "$CTS/build-cts" -G Ninja -DDEQP_TARGET=mobilegl-desktop \
+      -DCMAKE_BUILD_TYPE=Release \
+      -DCMAKE_C_COMPILER_LAUNCHER=ccache -DCMAKE_CXX_COMPILER_LAUNCHER=ccache
+ninja -C "$CTS/build-cts" glcts
+```
+
+Confirm the configure output says `*** Using MobileGL desktop target`. Budget a
+couple of hours for the `glcts` link on a small machine; it is a one-time cost
+that ccache makes cheap afterwards.
+
+If `fetch_sources.py` dies with `HTTP Error 403` it is an egress policy blocking
+the GitHub archive downloads (zlib, libpng), not a broken checkout: `git clone`
+still works, so clone the package at the tag the script pins into
+`external/<pkg>/src` by hand — for libpng also copy
+`scripts/pnglibconf.h.prebuilt` to `src/pnglibconf.h`, which is what the
+script's post-extract step does.
+
+## Step 3 — run, once per backend
+
+```sh
+cd "$MG"
+for BACKEND in DirectGLES DirectVulkan; do
+  python3 tools/cts/scripts/run_cts_local.py --backend "$BACKEND" \
+      --glcts "$CTS/build-cts/external/openglcts/modules/glcts" \
+      --lib   build-linux/libMobileGL.so \
+      --caselist cases.txt --outdir "runs/${BACKEND}" \
+      --env EGL_PLATFORM=surfaceless
+  python3 tools/cts/scripts/qpa_report.py "runs/${BACKEND}" --label "$BACKEND"
+done
+```
+
+`cases.txt` is any subset of a mustpass list. For one group, filter the list
+rather than running the whole suite:
+
+```sh
+grep direct_state_access \
+  "$CTS"/external/openglcts/data/gl_cts/data/mustpass/gl/khronos_mustpass/main/gl45-main.txt \
+  > cases.txt
+```
+
+`run_cts_local.py` re-invokes `glcts` with only the cases that have no result
+yet, so a crash costs one case rather than the run, and it records the case that
+was open when the process died as `Crash`. `qpa_report.py` scores `Pass`,
+`NotSupported` and the warning statuses as non-failures, the way Khronos scores
+a submission, and reports the strict `Pass`-only rate alongside.
+
+## Required flags, and why
+
+| Flag | Why it is not optional |
+| --- | --- |
+| `--env EGL_PLATFORM=surfaceless` | For DirectGLES. With no `/dev/dri` node Mesa's EGL fails `eglInitialize` on the default display, and MobileGL surfaces that as `EGL_BAD_ALLOC` out of `eglCreatePbufferSurface` — which points at the wrong call entirely. Harmless for DirectVulkan, so pass it to both. |
+| `--deqp-surface-type=fbo` (the runner's default) | DirectVulkan reads back zeros from the **default** framebuffer. dEQP verifies nearly everything through `glReadPixels`, so rendering to the surface scores Magma near zero for a reason unrelated to conformance. Use it for both backends so the two numbers stay comparable. |
+| `--deqp-terminate-on-device-lost=disable` (supplied by the runner) | Its default calls `glGetGraphicsResetStatus()` after every case. That is GL 4.5 / `KHR_robustness`, absent from what MobileGL exports, so the pointer is null and the process segfaults on the first case. |
+
+## What this environment does and does not tell you
+
+Reproducible here, and MobileGL's own rather than a driver quirk:
+
+- DirectVulkan's default-framebuffer readback returns zeros; a user FBO,
+  renderbuffer- or texture-attached, is correct on both backends. This is the
+  same defect the Android runs work around, so it can be debugged without a
+  phone.
+
+Different from a real GPU, so do not read conformance into it:
+
+- lavapipe supports renderbuffer formats Adreno reports as unsupported, so the
+  Android runs see `NotSupported` where these do not, and vice versa.
+- Backend limits differ. `GL_TEXTURE_BUFFER_OFFSET_ALIGNMENT` is 16 on llvmpipe
+  and on lavapipe; a device that reports 1 will not exercise the same paths.
+- Everything is a software rasterizer, so a case that fails only under real
+  timing or real tiling will not fail here.
+
+## Reference results: KHR-GL45.direct_state_access
+
+`opengl-cts-4.6.8.1`, the 371 `direct_state_access` cases of the `gl45-main`
+mustpass list, Mesa 25.2.8, `--deqp-surface-type=fbo`.
+
+| backend | renderer | conformance | strict Pass | Fail | InternalError | Crash |
+| --- | --- | ---: | ---: | ---: | ---: | ---: |
+| DirectGLES | Espryt | **82.48%** | 74.93% | 57 | 8 | 0 |
+| DirectVulkan | Magma | **73.58%** | 73.32% | 87 | 8 | 3 |
+
+## Contents
+
+    platform/tcuMobileGLPlatform.{cpp,hpp}   dEQP tcu::Platform for MobileGL
+    targets/mobilegl-desktop.cmake           VK-GL-CTS target (-DDEQP_TARGET=mobilegl-desktop)
+    scripts/sync_to_cts.py                   inject the port into a CTS checkout
+    scripts/run_cts_local.py                 crash-resuming local-host runner
+    scripts/qpa_report.py                    .qpa -> conformance rate

tools/cts/skills/linux-gl-cts-on-mobilegl/agents/openai.yaml

@@ -0,0 +1,4 @@
+interface:
+  display_name: "OpenGL CTS on MobileGL (desktop Linux)"
+  short_description: "Run VK-GL-CTS glcts against MobileGL headlessly and report Espryt and Magma separately"
+  default_prompt: "Use $linux-gl-cts-on-mobilegl to run the selected OpenGL CTS group against MobileGL on this Linux host and report the conformance rate for DirectGLES (Espryt) and DirectVulkan (Magma) separately."