dolphin/Source/Core/VideoCommon/BPFunctions.cpp
JosJuice 6e88c44d5d Move SmallVector to Common
We had one implementation of this type of data structure in Arm64Emitter
and one in VideoCommon. This moves the Arm64Emitter implementation to
its own file and adds begin and end functions to it, so that VideoCommon
can use it.

You may notice that the license header for the new file is CC0. I wrote
the Arm64Emitter implementation of SmallVector, so this should be no
problem.
2023-08-22 13:19:49 +02:00

446 lines
15 KiB
C++

// Copyright 2009 Dolphin Emulator Project
// SPDX-License-Identifier: GPL-2.0-or-later
#include "VideoCommon/BPFunctions.h"
#include <algorithm>
#include <cmath>
#include <string_view>
#include "Common/Assert.h"
#include "Common/CommonTypes.h"
#include "Common/Logging/Log.h"
#include "Common/SmallVector.h"
#include "VideoCommon/AbstractFramebuffer.h"
#include "VideoCommon/AbstractGfx.h"
#include "VideoCommon/BPMemory.h"
#include "VideoCommon/FramebufferManager.h"
#include "VideoCommon/RenderBase.h"
#include "VideoCommon/RenderState.h"
#include "VideoCommon/VertexManagerBase.h"
#include "VideoCommon/VertexShaderManager.h"
#include "VideoCommon/VideoCommon.h"
#include "VideoCommon/VideoConfig.h"
#include "VideoCommon/XFMemory.h"
namespace BPFunctions
{
// ----------------------------------------------
// State translation lookup tables
// Reference: Yet Another GameCube Documentation
// ----------------------------------------------
void FlushPipeline()
{
g_vertex_manager->Flush();
}
void SetGenerationMode()
{
g_vertex_manager->SetRasterizationStateChanged();
}
int ScissorRect::GetArea() const
{
return rect.GetWidth() * rect.GetHeight();
}
int ScissorResult::GetViewportArea(const ScissorRect& rect) const
{
int x0 = std::clamp<int>(rect.rect.left + rect.x_off, viewport_left, viewport_right);
int x1 = std::clamp<int>(rect.rect.right + rect.x_off, viewport_left, viewport_right);
int y0 = std::clamp<int>(rect.rect.top + rect.y_off, viewport_top, viewport_bottom);
int y1 = std::clamp<int>(rect.rect.bottom + rect.y_off, viewport_top, viewport_bottom);
return (x1 - x0) * (y1 - y0);
}
// Compare so that a sorted collection of rectangles has the best one last, so that if they're drawn
// in order, the best one is the one that is drawn last (and thus over the rest).
// The exact iteration order on hardware hasn't been tested, but silly things can happen where a
// polygon can intersect with itself; this only applies outside of the viewport region (in areas
// that would normally be affected by clipping). No game is known to care about this.
bool ScissorResult::IsWorse(const ScissorRect& lhs, const ScissorRect& rhs) const
{
// First, penalize any rect that is not in the viewport
int lhs_area = GetViewportArea(lhs);
int rhs_area = GetViewportArea(rhs);
if (lhs_area != rhs_area)
return lhs_area < rhs_area;
// Now compare on total areas, without regard for the viewport
return lhs.GetArea() < rhs.GetArea();
}
namespace
{
using RangeList = Common::SmallVector<ScissorRange, 9>;
static RangeList ComputeScissorRanges(int start, int end, int offset, int efb_dim)
{
RangeList ranges;
for (int extra_off = -4096; extra_off <= 4096; extra_off += 1024)
{
int new_off = offset + extra_off;
int new_start = std::clamp(start - new_off, 0, efb_dim);
int new_end = std::clamp(end - new_off + 1, 0, efb_dim);
if (new_start < new_end)
{
ranges.emplace_back(new_off, new_start, new_end);
}
}
return ranges;
}
} // namespace
ScissorResult::ScissorResult(const BPMemory& bpmemory, const XFMemory& xfmemory)
: ScissorResult(bpmemory,
std::minmax(xfmemory.viewport.xOrig - xfmemory.viewport.wd,
xfmemory.viewport.xOrig + xfmemory.viewport.wd),
std::minmax(xfmemory.viewport.yOrig - xfmemory.viewport.ht,
xfmemory.viewport.yOrig + xfmemory.viewport.ht))
{
}
ScissorResult::ScissorResult(const BPMemory& bpmemory, std::pair<float, float> viewport_x,
std::pair<float, float> viewport_y)
: scissor_tl{.hex = bpmemory.scissorTL.hex}, scissor_br{.hex = bpmemory.scissorBR.hex},
scissor_off{.hex = bpmemory.scissorOffset.hex}, viewport_left(viewport_x.first),
viewport_right(viewport_x.second), viewport_top(viewport_y.first),
viewport_bottom(viewport_y.second)
{
// Range is [left, right] and [top, bottom] (closed intervals)
const int left = scissor_tl.x;
const int right = scissor_br.x;
const int top = scissor_tl.y;
const int bottom = scissor_br.y;
// When left > right or top > bottom, nothing renders (even with wrapping from the offsets)
if (left > right || top > bottom)
return;
// Note that both the offsets and the coordinates have 342 added to them internally by GX
// functions (for the offsets, this is before they are divided by 2/right shifted). This code
// could undo both sets of offsets, but it doesn't need to since they cancel out when subtracting
// (and those offsets actually matter for the left > right and top > bottom checks).
const int x_off = scissor_off.x << 1;
const int y_off = scissor_off.y << 1;
RangeList x_ranges = ComputeScissorRanges(left, right, x_off, EFB_WIDTH);
RangeList y_ranges = ComputeScissorRanges(top, bottom, y_off, EFB_HEIGHT);
m_result.reserve(x_ranges.size() * y_ranges.size());
// Now we need to form actual rectangles from the x and y ranges,
// which is a simple Cartesian product of x_ranges_clamped and y_ranges_clamped.
// Each rectangle is also a Cartesian product of x_range and y_range, with
// the rectangles being half-open (of the form [x0, x1) X [y0, y1)).
for (const auto& x_range : x_ranges)
{
DEBUG_ASSERT(x_range.start < x_range.end);
DEBUG_ASSERT(static_cast<u32>(x_range.end) <= EFB_WIDTH);
for (const auto& y_range : y_ranges)
{
DEBUG_ASSERT(y_range.start < y_range.end);
DEBUG_ASSERT(static_cast<u32>(y_range.end) <= EFB_HEIGHT);
m_result.emplace_back(x_range, y_range);
}
}
auto cmp = [&](const ScissorRect& lhs, const ScissorRect& rhs) { return IsWorse(lhs, rhs); };
std::sort(m_result.begin(), m_result.end(), cmp);
}
ScissorRect ScissorResult::Best() const
{
// For now, simply choose the best rectangle (see ScissorResult::IsWorse).
// This does mean we calculate all rectangles and only choose one, which is not optimal, but this
// is called infrequently. Eventually, all backends will support multiple scissor rects.
if (!m_result.empty())
{
return m_result.back();
}
else
{
// But if we have no rectangles, use a bogus one that's out of bounds.
// Ideally, all backends will support multiple scissor rects, in which case this won't be
// needed.
return ScissorRect(ScissorRange{0, 1000, 1001}, ScissorRange{0, 1000, 1001});
}
}
ScissorResult ComputeScissorRects()
{
return ScissorResult{bpmem, xfmem};
}
void SetScissorAndViewport()
{
auto native_rc = ComputeScissorRects().Best();
auto target_rc = g_framebuffer_manager->ConvertEFBRectangle(native_rc.rect);
auto converted_rc = g_gfx->ConvertFramebufferRectangle(target_rc, g_gfx->GetCurrentFramebuffer());
g_gfx->SetScissorRect(converted_rc);
float raw_x = (xfmem.viewport.xOrig - native_rc.x_off) - xfmem.viewport.wd;
float raw_y = (xfmem.viewport.yOrig - native_rc.y_off) + xfmem.viewport.ht;
float raw_width = 2.0f * xfmem.viewport.wd;
float raw_height = -2.0f * xfmem.viewport.ht;
if (g_ActiveConfig.UseVertexRounding())
{
// Round the viewport to match full 1x IR pixels as well.
// This eliminates a line in the archery mode in Wii Sports Resort at 3x IR and higher.
raw_x = std::round(raw_x);
raw_y = std::round(raw_y);
raw_width = std::round(raw_width);
raw_height = std::round(raw_height);
}
float x = g_framebuffer_manager->EFBToScaledXf(raw_x);
float y = g_framebuffer_manager->EFBToScaledYf(raw_y);
float width = g_framebuffer_manager->EFBToScaledXf(raw_width);
float height = g_framebuffer_manager->EFBToScaledYf(raw_height);
float min_depth = (xfmem.viewport.farZ - xfmem.viewport.zRange) / 16777216.0f;
float max_depth = xfmem.viewport.farZ / 16777216.0f;
if (width < 0.f)
{
x += width;
width *= -1;
}
if (height < 0.f)
{
y += height;
height *= -1;
}
// The maximum depth that is written to the depth buffer should never exceed this value.
// This is necessary because we use a 2^24 divisor for all our depth values to prevent
// floating-point round-trip errors. However the console GPU doesn't ever write a value
// to the depth buffer that exceeds 2^24 - 1.
constexpr float GX_MAX_DEPTH = 16777215.0f / 16777216.0f;
if (!g_ActiveConfig.backend_info.bSupportsDepthClamp)
{
// There's no way to support oversized depth ranges in this situation. Let's just clamp the
// range to the maximum value supported by the console GPU and hope for the best.
min_depth = std::clamp(min_depth, 0.0f, GX_MAX_DEPTH);
max_depth = std::clamp(max_depth, 0.0f, GX_MAX_DEPTH);
}
if (VertexShaderManager::UseVertexDepthRange())
{
// We need to ensure depth values are clamped the maximum value supported by the console GPU.
// Taking into account whether the depth range is inverted or not.
if (xfmem.viewport.zRange < 0.0f && g_ActiveConfig.backend_info.bSupportsReversedDepthRange)
{
min_depth = GX_MAX_DEPTH;
max_depth = 0.0f;
}
else
{
min_depth = 0.0f;
max_depth = GX_MAX_DEPTH;
}
}
float near_depth, far_depth;
if (g_ActiveConfig.backend_info.bSupportsReversedDepthRange)
{
// Set the reversed depth range.
near_depth = max_depth;
far_depth = min_depth;
}
else
{
// We use an inverted depth range here to apply the Reverse Z trick.
// This trick makes sure we match the precision provided by the 1:0
// clipping depth range on the hardware.
near_depth = 1.0f - max_depth;
far_depth = 1.0f - min_depth;
}
// Lower-left flip.
if (g_ActiveConfig.backend_info.bUsesLowerLeftOrigin)
y = static_cast<float>(g_gfx->GetCurrentFramebuffer()->GetHeight()) - y - height;
g_gfx->SetViewport(x, y, width, height, near_depth, far_depth);
}
void SetDepthMode()
{
g_vertex_manager->SetDepthStateChanged();
}
void SetBlendMode()
{
g_vertex_manager->SetBlendingStateChanged();
}
/* Explanation of the magic behind ClearScreen:
There's numerous possible formats for the pixel data in the EFB.
However, in the HW accelerated backends we're always using RGBA8
for the EFB format, which causes some problems:
- We're using an alpha channel although the game doesn't
- If the actual EFB format is RGBA6_Z24 or R5G6B5_Z16, we are using more bits per channel than the
native HW
To properly emulate the above points, we're doing the following:
(1)
- disable alpha channel writing of any kind of rendering if the actual EFB format doesn't use an
alpha channel
- NOTE: Always make sure that the EFB has been cleared to an alpha value of 0xFF in this case!
- Same for color channels, these need to be cleared to 0x00 though.
(2)
- convert the RGBA8 color to RGBA6/RGB8/RGB565 and convert it to RGBA8 again
- convert the Z24 depth value to Z16 and back to Z24
*/
void ClearScreen(const MathUtil::Rectangle<int>& rc)
{
bool colorEnable = (bpmem.blendmode.colorupdate != 0);
bool alphaEnable = (bpmem.blendmode.alphaupdate != 0);
bool zEnable = (bpmem.zmode.updateenable != 0);
auto pixel_format = bpmem.zcontrol.pixel_format;
// (1): Disable unused color channels
if (pixel_format == PixelFormat::RGB8_Z24 || pixel_format == PixelFormat::RGB565_Z16 ||
pixel_format == PixelFormat::Z24)
{
alphaEnable = false;
}
if (colorEnable || alphaEnable || zEnable)
{
u32 color = (bpmem.clearcolorAR << 16) | bpmem.clearcolorGB;
u32 z = bpmem.clearZValue;
// (2) drop additional accuracy
if (pixel_format == PixelFormat::RGBA6_Z24)
{
color = RGBA8ToRGBA6ToRGBA8(color);
}
else if (pixel_format == PixelFormat::RGB565_Z16)
{
color = RGBA8ToRGB565ToRGBA8(color);
z = Z24ToZ16ToZ24(z);
}
g_framebuffer_manager->ClearEFB(rc, colorEnable, alphaEnable, zEnable, color, z);
}
}
void OnPixelFormatChange()
{
// TODO : Check for Z compression format change
// When using 16bit Z, the game may enable a special compression format which we might need to
// handle. Only a few games like RS2 and RS3 even use z compression but it looks like they
// always use ZFAR when using 16bit Z (on top of linear 24bit Z)
// Besides, we currently don't even emulate 16bit depth and force it to 24bit.
/*
* When changing the EFB format, the pixel data won't get converted to the new format but stays
* the same.
* Since we are always using an RGBA8 buffer though, this causes issues in some games.
* Thus, we reinterpret the old EFB data with the new format here.
*/
if (!g_ActiveConfig.bEFBEmulateFormatChanges)
return;
const auto old_format = g_framebuffer_manager->GetPrevPixelFormat();
const auto new_format = bpmem.zcontrol.pixel_format;
g_framebuffer_manager->StorePixelFormat(new_format);
DEBUG_LOG_FMT(VIDEO, "pixelfmt: pixel={}, zc={}", new_format, bpmem.zcontrol.zformat);
// no need to reinterpret pixel data in these cases
if (new_format == old_format || old_format == PixelFormat::INVALID_FMT)
return;
// Check for pixel format changes
switch (old_format)
{
case PixelFormat::RGB8_Z24:
case PixelFormat::Z24:
{
// Z24 and RGB8_Z24 are treated equal, so just return in this case
if (new_format == PixelFormat::RGB8_Z24 || new_format == PixelFormat::Z24)
return;
if (new_format == PixelFormat::RGBA6_Z24)
{
g_renderer->ReinterpretPixelData(EFBReinterpretType::RGB8ToRGBA6);
return;
}
else if (new_format == PixelFormat::RGB565_Z16)
{
g_renderer->ReinterpretPixelData(EFBReinterpretType::RGB8ToRGB565);
return;
}
}
break;
case PixelFormat::RGBA6_Z24:
{
if (new_format == PixelFormat::RGB8_Z24 || new_format == PixelFormat::Z24)
{
g_renderer->ReinterpretPixelData(EFBReinterpretType::RGBA6ToRGB8);
return;
}
else if (new_format == PixelFormat::RGB565_Z16)
{
g_renderer->ReinterpretPixelData(EFBReinterpretType::RGBA6ToRGB565);
return;
}
}
break;
case PixelFormat::RGB565_Z16:
{
if (new_format == PixelFormat::RGB8_Z24 || new_format == PixelFormat::Z24)
{
g_renderer->ReinterpretPixelData(EFBReinterpretType::RGB565ToRGB8);
return;
}
else if (new_format == PixelFormat::RGBA6_Z24)
{
g_renderer->ReinterpretPixelData(EFBReinterpretType::RGB565ToRGBA6);
return;
}
}
break;
default:
break;
}
ERROR_LOG_FMT(VIDEO, "Unhandled EFB format change: {} to {}", old_format, new_format);
}
void SetInterlacingMode(const BPCmd& bp)
{
// TODO
switch (bp.address)
{
case BPMEM_FIELDMODE:
{
// SDK always sets bpmem.lineptwidth.lineaspect via BPMEM_LINEPTWIDTH
// just before this cmd
DEBUG_LOG_FMT(VIDEO, "BPMEM_FIELDMODE texLOD:{} lineaspect:{}", bpmem.fieldmode.texLOD,
bpmem.lineptwidth.adjust_for_aspect_ratio);
}
break;
case BPMEM_FIELDMASK:
{
// Determines if fields will be written to EFB (always computed)
DEBUG_LOG_FMT(VIDEO, "BPMEM_FIELDMASK even:{} odd:{}", bpmem.fieldmask.even,
bpmem.fieldmask.odd);
}
break;
default:
ERROR_LOG_FMT(VIDEO, "SetInterlacingMode default");
break;
}
}
}; // namespace BPFunctions