Files
MobileGL/MobileGL/MG_Impl/GLImpl/Sync/GL_Sync.cpp
T
swung0x48andClaude Fable 5 139de76347 [Fix] (MG_State/MG_Impl/MG_Backend): render Flywheel instanced+indirect on both backends
Create 6 / Flywheel 1.0.6 now renders correctly with both flywheel:instancing
and flywheel:indirect on DirectGLES and DirectVulkan (verified in-game on
Adreno 830: waterwheels and cogwheels solid, animated, correct pairing, no
crashes across all four combinations).

- MG_State/MG_Impl: sync explicitly-ranged SSBO bindings of FLUSH_EXPLICIT
  persistent maps to the backend before compute dispatches. Flywheel writes
  its scatter-copy descriptors into the staging ring's persistent map and
  never flushes that span (UB per spec, works on drivers whose maps alias
  GPU-visible memory); our maps alias the CPU shadow, so the descriptors
  never reached the GPU: the scatter compute copied nothing (GLES: empty
  draw commands) or stale garbage (Vulkan: wild indirect commands ending in
  VK_ERROR_DEVICE_LOST).
- MG_Impl/MG_Backend: real glFenceSync objects backed by backend fences
  (GLES: native ES syncs guarded by context generation and owner thread;
  Vulkan: buffer-manager frame serials), replacing always-signaled stubs
  that let Flywheel reclaim staging memory the GPU still reads.
- MG_Backend/DirectGLES: compute dispatches now run the same per-program
  resource sync as draws (uniform-block bindings and sampler units must be
  re-established through the API because layout(binding) is stripped from
  transpiled ESSL) and rebind texture units afterwards; the cull shader
  used to read a stale _FlwFrameUniforms binding and the depth-pyramid
  downsample sampled a stale unit-0 texture, zeroing the Hi-Z pyramid and
  occlusion-culling all Flywheel geometry. Image uniforms are excluded from
  glUniform1i (ES bakes their unit via layout(binding)); image-unit sync is
  clamped to the device limit; eliminated/SSBO-classified uniform blocks
  are skipped.
- MG_Backend/DirectGLES: gl_BaseInstance in native indirect draws reads the
  GPU-written command buffer through an injected mg_IndirectParams SSBO
  view addressed per draw instead of the zero CPU shadow; layout(binding)
  is preserved for SSBO/image declarations (ES has no API rebinding for
  them); the ES context ownership claim moved to a global atomic owner
  thread with an EGL ground-truth check, and deferred buffer op state is
  mutex-guarded, so ops cannot silently no-op after context migration.
- MG_Backend/DirectVulkan: new RebaseInstanceIndexPass rewrites vertex
  InstanceIndex loads to (InstanceIndex - BaseInstance). glslang's relaxed
  Vulkan mode aliases gl_InstanceID to InstanceIndex, which includes
  firstInstance, but GL's gl_InstanceID is zero-based - draws with nonzero
  baseInstance paired meshes with wrong instance data (cogwheel drawn as a
  waterwheel, another wheel collapsed invisible). Gated on the
  shaderDrawParameters device feature. Sampled-read barriers additionally
  cover the compute stage (the Hi-Z downsample samples the depth
  attachment from compute), and short uniform-buffer ranges keep the
  existing zero-padding.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-09 06:30:10 +00:00

137 lines
5.1 KiB
C++

// MobileGL - MobileGL/MG_Impl/GLImpl/Sync/GL_Sync.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_Sync.h"
#include <MG_Backend/BackendObjects.h>
namespace MobileGL::MG_Impl::GLImpl {
namespace {
// Frontend sync object: wraps an optional backend fence handle. A null
// backend handle (backend has no fence support, or could not create a
// fence at call time) keeps the legacy always-signaled behavior.
struct SyncObject {
MG_Backend::BackendSyncHandle backendHandle = nullptr;
GLenum condition = GL_SYNC_GPU_COMMANDS_COMPLETE;
GLbitfield flags = 0;
};
// Sync calls may arrive from any thread (launchers migrate the context
// across JVM threads), so the live-object registry is mutex-guarded.
// Entries left at process shutdown are simply dropped; their backend
// handles die with the backend.
std::mutex g_syncObjectsMutex;
UnorderedMap<GLsync, SyncObject*> g_liveSyncObjects;
SyncObject* FindSyncObject(GLsync sync) {
const std::lock_guard<std::mutex> lock(g_syncObjectsMutex);
const auto it = g_liveSyncObjects.find(sync);
return it != g_liveSyncObjects.end() ? it->second : nullptr;
}
} // namespace
GLsync FenceSync(GLenum condition, GLbitfield flags) {
auto* syncObject = new SyncObject;
syncObject->condition = condition;
syncObject->flags = flags;
if (const auto backendFenceSync = MG_Backend::gBackendFunctionsTable.GL.FenceSync) {
syncObject->backendHandle = backendFenceSync();
}
const GLsync handle = reinterpret_cast<GLsync>(syncObject);
const std::lock_guard<std::mutex> lock(g_syncObjectsMutex);
g_liveSyncObjects[handle] = syncObject;
return handle;
}
GLboolean IsSync(GLsync sync) {
return FindSyncObject(sync) != nullptr ? GL_TRUE : GL_FALSE;
}
GLenum ClientWaitSync(GLsync sync, GLbitfield flags, GLuint64 timeout) {
const auto* syncObject = FindSyncObject(sync);
if (!syncObject) {
return GL_WAIT_FAILED;
}
const auto backendClientWaitSync = MG_Backend::gBackendFunctionsTable.GL.ClientWaitSync;
if (!backendClientWaitSync || !syncObject->backendHandle) {
return GL_ALREADY_SIGNALED; // legacy always-signaled fallback
}
return backendClientWaitSync(syncObject->backendHandle, flags, timeout);
}
void WaitSync(GLsync sync, GLbitfield flags, GLuint64 timeout) {
const auto* syncObject = FindSyncObject(sync);
if (!syncObject) {
return;
}
const auto backendWaitSync = MG_Backend::gBackendFunctionsTable.GL.WaitSync;
if (backendWaitSync && syncObject->backendHandle) {
backendWaitSync(syncObject->backendHandle, flags, timeout);
}
}
void DeleteSync(GLsync sync) {
if (sync == nullptr) {
return; // glDeleteSync(0) is silently ignored
}
SyncObject* syncObject = nullptr;
{
const std::lock_guard<std::mutex> lock(g_syncObjectsMutex);
const auto it = g_liveSyncObjects.find(sync);
if (it == g_liveSyncObjects.end()) {
return;
}
syncObject = it->second;
g_liveSyncObjects.erase(it);
}
const auto backendDeleteSync = MG_Backend::gBackendFunctionsTable.GL.DeleteSync;
if (backendDeleteSync && syncObject->backendHandle) {
backendDeleteSync(syncObject->backendHandle);
}
delete syncObject;
}
void GetSynciv(GLsync sync, GLenum pname, GLsizei bufSize, GLsizei* length, GLint* values) {
const auto* syncObject = FindSyncObject(sync);
if (!syncObject) {
if (length) {
*length = 0;
}
return;
}
GLint value = 0;
switch (pname) {
case GL_OBJECT_TYPE:
value = GL_SYNC_FENCE;
break;
case GL_SYNC_STATUS: {
const auto backendGetSyncStatus = MG_Backend::gBackendFunctionsTable.GL.GetSyncStatus;
const Bool signaled = !backendGetSyncStatus || !syncObject->backendHandle ||
backendGetSyncStatus(syncObject->backendHandle);
value = signaled ? GL_SIGNALED : GL_UNSIGNALED;
break;
}
case GL_SYNC_CONDITION:
value = static_cast<GLint>(syncObject->condition);
break;
case GL_SYNC_FLAGS:
value = static_cast<GLint>(syncObject->flags);
break;
default:
break;
}
if (length) {
*length = bufSize > 0 && values ? 1 : 0;
}
if (bufSize > 0 && values) {
values[0] = value;
}
}
} // namespace MobileGL::MG_Impl::GLImpl