<tdalign="right"valign="center"><imgsrc="logo32.png"align="right"width="64"height="32"border="0"></td></tr></table><h1align=center>Adding an accelerated graphics driver to TQt/Embedded</h1>
<p> TQt/Embedded has the capacity to make use of hardware accelerators.
To use a hardware accelerator for a PCI or AGP driver, you must
perform the following steps:
<p><oltype=1>
<li>
Define an accelerated descendant of TQLinuxFbScreen.
<p> This should implement <tt>TQVoodooScreen::connect()</tt> to map its
registers. Use <tt>qt_probe_bus</tt> to get a pointer to the PCI config
space. This is where you should check that you're being pointed to the
right device (using the PCI device/manufacturer ID information). Then
use PCI config space to locate your device's accelerator registers in
physical memory and mmap the appropriate region from <tt>/dev/mem</tt>.
There is no need to map the framebuffer, <tt>TQLinuxFbScreen</tt> will do
this for you. Return <tt>FALSE</tt> if a problem occurs at any point. <tt>TQVoodooScreen::initDevice()</tt> will be called only by the TQWS server and
is guaranteed to be called before any drawing is done (and so is a
good place to set registers to known states). <tt>connect()</tt> will be called
by every connecting client.
<p><li>
Define an accelerated descendant of TQGfxRaster.
<p> This is where the actual drawing code goes. Anything not implemented
in hardware can be passed back to <tt>TQGfxRaster</tt> to do in software. Use
the optype variable to make sure that accelerated and unaccelerated
operations are synchronised (if you start drawing via software into an
area where the hardware accelerator is still drawing then your drawing
operations will appear to be in the wrong order). optype is stored in
shared memory and is set to 0 by unaccelerated operations; accelerated
operations should set it to 1. When a software graphics operation is
requested and optype is 1, <tt>TQGfxRaster::sync()</tt> is called; you should
provide your own implementation of this that waits for the graphics
engine to go idle. lastop is also available for optimisation and is
stored in the shared space: this will not be set by the software-only
<tt>TQGfx</tt> and can be used to store the type of your last operation (e.g.
drawing a rectangle) so that part of the setup for the next operation
can be avoided when many of the same operations are performed in
sequence.
<p> All drawing operations should be protected via a <tt>TQWSDisplay::grab()</tt>
before any registers, lastop or optype are accessed, and <tt>ungrabbed()</tt> at the end. This prevents two applications trying to
access the accelerator at once and possibly locking up the machine.
It's possible that your source data is not on the graphics card so you
should check in such cases and fall back to software if necessary.
Note that <tt>TQGfxRaster</tt> supports some features not directly supported
by <ahref="ntqpainter.html">TQPainter</a> (for instance, alpha channels in 32-bit data and
stretchBlt's). These features are used by TQt; stretchBlt speeds up <ahref="ntqpixmap.html#xForm">TQPixmap::xForm</a>() and <tt>drawPixmap()</tt> into a transformed <ahref="ntqpainter.html">TQPainter</a>,
alpha channel acceleration is supported for 32-bit pixmaps.
<p><li>
If you wish, define an accelerated descendant of <tt>TQScreenCursor</tt>. <tt>restoreUnder()</tt>, <tt>saveUnder()</tt>, <tt>drawCursor()</tt> and <tt>draw()</tt> should
be defined as null operations. Implement <tt>set()</tt>, <tt>move()</tt>, <tt>show()</tt> and <tt>hide()</tt>. 4KB is left for your cursor at the end of the
visible part of the framebuffer (i.e. at (width*height*depth)/8 )
<p><li>
Implement <tt>initCursor()</tt> and <tt>createGfx()</tt> in your <ahref="qscreen.html">TQScreen</a>
descendant. Implement <tt>useOffscreen()</tt> and return <tt>TRUE</tt> if you can
make use of offscreen graphics memory.
<p><li>
Implement a small function <tt>qt_get_screen_mychip()</tt>, which simply
returns a new <tt>TQMychipScreen</tt>
<p><li>
Add your driver to the DriverTable table in <tt>qgfxraster_qws.cpp</tt>,
e.g.
<pre>
{ "MyChip", qt_get_screen_mychip,1 },
</pre>
<p> The first parameter is the name used with TQWS_DISPLAY to request your
accelerated driver.
<p><li>
To run with your new driver,
<pre>
export TQWS_DISPLAY=MyChip
</pre>
(optionally MyChip:/dev/fb<n> to request a different Linux
framebuffer than <tt>/dev/fb0</tt>), then run the program
<p></ol>
<p> If your driver is not PCI or AGP you'll need to inherit <ahref="qscreen.html">TQScreen</a>
instead of <tt>TQLinuxFbScreen</tt> and implement similar functionality to <tt>TQLinuxFbScreen</tt>, but otherwise the process should be similar. The most
complete example driver is <tt>qgfxmach64_qws.cpp</tt>; <tt>qgfxvoodoo_qws.cpp</tt> may provide a smaller and easier-to-understand
