base: Redesign internal frame buffer handling

Currently, frame buffer handling in gem5 is quite ad hoc. In practice,
we pass around naked pointers to raw pixel data and expect consumers
to convert frame buffers using the (broken) VideoConverter.

This changeset completely redesigns the way we handle frame buffers
internally. In summary, it fixes several color conversion bugs, adds
support for more color formats (e.g., big endian), and makes the code
base easier to follow.

In the new world, gem5 always represents pixel data using the Pixel
struct when pixels need to be passed between different classes (e.g.,
a display controller and the VNC server). Producers of entire frames
(e.g., display controllers) should use the FrameBuffer class to
represent a frame.

Frame producers are expected to create one instance of the FrameBuffer
class in their constructors and register it with its consumers
once. Consumers are expected to check the dimensions of the frame
buffer when they consume it.

Conversion between the external representation and the internal
representation is supported for all common "true color" RGB formats of
up to 32-bit color depth. The external pixel representation is
expected to be between 1 and 4 bytes in either big endian or little
endian. Color channels are assumed to be contiguous ranges of bits
within each pixel word. The external pixel value is scaled to an 8-bit
internal representation using a floating multiplication to map it to
the entire 8-bit range.

1 files changed, 355 insertions, 0 deletions

diff --git a/src/base/framebuffer.hh b/src/base/framebuffer.hh
new file mode 100644
index 000000000..457c6d06d
--- /dev/null
+++ b/src/base/framebuffer.hh
@@ -0,0 +1,355 @@
+/*
+ * Copyright (c) 2015 ARM Limited
+ * All rights reserved
+ *
+ * The license below extends only to copyright in the software and shall
+ * not be construed as granting a license to any other intellectual
+ * property including but not limited to intellectual property relating
+ * to a hardware implementation of the functionality of the software
+ * licensed hereunder.  You may use the software subject to the license
+ * terms below provided that you ensure that this notice is replicated
+ * unmodified and in its entirety in all distributions of the software,
+ * modified or unmodified, in source code or in binary form.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are
+ * met: redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer;
+ * redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution;
+ * neither the name of the copyright holders nor the names of its
+ * contributors may be used to endorse or promote products derived from
+ * this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ * Authors: Andreas Sandberg
+ */
+
+#ifndef __BASE_FRAMEBUFFER_HH__
+#define __BASE_FRAMEBUFFER_HH__
+
+#include <cmath>
+#include <cstdint>
+
+#include <vector>
+
+#include "base/types.hh"
+
+/**
+ * Internal gem5 representation of a Pixel.
+ */
+struct Pixel
+{
+    Pixel()
+        : red(0), green(0), blue(0), padding(0) {}
+
+    Pixel(uint8_t _red, uint8_t _green, uint8_t _blue)
+        : red(_red), green(_green), blue(_blue), padding(0) {}
+
+    uint8_t red;
+    uint8_t green;
+    uint8_t blue;
+    uint8_t padding;
+};
+
+inline bool
+operator==(const Pixel &lhs, const Pixel &rhs)
+{
+    return lhs.red == rhs.red &&
+        lhs.green == rhs.green &&
+        lhs.blue == rhs.blue &&
+        lhs.padding == rhs.padding;
+}
+
+
+/**
+ * Configurable RGB pixel converter.
+ *
+ * This class converts between external RGB representations and gem5's
+ * internal Pixel representation. The class assumes that pixels are
+ * stored in a word of configurable length (up to 32 bits). Individual
+ * pixels are assumed to be represented by contiguous bit ranges in
+ * the word (i.e., it is possible to shift and mask out a contiguous
+ * bit range for each pixel).
+ */
+class PixelConverter
+{
+  public:
+    /**
+     * Color channel conversion and scaling helper class.
+     */
+    struct Channel {
+        /**
+         * @param offset Offset in bits.
+         * @param width Width in bits.
+         */
+        Channel(unsigned offset, unsigned width);
+
+        /**
+         * Get the value of a single color channel represented as an
+         * 8-bit number.
+         */
+        uint8_t toPixel(uint32_t word) const {
+            return round(((word >> offset) & mask) * factor);
+        }
+
+        /**
+         * Convert an 8-bit representation of a color into an external
+         * format.
+         */
+        uint32_t fromPixel(uint8_t ch) const {
+            return (static_cast<uint8_t>(round(ch / factor)) & mask) << offset;
+        }
+
+        /** Offset in bits */
+        unsigned offset;
+        /** Bit mask (after shifting) */
+        unsigned mask;
+        /**
+         * Scaling factor when converting to the full range of an
+         * 8-bit color channel
+         */
+        float factor;
+    };
+
+    PixelConverter(unsigned length,
+                   unsigned ro, unsigned go, unsigned bo,
+                   unsigned rw, unsigned gw, unsigned bw,
+                   ByteOrder byte_order = LittleEndianByteOrder);
+
+    /** Get the Pixel representation of a color word. */
+    Pixel toPixel(uint32_t word) const {
+        return Pixel(ch_r.toPixel(word),
+                     ch_g.toPixel(word),
+                     ch_b.toPixel(word));
+    }
+
+    /** Get a Pixel representation by reading a word from memory. */
+    Pixel toPixel(const uint8_t *rfb) const {
+        return toPixel(readWord(rfb));
+    }
+
+    /** Convert a Pixel into a color word */
+    uint32_t fromPixel(const Pixel &pixel) const {
+        return ch_r.fromPixel(pixel.red) |
+            ch_g.fromPixel(pixel.green) |
+            ch_b.fromPixel(pixel.blue);
+    }
+
+    /**
+     * Convert a pixel into a color word and store the resulting word
+     * in memory.
+     */
+    void fromPixel(uint8_t *rfb, const Pixel &pixel) const {
+        writeWord(rfb, fromPixel(pixel));
+    }
+
+    /**
+     * Read a word of a given length and endianness from memory.
+     *
+     * The number of bytes read from memory is determined by the
+     * length of a color word. Note that some of the bytes may be
+     * padding.
+     *
+     * @param p Pointer to the first byte in the word.
+     * @return Word in host endianness.
+     */
+    uint32_t readWord(const uint8_t *p) const;
+    /**
+     * Write a word of a given length and endianness to memory.
+     *
+     * @param p Pointer to the first byte in memory.
+     * @param word Word to store (host endianness).
+     */
+    void writeWord(uint8_t *p, uint32_t word) const;
+
+    /** Bytes per pixel when stored in memory (including padding) */
+    unsigned length;
+    /**
+     * Number of bits used to represent one pixel value (excluding
+     * padding). This could be less than length * 8 if the pixel value
+     * is padded.
+     */
+    unsigned depth;
+    /** Byte order when stored to memory. */
+    ByteOrder byte_order;
+
+    /** Red channel conversion helper */
+    Channel ch_r;
+    /** Green channel conversion helper */
+    Channel ch_g;
+    /** Blue channel conversion helper */
+    Channel ch_b;
+
+    /** Predefined 32-bit RGB (red in least significant bits, 8
+     * bits/channel, little endian) conversion helper */
+    static const PixelConverter rgba8888_le;
+    /** Predefined 16-bit RGB565 (red in least significant bits,
+     * little endian) conversion helper */
+    static const PixelConverter rgb565_le;
+
+    /** Predefined 32-bit RGB (red in least significant bits, 8
+     * bits/channel, big endian) conversion helper */
+    static const PixelConverter rgba8888_be;
+    /** Predefined 16-bit RGB565 (red in least significant bits,
+     * big endian) conversion helper */
+    static const PixelConverter rgb565_be;
+};
+
+/**
+ * Internal gem5 representation of a frame buffer
+ *
+ * Display controllers and other devices producing images are expected
+ * to use this class to represent the final image.
+ *
+ * Pixels are indexed relative to the upper left corner of the
+ * image. That is, the pixel at position (0, 0) is the upper left
+ * corner. The backing store is a linear vector of Pixels ordered left
+ * to right starting in the upper left corner.
+ */
+class FrameBuffer
+{
+  public:
+    /**
+     * Create a frame buffer of a given size.
+     *
+     * @param width Width in pixels
+     * @param height Height in pixels
+     */
+    FrameBuffer(unsigned width, unsigned height);
+    /** Create an empty (0x0) frame buffer */
+    FrameBuffer();
+
+    virtual ~FrameBuffer();
+
+    /**
+     * Resize the frame buffer.
+     *
+     * This method resizes frame buffer including the backing
+     * store. The contents of the backing store are undefined after
+     * this operation.
+     *
+     * @param with Width in pixels.
+     * @param height Height in pixels.
+     */
+    void resize(unsigned width, unsigned height);
+
+    /** Frame buffer width in pixels */
+    unsigned width() const { return _width; }
+    /** Frame buffer height in pixels */
+    unsigned height() const { return _height; }
+    /** Total number of pixels in frame buffer */
+    unsigned area() const { return _width * _height; }
+
+    /**
+     * Fill the frame buffer with a single pixel value
+     *
+     * @param pixel Pixel value to fill with.
+     */
+    void fill(const Pixel &pixel);
+    /**
+     * Fill the frame buffer with black pixels
+     */
+    void clear();
+
+    /**
+     * Fill the frame buffer with pixel data from an external buffer
+     * of the same width and height as this frame buffer.
+     *
+     * @param fb External frame buffer
+     * @param conv Pixel conversion helper
+     */
+    void copyIn(const uint8_t *fb, const PixelConverter &conv);
+    /**
+     * Fill the frame buffer with pixel data from an external buffer
+     * of the same width and height as this frame buffer.
+     *
+     * @param fb External frame buffer
+     * @param conv Pixel conversion helper
+     */
+    void copyIn(const std::vector<uint8_t> &fb, const PixelConverter &conv) {
+        copyIn(fb.data(), conv);
+    }
+
+    /**
+     * Store the contents of this frame buffer in an external buffer
+     * of the same width and height as this frame buffer.
+     *
+     * @param fb External frame buffer
+     * @param conv Pixel conversion helper
+     */
+    void copyOut(uint8_t *fb, const PixelConverter &conv) const;
+    /**
+     * Store the contents of this frame buffer in an external buffer
+     * of the same width and height as this frame buffer.
+     *
+     * @param fb External frame buffer
+     * @param conv Pixel conversion helper
+     */
+    void copyOut(std::vector<uint8_t> &fb, const PixelConverter &conv) const {
+        copyOut(fb.data(), conv);
+    }
+
+    /**
+     * Get a pixel from an (x, y) coordinate
+     *
+     * @param x Distance from the left margin.
+     * @param y Distance from the top of the frame.
+     */
+    const Pixel &pixel(unsigned x, unsigned y) const {
+        assert(x < _width);
+        assert(y < _height);
+
+        return pixels[y * _width + x];
+    }
+
+    /**
+     * Get a pixel from an (x, y) coordinate
+     *
+     * @param x Distance from the left margin.
+     * @param y Distance from the top of the frame.
+     */
+    Pixel &pixel(unsigned x, unsigned y) {
+        assert(x < _width);
+        assert(y < _height);
+
+        return pixels[y * _width + x];
+    }
+
+    /**
+     * Create a hash of the image that can be used for quick
+     * comparisons.
+     */
+    uint64_t getHash() const;
+
+    /**
+     * Static "dummy" frame buffer.
+     *
+     * This is a dummy frame buffer that can be used as a place holder
+     * for devices that always expect a frame buffer to be present.
+     */
+    static const FrameBuffer dummy;
+
+    /** Frame buffer backing store */
+    std::vector<Pixel> pixels;
+
+  protected:
+    /** Width in pixels */
+    unsigned _width;
+    /** Height in pixels */
+    unsigned _height;
+};
+
+#endif // __BASE_FRAMEBUFFER_HH__