Metal: Resolve runtime issues with texture views

When creating a texture view, Metal may require that the original
GPUTexture state is modified in some way. This may be a result
of deferred creation, or, to cache the texture view against the
source.

As a result, GPUTexture passed into GPU_texture_create_view
cannot be const.

Small fixes have also been made in the Metal texture
implementation to ensure correct function of texture views.

Authored by Apple: Michael Parkin-White

Pull Request: https://projects.blender.org/blender/blender/pulls/107167

source/blender/gpu/GPU_texture.h

@@ -703,7 +703,7 @@ void GPU_texture_free(GPUTexture *texture);
  * TODO(fclem): Target conversion (ex: Texture 2D as Texture 2D Array) is not implemented yet.
  */
 GPUTexture *GPU_texture_create_view(const char *name,
-                                    const GPUTexture *source_texture,
+                                    GPUTexture *source_texture,
                                     eGPUTextureFormat view_format,
                                     int mip_start,
                                     int mip_len,

source/blender/gpu/intern/gpu_texture.cc

@@ -132,7 +132,7 @@ bool Texture::init_buffer(GPUVertBuf *vbo, eGPUTextureFormat format)
   return this->init_internal(vbo);
 }
 
-bool Texture::init_view(const GPUTexture *src_,
+bool Texture::init_view(GPUTexture *src_,
                         eGPUTextureFormat format,
                         eGPUTextureType type,
                         int mip_start,
@@ -448,7 +448,7 @@ GPUTexture *GPU_texture_create_error(int dimension, bool is_array)
 }
 
 GPUTexture *GPU_texture_create_view(const char *name,
-                                    const GPUTexture *src,
+                                    GPUTexture *src,
                                     eGPUTextureFormat format,
                                     int mip_start,
                                     int mip_len,

source/blender/gpu/intern/gpu_texture_private.hh

@@ -132,7 +132,7 @@ class Texture {
   bool init_3D(int w, int h, int d, int mip_len, eGPUTextureFormat format);
   bool init_cubemap(int w, int layers, int mip_len, eGPUTextureFormat format);
   bool init_buffer(GPUVertBuf *vbo, eGPUTextureFormat format);
-  bool init_view(const GPUTexture *src,
+  bool init_view(GPUTexture *src,
                  eGPUTextureFormat format,
                  eGPUTextureType type,
                  int mip_start,
@@ -313,7 +313,7 @@ class Texture {
  protected:
   virtual bool init_internal() = 0;
   virtual bool init_internal(GPUVertBuf *vbo) = 0;
-  virtual bool init_internal(const GPUTexture *src, int mip_offset, int layer_offset) = 0;
+  virtual bool init_internal(GPUTexture *src, int mip_offset, int layer_offset) = 0;
 };
 
 /* Syntactic sugar. */

source/blender/gpu/metal/mtl_texture.hh

@@ -181,7 +181,6 @@ class MTLTexture : public Texture {
   bool is_baked_ = false;
   MTLTextureDescriptor *texture_descriptor_ = nullptr;
   id<MTLTexture> texture_ = nil;
-  MTLTextureUsage usage_;
 
   /* Texture Storage. */
   id<MTLBuffer> texture_buffer_ = nil;
@@ -284,7 +283,7 @@ class MTLTexture : public Texture {
  protected:
   bool init_internal() override;
   bool init_internal(GPUVertBuf *vbo) override;
-  bool init_internal(const GPUTexture *src,
+  bool init_internal(GPUTexture *src,
                      int mip_offset,
                      int layer_offset) override; /* Texture View */
 

source/blender/gpu/metal/mtl_texture.mm

@@ -159,7 +159,7 @@ void gpu::MTLTexture::bake_mip_swizzle_view()
     }
 
     int range_len = min_ii((mip_texture_max_level_ - mip_texture_base_level_) + 1,
-                           texture_.mipmapLevelCount);
+                           texture_.mipmapLevelCount - mip_texture_base_level_);
     BLI_assert(range_len > 0);
     BLI_assert(mip_texture_base_level_ < texture_.mipmapLevelCount);
     BLI_assert(mip_texture_base_layer_ < num_slices);
@@ -1152,7 +1152,7 @@ void gpu::MTLTexture::ensure_mipmaps(int miplvl)
 void gpu::MTLTexture::generate_mipmap()
 {
   /* Fetch Active Context. */
-  MTLContext *ctx = reinterpret_cast<MTLContext *>(GPU_context_active_get());
+  MTLContext *ctx = static_cast<MTLContext *>(unwrap(GPU_context_active_get()));
   BLI_assert(ctx);
 
   if (!ctx->device) {
@@ -1265,7 +1265,7 @@ void gpu::MTLTexture::clear(eGPUDataFormat data_format, const void *data)
 
   /* Create clear framebuffer. */
   GPUFrameBuffer *prev_fb = GPU_framebuffer_active_get();
-  FrameBuffer *fb = reinterpret_cast<FrameBuffer *>(this->get_blit_framebuffer(0, 0));
+  FrameBuffer *fb = unwrap(this->get_blit_framebuffer(0, 0));
   fb->bind(true);
   fb->clear_attachment(this->attachment_type(0), data_format, data);
   GPU_framebuffer_bind(prev_fb);
@@ -1805,7 +1805,7 @@ bool gpu::MTLTexture::init_internal(GPUVertBuf *vbo)
   return true;
 }
 
-bool gpu::MTLTexture::init_internal(const GPUTexture *src, int mip_offset, int layer_offset)
+bool gpu::MTLTexture::init_internal(GPUTexture *src, int mip_offset, int layer_offset)
 {
   BLI_assert(src);
 
@@ -1817,12 +1817,17 @@ bool gpu::MTLTexture::init_internal(const GPUTexture *src, int mip_offset, int l
   source_texture_ = src;
   mip_texture_base_level_ = mip_offset;
   mip_texture_base_layer_ = layer_offset;
+  texture_view_dirty_flags_ |= TEXTURE_VIEW_MIP_DIRTY;
 
   /* Assign usage. */
   gpu_image_usage_flags_ = GPU_texture_usage(src);
+  BLI_assert_msg(
+      gpu_image_usage_flags_ & GPU_TEXTURE_USAGE_MIP_SWIZZLE_VIEW,
+      "Source texture of TextureView must have GPU_TEXTURE_USAGE_MIP_SWIZZLE_VIEW usage flag.");
 
   /* Assign texture as view. */
-  const gpu::MTLTexture *mtltex = static_cast<const gpu::MTLTexture *>(unwrap(src));
+  gpu::MTLTexture *mtltex = static_cast<gpu::MTLTexture *>(unwrap(src));
+  mtltex->ensure_baked();
   texture_ = mtltex->texture_;
   BLI_assert(texture_);
   [texture_ retain];

source/blender/gpu/opengl/gl_texture.cc

@@ -178,7 +178,7 @@ bool GLTexture::init_internal(GPUVertBuf *vbo)
   return true;
 }
 
-bool GLTexture::init_internal(const GPUTexture *src, int mip_offset, int layer_offset)
+bool GLTexture::init_internal(GPUTexture *src, int mip_offset, int layer_offset)
 {
   BLI_assert(GLContext::texture_storage_support);
 

source/blender/gpu/opengl/gl_texture.hh

@@ -117,7 +117,7 @@ class GLTexture : public Texture {
   /** Return true on success. */
   bool init_internal(GPUVertBuf *vbo) override;
   /** Return true on success. */
-  bool init_internal(const GPUTexture *src, int mip_offset, int layer_offset) override;
+  bool init_internal(GPUTexture *src, int mip_offset, int layer_offset) override;
 
  private:
   bool proxy_check(int mip);

source/blender/gpu/vulkan/vk_texture.cc

@@ -174,7 +174,7 @@ bool VKTexture::init_internal(GPUVertBuf * /*vbo*/)
   return false;
 }
 
-bool VKTexture::init_internal(const GPUTexture * /*src*/, int /*mip_offset*/, int /*layer_offset*/)
+bool VKTexture::init_internal(GPUTexture * /*src*/, int /*mip_offset*/, int /*layer_offset*/)
 {
   return false;
 }

source/blender/gpu/vulkan/vk_texture.hh

@@ -61,7 +61,7 @@ class VKTexture : public Texture {
  protected:
   bool init_internal() override;
   bool init_internal(GPUVertBuf *vbo) override;
-  bool init_internal(const GPUTexture *src, int mip_offset, int layer_offset) override;
+  bool init_internal(GPUTexture *src, int mip_offset, int layer_offset) override;
 
  private:
   /** Is this texture already allocated on device. */