Files
Xenia-Canary/src/xenia/gpu/vulkan/render_cache.h
Dr. Chat d18c99aab6 RenderCache: Account for MSAA when calculating tile sizes.
Add a new flag to enable native MSAA (this does not work properly at the moment)
2016-05-03 14:05:34 -05:00

388 lines
16 KiB
C++

/**
******************************************************************************
* Xenia : Xbox 360 Emulator Research Project *
******************************************************************************
* Copyright 2016 Ben Vanik. All rights reserved. *
* Released under the BSD license - see LICENSE in the root for more details. *
******************************************************************************
*/
#ifndef XENIA_GPU_VULKAN_RENDER_CACHE_H_
#define XENIA_GPU_VULKAN_RENDER_CACHE_H_
#include "xenia/gpu/register_file.h"
#include "xenia/gpu/shader.h"
#include "xenia/gpu/texture_info.h"
#include "xenia/gpu/vulkan/vulkan_shader.h"
#include "xenia/gpu/xenos.h"
#include "xenia/ui/vulkan/vulkan.h"
#include "xenia/ui/vulkan/vulkan_device.h"
namespace xe {
namespace gpu {
namespace vulkan {
// TODO(benvanik): make public API?
class CachedTileView;
class CachedFramebuffer;
class CachedRenderPass;
// Uniquely identifies EDRAM tiles.
struct TileViewKey {
// Offset into EDRAM in 5120b tiles.
uint16_t tile_offset;
// Tile width of the view in base 80x16 tiles.
uint16_t tile_width;
// Tile height of the view in base 80x16 tiles.
uint16_t tile_height;
// 1 if format is ColorRenderTargetFormat, else DepthRenderTargetFormat.
uint16_t color_or_depth : 1;
// Surface MSAA samples
uint16_t msaa_samples : 2;
// Either ColorRenderTargetFormat or DepthRenderTargetFormat.
uint16_t edram_format : 13;
};
static_assert(sizeof(TileViewKey) == 8, "Key must be tightly packed");
// Cached view representing EDRAM memory.
// TODO(benvanik): reuse VkImage's with multiple VkViews for compatible
// formats?
class CachedTileView {
public:
// Key identifying the view in the cache.
TileViewKey key;
// Image
VkImage image = nullptr;
// Simple view on the image matching the format.
VkImageView image_view = nullptr;
// Memory buffer
VkDeviceMemory memory = nullptr;
// Image sample count
VkSampleCountFlagBits sample_count = VK_SAMPLE_COUNT_1_BIT;
CachedTileView(ui::vulkan::VulkanDevice* device,
VkCommandBuffer command_buffer, VkDeviceMemory edram_memory,
TileViewKey view_key);
~CachedTileView();
bool IsEqual(const TileViewKey& other_key) const {
auto a = reinterpret_cast<const uint64_t*>(&key);
auto b = reinterpret_cast<const uint64_t*>(&other_key);
return *a == *b;
}
bool operator<(const CachedTileView& other) const {
return key.tile_offset < other.key.tile_offset;
}
private:
VkDevice device_ = nullptr;
};
// Parsed render configuration from the current render state.
struct RenderConfiguration {
// Render mode (color+depth, depth-only, etc).
xenos::ModeControl mode_control;
// Target surface pitch multiplied by MSAA, in pixels.
uint32_t surface_pitch_px;
// ESTIMATED target surface height multiplied by MSAA, in pixels.
uint32_t surface_height_px;
// Surface MSAA setting.
MsaaSamples surface_msaa;
// Color attachments for the 4 render targets.
struct {
bool used;
uint32_t edram_base;
ColorRenderTargetFormat format;
} color[4];
// Depth/stencil attachment.
struct {
bool used;
uint32_t edram_base;
DepthRenderTargetFormat format;
} depth_stencil;
};
// Current render state based on the register-specified configuration.
struct RenderState {
// Parsed configuration.
RenderConfiguration config;
// Render pass (to be used with pipelines/etc).
CachedRenderPass* render_pass = nullptr;
VkRenderPass render_pass_handle = nullptr;
// Target framebuffer bound to the render pass.
CachedFramebuffer* framebuffer = nullptr;
VkFramebuffer framebuffer_handle = nullptr;
bool color_attachment_written[4] = {false};
bool depth_attachment_written = false;
};
// Manages the virtualized EDRAM and the render target cache.
//
// On the 360 the render target is an opaque block of memory in EDRAM that's
// only accessible via resolves. We use this to our advantage to simulate
// something like it as best we can by having a shared backing memory with
// a multitude of views for each tile location in EDRAM.
//
// This allows us to have the same base address write to the same memory
// regardless of framebuffer format. Resolving then uses whatever format the
// resolve requests straight from the backing memory.
//
// EDRAM is a beast and we only approximate it as best we can. Basically,
// the 10MiB of EDRAM is composed of 2048 5120b tiles. Each tile is 80x16px.
// +-----+-----+-----+---
// |tile0|tile1|tile2|... 2048 times
// +-----+-----+-----+---
// Operations dealing with EDRAM deal in tile offsets, so base 0x100 is tile
// offset 256, 256*5120=1310720b into the buffer. All rendering operations are
// aligned to tiles so trying to draw at 256px wide will have a real width of
// 320px by rounding up to the next tile.
//
// MSAA and other settings will modify the exact pixel sizes, like 4X makes
// each tile effectively 40x8px / 2X makes each tile 80x8px, but they are still
// all 5120b. As we try to emulate this we adjust our viewport when rendering to
// stretch pixels as needed.
//
// It appears that games also take advantage of MSAA stretching tiles when doing
// clears. Games will clear a view with 1/2X pitch/height and 4X MSAA and then
// later draw to that view with 1X pitch/height and 1X MSAA.
//
// The good news is that games cannot read EDRAM directly but must use a copy
// operation to get the data out. That gives us a chance to do whatever we
// need to (re-tile, etc) only when requested.
//
// To approximate the tiled EDRAM layout we use a single large chunk of memory.
// From this memory we create many VkImages (and VkImageViews) of various
// formats and dimensions as requested by the game. These are used as
// attachments during rendering and as sources during copies. They are also
// heavily aliased - lots of images will reference the same locations in the
// underlying EDRAM buffer. The only requirement is that there are no hazards
// with specific tiles (reading/writing the same tile through different images)
// and otherwise it should be ok *fingers crossed*.
//
// One complication is the copy/resolve process itself: we need to give back
// the data asked for in the format desired and where it goes is arbitrary
// (any address in physical memory). If the game is good we get resolves of
// EDRAM into fixed base addresses with scissored regions. If the game is bad
// we are broken.
//
// Resolves from EDRAM result in tiled textures - that's texture tiles, not
// EDRAM tiles. If we wanted to ensure byte-for-byte correctness we'd need to
// then tile the images as we wrote them out. For now, we just attempt to
// get the (X, Y) in linear space and do that. This really comes into play
// when multiple resolves write to the same texture or memory aliased by
// multiple textures - which is common due to predicated tiling. The examples
// below demonstrate what this looks like, but the important thing is that
// we are aware of partial textures and overlapping regions.
//
// TODO(benvanik): what, if any, barriers do we need? any transitions?
//
// Example with multiple render targets:
// Two color targets of 256x256px tightly packed in EDRAM:
// color target 0: base 0x0, pitch 320, scissor 0,0, 256x256
// starts at tile 0, buffer offset 0
// contains 64 tiles (320/80)*(256/16)
// color target 1: base 0x40, pitch 320, scissor 256,0, 256x256
// starts at tile 64 (after color target 0), buffer offset 327680b
// contains 64 tiles
// In EDRAM each set of 64 tiles is contiguous:
// +------+------+ +------+------+------+
// |ct0.0 |ct0.1 |...|ct0.63|ct1.0 |ct1.1 |...
// +------+------+ +------+------+------+
// To render into these, we setup two VkImages:
// image 0: bound to buffer offset 0, 320x256x4=327680b
// image 1: bound to buffer offset 327680b, 320x256x4=327680b
// So when we render to them:
// +------+-+ scissored to 256x256, actually 320x256
// | . | | <- . appears at some untiled offset in the buffer, but
// | | | consistent if aliased with the same format
// +------+-+
// In theory, this gives us proper aliasing in most cases.
//
// Example with horizontal predicated tiling:
// Trying to render 1024x576 @4X MSAA, splitting into two regions
// horizontally:
// +----------+
// | 1024x288 |
// +----------+
// | 1024x288 |
// +----------+
// EDRAM configured for 1056x288px with tile size 2112x567px (4X MSAA):
// color target 0: base 0x0, pitch 1080, 26x36 tiles
// First render (top):
// window offset 0,0
// scissor 0,0, 1024x288
// First resolve (top):
// RB_COPY_DEST_BASE 0x1F45D000
// RB_COPY_DEST_PITCH pitch=1024, height=576
// vertices: 0,0, 1024,0, 1024,288
// Second render (bottom):
// window offset 0,-288
// scissor 0,288, 1024x288
// Second resolve (bottom):
// RB_COPY_DEST_BASE 0x1F57D000 (+1179648b)
// RB_COPY_DEST_PITCH pitch=1024, height=576
// (exactly 1024x288*4b after first resolve)
// vertices: 0,288, 1024,288, 1024,576
// Resolving here is easy as the textures are contiguous in memory. We can
// snoop in the first resolve with the dest height to know the total size,
// and in the second resolve see that it overlaps and place it in the
// existing target.
//
// Example with vertical predicated tiling:
// Trying to render 1280x720 @2X MSAA, splitting into two regions
// vertically:
// +-----+-----+
// | 640 | 640 |
// | x | x |
// | 720 | 720 |
// +-----+-----+
// EDRAM configured for 640x736px with tile size 640x1472px (2X MSAA):
// color target 0: base 0x0, pitch 640, 8x92 tiles
// First render (left):
// window offset 0,0
// scissor 0,0, 640x720
// First resolve (left):
// RB_COPY_DEST_BASE 0x1BC6D000
// RB_COPY_DEST_PITCH pitch=1280, height=720
// vertices: 0,0, 640,0, 640,720
// Second render (right):
// window offset -640,0
// scissor 640,0, 640x720
// Second resolve (right):
// RB_COPY_DEST_BASE 0x1BC81000 (+81920b)
// RB_COPY_DEST_PITCH pitch=1280, height=720
// vertices: 640,0, 1280,0, 1280,720
// Resolving here is much more difficult as resolves are tiled and the right
// half of the texture is 81920b away:
// 81920/4bpp=20480px, /32 (texture tile size)=640px
// We know the texture size with the first resolve and with the second we
// must check for overlap then compute the offset (in both X and Y).
class RenderCache {
public:
RenderCache(RegisterFile* register_file, ui::vulkan::VulkanDevice* device);
~RenderCache();
// Call this to determine if you should start a new render pass or continue
// with an already open pass.
bool dirty() const;
// Begins a render pass targeting the state-specified framebuffer formats.
// The command buffer will be transitioned into the render pass phase.
const RenderState* BeginRenderPass(VkCommandBuffer command_buffer,
VulkanShader* vertex_shader,
VulkanShader* pixel_shader);
// Ends the current render pass.
// The command buffer will be transitioned out of the render pass phase.
void EndRenderPass();
// Updates current render state. Call this every draw with an open render pass
void UpdateState();
// Clears all cached content.
void ClearCache();
// Queues commands to copy EDRAM contents into an image.
// The command buffer must not be inside of a render pass when calling this.
void RawCopyToImage(VkCommandBuffer command_buffer, uint32_t edram_base,
VkImage image, VkImageLayout image_layout,
bool color_or_depth, VkOffset3D offset,
VkExtent3D extents);
// Queues commands to blit EDRAM contents into an image.
// The command buffer must not be inside of a render pass when calling this.
void BlitToImage(VkCommandBuffer command_buffer, uint32_t edram_base,
uint32_t pitch, uint32_t height, MsaaSamples num_samples,
VkImage image, VkImageLayout image_layout,
bool color_or_depth, uint32_t format, VkFilter filter,
VkOffset3D offset, VkExtent3D extents);
// Queues commands to clear EDRAM contents with a solid color.
// The command buffer must not be inside of a render pass when calling this.
void ClearEDRAMColor(VkCommandBuffer command_buffer, uint32_t edram_base,
ColorRenderTargetFormat format, uint32_t pitch,
uint32_t height, MsaaSamples num_samples, float* color);
// Queues commands to clear EDRAM contents with depth/stencil values.
// The command buffer must not be inside of a render pass when calling this.
void ClearEDRAMDepthStencil(VkCommandBuffer command_buffer,
uint32_t edram_base,
DepthRenderTargetFormat format, uint32_t pitch,
uint32_t height, MsaaSamples num_samples,
float depth, uint32_t stencil);
// Queues commands to fill EDRAM contents with a constant value.
// The command buffer must not be inside of a render pass when calling this.
void FillEDRAM(VkCommandBuffer command_buffer, uint32_t value);
private:
// Parses the current state into a configuration object.
bool ParseConfiguration(RenderConfiguration* config);
// Finds a tile view. Returns nullptr if none found matching the key.
CachedTileView* FindTileView(const TileViewKey& view_key) const;
// Gets or creates a tile view with the given parameters.
CachedTileView* FindOrCreateTileView(VkCommandBuffer command_buffer,
const TileViewKey& view_key);
void UpdateTileView(VkCommandBuffer command_buffer, CachedTileView* view,
bool load, bool insert_barrier = true);
// Gets or creates a render pass and frame buffer for the given configuration.
// This attempts to reuse as much as possible across render passes and
// framebuffers.
bool ConfigureRenderPass(VkCommandBuffer command_buffer,
RenderConfiguration* config,
CachedRenderPass** out_render_pass,
CachedFramebuffer** out_framebuffer);
RegisterFile* register_file_ = nullptr;
ui::vulkan::VulkanDevice* device_ = nullptr;
// Entire 10MiB of EDRAM.
VkDeviceMemory edram_memory_ = nullptr;
// Buffer overlayed 1:1 with edram_memory_ to allow raw access.
VkBuffer edram_buffer_ = nullptr;
// Cache of VkImage and VkImageView's for all of our EDRAM tilings.
// TODO(benvanik): non-linear lookup? Should only be a small number of these.
std::vector<CachedTileView*> cached_tile_views_;
// Cache of render passes based on formats.
std::vector<CachedRenderPass*> cached_render_passes_;
// Shadows of the registers that impact the render pass we choose.
// If the registers don't change between passes we can quickly reuse the
// previous one.
struct ShadowRegisters {
uint32_t rb_modecontrol;
uint32_t rb_surface_info;
uint32_t rb_color_info;
uint32_t rb_color1_info;
uint32_t rb_color2_info;
uint32_t rb_color3_info;
uint32_t rb_depth_info;
uint32_t rb_depthcontrol;
uint32_t pa_sc_window_scissor_tl;
uint32_t pa_sc_window_scissor_br;
ShadowRegisters() { Reset(); }
void Reset() { std::memset(this, 0, sizeof(*this)); }
} shadow_registers_;
bool SetShadowRegister(uint32_t* dest, uint32_t register_name);
// Configuration used for the current/previous Begin/End, representing the
// current shadow register state.
RenderState current_state_;
// Only valid during a BeginRenderPass/EndRenderPass block.
VkCommandBuffer current_command_buffer_ = nullptr;
};
} // namespace vulkan
} // namespace gpu
} // namespace xe
#endif // XENIA_GPU_VULKAN_RENDER_CACHE_H_