llvertexbuffer.cpp

/**
 * @file llvertexbuffer.cpp
 * @brief LLVertexBuffer implementation
 *
 * $LicenseInfo:firstyear=2003&license=viewerlgpl$
 * Second Life Viewer Source Code
 * Copyright (C) 2010, Linden Research, Inc.
 *
 * This library is free software; you can redistribute it and/or
 * modify it under the terms of the GNU Lesser General Public
 * License as published by the Free Software Foundation;
 * version 2.1 of the License only.
 *
 * This library is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
 * Lesser General Public License for more details.
 *
 * You should have received a copy of the GNU Lesser General Public
 * License along with this library; if not, write to the Free Software
 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301  USA
 *
 * Linden Research, Inc., 945 Battery Street, San Francisco, CA  94111  USA
 * $/LicenseInfo$
 */

#include "linden_common.h"

#include "llfasttimer.h"
#include "llsys.h"
#include "llvertexbuffer.h"
// #include "llrender.h"
#include "llglheaders.h"
#include "llrender.h"
#include "llvector4a.h"
#include "llshadermgr.h"
#include "llglslshader.h"
#include "llmemory.h"
#include <glm/gtc/type_ptr.hpp>

//Next Highest Power Of Two
//helper function, returns first number > v that is a power of 2, or v if v is already a power of 2
U32 nhpo2(U32 v)
{
    U32 r = 1;
    while (r < v) {
        r *= 2;
    }
    return r;
}

//which power of 2 is i?
//assumes i is a power of 2 > 0
U32 wpo2(U32 i)
{
    llassert(i > 0);
    llassert(nhpo2(i) == i);

    U32 r = 0;

    while (i >>= 1) ++r;

    return r;
}

struct CompareMappedRegion
{
    bool operator()(const LLVertexBuffer::MappedRegion& lhs, const LLVertexBuffer::MappedRegion& rhs)
    {
        return lhs.mStart < rhs.mStart;
    }
};

#define ENABLE_GL_WORK_QUEUE 0

#if ENABLE_GL_WORK_QUEUE

#define THREAD_COUNT 1

//============================================================================

// High performance WorkQueue for usage in real-time rendering work
class GLWorkQueue
{
public:
    using Work = std::function<void()>;

    GLWorkQueue();

    void post(const Work& value);

    size_t size();

    bool done();

    // Get the next element from the queue
    Work pop();

    void runOne();

    bool runPending();

    void runUntilClose();

    void close();

    bool isClosed();

    void syncGL();

private:
    std::mutex mMutex;
    std::condition_variable mCondition;
    std::queue<Work> mQueue;
    bool mClosed = false;
};

GLWorkQueue::GLWorkQueue()
{

}

void GLWorkQueue::syncGL()
{
    /*if (mSync)
    {
        std::lock_guard<std::mutex> lock(mMutex);
        glWaitSync(mSync, 0, GL_TIMEOUT_IGNORED);
        mSync = 0;
    }*/
}

size_t GLWorkQueue::size()
{
    LL_PROFILE_ZONE_SCOPED_CATEGORY_THREAD;
    std::lock_guard<std::mutex> lock(mMutex);
    return mQueue.size();
}

bool GLWorkQueue::done()
{
    return size() == 0 && isClosed();
}

void GLWorkQueue::post(const GLWorkQueue::Work& value)
{
    LL_PROFILE_ZONE_SCOPED_CATEGORY_THREAD;
    {
        std::lock_guard<std::mutex> lock(mMutex);
        mQueue.push(std::move(value));
    }

    mCondition.notify_one();
}

// Get the next element from the queue
GLWorkQueue::Work GLWorkQueue::pop()
{
    LL_PROFILE_ZONE_SCOPED_CATEGORY_THREAD;
    // Lock the mutex
    {
        std::unique_lock<std::mutex> lock(mMutex);

        // Wait for a new element to become available or for the queue to close
        {
            mCondition.wait(lock, [=] { return !mQueue.empty() || mClosed; });
        }
    }

    Work ret;

    {
        std::lock_guard<std::mutex> lock(mMutex);

        // Get the next element from the queue
        if (mQueue.size() > 0)
        {
            ret = mQueue.front();
            mQueue.pop();
        }
        else
        {
            ret = []() {};
        }
    }

    return ret;
}

void GLWorkQueue::runOne()
{
    LL_PROFILE_ZONE_SCOPED_CATEGORY_THREAD;
    Work w = pop();
    w();
    //mSync = glFenceSync(GL_SYNC_GPU_COMMANDS_COMPLETE, 0);
}

void GLWorkQueue::runUntilClose()
{
    while (!isClosed())
    {
        runOne();
    }
}

void GLWorkQueue::close()
{
    LL_PROFILE_ZONE_SCOPED_CATEGORY_THREAD;
    {
        std::lock_guard<std::mutex> lock(mMutex);
        mClosed = true;
    }

    mCondition.notify_all();
}

bool GLWorkQueue::isClosed()
{
    LL_PROFILE_ZONE_SCOPED_CATEGORY_THREAD;
    std::lock_guard<std::mutex> lock(mMutex);
    return mClosed;
}

#include "llwindow.h"

class LLGLWorkerThread : public LLThread
{
public:
    LLGLWorkerThread(const std::string& name, GLWorkQueue* queue, LLWindow* window)
        : LLThread(name)
    {
        mWindow = window;
        mContext = mWindow->createSharedContext();
        mQueue = queue;
    }

    void run() override
    {
        mWindow->makeContextCurrent(mContext);
        gGL.init(false);
        mQueue->runUntilClose();
        gGL.shutdown();
        mWindow->destroySharedContext(mContext);
    }

    GLWorkQueue* mQueue;
    LLWindow* mWindow;
    void* mContext = nullptr;
};


static LLGLWorkerThread* sVBOThread[THREAD_COUNT];
static GLWorkQueue* sQueue = nullptr;

#endif

//============================================================================
// Pool of reusable VertexBuffer state

// batch calls to glGenBuffers
static GLuint gen_buffer()
{
    LL_PROFILE_ZONE_SCOPED_CATEGORY_VERTEX;

    GLuint ret = 0;
    constexpr U32 pool_size = 4096;

    thread_local static GLuint sNamePool[pool_size];
    thread_local static U32 sIndex = 0;

    if (sIndex == 0)
    {
        LL_PROFILE_ZONE_NAMED_CATEGORY_VERTEX("gen buffer");
        sIndex = pool_size;
#if !LL_DARWIN
        if (!gGLManager.mIsAMD)
        {
            glGenBuffers(pool_size, sNamePool);
        }
        else
#endif
        { // work around for AMD driver bug
            for (U32 i = 0; i < pool_size; ++i)
            {
                glGenBuffers(1, sNamePool + i);
            }
        }
    }

    ret = sNamePool[--sIndex];
    return ret;
}

static void delete_buffers(S32 count, GLuint* buffers)
{
    LL_PROFILE_ZONE_SCOPED_CATEGORY_VERTEX;
    // wait a few frames before actually deleting the buffers to avoid
    // synchronization issues with the GPU
    static std::vector<GLuint> sFreeList[4];

    if (gGLManager.mInited)
    {
        U32 idx = LLImageGL::sFrameCount % 4;

        for (S32 i = 0; i < count; ++i)
        {
            sFreeList[idx].push_back(buffers[i]);
        }

        idx = (LLImageGL::sFrameCount + 3) % 4;

        if (!sFreeList[idx].empty())
        {
            glDeleteBuffers((GLsizei)sFreeList[idx].size(), sFreeList[idx].data());
            sFreeList[idx].resize(0);
        }
    }
}


#define ANALYZE_VBO_POOL 0

// VBO Pool interface
class LLVBOPool
{
    public:
    virtual ~LLVBOPool() = default;
    virtual void allocate(GLenum type, U32 size, GLuint& name, U8*& data) = 0;
    virtual void free(GLenum type, U32 size, GLuint name, U8* data) = 0;
    virtual U64 getVramBytesUsed() = 0;
};

// VBO Pool for Apple GPUs (as in M1/M2 etc, not Intel macs)
// Effectively disables VBO pooling
class LLAppleVBOPool final: public LLVBOPool
{
public:
    U64 mAllocated = 0;

    U64 getVramBytesUsed() override
    {
        return mAllocated;
    }

    void allocate(GLenum type, U32 size, GLuint& name, U8*& data) override
    {
        LL_PROFILE_ZONE_SCOPED_CATEGORY_VERTEX;
        STOP_GLERROR;
        llassert(type == GL_ARRAY_BUFFER || type == GL_ELEMENT_ARRAY_BUFFER);
        llassert(name == 0); // non zero name indicates a gl name that wasn't freed
        llassert(data == nullptr);  // non null data indicates a buffer that wasn't freed
        llassert(size >= 2);  // any buffer size smaller than a single index is nonsensical

        mAllocated += size;

        { //allocate a new buffer
            LL_PROFILE_GPU_ZONE("vbo alloc");
            // ON OS X, we don't allocate a VBO until the last possible moment
            // in unmapBuffer
            data = (U8*) ll_aligned_malloc_16(size);
            STOP_GLERROR;
        }
    }

    void free(GLenum type, U32 size, GLuint name, U8* data) override
    {
        LL_PROFILE_ZONE_SCOPED_CATEGORY_VERTEX;
        llassert(type == GL_ARRAY_BUFFER || type == GL_ELEMENT_ARRAY_BUFFER);
        llassert(size >= 2);

        if (data)
        {
            ll_aligned_free_16(data);
        }

        mAllocated -= size;
        STOP_GLERROR;
        if (name)
        {
            delete_buffers(1, &name);
        }
        STOP_GLERROR;
    }
};

// VBO Pool for GPUs that benefit from VBO pooling
class LLDefaultVBOPool final : public LLVBOPool
{
public:
    typedef std::chrono::steady_clock::time_point Time;
    struct Entry
    {
        U8* mData;
        GLuint mGLName;
        Time mAge;
    };

    ~LLDefaultVBOPool() override
    {
        clear();
    }

    typedef std::unordered_map<U32, std::list<Entry>> Pool;

    Pool mVBOPool;
    Pool mIBOPool;

    U32 mTouchCount = 0;

    U64 mDistributed = 0;
    U64 mAllocated = 0;
    U64 mReserved = 0;
    U32 mMisses = 0;
    U32 mHits = 0;

    U64 getVramBytesUsed() override
    {
        return mAllocated + mReserved;
    }

    // increase the size to some common value (e.g. a power of two) to increase hit rate
    void adjustSize(U32& size)
    {
        // size = nhpo2(size);  // (193/303)/580 MB (distributed/allocated)/reserved in VBO Pool. Overhead: 66 percent. Hit rate: 77 percent

        //(245/276)/385 MB (distributed/allocated)/reserved in VBO Pool. Overhead: 57 percent. Hit rate: 69 percent
        //(187/209)/397 MB (distributed/allocated)/reserved in VBO Pool. Overhead: 112 percent. Hit rate: 76 percent
        U32 block_size = llmax(nhpo2(size) / 8, (U32) 16);
        size += block_size - (size % block_size);
    }

    void allocate(GLenum type, U32 size, GLuint& name, U8*& data) override
    {
        LL_PROFILE_ZONE_SCOPED_CATEGORY_VERTEX;
        llassert(type == GL_ARRAY_BUFFER || type == GL_ELEMENT_ARRAY_BUFFER);
        llassert(name == 0); // non zero name indicates a gl name that wasn't freed
        llassert(data == nullptr);  // non null data indicates a buffer that wasn't freed
        llassert(size >= 2);  // any buffer size smaller than a single index is nonsensical

        mDistributed += size;
        adjustSize(size);
        mAllocated += size;

        auto& pool = type == GL_ELEMENT_ARRAY_BUFFER ? mIBOPool : mVBOPool;

        Pool::iterator iter = pool.find(size);
        if (iter == pool.end())
        { // cache miss, allocate a new buffer
            LL_PROFILE_ZONE_NAMED_CATEGORY_VERTEX("vbo pool miss");
            LL_PROFILE_GPU_ZONE("vbo alloc");

            mMisses++;
            name = gen_buffer();
            glBindBuffer(type, name);
            glBufferData(type, size, nullptr, GL_DYNAMIC_DRAW);
            if (type == GL_ELEMENT_ARRAY_BUFFER)
            {
                LLVertexBuffer::sGLRenderIndices = name;
            }
            else
            {
                LLVertexBuffer::sGLRenderBuffer = name;
            }

            data = (U8*)ll_aligned_malloc_16(size);
        }
        else
        {
            mHits++;
            llassert(mReserved >= size); // assert if accounting gets messed up
            mReserved -= size;

            std::list<Entry>& entries = iter->second;
            Entry& entry = entries.back();
            name = entry.mGLName;
            data = entry.mData;

            entries.pop_back();
            if (entries.empty())
            {
                pool.erase(iter);
            }
        }

        clean();
    }

    void free(GLenum type, U32 size, GLuint name, U8* data) override
    {
        LL_PROFILE_ZONE_SCOPED_CATEGORY_VERTEX;
        llassert(type == GL_ARRAY_BUFFER || type == GL_ELEMENT_ARRAY_BUFFER);
        llassert(size >= 2);
        llassert(name != 0);
        llassert(data != nullptr);

        clean();

        llassert(mDistributed >= size);
        mDistributed -= size;
        adjustSize(size);
        llassert(mAllocated >= size);
        mAllocated -= size;
        mReserved += size;

        auto& pool = type == GL_ELEMENT_ARRAY_BUFFER ? mIBOPool : mVBOPool;

        Pool::iterator iter = pool.find(size);

        if (iter == pool.end())
        {
            std::list<Entry> newlist;
            newlist.push_front({ data, name, std::chrono::steady_clock::now() });
            pool[size] = newlist;
        }
        else
        {
            iter->second.push_front({ data, name, std::chrono::steady_clock::now() });
        }

    }

    // clean periodically (clean gets called for every alloc/free)
    void clean()
    {
        mTouchCount++;
        if (mTouchCount < 1024) // clean every 1k touches
        {
            return;
        }
        mTouchCount = 0;

        LL_PROFILE_ZONE_SCOPED_CATEGORY_VERTEX;

        std::unordered_map<U32, std::list<Entry>>* pools[] = { &mVBOPool, &mIBOPool };

        using namespace std::chrono_literals;

        Time cutoff = std::chrono::steady_clock::now() - 5s;

        for (auto* pool : pools)
        {
            for (Pool::iterator iter = pool->begin(); iter != pool->end(); )
            {
                auto& entries = iter->second;

                while (!entries.empty() && entries.back().mAge < cutoff)
                {
                    LL_PROFILE_ZONE_NAMED_CATEGORY_VERTEX("vbo cache timeout");
                    auto& entry = entries.back();
                    ll_aligned_free_16(entry.mData);
                    delete_buffers(1, &entry.mGLName);
                    llassert(mReserved >= iter->first);
                    mReserved -= iter->first;
                    entries.pop_back();

                }

                if (entries.empty())
                {
                    iter = pool->erase(iter);
                }
                else
                {
                    ++iter;
                }
            }
        }

#if 0
        LL_INFOS() << llformat("(%d/%d)/%d MB (distributed/allocated)/total in VBO Pool. Overhead: %d percent. Hit rate: %d percent",
            mDistributed / 1000000,
            mAllocated / 1000000,
            (mAllocated + mReserved) / 1000000, // total bytes
            ((mAllocated+mReserved-mDistributed)*100)/llmax(mDistributed, (U64) 1), // overhead percent
            (mHits*100)/llmax(mMisses+mHits, (U32)1)) // hit rate percent
            << LL_ENDL;
#endif
    }

    void clear()
    {
        for (auto& entries : mIBOPool)
        {
            for (auto& entry : entries.second)
            {
                ll_aligned_free_16(entry.mData);
                delete_buffers(1, &entry.mGLName);
            }
        }

        for (auto& entries : mVBOPool)
        {
            for (auto& entry : entries.second)
            {
                ll_aligned_free_16(entry.mData);
                delete_buffers(1, &entry.mGLName);
            }
        }

        mReserved = 0;

        mIBOPool.clear();
        mVBOPool.clear();
    }
};

static LLVBOPool* sVBOPool = nullptr;

void LLVertexBufferData::drawWithMatrix()
{
    if (!mVB)
    {
        llassert(false);
        // Not supposed to happen, check buffer generation
        return;
    }

    if (mTexName)
    {
        gGL.getTexUnit(0)->bindManual(LLTexUnit::TT_TEXTURE, mTexName);
    }
    else
    {
        gGL.getTexUnit(0)->unbind(LLTexUnit::TT_TEXTURE);
    }

    gGL.matrixMode(LLRender::MM_MODELVIEW);
    gGL.pushMatrix();
    gGL.loadMatrix(glm::value_ptr(mModelView));
    gGL.matrixMode(LLRender::MM_PROJECTION);
    gGL.pushMatrix();
    gGL.loadMatrix(glm::value_ptr(mProjection));
    gGL.matrixMode(LLRender::MM_TEXTURE0);
    gGL.pushMatrix();
    gGL.loadMatrix(glm::value_ptr(mTexture0));

    mVB->setBuffer();
    mVB->drawArrays(mMode, 0, mCount);

    gGL.popMatrix();
    gGL.matrixMode(LLRender::MM_PROJECTION);
    gGL.popMatrix();
    gGL.matrixMode(LLRender::MM_MODELVIEW);
    gGL.popMatrix();
}

void LLVertexBufferData::draw()
{
    if (!mVB)
    {
        llassert(false);
        // Not supposed to happen, check buffer generation
        return;
    }

    if (mTexName)
    {
        gGL.getTexUnit(0)->bindManual(LLTexUnit::TT_TEXTURE, mTexName);
    }
    else
    {
        gGL.getTexUnit(0)->unbind(LLTexUnit::TT_TEXTURE);
    }

    mVB->setBuffer();
    mVB->drawArrays(mMode, 0, mCount);
}

//============================================================================

//static
U64 LLVertexBuffer::getBytesAllocated()
{
    return sVBOPool ? sVBOPool->getVramBytesUsed() : 0;
}

//============================================================================
//
//static
U32 LLVertexBuffer::sGLRenderBuffer = 0;
U32 LLVertexBuffer::sGLRenderIndices = 0;
U32 LLVertexBuffer::sLastMask = 0;
U32 LLVertexBuffer::sVertexCount = 0;

// The viewer keeps a single VAO bound. Attribute pointer calls store state in
// that VAO and capture the current array buffer. Track the formats configured
// for sGLRenderBuffer to avoid resubmitting identical state to OpenGL-on-Metal.
// If VAO switching is introduced, invalidate this cache when the VAO changes.
static U32 sVertexAttribsConfigured = 0;
static U32 sColorPointerSource = 0;
static constexpr U32 COLOR_POINTER_COLOR = 1;
static constexpr U32 COLOR_POINTER_EMISSIVE = 2;


//NOTE: each component must be AT LEAST 4 bytes in size to avoid a performance penalty on AMD hardware
const U32 LLVertexBuffer::sTypeSize[LLVertexBuffer::TYPE_MAX] =
{
    sizeof(LLVector4), // TYPE_VERTEX,
    sizeof(LLVector4), // TYPE_NORMAL,
    sizeof(LLVector2), // TYPE_TEXCOORD0,
    sizeof(LLVector2), // TYPE_TEXCOORD1,
    sizeof(LLVector2), // TYPE_TEXCOORD2,
    sizeof(LLVector2), // TYPE_TEXCOORD3,
    sizeof(LLColor4U), // TYPE_COLOR,
    sizeof(LLColor4U), // TYPE_EMISSIVE, only alpha is used currently
    sizeof(LLVector4), // TYPE_TANGENT,
    sizeof(F32),       // TYPE_WEIGHT,
    sizeof(LLVector4), // TYPE_WEIGHT4,
    sizeof(LLVector4), // TYPE_CLOTHWEIGHT,
    sizeof(U64),       // TYPE_JOINT,
    sizeof(LLVector4), // TYPE_TEXTURE_INDEX (actually exists as position.w), no extra data, but stride is 16 bytes
};

static const std::string vb_type_name[] =
{
    "TYPE_VERTEX",
    "TYPE_NORMAL",
    "TYPE_TEXCOORD0",
    "TYPE_TEXCOORD1",
    "TYPE_TEXCOORD2",
    "TYPE_TEXCOORD3",
    "TYPE_COLOR",
    "TYPE_EMISSIVE",
    "TYPE_TANGENT",
    "TYPE_WEIGHT",
    "TYPE_WEIGHT4",
    "TYPE_CLOTHWEIGHT",
    "TYPE_JOINT"
    "TYPE_TEXTURE_INDEX",
    "TYPE_MAX",
    "TYPE_INDEX",
};

const U32 LLVertexBuffer::sGLMode[LLRender::NUM_MODES] =
{
    GL_TRIANGLES,
    GL_TRIANGLE_STRIP,
    GL_TRIANGLE_FAN,
    GL_POINTS,
    GL_LINES,
    GL_LINE_STRIP,
    GL_LINE_LOOP,
};

//static
void LLVertexBuffer::setupClientArrays(U32 data_mask)
{
    if (sLastMask != data_mask)
    {
        for (U32 i = 0; i < TYPE_MAX; ++i)
        {
            S32 loc = i;

            U32 mask = 1 << i;

            if (sLastMask & (1 << i))
            { //was enabled
                if (!(data_mask & mask))
                { //needs to be disabled
                    glDisableVertexAttribArray(loc);
                }
            }
            else
            {   //was disabled
                if (data_mask & mask)
                { //needs to be enabled
                    glEnableVertexAttribArray(loc);
                }
            }
        }
    }

    sLastMask = data_mask;
}

//static
void LLVertexBuffer::drawArrays(U32 mode, const std::vector<LLVector3>& pos)
{
    LL_PROFILE_ZONE_SCOPED_CATEGORY_VERTEX;
    gGL.begin(mode);
    for (auto& v : pos)
    {
        gGL.vertex3fv(v.mV);
    }
    gGL.end();
    gGL.flush();
}

//static
void LLVertexBuffer::drawElements(U32 mode, const LLVector4a* pos, const LLVector2* tc, U32 num_indices, const U16* indicesp)
{
    LL_PROFILE_ZONE_SCOPED_CATEGORY_VERTEX;
    llassert(LLGLSLShader::sCurBoundShaderPtr != NULL);

    STOP_GLERROR;

    gGL.syncMatrices();

    unbind();

    gGL.begin(mode);

    if (tc != nullptr)
    {
        for (U32 i = 0; i < num_indices; ++i)
        {
            U16 idx = indicesp[i];
            gGL.texCoord2fv(tc[idx].mV);
            gGL.vertex3fv(pos[idx].getF32ptr());
        }
    }
    else
    {
        for (U32 i = 0; i < num_indices; ++i)
        {
            U16 idx = indicesp[i];
            gGL.vertex3fv(pos[idx].getF32ptr());
        }
    }
    gGL.end();
    gGL.flush();
}

bool LLVertexBuffer::validateRange(U32 start, U32 end, U32 count, U32 indices_offset) const
{
    if (!gDebugGL)
    {
        return true;
    }

    llassert(start < mNumVerts);
    llassert(end < mNumVerts);

    if (start >= mNumVerts ||
        end >= mNumVerts)
    {
        LL_ERRS() << "Bad vertex buffer draw range: [" << start << ", " << end << "] vs " << mNumVerts << LL_ENDL;
    }

    if (indices_offset >= mNumIndices ||
        indices_offset + count > mNumIndices)
    {
        LL_ERRS() << "Bad index buffer draw range: [" << indices_offset << ", " << indices_offset+count << "]" << LL_ENDL;
    }

    {
#if 0  // not a reliable test for VBOs that are not backed by a CPU buffer
        U16* idx = (U16*) mMappedIndexData+indices_offset;
        for (U32 i = 0; i < count; ++i)
        {
            llassert(idx[i] >= start);
            llassert(idx[i] <= end);

            if (idx[i] < start || idx[i] > end)
            {
                LL_ERRS() << "Index out of range: " << idx[i] << " not in [" << start << ", " << end << "]" << LL_ENDL;
            }
        }

        LLVector4a* v = (LLVector4a*)mMappedData;

        for (U32 i = start; i <= end; ++i)
        {
            if (!v[i].isFinite3())
            {
                LL_ERRS() << "Non-finite vertex position data detected." << LL_ENDL;
            }
        }

        LLGLSLShader* shader = LLGLSLShader::sCurBoundShaderPtr;

        if (shader && shader->mFeatures.mIndexedTextureChannels > 1)
        {
            LLVector4a* v = (LLVector4a*) mMappedData;

            for (U32 i = start; i < end; i++)
            {
                U32 idx = (U32) (v[i][3]+0.25f);
                if (idx >= (U32)shader->mFeatures.mIndexedTextureChannels)
                {
                    LL_ERRS() << "Bad texture index found in vertex data stream." << LL_ENDL;
                }
            }
        }
#endif
    }

    return true;
}

#if LL_PROFILER_ENABLE_RENDER_DOC
void LLVertexBuffer::setLabel(const char* label) {
    LL_LABEL_OBJECT_GL(GL_BUFFER, mGLBuffer, strlen(label), label);
}
#endif

void LLVertexBuffer::clone(LLVertexBuffer& target) const
{
    target.mTypeMask = mTypeMask;
    target.mIndicesType = mIndicesType;
    target.mIndicesStride = mIndicesStride;
    if (target.getNumVerts() != getNumVerts() ||
        target.getNumIndices() != getNumIndices())
    {
        target.allocateBuffer(getNumVerts(), getNumIndices());
    }
}

void LLVertexBuffer::drawRange(U32 mode, U32 start, U32 end, U32 count, U32 indices_offset) const
{
    llassert(validateRange(start, end, count, indices_offset));
    llassert(mGLBuffer == sGLRenderBuffer);
    llassert(mGLIndices == sGLRenderIndices);
    gGL.syncMatrices();
    STOP_GLERROR;
    glDrawRangeElements(sGLMode[mode], start, end, count, mIndicesType,
        (GLvoid*) (indices_offset * (size_t) mIndicesStride));
    STOP_GLERROR;
}

void LLVertexBuffer::drawRangeFast(U32 mode, U32 start, U32 end, U32 count, U32 indices_offset) const
{
    glDrawRangeElements(sGLMode[mode], start, end, count, mIndicesType,
        (GLvoid*)(indices_offset * (size_t)mIndicesStride));
}


void LLVertexBuffer::draw(U32 mode, U32 count, U32 indices_offset) const
{
    drawRange(mode, 0, mNumVerts-1, count, indices_offset);
}


void LLVertexBuffer::drawArrays(U32 mode, U32 first, U32 count) const
{
    llassert(first + count <= mNumVerts);
    llassert(mGLBuffer == sGLRenderBuffer);
    llassert(mGLIndices == sGLRenderIndices);

    gGL.syncMatrices();
    STOP_GLERROR;
    glDrawArrays(sGLMode[mode], first, count);
    STOP_GLERROR;
}

//static
void LLVertexBuffer::initClass(LLWindow* window)
{
    llassert(sVBOPool == nullptr);

#if LL_DARWIN || LL_ARM64
    if (gGLManager.mIsApple)
    {
        LL_INFOS() << "VBO Pooling Disabled" << LL_ENDL;
        sVBOPool = new LLAppleVBOPool();
    }
    else
#endif
    {
        LL_INFOS() << "VBO Pooling Enabled" << LL_ENDL;
        sVBOPool = new LLDefaultVBOPool();
    }

#if ENABLE_GL_WORK_QUEUE
    sQueue = new GLWorkQueue();

    for (int i = 0; i < THREAD_COUNT; ++i)
    {
        sVBOThread[i] = new LLGLWorkerThread("VBO Worker", sQueue, window);
        sVBOThread[i]->start();
    }
#endif
}

//static
void LLVertexBuffer::unbind()
{
    STOP_GLERROR;
    glBindBuffer(GL_ARRAY_BUFFER, 0);
    glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, 0);
    STOP_GLERROR;
    sGLRenderBuffer = 0;
    sGLRenderIndices = 0;
    sVertexAttribsConfigured = 0;
    sColorPointerSource = 0;
}

//static
void LLVertexBuffer::cleanupClass()
{
    unbind();

    delete sVBOPool;
    sVBOPool = nullptr;

#if ENABLE_GL_WORK_QUEUE
    sQueue->close();
    for (int i = 0; i < THREAD_COUNT; ++i)
    {
        sVBOThread[i]->shutdown();
        delete sVBOThread[i];
        sVBOThread[i] = nullptr;
    }

    delete sQueue;
    sQueue = nullptr;
#endif
}

//----------------------------------------------------------------------------

LLVertexBuffer::LLVertexBuffer(U32 typemask)
:   LLRefCount(),
    mTypeMask(typemask)
{
    //zero out offsets
    for (U32 i = 0; i < TYPE_MAX; i++)
    {
        mOffsets[i] = 0;
    }
}

// list of mapped buffers
// NOTE: must not be LLPointer<LLVertexBuffer> to avoid breaking non-ref-counted LLVertexBuffer instances
static std::vector<LLVertexBuffer*> sMappedBuffers;

//static
void LLVertexBuffer::flushBuffers()
{
    LL_PROFILE_ZONE_SCOPED_CATEGORY_VERTEX;
    // must only be called from main thread
    for (auto& buffer : sMappedBuffers)
    {
        buffer->_unmapBuffer();
        buffer->mMapped = false;
    }

    sMappedBuffers.resize(0);
}

//static
U32 LLVertexBuffer::calcOffsets(const U32& typemask, U32* offsets, U32 num_vertices)
{
    U32 offset = 0;
    for (U32 i=0; i<TYPE_TEXTURE_INDEX; i++)
    {
        U32 mask = 1<<i;
        if (typemask & mask)
        {
            if (offsets && LLVertexBuffer::sTypeSize[i])
            {
                offsets[i] = offset;
                offset += LLVertexBuffer::sTypeSize[i]*num_vertices;
                offset = (offset + 0xF) & ~0xF;
            }
        }
    }

    offsets[TYPE_TEXTURE_INDEX] = offsets[TYPE_VERTEX] + 12;

    return offset;
}

//static
U32 LLVertexBuffer::calcVertexSize(const U32& typemask)
{
    U32 size = 0;
    for (U32 i = 0; i < TYPE_TEXTURE_INDEX; i++)
    {
        U32 mask = 1<<i;
        if (typemask & mask)
        {
            size += LLVertexBuffer::sTypeSize[i];
        }
    }

    return size;
}

// protected, use unref()
//virtual
LLVertexBuffer::~LLVertexBuffer()
{
    if (mMapped)
    { // is on the mapped buffer list but doesn't need to be flushed
        mMapped = false;
        unmapBuffer();
    }

    destroyGLBuffer();
    destroyGLIndices();

    if (mMappedData)
    {
        LL_ERRS() << "Failed to clear vertex buffer's vertices" << LL_ENDL;
    }
    if (mMappedIndexData)
    {
        LL_ERRS() << "Failed to clear vertex buffer's indices" << LL_ENDL;
    }
};

//----------------------------------------------------------------------------

void LLVertexBuffer::genBuffer(U32 size)
{
    LL_PROFILE_ZONE_SCOPED_CATEGORY_VERTEX;
    llassert(sVBOPool);

    if (sVBOPool)
    {
        llassert(mSize == 0);
        llassert(mGLBuffer == 0);
        llassert(mMappedData == nullptr);

        mSize = size;
        sVBOPool->allocate(GL_ARRAY_BUFFER, mSize, mGLBuffer, mMappedData);
    }
}

void LLVertexBuffer::genIndices(U32 size)
{
    LL_PROFILE_ZONE_SCOPED_CATEGORY_VERTEX;
    llassert(sVBOPool);

    if (sVBOPool)
    {
        llassert(mIndicesSize == 0);
        llassert(mGLIndices == 0);
        llassert(mMappedIndexData == nullptr);
        mIndicesSize = size;
        sVBOPool->allocate(GL_ELEMENT_ARRAY_BUFFER, mIndicesSize, mGLIndices, mMappedIndexData);
    }
}

bool LLVertexBuffer::createGLBuffer(U32 size)
{
    if (mGLBuffer || mMappedData)
    {
        destroyGLBuffer();
    }

    if (size == 0)
    {
        return true;
    }

    bool success = true;

    genBuffer(size);

    if (!mMappedData)
    {
        success = false;
    }
    return success;
}

bool LLVertexBuffer::createGLIndices(U32 size)
{
    if (mGLIndices)
    {
        destroyGLIndices();
    }

    if (size == 0)
    {
        return true;
    }

    bool success = true;

    genIndices(size);

    if (!mMappedIndexData)
    {
        success = false;
    }
    return success;
}

void LLVertexBuffer::destroyGLBuffer()
{
    if (mGLBuffer || mMappedData)
    {
        LL_PROFILE_ZONE_SCOPED_CATEGORY_VERTEX;
        //llassert(sVBOPool);
        if (sVBOPool)
        {
            sVBOPool->free(GL_ARRAY_BUFFER, mSize, mGLBuffer, mMappedData);
        }

        mSize = 0;
        mGLBuffer = 0;
        mMappedData = nullptr;
    }
}

void LLVertexBuffer::destroyGLIndices()
{
    if (mGLIndices || mMappedIndexData)
    {
        LL_PROFILE_ZONE_SCOPED_CATEGORY_VERTEX;
        //llassert(sVBOPool);
        if (sVBOPool)
        {
            sVBOPool->free(GL_ELEMENT_ARRAY_BUFFER, mIndicesSize, mGLIndices, mMappedIndexData);
        }

        mIndicesSize = 0;
        mGLIndices = 0;
        mMappedIndexData = nullptr;
    }
}

bool LLVertexBuffer::updateNumVerts(U32 nverts)
{
    llassert(nverts >= 0);

    bool success = true;

    U32 needed_size = calcOffsets(mTypeMask, mOffsets, nverts);

    if (needed_size != mSize)
    {
        success &= createGLBuffer(needed_size);
    }

    llassert(mSize == needed_size);
    mNumVerts = nverts;
    return success;
}

bool LLVertexBuffer::updateNumIndices(U32 nindices)
{
    llassert(nindices >= 0);

    bool success = true;

    U32 needed_size = sizeof(U16) * nindices;

    if (needed_size != mIndicesSize)
    {
        success &= createGLIndices(needed_size);
    }

    llassert(mIndicesSize == needed_size);
    mNumIndices = nindices;
    return success;
}

bool LLVertexBuffer::allocateBuffer(U32 nverts, U32 nindices)
{
    if (nverts < 0 || nindices < 0)
    {
        LL_ERRS() << "Bad vertex buffer allocation: " << nverts << " : " << nindices << LL_ENDL;
    }

    bool success = true;

    success &= updateNumVerts(nverts);
    success &= updateNumIndices(nindices);

    return success;
}

//----------------------------------------------------------------------------

// if no gap between region and given range exists, expand region to cover given range and return true
// otherwise return false
bool expand_region(LLVertexBuffer::MappedRegion& region, U32 start, U32 end)
{

    if (end < region.mStart ||
        start > region.mEnd)
    { //gap exists, do not merge
        return false;
    }

    region.mStart = llmin(region.mStart, start);
    region.mEnd = llmax(region.mEnd, end);

    return true;
}


// Map for data access
U8* LLVertexBuffer::mapVertexBuffer(LLVertexBuffer::AttributeType type, U32 index, S32 count)
{
    LL_PROFILE_ZONE_SCOPED_CATEGORY_VERTEX;
    _mapBuffer();

    if (count == -1)
    {
        count = mNumVerts - index;
    }

#if LL_DARWIN || LL_ARM64
    // Region tracking not needed on apple silicon - it recreates entire buffer
    // While mIsApple can be encountered under windows, this is a
    // macOS OpenGL behavior workaround. LL_ARM64 check might be not needed
    if (!gGLManager.mIsApple)
#endif
    {
        U32 start = mOffsets[type] + sTypeSize[type] * index;
        U32 end = start + sTypeSize[type] * count-1;

        bool flagged = false;
        // flag region as mapped
        for (U32 i = 0; i < mMappedVertexRegions.size(); ++i)
        {
            MappedRegion& region = mMappedVertexRegions[i];
            if (expand_region(region, start, end))
            {
                flagged = true;
                break;
            }
        }

        if (!flagged)
        {
            //didn't expand an existing region, make a new one
            mMappedVertexRegions.push_back({ start, end });
        }
    }
    return mMappedData+mOffsets[type]+sTypeSize[type]*index;
}


U8* LLVertexBuffer::mapIndexBuffer(U32 index, S32 count)
{
    LL_PROFILE_ZONE_SCOPED_CATEGORY_VERTEX;
    _mapBuffer();

    if (count == -1)
    {
        count = mNumIndices-index;
    }

#if LL_DARWIN || LL_ARM64
    if (!gGLManager.mIsApple)
#endif
    {
        U32 start = sizeof(U16) * index;
        U32 end = start + sizeof(U16) * count-1;

        bool flagged = false;
        // flag region as mapped
        for (U32 i = 0; i < mMappedIndexRegions.size(); ++i)
        {
            MappedRegion& region = mMappedIndexRegions[i];
            if (expand_region(region, start, end))
            {
                flagged = true;
                break;
            }
        }

        if (!flagged)
        {
            //didn't expand an existing region, make a new one
            mMappedIndexRegions.push_back({ start, end });
        }
    }

    return mMappedIndexData + sizeof(U16)*index;
}

// flush the given byte range
//  target -- "target" parameter for glBufferSubData
//  start -- first byte to copy
//  end -- last byte to copy (NOT last byte + 1)
//  data -- data to be flushed
//  dst -- mMappedData or mMappedIndexData
void LLVertexBuffer::flush_vbo(GLenum target, U32 start, U32 end, void* data, U8* dst)
{
#if LL_DARWIN || LL_ARM64
    if (gGLManager.mIsApple)
    {
        // on OS X, flush_vbo doesn't actually write to the GL buffer, so be sure to call
        // _mapBuffer to tag the buffer for flushing to GL
        _mapBuffer();
        LL_PROFILE_ZONE_NAMED_CATEGORY_VERTEX("vb memcpy");
        STOP_GLERROR;
        // copy into mapped buffer
        memcpy(dst+start, data, end-start+1);
    }
    else
#endif
    {
        llassert(target == GL_ARRAY_BUFFER ? sGLRenderBuffer == mGLBuffer : sGLRenderIndices == mGLIndices);

        // skip mapped data and stream to GPU via glBufferSubData
        if (end != 0)
        {
            LL_PROFILE_ZONE_NAMED_CATEGORY_VERTEX("glBufferSubData");
            LL_PROFILE_ZONE_NUM(start);
            LL_PROFILE_ZONE_NUM(end);
            LL_PROFILE_ZONE_NUM(end-start);

            constexpr U32 block_size = 65536;

            for (U32 i = start; i <= end; i += block_size)
            {
                //LL_PROFILE_ZONE_NAMED_CATEGORY_VERTEX("glBufferSubData block");
                //LL_PROFILE_GPU_ZONE("glBufferSubData");
                U32 tend = llmin(i + block_size, end);
                U32 size = tend - i + 1;
                glBufferSubData(target, i, size, (U8*) data + (i-start));
            }
        }
    }
}

void LLVertexBuffer::unmapBuffer()
{
    flushBuffers();
}

void LLVertexBuffer::_mapBuffer()
{
    if (!mMapped)
    {
        mMapped = true;
        sMappedBuffers.push_back(this);
    }
}

void LLVertexBuffer::_unmapBuffer()
{
    STOP_GLERROR;
    if (!mMapped)
    {
        return;
    }

    struct SortMappedRegion
    {
        bool operator()(const MappedRegion& lhs, const MappedRegion& rhs)
        {
            return lhs.mStart < rhs.mStart;
        }
    };

#if LL_DARWIN || LL_ARM64
    if (gGLManager.mIsApple)
    {
        STOP_GLERROR;
        if (mMappedData)
        {
            if (mGLBuffer)
            {
                delete_buffers(1, &mGLBuffer);
            }
            mGLBuffer = gen_buffer();
            glBindBuffer(GL_ARRAY_BUFFER, mGLBuffer);
            sGLRenderBuffer = mGLBuffer;
            sVertexAttribsConfigured = 0;
            sColorPointerSource = 0;
            glBufferData(GL_ARRAY_BUFFER, mSize, mMappedData, GL_STATIC_DRAW);
        }
        else if (mGLBuffer != sGLRenderBuffer)
        {
            glBindBuffer(GL_ARRAY_BUFFER, mGLBuffer);
            sGLRenderBuffer = mGLBuffer;
            sVertexAttribsConfigured = 0;
            sColorPointerSource = 0;
        }
        STOP_GLERROR;

        if (mMappedIndexData)
        {
            if (mGLIndices)
            {
                delete_buffers(1, &mGLIndices);
            }

            mGLIndices = gen_buffer();
            glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, mGLIndices);
            sGLRenderIndices = mGLIndices;

            glBufferData(GL_ELEMENT_ARRAY_BUFFER, mIndicesSize, mMappedIndexData, GL_STATIC_DRAW);
        }
        else if (mGLIndices != sGLRenderIndices)
        {
            glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, mGLIndices);
            sGLRenderIndices = mGLIndices;
        }
        STOP_GLERROR;
    }
    else
#endif // LL_DARWIN || LL_ARM64
    {
        if (!mMappedVertexRegions.empty())
        {
            LL_PROFILE_ZONE_NAMED_CATEGORY_VERTEX("unmapBuffer - vertex");

            if (sGLRenderBuffer != mGLBuffer)
            {
                glBindBuffer(GL_ARRAY_BUFFER, mGLBuffer);
                sGLRenderBuffer = mGLBuffer;
            }

            U32 start = 0;
            U32 end = 0;

            std::sort(mMappedVertexRegions.begin(), mMappedVertexRegions.end(), SortMappedRegion());

            for (U32 i = 0; i < mMappedVertexRegions.size(); ++i)
            {
                const MappedRegion& region = mMappedVertexRegions[i];
                if (region.mStart == end + 1)
                {
                    end = region.mEnd;
                }
                else
                {
                    flush_vbo(GL_ARRAY_BUFFER, start, end, (U8*)mMappedData + start, mMappedData);
                    start = region.mStart;
                    end = region.mEnd;
                }
            }

            flush_vbo(GL_ARRAY_BUFFER, start, end, (U8*)mMappedData + start, mMappedData);
            mMappedVertexRegions.clear();
        }

        if (!mMappedIndexRegions.empty())
        {
            LL_PROFILE_ZONE_NAMED_CATEGORY_VERTEX("unmapBuffer - index");

            if (mGLIndices != sGLRenderIndices)
            {
                glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, mGLIndices);
                sGLRenderIndices = mGLIndices;
            }
            U32 start = 0;
            U32 end = 0;

            std::sort(mMappedIndexRegions.begin(), mMappedIndexRegions.end(), SortMappedRegion());

            for (U32 i = 0; i < mMappedIndexRegions.size(); ++i)
            {
                const MappedRegion& region = mMappedIndexRegions[i];
                if (region.mStart == end + 1)
                {
                    end = region.mEnd;
                }
                else
                {
                    flush_vbo(GL_ELEMENT_ARRAY_BUFFER, start, end, (U8*)mMappedIndexData + start, mMappedIndexData);
                    start = region.mStart;
                    end = region.mEnd;
                }
            }

            flush_vbo(GL_ELEMENT_ARRAY_BUFFER, start, end, (U8*)mMappedIndexData + start, mMappedIndexData);
            mMappedIndexRegions.clear();
        }
    }
}

//----------------------------------------------------------------------------

template <class T,LLVertexBuffer::AttributeType type> struct VertexBufferStrider
{
    typedef LLStrider<T> strider_t;
    static bool get(LLVertexBuffer& vbo,
                    strider_t& strider,
                    S32 index, S32 count)
    {
        if (type == LLVertexBuffer::TYPE_INDEX)
        {
            U8* ptr = vbo.mapIndexBuffer(index, count);

            if (ptr == NULL)
            {
                LL_WARNS() << "mapIndexBuffer failed!" << LL_ENDL;
                return false;
            }

            strider = (T*)ptr;
            strider.setStride(0);
            return true;
        }
        else if (vbo.hasDataType(type))
        {
            U32 stride = LLVertexBuffer::sTypeSize[type];

            U8* ptr = vbo.mapVertexBuffer(type, index, count);

            if (ptr == NULL)
            {
                LL_WARNS() << "mapVertexBuffer failed!" << LL_ENDL;
                return false;
            }

            strider = (T*)ptr;
            strider.setStride(stride);
            return true;
        }
        else
        {
            LL_ERRS() << "VertexBufferStrider could not find valid vertex data." << LL_ENDL;
        }
        return false;
    }
};

bool LLVertexBuffer::getVertexStrider(LLStrider<LLVector3>& strider, U32 index, S32 count)
{
    return VertexBufferStrider<LLVector3,TYPE_VERTEX>::get(*this, strider, index, count);
}
bool LLVertexBuffer::getVertexStrider(LLStrider<LLVector4a>& strider, U32 index, S32 count)
{
    return VertexBufferStrider<LLVector4a,TYPE_VERTEX>::get(*this, strider, index, count);
}
bool LLVertexBuffer::getIndexStrider(LLStrider<U16>& strider, U32 index, S32 count)
{
    llassert(mIndicesStride == 2); // cannot access 32-bit indices with U16 strider
    llassert(mIndicesType == GL_UNSIGNED_SHORT);
    return VertexBufferStrider<U16,TYPE_INDEX>::get(*this, strider, index, count);
}
bool LLVertexBuffer::getTexCoord0Strider(LLStrider<LLVector2>& strider, U32 index, S32 count)
{
    return VertexBufferStrider<LLVector2,TYPE_TEXCOORD0>::get(*this, strider, index, count);
}
bool LLVertexBuffer::getTexCoord1Strider(LLStrider<LLVector2>& strider, U32 index, S32 count)
{
    return VertexBufferStrider<LLVector2,TYPE_TEXCOORD1>::get(*this, strider, index, count);
}
bool LLVertexBuffer::getTexCoord2Strider(LLStrider<LLVector2>& strider, U32 index, S32 count)
{
    return VertexBufferStrider<LLVector2,TYPE_TEXCOORD2>::get(*this, strider, index, count);
}
bool LLVertexBuffer::getNormalStrider(LLStrider<LLVector3>& strider, U32 index, S32 count)
{
    return VertexBufferStrider<LLVector3,TYPE_NORMAL>::get(*this, strider, index, count);
}
bool LLVertexBuffer::getNormalStrider(LLStrider<LLVector4a>& strider, U32 index, S32 count)
{
    return VertexBufferStrider<LLVector4a, TYPE_NORMAL>::get(*this, strider, index, count);
}
bool LLVertexBuffer::getTangentStrider(LLStrider<LLVector3>& strider, U32 index, S32 count)
{
    return VertexBufferStrider<LLVector3,TYPE_TANGENT>::get(*this, strider, index, count);
}
bool LLVertexBuffer::getTangentStrider(LLStrider<LLVector4a>& strider, U32 index, S32 count)
{
    return VertexBufferStrider<LLVector4a,TYPE_TANGENT>::get(*this, strider, index, count);
}
bool LLVertexBuffer::getColorStrider(LLStrider<LLColor4U>& strider, U32 index, S32 count)
{
    return VertexBufferStrider<LLColor4U,TYPE_COLOR>::get(*this, strider, index, count);
}
bool LLVertexBuffer::getEmissiveStrider(LLStrider<LLColor4U>& strider, U32 index, S32 count)
{
    return VertexBufferStrider<LLColor4U,TYPE_EMISSIVE>::get(*this, strider, index, count);
}
bool LLVertexBuffer::getWeightStrider(LLStrider<F32>& strider, U32 index, S32 count)
{
    return VertexBufferStrider<F32,TYPE_WEIGHT>::get(*this, strider, index, count);
}

bool LLVertexBuffer::getWeight4Strider(LLStrider<LLVector4>& strider, U32 index, S32 count)
{
    return VertexBufferStrider<LLVector4,TYPE_WEIGHT4>::get(*this, strider, index, count);
}

bool LLVertexBuffer::getClothWeightStrider(LLStrider<LLVector4>& strider, U32 index, S32 count)
{
    return VertexBufferStrider<LLVector4,TYPE_CLOTHWEIGHT>::get(*this, strider, index, count);
}

//----------------------------------------------------------------------------


// Set for rendering
void LLVertexBuffer::setBuffer()
{
    STOP_GLERROR;

    if (mMapped)
    {
        LL_WARNS_ONCE() << "Missing call to unmapBuffer or flushBuffers" << LL_ENDL;
        _unmapBuffer();
    }

    // no data may be pending
    llassert(mMappedVertexRegions.empty());
    llassert(mMappedIndexRegions.empty());

    // a shader must be bound
    llassert(LLGLSLShader::sCurBoundShaderPtr);

    U32 data_mask = LLGLSLShader::sCurBoundShaderPtr->mAttributeMask;

    // this Vertex Buffer must provide all necessary attributes for currently bound shader
    llassert_msg((data_mask & mTypeMask) == data_mask,
        "Attribute mask mismatch! mTypeMask should be a superset of data_mask.  data_mask: 0x"
                << std::hex << data_mask << " mTypeMask: 0x" << mTypeMask << " Missing: 0x" << (data_mask & ~mTypeMask) <<  std::dec);

    if (sGLRenderBuffer != mGLBuffer)
    {
        glBindBuffer(GL_ARRAY_BUFFER, mGLBuffer);
        sGLRenderBuffer = mGLBuffer;
        sVertexAttribsConfigured = 0;
        sColorPointerSource = 0;

        setupVertexBuffer();
    }
    else if (gGLManager.mIsApple || sLastMask != data_mask)
    {
        setupVertexBuffer();
        sLastMask = data_mask;
    }

    if (mGLIndices != sGLRenderIndices)
    {
        glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, mGLIndices);
        sGLRenderIndices = mGLIndices;
    }

    STOP_GLERROR;
}


// virtual (default)
void LLVertexBuffer::setupVertexBuffer()
{
    STOP_GLERROR;
    U8* base = nullptr;

    U32 data_mask = LLGLSLShader::sCurBoundShaderPtr->mAttributeMask;
    U32 setup_mask = gGLManager.mIsApple ?
        data_mask & ~sVertexAttribsConfigured : data_mask;

    if (setup_mask & MAP_NORMAL)
    {
        AttributeType loc = TYPE_NORMAL;
        void* ptr = (void*)(base + mOffsets[TYPE_NORMAL]);
        glVertexAttribPointer(loc, 3, GL_FLOAT, GL_FALSE, LLVertexBuffer::sTypeSize[TYPE_NORMAL], ptr);
    }
    if (setup_mask & MAP_TEXCOORD3)
    {
        AttributeType loc = TYPE_TEXCOORD3;
        void* ptr = (void*)(base + mOffsets[TYPE_TEXCOORD3]);
        glVertexAttribPointer(loc, 2, GL_FLOAT, GL_FALSE, LLVertexBuffer::sTypeSize[TYPE_TEXCOORD3], ptr);
    }
    if (setup_mask & MAP_TEXCOORD2)
    {
        AttributeType loc = TYPE_TEXCOORD2;
        void* ptr = (void*)(base + mOffsets[TYPE_TEXCOORD2]);
        glVertexAttribPointer(loc, 2, GL_FLOAT, GL_FALSE, LLVertexBuffer::sTypeSize[TYPE_TEXCOORD2], ptr);
    }
    if (setup_mask & MAP_TEXCOORD1)
    {
        AttributeType loc = TYPE_TEXCOORD1;
        void* ptr = (void*)(base + mOffsets[TYPE_TEXCOORD1]);
        glVertexAttribPointer(loc, 2, GL_FLOAT, GL_FALSE, LLVertexBuffer::sTypeSize[TYPE_TEXCOORD1], ptr);
    }
    if (setup_mask & MAP_TANGENT)
    {
        AttributeType loc = TYPE_TANGENT;
        void* ptr = (void*)(base + mOffsets[TYPE_TANGENT]);
        glVertexAttribPointer(loc, 4, GL_FLOAT, GL_FALSE, LLVertexBuffer::sTypeSize[TYPE_TANGENT], ptr);
    }
    if (setup_mask & MAP_TEXCOORD0)
    {
        AttributeType loc = TYPE_TEXCOORD0;
        void* ptr = (void*)(base + mOffsets[TYPE_TEXCOORD0]);
        glVertexAttribPointer(loc, 2, GL_FLOAT, GL_FALSE, LLVertexBuffer::sTypeSize[TYPE_TEXCOORD0], ptr);
    }
    const U32 desired_color_source = (data_mask & MAP_EMISSIVE) ?
        COLOR_POINTER_EMISSIVE :
        ((data_mask & MAP_COLOR) ? COLOR_POINTER_COLOR : 0);
    if (desired_color_source &&
        (!gGLManager.mIsApple || desired_color_source != sColorPointerSource))
    {
        AttributeType loc = TYPE_COLOR;
        void* ptr = desired_color_source == COLOR_POINTER_EMISSIVE ?
            (void*)(base + mOffsets[TYPE_EMISSIVE]) :
            (void*)(base + mOffsets[TYPE_COLOR]);
        glVertexAttribPointer(loc, 4, GL_UNSIGNED_BYTE, GL_TRUE, LLVertexBuffer::sTypeSize[TYPE_COLOR], ptr);
        sColorPointerSource = desired_color_source;
    }
    if (setup_mask & MAP_EMISSIVE)
    {
        AttributeType loc = TYPE_EMISSIVE;
        void* ptr = (void*)(base + mOffsets[TYPE_EMISSIVE]);
        glVertexAttribPointer(loc, 4, GL_UNSIGNED_BYTE, GL_TRUE, LLVertexBuffer::sTypeSize[TYPE_EMISSIVE], ptr);
    }
    if (setup_mask & MAP_WEIGHT)
    {
        AttributeType loc = TYPE_WEIGHT;
        void* ptr = (void*)(base + mOffsets[TYPE_WEIGHT]);
        glVertexAttribPointer(loc, 1, GL_FLOAT, GL_FALSE, LLVertexBuffer::sTypeSize[TYPE_WEIGHT], ptr);
    }
    if (setup_mask & MAP_WEIGHT4)
    {
        AttributeType loc = TYPE_WEIGHT4;
        void* ptr = (void*)(base + mOffsets[TYPE_WEIGHT4]);
        glVertexAttribPointer(loc, 4, GL_FLOAT, GL_FALSE, LLVertexBuffer::sTypeSize[TYPE_WEIGHT4], ptr);
    }
    if (setup_mask & MAP_JOINT)
    {
        AttributeType loc = TYPE_JOINT;
        void* ptr = (void*)(base + mOffsets[TYPE_JOINT]);
        glVertexAttribIPointer(loc, 4, GL_UNSIGNED_SHORT, LLVertexBuffer::sTypeSize[TYPE_JOINT], ptr);
    }
    if (setup_mask & MAP_CLOTHWEIGHT)
    {
        AttributeType loc = TYPE_CLOTHWEIGHT;
        void* ptr = (void*)(base + mOffsets[TYPE_CLOTHWEIGHT]);
        glVertexAttribPointer(loc, 4, GL_FLOAT, GL_TRUE, LLVertexBuffer::sTypeSize[TYPE_CLOTHWEIGHT], ptr);
    }
    if (setup_mask & MAP_TEXTURE_INDEX)
    {
        AttributeType loc = TYPE_TEXTURE_INDEX;
        void* ptr = (void*)(base + mOffsets[TYPE_VERTEX] + 12);
        glVertexAttribIPointer(loc, 1, GL_UNSIGNED_INT, LLVertexBuffer::sTypeSize[TYPE_VERTEX], ptr);
    }
    if (setup_mask & MAP_VERTEX)
    {
        AttributeType loc = TYPE_VERTEX;
        void* ptr = (void*)(base + mOffsets[TYPE_VERTEX]);
        glVertexAttribPointer(loc, 3, GL_FLOAT, GL_FALSE, LLVertexBuffer::sTypeSize[TYPE_VERTEX], ptr);
    }
    if (gGLManager.mIsApple)
    {
        sVertexAttribsConfigured |= data_mask;
        if (desired_color_source)
        {
            sVertexAttribsConfigured |= MAP_COLOR;
        }
    }
    STOP_GLERROR;
}

void LLVertexBuffer::setPositionData(const LLVector4a* data)
{
    flush_vbo(GL_ARRAY_BUFFER, 0, sizeof(LLVector4a) * getNumVerts()-1, (U8*) data, mMappedData);
}

void LLVertexBuffer::setTexCoord0Data(const LLVector2* data)
{
    flush_vbo(GL_ARRAY_BUFFER, mOffsets[TYPE_TEXCOORD0], mOffsets[TYPE_TEXCOORD0] + sTypeSize[TYPE_TEXCOORD0] * getNumVerts() - 1, (U8*)data, mMappedData);
}

void LLVertexBuffer::setTexCoord1Data(const LLVector2* data)
{
    flush_vbo(GL_ARRAY_BUFFER, mOffsets[TYPE_TEXCOORD1], mOffsets[TYPE_TEXCOORD1] + sTypeSize[TYPE_TEXCOORD1] * getNumVerts() - 1, (U8*)data, mMappedData);
}

void LLVertexBuffer::setColorData(const LLColor4U* data)
{
    flush_vbo(GL_ARRAY_BUFFER, mOffsets[TYPE_COLOR], mOffsets[TYPE_COLOR] + sTypeSize[TYPE_COLOR] * getNumVerts() - 1, (U8*) data, mMappedData);
}

void LLVertexBuffer::setNormalData(const LLVector4a* data)
{
    flush_vbo(GL_ARRAY_BUFFER, mOffsets[TYPE_NORMAL], mOffsets[TYPE_NORMAL] + sTypeSize[TYPE_NORMAL] * getNumVerts() - 1, (U8*) data, mMappedData);
}

void LLVertexBuffer::setTangentData(const LLVector4a* data)
{
    flush_vbo(GL_ARRAY_BUFFER, mOffsets[TYPE_TANGENT], mOffsets[TYPE_TANGENT] + sTypeSize[TYPE_TANGENT] * getNumVerts() - 1, (U8*) data, mMappedData);
}

void LLVertexBuffer::setWeight4Data(const LLVector4a* data)
{
    flush_vbo(GL_ARRAY_BUFFER, mOffsets[TYPE_WEIGHT4], mOffsets[TYPE_WEIGHT4] + sTypeSize[TYPE_WEIGHT4] * getNumVerts() - 1, (U8*) data, mMappedData);
}

void LLVertexBuffer::setJointData(const U64* data)
{
    flush_vbo(GL_ARRAY_BUFFER, mOffsets[TYPE_JOINT], mOffsets[TYPE_JOINT] + sTypeSize[TYPE_JOINT] * getNumVerts() - 1, (U8*) data, mMappedData);
}

void LLVertexBuffer::setIndexData(const U16* data)
{
    flush_vbo(GL_ELEMENT_ARRAY_BUFFER, 0, sizeof(U16) * getNumIndices() - 1, (U8*) data, mMappedIndexData);
}

void LLVertexBuffer::setIndexData(const U32* data)
{
    if (mIndicesType != GL_UNSIGNED_INT)
    { // HACK -- vertex buffers are initialized as 16-bit indices, but can be switched to 32-bit indices
        mIndicesType = GL_UNSIGNED_INT;
        mIndicesStride = 4;
        mNumIndices /= 2;
    }
    flush_vbo(GL_ELEMENT_ARRAY_BUFFER, 0, sizeof(U32) * getNumIndices() - 1, (U8*)data, mMappedIndexData);
}

void LLVertexBuffer::setPositionData(const LLVector4a* data, U32 offset, U32 count)
{
    flush_vbo(GL_ARRAY_BUFFER, offset * sizeof(LLVector4a), (offset + count) * sizeof(LLVector4a) - 1, (U8*)data, mMappedData);
}

void LLVertexBuffer::setNormalData(const LLVector4a* data, U32 offset, U32 count)
{
    flush_vbo(GL_ARRAY_BUFFER, mOffsets[TYPE_NORMAL] + offset * sTypeSize[TYPE_NORMAL], mOffsets[TYPE_NORMAL] + (offset + count) * sTypeSize[TYPE_NORMAL] - 1, (U8*)data, mMappedData);
}

void LLVertexBuffer::setTexCoord0Data(const LLVector2* data, U32 offset, U32 count)
{
    flush_vbo(GL_ARRAY_BUFFER, mOffsets[TYPE_TEXCOORD0] + offset * sTypeSize[TYPE_TEXCOORD0], mOffsets[TYPE_TEXCOORD0] + (offset + count) * sTypeSize[TYPE_TEXCOORD0] - 1, (U8*)data, mMappedData);
}

void LLVertexBuffer::setTexCoord1Data(const LLVector2* data, U32 offset, U32 count)
{
    flush_vbo(GL_ARRAY_BUFFER, mOffsets[TYPE_TEXCOORD1] + offset * sTypeSize[TYPE_TEXCOORD1], mOffsets[TYPE_TEXCOORD1] + (offset + count) * sTypeSize[TYPE_TEXCOORD1] - 1, (U8*)data, mMappedData);
}

void LLVertexBuffer::setColorData(const LLColor4U* data, U32 offset, U32 count)
{
    flush_vbo(GL_ARRAY_BUFFER, mOffsets[TYPE_COLOR] + offset * sTypeSize[TYPE_COLOR], mOffsets[TYPE_COLOR] + (offset + count) * sTypeSize[TYPE_COLOR] - 1, (U8*)data, mMappedData);
}

void LLVertexBuffer::setTangentData(const LLVector4a* data, U32 offset, U32 count)
{
    flush_vbo(GL_ARRAY_BUFFER, mOffsets[TYPE_TANGENT] + offset * sTypeSize[TYPE_TANGENT], mOffsets[TYPE_TANGENT] + (offset + count) * sTypeSize[TYPE_TANGENT] - 1, (U8*)data, mMappedData);
}

void LLVertexBuffer::setWeight4Data(const LLVector4a* data, U32 offset, U32 count)
{
    flush_vbo(GL_ARRAY_BUFFER, mOffsets[TYPE_WEIGHT4] + offset * sTypeSize[TYPE_WEIGHT4], mOffsets[TYPE_WEIGHT4] + (offset + count) * sTypeSize[TYPE_WEIGHT4] - 1, (U8*)data, mMappedData);
}

void LLVertexBuffer::setJointData(const U64* data, U32 offset, U32 count)
{
    flush_vbo(GL_ARRAY_BUFFER, mOffsets[TYPE_JOINT] + offset * sTypeSize[TYPE_JOINT], mOffsets[TYPE_JOINT] + (offset + count) * sTypeSize[TYPE_JOINT] - 1, (U8*)data, mMappedData);
}

void LLVertexBuffer::setIndexData(const U16* data, U32 offset, U32 count)
{
    flush_vbo(GL_ELEMENT_ARRAY_BUFFER, offset * sizeof(U16), (offset + count) * sizeof(U16) - 1, (U8*)data, mMappedIndexData);
}

void LLVertexBuffer::setIndexData(const U32* data, U32 offset, U32 count)
{
    if (mIndicesType != GL_UNSIGNED_INT)
    { // HACK -- vertex buffers are initialized as 16-bit indices, but can be switched to 32-bit indices
        mIndicesType = GL_UNSIGNED_INT;
        mIndicesStride = 4;
        mNumIndices /= 2;
    }
    flush_vbo(GL_ELEMENT_ARRAY_BUFFER, offset * sizeof(U32), (offset + count) * sizeof(U32) - 1, (U8*)data, mMappedIndexData);
}