We are currently migrating Bugzilla to GitHub issues.
Any changes made to the bug tracker now will be lost, so please do not post new bugs or make changes to them.
When we're done, all bug URLs will redirect to their equivalent location on the new bug tracker.

Example from current 2.0.1 source code version SDL_render_gles2.c:

static int
GLES2_RenderFillRects(SDL_Renderer *renderer, const SDL_FRect *rects, int count)
{
    GLES2_DriverContext *data = (GLES2_DriverContext *)renderer->driverdata;
    GLfloat vertices[8];
    int idx;

    if (GLES2_SetDrawingState(renderer) < 0) {
        return -1;
    }

    /* Emit a line loop for each rectangle */
    for (idx = 0; idx < count; ++idx) {
        const SDL_FRect *rect = &rects[idx];

        GLfloat xMin = rect->x;
        GLfloat xMax = (rect->x + rect->w);
        GLfloat yMin = rect->y;
        GLfloat yMax = (rect->y + rect->h);

        vertices[0] = xMin;
        vertices[1] = yMin;
        vertices[2] = xMax;
        vertices[3] = yMin;
        vertices[4] = xMin;
        vertices[5] = yMax;
        vertices[6] = xMax;
        vertices[7] = yMax;
        data->glVertexAttribPointer(GLES2_ATTRIBUTE_POSITION, 2, GL_FLOAT, GL_FALSE, 0, vertices);
        data->glDrawArrays(GL_TRIANGLE_STRIP, 0, 4);
    }
    return GL_CheckError("", renderer);
}


-------

If I do call glDisableVertexAttribArray(0) before this code is run, glVertexAttribPointer will fail. All OpenGL vertex attrib calls should properly encapsulated in glEnable/glDisable before using attrib data.

Does not fail on all system as some drivers behave different on disabling attrib 0, but encapsulation will solve this issue always.